Species differences in the hepatic microsomal enzyme metabolism of the pyrrolizidine alkaloids

Article abstract of DOI:10.1016/S0378-4274(98)00152-0

Species differences in pyrrolic metabolites and senecionine (SN) N-oxide formation among eight animal species (sheep, cattle, gerbils, rabbits, hamsters, Japanese quail, chickens, rats) varying in susceptibility to pyrrolizidine alkaloid (PA) intoxication were measured in vitro by hepatic microsomal incubations. The results suggested that there is not a strong correlation between the production of pyrrolic metabolites and susceptibility of animals to PA toxicity. The rate of PA activation in hamsters, a resistant species, measured by formation of (±)6,7-dihydro-7-hydroxy-1-hydroxymethyl-5H-pyrrolizine (DHP) far exceeded the rate of SN N-oxide formation (detoxification) (DHP/N-oxide=2.29). In contrast, SN N-oxide was the major metabolite in sheep, another resistant species, with much lower production of DHP (DHP/N-oxide=0.26). The roles of cytochrome P450s and flavin-containing monooxygenases (FMO) in bioactivation and detoxification of pyrrolizidine alkaloids (PA) were studied in vitro using sheep and hamster hepatic microsomes. Chemical and immunochemical inhibition data suggested that the conversion of SN to DHP is catalyzed mainly by cytochrome P450s (68-82%), whereas the formation of SN N-oxide is carried out largely by FMO (55-71%). There also appeared to be a high rate of glutathione-DHP conjugation in hamster (63%) and sheep (79%) liver microsomal incubation mixtures. Therefore, low rates of pyrrole metabolite production coupled with glutathione conjugation in sheep may explain the resistance of sheep to SN, whereas the high rate of GSH-DHP conjugation may be one of the factors contributing to the resistance of hamsters to intoxication by this PA. Copyright (C) 1998 Elsevier Science Ireland Ltd.

Full text of DOI:10.1016/S0378-4274(98)00152-0

Toxicology Letters 99 (1998) 127–137
Species differences in the hepatic microsomal enzyme
metabolism of the pyrrolizidine alkaloids
Jian-Ya Huan 1,a,b, Cristobal L. Miranda c, Donald R. Buhler a,c, Peter R. Cheeke a,b,
*
a Toxicology Program, Oregon State Uni6ersity, Cor6allis, OR 97331, USA
b Department of Animal Sciences, Oregon State Uni6ersity, Cor6allis, OR 97331, USA
c Department of Agricultural Chemistry, Oregon State Uni6ersity, Withycombe Hall 112, Cor6allis, OR 97331, USA
Received 20 April 1998; received in revised form 20 July 1998; accepted 24 July 1998
Abstract
Species differences in pyrrolic metabolites and senecionine (SN) N-oxide formation among eight animal species
(sheep, cattle, gerbils, rabbits, hamsters, Japanese quail, chickens, rats) varying in susceptibility to pyrrolizidine
alkaloid (PA) intoxication were measured in vitro by hepatic microsomal incubations. The results suggested that there
is not a strong correlation between the production of pyrrolic metabolites and susceptibility of animals to PA toxicity.
The rate of PA activation in hamsters, a resistant species, measured by formation of (9)6,7-dihydro-7-hydroxy-1-hy-
droxymethyl-5H-pyrrolizine (DHP) far exceeded the rate of SN N-oxide formation (detoxification) (DHP/N-oxide=
2.29). In contrast, SN N-oxide was the major metabolite in sheep, another resistant species, with much lower
production of DHP (DHP/N-oxide=0.26). The roles of cytochrome P450s and flavin-containing monooxygenases
(FMO) in bioactivation and detoxification of pyrrolizidine alkaloids (PA) were studied in vitro using sheep and
hamster hepatic microsomes. Chemical and immunochemical inhibition data suggested that the conversion of SN to
DHP is catalyzed mainly by cytochrome P450s (68–82%), whereas the formation of SN N-oxide is carried out largely
by FMO (55–71%). There also appeared to be a high rate of glutathione–DHP conjugation in hamster (63%) and
sheep (79%) liver microsomal incubation mixtures. Therefore, low rates of pyrrole metabolite production coupled
with glutathione conjugation in sheep may explain the resistance of sheep to SN, whereas the high rate of GSH-DHP
conjugation may be one of the factors contributing to the resistance of hamsters to intoxication by this PA. © 1998
Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Pyrrolizidine alkaloids; Cytochrome P450s; Flavin-containing monooxygenases; Glutathione conjugation;
Species
* Corresponding author. Tel.: +1 541 7371917; fax: +1 541 7374174; e-mail: cheekep@ccmail.orst.edu.
1 Present address: Department of Physiology and Pharmacology, Oregon Health Science University, 3181 SW Sam Jackson Park
Road, Portland, OR 97201, USA.
0378-4274/98/$ - see front matter © 1998 Elsevier Science Ireland Ltd. All rights reserved.
