Nicotinamide N-oxides as CXCR2 antagonists

Article abstract of DOI:10.1016/S0960-894X(01)00326-2

A series of nicotinamide N-oxides was synthesized and shown to be novel, potent, and selective antagonists of the CXCR2 receptor. Furthermore, these compounds showed significant functional activity against GRO-α-driven human neutrophil chemotaxis. Compounds of this class may be useful for the treatment of inflammatory, auto-immune, and allergic disorders.

Full text of DOI:10.1016/S0960-894X(01)00326-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1951–1954
Nicotinamide N-Oxides as CXCR2 Antagonists
Neil S. Cutshall,a,* Rocky Ursino,b,y Kristin A. Kucera,b,y John Lathamb and
Nathan C. Ihlea
aDepartment of Chemistry, Celltech R&D, Inc., 1631 220th Street SE, Bothell, WA98021, USA
bDepartment of Gene Function & Target Validation, Celltech R&D, Inc., 1631 220th Street SE, Bothell, WA98021, USA
Received 19 March 2001; accepted 9 May 2001
Abstract—A series of nicotinamide N-oxides was synthesized and shown to be novel, potent, and selective antagonists of the
CXCR2 receptor. Furthermore, these compounds showed significant functional activity against GRO-a-driven human neutrophil
chemotaxis. Compounds of this class may be useful for the treatment of inflammatory, auto-immune, and allergic disorders. # 2001
Elsevier Science Ltd. All rights reserved.
Introduction
neutrophil chemotaxis (IC50=10 nM) and was effica-
cious in vivo in an LPS-induced rabbit model of neu-
trophilia.2 Compound 2 was recently claimed as a
selective antagonist of IL-8 binding (IC50=110 nM) and
potent inhibitor of neutrophil chemotaxis (IC50=170
nM).3
Chemokines selectively recruit and activate a variety of
cells during inflammation. Interleukin-8 (IL-8) and
other Glu-Leu-Arg (ELR)-containing CXC chemo-
kines, including epithelial cell-derived neutrophil-acti-
vating peptide-78 (ENA-78), and growth-related
oncogene (GRO)-a, GRO-b, and GRO-g, attract and
activate neutrophils. In response to traumatic injury or
infection, human neutrophils are directed to the site of
injury or infection by CXC chemokines that signal via
two distinct seven-transmembrane G-protein coupled
receptors (GPCRs), CXCR1and CXCR2. Both
CXCR1and CXCR2 bind IL-8 with high affinity; how-
ever, only CXCR2 binds the other neutrophil-activating
and ELR motif-bearing chemokines with high affinity.
Although the exact role of IL-8 and GRO-a in human
disease is yet to be determined, these and other pro-
inflammatory chemokines are observed in association
with chronic and acute inflammatory conditions and
correlate with neutrophil infiltration in a wide range of
diseases including psoriasis, rheumatoid arthritis, and
adult respiratory distress syndrome.1
In this paper, we report the preliminary results of a
project to discover selective antagonists of the CXCR2
receptor. An effort to identify inhibitors of chemokine-
driven neutrophil responses found compound 4a, a
nicotinamide N-oxide with promising activity. A series
of studies were then conducted to define the SAR of
these compounds as potential new antiinflammatory
therapeutics.
Chemistry
Small-molecule CXCR2 antagonists are potentially
valuable novel therapeutics, although relatively few
have been reported in the literature. Most noteworthy is
compound 1 that has demonstrated potent inhibition of
6-Chloronicotinamide N-oxides were prepared accord-
ing to the route shown in Scheme 1. Commercially
available 6-chloronicotinic acid was converted to the
corresponding N-oxide 3 using trifluoroperacetic acid.4
Acid 3 was subsequently reacted with a variety of
amines or anilines using EEDQ as the coupling reagent
of choice to provide compounds 4a–h.
