Chem. Res. Toxicol. 1992,5, 823-827
Characterization of (*)-7,8,10-Trihydroxy-7,8,9,10tetrahydrobenzo[alpyrene-9-sulfonate Justin L. Green and Gregory A. Reed* Department of Pharmacology, Toxicology, and Therapeutics and Center for Environmental and Occupational Health, University of Kansas Medical Center, Kansas City, Kansas 66160-7417 Received June 29, 1992
The genotoxicity of certain benzo[a] pyrene (BP) derivatives is significantly enhanced in strains of Salmonella typhimurium following addition of sulfite to the incubations. The interaction between sulfite and those B P derivatives also results in the formation of isomeric BP sulfonates. As these trihydroxy sulfonates are formed in incubations of BP derivatives and sulfite in which a marked potentiation of bacterial mutagenicity occurs, we have investigated the properties of these novel intermediates. The compound (*)-7,8,1O-trihydroxy-7,8,9,10tetrahydrobenzo[alpyrene-9-sulfonate(BPT-9-sulfonate)was isolated and characterized in terms of its chemical and biological activity. This BPT sulfonate isomer is formed by the addition of the sulfite anion radical to the 9,lO-double bond of the known promutagen, (*)-7,8-dihydroxy7,8-dihydrobenzo[alpyrene(BP-7,8-diol). Evidence for the free radical character of this addition includes the initiation of the reaction by either peroxidase-catalyzed or chemical one-electron oxidation of sulfite, the inhibition of the reaction by phenolic antioxidants, and the isolation and characterization of the chain termination product, 7,8-dihydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene-9,10-disulfonate(BPD-disulfonate). Analysis of incubations of S. typhimurium strain TA98 with BP-7,8-diol and sulfite, which resulted in a 10-fold increase in revertant bacterial colonies above control levels, showed that BPT-9-sulfonate and BPD-disulfonate were the only isolable products derived from BP-7,8-diol. This prompted a further investigation of the chemistry of these products. BPT-9-sulfonate was found to be quite stable in aqueous media, being refractory to acid- or base-catalyzed hydrolysis over a pH range of 3-11. Further, it was unreactive in aqueous incubations with a diverse range of chemical and biological nucleophiles of varying strengths. This BP-triol sulfonate was, however, capable of covalent binding to purified, double-stranded calf thymus DNA. Binding levels of 335 82 pmol/mg of DNA were achieved, which represented approximately 25 % of the levels obtained with 7r,8t-dihydroxy9t,10t-epoxy-7,8,9,1O-tetrahydrobenzo[al pyrene under identical conditions. Rather than representing a terminal or inactive detoxification product of BP-7,8-diol, these data suggest that BPT-9-sulfonate is a selectively reactive intermediate which may play a role in the enhanced genotoxicity observed during the sulfite-mediated activation of benzo[a]pyrene derivatives.
