2-Azetidinone cholesterol absorption inhibitors: Increased potency by substitution of the C-4 phenyl ring

Article abstract of DOI:10.1016/S0968-0896(98)00073-X

SAR studies directed towards the optimization of 2-azetidinone cholesterol absorption inhibitors led to the discovery of 11a, the most potent cholesterol absorption inhibitor yet identified. Copyright (C) 1998 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(98)00073-X

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 1429±1437
2-Azetidinone Cholesterol Absorption Inhibitors:
Increased Potency by Substitution of the
C-4 Phenyl Ring
Wayne D. Vaccaro,* Rosy Sher and Harry R. Davis, Jr.
Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033-0539, USA
Received 14 November 1997; accepted 3 February 1998
AbstractÐSAR studies directed towards the optimization of 2-azetidinone cholesterol absorption inhibitors led to the
discovery of 11a, the most potent cholesterol absorption inhibitor yet identi®ed. # 1998 Elsevier Science Ltd. All rights
reserved.
Introduction
The 2-azetidinone 1a (Sch 48461) was identi®ed as a
potent cholesterol absorption inhibitor in the choles-
terol-fed hamster assay.1 Subsequently, 1a was shown to
reduce serum cholesterol in human trials.2 Compound
1a demonstrated a synergistic e€ect with HMG-CoA
reductase inhibitors in reducing plasma lipoprotein
levels in chow fed dogs and rhesus monkeys.3 Although
the mechanism of cholesterol absorption inhibition of
1a has yet to be determined, studies suggest that the site
of action is the intestinal wall.4
Chemistry
The 2-azetidinones reported in Table 1 were prepared
Previous SAR studies found that an alkoxy group at the
40-position of 2-azetidinones such as 1 is crucial for
in vivo activity.1 Attempts to change the position of
the 40-alkoxy group, introduce additional substituents
on to the 40-alkoxy substituted phenyl ring or replace
the 40-alkoxy group with other functional groups gen-
erally resulted in a loss of potency.5,6 However, the
20,40-dimethoxy substituted analogue 5 was determined
to be equipotent with 1. The biological activity of com-
pound 5 demanded further investigation of the e€ect
of 20,40-substitution on cholesterol absorption inhibi-
tion. The results of this investigation are reported
below.
with high trans diastereoselectivity by modi®cation of
the ketene-imine reaction as previously reported
(Scheme 1).7a,b Compounds displaying signi®cant bio-
logical activity were resolved by chiral HPLC on either
Chiracel1 OD or Chiralpak1 AS columns and retested
as single enantiomers.8 The biologically most active
enantiomer of each pair was assigned the 3R,4S-abso-
lute stereochemistry in analogy to 1a.9
Compounds reported in Table 2 were prepared as
shown in Scheme 2. The trans stereochemistry of 12 was
established by the ketene±imine reaction. Hydrolysis of
ester 12 a€ords the acid 13. Conversion of 13 to the
corresponding acid chloride and subsequent palladium
mediated coupling with an arylzinc reagent provided
ketone 14. Ketone 14 was resolved into its enantiomers
14a and 14b by HPLC on a Chiralpak1 AS column.
*Corresponding author.
0968-0896/98/$Ðsee front matter # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(98)00073-X

1430
W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
Compounds 14a and 14b were independently reduced
with a chiral oxazaborolidine catalyst derived from (S)-
diphenyl-2-pyrrolidinemethanol and borane dimethyl-
sul®de to set the stereochemistry at the 30-positions of
15a and 15b.10a,b Subsequent hydrogenation a€orded
the desired bisphenols 11a and 11b. The absolute
stereochemistry was assigned by analogy to compound
10 (Sch 58235).11
Table 1. Cholesterol absorption activity of analogues of 1
Cholesterol-fed Hamster12
Compd
R1
R2
R3
Ar
SCa
LCEb
Dose (mpk)c
ED50 (mpk)c
2.2
1(‹)
1a
1b
H
H
MeO
MeO
MeO
MeO
MeO
HO
H
H
4-MeOPh
4-MeOPh
4-MeOPh
Ph
29
43
0
78
93
0
50
10
50
10
10
10
50
50
10
3
H
H
2(‹)
3(‹)
4a
H
H
41
24
10
26
13
30
50
0
90
77
70
78
0
1.5
3.9
5.0
H
H
4-FPh
H
H
4-MeOPh
4-MeOPh
4-MeOPh
4-MeOPh
4-MeOPh
4-MeOPh
4-FPh
5(‹)
6(‹)
7(‹)
7a
MeO
MeO
HO
HO
HO
HO
HO
HO
HO
MeO
MeO
MeO
MeO
MeO
HO
H
MeO
H
95
97
31
95
0
H
0.22
0.37
0.08
7b
8a
H
3
H
34
0
10
10
10
10
8b
9a
HO
HO
H
4-FPh
Ph
Ph
H
45
14
94
69
9b
HO
H
a% Reduction serum cholesterol.
b% Reduction liver cholesterol esters.
cmpk=mg/Kg/day.
Table 2. Cholesterol absorption activity of analogues of 10
Cholesterol-fed Hamster12
Dose (mpk)c
Compd
Stereo
R1
R2
SCa
LCEb
ED50 (mpk)c
10
30(S),3(R),4(S)
30(S),3(R),4(S)
30(S),3(S),4(R)
H
OH
OH
OH
59
65
33
95
96
42
1
1
1
0.04
0.005
11a
11b
OH
OH
a% Reduction serum cholesterol.
b% Reduction liver cholesterol esters.
cmpk=mg/Kg/day.

W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
1431
Biological Results
achieve a 50% reduction in liver cholesterol ester levels,
were determined for select compounds.
Cholesterol absorption inhibition was assessed in orally
dosed, cholesterol-fed hamsters as previously descri-
bed.1 Both serum cholesterol and liver cholesterol ester
levels were determined. ED50s, the dose required to
In Table 1, the phenylpropyl substitution at C-3 was
held constant while the N-1 and C-4 substituents were
varied. Whereas the 20,40-dimethoxy substituted analogue
Scheme 1. Preparation of trans-2-azetidinones.
