Pharmacology & Pharmacy, 2011, 2, 361-369
doi:10.4236/pp.2011.24047 Published Online October 2011 (http://www.SciRP.org/journal/pp)
361
Bucolome N-Glucuronide Formation: Species
Differences and Identification of Human
UDP-Glucuronosyltransferase Isoforms
Humihisa Kanoh, Makiko Tada, Yoshihiro Uesawa, Kiminori Mohri*
Clinical Pharmaceutics Laboratory, Department of Pharmacy and Health Sciences, Faculty of Pharmacy and Pharmaceutical Sciences,
Meiji Pharmaceutical University, Tokyo, Japan.
Received August 4th, 2011; revised September 13th, 2011; accepted September 20th, 2011.
ABSTRACT
The barbituric acid derivative bucolome (BCP) is a nonsteroidal anti-inflammatory drug. The present study investi-
gated whether BCP N-glucuronide (BCP-NG, the primary metabolite of BCP) was produced in mammalian species
other than rats, and attempted to identify the UDP-glucuronosyltransferase (UGT) isoform (s) responsible for forma-
tion of BCP-NG in humans. BCP-NG was detected in all species tested. The results were as follows (pmol equivalent/
min/mg protein): rat, 479 ± 83; Mongolian gerbil, 378 ± 9; rabbit, 275 ± 26; guinea pig, 257 ± 10; human, 242 ± 18;
hamster, 177 ± 22; and mouse, 167 ± 15. Since human liver microsomes formed BCP-NG, we investigated the metabo-
lites of BCP excreted in the urine of a patient after oral administration of BCP (600 mg). BCP and BCP-NG were ex-
creted in the urine at amounts of 2.9 mg (about 0.5% of the do se) and 14.4 mg (about 2.5% of the dose) over 12 hours.
In order to identify the UGT isofo rms involved in formation of BCP-NG in humans, we investigated BCP-NG formation
by the microsomes of insect cells expressing each of twelve UGT isoforms (hUGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9,
1A10, 2B4, 2B7, 2B15, and 2B17). As a result, BCP-NG formation (pmol equivalents/min/mg protein) was observed
with microsomes expressing hUGT1A1 (142), 1A3 (196), 1A4 (8), 1A7 (8), 1A8 (66), 1A9 (38), 1A10 (9), 2B4 (7) and
2B7 (16). In particular, the activity of hUGT1A1 and 1A3 was high. These results suggest that the UGT isoforms re-
sponsible for formation of BCP-NG exist in various mammalian species, including humans, and that the UGT 1A family
is primarily responsible for BCP N-glucuronide formation in humans.
Keywords: Bucolome, Bucolome N-Glucuronide, UGT Isoforms, Barbiturate, Barb iturate N-Glucuronide
1. Introduction
Glucuronic acid conjugation is an enzymatic reaction
catalyzed by UDP-glucuronosyltransferase (UGT; EC
2.4.1.17), and it is one of the most important reactions in
phase II drug metabolism. UGT is widely present in
many species from bacteria to humans, and glucuronic
acid conjugation is estimated to account for about 35% of
phase II drug metabolism [1]. Bucolome (BCP, Figure 1)
is a barbituric acid derivative nonsteroidal anti-inflam-
matory drug that exhibits analgesic and anti-inflamma-
tory actions without having sedative or hypnotic effects,
unlike many barbiturates. It has been used for the treat-
ment of rheumato id arthritis [2]. It has been reported that
BCP promotes the secretion of bile in dogs and rats [3,4]
and induces uric acid excretion in humans [5], so BCP is
also used as an antipodagric. In recent years, it has been
shown that BCP inhibits CYP2C9 [6]. In patients on
warfarin therapy in Japanese hospitals, BCP is often ad-
ministered concomitantly to maintain the plasma war-
farin concentration through inhibition of the metabolism
of S-warfarin by CYP2C9 [7]. Although BCP is used for
various purposes, as noted above, its metabolic pathway
and the enzymes involved in humans have not been
evaluated in detail, with only old references being avail-
able [5,8-11].
In order to clarify whether the formation of BCP-NG
occurs in mammals other than rats, we investigated BCP-
NG formation in vitro using liver microsomes from rats,
guinea pigs, mice, hamsters, Mongolian gerbils, rabbits,
and humans. Although the N-glucosides of amobarbital
[12] and phenobarbital [13] have been reported to be the
primary metabolites of these barbiturate derivatives in

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human 362 UDP-Glucuronosyltransferase Isoforms
Figure 1. Chemical structures of bucolome and bucolome
N-glucuronide.
humans, their N-glucuronides have not been identified.
Based on the results of BCP-NG formation by human
liver microsomes in the present study, we also assessed
BCP metabolites in the urine of a patient with hyperu-
ricemia who was administered BCP at 600 mg/day. We
found BCP-NG excretion in the urine at 2.5% of th e dose
over 12 hours. In order to identify the UGT isoforms
involved in N-glucuronidation of BCP, we also studied
BCP-NG formation using insect cell microsomes ex-
pressing 12 different human UGT isoforms (hUGT1A1,
1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B15,
and 2B17).
This report describes the UGT isoforms catalyzing
BCP-NG formation in humans.
2. Materials and Methods
2.1. Chemicals, Experimental Animals, and
Enzymes
All animal procedures were approved by the Meiji
Pharmaceutical University Committee for Ethics of Ex-
perimentation and Animal Care (approved No. 2304).
BCP was synthesized from cyclohexylurea (Tokyo Che-
mical Industry Co., Ltd., Tokyo) and n-butylmalonate
(Tokyo Chemical Industr y) by the method of Send a et al.
