The Pharmacological Heterogeneity of Nepenthone Analogs in Conferring Highly Selective and Potent κ-Opioid Agonistic Activities

Article abstract of DOI:10.1021/acschemneuro.6b00321

To develop novel analgesics with no side effects or less side effects than traditional opioids is highly demanded to treat opioid receptor mediated pain and addiction issues. Recently, κ-opioid receptor (KOR) has been established as an attractive target,

Full text of DOI:10.1021/acschemneuro.6b00321

Subscriber access provided by Olson Library | Northern Michigan University
Article
The Pharmacological Heterogeneity of Nepenthone Analogs in
Conferring Highly Selective and Potent #-Opioid Agonistic Activities
Wei Li, Jian-Dong Long, Yuan-Yuan Qian, Yu Long, Xue-Jun Xu, Yu-jun Wang, Qing
Shen, Zuo-Neng Wang, Xi-Cheng Yang, Li Xiao, Hong-Peng Sun, Yulong Xu, Yiyi
Chen, Qiong Xie, Yonghui Wang, Liming Shao, Jing-gen Liu, Zhuibai Qiu, and Wei Fu
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.6b00321 • Publication Date (Web): 29 Dec 2016
Downloaded from http://pubs.acs.org on December 30, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Chemical Neuroscience is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 37
1
2
3
4
5
6
7
8
9
The Pharmacological Heterogeneity of
Nepenthone Analogs in Conferring Highly
Selective and Potent κꢀOpioid Agonistic
Activities
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Wei Li1#, Jian-Dong Long2#, Yuan-Yuan Qian1#, Yu Long3, Xue-Jun Xu2, Yu-Jun
Wang2, Qing Shen1, Zuo-Neng Wang1, , Xi-Cheng Yang1, Li Xiao1, Hong-Peng Sun1,
Yu-Long Xu1, Yi-Yi Chen1, Qiong Xie1, Yong-Hui Wang1, Li-Ming Shao1, Jing-Gen
Liu2*, Zhui-Bai Qiu1, Wei Fu1*
1
Department of Medicinal Chemistry, School of Pharmacy, Fudan University,
Shanghai, 201203, China
2
Key Laboratory of Receptor Research, Shanghai Institute of Materia Medica,
Chinese Academy of Sciences and Collaborative Innovation Center for Brain Science,
Shanghai, 201203, China
3 Department of Pharmacy, Dalian Medical University, Dalian, 116044, China
ACS Paragon Plus Environment

Page 3 of 37
ACS Chemical Neuroscience
1
2
3
4
ABSTRACT
5
6
7
8
9
To develop novel analgesics with no or less side effects than that of traditional
opioids is highly demanded to treat opioid receptors mediated pain and addiction
issues. Recently, κꢀopioid receptor (KOR) is established as an attractive target,
although its selective agonists could bear heterogeneous pharmacological activities. In
this study, we designed and synthesized two new series of nepenthone derivatives by
inserting a spacer (carbonyl) between 6α,14αꢀendoꢀethenylꢀthebaine and the
7αꢀphenyl substitution of the skeleton, and by substituting the 17ꢀNꢀmethyl group
with a cyclopropylmethyl group. We performed in vitro tests (binding and functional
assays) and molecular docking operations on our newly designed compounds. The
results of wetꢀexperimental measures and modeled binding structures demonstrate
that these new compounds are selective KOR agonists with nM level affinities.
Compound 4 from these new derivatives, showed highest affinity (Ki= 0.4 ± 0.1) and
the highest selectivity (ꢀ/κ = 339, δ/κ = 2034) toward KOR. The in vivo tests turned
out that compound 4 is able to induce stronger (ED50 = 2.1 mg/kg) and much longer
antinociceptive effect than that of the typical KOR agonist U50488H (ED50 = 4.4
mg/kg). Therefore, compound 4 can be used as a perfect lead compound for future
design of potent analgesics acting through KOR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS
κꢀopioid receptor, ꢀꢀopioid
receptor, nepenthones, 4,5ꢀepoxymorphinans,
pharmacological heterogeneity, structureꢀactivity relationship, binding structure
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 37
1
2
3
4
Introduction
5
6
7
8
9
Opioid receptors are composed of three major subtypes (ꢀ, δ, and κ), they modulate
various physiological and pathophysiological activities in the central nervous system
(CNS), including pain, psychotomimetic effects and neuroprotection1, 2. When bound
with opioids and opioid analogs, these receptors demonstrated heterogeneous
pharmacological profiles1. Typical narcotics as morphine, fentanyl and mepridine,
which have been used as analgesics, are the agonists of ꢀꢀopioid receptor (MOR),2
while some other opioids like butorphanol3 and nalbuphine4 (Chart 1) are also
agonists of κꢀopioid receptor (KOR) with lower affinities than that with the MOR. In
order to avoid the opioidꢀlike adverse effects (e.g. addiction, respiratory depression
and constipation) mainly from the binding of MOR with opioids1, 2, recent research
efforts are focused on the design of potent and selective KOR agonists5, 6. Early
efforts led to the discovery of KORꢀselective agonists as spiradoline7 and enadoline8,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
(Chart 1). However, these two KORꢀselective agonists have failed to survive
clinical trials of analgesia due to their negative benefitꢀrisk ratios. A successful
example is discovery of the KORꢀselective agonist nalfurafine (Chart 1).10, 11
Chemically, this compound is a 4,5ꢀepoxymorphinan derivative. It has been approved
and marketed in Japan, as it is able to produce a longꢀterm suppression of pruritus, but
can only be used for hemodialysis patients showing resistance to conventional
treatments12. On the other hand, a lot of studies are continuously performed to
develop dynorphin A analogs as potential treatment of chronic neuropathic pain,6 as
the KORꢀdynorphin system has been found to have counterꢀmodulatory effects on the
ACS Paragon Plus Environment

