Potent small molecule CCR1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2004.02.021

The present manuscript details structure-activity relationship studies of lead structure 1, which led to the discovery of CCR1 antagonists >100-fold more potent than 1.

Full text of DOI:10.1016/j.bmcl.2004.02.021

Bioorganic & Medicinal Chemistry Letters 14 (2004) 2169–2173
Potent small molecule CCR1 antagonists
John C. Kath,* William H. Brissette, Matthew F. Brown, Maryrose Conklyn,
Amy P. DiRico, Peter Dorff, RonaldP. Gladue, Brett M. Lillie, Paul D. Lira,
Erin N. Mairs, William H. Martin, Eric B. McElroy, Molly A. McGlynn,
Timothy J. Paradis, Christopher S. Poss, Ingrid A. Stock, Laurie A. Tylaska and
Deye Zheng
Pfizer Global Research and Development, Eastern Point Road, Groton, CT 06340, USA
Received8 January 2004; revised4 February 2004; accepted5 February 2004
Abstract—The present manuscript details structure–activity relationship studies of lead structure 1, which ledto the discovery of
CCR1 antagonists >100-foldmore potent than 1.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The preceding manuscript in this series highlighted
fundamental SAR work in our CCR1 (CC-chemokine
receptor-1) antagonist program aimedat finding potent
small molecule drugs for the treatment of autoimmune
diseases and organ transplant rejection.1 This effort
began with the identification of the structurally unique
screening hit, compound 1 (Fig. 1). This compoundwas
shown to be a micromolar inhibitor of both CCL3
binding and CCL3 induced chemotaxis of THP-1
cells.2;3 Initial medicinal chemistry efforts demonstrated
that the absolute stereochemistry present in compound 1
(2S,4S,5R) was required for optimal potency. Addi-
tional work provided compound 2, the 5-benzyl variant
of 1 whose increasedactivity in both CCL3 binding and
chemotaxis assays validated it as a suitable lead for
further optimization.
Aiming to improve upon the potency of 2, we set out to
investigate SAR involving the N-terminal amide func-
tionality. In doing so, our primary aim was to system-
atically investigate the heteroaromatic moiety. A series
of amide derivatives (Table 1) was prepared in two steps
from lactone 3.4;5 The loss in activity observedwith the
naphthalene-2-carboxamide analog 4 highlightedthe
importance of the quinoline nitrogen present in 2. Fur-
thermore, it was also discovered that the positioning of
Figure 1.
Keywords: CCR1; Chemokine.
* Corresponding author. Tel.: +1-860-441-3852; fax: +1-860-441-4111; e-mail: john_c_kath@groton.pfizer.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.02.021

2170
J. C. Kath et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2169–2173
Table 1.
CompoundR
inCg CICL3 bind
(lM)
CCL3 chemotaxis IC50 (lM)
50
2
0.65
0.46
4
5
10.3
>25
0.89
0.13
6
7
18.5
4.3
2.49
3.30
8
>32
>25
9
16.7
>25
>25
10
>32
11
12
0.51
NA
5.15
6.60
13
14
0.58
2.40
0.064
0.064
15
16
2.57
1.10
>32
>25

