2-Arylbenzoxazoles as CETP inhibitors: Substitution and modification of the α-alkoxyamide moiety

Article abstract of DOI:10.1016/j.bmcl.2009.12.046

The development of a series of 2-arylbenzoxazole α-alkoxyamide and β-alkoxyamine inhibitors of cholesteryl ester transfer protein (CETP) is described. Highly fluorinated α-alkoxyamides proved to be potent inhibitors of CETP in vitro, and the highly fluorinated 2-arylbenzoxazole β-alkoxyamine 4 showed a desirable combination of in vitro potency (IC₅₀ = 151 nM) and oral bioavailability in the mouse.

Full text of DOI:10.1016/j.bmcl.2009.12.046

Bioorganic & Medicinal Chemistry Letters 20 (2010) 1019–1022
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
2-Arylbenzoxazoles as CETP inhibitors: Substitution and modification of the
a-alkoxyamide moiety
*
Julianne A. Hunt , Silvia Gonzalez, Florida Kallashi, Milton L. Hammond, James V. Pivnichny, Xinchun Tong,
Suoyu S. Xu, Matt S. Anderson, Ying Chen, Suzanne S. Eveland, Qiu Guo, Sheryl A. Hyland, Denise P. Milot,
Carl P. Sparrow, Samuel D. Wright, Peter J. Sinclair
Merck Research Laboratories, Rahway, NJ 07065, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The development of a series of 2-arylbenzoxazole
lesteryl ester transfer protein (CETP) is described. Highly fluorinated
inhibitors of CETP in vitro, and the highly fluorinated 2-arylbenzoxazole b-alkoxyamine 4 showed a desir-
able combination of in vitro potency (IC50 = 151 nM) and oral bioavailability in the mouse.
Ó 2009 Elsevier Ltd. All rights reserved.
a
-alkoxyamide and b-alkoxyamine inhibitors of cho-
Received 28 October 2009
Revised 6 December 2009
Accepted 10 December 2009
Available online 14 December 2009
a
-alkoxyamides proved to be potent
Keywords:
CETP
Cholesteryl ester transfer protein
High-density lipoprotein cholesterol
Low-density lipoprotien cholesterol
Reverse cholesterol transport
2-Arylbenzoxazole
Cardiovascular disease (CVD) is the leading cause of morbidity
and mortality in developed countries. Therapies targeting eleva-
tion of high-density lipoprotein cholesterol (HDL-C) levels may re-
duce cardiovascular risk by slowing the development of
atherosclerosis.1 Epidemiological studies indicate that each 1 mg/
dl increase in plasma HDL-C concentration is associated with a
6% decrease in the risk of death from CVD, independent of low-
density lipoprotein cholesterol (LDL-C) concentration.2
One approach to treating CVD via elevation of HDL-C levels is to
inhibit cholesteryl ester transfer protein (CETP).3 CETP, a hydro-
phobic plasma glycoprotein that is physically associated with lipo-
protein particles, facilitates the transfer of cholesteryl esters and
triglycerides between HDL and the apo B-containing lipoproteins.
Since the net effect of CETP activity is a reduction in plasma
HDL-C levels, inhibition of CETP activity is expected to result in in-
creased HDL-C levels. In turn, elevation of HDL-C is expected to re-
duce cardiovascular risk.
In this Letter, we describe optimization of the
moiety of the screening lead 1, resulting in identification of potent
alkyl-substituted analogs such as 3. However, both the -aryloxya-
mides and the b-alkyloxyamides showed extremely limited bio-
availability. Further optimization of 3 was undertaken to improve
the pharmacokinetic properties of this series, leading to the
a-aryloxyamide
a
replacement of the central
as in compound 4, a potent and orally bioavailable CETP inhibitor.
We began our optimization of the -aryloxyamide moiety of 1
a-alkoxyamide with a b-alkoxyamine,
a
with a straightforward survey of replacements for the o-cresol sub-
structure (Table 1). While removal of the o-methyl substituent re-
sulted in only a fourfold loss of inhibitor potency (5), transposition
of the methyl substituent to the meta (1m) or para (1p) positions
OH
O
O
O
O
O
N
O
N
Our co-workers have recently communicated the discovery of a
series of 2-arylbenzoxazole inhibitors of CETP.4 Structure–activity
optimization of the substituents on the benzoxazole moiety of
screening lead 1 resulted in the identification of compound 2 as
a potent CETP inhibitor (see Fig. 1).4
NH
NH
Cl
NC
NC
1, CETP IC50=1107 nM
2, CETP IC50=28 nM
CF3
CF3
CF3
CF3
O
O
O
O
O
N
NH
NH
N
NC
3, CETP IC50=44 nM
4, CETP IC50=151 nM
* Corresponding author. Tel.: +1 732 594 0463; fax: +1 732 594 9473.
E-mail addresses: julianne_hunt@merck.com, huntjuli1205@gmail.com (J.A.
Hunt).
Figure 1. 2-Arylbenzoxazole CETP inhibitors.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.12.046

