The discovery of a selective, high affinity A2B adenosine receptor antagonist for the potential treatment of asthma

Article abstract of DOI:10.1016/j.bmcl.2004.11.044

Adenosine has been suggested to play a role in asthma, possibly via activation of A_2B adenosine receptors on mast cells and other pulmonary cells. We describe our initial efforts to discover a xanthine based selective A_2B AdoR antagonist that resulted in the discovery of CVT-5440, a high affinity A_2B AdoR antagonist with good selectivity (A_2B AdoR K_i = 50 nM, selectivity A₁ > 200: A_2A > 200: A₃ > 167).

Full text of DOI:10.1016/j.bmcl.2004.11.044

Bioorganic & Medicinal Chemistry Letters 15 (2005) 609–612
The discovery of a selective, high affinity A2B adenosine
receptor antagonist for the potential treatment of asthma
Jeff Zablocki,a,* Rao Kalla,a Thao Perry,a Venkata Palle,a Vaibhav Varkhedkar,a
Dengming Xiao,d Anthony Piscopio,d Tenning Maa,b Art Gimbel,b Jia Hao,c Nancy Chu,c
Kwan Leungc and Dewan Zengb
aDepartment of Bioorganic Chemistry, CV Therapeutics, Inc., 3172 Porter Dr., Palo Alto, CA 94304, USA
bDepartment of Drug Research and Pharmacological Sciences, CV Therapeutics, Inc., 3172 Porter Dr., Palo Alto, CA 94304, USA
cDepartment of Preclinical Development, CV Therapeutics, Inc., 3172 Porter Dr., Palo Alto, CA 94304, USA
dArray Biopharma, Inc., 3200 Walnut Street, Boulder, CO 80301
Received 27 September 2004; revised 14 November 2004; accepted 17 November 2004
Available online 9 December 2004
Abstract—Adenosine has been suggested to play a role in asthma, possibly via activation of A2B adenosine receptors on mast cells
and other pulmonary cells. We describe our initial efforts to discover a xanthine based selective A2B AdoR antagonist that resulted in
the discovery of CVT-5440, a high affinity A2B AdoR antagonist with good selectivity (A2B AdoR Ki = 50 nM, selectivity A1 > 200:
A
2A > 200: A3 > 167).
ꢀ 2004 Elsevier Ltd. All rights reserved.
Adenosine is an endogenous agonist that can activate all
four adenosine receptor subtypes—A1, A2A, A2B, and
A3.1 Activation of the A2B adenosine receptors (AdoR)
on mast cells may play a putative role in asthma by
enhancing mast cell degranulation and releasing inflam-
matory cytokines (e.g. interleukin-4, 8, and 13).2 Fur-
thermore, activation of A2B AdoR on bronchial
smooth muscle cells (BSMC) has been demonstrated
to lead to the release of interleukin-6 and monocytic
chemotactic peptide-1 (MCP-1).3
chronic oral treatment.4–6 Although the mechanism of
action of theophylline is not completely understood, it
is known to be a nonselective inhibitor for phosphodies-
terases (PDE) and a nonselective AdoR antagonist (A2B
AdoR Ki = 7059 nm, selectivity A1 0.6: A2A 0.6: A3 14,
Fig. 1).
We hypothesize that the low therapeutic index of theo-
phylline, due to both CNS and cardiac side effects,
may be the result of its poor selectivity (for both PDEs
and AdoRs). Based on our understanding of the role of
A2B AdoRs in asthma,2,3 we hypothesize that a potent
and selective A2B AdoR would provide a better
Theophylline 1, is used for the treatment of asthma in
both IV rescue therapy for acute asthma attacks and
O
O
H
H
N
CN
O
2 R =
MRS-1754
CVT-5440
N
N
N
N
H
N
H
O
O
R
N
O
N
O
N
O
N
3 R =
N
1 Theophylline
Figure 1. Structure of theophylline, MRS-1754 (2), and CVT-5440 (3).
Keywords: A2B; Adenosine; Antagonist; Asthma.
