Novel 4,1-benzoxazepine derivatives with potent squalene synthase inhibitory activities

Article abstract of DOI:10.1016/S0968-0896(01)00290-5

A series of (3,5-trans)-2-oxo-5-phenyl-1,2,3,5-tetrahydro-4,1-benzoxazepine derivatives were synthesized and evaluated for squalene synthase inhibitory and cholesterol biosynthesis inhibitory activities. Through modification of substituents of the lead compounds 1a and 1b, it was found that 4,1-benzoxazepine-3-acetic acid derivatives with isobutyl and neopentyl groups at the 1-position, the chloro atom at the 7-position, and the chloro and methoxy groups at the 2'-position on the 5-phenyl ring, had potent squalene synthase inhibitory activity. Among such compounds, the 5-(2,3-dimethoxyphenyl) derivative 2t exhibited potent inhibition of cholesterol biosynthesis in HepG2 cells. As a result of optical resolution study of 2t, the absolute stereochemistry required for inibitory activity was determined to be 3R,5S. In vivo study showed that the sodium salt of (3R,5S)-7-chloro-5-(2,3-dimethoxyphenyl)-1-neopentyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-aceitc acid 20 effectively reduced plasma cholesterol in marmosets. Copyright

Full text of DOI:10.1016/S0968-0896(01)00290-5

Bioorganic & Medicinal Chemistry 10 (2002) 401–414
Novel 4,1-Benzoxazepine Derivatives with Potent Squalene
Synthase Inhibitory Activities
Takashi Miki,* Masakuni Kori, Hiroshi Mabuchi, Hiroshi Banno, Ryu-ichi Tozawa,
Masahira Nakamura, Shigekazu Itokawa, Yasuo Sugiyama andHidefumi Yukimasa
Takeda Chemical Industries, Ltd. Pharmaceutical Research Division, 2-17-85, Juso-Honmachi, Yodogawa-ku, Osaka 532-8686, Japan
Received8 June 2001; accepted11 August 2001
Abstract—A series of (3,5-trans)-2-oxo-5-phenyl-1,2,3,5-tetrahydro-4,1-benzoxazepine derivatives were synthesized and evaluated
for squalene synthase inhibitory andcholesterol biosynthesis inhibitory activities. Through modification of substituents of the lead
compounds 1a and 1b, it was foundthat 4,1-benzoxazepine-3-acetic acidderivatives with isobutyl andneopentyl groups at the 1-
position, the chloro atom at the 7-position, andthe chloro andmethoxy groups at the 2 0-position on the 5-phenyl ring, hadpotent
squalene synthase inhibitory activity. Among such compounds, the 5-(2,3-dimethoxyphenyl) derivative 2t exhibitedpotent inhibi-
tion of cholesterol biosynthesis in HepG2 cells. As a result of optical resolution study of 2t, the absolute stereochemistry required
for inhibitory activity was determined to be 3R,5S. In vivo study showed that the sodium salt of (3R,5S)-7-chloro-5-(2,3-dime-
thoxyphenyl)-1-neopentyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-acetic acid 20 effectively reduced plasma cholesterol in
marmosets. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
acetic acid 20, the oral administration of which lowered
plasma cholesterol in marmosets.
Squalene synthase [EC 2.5.1.21] catalyzes the formation
of squalene from farnesyl diphosphate in cholesterol
biosynthesis. This enzymatic step occurs after the path-
way branches to other isoprene derivatives such as
dolicol, ubiquinones and isopentenyl t-RNA. Since
squalene synthase inhibitors do not interfere with the
biosynthesis of these isoprene derivatives, inhibition of
this step arrests only cholesterol biosynthesis andmight
be useful for the treatment of hyperlipidemia.
Chemistry
The syntheses of the 4,1-benzoxazepine-3-acetic acid
derivatives 2 listedin Tables 1 and3 are shown in
Scheme 1.1 Reduction of 2-aminobenzophenones 4 with
sodium borohydride (NaBH4) yielded aminoalcohols 5,
which were treated with aldehydes or ketones and
NaBH4 or sodium cyano borohydride (NaBH3CN) to
We previously reported1 that a novel class of squalene
synthase inhibitors, 4,1-benzoxazepine-3-acetic acid
derivatives 1a,b (Fig. 1), exhibitedpotent inhibition of
rat enzyme (IC50=0.061–0.072 mM) andHepG2 enzyme
(IC50=0.024–0.034 mM).
affordthe alkylatedcompounds
were also synthesizedby acylation of 5 with acidchlo-
ride, followed by reduction of the amides 11 with
6. The compounds 6
In this paper, we describe our synthetic studies of var-
ious 4,1-benzoxazepine derivatives,2 which ledto the
discovery of a potent squalene synthase inhibitor, the
sod ium salt of (3R,5S)-7-chloro-5-(2,3-dimethoxyphenyl)-
1-neopentyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-
*Corresponding author. Tel. +81-6-6300-6637; fax:+81-6-6300-6306;
e-mail: Miki_Takashi@takeda.co.jp
Figure 1. Structures of 4,1-benzoxazepine-3-acetic acidderivatives.
0968-0896/02/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00290-5

402
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
Table 1. Physicochemical andbiological properties of 1-substituted(3,5- trans)-7-chloro-5-(2-chlorophenyl)-2-oxo-1,2,3,5-tetrahydro-4,1-benzox-
azepine-3-acetic acidderivatives 1a,b and 2a–l
Mp (ꢀC)
Formulab
IC50 (mM)c
(rat enzyme)
a
CompdR
Yield
(%)
2a
2b
2c
2d
2e
2f
2g
2h
2i
CH2Ph
Me
Et
Pr
Bu
Pr
63
45
75
59
79
61
76
80
63
95
48
63
241–242
211–213
215–216
189–190
195–196
208–209
188–189
230–232
241–242
220–221
166–167
176–177
C24H25Cl2NO4
C18H15Cl2NO4
C19H17Cl2NO4
C20H19Cl2NO4
C21H21Cl2NO4
C20H19Cl2NO4
C23H25Cl2NO4
C21H19Cl2NO4
C24H25Cl2NO4
C22H23Cl2NO4
C22H23Cl2NO4
C23H25Cl2NO4
8.3
6.4
0.53
0.14
2.2
0.71
0.79
0.21
CH2CHEt2
CH2–cyclopropyl
CH2–cyclohexyl
CHEt2
CH2CH2CHMe2
CH2CH2CMe3
CH2CMe3
Bui
6.9
0.92
2.3
2j
2k
2l
1a1
1b1
>10 (25.6)d
0.072
0.061
aYieldof final step.
bAnalysis for C, H, N were correct within Æ0.4%.
cIC50 values were determined by a single experiment run in duplicate.
d%Inhibition at 10 mM.
lithium aluminum hydride (LiAlH4). The N-methyl and
N-ethyl analogues 6b,c were preparedby alkylation of
the diacetylated compound 10a obtainedfrom 5a, fol-
lowedby alkaline hyrdolysis of the alkylatedcom-
pounds 10b,c. After condensation of 6 andfumaric acid
chloride monoethyl ester 7, intramolecular Michael
addition of the obtained amides 8 afforded 4,1-benzox-
azepine-3-acetates 9. In this reaction, thermo-
dynamically stable 3,5-trans-isomers were obtained,
except with 2m, which was obtainedas a mixture of
trans/cis isomers (1:1). Hydrolysis of the esters 9 gave
the carboxylic acids 2.
Next, we focusedon the synthesis of 4,1-benzoxazepine
derivatives having various substituents at the 3-position
of leadcompound 1a (Scheme 2). The ethyl ester 1c has
1
been reportedin the previous paper. Modification of
the tether length between the carboxyl moiety andthe
Scheme 1. Reagents andconditions: (a) NaBH 4; (b) aldehyde, ketone,
NaBH4 or NaBH3CN, AcOH; (c) fumalic acidchloride monoethyl
ester 7, NaHCO3, CH2Cl2; (d ) KCO3, EtOH; (e) K2CO3, MeOH–
2
H2O; (f) Ac2O, DMAP, pyridine; (g) MeI, NaH, DMF; (h) EtI, NaH,
DMF; (i) NaOH, EtOH; (j) R’COCl; (k) LiAlH4.
Scheme 2. Reagents andconidtions: (a) NaOH, EtOH; (b) (1)
(PhO)2P(O)N3; (2) benzene, reflux; (3) concdHCl; (c) NaNO 2, AcOH–
H2O; (d) Jones’ oxidation; (e) (1) ClCOOEt, N-methylmorpholine; (2)
NaBH4; (f) SOCl2, pyridine; (g) NaCN; (h) HCl, EtOH; (i) K2CO3,
MeOH–H2O; (j) NH4Cl, DEPC, Et3N, DMF; (k) Swern’s oxidation.

