Core exploration in optimization of chemokine receptor CCR4 antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.10.091

The design, synthesis, and SAR studies of 'core' variations led to identification of novel, selective, and potent small molecule antagonist (22) of the CC chemokine receptor-4 (CCR4) with improved in vitro activity and liability profile. Compound 22 was e

Full text of DOI:10.1016/j.bmcl.2006.10.091

Bioorganic & Medicinal Chemistry Letters 17 (2007) 679–682
Core exploration in optimization of chemokine
receptor CCR4 antagonists
Ashok V. Purandare,* Honghe Wan, John E. Somerville, Christine Burke, Wayne Vaccaro,
XiaoXia Yang, Kim W. McIntyre and Michael A. Poss
Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543, USA
Received 27 September 2006; revised 26 October 2006; accepted 30 October 2006
Available online 2 November 2006
Abstract—The design, synthesis, and SAR studies of ‘core’ variations led to identification of novel, selective, and potent small
molecule antagonist (22) of the CC chemokine receptor-4 (CCR4) with improved in vitro activity and liability profile. Compound
22 was efficacious in a murine allergic inflammation model (ED50 ꢀ 10 mg/kg).
Ó 2006 Elsevier Ltd. All rights reserved.
The chemokine receptor CCR4 works together with its
ligands from the CC chemokine family, macrophage-de-
Cl
N
N
rived chemokine (MDC; CCL22), and thymus and acti-
vation regulated chemokine (TARC; CCL17), to
promote recruitment, homing, and education of activat-
ed leukocytes (mainly CD4+ Th2 lymphocytes).1 In
addition, a new ligand for CCR4 has recently been de-
scribed called CKLF1 (chemokine-like factor 1)2 whose
role in immuno-modulation is unclear. Studies3 using
monoclonal antibodies for the CCR4 receptor and its
ligand TARC have demonstrated efficacy in OVA-in-
duced murine asthma models.4 We5 and others6 have
demonstrated targeted antagonism of the CCR4 recep-
tor as a mechanism of inhibiting recruitment of activat-
ed leukocytes. These studies support the mechanism of
CCR4 antagonism as a potential therapeutic treatment
for diseases such as asthma and atopic dermatitis.
R
N
H
Cl
R
CCR4 IC50 (μM)
Compd
1
N(Et)2
0.27
N
H
N
2
0.05
N
N
H
O
Figure 1.
(
)
n
Recently, we reported5 the identification of pyrimidine
analogs (1 and 2 in Fig. 1) as potent antagonists of
CCR4, wherein we optimized the ‘wing’ or sidechain
portions. We envisaged that the basic pharmacophore
for CCR4 binding of such compounds consisted of
two ‘wings’ attached to a central ‘core’ in a defined con-
figuration (Fig. 2). Such an arrangement presented the
X
R5
R3R4N
Linker
Y= N
X= N or CH
N
R1
N
R2
Y
Figure 2.
‘wing’ groups to the desired residues in the receptor pro-
tein for favorable interactions.
We herein report our efforts to explore and further opti-
mize the ‘core’ that led to the identification of a potent
CCR4 antagonist 22 with an improved liability profile
and in vivo activity.
Keywords: Chemokine receptor CCR4; Antagonist; MDC; TARC;
Asthma; Atopic dermatitis.
*
Corresponding author. Tel.: +1 609 252 4320; fax: +1 609 252
7446; e-mail: ashok.purandare@bms.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.091

680
A. V. Purandare et al. / Bioorg. Med. Chem. Lett. 17 (2007) 679–682
Cl
H
NH
N
H
N
N
a
N
N
Cl
N
N
N
Cl
N
N
N(Et)2
Cl
N
N
N
4
7
H
H
Cl
Cl
Cl
Cl
3 CCR4 IC50 (μM)12 3.6 μM
Cl
b
Cl
N
+
N
Figure 3.
N
N
N
N
N
N
N
Cl
N
N
N
R
R
a, b
9
8
N
N
R2
R1
Cl
N
Cl
Cl
N
H
N
N
H
Cl
4 R = H
5 R = Et
6a-d
H
N
N
c
N
(Et)2N
Scheme 1. Reagents and conditions: (a) R1NH2, 1,2-dichloroethane,
rt, 75–80%; (b) R2NH2, NMP, 85 °C, 60–65%.