[2]. Paramidine (300 mg) was purchased from Aska
Pharmaceutical Co., Ltd., (Tokyo). BCP-NG was ob-
tained by the previously reported method [14]. Lubrol
WX was purchased from Sigma-Aldrich Co. (St. Louis,
MO). Phenylbutazone (PBZ, internal standard, I.S.) and
saccharo-1,4-lactone were purchased from Nacalai Tes-
que, Inc. (Kyoto), while magnesium chloride (MgCl2),
2-amino-2-hydroxymethyl-1,3-propanediol (Tris), am-
monium sulfate, UDP-glucuronic acid (UDP-Ga), metha-
nol (MeOH), and ethanol (EtOH) were all from Wako
Pure Chemical Industries Ltd. (Osaka). Alamethicin was
purchased from Enzo Life Sciences Inc. (New York).
Solutions of BCP and PBZ adjusted to 1 mg/ml with
DMSO were used as standard solutions, which were di-
luted with the solvent immediately before use. Reagents
were stored at 30˚C and were used within 3 months.
Male Wistar-ST rats weighing 300 - 500 g (clean), male
Hartley guinea pigs weighing 550 - 700 g (clean), male
ddY mice weighing 60 - 80 g (SPF), male Syrian ham-
sters weighing 185 - 200 g (SPF), male Mongolian ger-
bils weighing 60 - 63 g (SPF), and male Japanese white
rabbits weighing 2.0 - 2.2 kg (clean) were all purchased
from Japan SLC Inc. (Shizuoka). The animals were
housed under conventional conditions until use. Recom-
binant hUGT isoform microsomes for various human
UGT isoforms (UGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8,
1A9, 1A10, 2B4, 2B7, 2B15, and 2B17) were expressed
in insect cells using a baculovirus and 50-donor pooled
human liver microsomes were obtained from BD Gentest
(San Jose, CA). All chemicals used were analytical or
special grade products, and the water was double-dis-
tilled. Liver microsomes of rats, mice, guinea pigs, ham-
sters, Mongolian gerbils, and rabb its were prepared from
100 g of liver tissue harvested from four animals by the
previously reported method [15]. Protein concentrations
were determined by the method of Lowry et al. [16].
2.2. HPLC and Chromatography Conditions
BCP-NG formation in vitro and the levels of BCP and
BCP-NG in the urine of a patient were measured by the
method of Mohri et al. with slight modifications (2001).
The HPLC system consisted of a JASCO Intelligent
HPLC pump (Model PU-1580, JASCO Co. Ltd., Tokyo)
equipped with a JASCO Intelligent UV detector (Model
UV-1570), a JASCO Intelligent Sampler (Model 851-AS)
and a JASCO Chromatography Data Station (Chrom
NAV). The detector was set at 268 nm, with a sensitivity
of 0.005 absorbance units full scale (a.u.f.s.). BCP, PBZ
(I.S.), and BCP-NG were separated at room temperature
(23˚C) on a reversed-phase Capcell Pak ODS column
(UG120) [6.0 mm in internal diameter (I.D.)  15 cm in
length and particle size of 5 m] (Shiseido Co. Ltd., To-
kyo) equipped with a guard column packed with the
same stationary phase [4.6 mm I.D.  1 cm]. The mobile
phase (0.05 M phosphate buffer (pH 5.7):MeOH:THF;
50:40:4 v/v) was pumped through the column at a rate of
1.5 ml/min, after being passed through a 0.45 mm filter
(Millipore, Bedfo rd, MA) prior to use and degassed with
an ERC-3322 degasser (Erma Co. Ltd., Saitama) under
reduced pressure. All HPLC analyses were performed in
triplicate.
2.3. LC/MS Analysis of BCP-NG
A Shimadzu HPLC mass spectrometer (LCMS-2010EV,
Shimadzu Corp., Kyoto) was used, with an LC/MS pump
(LC-20AB) attached to an autosampler (SIL-20A), as
well as a degasser (DGU-20A5), column oven (CTO-
20SA), and system controller (CBM-20A). A Shim-pack

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human 363
UDP-Glucuronosyltransferase Isoforms
VP-ODS column (2.0 mm i.d. × 50 mm) attached to a
GVP-ODS guard column was used at an oven tempera-
ture of 40˚C, and analysis was performed with the ESI
method in negative ion mode after setting the mobile
phase flow rate at 0.2 ml/min, N2 gas flow rate at 1.5
l/min, probe voltage at 1.5 kV, probe temperature at
250˚C, CDL voltage at 20.0 V, and block temperature at
200˚C. Reversed-phase HPLC was done by flow injec-
tion and the elution conditions were as follows: the ratio
of mobile phase A (0.1 % formic acid) to mobile phase B
(acetonitrile) was altered from 7:3 at 0 minutes to 5:5 at 5
minutes according to a linear gradient and further to 2:8
at 8 minutes according to a linear gradient by increasing
acetonitrile. It was then maintained at this level for 5
minutes, reduced stepwise to 7:3 (v/v), and maintained at
that level for 6 minutes. Ion detection was performed by
the electrospray ionization procedure at an m/z value of
265 for BCP, 307 for PBZ (I.S.), and 441.20 for BCP-
NG.
2.4. BCP-NG Formation by Liver Microsomes of
Humans and Experimental Animals
The concentration of each reagent in the reaction mixture
yielding maximum BCP-NG formation was determined
in advance using rat liver microsomes (final concentra-
tion range tested: lubrol 0 - 0.4 mg/ml, rat liver micro-
somes 0 - 5 mg/ml, MgCl2 0 - 10 mM, saccharolactone
0 - 8 mM, BCP 0 - 5 mM, 0.5 M Tris-HCl pH6 - pH8,
UDP-Ga 0 - 10 mM, and incubation time 0 - 60 min).
