3-Trifluoromethyl-4-nitro-5-arylpyrazoles are novel KATP channel agonists

Article abstract of DOI:10.1016/j.bmcl.2003.10.066

This communication describes the discovery and synthesis of a series of 3-trifluoromethyl-4-nitro-5-arylpyrazoles as potent K_ATP channel agonists. The most potent compound reported is ca. 100-fold more potent than diazoxide and exhibits selectivity for the SUR1 K_ATP channel subtype. The 4-nitro substitutent on the pyrazole ring was required for activity, and limited SAR suggests that the de-protonated pyrazole maybe the active species.

Full text of DOI:10.1016/j.bmcl.2003.10.066

Bioorganic & Medicinal Chemistry Letters 14 (2004) 813–816
3-Trifluoromethyl-4-nitro-5-arylpyrazoles are novel KATP
channel agonists
Andrew J. Peat, Claire Townsend, M. Craig McKay, Dulce Garrido,
Christopher M. Terry, Jayme L. R. Wilson and Stephen A. Thomson*
GlaxoSmithKline Research & Development, 5 Moore Drive, Research Triangle Park, NC 27709, USA
Received 11 July 2003; accepted 15 October 2003
Abstract—This communication describes the discovery and synthesis of a series of 3-trifluoromethyl-4-nitro-5-arylpyrazoles as
potent KATP channel agonists. The most potent compound reported is ca. 100-fold more potent than diazoxide and exhibits selec-
tivity for the SUR1 KATP channel subtype. The 4-nitro substitutent on the pyrazole ring was required for activity, and limited SAR
suggests that the de-protonated pyrazole maybe the active species.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
muscle, comprised primarily of SUR2B/Kir6.1,10 as well
Potassium channels represent a subfamily of ion chan-
nels that selectively permit transport of potassium ions
across the cellular membrane. A subset of this class of
channels are the ATP-sensitive potassium channels
(KATP), which have been identified in a number of tis-
sues such as pancreatic b-cells,1 smooth muscle,2 cardiac
muscle,3 kidney,4 and both peripheral and central neu-
rons.5 The KATP channels present in pancreatic b-cells
serve to regulate glucose-mediated insulin secretion by
coupling cellular glucose metabolism directly to mem-
brane potential.6 In this manner, b-cell KATP channels
link plasma glucose levels (energy) to intracellular
[Ca2+] and ultimately insulin secretion.7
as channels found in the heart, consisting of SUR2A/
Kir6.2.11,12 Diazoxide activates the b-cell KATP chan-
nels thereby inhibiting pancreatic insulin release and
is used by physicians to treat severe cases of hyper-
insulinemia.13 In addition, diazoxide has also shown
beneficial results as a treatment for diseases asso-
ciated with excessive insulin levels such as diabetes14
and obesity.15 However, the use of diazoxide is lim-
ited by side effects such as severe oedema,16 and
excessive hair growth,17 which may be related to dia-
zoxide’s vasodilatory effects via activity at SUR2B/
Kir6.1.18
To date, there are only a few reports of potent SUR1/
Kir6.2 (b-cell KATP) agonists. Modifications to diaz-
oxide have been reported which result in increased
potency and selectivity for SUR1/Kir6.2.19 Recently a
series of potent and selective diazoxide analogues were
described with potencies in the low nano-molar range.20
Our desire was to find novel structures, distinct from the
diazoxide template, that would serve as b-cell KATP
channel agonists. In this effort, we used a high
throughput screening assay, based on changes in mem-
brane potential.21 This resulted in the discovery of
compound 1. Since compound 1 represented a novel
class of KATP channel agonist we were motivated to
investigate. Herein, we report the activity of this new
series of SUR1/Kir6.2 agonists and a limited SAR
(Fig. 1).
The b-cell KATP channel is comprised of two sub-units,
the sulfonylurea receptor-1 (SUR1) and the potassium
selective inward rectifier (Kir6.2).8 The SUR1 protein is
the molecular target for the sulfonylurea class of anti-
hyperglycemic drugs such as glipizide, which have been
widely used in the treatment of Type 2 diabetes melli-
tus.9 These compounds are antagonists of the b-cell
KATP channel and promote insulin secretion. In addi-
tion, a number of drugs act as agonists of KATP chan-
nels. Compounds such as cromakalim and pinacidil
activate KATP channels found in vascular smooth
* Corresponding author. Fax: +1-919-483-6053; e-mail: stephen.a.
thomson@gsk.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.10.066

