Hydrocarbon-Deoxyadenosine Adducts from DNA by the 32P

Chem. Res. Toxicol. 1990, 3, 545-550

545

Stereoselective Release of Polycyclic Aromatic Hydrocarbon-Deoxyadenosine Adducts from DNA by the 32P Postlabeling and Deoxyribonuclease I/Snake Venom Phosphodiesterase Digestion Methods Albert M. Cheh,*J Haruhiko Yagi, and Donald M. Jerina Laboratory of Bioorganic Chemistry, NZDDK, National Institutes of Health, Bethesda, Maryland 20892 Received J u n e 7, 1990

T h e restricted ability of deoxyribonuclease I/snake venom phosphodiesterase digestion t o liberate deoxyadenosine (dA) nucleotide adducts of polycyclic aromatic hydrocarbons from DNA, first observed by Dipple and Pigott with the bay-region diol epoxide adducts of 7,12-dimethylbenz[a]anthracene,has been observed with t h e dA adducts of benz[a]anthracene and benzo[c]phenanthrene diol epoxides. T h e micrococcal nuclease/spleen phosphodiesterase digestion used in the original 32Ppostlabeling procedure developed by Randerath t o determine DNA adducts also failed t o liberate dA nucleotide adducts quantitatively. Thus either method can potentially lead t o an underestimation of the extent t o which dA has been modified in DNA. T h e two digestion procedures exhibit systematic and mostly opposite stereoselectivity in the pattern of which dA adducts are resistant to digestion, which suggests t h a t these adducts may have preferred orientations within modified DNA t h a t are determined by whether they have the R or S configuration at C-1, t h e point of attachment between the exocyclic amino group of dA and the hydrocarbon; this in turn is dictated by the configuration about the precursor benzylic epoxide carbon and the cis versus trans nature of epoxide opening during adduct formation.

The adducts formed from the four optically active stereoisomers of BPDE, DBADE, BcPhDE, and BADE have The ability to form DNA adducts is an important debeen well characterized (5-10, 12-14). The relative prefterminant of chemical carcinogenicity (1). For polycyclic erence for adduct formation at dG or dA depends on the aromatic hydrocarbons, the ultimate carcinogens and diol epoxide stereoisomer. Attack by an exocyclic purine DNA-reactive forms, bay-region diol epoxides (2,3),show nitrogen results in both cis and trans opening of the epdifferent specificities in the type and amount of DNA oxide to produce adducts which we designate as cis or trans adducts formed. Nearly all of these adducts arise by attack (Figure 2). Individual diol epoxide isomers also form of the exocyclic amino group of deoxyguanosine (dGI2and different amounts of cis or trans adducts a t dG and dA. deoxyadenosine (dA) at the benzylic position of the epWith cis or trans addition by dG or dA, four different oxide group. Metabolically, diastereomeric bay-region diol adducts may be formed from a single diol epoxide isomer. epoxides in which the benzylic hydroxyl group is either cis Determination of the differences in the types of adducts (DE-1) or trans (DE-2) to the epoxide oxygen are possible formed by different diol epoxides assumes that the adducts (41, and each can be formed as a pair of enantiomers can be recovered quantitatively. DNA that contains ad(Figure 1). Deoxyguanosine adducts are the principal ducts is digested enzymatically to nucleotides (or nucleoproducts formed by the diastereomers of the benzo[a]sides) plus nucleotide (nucleoside) adducts. The adducts pyrene diol epoxides (BPDE) (5-7), dibenz[aj]anthracene are then separated from the normal, nonadducted nudiol epoxides (DBADE) ( 8 , 9 ) ,and benz[a]anthracene diol cleotides (nucleosides) and are quantitated. One method epoxides (BADE) (9, 10). Although only the bay-region for liberating adducts as nucleosides employs sequential diol epoxide-’) isomers of 5-methylchrysene appear to have digestion by deoxyribonuclease I (EC 3.1.21.1) (DNase I), been studied, these also prefer to attack dG (11). In snake venom phosphodiesterase (EC 3.1.4.1) (VPD), and contrast, substantial amounts of dA adducts are formed alkaline phosphatase (EC 3.1.3.1) (AP) (16). Dipple and from benzolclphenanthrene diol epoxide (BcPhDE) (9, Pigott first reported the resistance of DMBADE-dA ad12-14) and 7,i!2-dimethylbenz[a]anihracene diol epoxide ducts to this digestion procedure; low amounts of VPD that (DMBADE) ( 1 5 ) . were sufficient to liberate dG adducts failed to liberate all of the dA adducts, leading to their underestimation (17). The 32Ppostlabeling procedure of Randerath (18-20) is Work done while on sabbatical leave from the Department of a very valuable method for quantitating adducts formed Chemistry, American University, Washington, DC 20016. * Abbreviations: dA, deoxyadenosine; dG, deoxyguanosine; DE, diol in vivo or in vitro; it is especially useful with human samepoxide; BPDE, 7,8-dihydroxy-9,10-epoxy-7,8,9,lO-tetrahydrobenzo[a]- ples, because of its ability to detect adduct formation at pyrene; DBADE, 3,4-dihydroxy-1,2-epoxy-1,2,3,4-tetrahydrodibenz[a~]anthracene; BADE, 3,4-dihydroxy-1,2-epoxy-l,2,3,4-tetrahydrobenz[a]- levels as low as one adduct per lo8DNA base pairs, without anthracene; BcPhDE, 3,4-dihydroxy-1,2-epoxy-1,2,3,4-tetrahydrobenzo- requiring a radiolabeled carcinogen ( I ) . In its original form [clphenanthrene; DMBADE, 7,12-dimethyl-3,4-dihydroxy-l,2-epoxy- it used micrococcal nuclease (EC 3.1.31.1)(MN) and spleen 1,2,3,4-tetrahydrobenz[a]anthracene;DNase I, deoxyribonucleaseI; VPD, phosphodiesterase (EC 3.1.16.1) (SPD) to digest adducted snake venom phosphodiesterase: AP, alkaline phosphatase; MN, micrococcal nuclease; SPD, spleen phosphodiesterase. DNA to normal and adducted deoxynucleoside 3’-phos-