T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
403
benzoxazepine ring was carriedout as follows. Curtius
were measuredaccording to the methodof Cohen et al.
rearrangement of the 3-acetic acidderivatives 1a using
diphenylphosphorylazide (DPPA), and subsequent acid
hydrolysis of the resulting isocyanate gave the 3-amino-
methyl analogue 12.3 The diazotation of 12 with sodium
nitrite (NaNO2) gave the 3-hydroxymethyl analogue 13
which was convertedto the 3-carboxylic acidderivative
3a by Jones’ oxidation. The treatment of the 3-acetic
acidderivative 1a with ethyl chloroformate, followedby
reduction with NaBH4 gave the 3-(2-hydroxyethyl)
analogue 14.4 Compound 14 was convertedto the 3-(2-
chloroethyl) analogue 15 which was reactedwith
sodium cyanide to provide the 3-propionitrile analogue
16. Acidhydrolysis of 16 afforded the ethyl ester 17, and
subsequent alkaline hydrolysis gave the 3-propionic acid
3b. The amide derivative 3c were obtainedby con-
densation of 1a with ammonium chloride using diethyl
cyanophosphonate (DEPC). Compound 14 was con-
verted to the aldehyde 3d by Swern’s oxidation.5
with slight modification.6 The selectedcompounds were
evaluatedfor inhibition of cholesterol biosynthesis in
HepG2 cells according to the method of Kuroda et al.7
Squalene synthase inhibitory activity (in vitro)
Modification at the 1-position had a great influence on
potency of inhibition of rat derived squalene synthase
(Table 1). The 1-benzyl compound 2a showedonly
weak activity. In the series of N-alkyl derivatives
(2b,c,d,e), n-propyl compound 2d was the most potent
inhibitor of squalene synthase, with an IC50 value of
0.14 mM, indicating that the optimal carbon chain
number is three. The derivatives with b-branchedalkyl
groups such as neopentyl (1a), isobutyl (1b) and2-
ethylbutyl (2g) exhibitedmuch more potent activities
than n-alkylatedcompounsd
2d and 2e, while the
cyclopropylmethyl derivative 2h did not exhibit
improved activity. On the other hand, compounds hav-
ing a- and g-branchedalkyl substituents ( 2f, 2j and 2k,
2l) did not exhibit improved activity, probably due to
the bulkiness of these groups. Unsurprisingly, the
cyclohexylmethyl compound 2i was not a goodinhi-
bitor. The most favorable groups at the 1-position were
thus foundto be neopentyl ( 1a) andisobutyl ( 1b).
To determine which enantiomer is an active inhibitor,
we performedoptical resolution of the compound 2t,
which exhibitedthe most potent inhibiting activity. We
initially investigatedthe separation of the idaster-
eomeric amide which could be prepared by condensa-
tion of racemate 2t with various chiral amines and a-
aminoacidesters. After several trials, we foundthat the
diastereomeric amides of the acid 2t condensed with
methyl l-leucinate couldeasily be separatedby silica gel
column chromatography to give the less polar amide
18a andthe polar amide 18b in diastereomerically pure
forms. Then the amides 18a and 18b were hydrolyzed to
yieldenantiomerically pure acid 19a andacid 19b,
respectively (Scheme 3). Determination of the absolute
stereochemistry was performedby X-ray idffraction
analysis of the compound 19a. As depicted in Figure 2,
the absolute configuration of 19a (18a) was determined
to be (3R,5S). Therefore, 19b (18b) was (3S,5R).
Modification of the 3-acetic acid moiety of compound
1a gave the following results (Table 2). Elongation and
shortening of the tether length (3a,b) resultedin 40- to
50-foldedcrease in activity comparedwith
1a. The
activities of the ethyl ester analogue 1c1 andthe car-
boxamide analogue 3c were significantly weaker than
that of the corresponding 3-acetic acid analogues 1a,
but retainedIC values at the 10À6–10À7 M level. The
50
3-acetaldehyde analogue 3d exhibitedonly weak activ-
ity, with 39% inhibition at 10À5 M. The above results
indicate that a carboxymethyl group is the best sub-
stituent at the 3-position.
Migration of the 7-chloro atom in the fusedbenzene
ring to the 6- and8-positions yieldedcompounds 2m (a
mixture of 3,5-cis and-trans isomers) and 2n, which had
Biological Results and Discussion
The compounds synthesized were evaluated for inhibi-
tion of squalene synthase preparedfrom rat liver and
human hepatoma (HepG2) cells. Inhibitory activities
poor activities (Table 3). For the chloro atom, the 7-
8
position was foundto be best.
Scheme 3. Reagents andconditions: (a) l-Leu-OMe, DEPC, Et3N then separation; (b) (1) HCl, MeOH; (2) NaOH; (c) NaOH.

404
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
human HepG2 enzymes, while the (3S,5R)-enantiomer
19b exhibitedmodest inhibition (Table 4). Thus the ste-
reochemistry at C(3) andC(5) on the benzoxazepine
skeleton is of crucial importance in obtaining potent
inhibitory activity.
The IC50 values of the sodium salt 20 for human enzyme
9
andrat enzyme were 7.7 and16 nM, respectively.
Lineweaver–Burk analysis of HepG2 enzyme inhibition
revealedthat compound 20 was a competitive inhibitor
with respect to farnesyl pyrophosphate. The Ki value of
compound 20 was 1.1 nM andwas inedpenednt of
enzyme concentration.10
Figure 2. Stereoscopic molecular view of compound 19a.
Biological data for compounds having various substituents
on the 5-phenyl ring are also shown in Table 3. Compound
2o with no substituent at the 20-position was weaker than
the 20-chloro analogue 1a. Replacement of a chloro atom
at the 20-position with a bromo and a methoxy group led to
compounds 2p,q, which exhibitedpotent activity. In order
to improve the activity of the 20-methoxy analogue 2q,
additional substituents such as fluoro, methyl and meth-
oxy groups were introduced. Introduction of a fluoro atom
or a methoxy group at the 50- or 60-position causeda loss of
potency (2s,v,w), while introduction of a methyl group at
the 30-position resultedin a slight decrease of activity ( 2y).
On the other hand, introduction of a methoxy group at the
HepG2 cell assay
Since cholesterol is synthesizedintracellularly, goodcell
membrane permeability is necessary for squalene syn-
thase inhibitors. Selected1-neopentyl compounds ( 1a
and 2q,t,u) which exhibitedpotent inhibitory activity at
enzyme level (IC50=18–24 nM) were assayedfor their
ability to inhibit cholesterol synthesis in HepG2 cells
(Table 5). The 20-chloro compound 1a exhibitedonly
moderate inhibitory activity, with an IC50 value of
3.2 mM, despite potent activity (IC50=24 nM) at the
enzyme level. The two-order difference in potency
between the enzyme andcell assay might be due to cell
permeability. Replacement of the chloro atom in com-
pound 1a with a methoxy group to yieldcompound 2q
resultedin approximately 3-foldincrease in activity.
Furthermore, introduction of an additional methoxy
group into the 30- or 40-position gave rise to the much
more potent inhibitors 2t,u, with IC50 values of 0.12 and
0.40 mM, respectively. Compound 2t was thus foundto
be the most potent inhibitor at cell andenzyme levels.
0
30- and 40-positions (2t,u) andof a fluoro atom at the 4 -
position (2r) yielded the same level of potency as the 20-
methoxy analogue 2q. The 20,40,60-trimethoxy analogue
2x was less active than the 20,40-dimethoxy analogue 2u
but more active than the 20,60-dimethoxy analogue 2w.
A fluoro atom anda methoxy group are thus favorable
0
as additional substituents at the 30- and4 -positions of
the 20-methoxy analogue 2q.
In order to clarify the absolute stereochemistry required
for inhibitory activity, optical resolution of the com-
pound 2t was performed. The (3R,5S)-enantiomer 19a
exhibitedremarkably potent inhibition of both rat and
In vivo activity
In vivo activities in rats andmarmosets were studied
using the sodium salt 20, because of its goodoral bio-
Table 2. Physicochemical andbiological properties of (3,5- trans)-7-chloro-5-(2-chlorophenyl)-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine deriva-
tives 1c and 3a–d
Mp (ꢀC)
Formulab
IC50 (mM)c
(rat enzyme)
a
CompdR
Yield
(%)
3a
3b
3c
3d
1c1
COOH
48
94
65
64
166–167
225–227
291–292
173–176
C21H21Cl2NO4
C23H25Cl2NO4
C22H24Cl2N2O3
C22H23Cl2NO3
3.7
3.1
0.20
>10 (39.3)d
3.8
CH2CH2COOH
CH2CONH2
CH2CHO
CH2COOEt
aYieldof final step.
bAnalysis for C, H, N were correct within Æ0.4%.
cIC50 values were determined by a single experiment run in duplicate.
d%Inhibition at 10 mM.

T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
405
Table 3. Physicochemical andbiological properties of (3,5- trans)-1-neopentyl-2-oxo-5-phenyl-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-acetic acid
derivatives 2m–y
CompdX
Y
Yield
a (%)
Mp (ꢀC)
223–224
Formulab
IC50 (mM)c
Rat enzyme
HepG2
enzyme
2m
6-Cl
20–Cl
74
C22H23Cl2NO4
>10 (24.5)d
—
e
[a mixture of cis and trans (1:1)]
20-Cl
H
63
72
84
64
97
88
76
90
77
79
92
85
186–187
247–248
251–253
167–168
237–238
223–224
244–247
260–263
230–232
C22H23Cl2NO4
C22H24ClNO4
C22H23BrClNO4
C23H26ClNO5
C23H25ClFNO5
C23H25ClFNO5
C24H28ClNO6
C24H28ClNO6
C24H28ClNO6
>10 (41.9)d
0.76
—
e
2n
2o
2p
2q
2r
8-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
0.55
0.019
0.023
0.027
20-Br
20-OMe
0.027
0.028
0.044
2.0
20-OMe, 40-F
20-OMe, 50-F
20,30-diOMe
20,40-diOMe
20,50-diOMe
20,60-diOMe
20,40,60-triOMe
20-OMe, 30-Me
e
2s
2t
—
0.018
0.020
0.038
0.049
>10 (33.7)d
1.2
2u
2v
2w
2x
2y
e
—
—
e
263–270 (dec.)
260–270 (dec.)
230–231
C24H28ClNO6 3/10H2O
C25H30ClNO7
C24H28ClNO5
e
0.32
0.18
—
0.093
aYieldof final step.
bAnalysis for C, H, N were correct within Æ0.4%.
cIC50 values were determined by a single experiment run in duplicate.
d% Inhibition at 10 mM.
eNot tested.
availability (BA). The BAs of sodium salt 20 andfree
acid 19a in rats (non-fastedstate) were 87 and16%,
respectively.
Table 4. Inhibition of squalene synthase by compounds 19a and 19b
(mM)a
CompdStereochemistry
IC
50
Rat enzyme
HepG2
enzyme
The inhibition by compound 20 of cholesterogenesis
from [14C] acetate at 1 h after oral dosing was examined
in Wistar rats. The ED50 value was 3.0 mg/kg.
19a
19b
3R,5S
3S,5R
0.025
7.7
0.013
3.9
aIC50 values were determined by a single experiment run in duplicate.
Compound 20 was administered orally to marmosets at
a dose of 50 mg/kg/day for 4 days and found to
decrease plasma total cholesterol and non-HDL choles-
terol levels by 32 and64%, respectively, comparedto
control (Fig. 3). Basedon these responses in marmosets,
a primate whose lipoprotein profile is similar to that of
humans, efficacy in humans is expected.
1,2,3,5-tetrahydro-4,1-benzoxazepine-3-acetic acid 20
effectively reduced plasma cholesterol by oral adminis-
tration in marmosets. This compoundis one of the
hopeful candidates for a cholesterol-lowering and anti-
atherosclerotic agent. Further modification of 4,1-ben-
zoxazepine derivatives will be reported.
Conclusions
Experimental
We performedchemical moidfication of 4,1-benzox-
azepine derivatives and evaluated squalene synthase
inhibitory activity. Compounds 1a and 2q,t,u exhibited
potent inhibition of rat andhuman squalene synthases.
Among these, 2t was the most effective inhibitor of
cholesterol synthesis in HepG2 cells. The absolute
stereochemistry requiredfor inhibitory activity was
determined to be 3R,5S. The sodium salt of the (3R,5S)-
7-chloro-5-(2,3-dimethoxyphenyl)-1-neopentyl-2-oxo-
All melting points were determined on a Yanagimoto
micro melting point apparatus andare uncorrecta.d
Proton nuclear magnetic resonance (1H NMR) spectra
were recorded on a Varian GEMINI-200 (200 MHz)
spectrometer (with tetramethylsilane as an internal
standard). Infrared (IR) absorption spectra were recor-
ded on a JASCO IR-810. [a]D values were determined in
the indicated solvents on a JASCO DIP-370 polarimeter.