Enzymatic reactions were performed in a total volume of
250 l containing UDP-Ga (final concentration, 8 mM),
liver microsomes solubilized with lubrol (final concen-
tration, 0.1 mg/ml) of each species (rat, mouse, guinea
pig, hamster, Mongolian gerbil, rabbit, and human; 2
mg/ml microsomal protein), MgCl2 (final concentration,
8 mM), saccharo-1,4-lactone (final concentration, 4 mM),
BCP (final concentration, 2 mM), and 0.5 M Tris-HCl
(pH 7.4) (final concentration, 0.2 M). After the reaction
mixture had been preincubated at 37˚C for 5 minutes, the
reaction was started by addition of UDP-Ga or BCP at
37˚C for 20 minutes, and then the reaction was stopped
by adding 0.5 g ammonium sulfate. After adding 200
ml of an EtOH solution containing 24 M PBZ, the
resulting mixture was mixed vigorously for 30 seconds.
It was then centrifuged at 18,000 × g for 10 minutes at
4˚C, and 10 l of 0.1% NaHCO3 was added to 150 l of
the supernatant. Next, 10 l of the supernatant was di-
rectly injected into the HPLC apparatus and the concen-
tration of BCP-NG formed by the enzymatic reaction
was determined using the BCP calibration curve, as de-
scribed previously [15]. All measurements were per-
formed in triplicate.
2.5. Stabilization of BCP-NG
Since BCP-NG was stable under weakly basic conditions,
10 l of 0.1% NaHCO3 was added to 150 l of super-
natant.
2.6. Ethanol Extraction Method
An excess of solid ammonium sulfate (0.5 g) was added
to the reaction solution, wh ich was mixed vigorously an d
then centrifuged at 18,000  g for 10 minutes at 4˚C to
separate water and EtOH. This procedure using the salt-
ing out method is able to transfer highly water-soluble
compounds such as glucuronides into the EtOH layer at
high concentrations.
2.7. Km, Vmax, and Intrinsic Metabolic
Clearance (CLmet) of BCP and UDP-Ga
in Experimental Animals
In order to compare BCP-NG formation among the ex-
perimental animals tested (rat, guinea pig, mouse, ham-
ster, Mongolian gerbil, and rabbit), the Km, Vmax, and
intrinsic metabolic clearance (CLmet = Vmax/Km) were
obtained. To obtain Km and Vmax values fo r BCP, its con-
centration was varied over the range from 0.01 mM to 3
mM. The amount of UDP-Ga used in this reaction was
22 mM for rats, 8 mM for mice, 8 mM for guinea pigs, 6
mM for hamsters, 10 mM for Mongolian gerbils, and 6
mM for rabbits. To obtain Km and Vmax values for
UDP-Ga, its concentration was varied over the range
from 0.01 mM to 24 mM. BCP was used at 2 mM in this
enzymatic reaction. Human microsomes were pooled
from 50 persons. Since both racial differences and inter-
individual variations are very large, we did not obtain Km
and Vmax values for BCP N-glucuronidation using human
liver microsomes, because Homo sapiens is not a homo-
geneous species. The initial velocity of BCP-NG forma-
tion in relation to the BCP dose and UDP-Ga dose was
analyzed using Lineweaver-Burk plots. All measure-
ments were performed in triplicate.
2.8. Analysis of Urinary BCP and BCP-NG in a
Patient with Hyperuricemia
We received written informed consent from a patient to
measure the levels of BCP and BCP-NG in the urine.
Total urine was collected from a 23-year-old patient with
hyperuricemia at 0, 1, 2, 3, 4, 7, 8, 9, and 12 hours after
administration of 600 mg of BCP (2 tablets of Para-
midin™ 300 mg), and the urine volume and the urinary
levels of BCP and BCP-NG concentrations were deter-
mined at each sampling point. The pH of the urine sam-
ples was adjusted to about 8 by adding solid NaHCO3.
To 250 l of urine, 200 l of an EtOH solution contain-

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human 364 UDP-Glucuronosyltransferase Isoforms
ing 24 mM PBZ (IS) and 0.5 g ammonium sulfate were
added, followed by vigorous mixing for 30 seconds. Af-
ter centrifugation at 18,000  g for 10 minutes at 4˚C, 10
l of the supernatant was directly injected into the HPLC
apparatus. Urinary BCP and BCP-NG concentrations
were determined from the calibration curve for BCP, as
previously reported [15]. All measurements were per-
formed in triplicate.
2.9. BCP-NG Formation by Human
Recombinant UGT Isoform Microsomes
Twelve different human recombinant UGT isoform mi-
crosomes (1 mg/ml microsomal protein) were solubilized
by adding alamethicin (0.025 mg/ml DMSO solution).
Reactions were performed under the same conditions as
in the experiment using human liver microsomes and
were stopped by adding 500 l of chloroform. After the
reaction mixture had been mixed vigorously for 30 sec-
onds, it was centrifuged at 18,000  g for 10 minutes at
4˚C. To 150 l of the supernatant, the same volume of 5
M acetonitrile solution containing PBZ was added, fol-
lowed by vigorous mixing for 30 seconds and centri-
fuged at 18,000  g for 10 minutes at 4˚C. To 50 l of
the resulting supernatant, 5 volumes of acetonitrile was
added. Then the mixture was mixed vigorously again for
30 seconds and centrifuged at 18,000  g for 10 minutes
at 4˚C, after which 10 l of the supernatant was injected
into the LC/MS apparatus. Measurements were per-
formed in triplicate. Since BCP-NG formation by recom-
binant human UGT isoform microsomes was very low,
the LC/MS system was used to measure BCP-NG levels.
