Metabolism and excretion of 1-hydroxymethylpyrene, the proximate metabolite of the carcinogen 1-methylpyrene, in rats

Toxicology 366 (2016) 43–52

Contents lists available at ScienceDirect

Toxicology journal homepage: www.elsevier.com/locate/toxicol

Metabolism and excretion of 1-hydroxymethylpyrene, the proximate metabolite of the carcinogen 1-methylpyrene, in rats Carolin Bendadania,1, Lisa Steinhauserb , Klaus Albertb , Hansruedi Glatta,c , Bernhard H. Moniena,c,* a b c

Department of Molecular Toxicology, German Institute of Human Nutrition (DIfE) Potsdam-Rehbrücke, 14558 Nuthetal, Germany Institute of Organic Chemistry, University of Tuebingen, 72076 Tuebingen, Germany Department of Food Safety, Federal Institute for Risk Assessment (BfR), 10589 Berlin, Germany

A R T I C L E I N F O

Article history: Received 6 July 2016 Received in revised form 4 August 2016 Accepted 4 August 2016 Available online 5 August 2016 Keywords: 1-Methylpyrene 1-Hydroxymethylpyrene Rat metabolism Alkylated polycyclic aromatic hydrocarbons

A B S T R A C T

1-Methylpyrene, an alkylated polycyclic aromatic hydrocarbon and environmental carcinogen, is activated by side-chain hydroxylation to 1-hydroxymethylpyrene (1-HMP) and subsequent sulfo conjugation to the DNA-reactive 1-sulfooxymethylpyrene. In addition to the bioactivation, processes of metabolic detoxification and transport greatly influence the genotoxicity of 1-methylpyrene. For a better understanding of 1-HMP detoxification in vivo we studied urinary and fecal metabolites in rats following intraperitoneal doses of 19.3 mg 1-HMP/kg body weight (5 rats) or the same dose containing 200 mCi [14C]1-HMP/kg body weight (2 rats). After 48 h, 48.0% (rat 1) and 29.1% (rat 2) of the radioactivity was recovered as 1-HMP in the feces. Six major metabolites were observed by UV and on-line radioactivity detection in urine samples and feces after HPLC separation. The compounds were characterized by mass spectrometry, 1H NMR and 1H-1H COSY NMR spectroscopy, which allowed assigning tentative molecular structures. Two prominent metabolites, 1-pyrene carboxylic acid (M-6) and the acyl glucuronide of 1pyrene carboxylic acid (M-5) accounted for 17.7% (rat 1) and 25.2% (rat 2) of the overall radioactive dose. Further, we detected the acyl glucuronide of 6-hydroxy-1-pyrene carboxylic acid (M-1) and 8-sulfooxy-1pyrene carboxylic acid (M-3) together with two regioisomers of M-3 (M-2 and M-4) differing in position of the sulfate group at the pyrene ring. In urine samples, the radioactivity of 1-pyrene carboxylic acid and its five derivatives amounted to 32.4% (rat 1) or 45.5% (rat 2) of the total [14C]1-HMP dose. ã 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are formed as a result of incomplete combustion of organic material and can be classified as purely aromatic and alkylated derivatives. 1-Methylpyrene (1MP) is a prototype of an alkylated PAH. It is a common environmental carcinogen that has been found in cigarette and marijuana smoke (Bi et al., 2005; Husgafvel-Pursiainen et al., 1986; Lee et al., 1976; Severson et al., 1976), exhaust of diesel engines (Jensen and Hites, 1983). Levels of 1-MP at 4.1–36.2 mg/100

* Corresponding author at: Federal Institute for Risk Assessment (BfR), MaxDohrn-Strasse 8-10, 10589 Berlin, Germany. E-mail addresses: [email protected] (C. Bendadani), [email protected] (L. Steinhauser), [email protected] (K. Albert), [email protected] (H. Glatt), [email protected] (B.H. Monien). 1 Present address: Federal Office of Consumer Protection and Food Safety (BVL), Department of Food Safety – General Affairs, 10117 Berlin, Germany. http://dx.doi.org/10.1016/j.tox.2016.08.006 0300-483X/ã 2016 Elsevier Ireland Ltd. All rights reserved.

cigarettes were 3–10 times higher compared to those of benzo[a] pyrene in eight different brands of cigarettes (Grimmer 1979; Lee et al., 1976; Severson et al., 1976). 1-MP was also detected at concentrations similar to those of benzo[a]pyrene in restaurants and other locations with tobacco smoke exposure (1.3–8.0 ng/m3) (Husgafvel-Pursiainen et al., 1986), in samples of smoked cheese (0.04–0.3 mg/kg) (Guillen and Sopelana, 2004), in olive oil (1.3– 35 mg/l) (Guillen et al., 2004) and as bio-accumulated pollutant in marine tissues (Pancirov and Brown, 1977). In contrast to pyrene, which is considered to be noncarcinogenic (IARC, 1983; Rice et al., 1988), 1-MP induced formation of hepatic tumors in newborn mice (Rice et al., 1987). This activity may originate from metabolic activation by benzylic hydroxylation and subsequent sulfo conjugation. The hydroxylation at the exocyclic carbon leads to formation of 1-hydroxymethylpyrene (1-HMP). It was observed in rat hepatic homogenates (Rice et al., 1988) and genetically engineered Chinese Hamster V79 cell lines expressing human (h) or rat (r) cytochromes P450 (CYP). From nine CYPs studied hCYP1A1, hCYP1B1 and

