Synthesis and structure-activity relationships of 3,4-diaminocyclobut-3-ene-1,2-dione CXCR2 antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.04.082

A novel series of 3,4-diaminocyclobut-3-ene-1,2-diones was prepared and found to show potent inhibitory activity of CXCR2 binding and IL-8-mediated chemotaxis of a CXCR2-expressing cell line. Microsome stability and Caco2 studies were subsequently used to

Full text of DOI:10.1016/j.bmcl.2006.04.082

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4107–4110
Synthesis and structure–activity relationships
of 3,4-diaminocyclobut-3-ene-1,2-dione CXCR2 antagonists
J. Robert Merritt,a,* Laura L. Rokosz,a Kingsley H. Nelson, Jr.,a Bernd Kaiser,a
Wei Wang,a Tara M. Stauffer,a Lynne E. Ozgur,a Adriane Schilling,a Ge Li,a
John J. Baldwin,a Arthur G. Taveras,b Michael P. Dwyerb and Jianping Chaob
aPharmacopeia Drug Discovery, Inc., 3000 Eastpark Blvd., Cranbury, NJ 08512, USA
bSchering-Plough Research Institute, 2015 Galloping Hill Rd., Kenilworth, NJ 07033, USA
Received 6 April 2006; revised 26 April 2006; accepted 26 April 2006
Available online 11 May 2006
Abstract—A novel series of 3,4-diaminocyclobut-3-ene-1,2-diones was prepared and found to show potent inhibitory activity of
CXCR2 binding and IL-8-mediated chemotaxis of a CXCR2-expressing cell line. Microsome stability and Caco2 studies were
subsequently used to show that compounds of this chemotype are predicted to have good oral bioavailability and are thus suitable
for pharmaceutical development.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Inflammatory disease is mediated principally via the
migration of leukocytes into the affected tissues.1 This
influx is sustained by chemotactic cytokines, otherwise
known as chemokines.2 Neutrophils are a subset of leu-
kocytes implicated in the etiology of conditions such as
chronic obstructive pulmonary disease (COPD),3,4 rheu-
matoid arthritis (RA),3,5 inflammatory bowel disease,3,6
sepsis,3,7 and psoriasis.3,8 CXCL8 (formerly interleukin-
8 or IL-8) is a chemoattractant for neutrophils that is
readily detected at sites of inflammation.9 CXCR1 and
CXCR2 are the only known neutrophil receptors that
contain high-affinity binding sites for CXCL8.10 Specific
roles for each receptor remain elusive, but a correlation
between receptor activation and neutrophil-trafficking
has been firmly established through animal models of
human disease using either CXCR2 knockout mice,11
monoclonal antibodies,12 and CXCR2 antagonists.13
Such promising results make the development of
CXCL8 receptor antagonists an intriguing opportunity
for pharmaceutical development.
attention, as an analog related to this series is now in hu-
man trials for the treatment of COPD.15 We surmised
that 3,4-diaminocyclobut-3-ene-1,2-dione would be a
good bioisostere for this urea based on its previous use
as a cyanoguanidine replacement by researchers at
Wyeth.16
We prepared a series of cyclobutenediones in a fashion
similar to that exemplified for compound 19 in
Scheme 1. The corresponding cyclobutenedione ana-
log, 1, of GSK’s urea in Figure 1 was a potent inhib-
itor of CXCR2 binding with an IC50 of 0.036 lM.17
Further investigation of the left side aniline, Table 1,
revealed that the aniline NH and hydroxyl were criti-
cal since substitution or removal of these groups, as in
compounds 2 and 3, resulted in dramatic loss of potency.
Removal of an electron withdrawing group reduced
potency by 10-fold as in the des-nitro compound, 4, but
shifting the nitro group from position 4- to position 5-
had no impact on binding activity, 5. Replacement of
the nitro group at either the 4- or 5- position with other
electron withdrawing groups such as cyano was well toler-
ated, 6 and 7. Carboxylic acid was not well tolerated at
either position 3- or 4-, as shown in compounds 8 and 9.
