Characterization of emodin metabolites in Raji cells by LC-APCI-MS/MS

Article abstract of DOI:10.1016/j.bmc.2009.09.024

A rapid, simple, and sensitive liquid chromatography-atmospheric pressure chemical ionization tandem mass spectrometry (LC-APCI-MS/MS) method was developed for the identification and quantification of emodin metabolites in Raji cells, using aloe-emodin as an internal standard. Analyses were performed on an LC system employing a Cosmosil 5C¹⁸ AR-II column and a stepwise gradient elution with methanol-20 mM ammonium formate at a flow rate of 1.0 mL/min operating in the negative ion mode. As a result, the starting material emodin and its five metabolites were detected by analyzing extracts of Raji cells that had been cultivated in the presence of emodin. The identification of the metabolites and elucidation of their structures were performed by comparing their retention times and spectral patterns with those of synthetic samples. In addition to the major metabolite 8-O-methylemodin, four other metabolites were assigned as ω-hydroxyemodin, 3-O-methyl-ω-hydroxyemodin, 3-O-methylemodin (physcion), and chrysophanol.

Full text of DOI:10.1016/j.bmc.2009.09.024

Bioorganic & Medicinal Chemistry 17 (2009) 7493–7499
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Characterization of emodin metabolites in Raji cells by LC–APCI-MS/MS
*
Junko Koyama , Atsuko Takeuchi, Izumi Morita, Yu Nishino, Maki Shimizu, Munetaka Inoue,
Norihiro Kobayashi
Kobe Pharmaceutical University, Higashinada, Kobe 658-8558, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A rapid, simple, and sensitive liquid chromatography–atmospheric pressure chemical ionization tandem
mass spectrometry (LC–APCI-MS/MS) method was developed for the identification and quantification of
emodin metabolites in Raji cells, using aloe-emodin as an internal standard. Analyses were performed on
an LC system employing a Cosmosil 5C18 AR-II column and a stepwise gradient elution with methanol–
20 mM ammonium formate at a flow rate of 1.0 mL/min operating in the negative ion mode. As a result,
the starting material emodin and its five metabolites were detected by analyzing extracts of Raji cells that
had been cultivated in the presence of emodin. The identification of the metabolites and elucidation of
their structures were performed by comparing their retention times and spectral patterns with those
of synthetic samples. In addition to the major metabolite 8-O-methylemodin, four other metabolites were
Received 6 August 2009
Revised 7 September 2009
Accepted 10 September 2009
Available online 18 September 2009
Keywords:
LC–APCI-MS/MS
Emodin
Raji cell
Metabolite
assigned as
x
-hydroxyemodin, 3-O-methyl-
x
-hydroxyemodin, 3-O-methylemodin (physcion), and
chrysophanol.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
2–7 possessed greater inhibitory effects compared with inhibition
by emodin. Therefore, we are interested in characterization of the
metabolites of emodin in Raji cells.
The polyhydroxylated anthraquinone emodin (1,3,8-trihy-
droxy-6-methylanthraquinone) (1) is a widely occurring natural
product found in several higher plant families, fungi, lichens, and
animals. In higher plants emodin is present chiefly as the glycoside
conjugate, which is an active component in various Chinese herbs,
such as rhubarb, Rheum palmatum L., and Polygonum multiflorum
Thunb. Pharmacological studies using crude and pure emodin have
shown that emodin is a selective receptor tyrosine kinase inhibi-
tor1 and exhibits tumor cell-growth inhibition,2 as well as
antibiotic,3 anti-viral,4 and cytostatic5 activity. Thus, emodin is ex-
pected to be a promising antitumor agent for patients who develop
drug resistance to other traditional remedies.6–9
Because of their low concentrations in biological samples, quan-
tification of the metabolites of emodin in cells requires a sensitive
analytical method. High-performance liquid chromatography
(HPLC)17,18 and capillary electrophoresis (CE)19 are used most com-
monly for the detection of anthraquinone metabolites. Liquid chro-
matography combined with tandem mass spectrometry (LC–MS/
MS) is a highly specific and sensitive method that has been widely
used as a powerful analytical tool for the identification of drug
metabolites in biological matrices.15,20–23 MS technique has made
possible the acquisition of structurally informative data from pro-
tonated molecules of analytes of interest.
The biotransformation of emodin with rat liver microsomes has
been studied, and six oxidative emodin metabolites (2-hydroxy-
emodin (2), 4-hydroxyemodin (3), 5-hydroxyemodin (4),
In this paper, a simple, sensitive, and specific LC–APCI-MS/MS
method was employed to investigate the in vitro metabolism of
emodin (an overview of which is shown in Fig. 1). In Raji cells, five
emodin metabolites, 6 and 8–11, together with unchanged emodin
were detected. The results were quite different from the reported
results from the biotransformation of emodin incubated in rat liver
microsomes.10
7-hydroxyemodin (5),
x-hydroxyemodin (6), and emodic acid
(7)), generated by cytochrome P450 were observed.10 The hepatic
microsomes derived from various animal species transformed
emodin into several anthraquinone metabolites.11–15 Though emo-
din is a selective inhibitor of protein kinase, few biotransformation
studies of emodin in human cells have been performed. We have
previously shown that emodin inhibits Epstein-Barr virus early
antigen (EBV-EA) activation in Raji cells, which are derived from
the human B-lymphoid cell line.16 Oxidative emodin derivatives
2. Materials and methods
2.1. Chemicals
Emodin (1) was purchased from Extrasynthese (Genay, France).
