QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.107.016311v1 35/10/1832    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Tang, J. Articles by Hattori, M. Search for Related Content PubMed PubMed Citation Articles by Tang, J. Articles by Hattori, M.

In Vitro Metabolism of Isoline, a Pyrrolizidine Alkaloid from Ligularia duciformis, by Rodent Liver Microsomal Esterase and Enhanced Hepatotoxicity by Esterase Inhibitors

Institute of Natural Medicine (J.T., N.N., M.H.), Graduate School of Medicine and Pharmaceutical Sciences (T.A., K.T., M.S.), University of Toyama, Toyama, Japan; Department of Pharmacognosy, China Pharmaceutical University, Nanjing, China (J.T., Z.W.); and College of Pharmacy, Wuhan University, Wuhan, China (J.T.)

(Received April 18, 2007; accepted July 13, 2007)

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Isoline, a major retronecine-type pyrrolizidine alkaloid (PA) from the Chinese medicinal herb Ligularia duciformis, was suggested to be the most toxic known PA. Its in vitro metabolism was thus examined in rat and mouse liver microsomes, and its toxicity was compared with that of clivorine and monocrotaline after i.p. injection in mice. Isoline was more rapidly metabolized by both microsomes than clivorine and monocrotaline and converted to two polar metabolites M1 and M2, which were spectroscopically determined to be bisline (a deacetylated metabolite of isoline) and bisline lactone, respectively. Both metabolites were formed in the presence or absence of an NADPH-generating system with liver microsomes but not cytosol. Their formation was completely inhibited by the esterase inhibitors, triorthocresyl phosphate (TOCP) and phenylmethylsulfonyl fluoride, but not at all or partially by cytochrome P450 (P450) inhibitors, -naphthoflavone and proadifen (SKF 525A), respectively. These results demonstrated that both metabolites were produced by microsomal esterase(s) but not P450 isozymes. The esterase(s) involved showed not only quite different activities but also responses to different inhibitors in rat and mouse liver microsomes, suggesting that different key isozyme(s) or combinations might be responsible for the deacetylation of isoline. Isoline injected i.p. into mice induced liver-specific toxicity that was much greater than that with either clivorine or monocrotaline, as judged by histopathology as well as serum alanine aminotransferase and aspartate aminotransferase levels. Isoline-induced hepatotoxicity was remarkably enhanced by the esterase inhibitor TOCP but was reduced by the P450 inhibitor SKF 525A, indicating that rodent hepatic esterase(s) played a principal role in the detoxification of isoline via rapid deacetylation in vivo.

Toxic PAs are generally retronecine and otonecine esterified PAs. The metabolism and toxicity of various PAs have been studied in vitro and in vivo (Hayes et al., 1984; Buhler et al., 1990; Lin et al., 2000; Fu et al., 2004). The hepatotoxicity of PAs largely depends on their metabolic activation by hepatic P450s to become chemically reactive pyrroles (dehydro-PAs) (Mattocks, 1968; Mattocks and Bird, 1983; Glowaz et al., 1992; Lin et al., 2000) that form covalent adducts with cellular nucleophiles including DNA (Yang et al., 2001; Xia et al., 2004). On the other hand, alternative metabolic routes, including ester hydrolysis to the subgroup retronecine by guinea pig liver carboxylesterase GPH1 (Dueker et al., 1992a, 1995; Chung and Buhler, 1995) and N-oxidation by monooxygenase-containing flavine or P450s (Williams et al., 1989; Miranda et al., 1991; Huan et al., 1998), appear to constitute a detoxification mechanism. More recently, clivorine, a representative otonecine-type PA, was found to be hydrolyzed by microsomal esterase to generate a female Sprague-Dawley rat-specific metabolite deacetylclivorine (Lin et al., 2007). Nevertheless, unlike metabolic activation by P450s, the ester hydrolysis of PAs by hepatic esterase and its physiological role in hepatotoxicity are still largely unknown.

