In vitro biotransformation of 2,6,9-trisubstituted purine-derived cyclin-dependent kinase inhibitor bohemine by mouse liver microsomes

Article abstract of DOI:10.1080/0049825021000012600

1. Biotransformation pathways of the cyclin-dependent kinase inhibitor 6-benzylamino-2-(3-hydroxypropylamino)-9-isopropylpurine (bohemine) by mouse liver microsomes in vitro were investigated. 2. Metabolite profiles of [8-³H]-labelled bohemine were established by TLC/³H-autoradiography and enzymatic and MS analyses were used to elucidate the chemical structures of the metabolites. The structures of the main primary metabolites were confirmed by synthesis of authentic compounds. 3. A schema of the primary NADPH-dependent pathways has been proposed involving N²- and N9-dealkylation, N⁶-debenzylation, aromatic hydroxylation, and C2 side chain oxidation of bohemine. Three of the primary metabolites detected, 6-(benzylamino)-2-(3-hydroxypropylamino)purine (M4), 6-amino-2-(3-hydroxypropylamino)-9-isopropylpurine (M5) and 6-(4-hydroxybenzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine (M6), all retaining their parent primary hydroxyl group, were subsequently shown to be converted, by a liver cytosolic NAD⁺-dependent system, into their corresponding carboxylic acids. M6 was subject to microsomal glycosidations requiring UDP-sugar donors. NADPH-dependent conversion of M6 into M5 by microsomes was also demonstrated. 4. Cytochrome P450 (CYP) enzymes-selective inhibitors were used to identify CYPs involved in bohemine biotransformation. The findings suggested that CYP2a and CYP3a substantially contributed to the NADPH-dependent bohemine transformation in vitro. 5. The findings will facilitate experiments designed to dissect enzymatic systems catalysing clearance of C2,C6,N9-trisubstituted purine compounds from mammalian tissues.

Full text of DOI:10.1080/0049825021000012600

xenobiotica,
vol.
no.
32, 11, 1017±1031
2002,
In vitro biotransformation of 2,6,9-trisubstituted
purine-derived cyclin-dependent kinase inhibitor
bohemine by mouse liver microsomes
{
M. RYPKA , J. VESELY *, Z. CHMELA , D. RIEGROVA ,
K. CERVENKOVA , L. HAVLICEK , K. LEMR , J. HANUS ,
{
{
{
{
}
{
}
}
B. CERNY , J. LUKES|| and K. MICHALIKOVA
Department of Pathological Physiology, Medical Faculty, Palacky University, Olomouc,
{
{
Czech Republic
{
Republic
Institute of Nuclear Medicine, Medical Faculty I, Charles University, Prague, Czech
}
Olomouc, Czech Republic
Department of Analytical Chemistry, Faculty of Natural Science, Palacky University,
}
Institute of Experimental Botany, Academy of Science, Prague, Czech Republic
|| Department of Nuclear Medicine, University Hospital in Olomouc, Czech Republic
Received 17 April 2002
1. Biotransformation pathways of the cyclin-dependent kinase inhibitor 6-benzylami-
no-2-(3-hydroxypropylamino)-9-isopropylpurine (bohemine) by mouse liver microsomes
in vitro were investigated.
2. Metabolite pro®les of [8-3H]-labelled bohemine were established by TLC/3H-
autoradiography and enzymatic and MS analyses were used to elucidate the chemical
structures of the metabolites. The structures of the main primary metabolites were
con®rmed by synthesis of authentic compounds.
3. A schema of the primary NADPH-dependent pathways has been proposed
involving N2- and N9-dealkylation, N6-debenzylation, aromatic hydroxylation, and C2
side chain oxidation of bohemine. Three of the primary metabolites detected, 6-
(benzylamino)-2-(3-hydroxypropylamino)purine (M4), 6-amino-2-(3-hydroxypropylami-
no)-9-isopropylpurine (M5) and 6-(4-hydroxybenzylamino)-2-(3-hydroxypropylamino)-
9-isopropylpurine (M6), all retaining their parent primary hydroxyl group, were
subsequently shown to be converted, by a liver cytosolic NAD‡-dependent system, into
their corresponding carboxylic acids. M6 was subject to microsomal glycosidations
requiring UDP-sugar donors. NADPH-dependent conversion of M6 into M5 by
microsomes was also demonstrated.
4. Cytochrome P450 (CYP) enzymes-selective inhibitors were used to identify CYPs
involved in bohemine biotransformation. The ®ndings suggested that CYP2a and CYP3a
substantially contributed to the NADPH-dependent bohemine transformation in vitro.
5. The ®ndings will facilitate experiments designed to dissect enzymatic systems
catalysing clearance of C2,C6,N9-trisubstituted purine compounds from mammalian
tissues.
Introduction
Recently, several types of selective synthetic inhibitors of cyclin-dependent
kinases (CDKs) have been implicated as novel and invaluable reagents for mol-
ecular biological, biochemical and morphological studies of the cell cycle (reviewed
in Meijer et al. 1999, Sielecki et al. 2000). The antiproliferative activity of the
inhibitors has been well documented both in vitro and in vivo, a fact that has
* Author for correspondence; e-mail: vesely@tunw.upol.cz
Xenobiotica ISSN 0049±8254 print/ISSN 1366±5928 online
http://www.tandf.co.uk/journals
2002 Taylor & Francis Ltd
#
DOI: 10.1080/0049825021000012600

1018
M. Rypka et al.
encouraged a number of authors to consider the inhibitors as potentially new
anticancer drugs (reviewed in Buolamwini 2000, Fischer and Lane 2000). Unlike
substances that kill transformed cells, CDK inhibitors seem able to re-establish
appropriate mechanisms governing cells’ fate. The potential use of the inhibitors is
also being investigated in various clinical domains such as nephrology, dermatol-
ogy and vascular diseases (reviewed in Gray et al. 1998, 1999). The survival-
promoting, anti-apoptotic activity of the inhibitors on nerve cells seems note-
worthy in respect of Alzheimer disease (reviewed in Meijer 2000). Furthermore,
these inhibitors have been reported to alter the rhythm of a molluscan circadian
pacemaker and inhibit both virus and unicellular parasite cycling (reviewed in
Meijer et al. 2000, Rosania and Chang 2000).
The ®rst CDK-selective inhibitor derived from purines to be discovered was
the 2,6,9-trisubstituted purine olomoucine (Vesely et al. 1994). Since then,
thousands of new purine analogues have been synthesized, using a solid-phase
approach and/or computer-aided drug design combined with a structure-based
library strategy. The new compounds have been screened for their activity as well
as for di€erent aspects of their cellular e€ects (Gray et al. 1998, 1999, Chang et al.
1999, Legraverend et al. 1999, Meijer et al. 1999, Sausville et al. 1999). To date,
among the CDK inhibitors, the purines are best understood in terms of their
structure±activity relationships (Fischer and Lane 2000, Garcõa-Echeverrõa et al.
2000). Nonetheless, the metabolic pro®le of these inhibitors remains poorly
understood.
Chmela et al. (2001) studied the metabolism of an olomoucine-type CDK-
selective inhibitor, 6-benzylamino-2-(3-hydroxypropylamino)-9-isopropylpurine,
designated as bohemine (Sielecki et al. 2000), in mouse in vivo. Bohemine was
metabolized rapidly and was almost exclusively excreted in the form of metabo-
lites. The liver and kidney were the main organs responsible for fast bohemine
clearance. The principal Phase I bohemine reaction was bohemine oxidation to a
carboxylic acid, which could be catalysed by a cytosolic, 4-methylpyrazole-
sensitive, alcohol dehydrogenase. The Phase II-predominant biotransformation
route was UDP-glucoside-dependent bohemine -glucosidation. The third meta-
-
bolite identi®ed by Chmela et al. was bohemine glucuronide.
Generally, among Phase I drug-metabolizing enzymes, microsomal cyto-
chromes P450 (CYPs) are known to exert the most prominent role (reviewed in
Rendic 2002), are widely distributed in the tissues of all mammalian species and
are most abundant in mammalian liver (Venkatakrishnan et al. 2001). It was a
challenge to examine the participation of mouse liver microsomes in bohemine
metabolism since, surprisingly, the aforementioned in vivo study (Chmela et al.
2001) failed to reveal any speci®c individual products of their activity. Some
preliminary results of the current study were published as conference abstracts
(Chmela et al. 1998, Belejova et al. 1999, Michalõkova et al. 1999, Cervenkova et al.
2000).
Materials and methods
Chemicals
2,6-Dichloropurine, 2-amino-6-chloropurine, -NADPH (N1630), -glucosidase, EC 3.2.1.21
-
(G4511), -glucosidase, EC 3.2.1.20, recombinant form, from Saccharomyces cerevisiae (G0660),
-
¬
UDP-glucuronic acid, UDP-glucose, coumarin, 8-methoxypsoralen, troleandomycin and ketoconazole