2.10. Data Analysis
Km and Vmax values for BCP and UDP-Ga in the liver
microsomes of experimental animals were determined
from the experimental data using Michaelis-Menten hy-
perbolic kinetics and Lineweaver-Burk plots. Kinetic
data are expressed as the mean ± S.D. CLmet values were
calculated with the Km and Vma x values obtained from
Lineweaver-Burk plots. Multiple post-hoc comparisons
among three or more groups were done with Scheffe’s
test [17]. JMP Statistical Discovery Software (SAS In-
stitute Japan Co., Tokyo) was used for all analyses and
the level of significance was p < 0.05 (כ).
3. Results
3.1. HPLC and LC/MS
3.1.1. HPLC
The retention times of BCP, the I.S., and BCP-NG peaks
in the urine of a patient administered BCP were about 7.2,
9.9, and 12.9 minutes, respectively. Calibration curves
for BCP and BCP-NG were linear over the range from
0.01 to 100 µM with correlation coef ficients > 0.99. The
detection limits for BCP and BCP-NG were 0.25 ng and
0.06 ng, respectively. Intra-day and inter-day variation of
the calibration curves for BCP and BCP-NG was < 5% in
both cases.
3.1.2. LC/MS
The retention times of the BCP-NG (m/z = 441.20) and
PBZ (I.S., m/z = 307) peaks were about 9.8 and 13.9
minutes, respectively. Calibration curves for BCP-NG
were linear over th e range from 1.67 ng/ml to 2.07 g/ml
with correlation coefficients > 0.99. The detection limit
for BCP-NG was 16.7 pg. Intra-day and inter-day varia-
tion of the calibration curve for BCP-NG was <5%.
3.2. BCP-NG Formation in the Liver
Microsomes of Human and Experimental
Animals
BCP-NG formation was detected in the liver microsomes
of rats, Mongolian gerbils, rabbits, guinea pigs, humans,
hamsters, and mice. The level of BCP-NG formation
(pmol equivalents/min/mg protein, mean ± S.D.) was as
follows: rat; 479 ± 83, Mon golian gerbil; 378 ± 9, rabbit;
275 ± 26, guinea pig; 257 ± 10, human; 242 ± 18, ham-
ster; 177 ± 22, and mouse; 167 ± 15 (Figure 2). Liver
microsomal BCP-NG formation was highest in the rat
(relative comparison = 1), followed by the Mongolian
gerbil (0.83), guinea pig (0.54), rabbit (0.54), human
(0.50), hamster (0.37), and mouse (0.35).
0
100
200
300
400
500
600
Rat
Mongolian gerbil
Rabbit
Guinea pig
Human
Hamster
Mouse
BCP-NG (pm ol/mg/min
)
Figure 2. BCP N-glucuronidation activity in rat (a), mouse
(b), guinea pig (c), Mongolian gerbil (d), Rabbit (e), hamster
(f), and human (g) liver microsomes. Each column repre-
sents the mean ± S.D. (vertical bars) of triplicate determi-
nations. Values were compared among rats, guinea pigs,
mice, hamsters, Mongolian gerbils, and humans.

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human
UDP-Glucuronosyltransferase Isoforms
365
3.3. Kinetic Analysis of BCP-NG Formation by
Liver Microsomes in Experimental Animals
The Km, Vmax, and CLmet v alues for BCP and UDP-Ga are
shown in Tables 1(a) and 1(b). The kinetic profile of
BCP N-glucuronidation for BCP and UDPGa in the liver
microsomes of the 6 mammalian species tested showed a
single-enzyme Michaelis-Menten pattern, and no sig-
moidal kinetics were noted. The rate of BCP-NG forma-
tion in relation to the BCP dose and UDP-Ga dose was
analyzed using Lineweaver-Burk plots. Statistically sig-
nificant species differences were observed for the Km,
Vmax, and CLmet values of BCP and UDP-Ga.
3.4. Analysis of BCP and BCP-NG in the Urine
of a Patient with Hyperuricemia
The HPLC chromatogram of urine obtained 2 hours after
administration of BCP is shown in Figure 4. Cumulative
BCP excretion in the urine up to 12 hours after admini-
stration was about 3 mg, which was about 0.5% of the
dose administered. Cumulative BCP-NG excretion was
14.4 mg, which was about 2.5% of the BCP dose.
3.5. BCP-NG Formation by Recombinant
Human UGT Isoforms
BCP-NG activity was noted in 9 (hUGT 1A1, 1A3, 1A4,
1A7, 1A8, 1A9, 1A10, 2B4, and 2B7) of the 12 hUGT
isoforms (UGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9,
1A10, 2B4, 2B7, 2B15, and 2B17), and was particularly
high for hUGT1A1 and 1A3 (Figure 3). The initial ve-
locities of the BCP-NG activity of hUGT1A1, 1A3, 1A4,
1A7, 1A8, 1A9, 1A10, 2B4, and 2B7 were 142, 196, 8, 8,
66, 38, 9, 7, and 16 pmol equivalents/min/mg protein,
respectively, while those of the other recombinant UGT
isoforms were 0.2 pmol equivalents/min/mg protein.
Statistically significant differences were observed for
apparent BCP-NG formation by the hUGT isoforms.
Table 1. Kinetic parameters for BCP and UDP-Ga during BCP N-glucuronidation by liver microsomes pre-
pared from each animal species.
