Novel amino-β-lactam derivatives as potent cholesterol absorption inhibitors

Article abstract of DOI:10.1016/j.ejmech.2014.10.014

Two new trans-(3R,4R)-amino-β-lactam derivatives and their diastereoisomeric mixtures were synthesized as ezetimibe bioisosteres and tested in in vitro and in vivo experiments as novel β-lactam cholesterol absorption inhibitors. Both compounds exhibited l

Full text of DOI:10.1016/j.ejmech.2014.10.014

European Journal of Medicinal Chemistry 87 (2014) 722e734
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Novel amino-
inhibitors
b
-lactam derivatives as potent cholesterol absorption
,
*
a
b
a
ꢀ ꢁ a
ꢀ
ꢀ
Tonko Drazic , Kresimir Molcanov , Vinay Sachdev , Martina Malnar ,
Silva Hecimovic , Jay V. Patankar 1, Sascha Obrowsky b, Sanja Levak-Frank b,
b,
ꢁ
ꢁ a
ꢀ a
b, *
Ivan Habus , Dagmar Kratky
a
ꢀ
ꢁ
ꢀ
RuCer Boskovic Institute, Bijenicka c. 54, HR-10002 Zagreb, Croatia
b Institute of Molecular Biology and Biochemistry, Medical University of Graz, Graz, Austria
a r t i c l e i n f o
a b s t r a c t
Article history:
Two new trans-(3R,4R)-amino-
b-lactam derivatives and their diastereoisomeric mixtures were synthe-
Received 13 February 2014
Received in revised form
2 October 2014
Accepted 6 October 2014
Available online 7 October 2014
sized as ezetimibe bioisosteres and tested in in vitro and in vivo experiments as novel
b
-lactam
cholesterol absorption inhibitors. Both compounds exhibited low cytotoxicity in MDCKII, hNPC1L1/
MDCKII, and HepG2 cell lines and potent inhibitory effect in hNPC1L1/MDCKII cells. In addition, these
compounds markedly reduced cholesterol absorption in mice, resulting in reduced cholesterol concen-
trations in plasma, liver, and intestine. We determined the crystal structure of one amino-b-lactam
derivative to establish unambiguously both the absolute and relative configuration at the new stereo-
genic centre C17, which was assigned to be S. The pKa values for both compounds are 9.35, implying that
the amino-b-lactam derivatives and their diastereoisomeric mixtures are in form of ammonium salt in
Keywords:
b-lactam
Cholesterol absorption inhibitor
Hyperlipidemia
Cardiovascular heart disease
blood and the intestine. The IC50 value for the diastereoisomeric mixture is 60
inhibited cholesterol absorption comparable to ezetimibe.
mM. In vivo, it efficiently
© 2014 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/3.0/).
1. Introduction
approved in 2002) (Fig. 1) is the only cholesterol absorption in-
hibitor on the market today [5]. It can be applied either as a
Cardiovascular heart disease (CHD) is the leading cause of death
worldwide [1]. One of the major risk factors for CHD are elevated
serum cholesterol concentrations [2]. Lowering the level of serum
cholesterol is an established clinical practice for CHD treatment,
intervention, and prevention. Pharmacologically, circulating
cholesterol concentrations are reduced by statins, which are 3-
hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase in-
hibitors, affecting biosynthesis of endogenous cholesterol [3,4].
Another approach is to block the absorption of dietary cholesterol,
which is the other major contributor to serum cholesterol con-
centrations in the small intestine. Ezetimibe 1 (Zetia, Ezetrol;
monotherapy or in combination with statins [6]. Ezetimibe 1 was
originally discovered without a known molecular target through
in vivo screening of cholesterol-fed hamsters [7]. In 2004, re-
searchers from the Schering-Plough Research Institute identified
Niemann Pick C1-like1 (NPC1L1) protein as a molecular target of
ezetimibe 1 [8]. Ezetimibe 1 acts by blocking the internalization of
NPC1L1, thereby preventing cholesterol from entering the cyto-
plasm of enterocytes [9].
A thorough structureeactivity relationship (SAR) study [5] of
the
hamsters revealed that the 2-azetidinone backbone as well as the
aryl group at the N-1 and C-4 position of the -lactam ring are
required for activity. The aryl group at the C-4 position of the
lactam ring is optimally substituted with alkoxy or hydroxy groups
at the para-position. The side chain at the C-3 position of the
lactam ring with three linking atoms bearing a pendent aryl group
is optimal. Preferred configuration at the C-4 chiral center of the
b-lactam cholesterol absorption inhibitors in cholesterol-fed
b
b
-
Abbreviations: TBDMSCl, tert-butyldimethylsilyl chloride; CBS, Cor-
eyeBakshieShibata catalyst; CHD, cardiovascular heart disease; TEAA, tetraethy-
lammonium acetate; FCS, fetal calf serum; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide; NPC1L1, Niemann-Pick C1-like protein 1; THAP,
2,4,6-trihydroxyacetophenone; DAC, ammonium citrate dibasic.
* Corresponding authors.
b
-
b
-
lactam ring is S and the C-3 atom tolerates S or R configurations [5].
Introduction of a heteroatom at the 10-position of the C-3 side chain
can also contribute to the activity, whereas isosteric groups at the
30-position of the side chain decrease the activity [10].
ꢀ ꢁ
E-mail addresses: tdrazic@irb.hr (T. Drazic), dagmar.kratky@medunigraz.at
(D. Kratky).
1
Present address: Centre for Molecular Medicine and Therapeutics, Department
of Medical Genetics, University of British Columbia, Vancouver, Canada.
http://dx.doi.org/10.1016/j.ejmech.2014.10.014
0223-5234/© 2014 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/3.0/).

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
723
6/5 from trans-amino-
-lactam with commercially available 2-bromo-1-(4-
fluorophenyl)ethan-1-one 3 followed by stereoselective reduction
of the side chain keto group at the C-3 of the -lactam ring with
b-lactam 2: (i) N-alkylation of trans-amino-
b
2
b
CBS-catalyst or (ii) stereoselective reduction of the keto group of 3
preceding N-alkylation reaction. In the present study we applied
both approaches.
(i) N-alkylation of trans-amino-
subsequent reduction of the side chain keto group. N-alkyl-
ation of trans-amino- -lactam 2 was performed in mild
b-lactam 2 with ketone 3 and
b
conditions using NaI for in situ generation of 2-iodo-deriva-
tive of 2-bromo-1-(4-fluorophenyl)ethan-1-one 3 in the
presence of Et3N at room temperature and provided 4 in a
moderate yield (46%) (Scheme 1A). A mixture of THF and
DMF in ratio 9:1 was found optimal for the reaction. The C-3
side chain hydroxy group was obtained by stereoselective
reduction of the keto group with CBS catalyst (Scheme 1B)
[19e21]. However, addition of CBS-catalyst (0.1 eq.) and
BH3$Me2S (1 eq.) provided the diastereoisomeric mixture 5/6
(50:50), determined by 1H NMR. BH3$Me2S reduction of
Fig. 1. Structure of ezetimibe (30S,3R,4S)-1.
In continuation of our research in the field of b-lactam chem-
istry [11e15] and taking into consideration the requirements
determined by SAR studies [5], we synthesized bioisosteres 5 and 6
(Fig. 2) of ezetimibe 1 bearing a eNHe group at the C-3 position of
the
b-lactam ring.
Bioisosterism is a useful approach for lead compound modifi-
amino-
stereoselectivity, providing mixture 5/6 (50:50). The absence
of stereoselectivity in the reduction of amino- -lactam ke-
b-lactam ketone 4 (1:1 eq. ratio) had no effect on the
cation that can result in improved pharmacological activity,
decreased toxicity, and optimized pharmacokinetics. With the
classical bioisosteric exchange of the eCH2e with a eNHe group
we aimed at investigating whether the change in polarity of the
side chain, the ability of additional H-bond, and ammonium salt
formations would affect their cholesterol absorption inhibition and
cytotoxicity. Here we show that our new ezetimibe analogs 5, 6 and
their diastereoisomeric mixture 5/6 (70:30) are potent novel
cholesterol absorption inhibitors.
b
tone 4 was probably due to nitrogen proximity to the keto
group and the ability of borane to form a complex with
amine, which allowed a direct hydrogen delivery to the keto
group without participation of a chiral catalyst [22,23].
Addition of BH3$Me2S (2 eq.) to CBS-catalyst (0.1 eq.) did not
result in improvement of stereoselectivity in keto group
reduction of 4 either. Improvement of stereoselectivity was
accomplished with the complex formation between CBS-
catalyst and BH3$Me2S, (1:1 eq. ratio), followed by drop-
wise addition of 4. Reaction with (R)-CBS-catalyst provided a
diastereoisomeric mixture of amino alcohols 5/6 (70:30)
at ꢀ20 ꢁC, whereas (S)-CBS-catalyst afforded 6/5 (85:15)
(Fig. 4). Lowering the temperature to ꢀ80 ꢁC did not improve
the stereoselectivity of the reduction. Recrystallization of 6/5
(85:15) provided pure amino alcohol 6.
2. Results and discussion
2.1. Chemistry
2.1.1. Synthesis
We synthesized two novel ezetimibe analogs 5 and 6 (Fig. 2) and
their diastereoisomeric mixtures 5/6 (70:30) and 6/5 (85:15) from
enantiomerically pure trans-(3R,4R)-amino-b-lactam 2 (Fig. 3) and
determined their in vitro and in vivo activities as cholesterol ab-
sorption inhibitors. We extensively studied the stereoselectivity of
the side chain keto group reduction with CBS-catalyst in proximity
of the eNHe group at the C-3 position of 2-azetidinone. Enantio-
(ii) Stereoselective reduction of ketone
alkylation.
3 followed by N-
Commercially available ketone 3 was reduced following the
original protocol for CBS-reduction developed by Corey et al. and
yielded alcohols 7a and 7b in >99% ee (Scheme 2) [19]. Protection of
the OH group in 7a,b with TBDMSCl was carried out in DMF in the
presence of imidazole [24] to afford OTBDMS bromo derivatives 8a
and 8b. Exchange of bromine in 8a,b with iodine (in the presence of
NaI in acetone at 55 ꢁC) yielded OTBDMS iodo derivatives 9a and
9b. Iodo-bromo exchange proceeded very slowly, providing the
merically pure trans-(3R,4R)-amino-b-lactam 2 (Fig. 3) was syn-
thesized applying the chiral ester enolate-imine condensation
[16,17]. Enantiomeric purity of 2 (>99% ee) was determined by
the Mosher's MTPA method [18] on 19F NMR (ꢀ67.79 ppm, s, 3F, CF3
and -116.22 ppm, s, 1F, NeC6H4F).
There are two possible approaches for the synthesis of ezeti-
mibe analogs 5 and 6 and their diasteroisomeric mixtures 5/6 and
Fig. 2. Structure of novel amino bioisosteres (2R,3R,4R)-5 and (2S,3R,4R)-6.

