Structural and Stereoisomer Effects of Model Estrogen Quinone

Feb 24, 2004 - small proportion of the incorrect base dC opposite the lesion, and they ...... prediction by random periareolar fine-needle aspiration ...
0 downloads 0 Views 1MB Size
Chem. Res. Toxicol. 2004, 17, 311-324

311

Structural and Stereoisomer Effects of Model Estrogen Quinone-Derived DNA Adducts: 6 N -(2-Hydroxyestron-6(r,β)-yl)-2′-deoxyadenosine and N2-(2-Hydroxyestron-6(r,β)-yl)-2′-deoxyguanosine Lihua Wang,† Brian E. Hingerty,‡ Robert Shapiro,§ and Suse Broyde*,† Biology Department and Chemistry Department, New York University, New York, New York 10003, and Life Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830-6480 Received October 26, 2003

An extensive conformational analysis has been carried out for two diastereoisomeric pairs of model estrogen quinone-derived DNA adducts, N6-(2-hydroxyestron-6(R,β)-yl)-2′-deoxyadenosine (2-OHE1-6(R,β)-N6-dA) and N2-(2-hydroxyestron-6(R,β)-yl)-2′-deoxyguanosine (2-OHE16(R,β)-N2-dG), in a B-DNA duplex and at a primer-template junction in a pol R family DNA polymerase. In vitro primer extension studies in pol R [Terashima, I., et al. (1998) Biochemistry 37, 13807-13815] have shown that the adenine adducts can incorporate dT, together with a small proportion of the incorrect base dC opposite the lesion, and they block less strongly than the guanine adducts. We have carried out conformational searches with energy minimization for four DNA duplexes containing 2-OHE1-6R-N6-dA, 2-OHE1-6β-N6-dA, 2-OHE1-6R-N2-dG, or 2-OHE1-6β-N2-dG. Our searches revealed that the four-ring nonplanar 2-hydroxyestrone (2OHE1) moiety strongly prefers to reside in the major groove of the adenine adducts or the minor groove of the guanine adducts in a B-DNA duplex, with stereochemistry-dependent orientational differences in each case. No low energy conformations involving intercalation of the 2-OHE1 moiety were located in the searches. This stems from the largely nonplanar, nonaromatic nature of the 2-OHE1 ring system and implies that the proclivity for such bulky, nonplanar adducts to reside at the DNA helix exterior is a plausible conformational feature of other structurally similar estrogen quinone-derived DNA adducts, independent of base sequence context. In addition, the adenine adduct isomers, located in the major groove, manifest serious disturbance to the Watson-Crick base pairs at and near the lesion site, suggesting repair susceptibility. Possible structures of these adducts in a pol R family polymerase were also investigated through molecular modeling. The results rationalized the experimental in vitro primer extension studies. In addition, poor accommodation of the β-stereoisomers within the polymerase was noted, suggesting that these stereoisomers would be more prone to cause blockage. Stereochemistry-dependent differences in adduct orientation could be expected to produce different biochemical effects, as has been observed in adducts derived from polycyclic aromatic hydrocarbons.

Introduction Epidemiological studies have shown that endogenous as well as synthetic estrogens contribute to the formation and development of mammary and endometrial cancers (1-3). The potent mitogenic effects of estrogens in the target organs are initiated through an interaction between the estrogen and its cytosolic receptor, which subsequently activates the expression of downstream genes (4-6). Excessive and prolonged exposure of the susceptible organs to mitogenic stimulation by natural or artificial estrogens has been shown to correlate with the induction of estrogen-associated cancers in experimental animals and humans (2, 3, 7-11). Malignant phenotypes, however, are a rare consequence of estrogeninduced proliferation, and the hormonal potencies do not * To whom correspondence should be addressed. Tel: 212-998-8231. Fax: 212-995-4015. E-mail: [email protected]. † Biology Department, New York University. ‡ Life Sciences Division, Oak Ridge National Laboratory. § Chemistry Department, New York University.

always correlate with tumor incidence (12, 13). Therefore, it has been proposed that metabolic activation to either redox active and/or electrophilic metabolites may also play an etiologic role in estrogen-related carcinogenesis through the oxidation of the DNA backbone or bases, or the formation of DNA adducts (14-16). Such adducts may produce mutations that can initiate the process of carcinogenesis (17-21). In particular, mutations may inactivate human tumor suppressor genes, such as p53 (22-25), or activate protooncogenes such as N-ras (26, 27). Mutations in p53, which are common genetic alterations in human cancers, have been suggested to precede the development of breast cancer (28, 29). Codon 273 (CGT) of p53 is an important mutational hotspot in breast cancer, as well as in total human cancer (25, 30). Activation of proto-oncogenes eventually may confer a selective growth advantage on cells, resulting ultimately in tumorigenesis upon clonal expansion. The activation of N-ras genes in many tumors, including breast cancer, has been found to be mainly due

10.1021/tx034218l CCC: $27.50 © 2004 American Chemical Society Published on Web 02/24/2004

312

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

Figure 1. Structures of (A) 2-OHE1, (B) the 2-OHE1-6(R,β)N6-dA and 2-OHE1-6(R,β)-N2-dG adducts, and (C) the sequence contents employed in the conformational searches. Torsion angle definitions are as follows: R′, N1-C6-N6-CE6 for the adenine adducts, and N1-C2-N2-CE6 for the guanine adducts; β′, C6N6-CE6-CE5 for the adenine adducts, and C2-N2-CE6-CE5 for the guanine adducts; γ′, CE4-CE3-OE3-HOE3; χ, O4′-C1′-N9C4, where CE3, CE4, CE5, CE6, OE3, and HOE3 are from the 2-OHE1 moiety and the other atoms are from dA or dG. Abbreviations: A*, 2-OHE1-N6-dA; G*, 2-OHE1-N2-dG; CM, methylated cytosine.

to mutations in codon 12, 13, or 61, resulting in single amino acid substitutions (27, 31-35). Endogenous estrogens, E11 and E2, are hydroxylated by cytochrome P450 enzymes to form catechol metabolites, 2-hydroxyestrogens (2-OHE1, Figure 1A, and 2-OHE2) and 4-hydroxyestrogens (4-OHE1 and 4-OHE2) (36, 37), which are readily oxidized to o-quinones (36, 38, 39). It has been shown that DNA adducts can be induced in Syrian hamster kidney treated with E2 (40). Furthermore, Tsutsui et al. (41-43) have demonstrated that abilities to induce DNA adducts in Syrian hamster embryo cells are ranked as follows: 4-OHE1 > 2-OHE1 > 4-OHE2 > 2-OHE2 . E2, E1, which corresponds well to the respective transforming and carcinogenic abilities of these substances. Adducts were also formed in the mammary glands of female Sprague-Dawley rats treated with 4-OHE2 or its quinone (44). The carcinogenicity of these metabolites has been demonstrated by the induction of renal carcinomas in Syrian hamsters by 4-OHE1 and 4-OHE2 (45, 46), and uterine adenocarcinomas in CD-1 mice by E2, 2-OHE2, and 4-OHE2 (47), suggesting an intrinsic and organ-dependent carcinogenicity of the 2- and 4-hydroxy catechol estrogens. In vitro, quinones of 2- and 4-hydroxyestrogens also have been shown to react with DNA to form covalent 1 Abbreviations: 2-OHE -6(R,β)-N6-dA, N6-(2-hydroxyestron-6(R,β)1 yl)-2′-deoxyadenosine; 2-OHE1-6(R,β)-N2-dG, N2-(2-hydroxyestron6(R,β)-yl)-2′-deoxyguanosine; 2-OHE1, 2-hydroxyestrone; E1, estrone; E2, 17β-estradiol; 2-OHE2, 2-hydroxyestradiol; 4-OHE1, 4-hydroxyestrone; 4-OHE2, 4-hydroxyestradiol; CSD, Cambridge Structural Database; BcPh, benzo[c]phenanthrene; BP, benzo[a]pyrene; NER, nucleotide excision repair.

