Pyrrolinone derivatives as a new class of P2X3 receptor antagonists. Part 3: Structure-activity relationships of pyrropyrazolone derivatives

Article abstract of DOI:10.1016/j.bmcl.2020.127636

The P2X3 receptor is an attractive target for the treatment of pain and chronic coughing, and thus P2X3 antagonists have been developed as new therapeutic drugs. We previously reported selective P2X3 receptor antagonists by derivatization of hit compound 1. As a result, we identified hit compound 3, the structure of which was similar to hit compound 1. On the basis of SAR studies of hit compound 1, we modified hit compound 3 and compound 42 was identified as having analgesic efficacy by oral administration.

Full text of DOI:10.1016/j.bmcl.2020.127636

Bioorganic&MedicinalChemistryLetters30(2020)127636
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrrolinone derivatives as a new class of P2X3 receptor antagonists. Part 3:
Structure-activity relationships of pyrropyrazolone derivatives
⁎
Hiroyuki Tobinagaa, , Takayuki Kameyamab, Kentarou Asahia, Tohru Horiguchia, Miho Ooharaa,
Yoshiyuki Taodaa, Kayoko Hataa, Tsuyoshi Hasegawac, Yukio Tadaa, Naoko Kuriharaa,
Yasuhiko Kandaa, Shigenori Yagia, Maki Tomaric, Yoshikazu Tanakad, Fumiyo Takahashie,
Emiko Taniguchie, Yukio Takaharae, Shinji Shimadac, Chie Takeyamac, Shoichi Yamamotof,
⁎
Shunji Shinoharac, Hiroyuki Kaia,
a Laboratory for Medicinal Chemistry Research, Shionogi & Co., Ltd., 1-1 Futaba-cho 3-chome, Toyonaka, Osaka 561-0825, Japan
b Office for Children’s Bright Future, Shionogi & Co., Ltd., 2F, Nissay Yodoyabashi East, 3-13, Imabashi 3-chome, Chuo-ku, Osaka 541-0042, Japan
c Shionogi TechnoAdvance Research Co., Ltd., 1-1 Futaba-cho, 3-chome, Toyonaka, Osaka 561-0825, Japan
d Research Planning Department, Shionogi & Co., Ltd., 1-1 Futaba-cho, 3-chome, Toyonaka, Osaka 561-0825, Japan
e Laboratory for Drug Discovery & Disease Research, Shionogi & Co., Ltd., 1-1 Futaba-cho, 3-chome, Toyonaka, Osaka 561-0825, Japan
f Shionogi Marketing Solutions Co., Ltd., 8F, Nissay Yodoyabashi East, 3-13, Imabashi 3-chome, Chuo-ku, Osaka 541-0042, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
The P2X3 receptor is an attractive target for the treatment of pain and chronic coughing, and thus P2X3 an-
tagonists have been developed as new therapeutic drugs. We previously reported selective P2X3 receptor an-
tagonists by derivatization of hit compound 1. As a result, we identified hit compound 3, the structure of which
was similar to hit compound 1. On the basis of SAR studies of hit compound 1, we modified hit compound 3 and
compound 42 was identified as having analgesic efficacy by oral administration.
Purinergic receptors
Ion channels
P2X3 receptor antagonists
Pyrrolopyrazolone derivatives
Pain
Adenosine triphosphate (ATP) is created from high-energy phos-
phate bonds and consumed as a source of energy by all animals and
plants. Intracellular ATP is involved various processes including the
synthesis of biomolecules as well as the active transport of chemical
substances and bioluminescence. Extracellular ATP acts as a neuro-
transmitter by binding to purinergic P2 receptors. Two types of P2 re-
ceptors are known, P2X is a ligand-gated ion channel and P2Y is a G
protein-coupled receptor. P2X receptors are classified into seven sub-
types, from P2X1 to P2X7, and each subtype constitutes a homotrimer
or heterotrimer1. P2X receptors are widely expressed in living organ-
isms and are known to be involved in a variety of physiological phe-
nomena, including neurotransmission2, muscle contraction3, pain4,
taste5, and inflammatory responses6. Therefore, P2X receptors are re-
garded as promising targets for novel therapeutic drugs targeted toward
neurological, cardiovascular and inflammatory diseases. P2X3 re-
ceptors are primarily expressed in the peripheral sensory nerves and are
involved in neurotransmission for pain7,8. Therefore, many reported
P2X3 receptor antagonists may have applications as analgesics. Among
them, gefapixant showed improvement of interstitial cystitis/bladder
pain syndrome (IC/BPS) in a Phase 2 clinical trial. Gefapixant also
showed improvement of refractory chronic cough in a Phase 3 clinical
trial. Therefore, P2X3 receptor antagonists have the potential to be
useful for the treatment of various diseases9.
