Design and synthesis of novel isoquinoline-3-nitriles as orally bioavailable Kv1.5 antagonists for the treatment of atrial fibrillation

Article abstract of DOI:10.1021/jm060927v

Novel 3-cyanoisoquinoline Kv1.5 antagonists have been prepared and evaluated in in vitro and in vivo assays for inhibition of the Kv1.5 potassium channel and its associated cardiac potassium current, I_Kur. Structural modifications of isoquinoli

Full text of DOI:10.1021/jm060927v

6954
J. Med. Chem. 2006, 49, 6954-6957
Design and Synthesis of Novel
Isoquinoline-3-nitriles as Orally Bioavailable
Kv1.5 Antagonists for the Treatment of
Atrial Fibrillation
B. Wesley Trotter,*,† Kausik K. Nanda,† Nathan R. Kett,†
Christopher P. Regan,‡ Joseph J. Lynch,‡ Gary L. Stump,‡
Laszlo Kiss,§ Jixin Wang,# Robert H. Spencer,#
Stefanie A. Kane,# Rebecca B. White,| Rena Zhang,|
Kenneth D. Anderson,† Nigel J. Liverton,†
Figure 1. Profiles of isoquinolinones 1 and 2.
Charles J. McIntyre,† Douglas C. Beshore,†
George D. Hartman,† and Christopher J. Dinsmore†
Screening efforts identified compound 1 as a potent Kv1.5
antagonist10 with moderate plasma clearance and low bioavail-
ability in rat pharmacokinetic experiments. Of primary concern
was the moderate selectivity observed with respect to the human
ether-a-go-go-related gene (hERG) channel as measured by
voltage clamp experiments (Figure 1). hERG and the IKr current
that it mediates are present in human atrium and ventricle.
Because the proarrhythmic potential of existing antiarrhythmic
drugs is chiefly due to their high potency against IKr,11 high
selectivity for Kv1.5 over hERG was considered a key feature
for atrial selective therapy. Structure-activity relationships
(SAR) in this series suggested that the dimethylaminomethyl
moiety was associated with the undesired hERG blockade by
1, and synthetic modifications identified replacement of this
functional group with a cyano substituent as a viable strategy
for increasing selectivity over hERG.
Accordingly, cyanoisoquinolinone 2 (Figure 1) exhibited
Kv1.5 potency similar to that of 1 and no measurable activity
in a hERG binding assay.12 An effort to expand the SAR around
this lead found that substitutions and replacements of the phenyl
substituent did not improve potency or physical properties nor
did alterations of the substitution pattern of the fused aryl region.
The best opportunity for alterations of potency and physical
properties was found in the modification of the N-methyl group
of 2. Because of the empirically observed low solubility of 2,
which limited its use in animal experiments, polar groups were
incorporated with the aim of decreasing overall compound
lipophilicity. Synthesis of these compounds proceeded as
outlined in Scheme 1.
Ortho-lithiation of 4-methoxybenzoic acid (3) and treatment
with ethyl benzoate or methyl 3-fluorobenzoate provided
addition products 4a,b without recourse to protection of the
carboxylic acid moiety.13 Thionyl chloride treatment gave 5a,b,
which were treated with allylaminoacetonitrile14 to provide the
corresponding amide derivatives.15 Treatment with sodium
methoxide resulted in cyclization to provide key isoquinolinone
intermediates 6a,b. Osmium-catalyzed dihydroxylation and
periodate cleavage of 6b provided aldehyde 7, which underwent
reductive transformations under standard conditions to give
adducts 8c and 8d. Alternatively, dihydroxylation of 6a,b gave
diols 8a,b, which could be transformed via standard protecting
group manipulations and displacement chemistry to amines such
as 8e. 8a was also readily resolved by chiral HPLC to provide
enantiomerically pure diols 8f and 8g. Use of alternatively
substituted benzoate electrophiles in the first step of this
sequence provided analogues such as 8h (Table 1).