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
724
mixtures of 9a or 9b and unreacted 8a or 8b, respectively, in ratio
93:7 and 96% yield after 4 days. This mixture was used in the N-
alkylation reaction of amino- -lactam 2 without further purifica-
b
tion. The reaction proceeded for 7 days in CH3CN to afford silyl
intermediates 10a and 10b with a moderate yield (20%). Silyl in-
termediates 10a,b were further deprotected with 3% HCl in ethanol
to produce ezetimibe bioisosteres 5 and 6 (Scheme 2).
2.1.2. Molecular and crystal structure
The crystal structure of (3R,4R)-3-[(2S)-2-(4-fluorophenyl)-2-
hydroxyethylamino]-1-(4-fluorophenyl)-4-(4-hydroxyphenyl)aze-
tidin-2-one (2S,3R,4R)-6 was determined to establish unambigu-
ously both absolute and relative configurations at the stereogenic
centres C17 [25] and N2.
Two symmetry independent molecules of 6, related by a
pseudo-twofold rotation axis, are present in the crystal structure
designated as 6A and 6B (Fig. 5); they are homochiral and of similar
conformations (Fig. 6). Configuration at the stereogenic centre C17
Fig. 3. trans-(3R,4R)-amino-b-lactam 2.
Scheme 1. (A) N-alkylation of trans-amino-b-lactam (3R,4R)-2 and (B) subsequent reduction of the keto group in (3R,4R)-4 with CBS-catalysts.

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
725
Fig. 4. 1H NMR spectra of the (A) C-4
b-lactam (300 MHz, CDCl3) and (B) side chain C-2 OH group (300 MHz, DMSO-d6) proton signals for 5/6 (70:30) and (C) C4
b-lactam (600 MHz,
CDCl3) and (D) side chain C-2 OH group (300 MHz, DMSO-d6) well-resolved proton signals for 6/5 (85:15).
was assigned to be S in both 6A and 6B, according to the known R-
configurations at the stereogenic centres C2 and C3.
LC50 values were determined in MDCKII wild type, hNPC1L1/
MDCKII, and HepG2 cells (Table 2). MDCKII cells stably expressing
Molecules 6A and 6B are linked bya pairof NeH/O]C hydrogen
bonds into C2-symmetric dimers (Fig. 7, Table 1). Each molecule has
two hydroxy groups, which act as proton donors toward symmetry-
equivalent molecules; one toward a carbonyl and one toward
another hydroxy group, giving a total of four OeH$$$O hydrogen
bonds (Table 1) that link the dimers into layers parallel to (011).
There are only dispersion interactions between the layers (Fig. 8).
human Niemann-Pick C1-like protein 1 (hNPC1L1/MDCKII)
(Figure S1) are a pharmacologically validated system for investi-
gating NPC1L1-mediated cholesterol uptake [27]. The LC50 values
were higher than 100
lines. LC50 values for ezetimibe 1 were 62.29
MDCKII and 69.74 M in HepG2 cells. In MDCKII wild type cells,
mM and considered nontoxic in all three cell
mM in hNPC1L1/
m
ezetimibe 1 showed no toxicity (Table 2). In addition, we tested the
in vitro cytotoxicity of ezetimibe 1 and compounds 5, 6, and 5/6
(70:30) in combination with micelles and found all LC50 values to
2.1.3. pKa values
be above 100 mM (Figure S2).
We determined pKa values of the amino alcohols 5 (Fig. 9) and 6
(data not shown) using spectrophotometric titration. We measured
an increase of the absorbance intensity at l 247 nm by the addition
2.2.2. In vitro activity
of NaOH in the pH range 6e12 (Fig. 9A). pKa values calculated from
the obtained sigmoidal curves for amino alcohol 5 (Fig. 9B) revealed
that the pKa value was 9.35. We got the same pKa value for amino
alcohol 6 (data not shown), indicating that both compounds are
present in the form of NHþ4 in the blood and small intestine.
First we tested ezetimibe analogs 5 and 6 and their diastereo-
isomeric mixtures 5/6 (70:30) and 6/5 (85:15) for their ability to
inhibit cholesterol uptake. In MDCKII wildtype cells, ezetimibe 1
had no effect, but inhibited cholesterol uptake in hNPC1L1/MDCKII
cells (Figure S3). When hNPC1L1/MDCKII were treated with 5
(Fig. 10A), 6 (Fig. 10B), and their diastereoisomeric mixtures 5/6
(Fig. 10C) and 6/5 (Fig. 10D), we observed 50e55% inhibition of
2.2. Biochemical section
cholesterol uptake, reaching its maximum at 120
without a significant difference between the compounds. IC50
values were in the range of 60e80 M (Fig. 10). These results show
that the novel ezetimibe bioisosteres 5, and their
mM concentration
2.2.1. Cytotoxicity measurement
Cytotoxicity of amino alcohols 5, 6 and their diastereoisomeric
mixture 5/6 was analyzed using MTTcell proliferation assay and the
m
6

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
726
Scheme 2. N-alkylation of trans-amino-b-lactam (3R,4R)-2 with (1R)-9a and (1S)-9b for the preparation of amino alcohols (2R,3R,4R)-5 and (2S,3R,4R)-6.
diastereoisomeric mixtures 5/6 and 6/5 are potent inhibitors of
cholesterol uptake in vitro.
relative to vehicle, respectively. Both 5 and 6 showed significant
inhibition of cholesterol absorption (Table 3). Treatment of mice
with either 5 and 6 resulted in reductions of [3H]cholesterol in
plasma by 50% and 32%, respectively. Radioactivities in the liver
were decreased by 44% and 47%, respectively, and were comparable
to ezetimibe 1. Amino alcohol 6 markedly lowered [3H]cholesterol
in the small intestine with highest inhibition in the ileum by
58e60%, whereas amino alcohol 5 reduced radioactivity in the
small intestine by ~22% without reaching statistical significance in
either part of the intestine. We therefore increased the dose to
2.2.3. In vivo activity
In vivo, we first determined the consequences of ezetimibe 1 and
amino alcohols 5 and 6 on cholesterol absorption. We therefore
gavaged mice with 2 m
Ci [3H]cholesterol and 10 mg/kg/day of each
compound or vehicle, and radioactivity was measured in plasma,
liver, and three equal parts of the intestine (duodenum, jejunum,
ileum). Percentage of [3H]cholesterol reduction is presented
Fig. 5. ORTEP-3 [26] drawing of 6A and 6B symmetry-indepentent molecules of (2S,3R,4R)-6. Displacement ellipsoids are drawn for the probability of 50% and hydrogen atoms are
shown as spheres of arbitrary radii.