44

C. Bendadani et al. / Toxicology 366 (2016) 43–52

rCYP1A1 contributed most to side-chain hydroxylation of 1-MP, whereas hCYP1A2, hCYP2A3, hCYP3A4 and hCYP2E1 as well as rCYP1A2 and rCYP2B1 were less efficient enzymes (Engst et al., 1999; Glatt et al., 1994). In incubations of 1-MP with human hepatic microsomes, 1-HMP and its derivatives formed by further oxidation accounted for 38–64% of all metabolites, indicating that side-chain oxidation is a major pathway of 1-MP metabolism (Engst et al., 1999). The 1-HMP was essentially inactive in standard in vitro genotoxicity tests, whereas it was mutagenic in bacteria in the presence of rat (Glatt et al., 1990; Surh et al., 1990) and human (Czich et al., 1994) liver cytosolic preparations supplemented with 30 -phosphoadenosine-50 -phosphosulfate (PAPS), the cofactor for sulfotransferases (SULTs). Strong mutagenic effects were also observed in various Salmonella typhimurium strains (Glatt et al., 2002) and V79 cells (Teubner et al., 2002) genetically engineered for the expression of mouse, rat and human SULT forms. The findings indicated that 1-HMP is bioactivated to 1-sulfooxymethylpyrene (1-SMP). Chemically synthesized 1-SMP initiated tumor growth at the site of subcutaneous injection in rats (Horn et al., 1996) and in a two-stage mouse model (Surh et al., 1990). Characteristic DNA adducts were found in hepatic DNA of rats that were treated with 1-HMP or 1-SMP by 32P-postlabeling (Ma et al., 2002; Ma et al., 2000; Monnerjahn et al., 1993). Later, ultra performance liquid chromatography-tandem mass spectrometry (UPLC–MS/MS) (Monien et al., 2008) and 1H NMR spectroscopy were used to confirm the molecular structures of the most abundant adducts, N2-((pyrene-1-yl)methyl)- 20 -deoxyguanosine (N2-MP-dG) and N6-((pyrene-1-yl)methyl)-20 -deoxyadenosine (N6-MP-dA). These adducts were also detected in liver, lung and kidney of mice and rats treated with 1-MP corroborating the hypothesis of the bioactivation of 1-MP via 1-HMP to the ultimate genotoxic carcinogen 1-SMP (Bendadani et al., 2014b). When 1-HMP and 1-SMP were administered at equimolar doses to rats 1-SMP formed approximately 15-fold higher levels of DNA adducts in the liver compared to 1-HMP (Surh et al., 1990). This observation suggests that 1-HMP may be detoxified by yet undisclosed metabolic pathways or that continuously formed 1SMP may be more efficiently detoxified compared to a bolus injection of 1-SMP. A previous study showed that incubation of 1HMP in the presence of human and rat hepatic microsomes yielded 1-pyrene carboxylic acid (1-PCA) as a product of side chain oxidation and, also, that 1-HMP may be hydroxylated in part at the pyrene ring by V79 cell lines expressing human or rodent CYPs (Engst et al., 1999). Here, we studied the metabolic pathways of 1HMP in vivo. Male rats were treated with [14C]1-HMP and unlabeled 1-HMP. The metabolites were isolated from urine and feces and their putative molecular structures were deduced from mass spectrometric and 1H NMR spectroscopic data. 2. Materials and methods 2.1. Chemicals [14C]1-formylpyrene (specific activity 56.78 mCi/mmol) was purchased from NEN Radiochemicals (Boston, MA, USA). HionicFluorTM and SolvableTM were from Canberra Packard (Dreieich, Germany) and Ultima-FloTM AF and Count-offTM were from PerkinElmer (Rodgau-Jügesheim, Germany). HPLC-grade methanol and acetonitrile were obtained from Carl Roth GmbH (Karlsruhe, Germany). 1-PCA and 1-HMP and all other reagents (analytical grade or better) were from Sigma-Aldrich (Steinheim, Germany). 2.2. Synthesis of [14C]1-HMP A portion of 40 mg (172 mmol) [14C]1-formylpyrene was dissolved in 4 ml ethanol and reduced by adding 10 mg

(264 mmol) sodium borohydride as described previously (Ashby et al., 1990). The solution was stirred at room temperature for 16 h and dried under reduced pressure. The product was dissolved in 5 ml methanol and 2.5 ml water and purified by solid-phase extraction. A 500 mg Chromabond C18 column (Macherey & Nagel, Düren, Germany) was conditioned with 2 ml methanol, 2 ml water and 2 ml water/methanol (1:1). An aliquot of 500 ml of the product solution was loaded and the column was washed with 2 ml water/methanol (1:1). [14C]1-HMP was eluted with 2 ml water/methanol (1:4) and the purity (>99%) was determined via HPLC-UV. 2.3. Animal treatment with 1-HMP and [14C]1-HMP Seven male Wistar rats (7–8 week old, body weight 200 g) from Charles River Laboratories (Sulzfeld, Germany) were acclimatized for five days and then placed into metabolic cages. Five animals received an intraperitoneal dose of 19.3 mg (83 mmol) 1HMP/kg body weight. For the treatment of two animals, the specific radioactivity was adjusted to 200 mCi in 19.3 mg 1-HMP/kg body weight by the addition of unlabeled 1-HMP. Urine and feces were collected after 24 and 48 h and stored at 80  C. The animals were sacrificed and the dissected tissues were stored at 80  C. The experiment was approved by the Ministry of Environment, Health and Consumer Protection (Landesamt für Umwelt, Gesundheit und Verbraucherschutz) of the state of Brandenburg under the reference number 32-44457 + 15. 2.4. Radioactivity in tissues Tissue samples were mixed with two volumes of water and homogenized by an Ultra-Turrax homogenizer (IKA, Staufen, Germany). Aliquots of 0.1 g of homogenized tissue were mixed with 400 ml scintillation cocktail SolvableTM and incubated for 5 h at 50  C. After addition of 500 ml 2-propanol the incubation was continued for 2 h at 50  C. The mixture was decolorized by dropwise addition of 200 ml of 30% hydrogen peroxide. Finally, the samples were mixed with 4 ml Hionic-FluorTM. The radioactivity was measured by liquid scintillation counting in a b-counter LS 6500 (Beckman Coulter, Krefeld, Germany) with a calibration line of [14C]1-HMP in the dose range from 0.25 to 2.0 mCi. 2.5. Preparation of HPLC samples Aliquots of urinary samples were centrifuged at 5000g for 5 min and 25 ml of the supernatants were directly injected. For the extraction of fecal 1-HMP metabolites, feces samples were mixed with approximately two volumes of water and homogenized by an Ultra-Turrax homogenizer (IKA). The compounds were extracted using a matrix solid-phase dispersion (MSPD) protocol. Aliquots of 0.25 g of homogenized feces were blended with 1 g Isolute C18 resin (Biotage, Uppsala, Sweden). The mixture was transferred to an 8-ml reservoir tube with a bottom frit (Phenomenex, Aschaffenburg, Germany) that was coupled with a 100 mg C18ec solid-phase extraction cartridge (Macherey & Nagel), conditioned with 1 ml methanol and 1 ml of 20 mM sodium citrate buffer (pH 1.8). The coupled columns were washed with 5 ml of 20 mM sodium citrate buffer (pH 1.8) and the metabolites were eluted with 5 ml methanol. The radioactivity of the washing water and the eluate were checked by scintillation counting. The extraction efficiency was >90%. The eluate was dried under reduced pressure and the residue was taken up in 180 ml methanol and centrifuged for 5 min at 12,000g. A 50-ml portion of the supernatant was diluted with 100 ml methanol/water (1:1) and 50 ml were injected for HPLC analysis.

C. Bendadani et al. / Toxicology 366 (2016) 43–52

45

2.6. Analytical HPLC and detection of radioactivity

2.9. 1H NMR spectroscopic characterization of 1-HMP metabolites

Rat urinary and fecal metabolites were analyzed using an HPLC system consisting of two pump modules PU-980, a gradient mixer LG-980-02 and a multiwavelength detector MD-910 (all from Jasco, Groß-Umstadt, Germany) equipped with a Sunfire C18 column (150  3 mm, 5 mm, Waters, Eschborn, Germany). The solvent gradient of eluent A (10 mM ammonium acetate/acetonitrile, 95:5, pH 5.5) and eluent B (acetonitrile) at a flow rate of 1 ml/min was as follows: 0–10 min: 2% B, 10–25 min: increase to 40% B, 25–30 min: 100% B. UV spectra were recorded between 200 and 400 nm. Data acquisition and peak integration were performed with Borwin chromatography software (Jasco). The radioactivity of the eluate was measured on-line with a Radiomatic A-500 flow scintillation analyzer (Canberra Packard) with a 500-ml solid scintillation flow cell. The scintillation cocktail Ultima-FloTM was applied at a flow rate of 6 ml/min. Peak areas of the chromatograms from on-line radioactivity detection (units in cpm) were converted into radioactivity (in dpm or mCi) using a quench curve resulting from injection of specific amounts of [14C]1-HMP. The areas of all compound peaks considered for further investigation were at least three times above the average background in the chromatograms of the radioactivity detection [limit of detection (LOD) >3signal-to-noise ratio (S/N)]. The LOD was determined individually for each of the urine samples injected: 458 cpm (rat 1, urine 0–24 h), 268 cpm (rat 1, urine 24–48 h), 625 cpm (rat 2, urine 0–24 h) and 555 cpm (rat 2, urine 24–48 h). Considering the overall urine volume of each sample and the HPLC injection of 25 ml, the LOD values corresponded to 0.19% (rat 1, urine 0–24 h), 0.16% (rat 1, urine 24–48 h), 0.18% (rat 2, urine 0–24 h) and 0.17% (rat 2, urine 24–48 h) of the original dose of 200 mCi 1-HMP/kg body weight.

