[CANCER RESEARCH 50. 4611-4618. August 1. 1990]

Molecular Dosimetry of Polycyclic Aromatic Hydrocarbon Epoxides and DiolEpoxides via Hemoglobin Adducts1

Billy W. Day,2 Stephen Naylor,3 Liang-Shang Can,4 Yousif Sahali, Thanh T. Nguyen, Paul L. Skipper,John S. Wishnok, and Steven R. Tannenbaum5

Department of Chemistry, Division of Toxicology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

ABSTRACT

Ten reactive metabolites of five polycyclic aromatic hydrocarbons andstyrènewere investigated to determine the generality of ester adductformation with human hemoglobin in the form of RBC and hydrolysis tothe corresponding tetrahydrotetrols or dihydrodiols. No exceptions werenoted among the compounds tested, which included the anr/'-diol epoxidesof benzojajpyrene (Hal'), chrysene, and benzjajanthracene; the \rn-diolepoxide of BaP; a mixture of syn- and anr/-diol epoxides of benzo|e|-pyrene; and cpoxides of styrène,benzo[f jpyrene, BaP, and cyclopenta-

|c,rf|pyrene. A test of the propensity of the simplest benzylic epoxide,styrèneoxide, to form esters that hydrolyze via a /(u ' mechanism wasperformed. Hydrolysis of styrèneoxide-adducted hemoglobin in 11.'*<)at neutral pi I yielded 18Oincorporation results that suggest this mechanism of hydrolysis is opérantto a minor degree in styrèneoxide-hemoglobin ester adducts. A method was developed for the isolation andquantification of the polycyclic aromatic alcohols, which consists ofenzymatic proteolysis, immunoaffmity chromatography, and gas chro-matography-mass spectrometry or fluorimetry. The method allows forroutine analysis of hemoglobin from individual samples as small as 1 mlof whole blood. Analysis of blood from different human populationsrevealed that hemoglobin adducts of the anfi-diol epoxide of BaP dominated the spectrum of adducts formed by the selected metabolites.

INTRODUCTION

Epoxides and diol epoxides are the ultimate carcinogenicforms of many PAH.6 These oxiranes are produced by themetabolic activation of PAH by cytochromes P-450 and have

Received 1/12/90; revised 4/10/90.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This investigation was supported by DHHS Shared Instrument ProgramGrant 1-S10-RR1901 and NIH Grants ES01640. ES02109. ES04675, andCA44306.

2Supported by postdoctoral fellowships from NIH Grant ES 17020 and American Cancer Society Grant SIG-11-1.

3 Present address: MRC Toxicology Unit. Carshalton. Surrey SMS 4EF.England.

4 Present address: Glaxo Inc.. Research Triangle Park, NC 27709.' To whom requests for reprints should be addressed, at Room 56-309. Mas

sachusetts Institute of Technology. Department of Chemistry. Division of Toxicology. Cambridge, MA 02139.

6The abbreviations used are: PAH, polycyclic aromatic hydrocarbon(s);BaADE-I. (±)-rrani-3,4-dihydroxy-anr/'-l,2-epoxy-l,2,3,4-tetrahydrobenz[a]an-

thracene; BaADE-11, (±)-rrani-8,9-dihydroxy-anf/-10.1 l-epoxy-8.9,10,11-tetra-hydrobenz[fl]anthracene;BaAT-l, l,2,3,4-tetrahydroxy-1.2.3,4-Ietrahydrobenz(a]-anthracene; BaAT-II, 8,9.10,1 l-tetrahydroxy-8.9.10,1 l-tetrahydrobenz[a]anthra-cene; BaP45D. benzo[u]pyrene-4.5-dihydrodiol: BaP-7.8-diol. benzo[a]pyrene-fran.c-7,8-dihydrodiol; BaP-9,10-diol. benzo(fl]pyrene-rram-9,10-dihydrodiol;BaP45E. henzo|a)pyrene-4.5-dihydroepoxide; onr/'-BaPDE, (±)-7/j,8«-dihydroxy-

9«.10«-epoxy-7,8.9,10-tetrahydrobcnzo[a]pyrene: iyn-BaPDE, (±)-70.8«-dihy-droxy-9rf. 1Orf-epoxy-7,8,9.10-tetrahydrobenzo[a]pyrene; BaPT. 7,8.9.10-tetrahv-droxy-7,8.9.10-tetrahydrobcnzo|a|pyrene: BePD, benzo(e|pyrene-9,10-dihydro-diol; BePDE. 9.10-dihydroxy-l l,12-epoxy-9,10.1 l,12-tetrahydrobenzo[r|pyrene;BePE, benzo[fjpyrene-9.10-dihydroepoxide: BePT, 9.10,11,12-tetrahydroxy-9,10,1 l,12-tetrahydrobenzo(e]pyrene; CDE, (±)-rnuu-1.2-dihydroxy-afirf-3,4-epoxy-l,2,3,4-tetrahydrochrysene; BaP, benzo[a]pyrene; CPPE, cyclopenta[c,i/]-pyrene-3,4-dihydroepoxide; CPPD. cyclopenta[c,rf]pyrene-3,4-dihydrodiol; CT,l,2,3,4-tetrahydroxy-l,2,3,4-tetrahydrochrysene; El, electron ionization; FaT,1Ob-//-1.2,3,1 Ob-tetrahydroxy-1,2,3-trihydrofluoranthene; IAC. immunoafTinitychromatography; NICI. negative ion chemical ionization; PBS. phosphate-buffered normal saline: PCI, positive chemical ionization; SD. styrene-7.8-dihydro-diol; SO, styrene-7,8-oxide; SFS, synchronous fluorescence spectrometry; TMS.trimethylsilyl; GC. gas chromatography; MS, mass spectrometry; HPLC. highperformance liquid chromatography.

been shown to react with DNA via nucleophilic attack of DNAbases, usually at the more electropositive (Lewis acidic) benzyliccarbon of the epoxide (1, 2). Similar reactions of alkylatingcompounds occur with the nucleophilic (thiol, thioether, amino,carboxylate, and hydroxyl) sites of hemoglobin, and it has beenproposed to use the adducts of this readily obtainable proteinas a dosimeter for exposure and metabolism (3).

