Imidazolylpyrimidine based CXCR2 chemokine receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.02.028

An imidazolylpyrimidine was identified in a CXCR2 chemokine receptor antagonist screen and was optimized for potency, in vitro metabolic stability, and oral bioavailability. It was found that subtle structural modification within the series affected the o

Full text of DOI:10.1016/j.bmcl.2006.02.028

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2724–2728
Imidazolylpyrimidine based CXCR2 chemokine
receptor antagonists
Koc-Kan Ho,a,* Douglas S. Auld,a Adolph C. Bohnstedt,a Paolo Conti,b Wim Dokter,b
Shawn Erickson,a Daming Feng,a Jim Inglese,a Celia Kingsbury,a Steven G. Kultgen,a
Rong-Qiang Liu,a Christopher M. Masterson,a Michael Ohlmeyer,a Yajing Rong,a
Martijn Rooseboom,b Andrew Roughton,a Philippe Samama,a Martin-Jan Smit,b
Ellen Son,a Jaap van der Louw,b Gerard Vogel,b Maria Webb,a
Jac Wijkmansb and Ming Youa
aPharmacopeia Drug Discovery, Inc., PO Box 5350, Princeton, NJ 08543-5350, USA
bNV Organon, Molenstraat 110, PO Box 20, 5340 BH Oss, The Netherlands
Received 13 January 2006; revised 3 February 2006; accepted 6 February 2006
Available online 15 March 2006
Abstract—An imidazolylpyrimidine was identified in a CXCR2 chemokine receptor antagonist screen and was optimized for poten-
cy, in vitro metabolic stability, and oral bioavailability. It was found that subtle structural modification within the series affected the
oral bioavailability. Potent and orally available CXCR2 antagonists are herein reported.
Ó 2006 Elsevier Ltd. All rights reserved.
Chemokines,1 by interaction with their respective G-
protein-coupled receptor (GPCR), have a profound
influence on leukocyte migration and upregulation of
adhesion receptors. They are predominantly classified
into two large families, CXC and CC, based on their
amino acid sequence. The actions of CXC chemokines
are mediated through six cell surface receptors, CXCR1
to CXCR6.
pharmacokinetic properties for proof-of-principle effica-
cy studies.
A binding assay based on membranes prepared from
CHO cell transfected with human CXCR2 receptor
and labeled IL-8 was used to screen the Pharmacopeia
libraries (ꢀ500,000 compound, 1996). The initially ac-
tive compound identified, imidazolylpyrimidine 1, had
a Ki of 60 nM and it served as the starting point for
an optimization program. Compound 1 was divided into
four domains: the pyrimidine core and three pendant
groups R1, R2, and R3. Presented herein is the optimiza-
tion of the R1 and R2 domains.
Ligands for the CXCR2 receptor, for example, interleu-
kin-8 (IL-8, CXCL8), induce migration of neutrophils,
monocytes, and T lymphocytes toward a site of inflam-
mation. As such, the CXCR2 receptor and IL-8 are
thought to play a crucial role in inflammatory diseases.
Therefore, antagonism of CXCR2 could be beneficial to
these inflammatory diseases.2
R3
R2
N
O
H
N
S
N
N
N
O
H
N
The goal was to identify small molecule antagonists of
human CXCR2 that possessed suitable potency and
R1
1
Ki = 60 nM
Keywords: Chemokine; CXCR2; Pyrimidine; Microsomal stability;
Oral bioavailability.
Preparation of the pyrimidine analogs is exemplified for
compound 40 (Scheme 1). 2,4-Dichloro-6-propylpyrimi-
dine (4) was prepared from reaction of 2,6-dichloropyr-
*
Corresponding author. Tel.: +1 609 452 3695; fax: +1 609 655
4187; e-mail: kanho@pcop.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.02.028