However, installation of methyl ester at position 3- result-
ed in a more potent compound 10. Based on this result, the
methyl ester was replaced with a series of simple amides,
11–15. The dimethylamide 15 was found to have the best
potency within this series. When the dimethylamide
The series of ureas (Fig. 1) developed by researchers at
GlaxoSmithKline (GSK)14 has recently garnered much
Keywords: CXCR2; CXCR1; Cyclobutenedione; Squarate; 3,4-Diami-
nocyclobut-3-ene-1,2-dione; IL-8; CXCL8; Chemokine; Neutrophil.
*
Corresponding author. Tel.: +1 609 452 3702; e-mail: bmerritt@
pcop.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.082

4108
J. R. Merritt et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4107–4110
Table 2. Effect of right side substituents on CXCR2 binding versus
IL-8
O2N
O
O
O
N
N
H
H
OH
O
N
R
H
Figure 1. GSK urea for CXCR2.
N
OH
Compounds
R
CXCR2 IC50 (lM)
c
a,b
O
O
NH2
17
18
19
20
21
22
23
24
25
26
7.200
NO2
N
N
OH
OH OH
0.010
0.008
0.171
0.548
0.005
0.010
0.014
0.041
0.926
O
O
O
O
N
H
d
O
O
N
H
N
N
OMe
H
H
N
H
19
N
OH
N
OH
O
Scheme 1. Reagents and conditions: (a) 2 M dimethylamine in THF,
PyBroP, DIEA, DCM; (b) H2, 5%Pd on C, CH3OH; (c) 3,4-
dimethoxy-3-cyclobutene-1,2-dione, CH3OH; (d) cyclopentylamine,
DIEA, CH3OH, 60 ꢁC.
N
H
N
H
aniline was further substituted with halide, potency was
maintained as exemplified with compound 16.
N
H
With dimethylamide aniline as the optimized left side,
we prepared a series of cyclobutenediones to explore
the right-hand side SAR, Table 2. The right side NH
was found to be critical since substitution with methyl,
17, resulted in dramatic loss of potency. Since the elec-
tron-rich aniline moiety is well-known to have an unfa-
vorable metabolic profile,18 we chose to focus on alkyl
replacements. Introduction of cyclohexyl and cyclopen-
tyl replacements for aniline provided potent compounds
18 and 19. Introduction of a heteroatom as in 20 or a
fused ring system as in 21 resulted in significant loss of
N
H
N
H
O
N
H
N
H
Table 1. Effect of aniline substituents on CXCR2 binding versus IL-8
R5
O
O
R4
R3
N
N
H
R2
R1
Compounds
R1
R2
R3
R4
R5
CXCR2 IC50 (lM)
1
2
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
OH
OH
H
H
NO2
H
H
0.036
28.6
8.6
H
H
3
H
NO2
H
H
4
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
H
H
0.476
0.056
0.074
0.003
3.1
5
H
H
NO2
CN
H
6
H
H
7
H
CN
CO2H
H
8
H
H
9
CO2H
CO2Me
CONH2
H
1.0
10
11
12
13
14
15
16
H
H
0.053
0.034
0.022
0.065
0.139
0.002
0.002
H
H
CONHMe
CONHEt
H
H
H
H
CONHCH2Ph
CONMe2
CONMe2
H
H
H
H
Cl
H

J. R. Merritt et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4107–4110
4109
Table 3. In vitro chemotaxis, Caco2, and liver microsome results for selected compounds (data shown are means standard deviation of two or
more measurements)
Compounds
CXCR2 chemotaxis
IC50 (lM)
Caco2 Papp A:B (nm/s)
(predicted absorption)
Human microsome
(% remaining)
Rat microsome
(% remaining)
15
19
22
0.026 0.011
0.1448 0.0002
0.0191 0.0099
24 1.7 (low)
101 4.6 (high)
120 2.4 (high)
64
72
82
2
3
5
67
56
75
7
5
7
Steinberg, K. P.; Goodman, R. B. J. Immunol. 1999, 162,
2341.
8. Glinski, W.; Jablonska, S. J. Invest. Dermatol. 1984, 82,
386.
