Discovery of a novel bicyclic compound, DS54360155, as an orally potent analgesic without mu-opioid receptor agonist activity

Article abstract of DOI:10.1016/j.bmcl.2019.126748

We synthesized derivatives of a natural alkaloid, conolidine, and evaluated these derivatives in the acetic acid-induced writhing test and formalin test in ddY mice after oral administration. As a result, we identified (5S)-6-methyl-1,3,4,5,6,8-hexahydro-7H-2,5-methano[1,5]diazonino[7,8-b]indol-7-one sulfate salt, 15a (DS54360155), with a unique and original bicyclic skeleton, as an analgesic more potent than conolidine. Moreover, 15a did not exhibit mu-opioid receptor agonist activity.

Full text of DOI:10.1016/j.bmcl.2019.126748

Journal Pre-proofs
Discovery of a novel bicyclic compound, DS54360155, as an orally potent an-
algesic without mu-opioid receptor agonist activity
Tsuyoshi Arita, Masayoshi Asano, Kazufumi Kubota, Yuki Domon, Nobuo
Machinaga, Kousei Shimada
PII:
S0960-894X(19)30711-5
DOI:
https://doi.org/10.1016/j.bmcl.2019.126748
BMCL 126748
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
27 March 2019
27 September 2019
8 October 2019
Please cite this article as: Arita, T., Asano, M., Kubota, K., Domon, Y., Machinaga, N., Shimada, K., Discovery of
a novel bicyclic compound, DS54360155, as an orally potent analgesic without mu-opioid receptor agonist activity,
Bioorganic & Medicinal Chemistry Letters (2019), doi: https://doi.org/10.1016/j.bmcl.2019.126748
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.
© 2019 Published by Elsevier Ltd.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com
Discovery of a novel bicyclic compound, DS54360155, as an orally potent
analgesic without mu-opioid receptor agonist activity
Tsuyoshi Aritaa, Masayoshi Asanoa, Kazufumi Kubotab, Yuki Domonb, Nobuo Machinagaa, Kousei Shimadaa
a
Medicinal Chemistry Research Laboratories, R&D Division, Daiichi Sankyo Co., Ltd., 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan.
Specialty Medicine Research LaboratoriesⅠ, R&D Division, Daiichi Sankyo Co., Ltd., 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan.
b
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
We synthesized derivatives of a natural alkaloid, conolidine, and evaluated these derivatives in
the acetic acid-induced writhing test and formalin test in ddY mice after oral administration. As a
result, we identified (5S)-6-methyl-1,3,4,5,6,8-hexahydro-7H-2,5-methano[1,5]diazonino[7,8-
b]indol-7-one sulfate salt, 15a (DS54360155), with a unique and original bicyclic skeleton, as an
analgesic more potent than conolidine. Moreover, 15a did not exhibit mu-opioid receptor agonist
activity.
Keywords:
Conolidine
Indole alkaloid
Analgesic
2019 Elsevier Ltd. All rights reserved.
Acetic acid writhing test
Non-opioid analgesic
———
 Corresponding author. Tel: +81-3-3492-3131; fax: +81-3-5436-8563; e-mail: arita.tsuyoshi.sg@daiichisankyo.co.jp

Conolidine (1; Figure 1) is a natural indole alkaloid that was first isolated from Tabernaemontana divaricata by Kam et al.1 However,
few studies have evaluated the biological activity of 1. This is because 1 can only be isolated in small amounts from the bark of the plant.
In 2011, Micalizio et al. reported the first successful asymmetric total synthesis of 1. They also reported that synthetic 1 exhibited potent
non-opioid analgesic effects in mouse models.2,3 Since these studiesy, several other groups have reported efficient synthesis4-6 of this
compound and elucidated its mechanism of action.7 Although the mechanism of action of 1 is not completely understood, we are interested
in its unique analgesic properties that may render it a promising alternative to opioids, which have severe opioid-related adverse effects,8,9
such as constipation, nausea, vomiting, pruritus, somnolence, cognitive impairment, respiratory depression, tolerance, physical
dependence, and addiction. In this study, we synthesized a series of novel derivatives of 1 and evaluated their structure-activity
relationships (SARs) in vivo. As a result, we obtained a potent analgesic 15a (DS54360155) without mu-opioid receptor agonist activity.