8
N
N
10
Cl
We first embarked on a systematic study of the ‘core’
while retaining the 2,4-dichlorobenzyl and the diamine
sidechain present in 1. Our attempt to convert the
pyrimidine to a triazine core (3) led to ꢀ15-fold loss of
activity (see Fig. 3).
Cl
N
N
c
9
(Et)2N
N
N
N
H
11
We then surveyed the purine nucleus (N-7 alkylated and
unsubstituted) while maintaining the desired 1,3-direc-
tionality. These analogs could be readily accessed from
reported intermediates7 (4–5) through sequential dis-
placement of the chlorine atoms (Scheme 1).8
Scheme 2. Reagents and conditions: (a) Ref. 8, 70%; (b) 2,4-dichlo-
robenzyl chloride, K2CO3, NMP, rt, 60–65%; (c) H2N–(CH)Me–
(CH2)3N(Et)2, NMP, 90 °C, 55–60%.
In either regioisomeric ‘wing’ arrangements, N-7 alkyl-
ated analogs (6a and b) were 4- to 5-fold less active than
1. (Table 1) Also, N-7 unsubstituted analogs (6c–d) were
about 2-fold less active than the corresponding N-7
alkylated analogs.
N
H
a
N
N
NH2
CN
H2N
Cl
13
12
Alternatively, we maintained the 1,3-disposition of the
sidechains in the purine nucleus by directly attaching
the benzyl group to the N-7 atom. The synthesis of this
analog was carried out as shown in Scheme 2.8,9 During
the synthesis, about 25% of regioisomer 9 was also ob-
tained, which was used to prepare analog 11 having a
1,4-disposition. In addition, to further understand the
effect of relocating the nitrogen atom, we prepared the
corresponding pyrazolo[3,4-d]pyridine analog using the
approach depicted in Scheme 3.8,10
Cl
d
b, c
Cl
N
N
Cl
N
Cl
14
N
(Et)2N
N
N
N
H
Cl
These analogs showed a 3-fold reduction in activity as
compared to 1 (Table 2). Interestingly, there was no dif-
ference in the activity when the sidechains were disposed
in meta (1,3-) (10) or para (1,4-) (11) configurations.
Pyrazolo[3,4-d]pyridine analog (15) also showed a simi-
lar level of activity as 1.
15
Cl
Scheme 3. Reagents and conditions: (a) 2,4-dichlorobenzaldehyde,
H2, Pd/C; 70%; (b) ethyl propiolate, acetic acid, 110 °C, 55%; (c)
POCl3, reflux, 80%; (d) H2N–(CH)Me–(CH2)3N(Et)2, NMP, 130 °C,
55–60%.
Table 1. SAR for purines
Compound
R
R1
R2
CCR4 IC50 (lM)12
6a
6b
6c
6d
H
–(CH)Me–(CH2)3N(Et)2
2,4-Di-Cl-benzylamine
2,4-Di-Cl-benzylamine
–(CH)Me–(CH2)3N(Et)2
2,4-Di-Cl-benzylamine
–(CH)Me–(CH2)3N(Et)2
–(CH)Me–(CH2)3N(Et)2
2,4-Di-Cl-benzylamine
2.1
4.9
1.1
2.6
H
Et
Et

A. V. Purandare et al. / Bioorg. Med. Chem. Lett. 17 (2007) 679–682
681
Table 2. SAR for N-7 benzyl analogs
Table 4. SAR for terminal amine
Compound
CCR4 IC50 (lM)12
Compound
R
CCR4 IC50 (lM)12
10
11
15
0.82
0.83
0.28
O
H
22
0.02
0.08
HN
Cl
O
H
23
HN
Cl
Cl
NH
N
N
a
b, c
N
Table 5. In vitro profile of 1 and 22
Cl
N
N
Cl
N
Assay
Compound 1
Compound 22
CCR4 IC50 (lM)12
Chemotaxis inh. IC50 (lM)
Inh. of Ca2+ mobilization
IC50 (lM)
0.28
5
0.02
0.007
0.003
20
19
Cl
Cl
0.8
Human microsome stability
(nmol/min/mg)
0.10
0.003
Cl
Cl
HERG IC50 (lM)
CYP IC50 (lM)
HHA IC50 (lM)
SOS Chromotest
0.3
>30
All >1
>40
Negative
All >8
>40
Negative
NH
N
NH
N
N
N
d, e
N
N
N
N
N
HN
Encouraged by these findings, we introduced the opti-
mized wings5b from 2 onto this core. Synthesis of these
analogs was carried out using the route in Scheme 4.8
The resulting pyrido[2,3-d]pyrimidine 22 was 2-fold
more active in the CCR4 binding assay compared to
the pyrimidine analog 2 (Table 4). A similar trend in
the activity as seen in the case of 2 through variation
of the terminal amino acid sidechain was also observed
(22 vs. 23) for CCR4 binding.5b Compound 22 also
blocked MDC-mediated chemotaxis (IC50 0.007 lM)
and Ca2+ mobilization (IC50 0.003 lM).12 It was >500-
fold selective against related chemokine receptors
(CCR3, CCR2, and CXCR3) and GPCRs (5-HT1A,
5-HT6, and 5-HT7).
