Discovery of Soticlestat, a Potent and Selective Inhibitor for Cholesterol 24-Hydroxylase (CH24H)

pubs.acs.org/jmc

Article

Discovery of Soticlestat, a Potent and Selective Inhibitor for

Cholesterol 24-Hydroxylase (CH24H)

Tatsuki Koike,* Masato Yoshikawa, Haruhi Kamisaki Ando, William Farnaby, Toshiya Nishi,

Etsurou Watanabe, Jason Yano, Maki Miyamoto, Shigeru Kondo, Tsuyoshi Ishii, and Takanobu Kuroita

Cite This: J. Med. Chem. 2021, 64, 12228 −12244

Read Online

sı

*

Metrics & More

Article Recommendations

Supporting Information

ACCESS

ABSTRACT: Cholesterol 24-hydroxylase (CH24H, CYP46A1), a

brain-speciﬁc cytochrome P450 (CYP) family enzyme, plays a role

in the homeostasis of brain cholesterol by converting cholesterol to

24S-hydroxycholesterol (24HC). Despite a wide range of potential

of CH24H as a drug target, no potent and selective inhibitors have

been identiﬁed. Here, we report on the structure-based drug design

(SBDD) of novel 4-arylpyridine derivatives based on the X-ray co-

crystal structure of hit derivative 1b. Optimization of 4-arylpyridine

derivatives led us to identify 3v ((4-benzyl-4-hydroxypiperidin-1-

yl)(2,4′-bipyridin-3-yl)methanone, IC₅₀= 7.4 nM) as a highly

potent, selective, and brain-penetrant CH24H inhibitor. Following

oral administration to mice, 3v resulted in a dose-dependent

reduction of 24HC levels in the brain (1, 3, and 10 mg/kg). Compound 3v (soticlestat, also known as TAK-935) is currently under

clinical investigation for the treatment of Dravet syndrome and Lennox-Gastaut syndrome as a novel drug class for epilepsies.

syndrome (LGS) with a novel mechanism of action.²⁰⁻²²In

this paper, we report the design, synthesis, and discovery of

soticlestat starting from a high-throughput screening (HTS)

INTRODUCTION

■

CYP46A1, also known as cholesterol 24-hydroxylase

(CH24H), is a brain-speciﬁc cytochrome P450 (CYP) family

campaign, followed by further optimization utilizing structure-

enzyme that converts cholesterol to 24S-hydroxycholesterol

based drug design (SBDD).

(24HC) and plays a role in the homeostasis of brain

To design and develop therapeutic CH24H inhibitor, not

cholesterol.¹It has been reported that polymorphism in the

only the potent CH24H inhibitory activity but also the

CH24H gene is associated with the risk of Alzheimer’s disease

selectivity over oﬀ-target CYPs is essential because CYP family

(AD).^2,3In AD patients, 24HC levels in cerebrospinal ﬂuid

enzymes are broadly responsible for drug metabolism as well as

(CSF) are elevated compared with healthy control,⁴⁻⁶and the

hormone synthesis.²³In recent years, crystal structures have

expression of CH24H is reported to be increased in the

been reported for substrate-bound and substrate-free

reactive astrocyte and results in disruption of the glial

CH24H,²⁴and for the CH24H complex with some marketed

glutamate transporter EAAT2 association with lipid rafts.⁷

drugs such as the antifungal agent voriconazole.^25,26Although

The potential role of CH24H in glutamate regulation was also

voriconazole is reported to reduce brain 24HC,²⁷⁻²⁹the

supported in our previous study.⁸It is also suggested that

pharmacological eﬀects are also attributable to other

24HC is involved in the regulation of various receptors such as

mechanisms in the mevalonate pathway, given the voriconazole

the liver X receptors,⁹estrogen receptors,¹⁰retinoid orphan

activity of CYP51 inhibition. Therefore, it was important to

receptors,¹¹and N-methyl-D-aspartate (NMDA) receptors.¹²

identify a potent, highly selective and brain-penetrant CH24H

This mounting evidence links CH24H with various neuro-

inhibitor that merits clinical and preclinical investigation into

logical diseases^2−6,8,13such as AD and epilepsy.

the therapeutic potential of the central CH24H inhibition.

The therapeutic potential of CH24H activation has been

Since no potent and selective CH24H inhibitors have been

extensively investigated;¹⁴⁻¹⁹however, inhibitors of the

CH24H enzyme have not been fully studied as a central

nervous system (CNS) drug target. Recently, we reported that

soticlestat (also known as TAK-935, Figure 1), developed as a

Received: May 12, 2021

Published: August 13, 2021

potent and selective inhibitor for CH24H, shows therapeutic

potential for diseases associated with neural hyperexcitation in

mice.⁸Soticlestat is currently being investigated as a drug for

treatment of Dravet syndrome (DS) and Lennox-Gastaut

https://doi.org/10.1021/acs.jmedchem.1c00864

American Chemical Society

J. Med. Chem. 2021, 64, 12228−12244

12228

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 1. Chemical structures of cholesterol, 24S-hydroxycholesterol (24HC), and soticlestat.

Figure 2. Chemical structure of hit compound 1a and X-ray co-crystal structure of CH24H with hit derivative 1b (PDB ID 7LS4).

Figure 3. Scaﬀold hopping to ﬁnd potent and selective CH24H inhibitors.

described in publication so far, we performed HTS campaign,

where human CH24H inhibitory activities were evaluated by

measuring the amount of [¹⁴C] 24HC produced from [¹⁴C]

cholesterol in the CH24H-expressed cell lysate. As a result,

compound 1a was identiﬁed with moderate CH24H inhibitory

activity (IC₅₀= 350 nM, Figure 2), whereas voriconazole did

not show an activity potent enough for determination of IC₅₀

(>10 000 nM). Subsequent structure−activity relationship

12229

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Scheme 1

a

Reagents and conditions: (a) Pd(PPh₃)₄, Na₂CO₃, 4-pyridineboronic acid 5a, DME/water, reﬂux, overnight; (b) Pd(PPh₃)₄, Na₂CO₃, 5a, DME/

water, MW 140 °C, 1 h; (c) 6 M HCl aq. AcOH, reﬂux, 5 h−overnight; (d) Pd(dppf)Cl₂/CH₂Cl₂, KOAc, bis(pinacolato)diboron, THF/DMSO

(20:1), 80 °C, 5 h; (e) Pd(PPh₃)₄, Na₂CO₃, 4,6-dichloropyrimidine, DME/water, MW 150 °C, 1 h; (f) 10% Pd-C, Et₃N, MeOH, H₂, rt, 1 h; (g)

Pd(OAc)₂, t-Bu₃P/HBF₄, CuCN, K₂CO₃, pyrimidine N-oxide, 1,4-dioxane, MW 150 °C, 2 h.

(SAR) studies identiﬁed compound 1b with improved CH24H

inhibitory activity (IC₅₀= 60 nM), and the co-crystal structure

of compound 1b with CH24H was obtained (Figure 2).

Compound 1b is shown to bind in the active site of CH24H

with direct ligation to the heme iron by its pyridine nitrogen at

a distance of 2.0 Å. In addition to ligation to heme,

hydrophobic interactions such as an interaction with a pocket

under Helix F by the dimethyl moiety of compound 1b was

revealed to be important.

This co-crystal structure of compound 1b with CH24H

inspired us to design 4-arylpyridine derivatives as novel

CH24H inhibitors (Figure 3). 4-Arylpyridine was designed

as a rigid and eﬀective fragment that makes two important

interactions: one is the direct ligation to the heme iron by its

pyridine nitrogen and the other is the hydrophobic interaction

with the pocket under Helix F by its aryl moiety. To validate

this hypothesis, methyl 5-methyl-2-(pyridine-4-yl) benzoate 2a

was designed and synthesized. Compound 2a showed CH24H

inhibitory activity (IC₅₀= 1900 nM), suggesting that the 4-

arylpyridine could be a novel fragment for further CH24H

inhibitor design.

these hydrogen-bonding interactions, we also focused on a

hydrophobic interaction with a large hydrophobic cavity, in

which the steroidal scaﬀold of cholesterol is placed when acting

as a substrate (steroidal cavity). To obtain these additional

interactions, we designed compounds 3a and 3b which have 4-

benzylpiperidine amide or 4-benzylpiperazine amide as a

substituent (Figure 3). A benzyl group on the 4-position of

piperidine or piperazine can access the steroidal cavity as a

hydrophobic substituent. In addition to this hydrophobic

interaction, introduction of a carbonyl group (3c, 3d) as a

hydrogen bond acceptor (HBA) or a hydroxyl group (3e, 3f)

as a hydrogen bond donor (HBD) on the piperidine or

piperazine ring was designed to form additional interaction

with Gly369, Ala474, or heme propionate.

RESULTS AND DISCUSSION

■

Chemistry. The syntheses of the biarylcarboxylic acids 2f−l

are illustrated in Scheme 1. Compounds 2a−e were obtained

by Suzuki coupling reaction of commercially available 2-

bromobenzoates 4a−e with 4-pyridineboronic acid 5a.

Hydrolysis of 2a−e in the presence of hydrochloric acid

aﬀorded acids 2f−j as HCl salts. Compound 2k was obtained

starting from commercially available 2-bromomethylbenzoate

4f. Compound 4f was converted to boronic ester 6 by the

borylation reaction using bis(pinacolato)diboron and Pd-

(dppf)Cl₂. Suzuki coupling reaction of compound 6 with

4,6-dichloropyrimidine, followed by hydrogenation and

Introduction of a substituent on the ester moiety of

compound 2a was designed to improve CH24H inhibitory

activity. Co-crystal structure of compound 1b with CH24H

indicates that CH24H has several residues around the active

site such as Gly369, Ala474, and heme propionate, which can

make further hydrogen bonds with the ligand. In addition to

12230

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Scheme 2

a

Reagents and conditions: (a) DMT-MM, amine 9, DMF, rt, 3 h−overnight; (b) HATU, Et₃N, 9, DMF, rt, overnight.

Scheme 3

a

Reagents and conditions: (a) DMT-MM, amine 9, DMF, rt, 2 days; (b) HATU, Et₃N, 9, DMF, rt, overnight; (c) oxalyl chloride, DMF (cat.),

toluene, rt; Et₃N, 9, THF, rt, overnight; (d) Thionyl chloride, DMF, DME, 90 °C; Et₃N, 9, THF, rt, overnight; (e) NaH, MeI, DMF, 0 °C to rt; (f)

Pd(PPh₃)₄, Na₂CO₃, pyridineboronic acids 5, DME/water, MW 120 or 140 °C, 1 h; (g) Pd(PPh₃)₄, Na₂CO₃, 5, DMF/water, 100 °C, overnight;

(h) Pd(dppf)₂Cl₂·CH₂Cl₂, 2 M Na₂CO₃, 5, diglyme, 120 °C, overnight.

hydrolysis, aﬀorded compound 2k as the HCl salt. Compound

2l was obtained as the di-HCl salt by the coupling reaction of

commercially available methyl 2-chloropyridine-3-carboxylate

4g and pyrimidine N-oxide,³⁰followed by hydrogenation and

hydrolysis.

Compounds 3a, 3d−f, 3h, 3i, 3q−s, 3u, 3w, 3x, and 3z were

obtained by the amide coupling reaction of acids 2f−l with

commercially available amines 9 using 4-(4,6-dimethoxy-1,3,5-

triazin-2-yl)-4-methylmorpholinium chloride (DMT-MM) or

(dimethylamino)-N,N-dimethyl(3H-[1,2,3]triazolo[4,5-b]-

12231

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Scheme 4

a

Reagents and conditions: (a) HATU, Et₃N, t-butylpiperadine-1-carboxylate, DMF, rt, overnight; (b) Pd(PPh₃)₄, Na₂CO₃, 4-pyridineboronic acids

5a, DME/water, MW 150 °C, 1 h; (c) HCl, EtOAc, rt, overnight; (d) NaBH(OAc)₃, benzaldehyde, EtOAc, rt, overnight; (e) HATU, DIPEA,

benzoic acid, DMF, rt, overnight.

pyridine-3-yloxy)-methaniminium hexaﬂuorophosphate

(HATU) as condensation agents (Scheme 2).

In fact, methylation of the hydroxyl group on compound 3f

gives about 30 times decrease in activity (3g, IC₅₀= 74 nM).

Next, the eﬀect of the benzyl group on CH24H inhibition

was investigated by changing the 4-benzyl group on the

piperidine ring while keeping the 4-hydroxyl group on the

piperidine. Removal of the benzyl group (3h, IC₅₀= 7700 nM)

Compounds 3g, 3j−p, 3t, 3v, and 3y were obtained by the

amide coupling reaction of commercially available acids 4h−m

with amines 9 using DMT-MM or HATU as condensation

agents followed by Suzuki coupling reaction with 4-

pyridineboronic acid 5a−d (Scheme 3).

and the replacement of the benzyl group with ethyl (3i, IC₅₀

=

540 nM) or phenyl (3j, IC₅₀= 1200 nM) groups resulted in a

decrease in CH24H inhibitory activity. These results suggested

that the 4-benzyl group on the piperidine ring could access the

steroidal cavity to form hydrophobic interactions with

appropriate distance from the active site. As a result, we

could identify 4-hydroxyl-4-benzylpiperidine as a promising

substituent to form additional interactions with CH24H from a

4-arylpyridine scaﬀold.

Compound 3f was selected as the lead compound for further

optimization, and its co-crystal structure with CH24H was

obtained (Figure 4). Our design concept has been validated by

conﬁrming the four key interactions of compound 3f with

CH24H; that is, direct ligation to the heme iron by pyridine

nitrogen, hydrophobic interaction with the pocket under Helix

F by a tolyl moiety, hydrophobic interaction with the steroidal

cavity by a benzyl group, and a hydrogen bond between

Gly369 and the hydroxyl group. With these features,

compound 3f demonstrated in vitro nanomolar CH24H

inhibitory activity for the human CH24H enzyme.

Lead Optimization of the Biaryl Scaﬀold. Having

identiﬁed 4-hydroxyl-4-benzylpiperidine as a promising sub-

stituent on the 4-arylpyridine scaﬀold, we then looked to

optimize the 4-arylpyridine moiety starting from lead

compound 3f. CH24H inhibitory activities as well as the

calculated lipophilicity (c log P) and CYP3A4 inhibitory

activities (% inhibition at 10 μM) for each compound are

shown in Table 2.

