Synthesis and in vitro evaluation of inhibitors of intestinal cholesterol absorption

J. Med. Chem. 2005, 48, 6035-6053

6035

Synthesis and in Vitro Evaluation of Inhibitors of Intestinal Cholesterol

Absorption

Lisbet Kværnø,^†Moritz Werder,^‡Helmut Hauser,^‡and Erick M. Carreira*^,†

Laboratorium fu¨r Organische Chemie der ETH-Zu¨rich, HCI H 335, Wolfgang Pauli Strasse 10, CH-8093, Zurich, Switzerland,

and Lipideon Biotechnology AG, Hermann-Hiltbrunnerweg 25, CH-8713 Uerikon, Switzerland

Received May 4, 2005

We have utilized our recently developed in vitro assay to address two key questions in the

design of small-molecule cholesterol absorption inhibitors using ezetimibe, the only drug yet

approved for the inhibition of cholesterol absorption in the small intestine, as a starting point:

(1) the role of glycosylation and (2) the importance of the â-lactam scaffold of ezetimibe for

inhibitory activity. A wide range of nonhydrolyzable phenolic glycosides of ezetimibe were

synthesized and demonstrated to be active inhibitors of cholesterol absorption using the brush

border membrane vesicle assay. The analogous azetidines provided access to a variety of

inhibitors in vitro, suggesting that the â-lactam of ezetimibe merely serves as a ring scaffold

to appropriately position the required substituents. Our findings highlight several promising

strategies for the design of alternative small-molecule cholesterol absorption inhibitors that

could ultimately be useful in preventing cardiovascular disease by lowering blood cholesterol

levels.

Introduction

High blood-cholesterol levels constitute a major risk

factor for cardiovascular disease,^1,2the leading cause

of death in the Western industrialized world.³The total

blood cholesterol level is primarily regulated by two

complementary mechanisms: (1) cholesterol biosynthe-

sis in the liver and (2) absorption of dietary cholesterol

in the small intestine.^4,5Since their introduction in the

Figure 1. Development of ezetimibe (1) from SCH48461 (2).

late 1980s, statins have by far become the predominant

structure-activity relationships,^11,13-15and the optimi-

class of current lipid-lowering drugs (96% of total sales

zation of the inhibitory activity was consequently ac-

in 2001).⁶Statins inhibit HMG-CoA reductase, the

complished by relying solely on a 7-day cholesterol-fed

enzyme that catalyzes the rate-limiting step of choles-

hamster model^14,16as a screen for active compounds. It

terol biosynthesis in the liver.^2,7However, the patient

is important to note that in such an in vivo optimization

response to statins varies greatly, with half of all

process, the intrinsic activity of a compound is indis-

patients on statin therapies failing to reach their

tinguishable from differences in bioavailability and/or

cholesterol goals.^2,8Ezetimibe (1, Figure 1), which was

ease of conversion into active metabolites,^13,17making

approved in late 2002 for use either alone and in

it a very challenging medicinal chemistry effort.¹⁰In

combination with a statin, is the only example to date

1994, the â-lactam SCH48461 (2, Figure 1) was reported

of a drug that involves inhibition of intestinal cholesterol

to effect remarkable reduction in liver cholesterol ester

absorption.^9-11

(LCE) in the 7-day cholesterol-fed hamster model (LCE

The discovery and development of ezetimibe is an

ED₅₀2.2 mg/kg/day).¹⁴By a combination of isolation of

interesting story, as the â-lactam scaffold was investi-

highly active metabolites of SCH48461 (2)¹⁸and the

gated within a program aimed at identifying cholesterol

sites of metabolism through an extensive synthesis

acylCoA:acyltransferase (ACAT) inhibitors. ACAT is the

effort, this original lead structure was optimized as

enzyme in the enterocytes that is responsible for esteri-

fying cholesterol before its assembly into chylomicrons.

However, in the initial studies no correlation between

the high in vivo activity and the in vitro inhibition of

summarized in Figure 1 into SCH58235 (1),¹¹which

later came to be known as ezetimibe. Compared to

SCH48461 (2), ezetimibe (1) showed a 50-fold increase

in activity in hamsters (ED₅₀0.04 mg/kg/day)¹¹and as

ACAT was observed. Further experimentation led to the

much as a 400-fold increase (ED₅₀0.0005 mg/kg/day)¹⁹

conclusion that these â-lactams inhibit intestinal cho-

in the cholesterol-fed rhesus monkey.¹⁸

The process of cholesterol absorption in the small

intestine was traditionally viewed as exclusively pro-

lesterol absorption by a unique mechanism that is

upstream or preceding ACAT’s site of action.^10,12,13The

unknown protein target nevertheless elicited consistent

ceeding via passive diffusion. This paradigm has been

negated by more recent evidence, such as the existence

of highly efficacious small-molecule cholesterol absorp-

* Corresponding author. Phone: +41 446322830; Fax: +41

446321328. E-mail: carreira@org.chem.ethz.ch.

tion inhibitors, which strongly suggests cholesterol

^†Laboratorium fu¨r Organische Chemie der ETH-Zu¨rich.

^‡Lipideon Biotechnology AG.

absorption to be protein-mediated.^5,10,20-23The identity

Published on Web 08/25/2005

6036 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

Figure 2. Schematic outline of the preparation of brush border membrane vesicles

of the intestinal proteins effecting cholesterol absorption

is still a matter of active debate. A number of proteins

have been suggested to be involved in cholesterol

transport in the small intestine, including the scavenger

receptors SR-BI^21,22,24,25and CD36,^21,24,26an annexin

2/caveolin 1 complex,²⁷and most recently Niemann-

Pick C1-like 1 (NPC1L1) protein.^28-30

Ezetimibe (1) inhibits intestinal cholesterol absorption

by a mechanism that is currently not fully elucidated

at a molecular level.^10,28NPC1L1 knock-out (-/-) mice

showed significantly lowered cholesterol uptake com-

pared to wild-type mice, and no further reduction of

cholesterol absorption was observed when NPC1L1

knock-out (-/-) mice were treated with ezetimibe.

These observations led to the suggestion of NPC1L1 as

the molecular target of ezetimibe. However, reconstitu-

tion of cholesterol uptake into cells overexpressing

NPC1L1 was unsuccessful, suggesting the involvement

of additional proteins in intestinal cholesterol absorp-

tion.^28,29More recently, the development of an in vitro

binding assay using a number of ezetimibe analogues

further strongly indicated NPC1L1 to be critically

involved. Thus, these ezetimibe analogues were shown

to bind to the brush border membranes of several

species and to cells expressing NPC1L1 with virtually

identical binding affinities. In contrast, no binding was

observed to the brush border membrane of NPC1L1

knock-out mice.³⁰However, the direct binding of

ezetimibe analogues to NPC1L1 using radiolabeled and

fluorescent analogues remains elusive.²⁸In a recent

intriguing finding, it was alternatively suggested that

NPC1L1 is not located on the brush border membrane

but inside the enterocyte.³¹This latter finding highlights

the complexity of intestinal cholesterol absorption,

indicating a multistep process that presumably involves

a number of proteins. In independent recent investiga-

tions using photoreactive ezetimibe analogues,³²

ezetimibe was demonstrated to interact with aminopep-

tidase N (CD13) on the enterocyte brush border mem-

brane. However, aminopeptidase N, which is also in-

volved in the uptake of viruses, does not mediate

cholesterol absorption, and it remains to be established

whether aminopeptidase N is a vital molecular target

of ezetimibe.³³

Figure 3. Relative correlation of inhibitory activities

in vitro³⁶and in vivo.^13,17

Consequently, cholesterol uptake into such vesicles

takes place regardless of the actual identity of the

proteins involved in intestinal cholesterol uptake. We

have successfully validated our brush border membrane

vesicle assay by correlation of the relative inhibitory

activities of a number of known ezetimibe analogues for

which the corresponding in vivo data were available

(Figure 3).³⁶We now report a full account of our findings

using this in vitro assay to evaluate a number of novel

small molecules as potential inhibitors of cholesterol

absorption. These new inhibitors can be divided into two

general compound classes: (1) ezetimibe derivatives in

which the phenol is conjugated to various carbohydrates

or derivatized with simple substituents and (2) ana-

logues of ezetimibe with the â-lactam of ezetimibe

replaced by the corresponding azetidine. These struc-

ture-activity-driven investigations have resulted in a

number of tangible results of immediate relevance to

understanding ezetimibe’s mode of action and, addition-

ally, may lead to potentially wider application.

Given our expertise with brush border membrane

vesicles as an in vitro model to mimic intestinal

cholesterol uptake,^22-24,34,35we initiated a program

aimed at the development of an in vitro screening assay

for evaluating small-molecule inhibitors of cholesterol

uptake.^36,37An important advantage of using brush

border membrane vesicles is their ready preparation

from animal or human whole small intestine, leaving

the relevant membrane receptors intact (Figure 2).^23,38

Results and Discussion

A number of phenolic O-glycosides of ezetimibe (1)

such as the cellobioside 4 (Figure 3) display similar or

increased activity profiles both in vivo^17,39and in vitro.³⁶

Furthermore, ezetimibe has been shown to undergo

metabolism in the intestine within 1 min into its

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6037

Figure 4. Inhibition in the brush border membrane vesicle assay (%) using rabbit small intestine at nominal concentrations of

6 µM. Average standard deviations were (3% inhibition.

glucuronide derivative 20 (Figure 4) before entering the

portal plasma.³⁹Due to enterohepatic cholesterol recy-

cling,^4,5the glucuronide 20 is transported in bile back

to the small intestine, where it repeatedly exerts its

mode of action.³⁹The exact nature of the active species,

i.e. whether ezetimibe (1), its glucuronide 20, or both

are active as cholesterol absorption inhibitors, remains

however elusive. We were intrigued by the role of this

phenolic glycosylation with respect to inhibitory activity.

To rule out rapid metabolic hydrolysis of the glycosidic

bond, we synthesized the structurally closely related

glycosides 7, 9, and 11 with nonhydrolyzable ether

linkages (Scheme 1).^36,40,41In the synthetic sequence we

developed, the phenol of ezetimibe (1)⁴²was transiently

protected as the acetate ester by treatment with Ac₂O

and NaOH in iPrOH⁴³(97% yield), following silylation

of the secondary alcohol (91% yield). A number of

reaction conditions were investigated for the subsequent

deacetylation of the phenol. Adsorption of the phenol

acetate on neutral alumina and thermal heating to 70

°C⁴⁴for 5 h gave phenol 5 in 83% yield. This protocol

was shown to be superior to KCN/MeOH,⁴⁵NaHCO₃/

MeOH,⁴⁶Et₃N/MeOH/H₂O,¹⁷and adsorption on neutral

alumina and heating (3 min) under microwave irradia-

tion.⁴⁴In the course of these studies we observed that

the secondary silyl ether was prone to elimination under

a variety of different basic reaction conditions. Conse-

quently, this base-sensitivity of the protected ezetimibe

derivative governed the strategy for the subsequent

introduction of the C-glycoside subunits in the desired

targets.

dipiperidine/Bu₃P combination^41,47effected conversion

of the starting materials. Although the yields obtained

were unsatisfactory (∼20%), enough material could be

obtained to allow testing of these compounds. The source

of the modest yields was due to a predominant side

reaction involving opening of the â-lactam ring by the

hydrazide anion generated in the course of the reaction,

highlighting the susceptibility of â-lactams with this

substituent pattern to nucleophilic opening. In the

subsequent deprotections, hydrogenolysis of the benzyl

ethers was performed prior to the desilylation to furnish

the ether-linked glycosides 7, 9, and 11 in good yields.

Following the synthesis of the C-glycosides, we then

sought to examine an alternative, novel conjugation

strategy, which would allow the use of mild conditions

compatible with the base-sensitive core molecule. In

developing such a conjugation strategy, it was important

to identify a method for introduction of the carbohy-

drates that would obviate issues relating to reactivity

and diastereoselectivity associated with traditional gly-

cosylation methods.⁵¹We decided to examine the use of

carbohydrate-derived sulfonyl chlorides, because a con-

jugation strategy based on sulfonate ester formation

with a phenol would meet both stated requirements.³⁷

Related precedence for the use of a sulfonylation reac-

tion was reported in the preparation of sulfonate-linked

oligonucleotides^52,53while, to the best of our knowledge,

there have been no previous reports describing conjuga-

tion to simple carbohydrates.⁵⁴

Sulfonylated carbohydrate derivatives 14, 16, and 18

(Schemes 2 and 3)³⁷were assembled following a straight-

forward sequence of reactions. It is worth noting that

the selection of these carbohydrates was based on our

desire to introduce a minimum of structural modifica-

tions on the candidates for study, apart from the

Reaction of the free phenol with primary alcohols

under the standard Mitsunobu conditions (DEAD, Ph₃P)

for the construction of the ether bond did not afford any

conversion. However, the use of the 1,1′-(azodicarbonyl)-

6038 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

Scheme 1^a

^a(a) 1,⁴²Ac₂O, NaOH, iPrOH, 97%; (b) TBDMSCl, imid, DMF, 91%; (c) Al₂O₃(neutral), 70 °C, 83%; (d) 6,⁴⁸1,1′-(azodicarbonyl)dipiperidine,

Bu₃P, THF, 21%; (e) H₂, Pd(OH)₂/C, EtOAc/EtOH, 87%; (f) HF·pyridine, pyridine, THF, 86%; (g) 8,⁴⁹1,1′-(azodicarbonyl)dipiperidine,

Bu₃P, THF, 21%; (h) H₂, Pd(OH)₂/C, EtOAc/EtOH, 59%; (i) HF·pyridine, pyridine, THF, 62%; (j) 10,⁵⁰1,1′-(azodicarbonyl)dipiperidine,

Bu₃P, THF; (k) H₂, Pd(OH)₂/C, EtOAc/EtOH, 21%, two steps; (l) HF‚pyridine, pyridine, THF, 65%.

Scheme 2^a

Scheme 3^a

^a(a) MsCl, pyridine, CH₂Cl₂, 94%; (b) AcSK, EtOH, reflux, 96%;

(c) Oxone, AcOK, AcOH, 90%; (d) SOCl₂, Ph₃P, CH₃CN, CH₂Cl₂,

95%; (e) 5, pyridine, CH₂Cl₂, 91%; (f) H₂, Pd(OH)₂/C, EtOH, 81%;

(g) HF·pyridine, pyridine, THF, 90%.

sulfonate linkages, when compared to the ether-linked

C-glycosides 7, 9, and 11. The synthesis commenced

with mesylation of the alcohol in 6⁴⁸(MsCl, pyridine,

94% yield) and conversion into its thioacetate 12⁵⁵in

96% yield (Scheme 2). Thioacetate 12 was oxidized with

oxone⁵⁶to give the intermediate sulfonate salt⁵⁷(90%).

Conversion into the corresponding sulfonyl chloride 13

was achieved in 95% yield using SOCl₂/Ph₃P^52,53as a

mild reaction protocol. It has been previously reported

that the conversion rates in CH₂Cl₂to the sulfonyl

chlorides were much higher using more soluble tet-

raalkylammonium sulfonate salts instead of alkali-

metal sulfonate salts.⁵³In the present case, addition of

CH₃CN as a polar cosolvent likewise facilitated a

smooth transformation to 13. This aliphatic sulfonyl

chloride was only subjected to filtration through a short

silica gel plug to avoid any partial hydrolysis, before it

was used directly in the sulfonylation reaction.

^a(a) MsCl, pyridine, CH₂Cl₂, 99%; (b) AcSK, EtOH, reflux, 92%;

(c) Oxone, AcOK, AcOH, 55%; (d) SOCl₂, Ph₃P, CH₃CN, CH₂Cl₂,

quant.; (e) 5, pyridine, CH₂Cl₂, 92%; (f) H₂, Pd(OH)₂/C, EtOH, 90%;

(g) HF·pyridine, pyridine, THF, 37%; (h) MsCl, pyridine, CH₂Cl₂,

86%; (i) AcSK, EtOH, reflux, 79%; (j) Oxone, AcOK, AcOH, 96%;

(k) SOCl₂, Ph₃P, CH₃CN, CH₂Cl₂, 66%; (l) 5, pyridine, CH₂Cl₂; (m)

H₂, Pd(OH)₂/C, EtOH, 52% (two steps); (n) HF·pyridine, pyridine,

THF, 62% 18.

DMAP or alternatively Et₃N was used as the activating

amine base. The order of the final deprotection steps

(hydrogenolysis of the carbohydrate benzyl ether pro-

tecting groups followed by desilylation) proved crucial,

because desilylation prior to the hydrogenolysis of the

benzyl protecting groups led to concomitant hydro-

genolytic removal of the benzylic hydroxyl group. In-

terestingly, the benzylic silyloxy group is considerably

more resistant to hydrogenolysis. In this regard, pro-

tecting group debenzylation (81% yield) followed by

The coupling of sulfonyl chloride 13 to the phenol of

5 proceeded in 91% yield using pyridine only as the

catalyst while the yields were considerably lower when

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6039

Figure 5. Geometry-optimized overlays of ezetimibe (1) and the azetidine 23 (right, side-view left) with the flexible C3 side

chain omitted for clarity.

desilylation using HF‚pyridine smoothly afforded the

sulfonylated glycoside 14 in 90% yield.

â-lactam is indeed an essential pharmacophore or

whether it serves merely as a ring scaffold to ap-

propriately position the required substituents in the

context of the optimized substituents of ezetimibe (1).

In this regard, we chose to reduce the â-lactam to the

corresponding azetidine in order to investigate whether

the carbonyl oxygen plays a pivotal role in inhibitory

activity. When comparing the spatial arrangement of

the substituents around these two ring scaffolds, the

overlays of the geometry-optimized structures (ab initio

minimization, B3LYP/6-31G*)⁶⁰of the â- lactam in 1

with azetidine 23 illustrate that the slightly puckering

of the azetidine results in only minor perturbations of

the positioning of the side chains compared to the more

planar â-lactam (Figure 5). On the basis of this analysis,

the azetidine ring could potentially serve as an efficient

replacement of the â-lactam if the hydrogen-bond-

accepting capabilities of the carbonyl oxygen were not

essential. Such an investigation could shed further light

on the important structural question of whether the

â-lactam itself is important for activity or whether it

potentially could be substituted by appropriate ring

scaffolds that retain the required exit vectors of the

substituents.

The sulfonate-linked glycoconjugates 16 and 18 were

subsequently prepared by analogous reaction sequences

from the alcohols 8⁴⁹and 10,⁵⁰respectively, via sulfonyl

chlorides 15 and 17 (Scheme 3). However, in the

disaccharide case prolonged hydrogenolysis afforded the

deoxygenated disaccharide 19 as a side product, which

could be separated from the major product 18 after

desilylation.

The various novel glycosides of ezetimibe were sub-

sequently evaluated in the brush border membrane

vesicle assay^36,58and shown to be efficient inhibitors of

intestinal cholesterol absorption (Figure 4). The evalu-

ation furthermore revealed that their inhibitory activi-

ties were largely independent of the nature of the

linkage between the phenol and the carbohydrate

domain. The O-monosaccharides 20^17,36and 21^17,36(19%

and 14% inhibition, respectively), the ether-linked

monosaccharides 7 and 9 (13% and 15% inhibition,

respectively), and the sulfonate-linked monosaccharides

14 and 16 (20% and 15% inhibition, respectively) were

all of comparable inhibitory activity. The highest effica-

cies in each series were obtained for the cellobioside

derivatives 4^17,36(27% inhibition), 11 (28% inhibition),

and 18 (41% inhibition), demonstrating a correlation

between the size of the carbohydrate moiety and the

inhibitory activity.

A structural feature that additionally proved vital for

inhibitory activity was the benzylic hydroxyl group, as

also demonstrated in the developmental work toward

ezetimibe (1) using the animal screen (Figure 1).¹¹As

an additional correlation between our brush border

membrane vesicle results and in vivo data, deletion of

this hydroxyl group (22³⁶vs 4, and 19 vs 18) likewise

resulted in drastic decreases of the in vitro inhibitory

activity (27% f 3% and 41% f 10% inhibition, respec-

tively). In summary, the in vitro activities of these

nonhydrolyzable glycosides 7, 9, 11, 14, 16, and 18

suggest that a wide range of non-natural glycosides of

ezetimibe are indeed capable of being potent cholesterol

absorption inhibitors.

As part of the investigations of the ezetimibe precur-

sor SCH48461 (2), the â-lactam ring was suggested to

be an integral and essential pharmacophore for inhibi-

tion of cholesterol absorption,^13,15with the ring-opened

â-amino acid derivative of 2 being completely void of

activity.¹³Beyond these observations, the â-lactam was

retained in all subsequent structural modifications

resulting in the development of ezetimibe (1).^11,59Given

our convenient in vitro brush border membrane vesicle

assay, we decided to revisit the question of whether the

A most convenient means of examining the replace-

ment of the â-lactam ring by an azetidine was to effect

reduction of ezetimibe. The suitable reductant for the

conversion of N-substituted â-lactams to azetidines

without effecting ring cleavage has been reported to be

AlH₂Cl.⁶¹In a simple model study with the appropriate

substituent pattern of the â-lactam ring (eq 1), racemic

â-lactam (()-3⁶²was thus smoothly converted into the

azetidine (()-24 in 81% yield, using AlH₂Cl prepared

in situ from AlCl₃and LiALH₄. As observed for the

inactive â-lactam (()-3 when evaluated in the brush

border membrane vesicle assay (2% inhibition),³⁶aze-

tidine (()-24 likewise proved void of activity (2% inhibi-

tion).

In contrast to the reduction of â-lactam (()-3, treat-

ment of ezetimibe (1) with AlH₂Cl afforded a complex

mixture of products. In examining the reduction further,

the hydroxy groups were protected as their TBDMS

6040 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

Scheme 4^a

^a(a) TBDMSCl, imid, DMF, 97%; (b) AlH₂Cl, ether, 0 °C, 63% 26, 16% 28; (c) LiAlH₄, THF, 0 °C, 79%; (d) CBr₄, Ph₃P, Et₃N, CH₃CN,

60%.

Scheme 5^a

^a(a) MsCl, pyridine, CH₂Cl₂, 92%; (b) AlH₂Cl, ether, 0 °C, 77%; (c) HF·pyridine, pyridine, THF, 85%; (d) TBAF, THF, 68%; (e) H₂,

Pd(OH)₂/C, EtOH/EtOAc, 33%.

ethers in 97% yield (25, Scheme 4). Interestingly,

treatment of the silyl-protected ezetimibe derivative 25

with AlH₂Cl either at reflux in ether or at 0 °C effected

conversion primarily to the bicyclic aniline 26, with no

azetidine 30 isolated. As a working hypothesis, we

believe that 30 is formed as an intermediate that

undergoes ring opening to give an intermediate benzylic

carbocation 27, a process that is greatly facilitated by

electron-donating substituents of the corresponding

arene. This carbocation can subsequently undergo

Friedel-Crafts reaction with the adjacent aniline to

furnish the trans-substituted bicycle 26. The interme-

diacy of this carbocation finds support in the formation

of alkene 28, which was isolated in 16% yield. This may

be formed directly by fragmentation of the carbocation

27, via a retro-aza-Prins fragmentation. Given the

difficulties encountered and the dependency of the direct

reduction strategy on subtle structural variations, an

alternative strategy was pursued. Reductive ring open-

ing of 25 with LiAlH₄furnished 1,3-amino alcohol 29

in 79% yield. Subsequent reclosure of the ring by

conversion into the activated phosphonium salt via the

Appel reaction (CBr₄, PPh₃, and Et₃N in CH₃CN)

successfully yielded azetidine 30 in 60% isolated yield.