PII S0378-4274(98)00152-0

128
J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
1. Introduction
However, data from other research (Winter et al.,
1988) using gas chromatography/mass spectrome-
try (GS/MS) and MS/MS to identify the pyrrolic
metabolite (DHP), PA N-oxide, and hydrolytic
metabolite from hepatic microsomal incubation
showed that there was a high rate of DHP pro-
duction in guinea pigs, whereas the DHP produc-
tion in rats was relatively low. They suggested
that the resistance to certain PA such as SN in the
guinea pig is due to resistance to pyrrole toxicity
rather than low pyrrole formation. It seems that
the relative levels of DHP or N-oxide produced
from in vitro hepatic microsomal metabolism do
not reflect the resistance or susceptibility among
the animals. Furthermore, flavin-containing
monooxygenase (FMO), another hepatic microso-
mal monooxygenase, accounted for no more than
20% of the PA N-oxidase activity in Sprague–
Dawley rat liver microsomes (Williams et al.,
1989b), but this enzyme played a major role in
N-oxidation of SN in guinea pigs (Miranda et al.,
1991).
A number of indirect methods have been uti-
lized in studies aimed at determination of the
relative role of cytochrome P450s and FMO in the
metabolism of PA. These include the use of chem-
ical inhibitors (such as SKF-525A, etc.); pH or
thermal optima determination; and immunochem-
ical inhibition. Evidence based on chemical and
immunochemical inhibition data from guinea pigs
(Miranda et al., 1991) suggested that SN N-oxide
production was carried out largely by FMO in
guinea pig liver, whereas DHP formation from
SN was catalyzed mainly by liver cytochrome
P450.
Hydrolysis of a toxic pyrrolizidine ester to the
necine and necic acid moieties is another primary
detoxification pathway (Mattocks, 1986). Evi-
dence from chemical inhibition data (Dueker et
al., 1992a,b) showed that esterase hydrolysis con-
tributed 92% of moncrotaline metabolism in the
guinea pig, but only minimal activity was seen for
SN. Moreover, Chung and Buhler (1995) using
purified guinea pig hepatic carboxylesterases
showed that hepatic carboxylesterase GPH1 hy-
drolyzed [3H]-labeled jacobine (JB) and SN at
rates of 4.5 and 11.5 nmol/min per mg protein,
respectively, while carboxylesterase GPL1 had no
activity toward PAs.
The pyrrolizidine alkaloids (PAs) are a large
and important family of natural toxicants pro-
duced by a variety of plant species. Most PA
which produce toxic effects in livestock and hu-
mans are in the genera Senecio, Crotalaria, He-
liotropium and Echium (Cheeke, 1998). The
primary routes of PA metabolism have been es-
tablished in laboratory animals: dehydrogenation
to pyrrolic derivatives, conversion to N-oxides
and hydrolysis (Mattocks, 1986). PAs such as
senecionine require metabolic activation to pro-
duce toxicity. Dehydrogenation of PA to yield
unstable dehydropyrrolizidine alkaloids (PA
pyrroles) by hepatic monooxygenases located in
the endoplasmic reticulum is the classic pathway
of PA metabolism. The PA pyrroles either react
with cellular macromolecules or are hydrolyzed to
form a secondary pyrrole such as (9)6,7-dihy-
dro-7-hydroxy-1-hydroxymethyl-5H-pyrrolizine
(DHP). Hydrolysis of the ester groups by es-
terases such as carboxylesterase with excretion of
the acid and amino alcohol products, and forma-
tion and excretion of highly water soluble N-ox-
ides are detoxification mechanisms (Mattocks,
1986).
There is a marked variation in the susceptibility
of animal species to the toxic effects of PA expo-
sure. Sheep, guinea pigs, gerbils, rabbits, hamsters
and Japanese quail are highly resistant to PA
whereas rats, cattle, horses and chickens are
highly susceptible (Cheeke, 1998). The species dif-
ferences in resistance or susceptibility appear to
be primarily a consequence of differences in hep-
atic PA metabolism rather than to microbial PA
metabolism in the rumen or intestine (Cheeke,
1998). In general, those susceptible species such as
rats, mice, cattle and horses have a high rate of
pyrrole production, whereas PA-resistant animals
such as sheep, guinea pigs and Japanese quail
have low rates of pyrrole formation (White et al.,
1973; Shull et al., 1976), as assessed by colorimet-
ric analysis of pyrroles. In some cases, however,
for example with the rabbit, the in vitro pyrrole
production of a resistant species is high, whereas
with chickens, a relatively susceptible species, the
pyrrole formation is relatively low (Cheeke, 1998).

J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
129
There are conflicting reports on the correlation
of pyrrolic metabolite formation and the suscepti-
bility of animals to PA toxicity. In order to better
understand the mechanisms of species differences
in PA metabolism and the relative contribution of
hepatic microsomal enzymes involved in PA
metabolism as well as other possible biotransfor-
mation pathway, in vitro hepatic microsomal
metabolism and HPLC analytic methods devel-
oped by Kedzierski and Buhler (1986) were used
in this study to allow for the specific identification
and quantitation of important hepatic microsomal
metabolites of the PA senecionine. Hamsters and
sheep hepatic tissues were used for further study
because while both species are resistant to PA
toxicity, the mechanisms of resistance appear to
be different.
