Identification of (R)-N-(4-(4-methoxyphenyl)thiazol-2-yl)-1- tosylpiperidine-2-carboxamide, ML277, as a novel, potent and selective K v7.1 (KCNQ1) potassium channel activator

Article abstract of DOI:10.1016/j.bmcl.2012.07.060

A high-throughput screen utilizing a depolarization-triggered thallium influx through KCNQ1 channels was developed and used to screen the MLSMR collection of over 300,000 compounds. An iterative medicinal chemistry approach was initiated and from this effort, ML277 was identified as a potent activator of KCNQ1 channels (EC₅₀ = 260 nM). ML277 was shown to be highly selective against other KCNQ channels (>100-fold selectivity versus KCNQ2 and KCNQ4) as well as against the distantly related hERG potassium channel.

Full text of DOI:10.1016/j.bmcl.2012.07.060

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5936–5941
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of (R)-N-(4-(4-methoxyphenyl)thiazol-2-yl)-1-tosylpiperidine-2-
carboxamide, ML277, as a novel, potent and selective Kv7.1 (KCNQ1) potassium
channel activator
Margrith E. Mattmann a,c,d, Haibo Yu e,f, Zhihong Lin e,f, Kaiping Xu e,f, Xiaofang Huang e,f, Shunyou Long e,f
,
Meng Wu e,f, Owen B. McManus e,f, Darren W. Engers a,c,d, Uyen M. Le a,c,d, Min Li e,f,
,
⇑
Craig W. Lindsley a,b,c,d, Corey R. Hopkins a,b,c,d,
⇑
a Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37232, USA
b Department of Chemistry, Vanderbilt University, Nashville, TN 37232, USA
c Vanderbilt Center for Neuroscience Drug Discovery, Vanderbilt University Medical Center, Nashville, TN 37232, USA
d Vanderbilt Specialized Chemistry Center for Probe Development (MLPCN), Nashville, TN 37232, USA
e Department of Neuroscience, High Throughput Biology Center, Johns Hopkins University, Baltimore, MD 21205, USA
f Johns Hopkins Ion Channel Center (JHICC), Johns Hopkins University, Baltimore, MD 21205, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A high-throughput screen utilizing a depolarization-triggered thallium influx through KCNQ1 channels
was developed and used to screen the MLSMR collection of over 300,000 compounds. An iterative medic-
inal chemistry approach was initiated and from this effort, ML277 was identified as a potent activator of
KCNQ1 channels (EC50 = 260 nM). ML277 was shown to be highly selective against other KCNQ channels
(>100-fold selectivity versus KCNQ2 and KCNQ4) as well as against the distantly related hERG potassium
channel.
Received 21 June 2012
Revised 11 July 2012
Accepted 16 July 2012
Available online 2 August 2012
Keywords:
KCNQ1 activator
MLPCN probe
Ó 2012 Elsevier Ltd. All rights reserved.
Potassium channels
Voltage-gated ion channels
ML277
Long QT syndrome (LQTS) is a disorder of the electrical activity
of the heart in which the heart muscles take longer than normal to
recover after each heart beat and is named from the abnormal pat-
tern on an electrocardiogram (EKG).1 In response to exercise or
stress, LQTS can lead to dangerous arrhythmias causing uncontrol-
lable heartbeats which can be fatal. This syndrome can either be
caused by genetic mutations, or from pharmacological interven-
tion. To date, six genes have been identified that are responsible
for LQTS and are subclassified as LQT1–LQT6 depending on the
responsible gene. The two most common forms are LQTS1—caused
by mutations in the KCNQ1 gene (nearly half of the cases), and
LQTS2—caused by human Ether-a-go-go-related gene (hERG)
(responsible for ꢀ35% of the cases).1a KCNQ1 (Kv7.1; KvLQT1) is a
voltage-gated potassium channel gene that plays a key physiolog-
ical role in the repolarization of the cardiac tissue following an
action potential.1c,2 In cardiac muscle cells, KCNQ1 forms a channel
complex with the KCNE1 b-subunit (minK) to produce the delayed
rectifier current IKs, and is responsible, in part, for terminating the
action potential.1c,3 Thus, activators of KCNQ1 channels could be
useful treatments of LQTS.
