Chlorine Functionalization of a Model Phenolic C8 ... - ACS Publications

May 24, 2015 - Aaron A. Witham,. †. Anne M. R. Verwey,. †. Michael Sproviero,. †. Richard A. Manderville,*. ,†. Purshotam Sharma,. ‡ and Sta...
0 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/crt

Chlorine Functionalization of a Model Phenolic C8-Guanine Adduct Increases Conformational Rigidity and Blocks Extension by a Y‑Family DNA Polymerase Aaron A. Witham,† Anne M. R. Verwey,† Michael Sproviero,† Richard A. Manderville,*,† Purshotam Sharma,‡ and Stacey D. Wetmore*,‡ †

Departments of Chemistry and Toxicology, University of Guelph, Guelph, ON, Canada N1G 2W1 Department of Chemistry and Biochemistry, University of Lethbridge, Lethbridge, Alberta, Canada T1K 3M4

‡

S Supporting Information *

ABSTRACT: Certain phenoxyl radicals can attach covalently to the C8-site of 2′deoxyguanosine (dG) to afford oxygen-linked C8-dG adducts. Such O-linked adducts can be chemically synthesized through a nucleophilic displacement reaction between a phenolate and a suitably protected 8-Br-dG derivative. This permits the generation of model O-linked C8-dG adducts on scales suitable for insertion into oligonucleotide substrates using solid-phase DNA synthesis. Variation of the C8-aryl moiety provides an opportunity to derive structure− activity relationships on adduct conformation in duplex DNA and replication bypass by DNA polymerases. In the current study, the influence of chlorine C8dG functionalization on in vitro DNA replication by Klenow fragment exo− (Kf−) and the Y-family polymerase (Sulfolobus solfataricus P2 DNA polymerase IV (Dpo4)) has been determined. Model O-linked C8-dG adducts derived from the pentachlorophenoxyl radical ([PCP]G) and 2,4,6-trichlorophenoxyl radical ([TCP]G) were inserted into the reiterated G3-position of the NarI sequence (12-mer, NarI(12); and 22-mer, NarI(22)), which is a known hotspot for frameshift mutations mediated by N-linked polycyclic C8-dG adducts in bacterial mutagenesis. Within the NarI(12) duplex, the unsubstituted C8-phenoxy-dG ([PhO]G) adduct adopts a minimally perturbed B-form helix. Chlorination of [PhO]G to afford [PCP]G does not significantly change the adduct conformation within the NarI(12) duplex, as predicted by molecular dynamics simulations. However, when using NarI(22) for DNA synthesis in vitro, the chlorinated [PCP]G and [TCP]G lesions significantly block DNA replication by Kf− and Dpo4, whereas [PhO]G is readily bypassed. These findings highlight the impact that chlorine substituents impart to bulky C8-dG lesions.

■

INTRODUCTION

Bulky DNA adducts produced by chlorinated aromatic compounds include a C8-substituted biaryl ether 2′-deoxyguanosine (dG) adduct containing a C8-O-aryl linkage (denoted O-linked adduct) that is derived from reaction of dG with the pentachlorophenoxyl radical ([PCP]G, Figure 1a).7,8 Identification of [PCP]G prompted the development of synthetic methods to generate O-linked C8-dG adducts9,10 on scales suitable for conversion into phosphoramidites for insertion into oligonucleotides using solid-phase synthesis.11,12 Bulky adducts formed at the C8-site of dG are among the most common lesions produced by chemical carcinogens.13 Other bulky C8-dG lesions contain a C8-N-aryl linkage (denoted Nlinked adducts) or a C8-aryl linkage (denoted C-linked adducts). For N-Linked C8-dG adducts produced by arylamine carcinogens,14 common conformational motifs in duplex DNA (i.e., the major groove B-type,15 the intercalated base-displaced stacked (S-type),16,17 or the minor groove wedge (Wtype),18−20 Figure 1b) have been structurally characterized by

Chlorine substituents have a profound impact on the biological activities of organic compounds. Outcomes of chlorination may be positive, highlighted by their ability to enhance conformational rigidity and improve therapeutic potency,1,2 or negative, exemplified by the carcinogenic properties that chlorine substituents impart to biphenyl and phenolic ring systems.3,4 Properties of chlorine substituents that enhance toxicity5 include (1) increased electrophilicity of reactive intermediates due to the electron-withdrawing effect of chlorine, (2) augmented lipophilicity, which increases persistence in bodily tissues and binding interactions with hydrophobic active sites in enzymes to facilitate biotransformation, and (3) promotion of chemical reactions due to the leaving group ability of the chlorine substituent. For chlorophenols and polychlorinated biphenyls, enzymatic oxidative transformations produce electrophiles that promote formation of bulky DNA adducts.6−8 The resulting lesions contain chlorine substituents, and currently there is a lack of knowledge concerning the influence of chlorine functionalization on DNA replication. © 2015 American Chemical Society

Received: April 10, 2015 Published: May 24, 2015 1346

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology

Figure 1. (a) Structures of O-linked C8-dG adducts, oligonucleotide sequences of NarI(12) and NarI(22), and definition of important torsion angles in the O-linked C8-dG adduct, including χ at the nucleobase−sugar linkage and θ and ϕ at the C8-linkage. The χ dihedral angle defines the glycosidic bond orientation to be anti (χ = 180 ± 90°) or syn (χ = 0 ± 90°). (b) Depictions of the three major conformations produced by N-linked C8-dG adducts.

the G3-position (X) of the NarI sequence (Figure 1a),11,12 which is a hotspot for frameshift mutations induced by polycyclic N-linked C8-dG adducts in bacterial mutagenesis30 via a two-base slippage mechanism.31 Within the NarI(12) duplex, [PhO]G adopted the major groove B-type structure with minimal perturbation to the helix,11 as noted for the Nlinked C8-aniline-dG adduct.25 Replication past [PhO]G was examined using primer-elongation assays with the NarI(22) template (Figure 1a) annealed to a 15-mer primer in the presence of Kf− or the Y-family DNA polymerase IV (Dpo4) from Sulfolobus solfataricus.12 The elongation assays revealed that [PhO]G did not strongly block the progress of DNA replication. The relative frequencies of nucleotide incorporation opposite [PhO]G were also similar to insertion opposite G, suggesting that [PhO]G does not alter the misinsertion frequency of either polymerase. These observations correlate with the conformational/biological response of the N-linked C8-aniline-dG adduct,25 suggesting that replacement of the flexible amine linkage with an ether linkage does not strongly influence mutagenic outcome. To determine how chlorine substituents impact C8-dG conformational preferences and DNA synthesis by Kf− or Dpo4, we have incorporated [PCP]G and C8-(2,4,6-trichlorophenoxyl)-dG ([TCP]G, Figure 1a) into NarI(12) and NarI(22). Comparison with our previous studies on [PhO]G suggests that, although chlorine substituents do not dramatically change the C8-dG conformation within duplex DNA, the biochemical outcome is profoundly affected.

