Novel 2-aminooctahydrocyclopentalene-3a-carboxamides as potent CCR2 antagonists

Article abstract of DOI:10.1016/j.bmcl.2012.10.069

Novel CCR2 antagonists with a novel 2-aminooctahydrocyclopentalene-3a- carboxamide scaffold were designed. SAR studies led to a series of potent compounds. For example, compound 51 had a good PK profile in both dog and monkey, and exhibited excellent efficacy when dosed orally in an inflammation model in hCCR2 KI mice. In addition, an asymmetric synthesis to the core structures was developed.

Full text of DOI:10.1016/j.bmcl.2012.10.069

Bioorganic & Medicinal Chemistry Letters 23 (2013) 351–354
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel 2-aminooctahydrocyclopentalene-3a-carboxamides as potent
CCR2 antagonists
⇑
Chaozhong Cai , Fu-An Kang, Cuifen Hou, John C. O’Neill, Evan Opas, Sandra McKenney,
Dana Johnson, Zhihua Sui
Janssen Research and Development, L.L.C. Welsh & McKean Roads, Spring House, Box 776, PA 19477, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel CCR2 antagonists with a novel 2-aminooctahydrocyclopentalene-3a-carboxamide scaffold were
designed. SAR studies led to a series of potent compounds. For example, compound 51 had a good PK pro-
file in both dog and monkey, and exhibited excellent efficacy when dosed orally in an inflammation
model in hCCR2 KI mice. In addition, an asymmetric synthesis to the core structures was developed.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 27 August 2012
Revised 4 October 2012
Accepted 15 October 2012
Available online 24 October 2012
Keywords:
CCR2
Fused bicycle ring
1,3-Dipolar cycloaddition
Chemoattractant cytokines (chemokines) are a family of small
heparin-binding proteins and involved in recruiting, chemo-traf-
ficking and activation of leukocytes. There are four subfamilies of
chemokines defined by the arrangement of the cysteine (C) residue
of the mature proteins: the C, CC, CXC and CX3C chemokines.1 The
monocyte chemoattractant protein (MCP)-1 or CC ligand 2 (CCL2),
a member of the CC chemokine subfamily, and its receptor, CC-che-
mokine receptor 2 (CCR2), have been implicated in both acute and
chronic inflammatory and autoimmune diseases associated with
infiltration of monocytes, macrophages, lymphocytes, dendritic
cells, NK cells, esinophils, basophils, natural killer (NK) cells and
memory T-cells.2 While modified MCP-1 peptides, which were still
bound but no longer active CCR2, demonstrated the therapeutically
potential of MCP-1 inhibitors in animal models of arthritis,3 several
classes of small molecules with CCR2 antagonism have also been
shown to inhibit chemotaxis in response to MCP-1 in-vitro and
in animal models.4
A majority of these small molecule CCR2 antagonists with low
nanomolar potency in a receptor binding assay share a common
structural pattern: a basic center flanked either by two aromatic
rings or one aromatic and one aliphatic group (Fig. 1).4 Based on
this CCR2 pharmacophore model, we designed a series of fused
bicyclic chemotypes to define the linker 2. Here we reported a ser-
ies of novel 5/5 and 5/6 fused bicycles.
lopentanecarboxylate 1a with 1-chloropropan-2-one, followed by
subsequent annulation gave 2a (n = 1). The resulting fused bicycle
2 was hydrogenated to give ketone 3 as racemate. Reduction of ke-
tone 3 by NaBH4 generated the desired alcohol 4 as the major iso-
mer, which was converted with inversion of configuration to amine
5 by mesylation, azidation and reduction. Protection of the amine
with a Boc and hydrolysis led to acid 6 as a racemic mixture. The
5/6 fused bicycle was made in the same fashion.
The key intermediate 6 was coupled with a variety of amines to
give amides 7, which were subject to reductive amination with ke-
tone or aldehyde, or coupling with acids, to afford the final com-
pounds 8 Scheme 2.