Department of Animal Science in the Oregon
State University were used. All animals were
housed in 12 h cycles of light and dark. They were
allowed food and tap water ad libitum. Sheep
(75ꢀ90 kg) and cattle (500ꢀ600 kg) were raised
on the farm of the Department of Animal Science
at the Oregon State University.
2.3. Microsome preparation
Four adult male animals (except there were
eight gerbils used, with livers from two animals
combined for each sample because of small liver
size) from each species were killed by cervical
dislocation. The livers were immediately removed
and homogenized in ice-cold 0.1 M potassium
phosphate buffer (pH 7.4) containing 1.15% KCl,
and 1.0 mM EDTA in a Potter-Elvehjem homoge-
nizer. Liver tissues of four male sheep and cattle
were collected immediately after animals were
slaughtered at the Meat Science Laboratory of
Oregon State University, and liver samples were
homogenized immediately. The homogenate was
centrifuged at 9000×g for 30 min at 4°C and the
supernatant was then centrifuged at 105000×g at
4°C for 60 min. The centrifuged supernatant was
saved as the cytosol fraction and the microsomal
pellet was washed once by resuspension in the
above homogenizing buffer and recentrifuged at
105000×g as before. The washed microsomal
pellet was resuspended in 0.1 M potassium phos-
phate buffer (pH 7.4) containing 20% glycerol
(v/v) and 0.1 mM EDTA and stored at −80°C
until used. All the experiments were conducted
within an 8-month time period.
2. Materials and Methods
2.1. Chemicals and antibodies
Senecionine (purified from extracts of Senecio
triangularis) was provided by Dr J.N. Roitman
(USDA-ARS, Western Regional Research Labo-
ratory, Albany, CA). Synthetic standards of DHP
and SN N-oxide were prepared as described previ-
ously (Kedzierski and Buhler, 1986). Glucose-6-
phosphate, glucose-6-phosphate dehydrogenase,
NADP+, glutathione (GSH), SKF-525A, methi-
mazole, phenylmethylsulfonyl fluoride (PMSF)
and thiourea were obtained from Sigma (St.
Louis, MO). Tri-o-cresyl phosphate (TOCP) was
from Eastman Kodak (Rochester, NY). The anti-
body for rat liver NADPH cytochrome P450 re-
ductase was raised in rabbits and has been
characterized previously (Williams et al., 1989a).
2.4. In 6itro metabolism of SN by microsomes
Microsomal incubations were carried out in a
mixture containing 0.5 mg microsomal protein
with the final concentration of 0.1 M potassium
phosphate buffer (pH 7.6), 1.0 mM EDTA, 0.5
mM SN and a NADPH-generating system (10.0
mM glucose-6-phosphate, 1.0 U/ml glucose-6-
phosphate dehydrogenase and 1.0 mM NADP+
in a total volume of 0.5 ml. The mixture was
equilibrated to 37°C for 5 min and the reaction
was initiated by the addition of SN (SN was
2.2. Animals
Golden Syrian hamsters (100–120 g), Sprague–
Dawley rats (330ꢀ350 g), New Zealand white
rabbits (2000ꢀ2300 g) and gerbils (22ꢀ25 g)
were obtained from the Laboratory Animal Re-
search Center of the Oregon State University.
Broiler chickens (2400ꢀ3000 g) and Japanese
quail (85ꢀ110 g) from the Poultry Facility of the

130
J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
dissolved in 0.1 N HCl/NaOH mixture with ad-
justing pH between 6.5ꢀ6.8. It was sonicated to
aid its solubility before use). After 1 h incubation
at 37°C with shaking (100 cycle/min), the reaction
was terminated by rapid cooling with ice water.
The reaction mixture was then centrifuged at
46000×g for 45 min at 4°C and an aliquot of the
supernatant was analyzed by HPLC with a PRP-1
column at u=220 nm wavelength as previously
described (Kedzierski and Buhler, 1986). When
chemical inhibitors (SKF-525A, methimazole,
PMSF, TOCP, all dissolved in DMSO and
thiourea in dH2O) were applied to inhibit the
reaction, the mixture was preincubated with in-
hibitor at 37°C for 20 min before the addition of
SN. An equivalent volume of DMSO was used in
the control reactions. This protocol was also
slightly modified when rabbit anti-rat NADPH
cytochrome P450 reductase IgG was used to in-
hibit the microsomal enzyme activity. The concen-
trations of antibodies used in hepatic microsome
inhibition were 10, 20 and 30 mg/nmol total P450,
respectively. The concentration of hepatic micro-
somes used in incubations were 0.25 and 0.35
nmol of total P450 from hamsters and sheep,
respectively. The hepatic microsomes were prein-
cubated with various concentrations of anti-rat
NADPH cytochrome P450 reductase antibodies
for 20 min at room temperature, after which the
buffer system and SN were added as described
above. Control incubations with an equivalent
amount of preimmune rabbit IgG were also prein-
cubated as described above.
above. For the controls, the microsomal incuba-
tions were carried out in the absence of GSH or
cytosol fraction.
2.6. Statistics
Statistical analyses were performed using the
statistical software base SAS (SAS Institute).
Data were assessed for homogeneity of variance
using analysis of variance procedure with means
compared by Student-Newman–Keuls (SNK) test
at PB0.05. Other data were assessed by using
two-T-test from Statgragh (Statistical Graphics
Corporation).
