Steroidal glycoside cholesterol absorption inhibitors

Article abstract of DOI:10.1021/jm9702600

We have explored the use of steroidal glycosides as cholesterol absorption inhibitors which act through an unknown mechanism. The lead for this program was tigogenin cellobioside (1, tiqueside) which is a weak inhibitor (ED₅₀ = 60 mg/kg) as measured in an acute hamster cholesterol absorption assay. Modification of the steroid portion of the molecule led to the discovery of 11-ketotigogenin cellobioside (5, pamaqueside) which has an ED₅₀ of 2 mg/kg. Replacement of the cellobiose with other sugars failed to provide more potent analogs. However, large improvements in potency were realized through modification of the hydroxyl groups on the cellobiose. This strategy ultimately led to the 4'',6''-bis[(2-fluorophenyl)carbamoyl]-β-D- cellobiosyl derivative of 11-ketotigogenin (51) with an ED₅₀ of 0.025 mg/kg in the hamster assay, as well as the corresponding hecogenin analog 64 (ED₅₀ = 0.07 mg/kg).

Full text of DOI:10.1021/jm9702600

J . Med. Chem. 1997, 40, 2547-2554
2547
Ster oid a l Glycosid e Ch olester ol Absor p tion In h ibitor s
Michael P. DeNinno,* Peter A. McCarthy, Kimberly C. Duplantier, Cynthia Eller, J ohn B. Etienne,
Michael P. Zawistoski, Faan Wen Bangerter, Charles E. Chandler, Lee A. Morehouse, Eliot D. Sugarman,
Robert W. Wilkins, Heidi A. Woody, and Lawrence M. Zaccaro
Central Research Division, Pfizer Inc., Eastern Point Road, Groton, Connecticut 06340
Received April 18, 1997X
We have explored the use of steroidal glycosides as cholesterol absorption inhibitors which act
through an unknown mechanism. The lead for this program was tigogenin cellobioside (1,
tiqueside) which is a weak inhibitor (ED50 ) 60 mg/kg) as measured in an acute hamster
cholesterol absorption assay. Modification of the steroid portion of the molecule led to the
discovery of 11-ketotigogenin cellobioside (5, pamaqueside) which has an ED50 of 2 mg/kg.
Replacement of the cellobiose with other sugars failed to provide more potent analogs. However,
large improvements in potency were realized through modification of the hydroxyl groups on
the cellobiose. This strategy ultimately led to the 4′′,6′′-bis[(2-fluorophenyl)carbamoyl]-â-D-
cellobiosyl derivative of 11-ketotigogenin (51) with an ED50 of 0.025 mg/kg in the hamster assay,
as well as the corresponding hecogenin analog 64 (ED50 ) 0.07 mg/kg).
In tr od u ction
has been well studied.8 Their small size would also limit
compound requirements. A published procedure6d,9 for
assaying ACAT inhibitors proved useful; however, it was
modified10 to increase throughput. Thus, compounds
were tested for their ability to inhibit the intestinal
absorption of [3H]cholesterol in cholesterol-fed hamsters.
The compounds were tested at four concentrations with
N ) 4 per group. The ED50 was determined by compar-
ing liver radioactivity of control animals with that of
hamsters given experimental compounds and generat-
ing a dose-response curve. The ED50 of 1 in this
protocol was 60 mg/kg.
The treatment of hypercholesterolemia through the
inhibition of dietary and biliary cholesterol absorption
has been a popular area of research in recent years.
Specific approaches have involved both known molec-
ular targets (e.g., ACAT1 and CEH2) as well as optimiz-
ing the activity of compounds with unknown mecha-
nisms of action.3-6 Our work on the latter class of
agents has focused on the steroidal glycoside series of
cholesterol absorption inhibitors. As a class, these
compounds were attractive because they had extremely
low systemic exposure and were assumed to be lume-
nally active but lacked robust in vivo activity. One of
these compounds, tiqueside (1, CP-88,818),6 had pro-
gressed through phase II clinical trials where it dem-
onstrated adequate efficacy but inadequate potency7 (4
g/day was required to achieve >20% reduction of LDL
cholesterol).
Ch em istr y
The spirostane glycosides were prepared using one
of two glycosidation procedures. The preferred method
utilized an acetobromo sugar as the glycosyl donor and
the sterol under ZnF2 catalysis in acetonitrile.11 Under
these conditions high yields of the â-glycosides were
obtained. In some instances (e.g., where the sterol was
insoluble in acetonitrile), the more typical Konigs-
Knorr12 conditions (i.e., HgBr2, Hg(CN)2, CH2Cl2) were
employed. Deprotection using sodium methoxide in
methanol afforded compounds 5-12.
The synthesis of the carbohydrate-modified targets
proceeded through the coupled spirostane glycosides.
Two routes were used to prepare the 6′- and 6′′-sub-
stituted analogs as shown in Schemes 1 and 2. Com-
pounds with carbamoyl substituents could be prepared
from the unprotected glycoside by reaction with an iso-
cyanate in pyridine. From this reaction, a nonselective
acylation occurs, and often the 6′-mono-, 6′′-mono-, and
6′,6′′-di-substituted analogs were isolated from the same
reaction. A more selective preparation of the latter ser-
ies was achieved through the use of protecting groups.
Thus, the 6′- and 6′′-hydroxyls can be protected as their
silyl ethers using tert-butyldiphenylsilyl chloride. The
remaining hydroxyls are then protected in the same pot
with acetic anhydride to give the fully protected deriva-
tive 20. The silyl groups can be selectively cleaved with
HF/pyridine to afford the diol 21. This intermediate
was also used to prepare the dihalogenated and dideoxy
analogs 29-33 by standard methods.
As part of our discovery program, modifications to the
steroid and sugar portions of the lead agent 1 were
explored seeking improvements in potency. Modifica-
tion of both halves of the molecule resulted in potency
improvements which were additive, leading to some
highly potent cholesterol absorption inhibitors.
Biology
Since the molecular mechanism of action of 1 was
unknown, we relied on the in vivo screening of analogs.
Hamsters seemed to be well suited for our needs since
they have a gall bladder and their bile acid metabolism
* Corresponding author: tel, (860) 441-5858; fax, (860) 441-3803;
e-mail, deninm@pfizer.com.
X Abstract published in Advance ACS Abstracts, J uly 1, 1997.
S0022-2623(97)00260-4 CCC: $14.00 © 1997 American Chemical Society

2548 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16
DeNinno et al.
Sch em e 1
Sch em e 2
Sch em e 3
droxyls as a p-methoxybenzylidene followed by exhaus-
tive acylation with chloroacetic anhydride to furnish the
fully protected derivative 48. The benzylidene was then
cleaved with acid in high yield to provide the diol 49.
In general, the carbamates were prepared by reacting
the alcohols with the appropriate isocyanate using
copper(I) chloride catalysis.13 The acetates were cleaved
with sodium methoxide to furnish the final products.
Because of the complex NMR spectra of these com-
pounds, we relied heavily on FAB MS for structure
confirmation.
Resu lts a n d Discu ssion
With virtually no prior structure-activity relation-
ship (SAR) to guide us, we divided the molecule into
lipophilic and hydrophilic sectors and began to make
modifications. We first replaced tigogenin with other
commercially available steroids, and the results are
shown in Table 1. Complete removal of the E and F
rings (compound 9) or use of steroids with hydrocarbon
side chains (10, 11) was highly detrimental for activity,
indicating that the spiroketal was an important com-
ponent. Attempts to add oxygen functionality to that
region of the steroid led to the epiandrosterone deriva-
tive 12 which contained a carbonyl at C17 and showed
improved activity. However, aromatization of the A ring
(13) was not tolerated. All of the spirostane cellobio-
sides (1, 5-8) were active, with the most potent being
those containing a carbonyl in the C12 or C11 position.