NMR spectroscopy. Relating the adducted duplex structure to adduct-induced mutagenicity is a complicated undertaking because a single adduct can adopt multiple conformations, each with the potential of producing different mutagenic outcomes.21 Furthermore, mutagenic outcomes are greatly influenced by the types of DNA polymerases employed (i.e., high-fidelity versus lesion bypass polymerases) and the sequence context of the adduct.21 Nevertheless, great strides have been made to understand adduct-induced mutagenicity for N-linked C8-dG adducts.15−24 In general, N-linked C8-dG adducts that exhibit potent mutagenicity have planar polycyclic structures and, despite conformational heterogeneity, adopt the syn-conformation to produce the S-type or W-type duplexes. The single-ringed N-linked C8-aniline-dG adduct preferentially induces B-type conformation and lacks potent mutagenicity.25 In contrast to N-linked C8-dG adducts, C-linked lesions tend to induce B-type conformation, although conformational heterogeneity is observed depending on the size of the C8aryl ring,26 the nature and placement of substituents on the C8aryl ring,27 and the adduct ionization state.28 Thus, C- and Nlinked adducts have different mutagenic outcomes. For example, in contrast to the nonmutagenic N-linked C8aniline-dG adduct,25 the single-ringed C8-phenyl-dG ([Ph]G) adduct induces transversion and frameshift mutations26,29 and strongly blocks the progress of DNA replication by high-fidelity DNA polymerases: Klenow fragment exo− (Kf−)26,29 and DNA pol α.26 To establish the structural and biochemical impact of an Olinked C8-dG adduct, the unsubstituted C8-phenoxyl-dG adduct ([PhO]G, Figure 1a) was previously incorporated into 1347

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology

■

1 nm and scanning speed at 100 nm/min. Spectra of solutions containing 6.0 μM DNA duplexes, prepared as described above, were recorded in quartz glass cells (110-QS) and were the averages of five accumulations that were smoothed using the Jasco software. MD Simulations. MD simulations were performed on NarI(12) (5′-CTCGGCXCCATC) adducted oligonucleotides containing [PhO]G or [PCP]G at the G3 (X) position and paired against one of the four possible natural nucleotides (A, C, G, and T). Initial structures of damaged DNA were built using both the syn- and anticonformations of the adduct, where the initial orientation of the C8 moiety roughly corresponds to the lowest energy θ value of ∼180°. Full computational details are available in Supporting Information and followed strategies previously outlined.26−28 Radiolabeling and Annealing. T4 polynucleotide kinase and [γ-32P]ATP were used to label the 15-mer primer strands at the 5′ end. The unmodified and modified DNA primer/template duplexes were prepared by annealing the 15-mer primer and the [TCP]G- and [PCP]G-modified NarI(22) template strands (1.5 equiv of template strand) through heating the mixtures at 95 °C for 10 min, followed by slow cooling to room temperature overnight. Single-Nucleotide Incorporation Assays. Kf− or Dpo4 were used to perform primer-extension reactions on the annealed primer/ template duplex in the presence of dCTP, dGTP, dATP, or dTTP. Reactions were initiated by the addition of the dNTP (at a final concentration of 25 μM) to the enzyme/DNA mixture to give a final reaction volume of 10 μL. The final concentrations for Kf− assays were 50 mM NaCl, 10 mM Tris-HCl (pH 7.9), 10 mM MgCl2, 1 mM dithiothreitol (DTT), 100 nM duplex, and 20 nM Kf−. The final concentrations for Dpo4 assays were 50 mM NaCl, 50 mM Tris (pH 8.0), 2.5 mM MgCl2, 5 mM DTT, 100 μg/mL bovine serum albumin (BSA), 5% glycerol (v/v), 100 nM duplex, and 20 nM Dpo4. Reactions were incubated at 37 °C for 1 h with Kf− and 30 min with Dpo4, followed by 4 μL being transferred and mixed with 36 μL of loading dye (95% formamide, 20 mM EDTA, 0.05% xylene cyanol and bromophenol blue (v/v)) to terminate the reaction. Quenched reactions (4 μL) were then subjected to 15% (v/v) polyacrylamide gel electrophoresis in the presence of 7 M urea, and incorporation products were visualized from a phosphorimaging screen using a BioRad phosphorimager. Band quantification was performed using Quantity One software. Density analysis was performed by circumscribing each band of interest, followed by the software generating density values for each. Using Excel, the relative intensity of each band was determined as a percentage by dividing the density of a single band by the total density of all bands in the specific lane and multiplying by 100. Full-Length Extension Assays. Kf− or Dpo4 (20 nM) were used to perform primer-extension reactions on each previously labeled and annealed primer/template duplex in the presence of a 100 μM mixture of all four dNTPs. All other final concentrations in each reaction were the same as those outlined for the single-nucleotide incorporation assays. Reactions were initiated by the addition of the dNTP mix to enzyme/DNA mixtures to give a final reaction volume of 10 μL and were incubated at 37 °C for 1 h with Kf− or various time points (15, 30, 45, 60, and 90 min) with Dpo4. Following incubation, reactions were quenched and incorporation products were resolved in the same manner as above.

EXPERIMENTAL PROCEDURES

Materials. Pentachlorophenol (PCP), 2,4,6-trichlorophenol (TCP), and reagents for phosphoramidite synthesis (N,N-dimethylformamide diethyl acetal, 4,4′-dimethoxytrityl chloride, 2-cyanoethyl N,N-diisopropylchlorophosphoramidite) were purchased from commercial sources and used as received. Native NarI(12) and NarI(22) were purchased from Sigma-Aldrich Ltd. (Oakville, ON). The oligonucleotides were purified by Sigma-Aldrich using polyacrylamide gel electrophoresis (PAGE). All unmodified phosphoramidites (bz-dACE, ac-dC-CE, dmf-dG-CE, and dT-CE), activator (0.25 M 5(ethylthio)-1H-tetrazole in CH3CN), oxidizing agent (0.02 M I2 in THF/pyridine/H2O, 70:20:10, v/v/v), deblock (3% (v/v) dichloroacetic acid in dichloromethane), cap A (THF/2,6-lutidine/acetic anhydride), cap B (methylimidazole in THF), and 1000 Å controlled pore glass (CPG) solid supports were purchased from Glen Research (Sterling, VA). Escherichia coli pol I Klenow fragment exo− (Kf−) and T4 polynucleotide kinase were purchased from New England BioLabs, and Sulfolobus solfataricus P2 DNA polymerase IV (Dpo4) was purchased from Trevigen Inc. Isotopically labeled ATP ([γ-32P]-ATP) was purchased from PerkinElmer. Methods. Synthesis of [PCP]G and [TCP]G Phosphoramidites. The nucleoside adducts [PCP]G and [TCP]G were synthesized as previously outlined.10 They were converted into phosphoramidites (Scheme S1) using procedures for the synthesis of the [PhO]G phosphoramidite (see Supporting Information for experimental details and NMR spectra of synthetic intermediates) .11 Oligonucleotide Synthesis and Purification. All adducted oligonucleotide substrates were prepared on a 1 μmol scale using a BioAutomation MerMade 12 automatic DNA synthesizer using standard or modified β-cyanoethylphosphoramidite chemistry. Upon completion of DNA synthesis, the crude NarI oligonucleotides solutions were deprotected and cleaved from their solid support in aqueous ammonium hydroxide, filtered using syringe filters (PVDF 0.20 μm), and concentrated under diminished pressure. Samples were then resuspended in Milli-Q water (18.2 MΩ) and purified using an Agilent HPLC instrument equipped with an autosampler, a diode array detector (monitored at 258 and 310 nm), and autocollector. Separation was carried out at 50 °C using a 5 μm reversed-phase (RP) semipreparative C18 column (100 × 10 mm) with a flow rate of 3.5 mL/min and various gradients of buffer B in buffer A (buffer A = 95:5 aqueous 50 mM TEAA, pH 7.2/acetonitrile; buffer B = 30:70 aqueous 50 mM TEAA, pH 7.2/acetonitrile). Collected DNA samples were lyophilized to dryness and redissolved in 18.2 MΩ water for quantification by UV−vis measurement using ε260. Extinction coefficients were obtained from the following website: http://www. idtdna.com/analyzer/applications/oligoanalyzer. The C8-phenoxy-Gmodified oligonucleotides were assumed to have the same extinction coefficient as those of the natural NarI(12) (102 100 M−1 cm−1) and NarI(22) (185 700 M−1 cm−1) oligonucleotides. ESI-MS Analysis of Oligonucleotides. MS experiments for identification of the NarI oligonucleotides were conducted on a Bruker amaZon quadrupole ion trap SL spectrometer (Bruker Daltonics Ltd., Milton, ON). Oligonucleotide samples were prepared in 90% Milli-Q filtered water/10% methanol (v/v) containing 0.1 mM ammonium acetate. Masses were acquired in the negative ionization mode with an electrospray ionization source. Optical Measurements for NarI(12). All UV melting temperatures (Tm) of NarI(12) oligonucleotides were measured using a Cary 300Bio UV−vis spectrophotometer (Agilent Technologies, Santa Clara, CA) equipped with a 6 × 6 multicell block-heating unit using quartz (114-QS) 10 mm light path cells. Oligonucleotide samples were prepared in 50 mM phosphate buffer, pH 7, with 100 mM NaCl, using equivalent amounts (6.0 μM) of the unmodified or adducted oligonucleotide and its complementary strand. The Tm values of the duplexes were determined as previously outlined.11,26 Circular dichroism (CD) spectra were recorded on a Jasco J-815 CD spectropolarimeter (Jasco, Easton, MD) equipped with a 1 × 6 multicell block thermal controller and a water circulator unit. Spectra were collected at 10 °C between 200 and 400 nm, with a bandwidth of

