Identification and characterization of human UDP-glucuronosyltransferases responsible for the in-vitro glucuronidation of arctigenin

Article abstract of DOI:10.1111/jphp.12483

Objectives This study aimed to characterize the glucuronidation pathway of arctigenin (AR) in human liver microsomes (HLM) and human intestine microsomes (HIM). Methods HLM and HIM incubation systems were employed to catalyse the formation of AR glucuronide. The glucuronidation activity of commercially recombinant UGT isoforms towards AR was screened. A combination of chemical inhibition assay and kinetic analysis was used to determine the UGT isoforms involved in the glucuronidation of AR in HLM and HIM. Key findings AR could be extensively metabolized to one mono-glucuronide in HLM and HIM. The mono-glucuronide was biosynthesized and characterized as 4′-O-glucuronide. UGT1A1, 1A3, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7 and 2B17 participated in the formation of 4′-O-G, while UGT2B17 demonstrated the highest catalytic activity in this biotransformation. Both kinetic analysis and chemical inhibition assays demonstrated that UGT1A9, UGT2B7 and UGT2B17 played important roles in AR-4′-O-glucuronidation in HLM. Furthermore, HIM demonstrated moderate efficiency for AR-4′-O-glucuronidation, implying that AR may undergo a first-pass metabolism during the absorption process. Conclusion UGT1A9, UGT2B7 and UGT2B17 were the major isoforms responsible for the 4′-O-glucuronidation of AR in HLM, while UGT2B7 and UGT2B17 were the major contributors to this biotransformation in HIM.

Full text of DOI:10.1111/jphp.12483

Research Paper
Journal of Pharmacy
And Pharmacology
Identification and characterization of human
UDP-glucuronosyltransferases responsible for the in-vitro
glucuronidation of arctigenin
Hong Xina*, Yang-Liu Xiab*, Jie Houc, Ping Wangb, Wei Hea, Ling Yangb, Guang-Bo Geb and Wei Xua
b
aDepartment of Integrated Traditional and Western Medicine, First Affiliated Hospital of Harbin Medical University, Harbin, Laboratory of
c
Pharmaceutical Resource Discovery, Dalian Institute of Chemical Physics, Chinese Academy of Sciences and Department of Biochemistry and
Molecular Biology, Dalian Medical University, Dalian, China
Keywords
Abstract
arctigenin; glucuronidation; human intestine
Objectives This study aimed to characterize the glucuronidation pathway of
arctigenin (AR) in human liver microsomes (HLM) and human intestine
microsomes (HIM).
Methods HLM and HIM incubation systems were employed to catalyse the for-
mation of AR glucuronide. The glucuronidation activity of commercially recom-
binant UGT isoforms towards AR was screened. A combination of chemical
inhibition assay and kinetic analysis was used to determine the UGT isoforms
involved in the glucuronidation of AR in HLM and HIM.
microsomes; human liver microsomes;
UDP-glucuronosyltransferases
Correspondence
Guang-Bo Ge, Laboratory of Pharmaceutical
Resource Discovery, Dalian Institute of
Chemical Physics, Chinese Academy of
Sciences, No. 457 Zhongshan Road, Dalian,
116023, China.
E-mail: geguangbo@dicp.ac.cn
Key findings AR could be extensively metabolized to one mono-glucuronide in
HLM and HIM. The mono-glucuronide was biosynthesized and characterized as
4′-O-glucuronide. UGT1A1, 1A3, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7 and 2B17 par-
ticipated in the formation of 4′-O-G, while UGT2B17 demonstrated the highest
catalytic activity in this biotransformation. Both kinetic analysis and chemical
inhibition assays demonstrated that UGT1A9, UGT2B7 and UGT2B17 played
important roles in AR-4′-O-glucuronidation in HLM. Furthermore, HIM demon-
strated moderate efficiency for AR-4′-O-glucuronidation, implying that AR may
undergo a first-pass metabolism during the absorption process.
Wei Xu, First Affiliated Hospital of Harbin
Medical University, No. 23 Youzheng Street,
Harbin, 150001, China.
E-mail: 13796059990@163.com
Received April 17, 2015
Accepted July 28, 2015
doi: 10.1111/jphp.12483
Conclusion UGT1A9, UGT2B7 and UGT2B17 were the major isoforms respon-
sible for the 4′-O-glucuronidation of AR in HLM, while UGT2B7 and UGT2B17
were the major contributors to this biotransformation in HIM.
*These authors contributed equally to this
work.
Introduction
Fructus Arctii (Niubangzi), the dried fruits of Arctium lappa
L. (also known as burdock) (Asteraceae), is one of the most
popular traditional Chinese medicines. Fructus Arctii has
been widely used for dispelling pathogenic wind-heat, pro-
moting eruption, relieving sore throat, removing toxic sub-
stances and subduing swelling.[1] In China, a unique tea
made from Fructus Arctii is commonly used as a health care
product, which is useful for the treatment of tonsillitis,
pharyngolaryngitis and constipation.[2] Arctigenin (AR,
Figure 1) is one of the major lignin compounds in Fructus
Arctii, with a high content of 9.7 mg/g of dry materials.[3]
AR has been reported to exhibit anti-influenza,[4] anti-
HIV,[5] anticancer through inducing cytotoxicity[6] and
apoptosis,[7] antioxidative,[8] chemopreventive, anti-
inflammatory,[9] immunomodulative,[10] and anti-diabetic[11]
activity. AR and its glucoside arctiin (AR-4′-O-β-glucoside)
are phenylpropanoid dibenzylbutyrolactone lignans; they
often coexist in several traditional Chinese herbals, such as
Arctium lappa L., Torreya nucifera and Saussurea medusa,
etc.[12] The amount of arctiin in these herbals is much
higher than that of AR usually by an order of magnitude.[13]
A
previous study reported that arctiin (AR-4′-O-β-
glucoside) could be extensively hydrolysed to AR by
intestinal microflora, after oral administration of arctiin
in rat.[12] This finding suggested that the aglycone rather
than arctiin itself was the major constitute in circulating
system after oral administration of arctiin-containing
herbals; it also implied that the in-vivo pharmaceutical
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681
1673

In-vitro glucuronidation of arctigenin
Hong Xin et al.
