Discovery of novel 4-phenyl-2-(pyrrolidinyl)nicotinamide derivatives as potent Nav1.1 activators

Article abstract of DOI:10.1016/j.bmcl.2019.01.023

The voltage-gated sodium channel, Na_v1.1, is predominantly expressed in parvalbumin-positive fast spiking interneurons and has been genetically linked to Dravet syndrome. Starting from a high throughput screening hit isoxazole derivative 5, mod

Full text of DOI:10.1016/j.bmcl.2019.01.023

Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel 4-phenyl-2-(pyrrolidinyl)nicotinamide derivatives as
potent Na 1.1 activators
v
a,⁎
a
a
a
a
a
Tohru Miyazaki , Masanori Kawasaki , Atsushi Suzuki , Yuki Ito , Akio Imanishi ,
b
b
Takamitsu Maru , Tomohiro Kawamoto , Tatsuki Koike
a
b
Research, Takeda Pharmaceutical Company Ltd., Fujisawa, Kanagawa 251-8555, Japan
Axcelead Drug Discovery Partners, Inc., Fujisawa, Kanagawa 251-0012, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
The voltage-gated sodium channel, Na 1.1, is predominantly expressed in parvalbumin-positive fast spiking
v
Voltage-gated sodium channels
Dravet syndrome
interneurons and has been genetically linked to Dravet syndrome. Starting from a high throughput screening hit
isoxazole derivative 5, modifications of 5 via combinations of IonWorks and Q-patch assays successfully iden-
v
Na 1.1 activator
tified the nicotinamide derivative 4. Its increasing decay time constant (tau) of Na 1.1 currents at 0.03 μM along
v
Slow current decay of inactivation
BBB penetration
with significant selectivity against Na
v
1.2, Na 1.5, and Na 1.6 and acceptable brain exposure in mice was ob-
1.1 activator that can be used to analyze pathophysiological functions of
1.1 channel towards treating various central nervous system diseases.
v
v
served. Compound 4 is a promising Na
v
the Na
v
1
4
Na
v
1.1 is a voltage-gated sodium channel (Na
v
), comprising one
cardiac sodium channel and Na
v
1.2 is dominantly expressed in ex-
Therefore, high selectivity against Na 1.5 and
1.2 is preferable for drug candidates. On the other hand, the elec-
trophysiology data regarding Lu AE98134 (3) reveals promising po-
tency as a Na 1.1 activator for increasing the excitability of FSINs.
Herein, we report the discovery of a 4-phenyl-2-(pyrrolidinyl)nicoti-
namide derivative 4 as a highly potent Na 1.1 activator with improved
selectivity against Na 1.2 and Na 1.5 compared with previously re-
ported Na 1.1 activators.
To identify novel chemical lead compounds for Na 1.1 activators, a
high-throughput screening (HTS) was conducted at a single con-
1
5,16
pore-forming α-subunit encoded by SCN1A and two associated β-sub-
citatory neurons.
Na
v
units encoded by SCN1B-SCN4B. Na
v
1.1 as well as its subfamilies
v
(
Na 1.2, Na 1.3 and Na 1.6), is predominantly expressed in the central
v
v
v
1
,2
nervous system (CNS). Na
v
1.1 is largely expressed in parvalbumin-
v
positive fast spiking interneurons (FSINs) and is involved in membrane
3
depolarization and action potential (AP) firing. Therefore, loss of
v
function of the Na
v
1.1 channels could lead to disinhibition of excitatory
v
v
4
–6
pyramidal neurons causing various diseases of the CNS.
Dravet
v
syndrome is a rare genetic epileptic encephalopathy, where more than
v
7
0% of patients have de novo heterozygous mutations of the SCN1A
7
gene. In these mutations, a loss of function of the Na
v
1.1 channels has
centration (30 μM) using Na 1.1-expressing CHO cells on the IonWorks
v
8
been reported. The genetic link between Dravet syndrome patients and
platform followed by confirmation by Q-patch assays. From the HTS,
Na
v
1.1 channels suggest that a brain penetrant Na
v
1.1 activator could
isoxazole derivative 5 was identified as a hit compound which in-
possess significant therapeutic potential for treating Dravet syn-
creased the decay time constant, tau value (179%), of Na 1.1 currents
v
9
,10
drome.
been reported to date. Recently, a few Na
reported by Lundbeck: a 2-methylbenzamide derivative (1), AA43279
However, potent and selective Na
v
1.1 activators have not
relative to the control current at 10 μM without significant inhibition of
17
v
1.1 activators have been
the peak current (top peak current ratio = 86%, Table 1). An initial
structure–activity relationships (SAR) campaign identified pyrimidine
derivative 6 with an increased tau value (198%) at 1 μM by Q-patch
with lead-like physicochemical properties (LogD = 2.97 (pH7.4) and
solubility > 100 μg/mL (pH6.8)). Parallel relationships between in-
creasing tau values by the two methods were observed in 5 and 6. Thus,
we initially estimated the tau value of the tested compounds by Ion-
1
1
1
2
13
(
2), and Lu AE98134 (3) (Fig. 1). The most recently developed
activator, Lu AE98134 (3), increases the total area under the curve for
the duration of the depolarizing pulse from 1 μM in Na 1.1-expressing
1.5 and moderate
1.2 were observed. Biologically, Na 1.5 is a major
v
HEK cells, while issues of low selectivity against Na
v
selectivity against Na
v
v
⁎
Corresponding author.
E-mail address: tohru.miyazaki@takeda.com (T. Miyazaki).
https://doi.org/10.1016/j.bmcl.2019.01.023
Received 2 December 2018; Received in revised form 16 January 2019; Accepted 17 January 2019
0960-894X/ © 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: Miyazaki, T., Bioorganic & Medicinal Chemistry Letters, https://doi.org/10.1016/j.bmcl.2019.01.023