The 4′′,6′′-dicarbamoyl derivatives 52-64 were syn-
thesized from intermediate 49 in which the 4′′- and 6′′-
hydroxyls were uniquely freed (Scheme 3). This com-
pound was prepared by first protecting the 4′′,6′′-hy-

Cholesterol Absorption Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16 2549
Ta ble 3. 6′,6′′-Modified Cellobiose Analogs
Ta ble 1. Steroidal Cellobiosides
compda
steroid
tigogenin
ED50 (mg/kg)b
60
1
6
diosgenin
100
7
8
5
9
10
11
12
13
sarsasapogenin
hecogenin
11-ketotigogenin
androstane
75
10
2.0
compd
R
ED50 (mg/kg)
5
29
30
31
32
33
34
OH
H
2.0
6.5
3.0
0.7
0.5
1.5
5.0
37% @ 200 mg/kg
150
23% @ 200 mg/kg
25
OCH2OCH2CH3
cholestanol
F
Cl
I
stigmastanol
3-â-epiandrosterone
estrone
14% @ 100 mg/kg
(phenylcarbamoyl)oxy
a
All new compounds gave satisfactory C,H or C,H,N combustion
analyses. Values represent the ED50 or the percent inhibition at
b
the highest dose tested. The ED50 was determined from a four-
point dose-down with 4 hamsters/dose group. Reproducibility is
generally (20% of the average value.
compounds 38-40). Substitution at 6′, in fact, appeared
to be detrimental. One of the most potent compounds
was the 6′′-mono 2,6-dichlorophenyl carbamate 42 with
an ED50 of 0.2 mg/kg. Further exploration in this series
could not break this potency barrier. Analogs contain-
ing carbamates derived from non-aniline amines (45-
47) were much less active.
Ta ble 2. Cellobiose Replacements
The success of the approach to modify the cellobiose
led us to explore other sites on the sugar from which to
append carbamates. Another accessable site was the
4′′-position which along with the 6′′-hydroxyl could be
selectively freed using protecting group chemistry (vide
supra). The diol intermediate 49 was used to synthesize
a series of 4′′,6′′-dicarbamates listed in Table 5. In con-
trast to the 6′′-monocarbamates, the 2,6-dichlorophenyl
carbamate derivative 53 was not among the most potent
compounds in this series. The 2-fluorophenyl analog 51,
which had an ED50 of 0.025 mg/kg, was 2 orders of
magnitude more active than the unmodified cellobioside
5. Once again, carbamates derived from nonaromatic
amines were much less active. Even within the sub-
stituted phenyl carbamate series, small changes (e.g.,
F to Cl) led to dramatic effects on activity. This tight
SAR and the high levels of potency uncovered in this
series have helped to further define the mechanism of
action of these agents, which has yet to be fully
elucidated. The results provided here have effectively
ruled out a simple cholesterol sequestration mechanism
that was thought possible with compound 1.
The 4′′,6′′-bis[(2-fluorophenyl)carbamoyl] substitution
pattern was optimal in this series so the corresponding
tigogenin and hecogenin cellobioside analogs were also
prepared. The potencies of these compounds (63, ED50
) 1.0 mg/kg; 64, ED50 ) 0.07 mg/kg) were as predicted
based on the data in Table 1 indicating that the SAR
for both ends of the molecule was independent and
additive.
To determine whether the increases in potency real-
ized in the acute cholesterol absorption assay would
translate to improved potency upon multiple dosing, we
selected several compounds and administered them to
hamsters by diet admixture. Hamsters fed a cholesterol-
containing diet have increased hepatic and plasma
cholesterol levels compared to animals fed a cholesterol-
free diet. As shown in Figure 1, the administration of
1, 5, or 64 to cholesterol-fed hamsters prevented the
accumulation of hepatic cholesterol in a dose-respon-
sive fashion. The incremental increases in potency in
compd
sugar
glycos procedure
ED50 (mg/kg)
2.0
16
12
4.0
20
25
5
14
15
16
17
18
19
â-D-cellobioside
â-D-glucoside
â-D-galactoside
â-D-lactoside
â-D-maltoside
â-D-maltotrioside
R-D-cellobioside
A
A
A
A
B
B
23% @ 10 mg/kg
Compound 5 (pamaqueside, CP-148,623) was the most
potent with an ED50 of 2 mg/kg in the acute hamster
assay and was the subject of an earlier communication.14
Although simple steroids such as sitosterol have been
shown to have lipid-lowering properties,15 none of the
spirostane aglycones we examined inhibited cholesterol
absorption at 200 mg/kg. We then set out to investigate
the role of the cellobioside by replacing it with other
common sugars (Table 2) on the 11-ketotigogenin back-
bone. Although the potency of the lactoside derivative
16 was comparable to the cellobioside, none of the
compounds provided any advantage over compound 5.
Replacing the sugar had only limited success, so we
opted to make modifications to the cellobiose itself. We
began by modifying the primary hydroxyl groups which
were substituted or replaced with other functional
groups, and the results are shown in Table 3. Although
several of these analogs lost potency, the fluoro (31) and
chloro (32) derivatives provided some potency improve-
ments with compound 32 being the most potent of this
series (ED50 ) 0.5 mg/kg). The carbamate 34 lost
potency, but this result was nevertheless exciting as it
opened a series which allowed for the preparation of a
large number of analogs. We therefore synthesized a
number of mono- and dicarbamates at the 6′- and 6′′-
positions (Table 4). In addition to providing improve-
ments in potency, it was clear from the data that only
the 6′′-position needed to be substituted (e.g., see

2550 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16
Ta ble 4. 6′- and 6′′-Carbamoylcellobiose Analogs
DeNinno et al.
compd
R1
R2
ED50 (mg/kg)
2.0
5.0
1.0
5
35
36
37
38
39
40
41
42
43
44
45
46
47
H
H
(4-chlorophenyl)carbamoyl
(4-chlorophenyl)carbamoyl
H
(2-fluorophenyl)carbamoyl
(2-fluorophenyl)carbamoyl
H
(2,6-dichlorophenyl)carbamoyl
(2,6-dichlorophenyl)carbamoyl
(2,4-difluorophenyl)carbamoyl
(2-methylphenyl)carbamoyl
benzylcarbamoyl
(4-chlorophenyl)carbamoyl
H
(4-chlorophenyl)carbamoyl
(2-fluorophenyl)carbamoyl
37% @ 5 mg/kg
0.8
0.2
7.0
0.4
0.2
0.25
0.5
5.0
5.0
1.5
H
(2-fluorophenyl)carbamoyl
(2,6-dichlorophenyl)carbamoyl
H
H
H
H
H
H
phenethylcarbamoyl
1-pyrrolidenecarbonyl
Ta ble 5. 4′′,6′′-Dicarbamoylcellobiose Analogs
compd
R
ED50 (mg/kg)
2.0
5
52
53
54
51
55
56
57
58
59
60
61
62
H
phenylcarbamoyl
0.1
(2,6-dichlorophenyl)carbamoyl
(2,4-difluorophenyl)carbamoyl
(2-fluorophenyl)carbamoyl
(3-fluorophenyl)carbamoyl
(4-fluorophenyl)carbamoyl
(2-chlorophenyl)carbamoyl
benzylcarbamoyl
(2-furylmethyl)carbamoyl
[(ethoxycarbonyl)methyl]carbamoyl
1-morpholinocarbonyl
40% @ 5 mg/kg
0.04
0.025
0.1
0.2
1.0
0.3
0.3
0.6
4.0
F igu r e 1. Hamsters (n ) 6/group) were fed a 0.2% cholesterol,
0.1% cholic acid diet or a cholesterol-free diet for 4 days with
1 (9), 5 (b), or 64 (2) mixed in the chow at the indicated
percentages. The approximate dosage in mg/kg or µg/kg is also
included for reference. Livers were excised and saponified, and
the cholesterol content was expressed per gram of liver. The
data are a compilation of two separate experiments. Values
are mean ( SD. *p < 0.05 versus cholesterol-fed controls
(Student-Newman Keul’s test).
phenylthiocarbamoyl
0.04
this model mirror those discussed in the acute choles-
terol absorption protocol. Similarly, these compounds
prevented the elevation of plasma cholesterol levels due
to cholesterol feeding.16 Compounds 5 and 64 were both
progressed into clinical development. Details of these
studies will be presented elsewhere.