O
9’
O
O
9’
O
7’
2’
2’
5’
7’
8’
1’
O
3’
4’
3’
4’
O
8’
1’
6’
OH
6”
8
HLM: UGT1A9 2B7 2B17 1A1 1A3 2B4
8
6’
HO
5”O 1”
O
7
9
7
9
5’
2
O
1
1
HIM: UGT2B7 2B17 1A1 1A3 1A7 1A8 1A10
2”
4”
HO
2
3”
6
5
6
5
OH
3
3
O
O
OH
4
4
O
O
arctigenin
arctigenin-4’-O-glucuronide
Figure 1 Glucuronidation pathway of arctigenin.
effects of arctiin should be attributed to AR or its constitu-
tive metabolite(s).
androsterone and testosterone were purchased from Sigma-
Aldrich (St Louis, MO, USA). Pooled HLM from 11 donors
(Lot. SUBK) and pooled SD rat liver microsomes from 100
donors (Lot. BDVH) were purchased from the Research
Institute for Liver Diseases (Shanghai, China), and pooled
HIM from 10 donors (Lot. UGU) was purchased from
Celsis (Baltimore, MD, USA). A panel of recombinant
human UGT isoforms (UGT1A1, −1A3, −1A4, −1A6, −1A7,
−1A8, −1A9, −1A10, −2B4, −2B7, −2B15, and −2B17)
expressed in baculovirus-infected insect cells were pur-
chased from BD Gentest (Woburn, MA, USA). All other
reagents were of HPLC grade or of the highest grade com-
mercially available.
It was reported that AR could undergo both phase I and
phase II metabolisms after administration in the rat; the
biotransformations including hydrolysis, glucuronidation
and demethylation lead to formations of arctigenic acid,
AR-4′-O-glucuronide and 4-O-demethyl-AR, respec-
tively.[12] It also has been revealed that arctigenic acid and
AR-4′-O-glucuronide, rather than parent compound AR or
4-O-demethyl-AR, were the major compounds found in
plasma after oral administration of AR in rats.[14]
However, most of the previous studies concerning the
glucuronidation process of AR were conducted in rat or in
rat liver microsomes (RLM), and the glucuronidation
pathway of AR in human tissues has not been reported
yet. Therefore, it is necessary to characterize the
glucuronidation pathway(s) of AR in human tissues.
Glucuronidation assay of arctigenin in
human liver microsomes by LC-MS
The volumes of 200-μl incubation mixture contained
50 mM Tris-HCl buffer (pH 7.4), 5 mM MgCl2, 4 mM
UDPGA, 25 μg/ml alamethicin, 0.1 mg/ml HLM and
100 μM AR. The reaction was initiated by the addition of
UDPGA after pre-incubation at 37°C for 3 min. At 30-min
incubation, the reaction was terminated with 200 μl cold
acetonitrile, followed by centrifugation at 20 000 × g for
20 min to obtain the supernatant for high-performance
liquid chromatography spectrometry (HPLC) analysis.
Control incubations without UDPGA or without substrate
or without microsomes were performed to confirm
that the metabolite produced were microsomes- and
UDPGA-dependent.
It is well known that identification of major metabolites
and the involved drug-metabolizing enzymes is very helpful
for the elucidation of the metabolic and elimination path-
ways of a given drug or xenobiotic in human body, and for
the deep understanding of its in-vivo effects. Therefore, this
study aimed to characterize the glucuronidation pathway(s)
of AR and to identify the human UGT isoform(s) respon-
sible for the formation of the generated metabolite(s), by
using recombinant human UGTs, human liver microsomes
(HLM) and human intestine microsomes (HIM). In addi-
tion, the contributions of the major human UGTs in HLM
and HIM, as well as the hepatic and intestinal clearance of
AR via glucuronidation pathway, were also investigated.
Enzymatic hydrolysis was performed to confirm the
glucuronide of AR. After incubation for 30 min in 200-μl
glucuronidation reaction mixture (without D-saccharicacid
1,4-lactone), an equal volume of 0.15 M acetate buffer
(pH 5.0) with or without 1800 Fishman units of
β-glucuronidase was added to the reaction mixture. After
incubation for another 30 min at 37°C, the hydrolysis reac-
tion was terminated by addition of 200-μl cold acetonitrile,
and then centrifuged at 20 000 × g for 20 min; the superna-
tant was injected to HPLC-UV analysis.
Materials and Methods
Materials
AR, magnolol and nilotinib were purchased from
Sichuan Victory Biotechnology Co. Ltd. (Sichuan, China).
Alamethicin, Brij 58, magnesium chloride, D-saccharic acid
1,4-lactone, β-glucuronidase (EC No. 3.2.1.31), uridine-
diphospho-glucuronic acid trisodium salt (UDPGA),
1674
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681

Hong Xin et al.