Scheme 2. Preparation of 11a and 11b.

1432
W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
5 was determined to be equipotent with 1, the sterically
congested 20,40,60-trimethoxy substituted compound 6
was inactive. Substitution of hydroxy for the 20-methoxy
group of 5 as in compound 7 resulted in an increase in
potency. In general, we found that incorporation of a
20-hydroxy group signi®cantly increased potency. 2-
Hydroxy substituted analogues 7a, 8a and 9a are almost
an order of magnitude more potent than compounds 1a,
2, 3, and 4a which lack a 20-hydroxy group.
H2SO4 (10 mL), H2O (190 mL) and heating on a hot
plate). Chromatography was performed on Selecto Sci-
enti®c 40 Micron Flash silica gel (32±63). SGFC: Silica
gel ¯ash chromataography.
trans-1-(4-Methoxyphenyl)-3-(phenylpropyl)-4-(4-methoxy-
phenyl)-2-azetidinones1,7 (1,1a,1b), trans 1-(4-methoxy-
phenyl)-3-(phenylpropyl)-4-(4-hydroxy-phenyl)-2-azetidi-
nones13 (4a) and trans-1-(4-¯uorophenyl)-3(S)-[3(S)-
hydroxy-3-(4-¯uoro-phenyl)propyl]-4(R)-(4-hydroxyphenyl)-
2-azetidinone11 (10). These compounds were prepared as
previously described.
During the course of these studies the potent cholesterol
absorption inhibitor compound 10 (Sch 58235) was dis-
covered (Table 2).11 We were delighted to ®nd that the
addition of a 20-hydroxy group to compound 10
(Sch 58235) as in compound 11a increased potency
eightfold.11
Method 1: Preparation of aldehydes
2,4-Dibenzyloxybenzaldehyde (16). Benzylbromide (45.2
mL, 380 mmol) was added to a mixture of 2,4-dihydroxy
benzaldehyde (25 g, 181 mmol) and potassium carbonate
(75 g, 543 mmol) in acetone (300 mL). The mixture was
re¯uxed overnight. TLC (25% EtOAc/hexanes) indicated
consumption of starting material. The mixture was ®ltered
through Celite and the ®lter cake was well washed with
acetone. The ®ltrate was concentrated. The resulting resi-
due was redissolved in ethyl acetate, transferred to a
separatory funnel, washed with sodium carbonate (satd),
HCl (3 N) and brine, dried over anhydrous sodium sulfate
and concentrated to give 60.5 g, (ꢀ100%) of the title com-
pound as a solid of sucient purity to be used in the sub-
sequent step. 1H NMR (CDCl3, 400 MHz) d 5.10 (2H, s),
5.13 (2H, s), 6.59 (1H, d, J=2 Hz), 6.64 (1H, dd, J=2,
9 Hz), 7.39 (10H, m), 7.84 (1H, d, J=9 Hz), 10.4 (1H, s).
MS (CI): M+H, 319 (11), 89 (100).
Conclusion
The addition of a 20-hydroxy group to the C-4 phenyl
ring of 2-azetidinone cholesterol absorption inhibitors
increases potency, typically by an order of magnitude.
This study suggests that 20,40-dihydroxyphenyl is the
preferred C-4 substituent of 2-azetidinone cholesterol
absorption inhibitors as evidenced by compounds 9a
and 11a. We were grati®ed that the investigation of
20,40-substitution led to the discovery of 11a, the most
potent 2-azetidinone cholesterol absorption inhibitor yet
identi®ed. 11a is 440-fold more potent in the cholesterol-
fed hamster model than our initial lead compound 1a
(Sch 48461). However, since we are presently restricted
to an in vivo assay, interpretation of the cholesterol
absorption activity of the compounds in the tables may
not be straight forward. The observed cholesterol
absorption inhibition may be due to a compound's
bioavailability and/or ease of conversion to active
metabolites and not its intrinsic cholesterol absorption
inhibition activity.
In a similar fashion the following compounds were
prepared
2-Methoxy-4-benzyloxybenzaldehyde (17). Prepared
from 2-methoxy-4-hydroxybenzaldehyde. Yield: 88%.
1H NMR (CDCl3, 400 MHz) d 3.89 (3H, s), 5.14 (2H, s),
6.54 (1H, s), 6.62 (1H, dd, J=2, 9 Hz), 7.41 (5H, m),
7.82 (1H, d, J=9 Hz), 10.3 (1H, s). MS (CI): M+H, 243
(100). Puri®cation: SGFC: 25% EtOAc/hexane
TLC:(25% EtOAc/hexane, Rf=0.43), UV.
Experimental
Reagents were used as received unless stated otherwise.
Anhydrous solvents were purchased from Aldrich Che-
mical Company. Melting points were taken on a Tho-
mas Hoover Capillary Melting Point apparatus and are
uncorrected. 1H NMR spectra were taken on Varian
200, 300, 400 and 500 spectrometers and are reported in
ppm down®eld from internal tetramethylsilane. Ele-
mental analyses were carried out on Leemann Labs
CE440 elemental analyzer or a Fissions EA 1108
CHNS. IR Spectra were obtained on Perkin±Elmer
1320 or Nicolet MA-1 FT infrared spectrophotometers.
2-Benzyloxy-4-methoxybenzaldehyde (18). Prepared
from 2-hydroxy-4-methoxybenzaldehyde. Yield: 88%.
1H NMR (CDCl3, 400 MHz) d 3.85 (3H, s), 5.17 (2H, s),
6.51 (1H, s), 6.57 (1H, dd, J=2, 9 Hz), 7.41 (5H, m),
7.85 (1H, d, J=9 Hz), 10.4 (1H, s). MS (CI): M+H, 243
(29), 91 (100). Puri®cation: SGFC: 20% EtOAc/hexane
TLC:(25% EtOAc/hexane, Rf=0.36), UV.