K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2724–2728
2725
N
Cl
Cl
Cl
LiN
N
Cl
Cl
N
Cl
nPrLi, Et2O
DDQ
N
N
4
2
3
N
N
Cl
N
N
OCF3
F3CO
NH
N
+
6
N
DMF, DBU
N
N
Cl
F3CO
N
7
1. Isomer separation
N
O
H
N
N
N
F3CO
N
H
O
O
2.
N
H2N
N
O
H
,
8
40
DIEA, DMSO, 80 o
Scheme 1. Preparation of pyrimidine analogs.
C
imidine (2) with n-propyllithium followed by oxidation
of the intermediate (3) with 2,3-dichloro-5,6-dicyano-
1,4-benzoquinone (DDQ).3 Reaction of 4 with 4-trif-
luoromethoxyphenylimidazole (5) yielded a mixture
of two isomers which were separable by column chroma-
tography. The preferred 2-substituted pyrimidine isomer,
7, was reacted with a leucine derivative 8 to afford
product 40. The triazole analog 27 was prepared in an
analogous manner except substituting 1,2,4-triazole
for 5.
Analogs 28 and 29, the pyridylpyrimidines, were pre-
pared by a different route (Scheme 3). Condensation of
amidate 15 with b-ketoester 16 yielded a pyrimidinone
17, which was converted to the chloropyrimidine 18 with
phosphorus oxychloride. Displacement of the chloride
with the leucine derivative 8 gave the final products
(28 and 29).
Although imidazolylpyrimidine 1 possessed good in vitro
potency, it lacked stability with only 26% remaining
after 30 min in the presence of rat liver microsomes
(see Table 1) and had very poor bioavailability of much
less than 1% in rat. The lack of microsomal stability was
reflected with the relatively high clearance (1750 mL/h/
kg) observed in vivo after iv dosing. The highly lipophil-
ic n-octyl thioether moiety is prone to oxidation by the
liver enzymes and likely to contribute to the instability
of the compound. Thus, initial optimization efforts were
focused on this domain (R1) (Table 1). A twofold de-
crease in potency was observed when the thiol group
was removed (19). Compounds with successive removal
of two carbons from the chain, namely, the n-hexyl and
n-butyl derivatives (20 and 21, respectively), led to a
stepwise decrease in activity. With respect to microsom-
al stability, shortening of the chain length afforded more
stable compounds. Rat liver microsomal stability was
further enhanced via the addition of a branched alkyl
group (t-butyl, compound 22), however, this stability
gain was offset by a lack of potency.
The substituted phenylimidazoles (10) were prepared via
two routes; refluxing bromoacetophenone (9) in form-
amide4 or reacting benzaldehyde (11) with p-tolylsulfo-
nyl isocyanide (TosMIC) followed by heating of
intermediate 12 in methanolic ammonia (Scheme 2).5
R4
O
N
Br
formamide, reflux
R4
NH
9
10
R4 = 4-MeO-, 4-F3CO-, 4-Cl-, 4-F-, 4-F3C-, 4-NC-, 4-O2N-, 3-F3CO-
O
O
S
N
CHO
TosMIC, KCN, EtOH
R4
O
R4
11
12
R4 = 2-F3CO-
The n-butyl analog 21 showed a good balance between
potency (1.7 lM) and microsomal stability (68%
remaining after 30 min), and thus was used as the start-
ing point for the next round of optimization—where the
focus was to optimize the R2 domain, the imidazole moi-
ety (Table 2). A marked improvement in potency was
observed with the addition of a methyl substituent to
R4
NH3, MeOH
N
NH
10
Scheme 2. Preparation of substituted phenylimidazole from bromo-
acetophenones or benzaldehyde.