*Corresponding author. Tel.: +1-425-489-8065; fax: +1-425-489-
8019; e-mail: neil.cutshall@celltechgroup.com
yThese authors contributed equally to this work.
0960-894X/01/$ - see front matter # 2001Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00326-2

1952
N. S. Cutshall et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1951–1954
Scheme 2 outlines two synthetic routes toward 6-sulfo-
nyl-1-oxy-nicotinamides. 6-Chloronicotinoyl chloride
was reacted with 4-fluoroaniline to provide nicotina-
mide 5 in high yield. Route A was used to readily oxi-
dize the pyridine core using trifluoroperacetic acid4 at
elevated temperatures to provide 4a. Efforts to effect the
oxidation with other oxidizing reagents such as m-
CPBA,5 m-CPBA in the presence of HF,6 magnesium
monoperoxyphthalate hexahydrate,7 or urea hydrogen
peroxide complex and phthalic anhydride in methanol
failed to provide appreciable quantities of desired pro-
duct. Displacement of the 6-chloro substituent with
sodium methanesulfinate provided analogue 7a.
Because of the limited number of available sulfinates, a
second synthetic strategy was developed (route B).
Compound 5 could be reacted with a variety of alkyl or
aryl potassium thiolates to provide the corresponding 6-
thionicotinamides 6. Aqueous trifluoroperacetic acid
was then used as the preferred oxidant to provide com-
pounds 7b–g. This single-step, novel oxidation of both
the pyridine core to the corresponding N-oxide and the
thioether to the corresponding sulfone was readily
monitored by HPLC–MS. The oxidation reactions were
typically complete within 12 h at which time the pro-
ducts precipitated from solution. The 6-sulfonyl-1-oxy-
nicotinamides were then readily purified via trituration
to afford the products in yields >50%.
Ligand binding was evaluated using an SPA assay with
recombinant human [125I]-IL-8 and membranes pre-
pared from Sf9 cells expressing human CXCR2 and
Gai3b1g2 proteins.9 [125I]-IL-8 was chosen as the ligand
in the binding assay because it gave more reproducible
results than did [125I]-GRO-a.
SAR studies initially focused on amide modifications in
the 3-position of the pyridine (Table 1). It was found
that a number of substituted anilides blocked GRO-a-
driven neutrophil chemotaxis with activities in the low
micromolar range. Efforts to replace the anilide func-
tionality with other amides, however, led to compounds
with significantly decreased functional activity as illu-
strated by compounds 4g and 4h. In contrast to the
SAR flexibility observed for a variety of anilides in the
chemotaxis assay, no replacement of the 4-fluoroanilide
was found that gave acceptable activity in the IL-8
ligand binding assay. Based on our SAR survey, the 4-
fluoroanilide moiety of compound 4a was shown to be a
critical recognition element necessary for blocking IL-8/
CXCR2 interaction. The disparity between ligand bind-
ing affinity and functional activity is not without pre-
cedent in the chemokine field. It was shown by Jarnagin
et al. that mutants of the CC chemokine, monocyte
chemotactic protein, could function as high-affinity
binders to CCR2, yet were functionally inactive to sti-
mulate chemotaxis.10 It was also recently demonstrated
that high-affinity binding of IL-8 and GRO-a to
CXCR2 involves interaction with specific and different
amino acid residues of CXCR2. It was proposed that
the CXCR2 amino acid residues required for cell acti-
vation are not necessarily the same residues required for
ligand binding.11 Related work by Ahuja et al. demon-
strated that there are divergent regions of ligand contact
on CXCR2 for binding as well as cell activation and
concluded that distinct antagonists could be found to
independently block either high affinity binding or
receptor activation or both.12 Based on the possibility of
multiple binding domains for a small molecule with
CXCR2, there are several possible explanations for the
disparity between the functional and ligand binding
activities for the compounds in this series. Compounds
4b–e may interact at specific regions of CXCR2 that
block GRO-a but not IL-8 binding. Alternatively,
Results and Discussion
As part of a lead optimization strategy, nicotinamide N-
oxides were screened for both their ability to antagonize
the binding of IL-8 to CXCR2 and for their ability to
block GRO-a-driven human neutrophil chemotaxis.8
Scheme 1. Reactions and conditions: (a) TFA, 30% H2O2 in H2O, 0–
45 ꢀC, 64%; (b) R1NH2, EEDQ, CHCl3, rt.
