Base Sequence-Dependent Bends in Site-Specific Benzo[a]pyrene

Jan 15, 1996 - The site specifically modified oligonucleotides 5'-d(TCCTCCTG1G2CCTCTC) (I) and 5'-d(CTATG1G2G3TATC) (II) were synthesized with single ...
17 downloads 4 Views 216KB Size
Download PDF
Chem. Res. Toxicol. 1996, 9, 255-261

255

Base Sequence-Dependent Bends in Site-Specific Benzo[a]pyrene Diol Epoxide-Modified Oligonucleotide Duplexes Tongming Liu,† Jing Xu,† Hong Tsao,† Bin Li,† Rong Xu,† Cuijian Yang,† Shantu Amin,‡ Masaaki Moriya,‡ and Nicholas E. Geacintov*,† Chemistry Department, 31 Washington Place, New York University, New York, New York 10003, and American Health Foundation, Valhalla, New York 10595 Received June 21, 1995X

The site specifically modified oligonucleotides 5′-d(TCCTCCTG1G2CCTCTC) (I) and 5′d(CTATG1G2G3TATC) (II) were synthesized with single modified guanine residues at positions G1, G2, or G3, derived from the covalent binding reaction of 7R,8S-dihydroxy-9S,10R-epoxy7,8,9,10-tetrahydrobenzo[a]pyrene ((+)-anti-BPDE) with the exocyclic amino groups of the guanine residues. In denaturing 20% polyacrylamide gels, the electrophoretic mobilities of the (+)-anti-BPDE-modified oligonucleotides I and II are slower than the mobilities of the respective unmodified oligonucleotides and independent of the positions of the BPDE-modified guanines. However, in the double-stranded forms in native 8% polyacrylamide gels, the electrophoretic mobilities of the duplexes with lesions at G2 or G3 are remarkably slower (reductions in mobilities up to ∼40%) than to duplexes with lesions at G1 and are attributed to physical bends or flexible hinge joints at the sites of the BPDE lesions. These sequencedependent mobility effects occur whenever the BPDE-modified guanine residues with (+)trans-stereochemistry are flanked by unmodified G’s on the 5′-side. These retarded electrophoretic mobilities are attributed to bending induced by steric hindrance effects involving the bulky 5′-flanking guanines and the pyrenyl residues that are known to point into the 5′-direction relative to the modified G [Cosman, M., et al. (1992) Proc. Natl. Acad. Sci. U.S.A. 89, 19141918]. These anomalous electrophoretic mobility effects are not observed in the case of (-)anti-BPDE-modified sequences I with trans-(-)-anti-BPDE-N2-dG adduct stereochemistry.

Introduction The mutagenic characteristics of r7,t8-dihydroxy-t9,10epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (racemic (()anti-BPDE),1 a biologically active metabolite of the ubiquitous environmental pollutant benzo[a]pyrene, have been investigated extensively (1-7). Typically, Yang et al. (4) found, in the supF tRNA gene of the pZ189 shuttle vector, that the majority of the observed base substitutions were G f T transversions, with lesser amounts of G f C and A f T transversions, and deletions. In the dihydrofolate reductase gene in Chinese hamster ovary cells (6), and in the hprt gene of mutants in cultured human lymphoblastoid cells (7), G f T transversions were also found to be the dominant base substitution mutations. Bernelot-Moens et al. (5) studied the mutations induced by (()-anti-BPDE in the lacI gene of excision repair-deficient (Uvr-) Escherichia coli; single base pair deletions and base substitutions were almost equal in number and composed 76% of all mutations, with G f T transversions again being the dominant base substitutions. The mutation spectra of the (+)-antiBPDE (8-11) and the (-)-anti-BPDE (8) enantiomers have been assessed in mammalian (8-10) and bacterial (11) cells. Rodriguez and Loechler (11) analyzed the mutations induced by (+)-anti-BPDE in the supF gene * Corresponding author. † New York University. ‡ American Health Foundation. X Abstract published in Advance ACS Abstracts, December 1, 1995. 1 Abbreviations: (+)-anti-BPDE, 7R,8S-dihydroxy-9S,10R-epoxy7,8,9,10-tetrahydrobenzo[a]pyrene; (-)-anti-BPDE, 7S,8R-dihydroxy9R,10S-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene; TEAA, triethylammonium acetate; AF, 2-aminofluorene; AAF, N-acetyl-2-aminofluorene.

0893-228x/96/2709-0255$12.00/0

of the E. coli plasmid pUB3 and found frameshift mutations, deletions, and various base substitutions at A and G sites, with G f T transversions being more abundant than other types of base substitutions; the distribution of these base substitution mutations was base sequence-dependent (11). Wei et al. (8) compared the mutations induced by the chiral (+)- and (-)-antiBPDE stereoisomers at the HPRT locus in Chinese hamster V-79 cells; the (+)-isomer was found to be more mutagenic than (-)-anti-BPDE, and the former showed a higher selectivity for G f T transversions than the (-)enantiomer. When anti-BPDE reacts chemically with DNA, the major reaction products result from covalent binding reactions via trans-addition of the exocyclic amino group of guanine to the C10 position of BPDE, but minor proportions of cis-N2-dG and trans-N6-dA adducts are also observed (12, 13). The observed mutation spectra are generally consistent with the dominant modes of reaction of both (+)- and (-)-anti-BPDE with guanine residues in DNA and, to a lesser extent, binding to adenine residues. In mutagenesis experiments in which the DNA is randomly modified (1-11), it is difficult, if not impossible, to correlate the observed genomic sites of mutations with the sites of covalent binding of anti-BPDE (14). The fact that higher frequencies of mutation are observed in some sequence contexts than in others is well-established (2, 4-11). For example, mutation hotspots were found at the middle base pair of GGG triplets with six other hotspots located at GC sequences of the supF gene (4). In the dhfr gene of Chinese hamster ovary cells (6), most of the G f T transversions occurred in purine-rich © 1996 American Chemical Society