814
A. J. Peat et al. / Bioorg. Med. Chem. Lett. 14 (2004) 813–816
Figure 1.
2. Chemistry
Our initial goal was to systematically remove the nitro
groups on 1 to determine the contribution of each to
the activity. The 3-trifluoromethyl-5-phenylpyrazole
core, 2, was prepared by condensation of the available
4,4,4-trifluoro-1-phenylbutane-1,3-dione with hydrazine
mono-hydrate. As shown in Scheme 1, various nitra-
tion procedures were employed to give the desired
nitrated products. Nitration at 0 ꢀC gave a ca. 9:1
mixture of para and meta nitration products on the
phenyl ring. Reaction at room temperature revealed the
second nitration site to be the 4-position of the pyrazole
ring, giving 4 in good yield. More vigorous conditions
(100 ꢀC) were required to achieve the third nitration in
the synthesis of the original lead compound (1). For
selective nitration on the pyrazole ring we turned to a
two step method developed by Janssen.22 Treatment of
2 with a preformed mixture of nitric acid and acetic
anyhdride (CAUTION; pre-mix at 0 ꢀC slowly) led to
clean N-nitration of the pyrazole. Subsequent thermal
rearrangement (chlorobenzene, 115 ꢀC) gave the
4-nitropyrazole product 5.
Scheme 2. Reagents and conditions: (i) NH2NHTs, EtOH, TsOH–
H2O, rt, 91%; (ii) 2 equiv NaHMDS, À78 ꢀC, ArCHO, rt 1 h, reflux 3
h; (iii) HNO3, Ac2O premix at 0 ꢀC, HOAc, 100%; (iv) PhCl, 115 ꢀC.
with two equivalents of NaHMDS follow by the required
aldehyde, and subsequent heating effected the two-step
process and gave the pyrazoles 8 or 9 in good yield (65
and 60%, respectively). The procedure proved to be
quite general and a number of various benzaldehydes
could be employed. The pyrazoles were nitrated at the
4-position using the two step procedure described earlier
to give compounds 10 or 11.
To determine the contribution of the pyrazole NH to
the activity, the corresponding isoxazole 12 was pre-
pared via treatment of 4,4,4-trifluoro-1-phenylbutane-
1,3-dione with hydroxylamine, and subsequent nitration
using conditions similar to that for compound 1.27
3. Results and discussion
Replacement of the 2,4-dinitrophenyl group of 1 with
other electron deficient phenyl groups is shown in
Scheme 2. We desired a method which would allow for
the use of readily available starting materials and quick
construction of the pyrazole ring. The one-pot procedure
reported by Santagostino23 fulfilled these requirements.
The diethyl phosphonate 6 was prepared from diethyl
methylphosphonate and ethyl trifluoroacetate as descri-
bed in the literature.24,25 Treatment of 6 with tosylhy-
drazine in ethanol with a catalytic amount of TsOH
afforded the hydrazone 7 in good yield.26 Treatment of 7
Previous investigators of KATP channel agonists have
relied on either inhibition of insulin secretion from islets
(or other insulin secreting cells) or whole cell Rb+ efflux
methods to quantitatively assess agonist activity. Here
we desired a more direct measure of channel activity
and therefore we determined compound potency and
efficacy using the excised patch-clamp technique, in
which a portion of the cellular membrane containing a
functioning KATP channel is removed from the rest of the
cell. In these studies ATP-sensitive K+ currents were
Scheme 1. Reagents and conditions: (i) NH2NH2–H2O, EtOH, 81%; (ii) HNO3, H2SO4, 0 ꢀC, 72%; (iii) HNO3, H2SO4, rt, 65%; (iv) HNO3, H2SO4,
100 ꢀC, 90%; (v) HNO3, Ac2O premix at 0 ꢀC, HOAc, 100%; (vi) PhCl, 115 ꢀC, 78%.

