Species Differences in the in Vitro Metabolic ... - ACS Publications

Chem. Res. Toxicol. 2002, 15, 1421-1428

1421

Species Differences in the in Vitro Metabolic Activation of the Hepatotoxic Pyrrolizidine Alkaloid Clivorine Ge lin,*,† Yan-Yan Cui,† Xiao-Quan Liu,†,‡ and Zhen-Tao Wang§ Department of Pharmacology, The Chinese University of Hong Kong, Shatin, Hong Kong, Special Administrative Region, and Department of Pharmacognosy, China Pharmaceutical University, Nanjing, People’s Republic of China Received April 2, 2002

Clivorine is a representative naturally occurring hepatotoxic otonecine-type pyrrolizidine alkaloid. Our previous study has demonstrated that clivorine induces liver damage via metabolic activation to form the reactive pyrrolic ester followed by covalent binding to liver tissue constituents. The present study investigated species differences in the in vitro metabolic activation of clivorine in the male rat and guinea pig of both sexes. In the male rat, the activation of clivorine to form the reactive pyrrolic ester was found as the only metabolic pathway. Moreover, the toxic tissue-bound pyrroles and four isolatable metabolites identified, namely DHR, 7-GSH-DHR, 7,9-diGSH-DHR, and clivoric acid, were all generated from further metabolism of this reactive intermediate. In the case of both sexes of guinea pig, the same activation was observed as the minor biotransformation, while an additional metabolic pathway, a direct hydrolysis of clivorine to form novel clivopic acid was identified as the predominant detoxification pathway. Furthermore, the formation rates for the toxic tissue-bound pyrroles and less toxic DHR were significantly slower and higher, respectively, compared with those in the male rat. In addition, the formation of the reactive pyrrolic ester was mediated by the CYP3A subfamily in both animals, while carboxylesterases might be responsible for the detoxification hydrolysis in guinea pig. The results suggest that the higher metabolic rates for detoxification hydrolyses and the lower formation rate for the toxic tissue-bound pyrroles play the key roles in guinea pig resistance to clivorine intoxication. Therefore, the male rat and guinea pig should be the suitable animal models for further studies of bioactivation and deactivation of otonecine-type PA, respectively.

Introduction The majority of naturally occurring pyrrolizidine alkaloids (PAs)1 are hepatotoxic, and/or genotoxic, and carcinogenic. These natural toxins cause human and livestock poisonings through consumption of PA-containing plants, herbal teas, herbal medicines and contaminated food sources (1-7). Significant species differences in the susceptibility to PA intoxication have been reported in livestock and laboratory animals. For example, in general, cattle, horses, rats, and mice more resemble humans and are highly susceptible to PA intoxication, whereas sheep, rabbits, and guinea pigs are resistant to PA toxicity (2, 5, 7-9). The toxic PAs are generally classified structurally into two types, namely retronecinetype [which may further be divided into two subtypes: retronecine and heliotridine, the 7(S)-isomer of retronecine] and otonecine-type, where characteristically the * To whom correspondence should be addressed. Telephone: (852) 2609-6824. Fax: (852) 2603-5139. E-mail: [email protected]. † The Chinese University of Hong Kong. ‡ Current address: Center of Drug Metabolism and Pharmacokinetics, China Pharmaceutical University, Nanjing 210009, People’s Republic of China. § China Pharmaceutical University. 1 Abbreviations: PA, pyrrolizidine alkaloid; DHR, dehydroretronecine; GSH, glutathione; 7-GSH-DHR, 7-glutathionyldehydroretronecine; 7,9-diGSH-DHR, 7,9-diglutathionyldehydroretronecine; ESI, electrospray ionization; PDA, photodiode-array.

Figure 1. Structures of the necine bases of retronecine, heliotridine, and otonecine.

eight-membered heterocyclic necine base is bicyclic and monocyclic, respectively (2) (Figure 1). The mechanism of hepatotoxicity induced by retronecine-type PAs has been extensively investigated, and it is well established that hepatic metabolic activation of this type of PA to generate the reactive pyrrolic ester plays a key role in causing hepatotoxicity. Once formed, the reactive pyrrolic ester will rapidly bind hepatic tissue constituents to generate tissue-bound pyrroles leading to hepatotoxicity, and thus the tissue-bound pyrroles are directly related to PA-induced hepatotoxicity (1-3, 10-13). In addition, several publications reported that the species difference in susceptibility to retronecine-type PA intoxication is mainly due to variations in the balance between the formation of the toxic pyrrolic metabolites and the detoxification pathways to generate nontoxic N-oxides and/or hydrolyzed metabolites (8, 9, 14-19). However, there is lack of report to the investigation of otonecine-

10.1021/tx0255370 CCC: $22.00 © 2002 American Chemical Society Published on Web 10/10/2002

1422

Chem. Res. Toxicol., Vol. 15, No. 11, 2002

type PA induced hepatotoxicity (20) and the species differences in the susceptibility to this type of PA. Clivorine is a representative naturally occurring otonecine-type PA that is present in various Ligularia species (21-24), including those used as plant sources for traditional Chinese herbal medicines (25, 26). Clivorine is also known to be hepatotoxic and carcinogenic in ACI rats, and mutagenic in Salmonella typhimuriun in the presence of a mammalian microsomal enzyme system suggesting that metabolic activation is critical in the development of toxicity (23, 24, 27). Recently, our research group has investigated the mechanism of clivorine-induced hepatotoxicity in the male SD rat and demonstrated that similar to the retronecine-type PA, the hepatotoxicity induced by clivorine is also due to hepatic metabolic activation and formation of the reactive pyrrolic metabolite, pyrrolic ester dehydroclivorine (28, 29). In the present study, we aimed to investigate species differences in the in vitro metabolic activation of clivorine in the male rat and guinea pig of both sexes, to delineate the mechanism underlying these species differences and to identify suitable animal models for the study of otonecine-type PA intoxication and detoxification in human.

