Discovery of triarylethanolamine inhibitors of the Kv1.5 potassium channel

Article abstract of DOI:10.1016/j.bmcl.2010.03.005

A series of triarylethanolamine inhibitors of the Kv1.5 potassium channel have been prepared and evaluated for their effects in vitro and in vivo. The structure-activity relationship (SAR) studies described herein led to the development of potent, selecti

Full text of DOI:10.1016/j.bmcl.2010.03.005

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2493–2496
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of triarylethanolamine inhibitors of the Kv1.5 potassium channel
a
a
a
a
Douglas C. Beshore a, , Nigel J. Liverton , Charles J. McIntyre , Christopher F. Claiborne , Brian Libby ,
J. Christopher Culberson f, Joseph J. Salata g, Christopher P. Regan b, Joseph J. Lynch b, Laszlo Kiss d,
Robert H. Spencer e, Stephanie A. Kane e, Rebecca B. White c, Suzie Yeh c, George D. Hartman a,
Christopher J. Dinsmore a
*
a Department of Medicinal Chemistry, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
b Department of Central Pharmacology, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
c Department of Drug Metabolism and Pharmacokinetics, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
d Department of In Vitro Sciences, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
e Department of Pain, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
f Department of Chemical Modeling and Informatics, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
g Department of Safety and Exploratory Pharmacology, Merck Research Laboratories, Merck & Co., Inc., PO Box 4, West Point, PA 19486, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of triarylethanolamine inhibitors of the Kv1.5 potassium channel have been prepared and eval-
uated for their effects in vitro and in vivo. The structure–activity relationship (SAR) studies described
herein led to the development of potent, selective and orally active inhibitors of Kv1.5.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 15 December 2009
Revised 26 February 2010
Accepted 1 March 2010
Available online 4 March 2010
Keywords:
Kv1.5
TAEA
Triarylethanolamine
Arrhythmia
Atrial fibrillation
The most common sustained cardiac arrhythmia is atrial fibril-
lation (AF), which afflicts ꢀ1% of the overall US population with
higher prevalence in the elderly population and occurs more often
in men than women.1–4 The associated morbidity, mortality, qual-
ity of life and notably increased risk of stroke associated with AF
patients present a significant unmet medical need.2 Current antiar-
rhythmic therapies inhibit multiple ion channels present in both
the ventricle and atrium and have significant limitations and ad-
verse effects, most significantly the potential for ventricular proar-
rhythmia, which has led to the pursuit of atrial selective therapies
that offer potentially safer treatment options for patients suffering
from AF.5,6
The ultra-rapid delayed rectifier current IKur, which is expressed
in human atrium and not the ventricle, has emerged as a promising
target for atrial selective therapy. The voltage gated Kv1.5 ion
channel underlies the native cardiac IKur current that is involved
in the repolarization of atrial action potential.7,8 Additionally, inhi-
bition of IKur has been shown to increase atrial refractory period
and restore normal sinus rhythm during arrhythmia, a critical
objective to the development of new, more effective and safer
treatments for AF, as recurrent arrhythmia leads to persistent AF
and atrial remodelling, a condition that is resistant to existing drug
therapies.5 As such, the development of selective inhibitors of
Kv1.5 offer a promising strategy for the termination and/or preven-
tion of AF.9 Indeed, several research groups have disclosed Kv1.5
blockers that have shown promise for the development of novel at-
rial selective antiarrhythmic agents.10 Herein, we describe the dis-
covery
of
orally
bioavailable,
potent
and
selective
triarylethanolamine inhibitors of the Kv1.5 potassium channel.11
Investigations at Merck led to the identification of diphenyl
phosphine oxide (DPO)12 inhibitors (1, Fig. 1), which have been
well characterized both in vitro and in vivo for their inhibitory
activity against Kv1.5.13,14 While studies of isoquinoline blockers
(i.e., ISQ-1) were ongoing,15 a structurally diverse series was
sought. Compound 1 is highly lipophilic (cLog P = 5.67) and an im-
proved lead was desired that possessed properties suitable for oral
dosing. Employing 2-D scaffold hopping, similarity searches and
screening approaches led to the identification of triarylethanol-
amine 2, which is potent blocker of Kv1.5 (IC50 = 204 nM). While
2 possesses improved properties relative to 1 (i.e., increased PSA
and reduced cLog P), further optimization of physiochemical prop-
* Corresponding author. Tel.: +1 215 652 7474; fax: +1 215 652 3971.