Biotransformation of bohemine by mouse liver microsomes
1019
were purchased from Sigma-Aldrich (St Louis, MO, USA); bovine serum albumin (fraction V) was
from Bioveta (Ivanovice n.H., Czech Republic); and Silica gel 60 F254 TLC sheets and Silica gel for
m
column chromatography (Kieselgel 60, particle size 40±63 m) were purchased from Merck (Darmstadt,
Germany). All other chemicals and reagents used were of analytical or HPLC grade.
Chemical syntheses
Solvents were evaporated in vacuo at 80 C, unless otherwise speci®ed. Melting points (mp) were
8
determined on a Ko¯er block and were uncorrected.
<
Bohemine was synthesized as described by Havlõcek et al. (1997), and its carboxylic derivative 6-
benzylamino-2-(2-carboxyethylamino)-9-isopropylpurine and 6-benzylamino-2-(3-hydroxypropyla-
3
3
ˆ
mino)-9-isopropyl [8- H]purine ( [ H]bohemine) were prepared as described by Chmela et al.
¡
(2001). The speci®c activity of the [3H]bohemine was 587 TBq mol and its radiochemical purity
was at least 98% when checked by TLC on Silikagel 60 F254 sheets (Merck) with chloroform±methanol
(95:5).
Bohemine metabolite reference compounds were synthesized and labelled by 3H at position 8 of
their purine rings following the reaction strategy described previously (Havlõcek et al. 1997, Otyepka
1
§
et al. 2000, Chmela et al. 2001). All synthesized compounds gave a satisfactory element analysis ( 0.4%)
and their chemical structures were veri®ed by MS and NMR measurements.
2,6-Dichloro-9-isopropylpurine
2,6-Dichloropurine (5.1 g, 27 mmol), potassium carbonate (6.8 g, 49mmol) and isopropyliodide
(5 ml, 50mmol) were vigorously stirred in DMSO (50 ml) overnight. After evaporation of the volatile
components, the product (i.e. the principal spot on TLC) was isolated by means of column
8
chromatography (20% ethylacetate in heptane); yield 65%; mp 150±151 C. The compound was not
labelled by 3H.
1
ˆ
H-NMR (400 MHz, CD3OD): 1.672 (6H, d, J 6.8, (CH3)2), 4.932 (1H, m, CH(CH3)2), 8.673
(1H, s, HC8). 13C-NMR (100 MHz, CD3OD): 22.74 Qqd (J 128.0, J 4.6, J 4.6, CH3), 50.79 Dsep
ˆ
ˆ
ˆ
ˆ
(J 143.4, J 4.4, CH), 132.41 d (J 11.6, C-5), 147.67 Dd (J 214.0, J 4.4, C-8), 152 s, 153.91 s,
ˆ
ˆ
ˆ
ˆ
154.75 bm.
6-Benzylamino-2-(3-hydroxypropylamino)purine (an M4 spot metabolite reference)
6-Benzylamino-2-chloropurine (0.102 g, 0.39 mmol) prepared according to Hocart et al. (1991) and
8
3-hydroxypropylamine (3 ml) were heated to 160 C for 2 h. Excess amine was evaporated and the
product was crystallized from 1-butanol; yield 67%, mp 201±202 C. The speci®c activity of the 3H-
8
labelled M4 spot metabolite reference was 561 TBq mol¡1, a its radiochemical purity was at least 98%
when checked by TLC on Silikagel 60 F254 sheets (Merck) with chloroform±methanol±conc. aqueous
ammonia (90:10:0.2) and by HPLC on RP-18 with methanol±water (9:1).
MS (VG7070E): 298 (M‡:, 55%), 267 (17), 253 (26), 175 (17), 129 (18), 111 (15), 106 (18), 97 (33), 91
(77), 81 (42), 71 (48), 60 (43), 57 (95), 55 (88), 43 (100), 41 (84). ESI-MS (LCQ Finnigan MAT): m/z
2
‡
[M 1]
‡ ˆ
ˆ
299; ESI-MS : m/z (%) 91 (47.3), 135 (1.9), 136 (4.6), 150 (5.7), 163 (32.9), 164 (45.2), 165
(23), 175 (3.1), 177 (66.4), 178 (26.5), 189 (51.1), 191 (26.7), 199 (1.3), 200 (1.9), 203 (6.3), 207 (6.3), 208
(55.1), 221 (6.3), 225 (7.6), 239 (1.3), 240 (21.9), 241 (24.3), 253 (96.3), 257 (1.1), 264 (1.0), 281 (100).
1
ˆ
H-NMR (400MHz, CDCl3): 1.791 (2H, pen, J 6.4, CH2CH2CH2), 3.470 (2H, t, J 6.5, CH2O or
ˆ
ˆ
CH2NH), 3.641 (2H, t, J 6.3, CH2O or CH2NH), 4.768 (2H, s, CH2Ph), 7.22±7.42 (5H, m, Ph),
7.684 (1H, s, HC8).
6-Amino-2-(3-hydroxypropylamino)-9-isopropylpurine (an M5 spot metabolite reference)
2,6-Dichloro-9-isopropylpurine (199mg, 0.86 mmol) was heated in 15 ml conc. NH4OH to 85 C for
8
9 h. After cooling, the crude product was ®ltered and rinsed with water. The product (mp 257±259 C)
8
and 3-hydroxypropylamine (3 ml) were heated in a sealed ampoule to 165±170 C for 5 h. Excess amine
8
was evaporated, and the residue was crystallized from water; yield 30%, mp 143±144C. Recrystallization
was achieved in methanol±diethylether, mp¡143±145 C. The speci®c activity of the 3H-labelled M5
8
spot metabolite reference was 545 TBq mol 1, and its radiochemical purity was at least 98% when
checked by TLC on Silikagel 60 F254 sheets (Merck) with chloroform±methanol±conc. aqueous
ammonia (90:10:0.2) and by‡HPLC on RP-18 with methanol±water (9:1).
MS (VG7070E): 250 (M :, 83%, C11H18N6O; 0.0), 219 (60), 206 (73), 205 (98), 192 (25), 177 (49),
‡ ˆ
2
‡
163 (100), 150 (50), 134 (65). ESI-MS (LCQ Finnigan MAT): m/z [M 1]
251; ESI-MS : m/z
(%) 151 (1.3), 163 (7.3), 191 (17.4), 205 (8.8), 209 (100), 233 (9.7). 1H-NMR (400 MHz, CDCl3):
ˆ
ˆ
COSY [1.546 (6H, d, J 6.8, (CH3)2CH), 4.63 (1H, hept, J 6.8, (CH3)2CH)], COSY [1.73 (2H, m,
ˆ
ˆ
CH2CH2CH2), 3.59±3.67 (4H, m, CH2CH2CH2)], COSY [4.97 (1H, bt, J 6.7, NH), 3.59±3.67], 5.40
(2H, bs, NH2), 7.57 (1H, s, HC8).