Wang et al.

adducts (39). Stack et al. (48) established that 2-OHE1 quinone directly reacts with dA and dG to form the stable adducts 2-OHE1-6(R,β)-N6-dA and 2-OHE1-6(R,β)-N2-dG (Figure 1B), respectively, while reactions of 4-OHE1 and 4-OHE2 quinones with dG produce unstable and depurinating adducts (44, 48). One possible mechanism underlying the carcinogenicity of the estrogens may stem from the abasic sites resulting from reaction of estrogen 3,4-quinones with dG. An additional possibility is that stable adducts formed by 2-OHE1 quinone will, if not repaired, lead to replication errors that may initiate carcinogenesis. Terashima et al. (49) have studied the miscoding specificities of the 2-OHE1-6(R,β)-N6-dA and 2-OHE16(R,β)-N2-dG adducts using in vitro primer extension reactions catalyzed by mammalian DNA polymerases R, β, and δ. Translesional synthesis on site-specifically modified DNA templates containing a single 2-OHE16(R,β)-N6-dA or 2-OHE1-6(R,β)-N2-dG adduct showed that 2-OHE1-6(R,β)-N2-dG blocked primer extension reactions more strongly than 2-OHE1-6(R,β)-N6-dA and that 2-OHE16(R,β)-N6-dA can misincorporate dC opposite the lesion. Further in vivo studies carried out by the same group revealed that 2-OHE1-6(R,β)-N6-dA and 2-OHE1-6(R,β)N2-dG adducts are mutagenic, generating primarily A f T and G f T transversions, respectively, in simian kidney cells (50). In these studies, a mixture of the R- and β-stereoisomers was employed in each case. To study the conformational characteristics of model estrogen quinone-derived DNA adducts, we have carried out an extensive conformational search for 2-OHE1-6RN6-dA, 2-OHE1-6β-N6-dA, 2-OHE1-6R-N2-dG, and 2-OHE16β-N2-dG in DNA duplex heptamers, using N-ras codon 61 and p53 codon 273 sequences (Figure 1C). Cytosines in the CpG sequences of p53 are intrinsically methylated, which enhances guanine alkylation by a variety of carcinogens (51). One goal of our studies was to deduce conformational features of these adducts in solution, especially those likely to be shared with other estrogen quinone-derived DNA adducts, and consider them in relation to repair susceptibility. In this work, we employed an improved version of our molecular mechanics program DUPLEX, which accurately represents sequencedependent conformational features of B-DNA, and whose force field is reasonably good at estimating energies of carcinogen-DNA adducts according to experimental NMR benchmarks (52). The availability of the crystal structure of a pol R family DNA polymerase with primer-template DNA and dTTP (53) has also provided an opportunity to study the conformations of these adducts at a template-primer junction within the polymerase-DNA-dNTP ternary complex through a modeling study. Our results help rationalize the experimental in vitro findings in primer extension reactions catalyzed by pol R (49) and, more broadly, suggest how bulky estrogen-derived stable adducts may be accommodated within a model high fidelity polymerase for possible translesional synthesis or cause blockage that may lead to bypass by a low fidelity Y family polymerase.

Materials and Methods Construction of 2-OHE1-6(r,β)-N6-dA and 2-OHE1-6(r,β)N2-dG Adducts. Our modeling effort began with a search for a suitable crystal structure in the CSD (54) for 2-OHE1 (Figure

Conformations of Estrogen Quinone-Derived DNA Adducts Table 1. Starting Conformations for 2-OHE1-6(r,β)-N6-dA in Codon 61 Containing Heptamer Sequence of Human N-ras Gene and 2-OHE1-6(r,β)-N2-dG in Codon 273 Containing Heptamer Sequence of Human p53 Gene χ

R′

β′

265° (anti) 0, 18, 36, ..., 342° 0, 18, 36, ..., 342° 85° (syn)

0, 18, 36, ..., 342° 0, 18, 36, ..., 342°

HB bonding scheme Watson-Crick Hoogsteen No hydrogen bonding

1A) or its DNA adducts. Because no such crystal structures were available, we began with E1 (CSD ID: ESTRON10), which differs from 2-OHE1 only by the absence of a hydroxyl group on C2. The hydrogen at this position was replaced with a hydroxyl group. The hydrogen at C6 was replaced by R- or β-linkages to adenine N6 or guanine N2. Bond lengths, bond angles, and torsion angles of the 2-OHE1 moiety and linkage site were implemented in the DUPLEX linked atom algorithm (55) for generating coordinates in torsion angle space. Force Field Parametrization. Partial charges for the modified base were calculated with Gaussian 94 (56), employing the CNDO method, which is compatible with the partial charge set in DUPLEX. Tables S1 and S2 in the Supporting Information give the partial charges. Table S3 in the Supporting Information gives parameters for the contributions to the energy from the torsion angles involving the 2-OHE1 moiety (Figure 1). Conformational Search Strategies. A total of 400 orientations of the carcinogen base linkage torsion angles R′ and β′ were employed as starting conformations for the energy minimizations: R′ ) 0°, 18°, ..., 342° in combination with β′ ) 0°, 18°, ..., 342°. In addition, three different hydrogen-bonding possibilities at the modified base pair were searched for in each R′, β′ combination, using the DUPLEX hydrogen bond penalty function (57, 58): Watson-Crick, Hoogsteen, and a third one involving modified syn guanine (or adenine) opposite anti cytosine (or thymine), with no hydrogen bond. The hydrogen bond penalty function was also employed to locate standard Watson-Crick base pairs at the other sites. Torsion angles for the DNA starting conformation were those of the B-DNA fiber diffraction model (59), except that syn purines at the lesion site were oriented with the glycosidic torsion angle χ at 85° (Figure 1B). This combination of 400 R′ and β′ values, together with three different hydrogen-bonding possibilities at the lesion site, provided a total of 1200 starting structures for each type of adduct (Table 1). Each of these structures was energy-minimized with DUPLEX. A second minimization was performed for each resulting structure without the hydrogen bond penalty function. The searches for Hoogsteen G-C base pairing employed a protonated N3 of cytosine in the first minimization stages. The structure was deprotonated and minimized again in a terminal step, as in earlier work (57, 60). This procedure was necessary because energies of protonated and unprotonated forms cannot be compared as they contain different numbers of atoms. To ensure that intercalated structures were not missed due to the multiple minimum problem, we carried out additional searches from explicitly intercalated starting structures. For the adenine adducts, we employed NMR solution structures of S and R isomers of adenine adducts derived from BcPh (61, 62), which were classically intercalated on the 3′ or 5′ side of the modified adenine, respectively, as starting models. For the guanine adducts, we employed the NMR solution structures of S and R isomers of guanine adducts derived from BP (63, 64), which were intercalated between adjacent base pairs, and displaced the modified guanine residue into the major groove and minor groove, respectively, as starting structures; also, a classically intercalated structure was created using the NMR solution structure of an S isomer of guanine adducts derived from BcPh, as a starting model (65). In each case, the sequence was remodeled to our sequence context, and the carcinogen in the starting model was replaced with the 2-OHE1 moiety. The backbone torsion angles from the NMR solution structures or minimized B-DNA were used as the starting conformation for the DNA duplex. Lesion site torsion angles χ, R′, β′, and also

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 313 β′ + 180° from the NMR solution structures were used as the starting conformations for the 2-OHE1 adducts. Thus, each NMR solution structure provided four trials. To analyze the structures after minimization, final energy minima were classified into structural groups based on their similarity in the three lesion site torsion angles χ, R′, and β′. Structures whose χ, R′, and β′ values were all within (15° of the lowest energy variant were classified as one group. Visual inspection of all minima up to 20 kcal/mol was then employed to consolidate conformationally similar groups into one family. Molecular Modeling within the Active Site of a Pol r Family DNA Polymerase. A molecular modeling study was carried out to find possible conformations of 2-OHE1-6(R,β)-N6dA and 2-OHE1-6(R,β)-N2-dG within the pol R family RB69 DNA polymerase (53). The ternary complex containing primer/ template DNA and incoming dTTP, whose coordinates (PDB ID: 1IG9) were obtained from the Protein Data Bank (66), served as the starting model. The primer-template sequence, including the incoming dNTP, was remodeled to the codon 61 containing N-ras sequence or the codon 273 containing p53 sequence used in the conformational searches with DUPLEX (Figure 1C), so that the modified dA or dG is opposite an incoming dTTP or dCTP, respectively. Incoming dCTP was also investigated for the adenine adducts. The dideoxy end of the primer in the crystal was remodeled to a 3′-OH. We searched for sterically feasible orientations of the adducts within the polymerase by rotating R′ and β′ at 10° intervals, in combination, over their 360° range and evaluating each structure for collisions. In the searches for 2-OHE1-6(R,β)-N6-dA adducts, the glycosidic torsion angle of the modified dA was retained in its anti conformation at 196°, as in the crystal structure. In the searches for 2-OHE1-6(R,β)-N2-dG adducts, the modified dG had to adopt the less favored syn conformation because it was impossible to accommodate the 2-OHE1 moiety on the minor groove side, where the polymerase interacts closely with the DNA. The torsion angles of the template strand were slightly adjusted (Table S4, Supporting Information) so that the modified syn guanine aligns with the incoming dCTP without distorting the active site while maintaining the steric fit of a normal Watson-Crick base pair. In these models, the Ca2+ ions in the exonuclease domain and at the active site were replaced with catalytic Mg2+ ions. An extra Ca2+ fortuitously trapped by crystallization (CA 1003) was deleted. To bring the two ions at the active site into reasonable octahedral coordination with incoming dNTP and polymerase, minor adjustments were made to the torsion angles of the side chain of Asp 623 and Ser 624 (Table S4); the position of Mg2+ ion at x, -2.85; y, 45.18; z, 13.54 was relocated to x, -2.34; y, 46.28; z, 13.18, to achieve octahedral coordination, which was incomplete in the crystal structure. The torsion angles of the primer backbone were also slightly adjusted (Table S4, Supporting Information) so that the 3′-O of the primer was reasonably close (∼3.2 Å) to the R-phosphate of the incoming dNTP. All structural visualizations and modifications were made using the molecular modeling program Insight II (Accelrys Inc.).