The molecular structures of 1-HMP metabolites were studied using one-dimensional 1H nuclear magnetic resonance (NMR) and two-dimensional 1H-1H COSY NMR spectroscopy. Following the isolation from urinary samples the metabolites were desalted by the HPLC procedure described above, except that the eluent buffer A was replaced by water/acetonitrile (95:5). The products were thoroughly dried under reduced pressure and dissolved in 20 ml dimethyl sulfoxide-d6 (Sigma-Aldrich). Due to the limited sample amounts the 1H NMR spectra were measured on a home-build solenoidal microprobe (University of Tuebingen, Germany) with syringe injection into a capillary column (closed for measurement; inserted in a Bruker AMX 600, 14.1 T, proton resonance frequency 600.13 MHz, Bruker BioSpin, Rheinstetten, Germany). The experiment was controlled with an O2 workstation (Silicon Graphics, USA) and XWIN-NMR-Software (Bruker Daltonik, Bremen, Germany). The 1H/13C inverse microcoil NMR probe had an active detection volume of about 5 ml and a total sample volume of about 15 ml. The micro radio frequency coil was directly wrapped around a capillary column and positioned in a container filled with perfluorotributylamine (Sigma-Aldrich) for matching the susceptibility differences between copper and air (Gokay and Albert 2012; Kuhnle et al., 2009). All data sets were recorded and processed with Topspin 2.0 software (Bruker BioSpin). The temperature was set to 298 K and the spectra were referenced to the solvent signal of residual dimethyl sulfoxide at d = 2.50 ppm. 1H NMR spectra were recorded with 2–10 k transients and for the two-dimensional homonuclear correlation spectroscopy decoupling (1H-1H COSY) NMR spectra 128–512 transients were accumulated. Chemical shifts d are reported in parts per million (ppm) downfield from tetramethylsilane, and coupling constants J are reported in Hertz (Hz). The 1 H-1H COSY NMR experiments were conducted to assist in assigning hydrogen resonances. NMR spectra of the reference substances 1-PCA and 1-hydroxypyrene (1-HOP) allowed determining characteristic chemical shifts and coupling constants in the hydrogen spin systems. The 1H NMR spectrum of 1-HOP showed four isolated spin systems (Fig. S1 in the Supplementary information): three two-spin systems of hydrogens (H-2 and H3, H-4 and H-5, H-9 and H-10) and one three-spin system (H-6, H-7 and H-8). The assignment of the signals was achieved using a homonuclear 1H-1H COSY spectrum (Fig. S2 in the Supplementary information), in which strong cross-peaks corresponding to threebond couplings of the vicinal protons were observed. The 1H NMR and 1H-1H COSY spectra of the reference compound 1-PCA are shown in Figs. S3 and S4 of the Supplementary information, respectively.

2.7. Isolation of urinary 1-HMP metabolites by preparative HPLC Urine from animals treated with non-radioactive 1-HMP was centrifuged at 5,000g for 5 min. A volume of 5 ml of the supernatant was fractionated on a PrepLC 150 system coupled to a 996 photodiode array detector and a Fraction Collector III (all from Waters) with a SunFire Prep C18 OBD column (19  150 mm, 5 mm, Waters) using 10 mM ammonium acetate (pH 5.5)/acetonitrile (95:5) as eluent A and acetonitrile as eluent B. The gradient was as follows: 0–5 min: 5% B, 5–25 min: increase from 5 to 90% B, 25–30 min: 100% B. The flow rate was 15 ml/min. The individual metabolites were purified for a second time using similar chromatographic parameters and freeze-dried previous to further analyses. 2.8. Mass spectrometric characterization of 1-HMP metabolites

3. Results The metabolites were analyzed on a Quattro Premier XE tandem quadrupole mass spectrometer (Waters) with an electrospray interface operated in the negative ion mode. An Acquity UPLC system (Waters) was used for the direct injection of the analytes in a constant flow of 10 mM ammonium acetate (pH 7.0)/acetonitrile (95:5) at a flow rate of 0.35 ml/min. The molecular masses of single compounds were determined using a full-scan mode and then fragments of the parent ion were recorded by collision-induced dissociation. The tune parameters were as follows: temperature of the electrospray source 110  C; desolvation temperature 450  C; desolvation gas: nitrogen (850 l/h); cone gas: nitrogen (50 l/h); collision gas for collision-induced dissociation: argon (indicated cell pressure 5  103 mbar). The capillary voltage was set to 0.4 kV. The cone and RF1 lens voltage were 20 V and 0.1 V, respectively. Data acquisition and handling were performed with MassLynx 4.1 software (Waters).

3.1. Quantification of [14C]1-HMP metabolites in urine and feces Individual metabolites of 1-HMP in urine samples and extracts of feces from [14C]1-HMP treated animals were separated by analytical HPLC. The detection at 340 nm allowed distinguishing five major peaks M-2 to M-6 (Fig. 1A). The UV spectra of the signals showed that the pyrene chromophore was retained (data not shown). Peaks corresponding to the retention times of M-2 to M-6 were also observed by on-line radio detection (Fig. 1B). The relative intensities of the signals from both detection methods were similar. A sixth metabolite (M-1 in Fig. 1B) was clearly visible only by radio detection indicating the presence of a minor, relatively polar 1-HMP metabolite. In the course of this report the metabolites are denoted in the order of elution as M-1 to M-6. The peak areas allowed for the quantification of the six metabolites

46

C. Bendadani et al. / Toxicology 366 (2016) 43–52

Fig. 1. Chromatographic traces of UV detection at 340 nm (A) and radioactivity (B) following HPLC separation of a urine sample collected 24 h after treatment of a rat with 19.3 mg 1-HMP/kg body weight containing 200 mCi [14C]1-HMP/kg body weight. Metabolites assigned with M-1 to M-6 were identified tentatively via mass spectrometry and 1H NMR spectroscopy. The mixing process of the eluent flow from the UV detector with the scintillation cocktail Ultima-FloTM AF (PerkinElmer) caused peak broadening and a visible delay between UV and radioactivity detection of about 0.3 min.

by the relative radioactivity (Table 1). A seventh signal (M-7) detected only in extracts of feces showed a greater retention on the column than the other metabolites M-1 to M-6. Retention time and UV spectrum of the compound were identical to 1-HMP. 3.2. Characterization of urinary 1-HMP metabolites by mass spectrometry and 1H NMR spectroscopy The metabolites M-1 to M-6 were isolated from urine of male rats treated with unlabeled 1-HMP by preparative HPLC. To deduce putative chemical structures we recorded mass spectra and 1H