We previously reported that the exploration of hit compound 1 led
to the discovery of compound 2 as an orally available selective P2X3
antagonist10. The hit compound 3, a pyrrolopyrazolone derivative, was
also identified but its activity was less potent than hit compound 1.
Because of the high structural similarity between 1 and 3, a SAR study
was undertaken starting from hit compound 3 and with reference to the
historical results from the hit to lead SAR studies of hit compound 1.
Additionally, a docking study using a constructed P2X3 homology
model was applied to SAR studies of the pyrrolopyrazolone derivatives.
In this paper, we report the discovery of an orally available compound
from structure activity studies of pyrrolopyrazolone derivatives in
conjunction with docking studies (see Fig. 1).
Pyrrolopyrazolone derivatives were synthesized according to a
previously reported method (Scheme 1). Condensation of three com-
ponents under acidic conditions gave pyrrolinone intermediates, which
were cyclized with hydrazine hydrate to give the pyrropyrazolone de-
rivatives11. N-Me compounds were also prepared by cyclization with
⁎ Corresponding authors.
E-mail addresses: hiroyuki.tobinaga@shionogi.co.jp (H. Tobinaga), hiroyuki.kai@shionogi.co.jp (H. Kai).
https://doi.org/10.1016/j.bmcl.2020.127636
Received 3 August 2020; Received in revised form 11 October 2020; Accepted 18 October 2020
Availableonline22October2020
0960-894X/©2020ElsevierLtd.Allrightsreserved.

H. Tobinaga, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127636
Fig. 1. The structural similarity of hit compounds 1 and 3.
Scheme 1. Reagents and conditions for synthesis of the pyrrolopyrazolone skeletons: (a) AcOH, dioxane, and reflux; (b) hydrazine hydrate, AcOH, 95 °C.
Scheme 2. Reagents and conditions for synthesis of N-methylated pyrrolopyrazolone: (a) methylhydrazine, AcOH, 95 °C, 5 (44%), 6 (11%).
methylhydrazine to give 5 as a major product and 6 as a minor product
with other substituents at R1, including non-substituted (10), 3-
methoxy (11) and 4-methoxy (12), all showed reductions in activity.
Also, the chlorine substituted compounds (13–15) showed reductions in
activity. As a result of further exploration of the 2-substituted phenyl
compounds at R1, the ethoxy derivative (17) maintained similar activity
as compound 4, and the dimethylamino derivative (18) showed slightly
less activity. Introduction of a methoxy substituent to compound 9 at
position-3 or position-6 on the phenyl ring (19, 20) decreased the ac-
tivity. These results suggest that electron donating groups in the ortho
position of the phenyl ring are important for high antagonistic activity.
We explored the effects of varying the R2 substituent at position-3 of
the pyrrolopyrazolone derivatives (21–26). The n-butyl derivative (22)
showed almost the same activity as the tert-butyl derivative (9), but the
other smaller substituents (23: i-Pr, 24: Me) or larger substituents (21:
(Scheme 2)12
.
Table 1 shows the effects of substitutions at various positions
around the pyrropyrazolone ring structure on the antagonistic activity.
Compound 4 showed higher potency than hit compound 1. However,
the potency of 4 decreased compared with pyrrolinone derivatives 1
and 3. The N-methylated compounds (5, 6) showed large decreases in
potencies, which may indicate that the proton of the pyrazole ring plays
an important role in the antagonistic activity.