*
Introduction
form of S02, in bacterial mutagenicity assays activates (*)-7,8-dihydroxy-7,8-dihydrobenzo[alpyrene(BP-7,8Environmental and occupational exposure to complex diol) as a bacterial mutagen (5). These data support a mixtures of chemical contaminanta is of considerable role for SOZ,in the form of the sulfite anion, as a modulator interest in the assessment of risks to human health. The of the activation of BP derivatives to genotoxic forms. interactions between the common environmental pollutSuch enhanced activation is consistent with the observed anta sulfur dioxide (SO21 and benzo[alpyrene (BP)' are cocarcinogenic interaction. potentially important, as SO2 is thought to function as a Although sulfite-mediated epoxidation of BP-7,8-diol cocarcinogen with BP (1, 2). While SO2 itself is not is known to occur in chemical model systems (6),recent carcinogenic, chronic exposure to high ambient levels of data from this laboratory (7,8) suggest that additional SO2 correlates with an increased incidence of human reactions and products may also play a role in the enhanced respiratory tract neoplasia ( 3 , 4 ) . Moreover, concurrent genotoxicity seen in systems exposed to both sulfite and exposure to SO2 and BP potentiates the incidence of BP-7,8-diol. Curtis et al. have demonstrated that reactions respiratory tract carcinoma in rats (1)and hamsters (2) containingsulfite and BP-7,8-diolalso produce significant over that seen with BP alone. We have previously amounta of (f)-7,8,10-trihydroxy-7,8,9,lO-tetrahydrobendemonstrated that inclusion of sulfite, the physiological zo[al pyrene-9-sulfonate (BPT-9-sulfonate) (9). Formation of this metabolite has been reported in a number of 1 Abbreviations: B/E, magnetic sector flwdelectricsector field strength; biological systems, but it was previously thought to be BP, benzo[a]pyrene; BP-7,8-diol, (*)-trans-7,8-dihydroxy-7,8-dihydrobenzo[alpyrene; BPD-disulfonate, (+)-7,8dihydroxy-7,8,9,lC-~trahy- chemically and biologically inactive (9). We report here drobenzo[a]pyrene-9,lO-disulfonate;BPT-9-sulfonate, (f)-7,8,10-trithat BPT-9-sulfonateis a major end product in the sulfitehydroxy-7,8,9,lC-tetrahydrobenzo[a]pyrene-9-sulfonate; anti-BPDE, (*)dependent activation of BP-7,8-diolas a bacterial mutagen 7r,8t-dihydroxy-9t,lOt-epoxy-7,8,9,lO-tetrahydrobenzo[alpyrene; BPTIC-sulfonate,(~)-7r,8t,9t-trihydroxy-7,8,9,lC-tetrahydrobenzo~alpyren~ and that a second sulfonate product also is generated. We l0c-sulfonate; BP tetraol, (~)-7r,8t,9t,10e-tetrahydroxy-7,8,9,10- have identified this second product, and the mechanism tetrahydrobenzo[a]pyrene; CID, collision-induced decomposition; NIof formation of the sulfonate derivatives from BP-7,8-diol FAB, negative ion-fast atom bombardment; TBA, tetrabutylammonium has been characterized. These findings are discussed in ion. 0 1992 American Chemical Society
824 Chem. Res. Toxicol., Vol. 5, No. 6, 1992 relation to the observedgenotoxic effect of t h e interactions of sulfite with BP-7,8-diol.
Green a n d Reed
(i)-7r,8t-Dihydroxy-7,8,9,1O-tetrahydrobenzo[ alpyrene9,lO-disulfonate(BPD-disulfonate), A 25-mL volume of 0.1 M phosphate buffer containing 0.02% (w/v) Tween-20 was sparged vigorously with dry nitrogen gas for 45 min to remove Experimental Procedures dissolved oxygen. The following reagents were added sequentially Caution: BPand specific BPderivatiues are known mutagens to yield the indicated concentrations: 10 mM NazS03, 40 pM and carcinogens. All B P derivatives should be stored and BP-7,8-diol, 10pg/mL horseradish peroxidase, and 250 mM HzOz. handled with extreme care. Racemic BP-7,gdiol and 7r,8t-dihydroxy-9t,lOt-epoxy-7,8,9,10- The reaction was stirred in the dark, under nitrogen, at