T. Miyazaki et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Table 2
SAR of the amine moiety at the C-6 position of the pyrimidine core.
Cmpd
R
hNa
v
1.1 tau (%) by
hNa
v
1.1 tau (%) by Q-
a
b
IonWorks
patch
1 μM
NDc
3
μM
10 μM
7
110 (7)
105 (15)
Fig. 1. Small Molecule Nav1.1 Activators.
8
9
140 (14)
257 (1 0 2)
ND
111 (9)
96 (7)
90 (18)
74 (18)
ND
ND
Works with moderate throughput (80–100 data points, e.g. 30 com-
pounds with 3 concentrations per assay) followed by low throughput Q-
patch tests (12 data points per assay) to evaluate precise tau values of
the tested compounds.
1
1
1
1
0
1
2
3
257 (51)d
145 (7)
342 (167)
682 (235)
ND
Using 6 as a starting point, more potent compounds were in-
vestigated with increasing tau value at 10 nM. Initially, we investigated
a cyclic amine substituent at the C-6 position of the pyrimidine ring of 6
162 (64)d
(
Table 2). Regarding the ring size at the C-6 position, the 5-membered
pyrrolidine derivative 8 was preferable to a 6-, 7-, or 4-membered 9.
The ring-opened derivative 10 did not increase tau value compared
with the corresponding cyclic 8. Consequently, the introduction of
fluorine atoms on the pyrrolidine ring of 8 successfully identified a (S)-
184 (18)d
131 (31)
ND
3
1
-fluoropyrrolidine substituent in 11 with a higher tau value (682% at
μM) by Q-patch than that of 6 (198% at 1 μM).
a
Tau values in IonWorks assays are averages of N = 5 unless noted.
Standard deviation is provided in parentheses.
After identification of the (S)-3-fluoropyrrolidine moiety at the C-6
b
Tau values in Q-patch experiments are averages of N = 4 unless noted.
position of the pyrimidine ring, we replaced the 4-chlorophenyl moiety
Standard deviation is provided in parentheses.
c
d
of 11 to further improve potency (Table 3). We initially removed the
ND: not determined.
N = 4.
Table 1
Biological and Physicochemical Properties of HTS Hit Compound 5 and Lead Compound 6.
Cmpd
hNa
v
1.1 tau (%) by IonWorksa
10 μM
hNa
v
1.1 tau (%) and top current ratio (%) by Q-patchb
LogDe
Solubility (μg/mL)f
3
μM
30 μM
1 μM
3 μM
10 μM
c
c
d
5
6
99 (6)
102 (8)
138 (73)
555 (1 2 6)
113 (3)
120 (6)
179 (45)
3.43
2.97
0.90
d
9
2 (4)
86 (6)
86 (9)
258 (21)
370 (90)
198 (24)
13 (15)
731 (1 0 3)
115 (8)
774 (1 4 9)
131 (6)
> 100
1
a
Tau values in IonWorks assays are averages of N = 5 unless noted. Standard deviation is provided in parentheses.
b
c
d
e
f
Tau and top peak current values in Q-patch experiments are averages of N = 4 unless noted. Standard deviation is provided in parentheses.
N = 5.
N = 7.
LogD at pH7.4.
Solubility in pH6.8.
2

T. Miyazaki et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Table 3
Table 4
SAR of aryl moiety at the C-4 position of the pyrimidine core.
SAR of the core skeleton.
Cmpd
R
hNa
v
1.1 tau (%) by
hNa
v
1.1 tau (%) by Q-
Cmpd
Core skeleton
hNa
μM
v
1.1 tau (%) by IonWorksa
10 μM
a
b
IonWorks
patch
3
1
μM
3 μM
0.3 μM
170 (28)
1 μM
6
258 (21)
374 (81)
313 (46)
370 (90)
1
1
1
1
4
5
107 (10)
102 (35)
135 (36)
257 (51)c
682 (2 3 5)
302 (33)
391 (78)
255 (52)
332 (1 5 8)
2
3
4
767 (318)b
699 (370)
711 (2 4 7)
1402 (2 1 3)
1
6
111 (7)
271 (44)
NDd
ND
2
1
7
8
93 (5)
197 (17)
ND
ND
1
260 (25)
714 (105)
853 (130)
1904 (74)
25
97 (5)
107 (24)
110 (29)
2
6
99 (10)
1
9
114 (14)
298 (35)
ND
ND
2
2
0
1
80 (3)
86 (5)
115 (10)
98 (8)
ND
ND
ND
ND
a
Tau values in IonWorks assays are averages of N = 5 unless noted.
Standard deviation is provided in parentheses.
b
N = 4.
2
2
80 (4)e
87 (8)e
ND
ND
other hand, replacement of the phenyl ring at the C-4 with furan rings
(20 and 21) or a pyrazole ring (22) did not increase the tau value up to
a
Tau values in IonWorks assays are averages of N = 5 unless noted.
3
μM in IonWorks assays. Thus, it was concluded that 2,5-di-
Standard deviation is provided in parentheses.
b
fluorophenyl was the best substituent at the C-4 position.
Tau values in Q-patch experiments are averages of N = 4 unless noted.
In parallel with optimization of the substituent at the C-4 and C-6
positions of the pyrimidine ring, the pyrimidine core of 6 was replaced
with three types of pyridine rings (one of X or Y or Z = N) and a ben-
zene ring (X, Y , Z = CH, Table 4). The pyridine derivatives 23 and 24
exhibited higher tau values than 6, whereas the pyridine derivative 25
and benzene derivative 26 did not increase the tau value up to 10 μM.
This could suggest that the presence of a hydrogen bonding acceptor
such as a nitrogen atom on the 1- or 2-position of the 6-membered core
Standard deviation is provided in parentheses.
c
d
e
N = 4.
ND: not determined.
N = 8.
chlorine atom of 11 to yield 14, which exhibited an increased tau value
(
225% at 0.3 μM) equivalent to 11 (170% at 0.3 μM) as determined by
Q-patch assays. Next, fluorine atoms were introduced into the phenyl
ring of 14 to provide 15–19. Among them, the 2,5-difluoro substitution
pattern significantly enhanced the tau value (853%) at 0.3 μM. On the
is important for activity as a Na 1.1 activator. By replacing the pyr-
v
imidine core of 6, we identified the two promising pyridine ring can-
didates (Y = N or Z = N) as a core skeleton.
3

T. Miyazaki et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Table 5
SAR of the derivatives of compound 4.
Cmpd Core skeleton
hNa
v
1.1 tau (%) by
hNa
v
1.1 tau (%) by Q-
a
b
IonWorks
patch
0
.3 μM
1 μM
0.03 μM
124 (13)
0.1 μM
Fig. 2. Subtype selectivity of 4 towards other Na 1.X channels in Q-patch as-
v
says by increasing concentrations of the compound (N ≥ 4). Standard deviation
is indicated by error bars.
1
2
4
8
135 (19)
92 (12)
260 (25)
193 (83)
7
124 (6)
NDc
ND
(255%) than 18 at 0.03 μM, as determined by Q-patch assay. The
identification of 4 with 10 nM order activity encouraged us to further
evaluate the in vivo brain pharmacokinetics (PK) and subtype selectivity
against other sodium channels.
Intraperitoneal administration (30 mg/kg) of compound 4 in mice
resulted in sufficient brain exposure (193 ng/g 1 h after administra-
176 (32)
492 (31)
255 (62)
1204 (290)
1
8
tion), which corresponded to 13 nM of free brain concentration
comparable to the in vitro potency of 4 (Table 6). This suggests that
compound 4 has a potential use as an in vivo tool to investigate whether
this type of Na 1.1 activator can restore Nav1.1 functions and modify
v
a
Tau values in IonWorks assays are averages of N = 5 unless noted.
the disease state in animal models. Moreover, the multidrug resistance
protein 1 (MDR1) membrane permeability of 4 (efflux ratio = 1.0)
predicts that compound 4 can penetrate the blood–brain barrier (BBB)
Standard deviation is provided in parentheses.
b
Tau values in Q-patch experiments are averages of N = 4 unless noted.
Standard deviation is provided in parentheses.
1
9
c
in humans.
ND: not determined.
Finally, the subtype selectivity of 4 against Na
v
1.2, Na
v
1.3, Na
v
1.5,
1.2,
and Na
v
1.6 was evaluated in Q-patch assays (Fig. 2). While 4 activated
Na
Na
v
v
1.3 to a lesser extent than Na 1.1, it had no effect on Na
v
v
2
0
Finally, the optimal 2,5-difluorophenyl substituent at the C-4 posi-
tion and (S)-3-fluoropyrrolidine substituent at the C-6 position were
incorporated into the promising pyridine core (Y = N or Z = N) to af-
ford 27 and 4 (Table 5). While 27 did not increase the tau value at
1.5, or Na
v
1.6 up to 10 μM. The subtype selectivity of 4 was sig-
nificantly improved compared with previously reported Na
v
1.1 acti-
vators.
A typical synthetic route for the pyrimidine amide series is illu-
strated in Scheme 1 starting from commercially available methyl 4,6-
dichloropyrimidine-5-carboxylate to provide representative compound
6. A synthetic route for the nicotinamide derivative 4 from commer-
cially available methyl 2-chloro-4-iodopyridine-3-carboxylate is de-
monstrated in Scheme 2. Further synthetic details are available in the
Supporting Information.
0
.3 μM as determined by IonWorks, 4 exhibited a higher tau value
Table 6
PK Profiles in Mice and MDR1 Permeability of 4.
Exposure in plasma and brain in micea
MDR1 ERd
b
c
Plasma (ng/
mL)
Brain (ng/
g)
K
p
F
u,b
Free brain conc.
(nM)
1 μM
The novel and potent Na 1.1 activator 4 was designed and identified
v
starting from the HTS hit 5. Compound 4 showed sufficient brain ex-
posure in mice, achieving comparable concentrations to its in vitro
9
68
193
0.20 0.029 13
1.0
potency and exhibited improved subtype selectivity against Na
v
1.X
1.1 activators.
1.1 activator for further eva-
a
Exposure was measured at 1 h after 30 mg/kg intraperitoneal (i.p.) ad-
channels compared with previously reported Na
Therefore, compound 4 is a valuable Na
v
ministration.
v
b
c
d
Ratio of the total concentration in brain and that in plasma.
luation of pathophysiological functions of the Na 1.1 channel and has
v
Brain protein binding ratio.
potential for therapeutic treatments of CNS diseases such as Dravet
MDR1 efflux ratios (ER) in human MDR1-overexpressing LLC-PK1 cells at
syndrome.
1
μM of substrate.
4