256 Chem. Res. Toxicol., Vol. 9, No. 1, 1996

sequences (GGA, AGG, GGA, etc.). In exon 3 of the hprt gene in human lymphoblastoid cells, 16 hotspots of mutation were identified, with 6 of these occurring in the sequence GGGGG and 5 in the sequence GAAGAG (7). Hotspots of mutation were found in some, but not all GG sequences (the mutated G is underlined), especially in GGCC and GGG sequence contexts in the supF gene of the pUB3 plasmid treated with (+)-anti-BPDE (11). Using the optically resolved anti-BPDE isomers, hotspots of mutation for base substitutions occurred in GGTA, GCTG, CTGG, CGAG, GGCT, and GTGC ((+)-enantiomer) and in GGTG, GTGG, GTGC, and TGGG ((-)enantiomer) in the coding region of the hprt gene in V-79 cells (8). Clearly, runs of purines, especially runs of guanine, are associated with hotspots of mutation. The basis underlying these hotspot phenomena is presently not well understood (15). One possible explanation for the observations of mutation hotspots is that these same sequences are hotspots for covalent binding. By using photochemical strand scission techniques, Boles and Hogan reported a higher than average extent of covalent binding of racemic antiBPDE in runs of guanines (16) in the chicken adult β-globin gene. By using the same approach, Kootstra et al. (17) reported that a region containing the GC box consensus sequences (the motif GGGCGG) in the Chinese hamster ovary aprt gene was a hotspot for (()-anti-BPDE binding. Dittrich and Krugh used the photochemical mapping approach to map the (()-anti-BPDE binding patterns in enzyme restriction fragments of the plasmid pBR322 (18) and in another plasmid containing the same sequence as the human c-Ha-ras protooncogene (19); again, strand scission signaling the presence of covalently bound BPDE residues was found in G-rich sequences, but scission at pyrimidine sites was also observed (18, 19). It was shown recently that scission occurs not only at the BPDE-modified guanine residue but also at flanking unmodified bases, though at lower probabilities (20). The relationships, if any, between mutation hotspots and covalent binding hotspots thus have not yet been established. Site-directed mutagenesis experiments in vivo are, in principle, useful for establishing the effects of base sequence context on the mutagenic potential of a lesion. However, only a limited number of such studies with BPDE-modified guanine residues positioned in defined sequence contexts have been carried out so far (21-23). With a (+)-trans-anti-BPDE-N2-dG lesion (G*) in a TG*C sequence context in the E. coli plasmid pUC19, G f T transversions were observed almost exclusively (21), whereas in a CG*G sequence context in the supF gene of a pUB3 plasmid, G f C and G f A mutations were observed as well (22). Moriya et al. (23) investigated the mutagenic potentials of (+)- and (-)-trans-anti-BPDEN2-dG lesions positioned at either G1 or G2 in the sequence context TG1*G2C or TG1G2*C within the noncoding strand of the human c-Ha-ras protooncogene sequence; G f T transversions were dominant, although other base substitutions were observed at the G1* and G2* sites in varying proportions. The mutagenic potentials of the adducts derived from (+)-anti-BPDE and (-)anti-BPDE were generally different in E. coli than in simian kidney (COS) cells (23). As part of a long-range program to elucidate the structural features of adducts derived from the binding of (+)- and (-)-anti-BPDE to nucleic acids by NMR (2426) and other methods (27-29), we have synthesized site

Liu et al.

Figure 1. Structures of (+)-anti-BPDE, (-)-anti-BPDE, and the two trans-adducts that arise from the binding of the two anti-BPDE enantiomers to guanine residues in DNA.

specifically modified and stereochemically defined oligonucleotides with (+)-anti-BPDE-N2-dG lesions (Figure 1) positioned at different sites in sequences containing two or three contiguous guanine residues (30, 31). In this work, we show that the electrophoretic mobilities of BPDE-modified oligonucleotide duplexes with site specifically placed (+)-trans-anti-BPDE-N2-dG residues exhibit remarkable variations, depending on the position of the modified guanines in runs of two or three contiguous guanines. The anomalously slow electrophoretic mobilities are attributed to bends that are particularly pronounced whenever the (+)-trans-anti-BPDE-N2-dG lesions are flanked by guanine residues on the 5′-side. Interestingly, these sequence effects are not observed with the stereoisomeric (-)-trans-anti-BPDE-N2-dG (Figure 1) residues.

Materials and Methods Synthesis of anti-BPDE-Modified Oligonucleotide Adducts. Racemic anti-BPDE was synthesized according to published procedures (32). The (+)-anti-BPDE and (-)-antiBPDE enantiomers were separated from one another by HPLC using a 25 × 4.6 mm Regis Pirkle C D-phenylglycine (ionic) column (Regis Co., Morton Grove, IL), using separation protocols published by Weems and Yang (33). Caution: anti-BPDE is known to be mutagenic and carcinogenic in animal model systems and therefore should be handled with the utmost care. Two oligonucleotides, 5′-d(TCCTCCTGGCCTCTC) (I) and 5′d(CTATGGGTATC) (II), as well as their natural complementary strands, were purchased from Midland Certified Reagent Co. (Midland, TX). These oligonucleotides were repurified by reverse-phase HPLC methods (34) and desalted using a Sephadex G25 column. The synthesis and characterization of the (+)-anti-BPDEmodified oligonucleotide II with (+)-trans-anti-BPDE-N2-dG residues situated at either one of the three guanines G1, G2, and G3 have already been described in detail (31). The method of synthesis of the (+)- and (-)-anti-BPDEmodified oligonucleotides I was similar to the procedures previously described (27, 35). To summarize briefly, oligonucleotide I was dissolved in 0.75 mL of a 20 mM sodium phosphate buffer solution (strand concentration: 1.7 mM). About 100 µL of a 6 mM (+)- or (-)-anti-BPDE solution in tetrahydrofuran was added to the oligonucleotide solution; 13 µL of triethylamine (TEA) was added so that the pH of the solution was ∼11 (27). The reaction was allowed to proceed under gentle stirring for 24 h at 24 ( 1 °C in the dark. The solution was then extracted three times with water-saturated ether to remove the tetraol reaction products. This step was repeated one more time

