Carcinogenesisvol.il no.7 pp.1195-1201, 1990
Metabolism of (tttne potent carcinogen! 3-metIhykIhiollainitlhuryIeinie microsomes
Shen K.Yang, Fremakala Prasanna, Henri B.Weems, Maryce M.Jacobs1 and Peter P.Fu 2 Department of Pharmacology, F.Edward Hebert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, MD 20814-4799, 'The MITRE Corporation/Metrek Division, McLean, VA 22101 and 2National Center for Toxicological Research, Jefferson, AR 72079, USA 'Present address: American Institute for Cancer Research, 1759 R Street NW, Washington, DC 200O9, USA
Introduction The cyclopenta-ring of cyclopenta-polycyclic aromatic hydrocarbons (cyclopenta-PAHs*) may be a site of oxidative biotransformation to form the ultimate mutagenic and carcinogenic metabolites (1 - 7 ) . Examples of cyclopenta-PAHs studied to date include cyclopenta[c,d]pyrene (1 —3) and cyclopenta-fused isomers of benz[a]anthracene (4,5). Cholanthrylene (CE, an •Abbreviations: cyclopenta-PAHs, cyclopenta-polycyclic aromatic hydrocarbons; CE, cholanthrylene (also known as benz[/]aceanthrylene); 3MCE, 3-methylcholanthrylene; 3MC, 3-methylcholanthrene; 1-OH-3MC, 1-hydroxy-3MC—other hydroxylated products are similarly abbreviated; 3MC trans-] ,2-diol, trans-] ,2dihydroxy-3MC—the cis compound and the equivalent 3MCE compounds are similarly abbreviated; 3MCE 1,2-epoxide, l,2-epoxy-l,2-dihydro-3MCE; BA, benz[a]anthracene; DDQ, 2,3-dichloro-5,6-dicyano-l,4-benzoquinone; TCPO, 3,3,3-trichloropropylene 1,2-oxide; G-6-P, glucose-6-phosphate; THF, tetrahydroftiran; 3-OHMCE, 3-hydroxymethylcholanthrylene; 3MCE;rani-9,10-dihydrodiol, /ra«i-9,10-dihydroxy-9,10-dihydro-3MCE—other dihydrodiols are similarly abbreviated.
Materials and methods Materials 3MCE was prepared by dehydrogenation of 3MC by refluxing a benzene solution of 3MC with a molar excess of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ; Aldrich Chemical Co., Milwaukee, WI). 3MCE cu-l,2-diol was prepared by oxidation of 3MCE with OsO4 (Aldrich Chemical Co., Milwaukee, WI) in pyridine. 3MC-2-one was prepared by dehydrogenation of 3MCE ris-1,2-diol as described (6,11). 3MC trans-1,2-diol was prepared by dehydrogenation of 3MCEc/>l,2-diol with DDQ to form 3MC 1,2-quinone, followed by reduction of NaBH4 (Fisher Scientific Co., Fair Lawn, NJ) in methanol/water (9:1, v/v). Authentic 1-0H-3MC, 2-OH-3MC, 3MC-l-one, 3MC cis-] 1,12-dihydrodiol, 3MC 11,12-quinone, 11-OH-3MC, 3-OH-BA and 4-OH-BA were obtained from the Chemical Repository of the National Cancer Institute. Reaction of 3MC 11,12-quinone with NaBH4 in methanol/water (9:1, v/v) yielded a mixture of 3MC transA 1,12-dihydrodiol (47%) and 3MC cis-11,12-dihydrodiol (53%), which were separated by reversed-phase HPLC (see below). Incubation of 3MCE with rat liver microsomei The metabolites were isolated from a mixture of products formed by incubation of 3MCE with liver microsomes prepared from 3MC-treated rats and an
1195
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
The products formed in the metabolism of 3-methylcholanthrylene (3MCE), either in the presence or in the absence of an epoxide hydrolase inhibitor, 3,3,3-trichloropropylene 1,2-oxide (TCPO), with an NADPH-regenerating system and liver microsomes from 3-methylcholanthrene (3MC)-treated male Sprague-Oawley rats were separated by reversed-phase and normal-phase HPLC. The metabolites were characterized by UV — visible absorption spectral analysis, and by comparing their retention times on reversed-phase and normal-phase HPLC with authentic 3MC derivatives whenever available. In addition to 3MC fazns-l,2-diol, 3MC-l-one, and 3MC-2-one reported earlier by other investigators, 3-hydroxymethylcholanthrylene (3-OHMCE), 3-OHMCE fra/w-ll,12-dihydrodiol, 3MCE frans-ll,12-dihydrodiol, 3MCE frans-9,10-dihydrodiol, 9-and 10-hydroxy-3MCE, 3MC-2-one 3MC-2-one = 3MCE > 1-OH-3MC > 3MC-l-one (non-carcinogenic) (11). 3MCE is a minor metabolite of 3MC (12,13). 3MCE was reported to be metabolized to form rran.s-l,2-dihydroxyl-3MC (3MC rrans-l,2-diol), 3MC-2-one, 3MC-l-one (6,10), a minor amount of 3MC cis-\ ,2-diol (6) and other unidentified products (6,11). It appeared that the carcinogenic activity of 3MCE is due to metabolic formation of 3MC-2-one, a known carcinogenic compound (10,11). However, 3MCE may be metabolically activated via the formation of an unstable 1,2-epoxy-l ,2-dihydro3MCE (3MCE 1,2-epoxide) and/or via formation of the bay region of 3MCE 9,10-diol-7,8-epoxide (Figure 1). 3MCE 1,2-epoxide may be an activated metabolite of 3MCE since a C| benzylic carbonium ion may be an intermediate in the isomerization of 3MCE 1,2-epoxide to form 3MC-2-one and 3MC-l-one. Although qualitative metabolism studies of 3MCE (6,11) and other cyclopenta-fused isomers of benz[a]anthracene (BA) have been reported (4,5), pathways leading to the formation of carcinogenic metabolites have not been established. Enzymecatalyzed epoxidation at the cyclopenta rings of some cyclopentaPAHs is believed to be responsible for the mutagenicity in S. typhimurium (4,5,9). The aim of this study is to ascertain the potential pathway(s) of metabolism responsible for the mutagenic and carcinogenic activities of 3MCE.
S.K.Yang et at.
Catalytic hydrogenation of metabolites An individual metabolite of 3MCE (peak 2 in Figure 2), isolated by reversed-phase HPLC and further purified by normal-phase HPLC, was dissolved in THF (~ 3 ml) and hydrogenated by bubbling hydrogen gas at 1 atm through the solution in the presence of PtO2 • H2O (Adams catalyst; Alfa Products, Thiokol/Ventron Division, Danvers, MA) for 1 h.
Spectral analysis Mass spectral analysis was performed on a Finnigan Model 4000 GC —MS data system by electron impact at 70 eV and 250°C ionizer temperature using the solid probe inlet. UV —visible absorption spectra of samples in methanol were determined using a 1 cm path length quartz cuvette with a Varian Model Cary 118C spectrophotometer.
Reduction of metabolites with NaBH4 Individual metabolites of 3MCE, isolated by reversed-phase HPLC and further purified by normal-phase HPLC, were dissolved in methanol (~ 1 ml), followed by the addition of ~ 5 mg of NaBH4. After - 3 0 min, methanol was evaporated to dryness and the residue was partitioned with ethyl acetate and water. The ethyl acetate phase was washed several times with water and then evaporated to dryness. The resulting product was purified by reversed-phase HPLC.
Results Metabolism of 3MCE by rat liver microsomes The metabolites formed by incubation of 3MCE with rat liver microsomes and an NADPH-regenerating system were separated
HPLC HPLC was performed on a Waters Associates (Milford, MA) liquid chromatograph consisting of a Model 6000A solvent delivery system, a Model M45 solvent
I!, P-450
1
EH
P-450
%'':.
10
AE a o l o c /0= 0.794
cyclopenta[c,rf]pyrono
10
P-450
H*
R = CH,, 3-mathylcholanthrylono (3MCE) R s H, cholanthryleno (bonz[y]aceanthrylana)
P-450 H* ""OH
AE d 0 , o c //i= 0.856 OH
""OH 7 6 benz(£]anthracono
Fig. 1. Examples of known/potential metabolic activation pathways of some polycyclic aromatic hydrocarbons. Values of AEaeloc/IS for carbonium ion formation at benzylic positions are calculated as described by Jerina et al. (8). EH. epoxide hydrolase.