J. C. Kath et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2169–2173
2171
Table 2.
the carboxamide functionality on the quinoline ring
playedan important role in determining the potency of
analogs. The quinoline-2-carboxamide 5 was notable in
that it provided improved inhibition of CCL3 binding,
but in contrast, the corresponding quinoline-4-carbox-
amide, -6-carboxamide, and -8-carboxamide derivatives
(6–8, respectively) showeddecreasedactivity both in
CCL3 binding and chemotaxis assays. In addition, the
2-hydroxyquinoline-3-carboxamide analog 9 showeda
considerable loss in activity relative to the parent quin-
oline-3-carboxamide. Removing the fused phenyl ring
from the quinoline also proveddetrimental, as the nic-
otinamide 10 was completely inactive. However, the
corresponding 5-bromo-nicotinamide 11 provided
activity comparable to quinoline 2, possibly due to an
increase in lipophilicity and/or a decrease in the Lewis
basicity of the pyridine nitrogen. Converting the quin-
oline-carboxamide to an isoquinoline-carboxamide
provided mixed results with the isoquinoline-4-carbox-
amide 12 losing a log in activity andthe isoquinoline-1-
carboxamide 13 nearly equipotent to 2 in the CCL3
binding assay. Building on the finding that optimal
activity was achievedwhen the amide functionality was
one or two carbon atoms removedfrom the heteroaro-
matic nitrogen andthe implication that lipophilicity
and/or decreased Lewis basicity led to increased
potency, we preparedthe quinoxaline-2-carboxamied
14. This arrangement of the heteroaromatic nitrogens
relative to the carboxamide functionality and each other
provedquite favorable andprovideda log increase in
potency relative to 2. In comparison, the relatedcinn-
oline-3-carboxamide derivative 15 was slightly less
active than the parent quinoline-carboxamide. In addi-
tion to a favorable relative placement of the heteroaro-
matic nitrogens, lipophilicity also playedan important
role in the activity of 14. This was evident in that the
pyrazine-2-carboxamide derivative 16 was completely
inactive.
CompoundR
CCL3 bingd
IC50 (lM)
CCL3 chemo-
taxis IC50 (lM)
14
0.064
0.064
17
18
0.67
4.2
0.56
NA
19
20
21
2.6
3.0
1.3
2.0
0.61
2.9
22
0.16
0.15
boxamide did not alter the C-5 SAR trend reported
previously1 as the benzyl substituent in 14 provided the
best potency, followedby the cyclohexylmethyl ( 17) and
2-methyl-propyl (18) groups (Table 2). These results
focusedour attention squarely on the 5-benzyl deriva-
tive, which servedas the starting point for an investi-
gation into the effect of substitution of the phenyl ring.
Having discovered a more optimal N-terminal amide,
we turnedour attention to modifications of the C-5
benzyl group. The presence of the quinoxaline-2-car-
Scheme 1. Reagents andconditions: (a) LHMDS or LDA, THF, )78 ꢁC; (b) ROTf or RBr; (c) 10% Pd/C, 35 psi H2, THF/MeOH; (d) ketone,
Burgess reagent, benzene, reflux; (e) TFA, CH2Cl2; (f) quinoxalyl chloride, TEA, CH2Cl2; (g) amine, MeOH, rt; (h) O3, MeOH, SMe2; (i) NaBH4,
MeOH; (j) NaBH(OAc)3, HN(CH3)2, dichloroethane.

2172
J. C. Kath et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2169–2173
Table 3.
CompoundR1
R2
ing ICCCL3 bind
NHMe
(lM)
CCL3 chemotaxis IC50 (lM)
50
14
25
26
27
0.065
0.022
0.028
0.42
0.064
NHMe
NH2
0.008
0.002
0.069
NH2
28
NH2
0.048
0.012
29
30
NH2
NH2
0.026
0.008
0.028
0.005
31
NH2
44.8
4.6
32
33
34
NH2
NH2
NH2
0.28
4.6
0.12
>25
55.1
>25
35
36
NH2
NH2
3.5
12.5
0.015
0.005
To this end, 4 differentially substituted benzyl deriva-
tives (19–22) were preparedbeginning with the appro-
priately substitutedBoc-amino acid. With the exception
of the 3,4-dichlorobenzyl derivative 22, the remaining
three analogs were P10–foldless active than their
unsubstitutedcounterpart.
synthesis of C-2 alkyl derivatives was much more
straightforward, beginning with the common lactone
intermediate 234;5 (see Scheme 1). The differentiating step
was the addition of an alkylating agent (allyl/benzyl
bromide, alkyl triflate, or ketone) to the lithium enolate
of 23. The reaction proceeded with a high degree of
stereoselectivity with allyl/benzyl bromides or alkyl tri-
flates to provide 24. Condensation reactions of the lith-
ium enolate of 23 with cyclic ketones were less
stereoselective producing separable mixtures of diaste-
reomers, which, following a dehydration/hydrogenation
Concurrent with work aroundthe C-5 group was an
effort investigating alternative functionality at the C-2
position. Rather than the multi-step process involvedto
prepare analogs with alternative C-5 substituents, the