N
N
H
O
Conolidine (1)
Figure 1. Structure of conolidine (1)
This study aimed to identify the derivatives of 1 that are orally bioavailable and exhibit safe and potent analgesic effects in an animal
pain model. In this regard, to determine the in vivo SARs, we evaluated the effects of the synthesized derivatives on mouse performance
during an acetic acid-induced writhing test. The test was used to measure the reduction (% inhibition) in acetic acid-induced writhing
behavior in ddY mice, which are outbred mice maintained in closed colonies, in response to the oral administration of the compounds (30
mg/kg). Furthermore, our preliminary screening demonstrated that (−)-1 was a potent inhibitor of the human ether-a-go-go-related gene
(hERG) potassium channel (63% inhibition at 10 µM). The hERG encodes a potassium ion channel responsible for myocardial
repolarization. The inhibition of the hERG channel may cause arrhythmia due to QT prolongation. Therefore, overall, we aimed to identify
derivatives of 1 that exhibited strong in vivo analgesic activity as well as weak inhibition of the hERG channel.
We hypothesized that 1 derivatives, which exhibit lower lipophilicity (measured as a distribution coefficient; LogD), may reduce the
hERG channel inhibition. However, the introduction of polar groups, such as the hydroxyl or carboxyl group, to the indole group of 1
resulted in the attenuation of analgesic activity in vivo. In addition, the conversion of indole moiety was synthetically difficult. Hence, we
fixed the indole moiety and transformed the bicyclic skeleton. Although the conversion of the bicyclic ring was also not easy, we
tenaciously attempted the synthesis of various derivatives. Finally, we could stably synthesize several 8-membered ring derivatives. The
data of the synthesized conolidine derivatives (2–5) are summarized in Table 1. All the derivatives were sulfate salts, and compounds 2
and 5 were racemic forms. Compound 2 was a novel compound converted to the [4.3.2] bicyclic skeleton from the [4.2.2] bicyclic skeleton
of 1. Compounds 3 and 4 were simplified to a monocyclic 8-membered ring from a sterically bulky bicyclic structure to reduce the LogD
value. Compound 5 was an original compound that had an amide bond in the [5.2.1] bicyclic skeleton. Among the synthesized derivatives,
compound 5 had potent analgesic activity (83% inhibition at 30 mg/kg), whereas compounds 2, 3, and 4 showed weaker activity than 1.
From the results of compounds 1, 2, and 5, we considered that the difference in bicyclic 8-membered ring structure greatly affected the
analgesic activity in vivo. In addition, we observed a general correlation between the LogD value of the synthesized derivatives and hERG
inhibition. As a result, compound 5, which had the lowest LogD (0.7), exhibited the weakest hERG channel inhibition (26% inhibition at
10 µM). Bicyclic compounds 2 and 5 exhibited a higher metabolic stability than compounds 3 and 4. We speculated that the bicyclic
skeleton had a stable conformation by fixing the ring and hindering access of metabolic enzymes such as cytochrome P450 to substrates.
The low metabolic stability of (−)-1 may be due to the high LogD.

Table 1. In vivo structure activity relationship (SAR) of conolidine derivatives 2–5
8-membered ring
R
N
N
H
X
O
Inhibition of acetic acid-induced
writhing
8-membered
ring (R, X)
Metabolic
stability (%)b
hERG channel inhibition
(% inhibition at 10 µM)
Compound
Salt
LogDa
2.8
(% inhibition at 30 mg/kg)
N
(−)-1
H2SO4
49
21
31
36
83
34
63
88
50
73
26
O
N
2
3
4
5
H2SO4
H2SO4
H2SO4
H2SO4
1.5
1.1
1.3
0.7
91
O
N
53
O
N
49
O
N
>100
N
O
aThe distribution coefficient (LogD) was measured between 1-octanol and phosphate-buffered saline (pH 7.4).
bThe remaining percentage (%) of the tested compounds after 0.5-h incubation with mouse liver microsome (0.5 mg/mL).
