Structure-activity relationships of novel, highly potent, selective, and orally active CCR1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2007.03.104

Design and synthesis of a series of 3-amino-4-(2-(2-(4-benzylpiperazin-1-yl)-2-oxoethoxy)phenylamino)cyclobutenedione derivatives as novel CCR1 antagonists are described. Structure-activity relationship studies led to the identification of compound 22, which demonstrated potent binding activity, functional antagonism of CCR1 as well as good species cross-reactivity. In addition, compound 22 also showed desirable pharmacokinetic profiles and was selected for in vivo studies in the mouse collagen-induced arthritis model.

Full text of DOI:10.1016/j.bmcl.2007.03.104

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3367–3372
Structure–activity relationships of novel, highly potent,
selective, and orally active CCR1 antagonists
Yun Feng Xie,a, Kirk Lake,a Kathleen Ligsay,a Mallareddy Komandla,a Ila Sircar,a
Gobi Nagarajan,a Jian Li,b Kui Xu,b Jason Parise,b Lisa Schneider,b Ding Huang,b
Juping Liu,b Kevin Dines,b Naoki Sakurai,b Miguel Barbosab and Rick Jackb,*
aDepartment of Chemistry, Tanabe Research Laboratories, Inc., 4540 Towne Centre Court, San Diego, CA 92121, USA
bDepartment of Biology, Tanabe Research Laboratories, Inc., 4540 Towne Centre Court, San Diego, CA 92121, USA
Received 6 March 2007; revised 23 March 2007; accepted 29 March 2007
Available online 5 April 2007
Abstract—Design and synthesis of a series of 3-amino-4-(2-(2-(4-benzylpiperazin-1-yl)-2-oxoethoxy)phenylamino)cyclobutenedione
derivatives as novel CCR1 antagonists are described. Structure–activity relationship studies led to the identification of compound
22, which demonstrated potent binding activity, functional antagonism of CCR1 as well as good species cross-reactivity. In addition,
compound 22 also showed desirable pharmacokinetic profiles and was selected for in vivo studies in the mouse collagen-induced
arthritis model.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Chemokines are a specialized family of small cytokines
(8–15 kDa) that act as potent mediators of inflammation
by their ability to recruit and activate specific leukocyte
subpopulations. Chemokines activate leukocytes by
binding to specific chemokine receptors belonging to
the G protein-coupled receptors (GPCR) on the target
cell plasma membrane.1,2
CCR1-positive cells—mostly macrophages—are scat-
tered throughout the synovium in RA.4 MIP-1a null
mice exhibited lower clinical and histopathological
scores in anti-type II collagen monoclonal antibody
(anti-CII mAb)-induced arthritis.5 CCR1-deficient mice
had reduced incidence and ameliorated severity of
disease in a mouse EAE model.6 CCR1-deficiency also
significantly prolonged allograft survival in several
cardiac allograft models.7 The administration of
Met-RANTES, a CCR1/5 antagonist, reduced joint
inflammation, bone destruction, and cell recruitment
into joints in rodent arthritis models.8,9 Initial safety
and proof-of-concept was shown with a selective, orally
bioavailable CCR1 small molecule antagonist in
patients with RA.10 These data illustrate the potential
of CCR1 antagonists in these diseases, and thus make
it an attractive target for drug discovery research.
Chemokines are divided into four subfamilies: CC,
CXC, C, and CX3C defined by the position of the con-
served cysteine residues near the N-terminus. CCR1 is a
chemokine receptor expressed by a variety of immune
cells including monocyte-macrophages, T-lymphocytes,
neutrophils, basophils, eosinophils, NK cells, mast cells,
and dendritic cells. CCR1 is activated by CC chemo-
kines such as MIP-1a, RANTES, and MPIF-1.3 Sub-
stantial evidence regarding genetic and animal studies
has linked CCR1 and its ligands to the pathogenesis of
autoimmune diseases such as rheumatoid arthritis
(RA) and multiple sclerosis, and to allograft rejection.
Several chemotypes of CCR1 antagonists have been
reported: piperazine derivative I (BX-471)11 and II,12
xanthene derivative III,13 hydroxyethylene peptide
isostere compound IV14, and piperidine derivative V15
(Fig. 1). BX-471 and compound IV were advanced to
the clinic for multiple sclerosis and rheumatoid arthritis,
respectively.
Keywords: Chemokine receptor 1; CCR1 antagonist; Diaminocyclo-
butenedione.