E-mail address: douglas_beshore@merck.com (D.C. Beshore).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.03.005

2494
D. C. Beshore et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2493–2496
Table 1
O
In vitro inhibitory activity of triarylethanolamines
Pri
P
O
N
Scaffold
Hopping
Me
N
O
HO
B
A
HO
B
2, 5a-i
1 (DPO-1)
2
a,b
c
hKv1.5 IC50 = 30 nM
PSA ( 2) = 15
Compound
A
B
Kv1.5 IC50 (nM)
hERG IC50 (nM)
hKv1.5 IC50 = 204 nM
PSA ( 2) = 30
5a
5b
2
5c
5d
5e
5f
5g
5h
5i
Ph
Ph
Ph
iPr
Ph
Me
iPr
Bn
nBu
nHex
H
>60,000
>60,000
204
3242
256
>60,000
2140
4946
2869
3357
>60,000
>60,000
>65,000
>30,000
>60,000
>60,000
75,000
15,190
36,000
27,500
cLogP = 5.67
cLogP = 4.40
Me
Ph
Ph
3-Pyridyl
3-Pyridyl
3-Pyridyl
3-Pyridyl
3-Pyridyl
3-Pyridyl
Figure 1. DPO-1 (1) and triaryethanolamine (2) Kv1.5 blockers.
erties were required to obtain a compound suitable for in vivo
characterization.
Compounds were prepared according to the general procedures
described in Scheme 1, in which complimentary routes were uti-
lized to survey variations of the aryl and amino groups. Alkylation
a
Determined according to Ref. 17. Potency values (IC50s) were determined using
a high-throughput patch clamp assay with 10-point dose response curves, n = 3–4
of substituted a-bromo esters 3 with morpholine afforded amines
4, which were treated with organolithium reagents to afford mor-
cells per point.
b
ISQ-1 was used as a positive control, see Ref. 15.
pholinoethanols 5a–i. An additional subset of compounds were
prepared employing methyl a-bromo(phenyl) acetate via SN2 dis-
placement with various amines to afford 6, which upon subsequent
treatment with 3-pyridyl lithium afforded compounds 7a–j.16
Compound potency was determined in a high-throughput patch
clamp (HT-clamp) assay in which CHO cells expressing the human
Kv1.5 channel were employed.17 Selectivity over the human ether-
a-go-go-related channel (hERG), which underlies the native cardiac
potassium current IKr, present in both human atrium and ventricle,
was of primary concern for an atrial selective therapy and was
monitored due to the known proarrhythmic potential of IKr block-
ers.18 Table 1 depicts SAR related to the triphenyl moiety of 2.
These studies revealed that both geminal aromatic B-rings were re-
quired (5a–b vs 2) for Kv1.5 activity. The phenyl B-ring was readily
substituted with 3-pyridyl moieties to afford 5d without a signifi-
cant loss in potency. Replacement of the phenyl A-ring (5c, e–i)
was met with some success, although longer aliphatic groups only
offered moderately potent blockers. Also, replacement of a phenyl
with a benzyl group (5g) was a poorly tolerated, resulting in a 20-
fold loss in potency. Relative to 2, compound 5d was equipotent,
exhibited higher PSA and had a decreased cLog P, which we postu-
lated would improve physiochemical properties (vide infra).
Optimization of the amino group was explored in the context of
the improved bispyridyl scaffold (Table 2). Removal of the morpho-
lino moiety (i.e., 7a), as well as replacement with an unsubstituted
amino group (i.e., 7b), proved to decrease Kv1.5 blockade. How-
ever, the introduction of small aliphatic substituents on the amino
group restored Kv1.5 potency (7c–e). Introduction of the 1,4-oxa-
zepane (7g) or the ring-opened analogue N-methyl-2-(methyl-
oxy)ethanamine (7f) was well-tolerated, but selectivity over
hERG was diminished (ꢀ7-fold) for both compounds. Furthermore,
the introduction of a piperazine (7h) completely eroded selectivity
versus hERG. Fortuitously, removal of the basic nitrogen from 7h
afforded piperidine 7i, which exhibited good Kv1.5 potency and re-
stored hERG selectivity. Incorporation of a pyrrolidine was also
well-tolerated, affording the even more selective blocker versus
hERG 7j.