In-vitro glucuronidation of arctigenin
loading, the SPE cartridge was sequentially eluted with
12-ml Millipore water, 12-ml methanol and 12-ml metha-
nol containing 5% formic acid. The entire process was
monitored by HPLC, and the metabolite was assembled in
methanol containing 5% formic acid. After vacuum evapo-
ration, the residue was re-dissolved in 1-ml methanol and
separated by HPLC. The purity was greater than 95%
according to HPLC-UV analysis. The structure of metabo-
lite was determined by nuclear magnetic resonance (NMR),
and the experiment contained a Varian INOVA-500 NMR
spectrometer (Varian, Palo Alto, CA, USA). The purified
5.4-mg metabolites were stored at −20°C before dissolving
in dimethyl sulfoxide-d6 (Euriso-Top, Saint-Aubin, France)
for NMR analysis. Chemical shifts were given on δ scale and
referenced to tetramethylsilane at 0 ppm for 1H-NMR
(500 MHz) and 13C-NMR (100 MHz).
Analytical instruments and conditions
AR and its glucuronide were analysed by an LC system
(Shimadzu, Kyoto, Japan) containing an SCL-30A system
controller, two LC-30AD pumps, a SIL-30AC auto-injector,
a DGU-20AC vacuum degasser, a CTO-30AC column oven
and a SPD-20AVP UV detector. A Shim-pack VP-ODS
(5 μm, 150.0 mm × 2.1 mm, Shimadzu) analytical column
with an ODS guard column (2.2 μm, 5.0 mm × 2.0 mm)
was used to separate AR and its glucuronide. Column tem-
perature was kept at 40°C. Acetonitrile (A) and water with
0.2% formic acid (B) as mobile phase at a flow rate of
0.4 ml/min, with a gradient of 0–9.0 min, 90% B-45% B;
9.0–12.0 min, 45% B; 12.0–15.0 min, balance to 90% B. The
detection wavelength was set at 280 nm. Mass detection was
performed on a Shimadzu LC-MS-2010EV instrument with
an ESI interface both in positive and negative ion mode
from m/z 200 to 800. The detector voltage was set at
+1.55 kV and 11.55 kV for positive and negative ion detec-
tion, respectively. The curved desolvation line temperature
(CDL) and the block heater temperature were both set at
250°C, while the CDL voltage was set at 40V. Other MS
detection conditions were set as follows: interface voltage,
+4.5 kV and 14.0 kV for positive and negative ion detection,
respectively; nebulizing gas (N2) flow, tuned to be 1.5 L/
min; and the drying gas (N2) pressure 0.06 MPa. Data pro-
cessing was performed using HPLC-MS Solution version
3.41 software (Shimadzu).
Assay with recombinant UGTs
AR glucuronidation was measured in reaction mixtures
containing recombinant human UGT1A1, 1A3, 1A4, 1A6,
1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B15 and 2B17. The incuba-
tions were conducted as described above for the HLM
study, in the presence of 100 μm substrate. The final
protein concentration of the reactions was 0.1 mg/ml
and incubated at 37°C for 60 min. Liquid chromatography
with ultraviolet (LC-UV) and liquid chromatography-
electrospray ionization-mass spectrometry (LC-ESI-MS)
detection was used to monitor the metabolite.
Quantification of AR glucuronidation in the incubation
mixtures was accomplished by using the standard curve of
the glucuronide, which has a linear range from 0.1 to 10 μm
(the correlation coefficient was 0.999). The accuracy and
precision of the intra-day and inter-day error were both less
than 3%.
Chemical inhibition assays
Glucuronidation of AR in pooled HLM and HIM was
measured in the absence or presence of an inhibitor
(nilotinib, magnolol, androsterone or testosterone).
Nilotinib is a selective inhibitor of UGT1A1.[15] Magnolol is
an effective inhibitor for UGT1A9,[16] and androsterone is
reported to be able to inhibit both UGT1A9 and
UGT2B7.[17] Testosterone is a specific substrate of UGT2B17
with the Km value of 10 μm.[18] AR (15 μm in HLM and
25 μm in HIM) was incubated in the absence or presence of
nilotinib (10 μm), magnolol (1 μm) androsterone (10 μm)
or testosterone (25 μm). The incubations were performed
for 15 min at a protein concentration of 0.01 mg/ml in
HLM or of 0.05 mg/ml in HIM.
Biosynthesis of glucuronide and structural
characterization
The metabolite of AR was biosynthesized using RLM and
purified for structure elucidation and quantitative analysis.
In brief, 1 mM AR was incubated with RLM (0.5 mg
protein/ml), 0.1 M Tris-HCl (pH 7.4), 10 mM MgCl2, Brij
58 (0.5 mg/mg protein), 10 mM D-saccharic acid 1,4-
lactone, and 4 mM UDPGA in 20 ml of final incubations
for 12 h at 37°C. The stock solution of AR (100 mM) was
prepared in DMSO. The concentration of organic solvent in
the final incubation was 1%. The reaction was terminated
by adding 20 ml acetonitrile and then transferred the vessel
to an ice bath and cooled for 20 min. After removal of
protein by centrifugation at 20 000 × g for 30 min at 4°C,
the combined supernatants were loaded on an SPE car-
tridge (C18, 1000 mg, Agela Technologies, Radnor, PA,
USA), which was preconditioned by sequential washing
with 6-ml methanol and 6-ml Millipore water. After sample
Kinetic characterization
The formation rate of AR glucuronide was linear over
20 min of incubation and 0.01–0.1 mg of microsomal
protein. To ensure that less than 10% of substrate was
metabolized in all incubations, the kinetic determinations
were carried out using a microsomal protein concentration
of 0.01 mg/ml (HLM) and 0.05 mg/ml (HIM) with 15-min
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681
1675

In-vitro glucuronidation of arctigenin
Hong Xin et al.