Isoline is a major retronecine-type pyrrolizidine alkaloid found in a Chinese medicinal herb [Ligularia duciformis (C. Winkl.) Hand-Mazz.] that has been used as an expectorant and antitussive folk remedy (Zhao et al., 1998; Tang et al., 2004). However, little is known about its metabolism and toxicity. Even though isoline might be the most toxic known PA, its lethality was compared only with that of retrorsine, which was highly hepatotoxic (Sapiro, 1953; Mattocks, 1981; Mattocks and Bird, 1983). It is also interesting to note that isoline may not exert direct hepatotoxicity at a concentration of 0.1 mM by in vitro examination on hepatocytes (Ji et al., 2004). Because the toxicity of PAs differed significantly with respect to structural features (Mattocks, 1981), the toxicity of isoline should be compared with that of other structurally similar PAs, such as clivorine and monocrotaline, which become hepatotoxic after oxidation by microsomal P450s (Lame et al., 1990; Lin et al., 2000). The metabolism and liver and lung toxicities of monocrotaline as a retronecine-type PA have also been studied extensively (Huxtable, 1990; Glowaz et al., 1992; Fu et al., 2004). Moreover, animal resistance or susceptibility to PAs seems to be well known as a consequence of diversity in hepatic PA metabolism (Huan et al., 1998). Understanding the metabolic fate and balance of PAs in various pathways could be clinically important for prediction and prevention of emergent PA-induced threats or potent treatment of already present intoxication. Thus, we studied the in vitro metabolism of isoline by both rat and mouse liver microsomes compared with that of clivorine and monocrotaline. Isoline was most rapidly metabolized by rodent liver microsomes and was the most hepatotoxic of these three PAs in mice. Microsomal carboxylesterase(s) was the key enzyme(s) that contributed to the rapid metabolism and, consequently, the detoxification of isoline in vivo.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Chemicals. Isoline and clivorine were isolated from L. duciformis (C. Winkl.) Hand-Mazz. and L. hodgsonii Hook, respectively, in our laboratory. Monocrotaline, which was originally isolated from Crotalaria assamica Benth. (Tang et al., 2003b) was kindly supplied by Professor Zhiben Tu (Wuhan Botanical Institute, Academia Sinica, Wuhan, China). Their structures were confirmed using UV, IR, NMR, and MS analysis, and HPLC analysis showed that the purity of all three was >98%. Acetonitrile and methanol of HPLC grade were purchased from Wako Pure Chemicals (Osaka, Japan). Proadifen (SKF 525A) was purchased from Sigma Chemical (St. Louis, MO). Triorthocresyl phosphate (TOCP) and tetrachloro-o-benzoquinone (o-chloranil) were purchased from Tokyo Kasei Kogyo Co. (Tokyo, Japan). Phenylmethylsulfonyl fluoride (PMSF), -naphthoflavone (ANF), ß-NADH, ß-NADP⁺, glucose 6-phosphate, glucose-6-phosphate dehydrogenase, the reduced form of glutathione (GSH), alanine aminotransferase (ALT) and aspartate aminotransferase (AST) test kits, 10% formalin neutral buffer solution (pH 7.4 for tissue fixation), and phenobarbital sodium (PB) were all purchased from Wako Pure Chemicals. Dimethyl sulfoxide and all other chemicals and solvents were of the purest commercially available grade. For the isolation of the two metabolites of isoline, M1 and M2, a scaled-up rat microsomal incubation with isoline as described below was used. Both isolated and synthetic metabolites were subjected to IR, UV, MS, NMR, and HPLC-UV analysis for the confirmation of their identity and purities (>98%), respectively (Tang et al., 2003a,b,c).

Instrumentation. HPLC analysis was performed using a Tosoh HPLC system consisting of CCPM-2 dual pumps, a UV 8020 detector, an online degasser, and a Shimadzu C-R6A data processor under the same conditions as reported previously (Tang et al., 2004). TLC separation for preliminary study or for preparations purposes was performed on silica gel F₂₅₄ plates (0.25 mm; Merck, Darmstadt, Germany) using 1) chloroform-methanol-28% ammonia (8:2:0.1/2:8:0.1); 2) ethyl acetate-acetone-ethanol-28% ammonia (5:3:1:1), and 3) methanol-chloroform-water-triethylamine (8:0.85:1.1:0.05). Pyrrole metabolites were detected using Ehrlich reagent (Mattocks, 1968) sprayed directly. Parent PAs were detected using iodine or Dragendorff reagent. Pyrroles or parent PAs were also detected using iodine, acetic anhydride, and Ehrlich reagent in sequence in addition to o-chloranil in benzene, followed by Ehrlich reagent (for parent PAs). N-Oxides of PAs were visualized by spraying with acetic anhydride, followed by Ehrlich reagent. ¹H and ¹³C NMR spectra were measured using a Varian Unity Plus 500 (Varian Inc., Palo Alto, CA) spectrometer operating at 500 MHz for proton and 125 MHz for carbon in CDCl₃ or CD₃OD with trimethylsilane as an internal standard. Electron ionization-MS was measured using a JEOL JMS-GC-mate mass spectrometer at an ionization voltage of 70 eV (JEOL, Tokyo, Japan). High-resolution electron impact mass spectrometry and fast atom bombardment-MS were performed with a JEOL JMS-DX 300L spectrometer using glycerol as a matrix for the latter. Optical rotation was measured using a DIP-360 automatic polarimeter (Jasco, Tokyo, Japan). UV spectra were measured with a Shimadzu UV-2200 recording spectrophotometer (Shimadzu, Kyoto, Japan). IR spectra were recorded on a Jasco FT/IR-230 infrared spectrometer (Jasco). Circular dichroism spectra were measured with a Jasco J-805 spectropolarimeter (Jasco). Column chromatography was performed using silica gel BW-820MH, ODS (Fuji Silysia Co., Nagoya, Japan), and Diaion HP-20 (Mitsubishi Kasei Co., Tokyo, Japan). Preparative TLC was performed on silica gel 60 F₂₅₄ plates (0.5 mm, Merck).

Animals and Preparation of Liver Microsomes. Male Sprague-Dawley rats weighing 240 to 280 g and male ddY mice weighing 30 to 37 g were purchased from Nippon SLC (Hamamatsu, Japan). Microsomes were induced by injecting the rats i.p. with PB (80 mg/kg in saline) daily for 3 days (Hayes et al., 1984). Animals starved during the last 12 h were killed under pentobarbital anesthesia (50 mg/ml/kg), and then the livers were rapidly removed and weighed. Blood was removed by perfusion with cold 0.25 M sucrose, and the liver was homogenized in 4 volumes of ice-cold 0.25 M sucrose in 55-ml Wheaton tubes with a pestle. The homogenate was centrifuged at 11,472g for 30 min, and the supernatant was separated again by centrifugation at 105,000g for 90 min. After washing with 0.25 M sucrose, the pellets (microsomal fraction) were suspended in ice-cold 0.1 M potassium phosphate buffer (1:2 volume/liver weight, pH 7.4) containing 20% glycerol and 0.1 mM EDTA disodium and then were stored in portions at - 80°C. The supernatants (cytosol fraction) were also stored at - 80°C. Protein content was determined using the modified method of Lowry et al. (1951).