9. Murphy, P. M. Semin. Hematol. 1997, 34, 311.
10. Murphy, P. M. Ann. Rev. Immunol. 1994, 12, 593.
11. Cacalano, G.; Lee, J.; Kikly, K.; Ryan, A. M.; Pitts-Meek,
S.; Hultgren, B.; Wood, W. I.; Moore, M. W. Science
1994, 265, 682.
potency. Simple acyclic alkyls were found to be as
good as cyclic alkyls, as exemplified by 2-ethylpropyl,
isopropyl, and tert-butyl compounds 22, 23, and 24.
While unbranched heterobenzylics such as furanyl
compound 25 showed significant potency, further chain
elongation to phenethyl resulted in a much weaker com-
pound, 26.
Three potent compounds, 15, 19, and 22, were selected
for further evaluation in chemotaxis19, Caco220, and
rat and human liver microsome assays,21 Table 3. All
three compounds were potent inhibitors of CXCR2-
mediated chemotaxis. Although compound 15 was the
most potent (chemotaxis IC50 = 0.026 0.011 lM), its
predicted absorption based on Caco2 permeability was
low. This compound and others with right-side anilines
also had poor apparent solubility in most solvents.
Cyclobutenediones with right-side alkyls, such as 19
and 22, had much better apparent solubility and had
high predicted absorption based on Caco2 results. The
ethylpropyl compound 22 was a potent inhibitor of che-
motaxis (IC50 = 0.0191 0.0099 lM), predicted to be
well-absorbed based on Caco2 results, and predicted
to have good metabolic stability based on human and
rat microsome studies (>50% remaining after 30 min
at 37 ꢁC). These findings justify further SAR develop-
ment and in vivo studies.
12. Sekido, N.; Mukaida, N.; Harada, A.; Nakanishi, I.;
Watanabe, Y.; Matsushima, K. Nature 1993, 365, 654.
13. (a) Podolin, P. L.; Bolognese, B. J.; Foley, J. J.; Schmidt,
D. B.; Buckley, P. T.; Widdowson, K. L.; Jin, Q.; White, J.
R.; Lee, J. M.; Goodman, R. B.; Hagen, T. R.; Kajikawa,
O.; Marshall, L. A.; Hay, D. W.; Sarau, H. M. J. Immunol.
2002, 169, 6435; (b) White, J. R.; Lee, J. M.; Young, P. R.;
Hertzberg, R. P.; Jurewicz, A. J.; Chaikin, M. A.;
Widdowson, K.; Foley, J. J.; Martin, L. D.; Griswold,
D. E.; Sarau, H. M. J. Biol. Chem. 1998, 273, 10095; (c)
McColl, S. R.; Clark-Lewis, I. J. Immunol. 1999, 163,
2829; (d) Souza, D. G.; Bertini, R.; Vieira, A. T.; Cunha,
F. Q.; Poole, S.; Allegretti, M.; Colotta, F.; Teixeira, M.
M. Br. J. Pharmacol. 2004, 143, 132; (e) Bertini, R.;
Allegretti, M.; Bizzarri, C.; Moriconi, A.; Locati, M.;
Zampella, G.; Cervellera, M. N.; di Cioccio, V.; Cesta, M.
C.; Galliera, E.; Martinez, F. O.; di Bitondo, R.; Troiani,
G.; Sabbatini, V.; D’Anniballe, G.; Anacardio, R.; Cutrin,
J. C.; Cavalieri, B.; Mainiero, F.; Strippoli, R.; Villa, P.; di
Girolamo, M.; Martin, F.; Gentile, M.; Santoni, A.;
Corda, D.; Poli, G.; Mantovani, A.; Ghezzi, P.; Colotta,
F. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11791.