Introduction

This article not subject to U S . Copyright. Published 1990 by the American Chemical Society

546

Chem. Res. Toxicol., Vol. 3, No. 6, 1990

(R,S,R,S)-DE-1

Cheh et al.

(S.R,S,R)-DE-1

BENZ(a)ANTHRACENE DE Diol Epoxide-1 Diastereomers

BADE

BENZO(c)PHENANTHRENEDE BcPhDE

F i g u r e 3. Structures of the BADE and BcPhDE. For stereochemistry of isomers, see Figure 1.

(R,S,S.R)-DE-2

(S,R.R.S)-DE-2

Diol Epoxide-2 Diastereomers Figure I . Stereochemistry of the optically active, bay-region diol epoxides which can be formed in mammals. Absolute configurat ion is designated around the saturated ring from the benzylic hydroxyl group toward the epoxide oxygen.

(R)cis-adduct

&:

R-HN/,

(s)-trans-adduct

Figure 2. Formation of cis and trans adducts from a diol epoxide stereoisomer. The four diol epoxide isomers each give rise to eight such adducts from dA and from dG by attack of the exocyclic amin:. groups of the purines a t the benzylic epoxide carbon. Cis adducts retain the configuration of the incoming epoxide carbon, (2-1, while trans adducts have an inverted configuration. p h a t e s prior to s e p a r a t i n g a n d q u a n t i t a t i n g t h e adducts. Conversion of t h e nucleoside monophosphate adducts to nucleoside a d d u c t s w i t h AP allows d i r e c t comparison of t h e nucleoside a d d u c t yield after digestion with either the D N a s e I/VPD procedure o r the postlabeling digestion p r o c e d u r e (MN SPD). T h e present study examines the ease with which adducts derived from BADE and BcPhDE (Figure 3) a r e liberated f r o m DNA. The r e p o r t e d resistance of DMBADE-dA a d d u c t s t o VPD digestion m a y b e a general effect, as i t e x t e n d s to t h e d A a d d u c t s of B A D E a n d BcPhDE. T h e SPD used in t h e original postlabeling digestion procedure h a s also been shown t o e n c o u n t e r difficulty in releasing d A a d d u c t s W i t h both enzymes this resistance to cleavage d e p e n d s i n a systematic m a n n e r o n t h e R versus S s t e r eoisomerism a b o u t t h e benzylic a t t a c h m e n t s i t e i n the h y d r o c a r b o n a d d u c t , which i n t u r n d e p e n d s o n t h e stereoisomer of the diol epoxide f r o m which t h e a d d u c t s a r e formed a n d the cis versus trans nature of a d d u c t formation (Figure 2).