TLC was performed on Merck silica plates (60F F254
,
Method 2: Preparation of imines
250 mm) and are reported as Rf, (solvent system used to
develop plate), method of visualization UV, Ce stain
(dipping TLC plate into a mixture of CeSO4 (4 g) concd
N-(2-Methoxy-4-benzyloxybenzylidene)-4-¯uoroaniline
(19). A mixture of 2-methoxy-4-benzyloxybenzaldehyde

W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
1433
17 (10.12 g, 41.8 mmol) and 4-¯uoroaniline (3.96 mL,
41.8 mmol) in toluene (300 mL) was re¯uxed with azeo-
tropic removal of water via a Dean±Strark trap for 40 h.
NMR analysis of a small aliqout indicated imine for-
mation was ꢀ90% complete. The mixture was cooled to
room temperature and concentrated in vacuo to give a
brown±yellow solid. Recrystallizaion from ethyl acet-
ate:hexane provide 10.5 g (75%) of the title compound
as a yellow solid. 1H NMR (CDCl3, 400 MHz) d 3.86
(3H, s), 5.13 (2H, s), 6.56 (1H, d, J=2 Hz), 6.68 (1H, d,
J=9 Hz), 7.06 (2H, m), 7.20 (2H, m), 7.42 (5H, m), 8.19
(1H, d, J=9 Hz), 8.80 (1H, s). MS (CI): M+H, 336 (27),
257 (68), 75 (100).
N-(2,4-dimethoxybenzylidene)-4-methoxyaniline (25). Pre-
pared from 2,4-dimethoxybenzaldehyde and 4-methoxy-
aniline. Yield: 100%, yellow solid. 1H NMR (CDCl3,
300 MHz) d 3.88 (3H, s), 3.93 (3H, s), 3.94 (3H, s), 6.54
(2H, d, J=2Hz), 6.67 (1H, m), 6.98 (2H, d, J=9 Hz), 7.30
(2H, d, J=9 Hz), 8.20 (1H, d, J=9 Hz), 8.90 (1H, s). Used
without further puri®cation.
Method 3: Preparation of imines
N-(2,4-Dibenzyloxybenzylidene)-4-¯uoroaniline (26). A
mixture of 2,4-dibenzyloxybenzaldehyde 16 (57.6 g,
181 mmol) and 4-¯uoroaniline (17.2 mL, 181 mmol) in
isopropanol (400 mL) was heated to re¯ux, cooled to
room temperature, and diluted with hexanes and
allowed to stand overnight. The resulting precipitate
was collected via vacuum ®ltration, washed with cold
hexanes and dried under vacuum to give 70.4 g (95%) of
a gray/white solid. 1H NMR indicates the solid con-
tained ꢀ20% isopropanol. The solid was recrystallized
from ethyl acetate/hexanes to provide 52 g (70%) of the
In a similar fashion the following imines were prepared.
N-(2-Benzyloxy-4-methoxybenzylidene)-4-¯uoroaniline (20).
Prepared from 2-benzyloxy-4-methoxybenzaldehyde 18
1
and 4-¯uoroaniline. Yield: 100%, light-brown solid. H
NMR (CDCl3, 400 MHz) d 3.85 (3H, s), 5.15 (2H, s),
6.53 (1H, d, J=2 Hz), 6.62 (1H, d, J=9 Hz), 7.05 (2H,
m), 7.22 (4H, m), 7.40 (3H, m), 8.18 (1H, bs), 8.84 (1H,
s). MS (CI): M+H, 336 (39), 257 (96), 91 (100).
1
title compound free of isopropanol. H NMR (CDCl3,
400 MHz) d 5.10 (2H, s), 5.11 (2H, s), 6.61 (1H, d,
J=2 Hz), 6.68 (1H, dd, J=2, 9 Hz), 7.04 (2H, m), 7.16
(2H, m), 7.38 (10H, m), 8.13 (1H, m), 8.83 (1H, s). MS
(CI): M+H, 412 (100).
N-(2-Benzyloxy-4-methoxybenzylidene)-aniline (21). Pre-
pared from 2-benzyloxy-4-methoxybenzaldehyde 18 and
aniline. Yield: 85%, light-brown solid. 1H NMR
(CDCl3, 400 MHz) d 3.85 (3H, s), 5.15 (2H, s), 6.54 (1H,
d, J=2 Hz), 6.62 (1H, d, J=9 Hz), 7.21 (2H, m), 7.39
(8H, m), 8.22 (1H, bs), 8.87 (1H, s). MS (CI): M+H, 336
(39), 257 (96), 91 (100).
Method 4: Preparation of ꢀ-lactams via the ketene±imine
condensation
trans-1-(4-Fluorophenyl)-3-(phenylpropyl)-4-(4-methoxy-
2-benzyloxyphenyl)-2-azetidinone (27). 5-Phenylvaleryl
chloride (5.5 mL, 5.5 mmol, 1 M in toluene,) was added
to a room temperature solution of N-(2-methoxy-4-
benzyloxybenzylidene)-4-¯uoroaniline 19 (1.23 g, 3.67
mmol), n-tributylamine (1.75 mL, 7.33 mmol) and anhy-
drous toluene (50 mL). The mixture was heated to re¯ux
overnight. TLC (25% ethyl acetate/hexanes) indicated
consumption of starting materials. The mixture was
cooled to room temperature, 1 M HCl was added, the
resulting mixture was stirred for 15 min, transferred to a
separatory funnel, diluted with ethyl acetate, washed
with 1 M HCl, NaHCO3 (satd), water and brine, dried
over anhydrous sodium sulfate, and concentrated to an
oil. The residue was redissolved in 50% THF/MeOH,
cooled to 0 ꢁC and treated with sodium borohydride
(0.28 g, 7.5 mmol) to remove any aldehyde liberated
upon workup of unreacted imine. The aldehydes tend to
have Rfs similar to 2-azetidinones, reduction to the
alcohol usually simpli®es puri®cation. After 30 min,
NH4Cl (satd) was added. The mixture was transferred
to a separatory funnel and extracted with ethyl acetate.
The extracts were combined, washed with water and
brine, dried over anhydrous sodium sulfate and con-
centrated onto enough silica gel such that a free ¯owing
powder was obtained. The resulting powder was loaded
N-(2,4-Dibenzyloxybenzylidene)-aniline (22). Prepared
from 2,4-dibenzyloxybenzaldehyde 16 and aniline.