14. Widdowson, K. L.; Elliott, J. D.; Veber, D. F.; Nie, H.;
Rutledge, M. C.; McCleland, B. W.; Xiang, J.-N.;
Jurewicz, A. J.; Hertzberg, R. P.; Foley, J. J.; Griswold,
D. E.; Martin, L.; Lee, J. M.; White, J. R.; Sarau, H. M.
J. Med. Chem. 2004, 47, 1319.
15. (a) de Boer, W. I. Drug Discovery Today 2004, 10, 93(b)
Busch-Petersen, J.; Jin, Q.; McCleland, B. W.; Nie, H.;
Palovich, M. R.; Davis, R. S.; Fu, W.; Elliott, J. D.;
Burman, M.; Foley, J. J.; Schmidt, D. B.; Podolin, P.;
Bolognese, B. J.; Underwood, D. C.; Osborn, R. R.;
Dehaas, C. J.; Salmon, M.; Carpenter, D. C.; Killian, D.
J.; Sarau, H. M.; Widdowson, K. L. In Abstracts of
Papers, 228th National Meeting of the American Chem-
ical Society, Philadelphia, PA, United States, August 22–
26, 2004; American Chemical Society: Washington, DC,
2004; MEDI-192.
In summary, we have developed a potent series of
cyclobutenedione CXCR2 receptor antagonists. Preli-
minary pharmacokinetic studies suggest that this type
of compound is amenable to further drug develop-
ment. Additional developments and in vivo results
have been reported22 and will appear in a subsequent
publication.
References and notes
1. (a) Schall, T. J.; Bacon, K. B. Curr. Opin. Immunol. 1994,
6, 865; (b) Butcher, E. C.; Picker, L. J. Science 1996, 272,
60.
2. (a) Adams, D. H.; Lloyd, A. R. Lancet 1997, 349, 490; (b)
Saunders, J.; Tarby, C. M. Drug Discovery Today 1999, 4,
80; (c) Gura, T. Science 1996, 272, 954.
3. Luster, A. D. N. Eng. J. Med. 1998, 338, 436.
4. (a) Barnes, P. J.; Shapiro, S. D.; Pauwels, R. A. Eur.
Respir. J. 2003, 4, 672; (b) Donnelly, L. E.; Rogers, D. F.
Drugs 2003, 63, 1973.
5. Feldmann, M.; Brennan, F. M.; Maini, R. N. Annu. Rev.
Immunol. 1996, 14, 397.
16. Butera, J. A.; Antane, M. M.; Antane, S. A.; Argentieri, T.
M.; Freeden, C.; Graceffa, R. F.; Hirth, B. H.; Jenkins, D.;
Lennox, J. R.; Matelan, E.; Norton, N. W.; Quagliato, D.;
Sheldon, J. H.; Spinelli, W.; Warga, D.; Wojdan, A.;
Woods, M. J. Med. Chem. 2000, 43, 1187.
17. CXCR2 binding assay: For each 200 ll reaction, a
working mixture of 0.020 lg/ll hCXCR2-CHO over-
expressing membranes with
a
specific activity of
0.6 pmol/mg (Biosignal, Montreal [Quebec], Canada) and
2 lg/ll wheatgerm-agglutinin (WGA) coated SPA beads
(Amersham Biosciences, Piscataway, NJ) was prepared in
CXCR2 assay buffer (25 mM Hepes, pH 7.4, 2.0 mM
6. Kucharzik, T.; Walsh, S. V.; Chen, J.; Parkos, C. A.;
Nusrat, A. Am. J. Pathol. 2001, 159, 2001.
7. Cummings, C. J.; Martin, T. R.; Frevert, C. W.; Quan, J.
M.; Wong, V. A.; Mongovin, S. M.; Hagen, T. R.;

4110
J. R. Merritt et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4107–4110
CaCl2, and 1.0 mM MgCl2) (Sigma, St. Louis, MO). This
CellTiter GloTM (Promega) for 10 min and then reading
fluorescence on a Wallac Victor. Maximum chemotactic
response was determined by cells to which no compound
was added (positive control), whereas the negative control
(unstimulated) was defined by the absence of IL-8 in the
lower chamber.
mixture was incubated on ice for 5 min. A 0.40 nM stock
of ligand, [125I]IL-8 (NEN Life Sciences Products, Inc.,
Boston, MA), was prepared in the CXCR2 assay buffer.