Unfortunately, azetidine 30 was observed to be rather

unstable when compared to (()-24. During chromatog-

raphy on silica gel, azetidine 30 underwent fragmenta-

tion to give alkene 28; moreover, it was not possible to

find desilylation conditions that left the azetidine ring

intact to give the desired 23.

deficient phenol derivative could lead to a stable azeti-

dine analogue. In this respect, we had previously shown

that methane sulfonate esters are stable to a wide

variety of chemical transformations and can be conve-

niently removed with LDA at -78 °C.⁶³Mesylation of

phenol 5 with MsCl and Et₃N afforded a dimesylate

(data not shown), while MsCl and pyridine cleanly

furnished mesylate 31 in 92% yield (Scheme 5). Subse-

quent AlH₂Cl-promoted ring reduction proceeded

smoothly (77% yield) with 31, which following depro-

tection with HF‚pyridine furnished mesylated azetidine

32 in 85% yield. For comparison of biological activities,

the corresponding mesylated â-lactam 33 was synthe-

sized. When evaluated in the brush border membrane

vesicle assay, both mesylated azetidine 32 and mes-

ylated â-lactam 33 proved to be active inhibitors of

intestinal cholesterol absorption (22% and 30% inhibi-

tion, respectively). This notable preliminary finding

suggested that the azetidine ring scaffold can effectively

replace the â-lactam in small-molecule cholesterol ab-

sorption inhibitor analogues of ezetimibe. As a means

of examining the significance of the rigid azetidine

scaffold in correctly positioning the side chains, azeti-

dine 32 was converted into the acycle 34 by hydro-

genolytic cleavage of the benzylic C-N bond. This

acyclic counterpart 34 of the azetidine 32 displayed no

inhibitory activity when tested in the brush border

membrane vesicle assay (<2% inhibition), in excellent

agreement with the reported lack of activity in animals

for acyclic â-amino acids.¹³

The stability of phenyl-substituted azetidine 24 sug-

gested that the electron-rich nature of the arene in 30

was the cause of its observed instability. We reasoned

that conversion of the electron-rich phenol to an electron-

We were intrigued by the seemingly positive effect of

the mesyl substituent in enhancing the activity of

ezetimibe with respect to inhibition of cholesterol ab-

sorption. In this regard, a number of additional simple

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6041

Scheme 6

Scheme 7^a

^a(a) PhSO₂Cl, pyridine, CH₂Cl₂, 70%; (b) HF·pyridine, pyridine,

THF, 93%; (c) AlH₂Cl, ether, 0°C, 40%; (d) Me₂CdCH₂, TfOH, 87%,

(e) KCN, MeOH, 71%.

sulfonates were examined. The benzenesulfonylated

â-lactam 35 and the corresponding azetidine 36 were

thus synthesized through a sequence including sulfonate

ester formation (70% yield), desilylation (93% yield), and

ring reduction (40% yield) as outlined in Scheme 6. Both

benzenesulfonylated compounds 35 and 36 were shown

to be of comparable inhibitory activity (26% and 19%

inhibition, respectively) as the mesylates 33 and 32

(30% and 22% inhibition, respectively) when evaluated

in the brush border membrane vesicle assay. As a

further test, the tert-butyl-substituted inhibitor 38, as

a steric analogue of the mesylate 33, was synthesized

from the monoacetate 37¹⁷and subjected to evaluation

as a potential cholesterol absorption inhibitor. A con-

siderably lower inhibitory activity was obtained for 38

compared to the mesylate 33 (10% and 30% inhibition,

respectively), suggesting a favorable interaction between

the sulfonate linkage and the receptor at the brush

border membrane.

The electron-withdrawing nature of the linkage in the

sulfonyl glyconjugation approach described above

(Schemes 2 and 3) made it facile to investigate azetidine

glycoconjugates as cholesterol absorption inhibitors. As

outlined in Scheme 7, reduction of the â-lactam ring was

accomplished using AlH₂Cl to give azetidine sulfonyl

conjugates 39 and 41 in 78 and 94% yield, respectively,

following deprotection.³⁷Interestingly, ring reduction of

the corresponding sulfonylated cellobioside 18 afforded

a complex mixture of products that were inseparable

by standard chromatography on silica gel, suggesting a

different reactivity behavior compared to the sulfon-

ylated glycosides 14 and 40. When evaluated in the

brush border membrane vesicle assay, we were pleased

to observe that both azetidine glycoconjugates 33 and

35 displayed high activities as cholesterol absorption

inhibitors (27% and 20% inhibition, respectively) similar

to that observed for the â-lactams counterparts 21 and

22 (20% and 15% inhibition, respectively). Thus, the use

of a sulfonyl linkage to the carbohydrate moiety permit-

ted the synthesis of stable glycosylated azetidine ana-

logues of ezetimibe that were also active in vitro as

cholesterol absorption inhibitors.

^a(a) AlH₂Cl, ether, 0 °C, 78%; (b) AlH₂Cl, ether, 0 °C, 94%; (c)

HF·pyridine, pyridine, THF, 69%.

sulted in a detrimental effect on the inhibitory activity

(3% inhibition) compared to the azetidine counterpart

39 (27% inhibition). As observed in our earlier studies,

when desilylation of the secondary ether was performed

prior to the debenzylation of the carbohydrate domain,

it proved possible to isolate the deoxygenated azetidine

conjugate 45. In analogy to the deoxygenated cellobio-

sides 22 and 19, lack of the benzylic hydroxyl stereo-

center in 45 proved detrimental for inhibitory activity

(4% inhibition) compared to the azetidine glycoconjugate

39 (27% inhibition). These latter findings highlighted

the importance of an appropriate ring scaffold to cor-

rectly position the ring substituents as well as the

significance of the benzylic hydroxy group for inhibitory

activity, as expected from in vivo data of related

â-lactam derivatives.

Conclusion

We have prepared a number of derivatives of ezetimibe

(1), the first example of a cholesterol-lowering drug that

inhibits cholesterol absorption in the small intestine.

Our recently established brush border membrane vesicle

assay allowed us to rapidly address two key questions

for derivatives of this cholesterol absorption inhibitor:

(1) the role of glycosylation and (2) the importance of

the â-lactam as a scaffold. In the context of our

investigations, a new glycoconjugation method using

sulfonate ester formation was developed and utilized in

the synthesis of a number of phenolic glycoside deriva-

tives of ezetimibe. These nonhydrolyzable glycosides

proved active as inhibitors of cholesterol absorption in

the brush border membrane assay, demonstrating that

the phenol of ezetimibe tolerates a wide range of

structural modifications in molecules that retain inhibi-

tory activity. We furthermore investigated the reduction

of the â-lactam to the corresponding azetidine to assess

whether the â-lactam merely serves as a ring scaffold

In another demonstration of the importance of the

azetidine ring scaffold in optimally positioning the side

chains, azetidine glycoside 43 was converted into the

acyclic aniline 44 by hydrogenolytic cleavage of the

benzylic C-N bond followed by desilylation (Scheme 8).

This ring-opening to analogue glycoconjugate 44 re-

6042 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

Scheme 8^a

^a(a) AlH₂Cl, ether, 0 °C, 81%; (b) H₂, Pd(OH)₂/C, EtOH/EtOAc, 85%; (c) HF·pyridine, pyridine, THF, 73%; (d) TBAF, THF, 63%; (e) H₂,

Pd(OH)₂/C, EtOH/EtOAc, 31%.

was accomplished by dry column vacuum chromatography⁶⁶

to appropriately position the required substituents. In

on either Merck silica gel 60 (15-40 µm) or Brunschwig silica

the course of these studies the electron-withdrawing

nature of sulfonate ester linkage proved crucial to

achieve stability of the desired azetidine derivatives and

allowed the facile synthesis of several azetidine glyco-

18-32, 60 Å (18-32 µM). Concentration under reduced pres-

sure was performed by rotary evaporation at 40 °C, and the

purified compounds were subsequently dried under high

vacuum (<0.5 Torr). NMR spectra were recorded on a Varian

conjugates. Importantly, all the azetidines were dem-

onstrated to be as efficacious as their â-lactam coun-

terparts in inhibiting cholesterol absorption in vitro.

This latter finding suggests that the â-lactam of ezetimibe

can be easily replaced by alternative ring scaffolds. Our

findings suggest that a number of structural modifica-

tions based on the ezetimibe scaffold result in effective

cholesterol absorption inhibitors. The modest response

of ezetimibe as an inhibitor of cholesterol absorption

seems to be a general phenomenon in various in vitro

systems, as it has also been observed with Caco-2 cells,³³

CHO cells,^25cand COS-7 cells.²⁶One explanation to

ezetimibe’s impressingly high in vivo efficacy could be

its constant recycling in bile (after metabolism into its

glucuronide 20) back to the site of action, the small

intestine, as previously suggested.³⁹A detailed correla-

tion between in vivo inhibitory efficacies and the in vitro

inhibition results obtained in the brush border mem-

brane vesicle assay are part of ongoing investigations

and will be reported in due course.⁶⁴The alternative

strategies we thus devise for the design of new potent

cholesterol absorption inhibitors could ultimately find

use in the development of improved strategies for

lowering blood cholesterol levels and preventing car-

diovascular disease. Additionally, the use of phenol-

derived sulfonate esters as a conjugation strategy may

have broader applications in the design of bioactive

molecules.

Mercury 300 MHz apparatus operating at 300, 75, and 282

MHz for ¹H, ¹³C/DEPT and ¹⁹F, respectively, and chemical

shifts (δ) were referenced to the internal solvent signals for

¹H and ¹³C. Multiplicities are reported as follows: ¹H, s )

singlet, d ) dublet, t ) triplet, q ) quartet, m ) multiplet;

¹³C, C, CH, CH₂, CH₃(determined by DEPT). Coupling

constants are reported in hertz. IR spectra were recorded in

CHCl₃on a Perkin-Elmer Spectrum RX I FT-IR apparatus

(thin films on NaCl plates) and are reported as absorption

maxima in cm^-1. Elemental analysis was performed by the

Mikroelementaranalytisches Laboratorium at the ETH, Zu¨r-

ich. High-resolution matrix-assisted laser desorption ionization

mass spectrometry (MALDI-MS) and electrospray ionization

(ESI-MS) were performed by the mass spectrometry service

of the LOC at the ETH, Zu¨rich.

Phenol 5. Ezetimibe (1)⁴²(5.530 g, 13.5 mmol) was sus-

pended in 2-propanol (70 mL), to which aqueous NaOH (2 M,

15 mL) followed by Ac₂O (3.0 mL, 32 mmol) was added, and

the solution was stirred for 5 h followed by addition of

saturated aqueous NaHCO₃(200 mL). After extraction with

EtOAc (4 × 50 mL), the combined organic layer was washed

with saturated aqueous NaHCO₃(50 mL) and H₂O (50 mL),

evaporated on Celite, and purified by dry column vacuum

chromatography (5.2 × 5.5 cm) on silica gel eluting with a

gradient of 0-100% EtOAc in hexane (v/v) to give the

intermediary phenolic acetate (5.930 g, 97%) as a white foam.

R_f(1:1 EtOAc/hexane (v/v)): 0.35. ¹H NMR (300 MHz, CDCl₃)

δ: 7.31 (2H, d, J ) 8.7 Hz), 7.29-7.18 (4H, m), 7.09 (2H, d, J

) 8.7 Hz), 6.99 (2H, t, J ) 8.7 Hz), 6.92 (2H, t, J ) 8.7 Hz),

4.67 (1H, bs), 4.61 (1H, d, J ) 2.5 Hz), 3.08-3.04 (1H, m),

2.75 (1H, bs), 2.29 (3H, s), 1.97-1.85 (4H, m). ¹³C NMR (75

MHz, CDCl₃) δ: 169.16, 167.23, 163.56, 160.46, 160.32, 157.24,

150.58, 139.94, 139.90, 134.85, 133.53, 133.50 (C), 127.32,

127.21, 126.78, 122.38, 118.34, 118.23, 115.95, 115.65, 115.35,

115.07 (CH), 72.95, 60.81, 60.33 (CH), 36.61, 25.07 (CH₂), 21.19

(CH₃). IR (cm^-1): 3443, 3019, 2936, 2862, 1747, 1605, 1509,

1427, 1388, 1370, 1221, 1198, 1157, 1016, 835, 757, 668.

MALDI-MS (C₂₆H₂₃F₂NO₄): [MH - H₂O]⁺434.1556 (calcd

434.1568); [MNa]⁺474.1485 (calcd 474.1493). This acetate

(1.864 g, 4.13 mmol) was dissolved in anhydrous DMF (25 mL),

imidazole (939 mg, 13.8 mmol) and TBDMSCl (1.853 g, 12.3

mmol) were added sequentially, and the solution was stirred

for 3 h followed by addition of 50% saturated aqueous NaHCO₃

(150 mL). After extraction with EtOAc (4 × 40 mL), the

combined organic layer was washed successively with satu-

rated aqueous NaHCO₃(40 mL) and H₂O (40 mL), evaporated

on Celite, and purified by dry column vacuum chromatography

Experimental Section

Chemistry. Reactions in anhydrous solvents were all

performed using oven-dried glassware under an atmosphere

of argon. Reagent-grade solvents were all purchased from

chemical companies and used without prior purification.

Anhydrous THF, ether, toluene, CH₃CN and CH₂Cl₂were

dried and purified through activated alumina columns as

described.⁶⁵Diisopropylamine, triethylamine, and pyridine

were distilled from KOH. For chromatographic purification,

technical-grade solvents were distillated prior to use. TLC was

performed using Machery-Nagel Alugram Sil G/UV₂₅₄TLC

plates and visualized with ultraviolet light at 254 nm followed

by ceric ammonium molybdate, phosphomolybdic acid, or

H₂SO₄/MeOH stains. Chromatographic purification of products

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6043

(4.2 × 5.5 cm) on silica gel eluting with a gradient of 0-30%

EtOAc in hexane (v/v) to give the intermediary silylated

acetate (2.137 g, 91%) as a colorless oil. R_f(1:1 EtOAc/hexane

1139, 1087, 1028, 835, 775, 750, 698. MALDI-MS (C₅₈H₆₅F₂-

NO₈Si): [MNa]⁺992.4350 (calcd 992.4345). This fully protected

glycoside (134 mg, 0.138 mmol) was dissolved in EtOH (10

mL), Pd(OH)₂/C (20% (w/w), 36 mg) was added, and the

suspension was evacuated four times with H₂and stirred

under an H₂atmosphere for 15 h. The suspension was

evaporated on Celite and purified by dry column vacuum

chromatography (3.5 × 2.0 cm) on silica gel eluting with a

gradient of 0-100% EtOAc in hexane followed by 10% MeOH

in CH₂Cl₂(v/v) to give the intermediary silylated glycoside (84

mg, 87%) as a light yellow oil. R_f(10% MeOH in CH₂Cl₂(v/v)):

1

(v/v)): 0.69. H NMR (300 MHz, CDCl₃) δ: 7.31 (2H, d, J )

8.7 Hz), 7.26-7.20 (4H, m), 7.10 (2H, d, J ) 8.7 Hz), 6.98 (2H,

t, J ) 8.7 Hz), 6.91 (2H, t, J ) 8.7 Hz), 4.67 (1H, t, J ) 5.3

Hz), 4.58 (1H, d, J ) 1.9 Hz), 3.06-3.02 (1H, m), 2.28 (3H, s),

1.96-1.80 (4H, m), 0.88 (9H, s), 0.02 (3H, s), -0.16 (3H, s).

¹³C NMR (75 MHz, CDCl₃) δ: 169.16, 167.06, 163.42, 160.47,

160.16, 157.23, 150.62, 140.50, 135.10, 133.74, 133.70 (C),

127.26, 127.14, 126.77, 122.37, 118.27, 118.16, 115.89, 115.58,

115.03, 114.76 (CH), 73.74, 60.67, 60.53 (CH), 37.94 (CH₂),

25.73 (CH₃), 24.55 (CH₂), 20.99 (CH₃), 18.07 (C), -4.74, -5.05

(CH₃). IR (cm^-1): 2953, 2930, 2857, 1752, 1606, 1510, 1472,

1426, 1385, 1370, 1252, 1219, 1197, 1166, 1140, 1102, 1086,

1015, 912, 835, 777, 736. MALDI-MS (C₃₂H₃₇F₂NO₄Si): [MH

- TBDMSOH]⁺434.1556 (calcd 434.1568); [MNa]⁺588.2347

(calcd 588.2358). Anal. Calcd for C₃₂H₃₇F₂NO₄Si: C, 67.94; H,

6.59; N, 2.48. Found: C, 67.94; H, 6.64; N, 2.37. This silylated

acetate (5.123 g, 9.06 mmol) was dissolved in CH₂Cl₂(200 mL),

neutral alumina (50 g) was added, and the suspension was

evaporated to dryness. The coated alumina was dried shortly

under vacuum and then heated to 70 °C for 5.5 h. After cooling,

the alumina was extracted with 10% MeOH in CH₂Cl₂(8 ×

50 mL), and the combined organic extracts were evaporated

on Celite and purified by dry column vacuum chromatography

(5.4 × 5.5 cm) on silica gel eluting with a gradient of 0-30%

EtOAc in hexane (v/v) to give phenol 5 (3.919 g, 83%) as a

1

0.39. H NMR (300 MHz, CDCl₃) δ: 7.26-7.14 (6H, m), 6.96

(2H, t, J ) 8.7 Hz), 6.91-6.83 (4H, m), 5.02 (1H, bs), 4.70-

4.64 (2H, m), 4.49 (1H, s), 4.24-4.10 (3H, m), 3.92 (1H, d, J )

5.6 Hz), 3.84-3.74 (2H, m), 3.61-3.50 (2H, m), 3.34 (3H, s),

2.97 (1H, dd, J ) 5.6, 6.8 Hz), 1.92-1.76 (4H, m), 0.86 (9H, s),

0.00 (3H, s), -0.17 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

167.18, 158.64, 157.00, 156.00, 140.52, 133.71, 129.87 (C),

127.17, 127.06, 126.96, 118.20, 118.11, 115.76, 115.45, 115.23,

115.15, 114.98, 114.69, 99.30, 74.49, 73.81, 71.91, 69.91 (CH),

67.15 CH₂), 60.89, 60.47 (CH), 55.23 (CH₃), 38.10 (CH₂), 25.86

(CH₃), 24.73 (CH₂), 18.23 (C), -4.55, -4.84 (CH₃). IR (cm^-1):

3390, 2931, 2858, 1748, 1609, 1510, 1472, 1428, 1387, 1362,

1250, 1223, 1155, 1143, 1104, 1079, 1059, 1042, 835, 776, 757.

MALDI-MS (C₃₇H₄₇F₂NO₈Si): [MNa]⁺722.2940 (calcd

722.2937). This silylated glycoside (84 mg, 0.12 mmol) was

dissolved in THF (2.5 mL), TBAF (0.5 mL, 1 M in THF) was

added, and the solution was stirred for 20 h. After dilution

with EtOAc (25 mL), the organic phase was washed with

saturated aqueous NaHCO₃(10 mL) and H₂O (10 mL),

evaporated on Celite, and purified by dry column vacuum

chromatography (4.0 × 2.0 cm) on silica gel eluting with a

gradient of 0-20% MeOH in CH₂Cl₂(v/v) to give glycoside 7

(60 mg, 86%) as a white solid after coevaporation with hexane

(10 mL). Glycoside 7 could be further purified by chromatotron

eluting with 10% MeOH in CH₂Cl₂(v/v). R_f(10% MeOH in

CH₂Cl₂(v/v)): 0.26. ¹H NMR (300 MHz, CD₃OD) δ: 7.32-7.22

(6H, m), 7.04-6.93 (6H, m), 4.75 (1H, d, J ) 1.9 Hz), 4.66 (1H,

d, J ) 3.7 Hz), 4.59 (1H, dd, J ) 4.4, 5.6 Hz), 4.25 (1H, dd, J

) 1.2, 10.6 Hz), 4.13 (1H, dd, J ) 5.6, 10.6 Hz), 3.84-3.78

(1H, m), 3.64 (1H, t, J ) 9.3 Hz), 3.45-3.34 (2H, m), 3.37 (3H,

s), 3.08-3.03 (1H, m), 1.96-1.78 (4H, m). ¹³C NMR (75 MHz,

CD₃OD) δ: 169.14, 164.49, 160.11, 153.40, 141.76, 141.71,

134.74, 130.60 (C), 128.35, 128.24, 128.11, 119.49, 119.38,

116.37, 116.06, 115.98, 115.63, 115.35, 100.88, 74.80, 73.35,

73.10, 71.29 (CH), 68.35 (CH₂), 61.72, 60.85 (CH), 55.29 (CH₃),

37.21, 25.84 (CH₂). MALDI-MS (C₃₁H₃₃F₂NO₈): [MNa]⁺

608.2074 (calcd 608.2072).

1

white foam. R_f(1:3 EtOAc/hexane (v/v)): 0.24. H NMR (300

MHz, CDCl₃) δ: 7.26-7.14 (6H, m), 6.99-6.83 (6H, m), 6.16

(1H, bs), 4.65 (1H, t, J ) 5.3 Hz), 4.52 (1H, d, J ) 1.9 Hz),

3.04-2.98 (1H, m), 1.92-1.76 (4H, m), 0.86 (9H, s), 0.00 (3H,

s), -0.17 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ: 167.82, 163.28,

160.42, 156.12, 140.50, 140.45, 133.57 (C), 128.92, 127.19,

127.15, 127.08, 118.43, 118.32, 116.05, 115.85, 115.55, 115.01,

114.72 (CH), 73.82, 61.17, 60.35 (CH), 38.07 (CH₂), 25.89 (CH₃),

24.68 (CH₂), 18.25 (C), -4.54, -4.84 (CH₃). IR (cm^-1): 3351,

2953, 2938, 2857, 1722, 1615, 1604, 1510, 1450, 1391, 1361,

1252, 1223, 1156, 1103, 1087, 863, 834, 776, 760. MALDI-MS

(C₃₀H₃₅F₂NO₃Si): [MH

-

TBDMSOH]⁺392.1451 (calcd

392.1462); [MH]⁺524.2409 (calcd 524.2433); [MNa]⁺546.2242

(calcd 546.2252). Anal. Calcd for C₃₀H₃₅F₂NO₃Si: C, 68.81; H,

6.74; N, 2.67. Found: C, 68.61; H, 6.82; N, 2.66.

