Chem. Res. Toxicol. 2004, 17, 929-936
929
Synthesis, Characterization, and Identification of N7-Guanine Adducts of Isoprene Monoepoxides in Vitro Petra Begemann,†,‡ Nadia I. Christova-Georgieva,† Ramiah Sangaiah,† Hasan Koc,† Daping Zhang,§ Bernard T. Golding,§ Avram Gold,† and James A. Swenberg*,† Department of Environmental Sciences and Engineering, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-7431, and Department of Chemistry, Bedson Building, Newcastle University, Newcastle upon Tyne, NE1 7RU United Kingdom Received December 9, 2003
Isoprene (IP, 2-methylbuta-1,3-diene) is ubiquitous in the environment through emission by plants, combustion processes, and endogenous formation and exhalation by mammals, including humans. IP is also an industrial chemical, widely used in the manufacture of synthetic rubber and plastics. Like butadiene, IP is metabolized to reactive epoxides, which form adducts with macromolecules, and is a demonstrated carcinogen in mice. To date, DNA adducts of IP monoepoxides have not been reported. We report here on the formation of N7-guanine (N7Gua) adducts of isoprene-1,2-oxide (IP-1,2-O, 2-ethenyl-2-methyloxirane) and isoprene-3,4oxide (IP-3,4-O, propen-2-yloxirane). DNA adducts are useful as biomarkers to estimate exposure, as well as to investigate mechanisms of IP carcinogenesis. Incubation of 2′deoxyguanosine with the monoepoxides followed by deglycosylation gave four N7-Gua adducts that were isolated by HPLC and characterized by high-resolution FAB+-MS, ESI+-MS, ESI+MS/MS, and 1H NMR and two-dimensional heteronuclear 1H, 13C correlation NMR spectrometry. IP-1,2-O and IP-3,4-O reacted at both terminal and internal oxirane carbons to form the following regioisomeric adducts at Gua N7: N7-(2′-hydroxy-2′-methyl-3′-buten-1′-yl)guanine, N7-(1′-hydroxy-2′-methyl-3′-buten-2′-yl)guanine, N7-(1′-hydroxy-3′-methyl-3′-buten-2′-yl)guanine, and N7-(2′-hydroxy-3′-methyl-3′-buten-1′-yl)guanine. The same adducts were identified by UV spectra, HPLC retention times, and LC/ESI+-MS in the neutral thermal hydrolysates of single- and double-stranded calf thymus DNA after incubation with IP monoepoxides. Characterization of the N7-Gua adducts identified in incubations of DNA with IP monoepoxides represents the first step toward establishing biomarkers of IP metabolism and exposure.
Introduction IP1 (CAS No.78-79-5) is ubiquitous in the environment, with ambient concentrations of generally less than 10 ppb (1). Natural sources, including emissions by plants and trees and by combustion, contribute significantly to total IP emissions (2-4). Anthropogenic sources in the community also include combustion processes, e.g., automobile exhaust and cigarette smoke (2-5 and 0.5 mg/ cigarette in side stream and mainstream smoke, respectively) (5-8). Commercially, IP is used in the manufacture of synthetic rubber (cis-polyisoprene) and elastomers (copolymer with styrene or isobutene) (9). In occupational settings in the U.S., 8 h time-weighted average exposures are normally less than 1 ppm (10, 11). * To whom correspondence should be addressed. E-mail: jswenber@ email.unc.edu. † The University of North Carolina at Chapel Hill. ‡ Current Address: ENVIRON Health Sciences Institute, 4350 North Fairfax Drive, Arlington, VA 22203. § Newcastle University. 1 Abbreviations: dGuo, 2′-deoxyguanosine; ds, double-stranded; ESI+-MS, positive ion electrospray mass spectrometry; IP, isoprene; IP-1,2-O, isoprene-1,2-oxide; IP-3,4-O, isoprene-3,4-oxide; IPOcommercial, commercially available 95% isoprene-1,2-oxide with 5% isoprene-3,4oxide; MS/MS, tandem mass spectrometry; N7-Gua I, N7-(2′-hydroxy2′-methyl-3′-buten-1′-yl)guanine; N7-Gua II, N7-(1′-hydroxy-2′-methyl3′-buten-2′-yl)guanine; N7-Gua III, N7-(1′-hydroxy-3′-methyl-3′-buten2′-yl)guanine; N7-Gua IV, N7-(2′-hydroxy-3′-methyl-3′-buten-1′-yl)guanine; ss, single-stranded.