Unfortunately, there are relatively few selective KCNQ1 channel
activators that have been reported in the literature (Fig. 1).2 The re-
ported small molecule activators range from marketed drugs and
drug-like compounds such as the NSAID, mefenamic acid, 1,4 and
L364,373, 2,5 to nondrug-like molecules such as phenylboronic
acid, 36, and zinc pyrithione (ZnPy), 4.7 These compounds have re-
ported activities in the micromolar range against KCNQ1 and are
either not selective versus other members of the Kv7 family,4,6,7
or have no selectivity data reported.5 Based on this, we were inter-
ested in finding potent and selective activators of KCNQ1 and here-
in we report our efforts toward this endeavor.
As members of the Molecular Libraries Production Center Net-
work (MLPCN), a screen of the >300,000 NIH Molecular Library
Small Molecule Repository (MLSMR) compound collection using a
depolarization-triggered influx assay (Pubchem Assay Identifier,
AID: 2648) to identify activators of the KCNQ1 channel was per-
formed.8 From the initial library screen, 1082 compounds were
deemed ‘hits’ and after a round of triage and selectivity screens
⇑
Corresponding authors. Tel.: +1 615 936 6892; fax: +1 615 936 4381 (C.R.H.);
tel.: +1 410 614 5131; fax: +1 410 614 1001 (M.L.).
E-mail address: minli@jhmi.edu (M. Li).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.07.060

M. E. Mattmann et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5936–5941
5937
Table 1
O
N
N
Resynthesis and synthesized enantiomer activity
O
HO
a
Compd Structure
EC50
(lM)
% Maxb
H
N
N
S
N
H
F
O
O
O
N
N
N
HN
2
1
O
O
O
5
1.69 0.35
1.47 0.58
>30
285.0 30.3
N
S
OH
B
O
S
S
Zn2+
O
OH
N
S
HN
3
4
O
O
O
S
(R)-5
(S)-5
358.9 34.5
N
Figure 1. Literature reported activators of KNCQ1.
(parental CHO cells and KCNQ2 channels), 26 compounds were
identified as KCNQ1 activators. One particular series was consid-
ered a starting ‘hit’ series due to high potency against KCNQ1
(EC50 = 660 nM), chemical tractability and favorable selectivity
profile as deemed by PubChem assay statistics; compound 5 (Pub-
Chem Compound Identifier, CID: 4880560) has been evaluated in
597 assays and shows activity in only four confirmatory assays
(Fig. 2).
As 5 represents a racemic mixture, and initial testing was per-
formed from the DMSO stock solutions, we started our medicinal
chemistry campaign with a resynthesis of the original hit, as well
as both the R- and S-enantiomers (from the commercially available
starting materials). The resynthesized material reconfirmed with
S
HN
O
O
O
S
N
a
EC500s were generated from 8-point concentration response curves in duplicate
with three-fold dilutions starting from the maximal concentration (30 M).
l
b
The % Max values list the percent increase in KCNQ1 currents recorded at
+40 mV at saturating compound concentrations in dose-response experiments for
each compound. These maximal measured increases can vary for each compound
and along with the EC50 would contribute to compound efficacy in tissue or animal
experiments.
slightly lower potency than the original HTS hit (EC50 = 1.7
Interestingly, when the purified enantiomers were evaluated, only
the R-isomer showed any activity (R-isomer, EC50 = 1.5 M; S-iso-
mer, EC50 > 30 M) and was equipotent with the racemic material.
These results were very promising and we initiated an iterative
medicinal chemistry effort aimed at evaluating each of the three
distinct areas of the molecule—the upper amide portion, the piper-
idine, and the lower sulfonamide (see Fig. 1 and Table 1).
The first area of SAR evaluation was the sulfonamide portion of
the molecule. The synthesis of these molecules followed known
synthetic protocols and is shown in Scheme 1. Starting from the
commercially available (R)-N-Boc-2-piperidine carboxylic acid
which was coupled with 4-(4-methoxyphenyl)thiazol-2-amine, 7,
using HATU and DIEA to yield 8. The Boc group was removed
(TFA, DCM) and a library of sulfonamides (or amides) was com-
pleted via sulfonyl chloride coupling under basic conditions.