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
727
Table 1
Geometric parameters of the hydrogen bonds in (2S,3R,4R)-6.
DeH/Å H/A/Å D/A/Å
DeH/A/ꢁ Symm. op. on A
0.86(3) 2.11(3) 2.962(7) 172(4)
N2AeH2C/O1B
N2BeH2F/O1A
O2AeH2B/O3A
O2BeH2E/O3B
O3AeH3B/O1A
O3BeH3D/O1B
C2AeH2A/O2A
C2BeH2B/O2B
1þx, y, z
0.90
0.82
0.82
0.82
0.82
0.98
0.98
2.16
1.91
1.91
2.75
2.57
2.42
2.48
2.35
3.019(6) 159
2.692(7) 159
2.702(6) 162
3.334(6) 129
3.185(6) 133
3.365(8) 161
3.388(7) 155
3.280(13) 173
ꢀ1þx, y, z
2ex, 1/2 þ y, 1ez
ex, 1/2 þ y, ez
x, ꢀ1þy, z
x, 1 þ y, z
2ex, 1/2 þ y, 1ez
ex, 1/2 þ y, ez
ꢀ1þx, y, z
C12AeH12A/F2A 0.93
assigned to be S. Both diastereoisomeres 5 and 6 as well as their
diastereoisomeric mixture 5/6 showed significant cholesterol ab-
sorption inhibitory activity both in vitro and in vivo. Based on our
data and the pKa value for 5 and 6 being 9.35, indicating that both
compounds are present in the form of NHþ4 in the blood and small
intestine, other diastereoisomeric mixtures (e.g. 6/5 (85:15)) may
exhibit similar in vivo results as 5/6 (70:30). Results from this study
implicate a therapeutic potential of these novel compounds to
reduce cholesterol plasma concentrations and improve CHD.
Fig. 6. Overlay of molecules 6A (black) and 6B (grey).
20 mg/kg/day and determined inhibition of cholesterol absorption.
The inhibition in duodenum and jejunum was similar to the lower
dose, whereas in ileum it raised from 21% to 41% (Table 3).
Decreased cholesterol absorption of 5 in the small intestine
compared to ezetimibe 1 might be explained by the fact that 2-
azetidinones are moderate acyl-coenzyme A:cholesterol acyl-
transferase inhibitors and their level of activity is highly structure-
dependent [28]. It might therefore be speculated that acyl-
coenzyme A:cholesterol acyltransferase activity of amino alcohols
5 and 6 and their diastereoisomeric mixtures is altered compared to
ezetimibe 1. The effect of the diastereoisomeric mixture 5/6 (70:30)
on cholesterol absorption was comparable to ezetimibe 1 in plasma
and ileum, lower in duodenum and jejunum, and higher in the liver
(Table 3).
4. Experimental protocols
4.1. Materials and methods
All commercial reagent grade chemicals and solvents were used
without further treatment. Melting points were determined on a
Reichert Thermovar 7905 apparatus and are not corrected. The IR
spectra were recorded on a PerkinElmer Spectrum RX I FT-IR Sys-
tem spectrometer (KBr pellets technique) (PerkinElmer In-
struments, Norwalk, CT, USA). The 1H and 13C NMR spectra (in
CDCl3 and DMSO-d6 at RT) were measured on a Bruker AV 300 and/
or AV 600 spectrometer (Bruker BioSpin GmbH., Rheinstetten,
3. Conclusion
This report demonstrates that we have successfully synthesized
two novel ezetimibe bioisosteres 5, 6 and their diastereoisomeric
mixtures 5/6 (70:30) and 6/5 (85:15) from enantiomerically pure
trans-(3R,4R)-amino-b-lactam 2. Crystal structure of 6 was deter-
mined to establish unambiguously both absolute and relative
Germany),
d is given in ppm relative to tetramethylsilane as an
internal reference. Microanalysis was performed on a PE 2400 Se-
ries II CHNS/O Analyzer (PerkinElmer Instruments, Shelton, CT,
USA). Optical rotations: Optical Activity Automatic Polarimeter AA-
10 in 1 dm cell; c in g/100 mL (Optical Activity Ltd., Ramsey, En-
gland). Column chromatography on silica gel, 70e230 mesh, 60 Å
(SigmaeAldrich, St. Louis, MO, USA or Acros-Organics, New Jersey,
USA) was performed at RT. Thin layer chromatography was carried
out on TLC aluminium sheets, 20 ꢂ 20 cm, silica gel 60 F254 (Merck,
Darmstadt, Germany). RP-HPLC analysis was performed on HPLC
system (Knauer GmbH., Berlin, Germany) supplied with UV/VIS
WellChrom Diode Array Detector K-2800. UV/VIS measurements
were performed on a T80þ spectrophotometer (PG Instruments
Limited, England). Samples for HR-MS analysis were resuspended
configurations at the new stereogenic centre C17 and were
in 5 ml of THAP/DAC matrix and 1 ml was spotted onto a MALDI plate.
Mass spectra were obtained on a matrix-assisted laser desorption/
ionization-time-of-flight MALDI-TOF/TOF mass spectrometer
(4800 Plus MALDI-TOF/TOF Analyzer, Applied Biosystems, Foster
City, CA, USA) equipped with Nd:YAG laser operating at 355 nm
with firing rate 200 Hz in the positive ion reflector mode. 1600
shots per spectrum were taken covering mass range 100e1000 Da,
focus mass 500 Da and delay time 100 ns.
4.2. Crystal structure determination
Crystals of amino alcohol (2S,3R,4R)-6 suitable for data collec-
tion were grown from ethyl acetate and hexane by vapour diffusion.
Fig. 7. The two independent molecules 6A and 6B are connected by hydrogen bonds
into a dimer of pseudo-C2 symmetry. The pseudo-C2 axis is approximately parallel to
the crystallographic axis a, marked by an arrow ([). Symmetry operator i) 1 þ x, y, z.
The selected crystal was
a
needle with dimensions
0.10 ꢂ 0.03 ꢂ 0.02 mm3.

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
728
Fig. 8. Hydrogen-bonded layer in (2S,3R,4R)-6. Molecules 6A (red) and 6B (green). Position of the pseudo-C2 axis has been indicated by the black oval. Approximate position of the
pseudo axis is y z 0.20; z z 0.26.
Fig. 9. (A) UV spectra of 5 at pH 6.6 and pH 11 and (B) absorbance of 5 at
l 247 nm and pH 6e12.
Single crystal measurement was performed on an Oxford
Diffraction Xcalibur Nova R (microfocus Cu tube, Oxford Diffraction,
U.K.) at room temperature [293(2) K]. Program package CrysAlis
PRO [29] was used for data reduction. The structure was solved
refined with SHELXS97 [30]. The model was refined using the full-
matrix least squares refinement; all non-hydrogen atoms were
refined anisotropically. Hydrogen atoms bound to C and O were
modelled as riding entities using the AFIX command. Hydrogen
atoms bound to N2A and N2B could not be located from the elec-
tron density map due to the data poor quality. However, it can be
assumed that they are directed towards the nearest proton
acceptor: the carbonyl oxygen of the neighbouring molecule.
Therefore, they were generated at the appropriate positions and
refined using appropriate restraints (Table 1).
The structure comprises relatively large cavities filled by disor-
dered water molecules. Due to a very low electron density and low
data quality (due to small size of the crystal) the disordered water
was tentatively modelled as four isotropic oxygen atoms (hydro-
gens could not have been located) with occupancy of 0.5.
Molecular geometry calculations were performed by PLATON
[31], and molecular graphics were prepared using ORTEP-3 [26],
and CCDC-Mercury [32]. Crystallographic and refinement data for
amino alcohol (2S,3R,4R)-6 reported in this paper are shown in
Table 4.
Supplementary crystallographic data for this paper can be ob-
tained free of charge via www.ccdc.cam.ac.uk/conts/retrieving.
html (or from the Cambridge Crystallographic Data Centre, 12,
Union Road, Cambridge CB2 1EZ, UK; fax: (þ)44 1223 336033; or
deposit@ccdc.cam.ac.uk). CCDC-979536 contains the supplemen-
tary crystallographic data for this paper.
Configuration of novel stereogenic centre C17 was established
relative to the configurations of the known stereogenic centres C2
and C3. Friedel pairs were therefore not measured.
4.3. Synthesis of amino-b-lactam derivatives
Table 2
In vitro cytotoxicity assay expressing LC50
(
m
M).
4.3.1. (3R,4R)-3-amino-1-(4-fluorophenyl)-4-(4-(t-
butyldimethylsilyloxy)phenyl)azetidin-2-one (2)
Compound
MDCKII
hNPC1L1/MDCKII
HepG2
Amino-
b
-lactam 2 was synthesized following the procedure
1
5
6
>100
>100
>100
>100
62,29
>100
>100
>100
69,74
>100
>100
>100
ꢀ
described by Ojima et al. [16] and Habus et al. [17] and purified by a
silica gel column chromatography (hexane/ethyl acetate 9:1, grad-
ually increasing the content of ethyl acetate in eluent to pure ethyl
5/6 (70:30)

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
729
Fig. 10. Cholesterol uptake inhibition of novel b-lactam amino alcohols 5 and 6 and their diastereoisomeric mixtures 5/6 (70:30) and 6/5 (85/15). Cholesterol uptake was deter-
mined in hNPC1L1/MDCKII cells. Cells were treated with indicated concentrations of (A) 5, (B) 6, and their diastereoismeric mixtures (C) 5/6 (70:30) and (D) 6/5 (85:15) for 1 h. The
results are expressed as percentage of inhibition compared to untreated cells and represent mean S.E.M. of three independent experiments. IC50 values were determined for each
compound or mixture. *p < 0.05, **p < 0.01, ***p < 0.001 determined by one-way ANOVA followed by Dunnett's test.
Table 3
20
In vivo inhibition (%) of cholesterol absorption by compounds 1, 5, 6, and the dia-
stereoisomeric mixture 5/6 (70:30).
acetate) and obtained as yellow oil (1.3 g, 43%). [
a
]
¼ ꢀ26 (c ¼ 1,
D
EtOAc). FT-IR (KBr) cmꢀ1: 2931, 2858, 1739, 1609, 1510, 1390, 1266,
913, 834, 513. 1H NMR (300 MHz, CDCl3): 0.19 (s, 6H, Si-(CH3)2), 0.97
Compound
Plasma
Duodenum Jejunum
Ileum
Liver
1a
5a
5b
6a
6b
60 6***
50 8***
52 7***
32 5***
65 1***
23 18
18 14
61 2***
23 13
19 16
47 3*
21 20
41 12* 33 6***
52 3***
44 6***
(s, 9H, C-(CH3)3), 1.81 (bs, 2H, NH2), 4.05 (d, 1H, J ¼ 2.2 Hz, C4
b-
lactam), 4.58 (d, 1H, J ¼ 2.1 Hz, C3 -lactam), 6.83 (d, 2H, J ¼ 8.5 Hz,
b
AreH), 6.92 (t, 2H, J1,2 ¼ 8.7 Hz, AreH), 7.18 (d, 2H, J ¼ 8.5 Hz, AreH),
43 11**
37 11** 58 8**
47 0***
49 10***
70 5***
7.23e7.26 (m, 2H, AreH); 13C NMR (150 MHz, CDCl3): ꢀ4.27 (Si-
38 11*** 35 9*
49 7**
26
8
60 9**
41 21
(CH3)2), 18.32 (C-(CH3)3), 25.75 (C-(CH3)3), 66.69 (C4
b-lactam),
5/6c (70:30) 64 7***
48 9***
70.17 (C3
b
-lactam), 115.95 (d, J ¼ 22.8 Hz, 4-F-C6H4), 118.98 (d,
a,b5 mice or c4 mice per group were gavaged with compounds a10 mg/kg/day or
b,c20 mg/kg/day for 2 days.
*p < 0.05, **p < 0.01, ***p < 0.001 determined by one-way ANOVA followed by
Dunnett's test.
J ¼ 7.8 Hz, 4-F-C6H4), 120.86 (4-OTBDMS-C6H4), 127.31 (4-OTBDMS-
C6H4), 129.18 (4-OTBDMS-C6H4), 133.83 (d, J ¼ 2.2 Hz, 4-F-C6H4),
156.25 (4-OTBDMS-C6H4), 159.26 (d, J ¼ 243.9 Hz, 4-F-C6H4), 168.02

