Characterization of metabolism of (+)-praeruptorin B and (+)-praeruptorin E in human and rat liver microsomes by liquid chromatography coupled with ion trap mass spectrometry and time-of-flight mass spectrometry

Article abstract of DOI:10.1002/rcm.4922

Peucedani Radix is a Chinese medicinal herb noted for its effects on treatments of respiratory and pulmonary disorders. As a part of a systematic pharmacokinetic evaluation of the herb in our laboratory, the present study investigated, for the first time, the metabolic profile of (+)-praeruptorin B (dPB) and (+)-praeruptorin E (dPE), two main bioactive constituents of Peucedani Radix in pooled liver microsomes of rats (RLMs) and humans (HLMs). dPE was eliminated faster than dPB in both species. The incubation of dPB with RLMs and HLMs resulted in eight (B1-B8) and nine (B1-B9) metabolites, respectively, while both RLMs and HLMs converted dPE into 13 metabolites (E1-13). Structures of all the metabolites were proposed through comparing their mass data obtained via tandem mass spectrometry on an MSD ion trap system (IT-MS/MS) coupled with high-resolution mass measurement by time-of-flight mass spectrometry (TOF-MS) with those of the respective parent compound. B1 and E1 were unambiguously identified as (-)-cis-khellactone. The formations of all the metabolites were NADPH-dependent. Oxidation and hydrolysis were demonstrated to be two predominant metabolic pathways of dPB and dPE. Oxidation initiated at either the C-3′ or C-4′ substituent, while hydrolysis only started from the C-3′ substituent. Fragmentation of all metabolites followed similar pathways to those of the parent pyranocoumarins. The information on metabolic properties of dPB and dPE and the mass fragmentation profiles of their metabolites obtained in the present study will aid in characterization of metabolic profiles of other angular-type pyranocoumarins and further investigation of in vivo fates of these pyranocoumarins and the herb. Copyright

Full text of DOI:10.1002/rcm.4922

Research Article
Received: 4 September 2010
Revised: 25 December 2010
Accepted: 27 December 2010
Published online in Wiley Online Library: 2011
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
wileyonlinelibrary.com) DOI: 10.1002/rcm.4922
(
Characterization of metabolism of (þ)-praeruptorin B and
(
þ)-praeruptorin E in human and rat liver microsomes by liquid
chromatography coupled with ion trap mass spectrometry and
time-of-flight mass spectrometry
1
1
1
1
1,2
Yue-Lin Song , Ru Yan *, Wang-Hui Jing , Hai-Yu Zhao and Yi-Tao Wang *
1
Institute of Chinese Medical Sciences, University of Macau, Taipa, Macao, China
School of Chinese Materia Medica, Beijing University of Chinese Medicine, Beijing, China
2
Peucedani Radix is a Chinese medicinal herb noted for its effects on treatments of respiratory and pulmonary
disorders. As a part of a systematic pharmacokinetic evaluation of the herb in our laboratory, the present study
investigated, for the first time, the metabolic profile of (R)-praeruptorin B (dPB) and (R)-praeruptorin E (dPE), two
main bioactive constituents of Peucedani Radix in pooled liver microsomes of rats (RLMs) and humans (HLMs). dPE
was eliminated faster than dPB in both species. The incubation of dPB with RLMs and HLMs resulted in eight
(
(
B1–B8) and nine (B1–B9) metabolites, respectively, while both RLMs and HLMs converted dPE into 13 metabolites
E1–13). Structures of all the metabolites were proposed through comparing their mass data obtained via tandem mass
spectrometry on an MSD ion trap system (IT-MS/MS) coupled with high-resolution mass measurement by time-
of-flight mass spectrometry (TOF-MS) with those of the respective parent compound. B1 and E1 were unambiguously
identified as (S)-cis-khellactone. The formations of all the metabolites were NADPH-dependent. Oxidation and
hydrolysis were demonstrated to be two predominant metabolic pathways of dPB and dPE. Oxidation initiated at
either the C-3( or C-4( substituent, while hydrolysis only started from the C-3( substituent. Fragmentation of all
metabolites followed similar pathways to those of the parent pyranocoumarins. The information on metabolic
properties of dPB and dPE and the mass fragmentation profiles of their metabolites obtained in the present study will
aid in characterization of metabolic profiles of other angular-type pyranocoumarins and further investigation of in
vivo fates of these pyranocoumarins and the herb. Copyright ß 2011 John Wiley & Sons, Ltd.
Peucedani Radix (Chinese name Qian-hu), the dried roots of
Peucedanum praeruptorum Dunn (Apiaceae), is a herbal
medicine noted for its cough-relieving and sputum-reducing
effects and has been traditionally utilized as antitussive and
hepatoprotective, anti-tumor and anti-platelet aggregation
activities.
[
11–16]
In contrast to extensive pharmacologic studies, the
pharmacokinetic properties including the metabolism of
the coumarin constituents of Peucedani Radix have scarcely
been reported, hence their contribution to the herbal activities
and in vivo active forms remain unclear. As a part of an
ongoing project on systematic pharmacokinetic evaluation of
Peucedani Radix in our laboratory, metabolic profiling of the
main angular-type pyranocoumarins in rat liver microsomes
was carried out to aid their correlation in the in vivo fates of
the main constituents with the reported beneficial actions of
the herb on the rat. Further, the metabolic properties of the
compounds in human liver microsomes were characterized
in parallel to determine whether the data obtained from the
rat are extrapolative to humans.
[
1]
mucolytic agents for centuries in China. In addition to the
well-known actions on respiratory disorders, modern
pharmacologic evaluation of Peucedani Radix also revealed
significant beneficial effects of the herb on pulmonary
[
2–4]
[5,6]
hypertension in animal models
and patients.
So far, more than 50 coumarins have been isolated
from Peucedani Radix, including (ꢀ)-praeruptorin A, (ꢀ)-
praeruptorin B, (þ)-praeruptorin A, (þ)-praeruptorin B,
[
7–10]
(
þ)-praeruptorin E, etc.
Most of these coumarins belong
to angular-type pyranocoumarin, which has a khellactone
skeleton with different substituents at the C-3 and C-4
0
0
positions. Numerous pharmacologic studies have been
carried out on in vitro and in vivo models and demonstrated
diverse biological activities of these pyranocoumarins,
including anti-asthma, vasorelaxant, cardiac protective,
High-performance liquid chromatography/ion trap tan-
dem mass spectrometry (HPLC/IT-MS/MS) is a sensitive
and versatile analytical system that applies the ion trap
technique to identify both large and small molecules from
complex biological systems. The advantages of IT-MS/MS lie
in the ability to perform multiple stages of mass spectrometry
n
(
MS ) to facilitate sensitive and specific structural identifi-
*
Correspondence to: R. Yan or Y.-T. Wang, Institute of Chinese
Medical Sciences, University of Macau, Macao SAR, China.
E-mail: ruyan@umac.mo; ytwang@umac.mo
[
17]
cation.
In comparison, HPLC coupled with time-of-flight
mass spectrometry (HPLC/TOF-MS) uses the differences in
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
Copyright ß 2011 John Wiley & Sons, Ltd.