incubation. To estimate kinetic parameters, AR (1–200 μm)
was incubated with pooled HLM and HIM. For recombi-
nant human UGTs, AR (1–200 μm) was incubated with
each isoenzyme for 15 min; the final protein concentration
was 0.1 mg/ml for UGT1A1 and UGT1A7, 0.02 mg/ml for
UGT1A9, 0.05 mg/ml for UGT2B7, and 0.025 mg/ml for
UGT2B17. Kinetic parameters for AR glucuronidation in
HLM, HIM and UGTs were obtained by fitting the
Michaelis–Menten equation to experimental data using
GraphPad Prism 6.0 (GraphPad Software, San Diego, CA,
USA). The Michaelis–Menten equation is v = Vmax[S]/
(Km+[S]), where v is the rate of reaction, Vmax is the
maximum velocity, Km is the Michaelis constant (substrate
concentration at 0.5 Vmax), and [S] is the substrate concen-
tration. Results were expressed as mean S.E. of triplicate
independent determinations.
addition, AR glucuronide can be rapidly hydrolysed by
β-glucuronidase to the parent AR (Figure 2b), further
proving that it is the mono-glucuronide metabolite.
Then the metabolite was biosynthesized by using RLM,
and then its chemical structure was well characterized by
1
1H-NMR and 13C-NMR. The H-NMR and 13C-NMR spec-
tral data of the metabolite were listed in Table 1. Compared
with NMR data of AR, the 13C-NMR spectrum of the
glucuronide displayed the carbon signal of C-4′ (δ) shifted
upfield to δ144.75 (Δδ -0.32), which is probably due to a
glycosidation shift of a phenolic compound. In the sugar
moiety, the G1 protons and carbons exhibited characteristic
chemical shifts near 5 and 100 ppm, whereas the G6 carbon
(-COOH) showed chemical shifts of approximately
170 ppm. In addition, the β-configuration of the glucuronic
acid moiety was confirmed by the coupling constant of the
anomeric proton (δ5.01, J = 7.0 Hz), which agreed well with
previous reports regarding the O-glucuronides.[19] All these
evidence clearly demonstrated that the glucuronidation site
was located at the C-4′ phenolic group of AR and the glu-
curonide was identified as AR-4′-O-glucuronide (Figure 1).
Statistical analysis
Data from chemical inhibition studies are expressed as
means and SDs. Statistical analyses are performed with one-
way analysis of variance. Differences were considered statis-
tically significant when P < 0.05.
Assay with recombinant human UGTs
Results
Twelve recombinant UGTs, namely UGT1A1, 1A3, 1A4,
1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B15 and 2B17, were
used to catalyse AR glucuronidation and determine the
pivotal isoforms responsible for the metabolism pathway.
With the substrate concentration of 100 μm, UGT1A1,
UGT1A3, UGT1A7, UGT1A9, UGT1A10, UGT2B4,
UGT2B7 and UGT2B17 participated in the formation of
AR-4′-O-glucuronide (Figure 3). Compared with other
isoforms, UGT2B17 displayed a relative higher activity to
the formation of AR-4′-O-glucuronide, followed by
UGT1A9, UGT1A1 and UGT2B7, while UGT1A3, UGT1A8
and UGT1A10 demonstrated relative low efficiency in the
glucuronidation of AR.
Identification of arctigenin glucuronide
After incubation with AR (100 μm) and HLM (0.1 mg/ml)
in the presence of UDPGA, the metabolite was identified by
HPLC-ESI-MS, and the retention time of the one new peak
was 6.9 min (Figure 2a). The peak was not detected in the
control incubations without UDPGA, AR or microsomes.
The negative ion mode was adopted for identification AR
and its glucuronide, as it is more sensitive than the positive
ion mode for these two compounds. The mass spectra was
dominated by [M-H]− ion at m/z 547.4, corresponding to
the mono-glucuronide of AR with characteristic m/z 176. In
(a)
(b)
(×10,000)
(×10,000)
1.75
3.00
2.50
2.00
1.50
1.00
0.50
0
280 nm
Arctigenin
280 nm
1.50
Arctigenin
1.25
1.00
0.75
0.50
0.25
0
M
M
+UDPGA
–UDPGA
+ β-glucuronidase
0
2
4
6
8
10
12
0
2
4
6
8
10
12
Time (min)
Time (min)
Figure 2 Representative LC-UV profile of arctigenin and its glucuronide in human liver microsomes. (a) Arctigenin was incubated in human liver
microsomes with (solid line) and without (dotted line) UDPGA for 30 min. (b) Arctigenin was incubated in human liver microsomes without (solid
line) and with (dashed line) β-glucuronidase for 30 min.
1676
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681

Hong Xin et al.
In-vitro glucuronidation of arctigenin
Table 1 1H-NMR and 13C-NMR data for AR and its glucuronide
Control
Nilotinib 10 µM
Mangnolol 1 µM
Androsterone 10 µM
Testosterone 25 µM
AR
AR-O-glucuronide
Position
δ1HMult (J in Hz)
δ
13C
δ1HMult (J in Hz)
δ
13C
1
–
128.83
113.44
147.36
147.29
111.87
120.33
36.83
40.74
70.61
131.21
112.41
148.64
145.07
115.25
121.51
33.67
45.60
178.36
55.32
55.44
55.51
–
–
131.13
113.91
148.64
147.29
111.89
120.37
36.85
2
6.59 d (4.6)
6.58 d (8.0)
–
100
50
0
3
–
4
–
–
***
***
***
5
6.74
6.78
6
6.60 d (1.8)
6.58 d (8.0)
2.71
***
***
7
2.50
***
***
8
2.68
2.72
40.70
9
4.07
4.12
71.36
1′
2′
3′
4′
5′
6′
7′
8′
–
–
132.18
112.41
148.78
144.75
115.16
121.22
33.55
6.68 d(8.0)
6.67 d (7.9)
–
–
HLM
HIM
–
–
6.82 d(8.1)
6.82 d (8.1)
6.59 d (8.0)
3.05 d (7.0)
2.74
Figure 4 Inhibitory effects of nilotinib, magonolol, androsterone and
testosterone on the formation of arctigenin glucuronidated metabolite
in human liver microsomes and human intestine microsomes.