The SAR evaluation started with replacement of the initial hit
compound phenylsulfonamide portion with substituted aryl and
heteroaryl groups (Table 2). Addition of an ortho-methyl group
lM).
l
S
OH
l
O
(a)
N
HN
O
N
O
Boc
N
S
6
Boc
8
O
(b), (c)
N
HN
O
O
N
S
R
10a-gg,
11a-l
O
Scheme 1. Reagents: (a) HATU, DIEA, DMF, 4-(4-methoxyphenyl)thiazol-2-amine
(7) (42–77%), (b) TFA, DCM, (c) RSO2Cl (or ROCl), TEA, DCM.9
with the propyl group being well tolerated (10d, 700 nM); how-
ever, as the size increased the potency decreased (iso-propyl,
provided an equipotent compound (10a, 1.38 lM), while a meta-
substituted tolyl group provided a modest improvement in po-
tency (10b, 890 nM). Moving the methyl group to the 4-position
resulted in a significant improvement in potency (10c, 260 nM).
The para position tolerated an increase in bulk to a certain degree
10e, 2.59
group with the corresponding trifluoromethyl group led to a dim-
inution in potency for all regioisomers (10g–10i, 1–5 M). Other
substitution groups were tolerated in terms of potency (cyano,
methoxy, fluoro and chloro; 61 M), with the 4-chloro compound
lM; tert-butyl, 10f, >30 lM). Replacement of the methyl
l
l
(10t; 420 nM) being the most potent analog. Adding additional
substituents, such as the addition of a methyl spacer group (ben-
zyl) led to a complete loss of activity (10u–10z). Interestingly,
replacement of the aryl group with 1-isoquinoline led to the dis-
S
O
N
HN
O
O
N
covery of a weak KCNQ1 inhibitor (10aa; IC50 = 8
lM). The 3-pyri-
S
O
dyl replacement was equipotent with the phenyl (10bb; 1.36
l
M);
however, other heteroaryl groups were not active, nor were the
amide replacements for the sulfonamide group (10ee–10gg).
Utilizing the optimized 4-methylphenysulfonamide group,
along with the initial 4-(4-methoxyphenyl)thiazol-2-amine group,
the internal piperidine ring system was evaluated (Table 3). The
syntheses of compounds in Table 3 (11a–l) were completed as out-
5
CID: 4880560
KCNQ1 EC50 = 660 nM
Figure 2. Structure of the initial HTS hit.

5938
M. E. Mattmann et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5936–5941
Table 2
Table 2 (continued)
SAR evaluation of the sulfonamide analogs, 10a–gg
a
Compd
R
EC50
(lM)
% Maxb
20.9
S
O
S
O
N
HN
*
O
O
10m
>30
CN
CN
O
N
R
O
S
% Maxb
358.9
a
*
*
Compd
R
EC50 (lM)
10n
10o
10p
10q
0.88 0.14
1.63 0.10
1.16 0.03
1.03 0.03
132.5
88
O
S
*
O
O
O
(R)-5
1.47 0.58
1.38 0.24
0.89 0.05
O
S
O
O
S
*
*
*
10a
10b
145.6
252
CN
O
S
O
S
*
*
O
231.8
231.9
F
O
S
O
S
O
O
O
10c
10d
0.26 0.03
0.70 0.12
265.9
172
F
O
S
*
*
*
S
O
O
O
10r
10s
0.73 0.09
1.41 0.05
52
Cl
Cl
O
S
O
S
*
*
178.6
O
O
10e
10f
2.59 0.56
109
O
S
*
O
O
S
10t
0.42 0.06
161.0
Cl
O
S
>30
*
*
O
O
Cl
Cl
10u
10v
10w
>30
O
S
*
*
O
O
10g
10h
4.74 0.62
1.00 0.19
131.0
125.1
CF3
O
S
Cl
O
S
>30
F
*
O
CF3
S
O
O
S
*
F
*
>30
20.2
31.1
O
Cl
10i
1.71 1.00
60.4
O
S
O
CF3
10x
10y
10z
>30
>30
>30
O
S
*
*
O
F
10j
1.78 0.30
0.81 0.06
32.2
O
O
S
*
*
Cl
O
O
O
S
O
O
10k
256.6
O
S
O
O
O
S
*
Cl
10l
0.78 0.05
178
O

M. E. Mattmann et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5936–5941
5939
Table 3
Table 2 (continued)
SAR of internal ring system (11a–l)
a
Compd
R
EC50
(l
M)
% Maxb
S
O
S
O
N
HN
*
O
O
10aa
10bb
7.97 0.87
1.36 0.14
(inhibitor)
O
O
O
S
N
O
S
*
87.3
N
*
% Maxb
a
O
S
Compd
EC50
(
lM)
O
N
*
*
10cc
>30
11a
>30
N
N
*
O
S
*
11b
11c
11d
11e
11f
2.86 0.17
>30
64
N
O
*
10dd
