Capped diaminopropionamide-glycine dipeptides are inhibitors of CC chemokine receptor 2 (CCR2)

Article abstract of DOI:10.1016/j.bmcl.2007.07.028

A new series of CCR2 antagonists has been discovered that incorporates intramolecular hydrogen bonding as a strategy for rigidifying the scaffold. The structure-activity relationship was established through initial systematic modification of substitution pattern and chain length, followed by independent optimization of three different substituents (benzylamine, carboxamide, and benzamide). Several of the acyclic compounds display 10-30 nM binding affinity for CCR2. Moreover, these antagonists are able to block both MCP-1-induced Ca²⁺ flux and monocyte chemotaxis, and are selective for binding to CCR2 over CCR1 and CCR3.

Full text of DOI:10.1016/j.bmcl.2007.07.028

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5455–5461
Capped diaminopropionamide–glycine dipeptides are inhibitors
of CC chemokine receptor 2 (CCR2)
Percy H. Carter,* Gregory D. Brown, Sarah R. Friedrich, Robert J. Cherney,
Andrew J. Tebben, Yvonne C. Lo, Gengjie Yang, Heather Jezak, Kimberly A. Solomon,
Peggy A. Scherle and Carl P. Decicco
Research and Development, Bristol-Myers Squibb Company, Princeton, NJ 08543-4000, USA
Received 25 April 2007; revised 7 July 2007; accepted 9 July 2007
Available online 21 July 2007
Abstract—A new series of CCR2 antagonists has been discovered that incorporates intramolecular hydrogen bonding as a strategy
for rigidifying the scaffold. The structure–activity relationship was established through initial systematic modification of substitution
pattern and chain length, followed by independent optimization of three different substituents (benzylamine, carboxamide, and
benzamide). Several of the acyclic compounds display 10–30 nM binding affinity for CCR2. Moreover, these antagonists are able
to block both MCP-1-induced Ca2+ flux and monocyte chemotaxis, and are selective for binding to CCR2 over CCR1 and CCR3.
Ó 2007 Elsevier Ltd. All rights reserved.
CC chemokine receptor 2 (CCR2) is the primary chemo-
kine receptor on monocytes recruited to sites of inflam-
mation;1 it is also the exclusive receptor for monocyte
chemoattractant protein-1 (MCP-1), a well-charac-
terized pro-inflammatory cytokine. Targeting this
pro-inflammatory pair through both genetic and phar-
macologic approaches reduces inflammation in classic
rodent models of rheumatoid arthritis, multiple sclero-
sis, and atherosclerosis.2 These data, together with the
fact that both CCR2ꢀ/ꢀ and MCP-1ꢀ/ꢀ are viable,
fertile, and devoid of any overt phenotype, have
prompted researchers to pursue CCR2 antagonism as
an approach toward treating inflammatory diseases.2,3
(Fig. 1). In this Letter, we describe our initial test of this
hypothesis.
Scheme 1 illustrates representative examples of the
straightforward chemistry used to synthesize these mol-
ecules.5 The commercially available Na-Boc, Nx-Cbz
diaminoacids were esterified or amidated. The resulting
products were deprotected (TFA) and coupled with a
substituted hippuric acid. Hydrogenolysis of the Cbz
and reductive amination provided final products. In
the case of diaminopropionic acid (Dap)-derived prod-
ucts (n = 1), the compounds could be derivatized further
through reduction, hydrolysis, or amidation. Longer
chain lengths (n = 2 and 3) required that the amide be
introduced in the first step in order to avoid intramolec-
ular cyclization in the late-stage coupling. Since the
Others have disclosed that cyclic diamines such as 3-
aminopyrrolidine and 3-aminomethylpiperidine could
serve as scaffolding elements for potent CCR2 antago-
nists (binding IC50’s < 100 nM).4 We postulated that
the 3-aminopyrrolidine subunit could be replaced with
an acyclic diamine motif, provided that a functional
group capable of intramolecular hydrogen bonding
was suitably positioned along the diamine backbone
O
H
N
Ar'
N
H
N
H
Ar
Ar
O
O
O
H
N
Mode A
X
R
Ar'
N
N
H
Ar
O
H
N
R'
N
H
O
Ar'
N
H
Mode B
O
X
Keywords: CCR2; Small molecule; Antagonist; G-protein coupled re-
ceptor; Chemokine; Inflammation.
Figure 1. Use of intramolecular hydrogen bonding to constrain the
conformation of an acyclic scaffold while introducing additional
specificity/recognition elements.