Glycoside 7. Phenol 5 (257.5 mg, 0.492 mmol) and alcohol

6

^48,67(333 mg, 0.717 mmol) were dissolved in anhydrous THF

(15 mL) at 0 °C, Bu₃P (0.40 mL, 1.3 mmol) and 1,1′-(azodicar-

bonyl)dipiperidine (248 mg, 0.98 mmol) were added sequen-

tially, and the suspension was allowed to warm to ambient

temperature over several hours. After stirring at room tem-

perature for 14 h, EtOAc/hexane (1:4 (v/v), 30 mL) was added

and the suspension was filtered through Celite (2 × 15 mL

EtOAc/hexane (1:4 (v/v)) washings). The filtrate was evapo-

rated on Celite and purified by dry column vacuum chroma-

tography (3.8 × 3.3 cm) on silica gel eluting with a gradient

of 0-30% EtOAc in hexane (v/v) to give the intermediary fully

protected glycoside (98.5 mg, 21%) as a colorless oil. R_f(1:1

EtOAc/hexane (v/v)): 0.74. ¹H NMR (300 MHz, CDCl₃) δ:

7.40-7.16 (21H, m), 7.02-6.86 (6H, m), 5.03 (1H, d, J ) 10.6

Hz), 4.88 (1H, d, J ) 10.6 Hz), 4.86 (1H, d, J ) 11.2 Hz), 4.84

(1H, d, J ) 11.8 Hz), 4.70 (1H, d, J ) 12.5 Hz), 4.66 (1H, t, J

) 3.7 Hz), 4.68-4.65 (1H, m), 4.54 (1H, d, J ) 5.6 Hz), 4.52

(1H, d, J ) 3.1 Hz), 4.11 (2H, d, J ) 2.5 Hz), 4.06 (1H, t, J )

9.0 Hz), 3.92 (1H, dt, J ) 2.5, 10.0 Hz), 3.73 (1H, t, J ) 9.3

Hz), 3.62 (1H, dt, J ) 3.4, 9.7 Hz), 3.40 (3H, s), 3.07-3.00 (1H,

m), 1.96-1.78 (4H, m), 0.90 (9H, s), 0.03 (3H, s), -0.15 (3H,

s). ¹³C NMR (75 MHz, CDCl₃) δ: 167.20, 163.24, 160.24,

160.00, 158.53, 157.03, 140.50, 140.47, 138.42, 137.85, 137.80,

133.74, 129.70 (C), 128.26, 128.20, 127.97, 127.84, 127.62,

127.52, 127.16, 127.05, 126.91, 118.19, 118.09, 115.72, 115.43,

115.17, 114.97, 114.68, 98.14, 82.01, 79.77, 77.22 (CH), 75.76,

75.04 (CH₂), 73.79 (CH), 73.37 (CH₂), 69.06 (CH), 66.63 (CH₂),

60.97, 60.44 (CH), 55.25 (CH₃), 38.08 (CH₂), 25.87 (CH₃), 24.69

(CH₂), 18.22 (C), -4.56, -4.87 (CH₃). IR (cm^-1): 3031, 2929,

2857, 1749, 1608, 1510, 1454, 1386, 1361, 1250, 1220, 1156,

C-Glycoside 9. Phenol 5 (143 mg, 0.273 mmol) and alcohol

8

^49,68(180 mg, 0.325 mmol) were dissolved in anhydrous THF

(10 mL) at 0 °C, Bu₃P (0.20 mL, 0.80 mmol) and 1,1′-

(azodicarbonyl)dipiperidine (206 mg, 0.82 mmol) were added

sequentially, and the suspension was allowed to warm to

ambient temperature over several hours and stirred for 24 h.

EtOAc/hexane (1:4 (v/v), 20 mL) was added, the suspension

was filtered through Celite (2 × 10 mL EtOAc/hexane (1:4 (v/

v)) washings) and the filtrate was evaporated on Celite and

purified by dry column vacuum chromatography (4.1 × 3.3 cm)

on silica gel eluting with a gradient of 0-50% EtOAc in hexane

(v/v) to give the intermediary fully protected glycoside (60.1

mg, 21%) as a colorless oil. R_f(1:1 EtOAc/hexane (v/v)): 0.79.

¹H NMR (300 MHz, CDCl₃) δ: 7.37-7.17 (26H, m), 7.04-6.89

(6H, m), 4.96 (2H, bs), 4.89 (1H, d, J ) 9.3 Hz), 4.86 (1H, d, J

) 8.7 Hz), 4.69 (1H, t, J ) 5.3 Hz), 4.63-4.53 (5H, m), 4.21

(1H, d, J ) 10.6 Hz), 4.10 (1H, dd, J ) 5.0, 10.6 Hz), 3.85-

3.52 (7H, m), 3.07-3.02 (1H, m), 2.01-1.78 (4H, m), 0.91 (9H,

s), 0.05 (3H, s), -0.13 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

167.25, 158.74, 140.53, 140.49, 138.29, 137.85, 137.79, 137.65,

133.81 (C), 129.53, 128.32, 128.28, 128.18, 127.96, 127.81,

127.78, 127.74, 127.66, 127.54, 127.48, 127.18, 127.08, 126.90,

118.22, 118.12, 115.77, 115.47, 115.30, 114.98, 114.70 (CH),

87.12, 79.14, 78.25, 77.87, 77.71 (CH), 75.56, 75.11, 75.03

(CH₂), 73.82 (CH), 73.44, 68.93, 67.23 (CH₂), 61.02, 60.47 (CH),

38.10 (CH₂), 25.89 (CH₃), 24.71 (CH₂), 18.24 (C), -4.54, -4.83

6044 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

(CH₃). IR (cm^-1): 2951, 2929, 2858, 1749, 1608, 1510, 1454,

1386, 1361, 1250, 1223, 1156, 1141, 1101, 1028, 911, 835, 777,

735, 699. MALDI-MS (C₆₅H₇₁F₂NO₈Si): [MNa]⁺1082.4831

(calcd 1082.4815). This fully protected glycoside (72 mg, 0.068

mmol) was dissolved in EtOH (5 mL), Pd(OH)₂/C (20% (w/w),

40 mg) was added, and the suspension was evacuated four

times with H₂and stirred under an H₂atmosphere for 17 h.

The suspension was evaporated on Celite and purified by dry

column vacuum chromatography (3.8 × 2.0 cm) on silica gel

eluting with a gradient of 0-100% EtOAc in hexane followed

by 10% MeOH in CH₂Cl₂(v/v) to give the intermediary

silylated glycoside (28 mg, 59%) as a colorless oil. R_f(20%

chromatography (4.6 × 2.0 cm) on silica gel eluting with a

gradient of 0-20% MeOH in CH₂Cl₂(v/v) to give the inter-

mediary silylated glycoside (18.7 mg, 21% from 5) as a colorless

oil. R_f(20% MeOH in CH₂Cl₂(v/v)): 0.44. ¹H NMR (300 MHz,

CD₃OD) δ: 7.31-7.23 (6H, m), 7.04-6.94 (6H, m), 4.71 (1H,

d, J ) 1.9 Hz), 4.41 (1H, d, J ) 7.5 Hz), 4.12 (1H, dd, J ) 5.3,

10.9 Hz), 3.91-3.81 (3H, m), 3.66 (1H, d, J ) 5.6, 11.8 Hz),

3.57-3.47 (3H, m), 3.40-3.20 (7H, m), 3.07 (1H, t, J ) 5.9

Hz), 1.92-1.78 (4H, m), 0.87 (9H, s), 0.02 (3H, s), -0.18 (3H,

s). ¹³C NMR (75 MHz, CD₃OD) δ: 169.71, 160.66, 145.96,

142.43, 131.16, 131.05, 128.89, 128.80, 128.62, 120.00, 119.89,

116.83, 116.54, 116.41, 116.02, 115.74, 115.58, 104.65, 80.78,

80.43, 79.64, 78.16, 77.90, 75.13, 74.99, 71.43, 62.50, 62.08,

61.29, 38.96, 26.38, 25.75, 19.06, -4.40. MALDI-MS (C₄₃H₅₇F₂-

NO₁₃Si): [MNa]⁺884.3668 (calcd 884.3465). This silylated

glycoside (18.3 mg, 0.021 mmol) was dissolved in anhydrous

THF (2.5 mL, Teflon bottle) at 0 °C, anhydrous pyridine (0.50

mL) followed by HF‚pyridine complex (0.50 mL) was added,

and the solution was stirred for 17 h. NaHCO₃(s) was added

and the suspension was evaporated on Celite and purified by

dry column vacuum chromatography (4.6 × 2.0 cm) on silica

gel eluting with a gradient of 0-20% MeOH in CH₂Cl₂(v/v)

to give C-glycoside 11 (10.3 mg, 65%) as a white solid. R_f(20%

1

MeOH in CH₂Cl₂(v/v)): 0.64. H NMR (300 MHz, CDCl₃) δ:

7.24-7.11 (6H, m), 6.95 (2H, t, J ) 8.7 Hz), 6.85-6.79 (4H,

m), 5.22 (1H, bs), 4.93 (1H, bs), 4.65 (1H, t, J ) 5.3 Hz), 4.45

(1H, bs), 4.36 (1H, bs), 4.14 (1H, d, J ) 10.0 Hz), 3.97 (1H, d,

J ) 8.1 Hz), 3.69 (2H, bs), 3.51 (3H, bs), 3.22 (1H, d, J ) 6.2

Hz), 2.94 (1H, dd, J ) 5.6, 6.8 Hz), 2.76 (1H, bs), 1.88-1.68

(4H, m), 0.86 (9H, s), 0.00 (3H, s), -0.17 (3H, s). ¹³C NMR (75

MHz, CDCl₃) δ: 167.22, 163.28, 160.05, 158.36, 157.03, 140.57,

133.75, 130.30, 129.52, 127.22, 127.11, 118.23, 115.83, 115.54,

116.35, 115.05, 114.91, 114.76, 79.16, 78.33, 77.70, 73.88,

70.18, 69.52, 67.75, 61.54, 60.79, 60.57, 38.14, 25.91, 24.81,

18.27, -4.51, -4.80. IR (cm^-1): 3391, 2930, 2858, 1747, 1609,

1510, 1387, 1362, 1223, 1140, 1086, 1043, 1014, 835, 758.

MALDI-MS (C₃₇H₄₇F₂NO₈Si): [MH - TBDMSOH]⁺568.2132

(calcd 568.2147); [MNa]⁺722.2939 (calcd 722.2937). This

silylated glycoside (27.0 mg, 0.039 mmol) was dissolved in THF

(1.0 mL), TBAF (0.2 mL, 1 M in THF) was added, and the

solution was stirred for 15 h, diluted with CH₂Cl₂, evaporated

on Celite, and purified by dry column vacuum chromatography

(3.5 × 2.0 cm) on silica gel eluting with a gradient of 0-18%

MeOH in CH₂Cl₂(v/v) to give C-glycoside 9 (14.0 mg, 62%) as

a white solid after coevaporation with hexane (10 mL). R_f(20%

1

MeOH in CH₂Cl₂(v/v)): 0.31. H NMR (300 MHz, CD₃OD) δ:

7.33-7.24 (6H, m), 7.05-6.94 (6H, m), 4.78 (1H, d, J ) 1.9

Hz), 4.60 (1H, t, J ) 4.4 Hz), 4.41 (1H, d, J ) 7.5 Hz), 4.30

(1H, d, J ) 10.0 Hz), 4.12 (1H, dd, J ) 5.0, 10.6 Hz), 3.91-

3.84 (3H, m), 3.66 (1H, d, J ) 5.6, 11.8 Hz), 3.57-3.49 (3H,

m), 3.40-3.20 (6H, m), 3.10-3.06 (1H, m), 1.97-1.82 (4H, m).

¹³C NMR (75 MHz, CD₃OD) δ: 169.52, 164.87, 160.42, 142.07,

133.18 (C), 131.03, 130.87, 128.68, 128.59, 128.47, 123.36,

119.86, 119.74, 116.69, 116.38, 116.25, 116.22, 115.96, 115.88,

115.68, 104.54, 80.71, 80.36, 79.58, 78.11, 77.85, 74.94, 73.70,

71.38 (CH), 69.02, 62.47 (CH₂), 62.09 (CH₂+CH), 61.20 (CH),

37.54, 26.18 (CH₂). MALDI-MS (C₃₇H₄₃F₂NO₁₃): [MNa]⁺

770.2589 (calcd 770.2600).

1

MeOH in CH₂Cl₂(v/v)): 0.56. H NMR (300 MHz, CD₃OD) δ:

7.33-7.23 (6H, m), 7.05-6.94 (6H, m), 4.78 (1H, d, J ) 1.9

Hz), 4.59 (1H, t, J ) 5.3 Hz), 4.29 (1H, dd, J ) 1.5, 10.3 Hz),

4.13 (1H, dd, J ) 5.6, 10.6 Hz), 3.85 (1H, d, J ) 11.2 Hz), 3.67-

3.61 (1H, m), 3.57-3.51 (1H, m), 3.44-3.37 (2H, m), 3.31-

3.28 (2H, m), 3.11-3.06 (1H, m), 1.97-1.81 (4H, m). ¹³C NMR

(75 MHz, CD₃OD) δ: 169.20, 160.12, 130.69, 128.36, 128.25,

128.14, 119.52, 119.41, 116.35, 116.04, 115.93, 115.63, 115.35,

81.55, 79.49, 79.39, 73.35, 71.30, 71.23, 68.77, 62.66, 61.74,

60.86, 37.22, 25.84. MALDI-MS (C₃₁H₃₃F₂NO₈): [MH - TB-

DMSOH]⁺568.2143 (calcd 568.2147); [MNa]⁺608.2073 (calcd

608.2072).

Thioacetate 12.^55,68Alcohol 6^48,67(1.181 g, 2.54 mmol) was

dissolved in anhydrous CH₂Cl₂(25 mL) at 0 °C, anhydrous

pyridine (3.0 mL) followed by MsCl (0.50 mL, 6.4 mmol) was

added, and the solution was stirred at 0 °C for 1 h and at room

temperature for 7 h followed by addition of saturated aqueous

NaHCO₃(50 mL). The layers were separated, and the aqueous

layer was extracted with EtOAc (3 × 25 mL). The combined

organic layer was washed successively with saturated aqueous

NaHCO₃(25 mL) and H₂O (25 mL), evaporated on Celite, and

purified by dry column vacuum chromatography (4.1 × 3.3 cm)

on silica gel eluting with a gradient of 0-100% CH₂Cl₂in

hexane (v/v) followed by 0.25-1.0% MeOH in CH₂Cl₂(v/v) to

give the intermediary mesylate (1.303 g, 94%) as a colorless

oil after coevaporation with acetonitrile (3 × 10 mL). R_f(1%

C-Glycoside 11. Phenol 5 (80.3 mg, 0.153 mmol) and

alcohol 10⁵⁰(101.5 mg, 0.103 mmol) were dissolved in anhy-

drous THF (10 mL) at 0 °C, Bu₃P (50 mg, 0.20 mmol) and 1,1′-

(azodicarbonyl)dipiperidine (39.5 mg, 0.17 mmol) were added

sequentially, and the suspension was allowed to warm to

ambient temperature over several hours. After stirring at room

temperature for 26 h, EtOAc/hexane (1:4 (v/v), 30 mL) was

added, and the suspension was filtered through Celite (2 ×

10 mL EtOAc/hexane (1:4 (v/v)) washings). The filtrate was

evaporated on Celite and purified by dry column vacuum

chromatography (4.5 × 2.0 cm) on silica gel eluting with a

gradient of 0-25% EtOAc in hexane (v/v) to give a 1:1 mixture

of the intermediary fully protected glycoside and unreacted

phenol 5 (49.7 mg) as a white foam. R_f(1:1 EtOAc/hexane (v/

v)): 0.64; ¹³C NMR (75 MHz, CDCl₃) δ: 167.39, 163.27, 160.31,

158.82, 157.09, 140.54, 140.49, 139.05, 138.37, 138.29, 138.19,

137.85, 133.78, 133.73, 129.40, 128.96, 128.23, 128.12, 128.04,

127.94, 127.86, 127.73, 127.63, 127.57, 127.49, 127.41, 127.20,

127.10, 126.87, 118.30, 118.19, 116.01, 115.78, 115.49, 115.30,

114.99, 114.71, 102.41, 85.35, 84.84, 82.70, 79.29, 78.01, 77.82,

77.19, 75.64, 75.25, 75.10, 75.02, 74.96, 74.81, 73.84, 73.26,

68.99, 68.15, 67.49, 61.07, 60.44, 38.09, 25.90, 24.72, 18.25,

-4.53, -4.83. MALDI-MS (C₉₂H₉₉F₂NO₁₃Si): [MNa]⁺1514.6763

(calcd 1514.6751). This mixture (49.7 mg) was dissolved in

EtOH/EtOAc (10 mL, 1:1 (v/v)), Pd(OH)₂/C (20% (w/w), 31 mg)

was added, and the suspension was evacuated four times with

H₂and stirred under an H₂atmosphere for 3 h. The suspension

was evaporated on Celite and purified by dry column vacuum

1

MeOH in CH₂Cl₂(v/v)): 0.60. H NMR (300 MHz, CDCl₃) δ:

7.39-7.26 (15H, m), 5.02 (1H, d, J ) 10.6 Hz), 4.92 (1H, d, J

) 10.6 Hz), 4.84 (1H, d, J ) 10.6 Hz), 4.80 (1H, d, J ) 12.5

Hz), 4.66 (1H, d, J ) 11.8 Hz), 4.63 (1H, d, J ) 10.6 Hz), 4.60

(1H, d, J ) 3.7 Hz), 4.41-4.32 (2H, m), 4.02 (1H, t, J ) 9.3

Hz), 3.85 (1H, dt, J ) 3.7, 10.0 Hz), 3.52 (1H, dt, J ) 3.7, 6.2

Hz), 3.50 (1H, bs), 3.39 (3H, s), 2.98 (3H, s). ¹³C NMR (75 MHz,

CDCl₃) δ: 138.30, 137.75, 137.56 (C), 128.36, 128.30, 127.94,

127.84, 127.76, 127.57 (CH), 98.06, 81.73, 79.69, 76.86 (CH),

75.73, 75.09, 73.44 (CH₂), 68.59 (CH), 68.36 (CH₂), 55.46, 37.54

(CH₃). IR (cm^-1): 3031, 2913, 1497, 1454, 1359, 1177, 1089,

1074, 1046, 1003, 965, 931, 813, 739, 699. MALDI-MS

(C₂₉H₃₄O₈S): [MNa]⁺565.1873 (calcd 565.1872). Anal. Calcd

for C₂₉H₃₄O₈S: C, 64.19; H, 6.32. Found: C, 63.99; H, 6.27.

This mesylate (1.290 g, 2.38 mmol) was dissolved in EtOH (25

mL), KOSCMe (869 mg, 7.61 mmol) was added, and the

unclear solution was stirred at reflux for 4 h (orange precipi-

tate). After cooling, 50% saturated aqueous NaHCO₃(100 mL)

was added and the suspension was extracted with EtOAc (3

× 50 mL). The combined organic layer was washed succes-

sively with saturated aqueous NaHCO₃(50 mL) and H₂O (50

mL), evaporated on Celite, and purified by dry column vacuum

chromatography (4.1 × 3.3 cm) on silica gel eluting with a

gradient of 0-30% EtOAc in hexane (v/v) to give thioacetate

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6045

12 (1.189 g, 96%) as a light orange solid. R_f(1:1 EtOAc/hexane

m), 7.00 (2H, t, J ) 8.7 Hz), 6.95 (2H, t, J ) 8.7 Hz), 5.05 (1H,

d, J ) 11.2 Hz), 4.97 (1H, d, J ) 11.2 Hz), 4.84 (1H, d, J )

11.8 Hz), 4.82 (1H, d, J ) 10.6 Hz), 4.69 (1H, t, J ) 6.8 Hz),

4.67 (1H, d, J ) 12.5 Hz), 4.60 (1H, d, J ) 3.7 Hz), 4.56 (1H,

d, J ) 12.5 Hz), 4.54 (1H, d, J ) 10.6 Hz), 4.29 (1H, t, J ) 9.5

Hz), 4.06 (1H, t, J ) 9.0 Hz), 3.57 (1H, t, J ) 3.1 Hz), 3.53

(1H, d, J ) 3.1 Hz), 3.46 (3H, s), 3.26 (1H, t, J ) 9.3 Hz), 3.14

(1H, dd, J ) 10.0, 14.3 Hz), 2.96 (1H, dt, J ) 1.9, 6.8 Hz),

1.97-1.78 (4H, m), 0.90 (9H, s), 0.04 (3H, s), -0.13 (3H, s).

¹³C NMR (75 MHz, CDCl₃) δ: 166.62, 163.27, 160.37, 160.03,

157.14, 148.91, 140.33, 138.05, 137.63, 137.29, 136.67, 133.45,

133.42 (C), 128.44, 128.31, 128.18, 128.04, 127.96, 127.86,

127.65, 127.15, 127.03, 126.97, 123.15, 118.13, 118.03, 115.93,

115.64, 115.02, 114.75 (CH), 97.92, 81.67, 79.60, 79.23 (CH),

75.78, 74.86 (CH₂), 73.78 (CH), 73.37 (CH₂), 65.64, 60.66, 60.48

(CH), 55.73 (CH₃), 51.63, 38.06 (CH₂), 25.85 (CH₃), 24.69 (CH₂),

18.22 (C), -4.54, -4.87 (CH₃). IR (cm^-1): 3032, 2930, 2858,

1750, 1605, 1510, 1455, 1386, 1252, 1220, 1153, 1086, 1073,

1048, 870, 836, 755, 699. MALDI-MS (C₅₈H₆₅F₂NO₁₀SiS):

[MNa]⁺1056.3969 (calcd 1056.3964). Anal. Calcd for C₅₈H₆₅F₂-

NO₁₀SiS: C, 67.35; H, 6.33; N, 1.35. Found: C, 67.43; H, 6.44;

N, 1.33.

â-Lactam 14. Sulfonate 42 (105.1 mg, 0.102 mmol) was

dissolved in EtOH (5 mL), Pd(OH)₂/C (20% (w/w), 33 mg) was

added, and the suspension was evacuated four times with H₂

and stirred under an H₂atmosphere for 6 h. The suspension

was evaporated on Celite and purified by dry column vacuum

chromatography (4.2 × 2.0 cm) on silica gel eluting with a

gradient of 0-10% MeOH in CH₂Cl₂(v/v) to give the inter-

mediary silylated â-lactam (63.2 mg, 81%) as a colorless oil.

