Discovery of conolidine derivative DS39201083 as a potent novel analgesic without mu opioid agonist activity

Article abstract of DOI:10.1016/j.bmcl.2019.05.045

We discovered a novel compound, 5-methyl-1,4,5,7-tetrahydro-2,5-ethanoazocino[4,3-b]indol-6(3H)-one sulfuric acid salt (DS39201083), which was formed by derivatization of a natural product, conolidine. DS39201083 had a unique bicyclic skeleton and was a m

Full text of DOI:10.1016/j.bmcl.2019.05.045

Accepted Manuscript
Discovery of conolidine derivative DS39201083 as a potent novel analgesic
without mu opioid agonist activity
Tsuyoshi Arita, Masayoshi Asano, Kazufumi Kubota, Yuki Domon, Nobuo
Machinaga, Kousei Shimada
PII:
S0960-894X(19)30342-7
https://doi.org/10.1016/j.bmcl.2019.05.045
BMCL 26465
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
13 March 2019
16 May 2019
21 May 2019
Please cite this article as: Arita, T., Asano, M., Kubota, K., Domon, Y., Machinaga, N., Shimada, K., Discovery of
conolidine derivative DS39201083 as a potent novel analgesic without mu opioid agonist activity, Bioorganic &
Medicinal Chemistry Letters (2019), doi: https://doi.org/10.1016/j.bmcl.2019.05.045
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com
Discovery of conolidine derivative DS39201083 as a potent novel analgesic
without mu opioid agonist activity
Tsuyoshi Arita, Masayoshi Asano, Kazufumi Kubota, Yuki Domon, Nobuo Machinaga, Kousei Shimada
R&D Division, Daiichi Sankyo Co., Ltd., 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
We discovered a novel compound, 5-methyl-1,4,5,7-tetrahydro-2,5-ethanoazocino[4,3-b]indol-
6(3H)-one sulfuric acid salt (DS39201083), which was formed by derivatization of a natural
product, conolidine. DS39201083 had a unique bicyclic skeleton and was a more potent
analgesic than conolidine, as revealed in the acetic acid-induced writhing test and formalin test
in ddY mice. The compound showed no agonist activity at the mu opioid receptor.
Keywords:
Conolidine
2019 Elsevier Ltd. All rights reserved.
Indole alkaloid
Analgesic
Acetic acid writhing test
Non-opioid
———
 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 in 2004 by Kam et al.1 from Tabernaemontana divaricata,
along with a number of other monoterpenoid indole alkaloids having opioid analgesic properties. However, research on the biological
activity of 1 has not been promoted because it is an extremely rare compound that has been isolated in only 0.00014% yield from the
stem bark of the plant. In 2011, Micalizio and coworkers reported the first successful asymmetric total synthesis of conolidine, which
involved the formation of a C-ring via the Mannich reaction of a piperidine-indole compound and formaldehyde. They also revealed
that the synthetic conolidine showed powerful non-opioid analgesic effects in mouse models.2,3 Following this work, several other
groups have reported the efficient synthesis of this compound.4,5,6,7 In addition, the research to elucidate the mechanism of action of 1
has also been reported recently.8 Although the mechanism of action of 1 is not yet completely clear, we are interested in its unique
analgesic properties that may render it a promising alternative to opioids which impart severe opioid-related adverse effects9,10 such as
constipation, nausea, vomiting, pruritus, somnolence, cognitive impairment, respiratory depression, tolerance, physical dependence, and
addiction. In this letter, we describe the synthesis and in vivo structure activity relationship (SAR) study of a series of novel conolidine
derivatives.
Conolidine (1)
Figure 1. Structure of conolidine.
Our objective was to identify orally bioavailable conolidine derivatives that show more potent analgesic efficacy in the in vivo
animal pain model. Therefore, we designed and synthesized several varieties of conolidine derivatives (2a–4d) that were synthesized
from (±)-1 or aldehyde 5 that can be synthesized by the method already reported using the process illustrated in Scheme 1. All the
derivatives were sulfate salts, and the compounds, except for 4a,11 were either racemic (2a–3d, 4c–d) or diastereoisomeric 4b.
We initiated the SAR study by evaluating the synthesized derivatives in the acetic acid writhing test, which measures the reduction
(% inhibition) in the acetic acid induced-writhing behavior upon oral administration of the compounds (30 mg/kg) in Slc:ddY mice for
efficient derivatization. The effect of the substituted conolidine derivatives on the various parameters of the SAR study are summarized
in Table 1.
Introduction of the alkyl group at the R1 position (2a–b) reduced the in vivo activities of the compounds as compared with that of
(−)-1. Moreover, the substituent of the aromatic moiety in the core structure was investigated. For instance, introduction of a methoxy
group on the phenyl ring in compounds (3a–d) resulted in weaker activities as compared with that of (−)-1. We speculated that the
analgesic activities are associated with poor or moderate metabolic stabilities against mice cytochromes. Next, we synthesized several
derivatives (4a–d) focusing on the exo-olefin of conolidine. Among them, the unsubstituted compound 4a showed drastically improved
metabolic stability (97%) and more potent in vivo activity (73% inhibition at 30 mg/kg).
c, b
a, b
d, e, b
2a
4b
2b
1
1
i, j
g, h
b
f
3a-d
4c-d
8
7
6
5
Scheme 1. Reagents and conditions: (a) Pd-C, EtOH, 97%; (b) H2SO4 aq., EtOH, 68-98%; (c) MeI, NaH, DMF, 56%; (d) BrCH2CH2OAc, NaH, DMF, 58%;
(e) NaOH, THF/MeOH, 85%; (f) n-BuLi, benzenesulfonyl group -protected indole, THF, 72-94%; (g) Cs2CO3/MeOH, THF, 82-92%; (h) MnO2, CH2Cl2, 68-
89%; (i) ACECl, 1,2-dichloroethane, then MeOH; (j) paraformaldehyde, TFA, 1,2-dichloroethane, 25-38% from 7.
Table 1. In vivo SAR of substituted conolidine derivatives 2a–4d.
5
4
6
7
(-)-1 Conolidine