Results 2-OHE1-6r-N6-dA:

Major Groove Conformations. The 1208 starting structures were energy-minimized, and the resulting structures were sorted into conformational families. Families are numbered similarly if they belonged in the same χ, R′, and β′ domain for the R- and β-stereoisomer pair. One hundred twenty-three groups of conformations, sorted as described in Materials and Methods, were obtained. However, only eight conformational families were found up to 20 kcal/mol. In all of these families, the 2-OHE1 ring system was located on the major groove side of the DNA duplex. However, the modified adenine was able to adopt either the normal B-DNA anti conformation or the abnormal syn conforma-

314

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

Wang et al.

Figure 2. Space-filling (left) and stick (right) models of the lowest energy variant of (A) 2-OHE1-N6-6R-dA family 1; (B) 2-OHE1N6-6β-dA family 4; (C) 2-OHE1-N2-6R-dG family 1; and (D) 2-OHE1-N2-6β-dG family 2 conformers, shown in side-by-side stereoviews. Color code: silver, unmodified strand; cyan, modified strand; orange, modified dA or dG; purple, the 2-OHE1 moiety; and red, the methyl group of the 2-OHE1 moiety. Hydrogen atoms have been deleted for clarity. The corresponding energy, lesion torsion angles, and conformational features are given in Tables 2-5. An excellent, adaptable stereoviewer, Screen-Vu, is available from Berezin Stereo Photography Products (http://www.berezin.com/3d/screen.htm) to aid in viewing stereo images.

tion. Watson-Crick base pairing and stacking were severely disturbed or disrupted at the lesion, and the DNA duplex is bent toward the major groove to shield the adduct from solvent exposure. In families 1 and 2, the methyl face of the 2-OHE1 rings is oriented toward the solvent and the modified adenine is in the syn conformation. The difference between families 1 and 2 comes from the rotation of β′ by ∼180°, resulting in the opposite orientation of the methyl end of the 2-OHE1 moiety. Additionally, rotating R′ by ∼180° from family 1 not only reversed the orientation of the methyl end but also shielded the methyl face of the 2-OHE1 moiety from

solvent, as seen in family 5, a high energy conformational family included to illustrate the stereoisomer effect of the R- and β-linkages (see Discussion). The conformations in families 3 and 4 feature solvent exposure of the methyl edge of the 2-OHE1 rings; the modified adenine is in the anti conformation. The difference between families 3 and 4 stems from the rotation of R′ by ∼180°, again resulting in the opposite orientation of the methyl end of the 2-OHE1 moiety. The number of structures, energy minima with their corresponding energy, χ, R′, β′, methyl end orientation, and solvent exposure of families 1-5 are shown in Table 2. Stereo views of the overall lowest

Conformations of Estrogen Quinone-Derived DNA Adducts

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 315

Figure 3. Three-dimensional and two-dimensional χ, R′, and β′ energy topographies of 2-OHE1-N6-6R-dA up to 20 kcal/mol (∆E). Shaded square in the upper right panel denotes the feasible R′ and β′ ranges within the pol R family polymerase. Torsion angles cycle over 360°. Table 2. Minimum Energy Conformations for 2-OHE1-6r-N6-dAa family b

no. of structures c

energy (kcal/mol)

1 (2A) 2 (S1A)

88 184

-319.6 -316.6

3 (S1B) 4 (S1C)

30 138

-317.9 -314.7

5 (S1D)

*

-298.6

χ (°) 57 (45-67, syn) 50 (44-61, syn) 241 (223-283, anti) 208 (207-259, anti) 64 (syn)

R′ (°)d

β′ (°)d

methyl orientation

solvent exposure

349 (337-359, I) 338 (337-357, I)

210 (202-214, II) 54 (42-56, I)

3′ 5′

methyl face methyl face

346 (303-346, I) 186 (127-193, II)

206 (206-266, II) 196 (192-236, II)

5′ 3′

methyl edge methyl edge

191 (II)

257 (II)

5′

methyl edge

a

Energies and torsion angle values are for the lowest energy variant. Ranges for families are given in parentheses. b Corresponding figure numbers for the families are provided in parentheses. c This designates the number of energy minima up to 20 kcal/mol found for a given family. Families are consolidated from conformationally similar groups, each of which includes structures whose χ, R′, and β′ values are within (15° of the lowest energy variant of that group (See Materials and Methods). d Domains I and II are defined as follows: domain I, 0 ( 90°; domain II, 180 ( 90°. * Not counted for this high energy family.

energy conformation (family 1) are shown in Figure 2A. Figure S1 in the Supporting Information shows stereo views of the lowest energy conformers of other families in Table 2. The distribution of χ, R′, and β′ in energy minima up to 20 kcal/mol is given in Figure 3. No conformations with the 2-OHE1 moiety residing at the minor groove side or intercalated into the base pair stack were located up to 20 kcal/mol. 2-OHE1-6β-N6-dA: Major Groove Conformations. For this stereoisomer of the adenine adduct, 168 groups of energy-minimized conformations were obtained. Eight conformational families were found up to 20 kcal/mol. In the seven lowest energy families, the 2-OHE1 ring system was positioned on the major groove side of the DNA

duplex. As in the R-stereoisomer, the modified adenine could adopt either the anti or the syn conformation; Watson-Crick base pairing and stacking were severely disturbed or disrupted at the lesion, and the DNA duplex is bent toward the major groove to shield the adduct from solvent exposure. The conformations in families 1 and 2 feature solvent exposure of the nonmethyl face of the 2-OHE1 rings and the modified adenine is syn. The difference between families 1 and 2 again results from the rotation of β′ by ∼180°, causing the opposite orientation of the methyl end of the 2-OHE1 moiety. Additionally, rotating R′ by ∼180° from the value in family 1 not only reversed the orientation of the methyl end but also shielded the nonmethyl face of the 2-OHE1 moiety from

316

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

Wang et al.

Figure 4. Three-dimensional and two-dimensional χ, R′, and β′ energy topographies of 2-OHE1-N6-6β-dA up to 20 kcal/mol (∆E). Torsion angles cycle over 360°. Table 3. Minimum Energy Conformations for 2-OHE1-6β-N6-dAa familyb

no. of structuresc

energy (kcal/mol)

1 (S2A) 2 (S2B)

11 2

-311.1 -310.9

3 (S2C) 4 (2B)

28 35

-309.4 -320.9

5 (S2D)

85

-310.4

a-d

R′ (°)d

β′ (°)d

methyl orientation

solvent exposure

342 (331-351, I) 314 (305-314, I)

147 (119-151, II) 356 (356-360, I)

5′ 3′

nonmethyl face nonmethyl face

302 (254-326, I) 144 (112-207, II)

170 (138-180, II) 140 (78-163, II)

3′ 5′

methyl edge methyl edge

133 (106-133, II)

152 (147-167, II)

3′

methyl edge

χ (°) 56 (51-61, syn) 55 (54-55, syn) 275 (213-277, anti) 260 (227-307, anti) 59 (46-60, syn)

Same as in Table 2.

solvent, to produce family 5. In families 3 and 4, the methyl edge of the 2-OHE1 rings is facing the solvent and the modified adenine is anti. The difference between families 3 and 4 comes from the rotation of R′ by ∼180°, resulting in the opposite orientation of the methyl end of the 2-OHE1 moiety. The number of structures, energy minima with their corresponding energy, χ, R′, β′, methyl end orientation, and solvent exposure of families 1-5 are shown in Table 3. Stereo views of the overall lowest energy conformation (family 4) are shown in Figure 2B. Figure S2 in the Supporting Information shows stereo views of the lowest energy conformers of other families in Table 3. The distribution of χ, R′, and β′ in energy minima up to 20 kcal/mol is given in Figure 4. The lowest energy variants with the 2-OHE1 moiety residing at the minor groove side or intercalated into the base pair stack (data not shown) were at least ∼18 kcal/mol higher than

the overall lowest energy minimum (-320.9 kcal/mol). 2-OHE1-6r-N2-dG: Minor Groove Conformations. For this guanine adduct, 168 groups of energy-minimized conformations were obtained from 1212 starting structures. Four conformational families were found up to 20 kcal/mol. In the three lowest energy families, the 2-OHE1 ring system was located in the B-DNA minor groove; the modified guanine remained in the normal B-DNA anti conformation. However, Watson-Crick base pairing and stacking were disturbed or disrupted at the lesion site. In families 1 and 2, the methyl edge of the 2-OHE1 rings is solvent-exposed, while the rest of the 2-OHE1 moiety fits nicely in the minor groove. The difference between families 1 and 2 can be traced to the rotation of R′ by ∼180°, resulting in opposite orientations of the methyl end of the 2-OHE1 moiety. The conformations in families 3 and 4 feature extensive solvent exposure of the methyl

