Dihydropyrazolopyrimidine Inhibitors of KV1.5 (IKur)

Article abstract of DOI:10.1016/j.bmcl.2008.10.099

A series of dihydropyrazolopyrimidine inhibitors of K_V1.5 (I_Kur) have been identified. The synthesis, structure-activity relationships and selectivity against several other ion channels are described.

Full text of DOI:10.1016/j.bmcl.2008.10.099

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6381–6385
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Dihydropyrazolopyrimidine Inhibitors of KV1.5 (IKur
)
*
Wayne Vaccaro , Tram Huynh, John Lloyd, Karnail Atwal, Heather J. Finlay, Paul Levesque,
Mary Lee Conder, Tonya Jenkins-West, Hong Shi, Lucy Sun
Bristol-Myers Squibb, Route 206 & Provinceline Road, PO Box 4000, Princeton, NJ 08543-4000, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of dihydropyrazolopyrimidine inhibitors of KV1.5 (IKur) have been identified. The synthesis, struc-
ture–activity relationships and selectivity against several other ion channels are described.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 9 September 2008
Revised 16 October 2008
Accepted 17 October 2008
Available online 25 October 2008
Keywords:
Dihydropyrazolopyrimidine
KV1.5
IKur
Ion channels
Potassium channel
Atrial fibrillation
Atrial fibrillation and flutter are the most common cardiac
arrhythmias encountered in clinical practice.1 Although some pres-
ently available antiarrhythmic drugs such as ibutilide and dofeti-
lide can convert sustained fibrillation to sinus rhythm, these
drugs prolong the ventricular effective refractory period and there-
by increase the risk for life threatening ventricular arrhythmias.
Thus, there is presently an unmet medical need for safe and effica-
cious treatment of atrial fibrillation and other supraventricular
arrhythmias. The ultrarapid delayed rectifier potassium current
(IKur) is a rapidly activating sustained current that has been identi-
fied in human atrial myocytes, but is not functionally expressed in
human ventricular myocytes.2 IKur plays a significant role in the
repolarization of the atrial action potential and selective inhibition
of this current in human atrial myocytes prolongs action potential
duration.3 Prolongation of the action potential would consequently
prolong the atrial effective refractory period; that is, inhibition of
IKur would produce a class III antiarrhythmic effect. Importantly,
an inhibitor of IKur may have utility against atrial fibrillation with-
out having the attendant ventricular proarrhythmic risk associated
with other class III agents that have potassium channel targets that
are expressed in both the atrium and the ventricle.
It had been reported that the L-type calcium antagonist nifedi-
pine 1 was a weak inhibitor of KV1.5 (Table 1).6 Therefore, a collec-
tion of compounds assembled from Bristol–Myers Squibb ion
channel programs was evaluated for KV1.5 inhibitory activity using
EP voltage clamp techniques. Compounds that inhibited human
KV1.5 in a higher throughput Xenopus oocyte assay were further
evaluated against human KV1.5 in the lower throughput but more
sensitive mammalian cell (mouse fibroblast L929) assay. These ef-
forts identified 2, a compound with modest KV1.5 activity, but
encouraging KV1.5 versus L-type calcium channel selectivity. A
study was undertaken to determine if the KV1.5 activity and selec-
tivity of 2 could be improved by structural modification.
Table 1
Ion channel activity of compounds 1 and 2.9
Cl
O N
2
Cl
N
MeO C
2
CO Me
2
CO Et
2
N
IKur is most likely encoded by the potassium channel gene
KV1.5.4 Since screening compounds against IKur in human atrial
cells is impractical, recombinant human KV1.5 expressed in Xeno-
pus oocytes or mouse fibroblast L929 cells were used to screen for
inhibitors of IKur (KV1.5).5
1
2
N
H
N
H
Compound
L-type calcium HEK293, IC50
,
lM
KV1.5 L929, IC50, lM
1
2
0.06
6.1
9.38
1.1
* Corresponding author. Tel.: +1 609-252-5447.
