Rat liver metabolism of benzo[b] - American Chemical Society

Valhalla, New York 10595. Received August 1, 1991. Thioarenes, sulfur-containing polycyclic aromatic hydrocarbons, have been detected in a number...
3 downloads 0 Views 610KB Size
Chem. Res. Toxicol. 1992,5,491-495

491

Rat Liver Metabolism of Benzo[ b]naphtho[2,1-~lJthiophene Sharon E. Murphy,' Shantu Amin, Kristin Coletta, and Dietrich Hoffmann Naylor Dana Institute for Disease Prevention, American Health Foundation, 1 Dana Road, Valhalla, New York 10595 Received August 1, 1991

Thioarenes, sulfur-containingpolycyclic aromatic hydrocarbons, have been detected in a number of environmental sources. The metabolism of one thioarene, benzo[blnaphtho[2,l,dl thiophene ([2,1]BNT),byF344ratliver9000gsupernatant(S-9) wasstudied. [2,1lBNTwhichisstructurally analogous and has similar carcinogenic potency to chrysene, was metabolized to six ethyl acetateextractable metabolites when incubated with S-9from Aroclor 1254-treated F344 rata. Each metabolite was collected from reverse-phase HPLC, and their identities were determined by analysis of MS and NMR data. In order of elution from HPLC they are as follows: (1)trans1,2-dihydroxy-1,2-dihydrobenzo[blnaphtho[2,1-dlthiophene, (2) benzo[b]naphtho[2,1-d]thiophene sulfone, (3) benzo[blnaphtho[2,1-dlthiophene sulfoxide, (4) trans-3,4-dihydroxy3,4-dihydrobenzo[blnaphtho[2,1-dlthiophene, ( 5 ) 8- or 9-hydroxybenzo[blnaphtho[2,1d] thiophene, and (6) 7-hydroxybenzo[b]naphth0[2,1-d]thiophene. In addition, the identities of metabolites 1,2,3, and 4 were confirmed by comparison to standards. The syntheses of the sulfone and sulfoxide of [2,1lBNT are reported here. The syntheses of the dihydrodiols were reported previously. Metabolite 5, a hydroxy[B,l]BNT, was the major metabolite formed by liver S-9from untreated F344 rats. Microsomal preparations from these rata also produced significant amounts of the dihydrodiols, 1 and 4, and the sulfoxide, 3. Microsomes prepared from Wistar rata produced dihydrodiols and the sulfone and sulfoxide of [2,1lBNT. Therefore, [2,1]BNT is metabolized by both ring oxidation and sulfur oxidation in these two strains of rata.

Introduction

5

Thioarenes,sulfur-containingPAH,' have been detected in a number of environmental sources. They are present in the emission from brown coal-fired stoves (I), lubricating oils (2, 3),coal tar derivatives (4), and the exhaust from diesel engines (5). Measurable levels of benzo[blnaphtho[2,1-dlthiophene([2,1lBNT)have been found in rural air in Denmark (6). The carcinogenicity of this compound was suggested by Croisy (7),but the details were not published. Recently, [2,1lBNT was shown to be similar to chrysene in carcinogenic potency, when implanted in the lungs of rats (8). Little informationis available on the metabolism of thioarenes. Most PAH are metabolized by cytochrome P-450 enzymes to phenols, diols, and the arylating diol epoxides. The latter are responsible for the formation of some of the DNA adducts and the carcinogenicity of many of these compounds. Jacob and co-workers measured the metabolism of [2,1]BNT by microsomes from Wistar rats and found only oxidation of the sulfur atom (9). In the study presented here, we investigate the possible metabolism of [2,1lBNT to dihydroxydiols and phenols, analogous to the metabolites of chrysene, the structurally related nonsulfur-containing PAH.

* T o whom correspondenceshould be addressed.