Incubation of Pyrrolizidine Alkaloids with Microsomes and Preparation for TLC/HPLC. Incubation mixtures (0.2 ml) consisted of an NADPH-generating system, liver microsomes (0.48 mg of protein), and 1 mM isoline in potassium phosphate buffer (0.1 M, pH 7.4) in the presence or absence of 2.0 mM GSH. The NADPH-generating system contained 5 mM MgCl₂,1mM ß-NADH, 1 mM ß-NADP⁺, 10 mM glucose 6-phosphate, and 1.0 unit/ml glucose-6-phosphate dehydrogenase. Although the optimal pH value of the incubation system was 8.0 in the present study, the P450-mediated metabolism of PAs was usually examined at pH 7.4 (Kedzierski and Buhler, 1986). Other conditions including substrate concentration were the same as in previous reports (Kedzierski and Buhler, 1986; Williams et al., 1989; Dueker et al., 1992a; Lin et al., 2000). The reaction was initiated by adding substrate (isoline or other PA), performed at 37°C, and terminated by adding an equal volume of ice-cold methanol. Control incubations were performed without the NADPH-generating system and with or without rodent liver microsomes or substrate. The residue obtained by centrifugation at 10,000g at 4°C for 30 min was immediately filtered through 0.45-µm microdiscs (Ekicrodisc 3, 3 mm; Gelman Sciences, Inc., Ann Arbor, MI), and the filtrates were analyzed by TLC and HPLC.

Incubation with Esterase and P450 Inhibitors. The mixtures described above were incubated at 37°C for 60 min with one of the following inhibitors: TOCP (0.01, 0.1, 1, and 2.5 mM) and PMSF (0.1, 0.5, and 1 mM) for esterase(s) or SKF 525A (0.1 and 0.5 mM) and ANF (0.1 and 0.5 mM) for P450s, each with differential selectivity. The inhibitor concentrations were defined according to the preliminary TLC studies or published data to achieve significant inhibition effects by TOCP and PMSF (Dueker et al., 1992a,b; Lin et al., 2007) and SKF 525A and ANF (Hayes et al., 1984; Williams et al., 1989). Because these inhibitors were dissolved in dimethyl sulfoxide (<0.5% in the incubation mixture), the control incubation also contained this solvent. Reaction inhibition is expressed as percent decrease in formed metabolites on a molar basis.

Isolation of Metabolites of Isoline Using Rat Liver Microsomes. The procedure described above was scaled up (400 ml) with the slight modification of a 90-min incubation. The mixture was immediately extracted with 3 volumes of methanol and stored at 4°C. The extract was centrifuged at 10,000g at 2°C for 15 min. The supernatant was concentrated in vacuo and eluted through a Diaion HP-20 column with water, 30% methanol, 70% methanol, and 100% methanol. Each eluate was totally evaporated in vacuo to yield a residue. The 70% methanol fraction was eluted through a silica gel column with chloroform-methanol-25% ammonia (19:1:0.1 to 8:2:0.1). Fractions 2 to 4 were collected, concentrated, and purified by preparative TLC to yield a white crystalline compound, M1 (approximately 40 mg). Similarly, fractions 12 to 20 yielded an oily compound, M2 (approximately 20 mg). M1 and M2 were determined to be bisline and bisline lactone by use of UV, IR, MS, circular dichroism, and extensive use of NMR techniques, respectively, as described in our previous studies (Tang et al., 2003a, 2004).

View larger version (22K): [in this window] [in a new window]   FIG. 1. Structures of isoline, clivorine, monocrotaline, and two major metabolites from isoline.

Quantification of Metabolites. An HPLC assay method to qualitatively and quantitatively analyze M1, M2, and isoline in all incubates was developed as described previously (Tang et al., 2004).

In Vivo Animal Treatment and Dosing. Male ddY mice (6 weeks old, 30-37 g) were bred at the Experimental Animals Service Center of Toyama Medical and Pharmaceutical University on a 12-h light/dark cycle in a room with controlled temperature and humidity for 1 week before the experiments. Isoline, clivorine, and monocrotaline were dissolved in 0.9% saline containing a suitable volume of 0.2 M hydrochloric acid and neutralized with 0.2 M sodium hydroxide to give respective solutions of 12.7 (low), 25.3 (medium), and 50.6 mM (high) compounds. Each of these was administered once daily as an i.p. injection at a dose of 10 µl/g b.wt. for 1 or 3 days. Some mice were given undiluted TOCP (1 ml/kg b.wt.) via a stomach tube 4 h before administration of PAs (Mattocks, 1981; Mattocks et al., 1986; Chu et al., 1993). Others received SKF 525A at a dose of 50 mg/kg b.wt. by i.p. injection 15 min before PA administration as described by Ohhira et al. (2000). Controls were injected with saline alone. Thereafter, blood was collected via the abdominal vein under pentobarbital anesthesia, and the liver and other organs were immediately removed. Blood samples were stored at room temperature for >1 h to obtain serum for ALT and AST measurements. All animal experiments were conducted in accordance with the Guidelines of the Animal Care and Use Committee of Toyama Medical and Pharmaceutical University approved by the Japanese Association of Laboratory Animal Care.

Histopathology. The liver and other organs were fixed in buffered formalin (pH 7.4) for >24 h, embedded in paraffin, sliced into 4-µm sections, and stained with hematoxylin and eosin for light microscopic evaluation.

Statistical Analysis. Serum ALT and AST activity data were analyzed using one-way analysis of variance followed by Dunnett's test. Other variables were examined using Student's t test. Groups were considered significantly different when p was < 0.05.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Isolation and Characterization of Two Major Metabolites of Isoline Generated by Rat Liver Microsomes. Two major metabolites (M1 and M2) were generated by incubating isoline with rat liver microsomes in the presence of an NADPH-generating system. The amounts of the two metabolites produced almost equaled the amount by which isoline decreased on a molar basis. M1 was identified as (2S,3R,5R)-15-ethyl-12,15-dihydroxy-12,13-dimethylsenec-1-enine (bisline) (Fig. 1) by comparison with an authentic sample (see Materials and Methods) and with the reported composition of bisline (Susag et al., 2000). M2 was determined as the 9-O-(-lactone) form of [(12S,13R,15R)-15-ethyl-15-hydroxy-12,13-dimethyl-17-oxotetrahydropyran-12,15-dicarboxylic acid] retronecine (bisline lactone) (Fig. 1) (Tang et al., 2003a). M1 is a deacetylated metabolite of isoline, and M2 is formed by the 12-membered ring-fission of M1 and subsequent ring-closure to generate a 6-membered lactone ring. By HPLC, isoline and its two metabolites could be analyzed with retention times of approximately 43.4, 27.2, and 29.7 min, respectively (Tang et al., 2004).

Metabolism of Isoline, Clivorine, and Monocrotaline by Rodent Liver Microsomes. The metabolism of isoline, clivorine, and monocrotaline (Fig. 1) was investigated using both rat and mouse liver microsomes in the presence of an NADPH-generating system. Both types of microsomes metabolized isoline much more rapidly than either clivorine or monocrotaline, and mouse liver microsomes metabolized isoline more rapidly than those from rats (Table 1). For mouse liver microsomes, higher metabolic activation with clivorine and monocrotaline could be observed. It was noteworthy that two major metabolites (M1 and M2) of isoline generated by mouse liver microsomes were confirmed to be same as those generated by rat liver microsomes by using TLC (with at least three developing systems) and HPLC analysis as well as concurrent chromatography with the authentic samples. On the other hand, rat or mouse liver cytosol did not metabolize any of three PAs. Then the effects of PB on the microsomal metabolism of three compounds were examined by exposing rats to PB. Isoline metabolism was enhanced by approximately 1.5-fold, compared with that of clivorine (5-fold) but much lower than that of monocrotaline by a slight decrease of the parent alkaloid (Table 1).

View larger version (13K): [in this window] [in a new window]   FIG. 2. Time course of isoline metabolism by rat liver microsomes. Isoline (, 1 mM) was incubated with hepatic microsomes in the presence of an NADPH-generating system. The reaction mixture was sampled at the indicated times, and isoline and metabolites (M1, ; M2, ) as well as the sum of isoline and two metabolites (x) were quantified.

Enzyme Inhibition Studies. The metabolism of isoline to M1 and M2 by rat liver microsomes was dose dependent and was completely inhibited by esterase inhibitors such as TOCP and PMSF (Table 2). TOCP (a carboxylesterase inhibitor) inhibited the formation of M1 and M2 by 85.9 and 80.0%, respectively, at 0.01 mM and almost completely at 0.1 mM, whereas PMSF (selective to hydrolase A) inhibited their formation by 84.2 and 81.5%, respectively, at 0.1 mM and almost completely at 0.5 mM. Moreover, among the P450 inhibitors, SKF 525A (nonspecific P450 inhibitor) considerably inhibited the formation of the metabolites, in agreement with a previous report that SKF 525A partially inhibited rat microsomal esterase (Yeh, 1982). However, ANF (relatively selective to CYP2C8/9 and 1A/2) had no effect.

Mouse liver microsomes similarly tended to show an inhibitory effect of isoline metabolism, except for the absence of inhibition by SKF 525A and ANF and decreased inhibition by TOCP and PMSF, compared with rat liver microsomes (Table 2). TOCP had a dose-dependent inhibitory effect on the metabolism of isoline from 0.01 to 2.5 mM, but this inhibitory effect was much less than that of PMSF. Significant differences in formation of M1 and M2 between the control group without inhibitors and those with TOCP and PMSF were also observed (at least at p < 0.05). These findings showed that isoline was converted to M1 by rodent liver microsomal esterase(s).

Isoline Toxicity and Histopathology. The in vivo toxicity of isoline and other two PAs (clivorine and monocrotaline) to mice were examined. Isoline was lethal at a dose of 0.506 mmol/kg over 2 days (Table 3). Light microscopy showed liver damage consisting of hemorrhage and coagulative necrosis around the central vein at a medium dose of 0.253 mmol/kg isoline for 3 days. The coagulative necrosis was characterized by aggregative disappearance of cell membranes. Under the high-power lens, the cells in and around the area of necrosis showed enlarged nuclei and the changes in the cytoplasm as indicated by staining. The inflammatory cells were slightly increased. At a high single dose of 0.506 mmol/kg, the liver injury was more acute, extensive, and severe. Cage observation of the mice showed that they were very sick, listless, and inert. Livers were swollen and darkened after surgical removal. Hemorrhage and lesions consisting of areas of coagulative cell necrosis were greatly increased as shown on light microscopic examination (Fig. 3). Meanwhile, apoptotic bodies and enlarged nuclei were observed in hepatocytes outside the regions of necrosis. Even in those normal areas with fewer morphologic changes (approximately 50%), the cells with enlarged nuclei and the changed cytoplasm indicated by staining could still be observed. However, isoline did not damage other mouse tissues such as the heart, lung, and kidney (data not shown).