406
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
TLC analyses were carriedout on Merck Kieselgel 60
F254 plates. Elemental analyses were carriedout by
Takeda Analytical Laboratories, Ltd., and are within
Æ0.4% of the theoretical values unless otherwise noted.
For column chromatography, Merck Kieselgel 60 (70–230
mesh ASTM) was used. Yields were not maximized. The
following abbreviations are used: s=singlet, d=doublet,
t=triplet, q=quartet, m=multiplet, br=broad.
at 0 ꢀC. The mixture was stirredfor 30 min at 0 ꢀC and
concentrated under reduced pressure. The residue was
dissolved in EtOH (250 mL) and water (100 mL) and
concentrated HCl (140 mL) was added. After being
refluxedfor 3 h, the mixture was dilutedwith water
(200 mL) andextractedwith Et 2O (300 mLÂ3). The
extracts were washedwith 1N NaOH andbrine, and
then concentrated. The residue was dissolved in CH2Cl2
(200 mL), dried over Na2SO4 andthen concentrated
under reduced pressure. The residue was chromato-
graphed[eluent: hexane–AcOEt (4:1)] andcrystallized
with Et2O–petroleum ether (1:1) to give 4i (27 g,
92.6 mmol, 37%) as yellow plates. 4k–n were synthe-
sizedin a similar procedure.
2-Aminobenzophenone derivatives (4; Table 6).11
Method A12. A solution of Grignardreagent in THF
13
(15 mL) preparedfrom 2,3-dimethoxybromobenzene
(1.5 g, 6.9 mmol) and Mg (0.17 g, 6.9 mmol) was added
dropwise to a solution of 6-chloro-2-methyl-4H-3,1-
benzoxazin-4-one (2 g, 10.2 mmol) in THF (15 mL) at
the rate of 0.25 mL a minute with ice-cooling. After
being stirredfor 15 min at room temperature, the reac-
tion was quenchedwith 1N HCl (10 mL). The mixture
was extractedwith AcOEt, washedwith water, dried
over Na2SO4 andthen concentrated. The residue was
dissolved in EtOH (20 mL). After addition of 6N HCl
(19 mL), the solution was refluxedovernight, neu-
tralizedwith 1 N NaOH andextractedwith AcOEt. The
extract was washedwith water, driedover Na 2SO4, and
then concentrated. The residue was chromatographed
[eluent: hexane–AcOEt (3:1)] to give 4i (0.58 g,
2.0 mmol, 20%) as yellow prisms. Mp 91–95 ꢀC. IR nmax
2-Amino-ꢀ-phenylbenzyl alcohol derivatives (5a–n; Table
7). NaBH4 (1.5 g, 39.5 mmol) was added to a suspen-
sion of 4i (5.6 g, 19.2 mmol) in EtOH (150 mL). The
mixture was stirredfor 4 h, dilutedwith AcOEt, washed
with brine, dried over Na2SO4 andthen concentrated
under reduced pressure to give 5i (4.9 g, 16.7 mmol,
87%) as colorless powder. Mp 124–126 ꢀC. IR nmax
1
(KBr) cmÀ1: 3410, 3330 (NH), 3400–3200 (br, OH). H
NMR (CDCl3) d 3.17 (1H, br), 3.83 (3H, s), 3.88 (3H,
s), 4.21 (2H, br), 6.03 (1H, br), 6.57-7.11 (6H, m). Anal.
(C15H16ClNO3) C, H, N. 5a,15 b-–h,j–n were preparedin
a similar procedure.
1
(KBr) cmÀ1: 3470, 3340 (NH), 1635 (C¼O). H NMR
2-Acetylamino-5-chloro-ꢀ-(2-chlorophenyl)benzyl acetate
(10a). Ac2O (8.0 g, 78.1 mmol) andDMAP (0.4 g) was
added to a solution of 5a (9.5 g, 35.4 mmol) in
pyridine (100 mL). The mixture was stirred overnight at
room temperature. The solvent was removed, and the
(CDCl3) d 3.78 (3H, s), 3.92 (3H, s), 6.39 (2H, br), 6.63–
7.27 (6H, m). Anal. (C15H14ClNO3) C, H, N. 4b,12 c,e,14
f,12 g,h,j were preparedin a similar procedure.
were commercially available.
4a, d
Method B. n-BuLi hexane solution (1.6 M, 180 mL,
0.29 mol) was added to a solution of veratrol (52 g,
0.376 mol) in THF (200 mL) at the rate of 20 mL a
minute with ice-cooling. The mixture was stirredfor
30 min anda yellow solidwas precipitated. The suspen-
sion was added to a solution of 6-chloro-2-methyl-4H-
3,1-benzoxazin-4-one (49 g, 0.251 mol) in THF (200 mL)
Table 5. Effect of selectedcompounds on cholesterol synthesis from
14C] mevalonate in HepG2 cell
[
CompdY
Cholesterol
synthesis IC50 (mM)a
HepG2 enzyme
IC50 (mM)b
1a
2q
2t
20-Cl
3.2
0.024
0.023
0.018
0.020
20-OMe
0.91
0.12
0.40
20,30-diOMe
20,40-diOMe
Figure 3. Effect of oral administration (50 mg/kg/day, po, 4 days) of
the compound 20 on serum cholesterol levels in common marmosets
(male, 19–62 M old). Values are meanÆS.D. (n=5). **p<0.01 vs the
control value. TC, total cholesterol; non-HDL-C, non-HDL choles-
terol; HDL, high-density lipoprotein.
2u
aIC50 values were determined by a single experiment run in triplicate.
bIC50 values were determined by a single experiment run in duplicate.