Conformations of Estrogen Quinone-Derived DNA Adducts

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 317

Figure 5. Three-dimensional and two-dimensional χ, R′, and β′ energy topographies of 2-OHE1-N2-6R-dG up to 20 kcal/mol (∆E). Shaded squares in the upper right panel denote the feasible R′ and β′ ranges within the pol R family polymerase. Torsion angles cycle over 360°. Table 4. Minimum Energy Conformations for 2-OHE1-6r-N2-dGa familyb

no. of structuresc

energy (kcal/mol)

χ (°)

R′ (°)d

β′ (°)d

methyl orientation

solvent exposure

1 (2C) 2 (S3A)

86 1

-333.5 -316.9

250 (203-261, anti) 187 (anti)

165 (131-172, II) 319 (I)

215 (201-238, II) 225 (II)

5′ 3′

methyl edge methyl edge

3 (S3B) 4 (S3C)

8 *

-325.8 -299.5

235 (199-235, anti) 213 (anti)

169 (110-169, II) 28 (I)

3′ 5′

methyl face methyl face

a-d,*

58 (58-71, I) 34 (I)

Same as in Table 2.

face of the 2-OHE1 rings. Family 4 is a high energy conformational family included to illustrate the stereoisomer effect of R and β linkages (see Discussion). Because the bulky nonplanar ring system could not fit comfortably in the minor groove, the DNA duplex was severely bent toward the major groove, and the minor groove greatly widened to avoid collisions. The difference between families 3 and 4 can also be traced to the rotation of R′ by ∼180°, resulting in the opposite orientation of the methyl end of the 2-OHE1 moiety. Additionally, rotating β′ by ∼180° from families 1 and 2 not only reversed the orientation of the methyl end, but also exposed the methyl face of the 2-OHE1 moiety to solvent, as seen in family 3 and 4, respectively. In comparison to the adenine adducts, there are fewer structures among the tabulated families, likely reflecting less flexibility in the deep and narrow minor groove than in the shallow

and wide major groove. The number of structures, energy minima with their corresponding energy, χ, R′, β′, methyl end orientation, and solvent exposure of families 1-4 are shown in Table 4. Stereo views of the overall lowest energy conformation (family 1) are shown in Figure 2C. Figure S3 in the Supporting Information shows stereo views of the lowest energy conformers of other families in Table 4. The distribution of χ, R′, and β′ in energy minima up to 20 kcal/mol is given in Figure 5. The lowest energy variants with the 2-OHE1 moiety residing at the major groove side or intercalated in the base pair stack (data not shown) were at least ∼19 kcal/mol higher than the lowest energy minimum (-333.5 kcal/mol). 2-OHE1-6β-N2-dG: Minor Groove Conformations. For this stereoisomer, 177 groups of energy-minimized conformations were obtained. Seven conformational families were found with energy below 20 kcal/mol. In the

318

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

Wang et al.

Figure 6. Three-dimensional and two-dimensional χ, R′, and β′ energy topographies of 2-OHE1-N2-6β-dG up to 20 kcal/mol (∆E). Torsion angles cycle over 360°. Table 5. Minimum Energy Conformations for 2-OHE1-6β-N2-dGa family b

no. of structuresc

energy (kcal/mol)

χ (°)

R′ (°)d

β′ (°)d

methyl orientation

1 (S4A) 2 (2D)

51 33

-321.1 -322.1

246 (193-262, anti) 242 (210-252, anti)

157 (110-176, II) 296 (283-320, I)

152 (131-203, II) 165 (132-181, II)

3′ 5′

methyl edge methyl edge

3 (S4B) 4 (S4C)

6 4

-311.4 -310.2

225 (206-229, anti) 200 (167-245, anti)

113 (84-129, II) 277 (254-281, I)

5′ 3′

nonmethyl face nonmethyl face

a-d

5 (4-9, I) 33 (25-47, I)

solvent exposure

Same as in Table 2.

four lowest energy families, the 2-OHE1 ring system was positioned on the minor groove side of the DNA duplex; the modified guanine remained in the normal B-DNA anti conformation. However, Watson-Crick base pairing and stacking were disturbed or disrupted at the lesion. In families 1 and 2, the methyl edge of the 2-OHE1 rings is facing the solvent, while the rest of the 2-OHE1 moiety fits nicely in the minor groove, as in the case for the R-stereoisomer. Again, the difference between families 1 and 2 can be traced to the rotation of R′ by ∼180°, resulting in the opposite orientation of the methyl end of the 2-OHE1 moiety. The conformations in families 3 and 4 feature extensive exposure of the nonmethyl face of the 2-OHE1 rings. As in the R-isomer, the DNA duplex was severely bent toward the major groove, and the minor groove greatly widened to avoid collisions, because the nonplanar ring system could not fit well into the minor groove. The difference between families 3 and 4 can also be traced to the rotation of R′ by ∼180°, resulting

in the opposite orientation of the methyl end of the 2-OHE1 moiety. Additionally, rotating β′ by ∼180° from families 1 and 2 not only reversed the orientation of the methyl end but also exposed the nonmethyl face of the 2-OHE1 moiety to solvent, as seen in family 3 and 4, respectively. Again, there are fewer structures among the tabulated families as compared to the adenine adducts, likely reflecting less flexibility in the minor groove. The number of structures, energy minima with their corresponding energy, χ, R′, β′, methyl end orientation, and solvent exposure of families 1-4 are shown in Table 5. Stereo views of the overall lowest energy conformation (family 2) are shown in Figure 2D. Figure S4 in the Supporting Information shows stereo views of the lowest energy conformers of other families in Table 5. The distribution of χ, R′, and β′ in energy minima up to 20 kcal/mol is given in Figure 6. The lowest energy variants with the 2-OHE1 moiety residing at the major groove side or intercalated into the base pair stack (data not shown)

Conformations of Estrogen Quinone-Derived DNA Adducts

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 319

Figure 7. Stereo views of feasible conformations for (A) 2-OHE1-6R-N6-dA and (B and C) 2-OHE1-6R-N2-dG at the primer-template junction within the pol R family polymerase. Color code: silver, primer strand; cyan, template strand; pink, pol R family RB69 DNA polymerase; brown, catalytic Mg2+; purple, 2-OHE1 moiety; red, methyl group on 2-OHE1; green, incoming dTTP or dCTP; and orange, modified dA or dG.

were at least ∼14 kcal/mol higher than the overall lowest energy minimum (-322.1 kcal/mol). The 2-OHE1 Moiety in the r-Stereoisomer Adducts Can Reside on the Major Groove Side of the Template within a Pol r Family DNA Polymerase Ternary Complex. In the next stage of the work, our goal was to investigate molecular models of these adducts within the active site of a pol R family polymerase, in an

effort to rationalize the primer extension results of Terashima et al. (49). We began the study with the crystal structure of RB69 ternary complex (53), a pol R family polymerase, in the closed conformation needed for catalysis. We remodeled the base sequence, docked each of the stereoisomers to the templating adenine, and searched the R′ and β′ conformational space for structures with minimal collision, creating 1296 different structures

320

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

Wang et al.