>30
*
N
N
*
*
O
*
O
S
4.38 0.87
0.60 0.14
>30
54
88
10ee
10ff
6.13 3.98
47.8
68.0
N
*
*
F
O
O
*
*
N
*
8.16 12.4
>30
*
*
HN
N
*
O
10gg
11g
11h
11i
11j
>30
>30
>30
>30
N
*
a
EC50’s were generated from 8-point concentration response curves in duplicate
M). Activity
definition: The compound will be defined as inactive if the compound exhibits less
than 30% activation at 30 M. Otherwise, the compound will be defined as an
*
with three-fold dilutions starting from the maximal concentration (30
l
N
*
l
*
activator with the calculated EC50 value. EC50 values are expressed as EC50 SD,
using estimated standard deviations provided by the fitting software (Origin 6.0).
b
HN
*
The % Max values list the percent increase in KCNQ1 currents recorded at
+40 mV at saturating compound concentrations in dose-response experiments for
each compound. These maximal measured increases can vary for each compound
and along with the EC50 would contribute to compound efficacy in tissue or animal
experiments.
*
*
*
HN
11k
11l
>30
HN
*
*
lined in Scheme 1. Structural changes around this portion of the
molecule were not well tolerated, as shown with the one carbon
expansion of the ring (11a; >30
system (11b; 2.86 M; 11c; >30
12.9 1.4
121
HN
*
l
M); or the contraction of the ring
lM). Although the pyrrolidine ring
l
a
EC50’s were generated from 8-point concentration response curves in duplicate
with three-fold dilutions starting from the maximal concentration (30 M). Activity
definition: The compound will be defined as inactive if the compound exhibits less
than 30% activation at 30 M. Otherwise, the compound will be defined as an
system was active, 11b was ꢀ10-fold less potent than the piperi-
l
dine analog (10c). Addition of a heteroatom to the ring was some-
l
what tolerated with the morpholine (11d; 4.38 lM) and
activator with the calculated EC50 value. EC50 values are expressed as EC50 SD,
thiomorpholine (11e; 600 nM) producing moderately active com-
pounds. However, the piperazine analog was inactive (11f). Lastly,
moving the nitrogen and amide linker to a 1,3-arrangement was
detrimental to activity as was removal of the ring system and
replacement with straight (or branched) chain linkers (11g–l).
The final SAR library evaluated the upper amide portion of the
molecule and the synthesis of these molecules is outlined in
Scheme 2. For compounds 14a–h, the synthesis started with a sul-
fonamide formation with the commercially available (R)-piperi-
dine-2-carboxylic acid, 12, (RSO2Cl, TEA, DCM) followed by
aniline coupling (R1NH2, HATU, DIEA, DMF) to yield the desired
compounds 14a–h. In order to access compounds 14i–o, a further
Suzuki coupling step was employed. To this end, 13 was coupled
with an appropriately halogen substituted heteroaryl amine, 15
(HATU, DIEA, DMF) as outlined previously. Finally, 16 was sub-
jected to Suzuki cross-coupling conditions (R2B(OH)2, K3PO4,
Pd(OAc)2, THF:H2O, 80 °C) leading to the isolation of the desired
compounds.
using estimated standard deviations provided by the fitting software (Origin 6.0).
b
The % Max values list the percent increase in KCNQ1 currents recorded at
+40 mV at saturating compound concentrations in dose-response experiments for
each compound. These maximal measured increases can vary for each compound
and along with the EC50 would contribute to compound efficacy in tissue or animal
experiments.
