Discovery of ORN0829, a potent dual orexin 1/2 receptor antagonist for the treatment of insomnia

Article abstract of DOI:10.1016/j.bmc.2020.115489

Here, we present the design, synthesis, and SAR of dual orexin 1 and 2 receptor antagonists, which were optimized by balancing the antagonistic activity for orexin receptors and lipophilicity. Based on the prototype compound 1, ring construction and the i

Full text of DOI:10.1016/j.bmc.2020.115489

Journal Pre-proofs
Discovery of ORN0829, a potent dual orexin 1/2 receptor antagonist for the
treatment of insomnia
Aya Futamura, Ryo Suzuki, Yunoshin Tamura, Hiroshi Kawamoto, Mari
Ohmichi, Noriko Hino, Yuichi Tokumaru, Sora Kirinuki, Tetsuaki Hiyoshi,
Takeshi Aoki, Daiji Kambe, Dai Nozawa
PII:
S0968-0896(20)30310-2
https://doi.org/10.1016/j.bmc.2020.115489
BMC 115489
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
5 March 2020
2 April 2020
3 April 2020
Please cite this article as: A. Futamura, R. Suzuki, Y. Tamura, H. Kawamoto, M. Ohmichi, N. Hino, Y.
Tokumaru, S. Kirinuki, T. Hiyoshi, T. Aoki, D. Kambe, D. Nozawa, Discovery of ORN0829, a potent dual
orexin 1/2 receptor antagonist for the treatment of insomnia, Bioorganic & Medicinal Chemistry (2020), doi:
https://doi.org/10.1016/j.bmc.2020.115489
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Discovery of ORN0829, a potent dual orexin 1/2 receptor antagonist for the treatment of insomnia
Aya Futamuraa, Ryo Suzukia, Yunoshin Tamuraa, Hiroshi Kawamotoa, Mari Ohmichib, Noriko Hinoa, Yuichi
Tokumarua, Sora Kirinukia, Tetsuaki Hiyoshia, Takeshi Aokia, Daiji Kambea and Dai Nozawaa,*
a Discovery Research Laboratories, Taisho Pharmaceutical Co., Ltd. 1-403 Yoshino-cho, Kita-ku, Saitama 331-
9530, Japan
b Drug Safety and Pharmacokinetics Laboratories, Taisho Pharmaceutical Co., Ltd. 1-403 Yoshino-cho, Kita-ku,
Saitama 331-9530, Japan
Abstract
Here, we present the design, synthesis, and SAR of dual orexin 1 and 2 receptor antagonists, which were optimized
by balancing the antagonistic activity for orexin receptors and lipophilicity. Based on the prototype compound 1, ring
construction and the insertion of an additional heteroatom into the resulting ring led to the discovery of orexin 1 and
2 receptor antagonists, which were 3-benzoyl-1,3-oxazinane derivatives. Within these derivatives, (−)-3h enabled a
high dual orexin receptor antagonistic activity and a low lipophilicity. Compound (−)-3h exhibited potent sleep-
promoting effects at a po dose of 1 mg/kg in a rat polysomnogram study, and optimal PK properties with a rapid Tmax
and short half-lives in rats and dogs were observed, indicating a predicted human half-life of 0.9–2.0 h. Thus, (−)-3h
(ORN0829; investigation code name, TS-142) was selected as a viable candidate and is currently in clinical
development for the treatment of insomnia.
1

Abbreviations
OX1R, orexin 1 receptor; OX2R, orexin 2 receptor; DORA, dual orexin receptor antagonist; SORA, selective orexin
receptor antagonist; PSG, polysomnography; SAR, structure-activity relationship; Cbz, benzyloxycarbonyl; Boc,
tert-butoxycarbonyl; Ms, methanesulfonyl; IBX, 2-iodoxybenzoic acid; T3P®, propylphosphonic acid anhydride;
DIPEA, N,N-diisopropylethylamine; CDI, 1,1'-carbonyldiimidazole; PPL, lipase from Porcine Pancreas; MTBE,
methyl tert-butyl ether; REMD, replica-exchange molecular dynamics; CYP, cytochrome P450; CHO, Chinese
hamster ovary; ER, efflux ratio; P-gp, P-glycoprotein; RHS, right hand side; PPB, plasma protein binding; CSF,
cerebrospinal fluid; MED, minimum effective dose.
Introduction
Orexin neuropeptides (orexin-A and -B) were discovered in 1998 by two independent research groups1–2 and were
found to bind to seven-transmembrane G-protein-coupled receptors, orexin receptor 1 (OX1R) and orexin receptor 2
(OX2R).1 Both orexin-A and -B are proteolytically derived from a common precursor protein, prepro-orexin peptide,
produced in orexin neurons located in the lateral hypothalamus. OX1R and OX2R are widely distributed, including
the basal forebrain, limbic structures, and brainstem regions, and have partially overlapping but distinct distributions
throughout the brain, suggesting that each receptor subtype may have different physiological roles through different
neuronal pathways.3 The orexin system has been implicated in sleep-wake regulation, energy homeostasis, feeding,
2

and reward processing.4 The contribution of OXRs to sleep-wake regulation has been studied using orexin receptor-
knockout mice, and OX2R reportedly plays a larger role than OX1R.5–6 Experiments using receptor-selective
antagonists suggest that selective OX2R antagonists (2-SORAs) induce sleep, whereas selective OX1R antagonists
(1-SORAs) did not influence sleep architectures.7–8 However, emerging literature has reported that dual orexin
receptor antagonists (DORAs) might be more effective for sleep promotion than 2-SORA,9 and OX1R antagonism
has a subtle role in vigilance state gating, which is related to sleep onset, and influences stage transitions.10
Additionally, OX1R may have a role in the treatment of psychiatric disorders, such as stress and anxiety-related
behaviors.11–12
Figure 1. Representative DORAs and 2-SORA.
DORAs, as well as 2-SORAs, have been studied as new hypnotics (Figure 1). Almorexant13, SB-64986814,
Suvorexant15 and Lemborexant16 are DORAs for which proof-of-concepts for the treatment of insomnia in humans
have been completed. Moreover, Suvorexant and Lemborexant have received approval from the Food and Drug
Administration (FDA). ACT-541468 is presently being used in ongoing clinical activities, and it reportedly has a
short half-life in humans (approximately 6 h).17 Seltorexant, a representative 2-SORA, reportedly displays favorable
pharmacodynamic effects on sleep induction in addition to having a short half-life (approximately 2 h).18
3

In a regulatory review, the FDA has raised concerns about potential next-day residual effects associated with high
doses of Suvorexant (a plasma half-life of approximately 12 h), and the FDA-approved maximum dose of Suvorexant
is 20 mg once daily.19 This prompted us to explore compounds with optimal pharmacokinetic (PK) profiles, such as
rapid absorption and a favorable short half-life, to avoid the risks of long-acting sleep agents with next-day residual
effects.
Previously, we reported a new chemical class of DORAs with a pyrazolylethylbenzamide structure (Figure 2).20
These DORAs were identified through the optimization of pharmacophore A using a ligand-based drug design
approach and retaining the characteristic U-shaped conformation.21 Among this chemical class, compound 1 had
modest in vitro antagonistic activities for OX1R and OX2R, while it was estimated to have an optimal PK profile with
a rapid Tmax and a short half-life (predicted human T1/2: 1.2–1.4 h), which was a development target, based on animal
PK data. In a rat sleep study using polysomnography (PSG), compound 1 showed a marginal in vivo efficacy, which
was consistent with its moderate in vitro potency. On the other hand, compound 2 was identified as the most potent
DORA among this class; unfortunately, it also had a slightly higher lipophilicity.
Figure 2. Design of pharmacophore B.
Cyclic amine derivatives, such as SB-649868 (Figure 1), have been known to exhibit a high activity for OX1R and
OX2R, however these derivatives have relatively high lipophilicity. Based on pharmacophore A, construction of a
ring (as SB-649868) and insertion of an additional heteroatom into the resulting ring was designed, to maintain an
4