Yield: 83%, yellow solid. 1H NMR (CDCl3, 400 MHz)
d 5.11 (2H, s), 5.12 (2H, s), 6.62 (1H, d, J=2 Hz), 6.70
(1H, dd, J=2, 9 Hz), 7.20 (3H, m), 7.39 (12H, m), 8.19
(1H, bs), 8.87 (1H, s). Puri®cation: recrystallized from
ethyl acetate/hexanes.
N-(2-benzyloxy-4-methoxybenzylidene)-4-methoxyaniline
(23). Prepared from 2-benzyloxy-4-methoxybenzaldehye
1
18 and 4-methoxyaniline. Yield: 100%, yellow solid. H
NMR (CDCl3, 400 MHz) d 3.82 (3H, s), 3.85 (3H, s),
5.15 (2H, s), 6.53 (1H, d, J=2 Hz), 6.63 (1H, d, J=2,
9 Hz), 6.91 (2H, d, J=9 Hz), 7.22 (2H, d, J=9 Hz), 7.41
(5H, m), 8.21 (1H, bs), 8.90 (1H, s). Used without fur-
ther puri®cation.
N-(2,4,6-trimethoxybenzylidene)-4-methoxyaniline (24).
Prepared from 2,4,6-trimethoxybenzaldehyde and 4-
methoxyaniline. Yield: 100%, yellow solid. 1H NMR
(CDCl3, 300 MHz) d 3.88 (3H, s), 3.92 (3H, s), 3.94
(6H, s), 6.22 (2H, s), 6.97 (2H, d, J=9 Hz), 7.28
(2H, d, J=9 Hz), 8.83 (1H, s). Used without further
puri®cation.

1434
W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
onto a chromatography column prepacked with silica
gel and 20% ethyl acetate/hexane. Elution with the
same solvent provided 1.5 g (83%) of the title com-
hydrogenation of 29, 29a, and 29b, as described in
method 5, respectfully provided 7: Yield:100%, yellow
oil, 7a: Yield: 100%, foam and 7b: Yield 100%, oil.
TLC: (25% EtOAc/hexane, Rf=0.16, UV) 1H NMR
(CDCl3, 400 Mhz) d 1.85 (3H, m), 1.96 (1H, m), 2.64
(2H, t, J=7 Hz), 3.16 (1H, m), 3.74 (3H, s), 3.75 (3H, s),
4.93 (1H, d, J=2 Hz), 6.35 (1H, d, J=2 Hz), 6.44 (1H,
dd, J=2, 8 Hz), 6.78 (2H, d, J=9 Hz), 7.10 (1H, d,
J=9 Hz), 7.17 (2H, m), 7.25 (5H, m). HRMS (FAB):
M+H, C26H27NO4, calcd 417.1940, 7 found 417.1941,
7a found 417.1937, 7b found 417.1933.
1
pound. TLC: (25% EtOAc/hexane, Rf=0.45), UV. H
NMR (CDCl3, 400 MHz) d 1.86 (3H, m), 1.95 (1H, m),
2.65 (2H, t, J=7.0 Hz), 3.09 (1H, m), 3.80 (3H, s), 4.95
(1H, d, J=2 Hz), 5.03 (2H, s), 6.50 (1H, dd, J=2, 9 Hz),
6.57 (1H, d, J=2 Hz), 6.94 (2H, m), 7.06 (1H, d,
J=9 Hz), 7.19 (3H, m), 7.27 (5H, m), 7.39 (4H, m). MS
(CI): M+H, 496 (100).
Method 5: Preparation of phenols by debenzylation
trans-1-(4-Methoxyphenyl)-3-(phenylpropyl)-4-(2,4,-di-
methoxyphenyl)-2-azetidinone (5). Prepared from N-
(2,4-dimethoxybenzylidene)-4-methoxyaniline 25 and 5-
phenylvaleryl chloride as described in method 4. Yield:
88%, yellow oil. 1H NMR (CDCl3, 300 Mhz) d 1.93
(4H, m), 2.63 (2H, m), 3.18 (1H, m), 3.82 (3H, s), 3.85
(3H, s), 3.87 (3H, s), 4.94 (1H, d, J=2 Hz), 6.37 (1H, d,
J=2 Hz), 6.45 (1H, m), 6.79 (2H, d, J=9 Hz), 7.08 (1H,
d, J=8 Hz), 7.21 (8H, m). HRMS (EI): M+,
C27H30NO4, calcd 431.2097, found 431.2075.
trans-1-(4-Fluorophenyl)-3-(phenylpropyl)-4-(4-methoxy-
2-hydroxyphenyl)-2-azetidinone (28). trans-1-(4-Fluoro-
phenyl)-3-(phenylpropyl)-4-(4-methoxy-2-benzyloxy-
phenyl)-2-azetidinone 27 (0.16 g, 0.32 mmol) was
dissolved in ethyl acetate (15 mL) and diluted with
methanol (15 mL) and purged with nitrogen. Ten percent
palladium on carbon (0.016 g) was added, the resulting
mixture was purged with hydrogen and stirred under a
balloon of hydrogen overnight. TLC (25% ethyl acet-
ate/hexanes) indicated consumption of starting material.
The mixture was ®ltered through celite, and the ®lter-
cake was well washed with 50% THF/MeOH. The ®l-
trate was concentrated to provide 0.121 g (92%) of the
title compound as a yellow foam. TLC: (25% EtOAc/
trans-1-(4-Methoxyphenyl)-3-(phenylpropyl)-4-(2,4,6-tri-
methoxyphenyl)-2-azetidinone (6). Prepared from N-
(2,4,6-trimethoxybenzylidene)-4-methoxyaniline 24 and
5-phenylvaleryl chloride as dexcribed in method 4.
1
1
hexane, Rf=0.17, UV). H NMR (CDCl3, 400 MHz) d
Yield: 30%, orange oil. H NMR (CDCl3, 300 Mhz) d
1.85(3H, m), 1.96 (1H, m), 2.64 (2H, t, J=7.2 Hz), 3.10
(1H, m), 3.79 (3H, s), 4.94 (1H, d, J=2 Hz), 6.34 (1H,
dd, J=2, 8 Hz), 6.45 (1H, d, J=2 Hz), 6.95 (3H, m),
7.17 (3H, m), 7.26 (4H, m). HRMS (FAB): M+H,
C23H25NO3F, calcd 406.1818, found 406.1806. The fol-
lowing compound were prepared by combination of the
above methods as indicated.