Ammonium formate, acetonitrile, and methanol (MeOH) (HPLC
* Corresponding author. Tel.: +81 78 441 7549; fax: +81 78 441 7550.
E-mail address: j-koyama@kobepharma-u.ac.jp (J. Koyama).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.09.024

7494
J. Koyama et al. / Bioorg. Med. Chem. 17 (2009) 7493–7499
grade) were purchased from Wako (Osaka, Japan). Aloe-emodin
S-Adenosyl-L-methionine-d3 was purchased from C/D/N Isotopes
(12) was purchased from Sigma–Aldrich (St. Louis, MO).
(Quebec, Canada). Chrysophanol (11) was isolated from the root
Raji cells (2.0 x 106/mL)
+ 5 µM emodin
Incubated at 37°C in 5% CO2 for 72 h
Centrifuged at 1300 rpm for 10 min
Cells
Supernatant
1. Added MeOH
2. Sonicated for 15 min
3. Centrifuged at 3000 rpm for 15 min
4. Filtered through 0.45 µm filter
Bond Elut C18
Extract of cells
Extract of supernatant
LC-APCI-MS/MS
m/z = 283.0/240.0
8-O-Methylemodin
Figure 1. Schematic representation of the sample preparation and the determination of metabolites of emodin in Raji cells by LC–APCI-MS/MS.
R8
R1
R4
O
O
R7
R6
R2
R3
R5
R1
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OCH3
OCH3
OH
OH
R2
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
R3
R4
R5
R6
CH3
R7
H
H
H
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
R8
OH
OH
OH
OH
OH
OH
OH
OCH3
OH
OH
OH
OH
OH
OH
OH
OCH3
OH
1
2
OH
OH
OH
OH
OH
OH
OH
OH
OCH3
OCH3
H
H
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
3
CH3
4
CH3
5
CH3
6
CH2OH
COOH
CH3
7
8
9
CH2OH
CH3
10
11
12
13
14
15
16
17
CH3
H
CH2OH
CHO
OH
OH
OH
OH
OH
CH3
CH2OH
CH2OH
CH2OCH3
Figure 2. Structures of emodin derivatives.

J. Koyama et al. / Bioorg. Med. Chem. 17 (2009) 7493–7499
7495
OH
OH
OH
O
O
OH
OAc
OH
O
O
OAc O
1) Ac2O, NaOAc
2) KOH
1) Ac2O
CH3
CH2OH
HO
HO
CH2Br
AcO
2) N-bromo-
succinimide
O
(6)
(1)
(18)
1) NaOH
Boric acid
Ac2O
2) CH3I, K2CO3
3) Ac2O, KOH
OH
AcO
O
OAc
O
O
AcO
OAc
B
B
OH
O
O
NaOH
MeOH
OH O
+
CH2OH
H3CO
76%
CH2OAc
CH2OAc AcO
AcO
(9)
O
O
OH
OH
O
O
1) CH3I, K2CO3
2) NaOH
OH
CH2OCH3
OCH3
H3CO
O
O
HO
OH
O
+
(17)
CH2OH
CH2OH
HO
HO
O
(16)
(15)
Scheme 1. Synthesis of methyl ethers of
x-hydroxyemodin.
bark of Cassia siamea.24 2-Hydroxyemodin (2), 4-hydroxyemodin
(3), 5-hydroxyemodin (4), and 7-hydroxyemodin (5) were pre-
100 lg/mL streptomycin, and 10 mM HEPES buffer. Raji cells were
cultured at 37 °C in a humidified 5% CO2 incubator. The cells were
inoculated and maintained, and the medium was changed every
3–4 days. Raji cells (2.0 Â 106 cells mLÀ1) were incubated in
pared according to the method of Banks.25 Synthesis of
x-hydroxy-
emodin (6) was accomplished according to Hirose’s method,26 and
emodic acid (7) was prepared according to Wolfbeis’s method.27
Emodin aldehyde (13) was obtained by oxidation of 6 with
MnO2. 1-O-Methylemodin (14), 3-O-methylemodin (10), and 8-O-
25 mL of the serum-free medium containing 5 lM emodin at
37 °C in a CO2 incubator. Because the viability of Raji cells was
70% when exposed to 8.5 M emodin, the emodin concentration
l
methylemodin (8) were obtained by methylation of
iodomethane. 1-O-Methyl-
-hydroxyemodin (15)28 and 8-O-
methyl-
-hydroxyemodin (16)29 were obtained by methylation
of 6. 3-O-Methyl-
1
with
could not exceed this threshold for the metabolic assay. Incubation
with a medium containing 100 nM emodin was insufficient, making
the identification of the metabolites impossible. Further, emodin
x
x
x
-hydroxyemodin (9) was prepared from 1,3,8-
inhibits EBV-EA activation with an IC50 of 2.6 l
M.31 After incuba-
triacetoxy-6-bromomethylanthraquinone (18).30 The structures of
all emodin derivatives are provided in Figure 2, and the chemical
transformations are shown in Scheme 1.