Page 5 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
reward from narcotics exposure, and exhibits plasticity in addictiveꢀlike states.13
These discoveries5, 6 indicate that KORꢀselective agonists could bear heterogeneous
pharmacological activities. So far, it is still a hard work to design and develop novel
KOR agonists as potent therapeutics, but with no or much less adverse effects than
that of MOR agonists.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 1. Chemical structures of representative kappa opioid agonists
In our previous work14-16, we spent much efforts to design and chemically
synthesize novel derivatives of thebaine based on its morphineꢀlike structure, and
obtained MOR and KOR agonists.15, 16 Inspired by the structures and transformed
activities of GNTI and 6'ꢀGNTI in the literature,17,18 we introduced a 7αꢀaminophenyl
substitution to 6α,14αꢀendoꢀethenylꢀ7αꢀphenylthebaine (compound 1 in Scheme 1)
analogs, to see whether this substitution could help to increase KOR activity and
selectivity. Our initial pharmacological results suggested that, compound 2 (Scheme
1) with 7αꢀpꢀaminophenyl substitution has ex vivo KOR agonistic activities by the test
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 37
1
2
3
4
5
6
7
8
9
as rabbit vas deferens (RVD) preparations, and such ex vivo activity can be reversed
by the KORꢀselective antagonist norꢀBNI16. These initial results encouraged us to
finely tune the structure of compound 2, in order to design better KORꢀselective
agonists. For this purpose, we inserted a carbonyl group as a spacer extension
between the substructures of 6α,14αꢀendoꢀethenylꢀthebaine and the 7αꢀphenyl
substitution. Such spacer should help us to test the role of pꢀamino substitution in
conferring KOR agonistic activity of compound 2. We also wanted to test the effect
on KOR agonistic activity by substituting the 17ꢀNꢀmethyl group with a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
cyclopropylmethyl group, as the cyclopropymethyl group appeared in some opioid
20
analogs19,
.
Following these strategies, we chemically synthesized,
pharmacologically tested (in vitro and in vivo) the two series of compounds (3ꢀ4
series, Scheme 1). Furthermore, we performed molecular modeling of KOR structure,
and molecular docking our newly designed KORꢀselective agonists into the
agonistꢀbinding site of KOR, in order to elucidate the structural determinates for their
selective affinities toward KOR.
ACS Paragon Plus Environment

Page 7 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
Scheme 1 Initial design rationale for KOR agonists from nepenthone and
NꢀnorꢀNꢀcyclopropylmethylꢀnepenthone derivatives
7
8
9
RESULTS AND DISCUSSION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemical Designs.
4,5ꢀEpoxymorphinan has been recognized to be a privileged structural component
for ligands of opioid receptors 19. However, the DielsꢀAlder adducts (e.g., nepenthone
and thevinone) between thebaine and alkyl or aryl vinyl ketones would be much of
interest if the adducts were converted into orvinols first through Grignard reaction and
then the optimization strategies like Nꢀcyclopropylmethyl substitution,
3ꢀOꢀdemethylation or saturation of the 17,18ꢀethenyl group21-23. These synthetic
approaches led to the discovery of some important semiꢀsynthetic opioids, including
etorphine and buprenorphine19. Structurally, nepenthone and its analogs were not fully
explored and optimized, possibly due to their low pharmacological activities toward
opioid receptors. As outlined in Scheme 1, we started from synthesizing nepenthones
by DielsꢀAlder reactions, using thebaine and appropriate phenyl vinyl ketones as the
reactant materials. The reported approach24 for preparing phenyl vinyl ketone did not
work for nitrophenyl vinyl ketones, so we designed an alternative synthetic route
(Scheme 2). In our synthetic route, nitrobenzaldehydes reacted first with vinyl
magnesium bromide to afford the Grignard products, and then the products were
oxidized by Jones reagent to yield the expected nitrophenyl vinyl ketones. The
ketones underwent DielsꢀAlder reactions with thebaine to afford 5aꢀc. These
intermediates were further chemically reduced to generate compounds 3aꢀc.
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 8 of 37
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (a) vinyl magnesium bromide, THF, ꢀ65 °C; (b)
CrO3/H2SO4, acetone, <10 °C; (c) thebaine, toluene, reflux; (d) Raney Ni/85%
Hydrazine hydrate, ethanol, 60 °C.
Scheme 2 Synthesis route of compounds 3aꢀc
In order to synthesize NꢀnorꢀNꢀcylcopropylmethylꢀnepenthones, we designed
another route as shown in Scheme 3, starting from compound 5 and 5aꢀc. These
intermediates 5, 5aꢀc were demethylated with diethyl azodicarboxylate (DEAD) and
hydrolyzed to yield 6 and 6aꢀc, which were further alkylated with propyl methyl
bromide to afford 7aꢀc, and subsequently reduced to generate the expected
compounds 4aꢀc.
Reagents and conditions: (a) diethyl azodicarboxylate, benzene, reflux; (b) 1N HCl,
70ꢀ80 °C; (c) propylmethyl bromide, potassium carbonate, DMF, 100 °C ; (d). Raney
ACS Paragon Plus Environment

