Small molecule antagonists of the CC chemokine receptor 4 (CCR4)

Article abstract of DOI:10.1016/j.bmcl.2007.03.030

The identification, optimization, and structure-activity relationship (SAR) of small-molecule CCR4 antagonists is described. An initial screening hit with micromolar potency was identified that was optimized to sub-micromolar binding potency by enantiomer

Full text of DOI:10.1016/j.bmcl.2007.03.030

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3141–3145
Small molecule antagonists of the CC chemokine
receptor 4 (CCR4)
Douglas F. Burdi,* Shannon Chi, Karen Mattia, Celeste Harrington, Zhan Shi,
Shaowu Chen, Swanee Jacutin-Porte, Robert Bennett, Kenneth Carson, Wei Yin,
Vikram Kansra, Jose-Angel Gonzalo, Anthony Coyle, Bruce Jaffee, Timothy Ocain,
Marty Hodge, Gregory LaRosa and Geraldine Harriman
Millennium Pharmaceuticals, 40 Landsdowne Street, Cambridge, MA 02139, USA
Received 2 February 2007; revised 9 March 2007; accepted 12 March 2007
Available online 15 March 2007
Abstract—The identification, optimization, and structure–activity relationship (SAR) of small-molecule CCR4 antagonists is
described. An initial screening hit with micromolar potency was identified that was optimized to sub-micromolar binding potency
by enantiomer resolution, halogenation of the naphthalene ring, and extension of the alkyl chain linker between the central piper-
idine ring and the terminal aryl group. An antagonist was identified that showed good cross-reactivity against the mouse receptor
and inhibited CCR4-based cell recruitment in dose-dependent fashion.
Published by Elsevier Ltd.
Chemokines are small secreted proteins which stimulate
the directional migration of leukocytes through interac-
tions with a specific subset of G-protein coupled recep-
tors. Their role in leukocyte trafficking has made them
attractive therapeutic targets for both infectious and
inflammatory diseases.1 CCR4 is a chemokine receptor2
that is activated by the ligands MDC (macrophage
derived chemokine, CCL22), TARC (thymus and
activation-related chemokine, CCL17), and CKLF1
(chemokine-like factor 1),3 and has been shown to regu-
late the migration of CD4+ Th2 cells in disease settings.4
Studies in mouse models of allergic airway inflammation
using neutralizing antibodies against both TARC and
MDC,5 as well as receptor knockout studies6, have dem-
onstrated a role for CCR4 and its ligands in airway
hyperresponsiveness. Taken together, these studies sup-
port a role for CCR4 antagonism in the treatment of
allergic inflammatory conditions such as asthma and
atopic dermatitis.
onists of CCR4.8 A high-throughput screen of our small
molecule collection afforded the racemate 1 (Fig. 1) as a
compound of interest that was selected for further fol-
low-up.
In order to establish the initial SAR, we decided to
capitalize on the modular nature of the molecule and
to focus on the four major areas shown in Figure 1:
variants at the sulfonamide substituent (R1), variants
at the amino acid amide (R2), variants at the tertiary
amine (R3), and the absolute stereochemical require-
ments of the central amino acid (asterisk). The synthe-
sis of 1 and its analogs is straightforward9 and is
shown in Scheme 1. Coupling of commercially avail-
able naphthalene-2-sulfonyl chloride with racemic pro-
line-methyl ester afforded 2, which was hydrolyzed with
aqueous lithium hydroxide to afford the carboxylic acid
3. Coupling with BOC-protected 4-amino-piperidine
was achieved using EDCI to afford intermediate 4.
Deprotection with HCl/dioxane afforded 5, which was
reductively alkylated with benzaldehyde in the presence
of sodium triacetoxyborohydride to afford 1. Alkyl-
ation of the secondary amide to yield 6 was readily
achieved using methyl iodide in the presence of sodium
hydride and DMF.
As part of our efforts to identify new potential therapeu-
tics for CCR4-mediated inflammatory diseases, we and
others7 have sought to find novel small molecule antag-
Keywords: Chemokine receptor; Antagonist; Peritoneal recruitment
assay.
Before embarking on a complete SAR survey of the
molecule, we decided to examine the central amino acid
*
Corresponding author. E-mail: chengburdi@yahoo.com
0960-894X/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2007.03.030