To reduce the number of ring structures, a 4-methoxyphenyl moiety
at position-3 on the pyrrolopyrazolone was converted into a tert-butyl
group and the effects of substituents at R1 were investigated. The 2-
methoxyphenyl derivative (9) showed a large increase in activity, while
the cyclohexyl version (8) showed a decrease in activity. Compounds
2

H. Tobinaga, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127636
Table 1
understand its efficacy in vivo, the pharmacokinetics of compound 9
was tested in rats by oral administration at a 100 mg/kg/dose. The
plasma concentration of compound 9 was lower than the estimated
concentration from a low dose test, thus, its oral bioavailability was
greatly decreased. This suggested that the low oral bioavailability of
compound 9 was caused by its low solubility at pH 6.8. Therefore, the
introduction of hydrophilic substituents on the pyrrolopyrazolone ring
structure was needed to improve the solubility.
Initial structure activity relationship studies of the pyrrolopyrazolone deriva-
tives.
Before the introduction of a hydrophilic substituent to the pyrro-
lopyrazolone derivatives, we speculated which positions were amen-
able to substitution by using a P2X3 homology model. We previously
reported the SAR studies of pyrrolinone derivatives examined by a
docking study with a P2X3 homology model that was constructed as a
template for the zebrafish P2X4 receptor15. We also studied the SAR of
pyrrolopyrazolone derivatives using a docking study based on the same
P2X3 homology model16. As a result, the docking study using the P2X3
homology model agreed well with the SAR of the pyrrolopyrazolone
derivatives as well as the pyrrolinone derivatives. In this study, solvent
accessible regions were observed around the ortho methoxy substituent
on the phenyl ring at position-4 and the tert-butyl at position-3 (Fig. 2).
Therefore, we decided to introduce hydrophilic substituents at position-
4 and position-3.
Compound
R1
R2
R3
IC50
(μM)
4
c-Hexyl
4-MeO-Ph
4-MeO-Ph
4-MeO-Ph
4-MeO-Ph
t-Bu
H
0.211
1.56
5
c-Hexyl
1-Me
2-Me
H
Table 4 shows the optimization of compound 9. To improve its so-
lubility and PK profiles, hydrophilic substituents were introduced at R2
and R5 (38–43). The conversion of the hydrophilic R5 group was carried
out while R2 was fixed as a tert-butyl group. The resulting compounds
showed good activity (38–40). In particular, compounds 39 and 40
showed great improvements in their solubilities and inhibition of cy-
tochrome P450. However, these compounds were excreted rapidly,
resulting in low oral bioavailability in rat. The conversion of the hy-
drophilic R2 from a tert-butyl moiety was also carried out (41–43).
Compound 42, synthesized by introducing a hydroxyl group into the
tert-butyl moiety, maintained high activity and had improved solubility.
Additionally, compound 42 had a similar PK profile as compound 9,
and improved inhibition of cytochrome P450 2C9. Unlike compound
42, the hydroxyethyl (41) and diol derivatives (43) showed decreased
activities. It was speculated that the activity decrease observed for 41
was caused by reducing the bulkiness at R2 and the activity decrease
observed for 43 resulted from substituting a more hydrophilic moiety at
the R2 site. The introduction of hydrophilic groups at the R2 site did not
decrease the activity and agreed well with the prediction from the
docking study of the homology model.
6
c-Hexyl
> 10
0.376
0.916
0.025
0.304
0.69
7
2-MeO-Ph
c-Hexyl
8
H
9
2-MeO-Ph
Ph
t-Bu
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
t-Bu
H
3-MeO-Ph
4-MeO-Ph
2-Cl-Ph
t-Bu
H
t-Bu
H
1.59
t-Bu
H
0.466
1.68
3-Cl-Ph
t-Bu
H
4-Cl-Ph
t-Bu
H
0.633
0.226
0.27
2-Et-Ph
t-Bu
H
2-EtO-Ph
2-Me2N-Ph
2,3-(MeO)2-Ph
2,6-(MeO)2-Ph
2-MeO-Ph
2-MeO-Ph
2-MeO-Ph
2-MeO-Ph
2-MeO-Ph
2-MeO-Ph
t-Bu
H
t-Bu
H
0.137
1.69
t-Bu
H
t-Bu
H
0.538
0.105
0.021
0.112
1.633
0.36
t-BuCH2
n-Bu
H
H
i-Pr
H
Me
H
2-Furanyl
3-Pyridyl
H
H
0.665
neopentyl, 25: 2-furanyl, 26: 3-pyridyl) tended to decrease the activity.