We narrowed down on a [5.2.1] bicyclic skeleton containing an amide bond in the molecule, such as compound 5, as it was a novel
and original structure. Therefore, we decided to perform further derivatization based on the structure of compound 5. The synthesis of
derivatives from the novel bicyclic compound 5 is shown in Scheme 1. Compounds 5 and 10–14 were racemic, and compounds 15a and
15b were optically active compounds of compound 5, synthesized to investigate the differential efficacy between the enantiomers.
Moreover, the optically active compounds 19a and 19b were synthesized to examine the influence of changing the size of the bicyclic
ring. Although compounds 5, 10–14, 15a, and 15b were synthesized as sulfate salts and compounds 19a and 19b were synthesized as free
forms, we did not observe any difference in pharmacokinetic (PK) profile between the sulfate salts and free forms. We synthesized
compounds 10 and 11 to investigate the effects of the R2 and R3 substituents and compounds 12–14 to examine the optimal size of the R3
substituent. The commercially available amines 6a-c10, 6f, 16a, and 16b11 were protected by the 2-nitrobenzenesulfonyl (Ns)12 or
benzyloxycarbonyl (Cbz) group. These protected amines were subjected to alkylation and deprotection to obtain the corresponding amines
7a-f and 17a, 17b with a high yield from the starting material. Compounds 11 and 14 were synthesized from commercially available 7g13
and 7h14 as the starting materials, respectively. Amide formation with amines 7a-h, 17a, and 17b, and indole-2-carboxylic acid using 1-
ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDCI) yielded compounds 8a-h, 18a, and 18b, which were subjected to
Boc group deprotection to form the corresponding amine hydrochloride salt. Subsequently, the intramolecular Mannich reaction of amine
hydrochloride salt with formaldehyde resulted in the formation of bicyclic compounds 9a-h, 19a, and 19b. Only compound 9a-h, which
had a pyrrolidine ring in the molecule, was converted to sulfate salt to obtain the corresponding desired compounds (5, 10-14, 15a, and
15b).

H
O
O
HO
O
N
15a (S): R2=H, R3=Me
15b (R): R2=H, R3=Me
5 (rac): R2=H, R3=Me
12 (rac): R2=H, R3=Et
13 (rac): R2=H, R3=n-Pr
10 (rac): R2=Me, R3=Me
11 (rac): R2,R3= -(CH2)3-
N
O
O
O
O
N
O
e, f
g
a, b, c
d
N
*
R2
R3
N
N
H
N
R2
R3
*
N
N
H
*
NH2
R2
R3
NH
O
R2
*
R, S or racemic
14 (rac): R2=H, R3=c-Pr
7a (S): R2=H, R3=Me
7b (R): R2=H, R3=Me
7c (rac): R2=H, R3=Me
7d (rac): R2=H, R3=Et
7e (rac): R2=H, R3=n-Pr
7f (rac): R2=Me, R3=Me
7g (rac): R2,R3= -(CH2)3-
9a (S): R2=H, R3=Me
9b (R): R2=H, R3=Me
9c (rac): R2=H, R3=Me
9d (rac): R2=H, R3=Et
9e (rac): R2=H, R3=n-Pr
9f (rac): R2=Me, R3=Me
9g (rac): R2,R3= -(CH2)3-
6a (S): R2=H
6b (R): R2=H
6c (rac): R2=H
6f (rac): R2=Me
8a (S): R2=H, R3=Me
8b (R): R2=H, R3=Me
8c (rac): R2=H, R3=Me
8d (rac): R2=H, R3=Et
8e (rac): R2=H, R3=n-Pr
8f (rac): R2=Me, R3=Me
8g (rac): R2,R3= -(CH2)3-
7h (rac): R2=H, R3=c-Pr
9h (rac): R2=H, R3=c-Pr
8h (rac): R2=H, R3=c-Pr
H
N
HO
O
O
O
O
O
O
O
e, f
N
h, i, j
N
d
O
19a (S)
19b (R)
N
H
N
N
*
NH
NH2
*
*
R or S
18a (S)
18b (R)
17a (S)
17b (R)
16a (S)
16b (R)
N
N
N
N
N
N
H
N
N
H
N
N
H
N
N
H
N
N
H
N
O
O
O
O
O
10
5
11
12
13
N
N
N
N
N
N
H
N
N
H
N
H
N
H
N
N
H
N
N
N
O
O
O
O
O
19b
14
15a
15b
19a
Scheme 1. Reagents and conditions: (a) 2-Nitrobenzenesulfonyl chloride, Et3N, THF, 95–100%; (b) R2 halide, K2CO3, DMF, 60 °C, 96–99%; (c) 2-