*
Corresponding author. Tel.: +1 858 622 7022; fax: +1 858 558
0650; e-mail: rjack@trlusa.com
Present address: Genomics Institute of the Novartis Research
Foundation, 10675 John Jay Hopkins Drive, San Diego, CA
92121, USA.
Except compound II,12a,2b most CCR1 antagonists
reported so far suffered from poor species cross-reactivity
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.03.104

3368
Y. F. Xie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3367–3372
O
O
F
O
F
N
N
N
N
Cl
NH
Cl
NH
O
O
NH2
I (BX-471)
II
Cl
OH
O
O
CN
I-
F
N
O
N
N
H
NHMe
S
O
+
N
OH
HN
N
IV
Cl
III
V
Figure 1. Selected CCR1 antagonists.
or poor oral bioavailability, which may limit extensive
safety and pharmacological evaluation. For example,
compound III has shown in vivo efficacy only by sc
administration in the collagen-induced arthritis mouse
model. Our objective was to design a novel series with
an improved profile based on BX-471. A review of the
D
O
O
R2
O
O
F
N
N
R1
B
A
N
N
Cl
NH
NH
O
R3
O
NH2
BX-471
O
NHR4
C
Figure 2. Cyclobutenedione as a urea bioisostere and SAR strategy.
R2
R2
R2
OH
NH
OH
OH
R2
OH
a)
b)
HO
NH
R3
R2
R3
NH
R3
+
NH
R3
NHR4
NH2
O
O
O
OEt
R3
O
O
O
1a
R1
1
2
R1
R
N
R
Cl
d)
O
c)
N
R1
HN
N
Cl
3
O
R
R2
O
N
R1
e)
N
NH
R3
2
+
3
O
NHR4
O
R= Me, 4-11, 13-32
R= H, 12
Scheme 1. Reagents and conditions: (a) 1,2-diethoxycyclobutenedione, EtOH, reflux, 55–70%; (b) R4NH2, rt, 60–80%; (c) (R)-2-methylpiperazine or
piperazine, DIEA, DCM, 75–90%; (d) chloroacetylchloride, DIEA, DCM, 70–85%; (e) K2CO3, KI, DMF, 65–75%.

Y. F. Xie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3367–3372
3369
Table 1. Effect of substitutions on the N-benzylpiperazine moiety
literature related to BX-471 indicated that the urea
moiety is dispensable and can be replaced with various
groups including hydrogen. We thus sought a urea
replacement and identified diaminocyclobutenedione
as a novel urea mimetic moiety. Here, we report the
synthesis and structure–activity relationships (SAR) of
a novel series of diaminocyclobutenedione derivatives
as highly potent and selective CCR1 antagonists.
O
O
N
N
Cl
NH
R1
O
NHMe
O
Compound R1
IC50
(nM, hCCR1 binding)a
We separated the prototype structure into four discrete
areas—benzyl piperazine (part A), left aryl (part B),
aminocyclobutenedione (part C), linker (part D)—and
conducted methodical SAR studies (Fig. 2).
4
5
26
F
Compounds 4–32 were synthesized according to the gen-
eral Scheme 1, wherein the substitutions on the right and
left aryls (R1/R2/R3, parts A and B, Tables 1 and 2,
compounds 4–18) as well as the amine moiety on the
350
78
6
Cl
Br
Table 3. Effect of amine substitutions on the cyclobutenedione moiety
7
270
O
O
F
N
F
F
N
Cl
NH
8
97
F
O
NH
R4
F
9
2,340
2,910
220
O
Compound
R4
IC50 (nM, hCCR1 binding)a
10
11
4
19
20
21
22
Me
Et
26
72
Pr
CH2Ph
H
170
800
10
Cl
S
Cl
(des-methyl piperazine)
a Results shown are mean values of triplicate samples in a single
experiment.16
23
24
160
63
N
12
200
O
OH
25
26
20
18
NH2
Table 2. Effect of substitutions on the left aryl ring
27
28
64
12
N
N
O
R2
R3
O
F
N
O
N
NH
NH
O
NHMe
29
30
31
32
50
15
42
26
N
N
O
Compound
R2
R3
IC50 (nM, hCCR1 binding)a
4
13
14
15
16
17
18
H
H
H
H
H
F
Cl
H
26
920
N
N
Br
F
25
160
Me
F
90
140
H
Ph
>10,000
a Results shown are mean values of triplicate samples in a single
experiment.16
a Results shown are mean values of triplicate samples in a single
experiment.16

3370
Y. F. Xie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3367–3372
cyclobutenedione (part C, Table 3, compounds 19–32)
were varied.