NMR spectra of M-1 to M-6. The individual metabolites were analyzed in the negative ion mode by total ion scans and daughter ion scans using collision-induced dissociation. The results of the analyses are summarized in Fig. 2 together with molecular structures of M-1 to M-6 suggested on the basis of the fragmentations. The 1H NMR spectra of M-1 to M-6 supported the structural postulates and yielded additional information on the substitution pattern if a hydroxyl group was inserted at the pyrene ring. The assignments of hydrogen chemical shifts at the aromatic ring of the metabolites and the reference substances 1-PCA and 1HOP are summarized in Table 2. It is of note that 1-HMP (M-7) accounted for most of the radioactivity in the feces, however, M-7 was not detectable in the urine of [14C]1-HMP-treated rats. The results of the mass spectrometric and 1H NMR spectroscopic characterization of the 1-HMP metabolites are presented in the order of increasing complexity starting with the most conclusive case. The metabolite M-6 co-eluted with commercial 1-PCA. The daughter ion spectrum contained a primary ion at m/ z = 245 and a smaller fragment of m/z = 201 that may arise after decarboxylation (Fig. 2). These patterns were also observed in the mass spectrometric analysis of the reference compound 1-PCA. In accordance, the 1H NMR spectrum of M-6 (Fig. S5 in the Supplementary information) matches that of 1-PCA (Fig. S3 in the Supplementary information). Fig. S6 shows the 1H-1H COSY NMR spectrum of M-6, which allowed assigning the signals consistent to the reference 1-PCA (Fig. S4 in the Supplementary information). The primary mass spectrometric signal of the metabolite M-5 was at m/z = 421, suggesting the presence of the putative acyl glucuronide of 1-PCA (Fig. 2). The collision-induced dissociation at m/z = 421 yielded two fragments at m/z = 175 and m/z = 113, which are considered ‘fingerprint’ analytical proof for the presence of a glucuronic acid conjugate (Giessing and Lund, 2002; Staines et al., 2004; Yang et al., 1999), and one fragment of m/z = 245 matching the main signal of deprotonated 1-PCA. The fragment of m/z = 231 may correspond to 1-formylpyrene. However, the daughter ions with m/z = 151 and m/z = 80, possibly resulting from collisioninduced fragmentation of the pyrene ring, were not identified. The presence of nine ring hydrogen signals in the 1H NMR spectrum (Fig. S7 in the Supplementary information) confirmed that M-5 was not modified at the pyrene scaffold. The chemical shifts of the hydrogens roughly correspond to those of 1-PCA (Table 2). The

Table 1 Peak-related radioactivity (% of the dose) detected after separation by analytical HPLC of urinary and fecal metabolites of two male rats treated with 19.3 mg 1-HMP/kg body weight containing 200 mCi [14C]1-HMP/kg body weighta . Metabolite

retention time (min)b

radioactivity (% of the dose) in rat 1/rat 2 urine

feces

total

M-1 M-2 M-3 M-4 M-5 M-6 M-7 trace signalsd

19.3 19.9 21.1 23.5 24.8 27.6 30.2

1.0/2.2 3.8/5.8 6.8/11.2 2.1/1.1 2.7/2.1 6.9/5.2 n.d.c 7.0/9.6

n.d.c n.d.c n.d.c n.d.c 3.2/7.2 5.9/10.7 48.0/29.1 n.d.c

1.0/2.2 3.8/5.8 6.8/11.2 2.1/1.1 5.9/9.3 12.8/15.9 48.0/29.1 7.0/9.6

30.3/37.2

57.1/47.0

87.4/84.2e

a For each animal, urine and feces collected in the intervals between 0 and 24 h and 24–48 h were analyzed separately. The table contains the combined metabolite levels for both collection periods. b The retention time was determined by the UV detector. The radio detection was delayed by about 0.3 min due to the mixing process of the eluent with the scintillation cocktail. c not detected. d In the on-line radioactivity chromatograms minor peaks below the LOD were observed especially in the analyses of the urine samples. e Residual radioactivity in the metabolic cages is not included in these numbers. It was retained in two washing steps: the cages were washed with 250 ml water/methanol (1:1) and then with 250 ml of a 2% solution of Count-offTM (PerkinElmer) in water. Aliquots of 100 ml of the washing solutions were mixed with 4 ml Hionic-FluorTM (Canberra Packard) and the radioactivity was determined. The residual radioactivity for rat 1 and rat 2 was 1.5% and 3.5% of the overall dose.

C. Bendadani et al. / Toxicology 366 (2016) 43–52

47

Fig. 2. Mass spectra of the 1-HMP metabolites M-1 to M-6 recorded by direct injection are shown on the left. Notable fragmentation reactions occurred during the ionization process and showed the fragility of the molecules. The product ion spectra resulting from collision-induced dissociation supported hypotheses on the putative molecular structures of M-1 to M-6 and their fragmentation patterns, which are depicted on the right.

presence of the glucuronic acid group was supported by three additional 1H NMR signals at 5.76 ppm (1H, d, J = 7.6 Hz), 5.53 ppm (1H, d, J = 4.1 Hz) and 5.15 ppm (1H, m), however, they could not be assigned unambiguously. In summary, the data support the hypothesis that M-5 is the acyl glucuronide of 1-pyrene carboxylic acid (Gld-1-PCA). It is of note that also in fecal extracts peaks were detected at the retention times of 1-PCA (M-6) and Gld-1-PCA (M-5). The mass spectrometric analysis of M-1 also indicated the presence of a glucuronic acid group (Fig. 2). The primary mass spectrometric signal was at m/z = 437 suggesting that the

compound corresponds to M-5 with an additional hydroxyl group, inserted at the aromatic ring system. The daughter spectrum of m/z = 437 showed three major signals at m/z = 217, m/z = 175 and m/z = 113. The latter signals are characteristic for glucuronic acid conjugates (Giessing and Lund, 2002; Staines et al., 2004; Yang et al., 1999). The signal at m/z = 261 corresponds to the expected mass-to-charge ratio of the negatively charged 1-PCA with a single hydroxyl group as a substitution at the aromatic ring (Fig. 2). The fragmentation of the putative hydroxylated 1-PCA mainly led to decarboxylation (m/z = 217). The hypothesis that M-1 may be the glucuronic acid conjugate of the hydroxylated 1-PCA was

48

C. Bendadani et al. / Toxicology 366 (2016) 43–52

Table 2 1 H NMR (600 MHz, dimethyl sulfoxide-d6, chemical shift d values in ppm) spectral data of the ring hydrogens of 1-HMP metabolites M-1 to M-6 detected in rat urine and of the reference substances 1-PCA and 1-HOP. Hydrogen atoms were numbered from H-2 to H-10 (compare Fig. 3)a . position