room temperature for 1h. The entire reaction mixture thenwas applied tetrahydrobenzo[a]ppene (anti-BPDE) were purchased from to a C-18 solid-phase extraction column. The column was washed ChemSyn Science Laboratories, Lenexa, KS. Tritiated BP with 3 mL of water, followed by elution with 3 mL each of 10% derivatives were supplied by ChemSyn Science Laboratories aqueous methanol, 50%aqueous methanol, and 100% methanol. through the NCI Repository Program. The synthesis of 7r,8t,9tAll fractions were evaporated to dryness under reduced pressure trihydroxy-7,8,9,l0-tetrahydrobenzo[alpyrenel0-sdfonate(BPTand redissolved in 10% aqueous methanol. Tetrabutylammo10-sulfonate) was conducted as described previously (8). Calf nium chloride (TBA) was added to each fraction to yield a final thymus DNA, polyguanylic acid, and reduced glutathione were concentration of 0.1 M. The solution was stirred vigorously for from Sigma Chemical Co. (St.Louis, MO), andp-nitrothiophenol 30 min, transferred to a separatory funnel, and then extracted was from Aldrich Chemical Co. (Milwaukee, WI). All other five times with 2 volumes of 1-butanol. The butanol phases from chemicals and solvents were purchased from Fisher Scientific each fraction were combined and evaporated to dryness under (St. Louis, MO). reduced pressure. Residues were resuspended in 50 % aqueous Analytical Methodology. All synthetic and analytical promethanol, and individual reaction products were then purified cedures were performed in subdued light. Products were stored by reverse-phase HPLC utilizing a Whatman 9- X 250-mm Partisil under dry nitrogen a t -20 "C. Analytical and preparative HPLC ODs-2 semipreparative column eluted with a methanoliwater was conducted using a 4.6- X 250-mm Ultrasphere ODS column gradient, as described previously. (Beckman Instruments Co., Fullerton, CA) or a Whatman Chemical Stability Studies. The susceptibility of BPT-9Magnum-9 ODs-2 column (Whatman, Inc., Clifton, NJ), resulfonate to hydrolysis or nucleophilic attack was determined. spectively. Elution with a methanol/water gradient and detection BPT-9-sulfonate (10 pM) was added to 200 pL of 20 mM conditions were described previously (8). UV spectra were potassium phosphate buffer (pH 7.4, p = 0.1 M KC1). For obtained using a Shimdazu UV-260 spectrophotometer and hydrolysis studies, this buffer was adjusted to pH 3,5,9, and 11. fluorescence spectra with a Shimadzu RF-540 fluorometer. All Reactivity of B P T - h u l f o n a t e at pH 7.4 was examined further spectra were measured in 95% ethanol. Negative ion fast atom by the addition of various nucleophiles to a concentration of 50 bombardment mass spectra (NI-FAB) were determined on a ZAB mM. The nucleophiles examined included ethanol, chloride HS double-focusing mass spectrometer equipped with a 11/250 (KCl), bromide (KBr), thiocyanate (KSCN), phenol, p-nidata system (VG Analytical, Ltd., Manchester, U.K.). NI-FAB trothiophenol, and the biological nucleophiles reduced gluwas carried out using a xenon atom gun operated at 8-keV energy tathione and polyguanylic acid. All reaction mixtures were and 0.8-mA emission. Collision-induced decomposition (CID) incubated a t 37 "C for 24 h. Following this incubation, the NI-FAB MS experiments were carried out in the negative ion hydrolysis reaction mixtures were neutralized by the addition of mode. Linked scans at a constant ratio of magnetic sector flux 100 pL of 0.2 M phosphate buffer (pH 7.4). The samples were to electric sector field strength (B/E) were performed with then analyzed by HPLC as described above. All assays were precursor ions attenuated to 60% with helium in the first field performed in triplicate. free region gas cell. The scan range was 