active conformation and to lower the lipophilicity of the molecule (Pharmacophore B in Figure 2). This effort led to
the discovery of orexin 1 and 2 receptor antagonists, which were 3-benzoyl-1,3-oxazinane derivatives. Herein, we
report the design, synthesis, and structure-activity relationship of compounds with unique heterocyclic core ring
structures. We also provide the detailed PK profiles and results of PSG sleep studies for the clinical candidate (−)-
3h.
Chemistry
Scheme 1 summarizes the synthetic routes of (− or +)-3a, (− or +)-3b, (−)-3d, (−)-3f, and (−)-3g. Alcohol (R)-16d,
obtained from the reduction of mixed anhydride (prepared in situ from commercially available carboxylic acid (S)-
15d with NaBH4, was converted into mesylate, which was subsequently alkylated with 5-fluoro-2-(1H-pyrazol-3-
yl)pyridine 17 to yield (R)-18d. The deprotection of the Cbz group in (R)-18d followed by the amidation of the
resulting secondary amine with 5-methyl-2-(2H-1,2,3-triazol-2-yl)benzoyl chloride 19 yielded (−)-3d. Compounds
(− or +)-3a, (− or +)-3b, (−)-3f, and (−)-3g were prepared in a manner similar to that of (−)-3d but using (S or R)-
16a, (S or R)-16b, (R)-16f, and (R)-16g as a starting material.
Scheme 1. Reagents and conditions: (a) isobutyl chloroformate, 4-methylmorpholine, then NaBH4, THF-H2O, -
10 °C ; or BH3-THF, THF, 0°C–rt; (b) (1) MsCl, Et3N, CHCl3, 0°C, (2) 5-fluoro-2-(1H-pyrazol-3-yl)pyridine 17, t-
5

BuONa, DMSO, 80°C; or 17, Cs2CO3, DMF, 80°C; (c) Pd(OH)2/C, H2, Et3N, THF, rt; (d) 5-methyl-2-(2H-1,2,3-
triazol-2-yl)benzoyl chloride 19, Et3N, CHCl3, 0°C; (e) 4M HCl-EtOAc, EtOAc, rt.
Compounds (−)-3e and (−)-3i were synthesized as shown in Scheme 2. Homoserine lactone hydrochloride 21 was
treated with aqueous formaldehyde solution under an acidic condition to form (4S)-1,3-oxazinane-4-carboxylic acid,
which was acylated with 19 to yield (S)-22. Compounds (−)-3i and (−)-3e were prepared from (S)-22 and 25 in a
similar manner as described in Scheme 1.
Scheme 2. Reagents and conditions: (a) formaldehyde (37% in water), 1M hydrochloric acid, H2O, rt; (b) 19, Et3N,
CHCl3, 0°C; (c) isobutyl chloroformate, 4-methylmorpholine, then NaBH4, THF-H2O, -10°C; (d) (1) MsCl, Et3N,
CHCl3, 0°C, (2) 17, t-BuONa, DMSO, 80°C; (e) chiral separation using HPLC.
Oxazolidine and oxazinane derivatives with a pyrazole moiety [(− or +)-3c, (− or +)-3h, (−)-4, (−)-5 and (−)-10–
(−)-14] were synthesized as depicted in Scheme 3. The cyclic hemiaminal frameworks of these compounds were
constructed by reacting 32 or 35 with the corresponding aminoalcohols in the presence of molecular sieves, and the
subsequent amidation of the secondary amine moieties with the corresponding benzoic acids or benzoyl chlorides
yielded the target compounds (3c, 3h, 4, 5 and 10–14). Optical resolution of the racemates using chiral stationary
phase HPLC yielded the corresponding chiral compounds. Compound 32 (for 3c, 3h, 4 and 10–14) was prepared via
the alkylation of 17, 27, and 28 with bromoacetaldehyde diethyl acetal followed by hydrolysis under acidic conditions.
Compound 35 (for 5) was synthesized starting from the N-arylation of 33 with 2,4-difluoropyridine, followed by
6

oxidation with 2-iodoxybenzoic acid (IBX).
Scheme 3. Reagents and conditions: (a) bromoacetaldehyde diethylacetal, Cs2CO3, DMF, 80°C; (b) TFA, CHCl3,
rt; (c) 3-aminopropan-1-ol or 2-aminoethanol, molecular sieves 4A, CHCl3, rt; (d) R1-COCl (19, 36, 37), Et3N, CHCl3,
0 °C–rt; (e) R1-CO2H (38, 39), T3P®, DIPEA, CHCl3; (f) chiral separation using HPLC; (g) 2,5-difluoropyridine,
Cs2CO3, MeCN, 80°C; (h) IBX, DMSO, rt.
Scheme 4 outlined the synthetic routes of 1,3-oxazinan-2-yl derivatives (−)-6–(−)-9, which were prepared from the
chiral intermediate (−)-42. The 1,3-oxazinane ring was constructed by the cyclization of ethyl glyoxylate and 3-
aminopropan-1-ol in the presence of molecular sieves, followed by the acylation of the amine moiety with 19 to give
the racemate 41. Chiral separation of the racemate using the chiral preparative HPLC method yielded (−)-41, which
was reduced with NaBH4 to produce the primary alcohol (−)-42. Conversion of the alcohol moiety in (−)-42 into the
5-(5-fluoropydin-2-yl)-1,2,4-oxadiazol-3-ylmethyl group in (−)-6 was conducted using the conventional 4-step
reaction sequence via cyanation of the alcohol followed by the condensation of the hydroxyamidine with 5-
fluoropyridine-2-carboxylic acid procedures. Compounds (−)-7–(−)-9 were synthesized from (−)-42 by alkylation
with 1,2,3-trazole 43 and 1,2,4-triazol-4-one 44 in a similar procedure described above.
7

Scheme 4. Reagents and conditions: (a) 3-aminopropan-1-ol, molecular sieves 4A, CHCl3, rt; (b) 19, Et3N, CHCl3,
0 °C; (c) chiral separation using HPLC; (d) NaBH4, MeOH, rt; (e) (1) MsCl, Et3N, CHCl3, 0 °C, (2) NaCN, DMF,
80°C; (f) (1) NH2OH aq., EtOH, 80°C, (2) 5-fluoropyridine-2-carboxylic acid, 1,1'-carbonyldiimidazole (CDI), DMF,
40–80°C; (g) (1) MsCl, Et3N, CHCl3, 0°C, (2) R2H (43 or 44), t-BuONa, DMSO, 80°C.
The synthetic procedures used to prepare (−)-3h in bulk for its preclinical studies are shown in Scheme 5. In the
asymmetric synthesis of the key intermediate (−)-42, the enzymatically enantioselective acetylation of (+)-42 using
lipase from porcine pancreas (PPL) was accomplished in the presence of vinyl acetate to yield (−)-42 in high
enantiomeric excess (>99.5% ee). The following alkylation with 17 and EtOAc stallization in ethanol provided
enantiomerically pure (−)-3h.
Scheme 5. Reagents and conditions: (a) 3-aminopropan-1-ol, molecular sieves 4A, CHCl3, rt; (b) 19, Et3N, CHCl3,
0°C; (c) NaBH4, MeOH, rt; (d) PPL, vinyl acetate, MTBE; (e) (1) MsCl, Et3N, CHCl3, 0°C, (2) 17, t-BuONa, DMSO,
80°C; (f) Recrystallization from EtOH, rt–1°C.
The absolute stereo-chemical structure of (−)-3h was determined using its X-ray crystallography, and the absolute
8