Y.-L. Song et al.
transit time through a drift region to separate ions of different
masses, thus can provide accurate mass weight and the
molecular formula of the analyte to assist structural
[
18]
identification.
IT-MS/MS and TOF-MS have been widely
adopted in combination to characterize metabolites in
biological materials.
(
þ)-Praeruptorin B (dPB, Fig. 1) and (þ)-praeruptorin E
(
dPE, Fig. 2), the two main angular-type pyranocoumarins in
Peucedani Radix, have exhibited anti-tumor and calcium
channel antagonistic activities and were considered to
[
19,20]
contribute to the herbal activities.
The only structural
difference between these two compounds is the substituent at
0
the 4 position: it is an angeloyl group in dPB yet an isovaleryl
group in dPE. In the present study, the metabolic profiles of
dPB and dPE in rat and human liver microsomes were
investigated, for the first time, to find out whether this
structural difference results in distinct metabolic profiles. The
main focus of the study was on the elucidation of metabolic
pathways of dPB and dPE using HPLC/IT-MS/MS and
HPLC/TOF-MS and characterization of the metabolic
difference between the two compounds and the species
difference between rats and humans if any.
Figure 2. Representative HPLC chromatograms of incubates
of dPE with RLMs in the absence (A) and presence (B) of a
NADPH-regenerating system, or with HLMs in the presence
of a NADPH-regenerating system (C), at 323 nm.
EXPERIMENTAL
University (Guangzhou, China). The content of microsomal
Materials
[
22]
proteins was determined by Lowry’s method.
Both HLMs
dPB, dPE and (þ)-praeruptorin A (dPA) (purity >98%) were
purchased from Shanghai Traditional Chinese Medicine
Research Centre (Shanghai, China). Pooled human liver
microsomes (HLMs, from 15 healthy donors) were purchased
from Sigma-Aldrich (St. Louis, MO, USA). A pool of rat liver
microsomes (RLMs) was prepared from 30 healthy Sprague-
Dawley rats (male, 250–300 g) by differential centrifugation
and RLMs were stored at ꢁ808C until use.
Formic acid and methanol (both HPLC grade) were
purchased from Merck (Darmstadt. Germany). Ultra-pure
water was obtained from a Milli-Q plus water purification
system (Millipore, Bedford, MA, USA). Glucose 6-phosphate
(G-6-P), glucose-6-phosphate dehydrogenase (G-6-PD) and
nicotinamide adenine dinucleotide phosphate (NADP)
were purchased from Sigma-Aldrich (St. Louis, MO, USA).
All other chemicals were analytical grade and obtained
commercially.
[
21]
according to a standard procedure reported previously
at the School of Pharmaceutical Sciences, Sun Yat-sen
dPB and dPE metabolism in RLMs and HLMs
dPB or dPE (25 mM, final concentration) was incubated with
1
mg/mL of microsomal proteins in 0.1 M potassium
phosphate buffer (pH 7.4) in a final incubation volume of
00 mL, at 378C for 60 min. Reactions were initiated by
addition of a NADPH-regenerating system (1 mM NADP,
mM glucose 6-phosphate, 1 U/mL of glucose-6-phosphate
2
1
dehydrogenase) into the reaction mixtures and
terminated with 200 mL of ice-cold MeOH. The sample was
thoroughly mixed on a vortex mixer then centrifuged at
1
5 000 g for 10 min. An aliquot (70 mL) of the supernatant was
injected for HPLC/IT-MS/MS or HPLC/TOF-MS analysis.
Parallel incubations with no microsomes or no NADPH-
regenerating system, and 0 min incubations, served as
controls.
Preparation of (S)-cis-khellactone
Figure 1. Representative HPLC chromatograms of incubates
of dPB with RLMs in the absence (A) and presence (B) of a
NADPH-regenerating system, or with HLMs in the presence
of a NADPH-regenerating system (C), at 323 nm.
A basic hydrolysis study of dPA, another main pyranocou-
marin constituent from Peucedani Radix, was carried out
[
23]
according to a previous report
to aid structural identifi-
wileyonlinelibrary.com/journal/rcm
Copyright ß 2011 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730