R_f(10% MeOH in CH₂Cl₂(v/v)): 0.36. ¹H NMR (300 MHz,

acetone-d₆) δ: 7.55 (2H, d, J ) 8.7 Hz), 7.42 (2H, d, J ) 8.7

Hz), 7.37 (2H, dd, J ) 5.9, 8.4 Hz), 7.28 (2H, dd, J ) 5.0, 9.3

Hz), 7.11-7.01 (4H, m), 4.96 (1H, d, J ) 1.9 Hz), 4.84 (1H, t,

J ) 5.3 Hz), 4.69 (1H, d, J ) 3.7 Hz), 4.61 (1H, d, J ) 5.0 Hz),

4.35 (1H, d, J ) 3.1 Hz), 4.16 (1H, dt, J ) 1.2, 10.0 Hz), 3.87

(1H, dd, J ) 1.2, 14.9 Hz), 3.79 (1H, d, J ) 7.5 Hz), 3.65 (1H,

t, J ) 9.0 Hz), 3.56 (1H, dd, J ) 10.0, 14.9 Hz), 3.45-3.40

(1H, m), 3.38 (3H, s), 3.27-3.14 (2H, m), 2.00-1.88 (4H, m),

0.87 (9H, s), 0.05 (3H, s), -0.15 (3H, s). ¹³C NMR (75 MHz,

acetone-d₆) δ: 167.25, 163.96, 160.84, 160.75, 157.65, 150.14,

141.91, 141.87, 138.13, 134.95, 134.91 (C), 128.32, 128.23,

123.84, 118.98, 118.88, 116.43, 116.12, 115.49, 115.21 (CH),

100.74, 74.77, 74.42, 73.55, 73.04, 68.01, 61.25, 60.50 (CH),

55.56 (CH₃), 52.83, 38.50 (CH₂), 26.16 (CH₃), 25.34 (CH₂), 18.65

(C), -4.47, -4.71 (CH₃). IR (cm^-1): 3396, 2951, 2931, 2857,

1754, 1701, 1605, 1510, 1426, 1385, 1250, 1220, 1151, 1103,

1088, 1053, 1015, 988, 872, 836, 778. MALDI-MS (C₃₇H₄₇F₂-

NO₁₀SSi): [MNa]⁺786.2559 (calcd 786.2556). This silylated

â-lactam (58.9 mg, 0.077 mmol) was dissolved in anhydrous

THF (2.5 mL, Teflon bottle), anhydrous pyridine (0.5 mL)

followed by HF‚pyridine complex (0.5 mL) was added, and the

solution was stirred for 14.5 h, diluted with ether (20 mL),

and washed with saturated aqueous NaHCO₃(3 × 5 mL). The

organic layer was evaporated on Celite and purified by dry

column vacuum chromatography (4.2 × 2.0 cm) on silica gel

eluting with a gradient of 0-10% MeOH in CH₂Cl₂(v/v) to

give â-lactam 14 (44.9 mg, 90%) as a white solid. R_f(10%

MeOH in CH₂Cl₂(v/v)): 0.26. ¹H NMR (300 MHz, acetone-d₆)

δ: 7.56 (2H, d, J ) 8.7 Hz), 7.43 (2H, d, J ) 8.7 Hz), 7.37 (2H,

dd, J ) 5.6, 8.7 Hz), 7.30 (2H, dd, J ) 4.7, 9.0 Hz), 7.06 (2H,

d, J ) 9.3 Hz), 7.03 (2H, d, J ) 8.7 Hz), 4.99 (1H, d, J ) 2.5

Hz), 4.69 (1H, d, J ) 3.7 Hz), 4.61 (1H, d, J ) 5.0 Hz), 4.42

(1H, d, J ) 3.7 Hz), 4.34 (1H, bs), 4.15 (1H, dt, J ) 1.2, 8.7

Hz), 3.86 (1H, dd, J ) 1.2, 14.9 Hz), 3.79 (1H, d, J ) 8.1 Hz),

3.65 (1H, t, J ) 8.7 Hz), 3.57 (1H, dd, J ) 10.0, 14.9 Hz), 3.44-

3.38 (1H, m), 3.38 (3H, s), 3.32-3.14 (2H, m), 2.08-1.86 (4H,

m). ¹³C NMR (75 MHz, acetone-d₆) δ: 167.42, 163.87, 160.85,

157.67, 150.13, 142.52, 138.18, 134.93 (C), 128.35, 128.22,

128.13, 123.83, 119.01, 118.89, 116.44, 116.13, 115.40, 115.11

(CH), 100.74, 74.77, 73.56, 73.04, 72.77, 68.01, 61.27, 60.56

(CH), 55.56 (CH₃), 52.83, 37.54, 25.70 (CH₂). IR (cm^-1): 3395,

2925, 1732, 1604, 1509, 1365, 1219, 1148, 1103, 1051, 1014,

871, 834, 752. MALDI-MS (C₃₁H₃₃F₂NO₁₀S): [MNa]⁺672.1693

1

(v/v)): 0.64. H NMR (300 MHz, CDCl₃) δ: 7.41-7.32 (15H,

m), 5.03 (1H, d, J ) 10.6 Hz), 4.94 (1H, d, J ) 10.6 Hz), 4.86

(1H, d, J ) 10.6 Hz), 4.82 (1H, d, J ) 11.8 Hz), 4.69 (1H, d, J

) 11.8 Hz), 4.66 (1H, d, J ) 10.6 Hz), 4.58 (1H, d, J ) 3.1 Hz),

4.02 (1H, t, J ) 9.0 Hz), 3.81 (1H, dt, J ) 2.5, 7.5 Hz), 3.55

(1H, dd, J ) 3.7, 9.3 Hz), 3.48 (1H, dd, J ) 3.1, 13.7 Hz), 3.40

(3H, s), 3.35 (1H, t, J ) 9.5 Hz), 3.08 (1H, dd, J ) 7.5, 13.7

Hz), 2.36 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ: 194.67, 138.46,

137.90, 137.78 (C), 128.33, 128.29, 128.03, 127.94, 127.85,

127.81, 127.74, 127.53 (CH), 97.72, 81.69, 80.36, 79.78 (CH),

75.64, 75.04, 73.22 (CH₂), 69.23 (CH), 55.02 (CH₃), 30.73 (CH₂),

30.39 (CH₃). IR (cm^-1): 3063, 3031, 2908, 1694, 1497, 1454,

1358, 1201, 1156, 1136, 1092, 1072, 1050, 1029, 999, 955, 737,

698, 630. MALDI-MS (C₃₀H₃₄O₆S): [MNa]⁺545.1974 (calcd

545.1974). Anal. Calcd for C₃₀H₃₄O₆S: C, 68.94; H, 6.56.

Found: C, 68.77; H, 6.63.

Sulfonyl Chloride 13. Thioacetate 12^55,68(1.180 g, 2.26

mmol) was dissolved in AcOH (25 mL) to which KOAc (4.082

g, 41.6 mmol) followed by Oxone (2KHSO₅‚KHSO₄‚K₂SO₄,

4.019 g, 8.69 mmol) was added, and after stirring for 15 h,

saturated aqueous NaHCO₃(100 mL), H₂O (50 mL), and

saturated aqueous Na₂CO₃(50 mL) were carefully added. After

extraction with EtOAc (4 × 40 mL), the combined organic layer

was washed with saturated aqueous Na₂CO₃(50 mL), evapo-

rated on Celite, and purified by dry column vacuum chroma-

tography (4.0 × 3.3 cm) on silica gel eluting with a gradient

of 0-90% EtOAc in hexane (v/v) followed by 0-50% MeOH in

EtOAc (v/v) to give the intermediary sulfonate salt⁵⁷(1.116 g,

90%) as a white solid. R_f(1:3 MeOH/EtOAc (v/v)): 0.40. ¹H

NMR (300 MHz, CD₃OD) δ: 7.37-7.21 (15H, m), 4.90 (1H, d,

J ) 11.2 Hz), 4.86 (1H, d, J ) 10.6 Hz), 4.84 (1H, d, J ) 11.2

Hz), 4.73 (1H, d, J ) 3.1 Hz), 4.72 (1H, d, J ) 11.2 Hz), 4.64

(1H, d, J ) 12.5 Hz), 4.60 (1H, d, J ) 11.2 Hz), 4.16 (1H, t, J

) 9.2 Hz), 3.90 (1H, t, J ) 9.3 Hz), 3.55 (1H, dd, J ) 3.4, 9.3

Hz), 3.48 (3H, s), 3.30-3.22 (2H, m), 2.92 (1H, dd, J ) 10.0,

14.3 Hz). ¹³C NMR (75 MHz, CD₃OD) δ: 140.03, 139.57, 139.55

(C), 129.42, 129.31, 129.15, 128.93, 128.89, 128.84, 128.67,

128.59 (CH), 98.53, 83.03, 81.65, 81.52 (CH), 76.44, 75.83,

73.85 (CH₂), 68.52 (CH), 55.95 (CH₃), 53.65 (CH₂). IR (cm^-1):

3484, 3030, 2922, 1497, 1454, 1360, 1230, 1198, 1177, 1156,

1093, 1058, 1028, 736, 696. MALDI-MS (C₂₈H₃₁NaO₈S): [MNa]⁺

573.1536 (calcd 573.1535). This sulfonate salt (696 mg, 1.26

mmol) was suspended in anhydrous acetonitrile/CH₂Cl₂(10

mL, 1:1 (v/v)) at 0 °C, Ph₃P (1.002 g, 3.8 mmol) and thionyl

chloride (0.40 mL, 5.5 mmol) were added sequentially, and the

suspension was stirred at room temperature for 13 h. EtOAc/

hexane (1:4 (v/v), 100 mL) was added, the suspension was

filtered through Celite (4 × 15 mL EtOAc/hexane (1:3 (v/v))

washings), and the filtrate was evaporated and dried shortly

under vacuum to give sulfonyl chloride 13 (657 mg, 95%) as a

yellowish oil. R_f(1:1 EtOAc/hexane (v/v)): 0.65. ¹H NMR (300

MHz, CDCl₃) δ: 7.42-7.28 (15H, m), 5.05 (1H, d, J ) 10.6

Hz), 4.96 (1H, d, J ) 11.8 Hz), 4.85 (1H, d, J ) 10.6 Hz), 4.83

(1H, d, J ) 11.8 Hz), 4.67 (1H, d, J ) 12.5 Hz), 4.60 (1H, d, J

) 11.2 Hz), 4.60 (1H, d, J ) 3.1 Hz), 4.33 (1H, t, J ) 9.6 Hz),

4.07 (1H, t, J ) 9.0 Hz), 3.85 (1H, dd, J ) 1.2, 13.7 Hz), 3.55

(1H, d, J ) 9.3 Hz), 3.52 (1H, t, J ) 10.0 Hz), 3.46 (3H, s),

3.26 (1H, t, J ) 9.5 Hz). ¹³C NMR (75 MHz, CDCl₃) δ: 138.02,

137.57, 137.06 (C), 128.58, 128.36, 128.30, 128.23, 128.12,

127.92, 127.66 (CH), 98.00, 81.56, 79.41, 78.49 (CH), 75.85,

74.76, 73.38, 66.75 (CH₂), 65.93 (CH), 55.90 (CH₃). MALDI-

MS (C₂₈H₃₁ClO₇S): [MNa]⁺569.1378 (calcd 569.1377).

Sulfonate 42. Sulfonyl chloride 13 (197 mg, 0.36 mmol) was

suspended in anhydrous CH₂Cl₂(5 mL), anhydrous pyridine

(0.5 mL) followed by phenol 5 (70.0 mg, 0.13 mmol) was added,

and the solution was stirred for 22 h, diluted with EtOAc (25

mL), and washed sequentially with saturated aqueous NaH-

CO₃(10 mL) and H₂O (10 mL). The organic layer was

evaporated on Celite and purified by dry column vacuum

chromatography (4.3 × 2.0 cm) on silica gel eluting with a

gradient of 0-35% EtOAc in hexane (v/v) to give sulfonate 42

(125.5 mg, 91%) as a colorless oil/glass. R_f(1% MeOH in CH₂-

Cl₂(v/v)): 0.77. ¹H NMR (300 MHz, CDCl₃) δ: 7.37-7.14 (23H,

6046 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

(calcd 672.1691). Anal. Calcd for C₃₁H₃₃F₂NO₁₀S: C, 57.31; H,

5.12; N, 2.16. Found: C, 57.34; H, 5.26; N, 2.21.

sequentially at 0 °C, and the suspension was stirred at room

temperature for 2.5 h. EtOAc/hexane (1:4 (v/v), 100 mL) was

added, the suspension was filtered through Celite (2 × 12.5

mL EtOAc/hexane (1:4 (v/v)) washings), and the filtrate was

evaporated and dried shortly under vacuum to give sulfonyl

chloride 15 (871 mg, quant.) as a light yellow oil. R_f(10%

Sulfonyl Chloride 15. Alcohol 8⁴⁹(1.069 g, 1.93 mmol) was

dissolved in anhydrous CH₂Cl₂(25 mL), anhydrous pyridine

(3.0 mL) followed by MsCl (0.50 mL, 6.4 mmol) was added,

and after stirring for 2.5 h, saturated aqueous NaHCO₃(50

mL) was added. The layers were separated, and the aqueous

layer was extracted with EtOAc (3 × 25 mL). The combined

organic layer was washed successively with saturated aqueous

NaHCO₃(25 mL) and H₂O (25 mL), evaporated on Celite, and

purified by dry column vacuum chromatography (4.2 × 3.3 cm)

on silica gel eluting with a gradient of 0-50% EtOAc in hexane

(v/v) to give the intermediary mesylate⁵⁵(1.202 g, 99%) as a

colorless oil after coevaporation with acetonitrile (3 × 15 mL).

R_f(1:1 EtOAc/hexane (v/v)): 0.52. ¹H NMR (300 MHz, CDCl₃)

δ: 7.39-7.18 (20H, m), 4.94-4.83 (4H, m), 4.66 (1H, d, J )

10.9 Hz), 4.60 (1H, d, J ) 10.9 Hz), 4.56-4.50 (3H, m), 4.37

1

MeOH in CH₂Cl₂(v/v)): 0.84. H NMR (300 MHz, CDCl₃) δ:

7.48-7.21 (20H, m), 5.02-4.85 (4H, m), 4.68-4.55 (4H, m),

3.98-3.35 (9H, m). ¹³C NMR (75 MHz, CDCl₃) δ: 137.86,

137.72, 137.68, 136.96 (C), 128.64, 128.34, 128.32, 128.27,

128.24, 128.20, 127.99, 127.79, 127.67, 127.61, 127.49 (CH),

86.83, 79.20, 78.46, 77.59 (CH), 75.68, 74.96, 74.82 (CH₂), 74.20

(CH), 73.46, 68.04, 66.69 (CH₂). MALDI-MS (C₃₅H₃₇ClO₇S):

[MNa]⁺659.1849 (calcd 659.1846).

â-Lactam 40. Sulfonyl chloride 15 (871 mg, 1.26 mmol) was

suspended in anhydrous CH₂Cl₂(10 mL), anhydrous pyridine

(1.0 mL) followed by phenol 5 (334 mg, 0.634 mmol) was added,

and the solution was stirred for 13 h, diluted with EtOAc (50

mL), and washed sequentially with saturated aqueous NaH-

CO₃(20 mL) and H₂O (20 mL). The organic layer was

evaporated on Celite and purified by dry column vacuum

chromatography (4.3 × 3.3 cm) on silica gel eluting with a

gradient of 0-100% CH₂Cl₂in hexane (v/v) to give the

intermediary fully protected sulfonate (657 mg, 92%) as a

(1H, dd, J ) 3.7, 11.5 Hz), 3.77-3.45 (7H, m), 2.98 (3H, s). ¹³

C

NMR (75 MHz, CDCl₃) δ: 138.10, 137.70, 137.59, 137.34 (C),

128.44, 128.36, 128.33, 128.29, 127.95, 127.79, 127.75, 127.67,

127.62, 127.51 (CH), 86.76, 78.62, 77.86, 77.32, 76.86 (CH),

75.60, 75.21, 75.10, 73.36, 69.20, 68.68 (CH₂), 37.92 (CH₃). IR

(cm^-1): 3031, 2866, 1497, 1454, 1356, 1219, 1175, 1096, 964,

914, 772, 748, 698. MALDI-MS (C₃₆H₄₀O₈S): [MNa]⁺655.2344

(calcd 655.2342). Anal. Calcd for C₃₆H₄₀O₈S: C, 68.33; H, 6.37.

Found: C, 68.33; H, 6.46. This mesylate (1.190 g, 1.88 mmol)

was dissolved in EtOH (25 mL), KOSCMe (888 mg, 7.78 mmol)

was added, and the unclear solution was stirred at reflux for

16 h (orange precipitate). After cooling, 50% saturated aqueous

NaHCO₃(100 mL) was added and the suspension was ex-

tracted with EtOAc (3 × 50 mL). The combined organic layer

was washed successively with saturated aqueous NaHCO₃(50

mL) and H₂O (50 mL), evaporated on Celite, and purified by

dry column vacuum chromatography (4.0 × 3.3 cm) on silica

gel eluting with a gradient of 0-40% EtOAc in hexane (v/v)

to give the intermediary thioacetate⁶⁹(1.064 g, 92%) as a light

orange solid. R_f(1:1 EtOAc/hexane (v/v)): 0.69. ¹H NMR (300

MHz, CDCl₃) δ: 7.43-7.18 (20H, m), 4.93 (2H, s), 4.91 (1H, d,

J ) 11.8 Hz), 4.85 (1H, d, J ) 10.6 Hz), 4.68 (1H, d, J ) 10.6

Hz), 4.66 (1H, d, J ) 11.8 Hz), 4.62 (1H, d, J ) 10.0 Hz), 4.58

(1H, d, J ) 11.8 Hz), 3.78-3.40 (8H, m), 3.10 (1H, dd, J ) 6.2,

13.7 Hz), 2.36 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ: 194.82,

138.28, 138.10, 137.85, 137.60 (C), 128.35, 128.30, 128.26,

128.19, 128.13, 127.77, 127.72, 127.68, 127.62, 127.58, 127.53,

127.41 (CH), 86.85, 80.56, 79.12, 78.22, 77.86 (CH), 75.52,

75.18, 74.99, 73.42, 68.66, 31.10 (CH₂), 30.57 (CH₃). IR (cm^-1):

3064, 3031, 2902, 2865, 1693, 1497, 1454, 1360, 1210, 1134,

1099, 1069, 1028, 737, 698, 629. MALDI-MS (C₃₇H₄₀O₆S):

[MH]⁺613.2617 (calcd 613.2524); [MNa]⁺635.2445 (calcd

635.2443). This thioacetate (2.190 g, 3.57 mmol) was sus-

pended in AcOH (25 mL), KOAc (4.216 g, 43 mmol) followed

by Oxone (2KHSO₅‚KHSO₄‚K₂SO₄, 8.076 g, 17.5 mmol) was

added, and after stirring for 11 h, saturated aqueous Na₂CO₃

(100 mL) and H₂O (100 mL) were carefully added. After

extraction with EtOAc (5 × 100 mL), the combined organic

layer was washed with saturated aqueous Na₂CO₃(50 mL),

evaporated on Celite, and purified by dry column vacuum

chromatography (4.3 × 3.3 cm) on silica gel eluting with a

gradient of 0-90% EtOAc in hexane (v/v) followed by 0-50%

MeOH in EtOAc (v/v) to give the intermediary sulfonate salt

(467 mg, 20%) as a white solid. Further extractions of the

aqueous layer with CH₂Cl₂(100 + 3 × 50 mL), evaporation

on Celite and purification by dry column vacuum chromatog-

raphy (4.3 × 3.3 cm) on silica gel eluting with a gradient of

0-100% MeOH in EtOAc (v/v) followed by 20% MeOH in CH₂-

Cl₂(v/v) gave additional sulfonate salt (810 mg, 35%) as a

1

white foam. R_f(1:1 EtOAc/hexane (v/v)): 0.76. H NMR (300

MHz, CDCl₃) δ: 7.37-7.15 (28H, m), 7.01 (2H, t, J ) 8.7 Hz),

6.96 (2H, t, J ) 8.7 Hz), 5.03-4.81 (4H, m), 4.73-4.51 (6H,

m), 3.95 (1H, t, J ) 8.4 Hz), 3.78 (4H, bs), 3.57-3.53 (1H, m),

3.48 (1H, d, J ) 1.2 Hz), 3.40 (1H, t, J ) 9.0 Hz), 3.24 (1H, dd,

J ) 9.3, 14.9 Hz), 3.02-2.95 (1H, m), 1.97-1.80 (4H, m), 0.92

(9H, s), 0.06 (3H, s), -0.11 (3H, s). ¹³C NMR (75 MHz, CDCl₃)

δ: 166.72, 163.24, 160.35, 160.01, 157.13, 149.25, 140.37,

140.33, 137.90, 137.65, 137.58, 137.12, 136.97, 136.52, 133.52,

133.48 (C), 128.46, 128.32, 128.28, 128.17, 128.02, 127.97,

127.81, 127.76, 127.67, 127.63, 127.52, 127.13, 127.02, 123.32,

118.13, 118.02, 115.90, 115.60, 115.01, 114.72 (CH), 86.83,

79.13, 78.83, 77.73 (CH), 75.56, 75.00, 74.85 (CH₂), 74.19, 73.77

(CH), 73.31 (CH₂), 68.36, 60.57, 60.53 (CH), 51.31, 38.03 (CH₂),

25.85 (CH₃), 24.67 (CH₂), 18.20 (C), -4.57, -4.87 (CH₃). IR

(cm^-1): 2951, 2929, 2858, 1751, 1605, 1510, 1454, 1386, 1362,

1251, 1220, 1151, 1102, 871, 835, 776, 754, 699. MALDI-MS

(C₆₅H₇₁F₂NO₁₀SiS): [MNa]⁺1146.4440 (calcd 1146.4434). Anal.

Calcd for C₆₅H₇₁F₂NO₁₀SiS: C, 69.43; H, 6.36; N, 1.25.

Found: C, 69.27; H, 6.47; N, 1.28. This sulfonate (236 mg,

0.210 mmol) was dissolved in EtOH/EtOAc (10 mL, 1:1 (v/v)),

Pd(OH)₂/C (20% (w/w), 73 mg) was added, and the suspension

was evacuated four times with H₂and stirred under an H₂

atmosphere for 3.5 h. The suspension was evaporated on Celite

and purified by dry column vacuum chromatography (4.6 ×

2.0 cm) on silica gel eluting with a gradient of 0-20% MeOH

in CH₂Cl₂(v/v) to give â-lactam 40 (145 mg, 90%) as a white

foam. R_f(10% MeOH in CH₂Cl₂(v/v)): 0.25. ¹H NMR (300

MHz, acetone-d₆) δ: 7.55 (2H, dd, J ) 6.5, 8.7 Hz), 7.47 (2H,

d, J ) 8.4 Hz), 7.40-7.20 (4H, m), 7.11-6.98 (4H, m), 4.97

(1H, dd, J ) 2.3, 10.5 Hz), 4.83 (1H, bs), 4.61 (1H, bs), 4.48

(1H, bs), 4.30 (1H, bs), 3.90-3.81 (3H, m), 3.71-3.64 (1H, m),

3.56-3.38 (5H, m), 3.25-3.14 (2H, m), 2.66 (1H, t, J ) 7.2

Hz), 1.98-1.81 (4H, m), 0.88 (9H, s), 0.05 (3H, s), -0.15 (3H,

s). ¹³C NMR (75 MHz, acetone-d₆) δ: 168.30, 161.88, 158.69,

151.25, 142.96, 139.63, 139.16, 139.13, 135.98 (C), 131.66,

131.56, 129.36, 129.28, 124.92, 120.00, 119.90, 117.46, 117.16,

116.62, 116.52 (CH), 82.13, 80.16, 76.75, 75.44, 74.46, 72.35

(CH), 63.64 (CH₂), 61.60, 61.55 (CH), 54.03, 39.52 (CH₂), 27.20

(CH₃), 26.35 (CH₂), 19.68 (C), -3.44, -3.69 (CH₃). IR (cm^-1):

3380, 2930, 2858, 1749, 1604, 1510, 1385, 1363, 1220, 1172,

1149, 1088, 1032, 1016, 872, 835, 757. MALDI-MS (C₃₇H₄₇F₂-

NO₁₀SiS): [MNa]⁺786.2563 (calcd 786.2556). Anal. Calcd for

C₃₇H₄₇F₂NO₁₀SiS: C, 58.17; H, 6.20; N, 1.83. Found: C, 58.02;

H, 6.26; N, 1.85.