The generation of tetrols from DNA as proof of the priorexistence of adducts is legitimately based on the known susceptibility of the A^-guanosine adducts of BaPDE to hydrolysis

(4). Until recently, evidence for the reaction of metabolicallyactivated PAH or of synthetic epoxides with hemoglobin wasconfined to the analogous recovery of BaP tetrols followingtreatment of the protein with acid. The uncertainties inherentin this approach as applied in an experimental context havebeen discussed (5, 6). However, RBC isolated from normalhuman blood donors are unlikely to be contaminated with BaPtetrols or shorter lived nonhemoglobin adducts. Thus the isolation and identification of tetrols (7) from human hemoglobinhave provided convincing evidence for the formation of adductswith this protein.

Direct evidence for the chemical nature of PAH-hemoglobin

adducts in terms of structure and location in the protein islimited. Styrèneoxide has been shown to alkylate «-histidine-20 and /3-histidine-143 by mass spectrometric analysis of trypticfragments of adducted hemoglobin (8). The presence of carbox-ylic esters formed by anti-BaPDE alkylation of one or morecarboxylate groups has been demonstrated by fluorescence line-narrowing spectroscopy (9).

The formation of carboxylic esters by anf/'-BaPDE has also

been demonstrated by isotope incorporation (5, 6). These experiments made clear the labile nature of the ester adduct. Theyalso revealed that ester adducts are by far the most abundantadducts formed by ûH/7-BaPDE.Predominant alkylation ofcarboxylate groups and lability of the esters formed are bothphenomena which can be attributed to the nature of the carbon-ium ion formed at C-10 by, respectively, opening of the epoxidein a/if/'-BaPDE or cleavage of the C-10—O bond of the ester.

We herein report a methodological approach for the quantitative analysis of the tetrols released from carboxylic estersformed in vivo between human hemoglobin and the diol epoxides of BaP. The methodology is based on previous applicationsof immunoaffinity chromatography (10, 11) and fluorescencespectroscopy (12) to carcinogen adduct analysis. The application of negative ion chemical ionization mass spectrometry totrimethylsilyl derivatives of PAH alcohols is novel. Since it isreasonable to expect that other PAH epoxides and diol epoxideswill also alkylate carboxylate groups to form similarly labileesters, we broadened the scope of this investigation to encompass a variety of structures, including the simplest benzylicepoxide, styrèneoxide. As part of the method development, wedetermined the capillary GC-MS detection limits for the diolsand tetrols derived from hydrolysis of the PAH-hemoglobinesters and have determined the affinities of these alcohols for

4611

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MACROMOLECULAR ADDUCTS OF CARCINOGENS

the monoclonal antibody used in their isolation from enzymaticdigestions of hemoglobin.

MATERIALS AND METHODS

Materials. (±)-rra/M-l,2-Dihydroxy-an//-3,4-epoxy-l,2,3,4-tetrahy-drochrysene, (±)-frans-3,4-dihydroxy-anf/'-l,2-epoxy-l,2,3,4-tetrahy-

drobenz[a]anthracene, (±)-/rani-8,9-dihydroxy-a/i/(-10,1 l-epoxy-8,9,10,11 -tetrahydrobenz[fl]anthracene. (±)-7/S,8«-dihydroxy-9«,1Oa-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene, (±)-7/3,8a-dihydroxy-9/3,10/3-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene, benzo[a]pyrene-4,5-dihydroepoxide, ben7.o[e]pyrene-lrans-9,10-dihydrodiol, benzo[a|-pyrene-fra/ii-9,10-dihydrodiol, and benzo[a]pyrene-/ra/ii-7,8-dihydro-diol were purchased from the National Cancer Institute ChemicalCarcinogen Repository maintained by the Midwest Research Institute(Kansas City, MO). [7(n)-3H]Styrene-7,8-oxide (specific activity, 99

mCi/mmol) was purchased from Amersham Corporation (ArlingtonHeights, IL). 9,10-Dihydroxy-11,12-epoxy-9,10,11,12-tetrahydro-benzo[?)pyrene was synthesized in quantitative yield from BePD byepoxidation with m-chloroperbenzoic acid in tetrahydrofuran, by themethod of Yagi et al. (13). The BePDE was isolated as a 4:1 mixtureof anti:syn isomers, as determined by HPLC analysis of a neutralhydrolysate of the epoxides. Benzo[e]pyrene-9,10-dihydroepoxide wasprepared as an off-white gum in 82% yield from 2 mg of BePD bycyclization of the monotosylate of the diol with excess powdered NaOHin dry monoglyme at room temperature (2 h) (14), filtration, concentration under a stream of dry N2, and rapid purification by flashchromatography (N2) over 10 g A12O3(19/1 hexane/tetrahydrofuran).Purity was 97% as determined by GC-MS (EI; m/z 268, M+). Cyclo-penta[c,i/]pyrene-3,4-epoxide was prepared in our laboratory, and itsfull synthesis has been reported elsewhere (15). The three \Ob-H-l,2,3,10b-tetrahydroxy-l,2,3-trihydrofluoranthenes were obtained byincubating fluoranthene with rat microsomes and isolating the alcoholsby C-18 HPLC, as per previously described methods (16). The remaining diols (cis- and /rani-BaP45D, cis- and trans-CPPD, and SD) andtetrols (cis- and trans-anti-CT, cis- and trans-anti-BaAT-l, cis- and trans-anf/'-BaAT-II, cis- and trans-anti-BaPT, cis- and trans-syn-BaPT, and

the four BePTs) were obtained via hydrolysis of their correspondingepoxides in 0.9% aqueous formic acid/tetrahydrofuran or H2O/tetra-hydrofuran (99/1), extraction into ethyl acetate, and purification byHPLC on a C-18 column. Trisyl Z (1.5 meq /V-trimethylsilylimidazole/ml pyridine), purchased from Pierce (Rockford, IL), was used as thesilylation reagent at room temperature (5-10 n\; reaction time, <5min), and 1fj\ of this mixture was injected directly into the GC. PronaseE (protease type XXV) was obtained from Sigma Chemical Co. (St.Louis, MO). Sephadex G-25 was purchased from Pharmacia FineChemicals (Piscataway, NJ).