Arctigenin (15 μM in human liver microsomes and 25 μM in human
intestine microsomes) was incubated with human liver microsomes
(0.01 mg of protein/ml) or human intestine microsomes (0.05 mg of
protein/ml) with either nilotinib (10 μM), magonolol (1 μM),
androsterone (10 μM) or testosterone (25 μM) at 37°C for 15 min.
***P < 0.001 compared with control. Data represent the mean SD
(n = 3).
6.63 d(1.8)
2.84 d (5.3)
2.74
45.51
9′
–
–
178.32
55.36
CH3O
CH3O
CH3O
G1
G2
G3
G4
G5
G6
3.70 (3H, s)
3.81 (3H, s)
3.83 (3H, s)
3.86 (3H, s)
5.01 d (7.0)
3.37 m
3.31 m
3.39 m
3.69 m
–
3.70 (3H, s)
55.45
3.70 (3H, s)
55.59
–
–
–
–
–
–
99.88
–
72.93
–
75.75
–
71.36
–
76.01
Chemical inhibition assays
–
169.18
AR, arctigenin.
To reveal the importance of UGT1A1, UGT1A9, UGT2B7
and UGT2B17 in the formation of AR-4′-O-glucuronide in
HLM and in HIM, a series of inhibition assays were eluci-
dated by using selective inhibitors. As shown in Figure 4,
magnolol and testosterone exhibited similar inhibitory
effects on AR-4′-O-glucuronidation. The residual activity of
HLM was 65% and 73% for the formation of AR-4′-O-
glucuronide in the presence of magnolol and testosterone,
respectively, indicating that the contribution of UGT1A9 or
UGT2B17 in the formation of the glucuronide was about
30%, respectively. For the formation of AR-4′-O-
glucuronide, the remaining activity of HLM was less than
40% in the presence of androsterone, indicating that
UGT2B7 also played an important role in AR-4′-O-
glucuronidation. Furthermore, upon addition of nilotinib,
which could inhibit both UGT1A1 and UGT2B17 (data not
show), the formation of AR-4′-O-glucuronide in HLM
could be inhibited by about 40% of control, implying that
the contribution of UGT1A1 for AR-4′-O-glucuronidation
was about 10%. The similar inhibition assays were also con-
ducted in HIM, as shown in Figure 4; androsterone or tes-
tosterone could inhibit the formation of AR-4′-O-
glucuronide in HIM significantly with the residual activity
of 50%, while the formation of AR-4′-O-glucuronide in
HIM could be inhibited near 60% by addition of nilotinib.
8
6
4
2
0
100 µM
UGT isoforms
Figure 3 The formation of AR-4′-O-glucuronide catalysed by various
recombinant human UGTs. Arctigenin (100 μM) was incubated with
various recombinant human UGTs (0.1 mg/ml) at 37°C for 60 min.
Data represent the mean of triplicate incubations.
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681
1677

In-vitro glucuronidation of arctigenin
Hong Xin et al.
Table 2 Kinetic parameters of arctigenin AR glucuronidation in
human liver microsomes, human intestine microsomes and recombi-
nant UGTs
20 years. Modern pharmacological studies around the world
have clearly shown that AR possesses diverse pharmacologi-
cal activity, including anti-influenza, anti-oxidation and
immunomodulation, and most of these activities are ben-
eficial to human health.[20] In contrast to the extensive
studies on the pharmacological activity, the metabolic path-
ways and metabolic behaviours of AR in human and experi-
mental animals have not been well investigated. More
recently, a metabolic profiling study revealed that AR could
be metabolized to several phase I and phase II metabolites
after oral administration in rat, while glucuronidation was
one of the major metabolic pathways in rat.[12] Our prelimi-
nary study on in-vitro metabolism of AR also found that
AR could be rapidly glucuronidated in HLM, while the for-
mation of phase I metabolites in HLM was very slow. This
finding inspired us to characterize the glucuronidation
pathway(s) of AR in human tissues.
Our results demonstrated that AR could be rapidly
glucuronidated in both HLM and HIM in the presence of
UGPDA, while a mono-glucuronide was formed. The
glucuronide was biosynthesized and then fully characterized
as AR-4′-O-glucuronide. Reaction phenotyping by using
commercially available UGT enzymes demonstrated that at
least six UGT1A isoforms participated in AR-4′-O-
glucuronidation, while three UGT2B isoforms, namely
UGT2B4, UGT2B7 and UGT2B17, could catalyse this
biotransformation. Notably, three most abundantly
expressed UGT isoforms in human liver, namely UGT1A9,
UGT2B7 and UGT2B17, are major contributors to AR-4′-
O-glucuronidation in human liver, which is confirmed by
the chemical inhibition assays. Given the abundant expres-
sion of these three major UGT isoforms in human liver and
the good catalytic efficiency of these enzymes, the high
intrinsic clearance of AR-4′-O-glucuronidation in HLM
(4241 μl/min/mg protein) is not so surprising.
It is also worthy to mention that the corresponding Clint
of AR in RLM was 129 μl/min/mg according to a previous
report,[21] which is much lower than that in HLM as demon-
strated in this study. This could be partially explained by the
fact that rat UGT1A9 is a pseudogene and the propofol
O-glucuronidation activity in rat liver is much lower than
that in human liver.[22,23] These findings agreed well with the
observation of rapid formation of 4′-O-glucuronide after
both oral and IV administration of AR,[21] indicating that
AR may be rapidly eliminated from human body via
glucuronidation pathway.