Lacking an in vitro screen, we relied on a fairly high
throughput in vivo acute hamster model to rapidly
access the activity of the analogs. The spirostane,
tigogenin, contained in 1 appeared to be optimal;
however, incorporation of carbonyl groups at the C11
or C12 position led to improved activity. The cellobiose
unit in 1 was also found to be superior to other mono-
or disaccharides. Additional gains in potency were
registered upon modification of the cellobiose, preferably
with carbamoyl substituents at the 4′′- and 6′′-positions.
The availability of these highly potent inhibitors may
help to unravel some of the mysteries associated with
the process of cholesterol absorption. Lipid-lowering,
pharmacokinetic, and further SAR studies of these
spirostane glycosides will be the subject of future
publications.
Su m m a r y
Through iterative modifications to a weak cholesterol
absorption inhibitor lead (1), we have succeeded in
improving the potency by over 3 orders of magnitude.

Cholesterol Absorption Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16 2551
4 (2.0 g) in CH2Cl2 (35 mL), and the resulting mixture was
stirred at room temperature for 94 h. The reaction was
quenched by slow addition of saturated aqueous NaHCO3 (20
mL). The organic layer was separated, dried over magnesium
sulfate, and concentrated in vacuo to give 1.637 g of a black
solid. Purification by repeated flash chromatography (1:1
hexane:ethyl acetate eluent) provided 651 mg (33% yield) of
the title compound: mp 248-249 °C; FAB MS m/ z 1049 (M
+ H+), 1071 (M + Na+). Anal. Calcd for C53H76O21·H2O: C,
59.65; H, 7.37. Found: C, 59.66; H, 7.00.
Exp er im en ta l Section
Gen er a l. Melting points were obtained on a Buchi melting
point apparatus and are uncorrected. Reactions were run in
flame- or oven-dried glassware under nitrogen. Anhydrous
solvents were purchased from Aldrich Chemical Co. Silica gel
chromatography was performed using EM Science silica gel
60 (230-400 mesh). FAB mass spectra were run on a Kratos
Concept instrument. Proton NMR were recorded on a Bruker
AC300 or AC250 spectrometer. Elemental analyses were
obtained from Schwarzkopf Microanalytical Laboratory, Wood-
side, NY.
Glycosid a tion P r oced u r e: Meth od A. (3â,5r,25R)-3-
[(Hep ta a cetyl-â-D-cellobiosyl)oxy]sp ir osta n -11-on e (4). A
suspension of (3â,5R,25R)-3-hydroxyspirostan-11-one (2) (3.0
g, 6.97 mmol) and anhydrous zinc fluoride (2.88 g, 27.9 mmol)
in dry MeCN (175 mL) was dried by removal of 75 mL of MeCN
by distillation. The suspension was allowed to cool, hep-
taacetyl-â-D-cellobiosyl bromide (9.75 g, 13.9 mmol) was added,
and the resulting suspension was heated to 65 °C for 3 h. After
cooling to room temperature, CH2Cl2 (150 mL) was added; the
suspension was stirred for 10 min and filtered. The filtrate
was concentrated in vacuo to give 10 g of crude product. This
material was dissolved in 8:2 chloroform:methanol, pread-
sorbed on silica gel, and purified by flash chromatography
(eluent: 1:1 ethyl acetate:hexane followed by pure ethyl
(3â,5r,25R)-3-[(â-D-Cellobiosyl)oxy]spir ostan -11-on e (5).
A mixture of compound 4 (6.57 g, 6.26 mmol), NaOMe (68 mg,
1.25 mmol), MeOH (35 mL), and THF (75 mL) was heated to
reflux for 1 h followed by stirring at room temperature for 12
h. A white precipitate formed within 30 min. The final
suspension was concentrated in vacuo to give 6.0 g of crude
product. This material was purified by flash chromatography
(chloroform followed by 8:2 chloroform:MeOH) to give 2.71 g
1
(57% yield) of 5: mp >300 °C; H NMR (DMSO-d6) δ 5.22 (d,
1H, J ) 5 Hz), 5.00 (m, 3H), 4.64 (s, 1H), 4.58 (t, 1H, J ) 5
Hz), 4.54 (t, 1H, J ) 6 Hz), 4.34 (q, 1H, J ) 8 Hz), 4.27 (d, 1H,
J ) 8 Hz), 4.23 (d, 1H, J ) 8 Hz), 3.68-2.94 (m, 15H), 2.34
(m, 2H), 2.08-0.81 (m, 23H), 0.92 (s, 3H), 0.86 (d, 3H, J ) 7
Hz), 0.72 (d, 3H, J ) 6 Hz), 0.59 (s, 3H); FAB MS m/ z 777 (M
+ H+). Anal. Calcd for C39H62O14‚2H2O: C, 59.22; H, 8.41.
Found: C, 59.48; H, 8.48.
1
acetate) to give 6.81 g (93% yield) of 4: mp 210-212 °C; H
(3â,5r,25R )-3-[[6′,6′′-B is (t er t -b u t y ld ip h e n y ls ily l)-
2′,2′′,3′,3′′,4′′-p en t a a cet yl-â-D-cellob iosyl]oxy]sp ir ost a n -
11-on e (20). A mixture of compound 5 (25 g, 0.03 mol),
imidazole (15.78 g, 0.23 mol), 4-(dimethylamino)pyridine (2.0
g), and DMF (400 mL) was cooled to 0 °C. tert-Butyldiphe-
nylsilyl chloride (34.45 mL, 0.13 mol) was added; the mixture
was warmed to room temperature and stirred for 5 h. Pyridine
(53.57 mL, 0.66 mol) and acetic anhydride (54.87 mL, 0.50 mol)
were added, and the reaction mixture was stirred overnight.
The reaction was then quenched with water, diluted with ethyl
acetate, washed with water (2×), 1 N HCl (3×), and brine (1×),
dried over Na2SO4, filtered, and concentrated in vacuo to afford
20 as a foam (66.05 g, >100%): mp 135 °C; 1H NMR (250 MHz,
CDCl3) δ 7.85-7.3 (m, 20H), 5.3-3.3 (m, 18H), 2.6-1.0 (m,
61H), 0.90 (d, 3H, J ) 7 Hz), 0.8 (d, 3H, J ) 7 Hz), 0.7 (s, 3H);
FAB MS m/ z 1441 (M + H+).
(3â,5r,25R)-3-[(2′,2′′,3′,3′′,4′′-P en t a a cet yl-â-D-cellob io-
syl)oxy]sp ir osta n -11-on e (21). A mixture of compound 20
(15.04 g, 0.01 mol) and pyridine (100 mL) was cooled to 0 °C.
