Aryl sulfonamido indane inhibitors of the Kv1.5 ion channel

Article abstract of DOI:10.1016/j.bmcl.2007.02.052

A collection of aryl sulfonamido indanes based on the lead compound 1 was synthesized and evaluated for Kv1.5 inhibitory activity. Kv1.5 inhibitors have the potential to be atrium-selective agents for treatment of atrial fibrillation. (1R,2R)-1 has an IC₅₀ of 0.033 μM against Kv1.5 and is selective against other cardiac ion channels, including hERG.

Full text of DOI:10.1016/j.bmcl.2007.02.052

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2849–2853
Aryl sulfonamido indane inhibitors of the Kv1.5 ion channel
Michael F. Gross,a Serge Beaudoin,a,* Grant McNaughton-Smith,a George S. Amato,a
Neil A. Castle,a Christine Huang,b Anruo Zoua and Weifeng Yua
aIcagen Inc., PO Box 14487, Research Triangle Park, NC 27709, USA
bBristol-Myers Squibb Pharmaceutical Research Institute, Pennington, NJ 08534, USA
Received 23 January 2007; revised 16 February 2007; accepted 21 February 2007
Available online 27 February 2007
Abstract—A collection of aryl sulfonamido indanes based on the lead compound 1 was synthesized and evaluated for Kv1.5 inhib-
itory activity. Kv1.5 inhibitors have the potential to be atrium-selective agents for treatment of atrial fibrillation. (1R,2R)-1 has an
IC50 of 0.033 lM against Kv1.5 and is selective against other cardiac ion channels, including hERG.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Atrial fibrillation (AF) is one of the most common
forms of cardiac disease encountered in clinical practice.
Currently, over two million people in the United States
are affected by this serious cardiac arrhythmia.1
Although AF is typically not a fatal condition, ineffi-
cient emptying of the atrium leads to concomitant
patient risks, such as thrombogenesis and stroke.2 AF
is a major public health concern and there is a growing
medical need to reduce the occurrence of AF with drugs
that lack the ventricular proarrhythmic side effects (e.g.,
torsades de pointes) associated with currently used phar-
macological therapies.3
Figure 1. Structure of aryl sulfonamido indane Kv1.5 inhibitor.
regioisomers and scaffold variations were probed in an
attempt to define the Kv1.5 pharmacophore.
The preparation of aryl sulfonamido indane 1 is pre-
sented in Scheme 1. Nitration of 1-indanone 2 afforded
6-nitro-1-indanone 3 and 4-nitro-1-indanone 4 in a 3:1
ratio following column chromatography on silica gel.
Ketone reduction of 3 followed by dehydration gave
the indene 5. Treatment with m-CPBA to furnish epox-
ide 6 and subsequent ring opening with ammonium
hydroxide provided the racemic trans-amino alcohol 7.
Sulfonylation of 7 with 4-ethylphenylsulfonyl chloride
and reduction gave aromatic amine 8. Several methods
were used to reduce the nitro group. We found the use
of SnCl2–2H2O or NaBH4 with catalytic NiCl2 was
generally preferred over hydrogenation using Pd/C.
Coupling of 8 with m-anisoyl chloride provided 1.
In humans, the delayed rectifier potassium current
termed IKur has been shown to play an important role
in atrial muscle action potential repolarization.4 In con-
trast to other potassium currents present in the heart
such as IKs and IKr, IKur is functionally expressed only
in the atrium.5 The IKur current is conducted by the volt-
age-gated potassium channel encoded by Kv1.5.6 Thus,
an inhibitor of Kv1.5 would be expected to prolong the
action potential repolarization in the atrium but not in
the ventricle. Drugs that selectively block this ion chan-
nel should increase atrial action potential duration and
prevent and/or terminate AF without the risk of ventric-
ular effects leading to arrhythmogenesis.
Herein, we report a new class of Kv1.5 inhibitors based
on the indane scaffold (Fig. 1).7 In this study, amido
The 1,4-, 1,5-, and 1,7-regioisomers were synthesized to
determine the optimal positioning of the amido group.
Manipulation of 4 in a manner identical to that
described in Scheme 1 provided the 1,4-amido regio-
isomer 9 (Scheme 2).
Keywords: Kv1.5 inhibitors; Atrial fibrillation.
*
Corresponding author. E-mail: sbeaudoin@icagen.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.02.052