The formation of AR-4′-O-glucuronidation could also be
observed in HIM, which is consistent with previous report
that AR could be transformed to its 4′-O-glucuronide in rat
intestine microsomes.[21] The kinetic analysis demonstrated
that human intestine played an important role in AR-4′-O-
glucuronidation with Clint value of 776 μl/min/mg protein
in HIM, implying that AR may undergo a first-pass
Enzyme
Source
Vmax
Vmax/Km
(nmol/min/mg)
Km (μM)
(μl/min/mg)
HLM
67.0 1.0
19.1 0.3
0.69 0.01
2.3 0.08
6.5 0.2
15.8 1.0
24.6 1.3
6.0 0.7
61.1 5.4
19.1 1.7
5.2 0.6
5.4 0.5
4240
776
114
37
HIM
UGT1A1
UGT1A7
UGT1A9
UGT2B7
UGT2B17
342
548
2207
2.8 0.07
12.0 0.3
AR, arctigenin; HIM, human intestine microsomes; HLM, human liver
microsomes.
These findings suggested that UGT2B7 and UGT2B17 were
two major contributors in the formation of AR-4′-O-
glucuronide in HIM, and the contribution of UGT1A1 for
AR-4′-O-glucuronidation in HIM was about 10%. In sharp
contrast, magnolol had no effects on the formation of
AR-4′-O-glucuronide in HIM, suggesting that UGT1A9 did
not participate in intestinal glucuronidation of AR.
Kinetic characterization
The kinetic parameters including Km, Vmax and the intrinsic
clearance (Vmax/Km) for AR-4′-O-glucuronidation in HLM,
HIM and the potent UGT isoforms were determined and
shown in Table 2. The range of substrate concentrations was
from 1 to 200 μm for the kinetic analyses. In the concentra-
tion range tested, AR-4′-O-glucuronidation in HLM, HIM
and in human UGTs isoforms, including UGT1A1,
UGT1A7, UGT1A9, UGT2B7 and UGT2B17, exhibited
Michaelis–Menten kinetics, as evidenced by a linear Eadie–
Hofstee plot (Figure 5). The kinetic parameters of AR-4′-O-
glucuronidation in UGT1A3, UGT1A8, UGT1A10 and
UGT2B4 could not be characterized as the formation rates
of AR glucuronide in these isoforms were very slow. The Km
values for the formation of the metabolite in HLM and
HIM were 15.8 μm and 24.6 μm; the Vmax values were 67.0
and 19.1 nmol/min/mg protein; and the clearance values
were determined to be 4240 and 776 μl/min/mg protein,
respectively. In recombinant UGT1A1, UGT1A7, UGT1A9,
UGT2B7 and UGT2B17, the Km values were 6.0, 61.1, 19.1,
5.2 and 5.4 μm, respectively; the Vmax values were 0.69, 2.3,
6.5, 2.8 and 12.0 nmol/min/mg protein, respectively; and
the clearance values were calculated to be 114, 37, 342, 548
and 2207 μl/min/mg protein, respectively.
Discussion
As a major bioactive compound in several traditional
Chinese herbals, AR has drawn much attention in the past
1678
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681

Hong Xin et al.
In-vitro glucuronidation of arctigenin
(a)
(c)
(e)
(g)
(b)
(d)
(f)
20
15
10
5
80
60
40
20
0
80
60
40
20
0
20
15
10
5
>
>
0
0
1
2
3
4
0.0
0.5
V/S
1.0
V/S
0
0
50
100
150
200
0
50
100
150
200
Arctigenin (µM)
Arctigenin (µM)
0.8
0.6
0.4
0.2
0.0
2.0
1.5
1.0
0.5
0.0
0.8
0.6
2.0
1.5
1.0
0.5
0.0
>
>
0.4
0.2
0.0
0.00
0.05
V/S
0.10
0.00
0.02
V/S
0.04
0
50
100
150
200
0
50
100
150
200
Arctigenin (µM)
Arctigenin (µM)
4
3
2
1
0
8
6
4
2
0
8
6
4
2
0
3
2
>
1
>
0
0.0 0.2 0.4 0.6
V/S
0.0 0.1 0.2 0.3
V/S
0
50
100
150
200
0
50
100
150
200
Arctigenin (µM)
Arctigenin (µM)
15
10
5
15
10
>
5
0
0
1
V/S
2
0
0
50
100
150
Arctigenin (µM)
Figure 5 Enzyme kinetics of AR-4′-O-glucuronidation by human liver microsomes (a), human intestine microsomes (b), UGT1A1 (c), UGT1A7 (d),
UGT1A9 (e), UGT2B7 (f) and UGT2B17 (g). Arctigenin (1–200 μM) was incubated with pooled human liver microsomes (0.01 mg of protein/ml),
human intestine microsomes (0.05 mg of protein/ml), UGT1A1 (0.1 mg of protein/ml), UGT1A7 (0.1 mg of protein/ml), UGT1A9 (0.02 mg of
protein/ml), UGT2B7 (0.05 mg of protein/ml) and UGT2B17 (0.025 mg of protein/ml) at 37°C for 15 min. An Eadie–Hofstee plot was shown as an
inset to illustrate monophasic kinetics.
metabolism during the absorption process. Actually, an
extensive first-pass metabolism of AR to arctigenin-4′-O-
glucuronide was also observed in an intestinal perfusion
study in rat, with Cummins’s extraction ratio of
0.085 0.013.[14]
Apart from the rapid glucuronidation of AR in human
liver and intestine, kidney may also participate in AR-4′-O-
glucuronidationin human body. Taking into account the
fact that UGT1A9 demonstrated a high catalytic activity
towards AR-4′-O-glucuronidation and its abundant expres-
sion in kidney,[24] UGT1A9 may be the major contributor to
the metabolic clearance of AR in human kidney. Consider-
ing that UGT1A9 and UGT2B7 are highly expressed in
these metabolic organs of human,[25] as well as the high
catalytic activity of these UGT isoforms in AR-4′-O-
glucuronidation, it is readily conceivable that AR could be
extensively glucuronidated after oral administration in
human body. Therefore, the activity of AR-4′-O-
glucuronide and the excretion of this glucuronide should be
further investigated.