Fluorothiophenol, K2CO3, MeCN, 94–100%; (d) EDCI, THF, 88–99%; (e) HCl, 1,4-dioxane/AcOEt; (f) paraformaldehyde, TFA, 1,2-dichloroethane, 4–35% 2
steps from 8a-h, 18a, and 18b; (g) 1 M H2SO4 aq., EtOH, 0 °C, 80–99%; (h) carbobenzoxy chloride., Et3N, CH2Cl2, 70-94%; (i) MeI, NaH, THF, 92–97%; (j)
Pd/C, H2, MeOH, 98–100%
The data of the derivatives synthesized in Scheme 1 are shown in Table 2. As expected, the inhibition of the hERG channel tended to
be weaker as the LogD value was lower, albeit with some exceptions, as the basicity of the compound also affected the inhibition of the
hERG channel. The metabolic stability of most compounds also showed a good value due to the low LogD. Unfortunately, the in vivo
analgesic activity of compounds 10–14 was attenuated when compared with that of compound 5. Hence, the introduction of the alkyl
substituents to the R2 and R3 positions was not permitted (10 and 11). Moreover, from the results of 12–14, the introduction of a larger R3
substituent (Et, n-Pr, c-Pr) was found to be ineffective. These compounds generally have good metabolic stability; therefore, the
substituents do not appear to affect the physical properties of the compounds. Next, we investigated the differential efficacy between the
enantiomers (15a and 15b) of the optically active compounds of compound 5. We observed that compound 15a (S-form) exhibited
considerably strong analgesic activity (15a: 95% inhibition at 30 mg/kg) and weak inhibition of the hERG channel (14% inhibition at 10
μM). Furthermore, compound 19a (S-form), which included a piperidine ring in its structure, also showed stronger analgesic activity (19a:
77% inhibition at 30 mg/kg) than 19b (R-form). These results indicate that the target site was recognized by the enantiomer of the
compound, although the exact identity of target was unknown. Among these highly active compounds, the optically active compound
15a, which has a pyrrolidine ring, was the most active. Hence, we identified that a compact 8-membered bicyclic structure, such as a
[5.2.1] bicyclic skeleton, was preferred to ensure potent in vivo analgesic activity.

Table 2. In vivo SAR derivatives 5–19b
Inhibition of
acetic acid-
cNot tested.
hERG
channel
Metabolic
induced
writhing
Subsequently, we calculated the ED50 value of compounds (−)-
Compound.
LogDa
stability
(%)b
inhibition
(% inhibition
at 10 µM)
1, 5, 15a, 15b, and 19a, which exhibited strong analgesic activities
(Table 2) in the acetic acid writhing test. In addition, we also
measured the PK parameters for these compounds (Table 3). The
ED50 value was calculated from the dose-response curve of the
analgesic activities at 10, 30, and 100 mg/kg. We observed that
compound 15a exhibited the strongest dose-dependent analgesic
effect (ED50 = 12 mg/kg; 19%, 23%, and 95% inhibition at 3, 10,
and 30 mg/kg, respectively). The results presented in Table 2
further indicated the high potential of the analgesic efficacy of
compound 15a compared with that of other derivatives. In
addition, compound 15a exhibited a good PK profile (area under
the curve = 5.80 μM, Cmax = 2.15 μM, T1/2 = 2.27 h, Cl = 24.9
mL/min/kg, Vd = 1.72 L/kg). Between compounds 15a and 15b,
compound 15a had stronger analgesic efficacy despite its lower
exposure. This result suggests that compound 15a has an
intrinsically strong analgesic effect.