+

Experimental Procedures DNA Adduct Formation. The four optically pure bay-region BADE isomers, racemic (RSRS + SRSR) BcPhDE-1 and racemic

(SRRS + RSSR) BcPhDE-2 (see Figure l ) , were synthesized as described (21,221. DNA adducts were formed by reaction of 0.05 mL of diol epoxide in acetonitrile (1.0 mg/mL) with 0.50 mL of calf thymus DNA (0.80 mg/mL) that had been dialyzed against 0.01 M Tris-HC1 (pH 7.4). After 1h a t 37 "C, reaction mixtures were extracted three times with double the volume of ethyl acetate and two times with double the volume of ether to remove noncovalently bound hydrolysis products of the diol epoxide. Residual ether was removed by evaporation under a stream of nitrogen, and water was added to restore the original volume. Digestion of Adducted DNA. In accordance with the original postlabeling procedure (It?), MN and SPD were obtained from Sigma Chemical Co., St. Louis, MO, and Boehringer Mannheim Biochemicals, Indianapolis, IN, respectively, and were dialyzed against purified (Millipore MilliQ system) water. Enzymatic activities were determined with the assays used by the respective suppliers (23). In a scale-up of the standard procedure, to 0.40 mg of DNA in 0.50 mL of water were added 0.050 mL of 0.2 M sodium succinate (pH 5.9), 0.025 mL of 0.2 M calcium chloride, 24 units of MN, and 0.4 unit of SPD. Digestions proceeded at 37 "C for 4 h or overnight. Conversion to nucleosides plus nucleoside adducts occurred after adding 0.075 mL of l M Tris base, 0.005 mL of 1 % sodium azide, and 2.0 units of AP followed by incubation for 3 h a t 37 "C. The digestion with DNase I, VPD, and AP (Sigma Chemical Co.) was adapted from the procedure of Baird and Brookes (16) as follows: Enough water was added to the extracted and evaporated solution containing 0.40 mg of DNA to bring the volume to 1.0 mL. This was followed by the addition of 0.010 mL of 1 M magnesium chloride, 0.020 mL of 1% sodium azide, and 200 Kunitz units of DNase I dissolved in 0.025 mL of 0.01 M Tris-HC1 (pH 7.4). After digestion for 3 h a t 37 "C and addition of 0.030 mL of 1 M Tris base and 0.032 unit of VPD dissolved in 0.025 mL of 0.2 M Tris-HC1 (pH 9.0), digestion was continued for 40-44 h a t 37 "C. Then 2 units of AP was added, and the digest was incubated a t 37 "C for 3 more h or overnight. The enzymes were used as assayed and shipped by the supplier. Quantitation of Nucleoside Adducts. Waters (Milford, MA) C-18 Sep-paks were washed with 20 mL of methanol and 20 mL of water (I 7). Two milliliters of 0.2 M potassium phosphate buffer (pH 7.4) was added to each digested sample, and the digests were loaded onto the Sep-paks. After washing the Sep-paks three times with 5 mL of water and twice with 5 mL of 25% methanol in water, the nucleoside adducts were eluted with four 2-mL volumes of methanol. The combined methanol eluate was evaporated to dryness under a stream of nitrogen and redissolved in 1.0 mL of methanol followed by 1.0 mL of water. Nucleoside adducts formed from individual BADE isomers were separated by HPLC on a Du Pont (Wilmington, DE) Zorbax ODS column (0.46 X 25 cm) eluted at a flow rate of 1.0 mL/min under the following isocratic conditions: (RSRS)-BADE-1,32% water and 68% methanol (72% water, 26.5% acetonitrile, and 1.5% methanol in early experiments); (SRSR)-BADE-1,32% water and 68% methanol (71.5% water, 26% acetonitrile, and 1.5% methanol in early experiments); (SRRS)-BADE-2, 72% water, 26.5% acetonitrile, and 1.5% methanol; (RSSR)-BADE-2, 74% water, 24.5% acetonitrile, and 1.5% methanol. Nucleoside adducts formed from racemic BcPhDE-1 were separated a t a flow rate of 1.0 mL/min on the same column by using a linear gradient running from 55% water, 3% tetrahydrofuran, and 42% methanol to 30% water, 3% tetrahydrofuran, and 67% methanol over 25 min. Adducts formed from racemic BcPhDE-2 were separated isocratically with 68% water, 21 % acetonitrile, and 11% methanol.