The amount of IP from natural sources exceeds that from industrial output by about 300 times. IP is generated endogenously by mammals, including humans, probably during isoprenoid biosynthesis (1214). It is the main hydrocarbon exhaled from humans giving breath levels of 0.3-9 mg/24 h (14, 15). A 70 kg person generates approximately 17 mg/day, about 30 times the estimated dose received from ambient air (9). Blood concentrations in humans are 15-70 nmol/L (1-5 µg/L), considerably higher than in rats and other mammals (99.8%) was obtained by purification of IPOcommercial by preparative GC performed on a Varian series 2700 gas chromatograph (Varian Instrument Division, Palo Alto, CA) equipped with a thermal conductivity detector. The column used was 2.5 m × 4 mm i.d. 20 M 10% poly(ethylene glycol) (Applied Science Laboratories Inc., State College, PA); nitrogen was the carrier gas with a flow rate of 50 mL/min, the injection temperature was 20 °C, and an oven temperature of 20 °C with a temperature gradient of 8 °C/min to 150 °C (tR, 2.26 min)2 was used. IP-3,4-O (>99%) was synthesized and purified according to Harwood et al. (36). dGuo and calf thymus DNA were purchased from Sigma Chemical Co. (St. Louis, MO). HPLC grade acetonitrile was purchased from Mallinckrodt Baker, Inc. (Paris, KY). General Methods. 1. Reaction of dGuo with IP-1,2-O and IP-3,4-O. dGuo (5 mg, 19 µmol) was dissolved in 1 mL of 50 mM phosphate buffer (pH 7.4) by shaking at 45 °C for 2 h. Aliquots of the solution were transferred to glass vials with screw-top Teflon-coated caps and incubated with a 10-fold molar excess of IP-1,2-O or IP-3,4-O in an environmental shaker at 37 °C for 24 h. The reaction mixtures were diluted 5-fold with doubly distilled water to dissolve all precipitated material and extracted three times with 3 volumes of diethyl ether to remove excess IP monoepoxide. Residual ether was evaporated from the reaction mixture under nitrogen at room temperature. 2
Cottrell et al. Personal communication.
DNA Adducts of Isoprene Monoepoxides 2. Mild Acid Hydrolysis. The deglycosylation of the dGuo adducts was performed in 0.1 M HCl at 80 °C for 1 h. The neutralized hydrolysates were diluted 2-fold with doubly distilled water to dissolve the precipitates and additionally filtered through Uniflo syringe filters (Schleicher & Schuell, Keene, NH) as preparation for HPLC analysis. HPLC of the deglycosylation mixture from the reaction of dGuo with IP-1,2-O on the analytical column showed two major products, N7-Gua I and N-Gua II, with retention times of 28.8 and 32.5 min, respectively. The deglycosylation mixture from the reaction of dGuo with IP-3,4-O also yielded two major products, N7-Gua III and N7-Gua IV, with retention times of 33.9 and 36.2 min, respectively. Synthesis of N7-Gua Adducts of IP Monoepoxides. In a 15 mL vial with a screw-top Teflon-coated cap, dGuo (50 mg, 0.19 mmol) was dissolved in 10 mL of 50 mM phosphate buffer (pH 7.4) by shaking at 45 °C for 2 h. IPOcommercial (200 µL, 2 mmol) was added, and the mixture was incubated in an Orbit environmental shaker (Lab-Line Instruments, Inc., Melrose Park, IL) at 37 °C for 24 h. The reaction mixture was extracted three times with 3 volumes of diethyl ether to remove excess IP monoepoxides, and the residual ether was evaporated from the reaction mixture under nitrogen at room temperature. 