M-1

M2c

M-3

M4b

M-5

M-6

1-PCA

1-HOP

H-2 H-3 H-4 H-5 H-6 H-7 H-8 H-9 H-10

8.56 8.25 8.07 8.67 – 7.94 8.23 8.34 9.21

– – – – – – – – –

8.57 8.25 8.27 8.13 8.33 8.21 – 8.51 9.19

– – – – – – – – –

8.73 8.45  8.36 8.45  8.36 8.29 8.45  8.36 8.19 8.45  8.36 8.45  8.36 9.19

8.52 8.20–8.30 8.20–8.30 8.21 8.35 8.11 8.35 8.20–8.30 9.27

8.59 8.33–8.36 8.33–8.36 8.26 8.40 8.15 8.40 8.33–8.36 9.23

7.58 8.11 8.32 8.03 8.13 7.97 8.13 7.90 8.01

a The coupling constants of the hydrogen spin systems are summarized in Fig. 3 (M-3) and in the figure legends of the Supplementary information (M-1, M-2, M-5, M-6, 1PCA and 1-HOP). b The purification of M-4 did not provide enough material for the NMR analysis. c The low quality of the 1H NMR spectrum prohibited an unambiguous assignment of the signals.

supported by the 1H NMR spectrum (Fig. S8 in the Supplementary information). This showed only signals of eight aromatic ring hydrogens, which is in agreement to the suspected insertion of a hydroxyl group. Further, the three two-spin systems also observed in the reference compounds 1-PCA and 1-HOP were unchanged indicating that the substitution would be at positions C-6, C-7 or C-8. A hydroxylation at position C-7 was not plausible because two singlets from H-6 and H-8 would result. We tentatively assigned the suspected hydroxyl group to position 6, because the signal of the neighboring H-5 underwent a substantial downfield shift in contrast to the signal of H-9 (Table 2). The putative molecular structure of M-1 was that of the acyl glucuronide of 6-hydroxy-1pyrene carboxylic acid (Gld-6-HO-1-PCA). The mass spectra and daughter ion scans of M-2, M-3 and M-4 were similar, indicating that the metabolites were regioisomers with identical atomic compositions. The primary anion observed at m/z = 341 fragmented into four daughter ions with m/z = 297, 261, 217 and 97. The signal at m/z = 97 corresponds to the fragment of a protonated sulfate ion (HSO4), previously observed in the mass spectrometric analyses of reactive sulfate esters of 5-hydroxymethylfurfural (Monien et al., 2009a) and 1-HMP (Monien et al., 2009b). The sulfate ester may be formed after previous hydroxylation at an aromatic position of 1-PCA. This was supported by the presence of a fragment with m/z = 261 corresponding to the expected mass-to-charge ratio of hydroxylated 1-PCA (Fig. 2). In agreement, the integration of the 1H NMR signals of M-2 (Fig. S9 in the Supplementary information) and M-3 (Fig. 3) showed that only eight hydrogens were present at the pyrene ring. The quality of the

1 H NMR spectra of M-2 and M-4 was not sufficient to draw conclusions on the position of the suggested sulfate group. The 1H NMR and 1H-1H COSY NMR spectra recorded for M-3 allowed unambiguous assignments of the signals (Fig. 3). The spectra of M3 showed that the three-spin system of H-6, H-7 and H-8 was reduced to a two-spin system, indicating the substitution of either H-6 or H-8. The upfield shift of the NMR signals of H-7 and H-9 and the continuance of the H-5 signal at around 8.1 ppm indicated that the suspected sulfate group was located at position 8 of the pyrene ring system. The collected data supported the hypothesis that the metabolite M-3 was 8-sulfooxy-1-pyrene carboxylic acid (8-SO41-PCA).

4. Discussion The metabolic activation of the carcinogen 1-MP in mice and rats involves the benzylic hydroxylation to 1-HMP and the following sulfo conjugation to the DNA reactive 1-SMP (Bendadani et al., 2014b). Recently, the toxification of 1-HMP by two murine Sults, Sult1a1 and Sult1d1, was studied in vivo by UPLC–MS/MS quantification of 1-SMP in plasma and of the DNA adduct N2-MPdG in various tissues of respective knock out mice (Sult1a1 and Sult1d1) treated with a single intraperitoneal dose of 19.3 mg 1HMP/kg body weight (Bendadani et al., 2014a). The knock out of Sult1a1 led to a 97% reduction in 1-SMP plasma concentration 30 min after 1-HMP treatment compared to the wild-type mice. Also the N2-MP-dG levels were decreased due to the deficiency in Sult1a1, for example in liver (9.5-fold), lung (2.2-fold) and kidney

Fig. 3. The 1H NMR spectrum (A) and 1H-1H COSY NMR spectrum (B) of metabolite M-3 (600 MHz, dimethyl sulfoxide-d6). The number of hydrogens, the strong chemical shift of the signal of H-9 and the mass spectral data indicated that M-3 was 1-PCA containing a sulfate group at position C-8 of the pyrene ring. d (ppm) 9.19 (1H, d, J = 9.1 Hz, H-10), 8.57 (1H, d, J = 7.8 Hz, H-2), 8.51 (1H, d, J = 9.4 Hz, H-9), 8.33 (1H, d, J = 8.1 Hz, H-6), 8.25 (1H, d, J = 8.4 Hz, H-3), 8.27 (1H, d, J = 9.1 Hz, H-4), 8.21 (1H, d, J = 8.1 Hz, H-7), 8.13 (1H, d, J = 8.4 Hz, H-5).

C. Bendadani et al. / Toxicology 366 (2016) 43–52

(2.8-fold). The capacity of 1-HMP bioactivation by human SULT1A1/1A2 was studied with a transgenic mouse line expressing human SULT1A1/1A2 (Dobbernack et al., 2011). Thirty minutes after the injection of 1-HMP the plasma levels of 1-SMP were 12fold higher compared to those in wild-type mice and the N2-MP-dG levels exceeded those in wild-type mice in liver, lung and kidney by about 13.4-, 8.3- and 83-fold, respectively (Bendadani et al., 2014a). The extent of DNA alkylation by 1-HMP in a particular tissue in vivo may in part be determined by the SULT expression in that tissue. However, metabolic detoxification of 1-HMP and transport processes modulate the genotoxic effect. A previous study in which 1-HMP was administered to rats together with the uricosuric drug probenecid, an unspecific inhibitor of transmembrane transport proteins, showed that the directed ex- and import of 1-SMP play a primary role in the tissue distribution of DNA adducts (Monien et al., 2009b). The genotoxic effect may also be determined by the capacity of other metabolic pathways, e.g. the oxidation of 1-HMP to 1-formylpyrene and further to 1-PCA. This detoxification has been described for other relevant SULT-activated genotoxicants. For example, 5-hydroxymethylfurfural (Germond et al., 1987; Godfrey et al., 1999) and furfuryl alcohol (Nomeir et al., 1992), two rodent carcinogens abounding in human foodstuffs, were excreted in the urine of rats after a single administration mainly as carboxylic acids or conjugated derivatives. The most prominent urinary metabolite of 5-hydroxymethylfurfural was 5-hydroxymethylfuroic acid constituting about 80% of the eluted radioactivity (Godfrey et al., 1999). Likewise, 83%–88% of a single dose of [14C] furfuryl alcohol was excreted in the urine of rats as furoic acid or as one of its derivative metabolites, furoylglycine or furanacrylic acid (Nomeir et al., 1992). In the current work, we pursued the characterization and quantification of the major 1-HMP metabolites in rats. Following the intraperitoneal treatment of male rats with 200 mCi [14C]1HMP in 19.3 mg 1-HMP/kg body weight most of the radioactivity was excreted within two days in urine and feces (Table 1). The most prominent six urinary metabolites of 1-HMP-treated rats (23.3 and 27.6% of the radioactive dose in rat 1 and rat 2, respectively) were oxidized at the side chain resulting in 1-PCA (M-6) or derivatives thereof. The main metabolites retaining a 1-PCA scaffold were the acyl conjugate of glucuronic acid Gld-1-PCA (M-5; 5.9 and 9.3% of the radioactive dose in rat 1 and rat 2, respectively) and the phenolic sulfate ester with the putative structure of 8-SO4-1-PCA (M-3, 6.8 and 11.2% of the radioactive dose in rat 1 and rat 2, respectively). It is of note that most of the radioactivity was excreted in the feces as unmetabolized 1-HMP. Fig. 4 depicts a proposal for the metabolic pathways of 1-HMP deduced from the current data and including results from previous studies (Engst et al., 1999; Glatt et al., 2008). The radioactivity of all urinary and fecal metabolites added up to 80.4% (rat 1) and 74.6% (rat 2) of the overall [14C]1-HMP dose. Some of the radioactivity (1.5% for rat 1 and 3.5% for rat 2) was retained in the washing solutions of the metabolic cages, which were not further analyzed. In comparison the portion of radioactivity remaining in the body after 24 h was minute (Fig. S10). Less than 0.5% of the [14C]1-HMP dose was retained in the investigated tissues. The highest radioactivity was detected in fat tissues (0.21%) followed by liver (0.15%) and kidneys (0.05%). The discrepancy between the radioactive dose and the overall radioactivity of single metabolites may be explicable in part by the presence of various urinary and fecal metabolites at concentrations below the LOD imposed by the on-line radioactivity detection. For example, the sulfuric acid ester 1-SMP was not detectable. This was not surprising because 1-SMP hydrolyzes rapidly in aqueous solutions (t1/2 = 2.8 min in water at 37  C) (Ma et al., 2003). In part, 1-SMP is metabolized by glutathione conjugation. The resulting