Page 9 of 37
ACS Chemical Neuroscience
1
2
3
4
Ni/85% Hydrazine hydrate, ethanol, 60 °C.
5
6
Scheme 3 Synthesis route of compounds 4aꢀc
7
8
9
In Vitro Activities.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We expected the newly synthesized compounds (3ꢀ4 series, Scheme 1) to be KOR
agonists. This was based on the reports23 that nepenthones are active ligands of opioid
receptors. The only difference between the nepenthones and the
6α,14αꢀendoꢀethenylꢀ7αꢀphenylthebaines (Scheme 1) is the inserted carbonyl group
in the structural scaffold of nepenthones. Would these newly designed compounds
share similar pharmacological profiles as that of nepenthones?
To test the agonistic activities of these new compounds toward KOR, we performed
both the competitive binding assays and functional assays. The competitive binding
for each of the new compounds was measured in a similar way as that was used in our
previous studies15, that was, each compound was tested to competitively inhibit the
binding of [3H]DAMGO for MOR, [3H]DPDPE for δꢀopioid receptor (DOR) and
[3H]U69,593 for KOR 25. The functional assays for these new compounds were
performed through classic [35S]GTPγS binding. The results of these assays are
summarized and listed in Table 1 for the binding affinities, and Table 2 for their
abilities to stimulate [35S]GTPγS binding. As shown in Table 1, nepenthone, 3 has
higher affinity with MOR than that with KOR. Adding the aminoꢀsubstituent to the
7αꢀphenylcarbonyl group of compound 3 does not obviously change the receptor
affinity for the derivatives 3aꢀc, no matter the position of the aminoꢀsubstituent (oꢀ for
3a, mꢀ for 3b and pꢀ for 3c). However, substituting the 17ꢀNꢀmethyl group with a
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 10 of 37
1
2
3
4
5
6
7
8
9
cyclopropylmethyl group dramatically increased KOR affinity for the derived
compound 4 (Ki = 0.4 ± 0.1 nM for KOR), and also increased KOR selectivity (ꢀ/κ =
339, and δ/κ of 2034). Such increase in both affinity and selectivity for KOR were
counteracted by adding the aminoꢀsubstitutent to the 7αꢀphenylcarbonyl group, as
seen the effect on compounds 4aꢀc (Table 1). The effect of increasing KOR affinity
for the 17ꢀNꢀcyclopropylmethyl substitution was repeated when comparing the
difference in KOR affinity between the intermediate compounds 7aꢀc and compounds
5aꢀc (Table 1). The extent of affinity increase from 17ꢀNꢀcyclopropylmethyl group
can be roughly estimated to be the affinity ratio of 100/1 for compound 4 over
compound 6. Based on these KOR affinity data, the 17ꢀNꢀcyclopropylmethyl group
can be viewed as the KOR affinity enhancer for nepenthone derivatives.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1 Binding affinities of nepenthone and NꢀnorꢀNꢀcyclopropylmethylꢀnepenthone
analogs for ꢁ, δ and κ opioid receptors
Compd
R’
R
H
Ki±SEM (nM)
Selectivity
ratio
ꢁa
δb
κc
170.0±10.0 0.13 NA
80.0± 10.0 0.12 NA
370.0±50.0 0.11 NA
ꢁ/κ
δ/κ
CH3
CH3
CH3
21.9±5.1
ND
ND
ND
3
3a
3b
oꢀNH2 9.6±1.6
mꢀNH2 38.9±1.7
ACS Paragon Plus Environment

Page 11 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
CH3
pꢀNH2 30.1±0.8
ND
410.0±60.0 0.07 NA
339 2034
3c
CPMd
H
129.5±3.1 773.7±29.8 0.4±0.1
4
CPMd oꢀNH2 59.3±1.4
CPMd mꢀNH2 13.6±0.6
CPMd pꢀNH2 40.4±0.6
253.1±13.1 2.0±0.1
329.7±20.7 2.3±0.13
30.1 128
5.99 145
2.76 >342
7
8
9
4a
4b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4c
>5000
14.6±1.8
>5000
CH3
CH3
CH3
H
oꢀNO2 195.6±30.2 >5000
NA
NA
5a
mꢀNO2 13.5±0.7
pꢀNO2 21.1±1.1
617.9±6.7
918.4±62.5 0.02 0.67
5b
1547.0±65.5 1264.0±61.5 0.02 1.22
5c
H
55.8±0.8
962.9±11.4 39.8±0.9
1.40 24.2
1.00 1.43
4.62 33.7
8.77 83.6
0.089 1.5
6
7a
CPMd oꢀNO2 350.7±24.9 500.8±22.6 349.7±1.6
CPMd mꢀNO2 30.0±1.6
CPMd pꢀNO2 195.6±30.2 1865.0±31.0 22.3±0.6
219.3±18.5 6.5±1.3
7b
7c
morphine
6.9±1.3
116.8±15.4 76.9 ±16.3
aDisplacement of [3H]DAMGO from CHO cell membrane expressing MOR;
bDisplacement of [3H]DPDPE from CHO cell membrane expressing DOR;
cDisplacement of [3H]U69,593 from CHO cell membrane expressing KOR. d CPM
denotes cyclopropylmethyl; Values are expressed as the Mean ± S.E.M. for three
independent experiments performed in triplicate. ND: Not determined. NA: Not
applicable.
As shown in Table 2 for the results of functional assays, compound 3 showed poor
agonistic activity toward KOR, although it has better MOR agonistic activity.
Surprisingly, the addition of the aminoꢀsubstituent to the 7αꢀphenylcarbonyl group at
compound 3a brought in great agonistic activity toward KOR, i.e. ~ 600ꢀfold (EC50 =
1368 ± 248.9 nM of compound 3 vs. 2.3 ± 0.2 nM of compound 3a) more active. The
mꢀNH2 substitution (3b) and pꢀNH2 substitution (3c) showed similar increase of the
agonistic activity toward KOR. Could this positive functional effect from the
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 12 of 37
1
2
3
4
5
6
7
8
9
additional NH2ꢀsubstituent at the 7αꢀphenylcarbonyl group be repeated? The answer
is yes, as seen from the increased agonistic activity toward KOR for compounds 4aꢀc
(Table 2). Thus, the aminoꢀsubstituent at the 7αꢀphenylcarbonyl group of nepenthone
derivatives is a real KOR functionalizer. We also wondered how strong of the
functional effect from 17ꢀNꢀcyclopropylmethyl group of nepenthone derivatives
toward KOR. As shown in Table 2, this substitution can only enhance the MOR
agonistic activity, but not that toward KOR for compounds 4aꢀc. Very interestingly,
compound 4 showed highest binding affinity with the KOR, and also the highest
agonistic activity among all the new compounds.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2 Functional activity in stimulating [35S]GTPγS binding of nepenthone and
NꢀnorꢀNꢀcyclopropylmethylꢀnepenthone analogs for ꢁ, δ and κ opioid receptors
Compd
ꢁ
δ
κ
EC50 (nM)
NAa
Emax
EC50 (nM)
NAa
Emax
NAa
EC50 (nM)
Emax
1368.0±248.9
122.5±1.6
stimulation at
10 ꢁM
180.2±1.1
3
NAa
136.5±1.1
stimulation at
10 ꢁM
NAa
NAa
2.3±0.2
204.5±2.7
3a
NAa
NAa
NAa
NAa
NAa
30.2 ± 2.3
62.2 ± 7.4
205.6±10.7
180.5 ± 9.0
211.9±1.3
190.9±0.2
197.7±0.9
3b
3c
4
NAa
NAa
NAa
130.2±6.6
18.7±2.0
75.1±7.3
202.1±2.7
206.2±3.2
208.8±0.3
678.2±20.3
186.0±15.5
442.3±59.0
224.6±10.8 1.0±0.2
273.1±4.1
263.4±8.4
9.1±1.4
7.6±0.2
4a
4b
ACS Paragon Plus Environment

