Search tips
Search criteria 


Logo of carcinLink to Publisher's site
Carcinogenesis. 2008 May; 29(5): 937–943.
Published online 2008 April 13. doi:  10.1093/carcin/bgn015
PMCID: PMC2902377

Analysis of phenanthrene diol epoxide mercapturic acid detoxification products in human urine: relevance to molecular epidemiology studies of glutathione S-transferase polymorphisms


Many studies have investigated the effects of glutathione S-transferase (GST) polymorphisms on cancer incidence in people exposed to carcinogenic polycyclic aromatic hydrocarbons (PAHs). The basis for this is that the carcinogenic bay region diol epoxide metabolites of several PAH are detoxified by GSTs in in vitro studies. However, there are no reports in the literature on the identification in urine of the mercapturic acid metabolites that would result from this process in humans. We addressed this by developing a method for quantitation in human urine of mercapturic acids which would be formed from angular ring diol epoxides of phenanthrene (Phe), the simplest PAH with a bay region, and a common environmental pollutant. We prepared standard mercapturic acids by reactions of syn- or anti-Phe-1,2-diol-3,4-epoxide and syn- or anti-Phe-3,4-diol-1,2-epoxide with N-acetylcysteine. Analysis of human urine conclusively demonstrated that the only detectable mercapturic acid of this type—N-acetyl-S-(r-4,t-2,3-trihydroxy-1,2,3,4-tetrahydro-c/t-1-phenanthryl)-L-cysteine (anti-PheDE-1-NAC)—was derived from the ‘reverse diol epoxide’, anti-Phe-3,4-diol-1,2-epoxide, and not from the bay region diol epoxides, syn- or anti-Phe-1,2-diol-3,4-epoxide. Levels of anti-PheDE-1-NAC in the urine of 36 smokers were (mean ± SD) 728 ± 859 fmol/ml urine. The results of this study provide the first evidence for a mercapturic acid of a PAH diol epoxide in human urine, but it was not derived from a bay region diol epoxide as molecular epidemiologic studies have presumed, but rather from a reverse diol epoxide, representative of metabolites with little if any carcinogenic activity. These results demonstrate the need for integration of genotyping and phenotyping information in molecular epidemiology studies.


Beginning with the classic studies in the 1930s establishing their role as carcinogenic agents in coal tar and continuing through the discovery of diol epoxide metabolites as their major ultimate carcinogens to the more recent investigations demonstrating the selective reactivity of these metabolites at specific codons in the p53 tumor suppressor gene, polycyclic aromatic hydrocarbons (PAHs) have been firmly established as a major class of environmental carcinogens, which are most likely the cause of cancers of the skin and lung in occupationally exposed individuals, and a major cause of lung cancer in smokers (18). The representative PAH carcinogen, benzo[a]pyrene (BaP), certainly one of the most extensively investigated of all chemical carcinogens, has been evaluated recently as ‘carcinogenic to humans’ by the International Agency for Research on Cancer (9).

There is no doubt that PAH require metabolic activation to exert their carcinogenic effects, and there is massive evidence that DNA adduct formation is the critical step (1012). While various mechanisms of PAH metabolic activation have been proposed, the one for which there is the most convincing evidence in laboratory animals and humans proceeds through a diol epoxide metabolite in which one carbon terminus of the epoxide ring is in the bay region (bay region diol epoxide). One isomer of benzo[a]pyrene-7,8-diol-9,10-epoxide (BPDE), a bay region diol epoxide of BaP, is illustrated in Figure 1 (2,1315). There are many competing pathways of detoxification including the formation of phenols, diols and conjugation of phenols and diols as glucuronides (16). An extensively studied detoxification pathway, and the subject of this paper, involves conjugation of diol epoxides with glutathione, catalyzed by glutathione S-transferases (GSTs). Numerous in vitro studies demonstrate the activities of various GSTs, particularly GSTM1 and GSTP1 enzymes, as effective catalysts of PAH bay region diol epoxide detoxification (1722). This leads to the logical hypothesis that individuals who are null for the GSTM1 gene, or have polymorphisms in the GSTP1 gene leading to lower activity, are at higher risk for cancer upon exposure to PAH. Multiple molecular epidemiology studies have examined this hypothesis, but the overall results are generally mixed or show only modestly elevated cancer risks in exposed carriers of low-activity genotypes (2335). Remarkably, however, there does not appear to be a single report in the literature that has investigated the presence in human urine of a PAH diol epoxide-derived mercapturic acid, such as BPDE-10-N-acetyl-L-cysteine (NAC) (Figure 1), the metabolic end product of detoxification by the GST-catalyzed pathway.

Fig. 1.
Structures of the compounds discussed in this paper.

Therefore, in this study, we analyzed smokers’ urine for mercapturic acids derived from diol epoxide metabolites of phenanthrene (Phe, Figure 1). Phe is the simplest PAH with a bay region and its metabolism by the diol epoxide pathway shares many features with that of BaP. The same enzymes are involved and the metabolites are formed with the same stereoselectivity (3638). However, Phe is generally considered to be non-carcinogenic (39). A bay region diol epoxide of Phe [anti-phenanthrene-1,2-diol-3,4-epoxide (anti-Phe-1,2-D-3,4-E), Figure 1] is 250 times less reactive with DNA than BPDE, the corresponding bay region diol epoxide of BaP, based on our unpublished data, which may explain the low carcinogenicity of Phe. We have focused on the analysis of Phe metabolites because human exposure to Phe greatly exceeds that of BaP. All humans have r,1,t-2,3,c-4-tetrahydroxy-1,2,3,4-tetrahydrophenanthrene (PheT) (Figure 1), the major hydrolysis product of anti-Phe-1,2-D-3,4-E, in their urine, with smokers having higher levels than non-smokers (4042). The amounts of PheT in urine are ~10 000 times as great as those of r,7,t-8,9,c10-tetrahydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene, the corresponding tetraol derived from BPDE, facilitating analysis (42). Our overall goal is to develop a carcinogen metabolite phenotyping approach to examine the relationship of individual differences in Phe metabolism to cancer risk in smokers and others exposed to PAH. Mercapturic acids derived from anti-Phe-1,2-D-3,4-E, such as anti-PheDE-4-NAC (Figure 1), would logically be part of this phenotyping approach, representing detoxification by GST-catalyzed conjugation of a bay region diol epoxide.