tion of the Raji cells with emodin for 72 h, the cell suspensions were
centrifuged at 1300 rpm for 10 min, and the supernatant and cells
were separated. The cell pellet and supernatant were processed
separately to provide the cell extract and supernatant extract for
analysis.
Water (H2O) was purified using a Millipore water purification
system (Bedford, MA) with a resistance >18 MX/cm. Solid phase
extraction (SPE) cartridges (Bond Elut C18, 500 mg, 6 mL) were pur-
chased from Varian (Lake Forest, CA).
To generate the extract of cells, an internal standard (IS; 10 lM
aloe-emodin) was added to the cell pellet followed by the addition
of MeOH to provide a final volume of 0.5 mL, and the mixture was
processed ultrasonically for 15 min. After centrifugation at
3000 rpm for 15 min, the supernatant was filtered through a
2.2. Synthesis of 1,3,8-trihydroxy-6-methoxymethyl-
anthraquinone (17)
0.45-
of cells. To generate the extract of the supernatant, the supernatant
containing 2 M internal standard was loaded onto the SPE car-
lm filter and transferred into a vial to provide the extract
1,3,8-Triacetoxy-6-bromomethylanthraquinone (18) (50 mg)
was refluxed with NaOH (50 mg) in MeOH (20 mL) for 1 h. After
evaporation of the solvent, water was added, and the mixture
was stirred for 30 min. The aqueous solution was extracted with
chloroform, and the organic layer was washed with water, dried
over MgSO4, and evaporated to dryness. The residue was purified
with PTLC to afford 1,3,8-trihydroxy-6-methoxymethylanthraqui-
none (17) as yellow crystals (76%). Mp 195–197 °C (decomp.). 1H
NMR (500 MHz, CDCl3) d: 3.46 (3H, s, CH3), 4.53 (2H, s, CH2),
6.60 (1H, d, J = 2.0 Hz, H-2), 7.23 (1H, d, J = 2.0 Hz, H-4), 7.26 (1H,
s, H-7), 7.72 (1H, s, H-5). 13C NMR (CDCl3) d: 58.7 (OCH3), 73.5
(CH2O), 108.8 (C2), 109.2 (C1a), 110.2 (C4), 115.2 (C9a), 118.4
(C5), 122.4 (C7), 133.7 (C10a), 135.3 (C4a), 147.8 (C6), 162.4 (C8),
165.4 (C1), 166.8 (C3), 182.7 (C10), 190.3 (C9). HRMS m/z:
300.0641 (Calcd for C16H12O6, 300.0633).
l
tridge that had been activated and conditioned with 1.5 mL MeOH
followed by 1.5 mL water. The supernatant was drawn through the
column by using a low vacuum, then a full vacuum was applied
briefly, and the retained components were eluted from the car-
tridge with 1.5 mL MeOH into a vial to provide the extract of the
supernatant. The extracts (50 lL injection volume) were analyzed
by LC–APCI-MS/MS. Controls for the extracts were obtained
according to the above method using samples without emodin in
cells cultures and without Raji cells in a serum-free medium con-
taining emodin.
2.4. Instrumentation
The LC–APCI-MS/MS system used for the assays was a Shima-
dzu HPLC system (Shimadu, Kyoto, Japan) consisting of a binary
pump (LC-10AD liquid chromatograph), an automatic solvent deg-
asser (DGU-14A degasser), and an autosampler (SIL-10AD auto
injector) coupled to an API3000 LC-MS/MS system (Applied Biosys-
tems, Foster City, CA) equipped with an APCI interface. All MS data
were measured in the negative ion mode. Typically, source condi-
2.3. Cell culture and sample extraction
Raji cells (EBV genome-carrying lymphoblastoid cells) derived
from Burkitt’s lymphoma were cultured in RPMI-1640 (Gibco,
Grand Island, NY) supplemented with 10% fetal bovine serum
(GOLD, ICN Biomedicals, Aurora, Ohio), 100 units/mL penicillin G,

7496
J. Koyama et al. / Bioorg. Med. Chem. 17 (2009) 7493–7499
tions were set as follows: nebulizer gas (N2), 14; curtain gas, 10;
ionspray voltage, 5.0 kV; vaporizer temperature, 400 °C. Analyst
software version 1.4.2 (Applied Biosystems) was used for the quan-
titation procedure.
method in Section 2.3 from the supernatant (25 mL) containing
M IS and standard 8-O-methylemodin, -hydroxyemodin, or
3-O-methyl- -hydroxyemodin (1, 5, 50, and 500 nM, respectively).