Hydrogen fluoride in pyridine (36.64 mL) was added, and the
reaction mixture was gradually warmed to room temperature
and allowed to stir for 4 h. The reaction was cooled and
quenched with water. The mixture was then dissolved in ethyl
acetate, washed with water (1×), 1 N HCl (4×), and brine (1×),
dried over Na2SO4, filtered, and concentrated in vacuo. The
residual material was triturated with hexanes, filtered, washed
with hexanes, and dried to afford 7.22 g (74.8%) of 21: mp
233-234 °C; 1H NMR (250 MHz, CDCl3) δ 5.25-3.3 (m, 20H),
2.65-1.05 (m, 40H), 1.0 (s, 3H), 0.95 (d, 3H, J ) 8 Hz), 0.8 (d,
3H, J ) 8 Hz), 0.7 (s, 3H); FAB MS m/ z 965 (M + H+). Anal.
Calcd for C49H72O19: C, 60.98; H, 7.52. Found: C, 60.78; H,
7.54.
NMR (CDCl3) δ 5.11 (complex, 2H), 5.04 (t, J ) 9 Hz, 1H),
4.90 (t, J ) 9 Hz, 1H), 4.83 (t, J ) 8 Hz, 1H), 4.49 (complex,
4H), 4.34 (dd, J ) 4.5, 12.5 Hz, 1H), 4.04 (t, J ) 13 Hz, 1H),
4.03 (t, J ) 11 Hz, 1H), 3.72 (t, J ) 9.5 Hz, 1H), 3.65 (m, 1H),
3.56 (m, 1H), 3.45 (m, 1H), 2.47 (m, 1H), 2.22 (s, 2H), 2.08 (s,
3H), 2.06 (s, 3H), 2.00 (s, 6H), 1.99 (s, 6H), 1.96 (s, 3H), 2.00-
1.00 (m, 22H), 0.98 (s, 3H), 0.92 (d, J ) 7 Hz, 3H), 0.77 (d, J
) 7 Hz, 3H), 0.68 (s, 3H); FAB MS m/ z 1049 (M + H)+. Anal.
Calcd for C53H76O21‚H2O: C, 59.65; H, 7.37. Found: C, 59.86;
H, 7.25.
Meth od B: Step 1. (3â,5r,25R)-3-[(Tr im eth ylsilyl)oxy]-
sp ir osta n -11-on e. Trimethylsilyl chloride (3.27 mL, 25.8
mmol) was added to a solution of (3â,5R,25R)-3-hydrox-
yspirostan-11-one (4.0 g, 9.3 mmol) and triethylamine (6.5 mL,
46 mmol) in CH2Cl2 (60 mL) at room temperature. One gram
of 4-(dimethylamino)pyridine was added, and the reaction
mixture was stirred at room temperature for 12 h. The
reaction was quenched with MeOH (1 mL), diluted with ethyl
acetate, washed with water (5×) and brine (1×), dried (Na2-
SO4), filtered, and concentrated in vacuo. The product was
triturated with MeOH, filtered, and dried to afford 3.94 g (85%)
of product as a white solid: 1H NMR (250 MHz, CDCl3) δ 4.5
(q, 1H, J ) 6 Hz), 3.45 (m, 2H), 2.35 (t, 1H, J ) 10 Hz), 2.4
(dt, 1H, J ) 12, 2 Hz), 2.2 (s, 2H), 2.1-1.1 (m, 12H), 1.02 (s,
3H), 0.9 (d, 3H, J ) 7.0 Hz), 0.78 (d, 3H, J ) 7 Hz), 0.69 (s,
3H), 0.1 (s, 9H).
Step 2. (3â,5r,25R)-3-[(Hep ta a cetyl-â-D-m a ltosyl)oxy]-
sp ir osta n -11-on e. Powdered 4Å molecular sieves (4 g) were
added to a solution of 3-[(trimethylsilyl)oxy]-(3â,5R,25R)-
spirostan-11-one (3.90 g, 7.76 mmol) and acetobromomaltose
(8.15 g, 11.7 mmol) in CH2Cl2 (60 mL) at room temperature.
After the mixture stirred for 15 min, Hg(CN)2 (7.85 g, 31 mmol)
and HgBr2 (11.2 g, 31 mmol) were added, and the mixture
stirred at room temperature for 7 h. The mixture was diluted
with ethyl acetate, washed with 1 N HCl (3×) and brine (1×),
dried over Na2SO4, filtered, and concentrated in vacuo. The
product was purified by flash chromatography (30-60%
EtOAc/hexanes) and afforded 3.5 g (43%) of the title product:
1H NMR (250 MHz, CDCl3) δ 5.4 (d, 1H, J ) 3 Hz), 5.3 (dd,
1H, J ) 9, 9 Hz), 5.2 (dd, 1H, J ) 8, 8 Hz), 5.05 (dd, 1H, J )
9, 9 Hz), 4.88 (dd, 1H, J ) 11, 3 Hz), 4.7 (dd, 1H, J ) 10, 8
Hz), 4.4 (m, 2H), 4.2 (m, 2H), 4.0 (m, 3H), 3.5 (m, 4H), 2.5 (m,
1H), 2.2 (s, 2H), 2.16 (s, 3H), 2.1 (s, 3H), 2.02 (s, 3H), 2.0 (s,
3H), 1.98 (s, 3H), 1.96 (s, 3H), 1.95 (s, 3H), 1.95-1.0 (m, 23H),
1.0 (s, 3H), 0.9 (d, 3H, J ) 7 Hz), 0.65 (d, 3H, J ) 7 Hz), 0.55
(s, 3H); FAB MS 1049 (M + H+).
(3â,5r,25R)-3-[(6′,6′′-Did eoxy-6′,6′′-d iflu or o-2′,2′′,3′,3′′,4′′-
p en t a a cet yl-â-D-cellob iosyl)oxy]sp ir ost a n -11-on e (22).
DAST (0.34 mL, 2.59 mmol) was added to a solution of
compound 21 (250 mg, 0.259 mmol) in anhydrous dimethoxy-
ethane (5 mL), at 0 °C. After 20 min, the reaction mixture
was warmed to room temperature for 30 min and then to 40
°C for 1.5 h. The reaction mixture was cooled to 0 °C, diluted
with ethyl acetate (50 mL), and poured into ice water. The
organic layer was washed with 1 N HCl (1×), NaHCO3 solution
(1×), and brine (1×), dried over Na2SO4, filtered, and concen-
trated in vacuo. The crude product was purified by flash
chromatography (40% ethyl acetate/hexanes) to afford 190 mg
1
(76%) of 22 as a white solid: mp 267-269 °C; H NMR (250
MHz, CDCl3) δ 5.15 (dd, 1H, J ) 8, 8 Hz), 5.05 (dd, 1H, J ) 9,
9 Hz), 4.9 (dd, 1H, J ) 9, 8 Hz), 4.85 (dd, 1H, J ) 9, 8 Hz), 4.7
(m, 1H), 4.5 (m, 4H), 3.85 (dd, 1H, J ) 9.0, 9.0 Hz), 3.6 (m,
1H), 3.5 (m, 1H), 3.35 (dd, 1H, J ) 10.0, 11.0 Hz), 2.5 (m, 1H),
2.2 (s, 2H), 2.01 (s, 3H), 2.0 (s, 6H), 1.98 (s, 3H), 1.97 (s, 3H),
(3â,5r,25R )-3-[(H e p t a a c e t y l-r-D -c e llo b io s y l)o x y ]-
sp ir osta n -11-on e. Hydrobromic acid (30% in acetic acid, 1.2
mL) was added to a room temperature solution of compound

2552 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16
DeNinno et al.