Departments of Medicinal Chemistry, Stroke and
Neurodegeneration, Automated Biotechnology, Pain Research, and
Drug Metabolism, Merck Research Laboratories, WP14-2, P.O. Box
4, Sumneytown Pike, West Point, PennsylVania 19486
ReceiVed August 1, 2006
Abstract: Novel 3-cyanoisoquinoline Kv1.5 antagonists have been
prepared and evaluated in in vitro and in vivo assays for inhibition of
the Kv1.5 potassium channel and its associated cardiac potassium
current, IKur. Structural modifications of isoquinolinone lead 1 afforded
compounds with excellent potency, selectivity, and oral bioavailability.
Atrial fibrillation (AF) is the most common sustained
arrhythmia observed in clinical practice. The prevalence of AF
is rising as the population ages, and the disease has been
identified as a major contributor to stroke and resultant
morbidities and mortalities.1 Pharmacological approaches to AF
treatment have centered on ion channel blockers, as reentrant
electrical excitation appears to play a major role in disease
physiology.2 Currently available antiarrhythmic drugs inhibit
multiple ion channels and therefore exhibit frequent adverse
affects, including potentially lethal ventricular proarrhythmia.3
This limitation has led to a recent focus on “atrial selective”
therapy as a potentially safer treatment for AF.4
The ultrarapid delayed rectifier potassium current IKur, which
is observed in the human atrium but not in ventricle,5 has
emerged as a target for atrial selective therapy. The voltage gated
Kv1.5 channel is responsible for IKur and is also more prevalent
in human atrium than ventricle.6 IKur is a component of the
repolarization phase of the atrial action potential, and its
inhibition is known to effect prolongation of the atrial refractory
period and consequent restoration of normal sinus rhythm to
arrhythmic tissue. Thus, selective blockade of Kv1.5 may result
in termination and/or prevention of AF without exerting
undesired ventricular effects. Recent publications, most notably
from researchers at Sanofi-Aventis, have detailed the effects of
selective Kv1.5 antagonists in animal models of arrhythmia.7,8
Our efforts in this area have focused on the development of
compounds suitable for chronic oral dosing that would represent
a viable approach to prevention of AF recurrence in patients.
Here, we detail the discovery of a novel series of orally
bioavailable isoquinoline Kv1.5 blockers that have evolved via
optimization of isoquinolinone 1 (ISQ-1, Figure 1).7,9
* To whom correspondence should be addressed. Telephone: 215-652-
4035. Fax: 215-652-7310. E-mail: bwesley_trotter@merck.com.
† Department of Medicinal Chemistry.
Evaluation of compound potency was aided by the use of
high-throughput patch clamp (HT-clamp) measurements. This
technology provided determinations of potency versus the Kv1.5
channel with throughput many times that possible using
‡ Department of Stroke and Neurodegeneration.
§ Department of Automated Biotechnology.
# Department of Pain Research.
| Department of Drug Metabolism.
10.1021/jm060927v CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/27/2006

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 6955
Table 1. Heteroatom-Containing Isoquinolinone N-Substituents
Scheme 1a
a Determined according to ref 16. Values were determined from 10-point
dose response curves, n ) 3-4 cells per point. See Supporting Information
for assay protocol and precision assessment. b Compound 1 was employed
as a positive control; HT-clamp IC50 ) 190 nM. c See ref 12.
a Conditions: (a) sec-butyllithium, tetramethylethlyenediamine, tetra-
hydrofuran, -78 °C; (b) thionyl chloride, catalytic dimethylformamide, 1,2-
dichloroethane, reflux; (c) allylaminoacetonitrile, 2,6-lutidine, toluene, reflux;
(d) sodium methoxide, methanol; (e) osmium tetroxide, N-methylmorpholine
N-oxide, acetone/water, then sodium periodate; (f) sodium borohydride,
methanol to give 8c; (g) N-methyl-2-methoxyethylamine, sodium triaceto-
xyborohydride, 1,2-dichloroethane to give 8d; (h) osmium tetroxide,
N-methylmorpholine N-oxide, acetone/water; (i) (1) tert-butyldiphenyl-
silyl chloride, triethylamine, dimethylaminopyridine, dichloromethane, (2)
methanesulfonyl chloride, triethylamine, dichloromethane, (3) sodium azide,
dimethylformamide, (4) hydrogen, 10% palladium on carbon, ethyl acetate,
55 psi, (5) 3-pyridinecarboxaldehyde, sodium triacetoxyborohydride, 1,2-
dichloroethane to give 8e; (j) ChiralPak AD HPLC separation.