1. Mild Acid Hydrolysis. The extracted reaction mixture was centrifuged to remove the precipitated Gua formed by deglycosylation of dGuo during the incubation period. To 450 µL aliquots of the supernatant, 50 µL of 1 M HCl was added and the mixture was heated at 80 °C for 1 h. The hydrolysates were neutralized with 25 µL of 2 M KOH and filtered through Uniflo syringe filters (Schleicher & Schuell) before separation by semipreparative HPLC. The fractions containing N7-Gua I-IV were collected and dried in a Savant SVC100 Speed Vac (Thermo Savant, Milford, MA). The dried fractions were dissolved in a small amount of doubly distilled water, and the fractions belonging to each adduct were combined, lyophilized, and characterized by NMR and exact mass measurement. 2. N7-Gua I. 1H NMR (500 MHz, methyl sulfoxide-d6): 11.36 (bs, 1H, N1-H), 7.73 (s, 1H, C8-H), 6.27 (s, 2H, C2-NH2), 5.92 (dd, 1H, J3′-4′trans ) 17.8, J3′-4′cis ) 11.7 Hz, C3′-H), 5.42 (s, 1H, C2′-OH), 5.15 (dd, 1H, J3′-4′trans ) 17.8, J4′trans-4′cis ) 1.9 Hz, C4′Htrans), 4.96 (dd, 1H, J3′-4′cis ) 11.7, J4′trans-4′cis ) 1.9 Hz, C4′Hcis), 4.24 (d, 1H, J1′a-1′b ) 13.7 Hz, C1′-Hb), 4.18 (d, 1H, J1′a-1′b ) 13.7 Hz, C1′-Ha), 1.09(s, 3H, 2′-CH3) ppm. 13C NMR (125 MHz, methyl sulfoxide-d6): 158.9 (C4), 143.5 (C8), 142.8 (C3′), 112.9 (C4′), 108.7 (C5), 71.5 (C2′), 54.0 (C1′), 25.0 (CH3) ppm. FAB+ HRMS m/z calcd for C10H13N5O2 [M + H]+, 236.1147; found, 236.1164. 3. N7-Gua II. 1H NMR (500 MHz, methyl sulfoxide-d6): 11.25 (bs, 1H, N1-H), 7.85 (s, 1H, C8-H), 6.31 (dd, 1H, J3′-4′trans ) 17.6, J3′-4′cis ) 10.9 Hz, C3′-H), 6.28(s, 2H, C2-NH2), 5.11 (d, 1H, J3′-4′cis ) 10.9 Hz, C4′-Hcis), 5.02 (s, 1H, C1′-OH), 5.00 (d, 1H, J3′-4′trans ) 17.6 Hz, C4′-Htrans), 3.92 (d, 1H, J1′a-1′b ) 10.7 Hz, C1′-Hb), 3.75 (d, 1H, J1′a-1′b ) 10.7 Hz, C1′-Ha), 1.62 (s, 3H, 2′CH3) ppm. 13C NMR (125 MHz, methyl sulfoxide-d6) 161.0 (C4), 141.7 (C8), 139.9 (C3′), 114.9 (C4′), 108.1 (C5), 66.2 (C1′), 63.4 (C2′), 21.4 (CH3) ppm. FAB+ HRMS m/z calcd for C10H13N5O2 [M + H]+, 236.1147; found, 236.1160. 4. N7-Gua III. 1H NMR (500 MHz, methyl sulfoxide-d6): 11.33 (bs, 1H, N1-H), 7.90 (s, 1H, C8-H), 6.35 (s, 2H, C2-NH2), 5.22 (dd, 1H, J2′-1′a ) 6.4, J2′-1′b ) 3.9 Hz, C2′-H), 5.12 (bs, 1H, C1′-OH), 4.83 (bs, 1H, C4′-Ha), 4.58 (bs, 1H, C4′-Hb), 3.95 (ψt, 1H, J1a′-1′b ) 9.9, J2′-1b′ ) 6.4 Hz, C1′-Hb), 3.80 (dd, 1H, J1a′-1′b ) 9.9, J2′-1a′ ) 3.9 Hz, C1′-Ha), 1.53 (s, 3H, 3′-CH3) ppm. 13C NMR (125 MHz, methyl sulfoxide-d6): 159.5 (C4), 142.1 (C3′), 141.2 (C8), 112.6 (C4′), 108.2 (C5), 62.7 (C2′), 60.5 (C1′), 19.7 (CH3) ppm. FAB+ HRMS m/z calcd for C10H13N5O2 [M + H]+, 236.1147; found, 236.1135. 5. N7-Gua IV. 1H NMR (500 MHz, methyl sulfoxide-d6): 11.64 (bs, 1H, N1-H), 7.66 (s, 1H, C8-H), 6.28 (bs, 2H, C2-NH2), 5.34 (bs, 1H, C2′-OH), 4.80 (bs, 1H, C4′-Ha), 4.71 (bs, 1H, C4′Hb), 4.24 (dd, 1H, J1a′-1′b ) 14.0, J2′-1′b) 4.5 Hz, C1′-Hb), 4.17 (dd, 1H, J2′-1′a ) 7.4, J2′-1′b ) 4.5 Hz, C2′-H), 3.95 (dd, 1H, J1a′-1′b
Chem. Res. Toxicol., Vol. 17, No. 7, 2004 931 ) 14.0, J2′-1′a ) 7.4 Hz, C1′-Ha), 1.53 (s, 3H, 3′-CH3) ppm. 13C NMR (125 MHz, methyl sulfoxide-d6): 145.6 (C3′), 143.2 (C8), 111.0 (C4′), 108.5 (C5), 72.7 (C2′), 50.1 (C1′), 25.3 (CH3) ppm. FAB+ HRMS m/z calcd for C10H13N5O2 [M + H]+, 236.1147; found, 236.1151. Reaction of IP Monoepoxides with Calf Thymus DNA. Calf thymus DNA was hydrated in doubly distilled water overnight at 4 °C (4.6 mg/mL). An equal volume of 100 mM phosphate buffer (pH 7.4) was added, and the solution (2.3 mg/ mL) was sheared with a 19 gauge needle. 