Chem. Res. Toxicol. 1995,8,591-599
591
In Vitro Reaction with DNA of the Fjord-RegionDiol Epoxides of Benzo[$]chrysene and Benzo[c]phenanthrene As Studied by 32P-Postlabeling ~
Alison S. Giles,*$:Albrecht Seide1,t and David H. Phillips' Haddow Laboratories, Znstitute of Cancer Research, Cotswold Road, Sutton, Surrey, SM2 5NG U.K., and Znstitute of Toxicology, University of Mainz, Obere Zahlbacher Strasse 67,D-55131 Mainz, FRG Received October 14, 1994@ The chemical reactivities of the optically-pure fjord-region syn- and anti-benzolglchrysene 11,12-dihydrodiol 13,14-epoxides (BkICDEs) toward DNA in vitro have been compared with those of the optically-pure fjord-region syn- and anti-benzo[clphenanthrene3,4-dihydrodiol1,2epoxides (B[c]PhDEs), using the standard 32P-postlabelingassay. The (+)-anti-, (+)-syn-,and (-)-syn-isomers of the two sets of diol epoxides showed similar extents of reaction with DNA, but the (-)-anti-B[clPhDE was 2.5 times more reactive toward DNA than the corresponding BlglCDE isomer and was the most reactive of the eight diol epoxides studied. When the reactions of the BkICDEs with DNA were analyzed by the nuclease P1-enhanced method of 32P-postlabeling,the observed adduct levels were between 3 and 10 times lower than were obtained using the standard method of 32P-postlabeling. By analyzing by TLC and HPLC the 32P-postlabeledproducts of the reactions of the diol epoxides with synthetic polynucleotides, the relative reactions of the BlglCDEs and B[clPhDEs with guanine and adenine bases in DNA were determined. All four B[glCDE isomers reacted with adenine residues in similar proportions to those seen for the BkIPhDE isomers. Thus, the presence of a n additional benzene ring on the benzo[clphenanthrene structure, distant from the fjord region, does not radically alter the reactivity or base selectivity of the fjord-region diol epoxides, except in the case of the (-)-anti-isomer of benzolglchrysene. The reasons for the lower reactivity of this isomer compared with that of the corresponding isomer of benzo[c]phenanthrene are unclear. The analyses demonstrate the importance of the use of two chromatographic systems for the identification of hydrocarbon-DNA adducts, and the requirement of more than one TLC solvent system for adequate separation of all of the adducts formed by the benzolglchrysene diol epoxides with DNA.
Introduction Polycyclic aromatic hydrocarbons (PAHs)l are ubiquitous environmental pollutants (1)that exert their mutagenic and carcinogenic effect after metabolic activation (2, 3). Studies of over a dozen PAHs have either identified or implicated bay-region diol epoxides as the ultimate carcinogenic metabolites (3-6). These electrophiles react with nucleophilic sites in cells to form adducts (7, 8). It is the formation of carcinogenDNA adducts that is thought to initiate carcinogenesis (8). The bay-region diol epoxide metabolites of a PAT3 exist as a pair of diastereomers in which the benzylic hydroxyl group is either syn or anti to the epoxide oxygen. Each diastereomer exists as a pair of optically active enanti-
' Institute of Cancer Research. University of Mainz. Abstract published in Advance ACS Abstracts, May 1, 1995. Abbreviations: B[alADE, benz[alanthracene 3,4-dihydrodiol 1,Zepoxide; B[a]P, benzo[a]pyrene; B[alPDE, benzo[alpyrene 7,8-dihydrodiol9,lO-epoxide; B[clC, benzo[clchrysene; BIcIPh, benzorclphenanthrene; B[c]PhDE, benzo[c]phenanthrene 3,4-dihydrodioll,Z-epoxide; BlgIC, benzo[g]chrysene; BLgICDE, benzo[glchrysene 11,lZ-dihydrodiol 13,14-epoxide; ChDE, chrysene l,Z-dihydrodiol3,4-epoxide; DMBADE, 7,12-dimethylbenz[a]anthracene3,4-dihydrodiol 1,a-epoxide; dAp, 2'deoxyadenosine 3'-monophosphate; PAH, polycyclic aromatic hydrocarbon; poly(dA), polydeoxyadenylic acid; poly(dAdT1, poly(deoxyadenylic-deoxythymidylic acid); poly(dGdC1, poly(deoxyguany1ic-deoxycytidylic acid); poly(dG-mebdC), poly(deoxyguanylic-Me5-deoxycytidylic acid). @
omers. The syn-enantiomers of these bay-region diol epoxides prefer a conformation in which their hydroxyl groups are pseudodiaxial, whereas the anti-enantiomers prefer a pseudodiequatorial hydroxyl group conformation. The anti-enantiomers of the bay-region diol epoxides of benzo[a]pyrene (B[a]PDE), of benz[alanthracene (B[alADE), and of chrysene (ChDE) are markedly more mutagenic in mammalian cells (9-1 11, and tumorigenic in newborn mice (12-14), than are the syn-enantiomers. Absolute, as well as relative, configuration also has an effect on the biological activity of the diol epoxides. Thus, virtually all of the tumorigenic activity of the anti-B[alPDEs, B[a]ADEs, and ChDEs resides in the (+)-antienantiomer with (RS)-diol (SR)-epoxide absolute configuration (15-17). Benzo[clphenanthrene(B[clPh),on the other hand, has a sterically hindered bay region, termed a fjord region (18). The fjord region causes the molecule to be nonplanar. This results in both the syn- and anti-3,4-dihydrodiol 1,2-epoxide fjord-region diol epoxide diastereomers (B[c]PhDE) (Chart 1) adopting a preferred pseudodiequatorial conformation for the hydroxyl groups (19). The anti-diol epoxide with (RS)-diol (SR)-epoxide absolute configuration (la in Chart 1) is again the most biologically active of the four fjord-region diol epoxides, but all four are highly mutagenic to bacteria and mammalian cells (20)and are the most potent tumor initiators of all PAH diol epoxides thus far tested (21).
0893-228x/95/2708-0591$09.00/0 0 1995 American Chemical Society
Giles et al.
592 Chem. Res. Toxicol., Vol. 8, No. 4, 1995
Chart 1. Structures of the Four Configurational Isomers of B[c]PhDE
7 6 (-)-anti-
(-)-syn-
4R,35,2R,15
4R, 35,25, 1 R
1c. W H
1d. W H OH
/ \
/
\
\
/
/
OH
(+)-syn-
(+)-anti45,3R, 2R,15
45,3R, 25,1 R
Chart 2. Structures of the Four Configurational Isomers of B[glCDE
Sa. (-)-anfi-
I
11R,125,13R,145 HO-
0 /
Materials and Methods
(-)-syn-
11R,125,135,14R
\
I
Hn\"'
\
- L %
II
2c.