With several potent Kv1.5 blockers identified and mindful that
the initial lead compound 2 exhibited a poor solubility profile, an
assessment of physiochemical properties suitable for oral dosing
was conducted. Table 3 depicts solubility data of selected Kv1.5
blockers 5d, 7i and 7j. Unlike the poorly soluble lead compound
2, compound 5d exhibited moderate solubility in aqueous buffered
solution. Substitution of the morpholine for the piperidine 7i led to
a 10-fold decrease in aqueous solubility. Interestingly, the pyrroli-
dine blocker 7j exhibited a threefold increase in solubility at pH
7.4, relative to 5d, and decreasing the pH further to 5.2 led to en-
hanced aqueous solubility of greater than 1 mg/mL, presumably
as a result of protonation of the basic pyrrolidine. Due to the en-
hanced physiochemical properties, in addition to the low protein
binding of 7j relative to 7e (25% bound vs 82%, respectively), 7j
was selected for further evaluation.
O
N
O
N
Br
a
b
B
MeO
MeO
A
A
A
HO
O
O
B
3
4
5a-i
C
C
d
A = Ph, c
MeO
N
Rat pharmacokinetic studies were carried out on triarylethanol-
amine (TAEA) 7j, which exhibited high clearance (59 mL/min/kg)
but a high volume of distribution (3.4 L/kg), leading to a half life
of 1.1 h. Relative to the initial lead 2, 7j possesses increased polar
surface area (46 Å2), good solubility and reduced cLog P (2.44),
making it an attractive candidate for in vivo profiling. Electrophys-
iological (EP) studies were carried out on surgically instrumented,
anesthetized rats,20 which were infused intravenously with 7j
(Fig. 2). Various cardiac parameters were recorded, including atrial
HO
N
O
6
7a-j
Scheme 1. General preparation of Kv1.5 blockers. Reagents and conditions: (a)
morpholine, triethylamine, acetonitrile; (b) organolithium, tetrahydrofuran, À78 to
0 °C; (c) amine, triethylamine, acetonitrile; (d) 3-pyridyl lithium, tetrahydrofuran,
À78 to 0 °C.

D. C. Beshore et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2493–2496
2495
Table 2
100
%ARP
%VRP
%AVRP
In vitro inhibitory activity of triarylethanolamines
C
80
N
5d, 7a-j
HO
60
40
20
0
N
a,b
c
Compound
C
Kv1.5 IC50 (nM)
hERG IC50 (nM)
O
N
5d
256
>60,000
185
651
1777
4898
9618
Rat EP Plasma Level of 7j (nM)
7a
H
NH2
7107
16,390
>60,000
>60,000
7bd
Figure 2. Rat EP profile of TAEA 7j following 20 min infusions at 0.01, 0.03, 0.06, 0.2
and 0.6 mpk/min.
Me
Me
N
7c
7d
7e
850
2222
234
>60,000
15,000
N
curred without significantly affecting atrio-ventricular conduction,
as there were no changes observed in AV nodal refractoriness
(AVRP). At the highest dose tested, 7j exhibited a 79% increase in
Et
Et
N
>60,000
ARP and a 63% increase in VRP at plasma levels of <10 lM. These
OMe
results are superior to those previously reported for rat EP experi-
ments that evaluated isoquinoline-derived Kv1.5 blockers (see Ref.
15). The superiority of 7j is reflected in the magnitude of percent
change in ARP and VRP achieved at ꢀ3-fold lower plasma concen-
trations, which is likely due the large free fraction of 7j (hFF = 25%).
In a canine heart failure model, 7j also terminated pacing-induced
AF and induced significant prolongation of ARP but notably had no
effect on VRP at the highest dose tested, substantiating the atrial
selectivity associated with inhibition of the Kv1.5 channel.21 Thus,
these findings are consistent with the profile of an IKur selective
compound.
7f
583
517
3450
3550
Me
N
O
7g
N
H
N
7h
7935
2300
N
During the evaluation of 7j in additional rodent and canine
models, dose-dependent transient adverse effects (AEs) were ob-
served at exposures similar to those eliciting EP effects in the
aforementioned rat model, including shaking, tremors, twitching
and convulsions. Studies were carried out to determine if these
AEs were derived from neurological activity of the compound. In
vitro experiments revealed that compound 7j possesses good pas-
sive cell permeability and is not a rat P-gp substrate (Table 4).