(a)
Kinetic parameters for BCP
Species Km (mM) Vmax (nmol /min/m g) CLmet (L/min/mg)
Rat (a) 1.02 ± 0.04 1.46 ± 0.07b,c,d,e,f 1.43 ± 0.09
Mouse (b) 0.16 ± 0.07 0.17 ± 0.02 1.20 ± 0.49
Guinea pig (c) 0.35 ± 0.06 0.33 ± 0.02 0.96 ± 0.13
Mongolian gerbil (d) 1.25 ± 0.14b 0.91 ± 0.03b,c,e,f 0.74 ± 0.06
Rabbit (e) 0.49 ± 0.14 0.34 ± 0.01 0.72 ± 0.20
Hamster ( f) 1.93 ± 0.78b,c,e 0.56 ± 0.16b,c 0.30 ± 0.06a,b
Correlation coefficient with BCP-NG formation 0.144 0.893 0.461
(b)
Kinetic parameters for UDP-Ga
Species Km (mM) Vmax (nmol/min/m g ) CLmet (L/min/mg)
Rat (a) 8.350 ± 0.490b,c,d,e,f 1.220 ± 0.130b,c,d,e,f 0.146 ± 0.017
Mouse (b) 0.505 ± 0.070 0.177 ± 0.018 0.351 ± 0.014a,c,d,e,f
Guinea pig (c) 1.210 ± 0.230 0.315 ± 0.036 0.262 ± 0.018a,f
Mongolian gerbil (d) 2.200 ± 0.270b,c 0.558 ± 0.035b,c,e,f 0.255 ± 0.016a,f
Rabbit (e) 1.370 ± 0.110 0.341 ± 0.027 0.250 ± 0.032a,f
Hamster ( f) 1.560 ± 0.140b 0.214 ± 0.021 0.138 ± 0.017
Correlation coefficient with BCP-NG formation 0.827 0.925 –0.384
Values are the mean ± S.D. of triplicate experiments. CLmet was calculat ed as Vmax/Km. Small su perscript l etters (a, b, c, d, e, f) indicate th e
results of multiple comparison by Scheffé’s test in each species (p < 0.05).

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human 366 UDP-Glucuronosyltransferase Isoforms
0
50
100
150
200
250
1A1 1A3 1A4 1A6 1A7 1A8 1A9
1A10
2B4 2B7
2B15 2B17
BCP-NG （pmol/min/mg
）
Figure 3. BCP N-glucuronidation activity of recombinant
human UGT isoforms. Each column represents the mean ±
S.D. (vertical bars) of triplicate determinations. The lower
limit of quantitation for the assay under these conditions
was 1.67 ng of BCP-NG.
0
5000
10000
15000
0510 15 20
Time (min)
Signal intensity (v)
IS
(a)
0
5000
10000
15000
0510 15 20
Time (min)
Signal intensity (v)
BCP
ISBCP-NG
(b)
Figure 4. HPLC chromatogram of a urine sample from a
patient with hyperuricemia administered BCP. (a) Before
dosing; (b) 2 h after dosing.
4. Discussion
Although many N-glucuronides have been reported, most
of them are compounds that undergo N-glucuronidation,
including primary aromatic amines (UGT1A1, 1A4,
1A9), hydroxylamines (UGT1A1, 1A10), amides (2B7),
tertiary aliphatic amines (1A4, 2B10), and aromatic
N-heterocycles (1A4, 2B7, 2B10). A number of aromatic
N-heterocycles with five- and six-member rings, such as
imidazoles, pyrazoles, triazines, tetrazoles, and pyridines
are subject to N-glucuronidation. However, N-glucuro-
nidation of the pyrimidine skelton of barbiturates has not
been mentioned in recent review articles [18-20].
There have been a few reports of N-glucosides and
N-glucuronides in which glucose or glucuronic acid is
directly attached to a nitrogen (N) atom in the pyrimidine
skelton of barbiturates. It has b een reported that amobar-
bital and phenobarbital N-glucosid es, in which glucose is
directly bound to an N atom of the pyrimidine skeleton,
are the primary metabolites of barbiturate derivatives in
humans, but N-glucuronides of barbiturate derivatives
have not been reported to date in humans [12,13,21,22].
Neighbors et al. [23] reported that an N-glucoside and an
N-glucuronide were found in the urine of mice adminis-
tered phenobarbital. BCP-NG, the first N-glucuronide of
a barbiturate derivative, has only been reported in rats,
but there had been no investigation of whether or not
BCP-NG is produced in other species. Therefore, we
established an in vitro method to assay BCP-NG forma-
tion using rat liver microsomes, and studied BCP-NG
formation in the liver microsomes of Mongolian gerbils,
guinea pigs, rabbits, hamsters, mice, and humans.
BCP-NG formation was observed in all species, but the
activity varied greatly among the different species (Fig-
ure 2). These results suggest that UGT isoforms forming
BCP-NG exist in a wide range of mammalian species.
We calculated Km, Vmax, and CLmet valu es for BCP and
UDP-Ga under the conditions showing maximum BCP-
NG formation by the liver microsomes of rats, Mongo-
lian gerbils, guinea pigs, rabbits, hamsters, and mice
(Tables 1(a) and 1(b)). A significant relation was ob-
served between BCP-NG formation in these 6 animals
and the Vmax and Km values for UDP-Ga (R = 0.925 and
R = 0.827, respectively) (Table 1(b). However, although
a significant relation was observed between BCP-NG
formation in the 6 animals and Vmax values for BCP (R =
0.894), no relation was observed between BCP-NG for-
mation and Km values for BCP (R = 0.144). Glucuronic
acid conjugation is a two-substrate reaction (BCP and
UDP-Ga). It is thus considered that BCP-NG formation
occurs as follows: first, UDP-Ga binds to UGT in the
endoplasmic reticulum, and then BCP binds to UGT-
UDP-Ga complex, resulting in the formation of BCP-NG.
Thus, if UGT binds more readily with UDP- Ga, an ani-
mal will produce BCP-NG easily, resulting in a high
Vmax value for UDP-Ga.