configurations of other compounds were tentatively assigned by comparing their optical rotations with that of (−)-
3h. We found that the (−)-enantiomer was the active form by assessing the inhibitory activities of both enantiomers
of 3a–3c, 3h, 4, 10, and 12.
Results and Discussion
To explore the SARs of the compounds, the antagonist activities (IC50s) against OX1R and OX2R were obtained
using functional assays in which intracellular Ca2+ mobilization was measured in CHO-K1 cells overexpressing the
human orexin receptors. In addition, the metabolic stability in human liver microsomes (hMS) and the efflux ratio
(ER) of P-glycoprotein (P-gp) were evaluated for each compound. Also, the LogD7.4 value of each compound was
assessed using high-performance liquid chromatography (LogDHPLC) as an indicator of lipophilicity.22
Figure 3. Conformational analysis of the derivatives.
The red meshes show the distribution of the pyridine moiety under the alignment of the 5-tolyl triazole substructures.
The black circles show the approximate positions of the pyridine moieties in the U-shape conformations.
Conformations of the compounds were calculated using the replica-exchange molecular dynamics (REMD)
simulation.22
As stated in the Introduction, we have been focusing on the development of a potent and orally bioavailable DORA
that possesses an optimal PK profile, including rapid absorption and a favorable short half-life. To achieve this goal,
9

we aimed to enhance the antagonistic activity without increasing the lipophilicity of the linear compound (e.g.,
compound 1 and 2 in Figure 2), as a lower lipophilicity reportedly indicates a lower Vd, resulting in a short half-
life.20 At first, to maintain an active conformation, cyclization of the central core (as SB-649868 in Figure 1) was
considered based on pharmacophore A. Conformational analysis revealed that the compound having a pyridyl-
pyrazolyl-methyl side chain in its piperidine ring had a U-shape conformation, which is expected to contribute to the
active conformation (3b in Figure 3). From these results, we initially built a simple cyclic amine motif, such as
pyrrolidine or piperidine, and evaluated the functional activities for OX1R and OX2R of the resulting compounds
[(−)-3a, (+)-3a, (−)-3b, (+)-3b] (Table 1). The enantiomers [(−)-3a and (−)-3b] with an (S)-configuration showed
more potent inhibitory activities than those of the corresponding (R)-enantiomers [(+)-3a and (+)-3b]. The cyclic
amine showed a potent antagonistic activity with single-digit, nano-molar potencies, as in (−)-3b, but also increased
lipophilicity, with (−)-3b showing the most susceptible metabolic stability among these compounds (Table 1).
Next, a carbon atom on the pyrrolidine or piperidine core ring of (−)-3a and (−)-3b was replaced with a heteroatom,
such as an oxygen or sulfur atom, to reduce lipophilicity.23–24 Unfortunately, the 5-membered 1,3-oxazolidine rings
[(−)-3c and (−)-3d] decreased the functional activities of both OX1R and OX2R, while they contributed to lower
lipophilicity, to some extent, when compared with the pyrrolidine ring [(−)-3a]. In contrast, the 1,3-thiazolidine rings
[(−)-3e and (−)-3f] did not show any virtues in either functional activity or lipophilicity. In the case of a 6-membered
morpholine ring [(−)-3g], an approximately 10-fold reduction in the antagonist potency for OX2R was observed,
compared with that of piperidine (−)-3b. On the other hand, the 6-membered 1,3-oxazinane motifs [(−)-3h and (−)-
3i] were well tolerated in terms of the antagonist activities for both OX1R and OX2R; in particular, (−)-3h was about
3-fold more potent than (−)-3i (Table 1).
Table 1. SAR of amino core A
10

b
Compound
A
IC50 (nM)a
LogDHPLC
hMS(%)c
20.2
P-gp (ER)
1.51
OX1R
3.22
OX2R
2
2.81
3.15
313
2.78
3.08
3.06
2.96
2.95
2.67
2.68
2.52
(−)-3a
(+)-3a
(−)-3b
(+)-3b
(−)-3c
(+)-3c
(−)-3d
3.30
363
37.2
NT
1.96
NT
2.27
20.8
8.08
2690
149
1.64
52.4
12.1
2320
133
51.9
NT
1.63
NT
15.7
NT
1.21
NT
18.0
1.65
11

(−)-3e
(−)-3f
61.1
4.60
2300
13.1
3.51
3.08
53.0
73.7
1.34
1.13
(−)-3g
3.50
15.0
2.33
24.5
3.56
(−)-3h
(+)-3h
(−)-3i
1.05
344
1.27
274
2.58
2.56
2.39
34.4
NT
2.43
NT
3.36
3.41
24.1
3.08
aThe experiments were performed two to five times, and each test was performed in duplicate or triplicate. bLogD at
c
pH 7.4. Metabolized % after 15 min incubation in human liver microsomes (0.25 mg protein/mL, 1 μmol/L in
DMSO).
Interestingly, (−)-3h retained highly potent antagonistic activities despite a lower lipophilicity, compared with the
parent piperidine (−)-3b. Moreover, (−)-3h had the lowest P-gp ER value (2.43) among the oxygen-inserted
heterocyclic compounds [(−)-3g, (−)-3h, and (−)-3i]. While 1,3-oxazinane (−)-3h has a hemiaminal ether structure,
fortunately there was no problem with its chemical stability under acidic conditions (JP 1st fluid (pH 1.2), 37°C,
24h). Since (−)-3h was one of the most potent DORAs in this class and had a relatively low lipophilicity and well-
balanced metabolic stability (hMS), we were motivated to conduct further optimization (Table 1).
Optimization of the right hand side (RHS) of (−)-3h was performed, focusing on the internal 5-membered
heterocycle between the 1,3-oxazinan-2-ylmethyl and 5-fluoropyridin-2-yl moieties, including pyrazole [(−)-4, (−)-
12

5; regioisomers of (−)-3h], oxadiazole [(−)-6], triazole [(−)-7, (−)-8] and triazolone [(−)-9] (Table 2). Compounds
(−)-4–(−)-7 were potent DORAs with single-digit, nano-molar potencies, while antagonistic activities of these
compounds were lower than that of (−)-3h. Compound (−)-4 had a lower lipophilicity (LogDHPLC, 2.38), but (−)-4
had a higher P-gp ER value than (−)-3h [(−)-4: 3.54, (−)-3h: 2.43]. The replacement of the 4-fluoropyridine of (−)-4
with a lipophilic 4-fluorobenzene led to a potent DORA [(−)-10] without any P-gp issues; however, further
evaluations of (−)-10 suggested concerns in terms of both the time-dependent inhibition of CYP 3A4 and its binding
affinity to a hERG K+ channel, which probably occurred because of a high lipophilicity.25 From these results, we
confirmed that 3-(5-fluoropyridin-2-yl)-1H-pyrazole was the most suitable RHS moiety.
Table 2. SAR of RHS
b
Compound
RHS
IC50 (nM)a
LogDHPLC
hMS(%)c
P-gp (ER)
OX1R
OX2R
(−)-3h
(−)-4
(−)-5
(−)-6
1.05
5.39
3.98
4.54
1.27
2.58
2.38
3.03
2.53
34.4
15.8
29.7
19.5
2.43
3.54
1.56
3.06
2.21
2.03
3.60
13

