Potent antagonists of the Kv1.5 potassium channel: Synthesis and evaluation of analogous N,N-diisopropyl-2-(pyridine-3-yl)acetamides

Article abstract of DOI:10.1016/j.bmcl.2006.08.054

This letter describes the discovery of a novel series of potent Kv1.5 ion channel antagonists based on a diisopropyl amide scaffold. Structure-activity relationships of functionalized analogs are discussed. Key compound 1-(3-(diisopropylcarbamoyl)-2-pheny

Full text of DOI:10.1016/j.bmcl.2006.08.054

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5897–5901
Potent antagonists of the Kv1.5 potassium channel: Synthesis
and evaluation of analogous N,N-diisopropyl-2-
(pyridine-3-yl)acetamides
Kausik K. Nanda,a, M. Brad Nolt,a,*, Matthew J. Cato,b Stefanie A. Kane,b
Laszlo Kiss,b Robert H. Spencer,b Jixin Wang,b Joseph J. Lynch,c Christopher P. Regan,c
Gary L. Stump,c Bing Li,d Rebecca White,d Suzie Yeh,d Michael J. Bogusky,a
Mark T. Bilodeau,a Christopher J. Dinsmore,a Craig W. Lindsley,a George D. Hartman,a
Scott E. Wolkenberga and B. Wesley Trottera
aDepartment of Medicinal Chemistry, Merck Research Laboratories, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
bDepartment of Molecular Pharmacology, Merck Research Laboratories, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
cDepartment of Pharmacology, Merck Research Laboratories, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
dDepartment of Drug Metabolism, Merck Research Laboratories, Merck and Co., Inc., PO Box 4, West Point, PA 19486, USA
Received 10 July 2006; revised 8 August 2006; accepted 10 August 2006
Available online 1 September 2006
Abstract—This letter describes the discovery of a novel series of potent Kv1.5 ion channel antagonists based on a diisopropyl amide
scaffold. Structure–activity relationships of functionalized analogs are discussed. Key compound 1-(3-(diisopropylcarbamoyl)-2-
phenyl-3-(pyridin-3-yl)propyl)-3-(2-fluorobenzyl)urea (10) exhibits significant atrial-selective effects in an in vivo model.
Ó 2006 Elsevier Ltd. All rights reserved.
Atrial fibrillation (AF) is the most common sustained
cardiac arrhythmia in clinical practice, and is implicated
in approximately 15% (>80,000 per year) of all strokes
in the United States.1 Intervention with non-selective
ion channel blockers can be effective in treating and pre-
venting arrhythmias, but at present, available drug ther-
apies carry a significant risk of potentially lethal
ventricular proarrhythmia.2 In an effort to minimize
the risk of proarrhythmia, more selective treatments
are being pursued which do not act on the ventricle.
Herein, we describe our efforts to identify a selective
antagonist of Kv1.5, the molecular ion channel correlate
of IKur, that delays atrial repolarization in vivo. Other
members of the Kv1.x subfamily of potassium channels
are expressed in the central nervous system (CNS),
where their function is believed to be unrelated to cardi-
ac currents. In an effort to reduce brain exposure and
minimize the possibility of antagonizing CNS Kv1.x
channels, we sought inhibitors which were substrates
for human P-glycoprotein (P-gp), the most well-charac-
terized transporter that pumps xenobiotics out of the
CNS across the blood/brain barrier.4
A promising therapeutic approach is to selectively block
the ultra-rapid delayed rectifier potassium ion current
(IKur), thereby delaying atrial repolarization and con-
verting AF to normal sinus rhythm. The IKur current
has been observed in human atrial, but not ventricular
myocytes. This difference presents the possibility of
selectively targeting atria.3
High-throughput screening of the Merck sample col-
lection identified methyl 4-(diisopropylcarbamoyl)-3,4-
diphenylbutanoate (1)5 as
a potent inhibitor of
Kv1.5 (IC50 = 252 nM),6 although in vivo clearance
was high in dog (43 ml/min/kg). This compound is
structurally distinct from two recently disclosed cardi-
ac ion channel antagonists currently under investiga-
tion (Fig. 1).7
Keywords: Arrhythmia; Atrial fibrillation; Potassium channel; Kv1.5;
IKur.
