Solubilized phenyl-pyrazole ureas as potent, selective 5-HT2A inverse-agonists and their application as antiplatelet agents

Article abstract of DOI:10.1016/j.bmcl.2009.07.073

Potent 5-HT_2A inverse-agonists containing phenyl-pyrazole ureas with an amino side chain were identified. Optimization of this series resulted in selective compounds that proved effective in modulating 5HT-induced amplification of ADP-stimulate

Full text of DOI:10.1016/j.bmcl.2009.07.073

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5486–5489
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Solubilized phenyl-pyrazole ureas as potent, selective 5-HT2A
inverse-agonists and their application as antiplatelet agents
*
Peter I. Dosa , Sonja Strah-Pleynet, Honnappa Jayakumar, Martin Casper, Marc Decaire, Yifeng Xiong,
Juerg Lehmann, Karoline Choi, Katie Elwell, Amy Wong, Robert R. Webb, John W. Adams, Juan Ramirez,
Jeremy G. Richman, William Thomsen, Graeme Semple, Bradley R. Teegarden
Arena Pharmaceuticals Inc., 6166 Nancy Ridge Drive, San Diego, CA 92121, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Potent 5-HT2A inverse-agonists containing phenyl-pyrazole ureas with an amino side chain were identi-
fied. Optimization of this series resulted in selective compounds that proved effective in modulating 5HT-
induced amplification of ADP-stimulated human platelet aggregation.
Received 12 June 2009
Revised 10 July 2009
Accepted 14 July 2009
Available online 18 July 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
5-HT2A
Antiplatelet
Antithrombosis
The 5-HT2A receptor is one of at least 15 different serotonin (5-
HT) receptor subtypes, several of which regulate important behav-
ioral responses.1,2 Inverse-agonists of the 5-HT2A receptor are
known to improve sleep maintenance3,4 and to alleviate negative
symptoms in schizophrenia.5,6 Several clinical candidates have
been developed for these central nervous system (CNS) indica-
tions,4 including volinanserin (Sanofi-Aventis),7 and pimavanserin
(Acadia, Fig. 1).8,9
In addition to its CNS activities, the 5-HT2A receptor plays an
important role in regulating cardiovascular functions, including
platelet aggregation.10,11 By itself, serotonin does not significantly
alter platelet function, but has been found to amplify aggregation
induced by an agonist such as collagen, epinephrine, thrombin or
ADP. 5-HT2A inverse-agonists are known to prevent this amplifica-
tion and reduce the formation of blood clots (thrombogenesis).12
Previously, we identified a novel series of phenyl-pyrazole urea-
based 5-HT2A inverse-agonists which unlike most known inverse-
agonists of the receptor, lack a basic amine function.13 While these
compounds are potent 5-HT2A inverse-agonists, they possess
limited intrinsic solubility. This study focuses on our attempts to
mitigate this potential liability by incorporating an amino moiety
onto the phenyl-pyrazole urea scaffold. The ability of the resulting
compounds to attenuate serotonin-induced amplification of plate-
let activation was then evaluated.
We began our investigation by modifying compounds we
already had available as part of our 5-HT2A insomnia program.13
Using the potent (Ki = 1.5 nM) but insoluble inverse-agonist 1 as
a starting point, demethylation with boron tribromide furnished
phenol 2 in 45% yield (Scheme 1). Initial attempts at alkylation of
2 using Williamson ether synthesis conditions (DBU, alkyl halide)
led to a mixture of products, but utilizing a Mitsunobu protocol14
15
afforded 3 in 57% yield.
In cultured HEK cells transfected with the constitutively active
5-HT2A receptor, 3 inhibited serotonin-independent inositol phos-
phate accumulation (IP) with an IC50 of 1.9 nM, demonstrating that
it is a functional inverse-agonist.16 A Ki of 0.28 nM was determined
by a competitive binding assay against 125I-radiolabeled DOI at the
5-HT2A receptor (Table 1). However, 3 showed only sixfold selectiv-
ity at the related 5-HT2C receptor (Ki = 1.6 nM). Further evaluation
of 3 in a multi-receptor panel (CEREP) found submicromolar activ-
* Corresponding author at present address: Institute for Therapeutics Discovery
and Development, Room 455, College of Pharmacy, University of Minnesota, 717
Delaware Street SE, Minneapolis, MN 55414, United States.
E-mail address: pidosa@umn.edu (P.I. Dosa).
Figure 1. Structures of known 5-HT2A inverse-agonists.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.073