(−)-7
(−)-8
(−)-9
(−)-10
2.90
12.1
85.5
2.21
1.88
3.07
2.80
2.38
2.52
3.15
16.4
4.3
1.38
4.97
2.88
1.35
5.40
15.6
24.8
0.784
aThe experiments were performed two to five times, and each test was performed in duplicate or triplicate. bLogD at
c
pH 7.4. Metabolized % after 15 min incubation in human liver microsomes (0.25 mg protein/mL, 1 μmol/L in
DMSO).
Finally, we optimized the acyl moiety of (−)-3h as described in Table 3. Replacement of the 5-methyl-2-
triazolylbenzamide in (−)-3h with 5-fluoro-2-triazolylbenzamide [(−)-11], 2-triazolylbenzamide [(−)-12], 5-fluoro-
2-pyrimidinylbenzamide [(−)-13], or 5-methyl-2-pyrimidinylbenzamide [(−)-14] was well tolerated in terms of the
OX1R and OX2R potency, while these replacements led to an increase in the P-gp susceptibility. Taking these results
into consideration, we selected compound (−)-3h for further evaluation.
Table 3. SAR of acyl moiety
Compound
Ar
LogDHPLC
hMS(%)c
IC50 (nM)a
b
P-gp (ER)
OX1R
OX2R
14

(−)-3h
(−)-11
(−)-12
(−)-13
(−)-14
1.05
3.95
3.76
3.29
3.22
1.27
1.96
2.62
3.18
4.58
2.58
2.46
2.30
2.36
2.30
34.4
43.2
14.6
26.3
28.9
2.43
2.64
3.25
6.74
5.36
aThe experiments were performed two to five times, and each test was performed in duplicate or triplicate. bLogD
c
at pH 7.4. Metabolized % after 15 min incubation in human liver microsome (0.25 mg protein/mL, 1 μmol/L in
DMSO).
Pharmacokinetic parameters of (−)-3h in male rats and dogs are shown in Table 4. The intravenous administration
(iv) of (−)-3h exhibited a plasma clearance of 2370 and 159 mL/h/kg and Vd values of 853 and 294 mL/Kg (rats and
dogs), which were consistent with both microsomal stability results (rat: 51.0%, dog: 10.5%)26 and the plasma protein
binding (PPB) ratios. The plasma half-lives were as short as 0.238 h for rats and 1.16 h for dogs. For oral
administration, a rapid Tmax of 0.333 h and 0.583 h with oral bioavailabilities (F) of 5.5% and 61.6% were observed
in rats and dogs, respectively. Regarding cerebral exposure, we evaluated the brain penetration (B/P: 13.5%) and
cerebrospinal fluid penetration (CSF/P: 3.1%) after oral administration to rats (3 mg/kg). The concentration of (−)-
3h reached 2.23 nM in CSF at 1 h, suggesting that (−)-3h had an acceptable brain penetrability enabling exposure
above the IC50 to be attained. Based on these PK profiles, the efficacy of (−)-3h was next evaluated.
Table 4. Pharmacokinetic parameters of (−)-3h in rats and dogs
15

iv
po
PPB
(%)d
dose
Vd
CL
T1/2
dose
Tmax
B/P CSF/P
F
species
(mg/kg) (mL/Kg) (mL/h/kg)
(h)
(mg/kg)
(h)
(%)a
(%)b
(%)c
rat
1
853
294
2370
159
0.238
1.16
3
3
0.333 13.5
0.583
3.1
-
5.5
88.0
96.4
dog
0.5
-
61.6
The pharmacokinetic study was conducted in fasted Sprague-Dawley rats (n = 3) and dogs (n = 3). aB/P (%) was
defined as the brain/plasma ratio. bCSF/P (%) was defined as the cerebrospinal fluid/plasma ratio. cF (%) was
compared between intravenous administration and oral administration. dPPB (%) was defined as plasma protein
binding, conducted at 2g/mL.
Given the results of appropriate brain exposure and half-life in rats, the sleep-promoting effect of compound (−)-3h
was evaluated using PSG (Figure 4). The vehicle or compound (−)-3h at a dose of 1, 3, or 10 mg/kg was orally
administered to rats prior to turning the light off (start of the active phase). By measuring the sleep latency until sleep
lasting 120 seconds or more, a dosing of (−)-3h at 1–10 mg showed a fast onset of action. Furthermore, a dose-
dependent increase in the percentage of total sleep was observed. These data suggested that compound (−)-3h
possessed promising sleep-inducing and sleep-promoting effects.
Figure 4. Effects of compound (−)-3h on sleep parameters in a rat PSG study.
The sleep-promoting effects on the sleep latency (A) and the percentage of total sleep (B) for the 2 h period after the
oral administration of the vehicle or compound (−)-3h are shown. Values are shown as the mean and SEM (n = 10
animals/group). *P <0.05, **P <0.01, ***P <0.001 versus the vehicle-treated group, based on the Bartlett’s test
followed by the Dunnett multiple comparison post-hoc test.
16

Table 5 summarizes the profiles of prototype 1 and (−)-3h. During the course of the generation of (−)-3h, the 1,3-
oxazinane ring in the central core played critical roles in both the high antagonistic activities for the orexin receptors
and the low lipophilicity. These findings demonstrated that (−)-3h had both a strong pharmacological activity in
animal models and a short plasma half-life because of its low Vd. Thus, (−)-3h could be a useful hypnotic with a low
risk of next-day drowsiness. For this reason, (−)-3h was selected as the candidate compound ORN0829 (investigation
code name, TS-142) and was used for further clinical evaluations.
Table 5. Profiles of compounds 1 and (−)-3h
Compound
hOX1R (IC50, nM)
hOX2R (IC50, nM)
rOX1R (Kb, nM)a
rOX2R (Kb, nM)a
LogDHPLC (pH 7.4)
20.6
38.7
5.08
6.06
2.53
1.05
1.27
0.44
0.80
2.58
rat:
dog:
745
598
1.07
rat:
dog:
294
159
1.16
Vd (mL/kg)b
CL (mL/h/Kg)b
T1/2 (h)b
1710
3960
0.412
853
2370
0.238
Predicted humanc
T1/2 (h)
from rat: from dog:
1.1 1.4
from rat: from dog:
0.9 2.0
Rat PSG model
MED
3 mg/kg ip
1 mg/kg po
17

aThe Kb values are the mean of multiple results (at least three independent determinations performed in duplicate)
with the standard error of the means. bThe pharmacokinetic study was conducted in fasted Sprague-Dawley rats (n =
3) and dogs (n = 3) dosed at 2 mg/kg and 0.5 mg/kg for intravenous dosing. Csss method.
Conclusion
In conclusion, we identified 3-benzoyl-1,3-oxazinane (−)-3h as a highly potent DORA through iterative design
and compound optimization based on the structural framework of pyrazolylethylbenzamide 1, keeping the high
antagonist activity, low lipophilicity and short half-life PK profile in mind. The unique structure of (−)-3h, an 1,3-
oxazinane moiety, may play key roles in the formation of its bioactive U-shaped conformation and in lowering the
lipophilicity. Compound (−)-3h exhibited potent sleep-promoting effects at a po dose from 1 mg/kg in a rat
polysomnogram study, and optimal PK properties with rapid Tmax and short half-lives were observed in rats and dogs,
indicating a predicted human half-life of 0.9–2.0 h. Given the preclinical profiles as a useful hypnotic with a low risk
of next-day drowsiness, (−)-3h (ORN0829; investigation code name, TS-142) was selected as a candidate compound
and is currently in clinical development toward the treatment of insomnia.
Experimental section
Biology
Antagonistic activity against human and rat OX1R and OX2R
The antagonistic activities of the test compounds were determined as described previously.20 Chinese hamster ovary
(CHO)-K1 cells stably expressing recombinant human OX1R, human OX2R, rat OX1R or rat OX2R were seeded into
a 96-well black/clear bottom plate at a density of 2.4 × 104 cells/well one day before the experiment. The cells were
incubated with loading buffer (Hank’s balanced salt solution [HBSS] [pH7.4] containing 20 mM HEPES, 0.1% BSA,
0.2 mg/ml amaranth, 2.5 mM probenecid, 0.01% pluronic F-127 and 0.5 μM of Fluo-4AM [Invitrogen]) at 37°C for
1 h. After the loading buffer was removed, the cells were incubated with assay buffer (HBSS [pH7.4] containing 20
mM HEPES, 0.1% BSA, 0.2 mg/ml amaranth, and 2.5 mM probenecid) with or without various concentrations of
the test compounds for 30 min at room temperature. The changes in the intracellular Ca2+ concentration were
determined by monitoring the changes in fluorescence using a Functional Drug Screening System (Hamamatsu
18