E-mail address: wayne.vaccaro@bms.com (W. Vaccaro).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.099

6382
W. Vaccaro et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6381–6385
The requisite dihydropyrazolopyrimidines 5-8 were readily
overnight at room temperature. Concentration of the resulting
slurry provides the carboxylic acids which were used without fur-
ther purification. Attempts to purify the acids resulted in complex
mixture formation. In general, Route 1 was used to explore the aryl
group, and Route 2 was used to explore the amide group of 4. Occa-
sionally the prepared dihydropyrazolopyrimidines 5–8 would oxi-
dize to the corresponding pyrazolopyrimidines 9. Oxidation was
not predictable and seemed to be dependent on the nature of the
aryl and amide groups. For most compounds, oxidation was not
an issue and could be avoided by careful handling of dihydropyraz-
olopyrimidines 5–8.
assembled by modification of the Knovenegal reaction as shown
in Scheme 1.5,7,11,12 The ester/amide exchange reaction to prepare
b-ketoamides 3 as shown in Route 1 works well with secondary
amines but gives mixtures requiring purification with primary
amines.8 Thus, mono-substituted amide analogs 5-8 were most of-
ten prepared by Route 2. The seemingly straight forward conver-
sion of dihydropyrazolopyrimidine esters such as
4 to the
corresponding acids proved to be anything but straight forward.
Attempts to convert either the ethyl, 2-(cyano)ethyl or benzyl ester
analogs of 4 to the corresponding carboxylic acid under a variety of
conditions often resulted in multi-component mixtures. Eventually
we found that 4 could be smoothly converted to the corresponding
acid by treatment with 4 M HCl in dioxane followed by stirring
Initial SAR studies identified piperazinyl amides such as 5 as
submicromolar inhibitors of KV1.5 (Table 2). Exploration of the C-
7 position of 5 found that unsubstituted aryl analogs had little
Ar
O
Route 1: To
O
O
ii) ArCHO
N
explore Ar
N
N
NRR
5-8
NRR
i) HNRR
NH
3
N
H
O
O
O
iii)
NH2
iv) ArCHO
Ar
O
Ar
O
Route 2: To
explore NRR
N
NH
N
N
N
O
N
NRR
9
NH2
4
N
H
N
Scheme 1. Reagents and conditions: (i) HNRR, Toluene, reflux, overnight. (ii) ArCHO, piperidine, acetic acid, toluene, azeotrope, overnight. (iii) 3-Aminopyrazole, NaOAc,
DMF, 70 °C, overnight. (iv) ArCHO, 3-aminopyrrazole, NaHCO3, DMF, 70 °C, overnight. (v) 4 M HCl in dioxane, overnight, rt. (vi) HNRR, EDCI, DMAP, CH2Cl2
Table 2
KV1.5 inhibition of dihydropyrazolopyrimidines 5 and pyrazolopyrimidines 9.9
Ar
O
O
N
N
N
N
N
N
N
H
5
F
F
Ar
N
N
N
9
Compound
Ar
KV1.5 Oocytes % Inh, 3
l
M
KV1.5 L929 cells IC50, lM
5a
5b
5c
5d
5e
5f
2-Thiophene
2-(5-Me-thiophene)
2-(5-Cl-thiophene)
2-(3-Me-thiophene)
2-(3-Me-benzthiophene)
2-Furan
7
1
51
29
56
1
0.23
0.20
5g
5h
5i
5j
5k
5l
5m
5n
5o
5p
5q
2-(5-Me-Furan)
2-Benzofuran
Ph
2-ClPh
3-ClPh
4-ClPh
2,3-diClPh
2,4-diClPh
2,5-diClPh
7
33
24
30
68
42
72
50
41
65
62
0.16
0.19
0.44
0.16
0.28
0.82
0.12
0.14
3,4-diClPh
3,5-diClPh

W. Vaccaro et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6381–6385
6383
Table 2 (continued)
Compound
Ar
KV1.5 Oocytes % Inh, 3
lM
KV1.5 L929 cells IC50, lM
5r
5s
5t
5u
5v
5w
5x
5y
5z
5aa
5bb
5cc
5dd
5ee
5ff
5gg
5hh
5ii
5jj
5kk
5ll
5mm
9a
2-MePh
3-MePh
4-MePh
3,4-diMePh
3,5-diMePh
2-MeOPh
3-MeOPh
4-MeOPh
2,3-diMeOPh
3,4-diMeOPh
3-CNPh
4-CNPh
2-NO2Ph
37
57
27
34
58
2
30
11
7
48% (1
0.54
lM)
0.75
0.42
1
11
14
4
15
24
3
3-NO2Ph
4-NO2Ph
2-Pyrridine
2-Quinoline
3-Pyridine
3-Quinoline
2-Napthalene
2,3-(Methylenedioxy)Ph
3,4-(Methylenedioxy)Ph
2,3-diClPh
29
3
17
39
18
26
17
0.65
Table 3
KV1.5 inhibition of dihydropyrrolopyrimidines 6 and pyrazolopyrimidines 9.9