■

RESULTS NarI(12) Properties. UV Thermal Melting. Solid-phase DNA synthesis was employed to incorporate the O-linked C8dG adducts ([PCP]G and [TCP]G) into the G3-position (X) of NarI(12) (see Supporting Information for NMR spectra of synthetic intermediates and electrospray ionization (ESI) MS spectra of adducted oligonucleotides). UV-derived thermal melting parameters for the [PCP]G- and [TCP]G-modified NarI(12) oligonucleotides are summarized in Table 1, which also includes previously published Tm values for the unmodified and [PhO]G-modified NarI(12) duplexes.11 The chlorinated

1348

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology Table 1. Thermal Melting Parameters of C8-Phenoxyl-dG Modified NarI(12)

X

N

Tm (°C)a

G [PhO]G [TCP]G [PCP]G G [PhO]G [TCP]G [PCP]G G [PhO]G [TCP]G [PCP]G

C C C C G G G G A A A A

63.6 52.8 46.0 46.3 54.0 52.1 45.3 45.5 51.4 44.4 44.0 45.1

ΔTmb −10.8 −17.6 −17.3 −1.9 −8.7 −8.5 −7.0 −7.4 −6.2

X

N

Tm (°C)a

G [PhO]G [TCP]G [PCP]G G [PhO]G [TCP]G [PCP]G G [PhO]G [TCP]G [PCP]G

T T T T THF THF THF THF −2 −2 −2 −2

53.4 47.1 43.9 43.7 45.7 42.3 42.7 46.4 39.4 35.0 38.2 39.9

ΔTmb −6.3 −9.5 −9.9 −3.4 −3.0 +0.7 −4.4 −1.2 +0.5

Figure 2. CD spectral overlays of NarI(12) duplexes with X = G (dashed black lines), X = [PhO]G (red lines), X = [TCP]G (blue lines), and X = [PCP]G (green lines). All spectra of duplexes (6 μM) were recorded in 50 mM sodium phosphate buffer, pH 7, with 0.1 M NaCl at 10 °C.

Errors in Tm are ±1 °C. bΔTm = Tm (modified duplex) − Tm (unmodified duplex). a

amplitudes of the bands at 240 and 275 nm in the CD spectra of NarI(12) duplexes in which the adducts were paired with THF exhibited increased intensity. Overall, the modified duplexes containing the various C8-phenoxy-dG adducts displayed similar CD curves, indicating that chlorination of the phenyl ring did not strongly perturb the global and local DNA structure at the lesion site. In all spectra, it was not possible to resolve ellipticities due to the C8-aryl-dG modified bases. MD Simulations. To model the structural impact of chlorine substitution on the duplex conformation, MD simulations and free energy calculations were carried out on different NarI(12) conformations with (anti and syn) [PhO]G or [PCP]G paired opposite C, G, A, and T. Adducts Paired against C. Our simulations reveal that the major groove B-type conformation is preferentially induced by both [PhO]G and [PCP]G by at least 24.7 kJ mol−1 compared to the S- and W-type conformations (Table 2). Relative to Btype helices, the W-type conformation associated with [PCP]G and [PhO]G are equally unstable (i.e., 25−28 kJ mol−1). However, the S-type conformation is significantly more unstable for [PhO]G (53 kJ mol−1) than [PCP]G (29 kJ

lesions strongly decreased duplex stability compared to that for [PhO]G when paired opposite C (i.e., ΔTm = −10.8 °C for [PhO]G, −17.6 for [TCP]G, and −17.3 for [PCP]G) and G (i.e., ΔTm = −1.9 °C for [PhO]G, −8.7 for [TCP]G, and −8.5 for [PCP]G). However, when paired opposite T, the chlorinated lesions only moderately decreased duplex stability compared to that for [PhO]G (i.e., ΔTm = −6.3 °C for [PhO]G, −9.5 for [TCP]G, and −9.9 for [PCP]G), whereas similar Tm values for the adducted duplexes were observed when mismatched with A (Tm = 44−45 °C; ΔTm ∼ −7.0). To assess the influence of π-stacking interactions in the absence of H-bonding with the opposing base, adducts were paired opposite the stable tetrahydrofuran (THF) model of an abasic site.32 The ΔTm values ranged from [PhO]G (−3.4 °C) < [TCP]G (−3.0 °C) < [PCP]G (+0.7 °C), indicating increased duplex stability upon chlorination. The adducted NarI(12) strands were also hybridized with the truncated 10mer sequence (−2). Pairing NarI(12) with the 10-mer complementary strand provides a bulged duplex that models the intermediate state involved in a slippage-mediated mechanism of deletion mutation.33 Enhanced stability of the slipped mutagenic intermediate (SMI) is believed to correlate with an ability to promote −2 frameshift mutagenesis mediated by N-linked C8-dG adducts in the NarI sequence.33 For example, the N-linked C8-dG adduct of N-acetyl-2-aminofluorene ([AAF]G) is a potent frameshift mutagen and affords a ΔTm value of +15 °C for the truncated duplex.34 In contrast, the O-linked C8-phenoxy-dG adducts failed to stabilize the truncated NarI(12) duplex compared to that of the unmodified control (Table 1). Circular Dichroism. The NarI(12) duplexes with X (G, [PhO]G, [TCP]G, and [PCP]G) paired opposite N = C, G, THF, and −2 were analyzed using CD spectroscopy (Figure 2) to determine the impact of the adduct on the global tertiary structures. All duplexes showed B-DNA characteristics, with positive (275 nm) and negative (244 nm) S-shaped CD curves with a crossover at approximately 260 nm (Figure 2).35 For NarI(12) paired opposite C and −2, modified duplexes displayed significant hypochromicity in the positive CD band at 275 nm, which is indicative of partial unwinding of the helix and decreased stacking interactions.35 Conversely, the

Table 2. Comparison of the Relative Free Energies of Adducted NarI(12) Conformations with [PCP]G or [PhO]G Paired Opposite N DNA

adduct

N

Grel ([PCP]G)a

Grel ([PhO]G)a

B W S B W B B W

anti syn syn anti syn anti anti syn

anti-C syn-C; anti-C syn-C syn-G anti-G syn-A anti-T anti-T

0.0 28.0 29.3 26.8 0.0

0.0b 24.7b 53.1 34.3b 0.0 c

0.0 13.8

0.0 47.7

Relative free energy in kJ mol−1 with respect to the lowest energy conformation when the adduct is paired opposite the same nucleobase (A, C, G, or T). bRelative energies taken from ref 11. cRelative energy not available when the adduct is opposite A since only one conformation was characterized. a