Base Sequence Dependence of Bending

Chem. Res. Toxicol., Vol. 9, No. 1, 1996 257

Figure 2. Reverse-phase HPLC elution profile of (+)-antiBPDE-modified oligonucleotide I fraction (ODS column, 11-14% acetonitrile/TEAA buffer solution gradient). Only the eluates labeled I(G1+) and I(G2+) were identified (Table 1). in the case of (+)-anti-BPDE and seven more times in the case of (-)-anti-BPDE. The overall final molar ratio of (BPDE)/(DNA strand) was 1:1 in the case of (+)-anti-BPDE and 4:1 in the case of the (-)-enantiomer. The reaction mixtures were subjected to several successive HPLC separation procedures as described in the Results section. A Waters HPLC system (Millipore Corp., Milford, MA), including a Waters Model 991 photodiode array detector, was employed in these experiments. In the first step, the unmodified oligonucleotides were separated from the BPDEmodified ones by means of a semi-preparative 75 Å poly(styrene-divinylbenzene) reverse-phase PRP-1 HPLC column (Hamilton Co., Reno, NE). In subsequent steps, the different BPDE-modified oligonucleotides were separated from one another by using a Hypersil-ODS 5 µm 250 × 4.6 mm column (Keystone Scientific, Inc., Bellefonte, PA). Gel Electrophoresis. The (+)-BPDE-oligonucleotide adducts and unmodified oligonucleotides were labeled at the 5′ends with [32P]ATP purchased from New England Nuclear Corporation (Boston, MA), employing a T4 polynucleotide kinase 5′-terminus labeling system (Bethesda Research Laboratories, Gaithersburg, MD). The single-stranded oligonucleotide samples were purified by gel electrophoresis with 20% polyacrylamide denaturing gels (7 M urea) using a Pocker-Face II Model SE 1600 gel electrophoresis apparatus 40 cm long (Hoefer Scientific Instruments, San Francisco, CA); the applied voltage was 2500 V and the temperature was 45 °C. The bands were detected by standard autoradiography techniques using Kodak film No. XAR 5 (Eastman Kodak, Co., Rochester, NY). The double-stranded oligonucleotides were prepared by heating equimolar mixtures of the anti-BPDE-modified strands I and II and their complementary strands to 75 °C and by allowing the mixture to cool to 4 °C. The electrophoretic mobilities were determined using 8% native polyacrylamide gels at 4 °C.

Results Purification of anti-BPDE-Modified Oligonucleotides I. The singly modified (+)-anti-BPDE-modified strands I were separated from unmodified I by using the PRP-1 column and a linear 0-90% methanol/20 mM sodium phosphate buffer solution (pH 7.0) with a 2 mL/ min flow rate over 90 min. The fraction eluting between 39 and 41 min was collected and subjected to a second HPLC separation step (ODS column) using a linear 1014% acetonitrile/TEAA buffer solution (50 mM, pH 7.0) over 90 min (2 mL/min); this elution profile is shown in Figure 2. Two major products appear at 22 and 25 min with absorption spectra similar to those exhibited by other anti-BPDE-modified oligonucleotides (27, 30, 31, 35). Each of these products was subjected to a final HPLC purification step using the ODS column and a 1630% methanol/phosphate buffer gradient (2 mL/min, 60

Figure 3. Reverse-phase HPLC elution profile of enzyme digests (snake venom phosphodiesterase and bacterial phosphatase) of (A) unmodified oligonucleotide I, and (B) (+)-antiBPDE-modified oligonucleotide I(G2+). Elution conditions: 0-90% methanol/20 mM sodium phosphate buffer solution (pH 7.0), linear gradient over 90 min (1 mL/min). The order of elution (from left to right) is dC, dG, dT, and (+)-anti-BPDE-N2-dG mononucleoside adduct.

min). By using modified Maxam-Gilbert sequencing techniques (30, 36), we determined (23) that the 22 min fraction contains a product with the (+)-trans-antiBPDE-N2-dG lesion at the first guanine residue counted from the 5′-side; this BPDE-modified oligonucleotide is labeled I(G1+). The 25 min eluate contains the stereochemically identical lesion, but at guanine G2; this product is called I(G2+). Enzyme Digestion to Deoxynucleosides. To determine and confirm the nature of the BPDE-modified base, about 10 µg of each of the anti-BPDE-modified, purified oligonucleotides I was digested with snake venom phosphodiesterase and bacterial alkaline phosphatase (Pharmacia LKB Biotechnology Inc.), as described previously (27, 35); the nucleoside mixtures were then subjected to reverse-phase HPLC analysis using the Hypersil 5 µm ODS column. Typical elution profiles of the enzyme digestion products of unmodified I and the product I(G2+), employing an absorption detector at 254 nm, are shown in Figure 3A,B, respectively. The elution profile for the enzyme-digested I exhibits three maxima, as expected from the composition of I (Figure 3A); the order of elution of the nucleosides is dC, dG, and dT. An elution profile of the enzyme digest of the product I(G2+) is shown in Figure 3B; an identical result was obtained for the enzyme digest of product I(G1+) (data not shown). In Figure 3B, the amplitude of the unmodified dG elution maximum decreases, and another peak with the same elution time as an authentic (13) (+)-trans-anti-BPDEN2-dG mononucleoside adduct (data not shown) appears. The relative proportions of dC, dG, dT, and BPDEmodified 2′-deoxyguanosine (BPDE-N2-dG) were estimated as described previously (35). For the samples in Figure 3, the relative ratios of dC:dG:dT:BPDE-N2-dG are 8:2.1:4.5:0 for the digested, unmodified I and 8:1.2: 5.0:1.1 for the digested I(G2+) product. This latter result is within experimental error of the expected value of 8:1: 5:1.

258 Chem. Res. Toxicol., Vol. 9, No. 1, 1996

Figure 4. CD spectra of the late-eluting fraction of enzymedigested (+)- and (-)-anti-BPDE-modified oligonucleotides, obtained from the type of experiments described in Figure 3: fraction eluting at 84 min and containing the (+)-trans-antiBPDE-N2-dG mononucleoside adduct [from the enzyme digest of I(G2+)); (thick line); fraction eluting at 81 min and containing the (-)-trans-anti-BPDE-N2-dG mononucleoside adduct [from the enzyme digest of oligonucleotide I(G2-)] (dashed line).