1.9-1.0 (m, 27H), 1.0 (s, 3H), 0.92 (d, 3H, J ) 7 Hz), 0.75 (d,
3H, J ) 7 Hz), 0.7 (s, 3H); FAB MS m/ z 969 (M + H+). Anal.
Calcd for C49H70F2O17: C, 60.73; H 7.28. Found: C, 61.06; H,
7.19.
was then dried over Na2SO4, filtered, and concentrated in
vacuo. The crude product was purified by flash chromatog-
raphy (60% ethyl acetate/40% hexanes) to afford 0.505 g of 27
(46.7%): 1H NMR (250 MHz, CDCl3) δ 5.15 (dd, 1H, J ) 8, 8
Hz), 5.1 (m, 1H), 4.85 (m, 2H), 4.7 (s, 1H), 4.6 (s, 1H), 4.5 (dd,
1H, J ) 18, 7 Hz), 4.45 (m, 1H), 4.1 (q, 2H, J ) 6 Hz), 3.8 (m,
2H), 3.5 (m, 10H), 2.5 (m, 1H), 2.2 (s, 2H), 2.01 (s, 3H), 2.0 (s,
6H), 1.98 (s, 3H), 1.97 (s, 3H), 1.9-1.0 (m, 26H), 1.2 (m, 6H),
1.0 (s, 3H), 0.92 (d, 3H, J ) 7 Hz), 0.75 (d, 3H, J ) 7 Hz), 0.7
(s, 3H); FAB MS m/ z 1104 (M + Na+). Anal. Calcd for
C55H84O21: C, 61.10; H, 7.83. Found: C, 61.05; H, 7.92.
(3â,5r,25R)-3-[[6′,6′′-Bis(ph en ylcar bam oyl)-2′,2′′,3′,3′′,4′′-
p en ta a cetyl-â-D-cellobiosyl]oxy]sp ir osta n -11-on e (28). A
mixture of compound 21 (0.50 g, 0.52 mmol), CH2Cl2 (5 mL),
triethylamine (0.50 mL, 3.62 mmol), and 4-(dimethylamino)-
pyridine (0.10 g) was cooled to 0 °C. Phenyl isocyanate (0.34
mL, 3.12 mmol) was added, and the reaction mixture was
stirred at 0 °C for 15 min and then at room temperature for
30 min. The reaction was quenched with MeOH, diluted with
ethyl acetate, washed with water (1×), 1 N HCl (3×), saturated
NaHCO3 solution (2×), and brine (1×), dried over Na2SO4,
filtered, and concentrated in vacuo. The residual material was
triturated with hexanes, filtered, washed with hexanes, and
dried to afford >100% yield of compound 28: 1H NMR (250
MHz, CDCl3) δ 7.55-7.0 (m, 10H), 6.85 (s, 1H), 6.65 (s, 1H),
5.3-3.52 (m, 17H), 3.45 (t, 1H, J ) 10 Hz), 2.5-1.05 (m, 40H),
1.0 (s, 3H), 0.95 (d, 3H, J ) 8 Hz), 0.8 (d, 3H, J ) 8 Hz), 0.7
(s, 3H). Anal. Calcd for C63H82N2O21: C, 62.88; H, 6.87; N,
2.33. Found: C, 62.51; H, 6.91; N, 2.30.
Non selective Ca r ba m oyla tion . A mixture of compound
5 (0.50 g, 0.66 mmol), pyridine (3 mL), and 4Å molecular sieves
(0.50 g) was stirred for 10 min at room temperature. The reac-
tion mixture was then cooled to -40 °C, and 4-chlorophenyl
isocyanate (0.122 g, 0.8 mmol) was added. The reaction mix-
ture was gradually warmed to room temperature and stirred
for 1.5 h. The reaction was quenched with MeOH, filtered,
and concentrated in vacuo twice with toluene in order to
remove the pyridine. The crude material was purified by
gradient flash chromatography (2-10% MeOH:chloroform).
The products were concentrated in vacuo, triturated with
methanol/water, filtered, and washed with water and dried
to afford the products listed below in order of elution.
(3â,5r,25R)-3-[(6′,6′′-Did eoxy-6′,6′′-d iiod o-2′,2′′,3′,3′′,4′′-
p en ta a cetyl-â-D-cellobiosyl)oxy]sp ir osta n -11-on e (25). A
mixture of compound 21 (2.00 g, 2.08 mmol), imidazole (0.85
g, 12.0 mmol), and triphenylphosphine (3.26 g, 12.0 mmol) was
dissolved in toluene (40 mL). Iodine (2.10 g, 8.30 mmol) was
added, and the reaction mixture was heated at reflux temper-
ature overnight. The reaction mixture was cooled, diluted with
ethyl acetate, washed with 1 N HCl (1×), sodium bicarbonate
solution (1×), and brine (1×), dried over Na2SO4, filtered, and
concentrated in vacuo. The crude product was purified by
flash chromatography (10% ethyl acetate/methylene chloride)
to afford 1.78 g (72.3%) of 25 as a white solid: 1H NMR (250
MHz, CDCl3) δ 5.25-3.0 (m, 18H), 2.6-1.05 (m, 42H), 0.95
(d, 3H, J ) 7 Hz), 0.80 (s, 3H), 0.75 (d, 3H, J ) 7 Hz), 0.60 (s,
3H); FAB MS m/ z 1171 (M + H+).
(3â,5r,25R)-3-[(6′,6′′-Did eoxy-2′,2′′,3′,3′′,4′′-p en ta a cetyl-
â-D-cellobiosyl)oxy]spir ostan -11-on e (26). Azoisobutyrylni-
trile (10 mg) was added to a solution of compound 25 (150 mg,
0.128 mmol) and tri-n-butyltin hydride (0.105 mL, 0.39 mmol)
in anhydrous toluene (5 mL) at room temperature. The
reaction mixture was heated at reflux temperature for 3 h,
cooled and concentrated in vacuo. The residual material was
triturated with hexanes, filtered, dried, and concentrated. The
crude product was purified by flash chromatography (5% ethyl
acetate/methylene chloride) to afford 0.1 g (82%) of 26 as a
white solid: 1H NMR (250 MHz, CDCl3) δ 5.1 (m, 2H), 4.8 (m,
2H), 4.5 (m, 2H), 4.1 (q, 1H, J ) 7 Hz), 3.4 (m, 4H), 2.45 (bd,
1H, J ) 14 Hz), 2.25 (s, 2H), 2.12 (s, 3H), 2.10 (s, 3H), 2.09 (s,
3H), 2.06 (s, 3H), 1.98 (s, 3H), 1.9-1.3 (m, 25H), 1.3 (d, 3H, J
) 6 Hz), 1.22 (d, 3H, J ) 6 Hz), 1.0 (s, 3H), 0.9 (d, 3H, J ) 7
Hz), 0.8 (d, 3H, J ) 7 Hz), 0.7 (s, 3H); FAB MS m/ z 956 (M +
Na+). Anal. Calcd for C49H72O17: C, 63.07; H, 7.78. Found:
C, 63.11; H, 7.84.