*
The general synthesis of compounds in the diisopro-
pyl amide series is represented in Scheme 1. After
Corresponding author. E-mail: brad_nolt@merck.com
These authors contributed equally to this work.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.054

5898
K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5897–5901
R1
H
R1
H
N
N
O
O
N
O
O
N
O
O
H
H
R2
R2
OH
RSD-1235 (Cardiome)
O
O
4a
5
Kv1.5 IC50 = 252.2 nM
1
N
N
H
N
H
H
MeO
N
N
O
O
O
O
N
N
H
H
H
MeO2C
MeO2C
AVE0118 (Aventis)
4a
5
Figure 1. Cardiac ion channel antagonists.
Figure 2. Relative stereochemistry of 1,4-addition reaction products;
arrows indicate diagnostic NOE correlations. R1 = diisopropyl amine,
R2 = CO2CH3. The reaction temperature for the 1,4-addition was
À78 °C for 4a and 0 °C for structure 5.
straightforward preparation of 3 via an amide coupling
of an arylacetic acid with diisopropyl amine, deprotona-
tion and treatment with b-aryl vinyl esters generated 1,4-
addition products 4. Competing 1,2-addition products
were not observed.8
idine was tolerated (see 4a and 4b), and incorporation of
other pyridine isomers resulted in a ca. 20-fold decrease
in potency (4c–4f). Compounds containing two pyridine
rings, including those in which one 3-pyridine ring is re-
tained, lost potency versus the 3-pyridyl, phenyl analogs
(compare 4a and 4g/4h). Substituent effects were also
briefly investigated: 3-bromophenyl analogs 4i and 4k
exhibited 2-fold greater potency than the parent phenyls
4a and 4b. The corresponding nitriles, 4j and 4l, were
more potent and less potent, respectively, than 4a and
4b. The 4-cyanophenyl analog 4m was slightly more ac-
tive than the parent phenyl, and the 2-cyano substituent
(4n) was not well tolerated. Additionally, 2-biaryls such
as 4o exhibit improved potency versus Kv1.5.
It is noteworthy that the 1,4-addition exhibits tempera-
ture-dependent diastereoselectivity in the absence of
additional cation solvating agents.9 At À78 °C, a single
racemic diastereomer of 4 is generated while at higher
temperatures (0 °C) the alternative isomer is formed
selectively. It has been reported that E-enolates give rise
to syn conjugate addition products, while the Z-enolates
give anti conjugate addition products.10,11 Assignment
of relative stereochemistry was carried out spectroscop-
ically for 4a and 5 (Fig. 2) on the basis of vicinal benzyl-
ic proton coupling constants and ROESY correlations.
The selective 1,4-addition provided a convenient route
to the diastereomeric racemates 4a and 5, and high selec-
tivity was also observed across a range of arylacetamide
and cinnamate substrates. Generally, isomers formed at
low temperature showed increased Kv1.5 potency com-
pared with their analogs formed at 0 °C. For example,
4a (Kv1.5 IC50 265 nM) exhibits 10-fold higher potency
than 5 (Kv1.5 IC50 3698.0 nM). Although the magnitude
of this difference in potency varied across the series, gen-
erally the low-temperature diastereomers were more
potent.
N
R1
O
R2
MeO2C
4
Having established that polar groups could be incorpo-
rated into the two aryl rings of 4, a search for replace-
ments of the diisopropyl amide was undertaken.
Despite significant attempts, no potent replacement
was identified, and the diisopropylamide moiety was
found to be an essential component for potent Kv1.5
activity.
Utilizing this methodology, attention was focused on
exploring the SAR of the two aryl rings in 4 (Table 1;
data for the more potent diastereomer are presented).