Cl
Cl
Cl
Cl
O
O
N
N
N
NRR
6
N
NRR
9
N
H
N
Compound
NRR
KV1.5 Oocytes % Inh at 3
lM
KV1.5 L929 cells IC50 (lM)
6a
6b
6c
6d
6e
6f
6g
6h
6i
NH-Propyl
NH-Butyl
88 (10
56
48
61
72
45
69
83 (10
62
84 (10
l
M)
0.49
0.44
0.59
0.44
0.07
0.31
0.58
0.17
0.16
0.10
NH-Pentyl
N-(Ethyl)2
N-(Propyl)2
N-(Butyl)2
N-(Pentyl)2
NH-Ph
NH-CH2Ph
NH-(CH2)2Ph
NH-(3-Pyridyl)
l
M)
M)
6j
6k
l
9 (10
lM)
6l
52
0.97
N
6m
6n
67
77
0.66
0.09
N
N
6o
6p
90
17
0.19
N
N
O
(continued on next page)

6384
W. Vaccaro et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6381–6385
Table 3 (continued)
Compound
NRR
KV1.5 Oocytes % Inh at 3
5
lM
KV1.5 L929 cells IC50 (lM)
6q
6r
6s
6t
N
N
N
N
17
N
N
N
82 (10
59
lM)
0.085
0.159
CH O
3
6u
39
32% (1
0.20
lM)
N
CH O
3
6v
9b
88
17
N
N
KV1.5 inhibitory activity at the concentrations tested. However,
certain substituted or fused aryl analogs of 5 were found to be po-
tent inhibitors of KV1.5. For example, thiophene 5a and 5-methyl-
thiophene 5b have little KV1.5 inhibitory activity while 5-
chlorothiophene 5c is a potent inhibitor. Fusion of a phenyl ring
on to the weakly active thiophene 5d or furan 5f provides benzo-
thiophene 5e and benzofuran 5h, both potent inhibitors of KV1.5.
The phenyl derivative 5i was also found to be a weak inhibitor
KV1.5, however phenyl analogs with 3-chloro substitution such
as 5k, 5m, 5p, and 5q were among the most potent inhibitors of
KV1.5 identified in this study. Fusion of an additional phenyl ring
onto 5i provides the napthalene 5kk, and as was seen in the corre-
sponding thiophene (5a, 5e) and furan (5f, 5h) series, improves
activity. As was observed for chloro substituted analogs, substitu-
tion of a methyl group at the 3-position improves activity in the
phenyl series, compare 5i with 5s, 5u–v. Aryl analogs substituted
with more polar functionality (5w–5ff) or heteroaryls (5gg–5ii)
were weak inhibitors of KV1.5. We found that in general 2,3- and
3,4-dichlorophenyl were the preferred 7-position substituents for
KV1.5 inhibitory activity. It was also observed that pyrazolopyrim-
idines 9 were less active inhibitors of KV1.5 than the corresponding
dihydropyrazolo-pyrimidines 5, compare 5m and 9a.
group as in 6s. The distal nitrogen of piperazine 6s is not required
for KV1.5 activity as demonstrated by the 4-phenylpiperidine ana-
log 6t. KV1.5 activity in the pyrrolidine series (6m, 6u–v) suggests
that the nature of the substituent and stereochemistry are impor-
tant determinates of KV1.5 activity. The wide variety of amides that
are tolerated should assist in the fine tuning of specificity for KV1.5,
as well as, the ADMET properties for evaluation of in vivo efficacy.
Both the 2,3- and 3,4-dichlorosubstituted analogs 5m and 5p
were resolved by chiral HPLC to provide the corresponding pure
enantiomers 7a,b and 8a,b (Table 4). Most of the KV1.5 activity
was found to reside in a single enantiomer. The absolute stereo-
chemistry of the more active enantiomer was determined by X-ray
crystallography and circular dichroism. The priority rules designate
Table 4
The effect of stereochemistry on KV1.5 inhibition.9
R
O
N
N
N
Holding the 3,4-dichlorophenyl group constant, we next turned
our attention to the exploration of the amide fragment. Amide SAR
for dihydropyrazolopyrimidine 6 is summarized in Table 3.
N
N
H
Dihydropyrazolopyrimidine 6 was found to be quite tolerant of
variation of the amide group. A variety of mono- and disubstituted
amides are inhibitors of KV1.5. Diisopropylamide 6e was the most
potent analog identified in the acyclic alkylamide series (6a–g). A
series of phenyl substituted amides (6h–j) were also quite active.