49

conjugate is degrading to the mercapturic acid S-((pyrene-1-yl) methyl)N-acetylcysteine (MPMA, Fig. 4), which was detected previously by LC–MS/MS multiple reaction monitoring in plasma and urine samples of three male rats which received a single intraperitoneal dose of 58.1 mg 1-HMP/kg body weight (Ma et al., 2000). The overall MPMA amounts excreted within 7 days were between 0.005 and 0.059% of the 1-HMP dose. The LOD values of the current study (0.16% to 0.19% of the total [14C]1-HMP dose in different urine samples) indicated that minor amounts of [14C] MPMA were possibly not detected. In contrast to 1-SMP and MPMA, we were surprised that the glycine conjugate of 1-PCA was not detectable. In case of furfuryl alcohol most of the oxidation product furoic acid was present as glycine conjugate, the main urinary metabolite in rats accounting for 73.2% (after treatment with 27.5 mg furfuryl alcohol/kg body weight) and 75.8% (after treatment with 0.275 mg furfuryl alcohol/kg body weight) of the radioactive dose (Nomeir et al., 1992). An analogous glycine conjugate was also observed in rats that were treated with 5hydroxymethylfurfural (Godfrey et al., 1999). However, we were not able to detect N-pyrenoyl-glycine, which was chemically synthesized as reference substance, in urine or feces of 1-HMP treated rats. Another presumed metabolite of 1-HMP results from epoxidation at C-4 and C-5 (K-region) of the pyrene scaffold and the following detoxification by epoxide hydrolases. The reaction yielded trans-4,5-dihydroxy-4,5-dihydropyrene from pyrene in the basidiomycete Crinipellis stipitaria JK375 (Lange et al., 1994) and the analogous product from 1-MP in incubations with S9-mix from rats (Rice et al., 1988). However, we found no evidence for K-region oxidation of the pyrene structure. Previous studies highlighted the importance of mouse Sult1a1 but also of human SULT1A1/1A2 for the bioactivation of 1-HMP to the reactive 1-SMP (Bendadani et al., 2014a). The current data demonstrated the importance of the side-chain oxidation for the detoxification of 1-HMP. The total radioactivity of 1-PCA and its five derivatives accounted for at least 32. 4% (rat 1) or 45.5% (rat 2) of the total [14C]1-HMP dose. This indicated that alcohol dehydrogenases (ADHs) and aldehyde dehydrogenases (ALDHs) may play primary roles in the detoxification of 1-HMP. Previously, the capacity of 1-HMP oxidation by ADHs and ALDHs was studied at the level of single human enzymes. Using ADHs expressed in bacteria it was shown that ADH2 (Kollock et al., 2008b) but also ADH1C, ADH3 and ADH4 (Kollock et al., 2008a) efficiently oxidize 1-HMP to 1-formylpyrene. The transient 1-formylpyrene may be reduced to 1-HMP or further oxidized by ALDH2 and ALDH3A1 to yield 1-PCA (Glatt et al., 2008). Consequently, the carcinogenic effect of 1-HMP is not only affected by the formation of 1-SMP and its transport but also by the efficiency of metabolic detoxification of the precursor 1-HMP. Importantly, the genotoxic effect of 1-HMP may be modulated strongly by compounds which interfere with this detoxification, a competitive pathway of the sulfo conjugation. Ma et al. demonstrated the co-administration of 4-methylpyrazole or disulfiram to 1-HMP treated rats increased the hepatic DNA adduct levels by factors 28 and 3.8, respectively (Ma et al., 2002). In contrast to these rarely used drugs ethanol is a widely consumed intoxicant. A dose of 0.8 g ethanol/kg body weight led to a 15-fold increase in the hepatic DNA adduct level in 1-HMP treated rats in comparison to those that received 1-HMP alone. Ethanol consumed in alcoholic beverages is recognized as a direct human carcinogen by the International Agency for Research on Cancer (IARC, 2010). The data from Ma et al. (2002) indicated that ethanol plays an important role as potent co-carcinogen when it modulates the oxidative detoxification of other carcinogens. Also, oxidative detoxification of food carcinogen furfuryl alcohol was inhibited in a competitive manner in ethanol-treated mice (Sachse et al., 2016). The dominant role of side chain oxidation in 1-HMP metabolism supports the hypothesis that ethanol in alcoholic beverages may

50

C. Bendadani et al. / Toxicology 366 (2016) 43–52

Fig. 4. Evolving scheme of 1-HMP metabolism. The most prominent urinary metabolite of 1-HMP in rats was 1-PCA (M-1). 1-PCA was partially metabolized by UDPglucuronosyl transferases (UGTs) resulting in the formation of Gld-1-PCA (M-5). Ring-hydroxylated metabolites of 1-HMP, 1,n-dihydroxymethylpyrene (1,n-dHMP) and nhydroxy-1-pyrenecarboxylic acid (n-HO-1-PCA) were postulated from the current data. From n-HO-1-PCA three regioisomers of n-SO4-1-PCA were formed (M-2, M-3 and M4), the most prominent one of which was 8-SO4-1-PCA (M-3). One ring-hydroxylated isomer of Gld-n-HO-1-PCA was detected with a hydroxyl group at the C-6 position (Gld6-HO-1-PCA, M-1). The reactive sulfate ester 1-SMP was previously observed in plasma samples of 1-HMP treated rats (Monien et al., 2009b). The loss of the sulfate group yields a transient carbocation. Water addition may return 1-HMP and glutathione conjugation yields the mercapturic acid MPMA. The DNA adducts N2-MP-dG and N6-MP-dA (not shown) were detected in various tissues of 1-MP and 1-HMP treated rats (Monien et al., 2008) and mice (Bendadani et al., 2014b).