(ARP), ventricular refractory period (VRP), and a variety of
other cardiac intervals. Importantly, IKur is known to function
in rat atrium and ventricle; thus, the rat model provided a
measure of in vivo efficacy but no direct measure of selectivity.
Infusion of 8f produced a distinct increase in ARP and VRP
with no effects on cardiac conduction (Figure 2). At the highest
dose administered ([plasma] > 35 µM), the compound did not
affect AV node refractoriness (AVRP), a parameter that is
significantly increased in this model by IKr blockers.18
Given this desirable in vivo profile, we sought to identify
more potent Kv1.5 antagonists based on the dihydroxypropyl
structural motif. A significant advance with regard to potency
and pharmacokinetic properties was realized with the synthesis
of isoquinoline 12 (Scheme 2), the result of a formal migration
of the dihydroxypropyl group. Isoquinolines could be synthe-
sized in a straightforward manner from isoquinolinone 9,
prepared by palladium-catalyzed reductive deallylation of 6a.19
Treatment with phosphorus oxychloride provided chloroiso-
quinoline 10, which served as a precursor to alkoxy- and amino-
substituted isoquinolines. Reaction of 10 and the sodium
alkoxide of (S)-solketal provided adduct 11. Brief exposure of
11 to acid gave diol 12.20 Addition of aminopropane-2,3-diol
to 10 proceeded under microwave-assisted conditions to afford
13. Both 12 and 13 were potent Kv1.5 antagonists with good
selectivity versus hERG. We were pleased to find that 12 and
13 each exhibited substantially reduced plasma clearance and
markedly improved bioavailability relative to 8f when dosed
to rats (Figure 3).
Rat electrophysiology studies with 12 and 13 revealed
unexpected and undesirable effects on cardiac parameters (Fig-
ure 3). Infusion of each compound decreased ARP. Prolongation
of VRP was observed, and in the case of 13, AVRP was also
increased significantly at high plasma levels. While the cause
of this anomalous cardiac profile was unknown, this undesirable
effect was consistently observed in the rat model for a variety
of additional analogues containing the vicinal dihydroxypropyl
moiety. Subsequent optimization efforts therefore focused on
replacement of the dihydroxypropyl functionality.
Figure 2. In vivo and in vitro profile of 8f.
traditional patch clamp methods.16 Data obtained using this
assay17 indicated that, while incorporation of polar groups could
reduce Kv1.5 antagonist potency significantly, a variety of
compounds retained appreciable potency and good selectivity
versus hERG (Table 1). Hydroxyl groups proved particularly
advantageous as a means of incorporating polarity without
increasing undesirable hERG binding activity (in contrast to
basic amines such as 8d).
Rat pharmacokinetic profiling identified enantiomerically pure
diol 8f as a compound with moderate plasma clearance (Figure
2) and suitable physical properties for evaluation in a rat in
vivo experiment. Anesthetized rat electrophysiology18 was
employed to characterize the cardiac electrophysiological effects
of 8f. Infusion of the compound to surgically instrumented rats
provided measurements of effects on the atrial refractory period
Chloroisoquinoline 15 was therefore prepared from 6b via
the established deallylation and chlorination procedures (Scheme
3). Addition of nitrogen-containing heterocycles to 15 using the

6956 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24
Letters
Scheme 2a
a Conditions: (a) triethylamine, formic acid, tetrakis(triphenylphosphine)-
palladium(0), dioxane, 100 °C; (b) phosphorus oxychloride, 90 °C; (c) (S)-
3-amino-1,2-propanediol, n-butanol, microwave, 210 °C, 1 h; (d) (S)-
solketal, sodium hydride, THF, 66 °C; (e) concentrated hydrochloric acid,
tetrahydrofuran, 0 °C, 10 min.
Scheme 3a
Figure 3. In vivo and in vitro profiles of 12 and 13.
a Conditions: (a) triethylamine, formic acid, tetrakis(triphenylphosphine)-
palladium(0), dioxane, 100 °C; (b) phosphorus oxychloride, 90 °C; (c)
4-methylimidazole, n-butanol, microwave, 200 °C, 3 h.