1.85 (3H, m), 1.96 (1H, m), 2.69 (2H, t, J=8 Hz), 3.78
(1H, m), 3.84 (6H, s), 3.85 (3H, s), 5.25 (1H, d,
J=2 Hz), 6.16 (2H, s), 6.80 (2H, d, J=9 Hz), 7.27 (10H,
m). HRMS (EI): M+, C28H31NO5, calcd 461.2202,
found 461.2210.
trans-1-(4-Fluorophenyl)-3-(phenylpropyl)-4-(2,4-dihydroxy-
phenyl)-2-azetidinone (8a and 8b). Ketene±imine con-
densation of N-(2,4-dibenzyloxybenzylidene)-4-¯uoro-
aniline 26 and 5-phenylvaleryl chloride as described in
method 4 provided trans-1-(4-¯uorophenyl)-3-(phenyl-
trans-1-(4-Methoxyphenyl)-3-(phenylpropyl)-4-(4-methoxy-
2-hydroxy-phenyl)-2-azetidinone (7, 7a, and 7b). trans-1-
(4-Methoxyphenyl)-3-(phenylpropyl)-4-(4-methoxy-2-
benzyloxyphenyl)-2-azetidinone 29 was prepared from
N-(2-benzyloxy-4-methoxybenzylidene)-4-methoxyaniline
23 and 5-phenylvaleryl chloride as described in method
4. Yield: 93% oil. Puri®cation: SGFC: 30% EtOAc/
hexane TLC: (30% EtOAc/hexane, Rf=0.3), Ce stain,
UV. Yield: 93%. 1H NMR (CDCl3, 400 Mhz) ( 1.77
(4H, m), 2.52 (2H, t, J=7 Hz), 3.06 (1H, m), 3.75 (3H,
s), 3.77 (3H, s), 5.01 (1H, d, J=2 Hz), 5.08 (2H, dd,
J=12, 20 Hz), 6.43 (1H, dd, J=2, 9 Hz), 6.57 (1H, d,
J=2 Hz), 6.78 (2H, dd, J=2, 7 Hz), 7.08 (3H, m), 7.20
(5H, m), 7.38 (5H, m). HRMS (FAB): M+, C33H33NO4,
calcd 507.241, found 507.2395. The enantiomers of 29
were resolved on a preparative Chiracel OD column
(3% isopropanol/hexanes, 65 mL/min). Enantiomeric
purity was veri®ed on a analytical Chiracel AS column
propyl)-4-(2,4-dibenzyloxyphenyl)-2-azetidinone
30.
Yield: 56%. 1H NMR (CDCl3, 400 Mhz) d 1.82 (4H, m),
2.52 (2H, m), 3.10 (1H, m), 5.07 (5H, m), 6.54 (1H, d),
6.61 (1H, s), 6.66 (1H, s), 6.95 (2H, m), 7.07 (2H, m),
7.25 (5H, m), 7.39 (10H, m). MS (CI), M+H: 572 (100).
Puri®cation: SGFC: 18% EtOAc/hexane. TLC:(25%
EtOAc/hexane, Rf=0.47), UV. 30 was resolved on a
preparative Chiracel1 OD column (10% isopropanol/
hexanes,
70 mL/min).
30a
Rt=44.5 min.
30b
Rt=52.8 min. Subsequent idependent hydrogenation of
each enantiomer as described in method 5 provided 8a:
Yield: 100%, yellow foam and 8b: yield: 100%, oil.
TLC:(25% EtOAc/hexane, Rf=0.10, UV). 1H NMR
(CDCl3, 400 Mhz) d 1.83 (3H, m), 1.96 (1H, m), 2.62
(2H, t, J=7 Hz), 3.19 (1H, m), 4.91 (1H, d, J=2 Hz),
6.31 (1H, s), 6.35 (1H, dd, J=2, 8 Hz), 6.91 (2H, m),
7.02 (1H, d, J=8 Hz), 7.15 (3H, m), 7.26 (4H, m).
(5%
Rt=39.7 min,
isopropanol/hexane,
1.0 mL/min),
29a
Independent
29b Rt=51.1 min.

W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
1435
HRMS (FAB): M+H, C24H23NO3F, calcd 392.1662, 8a:
found 392.1661, 8b: found 392.1657.
and added to a room temperature solution of trans-
methyl 3-(3-[2-oxo-4-(2,4-dibenzyloxyphenyl)-1-(4-¯uoro-
phenyl)-azetidinyl]propanoate 12 (8.8 g, 16.3 mmol) in
THF (60 mL). TLC (30% EtOAc/hexanes) after ꢀ6 h
indicated consumption of starting material. The reac-
tion was quenched with HCl (1 M), transferred to a
separatory funnel, diluted with ethyl acetate, washed
with HCl (1 M), water and brine, dried over anhydrous
sodium sulfate, and concentrated to give 8.66 g of the
title compound of sucient purity to be used without
trans-1-Phenyl-3-(3-phenylpropyl)-4-(2,4-dihydroxyphenyl)-
2-azetidinone (9a and 9b). Ketene±imine condensation of
N-(2,4-dibenzyloxybenzylidene)-aniline 22 and 5-phen-
ylvaleryl chloride as described in method 4 provided
trans-1-phenyl-3-(3- phenylpropyl) - 4 - (2, 4 - dibenzyloxy-
1
phenyl)-2-azetidinone 31. Yield: 60%. H NMR (CDCl3,
400 Mhz) d 1.81 (4H, m), 2.52 (2H, t, J=7 Hz), 3.09 (1H,
m), 5.07 (5H, m), 6.51 (1H, dd, J=2, 9 Hz), 6.67 (1H, s),
7.23 (21H, m). MS (CI), M+H: 554 (100). Puri®cation:
SGFC: 18% EtOAc/hex. TLC:(25% EtOAc/hexane,
Rf=0.48), UV. 31 was resolved on a preparative Chir-
acel OD column (7% isopropanol/hexanes). Enantio-
meric purity was veri®ed on an analytical Chiracel1 OD
column (7% isopropanol/hexanes, 1.0 mL/min) 31a
Rt=18.49 min. 31b Rt=21.65 min. Subsequent indepen-
dent hydrogenation of each enantiomer as described in
method 5 provided 9a: Yield: 100%, oil and 9b: Yield
100%, oil. TLC: (25% EtOAc/hexane, Rf=0.20, UV).