1. Preparation of ss DNA. One milliliter aliquots of the DNA solution in Eppendorf vials were heated in a water bath to 100 °C for 5 min and then immediately cooled on ice for 10 min (37, 38). One milliliter aliquots of ss or ds DNA solution were transferred into glass vials with screw-top Teflon-coated caps and incubated with 90 µL of IPOcommercial in an environmental shaker at 37 °C for 24 h. The reaction mixture was worked up as described above for dGuo. 2. Neutral Thermal Hydrolysis of Unstable DNA Adducts (N7-dGuo). Aliquots of 500 µL of the extracted reaction mixture in Eppendorf vials were heated to 100 °C for 30 min. After they were cooled, the samples were transferred to Microcon YM-10 filters (Millipore Corp., Bedford, MA), centrifuged at 22 000g, and washed once with 30 µL of doubly distilled water. The filtrate was analyzed by HPLC and LC/MS. 3. HPLC Analysis and Purification. HPLC separations and purification were performed on Platinum EPS columns (Alltech, Deerfield, IL), using an Agilent Technologies (Palo Alto, CA) 1100 series quaternary pump and a Hewlett-Packard (Avondale, PA) 1040A diode array detector (DAD) and HP Chem Station Software (Rev. A.05.04). Chromatograms were recorded with detector wavelengths set at 285 (λmax for the N7-Gua adducts of IP monoepoxides) and 235 nm (to record the coelution of byproducts). The following linear gradient program was used for all separations: 0-1 min 0% B; 1-20 min to 12% B; 20-50 min to 22% B. Mobile phase A was 17 mM ammonium acetate buffer (pH 5.5) containing 1.5% acetonitrile; mobile phase B was 17 mM ammonium acetate buffer (pH 5.5) containing 30% acetonitrile. Analytical separations were performed by injection of 80-100 µL of the hydrolysates onto a 4.6 mm × 250 mm, 5 µm, 100 Å reversed phase column with a flow of 1 mL/min. Semipreparative separations were performed by injection of 1 mL amounts of the hydrolysate onto a 10 mm × 250 mm, 5 µm, 100 Å reversed phase column with a flow of 3 mL/min. 4. LC/ESI+-MS. LC/ESI+-MS analyses were performed on an LCQDECA quadrupole ion trap mass spectrometer (Thermo Finnigan, San Jose, CA) equipped with an API2 electrospray ionization source operating in the positive ion mode. The heated capillary of the ESI source was kept at 350 °C. Nitrogen was used as both the sheath and the auxiliary gas and maintained at 80 and 20 arbitrary units, respectively. The spray voltage was 4.5 kV. Reaction mixtures (10 µL) were analyzed by on-line LC/ESI+MS to acquire full scan ESI spectra. Online LC separations were carried out as described above for analytical HPLC conditions. A postcolumn split of the LC flow was accomplished using a Valco (Houston, TX) T-connector to allow only about one-third (350 µL/min) of the LC effluent into the ESI source. Direct injection of the HPLC-purified adducts was performed by introducing 5 µL of each sample dissolved in 30% acetonitrile in water into the mass spectrometer in order to obtain full scan and product ion ESI+-MS spectra for the adducts. The MS/MS collision energy was 35 V, and the argon pressure was 2.5 mTorr. 5. Exact Mass Measurement. High-resolution mass measurements were obtained using a VG70-250 SEQ mass spectrometer (Vacuum Generators, Wythenshawe, Cheshire, U.K.) with a FAB source by direct insertion probe. 6. NMR Spectrometry. 1H NMR spectra were recorded at 500 MHz on a Varian INOVA 500 spectrometer using 1-2 mg of the purified, lyophilized adducts dissolved in methyl sulfoxided6. The chemical shifts are reported in ppm relative to the proton
932
Chem. Res. Toxicol., Vol. 17, No. 7, 2004
Begemann et al.