Materials and Methods Materials. Monocrotaline, retrorsine, ketoconazole, orphenadrine, and all other chemicals and solvents were purchased from Sigma Chemical Co. (St. Louis, MO). Clivorine was isolated from Ligularia hodgsonii Hook by a standard alkaloid extraction procedure for PAs (26). The identity of the isolated clivorine was confirmed by UV, IR, NMR, and MS analyses, and its purity was determined to be higher than 99% by TLC, HPLC, and NMR spectroscopy. Instrumentation. 1H and 13C NMR spectra in either D2O or CD3OD were recorded on a Bruker ARX-500 spectrometer using the peak of TMS as reference. Both positive and negative ion ESI-MS analyses with direct loop injection were carried out in the Finnigan TSQ 7000 triple-stage quadruple mass spectrometer. The on-line HPLC/MS analyses for the identification of metabolites generated from microsomal incubations of clivorine were performed on a Hewlett-Packard (HP) 1100 Liquid Chromatograph connected to a Finnigan TSQ 7000 mass spectrometer coupled with an electrospray ionization (ESI) interface. The HP 1100 chromatographic system coupled with a photodiode-array (PDA) multiple wavelength UV detector set at 230 mn was utilized for all HPLC quantitative analyses. Preparation of Authentic Samples for the Putative Metabolites. Dehydroretronecine (DHR), 7-glutathionyldehydroretronecine (7-GSH-DHR), and 7,9-diglutathionyldehydroretronecine (7,9-diGSH-DHR) were all synthesized using monocrotaline as the starting material. The detailed synthetic steps and confirmation of their identities have been reported previously (29). Clivoric acid was isolated from a scaled-up rat microsomal incubation with clivorine as described previously (29). For the novel metabolite, clivopic acid, a scaled-up male guinea pig microsomal incubation with clivorine in the absence of the NADPH-generating system as described below was conducted to generate enough quantity of sample for isolation. The isolated and purified clivopic acid was subjected to 1H and 13C NMR and MS analyses for the confirmation of its identity. Animals and Preparation of Liver Microsome. Adult male Sprague-Dawley rats (body weight 200-250 g), and Dunkin-Hartley guinea pigs of both sexes (body weight 280300 g) supplied by the Laboratory Animal Services Centre at the Chinese University of Hong Kong were utilized in the present study. Microsomes were prepared from either rat or guinea pig by a standard procedure (28, 29, 33), and protein content was determined using a modified procedure of the Lowry

Lin et al. method (34). Cytochrome P450 content was determined by a published standard procedure (35). The prepared microsomes were stored at -80 °C until use. In addition, whole liver homogenates and cytosolic fractions of the male guinea pig were also prepared by a standard procedure (33) and utilized in the present study. Microsomal Incubation and Treatment of Incubated Samples. Our previously reported procedures were used (28, 29). Briefly, typical microsomal incubation mixtures (10.0 mL) included potassium phosphate buffer (100 mM, pH 7.4) containing liver microsomes (2 mg protein/mL), 0.25 mM clivorine, 1 mM NADH, 4 mM MgCl2, 1 mM NADP, 10 mM glucose 6-phosphate, 1.0 unit/mL glucose 6-phosphate dehydrogenase, and 2.0 mM GSH (the reduced form of glutathione). The addition of GSH into the incubation system enabled this nucleophilic agent to trap the reactive pyrrolic ester forming the corresponding GSH conjugates. The male rat microsomal metabolism of clivorine in the absence of GSH has been previously reported by our research group (29). Incubation was performed at 37 °C for 60 min, and terminated by chilling in an ice bath. Furthermore, incubations under similar conditions but using the same volume (1 mL) of either male guinea pig whole liver homogenates or male guinea pig cytosolic fractions or male guinea pig microsomes in the absence of the NADPH-generating system were also performed. In addition, various controls including using denatured microsomes and incubation in the absence of substrate were conducted. In the enzyme inhibition study, ketoconazole (5 µM) or orphenadrine (30 µM) was preincubated for 10 min in the microsomal system prior to the addition of clivorine. Furthermore, since activities of the metabolizing enzymes may be altered by preincubation, incubation samples without the addition of inhibitor were treated identically to enable a direct comparison with the results obtained from the enzyme inhibition experiments. The resultant incubates were centrifuged at 105000g at 2 °C for 20 min. Aliquots (200 µL) of the supernatant were directly subjected to HPLC/MS for qualitative studies and HPLC/PDA for quantitative determinations. In addition, the resultant pellets from all different incubations were utilized for the determination of the total hepatic tissue-bound pyrroles as previously reported (29, 36). Incubation for each individual condition was conducted in triplicate. Identification of Metabolites. The supernatants of incubates were directly subjected to HPLC/MS analysis for the identification of metabolites. The analytes were eluted on a Hamilton PRP-1 reversed-phase HPLC column (150 × 4.0 mm, 5 µm) with the mobile phase consisting of (A) 1% acetic acid and (B) acetonitrile, using the following gradient elution: initial 0-5 min, 100% A; 5-35 min, linear change to 75% A, and maintained for 5 min. The flow rate was kept constant at 0.8 mL/min. ESI-MS was performed under the operating conditions of 5 kV of spray voltage, sheath gas setting at 50 psi, 15 units of auxiliary gas, and a heated capillary temperature at 250 °C. Full scan mass spectra were obtained over the mass range of m/z 150-850 in a negative ion mode for the identification of 7-GSH-DHR, 7,9-diGSH-DHR, clivoric acid, and clivopic acid, or in a positive ion mode for the identification of DHR and intact clivorine. A direct comparison of the retention time and mass spectrometric data of each metabolite with that for the authentic sample was also conducted. In the case of the identification of the novel metabolite, clivopic acid, the supernatant obtained from a scaled-up male guinea pig microsomal incubation (100 mL) with 2.0 mM clivorine in the absence of an NADPH-generating system was adjusted to pH 3.0 by 0.1 M phosphate acid and extracted three times with ethyl acetate. The combined organic extracts were dehydrated by Na2SO4 and evaporated to dryness under reduced pressure. The residues were reconstituted into 50% aqueous acetonitrile, and then separated by preparative HPLC using an APEX C18 column (250 × 10 mm, 5 µm) with a mobile phase of acetonitrile: 0.2% ammonia formate (2:8) at a flow rate of 2 mL/