Introduction of a methyl group on the 2-position of the

pyridine ring resulted in a pronounced decrease in the CH24H

inhibitory activity (3k, IC₅₀> 10 000 nM), suggesting that the

methyl group next to the pyridine nitrogen prevented

formation of a ligand to the heme iron. Introduction of a

ﬂuoro group on the 3-position of the pyridine ring slightly

diminished the CH24H inhibitory activity (3l, IC₅₀= 17 nM),

while a methyl group maintained the activity (3m, IC₅₀= 4.9

nM). These results indicate that the electron-withdrawing

Compounds 3b and 3c were obtained starting from 2-

bromo-5-methylbenzoic acid 4h. Compound 4h was coupled

with t-butylpiperazine-1-carboxylate using HATU to generate

compound 11, followed by Suzuki coupling reaction with 4-

pyridineboronic acid 5a and deprotection with hydrochloric

acid to aﬀord compound 12 (Scheme 4). Compound 3b was

obtained by the reductive amination of compound 12 with

benzaldehyde. Compound 3c was obtained by the amide

coupling of compound 12 with benzoic acid.

Lead Generation by Exploration of Heterocyclic

Amide. The results for the human CH24H inhibitory

activities of 5-methyl-2-(pyridine-4-yl)benzamide derivatives

3a−j are shown in Table 1.

4-Benzylpiperidine derivative 3a (IC₅₀= 110 nM) and 4-

benzylpiperazine derivative 3b (IC₅₀= 93 nM) showed

improved CH24H inhibitory activities compared with that of

the lead fragment 2a (IC₅₀= 1900 nM). These results

suggested that the benzyl group on the 4-position of piperidine

or piperazine ring could access the steroidal cavity to make a

hydrophobic interaction in addition to the heme coordination

and hydrophobic interaction by the biaryl scaﬀold. We then

explored an additional interaction with the enzyme by adding

an HBA or HBD on the piperidine or piperazine ring.

Piperazine amide derivative 3c (IC₅₀> 10 000 nM) and

piperazinone derivative 3d (IC₅₀= 89 nM), in which the

carbonyl group was added as an HBA, did not show increased

activities compared with that of the nonsubstituted compound

3b (IC₅₀= 93 nM). On the other hand, introduction of a

hydroxyl group on the 3-position (3e, IC₅₀= 16 nM) and the

4-position (3f, IC₅₀= 2.7 nM) of the piperidine ring resulted in

a signiﬁcant increase of the CH24H inhibitory activities

compared with that of the nonsubstituted compound 3a (IC₅₀

= 110 nM). These results suggested that a new hydrogen

bonding with Gly369, Ala474, or heme propionate could be

obtained by adding a hydroxyl group as an HBD in the ligand.

12232

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a b

,

Table 1. Inhibitory Activities of 3a−j for Human CH24H Enzyme

a

b

Human enzyme. IC₅₀values and 95% conﬁdence intervals (given in parentheses) were calculated from duplicate measurements by a four-

parameter logistic curve using XLﬁt software (IDBS, London, UK).

ﬂuoride lowers the heme ligation ability of the pyridine

nitrogen.

We then obtained SAR on the benzene ring of the biaryl

scaﬀold. Introduction of a methyl group on the 3′-position of

the benzene ring resulted in a decrease in the CH24H

inhibitory activity (3n, IC₅₀= 130 nM), suggesting that the

methyl group on the 3′-position is not tolerated because of the

steric conﬂict with Helix I of CH24H (Figure 4). On the other

hand, introduction of a methyl group on the 4′-position (3o,

IC₅₀= 4.4 nM) and 6′-position (3p, IC₅₀= 7.8 nM) of the

12233

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

which is known as a main CYP enzyme for drug metabolism in

the liver, were evaluated at a concentration of 10 μM (Table

2). Compared with its potent CH24H inhibition (IC₅₀= 2.7

nM), lead compound 3f showed good selectivity against

CYP3A4 with 67% inhibition at 10 μM. However, a greater

degree of selectivity against CYP3A4 was desired to develop

these compounds as clinical drug candidates to avoid drug−

drug interaction (DDI) risk in humans.

Comparison of CYP3A4 inhibitory activity of compound 3k

(60%) with compound 3f (67%) indicates that the

introduction of the methyl group at the 2-position of the

pyridine ring does not aﬀect the CYP3A4 inhibitory activity in

stark contrast to the SAR in CH24H (Table 2). This

observation suggests that this series of CH24H inhibitors

does not bind CYP3A4 with heme ligation, but binds by some

nonspeciﬁc hydrophobic interactions without heme ligation. In

addition, we observed a tendency of reduced CYP3A4

inhibitory activities for compounds 3q−t in correlation with

their lower lipophilicity (c log P) while these compounds

maintain CH24H inhibitory activity regardless of their

lipophilicity. These results suggest that the binding of these

compounds to CYP3A4 is dominated by nonspeciﬁc hydro-

phobic interactions. This observation led us to reduce the

CYP3A4 inhibition by designing less lipophilic compounds

while maintaining the key interactions necessary for CH24H

binding. Based on this strategy to reduce the lipophilicity of

the compounds, we continued to optimize compound 3t,

which showed the lowest CYP3A4 inhibitory activity.

Figure 4. X-ray co-crystal structure of CH24H with 3f (PDB ID

7LS3).

Table 2. CH24H Inhibitory Activities, Calculated

Lipophilicity (c log P), and CYP3A4 Inhibitory Activities of

Compounds 3f and 3k−t

a

CH24H

CYP3A4

Lead Optimization to Improve CYP Selectivity. To

reduce the lipophilicity of the molecule, we replaced the

benzene and/or pyridine rings of compound 3t with more

polar pyridine or pyrimidine rings (Table 3). Because this

design causes only slight changes in the molecular shape, these

compounds were expected to maintain the key interaction with

Gly369 and heme iron of CH24H. Regarding the positioning

of the nitrogen atom on each aryl group, we opted to use the

most sterically hindered positions (X¹, X², or X³), which were

expected to allow the hydrophobic interaction with the pocket

under Helix F and the steroidal cavity to be maintained. Also,

by introducing nitrogen at the sterically hindered positions, we

could limit the polar surface area value, which is beneﬁcial for

better blood−brain barrier (BBB) permeability.

c

b

Cpd

R¹

R²

c log P

IC₅₀(nM)

% (10 μM)

3f

3k

3l

3m

3n

3o

3p

3q

3r

H

5′-Me

3′-Me

4′-Me

6′-Me

5′-Cl

5′-OMe

5′-F

2.8

3.3

3.0

2.5

2.8

3.0

2.5

2.3

2.7 (1.8−4.2)

>10 000

67

60

77

91

77

65

71

75

67

55

45

2-Me

3-F

3-Me

H

17 (13−21)

4.9 (2.6−9.2)

130 (65−260)

4.4 (2.5−7.5)

7.8 (5.4−11)

2.4 (1.7−3.5)

4.7 (2.2−10)

6.1 (4.0−9.2)

7.9 (2.6−24)

3s

3t

H

a

b

Human enzyme. IC₅₀values and 95% conﬁdence intervals (given in

parentheses) were calculated from duplicate measurements using a

Based on this design, compounds with one additional

nitrogen (X¹: 3u, X²: 3v, and X³: 3w) and with two additional

nitrogens (X¹and X²: 3x, X²and X³: 3y, X¹and X³: 3z) were

synthesized. As we expected, these compounds showed lower

lipophilicity (c log P) than lead compound 3t, as well as

improved CYP3A4 inhibitory activities (5−28% inhibition at

10 μM), which were considered low enough to be developed

as candidates. Regarding the CH24H inhibitory activity,

introduction of one nitrogen at X¹(3u, IC₅₀= 4.3 nM) and

X²(3v, IC₅₀= 7.4 nM) positions maintained the activity. On

the other hand, introduction of one nitrogen at the X³(3w,

IC₅₀= 21 nM) position and the introduction of two nitrogens

decreased the activity (3x: IC₅₀= 19 nM, 3y: IC₅₀= 100 nM,

3z: IC₅₀= 36 nM). The clog P values for the compounds 3w−

z (between −0.1 and 0.8) were much lower than that of

compound 3t (2.3), which resulted in lower inhibitory activity

not only for CYP3A4 but also for CH24H.

four-parameter logistic curve using XLﬁt software (IDBS, London,

c

UK). Calculated log P values were obtained using Daylight.³¹

benzene ring maintained CH24H inhibitory activity. SAR of

the 5′-position of the benzene ring was also obtained.

Replacement of the methyl group of compound 3f (IC₅₀

2.7 nM) with chloro (3q, IC₅₀= 2.4 nM), methoxy (3r, IC₅₀

=

4.7 nM), and ﬂuoro (3s, IC₅₀= 6.1 nM) groups maintained the

CH24H inhibitory activity. Interestingly, removal of a methyl

group (3t, IC₅₀= 7.9 nM) also maintained the activity. This

SAR information and a co-crystal structure of compound 3f

indicate that a wide variety of small substituents is acceptable

on the 4′-position, 5′-position, and 6′-position of the benzene

ring; however, contribution of the substituents on these

positions may be minimal and given that the nonsubstituted

compound 3t (IC₅₀= 7.9 nM) showed potent nanomolar

activity.

For compounds 3u and 3v, which showed potent CH24H

inhibitory activity and suﬃcient CYP3A4 selectivity, we

evaluated in vitro metabolic clearance in human hepatic

microsomes. The results showed that compound 3v had

With novel and potent CH24H inhibitors developed by our

SBDD approach, we then focused on the oﬀ-target CYP

selectivity of our compounds. Inhibitory activities for CYP3A4,

12234

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Table 3. CH24H Inhibitory Activities, Lipophilicity (c log P), CYP3A4 Inhibitory Activities, and Human Metabolic Stability of

Compounds 3t−z

a

c

CH24H

CYP3A4

clearance

d

b

Cpd

X¹

X²

X³

c log P

IC₅₀(nM)

% (10 μM)

μL/(min mg)

3t

CH

N

CH

N

CH

N

CH

N

CH

N

CH

N

2.3

1.5

1.4

0.8

0.6

7.9 (2.6−24)

4.3 (2.8−6.5)

7.4 (4.8−11)

21 (14−31)

19 (12−28)

100 (63−180)

36 (23−60)

45

17

28

18

13

5.3

8.8

120

70

22

17

3u

3v

3w

3x

3y

3z

e

CH

N

CH

−0.1

0.0

ND

9

N

a

b

Human enzyme. IC₅₀values and 95% conﬁdence intervals (given in parentheses) were calculated from duplicate measurements using a four-

c

d

parameter logistic curve using XLﬁt software (IDBS, London, UK). In vitro metabolic clearance in human hepatic microsomes. Calculated log P

values were obtained using Daylight.³¹Not determined.

e

superior metabolic stability (22 μL/(min mg)) and therefore

was selected for further characterization including detailed

CYP selectivity studies and a PK/PD study in mice.

Co-crystal structure of compound 3v was obtained and it

was conﬁrmed that compound 3v makes multiple binding

interactions with CH24H, in a manner analogous to the

binding mode obtained with the lead compound 3f (Figure 5).

66 000 nM), but also for other CYP enzymes related to drug

metabolism (2C8, 2C9, 2D6, 1A2, 2C19). To verify the

absence of inhibitory eﬀects of hormone synthesis, several key

steroid hormones, such as cortisol, testosterone, aldosterone,

and corticosterone, were analyzed in the adrenocortical H295R

cell line.^32,33Ketoconazole, known as a CYP3A4/5 inhibitor,

inhibited the entire pathway, whereas 3v did not show any

inhibitory activity on the hormone synthesis in cells (Table 5).

As a result, we selected compound 3v (IC₅₀= 7.4 nM) as a

potent and highly selective CH24H inhibitor.

Table 5. Steroidogenic Disruption in H295R Cells

3v

ketoconazole

hormone

% inhibition (10 μM) (SE)

IC₅₀(μM) (95% CI)

testosterone

corticosterone

cortisol

−0.4 (9.5)

−6.0 (5.0)

0.2 (7.3)

0.34 (0.30−0.39)

2.5 (2.3−2.7)

0.44 (0.39−0.49)

1.4 (1.0−2.2)

aldosterone

−3.8 (10.7)

Eﬀect of Compound 3v on Reducing Brain 24HC

Levels. The eﬀect of compound 3v on the reduction of 24HC

concentrations was evaluated in the mouse brain (1, 3, and 10

mg/kg, po). The 24HC levels in the brain as well as the brain

concentrations of compound 3v after oral administration to

mice are illustrated in Figure 6. Compound 3v showed

adequate brain exposure for CH24H inhibition, and a dose-

dependent concentration of compound 3v was observed in the

brain. Consistent with the brain concentration, compound 3v

showed dose-dependent reduction in 24HC levels in the brain

and exhibited signiﬁcant reduction at 24 h after single-dose

administration (33% reduction at 10 mg/kg and 25% reduction

at 3 mg/kg from control level, po). This result has

demonstrated that compound 3v is a promising CH24H

inhibitory agent with dose-dependent 24HC reduction in the

brain.

Figure 5. X-ray co-crystal structure of 3v with CH24H (PDB ID

7LRL).

The inhibitory activities (IC₅₀) of compound 3v for oﬀ-target

CYPs were evaluated (Table 4). Compared with the potent

CH24H inhibitory activity (IC₅₀= 7.4 nM), compound 3v

showed negligible inhibition not only for CYP3A4 (IC₅₀

=

Table 4. Inhibitory Activity (IC₅₀) of Compound 3v for

Other CYPs

CYP

IC₅₀(nM)

2C8

2C9

2D6

3A4

1A2

2C19

62 000

19 000

>100 000

66 000

>100 000

14 000

CONCLUSIONS

■

Here, we report on the design and synthesis of a novel, potent

and selective CH24H inhibitor 3v (IC₅₀= 7.4 nM) starting

12235

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 6. PK/PD study in C57BL/6N mice (7−8 weeks). Brain concentration of (A) 3v and (B) 24HC level (% control) in mouse brain after oral

administration of 3v at 1, 3, and 10 mg/kg, po. Black (1 mg/kg), white (3 mg/kg), and gray (10 mg/kg) circles represent mean + SD (n = 3).

inence UFLC high-pressure gradient system/LCMS-2020) or Agilent

LC/MS system (Agilent 1200SL/Agilent 6130MS), operating in ESI

(+ or −) or APCI (+ or −) ionization mode. Analytes were eluted

using a linear gradient of 0.05% TFA containing water/acetonitrile or

5 mM ammonium acetate containing water/acetonitrile mobile phase

and detected at 220 nm. Preparative HPLC was performed by an

automated HPLC system or MS-trigger using ODS column with 10−

100% gradient water−acetonitrile containing 0.1% TFA. Elemental

analyses were carried out by Takeda Analytical Laboratories, and the

results were within 0.4% of theoretical values. ¹H NMR spectra were

in all cases consistent with the proposed structures. The purities of all

tested compounds in biological systems were conﬁrmed to be more

than 95% pure as determined by analytical HPLC.

from a hit derivative 1b (IC₅₀= 60 nM) by an SBDD

approach. Compound 3v exhibited potent and selective

inhibition through four key interactions with CH24H. These

interactions include direct ligation to the heme iron by a

pyridine nitrogen, hydrophobic interaction with the pocket

under Helix F by a pyridine ring, hydrophobic interaction with

the steroidal cavity by a benzyl group, and a hydrogen bond

between Gly369 and a hydroxyl group. In addition to these

speciﬁc interactions to CH24H, CYP selectivity of 3v has been

achieved by the adjustment of the lipophilicity of the molecule.