1020
M. Rypka et al.
6-(4-Hydroxybenzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine (an M6 spot metabolite
reference)
2,6-Dichloro-9-isopropylpurine (508 mg, 2.2 mmol), 4-hydroxybenzylamine (406 mg, 3.3 mmol)
prepared according to Jones and Pyman (1925), and triethylamine (1 ml) were heated and stirred in
8
6 ml 1-butanol (115 C, 2 h). After evaporation of the volatile components, the crude product was
isolated by stepwise gradient column chromatography (0, 1, 2, 3, 4% methanol in CHCl3) and
8
crystallized from benzene; yield 95%, mp 100±105 C. The product (115 mg) and 3-amino-1-propanol
(3 ml) were heated in a sealed ampoule to 160 C for 2 h. Excess amine was evaporated. The ®nal
8
product was isolated by stepwise gradient column chromatography (1, 2, 3, 4% methanol in CHCl3) and
crystallized from ethylacetate; yield 50% when calculated on 2,6-dichloro-9-isopropylpurine, mp 169±
¡
170 C. The speci®c activity of the 3H-labelled M6 spot metabolite reference was 560 TBq mol 1, and
8
its radiochemical purity was at least 98% when checked by TLC on Silikagel 60 F254 sheets (Merck)
with chloroform±methanol (9:1) and by TLC on aluminium oxide 60 F254 neutral sheets (Merck) with
toluene±methanol (9:1).
MS (Jeol JMS-D100): 356 (M‡:, 43%, C18H24N6O2, 0 3), 250 (46%, C11H18N6O, 1 5), 219 (62%,
‡
‡
:
C10H15N6O, 0 4), 164 (45), 163 (100), 150 (55), 134 (73), 122 (91), 108 (48), 107 (65). ESI-MS (LCQ
:
‡
:
Finnigan MAT): m/z [M 1]
‡ ˆ
ˆ
357; ESI-MS : m/z (%) 209 (5.8), 251 (100), 263 (4.4). 1H-NMR
2
‡
ˆ
(200MHz, CDCl3): 1.509 (6H, d, J 6.8, (CH3)2CH), 1.70 (2H, m, CH2CH2CH2), 3.60 (4H, t,
‡
CH2CH2CH2), 4.60 (3H, m, CH2Ph CH(CH3)), 6.71 (2H, d, 7.5, Ph), 7.10 (2H, d, 7.5, Ph), 7.512
(1H, s, HC8).
6-(3-Hydroxybenzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine
2,6-Dichloro-9-isopropylpurine (487 mg, 2.11mmol), 3-hydroxybenzylamine (431 mg, 3.5 mmol)
prepared according to Epstein et al. (1964), and triethylamine (1 ml) were heated and stirred in 10 ml 1-
8
butanol (115 C, 2 h). After evaporation of volatile components the crude product was isolated by
stepwise gradient column chromatography (0, 1, 2, 3, 4% methanol in CHCl3) and crystallized from
8
CHCl3; yield 95%, mp 217±218 C. The product (640mg) and 3-amino-1-propanol (3 ml) were heated
in a sealed ampoule to 160 C for 4 h. Excess amine was evaporated. The ®nal product was isolated by
8
stepwise gradient column chromatography (1, 2, 3, 4% methanol in CHCl3) and crystallized from
ethylacetate-diethylether; yield 17% when calculated on 2,6-dichloro-9-isopropylpurine, mp 133±
135 C. The compound was ‡not labelled by 3H.
8
MS (VG7070E): 356 (M :, 100%, C18H24N6O2; 0,1), 325 (43), 312 (42), 311 (43), 298 (24), 191
¡
(17), 122 (24), 107 (62). 1H-NMR (400MHz, CDCl3): 1.562 (6H, d, J 6.8, (CH3)2CH), 1.747 (2H,
ˆ
‡
ˆ
m, CH2CH2CH2), 3.65 (4H, m, CH2CH2CH2), 4.59 (2H, bd, CH2Ph), 4.66 (1H, sept, J 6.8,
ˆ
CH(CH3)2), 5.034 (1H, t, J 6.9, exc. H), 6.4 (1H, bs, exc. H), 6.73±6.82 (3H, m, Ph), 7.06±7.12
(1H, m, Ph), 7.265 (1H, s, HC8), 7.605 (1H, s, phenolic H). COSY [1.562 d and 4.66 sept], COSY
[1.747m and 3.65 m], COSY [4.59 bd and 6.4 bs], COSY [3.65 m and 5.034 t], COSY [7.09 m and
6.77 m].
6-(2-Hydroxybenzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine
2,6-Dichloro-9-isopropylpurine (231 mg, 1 mmol), 2-hydroxybenzylamine (185 mg, 1.5 mmol) pre-
pared according to Raiford and Clark (1923) [17] and triethylamine (0.41ml) were heated and stirred in
8
10 ml 1-butanol (115 C, 3 h). After evaporation of the volatile components, the crude product was
isolated by column chromatography (4% MeOH in CHCl3) and crystallized from diethylether; yield
8
95%, mp 170±171 C. This product (226 mg) and 3-amino-1-propanol (2ml) were heated in a sealed
ampoule to 160 C for 3h. Excess amine was evaporated. The residue was puri®ed by stepwise gradient
8
column chromatography (0, 1, 1.5% methanol in CHCl3) and a ®nal syrupy product (125mg) was
obtained; yield 35% when calculated on 2,6-dichloro-9-isopropylpurine. The compound was not
labelled by 3H.
MS (VG7070E): 356 (M‡:, 57%), 325 (6), 312 (8), 298 (7), 250 (69), 219 (65), 206 (71), 205 (100), 177
(37), 163 (72), 150 (37), 134 (50), 108 (38), 106 (28), 78 (53). 1H-NMR (200 MHz, CDCl3): 1.52 (6H, d,
ˆ
J
CH2Ph), 4.57 (1H, hept, J 6.6, (CH3)CH), 5.11 (1H, bt, exc. H), 6.53 (1H, bs, exc. H), 6.82±6.93 (2H,
6.6, (CH3)2CH), 1.78 (2H, m, CH2CH2CH2), 3.66 (4H, m, CH2CH2CH2), 4.54±4.60 (2H, bs,
ˆ
m, Ph), 7.17±7.27 (2H, m, Ph), 7.52 (1H, s, HC8).
2-Amino-6-benzylamino-9-isopropylpurine (an M8 spot metabolite reference)
2-Amino-6-chloropurine (170 mg, 1.4 mmol) and benzylamine (3.8 ml, 35mmol) were heated with
8
stirring in 3ml n-butanol to 110 C for 3h. After cooling, the crude product, 2-amino-6-benzylamino-
purine, was ®ltered. This product (170mg, yield 51% when calculated on 2-amino-6-chloropurine) was
vigorously stirred with powdered potassium carbonate (0.5 g, 3.6 mmol), and isopropyliodide (0.5ml,
5 mmol) in 5 ml DMSO overnight. After evaporation of the volatile components the ®nal product was
puri®ed by column chromatography with 1% methanol in CHCl3 and crystallized from methanol±