Species Differences in Metabolism of Clivorine

Chem. Res. Toxicol., Vol. 15, No. 11, 2002 1423

min. The collected acidic fractions were further purified by a HPLC system coupled with a Radial-Pak Cartridge C18 (150 × 6 mm, 4 µm). The cartridge was eluted with d.d. water for 15 min to remove ammonia formate, followed by 20% aqueous acetonitrile for 15 min to elute the analyte. The aqueous acetonitrile fractions containing clivopic acid were collected and dried under reduced pressure followed by lyophilization to give a white powder. The obtained pure clivopic acid was characterized by NMR and MS with a direct loop injection. Quantification of Metabolites. The HPLC assay recently developed in our research laboratory (37) using a specific twocolumn setup with a Hamilton PRP-1 analytical column (250 × 4.6 mm, 5 µm) and Hamilton PRP-1 guard column (50 × 4.1 mm, 5 µm) was adopted for the quantitative studies of each individual incubation in triplicate. All peak responses were measured at 230 nm by a PDA detector. The incubated blank samples spiked with standards were also analyzed under identical conditions, and a direct comparison of the retention time of each analyte with that of the corresponding standard was also conducted for each analyte. Samples for the construction of calibration curves for clivorine, four metabolites, namely DHR, 7-GSH-DHR, 7,9-diGSH-DHR, and clivoric acid, were prepared as described previously (37). In the case of clivopic acid, similarly the purified authentic sample was spiked into the blank microsomal incubation mixtures at five different concentrations over a range of 6.25 to 150.0 µg/mL. All calibration curves with the concentration of analyte as a function of peak area ratio (analyte/internal standard) using retrorsine as an internal standard were constructed using a previously reported procedure (37). The quantity of each analyte present in different incubates was determined from the corresponding calibration curve. Data Analysis. Results are expressed as means ( SD throughout the text. Analysis of variance (ANOVA) with the Bonferroni test was used to compare results in different groups. A probability (p) value less than 0.05 was considered statistically significant.

Figure 2. The representative chromatograms of HPLC/PDA analysis of control male rat microsomal incubate spiked with authentic samples (A), male rat microsomal incubate (B), and male guinea pig microsomal incubate (C). Peak 1, clivorine; peak 2, DHR; peak 3, 7-GSH-DHR; peak 4, 7,9-diGSH-DHR; peak 5, clivoric acid; peak 6, clivopic acid; IS, internal standard retrorsine.

Results

ion at m/z 151 corresponding to a loss of a formic acid from the molecule. Therefore, its molecular weight was unequivocally confirmed to be 196 Da. Compared with previously published 1H NMR data for clivoric acid (29) (Table 1), the chemical shift of H-4 in the new metabolite significantly shifted downfield (δ 6.62 ppm), indicating an R,β-unsaturated six-membered lactone structure. The other proton signals compared well with comparable signals assigned for clivoric acid (Table 1). Therefore, based on the spectroscopic data the structure of this novel metabolite with an empirical formula of C10H12O4 was unequivocally identified as 6-carboxy-5,6-dihydro-5,6,dimethyl-3-ethenyl-2H-pyran-2-one. This in vitro guinea pig metabolite of clivorine was named as clivopic acid (Figure 3). However, its absolute structural configurations were not determined. The tissue-bound pyrroles were also determined after microsomal incubations of clivorine in both rat and guinea pig. The in vitro microsomal metabolic pathways of clivorine in male rat and guinea pig of both sexes were delineated as shown in Figure 3. Quantification of Metabolites. Using our previously developed HPLC assay (37), the amounts of intact clivorine and the four metabolites formed in various incubations in male rat microsomes were determined. After 1 h incubation about 49% of clivorine was metabolized. As illustrated in Figure 4, clivoric acid was the major metabolite with a formation rate of 40.61 nmol/ mg/h; the formation rates for the others were 8.91 (DHR), 7.43 (7-GSH-DHR), and 1.82 (7,9-diGSH-DHR) nmol/mg/

Identification of Microsomal Metabolites. The supernatants of different incubates were directly analyzed using a previously developed HPLC/PDA assay (37). Representative HPLC chromatograms of the microsomal incubated samples are shown in Figure 2. Similar to our previous findings (29), by means of comparison of the chromatographic patterns of authentic standards (Figure 2A), four known metabolites of clivorine were found in male rat microsomal incubations (Figure 2B). In the case of male guinea pig microsomal metabolism of clivorine, in addition to these four known metabolites, a new peak (tR 27.1 min) was observed, whereas the intact clivorine was not detected in the incubated samples (Figure 2C). A similar HPLC chromatographic pattern was also observed in the female guinea pig (data not shown). The identities of four known metabolites were further confirmed by HPLC/MS analysis with direct comparisons of the retention time and mass spectrometric data of the corresponding authentic samples. These metabolites were identified as DHR, 7-GSH-DHR, 7,9-diGSH-DHR, and clivoric acid, respectively (Figure 3). For the characterization of the newly found metabolite, a pure sample was isolated from a large scale incubation with male guinea pig microsomes as described in the Materials and Methods. The isolated sample was analyzed by negative ion mode ESI-MS with a direct loop injection. The MS spectrum exhibited the pseudo-molecular ion at m/z 195 ([M - 1]-) as the base peak and an

1424


Lin et al.