Luukkanen et al. [24] reported that the glucuronidation
of entacapone by UGT1A9 was inhibited by 1-naphthol
in an entacapone-competitive fashion and was noncom-
petitive with respect to UDP-Ga. Inhibition by UDP, on
the other hand, was noncompetitive with respect to enta-

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human 367
UDP-Glucuronosyltransferase Isoforms
capone and competitive with respect to UDPGA. They
stated that the reaction involved a compulsory ordered bi
bi mechanism based on analysis of inhibition profiles, in
which UDP-Ga was the first binding substrate and enta-
capone was the second binding substrate. The mecha-
nism of BCP-NG formation, in which the first binding
substrate is UDP-Ga and the second binding substrate is
BCP, agrees well with the results of Luukkanen et al.
[24]. The differences of kinetic parameters for BCP and
UDP-Ga among animal species are believed to be due to
complex reactions caused by species differences of UGT
isoforms, so when drug metabolism data obtained from
laboratory animals are extrapolated to human, differences
in the metabolism of animal species must be taken into
consideration.
Since BCP-NG was formed by human liver micro-
somes, we investigated metabolites of BCP in the urine
of a patient with hyperuricemia who was administered
600 mg of BCP (2 tablets of Paramidin 300 mg). Our
results demonstrated that BCP was metabolized to
BCP-NG and excreted in the urine (about 2.5% of the
BCP dose over 12 hours) (Figure 4). Thus, it was clari-
fied that N-glucuronidation is also the primar y metabolic
pathway of BCP in humans. It is known that N-glucoside
is the primary human metabolite of barbiturates, so we
investigated BCP N-glucoside formation in human liver
microsomes by using UDP-glucose (8 mM) instead of
UDP-Ga. After 2 hours of incubation under the same
conditions as for BCP-NG formation in vitro, HPLC was
performed, but an unknown new peak could not be de-
tected on the chromatogram (data not shown). Further
investigation will b e needed to determine the factors that
select N-glucoside and/or N-glucuronide as the metabo-
lite of barbituric acid derivatives.
In order to identify UGT isoforms that catalyze the
N-glucuronidation of BCP in humans, we studied BCP-
NG formation using microsomes of recombinant insect
cells expressing each human UGT isoform (hUGT1A1,
1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B15,
and 2B17) (Figure 3). As a result, BCP-NG formation
was noted in hUGT1A1, 1A3, 1A4, 1A7, 1A8, 1A9,
1A10, 2B4, and 2B7, with the specific activity (pmol
equivalents/min/mg protein) being particularly high for
1A1 (142.0) and 1A3 (196.2), followed by 1A8 (66.2),
1A9 (38.3), and 2B7 (15.9). The activity of hUGT1A4,
1A7, 1A10, and 2B4 was 10 pmol equivalents/min/mg
protein or lo wer.
Ohno et al. [25] established a method for the quantifi-
cation of mRNAs for hUGT isoforms using real-time
RT-PCR, and identified and quantified the levels of
hUGT mRNA isoforms in various human organs. Ac-
cording to their report, the mRNAs for hUGT1A1, 1A3,
1A5, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B10, 2B11,
2B15, and 2B17 were found in human liver tissue. The
expression of hUGT1A1 mRNA was about 30% of that
for hUGT2B7 mRNA, while that of hUGT2B4 mRNA
was about 9 times higher than that of 2B7 mRNA.
BCP-NG formation by 2B4 and 2B7 was lower than by
the UGT1A family, but when their relative abundance in
the liver (approximately 27 times that of 1A1 for 2B4
and approximately 3 times that of 1A1 for 2B7) is taken
into consideration, 2B4 and 2B7 also seem likely to con-
tribute to the formation of BCP-NG in humans.
In recent years, it has been reported that hepatic
UGT1A4 and 2B10 catalyze the N-glucuronidation of
aromatic N-heterocycles in humans [18,19]. However,
BCP-NG formation by hUGT1A4 was <10 pmol equiva-
lents/min/mg protein (Figure 3). The level of hUGT1A4
mRNA in tissues other than the liver is either very low
[18,19] or below the detection limit [25]. Therefore, it
seems that the contribution of UGT1A4 to BCP-NG
formation is negligible. hUGT2B10 has only been de-
tected in the liver and small intestine [18,19], and the
level of hUGT2B10 mRNA in the small intestine is on ly
0.05% of that in the liver. The ortholog of human 2B10
has not been detected in animals [20], while the
UGT1A1 family has been detected in various species
[20,26-27] and a number of different tissues [28-30].
We found that BCP-NG was formed in the liver mi-
crosomes of rats, Mongolian gerbils, rabbits, guinea pigs,
hamsters and mice, as well as in hu mans (Figure 2). Pre-
viously, we reported that BCP-NG formation was ob-
served with the microsomal fractions of the liver, small
and large intestines, and kidney in rats [30]. It is sup-
posed that the hUGT isoforms catalyzing BCP-NG for-
mation in rats differ from UGT2B10 that is localized to
the liver and small intestine in humans. However, mi-
crosomes expressing hUGT2B10 are not commercially
available, so we could not examine BCP-NG formation
directly. Further investigations will be needed to clarify
the role of BCP-NG f ormation by hUGT2B10.
We previously reported that the liver microsomes of
UGT1A family-deficient Gunn rats [29] show dramati-
cally low BCP-NG formation (only 8.5% of that in nor-
mal rats), while BCP-NG formation by phenobarbital-
and clofibric acid-pretreated microsomes was 1.5- and
1.6-fold higher than by untreated microsomes, respec-
tively [31].
5. Conclusions
The results obtained in this study suggest that the
UGT1A family plays the primary role in the formation of
BCP-NG in mammals, including humans, and that
UGT2B isoforms may have a complementary role. This

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human 368 UDP-Glucuronosyltransferase Isoforms
is the first report about detection of the N-glucuronid e of
a barbiturate derivative in humans.