/
I
residues in DNA must be accommodated in the minor groove of DNA, which is structurally difficult for a nonplanar molecule. Adducts with adenine residues are accommodated in the wide groove of DNA, which is possible for nonplanar molecules. Extensive reaction with both adenine and guanine residues in DNA has also been observed with the sterically-hindered bayregion diol epoxides of the potent carcinogen, 7,12dimethylbenz[alanthracene (DMBADE) (26). This molecule is also distorted from planarity. It has been postulated that there is a strong association between reactivity toward adenine in DNA and tumor-initiating activity, linked to the conformation of the bay-region diol epoxide (25, 27). The aim of the present paper was to study further the effect of distortion from planarity of a PAH on the behavior of its fjord-region diol epoxides. We have used 32P-postlabelingto study relative extents of formation of adducts in DNA by the four individual fjord-region 11,12-dihydrodiol 13,14-epoxides of B[glC in uitro. By using synthetic polynucleotides and analysis of adducts by TLC and HPLC, we have determined the nature of the adducts, and the relative extents of reaction of each of the diol epoxides with guanine and adenine bases in DNA. Throughout the study, we have carried out parallel experiments with the four B[clPhDE isomers, as a comparison.
2d. (+)-anti-
(+)-syn-
115,12R,13R,145
11S,12R,135,14R
Benzo[g]chrysene (B[g]C) and benzo[clchrysene (B[clC) are two PAHs whose structures are also distorted from planarity by a fjord region. Studies of the syn- and antifjord-region diol epoxide diastereomers of benzo[glchrysene (B[glCDE) (Chart 2) and benzo[clchrysene (BklCDE) (22,231have shown that the properties exhibited by B[c]PhDE are characteristic of diol epoxides of other PAH distorted from planarity by a fjord region, i.e., low chemical reactivity, as measured by their rates of reaction with water, coupled with high mutagenic and carcinogenic activity. The potent mutagenicity of the fjordregion diol epoxides was concluded to be due to the high frequency with which they form DNA adducts in V79 cells, rather than to the formation of adducts with greater mutagenic potential (24). In these studies racemic mixtures of the diol epoxides, rather than individual enantiomers, were used. The B[c]PhDEs react extensively with both deoxyadenosine and deoxyguanosine residues in DNA in vitro and in mammalian embryo cells (25),whereas the diol epoxides of planar PAHs react almost exclusively with guanine residues in DNA i n vitro and in vivo. These differences are best explained in terms of the secondary structure of DNA. Diol epoxide adducts with guanine
Materials. The diol epoxides and radiolabel used in these experiments were in liquid form. For both, contamination must be eliminated by wearing impermeable laboratory gloves, laboratory coats, and splash-proof goggles. Hazardous materials should be sequestered from the rest of the laboratory. Radiolabel exposure is kept to a minimum by working with Perspex shielding and by monitoring with the proper dosimeters (Geiger counters). Wastes must be placed only in approved containers and disposed of by Health and Safety personnel. Racemic B[clPh-3,4-dihydrodiol and B[g]C-11,12-dihydrodiol were synthesized from the corresponding o-quinone as described previously (22, 28, 29). The separation of the enantiomeric dihydrodiols was accomplished as described for B[c]Ph-3,4dihydrodiol (30). The optically pure syn- and anti-11,12dihydrodiol 13,l.i-epoxides of B[g]C, and 3,4-dihydrodiol 1,2epoxides of B[c]Ph, were prepared from enantiomerically pure BlglC-11,12-dihydrodiolsand B[c]Ph-3,4-dihydrodiols according to published procedures (29, 30). The specific rotation of each stereoisomer was in good accordance with published data (29, 30). A 21-mer oligonucleotide (sequence: 5'-ACCGCCGGCCA*GGAGGAGTAC-3') was modified at the highlighted adenine residue to give the trans-N6-adduct of (a) (-)-anti-B[c]PhDE or (b) (+)-anti-B[c]PhDE (A. Seidel, unpublished). Calf spleen phosphodiesterase was purchased from Boehringer Corp. Ltd. (Lewes, East Sussex, U.K.); micrococcal nuclease, nuclease P1, potato apyrase, salmon sperm DNA, poly(deoxyadenylic-deoxythymidylic acid) poly(dAdT) and poly(deoxyguanylic-deoxycytidylic acid) (poly(dGdC)) (both alternating copolymers) from Sigma Chemical Co. Ltd. (Poole, Dorset, U.K.); poly(deoxyguany1ic-Me5-deoxycytidylic acid) (polydGme5dC) (alternating copolymer) from Pharmacia Biotech (St. Albans, Herts, U.K.); T4 polynucleotide kinase from NBL Gene Sciences Ltd. (Cramlington, Northumberland, U.K.), and carrier-free [ Y - ~ ~ P I Afrom T P ICN Biomedicals Ltd. (High Wycombe, Buckinghamshire, U.K.). The PEI-cellulose plates, manufactured by Macherey-Nagel, were obtained from Camlab Ltd. (Cambridge, U.K.). Reaction of Diol Epoxides with DNA, Poly(dAdT),and Poly(dGdC). Aliquots (100pL) of solutions of either DNA, poly(dGdC), poly(dAdT), or poly(dG-me5dC) (1 mg/mL) in 0.1 M Tris-
DNA Adducts of Fjord-Region Diol Epoxides HC1 (pH 7.4) were mixed with solutions of the isomeric diol epoxides of B[g]C and B[c]Ph (10 pg) in THF/ethanol (1:8 v/v; 22.5 pL). The reaction mixtures were kept in the dark at room temperature overnight and then extracted with water-saturated diethyl ether (6 x 130 ,uL). The aqueous solutions were stored at -20 "C prior to analysis. s2P-PostlabelingAnalysis. (A) Standard Method. Carcinogen-modified DNA, poly(dGdC), poly(dAdT), or poly(dG-me5dC) samples (1 pg) were digested to deoxyribonucleoside 3'monophosphates with micrococcal nuclease (0.36 unit) and spleen phosphodiesterase (3,ug) a t 37 "C for 2 h in 10 pL of 20 mM sodium succinate/lO mM CaC12, (pH 6.0). An aliquot (1.7 ,uL) was then 32P-labeledby incubation with [ Y - ~ ~ P I A(75 T PpCi) and T4 polynucleotide kinase (0.75 pL; 10 unitslpL), using the postlabeling procedure essentially as described by Gupta et al. (31). All reactions were terminated by incubation with 2 p L potato apyrase (20 milliunitdpl) at 37 "C for 30 min. The carcinogen-modified oligonucleotides were digested as above and then diluted with distilled water to give DNA digest (0.0085pg) in a volume of 1.7 pL. Postlabeling