T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
407
residue was dissolved in AcOEt. The solution was
washedwith 1 N HCl, saturatedNaHCO andbrine,
2-Acylamino-5-chloro-ꢀ-(2-chlorophenyl)benzyl alcohol
derivatives.11 A mixture of 5a (2.0 g, 7.46 mmol),
NaHCO3 (0.94 g, 11.2 mmol) andCH Cl2 (50 mL) was
3
dried over Na2SO4, andthen concentratedunedr
reduced pressure. The residue was recrystallized from
CH2Cl2–hexane (1:1) to give 10a (12 g, 34.1 mmol, 96%)
as colorless prisms. Mp 131–133 ꢀC (CH2Cl2–hexane).
IR nmax(KBr) cmÀ1: 1735, 1650 (C¼O). 1H NMR
(CDCl3) d 2.20 (3H, s), 2.25 (3H, s), 7.02–7.86 (8H, m),
8.72 (1H, br). Anal. (C17H15Cl2NO3) C, H, N.
2
added to a solution of cyclopropanecarbonyl chloride
(0.82 g, 7.83 mmol) in CH2Cl2 (10 mL) at 0 ꢀC at the rate
of 0.5 mL a minute. The reaction mixture was stirredat
room temperature for 30 min, washedwith water, dried
over Na2SO4, and then concentrated under reduced
pressure to give 11a (2.1 g, 6.25 mmol, 84%) as a color-
less powder. 11b was preparedin a similar procedure.
11a: mp 115–116 ꢀC. Anal. (C17H15Cl2NO2) C, H, N.
11b: mp 109–110 ꢀC. Anal. (C19H21Cl2NO2) C, H, N.
2-Acetyl(methyl)amino-5-chloro-ꢀ-(2-chlorophenyl)benzyl
acetate (10b). NaH (38 mg, 1.57 mmol) was added to a
mixture of 10a (0.5 g, 1.42 mmol) andMeI (0.22 g,
1.57 mmol) in DMF (5 mL). After stirring for 30 min at
room temperature, diluted with AcOEt, washed with
water, 5% KHSO4, saturatedNaHCO 3 andbrine, dried
over Na2SO4, and then concentrated under reduced
pressure. The residue was chromatographed [eluent:
hexane–AcOEt (3:1)] to give 10b (0.52 g, 1.42 mmol,
quant) as a colorless oil. IR nmax(KBr) cmÀ1: 1747, 1666
N-Substituted 2-amino-5-chloro-ꢀ-phenylbenzyl alcohol
derivatives (6a–y; Table 7). NaBH3CN (2.0 g,
31.7 mmol) was added to an ice-cooled solution of 2-
amino-5-chloro-a-(2,3-dimethoxyphenyl)benzyl alcohol
5i (6.6 g, 22.5 mmol), pivalaldehyde (2.1 g, 23.9 mmol)
andAcOH (3.8 g, 63.4 mmol) in MeOH (70 mL). After
being stirredfor 1 h at room temperature, the reaction
was quenchedwith 5% KHSO 4. The mixture was
extractedwith AcOEt (100 ml). The extract was washed
1
(C¼O). H NMR (CDCl3) d 1.21 (1/2Â3H, s), 1.88 (1/
2Â3H, s), 2.16 (1/2Â3H, s), 2.17 (1/2Â3H, s), 2.75 (1/
2Â3H, s), 3.27 (1/2Â3H, s), 7.07–7.56 (8H, m).
with saturatedNaHCO andbrine, driedover Na 2SO4,
3
andthen concentratedin vacuo to give
6t (8.2 g,
2-Acetyl(ethyl)amino-5-chloro-ꢀ-(2-chlorophenyl)benzyl
acetate (10c). Compound 10c (0.54 g, 1.42 mmol,
quant) was preparedfrom 10a (0.5 g, 1.42 mmol) and
EtI (0.24 g, 1.57 mmol) in a similar manner as described
22.5 mmol, quant) as a pale yellow prisms. Mp 120–
121 ꢀC. IR nmax (KBr) cmÀ1: 3550, (NH), 3500–3200
(br, OH). 1H NMR (CDCl3) d 0.93 (9H, s), 2.82 (2H, s),
3.23 (1H, br), 3.83 (3H, s), 3.89 (3H, s), 4.86 (1H, br),
5.99 (1H, s), 6.57–7.13 (5H, m). Anal. (C20H26ClNO3)
C, H, N. 6a,15 d–g,i–k,m–s,u–y were preparedin a
similar procedure.
for the preparation of 10b as a colorless oil. IR
1
n
max(KBr) cmÀ1: 1747, 1666 (C¼O). H NMR (CDCl3)
d 1.00 (1/2Â3H, t, J=7.4 Hz), 1.08 (1/2Â3H, s), 1.15 (1/
2Â3H, t, J=7.4 Hz), 1.86 (1/2 3H, s), 2.14 (1/2Â3H, s),
2.17 (1/2Â3H, s), 2.31–2.48 (1/2Â1H, m), 3.11–3.28 (1/
2Â1H, m), 4.00–4.18 (1/2Â1H, m), 4.39–4.56 (1/2Â1H,
m), 7.02–7.62 (8H, m).
A mixture of 10b (0.52 g, 1.42 mmol), 1 N NaOH (6 mL)
andEtOH (6 mL) was refluxedfor 1 h. The reaction
mixture was diluted with AcOEt, washed with water
Table 6. Physicochemical properties of 2-aminobenzophenone derivatives 4a–n
CompdX
Y
Method
Yield
Mp
(ꢀC)
Formulaa
(%)
78
4a
5-Cl
6-Cl
4-Cl
5-Cl
5-Cl
4-Cl
5-Cl
5-Cl
5-Cl
20-Cl
20-Cl
(commercially available)
A
A
(commercially available)
4b12
4c
20-Cl
112–113
C13H9Cl2NO
4d
H
20-Br
20-OMe
20-OMe,40-F
20-OMe,50-F
20,30-diOMe
4e14
4f12
4g
A
A
A
A
A
B
A
B
B
B
B
86
96
20
37
Quant.
22
32
90
78
106–107
89–90
91–95
C14H11ClFNO2
C14H11ClFNO2
C15H14ClNO3
4h
4i
4j
4k
4l
4m
4n
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
20,40-diOMe
20,50-diOMe
102-103
oil
172
188-189
80–81
C15H14ClNO3
C15H14ClNO3
C15H14ClNO3
C16H16ClNO4
C15H14ClNO2
20,60-diOMe
20,40,60-triOMe
20-OMe,30-Me
aAnalysis for C, H, N were correct within Æ0.4% except for oily compounds.

408
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
andbrine, driedover Na 2SO4, andthen concentrated
under reduced pressure. The residue was chromato-
was stirredfor 2 h at room temperature. After cooling
with ice-bath, the reaction was quenchedwith water
(0.3 mL), 15% NaOH (0.3 mL) andwater (1 mL). The
mixture was filteredandthe filtrate was concentrated
under reduced pressure. The residue was chromato-
graphed[eluent: hexane–AcOEt (3:1)] to give 6h (2.1 g,
6.52 mmol, 78%) as a colorless oil. 1H NMR (CDCl3) d:
0.1–1.3 (5H, m), 2.92 (2H, d, J=6.8 Hz), 6.16 (1H, s),
6.57–7.60 (7H, m). 6l was similarly synthesizedfrom
11b.
graphed[eluent: hexane–AcOEt (10:1)] to give
6b
(0.38 g, 1.35 mmol, 95%) as a colorless oil. 6c was simi-
larly synthesizedfrom 10c.
1
6b: IR nmax(KBr) cmÀ1: 3600–3200 (br, NH, OH). H
NMR (CDCl3) d 2.85 (3H, s), 6.12 (1H, s), 6.61 (1H, d,
J=8.8 Hz), 6.86 (1H, d, J=2.6 Hz), 7.18 (1H, dd,
J=2.6, 8.8 Hz), 7.26–7.48 (4H, m). 6c: IR nmax(KBr)
1
cmÀ1: 3600–3200 (br, NH, OH). H NMR (CDCl3) d:
1.24 (3H, t, J=7.6 Hz), 3.13 (2H, q, J=7.6 Hz), 6.13
(1H, s), 6.60 (1H, d, J=8.8 Hz), 6.84 (1H, d, J=2.2 Hz),
7.15 (1H, dd, J=2.2, 8.8 Hz), 7.26–7.51 (4H, m).
Ethyl 3-[N-[4-chloro-2-(ꢀ-hydroxybenzyl)phenyl]carba-
moyl]acrylate (8a–y; Table 8). A solution of fumaric
acidchloride monoethyl ester
CH2Cl2 (15 mL) was added dropwise to a solution of 6t
(3.9 g, 10.7 mmol) andNaHCO (2.5 g, 30 mmol) in
7 (2.0 g, 12.3 mmol) in
A solution of 11a (2.8 g, 8.33 mmol) in Et2O (20 mL)
was added to a suspension of LiAlH4 (0.3 g) in Et2O
(30 mL) over a periodof 30 min. The reaction mixture
3
CH2Cl2 (100 mL). The reaction mixture was stirredfor
2 h at room temperature andfiltered. The filtrate was
Table 7. Physicochemical properties of 2-amino-a-phenylbenzyl alcohol derivatives 5a-–n and 6a–y
CompdX
Y
R
(%) Yield Mp (
ꢀC)
Formulaa
5a15
5b
5c
5d
5e
5f
5g
5h
5i
5-Cl
6-Cl
4-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
6-Cl
4-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
5-Cl
20-Cl
20-Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
94
70–71
Oil
107–109
87–88
81–82
C13H11Cl2NO
C13H11Cl2NO
C13H12ClNO
C13H11BrClNO
C14H14ClNO2
C14H13ClFNO2
C14H13ClFNO2
C15H16ClNO3
C15H16ClNO3
C15H16ClNO3
C15H16ClNO3
C16H18ClNO4
C15H16ClNO2
20-Cl
Quant.
98
83
87
98
92
87
99
85
H
20-Br
20-OMe
20-OMe,40-F
20-OMe,50-F
20,30-diOMe
20,40-diOMe
20,50-diOMe
20,60-diOMe
20,40,60-triOMe
20-OMe,30-Me
20-Cl
oil
118–119
124–126
127–128
177–178
130–135
163–164
114–115
5j
5k
5l
72
96
95
5m
5n
6a15
6b
6c
6d
6e
6f
6g
6h
6i
6j
6k
6l
6m
6n
6o
6p
6q
6r
H
CH2Ph
Me
Et
Pr
Bu
20-Cl
95
93
85
75
Oil
Oil
Oil
Oil
Oil
Oil
Oil
91–92
Oil
Oil
Oil
Oil
Oil
Oil
99–100
Oil
Oil
C14H13Cl2NO
C15H15Cl2NO
C16H17Cl2NO
C17H19Cl2NO
C16H17Cl2NO
C19H23Cl2NO
C17H17Cl2NO2
C20H23Cl2NO
C18H21Cl2NO
C18H21Cl2NO
C19H23Cl2NO
C18H21Cl2NO
C18H21Cl2NO
C18H22ClNO
20-Cl
20-Cl
20-Cl
20-Cl
Pr
96
20-Cl
CH2CHEt2
CH2-cyclopropyl
CH2-cyclohexyl
CHEt2
Quant.
78
91
87
97
83
79
Quant.
Quant.
79
84
98
98
Quant.
83
64
quant.
72
73
20-Cl
20-Cl
20-Cl
20-Cl
CH2CH2CHMe2
CH2CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
20-Cl
20-Cl
20-Cl
H
20-Br
C18H21BrClNO
C19H24ClNO2
C18H23ClFNO2
C18H23ClFNO2
C20H26ClNO3
C20H26ClNO3
C20H26ClNO3
C20H26ClNO3
C21H28ClNO4
C20H26ClNO2 1/4H2O
20-OMe
20-OMe,40-F
20-OMe,50-F
20,30-diOMe
20,40-diOMe
20,50-diOMe
20,60-diOMe
20,40,60-triOMe
20-OMe,30-Me
6s
6t
Oil
120–121
86-87
126-128
oil
150-151
130-131
6u
6v
6w
6x
6y
aAnalysis for C, H, N were correct within Æ0.4% except for oily compounds.