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
730
Table 4
b
-lactam), 166.23 (d, J ¼ 254.8 Hz, 4-F-C6H4), 195.58 (C]O). HRMS
for C29H32F2N2O3Si (Mr ¼ 522.65829): calcd. m/z [MþKþ] 561.1782,
Crystallographic data collection and structure refinement details for amino
alcohol (2S,3R,4R)-6.
found 561.1788.
Compound
6
Empirical formula
Formula wt./g molꢀ1
Crystal dimensions/mm
Space group
a/Å
C
23H23F2N2O4
4.3.3. (3R,4R)-3-[(2R/S)-2-(4-fluorophenyl)-2-
hydroxyethylamino]-1-(4-fluorophenyl)-4-(4-hydroxyphenyl)
azetidin-2-one (5/6, 70:30)
The reaction was carried out in dry conditions under argon at-
mosphere. To a solution of (R)-(þ)-2-methyl-CBS-oxazaborolidine
catalyst (251 mg, 0.91 mmol) in anhydrous THF (2 mL), 2M THF
426.41
0.10 ꢂ 0.03 ꢂ 0.02
P 21
16.3185(9)
6.0148(3)
25.0538(14)
90
95.139(5)
90
b/Å
c/Å
ꢁ
ꢁ
ꢁ
/
a
/
solution of BH3$Me2S (454
mL, 0.91 mmol) was added and the
b
/
mixture stirred at room temperature for 10 min. The temperature
was then lowered to ꢀ20 ꢁC and a solution of 4 (474 mg, 0.91 mmol)
in anhydrous THF (5 mL) added dropwise. After 24 h the reaction
mixture was warmed to room temperature and quenched with
methanol. The solution was concentrated in vacuo, distilled water
(15 mL) added and the product extracted with ethyl acetate
(3 ꢂ 40 mL). Collected organic layers were dried over Na2SO4 and
solvent evaporated to dryness. Crude product was dissolved in
ethanol (22 mL) and 1 M HCl (5 mL) added. Reaction mixture was
stirred at room temperature overnight, solvent evaporated to dry-
ness and crude product purified by a silica gel column chroma-
tography (hexane/ethyl acetate 1:3) to afford diastereoisomeric
mixture 5/6 (70:30) (283 mg, 76%). mp 125e127 ꢁC. FT-IR (KBr)
cmꢀ1: 3341, 1736, 1615, 1601, 1509, 1390, 1223, 1154, 1101, 832. 1H
NMR (600 MHz, DMSO-d6): 2.72e2.83 (m, 2H, CH(OH)eCH2eNH,
g
Z
4
V/Å3
2449.2(2)
1.156
0.758
3.11e76.27
293(2)
1.54179 (CuKa)
Xcalibur Nova
ꢀ20 < h < 18;
ꢀ7 < k < 6;
ꢀ30 < l < 31
12653
Dcalc/g cmꢀ3
m
/mmꢀ1
Q
range/ꢁ
T/K
Radiation wavelength
Diffractometer type
Range of h, k, l
Reflections collected
Independent reflections
Observed reflections
6976
5385
(I ꢃ 2
s)
Absorption correction
Rint
Multi-scan
0.0429
0.1013
0.3305
1.326
Mixed
558
39
R (F)
5/6), 2.93 (bs, 1H, CH(OH)eCH2eNH, 5/6), 3.97 (s, 1H, C4
5/6), 4.55e4.57 (m, 1H, CH(OH)eCH2eNH, 6), 4.59e4.62 (m, 1H,
-lactam, 5/6), 5.30
b-lactam,
Rw (F2)
Goodness of fit
H atom treatment
No. of parameters
No. of restraints
CH(OH)eCH2eNH, 5), 4.83 (d, 1H, J ¼ 2.0 Hz, C3
b
(d, 1H, J ¼ 4.3 Hz, CH(OH)eCH2eNH, 5), 5.33 (d, 1H, J ¼ 4.3 Hz,
CH(OH)eCH2eNH, 6), 6.73e6.75 (m, 2H, AreH, 5/6), 7.09e7.23 (m,
8H, AreH, 5/6), 7.30e7.36 (m, 2H, AreH, 5/6), 9.43 (s, 1H, AreOH, 5/
6); 13C NMR (150 MHz, DMSO-d6): 54.53 (CH(OH)eCH2eNH, 5),
Drmax
,
Drmin (eÅꢀ3
)
0.646; ꢀ0.388
54.78 (CH(OH)eCH2eNH, 6), 62.36 (C4
lactam, 6), 71.43 (CH(OH)eCH2eNH, 5), 71.61 (CH(OH)eCH2eNH,
6), 75.43 (C3 -lactam, 5), 75.63 (C3 -lactam, 6), 114.60 (d,
b-lactam, 5), 62.58 (C4 b-
(C]O,
b
-lactam). HRMS for C21H27FN2O2Si (Mr ¼ 386.53518): calcd.
m/z [MþHþ] 387.1898, found 387.1901.
b
b
J ¼ 20.9 Hz, 4-F-C6H4, 5), 114.62 (d, J ¼ 21.2 Hz, 4-F-C6H4, 6), 115.68
(4-OH-C6H4, 5/6), 115.79 (d, J ¼ 22.7 Hz, 4-F-C6H4, 5/6), 118.68 (d,
J ¼ 7.7 Hz, 4-F-C6H4, 5/6),127.40 (4-OH-C6H4, 6),127.48 (4-OH-C6H4,
5), 127.52 (4-OH-C6H4, 5), 127.58 (4-OH-C6H4, 6), 127.85 (d,
J ¼ 7.8 Hz, 4-F-C6H4, 5/6),133.86 (d, J ¼ 1.9 Hz, 4-F-C6H4, 5/6),140.33
(d, J ¼ 2.7 Hz, 4-F-C6H4, 6), 140.36 (d, J ¼ 2.8 Hz, 4-F-C6H4, 5), 157.34
(4-OH-C6H4, 5/6), 158.12 (d, J ¼ 240.3 Hz, 4-F-C6H4, 5/6), 161.19 (d,
4.3.2. 2-[(3R,4R)-1-(4-fluorophenyl)-4-(4-(t-butyldimethylsilyloxy)
phenyl)-2-oxo-3-azetidinylamino]-1-(4-fluorophenyl)ethan-1-one
(4)
To a solution of 2 (159 mg, 0.41 mmol) in a mixture of anhydrous
THF and DMF (9:1, 7.8 mL), 2-bromo-1-(4-fluorophenyl)ethan-1-
one 3 (98 mg, 0.45 mmol), sodium iodide (68 mg, 0.45 mmol)
and Et3N (114
mL, 0.82 mmol) were added. The reaction mixture was
J ¼ 242.4 Hz, 4-F-C6H4, 5/6), 167.06 (C]O,
b-lactam, 5), 167.08 (C]
stirred at room temperature for 3 h after which solvent was
evaporated to dryness. Distilled water (10 mL) was added and the
product extracted with ethyl acetate (3 ꢂ 20 mL). Collected organic
layers were dried over Na2SO4 and solvent evaporated to dryness. 4
was purified by a silica gel column chromatography (hexane/ethyl
O,
b
-lactam, 6). HRMS for C23H20F2N2O3 (Mr ¼ 410.41331): calcd. m/
z [MþHþ] 411.1515, found 411.1513.
4.3.4. (3R,4R)-3-[(2S/R)-2-(4-fluorophenyl)-2-
hydroxyethylamino]-1-(4-fluorophenyl)-4-(4-hydroxyphenyl)
azetidin-2-one (6/5, 85:15)
acetate 2:1) and obtained as light brown oil (99 mg, 46%).
20
[
a]
¼ þ20 (c ¼ 1, EtOAc). FT-IR (KBr) cmꢀ1: 3332, 2931, 1748, 1693,
Reaction was carried out following the procedure for 5/6. To a
D
1600, 1506, 1385, 1260, 1156, 913, 834, 515. 1H NMR (600 MHz,
solution
(275 mg, 0.99 mmol) in anhydrous THF (2 mL), 2 M THF solution of
BH3$Me2S (496 L, 0.99 mmol) and 4 (519 mg, 0.99 mmol) in
of
(S)-(ꢀ)-2-methyl-CBS-oxazaborolidine
catalyst
CDCl3): 0.18 (s, 6H, Si-(CH3)2), 0.96 (s, 9H, C-(CH3)3), 1.60 (bs, 1H,
NH), 4.09 (d, 1H, J ¼ 2.1 Hz, C4
b
-lactam), 4.26 and 4.34 (ABq, 2H,
-lactam),
m
JAB ¼ 18.5 Hz, C(¼O)eCH2eNH), 4.80 (d,1H, J ¼ 1.9 Hz, C3
b
anhydrous THF (5 mL) were added. Diastereoisomeric mixture 6/5
6.80 (d, 2H, J ¼ 8.5 Hz, AreH), 6.93 (t, 2H, J1,2 ¼ 8.6 Hz, AreH),
7.12e7.16 (m, 4H, AreH), 7.25e7.27 (m, 2H, AreH), 7.94e7.97 (m,
2H, AreH); 13C NMR (75 MHz, CDCl3): ꢀ4.28 (Si-(CH3)2), 18.30 (C-
(85:15) (278 mg, 68%) was obtained. mp 118e120 ꢁC.
4.3.5. (3R,4R)-3-[(2S)-2-(4-fluorophenyl)-2-hydroxyethylamino]-
1-(4-fluorophenyl)-4-(4-hydroxyphenyl)azetidin-2-one (6)
Amino alcohol 6 was obtained (111 mg, 25%) by recrystallization
(CH3)3), 25.74 (C-(CH3)3), 53.00 (C(]O)eCH2eNH), 63.79 (C4
b-
lactam), 75.44 (C3
b
-lactam), 115.98 (d, J ¼ 22.6 Hz, 4-F-C6H4),
116.16 (d, J ¼ 21.7 Hz, 4-F-C6H4), 119.06 (d, J ¼ 7.9 Hz, 4-F-C6H4),
120.87 (4-OTBDMS-C6H4), 127.30 (4-OTBDMS-C6H4), 129.12 (4-
OTBDMS-C6H4), 130.62 (d, J ¼ 9.4 Hz, 4-F-C6H4), 131.57 (d,
J ¼ 3.3 Hz, 4-F-C6H4), 133.68 (d, J ¼ 2.8 Hz, 4-F-C6H4), 156.19 (4-
OTBDMS-C6H4), 159.31 (d, J ¼ 243.9 Hz, 4-F-C6H4), 166.17 (C]O,
of diastereoisomeric mixture 6/5 (85:15) in CH2Cl2. mp 118e120 ꢁC.
20
[a
]
¼ þ9 (c ¼ 1, EtOAc). FT-IR (KBr) cmꢀ1: 3344, 1723, 1615, 1510,
D
1394, 1233, 1155, 829. 1H NMR (300 MHz, DMSO-d6): 2.70e2.86 (m,
2H, CH(OH)eCH2eNH), 3.08 (bs, 1H, CH(OH)eCH2eNH), 3.97 (d,
1H, J ¼ 1.9 Hz, C4
b-lactam), 4.54e4.59 (m, 1H, CH(OH)eCH2eNH),