1H NMR (CDCl3, 400 Mhz) d 1.86 (3H, m), 1.97 (1H,
m), 2.64 (2H, t, J=7 Hz), 3.16 (1H, m), 4.95 (1H, d,
J=2 Hz), 6.34 (2H, m), 7.03 (2H, m), 7.17 (2H, m), 7.26
(7H, m). HRMS (FAB): M+H, C24H24NO3, calcd
374.1756, 9a: found 374.1759, 9b: found 374.1751.
1
further puri®cation. H NMR (CDCl3, 400 Mhz) d 2.13
(2H, m), 2.54 (2H, m), 3.12 (1H, dt, J=2, 6 Hz), 4.99
(1H, d, J=2 Hz), 5.01 (2H, s), 5.09 (2H, s), 6.50 (1H, dd,
J=2, 9 Hz), 6.66 (1H, d, J=2 Hz), 6.94 (2H, app. t,
J=9 Hz), 7.05 (1H, d, J=8 Hz), 7.24 (2H, m), 7.36
(10H, m). MS (FAB): M+H 526 (100).
trans-1-(4-Fluorophenyl)-3-[3-oxo-3-(4-¯uorophenyl)pro-
pyl]-4-(2,4-dibenzyloxyphenyl)-2-azetidinone (14, 14a, and
14b). Oxalyl chloride (1.85 mL, 21.2 mmol) was added
to a 0 ꢁC solution of the trans-3-(3-[2-oxo-4-(2,4-diben-
zyloxyphenyl)-1-(4-¯uorophenyl)-azetidinyl]propanoic
acid 13 (8.57 g, 16.3 mmol) in methylene chloride
(50 mL). The mixture was allowed to warm to room
temperature overnight. The mixture was concentrated in
vacuo and dried under vacuum. The residue was redis-
solved in THF (50 mL), cooled to 0 ꢁC and transferred
via cannula to a 0 ꢁC solution prepared by the sequential
addition 4-¯uorophenylmagnesium bromide (24.5 mL,
24.5 mmol, 1 M in THF, Aldrich) and palladium (II)
trans-Methyl-3-(3-[2-oxo-4-(2,4-dibenzyloxyphenyl)-1-(4-
¯uorophenyl)-azetidinyl]propanoate (12). Methyl glutaryl
chloride (4.0 mL, 29 mmol) was added to a solution of
the N-(2,4-dibenzyloxybenzylidene)-4-¯uoroaniline 26
(7.95 g, 19.3 mmol) and n-tributylamine (9.2 mL,
38.6 mmol) in anhydrous toluene (200 mL) and the
mixture was re¯uxed overnight. the mixture was cooled
to room temperature, quenched with HCl (1 M), rapidly
stirred for 15 min, transferred to a separatory funnel,
diluted with ethyl acetate, washed with HCl (1 M),
sodium bicarbonate (satd), water and brine, dried over
anhydrous sodium sulfate, and concentrated to give
12.2 g of a dark oil. The oil was redissolved in ethyl
acetate, concentrated onto enough ¯ash grade silica gel
such that a free ¯owing powder was obtained. The
powder was loaded onto a chromatography column
packed with silica gel and 25% ethyl acetate/hexanes.
Elution with the same solvent gave 8.8 g (84%) of the
title compound. 1H NMR indicates the product is all
tetrakistriphenylphosphine (0.57 g, 0.49 mmol) to
a
solution of anhydrous zinc chloride (4.0 g, 29.3 mmol) in
THF (50 mL). TLC (2.5% EtOH/0.5% HOAc/CH2Cl2)
after 2 h indicated consumption of starting acid. The
reaction was quenched with HCl (2 M), transferred to a
separatory funnel, diluted with ethyl acetate, washed
with HCl (2 M), water and brine, dried over anhydrous
sodium sulfate, and concentrated onto enough ¯ash
grade silica gel such that a free ¯owing powder was
obtained. The powder was loaded onto a chromatography
column packed with silica gel and 20% ethyl acetate/
hexanes. Elution with the same solvent gave 1.05 g
(62%) of the title compound. 14 was resolved on a pre-
parative Chiralpak1 AS column (40% isopropanol/
hexanes) to provide trans-1-(4-¯uorophenyl)-3(S)-[3-
oxo-3-(4-¯uorophenyl)propyl]-4(R)-(2,4-dibenzyloxy-
phenyl)-2-azetidinone 14b (0.46 g), further elution with
100% isopropanol failed to elute trans-1-(4-¯uoro-
phenyl)-3(R)-[3-oxo-3-(4-¯uorophenyl)propyl]-4(S)-(2,4-
dibenzyloxyphenyl)-2-azetidinone 14a. Finally elution
with 100% ethanol provided enatiomer 14a (0.42 g).
Enantiomeric purity was determined on a Chiralpak1
AS analytical column (40% isopropanol/hexanes,
0.8 mL/min). Enantiomer 14b Rt=15.7 min, enantiomer
14a Rt=49.8 min and on an analytical Chiracel1 OD
column (30% isopropanol/hexanes, 1.0 mL/min).