1196
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
delivery system, a Model 660 solvent programmer and a Model 440 absorbance (254 or 280 m) detector. Samples were injected via a Valco model N60 loop injector (Valco Instruments, Houston, TX). Reversed-phase HPLC. The products formed in the incubation of 3MCE by rat liver microsomes were separated using a Waters Associates RCM-100 Radial Compression Module fitted with a Nova-Pak C18 4 (im cartridge (8 mm i.d. x 10 cm). The column was eluted with a 30 min linear gradient of methanol/water (3:2, v/v) to methanol at 2 ml/min. The product, derived either from catalytic hydrogenation or NaBH4 reduction of a 3MCE metabolite, was isolated by reversed-phase HPLC under identical chromatographic conditions. Normal-phase HPLC. The products formed in the metabolism of 3MCE by rat liver microsomes in the presence of TCPO were separated using a Du Pont Zorbax SIL column (6.2 mm i.d. x 25 cm, packed with 5 ^m porous silica microparticles). The column was eluted for 20 min with ethyl acetate/hexane/ triethylamine (100:897:3, v/v/v) at 2 ml/min, followed by elution with ethyl acetate in order to wash off more polar metabolites (14). The metabolites, formed by incubation of 3MCE with rat liver microsomes in the absence of TCPO, were first separated by reversed-phase HPLC as described above and were further purified by normal-phase HPLC (15). 3MC-2-one, 3MC-l-one, 1-0H-3MC and 2-OH-3MC were separated with retention times of 7.5, 7.9, 17.2 and 19.8 min respectively on a Du Pont Zorbax SIL column (6.2 mm i.d. x 25 cm) by elution with THF/hexane (1:9, v/v) at 2 ml/min.
NADPH-regenerating system, either in the absence or in the presence of a microsomal epoxide hydrolyase inhibitor, 3,3,3-trichloropropylene 1,2-oxide (TCPO; Sigma Chemical Co., St Louis, MO), similarly as described (14). A 100 ml reaction mixture contained 100 mg protein equivalent of rat liver microsomes, 5 mmol Tris-HCl (pH 7.5), 0.3 mmol of MgCI2, 10 units of glucose-6-phosphate (G-6-P) dehydrogenase (type XII, Sigma), 10 mg of NADP + , 48 mg of G-6-P, and with or without the addition of 0.06 mmol of TCPO. The reaction mixture was preincubated at 37°C for 5 min in a water shaker bath. 3MCE (8 /xmol in 4 ml of acetone) was then added and the mixture was incubated for 30 min. Residual 3MCE and its metabolites were extracted by sequential additions of 100 ml of acetone/triethylamine (125:1, v/v) and 200 ml of ethyl acetate/triethylamine (250:1, v/v). The resulting aqueous phase was extracted with an additional 200 ml of ethyl acetate/triethylamine (250:1, v/v). Organic solvent extracts were combined and dehydrated with anhydrous MgSO4, filtered and evaporated to dryness under reduced pressure. The residue was redissolved in either tetrahydrofuran (THF)/methanol (1:1, v/v) for reversedphase HPLC or ethyl acetate/hexane/triethylamine (20:79.7:0.3, v/v/v) for normal-phase HPLC separation of metabolites.
Metabolism of 3-methylcholanthrylene
LoJULJU 10 20 RETENTION TIME ( mln)
Fig. 2. Reversed-phase HPLC separation of products formed in the metabolism of 3MCE with liver microsomes from 3MC-treated rats and an NADPH-regenerating system. Unnumbered metabolite peaks have not been identified. Metabolites contained in various chromatographic peaks are: 3-OHMCE trans-11,12-dihydrodiol (peak 1), 3MC-2-one w-9,10-dihydrodiol previously identified as a principal rat liver microsomal metabolite of 3MC-2-one (16). Mass spectral analysis indicated M + at m/z 316 and a characteristic fragment ion at m/z 1197
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
by reversed-phase HPLC (Figure 2). In order to obtain chromatographically pure products for subsequent analyses, each chromatographic peak was further purified by normal-phase HPLC (15). For the purpose of structural identification, several metabolites were derivatized either by catalytic hydrogenation or by reduction with NaHB4. Identities of unmarked chromatographic peaks in Figure 2 have not been established. To date, none of the 3MCE derivatives which retain the 1,2-double bond have been chemically synthesized. Hence, products formed in the oxidative metabolism of 3MCE at positions other than the 1,2-double bond cannot be readily identified. However, because the 1,2-double bond of 3MCE is readily saturated by a catalytic hydrogenation reaction and many authentic derivatives of 3MC were available for this study, metabolites of 3MCE can be easily identified upon derivatization to the corresponding 3MC derivatives. Peak 10 contained the unreacted substrate, 3MCE. The metabolite contained in peak 6 had a UV-visible absorption spectrum essentially the same as that of 3MCE (Figure 3). Mass spectral analyses indicated M + at m/z 282 and characteristic fragment ions at m/z 265 (loss of OH), 253, 252, 250 and 239. It was hydrogenated to a product with a UV -visible absorption spectrum essentially the same as that of 3MC (Figure 3) and M + at m/z 284 and characteristic fragment ions at m/z 267 (loss of OH), 265, 263, 253, 252 and 250 by mass spectral analysis. Hence the metabolite contained in peak 6 was identified as 3-hydroxymethylcholanthrylene (3-OHMCE). The assignment of the metabolite in peak 6 as either 1-OH-3MCE or 2-OH-3MCE can be ruled out, because 1-OH-3MCE and 2-OH-3MCE are expected to tautomerize readily to the corresponding keto isomers. The metabolites contained in peaks 1 and 4 were identified as 3-OHMCE rra>w-11,12-dihydrodiol and 3MCE trans11,12-dihydrodiol respectively. The UV-visible absorption spectra of metabolite peaks 1 and 4 were nearly identical to each other (Figure 4). Mass spectral analyses indicated that peak 1 had a molecular ion (M + ) at m/z 316 and characteristic fragment ions at m/z 298 (loss of H2O) and 282, while peak 4 had a molecular ion (M + ) at m/z 300 and characteristic fragment ions at m/z 282 (loss of H2O) and 253 (loss of CHO). Peaks 1 and 4 of Figure 2 were converted by catalytic hydrogenation to products (M + at m/z 318 and 302 respectively) that had U V visible absorption spectra (Figure 4) similar to that of an authentic 3MC trans-11,12-dihydrodiol. Thus metabolites contained in peaks 1 and 4 were identified as 3-OHMCE trans-X 1,12-dihydrodiol and 3MCE trans-l 1,12-dihydrodiol respectively. 3-OHMCE trans-11,12-dihydrodiol may be derived from further metabolism of 3-OHMCE and/or 3MCE trans-11,12-dihydrodiol. The metabolites contained in peak 9 of Figure 2 were identified as a mixture of 3MC-2-one (-95%) and 3MC-l-one ( - 5 % ) by normal-phase HPLC analysis. However, peak 9 collected from reversed-phase HPLC had a UV-visible absorption spectrum either nearly identical to that of the authentic 3MC-2-one (Figure 5) in some chromatographic runs or behaved as shown by spectrum C in Figure 5 (note the characteristic absorption bands between 320 and 420 nm) in other chromatographic analyses. Spectrum C was taken within 10 min upon its isolation by reversed-phase HPLC. Mass spectral analysis indicated a molecular ion (M + ) at m/z 282 (base peak) and characteristic fragment ions at m/z 254 (loss of CO), 253 (loss of CHO) and 252. Since 3MC-2-one, or 3MC-l-one, and 3MCE 1,2-epoxide have an identical mol. wt, the metabolite contained in peak 9 can be one or a mixture of these three compounds. However, when the metabolite contained in peak 9 was stored in methanol for a period of time or was incubated with rat liver microsomes,
S.K.Yang et al.
B
A
l\
A
1
All
J\ j
ui
o z < m a. % m
MOM,
/
/'"I
-
rtXX
, / | '
/ J/ I
I
I
4
1
1| t |
i
'
1 \_
|
\ \ \
I
J
y \ \ i 250
i
•
^
'
^
\
\ \ 1
1
350
250
350
450
WAVELENGTH (nm)
250
350
450
WAVELENGTH (nm)
250
350
450
WAVELENGTH (nm)
Fig. 6. UV —visible absorption spectra of 3MC-2-one (ra/ii-9,10-dihydrodiol ( , peak 2 of Figure 2) and its NaBH4 reduction product 2-OH-3MC rram-9,10-dihydrodiol (-—).
250
3S0 WAVELENGTH ( n m )
450
Fig. 5. UV-visible absorption spectra of 3MC-l-one (curve B; -—, authentic compound), 3MC-2-one (curve A; , authentic compound) and metabolite peak 9 of Figure 2 (curve C; , tentatively identified as 3MCE 1,2-epoxide).
Fig. 7. UV-visible absorption spectra of 3MCE /ran.s-9,10-dihydrodiol (peak 5 of Figure 2, ) and its hydrogenation product 3MC 7,8,9,10-tetrahydro-/ranj-9,10-diol (—).
298 (loss of H 2 O). It was reduced by NaBH4 to two diastereomeric 2-OH-3MC //ww-9,10-dihydrodiols which had UV-visible absorption spectra (Figure 6) similar to those of 3MC rra/w-9,10-dihydrodiol (17,18) and 1-OH-3MC trans9,10-dihydrodiol (19). Hence the metabolite contained in peak 2 was identified as a 3MC-2-one fra/u-9,10-dihydrodiol. The metabolite contained in peak 5 was identified as 3MCE fra/w-9,10-dihydrodiol. Mass spectral analysis indicated M + at miz 300 and characteristic fragment ions at miz 282 (loss of H2O) and 253 (loss of H2O and CHO). It was converted by catalytic hydrogenation to a product (M + at miz 304 and characteristic fragment ions at miz 286 and 257) with a UV visible absorption spectrum characteristic of an anthracene nucleus (Figure 8). The hydrogenation product (3MC 7,8,9,10-
tetrahydro-frans-9,10-diol) was identical (with respect to UV —visible absorption spectrum, mass spectrum and retention times on both reversed-phase and normal-phase HPLC) to the hydrogenation product derived from 3MC trans-9,10-dihydrodiol, a known metabolite of 3MC (17,20). The metabolite contained in peak 3 was identified as 3MC trans-l,2-d\o\. It had a UV-visible absorption spectrum and a retention time that were identical to those of the authentic 3MC trans-l,2-d\o\. Mass spectral analysis indicated M + at miz 300 and a characteristic fragment ion at miz 282 (loss of H2O). This metabolite was unchanged by the catalytic hydrogenation reaction. Furthermore, its retention time was different from that of 3MC cis-l ,2-diol (with a retention time indicated by an arrow in Figure 2), which was not a detectable metabolite.