J. C. Kath et al. / Bioorg. Med. Chem. Lett. 14 (2004) 2169–2173
2173
sequence (Burgess reagent; Pd/C, 35 psi H2), provided 24.
Subsequent Boc deprotection, N-acylation with qui-
noxaline-2-carbonyl chloride and lactone opening with
the appropriate amine (MeNH2 or NH3) provided the
final compounds. For the preparation of 33 and 34, the
allyl intermediate 24a was convertedto the correspond-
ing aldehyde via ozonolysis. Reduction with NaBH4
provided the alcohol functionality of 33; reductive ami-
nation provided the amine functionality of 34.
leads to upregulation of the integrin CD11b on mono-
cytes. Expression of this integrin plays an important role
in monocyte adhesion and extravasation thereby leading
to the delivery of these cells to inflammatory sites. Lead
compound 1 demonstrated that CCR1 antagonism
effectively inhibits CCL3 induced CD11b upregulation
on monocytes in human whole blood(IC 50 ꢀ 5 lM).
Compound 26 was also evaluatedin this assay and
foundto be 10–20 ·more potent than leadcompound 1.6
A variety of lipophilic substituents were well toleratedat
C-2. Homologation of the isobutyl C-2 substituent of 14
ledto a modest improvement in potency ( 25) (Table 3).
In an attempt to reduce lipophilicity, the primary amide
26 was preparedandfoundto be equipotent to the
methyl amide 25. Homologation of 26 ledto diminished
potency (27) suggesting that the C-2 substituent binds in
a lipophilic pocket of defined dimensions. A variety of
cycloalkyl substituents were well toleratedat C-2 ( 28–
30). However, as alluded to above, large groups were
not well tolerated(compare 30 and 31). Benzyl substi-
tution at C-2 provided moderate potency (32), however,
incorporation of polar functionality at C-2 imparted
greatly reduced potency (33–35). In some cases, masking
polar functionality ledto improvedpotency (compare 35
and 36).
In summary, optimization of the N-terminal heterocycle
andC-2 alkyl group of leadcompound 1 ledto a
number of CCR1 antagonists with enhancedpotency.
References and notes
1. Kath, J. C.; DiRico, A. P.; Gladue, R. P.; Martin, W. H.;
McElroy, E. B.; Stock, I. A.; Tylaska, L. A.; Zheng, D.
Bioorg. Med. Chem. Lett. 2004, preceding manuscript. doi:
10.1016/j.bmcl.2004.02.020.
2. Neote, K.; DeGregorio, D.; Mak, J. Y.; Horuk, R.; Schall,
T. J. Cell 1993, 72, 415.
3. Cole, K.; Strick, C.; Paradis, T.; Ogborne, K. T.; Loetscher,
M.; Gladue, R. P.; Lin, W.; Boyd, J. G.; Moser, B.; Wood,
D. E.; Sahagan, B. G.; Neote, K. J. Exp. Med. 1998, 187,
2009.
4. Fray, A. H.; Kaye, R. L.; Kleinman, E. F. J. Org. Chem.
1986, 51, 4828.
Potent analogs were evaluatedin a number of secondary
assays. For example, in addition to inhibiting CCL3
induced chemotaxis of THP-1 cells, compound 26 also
effectively blockedchemotaxis inudcedby CCL5
(RANTES), the other primary ligandfor CCR1. This
represents a >100-foldincrease in activity relative to the
leadcompound 1. Activation of the CCR1 receptor
5. DeCamp, A. E.; Kawaguchi, A. T.; Volante, R. P.; Shinkai,
I. Tetrahedron Lett. 1991, 32, 1867.
6. Conklyn, M. J.; Neote, K.; Showell, H. J. Cytokine 1996, 8,
762.