*
Corresponding author. Tel.: +1 609 252 4144; e-mail: percy.carter@
bms.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.028

5456
P. H. Carter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5455–5461
O
O
H
H
c, d
e, f
N
CF3
N
CF3
NHBoc
N
N
n
CbzHN
n
N
CbzHN
n
H
H
H
X
O
X
O
X
g
X = CO2H
X = CO2Me
h
X = CH2OH
X = CONH2
a
b
X = CO2Me
X = CONHR
for n = 1
i
X = CO2H
Y
O
Y
O
H
H
N
l - n
j, k
N
CF3
CF3
NHFmoc
AllocHN
Z
N
Z
n
n
AllocHN
n
H
O
O
NH2
O
NH
O
O
NH
PAL-R
PAL-R
Scheme 1. Representative solution- and solid-phase synthesis of acyclic CCR2 inhibitors. Non-standard abbreviations: PAL-R = PAL polystyrene
resin. Reagents and conditions: (a) EDC, MeOH; (b) H2NR, HATU, i-Pr2NEt, CH2Cl2/DMF; (c) TFA, CH2Cl2; (d) HO2CCH2NHCO(3-CF3Ph),
BOP, i-Pr2NEt, CH2Cl2/DMF; (e) H2, 5% Pd/C (Degussa), MeOH; (f) 2,4-Me2PhCHO, NaCNBH3, MeOH; (g) LiBH4, MeOH, THF; (h) LiOH,
THF/MeOH/H2O; (i) H3N/MeOH, HATU; (j) piperidine/NMP wash, then Fmoc-amino acid, HBTU, HOBt, i-Pr2NEt, NMP; (k) piperidine/NMP
wash, then HO2C(3-CF3Ph), HBTU, HOBt, i-Pr2NEt, NMP; (l) Pd(PPh3)4, 37:2:1 CHCl3/AcOH/N-methyl morpholine; (m) 2,4-Me2PhCHO,
NaCNBH3, 1% AcOH/dimethylacetamide; (n) 95:5 TFA/H2O, triethylsilane.
primary amide derivatives proved to be potent inhibitors
(see Table 1), we employed solid-phase peptide synthesis
protocols to access some compounds.6 This approach al-
lowed for the rapid variation of all regions of the scaf-
fold from the starting Na-Fmoc, Nx-alloc diaminoacids
(Scheme 1).
simple alkyl sidechains (1g, 1h) are at least 10-fold less
active than their functionalized counterparts.
To test the hypothesis further, we synthesized a group of
compounds that altered the proposed hydrogen bonding
inherent in the diaminopropionic acid subunit (Fig. 2).
Alternate positioning of the amide or ester grouping
along the diaminoethane moiety provides inhibitors
with weaker potency (see 3a, 3b in Fig. 2;7 the (S)-vari-
ant of an analog of 3b exhibits similar activity, data not
shown). Inhibitors derived from elongation of either the
sidechain or main chain are less potent than their Dap
analogs (cf. 2 and 4, 5; analogs from longer main chain
lengths are weaker still). Notably, the rank-order activ-
ity for the three surveyed amide sidechain groups is
altered as the length of the main chain increases (cf.
2d–2f and 5d–5f). Furthermore, amine methylation ex-
erts differential effects on products from the Dap
(n = 1) and diaminobutyric acid (Dab, n = 2) scaffolds:
Table 1 lists the CCR2 binding data obtained with a pi-
lot set of compounds derived from Dap. The 2,4-dim-
ethylbenzyl
group,
glycine
spacer,
and
3-
trifluoromethylbenzamide grouping were standardized
for the initial study. Three observations stand out: (1)
the inhibitors derived from (S)-Dap are more potent
than those from (R)-Dap; and (2) sidechain amides are
more potent than the methyl ester, carboxylic acid,
and primary alcohol, with the large activity difference
between the isosteric primary amide and carboxylic acid
being particularly noteworthy; (3) the compounds with
Table 1. Initial test of the hypothesis shown in Figure 1 through
probing of sidechain identity and absolute stereochemistry
X
O
O
H
H
N
N
O
O
Ar'
N
H
N
Ar
Ar'
N
H
N
Ar
H
N
H
N
H
H
O
O
Ar'
N
H
N
H
Ar
Ar'
N
H
N
H
Ar
t-BuHN(O)C
3a, X = CO2Me, 15 μM (1)
3b, X = CONHEt, 22 μM (1)
X
O
X
O
4, 0.59 0.23 μM
Ar' = 2,4-Me2Ph, Ar = 3-CF3Ph
1
2
O
O
H
H
N
N
a
Series, X group
CCR2 binding IC50 (lM)
Ar'
N
H
N
Ar
Ar'
N
N
Ar
n
2
H
H
O
O
Me
1
2
O
Y
O
Y
5d, Y= NH2, 3.0 1.6 μM
5e, Y= NHEt, 0.41 0.09 μM (5)
5f, Y= NHt-Bu, 0.77 0.17 μM
6, n = 1, Y = NHtBu, 12% @ 1 μM
a, CO2Me
b, CH2OH
c, CO2H
1.0 0.9 (2)
2.2 0.1 (2)
1.5 0.5 (2)
0.32 0.22 (2)
5.5 1.8 (2)
7, n = 2, Y = NHEt, 0.15 0.08 μM
31% I at 10 lM
0.63 0.10 (2)
0.56 0.29 (2)
0.11 0.03 (3)
35% I at 10 lM
20% I at 10 lM
O
O
d, CONH2
e, CONHEt
f, CONH-t-Bu
g, Me
0.066 0.005 (4)
0.11 0.04 (2)
0.048 0.011 (13)
Not tested
H
H
N
N
Ar'
N
H
N
Ar
Ar'
N
N
Ar
H
H
O
O
O
Y
8, 0.017 0.003 μM (20)
2i, Y= NHMe, 0.23 0.02 μM
2j, Y= NMe2, 0.25 0.01 μM
h, i-Pr
Not tested
a Binding was performed using 0.3 nM
[
125I]MCP-1 and human
peripheral blood mononuclear cells at 23 °C (Ref. 5). IC50 values (n)
are reported as mean SD (n = 2) or mean SEM (n > 2). For less
potent compounds, % inhibition (% I) at a fixed inhibitor concen-
tration is listed (n = 1).