Analogous results were obtained with (-)-anti-BPDEmodified oligonucleotide products I(G1-) and I(G2-). The only difference was the retention time of the modified nucleoside (-)-anti-BPDE-N2-dG obtained from the enzyme digests of I(G1-) and I(G2-), which eluted at 81 min (as compared to the ∼84 min elution time of the (+)trans-anti-BPDE-N2-dG mononucleoside adduct, Figure 3B); this mononucleoside adduct exhibited the same retention time as the authentic (-)-trans-anti-BPDEN2-dG mononucleoside adduct standard (data not shown). CD Spectra of anti-BPDE-dG Mononucleosides. The identities of the I(G1+), I(G2+), I(G1-), and I(G2-) products were verified further by examining the CD spectra of the late-eluting BPDE-dG mononucleoside products of the enzyme digestions. Examples of the CD spectra of these BPDE-N2-dG products obtained from the enzyme digests of products I(G2+) and I(G2-) are shown in Figure 4. The CD spectra correspond to those of authentic (+)-trans-anti-BPDE-N2-dG and (-)-transanti-BPDE-N2-dG mononucleoside adducts, respectively (13). Electrophoretic Mobilities of Duplexes Derived from I. The gel migration patterns of the single-stranded unmodified I and anti-BPDE-modified I on denaturing 20% polyacrylamide gels (7 M urea) are compared in Figure 5 (lanes 1-4). It is evident that the unmodified oligonucleotide migrates faster than all four of the modified ones [(I(G1+), I(G2+), I(G1-), and I(G2-)]. Furthermore, all four of the anti-BPDE-modified oligonucleotides exhibit approximately the same electrophoretic mobilities. The slower mobilities of the modified DNA sequences are due to the increased mass of these fragments (30). The gel migration patterns of these same oligonucleotides in the double-stranded forms in native 8% polyacrylamide gels are compared to one another in Figure 5, lanes 6-10. The mobilities of the two (-)-transmodified I(G1-) and I(G2-) duplexes (lanes 9 and 10) are equal to one another and are somewhat greater than the mobility of the I(G1+) duplex (lane 7), but slower than that of the unmodified I duplex (lane 6). In contrast to all of the other BPDE-modified I duplexes, the mobility of the I(G2+) duplex (lane 8) is remarkably slower (relative mobility only ∼60% of the mobility of the unmodified I duplex).

Liu et al.

Figure 5. Polyacrylamide gel electrophoresis of single-stranded and double-stranded unmodified and anti-BPDE-modified oligonucleotides I. Lanes 1-5 (20% polyacrylamide denaturing gels, single-stranded oligonucleotides); lane 1, unmodified I; lane 2, I(G1+); lane 3, I(G2+); lane 4, I(G1-); lane 5, I(G2-). Lanes 6-10 (8% native polyacrylamide gels): lane 10, unmodified I, lane 6, I(G1+); lane 7, I(G2+); lane 8, I(G1-); lane 9, I(G2-).

Figure 6. Polyacrylamide gel electrophoresis of single-stranded and double-stranded unmodified and anti-BPDE-modified oligonucleotides II. Lanes 1-4 (20% polyacrylamide denaturing gels, single-stranded oligonucleotides): lane 1, unmodified II; lane 2, II(G3+); lane 3, II(G2+); lane 4, II(G1+). Lanes 5-8 (8% native polyacrylamide gels): lane 5, unmodified II; lane 6, II(G3+); lane 7, II(G2+); lane 8, II(G1+).

Electrophoretic Mobilities of Duplexes Derived from II. The results obtained with anti-BPDE-modified duplexes I suggest that a G residue flanking the (+)trans-anti-BPDE-N2-dG lesions on the 5′-side gives rise to the observed, remarkable slowing in the electrophoretic mobility of this duplex. To explore the validity of this hypothesis further, we compared the electophoretic mobilities of duplexes derived from oligonucleotide II with (+)-trans-anti-BPDE lesions situated at either G1, G2, or G3. The relative mobilities of the single-stranded unmodified II (lane 1) and three (+)-anti-BPDE-modified II adducts (lanes 2-4) on denaturing gels are compared in Figure 6. All three (+)-anti-BPDE-modified singlestranded II exhibit approximately similar but slower mobilities than that of the unmodified single-stranded II. On native gels, however, the duplexes exhibit rather striking differences in their electrophoretic mobilities (lanes 6-8). Interestingly, Suh et al. (28) reported that, in native 20% polyacrylamide gels, the single-stranded oligonucleotides also exhibit the same order of relative mobilities as the double-stranded forms in native 8% gels (Figure 6). The results of Suh et al., obtained with singlestranded oligonucleotides, may be due to secondary

Base Sequence Dependence of Bending

Chem. Res. Toxicol., Vol. 9, No. 1, 1996 259

Table 1. Relative Electrophoretic Mobilities of Unmodified and anti-BPDE-Modified Duplexes I and II with Their Natural Complementary Strands in Native 8% Polyacrylamide Gels (Data from Figures 5 and 6) oligonucleotide adducta

relative mobility (%) (unmodified ) 100%) single-stranded formb

relative mobility (%) (unmodified ) 100%) double-stranded formc

I(G1+) I(G2+) I(G1-) I(G2-) II(G1+) II(G2+) II(G3+)

92 92 92 92 92 93 92

96 59 99 99 98 88 82

a Sequence I: 5′-d(TCCTCCTG G CCTCTC). Sequence II: 5′1 2 d(CTATG1G2G3TATC). The different stereochemically distinct and site specifically modified oligonucleotide adducts are specified by either I or II to designate the sequence, with the modified G residue in parentheses. The superscripts “+” or “-” designate the (+)-trans- or (-)-trans-adduct stereochemistry. b Denaturing gels. c Native gel.

structures formed in the native gels, since these effects are not observed in denaturing 20% polyacrylamide gels (Figure 6). The II(G1+) duplex (lane 8) is characterized by a mobility that is somewhat slower than that of the unmodified II duplex (lane 5). We characterize this difference in mobility as “normal”, since (+)-anti-BPDEmodified oligonucleotide duplexes with one single guanine residue per oligonucleotide exhibit similar behavior (37). In contrast, the II(G2+) (lane 7) and II(G3+) duplexes (lane 6) exhibit anomalously slow electrophoretic mobilities; in both cases, the (+)-trans-anti-BPDE-N2-dG adducts are flanked by guanines on the 5′-side. The mobilities of the anti-BPDE-modified I and II single strands and duplexes, expressed relative to the mobilities of the unmodified sequences I and II, respectively, are summarized in Table 1.