Single oral administration of compound 3v (1, 3, and 10 mg/

kg) to mice exhibited a dose-dependent reduction in 24HC

levels in the mouse brain, proving that it is brain penetrable.

Compound 3v (soticlestat) is currently under clinical

investigation for the treatment of Dravet syndrome and

Lennox-Gastaut syndrome.

Methyl 5-Methyl-2-(pyridine-4-yl)benzoate (2a). To a mixture of

methyl 2-bromo-5-methylbenzoate 4a (712 mg, 3.11 mmol), DME

(15.0 mL), and water (3.00 mL) were added pyridine-4-boronic acid

5a (573 mg, 4.66 mmol), sodium carbonate (659 mg, 6.22 mmol),

and Pd(PPh₃)₄(180 mg, 0.16 mmol) at room temperature. The

mixture was reﬂuxed under N₂overnight. The reaction mixture was

diluted with ethyl acetate (EtOAc), ﬁltered with silica pad, and

concentrated in vacuo. The residue was puriﬁed by column

chromatography (silica gel, 30−50% EtOAc/hexane) to give 2a

EXPERIMENTAL SECTION

■

General. All commercially available reagents and solvents were

used without further puriﬁcation. Yields were not optimized. All

reactions were monitored by thin-layer chromatography (TLC)

analysis on Merck Kieselgel 60 F254 plates or Fuji Silysia NH plates,

or liquid chromatography−mass spectrometry (LC−MS) analysis.

Microwave-assisted reactions were carried out in a single-mode

reactor, Biotage Initiator 2.0 or 2.5 microwave synthesizer. Proton

Nuclear Magnetic Resonance (¹H NMR) spectra were recorded on

Varian Mercury-300 (300 MHz), Varian (400 MHz), Bruker DPX300

(300 MHz), Bruker Avance III (300 MHz), or Bruker Advance III

plus (400 MHz) spectrometer. Chemical shifts are given in parts per

million (ppm) downﬁeld from tetramethylsilane (δ) as the internal

standard in deuterated solvent, and coupling constants (J) are in hertz

(Hz). Data are reported as follows: chemical shift, integration,

multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, quint =

quintet, m = multiplet, dd = doublet of doublets, td = triplet of

doublets, and brs = broad signal), and coupling constants. Analytical

TLC was performed on silica gel 60 F₂₅₄plates (Merck) or NH TLC

plates (Fuji Silysia Chemical Ltd.). Extraction by organic solvent was

monitored by TLC. Chromatographic puriﬁcation was performed on

silica gel columns [(Merck Kieselgel 60, 70-230 mesh size or 230−

400 mesh size, Merck) or (Chromatorex NH-DM 1020, 100−200

mesh size)] or on Purif-Pack (SI or NH, Shoko Scientiﬁc). Melting

1

(548 mg, 2.41 mmol, 78%) as a colorless oil. H NMR (CDCl₃) δ

2.44 (3H, s), 3.65 (3H, s), 7.22 (3H, dd, J = 6.6, 4.7 Hz), 7.39 (1H, d,

J = 8.3 Hz), 7.73 (1H, s), 8.61 (2H, d, J = 6.0 Hz); MS (ESI) m/z:

228 (M + H)⁺.

Methyl 5-Chloro-2-(pyridin-4-yl)benzoate (2b). Compound 2b

(1.46 g, 5.89 mmol, 49%) was obtained as a colorless oil by a method

1

similar to that described for 2a. H NMR (CDCl₃) δ 3.68 (3H, s),

7.17−7.30 (3H, m), 7.55 (1H, dd, J = 8.2, 2.2 Hz), 7.91 (1H, d, J =

2.2 Hz), 8.61−8.67 (2H, m); MS (ESI) m/z: 248 (M + H)⁺.

Methyl 5-Methoxy-2-(pyridin-4-yl)benzoate (2c). A mixture of

methyl 2-bromo-5-methoxybenzoate 4c (3.00 g, 12.2 mmol),

Pd(PPh₃)₄(0.707 g, 0.610 mmol), sodium carbonate (2.59 g, 24.5

mmol), pyridine-4-boronic acid 5a (1.81 g, 14.7 mmol), DME (16.0

mL), and water (4.0 mL) was heated at 140 °C for 1 h under

microwave irradiation. The mixture was quenched with water at room

temperature and extracted with EtOAc. The organic layer was

separated, washed with brine, dried over Na₂SO₄, and concentrated in

vacuo. The residue was puriﬁed by column chromatography (NH

silica gel, 5−20% EtOAc/hexane) to give 2c (1.14 g, 4.69 mmol,

1

38%) as a pale yellow oil. H NMR (CDCl₃) δ 3.66 (3H, s), 3.89

(3H, s), 7.11 (1H, dd, J = 8.3, 2.6 Hz), 7.16−7.22 (2H, m), 7.23−

7.30 (1H, m), 7.42 (1H, d, J = 2.6 Hz), 8.56−8.63 (2H, m); MS

(ESI) m/z: 244 (M + H)⁺.

̈

points were determined on a Buchi melting point apparatus B-545.

LC−MS analysis was performed on Shimadzu UFLC/MS (Prom-

12236

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Methyl 5-Fluoro-2-(pyridin-4-yl)benzoate (2d). Compound 2d

60.0 mmol) was reﬂuxed for 5 h. The mixture was concentrated in

vacuo. The residual solid was washed with EtOAc to give 2k (0.446 g,

1.89 mmol, 82%) as a beige solid. ¹H NMR (DMSO-d₆) δ 7.57−7.71

(3H, m), 7.75 (1H, dd, J = 5.3, 1.5 Hz), 7.78−7.85 (1H, m), 8.87

(1H, d, J = 5.3 Hz), 9.22 (1H, s), 10.59−11.85 (1H, brs).

(1.08 g, 4.65 mmol, 30%) was obtained as a white solid by a method

1

similar to that described for 2c. H NMR (CDCl₃) δ 3.69 (3H, s),

7.16−7.26 (2H, m), 7.31 (2H, ddt, J = 7.1, 3.7, 1.6 Hz), 7.55−7.74

(1H, m), 8.44−8.85 (2H, m).

Ethyl 2-(Pyridin-4-yl)benzoate (2e). Compound 2e (2.84 g, 12.5

mmol, 48%) was obtained as a white solid by a method similar to that

6-[3-(Methoxycarbonyl)pyridin-2-yl]pyrimidine 1-Oxide (8). A

mixture of methyl 2-chloropyridine-3-carboxylate 4g (1.52 mL, 11.7

mmol), potassium carbonate (3.22 g, 23.3 mmol), pyrimidine N-oxide

(3.36 g, 35.0 mmol), copper(I) cyanide (0.104 g, 1.17 mmol),

Pd(OAc)₂(0.131 g, 0.580 mmol), and P(t-Bu)₃·HBF₄(0.507 g, 1.75

mmol) in dioxane (20.0 mL) was heated to 150 °C under microwave

irradiation for 2 h. The mixture was diluted with EtOAc and the

precipitate was ﬁltered through celite. The ﬁltrate was concentrated in

vacuo. The residue was puriﬁed by column chromatography (silica

gel, 30−100% EtOAc/hexane) to give 8 (0.480 g, 2.08 mmol, 18%) as

1

described for 2c. H NMR (CDCl₃) δ 1.04 (3H, t, J = 7.2 Hz), 4.12

(2H, q, J = 7.2 Hz), 7.18−7.26 (2H, m), 7.32 (1H, dd, J = 7.5, 1.1

Hz), 7.44−7.53 (1H, m), 7.53−7.63 (1H, m), 7.93 (1H, dd, J = 7.5,

1.1 Hz), 8.59−8.66 (2H, m); MS (ESI) m/z: 228 (M + H)⁺.

5-Methyl-2-(pyridin-4-yl)benzoic Acid Hydrochloride (2f). A

mixture of 2a (8.81 g, 38.8 mmol), 6 M HCl (64.6 mL, 388

mmol), and AcOH (100 mL) was reﬂuxed for 5 h. The solvent was

removed in vacuo, and the obtained precipitate was washed with

EtOAc and ﬁltered to give 2f (6.55 g, 26.2 mmol, 68%) as a white

solid. ¹H NMR (DMSO-d₆) δ 2.44 (3H, s), 7.38 (1H, d, J = 7.9 Hz),

7.55 (1H, d, J = 7.9 Hz), 7.82 (1H, s), 7.90 (2H, brs), 8.78−8.95 (2H,

m); MS (ESI) m/z: 214 (M + H)⁺−(HCl).

5-Chloro-2-(pyridin-4-yl)benzoic Acid Hydrochloride (2g). Com-

pound 2g (1.36 g, 5.03 mmol, 86%) was obtained as a white solid by a

method similar to that described for 2f. ¹H NMR (DMSO-d₆) δ 7.51

(1H, d, J = 8.2 Hz), 7.83 (1H, dd, J = 8.2, 2.2 Hz), 7.98 (3H, s),

8.74−9.09 (2H, m).

1

a brown solid. H NMR (CDCl₃) δ 3.87 (3H, s), 7.54 (1H, dd, J =

7.9, 4.9 Hz), 7.71 (1H, dd, J = 4.9, 0.8 Hz), 8.28 (1H, dd, J = 7.9, 1.5

Hz), 8.35 (1H, d, J = 4.9 Hz), 8.85 (1H, dd, J = 4.9, 1.5 Hz), 9.00

(1H, d, J = 0.8 Hz); MS (ESI) m/z: 232 (M + H)⁺.

2-(Pyrimidin-4-yl)nicotinic Acid Dihydrochloride (2l). Compound

2l (0.260 g, 0.949 mmol, quant) was obtained as a brown oil by a

method similar to that described for 2k. MS (ESI) m/z: 202 (M +

H)⁺.

(4-Benzylpiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)phenyl)-

methanone (3a). A mixture of 2f (300 mg, 1.20 mmol), 4-

benzylpiperidine 9a (0.232 mL, 1.32 mmol), DMT-MM (399 mg,

1.44 mmol), and DMF (5.00 mL) was stirred at room temperature for

3 h. Additional 4-benzylpiperidine (0.232 mL, 1.32 mmol) was added

to this solution. After stirring overnight, the mixture was quenched

with saturated aqueous NaHCO₃at room temperature and extracted

with EtOAc. The organic layer was separated, washed with brine,

dried over Na₂SO₄, and concentrated in vacuo. The residue was

puriﬁed by column chromatography (NH silica gel, 30−50% EtOAc/

hexane) to give 3a (287 mg, 0.775 mmol, 65%) as a colorless oil. ¹H

NMR (CDCl₃) δ 0.37−1.04 (1H, m), 1.06−1.67 (3H, m), 2.03−2.76

(8H, m), 3.07−3.27 (1H, m), 4.64 (1H, t, J = 12.4 Hz), 6.92−7.55

(10H, m), 8.52−8.76 (2H, m); ¹³C NMR (CDCl₃) δ 18.43, 21.09,

21.13, 30.92, 31.52, 32.09, 37.60, 37.83, 41.56, 41.86, 42.72, 46.48,

47.06, 58.30, 123.23, 123.82, 125.98, 126.06, 128.17, 128.22, 128.25,

128.27, 128.88, 128.94, 128.99, 130.23, 130.35, 132.49, 132.89,

135.59, 135.85, 139.42, 139.77, 147.55, 147.64, 149.77, 149.82,

169.08, 169.13; MS (ESI) m/z: 371 (M + H)⁺.

5-Methoxy-2-(pyridin-4-yl)benzoic Acid Hydrochloride (2h).

Compound 2h (1.19 g, 4.48 mmol, 96%) was obtained as a white

solid by a method similar to that described for 2f. ¹H NMR (DMSO-

d₆) δ 3.90 (3H, s), 7.17−7.37 (1H, m), 7.41−7.54 (2H, m), 7.83−

8.02 (2H, m), 8.76−8.98 (2H, m).

5-Fluoro-2-(pyridin-4-yl)benzoic Acid Hydrochloride (2i). Com-

pound 2i (1.10 g, 4.34 mmol, 95%) was obtained as a pale yellow

solid by a method similar to that described for 2f. ¹H NMR (DMSO-

d₆) δ 7.46−7.69 (2H, m), 7.79 (1H, d, J = 2.7 Hz), 7.99 (2H, d, J =

6.6 Hz), 8.84−8.99 (2H, m).

2-(Pyridin-4-yl)benzoic Acid Hydrochloride (2j). Compound 2j

(3.62 g, 15.4 mmol, 76%) was obtained as a white solid by a method

similar to that described for 2f. ¹H NMR (DMSO-d₆) δ 7.49 (1H, dd,

J = 7.6, 1.1 Hz), 7.62−7.84 (2H, m), 7.94−8.12 (3H, m), 8.82−9.04

(2H, m); MS (ESI) m/z: 200 (M + H)⁺−(HCl).

Methyl 2-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)benzoate

(6). A mixture of methyl 2-bromobenzoate 4f (5.30 g, 24.7 mmol),

bis(pinacolato)diboron (9.39 g, 37.0 mmol), Pd(dppf)Cl₂·CH₂Cl₂

(1.01 g, 1.23 mmol), and potassium acetate (7.26 g, 73.9 mmol) in a

mixture of THF (100 mL) and DMSO (5.00 mL) was stirred at 80 °C

under Ar for 5 h. The mixture was poured into water and extracted

with EtOAc. The organic layer was dried over MgSO₄, ﬁltered, and

concentrated in vacuo. The residue was puriﬁed by column

chromatography (silica gel, 1−10% EtOAc/hexane) to give 6 (6.30

g, 24.0 mmol, 98%) as a colorless oil. ¹H NMR (CDCl₃) δ 1.42 (12H,

s), 3.91 (3H, s), 7.36−7.46 (1H, m), 7.47−7.55 (2H, m), 7.94 (1H, d,

J = 7.6 Hz).