12

4

7

10

[2,1]BNT

1

6

chrysene

Experimental Section

Abbreviations [2,1]BNT,benzo[b]naphtho[2,1-d]thiophene;[2,11BNT-l,2-diol,trane-1,2-dihydroxy-l,2-dihydrobenzo[blnaphtho[2,1-d]thiophene;[2,1lBNT-3,4-diol,trane-3,4-dihydroxy-3,4dhydrobenzo[blnaphtho[2,1-d]thiophene; 2-bromo[2,1]BNT, P-bromobenzo[blnaphtho[2,1-dIthiophene;[SHl-[2,1lBNT,[2-3Hlbenzo[blnaphtho[2,1-dlthiophene;I-,2-,3-,4-,7-,8-,or 9-hydroxy[2,llBNT,1-,2-,3-,4-, 7-,8-,or 9-hydroxybenzo[blnaphtho[2,1-dlthiophene; BbF,benzo[ blfluoranthene;S-9,9OOOgsupernatant;PAH,polycyclic aromatic hydrocarbons.

Melting points are uncorrected. Proton NMR spectra were recorded on a Bruker AM 360-MHz instrument. The proton chemical shifts are reported in ppm downfield from TMS at 0 ppm. High-resolution MS was determined on a VG-7OSE highresolution mass spectrometer interfaced to a VG 11/250 data system at NYU. Other MS were performed with a Hewlett Packard Model 5988A. Elemental analysis was determined at Schwarzkopf Microanalytical Laboratoratory in Woodside, NY. UV spectra were obtained on an HP 89530A photodiode array spectrophotometer (Hewlett Packard, Paramus, NJ). Chemicals. [2,1]BNT was purchased from Aldrich (Milwaukee, WI), Aroclor 1254 was purchased from Accu Standard (New Haven, CT), and glucose 6-phosphate, NADP, and glucose6-phosphate dehydrogenase were purchased from Sigma (St. Louis, MO). Metabolite Synthesis. [2,l]BNT-l,%-diol, [2,1]BNT-3,4diol, and l-hydroxy-, 2-hydroxy-, 3-hydroxy-, and 4-hydroxy[2,1lBNT were synthesized as described (10). [2,llBNT-sulfone was prepared by oxidation of [2,1lBNT with excess 3-chloroperoxybenzoic acid in chloroform, mp 218-220 "C. [BJIBNTsulfoxide was prepared by oxidation of [2,1lBNT with 20% hydrogen peroxide in acetic acid, mp 178-180 OC. The purity of the two compounds was determined by reverse-phase HPLC (system I) to be greater than 98%. NMR spectra were obtained and are identical to those of metabolites 2 and 3 in Table I. High-resolution MS analysis was obtained for both the sulfone, M+ (266.0417), calculated M+ (266.0402), and sulfoxide, M+ (250.0448), calculated M+ (250.0452). To obtain [3H]-[2,1]BNT, 2-bromo[2,l]BNT was synthesized. The methodusedwassimilartothatusedtosynthesize2-hydroxy-

0 1992 American Chemical Society

492 Chem. Res. Toxicol., Vol. 5, No. 4, 1992

Murphy et al.

I

Peak#l

A

Peak 12

Peak #3

\

220

,

nm

400 220

nm

400

Figure 2. UV spectra of [2,1]BNT metabolites collected from HPLC: (1) [2,1lBNT-l,2-diol, (2) [2,l]BNT-sulfone, (3) [2,1]BNT-sulfoxide, (4) [2,1]BNT-3,4-diol, (5) 8- or 9-hydroxy[2,1]BNT, and (6) 7-hydroxy[2,1lBNT. [2,1]BNT (10). 3-Methylbenzo[b]thiophenewas brominated by reaction with N-bromosuccinimide in carbon tetrachloride. This was then reacted with triphenylphosphine in benzene to form (3-thionaphthylmethy1)triphenylphosphoniumbromide. This compound (4.9 g, 0.01 mol), p-bromobenzaldehyde (1.79 g, 0.01 mol), and sodium methoxide (1.1g, 0.02 mol) in methanol (150 mL) were stirred at ambient temperature for 24 h. After the addition of 150 mL of HzO, the organic layer was extracted into methylene chloride (2 X 150mL). The solvent was removed, and the resulting residue was filtered through silica gel column with elution by hexane. The filtrate was concentrated under reduced pressure to yield a thick oil, 1-(3-thionaphthy1)-2-@-bromophenybethylene, MS mlz (relative intensity) 316 (M+, 58), 314 (58), 234 (100). A solution of this compound (1.0 g, 3 mmol) and 12 (5 mg) in dry benzene (1L) was irradiated with a Pyrex-filtered Havonia 450-W medium-pressure UV lamp while bubbling dry air into the solution. Formation of 2-bromo[2,l]BNT by cyclization was monitored by TLC. After 16 h, the solvent was removed under reduced pressure and the yellow residue was recrystallized fromethanol to yield P-bromo[P,l]BNT (0.56g,60%): mp 152-153 "C; 'HNMR (360 MHz) 7.50-7.55, (m, 2 H, HBand He),7.65 (dd, 1 H, H3, J1,3 = 1.9 Hz, J 3 , 4 8.74 Hz), 7.84 (d, 1 H, Hs, J5,e = 8.6 Hz), 7.88 (d, 1 H, Hq, 53.4 = 8.5 Hz), 7.94-8.00 (m, 1H, H7 or H d , 8.18 (d, 1H, He, 55.6 = 8.6 Hz), 8.20-8.28 (m,