Metabolism of praeruptorin B and E in liver microsomes
cation of B1 (one of dPB metabolites) or E1 (one of dPE
metabolites) formed by liver microsomes of rats and humans.
In brief, dPA (5 mg/mL in tetrahydrofuran) was added
into 100 mL of 0.5 M KOH and the reaction mixture was
RESULTS AND DISCUSSION
HPLC/MS/MS and TOF-MS analysis of dPB, dPE and
(
S)-cis-khellactone
stirred at 608C for 30 min. Then 10% H
acidify the reaction solution (pH ꢂ6) and stirred at room
temperature for 120 min before extraction with CHCl
2 4
SO was added to
Under the developed analytical conditions, dPB and dPE were
eluted at 38.0 min and 42.6 min, respectively, and most of
metabolites and their respective parent compounds obtained
baseline separation (Figs. 1–4). The accurate mass, estimated
elemental composition, error in terms of ppm and sequential
3
.
The CHCl layer was collected, evaporated and the residue
3
reconstituted with methanol. The main product P1 was
isolated using silica gel column chromatography (petroleum
ether/ethyl acetate, 4:1 ! 1:1, v/v) and identified using
HPLC/MS/MS and polarimetric analysis (PerkinElmer, The
Netherlands).
1
2
mass (MS and MS ) data of the parent compounds are listed
in Tables 1 and 2 and provided additional information for
identification of their metabolites produced in liver micro-
somes of rats and humans.
1
The MS spectrum of dPB exhibited adduct ions at m/z 444
þ
þ
(
[MþNH ] ) and m/z 449 ([MþNa] ), corresponding to the
4
HPLC/MS analysis
molecular weight of 426 Da for dPB. And TOF-MS analysis
revealed a sodiated adduct ion at m/z 449.1567 and provided
All samples were analyzed on an Agilent series 1200
2
an elemental composition of C24
H
26
O
7
. MS of the sodiated
(
Agilent Technologies, USA) liquid chromatography,
þ
adduct ion at m/z 449 ([MþNa] ) revealed two major product
which was equipped with a vacuum degasser, a binary
pump, an autosampler and a diode-array detector (DAD)
system, and operated with Agilent ChemStation B 3.0
software. The analytical column was an ODS reversed-phase
ions at m/z 349 and m/z 327 resulting from the loss of the
4 7 4 7
C H COOH and C H COONa moieties, respectively, from
0
0
the C-3 or C-4 position (Table 1, Fig. 5). The characteristic ion
at m/z 245 corresponds to the cleavage of an angeloyl group
from m/z 327. Moreover, the characteristic ion at m/z 227
C
18 column (250 mm ꢃ 4.6 mm i.d., particle size 5 mm, Agilent,
USA). The column temperature was maintained at 358C.
corresponds to loss of a water molecule and a C
from the khellactone skeleton or an angelic acid molecule
COOH) from m/z 327.
dPE was eluted at 42.6 min. Its mass spectra revealed a
fragmentation pattern similar to that of dPB: adduct ions at
3 6
H molecule
The mobile phase consisted of 0.1% aqueous formic acid
(
A) and methanol (B). Samples were analyzed using a
4 7
(C H
gradient elution as follows: 0–4 min, 25%–50% B; 4–16 min,
5
2
7
0%–53% B; 16–17 min, 53%–56% B; 17–20 min, 56% B;
0–23 min, 56%–69% B; 23–40 min, 69%–76% B; 40–41 min,
6%–100% B; 41–44min, 100% B. The flow rate was set
þ
þ
m/z 446 ([MþNH ] ) and m/z 451 ([MþNa] ) observed in the
4
1
MS spectrum correspond to a molecular weight of 428 Da for
at 1.0 mL/min and the injection volume was 70 mL. The
sample chamber temperature was set at 48C. The respective
parent compound and its metabolites were monitored at
dPE. Accurate mass data showed the adduct ions at m/z
4
46.2188 and m/z 451.1737 and gave a molecular formula of
2
24 28 7
C H O . The MS spectrum of the sodiated adduct ion
3
23 nm.
exhibited product ions at m/z 349 and m/z 327, which indicate
Ion trap tandem mass spectrometry analysis (IT-MS/MS)
0
the loss of C H COOH and C H COONa from the C-4
was performed on an Agilent MSD Ion Trap system (Palo
Alto, CA, USA) coupled to an electrospray ionization (ESI)
interface. A quarter of the column eluent was directly
introduced into the ESI interface through a PEEK tubing
4
9
4
9
position, respectively. The presence of the characteristic ions
at m/z 245 and m/z 227 suggests further loss of a C
4
H
6
CO
0
moiety and a C H COOH molecule at the C-3 position,
4
7
respectively (Table 2, Fig. 6).
(
0.13 mm i.d.). Nitrogen was used as both the nebulizing
The sequential mass spectrometric analysis of dPB and dPE
gas at 45 psi and the drying gas at a flow rate of 8 L/min
at a temperature of 3508C. The mass spectrometer was
operated in positive ion mode over a mass range of
revealed common characteristic product ions at m/z 349, m/z
2
3
27, m/z 245 and m/z 227 in MS spectra (Tables 1 and 2),
supporting removal of two RCOOH units substituted at the
1
00–500 m/z. Helium was used as the collision gas for the
0
0
n
C-3 and C-4 positions, respectively. In an MS study of
pyranocoumarin constituents from Peucedani Radix, the
authors proposed that the fragmentation of an angular-type
tandem mass spectrometry experiments. Fragmentation was
induced with the resonant excitation amplitude of 0.6 V,
followed by an isolation of the precursor ion over a selected
mass window of 1 Da. Data acquisition was performed
using Agilent ChemStation software (Agilent Technologies,
MA, USA).
0
pyranocoumarin usually initiates from the 4 position when
0
0
the two substituents connected to the C-3 and C-4 positions
were different.
[
24]
In the present study, the mass fragmenta-
tion preference of dPE agrees well with the speculation of Tao
High-resolution mass spectrometry (HR-MS) analysis
was performed on an Agilent 6210 TOF mass spectrometer
[
24]
et al.,
whereas the mass data obtained for dPB could not
definitely support such mass fragmentation preference due to
the identical groups substituted at the C-3 and C-4 positions
of this compound.
(
Agilent Technologies, USA) that operated in positive ESI
mode. The parameters of the ion source were set as capillary
voltage of 4500 V and drying gas (N ) pressure of 20 psi.
0
0
2
When dPA, an angular-type pyranocoumarin which has an
MS conditions were as follows: mass range, 100–1000 m/z;
fragmentor voltage, 250 V; skimmer voltage, 60 V. The
instrument performed automatic runs by introducing a
constant flow of calibrating solution containing the ESI-TOF
tuning reference solution and API-TOF reference mass
solution (Agilent Technologies, USA).
0
angeloyl moiety and an acetyl moiety substituted at the C-3
0
and C-4 positions, respectively, was treated with KOH,
three products were yielded from the basic hydrolysis
(
Fig. 7). The product P1 was eluted at 11.9 min and exhibited
þ
þ
adduct ions at m/z 285 ([MþNa] ) and m/z 301 ([MþK] ) and
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

Y.-L. Song et al.
Figure 3. Representative total ion current chromatogram and extracted ion chromatograms of dPB and its metabolites
produced following incubation with RLMs in the presence of a NADPH-regenerating system.
þ
the protonated ion at m/z 263 ([MþH] ), indicating a
from removal of both angeloyl and acetyl moieties from the
C-3 and C-4 positions of dPA.
0
0
molecular weight of 262 Da. Accurate mass analysis indicated
2
that the elemental composition of P1 is C14
H
14
O
5
. MS of the
protonated ion resulted in two product ions: m/z 245
corresponding to loss of a water molecule and m/z 203 for
loss of a water molecule plus a C H molecule. These mass
3 6
In vitro metabolism of dPB and dPE in liver microsomes of
rats and humans
spectra, the retention time and UV spectrum of P1 were
identical with those of (ꢁ)-cis-khellactone, which was
prepared from dPA by the same method and unambiguously
identified on the basis of HPLC/MS/MS and NMR analyses
The representative HPLC-UV chromatograms, total ion
current (TIC) chromatograms and extracted ion current
(EIC) chromatograms of samples obtained from reactions of
dPB and dPE with liver microsomes are shown in Figs. 1–4.
[
23]
1
2
by Wu and co-workers.
Thus, P1 produced in the present
The MS and MS spectra of dPB and dPE metabolites are
study was assigned as (ꢁ)-cis-khellactone, which was derived
shown in Figs. 5 and 6 and chromatographic and mass
wileyonlinelibrary.com/journal/rcm
Copyright ß 2011 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730