Figure 3. The in vitro microsomal metabolic pathways of clivorine in the male rat (R) and guinea pig of both sexes (GP). NuS: nucleophilic biological macromolecules. GS: glutathionyl. The unstable intermediates are shown in the large brackets. A racemic mixture with 1:1 ratio of 7(S)- and 7(R)-enantiomers was identified for both 7-GSH-DHR and 7,9-diGSH-DHR. Table 1. 1H NMR Data for Clivopic Acid and Clivoric Acid position

clivopic acid (CD3OD)

position

4 5 8 9 10 11

6.62 (1H, d, J ) 3.9 Hz) 2.85 (1H, m) 1.49 (3H, s) 1.19 (3H, d, J ) 7.5 Hz) 6.35 (1H, dd, J ) 17.7, 11.1 Hz) 5.14 (1H, d, J ) 11.1 Hz, cis) 5.74 (1H, d, J ) 17.7 Hz, trans)

4 3 8 9 6 7 12

a

clivoric acid (D2O)a 5.63 (1H, d) 2.80 (1H, m) 1.56 (3H, s) 1.08 (3H, d) 6.35 (1H, dd) 5.18 (1H, d) 5.19 (1H, d) 2.10 (3H, s)

Previously published data (29).

h, respectively. In addition, the total amount of tissuebound pyrroles determined per incubate was 1.3 nmol, which accounts for about 0.5% of the incubated clivorine. In the cases of the enzyme inhibition experiment, compared with the data obtained for the controls with a similar 10 min preincubation prior to the addition of substrate (Table 2), the metabolic rate of clivorine (23 vs 39%, p < 0.001) and the formation of tissue-bound pyrroles (0.02 ( 0.02 vs 0.7 ( 0.07 nmol, p < 0.001) were significantly inhibited by a selective cytochrome P450 3A inhibitor ketoconazole. However, preincubation with a selective cytochrome P450 2B inhibitor orphenadrine did

not markedly affect the in vitro microsomal metabolism of clivorine in the male rat. The results for the quantification of the metabolites of clivorine in guinea pig are also illustrated in Figure 4. In both male and female guinea pig microsomal incubations, clivorine was completely metabolized after 1 h. Clivopic acid was found as the predominant metabolite accounting for about 60% (male) and 48% (female) of the incubated clivorine. The formation rates for the other metabolites were similar between two sexes, except that the formation rates for DHR and clivoric acid were higher in the male guinea pig. In the male guinea pig microsomal incubation, the formation rates of the other metabolites determined were 43.3, 21.1, 6.8, and 5.0 nmol/mg/h for clivoric acid, DHR, 7-GSH-DHR, and 7,9-diGSH-DHR, respectively. Moreover, the total amount of tissue-bound pyrroles was 0.3 nmol/incubate, which only accounts for about 0.1% of the incubated clivorine. In the enzyme inhibition study in the male guinea pig, ketoconazole did not affect the metabolic rate of clivorine (Table 2). However, it abolished the formations of DHR, 7-GSH-DHR, 7,9-diGDH-DHR, and tissue-bound pyrroles and significantly inhibited the formation of clivoric acid (4.5 ( 0.3 vs 50.5 ( 1.4 nmol/mg/h, p < 0.001). Whereas,



Table 2. Effect of Enzyme Inhibitors on the in Vitro Metabolism of Clivorine in the Male Rat and Male Guinea Pig

sample rat controla ketoconazole orphenadrine guinea pig controla ketoconazole orphenadrine a

DHR 4.8 ( 0.1 4.7 ( 0.04 9.9 ( 0.1 10.6 ( 0.5

formation rate of metabolite (nmol/mg of microsomal protein/1 h) 7-GSH-DHR 7,9-diGSH-DHR clivoric acid

clivopic acid

3.8 ( 0.2 2.6 ( 0.04

2.3 ( 0.1 1.9 ( 0.1

21.9 ( 0.2 17.3 ( 0.2

-b -

8.4 ( 0.2 6.2 ( 0.1c

5.8 ( 0.2 7.7 ( 0.2c

50.5 ( 1.4 4.5 ( 0.3c 39.2 ( 0.6c

70.2 ( 1.7 91.7 ( 5.9c 55.4 ( 0.6c

intact clivorine (nmol/mL)

total tissue-bound pyrroles (nmol)

152 ( 5 193 ( 10c 162 ( 0.3

0.7 ( 0.1 0.02 ( 0.02c 0.5 ( 0.2

-

0.6 ( 0.2 0.7 ( 0.1

Preincubated for 10 min prior to the addition of substrate. b (-) Not found. c p < 0.001 compared with the corresponding control. Table 3. Formation of Clivopic Acid in the Male Guinea Pig

sample

content of clivopic acid (nmol/mL)

percentage of the incubated clivorine (%)

whole live homogenate microsome cytosolic fraction

36.3 ( 21.2 14.8 ( 1.5 220.9 ( 21.8a

14.0 ( 8.3 5.8 ( 0.6 86.6 ( 8.6a

a

p < 0.001 compared with microsomal incubation.

metabolite, where it accounted for about 6, 14, and 87% of the incubated clivorine, respectively (Table 3).