was then carried out as above, but using 50 pCi of [ Y - ~ ~ P I Aper T P sample instead of 75 pCi. (B) Nuclease P1 Enhancement Method. DNA, poly(dAdT), or poly(dGdC) samples (1 ,ug) were digested with micrococcal nuclease and spleen phosphodiesterase as described for the standard method. This was followed by digestion for 1 h with nuclease P1 (0.04 unit) as described previously (32, 33). Samples were then 32P-labeledby incubation with carrier-free [ Y - ~ ~ P I A(50 T PpCi) and T4 polynucleotide kinase (7.5 units) a t 37 "C for 30 min. Reaction was terminated by addition of apyrase as above. The carcinogen-modified oligonucleotides were treated as above up to the 32P-postlabeling stage. At this point, they were diluted with distilled water to give a sample of DNA digest (0.01 ,ug) in a volume of 5 pL. They were then postlabeled as above, using 10 ,uCi of [ Y - ~ ~ P I Aper T P sample in place of 50 pCi. (C) Thin-Layer Chromatography. Resolution of 32Plabeled adducts was performed on PEI-cellulose TLC sheets (20 x 20 cm) using a 4-directional anion-exchange solvent system (31, 34) with the following solvents: D1, 1.7 M sodium phosphate (pH 6.0); D2, 3.5 M lithium formate and 8.5 M urea (pH 3.5); D3, 0.8 M lithium chloride, 0.5 M Tris-HC1, and 8.5 M urea (pH 8.0); D4, 1.7 M sodium phosphate (pH 6.0). In some experiments, 0.8 M sodium phosphate was used in place of 0.8 M lithium formate in solvent D3. Adduct spots on the chromatograms were visualized by autoradiography for 1 h at -70 "C using screen intensifiers. Quantitation of Adducts. The adduct spots were excised from the chromatogram and levels of radioactivity determined by Cerenkov counting. Appropriate blank areas of the chromatogram were counted to obtain background levels, which were subtracted. (A) Standard Method. The levels of adducts present were determined by relating the cpm in normal nucleotides to the cpm in adduct nucleotides as previously described (31). (B) Nuclease P1 Enhancement Method. The specific activity of the [y-32PlATPwas determined by measuring the T4 polynucleotide kinase-catalyzed incorporation of radioactivity into a known amount of 2'-deoxyadenosine 3'-monophosphate ( U p ) (32). The levels of adducts were calculated from the formula:
pmol of adductslpg of DNA = adduct radioactivity ( d p d f i g of DNA) specific activity of [ Y - ~ ~ P I A T( dPp d p m o l ) HPLC Analysis. HPLC analyses of B[g]CDE adduct spots were carried out with the apparatus described by Pfau et al. (35),consisting of two Waters 501 HPLC pumps, a Waters 712 WISP autosampler, a Waters 440 absorbance detector at 280 nm, and a Berthold LB 507 HPLC radioactivity monitor.
Chem. Res. Toxicol., Vol. 8, No. 4, 1995 593 Gradient control and other data processing were achieved with Waters Baseline-810 software. After quantitation on TLC, adduct spots were eluted with pyridinium formate (4 M, pH 4.5; 500 pL) overnight, and the eluates were filtered and evaporated to dryness. The residues were redissolved in deionized water (100 pL) before injection onto a Zorbax phenyl-modified reverse-phase column (250 x 4.6 mm; particle size, 5 pm). Elution was with the following solvent system: 0-15 min, a linear gradient of 10-46470 buffer B (methanolhuffer A, 9:l) in buffer A (the composition of buffer A was 0.3 M sodium dihydrogen orthophosphate and 0.2 M orthophosphoric acid, adjusted to pH 2.0); 15-60 min, a linear gradient of 46-48% B in buffer A, 60-80 min, a linear gradient of 48-80% B in buffer A all at a flow rate of 1.2 mumin. Phenanthrene-9,lO-diol was used as an internal marker, with a retention time of 40 min.
Results Reactions with DNA. TLC Analyses. When DNA modified by (-)-anti-BlglCDE was analyzed by the standard 32P-postlabelingmethod, four major adducts and two minor adducts were detected on TLC (Figure 1, panel A). Forty percent of the total radioactivity on the plate was contained in adduct spot 6, with spots 1and 4 accounting for 29% and 22% of total radioactivity, respectively, and spot 3,6% of total radioactivity. When the other three BlglCDEs were analyzed, the adduct maps shown in Figure 1, panels D, G, and J, were obtained. These showed similar but not identical patterns of adduct spots, with each isomer forming one major adduct spot containing greater than 50% of the total radioactivity, and one adduct spot accounting for between 20% and 25% of total radioactivity. The (-)-syn- and (+Ianti-isomers formed a further five minor adduct spots, and the (+)-syn-isomer formed a further four adduct spots. Adduct maps for the four corresponding B[clPhDEs reacted with DNA are shown in Figure 2, panels A-D. The (-)-anti-B[c]PhDE isomer formed five adduct spots (panel A), (-)-syn-B[clPhDE formed eight adduct spots (panel B), and (+)-anti- and (+)-syn-B[c]PhDEs formed six adduct spots each (panels C and D). As a group, the adducts formed by the B[clPhDEs exhibited greater mobility with the TLC solvent system used than did the B[glCDE adducts (Figure 1). Figure 3 shows the total reaction of each diol epoxide of BlgIC and B[c]Ph with DNA, quantitated using the standard method of postlabeling and expressed in pmol of adductslpg of DNA. The reactions with DNA of the (+)-syn-, (-)-syn-, and (+)-anti-isomers of BlgICDE are similar to those of the (+)-syn-, (-1-syn-, and (+)-antiisomers of B[c]PhDE, respectively. However, the (-)anti-B[c]PhDE isomer gave a 2.5-fold greater reaction toward DNA than the (-)-anti-BlglCDE isomer. The former was the most reactive of all the diol epoxides in this study, producing 13.7 pmol of adductdpg of DNA. Of the four configurational isomers of B[g]CDE, (-)-synB[g]CDE was the most DNA reactive isomer, followed by (-)-anti-B[g]CDE, then (+)-syn-BlglCDE, with the (+Ianti-isomer giving the least reaction with DNA. Similarly, (+I-anti-B[clPhDE was the least reactive of the four B[clPhDEs, with only 18%of the reaction of the (-)-antiB[c]PhDE isomer with DNA. When DNA samples that had been modified with one of the four BlgICDEs or one of the four B[clPhDEs were postlabeled using the nuclease P1-enhanced method and resolved on TLC, the same adduct spots were obtained,
594 Chem. Res. Toxicol., Vol. 8,No. 4, 1995
D
Giles et al.