Test compounds were first serial diluted by half-log
dilutions in DMSO (Sigma) and then diluted 8.3-fold in
CXCR2 assay buffer. The above solutions were added to a
Corning #3604 NBS (non-binding surface) 96-well assay
plate as follows: 50 ll test compound or 12% DMSO,
100 ll of membrane and SPA bead mixture (Final
[membrane] = 2.0 lg/reaction; Final [SPA bead] = 200 lg/
reaction), and 50 ll [125I]IL-8 (Final [IL-8] = 0.10 nM).
The assay plates were incubated for 2 h at room temper-
ature. Binding of [125I]IL-8 to the bead/membrane mixture
was detected using a Perkin Elmer–Wallac Microbeta 1450
liquid scintillation counter. The data were fit to a one-site
competition binding model for IC50 determination using
the program GraphPad Prism (GraphPad Software, Inc.,
San Diego, CA). All IC50s represent the average of two or
more determinations and the standard deviations were no
greater than 50% from the mean.
20. Caco2 assay: Caco2 assays were conducted using BD
BioCoatTM HTS Caco-2 Assay System provided by BD
Sciences (Bedford, MA) according to the manufacturer’s
instructions.
21. Microsome assay: microsome stability studies using
human and rat microsomes were conducted as follows:
test compounds (1 lM final concentration) were incubated
for 30 min at 37 ꢁC in a mixture containing 100 mM
potassium phosphate buffer (pH 7.4), 1 mM b-NADPH
(Sigma, St. Louis, MO), 0.4% bovine serum albumin
(Intergen, Purchase, NY), and human or rat liver micro-
somes (BD Gentest, Bedford, MA) with a final P450
content of 300 nM in a total volume of 0.5 ml. The
reaction was terminated by the addition of 0.1 ml meth-
anol containing 0.003 mg/ml ketanserin (Sigma) as inter-
nal standard (IS) and followed by addition of 0.75 ml ethyl
acetate for liquid extraction. After a 10-min vortex and 10-
min centrifugation at 20,000g, 0.5 ml of the ethyl acetate
phase was removed and evaporated under a nitrogen
stream. The residue was reconstituted in 0.1 ml methanol/
H2O (1:1) solution and subjected to HPLC-MS analysis.
The percentage of compound remaining was calculated by
taking the percentage of the average peak area of 30-min
samples from the average peak area of the corresponding
0-min samples.
22. Dwyer, M. P.; Yu, Y.; Chao, J.; Aki, C.; Purakkattle,
B.; Rindgen, D.; Bond, R.; Jakway, J.; Hipkin, R. W.;
Fosetta, J.; Gonsiorek, W.; Bian, H.; Fine, J.; Merritt, J.
R.; Rokosz, L. L.; Kaiser, B.; Li, G.; Wang, W.;
Stauffer, T.; Ozgur, L.; Taveras, A. In Abstracts of
Papers, 231st National Meeting of the American Chem-
ical Society, Atlanta, GA, United States, March 26–30,
2006; American Chemical Society: Washington, DC,
2006; MEDI-19.
18. (a) Mutschler, E.; Derendorf, H. Drug Actions: Basic
Principles and Therapeutic Aspects; Medpharm Scientific
Publishers: Stuttgart, Germany, 1995, p. 20; (b) Silverman,
R. B. The Organic Chemistry of Drug Design and Drug
Action; Academic Press: San Diego, 1992, pp. 295–296.
19. CXCR2 chemotaxis assay: test compounds were dissolved
in 100% DMSO and diluted by half-logs. The compounds
were subsequently diluted 100· in assay medium to bring
the DMSO concentration to 1.0%. 6.6 ll of test compound
was then mixed with 60 ll hCXCR2-CHO cells. The lower
chamber of a ChemoTx plate (Neuroprobe, Gaithersburg,
ND) was filled with 30 ll of 18 nM IL-8 (R&D Systems,
Minneapolis, MN). The empty upper chambers were
affixed to the lower plate and 25 ll of 1.5E6 hCXCR2-
CHO cells, pre-incubated for 100 min with compound,
was added to the upper well. Migration proceeded for
40 min at 37 ꢁC in a humidified incubator with 5% CO2.
After removing non-migrated cells from the top of the
plate, migrated cells were quantified by adding 25 ll