1

white solid. R_f(1:3 MeOH/EtOAc (v/v)): 0.49. H NMR (300

MHz, suspension in CD₂Cl₂/CD₃OD) δ: 7.37-7.18 (20H, m),

4.86-4.35 (8H, m), 3.86-3.21 (8H, m), 2.97 (1H, dd, J ) 8.7,

14.3 Hz). MALDI-MS (C₃₅H₃₇NaO₈S): [MH]⁺641.1467 (calcd

641.2185); [MNa]⁺663.1206 (calcd 663.2005). This sulfonate

salt (810 mg, 1.26 mmol) was suspended in anhydrous aceto-

nitrile/CH₂Cl₂(30 mL, 2:1 (v/v)) at 0 °C, Ph₃P (2.087 g, 7.96

mmol) and thionyl chloride (1.50 mL, 21 mmol) were added

â-Lactam 16. â-Lactam 40 (31.5 mg, 0.041 mmol) was

dissolved in anhydrous THF (2.5 mL, Teflon bottle), anhydrous

pyridine (0.5 mL) followed by HF‚pyridine complex (0.5 mL)

was added, and the solution was stirred for 24 h, diluted with

ether (20 mL), and washed with saturated aqueous NaHCO₃

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6047

(3 × 5 mL). The organic layer was evaporated on Celite and

purified by dry column vacuum chromatography (4.3 × 2.0 cm)

on silica gel eluting with a gradient of 0-20% MeOH in CH₂-

Cl₂(v/v) to give â-lactam 16 (9.8 mg, 37%) as a white solid. R_f

in AcOH (10 mL), KOAc (933 mg, 9.5 mmol) followed by Oxone

(2KHSO₅·KHSO₄·K₂SO₄, 1.179 g, 2.55 mmol) was added, and

after stirring for 18 h, saturated aqueous Na₂CO₃(50 mL) and

H₂O (50 mL) were carefully added. After extraction with CHCl₃

(4 × 25 mL), the combined organic layer was washed with

saturated aqueous Na₂CO₃(25 mL), evaporated on Celite, and

purified by dry column vacuum chromatography (4.1 × 3.3 cm)

on silica gel eluting with a gradient of 0-20% MeOH in CH₂-

Cl₂(v/v) to give the intermediary sulfonate salt (622 mg, 96%)

as a colorless oil. R_f(10% MeOH in CH₂Cl₂(v/v)): 0.29. ¹H

NMR (300 MHz, CDCl₃) δ: 7.40-7.14 (35H, m), 5.19-4.34

(15H, m), 4.17-3.22 (15H, m). ¹³C NMR (75 MHz, CDCl₃) δ:

138.97, 138.32, 138.21, 138.06, 137.88, 137.84, 128.70, 128.36,

128.18, 128.05, 127.86, 127.76, 127.63, 127.57, 127.44, 127.29,

127.20, 126.94, 84.53, 84.45, 82.01, 79.48, 77.96, 77.75, 76.06,

76.01, 75.46, 74.94, 74.79, 74.67, 74.57, 73.28, 73.08, 73.02,

53.42. IR (cm^-1): 3478, 3063, 3030, 2870, 1497, 1454, 1361,

1315, 1210, 1174, 1069, 1048, 1028, 736, 698, 621. MALDI-

MS (C₆₂H₆₅NaO₁₃S): [MH]⁺1073.4098 (calcd 1073.4122);

[MNa]⁺1095.3926 (calcd 1095.3941). This sulfonate salt (334

mg, 0.311 mmol) was dissolved in anhydrous acetonitrile/CH₂-

Cl₂(4 mL, 1:1 (v/v)) at 0 °C, Ph₃P (264 mg, 1.01 mmol) and

thionyl chloride (0.10 mL, 1.37 mmol) were added sequentially

at 0 °C, and the suspension was stirred at room temperature

for 6 h. EtOAc/hexane (1:4 (v/v), 30 mL) was added, the

suspension was filtered through a short pad of silica gel (4 ×

5 mL EtOAc/hexane (1:3 (v/v)) washings), and the filtrate was

evaporated and dried shortly under vacuum to give sulfonyl

chloride 17 (220 mg, 66%) as a light yellow foam. R_f(1:3 EtOAc/

hexane (v/v)): 0.38. ¹H NMR (300 MHz, CDCl₃) δ: 7.50-7.26

(35H, m), 5.30 (1H, d, J ) 11.2 Hz), 4.98 (1H, d, J ) 10.6 Hz),

4.96-4.81 (5H, m), 4.79 (1H, d, J ) 10.6 Hz), 4.67-4.50 (6H,

m), 4.48 (1H, d, J ) 11.8 Hz), 4.23-4.15 (1H, m), 3.98-3.91

(2H, m), 3.85-3.57 (8H, m), 3.51-3.38 (3H, m), 3.30 (1H, t J

) 9.0 Hz). ¹³C NMR (75 MHz, CDCl₃) δ: 138.77, 138.45,

138.17, 138.11, 137.78, 137.27 (C), 128.63, 128.38, 128.31,

128.18, 128.12, 127.94, 127.78, 127.70, 127.63, 127.55, 127.42,

127.29, 102.32, 84.98, 84.80, 82.66, 79.23, 77.95, 77.82, 75.78

(CH), 75.60, 75.38 (CH₂), 75.12 (CH), 74.99, 74.78, 74.70 (CH₂),

74.21 (CH), 73.24, 68.95, 67.35, 66.79 (CH₂). IR (cm^-1): 3089,

3063, 3030, 2868, 1496, 1454, 1362, 1313, 1280, 1209, 1167,

1091, 1067, 1028, 913, 771, 736, 698, 601. MALDI-MS (C₆₂H₆₅-

ClO₁₂S): [MNa]⁺1091.3767 (calcd 1091.3783).

â-Lactam 18. Sulfonyl chloride 17 (271 mg, 0.253 mmol)

was dissolved in anhydrous CH₂Cl₂(3 mL), anhydrous pyridine

(0.5 mL) followed by phenol 5 (75.7 mg, 0.145 mmol) was

added, and the solution was stirred for 38 h, diluted with

EtOAc (50 mL), and washed sequentially with saturated

aqueous NaHCO₃(15 mL) and H₂O (15 mL). The organic layer

was evaporated on Celite and purified by dry column vacuum

chromatography (4.5 × 3.3 cm) on silica gel eluting with a

gradient of 0-20% EtOAc in toluene (v/v) to give a 4:1 mixture

of the intermediary fully protected sulfonate and unreacted

phenol 5 (166 mg) as a white foam. R_f(1:1 EtOAc/hexane (v/

v)): 0.73. ¹H NMR (300 MHz, CDCl₃) δ: 7.49-7.17 (41H, m),

7.06 (2H, d, J ) 8.7 Hz), 7.02 (2H, t, J ) 8.1 Hz), 6.96 (2H, d,

J ) 8.7 Hz), 5.31 (1H, d, J ) 11.2 Hz), 5.01-4.74 (7H, m),

4.65-4.45 (8H, m), 4.21 (1H, t, J ) 9.3 Hz), 4.02-3.96 (2H,

m), 3.86-3.60 (6H, m), 3.53-3.47 (4H, m), 3.33 (1H, d, J )

9.3 Hz), 3.26 (1H, t, J ) 9.0 Hz), 3.19 (1H, d, J ) 9.3 Hz),

3.06-3.00 (1H, m), 2.06-1.84 (4H, m), 0.96 (9H, s), 0.10 (3H,

s), -0.07 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ: 166.70, 160.35,

160.00, 156.27, 149.33, 140.35, 140.31, 138.63, 138.26, 138.00,

137.90, 137.59, 137.45, 137.29, 136.51, 133.47 (C), 128.82,

128.73, 128.34, 128.19, 128.08, 127.98, 127.85, 127.66, 127.56,

127.45, 127.30, 127.25, 127.12, 127.01, 125.10, 123.32, 118.11,

118.01, 115.91, 115.60, 115.00, 114.93, 114.72, 102.39, 84.93,

84.80, 82.56, 78.82, 78.55, 77.95, 75.99 (CH), 75.60, 75.31

(CH₂), 75.15 (CH), 74.96, 74.76 (CH₂), 74.23, 73.77 (CH), 73.21,

73.08, 68.97, 67.62 (CH₂), 61.02, 60.57, 60.39 (CH), 51.26, 38.02

(CH₂), 25.85 (CH₃), 24.67 (CH₂), 18.19 (C), -4.56, -4.87 (CH₃).

¹⁹F NMR (282 MHz, CDCl₃) δ: -114.94 (1F, septet, J ) 4.3

Hz), -117.10 (1F, septet, J ) 4.3 Hz). MALDI-MS (C₉₂H₉₉F₂-

NO₁₅SiS): [MNa]⁺1578.6365 (calcd 1578.6370). This fully

1

(10% MeOH in CH₂Cl₂(v/v)): 0.22 (run twice). H NMR (300

MHz, acetone-d₆) δ: 7.55 (2H, d, J ) 8.7 Hz), 7.47 (2H, d, J )

8.7 Hz), 7.36 (2H, dd, J ) 5.6, 8.7 Hz), 7.29 (2H, dd, J ) 4.8,

9.2 Hz), 7.06 (2H, d, J ) 8.7 Hz), 7.03 (2H, d, J ) 9.0 Hz),

4.98 (1H, d, J ) 2.5 Hz), 4.68 (1H, bs), 4.58 (1H, bs), 4.38 (1H,

bs), 4.27 (1H, bs), 3.89-3.80 (3H, m), 3.66 (1H, d, J ) 10.6

Hz), 3.54-3.36 (5H, m), 3.24-3.14 (2H, m), 2.00-1.86 (4H,

m). ¹³C NMR (75 MHz, acetone-d₆) δ: 168.48, 151.29, 143.63,

139.23, 136.09 (C), 129.37, 129.29, 129.19, 124.97, 120.05,

119.94, 117.49, 117.18, 116.46, 116.18 (CH), 82.17, 80.18,

76.78, 74.49, 73.79, 72.42 (CH), 63.67 (CH₂), 62.35, 61.63 (CH),

54.06, 38.62, 26.75 (CH₂). IR (cm^-1): 3364, 2924, 1734, 1509,

1388, 1220, 1148, 1102, 872, 835, 769. MALDI-MS (C₃₁H₃₃F₂-

NO₁₀S): [MNa]⁺672.1744 (calcd 672.1691).

Sulfonyl Chloride 17. Alcohol 10⁵⁰(895.3 mg, 0.907 mmol)

was dissolved in anhydrous CH₂Cl₂(10 mL), anhydrous

pyridine (1.0 mL) followed by MsCl (0.20 mL, 2.6 mmol) was

added, and after stirring for 1 h, saturated aqueous NaHCO₃

(40 mL) was added. The layers were separated and the

aqueous layer extracted with EtOAc (3 × 20 mL). The

combined organic layer was washed successively with satu-

rated aqueous NaHCO₃(20 mL) and H₂O (20 mL), evaporated

on Celite, and purified by dry column vacuum chromatography

(4.2 × 3.3 cm) on silica gel eluting with a gradient of 0-50%

EtOAc in hexane (v/v) to give the intermediary mesylate (830.7

mg, 86%) as a white solid. R_f(1:1 EtOAc/hexane (v/v)): 0.67.

¹H NMR (300 MHz, CDCl₃) δ: 7.49-7.24 (35H, m), 5.31 (1H,

d, J ) 11.2 Hz), 5.00 (1H, d, J ) 11.2 Hz), 4.98-4.79 (6H, m),

4.66-4.36 (9H, m), 4.09 (1H, t, J ) 9.3 Hz), 3.90 (1H, dd, J )

2.8, 10.9 Hz), 3.83 (1H, d, J ) 10.0 Hz), 3.75-3.62 (5H, m),

3.55-3.39 (5H, m), 2.97 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

138.97, 138.37, 138.21, 138.04, 137.61 (C), 128.37, 128.29,

128.18, 128.08, 127.93, 127.84, 127.76, 127.38, 127.34, 127.24,

102.51, 84.86, 82.64, 78.70, 77.94, 76.84, 76.53, 76.38, 75.57

(CH), 75.22, 75.09 (CH₂), 74.96, 74.78 (CH₂, CH), 73.21, 73.02,

69.22, 68.89, 67.76 (CH₂), 37.74 (CH₃). IR (cm^-1): 3063, 3030,

2867, 1497, 1454, 1358, 1277, 1209, 1174, 1150, 1092, 1071,

1028, 984, 922, 812, 737, 698, 527. MALDI-MS (C₆₃H₆₈O₁₃S):

[MNa]⁺1087.4284 (calcd 1087.4278). Anal. Calcd for C₆₃H₆₈-

O₁₃S: C, 71.03; H, 6.43. Found: C, 70.94; H, 6.62. This

mesylate (825 mg, 0.774 mmol) was dissolved in EtOH (20

mL), KOSCMe (278 mg, 2.43 mmol), iPrOH (10 mL) and THF

(10 mL) were added, and the orange solution was stirred at

reflux for 3 h (orange precipitate). Additional KOSCMe (512

mg, 4.48 mmol) was added and the suspension was stirred at

reflux for 16 h. After cooling, 50% saturated aqueous NaHCO₃

(100 mL) was added and the suspension was extracted with

ether (4 × 30 mL). The combined organic layer was washed

successively with saturated aqueous NaHCO₃(50 mL) and

H₂O (50 mL), evaporated on Celite, and purified by dry column

vacuum chromatography (4.2 × 3.3 cm) on silica gel eluting

with a gradient of 0-50% EtOAc in hexane (v/v) to give the

intermediary thioacetate (637 mg, 79%) as a light orange solid.

R_f(1:3 EtOAc/hexane (v/v)): 0.45. ¹H NMR (300 MHz, CDCl₃)

δ: 7.43-7.19 (35H, m), 5.22 (1H, d, J ) 11.2 Hz), 4.92 (1H, d,

J ) 11.2 Hz), 4.88 (1H, d, J ) 11.2 Hz), 4.87-4.71 (5H, m),

4.62 (1H, d, J ) 12.5 Hz), 4.60-4.43 (5H, m), 4.41 (1H, d, J )

11.8 Hz), 4.06 (1H, t, J ) 9.3 Hz), 3.86 (1H, dd, J ) 3.7, 11.2

Hz), 3.75 (1H, dd, J ) 1.6, 10.9 Hz), 3.69-3.55 (5H, m), 3.51-

3.31 (6H, m), 3.05 (1H, dd, J ) 6.8, 13.7 Hz), 2.34 (3H, s). ¹³

C

NMR (75 MHz, CDCl₃) δ: 195.04, 139.19, 138.53, 138.30,

138.24, 138.17, 137.96 (C), 128.33, 128.26, 128.20, 128.04,

127.79, 127.71, 127.63, 127.55, 127.47, 127.29, 127.19, 102.40,

85.12, 84.88, 82.71, 79.85, 79.30, 78.05, 77.87 (CH), 75.62,

75.18 (CH₂), 75.09 (CH), 74.94, 74.81, 73.26, 73.21, 68.96,

67.86, 31.12 (CH₂), 30.49 (CH₃). IR (cm^-1): 3030, 2868, 1692,

1496, 1454, 1358, 1210, 1067, 1028, 773, 735, 698, 626.

MALDI-MS (C₆₄H₆₈O₁₁S): [MNa]⁺1067.4365 (calcd 1067.4380).

Anal. Calcd for C₆₄H₆₈O₁₁S: C, 73.54; H, 6.56. Found: C, 73.50;

H, 6.60. This thioacetate (631 mg, 0.604 mmol) was suspended

6048 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

protected sulfonate (166 mg 4:1 mixture) was dissolved in

EtOH (5 mL), Pd(OH)₂/C (20% (w/w), 94 mg) was added, and

the suspension was evacuated four times with H₂and stirred

under an H₂atmosphere for 11.5 h. The suspension was

evaporated on Celite and purified by dry column vacuum

chromatography (4.3 × 2.0 cm) on silica gel eluting with a

gradient of 0-10% MeOH in CH₂Cl₂(v/v) to give the inter-

mediary silylated â-lactam (69.5 mg, 52% from 5) as a colorless

oil. R_f(20% MeOH in CH₂Cl₂(v/v)): 0.46. ¹H NMR (300 MHz,

CD₃OD) δ: 7.46-7.38 (4H, m), 7.31-7.23 (4H, m), 7.04-6.95

(4H, m), 4.75-4.68 (1H, m), 4.44 (1H, d, J ) 8.1 Hz), 3.92-

3.80 (5H, m), 3.69-3.18 (11H, m), 3.10-3.05 (1H, m), 1.95-

1.75 (4H, m), 0.86 (9H, s), 0.01 (3H, s), -0.19 (3H, s). ¹³C NMR

(75 MHz, CD₃OD) δ: 169.31, 169.21, 161.76, 158.91, 150.96,

142.28, 138.45, 135.01, 134.98, 131.06, 130.95 (C), 128.83,

124.50, 119.92, 119.83, 116.99, 116.68, 116.10, 116.04, 115.81,

115.74, 104.54, 80.33, 80.10, 78.11, 77.81, 77.72, 76.30, 75.13,

74.89, 73.61, 71.38 (CH), 62.47, 61.63 (CH₂), 61.56, 61.47 (CH),

53.26, 38.83 (CH₂), 26.38 (CH₃), 25.75 (CH₂), 19.04 (C), -4.40,

-4.70 (CH₃). ¹⁹F NMR (282 MHz, CD₃OD) δ: -117.94 (1F,

septet, J ) 4.3 Hz), -120.10 (1F, septet, J ) 4.3 Hz). MALDI-

MS (C₄₃H₅₇F₂NO₁₅SiS): [MNa]⁺948.3088 (calcd 948.3084).

This silylated â-lactam (59.5 mg, 0.073 mmol) was dissolved

in anhydrous THF (2.0 mL, Teflon bottle), anhydrous pyridine

(0.40 mL) followed by HF‚pyridine complex (0.40 mL) was

added, and the solution was stirred for 14 h. Saturated

aqueous NaHCO₃(5 mL) was added and the suspension was

evaporated on Celite and purified by dry column vacuum

chromatography (4.4 × 2.0 cm) on silica gel eluting with a

gradient of 10-20% MeOH in CH₂Cl₂(v/v) to give â-lactam

18 (38.1 mg, 64%) as a white solid. R_f(10% MeOH in CH₂Cl₂

(v/v)): 0.17 (eluted three times). ¹H NMR (300 MHz, CD₃OD)

δ: 7.45 (2H, t, J ) 9.3 Hz), 7.40 (2H, d, J ) 8.7 Hz), 7.33-

7.24 (4H, m), 7.02 (2H, t, J ) 8.1 Hz), 6.98 (2H, d, J ) 8.7

Hz), 4.90 (1H, d, J ) 1.9 Hz), 4.60 (1H, dd, J ) 5.0, 6.2 Hz),

4.43 (1H, d, J ) 7.5 Hz), 3.92-3.79 (5H, m), 3.69-3.49 (4H,

m), 3.44-3.18 (6H, m), 3.12-3.06 (1H, m), 1.99-1.82 (4H, m).

¹³C NMR (75 MHz, CD₃OD) δ: 169.31, 165.08, 162.17, 161.85,

158.96, 150.98, 142.15, 138.51, 135.01 (C), 128.88, 128.76,

124.46, 119.97, 119.86, 116.99, 116.68, 116.13, 115.84, 104.54,

80.35, 80.06, 78.11, 77.81, 77.71, 76.31, 74.91, 73.77, 73.63,

71.39 (CH), 62.45, 61.50 (CH₂), 61.42 (CH), 53.26, 37.45, 26.12

(CH₂). ¹⁹F NMR (282 MHz, CD₃OD) δ: -118.08 (1F, septet, J

) 4.3 Hz), -120.21 (1F, septet, J ) 4.3 Hz). MALDI-MS

(C₃₇H₄₃F₂NO₁₅S): [MNa]⁺834.2223 (calcd 834.2219).

Azetidine (()-24. LiAlH₄(114 mg, 3.0 mmol) and AlCl₃

(390 mg, 2.9 mmol) were suspended in anhydrous ether (15

mL) and refluxed for 30 min. Azetidinone (()-3⁶²(361 mg, 1.00

mmol) dissolved in anhydrous ether (15 mL) was added, and

after stirring at reflux for 30 min, the suspension was cooled,

and H₂O (5 mL) was added dropwise followed by addition of

50% saturated aqueous NaHCO₃(30 mL). The layers were

separated, the aqueous layer was extracted with EtOAc/hexane

and ether, and the combined organic layer was washed

successively with saturated aqueous NaHCO₃(20 mL) and

H₂O (20 mL), evaporated on Celite, and purified by dry column

vacuum chromatography (3.7 × 3.3 cm) on silica gel eluting

with a gradient of 0-10% EtOAc in hexane (v/v) to give

azetidine (()-24 (281 mg, 81%) as a colorless oil. R_f(1:9 EtOAc/

hexane (v/v)): 0.53. ¹H NMR (300 MHz, CDCl₃) δ: 7.51-7.14

(10H, m), 6.87 (2H, t, J ) 8.7 Hz), 6.38 (2H, dd, J ) 4.7, 9.0

Hz), 4.46 (1H, d, J ) 6.8 Hz), 4.17 (1H, t, J ) 6.8 Hz), 3.35

(1H, dd, J ) 6.8, 7.5 Hz), 2.69-2.58 (3H, m), 1.85-1.56 (4H,

m). ¹³C NMR (75 MHz, CDCl₃) δ: 157.64, 154.52, 148.53,

142.69, 141.95 (C), 128.66, 128.25, 127.47, 125.99, 125.73,

115.41, 115.12, 113.04, 112.94 (CH), 74.37 (CH), 56.05 (CH₂),

42.09 (CH), 35.85, 33.52, 28.92 (CH₂). IR (cm^-1): 3026, 2933,

2852, 1603, 1508, 1473, 1453, 1321, 1222, 1120, 823, 773, 747,

699. MALDI-MS (C₂₄H₂₄FN): [MH]⁺346.1982 (calcd 346.1971).

Anal. Calcd for C₂₄H₂₄FN: C, 83.44; H, 7.00; N, 4.05. Found:

C, 83.45; H, 7.06; N, 4.27.

sequentially, and the solution was stirred for 5 h followed by

addition of 50% saturated aqueous NaHCO₃(50 mL). After

extraction with EtOAc (4 × 20 mL), the combined organic layer

was washed successively with saturated aqueous NaHCO₃(20

mL) and H₂O (20 mL), evaporated on Celite, and purified by

dry column vacuum chromatography (3.8 × 3.3 cm) on silica

gel eluting with a gradient of 0-10% EtOAc in hexane (v/v)

to give silyl ether 25 (424 mg, 97%) as a colorless oil. R_f(1:3

EtOAc/hexane (v/v)): 0.65. ¹H NMR (300 MHz, CDCl₃) δ:

7.25-7.21 (4H, m), 7.17 (2H, d, J ) 8.1 Hz), 6.98 (2H, t, J )

8.7 Hz), 6.91 (2H, t, J ) 8.7 Hz), 6.83 (2H, d, J ) 8.1 Hz), 4.66

(1H, t, J ) 5.6 Hz), 4.51 (1H, d, J ) 2.5 Hz), 3.08-3.02 (1H,

m), 1.96-1.78 (4H, m), 0.98 (9H, s), 0.88 (9H, s), 0.20 (6H, s),

0.02 (3H, s), -0.16 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

167.27, 163.28, 160.27, 160.04, 157.06, 155.71, 140.58, 140.54,

133.89, 133.86 (C), 129.99, 127.22, 127.11, 126.94, 120.56,

118.24, 118.15, 115.74, 115.44, 114.99, 114.72 (CH), 73.84,

61.08, 60.44 (CH), 38.08 (CH₂), 25.90, 25.68 (CH₃), 24.75 (CH₂),

18.26, 18.24 (C), -4.28, -4.52, -4.83 (CH₃). IR (cm^-1): 2954,

2930, 2858, 1752, 1607, 1510, 1385, 1259, 1223, 1101, 1085,

914, 834, 778. MALDI-MS (C₃₆H₄₉F₂NO₃Si₂): [MH - TBDM-

SOH]⁺506.2329 (calcd 506.2327); [MH]⁺638.3289 (calcd

638.3297); [MNa]⁺660.3117 (calcd 660.3117). Anal. Calcd for

C₃₆H₄₉F₂NO₃Si₂: C, 67.78; H, 7.74; N, 2.20. Found: C, 67.70;

H, 7.60; N, 2.02.