Benzo[b]naphtho [2,1-dl thiophene Metabolism

1111

lle4

Time (min)

Figure 3. Chromatographic separation of phenols of [2,1]BNT. Ethyl acetate-extractable metabolites of [2,1]BNT formed by rat liver S-9 were coinjected with 1-, 2-, 3-, and 4-hydroxy[2,1]BNT. HPLC system 111,described in the Experimental Section, was used. 1H, Hlo or H,), 8.30 (d, 1H, HI, 51.3 = 1.8 Hz); MS m/z (relative intensity) 314 (M+, loo), 312 (M+, 100). Anal. Calcd for C&9BrS: C, 61.36; H, 2.90; S, 10.24. Found C, 61.50; H, 2.74; S, 10.24. 2-Bromo[2,1]BNT was sent to Amersham (Chicago, IL) for tritium labeling. The method used was catalyzed halogen displacement with tritium gas. The purity of the crude [3H][ 2,llBNT returned to us was analyzed by reverse-phase HPLC. When an aliquot of this solution, 3 X lo7 dpm, was coinjected with pure [2,1]BNT using HPLC system I, 80% of the radioactivity coeluted with [2,1]BNT standard. A shorter HPLC system was designed (system IV) to purify the [3H]-[2,1]BNT. The column eluant was collected at the appropriate retention time. After multiple injections several millicuries of [3H]-[2,1]BNT was collected. The purity of this was determined by HPLC analysis on system 11. Greater than 98% of the radioactivity coeluted with the [2,1]BNT standard. The solvent was removed by evaporation under a stream of nitrogen. The residue was redissolved in hexane, and the UV spectrum was determined. It was identical to the published spectrum. The absorbance at 276 nm was measured, and the concentration of the solution was determined (e276 = 4.18 X lo4 cm-l M-l; 11). The radioactivity present was determined by liquid scintillation counting, and the specific activity was calculated to be 2.9 Ci/mmol. HPLC Systems. Reverse-phase HPLC analysis was performed with UV detection using four different systems. System I was used to collect [2,1]BNT metabolites for structural analysis. The column, a Beckman Ultrasphere ODS 5 pm, 10 mm X 25 cm, was equilibrated in 5050 methanol/HnO. The sample was eluted with a two-step linear gradient from 5050 methanol/H20 to 70: 30 methanol/HzO in 60 min and then to 100% methanol in 30 min. The flow rate was 2 mL/min. The analysis of [3H]-[2,1]BNT metabolites was performed on a Zorbax ODS analytical column (4.6 mm X 25 cm, Dupont, Wilmington, DE; system 11). The sample was eluted with the same gradient as system I, with a flow rate of 1mL/min. Further resolution of the phenols was obtained on a Vydac C18 column (Alltech, Deerfield, IL) with a shallow gradient. The gradient was linear from 7426 to 8020 methanol/H*O in 40 min and was held for 10 min, and then the percent methanol increased to 100 over the next 10 min (system 111).