2.7. Other assays
Microsomal protein content was determined by
the method of Lowry et al. (1951) with bovine
serum albumin as the standard. Total cytochrome
P450 content was estimated using the method of
Omura and Sato (1964). The concentration of
reduced glutathione in liver cytosol was measured
with the method of Hissin and Hilf (1976).
3. Results
3.1. Species differences in PA metabolism
There is no strong correlation between PA
pyrrole formation and species susceptibility to PA
intoxication among tested species (PB0.05)
(Table 1). In some animals such as the hamster,
hepatic microsome DHP formation far exceeded
the rate of SN N-oxide formation. In contrast, SN
N-oxide was the major metabolite in other species
such as sheep. There appeared to be a different
rate in metabolizing the parent compound. The
data from Table 1 show that the remaining parent
SN expressed as SNA/SNB (SNA and SNB are
expressed as amount of SN after and before mi-
crosomal metabolism) varies from 23.1 to 91.5%
in hamster and Japanese quail, respectively. The
variations among species in DHP and N-oxide
formation are mainly dependent on the efficiency
of catalytic capability of liver microsomes to par-
ticular PA compounds.
2.5. Effect of GSH in the metabolism of SN
To assay the effect of GSH in the metabolism
of SN, hepatic microsomal incubations with dif-
ferent concentration of GSH were carried out.
Microsomes from sheep or hamsters were incu-
bated in the presence of 2.0 mM GSH in final
incubation mixture, 100 vl cytosol with average
GSH concentrations of 4856.3 vg/ml for hamsters
and 4731 vg/ml for sheep in the total of 500 vl
incubation mixture or 2.0 mM GSH plus 100 vl
cytosol in final incubation mixture, respectively.
After incubation, the production of DHP and SN
N-oxide was assayed by HPLC as described

J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
131
Table 1
In vitro metabolism of senecionine by liver microsomes to form DHP and N-oxide in different species
Species (four, male)
DHP (nmol/min per
N-oxide (nmol/min per mg)
Ratio of DHP/N-oxide
SNA/SNB (%)
mg)
Susceptible:
Cattle
Chicken
Rat
0.2390.03a,b
0.2290.03a,b
0.8590.22c
0.5990.04b,c
0.5290.05a,b
0.7090.20b,c
0.3890.02a,b
0.4490.06b
1.2590.08c
61.194.3c
84.591.2d
69.799.3c
Resistant:
Sheep
Hamster
Rabbit
0.4590.07b
3.5590.08f
1.7190.14e
0.0890.03a
1.3490.06d
0.46
1.7690.08e
1.5590.01e
0.8190.05c,d
0.2890.05a
0.9790.01d
0.87
0.2690.05a
2.2990.07d
2.1190.07d
0.3090.05a,b
1.3990.08c
0.53
27.591.5a
23.191.2a
30.092.3a
91.592.3d
42.293.0b
—
Japanese quail
Gerbil
Guinea piga
Mean9S.E. in the same column followed by different superscripts are different (PB0.05). Hepatic microsomal protein was 0.5 mg
in the microsomal mixtures. SNA and SNB are expressed as amount of SN after and before microsomal metabolism.
a Data were adapted from Chung and Buhler (1995).
3.2. Effect of chemical inhibitors in DHP and
N-oxide formation
3.3. Effect of anti-NADPH cytochrome P450
IgG in DHP and N-oxide formation
In order to examine the relative contribution
of cytochrome P450, FMO and esterase in the
metabolism of SN by sheep and hamster hepatic
microsomes, several chemicals have been used
with in vitro microsomal incubation. Table 2
shows that there was an almost complete inhibi-
tion in DHP formation with SKF-525A (PB
0.05) in both species, whereas SN N-oxidation
was only reduced by 7.6% in sheep and 34% in
hamsters. With two FMO inhibitors (methima-
zole and thiourea), the SN N-oxide formation
was reduced 80 and 71% in hamsters, respec-
tively, whereas in sheep SN N-oxide production
was reduced about 37.7 and 55.8%, respectively.
The involvement of hepatic esterase hydrolysis
in PA metabolism has also been examined. With
TOCP, an esterase inhibitor, DHP formation
was reduced dramatically in sheep (90.8%), but
with only 23% reduction in hamsters. In con-
trast, with PMSF, another esterase inhibitor, the
DHP production was decreased by 30.8 and
10.4% in sheep and hamsters, respectively. There
were no marked decreases in SN N-oxide for-
mation with the two inhibitors in either species
(Table 2).
In order to further determine the relative
contribution of cytochrome P450 and FMO in the
metabolism of PA, an immunoinhibition
experiment with rabbit anti-rat NADPH
cytochrome P450 reductase IgG was carried out.
The results from this experiment show (Fig. 1) that
there was
a maximum inhibition in DHP
production with 67.2 and 82.2% in sheep and
hamsters (PB0.05), respectively, whereas only 8.1
and 54.7% of SN N-oxide formation was eliminated
in sheep and hamsters, respectively.