T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
409
Table 8. Physicochemical properties of ethyl 3-[N-[2-(hydroxymethyl)phenyl]carbamoyl]acrylate derivatives 8a-–y
CompdX
Y
R
(%) Yield Mp (
CH2Ph
ꢀC)
Formulaa
8a15
8b
8c
8d
8e
8f
8g
8h
8i
8j
8k
8l
8m
8n
8o
8p
8q
8r
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
3-Cl
5-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
20-Cl
20-Cl
Me
Et
Pr
Bu
85
Quant.
78
128–130
Oil
Oil
C20H19Cl2NO4 1/4H2O
C21H21Cl2NO4
C22H23Cl2NO4
C23H25Cl2NO4
C22H23Cl2NO4
C25H29Cl2NO4
C23H23Cl2NO4
C26H29Cl2NO4
C24H27Cl2NO4
C24H27Cl2NO4
C25H29Cl2NO4
C24H27Cl2NO4
C24H27Cl2NO4
C24H28ClNO4
C24H27BrClNO4
C25H30ClNO5
C25H29ClFNO5
C25H29ClFNO5
C26H32ClNO6
C26H32ClNO6
C26H32ClNO6
C26H32ClNO6
C27H34ClNO7
C26H32ClNO5
20-Cl
20-Cl
20-Cl
96
87
82
79
77
82
92
Oil
20-Cl
Pr
180-182
Oil
Oil
Oil
Oil
Oil
Oil
Oil
152–153
Oil
20-Cl
CH2CHEt2
CH2-cyclopropyl
CH2-cyclohexyl
CHEt2
20-Cl
20-Cl
20-Cl
20-Cl
CH2CH2CHMe2
CH2CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
20-Cl
Quant.
88
90
20-Cl
20-Cl
H
20-Br
86
93
96
97
88
92
99
91
168–169
Oil
Oil
20-OMe
20-OMe,40-F
20-OMe,50-F
20,30-diOMe
20,40-diOMe
20,50-diOMe
20,60-diOMe
20,40,60-triOMe
20-OMe,30-Me
8s
8t
Oil
127–129
137–139
140–142
Oil
Oil
Oil
8u
8v
8w
8x
8y
68
95
57
washedwith water, driedover Na 2SO4, andthen con-
centrated under reduced pressure to give 8t (4.8 g,
9.8 mmol, 92%) as colorless prisms. Mp 127–129 ꢀC. IR
1-Substituted (3,5-trans)-7-chloro-2-oxo-5-phenyl-1,2,3,5-
tetrahydro-4,1-benzoxazepine-3-acetic acid derivatives
(2a–y; Tables 1 and 3). A mixture of 9t (2.5 g,
5.10 mmol), K2CO3 (2.8 g, 20.4 mmol), MeOH (70 mL)
andwater (10 mL) was stirredovernight at room tem-
perature. The reaction mixture was diluted with water,
acidified, extracted with AcOEt. The extract was
washedwith water, driedover Na 2SO4, andthen con-
centrated under reduced pressure to give 2t (1.8 g,
3.9 mmol, 76%) as colorless needles. Mp 244–247 ꢀC. IR
n
(KBr) cmÀ1: 3470 (OH); 1725, 1660, 1630 (C¼O,
max
1
C¼C). H NMR (CDCl3) d 0.89 (9H, s), 1.24 (3H, t,
J=7.0 Hz), 2.81 (1H, d, J=13.4 Hz), 3.72 (3H, s), 3.88
(3H, s), 4.15 (2H, q, J=7.0 Hz), 4.42 (1H, d,
J=13.4 Hz), 6.02 (1H, s), 6.78–7.31 (8H, m). Anal.
(C26H32ClNO6) C, H, N. The compounds 8a,15 b–s,u–y
were synthesizedby acylation similar to that for the
preparation of 8t.
n
(KBr) cmÀ1: 3600–2200 (br, COOH), 1740, 1650
max
(C¼O). 1H NMR (CDCl3) d: 0.95 (9H, s), 2.84 (1H, dd,
J=16.6, 5.6 Hz), 3.08 (1H, dd, J=16.6, 7.4 Hz), 3.39
(1H, d, J=13.8 Hz), 3.63 (3H, s), 3.89 (3H, s), 4.34 (1H,
dd, J=7.4, 5.6 Hz), 4.53 (1H, d, J=13.8 Hz), 6.28 (1H,
s), 6.64 (1H, d, J=1.6 Hz), 6.97–7.36 (5H, m). Anal.
(C24H28ClNO6) C, H, N. 2a–s,u–y were synthesizedin a
similar manner.
Ethyl (3,5-trans)-7-chloro-2-oxo-5-phenyl-1,2,3,5-tetra-
hydro-4,1-benzoxazepine-3-acetate derivatives (9a–y; Ta-
ble 9). A mixture of 8t (7.2 g, 14.7 mmol) andK CO3
2
(2 g) in EtOH (100 mL) was stirredfor 24 h. The reac-
tion mixture was idlutedwith AcOEt, washedwith
water, dried over Na2SO4, andthen concentrated. The
residue was subjected to column chromatography
[eluent:hexane–AcOEt (3:1)] andrecrystallizedfrom
AcOEt–hexane (1:5) to give 9t (6.5 g, 13.3 mmol, 90%)
as colorless prisms. Mp 184–185 ꢀC. IR nmax (KBr)
cmÀ1: 1720, 1680 (C¼O). 1H NMR (CDCl3) d: 0.95
(9H, s), 1.24 (3H, t, J=7.2 Hz), 2.77 (1H, dd, J=16.4,
5.8 Hz), 3.04 (1H, dd, J=16.4, 7.6 Hz), 3.37 (1H, d,
J=14.0 Hz), 3.63 (3H, s), 3.89 (3H, s), 4.13 (2H, dq,
J=7.2, 1.8 Hz), 4.39 (1H, dd, J=7.6, 5.8 Hz), 4.52 (1H,
d, J=14.0 Hz), 6.28 (1H, s), 6.62 (1H, s), 6.96–7.33 (5H,
m). Anal. (C26H32ClNO6) C, H, N. 9a,15 b–s,u-–y were
preparedfrom 8a–s,u–y in a similar procedure.
(3,5-trans)-3-(Aminomethyl)-7-chloro-5-(2-chlorophenyl)-
1-neopentyl-1,5-dihydro-4,1-benzoxazepine-2(3H)-one
hydrochloride (12). DPPA (0.74 mL, 3.44 mmol) and
Et3N (0.48 mL, 3.44 mmol) was added to a solution of
1a (1.0 g, 2.29 mmol) in DMF (20 mL) at 0 ꢀC. The
mixture was stirredfor 1 h at room temperature, diluted
with water andextractedwith AcOEt. The extract was
washedwith 5% KHSO 4 andsaturatedNaHCO 3, dried
over MgSO4 andthen concentrate.d The resiude in
benzene (100 mL) was refluxedfor 1 h. The solvent was
removed under reduced pressure. A mixture of the

410
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
residue, 6N HCl (50 mL) and dioxane (30 mL) was
refluxedfor 20 min andthen concentratedin vacuo. The
residue was recrystallized from EtOH–hexane to give 12
(0.87 g, 1.96 mmol, 86%) as colorless plates.
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-3-(2-hydroxyethyl)
-1-neopentyl-1,5-dihydro-4,1-benzoxazepin-2(3H)-one (14).
N-Methylmorpholine (4.26 mL, 38.8 mmol) and
ClCOOEt (3.71 mL, 38.8 mmol) was added dropwise to a
solution of 1a (14 g, 32.1 mmol) in THF (150 mL) at 0 ꢀC.
After stirring for 15 min, NaBH4 (4.1 g, 96.9 mmol) and
MeOH (150 mL) were added to the mixture. After being
stirredfor 2 h at room temperature, the reaction was
quenchedwith 1N HCl. The mixture was dilutedwith
water andextractedwith AcOEt. The extract was
washedwith water, driedover MgSO 4, andthen con-
centrated. The residue was chromatographed [eluent:
hexane–AcOEt (1:1)] andrecrystallizedfrom hexane–
AcOEt to give 14 (8.6 g, 20.4 mmol, 64%) as colorless
prisms. Mp 157–159 ꢀC. IR nmax (KBr) cmÀ1: 3460
Mp 173–175 ꢀC. IR nmax (KBr) cmÀ1: 1665 (C¼O). H
1
NMR (DMSO-d6) d 0.88 (9H, s), 3.0–3.3 (2H, m), 3.68
(1H, d, J=13.8 Hz), 4.2–4.3 (1H, m), 4.31 (1H, d,
J=13.8 Hz), 6.17 (1H, s), 6.36 (1H, d, J=2.6 Hz), 7.5–
.
.
8.1 (6H, m). Anal. (C21H24Cl2N2O2 HCl H2O) C, H, N.
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-3-(hydroxyme-
thyl)-1-neopentyl-1,5-dihydro-4,1-benzoxazepin-2(3H)-
one (13). A solution of NaNO2 (2.0 g) in water (2 mL)
was added dropwise to a mixture of 12 (4.0 g,
9.01 mmol), AcOH (24 mL) andwater (35 mL) at 0 ꢀC.
The mixture was stirredfor 2 h at room temperature,
pouredinto water andextractedwith AcOEt. The
extract was washedwith water, driedover MgSO 4, and
then concentrated under reduced pressure. The residue
was chromatographed[eluent: hexane–AcOEt (4:1)] to
give 13 (0.85 g, 2.08 mmol, 23%) as colorless prisms. Mp
169–170 ꢀC. IR nmax (KBr) cmÀ1: 1670 (C¼O). 1H
NMR (CDCl3) d 0.95 (9H, s), 3.38 (1H, d, J=13.8 Hz),
3.8–4.1 (3H, m), 4.55 (1H, d, J=13.8 Hz), 6.27 (1H, s),
6.56 (1H, d, J=2.2 Hz), 7.3–7.9 (6H, m). Anal.
(C21H23Cl2NO3) C, H, N.
1
(OH), 1660 (C¼O). H NMR (CDCl3) d 0.94 (9H, s),
2.15 (2H, q, J=5.8 Hz), 2.25 (1H, t, J=5.3 Hz), 3.38
(1H, d, J=14.0 Hz), 3.71–3.90 (2H, m), 4.15 (1H, t,
J=6.2 Hz), 4.52 (1H, d, J=14.0 Hz), 6.27 (1H, s), 6.53
(1H, d, J=1.8 Hz), 7.29–7.49 (5H, m), 7.70–7.77 (1H,
m). Anal. (C22H25Cl2NO3) C, H, N.
(3,5-trans)-7-Chloro-3-(2-chloroethyl)-5-(2-chlorophe-
nyl)-1-neopentyl-1,5-dihydro-4,1-benzoxazepin-2(3H)-one
(15). A mixture of 14 (11 g, 26.0 mmol), SOCl2 (3.65 g,
50 mmol), pyridine (50 mg, 0.63 mmol) and toluene
(100 mL) was stirredfor 30 min at 90 ꢀC, pouredinto
Table 9. Physicochemical properties of 1-substitutedethyl (3,5- trans)-7-chloro-5-(2-chlorophenyl)-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-
acetate derivatives 9 a-–y
CompdX
Y
R
(%) Yield
CH2Ph
Me
Et
Prn
Bun
Prn
Mp (
ꢀC)
Formulaa
9a15
9b
9c
9d
9e
9f
9g
9h
9i
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
6-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
20-Cl
68
47
81
91
93
90
77
76
30
85
93
80
147–178
119–120
Oil
Oil
Amorphous
Oil
106–107
Amorphous
140–141
Oil
C20H19Cl2NO4
C21H21Cl2NO4
C22H23Cl2NO4
C23H25Cl2NO4
C22H23Cl2NO4
C25H29Cl2NO4
C23H23Cl2NO4
C26H29Cl2NO4
C24H27Cl2NO4
C24H27Cl2NO4
C25H29Cl2NO4
C24H27Cl2NO4
CH2CHEt2
CH2–cyclopropyl
CH2–cyclohexyl
CHEt2
CH2CH2CHMe2
CH2CH2CMe3
CH2CMe3
9j
9k
9l
Oil
Oil
9m
[a mixture of cis and trans (1:1)]
9n
9o
9p
9q
9r
8-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
7-Cl
20-Cl
H
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
CH2CMe3
86
84
82
81
83
86
90
80
78
71
80
88
142–143
Oil
C24H27Cl2NO4
C24H28ClNO4
C24H27BrClNO4
C25H30ClNO5
C25H29ClFNO5
C25H29ClFNO5
C26H32ClNO6
C26H32ClNO6
C26H32ClNO6
C26H32ClNO6
C27H34ClNO7
C26H32ClNO5
20-Br
128–129
173–174
114–115
153–154
184–185
117–119
165–166
175–177
154–155
190–191
20-OMe
20-OMe,40-F
20-OMe,50-F
20,30-diOMe
20,40-diOMe
20,50-diOMe
20,60-diOMe
20,40,60-triOMe
20-OMe,30-Me
9s
9t
9u
9v
9w
9x
9y
aAnalysis for C, H, N were correct within Æ0.4% except for oily compounds.