2850
M. F. Gross et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2849–2853
Scheme 1. Reagents: (a) KNO3, concd H2SO4, 74%; (b) NaBH4, MeOH, 99%; (c) TsOH, PhCH3, 100%; (d) m-CPBA, CH2Cl2, 96%; (e) NH4OH,
91%; (f) 4-Et-PhSO2Cl, Et3N, CH2Cl2, 83%; (g) SnCl2–2H2O, EtOH, 97%; (h) m-anisoyl chloride, Et3N, CH2Cl2, 93%.
oxide provided a 3:3:1 mixture of nitroindanones 16,
4, and 3. The desired 5-nitroindanone 16 was isolated
by column chromatography. Further elaboration of 16
proceeded as before to provide the 1,5-amido regio-
isomer 17.
The des-amido indane 19 was prepared in order to deter-
mine the contribution of the amido functionality to
Kv1.5 activity. Oxidation, ring opening, and sulfonyla-
tion of 1-indene 18 provided racemic 19 (Scheme 5).
Scheme 2. Synthesis of 4-amido regioisomer 9.
We wanted to ascertain if the 2-hydroxy group inter-
acted with the Kv1.5 channel protein. The racemic
des-hydroxy indane 20 was prepared via amination of
3 followed by sulfonylation, nitro reduction, and acyla-
tion as described in Scheme 6.
To further explore the SAR at the 2-position, the
cis-amino alcohol 22 was prepared as outlined in
Scheme 7. Treatment of indene oxide 6 with trifluorom-
ethanesulfonic acid in acetonitrile provided racemic 21.9
This intermediate was carried forward as described in
Scheme 1 to give 22.
Scheme 3. Reagents: (a) NaBH4, MeOH, 98%; (b) TsOH, toluene,
100%; (c) RhCl3–H2O, EtOH, 43%.
Scaffold variations were investigated (Fig. 2). The tetra-
lin 23 and benzosuberone 24 were synthesized in a fash-
ion identical to that illustrated for indane 1 (Scheme 1)
starting from the appropriate commercially available
precursors, 7-nitro-1-tetralone and 1-benzosuberone,
respectively.
The 1,7-amido regioisomer 12 was also prepared start-
ing from 4. Ketone reduction and dehydration of 4 gave
indene 10. Isomerization of the olefin using RhCl3 in
refluxing ethanol afforded indene 11, along with recov-
ered 10. Processing of 11 as described in Scheme 1 pro-
vided the 1,7-amido derivative 12 (Scheme 3).
The benzyl sulfonamide 26 was prepared in order to
determine whether conformational rigidity imposed by
the indane scaffold was important for Kv1.5 activity.
Sulfonylation of 3-nitrobenzyl amine 25, nitro reduction
and coupling with m-anisoyl chloride provided the
desired acyclic product (Scheme 8).
The 1,5-amido regioisomer 17 was synthesized as de-
scribed in Scheme 4 to complete the series. Nitration
of indane (13) gave a 2:3 mixture of nitroindanes 14
and 15.8 Oxidation of this mixture using chromium tri-

M. F. Gross et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2849–2853
2851
Scheme 4. Reagents: (a) HNO3, H2SO4, 37%; (b) CrO3, AcOH, 3% of 16.
Finally, we wanted to compare the ability of enantio-
mers (1R,2R)-1 and (1S,2S)-1 to inhibit the Kv1.5 chan-
nel. (1R,2R)-1 was prepared via Jacobsen asymmetric
epoxidation of indene 5.10 Under these conditions, the
epoxide (1S,2R)-6 was obtained in 70% ee.11 The stereo-
chemistry was assigned based on literature precedent
suggesting the major isomer expected from the (S,S)-
salen catalyst would be the (1S,2R) isomer.10 (1S,2R)-6
was carried on to give the amino alcohol (1R,2R)-7, then
processed as described in Scheme 1 to ultimately
provide (1R,2R)-1 in 95% ee, after recrystallization from
iPrOH/hexane (Scheme 9). In order to obtain both the
(1R,2R)-1 and (1S,2S)-1 expeditiously, the racemate
was subjected to preparative-scale chiral HPLC chroma-
tography to provide each enantiomer in >98% ee.
Scheme 5. Reagents: (a) m-CPBA, CH2Cl2, 93%; (b) NH4OH, 85%; (c)
4-Et-PhSO2Cl, NEt3, CH2Cl2, 81%.
Compounds were tested for inhibition of potassium cur-
rent in LtkÀ or mouse fibroblast L929 cells expressing hu-
man Kv1.5 using patch-clamp electrophysiological (EP)
techniques.12 The inhibitory activities of the compounds
examined in this study are summarized in Table 1.
Scheme 6. Reagents: (a) NH4OAc, Na(CN)BH3, MeOH, 35%;
(b) 4-Et-PhSO2Cl, NEt3, CH2Cl2, 88%; (c) SnCl2–2H2O, EtOH, 91%;
(d) m-anisoyl chloride; NEt3, CH2Cl2, 92%.
Scheme 7. Reagents: (a) CF3SO3H, CH3CN, 64%.
Figure 2. Tetralin and Benzosuberone Kv1.5 inhibitors.
Scheme 8. Reagents: (a) 4-Et-PhSO2Cl, NEt3, CH2Cl2, 99%; (b) SnCl2–2H2O, EtOH, 99%; (c) m-anisoyl chloride; NEt3, CH2Cl2, 95%.