(3â,5r,25R)-3-[(6′,6′′-Dim esyl-2′,2′′,3′,3′′,4′′-p en ta a cetyl-
â-D-cellobiosyl)oxy]sp ir osta n -11-on e (23). A solution of
compound 21 (1.00 g, 1.04 mmol) and triethylamine (1.50 mL,
10.40 mmol) in CH2Cl2 (10 mL) was cooled to 0 °C. Mesyl
chloride (0.48 mL, 6.22 mmol) and 4-(dimethylamino)pyridine
(0.02 g) were added, and the reaction mixture was stirred at
0 °C for 2 h. The reaction mixture was diluted with ethyl
acetate (50 mL), washed with 1 N HCl (2×), sodium bicarbon-
ate solution (1×), and brine (1×), dried over Na2SO4, filtered,
and concentrated in vacuo. The residual material was dis-
solved in CH2Cl2 (10 mL), and hexanes were added (20 mL).
The CH2Cl2 was removed in vacuo, and the resulting precipi-
tate was filtered, washed with hexanes, and dried under
vacuum to afford 1.10 g (94%) of 23 as a white solid: 1H NMR
(250 MHz, CDCl3) δ 5.25-3.3 (m, 18H), 3.1 (d, 6H, J ) 7 Hz),
2.6-1.1 (m, 40H), 1.0 (s, 3H), 0.95 (d, 3H, J ) 7 Hz), 0.77 (d,
3H, J ) 7 Hz), 0.7 (s, 3H).
(3â,5r,25R)-3-[[6′,6′′-Bis[(4-ch lor op h en yl)ca r b a m oyl]-
â-D-cellobiosyl]oxy]sp ir osta n -11-on e (35): 250 mg (36%);
FAB MS m/ z 1061 (M + H+). Anal. Calcd for C53H70Cl2N2O16
‚
1.5H2O: C, 58.51; H, 6.76; N, 2.57. Found: C, 58.55; H, 6.54;
N, 2.66.
(3â,5r,25R)-3-[[6′′-[(4-Ch lor op h en yl)ca r b a m oyl]-â-D-
cellobiosyl]oxy]sp ir osta n -11-on e (36): 93 mg (15%); FAB
MS m/ z 930 (M + Na+). Anal. Calcd for C46H66ClNO15: C,
60.82; H, 7.32; N, 1.54. Found: C, 60.71; H, 7.24; N, 1.33.
(3â,5r,25R)-3-[[6′-[(4-Ch lor op h en yl)ca r ba m oyl]-â-D-cel-
lobiosyl]oxy]sp ir osta n -11-on e (37): 90 mg (15%); FAB MS
m/ z 908 (M + H+). Anal. Calcd for C46H66ClNO15: C, 59.63;
H, 7.40; N, 1.51. Found: C, 59.54; H, 7.24; N, 1.33.
(3â,5r,25R)-3-[[6′,6′′-Bis(1-p yr r olid in ylca r b on yl)-â-D-
cellobiosyl]oxy]sp ir osta n -11-on e (47). Step 1: A mixture
of compound 21 (1.0 g, 1.04 mmol), carbonyldiimidazole (0.42
g, 2.6 mmol), and diisopropylethylamine (0.9 mL, 5.2 mmol)
in dichloroethane (7 mL) was stirred at room temperature for
2 h. Pyrrolidine (0.185 g, 2.6 mmol) was added, and the
mixture was stirred for 5 h at room temperature. The mixture
was diluted with ethyl acetate, washed with 1 N HCl (2×),
saturated NaHCO3 solution (1×), and brine (1×), dried over
Na2SO4, filtered, and concentrated in vacuo. The residue was
triturated from EtOAc/hexanes, filtered, and dried to afford
800 mg (88%) of product as a colorless solid: FAB MS m/ z
1183 (M + Na+).
(3â,5r,25R)-3-[(6′,6′′-Did eoxy-6′,6′′-d ich lor o-2′,2′′,3′,3′′,4′′-
p en ta a cetyl-â-D-cellobiosyl)oxy]sp ir osta n -11-on e (24). A
mixture of compound 23 (0.10 g, 0.09 mmol), lithium chloride
(0.15 g), and DMF (2 mL) was heated to 85 °C and stirred for
3 h. The reaction mixture was then cooled, diluted with ethyl
acetate (20 mL), washed with water (2×) and brine (1×), dried
over Na2SO4, filtered, and concentrated in vacuo. The residual
material was triturated with hexanes, filtered, and dried to
1
afford 0.079 g (88%) of 24 as a white solid: mp >275 °C; H
NMR (250 MHz, CDCl3) δ 5.2-3.3 (m, 18H), 2.6-1.1 (m, 40H),
1.0 (s, 3H), 0.90 (d, 3H, J ) 7 Hz), 0.75 (d, 3H, J ) 7 Hz), 0.70
(s, 3H); FAB MS m/ z 1001 (M + H+). Anal. Calcd for C49H70
Cl2O17: C, 58.74; H, 7.04. Found: C, 58.71; H, 7.12.
-
(3â,5r,25R)-3-[[6′,6′′-Bis(et h oxym et h yl)-2′,2′′,3′,3′′,4′′-
p en ta a cetyl-â-D-cellobiosyl)oxy]sp ir osta n -11-on e (27). A
mixture of compound 21 (0.50 g, 0.52 mmol), dichloroethane
(5 mL), diisopropylethylamine (1 mL), and ethoxymethyl
chloride (0.196 g, 2.07 mmol) was stirred at room temperature
for 4 h. The reaction mixture was warmed to 50 °C for 2 h
and cooled and the reaction quenched with MeOH. The
mixture was then diluted with ethyl acetate and washed with
water (1×), 1 N HCl (2×), and brine (1×). The organic layer
Step 2: Deprotection as described for compound 5 afforded
47 as a colorless solid: 400 mg (78%); FAB MS m/ z 852 (M +
H+). Anal. Calcd for C44H69NO15: C, 61.38; H, 8.19; N, 1.63.
Found: C, 61.10; H, 8.05; N, 1.70.
(3â,5r,25R)-3-[[4′′,6′′-(4-Meth oxyben zyliden e)-2′,2′′,3′,3′′,6′-
p en ta k is(ch lor oa cetyl)-â-D-cellobiosyl]oxy]sp ir osta n -11-
on e (48). Camphorsulfonic acid (3 g) was added to a mixture
of compound 5 (50 g, 0.066 mol) and anisaldehyde dimethyl

Cholesterol Absorption Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16 2553
acetal (50 mL, 0.29 mol) in 1,2-dichloroethane (1500 mL). The
suspension was heated to reflux temperature, and 200 mL of
solvent was distilled off. After 4 h at reflux temperature, the
dark, gelatinous mixture was cooled to 0 °C and treated with
pyridine (160 mL, 1.99 mol) and chloroacetic anhydride (170
g, 1 mol). The reaction mixture was allowed to warm to room
temperature and after 2 h was washed with 1 N HCl (3×),
NaHCO3 (1×), and brine (1×), dried (Na2SO4), filtered, and
a liquid scintillation counter. Total hepatic radioactivity was
calculated from the radioactivity in each aliquot after adjusting
for the total volume of the saponified liver. The percent
inhibition of cholesterol absorption was determined by com-
paring the hepatic radioactivity in hamsters receiving choles-
terol absorption inhibitors to that in hamsters receiving
vehicle.
2. Su bch r on ic Assa y. Male golden Syrian hamsters (LVG
strain; 100-120 g) (Charles Rivers Laboratory, Wilmington,
MA), housed 6/cage, were acclimated to an altered light cycle
(lights on from 3 p.m. to 3 a.m.) for 1 week and then placed on
a powdered AIN 76C semipurified diet (ICN Biomedicals,
Costa Mesa, CA) for 3 days to become accustomed to eating a
powdered diet. At the start of the study, food was switched
to AIN 76C plus 0.2% cholesterol/0.1% cholic acid that
contained various concentrations of powdered test compounds.