Figure 8. Stereo views into the active site of feasible conformations for (A) 2-OHE1-6R-N6-dA and (B and C) 2-OHE1-6R-N2-dG at the primer-template junction within the pol R family polymerase. Color code: same as in Figure 7.

for each case (see Materials and Methods). In the case of incoming dTTP, it was found that only a very narrow range of R′ and β′ could be adopted with minimal close contact, R′, ∼202 ( 10°; β′, ∼168 ( 10° (Figure 3, shaded square) for 2-OHE1-6R-N6-dA (Figures 7A and 8A; Figure S5, Supporting Information). The 2-OHE1 moiety is situated on the major groove side of the templating modified adenine in an open pocket of the polymerase. The methyl edge of the 2-OHE1 moiety is pointing outward from the DNA, while the methyl face of the rings is facing 5′ of the modified dA. Adjustments in the torsion angles CR-Cβ-Cγ-Cδ1 in Phe 282 of the polymerase, and γ′ (Figure 1A) of 2-OHE1 to 229° (from 274°) and to 300° (from 69°), respectively, were needed. Watson-Crick base pairing between the modified adenine and the incoming dTTP is maintained (Figure S8A, Supporting Information). In the case of incoming dCTP, somewhat distorted A-C wobble pairing (between N6 of the modified adenine and N3 of dCTP and between protonated N1 of the modified adenine and O2 of dCTP) is feasible (Figure S8B, Supporting Information). No conformation without serious collisions was found for 2-OHE1-6β-N6-dA within the polymerase. The minor groove side of the primer-template junction is crowded with critical protein-DNA interactions required for polymerase fidelity (67) and cannot accommodate a bulky ring system like that of the 2-OHE1. Thus, it is impossible to dock either of the stereoisomers to the minor groove side of the templating guanine without large rearrangements of protein residues in the active site while retaining the normal guanine anti conformation. However, a syn guanine orientation, plac-

ing the bulky 2-OHE1 ring system into the major groove, is feasible. To model the syn guanine conformation, the glycosidic torsion angle of the modified guanine was rotated to 14°, and the torsion angles of the neighboring template residues were slightly adjusted (Table S4, Supporting Information) to achieve coplanarity of the modified guanine and incoming dCTP and to minimize collision between the modified base and surrounding residues, while maintaining the neighboring template residues close to their crystal positions. Similar searches as for the adenine adducts were carried out for 2-OHE16R-N2-dG and 2-OHE1-6β-N2-dG. Two conformations were found with minimal collision: R′, ∼280 ( 10°; β′, ∼70 ( 10°; and R′, ∼270 ( 15°; β′, ∼160 ( 15° (Figure 5, shaded squares) for 2-OHE1-6R-N2-dG (Figures 7B,C and 8B,C; Figures S6 and S7, Supporting Information). In both cases, the methyl edge of the 2-OHE1 moiety is pointing outward from the DNA, with the methyl face of the rings facing the template strand. A hydrogen bond is formed between O6 of the modified guanine and N4 of the incoming dCTP. In addition, a second hydrogen bond would be possible between N7 of the modified guanine and protonated N3 of the incoming dCTP with modest shifts of the bases (Figure S8C, Supporting Information). No conformation without serious collisions was found for 2-OHE1-6β-N2-dG within this polymerase.

Discussion Helix Exterior Positions Are Preferred in All Adducts. Our conformational searches for the modified duplex revealed that the 2-OHE1 moiety strongly favors

Conformations of Estrogen Quinone-Derived DNA Adducts

residing at the exterior of the DNA duplex. An examination of the adduct structure provides a plausible explanation for this. Rings B, C, D, and the methyl group on C13 of the 2-OHE1 moiety (Figure 1A) are all nonplanar. This nonplanarity makes it unfavorable for the adducts to adopt an intercalated conformation, as accommodation of the nonplanar ring systems would require large distortion of the DNA duplex. The apparent proclivity for such bulky, nonplanar adducts to reside at the DNA helix exterior, irrespective of stereochemistry or base modified, is thus a plausible conformational feature of other structurally similar estrogen quinone-derived DNA adducts. Moreover, this conformational preference would unlikely be affected by base sequence context. Because of the multiple minimum problem, one can never be certain that all possible conformations have been located. However, in this work, we created a number of different intercalated starting structures (See Materials and Methods), but in every case, the obtained energy minimum either thrust the 2-OHE1 moiety to the helix exterior or was at least 40 kcal/mol above the overall energy minimum. Stereoisomeric Effects: Relationship between Adduct Orientation and Linkage Geometry. The conformations of the low energy 2-OHE1 adducts in the DNA duplex display a clear difference in the R- and β-stereoisomers. Tables 2-5 show that χ, R′, and β′ mainly adopt one of two domains, anti or syn for χ and domain I (0 ( 90°) or domain II (180 ( 90°) for R′ and β′, in the low energy families. The 2-OHE1 moiety adopts opposite orientations (in terms of whether the methyl end is at the 5′ or 3′ side of the modified base) in the R- and β-stereoisomers of the guanine and adenine adducts, when the corresponding χ, R′, and β′ values are in the same domain. Domain change in either one or all of the conformational determinants, χ, R′, and β′, reverses the orientation of the methyl end, while domain change in two of the determinants retains the orientation of the methyl end. The solvent exposure of the 2-OHE1 moiety also showed a correlation to the stereochemistry at the adduct linkage site. For the adenine adducts, the methyl edge of the 2-OHE1 moiety (for both stereoisomers) is solventexposed in families 3, 4, and 5, while the methyl face (for the R-stereoisomer) or nonmethyl face (for the β-stereoisomer) is exposed in families 1 and 2 (Tables 2 and 3; Figures 2A,B, S1, and S2). A similar correlation has been observed for the guanine adducts: the methyl edge of the 2-OHE1 moiety (for both stereoisomers) is solventexposed in families 1 and 2, while the methyl face (for the R-stereoisomer) or nonmethyl face (for the β-stereoisomer) is exposed in families 3 and 4 (Tables 4 and 5; Figures 2C,D, S3, and S4). The difference in solvent exposure between the R- and the β-stereoisomers is derived from the difference in the intrinsic connection at the linkage site. An energy difference of ∼11 kcal/mol is observed between the overall lowest energy variant of 2-OHE1-6RN2-dG and that of 2-OHE1-6β-N2-dG, favoring the R-stereoisomer (Tables 4 and 5). It can be seen that the 2-OHE1 moiety more closely contacts the unmodified strand in the lowest energy variant of the β-isomer than in that of the R-isomer, due to the intrinsic difference in the linkage connection at the lesion site. This produces a covalent constraint to accommodate the β-isomer in the minor groove and requires groove distortion to avoid

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 321 Table 6. Minor Groove Widthsa lowest energy variant of 2-OHE1-N2-dG adducts

width (Å)

R-stereoisomer β-stereoisomer

6.83, 7.29, 7.03 6.87, 8.29, 8.79

aPhosphate distances P12 to P4, P11 to P5, and P10 to P6 less 5.8 Å (to account for the phosphate group diameter in the groove), for the phosphate sequence: 5′ - P1 - P2 - P3 - P4 - P5 - P6 - 3′ 3′ - P12 - P11 - P10 - P9 - P8 - P7 - 5′.

collision. Indeed, the minor groove is significantly wider in the β-stereoisomer, as may be seen from Figure 2 and Table 6 (68). Analysis of energy components also supports the poorer accommodation of the β-stereoisomer in the minor groove. The energy from DNA-DNA interactions is ∼9.5 kcal/mol lower in the R-isomer than in the β-isomer, while the energy from DNA-[2-OHE1] and [2-OHE1]-[2-OHE1] interactions is not significantly different. The difference stems almost entirely from the van der Waals and torsional terms in the force field. Relationship of Duplex Structures to DNA Repair. The computed solution structures may be considered in relation to our current understanding of DNA repair. Such bulky adducts would be expected to be susceptible to NER (69, 70), which is now understood to recognize such lesions through the distortion that they impose on the DNA, as well as the chemical modification (71). Helix destabilization through disturbed WatsonCrick base pairing is one key factor that the NER machinery recognizes (72-74). In this connection, we note that the adenine adduct isomers, located in the major groove, manifest serious disturbance to the Watson-Crick pairs at and near the lesion site, suggesting significant repair susceptibility. NER excision assays in human cells (75) have shown that 1R(+)- and 1S(-)trans-anti-[BcPh]-N6-dA adducts are not repaired and this appears to relate to the fact that the Watson-Crick hydrogen bonds are not ruptured and also to the neatly intercalated, well-stacked conformation adopted by these nonplanar adducts whose topology permits a good fit into the DNA duplex (76). On the other hand, 10S(+)- and 10R(-)-trans-anti-[BP]-N6-dA adducts, also intercalated but with greater distortion, are susceptible to NER in this assay (77). In our current study, the guanine adducts situated neatly in the minor groove with slight WatsonCrick disturbance and less exposure of the bulk would appear to be less repair-susceptible. These resemble the minor groove 10S(+)- and 10R(-)-trans-anti-[BP]-N2-dG adducts, which have been shown to be modestly susceptible to NER excision in human cell NER extracts (78). Models of the Adducts within Pol r Can Rationalize the Experimental Primer Extension Data and Suggest Blockage by the β Stereoisomers. Our conformational study of the 2-OHE1-6(R,β)-N6-dA and 2-OHE1-6(R,β)-N2-dG within the pol R family polymerase provides a structural rationale for the primer extension assays of Terashima et al. (49) for the same adducts of unspecified R or β stereochemistry. The significant incorporation of dT opposite the adenine adducts (22.7%), together with a residual amount of the incorrect base, dC (0.54%), is consistent with our models, which place the 2-OHE1 ring system in an open pocket of the polymerase on the major groove side of the modified template without disruption of polymerase-DNA interactions, while maintaining the crystal geometry of the

