Discovery and synthesis of cyclohexenyl derivatives as modulators of CC chemokine receptor 2 activity

Article abstract of DOI:10.1016/j.bmcl.2015.11.051

A novel cyclohexenyl series of CCR2 antagonists has been discovered. This series of small, rigid compounds exhibits submicromolar binding affinity for CCR2. Modification of the substituents on the cyclohexene ring led to the identification of potent CCR2 antagonists. Progress from initial lead 5 (IC₅₀ = 700 nM) to (-)-38 (IC₅₀ = 9.0 nM) is discussed.

Full text of DOI:10.1016/j.bmcl.2015.11.051

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery and synthesis of cyclohexenyl derivatives as modulators
of CC chemokine receptor 2 activity
⇑
Gregory D. Brown , Qing Shi, George V. Delucca, Douglas G. Batt, Michael A. Galella, Mary-Ellen Cvijic,
Rui-Qin Liu, Feng Qiu, Qihong Zhao, Joel C. Barrish, Percy H. Carter
Research and Development, Bristol-Myers Squibb Company, Route 206 and Province Line Road, Princeton, NJ 08543-4000, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel cyclohexenyl series of CCR2 antagonists has been discovered. This series of small, rigid com-
pounds exhibits submicromolar binding affinity for CCR2. Modification of the substituents on the cyclo-
Received 16 September 2015
Revised 11 November 2015
Accepted 16 November 2015
Available online xxxx
hexene ring led to the identification of potent CCR2 antagonists. Progress from initial lead
5
(IC50 = 700 nM) to (ꢀ)-38 (IC50 = 9.0 nM) is discussed.
Ó 2015 Elsevier Ltd. All rights reserved.
Keywords:
Chemokine
CCR2
CCR2 antagonist
GPCR
CC chemokine receptor 2 (CCR2) is the primary chemokine
receptor on monocytes recruited to sites of inflammation.1 It is
the receptor for the pro-inflammatory cytokine, monocyte
chemoattractant protein-1 (MCP-1). On the basis of extensive
pre-clinical studies, CCR2 antagonism is a mechanism of interest
for the treatment of a number of diseases, including atherosclero-
sis, multiple sclerosis, rheumatoid arthritis, and diabetes.2,3 As a
result of the potential clinical relevance of CCR2 inhibition, CCR2
antagonism has been a goal of many medicinal and discovery
chemistry groups.4–6 Several unique classes of CCR2 antagonists
have been explored. Earlier CCR2 antagonists were found to have
similar basic, linear or cyclic structural features such as those
present in Merck’s cyclopentylamine series (1)7 and our own
CF3
O
O
N
H
H
N
O
CF3
N
H
N
N
O
N
N
N
N
1
2
O
O
O
OH
O
Cl
Cl
S
NH
Cl
N
N
O
F3C
N
N
H
Cl
previously reported c
-lactam linked series (2)8 (Fig. 1). Non-basic,
3
4
structurally diverse, unsaturated heterocyclic examples were
disclosed by various research groups. Researchers from
GlaxoSmithKline disclosed a novel series of diaryl-sulfonamide
(3) CCR2 antagonists.9 A rigid, heterocyclic series of triazolopyrim-
Figure 1. CCR2 antagonists.
idin-7-ones (4) was introduced by AstraZeneca.10
A more
comprehensive review of disclosed CCR2 antagonists has been
binding affinity of 700 nM and functional activity in CCR2-depen-
dent chemotaxis and calcium flux assays when tested at one
micromolar concentration (Fig. 2).12
Herein we describe the synthesis and structure–activity rela-
tionships (SAR) of these low molecular weight, rigid cyclohexenyl
derivatives. In an effort to enhance potency, initial synthetic efforts
focused on modification of three primary regions of the compound
(Fig. 3): (1) R-group, (2) aryl substituent (Ar), and (3) carboxylic
acid replacements or mimics (X).
summarized in an earlier report.11
With our continued interest in CCR2 as a therapeutic target, we
conducted a high throughput screening of our internal compound
collection. These efforts led to the identification of a structurally
unique racemic compound (5) exhibiting submicromolar CCR2
⇑
Corresponding author. Tel.: +1 609 252 3543.