Instrumentation. All samples were analyzed using a Hewlett-Packard5987A GC-MS with a standard El/chemical ionization source. Theinstrument was manually tuned daily with perfluorotributylamine formaximum sensitivity. Methane (99.999%; Med-Tech, Medford, MA)was the moderating bath and reagent gas for chemical ionizationexperiments. For the negative ion chemical ionization experiments thesource pressure was run at 0.4 to 0.6 torr, while for positive chemicalionization experiments the source pressure was 0.8 to 1.0 torr. Theelectron energy was 170 eV and the source temperature was 150°CforNICI experiments, 170 eV and 200°Cfor PCI experiments, and 70 eVand 240°Cfor El experiments. The emission current was 310 to 370

¿íA.Dwell times for selected ion monitoring runs were 200 ms/ionmonitored. The gas Chromatograph was interfaced to the MS with adirect capillary interface maintained at 240°C.The injection port andtransfer lines were operated at 250°C.The column was a 30-m x 0.25-

mm DB-1 fused silica capillary (J & W Scientific, Rancho Cordova,CA). Ultra-high purity helium (Med-Tech) was the carrier gas, using aflow rate of approximately 2 ml/min. Both the He and CH4 gas lineswere fitted with purifiers (no. 6406; Matheson, Secaucus, NJ). Samples were injected in the splitless mode (splitless valve open for 0.5min). SD extracted from in vitro globin digests and in standard solutionswas analyzed as its trimethylsilyl derivative with an initial GC oventemperature of 100°Cfor 1 min, followed by a temperature ramp to

280°Cat 20°C/min,with a 1-min isothermal phase at 280°C.All other

in vitro samples and standards were analyzed with an initial temperatureof 150°Cfor 1 min, temperature ramp to 330°Cat 20°C/min,and a 4-min isothermal phase at 330°C.In vivo samples were analyzed by

various temperature gradients, all culminating in a final isothermalphase at 330°Cfor 4 min. Quantitation and determination of the isomer

ratios of the diols/tetrols were accomplished by monitoring the M —162 (tetrols), M - 90 (diols), or M - 103 (SD) fragments of therespective compounds and comparing the measured Hewlett-Packardsoftware-generated corrected area of individual ions to a brief standardcurve generated on the same day for the compound. Detector responsewas linear over the fmol to nmol range.

HPLC analyses were performed on a Hewlett-Packard 1090 liquidChromatograph controlled by a Hewlett-Packard 300 Chem Station.UV detection was by a Hewlett-Packard 1040 diode array detector. AC-18 Nova-Pak (Waters, Milford, MA) reverse phase column, utilizing

a gradient of 0.1% trifluoroacetic acid/H2O (solvent A) and 0.1%trifluoroacetic acid/acetonitrile (solvent B) (time 0, 85% A; 10 min,75% A; 34 min, 72% A; 35 min, 72% A; 65 min, 70% A; 75 min, 100%B) at a flow rate of 0.7 ml/min, was used in the separations of both theformic acid hydrolysates of the epoxides and the extracts of the digests.Determination of the more hydrophilic SD was by a gradient of: time0, 100% A; 5 min, 85% A; 15 min, 75% A; 39 min, 72% A; 45 min,72% A; 75 min, 30% A; 80 min, 100% B. A Brownlee 1-mm x 15-cmC-18 column (Rainin Instruments, Woburn, MA) was used for micro-bore LC at a flow rate of 0.15 ml/min. The detector was operated at asampling interval of 640 ms over the range of 200-400 nm. Peaksignals were determined over a 4-nm window centered on the wavelength appropriate for the given chromophore, with the baseline reference a 100-nm window centered at 550 nm. Spectra of peaks of >0.2milli-absorbance units at the primary monitoring wavelength werestored by the Chem Station for later analysis. Determination of productratios was by comparison of the integrated peaks monitored at the Xm.»of the analyte. Quantitation was achieved by comparing integrated peakareas to a standard curve generated for each compound. The integrateddetector response was linear over the pmol to nmol range.

Fluorescence spectra were obtained using a SPEX Fluorolog 2 dou-ble-monochromator spectrofluorometer at room temperature (17).

Competitive Radioimmunoassays. Radioimmunoassays using the8E11 monoclonal antibody (a generous gift from Dr. R. Santella,Columbia University) were performed in triplicate by a standard(NH4)2SO4 precipitation method, using 7,8,9-trihydroxy-10-(2-[3H-ac-etamido]ethylthio)-7,8,9,10-tetrahydrobenzo[a]pyrene (specific activity, 3.3 Ci/mmol) as the tracer. Calculations of percentage of inhibitionwere made using the method of Muller (18).

Analysis of Human Blood Samples Reacted with Epoxides in Vitro.Twice washed RBC pelleted at 3000 x g from 1 ml of freshly obtainedhuman blood were suspended in 2 ml PBS and treated with 350 nmolof the epoxide in 35 n\ of dry tetrahydrofuran. The mixture wasincubated at 37°Cwith agitation for 1 h. The cells were then isolatedby centrifugation and lysed at 4°Cin 1 volume of double distilled H2O.

Cell debris was removed by centrifugation at 10,000 x g for 10 min.The lysate was cooled at 4°Cand added dropwise to 300 ml of rapidlystirred 0.015% HCl/acetone maintained at < -10°C. The precipitatedglobin was pelleted by centrifugation, dried overnight at 25°C,dissolved

in 40 ml H2O, and washed with 4 x 1 volume ethyl acetate followed by4x1 volume H2O-saturated 1-butanol to remove noncovalently boundPAH alcohols. Parallel incubations using 350 nmol of the corresponding diols or tetrols confirmed that this procedure removed all of thenoncovalently bound alcohols formed from simple water and phosphatehydrolyses of the epoxide. The 1-butanol extraction step also removessulfur adducts derived from reaction of the epoxide with glutathionethat co-precipitate with the globin. The globin solution was treated with4 ml lOx PBS and the pH was adjusted to 8.0 with 2 M NaOH. [In theSO 18Oincorporation experiments the globin solution was lyophilizedand redissolved in 2 ml 95-98% isotopie purity H2I8O (Aldrich Chem

ical Co., Milwaukee, WI), and the pH was adjusted to 8 with NH4CO,.]The protein was digested with 10% (w/w) Pronase (5% at 0 h, followedby another 5% at 6 h, after readjusting the pH to 8.0) for 12-16 h at37°Cwith stirring. The digest was extracted with 4 x 1 volume of ethyl

4612

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MACROMOLECULAR ADDUCTS OF CARCINOGENS

acetate. The high H2O solubility of SD required that the ethyl acetatebe supplemented with 10% of the H2O-saturated 1-butanol for itscomplete extraction from the protein digest solution. (In the SO 18O

incorporation experiments the fully extracted digest was additionallytreated with 200 ^1 triethylamine and incubated at 37°Cfor 1 h,

followed by extraction with the H2O/l-butanol/ethyl acetate mixture.)Alternatively, certain of the digest solutions were continuously extracted overnight with I volume of chloroform to yield the same results.The organic layers were combined and concentrated at 25°Con a rotary'

evaporator. The concentrated extract was resuspended in 9/1 H2O/methanol and subjected to HPLC analysis. Chromatographie fractionswere collected at the known retention times of the standards andconcentrated on a Speed-Vac lyophilizer.