2726
K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2724–2728
O
O
OEt
NH
HCl, dioxane
NH3, MeOH
N
N
N
16
CN
NH2
NaOEt, EtOH, reflux
14
15
O
H2N
N
O
N
H
H
N
O
POCl3
N
Cl
8
N
N
dimethylaniline, 125 o
C
DIEA, DMSO, 80 o
C
18
17
N
O
H
N
N
N
O
H
N
28 = 4-pyridyl
29 = 3-pyridyl
Scheme 3. Preparation of pyridylpyrimidines.
Table 1. Modification of R1 domain
Table 2. Modification of R2 domain
N
O
H
O
H
R2
N
N
N
N
N
N
O
N
H
O
H
N
N
R1
R2
Kia (nM)
RLMb (% rem)
Compound
R1
Kia (nM)
RLMb (% rem)
Compound
1
19
20
21
22
(n-Octyl)-S-
n-Octyl-
n-Hexyl-
n-Bu-
60
120
26
36
50
68
83
N
N
21
1700
68
N
290
1700
6000
t-Bu-
23
24
25
26
27
28
29
>10,000
90
27
25
13
2
a Values are means of at least two experiments with standard deviation
< 30%.
b Rat liver microsome stability was measured by incubation of com-
pound (10 lM) in enzyme (300 nM) for 30 min at 37 °C. The % of
compound remaining after 30 min was quantitated by a LC method.
Values are means of two experiments, with a standard deviation of
< 20%.
N
N
N
N
85
N
N
N
100
the 4-position of the imidazole ring (24). Conversely,
substitution at the 2-position (23) was not tolerated
and resulted in a significant loss of potency. Bulkier
groups, such as a t-butyl (25) and phenyl (26), were per-
mitted in the 4-position of the imidazole and afforded
compounds with potency similar to that of the methyl
analog 24.
N
N
>10,000
>5000
750
30
52
36
N
N
N
Replacement of the imidazole ring with a triazole (27) or
4-pyridine (28) group provided inactive compounds.
However, a twofold improvement in potency was seen
for the 3-pyridyl analog 29, indicating that the nitrogen
atom placement within the heteroaromatic R2 group
may be important for potency. Although many of the
R2 modifications yielded compounds that were more po-
tent than 21, all exhibited lower microsomal stability.
See Table 1 for footnote a and b.
ic nature of the aryl ring altered the potency and in vitro
metabolic profile of the series (Table 3). In general, all
modifications at the 4-position (termed R4) yielded com-
pounds with improved potency over 26 (R4 = H).
Further optimization of compound 26, the 4-phenyl
imidazole analog, illustrated that changing the electron-

K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2724–2728
Table 5. Positional isomers of the trifluoromethoxy substituent
2727
Table 3. Substituted phenyl analogs
N
N
O
O
R4
H
H
N
N
N
N
N
N
N
O
F3CO
N
O
H
H
N
N
Compound
R4
Kia (nM)
RLMb (% rem)
Compound
Position of
CF3O-
Kia
RLMb
(% rem)
% F
(rat)
(nM)
26
30
31
32
33
34
35
36
37
38
39
H-
100
40
2
35
30
65
66
54
20
80
50
48
nd
Me-
MeO-
HO-
40
44
45
para-
meta-
ortho-
25
35
50
74
55
13
33
7
42
32
5
F3CO-
Cl-
F-
17
27
See Table 1 for footnote a and b.
37
14
F3C-
NC-
O2N-
H2N-
H, gave compound 43 that was significantly less active
and less stable to rat liver microsomes.
43
160
600
The potency of the o-, m-, and p-trifluoromethoxy posi-
tional isomers was similar (Table 5), however the ortho
analog (45) was rapidly metabolized by the microsomes
with only 13% remaining after 30 min, compared with
55% and 74% for the meta and para compounds (44
and 40, respectively).
See Table 1 for footnote a and b.
Exceptions were the nitro (38) and amino (39) analogs.
More importantly, these compounds showed substantial
improvement in microsomal stability, with compounds
32, 33, 34, and 36 having greater than 50% of the com-
pound remaining after 30 min incubation with rat liver
microsome.
Oral bioavailability data (% F) for the trifluoromethoxy-
phenyl series are shown in Tables 4 and 5. As a general-
ization in this series, compounds with good in vitro
microsomal stability (>60% remaining after 30 min in
nmol/mL
10
With these more stable compounds in hand, the aliphat-
ic R1 group was re-visited with the aim to further reduce
the lipophilicity of the moiety and was exemplified by
the series based on compound 33 where carbons were
sequentially removed from R1 (Table 4). The calculated
AlogP reflects the overall change in lipophilicity
brought in by reducing the size of the R1 group.
i.v.
1
p.o.
The potency of the n-propyl analog (40) was comparable
with the n-butyl (33) derivative, however, compounds
with that of shorter R1 groups are two- to threefold less
active (compare 33 with 41 and 42). Modification of this
domain had little effect on microsomal stability.
However, complete removal of the alkyl chain, R1 =
0.1
0
2
4
6
8
time (h)
Figure 1. Pharmacokinetic profile of compound 40 in rats (n = 3) after
dosing 4 lmol/kg iv and po.
Table 4. R1 homologs
N
O
H
N
F3CO
N
N
Table 6. PK parameters of compound 40
N
O
H
N
Rat
Doga
R1
Dose (lmol/kg)
iv
po
4
4
2
Compound
R1
Kia
(nM)
RLMb
(% rem)
% F
(rat)c
AlogP
4
t1/2 (h)
Cl (mL/h/kg)
Vss (L/kg)
2.7 0.05
590 25
1.1 0.1
723
3
33
40
41
42
43
n-Bu-
n-Pr-
Et-
17
25
66
74
77
67
5
7
33
18
22
4
6.74
6.29
5.83
5.16
4.88
1.99 0.06
2.43 0.27
0.66 0.12
2.79 0.20
1.39 0.70
0.30 0.04
25 12
b
AUC (h nmol/mL)
1
b
Cmax (nmol/mL)
57
44
Me-
H-
F (%)
33
3
200
a n = 2.
b po.
See Table 1 for footnote a and b.
c Compound was dosed at 4 mg/kg iv and po.