Discussion

Figure 4. Quantification of the in vitro metabolites of clivorine in rat and guinea pig. (***) p < 0.001 compared with male rat. (###) p < 0.001 compared with male guinea pig.

the rate for producing clivopic acid was significantly accelerated (91.7 ( 5.9 vs 70.2 ( 1.7 nmol/mg/h, p < 0.001). In the case of orphenadrine inhibition, the metabolic rate of clivorine was not significantly affected. However, the formation rates of clivopic acid, clivoric acid, and 7-GSH-DHR were markedly inhibited, while that for 7,9-diGSH-DHR was significantly increased. In addition, incubations were also conducted for male guinea pig liver microsomes, whole liver homogenates, and cytosolic fractions in the absence of the NADPH-generating system. In all cases, clivopic acid was found as the only

Our recent studies on clivorine induced hepatotoxicity have proven that the in vitro metabolic activation of this otonecine-type PA is similar between human and male rat but not guinea pig. Moreover, rat was found to be much more susceptible to clivorine intoxication than guinea pig in a toxicology study in vivo. These detailed studies on in vitro human metabolic activation and in vivo toxicities in laboratory animals will be reported separately. The aims of the present study were to delineate and compare in vitro metabolic activation of clivorine in the male rat and guinea pig of both sexes and to identify which animal model is more suitable for the investigation of the mechanism underlying clivorine induced toxicity in man. Various studies on different retronecine-type PAs with male rat have demonstrated that the two major biotransformations involve the oxidative dehydrogenation to produce the reactive pyrrolic ester and N-oxidation to generate the corresponding PA N-oxide, and that the latter pathway is considered as a detoxification pathway (1, 2). In the case of clivorine, an otonecine-type PA, the reactive pyrrolic ester results from oxidative N-demethylation (28, 29), while N-oxidation was not found in its microsomal metabolism. The microsomal metabolic pathways of clivorine in the male rat found in the present study were consistent to that reported previously (29). Similar to the retronecine-type PA (1-3, 10-13), the formation of the reactive pyrrolic ester, dehydroclivorine, plays a key role in otonecine-type PA induced hepatotoxicity. As illustrated in Figure 3, the formation of this reactive pyrrolic ester was identified to be the only metabolic pathway of clivorine in the male rat. This reactive metabolite is unstable and cannot be directly determined (28, 29). Once formed, it either rapidly binds nucleophilic macromolecules to generate tissue-bound pyrroles leading to hepatotoxicity, or reacts with water or other soluble nucleophiles such as GSH to form DHR or glutathione conjugates, respectively. The two GSH

1426


conjugates are hydrophilic and may rapidly escape from liver into other body compartments and then readily excreted, thus they are generally considered to be nontoxic metabolites (36, 38, 39). The pyrrolic alcohol DHR can also alkylate nucleophiles in hepatocytes, but in a slower and probably more selective manner than the reactive pyrrolic ester, thus, DHR is generally recognized as less hepatotoxic (1, 2, 40). However, DHR is more stable and hydrophilic than the reactive pyrrolic ester and may be widely distributed through out the body to induce extrahepatic toxicity (41-43). In general the tissue-bound pyrroles are considered to be adducts related to PA intoxication (1-3, 5-7, 10, 24, 40). In the present study, all the pyrrolic metabolites were formed from the necine base moiety with concurrent generation of clivoric acid. Therefore, all the in vitro metabolites found in male rat resulted from the reactive pyrrolic ester. In the case of guinea pig, similar metabolic patterns were observed for both sexes. In addition to all the known reactive pyrrolic ester related metabolites found in male rat, a novel metabolite, clivopic acid, was identified for the first time. The formation of clivopic acid might arise from the necic acid moiety via direct hydrolysis of the three ester bonds of clivorine, followed by simultaneous intramolecular esterification to give a stable six-membered ring product (Figure 3). However, whether these three hydrolyses occurred one by one or simultaneously is unknown. On the other hand, the corresponding necine base, which should be produced concurrently during the formation of the acid metabolite, was not detected. Although appropriate configurational spectroscopic analyses of clivopic acid were not conducted in the present study, the two chiral centers are most likely R for C-5 and S for C-6 (Figure 3), since the configurations of the corresponding acidic moiety in clivorine would unlikely change during the hydrolyses and ring closure. Clivopic acid is hydrophilic and hence should be a readily excreted and nontoxic metabolite. Between two sexes, although overall metabolic patterns were quite similar, there were quantitative differences. Subsequent comparisons of the microsomal metabolism of clivorine between rat and guinea pig were only made for male animals. The quantitative studies indicate that about half of the clivorine was metabolized after 1 h incubation with male rat microsomes. The amount of the most abundant metabolite, clivoric acid, was determined to be about 33% of the incubated clivorine, which can be considered as the sum of all the other detectable metabolites, since they all likely simultaneously form with this acid metabolite. In the male guinea pig, clivorine was much more extensively metabolized, since no substrate could be detected after 1 h incubation at 0.25 and 1.25 mM concentration (data not shown in the latter case). The recovery of all metabolites formed measured by the sum of clivopic acid (60%) and clivoric acid (35%) was about 95%. As showed in Figure 4, similar formation rates of clivoric acid were determined, indicating that the rates of the key activation step to generate the reactive pyrrolic ester were about the same in both animals. However, the amount of the toxic tissue-bound pyrroles generated in guinea pig (0.3 ( 0.2 nmol) was significantly lower than that in rat (1.3 ( 0.3 nmol, p < 0.001), which demonstrated that the formation rate of the key toxic metabolite was significantly lower in guinea pig. On the other hand, a significantly greater amount of less toxic DHR was determined