Replacement of either of the phenyl rings with a 3-pyr-
N
X
a
O
b
O
O
X
X
N
OH
2
O
O
3
4
Scheme 1. Reagents and conditions: (a) EDC, HOBt, Et3N, diisopropylamine, rt, 18 h; (b) methyl cinnamate, LDA, THF, 0 ° or À78 °C, 2 h.

K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5897–5901
Table 1. Kv1.5 antagonist SAR of biaryl diisopropylamides 4
5899
Compound R1
R2
a
Kv1.5 IC50 (nM)
N
N
N
N
a
b
O
O
4a
4a
265
266
N
R1
N
R2
O
HO
O
4b
4c
4d
4eb
4f
6
7
N
Scheme 2. Reagents and conditions: (a) EtOH, NaOH (1 N aqueous),
rt 24 h, quantitative; (b) NHR1R2, PS-DCC, HOBt or HOAt, Hunig’s
base, DCM, rt, 18 h.
13760
5323
6348
6868
708
N
N
Table 2. Inhibition activity of 5-amides 7
Compound
HNR1R2
a
Kv1.5 IC50 (nM)
N
7a
249
NH2
N
7b
7c
410
432
NH2
NH2
2
4g
4h
4i
F
N
N
N
N
3836
137
Br
7d
589
NH2
3
N
N
S
NC
7e
7f
1594
2669
3379
NH2
4j
184
MeNH2
Br
4k
4l
147
O
N
7g
NH2
N
NC
326
a Values represent IC50 for more potent diastereomer, determined in
quadruplicate.
N
4m
4n
215
N
NC
in particular, conferred potency equivalent to the methyl
ester. The branched a-methyl benzylamine 7a was the
most potent of the series and, with the exception of 7e
and 7f, all compounds with IC50 < 5000 nM contained
one or more saturated carbon–carbon bonds linking a
substituted benzene ring. While potent, the representa-
tive amides from the series did not meet the minimal
P-gp susceptibility criteria (BA/AB ratio <3).
1773
N
N
CN
4o
113
O N
a Values represent IC50 for more potent diastereomer, determined in
quadruplicate.
In hopes of conferring P-gp susceptibility, reversed
amides were prepared via Curtius rearrangement of 6
(DPPA) in the presence of triethylamine in tert-buta-
nol.12 The tert-butoxycarbonyl protected Curtius rear-
rangement product was isolated in 81% yield,
deprotected (TFA) to give 8, and coupled with carbox-
ylic acids to give the reversed amides 9 (Scheme 3).
b Compound 4e was evaluated as a 2.5:1 mixture of diastereomers.
Subsequently maintaining the upper portion of the scaf-
fold, we addressed the metabolically labile ester func-
tionality within the 2-pyridin-3-yl series. Amide
analogs were prepared via saponification (NaOH) of
the methyl ester 4a and the coupling of carboxylic acid
intermediate 6 with various amines (Scheme 2).
The 5-reversed amides exhibited significant improve-
ments in Kv1.5 potency (Table 3). Homologous reversal
of the amide 7c resulted in 9a, the most potent com-
pound identified in this series to date (IC50 = 36 nM).
A comparison of 9c and analog 7b emphasizes the
While simple amides exhibited moderate Kv1.5 antago-
nist activity (7f, Table 2), larger groups, benzylic amines

5900
K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5897–5901
Table 4. Inhibition activity for 5-ureas 9g–q
a
Kv1.5 IC50 (nM)
N
N
Compound
R
N
N
a
b or c
O
R
O
O
6
9g
150
N
H
N
NH2
H
9
F
8
H
N
Scheme 3. Reagents and conditions: (a) i—DPPA, Et3N, tert-butyl
alcohol, reflux, 24 h, 81%; ii—TFA, DCM, quantitative; (b) carboxylic
acid, PS-DCC, HOBt or HOAt, DIPEA, DCM, rt, 18 h.; (c)
isocyanate, Et3N, DCM, rt, 30 min.