Figure 2. Additional tests of the hypothesis shown in Figure 1. CCR2
binding IC50 values are listed as described in the footnote to Table 1
(n = 2 unless indicated). Ar = 3-CF3–phenyl and Ar0 = 2,4-(Me)2–
phenyl.

P. H. Carter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5455–5461
5457
in the (S)-Dab series, tertiary amines are more active
Me
Me
O
Steps c - f,
Scheme 1
b
17a
than their secondary counterparts, whereas tertiary
amines are >10-fold less active than their secondary
counterparts in the (S)-Dap series (cf. 2f and 6 vs 5e
and 7; several other examples in both series exhibit sim-
ilar shifts, data not shown). Note, however, N-methyla-
tion of the amide is tolerated in the (S)-Dap series (cf. 2i
and 2j).
NHBoc
NHBoc
NHEt
HO
a
N3
42%
X
18 X = CO2H
20
19 X = CONHEt
90%
c, d
24%
Me
Me
O
Steps c - f,
Scheme 1
e, f
51%
17b
NHBoc
NHEt
NHBoc
NHEt
Taken together, Table 1 and Figure 2 show that acyclic
N-benzylated Dap–Gly dipeptides are potent CCR2
antagonists. Indeed, the binding potency of 2f ap-
proaches that of cyclic lead structure 8. Collectively,
the data are consistent with the involvement of the intra-
molecular hydrogen bonding postulated in Figure 1,
with Mode A being optimum (sidechain acts as H-bond
acceptor). This conclusion is also supported by confor-
mational modeling in water, in which the internal H-
bond was found in >99.8% of Boltzmann-weighted pop-
ulation (see Supporting Information). An alternative
hypothesis is that the amide sidechain moiety interacts
beneficially with CCR2.
HO
N3
O
21
22
Scheme 3. Synthesis of b-methyl compounds 17a and b. Reagents and
conditions: (a) EtNH2, BOP, CH2Cl2; (b) DEAD, Ph3P, HN3, PhH; (c)
DEAD, Ph3P, HO2C(p-NO2Ph), PhH; (d) LiOH, 2:2:1 THF/MeOH/
H2O; (e) Ms2O, i-Pr2NEt, CH2Cl2; (f) NaN3, DMSO, 65 °C.
As illustrated in Table 2, conservative modification of
the Dap and Gly does not enhance the potency of the
inhibitors. Methyl substitution on the a-carbon (10,
11) of the glycine leads to reductions in CCR2 binding,
as does substitution of the Gly a-nitrogen for carbon
(12). The methyl scan of the Dap subunit reveals that
only substitution of a b-methyl group in the (R)-config-
uration (17b) does not impact the binding affinity sub-
stantially. The data in Table 2 may suggest that these
inhibitors bind to CCR2 in a conformation that requires
the wide U, W-tolerance of the unsubstituted amino
acids.