These results showed that substitution with an appropriate size group at
R2 was necessary for good activity.
From the above SAR investigation, compound 42 was selected for
further evaluation (Table 5). The oral bioavailability of 42 was not
decreased by oral administration of a 30 mg/kg/dose, and the plasma
concentration of 42 was sufficient to test its efficacy in vivo. The results
suggested that the improvement of solubility of compound 42 led to a
good PK profile in rats. Compound 42 showed higher analgesic effects
by oral administration of a 30 mg/kg/dose than those observed for the
100 mg/kg/dose of compound 9. It should be noted that the analgesic
effect of compound 42 at 30/mg/kg/dose was higher than that of the
clinical dose of pregabalin.
Table 2 shows the effects of substituents at R4 on the pyrropyr-
azolone ring structure on antagonistic activity. We explored the iso-
xazole moiety by fixing a tert-butyl group at position-3 and 2-methox-
yphenyl at position-4 of the pyrrolopyrazolone ring (27–37). The
ethoxycarbonyl derivative (28) showed good activity, but the fluoro
(27) and amide (28, 29) derivatives showed decreased activity. In case
of the other 5-menbered hetero aromatic compounds, almost all com-
pounds showed good activity (IC50 < 0.1 μM) except for the 1,3,4-
oxadiazole derivative (35). In particular, the 1,2,4-oxazole-3-yl (33),
thiazole-2-yl (36) and thiophene-3-yl (37) derivatives showed high
activity, which was a similar pattern those observed for previously re-
ported SAR studies of pyrrolinone derivatives10. The solubility at pH
6.8 of compound 9 was poor, but these compounds showed good PK
profiles in rat pharmacokinetics.
To confirm the predictions from the docking study using the
homology model, both the enantiomers of compound 42 were synthe-
sized (Scheme 3)17. The β-hydroxyester (44) as the starting material
was protected by THP and reacted with methyl magnesium bromide in
the presence of triethylamine to give the ketone (45). The ketone
compound was reacted with diethyl oxalate under basic conditions to
give the α, γ-ketoester (46). The condensation of three components
shown in Scheme 1 gave the pyrrolinone (47). The Mitsunobu reaction
with (S)-methyl mandelate led to the ether (48), and then deprotection
of THP under acidic conditions gave a separable mixture of the dia-
stereomers (49). After separation of the each diastereomer, removal of
(S)-methyl mandelate with hydrogenation and cyclization with hy-
drazine hydrate gave both enantiomers, (S)-42 and (R)-42. The abso-
lute stereochemistry was determined by X-ray crystal structure analysis
Although the isoxazole compound 9 showed weak inhibition of
cytochrome P450 2C9, it was selected for further evaluation (Table 3).
We tested compound 9 for its analgesic effects against hyperalgesia in a
partial sciatic nerve ligation model for neuropathic pain in rats13,14
.
Compound 9 showed analgesic effects by oral administration of a
100 mg/kg/dose in rats, and its efficacy was almost the same as a
clinical dose of pregabalin (10 mg/kg/dose: 30% reversal). To further
3

H. Tobinaga, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127636
Table 2
Structure activity relationship studies on substituents at R4.