1198
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
Fig. 4. UV-visible absorption spectra of (A) 3MCE trans-X 1,12-dihydrodiol (peak 4 of Figure 2, ), 3MC trans-\ 1,12-dihydrodiol (-—; authentic compound) and (B) 3-OHMCE trans-11,12-dihydrodiol (peak 1 of Figure 2; ) and its hydrogenation product 3-OHMC trans-11,12-dihydrodiol (-—).
Metabolism of 3-methylcholanthrylene
10 RETENTION TIME ( min )
20
250 350 450 WAVELENGTH (nm)
Fig. 9. UV-visible absorption spectra of 3MCE 11,12-epoxide (peak C of Figure 8, ) and 3-OHMCE 11,12-epoxide (-—, a metabolite separated by reversed-phase HPLC from the more polar fraction eluted with ethyl acetate in Figure 8).
The metabolites contained in peaks 7 and 8 were identified as 9-OH-3MCE and 10-OH-3MCE respectively. Mass spectral analysis indicated M + at miz 282 and characteristic fragment ions at miz 253 0oss of CHO), 252 and 250. They were converted by catalytic hydrogenation to the products (9-OH-3MC and 10-OH-3MC) which had UV-visible absorption spectra in methanol as well as methanolic NaOH (0.1 N) similar to those of 3-OH-BA and 4-OH-BA respectively (not shown). Mass spectral data (M + at miz 284) of the hydrogenation products were consistent with the structural assignments. Effect of TCPO on 3MCE metabolism The metabolites formed by incubation of 3MCE with rat liver microsomes and an NADPH-regenerating system in the presence of TCPO, an epoxide hydrolase inhibitor, were separated by normal-phase HPLC (Figure 8). Unmarked chromatographic peaks in Figure 8 have not been identified. More polar metabolites were eluted with ethyl acetate. Peak A in Figure 8 contained an unmetabolized substrate,
Discussion 3MCE is a minor metabolite of 3MC (12,13). 3MCE was reported to be metabolized to form 3MC frans-1,2-diol, 1199
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
Fig. 8. Metabolites formed in the incubation of 3MCE by liver microsomes from MC-treated rats in the presence of TCPO. Identities of chromatographic peaks are: (A) 3MCE (substrate remaining); (B) 3MCE l,2-epoxide and/or a mixture of 3MC-2-one and 3MC-l-one (see text for discussion); (C) 3MCE 11,12-epoxide; (D) 3-OHMCE.
3MCE. The identified metabolites were: peak B, 3MCE 1,2-epoxide (which was readily converted to a mixture of 3MC-2-one and 3MC-l-one); peak C, 3MCE 11,12-epoxide; peak D, 3-OHMCE. The metabolite contained in peak B had a UV—visible absorption spectrum identical to what was observed in peak 9 of Figure 2; which is similar to that of 3MC-2-one or spectrum C of Figure 5. Mass spectral analysis indicated M + at miz 282 (base peak) and characteristic fragment ions at miz 254, 253 and 252. When this metabolite was incubated in a 0.05 M Tris-HCl (pH 7.5) buffer containing rat liver microsomes (2 mg protein per ml of incubation mixture; no NADPH was added or generated during the incubation), it was converted to a product with a UV—visible absorption spectrum similar to that of the authentic 3MC-2-one; no 3MC cis-l ,2-diol was formed. Upon storage in methanol, this metabolite was also converted to a product with a UV -visible absorption spectrum similar to that of 3MC-2-one. It was reduced with NaBH4 to 2-OH-3MC (-95%) and 1-OH-3MC ( - 5 % ) , determined by normal-phase HPLC as described in Materials and methods. Thus the metabolite with UV—visible absorption spectrum C in Figure 5 was probably an unstable 3MCE 1,2-epoxide. The metabolite contained in peak C had a UV—visible absorption spectrum (Figure 9) resembling that of 3MCE trans-11,12-dihydrodiol (Figure 4). Mass spectral analysis indicated M + at miz 282 (base peak) and fragment ions at miz 267, 266 and 253. It was converted by incubation with rat liver microsomes in the absence of NADPH to form a 3MCE trans-11,12-dihydrodiol, with properties (UV—visible absorption and mass spectra, retention times on both reversed-phase and normal-phase HPLC) identical to those of metabolite peak 4 in Figure 2, which was identified as 3MCE trans-l 1,12-dihydrodiol. Furthermore, the 3MCE trans-11,12-dihydrodiol derived from the metabolite contained in peak C was reduced by catalytic hydrogenation to form a product with HPLC chromatographic and spectral properties identical to those of the authentic 3MC trans-l 1,12-dihydrodiol. Thus the metabolite contained in peak C in Figure 8 was established to be 3MCE trans-l 1,12-epoxide. The metabolite contained in peak D was identical to the metabolite contained in peak 6 of Figure 2 with respect to retention times on both reversed-phase and normal-phase HPLC, UV—visible absorption spectrum and mass spectrum. It was also converted by catalytic hydrogenation to a product [M + (base peak) at miz 284 with characteristic fragment ions at miz 267 (loss of OH) and 266 (loss of H2O)] which had a UV-visible absorption spectrum similar to that of 3MC (Figure 3). The metabolite contained in peak D of Figure 8 was therefore identified as 3-OHMCE. The more polar metabolites eluted with ethyl acetate were analyzed by reversed-phase HPLC similarly as described in Figure 2. Identifiable metabolites were (in elution order): 3MC fra/w-1,2-diol, 3-OHMCE 11,12-epoxide (UV-visible absorption spectrum in Figure 9), 3-OHMCE, 9-OH-3MCE, 10-OH-3MCE, and some other as yet unidentified products. Incubation of 3-OHMCE 11,12-epoxide with rat liver microsomes yielded 3-OHMCE trans-11,12-dihydrodiol, which was identical to peak 1 of Figure 2. Identification of these metabolites was based on the analyses of their UV-visible absorption and mass spectral data, as well as the spectral properties of their hydrogenation products as described above.
S.K.Yang et al.
Similar products as shown for 3MCE
HOHjC
H,C
P-450
H3C
'OH
Fig. 10. Proposed pathways of 3MCE metabolism in rat liver microsomes. EH, epoxide hydrolase; NE, non-enzymatic rearrangement. Pathways indicated with a question mark have not been established. 3-OHMCE 11,12-epoxide and 3-OHMCE-rra/w-l 1,12-dihydrodiol are known products derived from further metabolism of 3-OHMCE (see Results).