We continued our study by examining conservative
modifications of both the Dap and Gly subunits. While
the synthesis of the majority of the Gly analogs was
facilitated by employing the aforementioned solid-phase
approach using PAL resin as the support (Scheme 1),
the backbone methylation scan of the (S)-Dap subunit
required that we synthesize specialized starting materials
(Schemes 2 and 3).5 The synthesis of the a-methyl com-
pound 16 proceeded from the commercially available
( )-a-methyl serine (Scheme 2); the choice of synthetic
route was ultimately dictated by problems relating to
the steric congestion of the a-alkyl amino acid. The syn-
theses of the b-methyl compounds 17a and 17b began
from L-threonine (Scheme 3). Amide formation followed
by Mitsuonobu inversion of the hydroxyl with hydra-
zoic acid8 provided the a, b-diaminobutyramide equiva-
lent 20; the complementary stereoisomer 22 was
Since the unmodified (S)-Dap–Gly backbone appeared
optimal for CCR2 binding, we focused our efforts on
improving the affinity of series 2 through modification
of the three capping groups: benzylamine, carboxamide,
and benzamide. These groups were readily altered via
chemistry analogous to that shown in Scheme 1,5,10
beginning with Na-Boc, Nb-Cbz (S)-Dap. Table 3 sum-
marizes the remainder of the SAR profile of the carbox-
amide moiety. Although tert-butyl amide 2f is more
potent than primary amide 2d, other simple alkyl amides
and tert-alkyl amides are not as potent (cf. 2f and 2k–
2o). Tertiary amides (see 2j, Fig. 2, and 2y–2aa, Table
3) likewise offer no advantage. Cycloalkyl amides 2p–
2s exhibit similar potencies to each other (90–100 nM),
despite large differences in lipophilicity, but phenyl
amide 2t is 10-fold weaker (1200 nM). The apparent
intolerance toward aromatic rings is also evident in the
100-fold potency difference between benzyl amide 2u
and cyclopropylmethyl amide 2w. The 2,2,2-trifluoro-
ethyl amide 2x shows similar potency to 2w and tert-bu-
tyl 2f. Notably, 2,2,2-trifluoroethylamide also serves as
an effective tert-butylamide replacement in a series of
HIV protease inhibitors.11
produced
via
a
double
inversion
sequence
(19 ! 21 ! 22).9 Compounds 20 and 22 were carried
forward using the chemistry shown in Scheme 1.
NH2
NHCbz
NHCbz
a, b
c - e
41%
HO
HO
O
Ar'
N
21%
Me CO2H
Me CONHEt
Me CONHEt
Boc
( )-14
( )-13
f, g
NH2
h - j
H
H
N
N
Ar'
N
N
Ar'
N
NHCbz
O
H
H
Me
EtHN
Me
O
Boc
O
O
EtHN
( )-16
( )-15
CF3
Scheme 2. Synthesis of a-methyl compound 16. Abbreviations:
Ar = (2,4-Me2)-phenyl. Reagents and conditions: (a) Cbz2O, NEt3,
THF/H2O; (b) EtNH2, HATU, DMAP, CH2Cl2; (c) Dess–Martin
periodinane, Pyr, CH2Cl2; (d) 2,4-Me2PhCH2NH2, NaCNBH3,
MeOH; (e) Boc2O, NEt3, THF/H2O; (f) 1 atm H2, 5% Pd/C (Degussa),
MeOH; (g) HO2CCH2NHCbz, HATU, i-Pr2NEt, cat. DMAP,
CH2Cl2; (h) 1 atm H2, 5% Pd/C (Degussa), MeOH; (i) HO2C(2-
NH2,5-CF3)Ph, HATU, i-Pr2NEt, CH2Cl2; (j) TFA/CH2Cl2.
Initial examination of the benzylamine moiety provided
guidance for further exploration (data not shown): (1)
removal of the benzyl grouping eliminates CCR2 activ-
ity; (2) carbamates, sulfonamides, and amides are inac-
tive, independent of ortho-substitution on the phenyl
ring; and (3) methyl substitution in either stereochemical
configuration on the benzyl methylene is tolerated, but

5458
P. H. Carter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5455–5461
Table 2. Modification of the (S)-Dap–Gly dipeptide backbone
Rβ
Y
O
H
N
Rα
X
CF3
N
H
Z
O
Xa
Y
Z
CCR2 binding IC50 (lM)
b
Compound
Rb
Ra
2d
10
H
H
CONH2
CONH2
CONH2
H
(R)-Me
(S)-Me
NH
NH
NH
CH2
NH
NH
NH
NH
0.066 0.005 (4)
40% I at 10 lM
9% I at 10 lM
7% I at 10 lM
0.11 0.04 (2)
8.7 (1)
H
H
11
12
H
H
c
H
H
CONHEt
CONHEt
CONHEta
CONHEt
CONHEt
H
H
H
H
H
2e
16
H
H
H
Mea,c
H
17a
17b
(S)-Me
(R)-Me
2% I at 10 lM
0.13 0.06 (2)
H
a All amide groups are in the (S)-configuration, save compound 16, which is racemic.
b Binding was performed using 0.3 nM [125I]MCP-1 and human peripheral blood mononuclear cells at 23 °C (Ref. 5). IC50 values (n) are reported as
mean SD (n = 2) or mean SEM (n > 2). For less potent compounds, % inhibition (% I) at a fixed inhibitor concentration is listed (n = 1).
c The 3-trifluoromethylbenzamide was replaced with 2-amino-5-trifluoromethylbenzamide, which confers ꢁ3-fold higher activity (see Table 5).
not enhancing, regardless of whether or not an ortho-
substituent is present on the phenyl ring. Accordingly,
we synthesized approximately fifty simple benzylamines;
the highlights of this investigation are presented in Table
4. Inhibitors containing either the 2-methylbenzyl or 4-
methylbenzyl group display binding affinities that are
within 2-fold of each other, but ꢁ20-fold lower than
antagonists containing the 2,4-dimethylbenzyl group
(cf. 2f and 23, 24), suggesting an additive effect. Notably,
the 10- to 20-fold enhancing effect of the ortho-methyl
group is general for a variety of para-substituents (cf.
29–32 and 34–37). The disubstituted analogs follow
the same rank order as the 4-substituted analogs. Other
comparisons reveal additional steric and electronic
requirements for the benzylamine moiety: (1) while ethyl
is allowed at the ortho position, methoxy is not preferred
there (cf. 24, 25 vs 26, 35); (2) electron-withdrawing
groups are not preferred at the para-position, (see 27,
28, 33; a larger scan confirms this, data not shown);
(3) para-substituents larger than ethyl are poorly toler-
ated (see 43, 44; a larger scan confirms this, data not
Table 4. Representative SAR of the benzylamine substituent
O
H
N
CF3
Ar1
N
N
H
H
O
O
NH
Table 3. Modification of the carboxamide moiety
a
Compound
Ar1
CCR2 binding IC50 (lM)
O
H
2f
2,4-Me2–Ph
4-Me–Ph
2-Me–Ph
2-Et–Ph
0.048 0.011 (13)
0.75 0.13 (2)
0.40 0.17 (2)
0.38 0.05 (4)
37% I at 1.0 lM
20% I at 1.0 lM
7.2 (1)
N
CF3
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
N
H
N
H
O
O
Xa
2,4-(MeO)2–Ph
4-CO2Me–Ph
4-SO2Me–Ph
4-Br–Ph
a
Compound
Xa
CCR2 binding IC50 (lM)
2k
2l
NH-i-Pr
NH-s-Bu
NHC(Me)2Et
0.17 0.03 (2)
0.12 0.05 (2)
0.12 0.04 (2)
0.19 0.07 (5)
0.10 0.08 (2)
0.088 0.034 (2)
0.097 0.054 (2)
0.099 0.029 (5)
0.15 0.01 (2)
1.2 0.8 (2)
1.7 1.1 (2)
1.1 0.1 (4)
4-MeO–Ph
4-Et–Ph
4-Me2-N–Ph
2m
2n
2o
2p
2q
2r
0.53 0.41 (2)
0.43 0.18 (2)
0.90 0.27 (3)
0.16 0.02 (2)
0.047 0.009 (2)
0.036 0.001 (2)
0.036 0.001 (2)
27% I at 1.0 lM
0.67 0.51 (2)
0.12 0.05 (2)
0.69 (1)
NHC(Me)2CH2OH
NH(a-Me)-c-Pr
NH-c-Pr
2-Me,4-NC–Ph
2-Me,4-Br–Ph
2-Me,4-MeO–Ph
2-Me,4-Et–Ph
2-Me,4-Me2N–Ph
3-F,4-Me–Ph
3-O2-N,4-Me–Ph
3-H2-N,4-Me–Ph
2,3-Me2,4-MeO–Ph
2,5-Me2,4-MeO–Ph
4-(n-Bu)–Ph
NH-c-Bu
NH-c-Pent
2s
NH-c-Hex
NHPh
NHCH2Ph
2t
2u
2v
2w
2x
2y
2z
2aa
7.0 0.3 (2)
NHCH2CH@CH2
NHCH2-c-Pr
NHCH2CF3
Pyrrolidine
Morpholine
N(OMe)Me
0.20 0.01 (2)
0.060 0.018 (2)
0.055 0.004 (4)
0.26 0.10 (2)
0.49 0.12 (2)
0.11 0.03 (4)
1.3 (1)
5.6 (1)
4-Ph–Ph
30% I at 1.0 lM
a Binding was performed using 0.3 nM 125I]MCP-1 and human
[
a Binding was performed using 0.3 nM 125I]MCP-1 and human
[
peripheral blood mononuclear cells at 23 °C (Ref. 5). IC50 values (n)
are reported as mean SD (n = 2) or mean SEM (n > 2). For less
potent compounds, % inhibition (% I) at a fixed inhibitor concen-
tration is listed (n = 1).
peripheral blood mononuclear cells at 23 °C (Ref. 5). IC50 values (n)
are reported as mean SD (n = 2) or mean SEM (n > 2). For less
potent compounds, % inhibition (% I) at a fixed inhibitor concen-
tration is listed (n = 1).

P. H. Carter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5455–5461
5459
shown); (4) electron-withdrawing groups are also not
preferred at the meta-position (see 38–40), and even
methyl substitution is modestly detrimental (cf. 35 and
41, 42). Although a number of potent benzylamines
are identified by this scan (Table 4), none clearly surpass
the initial 2,4-dimethylbenzylamine in CCR2 binding
affinity.