Scheme 4a
a Conditions: (a) zinc cyanide, zinc powder, tris(dibenzylideneacetone)-
dipalladium(0), dimethylacetamide, 120 °C; (b) sodium hydroxide, ethanol/
water, 40 °C, then HCl, 100 °C; (c) ethanolamine, N-(3-dimethylamino-
propyl)-N′-ethylcarbodiimide hydrochloride, 7-aza-1-hydroxybenzotriazole,
diisopropylethylamine, dimethylformamide.
Figure 4. In vitro and in vivo profile of 16.
aforementioned microwave-assisted conditions was employed
to access a variety of isoquinolines with improved in vivo
profiles. Imidazole 16 was typical of this class, exhibiting the
desired ARP prolongation upon infusion in the rat EP model
(Figure 4). No effects on other cardiac parameters were
observed.
Figure 5. In vitro and in vivo characterization of 19.
Extensive variation of the nitrogen-containing heterocycle
component of 16 did not identify compounds with improved
potencies or pharmacokinetic profiles. Improvements were
eventually realized by substitution of the isoquinoline 1-position
with carboxyl derivatives. Palladium-catalyzed cyanation21 of
10 (Scheme 4) could be employed to obtain bis-cyano inter-
mediate 17. Key to further elaboration was the discovery that
17 could be regioselectively hydrolyzed to provide carboxylic
acid 18. Peptide coupling chemistry then produced carboxamido-
substituted isoquinolines typified by ethanolamide 19.
Ethanolamide 19 exhibited improved potency, excellent
selectivity versus hERG, and good pharmacokinetic properties
(Figure 5). Rat EP experiments confirmed that the compound
potently increased ARP without significant effects on AVRP
(Supporting Information).
Figure 6. Effects on ARP and VRP upon infusion of 19 to anesthetized
dogs. Infusion regimens and associated plasma levels are noted in the
inset.
To more definitively assess the potential for atrial selectivity
in this series, 19 was evaluated in a canine electrophysiological
model.22 Kv1.5 and its associated IKur current figure prominently
in canine atrial repolarization,23 and anesthetized canine electro-
physiology experiments with Kv1.5 blockers have demonstrated
ARP prolongation without concomitant VRP prolongation.24
Consistent with these observations, administration of compound
19 to anesthetized dogs (Figure 6) selectively prolonged ARP
(plasma EC10 ≈ 110 nM). No effect on VRP was observed at

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 24 6957
Claremon, D. A. Preparation of isoquinoline derivatives as potassium
channel inhibitors. WO 05/030130, 2005.
any dose (data for highest dose shown, Figure 6). The combina-
tion of pharmacokinetic properties and in vivo effects observed
establish 19 as a promising atrial-selective agent.
(10) Kv1.5 voltage clamp measurements were obtained as detailed in:
Lagrutta, A.; Wang, J.; Fermini, B.; Salata, J. J. Novel, potent
inhibitors of human Kv1.5 K+ channels and ultrarapidly activating
delayed rectifier potassium current. J. Pharmacol. Exp. Ther. 2006,
317 (3), 1054-1063.
In conclusion, when a high-throughput patch clamp assay was
used to guide compound design, lead optimization provided
compounds that were efficacious in rat EP experiments. These
rodent experiments evaluated multiple cardiac parameters and
exposed undesirable in vivo effects of key lead compounds.
Given its low compound requirements and fast turnaround times,
the rat EP model served as an effective method for identifying
off-target effects and eliminating these effects without resort to
extensive in vitro counterscreening. Guidance of structural
optimization by rat EP provided compounds with good potency,
selectivity, pharmacokinetic properties, and in vivo efficacy.
Compound 19 represents a potent Kv1.5 antagonist with a
promising pharmacokinetic profile that has demonstrated the
ability to selectively increase ARP in canine electrophysiological
experiments.
(11) Redfern, W. S.; Carlsson, L.; Davis, A. S.; Lynch, W. G.; MacKenzie,
I.; Palethorpe, S.; Siegl, P. K. S.; Strang, I.; Sullivan, A. T.; Wallis,
R.; Camm, A. J.; Hammond, T. G. Relationships between preclinical
cardiac electrophysiology, clinical QT interval prolongation and
torsade de pointes for a broad range of drugs: evidence for a
provisional safety margin in drug development. CardioVasc. Res.