Acetic acid writhing test
Metabolic
Compound
R1
H
R2
R3
Log D a
N.T.c
(% inhibition at 30 mg/kg )
stability (%)b
(±)-1
H
49
49
30
27
42
3
N.T.c
78
(−)-1
2a
H
H
H
2.8
3.5
2.7
3.4
3.1
3.1
3.3
1.8
N.T.c
3.3
Me
40
2b
CH2CH2OH
H
38
3a
H
H
H
H
H
H
H
4-OMe
5-OMe
6-OMe
7-OMe
H
22
3b
45
3c
31
34
73
50
18
54
3d
N.T.c
97
4a
H
4b
H
N.T.c
56
4c d
H
4d
H
H
0
N.T.c
N.T.c
a The distribution coefficients (log D) were measured between 1-octanol and phosphate buffered saline (pH 7.4).
b Percentage of the tested compound remaining after 0.5 h of incubation with mouse liver microsome (0.1 mg/mL).
c Not tested, d Ervaticine5,12
These results encouraged us to synthesize further derivatives on the basis of 4a. Therefore we decided to investigate the influence of
an alkyl group at R4 position in detail. The synthetic route of compounds (17a-c) is shown in Scheme 2.13 Initially, the readily available
9 was reduced by LiAlH4 to produce alcohol 10, which was subsequently converted to the corresponding aldehyde 11 by Parikh–
Doering oxidation using SO3/pyridine.14 Addition of the benzenesulfonyl group protected-indole to 11 furnished the corresponding
alcohol 12 in a moderate yield. The reaction showed excellent selectivity, and almost no 3-position adduct was produced. Removal of
the benzenesulfonyl group under mild conditions using Cs2CO3/MeOH gave the desired product 13. Continuous oxidation of the
secondary alcohol with MnO2 and subsequent removal of the Boc group produced piperidine hydrochloride 15. Intramolecular Mannich
reaction of 15 with formaldehyde furnished the desired bicyclic compound 16. Finally, conversion of the bicyclic compound 16 to its
salt provided the desired compound (17a-c).
c
a
b
d
12
13
11
9
10
HCl
g
f
h
e
17a-c
16
15
14

Scheme 2. Reagents and conditions: (a) LiAlH4, THF, 72-90%; (b) SO3/pyridine, DMSO, (i-Pr)2EtN, CH2Cl2; (c) n-BuLi, benzenesulfonyl group -protected
indole, THF, 60-89% from 11; (d) Cs2CO3/MeOH, THF, 69-85%; (e) MnO2, CH2Cl2; (f) HCl, EtOAc; (g) paraformaldehyde, TFA, 1,2-dichloroethane, 29-40%
from 13; (h) H2SO4 aq., EtOH, 68%.
The results of derivatives (17a–c) are summarized in Table 2. We found that 17a bearing a methyl group showed the strongest
analgesic activity (90% inhibition at 30 mg/kg). In contrast, the introduction of larger substituents such as ethyl 17b and sec-butyl 17c
attenuated the in vivo activity presumably because of the low metabolic stability imparted by the increased lipophilicity. The
pharmacokinetic (PK) parameters of the compounds were also measured in mice. The most active compound, 17a, showed the best PK
profile (AUC = 3.21 μM, Cmax = 1.25 μM, T1/2 = 2.33 h). Interestingly, these data suggest that the PK profiles and the analgesic activity
tend to be almost positively correlated.
Furthermore, we evaluated some of the selected representative compounds ((−)-1, 4a, 17a) in detail by the acetic acid-induced
writhing test, in order to calculate the ED50 value (Table 3). As expected, 17a showed excellent dose-dependent activity (ED50 = 7.8
mg/kg; 23%, 52%, and 94% inhibition at 3, 10, and 30 mg/kg, respectively) four times higher than that of (−)-1 (ED50 = 32 mg/kg).
Consequently, (−)-1 and 17a were evaluated in the mice formalin test. In this test, 3.5% formalin (20 µL) was injected into the hind paw
pad of Slc:ddY mice (n = 8) 30 min after the compound was administered (30 mg/kg, p.o.). Reduction in the sum of the time spent in
paw licking and the biting responses were measured for the first 10 min (the initial phase) following the formalin injection. In general,
among the representative clinically used systemic analgesics such as pregabalin, gabapentin, duloxetine, and celecoxib, and opioids
such as morphine, only opioids are known to be effective in inhibiting the initial phase.15,16,17 If a compound without mu opioid agonist
activity can inhibit the initial phase, it is a potential analgesic that will also mimic the analgesic effect of opioids without mimicking
their side effects.
Compound 17a was found to have more potent analgesic efficacy (92% inhibition) as compared with (−)-1 (43% inhibition).
Furthermore, we investigated the binding affinity and functional activity of these representative compounds toward the mu opioid
receptor (MOR). We confirmed that the analgesic efficacies of these compounds are not mediated by the mu opioid activity of the MOR
binding test and MOR cAMP assay. Thus, the orally active 17a, DS39201083, was selected as a promising tool compound.
Table 2. In vivo SAR at R4 and pharmacokinetic (PK) parameters of conolidine derivatives 4a and 17a–c.
Acetic acid writhing test
(% inhibition at 30 mg/kg)
Metabolic
stability (%) b
Compound
R4
AUC (μM) a
Cmax (μM)a
T1/2 (h)a
Log D
4a
H
73
90
48
51
2.26
3.21
1.02
0.14
1.16
1.25
0.71
0.23
1.17
2.33
1.28
1.60
1.8
2.5
3.0
3.7
89
92
43
0
17a
17b
17c
Me
Et
sec-Bu
a Average of three values dosed at 10 mg/kg orally (p.o.) in ddY mice (0.5 w/v% methylcellulose suspension).
b Percentage of the tested compound remaining after 0.5 h of incubation with mouse liver microsome (0.5 mg/mL).
Table 3. In vivo analgesic evaluation and MOR assay of (−)-1, 4a and 17a.
Acetic acid writhing test
ED50 (mg/kg) a
Formalin test, initial phase1
(% inhibition at 30 mg/kg)
MOR binding test
MOR cAMP assay
MOR cAMP assay
Compound
IC50 (μM) b
EC50 (μM)c
Emax (%)d
32
11
43
N.T. e
92
70
>100
71
>100
>100
>100
13
13
12
(−)-1
4a
17a
7.8
a The ED50 value was calculated from the dose response curve of analgesic activity at 10, 30 and 100mpk.
b Binding 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.
c cAMP assays were carried out using human MOR-expressed CHO cells. DAMGO, with an EC50 0.088μM, was used as a positive control.
d Emax was calculated as the response (in %) obtained with DAMGO.
e Not tested
In conclusion, we identified a novel compound, 5-methyl-1,4,5,7-tetrahydro-2,5-ethanoazocino[4,3-b]indol-6(3H)-one sulfuric
acid salt 17a 18, derived from a natural product conolidine. Compound, 17a (DS39201083), has a unique and non-chiral bicyclic