Figure 1. HPLC analysis of the reaction mixtures of dGuo incubated with IP-1,2-O (a), IP-3,4-O (b), and IPOcommercial (c) after mild acid hydrolysis. Trace d shows the HPLC chromatogram from the neutral thermal hydrolysis of ds calf thymus DNA incubated with IPOcommercial. Separation was performed on an Alltech Platinum EPS column (4.6 mm × 250 mm, 5 µm, 100 Å), using an Agilent Technologies 1100 series quaternary pump and a Hewlett-Packard 1040A DAD measuring 285 and 235 nm. A linear gradient at 1 mL/min was used as follows: 0-1 min, 0% B; 1-20 min, to 12% B; 20-50 min, to 22% B. A, 17 mM ammonium acetate buffer (pH 5.5) containing 1.5% acetonitrile; B, 17 mM ammonium acetate buffer (pH 5.5) containing 30% acetonitrile. resonance of TMS. Heteronuclear single quantum coherence (HSQC) and heteronuclear multiple bond correlation (HMBC) NMR experiments were performed on the Varian INOVA 500 spectrometer with 13C signals acquired at 125 MHz. 1H and 13C signal assignments were made on the basis of HSQC and HMBC NMR experiments.
Results Reaction of Racemic IP Monoepoxides with dGuo and Characterization of N7-Gua Adducts. The HPLC trace of the reaction mixture resulting from treatment of dGuo with racemic IP-1,2-O followed by deglycosylation showed two major adduct peaks (N7-Gua I and II) with λmax at 285 and a shoulder at about 250 nm (Figure 1a) (35, 39). On this basis, the adducts were identified as N7 alkylation products. Because N7 can attack at
either the internal or the terminal oxirane carbons, N7Gua I and II represent pairs of racemic regioisomers. Four minor peaks with UV spectra different from N7Gua adducts remain to be identified (Figure 1a). Two peaks appeared in the deglycosylation reaction mixture from treatment of dGuo with racemic IP-3,4-O (N7-Gua III and IV; Figure 1b). The UV spectra were virtually identical to those observed for the major reaction products of dGuo with IP-1,2-O and also had the λmax at 285 and a shoulder at about 250 nm of N7-Gua adducts. Thus, N7-Gua III and IV were likewise identified as racemic regioisomeric N7 adducts. HPLC diode array analysis of the deglycosylated reaction mixture generated with IPOcommercial showed four major peaks corresponding to adducts N7-Gua I-IV in retention times and UV spectra. Peaks 1 and 3 cor-
DNA Adducts of Isoprene Monoepoxides
Chem. Res. Toxicol., Vol. 17, No. 7, 2004 933
Figure 2. Collision-induced MS/MS spectra of the molecular ion m/z 236 [M + H]+ of adducts N7-Gua I-IV. LC/ESI+-MS analyses were performed on an LCQDECA quadrupole ion trap mass spectrometer equipped with an API2 electrospray ionization source operating in the positive ion mode. The heated capillary of the ESI source was kept at 350 °C. Nitrogen was used as both the sheath and the auxiliary gas and maintained at 80 and 20 arbitrary units, respectively. The spray voltage was 4.5 kV. The MS/MS collision energy was 35 V, and the argon pressure was 2.5 mTorr.