Bicycle 26. LiAlH₄(57 mg, 1.5 mmol) and AlCl₃(200 mg,

1.5 mmol) were suspended in anhydrous ether (15 mL),

refluxed for 40 min, and cooled to 0 °C. Azetidinone 25 (180.8

mg, 0.283 mmol) dissolved in anhydrous ether (5 mL) was

added, and after stirring at 0 °C for 30 min, H₂O (1 mL) was

added dropwise. The suspension was evaporated on Celite and

purified by dry column vacuum chromatography (3.5 × 3.3 cm)

on silica gel eluting with a gradient of 0-50% CH₂Cl₂in

hexane (v/v) to give bicycle 26 (110.8 mg, 63%) and olefin 28

(24.1 mg, 16%) as colorless oils. 26: R_f(1:9 EtOAc/hexane (v/

1

v)): 0.23. H NMR (300 MHz, CDCl₃) δ: 7.18-7.14 (2H, m),

6.95 (2H, t, J ) 8.7 Hz), 6.88 (2H, d, J ) 8.7 Hz), 6.74 (2H, d,

J ) 8.1 Hz), 6.68 (1H, dd, J ) 2.8, 8.4 Hz), 6.44 (1H, dd, J )

6.5, 8.7 Hz), 6.38 (1H, dd, J ) 2.8, 9.6 Hz), 4.48 (1H, dd, J )

5.0, 6.8 Hz), 3.78 (1H, bs), 3.61 (1H, d, J ) 7.5 Hz), 3.26 (1H,

dd, J ) 3.1, 11.2 Hz), 2.91 (1H, dd, J ) 7.8, 11.5 Hz), 1.91-

1.85 (1H, m), 1.68-1.44 (3H, m), 1.16-1.04 (1H, m), 0.99 (9H,

s), 0.80 (9H, s), 0.20 (6H, s), 0.06 (3H, s), -0.21 (3H, s). ¹³C

NMR (75 MHz, CDCl₃) δ: 163.60, 160.36, 157.37, 154.27,

141.53, 141.01, 138.13 (C), 130.07, 127.56, 127.46, 125.58,

125.50, 120.01, 117.27, 116.98, 115.17, 114.89, 114.78, 114.08,

113.79 (CH), 74.64, 48.97 (CH), 44.52 (CH₂), 39.89 (CH), 38.67,

28.28 (CH₂), 26.00, 25.90 (CH₃), 18.38, 18.32 (C), -4.16, -4.43,

-4.77 (CH₃). IR (cm^-1): 2955, 2930, 2858, 1607, 1506, 1472,

1408, 1361, 1258, 1222, 1170, 1144, 1085, 1006, 915, 837, 808,

779, 735, 667. MALDI-MS (C₃₆H₅₁F₂NO₂Si₂): [MH - TBDM-

SOH]⁺492.2517 (calcd 492.2534); [M]⁺623.3414 (calcd

623.3426). Anal. Calcd for C₃₆H₅₁F₂NO₂Si₂: C, 69.30; H, 8.24;

N, 2.24. Found: C, 69.47; H, 8.32; N, 2.15. 28: R_f(1:9 EtOAc/

hexane (v/v)): 0.70. ¹H NMR (300 MHz, CDCl₃) δ: 7.29-7.25

(2H, m), 7.18 (2H, t, J ) 8.7 Hz), 6.19 (2H, t, J ) 8.7 Hz), 6.76

(2H, d, J ) 8.7 Hz), 6.30 (1H, d, J ) 15.6 Hz), 6.04 (1H, dd, J

) 6.8, 15.6 Hz), 4.68 (1H, dd, J ) 5.0, 7.5 Hz), 2.26-2.13 (2H,

m), 1.91-1.66 (2H, m), 0.98 (9H, s), 0.89 (9H, s), 0.19 (6H, s),

0.04 (3H, s), -0.16 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

163.42, 160.18, 154.72, 141.28, 131.10, 129.49 (C), 128.25,

127.42, 127.32, 126.87, 120.10, 114.97, 114.69, 73.85 (CH),

40.64, 28.94 (CH₂), 25.84, 25.68 (CH₃), 18.22, 18.18 (C), -4.42,

-4.60, -4.91 (CH₃). IR (cm^-1): 3030, 2956, 2930, 2887, 2858,

1605, 1509, 1472, 1362, 1258, 1223, 1169, 1155, 1088, 1006,

965, 915, 837, 804, 779, 701, 665.

Amino Alcohol 29. Azetidinone 25 (1.880 g, 2.95 mmol)

was dissolved in anhydrous THF (50 mL) at 0 °C, LiAlH₄(534

mg, 14.1 mmol) was added, and the mixture was stirred at 0

°C for 23 h. Saturated aqueous NHCO₃(2 mL) was carefully

added, and the suspension was evaporated on Celite and

purified by dry column vacuum chromatography (4.2 × 5.5 cm)

on silica gel eluting with a gradient of 0-20% EtOAc in hexane

(v/v) to give amino alcohol 29 (1.496 g, 79%) as a colorless oil.

Silyl Ether 25. Ezetimibe 1⁴²(279 mg, 0.681 mmol) was

dissolved in anhydrous DMF (5 mL), imidazole (262 mg, 3.84

mmol) and TBDMSCl (500 mg, 3.32 mmol) were added

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6049

R_f(1:3 EtOAc/hexane (v/v)): 0.47. ¹H NMR (300 MHz, CDCl₃)

δ: 7.19 (2H, dd, J ) 5.6, 8.7 Hz), 7.10 (2H, d, J ) 8.7 Hz),

6.98 (2H, t, J ) 8.7 Hz), 6.78 (2H, t, J ) 8.7 Hz), 6.76 (2H, d,

J ) 8.7 Hz), 6.42 (2H, dd, J ) 4.4, 9.3 Hz), 4.54 (1H, dd, J )

5.0, 6.8 Hz), 4.28 (1H, d, J ) 6.2 Hz), 3.73 (1H, dd, J ) 2.8,

10.9 Hz), 3.60 (1H, dd, J ) 5.3, 10.9 Hz), 1.84-1.76 (1H, m),

1.72-1.50 (3H, m), 1.37-1.20 (1H, m), 0.99 (9H, s), 0.85 (9H,

s), 0.20 (6H, s), -0.03 (3H, s), -0.17 (3H, s). ¹³C NMR (75 MHz,

CDCl₃) δ: 163.17, 159.94, 157.12, 154.33, 154.02, 143.62,

140.94, 140.91, 134.52 (C), 127.62, 127.16, 127.05, 119.84,

115.43, 115.13, 114.88, 114.73, 114.60, 74.41 (CH), 63.62 (CH₂),

61.91, 46.19 (CH), 38.51 (CH₂), 25.85, 25.69 (CH₃), 24.71 (CH₂),

18.22 (C), -4.30, -4.56, -4.88 (CH₃). IR (cm^-1): 3400, 2957,

2938, 2859, 1607, 1510, 1472, 1362, 1257, 1222, 1102, 1086,

914, 837, 779, 736, 668. MALDI-MS (C₃₆H₅₃F₂NO₃Si₂): [MH

- TBDMSOH]⁺510.2624 (calcd 510.2640); [MNa]⁺664.3420

(calcd 664.3430). Anal. Calcd for C₃₆H₅₃F₂NO₃Si₂: C, 67.35;

H, 8.32; N, 2.18. Found: C, 67.40; H, 8.23; N, 2.21.

(2H, t, J ) 8.7 Hz), 4.67 (1H, dd, J ) 4.4, 6.2 Hz), 4.59 (1H, d,

J ) 1.9 Hz), 3.16 (3H, s), 3.04-3.00 (1H, m), 1.93-1.79 (4H,

m), 0.87 (9H, s), 0.01 (3H, s), -0.16 (3H, s). ¹³C NMR (75 MHz,

CDCl₃) δ: 166.83, 163.46, 160.57, 160.21, 157.34, 148.88,

140.53, 140.49, 137.07, 133.59, 133.56 (C), 127.36, 127.28,

127.18, 122.94, 118.26, 118.16, 116.04, 115.73, 115.10, 114.81

(CH), 73.79, 60.67, 60.41 (CH), 37.97 (CH₂), 37.59, 25.76 (CH₃),

24.60 (CH₂), 18.11 (C), -4.71, -5.02 (CH₃). IR (cm^-1): 2952,

2931, 2857, 1752, 1605, 1509, 1371, 1252, 1220, 1176, 1153,

1102, 1086, 971, 871, 835, 777. MALDI-MS (C₃₁H₃₇F₂NO₅-

SiS): [MH - TBDMSOH]⁺470.1228 (calcd 470.12376); [MNa]⁺

624.2029 (calcd 624.2027). Anal. Calcd for C₃₁H₃₇F₂NO₅SiS:

C, 61.87; H, 6.20; N, 2.33. Found: C, 61.69; H, 6.19; N, 2.15.

Azetidine 32. LiAlH₄(58 mg, 1.5 mmol) and AlCl₃(202 mg,

1.5 mmol) were suspended in anhydrous ether (15 mL),

refluxed for 30 min, and cooled to 0 °C. Mesylate 31 (195.5

mg, 0.325 mmol) dissolved in anhydrous ether (5 mL) was

added, and after stirring at 0 °C for 15 min, saturated aqueous

NaHCO₃(1 mL) was added dropwise. The suspension was

evaporated on Celite and purified by dry column vacuum

chromatography (4.6 × 3.3 cm) on silica gel eluting with a

gradient of 0-50% EtOAc in hexane (v/v) to give the inter-

mediary silylated azetidine (146.4 mg, 77%) as a colorless oil.

R_f(1:3 EtOAc/hexane (v/v)): 0.37. ¹H NMR (300 MHz, CDCl₃)

δ: 7.49 (2H, d, J ) 8.7 Hz), 7.30 (2H, d, J ) 8.7 Hz), 7.18 (2H,

dd, J ) 5.0, 8.7 Hz), 6.98 (2H, t, J ) 8.7 Hz), 6.85 (2H, t, J )

8.7 Hz), 6.31 (2H, dd, J ) 4.4, 9.3 Hz), 4.58 (1H, t, J ) 5.3

Hz), 4.40 (1H, d, J ) 6.8 Hz), 4.11 (1H, t, J ) 7.2 Hz), 3.28

(1H, t, J ) 7.2 Hz), 3.17 (3H, s), 2.56-2.49 (1H, m), 1.77-1.50

(4H, m), 0.88 (9H, s), 0.01 (3H, s), -0.15 (3H, s). ¹³C NMR (75

MHz, CDCl₃) δ: 163.22, 159.99, 157.69, 154.57, 148.23, 148.07,

141.97, 140.62 (C), 127.41, 127.13, 127.03, 122.25, 115.43,

115.13, 114.96, 114.68, 113.00, 112.90 (CH), 73.86, 73.29 (CH),

55.88 (CH₂), 41.88 (CH), 37.90 (CH₂), 37.43 (CH₃), 29.43 (CH₂),

25.85 (CH₃), 18.24 (C), -4.53, -4.88 (CH₃). IR (cm^-1): 2932,

2856, 1605, 1509, 1473, 1372, 1331, 1252, 1222, 1198, 1171,

1151, 1090, 970, 870, 836, 776. MALDI-MS (C₃₁H₃₉F₂NO₄-

SiS): [MH - TBDMSOH]⁺456.1442 (calcd 456.14449); [MNa]⁺

610.2236 (calcd 610.22348). Anal. Calcd for C₃₁H₃₉F₂NO₄SiS:

C, 63.34; H, 6.69; N, 2.38. Found: C, 63.49; H, 6.87; N, 2.33.

This silylated azetidine (146.3 mg, 0.249 mmol) was dissolved

in anhydrous THF (5.0 mL, Teflon bottle) at 0 °C, anhydrous

pyridine (1.0 mL) followed by HF‚pyridine complex (1.0 mL)

was added, and the solution was stirred at 0 °C for 1 h and at

room temperature for 7 h, diluted with ether (30 mL), and

washed with saturated aqueous NaHCO₃(3 × 10 mL). The

organic layer was evaporated on Celite and purified by dry

column vacuum chromatography (4.2 × 2.0 cm) on silica gel

eluting with a gradient of 0-90% EtOAc in hexane (v/v) to

give azetidine 32 (100.0 mg, 85%) as a white foam. R_f(1:1

EtOAc/hexane (v/v)): 0.30. ¹H NMR (300 MHz, CDCl₃) δ: 7.50

(2H, d, J ) 8.7 Hz), 7.28 (2H, d, J ) 8.7 Hz), 7.22 (2H, dd, J

) 5.6, 8.7 Hz), 7.01 (2H, t, J ) 8.7 Hz), 6.84 (2H, t, J ) 8.7

Hz), 6.30 (2H, dd, J ) 4.3, 9.3 Hz), 4.57 (1H, t, J ) 5.6 Hz),

4.41 (1H, d, J ) 6.8 Hz), 4.12 (1H, t, J ) 6.8 Hz), 3.30 (1H, dd,

J ) 6.8, 7.5 Hz), 3.16 (3H, s), 2.55 (1H, dt, J ) 6.8, 7.5 Hz),

1.93 (1H, bs), 1.88-1.53 (4H, m). ¹³C NMR (75 MHz, CDCl₃)

δ: 163.62, 160.37, 157.74, 154.61, 148.22, 148.01, 141.89,

139.95, 139.91 (C), 127.46, 127.28, 127.17, 122.29, 115.46,

115.42, 115.13, 113.02, 112.92 (CH), 73.43, 73.28 (CH), 55.92

(CH₂), 41.81 (CH), 37.49 (CH₃), 36.28, 29.85 (CH₂). IR (cm^-1):

3416, 2938, 2853, 1508, 1367, 1221, 1196, 1171, 1149, 970, 871,

823. MALDI-MS (C₂₅H₂₅F₂NO₄S): [MH - H₂O]⁺456.1447

(calcd 456.1445); [M]⁺473.1481 (calcd 473.1472); [MNa]⁺

496.1380 (calcd 496.1370).

Azetidine 30. Amino alcohol 29 was dissolved in anhydrous

acetonitrile (5 mL), Ph₃P (99 mg, 0.38 mmol), CBr₄(133 mg,

0.40 mmol), and anhydrous Et₃N (0.10 mL, 0.71 mmol) were

added sequentially, and the mixture was stirred at room

temperature for 12 h. After dilution with EtOAc/hexane (20

mL, 1:4 (v/v)), the suspension was filtered through a short pad

of silica gel and the filter cake was washed with EtOAc/hexane

(2 × 25 mL, 1:4 (v/v)). The filtrates were evaporated on Celite

and purified by dry column vacuum chromatography (4.0 ×

3.3 cm) on silica gel eluting with a gradient of 0-7% EtOAc

in hexane (v/v) to give azetidine 30 (85 mg, 60%) and the

acyclic amino bromide (52.8 mg, 33%) as colorless oils. 30: R_f

1

(1:19 EtOAc/hexane (v/v)): 0.57. H NMR (300 MHz, CDCl₃)

δ: 7.32 (2H, d, J ) 8.1 Hz), 7.18-7.14 (2H, m), 6.98 (2H, t, J

) 8.7 Hz), 6.87-6.81 (4H, m), 6.34 (2H, dd, J ) 4.4, 8.7 Hz),

6.42 (2H, dd, J ) 4.4, 9.3 Hz), 4.55 (1H, dd, J ) 5.0, 6.2 Hz),

4.31 (1H, d, J ) 6.8 Hz), 4.08 (1H, dd, J ) 6.2, 7.5 Hz), 3.23

(1H, dd, J ) 6.8, 7.4 Hz), 2.57 (1H, dd, J ) 6.8, 7.5 Hz), 1.78-

1.50 (4H, m), 1.02 (9H, s), 0.89 (9H, s), 0.24 (6H, s), 0.02 (3H,

s), -0.16 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ: 163.24, 160.00,

158.64, 157.58, 154.97, 154.47, 148.67, 148.64, 140.89, 140.86,

135.39 (C), 127.18, 127.07, 126.78, 120.07, 120.02, 115.33,

115.04, 114.92, 114.63, 113.05, 112.95, 74.07, 73.87 (CH), 55.78

(CH₂), 41.91 (CH), 37.96, 29.52 (CH₂), 25.90, 25.77 (CH₃), 18.29

(C), -4.23, -4.48, -4.83 (CH₃). MALDI-MS (C₃₆H₅₁F₂NO₂-

Si₂): [MH - TBDMSOH]⁺492.2518 (calcd 492.2534); [M]⁺

623.3405 (calcd 623.3426); [MNa]⁺646.3314 (calcd 646.3324).

Acyclic Amino Bromide: R_f(1:19 EtOAc/hexane (v/v)):

0.23. ¹H NMR (300 MHz, CDCl₃) δ: 7.19 (2H, dd, J ) 5.6, 8.7

Hz), 7.09 (2H, d, J ) 8.7 Hz), 6.97 (2H, t, J ) 8.7 Hz), 6.79

(2H, t, J ) 8.7 Hz), 6.76 (2H, d, J ) 8.7 Hz), 6.47 (2H, dd, J )

4.4, 8.7 Hz), 4.51 (1H, dd, J ) 3.7, 7.5 Hz), 4.37 (1H, d, J )

6.8 Hz), 3.94 (1H, bs), 3.72 (1H, dd, J ) 3.7, 10.0 Hz), 3.40

(1H, dd, J ) 5.0, 10.6 Hz), 2.00-1.91 (1H, m), 1.74-1.47 (3H,

m), 1.32-1.15 (1H, m), 0.99 (9H, s), 0.83 (9H, s), 0.20 (6H, s),

-0.06 (3H, s), -0.20 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

163.36, 160.12, 157.44, 154.80, 154.32, 143.37, 141.15, 141.12,

133.11 (C), 127.93, 127.29, 127.19, 127.09, 126.87, 120.43,

120.01, 115.56, 115.27, 114.98, 114.86, 114.69, 114.64, 74.26,

60.13, 45.62 (CH), 38.34, 35.91, 25.83 (CH₂), 25.75, 25.59 (CH₃),

18.05 (C), -4.44, -4.72, -5.06 (CH₃). IR (cm^-1): 2956, 2930,

2858, 1607, 1509, 1472, 1362, 1257, 1222, 1100, 1086, 916, 837,

776. MALDI-MS (C₃₆H₅₂BrF₂NO₂Si₂): [MH - H₂NC₆H₄F]⁺

593.2283 (calcd 593.2282).

Mesylate 31. Phenol 5 (176 mg, 0.336 mmol) was dissolved

in anhydrous CH₂Cl₂(10 mL), anhydrous pyridine (0.5 mL)

followed by MsCl (0.1 mL, 1.29 mmol) was added, and the

solution was stirred for 22 h, diluted with EtOAc (50 mL), and

washed sequentially with saturated aqueous NaHCO₃(20 mL)

and H₂O (20 mL). The organic layer was evaporated on Celite

and purified by dry column vacuum chromatography (4.2 ×

3.3 cm) on silica gel eluting with a gradient of 0-50% EtOAc

in hexane (v/v) to give mesylate 31 (195.5 mg, 92%) as a

colorless oil. R_f(1% MeOH in CH₂Cl₂(v/v)): 0.74. ¹H NMR

(300 MHz, CDCl₃) δ: 7.35 (2H, d, J ) 8.7 Hz), 7.28 (2H, d, J

) 8.7 Hz), 7.26-7.18 (4H, m), 6.98 (2H, t, J ) 8.7 Hz), 6.93

â-Lactam 33. â-Lactam 31 (67.7 mg, 0.112 mmol) was

dissolved in THF (2 mL), TBAF (0.2 mL, 1 M in THF) was

added, and the solution was stirred for 1.5 h, diluted with

EtOAc (20 mL), and washed successively with saturated

aqueous NaHCO₃(10 mL) and H₂O (10 mL). The organic layer

was evaporated on Celite and purified by dry column vacuum

chromatography (4.2 × 2.0 cm) on silica gel eluting with a

gradient of 0-90% EtOAc in hexane (v/v) to give â-lactam 33

(37.0 mg, 68%) as a white solid. R_f(1:1 EtOAc/hexane (v/v)):

6050 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

0.17. ¹H NMR (300 MHz, CDCl₃) δ: 7.37-7.17 (8H, m), 7.03-

6.91 (4H, m), 4.69 (1H, t, J ) 5.9 Hz), 4.65 (1H, d, J ) 1.9

Hz), 3.16 (3H, s), 3.07-3.01 (1H, m), 2.63 (1H, bs), 2.03-1.84

(4H, m). ¹³C NMR (75 MHz, CDCl₃) δ: 167.11, 163.76, 160.68,

160.50, 157.44, 148.89, 139.92, 136.86, 133.41 (C), 127.40,

127.27, 122.98, 118.35, 118.24, 116.10, 115.79, 115.45, 115.18,

115.11 (CH), 73.03, 60.48, 60.41 (CH), 37.63 (CH₃), 36.48, 25.00

(CH₂). IR (cm^-1): 3428, 2937, 1744, 1604, 1510, 1426, 1369,

1221, 1176, 1152, 1103, 1016, 971, 912, 872, 835, 788, 734.

MALDI-MS (C₂₅H₂₃F₂NO₅S): [MH - H₂O]⁺470.1239 (calcd

470.1238); [MNa]⁺510.1164 (calcd 510.1163). Anal. Calcd for

C₂₅H₂₃F₂NO₅S: C, 61.59; H, 4.75; N, 2.87. Found: C, 61.79;

H, 4.89; N, 2.76.

chromatography (4.8 × 2.0 cm) on silica gel eluting with a

gradient of 0-50% EtOAc in hexane (v/v) to give azetidine 36

(24.5 mg, 40%) as a white foam. R_f(1:1 EtOAc/hexane (v/v)):

0.46. ¹H NMR (300 MHz, CDCl₃) δ: 7.88-7.82 (2H, m), 7.70-

7.63 (1H, m), 7.55-7.47 (2H, m), 7.38-7.30 (2H, m), 7.24-

7.19 (2H, m), 7.05-6.98 (4H, m), 6.83 (2H, t, J ) 8.7 Hz), 6.26

(2H, dd, J ) 4.4, 9.3 Hz), 4.56 (1H, dd, J ) 5.0, 7.5 Hz), 4.36

(1H, d, J ) 6.8 Hz), 4.09 (1H, dd, J ) 6.8, 7.5 Hz), 3.27 (1H,

dd, J ) 6.8, 7.5 Hz), 2.79 (1H, d, J ) 5.6 Hz), 2.52 (1H, dd, J

) 6.8, 7.5 Hz), 1.89-1.52 (4H, m). ¹³C NMR (75 MHz, CDCl₃)

δ: 163.65, 160.40, 157.73, 154.61, 148.65, 148.07, 141.59,

139.94, 135.36 (C), 134.11, 129.24, 129.03, 128.30, 127.28,

127.17, 122.54, 115.45, 115.15, 112.99, 112.90, 73.47, 73.32

(CH), 55.89 (CH₂), 41.74 (CH), 36.30, 29.93 (CH₂). IR (cm^-1):

3411, 2937, 2853, 1604, 1508, 1474, 1450, 1374, 1221, 1198,

1175, 1151, 1093, 1016, 867, 823, 752, 700, 686. MALDI-MS

(C₃₀H₂₇F₂NO₄S): [MH - H₂O]⁺518.1596 (calcd 518.1601); [M]⁺

535.1619 (calcd 535.1629); [MNa]⁺558.1512 (calcd 558.1527).

â-Lactam 35. Phenol 5 (104.0 mg, 0.199 mmol) was dis-

solved in anhydrous CH₂Cl₂(10 mL), anhydrous pyridine (0.5

mL) followed by PhSO₂Cl (0.10 mL, 0.78 mmol) was added,

and the solution was stirred for 19 h. Additional PhSO₂Cl (0.10

mL, 0.78 mmol) was added, and the solution was stirred for a

further 69 h, diluted with EtOAc (50 mL), and washed

sequentially with saturated aqueous NaHCO₃(20 mL) and

H₂O (20 mL). The organic layer was evaporated on Celite and

purified by dry column vacuum chromatography (4.2 × 3.3 cm)

on silica gel eluting with a gradient of 0-100% CH₂Cl₂in

hexane (v/v) followed by 0.5-1.0% MeOH in CH₂Cl₂(v/v) to

give the intermediary silylated benzene sulfonate (92.0 mg,

70%) as a colorless oil. R_f(1% MeOH in CH₂Cl₂(v/v)): 0.72.