The ratio of the standard to IS peak area was used for calibration by
linear regression analysis.
2
l
x
x
2.5. LC–APCI-MS/MS conditions
3. Results and discussion
A Cosmosil 5C18 AR-II column (5
l
m, 4.6 Â 150 mm, Nacalai
Tesque, Kyoto, Japan) was used for the LC–APCI-MS/MS. LC separa-
tion was carried out using a mobile phase consisting of 20 mM
ammonium formate (solvent A) and MeOH (solvent B). The follow-
ing stepwise gradient was employed at a flow rate of 1.0 mL/min:
initial A:B = 40:60, 30 min 20:80, 35 min 5:95. Sample volumes of
The LC–MS analyses described in the report represent the first
assay conditions allowing the chromatographic separation and
MS identification of the main metabolites of emodin transformed
in the human B-lymphoid cell line.
50
lL were injected.
3.1. Method development
2.6. S-Adenosyl-
L
-methionine-dependent methylation
An LC–APCI-MS/MS method was developed using a reverse-
phase LC C18 column (Cosmosil 5C18 AR-II) and a stepwise gradient
elution system composed of 20 mM ammonium formate as solvent
A and MeOH as solvent B (initial A:B = 40:60, 30 min 20:80, 35 min
5:95) at a flow rate of 1.0 mL/min. Several mobile phases were
evaluated for the ability to detect six oxidization metabolites of
emodin obtained from treatment of emodin with hepatic
microsomes. Evaluation of formic acid, acetic acid, ammonium for-
mate, and ammonium acetate as potential components of the elu-
tion system indicated that ammonium formate was preferable.
Table 1 provides an overview of MS parameters, including SRM
transitions and energies of the metabolites, and a listing of product
ions from the MS/MS spectra. These product ions were generated
from low-energy collision-induced dissociation of the protonated
molecules.
Raji cells (2.0 Â 106 cells mLÀ1) were incubated in 25 mL of the
serum-free medium containing 5 lM emodin and 300 lM S-aden-
osylmethionine-d3 (SAM-d3) at 37 °C in a CO2 incubator. After 72 h,
the cell suspensions were centrifuged at 1300 rpm for 10 min, and
the supernatant and cells were separated. The extracts were ob-
tained according to a method similar to the one described in Sec-
tion 2.3. The control was acquired using the extract that was
obtained according to the above method without Raji cells in the
serum-free medium containing emodin and SAM-d3.
2.7. Quantification
Quantitative analysis was carried out using the APCI-MS/MS
(selected reaction monitoring: SRM) mode for the precursor/prod-
uct ion of 8-O-methylemodin (283.0/240.0),
(285.0/210.8), 3-O-methyl- -hydroxyemodin (299.0/256.0) and
aloe-emodin (269.9/183.0). To generate the cell extracts for mak-
ing calibration curves, an internal standard (IS; 10 M aloe-emo-
din) and standard 8-O-methylemodin, -hydroxyemodin, or 3-O-
methyl- -hydroxyemodin (5, 50, 200, and 1000 nM, respectively)
were added to the cell pellet followed by the addition of MeOH
to provide a final volume of 0.5 mL. The mixtures were treated as
the similar method in Section 2.3 to obtain the cell extracts. The
supernatant extracts were generated according to the similar
x-hydroxyemodin
x
3.2. Identification of metabolites
l
The extracted ion chromatograms for the negative LC–APCI-MS
analysis of products obtained from the incubation of emodin in Raji
cells are shown inFigure 3. Compared with the control, three major
metabolites were detected in addition to unchanged emodin. Each
metabolite was identified on the basis of chromatographic behav-
ior and characteristic mass spectrometric fragmentation features
observed in the APCI-MS/MS spectra of these compounds. The
x
x
Table 1
LC–APCI-MS/MS parameters and retention times (TR)
Compound
Product ion (m/z)
Precursor ion (m/z)/product ion (m/z)
DPa (V)
FPb (V)
CEc (V)
CXPd (V)
TR (min)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Emodin
2-OHEM
4-OHEM
5-OHEM
7-OHEM
x-OHEM
Emodic acid
8-O-MeEM
241.1, 225.0 182.0
210.9, 201.2 183.0
257.0, 201.2 187.1, 158.8
256.9, 240.8 216.8, 213.0
257.0, 240.9 213.0
268.0, 256.8 241.0, 210.8
267.0, 254.9 227.1