*
Corresponding author. Tel.: +1 650 384 8547; fax: +1 650 858 0390; e-mail: jeff.zablocki@cvt.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.11.044

Table 1. Binding affinities of 3-phenyl (10–25) and 5-phenyl-1,2,4-oxadiazole analogues (26–39) for the A2B, A1, and A2A adenosine receptors
O
H
N
O
N
H
N
N
O
O
N
N
N
N
O
N
N
O
N
O
N
O
R
R
Compd
a
#
R
A2B nM SD
or (95% CI)b
A1 nM SD
or (95% CI)c
A2A nM SD
or (95% CI)d
Compd
a
#
R
A2B nM SD
or (95% CI)b
A1 nM SD
or (95% CI)c
A2A nM SD
or (95% CI)d
10
11
12
13
14
15
16
17
18
19
20
21e
22
23
24
25
2-Me
3-Me
4-Me
2-MeO
3-MeO
4-MeO
2-Cl
1120 (868–1451)
>6666
526 264
>6000
>6000
>5000
>5000
26
27
28
29
3e
2-Me
3-Me
4-Me
2-MeO
3-MeO
4-MeO
2-Cl
438 65
700 16
91 (80–100)
127 16
50 22
>10,000
—
>15,000
—
1010 (640–1590)
643 (256–1610)
>6000
>6000
>5000
>5000
113 (60–200)
1160 440
>10,000
2190 370
—
557 (370–830)
1680 (940–2980)
2780 (1700–4530)
1830 (1130–2950)
42 10
2940 480
>5000
—
>5000
>5000
>5000
>6000
>6000
30
31
32
33
34
35
36
—
37
38
39
250 67
5830
—
3-Cl
4-Cl
6.4 (3.2–12.8)
—
>6000
407 (120–1360)
—
>5000
3-Cl
4-Cl
2080 (1370–3170)
3660
>6000
—
—
>5000
—
—
5170
>2000
2-F
3-F
2-F
3-F
645 230
>2000
220 (140–355)
1450 (870–2400)
215 154
325
2010 (890–4540)
1830 1505
473 (290–770)
>6000
>5000
>5000
>6000
>6000
>5000
>5000
4-F
4-F
—
2-CF3
3-CF3
4-CF3
4-CN
3140 (2000–4940)
>5000
>5000
945 (540–1660)
485 (300–780)
4520 (2730–7480)
3-CF3
4-CF3
4-CN
1010 (603–1697)
255 40
207 (160–270)
>6000
5180 1193
288 (220–380)
>5000
4910
>5000
1170 (350–1430)
>6000
>5000
a All compounds were >95% pure by HPLC and characterized by 1H NMR, MS, and LC/MS. Data shown are mean with 95% confidence intervals (n = 4) or mean standard deviation (n = 6–8).
b Binding affinity for the human A2B AdoR was determined by competition for binding sites labeled by 3H-ZM241385 (14 nM) in membranes prepared from HEK-A2B (human embryonic kidney) cells.10
c Binding affinity for the human A1 AdoR was determined by competition for binding sites labeled by 3H-CPX (0.5 nM) in membranes prepared from Chinese hamster ovary (CHO)-A1 cells.10
d Binding affinity for the human A2A AdoR was determined by competition for binding sites labeled by 3H-ZM241385 (2 nM) in membranes prepared from HEK-A2A cells.1
e Select compounds were tested for their binding affinity for human A3 AdoR by displacement of specific binding of [125I]AB-MECA in membranes prepared from CHO-A3 cells.

J. Zablocki et al. / Bioorg. Med. Chem. Lett. 15 (2005) 609–612
611
therapeutic effect with fewer side effects. Jacobson and
co-workers recently described a selective A2B AdoR
antagonist, compound 2 (A2B AdoR Ki = 2 nm, selectiv-
ity A1 210: A2A 260: A3 290);7 however, this compound
is not metabolically stable. Our goal was to discover a
metabolically stable compound by replacing the anilide
moiety.7 Herein, we describe the preparation and affinity
of 3-phenyl-1,2,4-oxadiazoles and 5-phenyl-1,2,4-
oxadiazoles (Table 1) as amide surrogates for MRS-
1754.
nyl-1,2,4-oxadiazole series with the exception of the
meta-chloro analogues (17 vs 32). Although these series
were designed as constrained mimetics of p-cyanoanilide
2 (Fig. 1), the corresponding p-cyano analogues 25 and
39 did not have high affinity or selectivity for the A2B
subtype (Table 1). The oxadiazole is a constrained
amide surrogate that led to a different SAR result than
the anilide series previously described by Jacobson and
co-workers.7 The SAR of the 3-phenyl-1,2,4-oxadiazole
series (Table 1, left panel) for the A2B AdoR regarding
substitution on the phenyl ring demonstrates a slight
trend favoring electron withdrawing groups (EWGs),
16–25, over electron donating groups (EDGs), 10–15.