Page 13 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
29.7±5.9
176.9±0.5
1567.0±35.1 231.9±11.2 28.7±2.1
217.2±2.5
NAa
4c
morphine 123.5±30.5
U50488H NAa
209±15.9
122.1±1.1
NAa
NAa
NAa
NAa
112.6±2.0
1.8±0.6
262.0±0.6
Potency and efficacy in stimulating [35S]GTPγS binding to membranes of CHO cells
stably expressing opioid receptors. Data are expressed as the Mean ± S.E.M. of
independent experiments performed in triplicate. a NA: Not applicable.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Binding Structures
Based on the structureꢀactivity relationship (SAR) for our newly designed
nepenthone derivatives as described above, we wanted to explore how these
compounds bind with KOR, and what are the structural determinants for the measured
affinity differences. For this purpose, we performed molecular docking operations to
dock these new compounds into the agonistꢀbinding site of KOR. The active KOR
structure was homology modeled based on the Xꢀray structure of the active murine
MOR26. As shown in Figure 1, the docked structure of KOR binding with compound
4 indicates that compound 4 takes a similar binding mode as that for other opioid
ligands, which have been revealed in the published Xꢀray structures of opioid
receptors.27-30 In particular, compound 4 interacts with KOR residues Asp1383.32
,
Ile2906.51, Met1423.36, Tyr3207.43, Trp2876.48 and Tyr1393.33 (Figure 1a). In addition,
it forms the water bridge with His2916.52 through two water molecules, and the
oxygen atom of the 7αꢀphenylcarbonyl group of compound 4 could provide a
hydrogen bond acceptor to residue Gln1152.60 of KOR. This additional hydrogen
bonding interaction makes the 7αꢀphenylcarbonyl group orientated very well inside
the hydrophobic subpocket composed of residues Trp124ECL2, Leu1353.29, Val1182.63
,
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 14 of 37
1
2
3
4
5
6
7
8
9
Val1343.28 and Gln1152.60 of KOR (Figure 2(a)). Compounds 4aꢀc take similar
binding mode at the agonistꢀbinding site of KOR. The aminoꢀsubstituent at the
7αꢀphenylcarbonyl group of compounds 4aꢀc is elongated into the hydrophobic
subpocket between TM2 and TM3 as well. Due to the local steric hindrance, the
aminoꢀsubstituent is obviously not favorable to this subpocket, resulting in the
reduced KOR affinities for these compounds, especially for the KOR affinity of
compound 4c. On the contrary, compounds 4aꢀc can fit better to the agonistꢀbinding
site of MOR than that of compound 4 with MOR, as the aminoꢀsubstituent at the
7αꢀphenylcarbonyl group could form additional hydrogen bonds with the
nonꢀconserved residue Asn1272.63 around the top of this hydrophobic subpocket
(Figure 2(b)). So, the compounds 4aꢀc have higher affinities with MOR than that of
compound 4 (as listed in Table 1), this leads to higher KOR selectivity of compound
4 than that of compounds 4aꢀ4c. As measured from affinity assays, the KOR affinity
of compound 4 is ~400 fold stronger than that of compound 3 (Table 1). The
superimposed structures of 4 and 3 at the active site of KOR (Figure 1b) show that
compound 4 binds more tightly, oriented deeper to the binding pocket, and its
Nꢀcyclopropylmethyl group fits better inside the hydrophobic subpocket formed by
residues Tyr3207.43, Ile3167.39, W2876.48 and M1423.36 of KOR. Based on the binding
structure, compound 4 forms stronger electrostatic interaction and hydrogen bonding
interaction with the negative charged side chain of residue Asp1383.32, with a
hydrogen bond distance of 3.0 Å, while the distance for similar interation between
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 15 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
compound 3 and residue Asp1383.32 is much longer (4.0 Å). These differencs in the
interꢀmolecular interactions significantly make compound 4 a better KOR agonist.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
b
Figure 1 a) The binding mode for 17-ꢀnorꢀNꢀcyclopropylmethylꢀnepenthone
(compound 4, colored in orange) at the agonistꢀbinding site of KOR; b). The
superimposition of the binding structure of compound 3 (colored in grey; and its
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 16 of 37
1
2
3
4
interꢀmolecular interactions are colored yellow) with that of compound 4 (colored in
5
6
7
orange; and the interꢀmolecular interactions are colored blue).
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(a)
(b)
Figure 2 (a) Superimposition for binding structures of compounds 4ꢀ4c at the
agonistꢀbinding site of KOR; (b) superimposed structures for compounds 4aꢀc at the
agonistꢀbinding site of MOR (color patterns: 4 as orange; 4a as cyan; 4b as green;
and pink for 4c)
We also docked the compounds 3 and 3a into the agonistꢀbinding site of KOR.
As shown in Figure 3, compound 3a shares the conserved interactions with KOR as
that of compound 3 binding with KOR, for example, compound 3a interacts with
Asp1383.32, Ile2906.51, Met1423.36, Tyr3207.43, Trp2876.48, Tyr1393.33 and H2916.52 of
KOR. The difference is that the oꢀamino substitution at the 7αꢀphenylcarbonyl group
ACS Paragon Plus Environment