P. I. Dosa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5486–5489
5487
Table 1
Amino solubilizing group SAR
ity against several other GPCRs, including muscarinic M1, dopa-
mine D1 and D5, and opiate. Compound 3 also showed 73% inhi-
bition of the hERG channel at 1 M, likely preventing its use in any
l
l
clinical applications. The pharmacokinetic profile of 3 was evalu-
ated in male Sprague-Dawley rats. Compound 3 was dosed at
2.0 mg/kg (iv) or 10.0 mg/kg (po) using 80% PEG400 in PBS (pH
7.4) and had a t1/2 of 5.1 h and F = 19%. Compound 3 demonstrated
significantly improved solubility compared with 1, both under
neutral conditions (0.2 mg/ml in a pH 7.0 sodium phosphate buf-
fer) and acidic conditions (16 mg/ml in a pH 4.0 sodium acetate
buffer).
Compounds
X
R
5-HT2A Kia (nM)
5-HT2C Kia (nM)
3
Br
0.28
1.6
While the potency of 3 was adequate, a more selective com-
pound, particularly with respect to hERG channel blockade, was
needed for possible clinical applications. A new synthetic route
was developed in order to more readily access a larger variety of
solubilized phenyl-pyrazole ureas (Scheme 2). Starting with 4,15
demethylation with aluminum trichloride afforded phenol 5. A
Mitsunobu coupling of 5 with 2-bromoethanol led to 6. Pyrazole
halogenation could be achieved by treating 6 with NCS or NBS to
yield 7 or 8, respectively. Intermediates 6–8 were then reacted
with an amine to furnish 9a–9c. Amide hydrolysis under basic con-
ditions followed by coupling with an isocyanate led to the desired
final products 11a–c.
Our initial SAR efforts were directed at determining the neces-
sity of pyrazole halogenation for potency at the 5-HT2A receptor.
Pyrrolidine, piperidine, and morpholine analogs were prepared in
brominated, chlorinated, and nonhalogenated versions. From
examination of the data in Table 1 (compounds 3, 12–19), it is clear
that pyrazole halogenation (X = Br or Cl) is not required for potency
at the 5-HT2A receptor. In addition, removal of the pyrazole halo-
gen generally leads to compounds with increased selectivity for
the 5-HT2A receptor compared to the 5-HT2C receptor.
Holding the rest of the structure constant, a set of molecules
incorporating a wide range of amino substituents (R in Table 1)
were synthesized. Several compounds, including those substituted
with piperidine (16), morpholine (19), and 4-acetylpiperazine (27),
maintained subnanomolar potency in the competitive 5-HT2A
binding assay, but showed significantly improved selectivity over
5-HT2C compared to the pyrrolidine analog 13.16
Two analogs bearing a longer three carbon tether (21 and 22)
were prepared by a Mitsunobu coupling route similar to that
shown in Scheme 1. While 22 proved to be significantly less potent
and selective than its two carbon chain analog 19, no such differ-
ence was seen between 21 and 20.
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Cl
H
Br
Cl
H
Br
Cl
H
H
H
H
H
H
H
H
H
H
0.26
0.15
0.66
0.43
0.41
0.63
0.61
0.19
0.43
0.55
4.0
3.0
9.0
21.4
24.0
58.9
12.7
20.0
213
8.8
24.3
151
187
98.7
133
102.7
216
66.2
0.58
0.22
17.6
2.1
Next, a phenyl urea optimization was undertaken, using piperi-
dine as the amino side chain (Table 2). The results of this study show
that a range of electron withdrawing and donating substituents
were well tolerated on the phenyl ring of the urea. Compounds bear-
ing substituents on the 3- or 4-positions of the phenyl ring proved to
be significantly more potent than those bearing the same substitu-
ent at the 2-position. Several compounds (38, 39, and 42) were found
to have 5-HT2A binding affinities of less than 1 nM with selectivities
0.96
0.42
29
30
31
32
H
H
H
H
0.12
0.69
0.99
0.84
29.6
5.4
of better than 500-fold for 5-HT2A over 5-HT2c
.
Several of these compounds were evaluated for their ability to in-
hibit serotonin-induced amplification of ADP-stimulated human
platelet aggregation. Aggregation was measured turbidometrically
at 37 °C. Platelet rich human plasma was pre-incubated with the
phenyl-pyrazole urea for 1 min before aggregation was induced by
50.7
98
the simultaneous addition of 1 lM serotonin and 1 lM ADP. In the
absence of an inhibitor, addition of ADP by itself causes approxi-
mately 10–20% of maximal aggregation, while addition of both sero-
tonin and ADP causes maximal aggregation. The IC50 is defined as the
concentration of inhibitor (the phenyl-pyrazole urea) at which half
of the serotonin amplification effect was reversed.17
33
H
16.2
481
a
Mean of at least three determinations with s.d. < 0.4 log units.