1
trans. H NMR (CDCl3, 400 Mhz) d 2.14 (2H, m), 2.49
(2H, m), 3.11 (1H, dt, J=2, 8 Hz), 3.61 (3H, s), 4.99
(1H, d, J=2 Hz), 5.01 (2H, s), 5.10 (2H, s), 6.51 (1H, dd,
J=2, 8 Hz), 6.66 (1H, d, J=2 Hz), 6.93 (2H, app. t,
J=9 Hz), 7.06 (1H, d, J=9 Hz), 7.23 (2H, m), 7.38
(10H, m). MS (FAB): M+H 540 (100).
trans-3-(3-[2-Oxo-4-(2,4-dibenzyloxyphenyl)-1-(4-¯uoro-
phenyl)-azetidinyl]propanoic acid (13). Lithium hydrox-
ide (0.82 g, 19.6 mmol) was dissolved in water (20 mL)

1436
W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
Enantiomer 14b Rt=32.8 min, enantiomer 14a
Rt=23.2 min. (Note reversal of elution of enantiomers
14a and 14b.) The Chiracel1 OD column gave better
peak shape for the determination of enantiomeric purity
and the Chiralpak1 AS column gave better separation
fashion to provide 15b 0.52 g, (35%). 1H NMR (CDCl3,
400 MHz) d 1.85 (4H, m), 3.11 (1H, m), 4.57 (1H, m),
5.00 (1H, d, J=2 Hz), 5.02 (2H, s), 5.07 (2H, m), 6.52
(1H, dd, J=2, 8 Hz), 6.67 (1H, d, J=2 Hz), 6.95 (4H,
m), 7.07 (1H, d, J=9 Hz), 7.17 (2H, m), 7.23 (2H, m),
7.38 (10H, m).
1
of the two enantiomers. H NMR (CDCl3, 400 Mhz) d
2.22 (1H, m), 2.32 (1H, m), 3.16 (3H, m), 5.01 (2H, s),
5.05 (1H, d, J=2 Hz), 5.09 (2H, s), 6.51 (1H, dd, J=2,
9 Hz), 6.66 (1H, d, J=2 Hz), 6.94 (2H, app. t, J=9 Hz),
7.08 (3H, m), 7.32 (12H, m), 7.90 (2H, dd, J=6, 8 Hz).
MS (FAB): M+H 604 (100).
trans-1-(4-Fluorophenyl)-3(R)-[3(S)-hydroxy-3-(4-¯uoro-
phenyl)propyl]-4(S)-(2,4-dihydroxyphenyl)-2-azetidinone
(11a). A suspension of trans-1-(4-¯uorophenyl)-3(R)-
[3(S)-hydroxy-3-(4-¯uorophenyl)propyl]-4(S)-(2,4-di-
benzyloxyphenyl)-2-azetidinone 15a (0.75 g, 1.24 mL)
and 10% Pd/C (0.08 g) in a mixture of methanol
(20 mL) and ethyl acetate (20 mL) was hydrogenated on
a Parr apparatus at 60 pi overnight. TLC (50% ethyl
acetate/hexane's) indicated on a trace of product. The
mixture was ®ltered through Celite, concentrated, redis-
solved in methanol (20 mL) and ethyl acetate (20 mL).
10% Pd/C (0.18 g) was added and the mixture was
hydrogenated on a Parr apparatus at 60 psi overnight.
TLC indicated consumption of 15a and formation of
one major spot. The mixture was ®ltered through Celite,
the ®lter cake was well washed with 50% MeOH/
CH2Cl2. The ®ltrate was concentrated onto enough
¯ash grade silica gel such that a free ¯owing powder was
obtained. The powder was loaded onto a chromato-
graphy column packed with silica gel and 50% ethyl
acetate/hexanes. Elution with 50±75% ethyl acetate/
hexanes gave 0.39 g (74%) of the title compound as a
clear foam. 1H NMR (CDCl3/CD3OD, 400 Mhz) d 1.92
(4H, m), 3.10 (1H, m), 4.64 (1H, m), 4.96 (1H, d,
J=2 Hz), 6.28 (1H, dd, J=2, 8 Hz), 6.33 (1H, d,
J=2 Hz), 6.94 (5H, m), 7.26 (4H, m). HRMS (FAB):
M+H, C24H22NO4F2, calcd 426.1517, found 426.1503.
trans-1-(4-Fluorophenyl)-3(R)-[3(S)-hydroxy-3-(4-¯uoro-
phenyl)propyl]-4(S)-(2,4-dibenzyloxyphenyl)-2-azetidinone
(15a). Trimethylboroxine (0.2 mL, 1.4 mmol) was added
to a solution of (S)-a,a-diphenyl-2-pyrrolidinemethanol
(0.18 g, 0.71 mmol) in anhydrous toluene (3 mL). The
mixture was stirred for 30 min and a white precipitate
formed. A short path distillation head was added to the
reaction ¯ask, and the toluene was distilled under
nitrogen. Toluene (4 mL) was added to the reaction
¯ask and was again distilled. This process was repeated
and the residue was dried under vacuum overnight. The
white residue was redissolved in anhydrous toluene (8mL)
to yield a 0.09 M solution of oxazaborolidine catalyst.
The oxazaborolidine catalyst (4 mL, 0.09 M in toluene)
was added to a 0 ꢁC solution of trans-1-(4-¯uoro-
phenyl)-3(R)-[3(S)-hydroxy-3-(4-¯uorophenyl)propyl]-
4(S)-(2,4-dibenzyloxyphenyl)-2-azetidinone 14a (1.37 g,
2.26 mmol) in THF (15 mL). Borane dimethylsul®de
(1.19 mL, 2 M in THF) was added slowly via syringe
over 40 min. Reaction color changes from a clear yellow
to a clear gray/green. TLC (30% EtOAc/hexanes) after
an additional 15 min indicates the reduction is complete.
The reaction was quenched with methanol and allowed
to stir for 30 min (gas evolution), transfered to a
separatory funnel, diluted with ethyl acetate, washed
with HCl (1 M), sodium bicarbonate (satd), water and
brine, dried over anhydrous sodium sulfate, and con-
centrated onto enough ¯ash grade silica gel such that a
free ¯owing powder was obtained. The powder was
loaded onto a chromatography column packed with
silica gel and 30% ethyl acetate/hexanes. Elution with
the same solvent gave 1.21 g (88%) of the title com-
trans-1-(4-Fluorophenyl)-3(S)-[3(S)-hydroxy-3-(4-¯uoro-
phenyl)propyl]-4(R)-(2,4-dihydroxyphenyl)-2-azetidinone
(11b). In a similar manner enantiomer 15b was hydro-
1
genated to give 11b 0.237 g (67%) of a clear foam. H
NMR (CDCl3/CD3OD, 400 Mhz) d 1.78 (1H, m), 1.95
(3H, m), 3.09 (1H, m), 4.62 (1H, app. t, J=6 Hz), 4.94
(1H, d, J=2 Hz), 6.30 (1H, dd, J=2, 8 Hz), 6.33 (1H, d,
J=2 Hz), 6.94 (5H, m), 7.24 (4H, m). HRMS (FAB):
M+H, C24H22NO4F2, calcd 426.1517, found 426.1516.