Discussion Adduct-Induced DNA Bending. Some covalently bound ligands are known to cause significant local structural distortions in the modified DNA molecules. For example, the carcinogens 2-aminofluorene (AF) and N-acetyl-2-aminofluorene (AAF) cause bends or flexible hinge joints when covalently bound to double-stranded oligonucleotides (38, 39). Drug molecules that are known to cause bending upon binding covalently to DNA include cis-Pt (40) and CC-1065 (41). Adduct-induced structural distortions such as bends or flexible hinge joints result in smaller average end-to-end distances of the DNA duplexes, which, in turn, give rise to reduced electrophoretic mobilities (42). Earlier results obtained with randomly modified native DNA (43, 44) and synthetic polynucleotides (44, 45) showed that covalently bound anti-BPDE residues cause bends or kinks and a lowering the average persistence length of the modified doublestranded DNA (37, 44, 45). Recently, we have shown that oligonucleotide duplexes containing site-specific antiBPDE-N2-dG lesions with (+)-trans-adduct stereochemistry [oligonucleotide sequence 5′-d(CACAT[G*]TACAC) in a duplex with an unmodified complementary strand, where G* denotes the modified guanine residue] are significantly more bent than those with (-)-trans-, (+)cis-, or (-)-cis-stereochemistry (37). In this work we show that the degree of bending in site specifically modified oligonucleotide duplexes, as measured by high-resolution native polyacrylamide gel electrophoresis, depends not

only on the stereochemistry but also on the position of the anti-BPDE-N2-dG lesions in contiguous guanine sequences. However, these remarkable decreases in the electrophoretic mobilities are observed only in the case of (+)-trans- but not (-)-trans-adduct stereochemistry, and then only in those modified oligonucleotide sequences that have another guanine on the 5′-side of the (+)-antiBPDE-modified guanine residue. BPDE-N2-dG Adduct Configuration and Base Sequence-Dependent Bending. A plausible hypothesis that may account for the origins of these effects is based on the known orientations of the pyrenyl residues relative to the 5′ f 3′ strand polarity and the modified guanines in double-stranded DNA. In double-stranded (24, 25) and single-stranded (46) oligonucleotides containing anti-BPDE-N2-dG lesions with (+)-trans-adduct stereochemistry, the pyrenyl residues are tilted toward the 5′-end of the modified strand. In the (-)-trans-adduct case, the pyrenyl residue points toward the 3′-end of the modified strand rather than toward the 5′-end (25, 46). It was shown some time ago that BPDE-DNA adduct conformations can be spectroscopically classified into two groups: site I and site II (47). Intercalative conformations with considerable carcinogen-nucleic base stacking conformations are of the site I type, while external adduct conformations, e.g., the minor groove binding sites of the two trans-BPDE-N2-dG adducts (27, 45), are classified as site II. It is convenient to use these classifications to delineate differences in adduct conformations. There are experimental indications that the minor groove conformations of the pyrenyl residues in adducts with (+)-trans-adduct stereochemistry are significantly perturbed when a bulky guanine residue flanks the antiBPDE-N2-dG adduct on the 5′-side. By using lowtemperature fluorescence methods, Suh et al. (28) were able to distinguish between external site II adduct conformations and site I structures with significant pyrenyl-base stacking interactions in the duplexes II(G1+), II(G2+), and II(G3+). The duplex II(G1+), with the pyrimidine T on the 5′-side of the lesion is characterized by a dominant site II conformation. In contrast, the duplexes II(G2+) and II(G3+), both with G’s flanking the modified guanine residues, exhibit mixed site I and site II conformations. We hypothesize that the effects of 5′flanking G residues on the conformations of the pyrenyl residues are related to a steric hindrance effect. The presence of the bulkier guanine residues on the same strand flanking the (+)-anti-BPDE-modified G on the 5′side may interfere with the alignment of the pyrenyl residue in the minor groove. With less bulky C or T bases flanking the lesion on the 5′-side, the pyrenyl residue more easily assumes a minor groove site II conformation, as observed by NMR (24, 25) and by low-resolution optical spectroscopic methods (27-29). Taken together, the results in Table 1 and those of Suh et al. (28) indicate that 5′-guanines flanking (+)-transanti-BPDE-N2-dG lesions induce structural distortions in the duplexes that result in significant bending and, thus, a lowering of their electrophoretic mobilities. In contrast, guanine residues flanking these same lesions on the 3′-side do not appear to exert any unusual or detectable influence on the electrophoretic mobilities of anti-BPDE-modified oligonucleotide duplexes. While the pyrenyl residues in the (-)-trans-adduct duplexes point toward the 3′-direction of the modified strands (25), guanine residues flanking the (-)-transBPDE-N2-dG lesions on the 3′-side do not appear to

260 Chem. Res. Toxicol., Vol. 9, No. 1, 1996

influence the electrophoretic mobilities of the duplexes. This lack of bending associated with the (-)-anti-BPDEmodified adducts indicates that the conformations of the (+)-trans- and (-)-trans-anti-BPDE-N2-dG adducts are not merely symmetrical around the modified guanines, but differ in tertiary structural characteristics. Bent BPDE-DNA Adducts and Possible Biological Consequences. Oligonucleotide I, with BPDE lesions at either sites G1 or G2, constitutes a portion of the noncoding strand of the human c-Ha-ras 1 protooncogene encompassing codon 61 (in boldface type), as shown:

60 5′-d(....GCC d(....CCG2

61 CAG....) G1TC....)-5′

Liu et al.

adduct stereochemistry could play a role in this phenomenon as well. In closing, we note that research with site specifically modified PAH diol epoxide-oligonucleotide probes is now only in its early stages. The correlation of structural data with biological effects should lead to a better understanding of the mechanisms of mutations induced by this class of environmentally important chemicals.