Within the compounds containing EWGs there is not
a clear preference for ortho versus meta versus para sub-
stitution. In fact, the ortho-phenyl substitution was pre-
ferred for trifluoromethyl (22), meta substitution for
chloro (17), and para substitution for fluoro (21) with re-
spect to A2B AdoR affinity. Within the 3-phenyl-1,2,4-
oxadiazole series the meta-chloro analogue 17 had the
highest affinity for the A2B AdoR (Ki = 42 10 nM);
however, it was found to have higher affinity also for
the A1 AdoR (Ki = 6.4 nM). The compound from the
3-phenyl-1,2,4-oxadiazole series with the best selectivity
and affinity for the A2B AdoR is the para fluoro
analogue 21 (A2B AdoR Ki = 215 nM, selectivity A1 8:
A2A 23: A3 39).
All of the compounds containing an amide surrogate of
Table 1 were prepared by reacting the corresponding
chloromethyl derivatives III or VI with phenol 9. The
3-phenyl-1,2,4-oxadiazoles III and the 5-phenyl-1,2,4-
oxadiazoles VI were prepared from the corresponding
amide oximes and acid chlorides by acylation followed
by condensation.8,9 The commercially available 5,6-di-
aminouracil 6 was condensed with 4-benzyloxybenzoic
acid followed by sodium hydroxide mediated xanthine
ring closure to afford 7.10,11 After N-7 protection (tri-
methylsilylethoxymethyl, SEM) to afford 8 and hydrog-
enolysis of the benzyl protecting group, the resultant
phenol 9 was reacted with the chloromethyl 1,2,4-
oxadiazoles III and VI as shown in Scheme 1.12 The tar-
get compounds of Table 1 were obtained by deprotec-
tion of N-7 by treatment with acid.
The SAR of the 3-phenyl and 5-phenyl-1,2,4-oxadiazole
analogues with respect to A2B AdoR affinity is shown in
Table 1. A general comparison of similarly substituted
phenyl analogues between the 3-phenyl and 5-phenyl
series (e.g. 10 vs 26, 11 vs 27, etc) series favors the 5-phe-
In direct contrast to the 3-phenyl-1,2,4-oxadiazole ser-
ies, the 5-phenyl-1,2,4-oxadiazole series (Table 1, right
panel) favors EDGs on the phenyl ring (3, 26–30) over
EWGs (31–39) in relation to affinity for the A2B AdoR.
There is a clear trend with respect to phenyl substitution
OH
N
R
I
O
O
(a)
N
O
NH2
N
(b)
R
NH2
Cl
R
R
+
Cl
Cl
N
O
Cl
4
II
III
Cl
R
COCl
H2N
O
O
N
(a)
IV
(b)
R
O
N
+
Cl
NH2
N
HO
Cl
V
VI
N
5
O
O
H
NH2
NH2
N
c, d
N
e, f
N
O
Ph
N
O
N
O
N
7
6
O
N
O
H
N
SEM
N
X
Y
N
R
O
N
g, h
OH
N
N
O
N
O
N
VII
9
Scheme 1. Reagents and conditions: (a) DIEA, CH2Cl2, rt, 24 h; (b) toluene, reflux, 24 h; (c) 4-benzyloxybenzoic acid, EDCI, MeOH; (d) NaOH
(2 N), MeOH; (e) SEM-Cl, K2CO3, DMF, 23 ꢁC; (f) H2, Pd–C, MeOH; (g) III or VI, K2CO3, 56 ꢁC; (h) 1 N HCl, EtOH, 83 ꢁC.

612
J. Zablocki et al. / Bioorg. Med. Chem. Lett. 15 (2005) 609–612
8. Ibrahim, P.; Shenk, K.; Elzein, E.; Palle, V.; Zablocki, J.;
Rehder, K. World Patent 03/008411 A1, Jan. 30, 2003.