T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
411
saturatedNaHCO
andextractedwith AcOEt. The
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-1-neopentyl-2-
oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-propionic
3
extract was washedwith water, driedover Na 2SO4, and
then concentratedunder reducedpressure. The residue
was subjectedto column chromatography [eluent: hexane–
AcOEt (7:1)] to give 15 (7.0 g, 15.9 mmol, 61%) as color-
less needles. Mp 182–184 C. H NMR (CDCl3) d 0.94
(9H, s), 2.13–2.50 (2H, m), 3.39 (1H, d, J=14.0 Hz),
3.63–3.84 (2H, m), 4.17 (1H, dd, J=8.0, 5.0 Hz), 4.52
(1H, d, J=14.0 Hz), 6.27 (1H, s), 6.53 (1H, d,
J=1.8 Hz), 7.3–7.5 (5H, m), 7.7–7.77 (1H, m). Anal.
(C22H24Cl3NO2) C, H, N.
acid (3b; Table 2).
A
mixture of 17 (90 mg,
0.188 mmol), 10% K2CO3 (2 mL) andMeOH (5 mL)
was refluxed for 1 h, diluted with water, acidified,
extractedwith AcOEt. The extract was washedwith
water, dried over MgSO4, andthen concentratedto give
3b (80 mg, 0.178 mmol, 94%) as colorless crystals. Mp
225–227 ꢀC. IR nmax (KBr) cmÀ1: 1700, 1680 (C¼O). 1H
NMR (CDCl3) d 0.98 (9H, s), 2.0–2.7 (4H, m), 3.38 (1H,
d, J=14.0 Hz), 3.96 (1H, dd, J=7.2, 5.8 Hz), 4.51 (1H,
d, J=14.0 Hz), 6.24 (1H, s), 6.52 (1H, s), 7.2–7.8 (6H,
m). Anal. (C23H25Cl2NO4) C, H, N.
ꢀ
1
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-1-neopentyl-2-
oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-propionitrile
(16). A mixture of 15 (0.3 g, 0.681 mmol), NaCN (0.1 g)
andDMSO (6 mL) was stirredfor 1 h at 100 ꢀC, diluted
with water andextractedwith AcOEt. The extract was
washedwith 5% KHSO 4, saturatedNaHCO 3 andbrine,
dried over MgSO4, andthen concentratedunedr
reduced pressure. The residue was chromatographed
[eluent: hexane–AcOEt (4:1)] to give 16 (0.25 g,
0.58 mmol, 85%) as colorless crystals. Mp 194–195 ꢀC.
IR nmax (KBr) cmÀ1: 2240 (CN), 1680 (C¼O). 1H NMR
(CDCl3) d 0.94 (9H, s), 2.0–2.4 (2H, m), 2.59 (2H, t,
J=7.2Hz), 3.38 (1H, d, J=13.8 Hz), 4.05 (1H, dd,
J=7.6, 5.0 Hz), 4.51 (1H, d, J=13.8 Hz), 6.26 (1H, s),
6.54 (1H, d), 7.3–7.8 (6H, m). Anal. (C23H24Cl2N2O2)
C, H, N.
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-1-neopentyl-2-
oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-acetamide
(3c; Table 2). DEPC (0.46 g) andEt 3N (0.4 mL) was
added to a ice-cooled solution of 1a (1.0 g, 2.29 mmol),
NH4Cl (0.5 g) andEt N (0.5 mL) in DMF (8 mL). The
3
mixture was stirredfor 30 min at room temperature,
pouredinto water andextractedwith AcOEt. The
extract was washedwith 5% KHSO 4, saturatedNaHCO 3
andwater, driedover MgSO
4, andthen concentrated
under reduced pressure to give 3c (0.65 g, 1.49 mmol, 65%)
as colorless crystals. Mp 291–292ꢀC. IR nmax (KBr) cmÀ1
:
3340, 3200 (NH2), 1620 (C¼O). 1H NMR (CDCl3) d 0.96
(9H, s), 2.68 (1H, dd, J=14.6, 6.2 Hz), 2.88 (1H, dd,
J=14.6, 7.0 Hz), 3.37 (1H, d, J=14.0Hz), 4.39 (1H, dd,
J=7.0, 6.2 Hz), 4.50 (1H, d, J=14.0 Hz), 5.37 (1H, br),
5.92 (1H, br), 6.28 (1H, s), 6.61 (1H, d, J=1.8 Hz), 6.96–
7.33 (6H, m). Anal. (C22H24Cl2N2O3) C, H, N.
Ethyl (3,5-trans)-7-chloro-5-(2-chlorophenyl)-1-neopen-
tyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-propio-
nate (17). A mixture of 16 (0.2 g, 0.464 mmol), 6N HCl
(3 mL) andEtOH (3 mL) was refluxedfor 6 h, andcon-
centrated under reduced pressure. Water was added to
the residue and the resulting mixture was extracted with
AcOEt. The extract was washedwith saturated
NaHCO3, dried over MgSO4, andthen concentrated
under reduced pressure. The residue was chromato-
graphed[eluent:hexane–AcOEt (5:1)] to give 17 (0.18 g,
0.376 mmol, 81%) as colorless crystals. Mp 130–131 ꢀC.
IR nmax (KBr) cmÀ1: 1730, 1680 (C¼O). 1H NMR
(CDCl3) d 0.94 (9H, s), 1.17 (3H, t, J=7.2 Hz), 2.0–2.6
(4H, m), 3.37 (1H, d, J=14.0 Hz), 3.95–4.11 (3H, m), 4.52
(1H, d, J=14.0 Hz), 6.26 (1H, s), 6.52 (1H, d, J=2.2 Hz),
7.2–7.8 (6H, m). Anal. (C25H29Cl2NO4) C, H, N.
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-1-neopentyl-2-
oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-acetaldehyde
(3d; Table 2).
A solution of DMSO (4.73 mL,
66.6 mmol) in CH2Cl2 (10 mL) was added dropwise to a
solution of (COCl)2 (3.87 mL, 44.4 mmol) in CH2Cl2
(110 mL) at À78 ꢀC. After stirring for 10 min at À78 ꢀC,
a solution of 14 (9.4 g, 22.3 mmol) in CH2Cl2 (60 mL)
was added dropwise. The mixture was stirred for 1.5 h
at À70 ꢀC. After addition of Et3N (15.5 mL, 0.11 mol) at
À60 ꢀC, the mixture was stirredfor 5 min at À78 ꢀC and
for 3 h at room temperature. The mixture was poured
into water andextractedwith CH 2Cl2. The extract was
washedwith water, driedover MgSO 4, andthen con-
centrated. The residue was chromatographed [eluent:
hexane–AcOEt (3:1)] andrecrystallizedfrom hexane-
AcOEt to give 3d (8.6 g, 20.5 mmol, 92%) as colorless
prisms. Mp 173–176 ꢀ. IR nmax (KBr) cmÀ1: 1720, 1680
(C¼O). 1H NMR (CDCl3) d 0.94 (9H, s), 2.89 (1H, ddd,
J=17.6, 5.5, 1.5 Hz), 3.11 (1H, ddd, J=17.6, 6.4,
1.0 Hz), 3.41 (1H, d, J=14.0 Hz), 4.48 (1H, t,
J=6.1 Hz), 4.51 (1H, d, J=14.0 Hz), 6.28 (1H, s), 6.54
(1H, s), 7.31–7.49 (5H, m), 7.67–7.74 (1H, m), 9.83 (1H,
s). Anal. (C22H23Cl2NO3) C, H, N.
(3,5-trans)-7-Chloro-5-(2-chlorophenyl)-1-neopentyl-2-
oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-carboxylic
acid (3a; Table 2). Jones’ reagent (1.0 mL) was added to
a solution of 13 (0.5 g, 1.22 mmol) in acetone (10 mL).
The mixture was stirredfor 1.5 h at room temperature,
concentrated, dissolved in AcOEt and washed with
water. The solvent was removed under reduced pres-
sure. The residue was dissolved in saturated NaHCO3.
The solution was washedwith Et 2O, acidified and
extractedwith AcOEt. The extract was washedwith
water, dried over MgSO4, andthen concentratedto give
3a (0.25 g, 0.592 mmol, 48%) as colorless crystals. Mp
166–167 ꢀC. IR nmax (KBr) cmÀ1: 1750, 1670, 1630
Methyl N-[(3R,5S)-7-chloro-5-(2,3-dimethoxyphenyl)-1-
neopentyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-
acetyl]-L-leucinate (18a) and methyl N-[(3S,5R)-7-chloro-
5-(2,3-dimethoxyphenyl)-1-neopentyl-2-oxo-1,2,3,5-tetra-
1
(C¼O). H NMR (CDCl3) d 0.95 (9H, s), 3.43 (1H, d,
J=13.8 Hz), 4.53 (1H, d, J=13.8 Hz), 4.59 (1H, s), 6.39
(1H, s), 6.59 (1H, d), 7.3–7.9 (6H, m). Anal.
(C21H21Cl2NO4) C, H, N.
hydro-4,1-benzoxazepine-3-acetyl]-L-leucinate
DEPC (0.45 g, 2.6 mmol) was added to an ice-cooled
(18b).
solution of 2t (1.0 g, 2.16 mmol) andmethyl l-leucinate