3.4. Effect of GSH conjugation in DHP and
N-oxide formation
An indirect method for determining the effect of
GSH conjugation on DHP and N-oxide formation
was carried out by adding the reduced form of
GSH to the incubation mixture and then
comparing the DHP or N-oxide formation with
controls (Figs. 2 and 3). With 2.0 mM GSH in the
incubation mixture, the DHP production was
reduced 63 and 79% in hamsters and sheep
(PB0.05), respectively. In contrast, there was
little effect on SN N-oxide formation (P\0.05).

132
J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
Table 2
Chemical and antibody inhibition of DHP and N-oxide formation from senecionine by sheep and hamster liver microsomes
Inhibitors
DHP (as % of control)
Sheep
N-oxide (as % of control)
Hamster
Sheep
Hamster
SKF-525A (0.5 mM)
Methimazole (0.25 mM)
Thiourea (0.25 mM)
PMSF (1.0 mM)
N.D.
1.890.2*
N.D.
14.890.8*
89.691.3
76.992.5*
17.8*
92.493.5
62.394.9*
44.295.1*
74.093.7*
72.992.4*
91.9
65.993.0*
20.191.9*
29.092.2*
105.795.2
84.492.2
45.3*
6.190.7*
30.894.5*
69.296.4*
9.291.5*
32.7*
TOCP (0.1 mM)
Anti-P450 reductase IgGa
Results are given as the mean9S.E. of four animal liver microsomal incubations containing 0.5 and 1.0 mg microsomal protein
from hamster and sheep, respectively. The data are expressed as percentage of control. In control, the average production of DHP
and SN N-oxide from four hamster samples was 3.95 and 1.54 nmol/min per mg, respectively, the average values of DHP and SN
N-oxide from four sheep samples were 0.42 and 1.73 nmol/min per mg, respectively. The data with ‘*’ are significantly different from
the control (PB0.05). N.D. means the values are non-detectable.
a Each incubation was done in duplicate at a concentration of 30 mg/nmol P450 with preimmune IgG used as control.
4. Discussion
the hamster is resistant to oral administration of
PA (Shull et al., 1976), while this species is suscep-
tible to i.p. injection of PA (White et al., 1973).
With the data shown on Table 1, the hamster has
the highest capability to catalyze the SN to yield
the high rate of pyrrolic metabolites. The combi-
nation of these results could be explained such
that with i.p injection of PA and high capability
to metabolize the PA toward toxic metabolites,
the animal may accumulate a large amount of
toxic pyrrolic metabolites in a relatively short
time period, overwhelming the animal’s detoxifi-
cation system such as phase II conjugation en-
zymes, resulting in hepatic cytotoxicity. With the
oral administration of PA and high rate of
pyrrolic metabolite formation, the animal may be
capable of rapidly eliminating toxic metabolites
over the more prolonged period of PA absorption
without causing the cytotoxic effects. Another
example of PA resistance is that of Japanese
quail. In the case of this species, very little
metabolism of the parent alkaloids occurs (Buck-
master et al., 1997), as shown also in the present
experiment (Table 1); 91.5% of SN remained as
parent compound with Japanese quail micro-
somes. Thus Japanese quail are resistant to PA
because they form the pyrrolic metabolites at very
slow rate.
Species differences in the metabolism of PAs
and other xenobiotics are well known. Previous
studies (Chesney and Allen, 1973; White et al.,
1973; Shull et al., 1976; Cheeke, 1998) have indi-
cated that there was a correlation between the
pyrrolic metabolites from PA metabolism and
susceptibility among various species. Our data
(Table 1) in this study showed that some suscepti-
ble species such cattle and chickens have a low
DHP rather than a high DHP production, while
some resistant animals such as hamsters, rabbits
and gerbils have a relatively high pyrrole forma-
tion instead of low DHP production. These re-
sults suggested that there is no strong correlation,
at least in the in vitro hepatic microsomal
metabolism of SN using an HPLC method
(Kedzierski and Buhler, 1986), between the
pyrrole or N-oxide production and the resistance
or susceptibility among tested species (PB0.05).
It is interesting (Table 1) that resistant animals
such as sheep, hamsters, and rabbits catalyzed the
highest proportion of parent compound, whereas
those susceptible species such as cattle, chickens
and rats have a low capability toward metaboliz-
ing parent compounds. The dose of PA reaching
the target site and the catalytic capability of indi-
vidual species may play an important role in
susceptibility to PA toxicosis. As a good example,
The use of chemicals or antibodies as inhibitors
to identify the relative contribution of cytochrome

J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
133
Fig. 1. Immunoinhibition of SN metabolism in sheep and hamster liver microsomes by rabbit anti-rat NADPH cytochrome P450
reductase IgG. Hepatic microsomes from male adult sheep (0.35 nmol of total P450) or hamster (0.25 nmol of total P450) were
preincubated with rabbit anti-rat NADPH cytochrome P450 reductase IgG (10, 20 and 30 mg/nmol total P450, respectively), and
DHP and SN N-oxide were measured as described in Section 2. The means of experimental data were compared to their respective
control group mean. Results were expressed as a percentage of controls (preimmune rabbit IgG was used as control). In control, the
average production of DHP and SN N-oxide from sheep samples was 0.78 and 2.57 nmol/min per mg, respectively. The average
production of DHP and SN N-oxide from hamster samples was 2.06 and 0.89 nmol/min per mg, respectively.
bution toward the N-oxide formation. In an
earlier study, Miranda et al. (1991) reported that
in the guinea pig DHP formation was largely due
to cytochrome P450s. They also suggested that
the NADPH-dependent microsomal N-oxidation
of SN was carried out mainly by FMO in guinea
pig liver, lung and kidney. However, in another
study FMO accounted for no more than 20% of
the PA N-oxidase activity in Sprague–Dawley
rats (Williams et al., 1989b). In the present study
two chemical FMO inhibitors were used to deter-
mine the involvement of FMO in SN metabolism.