Acknowledgment. This work was supported by the Office of Health and Environmental Research, Department of Energy (DE-FGO2-86ER-60405). The synthesis of the BPDE-modified oligonucleotides was supported by Grant CA 20851 from the National Cancer Institute, National Institutes of Health.

References Vousden et al. (48) showed that racemic anti-BPDE causes mutations in codon 61 of the human H-ras 1 protooncogene. In codon 61, the G:C f T:A transversion was the major mutation observed at the first position of this codon (corresponding to position G1 in the oligonucleotide I). Mutations at position G2 in codon 60 of the ras gene were not probed in that experiment. Moriya et al. (23) recently incorporated the BPDE-modified oligonucleotides I(G1+), I(G2+), I(G1-), and I(G2-) into a singlestranded shuttle vector and investigated the site-specific mutagenic response of these lesions expressed in E. coli and simian kidney (COS) cells; the frequencies of mutation and mutagenic specificities were quite different in these bacterial and mammalian cells. In the COS cells, the mutation frequency was about 3 times greater in the case of the I(G2+) rather than the I(G1+) adduct; G f T transversions were the dominant types of mutations observed in all cases, but in the case of the (+)-transanti-BPDE lesion at G2, a significant proportion of G f A and G f C mutations were also observed in COS cells. This observation is consistent with the notions that the adduct conformations are more heterogeneous when guanine residues flank the (+)-trans-anti-BPDE-N2-dG moieties on the 5′-side (28) and that different BPDE adduct conformations may lead to different types of mutations (49). However, this difference in mutational specificities was not observed for these same vectors containing either the sequences I(G2+) or I(G1+) and expressed in E. coli cells (23). These differences suggest that the same lesions are processed quite differently by different polymerases, as shown earlier (50). It remains to be established whether the sequence-dependent bending effects described here also manifest themselves in the DNA-replication enzyme complexes and thus play a role in the reported mutation spectra and hotspot effects (10, 11, 15). Sequence-dependent adduct polymorphism, or heterogeneities in the conformations of chemically homogeneous BPDE-N2-dG adducts (28), may contribute to the diversity of point mutations in site-specific mutagenicity experiments, as discussed by Loechler and his co-workers (15, 22, 49). MacLeod et al. (51) recently reported that the transcription factor Sp1 binds more strongly to (()-antiBPDE-modified GC box sequences in the promoter of the hamster adenosine phosphoribosyl transferase gene than to the same unmodified sequences; the 5-10-fold enhancement in the binding of Sp1 was attributed to the existence of static bends or induced flexibility at the sites of anti-BPDE binding. The effects of base sequence and

(1) Wood, A. W., Chang, R. L., Levin, W., Yagi, H., Thakker, D. R., Jerina, D. M., and Conney, A. H. (1977) Differences in mutagenicity of the optical enantiomers of the diastereomeric benzo[a]pyrene 7,8-diol-9,10-epoxides. Biochem. Biophys. Res. Commun. 77, 1389-1396. (2) Eisenstadt, E., Warren, A. J., Porter, J., Atkins, D., and Miller, J. H. (1982) Carcinogenic epoxides of benzo[a]pyrene and cyclopenta[cd]pyrene induce base substitutions via specific transversions. Proc. Natl. Acad. Sci. U.S.A. 79, 1945-1949. (3) Brookes, P., and Osborne, M. R. (1982) Mutation in mammalian cells by stereoisomers of anti-benzo[a]pyrene diol epoxide in relation to the extent and nature of the DNA reaction products. Carcinogenesis 3, 1223-1226. (4) Yang, J.-L., Maher, V. M., and McCormick, J. J. (1987) Kinds of mutations formed when a shuttle vector containing adducts of (()-7β,8R-dihydroxy-9R,10R-epoxy-7,8,9,10- tetrahydrobenzo[a]pyrene replicates in human cells. Proc. Natl. Acad. Sci. U.S.A. 84, 3787-3791. (5) Bernelot-Moens, C., Glickman, B. W., and Gordon, A. J. E. (1990) Induction of specific frameshift and base substitution events by benzo[a]pyrene diol epoxide in excision-repair-deficient Escherichia coli. Carcinogenesis 11, 781-785. (6) Carothers, A. M., and Grunberger, D. (1990) DNA base changes in benzo[a]pyrene diol epoxide-induced dihydrofolate reductase mutants of Chinese hamster ovary cells. Carcinogenesis 11, 189192. (7) Keohavong, P., and Thilly, W. G. (1992) Mutational spectrometry: a general approach for hot-spot point mutations in selectable genes. Proc. Natl. Acad. Sci. U.S.A. 89, 4623-4627. (8) Wei, S.-J. C., Chang, R. L., Hennig, E., Cui, X. X., Merkler, K. A., Wong, C.-Q., Yagi, H., and Conney, A. H. (1994) Mutagenic selectivity at the HPRT locus in V-79 cells: comparison of mutations caused by bay-region benzo[a]pyrene 7,8-diol-9,10epoxide enantiomers with high and low carcinogenic activity. Carcinogenesis 15, 1729-1735. (9) Wei, S. J. C., Chang, R. L., Wong, C.-Q., Bhachech, N., Cui, X. X., Hennig, E., Yagi, H., Sayer, J. M., Jerina, D. M., Preston, B. D., and Conney, A. H. (1991) Dose-dependent differences in the profile of mutations induced by an ultimate carcinogen from benzo[a]pyrene. Proc. Natl. Acad. Sci. U.S.A. 88, 11227-11230. (10) Wei, S. J. C., Chang, R. L., Bachech, N., Cui, X. X., Merkler, K. A., Wong, C. Q., Hennig, E., Yagi, H., Jerina, D. M., and Conney, A. H. (1993) Dose-dependent differences in the profile of mutations induced by (+)-7R,8S-dihydroxy-9S,10R-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene in the coding region of the hypoxanthine (guanine) phosphoribosyltransferase gene in Chinese hamster V-79 cells. Cancer Res. 53, 3294-3301. (11) Rodriguez, H., and Loechler, E. L. (1993) Mutational specificity of the (+)-anti-diol epoxide of benzo[a]pyrene in a supF gene of an Escherichia coli plasmid: DNA sequence context influences hotspots, mutagenic specificity and the extent of SOS enhancement of mutagenesis. Carcinogenesis 14, 373-383. (12) Meehan, T., and Straub, K. (1979) Double stranded DNA stereoselectively binds benzo[a]pyrene diol epoxides. Nature 277, 410412. (13) Cheng, S. C., Hilton, B. D., Roman, J. M., and Dipple, A. (1989) DNA adducts from carcinogenic and noncarcinogenic enantiomers of benzo[a]pyrene dihydrodiol epoxide. Chem. Res. Toxicol. 2, 334-340. (14) Reardon, D. B., Bigger, C. A. H., Strandberg, J., Yagi, H., Jerina, D. M., and A. Dipple (1989) Sequence selectivity in the reaction of optically active hydrocarbon dihydrodiol epoxides with rat h-ras DNA. Chem. Res. Toxicol. 2, 12-14.