Analysis of Human Hemoglobin for in Vivo Adducts. Six I ml RBClysate samples from a group of cigarette smokers (19) were pooled andfiltered through a 3- x 30-cm Sephadex G-25 column, and the globinwas precipitated by the acidic acetone method. Thirty 1-ml bloodsamples obtained from nonsmoking mothers near term at a localhospital (courtesy of Drs. Peter Gann, Jacalyn Coghlin, and S. Katherine Hammond of the Department of Family and Community Medicine, University of Massachusetts Medical Center, Worcester, MA)were treated and analyzed on an individual basis. The globin samplewas dissolved in H2O at a concentration of 3 mg protein/ml, thesolution was made Ix with lOx PBS, and the pH was adjusted to 8.0to 8.3 with 2 M NaOH. Pronase was added (ratio of substrate toenzyme, 20:1) and the mixture was incubated with stirring for 6 h at37°C.The pH was readjusted to 8.3 with 2 M NaOH, an additional

aliquot of Pronase was added, and the incubation was continued for 6-12 h. Immunoaffinity chromatography using the 8E11 monoclonalantibody covalently linked to CNBr-activated Sepharose 4B (Sigma)was carried out at 100 to 150 mg of digested protein solution/ml ofcolumn (11). After the digest solution was applied to the column, thecolumn was washed with PBS, and the immunoreactive fraction waseluted with 9/1 methanol/H2O. The solution was concentrated, redis-solved in 15-25 ml of PBS, and reconcentrated on a second 0.25-mlIAC column. Immunoreactive material was concentrated and analyzedby synchronous fluorescence spectroscopy with a AX= 34 nm (17, 20).The amount of pyrene-like fluorescent species present in each samplewas calculated by comparison to a standard curve generated with trans-a/ifi-[14C]BaPT solutions of known concentration.

RESULTS

Ester Formation with Human Hemoglobin in Vitro. To ourknowledge there have been no reports on the interaction ofPAH epoxides and diol epoxides, other than those of BaP, withhuman hemoglobin. Since we were interested in the possibilityof analyzing in vivo hemoglobin samples for adduction by avariety of PAH at the same time we were analyzing for anti-BaPDE adducts, we undertook a limited structure-activity relationship study to evaluate the scope and generality of esterformation. Experiments were performed in vitro with preformedepoxides and diol epoxides to eliminate the confounding effectof metabolism. Various PAH epoxides and diol epoxides wereincubated with human RBC resuspended in PBS. The adductedglobin was isolated from the RBC lysates by precipitation inacidic acetone, freed of noncovalently bound PAH alcohols byexhaustive extraction with ethyl acetate and 1-butanol, andcompletely digested with Pronase under weakly basic to neutralconditions. The digest was extracted with ethyl acetate orchloroform and the extract was analyzed by C-18 HPLC and/or GC-MS. The amount of alcohol(s) recovered by this methodwas considered to represent the amount of ester formation bythe epoxide tested. The amounts of hemoglobin esters formedby each PAH epoxide or diol epoxide are summarized in Table1. The yield of other types of adducts was not determined, sincethe compounds used were not radiolabeled.

The relative yields of esters varied over a range of 200-fold.

Table 1 PAH alcohols released from human hemoglobin previously reacted withPAH epoxides and comparison of their stereochemistry with that of PAH alcohols

produced by hydrolysis of PAH epoxides

EpoxideSOCBaADE-ICDEa/m'-BaPDEBePDEyn-BaPDEBaADE-IICPPEBePEBaP45E%

oftotalepoxidere

actedreleasedfromestersasPAH

alcohols"2.70.9750.4020.2470.0770.0510.0240.0200.0150.014Trans:cis

ratioofextractedPAH

alcohols*NA"4.8:15:13.2:13.8:1'.

\:t>f1:2.56:1NjyND1.7:1Trans:cis

ratioofformicacidhydrolysisNA24:126:119:120:1',

1:14/1:1260:1>80:1>80:1>80:1

" Extracted with ethyl acetate from Pronase digest of human globin.* Determined from the ratios of integrated diode array-detected HPLC peaks

or of NICI-GC-MS peaks of the pertrimethylsilyl derivatives.' Ethyl acetate supplemented with 10% H2O-saturated 1-butanol.d NA, not applicable.' Tetrols identical to those derived from the anf/'-diol epoxide.•^Tetrolsidentical to those derived from the syn-d\o\ epoxide.* ND, not determined.

SO produced the greatest amount of esters, yielding over 2.5times as much alcohols releasable by proteolysis/hydrolysis asthe next most active compound, BaADE-I. All of the diolepoxides formed more ester(s) than did any of the epoxidesother than SO in our in vitro system. Overall, the order was:SD > BaADE-I > CDE > anf/'-BaPDE > BePDE > sy/z-BaPDE

> BaADE-II > CPPE > BePE > BaP45E. This order closelyresembles the inverse of the relative rates of hydrolysis of theepoxides.

Table 1 also gives the ratio of trans to CMalcohols extractedfrom the proteolysis of the in vitro PAH-hemoglobin esters.Trans and cis refer to the relative stereochemistry of the hy-droxyl moieties at the two carbons which were originally partof the oxirane system. The transicis ratio of the oxygen atomsin the ester adduct is determined by the relative amount of transto cis attack of the carboxylate aniónon the epoxide. The samerelative rates of attack by a nucleophile will be observed in theformic acid-catalyzed ring opening of the epoxide. The trans:cisratio of alcohols thus produced will, therefore, be the same asthe ratio of alcohols produced from adducted protein if theadducts are cleaved only by acyl hydrolysis mechanisms. If theBM_\ mechanism participates to any degree, it will serve toepimerize the carbon center from which the ester is formed and,thereby, drive the trans:cis ratio of alcohols produced nearer tounity (Fig. 1).

The ratios obtained from the proteolysis solutions and thecorresponding values obtained from formic acid hydrolysis ofthe epoxides/diol epoxides at pH 3 are compared in Table 1.The trans:cis ratios (or cis:trans in the case of s^n-diol epoxides)for the formic acid hydrolyses are uniformly higher than thoseseen in the hemoglobin proteolysis extracts. A general idea ofthe contribution of Z?AL!hydrolysis to each ester adduct can beinferred from the data, but we have closely examined thesediffering ratios only in the case of awi/'-BaPDE ester adducts

(6).All of the compounds tested appear to form esters with

human hemoglobin. While the yields varied considerably,within the more limited set of compounds consisting only ofdiol epoxides, the yield of ester was within 1 order of magnitudeof the yield of ester from awii'-BaPDE, a benchmark compound.