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
731
4.84 (d, 1H, J ¼ 1.9 Hz, C3
b
-lactam), 5.38 (d, 1H, J ¼ 4.3 Hz, CH(OH)e
834, 778, 724, 646. 1H NMR (300 MHz, CDCl3): ꢀ0.09 (s, 3H,
SieCH3), 0.10 (s, 3H, SieCH3), 0.88 (s, 9H, C-(CH3)3), 3.36e3.48 (m,
2H, CH(OTBDMS)-CH2), 4.82 (dd, 1H, J1 ¼ 7.4 Hz, J2 ¼ 4.8 Hz,
CH(OTBDMS)-CH2), 7.02 (t, 2H, J1,2 ¼ 8.7 Hz, AreH), 7.28e7.33 (m,
2H, AreH); 13C NMR (75 MHz, CDCl3): ꢀ4.75 (SieCH3), ꢀ4.60
(SieCH3), 18.35 (C-(CH3)3), 25.86 (C-(CH3)3), 39.48 (CH(OTBDMS)-
CH2), 74.76 (CH(OTBDMS)-CH2), 115.39 (d, J ¼ 21.6 Hz, 4-F-C6H4),
127.98 (d, J ¼ 8.1 Hz, 4-F-C6H4), 138.20 (d, J ¼ 3.2 Hz, 4-F-C6H4),
162.59 (d, J ¼ 245.8 Hz, 4-F-C6H4). Anal. Calcd. for C14H22BrFOSi
(Mr ¼ 333.31): C, 50.45; H, 6.65. Found: C, 50.17; H 7.05.
CH2eNH), 6.75 (d, 2H, J ¼ 8.5 Hz, AreH), 7.07e7.25 (m, 8H, AreH),
7.30e7.34 (m, 2H, AreH), 9.48 (s, 1H, AreOH); 13C NMR (75 MHz,
DMSO-d6): 54.76 (CH(OH)eCH2eNH), 62.57 (C4
(CH(OH)eCH2eNH), 75.57 (C3
b
-lactam), 71.56
b-lactam), 114.58 (d, J ¼ 21.1 Hz, 4-F-
C6H4), 115.65 (4-OH-C6H4), 115.74 (d, J ¼ 22.6 Hz, 4-F-C6H4), 118.63
(d, J ¼ 8.0 Hz, 4-F-C6H4), 127.34 (4-OH-C6H4), 127.52 (4-OH-C6H4),
127.78 (d, J ¼ 8.1 Hz, 4-F-C6H4), 133.85 (d, J ¼ 2.5 Hz, 4-F-C6H4),
140.25 (d, J ¼ 2.8 Hz, 4-F-C6H4), 157.33 (4-OH-C6H4), 158.08 (d,
J ¼ 240.7 Hz, 4-F-C6H4), 161.17 (d, J ¼ 242.2 Hz, 4-F-C6H4), 167.02
(C]O,
b
-lactam). HRMS for C23H20F2N2O3 (Mr ¼ 410.41331): calcd.
m/z [MþHþ] 411.1515, found 411.1522.
4.3.8. (1R)-2-iodo-1-(t-butyldimethylsilyloxy)-1-(4-fluorophenyl)
ethane (9a) and (1S)-2-iodo-1-(t-butyldimethylsilyloxy)-1-(4-
fluorophenyl)ethane (9b)
Reaction was carried out in a reaction flask protected from light.
To a solution of 8a (113 mg, 0.34 mmol) in acetone (3 mL), NaI
(254 mg, 1.68 mmol) was added. The reaction proceeded at 55 ꢁC
for 4 days after which distilled water (15 mL) was added and
product extracted with ethyl acetate (3 ꢂ 30 mL). Collected organic
layers were dried over Na2SO4 and solvent evaporated to dryness.
Thus obtained crude product was purified by a silicagel column
4.3.6. (1R)-2-bromo-1-(4-fluorophenyl)ethan-1-ol (7a) and (1S)-2-
bromo-1-(4-fluorophenyl)ethan-1-ol (7b)
Reaction was carried out in dry conditions under argon atmo-
sphere. For the synthesis of 7a, 2 M THF solution of BH3$Me2S
(1.15 mL, 2.30 mmol) was added to a solution of (R)-(þ)-2-methyl-
CBS-oxazaborolidine catalyst (64 mg, 0.23 mmol) in anhydrous THF
(2 mL). The mixture was stirred for 10 min and a solution of 2-
bromo-1-(4-fluorophenyl)ethan-1-one 3 (500 mg, 2.30 mmol) in
anhydrous THF (4 mL) was added dropwise. Reaction proceeded at
room temperature overnight. Reaction mixture was quenched with
methanol, solvent evaporated to dryness and distilled water
(15 mL) added. The product was extracted with CH2Cl2 (3 ꢂ 30 mL),
collected organic layers dried over Na2SO4 and solvent evaporated
to dryness. 7a was purified by a silica gel column chromatography
(hexane/ethyl acetate 6:1) and obtained as colourless oil (503 mg,
100%). The ee>99% of 7a was determined by chiral HPLC (200/4
Nucleodex beta-PM Column, MachereyeNagel, Germany, Method:
chromatography (hexane). Compound 9a was obtained as
a
mixture of 9a and unreacted 8a (93:7) as light brown oil (124 mg,
96%). 9b was synthesized following the same procedure starting
from 8b (75 mg, 0.22 mmol) and NaI (168 mg, 1.12 mmol) to give
after a silicagel column chromatography (hexane) a mixture of 9b
and unreacted 8b (94:6) as light brown oil (39 mg, 46%). FT-IR (KBr)
cmꢀ1: 3448, 2956, 2930, 2887, 2858, 1605, 1509, 1463, 1408, 1362,
1257, 1224, 1106, 997, 887, 837, 776. 1H NMR (300 MHz,
CDCl3): ꢀ0.13 (s, 3H, SieCH3), 0.10 (s, 3H, SieCH3), 0.89 (s, 9H, C-
(CH3)3), 3.28e3.30 (m, 2H, CH(OTBDMS)-CH2), 4.73 (t, 1H,
J1,2 ¼ 6.0 Hz, CH(OTBDMS)-CH2), 7.00 (t, 2H, J1,2 ¼ 8.7 Hz, AreH),
7.26e7.31 (m, 2H, AreH); 13C NMR (75 MHz, CDCl3): ꢀ4.68
(SieCH3), ꢀ4.56 (SieCH3), 15.03 (CH(OTBDMS)-CH2), 18.36 (C-
(CH3)3), 25.93 (C-(CH3)3), 74.65 (CH(OTBDMS)-CH2), 115.38 (d,
J ¼ 21.5 Hz, 4-F-C6H4), 127.80 (d, J ¼ 8.2 Hz, 4-F-C6H4), 138.87 (d,
J ¼ 3.1 Hz, 4-F-C6H4), 162.51 (d, J ¼ 246.0 Hz, 4-F-C6H4). HRMS for
0.1% TEAA in H2O: methanol 55: 45, 30 min, flow 0.7 mL/min,
20
l
¼ 254 nm, retention time 19.5 min). [
a]
¼ ꢀ28 (c ¼ 1, EtOAc). 7b
D
was synthesized following the same procedure using (S)-(ꢀ)-2-
methyl-CBS-oxazaborolidine catalyst. The ee>99% was deter-
20
mined by chiral HPLC (retention time 18.0 min). [
a]
¼ þ28 (c ¼ 1,
D
EtOAc). FT-IR (KBr) cmꢀ1: 3401, 2963, 2897, 1896, 1605, 1513, 1306,
1224, 1158, 1067, 992, 838, 779, 645, 556, 523. 1H NMR (300 MHz,
CDCl3): 2.75 (bs, 1H, CH(OH)eCH2), 3.46e3.61 (m, 2H, CH(OH)e
CH2), 4.89 (dd, 1H, J1 ¼8.7 Hz, J2 ¼ 3.5 Hz, CH(OH)eCH2), 7.05 (t, 2H,
J1,2 ¼ 8.6 Hz, AreH), 7.32e7.37 (m, 2H, AreH); 13C NMR (150 MHz,
CDCl3): 40.23 (CH(OH)eCH2), 73.33 (CH(OH)eCH2), 115.76 (d,
J ¼ 21.6 Hz, 4-F-C6H4), 127.88 (d, J ¼ 8.2 Hz, 4-F-C6H4), 136.25 (d,
J ¼ 3.1 Hz, 4-F-C6H4), 162.86 (d, J ¼ 247.0 Hz, 4-F-C6H4). Anal. Calcd.
for C8H8BrFO (Mr ¼ 219.05): C, 43.86; H, 3.68. Found: C, 43.63; H
4.01.
C
14H22FIOSi (Mr ¼ 380.31225): calcd. m/z [M-Hþ] 379.0396, found
379.0399.
4.3.9. (3R,4R)-3-[(2R)-2-(4-fluorophenyl)-2-(t-
butyldimethylsilyloxy)ethylamino]-1-(4-fluorophenyl)-4-(4-(t-
butyldimethylsilyloxy)phenyl)azetidin-2-one (10a)
Reaction was carried out in a reaction flask protected from light.
To a solution of 2 (154 mg, 0.40 mmol) in CH3CN (3 mL) 9a (151 mg,
0.40 mmol) was added. The reaction proceeded at 80 ꢁC for 7 days
after which the solvent was evaporated to dryness. Compound 10a
was purified by a silica gel column chromatography (hexane/ethyl
4.3.7. (1R)-2-bromo-1-(t-butyldimethylsilyloxy)-1-(4-
fluorophenyl)ethane (8a) and (1S)-2-bromo-1-(t-
butyldimethylsilyloxy)-1-(4-fluorophenyl)ethane (8b)
acetate 6:1) and obtained as light brown oil (52 mg, 20%).
¼ ꢀ9 (c ¼ 1, EtOAc). FT-IR (KBr) cmꢀ1: 3480, 2931, 2858, 1748,
20
Reaction was carried out in dry conditions under argon atmo-
sphere. For the synthesis of 8a a solution of imidazole (254 mg,
3.74 mmol) in DMF (0.5 mL) was added to a solution of 7a (327 mg,
1.49 mmol) in DMF (1 mL) and stirred for 10 min, followed by
dropwise addition of TBDMSCl (293 mg, 1.94 mmol) solution in
DMF (1.3 mL). The reaction proceeded for 72 h at room tempera-
ture. Solvent was evaporated to dryness, distilled water (15 mL)
added and the resulting mixture extracted with ethyl acetate
(3 ꢂ 30 mL). Collected organic layers were dried over Na2SO4 and
solvent evaporated to dryness. 8a was purified by a silica gel col-
[a
]
D
1607, 1505, 1464, 1385, 1259, 1226, 1101, 1006, 914, 834, 515. 1H
NMR (300 MHz, CDCl3): ꢀ0.15 (s, 3H, SieCH3), 0.02 (s, 3H, SieCH3),
0.18 (s, 6H, Si-(CH3)2), 0.86 (s, 9H, C-(CH3)3), 0.96 (s, 9H, C-(CH3)3),
2.01 (bs, 1H, CH(OTBDMS)eCH2eNH), 2.78e2.99 (m, 2H,
CH(OTBDMS)-CH2-NH), 4.00 (d, 1H, J ¼ 2.1 Hz, C4
b
-lactam), 4.61 (d,
1H, J ¼ 1.8 Hz, C3
b
-lactam), 4.77 (dd, 1H, J1 ¼ 7.8 Hz, J2 ¼ 4.3 Hz,
CH(OTBDMS)eCH2eNH), 6.81 (d, 2H, J ¼ 8.7 Hz, AreH), 6.91 (t, 2H,
J1,2 ¼ 8.8 Hz, AreH), 6.99 (t, 2H, J1,2 ¼ 8.8 Hz, AreH), 7.13 (d, 2H,
J ¼ 8.5 Hz, AreH), 7.21e7.30 (m, 4H, AreH); 13C NMR (75 MHz,
CDCl3): ꢀ4.75 (SieCH3), ꢀ4.41 (SieCH3), ꢀ4.28 (Si-(CH3)2), 18.29
(C-(CH3)3), 25.73 (C-(CH3)3), 25.96 (C-(CH3)3), 55.69 (CH(OTBDMS)-
umn chromatography (hexane/ethyl acetate 6:1) and obtained as
20
colourless oil (452 mg, 91%). [
a
]
¼ ꢀ46 (c ¼ 1, EtOAc). 8b was
D
synthesized following the same procedure starting from 7b
CH2-NH), 63.86 (C4
75.73 (C3
b-lactam), 74,18 (CH(OTBDMS)eCH2eNH),
-lactam), 115.24 (d, J ¼ 21.3 Hz, 4-F-C6H4), 115.90 (d,
(424 mg, 1.94 mmol) and obtained as colourless oil (602 mg, 94%).
b
20
[
a]
¼ þ46 (c ¼ 1, EtOAc). FT-IR (KBr) cmꢀ1: 2957, 2930, 2887,
J ¼ 22.6 Hz, 4-F-C6H4), 118.92 (d, J ¼ 7.9 Hz, 4-F-C6H4), 120.84 (4-
D
2858, 1606, 1509, 1463, 1417, 1362, 1257, 1226, 1155, 1113, 1012, 914,
OTBDMS-C6H4), 127.23 (4-OTBDMS-C6H4), 127.95 (d, J ¼ 8.0 Hz, 4-