1349

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology mol−1, Table 2), indicating that chlorine enhances the stacking interactions of the C8 moiety. The most stable (B-type) conformation of [PCP]G maintains Watson−Crick (WC) H-bonding with complementary C (>98% occupancy of three H-bonds; Figures 3 and S19,

restricting the conformational flexibility of the phenoxyl ring (Figure S18, Supporting Information), the steric bulk induced by chlorine affects the glycosidic torsion of the lesion. Adducts Paired against G. The W-type conformation is preferred when [PhO]G or [PCP]G is paired opposite G by at least 27 kJ mol−1 compared to the B-type conformation (Table 2). In this conformation, [PCP]G adopts a unimodal synglycosidic orientation (average χ ∼ 62°) and an orthogonal orientation at ϕ (peak at 100°), whereas θ deviates from planarity by ∼40° (peak at 150°, with an average value of 141°; Figure S20, Supporting Information). Although [PhO]G exhibits a χ distribution (average χ ∼ 56°) similar to that of [PCP]G, [PhO] is more planar about θ (average ∼165°) than [PCP]G, mainly due to steric repulsion between the ortho chlorine atoms of PCP and N7 of G. However, [PhO]G is more flexible about ϕ compared to that of [PCP]G (Figure S20, Supporting Information). In addition, the Hoogsteen Hbonding is more persistent for [PCP]G than it is for [PhO]G (Figure S21, Supporting Information). Adducts Paired against A. Similar to when the adduct is against C, [PCP]G adopts a high anti-orientation in the preferred B-type conformation against A (average χ ∼ 275°; Figure S22, Supporting Information). However, θ is close to planarity (average θ ∼ 159°), whereas ϕ is nearly orthogonal (average ϕ ∼ 98°; Figure S7). The lesion site is stabilized by three persistent Hoogsteen H-bonds (∼99% occupancy of two bonds and ∼70% occupancy of the third; Figures 1 and S23, Supporting Information). In contrast, [PhO]G adopts both the anti and high anti-conformations (average χ ∼ 248 ± 22°), an orthogonal (yet flexible) ϕ (Figure S22, Supporting Information), and weaker Hoogsteen H-bonding compared to those for [PCP]G (Figure S23, Supporting Information). Adducts Paired against T. Although a B-type conformation (Figure 3) is preferred when [PhO]G and [PCP]G are against T, the W-type conformation associated with [PCP]G is only 14 kJ mol−1 higher in energy (Table 1), whereas the [PhO]G Wtype conformation is likely energetically inaccessible (48 kJ mol−1 above the B-type conformation; Table 2). In the B-type conformation, [PCP]G adopts a high anti-glycosidic orientation (average χ ∼ 273°) and possesses a bimodal distribution with respect to θ (dominant peak at θ ∼ 160° and minor peak at θ ∼ 210°; Figure S24, Supporting Information). This conformation is stabilized by two H-bonds between the lesion and opposing T (>79% occupancy; Figure S25, Supporting Information). In contrast, the B-type conformation associated with [PhO]G adopts the anti-glycosidic orientation and possesses more persistent adduct−T H-bonding (>87% occupancy; Figure S25, Supporting Information). However, [PhO]G is more flexible about ϕ than is [PCP]G (Figure S24, Supporting Information). Primer Extension of NarI(22). Extension by Kf−. To test the fidelity of Kf− opposite the chlorinated O-linked C8phenoxy-dG adducts, single-nucleotide incorporation (SNI) assays were performed with the NarI(22)−15-mer duplexes containing X = [TCP]G or [PCP]G in the presence of 25 μM of individual dNTPs and 20 nM Kf− (Figure 4a). Lanes containing X = G and [PhO]G are shown for comparison.12 For the templates containing X = [TCP]G and [PCP]G, insertion of the correct base C was predominantly observed, with relative amounts of 78 and 70%, respectively (Figure 4b). In the presence of all four dNTPs (Figure 4c), full-length extension (FLE) by Kf− was observed in the X = G and [PhO]G lanes, as noted previously.12 A +8 band was also observed in the gel that is indicative of blunt-end extension and

Figure 3. Comparison of the most stable conformations of [PCP]Gand [PhO]G-adducted DNA with the lesion paired opposite different canonical nucleobases.

Supporting Information), χ of the adduct largely adopts the high anti-orientation (χ ∼ 250−300° for ∼90% simulation time), and θ deviates from planarity (180°) by a maximum of 50° (Figure S18, Supporting Information). In addition, the PCP moiety remains almost orthogonal with respect to the C8−O bond (ϕ ∼ 90 ± 30° for >92% simulation time), indicating significant ϕ conformational rigidity (Figure S18, Supporting Information). The [PCP]G-adducted DNA conformation differs from that induced by unsubstituted [PhO]G, which adopts the anti (rather than the high anti) orientation (Figure S18, Supporting Information). Therefore, in addition to 1350

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology

Figure 4. (a) SNI primer-extension assays by Kf− (20 nM) in the presence of individual dNTPs (indicated under each lane, 25 μM). (b) Relative amount of each nucleotide incorporated against X = G (white), [PhO]G (gray), [TCP]G (lined), or [PCP]G (black). (c) FLE of NarI(22)−15-mer duplexes, where X = G, [PhO]G, [TCP]G, or [PCP]G by Kf−. Reactions were carried out for 1 h at 37 °C.

Figure 5. (a) SNI primer-extension assays by Dpo4 in the presence of individual dNTPs (indicated under each lane, 25 μM). (b) Relative amount of each nucleotide incorporated against X = G (white), [PhO]G (gray), [TCP]G (lined), or [PCP]G (black). (c) FLE of NarI(22)−15-mer duplexes where X = G, [PhO]G, [TCP]G, or [PCP]G by Dpo4. Reactions were carried out for 30 min at 37 °C.

is independent of the nature of the polymerase employed and template strand (modified or unmodified).26 In contrast, the chlorinated O-linked adducts strongly stalled insertion one base across from the lesion (at position 1; see numbering of template in Figure 4a, FLE by the polymerase adds 7 bases (up to position 7 on the gel)). Replication past [PCP]G was more strongly blocked than replication past [TCP]G, as some +6 and

+7 (full-length) bands were observed in the X = [TCP]G lane (Figure 4c). Extension by Dpo4. Since extension past both [TCP]G and [PCP]G by Kf− was strongly stalled, replication of the modified NarI(22) templates was examined using the prototypical Yfamily polymerase Dpo4.36−38 Y-Family DNA polymerases contain a capacious active site and can bypass damage when 1351

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology

Figure 6. (a) SNI by Dpo4 in the presence of 20% DMSO on NarI(22)−15-mer duplexes in the presence of individual dNTPs (indicated under each lane, 25 μM). Reactions were incubated for 30 min at 37 °C. (b) Relative amounts of each dNTP incorporated opposite X = [PhO]G (gray), [TCP]G (lined), or [PCP]G (black). (c) Timed FLE by Dpo4 in the presence of 20% DMSO on NarI(22)−15-mer duplexes containing X = [PhO]G, [TCP]G, or [PCP]G, with aliquots of sample quenched after the time listed under the gel. The first lane on the left (5*) was included as a marker for each of the produced extension products by FLE by 20 nM of Dpo4 on the unmodified duplex after 5 min of incubation.