¹H NMR (300 MHz, CDCl₃) δ: 7.83 (2H, d, J ) 7.5 Hz), 7.66

(1H, t, J ) 7.5 Hz), 7.51 (2H, t, J ) 7.5 Hz), 7.25-7.14 (6H,

m), 7.00 (2H, d, J ) 8.7 Hz), 6.97 (2H, d, J ) 8.7 Hz), 6.91

(2H, t, J ) 8.7 Hz), 4.66 (1H, dd, J ) 4.4, 6.2 Hz), 4.55 (1H, d,

J ) 1.9 Hz), 3.02-2.96 (1H, m), 1.94-1.75 (4H, m), 0.87 (9H,

s), 0.00 (3H, s), -0.16 (3H, s). ¹³C NMR (75 MHz, CDCl₃) δ:

166.82, 163.44, 160.51, 157.29, 149.35, 140.54 (C), 136.77 (CH),

135.26, 134.32, 133.56, 129.13 (C), 128.31, 127.27, 127.17,

127.08, 123.15, 118.23, 118.12, 115.93, 115.63, 115.09, 114.82,

73.76, 60.53, 60.40 (CH), 37.92 (CH₂), 25.75 (CH₃), 24.59 (CH₂),

18.10 (C), -4.71, -5.03 (CH₃). IR (cm^-1): 2953, 2930, 2857,

1752, 1605, 1510, 1450, 1382, 1252, 1221, 1202, 1181, 1155,

1093, 1016, 868, 835, 776, 753, 700, 687. MALDI-MS (C₃₆H₃₉F₂-

NO₅SSi): [MH - TBDMSOH]⁺532.1395 (calcd 532.1394);

[MNa]⁺686.2185 (calcd 686.2184). This silylated benzene

sulfonate (90.0 mg, 0.136 mmol) was dissolved in anhydrous

THF (2.5 mL, Teflon bottle) at 0 °C, anhydrous pyridine (0.5

mL) followed by HF‚pyridine complex (0.5 mL) was added, and

the solution was allowed to warm slowly to room temperature.

After 14 h, the mixture was diluted with ether (20 mL) and

washed with saturated aqueous NaHCO₃(3 × 5 mL). The

organic layer was evaporated on Celite and purified by dry

column vacuum chromatography (5.0 × 2.0 cm) on silica gel

eluting with a gradient of 0-100% CH₂Cl₂in hexane (v/v)

followed by 1-7% MeOH in CH₂Cl₂(v/v) to give â-lactam 35

(69.2 mg, 93%) as a white foam. R_f(3% MeOH in CH₂Cl₂(v/

v)): 0.33. ¹H NMR (300 MHz, CDCl₃) δ: 7.82 (2H, dd, J ) 1.2,

7.5 Hz), 7.67 (1H, tt, J ) 1.2, 7.5 Hz), 7.51 (2H, t, J ) 7.5 Hz),

7.29-7.22 (4H, m), 7.15 (2H, dd, J ) 4.4, 8.7 Hz), 6.99 (2H, t,

J ) 8.7 Hz), 6.98 (2H, d, J ) 8.7 Hz), 6.92 (2H, t, J ) 8.7 Hz),

4.68 (1H, dd, J ) 5.6, 6.2 Hz), 4.60 (1H, d, J ) 1.9 Hz), 3.06-

2.98 (1H, m), 2.55 (1H, bs), 2.04-1.84 (4H, m). ¹³C NMR (75

MHz, CDCl₃) δ: 166.87, 163.56, 160.44, 160.31, 157.22, 149.23,

139.84, 139.79 (C), 136.46 (CH), 135.10, 134.26, 133.37 (C),

129.07, 128.22, 127.26, 127.16, 127.02, 123.07, 118.21, 118.11,

115.93, 115.62, 115.36, 115.07, 72.98, 60.50, 60.32 (CH), 36.54,

25.09 (CH₂). IR (cm^-1): 3440, 3069, 3017, 2927, 2862, 1747,

1604, 1510, 1450, 1426, 1378, 1221, 1201, 1180, 1154, 1094,

1016, 868, 835, 753, 700, 687, 668. MALDI-MS (C₃₀H₂₅F₂-

NO₅S): [MH - H₂O]⁺532.1388 (calcd 532.1394); [MNa]⁺

572.1302 (calcd 572.1319).

tert-Butyl Ether 38. 2-Methylpropene (10 mL) was con-

densed in a dried flask at -78 °C, and anhydrous CH₂Cl₂(5

mL), followed by phenol 37^17,36(83.7 mg, 0.185 mmol) and 5

drops of triflic acid, was added sequentially. After 5 min, the

suspension was transferred to a -20 °C cooling bath. After 40

min, Et₃N (0.5 mL) was added and the solution was stirred at

room temperature until all 2-methylpropene had evaporated.

The solution was evaporated on Celite and purified by dry

column vacuum chromatography (4.7 × 2.0 cm) on silica gel

eluting with a gradient of 0-50% EtOAc in hexane (v/v) to

give the intermediary acetylated tert-butyl ether (81.8 mg,

1

87%) as a white foam. R_f(1:1 EtOAc/hexane (v/v)): 0.47. H

NMR (300 MHz, CDCl₃) δ: 7.29-7.19 (6H, m), 7.01 (2H, t, J

) 8.7 Hz), 6.98 (2H, d, J ) 8.7 Hz), 6.91 (2H, t, J ) 8.7 Hz),

5.70 (1H, t, J ) 6.8 Hz), 4.55 (1H, d, J ) 1.9 Hz), 3.08 (1H, dt,

J ) 1.9, 7.5 Hz), 2.10-1.98 (2H, m), 2.05 (3H, s), 1.90-1.81

(2H, m), 1.35 (9H, s). ¹³C NMR (75 MHz, CDCl₃) δ: 170.11,

166.95, 163.97, 160.71, 160.50, 157.26, 155.84, 135.70, 133.81

(C), 131.72, 128.23, 128.11, 126.41, 124.46, 118.33, 118.23,

115.85, 115.58, 115.29 (CH), 78.79 (C), 74.75, 60.85, 59.90

(CH), 33.53 (CH₂), 28.78 (CH₃), 24.83 (CH₂), 21.13 (CH₃). IR

(cm^-1): 2979, 2935, 1747, 1608, 1510, 1389, 1368, 1234, 1159,

1102, 1015, 896, 835, 757. MALDI-MS (C₃₀H₃₁F₂NO₄): [MNa]⁺

530.2111 (calcd 530.2119). This acetylated tert-butyl ether

(78.2 mg, 0.154 mmol) was dissolved in anhydrous MeOH (5

mL), KCN (48 mg, 0.74 mmol) was added, and the solution

was stirred at room temperature for 4 h, evaporated on Celite,

and purified by dry column vacuum chromatography (4.6 ×

2.0 cm) on silica gel eluting with a gradient of 0-50% EtOAc

in hexane (v/v) to give tert-butyl ether 38 (51.1 mg, 71%) as a

1

colorless oil. R_f(1:3 EtOAc/hexane (v/v)): 0.21. H NMR (300

MHz, CDCl₃) δ: 7.30-7.18 (6H, m), 7.02-6.87 (6H, m), 4.70

(1H, t, J ) 5.6 Hz), 4.57 (1H, d, J ) 1.9 Hz), 3.10-3.04 (1H,

m), 2.61 (1H, bs), 2.03-1.87 (4H, m), 1.34 (9H, s). ¹³C NMR

(75 MHz, CDCl₃) δ: 167.65, 163.72, 160.46, 157.31, 155.76,

140.05, 133.81, 131.83 (C), 127.40, 127.29, 126.88, 126.45,

124.47, 124.13, 118.39, 118.29, 115.88, 115.58, 115.40, 115.29,

115.11, 114.30, 114.20 (CH), 78.84 (C), 72.95, 61.06, 60.13

(CH), 36.54 (CH₂), 28.79 (CH₃), 24.96 (CH₂). IR (cm^-1): 3424,

2979, 2934, 1737, 1606, 1510, 1390, 1367, 1222, 1158, 895, 835,

757. MALDI-MS (C₂₈H₂₉F₂NO₃): [MH - H₂O]⁺448.2082 (calcd

448.2088); [MNa]⁺488.2001 (calcd 488.2013). Anal. Calcd for

C₂₈H₂₉F₂NO₃: C, 72.24; H, 6.28; N, 3.01. Found: C, 72.39; H,

6.51; N, 2.95.

Azetidine 39. LiAlH₄(57 mg, 1.5 mmol) and AlCl₃(200 mg,

1.5 mmol) were suspended in anhydrous ether (15 mL),

refluxed for 30 min, and cooled to 0 °C. â-Lactam 14 (26.8 mg,

0.041 mmol) dissolved in anhydrous THF (1 mL, 2 × 0.5 mL

rinse) was added, and after stirring at 0 °C for 10 min,

saturated aqueous NaHCO₃(1 mL) was added dropwise. The

suspension was evaporated on Celite and purified by dry

column vacuum chromatography (4.7 × 2.0 cm) on silica gel

eluting with a gradient of 0-12% MeOH in CH₂Cl₂(v/v) to

give azetidine 39 (20.4 mg, 78%) as a colorless oil. R_f(10%

MeOH in CH₂Cl₂(v/v)): 0.20. ¹H NMR (300 MHz, acetone-d₆)

δ: 7.63-7.59 (2H, m), 7.49-7.42 (2H, m), 7.36-7.29 (2H, m),

Azetidine 36. LiAlH₄(57 mg, 1.5 mmol) and AlCl₃(202 mg,

1.5 mmol) were suspended in anhydrous ether (15 mL),

refluxed for 30 min, and cooled to 0 °C. â-Lactam 35 (62.8 mg,

0.114 mmol) dissolved in anhydrous ether (5 mL) was added,

and after stirring at 0 °C for 20 min, saturated aqueous

NaHCO₃(1 mL) was added dropwise. The suspension was

evaporated on Celite and purified by dry column vacuum

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6051

7.10-7.01 (2H, m), 6.92-6.77 (2H, m), 6.40-6.35 (2H, m), 4.72

(1H, d, J ) 3.7 Hz), 4.62 (1H, d, J ) 5.0 Hz), 4.61 (1H, bs),

4.52 (1H, d, J ) 6.9 Hz), 4.31 (2H, t, J ) 4.4 Hz), 4.21-4.15

(2H, m), 3.90 (1H, dd, J ) 1.2, 14.9 Hz), 3.76 (1H, d, J ) 8.1

Hz), 3.68 (1H, dd, J ) 3.7, 9.3 Hz), 3.66-3.57 (2H, m), 3.41

(3H, s, OMe), 3.38-3.31 (1H, m), 3.25 (1H, dt, J ) 5.0, 13.7

Hz), 2.62 (1H, dd, J ) 6.8, 14.3 Hz), 1.92-1.84 (1H, m), 1.74-

1.57 (3H, m). ¹³C NMR (75 MHz, acetone-d₆) δ: 163.90, 160.69,

158.31, 155.22, 149.93, 149.72, 149.52, 142.90, 142.84 (C),

129.60, 129.44, 128.30, 128.24, 128.13, 123.51, 122.99, 115.95,

115.91, 115.66, 115.40, 115.11, 113.87, 113.77, 113.67, 113.57

(CH), 100.84, 74.86, 74.03, 73.68, 73.14, 72.87, 68.09 (CH),

56.67 (CH₂), 55.63 (CH₃), 52.83 (CH₂), 42.78 (CH), 37.60, 29.83

(CH₂). IR (cm^-1): 3390, 2935, 2850, 1605, 1508, 1474, 1366,

1221, 1147, 1052, 1015, 874, 824, 755. MALDI-MS (C₃₁H₃₅F₂-

NO₉S): [MH - H₂O]⁺618.1968 (calcd 618.1973); [MH]⁺

636.2045 (calcd 636.2079); [MNa]⁺658.1901 (calcd 658.1898).

³H]cholesterol oleyl ether (37 Ci/mmol), [4-¹⁴C-cholesterol], and

Sepharose CL-4B were from Amersham Biosciences, the BCA

protein assay kit was from Pierce, and the glucose dehydro-

genase kit was from Diagnostic Systems.

Preparation of Brush Border Membrane Vesicles.

Brush border membrane vesicles were prepared and charac-

terized [total protein content by the BCA method, sucrase

activity and lipid uptake (4.2 mg of protein/mL, 0.20 mg SUV/

mL; see below)] essentially as previously described.^23,38The

source was small intestine (stored at -78 °C) from freshly

killed farm rabbits. The isolation buffer was 2 mM Tris-HCl

plus HCl to pH 7.1, 50 mM ^D-mannitol, and 0.83 mM EGTA;

10 mM MgC1₂was used in the precipitation step. The brush

border pellet was redispersed in 12 mM Tris-HCl plus HCl to

pH 7.1, 0.30 M ^D-mannitol, and 5 mM EGTA.

Preparation of Small Unilamellar Vesicles (SUV). A

total of 2 mg of egg phosphatidylcholine and cholesteryl oleate

(99:1 molar ratio) for control measurements and egg phos-

phatidylcholine, cholesteryl oleate, and inhibitor (90:1:9 molar

ratio) for inhibition experiments and in either case a trace

amount ³H-labeled cholesteryl oleyl ether (or ¹⁴C-labeled

cholesterol) were dried from a chloroform-methanol solution

(2:l v/v) by rotary evaporation. The lipid film was dried under

high vacuum for at least 1 h and then dispersed in PBS buffer

(2 mL). The suspension was sonicated with a microtip sonicator

(Branson 250) for 1-1.5 h (output 2.2, 60% duty cycle).⁷⁰After

sonication, the vesicles were centrifuged (pressure 3.0, 3 min)

in a Beckman airfuge and characterized by gel filtration

(Sepharose CL-4B, 45 × 1 cm) as previously reported.^35,71

Inhibition of Cholesterol Absorption by Brush Border

Membrane Vesicles. Brush border membrane vesicles (5.0

mg of protein/mL) were incubated at room temperature for 20

min with either control SUV (99:1 molar ratio egg phosphati-

dylcholine and cholesteryl oleate) or SUV containing inhibitors

(90:1:9 molar ratio egg phosphatidylcholine, cholesteryl oleate,

and inhibitor). The experiment was terminated by centrifuga-

tion (pressure 3.0, 3 min) in a Beckman airfuge. The donor

SUV remained in the supernatant under these conditions and

the brush border membrane vesicles precipitated. The radio-

activity present in both donor SUV and brush border mem-

brane vesicles was counted in triplicate in a Beckman LS 7500

liquid scintillation counter.

Azetidine 41. LiAlH₄(57 mg, 1.5 mmol) and AlCl₃(200 mg,

1.5 mmol) were suspended in anhydrous ether (15 mL),

refluxed for 30 min, and cooled to 0 °C. â-Lactam 40 (41.3 mg,

0.054 mmol) dissolved in anhydrous ether (5 mL) was added,

and after stirring at 0 °C for 10 min, saturated aqueous

NaHCO₃(1 mL) was added dropwise. The suspension was

evaporated on Celite and purified by dry column vacuum

chromatography (4.2 × 2.0 cm) on silica gel eluting with a

gradient of 0-20% MeOH in CH₂Cl₂(v/v) to give the inter-

mediary silylated azetidine (38.2 mg, 94%) as a white foam.

R_f(10% MeOH in CH₂Cl₂(v/v)): 0.31. ¹H NMR (300 MHz,

acetone-d₆) δ: 7.58 (2H, d, J ) 8.7 Hz), 7.47 (2H, d, J ) 8.7

Hz), 7.29 (2H, dd, J ) 5.6, 8.7 Hz), 7.05 (2H, t, J ) 8.7 Hz),

6.88 (2H, t, J ) 9.0 Hz), 6.37 (2H, dd, J 4.7, 9.0 Hz), 4.71 (1H,

t, J ) 5.5 Hz), 4.61 (1H, d, J ) 5.0 Hz), 4.49 (2H, d, J ) 6.8

Hz), 4.30 (1H, bs), 4.17 (1H, t, J ) 7.2 Hz), 3.92-3.83 (3H, m),

3.74-3.66 (1H, m), 3.57-3.40 (5H, m), 3.32-3.15 (2H, m),

2.63-2.56 (1H, m), 1.82-1.56 (4H, m), 0.87 (9H, s), 0.04 (3H,

s), -0.17 (3H, s). ¹³C NMR (75 MHz, acetone-d₆) δ: 164.97,

161.76, 159.31, 156.21, 150.76, 150.47, 150.45, 143.77, 143.11,

143.07 (C), 129.35, 129.22, 124.60, 116.95, 116.65, 116.48,

116.19, 114.86, 114.75 (CH), 82.15, 80.21, 76.81, 75.43, 74.99,

74.52, 72.41 (CH), 63.70, 57.54, 53.95 (CH₂), 43.62 (CH), 39.47,

31.22 (CH₂), 27.20 (CH₃), 19.70 (C), -3.40, -3.68 (CH₃). IR

(cm^-1): 3377, 2930, 2856, 1605, 1508, 1472, 1361, 1252, 1222,

1147, 1090, 1015, 871, 836, 776, 760. MALDI-MS (C₃₇H₄₉F₂-

NO₉SSi): [MNa]⁺772.2767 (calcd 772.2763). This silylated

azetidine (34.3 mg, 0.046 mmol) was dissolved in anhydrous

THF (2.5 mL, Teflon bottle), anhydrous pyridine (0.5 mL)

followed by HF‚pyridine complex (0.5 mL) was added, and the

solution was stirred for 14 h, diluted with ether (20 mL), and

washed with saturated aqueous NaHCO₃(3 × 5 mL). The

organic layer was evaporated on Celite and purified by dry

column vacuum chromatography (4.9 × 2.0 cm) on silica gel

eluting with a gradient of 0-18% MeOH in CH₂Cl₂(v/v) to

give azetidine 41 (20.2 mg, 69%) as a colorless oil. R_f(10%

MeOH in CH₂Cl₂(v/v)): 0.24. ¹H NMR (300 MHz, acetone-d₆)

δ: 7.61 (2H, d, J ) 8.1 Hz), 7.48 (2H, d, J ) 8.7 Hz), 7.30 (2H,

dd, J ) 5.6, 8.7 Hz), 7.04 (2H, t, J ) 8.7 Hz), 6.89 (2H, m),

6.38 (2H, dd, J ) 4.4, 8.7 Hz), 4.60 (2H, d, J ) 4.4 Hz), 4.52

(1H, d, J ) 6.8 Hz), 4.45 (1H, d, J ) 2.5 Hz), 4.29 (2H, d, J )

4.4 Hz), 4.19 (1H, t, J ) 6.8 Hz), 4.03-3.83 (3H, m), 3.80-

3.67 (1H, m), 3.60-3.31 (6H, m), 3.25 (1H, p, J ) 4.4 Hz), 2.62

(1H, dd, J ) 7.5, 14.3 Hz), 1.92-1.82 (1H, m), 1.78-1.61 (3H,

m). ¹³C NMR (75 MHz, acetone-d₆) δ: 164.04, 155.14, 149.92,

149.71, 149.47, 142.77, 129.48 (C), 128.19, 128.16, 128.05,

123.52, 123.03, 115.87, 115.58, 115.39, 115.32, 115.05, 113.78,

113.69, 113.61, 113.51 (CH), 81.09, 79.15, 75.76, 73.98, 73.46,

72.75, 71.36 (CH), 62.63, 56.60, 52.88 (CH₂), 42.68 (CH), 37.52,

29.61 (CH₂). IR (cm^-1): 3370, 2933, 1605, 1508, 1474, 1360,

1220, 1146, 1087, 1015, 873, 823, 771. MALDI-MS (C₃₁H₃₅F₂-

NO₉S): [MH - H₂O]⁺618.1973 (calcd 618.1973); [M]⁺635.1996

(calcd 635.2001); [MNa]⁺658.1900 (calcd 658.1898).

Percent inhibition was calculated from relative radioactivi-

ties in the supernatants and pellets according to the formula:

% inhibition ) [(% supernatant inhibitor SUV - % superna-

tant control SUV) × 100%]/% pellet control SUV.

The following are the obtained inhibitions: 1, 16 ( 4%; (()-

3, 2 ( 2%; 4,³⁶27 ( 4%; 7, 13 ( 4%; 9, 15 ( 3%; 11, 28 ( 4%;

14, 20 ( 5%; 16, 15 ( 3%; 18, 41 ( 4%; 19, 10 ( 3%; 20,³⁶19

( 4%; 21,³⁶14 ( 2%; 22,³⁶3 ( 4%; (()-24, 2 ( 3%; 32, 22 ( 2%;

33, 30 ( 4%; 34, <2 ( 2%; 35, 26 ( 3%; 36, 19 ( 3%, 38, 10

( 4%; 39, 27 ( 4%; 41, 20 ( 5%; 44, 3 ( 3%; 45, 4 ( 3%.

Acknowledgment. This research was supported by

a KTI grant (6813.2 BTS-LS) and Lipideon AG. L.K. was

supported by a fellowship from The Technical Univer-

sity of Denmark.

Supporting Information Available: Experimental pro-

cedures and spectral data for all described compounds (PDF).

This material is available free of charge via the Internet at

http://pubs.acs.org.

References

(1) Sudhop, T.; von Bergmann, K. Cholesterol absorption inhibitors

for the treatment of hypercholesterolaemia. Drugs 2002, 62,

2333-2347, and references therein.

(2) Bruckert, E. New advances in lipid-modifying therapies for

reducing cardiovascular risk. Cardiology 2002, 97, 59-66, and

references therein.

(3) American Heart Association. Heart Disease and Stroke. Statis-

ticss2003 Update. Available from The American Heart Associa-

tion at http://www.americanheart.org/downloadable/heart/1059-

0179711482003HDSStatsBookREV10590179711482007-105-

90179 711482003.pdf.

Brush Border Membrane Vesicle Assay. Materials. Egg

phosphatidylcholine was purchased from Avanti Polar Lipids,

cholesterol oleate and cholesterol were from Sigma, phosphate-

buffered saline (PBS) was from Invitrogen Corp., [1R,2R(N)-

6052 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19

Kværnø et al.

(4) Bays, H. Ezetimibe. Expert Opin. Investig. Drugs 2002, 11,

1587-1604.

(22) Hauser, H.; Dyer, J. H.; Nandy, A.; Vega, M. A.; Werder, M.;

Bieliauskaite, E.; Weber, F. E.; Compassi, S.; Gemperli, A.;

Boffelli, D.; Wehrli, E.; Schulthess, G.; Phillips, M. C. Identifica-

tion of a receptor mediating absorption of dietary cholesterol in

the intestine. Biochemistry 1998, 37, 17843-17850.

(23) Thurnhofer, H.; Hauser, H. Uptake of cholesterol by small

intestinal brush-border membrane is protein-mediated. Bio-

chemistry 1990, 29, 2142-2148.

(24) Werder, M.; Han, C. H.; Wehrli, E.; Bimmler, D.; Schulthess,

G.; Hauser, H. Role of scavenger receptors SR-BI and CD36 in

selective sterol uptake in the small intestine. Biochemistry 2001,

40, 11643-11650.

(25) (a) Cai, S. F.; Kirby, R. J.; Howles, P. N.; Hui, D. Y. Differentia-

tion-dependent expression and localization of the class B type I

scavenger receptor in intestine. J. Lipid Res. 2001, 42, 902-

909. (b) Cai, L.; Eckhardt, E. R. M.; Shi, W.; Zhao, Z.; Nasser,

M.; de Villiers, W. J. S.; van der Westhuyzen, D. R. Scavenger

receptor class B type I reduces cholesterol absorption in cultured

enterocyte CaCo-2 cells. J. Lipid Res. 2004, 45, 253-262. (c)

Altmann, S. W.; Davis, H. R.; Yao, X. R.; Laverty, M.; Compton,

D. S.; Zhu, L. J.; Crona, J. H.; Caplen, M. A.; Hoos, L. M.;

Tetzloff, G.; Priestley, T.; Burnett, D. A.; Strader, C. D.;

Graziano, M. P. The identification of intestinal scavenger

receptor class B, type I (SR-BI) by expression cloning and its

role in cholesterol absorption. Biochim. Biophys. Acta 2002,

1580, 77-93. (d) Levy, E.; Menard, D.; Suc, I.; Delvin, E.; Marcil,

V.; Brissette, L.; Thibault, L.; Bendayan, M. Ontogeny, immu-

nolocalisation, distribution and function of SR-BI in the human

intestine. J. Cell Sci. 2004, 117, 327-337. (e) Play, B.; Salvini,

S.; Haikal, Z.; Charbonnier, M.; Harbis, A.; Roussel, M.; Lairon,

D.; Jourdheuil-Rahmani, D. Glucose and galactose regulate

intestinal absorption of cholesterol. Biochem. Biophys. Res.