E-mail address: gregory.brown@bms.com (G.D. Brown).
http://dx.doi.org/10.1016/j.bmcl.2015.11.051
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Brown, G. D.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.11.051

2
G. D. Brown et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 1
O
Des-methyl (R = H) series SAR
OH
O
OH
R
Ar
Cl
(+/-)-5
(+/-)
CCR2 IC50 = 700 nM
Compound
R
Ar
CCR2
IC50 (nM)
a
Figure 2. CCR2 antagonist: initial lead (5) identified during a high throughput
collection screening. 4:1 trans/cis mixture (racemic).13
5
Me
H
H
H
H
4-Cl Ph
4-Cl Ph
3-Cl Ph
2-OMe Ph
3-(CF3) Ph
3,4-Me2 Ph
3,4-Cl2 Ph
700
15
16
17
18
19
20
3418
2967
>10,000
1600
8200
438
O
H
H
X
a
Binding was performed with 0.3 nM [125I]MCP-1 and human peripheral blood
R
Ar
mononuclear cells at room temperature.18 IC50 values reported as the average of
two or more determinations.
Figure 3. CCR2 antagonist general structure.
Synthesis of the initial cyclohexenyl derivatives in racemic form
was realized as outlined in Scheme 1. In this synthetic approach,
the R-group is determined by the chosen symmetric anhydride
(6a–6b); affording an opportunity to explore SAR of the aryl group
in the latter stage of the synthesis. Syntheses of the d-keto acids
8a–8b were realized by the reaction of symmetrical anhydrides
(6a–6b) with N,O-dimethylhydroxyl-amine, forming intermediates
7a–7b,14 followed by treatment with methyl lithium. The
d-ketoacids could be converted to the dihydropyranones (9a–9b)
by treatment with a reagent such as thionyl chloride or oxalyl
chloride in benzene.15 Condensation of intermediates 9a–9b with
elimination and subsequent sodium hydroxide hydrolysis, afforded
the desired racemic cyclohexene compounds (14a–14b). These
were obtained as a 4–5:1 mixture of diastereomers, similar to
the initial lead analog 5. These compounds were then tested in a
CCR2 binding assay to elucidate SAR trends leading toward
improved CCR2 binding potency.
The initial SAR optimization study focused on understanding
the relative importance of the methyl substitution (Table 1).
Replacing the methyl in 5 with hydrogen (15) resulted in a 5-fold
decrease in CCR2 potency. Preliminary aryl group SAR in the des-
methyl series indicated that meta and para substitutions were tol-
erated, while ortho substitutions were less active (15 and 16 versus
17).
We were particularly encouraged by the 8-fold improvement in
potency of the des-methyl compound 20 relative to the 4-chloro-
phenyl counterpart 15. Additionally, we viewed the marginal
potency enhancement observed in compound 18 compared to
compound 16 as an opportunity to increase CCR2 potency in future
exemplars. Based on these findings, we surmised that optimization
of the aryl group substitution in the original methyl series should
provide analogs having improved potency relative to 5. To that
end, more extensive aryl group SAR was subsequently conducted
within the more potent methyl series (Table 2). Mono-substituted
aryl group analogs (21–27) were prepared as previously described
methyl
a-lithioacetate in tetrahydrofuran yielded a mixture of
ring-opened methyl ketones (10a–10b) and cyclic ketone products
(11a–11b).16 Complete cyclization and dehydration of the mixture
to form enones 12a–12b were achieved by heating a benzene
solution of this mixture in the presence of methanesulfonic acid.17
The subsequent ketoesters 13a–13b were then prepared via
conjugate 1,4-addition employing aryl-magnesium halide reagents
in the presence of copper(I) chloride.