View larger version (158K): [in this window] [in a new window]   FIG. 3. Photomicrography of liver sections from mice treated with isoline, TOCP-isoline, and saline vehicle. Mice were administered with either isoline or saline vehicle, and tissues including liver, heart, lung, and kidney were removed 24 h after the last treatment and processed as described under Materials and Methods. A, liver of mouse treated with saline vehicle for 3 days. P, portal area; C, central vein. B, liver of mouse treated with 0.253 mmol/kg isoline for 3 days. *, coagulation necrosis; P, portal area. C, liver from mouse treated with a single dose of 0.506 mmol/kg isoline. C-1, extensive hemorrhage; C-2, apoptotic body (arrows); C-3, hemorrhage and coagulation necrosis. P, portal area. D, liver of mouse pretreated with 1 ml/kg TOCP followed by one dose of 0.253 mmol/kg isoline. Extensive hemorrhage and necrosis around the central vein was dose dependently enhanced by prior exposure to TOCP. A, B, C-1, and D, bar = 200 µm; C-2 and C-3, bar = 50 µm.

Monocrotaline had no effect on the liver even at a dose of 0.506 mmol/kg for 3 days. On the other hand, clivorine was lethal (Table 3), and body weight was decreased by 6.5 ± 1.1 g after the administration of 0.506 mmol/kg for 3 days, compared with increases of 0.2 to 0.5 g in the control and isoline-treated groups. However, light microscopy showed that a single high dose of clivorine did not significantly damage the liver.

Dose-Dependent Hepatotoxicity of Isoline and Enhancement by TOCP. The liver damage was quantified by measuring serum ALT and AST activities after administration of 0.127, 0.253, and 0.506 mmol/kg (low, medium, and high doses, respectively) of each of the three PAs. Table 3 shows that isoline remarkably increased serum ALT and AST activities (ALT 1003 ± 552 IU/l and AST 1065 ± 601 IU/l) after a single high dose. The increase induced by the low and medium doses of isoline for either 1 or 3 days was also dose dependently high. In contrast, monocrotaline did not increase serum activities of either ALT or AST even after a high dose for 3 days. On the other hand, clivorine slightly affected serum activities after a single high dose and after three medium doses.

When mice were given TOCP at a dose of 1 ml/kg 4 h beforehand, isoline was lethal at both high and medium doses and significantly increased serum ALT and AST activities even at a single low or medium dose (Table 3). Light microscopy confirmed the liver-specific damage in the TOCP-treated mice (Fig. 3). In addition, bisline (M1 of isoline) did not elicit any apparent toxic effects at a single medium dose. Pretreatment with TOCP did not affect the hepatotoxicity of monocrotaline but significantly enhanced that of clivorine only at a single high dose, which was also confirmed by light microscopy. On the other hand, prior exposure to SKF 525A at a dose of 50 mg/kg completely abolished the increases in serum ALT and AST activities induced by isoline at a single medium or low dose together with TOCP pretreatment and partially suppressed those induced by isoline at a single medium dose with TOCP (Table 3), suggesting that the hepatotoxicity of isoline depends on metabolic activation by P450s.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The toxicity of PAs is due to metabolic activation by an NADPH-dependent P450 enzymatic system, and detoxification is thought to include N-oxidation (Mattocks and Bird, 1983; Williams et al., 1989) and ester hydrolysis (Mattocks, 1970; Chung and Buhler, 1995). Understanding of the ester hydrolysis is primarily based on the formation of hydrolysis products such as retronecine, a feature of all retronecine-type PAs with various acid moieties, by purified carboxylesterase (GPH1) from guinea pig hepatic microsomes and the nontoxic and excretable property of these products (Dueker et al., 1992a, 1995).

The present study identified a distinct pathway catalyzed by rodent liver microsomal esterase(s) for PAs. Two major metabolites (M1 and M2) of isoline, a retronecine-type PA, were produced by both rat and mouse liver microsomes (either untreated or pretreated with PB) in the presence or absence of an NADPH-generating system but not in cytosolic incubations. Both structures (Fig. 1) showed hydrolysis of the acetyl group at the C-12 position of isoline. Interestingly, M2 was transformed from M1 nonenzymatically but without further hydrolysis at the C-9 position, which is generally believed to be especially susceptible to enzymic hydrolysis (Fu et al., 2004). Further HPLC analyses demonstrated their hydrophilicity to be higher than that of isoline (Tang et al., 2004). Esterase inhibitors, such as TOCP and PMSF, significantly inhibited M1 and M2 formation, whereas the P450s inhibitors with differential selectivity, ANF and SKF 525A, were not inhibitory and partially inhibitory, respectively. All of these data indicated that microsomal esterase(s) mediated the formation of both polar metabolites.

Because isoline is structurally similar to clivorine in the necic acid moiety and to monocrotaline in the necine base moiety (Fig. 1), their metabolism and hepatotoxicity were compared. Rat or mouse liver microsomes metabolized the three PAs at significantly different ratios. Isoline was the most rapidly metabolized, showing the high metabolic activity of liver microsomal esterase(s), especially in mice or by PB-induced rat liver microsomes. This finding indicated not only that the same inducer of P450s also induced carboxylesterase isozymes (Satoh and Hosokawa, 1998) but also that esterase activities are probably higher in mouse liver microsomes. The rat (PB-induced) and mouse microsomal esterase(s) could also metabolize clivorine, which contains an acetyl group at the same position as the necic acid moiety, confirmed by TOCP inhibition (with 74 and 80% at concentrations of 0.01 and 0.1 mM in mice, respectively). This observation indicated that hepatic esterase(s) also played a role in clivorine metabolism by deacetylation. In contrast, monocrotaline, which contains the same necine residue (retronecine base) but no acetyl group, was unchanged under the same conditions. Although monocrotaline was a good substrate for guinea pig liver carboxylesterase GPH1 (Dueker et al., 1992a), the macrocyclic ester of the necic acid moiety was hydrolyzed at a different position. In the experiments described above, we did not detect retronecine, a common PA hydrolysis product generated by GPH1 (Dueker et al., 1992a,b, 1995). This finding implies that the esterase(s) present in our microsomal incubation system might be different isoforms specially targeted for the acetyl substituent.