A brief look at the SAR on the right side revealed that bicyclic
amides were superior to mono substituted amides. We decided
to use 3-(trifluoromethyl)-5,6,7,8-tetrahydro-1,6-naphthyridine
for the SAR study on the left hand side (Table 1). 3-Pentanyl analog
9 showed only micro molar level binding affinity, while potency of
its cyclic analog 10 slightly increased. Expansion of the ring from 5
to 7 members led to a trend of increasing potency and arrived at a
submolar compound 12. Insertion of sulfur into the cyclohexane
ring 13, similar in size to 12, retained potency, but had weak hERG
inhibition. Increasing polarity of S to SO2 reduced hERG binding,
but also reduced the potency for CCR2. Replacing cyclohexyl with
tetrahydropyranyl afforded the most potent CCR2 compound 15
with IC50 of 125 nM. Further modification of 15 by either increas-
ing flexibility or removal of basicity of the nitrogen (16 and 17)
failed to provide active compounds. Since polarity seems to sup-
press hERG binding, the methoxy-tetrahydropyranyl group was
examined. To our delight, the CCR2 potency was maintained and
The first generation synthesis of the fused bicycle is illustrated
in Scheme 1. Alkylation of commercially available ethyl 2-oxocyc-
⇑
Corresponding author. Tel.: +1 215 628 7845; fax: +1 215 540 5612.
E-mail address: ccai@its.jnj.com (C. Cai).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.10.069

352
C. Cai et al. / Bioorg. Med. Chem. Lett. 23 (2013) 351–354
aromatic,
aliphatic
or absent
basic or positively
charged at
aromatic
essential
H
N
O
physiological PH
R
N
L
Ar
A
linker1
N
Ar
linker2
R1
n
R2
1-4 atoms
6-9 atoms
Figure 1. Design of fused ring systems.
Table 1
EtO2C
O
CO2Et
CO2Et
O
Left side SAR using the racemic bicycle core
O
d
a, b
c
CF3
n
n
H
n
O
H
1a, n = 1
1b. n = 2
N
N
N
3
R
2
*
*
*
H
n
CO2Et
CO2Et
CO2H
Compd
n
R
hCCR2 binding IC50
(nM)
hERG binding IC50
M)
H2N
HO
BocHN
h, i
e, f
(
l
g
9
1
1
1
3300
2329
1529
>50.0
33.5
33.6
n
H
n
n
H
H
10
11
6 (racemic)
4
5
Scheme 1. Regents and conditions: (a) CH3COCH2Cl, Na2CO3, acetone, rt (n = 1) or
CH@C(CH3)OAc, Mn(OAc)3, Cu(OAc)2, DCM, 40 °C (n = 2); (b) NaH, toluene, reflux;
(c) H2, Pd/C, EtOH and EtOAc; (d) NaBH4, THF, MeOH, 0 °C to rt; (e) MsCl, TEA, DCM,
0 °C; (f) NaN3, DMF, 50 °C; (g) H2, Pd/C, EtOH; (h) (Boc)2O, DCM; (i) LiOH, H2O,
MeOH, rt.
12
13
14
1
1
1
699
14.7
23.1
>50.0
874
S
hERG liability was eliminated (18 vs 15). To compare the 5/6 fused
bicycles to 5/5 analogs, we incorporated the two most potent sub-
stituents (tetrahydropyranyl and methoxy-tetrahydropyranyl) on
the 5/6 system to generate compounds 19 and 20. Because 19
and 20 were less potent, we focused our efforts on the 5/5 system
in later discussion.5
Compound 18 is a mixture of 8 possible diastereoisomers. In or-
der to determine the stereochemical influence of the fused ring on
biological activity and provide the handle for further modification
in the middle, we developed an asymmetric synthetic route to a
key enantiopure intermediate 27, which turned out to be the active
configuration.6 All SAR from this point on was conducted using the
pure active enantiomer for the central bicycle.
O
O
3558
S
15
16
1
1
125
25.7
47.0
O
>25,000
O
O
17
1
>25,000
>50.0
O
OMe
OMe
18a
19
1
2
2
170
>50.0
38.9
40.9
O
O
As shown in, the optically pure enantiomer 217 was protected
with a second Boc group.8 1,3-Dipolar addition of bis-Boc protected
22 with silylated allylic acetate gave the desired bicycle 23 as a
major isomer.9 Ozonolysis of olefin 23 gave ketone 24, which
was deoxygenated via tosyl hydrazone to 25.10 Removal of both
Boc groups, and reprotection of the amine with one Boc group
yielded 26, which was hydrolyzed to give 27. Acid 27 was subject
to coupling with a variety of amines to yield optically pure amide
28. TFA-mediated deprotection and subsequent reductive amina-
tion with previously optimized methoxy-tetrahydropyranyl ketone
provided candidate 29 for the right side aromatic ring SAR
investigation.
7927
1070
20a
O
a
Methoxytetrahydropyrans are mixtures of diastereoisomers.
As illustrated in Table 2, unsubstituted pyridine 30 was inactive.
When the pyridine ring was substituted with electron withdrawing
groups such as Cl (31), Br (32), CF3 (33) and NO2 (34) at the 3 posi-
O
O
Ar
Ar
b, c
N
a
BocHN
R
N
N
6
H
H
n
n
7
8 (n = 1 or 2)
Scheme 2. Regents and conditions: (a) the amine, EDAC, HOBt, TEA, THF; (b) TFA, DCM; (c) ketone/aldehyde, NaBH(OAc)3, TEA, DCM or acid, EDAC, HOBt, TEA, DCM.

C. Cai et al. / Bioorg. Med. Chem. Lett. 23 (2013) 351–354
353
Table 2
Table 3
Right side SAR using the enantiopure bicycle core
Bicycle core SARa
R
CF3
OMe
OMe
O
O
H
H
N
2
N
N
N
X
N
1
3
4
8
5
R1
R2
O
O
7
H
H
6
Compd R1
R2
hCCR2 binding IC50
(nM)
hERG binding IC50
M)
Compd
R
X
hCCR2 binding IC50 (nM)
hERG binding IC50 (lM)
(l
30
31
32
33
34
35
36
37
38
39
40
41
42
H
Cl
Br
CF3
NO2
OMe
OiPr
F
Cl
Br
CF3
CN
OCF3
N
N
N
N
N
N
N
CH
CH
CH
CH
CH
CH
>25,000
1100
200
>50.0
33.3
48.9
>50.0
>50.0
46.6
46.9
23.5
42.4
20.3
27.6
33.7
7.4
33
H
H
F
H
F
H
Cl
H
CF3
H
CO2NH2
H
OH
H
OMe
H
76
95
23
150
20
45
1258
263
88
130
59
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
43
F
76
44a
44b
45a
45b
46a
46b
47a
47b
48a
48b
49a
49b
50a
50b
F
H
Cl
H
1500
>25,000
5600
4400
400
290
34
1800
16
CF3
H
CO2NH2
H
OH
H
OMe
H
58
210
200
8095
247
⁄Single enantiomers for the bicycle core; mixtures of diastereoisomers for meth-
oxy-tetrahydropyranyl moiety.
NMe2
H
NMe2
tion, the potency of these analogs was found to be enhanced, with
CF3 being the most potent compound. Compounds with electron
donating groups (35, 36) at the same position were less potent.
This indicates that both lipophilicity and electronic effects influ-
ence the potency at this position. To investigate the influence of
electron density in the right aromatic ring, we synthesized a series
of phenyl analogs substituted with both electron withdrawing (F,
Cl, Br, CF3 and CN) and donating (OCF3) groups. This modification
resulted in two low nanomolar potent compounds 40 (34 nM)
and 42 (16 nM). However, both 40 and 42 suffered from strong
hERG inhibition and failed to provide enough separation between
CCR2 and hERG activity.
a
Single enantiomers for the bicycle core; mixtures of diastereoisomers for
methoxy-tetrahydropyranyl moiety.
cles. Taking advantage of the newly developed synthetic method,
we were able to conduct our SAR study at the 7-position by func-
tionalization from intermediates 23 and 24 (Scheme 3). It is worth
mentioning that mono substitutions on the prochiral 7-position
would generate two epimers with the substituent oriented to-
wards (R1) and away from (R2) the naphthyridine ring.