Alternatively, the ketoesters could be synthesized by treating
the enones with a pre-formed lithium diarylcuprate (Ar2CuLi).
The resulting b-ketoesters were reduced to the b-hydroxyesters
with sodium borohydride. Finally, conversion of the hydroxyl
group to a mesylate leaving group, followed by DBU-mediated
O
R
R
O
O
O
R
O
O
b
c
d
a
O
O
N
OH
OH
R
O
O
O
7a-7b
8a-8b
9a-9b
6a (R=H)
6b (R=Me)
O
O
O
O
O
O
O
O
e
f
g-j
OH
O
O
O
O
+
O
R
R
R
R
R
Ar
OH
11a-11b
Ar
13a-13b
(+/-)
(+/-)
10a-10b
12a-12b
14a-14b
Scheme 1. General synthesis of racemic cyclohexenyl derivatives. Reagents: (a) HN(Me)OMe, pyridine, CH2Cl2 (98%); (b) MeLi, THF (100%); (c) SOCl2, C6H6; (d) (i-Pr)2NH, n-
BuLi, CH3CO2CH3, THF (24% over 2-steps); (e) MeSO3H, C6H6 (72%); (f) ArMgBr, CuCl, THF or ArBr, n-BuLi, CuCN, Et2O (68%); (g) NaBH4, MeOH; (h) Ms2O, Et3N, CH2Cl2; (i) DBU,
C6H6; (j) 3.0 N NaOH, MeOH (46% over 3-steps).
Please cite this article in press as: Brown, G. D.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.11.051

G. D. Brown et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
Table 2
Table 3
Aryl group modifications in methyl series
Carboxylic acid replacements
O
O
O
O
S
N
R'
OH
H
Ar
(+/-)
(+/-)
Cl
Cl
R0
Compound
Ar
CCR2
IC50a(nM)
Compound
CCR2
IC50 (nM)
a
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Ph
3-F Ph
9737
3700
928
284
254
1685
2637
3700
550
328
301
40
41
42
Me
CF3
Ph
5406
6605
150
3-Me Ph
3-(CF3) Ph
3-Ph–Ph
4-F Ph
4-Me Ph
2,3-Me2 Ph
2,5-Me2 Ph
3,4-Me2 Ph
3,5-Me2 Ph
3,5-(CF3)2 Ph
3,4-F2 Ph
3-Me, 4-F Ph
3,5-Cl2 Ph
Binding was performed with 0.3 nM [125I]MCP-1 and human peripheral blood
a
mononuclear cells at room temperature.18 IC50 values reported as the average of
two or more determinations.
carboxylic acid counterpart 36 (Table 3). The most notable activity
from these efforts was the N-acyl phenyl substituted sulfonamide
42 which was within two-fold of the potency of 36 (IC50’s of
150 nM versus 80 nM, respectively).
744
3480
>10,000
114
80
69
Finally, R-group methyl replacement strategies were investi-
gated. These analogs, incorporating the 3-trifluoromethyl, 4-chloro
phenyl aryl sidechain, were synthesized as outlined in Scheme 3.