Brain penetration of 7j was also evaluated in rats after a 1 h intra-
venous infusion in tail vein and showed that 7j exhibits good brain
penetration with a brain/plasma (B/P) ratio of 0.58. In addition, rat
EEG studies were carried out, revealing abnormally high frequency
contributions from 40 to 50 Hz upon administration of 7j (data not
shown), suggesting a pharmacological CNS effect on Kv1.x chan-
nels due to presence of the compound in brain.
7i
7j
206
284
14,000
20,000
N
N
a
Determined according to Ref. 17. Potency values (IC50s) were determined in a
high-throughput patch clamp assay from 10-point dose response curves, n = 3–4
cells per point.
b
ISQ-1 was used as a positive control, see Ref. 15.
See Ref. 19.
c
Table 3
Aqueous solubility of selected triarylethanolamines
The brain penetration and EEG studies performed suggest that
the AEs observed with 7j are neurologically related. Furthermore,
voltage gated potassium channels are known to be expressed in
the CNS and both channel openers and blockers have been targeted
Compound
Solubilitya
pHb
5d
7i
7j
7j
0.043
0.0054
0.11
7.2
7.4
7.4
5.2
1.4
a
Table 4
Solubility is expressed in mg/mL in aqueous buffer at listed pH. All compounds
Brain penetration, passive permeability and P-gp efflux of compound 7j
tested were crystalline by microscopic examination under plain polarized light;
however the specific crystalline form was not determined.
Compound [Plasma]a,b
M)
[Brain]a,b
M)
B/P
%
FF
Pappd rP-
gpe
b
pH of buffered aqueous solution.
(l
(
l
ratioc
7j
23.4
13.7
0.58
25
36
1.2
a
Brain and plasma levels were determined following a 1 h tail vein iv infusion,
g/min kg.
n = 3 and is expressed as an average.
refractory period (ARP) and ventricular refractory period (VRP). Be-
cause IKur is functionally expressed in both the atria and ventricular
cardiac tissues of rats, this model is restricted only to the assess-
ment of efficacy and cannot address atrial selectivity. Compound
7j exhibited a dose-dependent increase in both ARP and VRP.
Importantly, these robust effects on cardiac repolarization oc-
435
l
b
c
d
e
B/P = [Brain]/[Plasma] expressed as a ratio.
Passive permeability is expressed as 10À6 cm/s.
Rat P-gp efflux values were determined in CHO cell lines over expressing rat P-
gp (mdr1a). Values shown are a ratio of (B to A)/(A to B).

2496
D. C. Beshore et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2493–2496
as potential therapeutic agents for CNS diseases.22–24 Therefore,
the development of a peripherally restricted Kv1.5 antagonist,
which minimizes the potential for central nervous system expo-
sure and inhibition of potassium channels in brain, is anticipated
to afford an improved safety profile. Efforts aimed at addressing
this hypothesis have been pursued25 and will be the subject of fu-
ture publications.
In conclusion, similarity searching of DPO inhibitors led to the
identification of a class of triarylethanolamine Kv1.5 antagonists.
Poor physical properties of the initial lead were overcome by
replacement of the bisphenyl alcohol with a bispyridyl alcohol
moiety and these studies led to the identification of TAEA 7j, a po-
tent inhibitor of Kv1.5 channels with desirable physical properties.
TAEA 7j exhibited robust electrophysiological effects on atrial
repolarization but not atrioventricular conduction (AVRP) in vivo
in rats and was found to be atrial selective in dogs. During the
course of in vivo studies, neurological side effects were observed
with TAEA 7j. Subsequent studies suggested that the optimization
of a peripherally restricted and selective Kv1.5 blocker that pos-
sesses good physical properties, robust electrophysiological effects
and reduced central penetration may provide an efficacious agent
with an acceptable safety profile.
4. Estrada, J. C.; Darbar, D. Eur. J. Clin. Pharmacol. 2008, 64, 1139.
5. Tamargo, J.; Caballero, R.; Gómez, R.; Delpón, E. Expert Opin. Investig. Drugs
2009, 18, 399.
6. Goldstein, R. N.; Stambler, B. S. Prog. Cardiovasc. Dis. 2005, 48, 193.
7. Fedida, D.; Wible, B.; Li, G.-R.; Wang, Z.; Nattel, S. Circ. Res. 1997, 80, 572.
8. Wang, Z.; Fermini, B.; Nattel, S. Circ. Res. 1993, 73, 1061.
9. Yang, Q.; Wang, X.; Du, L.; Li, M.; You, Q. Expert Opin. Ther. Patents 2007, 17,
1443.