750-10 amu acquired DNA Binding Studies. Silanized glassware was used at 10 s/dec, and 6 scans were acquired, at a collision energy of throughout the incubation and purification procedures. Calf 8 keV. thymus DNA was purified by exhaustive extraction with phenol/ Bacterial Mutagenicity Assays and Product Analysis. chloroform and precipitation with ice-cold ethanol. Purified calf Assays were performed using a liquid incubation procedure as thymus DNA was dissolved in 0.1 M potassium phosphate buffer described previously (8). Briefly, the bacteria were suspended (pH 7.4), at a concentration of 1mg/mL. Tritium-labeled antiin 1 mL of 0.1 M phosphate buffer (pH 7.4) at 37 "C, in the BPDE (302 mCi/mmol), BPT-9-sulfonate (57 mCi/mmol), and presence of 20 pM [3H]BP-7,8-diol. Where indicated, 10 mM BP tetraol(57 mCi/mmol) were dissolved in DMSO and added sulfite was added and the incubation continued for 30 min. A to the DNA. The concentration of the BP derivatives was 20 500-pL aliquot was removed for radiometric HPLC analysis, and pM, and the DMSO concentration was 1% (v/v). Control the remaining 500 pL of the incubation was plated and scored incubations contained 20 pM BPT-9-sulfonate without DNA. for mutagenicity. These reaction mixtures were incubated for 1h a t 37 "C and then BPT-9-sulfonate. Synthesis was accomplished based on the extracted with 10 volumes of buffer-saturated ethyl acetate to procedure of Curtis et al. (9). BP-7,8-diol(300 pg) dissolved in remove unbound BP derivatives. Pooled organic extracts from 10 pL of THF/TEA (99:l) was added to 1.0 mL of 2-propanol in each incubation were evaporated under reduced pressure and a micro reaction vessel. Calcium nitrate (20 mg) and sodium reconstituted in 250 p L of 50% aqueous methanol for HPLC sulfite (40 mg) were dissolved in 1.0 mL of triple-distilled analysis. DNA was denatured by heating at 100 "C for 10 min deionized water (pH adjusted to 6.5 with 6.0 N HCl), and the to release intercalated BP derivatives and then immediately mixture was added to the reaction vessel. The solution was placed on ice to prevent reannealing of the duplex DNA. DNA extensively sparged with dry nitrogen gas to remove residual was isolated by ethanol precipitation, and each ethanol-buffer oxygen. The sealed reaction vessel was stirred vigorously a t 85 supernatant was prepared as described above for HPLC analysis. "C for 72 h. The reaction mixture was applied to a C-18 solidThe DNA pellet was washed with 5 volumes of ice-cold ethanol phase extraction column (Analytichem International, Harbor and lyophilized. The purified DNA was resuspended in 10 mM City, CA) and washed with 6 mL of deionized water. Less than NaCl-1 mM EDTA and quantitated by absorbance at 260 nm 7% of BP-7,8-diol-derived material was lost in the water wash. [20 AU/(mg of DNAemL)], and the amount of incorporated label The remaining BP-7,8-diol-derived products were eluted from was quantitated by liquid scintillation counting. All incubations the column with 3 mL of ice-cold methanol. The methanoleluate were run in triplicate, and results are reported as mean f standard was evaporated to dryness under reduced pressure. The crude deviation. product mixture was dissolved in 1mL of 50 % aqueous methanol and purified by reverse-phase HPLC utilizing a 9- X 250-mm Results Partisil ODs-2 semipreparative column (Whatman) eluted with BP-7,8-diol is n o t itself mutagenic, and b o t h t h e a methanol/water gradient. Total yield of BPT-9-sulfonate was biological end point of mutation induction and t h e 37%.
Chem. Res. Toxicol., Vol. 5, No. 6, 1992 825
Benzo[a]pyrene Sulfonates
Table I. Characterization of BP-7.8-diol-DerivedProducts 20 rM BP-7,Bdlol 57 t 6 Revulantr/Plato
I!