The SAR evaluation started with minor changes to the right-
hand phenyl group with replacements for the 4-methoxy group
(Table 4). Unfortunately, alkyl, halogen and nitro replacements of
the 4-methoxy group led to a total loss of activity against KCNQ1
(14a–f). In addition, truncation of the molecule into the thiazole
(14g) or cyclization into the 2-benzo[d]thiazole (14h) were also
not tolerated. Cyclization of the 4-methoxy group into the benzo-
furan moiety, 14i, was tolerated; however there was 5-fold loss
of potency (1.32
4-trifluoromethoxy group was not tolerated (14j). Finally,
modification of the thiazole from the 2,4-substitution pattern to
lM). Even substituting the 4-methoxy with
a

5940
M. E. Mattmann et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5936–5941
1
Table 4
R
O
OH
HN
SAR evaluation of the amide analogs, 14a–x
OH
R1
O
O
HN
O
O
(a)
(b)
N
N
O
S
S
R
O
O
NH
12
O
N
S
O
14a-h
13
R
R
(c)
X
R1
R
EC50 (lM)
% Maxb
265.9
a
Compd
Y
H N
2
S
S
S
S
S
S
S
S
Z
Br
10c
Me
Me
Me
Me
Me
Me
Me
Me
0.26 0.03
>30
O
Y
15
N
N
N
N
N
N
N
X
*
*
*
*
*
*
*
*
Y
Z
X
2
R
Br
Z
HN
14a
14b
14c
14d
14e
14f
HN
O
O
(d)
O
>30
O
N
N
S
S
O
O
14i-o
>30
F
16
>30
R
R
Scheme 2. Reagents and conditions: (a) p-TsCl or p-chlorophenylSO2Cl, TEA, DCM
(R = CH3, 57–72%; R = Cl, 37–43%); (b) R1NH2, HATU, DIEA, DMF; (c) HATU, DIEA,
DMF (R = CH3, 12–79%; R = Cl, 19–75%); (d) R2B(OH)2, K3PO4, Pd(OAc)2, THF:H2O,
80 °C.9
>30
NO2
Cl
>30
14g
>30
2,5-substitution (14k and 14m), or addition of an nitrogen (thiadi-
azole moiety) were not tolerated (14l, 14n–o) highlighting the very
narrow SAR window in this portion of the molecule, suggesting the
4-(4-methoxyphenyl)-2-methylthiazole moiety is an integral part
of the binding on this molecule to KCNQ1.
To further profile our lead compound, 10c, we performed a
number of ancillary pharmacological assays to assess the overall
selectivity of 10c. Selectivity of 10c for blocking or activating KCNQ
channel family members was assessed utilizing automated electro-
physiology methods (AID: 493009) (Table 5). 10c was shown to be
>100-fold selective versus KCNQ2 and KCNQ4 (no significant
N
S
S
14h
14i
Me
Me
>30
N
N
*
*
1.32 0.42
30
O
S
14j
14k
14l
Me
Me
Me
>30
OCF3
O
N
S
*
*
N
27.4 61.3
>30
Inhibitor
N
N
N
change in amplitude, <25% activation up to 30
10c was evaluated against the hERG potassium channel and was
lM). In addition,
O
O
S
*
*
inactive against this channel up to 30 lM. Lastly, 10c was evalu-
ated against Ricerca Biosciences10 Lead Profiling assay which eval-
uates up to 68 GPCRs, ion channels and transports for % inhibition
14m
14n
14o
Me
Me
Cl
>30
>30
>30
S
N
N
N
at 10 lM. 10c was found to bind to only 6 of the 68 assays con-
O
O
S
ducted.11 However, these assays are single-point values and func-
tional assay selectivity will need to be evaluated. This point is
highlighted by the fact that 10c was shown to have 80% inhibition