Given the positive effects of ortho- and meta-amine sub-
stitution, we examined N-substituted derivatives. As
shown in Table 5, analogs with ortho-alkylamines, ur-
eas, or carbamates are all potent antagonists (represen-
tative groups shown in 53–58), but none are clearly
superior to 50 in binding. The meta-amine can also be
substituted with a variety of substituents (representative
groups shown in 59–64). Small groups are well-tolerated
(see 59, 60), but large groups appear to reduce activity
more sharply than in the ortho series (cf. 54–58 and
61–64). Attempts to improve the affinity of each of the
illustrated groups (alkyl, benzyl, urea, carbamate) in
both the ortho- and meta-series have been unsuccessful
(data not shown).
We examined approximately forty different benzamides
and highlights of this investigation are presented in
Table 5; more so than the benzylamine, our survey of
the benzamide was influenced by unpublished work in
a separate series.12 Replacing the 3-CF3 substituent with
less hydrophobic groups is detrimental, but the more
lypophilic pentafluoroethyl group does not improve
activity (cf. 2f and 46–49). Anthranilamide 50 and
meta-aniline 51 both exhibit improved affinity relative
to benzamide 2f. The ortho-amine also improves the po-
tency of the meta-trifluoromethoxy benzamide (cf. 47,
52). As expected, replacement of the 3-CF3 benzamide
with the 2-NH2, 5-CF3 benzamide enhances the activity
of analogs containing alternative amides and benzylam-
ines (see 65, 66, Table 6). Likewise, both the 2-NH2,
5-CF3 benzamide and 3-NH2, 5-CF3 benzamide enhance
the activity of the (S)-Dab-based series (i.e., analogs of
5f). However, none of the Dab-based inhibitors exhibit
equivalent potency to the best Dap-based inhibitors
(data not shown).
We studied the ability of our more potent compounds to
antagonize the functional activity of MCP-1 on CCR2-
bearing cells. As shown in Table 6, the potency of a gi-
ven compound for inhibiting the MCP-1-induced Ca2+
flux can be readily predicted from the binding IC50,
but the activity in the monocyte chemotaxis assay can-
not be as easily extrapolated. The installation of the
ortho-amine functionality on the benzamide appears to
enhance chemotaxis activity (cf. 2d and 65; also 2f and
50). Alkyl substitution on either the ortho- or meta-ami-
no groups improves chemotaxis antagonism still further
(cf. 50 and 53, 54, 57; cf. 51 and 60). Thus, simple block-
ade of chemokine binding is not sufficient to induce
equipotent functional antagonism in the chemotaxis as-
say. Rather, specific structural features enable enhanced
inhibition of cell migration. This phenomenon has been
documented previously with a series of CCR3 antago-
nists.13 Notably, the compounds do not induce Ca2+ flux
when incubated with monocytes in the absence of MCP-
1 (data not shown), indicating that they are antagonists,
and not partial agonists. Additionally, the apparent IC50
for Ca2+ flux inhibition by 50 increases with increasing
concentrations of MCP-1 (data not shown), indicating
that this compound is a competitive antagonist of
CCR2.
Table 5. Representative SAR of the benzamide moiety
O
H
N
N
N
Ar2
H
H
O
O
NH
Compound
Ar2
CCR2 binding
a
IC50 (lM)
2f
3-CF3Ph
0.048 0.011 (13)
0.065 0.015 (2)
0.19 0.06 (3)
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
3-CF2CF3Ph
3-OCF3Ph
3-SCF3Ph
3-CF2HPh
We also examined the selectivity profile of this series of
CCR2 antagonists. None of the studied inhibitors exhib-
its significant binding to CCR1 (all <30% inhibition at
10 lM). These compounds also generally show high
selectivity for binding to CCR2 relative to CCR3 (Table
6). The combination of tert-butylamide and ortho-ani-
line erodes this selectivity to ꢁ100-fold when compared
to the primary amide, substituted ortho-anilines, or the
meta-aniline. We screened 50 against a broad panel of
receptors, ion channels, and transporters (>100 targets).
Compound 50 showed <30% inhibition at 0.1 lM
against all of these targets. At higher concentrations
(10 lM), it showed >70% inhibition of the histamine
H2, sigma 1, CCR5, and opiate (mu) receptors, as well
as the Na+ and L-Ca2+ channels.