Thus, reactivity toward hemoglobin should not be the criticalfactor determining whether ester adducts can be observed inhuman blood specimens. Actual exposure to the parent hydro-

4613

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MACROMOLECULAR ADDLICTS OF CARCINOGENS

R, or R and R' aromatic

cyt P-450

Protein-COn"

YProtein

R' OH

0-•H

: OHH

Protein

BAC?

or

'- OH

H

t l

;iH

cis

OH

Htrans

Fig. 1. Formation of benzylic esters by reaction of PAH epoxides with car-boxylate groups on protein and subsequent hydrolysis to alcohols by differentmechanisms. The classification of mechanism is according to Ingold (27), inwhich A and B refer to acid and base catalysis, the subscripts AC and AL indicatewhether the bond between oxygen and the acyl or alkyl carbons is cleaved, and /and 2 are used to designate that the mechanism is unimolecular or bimolecular.cyt, cytochrome.

carbon as well as the exposure of RBC to epoxides and diolepoxides in vivo are more likely to be the major determinantsof adduci levels.

StyrèneOxide-Human Hemoglobin Ester Formation in Vitro.The formation of substantial amounts of m-alcohols from thevarious PAH-hemoglobin ester adducts during proteolysis suggested that the ßALlmechanism of PAH-hemoglobin esterhydrolysis may be general in nature for the benzylic PAHadducts. This hypothesis was tested by measuring the incorporation of I8O into styrene-7,8-dihydrodiol as the result of pro

teolysis of hemoglobin which had reacted with SO, since SO isthe benzylic epoxide least likely to form esters which wouldhydrolyze via the AAL1mechanism.

El mass spectrometry was used for analysis of SD in the formof its di-TMS derivative. NICI-MS is not a sensitive techniquefor this derivative of SD. The molecular ion in the El spectrumwas observed in low abundance, as was the M —CH, fragment.The abundances of these ions were insufficient for the determination of isotope incorporation. An abundant fragment ionwas present at m/z 147, which corresponds to (CH3)2Si+—O—

Si(CH.,)., and is characteristic of the di-TMS derivatives ofvicinal diols. The base peak at m/z 179 results from C-7—C-8bond cleavage and retains the oxygen attached at C-7.

The incorporation of I8O could only be reliably detected from

the mass fragmentogram if it occurred at C-7. Only estersformed at this site would be expected to undergo hydrolysis bya carbonium ion mechanism. A 4% increase in the m/: 181 ionof SD, as well as a 5% 18O isotope abundance in m/z 149,occurs as a consequence of proteolysis of SO-adducted globinin H2'8O, clearly indicating at least a minor BM\ componentin the hydrolysis of the C-7 SO-hemoglobin esters.

8E11 Monoclonal Antibody Cross-Reactivity. It has previously-

been shown that 8E11 monoclonal antibody exhibits significantcross-reactivity with DNA adducted by BaADE-I, BaADE-II,and CDE (21). Since other PAH epoxides in addition to anti-BaPDE formed esters with hemoglobin in vitro, we becameinterested in the possibility that the 8E11 monoclonal antibodyused in the IAC would also recognize PAH diols/tetrols otherthan those formed from A-ring metabolites of BaP (22).

We determined the cross-reactivity of 8E11 with several PAHalcohols by radioimmunoassay. Results are shown in Table 2.The group of compounds tested was too limited to develop ameaningful structure-binding relationship between PAH alcohols and the 8E11 monoclonal antibody. With the exception ofthe A-ring BaP dihydrodiols, PAH diols had poor (trans-Ba-P45D and trans-BePD) to essentially no (CPPD and SD) affinity for the antibody. High affinity was noted for all of the PAHtetrahydrotetrols tested, with the interesting exception of thethree FaTs, which had no measurable affinity. The sv/i-BaPDE-derived tetrols bound with high affinity to 8E11, even thoughit has been reported that the DNA adducts of this diol epoxideare not recognized by the antibody (21).

Mass Spectrometry and Detection Limits. Mass spectral characteristics have been reported for derivatives of a variety ofPAH metabolites. The El-mass spectral fragmentation patternsof the TMS derivatives of BaP diols, tetrols, and epoxides aswell as phenols have been reported (23). Also, the El-, NICI-,and PCI-MS fragmentations of the permethyl, pertrifluoroac-etyl, and peracetyl derivatives of the BaPTs, as well as the Elspectra of the permethyl and peracetyl derivatives of CT, BaAT-I, and BaAT-II, were recently published (24). None of thesereports, though, addresses the question of sensitivity.

We presumed that mass spectral analysis of PAH and PAH-derived metabolites would be most sensitive when NICI wasused, because the high electron affinity of the PAH ring systemsshould readily stabilize a negative charge (25). This property ofthe PAH also obviates the need to use electrophoric derivatizingagents, and we thus chose to use trimethylsilylation for conven-

Table 2 Affinities oj PAH alcohols for the SEI I monoclonal antibody, asdetermined by radioimmunoassay

Tracer was 7.8,9-trihydroxy-10-(2-(3H-acetamido]ethylthio)-7,8,9,10-tetrahy-

drobcnzoja ]p>rene.

Compoundan/i-BaPTBaP-9.10-diolyn-BaPTBaP-7.8-rf/o/BaAT-IIBePTCTBaAT

IBaP45DBePDCPPDSDFaT1C»

(pmol)4.748.5613.724.755.290.7151.7352.46.19.7650.783.000No

affinityNoaffinityRelative

bindingaffinity(%)10055351995310.70.70.005

4614

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MACROMOLECULAR ADDUCTS OF CARCINOGENS

ience. The GC oven programs described in "Materials andMethods," in combination with the use of the 30-m capillary

1% phenyl columns, allowed for baseline separation of theTMS derivatives of the PAH alcohols (data not shown).

NICI-MS data for the TMS derivatives of the standard andsynthetic diols and tetrols are recorded in Table 3. The NICImass spectrum of the TMS derivative of trans-anti-BaP tetrol,which is representative, is displayed in Fig. 2. The El massspectrum is also shown to illustrate the degree to which fragmentation is reduced under NICI conditions.

Fragment ions of the TMS derivatives are generated viadissociative electron-capture mechanisms. Under the conditions we utilized, the TMS derivatives of tetrahydrotetrols ofthe PAH examined uniformly lost a neutral fragment of di-TMS ether to yield a radical anión (M - 162)7. The dihydro-diols invariably lost TMS-OH to yield an aromatic TMS phenolradical anión (M —90)'. BePD, primarily in the form of thetrans-dio\, was unstable to derivatization and/or on-column.