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
732
F-C6H4), 129.40 (4-OTBDMS-C6H4), 133.78 (d, J ¼ 2.7 Hz, 4-F-C6H4),
138.86 (d, J ¼ 3.0 Hz, 4-F-C6H4), 156.11 (4-OTBDMS-C6H4),159.20 (d,
J ¼ 243.6 Hz, 4-F-C6H4), 162.33 (d, J ¼ 245.5 Hz, 4-F-C6H4), 166.38
water (100 mL, 0.1 mol/L). NaOH solutions (1.5e20
to each solution of amino alcohol 5 or 6, pH values measured and
UV spectra ( 200e350 nm) recorded at pH 6e12. The titrations
were performed at rt (20 ꢁC) and inflection points of sigmoid curves
at 247 nm were calculated as pKa values for both amino alcohols.
mL) were added
l
(C]O,
b
-lactam). HRMS for C35H48F2N2O3Si2 (Mr ¼ 638.93503):
calcd. m/z [MþNaþ] 661.3063, found 661.3082.
l
4.3.10. (3R,4R)-3-[(2S)-2-(4-fluorophenyl)-2-(t-
butyldimethylsilyloxy)ethylamino]-1-(4-fluorophenyl)-4-(4-(t-
butyldimethylsilyloxy)phenyl)azetidin-2-one (10b)
4.5. Cell culture and establishment of the cell line stably expressing
hNPC1L1 protein
10b was synthesized following the procedure for 10a and ob-
MadineDarby Canine Kidney II (MDCKII) cells, MDCKII cells
stably expressing human NPC1L1 (hNPC1L1/MDCKII), and human
HepG2 liver cells were maintained in DMEM (Dulbecco's Modified
Eagle Medium, Invitrogen, USA) containing 100 units/mL penicillin
20
tained as light brown oil (9 mg, 19%). [
a
]
¼ þ16 (c ¼ 1, EtOAc). FT-
D
IR (KBr) cmꢀ1: 3439, 2929, 2857, 1749, 1608, 1506, 1472, 1385, 1259,
1101, 1012, 913, 833, 516. 1H NMR (300 MHz, CDCl3): ꢀ0.14 (s, 3H,
SieCH3), 0.04 (s, 3H, SieCH3), 0.18 (s, 6H, Si-(CH3)2), 0.86 (s, 9H, C-
(CH3)3), 0.96 (s, 9H, C-(CH3)3),1.80 (bs,1H, CH(OTBDMS)eCH2eNH),
2.80e2.96 (m, 2H, CH(OTBDMS)-CH2-NH), 4.02 (d,1H, J ¼ 2.0 Hz, C4
and 100 mg/mL streptomycin supplemented with 10% FCS (Sigma-
eAldrich, Germany). The cells were grown at 37 ꢁC in a humidified
5% CO2 incubator. Stable transfection of MDCKII cells with human
NPC1L1 protein was performed using Lipofectamine LTX (Invi-
trogen, USA) according to the manufacturer's protocol. Selection
with 1 mg/mL G418 (SigmaeAldrich, Germany) was started 2 days
after transfection. The concentration of G418 was decreased to
b
-lactam), 4.62 (d, 1H, J ¼ 2.0 Hz, C3
b-lactam), 4.81 (dd, 1H,
J1 ¼ 7.6 Hz, J2 ¼ 4.3 Hz, CH(OTBDMS)eCH2eNH), 6.81 (d, 2H,
J ¼ 8.6 Hz, AreH), 6.90 (t, 2H, J1,2 ¼ 8.8 Hz, AreH), 6.99 (t, 2H,
J1,2 ¼ 8.7 Hz, AreH), 7.12 (d, 2H, J ¼ 8.5 Hz, AreH), 7.20e7.31 (m, 4H,
AreH); 13C NMR (75 MHz, CDCl3): ꢀ4.72 (SieCH3), ꢀ4.39
(SieCH3), ꢀ4.27 (Si-(CH3)2), 18.30 (C-(CH3)3), 25.74 (C-(CH3)3),
500
m
g/mL after 14 days. Stably transfected cells were maintained
g/mL
g/mL G418.
in DMEM containing 100 units/mL penicillin and 100
streptomycin supplemented with 10% FCS and 500
m
25.97 (C-(CH3)3), 55.68 (CH(OTBDMS)-CH2-NH), 63.35 (C4
b
-lac-
m
tam), 74,25 (CH(OTBDMS)eCH2eNH), 75.82 (C3
b-lactam), 115.25
(d, J ¼ 21.5 Hz, 4-F-C6H4), 115.91 (d, J ¼ 22.8 Hz, 4-F-C6H4), 118.94 (d,
J ¼ 7.9 Hz, 4-F-C6H4), 120.84 (4-OTBDMS-C6H4), 127.25 (4-OTBDMS-
C6H4), 127.96 (d, J ¼ 8.0 Hz, 4-F-C6H4), 129.35 (4-OTBDMS-C6H4),
133.77 (d, J ¼ 2.8 Hz, 4-F-C6H4), 138.89 (d, J ¼ 3.0 Hz, 4-F-C6H4),
156.13 (4-OTBDMS-C6H4), 159.22 (d, J ¼ 243.0 Hz, 4-F-C6H4), 162.34
4.6. Cytotoxicity assay
MDCKII, hNPC1L1/MDCKII, and HepG2 cells were seeded in 96-
well microtiter plates on day 0 at densities of 3000 cells/well. On
day 1, tested compounds (73 mM stock solution in DMSO) were
added in five consecutive 10-fold dilutions (10ꢀ8 to 10ꢀ4 mol/L) and
incubated for 72 h. Cytotoxicity of the compounds was assessed on
day 4 by the MTTcell proliferation assay (SigmaeAldrich, Germany)
which detects mitochondrial dehydrogenase activity in viable cells
[33]. The results are expressed as LC50 and the values for each
compound were calculated from doseeresponse curves using
linear regression analysis. Each concentration was tested in
quadruplicate in three independent experiments. Toxicity of the
compounds in combination with micelles was also tested using the
same method. MDCKII and hNPC1L1/MDCKII were seeded in 96-
(d, J ¼ 245.4 Hz, 4-F-C6H4), 166.36 (C]O,
b-lactam). HRMS for
C
35H48F2N2O3Si2 (Mr ¼ 638.93503): calcd. m/z [MþHþ] 639.3244,
found 639.3256.
4.3.11. (3R,4R)-3-[(2R)-2-(4-fluorophenyl)-2-hydroxyethylamino)]-
1-(4-fluorophenyl)-4-(4-hydroxyphenyl)azetidin-2-one (5)
Compound 10a (24 mg, 0.04 mmol) was dissolved in 3% HCl in
ethanol (1.59 mL) and the mixture was stirred at room temperature
overnight. Solvent was evaporated to dryness and crude product
purified by a silica gel column chromatography (hexane/ethyl ac-
etate 1:3, 5% MeOH). Product 5 was obtained as light yellow
20
well microtiter plates (150 mL medium/well) at a density of
powder (13 mg, 84%). mp 136e138 ꢁC. [
a]
¼ ꢀ18 (c ¼ 1, EtOAc).
D
2 ꢂ 105 cells/mL medium 24 h before the experiment. Micelles were
FT-IR (KBr) cmꢀ1: 3448, 1735, 1617, 1509, 1389, 1225, 1102, 833. 1H
prepared according to the modified method reported by Field et al.
NMR (300 MHz, DMSO-d6): 2.75e2.77 (m, 2H, CH(OH)eCH2eNH),
[34]. In brief, 0.25 mM oleic acid, 50
mM cholesterol, 10 mM com-
2.95 (bs, 1H, CH(OH)eCH2eNH), 3.97 (bs, 1H, C4
b-lactam)
pactin, 50 M mevalonate, 5 mM Na-taurocholate (SigmaeAldrich,
m
4.58e4.63 (m, 1H, CH(OH)eCH2eNH), 4.84 (d, 1H, J ¼ 2.0 Hz, C3
b-
Germany) in DMEM/10% FCS were mixed and sonicated. Following
the micellar formation, compounds 1, 5, 6, and 5/6 (70:30) (73 mM
stock solution in DMSO) were added to the medium and vortexed
to obtain the final concentrations of the compounds (25, 50, 100,
150, 200 mM) and applied to the cells for 1 h. Thereafter, cytotoxicity
was determined using the MTT assay.
lactam), 5.33 (d, 1H, J ¼ 4.4 Hz, CH(OH)eCH2eNH), 6.74 (d, 2H,
J ¼ 8.5 Hz, AreH), 7.08e7.24 (m, 8H, AreH), 7.33e7.37 (m, 2H,
AreH), 9.46 (s, 1H, AreOH); 13C NMR (75 MHz, DMSO-d6): 54.55
(CH(OH)eCH2eNH), 62.34 (C4
b
-lactam), 71.45 (CH(OH)e
CH2eNH), 75.47 (C3
b
-lactam), 114.60 (d, J ¼ 21.0 Hz, 4-F-C6H4),
115.66 (4-OH-C6H4), 115.80 (d, J ¼ 21.7 Hz, 4-F-C6H4), 118.65 (d,
J ¼ 8.0 Hz, 4-F-C6H4), 127.49 (4-OH-C6H4), 127.52 (4-OH-C6H4),
127.84 (d, J ¼ 8.1 Hz, 4-F-C6H4), 133.87 (d, J ¼ 2.3 Hz, 4-F-C6H4),
140.40 (d, J ¼ 2.7 Hz, 4-F-C6H4), 157.31 (4-OH-C6H4), 158.08 (d,
J ¼ 240.7 Hz, 4-F-C6H4), 161.17 (d, J ¼ 241.6 Hz, 4-F-C6H4), 167.10
4.7. Medium preparation
Medium A contained DMEM plus 100 units/mL penicillin and
100 mg/mL streptomycin supplemented with 5% lipoprotein-
deficient serum (density > 1.21 g/mL, prepared from FCS by ultra-
centrifugation). Cholesterol Replenishing Medium (Medium B) was
prepared according to the modified method reported by Field et al.
(C]O,
b
-lactam). HRMS for C23H20F2N2O3 (Mr ¼ 410.41331): calcd.
m/z [MþHþ] 411.1515, found 411.1517.
4.4. Spectrophotometric titration
[34]. In brief, 0.25 mM oleic acid, 50
compactin, 50 M mevalonate, 5 mM Na-taurocholate (Sigma-
eAldrich, Germany), and [3H]cholesterol (0.18
Ci/mL medium,
[1,2e3H(N)] cholesterol, 1 mCi/mL (ARC Inc., USA)) in Medium A
were mixed and sonicated. Following the micellar formation,
compounds 1, 5, 6, 5/6 (70:30), and 6/5 (85:15) (73 mM stock
mM free cholesterol, 10 mM
Stock solution of amino alcohols 5 and 6 (160
m
L, 2 mg/mL
m
methanol) was added to NaCl solution in water (20 mL, 0.1 mol/L),
thus resulting in a solution of each amino alcohol 5 and 6 in con-
centration of 4 ꢂ 10ꢀ5 mol/L. Furthermore, solutions of NaOH
(0.1 mol/L, 1 mol/L and 2 mol/L) were prepared in NaCl solution in
m