R
22-23
21
Scheme 4. Reagents and conditions: (a) 2,4-dichlorobenzylamine, 1,2-
dichloroethane, Hunig’s base, rt, 85%; (b) Boc-piperazine, NMP,
90 °C, 75%; (c) TFA–CH2Cl2 (1:1), rt, 90%; (d) N-Boc-R-homoproline
or N-Boc-R-proline, DIC, HOAt, DMF, rt, 90%; (e) TFA–CH2Cl2
(1:4), rt, 90%.
Next, we explored the quinoline nucleus in both regio-
isomeric forms. Syntheses of these analogs (16–18, Table
3) were achieved using general procedures as reported in
the literature.8,11 Regioisomer 17 was 2-fold less active
than 1, whereas the other regioisomer 16 was 8- to 9-fold
less active. Further, introduction of a nitrogen atom in
the quinoline core afforded pyrido[2,3-d]pyrimidine 18,
which was 2-fold more active than 1.
As evident (from Table 5), compound 22 showed im-
proved in vitro activity, reduced potential to cause
drug–drug interaction (IC50 > 8 lM for all CYP iso-
zymes), and likely lower clearance (human microsome
stability assay rate; 0.003 nmol/min/mg) as compared
to 1. Compound 22 also displayed lower possibility to
cause cardiac, hepatic, and carcinogenicity liabilities
based upon its in vitro activities in the hERG functional
(patch clamp) (IC50 > 30 lM), immortalized hepatocyte
(HHA) (IC50 > 40 lM), and SOS chromotest (+S9/ÀS9;
negative) assays, respectively.
NHR1
N
R2HN
N
X
16-18
Table 3. SAR for quinolines
Compound
X
R1
R2
CCR4 IC50 (lM)12
16
17
18
CH
–(CH)Me–(CH2)3N(Et)2
2,4-Di-Cl-benzylamine
2,4-Di-Cl-benzylamine
2,4-Di-Cl-benzylamine
–(CH)Me–(CH2)3N(Et)2
–(CH)Me–(CH2)3N(Et)2
1.91
0.5
CH
N
0.14

682
A. V. Purandare et al. / Bioorg. Med. Chem. Lett. 17 (2007) 679–682
J. Leukoc. Biol. 2000, 68, 400, and references cited
therein.
2. Wang, Y.; Zhang, Y.; Yang, X.; Han, W.; Liu, Y.; Xu,
Q.; Zhao, R.; Di, C.; Song, Q.; Ma, D. Life Sci. 2006,
78, 614.
3. (a) Gonzalo, J. A.; Pan, Y.; Lloyd, C. M.; Jia, G. Q.; Yu,
G.; Dussault, B.; Powers, C. A.; Proudfoot, A. E.; Coyle,
A. J.; Gearing, D.; Gutierrzez-Ramos, J. C. J. Immunol.
1999, 163, 403; (b) Kawasaki, S.; Takizawa, H.; Yoney-
ama, H.; Nakayama, T.; Fujisawa, R.; Izumizaki, M.;
Imai, T.; Yoshie, O.; Homma, I.; Yamamoto, K.; Matsu-
shima, K. J. Immunol. 2001, 166, 2055.
4. (a) Chvatchko, Y.; Hoogewerf, A. J.; Meyer, A.; Alouani,
S.; Juillard, P.; Buse, R.; Conquest, F.; Proudfoot, A. E.
I.; Wells, T. N. C.; Power, C. A. J. Exp. Med. 2000, 191,
1755; (b) Wakugawa, M.; Nakamura, K.; Kakinuma, T.;
Tamaki, K. Drug News Perspect. 2002, 15, 175, and
references cited therein.