The instability of dihydrodiols bearing unsaturation in thedihydro ring to TMS derivatization has previously been noted(23). The GC peak yielding MS data consistent with the di-TMS derivative of BePD constituted approximately 30% of thesignal, with the remainder corresponding to the two muchearlier eluting TMS phenols arising from dehydration of BePD.

When the appropriate ion of the TMS derivative of the PAHalcohols was followed by selected ion monitoring, detectionlimits in the low fmol range were obtained in most cases. Theonly exception was SD, which was detected with far greatersensitivity by EI-MS. As noted earlier, the predominant fragmentation was cleavage of the C-l—C-8 bond, with chargeretention by the fragment possessing the phenyl ring (m/z 179).A detection limit of 60 fmol was obtained.

It must be emphasized that the molar values listed in Table3 for the various alcohols are not to be construed as absolutedetection limits for the compounds; rather, they are the detection limits of our mass spectrometer in the NICI mode. Thevalues are, however, useful in the comparison of the relativeresponse of the negative ions generated by derivatives of different PAH ring systems.

Alcohols Obtained from in Vivo Adducted Hemoglobin. Amethod was developed for the analysis of human hemoglobin-

PAH ester adducts based on previous work in this laboratory(5, 6, 10, 11) and others (7, 12). It is summarized in Fig. 3.There are two key elements in the method. Enzymatic digestionusing Pronase E is most efficient in producing alcohols fromhemoglobin-PAH ester adducts when performed at a pH whichis optimal for the BALl hydrolysis mechanism (i.e., pH 7-8.5).

In our experience, also, a single incubation with the enzyme isnot sufficient for the complete hydrolysis of ani/-BaPDE-de-rived esters. A second incubation with Pronase, after readjustment of the pH of the digest medium (which is typically in thepH range of 5.8 to 6.5 after the first incubation), providescomplete hydrolysis of the ester adducts. Pronase, which hydro-lyzes 70-90% of the peptide bonds in proteins, is preferable tomore selective enzymes such as trypsin. When large peptidesremain in the digest, we have found that tetrol hydrolysisproducts are difficult to isolate. Presumably the difficulty arisesfrom their great hydrophobic affinity for globin and largepeptides (6).

The second element is key to accurate fluorometric analysisand is the use of two sequential concentrations of componentsin the digest medium by immunoaffinity chrornatography. Asingle IAC step produces samples with high background in SFSbecause the columns have some nonspecific affinity for hemeand aromatic amino acids. The second IAC step provides samples with background readings only slightly higher than standard PBS solutions. Fig. 4 depicts the difference in the SFSspectrum after one and two IAC steps.

The method in Fig. 3, with the exception of microbore HPLCand GC-MS analysis, was used on a set of RBC samplesobtained from 30 nonsmoking donors exposed only to ambientPAH. The levels of pyrene chromophores in these samples weremeasured by SFS and a histogram describing the results isshown in Fig. 5. The apparent adduction levels are in the 1-15pmol/g hemoglobin range, which is in agreement with a previous report (7). The apparent bimodality does not relate to anyobvious differences in environmental source such as occupation.

The purpose of this study was to assess the relative range anddistribution of adducts in a human population large enough toyield statistically meaningful results. The measured values cannot be interpreted as actual levels of BaP tetrols because othercomponents that bind to the IAC column contribute to thefluorescence (see below). In fact, this is a serious shortcomingof SFS; it is not an analytical technique with sufficiently highresolving power to determine single compounds in complexmixtures. Nevertheless, we feel that the results can be used forvalid interindividual comparisons. Three pooled blood samplesin addition to the one described in Fig. 4 have been analyzedby microbore HPLC and all have yielded essentially the sameresult, namely that about 50% of the fluorescence intensitymeasured by SFS can be attributed to BaP tetrols. While theremay be exceptional individuals, in general it appears that thedistribution of adducts which are fluorescent by AX = 34 nmSFS spectroscopy is relatively uniform. Moreover, the recoveryof the method used is nearly quantitative (>95%) and highly

Table 3 I\'IC/-mass spectral data for PAH alcohol TMS derivatives

CompoundMBePD

430(17)SD*282 (M*.1)BaP45DCPPDCT

588(2)BaAT-1BaAT-IIBePTsyn-BaPTa/m'-BaPTM

-90°340(10)340(100)314(100)m/2(%)M

-162*268

(32)242

(7)422(100)422(100)422(100)446(100)446(100)446(100)Major

ions267(100),

358(4)179(100),147(23)325(11),

309(20).239(32)312(3)414(4)334

(5)334(2)284(12)356

(8)356(5)Détection

limit''1

pmol60fmol50

fmol1fmol10

fmol100fmol1

fmol25fmol1fmol1

fmol°Loss of (CHj)jSi—OH.

*Lossof[(CH,)3Si)jO.f GC-MS analytical limit by selected ion monitoring of the base peak. Signal to noise ratio, > 10.* Determined by 70 e V electron ioni/ation MS.

4615

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

TMSOOTÕ-608

TMS = (CHjJjSi-

MACROMOLECULAR ADDUCTS OF CARCINOGENS

446 40

[M - (TMS-O-TMS)I -

30 i

73

CH3TMS-0-SÌ*

147 CHj

CH3TMS-O-CH = CH-0-Si

191[M - (TMS-O-CH=CH-0-TMS)l <

404

M*

608

100 300 400'soo

600

Fig. 2. Mass spectra of the trimethylsilyl derivative of trans-anti-BaP tetrolusing negative ion chemical ionization (upper) and electron ionization (lower).The base peak at m/: 446 in the N1CI spectrum is produced by loss of TMS—O—TMS and is characteristic of all the tetrols studied.

Whole Blood

RBCs

Lysate

Gel Filtration

Hemoqlobinnoql(

Globin

Enzymatic Digestion

Immunoaffiniry Chromatography

[Microbore HPLC]

Fluorescence Spectrometry

GC-NICI-SIM-MS

Fig. 3. Method for quantitative and qualitative analysis of PAH-hemoglobinester adducts. SIM, selected ion monitoring.

reproducible, at least for BaP tetrols.Asymmetry was a typical feature of the synchronous fluores

cence spectra of many individual samples. Pure compoundsexhibit symmetrical SFS spectra, so asymmetry is characteristicof mixtures of two or more chromophores or two or more ofthe same chromophores situated in very different microenviron-ments. We thus further analyzed pooled smokers' globin, after

digestion and two IAC concentrations, by C-18 microboreHPLC followed by SX = 34 nm SFS of collected Chromatographie fractions. The reconstructed chromatogram is shownin Fig. 6.