Photonics, Shizuoka, Japan) after the application of [Ala6, 12]orexin-A (Peptide Institute, Osaka, Japan) (final
concentration of 0.5 nM for human OX1R, 1 nM for human OX2R, 10 nM for rat OX1R or 30 nM for rat OX2R).
The concentration-response curves were fit using nonlinear regression analyses; the 50% inhibitory concentration
(IC50) was calculated using GraphPad Prism software (version 5.04; GraphPad Software Inc., San Diego, CA). The
equilibrium dissociation constant (Kb) values were calculated according to the equation “Kb = IC50/{1 +
([concentration of Ala-OXA used in the determination of IC50 value]/[50% effective concentration of Ala-OXA])}”
using SAS (version 9.2; SAS Institute Japan Ltd., Tokyo, Japan). The experiments were performed two to five times,
and each test was performed in duplicate or triplicate.
Effect on sleep-wake states during a dark period in rats
The effect of (−)-3h on sleep-wake states was evaluated during a dark period in rats. Male Wistar rats (10 rats/group,
10-12 weeks of age, Charles River Laboratories, Japan) that had been surgically implanted with electrodes for
electroencephalogram (EEG) and electromyogram (EMG) studies were used. The rats were individually transferred
to an acrylic chamber placed within an electrically shielded sound-proof box (light on: 5:00-17:00) 7-8 hours before
the administration of (−)-3h to allow them to acclimate to the measurement environment. Rats were treated with (−)-
3h at doses of 1, 3 and 10 mg/kg orally as a free base within 10 minutes prior to the onset of the dark period. The
EEG and EMG recordings were continued for 6 hours after the administration. The sleep–wake states were classified
every 20 seconds as wakefulness or sleep based on the EEG and EMG patterns. The rate of the sleep-wake state was
calculated for 2 hours after administration. Sleep onset latency was defined as latency until the first sleep persisting
for 120 seconds or more after administration. (−)-3h was suspended in 0.5% methyl cellulose 400 solution.
Chemistry
General methods
All the solvents and reagents were obtained from commercial suppliers and used without further purification unless
otherwise noted. The preparative HPLC purification conditions were as follows: Gilson preparative HPLC system;
column Waters ODS sunfire, 50 mm × 30 mm; eluent A, water + 0.1% CF3CO2H; eluent B, CH3CN + 0.1%
CF3CO2H; 10% B up to 95% B in 12 min; flow rate 40 mL/min. Silica gel column chromatography was performed
19

using a Kanto 60 N silica gel or an appropriately sized pre-packed silica cartridge (Biotage® SNAP Ultra or SNAP
KP-NH) on a Biotage system. Preparative supercritical fluid chromatography (SFC) was performed using a Waters
SFC30 instrument. Preparative HPLC for the chiral compounds was performed using a Gilson preparative HPLC
system. The 1H and 13C NMR spectra were recorded on a JEOL 600 MHz and 500 MHz NMR spectrometer, and all
the chemical shifts were reported in parts per million (ppm) relative to tetramethylsilane (TMS,  0.00 ppm) as an
internal standard. High resolution mass spectral data were acquired using a Shimadzu LCMS–IT–TOF equipped with
an ESI/APCI dual ion source. The final compounds exhibited ≥95% purity, as determined using HPLC and LC–MS
on an Agilent instrument using electrospray ionization. The HPLC conditions were as follows: Shimadzu LC–20A;
column Shim–Pack XR–ODS 2.2 μm, 3.0 mm × 75 mm; eluent A, water + 0.1% phosphoric acid; eluent B, CH3CN;
1) 10% B in 4.0 min, 10%–90% B in 5.0 min, 90% B in 2.0 min, 2) 20% B in 4.0 min, 20%–90% B in 5.0 min, 90%
B in 2.0 min or 3) 40% B in 4.0 min, 40%–90% B in 5.0 min, 90% B in 2.0 min; flow rate 0.8 mL/min; UV detection
(λ = 210 nm). The LC–MS conditions were as follows: Agilent 1290 infinity and Agilent 6150; column Waters
Acquity CSH C18, 1.7 μm, 2.1 mm × 50 mm; eluent A, water + 0.1% formic acid; eluent B, CH3CN + 0.1% formic
acid; 20%–99% B in 1.2 min, 99% B in 0.2 min; flow rate 0.8 ml/min; UV detection (λ = 254 nm). Optical rotations
were measured using a Rudolph Research Analytical AUTOPOL V. Infrared spectroscopy (IR) was performed using
a Shimadzu IRAffinity-1. A single crystal X-ray structure analysis was performed using Rigaku R-AXIS RAPIDⅡ.
The orexin receptor antagonists described in this paper exist in multiple conformations as a result of hindered
rotations that were slow on the NMR timescale. The 1H and 13C NMR spectra of some compounds consisted of broad
and complicated multiplets for which a detailed coupling constant was difficult to analyze. Thus, the NMR resonances
of these compounds were not listed in numerical format. Instead, we included pictures of the 1H and 13C NMR spectra
for these compounds at 25°C in the Supplementary Data. For compound (−)-3h only, the 1H and 13C chemical shifts
were assigned based on an analysis of the COSY, HSQC, and HMBC spectra acquired for the sample dissolved in
CDCl3 at 25°C. Compounds 17, 27, and 28 were prepared according to previously reported procedures.20
Benzyl (4R)-4-(hydroxymethyl)-1,3-oxazolidine-3-carboxylate ((R)-16d)
4-Methylmorpholine (1.0 mL, 9.2 mmol) and isobutyl chloroformate (1.4 mL, 11 mmol) were successively added
to a solution of (4S)-3-[(benzyloxy)carbonyl]-1,3-oxazolidine-4-carboxylic acid (S)-15d (2.1 g, 8.4 mmol) in 1,2-
20

dimethoxyethane (20 mL) at −15°C. The mixture was stirred at room temperature for 1 h and then filtered to remove
a colorless solid. To the filtrate, a solution of NaBH4 (0.47 g, 13 mmol) in water (10 mL) was added at −15°C within
30 min. The reaction mixture was stirred at −10°C for 1 h. The reaction was quenched by the addition of water, and
the mixture was extracted with EtOAc. The organic layer was washed with saturated NH4Cl aqueous solution,
saturated NaHCO3 aqueous solution, and brine. The organic layer was dried over MgSO4, filtered, and concentrated
under reduced pressure. The resulting residue was purified using column chromatography (SiO2, 20%–80% EtOAc
in hexanes) to yield the title compound (R)-16d as a colorless oil (1.2 g, 61% yield). HRMS calcd for C12H15NO4
[M+H] + 238.1074 found 238.1045; LC–MS t = 0.67 min, [M+H]+ = 238. 1H NMR (500 MHz, CDCl3) δ 7.30–7.43
(m, 5H), 5.12–5.21 (m, 2H), 5.00 (br s, 1H), 4.81 (d, J = 4.12 Hz, 1H), 4.01–4.10 (m, 2H), 3.75–3.83 (m, 1H), 3.63–
3.74 (m, 2H), 3.32–3.54 (m, 1H); 13C NMR (125 MHz, CDCl3) δ 135.9, 128.6, 128.4, 128.1, 79.3, 69.4, 67.6, 63.8,
58.0.
tert-Butyl (4R)-4-(hydroxymethyl)-1,3-thiazolidine-3-carboxylate ((R)-16f)
A borane-tetrahydrofuran complex (ca. 0.9 mol/L; 5.0 mL, 4.5 mmol) was added to a solution of (4R)-3-tert-
butoxycarbonylthiazolidine-4-carboxylic acid (R)-15f (1.0 g, 4.3 mmol) in THF (6.1 mL) at 0°C. After stirring at
0°C for 5 min, the reaction mixture was allowed to warm to room temperature and stirred for 3 h. The reaction was
quenched by the addition of water, and the mixture was extracted using EtOAc. The combined organic layer was
dried over Na2SO4 and then concentrated under reduced pressure. The resulting residue was purified using column
chromatography (SiO2, 30%–80% EtOAc in hexanes) to yield the title compound (R)-16f as a colorless oil (0.83 g,
88% yield). HRMS calcd for C9H17NO3S [M+Na] + 242.0821 found 242.0814; 1H NMR (500 MHz, CDCl3, 50 °C) δ
4.63 (d, J = 9.60 Hz, 1H), 4.31–4.39 (m, 1H), 4.26 (d, J = 9.61 Hz, 1H), 3.71 (t, J = 5.83 Hz, 2H), 3.14 (dd, J = 6.52,
11.66 Hz, 1H), 2.92 (br d, J = 9.61 Hz, 1H), 1.49 (s, 9H); 13C NMR (125 MHz, CDCl3, 50 °C) δ 154.7, 81.5, 64.5,
61.9, 49.2, 33.2, 28.6; MS (ESI/APCI dual): m/z 220 [M+H]+.
Benzyl (4R)-4-{[3-(5-fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}-1,3-oxazolidine-3-carboxylate ((R)-18d)
Et3N (0.84 mL, 6.0 mmol) and methanesulfonyl chloride (0.37 mL, 4.8 mmol) were added to a solution of (R)-16d
21

(0.95 g, 4.0 mmol) in THF (10 mL) at 0°C. The mixture was stirred at room temperature for 1 h and then filtered
through a pad of KC flock® (cellulose powder; Nippon paper group). The filtrate was concentrated under reduced
pressure to yield benzyl (4S)-4-{[(methanesulfonyl)oxy]methyl}-1,3-oxazolidine-3-carboxylate. To a solution of the
methanesulfonate in DMSO (10 mL) were added 5-fluoro-2-(1H-pyrazol-3-yl)pyridine 17 (0.85 g, 5.2 mmol) and t-
BuONa (0.58 g, 6.0 mmol) at room temperature. The mixture was stirred at 90°C for 1 h. The reaction was quenched
by the addition of water, and the mixture was extracted with EtOAc. The organic layer was washed with brine, dried
over MgSO4, filtered, and concentrated under reduced pressure. The resulting residue was purified using column
chromatography (KP-NH, 20%–80% EtOAc in hexanes) to yield the title compound (R)-18d as a pale yellow oil
(0.81 g, 53% yield in 2 steps). HRMS calcd for C20H19FN4O3 [M+H] + 383.1514 found 383.1491; LC–MS t = 0.98
min, [M+H]+ = 383. Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra
in CDCl3 at 25°C.
[(4R)-4-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}-1,3-oxazolidin-3-yl][5-methyl-2-(2H-1,2,3-
triazol-2-yl)phenyl]methanone ((−)-3d)
Et3N (0.55 mL, 3.9 mmol) and 20% Pd(OH)2 on carbon (0.10 g) were added to a mixture of (R)-18d (0.58 g, 1.5
mmol) in THF (5.0 mL). The mixture was charged with H2 and stirred at room temperature for 3 h. The reaction
mixture was filtered through a pad of Celite® to yield a colorless solution of 5-fluoro-2-(1-{[(4R)-1,3-oxazolidin-4-
yl]methyl}-1H-pyrazol-3-yl)pyridine. Thionyl chloride (0.095 mL, 1.3 mmol) and DMF (3 drops) were added to a
mixture of 5-methyl-2-(2H-1,2,3-triazol-2-yl)benzoic acid (0.24 g, 1.2 mmol) in toluene (5.0 mL) at room
temperature, and the mixture was stirred at 80°C for 1 h. The reaction mixture was cooled to room temperature, and
the volatiles were distilled under reduced pressure to yield 5-methyl-2-(2H-1,2,3-triazol-2-yl)benzoyl chloride 19.
A solution of 19 in toluene (5.0 mL) was added dropwise at 0°C to a mixture of Et3N (0.41 mL, 3.0 mmol) and the
colorless solution of 5-fluoro-2-(1-{[(4R)-1,3-oxazolidin-4-yl]methyl}-1H-pyrazol-3-yl)pyridine obtained in the
previous reaction. The mixture was allowed to warm to room temperature and then stirred for 1 h. The reaction was
quenched by the addition of saturated NaHCO3 aqueous solution, and the mixture was extracted with EtOAc. The
organic layer was washed with brine, dried over MgSO4, filtered, and concentrated under reduced pressure. The
22

resulting residue was purified by column chromatography (KP-NH, 10%–80% EtOAc in hexanes) to yield a
mixture of the title compound (−)-3d and its region-isomer (0.17 g, 33% yield, (−)-3d : isomer = 17 : 1). The
mixture (0.050 g, 0.12 mmol) was separated using preparative SFC (CHIRALPAK® AD-H column 2 cm * 25 cm,
40% MeOH/CO2, flow rate = 30 mL/min), providing (−)-3d as a colorless amorphous substance (0.037 g, 74%
yield). HRMS calcd for C22H20FN7O2 [M+H] + 434.1735 found 434.1723; LC–MS t = 0.91 min, [M+H]+ = 434; []
20
D
–65.6 (c 0.970, CHCl3); IR (KBr, cm−1) 825, 1111, 1226, 1427, 1495, 1651, 2877, 3118. Please see the
Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
[(2S)-2-(Hydroxymethyl)pyrrolidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((S)-20a)
A solution of 19 (0.554 g, 2.46 mmol) and Et3N (0.686 mL, 4.92 mmol) in CHCl3 (5.00 mL) was added in a dropwise
manner to a solution of L-prolinol (S)-16a (0.299 g, 2.95 mmol) in CHCl3 (3.00 mL) at 0°C and within 15 min. The
reaction mixture was stirred at room temperature for 1 h. The reaction was quenched by the addition of saturated
NaHCO3 aqueous solution, and the mixture was extracted with CHCl3. The organic layer was allowed to pass through
an ISOLUTE® phase separator, and the filtrate was concentrated under reduced pressure. The resulting residue was
purified using column chromatography (SiO2, 20%–100% EtOAc in hexanes) to yield the title compound (S)-20a as
a colorless oil (0.465 g, 66% yield). HRMS calcd for C15H18N4O2 [M+H] + 287.1503 found 287.1497; LC–MS t =
0.71 min, [M+H]+ = 287; []2D6 –76.9 (c 0.775, CHCl3). Please see the Supplementary Data for pictures of 500 MHz
1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
Preparation of compounds (R)-20a, (S or R)-20b, (R)-20f, and (R)-20g
These compounds were synthesized from their corresponding precursors according to the procedure described for
compound (S)-20a.
[(2R)-2-(Hydroxymethyl)piperidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((R)-20a)
The title compound was prepared from D-prolinol (R)-16a (78% yield, colorless oil). C15H18N4O2 [M+H]+ 287.1503
found 287.1486; LC–MS t = 0.71 min, [M+H]+ = 287; []2D0 +70.5 (c 1.02, CHCl3). Please see the Supplementary
Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25 °C.
23