50,000
1i
25,000
75,000
E
20 rM BP.7.8-diol
10 mM SO3= 573 t 63 Revertants/Plate
Figure 1. Sulfite-dependent activation and metabolism of BP7,8-diol. S. typhimurium strain TA98 was incubated with 20 gM PHIBP-7,8-diol,in either the presence (bottompanel)or the absence (top panel) of 10 mM sulfite. Scoring of mutagenicity was performed and is noted as the mean h standard deviation of triplicate determinations. The spontaneous revertant number observed was 31 f 5. Radiochromatographic analysis of BP7,8-diol-derived products was performed as described in the Experimental Procedures section. The chromatograms shown are representative of triplicate analyses. observed chromatographic profile demonstrate that Salmonella typhimurium are unable to activate this compound (Figure 1). In the presence of 10 mM sulfite, however, both of these end points were altered markedly. Efficient activation of BP-7,Sdiol as a mutagen was observed, and the radiochromatographic profile of the cellfree supernatant from those incubations demonstrated total consumption of the startingsubstrate. No BP tetraols were seen, and instead two more polar products were present (Figure 1). Isolated material eluting in the peaks at 2.2 and 11.7 min, when reconstituted in ethanol, exhibited UV spectra diagnostic for the presence of a pyrene chromophore. The maxima for absorbance, however, were at 347,330, and 315 nm for the material eluting at 2.2 min, and at 343, 327, and 314 nm for the latereluting compound. The bathochromic shift in the maxima for the 2.2-min peak, relative to the observed maxima for BP tetraols, can be diagnostic for the presence of a sulfonate group on C-10 of a tetrahydrobenzo[alpyrene derivative (8). Chromatographic and absorbance properties of the material eluting at 11.7 min were identical to those of BPT-9-sulfonate (8, 9). This identity was confirmed by synthesis of BPT-9-sulfonate, using a modification of the published procedure (91, and further characterization. Reaction of BP-7,8-diolwith sulfite and the calcium nitrate radical initiator (10) resulted in the formation of BPT9-sulfonate, with an isolated yield of nearly 40%. Unreacted BP-7,8-diol made up the remainder of the significant products detected. The major product from the BP-7,8-diol/sulfite/nitrate reaction (tR = 11.7 min) was dissolved in 0.1 M NH40H to provide a volatile counterion and then analyzed by NIFAB MS (Table I). This product demonstrated the expected molecular ion at m / z = 383. Absorbance spectroscopy of the product showed maxima a t 343.1, 327.0, and 313.0 nm, and fluorescence measurements demonstrated excitation maxima at 344.9, 328.9, and 315.2 nm and emissionmasimaat 398.5 and 380.1 nm (Em). Product
BPD-disulfonate
2.2
668 447
BPT-9-sulfonate 11.7
383
347.1 330.3 314.9 343.1 327.0 313.0
347.1 331.3 317.9 344.9 328.9 315.2
400.1 381.7 398.5 380.1
formation was highly dependent on the presence of nitrate and was inhibited by over 70% in the presence of 50 pM butylated hydroxyanisole, a phenolic antioxidant (data not shown). Since the more polar product observed in the bacterial extracts was not observed in the sulfitelnitrate synthetic system, an alternative synthetic approach was employed. Sulfite oxidation was initiated by the horseradish peroxidase/HzOz system under anaerobic conditions. This system generated amajor product with an HPLC retention time of 2.2 min, the same as that observed for the polar product from the bacterial incubations. Spectroscopic analysis of this product showed UV absorbance maxima at 347.1, 330.3, and 314.9 nm. Fluorescence spectra had maxima at 347.1,331.3, and 317.9 nm (Ex), and at 400.1 and 381.7 nm (Em). Mass spectrometry (NI-FAB, glycerol matrix) demonstrated prominent ions at mlz = 447 (M 1)and m / z = 668 (M + TBA- 1)which support the addition of two sulfonate groups to BP-7,B-diol. The ion at mlz = 668 was further examined by NI-FAB CID MS,and linked scans a t constant B/E of the 668 amu ion peak showed fragmentation into two primary peaks, one at 447 m u and another a t 503amu. The peak at 447 amu is consistent with the addition of two sulfonate groups (M - 11, while the 503 amu peak may be due to alkylation of the disulfonate with a butyl group transferred from the TBA counterion. The absorbance and fluorescencedata support the presence of a tetrahydrobenzo[alpyrenechromophore in this product, and the mass spectral characterization indicates the addition of two sulfonate groups. Taken together, the data support determination that the polar product is BPD-disulfonate. The conversion of BP-7,8-diol to sulfite addition products in the mutagenicity studies, but not to detectable quantities of diol epoxide-derived products, suggests that these addition products may themselves be chemically or biologically active BP derivatives. As a