Biotransformation of bohemine by mouse liver microsomes
1021
8
diethylether; yield 34% when calculated on 2-amino-6-chloropurine, mp 191±193 C. The compound
was not labelled by 3H.
MS (VG7070E): 282 (M‡:, 100%, C15H18N6, 0 2), 239 (52), 177 (42), 135 (62%, C H N , 0.0), 106
¡
:
(82), 91 (80). ESI-MS (LCQ Finnigan MAT): m/z [M 1]
5
‡ ˆ
5ˆ5
2
‡
283; ESI-MS : m/z (%) 91 (23.5), 150
1
(2.2), 163 (1.2), 192 (13.7), 205 (2.3), 241 (100). H-NMR (400 MHz, CDCl3): 1.54 (6H, d, J 6.8,
ˆ
ˆ
(CH3)2CH), 4.68 (1H, hept, J 6.8, (CH3)2CH), 4.68 (2H, bs, NH2), 4.81 (2H, bd, CH2), 5.82 (1H, bs,
NH), 7.26±7.40 (5H, m, Ph), 7.535 (1H, s, HC8).
Incubation of bohemine and its metabolites with liver microsomal and cytosolic subcellular fractions
All experiments using animals were authorized by the Institutional Ethical Committee and carried
out according to institutional guidelines in compliance with Czech national laws and in agreement with
the Declaration of Helsinki. Monitored male NMRI outbred mice of mixed genetic background and of
approximately 20±25 g body weight between 5 and 7 weeks of age (VUFB, Konarovice, Czech Republic)
were used to obtain subcellular fractions.
M
The fractions were prepared in 0.25 sucrose solution by di€erential centrifugation and their
protein content was determined as described by Chmela et al. (2001). The microsomes were transferred
M
by repeated centrifugation (100000g, 60 min) into phosphate bu€er (0.1 , pH 7.4) shortly before use.
Two sets of microsomal incubations were run in the phosphate bu€er: one in the presence of NADPH,
the other containing glycosyl donors. The incubations conducted in the presence of glycosyl donors
M
UDP-glucose (5m ). They were further supplemented with either about 10kBq 3H-labelled metabo-
(0.2 ml, ®nal volume) employed 0.4 mg microsomal protein, and either UDP-glucuronic acid (5 m ) or
M
lite spot extract or with a newly synthesized, 3H-labelled authentic reference compound (296 kBq; the
m
concentration was adjusted to 5 m by an addition of the appropriate unlabelled compound) as
substrates. The NADPH mixtures (0.2 ml, ®nal volume) employed 0.10 mg microsomal protein,
3
M
M
mm
5 m NADPH, 5 m MgCl2 and were supplemented with either 296 kBq [ H]bohemine (3 ), or
3
about 10kBq [ H]-labelled metabolite spot extract or with 296 kBq (5
3H-labelled authentic
mm
)
reference compound. Incubations with mouse liver cytosolic fraction (0.2 mg protein) in the presence
M
of NAD‡ coenzyme (5 m ) were performed as described by Chmela et al. (2001) in the above phosphate
3
mm
bu€er supplemented with about 10kBq [ H]-labelled metabolite spot extract, or with 296 kBq (5
)
3H-labelled authentic reference compound. Finally, to identify the CYPs involved in the bohemine
metabolism, CYP-selective inhibitors were included in the NADPH-containing assays at concentra-
tions indicated in the Results. Each of the mechanism-based inhibitors ketoconazole, 8-methoxypsor-
8
alen or troleandomycin were pre-incubated with microsomes in the presence of NADPH at 37 C for
15 min (Newton et al. 1995, Chauret et al. 1997, Court et al. 1997) before the reaction which was
m
8
initiated by the addition of radioactive substrate (20 l). All reactions were allowed to proceed at 37 C
under continuous mixing in open tubes. For the inhibition studies, aliquots removed within an
incubation interval between 5 and 20 min were selected after our preliminary tests with mouse liver
microsomes revealed a linear reaction during this period.
Extraction of metabolites and TLC analysis
For [3H]-metabolite analysis, incubations were stopped, homogenized and extracted with a ninefold
volume of ethanol. The ethanolic extracts (supernatants, 4000g, 5 min) were analysed by TLC with
subsequent 3H-autoradiography and densitometry. In particular, TLC was routinely run in acidic
mobile phase chloroform±methanol±acetic acid 90:10:1 (by vol.; phase A); this phase was used to resolve
the most polar spots localized in the vicinity of the origin. Acetic acid was replaced by 1 vol. conc.
aqueous ammonia (phase B) to diagnose (by changing TLC mobility of the spots) the acidic/basic
character of substances. The spots were visualized by autoradiography, scraped o€ and extracted to
identify metabolite chemical structures that were analysed by means of chemical modi®cation with
diazomethane, enzymatic analyses and mass spectroscopy (Chmela et al. 2001).
Enzymatic analyses
Incubates of metabolites extracted from TLC spots containing about 10kBq radioactivity were
supplemented either with 0.67 Sigma units -glucuronidase or 2.5 Sigma units -glucosidase in a citrate
M
-
bu€er (0.05 , pH 5.0). The -glucosidase incubation mixture contained the recombinant enzyme (2.5
-
¬
Sigma units) in the above phosphate bu€er. The mixtures were then incubated in open tubes under
8
continuous gentle mixing at 37 C. Aliquots were removed at 10, 30 and 120min. The volumes of
incubation mixtures were increased 10±20 times for preparatory purposes.
Spectroscopic methods
Electron ionization mass spectra (MS) of newly synthesized authentic compounds (see above) were
m
recorded on a VG-7070E instrument (EI, ionizing energy 70 eV, 50 or 100 A, 200 C, direct inlet;
purchased from Vacuum Generators, Manchester, UK) or a Jeol JMS-D100 (EI, 80eV, 300 A, 200 C,
8
m
8

1022
M. Rypka et al.
direct inlet; purchased from Japan Electron Optics Laboratory, Tokyo, Japan) with a peak-matching
method using per¯uorokerosene as a standard [m/z (%, elem. comp., deviation in ppm)]. A soft
ionization MS of both the newly synthesized references and metabolites extracted from TLC spots
(see the Results) was carried out using an ion-trap mass spectrometer LCQ Finnigan MAT (San Jose,
CA, USA). Samples were either standards diluted with 1% acetic acid in ethanol, or ethanolic extracts of
TLC sp¡ots acidi®ed with acetic acid to the 1% concentration. Samples were introduced with a ¯ow-rate
1
m
2
l min to the electrospray interface with source voltage 5.6 kV, sheath gas ¯ow 28 (arbitrary units)
8
and a capillary temperature of 200 C. The selected ions were fragmented in an ion trap (m/z isolation
width 6.0, collision energy 25%). Mass spectra were scanned in full-scan mode as well as in a zoom scan
and MSn. 1H- and 13C-NMR spectra of newly synthesized reference compounds were measured either
on a Varian VXR-400 instrument at 400 MHz (1H) or 100 MHz (13C), or on a Varian XL-200 (Palo
Alto, CA, USA) at 200 MHz. J were given in Hz.
Results
Metabolite pro®les in microsomal incubates supplemented with NADPH
Chmela et al. (2001) identi®ed three bohemine metabolites that were desig-
nated as M1 (bohemine -glucuronide), M2 (bohemine -glucoside) and M3
-
-
(bohemine carboxylic acid). The metabolites newly detected in this study were
marked according to the same scheme, i.e. M4±M8 as described below. From
®gure 1 it can be seen that most [3H]bohemine ( 85%) was metabolized into its
>
metabolites during a 120-min incubation of the parent compound in mouse liver
microsomes supplemented with NADPH. The incubation resulted in a complex
metabolite pro®le. The microsomes gave a spot M3 identical to that described by
Chmela et al. (2001), which was con®rmed as containing a carboxylic acid, 6-
benzylamino-2-(2-carboxyethylamino)-9-isopropylpurine. This aside, ®ve addi-
Figure 1. 3H-autoradiographic and densitometric analysis of [3H]bohemine incubation with mouse
liver microsomes in the presence of NADPH. The incubation interval was 120 min and incubates
were resolved by TLC in chloroform±methanol±acetic acid 90:10:1 (phase A, upper) or in
chloroform±methanol±conc. aqueous ammonia 90:10:1 (phase B, bottom). B, bohemine; S, start
(origin).