Base Sequence Dependence of Bending (15) Rodriguez, H., and Loechler, E. L. (1993) Mutagenesis by the (+)anti-diol epoxide of benzo[a]pyrene: what controls mutagenic specificity? Biochemistry 32, 1759-1769. (16) Boles, T. C., and Hogan, M. E. (1986) High-resolution mapping of carcinogen binding sites on DNA. Biochemistry 25, 3039-3043. (17) Kootstra, A., Lew, L. K., Nairn, R. S., and MacLeod, M. C. (1989) Preferential modification of GC boxes by benzo[a]pyrene-7,8-diol9,10-epoxide. Mol. Carcinogenesis 1, 239-244. (18) Dittrich, K. A., and Krugh, T. R. (1991) Analysis of site-specific binding of (()-anti-benzo[a]pyrene diol epoxide to restriction fragments of pBR322 DNA via photochemical mapping. Chem. Res. Toxicol. 4, 270-276. (19) Dittrich, K. A., and Krugh, T. R. (1991) Mapping of (()-anti-benzo[a]pyrene diol epoxide adducts to human c-Ha-ras1 protooncogene. Chem. Res. Toxicol. 4, 277-281. (20) Li, B., Mao, B., Liu, T.-M., Xu, J., Dourandin, S., Amin, S., and Geacintov, N. E. (1995) Laser pulse-induced photochemical strand cleavage of site-specifically and covalently modified (+)-antibenzo[a]pyrene diol epoxide-oligonucleotide adducts. Chem. Res. Toxicol. 8, 396-402. (21) Mackay, W., Benasutti, M., Drouin, E., and Loechler, E. L. (1992) Mutagenesis by (+)-ianti-B[a]P-N2-Gua, the major adduct of activated benzo[a]pyrene, when studied in Escherichia coli plasmid using site-directed methods. Carcinogenesis 13, 1415-1425. (22) Jelinsky, S. A., Liu, T., Geacintov, N. E., and Loechler, E. L. (1995) The major, N2-Gua adduct of the (+)-anti-benzo[a]pyrene diol epoxide is capable of inducing G f A and G f C, in addition to G f T mutations. Biochemistry 34, 13545-13553. (23) Moriya, M., Spiegel, S., Fernandes, A., Amin, S., Liu, T.-M., Geacintov, N. E., and A. P. Grollman (1995) Translesional synthesis past benzo[a]pyrene diol epoxide-2′-deoxyguanosine DNA adducts: Marked effects of host cell, sequence context, and chirality. Submitted. (24) Cosman, M., de los Santos, C., Fiala, R., Hingerty, B. E., Ibanez, V., Margulis, L. A., Live, D., Geacintov, N. E., Broyde, S., and Patel, D. J. (1992) Solution conformation of the major adduct between the carcinogen (+)-anti-benzo[a]pyrene diol epoxide and DNA. Proc. Natl. Acad. Sci. (U.S.A.) 89, 1914-1918. (25) de los Santos, Cosman, M., Hingerty, B. E., Ibanez, V., Margulis, L. A., Geacintov, N. E., Broyde, S., and Patel, D. J. (1992) Influence of benzo[a]pyrene diol epoxide chirality on solution conformations of DNA covalent adducts: the (-)-trans-anti[BP]G‚C adduct structure and comparison with the (+)-trans-antienantiomer. Biochemistry 31, 5245-5252. (26) Cosman, M., de los Santos, C., Fiala, R., Hingerty, B. E., Ibanez, V., Luna, E., Harvey, R. G., Geacintov, N. E., Broyde, S., and Patel, D. (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. (27) Geacintov, N. E., Cosman, M., Mao, B., Alfano, A., Ibanez, V., and Harvey, R. G. (1991) Spectroscopic characteristics and site I/site II classification of cis and trans benzo[a]pyrene diol epoxide enantiomer-guanosine adducts in oligonucleotides and polynucleotides. Carcinogenesis 12, 2099-2108. (28) Suh, M., Jankowiak, R., Ariese, F., Mao, B., Geacintov, N. E., and Small, G. J. (1994) Flanking base effects on the structural conformation of the (+)-trans-anti-benzo[a]pyrene diol epoxide adduct to N2-dG in sequence defined oligonucleotides. Carcinogenesis 15, 2891-2898. (29) Ya, N.-Q., Smirnov, S., Cosman, M., Bhanot, S., Ibanez, V., and Geacintov, N. E. (1994) Thermal stabilities of benzo[a]pyrene diol epoxide-modified oligonucleotide duplexes. Effects of mismatched complementary strands and bulges. In Structural Biology: The State of the Art. Proceedings of the 8th Conversation (Sarma, R. H., Sarma, M. H., Eds.) Vol. 2, pp 349-366, Adenine Press, Schenectady, NY. (30) Mao, B., Margulis, L. A., Li, B., Ibanez, V., Lee, H., Harvey, R. G., and Geacintov, N. E. (1992) Direct synthesis and identification of benzo[a]pyrene diol epoxide-deoxyguanosine binding sites in modified oligodeoxynucleotides. Chem. Res. Toxicol 5, 773-778. (31) Mao, B., Xu, J., Li, B., Margulis, L. A., Smirnov, S., Ya, N. Q., Courtney, S., and Geacintov, N. E. (1995) Synthesis and characterization of covalent adducts derived from the binding of benzo[a]pyrene diol epoxide to a -GGG- sequence in a deoxyoligonucleotide. Carcinogenesis 16, 357-365. (32) Yagi, H., Thakker, D. R., Hernandez, O., Koreeda, M., and Jerina, D. M. (1977) Synthesis and reactions of the highly mutagenic 7,8-

Chem. Res. Toxicol., Vol. 9, No. 1, 1996 261

(33)

(34)

(35)

(36) (37)

(38) (39) (40)

(41) (42) (43) (44)

(45)

(46)

(47)

(48)

(49)

(50) (51)

diol-9,10-epoxides of the carcinogen benzo[a]pyrene. J. Am. Chem. Soc. 99, 1604-1611. Weems, H. B., and Yang, S. K. (1989) Chiral stationary phase high-performance liquid chromatographic resolution of enantiomeric benzo[a]pyrene diol-epoxides and tetrols. Chirality 1, 276283. McLaughlin, L. W., and Piel, N. (1984) Chromatographic purification of synthetic oligonucleotides. In Oligonucleotide Synthesis: A Practical Approach (Gait, M. J., Ed.) pp 117-133, IRL Press, Oxford. Cosman, M., Ibanez, V., Geacintov, N. E., and Harvey, R. G. (1990) Preparation and isolation of adducts in high yield derived from the binding of two benzo[a]pyrene-7,8-dihydroxy-9,10-oxide stereoisomers to the oligonucleotide d(ATATGTATA) Carcinogenesis 11, 1667-1672. Maxam, A. M., and Gilbert, W. (1980) Sequencing end-labeled DNA with base-specific chemical cleavage. Methods Enzymol. 65, 499-560. Xu, R., Mao, B., Xu, J., Li, B., Birke, S., Swenberg, C. E., and Geacintov, N. E. (1995) Stereochemistry-Dependent Bending in Oligonucleotide Duplexes Induced by Site-Specific Covalent Benzo[a]pyrene Diol Epoxide-Guanine Lesions. Nucleic Acids Res. 23, 2314-2319. Schwartz, A., Marrot, L., and Leng, M. (1989) The DNA bending by acetylaminofluorene residues and by apurinic sites. J. Mol. Biol. 207, 445-450. Leng, M. (1990) DNA bending induced by covalently bound drugs. Gel electrophoresis and chemical probe studies. Biophys. Chem. 35, 155-163. Bellon, S. F., Coleman, J. H., and Lippard, S. J. (1991) DNA unwinding produced by site-specific intrastrand cross-links of the antitumor drug cis-diamminedichloroplatinum (II). Biochemistry 30, 8026-8035. Lin, C. H., Sun, D., and Hurley, L. H. (1991) (+)-CC-1065 produces bending of DNA that appears to resemble adenine/thymine tracts. Chem. Res. Toxicol. 4, 21-26. Lerman, L. S., and Frisch, H. L. (1982) Why does the electrophoretic mobility of DNA in gels vary with the length of the molecule? Biopolymers 21, 995-997. Hogan, M. E., Dattagupta, N., and Whitlock, J. P., Jr. (1981) Carcinogen-induced alteration of DNA structure. J. Biol. Chem. 256, 4504-4513. Eriksson, M., Norde´n, B., Jernstro¨m, B., and Gra¨slund, A. (1988) Binding geometries of benzo[a]pyrene diol epoxide isomers covalently bound to DNA. Orientational distribution. Biochemistry 27, 1213-1221. Roche, C. J., Jeffrey, A. M., Mao, B., Alfano, A., Kim, S. K., Ibanez, V., and Geacintov, N. E. (1991) Dependence of conformations of benzo[a]pyrene diol epoxide-DNA adducts derived from stereoisomers of different tumorigenicities on base sequence. Chem. Res. Toxicol. 4, 311-317. Mao, B., Li, B., Amin, S., Cosman, M., and Geacintov, N. E. (1993) Opposite stereoselective resistance to digestion by phosphodiesterases I and II of benzo[a]pyrene diolepoxide-modified oligonucleotide adducts. Biochemistry 32, 11785-11793. Geacintov, N. E., Gagliano, A. G., Ibanez, V., and Harvey, R. G. (1982) Spectroscopic characterizations and comparisons of the structures of the covalent adducts derived from the reactions of 7,8-dihydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene-9,10-oxide, and the 9,10-epoxides of 7,8,9,10-tetrahydrobenzo[a]pyrene and 9,10,11,12-tetrahydrobenzo[e]pyrene with DNA. Carcinogenesis 3, 247-253. Vousden, K. H., Bos, J. L., Marshall, C. J., and Phillips, D. H. (1986) Mutations activating human c-Ha-ras1 protooncogene (HRAS1) induced by chemical carcinogens and depurination. Proc. Natl. Acad. Sci. U.S.A. 83, 1222-1226. Drouin, E. E, and Loechler, E. L. (1993) AP sites are not significantly involved in mutagenesis by the (+)-anti-diol epoxide of benzo[a]pyrene: the complexity of its mutagenic specificity is likely to arise from adduct conformational polymorphism. Biochemistry 32, 6555-6562. Shibutani, S., Takeshita, M., and Grollman, A. P. (1991) Insertion of specific bases during DNA synthesis past the oxidationdamaged base 8-oxodG. Nature 349, 431-434. MacLeod, M. C., Powell, K. L., and Tran, N. (1995) Binding of the transcription factor, Sp1, to non-target sites in DNA modified by benzo[a]pyrene diol epoxide. Carcinogenesis 16, 975-983.

TX9501086