Chem. Res. Toxicol., Vol. 5, No. 4, 1992 493 To purify [3H]-[2,1]BNT, system IV was used. The column, a Hibar 11,LiChrosorb RP-18,lO pm (4.6 mm X 25 cm, EM) was equilibrated in 70:30 methanol/HzO, and [3Hl-[2,1]BNT was eluted with a linear gradient to 100% methanol in 20 min. The flow rate was 1 mL/min. Preparation of S-9 and Microsomal Fractions. Male F344 rata, 320-350 g, and male Wistar rata, 200-220 g, were obtained from Charles River Breeding Laboratories (Kingston, NY). Six F344 rats were given a single ip injection of Aroclor 1254, 500 mg/kg in corn oil for 5 consecutive days. Twenty-four hours following the last injection the animals were sacrificed by decapitation and livers removed. The S-9 and microsomal fractions were prepared immediately (12). Livers were minced and homogenized in 0.15 M KC1 (3 mL/g of tissue). The homogenate was centrifuged at 9OOOg at 4 "C for 15 min. A portion of the supernatant, the S-9, was frozen at -80 "C until used. A microsomal fraction was prepared from the remainder by centrifugation at 105000gfor 60 min. The pellet was resuspended in 100 mM potassium phosphate, pH 7.4, centrifuged at 105000g for 60 min, again resuspended in the buffer, and frozen at -80 "C until used. The S-9 and microsomal fractions were also prepared from the livers of untreated F344 and Wistar rats. The protein concentration of each preparation was determined using the Bio-Rad protein assay (Richmond, CA). For the identification of metabolites of [2,1]BNT,severallargescale incubations of [2,1lBNT with the S-9 fraction of livers from rats treated with Aroclor were carried out, at 37 "C with shaking. The total volume of each incubation mixture was 80 mL, containing 10 mg of BNT (0.5 mM, dissolved in 2 mL of DMSO), 100 mM potassium phosphate (pH 7.4), 0.11 M KC1,8 mM MgC12,5 mM glucose 6-phosphate, 1 mM NADP, and 9.7 mg/mL S-9. After 30 min, the reaction was stopped by the addition of an equal volume of ice-cold acetone. This mixture was then extracted with 160 mL of ethyl acetate. The aqueous layer was extracted twice with equal volumes of ethyl acetate. All ethyl acetate extracts were combined and evaporated to dryness on a rotary evaporator. The residue was dissolved in 2 mL of methanol and then filtered, and aliquot8 (20-400 pL) were analyzed by HPLC, using system I. Fractions containing UVabsorbing peaks were collected from multiple injections. Each was evaporated to dryness and stored dry until spectral analysis. Quantitation of the metabolites formed by liver S-9 from control and Aroclor-treated rats was performed using PHI- [2,11BNT. L3Hl-[2,11BNT was diluted with 2 mg of unlabeled [2,11BNT to a specific activity of 0.32 pCi/pmol. This was incubated under the conditions described above with a total volume of 16 mL. The metabolites were prepared for HPLC analysis as above and dissolved in a final volume of 1 mL of methanol. HPLC analysis was on system 11. The metabolites formed by liver microsomes from untreated F344 and Wistar rats were compared using [3Hl-[2,11BNT diluted with unlabeled [2,1lBNT to a specific activity of 1.9 pCi/pmol. The total volume of each incubation was 5 mL, containing 50 pM [2,1]BNT, 50 pM potassium phosphate (pH 7.4), 100mM KCl, 2 mM MgCl2,5mM glucose 6-phosphate, 1 mM NADP, 4 units/mL glucose-6phosphate dehydrogenase, and 2.0 mg/mL Fisher microsomes or 2.8 mg/mL Wistar microsomes. The conditions used were essentially those used by Jacob et al. in the study of [2,1lBNT metabolism by Wistar rata (9). After 30 min, reactions were stopped and prepared for HPLC analysis as above.

Results and Discussion The metabolism of [2,1]BNT was studied using the S-9 fraction from the liver homogenate of Aroclor-treated F344 rats. The ethyl acetate-extractable metabolites were analyzed by reverse-phase HPLC. Six major metabolites were detected (Figure 1A). After several large-scale incubations, sufficient quantity of each metabolite was collected from HPLC. The identity of each was determined by MS, NMR, and UV spectra. These spectra are presented in Figure 2 and Table I.