2003, 58 (1), 32-45. Jamieson, C.; Moir, E. M.; Rankovic, Z.;
Wishart, G. Medicinal chemistry of hERG optimizations: highlights
and hang-ups. J. Med. Chem. 2006, 49 (17), 5029-5046.
(12) Raab, C. E.; Butcher, J. W.; Connolly, T. M.; Karczewski, J.; Yu,
N. X.; Staskiewicz, S. J.; Liverton, N.; Dean, D. C.; Melillo, D. G.
Synthesis of the first sulfur-35-labeled hERG radioligand. Bioorg.
Med. Chem. Lett. 2006, 16 (6), 1692-1695.
(13) Bennetau, B.; Mortier, J.; Moyroud, J.; Guesnet, J.-L. Directed
lithiation of unprotected benzoic acids. J. Chem. Soc., Perkin Trans.
1 1995, 10, 1265-1271.
(14) Kirino, O.; Oshita, H.; Oishi, T.; Kato, T. Preventive activities of
N-substituted-alpha-aminonitriles against fusarium diseases. Agric.
Biol. Chem. 1980, 44 (1), 25-30.
Acknowledgment. We thank Scott D. Mosser, Wei Lemaire,
John F. Fay, Sandor L. Varga, Charles W. Ross, III, Harry G.
Ramjit, Suzie Yeh, Cynthia Miller-Stein, Debra McLoughlin,
Anne Taylor, Tamara Pittman, Todd Killino, Scott E. Fauty,
and Carl F. Homnick for technical support of this work.
(15) The tautomeric structure of chlorolactone 5 is consistent with previous
observations for this structural type. Reactions of chlorolactones with
secondary amines are known to proceed as expected for the acyclic
acid chloride tautomer of 5. Nishio, T.; Yamamoto, H. Preparation
of 3-hydroxyisoindole-1-ones and o-acelbenzamides. A study of ring-
chain tautomerism. J. Heterocycl. Chem. 1995, 32, 883-891. Yost,
W. L.; Burger, A. Rearrangment of diethyl 3-phenylphthalidyl-3-
malonate to derivatives of 3-phenylindone-2-carboxylic acid. J. Org.
Chem. 1950, 15 (5), 1113-1118.
Supporting Information Available: Spectral and analytical
data for new compounds, one scheme detailing the preparation of
2, rat EP data for 19, and the HT-clamp assay protocol. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(16) Zheng, W.; Spencer, R. H.; Kiss, L. High throughput assay
technologies for ion channel drug discovery. Assay Drug DeV.
Technol. 2004, 2 (5), 543-552.
(17) HT-clamp measurements were obtained using CHO cells expressing
human Kv1.5 channels. Human and rat Kv1.5 channels are 90%
homologous, with the largest sequence differences occurring in the
nonpore regions. See Supporting Information and: Garcia-Guzman,
M.; Sala, F.; Criado, M.; Sala, S. A delayed rectifier potassium
channel cloned from bovine adrenal medulla: Functional analysis
after expression in Xenopus oocytes and in a neuroblastoma cell line.
FEBS Lett. 1994, 354, 173-176.
(18) Regan, C. P.; Cresswell, H. K.; Zhang, R.; Lynch, J. J., Jr. Novel
method to assess cardiac electrophysiology in the rat: characterization
of standard ion channel blockers. J. CardioVasc. Pharmacol. 2005,
46 (1), 68-75.
(19) Koch, T.; Hesse, M. Synthesis of macrocyclic N-unsubstituted imides.
Synthesis 1992, 931-932.
(20) McClure, D. E.; Engelhardt, E. L.; Mensler, K.; King, S.; Saari, W.
S.; Huff, J. R.; Baldwin, J. J. Chiral heteroaryloxymethyloxiranes.
J. Org. Chem. 1979, 44 (11), 1826-1831. Barlow, J. J.; Block, M.
H.; Hudson, J. A.; Leach, A.; Longridge, J. L.; Main, B. G.;
Nicholson, S. Acid catalyzed racemization of 1-(heterocyclyloxy)-
2,3-propanediols. J. Org. Chem. 1992, 57 (19), 5158-5162.