Body weights and food jar weights were recorded daily, and
doses of compound (mg/kg) were estimated from mean body
weight, food consumed per day, and percent compound in the
feed. Hamsters were euthanized on study day 4, and liver was
excised for the determination of hepatic cholesterol levels.
Livers were saponified in 5 mL of 2.5 M KOH for 2 h at 70 °C.
Aliquots (200 µL) of the saponified liver (or homogenate) were
added to 1 mL of ethanol containing [14C]cholesterol (∼100 000
dpm; used as an internal recovery standard). Tubes were
mixed and placed at -20 °C for 15 min to facilitate precipita-
tion of protein before the addition of 4 mL of hexane. After
vigorous vortexing, samples were centrifuged at ∼1500g for 5
min to separate phases; 3 mL of the hexane layer was
evaporated to dryness under N2 before being redissolved in
300 µL of 1% (v/v) Triton X-100 in ethanol. Cholesterol levels
were analyzed using a commercial assay kit (Single Vial,
Boehringer Mannheim, Indianapolis, IN). Triplicate 15 µL
aliquots were analyzed in 96-well microtiter plates; 200 µL of
reagent was added, and the plates were allowed to incubate
at room temperature for 1 h. Absorbance was read at 490 nm
using a microtiter plate spectrophotometer (Molecular Devices,
Menlo Park, CA) and compared to a standard curve to calculate
cholesterol mass. Radioactivity in the 15 µL aliquots was
analyzed by liquid scintillation counting to correct for choles-
terol extraction efficiency.
concentrated in vacuo. The residue was dissolved in
a
minimum amount of ethyl acetate, and the product was
precipitated with hexanes. The solid was filtered, washed with
hexanes, and dried to afford 77 g of 48 as a white solid (93%):
1
mp 256-257 °C; H NMR (250 MHz, CDCl3) δ 7.35 (d, 2H, J
) 9.0 Hz), 6.88 (d, 2H, J ) 9.0 Hz), 5.45 (s, 1H), 5.3 (m, 2H),
5.0 (m, 2H), 4.7 (d, 1H, J ) 7.0 Hz), 4.6 (m, 2H), 4.5 (dd, 1H,
J ) 8.0, 8.0 Hz), 4.38 (dd, 1H, J ) 11.0, 6.0 Hz), 4.2-3.5 (m,
15H), 3.8 (s, 3H), 3.4 (dd, 1H, J ) 11.0, 10.0 Hz), 2.5 (m, 1H),
2.25 (s, 2H), 2.1-1.0 (m, 25H), 1.0 (s, 3H), 0.94 (d, 3H, J ) 7.0
Hz), 0.78 (d, 3H, J ) 7.0 Hz), 0.7 (s, 3H); FAB MS m/ z 1277
(M + Na+).
(3â,5r,25R)-3-[[2′,2′′,3′,3′′,6′-P en ta k is(ch lor oa cetyl)-â-D-
cellobiosyl]oxy]sp ir osta n -11-on e (49). Trifluoroacetic acid
(19 mL) was added to a solution of compound 48 (23.7 g, 0.019
mol) in CH2Cl2 (150 mL) and MeOH (50 mL). After 4 h, the
mixture was washed with water (3×), NaHCO3 (2×), and brine
(1×), dried (Na2SO4), filtered, and concentrated. The residue
was dissolved in a minimal amount of ethyl acetate and
precipitated with hexanes. The solid was filtered, washed with
hexanes, and dried to afford 19.7 g of 49 as a white solid
(92%): mp 228-229 °C; 1H NMR (250 MHz, CDCl3) δ 5.2 (dd,
1H, J ) 9.0, 9.0 Hz), 5.1 (dd, 1H, J ) 9.0, 9.0 Hz), 4.95 (m,
2H), 4.6 (m, 3H), 4.5 (ddd, 1H, J ) 8.0, 7.0, 7.0 Hz), 4.2-3.4
(m, 20H), 3.35 (dd, 1H, J ) 9.0, 9.0 Hz), 2.9 (d, 1H, J ) 6.0
Hz), 2.45 (m, 1H), 2.2 (s, 2H), 2.1-1.1 (m, 22H), 1.0 (s, 3H),
0.94 (d, 3H, J ) 7.0 Hz), 0.77 (d, 3H, J ) 7.0 Hz), 0.7 (s, 3H);
FAB MS m/ z 1159 (M + Na+).
(3â,5r,25R)-3-[[4′′,6′′-Bis[(2-flu or op h en yl)ca r ba m oyl]-
2′,2′′,3′,3′′,6′-p en ta k is(ch lor oa cetyl)-â-D-cellobiosyl]oxy]-
sp ir osta n -11-on e (50). Cuprous chloride (1.74 g, 18 mmol)
was added to a solution of compound 49 (5.0 g, 4.4 mmol) and
2-fluorophenyl isocyanate (1.98 mL, 18 mmol) in dry DMF (30
mL) at room temperature. After 2 h, the mixture was diluted
with ethyl acetate (100 mL), washed with 1 N HCl (2×) and
brine (1×), dried (Na2SO4), filtered, and concentrated in vacuo.
The residue was dissolved in CH2Cl2 (50 mL), and MeOH (50
mL) was added. The CH2Cl2 was removed in vacuo, and a solid
precipitated from the MeOH. The solid was filtered, washed
with MeOH, and dried to afford 4.82 g of 50 as a white solid
(78%): mp 234-234.5 °C; 1H NMR (250 MHz, CDCl3) δ 7.9
(m, 2H), 7.05 (m, 8H), 5.35 (dd, 1H, J ) 8.0, 7.0 Hz), 5.28 (dd,
1H, J ) 9.0, 8.0 Hz), 5.15 (dd, 1H, J ) 9.0, 9.0 Hz), 5.05 (dd,
1H, J ) 9.0, 8.0 Hz), 4.98 (dd, 1H, J ) 8.0, 7.0 Hz), 4.72 (d,
1H, J ) 9.0 Hz), 4.65-3.4 (m, 21H), 3.35 (dd, 1H, J ) 10.0,
9.0 Hz), 2.45 (m, 1H), 2.2 (s, 2H), 2.1-1.1 (m, 22H), 1.0 (s,
3H), 0.9 (d, 3H, J ) 7.0 Hz), 0.8 (d, 3H, J ) 7.0 Hz), 0.7 (s,
3H); FAB MS m/ z 1433 (M + Na+).
P h a r m a cology. 1. Acu te Ch olester ol Absor p tion Mea -
su r em en t in Ha m ster s. Cholesterol absorption was assessed
using a modification of the method reported by Kelly and Tsai9
as reported by Harwood et al.6d Male golden Syrian hamsters
(∼90-110 g) were fed AIN 76C semipurified diet containing
1% cholesterol and 0.5% cholic acid (ICN Biomedicals, Costa
Mesa, CA) for 3 days and then fasted overnight. The following
morning, they were administered various doses of test com-
pounds suspended in a 10% aqueous ethanol solution contain-
ing 0.25% methyl cellulose and 0.6% Tween 80 as suspending
agents. Dosing volume was 0.5 mL/100 g of body weight.