We report here the first identification in human urine of a mercapturic acid derived from a PAH diol epoxide. However, this mercapturic acid was not formed from the bay region diol epoxide, anti-Phe-1,2-D-3,4-E, but rather from anti-Phe-3,4-D-1,2-E (Figure 1), the Phe analogue of PAH ‘reverse diol epoxides’ generally found to be non-carcinogenic (2). These results challenge the widely held assumption, inherent in virtually all molecular epidemiologic studies on this topic, that GSTs are involved in the detoxification of carcinogenic PAH diol epoxides in humans.

Materials and methods

Apparatus and chromatography

High-performance liquid chromatography (HPLC) was carried out with a Waters instrument (Millipore, Waters Division, Milford, MA), equipped with a model 991 photodiode array detector, and parameters as follows. System 1: for isolation and purification of Phe metabolites, a 250 × 4.6 mm Luna 5 μm, C18 column (Phenomenex, Torrance, CA) was eluted at 0.7 ml/min with a linear gradient from 30 to 80% methanol in H2O over 50 min, then held at 80% methanol, with ultraviolet (UV) detection at 261 and 241 nm. System 2: for isolation and purification of standard mercapturic acids, a 250 × 4.6 mm Luna 5 μm, C18 column was eluted at 0.7 ml/min with a linear gradient from 5 to 80% methanol in 32 mM NH4OAc over 50 min, then held at 80% methanol, with UV detection at 231 nm. System 3: for purification of diol epoxides, a 250 × 4.6 mm Restek Pinnacle Cyano 5 μm column (Restek, Bellefonte, PA) was eluted with 15% tetrahydrofuran (THF) in hexane at a flow rate of 1 ml/min, with UV detection at 231 nm.

For solid-phase extraction, Oasis MAX and MCX LP extraction cartridges (6 ml, 500 mg) were purchased from Waters. Strata-X polymeric sorbent (33 μm, 30 mg/1 ml) was obtained from Phenomenex. Bond Elute phenylboronic acid cartridges (1 ml, 100 mg) were obtained from Varian, Lake Forest, CA.

Liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) for characterization and quantitative analysis of mercapturic acids was carried out with a Thermo Finnigan TSQ Quantum Discovery MAX or Ultra AM triple quad mass spectrometer (Thermo Electron, San Jose, CA) interfaced with an Agilent 1100 Capillary HPLC System (Agilent Technologies, Palo Alto, CA). A KrudKatcher disposable pre-column filter (Phenomenex) was used in all HPLC work.

The following HPLC Systems were used for LC-ESI-MS/MS experiments. System 4: for the analysis of urinary mercapturic acids, a 1.8 μm, 100 × 0.5 mm Zorbax SB-C18 column (Agilent Technologies, Santa Clara, CA) was eluted at 9 μl/min at 50°C with a linear gradient from 5 to 23.7% methanol in 64 mM NH4OAc over 30 min, then to 100% methanol in 5 min. System 5: a 5 μm, 250 × 0.5 mm Zorbax SB-C18 column was eluted at 10 μl/min at 30°C with a linear gradient from 5 to 30% methanol in 64 mM NH4OAc over 65 min, then to 100% methanol in 5 min. System 6: this was the same as System 5, except that a Synergi 4 μm, 150 × 0.5 mm Polar-RP 80A column (Phenomenex) was used.

The ESI source was set in the negative ion mode as follows: voltage 3.4 kV; current 42 mA and heated ion transfer tube 250°C. The metabolites were analyzed by MS/MS using the selected reaction monitoring (SRM) mode. The Ar collision gas pressure was 1.0 mTorr. The collision energy for PheDE-NAC was 17 eV for the transitions m/z 390 to 261 and 243 and 26 eV for m/z 390 to 225 and 209. The collision energy for [13C6]PheDE-NAC was 17 eV for the transitions m/z 396 to 267 and 249 and 26 eV for m/z 396 to 231 and 215.

Analysis of tetraols, as their trimethylsilyl derivatives, was carried out by gas chromatography–mass spectrometry, essentially as described (42).

Chemicals and enzymes

Racemic anti-Phe-1,2-D-3,4-E, racemic syn-Phe-1,2-D-3,4-E, racemic Phe-1,2-diol and racemic Phe-3,4-diol were kindly provided by Drs D.Jerina and H.Yagi, National Institutes of Health, Bethesda, MD. [13C6]Phe and [D10]Phe were obtained from Cambridge Isotope Laboratories, Andover, MA. [13C6]Phe-(1R, 2R)-diol and [13C6]Phe-(3R, 4R)-diol were obtained from the metabolism of [13C6]Phe as described below. NAC, m-chloroperoxybenzoic acid (MCPBA), NH4OAc (99.999% for LC-MS work) and THF (anhydrous, inhibitor free) were obtained from Sigma–Aldrich, St Louis, MO. Human P450 1A1 and microsomal epoxide hydrolase were purchased from BD Biosciences, San Jose, CA.