The data from Table 2 show that either methima-
zole or thiourea significantly reduced SN N-oxide
production in sheep and hamsters (PB0.05).
These results demonstrate that FMO may play a
major role in detoxification of PA in both species.
The hepatic esterase hydrolysis has been shown
to play an important role in PA metabolism
P450 and FMO has been reported in several
studies (Williams et al., 1989a,b; Miranda et al.,
1991). We found that 0.5 mM SKF-525A, a tradi-
tional P450 inhibitor, almost completely blocked
the DHP formation in both sheep and hamsters
(PB0.05), while SN N-oxide was only reduced by
7.6% in sheep and 34% in hamsters. This result
may suggest that cytochrome P450s mainly con-
tribute to the hepatic metabolism of SN to DHP.
The result also is in agreement with the study of
Miranda et al. (1991) in which they reported that
DHP formation was inhibited completely by
SKF-525A, but there was lack of inhibitory effect
on SN N-oxide formation in guinea pigs. These
results were further supported by immunoinhibi-
tion results from Table 2. In summary, we suggest
that the conversion of the SN to DHP is catalyzed
mainly by cytochrome P450s, and that these hep-
atic microsomal enzymes have much less contri-

134
J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
Fig. 2. The DHP and N-oxide formation from SN after incubation with sheep hepatic microsomes in presence of GSH, cytosol, or
GSH plus cytosol. Hepatic microsomes (0.35 nmol of total P450) from four male adult sheep were incubated with 0.5 mM
senecionine in presence of 2.0 mM GSH, 100 vl cytosol, or 2.0 mM GSH plus 100 vl cytosol, respectively, as described in Section
2. Results were given as the mean9S.E. of four animals and expressed as a percentage of controls. In control, the average
production of DHP and SN N-oxide from four sheep samples was 0.58 and 2.14 nmol/min per mg, respectively. The bars with ‘*’
are significantly different from the control (PB0.05).
DHP production was only decreased 30.8 and
10.4% in sheep and hamsters, respectively. The
SN N-oxide formation was inhibited about 26%
by PMSF in sheep whereas the production of SN
N-oxide was unaffected by this inhibitor. PMSF
has been presented earlier to be a potent inhibitor
for carboxylesterase (Dueker et al., 1992a). It was
shown that with 0.1 mM PMSF, 99% of guinea
pig hepatic carboxylesterase GPH1 activity was
inhibited. The differences in inhibition of esterase
hydrolysis to PA metabolism in two species may
be due to different specificity of the inhibitors
toward the different esterases in sheep and ham-
ster hepatic microsomes. The presence of esterlytic
cleavage in PA metabolism may partially explain
the sheep’s resistance to PA toxicosis. The high
rate of esterase hydrolysis may decrease the pro-
portion of PAs that are available for conversion
to toxic metabolites, and subsequently, a lower
toxicity is exhibited by the PAs. The absence of
significant hydrolytic activity in hamsters may
correlate to susceptibility in this animal when i.p.
(Dueker et al., 1992a). Previous research has
shown that hepatic microsomes hydrolyze
monocrotaline into retronecine and monocrotalic
acid in guinea pigs (Dueker et al., 1992b), while
the rat exhibits no such hydrolytic capabilities
(Lame et al., 1991). Chung and Buhler (1995) in a
recent study showed that hepatic carboxylesterase
GPH1 hydrolyzed [3H]-JB and [3H]-SN at rates of
4.5 and 11.5 nmol/min per mg protein, respec-
tively, while carboxylesterase GPL1 has no activ-
ity toward PA. In this study, two chemical
esterase inhibitors were used to identify the in-
volvement of hepatic esterases on PA metabolism
in sheep and hamsters. The results from Table 2
show that DHP formation was reduced dramati-
cally (90.2%) in sheep and only 23% in the ham-
sters with TOCP at a concentration of 0.1 mM.