of the hERG binding assay; however, when evaluated in a func-
*
N
S
*
a
tional assay 10c showed no inhibition up to 30 lM. Based on the
EC50’s were generated from 8-point concentration response curves in duplicate
with three-fold dilutions starting from the maximal concentration (30 M). Activity
definition: The compound will be defined as inactive if the compound exhibits less
than 30% activation at 30 M. Otherwise, the compound will be defined as an
l
potent activation of KCNQ1 and the clean selectivity profile, com-
pound 10c has been declared an MLPCN probe and redesignated
as ML277.12,13
l
activator with the calculated EC50 value. EC50 values are expressed as EC50 SD,
Finally, we evaluated ML277 in our Tier 1 in vitro pharmacoki-
netic assays (Table 6) in order to gauge the metabolic stability and
predicted clearance in rat (r) and human (h). Utilizing rapid equi-
librium dialysis, we determined the protein binding of ML277 in
human and rat plasma. ML277 showed high plasma protein bind-
ing in both human and rat (Fu < 0.01). ML277 was also assessed for
its intrinsic clearance in hepatic microsomes in order to predict the
in vivo clearance (CLHEP) in rat and human. ML277 was unstable to-
wards oxidative metabolism in both species and was therefore pre-
dicted to display high clearance in both human and rat microsomes
(18.0 and 64.7 mL/min/kg, respectively). In order to better under-
stand the metabolic instability of ML277 in vitro, we identified
the primary pathways of biotransformation using mass
spectrometric techniques. The results of this analysis revealed
using estimated standard deviations provided by the fitting software (Origin 6.0).
b
The % Max values list the percent increase in KCNQ1 currents recorded at
+40 mV at saturating compound concentrations in dose-response experiments for
each compound. These maximal measured increases can vary for each compound
and along with the EC50 would contribute to compound efficacy in tissue or animal
experiments.
the principal pathways of metabolism were NADPH-dependent
oxidation of the tolyl moiety and oxidative O-demethylation of
the methoxy group. In an effort to block these sites of oxidation,
and susbequently attenuate the metabolic instability, we replaced
both the tolyl moiety and 4-methoxy groups with 4-trifluorometh-
oxy which resulted in a compound with significantly improved
in vitro clearance (CLHEP, 7.9 and 20.6 mL/min/kg, respectively for

M. E. Mattmann et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5936–5941
5941
Table 5
4. Abitbol, I.; Peretz, A.; Lerche, C.; Busch, A. E.; Attali, B. EMBO J. 1999, 18, 4137.
5. Salata, J. J.; Jurkiewicz, N. K.; Wang, J.; Evans, B. E.; Orme, H. T.; Sanguinetti, M.
C. Mol. Pharmacol. 1998, 53, 220.
Effects of 10c on activation of KCNQ2, KCNQ4 and inhibition of hERG channels. EC50