Similarly, a recent study (Lin et al., 2007) showed that clivorine could undergo deacetylation by PMSF-sensitive microsomal hydrolase A, suggesting a possible detoxification pathway in female rats. In our study, M1 and M2 were deacetylated metabolites of isoline, with formation inhibited dose dependently in mice by either TOCP or PMSF. PMSF almost completely abolished the hydrolysis in mice at a concentration of 1 mM. Additionally mouse liver microsomes were more resistant to TOCP inhibition than PMSF, showing more PMSF sensitivity than rat liver microsomes (Table 2). This observation indicates that different key isozyme(s) or combinations might be responsible for the deacetylation of isoline. The extensive metabolism to clivorine, another PA with similar acetyl structures (both S-configuration at C-12 position) (Fig. 1), further indicates that esterase(s) may have broad and overlapping substrate specificity to esters (Satoh and Hosokawa, 1998). It is well established that the liver microsomal fraction contains the highest activity of esterase(s), but there are large species differences, especially among small laboratory animals. Although the corresponding isoform in mouse (MH1) could be found with homology and characteristics similar to those of rat hydrolase A (Satoh and Hosokawa, 1998; Lin et al., 2007), further research to verify its identity and functions is needed.

To better understand the toxicity of isoline among the known and possible detoxification feature of the esterase(s), the toxic effects of isoline and clivorine and monocrotaline on mice were examined. Isoline was the most hepatotoxic of the three, judging from both histopathological findings and serum ALT and AST values (Table 3; Fig. 3). Isoline induced a marked elevation in the serum ALT level in mice and caused liver lesions at a single medium dose (0.253 mmol/kg). At a single high dose (0.506 mmol/kg) or a medium dose for 3 days, isoline produced even more significant severe damage (Fig. 3). On the basis of the serum ALT and AST activities, the toxicity was increased in a dose-dependent manner from low to high doses. On the other hand, the high doses of clivorine and monocrotaline were not significantly hepatotoxic after 1 and 3 days, respectively. Moreover, at an equivalent molar dose (0.253 mmol/kg) for 3 days, clivorine and monocrotaline showed much lower toxicity than isoline. Furthermore, M1 (bisline), a metabolite of isoline produced by microsomal esterase, administered at a single medium dose of 0.253 mmol/kg, was not significantly toxic to the mouse liver. Together with in vitro incubations and the altered chemical and probably toxicokinetic properties of two metabolites, we suggested that the high toxicity of parent PA (isoline) may not be derived from its secondary metabolism. However, further in vivo studies using multiple dosing of M1 or M2 are warranted.

TOCP is a well-known carboxylesterase inhibitor in vivo (Mattocks et al., 1986) and has very low toxicity. Prior exposure to TOCP significantly enhanced liver damage at all tested doses of isoline (Fig. 3; Table 3). TOCP followed by a single high dose of clivorine also induced severe liver damage and sometimes death. Clivorine also possesses an acetyl group and was not significantly toxic alone. On the contrary, monocrotaline does not possess an acetyl group, and this compound did not elicit any effect, even at a single high dose, regardless of prior TOCP exposure. These results demonstrated the different responses of mice to the three PAs and in particular much less susceptibility to clivorine than in another report (Lin et al., 2007), corresponding with large species differences in vitro. These in vivo toxicity tests associated with in vitro metabolic findings indicated that the hydrolysis of the acetyl group by esterase was a detoxification process and suggested that an acetyl group attached to the macro ring of PAs constituted a vital part of the hepatotoxicity.

On the other hand, the real reason for isoline hepatotoxicity may depend on the bioactivation pathway largely mediated by P450 isozymes (Mattocks, 1968; Buhler et al., 1990). To confirm this point, the effects of P450 inhibition by SKF 525A on isoline-induced hepatotoxicity were studied. SKF 525A completely abolished the induction of hepatotoxicity by a single subsequent medium dose of isoline or a single low dose of isoline after pretreatment with TOCP and significantly repressed the severe hepatotoxicity induced by a single medium dose of isoline with prior dosing of TOCP (Table 3). Accordingly, the high in vivo toxicity of isoline might be related to the bioactivation pathway by P450s, which metabolized PAs into unstable and highly reactive "pyrroles" These compounds efficiently reacted with many nucleophiles such as GSH to spontaneously decompose or polymerize in seconds (Mattocks et al., 1989; Tang et al., 2003b,c) or attack macromolecules such as DNA (Kim et al., 1999; Yang et al., 2001). However, monocrotaline that was probably metabolized to the pyrrole (Glowaz et al., 1992) by P450s caused extrahepatic diseases with liver damage. Thus, the acute and severe hepatotoxicity induced by isoline or other PAs containing an acetyl group in the necic acid moiety might be due to their rapid metabolism predominately by the metabolic pathway involving P450s. Therefore, further investigation on P450 isozyme-mediated metabolic activation of isoline would be our next subject.