268.0, 240.0
256.0, 227.0
239.9, 183.1
225.0, 181.9
240.0, 183.0
255.0, 239.0 182.9, 199.0
240.0, 183.1
256.0, 239.0
256.0, 239.0
268.3, 239.9
268.9/225.0:- CO2
284.9/210.9:- (2CO + H2O)
284.9/257.0:- CO
À81
À51
À51
À91
À51
À51
À51
À61
À56
À61
À51
À51
À56
À61
À71
À290
À170
À180
À310
À170
À170
À170
À190
À190
À190
À170
À210
À200
À190
À250
À36
À48
À38
À38
À38
À46
À46
À34
À38
À36
À38
À42
À38
À34
À38
À13
À11
À19
À15
À13
À9
21.5
12.5
13.5
30.0
10.9
6.6
3.8
9.5
15.0
35.5
30.4
9.9
8.5
11.4
3.5
3.6
16.7
284.9/256.9:- CO
284.9/240.9:- (2CO + H2O)
285.0/210.8:- (2CO + H2O)
298.9/254.9:- CO2
283.0/240.0:- (CH3 + CO)
299.0/256.0:- (CH3 + CO)
282.9/239.9:- (CH3 + CO)
253.0/225.0:- CO
269.0/183.0:- (2CHO + CO)
283.0/255.0:- CO
282.9/240.0:- (CH3 + CO)
299.0/256.0:- (CH3 + CO)
299.0/256.0:- (CH3 + CO)
299.1/268.3:- OCH3
À21
À19
À15
À15
13
3-O-Me-
x
-OHEM
3-O-MeEM
Chrysophanol
Aloe-emodin
Emodin aldehyde
1-O-MeEM
À11
À17
À19
À15
1-O-Me-
8-O-Me-
x
x
-OHEM
-OHEM
l,3,8-TriOH-6-methoxymethylAQ
À56
À270
À40
À20
AQ: anthraquinone; EM: emodin; Me: methyl; OH: hydroxy.
a
Declustering potential.
Focusing potential.
Collision energy.
Collision exit potential.
b
c
d

J. Koyama et al. / Bioorg. Med. Chem. 17 (2009) 7493–7499
7497
Figure 3. LC–APCI-MS chromatograms of emodin metabolites.
retention times and characteristic fragments used to identify the
metabolites are summarized in Table 1.
the fragment with m/z 256 formed the fragment ion at m/z 227. Be-
cause anthraquinones having b-hydroxy groups are more acidic, it
The compound eluting at 6.6 min (Fig. 3d) gave a deprotonated
molecule at m/z 285, 16 Da more than that of emodin (m/z 285–
269), indicating that this compound was a monohydroxylated
metabolite of emodin. The appearances of product ions at m/z
268 ([MÀHÀOH]), m/z 257 ([MÀHÀCO]), m/z 241 ([MÀHÀCO–
O]), and m/z 211 ([241-HCHO]) suggested that this compound
should be a metabolite generated by oxidation of the exocyclic
methyl group of emodin. Thus, the metabolite was identified as
was thought that 1-O-methyl-
-hydroxyemodin would have earlier retention times in reverse-
phase chromatography. In order to elucidate the structure of the
metabolite at 15.0 min, samples of O-methylated -hydroxyemo-
dins (9, 15, 16) were prepared from emodin for comparison
(Scheme 1). According to the information obtained from the MS/
MS spectrum and the retention times of authentic samples, the
x-hydroxyemodin and 8-O-methyl-
x
x
compound at 15.0 min was identified as 3-O-methyl-
emodin (9). In order to determine the direct precursor of 3-O-
methyl- -hydroxyemodin, metabolites of -hydroxyemodin (6)
and 3-O-methylemodin (10) in Raji cells were studied using LC–
APCI-MS/MS method. These studies indicated that 3-O-methyl-
x-hydroxy-
x
-hydroxyemodin (6) from the information obtained from the
MS/MS spectrum and from comparison with a synthesized authen-
tic sample.
x
x
The compound eluting at 9.5 min (Fig. 3c) showed a deproto-
nated molecule at m/z 283, 14 Da more than that of emodin, indi-
cating that this compound was a monomethylated metabolite of
emodin. Authentic samples of three monomethylemodins (8, 10,
14) were prepared by chemical methylation of emodin to assist
with identification of these metabolites. According to the retention
times and MS/MS spectra of these authentic samples, the com-
pound eluting at 9.5 min was identified as 8-O-methylemodin
(8). The compound eluting at 35.5 min (not visible in Fig. 3c be-
cause of a very small peak) was identified as 3-O-methylemodin
(10).
x-
hydroxyemodin was generated as a metabolite from both 6 and 10.
In addition to the metabolites described above, the compound
eluting at 30.4 min showed a deprotonated molecule at m/z 253,
16 Da less than that of emodin. The product ion at m/z 225 formed
by loss of CO and an ion at m/z 182 ([MÀHÀ2CO–CH3]) were ob-
served in the MS/MS spectrum of this metabolite. According to
the information from the MS analysis and a comparison with
authentic samples, this compound was identified as chrysophanol
(11).