result, the stability of BPT-9-sulfonate as a function of pH and its susceptibility to attack by various nucleophiles were studied. Incubation of BPT-9-sulfonate a t pH ranging from 3 to 11 resulted in quantitative recovery of the compound, indicating hydrolytic stability over this range (data not shown). The susceptibility of BPT-9-sulfonate to nucleophilic attack was examined by performing aqueous incubations of the BP derivative with several model nucleophiles. The compounds examined were chosen to represent a wide range of nucleophilic activities, on the basis of their relative reactivities with methyl iodide (11). The biological nucleophiles reduced glutathione and polyguanylic acid were also tested. As was seen with the pH studies, however, no loss of BPT-9-sulfonate or appearance of new products was observed in any of these incubations (data not shown). Isolated quantities of BPDdisulfonate were not sufficient to carry out corresponding reactivity studies. The final study of the reactivity of [3H]BPT-9-sulfonate used calf thymus DNA as the target molecule. Parallel
826 Chem. Res. Toxicol., Vol. 5, No. 6,1992
Green and Reed
Table 11. DNA Binding by B[a]P Derivativesa BP derivative DNA binding, pmol/mg of DNA 335 f 82b BPT-9-sulfonate 1293f 40 anti-BPDE 27 k 32 BP tetraols Calf thymus DNA (1 mg/mL) in 0.1 M potassium phosphate buffer was incubated with 201M of each tritium-labeled BP derivative for 1h at 37 “C. DNA isolation was accomplished by ethyl acetate extraction followed by ethanol precipitation. Data represent means SD of triplicate determinations.
*
\so@ HO
on
-
Figure 2. Reaction pathways for BP-7,8-diol and sulfite derivatives.
incubations contained [3HlBPtetraol with DNA, or I3H1BPT-9-sulfonate in the absence of DNA. The DNA concentration was 1mg/mL, and all BP derivatives were added to a final concentration of 20 pM. Incubation of labeled BPT-9-sulfonate in the absence of DNA, followed by the standard extraction and ethanol precipitation procedure employed, resulted in no detectable tritium in the resuspended samples. This established that free BPT9-sulfonate was not carried over in the DNA isolation procedure. DNA isolated from incubations with labeled BP tetraol, however, did contain associated label (Table 11) equivalent to the presence of 27 pmol of the BP derivative/mg of DNA. Incubation of DNA with labeled BPT9-sulfonate, however, resulted in an apparent binding level of this derivative over 12-fold higher than the level observed with the nonreactive intercalating derivative, BP tetraol. The ability of BPT-9-sulfonate to bind to isolated double-stranded DNA, in contrast to the lack of reactivity with simple nucleophiles or polyguanylic acid, illustrates a strict and surprising specificity for this reactive BP derivative.
Discussion The initial studies of the reaction of sulfite and BP7,8-diol demonstrated the free radical oxidation of sulfite generating a sulfite peroxyl radical. This sulfite peroxyl radical served as an efficient epoxidizing agent for the conversion of BP-7,8-diol to the ultimate mutagenic and carcinogenic BP derivative, anti-BPDE (6). This reaction scheme (Figure 2: “Peroxyl Radical Pathway”) was supported by the isolation and characterization of the isomeric BP tetraols derived from anti-BPDE, and by the requirement for both molecular oxygen and sulfite oxidation in order for product formation to occur. Significant formation of water-soluble products also was noted, but those products were not isolated. Subsequent studies demon-
strated the ability of sulfite to activate BP-7,8-diol as a bacterial mutagen in incubations containing Ames’ tester strains of S. typhimurium (5). No analysis of BP-7,8diol-derived products was performed in that study, but an indirect characterization of the activation demonstrated a requirement for sulfite oxidation, paralleling the results in the chemical system (6). Curtis et al. first isolated the major water-soluble product from the reaction of sulfite oxidation products with BP-7,8-diol, and they tentatively identified this product as BPT-9-sulfonate (9). Their findings and results from this laboratory regarding the isomeric BPT-10-sulfonate (7, 8, 12) prompted further examination of the fate of BP-7,8-diol in the sulfitemediated mutagenicity studies, and of the mechanisms of these reactions. Our findings support the expanded scheme for the interactions of BP-7,8-diol and sulfite shown as the “AdditionPathway” in Figure 2. This reaction path shares the same initial reaction as occurs for the previously described peroxyl radical pathway, namely, the oxidation of the sulfite anion to form a sulfite anion radical. As an alternative to oxygen trapping and peroxyl radical formation, the sulfite anion radical itself may attack the 9,lOdouble bond of BP-7,8-diolto