322

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

active site in this catalytically competent closed polymerase. Watson-Crick base pairing between the modified adenine and the incoming dTTP is maintained in the model (Figures 7A and 8A; Figure S5 in the Supporting Information provides a larger nonstereo view of the ternary complex), and the distance between the R-phosphate of the incoming dTTP and the modeled 3′-O of the primer terminus is ∼3.2 Å, close to the range for reaction. The covalent linkage of the 2-OHE1 at N6 distorts but does not rupture the Watson-Crick hydrogen bond between N6 of the modified adenine and O4 of the incoming dTTP, which could affect the efficiency of incorporation of dTTP. On the other hand, an incoming dCTP can be incorporated opposite the modified adenine through wobble A-C pairing (79), possibly causing the misincorporation of dC in the primer extension reaction (49). In addition, the polymerase structure shows why 2-OHE1-6(R,β)-N2-dG is much more prone to block pol R, since this polymerase, like other polymerases, has close interactions with the DNA duplex region on the minor groove side of the template, which are needed for polymerase fidelity (67). The guanine adducts consequently cannot be placed in its normally preferred position with guanine anti on the minor groove side of the template without major disruption of these critical polymeraseDNA interactions. However, rotation of the guanine to the less favored syn domain would place the estrogen bulk on the major groove side, which could explain the ∼6% incorporation of the correct base dCTP opposite the modified guanine in primer extension reactions catalyzed by pol R (49). The searches with DUPLEX showed that the anti conformation is favored over syn by ∼20 kcal/ mol in the modified duplexes; however, the polymerase better accommodates the syn conformation. Modeling studies together with molecular dynamics simulation have previously found that the (+)-trans-anti-[BP]-N2dG adduct can be similarly accommodated within the replicative T7 DNA polymerase in the closed conformation, when the modified guanine is in the syn conformation (80). In the current modeling study, the syn-oriented modified guanine retains its coplanarity with the incoming dCTP, and a hydrogen bond is formed between O6 of the modified guanine and N4 of the incoming dCTP. A second hydrogen bond may be possible between N7 of the modified guanine and N3 of the incoming dCTP, if cytosine is protonated at the N3 position. This hydrogen bonding scheme has been observed in crystal structures (81, 82). Because the conformation of the 2-OHE1 moiety within the polymerase is constrained by the partially formed DNA duplex and by the walls of the polymerase pocket on the major groove side, the conformational preference at the linkage site of the adenine or guanine adducts might well differ from that in the fully formed DNA duplex in solution. We find that for the R-isomer of the adenine adducts, the feasible χ, R′, and β′ values in the polymerase model are in the vicinity of those of family 4 for the duplex (Figure 3, upper right panel). On the other hand, the feasible χ, R′, and β′ values in the polymerase model for the R-isomer of the guanine adducts are not near the energy minima located for the duplexes (Figure 5, upper right panel), due to the fact that the guanine adducts prefer to reside in the minor groove of the DNA duplex, where the DNA interacts closely with the protein in the polymerase model. Thus, the modified guanine had

Wang et al.

to adopt the less common syn conformation, placing the 2-OHE1 bulk on the major groove side, to be accommodated. The conformational searches within the pol R family polymerase also showed that no conformation without serious collisions was feasible for either β-stereoisomer. The β-linkage places the methyl face of the 2-OHE1 rings closer to the DNA and consequently causes more serious and more frequent collisions between the 2-OHE1 moiety and the DNA than the R-linkage. This implies that the β-stereoisomers may act as blocks in primer extension (49) and that the observed extension is due to the R-stereoisomer. While this polymerase modeling study did not allow protein flexibility through molecular dynamics, it is clear from our work that the β-stereoisomers would require significant protein distortion if they were to be accommodated in the polymerase active site region.

Conclusions Our conformational searches for modified duplexes containing estrogen quinone-derived adducts 2-OHE16(R,β)-N6-dA or 2-OHE1-6(R,β)-N2-dG show that the adenine adducts favor the major groove side of the B-DNA duplex, while the guanine adducts prefer to reside on the minor groove side, with stereochemistrydependent orientational differences in each case. Intercalation is disfavored due to the nonplanar, mainly nonaromatic nature of the 2-OHE1 ring system. The preference for the helix exterior is thus a plausible feature of other estrogen quinone-derived adducts with similar nonplanar structures. Modeling studies in a pol R family polymerase rationalize the observed in vitro primer extension studies, which revealed predominant blockage by the guanine adducts and misincorporation of dCTP opposite the adenine adducts. The modeling studies also indicated poor accommodation of the β-stereoisomers within the polymerase, suggesting that they would be more prone to cause blockage. Different orientations of R- and β-stereoisomers in both adenine and guanine adducts could be expected to produce different biochemical effects, as has been observed in stereochemistry-dependent differential treatment of adducts derived from polycyclic aromatic hydrocarbons (71, 83-85).

Acknowledgment. This research was supported by NIH Grants CA75449 and CA28038 to S.B. Computations were carried out at the DOE National Energy Research Supercomputer Center. Visualization, modeling, and data analysis were carried out on our own SGI workstations. Supporting Information Available: Partial charges employed for 2-OHE1-6(R,β)-N6-dA and 2-OHE1-6(R,β)-N2-dG, torsion energy contributions involving the 2-OHE1 moiety, torsion angle modifications in modeling the adducts within the pol R family polymerase, and structures of additional families in Tables 2-5. This material is available free of charge via the Internet at http://pubs.acs.org.

References (1) Henderson, B. E., Ross, R., and Bernstein, L. (1988) Estrogens as a cause of human cancer: the Richard and Hinda Rosenthal Foundation award lecture. Cancer Res. 48, 246-253. (2) Feigelson, H. S., and Henderson, B. E. (1996) Estrogens and breast cancer. Carcinogenesis 17, 2279-2284. (3) Liehr, J. G. (1990) Genotoxic effects of estrogens. Mutat. Res. 238, 269-276.

Conformations of Estrogen Quinone-Derived DNA Adducts (4) Weisz, A., Cicatiello, L., Persico, E., Scalona, M., and Bresciani, F. (1990) Estrogen stimulates transcription of c-jun protooncogene. Mol. Endocrinol. 4, 1041-1050. (5) Kumar, V., Green, S., Stack, G., Berry, M., Jin, J. R., and Chambon, P. (1987) Functional domains of the human estrogen receptor. Cell 51, 941-951. (6) Evans, R. M. (1988) The steroid and thyroid hormone receptor superfamily. Science 240, 889-895. (7) Nandi, S., Guzman, R. C., and Yang, J. (1995) Hormones and mammary carcinogenesis in mice, rats, and humans: a unifying hypothesis. Proc. Natl. Acad. Sci. U.S.A. 92, 3650-3657. (8) Li, J. J., and Li, S. A. (1996) The effects of hormones on tumor induction. In Chemical Induction of Cancer (Arcos, J. G., Argus, M. F., and Woo, Y. T., Eds.) pp 396-449, Birkhuaser, Boston. (9) Alberg, A. J., Visvanathan, K., and Helzlsouer, K. J. (1998) Epidemiology, prevention, and early detection of breast cancer. Curr. Opin. Oncol. 10, 492-497. (10) Li, J. J., and Li, S. A. (1990) Estrogen carcinogenesis in hamster tissues: a critical review. Endocr. Rev. 11, 524-531. (11) Grady, D., and Ernster, V. L. (1996) Endometrial cancer. In Cancer Epidemiology and Prevention (Schottenfeld, D., and Fraumeni, J. F., Jr., Eds.) pp 1058-1089, Oxford University Press, Oxford. (12) Liehr, J. G. (1983) 2-Fluoroestradiol. Separation of estrogenicity from carcinogenicity. Mol. Pharmacol. 23, 278-281. (13) Li, J. J., and Li, S. A. (1984) Estrogen-induced tumorigenesis in hamsters: roles for hormonal and carcinogenic activities. Arch. Toxicol. 55, 110-118. (14) Yager, J. D., and Liehr, J. G. (1996) Molecular mechanisms of estrogen carcinogenesis. Annu. Rev. Pharmacol. Toxicol. 36, 203232. (15) Zhu, B. T., and Conney, A. H. (1998) Functional role of estrogen metabolism in target cells: review and perspectives. Carcinogenesis 19, 1-27. (16) Li, J. J., and Li, S. A. (1996) Hormonal Carcinogenesis II, SpringVerlag, New York. (17) Cavalieri, E. L., Li, K. M., Balu, N., Saeed, M., Devanesan, P., Higginbotham, S., Zhao, J., Gross, M. L., and Rogan, E. G. (2002) Catechol ortho-quinones: the electrophilic compounds that form depurinating DNA adducts and could initiate cancer and other diseases. Carcinogenesis 23, 1071-1077. (18) Cavalieri, E. L., Rogan, E. G., and Chakravarti, D. (2002) Initiation of cancer and other diseases by catechol ortho-quinones: a unifying mechanism. Cell Mol. Life Sci. 59, 665-681. (19) Bolton, J. L., and Chang, M. (2001) Quinoids as reactive intermediates in estrogen carcinogenesis. Adv. Exp. Med. Biol. 500, 497-507. (20) Cavalieri, E., Frenkel, K., Liehr, J. G., Rogan, E., and Roy, D. (2000) Estrogens as endogenous genotoxic agentssDNA adducts and mutations. J. Natl. Cancer Inst. Monogr., 75-93. (21) Liehr, J. G. (2000) Role of DNA adducts in hormonal carcinogenesis. Regul. Toxicol. Pharmacol. 32, 276-282. (22) Beroud, C., and Soussi, T. (2003) The UMD-p53 database: new mutations and analysis tools. Hum. Mutat. 21, 176-181. (23) Soussi, T. (2003) Focus on the p53 gene and cancer: advances in TP53 mutation research. Hum. Mutat. 21, 173-175. (24) Soussi, T., Dehouche, K., and Beroud, C. (2000) p53 website and analysis of p53 gene mutations in human cancer: forging a link between epidemiology and carcinogenesis. Hum. Mutat. 15, 105113. (25) Olivier, M., Eeles, R., Hollstein, M., Khan, M. A., Harris, C. C., and Hainaut, P. (2002) The IARC TP53 database: new online mutation analysis and recommendations to users. Hum. Mutat. 19, 607-614. (26) Conti, C. J. (1992) Mutations of genes of the ras family in human and experimental tumors. Prog. Clin. Biol. Res. 376, 357-378. (27) Bos, J. L. (1989) ras oncogenes in human cancer: a review. Cancer Res. 49, 4682-4689. (28) Coles, C., Condie, A., Chetty, U., Steel, C. M., Evans, H. J., and Prosser, J. (1992) p53 mutations in breast cancer. Cancer Res. 52, 5291-5298. (29) Fabian, C. J., Kimler, B. F., Zalles, C. M., Klemp, J. R., Kamel, S., Zeiger, S., and Mayo, M. S. (2000) Short-term breast cancer prediction by random periareolar fine-needle aspiration cytology and the Gail risk model. J. Natl. Cancer Inst. 92, 1217-1227. (30) Lasky, T., and Silbergeld, E. (1996) P53 mutations associated with breast, colorectal, liver, lung, and ovarian cancers. Environ. Health Perspect. 104, 1324-1331. (31) Sivertsson, A., Platz, A., Hansson, J., and Lundeberg, J. (2002) Pyrosequencing as an alternative to single-strand conformation polymorphism analysis for detection of N-ras mutations in human melanoma metastases. Clin. Chem. 48, 2164-2170.