structure and shows more potent analgesic efficacy upon its oral administration in mice, as revealed by the acetic acid-induced
writhing test and formalin test; furthermore, it shows no agonist activity for the MOR. Although the mechanism of action is still
unknown, we consider DS39201083 to be a promising tool compound. Studies on further derivatization of this series of compounds
and target identification to understand the mechanism of action are ongoing, and the results will be reported in due course.
Reference and notes
1. Kam, T. S.; Pang, H. S.; Choo, Y. M.; Komiyama, K. Chem. Biodivers. 2004, 1, 646.
2. Tarselli, M. A.; Raehal, K. M.; Brasher, A. K.; Streicher, J. M.; Groer, C. E.; Cameron, M. D.; Bohn, L. M.; Micalizio, G.
C. Nat. Chem. 2011, 3, 449.
3. Bohn, L. M.; Micalizio, G. C. WO2012/088402 (patent).
4. Takanashi, N.; Suzuki, K.; Kitajima, M.; Takayama, H. Tetrahedron Lett. 2016, 57, 375.
5. Naoe, N.; Yoshida, Y.; Oishi, S.; Fujii, N.; Ohno, H. J. Org. Chem. 2016, 81, 5690.
6. Huang, Y.; Yang, Y.; Song, H.; Liu, Y1.; Wang, Q. Sci. Rep. 2015, 5, 13516.
7. Sugimoto, K.; Matsuya, Y. Tetrahedron Lett. 2017, 58, 4420.
8. Mendis G. D. C.; Berecki G.; Morrisroe E.; Pachernegg S.; Li M.; Varney M.; Osborne P. B.; Reid C.A.; Halgamuge S.;
Petrou S. Sci. Rep. 2019, 9, 121.
9. Crofford, L. J. Nat. Rev. Rheum. 2010, 6, 191.
10. Labianca, R.; Sarzi-Puttini, P.; Zuccaro, S. M.; Cherubino, P.; Vellucci, R.; Fornasari, D. Clin. Drug. Investig. 2012, 32,
53.
11. Scopes, D. I. C.; Allen, M. S.; Hignett, G. J.; Wilson, N. D. V.; Harris, M.; Joule, J. A. J. Chem. Soc. Perkin Trans. 1.
1977, 21, 2376.
12. Rahman, A.; Muzaffar, A. Heterocycles, 1985, 23, 2975.
13. See the supporting information for details of the experimental procedure of 17a.
14. Parikh, J.P.; Doering, W. E. J. Am. Chem. Soc. 1967, 89, 5505.
15. Field, M. J.; Oles, R. J.; Lewis, A.S.; McCleary, S.; Hughes, J.; Singh, L. Br. J. Pharmacol. 1997, 121, 1513.
16. Sun Y. H.; Dong Y. L.; Wang Y. T.; Zhao G. L.; Lu G. J.; Yang J.; Wu S. X.; Gu Z. X; Wang W. PLoS ONE. 2013, 8(10).
17. Le Bars D.; Gozariu M.; Cadden S. W. Pharmacol. Rev. 2001, 53, 597.
18. The data for 17a. 1H-NMR (400 MHz, DMSO-d6) δ: 11.58 (1H, br s), 7.65 (1H, d, J = 8.2 Hz), 7.48 (1H, d, J = 8.2 Hz),
7.34 (1H, t, J = 7.6 Hz), 7.12 (1H, t, J = 7.4 Hz), 4.72 (2H, s), 3.42–3.37 (2H, m), 3.22–3.19 (2H, m), 2.11–2.04 (2H, m),
1.88–1.86 (2H, m), 1.28 (3H, s). MS (ESI) m/z: 255 (M+H)+. Anal. Calcd for C16H18N2O. H2SO4. 1/3C2H5OH: C, 54.43;
H, 6.03; N, 7.62; S, 8.72. Found: C, 54.16; H, 5.80; N, 7.89; S, 8.73.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Discovery of conolidine derivative
DS39201083 as a potent novel analgesic
without mu opioid agonist activity
Leave this area blank for abstract info.
Tsuyoshi Arita, Masayoshi Asano, Yuki Domon, Kazufumi Kubota, Nobuo Machinaga, Kousei Shimada
(-)-Conolidine
DS39201083