Lin et al.

in guinea pig (21.4 ( 1.3 vs 8.9 ( 0.6 nmol/mg/h, p < 0.001), which suggests that hydrolysis of the reactive pyrrolic ester was significantly faster in guinea pig. Therefore, all these results suggested that the less susceptibility of guinea pig to clivorine intoxication is most likely due to the significantly faster metabolic rates of two detoxification hydrolysis pathways to form nontoxic clivopic acid and less toxic DHR, respectively, and the significantly slower formation rate for the toxic tissue-bound pyrroles. In the enzyme inhibition study (Table 2), ketoconazole, a selective cytochrome P450 3A inhibitor (44, 45), almost completely inhibited the activation of clivorine to generate the reactive pyrrolic ester in both animals. Furthermore, inhibition of the cytochrome P450 3A subfamily abolished the rat microsomal metabolism of clivorine, whereas, the metabolic rate of clivorine in guinea pig was not affected, where clivorine was completely metabolized to form only one metabolite, clivopic acid. These results demonstrated that the cytochrome P450 3A subfamily were the primary and key enzymes responsible for the activation of clivorine in both rat and guinea pig. In the case of preincubation of orphenadrine, a known rat cytochrome P450 B inhibitor (46), the overall metabolic pattern of clivorine in rat was not affected (Table 2), suggesting that the cytochrome P450 2B subfamily was not involved in the metabolism of clivorine in rat. In the case of guinea pig microsomal metabolism, although the overall metabolic pattern was not significantly changed, the formation rates of the major metabolites such as clivopic acid and clivoric acid were dramatically inhibited (Table 2). The cytochrome P450 2B subfamily was proposed to play a role in guinea pig microsomal metabolism of retronecine-type PA senecionine by using purified enzymes (47). It should be noted that there is lack of reports to the inhibitory specificity of orphenadrine toward the cytochrome P450 2B subfamily in guinea pig, although this inhibitor was proven to inhibit cytochrome P450 2B isoforms in both rat and human (46, 48). The definitive role of the cytochrome P450 2B subfamily on the metabolism of clivorine in guinea pig needs to be clarified. In the present study, only a trace amount of clivopic acid was obtained by a direct chemical hydrolysis of clivorine under either acidic or basic condition (data not shown) suggesting that this acid was generated enzymatically. When clivorine was incubated with guinea pig microsomes in the absence of the NADPH-generating system, clivopic acid was found as the only metabolite, suggesting that the formation of this acid metabolite might be catalyzed by the NADPH independent microsomal esterases. The direct hydrolysis of a retronecine-type PA, monocrotaline, to generate the corresponding necic acid metabolite in guinea pig microsomal metabolism has been reported (19). In this well-designed study, the direct hydrolysis was demonstrated to be mediated by microsomal carboxylesterase since the formation of the respective necic acid was significantly inhibited by triorthocresyl phosphate, a selective carboxylesterase inhibitor. However, this inhibitor is no longer commercially available. A similar inhibition study was not performed in the present study. Further studies were performed in male guinea pig whole liver homogenate, microsomes, and cytosolic fraction in the absence of a NADPH-generating system. As summarized in Table 3, the highest amount of clivopic


acid was found in the cytosolic incubation while the lowest amount was determined in the microsomal incubation. Thus, it is most likely that the generation of clivopic acid was mainly catalyzed by the soluble esterases, since the amount/activity of such enzymes should be present in the highest in the cytosolic fraction and the lowest in the microsome fraction when the same volume of biological samples is utilized for each incubation. Previous study on the microsomal carboxylesterases in several animal species to catalytic activity toward a variety of esters and amides (49) has shown that in general guinea pig displayed higher catalytic activity than rat toward many of the substrates. These reported differences in the catalytic activities of microsomal carboxylesterases may help to explain why the direct hydrolysis of clivorine catalyzed by esterases in the guinea pig microsomes was predominant whereas the same pathway was not observed in the rat. In conclusion, cytochrome P450 3A subfamily mediated metabolic activation of clivorine to generate the reactive pyrrolic ester was the only pathway found for the direct metabolism of clivorine in the male rat. Whereas, an additional direct hydrolysis of clivorine catalyzed by microsomal carboxylesterases was identified in guinea pig of both sexes. Therefore, our results demonstrated that the higher metabolic rates for both direct hydrolysis of clivorine to generate nontoxic clivopic acid and hydrolysis of the reactive pyrrolic ester to produce less toxic DHR, in combination with a lower formation rate for the toxic tissue-bound pyrroles play the key roles in guinea pig resistance to clivorine intoxication. While, the activation of clivorine as the only direct metabolic pathway in the male rat to generate the reactive pyrrolic ester followed by a higher formation rate for the toxic tissuebound pyrroles in the liver mainly contribute to the high susceptibility of the male rat to clivorine hepatotoxicity. Thus, the male rat should be a suitable animal model for further studies of otonecine-type PA intoxication while guinea pig should be considered as a good animal model for the investigation of detoxification of this type of PA in man.

Acknowledgment. The authors thank Mr. Tin-Yan Cheng (Department of Pharmacology, the Chinese University of Hong Kong) for the quantification of the tissuebound pyrroles and Prof. Edward M. Hawes (Drug Metabolism and Drug Disposition Group, University of Saskatchewan, Saskatoon, Canada) for critical review of this manuscript. This research project was supported by the Research Grant Council of Hong Kong (Earmarked Research Grant CUHK 415/95M).