[(2S)-2-(Hydroxymethyl)piperidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((S)-20b)
The title compound was prepared from [(2S)-piperidin-2-yl]methanol (S)-16b (32% yield, colorless solid). HRMS
calcd for C16H20N4O2 [M+H]+ 301.1659 found 301.1630; LC–MS t = 0.73 min, [M+H]+ = 301. Please see the
Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25 °C.
[(2R)-2-(Hydroxymethyl)piperidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((R)-20b)
The title compound was prepared from [(2R)-piperidin-2-yl]methanol (R)-16b (41% yield, colorless solid). HRMS
calcd for C16H20N4O2 [M+H]+ 301.1659 found 301.1629; LC–MS t = 0.73 min, [M+H]+ = 301. Please see the
Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25 °C.
[(4R)-4-(Hydroxymethyl)-1,3-thiazolidin-3-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((R)-20f)
Hydrogen chloride in EtOAc (4.0 mol/L, 7.33 mL, 29.3 mmol) was added to a solution of (R)-16f (0.643 g, 2.93
mmol) in EtOAc (7.00 mL) at 0°C, and the reaction mixture was stirred at room temperature for 2 h. The organic
solvent was removed under reduced pressure to yield [(4S)-1,3-thiazolidin-4-yl]methanol hydrochloride as a colorless
amorphous substance (0.345 g, 76% yield). The title compound was prepared from [(4S)-1,3-thiazolidin-4-
yl]methanol hydrochloride according to the procedure described for compound (S)-20a (79% yield, colorless oil).
HRMS calcd for C14H16N4O2S [M+H]+ 305.1067 found 305.1060; LC–MS t = 0.70 min, [M+H]+ = 305; []2D0 −144
(c 1.15, CHCl3). Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in
CDCl3 at 25°C.
[(3R)-3-(Hydroxymethyl)morpholin-4-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((R)-20g)
The title compound was prepared from [(3R)-morpholin-3-yl]methanol hydrochloride (R)-16g (40% yield, colorless
oil). HRMS calcd for C15H18N4O3 [M+H] + 303.1452 found 303.1431. LC–MS t = 0.57 min, [M+H]+ = 303. Please
see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
24

[(2S)-2-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}pyrrolidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-
yl)phenyl]methanone ((−)-3a)
Et3N (0.20 mL, 1.4 mmol) and methanesulfonyl chloride (0.082 mL, 1.1 mmol) were added to a solution of (S)-20a
(0.27 g, 0.96 mmol) in THF (3.2 mL) at 0°C. The reaction mixture was stirred at 0°C for 30 min and then filtered
through a pad of KC flock®. The filtrate was concentrated under reduced pressure to yield {(2S)-1-[5-methyl-2-(2H-
1,2,3-triazol-2-yl)benzoyl]pyrrolidin-2-yl}methyl methanesulfonate as a colorless oil. 5-Fluoro-2-(1H-pyrazol-3-
yl)pyridine 17 (0.16 g, 0.96 mmol) and Cs2CO3 (0.62 g, 1.9 mmol) were then added to a solution of the
methanesulfonate in DMF (0.96 mL) at room temperature. The reaction mixture was stirred at 80°C for 2 h. The
reaction was quenched by the addition of water, and the mixture was extracted with EtOAc. The organic layer was
washed with brine, dried over Na2SO4, filtered, and concentrated under reduced pressure. The resulting residue was
purified using column chromatography (SiO2, 20%–80% EtOAc in hexanes) to yield a mixture of the title compound
(−)-3a and its regioisomer (0.22 g, 63% yield in 2 steps, (−)-3a : regioisomer = 8 : 1). The mixture (0.22 g) was then
separated using a CHIRALPAK® AD-H column (2 cm * 25 cm) with 40% iPrOH in hexanes and a flow rate of 10
mL/min to yield the title compound (−)-3a as a colorless oil (0.093 g, 42% yield). HRMS calcd for C23H22FN7O
[M+H] + 432.1943 found 432.1923; LC–MS t = 1.01 min, [M+H]+ = 432; []2D6 –120 (c 1.05, CHCl3). Please see the
Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
Preparation of compounds (+)-3a, (− or +)-3b, (−)-3f, and (−)-3g
These compounds were synthesized from their corresponding precursors ((R)-20a, (S)-20b, (R)-20b, (R)-20f, and
(R)-20g) according to the procedure as described for compound (−)-3a.
[(2R)-2-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}pyrrolidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-
yl)phenyl]methanone ((+)-3a)
The title compound was prepared from (R)-20a (10% yield in 2 steps, colorless amorphous substance). HRMS calcd
for C23H22FN7O [M+H] + 432.1943 found 432.1924; LC–MS t = 1.01 min, [M+H]+ = 432; []2D0 +106 (c 0.410,
CHCl3). Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at
25

25°C.
[(2S)-2-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}piperidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-
yl)phenyl]methanone ((−)-3b)
The title compound was prepared from (S)-20b (20% yield in 2 steps, colorless solid). HRMS calcd for C24H24FN7O
[M+H] + 446.2099 found 446.2080; LC–MS t = 1.02 min, [M+H]+ = 446; []2D0 –35.6 (c 1.29, CHCl3); IR (KBr,
cm−1) 821, 1225, 1457, 1496, 1623, 2933, 3101. Please see the Supplementary Data for pictures of 500 MHz 1H and
125 MHz 13C NMR spectra in CDCl3 at 25°C.
[(2R)-2-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}piperidin-1-yl][5-methyl-2-(2H-1,2,3-triazol-2-
yl)phenyl]methanone ((+)-3b)
The title compound was prepared from (R)-20b (8.4% yield in 2 steps, colorless solid). HRMS calcd for
C24H24FN7O [M+H]+ 446.2099 found 446.2086; LC–MS t = 0.99 min, [M+H]+ = 446; []2D0 +34.0 (c 0.659, CHCl3).
Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
[(4R)-4-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}-1,3-thiazolidin-3-yl][5-methyl-2-(2H-1,2,3-
triazol-2-yl)phenyl]methanone ((−)-3f)
The title compound was prepared from (R)-20f (17% yield in 2 steps, colorless amorphous substance). HRMS calcd
for C22H20FN7OS [M+H]+ 450.1507 found 450.1500; LC–MS t = 0.99 min, [M+H]+ = 450; []2D0 –74.1 (c 0.800,
CHCl3). Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at
25°C.
[(3R)-3-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}morpholin-4-yl][5-methyl-2-(2H-1,2,3-triazol-2-
yl)phenyl]methanone ((−)-3g)
The title compound was prepared from (R)-20g (24% yield in 2 steps, colorless amorphous substance). HRMS calcd
26