1-Benzyl-4-(5-methyl-2-(pyridin-4-yl)benzoyl)piperazin-2-one

(3d). A mixture of 2f (600 mg, 2.40 mmol), piperazin-2-one (361 mg,

3.60 mmol), HATU (137 mg, 3.60 mmol), triethylamine (1.68 mL,

12.0 mmol), and DMF (6.00 mL) was stirred at room temperature for

3 h. The mixture was poured into water and extracted with EtOAc.

The organic layer was washed with brine, dried over Na₂SO₄, ﬁltered,

and concentrated in vacuo. To a mixture of the residual compound,

benzyl bromide (0.100 mL, 0.840 mmol), and DMF (2.00 mL) was

added NaH (30.5 mg, 0.76 mmol) at room temperature, and the

mixture was stirred at the same temperature for 3 h. The mixture was

poured into water and extracted with EtOAc. Then the mixture was

washed with brine, dried over Na₂SO₄, and ﬁltered on NH silica gel.

The ﬁltrate was concentrated in vacuo. The residue was puriﬁed by

preparative HPLC. The desired fraction was neutralized with

saturated aqueous NaHCO₃and extracted with EtOAc. The organic

layer was separated, dried over MgSO₄, and concentrated in vacuo to

Methyl 2-(6-Chloropyrimidin-4-yl)benzoate (7). A mixture of 6

(2.00 g, 7.63 mmol), 4,6-dichloropyrimidine (1.36 g, 9.16 mmol),

Pd(PPh₃)₄(0.441 g, 0.380 mmol), and sodium carbonate (2.43 g,

22.9 mmol) in DME (10.0 mL) and water (2.00 mL) was stirred at

150 °C under microwave irradiation for 1 h. The mixture was poured

into water and extracted with EtOAc. The organic layer was washed

with brine, dried over Na₂SO₄, ﬁltered, and concentrated in vacuo.

The residue was puriﬁed by column chromatography (silica gel, 3−

15% EtOAc/hexane) to give 7 (0.600 g, 2.41 mmol, 32%) as a pale

1

give 3d (90.0 mg, 0.233 mmol, 10%) as a colorless oil. H NMR

1

(CDCl₃) δ 2.34−2.48 (3H, m), 2.90−4.67 (8H, m), 7.09−7.16 (2H,

m), 7.21−7.40 (8H, m), 8.58−8.63 (2H, m); MS (ESI) m/z: 386 (M

+ H)⁺.

yellow oil. H NMR (CDCl₃) δ 3.75 (3H, s), 7.49−7.67 (4H, m),

7.90 (1H, dd, J = 7.3, 1.3 Hz), 9.01 (1H, d, J = 1.3 Hz).

2-(Pyrimidin-4-yl)benzoic Acid Hydrochloride (2k). A mixture of

7 (0.600 g, 2.41 mmol), triethylamine (1.68 mL, 12.1 mmol), and

10% Pd-C (0.257 g, 0.120 mmol) in MeOH (20.0 mL) was

hydrogenated under balloon pressure at room temperature for 1 h.

The catalyst was removed by ﬁltration and the ﬁltrate was

concentrated in vacuo. The mixture was diluted with EtOAc, washed

with brine, dried over Na₂SO₄, ﬁltered, and concentrated in vacuo. A

mixture of the residue, AcOH (2.00 mL), and 6 M HCl (10.0 mL,

(4-Benzyl-3-hydroxypiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3e). Compound 3e (257 mg, 0.665 mmol, 88%)

was obtained as a white solid by a method similar to that described for

1

3a. H NMR (300 MHz, CDCl₃) δ 0.31−0.07 (1H, m), 0.56−1.65

(3H, m), 1.78−2.75 (7H, m), 2.81−3.59 (3H, m), 4.39−4.80 (1H,

m), 6.94−7.55 (10H, m), 8.52−8.68 (2H, m); MS (ESI) m/z: 387

(M + H)⁺.

12237

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(4-Benzyl-4-hydroxypiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3f). A mixture of 2f (330 mg, 1.32 mmol), 4-

benzyl-4-hydroxypiperidine 9f (379 mg, 1.98 mmol), HATU (753

mg, 1.98 mmol), and triethylamine (0.917 mL, 6.60 mmol) in DMF

(5.00 mL) was stirred at room temperature overnight. The mixture

was quenched with water and extracted with EtOAc. The organic

layer was separated, washed with brine, dried over Na₂SO₄, and

concentrated in vacuo. The residue was puriﬁed by column

chromatography (NH silica gel, 50−80% EtOAc/hexane) to give 3f

119.97, 127.00, 127.45, 128.51, 129.33, 129.37, 130.47, 135.74,

157.27, 158.38; MS (ESI) m/z: 374 (M + H)⁺.

(4-Hydroxy-4-(pyridin-2-ylmethyl)piperidin-1-yl)(2-(pyridin-4-yl)-

phenyl)methanone (3w). Compound 3w (143 mg, 0.383 mmol,

40%) was obtained as a colorless oil by a method similar to that

described for 3e. ¹H NMR (CDCl₃) δ 0.02−1.57 (5H, m), 2.54 (1H,

s), 2.76−3.32 (4H, m), 4.42 (1H, d, J = 11.0 Hz), 6.98−7.10 (1H,

m), 7.10−7.20 (1H, m), 7.31−7.55 (6H, m), 7.62 (1H, t, J = 7.6 Hz),

8.42 (1H, d, J = 4.9 Hz), 8.57−8.72 (2H, m); ¹³C NMR (CDCl₃) δ

35.97, 36.79, 36.92, 37.56, 42.59, 42.86, 47.14, 47.44, 69.17, 69.48,

77.27, 121.72, 123.21, 123.95, 124.50, 127.31, 127.75, 128.86, 129.00,

129.17, 129.31, 129.42, 135.05, 136.09, 137.06, 147.70, 148.23,

148.36, 149.86, 158.64, 168.74; MS (ESI) m/z: 374 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-(pyrimidin-4-yl)pyridin-3-

yl)methanone (3x). Compound 3x (85.0 mg, 0.227 mmol, 26%) was

obtained as a pale yellow solid by a method similar to that described

for 3f. ¹H NMR (CDCl₃) δ 1.27−2.00 (5H, m), 2.80 (2H, s), 3.09−

3.52 (3H, m), 4.43−4.67 (1H, m), 7.12−7.22 (2H, m), 7.28−7.50

(4H, m), 7.61−7.75 (1H, m), 8.16−8.27 (1H, m), 8.73−9.23 (3H,

m); ¹³C NMR (CDCl₃) δ 35.65, 35.84, 36.19, 36.48, 37.56, 37.77,

42.84, 43.68, 49.29, 50.06, 69.43, 69.94, 118.99, 119.36, 124.64,

124.90, 127.03, 128.53, 130.52, 132.93, 133.12, 135.76, 135.90,

149.52, 150.49, 151.11, 157.56, 157.91, 158.22, 162.58, 162.85,

168.07, 169.14, 189.88, 190.53, 193.28; MS (ESI) m/z: 375 (M +

H)⁺.

(4-Hydroxy-4-(pyridin-2-ylmethyl)piperidin-1-yl)(2-(pyrimidin-4-

yl)phenyl)methanone (3z). Compound 3z (85.0 mg, 0.227 mmol,

26%) was obtained as a pale yellow solid by a method similar to that

described for 3f. ¹H NMR (CDCl₃) δ 1.30−1.81 (4H, m), 2.68−3.46

(5H, m), 4.43 (1H, d, J = 13.2 Hz), 6.06 (1H, brs), 7.09 (1H, d, J =

7.6 Hz), 7.13−7.20 (1H, m), 7.36−7.42 (1H, m), 7.47−7.56 (2H, m),

7.58−7.80 (3H, m), 8.46 (1H, d, J = 4.2 Hz), 8.76 (1H, brs), 9.25

(1H, d, J = 1.1 Hz); ¹³C NMR (CDCl₃) δ 121.77, 124.50, 127.37,

127.50, 129.27, 129.32, 130.47, 134.90, 136.81, 137.07, 148.36,

157.22, 158.46, 158.68, 158.80, 169.20; MS (ESI) m/z: 375 (M +

H)⁺.

1

(401 mg, 1.04 mmol, 79%) as a white solid. H NMR (CDCl₃) δ

0.03−1.61 (5H, m), 2.26−2.83 (6H, m), 2.87−3.16 (2H, m), 4.34−

4.59 (1H, m), 6.93−7.61 (10H, m), 8.50−8.74 (2H, m); MS (ESI)

m/z: 387 (M + H)⁺.

(4-Hydroxypiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)phenyl)-

methanone (3h). Compound 3h (112 mg, 0.378 mmol, 32%) was

obtained as a colorless oil by a method similar to that described for

1

3a. H NMR (CDCl₃) δ 0.54−1.31 (2H, m), 1.40−1.81 (2H, m),

2.42 (3H, s), 2.46−2.91 (1H, m), 2.95−3.35 (2H, m), 3.71 (1H, dd, J

= 7.9, 4.2 Hz), 3.93−4.08 (1H, m), 7.22 (1H, d, J = 7.9 Hz), 7.28−

7.36 (2H, m), 7.41 (2H, t, J = 5.7 Hz), 8.57−8.66 (2H, m); MS (ESI)

m/z: 297 (M + H)⁺.

(4-Ethyl-4-hydroxypiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3i). Compound 3i (197 mg, 0.607 mmol, 36%)

was obtained as a colorless oil by a method similar to that described

for 3f. ¹H NMR (CDCl₃) δ 0.12−1.63 (10H, m), 2.41 (3H, s), 2.58−

3.20 (3H, m), 4.30−4.49 (1H, m), 7.13−7.53 (5H, m), 8.60 (2H, d, J

= 4.5 Hz); ¹³C NMR (CDCl₃) δ 6.99, 15.24, 21.10, 34.97, 35.44,

35.56, 35.64, 35.92, 36.35, 37.51, 42.58, 42.98, 65.82, 69.29, 69.41,

123.26, 123.69, 128.00, 128.25, 128.89, 129.02, 130.25, 130.35,

132.30, 135.78, 139.38, 139.53, 147.61, 149.71, 168.97; MS (ESI) m/

z: 325 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(5-chloro-2-(pyridin-4-yl)-

phenyl)methanone (3q). Compound 3q (228 mg, 0.560 mmol, 76%)

was obtained as a white solid by a method similar to that described for

1

3f. H NMR (300 MHz, CDCl₃), H NMR (CDCl₃) δ 0.97−1.74

(5H, m), 2.43 (1H, d, J = 5.7 Hz), 2.57−3.18 (4H, m), 4.44 (1H, t, J

= 13.8 Hz), 6.99−7.18 (2H, m), 7.22−7.55 (8H, m), 8.57−8.78 (2H,

m); ¹³C NMR (CDCl₃) δ 36.28, 37.51, 37.58, 38.62, 42.49, 42.93,

49.06, 49.14, 68.79, 69.09, 123.06, 123.67, 126.94, 127.59, 127.89,

128.46, 129.70, 129.79, 130.33, 130.42, 130.50, 133.65, 134.33,

135.43, 135.46, 135.58, 137.09, 137.36, 146.40, 146.49, 149.93,

150.02, 167.20, 167.27; MS (ESI) m/z: 407 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-bromo-5-methylphenyl)-

methanone (10a). A mixture of 2-bromo-5-methylbenzoic acid 4h

(3.00 g, 14.0 mmol), HATU (6.37 g, 16.7 mmol), 4-benzylpiperidin-

4-ol (2.94 g, 15.4 mmol), triethylamine (9.72 mL, 69.8 mmol), and

DMF (50.0 mL) was stirred at room temperature for 5 h. The mixture

was quenched with water at room temperature and extracted with

EtOAc. The organic layer was separated, washed with brine, dried

over Na₂SO₄, and concentrated in vacuo. The residue was puriﬁed by

column chromatography (NH silica gel, 50−80% EtOAc/Hexane) to

give 10a (4.51 g, 11.6 mmol, 83%) as a white solid. ¹H NMR

(CDCl₃) δ 1.22−1.97 (5H, m), 2.31 (3H, d, J = 7.5 Hz), 2.79 (2H, s),

3.07−3.54 (3H, m), 4.54 (1H, t, J = 11.9 Hz), 6.96−7.23 (4H, m),

7.28−7.50 (4H, m); MS (ESI) m/z: 388, 390 (M, M + 2H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-chloropyridin-3-yl)-

methanone (10h). To a mixture of 2-chloronicotinic acid 4m (1.00 g,

6.35 mmol), toluene (15 mL), and DME (5 mL) was added thionyl

chloride (0.505 mL, 6.92 mmol), and the mixture was stirred at 90 °C

under N₂for 4 h. The reaction mixture was concentrated under

reduced pressure. The residue was dissolved in THF (15 mL); then,

triethylamine (0.965 mL, 6.92 mmol) and 4-benzyl-4-hydroxypiper-

idine 9f (1.10 g, 5.77 mmol) were added and the reaction mixture was

stirred at room temperature under N₂overnight. To the reaction

mixture, saturated aqueous NaHCO₃was added and extracted with

EtOAc. The organic layer was separated, washed with brine, dried

over Na₂SO₄, and concentrated in vacuo. The residue was puriﬁed by

column chromatography (silica gel, 50−100% EtOAc/hexane) to give

(4-Benzyl-4-hydroxypiperidin-1-yl)(5-methoxy-2-(pyridin-4-yl)-

phenyl)methanone (3r). Compound 3r (156 mg, 0.388 mmol, 34%)

was obtained as a white solid by a method similar to that described for

1

3f. H NMR (CDCl₃) δ 0.05−1.63 (5H, m), 2.31−2.84 (3H, m),

2.89−3.16 (2H, m), 3.79−3.93 (3H, m), 4.34−4.58 (1H, m), 6.83−

7.15 (4H, m), 7.18−7.53 (6H, m), 8.48−8.74 (2H, m); ¹³C NMR

(CDCl₃) δ 35.76, 36.22, 36.30, 36.88, 37.44, 37.50, 42.39, 42.86,

49.05, 49.18, 55.54, 68.86, 69.14, 112.66, 112.69, 115.53, 115.75,

123.10, 123.71, 126.87, 126.90, 127.50, 128.24, 128.42, 130.35,

130.39, 130.42, 130.52, 135.56, 135.71, 136.79, 137.12, 147.22,

147.33, 149.75, 149.84, 160.24, 160.35, 168.55, 168.61; MS (ESI) m/

z: 403 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(5-ﬂuoro-2-(pyridin-4-yl)-

phenyl)methanone (3s). Compound 3s (147 mg, 0.376 mmol, 64%)

was obtained as a white solid by a method similar to that described for

1

3f. H NMR (CDCl₃) δ 0.78−1.72 (5H, m), 2.33−2.51 (1H, m),

2.55−2.81 (2H, m), 2.87−3.21 (2H, m), 4.44 (1H, brs), 6.95−7.35

(8H, m), 7.36−7.58 (2H, m), 8.48−8.77 (2H, m); ¹³C NMR

(CDCl₃) δ 35.73, 36.28, 37.53, 42.86, 49.06, 49.17, 68.79, 114.86,

123.17, 123.77, 126.95, 128.47, 130.34, 131.12, 149.98, 167.28; MS

(ESI) m/z: 391 (M + H)⁺.