9h
9i
240
252
256
283
O
O
H
N
Table 3. Inhibition activity of 5-reversed amides 9a–f
O
a
Kv1.5 IC50 (nM)
Compound
R
H
N
9j
F
O
O
9a
36
O
2
9k
N
H
9b
9c
9d
9e
9f
94
100
O
O
H
N
2
9l
298
341
353
362
490
757
2
O
O
H
250
N
9m
9n
9o
9p
9q
F
F
3
O
O
498
H
N
2
Cl
O
O
1310
Br
NH2
H
F
N
a Values represent IC50 for more potent diastereomer, determined in
quadruplicate.
O
H
N
O
O
importance of the amide orientation, and reversed
amides generally gain 2- to 10-fold in activity. Despite
its high potency, compound 9a was not a P-gp substrate
(BA/AB ratio <3) and was not further pursued.
Cl
H
N
a Values represent IC50 for more potent diastereomer, determined in
quadruplicate.
It has been noted that P-gp susceptibility depends on a
number of factors, including increased polar functional-
ity and/or the availability of hydrogen-bond donating
groups on substrates. Because reversed amides were so
well tolerated with regard to Kv1.5 activity, it was rea-
soned that incorporation of an additional H-bond
donor would not affect potency and could generate a
P-gp substrate. Preparation of these analogs was
straightforward from amine 8 via reaction with the
appropriate isocyanate (Scheme 3). As we had hoped,
adding polar functionality did confer P-gp susceptibility
in this series (for 9g, BA/AB ratio = 6.6). Furthermore,
the resulting ureas generally exhibited potent Kv1.5
antagonism, with benzylic amine-derived ureas among
the most potent (Table 4). 2-Fluorobenzyl urea 9g dem-
onstrated the best potency with a Kv1.5 IC50 = 150 nM.
The diastereomerically pure racemic analog 9g was
resolved via chiral chromatography (ChiralPak AD,
n-hexanes/ethanol, 85:15) to yield a single active enan-
tiomer, 10, which exhibited potency equivalent to 9g
and was used for all further characterization. The dog
i.v. pharmacokinetic properties of this analog
(Cl = 9.81 mL/min/kg) were sufficient to allow evalua-
tion in vivo in a canine electrophysiology model.
Canine cardiac electrophysiological studies were con-
ducted as previously described.13 Compound 10 was in-
fused continuously over 1 h to anesthetized dogs, and
atrial refractory period (ARP), ventricular refractory
period (VRP), QT interval, heart rate, and mean arterial
pressure were monitored. Analog 10 elicited a significant
dose-responsive increase in atrial refractory period
(Fig. 3). Animals tested at all doses showed no changes
in ventricular refractory period (Fig. 4) or QT interval,
indicating a selective atrial effect. Plasma compound lev-
els associated with a 10% increase in atrial refractory
period were ꢀ20 nM.

K. K. Nanda et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5897–5901
5901
(diisopropylcarbamoyl)-2-phenyl-3-(pyridin-3-yl)propyl)-
3-(2-fluorobenzyl)urea (10) showed significant atrial-se-
lective effects in the dog electrophysiological model.
Compound 10 is also a P-gp substrate with low potential
for CNS exposure.
Effect of Vehicle vs 10,
0.15, 0.30 and 0.45 ug/kg/min, on ARP in Anesthetized Dog
(25% beta-CD vehicle, 20 ml over 60 min)
20
15
10
5
Acknowledgments
The authors thank Nathan R. Kett for the synthesis of
compound 4e, and Charles W. Ross, III and Joan S.
Murphy for obtaining HRMS data. We express our
gratitude to Todd Anderson, April L. Cox, Anna
Dudkina, and Ray T. McClain for assisting with the
reverse phase purification of the final products.