1020
J. A. Hunt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1019–1022
Table 1
poor activity in our in vitro assay, as did methylation of the amide
nitrogen (23), or of the carbon adjacent to the amide carbonyl (24).
Inhibition of CETP activity by compounds of the formula
R
Only a few of the more conservative linker variations were fairly
O
O
neutral in their effect on inhibitor potency. Replacement of the
oxygen with a nitrogen (14) led to only a twofold decrease in po-
tency, as compared to the analogous -aryloxyamide 5. Similarly,
addition of a methylene unit between the -oxygen and the phenyl
a-
O
N
NH
Cl
a
a
a
Compd
R
IC50 (nM)
Meta
ring (15) resulted in less than a twofold decrease in potency as
Ortho
Para
compared to 5.
1
5
6
Me
H
Et
1107
4230
2587
19,310
>3 Â 104
Inspired by the nearly equivalent potencies of the phenyl (5)
and benzyl (15) analogs in Table 2, we next turned to an explora-
nd
Nd
7
8
9
10
11
12
13
Pr
CF3
Cl
CN
F
OMe
NO2
2700
6390
2150
nd
Nd
810
tion of benzyl and other alkyl substituents on the a-oxygen. At this
1773
1248
>3 Â 104
1994
12,880
26,490
point in our research, our co-workers had demonstrated the dra-
matically increased potency of the 5-cyanobenzoxazoles as com-
pared to the analogous 5-chloro compounds (cf. compound 25 vs
compound 1),4 so we continued the remainder of our SAR explora-
tion in the 5-cyano series (see Table 3).
6810
>3 Â 104
1774
4570
2206
>3 Â 104
>3 Â 104
>3 Â 104
4396
a
Assay conditions have been described.5
Although 26, the benzyl analog of 25 was nearly 10-fold less
potent, replacement of the methyl substituent on the phenyl ring
with a trifluoromethyl (27) moved the potency back in the right
direction. Similarly, 28, the ethyl analog of 25 was tenfold less
potent, but elaboration to the isopropyl (29) and tert-butyl (30)
analogs restored potency. Combining the branched-alkyl strategy
with the trifluoromethyl strategy led to the series of compounds
shown at the bottom of Table 3, with the 1,1-bistrifluoromethyl-
ethyl analog 3 emerging as the most potent inhibitor in the
series.
Unfortunately, although we had significantly improved the
in vitro potency of our lead compound 1, we were still challenged
by the poor physicochemical properties of the 2-arylbenzoxazole
amide series. For example, the bioavailability (in mouse) of com-
pound 31 was less than 4%. Not surprisingly, even the most potent
inhibitors among the compounds shown in Tables 1–3 lacked any
appreciable activity when we tested them in vivo.6
caused a significant decline in potency. Larger alkyl substituents (6
and 7) were tolerated, but did not improve potency. Similarly, tri-
fluoromethyl (8), halo (9 and 11), nitrile (10), methoxy (12), and ni-
tro (13) substituents were either detrimental to potency or neutral.
Not only were we unable to improve the potency of our lead via
simple modification of the o-cresol substructure, we were also un-
able to determine an optimal potency-enhancing position for sub-
stituents on the phenyl ring, although it should be noted that a
number of these compounds suffered from a severe lack of aqueous
solubility, so it is possible that some of the inhibition constants
shown in Table 1 reflect this problem rather than the true inhibitor
potency.
Having encountered problems with compound solubility in our
initial survey of phenyl substituents, we turned next to replacing
the
a-alkoxyamide, or ‘linker’, region of the molecule, reasoning
that the amide bond might be contributing to the poor physico-
chemical properties of our lead series. Unfortunately, most of the
linker variations we surveyed were not successful; in fact, every
variation resulted in an analog that was a less potent inhibitor than
Table 3
Inhibition of CETP activity by compounds of the formula
O
O R
O
N
NH
the lead compound 1. As shown in Table 2, reversal of the
oxyamide to the carbamate (17), replacement of the amide with
a urea (18), replacement of the -oxygen with a carbon (19) or a
a-alk-
NC
a
a
Compd
R
IC50 (nM)
sulfur (20), reversal of the amide (21), and replacement of the
amide with a sulfonamide (22) all resulted in compounds with
25
130
1182
505
Table 2
26
27
Inhibition of CETP activity by compounds of the formula
R
CF3
X
X
X
O
N
X
28
29
1375
613
Cl
a
Compd
R
X–X–X–X
IC50 (nM)
1
5
Me
H
H
H
H
H
H
H
H
Me
H
Me
H
NH–C(O)–CH2–O
NH–C(O)–CH2–O
NH–C(O)–CH2–NH
NH–C(O)–CH2–O–CH2
NH–C(O)–CH2–NH–CH2
NH–C(O)–O–CH2
NH–C(O)–NH–CH2
NH–C(O)–CH2–CH2
NH–C(O)–CH2–S
C(O)–NH–CH2
NH–SO2–CH2
NMe–C(O)–CH2–O
NH–C(O)–CH(Me)–O
1107
4230
8132
6537
18,080
>3 Â 104
>3 Â 104
>3 Â 104
>3 Â 104
>3 Â 104
>3 Â 104
>3 Â 104
>3 Â 104
30
31
32
132
94
14
15
16
17
18
19
20
21
22
23
24
CF3
CF3
CF3
51
CF3
3
44
CF3
CF3
CF3
F3C
33
151
Assay conditions have been described.5
Assay conditions have been described.5
a
a

J. A. Hunt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1019–1022
1021
14
Early in our lead optimization efforts, we had prepared 34, a b-
aryloxyamine (or ‘reduced amide’) analog of our lead compound 1,
but we had been discouraged by its poor in vitro activity (Table 4).
However, in light of the improved potency of the compounds
shown in Table 3, we decided to revisit the reduced amide series.
As shown in Table 4, with optimized substituents in place on either
end of the 2-arylbenzoxazole scaffold, we finally succeeded in
identifying a compound (4) with both in vitro potency and accept-
able pharmacokinetic properties (for compound 4: AUC-
Ph-NH2,
DIPEA,
35%
O
Br
Ph-CH2NH2,
DIPEA
Ph-CH2OH,
KOtBu
O
N
16
NH
15
28%
Cl
16%
37
Niv = 1.2 lM h kg/mg, Cl = 31 ml/min/kg, t1/2 = 7.3 h).
18
17
As shown in Figure 2, one-step assembly7 of the 2-arylbenzox-
azole amine 36 followed by formation of the key bromoacetamide
Ph-CH2OCOCl,
pyr, 35%
Ph-CH2NCO,
pyr, 84%
intermediate 37 allowed rapid access to the
a-aryloxyamide ana-
logs shown in Table 1 via SN2 displacement of the
a
-bromine of 37.
Ph-CH2CH2COCl,
Ph-OCH(CH3)COCl,
As shown in Figure 3, nearly all of the linker variants in Table 2
were prepared from either aniline 36 or bromoacetamide 37 using
standard synthetic transformations and commercially available
starting materials. Reverse amide 21 was prepared from 2-methyl-
benzylamine and 2-arylbenzoxazole 428 in 74% yield. Finally, ter-
tiary amide 23 was prepared in 95% yield by deprotonation of
compound 1 with lithium hexamethyldisilazide and subsequent
alkylation with methyl iodide.
O
N
pyr
DIPEA
19
24
NH2
84%
96%
Cl
36
Ph-SCH2COCl,
DIPEA, <5%
Ph-CH2SO2Cl,
DIPEA, 45%
20
22
The key 5-cyano moiety in 2-arylbenzoxazole amine 38 proved
labile to the conditions shown for formation of 36 (Fig. 2), so a
milder, two-step protocol was developed for the assembly of this
arylbenzoxazole core (Fig. 4). Similarly, the alkyl substituents
shown in Table 3 were not generally amenable to the SN2 reaction
Figure 3. Synthesis of 2-arylbenzoxazoles 14–20, 22, and 24.
Pd/C
NH4HCO2
1. Pd/C, NH4HCO2
2.
conditions shown in Figure 2, so
a-alkyloxy acid chlorides 39 were
OH
O
N
ClOC Ph NO2
NO2
prepared from -bromoacetic acid and the appropriate alcohols.
a
3. pTsOH, toluene,
Dean-Stark trap
NC
NO2
NC
70%
The acid chlorides 39 were then coupled with aniline 38 under
standard conditions.
51%
Synthesis of the b-alkoxyamines shown in Table 4 required a
different arylbenzoxazole component, the bromide 40, which was
O
39
O
OR
OR
O
O
N
Cl
NH2
NH
N
DIPEA, CH2Cl2
NC
NC
38
Table 4
Inhibition of CETP activity by compounds of the formula
1. ROH, NaH, THF
O
R
O
O
O
N
Br
OR
HO
Cl
NH
2. (COCl)2, DMF,
CH2Cl2
39
X
F%b
nd
30
20
a
Compd
X
R
IC50 (nM)
75,900
890
Figure 4. Synthesis of 2-arylbenzoxazole a-alkyloxyamides 3 and 25–33.
34
35
4
H
prepared using the same two-step sequence shown in Figure 5
for aniline 38. Palladium-catalyzed amination of 40 provided the
desired 2-arylbenzoxazole b-alkoxyamines, as shown for
CN
CN
CF3
CF3
151
a
Assay conditions have been described.5
In mouse.
1. Pd/C, NH4HCO2
b
OH
O
N
2.
ClOC Ph Br
Br
3. pTsOH, toluene,
Dean-Stark trap
NC
NO2
NC
40
47%
OH
NH2
PPA, 150ºC
O
N
CF3
CF3
CF3
CF3
O
NH2
O
H2N
Cl
NH2
HO2C
O
Cl
41
O
36
NH
N
O
Br
Pd2(dba)3, BINAP, NaOtBu,
toluene, 80C
NC
Br
4
Br
O
N
61%
NH
DIPEA, CH2Cl2
55% (two steps)
Cl
37
R
CF3
CF3
1.
, K2CO3, DMF
R
O
HO
O
O
CF3
HO
O
Br
O
N
H2N
H2N
NH
2. LiAlH4, Et2O
30%
CF3
41
Cl
K2CO3, DMF
Figure 2. Synthesis of 2-arylbenzoxazole
a-aryloxyamides 1 and 5–13.
Figure 5. Synthesis of 2-arylbenzoxazole b-alkyloxyamine 4.

1022
J. A. Hunt et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1019–1022
compound 4. While some b-alkoxyamines were commercially
available, compound 41 was prepared from -bromoacetamide
and bistrifluoromethyl-tert-butanol via nucleophilic substitution
of the -bromide followed by reduction of the amide to the amine
(Fig. 5).
Our group4 and others9 have shown that 2-arylbenzoxazole
References and notes
a
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. Barter, P. J.; Kastelein, J. J. P. J. Am. Coll. Cardiol. 2006, 47, 492.
4. Smith, C. J.; Ali, A.; Chen, L.; Hammond, M. L.; Anderson, M. S.; Chen, Y.; Eveland,
S. S.; Guo, Q.; Hyland, S. A.; Milot, D. P.; Sparrow, C. P.; Wright, S. B.; Sinclair, P. J.
Bioorg. Med. Chem. Lett. 2009. Available online 29-Oct-09.
5. Eveland, S. S.; Milot, D. P.; Guo, Q.; Chen, Y.; Hyland, S. A.; Peterson, L. B.;
Jezequel-Sur, S.; O’Donnell, G. T.; Zuck, P. D.; Ferrer, M.; Strulovici, B.; Wagner, J.
A.; Tanaka, W. K.; Hilliard, D. A.; Laterza, O.; Wright, S. D.; Sparrow, C. P.;
Anderson, M. S. Anal. Biochem. 2007, 368.
a
a
-
aryloxyamides such as 1 and 2 can inhibit CETP activity in vitro.
However, the poor physicochemical properties of most compounds
in this structure class represent a barrier to the development of
analogs with in vivo activity. By reducing the central amide to
the corresponding amine and replacing the aryloxy moiety with
highly fluorinated alkyloxy substructures, we have identified
new, potent CETP inhibitors with significantly improved pharma-
cokinetic properties. Unfortunately, despite the in vitro potency
and moderate bioavailability of 2-arylbenzoxazole b-alkoxyamine
4, this compound (and others in the same structure class) did not
6. Data not shown.
7. Hein, D. W.; Alheim, R. J.; Leavitt, J. J. J. Am. Chem. Soc. 1957, 79, 427.
8.
Ph CO2Me
ClOC
1.
2. pTsOH, xylenes,
Dean-Stark trap
OH
O
N
COCl
1. LiOH
Cl
NH2
Cl
2. (COCl)2, DMF
42
show activity in our in vivo assays.6 Nevertheless, the
a-alkyloxya-
70%
mide and b-alkyloxyamine compounds shown in Tables 3 and 4
serve as leads for continuing development of this structure class.
We will report further modifications of this class of CETP inhibitors
aimed at improving in vivo activity in due course.
9. Harikrishnan, L. S.; Kamau, M. G.; Herpin, T. F.; Morton, G. C.; Liu, Y.; Cooper, C.
B.; Salvato, M. E.; Qiao, J. X.; Wang, T. C.; Adam, L. P.; Taylor, D. S.; Chen, A. Y. A.;
Yin, X.; Seethala, R.; Peterson, T. L.; Nirschl, D. S.; Miller, A. V.; Weigelt, C. A.;
Appiah, K. K.; O’Connell, J. C.; Lawrence, R. M. Bioorg. Med. Chem. Lett. 2008, 18,
2640.