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 323 (32) van Elsas, A., Scheibenbogen, C., van der Minne, C., Zerp, S. F., Keilholz, U., and Schrier, P. I. (1997) UV-induced N-ras mutations are T-cell targets in human melanoma. Melanoma Res. 7 (Suppl. 2), S107-S113. (33) Corradini, P., Ladetto, M., Inghirami, G., Boccadoro, M., and Pileri, A. (1994) N- and K-ras oncogenes in plasma cell dyscrasias. Leuk. Lymphoma 15, 17-20. (34) Rodenhuis, S. (1992) ras and human tumors. Semin. Cancer Biol. 3, 241-247. (35) Leon, J., and Pellicer, A. (1993) ras genes involvement in carcinogenesis: lessons from animal model systems. In The Ras Superfamily of GTPases (Lacal, J. C., and McCormick, F., Eds.) pp 1-36, CRC Press, Boca Raton, FL. (36) Martucci, C. P., and Fishman, J. (1993) P450 enzymes of estrogen metabolism. Pharmacol. Ther. 57, 237-257. (37) Liehr, J. G., Roy, D., Ari-Ulubelen, A., Bui, Q. D., Weisz, J., and Strobel, H. W. (1990) Effect of chronic estrogen treatment of Syrian hamsters on microsomal enzymes mediating formation of catecholestrogens and their redox cycling: implications for carcinogenesis. J. Steroid Biochem. 35, 555-560. (38) Kalyanaraman, B., Sealy, R. C., and Sivarajah, K. (1984) An electron spin resonance study of o-semiquinones formed during the enzymatic and autoxidation of catechol estrogens. J. Biol. Chem. 259, 14018-14022. (39) Dwivedy, I., Devanesan, P., Cremonesi, P., Rogan, E., and Cavalieri, E. (1992) Synthesis and characterization of estrogen 2,3- and 3,4-quinones. Comparison of DNA adducts formed by the quinones versus horseradish peroxidase-activated catechol estrogens. Chem. Res. Toxicol. 5, 828-833. (40) Liehr, J. G., Avitts, T. A., Randerath, E., and Randerath, K. (1986) Estrogen-induced endogenous DNA adduction: possible mechanism of hormonal cancer. Proc. Natl. Acad. Sci. U.S.A. 83, 53015305. (41) Tsutsui, T., Tamura, Y., Yagi, E., and Barrett, J. C. (2000) Involvement of genotoxic effects in the initiation of estrogeninduced cellular transformation: studies using Syrian hamster embryo cells treated with 17β-estradiol and eight of its metabolites. Int. J. Cancer 86, 8-14. (42) Yagi, E., Barrett, J. C., and Tsutsui, T. (2001) The ability of four catechol estrogens of 17β-estradiol and estrone to induce DNA adducts in Syrian hamster embryo fibroblasts. Carcinogenesis 22, 1505-1510. (43) Hayashi, N., Hasegawa, K., Komine, A., Tanaka, Y., McLachian, J. A., Barrett, J. C., and Tsutsui, T. (1996) Estrogen-induced cell transformation and DNA adduct formation in cultured Syrian hamster embryo cells. Mol. Carcinog. 16, 149-156. (44) Cavalieri, E. L., Stack, D. E., Devanesan, P. D., Todorovic, R., Dwivedy, I., Higginbotham, S., Johansson, S. L., Patil, K. D., Gross, M. L., Gooden, J. K., Ramanathan, R., Cerny, R. L., and Rogan, E. G. (1997) Molecular origin of cancer: catechol estrogen3,4-quinones as endogenous tumor initiators. Proc. Natl. Acad. Sci. U.S.A. 94, 10937-10942. (45) Liehr, J. G., Fang, W. F., Sirbasku, D. A., and Ari-Ulubelen, A. (1986) Carcinogenicity of catechol estrogens in Syrian hamsters. J. Steroid Biochem. 24, 353-356. (46) Li, J. J., and Li, S. A. (1987) Estrogen carcinogenesis in Syrian hamster tissues: role of metabolism. Fed. Proc. 46, 1858-1863. (47) Newbold, R. R., and Liehr, J. G. (2000) Induction of uterine adenocarcinoma in CD-1 mice by catechol estrogens. Cancer Res. 60, 235-237. (48) Stack, D. E., Byun, J., Gross, M. L., Rogan, E. G., and Cavalieri, E. L. (1996) Molecular characteristics of catechol estrogen quinones in reactions with deoxyribonucleosides. Chem. Res. Toxicol. 9, 851-859. (49) Terashima, I., Suzuki, N., Dasaradhi, L., Tan, C. K., Downey, K. M., and Shibutani, S. (1998) Translesional synthesis on DNA templates containing an estrogen quinone-derived adduct: N2(2-hydroxyestron-6-yl)-2′-deoxyguanosine and N6-(2-hydroxyestron6-yl)-2′-deoxyadenosine. Biochemistry 37, 13807-13815. (50) Terashima, I., Suzuki, N., and Shibutani, S. (2001) Mutagenic properties of estrogen quinone-derived DNA adducts in simian kidney cells. Biochemistry 40, 166-172. (51) Chen, J. X., Zheng, Y., West, M., and Tang, M. S. (1998) Carcinogens preferentially bind at methylated CpG in the p53 mutational hot spots. Cancer Res. 58, 2070-2075. (52) Wang, L., Hingerty, B. E., Srinivasan, A. R., Olson, W. K., and Broyde, S. (2002) Accurate representation of B-DNA double helical structure with implicit solvent and counterions. Biophys. J. 83, 382-406. (53) Franklin, M. C., Wang, J., and Steitz, T. A. (2001) Structure of the replicating complex of a pol R family DNA polymerase. Cell 105, 657-667.