0.15 0.03 (2)
0.40 0.18 (2)
2-NH2,5-CF3Ph
3-NH2,5-CF3Ph
2-NH2,5-OCF3Ph
2-NHEt,5-CF3Ph
2-NHBn,5-CF3Ph
0.011 0.004 (7)
0.023 0.000 (2)
0.028 0.019 (2)
0.013 0.003 (4)
0.019 0.007 (2)
0.023 0.004 (2)
0.065 0.004 (2)
0.015 0.001 (2)
0.090 0.016 (2)
0.016 0.008 (2)
0.019 0.008 (2)
0.11 0.07 (2)
2-NH-i-Bu,5-CF3Ph
2-(NHCO2-i-Pr),5-CF3Ph
2-(NHCONH-i-Pr),5-CF3Ph
2-(NHCO-c-Hex), 5-CF3Ph
3-NHMe,5-CF3Ph
3-NHEt,5-CF3Ph
3-NHBn,5-CF3Ph
3-NH-i-Bu,5-CF3Ph
3-(NHCO2Et),5-CF3Ph
3-(NHCOc-Hex),5-CF3Ph
0.13 0.06 (3)
0.31 0.10 (2)
A number of chemokine receptor antagonists interact
with a conserved glutamic acid in VII:06.14 Indeed, oth-
ers have shown that this residue is a key interaction site
for four distinct CCR2 antagonist series, including the
series represented by cyclic lead 8.14 In studies using a
CCR2/3 chimera,15 we have confirmed that 50 also
25% I at 1.0 lM
a Binding was performed using 0.3 nM 125I]MCP-1 and human
[
peripheral blood mononuclear cells at 23 °C (Ref. 5). IC50 values (n)
are reported as mean SD (n = 2) or mean SEM (n > 2). For less
potent compounds, % inhibition (% I) at a fixed inhibitor concen-
tration is listed (n = 1).

5460
P. H. Carter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5455–5461
Table 6. Functional antagonism and chemokine receptor selectivity of the (S)-Dap–Gly inhibitors
O
H
N
CF3
N
H
N
H
O
O
Xa
Y
Compound
Xa
Y
hCCR2 binding,
a
IC50 (nM)
Ca2+ Flux,
b
IC50 (nM)
Chemotaxis,
c
IC50 (nM)
hCCR3 binding,
d
IC50 (nM)
2d
65
2f
NH2
NH2
H
2-NH2
66 5 (4)
37 8 (6)
48 11 (13)
11 4 (7)
20 6 (6)
13 3 (4)
69 17
82 68
>1500 (n = 2)
320
>3000
>10,000
>10,000
>10,000
NH-t-Bu
NH-t-Bu
NH-t-Bu
NH-t-Bu
NH-t-Bu
NH-t-Bu
NH-t-Bu
NH-t-Bu
H
2-NH2
2-NH2
52 30 (3)
8
50
66e
53
54
57
51
60
7
260 20
0
750 180 (3)
1,500 370
>10,000
37 21
84
57
150
2-NHEt
2-NHBn
2-NHC(O)NH-i-Pr
3-NH2
100 50 (3)
39 22
240
19
15
23
19
7
1
0
8
>10,000
58
52
ꢁ10,000
>10,000
>10,000
720 130
190 30
3-NHEt
58
Data are reported in one of three formats: result of a single determination; mean SD (n = 2, n not listed); or mean SEM (n indicated). Indication
of an IC50 as ‘‘>’’ simply refers to <50% response at the indicated concentration (n = 1 unless otherwise indicated).
a Antagonism of binding of 0.3 nM [125I]MCP-1 to human PBMCs at 23 °C (Ref. 5).
b Antagonism of Ca2+ flux induced by 10 nM MCP-1 in with THP-1 cells at 23 °C (Ref. 5).
c Antagonism of chemotaxis of human PBMCs induced by 10 nM MCP-1 at 37 °C (Ref. 5).
d Antagonism of binding of 0.15 nM [125I]Eotaxin to CHO cells expressing human CCR3 (Ref. 13).
e 2-Me,4-Br–PhCH2 replaces 2,4-Me2PhCH2.
capitalizes on this interaction: its binding to E291Q and
E291A mutants was reduced by 470- and 345-fold,
respectively, relative to binding to the wildtype.
References and notes
1. Geissmann, F.; Jung, S.; Littman, D. R. Immunity 2003,
19, 71.
´
´
2. (a) Feria, M.; Dıaz-Gonzalez, F. Exp. Opin. Ther. Patents
2006, 16, 49; (b) Dawson, J.; Miltz, W.; Wiessner, C. Exp.
Opin. Ther. Targets 2003, 7, 35; (c) Carter, P. H. Curr.
Opin. Chem. Biol. 2002, 6, 510.
In summary, we have discovered a new series of CCR2
antagonists. Modification of the scaffold through back-
bone methylation and alteration of spacing was not
fruitful, whereas alteration of the terminal substituents
provided for enhancements in potency. The most active
compounds exhibit binding affinities <50 nM, and are
antagonists of both Ca2+ and chemotaxis responses in-
duced by MCP-1. These compounds also show good
selectivity for binding to CCR2 relative to CCR1 and
CCR3. Our hypothesis is that these compounds are act-
ing as acyclic mimics of the lead 8, as judged by the SAR
(Table 1 and Fig. 2), computational modeling (Support-
ing Information), and receptor mutagenesis (see above).
Importantly, these compounds introduce an additional
site for modification to tune selectivity and/or potency.
The exploitation of this observation will be the focus
of future publications.