The effect of substitutions on the phenyl ring (part B,
Table 2) was explored next. Deletion of the 4-chlorine
atom from 4 was detrimental to the potency, resulting
in a >30-fold loss in activity (13 vs 4). The 4-fluoro ana-
log (15) as well as 3,4-difluoro analog (17) led to a 6-fold
loss in potency. Replacement of the 4-chloro with 4-bro-
mo (14) retained the high potency with an IC50 value of
25 nM. On the other hand, replacement of the 4-chloro
with the bulky phenyl group resulted in total loss of
affinity, suggesting a space restriction around this site.
Interestingly, replacement of the 4-chloro group of 4
with a methyl group retained modest potency (16,
IC50 = 90 nM).
Requisite hydroxyanilines were reacted with 1,2-diet-
hoxycyclobutenedione under EtOH reflux to give
intermediate 1 and the corresponding diaminated
compound 1a. Treatment of 1 with desired amines at
room temperature led to the diaminocyclobutenedione
intermediate 2, which was subsequently reacted with
compound 3 (prepared in two steps) to yield target
compounds 4–11 and 13–32. The des-methyl piperazine
compound 12 was prepared in a similar fashion starting
from the 4-chloro-N-benzylpiperazine. The synthesis of
compounds 33–34 wherein the linker –OCH2– was
replaced with the bioisostere –C@C– is outlined in
Scheme 2.
Having identified the 4-fluorobenzyl and 4-chloro-
substituted phenyl as the two best moieties for parts A
The analogs prepared in this study were evaluated for
their inhibitory activity against human CCR1 and ex-
pressed as IC50 values. Compound 4, where a cyclob-
utenedione moiety was used to replace the urea
functionality in BX-471, showed high potency for hu-
man CCR1 binding (IC50 = 26 nM). It was selected as
the starting compound for systematic SAR studies.
The effect of substitutions on the aromatic ring (part
A, Table 1) was initially investigated. Unfortunately,
all but 4-halo substitutions resulted in reduced potency.
The potency decreased with the increase in size of the
halogen (compare compounds 4, 6, and 7). Compound
8, the 3,4-difluoro analog, was 3-fold less potent than
compound 4. The un-substituted piperazine analog 12
also exhibited a 3-fold loss in potency.
Table 4. Linker modification
O
F
N
N
Cl
NH
O
O
NHR
Compound
R
H
IC50 (nM, hCCR1 binding)a
33
34
7
76
Me
a Results shown are mean values of triplicate samples in a single
experiment.16
O
O
O
O
O
a)
H
b)
Cl
NO2
Cl
NH2
Cl
NO2
O
O
F
N
OH
c)
d)
N
Cl
NH2
Cl
NH2
O
O
F
F
N
N
N
N
f)
e)
Cl
NH
O
Cl
NH
O
NHR4
O
OEt
O
R4 = H, 33
R4 = Me, 34
Scheme 2. Reagents and conditions: (a) Ph3P@CHCO2CH3, toluene, reflux, 82%; (b) Zn, AcOH, EtOH, reflux, 30%; (c) 2 M NaOH, MeOH, 95%;
(d) (R)-1-(4-fluorobenzyl)-3-methylpiperazine, EDC, HOBt, DIEA, THF, 68%; (e) 1,2-diethoxycyclobutenedione, EtOH, reflux, 42%; (f) R4NH2,
EtOH, rt, 40% (R = H), 50% (R = Me).

Y. F. Xie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3367–3372
Table 5. Functional and species activities of selected compounds
3371
Compound
Ca2+ flux (h, IC50, nM)
Chemotaxis (h, IC50, nM)
mCCR1 binding (IC50, nM)
rCCR1 binding (IC50, nM)
49
—
4
14
18
14
4
2
5
1
1
2
1
3
3
1
1
2
17
25
7
615
—
22
24
260
—
20
19
13
6
54
47
23
12
39
38
—
18
—
25
26
960
968
1770
1670
1000
—
21
17
28
29
12
5
87
24
31
33
8
130
—
6
18
BX471
615
59
Table 6. Pharmacokinetic profiles in rodents for compound 22a
Species Cmax Tmax t1/2 (h, po) F (%)
AUC0!1
(lM, po) (h, po) (h*lM, po)
CCR2, CCR3, CCR4, and CCR5 (data not shown).