A. J. Peat et al. / Bioorg. Med. Chem. Lett. 14 (2004) 813–816
815
Figure 3. Dose–response curves for 1 versus KATP subtypes from
excised patch recordings.
Figure 2. Compound dose–response curves for activation of KATP
(SUR1/Kir6.2) from excised patch recordings.
recorded from inside-out macropatches pulled from
Xenopus oocytes expressing the human SUR1/Kir6.2
channel.28 This technique enables the measurement of
direct effects of the compound at the channel itself and
thus provides an unambiguous determination of the
on the pyrazole ring. Comparison of compound 4 with
compound 3 reveals that removal of the pyrazole 4-nitro
group results in a complete loss of activity. Activity is
also dependent on the nature of the substitutents on the
phenyl ring. Removal of the 2-nitro group on the phenyl
ring results in a 10-fold loss of activity (1 vs 4) and
removal of both phenyl nitro groups gives a compound
(5) with minimal agonist activity. Activity can be
restored by incorporation of other electron withdrawing
groups on the phenyl ring as exemplified by compounds
10 and 11. Within this limited compound set the elec-
tron withdrawing ability of the phenyl substitutents
appears to correlate with potency [2,4-(NO2)2>2,4-
(CF3)3>4-NO2>2,4-F2> >H]. These initial results sug-
gest that the ionization of the pyrazole N–H is critical to
agonist activity. We determined the pKa of 1 to be 4.78
(Æ0.02) using standard UV spectroscopic methods, which
suggests the deprotonated species is the active agent at
physiological pH. This hypothesis is also supported by the
lack of activity of the oxazole analogue (12).
activity of a compound at KATP
.
The relationship between the dose of a compound and
KATP channel activity recorded from excised patches is
illustrated in Figure 2 and calculated EC50 values are
listed in Table 1. All compounds were tested at least
three times over an appropriate range of four to five
doses to determine EC50 values. The percent max refers
to the percentage of maximal agonistic response elicited
by the compound as compared to a fully efficacious
effect of diazoxide (50 mM).
The lead compound 1 is ca. 100-fold more potent than
diazoxide and is comparable to the potent diazoxide
analogues reported by Nielsen.20 The potency of this
template is highly dependent on the 4-nitro substitutent
We also examined the selectivity of the lead compound
1 for the SUR1/Kir6.2 channel versus other KATP
channel subtypes. Shown in Figure 3 are the compound
1 dose–response curves for the three common KATP
subtypes (SUR1, SUR2A, and SUR2B). The calculated
EC50 values for 1 are 0.07, 12.30 and 2.74 micro-molar
for SUR1, SUR2A, and SUR2B, respectively. Under
the same assay conditions diazoxide was 2-fold selective
for SUR1 over SUR2A and SUR2B (data not shown).
Table 1. Compound activation of KATP (SUR1/Kir6.2) from
excised patch recordings
Compd
X
R1
R2
EC50, mM a,b
%Maxa,c
Diazoxide
1
3
4
5
10
11
12
8.80 (Æ0.25)
0.07 (Æ0.01)
>30
100
122 (Æ3)
NH NO2
NH
NH NO2
NH NO2
NH NO2
NH NO2
2,4-(NO2)2
4-NO2
4-NO2
H
2,4-F2
2,4-(CF3)2
4-NO2
4. Conclusion
H
8 (Æ4)d
0.70 (Æ0.21)
94 (Æ9)
38 (Æ3)d
113 (Æ12)
197 (Æ17)
In conclusion, we have discovered a new class of SUR1/
Kir6.2 (KATP) channel agonists, exemplified by com-
pound 1. The lead compound in this series is 100-fold more
potent than diazoxide, and shows significant selectivity
for the SUR1 KATP subtype. The limited SAR suggests
that the deprotonated pyrazole maybe the active form.
Further studies will expand the SAR of this series as well
as investigate the SUR structural binding requirements.
>30
1.15 (Æ0.16)
0.54 (Æ0.15)
>30
O
NO2
a Values are means from 3–6 experiments.
b Concentration which elicits 50% of the maximal response.
c Percent maximal agonist response as compared to 50 mM diazoxide.
d Maximal response at 30 mM.