0
-5
-10
10
20
30
40
50
60
70
Infusion Time (min)
References and notes
Vehicle
0.15 ug/Kg/min [10] = 4 - 9 nM
0.30 ug/Kg/min [10] = 12 - 18 nM
0.45 ug/kg/min [10] = 16 - 21 nM
1. Kannel, W. B.; Wolf, P. A.; Benjamin, E. J.; Levy, D. Am.
J. Cardiol. 1998, 82, 2N.
´
2. Marban, E. Nature 2002, 415, 213.
Figure 3. In vivo study of 10 and atrial refractory period.
3. (a) Wang, Z.; Fermini, B.; Nattel, F. B. Circ. Res. 1993,
73, 1061; (b) Feng, J.; Wible, B.; Li, G.-R.; Wang, Z.;
Nattel, S. Circ. Res. 1997, 80, 572.
Effect of Vehicle vs 10,
0.30 and 0.45 ug/kg/min, on VRP in Anesthetized Dog
(25% beta-CD vehicle, 20 ml over 60 min)
4. Ramakrishnan, P. Einstein Quart. J. Biol. Med. 2003, 19,
160.
5. Diastereomerically pure racemate. By analogy, this com-
pound has relative stereochemistry similar to that of 4a
(Fig. 2).
20
15
10
5
6. Determined by patch clamp electrophysiology (Molecular
Devices IonWorks): Zheng, W.; Spencer, R. H.; Kiss, L.
Assay Drug Dev. Technol. 2004, 2, 543.
´
7. (a) Wettwer, E.; Hala, O.; Christ, T.; Heubach, J. F.;
Dobrev, D.; Knaut, M.; Varro, A.; Revens, U. Circulation
´
2004, 110, 2299; (b) Roy, D.; Rowe, B. H.; Stiell, I. G.;
Coutu, B.; Ip, J. H.; Phaneuf, D.; Lee, J.; Vidaillet, H.;
Dickinson, G.; Grant, S.; Ezrin, A. M.; Beatch, G. N.
J. Am. Coll. Cardiol. 2004, 44, 2355.
0
8. Stefanovsky, Y.; Gospodova, T.; Viteva, L. Tetrahedron
1986, 42, 5355.
-5
9. (a) Chen, T.-Y. R.; Anderson, M. R.; Grossman, S.;
Peters, D. G. J. Org. Chem. 1987, 52, 1231; (b) Heathcock,
C. H.; Oare, D. A. J. Org. Chem. 1985, 50, 3022.
10. (a) Xie, L.; Isenberger, K. M.; Held, G.; Dahl, L. M.
J. Org. Chem 1997, 62, 7516; (b) Kanemasa, J.; Nomura,
M.; Taguchi, Y. Chem. Lett. 1992, 1801; (c) Kanemesa, S.;
Nomura, M.; Wada, E. Chem. Lett. 1991, 1735; (d)
Heathcock, C. H.; Henderson, M. A.; Oare, D. A.;
Sanner, M. A. J. Org. Chem. 1985, 50, 3019.
-10
10
20
30
40
50
60
70
Infusion Time (min)
Vehicle
0.30 ug/Kg/min, [10] = 12 - 18 nM
0.45 ug/Kg/min [10] = 16 - 21 nM
Figure 4. In vivo study of 10 and ventricular refractory period.
11. Spears, G. W.; Caufield, C. E.; Still, W. C. J. Org. Chem.
1987, 52, 1226.
12. Shioiri, T.; Ninomiya, K.; Yamada, S. J. Am. Chem. Soc.
1972, 94, 6203.
13. Stump, G. L.; Smith, G. R.; Tebben, A. J.; Jahansouz, H.;
Salata, J. J.; Selnick, H. G.; Claremon, D. A.; Lynch, J. J.
J. Cardiovasc. Pharmacol. 2003, 42, 105.
In summary, we have discovered a novel series of potent
Kv1.5 antagonists based on the diisopropyl amide scaf-
fold 4. These compounds exhibit in vitro IKur/Kv1.5
antagonist activity and, notably, key compound 1-(3-