Scheme 2. Reactions and conditions: (a) 4-FC6H4NH2, i-Pr2NEt, CH2Cl2, 0 ꢀC to rt, 92%; (b) 30% H2O2 in TFA, 0–45 ꢀC, 25%; (c) NaSO2Me,
DMF, 56%; (d) R2SH, t-BuOK, THF, 60–80%; (e) 30% H2O2 in TFA, 0–45 ꢀC, 50–70%.

N. S. Cutshall et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1951–1954
1953
compounds 4b–e may inhibit neutrophil chemotaxis
through interactions with amino acid residues on
CXCR2 that are required for receptor activation but
not required for agonist binding.
interact at sites on CXCR2 that specifically disrupt IL-8
binding or bind at a subsite of CXCR2 not required for
GRO-a-stimulated receptor activation.
As part of an ongoing assessment of these compounds,
the selectivity of representative 6-chloro- and 6-sulfo-
nylnicotinamide N-oxides was investigated. Ligand
binding antagonist selectivity was assessed using a panel
of phylogenetically diverse GPCRs (Table 3). In all
cases, the compounds were most effective in the CXCR2
assay. In general, little or no activity was seen at the
non-chemokine receptors. Compounds 4a and 7a did
inhibit binding of IL-8 to CXCR1, but to a lesser extent
than at CXCR2. The observation that 7b, 7d, and 7g
were highly selective for CXCR2 versus CXCR1sug-
gests that the compounds were in fact interacting with
CXCR2 rather than binding to the chemokine IL-8.
Functional selectivity was assessed in a neutrophil che-
motaxis assay using N-formyl-methionyl-leucyl-phenyl-
alanine (FMLP). This chemoattractant polypeptide
activates neutrophils via specific N-formyl peptide
receptors. In all cases, the compounds were more potent
as GRO-a antagonists than as FMLP antagonists.
Again, compound 4a was unique in that it displayed the
highest overall level of ligand binding and functional
selectivity.
As a complement to SAR exploration at the 3-position
of the nicotinamide N-oxides, it was of interest to look
at SARs around the 6-substituent of the N-oxide core. It
was found that 6-sulfonylnicotinamide N-oxides were
particularly interesting (Table 2). In general, a diverse
set of 6-alkyl and -alkylaryl substituted nicotinamide N-
oxide sulfones provided potent analogues, especially in
the ligand binding assay. As was observed with the 6-
chloronicotinamide N-oxides however, there were
examples of compounds that had disparate activity
between their binding and functional activities. In par-
ticular, the aryl sulfones 7e and 7f represented the most
potent ligand binding antagonists in the series with
IC50’s of 90 and 32 nM, respectively, yet they were
functionally inactive. Again, these compounds may
Table 1. Representative activities of 6-chloronicotinamide N-oxides
Table 2. Representative activities of 6-sulfonylnicotinamide N-oxides
Compound
R1
GRO-a chemotaxisa IL-8 SPA bindinga
(IC50, mM)
(IC50, mM)
4a
4b
1.1
1.0
0.5
>20
>20
Compound
R2
GRO-a chemotaxisa IL-8 SPA bindinga
(IC50, mM)
(IC50, mM)
7a
7b
2.0
0.13
4c
0.83
3.3
2.0
0.13
0.40
0.46
0.090
4d
4e
1.3
4.2
>20
>20
7c
7d
7e
1.9
>20
4f
>20
>20
>20
>20
>20
>20
7f
>20
0.032
0.28
4g
4h
7g
1.0
aValues were calculated from the geometric mean of at least two
experiments. An IC50 value of >20 indicates less than 50% activity at
20 mM.
aValues were calculated from the geometric mean of at least two
experiments.