responded to adducts N7-Gua I and II, and peaks 2 and 4 corresponded to adducts N7-Gua III and IV, respectively (Figure 1c). The adduct peaks were well-resolved under semipreparative HPLC conditions, and fractions were therefore collected and processed without further purification. The adducts isolated from treatment of dGuo with IPOcommercial were characterized by FAB+-HRMS, ESI+-MS, ESI+-MS/MS, and 1H NMR and heteronuclear 1H, 13C chemical shift correlation experiments. The elemental compositions of adducts N7-Gua I-IV by exact mass measurements on the protonated molecular ions were consistent with N7 adducts. Analysis of the reaction mixtures by LC/ESI+-MS in SCAN mode under the same chromatographic conditions as HPLC DAD showed a major ion at m/z ) 236 corresponding to the protonated molecular ion [M + H]+ expected for the IP monoepoxide-Gua adducts, along with a minor fragment at m/z ) 152 resulting from loss of the side chain. The collision-induced MS/MS spectra of [M + H]+ showed product ions at m/z ) 218 and 152, resulting from the loss of water and the side chain, respectively (Figure 2). The MS/MS of N7-Gua III showed, in addition, a minor product ion at m/z ) 206, from loss of formaldehyde (Figure 2). In the 1H NMR spectra of the IP monoepoxide adducts, signals of the Gua moiety were assigned by comparison with data reported for the N7-Gua adducts of butadiene monoepoxide (35). On this basis, an extremely broad, low intensity peak at 11.36 ppm is assigned to the imido N1H: a broad two proton singlet at 6.27 ppm to the protons of the exocyclic amino group, and a sharp one proton singlet at 7.73 ppm to C8-H (N7-Gua I). The assignment of C8-H was confirmed by 1H, 13C HSQC and HMBC experiments (Figure 3).
Figure 3. 1H, 13C HMBC spectrum (500 MHz, DMSO-d6) of N7-Gua I: cross-peaks between the side chain C-1′ methylene protons and purine carbons C-5 and C-8. HMBC NMR experiments were performed on a Varian INOVA 500 spectrometer with 13C signals acquired at 125 MHz.
For the regioisomers derived from IP-1,2-oxide, sufficient material was available to establish the point of attachment of the side chain to Gua-N7 by HSQC and HMBC NMR experiments. In the case of N7-Gua I, the diastereotopic C-1′-methylene protons show cross-peaks with C-5 and C-8 of the purine ring in the HMBC spectrum (Figure 3), fixing C-1′ of the side chain as the point of attachment. On the basis of the structural assignment of N7-Gua I, N7-Gua II can be assigned as the N7-C2′ adduct, although a cross-peak between C8-H of the purine and C-2′ of the side chain was not detected in the HMBC spectrum. Additionally, the point of attachment for regioisomers N7-Gua I and II can be confirmed by 13C chemical shifts derived from combined HSQC/HMBC data. In otherwise identical chemical
934
Chem. Res. Toxicol., Vol. 17, No. 7, 2004
Begemann et al.
Scheme 2. Adduct Formation on N7-Gua
environments, carbon will experience a larger downfield chemical shift when substituted with a hydroxy group than with an imido nitrogen. The signal of C-1′, which is assigned as the point of attachment to N7 in N7-Gua I, appears 12.2 ppm upfield with respect to C-1′ of N7-Gua II, which bears an -OH substituent. Comparison of the C-2′ shifts shows that the C-2′ resonance of N7-Gua II appears 8.1 ppm upfield with respect to C-2′ of N7-Gua I, consistent with the assignment of C-2′ as the point of attachment for N7-Gua II. No cross-peaks were observed in the HMBC spectra between the side chains and the purine nucleus in the dilute solutions of N7-Gua III and N7-Gua IV. However, the structures of the regioisomers could be assigned by 13 C shifts using the rationale described above for N7-Gua adducts I and II. At 50.1 ppm, C-1′ of N7-Gua IV is shifted 10.4 ppm upfield relative to C-1′ of N7-Gua III at 60.5 ppm, while at 62.7 ppm, C-2′ of N7-Gua III appears 10 ppm upfield from C-2′ of N7-Gua IV at 72.7 ppm. This pattern is consistent with the assignment of N7 substitution at C-2′ for N7-Gua III and at C-1′ for N7-Gua IV. Proton assignments on the side chains were straightforward, based on splitting patterns, coupling constants, and 1H chemical shifts as well as connectivities derived from HSQC/HMBC NMR experiments. In all of the N7 adduct side chains, the methylene protons show diastereotopic character, in accordance with proximity to a chiral center. The presence of three vinyl resonances in the spectra of N7-Gua I and N7-Gua II is in accord with expectations for nucleophilic addition of N7 at the oxirane carbons of IP-1,2-oxide. The signals of the vinyl
protons at C-3′ appear as well-resolved doublets-ofdoublets through coupling with the terminal vinyl protons at C-4′. The terminal vinyl signals of N7-Gua I show a small geminal coupling of