In summary, the rodent hepatic microsomal esterases were responsible for the biotransformation of isoline to two deacetylated metabolites (M1 and M2) in vitro, especially in mouse and PB-induced rat microsomes. Isoline was also metabolized much more extensively than clivorine and monocrotaline but showed characteristics some-what similar to those of clivorine, with the same acetyl structure. The in vivo experimental results indicated that isoline was the most toxic PA among the three and also was liver-specific. The hepatotoxicity of isoline was enhanced by esterase inhibition by TOCP as in the case with clivorine but not with monocrotaline. On the other hand, the isoline-induced hepatotoxicity could be remarkably decreased by SKF 525A, a P450 inhibitor. All of these data demonstrated that microsomal esterase isozymes played a key role in the detoxification of isoline associated with metabolic activation initiated by P450s. Further studies will be conducted to identify the metabolic fate of isoline by P450s and microsomal esterases, particularly their intrinsic role in vivo. The underlying regulatory mechanisms could have potential clinical implications.

	Footnotes

 
This research was supported by Grant 39825129 from the Natural Science Foundation of China for outstanding young scientists to Z.W.

doi:10.1124/dmd.107.016311.

ABBREVIATIONS: PA, pyrrolizidine alkaloid; P450, cytochrome P-450 enzyme; NMR, nuclear magnetic resonance; MS, mass spectrometry; HPLC, high-performance liquid chromatography; TOCP, triorthocresyl phosphate; PMSF, phenylmethylsulfonyl fluoride; ANF, -naphthoflavone; GSH, glutathione; ALT, alanine aminotransferase; AST, aspartate aminotransferase; PB, phenobarbital sodium; TLC, thin layer chromatography.

Address correspondence to: Dr. Teruaki Akao, Graduate School of Medicine and Pharmaceutical University, University of Toyama, Toyama 930-0194, Japan. E-mail: ta0113{at}pha.u-toyama.ac.jp

	References

Top Abstract Materials and Methods Results Discussion References

 


Buhler DR, Miranda CL, Kedzierski B, and Reed RL (1990) Mechanisms for pyrrolizidine alkaloid activation and detoxification, in Biological Reactive Intermediates IV (Witmer CM, Snyder RS, Jollow DJ, Kalf GF, Kocsis JJ, and Sipes IG eds) pp 597-603. Plenum Press, New York.

Chu PS, Lamé MW, and Segall HJ (1993) In vivo metabolism of retrorsine and retrorsine N-oxide. Arch Toxicol 67: 39-43.[CrossRef][Medline]

Chung WG and Buhler DR (1995) Major factor for the susceptibility of guinea pig to the pyrrolizidine alkaloid jacobine. Drug Metab Dispos 23: 1263-1267.[Abstract]

Dueker SR, Lamé MW, Morin D, Wilson DW, and Segall HJ (1992a) Guinea pig and rat hepatic microsomal metabolism of monocrotaline. Drug Metab Dispos 20: 275-280.[Abstract]

Dueker SR, Lamé MW, and Segall HJ (1992b) Hydrolysis of pyrrolizidine alkaloids by guinea pig hepatic carboxylesterases. Toxicol Appl Pharmacol 117: 116-121.[CrossRef][Medline]

Dueker SR, Lamé MW, and Segall HJ (1995) Hydrolysis rates of pyrrolizidine alkaloids derived from Senecio jacobaea. Arch Toxicol 69: 725-728.

Edgar JA, Roeder E, and Molyneux RJ (2002) Honey from plants containing pyrrolizidine alkaloids: a potential threat to health. J Agric Food Chem 50: 2719-2730.[CrossRef][Medline]

Fu PP, Xia QS, Lin G, and Chou MW (2004) Pyrrolizidine alkaloids-genotoxicity, metabolism enzymes, metabolic activation, and mechanisms. Drug Metab Rev 36: 1-55.[CrossRef][Medline]

Glowaz SL, Michnika M, and Huxtable RJ (1992) Detection of a reactive pyrrole in the hepatic metabolism of the PA, monocrotaline. Toxicol Appl Pharmacol 115: 168-173.[CrossRef][Medline]

Hayes MA, Roberts E, Jago MV, Safe SH, Farber E, and Cameron RC (1984) Influence of various xenobiotic inducers on cytocidal toxicity of lasiocarpine and senecionine in primary cultures of rat hepatocytes. J Toxicol Environ Health 14: 683-694.[Medline]

Huan JY, Miranda CL, Buhler DR, and Cheeke PR (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.[CrossRef][Medline]

Huxtable RJ (1989) Human health implication of pyrrolizidine alkaloids and herbs containing them, in Toxicants of Plant Origin, Vol. 1, Alkaloids (Cheeke PR ed) pp. 41-86. CRC Press, Boca Raton, FL.

Huxtable RJ (1990) Activation and pulmonary toxicity of pyrrolizidine alkaloids. Pharmacol Ther 47: 371-389.[CrossRef][Medline]

Ji LL, Tan AM, Tang J, Zhang M, and Wang ZT (2004) Toxicity of several pyrrolizidine alkaloids on hepatocytes. Chin J Nat Med 2: 239-241.