3142
D. F. Burdi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3141–3145
H
R2
N
N
N
N
S
*
O
O
N
N
S
R3
O
O
O
R1
O
1
Figure 1. Structure of the initial screening hit (Ki(CCR4) = 7.6 lM) and the initial strategy for establishing SAR.
COOH
CO2Me
N
S
N
S
O
O
SO2Cl
iii
ii
i
O
O
3
2
H
N
H
N
N
S
NH
N
N
N
BOC
v
iv
O
O
O
O
S
O
O
4
5
H
N
N
N
S
N
N
vi
O
O
O
O
S
O
O
1
6
Scheme 1. Reagents and conditions: (i) (+/À)-proline-methyl ester, TEA, DCM (94%); (ii) LiOH, THF/H2O (97%); (iii) tert-butyl 4-amino-
piperidine-1-carboxylate, EDCI, DCM (83%); (iv) 4 N HCl, dioxane (98%); (v) benzaldehyde, Na(AcO)3BH, DCM (89%); (vi) methyl iodide, NaH,
DMF (63%).
linker by preparing several analogs. Substitution of a
variety of amino acids in place of proline, including
N-methyl-alanine, resulted in a complete loss of activity
(data not shown) leading us to conclude that conforma-
tional rigidity was required at this position. We next
sought to establish the stereochemical requirements for
the proline linker and turned our attention to preparing
the enantiomers 7 and 8 (Fig. 2) from D- and L-proline-
methyl ester, respectively. We identified the S-enantio-
mer 8 as the more active enantiomer, and this
compound was used as a starting point for further
optimization.
An initial SAR survey of 8 was undertaken in order to
determine the best path for optimization (Table 1). We
were quickly able to establish a requirement for the sul-
fonamide moiety as a substantial loss in potency was
witnessed with the carboxamide 9, the amine 10, and
the urea 11. The requirement for a bulky aromatic group
at R1 also became apparent as the benzenesulfonamide
12 and the isopropyl sulfonamide 14 lost activity. Inser-
tion of a methylene spacer, as in 13, was also detrimen-
tal. Interestingly, positioning of the sulfonamide moiety
at the 1-position of the naphthalene ring, as exemplified
by 18, also resulted in a loss of potency. Although sub-
H
N
H
N
N
N
N
S
N
S
O
O
O
O
O
O
7
8
Ki (CCR4) = 20.2 uM
Ki (CCR4) = 4.6 uM
Figure 2. Identification of the more active enantiomer 8.

D. F. Burdi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3141–3145
3143
Table 1. Initial tepresentative SAR
R2
N
N
N
X
R3
O
R1
Compound
X
R1
R2
R3
CCR4 FLIPR Ki (lM)
8
9
SO2
CO
2-Naphthyl
2-Naphthyl
2-Naphthyl
H
CH2–Ph
CH2–Ph
CH2–Ph
CH2–Ph
CH2–Ph
CH2–Ph
CH2–Ph
CH2–Ph
H
4.6a
>29b
25b
H
10
11
12
13
14
15
16
17
18
19
20
21
CH2
(CO)NH
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
H
4-tert-Butylphenyl
Phenyl
Benzyl
H
>20a
>29b
15b
H
H
Isopropyl
H
>29b
10a
4-Isopropylphenyl
2-Naphthyl
2-Naphthyl
1-Naphthyl
2-Naphthyl
2-Naphthyl
2-Naphthyl
H
H
>33b
10b
H
Isobutyl
CH2–Ph
CH2-2-pyridyl
CH2-3-pyridyl
CH2–Ph
Me
H
26b
>10a
>10a
1.3a
H
Me
a Value determined10 using TARC as a ligand.
b Value determined10 using MDC as a ligand.
stitution of an isopropyl group on the phenyl ring, as
exemplified by 15, conveyed modest potency, a 2-naph-
thyl ring at R1 proved consistently more potent and we
proceeded to optimize the remainder of the molecule
with this substituent comprising an essential component
of the pharmacophore.
loss of activity leading us to conclude that a methyl
group was optimal at that position.
Armed with an understanding of the SAR surrounding
the central portions of the scaffold, we next sought to
optimize the substituents at the termini. The naphtha-
lene ring proved least accessible to diversification owing
to the limited number of synthetically accessible
5-substituted naphthalene-2-sulfonic acids. Neverthe-
less, the fluoro-substituted analog 30 could be accessed
by thermal decomposition of the diazonium ion in the
presence of BF3 etherate,11 but showed no improvement
in potency. Polar group substitutions, as in 31 and 32,
were not tolerated. We were gratified to see a substantial
increase in potency in the chloro-substituted derivative
33, and this functional group was maintained constant
as we sought to optimize the terminal arene ring.
We next turned our attention to the central and right-
hand portions of the molecule. Our most consistent
jump in activity was realized upon methylation at R2;
comparison of the methylated derivative 21 with its
hydrogen-substituted counterpart 8 revealed a modest
increase in activity. A cursory examination of the piper-
idine substituents revealed a requirement for alkylation
at R3. The secondary amine 16 was inactive, while bulk-
ier alkyl substituents, as in 17, restored some activity.
Heteroaryl substitutions, as in 19 and 20, were not toler-
ated. We returned ultimately to the benzyl substituent in
21, which showed a potency of 1.3 lM and was selected
for further optimization.
Our efforts to improve potency by substitution of the
phenethyl ring proved frustrating. Halogen substitution
at the ortho (34, 35), meta (36) or para (37) position led
to a modest decrease in potency, while electron donating
substituents (38) and electron withdrawing substituents
(42) had a similar effect. Polar substituents (40), both ba-
sic (41) and acidic (39), also led to a modest decrease in
potency, thereby leaving no clear path forward for opti-
mization. Nevertheless, the reasonable potency seen in
33 prompted us to profile this compound further. As
shown in Figure 3, 33 proved nearly equipotent when as-
sayed against murine CCR4 using murine MDC. This
cross-reactivity presented an attractive opportunity to
evaluate this molecule in a murine recruitment model
and prompted us to profile this compound in an
in vivo efficacy experiment.
With the lead compound 21 in hand, we sought to im-
prove the potency by optimizing the substituents at
R1, R2, R3, and R4 as shown in Table 2.
A length scan at R3 revealed an improvement in potency
in the 2-carbon linker as exemplified by 22. The 3-car-
bon linker in 23 offered no improvement in potency
and the additional rotatable bond and molecular weight
were deemed unnecessary. The acyl derivative 24 and
the sulfonamide 25, prepared from the corresponding
amine and acid chlorides, were inactive. We turned next
to the amide substituent at R2, mindful of the improved
potency conveyed by N-methylation. Synthesis of the
N-ethyl derivative 26, however, led to a significant loss
of potency, and therefore further methylene homologs
were not pursued. Substitution with polar groups (27),
both basic (29) and acidic (28), resulted in a complete
In order to evaluate the ability of CCR4 antagonists to
inhibit chemotaxis in vivo, we performed an early proof
of concept study utilizing a peritoneal recruitment assay.