10. Bilodeau, M. T.; Trotter, B. W. Curr. Topics Med. Chem. 2009, 9, 436.
11. Kv1.5 blockers described herein were disclosed in: Dinsmore, C. J.; Bergman, J.
M.; Beshore, D. C.; Trotter, B. W.; Nanda, K.; Isaacs, R.; Payne, L. S.; Neilson, L.
A.; Wu, Z.; Bilodeau, M. T.; Manley, P. J.; Balitza, A. E. WO2006015159A2, 2006.
12. Lagrutta, A.; Wang, J.; Fermini, B.; Salata, J. J. J. Pharmacol. Exp. Ther. 2006, 317,
1054.
13. Regan, C. P.; Cresswell, H. K.; Zhang, R.; Lynch, J. J. J. Cardiovasc. Pharmacol.
2005, 46, 68.
14. Regan, C. P.; Wallace, A. A.; Cresswell, H. K.; Atkins, C. L.; Lynch, J. J. J.
Pharmocol. Exp. Ther. 2006, 316, 727.
15. Trotter, B. W.; Nanda, K. K.; Kett, N. R.; Regan, C. P.; Lynch, J. J.; Stump, G. L.;
Kiss, L.; Wang, J.; Spencer, R. H.; Kane, S. A.; White, R. B.; Zhang, R.; Anderson, K.
D.; Liverton, N. J.; McIntyre, C. J.; Beshore, D. C.; Hartman, G. D.; Dinsmore, C. J.
J. Med. Chem. 2006, 49, 6954.
16. Data reported herein are for racemic triarylethanolamine Kv1.5 blockers.
17. Zheng, W.; Spencer, R. H.; Kiss, L. Assay Drug Dev. Technol. 2004, 2, 543.
18. McBride, B. F. J. Clin. Pharmacol. 2009, 49, 258.
19. Raab, C. E.; Butcher, J. W.; Connolly, T. M.; Karezewshki, J.; Yu, N. X.;
Staskiewicz, S. J.; Liveton, N.; Dean, D. C.; Melillo, D. G. Bioorg. Med. Chem. Lett.
2006, 16, 1692.
20. Regan, C. P.; Cresswell, H. K.; Zhang, R.; Lynch, J. J., Jr. J. Cardiovasc. Pharmacol.
2005, 46, 68.
21. Regan, C. P.; Kiss, L.; Stump, G. L.; McIntyre, C. J.; Beshore, D. C.; Liverton, N. J.;
Dinsmore, C. J.; Lynch, J. J., Jr. J. Pharmocol. Exp. Ther. 2009, 324, 322.
22. Judge, S. I. V.; Smith, P. J.; Stewart, P. E.; Bever, C. T., Jr. Recent Patents CNS Drug
Discov. 2007, 2, 200.
23. Brooke, R. E.; Atkinson, L.; Edwards, I.; Parson, S. H.; Deuchars, J. Brain Res.
2006, 1070, 101.
Acknowledgement
The authors would like to thank Dr. Richard C. Isaacs for prep-
aration of compound 7b and Kevin Nyugen for technical assistance.
24. Devaux, J.; Gola, M.; Jacquet, G.; Crest, M. J. Neurophysiol. 2002, 87, 1376.
25. See the following reference for peripherally restricted Kv1.5 antagonists:
Nanda, K. K.; Nolt, M. B.; Cato, M. J.; Kane, S. A.; Kiss, L.; Spencer, R. H.; Wang, J.;
Lynch, J. J.; Regan, C. P.; Stump, G. L.; Li, B.; White, R.; Yeh, S.; Bogusky, M. J.;
Bilodeau, M. T.; Dinsmore, C. J.; Lindsley, C. W.; Hartman, G. D.; Wolkenberg, S.
E.; Trotter, B. W. Bioorg. Med. Chem. Lett. 2006, 16, 5897.
References and notes
1. Valderrama, A. L.; Dunbar, S. B.; Mensah, G. A. Am. J. Prev. Med. 2005, 29, 75.
2. Waldo, A. L. Am. Heart J. 2006, 151, 771.
3. Chugh, S. S.; Blackshear, J. L.; Shen, W.-K.; Hammill, S. C.; Gersh, B. J. J. Am. Coll.
Cardiol. 2001, 37, 371.