3MC-2-one, 3MC-l-one (6,11), and a minor amount of 3MC c«-l, 2-diol (6) and other unidentified products (6,11). In this study, a large number of rat liver microsomal metabolites of 3MCE have been identified. These include 3-OHMCE, 3-OHMCE trans-] 1,12-dihydrodiol, 3-OHMCE 11,12-epoxide, 3MCE trans-11,12-dihydrodiol, 3MCE 11,12-epoxide, 3MC trans-\,2-dio\, 3MCE rrans-9,10-dihydrodiol, 9-and 10-OH3MCE, 3MC-l-one, 3MC-2-one, 3MC-2-one mzn.s-9,10-dihydrodiol, and an unstable metabolite tentatively identified as 3MCE 1,2-epoxide. Proposed pathways of 3MCE metabolism in rat liver microsomes are shown in Figure 10. In contrast, the metabolism of CE by liver microsomes from Aroclor-1254treated and phenobarbital-treated rats produced cholanthrene trans-1,2-diol, CE 9,10-dihydrodiol, CE 11,12-dihydrodiol and 10-OH-CE as the major metabolites (4,5). By comparing the metabolites formed in the metabolism of 3MCE and CE, it appears that a C3-methyl substituent in CE causes the metabolically formed 3MCE 1,2-epoxide to undergo isomerization to form 3MC-l-one and 3MC-2-one. However, the CE 1,2-epoxide intermediate formed in the metabolism of CE may also undergo isomerization to form cholanthrene-1-one and cholanthrene-2-one, but these were not detected in earlier reports (4,5). Thus it is possible that the metabolite pattern of CE may actually be more complex than has been reported (4,5). The detection of 3MC trans-1,2-diol suggested that the metabolically formed 3MCE 1,2-epoxide was either completely or partially hydrolyzed non-enzymatically to form 3MC trans-l,2-diol. Since an authentic 3MCE 1,2-epoxide was not available, it was not possible to establish if the metabolically formed 3MCE 1,2-epoxide was a substrate of the microsomal epoxide hydrolyase. In contrast, neither 3MCE trans-l 1,12-
1200
dihydrodiol nor 3-OHMCE trans-l 1,12-dihydrodiol was detected when the in vitro incubation was carried out in the presence of TCPO, indicating that non-enzymatic hydrolysis of the metabolically formed K-region 11,12-epoxides did not occur when microsomal epoxide hydrolase was inhibited by TCPO. The carcinogenic activity of 3MCE may be due to metabolic formation of 3MCE 1,2-epoxide and/or via the formation of the bay region 9,10-diol-7,8-epoxides of 3MCE and 3MC-2-one. Although not detected in this study, 3-OHMCE 1,2-epoxide, 3-OHMC-2-one, 3-OHMCE trans-9,10-dihydrodiol, 9-OH3-OHMCE, 10-OH-3-OHMCE and 3-OHMC-2-one trans9,10-dihydrodiol may be potential metabolites of 3-OHMCE. Thus there may be additional pathways of metabolic activation through the hydroxylation at the methyl group of 3MCE. A reactive C, benzylic carbonium ion may be formed as an intermediate (Figure 1) during the isomerization of 3MCE 1,2-epoxide to form 3MC-2-one and 3MC-l-one as well as during hydrolysis to form 3MC trans-l,2-diol. As shown in Figure 1, this carbonium ion has a AEdeioc/l3 value higher than those derived from benzo[a]pyrene 7,8-diol-9,10-epoxide, cyclopenta[c,d]pyrene 3,4-epoxide, BA 3,4-diol-l,2-epoxide and 3MCE 9,10-dioI-7,8-epoxide. Thus, based on theoretical predictions (8), 3MCE 1,2-epoxide is a potential ultimate carcinogenic metabolite of 3MCE and may exhibit higher carcinogenic activity than those derived from benzo[a]pyrene, cyclopenta[c,*/]pyrene and BA (Figure 1). A reactive C| benzylic carbonium ion may similarly be formed from 3-OHMCE 1,2-epoxide, a potential metabolite of 3-OHMCE. Additional studies on the mutagenicity and tumorigenicity of 3MCE metabolites and analysis of activated metabolite-DNA binding adducts should provide a clearer understanding of the activation pathways of 3MCE.
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
9-OH-3MCE and 10-OH-3MCE n t—
Metabolism of 3-methylcholanthrylene
Acknowledgements We thank Magang Shou for technical assistance. This work was supported by an Independent Research and Development award from the MITRE Corporation through the Henry M.Jackson Foundation for the Advancement of Military Medicine Protocol no. G17586 at the Uniformed Services University of the Health Sciences and US Public Health Service grant CA29133. The opinions or assertions contained herein are the private ones of the authors and are not to be construed as official or reflecting the views of the Department of Defense or the Uniformed Services University of the Health Sciences or the MITRE Corporation. The experiments reported herein were conducted according to the principles set forth in the 'Guide for the Care and Use of Laboratory Animals', Institute of Animal Resources, National Research Council, Department of Health, Education and Welfare Publication no. (NIH) 78-23.
formation of dihydrodiols by chemical or enzymic oxidation of 3-methylcholanthrene. Chem.-Biol. Interactions, 23, 121-135. 18.Jacobs,S.A., Cortez.C. and Harvey.R.G. (1983) Synthesis of potential proximate and ultimate carcinogenic metabolites of 3-methylcholanthrene. Carcinogenesis, 4, 519-522. 19. Thakker.D.R., Levin.W., Wood.A.W., Conney.A.H., Stoming,T.A. and Jerina,D.M. (1978) Metabolic formation of l,9,10-trihydroxy-9,10-dihydro3-methylcholanthrene: a potential proximate carcinogen from 3-methylcholanthrene. J. Am. Chem. Soc, 100, 645-647. 20. King.H.W.S., Osborne,M.R. and Brookes,P. (1978) The identification of 3-methylcholanthrene-9,10-dihydrodiol as an intermediate in the binding of 3-methylcholanthrene to DNA in cells in culture. Chem.-Biol. Interactions, 20, 367-371.
References
Received on January 22, 1990; revised on April 6, 1990; accepted on April 19, 1990
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
1. Gold,A. and Eisenstadt,E. (1980) Metabolic activation of cyclopenta(c,d)pyrene to 3,4-epoxycyclopenta(c,d)pyrene by rat liver microsomes. Cancer Res., 40, 3940-3944. 2Gold,A., Nesnow.S., Moore,M., Garland,H., Curtis.G., Howard,B., Graham,D. and Eisenstadt,E. (1980) Mutagenesis and morphological transformation of mammalian cells by a non-bay-region polycyclic cyclopenta(c,d)pyrene and its 3,4-oxide. Cancer Res., 40, 4482—4484. 3. Wood.A.W., Levin,W., Chang.R.L., Huang,M.-T., Ryan,D.E., Thomas, P.E., Lehr,R.E., Kumar.S., Koreeda,M., Akagi.H., Ittah,Y., Dansette.P., Yagi.H., Jerina.D.M. and Conney.A.H. (1980) Mutagenicity and tumorinitiating activity of cyclopenta(c,rf)pyrene and structurally related compounds. Cancer Res., 40, 642-649. 4. Nesnow.S., Leavitt.S., Easterling.R., Watts,R., Toney.S.H., Claxton.L., Sangaiah.R., Toney,G.E., Wiley.J., Fraher.P. and Gold,A. (1984) Mutagenicity of cyclopenta-fused isomers of benz(a)anthracene in bacterial and rodent cells and identification of the major rat liver microsomal metabolites. Cancer Res., 44, 4993-5003. 5. Nesnow.S., Easterling.R.E., Ellis,E., Watts.R. and Ross.J. (1988) Metabolism of benz[/]aceanthrylene and benz[/]aceanthrylene by induced rat liver S9. Cancer Lett., 39, 19-27. 6. Sims,P. (1966) The metabolism of 3-methylcholanthrene and some related compounds by rat-liver homogenates. Biochem. J., 98, 215—228. 7. Fu.P.P., Beland.F.A. and Yang.S.K. (1980) Cyclopenta-polycyclic aromatic hydrocarbons: potent carcinogens and mutagens. Carcinogenesis, 1, 725-727. 8. Jerina.D.M., Lehr.R.E., Yagi,H., Hernandez,O., Dansette.P.M., Wislocki.P.G., Wood.A.W., Chang.R.L., Levin,W. and Conney.A.H. (1976) Mutagenicity of benzo[a]pyrene derivatives and the description of a quantum mechanical model which predicts the ease of carbonium ion formation from diol epoxides. In de Serres.F.J., Fouts,J.R., BendJ.R. and Philpot,R.M. (eds), In Vitro Metabolic Activation in Mutagenesis Testing. Elsevier North Holland Biomedical Press, Amsterdam, pp. 159 — 177. 9. Prasanna.P., Jacobs,M.M. and Yang,S.K. (1987) Selenium inhibition of benzo[a]pyrene, 3-methylcholanthrene, and 3-methylcholanthrylene mutagenicity in Salmonella typhimurium strains TA98 and TA100. Mutat. Res., 190, 101-105. 10. Sims,P. (1967) The carcinogenic activities in mice of compounds related to 3-methylcholanthrene. Int. J. Cancer, 2, 505-508. ll.Cavalieri.E., Roth.R., Althoff.J., Grandjean.C, Patil.K., Marsh,S. and McLaughlin,D. (1978) Carcinogenicity and metabolic profiles of 3-methylcholanthrene oxygenated derivative at the 1 and 2 positions. Chem.-Biol. Interactions, 22, 6 9 - 8 1 . 12.Thakker,D.R., Levin,W., Stoming,T.A., Conney.A.H. and Jerina.D.M. (1978) Metabolism of 3-methylcholanthrene by rat liver microsomes and a highly purified monooxygenase system with and without epoxide hydrase. In Jones.P.W. and Freudenthal,R.I. (eds), Carcinogenesis, Vol. 3: Polynuclear Aromatic Hydrocarbons. Raven Press, New York, pp. 253-264. 13.Stoming,T.A., Bomstein.W. and Bresnick,E. (1977) The metabolism of 3-methylcholanthrene by rat liver microsomes—a reinvestigation. Biochem. Biophys. Res. Commun., 79, 461-469. 14. Mushtaq.M., Weems.H.B. and Yang.S.K. (1986) Metabolic and stereoselective formations of non-K-region benz[a]anthracene 8,9- and 10,11-epoxides. Arch. Biochem. Biophys., 246, 478-487. 15. Chou,M.W. and Yang,S.K. (1979) Combined reversed-phase and normalphase high performance liquid chromatography in the purification and identification of 7,12-dimethylbenz[a]anthracene metabolites. J. Chromaiogr., 185, 635-654. 16. Shou.M. and Yang,S.K. (1990) 9,10-Dihydroxy-9,10-dihydro-3-methylcholanthrene-2-one: a principal metabolite formed in the metabolism of the potent carcinogen 3-methylcholanthrene-2-one by rat liver microsomes. Carcinogenesis, 11, 689—695. 17,Tierney,B., Hewer.A., Rattle,A., Grover.P.L. and Sims.P. (1978) The
1201
Metabolism of (tttne potent carcinogen! 3-metIhykIhiollainitlhuryIeinie microsomes
Shen K.Yang, Fremakala Prasanna, Henri B.Weems, Maryce M.Jacobs1 and Peter P.Fu 2 Department of Pharmacology, F.Edward Hebert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, MD 20814-4799, 'The MITRE Corporation/Metrek Division, McLean, VA 22101 and 2National Center for Toxicological Research, Jefferson, AR 72079, USA 'Present address: American Institute for Cancer Research, 1759 R Street NW, Washington, DC 200O9, USA
Introduction The cyclopenta-ring of cyclopenta-polycyclic aromatic hydrocarbons (cyclopenta-PAHs*) may be a site of oxidative biotransformation to form the ultimate mutagenic and carcinogenic metabolites (1 - 7 ) . Examples of cyclopenta-PAHs studied to date include cyclopenta[c,d]pyrene (1 —3) and cyclopenta-fused isomers of benz[a]anthracene (4,5). Cholanthrylene (CE, an •Abbreviations: cyclopenta-PAHs, cyclopenta-polycyclic aromatic hydrocarbons; CE, cholanthrylene (also known as benz[/]aceanthrylene); 3MCE, 3-methylcholanthrylene; 3MC, 3-methylcholanthrene; 1-OH-3MC, 1-hydroxy-3MC—other hydroxylated products are similarly abbreviated; 3MC trans-] ,2-diol, trans-] ,2dihydroxy-3MC—the cis compound and the equivalent 3MCE compounds are similarly abbreviated; 3MCE 1,2-epoxide, l,2-epoxy-l,2-dihydro-3MCE; BA, benz[a]anthracene; DDQ, 2,3-dichloro-5,6-dicyano-l,4-benzoquinone; TCPO, 3,3,3-trichloropropylene 1,2-oxide; G-6-P, glucose-6-phosphate; THF, tetrahydroftiran; 3-OHMCE, 3-hydroxymethylcholanthrylene; 3MCE;rani-9,10-dihydrodiol, /ra«i-9,10-dihydroxy-9,10-dihydro-3MCE—other dihydrodiols are similarly abbreviated.