5488
P. I. Dosa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5486–5489
Scheme 1. Reagents and conditions: (a) BBr3, 1,2-dichloroethane, 100 °C, 45%; (b) PPh3, DIAD, 1-(2-hydroxyethyl)pyrrolidine, THF, rt, 57%.
Scheme 2. Reagents and conditions: (a) AlCl3, 1,2-dichloroethane, 0–80 °C, 83%; (b) PPh3, DIAD, THF, rt, 82%; (c) for 7: NCS, MeOH, 150 °C, 47%; for 8: NBS, DMA, rt, 2 h, 93%;
(d) R1-H, DMF, DIEA; (e) NaOH in MeOH/H2O, 150 °C, 30 min; (e) R2NCO, CH2Cl2, rt.
Ureas 38, 39, and 42 were found to inhibit serotonin-induced
amplification of ADP-stimulated human platelet aggregation with
IC50’s of 31, 17, and 6 nM, respectively. Table 3 shows inhibition
of platelet aggregation data for a range of compounds with 5-
HT2A binding affinities of less than 2 nM. In general, the addition
of a basic amine moiety led to compounds which showed greater
effects on platelet aggregation than related insoluble compounds
(for instance, compound 3 had an IC50 of 21 nM while compound
1 had an IC50 of 296 nM). These compounds are all less potent in
the platelet inhibition assay than in the binding assay, but the
reduction in potency varies widely from 3.5-fold (compound 49)
to over 500-fold (compound 55).
Table 3
Inhibition of 5HT-induced amplification of ADP-stimulated human platelet
aggregation
Compounds
R1
R2
X
5-HT2A
5-HT2c
Inhibition
of platelet
aggregation
IC50 (nM)
Kia (nM)
Kia (nM)
Table 2
Phenyl urea SAR
49
50
51
52
53
3
4-F
H
0.86
0.18
0.28
0.36
1.9
53.2
12.4
11.0
9.0
3
10
7
2,4-F
2,4-F
4-F
Br
Cl
Br
Br
Br
H
Compounds
R
5-HT2A Kia (nM)
5-HT2c Kia (nM)
21
17
21
6
34
35
36
16
37
38
39
40
41
42
43
44
45
46
47
48
H
2-Cl
3-Cl
4-Cl
2-F
3-F
4-F
4-Me
3-OMe
4-OMe
3-CF3
4-CF3
3-Acetyl
2,4-F
2-F, 5-Me
3,5-F
1.4
2.5
1090
223
21.7
58.9
1150
760
466
104
403
385
2.9
18.9
59.1
1170
210
147
4-OH
4-Cl
810
1.6
0.33
0.41
6.1
0.99
0.84
1.4
0.28
0.54
42
4-OMe
385
3.3
0.54
0.25
0.34
14.3
3.2
39
38
54
4-F
H
H
Cl
0.84
0.99
0.15
466
760
405
17
31
34
3-F
2.1
0.91
2,4-F
a
Mean of at least three determinations with s.d. < 0.4 log units.

P. I. Dosa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5486–5489
5489
2. Sanders-Bush, E.; Fentress, H.; Hazelwood, L. Mol. Interventions 2003, 3,
319.
3. Sharpley, A. L.; Solomon, R. A.; Fernando, A. I.; da Roza Davis, J. M.; Cowen, P. J.
Psychopharmacology 1990, 101, 568.
4. Teegarden, B. R.; Al-Shamma, H.; Xiong, Y. Curr. Topics Med. Chem. 2008, 8, 969.
5. Abi-Dargham, A.; Laruelle, M.; Aghajanian, G. K.; Charney, D.; Krystal, J. J.
Neuropsych. Clin. Neurosci. 1997, 9, 1.
Table 3 (continued)
Compounds
R1
R2
X
5-HT2A
5-HT2c
Inhibition
of platelet
aggregation
IC50 (nM)
Kia (nM)
Kia (nM)
6. Di Pietro, N. C.; Seamans, J. K. Pharmacopsych. 2007, 40, S27.
7. Kehne, J. H.; Baron, B. M.; Carr, A. A.; Chaney, S. F.; Elands, J.; Feldman, D. J.;
Frank, R. A.; Van Giersbergen, P. L. M.; McCloskey, T. C.; Johnson, M. P.;
McCarty, D. R.; Poirot, M.; Senyah, Y.; Siegel, B. W.; Widmaier, C. J. Pharmacol.
Exp. Ther. 1996, 277, 968.
55
3,5-F
3-F
Cl
H
0.31
0.82
37.8
183
89
56
1230
8. Vanover, K. E.; Weiner, D. M.; Makhay, M.; Veinbergs, I.; Gardell, L. R.; Lameh,
J.; Del Tredici, A. L.; Piu, F.; Schiffer, H. H.; Ott, T. R.; Burstein, E. S.; Uldam, A. K.;
Thygesen, M. B.; Schlienger, N.; Andersson, C. M.; Son, T. Y.; Harvey, S. C.;
Powell, S. B.; Geyer, M. A.; Tolf, B.; Brann, M. R.; Davis, R. E. J. Pharm. Exp. Ther.
2006, 317, 910.
a
Mean of at least three determinations with s.d. < 0.4 log units.
One of the compounds showing the greatest 5HT2A/5HT2C selec-
tivity, 54, was evaluated in multi-receptor panel (CEREP) and
proved to be significantly more selective than 3, giving submi-
cromolar activity against only the 5HT2A and 5HT2C receptors. In
addition, 54 showed only 7% inhibition of the hERG channel at
9. Roberts, C. Curr. Opin. Invest. Drugs 2006, 7, 653.
10. De Clerck, F. F.; Herman, A. G. Fed Proc. 1983, 42, 228.
11. Nagatomo, T.; Rashid, M.; Abul Muntasir, H.; Komiyama, T. Pharmacol. Ther.
2004, 104, 59.
12. De Clerck, F. F.; Janssen, P. A. J. Hypertens., 1990, S87.
13. Teegarden, B. R.; Li, H.; Jayakumar, H.; Strah-Pleynet, S.; Dosa, P. I.; Selaya, S. D.;
Kato, N.; Elwell, K. H.; Davidson, J.; Cheng, K.; Saldana, H.; Frazer, J. M.; Whelan,
K.; Foster, J.; Espitia, S.; Webb, R. R; Beeley, N.R.A.; Kilduff, T.S.; Morairty, S.;
Thomsen, W.; Al-Shamma, H., submitted for publication.; For an example of
another non-basic 5-HT2A antagonist see: Ladduwahetty, T.; Boase, A. L.;
Mitchinson, A.; Quin, C.; Patel, S.; Chapman, K.; MacLeod, A. M. Bioorg. Med.
Chem. Lett. 2006, 3201.
14. Hughes, D. L. In Organic Reactions; Paquette, L. A., Ed.; Wiley: New York, 1992;
Vol. 42, pp 335–656.
15. a For experimental details, see: Teegarden, B.; Jayakumar, H.; Li, H.; Strah-
Pleynet, S.; Dosa, P. I. Preparation of diaryl and arylheteroaryl urea derivatives, in
particular pyrazolylphenyl ureas, as 5-HT2A serotonin receptor modulators.
1 l
M18 and was soluble under acidic conditions (1.9 mg/ml in
0.01 N HCl), though not at neutral pH (0.01 mg/ml in a sodium
phosphate buffer). The pharmacokinetic profile of 54 was evalu-
ated in rats. Compound 54 had a t1/2 of 2.8 h and F = 11%.
In summary, addition of an amino group to the previously iden-
tified phenyl-pyrazole urea scaffold led to several compounds with
high 5-HT2A potency and good selectivity over the related 5-HT2C
receptor. Compounds were identified that were highly effective
at inhibiting serotonin-induced amplification of ADP-stimulated
human platelet aggregation and which had reduced off-target
liabilities.
WO2005012254, 2005.;
b Teegarden, B.; Xiong, Y.; Strah-Pleynet, S.;
Jayakumar, H.; Dosa, P. I.; Feichtinger, K.; Casper, M.; Lehmann, J.; Jones, R.
M.; Unett, D. J. Preparation of 3-phenylpyrazole derivatives as modulators of 5-
HT2A serotonin receptors for the treatment of related disorders. WO2006055734,
2006.
Acknowledgements
16. The solubilities of compound 13 (0.25 mg/ml in sodium phosphate buffer and
2.8 mg/ml in 0.01 N HCl) and compound 19 (0.15 mg/ml and 3.3 mg/ml) were
similar to that of compound 3.
17. Adams, J. W.; Ramirez, J.; Ortuno, D.; Shi, Y.; Thomsen, W.; Richman, J. G.;
Morgan, M.; Dosa, P.; Teegarden, B. R.; Al-Shamma, H.; Behan, D. P.; Connolly,
D. T. Eur. J. Pharmacol. 2008, 586, 234.
We thank Hong Zheng, John Frazer, Rosemarie Angeles and
Diane Yuskin for their excellent assistance in running assays and
Dan Connelly for useful discussions.
18. Inhibition of the hERG channel varied based on the substituents in the
compounds, but in general molecules containing the less basic morpholino
References and notes
group, like 54 and 56 (29% at 3
l
M), showed less inhibition than those with
1. Hoyer, D.; Hannon, J. P.; Martin, G. R. Pharmacol., Biochem. Behav. 2002, 71, 533.
more basic groups like 32 (90% at 3
l
M) and 50 (64% at 1 M).
l