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681
1679

In-vitro glucuronidation of arctigenin
Hong Xin et al.
Furthermore, among the major isoforms participated in
AR-4′-O-glucuronide, UGT2B17 is one of the major con-
tributors in both human liver and intestine, leading to a
rapid metabolic elimination of AR from human body. It has
been reported that the individuals possessing mutant
UGT2B17 (UGT2B17*2) exhibit the absence of UGT2B17
protein.[26] The UGT2B17*2/*2 genotype is strongly associ-
ated with low activity towards testosterone, a selective sub-
strate of UGT2B17.[27] The frequency of the UGT2B17*2/*2
genotype was reported to be 67% and >80% in Korean[28]
population and Japanese population,[29] respectively, indi-
cating that UGT2B17 protein may be absent in many Asian
people. It is easily conceivable that a prolongation of the
residence time and an increased exposure of AR may be fre-
quently observed in Asian people or other individuals pos-
sessing UGT2B17*2/*2. The in-vivo decreased efficiency
of UGT2B17-mediated AR-4′-O-glucuronidation in the
individuals possessing UGT2B17*2/*2 should be further
investigated.
assays demonstrated that UGT1A9, UGT2B7 and UGT2B17
played important roles in hepatic glucuronidation of AR,
while UGT2B7 and UGT2B17 were two major contributors
to the formation of AR-4′-O-glucuronide in HIM. In addi-
tion, the Clint of 4′-O-glucuronide of AR generated in HIM
indicated a first-pass metabolism in the intestinal border
after oral administration, and the following hepatic
glucuronidation should lead to a rapid elimination of
arctigenin from human body.
Conclusions
In conclusion, the glucuronidation pathway of AR in HLM
and HIM was well characterized for the first time. UGT1A9,
UGT2B7 and UGT2B17 were the major isoforms respon-
sible for the 4′-O-glucuronidation of AR in HLM, while
UGT2B7 and UGT2B17 were the major contributors to this
biotransformation in HIM.
Despite of UGT2B17, UGT1A1 and UGT2B7 also play
important roles in the hepatic and intestinal glucur-
onidation of AR. It has been reported that many therapeutic
drugs, natural bioactive compounds and herbal medicines
could potently inhibit UGT1A1 and UGT2B7, such as an
immunosuppressant drug tacrolimus,[30] a natural flavonoid
Declarations
Conflict of interest
The Author(s) declare(s) that they have no conflicts of
interest to disclose.
silybin,[31] and
a
ginsenoside20(S)-protopanaxatriol,[32]
as well as Andrographis paniculata extracts.[33] Co-
administration with these drugs or herbs could be helpful
for prolongation of the residence time and increasing the
exposure of AR in human body, via inhibiting UGT1A1-
and UGT2B7-mediated AR-4′-O-glucuronidation.
Taken together, AR could be extensively metabolized by
human UGTs and generates one mono-glucuronides (4′-O-
glucuronide). Both kinetic characterization and inhibition
Funding
This work was supported by the National S&T Major
Projects of China (2012ZX09501001, 2012ZX09506001),
the National Basic Research Program of China
(2013CB531800), NSF of China (81473181, 81273590,
81302793), and the Leading Talent Research Fund of Harbin
(2011RFXYS079).
competent and immunocompromised
mice infected with influenza A virus.
8. Shoeb M et al. Isolation, structure
elucidation and bioactivity of
schischkiniin, a unique indole alkaloid
from the seeds of Centaurea
schischkinii. Tetrahedron 2005; 61:
9001–9006.
References
1. Zhao
F
et al. In vitro anti-
Biol Pharm Bull 2010; 33: 1199–1205.
5. Schroder HC et al. Differential in vitro
anti-HIV activity of natural lignans. Z
Naturforsch C 1990; 45: 1215–1221.
inflammatory effects of arctigenin, a
lignan from Arctium lappa L., through
inhibition on iNOS pathway.
J
Ethnopharmacol 2009; 122: 457–462.
6. Awale
S
et al. Identification of
9. Kang K et al. The chemopreventive
effects of Saussurea salicifolia through
induction of apoptosis and phase II
detoxification enzyme. Biol Pharm
Bull 2007; 30: 2352–2359.
2. Yang YN et al. Hepatoprotective activ-
ity of twelve novel 7′-hydroxy lignan
glycosides from Arctii Fructus. J Agric
Food Chem 2014; 62: 9095–9102.
3. Kravtsova SS, Khasanov VV. Lignans
and fatty acid composition of Arctium
lappa seeds. Chem Nat Comp 2011; 47:
800–801.
arctigenin as an antitumor agent
having the ability to eliminate the tol-
erance of cancer cells to nutrient star-
vation. Cancer Res 2006; 66: 1751–
1757.
10. Cho MK et al. Arctigenin,
a
7. Wang L et al. Induction of apoptosis
of the human leukemia cells by
arctigenin and its mechanism of
action. Yao Xue Xue Bao 2008; 43:
542–547.
phenylpropanoid
dibenzylbutyrolactone lignan, inhibits
MAP kinases and AP-1 activation
via potent MKK inhibition: the
role in TNF-alpha inhibition. Int
4. Hayashi K et al. Therapeutic effect of
arctiin and arctigenin in immuno-
1680
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681

Hong Xin et al.