Only slight effects were seen in N-oxide formation
from SN. This may suggest that sheep hepatic
microsomes possess significant hydrolytic activity
toward PA. However, when PMSF was used as
an inhibitor at the concentration of 1.0 mM, the

J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
135
Fig. 3. The DHP and N-oxide formation from SN after incubation with hamster hepatic microsomes in presence of GSH, cytosol,
or GSH plus cytosol. Hepatic microsomes (0.25 nmol of total P450) from four male adult hamsters were incubated with 0.5 mM
senecionine in presence of 2.0 mM GSH, 100 vl cytosol or 2.0 mM GSH plus 100 vl cytosol, respectively, as described in Section
2. Results were given as the mean9S.E. of four animals and expressed as a percentage of controls. In control, the average
production of DHP and SN N-oxide from four hamster samples was 5.89 and 2.60 nmol/min per mg, respectively. The bars with
‘*’ are significantly different from the control (PB0.05).
there was no significant difference between the in
vitro rate of reaction of GSH conjugation with JB
in the presence of rat GST and the non-enzymatic
rate. In contrast, the guinea pig GST catalyzed
the formation of JB-GSH at approximately twice
the non-enzymatic rate. Furthermore, the levels of
DHP and SN N-oxide in the different treatment
groups were also compared (Figs. 2 and 3). It
appears that the relative GSH concentration
present in the incubation mixture could signifi-
cantly affect the production of DHP (PB0.05)
but not SN N-oxide (P\0.05) in both species.
Moreover, the earlier study reported that GSH
could react with DHP directly in the chemical
synthesis of the conjugate (Robertson et al.,
1977). This suggests that when an adequate con-
centration of GSH is present, GSH is capable of
directly reacting with pyrrolic metabolites even if
there is a low GST activity. Reed et al. (1992)
reported that less than 20% of DHP was conju-
gated with GSH in vitro, whereas 95–100% of
dehydrosenecionine was conjugated with GSH in
a very short time period. From these results, they
administration of PAs was given (White et al.,
1973).
Conjugation with cellular nucleophiles such as
GSH has been shown to be an important detoxifi-
cation process with respect to pyrrolic metabolites
(Lame et al., 1990; Reed et al., 1992; Dueker et
al., 1994; Yan and Huxtable, 1995). The results
from this study (Figs. 2 and 3) show that when 2.0
mM GSH was added in hepatic microsomal incu-
bation mixtures, the DHP formation was signifi-
cantly reduced in sheep and hamster (PB0.05),
while SN N-oxide formation from SN was unaf-
fected (P\0.05). The presence of the cytosol
fraction alone without additional GSH in the
incubation mixtures also significantly decreased
the production of pyrrolic metabolites in sheep
and hamster (PB0.05). However, the presence of
cytosol fraction with additional GSH compared
to the absence of the cytosol fraction with addi-
tional GSH did not alter the reduction of pyrrole
formation in either species (P\0.05). Similar re-
sults have been reported by Dueker et al. (1994).
Using purified rat hepatic GST, they found that

136
J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
E.O., Pierson, M.L., Shull, L.R., 1997. Response of
Japanese quail to dietary and injected pyrrolizidine (Sene-
cio) alkaloid. J. Anim. Sci. 45, 1322–1325.
suggested that SN could be catalyzed initially into
dehydrosenecionine, which then is either hy-
drolyzed to DHP or reacts with GSH to form the
DHP-GSH conjugate with cleavage of the senecic
acid. Thus, our results shown in Figs. 2 and 3
suggest that GSH conjugation may play an impor-
tant biological role in the detoxification of PA in
sheep and hamsters.
Cheeke, P.R., 1994. Review of the functional and evolutionary
roles of the liver in the detoxification of poisonous plants,
with special reference to pyrrolizidine alkaloids. Vet. Hu-
man Tox. 36, 240–247.
Cheeke, P.R., 1998. Natural Toxicants in Feeds, Forages, and
Poisonous Plants. Interstate, Danville, IL.
Chesney, C.F., Allen, J.R., 1973. Resistance of the guinea pig
to pyrrolizidine alkaloid intoxication. Toxicol. Appl. Phar-
macol. 26, 385–392.
Chung, W.G., Buhler, D.R., 1995. Major factors for the
susceptibility of guinea pig to the pyrrolizidine alkaloid
jacobine. Drug Metab. Dispos. 23, 1263–1267.
Dueker, S.R., Lame, M.W., Jones, D., Morin, D., Segall, H.J.,
1994. Glutathione conjugation with the pyrrolizidine alka-
loid, jacobine. Biochem. Biophy. Res. Commun. 198, 516–
522.
Dueker, S.R., Lame, M.W., Morin, D., Wilson, D.W., Segall,
H.J., 1992a. Guinea pig and rat hepatic microsomal
metabolism of monocrotaline. Drug Metab. Dispos. 20,
275–280.
Dueker, S.R., Lame, M.W., Segall, H.J., 1992b. Hydrolysis of
pyrrolizidine alkaloids by guinea pig hepatic car-
boxylesterases. Toxicol. Appl. Pharmacol. 117, 116–121.
Hissin, P.J., Hilf, R., 1976. A fluorometric method for deter-
mination of oxidized and reduced glutathione in tissues.
Anal. Biochem. 74, 214–226.
Kedzierski, B., Buhler, D.R., 1986. Method for determination
of pyrrolizidine alkaloids and their metabolites by high
performance liquid chromatography. Anal. Biochem. 152,
59–65.
Lame, M.W., Jones, A.D., Morin, D., Segall, H.J., 1991.
Metabolism of [14C]monocrotaline by isolated perfused rat
liver. Drug Metab. Dispos. 19, 516–524.