Materials and methods Materials 3MCE was prepared by dehydrogenation of 3MC by refluxing a benzene solution of 3MC with a molar excess of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ; Aldrich Chemical Co., Milwaukee, WI). 3MCE cu-l,2-diol was prepared by oxidation of 3MCE with OsO4 (Aldrich Chemical Co., Milwaukee, WI) in pyridine. 3MC-2-one was prepared by dehydrogenation of 3MCE ris-1,2-diol as described (6,11). 3MC trans-1,2-diol was prepared by dehydrogenation of 3MCEc/>l,2-diol with DDQ to form 3MC 1,2-quinone, followed by reduction of NaBH4 (Fisher Scientific Co., Fair Lawn, NJ) in methanol/water (9:1, v/v). Authentic 1-0H-3MC, 2-OH-3MC, 3MC-l-one, 3MC cis-] 1,12-dihydrodiol, 3MC 11,12-quinone, 11-OH-3MC, 3-OH-BA and 4-OH-BA were obtained from the Chemical Repository of the National Cancer Institute. Reaction of 3MC 11,12-quinone with NaBH4 in methanol/water (9:1, v/v) yielded a mixture of 3MC transA 1,12-dihydrodiol (47%) and 3MC cis-11,12-dihydrodiol (53%), which were separated by reversed-phase HPLC (see below). Incubation of 3MCE with rat liver microsomei The metabolites were isolated from a mixture of products formed by incubation of 3MCE with liver microsomes prepared from 3MC-treated rats and an
1195
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
The products formed in the metabolism of 3-methylcholanthrylene (3MCE), either in the presence or in the absence of an epoxide hydrolase inhibitor, 3,3,3-trichloropropylene 1,2-oxide (TCPO), with an NADPH-regenerating system and liver microsomes from 3-methylcholanthrene (3MC)-treated male Sprague-Oawley rats were separated by reversed-phase and normal-phase HPLC. The metabolites were characterized by UV — visible absorption spectral analysis, and by comparing their retention times on reversed-phase and normal-phase HPLC with authentic 3MC derivatives whenever available. In addition to 3MC fazns-l,2-diol, 3MC-l-one, and 3MC-2-one reported earlier by other investigators, 3-hydroxymethylcholanthrylene (3-OHMCE), 3-OHMCE fra/w-ll,12-dihydrodiol, 3MCE frans-ll,12-dihydrodiol, 3MCE frans-9,10-dihydrodiol, 9-and 10-hydroxy-3MCE, 3MC-2-one 3MC-2-one = 3MCE > 1-OH-3MC > 3MC-l-one (non-carcinogenic) (11). 3MCE is a minor metabolite of 3MC (12,13). 3MCE was reported to be metabolized to form rran.s-l,2-dihydroxyl-3MC (3MC rrans-l,2-diol), 3MC-2-one, 3MC-l-one (6,10), a minor amount of 3MC cis-\ ,2-diol (6) and other unidentified products (6,11). It appeared that the carcinogenic activity of 3MCE is due to metabolic formation of 3MC-2-one, a known carcinogenic compound (10,11). However, 3MCE may be metabolically activated via the formation of an unstable 1,2-epoxy-l ,2-dihydro3MCE (3MCE 1,2-epoxide) and/or via formation of the bay region of 3MCE 9,10-diol-7,8-epoxide (Figure 1). 3MCE 1,2-epoxide may be an activated metabolite of 3MCE since a C| benzylic carbonium ion may be an intermediate in the isomerization of 3MCE 1,2-epoxide to form 3MC-2-one and 3MC-l-one. Although qualitative metabolism studies of 3MCE (6,11) and other cyclopenta-fused isomers of benz[a]anthracene (BA) have been reported (4,5), pathways leading to the formation of carcinogenic metabolites have not been established. Enzymecatalyzed epoxidation at the cyclopenta rings of some cyclopentaPAHs is believed to be responsible for the mutagenicity in S. typhimurium (4,5,9). The aim of this study is to ascertain the potential pathway(s) of metabolism responsible for the mutagenic and carcinogenic activities of 3MCE.
S.K.Yang et at.
Catalytic hydrogenation of metabolites An individual metabolite of 3MCE (peak 2 in Figure 2), isolated by reversed-phase HPLC and further purified by normal-phase HPLC, was dissolved in THF (~ 3 ml) and hydrogenated by bubbling hydrogen gas at 1 atm through the solution in the presence of PtO2 • H2O (Adams catalyst; Alfa Products, Thiokol/Ventron Division, Danvers, MA) for 1 h.
Spectral analysis Mass spectral analysis was performed on a Finnigan Model 4000 GC —MS data system by electron impact at 70 eV and 250°C ionizer temperature using the solid probe inlet. UV —visible absorption spectra of samples in methanol were determined using a 1 cm path length quartz cuvette with a Varian Model Cary 118C spectrophotometer.
Reduction of metabolites with NaBH4 Individual metabolites of 3MCE, isolated by reversed-phase HPLC and further purified by normal-phase HPLC, were dissolved in methanol (~ 1 ml), followed by the addition of ~ 5 mg of NaBH4. After - 3 0 min, methanol was evaporated to dryness and the residue was partitioned with ethyl acetate and water. The ethyl acetate phase was washed several times with water and then evaporated to dryness. The resulting product was purified by reversed-phase HPLC.
Results Metabolism of 3MCE by rat liver microsomes The metabolites formed by incubation of 3MCE with rat liver microsomes and an NADPH-regenerating system were separated
HPLC HPLC was performed on a Waters Associates (Milford, MA) liquid chromatograph consisting of a Model 6000A solvent delivery system, a Model M45 solvent
I!, P-450
1
EH
P-450
%'':.
10
AE a o l o c /0= 0.794
cyclopenta[c,rf]pyrono
10
P-450
H*
R = CH,, 3-mathylcholanthrylono (3MCE) R s H, cholanthryleno (bonz[y]aceanthrylana)
P-450 H* ""OH
AE d 0 , o c //i= 0.856 OH
""OH 7 6 benz(£]anthracono
Fig. 1. Examples of known/potential metabolic activation pathways of some polycyclic aromatic hydrocarbons. Values of AEaeloc/IS for carbonium ion formation at benzylic positions are calculated as described by Jerina et al. (8). EH. epoxide hydrolase.
1196
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
delivery system, a Model 660 solvent programmer and a Model 440 absorbance (254 or 280 m) detector. Samples were injected via a Valco model N60 loop injector (Valco Instruments, Houston, TX). Reversed-phase HPLC. The products formed in the incubation of 3MCE by rat liver microsomes were separated using a Waters Associates RCM-100 Radial Compression Module fitted with a Nova-Pak C18 4 (im cartridge (8 mm i.d. x 10 cm). The column was eluted with a 30 min linear gradient of methanol/water (3:2, v/v) to methanol at 2 ml/min. The product, derived either from catalytic hydrogenation or NaBH4 reduction of a 3MCE metabolite, was isolated by reversed-phase HPLC under identical chromatographic conditions. Normal-phase HPLC. The products formed in the metabolism of 3MCE by rat liver microsomes in the presence of TCPO were separated using a Du Pont Zorbax SIL column (6.2 mm i.d. x 25 cm, packed with 5 ^m porous silica microparticles). The column was eluted for 20 min with ethyl acetate/hexane/ triethylamine (100:897:3, v/v/v) at 2 ml/min, followed by elution with ethyl acetate in order to wash off more polar metabolites (14). The metabolites, formed by incubation of 3MCE with rat liver microsomes in the absence of TCPO, were first separated by reversed-phase HPLC as described above and were further purified by normal-phase HPLC (15). 3MC-2-one, 3MC-l-one, 1-0H-3MC and 2-OH-3MC were separated with retention times of 7.5, 7.9, 17.2 and 19.8 min respectively on a Du Pont Zorbax SIL column (6.2 mm i.d. x 25 cm) by elution with THF/hexane (1:9, v/v) at 2 ml/min.