Immediately afterwards, hamsters received an oral bolus (1.5
mL/hamster) of hamster liquid diet (Bioserve, Frenchtown, NJ )
containing [3H]cholesterol (∼0.5 µCi), unlabeled cholesterol (15
mg), and cholic acid (7.5 mg). The degree of cholesterol
absorption was assessed by measuring the hepatic content of
[3H]cholesterol 24 h later. Livers were excised and saponified
in 5 mL of 2.5 M KOH for 2 h at 70°C. Aliquots (200 µL) of
the saponified livers were decolorized with H2O2 (200 µL of
30%), neutralized with 3 N HCl (200 µL), and then counted in
Refer en ces
(1) (a) Sliskovic, D. R.; Trivedi, B. K. ACAT Inhibitors: potential
antiatherosclerotic agents. Curr. Med. Chem. 1994, 1, 204-225.
(b) Matsuda, K. ACAT Inhibitors as antiatherosclerotic agents:
compounds and mechanisms. Med. Res. Rev. 1994, 14, 271-305.
(2) (a) Lopenzcandales, A.; Bosner, M. S.; Spilburg, C. A.; Lange,
L. G. Cholesterol transport function of pancreactic cholesterol
esterase: directed sterol uptake and esterifiction in enterocytes.
Biochemistry 1993, 32, 12085-12089. (b) Mckean, M. L.; Com-
mons, T. J .; Berens, M. S.; Hsu, P. L.; Ackerman, D. M.; Steriner,
K. E.; Adelman, S. J . Effects of inhibitors of pancreatic choles-
terol ester hydrolase (PCEH) on [14C]-cholesterol absorption in
animal models. FASEB J . 1992, 6, A1388. (c) Gallo, L. L. Sterol
ester hydrolase from rat pancreas. Methods Enzymol. 1981, 71,
664-674.
(3) (a) Salisbury, B. G.; Davis, 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-Polokoff, R.; Smith, A.
A.; Weig, B. C.; Zilli, D. L.; Clader, J . W.; Skybertz, E. J .
Hypocholesterolemic activity of a novel inhibitor of cholesterol
absorption, SCH 48461. Atherosclerosis 1995, 115, 45-63. (b)
Burnett, D. A.; Caplen, M. A.; Davis, H. R., J r.; Burrier, R. E.;
Clader, J . W. 2-Azetidinones as inhibitors of cholesterol absorp-
tion. J . Med. Chem. 1994, 37, 1733-1736.
(4) Himber, J .; Missano, B.; Rudling, M.; Hennes, U.; Kempen, H.
J . Effects of stigmastanyl-phosphocholine (Ro 16-6532) and
lovastatin on lipid and lipoprotein levels and lipoprotein me-
tabolism in the hamster on different diets. J . Lipid Res. 1995,
36, 1567-1585.
(5) Dressman, J . B.; Adair, C. H.; Barnett, J . L.; Berardi, R. R.;
Dunn-Kucharski, V. A.; J arvenpaa, K. M.; Parr, D. D.; Sowle,
C. A.; Swidan, S. Z.; Tobey, S. W.; Reppas, C. High molecular-
weight hydroxypropylmethylcellulose. Arch. Intern. Med. 1993,
153, 1345-1353.
(6) (a) Malinow, M. R. Effects of synthetic glycosides on cholesterol
absorption. Ann. N. Y. Acad. Sci. 1985, 454, 23-27. (b) Sidhu,
G. S.; Upson, B.; Malinow, M. R. Effect of soy saponins and
tigogenin cellobioside on intestinal uptake of cholesterol, cholate
and glucose. Nutr. Rep. Internat. 1987, 35, 615-623. (c) Malinow,
M. R.; Elliott, W. H.; McLaughlin, P.; Upson, B. Effects of

2554 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 16
DeNinno et al.
synthetic glycosides on steroid balance in Macaca fascicularis.
J . Lipid Res. 1987, 28, 1-9. (d) Harwood, H. J ., J r.; Chandler,
C. E.; Pellarin, L. D.; Bangerter, F. W.; Wilkins, R. W.; Long, C.
A.; Cosgrove, P. C.; Malinow, M. R.; Marzetta, C. A.; Pettini, J .
L.; Savoy, Y. E.; Mayne, J . T. Pharmacologic consequences of
cholesterol absorption inhibition: alteration in cholesterol me-
tabolism and reduction in plasma cholesterol concentration
induced by the synthetic saponin â-tigogenin cellobioside (CP-
88,818; tiqueside). J . Lipid Res. 1993, 34, 377-395.
(11) Walinsky, S. W.; Lambert, J . F.; Goggin, K. D. Zinc fluoride-ac-
tivated glycoside syntheses. Tigogenyl â-O-cellobioside as a novel
cholesterol lowering agent. Abstracts from the 208th ACS
National Meeting, Washington, DC, August 1994; CARB 11.
(12) Toshima, K.; Tatsuta, K. Recent progress in O-glycosidation
methods and its application to natural products synthesis. Chem.
Rev. 1993, 93, 1503-1531.
(13) Duggan, M. E.; Imagire, J . S. Copper(I) chloride catalyzed
addition of alcohols to alkyl isocyanates. A mild and expedient
method for alkyl carbamate formation. Synthesis 1989, 131-
132.
(14) McCarthy, P. A.; DeNinno, M. P.; Morehouse, L. A.; Chandler,
C. E.; Bangerter, F. W.; Wilson, T. C.; Urban, F. J .; Walinsky,
S. W.; Cosgrove, P. G.; Duplantier, K. C.; Etienne, J . B.; Fowler,
M. A.; Lambert, J . F.; O’Donnell, J . P.; Pezzullo, S. L.; Watson,
H. A., J r.; Wilkins, R. W.; Zaccaro, L. M.; Zawistoski, M. P. 11-
Ketotigogenin cellobioside (pamaqueside): a potent cholesterol
absorption inhibitor in the hamster. J . Med. Chem. 1996, 39,
1935-1937.
(7) Dujovne, C. A.; Harris, W. S.; Gelfand, R. A.; Morehouse, L. A.;
McCarthy, P. A.; Chandler, C. A.; DeNinno, M. P.; Harwood, H.
J ., J r.; Newton, F. A.; Shear, C. L. Investigations of the effects
of synthetic saponins on cholesterol absorption and serum
cholesterol levels. In Atherosclerosis X; Woodford, F. P., Davi-
gnon, J ., Sniderman, A., Eds.; Elsevier Science: New York, 1995;
pp 298-306.
(8) Turley, S. D.; Spady, D. K.; Dietschy, J . M. Regulation of fecal
bile acid excretion in male golden Syrian hamsters fed a cereal-
based diet with and without added cholesterol. Hepatology 1997,
25, 797-803 and references cited therein.
(9) Kelly, J . J .; Tsai, A. C. Effect of pectin, gum arabic and agar on
cholesterol absorption, synthesis and turnover in rats. J . Nutr.
1978, 108, 630-639.
(10) To achieve the necessary throughput of a primary screen, the
radioactivity of the plasma compartment was not measured
which historically only contained a fraction of the counts found
in the liver. Also to save processing time, the livers were
saponified instead of homogenized. Control experiments con-
firmed that these modifications did not alter ED50 values. See
the Experimental Section for complete details.
(15) (a) Grundy, S. M.; Mok, H. Y. I. In Lipoprotein Metabolism;
Greten, H., Ed.; Springer-Verlag: New York, 1976; pp 112-118.
(b) Lees, R. S.; Lees, A. M. In Lipoprotein Metabolism; Greten,
H., Ed.; Springer-Verlag: New York, 1976; pp 119-124. (c)
Oster, P.; et. al. In Lipoprotein Metabolism; Greten, H., Ed.;
Springer-Verlag: New York, 1976; pp 125-130.
(16) Chandler, C. E.; Morehouse, L. A. Unpublished results.
J M9702600