form the stabilized benzylic radical intermediate shown. This intermediate may either trap molecular oxygen to yield BPT-9-sulfonate via a peroxyl radical intermediate or undergo an additional electron transfer leading to a carbocationwhich traps water to yield the hydroxyl substituent. Formation of BPDdisulfonate through the coupling of a sulfite radical with the benzylic radical product is proposed as a chaintermination step. Several lines of evidence support the free radical nature of these reactions. First, all product formation was inhibited by phenolic antioxidant compounds. Second, product formation can be facilitated by either chemical or enzymatic systems which accomplish the one-electron oxidation of sulfite. The catalytic effect of CaN03 occurs due to the decomposition of nitrate to nitrogen dioxide (10). Nitrogen dioxide can abstract an electron from sulfite to form the sulfite anion radical, the key sulfite-derived species necessary for all of the pathways noted in Figure 2. The role of the nitrate ion in chain initiation is supported by the lack of significant product formation under the experimental conditions when CaN03 is omitted from the reaction mixture (data not shown). The second system shown to produce these sulfite-dependent BP derivatives from BP-7,&diol is peroxidase-catalyzed. The ability of either horseradish peroxidase (13) or the peroxidase activity of prostaglandin H synthase (14)to oxidize sulfite to the sulfite anion radical is well documented. The effect of oxygen tension on product distribution also is consistent with the radical nature of this pathway. Formation of BPD-disulfonate was particularly sensitive to inhibition by oxygen, as would be expected on the basis of the scheme shown. Both the sulfite anion radical and the postulated BPD sulfonate benzylic radical can be diverted from disulfonate formation by reaction with oxygen. Finally, the formation of the disulfonate product itself represents strong support for a radical addition mechanism. Although ionic addition of sulfite to olefins is known (101, these additions are freely reversible, and they cannot readily form an electrophilic intermediate for trapping of a second sulfite molecule. Similar formation of a disulfonate
Benzo[a]pyrene Sulfonates
termination product from the reaction of sulfite with a variety of simple olefins has been demonstrated (10). The ability of BPT-9-sulfonate to react selectivelywith double-stranded DNA, but not with a variety of other nucleophiles, parallels the behavior of the isomeric BPT10-sulfonate (8). Generation of an electrophilic center from BPT-9-sulfonate might result from an acid-catalyzed hydrolysis of the sulfonate moiety to form a transient carbocation at (2-9, but the stability of BPT-9-sulfonate over the pH range studied argues against this possibility. Direct nucleophilicdisplacement of the sulfonate also does not occur in aqueous incubations, even with a nucleophile as strong as the p-nitrothiophenolate anion. The overall stability of BPT-9-sulfonate to nucleophilic attack and to acid/base-catalyzedhydrolysis suggeststhat displacement of the sulfonate moiety may require a unique reaction environment not found in simple chemical systems. Isolation and structural characterization of actual BPT9-sulfonate-derived DNA adducts will provide invaluable insight into possible reaction mechanisms. If the adduct structure reflects nucleophilic attack and displacement of the sulfonate moiety at C-9,it might suggestthat areaction environment very different from that of the bulk aqueous medium is required for the reaction to progress. The helical structure of double-stranded DNA must provide a proper reaction environment which either heightens the electrophilicity of BPT-9-sulfonate or furnishes an extremely potent nucleophilic site. In conclusion, the present study summarizes the chemical characterization of BPT-9-sulfonate related to mechanism of formation and to structure and reactivity. The free radical pathway that converts BP-7,8-diol to BPT9-sulfonate and BPD disulfonate represents the only demonstrable transformation of this proximate mutagen in incubations where profound induction of mutagenicity occurs. Moreover, BPT-9-sulfonate itself showed substantial ability to covalently modify isolated doublestranded DNA. These data suggest that the formation of these sulfite addition products and their resultant covalent modification of DNA may contribute to the formation of a heritable genotoxic lesion. We suggest that the free radical-mediated formation of sulfite addition products, and their subsequent covalent interactions with cellular nucleic acids, may account, in part, for the enhanced
Chem. Res. Toxicol., Vol. 5, No.6, 1992 827
genotoxicity observed during the sulfite-mediated metabolism of BP-7,8-diol.