494 Chem. Res. Toxicol., Vol. 5, No. 4, 1992 5

Murphy et al.

4

I -

BNT

/ t HO

@

6

4

1

Figure 4. Proposed pathways for the formation of the major metabolites of [2,1]BNT formed by the S-9 fraction from the liver of of the dihydrodiols is illustrated, but the absolute configurations of the metabolites Aroclor-treated F344 rats. The R,R confimration have not been determined. The identities of metabolites 1-4 were confirmed by comparison of retention time and spectra to synthetic standards. The identification were made as follows. The MS of peak 1 had a molecular ion at mlz 268 and a prominent M - H20 peak, consistent with a dihydrodiol of [2,1]BNT. The UV, MS, NMR, and retention volume of the peak were identical to those of synthetic [2,1lBNT1,2-diol. Metabolite 4 was similarly identified as [2,11BNT-3,Cdiol. The MS of metabolite 2 had a molecular ion at mlz 266, consistent with [2,1]BNTplus 2 oxygen atoms, whichcould be a phenolic sulfoxide or a sulfone. Its UV spectrum was identical to that of synthetic [2,1]BNT-sulfone. This metabolite was collected and its identity confirmed by its proton NMR which was identical to that of the synthetic sulfone. The MS of metabolite 3 had a molecular ion a t mlz 250, consistent with [2,1]BNT plus 1 oxygen atom, suggesting [2,1]BNT-sulfoxide. Phenolic metabolites would also have a molecular ion at mlz 250, but the HPLC retention time of these compounds is much longer. The UV and proton NMR spectrum of peak 3 were identical to those of synthetic [2,1lBNT-sulfoxide. The MS of peak 6 had a molecular ion at mlz 250 and a prominent M-29 peak consistent with a hydroxy-BNT. CHO is a standard fragment, that is lost by phenolic polyaromatic hydrocarbons (23). Peak 6 did not coelute with 1-,2-, 3-, or 4-hydroxy[2,llBNT (Figure 3). The NMR data for this peak are summarized in Table I. The upfield doublet (J = 7.9 Hz) at 6.84 ppm is due to the proton ortho to the phenolic hydroxy group. The chemical shift and large coupling constant indicate that it is either HEor Hs. Therefore, the metabolite must be either 10-or 7-hydroxy[2,1lBNT. The doublet at 8.75 ppm was identified by decoupling as He. Its downfield shift relative to its position in [2,1]BNT indicates that it is adjacent to the hydroxy group. Therefore, this metabolite must be 7-hydroxy[2,1lBNT. All other assignments were confirmed by decoupling experiments. The MS of peak 5 was also consistent with a hydroxy[2,1lBNT; it had a molecular ion of mlz 250 and an M CHO peak. Peak 5 did not coelute with 1-, 2-, 3-, or 4hydroxy[2,1lBNT (Figure 3). The NMR spectrum of peak

Table 11. Formation of Ethyl Acetate-Extractable [2,1]BNT Metabolites by F344 Rat Liver 5-9. rates of production [nmol/(mg30 min)] metaboliteb Aroclor treatedc control [2,1]BNT-1,2-diol(l) 0.51 ndd [2,llBNT-sulfone (2) 0.27 nd 0.64 nd [2,1]BNT-sulfoxide (3) [2,1]BNT-3,4-diol (4) 0.82 nd 9-or 8-hydroxy[2,llBNT (5) 0.66 0.23 7-hydroxy[2,llBNT (6) 2.9 0.0025 total 5.8 0.23 Aliquots of ethyl acetate-extractable material from 30-min incubations with 430 pM [3Hl-[2,1]BNT and 1-2 mg/mL protein were analyzed by HPLC (details as in ExperimentalSection). b Metabolites are listed in the order of their elution from a Zorbax C-18 column, using the HPLC system in Figure 1 (peaks 1-6). 9OOOg supernatant was prepared from the liver of rats administered 500 mg/kg Aroclor by ip injection 5 days prior to sacrifice. dnd, not detected,