Many potent compounds were identified by this study. Both
lipophilic and polar groups are tolerated (Table 3). The bisfluoro
analog 43, which was designed to block metabolism on the fused
ring, displayed equal potency to the parent compound 33. Mono
Having executed both left hand side (R) and right hand side (Ar)
modifications in 8, we directed our attention to the middle bicy-
CO2Me
b
(Boc)2N
a
BocHN
CO2Me
CO2Me
(Boc)2N
H
21
22
23
CO2Me
CO2Me
CO2Me
d, e
f, g
(Boc)2N
BocHN
(Boc)2N
c
O
H
H
H
24
25
26
O
O
MeO
O
NHR
NHR
CO2H
H
N
BocHN
BocHN
j, k
h
i
H
H
H
27
28, pure enantiomer
29
Scheme 3. Regents and conditions: (a) (Boc)2O, DMAP, THF, 58 °C; (b) AcOCH2C(@CH2)CH2TMS, Pd(OAc)2, P(i-PrO)3, toluene, 100 °C; (c) O3, EtOH, À78 °C; (d) NH2NHTs,
Na2SO4, MeOH, 80 °C; (e) catecholborate, CHCl3, 0 °C to rt, then NaOAc, reflux; (f) TFA, DCM, rt; (g) (Boc)2O, DCM; (h) LiOH, H2O, MeOH, rt; (i) the amine, EDAC, HOBt, TEA,
THF; (j) TFA, DCM; (k) 3-methoxydihydro-2H-pyran-4(3H)-one, NaBH(OAc)3, TEA, DCM.

354
C. Cai et al. / Bioorg. Med. Chem. Lett. 23 (2013) 351–354
CF3
CF3
OMe
OMe
O
O
H
N
H
N
N
N
N
N
O
O
H
H
51
52
Figure 2. Enantiopure 51 and 52.
phages by 89%, 100%, 100% and 101% at 0.3, 1, 3 and 10 mpk dose,
respectively (ED50 = 0.1 mpk; EC50 = 22 nM in plasma)(Fig. 3).
In addition to its favorable in vivo efficacy, compound 51 dis-
played excellent pharmacokinetic profiles in higher species (dog
and monkey). In dog PK (10 mpk in 0.5% methocel, po and 2 mpk
in 20% HPbCD, iv), it has moderate systemic clearance (CL:
15.2 mL/min/kg), large volume of distribution (Vdss: 2.3 L/kg), rea-
sonable oral half-life (t1/2: 3.4 h) and good oral exposure (AUC:
12128 h⁄ng/mL). The oral bioavailability (F%) of compound 51
was 100%.
In conclusion, we identified novel 2-aminooctahydrocyclo-pen-
talene-3a-carboxamides as potent CCR2 antagonists. Modifications
in the middle bicycle as well as the left and right hand sides led to
the discovery of a series of promising CCR2 antagonists with excel-
lent in-vivo efficacy, and desirable PK and safety profiles. Further
modifications on the middle bicyclic ring are in progress and will
be reported in due course.
125
100
75
50
25
0
Basal
Acknowledgments
Tg+Vehicle po bid
Tg+1 mg/kTggp+3o mbigd/kg po bid
Tg+10 mg/kg po bid
Tg+0.3 mg/kg po bid
We are grateful to Dr. Christopher Teleha for providing large
quantities of the intermediate 23 and Dr. Derek Beauchamp for
the crystallography analysis of 51. We also thank the ADME/PK,
Secondary Pharmacology and Lead Generation Biology Teams for
their contributions to this work.
Figure 3. In vivo efficacy in thioglycollate induced peritonitis of 51.
substitution with small groups such as F, Cl, OH or a flat substitu-
ent such as CO2NH2 either increased or retained the binding po-
tency in both epimers. Increasing the size of the substitution
(OMe, CF3 and NMe2) resulted in decrease in potency regardless
of the electronic property of the substitution, which indicates the
receptor binding pocket is limited in size. In the cases of 46a and
50a, the significant loss of activity may be caused by repulsive
interaction between the substituent and the naphthyridine ring
leading to a disfavored confirmation.