3. For an excellent leading reference, see Yang, L.; Zhou, C.;
Guo, L.; Morriello, G.; Butora, G.; Pasternak, A.;
Parsons, W. H.; Mills, S. G.; MacCoss, M.; Vicario, P.
P.; Zweerink, H.; Ayala, J. M.; Goyal, S.; Hanlon, W. A.;
Cascieri, M. A.; Springer, M. S. Bioorg. Med. Chem. Lett.
2006, 14, 3735.
4. (a) Shiota, T.; Kataoka, K.; Imai, M.; Tsutsumi, T.;
Sudoh, M.; Sogawa, R.; Morita, T.; Hada, T.; Muroga,
Y.; Takenouchi, O.; Furuya, M.; Endo, N.; Tarby, C. M.;
Moree, W.; Teig, S. WO PCT 99/25686; (b) Moree, W.;
Kataoka, K.; Ramirez-Weinhouse, M. M.; Shiota, T.;
Imai, M.; Sudo, M.; Tsutsumi, T.; Endo, N.; Muroga, Y.;
Hada, T.; Tanaka, H.; Morita, T.; Greene, J.; Barnum, D.;
Saunders, J.; Kato, Y.; Myers, P. L.; Tarby, C. M. Bioorg.
Med. Chem. Lett. 2004, 14, 5413.
5. For details on the biological assays and chemical synthe-
sis, see: Carter, P. H.; Cherney, R. J. WO/PCT
2002050019.
Acknowledgments
6. An automated peptide synthesizer was used to load the
protected diaminoacid onto resin and to perform steps j
and k.
7. Compounds 3a and 3b were synthesized by performing
steps e and f before steps c and d (Scheme 1).
8. (a) Fabiano, E.; Golding, B. T.; Sadeghi, M. M. Synthesis
1987, 190; (b) Use of ZnN6, Ph3P/DEAD, or DBU/DPPA
failed; See also (c) Pearson, C.; Rinehart, K. L.; Sugano,
M. Tetrahedron Lett. 1999, 40, 411.
We thank Dr. Douglas G. Batt for insightful comments
and acknowledge Ms. Ruowei Mo and Mr. Dayton
Meyer for their assistance in the preparation of certain
benzoic acids.
Supplementary data
9. b-Elimination of serine can be surpressed using N-trityl
protection See Cherney, R. J. J. Org. Chem. 1996, 61,
2544, In our hand, use of N-Trityl Thr led to aziridine
formation.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2007.07.028.

P. H. Carter et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5455–5461
5461
10. Modification of the carboxamide used acid 2b. Modifica-
tion of the benzyl amine followed Scheme 1. Modification
of the benzamide entailed introducing the benzylamine
first (steps e, f), followed by reprotecting with CbzCl, Boc
deprotection and amide formation (steps c and d), and
hydrogenolysis.
11. Duffy, J. L.; Kevin, N. J.; Kirk, B. A.; Chapman, K. T.;
Schleif, W. A.; Olsen, D. B.; Stahlhut, M.; Rutkowski, C.
A.; Kuo, L. C.; Jin, L.; Lin, J. H.; Emini, E. A.; Tata, J. R.
Bioorg. Med. Chem. Lett. 2002, 12, 2423.
12. Cherney, R. J.; Mo, R.; Meyer, D. T.; Nelson, D. J.; Lo,
Y. C.; Yang, G.; Scherle, P. A.; Mandlekar, S.; Wasser-
man, Z.; Jezak, H.; Solomon, K. A.; Tebben, A. J.; Carter,
P. H.; Decicco, C. P. Abstracts of Papers, 233 rd ACS
National Meeting, Chicago, IL, 2007, MEDI-241.
13. De Lucca, G. V.; Kim, U. T.; Vargo, B. J.; Duncia, J. V.;
Santella, J. B., III; Gardner, D. S.; Zheng, C.; Liauw, A.;
Wang, Z.; Emmett, G.; Wacker, D. A.; Welch, P. K.;
Covington, M.; Stowell, N. C.; Wadman, E. A.; Das, A.
M.; Davies, P.; Yeleswaram, S.; Graden, D. M.; Solomon,
K. A.; Newton, R. C.; Trainor, G. L.; Decicco, C. P.; Ko,
S. S. J. Med. Chem. 2005, 48, 2194.
14. (a) Rosenkilde, M. M.; Swartz, T. W. Curr. Top. Med.
Chem. 2006, 6, 1319; (b) Berkhout, T. A.; Blaney, F. E.;
Bridges, A. M.; Cooper, D. G.; Forbes, I. T.; Gribble, A.
D.; Groot, P. H. E.; Hardy, A.; Ife, R. J.; Kaur, R.;
Moores, K. E.; Shillito, H.; Willetts, J.; Witherington, J.
J. Med. Chem. 2003, 46, 4070.
15. Dinchuk, J. E.; Davies, P.; Zhao, Q.; Carter, P. H.;
Solomon, K. A.; Scherle, P. A. PCT WO 2005049799.