(% inhibition
at 30 mg/kg)
5
83
37
47
43
44
55
0.7
0.7
1.2
1.1
1.6
1.1
>100
93
26
36
13
14
43
19
10
11
12
13
14
82
91
Consequently, compounds (−)-1, 15a, and 15b were evaluated
using a mouse formalin test (Table 4). In this test, 3.5% formalin
(20 µL) was injected in the hind-paw pad of ddY mice (n = 8) 30
min after the compound was administered (30 mg/kg, p.o.). The
reduction in the sum of the time spent in the paw licking and biting
responses was measured for the first 10 min (the initial phase) after
formalin injection. Generally, among the representative clinically
used systemic analgesics such as pregabalin, gabapentin,
duloxetine, and celecoxib, and opioids, including morphine, only
opioids are known to be effective in inhibiting the initial phase.15-
72
92
15a
15b
19a
19b
95
49
77
65
0.7
0.7
1.0
0.9
99
14
23
17
If a compound without mu-opioid receptor (MOR) agonist
activity could inhibit the initial phase, it was considered a potential
compound that could mimic the analgesic effect of opioids without
the opioid-associated side effects. Compound 15a was observed to
have more potent analgesic efficacy (83% inhibition) than (−)-1
(43% inhibition). Furthermore, we investigated the binding
affinity and functional activity of these representative compounds
toward the MOR. By using a MOR binding test and MOR cAMP
assay, we confirmed that the analgesic efficacy of these
compounds is not mediated by mu-opioid receptor activity. This
result indicated that 15a may be a powerful ideal analgesic without
side effects. Thus, orally active 15a, named DS54360155, was
selected as a promising candidate compound for analgesics.
>100
74
38
82
N.T.c
aThe distribution coefficient (LogD) was measured between 1-octanol and
phosphate-buffered saline (pH 7.4).
bThe remaining percentage (%) of the tested compounds after 0.5-h
incubation with mouse liver microsome (0.5 mg/mL).
Table 3. ED50 values and pharmacokinetic parameters of (−)-conolidine 1, 5, 15a, 15b, and 19a in the acetic acid writhing test
Analgesic activity against
Compound
acetic acid-induced writhing
ED50 (mg/kg)a
AUC (μM)b
Cmax (μM)b
T1/2 (h)b
Cl (mL/min/kg)c
Vd (L/kg)c
(−)-1
5
32
21
12
27
23
N.T.
3.29
5.80
12.2
1.59
N.T.
1.46
2.15
2.99
0.72
N.T.
1.69
2.27
2.52
2.18
N.T.
N.T.
24.9
12.9
N.T.
N.T.
N.T.
1.72
1.14
N.T.
15a
15b
19a
aThe ED50 value was calculated from the dose-response curve of the analgesic activities at 10, 30, and 100 mg/kg.
bAverage of three values dosed at 10 mg/kg orally in ddY mice (0.5 w/v % methylcellulose suspension).
cAverage of three values dosed at 1.0 mg/kg intravenously in the ddY mice (20% HP-b-CyD, solution).
N.T., not tested; AUC, area under the curve

Table 4. Mouse formalin test initial phase and MOR assay of compounds 1, 15a, and 15b
MOR binding assay
Formalin test initial phase
(% inhibition at 30 mg/kg)
MOR cAMP assay
MOR cAMP assay
Compound
EC50 (μM)b
Emax (%)c
IC50 (μM)a
(−)-1
15a
43
83
30
70
>100
>100
>100
13
19
18
>100
>100
15b
aBinding affinities (IC50) were obtained by the competitive displacement of radiolabeled [3H] diprenorphine. Morphine with an IC50 0.41 μM was used as a
positive control.
bCyclic adenine monophosphate (cAMP) assay was performed using human mu-opioid receptor (MOR)-expressing CHO cells. The EC50 of DAMGO, the
positive control, was 0.075 μM.
cEmax was calculated as the % of the response obtained with DAMGO.