1

10h (1.86 g, 5.62 mmol, 97%) as a white solid. H NMR (CDCl₃) δ

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-(pyrimidin-4-yl)phenyl)-

methanone (3u). Compound 3u (110 mg, 0.295 mmol, 35%) was

obtained as a yellow oil by a method similar to that described for 3f.

¹H NMR (CDCl₃) δ 1.29−1.42 (2H, m), 1.50−1.78 (3H, m), 2.61−

1.22−1.94 (5H, m), 2.80 (2H, d, J = 5.3 Hz), 3.08−3.60 (3H, m),

4.47−4.66 (1H, m), 7.14−7.23 (2H, m), 7.27−7.40 (4H, m), 7.55−

7.71 (1H, m), 8.43 (1H, dd, J = 4.5, 1.9 Hz); MS (ESI) m/z: 331 (M

+ H)⁺.

2.82 (2H, m), 2.86−3.40 (3H, m), 4.37−4.60 (1H, m), 7.06−7.46

(6H, m), 7.47−7.84 (4H, m), 8.66−8.81 (1H, m), 8.85−9.27 (1H,

m); ¹³C NMR (CDCl₃) δ 35.89, 36.12, 37.58, 43.47, 69.49, 69.60,

(4-Benzyl-4-methoxypiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3g). To a solution of 10a (314 mg, 0.810 mmol)

in DMF (5.00 mL) was added sodium hydride (52.9 mg, 1.21 mmol)

12238

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

at 0 °C and stirred for 30 min at the same temperature. Iodomethane

(0.252 mL, 4.04 mmol) was added to this solution at 0 °C, and the

reaction mixture was stirred at room temperature overnight. The

mixture was quenched with water at 0 °C and extracted with EtOAc.

The organic layer was separated, washed with brine, dried over

Na₂SO₄, and concentrated in vacuo. The resulting crude compound

(10b), Pd(PPh₃)₄(46.8 mg, 0.0400 mmol), sodium carbonate (172

mg, 1.62 mmol), pyridine-4-boronic acid 5a (110 mg, 0.890 mmol),

DME (5.00 mL), and water (1.00 mL) was heated at 140 °C for 1 h

under microwave irradiation. The mixture was quenched with water at

room temperature and extracted with EtOAc. The organic layer was

separated, washed with brine, dried over Na₂SO₄, and concentrated in

vacuo. The residue was puriﬁed by column chromatography (NH

silica gel, 50−80% EtOAc/hexane) to give 3g (231 mg, 0.577 mmol,

71%) as a colorless oil. ¹H NMR (CDCl₃) δ 1.23−1.45 (1H, m), 1.57

(5H, s), 2.29−2.49 (1H, m), 2.66−2.93 (2H, m), 3.68 (2H, s), 3.79

(3H, s), 3.89 (1H, d, J = 14.0 Hz), 4.74 (1H, d, J = 13.2 Hz), 6.78−

6.91 (2H, m), 7.10−7.31 (7H, m), 7.33−7.45 (1H, m), 8.60−8.68

(2H, m); ¹³C NMR (CDCl₃) δ 21.08, 32.05, 32.28, 33.56, 33.94,

37.17, 37.24, 41.56, 42.06, 42.17, 42.68, 48.64, 48.68, 73.54, 73.77,

123.15, 123.67, 126.40, 126.45, 128.03, 128.09, 128.28, 128.87,

128.96, 130.05, 130.19, 130.22, 130.32, 132.44, 133.03, 135.48,

135.76, 136.28, 136.48, 139.32, 139.44, 147.44, 147.49, 149.90,

168.94, 169.03; MS (ESI) m/z: 401 (M + H)⁺.

139.15, 139.33, 147.79, 147.91, 149.09, 149.25, 158.55, 168.99,

169.10; MS (ESI) m/z: 401 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-(3-ﬂuoropyridin-4-yl)-5-

methylphenyl)methanone (3l). A mixture of 10a (50.0 mg, 0.130

mmol), 2 M aqueous solution of sodium carbonate (0.193 mL, 0.390

mmol), 3-ﬂuoropyridin-4-ylboronic acid 5c (36.3 mg, 0.260 mmol),

Pd(dppf)Cl₂·CH₂Cl₂(10.5 mg, 0.0100 mmol), and diglyme (2.00

mL) was heated at 120 °C for 5 h under microwave irradiation. The

mixture was diluted with EtOAc, silica powder was added to this

suspension at 0 °C, and the resulting mixture was stirred at the same

temperature for 10 min. The resulting mixture was ﬁltered, and the

ﬁltrate was concentrated in vacuo. The residue was puriﬁed by

column chromatography (silica gel, 50−100% EtOAc/hexane)

followed by preparative HPLC to give 3l (12.5 mg, 0.0310 mmol,

24%) as a colorless oil. ¹H NMR (CDCl₃) δ 1.08−1.83 (5H, m), 2.43

(3H, s), 2.50−3.37 (5H, m), 4.25−4.45 (1H, m), 6.81−7.65 (9H, m),

8.23−8.61 (2H, m); MS (ESI) m/z: 405 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(5-methyl-2-(3-methylpyridin-

4-yl)phenyl)methanone (3m). Compound 3m (39.0 mg, 0.0970

mmol, 38%) was obtained as a colorless oil by a method similar to

that described for 3k. ¹H NMR (CDCl₃) δ 1.44 (2H, brs), 2.22 (4H,

brs), 2.43 (4H, s), 2.52 (1H, brs), 2.69 (2H, s), 2.81−3.04 (1H, m),

3.16 (2H, brs), 4.31 (1H, d, J = 13.3 Hz), 6.99−7.39 (9H, m), 8.34−

8.62 (2H, m); ¹³C NMR (CDCl₃) δ 17.09, 21.15, 37.37, 42.84, 49.21,

69.14, 126.90, 128.44, 129.42, 129.50, 130.42, 135.72, 136.02, 138.62,

146.80, 147.09, 151.14, 168.79; MS (ESI) m/z: 401 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(3-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3n). A mixture of 2-bromo-3-methylbenzoic acid

4i (300 mg, 1.40 mmol), DMT-MM (463 mg, 1.67 mmol), 4-

benzylpiperidin-4-ol 9f (267 mg, 1.40 mmol), N-methylmorpholine

(0.767 mL, 6.98 mmol), and DMF (5.00 mL) was stirred at room

temperature for 4 h. The mixture was quenched with water at room

temperature and extracted with EtOAc. The organic layer was

separated, washed with brine, dried over Na₂SO₄, and concentrated in

vacuo. The resulting crude product, Pd(PPh₃)₄(81.0 mg, 0.0700

mmol), sodium carbonate (297 mg, 2.80 mmol), pyridine-4-boronic

acid 5a (258 mg, 2.10 mmol), DME (5.00 mL), and water (1.00 mL)

were heated at 140 °C under microwave irradiation for 1 h. The

mixture was quenched with water at room temperature and extracted

with EtOAc. The organic layer was separated, washed with brine,

dried over Na₂SO₄, and concentrated in vacuo. The residue was

puriﬁed by column chromatography (NH silica gel, 0−20% EtOAc/

hexane) and by column chromatography (silica gel, 0−20% MeOH/

hexane), followed by crystallization (EtOAc/hexane) to give 3n (112

mg, 0.290 mmol, 20%) as a white solid. Mp 189 °C (EtOAc/hexane);

¹H NMR (CDCl₃) δ 0.64−1.52 (5H, m), 2.17 (3H, d, J = 15.9 Hz),

(4-Hydroxy-4-phenylpiperidin-1-yl)(5-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3j). A mixture of 2-bromo-5-methylbenzoic acid

4h (100 mg, 0.470 mmol), HATU (265 mg, 0.700 mmol), 4-hydroxy-

4-phenylpiperidine 9j (99.0 mg, 0.560 mmol), triethylamine (0.129

mL, 0.930 mmol), and DMF (3.00 mL) was stirred at room

temperature for 3 h. The mixture was quenched with water at room

temperature and extracted with EtOAc. The organic layer was

separated, washed with water and brine, dried over Na₂SO₄, and

concentrated in vacuo. The resulting crude compound, Pd(PPh₃)₄

(27.2 mg, 0.0200 mmol), 2 M aqueous solution of sodium carbonate

(0.470 mL, 0.940 mmol), pyridine-4-boronic acid 5a (87.0 mg, 0.710

mmol), and DME (5.00 mL) were heated at 120 °C under microwave

irradiation for 1 h. The mixture was quenched with water at room

temperature and extracted with EtOAc. The organic layer was

separated, washed with brine, dried over Na₂SO₄, and concentrated in

vacuo. The residue was puriﬁed by column chromatography (silica

gel, 50−100% EtOAc/hexane), followed by crystallization (EtOAc/

hexane) to give 3j (112 mg, 0.301 mmol, 64%) as a white solid. Mp

1

188 °C (EtOAc/hexane); H NMR (CDCl₃) δ 0.66−2.17 (5H, m),

2.43 (3H, s), 2.69−3.40 (3H, m), 4.46−4.76 (1H, m), 7.05 (1H, d, J

= 7.4 Hz), 7.19−7.45 (9H, m), 7.52 (1H, d, J = 6.0 Hz), 8.57−8.74

(1H, m); ¹³C NMR (CDCl₃) δ 21.12, 37.38, 37.56, 37.61, 37.68,

38.00, 38.64, 42.70, 43.24, 71.04, 71.08, 123.26, 123.81, 124.03,

124.39, 127.15, 127.34, 128.03, 128.24, 128.45, 129.04, 130.31,

130.45, 132.36, 133.03, 135.40, 135.72, 139.46, 139.58, 147.47,

147.61, 147.66, 149.67, 149.87, 169.09; MS (ESI) m/z: 373 (M +

H)⁺.

2.44−3.32 (5H, m), 4.25 (1H, d, J = 12.8 Hz), 7.01−7.58 (10H, m),

8.44−8.84 (2H, m); ¹³C NMR (CDCl₃) δ 20.42, 20.60, 35.95, 36.41,

36.71, 37.07, 37.14, 42.84, 42.92, 49.13, 49.36, 68.99, 69.27, 123.73,

124.13, 124.38, 125.05, 126.91, 128.45, 128.56, 130.42, 130.92,

134.95, 135.31, 135.73, 135.95, 136.56, 136.77, 146.77, 149.55,

168.69; MS (ESI) m/z: 387 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(5-methyl-2-(2-methylpyridin-

4-yl)phenyl)methanone (3k). A mixture of 10a (100 mg, 0.260

mmol), (2-methylpyridin-4-yl)boronic acid 5b (52.9 mg, 0.390

mmol), 3 M aqueous solution of potassium carbonate (0.172 mL,

0.520 mmol), Pd(PPh₃)₄(14.9 mg, 0.0100 mmol), and DME (1.00

mL) was heated at 120 °C for 1 h under microwave irradiation. The

mixture was poured into water at room temperature and extracted

with EtOAc. The organic layer was separated, washed with brine,

dried over MgSO₄, and concentrated in vacuo. The residue was

puriﬁed by column chromatography (NH silica gel, 50−100%

EtOAc/hexane) to give 3k (26.7 mg, 0.0670 mmol, 26%) as a

(4-Benzyl-4-hydroxypiperidin-1-yl)(4-methyl-2-(pyridin-4-yl)-

phenyl)methanone (3o). Compound 3o (737 mg, 1.91 mmol, 82%)

was obtained as a pink solid by a method similar to that described for

3n. ¹H NMR (CDCl₃) δ 0.20−1.74 (6H, m), 2.43 (3H, s), 2.69 (2H,

brs), 2.88−3.16 (2H, m), 4.43 (1H, brs), 6.92−7.60 (10H, m), 8.52−

8.74 (2H, m); ¹³C NMR (CDCl₃) δ 21.28, 35.80, 36.31, 36.88, 37.46,

42.53, 42.87, 49.08, 49.18, 68.88, 69.17, 123.24, 123.84, 126.86,

127.51, 127.75, 128.40, 129.56, 129.69, 129.79, 129.90, 130.39,

132.78, 133.08, 135.20, 135.65, 135.72, 135.88, 139.52, 139.62,

147.75, 149.81, 169.04; MS (ESI) m/z: 387 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-methyl-6-(pyridin-4-yl)-

phenyl)methanone (3p). To a mixture of 2-bromo-6-methylbenzoic

acid 4k (300 mg, 1.40 mmol), DMF (10.8 μl, 0.14 mmol), and THF

(5.00 mL) was added oxalyl chloride (0.134 mL, 1.53 mmol) at 0 °C.