Metabolism of praeruptorin B and E in liver microsomes
retention times, UV spectra, mass spectra and accurate
mass data (Figs. 1 and 3, Table 1), indicating that dPB
undergoes similar biotransformations in liver microsomes of
both species.
1
Similar to the MS spectrum of dPB, sodiated adduct ions
þ
(
[MþNa] ) at significant intensity were observed for B2–B8;
2
thus the sodiated adduct ions were used to produce MS
spectra of these metabolites (Fig. 3, Table 1). In contrast, a
þ
strong signal of the protonated ion ([MþH] ) was observed in
1
2
MS of B1 and applied for MS scan of this metabolite. In
addition, the characteristic ions at m/z 245, m/z 227 and m/z
2
2
03 were commonly observed in MS spectra of most
metabolites. The identity of B9, the metabolite only found
in HLMs, could not be elucidated based on the available mass
data (data not shown).
1
B1 was eluted at 11.9 min. Its MS spectrum showed
þ
the protonated ion at m/z 263 ([MþH] ), sodiated adduct
þ
ion at m/z 285 ([MþNa] ) and potassiated adduct ion at m/z
þ
3
01 ([MþK] ), suggesting that the molecular weight of B1
is 262 Da. TOF-MS also provided the accurate mass data
þ
þ
at m/z 263.0915 ([MþH] ), m/z 285.0731 ([MþNa] ) and
þ
m/z 301.0468 ([MþK] ), respectively, and gave a formula of
14 14 5
C H O , which corresponds to removal of both angeloyl
0
0
moieties from the C-3 and C-4 positions of dPB (Fig. 5). The
characteristic ions at m/z 245 and m/z 203 present in the MS
2
spectrum correspond to the loss of a water molecule, a water
molecule together with a C H molecule, respectively, from
3
6
the protonated ion. Thus, B1 should have resulted from
0
0
hydrolytic removal of both C-3 and C-4 angeloyl groups
from dPB by liver microsomal proteins. Furthermore, B1
showed identical retention times, UV and mass spectra, and
accurate mass data to those of (ꢁ)-cis-khellactone (P1), the
product yielded from basic hydrolysis of dPA (Fig. 7). Finally,
B1 was identified as (ꢁ)-cis-khellactone.
B2 was observed at a retention time of 22.2 min and
þ
exhibited adduct ions at m/z 383 ([MþNa] ) and m/z 399
þ
(
[MþK] ), suggesting a molecular weight of 360 Da, 66 Da
Figure 4. Representative total ion current chromatogram
and extracted ion chromatograms of dPE and its metabolites
produced following incubation with RLMs in the presence of
a NADPH-regenerating system.
lower than its parent compound (Figs. 1 and 5). Accurate
20 7
mass data supported a molecular formula of C19H O
Table 1). Further, the MS of the sodiated ion at m/z 383
2
(
revealed a major product ion at m/z 267, which was 116 mass
units less than the precursor ion and corresponds to loss of a
mono-oxidized angelic acid molecule. Thus B2 was specu-
lated as a product formed from cleavage of one angeloyl
group concomitant with mono-oxidation on the other
angeloyl group of dPB (Fig. 8). The oxidation on the side
spectrometric data are summarized in Tables 1 and 2 and
compared with those of the parent compounds. Metabolites
of dPB and dPE showed similar mass fragmentation profiles
to their respective parent compound.
0
00 000
000
The incubation of dPB with rat liver microsomal proteins
resulted in eight metabolites (B1–8), while one additional
metabolite (B9) that eluted between B4 and B5 was observed
in HLMs (Fig. 1). A total of 13 metabolites (E1–13) of dPE were
generated in liver microsomes of both species (Fig. 2). All
the metabolites were eluted earlier than the respective parent
compound and formed in a NADPH-dependent manner,
indicating that the metabolites are more polar than their
respective parent compound and their formations are
catalyzed by cytochrome P450 isozyme(s).
chain could be hydroxylation at the C-3 , 4 , or 5 position,
or cyclo-oxidation at the double bond.
1
2
B3 and B5 showed similar MS and MS spectra yet
different retention times (B3, 27.1 min; B5, 28.8 min),
1
suggesting that these two metabolites are isomers. The MS
þ
spectra showed adduct ions at m/z 362 ([MþNH
4
] ) and m/z
þ
367 ([MþNa] ), indicating the molecular weights of B3 and
B5 are 344 Da and 82 Da lower than dPB and 16 Da lower than
B2. TOF-MS analysis revealed identical sodiated adduct ions
at m/z 367.1154 and provided the same molecular formula
2
of C H O for both metabolites. In the MS spectrum of
1
9
20
6
the sodiated adduct ion, the presence of the predominant
product ion at m/z 267 indicates the loss of one C COOH
The metabolites of dPB formed in rat and human liver
microsomes
4 7
H
molecule. Thus B3 and B5 might be positional isomers
0
The eight metabolites (B1–B8) of dPB detected in RLMs were
identical with those found in HLMs as judged by their
with one being 3 -O-angeloyl khellactone and the other
0
0
4 -O-angeloyl khellactone, or stereoisomers of 3 -O-angeloyl
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