Tetrols derived from hydrolysis of anr/-BaPDE esters wereclearly in evidence, while tetrols arising from syn-BaPDE esterswere not detected. The predominant if not exclusive formationof ester adducts by the aw//-stereoisomer of BaPDE has alsobeen observed in mice dosed with BaP (6). The tentative identification based on HPLC-SFS of the tetrols was confirmed by

20o

10 •

320 340 360

Excitation Wavelength (nm)

Fig. 4. Effect of one repetition of the immunoaffinity chromatography purification step on the AA= 34 nm synchronous fluorescence spectrum. A 100-mgglobin sample from a single individual was used to obtain the material from whichspectra were recorded.

8 r

O

Concentration (pmoles/g Globin)Fig. 5. Distribution of pyrene-like fluorescence in the hemoglobin of 30 adult

blood donors, determined by SFS analysis of adducts isolated by inumimi.ill mil \chromatography (2 times) without further purification. Intensity of the fluorescence relative to that of BaP tetrols was used to calculate concentration, which isthus a relative, if not strictly accurate, quantity.

the subsequent observation of molecular ions in the GC-PCI-MS spectra of the isolated material. Nearly 50% of the SFSsignal in the immunorecognized fraction is due to materialother than the anf/'-BaPDE-derived tetrols. The remainder of

the chromatographed material detectable by AX = 34 nm SFSresides largely in fractions 19 and 20, which correspond, inretention time, to any of the following dihydrodiols: trans-7,8-and /rans-9,10-BaPD, m- and trans-9,\0-BePD, and as- andtrans-CPPD. We were unable to further characterize the material in fractions 19 and 20.

Since we had noted that the 8E11 monoclonal antibody usedin IAC could bind PAH alcohols, especially the tetrahydrote-trols, other than those derived from BaP and that PAH epoxidesand diol epoxides appeared to form esters with hemoglobin in

4616

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MACROMOLECULAR ADDUCTS OF CARCINOGENS

Peak Wavelength

5 10 15 20 25 30 35

Fraction Number

Fig. 6. Reconstructed HPLC chromatogram of the immunoreactive materialderived from pooled RBC from six anonymous donors. Each 2-min fraction wasanalyzed by AX= 34 nm SFS. Fractions 9 and 11 correspond to trans-ami- andc/j-anr/-BaPT, respectively. The tetrols from s>'H-BaPDE. which were not notedin this sample, elute in fractions 10 and 13-14.

vitro, in vivo samples taken through the method in Fig. 3 havebeen further examined by GC-NICI-MS methods designed todetect the PAH alcohols we had studied in the in vitro experiments. The details of one such study have been reported elsewhere (26). In brief, analysis of pooled blood samples fromcigarette smokers revealed the presence of both the cis- andfraws-tetrols of a/m-BaPDE, as well as r-l,f-2,f-3,c-4-tetrahy-droxy-l,2,3,4-tetrahydrochrysene (i.e., the fraws-tetrol fromanti-CDE). No other identifiable PAH alcohols were noted inthe chromatograms of these samples.

DISCUSSION

The results obtained from this study indicate that alkylationof carboxylic acids to form esters on hemoglobin is a generalreaction for epoxide and diol epoxide metabolites of PAH. Wehave not attempted to determine the balance between this andother reaction pathways; it is probable that the balance iscompound dependent. Thus, no implication is intended thatalkylation of carboxylate side chains is the major reactionproduct, although in the case of onf/'-BaPDE it is.

The principal reason for investigating the formation of esteradducts is that they are ideally suited for quantitative analysis.The hydrolysis products are polycyclic aromatic alcohols whichcontain no readily ionizable functional groups. They are thussuitable substrates for analysis by GC-MS, which is a techniquewith sufficient sensitivity and resolving power to provide a highdegree of confidence in the identity of each analyte. In addition,isotope dilution MS can be used to achieve quantitative accuracy. The application of this approach for quantitation of BaPtetrols released from macromolecules adducted in humans iscurrently under development in this laboratory.

Evidence was obtained in this and another (26) study forhuman exposure to and metabolic activation of two carcinogenic PAH, benzo[a]pyrene and chrysene. Benzo[a]pyrene-he-moglobin adducts in human blood have been previously beenreported by others (7). Our results confirm this finding andprovide a broader data base, indicating that adduct levels of 1to 15 pmol/g hemoglobin are typical, although these valuesmay be overestimated (see Fig. 6). Immunoaffinity chromatog-raphy was used in the work-up of human hemoglobin with lowadduct levels, representative of environmental exposures. Twopurifications on the Sepharose-bound monoclonal antibodywere required to obtain samples which were amenable to spec-

troscopic analysis. The less selective procedure of solvent extraction, which would isolate all of the PAH alcohols ratherthan a select few with cross-reactivity to the 8E11 monoclonalantibody, resulted in samples far too complex for analysis byGC-MS. Thus, we were not able to make a comprehensivesurvey of the background of ester adducts in human blood.Additional antibodies with relatively broad specificity would beextremely helpful in this regard. We are continuing to evaluatewhether other tetrols, which do bind to 8E11, can be detectedby GC-MS in digests of human globin.

ACKNOWLEDGMENTS

The authors wish to thank Annelise Skor for technical assistance.

REFERENCES

1. Dipple, A., Moschel, R. C, and Bigger, C. A. H. Polynuclear aromaticcarcinogens. In: Chemical Carcinogens, ACS Monograph 182, pp. 41-163.Washington. DC: American Chemical Society, 1984.

2. Harvey, R. G. Polycyclic Hydrocarbons and Carcinogens, ACS SymposiumSeries 283. Washington, DC: American Chemical Society, 1985.

3. Osterman-Golkar, S., Ehrenberg, L.. Segerbäck,D., and Hallstrom, I. Evaluation of genetic risks of alkylating agents. II. Haemoglobin as a dosemonitor. Mutât.Res., 34: 1-10, 1976.

4. Koreeda, M., Moore. P. D., Wislocki, P. G., Levin, W., Conney. A. H., Yagi,H., and Jerina, D. M. Binding of benzo[a]pyrene 7,8-diol-9,10-epoxides toDNA, RNA. and protein of mouse skin occurs with high stereoselectivity.Science (Wash. DC), 199: 778-781. 1978.