Acknowledgment. The work reported here was supported by NIH Grant ES-04092 (Awarded to G.A.R.) and by Biomedical Research Grant SO7 RR05373. References (1) Laskin, S.,Kuscher, M., Sellakumar, A., and Katz, G. V. (1976)
Combined carcinogen-irritant animal inhalation studies. In Air Pollution and the Lung (Aharanson, E. F., Ben-David, A., and Klingsberg, M. A., Eds.) pp 190-213,John Wiley and Sons, New York. (2) Pauluhn, J., Thyssen, J., Althoff, J., Kimmerle, G., and Mohr, U. (1985)Long-term inhalation study with benzo[alpyrena and SO2 in Syrian golden hamsters. Exp. Pathol. 28,31. (3) Lave,L.B.,andLiskin,E. P. (1970)Airpollutionandhealth.Science 169,723-733. (4) Higgins, I. T. T. (1971)Effects of the sulfur oxides and particulates on human health. Arch. Enuiron. Health 22, 584-590. (5)Reed,G. A. (1987)Sulfite-dependentmutagenicityofbenzo[alpyrene derivatives. Carcinogenesis 8,1145-1148. (6) Reed, G. A., Curtis, J. F., Mottley, C., Eling, T. E., and Mason, R. P. (1986)Epoxidation of (*)-7,8-dihydroxy-7,8-dihydrobenzo[al-
pyrene during (bi)suKte autoxidation: Activationof a procarcinogen by a cocarcinogen. Proc. Natl. Acad. Sci. U.S.A. 83,7499-7502. (7) Reed, G. A., and Ryan, M. J. (1990) Sulfite enhancement of diolepoxide mutagenicity: The role of altered glutathione metabolism. Carcinogenesis 11, 1635-1639. (8) Green, J. L., and Reed, G. A. (1990)Benzo[alpyrene bay-region sulfonates, a novel class of reactive intermediates. Chem. Res. Toxicol. 3,59-64. (9) Curtis, J. F.,Hughes, M. H., Mason, R. P., and Eling, T. E. (1988) Peroxidase-catalyzedoxidationof (bi)sulfite: reaction of free radical metabolites of (bi)sulfite with (*)-7,8-dihydroxy-7,8-dihydrobenzo[alpyrene. Carcinogenesis 9,2015-2021. (10) Norton, C. J., Seppi,N. F., and Reuter, M. J. (1967)Alkanesulfonate synthesis. I. Ion catalysis of sulfite radical-ion addition to olefins. J. Org. Chem. 33,41564165. (11) Pearson, R. G.,Sobel, H., and Songstad, J. (1968)Nucleophilic reactivity constants toward methyl iodide and trans-[Pt(py)2C121. J. Am. Chem. SOC. 90,319-326. (12) Green, J. L., Pan, Y. L., and Reed, G. A. (1991)Mutagenicity of benzo[a]pyrene bay-region sulfonates. Carcinogenesis 12, 13591362. (13) Mottley, C., Trice, T. B., and Mason, R. P. (1982)Direct detection
of the sulfurtrioxide radical anion during the horseradish peroxide hydrogen peroxideoxidationof sulfite (aqueoussulfur dioxide). Mol. Pharmacol. 22,732-737. (14) Mottley, C., Mason, R. P., Chignell, C. F., Sivarajah, K., and Eling, T. E. (1982)The formation of sulfur trioxide anion radical during the prostaglandin hydroperoxidase catalyzed oxidation of bisulfite. J. Biol. Chem. 257,5050-5055.