REFERENCES
[1] C. Guillemette, “Pharmacogenomics of Human UDP-
Glucuronosyltransferase Enzymes,” The Pharmacogeno-
mics Journal, Vol. 3, No. 3, 2003, pp. 136-158.
doi:10.1038/sj.tpj.6500171
[2] S. Senda, H. Izumi and H. Fujimura, “On Uracil Deriva-
tives and Related Compounds. 6,5-Alkyl-2,4,6-trioxoper-
hydrophyrimidine Derivatives as Antiphlogistics,” Arz-
neimittel-Forschung, Vol. 17, No. 12, 1967, pp. 1519-1523.
[3] K. Kitani, S. Tsuruoka, R. Miura and Y. Morita, “The
Effect of Bucolome on the Canalicular Bile Formation
and Sulfobromophthalein Transport Maximum in the
Dog,” Biochemical Pharmacology, Vol. 25, No. 12, 1976,
pp. 1377-1381. doi:10.1016/0006-2952(76)90107-6
[4] K. Kitani, M. Nokubo, S. Kanai, R. Miura and T. Uesugi,
“The Biliary Excretion of Bucolome in the Rat: A Possi-
ble Cause for Choleresis,” Biochemical Pharmacology,
Vol. 26, No. 24, 1977, pp. 2457-2461.
doi:10.1016/0006-2952(77)90461-0
[5] T. Yashiki, T. Matsuzawa, T. Kondo, Y. Uda and T.
Shima, “Studies on the Metabolic Fate and the Pharma-
cokinetics of 5-N-Butyl-1-cyclohexyl-2,4,6-trioxoperhy-
dropyrimidine (BCP) in Man. I. Identification of the Me-
tabolites of BCP in Human Urine,” Chemical & Pharma-
ceutical Bulletin, Vol. 19, No. 3, 1971, pp. 468-477.
[6] M. Kobayashi, M. Takagi, K. Fukumoto, R. Kato, K.
Tanaka and K. Ueno, “The Effect of Bucolome, a
CYP2C9 Inhibitor, on the Pharmacokinetics of Losartan,”
Drug Metabolism and Pharmacokinetics, Vol. 23, No. 2,
2008, pp. 115-119. doi:10.2133/dmpk.23.115
[7] H. Takahashi, T. Kashima, S. Kimura, N. Murata, T.
Takaba, K. Iwade, T. Abe, H. Tainaka, T. Yasumori and H.
Echizen, “Pharmacokinetic Interaction between Warfarin
and a Uricosuric Agent, Bucolome: Application of in Vi-
tro Approaches to Predicting in Vivo Reduction of (S)
Warfarin Clearance,” Drug Metabolism and Disposition,
Vol. 27, No. 10, 1999, pp. 1179-1186.
[8] T. Yashiki, T. Kondo, Y. Uda and H. Mima, “Studies on
the Metabolic Fate and the Pharmacokinetics of 5-N-Bu-
tyl-1-cyclohexyl-2,4,6-trioxoperhydropyrimidine (BCP) in
Man. II. Determination of BCP and Its Metabolite, 1-
Cyclohexyl-5-(3-hydroxybutyl)-2,4,6-trioxoperhydropyri-
midine by Ultraviolet Absorption Method,” Chemical &
Pharmaceutical Bulletin, Vol. 19, No. 3, 1971, pp. 478-
486.
[9] T. Yashiki, Y. Uda, T. Kondo and H. Mima, “Studies on
the Metabolic Fate and the Pharmacokinetics of 5-N-Bu-
tyl-1-cyclohexyl-2,4,6-trioxoperhydropyrimidine (BCP) in
Man. III. Gas-Liquid Chromatographic Determination of
BCP and Its Metabolites,” Chemical & Pharmaceutical
Bulletin, Vol. 19, No. 3, 1971, pp. 487-492.
[10] T. Yashiki, T. Matsuzawa, M. Yamada, T. Kondo and Y.
Uda, “Studies on the Metabolic Fate and the Pharma-
cokinetics of 5-N-Butyl-1-cyclohexyl-2,4,6-trioxoperhy-
dro-pyrimidine (BCP) in Man. IV. Pharmacokinetics of
BCP in Man Following Oral Administration,” Chemical
& Pharmaceutical Bulletin, Vol. 19, No. 5, 1971, pp. 869-
880.
[11] T. Yashiki, T. Matsuzawa, M. Yamada, T. Kondo, H.
Mima, M. Yamamoto, T. Yamada, M. Nakajima and K.
Doi, “Studies on the Metabolic Fate and the Pharmacoki-
netics of 5-N-Butyl-1-cyclohexyl-2,4,6-trioxoperhydro-
pyrimidine (BCP) in Man. V. Pharmacokinetics of BCP
in Man and in Rabbit Following Intravenous Administra-
tion of BCPNa,” Chemical & Pharmaceutical Bulletin,
Vol. 19, No. 5, 1971, pp. 881-891.
[12] B. K. Tang, W. Kalow and A. A. Grey, “Amobarbital
Metabolism in Man: N-Glucoside Formation,” Research
Communications in Chemical Pathology and Pharma-
cology, Vol. 21, No. 1, 1978, pp. 45-53.
[13] B. K. Tang, W. Kalow and A. A. Grey, “Metabolic Fate
of Phenobarbital in Man N-Glucoside Formation,” Drug
Metabolism and Disposition, Vol. 7, No. 5, 1979, pp. 315-
318.
[14] K. Mohri, T. Uesugi and K. Kamisaka, “Bucolome N-
Glucuronide: Purification and Identification of a Major
Metabolite of Bucolome in Rat Bile,” Xenobiotica, Vol.