stalling is alleviated in the presence of organic solvents, such as 20% (v/v) DMSO. Kirouac and co-workers determined that the bulky pyrene ring system in [AP]G associates with hydrophobic residues in the cleft between the finger and little finger domains of Dpo4,40 which is thought to cause stalling of the polymerase during primer-elongation assays. The addition of organic solvents such as DMSO decreases the dielectric constant of the aqueous buffer, which diminishes the stalling effect, possibly by stabilizing [AP]G in an anti-conformation to allow proper H-bonding with the opposing base C.40 Thus, SNI and timed FLE by Dpo4 were carried out on the NarI(22)−15mer duplexes containing X = [PhO]G, [TCP]G, or [PCP]G in the presence of 20% DMSO to determine if a nonpolar solvent would increase the extent of replication. In the presence of individual dNTPs (Figure 6a), the extent of misincorporation opposite all modified bases was increased and multiple incorporations of C were observed. For example, approximately 12 and 16% of the extended primers were further extended to include four C bases opposite [TCP]G and [PCP]G, respectively, which is indicative of polymerase slippage.26 In the presence of all four dNTPs (Figure 6c), [PhO]G was extended to create a full-length product. Extension past both chlorinated adducts increased, with faint bands detected corresponding to a full-length product. However, extension was still strongly stalled opposite the lesion (position 1) and at position 3 (indicated by the arrow). These results contrast the successful full-length extension past [AP]G by Dpo4 in the presence of 20% DMSO,40 suggesting that the chlorinated adducts ([TCP]G and [PCP]G) may in fact be stronger blocks than the [AP]G adduct, which contains the N-linked pyrene ring system.

high-fidelity polymerases are blocked. Thus, they act as adduct tolerance factors. Dpo4 has been shown to bypass a broad range of bulky DNA adducts, such as cisplatin-GG,39 N(deoxyguanosin-8-yl)-1-aminopyrene ([AP]G),40 and a benzo[a]pyrene diol epoxide (BPDE) adduct of dA ([BPDE]A).41 In SNI assays (Figure 5a), Dpo4 exhibited low fidelity with the unmodified NarI(22) template (X = G) and the X = [PhOG] template, as observed previously.12 The low fidelity with the undamaged template is consistent with the low geometric selection for correct base pairs by Dpo4.42 Interestingly, Dpo4 also exhibited low fidelity with the X = [PCP]G and [TCP]G NarI(22) templates. The relative amount of the correct base C (86% for [TCP]G and 78% for [PCP]G) was only slightly higher than the relative amount of incorporation for each incorrect base (79−82% for [TCP]G and 71−74% for [PCP]G) (Figure 5b) over the 30 min incubation time. This result differed markedly from our primerelongation assays utilizing NarI(22) containing various Clinked C8-dG adducts.26 Under identical conditions, the major product of insertion by Dpo4 opposite the C-linked adducts was the correct base C, with misincorporation of minor amounts of A and G. Opposite the N-linked [AP]G40 and [AAF]G43 lesions, Dpo4 also maintains correct C incorporation preference, as evidenced by primer-extension assays. In the presence of all four dNTPs (Figure 5c), Dpo4 was completely blocked following insertion of one base opposite the chlorinated adducts. Insertion opposite [TCP]G and [PCP]G was observed as a major stalling point for both templates, as a strong band was detected at position 1 after 90 min incubation time. Despite replication blockage, incorporation opposing the chlorinated adducts was not difficult for the polymerase, as the amount of remaining primer after 15 min of incubation was only ∼24% for [TCP]G and ∼40% for [PCP]G. Over the 90 min time frame, Dpo4 was able to fully extend the template containing [PhO]G. For other lipophilic bulky adducts, such as [AP]G40 and [BPDE]A,41 it has been previously found that polymerase

■

DISCUSSION Impact of Chlorination on Adduct Conformation in NarI(12). For [PhO]G in the NarI(12) duplex, we previously demonstrated that the adduct adopts the B-type conformation paired opposite C and the minor groove W-type conformation 1352

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology mismatched with G.11 The conformational analysis was based on MD simulations that were supported by CD and 19F NMR measurements.11 In the present study, MD simulations predict that the fully chlorinated [PCP]G adduct also adopts the Btype conformation paired opposite C and the W-type conformation paired opposite G. The major conformational change induced by chlorine attachment is predicted to be restriction of phenoxy ring flexibility with the PCP moiety almost orthogonal with respect to the C8−O bond. For [PCP]G paired opposite C, alternative syn-conformations (Stype and W-type) are predicted to be less stable by ∼28 kJ mol−1, whereas the alternative anti-conformation paired opposite G is less stable by 26.8 kJ mol−1 (Table 2). Therefore, such alternative conformations will likely be energetically inaccessible to [PCP]G-adducted DNA. The calculations appear to be consistent with the CD spectra (Figure 2), as the various C8-phenoxy-dG adducts displayed similar CD curves paired opposite C and G, indicating that chlorination of the phenyl ring did not strongly perturb the global and local DNA structure at the lesion site. Exposure of the lipophilic PCP moiety to bulk solvent in the major (B-type) or minor (Wtype) groove would also be expected to decrease helix stability (Table 1). For [PCP]G and [PhO]G paired opposite A and T, the MD structural data suggests a preference for B-type structures with the phenoxy ring located in the major groove (Table 2 and Figure 3). Paired opposite A, the Tm values (Table 1) do not strongly support a B-type structure for both adducts because solvent exposure of the PCP moiety in the major groove would be expected to decrease the Tm value compared to the corresponding Tm for the [PhO]G-adducted duplex. This suggests that an alternative (W-type) conformation with [PCP]G present in a syn-conformation may be accessible for [PCP]G paired opposite A. The calculations predict that a Wtype conformation for [PCP]G paired opposite T is energetically accessible (Table 2), suggesting conformational heterogeneity for the [PCP]G lesion paired opposite T. Impact of Chlorination on DNA Synthesis. The highfidelity polymerase Kf− mainly inserts the correct base C opposite the C8-phenoxy-dG lesions (Figure 4a). For the nonchlorinated [PhO]G adduct, the oxygen linker limits steric contacts between the phenyl moiety and the DNA helix, which allows [PhO]G to behave similar to that of natural dG.27 Consequently, Kf− is able to fully extend the 15-mer primer in the presence of the [PhO]G-damaged NarI(22) template (Figure 4c).12 Thus, [PhO]G is not expected to cause errorprone DNA synthesis in cells because it can be readily bypassed by a high-fidelity polymerase that will correctly insert C opposite the lesion. In contrast, the chlorinated adducts [PCP]G and [TCP]G block extension by Kf− following insertion of C opposite the adduct (Figure 4) due to the increased steric bulk and augmented lipophilicity provided by the chlorine substituents. Bulky N-linked43−45 and C-linked26,29 C8-dG adducts are also known to block extension by Kf− following C insertion opposite the adduct. The ability of [PCP]G and [TCP]G to block extension by the high-fidelity polymerase Kf− suggests that they would promote recruitment of Y-family translesion polymerases for potential adduct bypass.36−38 Crystal structures of Y-family polymerases show a more open active site with increased solvent exposure than replicative polymerases.37 This leads to fewer interactions with the DNA, significantly decreasing the fidelity of these enzymes, especially for undamaged templates,