Y.-L. Song et al.
Table 1. Chromatographic and mass spectrometric data of dPB and its metabolites formed in liver microsomes of rats and
humans
Rt
Measured
mass
Elemental
composition (ppm)
Error Molecular
formula
1
2
Compound (min)
MS
MS
349[MþNa–C
þ
þ
þ
4 7
H COOH]
H
dPB
38.0 444[MþNH
4
]
C
24
H
26
O
7
þ
4
49[MþNa]
449.1567
C
24
H
26
O
7
Na
0.97
327[MþH–C
4
7
COOH]
þ
2
45[327–C H COOH] , 227
4
7
þ
þ
þ
6
H ]
B1
11.9
22.2
263[MþH]
263.0915
285.0731
301.0468
C
C
C
14
14
14
H
H
H
15
14
14
O
5
ꢁ0.24
0.95
C
14
H
14
O
5
245[MþH–H
2
O] , 203[245–C
3
þ
2
3
85[MþNa]
O
5
Na
þ
01[MþK]
383[MþNa]
O
5
K
1.82
þ
þ
B2
B3
B4
383.1106 C H O Na ꢁ1.23 C H O
267[MþNa–HOC H COOH]
1
9
20
7
19 20
7
6
8
4
6
þ
3
99[MþK]
399.0851
C
19
H
20
O
7
K
ꢁ2.82
þ
þ
þ
27.1 362[MþNH
4
]
C
19
H
20
O
267[MþNa–C
4
H
7
COOH]
3
67[MþNa]
367.1153 C H O Na ꢁ0.24
1
9
20
26
6
8
þ
þ
þ
þ
27.8 460[MþNH ]
C H O
24 26
365[MþNa–C H COOH]
4
4
7
4
65[MþNa]
465.1524
345.1336
367.1154
C
24
H
O
Na ꢁ0.54
245[365–HOC
4
H
6
COONa] ,
2
27, 203
þ
þ
B5
B6
B7
B8
28.8
345[MþH]
þ
4
]
C H O
ꢁ0.98 C H O
267[MþNa–C H COOH]
1
9
21
6
19 20
6
8
8
8
4
7
3
3
62[MþNH
þ
67[MþNa]
C
19
H
20
O
6
Na ꢁ0.47
þ
þ
þ
þ
30.0 460[MþNH
4
]
C
C
C
24
H
24
H
24
H
26
26
26
O
O
O
365[MþNa–C
4
H
7
COOH]
4
65[MþNa]
465.1522 C H O Na ꢁ0.56
245[365–HOC H COONa] ,
2
4
26
8
4
6
2
27, 203
þ
þ
þ
þ
31.3 460[MþNH
4
]
365[MþNa–C
4 7
H COOH]
245[365–HOC H COONa] ,
4
65[MþNa]
465.1526 C H O Na ꢁ1.28
2
4
26
8
4
6
2
27, 203
349[MþNa–HOC
327[MþH–HOC
þ
þ
þ
33.8 460[MþNH
4
]
4
H
6
COOH]
þ
4
65[MþNa]
465.1521
C
24
H
26
O
8
Na ꢁ0.25
4
H
6
COOH] , 227
khellactone with one as the R-isomer and the other as the
S-isomer.
Chromatographic and mass spectrometric data of dPE
and its metabolites E1–E13 are summarized in Table 2. The
mass spectra of dPE and metabolites E4–6, three metabolites
formed via di-oxidation that were absent with dPB, are shown
B4, B6, B7 and B8 were eluted at 27.8 min, 30.0 min, 31.3 min
and 33.8 min, respectively, yet showed the same characteristic
þ
þ
1
ions, including m/z 465 ([MþNa] ), m/z 460 ([MþNH4] ), m/z
in Fig. 6. Similar to that of dPB, the MS spectra of most of its
þ
þ
3
43 ([M–C
4
H
7
COOH] ) and m/z 245 ([M–2C
4
H
7
COOH] ) in
metabolites exhibited intense sodiated adduct ions, which
2
their MS spectra, thus B4, B6, B7 and B8 should be four
isomers with molecular weights of 442 Da, 16 Da higher than
dPB, indicating that these four metabolites are all mono-
oxidized products of dPB. TOF-MS also resulted in identical
were used as precursor ions for MS analysis, yet E1 showed
the protonated ion as base peak, which was the precursor
2
ion of MS analysis. In addition, a potassium adduct ion
appeared in the mass spectra of dPE and all its metabolites.
Again, the characteristic ions at m/z 245 and m/z 227 were
sodiated adduct ions and the same elemental composition of
2
2
C
24
H
26
O
8
for the four metabolites. Interestingly, MS spectra
commonly found in MS spectra of most dPE metabolites,
of the sodiated adduct ion revealed distinct fragmentation
profiles: The characteristic ion at m/z 365 was found with B4,
indicating that these metabolites follow a similar fragmenta-
tion pathway to dPB metabolites.
B6 and B7 and supported the loss of one C
4
H
7
COOH
E1 was eluted at 11.9 min and showed identical retention
1
2
molecule while the fragment ion at m/z 349 in B8 corresponds
to the loss of the oxidized angeloyl group (Table 1, Fig. 5).
This distinct fragmentation profile might be a result of
mono-oxidation occurring at a different side chain. The
time, MS and MS spectra, and accurate mass data with those
of B1 formed from dPB and (ꢁ)-cis-khellactone produced
from dPA, thus E1 was also identified as (ꢁ)-cis-khellactone,
which was formed from hydrolysis of both the angeloyl
group and the isovaleryl group from the C-3 and C-4
positions of dPE, respectively (Table 2, Fig. 9).
0
00
0
0
mono-oxidation might occur via direct hydroxylation of C-4
0
00
or C-5 , or via cyclooxidation of the double bond of the side
0
0
chain substituted at either the C-3 or C-4 position.
E2 and E3 were observed at the retention times of 18.2 min
and 20.1 min, respectively, and exhibited similar character-
þ
þ
istic ions at m/z 385 ([MþNa] ) and m/z 401 ([MþK] ) in their
1
MS spectra, indicating that E2 and E3 are isomers with the
The metabolites of dPE formed by rat and human liver
microsomes
molecular weight of 362 Da, which was 66 Da lower than dPE
(
molecular weight: 428 Da). TOF-MS also revealed identical
elemental compositions and molecular formulas of C19
for both metabolites. In addition, the product ions at m/z 245
The incubation of dPE with RLMs and HLMs yielded the
same 13 metabolites as judged by identical retention times,
UV and mass spectra and accurate mass data (Figs. 2 and 4,
Table 2), indicating that dPE undergoes similar biotrans-
formations in liver microsomes of both species.
22 7
H O
1
2
in MS spectra and m/z 267 in MS of the sodiated adduct ion
correspond to the loss of HOC COONa and HOC
COOH molecules from the precursor ion at m/z 385,
4
H
8
4 8-
H
wileyonlinelibrary.com/journal/rcm
Copyright ß 2011 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730