9. Elzein, E.; Ibrahim, P.; Koltun, D. O.; Rehder, K.; Shenk,
K. D.; Marquart, T.; Jiang, B.; Li, X.; Natero, R.; Li, Y.;
Nguyen, M.; Kerwar, S.; Chu, N.; Soohoo, D.; Hao, J.;
Maydanik, V. Y.; Lustig, D. A.; Zeng, D.; Leung, K.;
Zablocki, J. A. Bioorg. Med. Chem. Lett. 2004, 14, 6017–
6021.
and affinity for the A2B AdoR within the 5-phenyl-1,2,4-
oxadiazole series: MeO > Me > F = CF3 > Cl. Again,
there is not a clear preference with regards to ortho ver-
sus meta versus para phenyl substitution in the 5-phenyl-
1,2,4-oxadiazole series. The para-Me analog 28 had
favorable affinity for the A2B AdoR (Ki = 91 nM); how-
ever, it had similar affinity for the A1 AdoR
(Ki = 113 nM). The ortho-MeO analogue 29 had favor-
able affinity for the A2B AdoR (Ki = 127 nM) and mod-
est selectivity (A1 9: A2A 23: A3 5). The meta-MeO
analogue 3 had the best overall affinity for the A2B
AdoR (Ki = 50 22 nM) and binding selectivity
(A1 > 200: A2A > 200: A3 > 167). Therefore, although
the SAR did not parallel the anilide series, MRS-1754
(2), comparable affinity and selectivity was obtained in
the 5-phenyl-1,2,4-oxadiazole series with the analogue
3.13
10. Kalla, R.; Perry, T.; Elzein, E.; Varkhedkar, V.; Li, X.;
Ibrahim, P.; Palle, V.; Xiao, D.; Zablocki, J. World patent
application 03/042214, May 22, 2003.
11. Alternative synthesis: 4-{2-[5-(2-methoxyphenyl)1,2,4-
oxadiazol-3-yl]methoxy}benzoic acid (3.0 g), 5,6-diamino-
1,3-dipropyl-1,3-dihydropyrimidine-2,4-dione (3.2 g), and
1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide hydro-
chloride (3.0 g) in DMF (50 mL) was stirred for 18 h
at 23 ꢁC. After concentration in vacuo (rotovap and
high vac), the residue was treated with NaOH (2 N,
150 mL) for 2 h. After cooling to 0 ꢁC, the reaction was
acidified to pH 2–3, extracted with ethyl acetate
(2 · 300 mL), washed with water, and concentrated to a
small volume (20 mL). The resultant solid was filtered,
washed with ethyl acetate (20 mL), and ethyl acetate–
methanol (20 mL) to afford 29: NMR (DMSO-d6) d 0.89
(t, J = 8.0 Hz, 3H), 0.92 (t, J = 8.0 Hz, 3H), 1.84–1.52 (m,
4H), 3.88 (t, J = 8.0 Hz, 2H), 3.96 (s, 3H), 4.03 (t,
J = 8.0 Hz, 2H), 5.46 (s, 2H), 7.18 (t, J = 8.0 Hz, 1H),
7.24 (d, J = 8.0 Hz, 2H), 7.33 (d, J = 8.0 Hz, 1H), 7.70 (t,
J = 8.0 Hz, 1H), 8.03 (d, J = 8.0 Hz, 1H), 8.12 (d,
J = 8.0 Hz, 2H).
12. To prepare 7 4-benzyloxybenzioc acid was substituted in
place of 4-{2-[5-(2-methoxyphenyl)1,2,4-oxadiazol-3-yl]-
methoxy}benzoic following the procedure outlined in
Ref.11. The N-7 SEM protection was accomplished by
treatment of 7 (3.8 g, 9.1 mmol), potassium carbonate
(6.27 g, 45.4 mmol) in DMF (100 mL) with SEM-Cl
(3.31 mL, 18 mmol) at 70 ꢁC for 72 h. After concentration
in vacuo, the residue was purified by applying flash
chromatography (ethyl acetate–hexane 3/7) to afford the
N-7 SEM protected 7. The N-7-SEM protected 7 (1.74 g,
3.2 mmol) was converted to 9 by hydrogenolysis of the
benzyloxy protecting group using palladium hydroxide
(10%, 1.0 g) in methanol at 23 ꢁC at 5 atm hydrogen for
16 h. The resultant suspension was filtered through Celite,
washed with methylene chloride–methanol (1/1,
2 · 40 mL) to afford 9 after concentration in vacuo.