R4
IC50
Solubility a
(μM)
CYP IC50 (μM)
1A2
CLt b (ml/min/kg)
16.6
t1/2
BA b
(%)
b
Compound
(μM)
2C9
10
3A4
(hr)
9
0.025
ND
> 20
> 20
2.1
33.8
27
28
29
30
31
F
9.3
20.4
0.2
> 20
> 20
NT
7
4
18
5
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
NT
COOEt
CONHMe
CONMe2
0.017
1.336
1.26
3.5
> 50
ND
> 20
> 20
> 20
1
> 20
2
0.072
32
33
0.087
0.026
0.9
ND
> 20
> 20
4
1
8
16.9
3.79
1.9
8.2
20
13
20.8
34
35
0.095
0.417
ND
2.6
> 20
> 20
2
5
2
8
NT
NT
NT
NT
NT
NT
36
37
0.025
0.017
ND
ND
10
2
6
13
10
19.3
15.4
2.2
3.1
26.5
36.8
> 20
CLt: total clearance; BA: oral bioavailability; fu: fraction unbound in rat serum; ND: not detected; NT: not tested.
a. Solubilities were measured in sodium phosphate buffer (pH 6.8, 10 mmol) containing 1% DMSO.
b. All compounds were administered at 0.5 mg/kg iv and 1.0 mg/kg po.
Compounds were administered as a mixture of three to five compounds.
Table 3
Biological test results of selected compounds.
Compound
Analgesic testa
Rat pharmacokinetics
dose, po
(mg/kg)
efficacy
dose, iv
(mg/kg)
t1/2
CLt
dose, po
(mg/kg)
Cmax
AUC
BA
(% reversal)
(hr)
(ml/min/kg)
(ng/ml)
(μg·hr/ml)
6.48
(%)
9
100
10
30.5
10
3.2
12.6
100
450
4.9
Pregabalin
20–30
a. Rat Seltzer model: Hyperalgesia (paw pressure test).
using similar compounds15. From the activity assessment, it was con-
firmed that (S)-42 showed higher activity compared with (R)-42. The
differences in potency between (S)-42 and (R)-42 agreed well with the
predictions from the docking study.
homology model was constructed as a template for the zebrafish P2X4
receptor, and then a docking study of the pyrrolopyrazolone derivatives
was carried out. The compounds predicted to bind to the receptor from
the docking study using the P2X3 homology model agreed well with the
experimental efficacy results of the compounds generated from the SAR
study. After further compound optimization based on the above results,
we discovered the selective P2X3 antagonist 42, which showed a good
In conclusion, the SAR studies of a hit compound containing a
pyrrolopyrazolone skeleton identified from an in-house compound li-
brary led to the discovery of lead compound 9. A human P2X3
4

H. Tobinaga, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127636
Fig. 2. Docking studies of two pyrrolopyrazolone derivatives.
Table 4
Optimization of compound 9.
R2
R4
IC50
Solubility a
(μM)
CYP IC50 (μM)
CLt b
t1/2
BA b
(%)
b
Compound
(μM)
1A2
2C9
3A4
(ml/min/kg)
(hr)
9
t-Bu
Me
0.025
0.031
0.101
0.09
ND
> 20
> 20
> 20
> 20
> 20
> 20
> 20
10
> 20
> 20
> 20
> 20
16
16.6
24.5
79.2
94.9
NT
2.1
1.3
0.8
0.1
NT
1.5
1
33.8
NC
38
39
40
41
42
43
t-Bu
HO(CH2)2
MeSO2(CH2)2
HO(O)CCH2
Me
0.8
11
t-Bu
40.5
> 50
> 50
> 50
> 50
> 20
> 20
15
NC
t-Bu
NC
HO(CH2)2
HOCH2(CH3) 2
0.66
NT
C
Me
0.063
0.304
16
> 20
> 20
18.6
33.7
27.2
1.7
(HOCH2)2(CH3)C
Me
> 20
Table 5
Characterization of compound 42.
Compound
Analgesic testa
Rat pharmacokinetics
dose, po
(mg/kg)
efficacy
dose, iv
(mg/kg)
t1/2
CLt
dose, po
Cmax
AUC
BA
(% reversal)
(hr)
(ml/min/kg)
(mg/kg)
(ng/ml)
(μg·hr/ml)
(%)
9
100
30
30.5
41
10
10
3.2
1.8
12.6
14.9
100
30
450
6.48
4.9
42
1410
10.45
30.4
Pregabalin
10
20–30
a. Rat Seltzer model: Hyperalgesia (paw pressure test).