Kedzierski B and Buhler DR (1986) Method for determination of pyrrolizidine alkaloids and their metabolites by high-performance liquid chromatography. Anal Biochem 152: 59-65.[CrossRef][Medline]

Kim HY, Stermitz FR, Li JK, and Coulombe RA (1999) Comparative DNA cross-linking by activated pyrrolizidine alkaloids. Food Chem Toxicol 37: 619-625.[CrossRef][Medline]

Lamé MW, Morin D, Jones AD, Segall HJ, and Wilson DW (1990) Isolation and identification of a pyrrolic glutathione conjugate metabolite of the pyrrolizidine alkaloid monocrotaline. Toxicol Lett 51: 321-329.[CrossRef][Medline]

Lin G, Cui YY, and Hawes EM (2000) Characterization of rat liver microsomal metabolites of clivorine, a hepatotoxic otonecine-type pyrrolizidine alkaloid. Drug Metab Dispos 28: 1475-1483.[Medline]

Lin G, Tang J, Liu XQ, Jiang Y, and Zheng J (2007) Deacetylclivorine: a gender-selective metabolite of clivorine formed in female Sprague-Dawley rat liver microsomes. Drug Metab Dispos 35: 607-613.[Abstract/Free Full Text]

Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265-275.[Free Full Text]

Mattocks AR (1968) Toxicity of pyrrolizidine alkaloids. Nature 217: 723-728.[CrossRef][Medline]

Mattocks AR (1970) Role of the acid moieties in the toxic actions of pyrrolizidine alkaloids on liver and lung. Nature 228: 174-175.[CrossRef][Medline]

Mattocks AR (1981) Relation of structural features to pyrrolic metabolites in livers of rats given pyrrolizidine alkaloids and derivatives. Chem Biol Interact 35: 301-310.[CrossRef][Medline]

Mattocks AR and Bird I (1983) Pyrrolic and N-oxide metabolites formed from pyrrolizidine alkaloids by hepatic microsomes in vitro: relevance to in vivo hepatotoxicity. Chem Biol Interact 43: 209-222.[CrossRef][Medline]

Mattocks AR, Driver HE, Barbour RH, and Robins DJ (1986) Metabolism and toxicity of synthetic analogues of macrocyclic diester pyrrolizidine alkaloids. Chem Biol Interact 58: 95-108.[CrossRef][Medline]

Mattocks AR, Jukes R, and Brown J (1989) Simple procedures for preparing putative toxic metabolites of pyrrolizidine alkaloids. Toxicon 27: 561-567.[Medline]

Miranda CL, Chung W, Reed RE, Zhao X, Henderson MC, Wang JL, Williams DE, and Buhler DR (1991) Flavin-containing monooxygenase: a major detoxifying enzyme for the pyrrolizidine alkaloid senecionine in guinea pig tissues. Biochem Biophys Res Commun 178: 546-552.[CrossRef][Medline]

Ohhira S, Matsui H, and Watanabe K (2000) Effects of pretreatment with SKF 525A on triphenyltin metabolism and toxicity in mice. Toxicol Lett 117: 145-150.[CrossRef][Medline]

Roeder E (2000) Medicinal plants in China containing pyrrolizidine alkaloids. Pharmazie 55: 711-726.[Medline]

Sapiro ML (1953) The alkaloids of Senecio ruwenzoriensis. J Chem Soc 1942-1943.

Satoh T and Hosokawa M (1998) The mammalian carboxylesterases: from molecules to functions. Annu Rev Pharmacol Toxicol 38: 257-288.[CrossRef][Medline]

Susag L, Parvez M, Mathenge S, and Benn MH (2000) Structure revision of isoline (ruwenine), bisline and isolinecic acid. Phytochemistry 54: 933-935.[CrossRef][Medline]

Tang J, Wang ZT, Akao T, Nakamura N, and Hattori M (2003a) Two new lactones metabolized from isoline by rat liver microsomes. Chin Chem Lett 14: 1271-1274.

Tang J, Zhang M, Wang ZT, Akao T, Nakamura N, and Hattori M (2004) Simultaneous determination of isoline and its two major metabolites using high performance liquid chromatography. J Anal Toxicol 28: 11-15.[Medline]

Tang J, Zhang M, Wang ZT, Akao T, Nakamura N, Hattori M (2003b) Identity of monocrotalic acid released from dehydromonocrotaline, a high chemically reactive pyrrolic ester. J China Pharm University 34: 499-502.

Tang J, Zhang M, Wang ZT, Akao T, Nakamura N, Hattori M, and Xu GJ (2003c) Preparation and characterization of two new metabolites of isoline. Chin J Nat Med 1: 21-23.

Williams DE, Reed RL, Kedzierski B, Dannan GA, Guengerich FP, and Buhler DR (1989) Bioactivation and detoxication of the pyrrolizidine alkaloid senecionine by cytochrome P450 enzymes in rat liver. Drug Metab Dispos 17: 387-392.[Abstract]

Xia Q, Chou MW, Lin G, and Fu PP (2004) Metabolic formation of DHP-derived DNA adducts from a representative otonecine type pyrrolizidine alkaloid clivorine and the extract of Ligularia hodgsonnii hook. Chem Res Toxicol 17: 702-708.[CrossRef][Medline]

Yang YC, Yan J, Doerge DR, Chan PC, Fu PP, and Chou MW (2001) Metabolic activation of the tumorigenic pyrrolizidine alkaloid, riddelline, leading to DNA adduct formation in vitro. Chem Res Toxicol 14: 101-109.

Yeh SH (1982) Localization and characterization of meperidine esterase of rats. Drug Metab Dispos 10: 319-325.[Abstract]

Zhao XG, Li ZM, Zhang M, Wang ZT, Ma JY, Xu LS, Yu GD, Xu GJ, and Lin G (1998) Studies on Chinese medicinal herbs: Shanziwan (Radix et Rhizoma Ligulariae). 1. Survey on resources and identification of original plants. Chin Trad Herb Drugs 29: 115-119.

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.107.016311v1 35/10/1832    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Tang, J. Articles by Hattori, M. Search for Related Content PubMed PubMed Citation Articles by Tang, J. Articles by Hattori, M.