The SRM chromatograms of three metabolites (6, 8, 9) are
shown in Figure 4. The same metabolites were observed when
the extract of cells was analyzed.
Another metabolite eluting at 15.0 min (Fig. 3e), exhibited a
deprotonated molecule at m/z 299, which corresponds to m/z of
emodic acid or to a mass shift of 14 Da (m/z 299–285) more than
The results from our study did not agree with our original pre-
that of
x
-hydroxyemodin. Identification of the metabolite as
diction that only x-hydroxyemodin (6) and 3-O-methyl-x-
emodic acid was ruled out because the retention time of emodic
acid was 3.8 min, not the observed retention time of 15.0 min.
The MS/MS spectrum of this metabolite displayed ions at m/z
299, 256, and 227. The ion at m/z 256 was formed by the loss of
CH3 and CO from m/z 299, and an additional loss of CHO from
hydroxyemodin (9) would be obtained as oxidative metabolites
of emodin.10 The hepatic microsomes derived from various animal
species transformed emodin into several oxidized metabolites cat-
alyzed by cytochrome P450. The reason why only a few oxidative
metabolites were observed in our study may be attributed to the

7498
J. Koyama et al. / Bioorg. Med. Chem. 17 (2009) 7493–7499
fact that the Raji cells have low cytochrome P450-related metabo-
lizing activity.32
d3) was used as the methyl donor to obtain a small amount of
[CD3]-8-O-methylemodin. It was concluded that O-methyl
transferase in Raji cells is different from the previously reported
EOMT.
Although O-methyl emodins were isolated from many fungi, li-
chens, and higher plants, most reports described the isolation of 1-
O-methylemodin (14) or 3-O-methylemodin (10). Only a few arti-
cles have reported that 8-O-methylemodin (8) was isolated as a
metabolite from fungi.33–36 From a biosynthetic point of view,
emodin is the common key intermediate in the biosynthesis of
many natural products, such as monomeric and dimeric anthraqui-
noids,37 xanthones,38 and benzophenones.39 Methylation of the 1-
hydroxy group of emodin is considered to be a key step in the con-
version of anthraquinone to benzophenone in fungi. Methyltrans-
ferases are an important class of enzymes that catalyze the
transfer of a methyl group from several donors including SAM, co-
lin, betain, 5-methyltetrahydrofolic acid, and dimethylrutin, to a
wide variety of substrates. SAM, the primary methyl donor present
in all living organisms, is generated by methionine adenosyl trans-
ferase. SAM is involved in the methylation of DNA, proteins, and
lipids, and in polyamine synthesis.40,41 Emodin O-methyltransfer-
ase (EOMT) from Penicillium frequentans was described by Geten-
beck and Malmstrom,42 and Chen at al. reported EOMT derived
from Aspergillus terreus.43 EOMT catalyzes methylation of the 1-hy-
droxy group of emodin to form 1-O-methylemodin. Therefore, it
was concluded that O-methylation of emodin to 8-O-methylemo-
din in Raji cells was catalyzed by another O-methyl transferase that
uses SAM as the methyl donor. An isotopically labeled SAM (SAM-
3.3. Validation results for emodin metabolites
The linearity, precision, and sensitivity of the method were ana-
lyzed by applying methanolic solutions of
x-hydroxyemodin (6),
8-O-methylemodin (8), 3-O-methyl- -hydroxyemodin (9), and IS
x
to the LC–APCI-MS/MS system. The linearity of the calibration
curves was calculated from the representation of the ratio of the
peak area of emodin metabolite to the peak area of IS versus emo-
din metabolite concentration. The data were collected for four dif-
ferent concentrations of emodin metabolites from the supernatant
(1, 5, 50, 500 nM) and cells (5, 50, 200, 1000 nM) with IS. Table 2
shows the slope, intercept, and correlation coefficient values ob-
tained from linear least-squares regression analysis of metabolites
(6, 8, 9). As shown in Table 3, the relative standard deviation (RSD)
values for intraday and interday retention times of metabolites and
IS were 0.48–1.05% and 2.45–4.82%, respectively. The intraday RSD
values of the area ratio were less than 6.59%. The detection limits
were 71.5, 14.3, and 150.0 pg/mL for 6, 8, and 9, respectively, and
these values were three times the signal/noise (S/N) ratio. These re-
sults proved that the method for the measurement of emodin
metabolites was reliable and reproducible.
Figure 4. SRM chromatograms of supernatant extract solution in MeOH: (a) total ion chromatogram, (b) 8-O-methylemodin, (c)
x-hydroxyemodin, (d) 3-O-methyl-x-
hydroxyemodin.