T. Miyazaki et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Scheme 1. Preparation of pyrimidine derivative 6.
Scheme 2. Preparation of nicotinamide derivative 4.
Acknowledgments
an Scn1a gene mutation. J Neurosci. 2007;27:5903–5914.
4
5
6
.
.
.
Han S, Tai C, Westenbroek RE, et al. Autistic-like behaviour in Scn1a+/− mice and
rescue by enhanced GABA-mediated neurotransmission. Nature. 2012;489:385–390.
Oakley JC, Kalume F, Catterall WA. Insights into pathophysiology and therapy from a
mouse model of Dravet syndrome. Epilepsia. 2011;52(Suppl. 2):59–61.
We appreciate Tatsuru Tomokuni (Takeda Pharmaceutical
Company Ltd.), Takeshi Matsubayashi, and Akihiko Naotsuka
Verret L, Mann EO, Hang GB, et al. Inhibitory interneuron deficit links altered net-
work activity and cognitive dysfunction in Alzheimer model. Cell.
(
Axcelead Drug Discovery Partners, Inc.) for the HTS experiments. We
also thank Masataka Yamamoto and Mika Inoue for conducting parallel
synthesis at Axcelead Drug Discovery Partners, Inc.
2
012;149:708–721.
7. Catterall WA. Sodium channels, inherited epilepsy, and antiepileptic drugs. Ann Rev
Pharmacol Toxicol. 2014;54:317–338.
8
.
Mantegazza M, Gambardella A, Rusconi R, et al. Identification of an Nav1.1 sodium
channel (SCN1A) loss-of-function mutation associated with familial simple febrile
seizures. Proc Natl Acad Sci USA. 2005;102:18177–18182.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2019.01.023.
v
9. Jensen HS, Grunnet M, Bastlund JF. Therapeutic potential of Na 1.1 activators.
Trends Pharmacol Sci. 2014;35:113–118.
1
1
1
1
0. Richards KL, Milligan CJ, Richardson RJ, et al. Selective Nav1.1 activation rescues
Dravet syndrome mice from seizures and premature death. Proc Natl Acad Sci USA.
2018;115:E8077–E8085.
References
1. Crestey F, Frederiksen K, Jensen HS, et al. Identification and electrophysiological
v
evaluation of 2-methylbenzamide derivatives as Na 1.1 modulators. ACS Chem
1
2
3
.
.
.
Catterall WA. Voltage-gated sodium channels at 60: structure, function and patho-
physiology. J Physiol. 2012;590:2577–2589.
Neurosci. 2015;6:1302–1308.
2. Frederiksen K, Lu D, Yang J, et al. A small molecule activator of Nav1.1 channels
increases fast-spiking interneuron excitability and GABAergic transmission in vitro
and has anti-convulsive effects in vivo. Eur J Neurosci. 2017;46:1887–1896.
3. Lybøl Nadia, von Schoubyea NL, Frederiksenb K, et al. The sodium channel activator
Catterall WA. Forty years of sodium channels: structure, function, pharmacology, and
epilepsy. Neurochem Res. 2017;42:2495–2504.
v
Ogiwara I, Miyamoto H, Morita N, et al. Na 1.1 localizes to axons of parvalbumin-
positive inhibitory interneurons: a circuit basis for epileptic seizures in mice carrying
5

T. Miyazaki et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Lu AE98134 normalizes the altered firing properties of fast spiking interneurons in
17. Further biological profiling of 5 including increase of the AP firing activity of PV-
positive fast-spiking GABAergic interneurons will be demonstrated in another
manuscript (in preparation).
Dlx5/6+/− mice. Neurosci Lett. 2018;662:29–35.
1
1
4. Vincent GM. The molecular genetics of the long QT syndrome: genes causing fainting
and sudden death. Annu Rev Med. 1998;49:263–274.
18. Free brain concentration was calculated by the following equation. Free brain
concentration = (Brain exposure/Molecular weight) * Fu,b.
5. Gong B, Rhodes KJ, Bekele-Arcuri Z, Trimmer JS. Type I and type II Na(+) channel
alpha-subunit polypeptides exhibit distinct spatial and temporal patterning, and as-
sociation with auxiliary subunits in rat brain. J Comp Neurol. 1999;412:342–352.
19. Abel S, Beaumont KC, Crespi CL, et al. Potential role for P-glycoprotein in the non-
proportional pharmacokinetics of UK-343,664 in man. Xenobiotica.
2001;31:665–676.
1
v
6. Hu W, Tian C, Li T, Yang M, Hou H, Shu Y. Distinct contributions of Na 1.6 and
Na
v
1.2 in action potential initiation and backpropagation. Nat Neurosci.
20. The detailed selectivity data of 4 at 1 μM is described in the Supporting Information.
2
009;12:996–1002.
6