with error rates increased up to 100 times compared to those for replicative polymerases.37 To avoid high levels of spontaneous mutagenesis, it is important that these low-fidelity enzymes be used only as required. In fact, lesion-induced mutations are believed to occur largely through error-prone bypass by Y-family polymerases following polymerase switching due to lesion blockage of replicative polymerases.46 The Y-family polymerase Dpo4 exhibited low fidelity with the NarI(22) templates containing G, [PhO]G, [PCP]G, and [TCP]G (Figure 5a). The low fidelity observed for Dpo4 with the unmodified NarI(22) template was to be expected given its low geometric selection for correct base pairs.42 However, Dpo4 generally inserts the correct base C opposite bulky Nlinked C8-dG adducts in primer-extension assays.40,43 For a variety of C-linked C8-dG adducts, we also found Dpo4 to mainly insert C opposite the adducts, under primer-extension conditions identical to those presented in Figure 5a for SNI opposite [PhO]G, [PCP]G, and [TCP]G.26 Although our gel data is qualitative in nature, it appears that chlorination of a C8dG adduct decreases the fidelity of Dpo4. In the reiterated G3 position of the NarI sequence, polycyclic N-linked C8-dG adducts promote two-base slippage in primerextension assays (i.e., incorporation of a second C in SNI assays that implies pairing of CC with GG in positions 3 and 4 in the template strand; Figure 5a).45 N-Linked C8-dG adducts that stabilize the SMI have a tendency to induce two-base slippage33,34 and produce the S-type conformation when paired opposite C in the duplex.21,22 C-Linked C8-dG adducts also have a strong tendency to induce polymerase slippage when inserted into the G3-position of NarI.26 In this case, the Clinked adducts do not stabilize the SMI, but they have a strong syn-conformational preference (25−27 kJ mol−1).26,27 Insertion of C opposite C-linked adducts requires the polymerase to flip the adduct from its stable syn-conformation into the energetically destabilized anti-conformation. The barrier for this process is proposed to cause polymerase stalling, an important aspect in the polymerase slippage mechanism.26,45 In contrast, the chlorinated O-linked C8-dG adducts failed to induce slippage in the SNI assays carried out in aqueous buffer (Figure 5a), although relatively weak intensity bands for multiple incorporations of C, which is indicative of polymerase slippage,26 was observed when 20% DMSO was added to the reaction (Figure 6a). Unlike polycyclic N-linked C8-dG adducts, the chlorinated adducts fail to stabilize the 2-base bulged duplex that mimics the SMI (Table 1). Furthermore, paired opposite C in the duplex, the S-type conformation is not a viable option for [PCP]G and [TCP]G due to their conformational rigidity, lack of planarity, and inability to effectively π-stack within the helix. Compared to C-linked C8-dG adducts, the O-linked variety possess a much weaker syn-conformational preference (∼16 kJ mol−1) due to the ether tether separating the phenyl moiety from the G nucleobase.27 This property is expected to decrease polymerase stalling and hence decrease polymerase slippage.26,45 Both chlorinated adducts ([PCP]G and [TCP]G) also strongly blocked FLE by Dpo4. While Dpo4 can readily insert a base opposite [PCP]G and [TCP]G, it cannot extend beyond that with any degree of efficiency. The N-linked [AAF]G adduct is also a strong block to extension by Dpo4.43 In the synconformation, the [AAF]G lesion is highly distorting to the Dpo4 active site, which inhibits the nucleotidyl transfer reaction, and causes replication to be aborted one base prior to the lesion site.47 In contrast, the anti-conformation of 1353

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology

this is clearly not the case, as both [PCP]G and [TCP]G strongly block FLE by Kf− and Dpo4. The gels also suggest that chlorination of a C8-dG adduct reduces the fidelity of Dpo4 in SNI assays. Given that both [PCP]G and [TCP]G strongly block extension by Dpo4, future studies will examine FLE of the adducted NarI(22) templates using additional Y-polymerase enzymes (pol η, pol κ, and Rev1) to determine which polymerase might extend the chlorinated lesion base pair. Such studies would also include kinetic assays to provide a quantitative estimate of the nucleotide selection preferences of Y-family polymerases opposite the chlorinated C8-dG adducts. Future computational studies attempting to provide a structural basis for error-prone DNA synthesis mediated by [PCP]G and [TCP]G will also explore the interaction of these bulky adducts within the active site of the DNA polymerase. The conformation of the lipophilic chlorinated adducts may vary within the active site of the enzyme compared to the Btype conformation within the fully paired duplex.

[AAF]G allows it to enter the active site of Dpo4 and base pair with the incoming dNTP for SNI opposite the adduct. Further extension past the lesion is then inhibited due to placement of the AAF moiety in the major groove, which is positioned to collide with the little finger domain of the polymerase, which impedes translocation.47 Our MD simulations (Figure 3) predict an anti-conformation for [PCP]G paired opposite C with the PCP moiety located in the major groove. Interaction of the chlorinated phenyl ring system with the little finger domain may provide a basis for the blocking ability of [PCP]G and [TCP]G. Nucleotide Excision Repair (NER). The mutational outcome of a DNA adduct depends on its ability to evade excision by DNA repair enzymes. NER is a common mechanism for the repair of bulky DNA lesions, which is believed to operate through recognition of local helical distortions or destabilizations caused by the damage rather than specific chemical characteristics of the lesion.48 For example, owing to greater helical distortions, the S-type conformation of N-linked polycyclic C8-dG adducts has a greater repair propensity than that of the less distorting B- and W-type structures.49 Since our MD simulations suggest that Olinked [PCP]G, and likely [TCP]G, favors the minimally perturbed B-type structure in NarI(12) paired opposite C (Figure 3 and Table 2), these lesions may be less repair prone than the N- and C-linked variants. A crystal structure of the recognition factor involved in the initial steps of eukaryotic NER bound to damaged DNA reveals that lesion recognition involves insertion of a β-hairpin into the major groove of DNA at the lesion site, which flips the lesion and opposing base(s) into the minor groove.50 In the case of the O-linked C8-dG adducts, the major groove location of the phenoxy moiety in the preferred B-type conformation may sterically obstruct major groove insertion of the β-hairpin.11 [PCP]G and [TCP]G would be expected to be stronger blocks to β-hairpin insertion than [PhO]G. In synchrony with the small helical distortions in the B-type conformation, this may prevent initial damage recognition, which may, in turn, prevent recruitment of downstream factors involved in subsequent NER steps and thereby provide repair resistance.

■

ASSOCIATED CONTENT

S Supporting Information *

Synthetic and computational details, NMR spectra of phosphoramidites, MS analysis of modified NarI oligonucleotides, and full description of MD simulation procedure and results. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/ acs.chemrestox.5b00143.

■

AUTHOR INFORMATION

Corresponding Authors

*(R.A.M.) Tel: 1-519-824-4120, ext. 53963; E-mail: [email protected]. *(S.D.W.) Tel: 1-403-329-2323; E-mail: stacey.wetmore@ uleth.ca. Funding

This work was supported by the Natural Sciences and Engineering Research Council of Canada [Discovery 3116002013 to R.A.M. and 249598-07 to S.D.W.]; Canada Research Chain Program [950-228175 to S.D.W.]; Canadian Foundation of Innovation [10679 to R.A.M. and 22770 to S.D.W.]; and the Ontario Research Fund [10679 to R.A.M.].