(21) Jin, F.; Confalone, P. N. Palladium-catalyzed cyanation reactions of
aryl chlorides. Tetrahedron Lett. 2000, 41 (18), 3271-3273.
(22) Stump, G. L.; Smith, G. R.; Tebben, A. J.; Jahansouz, H.; Salata,
J. J.; Selnick, H. G.; Claremon, D. A.; Lynch, J. J., Jr. In vivo canine
cardiac electrophysiologic profile of 1,4-benzodiazepine IKs blockers.
J. CardioVasc. Pharmacol. 2003, 42 (1), 105-112.
(23) Fedida, D.; Eldstrom, J.; Hesketh, J. C.; Lamorgese, M.; Castel, L.;
Steele, D. F.; Van Wagoner, D. R. Kv1.5 is an important component
of repolarizing K+ current in canine atrial myocytes. Circ. Res. 2003,
93 (8), 744-751.
(24) Stump, G. L.; Wallace, A. A.; Regan, C. P.; Lynch, J. J., Jr. In vivo
antiarrhythmic and cardiac electrophysiologic effects of a novel
diphenylphosphine oxide IKur blocker (2-isopropyl-5-methylcyclo-
hexyl)diphenylphosphine oxide. J. Pharmacol. Exp. Ther. 2005, 315
(3), 1362-1367.
References
(1) Greenlee, R. T.; Vidaillet, H. Recent progress in the epidemiology
of atrial fibrillation. Curr. Opin. Cardiol. 2005, 20 (1), 7-14. Stewart,
S.; Murphy, N.; Walker, A.; McGuire, A.; McMurray, J. Cost of an
emerging epidemic: an economic analysis of atrial fibrillation in the
UK. Heart 2004, 90, 286-292.
(2) Nattel, S. Atrial electrophysiology and mechanisms of atrial fibril-
lation. J. CardioVasc. Pharmacol. Ther. 2003, 8, S5-S11.
(3) Tamargo, J.; Ricardo, C.; Delpon, E. Pharmacological approaches
in the treatment of atrial fibrillation. Curr. Med. Chem. 2004, 11
(1), 13-28.
(4) Goldstein, R. N.; Stambler, B. S. New antiarrhythmic drugs for
prevention of atrial fibrillation. Prog. CardioVasc. Dis. 2005, 48 (3),
193-208.
(5) Fedida, D.; Wible, B.; Wang, Z.; Fermini, B.; Faust, F.; Nattel, S.
Identity of a novel delayed rectifier current from human heart with
a cloned K+ channel current. Circ. Res. 1993, 73, 210-216.
(6) Feng, J.; Wible, B.; Li, G.-R.; Wang, Z.; Nattel, S. Antisense
oligodeoxynucleotides directed against Kv1.5 mRNA specifically
inhibit ultrarapid delayed rectifier K+ current in cultured adult human
atrial myocytes. Circ. Res. 1997, 80, 572-579.
(7) Brendel, J.; Peukert, S. Blockers of the Kv1.5 channel for the
treatment of atrial arrhythmias. Curr. Med. Chem.: CardioVasc.
Hematol. Agents 2003, 1 (3), 273-287.
(8) Peukert, S.; Brendel, J.; Pirard, B.; Bruggemann, A.; Below, P.;
Kleemann, H. W.; Hemmerle, H.; Schmidt, W. Identification, syn-
thesis, and activity of novel blockers of the voltage-gated potassium
channel Kv1.5. J. Med. Chem. 2003, 46 (4), 486-98. Peukert, S.;
Brendel, J.; Pirard, B.; Strubing, C.; Kleemann, H.-W.; Bohme, T.;
Hemmerle, H. Pharmacophore-based search, synthesis, and biological
evaluation of anthranilic amides as novel blockers of the Kv1.5
channel. Bioorg. Med. Chem. Lett. 2004, 14 (11), 2823-2827.
(9) Claremon, D. A.; McIntyre, C. J.; Liverton, N. J. Isoquinolinone
potassium channel inhibitors. WO 02/24655, 2002. Isoquinoline
Kv1.5 blockers in this manuscript are covered in: Trotter, B. W.;
Nanda, K. K.; Kett, N. R.; Dinsmore, C. J.; Ponticello, G. S.;
JM060927V