1
pound as a clear foam. H NMR (CDCl3, 400 Mhz) d
Acknowledgements
1.84 (4H, m), 3.07 (1H, m), 4.57 (1H, m), 4.99 (1H, d,
J=2 Hz), 5.02 (2H, s), 5.08 (2H, m), 6.51 (1H, dd, J=2,
8 Hz), 6.67 (1H, d, J=2 Hz), 6.95 (4H, m), 7.06 (1H, d,
J=9 Hz), 7.18 (2H, m), 7.23 (2H, m), 7.38 (10H, m).
HRMS (FAB): M+H, C38H34NO4F2, calcd 606.2456,
found 606.2451.
We wish to thank Dr S. Dugar, L.-Y. Chen, Dr D.
Burnett, Dr J. Clader, Dr P. Das, Ms L. Hoos, Mr D.
McGregor and Dr S. Rosenblum for their assistance
and helpful discussions.
References and Notes
trans-1-(4-Fluorophenyl)-3(S)-[3(S)-hydroxy-3-(4-¯uoro-
phenyl)propyl]-4(R)-(2,4-dibenzyloxyphenyl)-2-azetidinone
(15b). Reduction of 14b was carried out in a similar
1. Burnett, D. A.; Caplan, M. A.; Davis, Jr. H. R.; Burrier, R.
E.; Clader, J. W. J. Med. Chem. 1994, 37, 1733.

W. D. Vaccaro et al./Bioorg. Med. Chem. 6 (1998) 1429±1437
1437
2. Bergman, M.; Morales, H.; Mellars, L.; Kosoglou, T.; Bur-
8. Burnett, D. A. Tetrahedron Lett. 1994, 35, 7339.
rier, R.; Davis, H. R.; Sybertz, E. J.; Pollare, T. 12th Interna-
tional Symposium on Drugs A€ecting Lipid Metabolism,
Houston, Texas, November 7±10, 1995.
9. Typically the more active enantiomer (3R,4S) elute ®rst on
the Chiracel1 OD column. The order is reversed on the
Chiralpak1 AS column.
3. Davis, Jr. H. R.; Watkins, R. W.; Salisbury, B. G.; Comp-
ton, D. S.; Sybertz, E. J. Burrier, R. E. Atherosclerosis 1994,
109, 162.
10. (a) Mathre, D. J.; Jones, T. K.; Xavier, L. C.; Blacklock, T.
J.; Blacklock, T. J.; Mohan, J. J.; Turner Jones, E. T.; Hoogs-
teen, K.; Baum, M. W.; Grabowski, E. J. J. J. Org. Chem.,
1991, 56, 751. (b)Jones, T. K.; Mohan, J. J.; Xavier, L. C.;
Blacklock, T. J.; Mathre, D. J.; Sohar, P.; Turner Jones, E. T.;
Reamer, R. A.; Roberts, F. E.; Grabowski, E. J. J. J. Org.
Chem. 1991, 56, 763.
4. Salisbury, B. G.; Davis, Jr. H. R.; Burrier, R. E.; Burnett, D.
A.; Boykow, G.; Caplen, M. A.; Clemmons, A. L.; Compton,
D. S.; Hoos, L. M.; McGregor, D. G.; Schnitzer-Poloko€, R.;
Smith, A. A.; Weig, B. C.; Zilli, D. L.; Clader, J. W.; Sybertz,
E. J. Atherosclerosis 1995, 115, 45.
11. Rosenblum, S. B.; Huynh, T.; Davis, H.; Yumibe, N.;
Clader, J.; Afonso, A.; Burnett, D. 12th International Sympo-
sium on Drugs A€ecting Lipid Metabolism, Houston, Texas,
November 7±10, 1995. Rosenblum, S. B.; Huynh, T.; Alfonso,
A.; Davis, H. R., Jr.; Yumibe, N.; Clader, J.; Burnett, D. J.
Med. Chem. 1998, 41, 973.
5. Clader, J. W.; Burnett, D. A.; Caplen, M. A.; Domalski, M.
S.; Dugar, S.; Vaccaro, W.; Sher, R.; Browne, M. E.; Zhao, H.;
Burrier, R. E.; Salisbury, B.; Davis, H. R. Jr. J. Med. Chem.
1996, 39, 3684.
6. Vaccaro, W.; Sher, R.; Clader, J. W.; Browne, M.; Burnett,
D. A.; Caplen, M. A.; Chen, L.-Y.; Domalski, M. S.; Dugar,
S.; Pushpavanem, P.; Zhao, H.; Davis, H. R. Jr. 210th ACS
National Meeting, Chicago, IL. August 20, 1995. Medicinal
Chemistry Session, Presentation 15.
12. Measured by reduction in liver cholesterol ester levels (L/
CE) in the 7-day cholesterol fed hamster model. All compounds
with indicated percent reductions were statistically di€erent
from the cholesterol-fed control group. The compounds were
evaluated in a series of separate 7-day cholesterol-fed hamster
studies hence, direct statistical comparisons among the com-
pounds were not performed. See ref. 1.
7. (a) Browne, M.; Burnett, D. A.; Caplen, M. A.; Chen, L.-Y.;
Clader, J. W.; Domalski, M. S.; Dugar, S.; Pushpavanem, P.;
Sher, R.; Vaccaro, W.; Vizziano, M.; Zhao, H. Tetrahedron
Lett. 1995, 36, 2555. (b) Best yields were obtained when the
imines were recrystallized prior to use in the ketene±imine
reaction.
13. Vaccaro, W. D.; Sher, R.; Davis, Jr. H. R. Bioorg. Med.
Chem. Lett. 1998, 8, 35.