5

H. Tobinaga, et al.
Bioorganic&MedicinalChemistryLetters30(2020)127636
Scheme 3. Reagents and conditions for the synthesis of (S)-42.
analgesic effect by oral administration because of its improved potency
and PK profile.
4. Tsuda M, Shigemoto-Mogami Y, Koizumi S, Mizokoshi A, Kohsaka S, Salter, Michael
W, Inoue K. Nature, 2003, 424, 778.
5. Finger Thomas E, Danilova V, Barrows J, et al. Science. 2005;310:1495.
6. Labasi JM, Petrushova N, Donovan C, McCurdy S, Lira P, Payette MM, Brissette W,
Wicks JR, Audoly L, Gabel CA. J Immunol. 2002;168:6436.
Declaration of Competing Interest
7. Cocckayne DA, Hamilton SG, Zhu QM, et al. Nature. 2000;407:1011.
8. Jarvis MF, Burgard EC, McGaraughty S, et al. PNAS. 2002;99:17179.
9. Abdulgawi R, Dockry R, Holt K, et al. Lancet. 2015;385:1198.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
10. Tobinaga H, Kameyama T, Oohara M, et al. Bioorg Med Chem Lett. 2018;28:2338.
11. Gein VL, Kasimova NN. Russ J Org Chem. 2005;75:254.
12. Experimental procedure for preparation of compound 5 and 6: To a suspension of
compound 3 (92 mg, 2 mmol) in AcOH (2 mL), methylhydrazine (0.032 mL, 6 mmol)
was added and the mixture was stirred at 95 °C for 2 h. The reaction mixture was
quenched with water, and extracted with EtOAc. The organic layer was washed with
H2O and brine, dried over MgSO4, and concentrated under reduced pressure. The
residue was purified by washing with EtOAc/n-hexane to give 5 (41 mg, yield 44%)
as a yellow solid and 6 (10 mg, yield 11%) as a yellow solid. Compound 5; 1H-NMR
(CDCl3) δ: 0.44–1.01 (6H, m), 1.26–1.53 (4H, m), 1.70–1.79 (2H, m), 3.93 (7H, s), 5.
22 (1H, d, J = 3.0 Hz), 6.70 (1H, d, J = 1.8 Hz), 7.06 (2H, dt, J = 9.3, 2.4 Hz), 7.36
(2H, dt, J = 9.2, 2.4 Hz), 7.67–7.70 (2H, m), 7.89–7.93 (2H, m), 8.48 (1H, t, J = 1.4
Hz).; MS-ESI (m/z) = 469 [M+H]+.; Compound 6; 1H-NMR (CDCl3) δ: 0.52–1.05
(4H, m), 1.23–1.55 (6H, m), 1.86 (1H, d, J = 11.9 Hz), 3.89 (3H, s), 4.14 (3H, s), 5.
38 (1H, d, J = 2.9 Hz), 6.72 (1H, d, J = 1.7 Hz), 6.98–7.00 (2H, m), 7.63–7.66 (4H,
m), 7.93–7.96 (2H, m), 8.50 (1H, d, J = 1.5 Hz).; MS-ESI (m/z) = 469 [M+H]+.
13. Randall L, Selitto J. Arch Int. Pharmacodun Ther. 1957;4:409.
Acknowledgement
We thank Edanz Group (https://en-author-services.edanz-
group.com/ac) for editing a draft of this manuscript.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2020.127636.
14. Shir Y, Seltzer Z. Neurosci Lett. 1990;115:62.
References
15. Tobinaga H, Kameyama T, Asahi K, et al. Bioorg Med Chem Lett. 2019;29:688.
16. Docking study was carried out using the program Glide (Schrödinger, LLC, New York,
NY, USA).
1. North, Alan North RA. Physiol Rev. 2002;82:1013.
2. Gu JG, MacDermott AB. Nature. 1997;389:749.
17. See the supplementary data.
3. Mulryan K, Gitterman DP, Vial C, et al. Nature. 2000;403:86.
6