Metabolism of praeruptorin B and E in liver microsomes
Table 2. Chromatographic and mass spectrometric data of dPE and its metabolites formed in liver microsomes of rats and
humans
Rt
Measured
Mass
Elemental
composition
Error Molecular
1
2
Compound (min)
MS
(ppm)
formula
MS
þ
þ
þ
4 9
H COOH]
dPE
42.6
11.9
446[MþNH
4
]
446.2188
451.1737
C
C
24
H
28
O
8
NH
Na
4
ꢁ1.34
ꢁ2.22
C
C
24
14
H
H
28
14
O
8
5
349[MþNa–C
327[MþH–C
þ
4
51[MþNa]
24
H
28
O
8
4
H
9
COOH]
þ
4
67[MþK]
þ
þ
E1
263[MþH]
263.0920
285.0737
301.0474
385.1266
401.1010
385.1258
401.1004
483.1625
499.1363
C
14
14
14
H
15
14
14
O
5
ꢁ2.23
ꢁ1.54
0.55
ꢁ2.26
ꢁ3.61
ꢁ2.85
ꢁ2.03
0.13
O
245[MþH–H
2
O]
þ
2
85[MþNa]
C
C
H
O
5
Na
þ
3
01[MþK]
385[MþNa]
H
O
5
K
þ
þ
E2
E3
E4
18.2
20.1
24.0
C H O Na
C
C
C H O K
C H O Na
C
C H O
267[MþNa–HOC H COOH]
þ
4 8
1
9
22
22
22
7
19 22
7
7
9
4
8
þ
4
01[MþK]
19
H
O
O
7
K
245[MþH–HOC
267[MþNa–HOC
H COOH]
þ
þ
4 8
H COOH]
385[MþNa]
19
H
7
Na
C
19
H
22
O
þ
þ
4
01[MþK]
245[MþH–HOC H COOH]
þ
4 8
1
9
22
7
4
8
þ
483[MþNa]
C H O
24 28
365[MþNa–HOC H COOH]
þ
4 8
2
4
28
9
þ
4
99[MþK]
24
H
28
O
9
K
0.45
343[MþH–HOC
H COOH]
þ
2
45[365–HOC
4
H
6
CONa] , 227
þ
þ
E5
E6
E7
E8
E9
E10
25.9
26.9
29.0
29.5
29.5
30.7
483[MþNa]
483.1646
499.1378
C H O Na
C
ꢁ4.48
ꢁ2.78
C H O
365[MþNa–HOC H COOH]
2
4
28
9
24 28
9
9
6
6
8
8
4
8
þ
þ
4
99[MþK]
24
H
28
O
9
K
343[MþH–HOC
45[365–HOC
365[MþNa–HOC
H COOH]
4 8
CONa] , 227
þ
2
4
H
6
þ
þ
483[MþNa]
483.1634
499.1364
C
24
H
28
O
9
Na
ꢁ1.81
C
C
C
C
24
19
19
24
H
H
H
H
28
22
22
28
O
O
O
O
4 8
H COOH]
þ
þ
4
99[MþK]
C H O K
0.23
343[MþH–HOC H COOH]
2
4
28
9
4
8
þ
2
45[365–HOC
267[MþNa–C
245[MþH–C H COOH]
4
H
6
CONa] , 227
þ
þ
þ
364[MþNH
4
]
4 9
H COOH]
þ
3
3
69[MþNa]
369.1309
385.1053
C H O Na
C H O K
19 22 6
ꢁ0.21
ꢁ1.53
1
9
22
6
4
9
þ
85[MþK]
þ
þ
þ
364[MþNH
4
]
267[MþNa–C
245[MþH–C
4 9
H COOH]
þ
4 9
H COOH]
3
69[MþNa]
369.1310
385.1055
462.2128
467.1617
483.1415
C
19
H
22
O
6
Na
ꢁ0.41
ꢁ2.10
ꢁ1.34
1.22
þ
3
85[MþK]
C H O K
C
C
C
1
9
22
28
28
28
6
þ
þ
þ
þ
462[MþNH
4
]
24
24
24
H
O
O
O
8
NH
4
365[MþNa–C
343[MþH–C
4 9
H COOH]
4
67[MþNa]
H
8
Na
4
H
9
COOH]
þ
4
83[MþK]
H
8
K
0.07
þ
þ
þ
þ
462[MþNH ]
C H O
24 28
365[MþNa–C H COOH]
4
4
9
4
4
67[MþNa]
467.1676
483.1416
467.1784
483.1427
C
C
C
24
24
24
H
28
28
28
O
8
Na
0.15
ꢁ0.01
ꢁ1.32
ꢁ2.56
343[MþH–C
4
H
9
COOH]
þ
83[MþK]
H
H
O
O
8
K
þ
þ
E11
E12
31.3
31.7
467[MþNa]
8
Na
C
C
24
24
H
H
28
28
O
8
8
349[MþNa–HOC
4 8
H COOH]
þ
þ
4
83[MþK]
C H O K
327[MþH–HOC H COOH]
2
4
28
8
4
8
þ
þ
þ
þ
462[MþNH
4
]
O
365[MþNa–C
343[MþH–C
4
H
9
COOH]
COOH]
4
67[MþNa]
467.1684
483.1426
462.2137
467.1683
483.1422
C
24
H
28
O
8
Na
ꢁ1.63
ꢁ2.24
ꢁ3.35
ꢁ1.40
ꢁ1.53
4
H
9
þ
4
83[MþK]
þ
4
C H O K
2
4
28
8
þ
þ
E13
34.0
462[MþNH ]
C H O NH
C H O
365[MþNa–C H COOH]
24 28
24 28
C H O K
24 28 8
8
4
24 28
8
4
9
þ
4
4
67[MþNa]
C
H
O
8
Na
343[MþH–C
4 9
H COOH]
þ
83[MþK]
0
the other oxygen atom should be introduced into the C-3
supporting occurrence of hydroxylation on the isovaleryl
group substituted at the C-4 position. Thus, E2 and E3 should
0
angeloyl moiety. Therefore, E3, E4 and E5 were tentatively
identified as positional isomers formed from di-oxidation of
dPE with one oxygen atom introduced into the isovaleryl
group and the other into the angeloyl group, respectively
(Table 1 and Fig. 9).
be positional isomers that are formed from a combination of
removal of the angeloyl group via hydrolysis and hydrox-
ylation on the isovaleryl substituent of dPE (Table 1 and
0
00
Fig. 9). The hydroxyl group could be introduced at the C-2 ,
0
00
000
000
1
2
C-3 , C-4 or C-5 position of the isovaleryl group.
E7 and E8 showed similar MS and MS spectra and
accurate mass data yet different retention times (E7, 29.0 min;
1
2
E4, E5 and E6 showed similar MS and MS spectra yet
different retention times (E4, 24.0 min; E5, 25.9 min; E6,
1
E8, 29.5 min). Their MS spectra showed adduct ions at m/z
þ
þ
þ
2
6.9 min), suggesting that these three metabolites are isomers.
364 ([MþNH ] ), m/z 369 ([MþNa] ) and m/z 385 ([MþK] ),
4
1
þ
In MS spectra, the adduct ions at m/z 483 ([MþNa] ) and m/z
indicating the molecular weight of 346 Da, 82 Da lower than
dPE and 16 Da lower than the mono-hydroxylated metab-
olites E2 and E3. The molecular formulas are C H O as
measured by TOF-MS. The MS of the sodiated adduct ion at
m/z 369 of the two compounds revealed a predominant
product ion at m/z 267, 102 Da less than the precursor ion,
þ
4
99 ([MþK] ) indicated the molecular weight of 460 Da, 32
Da higher than dPE, suggesting the occurrence of
1
9
22
6
2
di-oxidation, which was supported by molecular formulas
of C24
istic ion at m/z 365 in MS spectra corresponds to the loss of a
mono-hydroxylated C COOH molecule, indicating that
there is one hydroxyl substituent on the isovaleryl group and
28 9
H O as revealed by TOF-MS analysis. The character-
2
0
4
H
9
corresponding to the loss of the C-4 isovaleryl moiety
4 9
(C H COOH). Therefore, the identities of E7 and E8 were
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

Y.-L. Song et al.
1
2
Figure 5. MS (left) and MS (right) spectra of dPB and its metabolites B1–B8 formed in liver microsomes of rats and humans.
wileyonlinelibrary.com/journal/rcm
Copyright ß 2011 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730

Metabolism of praeruptorin B and E in liver microsomes
1
2
Figure 6. MS (left) and MS (right) spectra of dPE and its metabolites E4–E6 formed in liver microsomes of rats and humans.
0
þ
tentatively assigned as stereoisomers of 4 -O-isovaleryl
and m/z 327 ([MþH–HOC H COOH] ) for E11. These
4
8
khellactone (Fig. 9) with one as the R-isomer and the other
the S-isomer.
E9, E10, E11, E12 and E13 were the last five metabolites
eluted between 29–34 min sequentially. The MS spectra of
all five metabolites exhibited a sodiated adduct ion at
findings suggested that the mono-oxidation occurs on
the angeloyl group of E9, E10, E12 and E13 but on the
isovaleryl group of E11 (Table 1, Fig. 9). Since the C-3
0
1
angeloyl moiety has four potential sites for mono-oxidation,
00 00 00
including hydroxylation at C-3 , C-4 and C-5 , and
þ
00
00
m/z 467 ([MþNa] ), indicating their molecular weight of
cyclooxidation of the double bond between C-2 and C-3 ,
it is reasonable to deduce that three of the four metabolites
E9, E10, E12 and E13 were formed via hydroxylation on
the angeloyl group and one of them was generated via
cyclo-oxidation. In the case of E11, there are also four
4
44 Da, 16 Da higher than dPE. TOF-MS analysis indicated
28 8
the same molecular formula of C24H O for all five
metabolites. Thus, these five metabolites might be mono-
oxidized products of dPE. Interestingly, MS of the
sodiated ions of the five metabolites resulted in distinct
2
0
00
000
000
000
potential sites (C-2 , C-3 , C-4 and C-5 ) of hydroxylation
on the isovaleryl group but only one metabolite M11 was
observed, thus the exact position of mono-oxidation could
not be definitely assigned.
product ions: the major product ion at m/z 365
þ
(
[MþNa–C H COOH] ) for E9, E10, E12 and E13, yet the
4
9
þ
4 8
H COOH] )
major product ions at m/z 349 ([MþNa–HOC
1
2
Figure 7. The HPLC chromatogram of the sample from basic hydrolysis of dPA (A) and MS (B) and MS (C) spectra of the main
product (ꢁ)-cis-khellactone (P1).
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