3144
D. F. Burdi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3141–3145
Table 2. Optimizing the substituents at R1, R2, R3, and R4
R2
N
R4
N
N
S
R3
O
O
R1
O
Compound
R1
R2
R3
–CH2–
R4
CCR4 FLIPR Ki (lM)a
21
22
23
24
25
26
27
28
H
Me
Me
H
H
H
H
H
H
H
H
1.3
H
–CH2–CH2–
–(CH2)3–
0.59
0.62
>10
>10
2.80
>10
>10
H
Me
Me
5-Cl
H
–(CO)–CH2–
SO2
Me
Et
5-Cl
H
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CO2Et
–CH2–CO2H
H
29
H
–CH2–CH2–
H
>10
(H2C)2
N
O
30
31
32
33
34
35
36
37
38
39
40
41
42
5-F
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
H
0.85
>10
>10
0.10
0.26
0.46
0.40
0.82
0.47
0.53
0.24
0.28
0.42
5-NH2
5-NHAc
5-Cl
H
H
H
2-F
5-Cl
5-Cl
2-Cl
3-Cl
5-Cl
5-Cl
4-Cl
5-Cl
5-Cl
3-MeO
2-COOH
2-CONH2
2-NH2
2-CF3
5-Cl
5-Cl
5-Cl
a Values determined10 using TARC as a ligand.
400
350
300
250
200
150
100
50
N
N
N
S
O
O
O
33
Cl
Ki (hTARC/FLIPR) = 100 nM
Ki (hMDC/FLIPR) = 500 nM
Ki (mMDC/FLIPR) = 150 nM
0
Figure 3. Structure of 33, a potent CCR4 antagonist used in a proof of
concept study.
Briefly, compound 33 was administered sub-cutaneously
to BALB-C mice (6 animals/cohort). After 30 min,
mMDC (1 lg) was administered intraperitoneally; after
90 min, cells were withdrawn from the peritoneum,
sorted, and counted.
Figure 4. Dose response of 33 to peritoneal recruitment by mMDC.
nists of CCR4. In light of the ability of compound 33
to block recruitment of cells in response to mMDC, it
will be interesting to evaluate these compounds in ani-
mal models of Th2+ disease. These results will be re-
ported in due course.
As shown in Figure 4, we were pleased to see that 33
inhibited the recruitment of cells to the peritoneum in
dose-dependent fashion, with a sub-cutaneous dose of
5 mpk inhibiting recruitment nearly 90%. Based on
our PK analysis (data not shown), this dose delivered
a maximum exposure of 600 nM with exposure exceed-
ing the IC50 throughout the course of the experiment.
References and notes
1. Proudfoot, A. E. I. Nat. Rev. Immunol. 2002, 2, 106.
2. Power, C. A.; Meyer, A.; Nemeth, K.; Bacon, K. B.;
Hoogewerf, A. J.; Proudfoot, A. E. I.; Wells, T. N. C.
J. Biol. Chem. 1995, 270, 19495.
In summary, we have identified and optimized a series of
naphthalene-sulfonamides that are functional antago-