NADPH-regenerating system, either in the absence or in the presence of a microsomal epoxide hydrolyase inhibitor, 3,3,3-trichloropropylene 1,2-oxide (TCPO; Sigma Chemical Co., St Louis, MO), similarly as described (14). A 100 ml reaction mixture contained 100 mg protein equivalent of rat liver microsomes, 5 mmol Tris-HCl (pH 7.5), 0.3 mmol of MgCI2, 10 units of glucose-6-phosphate (G-6-P) dehydrogenase (type XII, Sigma), 10 mg of NADP + , 48 mg of G-6-P, and with or without the addition of 0.06 mmol of TCPO. The reaction mixture was preincubated at 37°C for 5 min in a water shaker bath. 3MCE (8 /xmol in 4 ml of acetone) was then added and the mixture was incubated for 30 min. Residual 3MCE and its metabolites were extracted by sequential additions of 100 ml of acetone/triethylamine (125:1, v/v) and 200 ml of ethyl acetate/triethylamine (250:1, v/v). The resulting aqueous phase was extracted with an additional 200 ml of ethyl acetate/triethylamine (250:1, v/v). Organic solvent extracts were combined and dehydrated with anhydrous MgSO4, filtered and evaporated to dryness under reduced pressure. The residue was redissolved in either tetrahydrofuran (THF)/methanol (1:1, v/v) for reversedphase HPLC or ethyl acetate/hexane/triethylamine (20:79.7:0.3, v/v/v) for normal-phase HPLC separation of metabolites.
Metabolism of 3-methylcholanthrylene
LoJULJU 10 20 RETENTION TIME ( mln)
Fig. 2. Reversed-phase HPLC separation of products formed in the metabolism of 3MCE with liver microsomes from 3MC-treated rats and an NADPH-regenerating system. Unnumbered metabolite peaks have not been identified. Metabolites contained in various chromatographic peaks are: 3-OHMCE trans-11,12-dihydrodiol (peak 1), 3MC-2-one w-9,10-dihydrodiol previously identified as a principal rat liver microsomal metabolite of 3MC-2-one (16). Mass spectral analysis indicated M + at m/z 316 and a characteristic fragment ion at m/z 1197
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
by reversed-phase HPLC (Figure 2). In order to obtain chromatographically pure products for subsequent analyses, each chromatographic peak was further purified by normal-phase HPLC (15). For the purpose of structural identification, several metabolites were derivatized either by catalytic hydrogenation or by reduction with NaHB4. Identities of unmarked chromatographic peaks in Figure 2 have not been established. To date, none of the 3MCE derivatives which retain the 1,2-double bond have been chemically synthesized. Hence, products formed in the oxidative metabolism of 3MCE at positions other than the 1,2-double bond cannot be readily identified. However, because the 1,2-double bond of 3MCE is readily saturated by a catalytic hydrogenation reaction and many authentic derivatives of 3MC were available for this study, metabolites of 3MCE can be easily identified upon derivatization to the corresponding 3MC derivatives. Peak 10 contained the unreacted substrate, 3MCE. The metabolite contained in peak 6 had a UV-visible absorption spectrum essentially the same as that of 3MCE (Figure 3). Mass spectral analyses indicated M + at m/z 282 and characteristic fragment ions at m/z 265 (loss of OH), 253, 252, 250 and 239. It was hydrogenated to a product with a UV -visible absorption spectrum essentially the same as that of 3MC (Figure 3) and M + at m/z 284 and characteristic fragment ions at m/z 267 (loss of OH), 265, 263, 253, 252 and 250 by mass spectral analysis. Hence the metabolite contained in peak 6 was identified as 3-hydroxymethylcholanthrylene (3-OHMCE). The assignment of the metabolite in peak 6 as either 1-OH-3MCE or 2-OH-3MCE can be ruled out, because 1-OH-3MCE and 2-OH-3MCE are expected to tautomerize readily to the corresponding keto isomers. The metabolites contained in peaks 1 and 4 were identified as 3-OHMCE rra>w-11,12-dihydrodiol and 3MCE trans11,12-dihydrodiol respectively. The UV-visible absorption spectra of metabolite peaks 1 and 4 were nearly identical to each other (Figure 4). Mass spectral analyses indicated that peak 1 had a molecular ion (M + ) at m/z 316 and characteristic fragment ions at m/z 298 (loss of H2O) and 282, while peak 4 had a molecular ion (M + ) at m/z 300 and characteristic fragment ions at m/z 282 (loss of H2O) and 253 (loss of CHO). Peaks 1 and 4 of Figure 2 were converted by catalytic hydrogenation to products (M + at m/z 318 and 302 respectively) that had U V visible absorption spectra (Figure 4) similar to that of an authentic 3MC trans-11,12-dihydrodiol. Thus metabolites contained in peaks 1 and 4 were identified as 3-OHMCE trans-X 1,12-dihydrodiol and 3MCE trans-l 1,12-dihydrodiol respectively. 3-OHMCE trans-11,12-dihydrodiol may be derived from further metabolism of 3-OHMCE and/or 3MCE trans-11,12-dihydrodiol. The metabolites contained in peak 9 of Figure 2 were identified as a mixture of 3MC-2-one (-95%) and 3MC-l-one ( - 5 % ) by normal-phase HPLC analysis. However, peak 9 collected from reversed-phase HPLC had a UV-visible absorption spectrum either nearly identical to that of the authentic 3MC-2-one (Figure 5) in some chromatographic runs or behaved as shown by spectrum C in Figure 5 (note the characteristic absorption bands between 320 and 420 nm) in other chromatographic analyses. Spectrum C was taken within 10 min upon its isolation by reversed-phase HPLC. Mass spectral analysis indicated a molecular ion (M + ) at m/z 282 (base peak) and characteristic fragment ions at m/z 254 (loss of CO), 253 (loss of CHO) and 252. Since 3MC-2-one, or 3MC-l-one, and 3MCE 1,2-epoxide have an identical mol. wt, the metabolite contained in peak 9 can be one or a mixture of these three compounds. However, when the metabolite contained in peak 9 was stored in methanol for a period of time or was incubated with rat liver microsomes,
S.K.Yang et al.
B
A
l\
A
1
All
J\ j
ui
o z < m a. % m
MOM,
/
/'"I
-
rtXX
, / | '
/ J/ I
I
I
4
1
1| t |
i
'
1 \_
|
\ \ \
I
J
y \ \ i 250
i
•
^
'
^
\
\ \ 1
1
350
250
350
450
WAVELENGTH (nm)
250
350
450
WAVELENGTH (nm)
250
350
450
WAVELENGTH (nm)
Fig. 6. UV —visible absorption spectra of 3MC-2-one (ra/ii-9,10-dihydrodiol ( , peak 2 of Figure 2) and its NaBH4 reduction product 2-OH-3MC rram-9,10-dihydrodiol (-—).
250
3S0 WAVELENGTH ( n m )
450
Fig. 5. UV-visible absorption spectra of 3MC-l-one (curve B; -—, authentic compound), 3MC-2-one (curve A; , authentic compound) and metabolite peak 9 of Figure 2 (curve C; , tentatively identified as 3MCE 1,2-epoxide).
Fig. 7. UV-visible absorption spectra of 3MCE /ran.s-9,10-dihydrodiol (peak 5 of Figure 2, ) and its hydrogenation product 3MC 7,8,9,10-tetrahydro-/ranj-9,10-diol (—).
298 (loss of H 2 O). It was reduced by NaBH4 to two diastereomeric 2-OH-3MC //ww-9,10-dihydrodiols which had UV-visible absorption spectra (Figure 6) similar to those of 3MC rra/w-9,10-dihydrodiol (17,18) and 1-OH-3MC trans9,10-dihydrodiol (19). Hence the metabolite contained in peak 2 was identified as a 3MC-2-one fra/u-9,10-dihydrodiol. The metabolite contained in peak 5 was identified as 3MCE fra/w-9,10-dihydrodiol. Mass spectral analysis indicated M + at miz 300 and characteristic fragment ions at miz 282 (loss of H2O) and 253 (loss of H2O and CHO). It was converted by catalytic hydrogenation to a product (M + at miz 304 and characteristic fragment ions at miz 286 and 257) with a UV visible absorption spectrum characteristic of an anthracene nucleus (Figure 8). The hydrogenation product (3MC 7,8,9,10-
tetrahydro-frans-9,10-diol) was identical (with respect to UV —visible absorption spectrum, mass spectrum and retention times on both reversed-phase and normal-phase HPLC) to the hydrogenation product derived from 3MC trans-9,10-dihydrodiol, a known metabolite of 3MC (17,20). The metabolite contained in peak 3 was identified as 3MC trans-l,2-d\o\. It had a UV-visible absorption spectrum and a retention time that were identical to those of the authentic 3MC trans-l,2-d\o\. Mass spectral analysis indicated M + at miz 300 and a characteristic fragment ion at miz 282 (loss of H2O). This metabolite was unchanged by the catalytic hydrogenation reaction. Furthermore, its retention time was different from that of 3MC cis-l ,2-diol (with a retention time indicated by an arrow in Figure 2), which was not a detectable metabolite.