Y.-L. Song et al.
mono-oxidation of these DCKs on the lipophilic camphanoyl
0
0
moieties substituted at the C-3 or C-4 positions was
also observed.
[
26]
In contrast to hydrolysis as one of the
predominant pathways of dPB and dPE metabolisms in both
RLMs and HLMs that were found in the present study, no
hydrolysis of either camphanoyl ester moiety of DCKs was
observed and the authors speculated a steric hindrance of the
bulky camphanoyl structure as the cause. Taken together,
0
0
these findings indicate that the C-3 and C-4 substituents
on khellactone are determinant factors for the metabolic
properties (metabolic pathway and extent of metabolism) of
these angular-type pyranocoumarins.
There were two isomers (B3 and B5, E7 and E8) formed
0
from hydrolytic removal of the C-3 ester group of dPB
and dPE, respectively (Figs. 8 and 9). The mass profiles of E7
and E8 indicated two stereoisomers with one as the R-isomer
and the other as the S-isomer. Generally, direct hydrolysis of
0
the C-3 moiety should result in the R-isomer without
changing the configuration. The formation of the S-isomer
was proposed as follows: The R-isomer loses a water
Figure 8. Proposed metabolic pathways of dPB in liver
microsomes of rats and humans. The intermediate is shown
in brackets.
0
molecule from the exposed hydroxyl group at the C-4
0
position and the adjacent C-4 ester group to form a
0
0
double bond between C-3 and C-4 . This newly formed
alkene functional group might be unstable and undergoes
hydration in the aqueous media to give both R- and
S-isomers. The formation of two stereoisomers from
Metabolic difference between dPB and dPE in rat and
human liver microsomes
0
removal of the C-3 ester group has been definitely supported
In general, dPE was eliminated slightly faster than dPB in liver
microsomes of both species (percentage of remaining: dPE,
by NMR analysis of the metabolites of l-praeruptorin A,
another pyranocoumarin isolated from Peucedani Radix, and
seems to be a common metabolic pathway of angular-type
pyranocoumarins in liver microsomal proteins (unpublished
data from our research group). Accordingly, the metabolic
pathway of dPB was proposed, and B3 and B5 were
1
5% (rat) and 41% (human); dPB, 25% (rat) and 51% (human)).
In the present study, the mass fragmentation profiles of
0
0
dPE, of which the C-3 and C-4 positions are substituted
with different groups (angeloyl and isovaleryl group,
respectively), and its metabolites, revealed a preference of
0
tentatively assigned as stereoisomers of 3 -O-angeloyl
0
0
cleavage of the C-3 angeloyl substituent. This finding agrees
khellactone. However, further hydrolysis of the C-4 ester
n
well with the ESI-MS profiles of pyranocoumarins from
Peucedani Radix reported by Tao et al. Thus, the metabolic
pathway of dPE was easily proposed (Fig. 9). dPB has two
angeloyl moieties substituted at the C-3 and C-4 positions.
Although hydrolysis and oxidation could be concluded as
the two predominant pathways of dPB metabolism by liver
microsomal proteins of rats and humans, the exact site of
hydrolysis was not readily speculated based on the mass
fragmentation patterns of dPB and its metabolites. Since dPB
metabolites (except for B1 which has two adjacent hydroxyl
group only resulted in one metabolite, (ꢁ)-cis-khellactone, of
which the configuration is the same as the parent compounds.
This might be a result of a subsequent stereoselective
hydrolysis of the isomers catalyzed by NADPH-independent
[
24]
0
0
[
27–29]
carboxylesterase(s).
Furthermore, slight species difference in dPB and dPE
metabolisms was evidenced by the metabolites formed and
the relative ratio of each metabolite. RLMs used in the present
study showed higher capability in catalyzing both dPB and
dPE metabolisms (percentages of the parent compound
remaining: 25% (dPB) and 15% (dPE) in the rat vs. 51% (dPB)
and 41% (dPE) in human). dPB produced eight metabolites
(B1–B8) in RLMs and nine metabolites (B1–B9) in HLMs. dPE
formed 13 identical metabolites in both species. Further, the
relative ratios of the metabolites, as calculated based on the
peak area ratio of each metabolite to that of B1 or E1 formed
from dPB or dPE in RLMs, were different in rats and humans
(Fig. 10). B3, one of the metabolites generated from hydrolysis
0
0
groups) possess different substituents at the C-3 and C-4
positions and exhibit similar mass fragmentation patterns
to those of dPE, dPE metabolites and the angular-type
[
24,25]
pyranocoumarin constituents from Peucedani herb,
0
the mass fragmentation preference at the C-4 position might
also be applicable to B2–B8. Thus the structures of these
metabolites and the metabolic pathway of dPB in RLMs and
HLMs were proposed (Fig. 8).
0
As shown in Figs. 8 and 9, oxidation and hydrolysis were
the two predominant pathways of dPB and dPE metabolism
in rat and human liver microsomal proteins and occurred
only on the aliphatic chain. Although oxidation could be
of one angeloyl group from the C-3 position of dPB, was
the most abundant metabolite in RLMs while B3 and B8,
0
the metabolites generated via mono-oxidation on the C-4
side chain of dPB, were two major metabolites in HLMs.
(ꢁ)-cis-Khallectone (B1), formed from loss of both angeloyl
0
0
initiated on either the C-3 or C-4 substituent, hydrolysis was
0
0
0
only initiated at the C-3 substituent. Moreover, di-oxidation
moieties from the C-3 and C-4 positions, was less than
one-tenth of B3 in both species. In dPE metabolism, the
metabolite formed via hydrolysis ((ꢁ)-cis-khallectone (E1)),
has the highest relative ratio in RLMs, while the metabolite
was observed with dPE yet was absent with dPB. In a recent
0
0
report on the metabolic study of 3 ,4 -di-O-(ꢁ)-camphanoyl-
þ)-cis-khellactone derivatives (DCKs) in HLMs, a rapid
(
wileyonlinelibrary.com/journal/rcm
Copyright ß 2011 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730

Metabolism of praeruptorin B and E in liver microsomes
Figure 9. Proposed metabolic pathways of dPE in liver microsomes of rats and humans. The intermediate is shown in brackets.
Figure 10. Relative ratios of the metabolites of dPB (left) and dPE (right) produced in liver microsomal proteins of rats and
humans.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/rcm