Biotransformation of bohemine by mouse liver microsomes
1023
tional distinct spots M4±M8 were detected. M4 and M5 represented the major
spots, whereas M6±M8 were minor components (®gure 1).
Elucidation of new metabolite structures
The chemical structures of the new metabolites formed in the NADPH-
microsomal system (®gure 1) were elucidated as follows. A metabolite in a spot
and was tentatively identi®ed by MSn
ˆ
M4 extract displayed m/z 299 M
‰
‡ Š‡
scan as 6-benzylamino-2-(3-hydroxypropylamino)purine, a product of bohemine
1
deisopropylation (table 1). A metabolite detected in an M5 extract exhibited
‡
ˆ
‡
m/z 251 [M 1] and was similarly identi®ed as 6-amino-2-(3-hydroxypropyl-
amino)-9-isopropylpurine, a product of bohemine debenzylation (table 1). The
chemical structures of both products were con®rmed by comparing their TLC
characteristics and MS spectra with those of authentic reference compounds (see
the Materials and methods) (table 1).
MSn analysis also suggested that a microsomal metabolite extracted from spot
‡
ˆ
‡
M6 and exhibiting m/z 357 [M 1] could be a product of hydroxylation
directed to the benzyl ring attached to the N6-atom of bohemine. This was
supported by a reaction with diazomethane (table 1). Subsequently, all three
potential positional isomers, i.e. 6-(4-hydroxybenzylamino)-, 6-(3-hydroxybenzy-
lamino)- and 6-(2-hydroxybenzylamino)-2-(3-hydroxypropylamino)-9-isopropyl-
purine were synthesized. A comparison of their TLC and MS characteristics with
Table 1. Primary products of NADPH-dependent transformation of bohemine: RF, methylation and
spectroscopic data.
Compound
Analytical method
Data
Boheminea
M3a
TLC
TLC
Rf: 0.63 (phase A), 0.75 (phase B)
Rf: 0.55 (phase A), 0.01 (phase B)
Rf: 0.07±0.1‡7 (phase A), 0.23 (phase B)
M4
TLC
ESI-MS
‡
m/z [M 1] : 299
ESI-MS2
m/z (%): 91 (34), 163 (25), 164 (34), 165 (18), 177 (54),
178 (20), 189 (45), 191 (25), 207 (6), 208 (47), 240 (11),
241 (20), 253 (79), 281 (100)
M5
M6
TLC
ESI-MS
Rf: 0.26 (ph‡ase A), 0.28 (phase B)
‡
m/z [M 1] : 251
ESI-MS2
m/z (%): 191 (14), 205 (9), 209 (100), 233 (12)
TLC
ESI-MS
Rf: 0.31 (ph‡ase A), 0.33 (phase B)
‡
m/z [M 1] : 357
ESI-MS2
m/z (%): 209 (7), 251 (100)
Rf: 0.62 (phase A), 0.75 (phase B)
TLC of an M6
methylderivative
M7
M8
TLC
ESI-MS
Rf: 0.41 (ph‡ase A), 0.38 (phase B)
‡
m/z [M 1] : 357
ESI-MS2
m/z (%): 191 (13), 192 (22), 205 (20), 206 (79),235 (63), 247 (8),
315 (100), 339 (41)
TLC
ESI-MS
Rf: 0.68 (ph‡ase A), 0.78 (phase B)
‡
m/z [M 1] : 283
ESI-MS2
m/z (%): 91 (14), 192 (17), 241 (100)
a For the spectroscopic data of bohemine and those of M1 to M3, see Chmela et al. (2001). TLC,
thin layer chromatography. For the mobile phase A (acidic) and phase B (alkaline), see the Materials and
methods. ESI, electrospray ionization.

1024
M. Rypka et al.
those of the metabolite revealed that the metabolite in spot M6 was 6-(4-hydro-
xybenzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine (see the Materials
and methods) (table 1). In detail, MSn data of 6-(4-hydroxybenzylamino)-2-(3-
hydroxypropylamino)-9-isopropylpurine were very alike to those of 6-(2-hydro-
xybenzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine. However, these
two could easily be di€erentiated from each other by their distinct TLC mobilities,
ˆ
the latter being chracterized by Rf’s 0.64 (phase A) and 0.75 (B). 6-(3-Hydroxy-
benzylamino)-2-(3-hydroxypropylamino)-9-isopropylpurine clearly di€ered from
these two isomers by both MSn and TLC characteristics (Rf 0.35 both in phases
ˆ
A and B).
Spot M8, representing about 5% of the total radioactivity in incubates, was
observed with a slightly greater mobility than that of parent bohemine (®gure 1).
Such migration suggested that a product in M8 was a less polar compound than
bohemine. Indeed, MS analysis indicated the presence of a metabolite localized in
‡
spot M8 and displaying m/z 283 [M 1] (table 1). An MSn scan suggested a
molecular structure of 2-amino-6-benzylamino-9-isopropylpurine, another prod-
uct of bohemine dealkylation. Again, comparison with an authentic, newly
synthesized reference compound corroborated this assumption (see the Materials
and methods) (table 1).
ˆ
‡
Identi®cation of a minor metabolite contained in minor spot M7 and repre-
senting 3% of the total radioactivity in the incubates (®gure 1) was hampered by
<
the small quantities extracted thus far. It was characterized by an MS peak of
‡
¹
ˆ
‡
Mr 356 (m/z 357 [M 1] ) suggesting attachment of a hydroxyl group to
bohemine. The molecular structure of this metabolite remains to be elucidated
(table 1).
Incubation of primary metabolites with formation of secondary metabolites
The primary metabolite spots M3±M6 and M8 were extracted and extracts
evaporated to dryness. The dry residues were then re-solved and re-incubated
with either mouse liver cytosol or microsomes in the presence of appropriate
coenzymes (NAD‡,‡NADPH) or UDP-glycosyl donors.
First, the NAD 0coenzyme-supplemented cytosolic fractions generated sec-
ondary products M4 ±M60 from the primary M4 to M6 extracts, respectively
(table 2). Each of the three secondary products M40±M60 could be subsequently
modi®ed by a reaction with diazomethane. Notably, diazomethane treatment of
M60 gave two derivatives with mutually di€erent Rf that indicated the presence
of two acidic hydroxyl gro0ups in M60 (table 2). These ®ndings suggested that
the secondary products M4 ±M60 could represent carboxylic acids, i.e. 6-(benzyl-
amino)-2-(2-carboxyethylamino)-9-purine, 6-amino-2-(2-carboxyethylamino)-9-
isopropylpurine and 6-(4-hydroxybenzylamino)-2-(2-carboxyethyl-amino)-9-iso-
propylpurine, derived from the primary metabolites M4±M6, respectively. MSn
analysis of M40±M60 further supported this deduction (table 2). No new spots
appeared when the cytosols were incubated with M3 or M8 extracts.
Second, M6 was converted into M5 by microsomes in the presence of
NADPH, as checked by TLC and MS characteristics of the latter (table 1). No
secondary products were detected when the primary metabolites M3±M5 and M8
were incubated with microsomes supplemented with NADPH-coenzyme.

Biotransformation of bohemine by mouse liver microsomes
1025
Table 2. Secondary products of NAD‡-dependent cytosolic transformation of bohemine primary
metabolites: Rf, methylation and spectroscopic data.
Compound
Analytical method
Data
Secondary
TLC
ESI-MS
Rf: 0.05 (ph‡ase A), 0.00 (phase B)
product M40
‡
m/z [M 1] : 313
ESI-MS2
m/z (%): 91 (4), 253 (91), 295 (100)
Rf: 0.16 (phase A), 0.44 (phase B)
TLC of an M40 methylester
Secondary
TLC
ESI-MS
Rf: 0.15 (ph‡ase A), 0.01 (phase B)
product M50
‡
m/z [M 1] : 265
ESI-MS2
m/z (%): 205 (100), 223 (17), 247 (20)
Rf: 0.48 (phase A), 0.53 (phase B)
TLC of an M50 methylester
Secondary
TLC
ESI-MS
Rf: 0.20 (ph‡ase A), 0.00 (phase B)
product M60
m/z [M 1] : 371
‡
ESI-MS2
m/z (%): 205 (8), 265 (100), 277 (16)
Rf: 0.55 (phase A), 0.52 (phase B)
Rf: 0.77 (phase A), 0.87 (phase B)
TLC of a presumed M600 methylester
TLC of a presumed M6 dimethylderivative
TLC, thin layer chromatography. For the mobile phase A (acidic) and phase B (alkaline), see the
Materials and methods. ESI, electrospray ionization.
Third, in the presence of UDP-glucoside, the microsomes readily produced a
new spot from the primary M6 spot extract. A structure of a -glucoside conjugate
-
with a glucoside moiety attached to the aliphatic hydroxyl of M6 was assigned to
this new product. This was based on the origin, sensitivity to -glucosidase,
-
insensitivity to -glucosidase, MS data of the assumed aliphatic M6 conjugate, and
¬
considering its ability to react with diazomethane that indicated the presence of a
free acidic hydroxyl group (presumably the aromatic one) in its molecule (table 3).
No new products were identi®ed when M3±M5 or M8 spot extracts were
incubated with UDP-glucoside-enriched microsomes.
Fourth, microsomes supplemented with UDP-glucuronic acid produced a new
metabolite spot from the primary M6 spot extract. This secondary metabolite of
M6 was tentatively identi®ed as M6 monoglucuronide based on its sensitivity to -
-
glucuronidase and MSn analysis (table 3). The fact that only one product was
detected after this secondary metabolite had been subject to methylation argues
for the occupancy of the aromatic hydroxyl on the M6 benzyl ring by the
Table 3. Secondary products of UDP-glucose- and UDP-glucuronic acid-dependent transformations
of primary bohemine M6: Rf, methylation and spectroscopic data.
Compound
Analytical method
Data
M6 ± UDP-glucose
reaction product
TLC
ESI-MS
Rf: 0.06 (ph‡ase A), 0.02 (phase B)
‡
m/z [M 1] : 519
ESI-MS2
m/z (%): 251 (8), 357 (9), 413 (100)
Rf: 0.16 (phase A), 0.09 (phase B)
TLC of a methylderivative of the
product
M6 ± UDP-
glucuronic acid
reaction product
TLC
ESI-MS
Rf: 0.00 (ph‡ase A), 0.00 (phase B)
‡
m/z [M 1] : 533
ESI-MS2
m/z (%): 209 (6), 251 (100), 357 (49)
Rf: 0.12 (phase A), 0.06 (phase B)
TLC of a methylderivative of the
product
TLC, thin layer chromatography. For the mobile phase A (acidic) and phase B (alkaline), see the
Materials and methods. ESI, electrospray ionization.