Page 17 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
forms hydrogen bond with Tyr3207.43 of KOR, which makes compound 3a show
stronger affinity toward KOR than that of compound 3.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
b
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 18 of 37
1
2
3
4
Figure 3 a). The binding mode of 3a with KOR; b). The binding mode of 3 with KOR
5
6
7
8
9
Based on the docked structures of our compounds binding with either KOR or
MOR, we performed binding energy calculations by means of XꢀScore 1.2, and
explored the correlation between the calculated binding energies and the
wetꢀexperimentally measured binding affinities using the twoꢀdimensional correlation
analyses. The calculated results are shown in Figure S2 and listed in Table S1, which
demonstrated favorable R2 (coefficient of determination) values (for MOR complex,
R2 = 0.7196; for KOR complex, R2 = 0.8484). The good correlation between the
calculated binding affinities and the measured binding affinities support that our
docked binding structures are reasonable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Vivo Activities
The encouraging results of in vitro tests and the modeled binding structures of our
newly designed nepenthone derivatives made us want to know whether or not these
compounds are able to produce analgesic effect in vivo. We selected compound 4,
compound 3, and compound U50488H as a control, for further animal tests using hot
plate approach. The results of ED50 values are listed in Table 3. Both compound 3a
and 4 showed strong antinocieceptive effect in hot plate test, and compound 4 showed
a doseꢀdependent ED50 value of 2.1 (1.5ꢀ3.1) mg/kg, which is approximately 2ꢀfold
more potent than that of U50488H31.
ACS Paragon Plus Environment

Page 19 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Table 3 ED50 values of the antinociception induced by Compounds 3a, 4 and
U50488H in the hot plate test. The ED50 value was calculated from data obtained at
15 min (3a) or 2h (4) after administration. In parentheses are 95% confidence limits.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compd
Antinociception ED50 (mg/kg)
6.8 (6.5ꢀ7.1)
3a
2.1 (1.5ꢀ3.1)
4
4.4 (2.5ꢀ7.8)a
U50488H
aThe data of (ꢀ)U50,488H were cited from our group’s results in previously published paper.31
More interestingly as shown in Figure 4, the antinocieceptive effect of compound 4
peaked from 2h to 8h after injection in the hot plate test and then gradually declined.
After 24 h, it still produced almost 35% of antinociception. In comparison,
U50488Hꢀinduced antinociception returned to the preꢀinjection level after 2h. These
results suggest that compound 4 is able to produce much longer active antinociception
than that of the classical agonist U50488H. Further in vivo tests showed that, the
compound 4ꢀinduced antinociception can be significantly inhibited by the
KORꢀselective antagonist norꢀBNI (as shown in Figure 5), we concluded that the
antinociceptive effect of compound 4 showed in the hot plate test should work
through its selective binding with KOR.
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 20 of 37
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4 Time courses for the effects of Compound 4 (5, 6, 7.5 mg/kg, s.c.) and
U50488H (7 mg/kg, s.c.) induced antinociception in the hot plate test.
Figure 5 Compoud 4ꢀinduced antinociception was mediated by kappa opioid
receptor. Mice were pretreated with norꢀBNI (10 mg/kg, s.c.) for 24 h, then injected
with compound 4 (6 mg/kg, s.c.). 2 h later, the antinociceptive effect was assessed in
the hot plate test.
ACS Paragon Plus Environment

Page 21 of 37
ACS Chemical Neuroscience
1
2
3
4
CONCLUSION
5
6
7
8
9
Starting from the morphineꢀlike structure of thebaine derivatives in our previous
work15, 16, we explored possibilities to make them as selective KOR agonists, by
designing novel nepenthone derivatives. We applied two strategies on the structure of
nepenthone, i.e., inserting a spacer (carbonyl group) between the substructure of
6α,14αꢀendoꢀethenylꢀthebaine and the 7αꢀphenylcarbonyl substitution; and replacing
the 17ꢀNꢀmethyl group with a cyclopropylmethyl group. We synthesized two series of
new compounds (3, 3aꢀc; and 4, 4aꢀc) and subjected them to both the in vitro tests
(binding assays and functional assays), and molecular docking operations to obtain
their binding structures with KOR. The measured data and binding structures showed
that compounds 4, 4aꢀc are potent selective KOR agonists at nM level of binding
affinities. We performed in vivo tests for antinociception on compounds 4, 3a with
U50488H as the control, and found that compound 4 is able to induce stronger and
much longer antinociceptive effect than that of U50488H, and that antinociception
can be reversed by the KOR antagonist norꢀBNI. All these results support that
compound 4 is a very good lead compound for future design of potential chemical
entitles acting through KOR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
METHODS
Chemistry
Reagents and solvents were purchased from Sinopharm Chemical Reagent Co., Ltd,
Acros and other suppliers, which were used as received. Tetrahydrofuran was boiled
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 22 of 37
1
2
3
4
5
6
7
8
9
with sodium and distilled before use. Melting points were determined in glass
capillary tubes and were uncorrected. IR data were taken on an AVATAR 360 FTꢀIR
spectrometer (KBr). NMR data were recorded with Mercury plus 400 MHz or Bruker
ASCEND 600 MHz NMR system. Chemical shifts (δ) are expressed in parts per
million (ppm) relative to tetramethylsilane (TMS) as an internal standard. Signals of
active hydrogen were examined by D2O exchange. Mass spectra were measured on an
Agilent 1100 series LC/MSD 1946D spectrometer. Highꢀresolution mass spectra
(HRMS) were conducted on an AB SCIEX TripleTOF 5600+ mass spectrometer. The
content of compounds for biological evaluation was examined by WATERS Breeze
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TM
2 HPLC system with methanol: water (80:20 or 85:15) as the eluent. Unless
specified, the purity of target compounds was >95% which was considered to be pure
enough for biological assays.
Synthesis of nepenthone, 3
Nepenthone was prepared according to the procedure described in the literature24.
Nepenthone, 5.92 g (83.4%), mp: 153ꢀ154°C (lit. mp: 155°C).
General Procedure for the synthesis of aminoꢀnepenthones, 3aꢀc
1 mL (17.3 mol) of 85% Hydrazine hydrate /Raney Ni (cat.) was added fractionally
to a solution of 1 g (2.05 mmol) of nitronepenthone in 50 mL ethanol at 60°C. The
reaction was completely confirmed by TLC. Raney Ni was removed by filtration and
the product was recrystallized from appropriate solvents to yield aminonepenthones.
o-aminoꢀnepenthone, 3a
ACS Paragon Plus Environment