values are plotted for KCNQ1, KCNQ2 and KCNQ4 channels. IC50 value is given for
hERG
6. Mruk, K.; Kobertz, W. R. PLoS One 2009, 4, e4236.
7. Gao, Z.; Xiong, Q.; Sun, H.; Li, M. J. Biol. Chem. 2008, 283, 22649.
8. Automated electrophysiology assay: KCNQ1 activity was examined in an
electrophysiological assay using the population patch clamp mode on the
IonWorks Quattro™ (MDC, Sunnyvale, CA), an automated patch clamp
instrument. CHO cells stably expressing KCNQ1 channels were freshly
dislodged from flasks and dispensed into a 384-well population patch clamp
(PPC) plate. The cell plating density was 6000 cells/well suspended in the
extracellular solution, composed of (in mM): 137 NaCl, 4 KCl, 1 MgCl2, 1.8
CaCl2, 10 HEPES, and 10 glucose, pH 7.4 adjusted with NaOH. After dispensing,
seal resistance of cells was measured for each well and cells were perforated by
EC50/IC50 (lM)
Compd
KCNQ1
0.26
KCNQ2
>30
KCNQ4
>30
hERG
>30
10c
Table 6
In vitro pharmacokinetic properties of 10c (ML277)
incubation with 50 lg/ml amphotericin B (Sigma, St. Louis, MO), which was
S
dissolved in the internal solution composed of (in mM): 40 KCl, 100 K-
Gluconate, 1 MgCl2, 5 HEPES, 2 CaCl2, pH 7.2 adjusted with KOH. Activity of
KCNQ1 was then measured with the recording protocol as followings. Leak
currents were linearly subtracted by extrapolating from the current elicited by
a 100-ms step to ꢁ90 mV from a holding potential of ꢁ80 mV. During the
voltage pulse protocol, cells were held at ꢁ80 mV, followed by a 2000 ms
depolarization from ꢁ80 mV to +40 mV, and then back to ꢁ80 mV. The
currents were measured at the end of the depolarization pulse to +40 mV
before and after the application of compounds for 3 min. Only cells with a
current amplitude more than 0.6–0.8 (depending on cell lot) nA at +40 mV and
a seal resistance >30 MO were included in the data analysis. Compound effects
were assessed by the percentage changes in the KCNQ1 steady state currents,
which were calculated by dividing the difference between pre- and post-
compound KCNQ1 currents by the respective pre-compound currents in the
same well. No corrections for liquid junction potentials (estimated as ꢁ20 mV
by comparing the KCNQ1 reversal potential with the calculated Nernst
potential for potassium) were applied. The current signal was sampled at
0.625 kHz. A similar protocol was used to evaluate compound effects on KCNQ2
and KCNQ4 channels. For KCNQ4, the cells were depolarized to +40 mV from a
holding potential of ꢁ70 mV. Currents were measured at the step current at
+40 mV. Compound effects on hERG channels were evaluated using an
IonWorks™ automated electrophysiology assay employing CHO cells stably
expressing hERG channels, which is similar to the assay used for KCNQ1
channels with the exceptions noted below. Compound effects were examined
using a pulse protocol consisting of a 100 ms step to ꢁ30 mV, a conditioning
prepulse (2 s duration, 45 mV) followed by a test pulse (2 s duration, ꢁ30 mV)
from a holding potential at ꢁ70 mV. hERG current amplitudes are calculated
from the initial peak currents measured during the test pulses to ꢁ30 mV
minus currents measured during the initial steps to ꢁ30 mV. Leak currents are
estimated using a 100 ms step to ꢁ80 mV from the holding potential (ꢁ70 mV)
and a linear correction is applied to the data.
O
HN
N
O
O
N
S
O
10c, ML277
MW
c Log P
hPPB (% fu)
rPPB (% fu)
hCLHEP (mL/min/kg)
rCLHEP (mL/min/kg)
407.5
5.04
0.6
0.7
18.0
64.7
human and rat). While this compound proved inactive against
KCNQ1, its improved DMPK properties provided valuable metabo-
lism SAR in order to improve the PK properties of ML277.
In conclusion, we report the discovery of a novel and potent
KCNQ1 activator (ML277) after a medicinal chemistry effort stem-
ming hits obtained from a high-throughput screen of the MLSMR
compound collection. The SAR analysis revealed that the (R)-iso-
mer was the active enantiomer and further SAR studies showed
the most active southern portion was the tolyl moiety. Further
studies revealed that the 4-(4-methoxyphenyl)thiazole right-hand
portion was critical for activity as all other replacements were
inactive. Lastly, we have shown that ML277 is selective against
KCNQ2 and KCNQ4. ML277 is an MLPCN probe and as such is freely
available upon request.
9. All final compounds were purified by high-throughput HPLC and characterized
by LCMS and/or 1H NMR and found to be in agreement with their structures
(>95% purity).
10. www.ricerca.com.
11. Ricerca Lead Profiling Resuts (% inhibition at 10 lM, species): Adenosine, A3
(73%, human); Calcium Channel L-type (52%, rat); Cannabinoid, CB1 (97%,
human); Potassium channel, hERG (80%, human); Serotonin (5-
hydroxytryptamine), 5-HT2B (66%, human); Transporter, dopamine (DAT)
(54%, human).