However, replacement of the phenyl of 6h with 3-pyridyl, as in
6k, resulted in a reduction in KV1.5 activity. In the cyclic amide ser-
ies (6l–o), piperidine 6n was preferred. Morpholine 6p and piper-
azines 6q and 6r were much less active. KV1.5 activity may be
restored in the piperazine series (6q–s) by the addition of a 4-aryl
F
Compound
Stereo
R
KV1.5 Oocytes %
KV1.5 L929 cells
IC50
Inh, 3
l
M
, lM
5m
7a
7b
5p
8a
8b
Racemic
(R)
(S)
Racemic
(S)
(R)
2,3-Cl
2,3-Cl
2,3-Cl
3,4-Cl
3,4-Cl
3,4-Cl
72
2
67
65
19
72
0.16
0.16
0.12
0.07

W. Vaccaro et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6381–6385
6385
Table 5
P.; Gregoratos, G.; Hiratzka, L. F.; Jacobs, A. K.; Russell, R. O.; Smith, S. C.;
Alsono-Garcia, A.; Blomstrom-Lundqvist, C.; Backer, G. D.; Flather, M.; Hradec,
J.; Oto, A.; Parkhomenko, A.; Silber, S.; Torbicki, A. Circulation 2001, 104, 2118.
2. Konarzewska, H.; Peeters, G. A.; Sanguinetti, M. C. Circulation 1995, 92, 1179;
Wang, Z.; Fermini, B.; Nattel, S. J. Pharmacol. Exp. Ther. 1995, 272, 184; Amos, G.
J.; Wettwer, E.; Metzger, F.; Li, Q.; Himmel, H. M.; Ravens, U. J. Physiol. 1996,
491, 31; Li, G.-R.; Feng, J.; Yue, L.; Carrier, M.; Nattel, S. Circ. Res. 1996, 4, 689.
3. Wang, Z.; Fermini, B.; Nattel, S. Circ. Res. 1993, 73, 1061.
4. Feng, J.; Wible, B.; Li, G. R.; Wang, Z.; Nattel, S. Circ. Res. 1997, 80, 572.
5. Atwal, K. S.; Vaccaro, W.; Lloyd, J.; Finlay, H. J.; Lin, Y.: Bandarau, R. S. PCT Int.
Appl. WO2001040231, 2001.
6. Grissmer, S.; Nguyen, A. N.; AIyar, J.; Hanson, D. C.; Mather, R. J.; Gutman, G. A.;
Karmilowicz, M. J.; Auperin, D. D.; Chandy, K. G. Mol. Pharmacol. 1994, 45,
1227.
Ion channel selectivity of 7b and 8b.9
R
O
H
N
N
N
N
N
H
F
7. All compounds gave suitable NMR and LC/MS data.
8. Witzeman, J. S.; Nottingham, W. D. J. Org. Chem. 1991, 56, 1713.
9. All data is an average of 2–4 determinations.
Compound
7b
8b
10. Sodium channel patch clamp experiments were performed at 0.2, 1 and 4 Hz.
1 Hz results are reported in Table 5.
R
2,3-Cl
IC50 0.16
3,4-Cl
IC50 0.07 lM
69% inh (10
42% inh (10
59% inh (10
21% inh (10
KV1.5
HERG
INa
ICa
IKS
IK1
l
M
11. Preparation of 5–8 via Route 1: Step 1A: A mixture of t-butoxyacetoacetate
(6.8 mL, 45 mmol) and the appropriate disubstituted amine (41 mmol) in
toluene (50 mL) was refluxed overnight. The mixture was cooled to room
temperature, transferred to a separatory funnel, diluted with ethyl ether and
extracted with aqueous HCl (1 M). The HCl extracts were combined and
washed with ethyl ether, made basic (pH 9) with aqueous NaOH (50% w/w)
and extracted with ethyl acetate. The ethyl acetate extracts were combined,
washed with water and brine, dried over anhydrous sodium sulfate, filtered
and concentrated to provide compound 3. Step 1B: A mixture of compound 3
(40 mmol), the appropriate arylaldehyde (45 mmol), piperidine (1.0 ml,
10 mmol), acetic acid (0.59 mL, 10 mmol) in toluene (100 mL) were refluxed
overnight with azeotropic removal of water via a Dean-Stark trap. The mixture
was cooled to room temperature and concentrated in vacuo. Step 1C: The
product of Step 1B (40 mmol) was dissolved in dimethylformamide (100 mL).