D. F. Burdi et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3141–3145
3145
3. (a) Imai, T.; Chantry, D.; Raport, C. J.; Wood, C. L.;
Nishimura, M.; Godiska, R.; Yoshie, O.; Gray, P. W.
J. Biol. Chem. 1998, 273, 1764; (b) Imai, T.; Baba, M.;
Nishimura, M.; Kakizaki, M.; Takagi, S.; Yoshie, O.
J. Biol. Chem. 1997, 272, 15036; (c) 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.
Dairaghi, D. J.; Schall, T. J.; Collins, T. L.; Medina, J. C.
Bioorg. Med. Chem. Lett. 2006, 16, 2800.
8. For a recent review, see Purandare, A. V.; Somerville, J. E.
Curr. Top. Med. Chem. 2006, 6, 1335.
9. All compounds were characterized by LC/MS and 1H
NMR. Chiral HPLC analysis of 7 and 8 revealed an
enantiomeric excess (ee) of >98%.
4. Kim, C. H.; Rott, L.; Kunkel, E. J.; Genovese, M. C.;
Andrew, D. P.; Wu, L.; Butcher, E. C. J. Clin. Invest.
2001, 108, 1331.
10. This screen is based on the use of a stable Chinese hamster
ovary cell line that overexpresses recombinant CCR4
receptor and Ga16 protein, whose activation causes
intracellular calcium mobilization. Agonist-promoted
increases in intracellular calcium were detected with a
fluorescence imaging plate reader (FLIPR) using a cal-
cium-sensitive dye. The ability of compounds to antago-
nize binding of MDC or TARC to cell surface CCR4 is
reflected in the reduction of fluorescence signal relative to
positive controls. Final assay volume is 60 lL. Final
concentrations of key components are as follows: MDC:
15 nM or TARC: 10 nM; Test compound: variable;
dimethylsulfoxide 1.0% (v/v), bovine serum albumin:
1.0 mg/mL; HEPES: 20 mM; probenecid: 2.5 mM;
10,000 cells per well. A counterscreen against UTP was
performed in parallel in order to screen out cytotoxic
molecules. Good correlations between Ki values were
observed when comparing assays utilizing TARC and
MDC as ligands.
5. (a) Gonzalo, J.-A.; Pan, Y.; Lloyd, C. M.; Jia, G.-Q.; Yu,
G.; Dussault, B.; Powers, C. A.; Proudfoot, A. E. I.;
Coyle, A. J.; Gearing, D.; Gutierrez-Ramos, J.-C.
J. Immunol. 1999, 163, 403; (b) Kawasaki, S.; Takizawa,
H.; Yoneyama, H.; Nakayama, T.; Fujisawa, R.; Izumi-
zaki, M.; Imai, T.; Yoshie, O.; Homma, I.; Yamamoto,
K.; Matsushima, K. J. Immunol. 2001, 166, 2055.
6. Schuh, J. M.; Power, C. A.; Proudfoot, A. E. I.; Kunkel,
S. L.; Lukacs, N. W.; Hogaboam, C. M. FASEB J. 2002,
16, 1313.
7. (a) Allen, S.; Newhouse, B.; Anderson, A. S.; Fauber, B.;
Allen, A.; Chantry, D.; Eberhardt, C.; Odingo, J.;
Burgess, L. E. Bioorg. Med. Chem. Lett. 2004, 14, 1619;
(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; (c) Wang,
X.; Xu, F.; Xu, Q.; Mahmud, H.; Houze, J.; Zhu, L.;
Akerman, M.; Tonn, G.; Tang, L.; McMaster, B. E.;
11. Mirsadeghi, S.; Prasad, G. K. B.; Whittaker, N.; Thakker,
D. R. J. Org. Chem. 1989, 54, 3091.