1954
N. S. Cutshall et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1951–1954
Table 3. Percent inhibition of bindinga,b at 20 mM and FMLP neutrophil chemotaxis inhibitiona of representative nicotinamide N-oxides
Compound
CXCR1c
CXCR2d
Calcitonine
Cholecystokinin
f
Endothelin
g
Neuropeptide
h
Somatostatini
FMLP
(chemotaxisj IC50, mM)
CCKB
ETB
Y1
4a
7a
7b
7d
7g
68
73
8
8
9
85
97
97
60
77
ꢁ1
ꢁ20
3
4
ꢁ23
11
40
1
1
29
14
8
15
58
-2
60
>40
9.6
13.3
4
1
5
ꢁ1541
k
ꢁ2
ꢁ1
ꢁ1
9
NT
8
32
38
13
aValues were calculated from the geometric mean of at least two experiments.
bBinding assays were conducted at Panlabs (Bothell, WA).
cDisplacement of [125I]-IL-8 from CHO cell membranes expressing human CXCR1receptors.
dDisplacement of [125I]-IL-8 from CHO cell membranes expressing human CXCR2 receptors.
eDisplacement of [125I]-calcitonin (salmon) from T-47D (human breast cancer) cell membranes expressing human calcitonin receptors.
fDisplacement of [3H]-CCK-8 from NIH-3T3 (mouse embryo cell membranes) expressing human cholecystokinin CCKB receptors.
gDisplacement of [125I]-endothelin-1from CHO cell membranes expressing human endothelin ET receptor.
B
hDisplacement of [125I]-peptide YY (PYY) from SK-N-MC (human neuroblastoma) cell membranes expressing human neuropeptide Y1 (NPY1)
receptors.
iDisplacement of [125I]-tyr1 somatostatin from AtT-20 (mouse pituitary) cell membranes expressing somatostatin receptors.
jSee ref 8.
kNot tested.
Abstracts of Papers, 220th National Meeting of the American
Since compound 4a was unique in that it showed
Chemical Society, Washington, DC, August 20–24, 2000;
appreciable selectivity over other GPCRs as well as
American Chemical Society: Washington, DC, 2000; MEDI
activity as both a ligand binding and functional
144.
antagonist, additional studies were undertaken. It was
shown that 4a was able to moderate IL-8 driven human
4. Liotta, R.; Hoff, W. S. J. Org. Chem. 1980, 45, 2887.
5. Takemoto, T.; Eda, M.; Okada, T.; Sakashita, H.; Matzno,
neutrophil chemotaxis (IC50=1.3–2.3 mM), was well
tolerated in mice at 100 mg/kg po, and was relatively
stable to rat liver microsomes (data not shown).
S.; Gohda, M.; Ebisu, H.; Nakamura, N.; Fukaya, C.; Hihara,
M.; Eiraku, M.; Yamanouchi, K.; Yokoyama, K. J. Med.
Chem. 1994, 37, 1 8.
6. Rhie, S. Y.; Rhy, E. K. Heterocycles 1995, 41, 323.
7. Brougham, P.; Cooper, M. S.; Cummerson, D. A.; Heaney,
H.; Thompson, N. Synthesis 1987, 11, 1015.
8. Neutrophil isolation and GRO-a or FMLP chemotaxis was
conducted according to Frevert, C. W.; Wong, V. A.; Good-
man, R. B.; Goodwin, R.; Martin, T. R. J. Immunol. Meth.
1998, 213, 41 .