In conclusion, our study demonstrates that there
is no strong correlation between the production of
pyrrolic metabolites and the susceptibility of ani-
mals to PA toxicity among the tested species. The
species differences in susceptibility to PA toxicity
appear to be dependent on the individual species
hepatic microsomal enzymes catalytic capability.
Chemical and antibody inhibition data suggest that
the conversion of SN to DHP is catalyzed mainly
by cytochrome P450s, whereas the formation of SN
N-oxide is carried out largely by FMO in sheep and
hamsters. The involvement of esterase hydrolysis
may have a significant impact on DHP formation
in sheep with much less effect in hamsters. The
indirect evidence shows that there is a high rate of
GSH-DHP conjugation in both species. It has been
proposed (Wachenheim et al., 1992) that the resis-
tance of sheep to PA toxicosis is due to microbial
metabolism of PA in the rumen. In the present
study, we worked with a variety of PA resistant
species, most of which (hamsters, gerbils, Japanese
quail, rabbits) are not ruminants. We believe that
in these species, resistance to PA toxicosis is asso-
ciated with hepatic PA metabolism. Cheeke (1994,
1998) discussed in detail his viewpoints on mecha-
nisms of PA resistance as a reflection of evolution-
ary history and feeding strategies of herbivores.
Lame, M.W., Morin, D., Jones, D., Segall, H.J., Wilson,
D.W., 1990. Isolation and identification of a pyrrolic glu-
tathione conjugate metabolite of the pyrrolizidine alkaloid
monocrotaline. Toxicol. Lett. 51, 321–329.
Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J.,
1951. Protein measurement with the Folin phenol reagent.
J. Biol. Chem. 193, 265–275.
Mattocks, A.R., 1986. Chemistry and Toxicology of
Pyrrolizidine Alkaloids. Academic Press, Orlando, FL.
Miranda, C.L., Chung, W.G., Reed, R.L., Zhao, X., Hender-
son, M.C., Wang, J.L., Williams, D.E., Buhler, D.R., 1991.
Flavin-containing monooxygenase: a major detoxifying en-
zyme for the pyrrolizidine alkaloid senecionine in guinea
pig tissues. Biochem. Biophys. Res. Commun. 178, 546–
552.
Omura, T., Sato, R., 1964. The carbon monoxide-binding
pigment of liver microsomes. I. Evidence for its hemo-
protein reagent. J. Biol. Chem. 239, 2370–2378.
Reed, R.L., Miranda, C.L., Kedzierski, B., Henderson, M.C.,
Buhler, D.R., 1992. Microsomal formation of a pyrrolic
alcohol glutathione conjugate of the pyrrolizidine alkaloid
senecionine. Xenobiotica 22, 1321–1327.
Acknowledgements
We thank Marilyn C. Henderson of the Depart-
ment of Agricultural Chemistry, Oregon State
University, for her excellent technical support dur-
ing the experimental period.
References
Buckmaster, G.W., Cheeke, P.R., Arscott, G.H., Dickinson,

J.-Y. Huan et al. / Toxicology Letters 99 (1998) 127–137
137
Robertson, K.A., Seymour, J.L., Hsia, M.T., Allen, J.R.,
1977. Covalent interaction of dehydroretronecine, a car-
cinogenic metabolite of the pyrrolizidine alkaloid
monocrotaline, with cysteine and glutathione. Cancer Res.
37, 3141–3144.
Williams, D.E., Reed, R.L., Kedzierski, B., Dannan, G.A.,
Guengerich, F.P., Buhler, D.R., 1989a. Bioactivation and
detoxification of the pyrrolizidine alkaloid senecionine by
cytochrome P450 enzymes in rat liver. Drug Metab. Dis-
pos. 17, 387–392.
Shull, L.R., Buckmaster, G.W., Cheeke, P.R., 1976. Factors
influencing pyrrolizidine (Senecio) alkaloid metabolism:
Species, liver sulfhydryls and rumen fermentation. J. Anim.
Sci. 43, 1247–1253.
Wachenheim, D.E., Blythe, L.L., Craig, A.M., 1992. Charac-
terization of rumen bacterial pyrrolizidine alkaloid trans-
formation in ruminants of various species. Vet. Human.
Tox. 34, 513–517.
Williams, D.E., Reed, R.L., Kedzierski, B., Ziegler, D.M.,
Buhler, D.R., 1989b. The role of flavin-containing
monooxygenase in the N-oxidation of the pyrrolizidine
alkaloid senecionine. Drug Metab. Dispos. 17, 380–386.
Winter, C.K., Segall, H.J., Jones, A.D., 1988. Species differ-
ences in the hepatic microsomal metabolism of the
pyrrolizidine alkaloid senecionine. Comp. Biochem. Phys-
iol. 90C, 429–433.
White, I.H.N., Mattocks, A.R., Butler, W.H., 1973. The con-
version of the pyrrolizidine alkaloid retrorsine to pyrrolic
derivatives in vivo and in vitro and its acute toxicity to
various animal species. Chem. Biol. Interact. 6, 207–218.
Yan, C.C., Huxtable, R.J., 1995. Relationship between glu-
tathione concentration and metabolism of the pyrrolizidine
alkaloid, monocrotaline, in the isolated, perfused liver.
Toxicol. Appl. Pharmacol. 130, 132–139.
.