Y.-L. Song et al.
formed from mono-oxidation (E13) predominated in HLMs
Fig. 10). The species difference could be attributed to the
well-characterized species difference in CYP450 expression
[5] J. Kang, R. J. Yu. Zhongguo Yi Ke Da Xue Xue Bao 1994, 23,
(
122.
[
[
[
[
6] Q. Y. Wang, E. R. Li, G. X. Zhao, J. Kang, X. M. Hou, R. J. Yu.
Zhongguo Yi Ke Da Xue Xue Bao 1998, 27, 588.
7] Z. X. Chen, B. S. Huang, Q. L. She, G. F. Zeng. Acta Pharm.
Sin. 1979, 14, 486.
8] J. S. Ye, H. Q. Zhang, C. Q. Yuan. Acta Pharm. Sin. 1982, 17,
431.
[
30,31]
between rats and humans.
CONCLUSIONS
9] H. T. Chang, Y. Okada, T. Okuyama, P. F. Tu. Magn. Reson.
Chem. 2007, 45, 611.
This is the first report on dPB and dPE metabolisms in rat
and human liver microsomes. The metabolites were tenta-
tively identified using liquid chromatography coupled with
electrospray tandem mass spectrometry and time-of-flight
[
[
10] H. T. Chang, Y. Okada, T. J. Ma, T. Okuyama, P. F. Tu. J.
Asian Nat. Prod. Res. 2008, 10, 577.
11] J. Y. Wu, W. F. Fong, J. X. Zhang, C. H. Leung, H. L. Kwong,
M. S. Yang, D. Li, H. Y. Cheung. Eur. J. Pharmacol. 2003, 473,
n
mass spectrometry. The ESI-MS fragmentation profiles of
9
.
dPE and its metabolites followed the same pathway as those
of the angular-type pyranocoumarins from Peucedani
Radix reported previously, thus might aid identification of
other pyranocoumarins and metabolites based on their
sequential mass fragmentation profiles. The current results
have also demonstrated that HPLC/IT-MS/MS coupling
with HPLC/TOF-MS is a rapid, sensitive and reliable tool for
the characterization of unknown metabolites in biological
samples. Oxidation and hydrolysis were proved to be the two
predominant pathways of both compounds and substituents
at the C-3 and C-4 positions determine the extent and the
type of reactions. Furthermore, species difference in dPB and
dPE biotransformation was insignificant, supporting the rat
as a suitable substitute for predicting the metabolic profile of
dPB and dPE in human. Further studies are ongoing in our
laboratory on identification of the main CYP450 isozyme(s)
involved and the in vivo metabolic profiles of dPB and dPE.
[12] Y. Aida, T. Kasama, N. Takeuchi, S. Tobinaga. Chem. Pharm.
Bull. 1995, 43, 859.
[
13] Z. Xu, X. Wang, Y. Dai, L. Kong, F. Wang, H. Xu, D. Lu, J.
Song, Z. Hou. Chem. Biol. Interact. 2010, 186, 239.
14] M. R. Rao, W. B. Liu, P. Q. Liu. Acta Pharm. Sin. 2001, 36, 165.
15] N. C. Zhao, W. B. Jin, X. H. Zhang, F. L. Guan, Y. B. Sun, H.
Adachi, T. Okuyama. Biol. Pharm Bull. 1999, 22, 984.
16] M. Yoshikawa, N. Nishida, K. Ninomiya, T. Ohgushi, M.
Kubo, M. Toshio, H. Matsuda. Bioorg. Med. Chem. 2006, 14,
456.
[
[
[
[
17] W. Chan, L. Cui, G. W. Xu, Z. W. Cai. Rapid Coummun. Mass
Spectram. 2006, 20, 1775.
0
0
[
18] J. F. Garda-Reyes, I. Ferrer, E. M. Thurman, A. Molina-Diaz,
A. R. Fernandez-Alba. Rapid Commun. Mass Spectram. 2005,
1
9, 2780.
[19] H. Nishino, T. Okuyama, M. Takata, S. Shibata, H. Tokuda, J.
Takayasu, T. Hasegawa, A. Nishino, H. Ueyama, A.
Iwashima. Carcinogenesis 1990, 11, 1557.
[
[
20] X. D. Wu, X. Wu. Chin. Pharm. J. 1996, 31, 234.
21] G. Lin, J. Tang, X. Q. Liu, Y. Jiang, J. Zheng. Drug Metab.
Dispos. 2007, 35, 607.
[
22] O. H. Lowry, N. J. Rosenbrough, A. L. Farr, R. J. Randall. J.
Biol. Chem. 1951, 193, 265.
Acknowledgements
[
[
23] X. L. Wu, L. Y. Kong, Z. D. Min. Acta Pharm. Sin. 2002, 37, 527.
24] Y. Tao, J. Luo, Y. Lu, D. Xu, Z. Hou, L. Y. Kong. Nat. Prod.
Commun. 2009, 4, 1079.
The research was supported by the National Basic Research
Program of China (973 program, Grant No. 2009CB522707)
and the Research Committee of University of Macau (Project
No. UL016/09-Y1/CMSWYT01/ICMS).
[
[
25] G. Y. Zhu, G. Y. Chen, Q. Y. Li, X. L. Shen, H. X. Fang. Chin. J.
Nat. Med. 2004, 2, 308.
26] M. Suzuki, Y. Li, P. C. Smith, J. A. Swenberg, D. E. Martin, S.
L. Morris-Natschke, K. H. Lee. Drug Metab. Dispos. 2005, 33,
REFERENCES
1
588.
[
27] Z. Sun, D. J. Murry, S. P. Sanghani, W. I. Davis, N. Y.
Kedishvili, Q. Zou, T. D. Hurley, W. F. Bosron. JPET
2004, 310, 469.
[
1] The State Pharmacopoeia Commission of P.R. China.
Peucedani Radix, in Pharmacopoeia of the People’s
Republic of China, vol. I, Chemical Industry Press, Beijing,
[28] Q. G. Zhu, J. H. Hu, H. W. Zeng. Acta Pharmaceutica Sinica
2005, 40, 322.
2
010, p. 248.
[
[
[
2] S. C. Xi, Y. M. Ruan, L. Z. Zhang. Zhongguo Zhong Xi Yi Jie He
Za Zhi. 1996, 16, 218.
3] R. Zhou, H. L. Wang, X. H. Zhang, J. Xing, C. M. Li.
Zhongguo Yi Ke Da Xue Xue Bao 2001, 30, 325.
4] N. C. Zhao, H. L. Wang, X. Jin, X. H. Zhang, K. J. Liu, F. L.
Guan. Zhongguo Yi Ke Da Xue Xue Bao 1994, 23, 197.
[29] Y. Yoshigae, T. Imai, M. Taketani, M. Otagiri. Chirality 1999,
11, 10.
[30] D. F. Lewis, C. Ioannides, D. V. Parke. Environ. Health
Perspect. 1998, 106, 633.
[31] M. Turpeinen, C. Ghiciuc, M. Opritoui, L. Tursas, O.
Pelkonen, M. Pasanen. Xenobiotica 2007, 37, 1367.
wileyonlinelibrary.com/journal/rcm
Copyright ß 2011 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2011, 25, 719–730