1198
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
Fig. 4. UV-visible absorption spectra of (A) 3MCE trans-X 1,12-dihydrodiol (peak 4 of Figure 2, ), 3MC trans-\ 1,12-dihydrodiol (-—; authentic compound) and (B) 3-OHMCE trans-11,12-dihydrodiol (peak 1 of Figure 2; ) and its hydrogenation product 3-OHMC trans-11,12-dihydrodiol (-—).
Metabolism of 3-methylcholanthrylene
10 RETENTION TIME ( min )
20
250 350 450 WAVELENGTH (nm)
Fig. 9. UV-visible absorption spectra of 3MCE 11,12-epoxide (peak C of Figure 8, ) and 3-OHMCE 11,12-epoxide (-—, a metabolite separated by reversed-phase HPLC from the more polar fraction eluted with ethyl acetate in Figure 8).
The metabolites contained in peaks 7 and 8 were identified as 9-OH-3MCE and 10-OH-3MCE respectively. Mass spectral analysis indicated M + at miz 282 and characteristic fragment ions at miz 253 0oss of CHO), 252 and 250. They were converted by catalytic hydrogenation to the products (9-OH-3MC and 10-OH-3MC) which had UV-visible absorption spectra in methanol as well as methanolic NaOH (0.1 N) similar to those of 3-OH-BA and 4-OH-BA respectively (not shown). Mass spectral data (M + at miz 284) of the hydrogenation products were consistent with the structural assignments. Effect of TCPO on 3MCE metabolism The metabolites formed by incubation of 3MCE with rat liver microsomes and an NADPH-regenerating system in the presence of TCPO, an epoxide hydrolase inhibitor, were separated by normal-phase HPLC (Figure 8). Unmarked chromatographic peaks in Figure 8 have not been identified. More polar metabolites were eluted with ethyl acetate. Peak A in Figure 8 contained an unmetabolized substrate,
Discussion 3MCE is a minor metabolite of 3MC (12,13). 3MCE was reported to be metabolized to form 3MC frans-1,2-diol, 1199
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
Fig. 8. Metabolites formed in the incubation of 3MCE by liver microsomes from MC-treated rats in the presence of TCPO. Identities of chromatographic peaks are: (A) 3MCE (substrate remaining); (B) 3MCE l,2-epoxide and/or a mixture of 3MC-2-one and 3MC-l-one (see text for discussion); (C) 3MCE 11,12-epoxide; (D) 3-OHMCE.
3MCE. The identified metabolites were: peak B, 3MCE 1,2-epoxide (which was readily converted to a mixture of 3MC-2-one and 3MC-l-one); peak C, 3MCE 11,12-epoxide; peak D, 3-OHMCE. The metabolite contained in peak B had a UV—visible absorption spectrum identical to what was observed in peak 9 of Figure 2; which is similar to that of 3MC-2-one or spectrum C of Figure 5. Mass spectral analysis indicated M + at miz 282 (base peak) and characteristic fragment ions at miz 254, 253 and 252. When this metabolite was incubated in a 0.05 M Tris-HCl (pH 7.5) buffer containing rat liver microsomes (2 mg protein per ml of incubation mixture; no NADPH was added or generated during the incubation), it was converted to a product with a UV—visible absorption spectrum similar to that of the authentic 3MC-2-one; no 3MC cis-l ,2-diol was formed. Upon storage in methanol, this metabolite was also converted to a product with a UV -visible absorption spectrum similar to that of 3MC-2-one. It was reduced with NaBH4 to 2-OH-3MC (-95%) and 1-OH-3MC ( - 5 % ) , determined by normal-phase HPLC as described in Materials and methods. Thus the metabolite with UV—visible absorption spectrum C in Figure 5 was probably an unstable 3MCE 1,2-epoxide. The metabolite contained in peak C had a UV—visible absorption spectrum (Figure 9) resembling that of 3MCE trans-11,12-dihydrodiol (Figure 4). Mass spectral analysis indicated M + at miz 282 (base peak) and fragment ions at miz 267, 266 and 253. It was converted by incubation with rat liver microsomes in the absence of NADPH to form a 3MCE trans-11,12-dihydrodiol, with properties (UV—visible absorption and mass spectra, retention times on both reversed-phase and normal-phase HPLC) identical to those of metabolite peak 4 in Figure 2, which was identified as 3MCE trans-l 1,12-dihydrodiol. Furthermore, the 3MCE trans-11,12-dihydrodiol derived from the metabolite contained in peak C was reduced by catalytic hydrogenation to form a product with HPLC chromatographic and spectral properties identical to those of the authentic 3MC trans-l 1,12-dihydrodiol. Thus the metabolite contained in peak C in Figure 8 was established to be 3MCE trans-l 1,12-epoxide. The metabolite contained in peak D was identical to the metabolite contained in peak 6 of Figure 2 with respect to retention times on both reversed-phase and normal-phase HPLC, UV—visible absorption spectrum and mass spectrum. It was also converted by catalytic hydrogenation to a product [M + (base peak) at miz 284 with characteristic fragment ions at miz 267 (loss of OH) and 266 (loss of H2O)] which had a UV-visible absorption spectrum similar to that of 3MC (Figure 3). The metabolite contained in peak D of Figure 8 was therefore identified as 3-OHMCE. The more polar metabolites eluted with ethyl acetate were analyzed by reversed-phase HPLC similarly as described in Figure 2. Identifiable metabolites were (in elution order): 3MC fra/w-1,2-diol, 3-OHMCE 11,12-epoxide (UV-visible absorption spectrum in Figure 9), 3-OHMCE, 9-OH-3MCE, 10-OH-3MCE, and some other as yet unidentified products. Incubation of 3-OHMCE 11,12-epoxide with rat liver microsomes yielded 3-OHMCE trans-11,12-dihydrodiol, which was identical to peak 1 of Figure 2. Identification of these metabolites was based on the analyses of their UV-visible absorption and mass spectral data, as well as the spectral properties of their hydrogenation products as described above.
S.K.Yang et al.
Similar products as shown for 3MCE
HOHjC
H,C
P-450
H3C
'OH
Fig. 10. Proposed pathways of 3MCE metabolism in rat liver microsomes. EH, epoxide hydrolase; NE, non-enzymatic rearrangement. Pathways indicated with a question mark have not been established. 3-OHMCE 11,12-epoxide and 3-OHMCE-rra/w-l 1,12-dihydrodiol are known products derived from further metabolism of 3-OHMCE (see Results).