Page 23 of 37
ACS Chemical Neuroscience
1
2
3
4
Colorless needle (from ethanol/water) 0.75g, 79.9%, mp: 182ꢀ183°C; IR (KBr) υ
5
6
7
1
3447.21, 3335.72, 1647.29 cmꢀ1; H NMR (DMSOꢀd6) δ 7.98 (1H, d, J=7.44 Hz),
8
9
7.21 (1H, m, J1=7.82 Hz, J2=7.64 Hz), 7.11 (2H, s, NꢀH, this peak is diminished when
treated with D2O), 6.70 (1H, brd, J1=1.17 Hz, J2=8.60 Hz), 6.59 (1H, d, J=8.22 Hz),
6.55 (1H, m, J1=7.63 Hz, J2=7.43 Hz), 6.48 (1H, d, J=8.21 Hz), 5.79 (1H, d, J=8.99
Hz), 5.35 (1H, d, J=8.61 Hz), 4.78 (1H, s), 3.68 (3H, s), 3.24 (3H, s), 3.12ꢀ3.06 (2H,
m), 2.98 (1H, dd, J1=10.12 Hz, J2=12.13 Hz), 2.46ꢀ2.35 (2H, m), 2.25 (3H, s),
2.25ꢀ2.23 (2H, m), 1.66 (1H, brd, J=9.78 Hz), 1.08 (1H, dd, J1=6.65 Hz, J2=12.33 Hz)
ppm; 13CNMR (DMSOꢀd6) δ 200.71, 151.23, 147.54, 141.04, 134.14, 133.68, 132.86,
131.45, 128.17, 127.79, 119.01, 117.21, 116.66, 114.17, 113.23, 92.21, 81.21, 59.09,
55.92, 51.70, 46.40, 44.94, 42.94, 42.62, 42.54, 32.63, 32.34, 21.67 ppm; MS
(LCMS) m/z 459.1 (M+H) +; HRMS (m/z): [M+H]+ Calculated for C28H30N2O4:
459.2283; found: 459.2258.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
m
ꢀaminoꢀnepenthone, 3b
Light yellow powder, 6.36g, 84.6%, mp: 172ꢀ173°C. IR (KBr) υ 3463.37, 3320.72,
1
1674.88 cmꢀ1; H NMR (DMSOꢀd6) δ 7.19ꢀ7.09 (3H, m), 6.76 (1H, brd, J=8.99 Hz),
6.60 (1H, d, J=8.21 Hz), 6.49 (1H, d, J=8.22 Hz), 5.75 (1H, d, J=9.00 Hz), 5.39 (1H,
d, J=8.99 Hz), 5.30 (2H, s, this peak is diminished when treated with D2O), 4.76 (1H,
d, J=0.78 Hz), 3.93 (1H, dd, J1=7.04 Hz, J2=9.39 Hz), 3.68 (3H, s), 3.21 (3H, s),
3.14ꢀ3.07 (2H, m), 2.99 (1H, dd, J1=9.78 Hz, J2=11.74 Hz), 2.46ꢀ2.36 (2H, m), 2.25
(3H, s), 2.22ꢀ2.17 (2H, m), 1.69 (1H, d, J=11.73 Hz), 1.09 (1H, dd, J1=6.65 Hz,
J2=12.14 Hz) ppm; 13CNMR (DMSOꢀd6) δ 209.94, 159.35, 158.10, 151.65, 149.11,
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 24 of 37
1
2
3
4
5
6
144.59, 144.11, 139.41, 138.77, 137.96, 129.66, 128.73, 128.64, 126.38, 123.93,
123.23, 123.13, 102.81, 91.83, 69.64, 66.54, 62.28, 57.03, 55.53, 53.51, 53.23, 43.06,
7
8
9
+
42.77, 32.27 ppm; MS (LCMS) m/z 459.2 (M+H) ; HRMS (m/z): [M+H]+
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Calculated for C28H30N2O4: 459.2283; found: 459.2257.
p
ꢀaminoꢀnepenthone, 3c
Offꢀwhite solid, 10.25g (76.2%), mp: 238ꢀ240°C; IR (KBr) υ 3482.74, 3347.15,
1
1595.69 cmꢀ1; H NMR (DMSOꢀd6) δ 7.74 (2H, d, J=8.61 Hz), 6.59 (1H, d, J=8.21
Hz), 6.54 (2H, d, J=8.60 Hz), 6.48 (1H, d, J=8.22 Hz), 6.00 (2H, s, NꢀH, this peak is
diminished when treated with D2O), 5.73 (1H, d, J=9.00 Hz), 5.35 (1H, d, J=9.00 Hz),
4.76 (1H, s, Hꢀ5), 3.89 (1H, dd, J1=7.43 Hz, J2=8.30 Hz), 3.67 (3H, s), 3.21 (3H, s),
3.11ꢀ3.06 (2H, m), 2.92 (1H, dd, J1=10.17 Hz, J2=11.74 Hz), 2.50ꢀ2.35 (2H, m), 2.25
(3H, s), 2.23ꢀ2.20 (2H, m), 1.68ꢀ1.65 (1H, m), 1.05 (1H, dd, J1=6.65 Hz, J2=11.94 Hz)
ppm; 13CNMR (DMSOꢀd6) δ 195.95, 153.30, 147.54, 141.02, 134.14, 133.04, 130.52,
128.19, 127.60, 125.30, 118.99, 113.21, 112.31, 92.30, 81.19, 59.10, 55.91, 51.64,
46.40, 44.96, 42.96, 42.62, 41.36, 32.42, 32.27, 21.66 ppm; MS (LCMS) m/z
459.1(M+H)+; HPLC purity: 92.6%, TR=11.58min (MeOH:water= 80:20); HRMS
(m/z): [M+H]+ Calculated for C28H30N2O4: 459.2283; found: 459.2256.
Synthesis of 17ꢀNꢀnorꢀN-cyclopropylmethyl nepenthones, 4
10 g (21.5 mmol) of Nꢀnorꢀnepenthone hydrochloride was dissolved in 100 mL
anhydrous DMF. Then anhydrous K2CO3 9.0 g (65.1 mmol) and 2.6 mL (26.8 mmol)
cyclopropylmethyl bromide were added. The mixture was heated to 80ꢀ90 °C and
ACS Paragon Plus Environment