324

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

(54) Allen, F. H. (2002) The Cambridge Structural Database: a quarter of a million crystal structures and rising. Acta Crystallogr. B. 58 (Pt. 3, Pt. 1), 380-388. (55) Scott, R. A., and Scheraga, H. A. (1965) Method for calculating internal rotation barriers. J. Chem. Phys. 42, 2209-2215. (56) Frisch, M. J., Trucks, G. W., Schlegel, H. B., Gill, P. M. W., Johnson, B. G., Robb, M. A., Cheeseman, J. R., Keith, T., Petersson, G. A., Montgomery, J. A., Raghavachari, K., Al-Laham, M. A., Zakrzewski, V. G., Ortiz, J. V., Foresman, J. B., Cioslowski, J., Stefanov, B. B., Nanayakkara, A., Challacombe, M., Peng, C. Y., Ayala, P. Y., Chen, W., Wong, M. W., Andres, J. L., Replogle, E. S., Gomperts, R., Martin, R. L., Fox, D. J., Binkley, J. S., Defrees, D. J., Baker, J., Stewart, J. P., Head-Gordon, M., Gonzalez, C., and Pople, J. A. (1995) Gaussian 94, Revision E.1, Gaussian, Inc., Pittsburgh, PA. (57) Shapiro, R., Hingerty, B. E., and Broyde, S. (1989) Minor-groove binding models for acetylaminofluorene modified DNA. J. Biomol. Struct. Dyn. 7, 493-513. (58) Hingerty, B. E., Figueroa, S., Hayden, T. L., and Broyde, S. (1989) Prediction of DNA structure from sequence: a build-up technique. Biopolymers 28, 1195-1222. (59) Arnott, S., Campbell-Smith, P. J., and Chandrasekaran, R. (1976) Atomic coordinates and molecular conformations for DNA-DNA, RNA-RNA, and DNA-RNA helices. In CRC Handbook of Biochemistry and Molecular Biology (Fasman, G., Ed.) Vol. 2, pp 411-422, CRC Press, Cleveland. (60) Shapiro, R., Ellis, S., Hingerty, B. E., and Broyde, S. (1995) Major and minor groove conformations of DNA trimers modified on guanine or adenine by 4-aminobiphenyl: adenine adducts favor the minor groove. Chem. Res. Toxicol. 8, 117-127. (61) Cosman, M., Laryea, A., Fiala, R., Hingerty, B. E., Amin, S., Geacintov, N. E., Broyde, S., and Patel, D. J. (1995) Solution conformation of the (-)-trans-anti-benzo[c]phenanthrene-dA ([BPh]dA) adduct opposite dT in a DNA duplex: intercalation of the covalently attached benzo[c]phenanthrenyl ring to the 3′-side of the adduct site and comparison with the (+)-trans-anti-[BPh]dA opposite dT stereoisomer. Biochemistry 34, 1295-1307. (62) Cosman, M., Fiala, R., Hingerty, B. E., Laryea, A., Lee, H., Harvey, R. G., Amin, S., Geacintov, N. E., Broyde, S., and Patel, D. (1993) Solution conformation of the (+)-trans-anti-[BPh]dA adduct opposite dT in a DNA duplex: intercalation of the covalently attached benzo[c]phenanthrene to the 5′-side of the adduct site without disruption of the modified base pair. Biochemistry 32, 12488-12497. (63) Cosman, M., Hingerty, B. E., Luneva, N., Amin, S., Geacintov, N. E., Broyde, S., and Patel, D. J. (1996) Solution conformation of the (-)-cis-anti-benzo[a]pyrenyl-dG adduct opposite dC in a DNA duplex: intercalation of the covalently attached BP ring into the helix with base displacement of the modified deoxyguanosine into the major groove. Biochemistry 35, 9850-9863. (64) Cosman, M., de los Santos, C., Fiala, R., Hingerty, B. E., Ibanez, V., Luna, E., Harvey, R., Geacintov, N. E., Broyde, S., and Patel, D. J. (1993) Solution conformation of the (+)-cis-anti-[BP]dG adduct in a DNA duplex: intercalation of the covalently attached benzo[a]pyrenyl ring into the helix and displacement of the modified deoxyguanosine. Biochemistry 32, 4145-4155. (65) Lin, C. H., Huang, X., Kolbanovskii, A., Hingerty, B. E., Amin, S., Broyde, S., Geacintov, N. E., and Patel, D. J. (2001) Molecular topology of polycyclic aromatic carcinogens determines DNA adduct conformation: a link to tumorigenic activity. J. Mol. Biol. 306, 1059-1080. (66) Berman, H. M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N., and Bourne, P. E. (2000) The Protein Data Bank. Nucleic Acids Res. 28, 235-242. (67) Morales, J., and Kool, E. (1999) Minor groove interactions between polymerase and DNA: More essential to replication than Watson-Crick hydrogen bonds. J. Am. Chem. Soc. 121, 2323-2324. (68) Fratini, A. V., Kopka, M. L., Drew, H. R., and Dickerson, R. E. (1982) Reversible bending and helix geometry in a B-DNA

Wang et al.

(69)

(70) (71)

(72)

(73) (74) (75)

(76)

(77)

(78)

(79)

(80)

(81) (82)

(83)

(84)

(85)

dodecamer: CGCGAATTBrCGCG. J. Biol. Chem. 257, 1468614707. Cline, S. D., and Hanawalt, P. C. (2003) Who’s on first in the cellular response to DNA damage? Nat. Rev. Mol. Cell Biol. 4, 361-372. Friedberg, E. C. (2001) How nucleotide excision repair protects against cancer. Nat. Rev. Cancer 1, 22-33. Geacintov, N. E., Broyde, S., Buterin, T., Naegeli, H., Wu, M., Yan, S., and Patel, D. J. (2002) Thermodynamic and structural factors in the removal of bulky DNA adducts by the nucleotide excision repair machinery. Biopolymers 65, 202-210. Sugasawa, K., Okamoto, T., Shimizu, Y., Masutani, C., Iwai, S., and Hanaoka, F. (2001) A multistep damage recognition mechanism for global genomic nucleotide excision repair. Genes Dev. 15, 507-521. Batty, D. P., and Wood, R. D. (2000) Damage recognition in nucleotide excision repair of DNA. Gene 241, 193-204. Wood, R. D. (1999) DNA damage recognition during nucleotide excision repair in mammalian cells. Biochimie 81, 39-44. Buterin, T., Hess, M. T., Luneva, N., Geacintov, N. E., Amin, S., Kroth, H., Seidel, A., and Naegeli, H. (2000) Unrepaired fjord region polycyclic aromatic hydrocarbon-DNA adducts in ras codon 61 mutational hot spots. Cancer Res. 60, 1849-1856. Wu, M., Yan, S., Patel, D. J., Geacintov, N. E., and Broyde, S. (2002) Relating repair susceptibility of carcinogen-damaged DNA with structural distortion and thermodynamic stability. Nucleic Acids Res. 30, 3422-3432. Yan, S., Shapiro, R., Geacintov, N. E., and Broyde, S. (2001) Stereochemical, structural, and thermodynamic origins of stability differences between stereoisomeric benzo[a]pyrene diol epoxide deoxyadenosine adducts in a DNA mutational hot spot sequence. J. Am. Chem. Soc. 123, 7054-7066. Hess, M. T., Gunz, D., Luneva, N., Geacintov, N. E., and Naegeli, H. (1997) Base pair conformation-dependent excision of benzo[a]pyrene diol epoxide-guanine adducts by human nucleotide excision repair enzymes. Mol. Cell Biol. 17, 7069-7076. Giri, I., and Stone, M. P. (2003) Wobble dC‚dA pairing 5′ to the cationic guanine N7 8,9-dihydro-8-(N7-guanyl)-9-hydroxyaflatoxin B1 adduct: implications for nontargeted AFB1 mutagenesis. Biochemistry 42, 7023-7034. Perlow, R. A., and Broyde, S. (2001) Evading the proofreading machinery of a replicative DNA polymerase: induction of a mutation by an environmental carcinogen. J. Mol. Biol. 309, 519536. Saenger, W. (1984) Principles of Nucleic Acid Structure, SpringerVerlag, New York. Cate, J. H., Gooding, A. R., Podell, E., Zhou, K., Golden, B. L., Kundrot, C. E., Cech, T. R., and Doudna, J. A. (1996) Crystal structure of a group I ribozyme domain: principles of RNA packing. Science 273, 1678-1685. Chary, P., and Lloyd, R. S. (1995) In vitro replication by prokaryotic and eukaryotic polymerases on DNA templates containing site-specific and stereospecific benzo[a]pyrene-7,8dihydrodiol-9,10-epoxide adducts. Nucleic Acids Res. 23, 13981405. Ponten, I., Sayer, J. M., Pilcher, A. S., Yagi, H., Kumar, S., Jerina, D. M., and Dipple, A. (2000) Factors determining mutagenic potential for individual cis and trans opened benzo[c]phenanthrene diol epoxide-deoxyadenosine adducts. Biochemistry 39, 4136-4144. Rechkoblit, O., Zhang, Y., Guo, D., Wang, Z., Amin, S., Krzeminsky, J., Louneva, N., and Geacintov, N. E. (2002) Translesion synthesis past bulky benzo[a]pyrene diol epoxide N2-dG and N6dA lesions catalyzed by DNA bypass polymerases. J. Biol. Chem. 277, 30488-30494.

TX034218L