As a result, the SAR on the left, right and the middle led to sev-
eral compounds with desirable profiles (33, 43, 44a, 45a, 45b, 47a,
48a and 48b). Some of these analogs were further evaluated as po-
tential clinical candidates. For example, compound 33, which was
still a mixture of two diastereomers due to the chirality of the
methoxy-tetrahydropyranyl moiety, was separated with chiral
KromasilÒ AD column (25% isopropanol in ethanol) to give two ma-
jor diastereoisomers 51 and 52 (Fig. 2).11 Both 51 and 52 exhibited
equal binding affinity (53 nM for 51, 61 nM for 52) against the
CCR2 receptor. In CCR2 functional assays, they are potent antago-
nists of MCP-1-induced chemtaxis (22 nM for 51, 24 nM for 52)
and sub-molar antagonists of MCP-1 induced calcium flux efficacy.
Compound 51 has no hERG activity in both the binding assay
References and notes
1. (a) Fernandez, E. J.; Lolis, E. Annu. Rev. Pharmacol. Toxicol. 2002, 42, 469; (b)
Laing, K. J.; Secombes, C. J. Dev. Comp. Immunol. 2004, 28, 443.
2. (a) Kinne, R. W.; Brauer, R.; Stuhlmuller, B.; Palombo-kinne, E.; Burmester, G.
Arthritis Res. 2000, 2, 189; (b) Barrera, P.; Blom, A.; van Lent, P. L.; van Bloois, L.;
Beijnen, J. H.; van Rooiken, N.; De Waal Malefijt, M. C.; van de Putte, L. B.; van
den Berg, W. B. Arthritis Rheum. 1951, 2000, 43.
3. Gong, J. H.; Clark-Lewis, I. J. Exp. Med. 1995, 181, 631.
4. (a) Xia, M.; Sui, Z. Expert Opin. Ther. Pat. 2009, 19, 295; for a recent review of
CCR2, see: (b) Hou, C.; Sui, Z. In ‘‘CCR2 Antagonists for the Treatment of Diseases
Associated with Inflammation’’ Anti-Inflammatory Drug Discovery; Levin, J.,
Laufer, S., Eds.; Royal Society of Chemistry, 2012; pp 350–380. and references
therein.
5. During our research process, scientists at Abbott Laboratories reported
octahydropentalene derivatives as chemokine receptor antagonists in
a
patent (George, D. M.; Wang, L.; Li, B.; Ericsson, A. M.; Ansell, G. K. PCT Int.
Appl. WO 2009042193). Although they did not have the compounds with the
same structures as in this paper and neither the specified active
stereochemistry, they shared the same 5/5 fused bicycles.
6. For example, compound 33 (IC50, 76 nM), which was made from intermediate
27, was more potent than its diastereoisomer (IC50, 4070 nM, structure not
shown) made from the other enantiomer of 27.
7. Compound 21 was prepared from (+)-2-azabicyclo[2.2.1]hept-5-ene-3-one
according to the procedure described in: Smith, M. E. B.; Derrien, N.; Lloyd, M.
C.; Taylor, S. J. C.; Chaplin, D. A.; McCague, R. Tetrahedron Lett. 2001, 42, 1347.
8. The second Boc group not only prevents the nucleophilic addition of the
nitrogen to the silylated allylic acetate to generate the undesired product, but
also plays an important role in controlling the stereochemistry in favor of the
desired, enantiomer 23.
(>50
It did not inhibit the major isoforms of human CYP enzymes with
an IC50 of >10 M against CYP1A2, 2C19, 2D6, 2C9 and 7.5
against CYP 3A4.
lM) and the functional assay (15.6% @ 3 lM, patch clamp).
l
lM
9. Trost, B. M.; Chan, D. M. T. J. Am. Chem. Soc. 1979, 101, 6429.
10. Corey, E. J.; Desai, M. C.; Engler, T. A. J. Am. Chem. Soc. 1985, 10, 4339.
11. The stereochemistry of compound 51 was determined by X-ray
crystallography of the single crystal of its semi-succinate from acetonitrile.
When orally administered in an acute inflammation model (thi-
oglycollate induced peritonitis) in mCCR2 knock-out/hCCR2 knock-
in mice, compound 51 markedly inhibited the infiltrate of macro-