3MC-2-one, 3MC-l-one (6,11), and a minor amount of 3MC c«-l, 2-diol (6) and other unidentified products (6,11). In this study, a large number of rat liver microsomal metabolites of 3MCE have been identified. These include 3-OHMCE, 3-OHMCE trans-] 1,12-dihydrodiol, 3-OHMCE 11,12-epoxide, 3MCE trans-11,12-dihydrodiol, 3MCE 11,12-epoxide, 3MC trans-\,2-dio\, 3MCE rrans-9,10-dihydrodiol, 9-and 10-OH3MCE, 3MC-l-one, 3MC-2-one, 3MC-2-one mzn.s-9,10-dihydrodiol, and an unstable metabolite tentatively identified as 3MCE 1,2-epoxide. Proposed pathways of 3MCE metabolism in rat liver microsomes are shown in Figure 10. In contrast, the metabolism of CE by liver microsomes from Aroclor-1254treated and phenobarbital-treated rats produced cholanthrene trans-1,2-diol, CE 9,10-dihydrodiol, CE 11,12-dihydrodiol and 10-OH-CE as the major metabolites (4,5). By comparing the metabolites formed in the metabolism of 3MCE and CE, it appears that a C3-methyl substituent in CE causes the metabolically formed 3MCE 1,2-epoxide to undergo isomerization to form 3MC-l-one and 3MC-2-one. However, the CE 1,2-epoxide intermediate formed in the metabolism of CE may also undergo isomerization to form cholanthrene-1-one and cholanthrene-2-one, but these were not detected in earlier reports (4,5). Thus it is possible that the metabolite pattern of CE may actually be more complex than has been reported (4,5). The detection of 3MC trans-1,2-diol suggested that the metabolically formed 3MCE 1,2-epoxide was either completely or partially hydrolyzed non-enzymatically to form 3MC trans-l,2-diol. Since an authentic 3MCE 1,2-epoxide was not available, it was not possible to establish if the metabolically formed 3MCE 1,2-epoxide was a substrate of the microsomal epoxide hydrolyase. In contrast, neither 3MCE trans-l 1,12-
1200
dihydrodiol nor 3-OHMCE trans-l 1,12-dihydrodiol was detected when the in vitro incubation was carried out in the presence of TCPO, indicating that non-enzymatic hydrolysis of the metabolically formed K-region 11,12-epoxides did not occur when microsomal epoxide hydrolase was inhibited by TCPO. The carcinogenic activity of 3MCE may be due to metabolic formation of 3MCE 1,2-epoxide and/or via the formation of the bay region 9,10-diol-7,8-epoxides of 3MCE and 3MC-2-one. Although not detected in this study, 3-OHMCE 1,2-epoxide, 3-OHMC-2-one, 3-OHMCE trans-9,10-dihydrodiol, 9-OH3-OHMCE, 10-OH-3-OHMCE and 3-OHMC-2-one trans9,10-dihydrodiol may be potential metabolites of 3-OHMCE. Thus there may be additional pathways of metabolic activation through the hydroxylation at the methyl group of 3MCE. A reactive C, benzylic carbonium ion may be formed as an intermediate (Figure 1) during the isomerization of 3MCE 1,2-epoxide to form 3MC-2-one and 3MC-l-one as well as during hydrolysis to form 3MC trans-l,2-diol. As shown in Figure 1, this carbonium ion has a AEdeioc/l3 value higher than those derived from benzo[a]pyrene 7,8-diol-9,10-epoxide, cyclopenta[c,d]pyrene 3,4-epoxide, BA 3,4-diol-l,2-epoxide and 3MCE 9,10-dioI-7,8-epoxide. Thus, based on theoretical predictions (8), 3MCE 1,2-epoxide is a potential ultimate carcinogenic metabolite of 3MCE and may exhibit higher carcinogenic activity than those derived from benzo[a]pyrene, cyclopenta[c,*/]pyrene and BA (Figure 1). A reactive C| benzylic carbonium ion may similarly be formed from 3-OHMCE 1,2-epoxide, a potential metabolite of 3-OHMCE. Additional studies on the mutagenicity and tumorigenicity of 3MCE metabolites and analysis of activated metabolite-DNA binding adducts should provide a clearer understanding of the activation pathways of 3MCE.
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
9-OH-3MCE and 10-OH-3MCE n t—
Metabolism of 3-methylcholanthrylene
Acknowledgements We thank Magang Shou for technical assistance. This work was supported by an Independent Research and Development award from the MITRE Corporation through the Henry M.Jackson Foundation for the Advancement of Military Medicine Protocol no. G17586 at the Uniformed Services University of the Health Sciences and US Public Health Service grant CA29133. The opinions or assertions contained herein are the private ones of the authors and are not to be construed as official or reflecting the views of the Department of Defense or the Uniformed Services University of the Health Sciences or the MITRE Corporation. The experiments reported herein were conducted according to the principles set forth in the 'Guide for the Care and Use of Laboratory Animals', Institute of Animal Resources, National Research Council, Department of Health, Education and Welfare Publication no. (NIH) 78-23.
formation of dihydrodiols by chemical or enzymic oxidation of 3-methylcholanthrene. Chem.-Biol. Interactions, 23, 121-135. 18.Jacobs,S.A., Cortez.C. and Harvey.R.G. (1983) Synthesis of potential proximate and ultimate carcinogenic metabolites of 3-methylcholanthrene. Carcinogenesis, 4, 519-522. 19. Thakker.D.R., Levin.W., Wood.A.W., Conney.A.H., Stoming,T.A. and Jerina,D.M. (1978) Metabolic formation of l,9,10-trihydroxy-9,10-dihydro3-methylcholanthrene: a potential proximate carcinogen from 3-methylcholanthrene. J. Am. Chem. Soc, 100, 645-647. 20. King.H.W.S., Osborne,M.R. and Brookes,P. (1978) The identification of 3-methylcholanthrene-9,10-dihydrodiol as an intermediate in the binding of 3-methylcholanthrene to DNA in cells in culture. Chem.-Biol. Interactions, 20, 367-371.
References
Received on January 22, 1990; revised on April 6, 1990; accepted on April 19, 1990
Downloaded from http://carcin.oxfordjournals.org/ at Indiana University Library on May 10, 2015
1. Gold,A. and Eisenstadt,E. (1980) Metabolic activation of cyclopenta(c,d)pyrene to 3,4-epoxycyclopenta(c,d)pyrene by rat liver microsomes. Cancer Res., 40, 3940-3944. 2Gold,A., Nesnow.S., Moore,M., Garland,H., Curtis.G., Howard,B., Graham,D. and Eisenstadt,E. (1980) Mutagenesis and morphological transformation of mammalian cells by a non-bay-region polycyclic cyclopenta(c,d)pyrene and its 3,4-oxide. Cancer Res., 40, 4482—4484. 3. Wood.A.W., Levin,W., Chang.R.L., Huang,M.-T., Ryan,D.E., Thomas, P.E., Lehr,R.E., Kumar.S., Koreeda,M., Akagi.H., Ittah,Y., Dansette.P., Yagi.H., Jerina.D.M. and Conney.A.H. (1980) Mutagenicity and tumorinitiating activity of cyclopenta(c,rf)pyrene and structurally related compounds. Cancer Res., 40, 642-649. 4. Nesnow.S., Leavitt.S., Easterling.R., Watts,R., Toney.S.H., Claxton.L., Sangaiah.R., Toney,G.E., Wiley.J., Fraher.P. and Gold,A. (1984) Mutagenicity of cyclopenta-fused isomers of benz(a)anthracene in bacterial and rodent cells and identification of the major rat liver microsomal metabolites. Cancer Res., 44, 4993-5003. 5. Nesnow.S., Easterling.R.E., Ellis,E., Watts.R. and Ross.J. (1988) Metabolism of benz[/]aceanthrylene and benz[/]aceanthrylene by induced rat liver S9. Cancer Lett., 39, 19-27. 6. Sims,P. (1966) The metabolism of 3-methylcholanthrene and some related compounds by rat-liver homogenates. Biochem. J., 98, 215—228. 7. Fu.P.P., Beland.F.A. and Yang.S.K. (1980) Cyclopenta-polycyclic aromatic hydrocarbons: potent carcinogens and mutagens. Carcinogenesis, 1, 725-727. 8. Jerina.D.M., Lehr.R.E., Yagi,H., Hernandez,O., Dansette.P.M., Wislocki.P.G., Wood.A.W., Chang.R.L., Levin,W. and Conney.A.H. (1976) Mutagenicity of benzo[a]pyrene derivatives and the description of a quantum mechanical model which predicts the ease of carbonium ion formation from diol epoxides. In de Serres.F.J., Fouts,J.R., BendJ.R. and Philpot,R.M. (eds), In Vitro Metabolic Activation in Mutagenesis Testing. Elsevier North Holland Biomedical Press, Amsterdam, pp. 159 — 177. 9. Prasanna.P., Jacobs,M.M. and Yang,S.K. (1987) Selenium inhibition of benzo[a]pyrene, 3-methylcholanthrene, and 3-methylcholanthrylene mutagenicity in Salmonella typhimurium strains TA98 and TA100. Mutat. Res., 190, 101-105. 10. Sims,P. (1967) The carcinogenic activities in mice of compounds related to 3-methylcholanthrene. Int. J. Cancer, 2, 505-508. ll.Cavalieri.E., Roth.R., Althoff.J., Grandjean.C, Patil.K., Marsh,S. and McLaughlin,D. (1978) Carcinogenicity and metabolic profiles of 3-methylcholanthrene oxygenated derivative at the 1 and 2 positions. Chem.-Biol. Interactions, 22, 6 9 - 8 1 . 12.Thakker,D.R., Levin,W., Stoming,T.A., Conney.A.H. and Jerina.D.M. (1978) Metabolism of 3-methylcholanthrene by rat liver microsomes and a highly purified monooxygenase system with and without epoxide hydrase. In Jones.P.W. and Freudenthal,R.I. (eds), Carcinogenesis, Vol. 3: Polynuclear Aromatic Hydrocarbons. Raven Press, New York, pp. 253-264. 13.Stoming,T.A., Bomstein.W. and Bresnick,E. (1977) The metabolism of 3-methylcholanthrene by rat liver microsomes—a reinvestigation. Biochem. Biophys. Res. Commun., 79, 461-469. 14. Mushtaq.M., Weems.H.B. and Yang.S.K. (1986) Metabolic and stereoselective formations of non-K-region benz[a]anthracene 8,9- and 10,11-epoxides. Arch. Biochem. Biophys., 246, 478-487. 15. Chou,M.W. and Yang,S.K. (1979) Combined reversed-phase and normalphase high performance liquid chromatography in the purification and identification of 7,12-dimethylbenz[a]anthracene metabolites. J. Chromaiogr., 185, 635-654. 16. Shou.M. and Yang,S.K. (1990) 9,10-Dihydroxy-9,10-dihydro-3-methylcholanthrene-2-one: a principal metabolite formed in the metabolism of the potent carcinogen 3-methylcholanthrene-2-one by rat liver microsomes. Carcinogenesis, 11, 689—695. 17,Tierney,B., Hewer.A., Rattle,A., Grover.P.L. and Sims.P. (1978) The
1201