5. Skipper, P. L., Naylor. S., Gan. L-S., Day, B. W., Pastorelli, R., andTannenbaum, S. R. Origin of tetrahydrotetrols derived from human hemoglobin adducts of benzo|a]pyrene. Chem. Res. Toxicol., 2: 220-221, 1989.

6. Naylor, S., Gan, L-S., Day. B. W.. Pastorelli, R., Skipper, P. L., andTannenbaum. S. R. Benzo[o]pyrene diol epoxide adducts of mouse andhuman hemoglobin: physicochemical basis for dosimetry. Chem. Res. Toxicol., 3: 111-117. 1990.

7. Weston, A., Rowe, M. L.. Manchester, D. K., Farmer, P. B., Mann, D. L.,and Harris, C. C. Fluorescence and mass spectral evidence for the formationof benzo[a]pyrene anfi'-diol epoxide-DNA and -hemoglobin adducts in hu

mans. Carcinogenesis (Lond.), 10: 251-257, 1989.8. Kaur. S.. Hollander. D.. Haas, R.. and Burlingame, A. L. Characterization

of structural xenobiotic modifications by high sensitivity tandem mass spec-trometry. J. Biol. Chem., 264: 16981-16984, 1989.

9. Jankowiak. R., Day. B. W.. Lu, P., Doxtader, M. M., Skipper, P. L.,Tannenbaum, S. R., and Small, G. J. Fluorescence line-narrowing spectralanalysis of in vivo human hemoglobin-benzo|a]pyrenc adducts: comparisonto synthetic analogues. J. Am. Chem. Soc., in press, 1990.

10. Gan, L-S., Skipper. P. L.. Peng. X., Groopman, J. D., Chen, J-S., Wogan.G. N.. and Tannenbaum. S. R. Serum albumin adducts in the molecularepidemiology of aflatoxin Carcinogenesis: correlation with aflatoxin B, intakeand urinary excretion of aflatoxin M,. Carcinogenesis (Lond.). 9:1323-1325,1988.

11. Pastorelli, R., Naylor, S., Skipper, P. L., and Tannenbaum, S. R. Humanhemoglobin modified in vitro by benzo[a)pyrene: isolation and properties ofthe adduct. Proc. Am. Assoc. Cancer Res., 29:96, 1988.

12. Vahakangas. K., Haugen, A., and Harris, C. C. An applied synchronousfluorescence spectrophotometric assay to study benzo(o]pyrene-diolepoxide-DNA adducts. Carcinogenesis (Lond.), 6: 1109-1116. 1985.

13. Yagi, H., Thakker, D. R., Lehr, R. E., and Jerina, D. M. Benzo-ringdiolepoxides of benzo[e]pyrene and triphenylene. J. Org. Chem., 44: 3439-3442. 1979.

14. Holand. S., and Epstein. R. Preparation of oxiranes from v/c-diols. Synthesis(Stuttgart), 706-708, 1977.

15. Sahali, Y., Skipper. P. L.. and Tannenbaum. S. R. New syntheses of cyclo-penta|o/|pyrene 3,4-oxide and 4-pyrenylacetic acid. J. Org. Chem.. 55: 2918-2920. 1990.

16. LaVoie, E. J., Hecht. S. S., Bedenko, V., and Hoffmann, D. Identificationof the mutagenic metabolites of fluoranthene. 2-methylfluoranthene. and 3-methylfluoranthene. Carcinogenesis (Lond.), 3: 841-846, 1982.

17. Gan, L-S., Otteson, M. S., Doxtader. M. M., Skipper. P. L.. Dasari, R. R..and Tannenbaum. S. R. Quantitation of carcinogen bound protein adductsby fluorescence measurements. Spectrochim. Acta, 45A: 81-86, 1989.

18. Muller, R. Calculation of average antibody affinity in antihapten sera fromdata obtained by competitive radioimmunoassay. J. Immunol. Methods. 34:345-352, 1980.

19. Bryant, M. S., Vineis, P., Skipper, P. L., and Tannenbaum, S. R. Hemoglobinadducts of aromatic amines: associations with smoking status and type oftobacco. Proc. Nati. Acad. Sci. USA, 85: 9788-9791. 1988.

20. Vo-Dinh, T. Multicomponent analysis by synchronous luminescence spec-trometry. Anal. Chem., 50: 396-401. 1978.

21. Santella, R. M., and Stefanidis. M. Immunological methods for the detectionand quantitation of exposure to aromatic hydrocarbons. /n:T. Vo-Dinh (ed.).

4617

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MACROMOLECULAR ADDUCTS OF CARCINOGENS

Chemical Analysis of Polycyclie Aromatic Compounds, pp. 391-410. New ron. Mass Spectrom.. /5: 485-493, 1988.York: John Wiley and Sons, 1989. 25. Buchanan, M. V., and Olerich, G. Differentiation of polycyclic aromatic

22. Lee, B. M., and Santella, R. M. Quantitation of protein adducts as a marker hydrocarbons using electron capture negative chemical ionization. Org. Massof genotoxic exposure: immunologie detection of benzo[a]pyrene globin Spectrom., 19: 486-489. 1984.adducts in mice. Carcinogenesis (Lond.), 9: 1773-1777, 1988. 26. Day. B. W., Skipper. P. L.. Wishnok, J. S.. Coghlin-Strom, J., Hammond,

23. Takahashi, G., Kinoshita, K., Hashimoto, K., and Yauhira, K. Identification S. K., Gann, P., and Tannenbaum, S. R. Identification of an in vivochryseneof benzo[a]pyrene metabolites by gas chromatograph-mass spectrometer. diol epoxide adduci in human hemoglobin. Chem. Res. Toxicol.. in press.Cancer Res., 39: 1814-1818, 1979. 1990.

24. Chess, E. K., Thomas, B. L.. Hendren. D. J., and Bean, R. M. Mass spectral 27. Ingold, C. K., Structure and Mechanism in Organic Chemistry, Ed. 2, pp.characteristics of derivatized metabolites of benzo|a]pyrene. Biomed. Envi- 1137-1142. 1157-1163. Ithaca, NY: Cornell University Press, 1969.

4618

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1990;50:4611-4618. Cancer Res   Billy W. Day, Stephen Naylor, Liang-Shang Gan, et al.   Epoxides and Diol Epoxides via Hemoglobin AdductsMolecular Dosimetry of Polycyclic Aromatic Hydrocarbon

  Updated version

  http://cancerres.aacrjournals.org/content/50/15/4611

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/50/15/4611To request permission to re-use all or part of this article, use this link

on June 24, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from