2852
M. F. Gross et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2849–2853
Table 2. Aqueous solubility of Kv1.5 inhibitors13
Compound
lg/mL
1
(1R,2R)-1
2.9
1.3
20
26
<0.1
0.8
Scheme 9. Reagents: (a) (S,S)-Mn-salen catalyst, PPNO, NaOCl,
0.05 M NaH2PO4, CH2Cl2, 71% yield, 70% ee; (b) NH4OH, 90%.
other ion channels including a human cardiac potassium
channel (hERG),15 a human cardiac voltage-dependent
sodium channel (hSCN5A),16 and calcium channels
found in a rat pituitary cell (GH3).17 Compound
(1R,2R)-1 was found to be at least 300-fold selective,
producing <50% inhibition at 10 lM versus all three
of these ion channels (Table 3).
Table 1. Kv1.5 Inhibitory activity
Compound
IC50 (lM) or inhibition at 1 lMa
1
(1R,2R)-1
(1S,2S)-1
0.16
0.033
0.76
Satisfied with the in vitro profile of (1R,2R)-1, the phar-
macokinetic (PK) properties were investigated in rat and
dog. Despite favorable calculated (e.g., clogP, H-bond
donors, H-bond acceptors, MW)18 and measured (e.g.,
CACO-2 permeability of 1.9 · 10À5 cm/s) properties,
we found that the oral bioavailability (F%) of (1R,2R)-
1 was low in both species. In addition, the intraarterial
infusion (INF) half-life was short, particularly in rat
(Table 4).
8
9
2
2%
17 1%
0.88
24 3%
12
17
19
20
22
23
24
26
3
1%
0.15
0.30
0.44
0.20
0.24
a Inhibition values are means of at least two experiments SEM.
In an effort to understand why the oral bioavailability of
(1R,2R)-1 was so low, a separate PK experiment was
performed using bile duct cannulated (BDC) rats. It
was discovered that the INF dose was rapidly and exten-
sively metabolized. Deacylation, O-dealkylation and
hydroxylation to give metabolites I, II, and III, respec-
tively, were predominant metabolic pathways (Fig. 3).
It was not determined whether hydroxylation of the sul-
fonamido fragment occurred on the phenyl ring or on
the ethyl substituent (III).
All four of the possible amido regioisomers were
examined in this study. The 1,7-isomer 12 retained a
reasonable level of Kv1.5 activity, whereas the 1,4-
and 1,5-isomers (9 and 17, respectively) were less potent
having IC50 values >1 lM. Clearly, the 1,6-isomer 1
arranged the sulfonamido and amido groups in the most
preferred orientation. Removal of the amido functional-
ity was not tolerated (see 8 and 19).
Thorough analysis of the oral (PO) dose in BDC rats
revealed that most of (1R,2R)-1 remained unabsorbed
from the gastrointestinal tract. This was attributed lar-
gely to the poor aqueous solubility of (1R,2R)-1 that
was measured to be 1.3 lg/mL (Table 2).
The des-hydroxy indane 20 and trans indane 1 were
essentially equipotent. The cis indane 22 was slightly less
potent (approximately 2-fold). Combined, these results
suggest that the 2-hydroxy group does not participate
in a critical binding interaction with the Kv1.5 channel
protein. Removing the hydroxyl group from the indane
template, however, resulted in a substantial loss of aque-
ous solubility (Table 2).
Table 3. Cardiac ion channel selectivity of (1R,2R)-1
Ion channel
Inhibition at 10 lMa
hERG
hSCN5A
GH3 (Ca)
37 5%
15 4%
27 4%
Of the bicyclic scaffold modifications that were explored,
the indane series retained the highest level of Kv1.5
inhibitory activity. Nevertheless, the tetralin and benzo-
suberone aryl sulfonamido templates (23 and 24, respec-
tively) have the potential to be optimized further to
provide interesting Kv1.5 inhibitors.14 It appeared that
conformational constraint is not required for Kv1.5
activity based on potency of 26. However, this com-
pound suffered from the same solubility limitations as
the des-hydroxy analog 20 (Table 2).
a Inhibition values are means of at least three EP experiments SEM.
Table 4. Pharmacokinetic parameters of (1R,2R)-1a
Parameter
Rat
Dog
5
F (%)
t1/2 (h)
1 0.6
0.4 0.01
2
1.5 0.2
12
1.4
Clearance (mL/min/kg)
Vd (L/kg)
200
4
The chirality of 1 significantly influenced Kv1.5 channel
blockade: (1R,2R)-1 was approximately 25 times more
potent than (1S,2S)-1. Encouraged by this level of activ-
ity, the inhibitory effect of (1R,2R)-1 was studied on
a Dose: 20 lmol/kg PO; 10 lmol/kg INF; vehicle: 1:1:1 PEG:EtOH:
H2O.