Commun. 2003, 310, 446-451.

(26) van Bennekum, A.; Werder, M.; Thuahnai, S. T.; Han, C.-H.;

Williams, D. L.; Wettstein, P.; Schulthess, G.; Phillips, M. C.;

Hauser, H. Two class B scavenger receptors, SR-BI and CD36,

mediate intestinal absorption of dietary cholesterol and beta-

carotene. Biochemistry 2005, 44, 4517-4525.

(27) Smart, E. J.; De Rose, R. A.; Farber, S. A. Annexin 2-caveolin

1 complex is a target of ezetimibe and regulates intestinal

cholesterol transport. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,

3450-3455.

(28) Altmann, S. W.; Davis, H. R.; Zhu, L. J.; Yao, X. R.; Hoos, L. M.;

Tetzloff, G.; Iyer, S. P. N.; Maguire, M.; Golovko, A.; Zeng, M.;

Wang, L. Q.; Murgolo, N.; Graziano, M. P. Niemann-Pick C1

like 1 protein is critical for intestinal cholesterol absorption.

Science 2004, 303, 1201-1204.

(29) Davis, H. R.; Zhu, L.-j.; Hoos, L. M.; Tetzloff, G.; Maguire, M.;

Liu, J.; Yao, X.; Iyer, S. P. N.; Lam, M.-H.; Lund, E. G.; Detmers,

P. A.; Graziano, M. P.; Altmann, S. W. Niemann-Pick C1 like

1 (NPC1L1) is the intestinal phytosterol and cholesterol trans-

porter and a key modulator of whole body cholesterol homeo-

stasis. J. Biol. Chem. 2004, 279, 33586-33592.

(30) Garcia-Calvo, M.; Lisnock, J. M.; Bull, H. G.; Hawes, B. E.;

Burnett, D. A.; Braun, M. P.; Crona, J. H.; Davis, H. R.; Dean,

D. C.; Detmers, P. A.; Graziano, M. P.; Hughes, M.; MacIntyre,

D. E.; Ogawa, A.; O’Neill, K. A.; Iyer, S. P. N.; Shevell, D. E.;

Smith, M. M.; Tang, Y. S.; Makarewicz, A. M.; Ujjainwalla, F.;

Altmann, S. W.; Chapman, K. T.; Thornberry, N. A. The target

of ezetimibe is Niemann-Pick Cl-like 1 (NPC1L1). Proc. Natl.

Acad. Sci. U.S.A. 2005, 102, 8132-8137.

(31) Davies, J. P.; Scott, C.; Oishi, K.; Liapis, A.; Ioannou, Y. A.

Inactivation of NPC1L1 causes multiple lipid transport defects

and protects against diet-induced hypercholesterolemia. J. Biol.

Chem. 2005, 280, 12710-12720.

(32) (a) Kramer, W.; Glombik, H.; Petry, S.; Heuer, H.; Schafer, H.

L.; Wendler, W.; Corsiero, D.; Girbig, F.; Weyland, C. Identifica-

tion of binding proteins for cholesterol absorption inhibitors as

components of the intestinal cholesterol transporter. FEBS Lett.

2000, 487, 293-297. (b) Frick, W.; Bauer-Schafer, A.; Bauer, J.;

Girbig, F.; Corsiero, D.; Heuer, H.; Kramer, W. Synthesis of a

biotin-tagged photoaffinity probe of 2-azetidinone cholesterol

absorption inhibitors. Bioorg. Med. Chem. 2003, 11, 1639-1642.

(c) Kramer, W.; Girbig, F.; Corsiero, D.; Burger, K.; Fahrenholz,

F.; Jung, C.; Muller, G. Intestinal cholesterol absorption: iden-

tification of different binding proteins for cholesterol and

cholesterol absorption inhibitors in the enterocyte brush border

membrane. Biochim. Biophys. Acta 2003, 1633, 13-26.

(33) Kramer, W.; Girbig, F.; Corsiero, D.; Pfenninger, A.; Frick, W.;

Jahne, G.; Rhein, M.; Wendler, W.; Lottspeich, F.; Hochleitner,

E. O.; Orso, E.; Schmitz, G. Aminopeptidase N (CD13) is a

molecular target of the cholesterol absorption inhibitor ezetimibe

in the enterocyte brush border membrane. J. Biol. Chem. 2005,

280, 1306-1320.

(5) Turley, S. D.; Dietschy, J. M. Sterol absorption by the small

intestine. Curr. Opin. Lipidol. 2003, 14, 233-240.

(6) Earl, J.; Kirkpatrick, P. Ezetimibe. Nat. Rev. Drug Discovery

2003, 2, 97-98.

(7) Maron, D. J.; Fazio, S.; Linton, M. F. Current perspectives on

statins. Circulation 2000, 101, 207-213.

(8) Euroaspire Principal results from Euroaspire II. Eur. Heart J.

2001, 22, 554-572.

(9) (a) Zetia (Ezetimibe). FDA Drug Approvals List [online] http://

www.fda.gov/cder/foi/label/2002/21445lbl.pdf. (b) Knopp, R. H.;

Gitter, H.; Truitt, T.; Bays, H.; Manion, C. V.; Lipka, L. J.;

LeBeaut, A. P.; Suresh, R.; Yang, B.; Veltri, E. P. Effects of

ezetimibe, a new cholesterol absorption inhibitor, on plasma

lipids in patients with primary hypercholesterolemia. Eur. Heart

J. 2003, 24, 729-741. (c) Bays, H. E.; Moore, P. B.; Drehobl, M.

A.; Rosenblatt, S.; Toth, P. D.; Dujovne, C. A.; Knopp, R. H.;

Lipka, L. J.; LeBeaut, A. P.; Yang, B.; Mellars, L. E.; Cuffie-

Jackson, C.; Veltri, E. P. Effectiveness and tolerability of

ezetimibe in patients with primary hypercholesterolemia: Pooled

analysis of two phase II studies. Clin. Ther. 2001, 23, 1209-

1230. (d) Gagne, C.; Bays, H. E.; Weiss, S. R.; Mata, P.; Quinto,

K.; Melino, M.; Cho, M.; Musliner, T. A.; Gumbiner, B. Efficacy

and safety of ezetimibe added to ongoing statin therapy for

treatment of patients with primary hypercholesterolemia. Am.

J. Cardiol. 2002, 90, 1084-1091.

(10) Clader, J. W. The discovery of ezetimibe: A view from outside

the receptor. J. Med. Chem. 2004, 47, 1-9.

(11) Rosenblum, S. B.; Huynh, T.; Afonso, A.; Davis, H. R.; Yumibe,

N.; Clader, J. W.; Burnett, D. A. Discovery of 1-(4-fluorophenyl)-

(3R)-[3-(4-fluorophenyl)-(3S)-hydroxypropyl]-(4S)-(4-hydroxy-

phenyl)-2-azetidinone (SCH 58235): A designed, potent, orally

active inhibitor of cholesterol absorption. J. Med. Chem. 1998,

41, 973-980.

(12) 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.; Schnitzerpolokoff, R.; Smith, A.

A.; Weig, B. C.; Zilli, D. L.; Clader, J. W.; Sybertz, E. J.

Hypocholesterolemic activity of a novel inhibitor of cholesterol

absorption, SCH-48461. Atherosclerosis 1995, 115, 45-63.

(13) Clader, J. W.; Burnett, D. A.; Caplen, M. A.; Domalski, M. S.;

Dugar, S.; Vaccaro, W.; Sher, R.; Browne, M. E.; Zhao, H. R.;

Burrier, R. E.; Salisbury, B.; Davis, H. R. 2-Azetidinone choles-

terol absorption inhibitors: Structure-activity relationships on

the heterocyclic nucleus. J. Med. Chem. 1996, 39, 3684-3693.

(14) Burnett, D. A.; Caplen, M. A.; Davis, H. R.; Burrier, R. E.; Clader,

J. W. 2-Azetidinones as inhibitors of cholesterol absorption. J.

Med. Chem. 1994, 37, 1733-1736.

(15) Dugar, S.; Kirkup, M. P.; Clader, J. W.; Lin, S. I.; Rizvi, R.; Snow,

M. E.; Davis, H. R.; McCombie, S. W. γ-Lactams and related

compounds as cholesterol absorption inhibitors: Homologs of the

â-lactam cholesterol absorption inhibitor SCH 48461. Bioorg.

Med. Chem. Lett. 1995, 5, 2947-2952.

(16) Schnitzerpolokoff, R.; Compton, D.; Boykow, G.; Davis, H.;

Burrier, R. Effects of acyl-CoA-cholesterol O-acyltransferase

inhibition on cholesterol absorption and plasma-lipoprotein

composition in hamsters. Comp. Biochem. Physiol. A 1991, 99,

665-670.

(17) Vaccaro, W. D.; Davis, H. R. Sugar-substituted 2-azetidinone

cholesterol absorption inhibitors: Enhanced potency by modi-

fication of the sugar. Bioorg. Med. Chem. Lett. 1998, 8, 313-

318.

(18) van Heek, M.; France, C. F.; Compton, D. S.; McLeod, R. L.;

Yumibe, N. P.; Alton, K. B.; Sybertz, E. J.; Davis, H. R. In vivo

metabolism-based discovery of a potent cholesterol absorption

inhibitor, SCH58235, in the rat and rhesus monkey through the

identification of the active metabolites of SCH48461. J. Phar-

macol. Exp. Ther. 1997, 283, 157-163.

(19) van Heek, M.; Compton, D. S.; Davis, H. R. The cholesterol

absorption inhibitor, ezetimibe, decreases diet-induced hyperc-

holesterolemia in monkeys. Eur. J. Pharmacol. 2001, 415, 79-

84.

(20) (a) Hernandez, M.; Montenegro, J.; Steiner, M.; Kim, D.;

Sparrow, C.; Detmers, P. A.; Wright, S. D.; Chao, Y. S. Intestinal

absorption of cholesterol is mediated by a saturable, inhibitable

transporter. Biochim. Biophys. Acta Mol. Cell. Biol. Lipids 2000,

1486, 232-242. (b) Detmers, P. A.; Patel, S.; Hernandez, M.;

Montenegro, J.; Lisnock, J.; Pikounis, B.; Steiner, M.; Kim, D.;

Sparrow, C.; Chao, Y. S.; Wright, S. D. A target for cholesterol

absorption inhibitors in the enterocyte brush border membrane.

Biochim. Biophys. Acta 2000, 1486, 243-252.

(21) Jourdheuil-Rahmani, D.; Charbonnier, M.; Domingo, N.; Luc-

cioni, F.; Lafont, H.; Lairon, D. Biliary anionic peptide fraction

and ApoA-I regulate intestinal cholesterol uptake. Biochem.

Biophys. Res. Commun. 2002, 292, 390-395.

(34) (a) Compassi, S.; Werder, M.; Boffelli, D.; Weber, F. E.; Hauser,

H.; Schulthess, G. Cholesteryl ester absorption by small intes-

tinal brush border membrane is protein-mediated. Biochemistry

1995, 34, 16473-16482. (b) Compassi, S.; Werder, M.; Weber,

Inhibitors of Intestinal Cholesterol Absorption

Journal of Medicinal Chemistry, 2005, Vol. 48, No. 19 6053

F. E.; Boffeli, D.; Hauser, H. Comparison of cholesterol and

sitosterol uptake in different brush border membrane models.

Biochemistry 1997, 36, 6643-6652. (c) Schulthess, G.; Compassi,

S.; Boffelli, D.; Werder, M.; Weber, F. E.; Hauser, H. A compara-

tive study of sterol absorption in different small-intestinal brush

border membrane models. J. Lipid Res. 1996, 37, 2405-2419.

(d) Werder, M.; Hauser, H.; Abele, S.; Seebach, D. â-Peptides as

inhibitors of small-intestinal cholesterol and fat absorption. Helv.

Chim. Acta 1999, 82, 1774-1783.

(57) Liptak, A.; Balla, E.; Janossy, L.; Sajtos, F.; Szilagyi, L. The first

synthesis of secondary sugar sulfonic acids by nucleophilic

displacement reactions. Tetrahedron Lett. 2004, 45, 839-842.

(58) Refer to ref 36 for a detailed description of the development of

the brush border membrane vesicle assay. Most importantly,

brush border membrane vesicles can inherently only mimic the

first uptake step, namely that of cholesterol into the brush

border membrane, of the overall multistep process of cholesterol

absorption. In summary, the small-molecule inhibitors were

incorporated into phospholipid unilamellar donor vesicles to-

gether with cholesteryl oleyl ester and trace amounts of radio-

(35) Schulthess, G.; Compassi, S.; Werder, M.; Han, C. H.; Phillips,

M. C.; Hauser, H. Intestinal sterol absorption mediated by

scavenger receptors is competitively inhibited by amphipathic

peptides and proteins. Biochemistry 2000, 39, 12623-12631.

(36) Kværnø, L.; Ritter, T.; Werder, M.; Hauser, H.; Carreira, E. M.

An in vitro assay for evaluation of small-molecule inhibitors of

cholesterol absorption. Angew. Chem., Int. Ed. 2004, 43, 4653-

4656.

labeled cholesteryl oleyl ether. The latter was used as

a

nonhydrolyzable analogue of cholesteryl oleyl ester since both

have been shown to exhibit identical brush border membrane

uptake kinetics (ref 34a). Cholesteryl ester instead of free

cholesterol was chosen for standard inhibition experiments, since

passive diffusion is negligible for cholesteryl ester (ref 22).

Cholesteryl ester is absorbed into brush border membrane

vesicles with essentially the same uptake kinetics as free

cholesterol (ref 34a) and into both SR-BI and CD36 transfected

COS-7 cells (ref 26).

(37) Kværnø, L.; Werder, M.; Hauser, H.; Carreira, E. M. Carbohy-

drate sulfonyl chlorides for the simple, convenient access to

glycoconjugates. Org. Lett. 2005, 7, 1145-1148.

(38) Hauser, H.; Howell, K.; Dawson, R. M. C.; Bowyer, D. E. Rabbit

small intestinal brush-border membrane preparation and lipid-

composition. Biochim. Biophys. Acta 1980, 602, 567-577.

(39) van Heek, M.; Farley, C.; Compton, D. S.; Hoos, L.; Alton, K.

B.; Sybertz, E. J.; Davis, H. R. Comparison of the activity and

dispositionofthenovelcholesterolabsorptioninhibitor,SCH58235,

and its glucuronide, SCH60663. Br. J. Pharmacol. 2000, 129,

1748-1754.

(40) Similar C-linked glycosides of ezetimibe have been previously

reported, see: Tomiyama, H.; Yokota, M.; Noda, A.; Ohno, A.

Preparation of â-lactam compounds as serum cholesterol-lower-

ing agents. WO2002066464, 113 pp.

(41) For a synthesis of a related C-linked glycoside of SCH48461,

see: Vaccaro, W. D.; Sher, R.; Davis, H. R. Sugar-substituted

2-azetidinones as cholesterol absorption inhibitors. Bioorg. Med.

Chem. Lett. 1998, 8, 35-40.

(42) Wu, G. Z.; Wong, Y. S.; Chen, X.; Ding, Z. X. A novel one-step

diastereo- and enantioselective formation of trans-azetidinones

and its application to the total synthesis of cholesterol absorption

inhibitors. J. Org. Chem. 1999, 64, 3714-3718.

(43) Srivastava, V.; Tandon, A.; Ray, S. Convenient and selective

acetylations of phenols, amines and alcohols. Synth. Commun.

1992, 22, 2703-2710.

(44) Varma, R. S.; Varma, M.; Chatterjee, A. K. Microwave-assisted

deacetylation on aluminasA simple deprotection methodology.

J. Chem. Soc., Perkin Trans. 1 1993, 999-1000.

(45) Herzig, J.; Nudelman, A.; Gottlieb, H. E.; Fischer, B. Studies in

sugar chemistry. 2. A simple method For O-deacylation of

polyacylated sugars. J. Org. Chem. 1986, 51, 727-730.

(46) Greene, T. W.; Wuts, P. G. Protective Groups in Organic

Synthesis, 3rd ed.; John Wiley & Sons Inc.: New York, 1999.

(47) Tsunoda, T.; Yamamiya, Y.; Ito, S. 1,1′-(Azodicarbonyl)dipiperi-

dine-tributylphosphine, a new reagent system for Mitsunobu

reaction. Tetrahedron Lett. 1993, 34, 1639-1642.

(59) Ezetimibe has been shown to be extensively metabolized into

its glucuronide 20 while hydrolytic opening of the â-lactam ring

is not a major metabolic pathway; see ref 39 and the following:

Kosoglou, T.; Statkevich, P.; Johnson-Levonas, A. O.; Paolini,

J. F.; Bergman, A. J.; Alton, K. B. EzetimibesA review of its

metabolism, pharmacokinetics and drug interactions. Clin.

Pharmacokinet. 2005, 44, 467-494.

(60) Ab initio calculations: Initial energy-minimization by molecular

mechanics using MacroModel v. 7.0 [pseudosystematical Monte

Carlo conformational search (100 steps, limit 30 kJ/mol) with

CHC1₃as solvent, MMFFs force field] was followed by a

quantum-mechanical geometry optimization using Jaguar v. 4.2

(B3LYP/6-31G*) in Maestro v. 4.1.

(61) (a) Yamashita, M.; Ojima, I. Effective Route to Azetidines From

Azetidin-2-Ones Using Hydroalanes As Specific Reducing Agents.

J. Am. Chem. Soc. 1983, 105, 6339-6342. (b) Ojima, I.; Zhao,

M. Z.; Yamato, T.; Nakahashi, K.; Yamashita, M.; Abe, R.

Azetidines and bisazetidinessTheir synthesis and use as the key

intermediates to enantiomerically pure diamines, amino-alco-

hols, and polyamines. J. Org. Chem. 1991, 56, 5263-5277. (c)

Ojima, I.; Yamato, T.; Nakahashi, K. Synthesis of novel chiral

disazetidines by the hydroalane reduction of bis-beta-lactams.

Tetrahedron Lett. 1985, 26, 2035-2038.

(62) Browne, M.; Burnett, D. A.; Caplen, M. A.; Chen, L. Y.; Clader,

J. W.; Domalski, M.; Dugar, S.; Pushpavanam, P.; Sher, R.;

Vaccaro, W.; Viziano, M.; Zhao, H. R. trans-Diastereoselective

synthesis of 3-alkyl substituted â-lactams via the acid chloride-

imine reaction of nonactivated acid-chlorides. Tetrahedron Lett.

1995, 36, 2555-2558.

(63) Ritter, T.; Stanek, K.; Larrosa, I.; Carreira, E. M. Mild cleavage

of aryl mesylates: Methanesulfonate as potent protecting group

for phenols. Org. Lett. 2004, 6, 1513-1514.

(64) In preliminary animal experiments using ezetimibe (1) as a

reference compound, essentially carried out as described in ref

26, we have established that inhibitor 18 is indeed an inhibitor

of cholesterol absorption in C57BL/6 mice at a dose of 10 mg/

kg/day (70% of the inhibition obtained for ezetimibe).

(65) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;

Timmers, F. J. Safe and convenient procedure for solvent

purification. Organometallics 1996, 15, 1518-1520.

(48) Jaramillo, C.; Chiara, J. L.; Martinlomas, M. An effective

strategy for the synthesis of 6-O-(2-amino-2-deoxy-alpha-D-

glucopyranosyl)-D-chiro and D-myo-inositol 1-phosphate related

to putative insulin mimetics. J. Org. Chem. 1994, 59, 3135-

3141.

(49) RajanBabu, T. V.; Reddy, G. S. 1-Methylene sugars as C-

glycoside precursors. J. Org. Chem. 1986, 51, 5458-5461.

(50) Spak, S. J.; Martin, O. R. The synthesis of an O,C-trisaccha-

ride: The O,C-analog of methyl 4′-O-â-D-glucopyranosylgentio-

bioside. Tetrahedron 2000, 56, 217-224.

(51) (a) Glycoscience; Fraser-Reid, B., Tatsuta, K., Thiem, J., Eds.;

Springer-Verlag: Berlin, Heidelberg, New York, 2001. (b) Bioor-

ganic Chemistry: Carbohydrates; Hecht, S. M., Ed.; Oxford

University Press: New York, Oxford, 1999.

(52) (a) Johnson, D. C.; Widlanski, T. S. Multiple routes for the

synthesis of N-acyl sulfonamide-bridged nucleosides. Synthesis

2002, 809-815. (b) Huang, J. X.; McElroy, E. B.; Widlanski, T.

S. Synthesis of Sulfonate-Linked DNA. J. Org. Chem. 1994, 59,

3520-3521.

(53) Huang, J. X.; Widlanski, T. S. Facile synthesis of sulfonyl

chlorides. Tetrahedron Lett. 1992, 33, 2657-2660.

(54) Although there is literature precedence for the synthesis of a

sugar sulfonyl chloride, its use was limited to the preparation

of simple esters (from 2-propanol, tert-butyl alcohol, methanol)

and amides (from N-butylamine, and an amino sugar); see:

Ulgar, V.; Maya, I.; Fuentes, J.; Fernandez-Bolanos, J. G.

Tetrahedron 2002, 2058, 7967-7973.

(66) Pedersen, D. S.; Rosenbohm, C. Dry column vacuum chroma-

tography. Synthesis 2001, 2431-2434.

(67) Bernet, B.; Vasella, A. Carbocyclic compounds from monosac-

charides. 1. Transformations in the glucose series. Helv. Chim.

Acta 1979, 62, 1990-2016.

(68) Griffin, F. K.; Paterson, D. E.; Taylor, R. J. K. Ramberg-Ba¨cklund

approaches to the synthesis of C-linked disaccharides. Angew.

Chem., Int. Ed. 1999, 38, 2939-2942.

(69) Gervay, J.; Flaherty, T. M.; Holmes, D. Studies on the synthesis

of C-glycoside sulfones as potential glycosyl transferase inhibi-

tors. Tetrahedron 1997, 53, 16355-16364.

(70) (a) Mu¨tsch, B.; Gains, N.; Hauser, H. Interaction of intestinal

brush-border membrane-vesicles with small unilamellar phos-

pholipid-vesiclessExchange of lipids between membranes is

mediated by collisional contact. Biochemistry 1986, 25, 2134-

2140. (b) Brunner, J.; Hauser, H.; Semenza, G. Single bilayer

lipid-protein vesicles formed from phosphatidylcholine and

small intestinal sucrase-isomaltase. J. Biol. Chem. 1978, 253,

7538-7546.

(71) Schurtenberger, P.; Hauser, H. Characterization of the size

distribution of unilamellar vesicles by gel-filtration, quasi-elastic

light-scattering and electron-microscopy. Biochim. Biophys. Acta

1984, 778, 470-480.

(55) Paterson, D. E.; Griffin, F. K.; Alcaraz, M. L.; Taylor, R. J. K. A

Ramberg-Ba¨cklund approach to the synthesis of C-glycosides,

C-linked disaccharides, and C-glycosyl amino acids. Eur. J. Org.

Chem. 2002, 1323-1336.

(56) Reddie, R. N. Peroxomonosulfate oxidation of acetylthio com-

pounds to sulfonates. Synth. Commun. 1987, 17, 1129-1139.

JM050422P

Article Doi

Synthesis and in vitro evaluation of inhibitors of intestinal cholesterol absorption

DOI: 10.1021/jm050422p

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm050422p

Full text of DOI:10.1021/jm050422p

Products guided by the article

R&D Labs maybe for 866918-13-0

Relevant to this article

Hot Product