Our synthetic approach, which relied upon the preparation of
the key hydroxymethyl analog 51, began with the one-pot
Michael–Michael–Dieckmann cyclization to form the functional-
ized cyclohexanone 44.19 Subsequent sodium chloride–dimethyl-
sulfoxide mediated decarboxylation of 44 followed by
methylation resulted in a 5:1 mixture of major, undesired (46a)
and minor, desired (46b) diastereomers. The diastereomers were
separated via normal phase chromatography methods. The unde-
sired diastereomer was epimerized under basic conditions and
separated by chromatography. Employing synthetic transforma-
tions well established in the chemical literature, the targeted com-
pounds 53a and 53b were realized. A detailed account of this
synthesis (Scheme 3) is described in an earlier report.20
The hydroxymethyl analog 53a was active versus CCR2
(IC50 = 210 nM); however, it was approximately 20-fold less potent
than the unfunctionalized methyl derivative 38 (Table 4). A limited
effort to introduce oxygen linked R-groups (53b) led to a more sig-
nificant decrease in CCR2 potency. In light of these results, further
R-group modifications were discontinued. Having completed a sur-
vey of three primary sites of diversity, compounds 36 and 38
(Table 2) were identified as the most potent compounds with
CCR2 binding affinity within the cyclohexenyl series. Subsequent
efforts were undertaken to resolve the racemic compounds into
their individual enantiomers for CCR2 binding affinity determina-
tions. Resolution of racemic 36 was carried out by chiral prepara-
tive HPLC to afford (ꢀ)-36 and (+)-36 in 99.9% and 99.5%
enantiomeric excess, respectively (Scheme 4). Subsequent testing
of the enantiomers for CCR2 binding affinity determined that the
(ꢀ)-antipode was the active isomer (IC50 = 36 nM) whereas the
(+)-antipode had significantly reduced activity toward CCR2
(IC50 = 5779 nM).
3,4-Cl2 Ph
3-Me, 4-Cl Ph
3-(CF3), 4-Cl Ph
11
249
6-(4-Methylnaphthalen-1-yl
Binding was performed with 0.3 nM [125I]MCP-1 and human peripheral blood
a
mononuclear cells at room temperature.18 IC50 values reported as the average of
two or more determinations.
and were tested for CCR2 binding affinity. Efforts were focused
mainly on meta and para substitutions since these appeared to
be tolerated in the des-methyl series. Most notably, the 3-trifluo-
romethyl and 3-phenyl substituted analogs 24 and 25 were found
to be the most potent mono substituted analogs having IC50’s of
284 and 254 nM, respectively.
Subsequent efforts investigated disubstituted analogs (28–39).
While many disubstituted analogs were active, analogs containing
a meta or para chloro group were found to be most potent (35–38).
Most noteworthy was the identification of the 3-trifluoromethyl,
4-chloro analog (38) having >50-fold improvement in binding
affinity relative to 5 (IC50 = 11 nM versus 700 nM, respectively).
Interestingly, analogs having much larger substituents such as
the naphthalene derivative 39 were more potent than 5, but were
not pursued further due to decreased potency relative to the most
potent analog 38. Having identified 38 as a promising, potent CCR2
antagonist, we explored carboxylic acid group modifications on a
limited basis. We elected to investigate acylsulfonamides in the
3,4-dichlorophenyl series using acid 36 and standard coupling con-
ditions to afford 40–42 as outlined in Scheme 2.
Unfortunately, when tested for CCR2 binding affinity, the acyl
sulfonamides 40–42 showed a loss in potency relative to their
O
O
O
O
S
N
R'
OH
Cl
The absolute configuration of the enantiomers was unambigu-
ously determined by single X-ray crystallography of (ꢀ)-36 which
showed this isomer to be ((1S,5S)-30,40-dichloro-1,5-dimethyl-
1,4,5,6-tetrahydro-[1,10-biphenyl]-2-carboxylic acid (Fig. 4).21
The more potent CCR2 analog, (+/ꢀ)-38, was resolved employ-
ing chromatography conditions similar to the conditions used in
the resolution of (+/ꢀ)-36. Similarly (ꢀ)-38 retained the CCR2
activity (IC50 = 9 nM) while (+)-38 was much less potent
(IC50 = 4500 nM) (Scheme 5).
H
a
(+/-)
(+/-)
Cl
Cl
36
Cl
40-42
Scheme 2. Synthesis of cyclohexenyl sulfonamide derivatives. Reagents: (a) H2NS
(O)2R, EDC, DMAP, CH2Cl2/DMF.