■

CONCLUSIONS In the present study, we have undertaken a series of experiments to determine how chlorine substituents impact C8-dG adduct conformational preferences and DNA synthesis by Kf− or Dpo4. Model O-linked C8-dG adducts derived from pentachlorophenol ([PCP]G) and 2,4,6-trichlorophenol ([TCP]G) were incorporated into the NarI sequence. A model O-linked C8-phenoxy-dG adduct ([PhO]G) was previously demonstrated to minimally perturb duplex DNA11 and DNA synthesis by the high-fidelity DNA polymerase: Klenow fragment exo− (Kf−).12 On the basis of these findings, [PhO]G is not expected to cause error-prone DNA synthesis in cells because it can be readily bypassed by a high-fidelity polymerase that will correctly insert C opposite the lesion. Our MD calculations suggest that chlorination of [PhO]G to afford [PCP]G does not dramatically impact adduct conformation in the NarI(12) duplex. Both [PhO]G and [PCP]G adopt a Btype structure paired opposite C, with the phenoxy ring located in the major groove. On the basis of this comparison alone, it would be tempting to suggest that [PhO]G and [PCP]G would have similar influences on DNA synthesis by Kf− or Dpo4. However, our primer-elongation experiments demonstrate that

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Computational resources provided by Westgrid and Compute/ Calcul Canada are greatly appreciated. Helpful comments by the reviewers are also appreciated.

■

ABBREVIATIONS dG, 2′-deoxyguanosine; [PCP]G, C8-pentachlorophenoxy-dG; [PhO]G, C8-phenoxy-dG; [TCP]G, C8-(2,4,6-trichlorophenoxy)-dG; CD, circular dichroism; MD, molecular dynamics; NER, nucleotide excision repair

■

REFERENCES

(1) Harris, C. M., Kannan, R., Kopecka, H., and Harris, T. M. (1985) The role of the chlorine substituents in the antibiotic vancomycin (preparation and characterization of mono- and didechlorovancomycin). J. Am. Chem. Soc. 107, 6652−6658. (2) Wulff, H., Miller, M. J., Hänsel, W., Grissmer, S., Cahalan, M. D., and Chandy, K. G. (2000) Design of a potent and selective inhibitor of 1354

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology the intermediate-conductance Ca2+-activated K+ channel, IKCa1: A potential immunosuppressant. Proc. Natl. Acad. Sci. U.S.A. 97, 8151− 8156. (3) Engel, L. S., Lan, Q., and Rothman, N. (2007) Polychlorinated biphenyls and non-hodgkin lymphoma. Cancer Epidemiol., Biomarkers Prev. 16, 373−376. (4) IARC (1979) Some halogenated hydrocarbons. IARC Monogr. Eval. Carcinog. Risks Hum. 20, 303−326. (5) Henschler, D. (1994) Toxicity of chlorinated organic compounds: effects of the introduction of chlorine in organic molecules. Angew. Chem., Int. Ed. Engl. 33, 1920−1935. (6) Pereg, D., Robertson, L. W., and Gupta, R. C. (2002) DNA adduction by polychlorinated biphenyls: adducts derived from hepatic microsomal activation and from synthetic metabolites. Chem.−Biol. Interact. 139, 129−144. (7) Dai, J., Wright, M. W., and Manderville, R. A. (2003) An oxygen bonded C8-deoxyguanosine nucleoside adduct of pentachlorophenol by peroxidase activation: evidence for ambident C8 reactivity by phenoxyl radicals. Chem. Res. Toxicol. 16, 817−821. (8) Dai, J., Sloat, A. L., Wright, M. W., and Manderville, R. A. (2005) Role of phenoxyl radicals in DNA adduction by chlorophenol xenobiotics following peroxidase activation. Chem. Res. Toxicol. 18, 771−779. (9) Dahlmann, H. A., and Sturla, S. J. (2011) Synthesis of 8-phenoxy2′-deoxyguanosine nucleoside adducts. Eur. J. Org. Chem. 2011, 2987− 2992. (10) Kuska, M. S., Majdi Yazdi, M., Witham, A. A., Dahlmann, H. A., Sturla, S. J., Wetmore, S. D., and Manderville, R. A. (2013) Influence of chlorine substitution on the hydrolytic stability of biaryl ether nucleoside adducts produced by phenolic toxins. J. Org. Chem. 78, 7176−7185. (11) Kuska, M. S., Witham, A. A., Sproviero, M., Manderville, R. A., Majdi Yazdi, M., Sharma, P., and Wetmore, S. D. (2013) Structural influence of C8-phenoxy-guanine in the NarI recognition DNA sequence. Chem. Res. Toxicol. 26, 1397−1408. (12) Verwey, A. M. R., Witham, A. A., Li, M., and Manderville, R. A. (2014) Mutagenicity analysis of C8-phenoxy-guanine in the NarI recognition DNA sequence. J. Toxins 1, 6. (13) Millen, A. L., Sharma, P., and Wetmore, S. D. (2012) C8-linked bulky guanosine DNA adducts: experimental and computational insights into adduct conformational preferences and resulting mutagenicity. Future Med. Chem. 4, 1981−2007. (14) Beland, F. A., and Kadlubar, F. F. (1990) Chemical carcinogenesis and mutagenesis, in Handbook of Experimental Pharmacology (Cooper, C. S., Grover, P. L., Eds.) Vol. 94/I, pp 267−325, Springer-Verlag, Heidelberg. (15) Cho, B. P., Beland, F. A., and Marques, M. M. (1992) NMR structural studies of a 15-mer DNA sequence from ras protooncogene modified at the first base of codon 61 with the carcinogen 4aminobiphenyl. Biochemistry 31, 9587−9602. (16) Mao, B., Hingerty, B. E., Broyde, S., and Patel, D. J. (1998) Solution structure of the aminofluorence [AF]-intercalated conformer of the syn-[AF]-C8-dG adduct opposite dC in a DNA duplex. Biochemistry 37, 81−94. (17) Wang, F., DeMuro, N. E., Elmquist, C. E., Stover, J. S., Rizzo, C. J., and Stone, M. P. (2006) Base-displaced intercalated structure of the food mutagen 2-amino-3-methylimidazo[4,5-f ]quinoline in the recognition sequence of the NarI restriction enzyme, a hotspot for −2 bp deletions. J. Am. Chem. Soc. 128, 10085−10095. (18) Shapiro, R., Hingerty, B. E., and Broyde, S. (1989) Minorgroove binding models for acetylaminofluorene modified DNA. J. Biomol. Struct. Dyn. 7, 493−513. (19) Elmquist, C. E., Wang, F., Stover, J. S., Stone, M. P., and Rizzo, C. J. (2007) Conformational differences of the C8-deoxyguanosine adduct of 2-amino-3-methylimidazo[4,5-f ]quinoline (IQ) within the NarI recognition sequence. Chem. Res. Toxicol. 20, 445−454. (20) Wang, F., Elmquist, C. E., Stover, J. S., Rizzo, C. J., and Stone, M. P. (2007) DNA sequence modulates the conformation of the food