5. (a) Purandare, A. V.; Gao, A.; Wan, H.; Sommer-
ville, J. E.; Seachord, C.; Burke, C.; Vaccaro, W.;
Wityak, J.; Poss, M. A. Bioorg. Med. Chem. Lett.
2005, 15, 2669; (b) Purandare, A. V.; Wan, H.; Gao,
A.; Somerville, J.; Burke, C.; Vaccaro, W.; Yang, X.;
McIntyre, K. W.; Poss, M. A. Bioorg. Med. Chem.
Lett. 2006, 16, 204.
6. Purandare, A. V.; Somerville, J. E. Curr. Top. Med. Chem.
2006, 6, 1335, and references cited therein.
7. Pitts, W. J.; Watson, A. J.; Dodd, J. H. WO2002088079,
2002.
8. All compounds were characterized by LC–MS and NMR
analysis. In addition, the yields were based on weight of
pure product unless mentioned otherwise.
9. Breshears, S. R.; Wang, S. S.; Bechtolt, S. G.; Christensen,
B. E. J. Am. Chem. Soc. 1959, 81, 3789.
Figure 4. Dose-dependent inhibition by compound 22 of eosinophil
infiltration into allergic lung airways. Total leukocytes (solid bars),
total eosinophils (hatched bars), and percent eosinophils (open bars) in
bronchoalveolar lavage (BAL) fluid were determined as described.13
Data represent means SEM of seven to eight mice per treatment
group. *p < 0.05 vs. Vehicle group, ANOVA with Dunnett’s test.
In vivo, compound 22 was administered subcutaneously
twice a day at doses of 3, 10, and 30 mg/kg to mice pre-
viously immunized with ovalbumin (Fig. 4).13 Dexa-
methasone administered orally, 5 mg/kg, was used as a
control to assess the relative efficacy. Compound 22 re-
duced the recruitment of eosinophils with an ED50 of
ꢀ10 mg/kg (as compared to ED50 of 30 mg/kg twice a
day for 25b). Additionally, at the 30 mg/kg dose, 22
was as efficacious as dexamethasone in reducing eosino-
philic infiltration into mouse BAL.
10. (a) Dorgan, R. J. J.; Parrick, J.; Hardy, C. R. J. Chem.
Soc., Perkin Trans. 1 1980, 4, 938; (b) Dorn, H.;
Ozegowski, R. J. Prakt. Chem. (Leipzig) 1982, 324,
557.
In summary, we have identified a potent and selective
CCR4 antagonist (compound 22) with an improved lia-
bility profile that showed in vivo efficacy in a murine
allergic lung inflammation model.
11. (a) Elslager, E. F.; Hess, C.; Johnson, J.; Ortwine, D.;
Chu, V.; Werbel, L. M. J. Med. Chem. 1981, 24, 127; (b)
Sanmartin, C.; Echeverria, M.; Mendivil, B.; Cordeu, L.;
Cubedo, E.; Garcia-Foncillas, J.; Font, M.; Palop, J. A.
Bioorg. Med. Chem. Lett. 2005, 13, 2031.
References and notes
12. In vitro assays
(
125I-MDC binding with stably
1. (a) Rossi, D.; Zlotnik, A. Annu. Rev. Immunol. 2000, 18,
217; (b) Zlontik, A.; Yoshie, O. Immunity 2000, 12, 121;
(c) Owen, C. Pulm. Pharmacol. Ther. 2001, 14, 193; (d)
Power, C. A.; Proudfoot, A. E. Curr. Opin. Pharmacol.
2001, 1, 417; (e) Berin, M. C. Drug News Perspect. 2002,
15, 10; (f) Editorial; Clin. Exp. Allergy 2001, 31, 1809;
(g) Lukacs, N. W. Nat. Rev. Immunol. 2001, 1, 108; (h)
Mantovani, A.; Gray, P. A.; Damme, J. V.; Sozzani, S.
expressed human CCR4 in HEK-293 cells; chemotaxis
with CCR4-transfected L1.2 cells; calcium mobilization
in CEMS529 cells and cytotoxicity in L1.2 cells) were
performed using conditions as described in Ref. 5a
and b. The values are reported from average of three
repeats.
13. OVA lung inflammation model in mice was carried out as
described in Ref. 5b.