412
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
hydrochloride (0.47 g, 2.6 mmol) in DMF (20 mL), fol-
lowed by addition of Et3N (0.75 mL, 5.41 mmol). The
mixture was stirredfor 30 min at room temperature,
pouredinto water andextractedwith AcOEt. The
extract was washedwith idlutedHCl andsaturated
NaHCO3, dried over MgSO4, andthen concentrated.
The residue was chromatographed [eluent: hexane–
AcOEt (2:1)] to give 18a (0.52 g, 0.883 mmol, 41%) as
colorless prisms from the first fraction and 18b (3S,5R,
0.55 g, 0.934 mmol, 43%) as a colorless oil from the
secondfraction.
Sodium (3R,5S)-7-chloro-5-(2,3-dimethoxyphenyl)-1-neo-
pentyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-ace-
tate (20). A solution of 19a (30 g, 64.9 mmol) and1 N
NaOH (64.9 mL, 64.9 mmol) in MeOH (400 mL) was
concentrated under reduced pressure. AcOEt was added
to the residue and the solvent was removed. The residue
was washedwith AcOEt to give 20 (31.8 g, 65.7 mmol,
23
quant) as a colorless powder. Mp>300ꢀ. ½aꢁD À235.1 ꢀ
(c 0.60, MeOH). IR nmax (KBr) cmÀ1: 1660, 1580
1
(C¼O). H NMR (CDCl3) d 0.95 (9H, s), 2.59 (1H, dd,
J=15.4, 6.4 Hz), 2.79 (1H, dd, J=15.4, 7.2 Hz), 3.53
(1H, d, J=14.0 Hz), 3.58 (3H, s), 3.88 (3H, s), 4.39 (1H,
dd, J=7.2, 6.4 Hz), 4.44 (1H, d, J=14.0 Hz), 6.21 (1H,
s), 6.51 (1H, d, J=2.4 Hz), 7.08 (1H, dd, J=7.8,
2.0 Hz), 7.19 (1H, t, J=7.8 Hz), 7.27 (1H, dd, J=7.8,
2.0 Hz), 7.40 (1H, dd, J=8.4, 2.4 Hz), 7.55 (1H, d,
J=8.8 Hz). Anal. (C24H28ClNNaO6) C, H, N.
18a: mp 150–151 ꢀC. IR nmax (KBr) cmÀ1: 1740, 1670
1
(C¼O). H NMR (CDCl3) d 0.8–1.0 (15H, m), 1.95–2.2
(1H, m), 2.68 (1H, dd, J=14.4, 5.8 Hz), 2.91 (1H, dd,
J=14.4, 7.0 Hz), 3.35 (1H, d, J=14.0 Hz), 3.62 (3H, s),
3.70 (3H, s), 3.89 (3H, s), 4.3–4.4 (1H, m), 4.49 (1H, d,
J=14.0 Hz), 4.5–4.65 (1H, m), 6.2–6.4 (2H, m), 6.62
(1H, d, J=2.2 Hz), 6.9–7.4 (5H, m). Anal.
Crystallographic data
.
(C31H41ClN2O7 1/2H2O) C, H, N. 18b: IR nmax (neat)
cmÀ1: 1740, 1670, 1660 (C¼O). H NMR (CDCl3) d
Crystallographic data (excluding structure factors) for
the structures in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supple-
mentary publication nos. CCDC 164792 (19a). Copies
of the data can be obtained, free of charge, on applica-
tion to CCDC, 12 Union Road, Cambridge CB2 1EZ,
UK (fax: +44-1223-336033 or e-mail: deposit@-
ccdc.cam.ac.uk).
1
0.8–1.0 (15H, m), 1.5–1.75 (1H, m), 2.70 (1H, dd,
J=14.4, 6.0 Hz), 2.89 (1H, dd, J=14.4, 6.6 Hz), 3.37
(1H, d, J=14.0 Hz), 3.62 (3H, s), 3.71 (3H, s), 3.89
(3H, s), 4.37 (1H, t, J=6.3 Hz), 4.52 (1H, d,
J=14.0 Hz), 4.5–4.7 (1H, m), 6.28 (1H, m), 6.41 (1H,
brd, J=8.4 Hz), 6.61 (1H, d, J=1.6 Hz), 6.9–7.4 (5H,
m).
(3R,5S)-7-Chloro-5-(2,3-dimethoxyphenyl)-1-neopentyl-
2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-acetic acid
(19a) and (3S,5R)-7-chloro-5-(2,3-dimethoxyphenyl)-1-
neopentyl-2-oxo-1,2,3,5-tetrahydro-4,1-benzoxazepine-3-
acetic acid (19b). A mixture of 18a (0.4 g, 0.679 mmol),
dioxane (6 mL), MeOH (10 mL) and concentrated HCl
(5 mL) was refluxedovernight, cooledto room tem-
perature, pouredinto water andextractedwith AcOEt.
Animals and materials
Animals were suppliedby Clea, Japan, Inc. unless
otherwise mentioned. Male Wistar rats were free access
to standard rodent chow (CE-2 in pellet form, Clea
Japan). RS-[2-14C] mevalonolactone and[1- 3H]-farnesyl
pyrophosphate were purchasedfrom New England
Nuclear. [2-14C] mevalonic acidwas synthesizedfrom
[2-14C] mevalonolactone by saponification with potas-
sium hydroxide. [2-14C] Sodium acetate was purchased
from Amersham. Farnesyl pyrophosphate was synthe-
sizedby the methoddescribedby V. J. Davisson and
coworkers16 (Nemoto & Co.). HepG2 cells were sup-
pliedby ATCC. Fetal bovine serum (FBS) andDulbec-
co’s modified Eagle’s medium (DMEM) were purchased
from GIBCO. Human lipoprotein deficient serum
(human LPDS) was purchasedfrom Sigma. All other
reagents were suppliedby Wako Pure Chemical Indus-
tries.
The extract was washedwith water, driedover MgSO
4
and then concentrated. The residue was dissolved in
DMF (5 mL), andMeI (0.042 mL, 0.68 mmol) and
K2CO3 (0.094 g, 0.68 mmol) was added. The reaction
mixture was stirredfor 20 min at room temperature,
pouredinto water, andextractedwith AcOEt. The
extract was washedwith idlutedHCl andsaturated
NaHCO3, dried over MgSO4 andthen concentrated.
The residue was chromatographed [eluent: hexane–
AcOEt (5:1)] to give methyl ester of 19a (85 mg,
0.179 mmol, 26%). A mixture of this compound, K2CO3
(50 mg, 0.36 mmol), MeOH (2 mL) andwater (2 mL)
was refluxed for 2 h, diluted with water, acidified and
extractedwith AcOEt. The extract was washedwith
water, dried over MgSO4, andthen concentratedto give
19a (60 mg, 0.13 mmol, 73%) as colorless crystals. Mp
Preparation of rat squalene synthase
An SD male rat (6 weeks old) was killed by bleeding,
andits liver was excised. About 10 g of the liver was
washedwith a saline solution cooledwith ice, which was
then homogenizedin 15 mL of an ice-cooledbuffer
solution [100 mM potassium phosphate (pH 7.4),
15 mM nicotinamide, 2 mM MgCl2], followedby cen-
trifugation for 20 min at 10,000g (4 ꢀC). The super-
natant layer was separatedandsubjectedto further
centrifugation for 90 min at 105,000g (4 ꢀC). The sedi-
ment was then suspended in an ice-cooled 100 mM
potassium phosphate buffer solution (pH 7.4), which
was again subjectedto centrifugation for 90 min at
24
218–222 ꢀC. ½ꢀꢁD À246.8 ꢀ (c 0.43, MeOH). Anal.
.
(C24H28ClNO6 3/4H2O) C, H, N.%ee 99.4% [HPLC
analysis; ULTRON ES-OVM (4.6IDÂ150; Shinwa
Chemical Industries, Ltd); Eluent, EtOH: 0.02 M
KH2PO4 (pH 3.5)=35:65; flow rate, 0.5 mL/min; reten-
tion time, 19.2 min for 19a and10.3 min for 19b].
19b was similarly synthesizedfrom 18b. Mp 227–230 ꢀC.
24
ꢀ
2H2O) C, H, N.%ee 98.1% (HPLC analysis).
.
½ꢀꢁD +242.7 (c 0.40, MeOH). Anal. (C24H28ClNO6 1/