N-Acetyl-S-(r-1,t-2,3-trihydroxy-1,2,3,4-tetrahydro-c/t-4-phen-anthryl)-L-cysteine (anti-PheDE-4-NAC, isomers 1 and 2) was prepared by adding racemic anti-Phe-1,2-D-3,4-E (0.46 mg) in 1.8 ml THF to 2 ml of 0.62 M NAC (adjusted to pH 7–7.5). The two-phase solution was stirred for 2 h and then stored at −20°C until analysis. Isomer 1, HPLC retention time (System 5) 51.3 min; UV, λmax 228, 284 nm; 1H nuclear magnetic resonance [D2O] δ 8.10 (d, J = 8.4 Hz, 1H, Phe-H), 7.79 (m, 2H, Phe-H), 7.4–7.5 (m, 3H, Phe-H), 4.65 (m, 2H, H1, 4), 4.3 [m, 3H, H2, 3 and CHNH(COOH)], 3.10 (dd, J = 7.2, 14 Hz, 1H, SCH2a), 2.98 (dd, J = 3.6, 13.8 Hz, 1H, SCH2b), 1.76 (s, 3H, CH3); MS (negative ion ESI) m/z (relative intensity) 390 [M − H] (86); 261 [M − CH2CH(COOH)NHCOCH3] (100); 243 (14); 209 (7). Isomer 2, HPLC retention time (System 5) 56.2 min; UV, λmax 228, 284 nm; 1H nuclear magnetic resonance [D2O] δ 8.00 (d, J = 8.4 Hz, 1H, Phe-H), 7.82 (m, 2H, Phe-H), 7.46–7.56 (m, 3H, Phe-H), 4.7 (m, partially obscured by H2O, H1, 4), 4.41 [m, 3H, H2, 3 and CHNH(COOH)], 3.39 (dd, J = 4.2, 13.8 Hz, 1H, SCH2a), 2.85 (dd, J = 9.6, 13.2 Hz, SCH2b), 1.82 (s, 3H, CH3); MS (negative ion ESI) m/z (relative intensity) 390 [M − H] (71); 261 [M − CH2CH(COOH)NHCOCH3] (100); 243 (19); 209 (8). N-Acetyl-S-(r-1,t-2,c-3-trihydroxy-1,2,3,4-tetrahydro-c/t-4-phenanthryl)-L-cysteine (syn-PheDE-4-NAC, isomers 1 and 2) were similarly prepared by reaction of racemic syn-Phe-1,2-D-3,4-E (1.62 mg) with NAC. Isomer 1, HPLC retention time (System 5) 62.1 min; UV, λmax 228, 288 nm; MS (negative ion ESI) m/z (relative intensity) 390 [M − H] (85); 261 [M − CH2CH(COOH)NHCOCH3] (100); 243 (4); 209 (50); 165 (15); 128 (13). Isomer 2, HPLC retention time (System 5) 66.3 min; UV, λmax 228, 288 nm; MS (negative ion ESI) m/z (relative intensity) 390 [M − H] (69); 261 [M − CH2CH(COOH)NHCOCH3] (100); 243 (4); 209 (54); 165 (15); 128 (8).

N-Acetyl-S-(r-4,t-2,3-trihydroxy-1,2,3,4-tetrahydro-c/t-1-phenanthryl)-L-cysteine (anti-PheDE-1-NAC, isomers 1 and 2) and N-acetyl-S-(r-4,t-3,c-2-trihydroxy-1,2,3,4-tetrahydro-c/t-1-phenanthryl)-L-cysteine (syn-PheDE-1-NAC, isomers 1 and 2) were similarly prepared by the reaction of NAC with anti-Phe-3,4-D-1,2-E and syn-Phe-3,4-D-1,2-E, respectively. The anti-Phe-3,4-D-1,2-E, isomer 1, HPLC retention time (System 4) 25.01 min; UV, λmax 230, 270 nm; MS/MS of m/z 390, m/z (relative intensity) 372 (8); 261 (92); 243 (100); 225 (70); 209 (85); 165 (4); 128 (12). Isomer 2, HPLC retention time (System 4) 26.55 min; UV, λmax 230, 270, 280 nm; MS/MS of m/z 390, m/z (relative intensity) 372 (4); 261 (96); 243 (100); 225 (66); 209 (92); 165 (6); 128 (6). The syn-PheDE-1-NAC, isomer 1, HPLC retention time (System 4) 32.24 min; UV, λmax 230, 270 nm; MS/MS of m/z 390, m/z (relative intensity) 372 (10); 261 (56); 243 (100); 225 (76); 209 (56); 165 (6); 128 (4). Isomer 2, HPLC retention time (System 4) 34.69 min; UV, λmax 230, 270 nm; MS/MS of m/z 390, m/z (relative intensity) 372 (12); 261 (79); 243 (100); 225 (78); 209 (82); 165 (8); 128 (3).

[13C6]Phe-(1R, 2R)-diol and [13C6]Phe-(3R, 4R)-diol from [13C6]Phe metabolism.

[13C6]Phe (125 μg) was incubated overnight at 37°C in a 50 ml polypropylene centrifuge tube containing 15.2 ml of 50 mM phosphate buffer, pH = 7.4, 150 μl dimethyl sulfoxide, 10 mM MgCl2, 0.6 mM NADP+, 0.6 U/ml glucose-6-phosphate-dehydrogenase, 15 mM glucose-6-phosphate, 0.50 ml human P450 1A1 and 0.30 ml microsomal epoxide hydrolase. The metabolites were extracted into ethyl acetate (5 × 25 ml) and the solvent was removed on a Speedvac. The residue was redissolved in 30% CH3OH and the metabolites were separated and collected using HPLC System 1.

PheDE from metabolically formed [13C6]Phe-diols and racemic Phe-diols.

To an insert vial containing a dry Phe-diol (~200 nmol) was added 10 μl of THF containing 800 nmol MCPBA (20 mg/ml) (dried with Na2SO4) and the mixture was allowed to react for 20 h at room temperature. The reaction mixture containing the diol epoxides was dissolved in 80 μl ethyl acetate and extracted with 80 μl of 10% aqueous NaOH to remove excess MCPBA. The ethyl acetate layer was washed with 80 μl H2O and dried with Na2SO4.


The [13C6]-labeled diol epoxides were converted to the corresponding anti-[13C6]PheDE-4-NAC and anti-[13C6]PheDE-1-NAC conjugates by adding the ethyl acetate layer containing the diol epoxides to a solution containing 600 μl THF, 600 μl CH3OH, 60 mg NAC and 36.4 mg KOH.


The study was approved by the University of Minnesota Research Subjects’ Protection Program Institutional Review Board Human Subjects Committee. Smokers were recruited through flyers posted on the University of Minnesota campus and near campus and from advertisements in newspapers and on radio and television. Interested participants called the Tobacco Use Research Center and were told that the study would involve submission of urine samples. All subjects signed a consent form. Samples were stored at −20°C until analysis.

Analysis of anti-PheDE-1-NAC in urine

A 5 ml urine sample was placed in an 8 ml silanized vial (Chrom Tech, Apple Valley, MN) and 334 pmol/ml of anti-[13C6]PheDE-1-NAC, isomer 2, was added as internal standard. HCOOH (167 μl) was added to the urine. The sample was loaded onto an Oasis MCX cartridge that was previously equilibrated using 6 ml of CH3OH and 6 ml of 2% aqueous HCOOH. The cartridge was then washed with the following solutions (% in H2O): 5 ml of 2% HCOOH, 5 ml of 10% CH3OH–2% HCOOH, 5 ml of 20% CH3OH–2% HCOOH, 3 ml of 30% CH3OH–2% HCOOH and 2 ml of 30% CH3OH. The cartridge was then eluted with 5 ml of 80% aqueous CH3OH to collect the analyte. The solvents were removed on a Speedvac using an 8 ml silanized vial.

The sample was redissolved in 1 ml of 1% aqueous NH4HCO3 and loaded on an Oasis MAX cartridge that had been previously equilibrated with 6 ml of CH3OH and 6 ml of 0.2 N KOH. The cartridge was washed twice with 6 ml of 0.2 N KOH, twice with 6 ml of 0.01 N methanolic KOH, twice with 6 ml of 1 M NH4OAc plus 1% CH3COOH, once with 6 ml of CH3OH, once with 6 ml of acetone and once with 6 ml of 2% HCOOH in acetone. The cartridge was then washed with 5 ml of 2% HCOOH in 60% CH3OH to elute anti-PheDE-1-NAC, and the solvents were removed on a Speedvac using an 8 ml silanized vial.

For desalting, the sample was dissolved in 800 μl of 0.1% aqueous NH4HCO3 and loaded onto a Strata-X polymeric sorbent that was previously activated using 1 ml of CH3OH and 1 ml of H2O. The cartridge was washed three times with 1 ml of H2O and the analyte was eluted using 1 ml of 90% CH3OH. The analyte was collected in a 2 ml silanized vial and the solvents were removed on a Speedvac.

The sample was dissolved in 500 μl of H2O and loaded onto a Bond Elute phenylboronic acid cartridge that was previously activated using 1 ml of CH3OH and 1 ml of H2O. The cartridge was washed with 100 μl of H2O and placed under vacuum overnight. The cartridge was then washed twice with 1 ml of acetone (that had been dried with Na2SO4) and anti-PheDE-1-NAC was eluted with 1 ml of 80% CH3OH in H2O. This fraction was collected in a 2 ml silanized vial and concentrated to ~300 μl on the Speedvac. Forty microliter of 2% aqueous NH4OAc was added to the sample which was transferred to an insert vial and concentrated to dryness. The residue was dissolved in 40 μl H2O and 8 μl was analyzed by LC-ESI-MS/MS-SRM System 4.

Control experiments demonstrated that all PheDE-NAC conjugates considered here were recovered from this procedure.


Since we expected to find anti-PheDE-4-NAC in human urine, we began this investigation by reacting synthetic racemic anti-Phe-1,2-D-3,4-E (Figure 1) with NAC to prepare standards. This could give four diastereomeric products, resulting from cis- or trans-addition of NAC at the 4 position of each enantiomer, but only two peaks were observed upon HPLC analysis. The overall structures of the two peaks—isomers 1 and 2—were established by MS, UV and 1H-nuclear magnetic resonance as anti-PheDE-4-NAC (Figure 1).

Further information on the stereochemistry of these products was obtained by preparing enantiomerically enriched (1R, 2S, 3S, 4R)-Phe-1,2-D-3,4-E, a single enantiomer of anti-Phe-1,2-D-3,4-E. This was prepared with a [13C6] label in anticipation of the eventual use of the [13C6]-labeled mercapturic acids as internal standards for quantitative analysis. The preparation of this diol epoxide was accomplished in two steps. First, [13C6]Phe was incubated with human P4501A1, human epoxide hydrolase and cofactors, conditions known to stereoselectively produce Phe-(1R, 2R)-diol (37). A HPLC chromatogram of the products is illustrated in Figure S1 (available at Carcinogenesis Online). [13C6]Phe-(1R, 2R)-diol was isolated and allowed to react with MCPBA to give [13C6](1R, 2S, 3S, 4R)-Phe-1,2-D-3,4-E, which was then reacted with NAC, and the reaction mixture analyzed by LC-ESI-MS. The products of reaction of [13C6](1R, 2S, 3S, 4R)-Phe-1,2-D-3,4-E with NAC coeluted with isomers 1 and 2 from racemic anti-Phe-1,2-D-3,4-E. Therefore, isomers 1 and 2 from the racemic diol epoxide must be the cis- and trans-addition products to the (1R, 2S, 3S, 4R)- and (1S, 2R, 3R, 4S)-diol epoxides, in which each (1R, 2S, 3S, 4R)- and (1S, 2R, 3R, 4S)-derived isomer coelute. Similar pairs of isomers, characterized by MS and UV, were formed from syn-Phe-1,2-D-3,4-E and, as described below, from anti- and syn-Phe-3,4-D-1,2-E.

Isomers 1 and 2 of anti-PheDE-4-NAC were used to establish conditions for analysis of this or related mercapturic acids in human urine. A series of four solid-phase extractions was used to enrich the mercapturic acid-containing fraction. The first was a mixed reverse-phase cation exchange (MCX) cartridge, which could retain anti-PheDE-4-NAC by virtue of its reverse phase properties. The second was a mixed reverse-phase anion exchange (MAX) cartridge, which was able to retain anti-PheDE-4-NAC due to its anion exchange properties. The third was a Strata-X cartridge for desalting. Finally, a phenylboronic acid cartridge preferentially retained materials containing cis-1,2- or cis-1,3-hydroxyl groups, as determined in experiments with standards. This sequential extraction procedure was tested on the urine of an individual who had taken a dose of [D10]Phe. LC-ESI-MS/MS-SRM analysis of the appropriate fraction produced the results illustrated in Figure 2A–C. Clean chromatograms were obtained with peaks corresponding to m/z 390 to 261 (Figure 2A) and m/z 400 to 271 (Figure 2B), consistent with mercapturic acids derived from a Phe diol epoxide and a [D10]Phe diol epoxide, respectively. The standards—anti-PheDE-4-NAC (isomers 1 and 2) and syn-PheDE-4-NAC (isomers 1 and 2)—run under the same conditions are illustrated in Figure 2C. Isomer 2 of anti-PheDE-4-NAC, retention time 56.20 min (shaded in the Figure), eluted with a similar retention time as the peak in Figure 2A—56.25 min. Many other human urine samples analyzed in this way gave traces similar to those in Figure 2A, as discussed below, but further analysis using different HPLC Systems demonstrated unequivocally that the mercapturic acid peak found in all of these samples did not in fact coelute with isomer 2 of anti-PheDE-4-NAC. This is illustrated in Figure S2 (available at Carcinogenesis Online), which used a different HPLC system for analysis of a urine sample from a creosote worker exposed to high levels of Phe. Analysis of this sample gave a single main peak with m/z 390 to 261, as in the other urine samples, but clearly it did not coelute with [13C6]-labeled isomer 2 of anti-PheDE-4-NAC. These results demonstrate conclusively that the mercapturic acid in human urine derived from a Phe diol epoxide is not an isomer of PheDE-4-NAC.

Fig. 2.
Chromatograms obtained upon LC-ESI-MS/MS-SRM analysis (System 5) of solid-phase extracted urine samples from a non-smoker dosed with 50 μg [D10]Phe: (A) SRM at m/z 390 to 261 for a mercapturic acid derived from a Phe diol epoxide; (B) SRM at ...

The UV spectrum of the material in the urine of the creosote worker was similar to those of PheDE-4-NAC and 2-hydroxymethylnaphthalene (λmax 226, 271 nm), indicating that it was a mercapturic acid derived from a Phe diol epoxide in which the diol and epoxide functionalities were in the same ring. Therefore, we hypothesized that the diol epoxide precursor to this product was an isomer of Phe-3,4-D-1,2-E (Figure 1). We prepared [13C6]Phe-(3R, 4R)-diol by human P450 1A1 and epoxide hydrolase-catalyzed metabolism of [13C6]Phe. This diol or the racemic synthetic Phe-3,4-diol was allowed to react with MCPBA, giving a mixture of syn- and anti-Phe-3,4-D-1,2-E (supplementary Figure S3 is available at Carcinogenesis Online), consistent with the literature on MCPBA oxidation of bay region diols (43). Their structures were established by their UV spectra and by hydrolysis of each peak to tetraols. Hydrolysis of syn-Phe-3,4-D-1,2-E gave a tetraol which (as its trimethylsilyl derivative) coeluted on gas chromatography–mass spectrometry with that formed by trans-addition of H2O to syn-Phe-1,2-D-3,4-E while hydrolysis of anti-Phe-3,4-D-1,2-E gave a tetraol enantiomeric with, and which (as its trimethylsilyl derivative) coeluted on gas chromatography–mass spectrometry with PheT. These relationships are summarized in Figure 3. Each diol epoxide was then allowed to react with NAC, producing the HPLC traces illustrated in Figure 4A and B. The MS and UV characteristics of these peaks were very similar to those of the PheDE-4-NAC isomers and, together with their method of preparation, conclusively established their overall structures as isomers of anti-PheDE-1-NAC (Figure 1) and syn-PheDE-1-NAC. The reactions of anti-Phe-3,4-D-1,2-E with NAC were also carried out at different pH. The formation of isomer 1 in Figure 4A was relatively insensitive to pH change while isomer 2 increased markedly at basic pH, indicating that it resulted from trans-addition of NAC to the epoxide ring. Isomer 2 of Figure 4A coeluted with the major peak in human urine in five different HPLC Systems (System 4, System 5 at 30 and 60°C and System 6 at 10 and 60°C), as tested by coinjection. It also had the same MS/MS characteristics as the peak from the creosote worker's urine (Figure 5). These results demonstrate that the mercapturic acid in human urine is an isomer of anti-Phe-DE-1-NAC (probably resulting from trans-addition of glutathione to anti-Phe-3,4-D-1,2-E, followed by normal metabolic processing). The material in human urine is not Phe-DE-4-NAC, which would have been formed from Phe-1,2-D-3,4-E.

Fig. 3.
Structures of tetraols formed by trans-addition of H2O to syn-Phe-3,4-D-1,2-E, syn-Phe-1,2-D-3,4-E, anti-Phe-3,4-D-1,2-E and anti-Phe-1,2-D-3,4-E.
Fig. 4.
Chromatograms obtained upon LC-ESI-MS/MS-SRM analysis (System 4) of the products of reaction of NAC with (A) anti-Phe-3,4-D-1,2-E and (B) syn-Phe-3,4-D-1,2-E. The products are isomers 1 and 2 of (A) anti-PheDE-1-NAC and (B) syn-PheDE-1-NAC.
Fig. 5.
MS/MS (daughters of m/z 390) of (A) anti-Phe-DE-1-NAC (isomer 2) and (B) the peak illustrated in Figure S2A (available at Carcinogenesis Online) from a creosote worker.

We then developed an analytical method for quantitation of anti-Phe-DE-1-NAC in human urine using [13C6]anti-Phe-DE-1-NAC, prepared as described above, as internal standard. Solid-phase extraction was carried out as described above and anti-Phe-DE-1-NAC was quantified by LC-ESI-MS/MS-SRM. A typical chromatogram is illustrated in Figure 6. All chromatograms were similar to this, with a major peak of varying intensity. The results of analyses of urine from 36 smokers are summarized in Table S1 (available at Carcinogenesis Online). Levels of anti-Phe-DE-1-NAC were (mean ± SD) 728 ± 859 fmol/ml urine (range 14–3290 fmol/ml).

Fig. 6.
Typical chromatogram obtained upon LC-ESI-MS/MS-SRM analysis (System 4) of human urine for anti-Phe-DE-1-NAC. (A) analyte, m/z 390 to m/z (261 + 243 + 225 + 209); (B) internal standard [13C6]anti-Phe-DE-1-NAC, m/z 396 to m/z (267 + 249 + 231 + 215).


We report here the first identification and quantitation in human urine of a mercapturic acid resulting from detoxification of a PAH diol epoxide. The origin of this mercapturic acid is undoubtedly GST-catalyzed reaction of glutathione with the diol epoxide followed by normal metabolic processing by γ-glutamyltranspeptidase, cysteinylglycine dipeptidase and cysteine S-conjugate N-acetyltransferase (44). However, the diol epoxide that was detoxified was not the bay region diol epoxide, Phe-1,2-D-3,4-E, but rather the ‘reverse’ diol epoxide Phe-3,4-D-1,2-E. The identification of the urinary mercapturic acid by comparison with standards derived from the reaction of each diol epoxide with NAC leaves no doubt about its overall structure, and there was no evidence at all for production of the bay region diol epoxide-derived mercapturic acids. If Phe is a good model for angular ring metabolism of carcinogenic PAH such as BaP, and the available literature data indicate that it is (37,38), then these results call into question the widely held assumption, based completely on in vitro studies, that bay region diol epoxides of carcinogenic PAH are detoxified by GSTs. This assumption underlies multiple literature studies that have investigated polymorphisms in GST enzymes, particularly GSTM1 and GSTP1, with respect to cancer outcomes in people exposed to PAH, such as smokers. Our results demonstrate that this assumption is untenable.

The formation and biological activities of reverse diol epoxides have been investigated only to a limited extent compared with bay region diol epoxides, and most studies indicate that they are less mutagenic and carcinogenic (2,45). The reverse diol epoxides of BaP and chrysene—BaP-9,10-diol-7,8-epoxide and chrysene-3,4-diol-1,2-epoxide—are generally less mutagenic than the bay region diol epoxides in Salmonella typhimurium strains and V79 cells, with the exception of chrysene-3,4-diol-1,2-epoxide in strain TA98 (2,45). Many studies have examined the carcinogenicity of diol precursors to bay region diol epoxides versus diol precursors to reverse diol epoxides and a consistent finding is that the precursors to bay region diol epoxides are more carcinogenic (2). The reverse diol epoxide of Phe, Phe-3,4-diol-1,2-E, has not been studied, whereas the bay region diol epoxide of Phe, Phe-1,2-D-3,4-E, is a weak mutagen which exhibits virtually no carcinogenic activity (46,47). Collectively, the available data indicate that detoxification of reverse diol epoxides by glutathione conjugation, as observed here, would probably have minimal impact on PAH carcinogenesis in exposed humans.

The flux of human Phe metabolism through the various pathways examined to date in the subjects in this study can be summarized. The most abundant metabolites are PheT (5.33 ± 4.37 pmol/ml urine) and total phenanthrols (HOPhe, 4.30 ± 3.20 pmol/ml urine, excluding 9-HOPhe) (40,42,48,49). Other studies have shown that levels of Phe-1,2-diol and Phe-3,4-diol are similar to those of HOPhe (50). The glutathione detoxification pathways giving rise to anti-PheDE-1-NAC (0.728 ± 0.859 pmol/ml urine) and N-acetyl-S-(9,10-dihydro-9-hydroxy-10-phenanthryl)-L-cysteine (PheONAC, 0.0497 ± 0.0993 pmol/ml urine) appear to be relatively minor pathways compared with the formation of oxidative metabolites, although they would still greatly exceed levels of DNA adduct formation by diol epoxides of Phe and other PAH.

Our finding that PheDE-1-NAC is the only detectable mercapturic acid product of the diol epoxide pathway raises some questions about which diol epoxides are actually formed in humans. All humans have PheT in their urine, and we have also detected r,7,t-8,9,c10-tetrahydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene, the corresponding tetraol formed from BPDE, although in much lower quantities (40,42,51). Our previous assumption was that PheT resulted from hydrolysis of anti-Phe-1,2-D-3,4-E, but as indicated in Figure 3, anti-Phe-3,4-D-1,2-E would be hydrolyzed to an enantiomer of PheT that would not have been distinguished from PheT in our studies to date. This requires further investigation. As mentioned above, there is evidence in the literature that both Phe-1,2-diol and Phe-3,4-diol are formed in humans (50), and the former would be expected to be a better substrate for further oxidation, based on studies with other PAH (37). We hypothesize that anti-Phe-1,2-D-3,4-E is produced more extensively in humans than anti-Phe-3,4-D-1,2-E, but that the latter is a better substrate for GST enzymes because the benzylic position of the epoxide ring is less hindered. This will be investigated in future studies.

Our original goal in this study was to examine the relationship of polymorphisms in GST genes to levels of mercapturic acid metabolites of Phe diol epoxide. This would have required the analysis of ~10 times the number of samples analyzed here. We did not pursue the analysis of more samples because of the likely low biological relevance of anti-PheDE-1-NAC, as discussed above.

In summary, the results of this study demonstrate that a PAH diol epoxide-derived mercapturic acid is present in human urine, but that the diol epoxide precursor is not a bay region diol epoxide of the type generally associated with PAH carcinogenesis. Our results demonstrate graphically the importance of integrating phenotype and genotype information in order to come to logical and firm conclusions regarding gene environment interactions in molecular epidemiologic studies.

Supplementary material

Supplementary Figures S1S3 and Table S1 can be found at


National Cancer Institute (CA-92025); American Cancer Society (RP-00-138 to S.S.H.).

Supplementary Material

[Supplementary Data]


We thank Dorothy Hatsukami and Joni Jensen of the University of Minnesota Transdisciplinary Tobacco Use Research Center, supported by grant DA-13333 from NIH, for providing the urine samples from ongoing studies with smokers.

Conflict of Interest Statement: None declared.



glutathione S-transferase
high-performance liquid chromatography
liquid chromatography-electrospray ionization-tandem mass spectrometry
m-chloroperoxybenzoic acid
polycyclic aromatic hydrocarbon
selected reaction monitoring


1. Phillips DH. Fifty years of benzo[a]pyrene. Nature. 1983;303:468–472. [PubMed]
2. Conney AH. Induction of microsomal enzymes by foreign chemicals and carcinogenesis by polycyclic aromatic hydrocarbons: G.H.A. Clowes Memorial Lecture. Cancer Res. 1982;42:4875–4917. [PubMed]
3. Pfeifer GP, et al. Tobacco smoke carcinogens, DNA damage and p53 mutations in smoking-associated cancers. Oncogene. 2002;21:7435–7451. [PubMed]
4. International Agency for Research on Cancer. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. 1985. Polynuclear aromatic compounds. Part 4. Bitumens, coal-tars and derived products, shale oils and soots. IARC, Lyon, Vol. 35, pp. 83–241. [PubMed]
5. International Agency for Research on Cancer. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. 1984. Polynuclear aromatic compounds. Part 3. Industrial exposures in aluminum production, coal gasification, coke production, and iron and steel founding. IARC, Lyon, Vol. 34, pp. 101–131.
6. International Agency for Research on Cancer. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. 1983. Polynuclear aromatic compounds. Part 1. Chemical, environmental and experimental data. IARC, Lyon, Vol. 32, pp. 419–430. [PubMed]
7. Hecht SS. Tobacco smoke carcinogens and lung cancer. J. Natl Cancer Inst. 1999;91:1194–1210. [PubMed]
8. Hecht SS. Tobacco carcinogens, their biomarkers, and tobacco-induced cancer. Nat. Rev.Cancer. 2003;3:733–744. [PubMed]
9. Straif K, et al. Carcinogenicity of polycyclic aromatic hydrocarbons. Lancet Oncol. 2005;6:931–932. [PubMed]
10. Dipple A, et al. Polynuclear aromatic hydrocarbons. In: Searle CE, editor. Chemical Carcinogens. 2nd edn. 1984. ACS Monograph 182. American Chemical Society, Washington, DC, Vol. 1, pp. 41–163.
11. Szeliga J, et al. DNA adduct formation by polycyclic aromatic hydrocarbon dihydrodiol epoxides. Chem. Res. Toxicol. 1998;11:1–11. [PubMed]
12. Geacintov NE, et al. NMR solution structures of stereoisomeric covalent polycyclic aromatic carcinogen—DNA adducts: principles, patterns, and diversity. Chem. Res. Toxicol. 1997;10:112–146. [PubMed]
13. Casale GP, et al. Detection and quantification of depurinated benzo[a]pyrene-adducted DNA bases in the urine of cigarette smokers and women exposed to household coal smoke. Chem. Res. Toxicol. 2001;14:192–201. [PubMed]
14. Jiang H, et al. Competing roles of aldo-keto reductase 1A1 and cytochrome P4501B1 in benzo[a]pyrene-7,8-diol activation in human bronchoalveolar H358 cells: role of AKRs in P4501B1 induction. Chem. Res. Toxicol. 2006;19:68–78. [PubMed]
15. Boysen G, et al. Analysis of DNA and protein adducts of benzo[a]pyrene in human tissues using structure-specific methods. Mutat. Res. 2003;543:17–30. [PubMed]
16. Cooper CS, et al. The metabolism and activation of benzo[a]pyrene. Prog. Drug Metabol. 1983;7:295–396.
17. Sundberg K, et al. Glutathione conjugation and DNA adduct formation of dibenzo[a,l]pyrene and benzo[a]pyrene diol epoxides in V79 cells stably expressing different human glutathione transferases. Chem. Res. Toxicol. 2002;15:170–179. [PubMed]
18. Sundberg K, et al. Expression of human glutathione transferases in V79 cells and the effect on DNA adduct-formation of diol epoxides derived from polycyclic aromatic hydrocarbons. Chem. Biol. Interact. 2001;133:91–94. 2–28.
19. Sundberg K, et al. Differences in the catalytic efficiencies of allelic variants of glutathione transferase P1-1 towards carcinogenic diol epoxides of polycyclic aromatic hydrocarbons. Carcinogenesis. 1998;19:433–436. [PubMed]
20. Sundberg K, et al. Glutathione conjugation of bay- and fjord-region diol epoxides of polycyclic aromatic hydrocarbons by glutathione transferase M1-1 and P1-1. Chem. Res. Toxicol. 1997;10:1221–1227. [PubMed]
21. Dreij K, et al. Catalytic activities of human alpha class glutathione transferases toward carcinogenic dibenzo[a,l]pyrene diol epoxides. Chem. Res. Toxicol. 2002;15:825–831. [PubMed]
22. Sundberg K, et al. Studies on the differential inhibition of glutathione conjugate formation of (+)-anti-benzo[a]pyrene 7,8-dihydrodiol 9,10-epoxide and 1-chloro-2,4-dinitrobenzene in V79 Chinese hamster cells. Biochem J. 2000;3:693–696. [PubMed]
23. Bartsch H, et al. Genetic polymorphism of CYP genes, alone or in combination, as a risk modifier of tobacco-related cancers. Cancer Epidemiol. Biomarkers Prev. 2000;9:3–28. [PubMed]
24. Smith GB, et al. Human lung microsomal cytochrome P4501A1 (CYP1A1) activities: impact of smoking status and CYP1A1, aryl hydrocarbon receptor, and glutathione S-transferase M1 genetic polymorphisms. Cancer Epidemiol. Biomarkers Prev. 2001;10:839–853. [PubMed]
25. Benhamou S, et al. Meta- and pooled analyses of the effects of glutathione S-transferase M1 polymorphisms and smoking on lung cancer risk. Carcinogenesis. 2002;23:1343–1350. [PubMed]
26. Hung RJ, et al. CYP1A1 and GSTM1 genetic polymorphisms and lung cancer risk in Caucasian non-smokers: a pooled analysis. Carcinogenesis. 2003;24:875–882. [PubMed]
27. Vineis P, et al. CYP1A1, GSTM1 and GSTT1 polymorphisms and lung cancer: a pooled analysis of gene-gene interactions. Biomarkers. 2004;9:298–305. [PubMed]
28. Stucker I, et al. Genetic polymorphisms of glutathione S-transferases as modulators of lung cancer susceptibility. Carcinogenesis. 2002;23:1475–1481. [PubMed]
29. Wang Y, et al. Association between glutathione S-transferase p1 polymorphisms and lung cancer risk in Caucasians: a case-control study. Lung Cancer. 2003;40:25–32. [PubMed]
30. Cote ML, et al. Combinations of glutathione S-transferase genotypes and risk of early-onset lung cancer in Caucasians and African Americans: a population-based study. Carcinogenesis. 2005;26:811–819. [PubMed]
31. Lewis SJ, et al. GSTM1, GSTT1 and GSTP1 polymorphisms and lung cancer risk. Cancer Lett. 2002;180:165–171. [PubMed]
32. Nazar-Stewart V, et al. A population-based study of glutathione S-transferase M1, T1 and P1 genotypes and risk for lung cancer. Lung Cancer. 2003;40:247–258. [PubMed]
33. Wang J, et al. GST genetic polymorphisms and lung adenocarcinoma susceptibility in a Chinese population. Cancer Lett. 2003;201:185–193. [PubMed]
34. Skuladottir H, et al. Polymorphisms in genes involved in xenobiotic metabolism and lung cancer risk under the age of 60 years. A pooled study of lung cancer patients in Denmark and Norway. Lung Cancer. 2005;48:187–199. [PubMed]
35. Schwartz AG, et al. The molecular epidemiology of lung cancer. Carcinogenesis. 2007;28:507–518. [PubMed]
36. Nordquist M, et al. Metabolism of chrysene and phenanthrene to bay-region diol epoxides by rat liver enzymes. Mol. Pharmacol. 1981;19:168–178. [PubMed]
37. Thakker DR, et al. Polycyclic aromatic hydrocarbons: metabolic activation to ultimate carcinogens. In: Anders MW, editor. Bioactivation of Foreign Compounds. New York: Academic Press, Inc.; 1985. pp. 177–242.
38. Shou M, et al. Regio- and stereo-selective metabolism of phenanthrene by twelve cDNA-expressed human, rodent, and rabbit cytochromes P-450. Cancer Lett. 1994;83:305–313. [PubMed]
39. LaVoie EJ, et al. Structure-activity relationships among tricyclic polynuclear aromatic hydrocarbons. In: Yang SK, Silverman BD, editors. Polycyclic Aromatic Hydrocarbon Carcinogenesis: Structure-Activity Relationships. 1988. CRC Press, Boca Raton, FL, Vol. 1, pp. 151–175.
40. Hecht SS, et al. Comparison of polymorphisms in genes involved in polycyclic aromatic hydrocarbon metabolism with urinary phenanthrene metabolite ratios in smokers. Cancer Epidemiol. Biomarkers Prev. 2006;15:1805–1811. [PubMed]
41. Hecht SS, et al. Longitudinal study of urinary phenanthrene metabolite ratios: effect of smoking on the diol epoxide pathway. Cancer Epidemiol. Biomarkers Prev. 2005;14:2969–2974. [PubMed]
42. Hecht SS, et al. r-1,t-2,3,c-4-Tetrahydroxy-1,2,3,4-tetrahydrophenanthrene in human urine: a potential biomarker for assessing polycyclic aromatic hydrocarbon metabolic activation. Cancer Epidemiol. Biomarkers Prev. 2003;12:1501–1508. [PubMed]
43. Lee H, et al. Stereoselective syn epoxidation of the dihydrodiols of dibenz[a,h]anthracene and 7-methylbenz[a]anthracene. Tetrahedron Lett. 1981;22:1657–1660.
44. Eaton DL, et al. Concise review of the glutathione S-transferases and their significance to toxicology. Toxicol. Sci. 1999;49:156–164. [PubMed]
45. Glatt H, et al. Genotoxicity characteristics of reverse diol-epoxides of chrysene. Carcinogenesis. 1993;14:11–19. [PubMed]
46. Buening MK, et al. Tumorigenicity of bay-region epoxides and other derivatives of chrysene and phenanthrene in newborn mice. Cancer Res. 1979;39:5063–5068. [PubMed]
47. Wood AW, et al. Mutagenicity and tumorigenicity of phenanthrene and chrysene epoxides and diol epoxides. Cancer Res. 1979;39:4069–4077. [PubMed]
48. Carmella SG, et al. Analysis of phenanthrols in human urine by gas chromatography-mass spectrometry: potential use in carcinogen metabolite phenotyping. Cancer Epidemiol. Biomarkers Prev. 2004;13:2167–2174. [PubMed]
49. Upadhyaya P, et al. Quantitation of N-acetyl-S-(9,10-dihydro-9-hydroxy-10-phenanthryl)-L-cysteine in human urine: comparison with glutathione-S-transferase genotypes in smokers. Chem. Res. Toxicol. 2006;19:1234–1240. [PMC free article] [PubMed]
50. Jacob J, et al. Profile of urinary phenanthrene metabolites in smokers and non-smokers. Biomarkers. 1999;4:319–327. [PubMed]
51. Simpson CD, et al. Determination of r-7,t-8,9,c-10-tetrahydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene in human urine by gas chromatography-negative ion chemical ionization-mass spectrometry. Chem. Res. Toxicol. 2000;13:271–280. [PubMed]

Articles from Carcinogenesis are provided here courtesy of Oxford University Press