Chem. Res. Toxicol., Vol. 3, No. 6, 2990 547

Stereoselective Digestion of PAH- Adducted D N A Table I . Ratio of dA/dG Adducts from Optically Active Benz[a ]anthracene Diol Epoxides a s a Function of t h e Amount of VPD Used i n t h e Digestion 70 of the av dAjdG ratio amount dA/dG obsd with of VPD ratiob 15X VPD BADE isomer" (RSRS)-BADE-I 0.G 0.25 39 I X 0.58 91 ~

(SRSR)-BADE-1

(SRRS)-BADE-2

15X 0.1x 1x 5X

0.64 0.13

100

lox

0.27 0.32 0.44 0.37

73 86 119 100

15X 0.1x I X

5X

lox lSX

(RSSR)-RADE-2

0.1x

1x 15X

0.12 0.27

35

27 60 107

0.48 0.45 0.45

100 100

0.23 0.27 0.30

90 100

77

"See Figure 1 lor the naming of the diol epoxide isomers. bAverage of two o r more determinations. Separations were performed with a Hewlett-Packard 1090M high-performance liquid chromatograph equipped with a diodearray detector. Relative amounts of adducts were determined by peak areas measured a t 255 nm.

Results and Discussion The data of Dipple and Pigott (17)indicate that levels of >0.5 Sigma unit of VPD/mg of DNA would fully release DMBADE-dA adducts, while levels of c0.05 unit/mg of DNA would release relatively little DMBADE-dA adduct; intermediate amounts of enzyme would release intermediate amounts of dA adducts. The protocol described under Experimental Procedures which uses 0.08 unit of VPD/mg of DNA is designated as 1 X in VPD; this amount would fall near the low end of the Dipple and Pigott curve for release of dA adducts derived from DMBADE. Increasing the amount of enzyme to 5-1OX would approach the upper plateau for adduct release. The amount of enzyme used by Baird and Brookes (26) corresponds to 1.2X and that of Sawicki et al. (24) to 2X. Table I illustrates recovery of BADE-dA adducts with increasing levels of VPD. Essentially the same amounts of dG adducts were released by VPD levels ranging from 0.1X upward. The rightmost column of Table I expresses the release of dA adducts as a percentage of the amount released by 15X VPD. The dA adducts of (SRSR)BADE-1 and (SRRS)-BADE-2 were somewhat more resistant to release than the dA adducts formed from their enantiomeric diol epoxides [ (RSRS)-BADE-l and (RSSRI-BADE-2, respectively]; at least 5X VPD was required for maximal release of the former, while 1 X was sufficient for the latter. The requirement for 5X VPD in order to liberate dA adducts fully is close to the amount predicted from the data of Dipple and Pigott (17);the amounts used in the procedures of Baird and Brookes (16) and Sawicki e t al. ( 2 4 ) may not have fully liberated dA adducts either. Figure 4 shows the chromatographic profile of adducts obtained from (SRRSl-BADE-2-treated DNA after digestion with l x VPD and 15X VPD. The cross-hatched area indicates the increase in dA adduct yield with increased amount of enzyme; in this case the trans dA adduct is more resistant to digestion. Table I1 shows the extents of release of the pairs of cis and trans dA adducts of BADE

I

I

6

1

I

10 TIME (min)

I

1

14

F i g u r e 4. HPLC comparison of the amounts of dA nucleoside adducts obtained from DNA treated with (SRRS)-BADE-l when l x DNase I / l X VPD or 1 X DNase I / 5 X VPD is used prior to AP. The subscripts c and t indicate cis and trans adducts of dG or dA. The designation T indicates residual tetraol, C a deoxycytidine adduct, and X an additional dG adduct. The crosshatched area indicates the digestion shortfall with inadequate VPD. Table 11. Extent of Release of Individual Benz[a ]anthracene and Benzo[c]phenanthrene Diol Epoxide-dA Adducts Obtained with 1X Compared to 15X (BADE) o r 5X (BcPhDE) VPD" 90 of adduct released by 1 X VPDb BADE isomersc (RSRS)-BADE-1 (SRSR)-BADE-l (SRRS)-BADE-2 (RSSR)-BADE-2 BcPhDE isomers (RSRS)-BcPhDE-I (SRSR)-BcPhDE-l (SRRS)-BcPhDE-2 (RSSR)-BcPhDE-2