In conclusion, the synthesis and evaluation of a novel
series of potent and selective CXCR2 antagonists has
been described. The molecules have shown potency in
an IL-8 ligand binding assay and show the ability to
block neutrophil chemotaxis using GRO-a as the ago-
nist. Compound 4a was selected for more advanced
investigation based on its in vitro functional, ligand
binding, and pharmacological profile. The nicotinamide
N-oxides continue to provide valuable tools in the
investigation of the role of CXCR2 on neutrophils.
9. [125I] IL-8 (human recombinant) was obtained from NEN
Life Science Products, Inc., Boston, MA, USA with specific
activity of 2200 Ci/mmol. Recombinant human IL-8 was
obtained from R&D Systems, Minneapolis, MN. CXCR2
receptor membranes were prepared from Sf9 cells co-expressed
with Gai3b1g2 proteins by BioSignalTM, Montreal, Canada.
Wheat Germ Agglutinin Scintillation Proximity Assay Beads
were obtained from Amersham Pharmacia Biotech, Piscat-
away, NJ. All assays were performed in a 96-well, solvent
resistant, white PicoPlate obtained from Packard Instruments.
Each reaction mixture contained [125I] IL-8 (0.16 nM), 5 mg/
well CXCR2 membranes and 1mg/well WGA-SPA beads in
25 mM HEPES (pH 7.4), containing 2 mM CaCl2, 1 mM
MgCl2, 0.1% BSA and 0.03% CHAPS. In addition, the drug
or compound of interest was added which had been pre-dis-
solved in DMSO so as to reach a final concentration of
between 0.01nM and 80 mM, with a maximum final DMSO
concentration of 1%. Nonspecific binding was defined by the
presence of 8–16 nM unlabelled IL-8. The assay was initiated
by the addition of WGA-SPA beads in 25 mM HEPES (pH
7.4) containing 2 mM CaCl2 and 1mM MgCl 2. After 4–6 h of
gentle agitation at room temperature, the plate was counted
on the Packard TopCount liquid scintillation counter.
10. Jarnagin, K.; Grunberger, D.; Mulkins, M.; Wong, B.;
Hemmerich, S.; Paavola, C.; Bloom, A.; Bhakta, S.; Diehl, F.;
Freedman, R.; McCarley, D.; Polsky, I.; Ping-Tsou, A.;
Kosaka, A.; Handel, T. M. Biochemistry 1999, 38, 16167.
11. Katancik, J. A.; Sharma, A.; De Nardin, E. Cytokine
2000, 12, 1480.
Acknowledgements
The authors would like to thank Mr. Ted Baughman
and Dr. J. Jeffry Howbert for helpful discussions.
References and Notes
1. (a) Harada, A.; Mukaida, N.; Matsushima, K. Mol. Med.
Today 1996, 2, 482. (b) Murdoch, C.; Finn, A. Blood 2000, 95,
3032. (c) Hauser, C. J.; Fekete, Z.; Goodman, E. R.; Klein-
stein, E.; Livingston, D. H.; Deitch, E. A. Shock 1999, 12, 428.
2. Palovich, M. R.; Widdowson, K.; Elliott, J. D.; White, J.
R.; Sarau, H. M.; Rutledge, M. C.; Bi, J.; Dede, K. A.; Ben-
son, G. M.; Griswold, D. E.; Martin, L. D.; Nie, H. Schmidt,
D. M.; Foley, J. J. Abstracts of Papers, Part 2, 219th National
Meeting of the American Chemical Society, San Francisco,
CA, March 26–30, 2000; American Chemical Society:
Washington, DC, 2000; MEDI 338.
3. Li, J. J.; Yue, W. S.; Trivedi, B. K.; Miller, S. R.; Connor,
D. T.; Roth, B. D.; Low, J. E.; Heilig, D. J.; Hunt, S.; Carson,
K. G.; Glynn, R. A.; Ye, Q.; Luly, J. R.; Yang, W.; Qin, S.
12. Ahuja, S. K.; Lee, J. C.; Murphy, P. M. J. Biol. Chem.
1996, 271, 225.