Page 25 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
reacted for 10 h. The K2CO3 was removed by filtration and 90 mL of 10% Na2CO3
and 150 mL ethyl acetate were added to the filtrate. The organic phase was separated
and washed with of appropriate amount of water and brine. After being dried with
anhydrous Na2SO4, it was filtered and evaporated to dryness. The residue was further
purified by flash column chromatography to afford the title compound.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
White foam, 3.6 g (34.6%), m.p. 57.5ꢀ58.9°C, IR (KBr) υ3435.62, 2925.92, 1683.13,
1212.49, 1102.81, 652.05 cmꢀ1, 1H NMR (CDCl3) δ 8.02(2H,d,J=7.04Hz),
7.61(1H,t,J1=7.73Hz, J2=7.44Hz), 7.51 (2H, t, J1=7.44Hz, J2=7.43Hz), 6.59(1H, d,
J=8.02Hz), 6.47(1H, d, J=8.22Hz), 5.75(1H, d, J=8.61Hz), 5.45(1H, d, J=8.61Hz),
4.81(1H,s), 4.16ꢀ4.13(1H,m), 3.68(3H,s), 3.47(1H, d, J=6.26Hz), 3.21(3H,s),
3.05ꢀ2.98(2H,m), 2.68(1H, d, J=6.26Hz), 2.45ꢀ2.39(1H, m), 2.31ꢀ2.25(4H, m),
1.69(1H, d, J=9.00Hz), 1.21ꢀ1.13(2H, m), 0.76(1H, brs), 0.45ꢀ0.39(2H,m),
0.09ꢀ0.06(2H,m) ppm. 13CNMR (DMSOꢀd6) δ 200.18, 147.679, 141.352, 138.089,
134.674, 134.445, 133.15, 128.842, 128.53, 128.393, 127.211, 119.512, 113.527,
92.716, 81.937, 59.227, 56.780, 56.201, 52.222, 47.518, 43.806, 42.899, 42.693,
32.752, 32.074, 22.773, 9.463, 4.066, 3.494 ppm. MS (LCMS) m/z 484.3 (M+H)+.
HRMS (m/z): [M+H]+ Calculated for C31H33N1O4: 484.2482; found: 484.2486.
General Procedure for the synthesis of 17ꢀNꢀnorꢀN-cyclopropylmethyl–
aminonepenthones, 4aꢀc
1 mL (17.3 mol) of 85% Hydrazine hydrate /Raney Ni (cat.) was added fractionally
to a solution of 500 mg (1.89mmol) of N-CPM-N-norꢀnitronepenthone in 50 mL
ethanol at 60 °C. The reaction was complete as shown by TLC analysis. Raney Ni
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 26 of 37
1
2
3
4
5
6
7
8
9
was removed by filtration. The crude product was further purified by flash column to
yield the appropriate NꢀcyclopropylmethylꢀNꢀ norꢀaminonepenthone.
N-cyclopropylmethyl-N-normethyl -o-aminonepenthone, 4a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Light yellow foam, 940 mg (89.1%), mp: 88ꢀ90°C, IR (KBr) υ 3453.41, 2915.07,
1614.78, 1210.36, 1102.48 cmꢀ1; 1H NMR (DMSOꢀd6) δ 8.02(1H, d, J=7.83 Hz), 7.21
(1H, t, J1=7.83 Hz, J2=7.43 Hz), 7.12 (2H, s), 6.70 (1H, d, J=7.83Hz), 6.60ꢀ6.54 (2H,
m), 6.47 (1H, d, J=8.20 Hz), 5.80 (1H, d, J=9.00 Hz), 5.37 (1H, d, J=8.61 Hz),
4.05ꢀ4.00(1H, m), 4.80 (1H, s), 3.68 (3H, s), 3.46 (1H, d, J=6.26 Hz), 3.25 (3H, s),
3.05ꢀ2.97 (2H, m), 2.67 (1H, brs), 2.44ꢀ2.38 (1H, m), 2.33ꢀ2.22 (4H, m), 1.68 (1H, d,
J=10.56 Hz), 1.14ꢀ1.07 (1H, m), 0.76 (1H, brs), 0.42 (2H, brs), 0.076 (2H, brs) ppm;
13CNMR (DMSOꢀd6) δ 201.331, 151.331, 147.824, 141.344, 134.567, 134.194,
133.500, 131.785, 128.576, 128.050, 119.428, 117.591, 116.935, 114.709, 113.474,
92.510, 81.678, 59.212, 56.735, 56.209, 52.008, 47.457, 43.859, 42.868, 42.685,
32.973, 32.821, 22.781, 9.471, 4.119, 3.441 ppm; MS (LCMS) m/z 499.3(M+H)+.
HRMS (m/z): [M+H]+ Calculated for C31H34N2O4: 499.2597; found: 499.2566.
N-cyclopropylmethyl -N-normethyl -m- aminonepenthone, 4b
Light yellow foam, 450 mg (92.4%), mp: 93ꢀ96°C; IR (KBr) υ3445.65, 3372.95,
2925.69, 1681.38, 1627.18, 1600.10, 1451.95, 1101.86 cmꢀ1; 1H NMR (DMSOꢀ d6) δ
7.21(1H, d, J=7.83 Hz), 7.16ꢀ7.10(2H, m), 6.77(1H, dd, J1=1.56 Hz, J2=7.82 Hz),
6.59(1H, d, J=8.22 Hz), 6.47(1H, d, J=8.21 Hz), 5.74(1H, d, J=8.61 Hz), 5.41(1H, d,
J=8.61Hz), 5.29 (2H, s, this peak is diminished when treated with D2O), 4.76 (1H, s),
3.97 (1H, dd, J1=7.04 Hz, J2=9.19 Hz), 3.68 (3H, s), 3.47 (1H, d, J=6.26 Hz), 3.22
ACS Paragon Plus Environment