" 15X VPD with BADE adducts, 5X VPD with BcPhDE adducts. *The percentages are averages of the data from at least two separate digests. See Figure lfor the naming of the diol epoxide isomers. dAdduct configuration about C-1 is shown in parentheses. e ND = not determined; insufficient adduct formed. that were obtained with 1 X VPD compared to 15X VPD. In all cases except (RSSR)-BADE-2,where almost no cis adduct is formed, it appears that one member of the cistrans pair is released adequately whereas the other is not. Since the trans dA adduct obtained from (RSSR)-BADE-2 is released adequately, it is suggested that the cis adduct might not be. If that is true, then for every cis-trans pair, the adduct that is resistant to digestion by inadequate VPD has the R configuration about the attachment site, C-1 on the hydrocarbon. Unlike BADE, BcPhDE forms predominantly dA as opposed to dG adducts; the extent of adduct formation is far greater as well. Table I1 also shows the extent of release of BcPhDE-1 or BcPhDE-2 adducts when 1 X VPD is used compared to 5X, and how this shortfall varies with the cis or trans isomerism of the BcPhDE-dA adducts. Again, there is a substantial difference in resistance to VPD digestion observed within a cis-trans pair of dA adducts, and greater resistance is encountered if the hydrocarbon adduct has the R configuration at C-1. The original Randerath postlabeling procedure used digestion by SPD and MN. The number of units of the two enzymes called for by that procedure (see Experimental Procedures) is designated IX. In the procedure, the digestive enzymes are dialyzed against distilled water.

548 Chem. Res. Toxicol., Vol. 3, No. 6, 1990

Cheh et al. Table IV. Summary of Which Stereoisomeric dA Adduct Encounters Greater Resistance to Digestionn greater resistance to greater resistance to VPD SPD BA or BcPh diol epoxideb BADE BcPhDE BADE BcPhDE RSRS R R S S SRSR R R S S SRRS R R S R?' RSSR R?d R S?d R

SRRS

I1 5

I

I

I

15 TIME (min)

I

I

25

F i g u r e 5. HPLC comparison of the amounts of dA nucleoside adducts ohtained from DNA treated with racemic BcPhDE-2 when IX hIN/lx SPD or l x DNase/5x VPD is used prior to Al'. The .subscripts c and t indicate cis or trans adducts formed at dC, or dA, while the absolute configurations shown refer to the parent diol epoxide. The cross-hatched areas indicate the SPD shortfall. Table 111. Extent of Release of Individual BADE and B c P h D E dA Adducts by 1 X (Postlabeling Method) Compared to 5x VPD" 9 G of adduct released bv 1 X SPD

BADE iscimersb ( R S R S )- R A DE- 1 (SRSR)-BADE-l

(SRRS)-BADE-2 (RSSR)-RADE-2 RrPhDE isomers (RSRS)-RrPhDF,-I

(SRSR)-BcPhDE-I (SRRS)-BcPhDE-2 (RSSR)-RcPhDE-2

104 (R)' 79 ( S )

48 ( S ) 96 (R) 57 ( S )

76 (R) 38 is)

NDd (R)

94 ( R ) 76 (SI 71 (R) 59 (S)

38 (S) 106 (R) 80 (S) 37 (R)

O'I'he percentages are averages of the data from at least two different digests. bSee Figure 1 for the naming of the diol epoxide isomers. Adduct configuration about C-1 is shown in parentheses. d N D = not determined; insufficient adduct formed.