Compound 9 (50 mg, 0.1 mmol) and potassium carbonate
(0.5 g) in acetone was reacted with 5-[(3-methoxy)phenyl]-
3-chloromethyl-oxadiazole (0.1 mmol) at 56 ꢁC for 16 h.
After concentration in vacuo, the residue was diluted with
ethyl acetate, filtered, and purified by applying preparative
TLC (ethyl acetate–hexane 3/7) to afford SEM protected 3.
The SEM protected 3 (50 mg) was dissolved in ethanol
(2 mL) and treated with HCl (1 N, 0.5 ml) for 2 h at 86 ꢁC.
After cooling to 23 ꢁC, the solid obtained was filtered,
washed with ethanol (3 · 2 mL) to afford pure 3: 1H NMR
(DMSO-d6) d 1.00–0.82 (m, 6H), 1.82–1.54 (m, 4H), 3.89
(s, 3H), 4.12–3.80 (m, 4H), 5.48 (s, 2H), 7.26 (d,
J = 8.0 Hz, 2H), 7.33 (d, J = 8.0 Hz, 1H), 7.59 (t,
J = 8.0 Hz, 1H), 7.63 (s, 1H), 7.75 (d, J = 8.0 Hz, 1H),
8.13 (d, J = 8.0 Hz, 2H).
We found the p-cyanoanilide 2 to be metabolically
unstable in liver S-9 incubations as compared to 29
and 3 (29ꢀ3 > 2). Although the metabolic stability
was enhanced by replacing the anilide by the 5-phenyl-
1,2,4-oxadiazole, we did not observe any appreciable
amounts of 29 in plasma following oral dosing in rats
(10 mg/kg). In conducting these studies, we discovered
that the 5-phenyl-1,2,4-oxadiazole series as a whole
had very poor aqueous solubility, and this may be con-
tributing to the unfavorable oral availability. The 5-phen-
yl-1,2,4-oxadiazole series including analogues 29 and 3
served as a starting point for subsequent series of selec-
tive, high affinity A2B AdoR antagonists with improved
oral bioavailability that will be described elsewhere. The
discovery of selective, high affinity A2B AdoR antago-
nists may help further define the role of the A2B AdoR
in asthma.14
Acknowledgements
We would like to thank Dr. Brent Blackburn, and Dr.
Luiz Belardinelli, for valuable input and discussion.
The authors would like to thank Yuzhi Wu and Marie
Nguyen for technical help.
References and notes
1. Fredholm, B. B.; Ijzerman, A. P.; Jacobson, K. A.; Klotz,
K. N.; Linden, J. Pharmacol. Rev. 2001, 53, 527–552.
2. Ryzhov, S.; Goldstein, A. E.; Matafonov, A.; Zeng, D.;
Biaggioni, I.; Feoktistov, I. J. Immunol. 2004, 172, 7726–
7733.
3. Zhong, H.; Belardinelli, L.; Maa, T.; Feoktistov, I.;
Biaggioni, I.; Zeng, D. Am. J. Respir. Cell Mol. Biol.
2004, 30, 118–125.
4. Stine, R. J.; Marcus, R. H.; Parvin, C. A. Ann. Emerg.
Med. 1989, 18, 640–646.
5. DÕAlonzo, G. E. Allergy Asthma Proc. 1996, 17, 335–339.
6. Powell, E. C.; Reynolds, S. L.; Rubenstein, J. S. Pediatr.
Emerg. Care 1993, 9, 129–133.
13. MRS-1754 was determined to have a Ki of 16 nm under
our assay conditions.
14. Baraldi, P. G.; Tabrizi, M. A.; Preti, D.; Bovero, A.;
Romagnoli, R.; Fruttarolo, F.; Zaid, N. A.; Moorman, A.
R.; Varani, K.; Gessi, S.; Merighi, S.; Borea, P. A. J. Med.
Chem. 2004, 47(6), 1434–1447.
7. Kim, Y. C.; Ji, X.; Melman, N.; Linden, J.; Jacobson, K.
A. J. Med. Chem. 2000, 43, 1165–1172.