1026
M. Rypka et al.
diazomethane-reactive glucuronide moiety. No new metabolites were observed
when M3±M5 or M8 extracts were incubated with these microsomes.
To strengthen the main ®ndings on secondary metabolites described, we used
[8-3H]-labelled authentic standards of the primary metabolites (see the Materials
and methods) and incubated them with either cytosol or microsomes in the
presence of the above coenzymes and donors. These incubations gave the same
carboxylated and glycosylated secondary products as were found in incubates
employing the primary spot extracts, based on their TLC mobility and MS
analysis (data not shown; cf. tables 1±3). Surprisingly, we were unable to force
M8 to react in any catalytic system used throughout this work.
E€ects of selective inhibitors of cytochrome P450 forms
CYP form-selective probes were used, namely coumarin, ketoconazole, 8-
methoxypsoralen and troleandomycin. Experiments using these inhibitors at
M
m
concentrations between 0.2 and 250
under linear reaction conditions revealed
ketoconazole and 8-methoxypsoralen as potent agents in inhibiting bohemine
biotransformation in vitro (®gure 2). 8-Methoxypsoralen substantially inhibited
¶
(®gure 2). The appearance of M4 was also prevented
M
m
the formation of spot M4 at concentrations
5
and formation of spots M3 and
¶
M5 at concentrations 25
by coumarin at concentrations 50 . Ketoconazole and troleandomycin pre-
M
m
M
m
>
ferentially blocked the formation of spots M3 and M5 rather then that of M4,
¶
M
m
provided the inhibitors were present at concentrations of about
1
(®gure 2).
Figure 2. E€ects of CYP form-selective chemical inhibitors on bohemine transformation by mouse
liver microsomes in the presence of NADPH. Data, which were obtained by TLC resolution and
consequent 3H-autoradiographic/densitometric evaluation of the metabolite spots, are the means
from triplicate experiments. The amount of each metabolite was compared with that formed in
parallel control incubations conducted in the absence of inhibitors, and expressed as a per cent of
the corresponding control.

Biotransformation of bohemine by mouse liver microsomes
1027
Discussion
Chmela et al. (2001) reported that a cytosolic, NAD‡-dependent alcohol
dehydrogenase class I could catalyse transformation of bohemine into an acid
and that UDP-sugar-dependent enzymatic mechanisms were responsible for
bohemine glucoside and glucuronide formation in mice in vivo. They were unable
to observe any speci®c individual products indicating participation of the micro-
somal, NADPH-dependent CYP system in bohemine clearance from the mouse.
However, in view of the well-known prominent expression of CYPs in animal
tissues, further experimentation was warranted.
The data presented here show that bohemine has been changed into its
metabolites during a 120-min incubation of the parent compound in mouse liver
microsomes supplemented with NADPH. The proposed primary NADPH-
dependent reactions of bohemine in this model have included oxidation (spot
M3), hydroxylation (spot M6), two types of dealkylation (spots M4 and M8) and
debenzylation (spot M5) (®gure 3). Importantly, not only a cytosolic alcohol
dehydrogenase, as found by Chmela et al. (2001), but also liver microsomes
enriched in NADPH, as shown here, have been found to be capable of producing
bohemine carboxyl acid, elucidated as a spot M3 metabolite.
Additionally, we have identi®ed several potential secondary pathways mediat-
ing primary metabolite biotransformations. First, all the primary metabolites that
retained their aliphatic h‡ydroxyl group at the C2 side chain could be oxidized by
liver cytosol in an NAD -dependent manner to their corresponding acids. This
®ts well with our previous results showing a substantial role of cytosolic alcohol
dehydrogenase in mediating bohemine metabolism (Chmela et al. 2001). Second,
primary metabolite transformations explored by us have suggested that the spot
M6 metabolite is exceptional owing to its predisposition to be modi®ed by both
Figure 3. Proposed bohemine primary biotransformation pathways.

1028
M. Rypka et al.
type I and type II reactions. Particularly, this primary metabolite could be
conjugated with UDP-glucoside. In this context, it is noteworthy that our current
®ndings show that bohemine glucosidation is species speci®c, in that microsomes
from rat, monkey and/or human origin do not support this reaction (Cervenkova
et al. 2002). Furthermore, the fact that M6 can serve as a precursor for product M5
is of particular importance since it makes M5 as the primary metabolite uncertain
(®gure 3).
Molecular structures of both the primary and secondary metabolites and,
subsequently, a scheme of bohemine transformation pathways have been proposed
based on several types of evidence: (1) TLC mobile p‡hase-dependent monitoring;
(2) enzymatic hydrolysis/oxidation (including NAD -dependent oxidation); (3)
chemical modi®cation (methylation) that can help unmask the presence of an acidic
hydroxyl, such as a carboxylic or aromatic one; and (4) MS scanning. Moreover,
use of authentic reference compounds has further substantiated our ®ndings
concerning primary metabolites M3±M6 and those characterizing M8. We can
refer back to known facts concerning bohemine metabolites M2 (the glucoside) and
M3 (the acid) that have molecular characteristics established by NMR (Chmela
et al. 2001). Nevertheless, as our new structural data have not yet been ultimately
validated by a direct molecular analysis, we feel obliged to declare here their
tentative character.
Several strategies may be followed to gain an insight into which CYPs are
responsible for metabolizing bohemine. Numerous investigators have substan-
tiated an inhibition-based approach using CYP-selective probes (Newton et al.
1995, Hickman et al. 1998). In mouse tissues, 8-methoxypsoralen has been used by
Pelkonen et al. (1997), Gu et al. (1998) and Patten et al. (1998) to inhibit mouse
CYP2a form(s). Furthermore, ketoconazole and troleandomycin, human
CYP3A4-selective inhibitors, are known to inhibit mouse CYP3a form(s) (Galla-
gher et al. 1994, Bourrie et al. 1996, Doostzadeh et al. 1998a, b, Eagling et al. 1998,
Boogaards et al. 2000, Wilson and LeBlanc 2000, Honda et al. 2001, Quintieri et al.
2001). In the present study, we inhibited NADPH-dependent bohemine degrada-
tion by varying concentrations of chemical inhibitors including 8-methoxypsoralen
(CYP2a), ketoconazole, troleandomycin (CYP3a) and coumarin, a CYP2A6/2a5
substrate (Boogaards et al. 2000, Juvonen et al. 2000).
We have observed notable inhibitory e€ects of 8-methoxypsoralen on M4
M
m
formation. This strong inhibitory e€ect at low concentrations (5 ) of 8-methoxy-
psoralen could reasonably be explained by a substantial role of a CYP2a form in
NADPH-dependent bohemine transformation in vitro. A suppressive e€ect of
M
m
selectivity of this inhibitor at high concentrations. Ketoconazole and troleando-
25
8-methoxypsoralen on M3 and M5 accumulation probably indicates a loss of
mm
mycin (at concentrations about 1 ) each blocked M5 and, partly, M3 formation.
These e€ects presumably suggest a contribution of a ketoconazole/troleandomy-
cin-sensitive CYP3a form in bohemine degradation. Thus, at least two mice
CYPs, namely CYP2a and CYP3a, could participate in bohemine transformation.
It can be speculated that CYP2a mainly catalysed deisopropylation. By contrast,
CYP3a seemed to be pluripotent and capable in mediating several pathways, e.g.
bohemine debenzylation and oxidation. Naturally, these ®ndings do not exclude a
possible participation of other CYPs in bohemine metabolism.
Metabolic information obtained from microsomes and cytosols concerns only
metabolism at the subcellular level. In spite of the ability of CYP systems to