Please cite this article in press as: Brown, G. D.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.11.051

4
G. D. Brown et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
O
O
O
O
O
O
O
OH
O
a
b
c
O
d
CF3
Cl
O
CF3
Cl
O
O
CF3
Cl
O
CF3
O
CF3
O
Cl
46b
enantiomeric pair
(1-RS, 3-RS)
Cl
46a
major diastereomer (1-SR, 3-RS)
46b minor diastereomer (1-RS, 3-RS)
45
43
44
O
O
OTf
O
O
OH
Ph
O
O
Si
e-g
Si
h
i-j
l
HO
R'O
Ph
Ph
CF3
CF3
CF3
CF3
Cl
Cl
Cl
k
Cl
51
49
50
l
53a (R = H)
53b (R = t-Bu)
O
O
O
CF3
Cl
52
Scheme 3. R-group modification of cyclohexenyl modulators of CCR2. Reagents: (a) (i-Pr)2NH, n-BuLi, CH2CHCO2CH3, THF (68%); (b) DMSO, NaCl, H2O (75%); (c) (i-Pr)2NH,
n-BuLi, MeI, THF (55%); (d) NaOMe, MeOH (51%); (e) NaBH4, MeOH (78%); (f) imidazole, Ph2Si(t-Bu)Cl, DMF (73%); (g) Celite, PCC, CH2Cl2 (99%); (h)NaHMDS, Comin’s reagent,
THF (99%); (i) MeOH, (i-Pr)2NEt, dppf, Pd(OAc)2, CO, DMF (76%); (j) TBAF, AcOH, THF (94%); (k) PTSA, isobutylene, DCM ꢀ78 °C—rt (52%); (l) 1.0 N NaOH, EtOH (63%).
Table 4
R-group functionalization
O
OH
R
(+/-)
CF3
Cl
Compound
R
CCR2
IC50 (nM)
a
38
CH3
11
53a
53b
CH2OH
CH2O(t-Bu)
210
6151
a
Binding was performed with 0.3 nM [125I]MCP-1 and human peripheral blood
mononuclear cells at room temperature.18 IC50 values reported as the average of
two or more determinations.
O
O
O
Figure 4. Single X-ray crystal structure of (ꢀ)-36.
OH
Cl
OH
Cl
OH
Cl
CCR5. Compound (ꢀ)-38 was also evaluated in CCR2-mediated
functional assays which showed potent antagonist activity in
both chemotaxis (IC50 = 20 nM) and calcium flux (IC50 = 15 nM)
consistent with it’s CCR2 binding affinity.12,22 Furthermore,
when screened against multiple ion channels (hERG, sodium, and
calcium) to assess any potential cardiovascular liabilities as well
as against a panel of cytochrome P450 isoforms for drug–drug
interaction concerns, no significant profiling liabilities were
Cl
Cl
Cl
(+/-)-36
IC50 = 80 nM
(-)-36
36.0 nM
(+)-36
5779 nM
Scheme 4. Resolution and CCR2 binding affinity of (+)-36 and (ꢀ)-36.
observed at concentrations up to 10
l
M. Compound (ꢀ)-38 also
showed marginal intrinsic permeability in vitro (PAMPA
Pc = 340 nm/s). However, when dosed orally in the mouse, (ꢀ)-38
exhibited reasonable bioavailability (%F = 67), but high clearance
(17.2 ml/min/kg) which was attributed to poor microsomal
stability since (ꢀ)-38 showed high clearance rates in in vitro
Having identified (ꢀ)-38 as having high affinity for CCR2, we
further profiled this analog in available selectivity assays which
showed (ꢀ)-38 to be >100 fold selective versus CCR1, CCR3, and
Please cite this article in press as: Brown, G. D.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.11.051

G. D. Brown et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
O
O
O
Acknowledgements
OH
CF3
OH
OH
The authors would like to thank Stephen Wrobleski and Wil-
liam Pitts for their guidance, insight, and continuous support dur-
ing manuscript preparation. We gratefully acknowledge Dauh-
Rurng Wu for chiral separations.