F
F
e '13 '12
'l\
e\lo
I
'9
'8
'7
H
I I
n
L
-.
25\(
0
''e
.'32 '24b
0
'26
'24
'22
'21 Figure 1. Autoradiograms from 32P-postlabeled digests of the BlgJCDEs reacted with DNA (lefi-hand column), poly(dGdC) (middle column), and poly(dAdT) (right-hand column) a s described in the text. Panels A-C, (-)-anti-B[g]CDE; D-F, t -)-syn-BkICDE; G-I, (+)-anti-B[g]CDE; J-L, (+)-syn-B[g]CDE. 32P-Postlabelingwas by the standard method. Autoradiography was for 1 h at -70 "C. The origin is located near the bottom left-hand corner of each map and was excised before autoradiography. Adduct spots have been numbered the same where they have been shown to be the same using cochromatography on TLC and HPLC. A suffix "b" is given where this adduct spot comigrates in the standard TLC solvent system with the corresponding spot from reaction of the diol epoxide with DNA and cannot be separated from this or detected by HPLC. Faint adduct spots are circled.
in the same proportions as those seen when labeling was by the standard method (results not shown). However, overall adduct levels were determined to be 3- 10 times lower with the nuclease P1-enhanced method of postlabeling than with the standard method. We attempted to determine the labeling efficiencies of some of the adducts, by labeling two individual 21-mer oligonucleotides containing a single adduct, either transN6-dA-( -)-anti-B[c]PhDE or trans-NG-dA-(+)-anti-B[clPhDE, by both the standard and the nuclease PIenhanced methods of postlabeling. In this instance, we found that the nuclease P1-enhanced method gave a higher level of modification than did the standard method. With the nuclease P1-enhanced method of postlabeling, the adduct with (+)-anti-B[c]PhDE was labeled with 50% efficiency, in contrast to the (-)-anti-
B[c]PhDE adduct which was labeled with only 35% efficiency. The standard method of postlabeling gave labeling efficiencies of 12% for both adducts. Reactions with Poly(dAdT)and Poly(dGdC). (A) TLC Analyses. Figure 1 illustrates how the major adducts formed by each BlgICDE with DNA (panels A, D, G, and J) can be tentatively identified by chromatographic comparison with those adducts arising from reaction of each isomer with either poly(dGdC) (panels B, E, H, and K) or poly(dAdT) (panels C, F, I, and L), after 32P-postlabeling by the standard procedure. It should be noted that only very low levels of adducts were detected after reaction of the B[c]PhDEs and BlgICDEs with poly(deoxyadeny1ic acid) (polydA), apparently demonstrating the requirement of double stranded nucleic acid for the diol epoxides to react. It was assumed, by
DNA Adducts of Fjord-Region Diol Epoxides
Chem. Res. Toxicol., Vol. 8, No. 4, 1995 595
A 4-0
2”
6
n
3” !J 1”
0
4
(-)-anti-
(-)-syn-
( + ) - a n tI -
(+)-syn-
Figure 3. Comparison of the relative levels of adducts formed with DNA by each enantiomer of B[g]CDE (filled bars) and BklPhDE (hatched bars), as analyzed by the standard method of 32P-postlabeling. TLC and quantitation were as described in the 14-
0
text. 0
I
Figure 2. Autoradiographs of 32P-labeled digests of DNA reacted with A, (-)anti-B[clPhDE; B, (-)-syn-B[clPhDE; C, (+Ianti-BkIPhDE; D, (+)-syn-B[clPhDE. 32P-Postlabeling was by the standard method. The origin is located near the bottom left-hand corner of each map and was excised before autoradiography, which was for 1h at -70 “C. Faint adduct spots are circled.
analogy with the properties of other, previously studied, diol epoxides, that principally the guanine residues in poly(dGdC) and the adenine residues in poly(dAdT)would be modified. Thus, all four BlgICDEs reacted extensively with adenine residues in poly(dAdT)and in DNA, to form between two and five adduct spots that, in most cases, showed greater mobility in the TLC solvent system than the guanine adducts. The reactions of (-)-anti- and (+)syn-diol epoxides with poly(dGdC)gave rise to one extra minor spot each (Figure 1, panels B and K, spots X and Y, respectively), which did not comigrate with any spots from their reactions with DNA (Figure 1, panels A and J). These additional adduct spots indicate that there may have been some minor reaction with cytosine residues. To test this, (-)-anti-B[g]CDE was reacted with poly(dG me5dC), under the same conditions as the other in vitro reactions, and postlabeled by the standard method. The pattern of adduct spots obtained was the same as that of (-)-anti-B[g]CDE reacted with poly(dGdC), although in each case, the adduct levels were 4 times higher in the poly(dG-me5dC)sample. Each B[glCDE formed a major adduct with poly(dAdT) (Figure 1, panels C, F, I, and L, adduct spots 4, 10, 16, and 24) which comigrated with one from reaction with poly(dGdC)(the same spot numbers given the suffix b in Figure 1, panels B, E, H, and K). Attempts were made to separate these comigrating spots by varying the solvent system employed in the TLC separation, using previously published methods (36-39). By replacing the lithium chloride (0.8 M)in D3 of our solvent system with 0.8 M sodium phosphate (361, we resolved the (-)-antiB[g]CDE/poly(dAdT)spot 4 (Figure 1,panel c)from the poly(dGdC) spot 4b (Figure 1, panel B). The resolved guanine adduct accounted for the extra spot seen when DNA modified with this diol epoxide was analyzed using the new TLC solvent and accounted for 7% of the radioactivity level of spot 4. Results are shown in Figure 4, panels A-C. When poly(dAdT) modifed with (-)-synBlgJCDE was postlabeled and chromatographed using
the new D3 solvent, spot 10 (Figure 1, panel F) was resolved into two adduct spots, one major and one minor (Figure 4, panel F, spots 10 and 1Oc). The minor spot accounted for the extra spot seen when DNA modified by this diol epoxide was analyzed using this new TLC solvent (Figure 4, panel D, spot 1Oc) and accounted for 8%of the radioactivity level of spot 10. This solvent also resolved the (-)-syn-B[g]CDE/poly(dAdT) adduct spot 8 from the poly(dGdC) adduct spot 8b. The effect of this new solvent system was to retard the mobility of the adenine adducts. However, the major adduct spot 10 from poly(dAdT) still comigrated with the corresponding adduct spot 10b from reaction of the diol epoxide with poly(dGdC)(Figure 4, panels D-F). Adduct spots 16 and 16b arising from reaction of (+)-anti-B[g]CDE with poly(dAdT) and poly(dGdC), respectively, and adduct spots 24 and 24b arising from reaction of (+)-syn-BkJCDEwith poly(dAdT) and poly(dGdC), respectively, could not be resolved from each other using this new solvent system (results not shown). (B)HPLC Analyses. Areas containing corresponding pairs of major adduct spots from reaction of each isomer with DNA and with a poly(deoxynuc1eic acid) were excised from the TLC plate and eluted. Aliquots of each eluate, containing the same level of radioactivity, were then examined both separately and together on HPLC using the conditions described previously. Also examined were the pairs of adduct spots formed by each diol epoxide with poly(dAdT) and poly(dGdC) which comigrated on TLC (Figure 1,middle and right-hand columns, adduct spots 4 and 4b; 10 and lob; 16 and 16b; 24 and 24b). We found that each pair could be resolved on HPLC and that, in each case, it was the adduct with poly(dAdT) which coeluted with the corresponding adduct from reaction of the diol epoxide with DNA. When adduct spot 4 from reaction of (-)-anti-B[g]CDE with DNA was examined on HPLC, we found no evidence of the presence of the poly(dGdC) adduct, which had been resolved from the poly(dAdT) adduct when the new TLC solvent system was employed. Similarly, adduct spots 8 and 10 from reaction of (-)-syn-B[g]CDE with DNA also eluted as single peaks on HPLC, despite each being resolved into two adduct spots when run on TLC using the modified solvent system. The extra spot X from reaction of the (-)-anti-BCgICDE isomer with poly(dGdC) (Figure 1, panel B) eluted as two distinct peaks on HPLC. The same two peaks were obtained when the correspondingspot formed by reaction of the same isomer with poly(dG-me5dC)was run on HPLC.