12. (R)-N-(4-(4-methoxyphenyl)thiazol-2-yl)-1-tosylpiperidine-2-carboxamide
(ML277, CID 53347902). To
amine, 2, in DMF (0.65 mmol, 0.4 M) was added in order O-(7-
azabenzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium
hexafluorophosphate
a solution of 4-(4-methoxyphenyl)thiazol-2-
Acknowledgments
(HATU; 0.72 mmol), N,N-diisopropylethylamine (DIEA; 2 mmol), and (R)-1-
(tert-butoxycarbonyl)piperidine-2-carboxylic acid, 1, (0.65 mmol). The
mixture was stirred overnight at room temperature. The reaction mixture
was dissolved in 5ꢂ volume H2O and extracted with dichloromethane (DCM),
then concentrated in vacuo. The product was purified by column
chromatography (ethyl acetate and hexanes, ramp to 50% ethyl acetate). The
Boc-protected intermediate was subsequently dissolved in minimal DCM. To
this was added equivolume TFA and the deprotection proceeded at room
temperature for 2 h, at which time the reaction mixture was concentrated in
vacuo. The final compound was afforded by dissolving the deprotected
intermediate (0.57 mmol) in DCM (0.1 M). To this solution was added
triethylamine (1.14 mmol) and p-toluenesulfonyl chloride (TsCl; 0.63 mmol).
After stirring overnight at room temperature, the reaction mixture was washed
with saturated sodium bicarbonate and brine and passed through a phase
separator. The mixture was concentrated in vacuo and purified by HPLC. 1H
NMR: (400 MHz, CDCl3) d (ppm): 9.98 (br s, 1H), 7.81 (m, 4H), 7.39 (d,
J = 8.1 Hz, 2H), 7.05 (s, 1H), 6.98 (dt, J = 8.9, 2.6 Hz, 2H), 4.75 (d, J = 4.8 Hz, 1H),
4.08 (dd, J = 14.7, 2.2 Hz, 1H), 3.88 (s, 3H), 3.14 (td, J = 14, 2.4 Hz, 1H), 2.48 (s,
3H), 2.33 (d, J = 14 Hz, 1H), 1.63–1.12 (m, 5 H); 13C NMR: (100 MHz, CDCl3) d
(ppm): 168.06, 159.55, 156.86, 149.71, 144.18, 136.77, 130.11, 127.33, 127.19,
126.93, 114.05, 105.90, 56.21, 55.28, 43.94, 23.54, 23.07, 21.55, 19.75. LCMS:
RT = 0.845 min, m/z = 472.1 [M+H]+ (>99% @ 215 and 254 nm).
The authors would like to thank Tammy Santomango, Katrina
Brewer and Kaustubh Kulkarni for technical assistance with the
PK experiments and Nathan Kett and Sichen Chang for the purifica-
tion of compounds. This work was supported, in whole or in part,
by National Institutes of Health and MLPCN grants U54MH084659
(C.W.L.), U54MH084691 (M.L.), and 1R03MH090837-01,
1R03MH090849-01, and 1R03DA031670-01 (M.W.). Vanderbilt is
a member of the MLPCN and houses the Vanderbilt Specialized
Chemistry Center for Accelerated Probe Development. Johns Hop-
kins is a member of the MLPCN and houses the Johns Hopkins
Ion Channel Center.
References
´
1. (a) Herbert, E.; Trusz-Gluza, M.; Moric, E.; Smiłowska-Dzielicka, E.; Mazurek,
U.; Wilczok, T. Med. Sci. Monit. 2002, 8, RA240; (b) Chiang, C.-E.; Roden, D. M. J.
Am. Coll. Cardiol. 2000, 36, 1; (c) Jespersen, T.; Grunnet, M.; Olesen, S.-P.
Physiology (Bethesda) 2005, 20, 408.
13. 10c, VU0458298 (ML277) has been declared
a probe via the Molecular
2. Xiong, Q.; Gao, Z.; Wang, W.; Li, M. Trends Pharmacol. Sci. 2008, 29, 99.
3. Sanguinetti, M. C.; Curran, M. E.; Zou, A.; Shen, J.; Spector, P. S.; Atkinson, D. L.;
Keating, M. T. Nature 1996, 384, 80.
Libraries Probe Production Centers Network (MLPCN) and is available through
the network, see: http://.mli.nih.gov/mli/.