In-vitro glucuronidation of arctigenin
whereas UGT2A1 conjugates both
androgens similarly. Drug Metab
Dispos 2009; 37: 417–423.
26. Wilson W 3rd et al. Characterization
of a common deletion polymorphism
of the UGT2B17 gene linked to
UGT2B15. Genomics 2004; 84: 707–
714.
Immunopharmacol 2004; 4: 1419–
1429.
11. Xu ZH et al. The antidiabetic activity
of total lignan from Fructus arctii 19. Xia YL et al. Identification and
against alloxan-induced diabetes in
mice and rats. Phytother Res 2008; 22:
97–101.
characterization of human UDP-
glucuronosyltransferases responsible
for the glucuronidation of fraxetin.
Drug Metab Pharmacokinet 2014; 29:
135–140.
27. Schulze JJ et al. Doping test results
dependent on genotype of uridine
diphospho-glucuronosyl transferase
2B17, the major enzyme for testoster-
one glucuronidation. J Clin Endocrinol
Metab 2008; 93: 2500–2506.
12. Gao Q et al. Hydrolysis is the domi-
nating in vivo metabolism pathway for
arctigenin: identification of novel 20. Chan YS et al. A review of the phar-
metabolites of arctigenin by
macological effects of Arctium lappa
(burdock). Inflammopharmacology
2011; 19: 245–254.
28. Jakobsson J et al. Large differences in
testosterone excretion in Korean and
Swedish men are strongly associated
with a UDP-glucuronosyl transferase
2B17 polymorphism. J Clin Endocrinol
Metab 2006; 91: 687–693.
LC/MS/MS after oral administration
in rats. Planta Med 2013; 79: 471–479.
13. Zhou XY et al. Determination of 21. Gao Q et al. Elucidation of arctigenin
arctiin and arctigenin contents in
Arctium tomentosum Mill. by HPLC
method. J Chem 2011; 8: 372–376.
pharmacokinetics after intravenous
and oral administrations in rats: inte-
gration of in vitro and in vivo findings
via semi-mechanistic pharmacokinetic
modeling. AAPS J 2014; 16: 1321–
1333.
29. Terakura S et al. A UGT2B17-positive
donor is a risk factor for higher
transplant-related mortality and lower
survival after bone marrow transplan-
tation. Br J Haematol 2005; 129: 221–
228.
14. Gao Q et al. Extensive intestinal first-
pass metabolism of arctigenin: evi-
denced by simultaneous monitoring
of both parent drug and its major 22. Shiratani H et al. Species differences in
metabolites. J Pharm Biomed Anal
2014; 91: 60–67.
UDP-glucuronosyltransferase activ-
ities in mice and rats. Drug Metab
Dispos 2008; 36: 1745–1752.
30. Liu XY et al. Tacrolimus strongly
inhibits multiple human UDP-
15. Fujita K et al. The small-molecule
tyrosine kinase inhibitor nilotinib is a 23. Liang SC et al. Determination of pro-
pofol UDP-glucuronosyltransferase
(UGT) activities in hepatic
glucuronosyltransferase
(UGT)
potent noncompetitive inhibitor of
the SN-38 glucuronidation by human
isoforms. Pharmazie 2012; 67: 804–
808.
UGT1A1.
Cancer
Chemother
microsomes from different species by
UFLC-ESI-MS. J Pharm Biomed Anal
2011; 54: 236–241.
31. Jancova P et al. Evidence for differ-
Pharmacol 2011; 67: 237–241.
ences
in
regioselective
and
16. Zhu LL et al. Potent and selective inhi-
stereoselective glucuronidation of
silybin diastereomers from milk thistle
(Silybum marianum) by human UDP-
glucuronosyltransferases. Xenobiotica
2011; 41: 743–751.
bition of magnolol on catalytic activ- 24. Knights KM et al. Renal drug metabo-
ities of UGT1A7 and 1A9. Xenobiotica
2012; 42: 1001–1008.
lism in humans: the potential for
drug-endobiotic interactions involv-
ing cytochrome P450 (CYP) and
UDP-glucuronosyltransferase (UGT).
Br J Clin Pharmacol 2013; 76: 587–
602.
17. Uchaipichat V et al. Selectivity of sub-
strate (trifluoperazine) and inhibitor
32. He YJ et al. The inhibitory effect of
20(S)-protopanaxatriol (ppt) towards
UGT1A1 and UGT2B7. Phytother Res
2013; 27: 628–632.
(amitriptyline,
androsterone,
canrenoic acid, hecogenin, phenylb-
utazone, quinidine, quinine, and 25. Sato Y et al. Optimized methods for
sulfinpyrazone) ‘probes’ for human
UDP-glucuronosyltransferases. Drug
Metab Dispos 2006; 34: 449–456.
targeted peptide-based quantification
of human uridine 5′-diphosphate-
33. Ismail S et al. Effects of Andrographis
paniculata and Orthosiphon stamineus
extracts on the glucuronidation of
4-methylumbelliferone in human
UGT isoforms. Molecules 2010; 15:
3578–3592.
glucuronosyltransferases
in
bio-
18. Sten
T
et al.
UDP-
logical specimens using liquid
chromatography-tandem mass spec-
trometry. Drug Metab Dispos 2014; 42:
885–889.
glucuronosyltransferases (UGTs) 2B7
and UGT2B17 display converse
specificity in testosterone and
epitestosterone
glucuronidation,
© 2015 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 67, pp. 1673–1681
1681