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
733
[3] S. Lamon-Fava, Statins and lipid metabolism: an update, Curr. Opin. Lipidol. 24
(2013) 221e226.
[4] C.M. Minder, R.S. Blumenthal, M.J. Blaha, Statins for primary prevention of
cardiovascular disease: the benefits outweigh the risks, Curr. Opin. Cardiol. 28
(2013) 554e560.
solution in DMSO) were added to Medium B and vortexed to obtain
the final concentrations (10, 30, 60, 90, 120, 150, 200 mM).
4.8. In vitro cholesterol uptake assay
[5] D.A. Burnett, b-Lactam cholesterol absorption inhibitors, Curr. Med. Chem. 11
(2004) 1873e1887.
MDCKII wildtype and hNPC1L1/MDCKII cells were seeded in 24-
[6] D.P. Mikhailidis, R.W. Lawson, A.L. McCormick, G.C. Sibbring,
A.M. Tershakovec, G.M. Davies, K. Tunceli, Comparative efficacy of the addition
of ezetimibe to statin vs statin titration in patients with hyper-
cholesterolaemia: systematic review and meta-analysis, Curr. Med. Res. Opin.
27 (2011) 1191e1210.
well plates at a density of 5 ꢂ 105 cells/mL medium (500
mL me-
dium/well) in Medium A 24 h before the treatment. Cells were
washed once with PBS and incubated in Medium B (500 mL me-
dium/well) containing micelles and tested compounds in indicated
concentrations for 1 h. Following the incubation, the medium was
removed and cells were washed three times with ice-cold 0.2% free
fatty acid-free BSA. Cells were lysed and radioactivity in the lysate
[7] S.B. Rosenblum, T. Huynh, A. Afonso, H.R. Davis Jr., N. Yumibe, J.W. Clader,
D.A. Burnett, Discovery of 1-(4-Fluorophenyl)-(3R)-[3-(4-fluorophenyl)-(3S)-
hydroxypropyl]-(4S)-(4-hydroxyphenyl)-2-azetidinone (SCH 58235):
A
designed, potent, orally active inhibitor of cholesterol absorption, J. Med.
Chem. 41 (1998) 973e980.
[8] S.W. Altmann, H.R. Davis Jr., L. Zhu, X. Yao, L.M. Hoos, G. Tetzloff, S.P.N. Iyer,
M. Maguire, A. Golovko, M. Zeng, L. Wang, N. Murgolo, M.P. Graziano, Nie-
mann-Pick C1 like 1 protein is critical for intestinal cholesterol absorption,
Science 303 (2004) 1201e1204.
[9] C. Xie, Z.-S. Zhou, N. Li, Y. Bian, Y.-J. Wang, L.-J. Wang, B.-L. Li, B.-L. Song,
Ezetimibe blocks the internalization of NPC1L1 and cholesterol in mouse small
intestine, J. Lipid Res. 53 (2012) 2092e2101.
[10] B.A. McKittrick, K. Ma, K. Huie, N. Yumibe, H. Davis Jr., J.W. Clader,
M. Czarniecki, Synthesis of C3 heteroatom-substituted azetidinones that
display potent cholesterol absorption inhibitory activity, J. Med. Chem. 41
(1998) 752e759.
(100 mL) was determined by liquid scintillation counting. Protein
concentrations of the lysates were determined with the DC Protein
Assay (Bio-Rad, USA). The results were calculated as cpm/mg pro-
tein and expressed as percentage of inhibition compared to un-
treated cells. Each concentration was tested in duplicate in three
independent experiments.
4.9. In vivo acute cholesterol absorption
ꢀ
ꢁ
ꢀ
[11] T. Poljak, K. Molcanov, D. Margetic, I. Habus, Study on the Lewis acid-
promoted aza-Diels-Alder reaction of azetidin-2-one-tethered imines with
siloxydienes in the asymmetric synthesis of 2-aryl(alkyl)-2,3-dihydro-4-
pyridones, Croat. Chem. Acta 81 (2008) 539e558.
Male C57BL/6 mice aged 25e28 weeks were maintained under a
12 h light/12 h dark cycle in a temperature-controlled environment
with free access to chow diet (Ssniff, Germany) and water. For 2
days, mice (n ¼ 3e5) were fasted for 4 h and gavaged with corn oil
ꢁ
ꢀ
[12] K. Radolovic, I. Habus, B. Kralj, New thiazolidinone and triazinethione con-
jugates derived from amino- -lactams, Heterocycles 78 (2009) 1729e1759.
[13] V. Kovac, K. Radolovic, I. Habus, D. Sieber, K. Heinze, V. Rapic, Conformational
analysis of -lactam-containing ferrocene peptides, Eur. J. Inorg. Chem. 3
b
ꢀ
ꢁ
ꢀ
ꢁ
(100 ml) (vehicle) containing ezetimibe 1 (10 mg/kg/day) or com-
b
pounds 5, 6, and 5/6 (70:30) (10 or 20 mg/kg/day). On day 2, 90 min
(2009) 389e399.
ꢁ
ꢀ
ꢀ
post vehicle or compound treatment, mice were gavaged with corn
[14] K. Radolovic, K. Molcanov, I. Habus, Synthesis and structure determination of
methyl (3S)- and (3R)-(trans-(30R,40R)-3-amino-1-(4-methoxyphenyl)-2-oxo-
4-phenylazetidin-3-yl)-2,2-dimethyl-3-ferrocenylpropanoate, J. Mol. Struct.
966 (2010) 8e13.
[15] K. Vazdar, D. Margetic, I. Habus, Lewis acid promoted diastereoselective
Mannich reaction of -lactam-Tethered aldimines with 1-methoxy-1-
oil (200
m mCi) ([1,2e3H(N)]
l) containing [3H]cholesterol (2
cholesterol, 1 mCi/mL, ARC Inc., St. Louis, USA). Plasma, liver, and
small intestines were collected 4 h post-gavage. Cholesterol uptake
and absorption were determined as previously described [35].
ꢁ
ꢀ
b
trimethylsilyloxy-2,2-dimethylethene, Heterocycles 83 (2011) 63e90.
ꢀ
[16] I. Ojima, I. Habus, Asymetric synthesis of
b-lactams by chiral ester enolate-
4.10. Statistical analyses
imine condensation, Tetrahedron Lett. 31 (1990) 4289e4292.
ꢀ
ꢀ
ꢀ
ꢀ
ꢁ
[17] T. Poljak, K. Molcanov, A. Visnjevac, I. Habus, V. Kovac, V. Rapic, Synthesis and
structure of
trans-(þ)-(3S,4S)-3-amino-4-ferrocenyl-1-p-methox-
yphenylazetidin-2-one, J. Mol. Struct. 751 (2005) 60e64.
[18] J.A. Dale, D.L. Dull, H.S. Mosher, -Methoxy- -trifluoromethylphenylacetic
The results are presented as mean S.D. or mean S.E.M. Sta-
tistical analyses were performed using unpaired two-tailed t-test or
one-way analysis of variance (ANOVA) followed by Dunnett's test
for multiple comparisons using GraphPad Prism 5.0 software (San
Diego, CA, USA). Differences were considered significant at P < 0.05.
IC50 values were determined with GraphPad Prism 5.0.
a
a
acid, a versatile reagent for the determination of enantiomeric composition of
alcohols and amines, J. Org. Chem. 34 (1969) 2543e2549.
[19] E.J. Corey, R.K. Bakshi, S. Shibata, Highly enantioselective borane reduction of
ketones catalyzed by chiral oxazaborolidines. Mechanism and synthetic im-
plications, J. Am. Chem. Soc. 109 (1987) 5551e5553.
[20] G. Wu, Y. Wong, X. Chen, Z. Ding, A novel one-step diastereo- and enantio-
selective formation of trans-azetidinones and its application to the total
synthesis of cholesterol absorption inhibitors, J. Org. Chem. 64 (1999)
3714e3718.
Acknowledgements
[21] T. Ritter, L. Kværnø, M. Werder, H. Hauser, E.M. Carreira, Heterocyclic ring
scaffolds as small-molecule cholesterol absorption inhibitors, Org. Biomol.
Chem. 3 (2005) 3514e3523.
This research was funded by the Croatian Ministry of Science,
Education and Sports (Programmes 098-0982915-2948 and 098-
0982522-2525), the Austrian Science Fund FWF (DK-MCD W1226,
SFB LIPOTOX F30, P22832), and the Croatian-Austrian Intergov-
ernmental S&T Cooperation Programme (HR 08/12). The authors
[22] G.J. Quallich, T.M. Woodall, Enantioselective oxazaborolidine reduction of
ketones containing heteroatoms, Tetrahedron Lett. 34 (1993) 785e788.
[23] E.J. Corey, C.J. Helal, Asymmetric synthesis of (S)-carbinoxamine. New aspects
of oxazaborolidine-catalyzed enantioselective carbonyl reduction, Tetrahe-
dron Lett. 37 (1996) 5675e5678.
ꢁ
ꢁ
thank Dr. Biserka Kojic-Prodic, Dr. Marijeta Kralj, and Dr. Vesna
€
[24] S. Karlsson, J.H. Sorensen, Selection and development of a route for cholesterol
ꢀ
ꢀ
ꢁ
Blazek (RuCer Boskovic Institute) for scientific discussions and
absorption inhibitor AZD4121, Org. Process Res. Dev. 16 (2012) 586e594.
[25] Atom numbering in crystal structure does not follow conventional 2-
azetidinone atom numbering
Anton Ibovnik (Medical University of Graz) for technical assistance.
[26] L.J. Farrugia, ORTEP-3 for windows e a version of ORTEP-III with a graphical
user interface (GUI), J. Appl. Cryst. 30 (1997) 565.
[27] A.B. Weinglass, M.G. Kohler, E.O. Nketiah, J. Liu, W. Schmalhofer, A. Thomas,
Appendix A. Supplementary data
€
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.10.014.
B. Williams, L. Beers, L. Smith, M. Hafey, K. Bleasby, J. Leone, Y.S. Tang,
M. Braun, F. Ujjainwalla, M.E. McCann, G.J. Kaczorowski, M.L. Garcia,
MadineDarby Canine kidney cells: a pharmacologically validated system for
NPC1L1-mediated cholesterol uptake, Mol. Pharmacol. 73 (2008)
1072e1084.
References and notes
[28] D.A. Burnett, M.A. Caplen, H.R. Davis Jr., R.E. Burrier, J.W. Clader, 2-
Azetidinones as inhibitors of cholesterol absorption, J. Med. Chem. 37
(1994) 1733e1736.
[1] J.A. Finegold, P. Asaria, D.P. Francis, Mortality from ischaemic heart disease by
country, region, and age: statistics from World Health Organisation and
United Nations, Int. J. Cardiol. 168 (2013) 934e945.
[2] E. Bruckert, New advances in lipid-modifying therapies for reducing cardio-
vascular risk, Cardiology 97 (2002) 59e66.
[29] CrysAlis PRO, Oxford Diffraction Ltd., U.K, 2007.
[30] G.M. Sheldrick,
A short history of SHELX, Acta Crystallogr. A64 (2008)
112e122.