Biotransformation of bohemine by mouse liver microsomes
1029
metabolize bohemine extensively in vitro, participation of the microsomal CYP
systems in bohemine metabolism in vivo may be of a secondary importance, at least
in mouse. This follows from the results obtained by Chmela et al. (2001).
Recently, we obtained data indicating that metabolic pathways contributing to
the bohemine transformation also mediate metabolism of other olomoucine-type
derivatives such as olomoucine, roscovitine and some others (unpublished data).
Hopefully, the present ®ndings will facilitate future experiments designed to
dissect enzymatic systems contributing to the clearance of C2,C6,N9-trisubsti-
tuted purine compounds from mammalian tissues.
Acknowledgements
The work was supported, in part, by the Ministry of Education, Youth and
Sports of the Czech Republic (MSM 153110001, COST B5/B17) and by the Grant
Agency of the Czech Republic (301/02/0475/A). The authors thank H. Rozsypa-
lova, S. Snaselova and V. Langerova for excellent technical assistance. They also
thank P. Sedmera, M. Kuzma and V. Havlõcek, Institute of Microbiology,
Academy of Sciences, Czech Republic, for NMR measurements and valuable
contribution to the interpretation of MS spectra. Last, but not least, they are
indebted to Alexander Oulton and Jarmila Potomkova for editorial assistance.
References
A
S
S
Belejov , M., Chmela, Z., Rypka, M., Lemr, K., Luke , J., Hanu , J., Ulrichov , J.
A
Y
Vesel , J.,
and
1999, Metabolism of synthetic CDK-inhibitors in mammalian cells. Paper presented at the 15th
Biological Days, 6±8 September, Ceske Budejovice, Czech Republic; repr. J. Berger (ed.), Cells
(Ceske Budejovice: KOOP), p. 39.
Boogaards, J. J. P., Bertrand, M., Jackson, P., Oudshoorn, M. J., Weaver, R. J., Van Bladeren, P. J.
Walther, B.,
and 2000, Determining the best animal model for human cytochrome P450
activities: a comparison of mouse, rat, rabbit, dog, micropig, monkey and man. Xenobiotica, 30,
1131±1152.
E
Bourri , M., Meunier, V., Berger, Y.
Fabre, G.,
1996, Cytochrome P450 isoform inhibitors as a
and
tool for the investigation of metabolic reactions catalyzed by human liver microsomes. Journal of
Pharmacology and Experimental Therapeutics, 277, 321±332.
Buolamwini, J. K.,
2000, Cell cycle molecular targets in novel anticancer drug discovery. Current
Pharmaceutical Design, 6, 379±392.
A
A
A
A
I
A
ervenkov , K., Belejov , M., Chmela, Z., Rypka, M., Riegrov , D., Michnov , K., Michal kov ,
C
U
A
S
Y
A
IC
Y
K., rov , I., Brejcha, A., Hanu , J., ern , B., Fuksov , K., Havl ek, L., Vesel , J
., 2002,
S
In vitro glycosidation potential towards olomoucine-type cyclin-dependent kinase inhibitors in
C
rodent and primate microsomes. Physiological Research, accepted for publication.
A
S
IC
A
A
ervenkov , K., Chmela, Z., Rypka, M., Lemr, K., Hanu , J., Havl ek, L., Belejov , M., Lotkov ,
C
A
H., ervinkov , Z.
Y
Vesel , J.,
2000, Metabolism of cyclin-dependent kinase inhibitor
C
and
bohemine at the cellular and subcellular levels. Paper presented at the 2nd FEPS Congress, 29
June±4 July 1999, Prague, Czech Republic; abst. Physiological Research, 49, P42.
Chang, Y. T., Gray, N. S., Rosania, G. R., Sutherlin, D. P., Kwon, S., Norman, T. C., Sarohia, R.,
Leost, M., Meijer, L. Schultz, P. G.,
and 1999, Synthesis and application of functionally
diverse 2,6,9-trisubstituted purine libraries as CDK inhibitors. Chemistry and Biology, 6, 361±
375.
Chauret, N., Gauthier, A., Martin, J.
Nicoll Griffith, D. A.,
- 1997, In vitro comparison of
and
cytochrome P450-mediated metabolic activities in human, dog, cat, and horse. Drug Metabolism
and Disposition, 25, 1130±1136.
Y
IC
S
A
Chmela, Z., Vesel , J., Lemr, K., Hanu, J., Havl ek, L., Rypka, M., Kry tof, V., Ulrichov , J.,
A
Walterov , D., Fuksov , K., Luke , J., Kol ov , H., Strnad, M.
A
S
AR
A
Lenobel, R.,
1998,
and
Metabolism of olomoucine, roscovitine, and bohemine in small laboratory rodents. Paper
presented at the 1st International Conference on Inhibitors of Protein Kinases, 15±20
September, Warsaw, Poland; abst. Cellular and Molecular Biology Letters, 3, 280.