mutagen 2-amino-3-methylimidazo[4,5-f ]quinoline in the recognition sequence of the NarI restriction enzyme. Biochemistry 46, 8498−8516. (21) Cho, B. (2010) Structure−function characteristics of aromatic amine−DNA adducts, in The Chemical Biology of DNA Damage (Geacintov, N. E., and Broyde, S., Eds) pp 217−238, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany. (22) Patel, D. J., Mao, B., Gu, Z., Hingerty, B. E., Gorin, A., Basu, A. K., and Broyde, S. (1998) Nuclear magnetic resonance solution structures of covalent aromatic amine−DNA adducts and their mutagenic relevance. Chem. Res. Toxicol. 11, 391−407. (23) Cho, B. P. (2004) Dynamic conformational heterogeneities of carcinogen-DNA adducts and their mutagenic relevance. J. Environ. Sci. Health, Part C 22, 57−90. (24) Rechkoblit, O., Kolbanovskiy, A., Malinina, L., Geacintov, N. E., Broyde, S., and Patel, D. J. (2010) Mechanism of error-free and semitargeted mutagenic bypass of an aromatic amine lesion by Yfamily polymerase Dpo4. Nat. Struct. Biol. 17, 379−389. (25) Shapiro, R., Ellis, S., Hingerty, B. E., and Broyde, S. (1998) Effect of ring size on conformations of aromatic amine−DNA adducts: the aniline−C8 guanine adduct resides in the B-DNA major groove. Chem. Res. Toxicol. 11, 335−341. (26) Sproviero, M., Verwey, A. M. R., Rankin, K. M., Witham, A. A., Soldatov, D. V., Manderville, R. A., Fekry, M. I., Sturla, S. J., Sharma, P., and Wetmore, S. D. (2014) Structural and biochemical impact of C8-aryl-guanine adducts within the NarI recognition DNA sequence: influence of aryl ring size on targeted and semi-targeted mutagenicity. Nucleic Acids Res. 42, 13405−13421. (27) Sharma, P., Majdi Yazdi, M., Merriman, A., Manderville, R. A., and Wetmore, S. D. (2015) Influence of the linkage type and functional groups in the carcinogenic moiety on the conformational preferences of damaged DNA: structural and energetic characterization of carbon- and oxygen-linked C8 phenolic-guanine adducts. Chem. Res. Toxicol. 28, 782−796. (28) Sharma, P., Manderville, R. A., and Wetmore, S. D. (2014) Structural and energetic characterization of the major DNA adduct formed from the food mutagen ochratoxin A in the NarI hotspot sequence: influence of adduct ionization on the conformational preferences and implications for the NER propensity. Nucleic Acids Res. 42, 11831−11845. (29) Kohda, K., Tsuomoto, H., Kasamatsu, T., Sawamura, F., Terahima, I., and Shibutani, S. (1997) Synthesis and miscoding specificity of oligodeoxynucleotide containing 8-phenyl-2′-deoxyguanosine. Chem. Res. Toxicol. 10, 1351−1358. (30) Fuchs, R. P. P., Schwartz, N., and Daune, M. P. (1981) Hot spots of frameshift mutations induced by the ultimate carcinogen Nacetoxy-N-2-acetylaminofluorene. Nature 294, 657−659. (31) Hoffmann, G. R., and Fuchs, R. P. P. (1997) Mechanisms of frameshift mutations: insight from aromatic amines. Chem. Res. Toxicol. 10, 347−359. (32) Takeshita, M., Chang, C. N., Johnson, F., Will, S., and Grollman, A. P. (1987) Oligodeoxynucleotides containing synthetic abasic sites. Model substrates for DNA polymerases and apurinic/apyrimidinic endonucleases. J. Biol. Chem. 262, 10171−10179. (33) Elmquist, C. E., Stover, J. S., Wang, Z., and Rizzo, C. J. (2004) Site-specific synthesis and properties of oligonucleotides containing C8-deoxyguanosine adducts of the dietary mutagen IQ. J. Am. Chem. Soc. 126, 1189−11201. (34) Milhe, C., Fuchs, R. P. P., and Lefevre, J. F. (1996) NMR data show that the carcinogen N-2-acetylaminofluorene stabilises an intermediate of −2 frameshift mutagenesis in a region of high mutation frequency. Eur. J. Biochem. 235, 120−127. (35) Gray, D. M., Ratliff, R. L., and Vaughan, M. R. (1992) Circular dichroism spectroscopy of DNA. Methods Enzymol. 211, 389−406. (36) Broyde, S., Wang, L., Rechkoblit, O., Geacintov, N. E., and Patel, D. J. (2008) Lesion processing: high-fidelity versus lesion-bypass DNA polymerases. Trends Biochem. Sci. 33, 209−219. (37) Ling, H., Boudsocq, F., Woodgate, R., and Yang, W. (2001) Crystal structure of a Y-family DNA polymerase in action: a 1355

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356

Article

Chemical Research in Toxicology mechanism for error-prone and lesion-bypass replication. Cell 107, 91−102. (38) Guengerich, F. P. (2006) Interactions of carcinogen-bound DNA with individual DNA polymerases. Chem. Rev. 106, 420−452. (39) Wong, J. H. Y., Brown, J. A., Suo, Z., Blum, P., Nohmi, T., and Ling, H. (2010) Structural insight into dynamic bypass of the major cisplatin-DNA adduct by Y-family polymerase Dpo4. EMBO J. 29, 2059−2069. (40) Kirouac, K. N., Basu, A. K., and Ling, H. (2013) Structural mechanism of replication stalling on a bulky amino-polycyclic aromatic hydrocarbon DNA adduct by a Y-family DNA polymerase. J. Mol. Biol. 425, 4167−4176. (41) Ling, H., Sayer, J. M., Plosky, B. S., Yagi, H., Boudsocq, F., Woodgate, R., Jerina, D. M., and Yang, W. (2004) Crystal structure of a benzo[a]pyrene diol epoxide adduct in a ternary complex with a DNA polymerase. Proc. Natl. Acad. Sci. U.S.A. 101, 2265−2269. (42) Kokoska, R. J., Bebenek, K., Boudsocq, F., Woodgate, R., and Kunkel, T. A. (2002) Low fidelity DNA synthesis by a Y family DNA polymerase due to misalignment in the active site. J. Biol. Chem. 277, 19633−19638. (43) Boudsocq, F., Iwai, S., Hanaoka, F., and Woodgate, R. (2001) Sulfolobus solfataricus P2 DNA polymerase IV (Dpo4): an archaeal DinB-like DNA polymerase with lesion-bypass properties akin to eukaryotic polh. Nucleic Acids Res. 29, 4607−4616. (44) Miller, H., and Grollman, A. P. (1997) Kinetics of DNA polymerase I (Klenow fragment exo−) activity on damaged DNA templates: effect of proximal and distal template damage on DNA synthesis. Biochemistry 36, 15336−15342. (45) Stover, J. S., Chowdhury, G., Zang, H., Guengerich, F. P., and Rizzo, C. J. (2006) Translesion synthesis past the C8- and N2deoxyguanosine adducts of the dietary mutagen 2-amino-3methylimidazo[4,5-f ]quinoline in the NarI recognition sequence by prokaryotic DNA polymerases. Chem. Res. Toxicol. 19, 1506−1517. (46) Prakash, S., Johnson, R. E., and Prakash, L. (2005) Eukaryotic translesion synthesis DNA polymerases: specificity of structure and function. Annu. Rev. Biochem. 74, 317−353. (47) Wang, L., and Broyde, S. (2006) A new anti conformation for N-(deoxyguanosin-8-yl)-2-acetylaminofluorene (AAF-dG) allows Watson-Crick pairing in the Sulfolobus solfataricus P2 DNA polymerase IV (Dpo4). Nucleic Acids Res. 34, 785−795. (48) Liu, Y., Reeves, D., Kropachev, K., Cai, Y., Ding, S., Kolbanovskiy, M., Kolbanovskiy, A., Bolton, J. L., Broyde, S., Van Houten, B., and Geacintov, N. E. (2011) Probing for DNA damage with β-hairpins: similarities in incision efficiencies of bulky DNA adducts by prokaryotic and human nucleotide excision repair systems in vitro. DNA Repair 10, 684−696. (49) Jain, V., Hilton, B., Lin, B., Patnaik, S., Liang, F., Darian, E., Zou, Y., MacKerell, A. D., and Cho, B. P. (2013) Unusual sequence effects on nucleotide excision repair of arylamine lesions: DNA bending/ distortion as a primary recognition factor. Nucleic Acids Res. 41, 869− 880. (50) Min, J.-H., and Pavletich, N. P. (2007) Recognition of DNA damage by the Rad4 nucleotide excision repair protein. Nature 449, 570−575.

1356

DOI: 10.1021/acs.chemrestox.5b00143 Chem. Res. Toxicol. 2015, 28, 1346−1356