T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
413
105,000 g (4 ꢀC). The sediment thus obtained (micro-
some fraction) was suspended in an ice-cooled 100 mM
potassium phosphate buffer (pH 7.4) (about 40 mg/mL
protein concentration, determined using BCA protein
assay kit of Pierce Co., Ltd.). This suspension was used
as the enzyme solution.
400 mL of 15% potassium hydroxide in 80% ethanol,
followedby aidng 300 mL of distilled water. Non-
saponifiables were extractedwith n-hexane (800 mL).
The hexane layer (400 mL) was concentratedunedr
reduced pressure. The residue was dissolved in the 0.1%
solution of cholesterol in acetone–ethanol (1:1)
(200 mL). The 0.5% solution of digitonin in 50% etha-
nol (400 mL) was added to the solution. After leaving at
room temperature, cholesterol fraction was obtainedas
digitonin precipitate. The radioactivity taken into digi-
tonin precipitate was measured.
Preparation of human squalene synthase
HepG2 cells (about 1Â109 cells) obtainedby incubation
(37 ꢀC in the presence of 5% CO2) in a DMEM contains
10% FBS, penicillin G (100 units/mL) andstreptomycin
(10 g/mL) were suspended in 10 mL of ice-cooled buffer
solution [100 mM potassium phosphate buffer (pH 7.4),
30 mM nicotinamide and 2.5 mM MgCl2]. The cells
were crashedby means of ultrasonication (for 30 s,
twice). From the sonicate thus obtained, the microsome
fraction was obtainedby the same procedure as in pre-
paration of rat-derived enzyme, which was suspended in
an ice-cooled100 mM potassium phosphate buffer (pH
7.4) (about 4 mg/mL protein concentration). This sus-
pension was usedas the enzyme solution.
Cholesterogenesis in the liver in Wistar rats
Six-week-oldmale Wistar rats were orally administered
with a test compound0.5%–methylcellulose emulsion,
andwere intravenously injectedwith [ 14C] acetate
(10 mCi/rat) 1 h after administration. Animals were kil-
led1 h after the injection. The livers were removed,
saponified, extracted with petroleum ether and then
dried under nitrogen vapor. The residue was dissolved
in the ethanol–acetone (1:1) (3 mL). The 0.5% solution
of digitonin in 50% ethanol (2 mL) was added to the
solution. After leaving at room temperature for 4 h,
cholesterol fraction was obtainedas idgitonin pre-
cipitate. The radioactivity taken into digitonin pre-
cipitate was measured.
Assay of squalene synthase inhibitory activity
Squalene synthase activity was monitoredby the for-
mation of [3H]squalene from [1-3H]FPP. Fifty microliter
of assay mixture included 5 mM[1-3H]FPP (25 mCi/mol),
1 mM NADPH, 5 mM MgCl2, 6 mM glutathione,
100 mM buffer solution of potassium phosphate (pH
7.4), the test compounddissolvedin DMSO (a final
concentration of DMSO was 2%) andenzyme solution
preparedfrom rat or HepG2 cells (protein content
0.8 g). The assay ran 45 min at 37 ꢀC andstoppedby
adding 150 mL of CHCl3–MeOH (1:2) containing 0.2%
coldsqualene as carrier. Aqueous solution of 3 N
Effect on plasma cholesterol levels in marmosets
By a stomach tube, a test compound0.5%–methylcel-
lulose emulsion was administered to male Common
marmosets (19–62 months old, purchased from Japan
EDM) for 4 days. Under non-fasted condition, blood
was taken from the femoralvain andthe plasma total
cholesterol, HDL (high-density lipoprotein)-cholesterol,
andtriglyceride were enzymatically measuredusing an
assay kit (Wako Pure Chemical Industries). The non-
HDL-cholesterol was calculatedby subtracting HDL-
cholesterol from total cholesterol.
NaOH (50 mM) andCHCl (50 mM) were added to the
3
mixture. The chloroform layer containing the reaction
mixture having squalene as the principal component
and3 mL of toluene-basedliquidscintillator were
mixed, and its radioactivity was determined by means of
a liquidscintillation counter. The squalene synthase
inhibitory activity was expressedin terms of the con-
centration of the test compoundinhibiting by 50% the
radioactivity taken into the chloroform layer [IC50,
molar concentration (M)].
Acknowledgements
The authors wish to thank Dr. Kanji Meguro for his
encouragement andhelpful discussion throughout this
study. We also thank Mr. Akira Fujishima and Ms.
Keiko Higashikawa for X-ray crystallographic analysis.
Inhibition of cholesterol synthesis in HepG2 cells
HepG2 cells were culturedin 24-well cell culture plates
(105 cells/well) in a DMEM contains 10% FBS, peni-
cillin G (100 units/mL) andstreptomycin (10 g/mL) for
6 days and preincubated overnight in a DMEM con-
taining 10% human LPDS (Sigma). The medium was
replacedwith the meidum containing the test com-
pounds that previously had been dissolved in DMSO (a
final concentration of DMSO was 0.4% or less). After
incubation for 1 h at 37 ꢀC, 10 ml of 25 mM [14C] meva-
lonic acid(2 mCi/ mmol) 10 mL was added, and further
incubatedfor 2 h. The cells were washedwith phos-
phate-bufferedsaline (twice), dissolvedin 15% aqueous
solution of potassium hydroxide (100 m) at 37 ꢀC. The
cell lysates were saponifiedfor 1 h at 75 ꢀC by adding
References and Notes
1. Miki, T.; Kori, M.; Fujishima, A.; Mabuchi, H.; Tozawa,
R.; Nakamura, M.; Sugiyama, Y.; Yukimasa, H. Bioorg. Med.
Chem. 2002, 10 (this issue).
2. (a) Yukimasa, H.; Tozawa, R.; Kori, M.; Kitano, K.
(Takeda) Eur. Patent, 567026; Chem. Abstr. 1994, 120,
164246g. (b) Yukimasa, H.; Kori, M.; Tozawa, R.; Sugiyama,
Y. (Takeda) Eur. Patent, 645377; Chem. Abstr. 1995, 123,
256777w. (c) Yukimasa, H.; Tozawa, R.; Sigiyama, Y.; Kori,
M. (Takeda) Eur. Patent, 645378; Chem. Abstr. 1995, 123,
143931y. (d) Yukimasa, H.; Tozawa, R.; Kori, M.; Kitano, K.
(Takeda) PCT Int. Appl. WO9521834; Chem. Abstr. 1996,
124, 55994m.

414
T. Miki et al. / Bioorg. Med. Chem. 10 (2002) 401–414
3. (a) Shioiri, T.; Ninomiya, K.; Yamada, S. J. Am. Chem.
Soc. 1972, 94, 6203. (b) Yamada, S.; Ninomiya, K.; Shioiri, T.
Tetrahedron Lett. 1973, 26, 2343.
does not reach such a 2-fold increase over 2t, 19a is 1.5–1.7-
foldmore potent than 2t. It is assumedthat 19a wouldbe as
potent an inhibitor as 20.
4. Kokotos, G. Synthesis 1990, 299. Rodoriguez, M.; Llinares,
M.; Doulut, S.; Heitz, A.; Martines, J. Tetrahedron Lett. 1991,
32, 923.
10. (a) Tozawa, R.; Nakamura, M. unpublisheddata. (b)
Petras, S. F.; Lindsey, S.; Harwood, Nr. H. J. J. Lipid Res.
1999, 40, 24.
5. Omura, K.; Swern, D. Tetrahedron 1978, 34, 1651.
6. Cohen, L. H.; Griffioen, A. M.; Wanders, R. J. A.; Van
Roermund, C. W. T.; Huysmans, C. M. G.; Princen, H. M. G.
Biochem. Biophys. Res. Commun. 1986, 138, 335.
7. Kuroda, M.; Endo, A. Biochim. Biophys. Acta 1977, 486, 70.
8. Removal of the 7-chloro atom of 1-isobutyl derivative 1b
causeda significant loss of activity (40% inhibition at 10 mM
against rat-derived enzymes). The 7-chloro atom was found to
be essential to exhibit inhibitory activity (unpublisheddata).
9. The assay conditions for the sodium salt 20 are the same as
those for the free acid 19a. Inhibitory activity of 20 is 2-fold
more potent than the racemate 2t. Although activity of 19a
11. Walch, D. A. Synthesis 1980, 677.
12. Sternbach, L. H.; Fryer, R. I.; Metlesics, W.; Sach, G.;
Stempel, A. J. Org. Chem. 1962, 27, 3781.
13. Mason, H. S. J. Am. Chem. Soc. 1947, 69, 2241.
14. Bogatskii, A. V.; Andronati, S. A.; Vostrova, L. N.; Lit-
vinova, L. A.; Yokubovskaya, L. N.; Yasinenko, N. E.; Iva-
nov, E. I.; Sharbatyan, P. A. Zh. Obshch. Khim. 1976, 46, 1893
Chem. Abstr., 1976, 85, 177288p.
15. Masuoka, Y.; Asako, T.; Goto, G.; Noguchi, S. Chem.
Pharm. Bull. 1986, 34, 140.
16. Davisson, V. J.; Woodside, A. B.; Poulter, C. D. Methods
Enzymol. 1995, 110, 130.