J. Koyama et al. / Bioorg. Med. Chem. 17 (2009) 7493–7499
7499
Table 2
Regression analysis equations for emodin metabolites
Compound
Regression equation supernatant extractb
Correlation coefficient
Regression equation cells extractc
Correlation coefficient
6a
8a
9a
y = 0.0021x À 0.004
y = 0.0466x + 0.1156
y = 0.0018x + 0.0063
r
r
r
2 = 0.9999
2 = 0.9998
2 = 0.9995
y = 0.0003x + 0.0069
y = 0.0103x + 0.1191
y = 0.0004x À 0.0066
r
r
r
2 = 0.9976
2 = 0.9991
2 = 0.9992
y: peak area ratio of metabolite to IS, x: concentration in nM.
a
Quantitative analysis was carried out using the precursor/product ion: m/z 285/211, 283/240, 299/256.
1, 5, and 50, 500 nM.
5, 50, 200, and 1000 nM.
b
c
Table 3
4. Cohen, P. A.; Hudson, J. B.; Toweres, G. H. N. Experientia 1996, 52, 180.
Relative standard deviation (RSD) and limit of detection (LOD) of emodin related
compounds
5. Jing, X.; Ueki, N.; Cheng, J.; Imanishi, H.; Hada, T. Jpn. J. Cancer Res. 2002, 93,
874.
6. Caldon, C. E.; Daly, R. J.; Sutherland, R. L.; Musgrove, E. A. Cell Biochem. 2006, 97,
261.
7. Jayasuriya, H.; Koonchanok, N. M.; Geahlen, R. L.; McLaughlin, J. L.; Chang, C. J. J.
Nat. Prod. 1992, 55, 696.
Compound RSD of Tr/
intraday
RSD of Tr/
interday
(n = 5, %)
RSD of area ratio
(supernatant,
n = 3)
LOD (pg/
mL) (S/
N = 3)
(n = 10, %)
8. Majsterek, I.; Pytel, D.; Blasiak, J. Postepy Biochem. 2005, 51, 251.
9. Wakeling, A. E. Endocr.-Relat. Cancer 2005, 12, S183.
10. Morooka, N.; Nakano, S.; Itoi, N.; Ueno, Y. Agric. Biol. Chem. 1990, 54, 1247.
11. Masuda, T.; Haraikawa, K.; Morooka, N.; Nakano, S.; Ueno, Y. Mut. Res. 1985,
149, 327.
12. Tanaka, H.; Morooka, N.; Haraikawa, K.; Ueno, Y. Mut. Res. 1987, 176, 165.
13. Murakami, H.; Kobayashi, J.; Masuda, T.; Morooka, N.; Ueno, Y. Mut. Res. 1987,
180, 147.
1
6
8
9
IS
1.01
1.05
0.82
0.55
0.48
2.53*
4.82
2.58
2.45
3.36
2.00
5.45
6.59
4.26
—
13.5
71.5
14.3
150.0
—
*
n = 3.
14. Mueller, S. O.; Stopper, H.; Dekant, W. Drug Metab. Dispos. 1998, 26, 540.
15. Song, R.; Xu, F.; Zhang, Z.; Liu, Y.; Dong, H.; Tian, Y. Biomed. Chromatogr. 2008,
22, 1230.
The detection limit of emodin for HPLC, capillary electrophore-
sis (CE), and the present LC–APCI-MS/MS method were 40 ng/mL,44
100 ng/mL,45 and 13.5 pg/mL, respectively. Thus, the sensitivity of
LC–MS/MS analysis was superior to that of HPLC and CE. Compared
with spectrophotometric detection methods of HPLC and CE, the
LC–MS/MS method is more specific to the detection of prespecified
structures because specific ions can be selected. Therefore, the LC–
MS/MS method described herein is more useful for the detection of
emodin derivatives than standard approaches.
16. Koyama, J.; Inoue, M.; Morita, I.; Kobayashi, N.; Osakai, T.; Nishino, H.; Tokuda,
H. Cancer Lett. 2006, 241, 263.
17. Tian, T.; Sun, O.; Shen, J.; Zhang, T.; Gao, P.; Sun, Q. World J. Microbiol.
Biotechnol. 2008, 24, 861.
18. Tang, W. F.; Yu, Q.; Wan, M.-. H.; Qin, F.; Wang, Y. G.; Chen, G. Y.; Liang, M. Z.;
Huang, X. Biomed. Chromatogr. 2007, 21, 701.
19. Malone, M. A.; Weber, P. L.; Smyth, M. R.; Lunte, S. M. Anal. Chem. 1994, 66,
3782.
20. Heavner, D. L.; Richardson, J. D.; Morgan, W. T.; Ogden, N. W. Biomed.
Chromatogr. 2005, 19, 312.
21. Hoizey, G.; Goglin, A.; Malinovsky, J.-M.; Robinet, A.; Binet, L.; Kaltenbach, M.
L.; Milart, H.; Lamiable, D. J. Pharm. Biomed. 2006, 42, 593.
