2-(4-Carbonylphenyl)benzoxazole inhibitors of CETP: Attenuation of hERG binding and improved HDLc-raising efficacy

Article abstract of DOI:10.1016/j.bmcl.2011.02.049

The development of 2-phenylbenzoxazoles as inhibitors of cholesteryl ester transfer protein (CETP) is described. Efforts focused on finding suitable replacements for the central piperidine with the aim of reducing hERG binding: a main liability of our benchmark benzoxazole (1a). Replacement of the piperidine with a cyclohexyl group successfully attenuated hERG binding, but was accompanied by reduced in vivo efficacy. The approach of substituting a piperidine moiety with an oxazolidinone also attenuated hERG binding. Further refinement of this latter scaffold via SAR at the pyridine terminus and methyl branching on the oxazolidinone led to compounds 7e and 7f, which raised HDLc by 33 and 27 mg/dl, respectively, in our transgenic mouse PD model and without the hERG liability of previous series.

Full text of DOI:10.1016/j.bmcl.2011.02.049

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2597–2600
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
2-(4-Carbonylphenyl)benzoxazole inhibitors of CETP: Attenuation of hERG
binding and improved HDLc-raising efficacy
Ramzi F. Sweis a, , Julianne A. Hunt a, Peter J. Sinclair a, Ying Chen b, Suzanne S. Eveland b, Qiu Guo b,
⇑
Sheryl A. Hyland b, Denise P. Milot b, Anne-Marie Cumiskey c, Melanie Latham c, Raymond Rosa c,
Larry Peterson c, Carl P. Sparrow b, Matt S. Anderson b
a Department of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
b Department of Cardiovascular Diseases, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
c Department of Pharmacology, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The development of 2-phenylbenzoxazoles as inhibitors of cholesteryl ester transfer protein (CETP) is
described. Efforts focused on finding suitable replacements for the central piperidine with the aim of
reducing hERG binding: a main liability of our benchmark benzoxazole (1a). Replacement of the piperi-
dine with a cyclohexyl group successfully attenuated hERG binding, but was accompanied by reduced
in vivo efficacy. The approach of substituting a piperidine moiety with an oxazolidinone also attenuated
hERG binding. Further refinement of this latter scaffold via SAR at the pyridine terminus and methyl
branching on the oxazolidinone led to compounds 7e and 7f, which raised HDLc by 33 and 27 mg/dl,
respectively, in our transgenic mouse PD model and without the hERG liability of previous series.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 27 December 2010
Revised 7 February 2011
Accepted 14 February 2011
Available online 23 February 2011
Keywords:
Cholesteryl ester transfer protein
Cholesterol
hERG
HDLc
LDLc
Atherosclerosis
For years, cardiovascular disease (CD) has remained the leading
cause of mortality in the world. Noteworthy is the fact that despite
the wide-spread use of methods for treatment (i.e., HMG-CoA
reductase inhibitors), there remains a large unmet medical need
due to the continued and prevalent impact of this disease on hu-
man health. For some time, it has been known that an inverse cor-
relation exists between levels of high-density lipoprotein
cholesterol (HDLc) and the relative risk for developing atheroscle-
rosis (and hence CD).1 Epidemiological studies have indicated that
a 1 mg/dl increase in HDLc is associated with a 6% reduction in the
risk of death from coronary heart disease (CHD).2 Despite this, the
current primary therapy for HDLc elevation has remained un-
changed for years (niacin), and achieves only modest efficacy
(ꢀ20%).
of HDLc.3 Indeed, clinical studies have been conducted for small-
molecule inhibitors validating this approach in humans.4 Hence,
there is ongoing and continued interest in the development of
small-molecule inhibitors of CETP, particularly if clinical studies
determine that this approach to HDLc treatment results in reduced
risk for development of CHD.
Previously, we have disclosed the discovery and development of
a 2-phenylbenzoxazole class of CETP inhibitors.5 This work lead to
the development of a series of benzoxazoles containing the 2-aryl-
pyridyl moiety in the eastern half of the molecule (Fig. 1). These
compounds were potent (<50 nM) inhibitors of CETP and demon-
strated robust elevation of HDLc (>20 mg/dl at 10 mpk) in our
transgenic mouse PD model.5b,6 Unfortunately, these compounds
were also plagued by severe human ether-a-go-go-related gene
(hERG) binding as reported in our MK499 assay.7 Compound 1
inhibited hERG with an IKr binding IC50 of 56 nM. Our approach
to address this issue focused on finding a suitable modification to
the piperidine core. Ultimately, this led to the discovery of oxazo-
lidinones 7e and 7f, which possessed improved efficacy in vivo,
reduced log D value,8 and no hERG liability.
In the search for modern approaches to HDLc-raising therapy,
cholesteryl ester transfer protein (CETP) has risen to the forefront
of potential targets. CETP is a plasma glycoprotein that facilitates
exchange of cholesteryl ester from HDL to low-density- and very-
low-density lipoproteins (LDL and VLDL). This transfer is accompa-
nied by a reverse exchange of triglycerides. As such, inhibition of
CETP has been proposed as an approach for raising plasma levels
Recent literature has provided a wealth of information on
successful tactics to deal with the issue of hERG binding.9 There
have been thorough analyses of the types of chemical moieties
and physical properties that molecules with this liability generally
⇑
Corresponding author. Tel.: +1 732 594 2742.
E-mail address: rfsweis@gmail.com (R.F. Sweis).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.049

2598
R. F. Sweis et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2597–2600
HN
N
O
OiPr
F
1a: IC50 : 13 nM
N
O
Δ
HDL: +24 mg/dl
hERG(MK499): 56 nM
N
NC
O
Me
N
7e(ent a)
HDL: +33 mg/dl
hERG(MK499): >30,000 nM
: IC50 : 22 nM
N
H
Δ
O
O
N
OCF3
N
O
7f(ent b)
: IC50 : 18 nM
Δ
HDL: +27 mg/dl
NC
hERG(MK499): >30,000 nM
Figure 1. 2-Phenylbenzoxazole CETP inhibitors.
have.10 In the previously discovered bi-aryl piperidine subclass (1),
the source of the hERG liability was believed to be the piperidine
moiety since the hERG binding issue only became serious after
its introduction in our prior SAR work. As a result, our approach
to address this liability centered on finding a suitable replacement
for this embedded unit. The first approach was to remove the nitro-
gen and employ a cyclohexyl group as a replacement. Our second
approach utilized an oxazolidinone replacement as inspired by
the success of its use in clinical compound MK-0859.11
The synthesis of the cyclohexyl replacement began with 4-N-
Boc-aminomethyl-cyclohexanone. Removal of the Boc group with
TFA followed by amide formation with the previously synthesized
2-arylbenzoxazole fragment provided the elaborated cyclohexa-
none 2. Vinyl triflate formation with trifluoromethanesulfonic
anhydride in the presence of 2,6-lutidine allowed for further elab-
oration via Suzuki cross-coupling with various aryl boronic acids.
Finally, hydrogenation of the cycloalkene furnished both cis- and
trans- isomers of all compounds evaluated in this series
(Scheme 1).
approach of azide displacement and reduction with triphenylphos-
phine which provided pyridyl oxazolidinones 5. This was coupled
to the left-hand fragment, 4-(5-cyano-7-isopropyl-benzooxazol-
2-yl)-benzoic acid, via the intermediate acyl chloride. Finally, the
terminal 4-chloropyridine on the right-hand terminus of the mol-
ecule was used as a handle for further elaboration via Suzuki cross-
coupling reactions to give final compounds 6–7 (Scheme 2).
The initial survey into the cyclohexyl class of molecules focused
on simple substitution of the terminal phenyl ring rather than the
more-elaborated biphenyl unit found in benchmark compound 1 in
order to quickly determine the feasibility of this approach to ad-
dress hERG.14 As shown in Table 1, CETP inhibition prefers the
trans isomer in general, with the para t-Bu group being the best
among the initial group of substituents. This strategy proved suc-
cessful in addressing hERG binding. However, the two most potent
compounds, 3a and 3b, showed only modest efficacy towards rais-
ing HDLc in our mouse PD model (17 and 18 mg/dl, respectively).
In addition, the removal of the nitrogen from the original piperi-
dine led to an even more lipophilic class of molecules as evidenced
by high log D values (7.1–7.3). For these reasons, we did not envi-
sion the strategy of piperidine-to-cyclohexane replacement as a
viable one going forward, and we shifted our focus towards evalu-
ating the oxazolinone class of molecules.
Our approach towards incorporation of an oxazolidinone
replacement commenced with epichlorohydrin (R- or S-) or 2-chlo-
romethyl-2-methyl-oxirane. Potassium cyanate-mediated ring
expansion furnished the corresponding chloromethyl oxazolidi-
nones 4.12 This intermediate was coupled to 2-bromo-4-chloropyr-
Our strategy to incorporate the oxazolidinone fragment did not
look promising at first. The chloro-pyridine analog 6a showed
minimal CETP inhibition (19 lM) leading us to question the
idine via
chloride was then converted to the primary amine via the two-step
a
palladium/Xanthphos-catalyzed amidation.13 The
O
NH
c, d,
e
O
a, b
Boc
N
H
N
O
O
2
NC
O
NH
O
N
R
NC
3
Scheme 1. Synthesis of benzoxazoles 3a–k. Reagents and conditions: (a) triflouroacetic acid, CH2Cl2; (b) HBTU, HOBT, (i-Pr)2NEt, 4-(5-cyano-7-isopropylbenzooxazol-2-yl)-
benzoic acid (71%); (c) Tf2O, 2,6-lutidine (60%); (d) ArB(OH)2, K2CO3, 1,10-bis(di t-butylphosphino)ferrocene palladium dichloride (58–89%); (e) 10% Pd/C, H2 (1 atm), MeOH,
THF (31–77%).

R. F. Sweis et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2597–2600
2599
R
Cl
R
O
a
Cl
b, c
d
O
H2N
NH
(
R
) or ( )
S
N
Cl
O
N
Me
O
O
Cl
5
4
+
O
O
N
CO2H
NC
O
R2
N
H
e, f
O
O
N
N
N
O
R1
6-7
NC
Scheme 2. Synthesis of benzoxazoles 6–8. Reagents and conditions: (a) potassium cyanate, water, reflux (48–60%); (b) cesium carbonate, Pd2dba3, Xanthphos, 2-bromo-5-
chloropyridine, dioxane, 80 °C (54–61%); (c) sodium azide, DMF (88–96%); (d) triphenylphosphine, THF/H2O (84–89%); (e) oxalyl chloride, DMF, di-iso propylethylamine (78–
86%); (f) ArB(OH)2, K2CO3, 1,10-bis(di-t-butylphosphino)ferrocene palladium dichloride (48–93%).
Table 1
Table 2
SAR of the aryl-cyclohexyl class of benzoxazole CETP inhibitors
SAR analysis of racemic aryl-oxazolidinones
O
O
H
NH
N
H
O
O
O
N
N
N
N
O
R
R
NC
NC
Entry
R
Config.
IC50 (nM)
IKr IC50 (MK499) (nM)
Entry
1
R
IC50 (nM)
19,160
IKr IC50 (nM) (MK499)
>30,000
Log D
1
2
3
4
5
6
7
8
4-t-Bu
4-t-Bu
4-CF3
4-CF3
3-CF3
3-CF3
3-CH3
3-CH3
cis
trans
cis
trans
cis
trans
cis
trans
cis
trans
trans
37
38
134
40
78
100
145
133
1038
84
>30,000
>30,000
>30,000
>30,000
2372
>30,000
11,330
>30,000
>30,000
>30,000
>30,000
3a
3b
3c
3d
3e
3f
3g
3h
3i
Cl
5.8
6a
6b
2
3
2342
829
>30,000
1188
5.4
5.1
NC
6c
9
10
11
3,5-bis-CF3
3,5-bis-CF3
4-Cl
H2NOC
3j
3k
4
5
–
>30,000
12,650
4.8
6.2
6d
6e
77
955
S
Me
i-PrO
likelihood of success via this approach (Table 2). Upon elaborating
the terminus with an additional aryl group, potency began to in-
crease to the sub-micromolar range (entries 3 and 5) although
slightly polar groups such as the primary amide 6d were not toler-
ated. Modifying the aryl group to resemble the substituent pattern
found in our initial potent compound (1a) led to a series of potent
compounds: 6f, 6g, and 6h (entries 6–8). In addition, hERG binding
was minimal for most of these compounds. These encouraging re-
sults, coupled with the slightly reduced log D values of this class of
compounds (5.1–6.9) led to further evaluation of the individual
enantiomers of the more potent compounds.
6
171
>30,000
6.9
6f
Me
i-PrO
F3CO
7
8
124
67
>30,000
>30,000
6.6
6.4
6g
6h
F
The enantiomers of the 2-isopropoxy-5-fluoro analog clearly
showed a preference for R-configuration in our primary assay
(Table 3, entries 1–2). This configuration was also preferred for
the 2-trifluoromethoxy analog (entries 3–4). The potent configura-
tions of both also showed reasonable response on our mouse PD
assay: raising HDLc by 24 and 19 mg/dl, respectively. Finally, the
methyl-substituted oxazolidinone was evaluated with the 2-triflu-
oromethoxy substitution pattern (entries 5 and 6). The addition of
the branching methyl group enhanced potency >2-fold and
resulted in a much more robust HDLc elevation of 33 and

2600
R. F. Sweis et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2597–2600
Table 3
Analysis of enantiopure aryl-oxazolidinones for CETP inhibition and HDLc raising efficacy
O
R2
N
H
O
O
N
N
N
O
R1
NC
Entry
R1
R2
Config.
IC50 (nM)
IKr IC50 (nM) (MK499)
DHDLc mg/dl
1
2
H
H
R
S
66
154
>30,000
>30,000
7a
7b
i-PrO
24
F
19
3
4
H
H
R
S
46
140
>30,000
3580
7c
7d
F3CO
F3CO
5
6
Me
Me
R
S
22
18
>30,000
>30,000
33
27
7e
7f
van, L. S. I.; Burgess, L.; Evans, G. W.; Kuivenhoven, J. A.; Barter, P. J.; Revkin, J.
H.; Grobbee, D. E.; Riley, W. A.; Grobbee, D. E.; Shear, C. L.; Duggan, W. T.; Bots,
M. L. N. Engl. J. Med. 2007, 356, 1620–1630; (i) Nissen, S. E.; Tardif, J.-C.;
Nicholls, S. J.; Revkin, J. H.; Shear, C. L.; Duggan, W. T.; Ruzyllo, W.; Bachinsky,
W. B.; Lasala, G. P.; Tuzcu, E. M. N. Engl. J. Med. 2007, 356, 1304–1316; (j) Bots,
M. L.; Visseren, F. L.; Evans, G. W.; Riley, W. A.; Revkin, J. H.; Tegeler, C. H.;
Shear, C. L.; Duggan, W. T.; Vicari, R. M.; Grobbee, D. E.; Kastelein, J. J. Lancet
2007, 370, 153–160.
27 mg/dl, respectively. In addition, the hERG binding was also
attenuated for both enantiomers.
In summary, we have highlighted a dual strategy to address the
issue of hERG binding that plagued our benchmark compound 1a.
By replacing the piperidine moiety with a cyclohexyl group, hERG
binding was successfully attenuated. The most potent compounds
from a brief survey of analogs showed only modest efficacy in
HDLc raising (17 and 18 mg/dl for compounds 3a and 3b) and pos-
sessed high log D values (7.1–7.3). Replacement of the piperidine
moiety with an oxazolidinone also led to a successful attenuation
of hERG binding. SAR optimization resembled that for the piperi-
dine class, resulting in 2-alkoxyphenyl-pyridine substituted com-
pounds 7a–7d which were all potent inhibitors of CETP. The
potency was generally enhanced for the R enantiomer, and both
7a and 7c led to an increase in HDLc in our mouse PD assay. Finally,
methyl substitution on the oxazolidinone led to improved potency
in compounds 7e and 7f (22 and 18 nM), robust HDLc-raising effi-
cacy (33 and 27 mg/dl, respectively), and no hERG liability. Further
optimization of this class of compounds in the context of CETP
inhibition will be reported in due course.
5. Sweis, R. F.; Hunt, J. A.; Kallashi, F.; Hammond, M. L.; Chen, Y.; Eveland, S. S.;
Guo, Q.; Hyland, S. A.; Milot, D. P.; Cumiskey, A.-M.; Latham, M.; Rosa, R.;
Peterson, L.; Sparrow, C. P.; Wright, S. D.; Anderson, M. S.; Sinclair, P. J. Bioorg.
Med. Chem. Lett. 2010. doi:10.1016/j.bmcl.2010.11.090; (b) Kallashi, F.; Ali, A.;
Kim, D.; Kowalchick, J.; Park, Y. J.; Hunt, J.; Smith, C. J.; Hammond, M. L.;
Pivnichny, J. V.; Tong, X.; Xu, S. S.; Anderson, M. S.; Chen, Y.; Eveland, S. S.; Guo,
Q.; Hyland, S. A.; Milot, D. P.; Cumisky, A.; Latham, M.; Rosa, R.; Peterson, L.;
Sparrow, C. P.; Wright, S. D.; Sinclair, P. Bioorg. Med. Chem. Lett. 2011, 21, 558;
(c) Smith, C. J.; Ali, A.; Chen, L.; Hammond, M. L.; Anderson, M.; Chen, Y.;
Eveland, S. S.; Guo, Q.; Hyland, S.; Milot, D. P.; Sparrow, C. P.; Wright, S. D.;
Sinclair, P. J. Biorg. Med. Chem. Lett. 2010, 20, 346; (d) Hunt, J. A.; Gonzalez, S.;
Kallashi, F.; Hammond, M. L.; Pivnichny, J. V.; Tong, X.; Xu, S. S.; Anderson, M.
S.; Chen, Y.; Eveland, S. S.; Guo, Q.; Hyland, S. A.; Milot, D.; Sparrow, C. P.;
Wright, S. D.; Sinclair, P. J. Biorg. Med. Chem. Lett. 2010, 20, 1019.
6. All compounds reported were evaluated in our transgenic mouse PD model at
10 mpk.
7.
I
Kr binding data was obtained by measuring displacement of [35S]-radiolabeled
MK499 from HEK cells stably expressing the hERG gene: (a) Lynch, J. J., Jr.;
Wallace, A. A.; Stupienski, R. F., III; Baskin, E. P.; Beare, C. M.; Appleby, S. D.;
Salata, J. J.; Jurkiewicz, N. K.; Sanguinetti, M. C.; Stein, R. B.; Gehret, J. R.;
Kothstein, T.; Claremon, D. A.; Elliott, J. M.; Butcher, J. W.; Remy, D. C.; Baldwin,
J. J. J. Pharmacol. Exp. Ther. 1994, 269, 541–553; (b) Butcher, J. W.; Claremon, D.
A.; Connolly, T. M.; Dean, D. C.; Karczewski, J.; Koblan, K. S.; Kostura, M. J.;
Liverton, N. J.; Melillo, D. G. Radioligand and Binding Assay. World Patent
Application WO 02/05860, 2002.
References and notes
1. Singh, I. M.; Shishehbor, M. H.; Ansell, B. J. J. Am. Med. Assoc. 2007, 298, 786.
2. Gordon, D. J.; Rifkind, B. M. N. Eng. J. Med. 1989, 321, 1311.
3. (a) Sikorski, J. A. J. Med. Chem. 2006, 49, 1; (b) Tall, A. R. N. Engl. J. Med. 2007,
356, 1362; (c) Hunt, J. A.; Lu, Z. Curr. Top. Med. Chem. 2009, 9, 419.
8. All log D values are calculated by HPLC at pH 7.3. The determined log D for
compound 1a was 7.2.
4. (a) de Grooth, G. J.; Kuivenhoven, J. A.; Stalenhoef, A. F. H.; de Graaf, J.;
Zwinderman, A. H.; Posma, J. L.; van Tol, A.; Kastelein, J. J. P. Circulation 2002,
105, 2159–2165; (b) Kuivenhoven, J. A.; de Grooth, G. J.; Kawamura, H.; Klerkx,
A. H.; Wilhelm, F.; Trip, M. D.; Kastelein, J. J. P. Am. J Cardiol. 2005, 95, 1085–
1088; (c) Clark, R. W. Curr. Opin. Pharmacol. 2006, 6, 162–168; (d) Brousseau, M.
E.; Schaefer, E. J.; Wolfe, M. L.; Bloedon, L. T.; Digenio, A. G.; Clark, R. W.;
Mancuso, J. P.; Rader, D. J. N. Engl. J. Med. 2004, 350, 1505–1515; (e) Davidson,
M. H.; McKenney, J. M.; Shear, C. L.; Revkin, J. H. J. Am. Coll. Cardiol. 2006, 48,
1774–1781; (f) McKenney, J. M.; Davidson, M. H.; Shear, C. L.; Revkin, J. H. J. Am.
Coll. Cardiol. 2006, 48, 1782–1790; (g) Barter, P. J.; Caulfield, M.; Eriksson, M.;
Grundy, S. M.; Kastelein, J. J. P.; Komajda, M.; Lopez-Sendon, J.; Mosca, L.;
Tardif, J.-C.; Waters, D. D.; Shear, C. L.; Revkin, J. H.; Buhr, K. A.; Fisher, M. R.;
Tall, A. R.; Brewer, B. N. Engl. J. Med. 2007, 357, 2109–2122; (h) Kastelein, J. J. P.;
9. Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G. J. Med. Chem. 2006, 49, 5029.
10. Waring, M. J.; Johnstone, C. Biorg. Med. Chem. Lett. 2007, 17, 1759.
11. Krishna, R.; Anderson, M. S.; Bergman, A. J.; Jin, B.; Fallon, M.; Cote, J.; Rosko, K.;
Chavez-Eng, C.; Lutz, R.; Bloomfield, D. M.; Gutierrez, M.; Doherty, J.;
Bieberdorf, F.; Chodakewitz, J.; Gottesdiener, K. M.; Wagner, J. A. Lancet 2007,
370, 1907.
12. Schierle Arndt, K.; Kolter, D.; Danielmeier, K.; Steckhan, E. Eur. J. Org. Chem.
2001, 13, 2425.
13. Yin, J.; Buchwald, S. L. Org. Lett. 2000, 2, 1101.
14. Incorporation of the 2-isopropxy-5-fluorophenyl group found in compound 1a
led to only micromolar potency. This result suggested distinct SAR for the
cyclohexyl series of compounds.