1030
M. Rypka et al.
Y
Chmela, Z., Vesel , J., Lemr, K., Rypka, M., Hanu , J., Havl ek, L., Kry tof, V., Michnov , L.,
S
IC
S
A
A
Fuksov , K.
S
Luke , J.,
2001, In vivo metabolism of 2,6,9-trisubstituted purine-derived
and
cyclin-dependent kinase inhibitor bohemine in male mice: glucosidation as the principal
metabolic route. Drug Metabolism and Disposition, 29, 326±334.
Court, M. H., Von Moltke, L. L., Shader, R. I.
Greenblatt, D. J.,
1997, Biotransformation of
chlorzoxazone by hepatic microsomes from humans and ten other mammalian species.
and
Biopharmaceutics and Drug Disposition, 18, 213±226.
E
Doostzadeh, J., Cotillon, A. C., Benalych rif, A.
Morfin, R.,
1998a, Inhibition studies of
dehydroepiandrosterone 7 - and 7 hydroxylation in mouse liver microsomes. Steroids, 63, 608±
-
and
¬
-
614.
Doostzadeh, J., Cotillon, A. C.
alpha- and beta-positions by mouse liver microsomes. E€ects of cytochrome p450 inhibitors and
Morfin, R.,
1998b, Hydroxylation of pregnenolone at the 7
-
and
structure-speci®c inhibition by steroid hormones. Steroids, 63, 383±392.
and
Eagling, V. A., Tjia, J. F.
Back, D. J.,
1998, Di€erential selectivity of cytochrome P450 inhibitors
against probe substrates in human and rat liver microsomes. British Journal of Clinical
Pharmacology, 45, 107±114.
Epstein, J., Plapinger, R. E., Michel, H. O., Cable, J. R., Stephani, R. A., Hester, R. J.,
Billington. C.,
List, G. R.,
1964, Reactions of isopropyl methylphosphono¯uoridate
Jr and
with substituted phenols. I. Journal of the American Chemical Society, 86, 3075±3084.
Fischer, P. M. Lane, D. P.,
and
therapeutics. Current Medicinal Chemistry, 7, 1213±1245.
Gallagher, E. P., Wienkers, L. C., Stapleton, P. L., Kunze, K. L.
2000, Inhibitors of cyclin-dependent kinases as anti-cancer
Eaton, D. L.,
and
1994, Role of
human microsomal and human complementary DNA-expressed cytochromes P4501A2 and
P4503A4 in the bioactivation of a¯atoxin B1. Cancer Research, 54, 101±108.
I
Garc a Echeverr a, C., Traxler, P.
I
Evans, D. B.,
2000, ATP site-directed competitive and
-
and
irreversible inhibitors of protein kinases. Medical Research Reviews, 20, 28±57.
Gray, A. S., Wodicka, L., Thunnissen, A. M. W. H., Norman, T. C., Kwon, S., Espinoza, F. H.,
-
Morgan, D. O., Barnes, G., Leclerc, S., Meijer, L., Kim, S. H., Lockhart, D. J.
Schultz,
-
and
1998, Exploiting chemical libraries, structure, and genomics in the search for kinase
inhibitors. Science, 281, 533±538.
Gray, N., Detivaud, L. Meijer, L.,
P. G.,
and
kinases. Current Medicinal Chemistry, 6, 859±875.
Gu, J., Zhang, Q. Y., Genter, M. B., Lipinskas, T. W., Negishi, M., Nebert, D. W.
1999, ATP-site directed inhibitors of cyclin-dependent
Ding, X.,
and
1998, Puri®cation and characterization of heterogously expressed mouse CYP2A5 and CYP2G1:
role in metabolic activation of acetaminophen and 2,6-dichlorobenzonitrile in mouse olfactory
mucosal microsomes. Journal of Pharmacology and Experimental Therapeutics, 285, 1287±1295.
IC
Havl ek, L., Hanu , J., Vesel , J., Leclerc, S., Meijer, L., Shaw, G.
S
Y
Strnad, M.,
1997,
Cytokinin-derived cyclin dependent kinase inhibitors: synthesis and cdc2 activity of olomoucine
and
and related compounds. Journal of Medicinal Chemistry, 40, 408±412.
Hickman, D., Wang, J. P., Wang, Y.
probes and suitability of organic solvents for the measurement of human cytochrome P450
Unadkat, J. D.,
1998, Evaluation of the selectivity of in vitro
and
monoxygenase activities. Drug Metabolism and Disposition, 26, 207±215.
Hocart, C. H., Letham, D. S.
Parker, C. W.,
1991, Substituted xanthines and cytokinin analogues
and
as inhibitors of cytokinin N-glucosylation. Phytochemistry, 30, 2477±2486.
Honda, A., Salen, G., Matsuzaki, Y., Batta, A. K., Xu, G., Leitersdorf, E., Tint, G. S., Erickson,
S. K., Tanaka, N. Shefer, S.,
and
2001, Side chain hydroxylations in bile acid biosynthesis
catalyzed by CYP3A are markedly up-regulated in Cyp27±/± mice but not in cerebrotendinous
xanthomatosis. Journal of Biological Chemistry, 276, 34579±34585.
Jones, E. C. S.
Pyman, F. L.,
1925, The relation between chemical constitution and pungency in
and
acid amides. Journal of the Chemical Society, 127, 2588±2598.
Juvonen, R. O., Gynther, J., Pasanen, M., Alhava, E. Poso, A.,
and
2000, Pronounced di€erences in
inhibition potency of lactone and non-lactone compounds for mouse and human coumarin 7-
hydroxylases (CYP2A5 and CYP2A6). Xenobiotica, 30, 81±92.
Legraverend, M., Ludwig, O., Bisagni, E., Leclerc, S., Meijer, L., Giocanti, N., Sadri, R.
Favaudon, V.,
1999, Synthesis and in vitro evaluation of novel 2,6,9-trisubstituted purines acting
and
as cyclin-dependent kinase inhibitors. Bioorganic and Medicinal Chemistry, 7, 1281±1293.
Meijer, L.,
2000, Cyclin-dependent kinases inhibitors as potential anticancer, antineurodegenerative,
antiviral and antiparasitic agents. Drug Resistance Updates, 3, 83±88.
Meijer, L., Leclerc, S.
Leost, M.,
1999, Properties and potential applications of chemical
inhibitors of cyclin-dependent kinases. Pharmacology and Therapeutics, 82, 279±284.
and
I
Michal kov , K., Michnov , L., Belejov , M., Chmela, Z., Rypka, M., Lemr, K., Havl k, L.,
A
A
A
IC
S
Luke , J., Vesel , J.
Y
A
Ulrichov , J.,
1999, Enzyme systems involved in metabolism of
and
purine-related cyclin-dependent kinase inhibitors. Paper presented at the 15th Biological Days,
6±8 September, Ceske Budejovice, Czech Republic; abst. J. Berger (ed.), Cells (Ceske
Budejovice: KOOP), p. 97.

Biotransformation of bohemine by mouse liver microsomes
1031
Newton, D. J., Wang, R. W.
speci®ties in the in vitro metabolism of therapeutic agents by human liver microsomes. Drug
Lu, A. Y. H.,
1995, Cytochrome P450 inhibitors. Evaluation of
and
Metabolism and Disposition, 23, 154±158.
S
Otyepka, M., Kry tof, V., Havl ek, L., Iglerov , V., Strnad, M.
IC
A
C
Ko a, J.,
2000, Docking-
and
based development of purine-like inhibitors of cyclin-dependent kinase-2. Journal of Medicinal
Chemistry, 43, 2506±2513.
Patten, C. J., Peterson, L. A.
Murphy, S. E.,
1998, Evidence for metabolic activation of N-
and
nitrosonornicotine and N-nitrosobenzyl methylamine by rat nasal coumarin hydroxylase. Drug
Metabolism and Disposition, 26, 177±180.
Pelkonen, O., Maenpaa, J., Taavitsainen, P., Rautio, A.
Raunio, H.,
1998, Inhibition and
and
induction of human cytochrome P450 (CYP) enzymes. Xenobiotica, 28, 1203±1253.
Quintieri, L., Rosato, A., Napoli, E., Sola, F., Geroni, C., Floreani, M. Zanvello, P.,
and
2000, In
vivo antitumor activity and host toxicity of methoxymorpholinyl doxorubicin: role of cytochrome
P450 3A. Cancer Research, 60, 3232±3238.
Raiford, L. C.
Clark, E. P.,
1923, Diacyl derivatives of ortho-hydroxybenzylamine. Journal of the
and
American Chemical Society, 45, 1738±1743.
Rendic, S.,
2002, Summary of information on human CYP enzymes: human P450 metabolism data.
Drug Metabolism Reviews, 34, 83±448.
Rosania, G. R.
Chang, Y. T.,
- 2000, Targeting hyperproliferative disorders with cyclin-dependent
and
kinase inhibitors. Expert Opinion on Therapeutic Patents, 10, 215±230.
Sausville, E. A., Zaharevitz, D., Gussio, R., Meijer, L., Louarn Leost, M., Kunick, C., Schultz,
-
R., Lahusen, T., Headlee, D., Stinson, S., Arbuck, S. G.
Senderovicz, A.,
1999, Cyclin-
and
dependent kinases: initial approaches to exploit a novel therapeutic target. Pharmacology and
Therapeutics, 82, 285±292.
Sielecki, T. M., Boylan, J. F., Beneld, P. A.
Trainor, G. L.,
2000, Cyclin-dependent kinase
and
inhibitors: useful targets in cell cycle regulation. Journal of Medicinal Chemistry, 43, 1±18.
Venkatakrishnan, K., Von Moltke, L. L. Greenblatt, D. J.,
and
2001. Human drug metabolism and
the cytochromes P450: application and relevance of in vitro models. Journal of Clinical
Pharmacology, 41, 1149±1179.
Y
Vesel , J., Havl ek, L., Strnad, M., Blow, J. J., Donella Deana, A., Pinna, L., Letham, D. S.,
IC
-
Kato, J., Detivaud, L., Leclerc, S.
Meijer, L.,
1994, Inhibition of cyclin-dependent
and
kinases by purine analogues. European Journal of Biochemistry, 224, 771±786.
Wilson, W. S. LeBlanc, G. A.,
and
2000, The contribution of hepatic inactivation of testosterone to
the lowering of serum testosterone levels by ketoconazole. Toxicological Sciences, 54, 128±137.