The mixture was stirred at 0 °C to room temperature for 4 h. The

mixture was concentrated in vacuo, and the residue was dissolved in

THF (2.00 mL). This solution was added to a mixture of 4-

benzylpiperidin-4-ol 9f (295 mg, 1.54 mmol), triethylamine (0.893

1

colorless oil. H NMR (CDCl₃) δ 1.14−1.72 (6H, m), 2.31−2.51

(4H, m), 2.55−2.82 (4H, m), 2.88−3.16 (2H, m), 4.36−4.60 (1H,

m), 7.04 (1H, d, J = 6.1 Hz), 7.14 (1H, d, J = 8.0 Hz), 7.18−7.41

(8H, m), 8.41−8.61 (1H, m); ¹³C NMR (CDCl₃) δ 21.09, 24.46,

24.57, 35.93, 36.12, 36.38, 36.88, 37.36, 37.43, 42.43, 42.84, 49.17,

68.91, 69.19, 120.34, 120.99, 122.58, 123.20, 126.86, 128.04, 128.40,

128.80, 128.99, 130.32, 132.59, 133.31, 135.42, 135.59, 135.71,

12239

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

mL, 6.42 mmol), and THF (5.00 mL) at room temperature. The

mixture was stirred at the same temperature overnight. The solvent

was removed in vacuo, and the obtained compound was dissolved in a

microwave vessel with DME (5.00 mL). To this mixture was added

Pd(PPh₃)₄(74.0 mg, 0.0600 mmol), sodium carbonate (271 mg, 2.56

mmol), pyridine-4-boronic acid 5a (236 mg, 1.92 mmol), and water

(1.00 mL) and the resulting mixture was heated at 140 °C under

microwave irradiation for 1 h. The mixture was quenched with water

at room temperature and extracted with EtOAc. The organic layer was

separated, washed with brine, dried over Na₂SO₄, and concentrated in

vacuo. The residue was puriﬁed by column chromatography (NH

silica gel, 0−20% EtOAc/hexane) followed by column chromatog-

raphy (silica gel, 0−20% MeOH/ EtOAc) to give 3p (283 mg, 0.732

(1H, m), 7.56−7.68 (2H, m), 7.77 (2H, d, J = 5.3 Hz), 8.42 (1H, d, J

= 4.9 Hz), 8.66−8.80 (3H, m); MS (ESI) m/z: 375 (M + H)⁺.

tert-Butyl 4-(2-bromo-5-methylbenzoyl)piperazine-1-carboxy-

late (11). A mixture of 24h (500 mg, 2.33 mmol), DIPEA (0.609

mL, 3.49 mmol), HATU (1.06 g, 2.79 mmol), tert-butylpiperazine-1-

carboxylate (520 mg, 2.79 mmol), and DMF (5.00 mL) was stirred at

room temperature overnight. The reaction mixture was diluted with

water and extracted with EtOAc. The extract was washed with water

and brine, dried over MgSO₄, ﬁltered, and concentrated in vacuo. The

residue was puriﬁed by column chromatography (silica gel, 20−50%

EtOAc/hexane) to give 11 (877 mg, 2.29 mmol, 98%) as a white

solid. ¹H NMR (CDCl₃) δ 1.47 (9H, s), 2.32 (3H, s), 3.10−3.61 (6H,

m), 3.65−3.90 (2H, m), 7.01−7.11 (2H, m), 7.39−7.50 (1H, m).

(5-Methyl-2-(pyridin-4-yl)phenyl)(piperazin-1-yl)methanone

(12). A mixture of 11 (500 mg, 1.30 mmol), Pd(PPh₃)₄(75.0 mg,

0.0700 mmol), sodium carbonate (277 mg, 2.61 mmol), pyridine-4-

boronic acid 5a (176 mg, 1.43 mmol), DME (10.0 mL), and water

(2.00 mL) was heated at 150 °C for 1 h under microwave irradiation.

The mixture was quenched with water at room temperature and

extracted with EtOAc. The organic layer was separated, washed with

brine, dried over MgSO₄, and concentrated in vacuo. The residue was

puriﬁed by column chromatography (silica gel, 50−100% EtOAc/

hexane). A solution of this crude compound in EtOAc (5.00 mL) was

added to 4 M HCl solution of EtOAc (5.00 mL, 20.0 mmol) at room

temperature. The mixture was stirred at the same temperature

overnight, and concentrated in vacuo. The residue was dissolved with

water and washed with EtOAc. The aqueous layer was basiﬁed with 1

M NaOH, diluted with brine, and extracted with EtOAc−THF. The

extract was washed with brine, dried over MgSO₄, ﬁltered, and

concentrated in vacuo to give 12 (283 mg, 1.01 mmol, 78%) as a

1

mmol, 57%) as a colorless oil. H NMR (CDCl₃) δ 0.34−2.10 (5H,

m), 2.19−2.81 (6H, m), 2.86−3.16 (2H, m), 4.36−4.53 (1H, m),

6.93−7.75 (10H, m), 8.49−8.77 (2H, m); ¹³C NMR (CDCl₃) δ

19.27, 19.37, 35.87, 36.45, 36.47, 36.95, 37.03, 37.25, 42.20, 42.29,

49.08, 49.15, 68.92, 69.11, 123.46, 124.08, 126.62, 126.88, 126.91,

128.42, 128.88, 128.90, 130.35, 130.41, 130.80, 135.08, 135.34,

135.29, 135.70, 135.87, 147.96, 148.09, 149.60, 149.69, 168.01,

168.04; MS (ESI) m/z: 387 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2-(pyridin-4-yl)phenyl)-

methanone (3t). Compound 3t (763 mg, 2.05 mmol, 82%) was

obtained as a pink solid by a method similar to that described for 3n.

¹H NMR (CDCl₃) δ 0.11−1.85 (5H, m), 2.25−3.31 (5H, m), 4.43

(1H, brs), 6.77−7.81 (11H, m), 8.40−8.84 (2H, m); ¹³C NMR

(CDCl₃) δ 35.80, 36.32, 36.86, 37.47, 42.50, 42.86, 49.08, 49.18,

68.87, 69.15, 123.25, 123.86, 126.89, 127.45, 127.72, 128.42, 128.99,

129.13, 129.25, 129.51, 129.57, 130.37, 135.22, 135.60, 135.70,

135.91, 147.58, 149.81, 149.87, 168.77; MS (ESI) m/z: 373 (M +

H)⁺.

1

colorless oil. H NMR (CDCl₃) δ 1.43 (1H, s), 1.93 (1H, ddd, J =

11.5, 7.8, 3.0 Hz), 2.42 (3H, s), 2.44−2.60 (2H, m), 2.64−2.85 (2H,

m), 2.89−3.03 (1H, m), 3.48−3.66 (2H, m), 7.24 (1H, d, J = 0.8 Hz),

7.30−7.34 (2H, m), 7.38−7.43 (2H, m), 8.57−8.68 (2H, m); MS

(ESI) m/z: 282 (M + H)⁺.

(4-Benzyl-4-hydroxypiperidin-1-yl)(2,4′-bipyridin-3-yl)-

methanone (3v). A mixture of 10h (5.00 g, 15.1 mmol), pyridine-4-

boronic acid 5a (2.23 g, 18.1 mmol), sodium carbonate (4.81 g, 45.3

mmol), Pd(PPh₃)₄(0.873 g, 0.76 mmol), DMF (50 mL), and water

(10 mL) was heated at 100 °C under N₂overnight. To the reaction

mixture, brine was added and extracted with EtOAc. The organic layer

was separated, washed with brine, dried over Na₂SO₄, and

concentrated in vacuo. The residue was puriﬁed by column

chromatography (NH silica gel, 50−100% EtOAc/hexane) to give

3v (3.42 g, 9.16 mmol, 61%) as a white solid. Crystallization from

EtOAc/hexane aﬀorded compound 3v as a white crystal. Mp 150 °C

(EtOAc/hexane); ¹H NMR (CDCl₃) δ 0.07-1.67 (5H, m), 2.35−3.17

(5H, m), 4.41−4.60 (1H, m), 6.98−7.15 (2H, m), 7.22−7.32 (3H,

m), 7.41 (1H, dd, J = 7.5, 5.0 Hz), 7.61 (1H, d, J = 4.2 Hz), 7.70−

7.83 (2H, m), 8.60−8.82 (3H, m); ¹³C NMR (CDCl₃) δ 35.6, 36.0,

36.2, 36.7, 37.6, 42.5, 43.0, 49.1, 49.2, 68.8, 69.0, 122.8, 123.5, 126.9,

128.4, 130.3, 131.5, 131.8, 135.4, 135.6, 136.0, 136.1, 146.3, 150.0,

150.3, 151.6, 152.2, 167.2, 167.4; MS (ESI) m/z: 374 (M + H)⁺;

Anal. Calcd for C₂₃H₂₃N₃O₂: C, 73.97; H, 6.21; N, 11.25. Found: C,

73.81; H, 6.27; N, 11.15.

(4-Benzylpiperazin-1-yl)(5-methyl-2-(pyridin-4-yl)phenyl)-

methanone (3b). To a mixture of 12 (50.0 mg, 0.180 mmol) and

benzaldehyde (0.0220 mL, 0.210 mmol) in EtOAc (2.00 mL) was

added NaBH(OAc)₃(56.5 mg, 0.270 mmol) at room temperature.

The mixture was stirred at the same temperature overnight. Then the

mixture was quenched with water, diluted with 1 M NaOH, and

extracted with EtOAc. The organic layer was separated, washed with

brine, dried over MgSO₄, and concentrated in vacuo. The residue was

puriﬁed by column chromatography (NH silica gel, 10−50% EtOAc/

hexane) to give 3b (60.6 mg, 0.163 mmol, 92%) as a colorless oil. ¹H

NMR (CDCl₃) δ 1.52 (1H, t, J = 7.6 Hz), 2.01−2.19 (2H, m), 2.31−

2.45 (4H, m), 2.66−2.77 (1H, m), 2.94−3.05 (1H, m), 3.31 (2H, s),

3.47−3.71 (2H, m), 7.17−7.26 (5H, m), 7.27−7.32 (3H, m), 7.37−

7.41 (2H, m), 8.59−8.66 (2H, m); MS (ESI) m/z: 372 (M + H)⁺.

(4-Benzoylpiperazin-1-yl)(5-methyl-2-(pyridin-4-yl)phenyl)-

methanone (3c). A mixture of 12 (50.0 mg, 0.180 mmol), benzoic

acid (26.0 mg, 0.210 mmol), HATU (101 mg, 0.270 mmol), DIPEA

(0.0930 mL, 0.530 mmol), and DMF (1.00 mL) was stirred at room

temperature overnight. The reaction mixture was diluted with water

and extracted with EtOAc. The extract was washed with brine, dried

over MgSO₄, and concentrated in vacuo. The residue was puriﬁed by

column chromatography (NH silica gel, 50−100% EtOAc/hexane)

and crystallized from EtOAc−hexane to give 3c (50.3 mg, 0.130

mmol, 73%) as a white solid. Mp 184 °C (EtOAc/hexane); ¹H NMR

(CDCl₃) δ 2.43 (3H, s), 2.56−4.01 (8H, m), 7.22−7.45 (10H, m),

8.60−8.70 (2H, m); ¹³C NMR (CDCl₃) δ 21.11, 41.59, 46.38,

123.34, 126.98, 128.37, 128.59, 129.06, 130.06, 130.85, 132.79,

134.57, 134.95, 139.74, 147.21, 150.13, 169.56, 170.35; MS (ESI) m/

z: 386 (M + H)⁺.

2,4′-Bipyridin-3-yl(4-hydroxy-4-(pyridin-2-ylmethyl)piperidin-1-

yl)methanone (3y). A solution of 2-chloronicotinic acid 4m (150 mg,

0.950 mmol), HATU (434 mg, 1.14 mmol), triethylamine (0.663 mL,

4.76 mmol), and 4-(pyridin-2-ylmethyl)piperidin-4-ol dihydrochlor-

ide 9k (201 mg, 1.05 mmol) in DMF (5.00 mL) was stirred at room

temperature for 4 h. The mixture was quenched with water at room

temperature and extracted with EtOAc. The organic layer was

separated, washed with brine, dried over Na₂SO₄, and concentrated in

vacuo. The resulting crude product (10i), Pd(PPh₃)₄(54.9 mg,

0.0500 mmol), sodium carbonate (201 mg, 1.90 mmol), pyridine-4-

boronic acid 5a (128 mg, 1.05 mmol), DME (5.00 mL), and water

(1.00 mL) were heated at 140 °C for 1 h under microwave irradiation.

The mixture was quenched with water at room temperature and

extracted with EtOAc. The organic layer was separated, washed with

brine, dried over Na₂SO₄, and concentrated in vacuo. The residue was

puriﬁed by column chromatography (NH silica gel, 50−80% EtOAc/

Protein Crystallization and Structure Analyses. Human

CH24H (aa 28-494) with an N-terminal 6xHIS tag was cloned into

pCWOri vector (Addgene). The above plasmid was co-transformed

together with pGro7 (TaKaRa) into E. coli stain DH5α. To express

the protein, the transformed cells were cultured in 2XYT media and

induced at OD 0.8−1 with 1 mM IPTG, 0.5 mM 5-aminolevulinic

1

hexane) to give 3y (116 mg, 0.310 mmol, 33%) as a colorless oil. H

NMR (CDCl₃) δ 0.84−1.58 (5H, m), 2.43−3.36 (5H, m), 4.49 (1H,

d, J = 10.6 Hz), 7.00−7.10 (1H, m), 7.12−7.21 (1H, m), 7.35−7.47

12240

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

acid hydrochloride, and 0.02% arabinose, and then further cultured at

30 °C for 72 h. The cells were harvested and puriﬁed.³⁴The ﬁnal

protein concentration for crystallization was 28 mg in the buﬀer

containing 50 mM KPi 7.4, 500 mM NaCl, 0.5 mM EDTA, 2 mM

DTT, and 10% glycerol. The CH24H co-crystal complex was

prepared by addition of 2 mM compound and grown from a reservoir

solution containing 0.4 M calcium chloride and 18% (w/v) PEG 3350

at 20 °C by the sitting-drop vapor diﬀusion method. Prior to data

collection, crystals were immersed in the reservoir solution containing

25% ethylene glycol as a cryoprotectant and ﬂash-frozen in liquid

nitrogen. Diﬀraction data were collected from a single crystal using

the CCD detector Quantum 315R (ADSC) at beamline 5.0.3 of the

Advanced Light Source (Berkeley, CA) under a 100 K nitrogen

cryostream. The data were reduced and scaled with the program

HKL2000.³⁵The structure was solved by the molecular replacement

method with Molrep³⁶of the CCP4 program suites³⁷using the

CH24H structure (PDB code 2Q9F) as a search model. The structure

was reﬁned through an iterative procedure utilizing REFMAC³⁸

followed by interactive model building using the program COOT.³⁹

The dictionary ﬁles for the ligands were prepared using AFITT

(OpenEye Scientiﬁc Software). The ﬁnal models were validated using

MolProbity.⁴⁰Crystallographic processing and reﬁnement statistics

are summarized in Supplementary Table 4. All graphical ﬁgures were

CYP2C19, CYP2D6, or CYP3A4; BD Biosciences), substrates

(phenacetin, amodiaquine, tolbutamide, (S)-mephenytoin, bufuralol,

or testosterone), and 1−100 μmol/L test compounds. The substrates

and inhibitors were dissolved in methanol and dimethylsulfoxide,

respectively, and added to the incubation mixture with the ﬁnal

solvent concentration of 0.5%. Incubations were conducted at 37 °C

for 10−20 min and terminated by adding acetonitrile. After

centrifugation, aliquots of the supernatants were subjected to

measurement of the LC/MS/MS. All incubations were made in

duplicate. The activities of CYP1A2, CYP2C8, CYP2C9, CYP2C19,

CYP2D6, and CYP3A4 were determined by the peak of

acetaminophen, N-desethylamodiaquine, 1′-hydroxydiclofenac, 4′-

hydoroxymephenytoin, 1′-hydroxybufuralol, and 6β-hydroxytestoster-

one, respectively. The activities of test samples were expressed as the

percentage of activity remaining compared with a control sample

containing no inhibitor.