ꢀ ꢁ
T. Drazic et al. / European Journal of Medicinal Chemistry 87 (2014) 722e734
734
[31] A.L. Spek, Single-crystal structure validation with the program PLATON,
J. Appl. Cryst. 36 (2003) 7e13.
[32] C.F. Macrae, P.R. Edgington, P. McCabe, E. Pidcock, G.P. Shields, R. Taylor,
M. Towler, J. van de Streek, Mercury: visualization and analysis of crystal
structures, J. Appl. Cryst. 39 (2006) 453e457.
[33] T. Mosmann, Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays, J. Immunol. Methods 65
(1983) 55e63.
[34] F.J. Field, T. Shreves, D. Fujiwara, S. Murthy, E. Albright, S.N. Mathur, Regula-
tion of gene expression and synthesis and degradation of 3-hydroxy-3-
methylglutaryl coenzyme A reductase by micellar cholesterolin CaCo-2 cells,
J. Lipid Res. 32 (1991) 1811e1821.
[35] S. Obrowsky, P.G. Chandak, J.V. Patankar, S. Povoden, S. Schlager, E.E. Kershaw,
J.G. Bogner-Strauss, G. Hoefler, S. Levak-Frank, D. Kratky, Adipose triglyceride
lipase is a TG hydrolase of the small intestine and regulates intestinal PPAR
a
signaling, J. Lipid Res. 54 (2013) 425e435.