for C23H22FN7O2 [M+H] + 448.1892 found 448.1871; LC–MS t = 0.90 min, [M+H]+ = 448; []2D0 –47.9 (c 0.791,
CHCl3); IR (KBr, cm−1) 825, 1114, 1228, 1458, 1496, 1642, 2863, 3117. Please see the Supplementary Data for
pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
(4S)-3-[5-Methyl-2-(2H-1,2,3-triazol-2-yl)benzoyl]-1,3-oxazinane-4-carboxylic acid ((S)-22)
L-Homoserine lactone hydrochloride 21 (2.55 g, 18.5 mmol) was added to a mixture of formaldehyde (37% in water,
9.22 mL, 90.8 mmol) and hydrochloric acid solution (1.0 mol/L, 1.74 mL) in water (44.0 mL). The reaction mixture
was stirred at room temperature for 17 h. The solvent was then distilled under reduced pressure to yield a white solid,
which was then suspended in EtOH (100 mL). After stirring for 15 min, the insoluble matter was filtered off, and the
filtrate was concentrated under reduced pressure to yield approximately 25 mL of crude residue. After the addition
of EtOAc (25 mL), the crude residue was stored at −10°C overnight. The precipitate was then collected by filtration
to yield (4S)-1,3-oxazinane-4-carboxylic acid hydrochloride as a colorless solid (2.70 g, 86% yield).
A solution of benzoyl chloride 19 (0.38 g, 1.7 mmol) and Et3N (0.41 mL, 3.0 mmol) in CHCl3 (11 mL) were added
to a mixture of (4S)-1,3-oxazinane-4-carboxylic acid hydrochloride (0.40 g, 2.4 mmol) and Et3N (0.83 mL, 5.9 mmol)
in CHCl3 (5.0 mL) in a dropwise manner at 0°C within 20 min. The solvent was then distilled under reduced pressure.
The resulting residue was purified using preparative HPLC to yield the title compound (S)-22 as a colorless solid
(0.18 g, 33% yield). HRMS calcd for C15H16N4O4 [M+H] + 317.1244 found 317.1223; LC–MS t = 0.67 min, [M+H]+
= 317. Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in DMSO-d6
at 25°C.
[(4S)-4-(Hydroxymethyl)-1,3-oxazinan-3-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone ((S)-23)
The title compound was prepared from (S)-22 according to the procedure described for compound 16d (74% yield,
colorless solid). HRMS calcd for C15H18N4O3 [M+H] + 303.1452 found 303.1456. LC–MS t = 0.59 min, [M+H]+ =
1
303. Please see the Supplementary Data for pictures of 500 MHz H and 125 MHz 13C NMR spectra in CDCl3 at
25°C.
27

[(4S)-4-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}-1,3-oxazinan-3-yl][5-methyl-2-(2H-1,2,3-triazol-
2-yl)phenyl]methanone ((−)-3i)
The title compound was prepared from (S)-23 according to the procedure described for compound 18d (17% yield,
colorless solid). HRMS calcd for C23H22FN7O2 [M+H] + 448.1892 found 448.1880; LC–MS t = 0.87 min, [M+H]+ =
448; []2D0 –50.0 (c 0.730, CHCl3); IR (KBr, cm−1) 820, 1055, 1228, 1431, 1458, 1625, 2958, 3134. Please see the
Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in CDCl3 at 25°C.
[2-(Hydroxymethyl)-1,3-thiazolidin-3-yl][5-methyl-2-(2H-1,2,3-triazol-2-yl)phenyl]methanone (26)
3-[5-Methyl-2-(2H-1,2,3-triazol-2-yl)benzoyl]-1,3-thiazolidine-2-carboxylic acid 25 was prepared from 1,3-
thiazolidine-2-carboxylic acid 24 according to the procedure described for compound (S)-20a (80% yield in 2 steps,
colorless oil). LC–MS t = 0.73 min, [M+H]+ = 319.
The title compound was prepared from 25 according to the procedure described for compound (R)-16d (79% yield,
colorless oil). LC–MS t = 0.69 min, [M+H]+ = 305. 1H NMR (600 MHz, CDCl3) δ 7.89 (d, J = 8.26 Hz, 1H), 7.81 (s,
2H), 7.37 (br d, J = 8.26 Hz, 1H), 7.31 (br s, 1H), 5.74 (s, 1H), 3.63 (br s, 1H), 3.23–3.52 (m, 1H), 3.10–3.21 (m,
1H), 2.81–2.90 (m, 1H), 2.44 (s, 3H); 13C NMR (150 MHz, CDCl3) δ 173.8, 168.1, 138.8, 136.0, 136.0, 131.3, 128.5,
128.2, 128.2, 121.9, 59.6, 51.3, 30.6, 20.9.
[(2S)-2-{[3-(5-Fluoropyridin-2-yl)-1H-pyrazol-1-yl]methyl}-1,3-thiazolidin-3-yl][5-methyl-2-(2H-1,2,3-
triazol-2-yl)phenyl]methanone ((−)-3e)
The title compound 3e (racemate) was prepared from 26 according to the procedure described for compound (−)-
3a (26% yield, colorless solid). The chiral resolution of racemate 3e (41 mg, 0.091 mmol) was performed using a
CHIRALPAK® AD-H column (2 cm * 25 cm) with 50% iPrOH in hexanes with a flow rate of 10 mL/min to yield
the title compound (−)-3e as a colorless solid (9.8 mg, 24% yield, tR = 23–26 min [(+)-form: 15–19 min]). HRMS
calcd for C22H20FN7OS [M+H] + 450.1507 found 450.1495; LC–MS t = 1.08 min, [M+Na]+ = 472; []2D0 –247 (c
0.490, CHCl3). Please see the Supplementary Data for pictures of 500 MHz 1H and 125 MHz 13C NMR spectra in
CDCl3 at 25°C.
28

2-[1-(2,2-Diethoxyethyl)-1H-pyrazol-3-yl]-5-fluoropyridine (29)
2-Bromo-1,1-diethoxyethane (6.0 mL, 40 mmol) was added to a stirred suspension of 5-fluoro-2-(1H-pyrazol-3-
yl)pyridine 17 (5.0 g, 31 mmol) and Cs2CO3 (20 g, 61 mmol) in DMF (100 mL) in a dropwise manner within 5 min
at room temperature. The reaction mixture was stirred at 80°C for 4 h. After cooling to room temperature, water was
added, and the mixture was extracted with EtOAc. The organic layer was dried over Na2SO4, filtered, and
concentrated under reduced pressure. The resulting residue was purified using column chromatography (SiO2, 30%–
80% EtOAc in hexanes) to yield the title compound 29 as a colorless solid (6.4 g, 75% yield). HRMS calcd for
C14H18FN3O2 [M+H] + 280.1456 found 280.1450; LC–MS t = 0.91 min, [M+H]+ = 280; 1H NMR (500 MHz, CDCl3,
25 °C) δ 8.48 (d, J = 2.74 Hz, 1H), 7.92 (dd, J = 4.12, 8.92 Hz, 1H), 7.52 (d, J = 2.40 Hz, 1H), 7.43 (dt, J = 3.10,
7.90 Hz, 1H), 6.79 (d, J = 2.40 Hz, 1H), 4.82 (t, J = 5.49 Hz, 1H), 4.26 (d, J = 5.49 Hz, 2H), 3.72 (qd, J = 7.09, 9.26
Hz, 2H), 3.44 (qd, J = 7.03, 9.43 Hz, 2H), 1.17 (t, J = 7.03 Hz, 6H); 13C NMR (125 MHz, CDCl3, 25 °C) δ 163.7,
161.7, 151.0, 132.3, 130.1, 130.0, 127.5, 127.4, 115.8, 115.6, 102.8, 101.8, 64.0, 55.5, 15.5.
Preparation of compounds 30 and 31
These compounds were synthesized from their corresponding precursors according to the procedure described for
compound 29.
2-[1-(2,2-Diethoxyethyl)-1H-pyrazol-4-yl]-5-fluoropyridine (30)
The title compound was prepared from 5-fluoro-2-(1H-pyrazol-4-yl)pyridine 27 (91% yield, colorless solid). HRMS
calcd for C14H18FN3O2 [M+H]+ 280.1456 found 280.1436; LC–MS t = 0.88 min, [M+H]+ = 280; 1H NMR (600 MHz,
CDCl3) δ 8.41 (d, J = 2.48 Hz, 1H), 7.94 (d, J = 3.72 Hz, 2H), 7.44 (dd, J = 4.13, 8.67 Hz, 1H), 7.38 (dt, J = 3.30,
8.67 Hz, 1H), 4.82 (t, J = 5.37 Hz, 1H), 4.24 (d, J = 5.37 Hz, 2H), 3.68–3.76 (m, 2H), 3.43–3.50 (m, 2H), 1.17 (t, J
= 7.02 Hz, 6H).
1-(2,2-Diethoxyethyl)-4-(4-fluorophenyl)-1H-pyrazole (31)
29