In Vitro Steroidogenic Disruption Model in H295R Cells.

The H295R human adrenocortical carcinoma cell line was obtained

from American Type Culture Collection (ATCC, #CRL-2128). Cells

were cultured in 75 mL ﬂasks (Corning Costar, Cambridge, MA,

430641), with 30 mL of media at 37 °C with a 5% CO₂atmosphere.

Brieﬂy, cells were grown in Dulbecco’s modiﬁed Eagle’s medium and

Ham’s F-12 Nutrient mixture (DMEM/F12) medium (COSMO-

BIO, Tokyo, Japan, 16007100) supplemented with 2.5% Nu-serum

(BD Falcon, Franklin, Lakes, NJ, 355100), 1% ITS-plus (BD Falcon,

354352), 100 units/mL penicillin, and 100 μg/mL streptomycin

(Gibco, Grand Island, New York, 15140-122). The H295R steroid

hormones synthesis assay was performed as follows: The cells were

grown in 24-well plates (Corning Costar, 3526) with 5 × 10⁴cells in

0.5 mL of medium for each well, and were incubated three overnights.

After incubation, the medium was changed with 0.5 mL of culture

medium containing 20 μM forskolin. After two additional overnight

incubations, cells were washed with 500 μL of DMEM/F12 medium

supplemented with 1% ITS-plus twice. And then, the cells were

treated with 500 μL of DMEM/F12 medium supplemented with 1%

ITS-plus 20 μM forskolin and varying concentrations of the

compounds. After overnight incubation, the supernatant of the

culture media was collected and stored at −30 °C. Commercially

available enzyme immunoassay (EIA) kits were used with steroid

hormones, such as cortisol, testosterone, aldosterone, and cortico-

sterone (Cayman, San Diego, CA, 500360, 582701, 10004377, and

500655, respectively), according to the manufacturer’s instructions.

The procedure followed the basic principle of EIA, in which there is

competition between an acetylcholinesterase-linked hormone (tracer)

and a nonradioactive hormone for a ﬁxed number of antibody binding

sites. Amounts of hormones in the above reserved cell culture media

were quantitatively measured in the use of a calibration curve.

Percentage inhibition was calculated from amounts of hormones using

the following formula: % inhibition = (A − X) × 100/(A)

̈

generated using PyMOL (Schrodinger LLC, Cambridge, MA) and

LIGPLOT.⁴¹

Measurement of CH24H Inhibitory Activity. CH24H enzyme

was expressed by transducing full-length CH24H gene (NCBI

Accession Num. BC022539) into FreeStyle 293-F cells (Invitrogen,

Carlsbad, CA). A CH24H lysate product was prepared from

supernatant isolated by centrifugation of the homogenate. Catalytic

activity of CH24H enzyme was measured using TLC. To evaluate the

inhibitory activity, 2 μL of serial diluted compounds were incubated

with 3 μL of CH24H enzyme in assay buﬀer (50 mM Potassium

phosphate buﬀer (pH 7.4) supplemented with 0.1% BSA and

Complete EDTA-free (Roche, Switzerland)) for about 15 min at

room temperature. The ﬁnal concentration of DMSO in the assay was

0.2% when the compound was tested in duplicate in 384-well plates.

The reaction was started with the addition of 5 μL of substrate [14C]

Cholesterol (PerkinElmer, Foster City, CA, NEC018250UC) at a

ﬁnal concentration of 15 μM, which was dissolved in assay buﬀer

containing 2 mM β-NADPH. After 5-h incubation at 37 °C, 35 μL of

chloroform: methanol (1:2 v/v) was added to terminate the CH24H

reaction. After mixing, 25 μL of distilled water containing 0.0024%

Trypan blue was added to the mixture. Based on the Bligh and Dyer

total lipid extraction method, 4.5 μL of the lower layer was spotted on

the TLC plate. The TLC plates were developed using ethyl

acetate:toluene (2:3 v/v) and visualized by FLA7000 (Fujiﬁlm

Corporation, Japan).

Measurement of CYP3A4 Inhibitory Activity. Incubation

mixtures were prepared in a total volume of 40 μL with ﬁnal

component concentrations as follows: 50 mmol/L phosphate buﬀer

(pH 7.4), NADPH-generating system (a mixture of MgCl₂, β-NADP⁺,

glucose-6-phosphate, and glucose-6-phosphate dehydrogenase),

CYP3A4-expressing microsome (BD Biosciences), substrate (testos-

terone), and 10 μmol/L test compounds. The substrate and inhibitors

were dissolved in methanol and dimethylsulfoxide, respectively, and

added to the incubation mixture with the ﬁnal solvent concentration

of 0.5%. Incubations were conducted for 60 min and terminated by

adding acetonitrile. After centrifugation, aliquots of the supernatants

were subjected to measurement of the LC/MS/MS. All incubations

were made in triplicate. The activity of CYP3A4 was determined by

the peak of 6β-hydroxytestosterone. The activities of test samples

were expressed as the percentage of activity remaining compared with

a control sample containing no inhibitor.

A: 20 μM forskolin, X: 20 μM forskolin plus test inhibitor. IC₅₀

values were calculated with XLﬁt software (IDBS, Emeryville, CA)

using a four-parameter logistic curve. Data are provided as the average

and standard error of mean (SEM) performed in triplicate.

In Vitro Metabolic Clearance in Human Hepatic Micro-

somes. Human liver microsomes were purchased from Sekisui

XenoTech, LLC. (Kansas City, KS). The microsomes (0.2 mg

protein/mL) and the compound (1 μM) were mixed in phosphate

buﬀer (pH 7.4). The reactions were initiated by adding an NADPH-

generating system (a mixture of MgCl₂, glucose-6-phosphate, β-

NADP⁺, and glucose-6-phosphate dehydrogenase) to the mixtures

before incubation. Incubations were conducted at 37 °C for 15 and 30

min and terminated by adding acetonitrile. The zero-time incubations,

which served as the controls, were terminated by adding acetonitrile

before adding each compound. After the samples were mixed and

centrifuged, the compound concentrations in the supernatant

fractions were measured by LC/MS/MS.

Measurement of CYP Family Enzyme Inhibitory Activity.

Incubation mixtures were prepared in a total volume of 100 μL with

ﬁnal component concentrations as follows: 50 mmol/L phosphate

buﬀer (pH 7.4), NADPH-generating system (a mixture of MgCl₂, β-

NADP⁺, glucose-6-phosphate, and glucose-6-phosphate dehydrogen-

ase), CYP-expressing microsomes (CYP1A2, CYP2C8, CYP2C9,

Ethics Statement. The care and use of the animals and the

experimental protocols used in this research were approved by the

Experimental Animal Care and Use Committee of Takeda

Pharmaceutical Company Limited.

12241

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

In Vivo PK/PD Study of Compound 3v in Mice. Female

C57BL/6N mice (7- to 8-week-old) (3 mice/group) were used.

Compound 3v was suspended in a 0.5% aqueous methylcellulose

[133-14255 WAKO] solution (10 mL/kg). The solution was forcibly

administered orally at 1, 3, and 10 mg/kg. At 0.5 and 1, 2, 4, 8, and 24

h for 1 and 3 mg/kg, and at 0.25, 0.5, 1, 2, 4, 8, and 24 h for 10 mg/kg

after administration, the animals were sacriﬁced and the brain samples

were collected. The brain was homogenized with about 4-fold amount

of saline. The concentrations of 3v in the brain were measured with

LC-MS/MS, and the brain 24HC was measured with HPLC

described below.

Measurement of Plasma and Brain Concentrations of

Compound 3v in Mice. The aliquots of the brain homogenate

were mixed with acetonitrile containing internal standard. The

mixtures were centrifuged. The supernatants were diluted with

solvents for LC-MS/MS (mobile phase A: 10 mM ammonium

formate/formic acid (100/0.2, v/v), mobile phase B: acetonitrile/

formic acid (100/0.2, v/v)). The diluted solutions were injected into

LC-MS/MS (API5000, AB Sciex, CA) equipped with Shimadzu

Shim-pack XR-ODS (2.2 μm, 2.0 × 30 mm, Shimadzu Corporation,

Japan) at 50 °C. Compound 3v was detected using a multiple reaction

monitoring mode using the transition: m/z 374.20 → 183.10. Analyst

software TM (version 1.4.2) was used for data acquisition and

processing.

William Farnaby − Research, Takeda Pharmaceutical

Company Ltd., Fujisawa, Kanagawa 251-8555, Japan

Toshiya Nishi − Research, Takeda Pharmaceutical Company

Ltd., Fujisawa, Kanagawa 251-8555, Japan

Etsurou Watanabe − Research, Takeda Pharmaceutical

Company Ltd., Fujisawa, Kanagawa 251-8555, Japan

Jason Yano − Takeda California Inc., San Diego, California

92121, United States

Maki Miyamoto − Research, Takeda Pharmaceutical

Company Ltd., Fujisawa, Kanagawa 251-8555, Japan

Shigeru Kondo − Research, Takeda Pharmaceutical Company

Ltd., Fujisawa, Kanagawa 251-8555, Japan

Tsuyoshi Ishii − Research, Takeda Pharmaceutical Company

Ltd., Fujisawa, Kanagawa 251-8555, Japan

Takanobu Kuroita − Research, Takeda Pharmaceutical

Company Ltd., Fujisawa, Kanagawa 251-8555, Japan

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.1c00864

Author Contributions

The manuscript was written through contributions of all

authors. All authors have given approval to the ﬁnal version of

the manuscript.

Measurement of 24HC in the Mouse Brain. 24HC in the brain

homogenate was extracted with an acetonitrile solution (98%

acetonitrile, 1.98% methanol, 0.02% formic acid), and quantiﬁed by

HPLC (UPLC, Waters corporation, MA) that was equipped with

Acquity UPLC HSS C₁₈SB (1.8 μm, 2.1 × 100 mm) maintained at 40

°C. The average value of 24HC amount was calculated and the results

are shown in relative values with the control group as 100%. UV

detection was performed at 201 nm.

Notes

The authors declare the following competing ﬁnancial

interest(s): This study was funded by Takeda Pharmaceutical

Company Limited (TAKEDA). T.K., M.Y., H.K.A, W.F., T.N.,

E.W., J.Y., M.M., S.K., T.I. and T.K. are current and former

employees of TAKEDA. No other potential conﬂicts of interest

relevant to this article exists.

ASSOCIATED CONTENT

* Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.jmedchem.1c00864.

■

sı

ACKNOWLEDGMENTS

■

X-ray data collection and reﬁnement statistics for the

crystal structures of CH24H in complex with com-

pounds 1b, 3f, and 3v; brain concentration of compound

3v after oral administration in C57BL/6N mice (8

weeks) and the eﬀect on 24HC reduction; chromato-

grams for puriﬁcation and analysis of 3a−z; and ¹H and

¹³C NMR charts for 3v (PDF)

The authors thank Tomohiro Kaku, Katsumi Kobayashi, and

Keisuke Imamura for obtaining SAR around hit compound 1a;

Sayuri Watanabe and Shigeo Hasegawa for obtaining biological

results; Weston Lane for depositing crystal structures to pdb;

and Takashi Miki and Shinichi Kondo for helpful discussions.

GM/CA@APS has been funded by the National Cancer

Institute (ACB-12002) and the National Institute of General

Medical Sciences (AGM-12006, P30GM138396). This

research used resources of the Advanced Photon Source, a

US Department of Energy (DOE) Oﬃce of Science User

Facility operated for the DOE Oﬃce of Science by Argonne

National Laboratory under Contract No. DE-AC02-

06CH11357. The Berkeley Center for Structural Biology is

supported in part by the Howard Hughes Medical Institute.

The Advanced Light Source is a Department of Energy Oﬃce

of Science User Facility under Contract No. DE-AC02-

05CH11231.

Molecular formula strings (CSV)

Accession Codes

The coordinates of the crystal structures of CH24H in complex

with compounds 1b (7LS4), 3f (7LS3), and 3v (7LRL) have

been deposited in the Protein Data Bank. The authors will

release the atomic coordinates and experimental data upon

article publication.

AUTHOR INFORMATION

Corresponding Author

■

Tatsuki Koike − Research, Takeda Pharmaceutical Company

Ltd., Fujisawa, Kanagawa 251-8555, Japan; orcid.org/

0000-0002-3264-1406; Phone: +81-466-32-1132;

Email: tatsuki.koike@takeda.com

ABBREVIATIONS

■

AD, Alzheimer’s disease; BBB, blood−brain barrier; CNS,

central nervous system; CSF, cerebrospinal ﬂuid; DDI, drug−

drug interaction; HBA, hydrogen-bonding acceptor; HBD,

Authors

1

Masato Yoshikawa − Research, Takeda Pharmaceutical

Company Ltd., Fujisawa, Kanagawa 251-8555, Japan;

orcid.org/0000-0002-7175-1173

Haruhi Kamisaki Ando − Research, Takeda Pharmaceutical

Company Ltd., Fujisawa, Kanagawa 251-8555, Japan

hydrogen-bonding donor; 24HC, 24S-hydroxycholesterol; H

NMR, proton nuclear magnetic resonance; HTS, high-

throughput screening; SAR, structure−activity relationship;

SBDD, structure-based drug design; TLC, thin-layer chroma-

tography

12242

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Miles, R.; Aubourg, P.; Cartier, N. CYP46A1 inhibition, brain

cholesterol accumulation and neurodegeneration pave the way for

Alzheimer ’s disease. Brain 2015, 138, 2383−2398.