References (1) Mattocks, A. R. (1968) Toxicity of pyrrolizidine alkaloids. Nature 217, 723-728. (2) Mattocks, A. R. (1986) Chemistry and Toxicology of Pyrrolizidine Alkaloids, Academic Press, New York. (3) Huxtable, R. J. (1989) Human health implications of pyrrolizidine alkaloids and herbs containing them. In Toxicants of Plant Origin (Cheeke, P. R., Ed.) Vol. 1, pp 41-86, CRC Press Inc., Boca Raton, Florida. (4) Huxtable, R. J. (1980) Herbal teas and toxins: Novel aspects of pyrrolizidine poisoning in the United States. Perspect. Biol. Med. 24, 1-14. (5) McLean, E. K. (1970) The toxic actions of pyrrolizidine (senecio) alkaloids. Pharmacol. Rev. 22, 429-483. (6) White, I. N. H., Mattocks, A. R., and Butler, W. H. (1973) The conversion of the pyrrolizidine alkaloid retrorsine to pyrrolic


(7)

(8) (9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

derivatives in vivo and in vitro and its acute toxicity to various animal species. Chem.-Biol. Interact. 6, 207-218. Stegelmeier, B. L., Edgar, J. A., Colegate, S. M., Gardner, D. R., Schoch, T. K., Coulombe, R. A., and Molyneux, R. J. (1999) Pyrrolizidine alkaloid plants, metabolism and toxicity. J. Nat. Toxins 8, 95-116. Cheeke, P. R. (1998) Natural toxicants in feeds, forages and poisonous plants, Interstate, Danville. Huan, J.-Y., Miranda, C. L., Buhler, D. R., and Cheeke, P. R. (1998) The roles of CYP3A and CYP2B isoforms in hepatic bioactivation and detoxification of the pyrrolizidine alkaloid senecionine in sheep and hamsters. Toxicol. Appl. Pharmacol. 151, 229-235. Castagnoli, N., Jr., Rimoldi, J. M., Bloomquist, J., and Castagnoli, K. P., (1997) Potential metabolic bioactivation pathways involving cyclic tertiary amines and azaarenes. Chem. Res. Toxicol. 10, 924-940. Hinson, J. A., Pumford, N. R., and Nelson, S. D. (1994) The role of metabolic activation in drug toxicity. Drug Metab. Rev. 26, 395412. Mattocks, A. R., and White, I. N. H. (1971) The conversion of pyrrolizidine alkaloids to N-oxides and to dihydropyrrolizidine derivatives by rat-liver microsomes in vitro. Chem.-Biol. Interact. 3, 383-396. White, I. N. H., and Mattocks, A. R. (1971) Factors affecting the conversion of pyrrolizidine alkaloids to N-oxides and to pyrrolic derivatives in vitro. Xenobiotica 1, 503-505. Chung, W.-G., and Buhler, D. R. (1995) Major factors for the susceptibility of guinea pig to the pyrrolizidine alkaloid jacobine. Drug Metab. Dispos. 23, 1263-1267. Chu, P. S., and Segall, H. J. (1991) Species difference in the urinary excretion of isatinecic acid from the pyrrolizidine alkaloid retrorsine. Comp. Biochem. Physiol. 100, 683-686. Lame´, M. W., Jones, A. D., Morin D., and Segall, H. J. (1991) Metabolism of [14C]monocrotaline by isolated perfused rat liver. Drug Metab. Dispos. 19, 516-524. Reid, M. J., Lame´, M. W., Morin, D., Wilson, D. W., and Segall, H. J. (1997) Mopnocrotaline metabolism and distribution in Fisher 344 and Sprague-Dawley rats. Comp. Biochem. Physiol., Part B: Biochem. Mol. Biol. 117, 115-123. Miranda, C. L., Chung, W., Reed, R. E., Zhao, X., Henderson, M. C., Wang, J. L., Willams, D. E., and Buhler, D. R (1991) Flavincontaining monooxygenase: a major detoxifying enzyme for the pyrrolizidine alkaloid senecionine in guinea pig tissues. Biochem. Biophys. Res. Commun. 178, 546-552. Dueker, S. R., Lame´, M. W., Morin, D., Wilson, D. W., and Segall, H. J. (1992) Guinea pig and rat hepatic microsomal metabolism of monocrotaline. Drug Metab. Dispos. 20, 275-280. Culvenor, C. C. J., Edgar, J. A., Smith, L. W., Jago, M. V., and Peterson, J. E. (1971) Active metabolites in the chronic hepatotoxicity of pyrrolizidine alkaloids, including otonecine esters. Nat. New Biol. 229, 255-256. Birnbaum, K. B., Kla´sek, A., Sedmera, P., Snatzke, G., Johnson, L. F., and Santavy, F. (1971) Revised structure of the alkaloid clivorine. Tetrahedron Lett. 3421-3424. Kla´sek, A., Vrublovsky, P., and Santavy, F. (1967) Isolation of pyrrolizidine alkaloids from the plants Senecio vivularis D. C. and Ligularia clivorum Maxim. Collect. Czech. Chem. Commun. 32, 2512-2522. Kuhara, K., Takanashi, H., Hirono, I., Furuya, T., and Asada, Y. (1980) Carcinogenic activity of clivorine, a pyrrolizidine alkaloid isolated from Ligularia dentata. Cancer Lett. 10, 117-122. Mori, H., Sugie, S., Yoshimi, N., Asada, Y., Furuya, T., and Williams, G. M. (1985) Genotoxicity of a variety of pyrrolizidine alkaloids in the hepatocyte primary culture-DNA repair test using rat, mouse, and hamster hepatocytes. Cancer Res. 45, 31253129. Lin, G., Zhou, K.-Y., Zhao, X.-G., Wang, Z-.T., and But, P. P. H. (1998) Determination of hepatotoxic pyrrolizidine alkaloids by online high performance liquid chromatography mass spectrometry with an electrospray interface. Rapid Commun. Mass Spectrom. 12, 1445-1456. Lin, G., Rose, P., Chatson, K., B., Hawes, E. M., Zhao, X.-G., and Wang, Z.-T. (2000) Characterization of two structural forms of otonecine-type pyrrolizidine alkaloids from Ligularia hodgsonii by NMR spectroscopy. J. Nat. Prod. 63, 857-860. Yamaka, H., Nagao, M., Sugimura, T., Furuya, T., Shiral, A., and Matsushima, T. (1979) Mutagenicity of pyrrolizidine alkaloids inthe Salmonella/mammalian-microsome test. Mutat. Res. 68, 211-216.