3-aminopyrazole (5.1 g, 62 mmol) and sodium acetate (10.1 g, 123 mmol) were
added, and the mixture was stirred at 70 °C overnight (17 h). The reaction was
cooled to room temperature, transferred to a separatory funnel, diluted with
water and ethyl acetate, washed with water (a small amount of methanol was
sometimes added to breakup emulsions that may form) and brine, dried over
anhydrous sodium sulfate and concentrated. The resulting residue was purified
by silica gel chromatography to provide compounds 5–8.
12. Preparation of 5–8 via Route 2: Step 2A: A mixture of t-butoxyacetoacetate
(23.4 mL, 141 mmol), the appropriate arylaldehyde (141 mmol), piperidine
(3.5 ml, 35.3 mmol), and acetic acid (2.01 mL, 35.3 mmol) in toluene (300 mL)
was refluxed overnight with azeotropic removal of water via a Dean-Stark trap.
The mixture was cooled to room temperature and concentrated in vacuo. Step
2B: A mixture of the product of Step 2A (141 mmol), 3-aminopyrazole (17.6 g
212 mmol) and sodium acetate (46.3 g, 564 mmol) in dimethylformamide
(300 mL) was stirred at 70 °C overnight (17 h). The mixture was cooled to room
temperature, transferred to a separatory funnel, diluted with water and ethyl
acetate, washed with water (a small amount of methanol was sometimes
added to breakup emulsions that may form) and brine, dried over anhydrous
sodium sulfate and concentrated. The resulting residue was purified by silica
gel chromatography to provide compound 4. Step 2C: HCl (4 M in dioxane) was
added to compound 4 (2.97 mmol) at room temperature. The resulting thick
reaction mixture was allowed to stir overnight. The mixture was concentrated
in vacuo and used without further purification. Step 2D: The appropriate amine
(0.47 mmol) was added to a suspension of the product of Step 2C (0.32 mmol),
EDCI (0.09 g, 0.47 mmol), DMAP (0.004 g, 0.03 mmol) in dichloromethane
(1 mL). When LC/MS analysis indicated the reaction was complete the mixture
was loaded directly onto a silica cartridge (Worldwide Monitoring Clean-up
cartridge, CUSIL12M6) which had been equilibrated with 100% hexanes.
Elution with 100% hexanes (40 mL), followed by 50% Ethyl acetate/hexanes
(40 mL) and 100% ethyl acetate (70 mL). The purest fractions (TLC analysis)
were combined to give compounds 5–8.
41% inh (10
14% inh (10
55% inh (10
7% inh (10
1% inh (10
l
l
l
M)
M)
M)
l
l
l
l
M)
M)
M)
M)
10
(L)
lM)
lM)
3% inh (10 lM)
an (S)-configuration for 2,3-dichloro analog 7b and an (R)-configu-
ration for the 3,4-dichloro analog 8b. For both compounds 7b and
8b, the C-4 aryl group is oriented back behind the plane of the paper.
Compounds 7b and 8b were evaluated for selectivity versus a
panel of ion channels as reported in Table 5. Compounds 7b and
8b are both greater than 50 fold selective for KV1.5 versus HERG,
INa, ICa (L-type), IKs, and IK1 ion channels. The ion channel selectivity
of these compounds suggests that they may be useful for the treat-
ment of atrial fibrillation without the risk of ventricular
proarrhythmia.
In summary, we have reported the discovery, initial SAR and
optimization studies of a novel series of dihydropyrazolopyrimi-
dine inhibitors of KV1.5 (IKur). Future reports from these labs will
describe our efforts to further improve the potency, selectivity
and ADMET properties that will permit in vivo evaluation of the
described dihydropyrazolopyrimidine class of KV1.5 (IKur
inhibitors.
)
Acknowledgment
We thank Ruth Wexler and Michael Poss for their suggestions in
the preparation of this manuscript.
References and notes
1. Fuster, V.; Ryden, L. E.; Asinger, R. W.; Cannom, D. S.; Crijins, H. J.; Frye, R. L.;
Halperin, J. L.; Kay, N. G.; Klein, W. W.; Levy, S.; McNamara, R. L.; Prystowsky, E.
N.; Wann, L. S.; Wyse, D. G.; Gibbons, R. J.; Antman, E. M.; Alpert, J. S.; Faxon, D.