Based on its excellent in vitro potency, functional activ-
ity, species cross-reactivity, and pharmacokinetic pro-
file (Table 6), compound 22 was selected for in vivo
evaluation in a mouse collagen-induced arthritis model.
It showed a statistically significant reduction of arthri-
tis score at 50 mg/kg via oral administration. The de-
tails of these studies will be published in a separate
paper.
Mouse 11.6
Rat 5.3
1
1
95.1
43.6
2.3
2.6
100
ND
ND, not determined.
a Male Balb/C mice and male Sprague–Dawley rats were used. Com-
pound was administered as a solution in methyl cellulose with 10%
Tween 80. AUC, area under the curve; F, bioavailability determined
by iv/po ratio. Dose: 5 mg/kg iv; 50 mg/kg po.
In summary, we have identified a series of diaminocy-
clobutenedione analogs as novel, highly potent, selec-
tive, and orally active CCR1 receptor antagonists with
functional activity and good species cross-reactivity
against rat CCR1.
and B, respectively, the focus was switched to modify
the amino group on part C (Table 3). Increasing the size
of the alkyl from Me (4) to Et (19) and Pr (20) gradually
reduced the potency, whereas a 3-fold increase in
potency was observed with the primary amine
compound 22. Incorporation of additional substitutions
on the Et-chain of 19 either retained (24, 27, and 29) or
enhanced potency (25, 26, and 28). The enhanced
potency observed for the pyridyl analogs suggests that
a hydrogen bond acceptor in this region of the molecule
is favorable. SAR at this site also indicates the presence
of a binding pocket that can accommodate diverse
groups.
Acknowledgment
The authors thank Dr. Jie-Fei Cheng for valuable sug-
gestions in the preparation of the manuscript.
References and notes
1. Charo, I. F.; Ransohoff, R. M. N. Eng. J. Med. 2006, 354,
610.
2. Tarrant, T. K.; Patel, D. D. Pathophysiology 2006, 13, 1.
3. Saeki, T.; Naya, A. Curr. Pharm. Des. 2003, 9, 1201.
4. Katschke, K. J.; Rottman, J. B., et al. Arthritis Rheum.
2001, 44, 1022.
5. Chintalacharuvu, S. R.; Wang, J. X.; Giaconia, J. M.;
Venkataraman, C. Immunol. Lett. 2005, 100, 202.
6. Rottman, J. B.; Slavin, A. J.; Silva, R.; Weiner, H. L.;
Gerard, C. G.; Hancock, W. W. Eur. J. Immunol. 2000, 30,
2372.
Table 4 highlights the modification of linker part D.
Replacement of –OCH2– in 22 with the bioisostere
–C@C– in the linker part D produced the corresponding
cinnamide analog 33 that retained similar potency
(IC50 = 7 nM). However, compound 34, the cinnamide
analog of 4, showed a 3-fold reduction in potency. Sim-
ilar SAR with cinnamide analogs was observed
previously.12
Selected compounds were examined for functional activ-
ity and species cross-reactivity17 (Table 5). All com-
pounds showed the inhibition of the MIP-1a-induced
Ca2+ mobilization and inhibition of chemotaxis, which
indicated functional antagonism for hCCR1. Com-
pounds also showed higher affinity for rCCR1 than
mCCR1. Compound 22 showed the best overall profile
with IC50 values of 4 nM (Ca2+ mobilization), 7 nM
(chemotaxis), 260 nM (mCCR1 binding), and 20 nM
(rCCR1 binding), respectively.
7. Gao, W.; Topham, P. S.; King, J. A.; Smiley, S. T.;
Csizmadia, V.; Lu, B.; Gerard, C. J.; Hancock, W. W.
J. Clin. Invest. 2000, 105, 35.
8. Shahrara, S.; Proudfoot, A. E.; Woods, J. M.; Ruth, J. H.;
Amin, M. A.; Park, C. C.; Haas, C. S.; Pope, R. M.;
Haines, G. K.; Zha, Y. Y., et al. Arthritis Rheum. 2005,
52, 1907.
9. Plater-Zyberk, C.; Hoogewerf, A. J.; Proudfoot, A. E.;
Power, C. A.; Wells, T. N. Immunol. Lett. 1997, 57, 117.