22. Li, L.; Ye, M.; Bi, K.; Guo, D. Biomed. Chromatogr. 2006, 20, 95.
23. Lee, S. K.; Kim, G. H.; Kim, D. H.; Jahng, Y.; Jeong, T. C. Biol. Pharm. Bull. 2007, 30,
1991.
24. Koyama, J.; Morita, I.; Tagahara, K.; Aqil, M. Phytochemistry 1997, 56, 849.
25. Banks, H. J.; Cameron, D. W.; Raverty, W. D. Aust. J. Chem. 1978(31), 2271.
26. Hirose, Y.; Suehiro, Y.; Furukawa, Y.; Murakami, T. Chem. Pharm. Bull. 1982, 30,
4186.
4. Conclusion
A rapid, simple, and sensitive LC–APCI-MS/MS method for the
determination of emodin metabolites in Raji cells, derived from a
human B-lymphoid cell line, has been developed. The starting
material emodin and its five metabolites were detected in the Raji
cell culture. In addition to the major metabolite 8-O-methyl emo-
27. Oswald, B.; Gruber, M.; Böhmer, M.; Lehmann, F.; Probst, M.; Wolfbeis, O. S.
Photochem. Photobiol. 2001, 74, 237.
28. Kalidhar, S. B. Phytochemistry 1989, 28, 3459.
29. Fujimoto, H.; Nakamura, E.; Okuyama, E.; Ishibashi, M. Chem. Pharm. Bull. 2004,
52, 1005.
30. Hadj-Hamdri, A.; Vidal-Cros, A.; Gaudry, M.; Sobrio, F.; Rousseau, B. J. Labelled
Compd. Radiopharm. 1995, 36, 795.
31. Koyama, J.; Nishino, Y.; Morita, I.; Kobayashi, N.; Osakai, T.; Tokuda, H. Bioorg.
Med. Chem. Lett. 2008, 18, 4106.
din,
x-hydroxyemodin, 3-O-methyl-x-hydroxyemodin, 3-O-meth-
ylemodin (physcion), and chrysophanol were detected. This is the
first report describing the elucidation of metabolites of emodin in
a human cell line. Investigation of the novel O-methyl transferase
species found in Raji cells will be continued, and the results will
be reported in due course.
32. Yang, Q.; Hergenhahn, M.; Weninger, A.; Bartsch, H. Carcinogenesis 1999, 20,
1769.
33. Brunauer, G.; Hager, A.; Grube, M.; Tuerk, R.; Stocker-Woergoetter, E. Plant
Physiol. Biochem. 2007, 45, 146.
Acknowledgements
34. Gill, M.; Morgan, P. M. ARKIVOC 2001, 7, 145.
This study was supported in part by grants from the Ministry of
Education, Culture, Sports, Science and Technology of Japan, and by
the ‘High-Tech Research Center’ Project for Private Universities: a
matching fund subsidy from the Ministry of Education, Culture,
Sports, Science and Technology, 2004–2008.
35. Ayer, W. A.; Trifonov, L. S. J. Nat. Prod. 1994, 57, 317.
36. Shu, Y.-. Z.; Arcuri, M.; Kozlowski, M. R.; Wang, R. R.; Lam, K. S.; Chang, L.-. P.;
Pirnik, D. M.; Rose, P. M. J. Antibiot. 1994, 47, 1328.
37. Sankawa, U.; Ebizuka, Y.; Shibata, S. Tetrahedron Lett. 1973, 23, 2125.
38. Birch, A. J.; Baldas, J.; Hlubucek, J. R.; Simpson, T. J.; Westerman, P. W. J. Chem.
Soc., Perkin Trans. 1 1976, 898.
39. Franck, B.; Hueper, F. Angew. Chem. 1966, 78, 752.
40. Lu, S. C. Int. J. Biochem. Cell Biol. 2000, 32, 391.
References and notes
41. Miner, S. E.; Evrovsky, J.; Cole, D. E. Clin. Biochem. 1997, 30, 189.
42. Gatenbeck, S.; Malmstrom, L. Acta Chem. Scand. 1969, 23, 3493.
43. Chen, Z. G.; Fujii, I.; Ebizuka, Y.; Sankawa, U. Arch. Microbiol. 1992, 158, 29.
44. Koyama, J.; Morita, I.; Kobayashi, N. J. Chromatogr., A 2007, 1145, 183.
45. Koyama, J.; Morita, I.; Fujiyoshi, H.; Kobayashi, N. Chem. Pharm. Bull. 2005, 53,
573.
1. Zhang, L.; Lau, Y. K.; Xi, L.; Hong, R. L.; Kim, D. S.; Chen, C. F.; Hortobagyi, G. N.;
Chang, C. J.; Hung, M. C. Oncogene 1998, 16, 2855.
2. Kuo, Y. C.; Sun, C. M.; Ou, J. C.; Tsai, W. J. Life Sci. 1997, 61, 2335.
3. Ubbink-Kok, T.; Anderson, J. A.; Konings, W. N. Antimicrob. Agents Chemother.
1986, 30, 147.