15, No. 7, 1985, pp. 615-621.
doi:10.3109/00498258509045891
[15] K. Mohri, Y. Uesawa and T. Uesugi, “Metabolisme of
Bucolome in Rats Stability and Biliary Excretion of Bu-
colome N-Glucuronide,” Journal of Chromatography B:
Biomedical Sciences and Applications, Vol. 759, No. 1,
2001, pp. 153-159. doi:10.1016/S0378-4347(01)00218-3
[16] O. H. Lowry, N. J. Rosebrough, A. L. Farr and R. J. Ran-
dall, “Protein Measurement with the Folin Phenol Re-
agent,” The Journal of Biological Chemistry, Vol. 193,
No. 1, 1951, pp. 265-275.
[17] H. Scheffe, “The Analysis of Variance,” John Wiley &
Sons Inc., New York, 1959.
[18] S. Kaivosaari, P. Toivonen, L. M. Hesse , M. Koskinen, M.
H. Court and M. Finel, “Nicotine Glucuronidation and the
Human UDP Glucuronosyltransferase UGT2B10,” Mo-
lecular Pharmacology, Vol. 72, No. 3, 2007, pp. 761-768.
doi:10.1124/mol.107.037093
[19] S. Kaivosaari, P. Toivonen, O. Aitio, J. Sipilä, M. Koski-
nen, J. S. Salonen and M. Finel, “Regio- and Stereospeci-
fic N-Glucuronidation of Medetomidine: The Differences
between UDP Glucuronosyltransferase (UGT) 1A4 and
UGT2B10 Account for the Complex Kinetics of Human
Liver Microsomes,” Drug Metabolism and Disposition,
Vol. 36, No. 3, 2008, pp. 1529-1537.
doi:10.1124/dmd.108.021709
[20] S. Kaivosaari, M. Finel and M. Koskinen, “N-Glucuroni-
dation of Drugs and Other Xenobiotics by Human and
Animal UDP-Glucuronosyltransferases,” Xenobiotica, Vol.
41, No. 8, 2011, pp. 652-659.
doi:10.3109/00498254.2011.563327
[21] B. K. Tang, “Drug Glucosidation,” Pharmacology & The-

Bucolome N-Glucuronide Formation: Species Differences and Identification of Human
UDP-Glucuronosyltransferase Isoforms
369
rapeutics, Vol. 46, No. 1, 1990, pp. 53-56.
doi:10.1016/0163-7258(90)90034-Y
[22] S. G. Paibir and W. H. Soine, “High-Performance Liquid
Chromatographic Analysis of Phenobarbital and pheno-
barbital Metabolites in Human Urine,” Journal of Chro-
matography B: Biomedical Sciences and Applications,
Vol. 691, No. 1, 1997, pp. 111-117.
[23] S. M. Neighbors and W. H. Soine, “Identification of Phe-
nobarbital N-Glucuronides as Urinary Metabolites of Phe-
nobarbital in Mice,” Drug Metabolism and Disposition,
Vol. 23, No. 5, 1995, pp. 548-552.
[24] L. Luukkanen, J. Taskinen, M. Kurkela, R. Kostiainen, J.
Hirvonen and M. Finel, “Kinetic Characterization of the
1A Subfamily of Recombinant Human UDP-Glucurono-
syltransferases,” Drug Metabolism and Disposition, Vol.
33, No. 7, 2005, pp. 1017-1026.
doi:10.1124/dmd.105.004093
[25] S. Ohno and S. Nakajin, “Determination of mRNA Ex-
pression of Human UDP-Glucuronosyltransferases and
Application for Localization in Various Human Tissues
by Real-Time Reverse Transcriptase-Polymerase Chain
Reaction,” Drug Metabolism and Disposition, Vol. 37,
No. 1, 2009, pp. 32-40. doi:10.1124/dmd.108.023598
[26] T. Zhang, P. Haws and Q. Wu, “Multiple Variable First
Exons: A Mechanism for Cell- and Tissue-Specific Gene
Regulation,” Genome Research, Vol. 14, No. 1, 2004, pp.
79-89. doi:10.1101/gr.1225204
[27] P. I. Mackenzie, K. W. Bock, B. Burchell, C. Guillemette,
S. Ikushiro, T. Iyanagi, J. O. Miners, I. S. Owens, D. W.
Nebert, “Nomenclature Update for the Mammalian UDP
Glycosyltransferase (UGT) Gene Superfamily,” Phar-
macogenetics and Genomics, Vol. 15, 2005, pp. 677-685.
doi:10.1097/01.fpc.0000173483.13689.56
[28] C. D. King, M. D. Green, G. R. Rios, B. L. Coffman, I. S.
Owens, W. P. Bi sho p and T. R. Tephly , “The Gluc uronida-
tion of Exogenous and Endogenous Compounds by Stably
Expressed Rat and Human UDP-Glucuronosyltransferase
1.1,” Archives of Biochemistry and Biophysics, Vol. 332,
No. 1, 1996, pp. 92-100. doi:10.1006/abbi.1996.0320
[29] J. R. Chowdhury, R. Kondapalli and N. R. Chowdhury,
“Gunn Rat: A Model for Inherited Deficiency of Bilirubin
Glucuronidation,” Advances in Veterinary Science & Com-
parative Medicine, Vol. 37, 1993, pp. 149-173.
[30] H. Kanoh, M. Tada, S. Ikushiro and K. Mohri, “Charac-
terization of Bucolome N-Glucuronide Formation: Tissue
specificity and Identification of Rat UDP-Glucuronosyl-
transferase Isoform (s),” Pharmacology & Pharmacy, Vol.
2, No. 3, 2011, pp. 151-158. doi:10.4236/pp.2011.23021
[31] H. Kanoh, K. Okada and K. Mohri, “Identification of the
UDP-Glucuronosyltransferase Responsible for Bucolome
N-Glucuronide Formation in Rats,” Pharmazie, Vol. 65,
No. 11, 2010, pp. 840-844.