10. Haringman, J. J.; Kraan, M. C.; Smeets, T. J. M.;
Zwinderman, K. H.; Tak, P. P. Ann. Rheum. Dis. 2003,
62, 715.
Compound 22 was found to be a highly selective CCR1
antagonist with greater than 500-fold selectivity over
11. Lians, M.; Mallari, C.; Rosser, M.; Ng, H. P.; May, K.;
Monahan, S., et al. J. Biol. Chem. 2000, 175, 19000.

3372
Y. F. Xie et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3367–3372
12. (a) Revesz, L.; Bollbuck, B.; Buhl, T.; Eder, J.; Esser, R.;
Feifel, R.; Heng, R.; Hiestand, P. Bioorg. Med. Chem.
Lett. 2005, 15, 5160; (b) Revesz, L.; Bollbuck, B.; Buhl, T.;
Dawson, J.; Feifel, R.; Heng, R.; Hiestand, P.; Schlap-
bach, A.; Waelchli, R.; Loetscher, P. Lett. Drug Des.
Discov. 2006, 3, 689.
13. Naya, A.; Sagara, Y.; Ohwaki, K.; Saeki, T.; Ichikawa,
D.; Iwasawa, Y.; Noguchi, K.; Ohtake, N. J. Med. Chem.
2001, 44, 1429.
14. Kath, J.; Brissette, W.; Brown, M.; Conklyn, M.; DiRico,
A.; Dorff, P.; Gladue, R.; Lillie, B.; Lira, P., et al. Bioorg.
Med. Chem. Lett. 2004, 14, 2169.
15. Schwender, C. F.; Mackay, C. R.; Pinto, J. C.; Newman,
W. WO9802151.
16. Results shown are mean values of triplicate samples in a
single experiment. THP-1 cells expressing CCR1 were
harvested and washed once with PBS. Cells (5 · 106/ml)
were mixed with test compound or vehicle and 50 pM
human [125I]MIP-1a in a 96-well tissue culture plate in a
total volume of 120 ll of binding buffer which contained
50 mM Hepes, pH 7.4, 5 mM MgCl2, 1 mM CaCl2, and
0.5% BSA (Hepes-BSA). The mixture was incubated at
room temperature for 90 min and transferred to 96-well
glass fiber filter plate pre-coated with 0.5% polyethylene-
imine (PEI). Cells were separated by vacuum aspiration
and washed twice with Hepes-BSA + 0.5 M NaCl. Radio-
activity was measured on TopCount plate reader with
40 ll Microscint 40 scintillation fluid. Non-specific bind-
ing was determined in the presence of 150 nM unlabeled
human MIP-1a.
17. Results shown for chemotaxis, mouse binding, and rat
binding are the mean values of at least two independent
experiments. Ca2+ flux assay: THP-1 cells in RPMI-1640
were adjusted to 2 · 106/ml, mixed 1:1 with Calcium
Assay Kit buffer (Molecular Devices, Sunnyvale, CA),
and incubated for 30 min at 37 ꢁC in the presence of
2.5 mM probenicid (Sigma). Dye-loaded cells were then
collected by centrifugation at 1000 rpm for 5 min,
resuspended in assay buffer (RPMI-1640 containing
0.1% BSA), and seeded at 1.5 · 105/well on poly-
D-lysine-coated 96-well plates (Molecular Devices) with
or without test compounds. The plates were centrifuged
at 1000 rpm for 5 min. After stimulation with 5 nM
human MIP-1a, changes in intracellular free Ca2+
concentration were measured using FLIPR (Molecular
Devices). Chemotaxis assay: THP-1 cells at 5 · 106/ml
were incubated with 1 lM calcein (Molecular Devices)
for 30 min at 37 ꢁC. Cells were collected by centrifuga-
tion, washed once with RPMI-1640, and resuspended at
5 · 106/ml. The calcein-labeled cells were then mixed
with compound or DMSO control for 30 min at 37 ꢁC.
and loaded in the wells on the top of
a 5-lm
polycarbonate filter in 96-well Boyden chamber
a
(NeuroProbe), in which 0.5 nM human MIP-1a with
or without compounds had been added to the corres-
ponding wells beneath the filter. The sealed chamber
was incubated for 2 h at 37 ꢁC. Non-migrated cells on
top of the filter were removed by wiping. The filter was
reversed and read at 485/530 nm emission/excitation
wavelengths with FlexStation II (Molecular Devices).