2728
K.-K. Ho et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2724–2728
rat liver microsomes) showed reasonable oral bioavail-
ability (% F > 20). Exception was seen with compounds
33 which illustrate that although subtle structural mod-
ification may not dramatically affect metabolic stability,
they may profoundly affect oral bioavailability, as seen
in the comparison of n-butyl and n-propyl analogs (33
and 40). It may be hypothesized that the addition of
an extra carbon atom negatively affects the physico-
chemical properties of 33 thereby reducing drug absorp-
tion leading to the observed difference in oral
bioavailability.
having little effect on potency or microsomal stability.
Two compounds (40 and 42)6 were identified with both
good potency (Ki < 50 nM) and oral bioavailability (>20
% in rat).
Acknowledgments
We thank Sharon Chen, Julie Lao, and Sanjay Patel of
Pharmacopeia Drug Discovery, Inc., for assistance in
setting up the assays.
The pharmacokinetic (iv and po) profile in rats and rel-
evant PK parameters (rat and dog) of compound 40 are
shown in Figure 1 and Table 6, respectively.
References and notes
1. Ono, J. S.; Nakamura, T.; Miyazaki, M. O.; Ohbayashi,
M.; Dawson, M.; Toda, M. J. Allergy Clin. Immunol. 2003,
111, 1185.
2. D’Ambrosio, D.; Panina-Bordignon, P.; Sinigaglia, F.
J. Immunol. Methods 2003, 273, 3.
3. Harden, D. B.; Mokrosz, J. M.; Strekowski, L. J. Org.
Chem. 1988, 53, 4137.
4. Lynch, W. E., Jr.; Kurtz, D. M.; Wand, S.; Scott, R. A.
J. Am. Chem. Soc. 1994, 116, 11030.
Compound 1, obtained from the CXCR2 chemokine
receptor antagonist screening program, was optimized
for potency and in vitro metabolic stability. Metabolism
related to the bulky and lipophilic n-octyl thio group
was addressed by replacing it with smaller aliphatic moi-
eties (R1). Optimization of the imidazole domain (R2) by
introduction of the 4-(4-substituted-phenyl)imidazole
group yielded particularly potent molecules. Pharmaco-
kinetic studies performed on the series highlighted the
importance of in vitro microsomal stability, which in
this series appears to be one of the key determining fac-
tors for good oral availability. However, subtle structur-
al changes within a series, such as altering chain length,
can drastically affect pharmacokinetic profiles while
5. Horne, D. A.; Yakushijin, K.; Buechi, C. Heterocycles
1994, 39, 139.
6. Analytical data for a representative compound 40 (HCl
1
salt): H NMR (CD3OD) d 9.90 (s, 1H), 8.75 (s, 1H), 8.10
(overlapping s and d, 2H), 7.60 (d, 2H), 6.65 (s, 1H), 4.65
(m, 1H), 3.40 (m overlapped with solvent, 6H), 2.75 (t, 2H),
1.85 (m, 7H), 1.10 (m, 12H); MS (ESI) m/z 563.3 (M+H)+.