(+/-)
CF3
CF3
Cl
Cl
Cl
References and notes
(+/-)38
CCR2 IC50 = 11 nM
(-)-38
(+)-38
1. Geissmann, F.; Jung, S.; Littman, D. R. Immunity 2003, 19, 71.
2. Charo, I. F.; Ransohoff, R. M. N. Engl. J. Med. 2006, 354, 610.
3. Feria, M.; Díaz-González, F. Expert Opin. Ther. Pat. 2006, 16, 49.
4. Carter, P. H.; Cherney, R. J.; Mangion, I. K. Annu. Rep. Med. Chem. 2007, 42, 211.
5. Xia, M.; Sui, Z. Expert Opin. Ther. Pat. 2009, 19, 295.
6. Struthers, M.; Pasternak, A. Curr. Top. Med. Chem. 2010, 10, 1278.
7. Cai, Dongwei; Fleitz, Fred; et al. WO2005044795, 2005.
8. (a) Carter, P. H.; Duncia, J. V. et al. WO 2008014360, 2008.; (b) Cherney, R. J.;
Mo, R.; Meyer, D. T., et al. Bioorg. Med. Chem. Lett. 2010, 20, 2425; (c) Carter, P.
H.; Brown, G. D., et al. ACS Med. Chem. Lett. 2015, 6, 439.
9. Goodman, K. B.; Sehon, C. A.; Cleary, P. A.; et al. WO 2007067875, 2007.
10. Bengtsson, Boel Ase.; Blackaby, Wesley; et al. WO 2011114148, 2011.
11. Carter, P. H. Expert Opin. Ther. Pat. 2013, 23, 549.
12. For biological assay details, see: Carter, P. H.; Brown, G. D.; et al. WO
2010009068, 2010.
13. The trans-to-cis ratio was determined using standard analytical chiral
separation methods.
14. (a) Jacobi, P. A. et al Tetrahedron 1987, 43, 5475; (b) Denmark, S. E.; Fujimori, S.
Synlett 2001, 1024. Special issue.
15. Belmont, D. T. et al J. Org. Chem. 1985, 50, 4102.
16. Methyl a-lithioacetate was prepared from lithium diisopropylamide (LDA) and
methyl acetate as described in Ref. 14 and Reference 14 therein.
17. Thompson, S. K.; Heathcock, C. H. J. Org. Chem. 1992, 57, 5979.
18. Yoshimura et al J. Immunol. 1990, 145, 292.
19. Posner, G. H.; Shulman-Roskes, E. M. J. Org. Chem. 1989, 54, 3514.
20. Carter, P. H.; Brown, G. D. et al. WO 2010009068, 2010.
21. The Cambridge Crystallographic Data Centre (CCDC) number is 1419348.
22. When compounds were incubated with monocytes without MCP-1 present,
calcium flux inhibition was not changed, indicating the compounds were
antagonists and not partial agonists.
9.0 nM
4500 nM
Scheme 5. Resolution and CCR2 binding affinity of (+)- and (ꢀ)-38.
microsomal assays across four species (human/mouse/rat/cyno =
0.032/0.076/0.114/0.084 nM/min, respectively).
In summary, a novel cyclohexenyl series of low molecular
weight, rigid CCR2 antagonists has been discovered. Three
areas of modification (R-group, aryl-group, and acid deriva-
tives) were explored. Significant improvements in CCR2 binding
affinity were realized by structure–activity relationship studies
resulting in the identification of (+/ꢀ)-38 as a potent CCR2
ligand (IC50 = 11 nM). Resolution and determination of the
absolute configuration by single X-ray crystallography identi-
fied (ꢀ)-38, (4S,6S)-6-(4-chloro-3-(trifluoromethyl)phenyl)-4,6-
dimethyl-cyclohex-1-enecarboxylic acid, as having the most
potent CCR2 binding affinity within this novel series
(IC50 = 9.0 nM) and showing potent CCR2-mediated functional
activity in both chemotaxis (IC50 = 20 nM) and calcium flux
(IC50 = 15 nM). Further efforts to identify additional promising
CCR2 antagonists will be reported in due course.
Please cite this article in press as: Brown, G. D.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.11.051