816
A. J. Peat et al. / Bioorg. Med. Chem. Lett. 14 (2004) 813–816
17. Uno, H.; Kemnitz, J. W.; Cappas, A.; Adachi, K.;
References and notes
Sakuma, A.; Kamoda, H. J. Dermatol. Sci. 1990, 1, 183.
18. Buhl, A. E.; Waldon, D. J.; Conrad, S. J.; Mulholland,
MlJ.; Shull, K. L.; Kubicek, M. F.; Johnson, G. A.;
Brunden, M. N.; Stefanski, K. J.; Stehle, R. G. J. Invest.
Dermatol. 1992, 98, 315.
1. Cook, D. L.; Hales, C. N. Nature 1984, 311, 271.
2. Standen, N. B.; Quayle, J. M.; Davies, N. W.; Brayden,
J. E.; Huang, Y.; Nelson, M. T. Science 1989, 245, 177.
3. Noma, A. Nature 1983, 305, 147.
4. Quast, U. Naunyn-Schmiedebergs Arch. Pharmacol. 1996,
354, 213.
5. Bernadi, H.; Fosset, M.; Lazdunski, M. Proc. Natl. Acad.
Sci. U.S.A. 1988, 85, 9816.
6. Aguilar-Bryan, L.; Bryan, J.; Nakazaki, M. Recent Prog.
Horm. Res. 2001, 56, 47.
7. Wollheim, C. B.; Sharp, G. W. G. Physiol. Rev. 1981, 61,
914.
8. (a) For recent reviews, see: Bryan, J.; Aguilar-Bryan, L.
Biochim. Biophys. Acta 1999, 1461, 285. (b) Seino, S.;
Miki, T.; Yano, H. Endocrine Updates 2001, 11, 91. (c)
Ashcroft, S. J. H. J. Membr. Biol. 2000, 176, 187.
9. Gribble, F. M.; Ashcroft, F. M. Metab., Clin. Exp. 2000,
49, 3.
10. Newgreen, D. T.; Bray, K. M.; McHarg, A. D.; Weston,
A. H.; Duty, S.; Brown, B. S.; Kay, P. B.; Edwards, G.;
Longmore, J.; Southerton, J. S. Br. J. Pharmacol. 1990,
100, 605.
11. Escande, D.; Thuringer, D.; Leguern, S.; Cavero, I. Bio-
chem. Biophys. Res. Commun. 1988, 154, 620.
12. Yokoshiki, H.; Sunagawa, M.; Seki, T.; Sperelakis, N.
Am. J. Physiol. 1998, 274, C25.
13. Garrino, M. G.; Plant, T. D.; Henquin, J. C. Br. J. Phar-
macol. 1989, 98, 957.
19. (a) Lebrun, P.; Arkhammar, P.; Antoine, Q. A.; Hansen,
J. B.; Pirotte, B. Diabetologia 2000, 43, 723. (b) De Tullio,
P.; Pirotte, B.; Lebrun, P.; Fontaine, J.; Dupont, L.;
Antoine, M.-H.; Ouedraogo, R.; Khelili, S.; Maggetto, C.;
Masereel, B.; Diouf, O.; Podona, T.; Delarge, J. J. Med.
Chem. 1996, 39, 937. (c) Peat, A. J.; Townsend, C.; Wor-
ley, J. F., III; Allen, S. H.; Garrido, D.; Mertz, R. J.;
Pfohl, J. L.; Terry, C. M.; Truax, J. F.; Veasey, R. L.;
Thomson, S. A. Bioorg. Med. Chem. Lett. 2002, 12, 2977.
20. Nielsen, F. E.; Bodvarsdottir, T. B.; Worsaae, A.; MacKay,
P.; Stidsen, C. E.; Boonen, H. C. M.; Pridal, L.; Arkham-
mar, P. O. G.; Wahl, P.; Ynddal, L.; Junager, F.; Dragsted,
N.; Tagmose, T. M.; Mogensen, J. P.; Koch, A.; Treppen-
dahl, S. P.; Hansen, J. B. J. Med. Chem. 2002, 45, 4171.
21. Pfohl, J. L.; Worley, J. F., III; Condreay, J. P.; An, G.;
Apolito, C. J.; Kost, T. A.; Truax, J. F. Receptors Chan-
nels 2002, 8, 99.
22. Janssen, J. W. A. M.; Koeners, H. J.; Habraken, C. L. J.
Org. Chem. 1973, 38, 1777.
23. Almirante, N.; Cerri, A.; Fedrizzi, G.; Marazzi, G.; San-
tagostino, M. Tetrahedron Lett. 1998, 39, 3287.
24. Compound 6 was isolated as a mixture of keto and
hydrate forms.
25. Nickson, T. E. J. Org. Chem. 1988, 53, 3870.
26. Compound 7 contains ca. 5% of the difluoro analogue.
Use of 0.1% aqueous HCl gave ca. 35% of the difluoro
analogue.
27. All compounds gave spectroscopic data (1H NMR and
LC/MS) consistent with the reported structures and com-
pound purity was determined to be >95%. Compounds 2
and 5 are available commercially.
14. (a) Greenwood, R. H.; Mahler, R. F.; Hales, C. N. Lancet
1976, 28, 444. (b) Bjork, E.; Berne, C.; Kampe, O.; Wibell,
L.; Oskarsson, P.; Karlsson, F. A. Diabetes 1996, 45,
1427.
15. Alemzadeh, R.; Langley, G.; Upchurch, L.; Smith, P.;
Slonim, A. E. J. Clin. Endocrinol. Metab. 1998, 83, 1911.
16. (a) Honey, M.; Cotter, L.; Davies, N.; Denison, D.
THORAX 1980, 35, 269. (b) Falk, J. L.; Tang, M. Physiol.
Behavior 1972, 9, 259.
28. Gribble, F. M.; Ashfield, R.; Ammala, C.; Ashcroft, F. M.
J. Physiol. 1997, 487, 87.