In our hands, this resulted in loss of up to half of the enzymatic activity. The amount of enzyme added was adjusted to compensate for this loss in activity. The BcPhDE-2 adduct profiles obtained after digestion with l x DNase I/5X VPD or l x S P D / l x MN are shown in Figure 5. The release of dA adducts by the postlabeling digestion method falls short of the release seen with 5X VPD. Figure 5 also shows that resistance to digestion is not simply a function of which adduct, cis or trans, is formed in greater yield; the (RSSR)-BcPhDE-2cis adduct, which is formed in the least quantity, is the most resistant to digestion. Still less dA adduct was released with 0.5X MN and SPD compared to IX, which would correspond to not compensating for the loss in enzyme activity upon dialysis. As long as the same amount of enzyme activity was used, microdrop dialysis ( 2 5 ) of the postlabeling enzymes or omission of the dialysis had little effect on the extent of' failure to liberate completely the BADE-dA adducts. Table I11 compares the extent of release of BADE or BcPhDE cis and trans adducts obtained with 1x M N / l x SPD compared to 1x DNase I / 5 X VPD. With the majority of pairs of cis-trans adducts, a greater failure of SPD to release the adduct occurs if it has the S configuration at the point of attachment, C-1. Comparisons were made between DNAs that were reacted with the usual amount of diol epoxide and 1 / 3 of that amount prior to digestion. After the digests were processed through Sep-pak recovery and the methanol

" R and S refer to the configurations of C-1 of the hydrocarbon adduct. See Figure 1 for the naming of the diol epoxide isomers. 'The difference in cis-trans resistance to digestion may not be larger than experimental error. Insufficient cis adduct formed to allow accurate comparison of whether the cis or the trans adduct is more resistant to digestion; the assignment is tentatively suggested from the behavior of the trans adduct alone.

eluate was evaporated, the latter samples were redissolved in 1 / 3 of the volume used for the former. Virtually identical chromatograms were obtained from the two DNA samples. Since the probability of having adjacent adducts is reduced when the extent of adduct formation is reduced, this suggests that clusters of adjacent adducts do not cause the resistance to digestion. Calcium(I1) is required by MN. Treatment of the deoxynucleotide 3'-phosphates released by MN + SPD with AP caused rapid production of a white precipitate, presumed to be calcium phosphate. The possibility that binding to this precipitate might be the source of poor recovery of dA adducts was ruled out when it was found that adding the same amount of Ca(1I) to the DNase I + VPD digests caused the same precipitate to form upon AP digestion but did not affect the yield of individual adducts. One caution is in order. The postlabeling digestions performed here represent a 40-fold scale-up of the standard procedure. While it is possible this could introduce resistance to digestion, it is suggestive that the shortfalls observed upon digesting BADE- and BcPhDE-adducted DNA with DNase I/VPD are quite similar to those obtained by Dipple and Pigott (17) with 20-fold less DMBADE-adducted DNA. The pattern of resistance to digestion exhibited by pairs of cis versus trans dA adducts is quite interesting. Table IV summarizes the pattern of resistance to digestion as a function of the stereoisomerism of the attacking diol epoxide, the adduct configuration ( R or S ) a t the point of attachment to the hydrocarbon (C-1, the N-substituted benzylic carbon atom of the hydrocarbon moiety), and the choice of exonuclease. With VPD digestion, greater resistance of dA adducts occurs if the adduct has an R configuration at C-1. With the 3'-exonuclease SPD, which cleaves in the opposite direction from the 5'-exonuclease VPD, in six out of eight cases greater resistance occurs if the adduct has the opposite, or S, configuration. A recent report notes that the revised postlabeling digestion procedure (ZO), which has replaced digestion by MN and SPD with digestion by nuclease P1 and potato apyrase followed by 5'-phosphorylation with polynucleotide kinase and digestion with VPD, significantly increases the yield of some, but not all, DMBA diol epoxide adducts (26). When these adducts are characterized, it will be interesting to note if the adducts that suffered the greatest digestion failure with MN + SPD have an S configuration about C-1. Structural characterization of diol epoxide-nucleoside adducts (ribose and deoxyribose) at the exocyclic amino groups of the purine bases has provided strong physical evidence that these adducts adopt preferred conformations in solution. The sign of an intense circular dichroism band at 25&290 nm is either positive or negative depending on