Page 27 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
(3H, s), 3.05ꢀ2.98 (2H, m), 2.68(1H, brd, J=9.00 Hz), 2.42 (1H, dd, J1=6.65 Hz,
J2=18.78 Hz), 2.26ꢀ2.18 (4H, m), 1.70 (1H, brd, J=12.13 Hz), 1.12ꢀ1.10 (1H, m), 0.77
(1H, s), 0.46ꢀ0.39(2H, m), 0.079(2H, brs) ppm; 13CNMR (DMSOꢀd6) δ 200.165,
148.922, 147.740, 141.344, 138.875, 134.415, 134.186, 129.307, 128.568, 127.630,
119.496, 118.604, 116.325, 113.489, 112.933, 92.434, 81.723, 59.220, 56.712,
56.193, 51.970, 47.472, 43.844, 42.868, 42.670, 32.897, 32.485, 22.758, 9.456, 4.104,
3.471 ppm; MS (LCMS) m/z 499.3 (M+H)+. HRMS (m/z): [M+H]+ Calculated for
C31H34N2O4: 499.2597; found: 499.2565.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N-cyclopropylmethyl -N-normethyl -p-aminoꢀnepenthone, 4c
White powder, 550 mg (99.3%), mp 214ꢀ217°C; IR (KBr) υ3454.36, 3367.90,
1
1660.69, 1596.93, 1178.60 cmꢀ1; H NMR (DMSOꢀd6) δ 7.74 (2H, d, J=8.61 Hz),
6.58 (1H, d, J=8.22 Hz), 6.56 (2H, d, J=9.00 Hz), 6.46 (1H, d, J=7.82 Hz), 6.00(2H, s,
this peak is diminished when treated with D2O), 5.72 (1H, d, J=9.00 Hz), 5.37 (1H, d,
J=8.61 Hz), 4.77 (1H, s), 3.94ꢀ3.90 (1H, m), 3.68 (3H, s), 3.44 (1H, d, J=6.26), 3.22
(3H, s), 3.02ꢀ2.94(2H, m), 2.67(1H, brd, J=7.04 Hz), 2.41 (1H, dd, J1=6.26 Hz,
J2=18.58 Hz), 2.29ꢀ2.20 (4H, m), 1.68 (1H, brd, J=9.78), 1.07 (1H, dd, J1=6.65 Hz,
J2=12.12 Hz), 0.76 (1H, brs), 0.45ꢀ0.39 (2H, m), 0.051 (2H, brs) ppm; 13CNMR
(DMSOꢀd6) δ196.727,153.534, 147.809, 141.322, 134.575, 133.683, 130.931,
128.583, 127.882, 125.717, 119.405, 113.436, 112.681, 103.678, 92.571, 81.655,
59.220, 56.780, 56.193, 51.940, 47.442, 43.867, 42.860, 41.526, 32.874, 32.781,
9.478, 4.089, 3.471 ppm; MS (LCMS) m/z 499.3 (M+H)+. HRMS (m/z): [M+H]+
Calculated for C31H34N2O4: 499.2597; found: 499.2569.
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 28 of 37
1
2
3
4
Computational Approaches
5
Homology Modeling and Structure Validation:
6
7
8
9
The active murine MOR was structurally determined at a resolution of 2.1
Å ( Protein
(
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Data Bank entry: 5C1M 26), it provides a solid template for modeling the active 3D
structures of MOR and KOR. The active human 3D MOR model and the active human
3D KOR model were constructed using a homologyꢀmodeling approach based on the
xꢀray crystal structure of murine active MOR. The “canonical” sequences of human
MOR (P35372ꢀ1) and human KOR (P41145ꢀ1) were downloaded from UniProt
database (www.uniprot.org). The residues of human MOR sequence before P65 at the
N terminus, as well as the residues after I349 at the C terminus were truncated. The
Blast program encoded in Discovery Studio 3.5 (Discovery Studio 3.5 (San Diego,
CA)) was used to align the sequences of murine active MOR and human MOR and
KOR, the Modeller program encoded in Discovery Studio 3.5 was used to build the
models of human MOR and KOR. 50 homology models were built for each receptor.
The one with the lowest DOPE scores was submitted to energy minimization (1000
steps steepest descent with backbone constrained in CHARMm force field). The
resulting models were then validated by PROCHECK32 Ramachandran plots for their
stereochemical qualities (See Figure S1). Energy minimizations of the models were
carried out using the Discover module encoded in Discovery Studio 3.5 with the same
parameters as that of our previous studies 33.
Molecular Docking: The structures of compound 3a, 4, 4a, 4b and 4c were sketched
in SYBYL (Tripos, St Louis, MO, USA) and energyꢀminimized using the Tripos
ACS Paragon Plus Environment

Page 29 of 37
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Force Field (GasteigerꢀHückel charges, distanceꢀdependent dielectric constant is 4.0;
nonꢀbonded interaction cutoff is 8 Å; termination criterion: the energy gradient <0.05
kcal/(mol×Å) for 10,000 iterations).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The docking program GOLD Suite 5.1 (Cambridge Crystallographic Data Centre,
Cambridge, UK) with the genetic algorithm was employed to dock the ligands into the
active sites of human MOR and KOR. The binding site was defined as a sphere with the
radius of 15 Å around the atom Cβ of Asp138 3.32. Side chains of residues Glu2.60
,
Asp3.32, Tyr3.33 and Tyr7.43 were set to be flexible during the docking run, referring to the
rotamer library in GOLD. Full flexibility was allowed for all the rotatable bonds of
ligands. The number of genetic algorithm (GA) runs was set to 10 and the GoldScore
was used as the scoring function. The topꢀranked poses were visually inspected by
considering favorable interactions with the key residues demonstrated by available
mutagenesis studies. The Binding affinities were predicted by XꢀScore 1.2 with
empirical scoring functions. The predicted affinities and all of available experimental
data were used for the selection of the final complex. The resulting ligandꢀreceptor
complexes were plotted using the PyMOL (The PyMOL molecular graphics system,
version 0.98, DeLano Scientific, Palo Alto, CA) software.
Pharmacological assays
U50488H, norꢀBNI (norꢀbinaltorphimine) were purchased from SigmaꢀAldrich.
Morphine hydrochloride was supplied by Qinghai Pharmaceutical General Factory
(Qinghai province, China).
Cell culture and membrane preparation
ACS Paragon Plus Environment