exert a strong co-carcinogenic effect via competitive inhibition of alcohol and aldehyde dehydrogenases, which are crucial for the detoxification of 1-HMP. 5. Conclusion 1-HMP is a main metabolite of carcinogenic 1-MP in rats and humans (Rice et al., 1988; Engst et al., 1999). It is bioactivated to the

proximal carcinogen 1-SMP, however, previous studies indicated that sulfo conjugation contributes little to the overall metabolism of 1-HMP (Surh et al., 1990). This study describes the main pathways of metabolic detoxification of 1-HMP in rats. Great parts of a [14C]1-HMP dose, 32.4% (rat 1) and 45.5% (rat 2), were metabolized to PCA and five different derivatives and excreted in the urine. In the feces, 48.0% (rat 1) and 29.1% (rat 2) of the radioactivity was recovered as 1-HMP. This may be released by

C. Bendadani et al. / Toxicology 366 (2016) 43–52

bacterial b-glucuronidases in the gut after formation of the glucuronic acid conjugate of 1-HMP in the liver and subsequent transport into the bile duct (Roberts et al., 2002). However, the putative glucuronic acid conjugate of 1-HMP was not detected. The genotoxic effect of 1-HMP in different species is determined principally by the ratio of sulfo conjugation and oxidative detoxification. We will determine this ratio in tissue homogenates in vitro in order to compare the susceptibility of humans, rats and mice regarding the genotoxicity of 1-HMP. Conflict of interest The authors declare that there is not conflict of interest. Acknowledgements Research described in this article was supported by Philip Morris USA Inc. and by Philip Morris International. The authors thank Brigitte Knuth and Elke Thom for their excellent technical assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.tox.2016.08.006. References Ashby, J., Gallagher, J.E., Kohan, M., Tinwell, H., Kimber, I., Paton, D., Callander, R.D., Chouroulinkov, I., 1990. 1-Chloromethylpyrene: a reference skin sensitizer and genotoxin. Mutat. Res. 243, 281–289. Bendadani, C., Meinl, W., Monien, B., Dobbernack, G., Florian, S., Engst, W., Nolden, T., Himmelbauer, H., Glatt, H.R., 2014a. Determination of sulfotransferase forms involved in the metabolic activation of the genotoxicant 1hydroxymethylpyrene using bacterially expressed enzymes and genetically modified mouse models. Chem. Res. Toxicol. 27, 1060–1069. Bendadani, C., Meinl, W., Monien, B.H., Dobbernack, G., Glatt, H.R., 2014b. The carcinogen 1-methylpyrene forms benzylic DNA adducts in mouse and rat tissues in vivo via a reactive sulphuric acid ester. Arch. Toxicol. 88, 815–821. Bi, X., Sheng, G., Feng, Y., Fu, J., Xie, J., 2005. Gas- and particulate-phase specific tracer and toxic organic compounds in environmental tobacco smoke. Chemosphere 61, 1512–1522. Czich, A., Bartsch, I., Dogra, S., Hornhardt, S., Glatt, H.R., 1994. Stable heterologous expression of hydroxysteroid sulphotransferase in Chinese hamster V79 cells and their use for toxicological investigations. Chem. Biol. Interact. 92, 119–128. Dobbernack, G., Meinl, W., Schade, N., Florian, S., Wend, K., Voigt, I., Himmelbauer, H., Gross, M., Liehr, T., Glatt, H.R., 2011. Altered tissue distribution of 2-amino-1methyl-6-phenylimidazo[4,5-b]pyridine DNA adducts in mice transgenic for human sulfotransferases 1A1 and 1A2. Carcinogenesis 32, 1734–1740. Engst, W., Landsiedel, R., Hermersdörfer, H., Doehmer, J., Glatt, H.R., 1999. Benzylic hydroxylation of 1-methylpyrene and 1-ethylpyrene by human and rat cytochromes P450 individually expressed in V79 Chinese hamster cells. Carcinogenesis 20, 1777–1785. Germond, J.E., Philippossian, G., Richli, U., Bracco, I., Arnaud, M.J., 1987. Rapid and complete urinary elimination of [14C]-5-hydroxymethyl-2-furaldehyde administered orally or intravenously to rats. J. Toxicol. Environ. Health 22, 79– 89. Giessing, A.M., Lund, T., 2002. Identification of 1-hydroxypyrene glucuronide in tissue of marine polychaete Nereis diversicolor by liquid chromatography/ion trap multiple mass spectrometry. Rapid Commun. Mass Spectrom. 16, 1521– 1525. Glatt, H., Henschler, R., Phillips, D.H., Blake, J.W., Steinberg, P., Seidel, A., Oesch, F., 1990. Sulfotransferase-mediated chlorination of 1-hydroxymethylpyrene to a mutagen capable of penetrating indicator cells. Environ. Health Perspect. 88, 43–48. Glatt, H.R., Seidel, A., Harvey, R.G., Coughtrie, M.W., 1994. Activation of benzylic alcohols to mutagens by human hepatic sulphotransferases. Mutagenesis 9, 553–557. Glatt, H.R., Pabel, U., Muckel, E., Meinl, W., 2002. Activation of polycyclic aromatic compounds by cDNA-expressed phase I and phase II enzymes. Polycyclic Aromat. Compd. 22, 955–967. Glatt, H.R., Rost, K., Frank, H., Seidel, A., Kollock, R., 2008. Detoxification of promutagenic aldehydes derived from methylpyrenes by human aldehyde dehydrogenases ALDH2 and ALDH3A1. Arch. Biochem. Biophys. 477, 196–205.

51

Godfrey, V.B., Chen, L.J., Griffin, R.J., Lebetkin, E.H., Burka, L.T., 1999. Distribution and metabolism of (5-hydroxymethyl)furfural in male F344 rats and B6C3F1 mice after oral administration. J. Toxicol. Environ. Health A 57, 199–210. Gokay, O., Albert, K., 2012. From single to multiple microcoil flow probe NMR and related capillary techniques: a review. Anal. Bioanal. Chem. 402, 647–669. Grimmer, G., 1979. Prozesse, bei denen PAH entstehen, Luftqualitätskriterien für ausgewählte polyzyklische aromatische Kohlenwasserstoffe. Erich SchmidtVerlag, Berlin, Germany, pp. 54–76. Guillen, M.D., Sopelana, P., 2004. Occurrence of polycyclic aromatic hydrocarbons in smoked cheese. J. Dairy Sci. 87, 556–564. Guillen, M.D., Sopelana, P., Palencia, G., 2004. Polycyclic aromatic hydrocarbons and olive pomace oil. J. Agric. Food Chem. 52, 2123–2132. Horn, J., Flesher, J.W., Lehner, A.F., 1996. 1-Sulfooxymethylpyrene is an electrophilic mutagen and ultimate carcinogen of 1-methyl- and 1-hydroxymethylpyrene. Biochem. Biophys. Res. Commun. 228, 105–109. Husgafvel-Pursiainen, K., Sorsa, M., Moller, M., Benestad, C., 1986. Genotoxicity and polynuclear aromatic hydrocarbon analysis of environmental tobacco smoke samples from restaurants. Mutagenesis 1, 287–292. IARC, 1983. Polynuclear aromatic compounds, part 1, chemical, environmental and experimental data. IARC Monogr. Eval. Carcinog. Risk Chem. Man., Lyon, France, pp. 1–477. IARC, 2010. Alcohol consumption and ethyl carbamate. IARC Monogr. Eval. Carcinog. Risks Hum., Lyon, France, pp. 3–1383. Jensen, T.E., Hites, R.A., 1983. Aromatic diesel emissions as a function of engine conditions. Anal. Chem. 55, 594–599. Kollock, R., Frank, H., Seidel, A., Meinl, W., Glatt, H.R., 2008a. Oxidation of alcohols and reduction of aldehydes derived from methyl- and dimethylpyrenes by cDNA-expressed human alcohol dehydrogenases. Toxicology 245, 65–75. Kollock, R., Meinl, W., Schneider, H., Batke, M., Frank, H., Seidel, A., Glatt, H.R., 2008b. Efficient oxidation of promutagenic hydroxymethylpyrenes by cDNA-expressed human alcohol dehydrogenase ADH2 and its inhibition by various agents. Biochem. Pharmacol. 75, 527–537. Kuhnle, M., Holtin, K., Albert, K., 2009. Capillary NMR detection in separation science. J. Sep. Sci. 32, 719–726. Lange, B., Kremer, S., Sterner, O., Anke, H., 1994. Pyrene metabolism in Crinipellis stipitaria identification of trans-4,5-dihydro-4,5-dihydroxypyrene and 1pyrenylsulfate in strain JK364. Appl. Environ. Microbiol. 60, 3602–3607. Lee, M.L., Novotny, M., Bartle, K.D., 1976. Gas chromatography/mass spectrometric and nuclear magnetic resonance spectrometric studies of carcinogenic polynuclear aromatic hydrocarbons in tobacco and marijuana smoke condensates. Anal. Chem. 48, 405–416. Ma, L., Landsiedel, R., Seidel, A., Glatt, H.R., 2000. Detection of mercapturic acids and nucleoside adducts in blood: urine and faeces of rats treated with 1hydroxymethylpyrene. Polycyclic Aromat. Compds. 21, 135–149. Ma, L., Kuhlow, A., Glatt, H.R., 2002. Ethanol enhances the activation of 1hydroxymethylpyrene to DNA adduct-forming species in the rat. Polycyclic Aromat. Compds. 22, 933–946. Ma, L., Kuhlow, A., Glatt, H.R., 2003. Albumin strongly prolongs the lifetime of chemically reactive sulphuric acid esters and affects their biological activities in the rat. Nova Acta Leopold. 329, 265–272. Monien, B.H., Müller, C., Engst, W., Frank, H., Seidel, A., Glatt, H.R., 2008. Time course of hepatic 1-methylpyrene DNA adducts in rats determined by isotope dilution LC-MS/MS and 32P-postlabeling. Chem. Res. Toxicol. 21, 2017–2025. Monien, B.H., Frank, H., Seidel, A., Glatt, H.R., 2009a. Conversion of the common food constituent 5-hydroxymethylfurfural into a mutagenic and carcinogenic sulfuric acid ester in the mouse in vivo. Chem. Res. Toxicol. 22, 1123–1128. Monien, B.H., Müller, C., Bakhiya, N., Donath, C., Frank, H., Seidel, A., Glatt, H.R., 2009b. Probenecid, an inhibitor of transmembrane organic anion transporters, alters tissue distribution of DNA adducts in 1-hydroxymethylpyrene-treated rats. Toxicology 262, 80–85. Monnerjahn, S., Seidel, A., Steinberg, P., Oesch, F., Hinz, M., Stezowsky, J.J., Hewer, A., Phillips, D.H., Glatt, H.R., 1993. Formation of DNA adducts from 1hydroxymethylpyrene in liver cells in vivo and in vitro. In: Phillips, D.H., Castegnaro, M., Bartsch, H. (Eds.), Postlabelling Methods for Detection of DNA Adducts. IARC, Lyon, France, pp. 189–193. Nomeir, A.A., Silveira, D.M., McComish, M.F., Chadwick, M., 1992. Comparative metabolism and disposition of furfural and furfuryl alcohol in rats. Drug Metab. Dispos. 20, 198–204. Pancirov, R.J., Brown, R.A., 1977. Polynuclear aromatic hydrocarbons in marine tissues. Environ. Sci. Technol. 11, 989–992. Rice, J.E., Rivenson, A., Braley, J., LaVoie, E.J.R., 1987. Methylated derivatives of pyrene and fluorene: evaluation of genotoxicity in the hepatocyte/DNA repair test and tumorigenic activity in newborn mice. J. Toxicol. Environ. Health 21, 525–532. Rice, J.E., Geddie, N.G., DeFloria, M.C., LaVoie, E.J., 1988. Structural requirements favoring mutagenic activity among methylated pyrenes in S. typhimurium. In: Cooke, M., Dennis, A. (Eds.), Polynuclear Aromatic Hydrocarbons: A Decade of Progress. Battelle Press Columbus, Ohio, USA, pp. 773–785. Roberts, M.S., Magnusson, B.M., Burczynski, F.J., Weiss, M., 2002. Enterohepatic circulation: physiological, pharmacokinetic and clinical implications. Clin. Pharmacokinet. 41, 751–790. Sachse, B., Meinl, W., Glatt, H.R., Monien, B.H., 2016. Ethanol and 4-methylpyrazole increase DNA adduct formation of furfuryl alcohol in FVB/N wild-type mice and in mice expressing human sulfotransferases 1A1/1A2. Carcinogenesis 37, 314– 319.

52

C. Bendadani et al. / Toxicology 366 (2016) 43–52

Severson, R.F., Snook, M.E., Arrendale, R.F., Chortyk, O.T., 1976. Gas chromatographic quantitation of polynuclear aromatic hydrocarbons in tobacco smoke. Anal. Chem. 48, 1866–1872. Staines, A.G., Sindelar, P., Coughtrie, M.W., Burchell, B., 2004. Farnesol is glucuronidated in human liver, kidney and intestine in vitro, and is a novel substrate for UGT2B7 and UGT1A1. Biochem. J. 384, 637–645. Surh, Y.J., Blomquist, J.C., Liem, A., Miller, J.A., 1990. Metabolic activation of 9hydroxymethyl-10-methylanthracene and 1-hydroxymethylpyrene to

electrophilic, mutagenic and tumorigenic sulfuric acid esters by rat hepatic sulfotransferase activity. Carcinogenesis 11, 1451–1460. Teubner, W., Meinl, W., Glatt, H.R., 2002. Stable expression of rat sulfotransferase 1B1 in V79 cells: activation of benzylic alcohols to mutagens. Carcinogenesis 23, 1877–1884. Yang, Y., Griffiths, W.J., Midtvedt, T., Sjovall, J., Rafter, J., Gustafsson, J.A., 1999. Characterization of conjugated metabolites of benzo[a]pyrene in germ-free rat urine by liquid chromatography/electrospray tandem mass spectrometry. Chem. Res. Toxicol. 12, 1182–1189.