M. F. Gross et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2849–2853
2853
Figure 3. In vivo metabolites of (1R,2R)-1.
5. Li, G.-R.; Feng, J.; Yue, L.; Carrier, M.; Nattel, S. Circ.
Res. 1996, 78, 689.
6. Fedida, D.; Wible, B.; Wang, Z.; Fermini, B.; Faust, F.;
Nattel, S.; Browm, A. M. Circ. Res. 1993, 73, 210.
7. Compounds in this letter were first disclosed in patent
application WO9804521.
8. Markgraf, J. H.; Greeno, E. W.; Miller, M. D.; Zaks, W.
J.; Lee, G. A. Tetrahedron Lett. 1983, 24, 241.
9. Senanayake, C. H.; DiMichele, L. M.; Liu, J.; Freden-
burgh, L. E.; Ryan, K. M.; Roberst, F. E.; Larsen, R. D.;
Verhoeven, T. R.; Reider, P. J. Tetrahedron Lett. 1995, 36,
7615.
10. Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.;
Deng, L. J. Am. Chem. Soc. 1991, 113, 7063.
11. Enantiomeric excess (ee) was determined by chiral HPLC
using a CHIRALPAK AS column.
In summary, we investigated the Kv1.5 pharmacophoric
elements of the aryl sulfonamido indane series. It was
discovered that (1R,2R)-1 is a potent Kv1.5 inhibitor
with an IC50 of 0.033 lM. The selectivity of (1R,2R)-1
versus other cardiac ion channels indicates this com-
pound may have the potential for treating AF without
the risk of unwanted ventricular effects. However, the
unacceptable pharmacokinetic profile of (1R,2R)-1 pre-
cludes its further development. In subsequent publica-
tions we will discuss our efforts to design Kv1.5
inhibitors having more desirable physiochemical and
PK properties.
References and notes
12. Synders, D. J.; Tamkun, M. M.; Bennett, P. B. J. Gen.
Physiol. 1993, 101, 513.
1. (a) Page, R. L.; Roden, D. M. Nat. Rev. Drug Disc. 2005,
4, 889; (b) Lloyd-Jones, D. M.; Wang, T. J.; Leip, E. P.;
Larson, M. G.; Levy, D.; Vasan, R. S.; D’Agostino, R. B.;
Massaro, J. M.; Beiser, A.; Wolf, P. A.; Benjamin, E. J.
Circulation 2004, 110, 1042.
2. Crystal, E.; Connolly, S. J. Heart 2004, 90, 813.
3. (a) Recanatini, M.; Poluzzi, E.; Masetti, M.; Cavalli, A.;
De Ponti, F. Med. Res. Rev. 2005, 25, 133; (b) Clancy, E.
E.; Kurokawa, J.; Tateyama, M.; Wehrens, X. H. T.;
Kass, R. S. Annu. Rev. Pharmacol. Toxicol. 2003, 43, 441;;
(c) Yang, T.; Snyders, D.; Roden, D. M. J. Cardiovasc.
Pharmcol. 2001, 38, 737.
13. Aqueous solubility of the crystalline solids was determined
in 50 mM phosphate-buffered saline at pH 6.8 using the
shake-flask method and LC/MS/MS detection.
14. Gross, M. F.; Castle, N. A.; Zou, A.; Wickenden, A. D.;
Yu, W.; Spear, K. L. Bioorg. Med. Chem. Lett. 2006, in
preparation.
15. Nerbonne, J. M. J. Physiol. 2000, 525, 285.
16. Gellens, M. E.; George, A. L.; Chen, L.; Chahine, M.;
Horn, R.; Barchi, R. L.; Kallen, R. G. Proc. Natl. Acad.
Sci. U.S.A. 1992, 89, 554.
17. Glassmeier, G.; Hauber, M.; Wulfsen, I.; Weinsberg, F.;
Bauer, C. K.; Schwarz, J. R. Pflugers Arch. 2001, 442, 577.
18. Lipinski, C. A. Adv. Drug Delivery Rev. 2001, 46, 3.
4. Nerbonne, J. M.; Kass, R. S. Physiol. Rev. 2005, 85, 1205.