1428


(28) Lin, G., Cui, Y. Y., and Hawes, E. M. (1998) Microsomal formation of a pyrrolic alcohol glutathione conjugate of clivorine, firm evidence for the formation of a pyrrolic metabolite of an otonecinetype pyrrolizidine alkaloid. Drug Metab. Disps. 26, 181-184. (29) Lin, G., Cui, Y. Y., and Hawes, E. M. (2000) Characterization of rat liver microsomal metabolism of clivorine, an hepatotoxic otonecine-type pyrrolizidine alkaloid. Drug Metab. Disps. 28, 1475-1483. (30) Mattocks, A. R. (1982) Hydrolysis and hepatotoxicity of retrornecine diesters. Toxicol. Lett. 14, 111-116. (31) Mattocks, A. R., Croswell, S., Jukes, R., and Huxtable, R. J. (1991) Identity of a biliary metabolite formed from monocrotaline in isolated, perfused rat liver. Toxicon 29, 409-415. (32) Mattocks, A. R., Jukes, R., and Brownm J. (1989) Simple procedures for preparing putative toxic metabolites of pyrrolizidine alkaloids. Toxicon 27, 561-567. (33) Williams, D. E., Reed, R. L., Kedzierski, B., Dannan, G. A., Guengerich, F. P., and Buhler, D. R. (1989) Bioactivation and detoxication of the pyrrolizidine alkaloid senecionine by cytochrome P-450 enzymes in rat liver. Drug Metab. Dipos. 17, 387392. (34) Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. (35) Omura, T., and Sato, R. (1964) The carbon monoxide-binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J. Biol. Chem. 239, 370-378. (36) Yan, C. C., and Huxtable, R. J. (1995) Relationship between glutathione concentration and metabolism of pyrrolizidine alkaloid, monocrotaline, in the isolated, perfused liver. Toxicol. Appl. Pharmacol. 130, 132-139. (37) Cui, Y. Y., and Lin, G. (2000) Simultaneous analysis of clivorine and its four microsomal metabolites by high-performance liquid chromatography. J. Chromatogr. A 903, 85-92. (38) White, I. N. H. (1976) The role of liver glutathione in the acute toxicity of retrorsine to rats. Chem.-Biol. Interact. 13, 333-342. (39) Yan, C. C., and Huxtable, R. J. (1995) The effect of pyrrolizidine alkaloids, monocrotaline and trichodesmine, on tissue pyrrole

Lin et al.

(40) (41) (42)

(43) (44)

(45) (46)

(47)

(48)

(49)

binding and glutathione metabolism in the rat. Toxicon 33, 627634. Mattocks, A. R., and White, I. N. H. (1973) Toxic effects and pyrrolic metabolites in the liver of young rats given the pyrrolizidine alkaloid retrorsine. Chem.-Biol. Interact. 6, 297-306. Allen, J. R., Hsu, I. C., and Carstens, L. A. (1975) Dehydroretronecine-induced rhabdomyosarcomas in rats. Cancer Res. 35, 997-1002. Shumaker, R. C., Robertson, K. A., Hsu, I. C., and Allen, J. R. (1976) Neoplastic transformation in tissues of rats exposed to monocrotaline or dehydroretronecine. J. Natl. Cancer Inst. 56, 787-790. Johnson, W. D., Robertson, K. A., Pounds, J. G., and Allen, J. R. (1978) Dehydroretronecine-induced skin tunors in mice. J. Natl. Cancer Inst. 61, 85-89. Ghosal, A., Satoh, H., Thomas, P. E., Bush, E., and Moore, D. (1996) Inhibition and kinetics of cytochrome P4503A activity in microsomes from rat, human, and cDNA-expressed human cytochrome P450. Drug Metab. Dispos. 24, 940-947. Ervine, C. M., and Matthew, D. E. (1996) Comparison of ketoconazole and fluconazole an Cytochrome P450 inhibitor. Drug Metab. Dispos. 24, 211-215. Reidy, G. F., Mehta, I., and Murray, M. (1989) Inhibition of oxidative drug metabolism by orphenadrine; in vitro and in vivo evidence for isoforms-specific complexation of cytochrome P450 and inhibition kinetics. Mol. Pharmacol. 35, 736-743. Chung, W.-G., Miranda, C. L., and Buhler, D. R. (1995) A cytochrome P4502B form is the major bioactivation enzyme for the pyrrolizidine alkaloid senecionine in guinea pig. Xenobiotica 25, 929-939. Guo, Z., Raeissi, S., White, R. B., and Stevens, J. C. (1997) Orphenadrine and methimazole inhibit multiple cytochrome P450 enzymes in human liver microsomes. Drug Metab. Dispos. 25, 390-393. Hosokawa, M., Maki, T., and Satoh, T. (1990) Characterization of molecular species of liver microsomal carboxylesterases of several animal species and humans. Arch. Biochem. Biophys. 277, 219-227.

TX0255370

Species Differences in the in Vitro Metabolic ... - ACS Publications

Recommend Documents