596 Chem. Res. Toxicol., Vol. 8, No. 4, 1995
Giles et al.
B
O'X
F
E
r
\
Figure 4. Autoradiographs from 32P-postlabeleddigests of (-)anti-BlgICDE (A-C) and (-)-syn-BlgICDE (D-F) reacted with DNA (left-hand column), poly(dGdC) (middle column), and poly(dAdT) (right-hand column). The standard method of 32P-postlabelingwas used. TLC was carried out as described in the text, with 0.8 M sodium phosphate in place of 0.8 M lithium formate in solvent D3. The origin is located near the bottom left-hand corner of each map and was excised before autoradiography, which was for 1 h a t -70 "C.Adduct spots have been numbered the same where they have been shown to be the same, according to the legend of Figure *1.
Table 1. HPLC Retention Times of the Major Adduct Spots Resolved by PEI-Cellulose TLC of S2P-Labeled Digests of the B[g]CDEs Reacted with DNA, Poly(dGdC), and Poly(dAdT) retention timedmin adductno." DNA pdGC pdAT mixb compd -anti 1 37.4 37.6 37.5 3 29.7 30.2 30.0 4 39.4 39.6 39.6 6 59.6 58.6 58.6 -syn 7 52.0 51.6 52.0 8 38.2 38.6 38.7 9 28.5 28.6 28.5 10 39.6 39.4 39.8 12 47.4 47.8 47.7 13 57.2 56.9 57.1 +anti 14 36.5 36.2 36.3 15 27.8 28.0 28.0 16 39.1 39.0 39.2 19 57.9 57.8 58.3 20 49.4 49.9 49.5 +syn 21 37.8 38.4 38.4 22 37.9 37.6 38.0 24 33.0 33.0 33.1 26 50.5 50.6 50.7 a Adduct numbers correspond to those in Figure 1. Mix refers to the coelution of the adduct formed with DNA, and the corresponding adduct formed with either poly(dAdT) or poly(dGdC).
Table 1 gives the retention times of all of the major adducts formed by each isomer with DNA, and with the corresponding polydeoxyribonucleotide. The extent of reaction of each isomer toward adenine and guanine residues has been calculated as a percentage of the total reaction with DNA. Results for reaction with adenine residues are given in Figure 5 and are compared with the results for the B[c]PhDEs. The (+)-syn-B[g]CDE isomer shows a marked preference of 70% for reaction with adenine (1.9pmol of adductdmg of DNA). Reaction of (-)-anti-B[glCDE was more evenly distributed between the two purines. Both (-)-syn- and (+)-anti-B[g]CDE react preferentially with guanine residues. The (+)-syn-
U z 0
100
C
.-5 3
#!
(-)-anti-
(-)-syn-
(+)-anti-
(+)-syn-
Figure 6. Comparison of the relative levels of adducts formed by the BlgICDEs (filled bars) and the B[clPhDEs (hatched bars) with adenine residues in DNA, as a percentage of the total adduct level with DNA. The standard method of 32P-postlabeling and quantitation were used, a s described in the text.
B[c]PhDE isomer also showed a marked preference for reaction with adenine residues, with 80% of adduct formation by this diol epoxide occurring at this base (2.9 pmol of adducts/,g of DNA). The other three B[c JPhDEs showed no preference for reaction with adenine residues. The (+)-anti-B[c]PhDE isomer showed a marked preference of 70% for guanine residues.
Discussion In 1980,Wood et al. (20) demonstrated that all four fjord-region B[c]PhDEs are highly mutagenic in bacterial and mammalian cells. This is thought to be the result of steric hindrance in the bay-region causing a slight distortion of the parent molecule, and thus its diol epoxide derivatives, from planarity. A distorted diol epoxide molecule is less ionic than a planar equivalent, due to there being less stabilization of the developing ionic charge. This makes a distorted diol epoxide molecule more reactive toward strong nucleophiles such as DNA, and resistant to attack by water, a weaker nucleophile. The PAHs B[a]P, 7-methylbenz[a]anthracene,
DNA Adducts of Fjord-Region Diol Epoxides 5-methylchrysene, and B[c]Ph show an increase in distortion from planarity from B[alP through to B[clPh, respectively, and so a decreasing ionic strength. There is, therefore, an observed increase in reaction of their bayregion diol epoxides with DNA. In the present study, we have demonstrated that the sterically-hindered, and therefore distorted, B[glCDEs behave similarly to the B[clPhDEs. The (-1-syn-, (+)-anti-,and (+I-syn-isomers of B[g]CDE show levels of reaction with DNA very similar to those of the corresponding isomers of B[clPhDE. In the case of the syn-B[glCDE enantiomers, this finding is similar to that reported recently by Szeliga et al. (23). However, the (-1-anti-B[clPhDE isomer is 2.5 times as reactive toward DNA as the (-)-anti-Blg]CDE isomer and is the most reactive of all the diol epoxides studied in this paper. The standard method of 32P-postlabelingwas used for the quantitative studies, as adduct levels were high enough for detection by this procedure. As a comparison, however, adduct levels were also determined using the nuclease P1-enhanced method of postlabeling. By the latter method, we recovered lower levels of adducts. We feel this may be as a result of the experimental conditions employed. A high level of dilution of both sample and reaction mixtures was required in order to make possible adduct measurements of such highly modified DNA by this method. This may have affected the relative efficiency of the standard and nuclease P1-enhanced procedures. In contrast, the labeling efficiencies of two B[clPhDE adducts contained in synthetic oligonucleotides were higher using the nuclease P1-enhanced method of postlabeling than with the standard method. Experimental conditions may again be responsible for these results. For both methods, extensive dilution of the samples was required in order to measure the adduct levels in these very highly modified (5%)samples. The observation that one of the adducts labeled more efficiently than the other in this study, using the nuclease P1-enhanced method, is one that does not mirror the results of our main study. Our observed ratios of reaction of the B[c]PhDEs with adenine and guanine bases in DNA, as detected by both postlabeling methods, are similar to those reported previously, obtained by a different experimental method (25). In addition, the labeling efficiency of the two synthetic adduct-containing oligonucleotides was the same by the standard method. Taken together, these results indicate that, under the normal postlabeling reaction conditions, the labeling efficiencies of our adducts were comparable. Thus, we have assumed the labeling efficiencies for adducts formed by the four B[glCDEs to be comparable and have made comparisons between the diol epoxides of the two compounds from the results of quantitation of postlabeling experiments. An important point is that we have detected the same BLcIPhDE adducts, and in the same proportions, using both the standard method and the nuclease P1 digestion enhancement method of postlabeling. This is in contrast to a study in which B[c]PhDE adducts were reported to be undetectable by the nuclease P1 digestion method (37). Adducts formed with exocyclic amine groups by PAHs are generally nuclease P1 resistant, as shown by Reddy et al. (32); thus our demonstration that adducts formed by B[clPhDEs and B[g]CDEs are also nuclease P1 resistant is in accord with this empirical rule. The finding that we could only detect very low levels of adenine adducts after reaction of the diol epoxides with
Chem. Res. Toxicol., Vol. 8, No. 4, 1995 597 poly(&) suggests that adduct formation requires the nucleic acid to have a double helical structure. This is possibly because intercalation of the diol epoxide is maximized when the nucleic acid has this tertiary structure. However, adduct formation of some PAH diol epoxides with poly(&) has been reported (40),as has adduct formation with purine monophosphates (36, 37). It is of interest that the reaction of the B[clPhDEs with DNA, poly(dGdC), and poly(dAdT) gave rise to more adduct spots after 32P-postlabelingand TLC, using the methods described in this paper, than have been observed previously when the B[c]PhDEs were reacted with DNA, GMP, or AMP and analyzed by a modified postlabeling method (37). This fact may explain why our values for the reaction of each B[clPhDE with adenine residues in DNA differ slightly (approximately 10% lower) from those previously published (37, 41). However, they are not varied enough to be accounted for by a resistance of some of the adenine adducts to the action of the micrococcal nuclease and spleen phosphodiesterase enzymes, as has been reported previously by Cheh et al. (42). The difference in results may be due to differences in the DNA digestion procedures employed in the different laboratories. Despite adduct spots 4 of (-)-anti-B[g]CDE reacted with DNA, and spots 8 and 10 of (-)-syn-B[g]CDE reacted with DNA, being resolved into two discrete spots when using the modified D3 TLC solvent, the material in these spots eluted as single peaks on HPLC. This is likely to be due to the fact that the minor spot in each case contained low levels of radioactivity below the background noise level. Spots 16 from (+)-anti-B[g]CDE and 24 from (+)-syn-B[g]CDE eluted as a single peak on HPLC and could not be resolved into their adenine and guanine components when run on TLC, in the original or the modified solvents. Thus, since the single peak coeluted on HPLC with that of the spot from poly(dAdT), it is likely that the DNA spot arises from adducts with adenine only, for both isomers. Spots 16b and 24b arising from reaction of (+)-anti-B[glCDE and (+)-syn-BlglCDE, respectively, with poly(dGdC1 are likely to be additional guanine adducts not formed by these diol epoxides with guanine residues in DNA. Adducts arising from the reaction of the (+)-anti- and (-1-anti-B[c]PhDEs with cytosine residues have been reported previously (41). However, we found no evidence of reaction of (-)-anti-BlglCDE with cytosine residues when poly(dGdC) was replaced with poly(dG-me5dC),as the additional adduct spot was still present. We found that the additional adduct spot contained higher levels of radioactivity in the analysis of the poly(dG-me5dC) sample than in the poly(dGdC) sample, indicating that a protected cytosine residue favors formation of this adduct. It is possible that the apparent extra spots from the reaction of (-)-anti- and (+I-syn-B[glCDEs with poly(dGdC) are in fact present in the DNA sample but below the level of detection of discrete spots by autoradiography. The results presented in this paper emphasize the importance of using at least two chromatographic systems in the analysis of DNA adducts. The presence of large adduct spots after autoradiography of TLC plates can often mask smaller spots, even when, as here, 20 x 20 cm TLC plates are employed to obtain a better resolution than is achievable with 10 x 10 cm plates. However, where an adduct spot contains only a low level of radioactivity, we have shown that it may be below the limits of detection using HPLC. The use of a second
598 Chem. Res. Toxicol., Vol. 8, No. 4, 1995
chromatographic system also provides an opportunity to confirm the comigration of adduct spots, thereby increasing confidence that they are the same adduct. In summary, we have shown that each B[g]CDE reacts extensively with DNA in vitro, forming adducts with both adenine and guanine bases, similar to the reactivities of the B[c]PhDEs. We have demonstrated that a combination of TLC and HPLC can be used to separate and identify the origins of most, but not all, of the adduct spots obtained by reaction of each diol epoxide with DNA, poly(dAdT), and poly(dGdC). Several of the adduct spots from the pairs of B[g]CDE diastereomers reacted with DNA show similar mobilities on TLC and similar retention times on HPLC. Further improvements in chromatographic separation of 32P-postlabeled PAH-DNA adducts are being developed in order to achieve separation of all of the B[g]CDE-DNA adducts. These procedures will assist in the identification of DNA adducts arising in vivo from metabolic activation of the parent hydrocarbon.
Acknowledgment. A.S.G. gratefully acknowledges receipt of a Research Studentship from the Institute of Cancer Research.
References Lunde, G., and Bjorseth, A. (1977) Polycyclic aromatic hydrocarbons in long-range transported aerosols. Nature CLondoni 268, 518-519. Dipple, A,, Moschel, R. C., and Bigger, C. A. H. (1984) Polynuclear aromatic carcinogens. In Chemical Carcinogens (Searle, C. E., Eds.) 2nd ed., pp 41-163, American Chemical Society, Washington. Thakker, D. R., Nordqvist, M., Yagi, H., Levin, W., Ryan, D., Thomas, P. E., Conney, A. H., and Jerina, D. M. (1979) Comparative metabolism of a series of polycyclic aromatic hydrocarbons by rat liver microsomes and purified cytochrome P450. In Polynuclear Aromatic Hydrocarbons: Third International Symposium on Chemistry and Biology-Carcinogenesis and Mutagenesis (Jones, P. W., and Leber, P., Eds.) pp 455-472, Ann Arbor Science Publishers, Ann Arbor, MI. Jerina, D. M., and Daly, J. W. (1976) Oxidation at carbon. InDrug Metabolism (Parke, D. V., and Smith, R. L., Eds.) pp 13-32, Taylor and Francis, London. Sims, P., and Grover, P. L. (1981) Involvement of dihydrodiols and diol epoxides in the metabolic activation of polycyclic hydrocarbons other than benzo[a]pyrene. In Polycyclic Hydrocarbons and Cancer (Gelboin, H. V., and Ts’o, P. 0. P., Eds.) Val. 3, pp 117-181, Academic Press, New York. Sims, P., Grover, P. L., Swaisland, A,, Pal, K., and Hewer, A. (1974) Metabolic activation of benzo[alpyrene proceeds by a diol epoxide. Nature (London) 252, 326-328. Brookes, P., and Lawley, P. D. (1964) Evidence for the binding of polynuclear aromatic hydrocarbons to the nucleic acids of mouse skin. Nature (London) 202, 781-784. ( 8 ) Dipple, A,, Peltonen, K., Cheng, S.C., and Hilton, B. D. (1993) Chemistry of DNA adduct formation by dihydrodiol epoxides of polycyclic aromatic hydrocarbons. In Polycyclic Aromatic Hydrocarbons: Synthesis, Properties, Analytical Measurements, Occurrence and Biological Efects (Garrigues, P., and Lamotte, M., Eds.) pp 807-816, Gordon and Breach, Philadelphia. (9) Malaveille, C., Kuroki, T., Sims, P., Grover, P. L., and Bartsch, H. (1977) Mutagenicity of isomeric diol epoxides of benzo[alpyrene and benz[alanthracene in S. Typimurium TA98 and TAlOO and in V79 Chinese hamster cells. Mutat. Res. 44, 313-326. (10) Newbold, R. F., and Brookes, P. (1976) Exceptional mutagenicity of a benzo[alpyrene diol epoxide in cultured mammalian cells. Nature (London) 261, 52-54. (11) Wood, A. W., Chang, R. L., Levin, W., Yagi, H., Tada, M., Vyas, K. P., Jerina, D. M., and Conney, A. H. (1982) Mutagenicity of the optical isomers of the diastereomeric bay-region chrysene 12diol-3,4-epoxidesin bacterial and mammalian cells. Cancer Res. 42, 2972-2976. (12) Chang, R. L., Wood, A. W., Conney, A. H., Yagi, H., Sayer, J. M., Thakker, D. R., Jerina, D. M., and Levin, W. (1987) Role of diaxial versus diequatorial groups in the tumorigenic activity of a benzo-
Giles et al. [alpyrene bay-region diol epoxide. Proc. Natl. Acad. Sci. U.S.A. 84, 8633-8636. (13) Wislocki, P. G., Buening, M. K., Levin, W., Lehr, R. E., Thakker, D. R., Jerina, D. M., and Conney, A. H. (1979) Tumorigenicity of the diastereomeric benz[alanthracene 3,4-diol-l,Z-epoxides and the (+)- and (-)-enantiomers of benz[a]anthracene 3,4-dihydrodiol in newborn mice. J . Natl. Cancer Inst. 63,201-204. (14) Buening, M. K., Levin, W., Karle, J . M., Yagi, H., Jerina, D. M., and Conney, A. H. (1979) Tumorigenicity of bay-region epoxides and other derivatives of chrysene and phenanthrene in newborn mice. Cancer Res. 39, 5063-5068. (15) Buening, M. K., Wislocki, P. G., Levin, W., Yagi, H., Thakker, D. R., Akagi, H., Koreeda, M., Jerina, D. M., and Conney, A. H. (1978) Tumorigenicity of the optical enantiomers of the diastereomeric benzo[a]pyrene 7,8-diol-9,10-epoxides in newborn mice: exceptional activity of (+)-7P,8a-dihydroxy-9a,lOa-epoxy-7,8,9,10-tetrahydrobenzo[alpyrene.Proc. Natl. Acad. Sci. U.S.A. 75, 5358-5361. (16) Chang, R. L., Levin, W., Wood, A. W., Yagi, H., Tada, M., Vyas, K. P., Jerina, D. M., and Conney, A. H. (1983) Tumorigenicity of enantiomers of chrysene 1,2-dihydrodiol and of the diastereomeric bay-region chrysene 1,2-diol-3,4-epoxideson mouse skin and in newborn mice. Cancer Res. 43, 192-196. (17) Levin, W., Chang, R. L., Wood, A. W., Yagi, H., Thakker, D. R., Jerina, D. M., and Conney, A. H. (1984) High stereoselectivity among the optical isomers of the diastereomeric bay-region diol epoxides of benz[a]anthracene in the expression of tumorigenic activity in murine tumor models. Cancer Res. 44, 929-933. (18) Bartle, K. D., and Jones, D. W. (1972) The application of proton magnetic resonance spectroscopy to structure identification in polycyclic aromatic molecules. Advances in Organic Chemistry : Methods and Results 8, 317-423. (19) Sayer, J. M., Yagi, H., Croisy-Delcey, M., and Jerina, D. M. (19811 Novel bay-region diol epoxides from benzo[c]phenanthrene. J. Am. Chem. Sac. 103,4970-4972. ( 2 0 ) Wood, A. W., Chang, R. L., Levin, W., Thakker, D. R., Yagi, H., Sayer, J. M., Jerina, D. M., and Conney, A. H. (1984)Mutagenicity of the enantiomers of the diastereomeric bay-region benzo[clphenanthrene-3,4-diol-l ,%epoxides in bacterial and mammalian cells. Cancer Res. 44, 2320-2324. (21) Levin, W., Wood, A. W., Chang, R. L., Ittah, Y., Croisy-Delcey, M., Yagi, H., Jerina, D. M., and Conney, A. H. (1980) Exceptionally high tumor-initiating activity of benzo[c]phenanthrene bayregion diol epoxides on mouse skin. Cancer Res. 40, 3910-3914. (22) Glatt, H., Piee, A,, Pauly, K., Steinbrecher, T., Schrode, R., Oesch, F., and Seidel, A. (1991) Fjord- and bay-region diol epoxides investigated for stability, SOS induction in Escherichia coli, and mutagenicity in Salmonella typhimurium and mammalian cells. Cancer Res. 51, 1659-1667. (23) Szeliga, J., Lee, H., Harvey, R. G., Page, J . E., Ross, H. L., Routledge, M. N., Hilton, B. D., and Dipple, A. (1994) Reaction with DNA and mutagenic specificity of syn-benzo[glchrysene 11,12-dihydrodiol 13,14-epoxide. Chem. Res. Toxicol. 7,420-427. (24) Phillips, D. H., Hewer, A,, Seidel, A,, Steinbrecher, T., Schrode, R., Oesch, F., and Glatt, H. (1991) Relationship between mutagenicity and DNA adduct formation in mammalian cells for fjordand bay-region diol epoxides of polycyclic aromatic hydrocarbons. Chem.-Biol. Interact. SO, 177-186. (25) Dipple, A,, Pigott, M. A,, Agarwal, S. K., Yagi, H., Sayer, J . M., and Jerina, D. M. (1987) Optically active benzo[c]phenanthrene diol epoxides bind extensively to adenine in DNA. Nature (London) 327, 535-536. (26) Bigger, C. A. H., Sawicki, J . T., Blake, D. M., Raymond, L. G., and Dipple, A. (1983) Products of binding of DMBA to DNA in mouse skin. Cancer Res. 43, 5647-5651. (27) DiGiovanni, J., Sawyer, T. W., and Fisher, E. P. (1986) Correlation between formation of a specific hydrocarbon-deoxyribonucleoside adduct and tumor-initiating activity of 7,12-dimethylbenz[alanthracene and its 9- and 10-monofluoro derivatives in mice. Cancer Res. 46, 4336-4341. ( 2 8 ) Platt, K. L., and Oesch, F. (1983) Efficient synthesis of non-Kregion-dihydrodiols of polycyclic aromatic hydrocarbons from o-quinones and catechols. J . Org. Chem. 48, 256-268. (29) Bushman, D. R., Grossman, S. J., Jerina, D. M., and Lehr, R. E. (1989) Synthesis of optically active fjord-region 11,lZ-diol 13,14epoxides and the K-region 9,lO-oxide of the carcinogen benzoklchrysene. J . Org. Chem. 54, 3533-3544. (30) Yagi, H., Thakker, D. R., Ittah, Y., Croisy-Delcey, M., and Jerina, D. M. (1983) Synthesis and assignment of absolute configuration to the trans-3,4-dihydrodiols and 3,4-diol-l,Z-epoxidesof benzo[clphenanthrene. Tetrahedron Lett. 24, 1349-1352. (31) Gupta, R. C., Reddy, M. V., and Randerath, K. (1982) 32PPostlabelling analysis of non-radioactive aromatic carcinogenDNA adducts. Carcinogenesis 3, 1081-1092.
DNA Adducts of Fjord-Region Diol Epoxides (32) Reddy, M. V., and Randerath, K. (1986) Nuclease P1-mediated enhancement of sensitivity of 32P-postlabellingtest for structurally diverse DNA adducts. Carcinogenesis 7, 1543-1551. (33) Hughes, N. C., and Phillips, D. H. (1993) 32P-Postlabellinganalysis of the covalent binding of benzo[ghi]perylene to DNA in vivo and in uitro. Carcinogenesis 14, 127-133. (34) Phillips, D. H., Glatt, H. R., Seidel, A., Bochnitschek, W., Oesch, F., and Grover, P. L. (1986) Mutagenic potential of DNA adducts formed by diol epoxides, triol-epoxides and the K-region epoxide of chrysene in mammalian cells. Carcinogenesis 7, 1739-1743. (35) Pfau, W., and Phillips, D. H. (1991) Improved reversed-phase high-performance liquid chromatographic separation of 32Ppostlabelled nucleoside 3’,5’-bisphosphate adducts of polycyclic aromatic hydrocarbons. J . Chromatogr. 570, 65-76. (36) Canella, K., Peltonen, K., and Dipple, A. (1991) Identification of (+) and (-) anti benzo[alpyrene dihydrodiol epoxide-nucleic acid adducts by the 32P-postlabellingassay. Carcinogenesis 12,11091114. (37) Canella, K. A,, Peltonen, K., Yagi, H., Jerina, D. M., and Dipple, A. (1992) Identification of individual benzo[clphenanthrene dihydrodiol epoxide-DNA adducts by the 32P-postlabeling assay. Chem. Res. Toxicol. 5, 685-690. (38) Schoepe, K. B., Friesel, H., Schurdak, M. E., Randerath, K., and Hecker, E. (1986) Comparative DNA binding of 7,12-dimethylbenz[a]anthracene and some of its metabolites in mouse epidermis
Chem. Res. Toxicol., Vol. 8, No. 4, 1995 599 in uivo as revealed by the 32P-postlabelingtechnique. Carcinogenesis 7, 535-540. (39) Hughes, N. C., and Phillips, D. H. (1990) Covalent binding of dibenzpyrenes and benzo[alpyrene to DNA: evidence for synergistic and inhibitory interactions when applied in combination to mouse skin. Carcinogenesis 11, 1611-1619. (40) Reddy, A. P., Preuss-Schwartz, D., and Marnett, L. J . (1992) Separation of (+)-syn- and (-)-anti-benzo[a]pyrene dihydrodiol epoxide-DNA adducts in 32P-postlabelinganalysis. Chem. Res. Toxicol. 5, 19-25. (41) Jerina, D. M., Sayer, J . M., Agarwal, S. K., Yagi, H., Levin, W., Wood, A. W., Conney, A. H., Pruess-Schwartz, D., Baird, W. M., Pigott, M. A., and Dipple, A. (1986) Reactivity and tumorigenicity of bay-region diol epoxides derived from polycyclic aromatic hydrocarbons. In Biological Reactive Intermediates III (Kocsis, J . J., Jollow, D. J., Witmer, C. M., Nelson, J. O., and Snyder, R., Eds.) pp 11-30, Plenum Publishing Corp., New York. (42) Cheh, A. M., Yagi, H., and Jerina, D. M. (1990) Stereoselective release of polycyclic aromatic hydrocarbon-deoxyadenosine adducts from DNA by the 32P-postlabeling and deoxyribonuclease l h n a k e venom phosphodiesterase digestion methods. Chem. Res. Toricol. 3, 545-550.
TX940135G