Stereoselective Digestion of PAH-Adducted DNA

whether the absolute configuration at the N-substituted benzylic carbon of the tetrahydroaromatic moiety is S or R (7). These circular dichroism bands arise from exciton interactions between the chromophores of the purine base and the hydrocarbon, and their signs are determined by the skew sense of the electric dipole moments of these chromophores. These conformational preferences, the exact nature of which is presently unknown but which probably involve opposite orientations of adducts having either the R or S configuration a t C-1, with respect to the base to which they are attached, are probably maintained in short single-stranded oligomers such as those produced by DNase I and MN. Thus, an alternating orientation of the adduct either facing the exonuclease or facing away, depending on whether the adduct is R or S a t C-1, could be the basis for the pattern of resistance encountered during exonuclease (VPD or SPD) digestion. On the other hand, the pattern of resistance to exonuclease digestion could be set by the orientation of the adducts in the intact DNA duplex prior to endonuclease cleavage. Assuming that the direction the adduct lies in the groove of intact DNA alternates depending on whether the adduct is R or S a t C-1, and assuming that endonuclease cleavage brackets an adduct, the resulting short oligomer product will have the adduct attached near or at one end of its DNA strand (5’ or 3’) if it has the R configuration and a t or near the opposite end if it has the S configuration. Resistance to exonucleolytic activity could then vary depending on whether the exonuclease encounters the adduct at the beginning or the end of its oligomeric substrate, independently of whether the adduct faces the exonuclease or faces away, or both factors could play a role. These data suggest that adducts having R versus S configurations at C-1 have opposite orientations in either single-stranded or duplex DNA or both. Since the R or S adduct configuration is determined by the stereochemistry of the attacking diol epoxide and the cis versus trans nature of epoxide opening (Figure 21, the orientation of the adduct would in turn be determined by these latter factors. Further experiments are needed to determine whether the resistance to digestion stems from the directionality of adduct orientation in the intact duplex or in the small oligomers resulting from endonuclease cleavage, or both, and what the absolute orientations of the adducts might be. [Jnderstanding the basis for the greater resistance to exonuclease cleavage by dA adducts relative to dG adducts will be aided by improved concepts of the steric constraints of the catalytic binding sites of these enzymes as well a5 more exacting information about the conformation of these adducts.

Acknowledgment. We thank Dr. Jane Sayer for reading the manuscript and for helpful discussions.

References (1) Perera, F. P. (1987) Molecular Cancer Epidemiology: A New Tool in Cancer Prevention. J . Natl. Cancer Inst. 78, 887-898. (2) Levin, W., Wood, A., Chang, R., Ryan, D., Thomas, P., Yagi, H., Thakker, D., Vyas, K., Boyd, C., Chu, S.-Y., Conney, A., and Jerina, D. (1982) Oxidative metabolism of polycyclic aromatic hydrocarbons to ultimate carcinogens. Drug Metab. Reu. 13, 555-580. (3) Jerina, D. M., Yagi, H., Thakker, D. R., Sayer, J. M., van Bladeren, P. J., Lehr. R. E., Whalen, D. L., Levin, W., Chang, R. L., Wood, A. W., and Conney, A. H. (1984) Identification of the ultimate carcinogenic metabolites of the polycyclic aromatic hydrocarbons: Bay-region (R,S)-diol-(S,R)-epoxides.In Foreign Conipoicnd Mctabo/i.jm (Caldwell, J., and Paulson, G. D., Eds.) pp 267-266, Taylor and Francis, London. (4) ‘I’hakker, D. H.,Yagi, H., Levin, W., Wood, A. W., Conney, A. H., and ,Jerina, D. M. (1985) Polycyclic aromatic hydrocarbons:

Chem. Res. Toricol., Vol. 3, No. 6, 1990 549 Metabolic activation to ultimate carcinogens. In Rioactiuation of Foreign Compounds (Anders, M. W., Ed.) pp 177-242, Academic Press, New York. ( 5 ) Meehan, T., and Straub, K.(1979) Double-stranded DNA stereoselectively binds benzo(a)pyrene diol epoxides. Nature 277, 4 10-4 12. (6) 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. (7) Sayer, J. M., Chadha, A,, Agarwal, S. K., Yeh, H. J. C., Yagi, H., and Jerina, D. M. (1990) Covalent adducts of benzo(a)pyrene 7,8-diol9,10-epoxides: Structural reinvestigation and characterization of a novel adenosine adduct on the ribose moiety. J . Org. Chem. (in press). (8) Chadha, A., Sayer, J. M., Yeh, H. J. C., Yagi, H., Cheh, A. M., Pannell, L. K., and Jerina, D. M. (1989) Structures of covalent nucleoside adducts formed from adenine, guanine and cytosine bases of DNA and the optically active bay-region 3,4-diol 1,2-epoxides of dibenz(a,j)anthracene. J . Am. Chem. Soc. 111, 5456-5463. (9) Jerina, D. M., Cheh, A. M., Chadha, A,, Yagi, H., and Sayer, J. M. (1988) Binding of metabolically formed bay-region diol epoxides to DNA. In Microsomes and Drug Oxidations: Proceedings of the 7th International Symposium (Miners, J. O., Birkett, D. J., Drew, R., May, B. K., and McManus, M. E., Eds.) pp 354-362, Taylor and Francis, London. (10) Cheh, A. M., Chadha, A., Sayer, J. M., Yagi, H., Yeh, H., Pannell, L., and Jerina, D. M., manuscript in preparation. (11) Reardon, D. B., Prakash, A. S., Hilton, B. D., Roman, J. M., Pataki, J., Harvey, R. G., and Dipple, A. (1987) Characterization of 5-methylchrysene-1,2-dihydrodiol-3,4-epoxide-DNA adducts. Carcinogenesis 8, 1317-1322. (12) .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 Press, New York. (13) Agarwal, S.K., Sayer, J. M., Yeh, H. J. C., Pannell, L. K., Hilton, B. D., Pigott, M. A., Dipple, A., Yagi, H., and Jerina, D. M. (1987) Chemical characterization of DNA adducts derived from the configurationally isomeric benzo(c)phenanthrene 3,4diol-1,2-epoxides. J . Am. Chem. Soc. 109, 2497-2505. (14) 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 327, 535-536. (15) Dipple, A,, Pigott, M. A., Moschel, R. C., and Constantino, N. (1983) Evidence that binding of 7,12-dimethylbenz(a)anthracene to DNA in mouse embryo cell cultures results in extensive substitution of both adenine and guanine residues. Cancer Res. 43, 4132-4135. (16) Baird, W. M., and Brookes, P. (1973) Studies on the formation of hydrocarbon-deoxyribonucleoside products by the binding of derivatives of 7-methylbenz(a)anthraceneto DNA in aqueous solution and in mouse embryo cells in culture. Cancer Res. 33, 2378-2385. (17) Dipple, A., and Pigott, M. A. (1987) Resistance of 7,12-dimethylbenz(a)anthracene-deoxyadenosine adducts in DNA to hydrolysis by snake venom phosphodiesterase. Carcinogenesis 8, 491-493. (18) Gupta, R. C., Reddy, M. V., and Randerath, K. (1982) 32Ppostlabeling analysis of non-radioactive aromatic carcinogen-DNA adducts. Carcinogenesis 3, 1081-1092. (19) Watson, W. P. (1987) Post-radiolabeling for detecting DNA damage. Mutagenesis 2, 319-331. (20) Randerath, K., Randerath, E., Danna, T. F., van Golen, K. L., and Putman, K. L. (1989) A new sensitive 32P-postlabelingassay based on the specific enzymatic conversion of bulky DNA lesions to radiolabeled dinucleotides and nucleoside 5’monophosphates. Carcinogenesis 10, 1231-1239. (21) Yagi, H., Vyas, K. P., Tada, M., Thakker, D. R., and Jerina, D. M. (1982) Synthesis of the enantiomeric bay-region diol epoxides of benz(a)anthracene and chrysene. J . Org. Chem. 47, 1110-1117. (22) 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-1,2-epoxides of benzo(c)phenanthrene. Tetrahedron Lett. 24, 1349-1352.

550 Chem. Res. Toxicol., Vol. 3, No. 6, 1990 (23) Procedures obtained from the Sigma Chemical Co. and Boehringer Mannheim Biochemicals Technical Services Departments. (24) Sawicki, J. T., Moschel, R. C., and Dipple, A. (1983) Involvement of both syn- and anti-dihydrodiol-epoxidesin the binding of 7,12-dimethylbenz(a)anthraceneto DNA in mouse embryo cell cultures. Cnncrr Rrs. 43, 3212-3218.

Cheh et al. (25) Marusyk, R., and Sergeant, A. (1980) A simple method for dialysis of small-volume samples. Anal. Biochem. 105, 403-405. (26) Yang, P.-F., and Randerath, K. (1990) High-resolution TLC

mapping and characterization of 32P-postlabeled monophosphate 7,12-dimethylbenz(a)anthracene-DNAadducts. Carcinogenesis 11, 159-164.