REFERENCES

■

(1) Russell, D. W.; Halford, R. W.; Ramirez, D. M.; Shah, R.; Kotti,

T. Cholesterol 24-hydroxylase: An enzyme of cholesterol turnover in

the brain. Annu. Rev. Biochem. 2009, 78, 1017−1040.

(16) Hudry, E.; Van Dam, D.; Kulik, W.; De Deyn, P. P.; Stet, F. S.;

̀

Ahouansou, O.; Benraiss, A.; Delacourte, A.; Bougneres, P.; Aubourg,

(2) Kölsch, H.; Lutjohann, D.; Ludwig, M.; Schulte, A.; Ptok, U.;

̈

P.; Cartier, N. Adeno-associated virus gene therapy with cholesterol

24-hydroxylase reduces the amyloid pathology before or after the

onset of amyloid plaques in mouse models of Alzheimer ’s disease.

Mol. Ther. 2010, 18, 44−53.

Jessen, F.; von Bergmann, K.; Rao, M. L.; Maier, W.; Heun, R.

Polymorphism in the cholesterol 24S-hydroxylase gene is associated

with Alzheimer ’s disease. Mol. Psychiatry 2002, 7, 899−902.

(3) Li, L.; Yin, Z.; Liu, J.; Li, G.; Wang, Y.; Yan, J.; Zhou, H.

CYP46A1 T/C polymorphism associated with the APOEepsilon4

allele increases the risk of Alzheimer ’s disease. J. Neurol. 2013, 260,

1701−1708.

(17) Mast, N.; Saadane, A.; Valencia-Olvera, A.; Constans, J.;

Maxfield, E.; Arakawa, H.; Li, Y.; Landreth, G.; Pikuleva, I. A.

Cholesterol-metabolizing enzyme cytochrome P450 46A1 as a

pharmacologic target for Alzheimer ’s disease. Neuropharmacology

2017, 123, 465−476.

(18) Mast, N.; Verwilst, P.; Wilkey, C. J.; Guengerich, F. P.;

Pikuleva, I. A. In Vitro Activation of Cytochrome P450 46A1

(CYP46A1) by Efavirenz-Related Compounds. J. Med. Chem. 2020,

63, 6477−6488.

(19) Petrov, A. M.; Lam, M.; Mast, N.; Moon, J.; Li, Y.; Maxfield, E.;

Pikuleva, I. A. CYP46A1 Activation by Efavirenz Leads to Behavioral

Improvement without Significant Changes in Amyloid Plaque Load in

the Brain of 5XFAD Mice. Neurotherapeutics 2019, 16, 710−724.

(20) Bialer, M.; Johannessen, S. I.; Koepp, M. J.; Levy, R. H.;

Perucca, E.; Tomson, T.; White, H. S. Progress report on new

antiepileptic drugs: A summary of the Fourteenth Eilat Conference on

New Antiepileptic Drugs and Devices (EILAT XIV). I. Drugs in

preclinical and early clinical development. Epilepsia 2018, 59, 1811−

1841.

(21) Strzelczyk, A.; Schubert-Bast, S. Therapeutic advances in

Dravet syndrome: a targeted literature review. Expert Rev. Neurother.

2020, 20, 1065−1079.

(22) Halford, J. J.; Sperling, M. R.; Arkilo, D.; Asgharnejad, M.;

Zinger, C.; Xu, R.; During, M.; French, J. A. A phase 1b/2a study of

soticlestat as adjunctive therapy in participants with developmental

and/or epileptic encephalopathies. Epilepsy Res. 2021, 174,

No. 106646.

(23) Guengerich, F. P. Human Cytochrome P450 Enzymes. In

Cytochrome P450: Structure, Mechanism, and Biochemistry, 3rd ed.;

Ortiz de Montellano, P. R., Ed.; Kluwer Academic: New York, 2005;

pp 377−530.

(24) Mast, N.; White, M. A.; Bjorkhem, I.; Jhonson, E. F.; Stout, C.

D.; Pikuleva, I. A. Crystal structures of substrate-bound and substrate-

free cytochrome P450 46A1, the principal cholesterol hydroxylase in

the brain. Proc. Natl. Acad. Sci. USA 2008, 105, 9546−9551.

(25) Mast, N.; Charvet, C.; Pikuleva, I. A.; Stout, C. D. Structural

basis of drug binding to CYP46A1, an enzyme that controls

cholesterol turnover in the brain. J. Biol. Chem. 2010, 285, 31783−

31795.

(4) Schönknecht, P.; Lutjohann, D.; Pantel, J.; Bardenheuer, H.;

̈

Hartmann, T.; von Bergmann, K.; Beyreuther, K.; Schröder, J. 24S-

hydroxycholesterol is increased in patients with Alzheimer ’s disease

compared to healthy controls. Neurosci. Lett. 2002, 324, 83−85.

(5) Papassotiropoulos, A.; Lutjohanm, D.; Bagli, M.; Locatelli, S.;

Jessen, F.; Buschfort, R.; Ptok, U.; Bjorkhem, I.; Bergmann von, K.;

Heun, R. 24 S -hydroxycholesterol in cerebrospinal fluid is elevated in

early stages of dementia. J. Psychiatr. Res. 2002, 36, 27−32.

(6) Leoni, V.; Shafaati, M.; Salomon, A.; Kivipelto, M.; Björkhem, I.;

Wahlund, L. O. Are the CSF levels of 24S-hydroxycholesterol a

sensitive biomarker for mild cognitive impairment? Neurosci. Lett.

2006, 397, 83−87.

(7) Tian, G.; Kong, Q.; Lai, L.; Ray-Chaudhury, A.; Lin, C.-L. G.

Increased expression of cholesterol 24S-hydroxylase results in

disruption of glial glutamate transporter EAAT2 association with

lipid rafts: a potential role in Alzheimer ’s disease. J. Neurochem. 2010,

113, 978−989.

(8) Nishi, T.; Kondo, S.; Miyamoto, M.; Watanabe, S.; Hasegawa, S.;

Kondo, S.; Yano, J.; Watanabe, E.; Ishi, T.; Yoshikawa, M.; Ando, H.

K.; Farnaby, W.; Fujimoto, S.; Sunahara, E.; Ohori, M.; During, M. J.;

Kuroita, T.; Koike, T. Soticlestat, a novel cholesterol 24-hydroxylase

inhibitor shows a therapeutic potential for neural hyperexcitation in

mice. Sci. Rep. 2020, 10, No. 17081.

(9) Lehmann, J. M.; Kliewer, S. A.; Moore, L. B.; Smith-Oliver, T.

A.; Oliver, B. B.; Su, J. L.; Sundseth, S. S.; Winegar, D. A.; Blanchard,

D. E.; Spencer, T. A.; Willson, T. M. Activation of the nuclear

receptor LXR by oxysterols defines a new hormone response pathway.

J. Biol. Chem. 1997, 272, 3137−3140.

(10) Umetani, M.; Domoto, H.; Gormley, A. K.; Yuhanna, I. S.;

Cummins, C. L.; Javitt, N. B.; Korach, K. S.; Shaul, P. W.;

Mangelsdorf, D. J. 27-Hydroxycholesterol is an endogenous SERM

that inhibits the cardiovascular effects of estrogen. Nat. Med. 2007, 13,

1185−1192.

(11) Wang, Y.; Kumar, N.; Crumbley, C.; Griffin, P. R.; Burris, T. P.

A second class of nuclear receptors for oxysterols: Regulation of

RORalpha and RORgamma activity by 24S-hydroxycholesterol

(cerebrosterol). Biochim. Biophys. Acta, Mol. Cell Biol. Lipids 2010,

1801, 917−923.

(12) Paul, S. M.; Doherty, J. J.; Robichaud, A. J.; Belfort, G. M.;

Chow, B. Y.; Hammond, R. S.; Crawford, D. C.; Linsenbardt, A. J.;

Shu, H. J.; Izumi, Y.; Mennerick, S. J.; Zorumski, C. F. The major

brain cholesterol metabolite 24(S)-hydroxycholesterol is a potent

allosteric modulator of N-methyl-D-aspartate receptors. J. Neurosci.

2013, 33, 17290−300.

(26) Mast, N.; Linger, M.; Clark, M.; Wiseman, J.; Stout, C. D.;

Pikuleva, I. A. In silico and intuitive predictions of CYP46A1

inhibition by marketed drugs with subsequent enzyme crystallization

in complex with fluvoxamine. Mol. Pharmacol. 2012, 82, 824−834.

(27) Shafaati, M.; Mast, N.; Beck, O.; Nayef, R.; Heo, G. Y.;

Björkhem-Bergman, L.; Lutjohann, D.; Björkhem, I.; Pikuleva, I. A.

̈

The antifungal drug voriconazole is an efficient inhibitor of brain

cholesterol 24S-hydroxylase in vitro and in vivo. J. Lipid Res. 2010, 51,

318−323.

(13) Björkhem, I.; Lövgren-Sandblom, A.; Leoni, V.; Meaney, S.;

Brodin, L.; Salveson, L.; Winge, K.; Pålhagen, S.; Svenningsson, P.

Oxysterols and Parkinson ’s disease: evidence that levels of 24S-

hydroxycholesterol in cerebrospinal fluid correlates with the duration

of the disease. Neurosci. Lett. 2013, 555, 102−105.

(14) Boussicault, L.; Alves, S.; Lamaziere, A.; Planques, A.; Heck, N.;

Moumné, L.; Despres, G.; Bolte, S.; Hu, A.; Pages, C.; Galvan, L.;

Piguet, F.; Aubourg, P.; Cartier, N.; Caboche, J.; Betuing, S.

CYP46A1, the rate-limiting enzyme for cholesterol degradation, is

neuroprotective in Huntington ’s disease. Brain 2016, 139, 953−970.

(15) Djelti, F.; Braudeau, J.; Hudry, E.; Dhenain, M.; Varin, J.;

(28) Mast, N.; Zheng, W.; Stout, C. D.; Pikuleva, I. A. Antifungal

Azoles: Structural Insights into Undesired Tight Binding to

Cholesterol-Metabolizing CYP46A1. Mol. Pharmacol. 2013, 84, 86−

94.

̀

(29) Mast, N.; Zheng, W.; Stout, C. D.; Pikuleva, I. A. Binding of a

cyano- and fluoro-containing drug bicalutamide to cytochrome P450

46A1: unusual features and spectral response. J. Biol. Chem. 2013,

288, 4613−4624.

(30) Leclerc, J. P.; Fagnou, K. Palladium-Catalyzed Cross-Coupling

Reactions of Diazine N -Oxides with Aryl Chlorides, Bromides, and

Iodides. Angew. Chem., Int. Ed. 2006, 45, 7781−7786.

̀

Bieche, I.; Marquer, C.; Chali, F.; Ayciriex, S.; Auzeil, N.; Alves, S.;

Langui, D.; Potier, M. C.; Laprevote, O.; Vidaud, M.; Duyckaerts, C.;

12243

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(31) Leo, A.; Weininger, D. Clogp Reference Manual, Daylight,

version 4.9. http://www.daylight.com/dayhtml/doc/clogp/.

(32) Nielsen, F. K.; Hansen, C. H.; Fey, J. A.; Hansen, M.; Jacobsen,

N. W.; Halling-Sorensen, B.; Bjorklund, E.; Styrishave, B. H295R cells

as a model for steroidogenic disruption: a broader perspective using

simultaneous chemical analysis of 7 key steroid hormones. Toxicol. In

Vitro 2012, 26, 343−350.

(33) Oskarsson, A.; Ullerås, E.; Plant, K. E.; Hinson, J. P.; Goldfarb,

P. S. Steroidogenic gene expression in H295R cells and the human

adrenal gland: adrenotoxic effects of lindane in vitro. J. Appl. Toxicol.

2006, 26, 484−492.

(34) White, M. A.; Mast, N.; Bjorkhem, I.; Johnson, E. F.; Stout, C.

D.; Pikuleva, I. A. Use of complementary cation and anion heavy-

atom salt derivatives to solve the structure of cytochrome P450 46A1.

Acta Crystallogr., Sect. D: Biol. Crystallogr. 2008, 64, 487−495.

(35) Otwinowski, Z.; Minor, W. [20] Processing of X-ray diffraction

data collected in oscillation mode. Methods Enzymol. 1997, 276, 307−

326.

(36) Vagin, A.; Teplyakov, A. MOLREP: an Automated Program for

Molecular Replacement. J. Appl. Crystallogr. 1997, 30, 1022−1025.

(37) Project CC Collaborative Computational Project, N The CCP4

suite: programs for protein crystallography Acta Crystallogr., Sect. D:

Biol. Crystallogr. 1994, 50, 760 763

DOI: 10.1107/

S0907444994003112.

(38) Murshudov, G. N.; Skubák, P.; Lebedev, A. A.; Pannu, N. S.;

Steiner, R. A.; Nicholls, R. A.; Winn, M. D.; Long, F.; Vagin, A. A.

REFMAC5 for the refinement of macromolecular crystal structures.

Acta Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 355−367.

(39) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features

and development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr.

2010, 66, 486−501.

(40) Chen, V. B.; Arendall, W. B., 3rd; Headd, J. J.; Keedy, D. A.;

Immormino, R. M.; Kapral, G. J.; Murray, L. W.; Richardson, J. S.;

Richardson, D. C. MolProbity: all-atom structure validation for

macromolecular crystallography. Acta Crystallogr., Sect. D: Biol.

Crystallogr. 2010, 66, 12−21.

(41) Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. LIGPLOT: a

program to generate schematic diagrams of protein-ligand inter-

actions. Protein Eng., Des. Sel. 1995, 8, 127−134.

NOTE ADDED AFTER ASAP PUBLICATION

This article published August 13, 2021 with an incorrect Figure

5 ﬁle. The correct ﬁgure published August 16, 2021.

■

12244

https://doi.org/10.1021/acs.jmedchem.1c00864

J. Med. Chem. 2021, 64, 12228−12244

Article Doi

Discovery of Soticlestat, a Potent and Selective Inhibitor for Cholesterol 24-Hydroxylase (CH24H)

DOI: 10.1021/acs.jmedchem.1c00864

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.1c00864

Full text of DOI:10.1021/acs.jmedchem.1c00864

Products guided by the article

R&D Labs maybe for 1429504-71-1

Relevant to this article

Hot Product