In conclusion, we identified a novel compound, (5S)-6-methyl-
1,3,4,5,6,8-hexahydro-7H-2,5-methano[1,5]diazonino[7,8-
Furthermore, the compound exhibited weak hERG channel
inhibition and no MOR agonist activity. Although the mechanism
of action of DS54360155 is not completely understood, we
consider it a promising candidate compound for application as an
analgesic. Currently, studies on further derivatization of this series
of compounds and target identification are ongoing to understand
the mechanism of action, and the results will be reported in due
course.
b]indol-7-one sulfate salt 15a, derived from the natural product
conolidine. Compound 15a (DS54360155)18 has a unique and
original [5.2.1] bicyclic structure comprising an amide bond and a
pyrrolidine ring. The compound exhibited potent analgesic
efficacy following oral administration in mice, as revealed by both
the acetic acid-induced writhing test and formalin test.
References and notes
1. Kam TS, Pang HS, Choo YM, Komiyama K. Chem
Biodivers. 2004;1:646.
2. Tarselli MA, Raehal, KM, Brasher AK, Streicher JM,
Groer CE, Cameron MD, Bohn LM, Micalizio GC. Nat
Chem. 2011;3:449.
aminopyrrolidine (>97.0 ee%) which were purchased
from Tokyo Chemical Industry Co., Ltd.
11. (S) and (R)-1-(tert-butoxycarbonyl)-3-aminopiperidine
(>97 ee%) were purchased from FUJIFILM Wako Pure
Chemical Corporation.
3. Bohn LM, Micalizio GC. WO2012088402.
4. Takanashi N, Suzuki K, Kitajima M. Takayama, H.
Tetrahedron Lett. 2016;57:375.
12. Kan T, Fukuyama T. Chem Commun. 2004;353.
13. Wu Y, Yang J, Lin Y, Qian Z, Shi Y, Ma R, Chen S.
CN102070640.
5. Naoe N, Yoshida Y, Oishi S, Fujii N. Ohno, H. J Org
Chem. 2016;81:5690.
14. Fish V, Ryckmans T, Stobie A, Wakenhut F.
WO2006064336.
6. Huang Y, Yang Y, Song H, Liu Y1, Wang Q. Sci Rep.
2015;5:13516.
7. Mendis GDC, Berecki G, Morrisroe E, Pachernegg S, Li
M, Varney M, Osborne PB, Reid CA, Halgamuge S,
Petrou S. Sci Rep. 2019;9:121.
8. Crofford LJ. Nat Rev Rheum. 2010;6:191.
9. Labianca R, Sarzi-Puttini P, Zuccaro SM, Cherubino P,
Vellucci R, Fornasari D. Clin Drug Investig. 2012;32:53.
10. (3S)-(-)-1-(tert-butoxycarbonyl)-3-aminopyrrolidine
(>98.0 ee%) and (3R)-(-)-1-(tert-butoxycarbonyl)-3-
15. Field MJ, Oles RJ, Lewis AS, McCleary S, Hughes J,
Singh L. Br J Pharmacol. 1997;121;1513.
16. Sun YH, Dong YL, Wang YT, Zhao GL, Lu GJ, et al.
PLoS ONE. 2013;8:e76603.
17. Le Bars D, Gozariu M, Cadden SW. Pharmacol Rev.
2001:53:597.
18. See supporting information for details of the
experimental procedure of 15a.
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or
Discovery of a novel bicyclic compound,
DS54360155, as an orally potent analgesic
without mu opioid receptor agonist activity
Leave this area blank for abstract info.
Tsuyoshi Arita, Masayoshi Asano, Yuki Domon, Kazufumi Kubota, Nobuo Machinaga, Kousei Shimada
N
Analgesic activity against
acetic acid-induced writhing ED50 12 mg/kg
N
N
H
Formalin test initial phase
83% inhibition at 30 mg/kg
O
DS54360155

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships that
could have appeared to influence the work reported in this paper.
☒The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests: