Synthesis and Structure-Activity Relationships of 5′-Aryl-14-alkoxypyridomorphinans: Identification of a μ Opioid Receptor Agonist/δOpioid Receptor Antagonist Ligand with Systemic Antinociceptive Activity and Diminished Opioid Side Effects

Article abstract of DOI:10.1021/acs.jmedchem.0c00503

We previously identified a pyridomorphinan (6, SRI-22138) possessing a 4-chlorophenyl substituent at the 5′-position on the pyridine and a 3-phenylpropoxy at the 14-position of the morphinan as a mixed μ opioid receptor (MOR) agonist and /κ opioid receptor (DOR/KOR) antagonist with potent antinociceptive activity and diminished tolerance and dependence in rodents. Structural variations at the 5′- and 14-positions of this molecule gave insights into the structure-activity relationships for binding and functional activity. Subtle structural changes exerted significant influence, particularly on the ability of the compounds to function as agonists at the MOR. In vivo evaluation identified compound 20 (SRI-39067) as a MOR agonist/DOR antagonist that produced systemically active potent antinociceptive activity in tail-flick assay in mice, with diminished tolerance, dependence/withdrawal, reward liability, and respiratory depression versus morphine. These results support the hypothesis that mixed MOR agonist/DOR antagonist ligands may emerge as novel opioid analgesics with reduced side effects.

Full text of DOI:10.1021/acs.jmedchem.0c00503

Subscriber access provided by Uppsala universitetsbibliotek
Article
Synthesis and Structure–Activity Relationships of 5#-Aryl-14-
alkoxypyridomorphinans: Identification of a Mu Opioid Receptor
Agonist/Delta Opioid Receptor Antagonist Ligand with Systemic
Antinociceptive Activity and Diminished Opioid Side Effects.
Rakesh H. Vekariya, Wei Lei, Abhisek Ray, Surendra K. Sainai, Sixue Zhang,
Gabriella Molnar, Deborah Barlow, Kelly L. Karlage, Edward Bilsky, Karen
Houseknecht, Tally M. Largent-Milnes, John M. Streicher, and Subramaniam Ananthan
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00503 • Publication Date (Web): 12 Jun 2020
Downloaded from pubs.acs.org on June 13, 2020
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 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 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.
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 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis and Structure–Activity Relationships of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
′-Aryl-14-alkoxypyridomorphinans: Identification
of a Mu Opioid Receptor Agonist/Delta Opioid
Receptor Antagonist Ligand with Systemic
Antinociceptive Activity and Diminished Opioid
Side Effects.
†
‡
†
†
†
Rakesh H. Vekariya, Wei Lei, Abhisek Ray, Surendra K. Saini, Sixue Zhang, Gabriella
‡
§
‡
‡
§
Molnar, Deborah Barlow, Kelly L. Karlage, Edward J. Bilsky, Karen L. Houseknecht, Tally
‡
‡
,†
M. Largent-Milnes, John M. Streicher, and Subramaniam Ananthan*
†Chemistry Department, Southern Research, Birmingham, Alabama 35205, United States,
‡Department of Pharmacology, College of Medicine, University of Arizona, Tucson, Arizona
§
8
5724, United States, Department of Biomedical Sciences, College of Osteopathic Medicine,
University of New England, Biddeford, Maine 04005, United States.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
We previously identified a pyridomorphinan (6, SRI-22138) possessing a 4-chlorophenyl
substituent at the 5′-position on the pyridine and a 3-phenylpropoxy at the 14-position of the
morphinan as a mixed mu opioid receptor (MOR) agonist and delta/kappa opioid receptor
(DOR/KOR) antagonist with potent antinociceptive activity and diminished tolerance and
dependence in rodents. Structural variations at the 5′- and 14-positions of this molecule gave
insights into the structure–activity relationships for binding and functional activity. Subtle
structural changes exerted significant influence, particularly on the ability of the compounds to
function as agonists at MOR. In vivo evaluation identified compound 20 (SRI-39067) as a MOR
agonist/DOR antagonist that produced systemically active potent antinociceptive activity in the
tail-flick assay in mice, with diminished tolerance, dependence/withdrawal, reward liability, and
respiratory depression versus morphine. These results support the hypothesis that mixed MOR
agonist/DOR antagonist ligands may emerge as novel opioid analgesics with reduced side effects.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
2

Page 3 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
For relief of moderate to severe acute pain, opioids remain the mainstay of treatment. However,
the use of opioid analgesics is associated with severe limiting side effects including abuse liability,
respiratory depression, constipation and development of tolerance and dependence.1-7 Therefore,
considerable effort has been expended on the discovery of novel opioid analgesics that are as
efficacious as the prototypic opioid analgesic morphine but possessing reduced side effects. Opioid
drugs exert their analgesic effects through binding and activation of the three opioid receptor
subtypes, namely the mu opioid receptor (MOR), the delta opioid receptor (DOR) and the kappa
opioid receptor (KOR).8,9 Several lines of evidence suggest that there are physical and/or
functional interactions between these G-protein coupled receptors (GPCRs), and molecules that
interact with multiple receptors may prove advantageous in the search for opioid ligands with
diminished side effect profiles.10-15 Considerable evidence from gene knockout/knockdown studies
and studies using pharmacological probes suggests that activation of MOR with simultaneous
inhibition of DOR leads to MOR-mediated analgesic activity with diminished development of
tolerance, dependence, and potentially other side effects.13,15-19 Based on these observations,
we20,21 and others13,22-28 pursued the development of mixed function ligands possessing a MOR
agonist/DOR antagonist profile of activity using peptidic, peptidomimetic, and nonpeptide opioid
templates. Exemplary MOR agonist/DOR antagonist ligands evaluated in animal models is shown
in Figure 1.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
HO
HO
OH
O
O
O
O
HO
N
O
OH
OH
O
S
O
HN
N
O
H
H
N
H N
2
H
N
N
NH2
NH2
O
N
N
H N
2
OH
N
NH2
H
H
H
O
O
O
S
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
, DIPP
ψ
NH2
2, VRP26
3, AAH8
OCH3
CH3
OH
O
N
17
N
N
O
OH
N
N
H
14
4'
CH3
3
Cl
Cl
H CO
3
O
O
5'
O
OCH3
HO
O
HO
N
HO
N
HO
O
4
, Mitragynine pseudoindoxyl
5, UMB 425
6, SRI-22138
7, SRI-9409
Figure 1. Structures of MOR agonist/DOR antagonist ligands 1–6 and the MOR-DOR antagonist
7
.
We previously synthesized and evaluated a series of 14-alkoxypyridomorphinans possessing a 4-
(chlorophenyl) group at the 5′-position on the pyridine ring that led to the identification of
compound 6 as a ligand possessing a balanced binding affinity profile at MOR and DOR with
2
1
agonist activity at MOR and potent antagonist activity at DOR. Pharmacological evaluation with
this compound demonstrated antinociceptive activity comparable to morphine in the mouse warm
water tail-flick assay. Moreover, on repeated administration, compound 6 displayed diminished
development of analgesic tolerance compared to morphine. These studies with compound 6 were
performed by intracerebroventricular (icv) administration due to its poor blood brain barrier
penetration capacity. The current effort was undertaken to expand the structure–activity
relationship (SAR) in the pyridomorphinan scaffold of this prototype compound 6 and to identify
improved ligands with mixed MOR agonist/DOR antagonist profile and acceptable absorption,
ACS Paragon Plus Environment
4

Page 5 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
distribution, metabolism, and excretion (ADME) and pharmacokinetic (PK) properties to enable
proof-of-concept studies with systemic administration.
RESULTS AND DISCUSSION
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound Design. The design principle that led to the identification of compound 6 was that a
non-selective dual MOR/DOR antagonist pyridomophinan 7 (Figure 1) possessing a C-14
hydroxyl group could be transformed to a MOR agonist with retention of DOR antagonist profile
2
1
by installation of a 3-(phenylpropyl) group on the C-14 oxygen atom. In our previous DOR
antagonist-focused effort, we had demonstrated that a number of compounds related to 7
possessing various aryl and heteroaryl groups at the 5′-position displayed a dual MOR/DOR
antagonist profile similar to that of 7.20,29-31 Therefore, we pursued the installation of a 3-
(phenylpropyl) group on C-14 oxygen on a series of pyridomorphinans possessing substituent
variations on the phenyl group at the 5′-position (compounds 8–29, Table 1) and a few heteroaryl
variants in place of the phenyl group at the 5′-position (compounds 30–42, Table 2). In addition,
we pursued targeted changes to the alkyl group on the oxygen at C-14 using 6 and its deschloro
analogue 8 as the templates. This study was intended to explore the consequence of installation of
a substituent other than a 3-(phenylpropyl) group (compounds 43–52, Table 3) on opioid receptor
binding and functional activity.
Synthesis. Several synthetic approaches were utilized for the preparation of the desired target
compounds. The original synthetic route developed for the preparation of compound 6 was adopted
for the preparation of the 5′-aryl target compounds 8–10 and the 5′-heteroaryl target compounds
3
6 and 38, as depicted in Scheme 1. The starting materials 54–58 were synthesized by annulation
of the pyridine ring on the C-ring of naltrexone (53) by condensation with substituted
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
malondialdehydes and ammonium acetate in acetic acid. Dialkylation of the phenol and the tertiary
alcohol groups with phenylpropyl bromide followed by selective removal of the phenypropyl
group from the phenolic oxygen using BBr yielded the desired target compounds.
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of Pyridomorphinans 8–10, 36, and 38a
N
N
N
OH
OH
a
O
b, c
R
R
O
O
O
O
HO
N
HO
N
HO
53
54, R = phenyl
8, R = phenyl
55, R = 4-bromophenyl
56, R = 4-fluorophenyl
57, R = 4-pyrimidinyl
58, R = 2-pyrazinyl
9, R = 4-bromophenyl
10, R = 4-fluorophenyl
36, R = 4-pyrimidinyl
38, R = 2-pyrazinyl
aReagents and conditions: (a) RCH(CHO)
DMF, 0 °C to rt, 4 h; (c) BBr , CH Cl , –78 °C to 0 °C, 1 h.
, AcONH
, AcOH, reflux, 16 h; (b) NaH, Ph(CH
2
4
2
3
) Br,
3
2
2
We also developed an alternative, more versatile approach utilizing the 3-bromopyridine 60,
3
0
prepared by dialkylation of the previously described 59, as the key starting material. Suzuki
coupling of 60 with arylboronic acids or esters, followed by deprotection of the phenol, as depicted
in Scheme 2, gave the desired target compounds 11–18, 22–29, 31–34, 37, and 39–41. Similarly,
coupling of 60 with pyrrole using Pd
2
(dba)
3
as the catalyst in the presence of Cs
gave 30 (Table 2).
2
CO
3
and P(t-Bu)
3
in toluene followed by 3-O-dealkylation with BBr
3
ACS Paragon Plus Environment
6

Page 7 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 2. Synthesis of Pyridomorphinans 11–18, 22–29, 31–34, 37, and 39–41a
N
N
N
N
O
OH
a
b
c
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Br
Br
O
O
O
N
R
HO
N
O
R
O
N
O
HO
N
60
59
6
1, 2,4-difluorophenyl
11, 2,4-difluorophenyl
12, 3,4-difluorophenyl
13, 2-methylphenyl
14, 3-methylphenyl
15, 4-methylphenyl
16, 4-fluoro-2-methylphenyl
17, 2-cyclopropylphenyl
18, 2,6-dimethylphenyl
62, 3,4-difluorophenyl
6
6
6
6
6
6
6
3, 2-methylphenyl
4, 3-methylphenyl
5, 4-methylphenyl
6, 4-fluoro-2-methylphenyl
7, 2-cyclopropylphenyl
8, 2,6-dimethylphenyl
9, 3-hydroxyphenyl
22, 3-hydroxyphenyl
70, 2-dimethylaminophenyl
23, 2-dimethylaminophenyl
24, 3-dimethylaminophenyl
25, 4-dimethylaminophenyl
71, 3-dimethylaminophenyl
2, 4-dimethylaminophenyl
7
73, 2-(dimethylaminomethyl)phenyl 26, 2-(dimethylaminomethyl)phenyl
7
7
4, 2-aceylaminophenyl
5, 3-acetylaminophenyl
27, 2-aceylaminophenyl
28, 3-acetylaminophenyl
29, 4-acetylaminophenyl
31, 2-pyridyl
32, 3-pyridyl
33, 4-pyridyl
34, 2-methyl-4-pyridyl
37, 5-pyrimidinyl
39, 4-pyrazolyl
40, 1-methyl-4-pyrazolyl
41, 3-methyl-4-isoxazolyl
76, 4-acetylaminophenyl
7
7
7
8
8
8
8
7, 2-pyridyl
8, 3-pyridyl
9, 4-pyridyl
0, 2-methyl-4-pyridyl
1, 5-pyrimidinyl
2, 4-pyrazolyl
3, 1-methyl-4-pyrazolyl
84, 3-methyl-4-isoxazolyl
aReagents and conditions: (a) NaH, Ph(CH
)
3
Br, DMF, 0 °C to rt, 4 h; (b) RB(OH)
, CH Cl , –78 °C to 0 °C, 1 h.
or RBpin,
2
2
Pd(PPh , K CO , 10:1 DMF/H O, mw, 100 °C, 1 h; (c) BBr
3
)
4
2
3
2
3
2
2
For the preparation of the targets containing a methoxy group (19–21 and 35), intermediates
carrying benzyl as the phenolic protecting group were utilized to facilitate selective deprotection
under reductive or acidic conditions (Scheme 3). The benzyl protected bromopyridine 85 was
accessible from 59 via benzylation of the phenolic OH followed by phenypropylation of the 14-
OH group.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 3. Synthesis of Pyridomorphinans 19–21 and 35a
N
N
N
O
O
a
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
R
Br
R
O
O
O
N
N
HO
BnO
N
BnO
85
86, R = 2-methoxyphenyl
19, R = 2-methoxyphenyl
20, R = 3-methoxyphenyl
21, R = 4-methoxyphenyl
35, R = 2-methoxy-4-pyridyl
87, R = 3-methoxyphenyl
88, R = 4-methoxyphenyl
89, R = 3-methoxy-4-pyridyl
aReagents and conditions: (a) RBpin, Pd(PPh
TFA, reflux, 1.5 h or H , 10% Pd/C, 1:1 CH Cl
)
4
, K
CO
, 10:1 DMF/H O, mw, 100 °C, 1 h; (b)
3
2
3
2
2
2
2
/MeOH, 20 h.
The oxadiazolyl compound 42 was prepared from the carbethoxypyridine 90 via conversion to
hydrazide, cyclization to oxadiazole with triethyl orthoacetate, followed by etherification of the
1
4-OH and deprotection of the phenol as shown in Scheme 4.
Scheme 4. Synthesis of Oxadiazolyl Pyridomorphinan 42a
Ph
N
N
N
OH
OH
O
c, d
a, b
CH3
CH3
O
N
O
N
CO Et
2
O
O
N
O
N
BnO
N
BnO
N
HO
N
90
91
42
aReagents and conditions: (a) N
h; (c) NaH, Ph(CH Br, DMF, 0 °C to rt, 4 h; (d) H
H
4
·H
O, EtOH, reflux, 16 h; (b) CH
C(OEt)
, AcOH, 150 °C, 3
/MeOH, 20 h.
2
2
3
3
2
)
3
2
, 10% Pd/C, 1:1 CH
2
Cl
2
ACS Paragon Plus Environment
8

Page 9 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
For the synthesis of the target compound 43 possessing a pendant 4-pyridyl group in the 14-O-
alkyl substituent, we initially attempted dialkylation of 7 or monoalkylation of 92 with 3-(4-
pyridylpropyl)bromide and related alkylating agents. However, all our attempts under a variety of
reaction conditions failed to yield the desired alkylated intermediates, possibly due to base
promoted elimination reaction predominating over nucleophilic substitution reaction. This
difficulty was overcome by a stepwise derivatization involving 14-O-allylation, Heck coupling
with 4-bromopyridine, saturating the double bond, followed by acid catalyzed deprotection of the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
-O-benzyl group as shown in Scheme 5.
Scheme 5. Synthesis of 14-(3(4-Pyridyl)propoxy)pyridomorphinan 43a
N
N
N
OH
OH
O
c
a
b
Cl
Cl
Cl
O
O
O
HO
N
BnO
N
BnO
N
7
92
93
N
N
N
N
N
N
O
O
O
d
e
Cl
Cl
Cl
O
O
O
BnO
N
BnO
N
HO
N
94
95
43
aReagents and conditions: (a) BnBr, K
DMF, 0 °C to rt, 4 h; (c) 4-bromopyridine hydrochloride, Pd(PPh
1 h; (d) H , 10% Pd/C, 1:1 CH Cl /MeOH, 20 h; (e) TFA, reflux, 1.5 h.
CO
, Me
CO, reflux, 3h; (b) CH
=CHCH
, K
Br, NaH,
2
3
2
2
2
3
)
4
, Pd(OAc)
2
2
CO , 130 °C,
3
2
2
2
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A dialkylation and selective mono-dealkylation sequence (Scheme 6) was successful for the
preparation of the 2-phenoxyethoxy (44) and the 2-quinolinylmethoxy (45) target compounds
using 2-phenoxyethyl triflate and 2-(chloromomethyl)quinoline as the alkylating agent,
respectively.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 6. Synthesis of 14-Alkoxypyridomorphinans 44 and 45a
N
N
N
O
O
OH
O
O
a
b
Cl
Cl
O
Cl
HO
N
O
O
N
HO
N
O
O
7
44
96
N
N
N
N
O
O
c
b
Cl
Cl
O
O
O
N
HO
N
N
97
45
aReagents and conditions: (a) PhOCH
CH
OSO
CF
, PMP, CH
NO
, 50 °C, 1 h; (b) BBr ,
2
2
2
3
3
2
3
o
CH
2
Cl
2
, –78 C to 0 °C, 1 h; (c) 2-(chloromethyl)quinoline, NaH, DMF, 0 °C to rt, 4 h.
The allylation–Heck coupling route shown in Scheme 5 for the preparation of 43 was adapted for
the synthesis of the 3-(3-pyridylpropoxy) compound 46 and the 3-(4-pyridylpropoxy) compound
29
4
7 (Table 3) using the 5′-phenylpyridomorphinan 54 as the starting material. In the preparation
of these compounds, saturation of the double bond and removal of the phenolic-O benzyl
protecting group were accomplished by catalytic hydrogenation in a single step. Dialkylation of
ACS Paragon Plus Environment
1
0

Page 11 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
4 with 4-(fluorophenyl)propyl bromide followed by treatment with BBr to selectively remove
3
the alkyl group from the phenolic oxygen delivered the desired target compound 48 (Table 3). The
cyclohexylpropoxy (49) and the 4-phenylbutoxy (51) compounds were synthesized as depicted in
Scheme 7.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 7. Synthesis of 14-Alkoxypyridomorphinans 49 and 51a
N
R
N
N
OH
O
OH
a
b
O
N
O
O
HO
N
BnO
N
BnO
54
99
, R = cyclohexyl
98
100, R = benzyl
c
R
N
O
O
N
HO
49, R = cyclohexyl
1, R = benzyl
5
aReagents and conditions: (a) BnBr, K
CO
, Me CO, reflux, 3h; (b) 3-cyclohexylpropyl bromide
2
3
2
or 4-phenylbutyl bromide, NaH, DMF, 0 °C to rt, 4 h; (c) BBr
3
, CH
2
Cl , –78 °C to rt (for 49) or
2
TFA, 70 °C, 2 h (for 51).
A synthetic sequence similar to that described in Scheme 5 was deployed for the preparation of
compound 50 using compound 98 as the starting material. Thus, allylation of 98 and palladium
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
catalyzed coupling of the allyl derivative with 4-bromo-3,6-dihydro-2H-pyran gave benzyl
protected diene intermediate which was subjected to catalytic hydrogenation to saturate the double
bonds with simultaneous removal of the benzyl protecting group to obtain the desired compound
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0. Similarly, catalytic hydrogenation of the 14-O-allyl derivative of 98 gave the propoxy
compound 52 (Table 3).
In Vitro Pharmacology and ADME Evaluation. The binding affinity of the compounds at MOR,
DOR, and KOR were evaluated using well-established radioligand displacement assays.
Membranes prepared from CHO cells stably expressing human MOR, DOR, or KOR and
[3
H]diprenorphine as the radioligand were used in determining the binding affinities.32,33 The
35
functional activity evaluations were performed using the [ S]-GTPγS coupling assay using
membranes from the same cells. For MOR agonist experiments, DAMGO was used as the
reference agonist and the Emax values were calculated relative to DAMGO as the full agonist (Emax
100%). For DOR antagonist evaluation, SNC80 was used as the agonist ligand to determine the
3
5
ability of the test compounds to inhibit SNC80 mediated stimulation of [ S]-GTPγS binding. KOR
agonist evaluations were performed similar to MOR agonist experiments with (±)-U50,488 as the
standard agonist ligand (Emax 100%). Since the goal of our effort was to identify compounds with
agonist activity at MOR and antagonist activity at DOR, the compounds were evaluated first as
agonists at MOR and antagonists at DOR. Only selected compounds that were of interest based on
binding and functional profiles at MOR and DOR were then evaluated at KOR to determine
whether they displayed undesirable agonist activity.
All compounds were also evaluated via in vitro ADME assays to determine aqueous solubility, log
D, and mouse liver and human liver microsomal stability. Mouse liver microsomal stability and
log D data are presented along with binding affinity and functional activity data of the compounds
ACS Paragon Plus Environment
1
2

Page 13 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
in Tables 1–3. Human liver microsomal stability, solubility, and calculated physicochemical
properties of the target compounds are given in Table S1 (Supplementary Information).
Structure–Activity Relationships (SAR).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Binding affinity of 5′-aryl compounds at MOR. The data for the group of compounds possessing
structural variations on the 5′-phenyl ring are presented in Table 1. Most of the 5′-aryl compounds
displayed only modest binding affinity at MOR. The lead compound 6 possessing the 4-
chlorophenyl substituent displayed single-digit nanomolar binding affinity at MOR (K
Among compounds with other substituent groups on the phenyl ring, only three compounds, the
-fluorophenyl compound 10, the 2-(dimethylamino)phenyl compound 23, and the 4-
(dimethylamino)phenyl compound 25 showed improved binding affinity at MOR with K values
of 2.33 nM, 6.97 nM, and 4.70 nM, respectively. Compounds that displayed moderate binding
affinity with K value in the range of 10 nM – 100 nM include the unsubstituted (8), 3-methyl (14)
2-methyl-4-fluoro (16), 2-methoxy (19), 3-methoxy (20), 3-hydroxy (22), and 3-dimethylamino
24) phenyl derivatives. In contrast to the 4-fluoro compound with high affinity, the difluoro
i
= 7.55 nM).
4
i
i
(
compounds 11 and 12 displayed a marked reduction in binding affinity. Other substituents that
also had a deleterious effect on MOR affinity include 4-bromo (9), 2-methyl (13), 2,6-dimethyl
(18), 4-methyl (15), 2-cyclopropyl (17), 4-methoxy (21), and 2-, 3- or 4-acetylamino (compounds
2
7, 28 and 29) groups.
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Binding, Functional Activity, and In Vitro ADME Data for 5′-Aryl-14-
(phenylpropoxy) Compounds 8–29.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
O
R
O
HO
N
ACS Paragon Plus Environment
1
4

Page 15 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
3
Binding affinities determined by competitive displacement of [ H]diprenorphine in membrane
b 35
preparations from CHO cells expressing human MOR, DOR, or KOR. [ S]GTPγS binding for
functional activity performed using the same membrane preparations. MOR and KOR agonist
efficacy expressed as percent stimulation versus standard agonists DAMGO and U50,4888,
respectively. DOR antagonist potency (IC50) and efficacy (Imax) determined versus the standard
agonist SNC80 (100 nM). DOR antagonist efficacy (Imax) calculated as percent of the reference
antagonist naloxone. KOR efficacy in parenthesis is the maximum stimulation produced at 10 µM.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
All values expressed as mean ± SEM for N=3–4 independent experiments. The log D measured
d
e
at pH 7.4. In vitro mouse liver microsomal half-life values in minutes. Data for compound 6
f
g
h
included for comparison. Percent maximum stimulation at 10 µM. ND = Not Determined. NC
Not Converged (no agonist activity detected).
=
Binding affinity of 5′-aryl compounds at DOR. At DOR, the structure of the ligands had significant
effects on the binding affinities. Removal of the chlorine at the para-position on the phenyl ring
in the lead compound 6 yielded compound 8 that displayed nearly 10-fold reduction in affinity.
Replacement with other substituents such as bromine (compound 9), methyl (compound 15),
methoxy (compound 21), dimethylamino (compound 25) or acetylamino (compound 29) all led to
moderate to significant reductions in affinity at DOR. While the 4-fluoro compound 10 maintained
high affinity, the 3,4-difluoro (12) and the 2,4-difluoro (11) compounds displayed ~3- and ~25-
fold reductions in affinity, respectively. Substitution at the 2-position of the phenyl ring in general
appears to result in significant reductions in binding affinity at DOR as exemplified by the data on
the 2-methyl, 2-cyclopropyl, 2,6-dimethyl, 2-methoxy, 2-dimethylamino and 2-acetylamino
compounds 13, 17, 18, 19, 23, and 27, respectively. Among ortho-, meta- and para- substituted
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
isomers, the ortho- methyl compound 13 shows lower affinity (DOR K = 97.2 nM) than the meta-
i
and para- methyl compounds 14 and 15 (DOR K = 26.8 nM and 34.4 nM, respectively). In
i
contrast, the methoxy (19–21) and dimethylamino (23–25) positional isomers did not display
significant differences in their binding affinity at DOR. The acetylamino substituent at the ortho-,
meta-, or para- position (compounds 27, 28, and 29) proved detrimental to binding at DOR.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Binding Affinity, Functional Activity, and In Vitro ADME Data for 5′-Heteroaryl-
1
4-(phenylpropoxy) Compounds 30–42
N
O
R
O
HO
N
a
35
b
Binding K i(nM)
[
S]GTPγS Coupling
MLM
c
compd
R
MOR Agonist
EC50 (nM)
MOR Agonist DOR Antagonist DOR Antagonist KOR Agonist
KOR Agonist
max (%)
log D
d
MOR
14.7 ± 1.5 < 0.169
105 ± 1 6.52 ± 0.28 19.2 ± 4.3 526 ± 83
82.7 ± 2.5 30.6 ± 7.9 27.8 ± 0.8 >3333
47.1 ± 2.1 6.43 ± 1.49 15.8 ± 1.8 56.3 ± 26.5
1.05 ± 0.37 63.5 ± 51.2 <0.51 1.88 ± 0.18
10.7 ± 0.6 3.50 ± 1.14 6.62 ± 2.88 2.69± 0.37
66.2 ± 4.7 22.1 ± 7.8 30.8 ± 8.5 >3333
47.9 ± 5.8 5.18 ± 0.37 24.8 ± 1.3 >3333
DOR
KOR
t ½ (min)
E
max (%)
IC50 (nM)
0.98 ± 0.12
0.51 ± 0.05
0.67 ± 0.03
18.3 ± 0.6
I
max (%)
EC50 (nM)
E
N
e
e
3
3
3
3
3
0
1
2
3
4
13.8 ± 2.6 1.61 ± 0.28
84.7 ± 8.7
37.7 ± 4.6
101 ± 0.3
99.3 ± 0.7
93.3 ± 0.7
91.0 ± 2.0
86.7 ± 0.9
NC
NC
3.30
2.32
3.67
3.53
1.74
7.07
6.25
8.22
11.1
9.60
f
f
N
ND
ND
g
e
e
e
N
(30.2 ± 5.3)
NC
NC
NC
NC
NC
NC
N
e
e
49.3 ± 8.7
N
e
e
81.0 ± 4.0
93.5 ± 3.5
0.39 ± 0.03
CH
3
N
e
3
3
5
6
0.70 ± 0.08
0.71 ± 0.02
0.87 ± 0.14
1.27 ± 0.15
0.62 ± 0.08
19.6 ± 8.7
0.68 ± 0.05
< 0.17
89.3 ± 0.3
92.0 ± 2.1
92.3 ± 0.3
96.7 ± 0.7
92.3 ± 1.2
106 ± 6
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
4.32
3.30
2.00
3.55
0.707
0.943
2.11
2.85
13.4
10.1
26.8
8.72
7.00
21.7
18.5
6.35
OCH
3
N
N
g
e
e
e
e
e
f
e
e
e
e
e
f
(37.3 ± 5.1)
N
g
37
N
(29.3 ± 5.9)
N
g
3
3
4
4
8
9
0
1
N
88.7 ± 1.4 40.1 ± 8.8
121 ± 66 257 ± 211
19.8 ± 0.9 >3333
(40.8 ± 3.8)
N
NH
29.8 ± 21.3 3.29 ± 0.82
115 ± 3
N
N
CH
3
3.79 ± 1.63 7.96 ± 2.22 8.40 ± 5.25 659 ± 484
99.3 ± 17.8 64.5 ± 3.3 23.1 ± 4.0 3.46 ± 1.19
44 ± 2
O
N
74.7 ± 3.4
92.0 ± 3.1
101 ± 3
ND
ND
H
3
C
O
N
CH
3
e
e
42
N
3.18 ± 0.07 8.12 ± 1.97 14.8 ± 4.8 17.5 ± 6.1
70.0 ± 2.5
100
NC
NC
DAMGO
Naloxone
U50,488
25.5 ± 1.4
59.5 ± 2.9 98.1 ± 3.0
153 ± 6
100
31.9 ± 9.0
35.8 ± 2.0
100
a-d
f
e
See corresponding footnotes in Table 1. NC = Not Converged (no agonist activity detected).
g
ND = Not Determined. Percent maximum stimulation at 10 µM.
ACS Paragon Plus Environment
1
6

Page 17 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. Binding Affinity, Functional Activity, and In Vitro ADME Data for 5′-Aryl-14-
alkoxy Compounds 43–52.a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
R1
O
R
O
HO
N
a
35
b
Binding K
i
(nM)
S-GTPγS Coupling
MOR Agonist MOR Agonist DOR Antagonist DOR Antagonist KOR Agonist KOR Agonist log D
MLM
c
compd
R
R
1
d
MOR
DOR
KOR
t ½ (min)
EC50 (nM)
Emax (%)
IC50 (nM)
I
max (%)
EC50 (nM)
Emax (%)
Cl
N
4
4
4
3
4
5
5.37 ± 0.51 5.24 ± 1.65 10.3 ± 5.7 NCe
NC
NC
NC
e
4.69 ± 2.45
16.7 ± 1.2
102 ± 3.1
101 ± 0.6
ND
f
ND
f
1.73
12.2
160
Cl
Cl
O
N
819 ± 96 172 ± 85 21.3 ± 2.3 NCe
e
f
f
25.4 ± 21
10.8 ± 9
ND
ND
2.72
2.79
e
e
f
f
799 ± 131 32.5 ± 12.6 31 ± 4.6
3.96 ± 0.43 5.89 ± 3.70 5.3 ± 1.4
NC
ND
ND
9.40
N
>3333
64 ± 9.8
< 0.51
93 ± 1
96 ± 2
100
e
e
0.0434
0.818
2.38
8.60
8.88
81.1
27.6
6.49
46
NC
NC
N
47
48
49
50
3.82 ± 0.06 29.2 ± 9.2 21.9 ± 5.6 NCe
70.9 ± 3.4 11.0 ± 0.8 60.2 ± 20.3 93.9 ± 9.1
95.3 ± 8.2 12.9 ± 2.1 57.9 ± 3.8 23.6 ± 1.4
NC
e
< 0.51
ND
f
ND
f
F
e
e
60.0 ± 1.5
66.0 ± 4.6
63.8 ± 40.3
28.7 ± 11.9
< 0.17
NC
NC
100
34.5 ± 14.2
9.54 ± 0.19
3.27
O
f
f
5.99 ± 0.53 64.5 ± 28.9 24.8 ± 7.3 9.12 ± 4.93 68.3 ± 1.2
96.3 ± 2.7
ND
ND
3.37
f
f
5
5
1
2
92.1 ± 4.1 14.6 ± 0.2 57.1 ± 12.4 24.8 ± 2.7
24.3 ± 2.6
66.6 ± 16.5
100
ND
ND
2.33
2.60
16.8
6.67
CH
3
f
f
f
f
f
f
499 ± 47 7.48 ± 0.97 252 ± 20
ND
ND
ND
ND
ND
ND
DAMGO
Naloxone
U50,488
25.5 ± 1.4
100
59.5 ± 2.9 98.1 ± 3.0
31.9 ± 9.0
153 ± 6
100
35.8 ± 2.0
100
a-d
e
See corresponding footnotes in Table 1. NC = Not Converged (no agonist activity detected).
fND = Not Determined.
Binding affinity of 5′-aryl compounds at KOR. Most of the 5′-aryl compounds displayed weak
binding affinity at KOR. Exceptions include the 4-fluoro (10), 2-dimethylamino (23), and 4-
dimethylamino (25) compounds that display KOR binding K values of <10 nM. Significant
i
binding affinity at KOR was also displayed by compounds possessing a 3-dimethylamino (24, K
i
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
=
11.9 nM), 2-acetylamino (27, K
= 18.1 nM) group.
i
= 13.5 nM), 4-acetylamino (29, K = 15.1 nM), and 3-OH (22,
i
K
i
Functional activity of 5′-aryl compounds at MOR. The evaluated 5′-arylpyridomorphinans
displayed a partial to full agonist profile with efficacy (Emax) ranging from 37% to 122%. A wide
range in potencies were also displayed by these ligands (EC50 ranging from 1.12 nM to 233 nM).
Surprisingly, the functional potency of the compounds did not always reflect the binding affinity
of the compounds. Except for compounds 10, 17, and 23, all of the phenyl substituted compounds
displayed much stronger functional potency compared to their binding affinity. For example, the
MOR affinity of compounds 4-methyl (15) and 2,6-dimethyl (18) compounds are in the 200 nM
range but are >70-fold more potent as functional full agonists with EC50 < 4 nM. Interestingly,
among the dimethylamino isomers, while the ortho-substituted compound 23 displayed greatly
diminished agonist potency (EC50 = 233 nM) the meta- and para-substituted compounds 24 and
25 emerged as highly potent, full agonists with EC50 values of 1.50 nM and 2.10 nM, respectively.
These findings suggest that these ligands have high intrinsic efficacy at MOR, enabling high
potency functional activity even with modest binding affinity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Functional activity of 5′-aryl compounds at DOR. At DOR, most of these compounds consistently
displayed an antagonist profile of activity with high potency and efficacy (Imax >70%).
Subnanomolar antagonist IC50 were displayed by the unsubstituted (8), 4-bromo (9), 3-methyl (14),
and 2-methoxy (19) compounds. The presence of a 4-fluoro (10), 2- methyl (13), 2,6-dimethyl
(18), or 4-methoxy (21) group tended to diminish DOR antagonist potency. The 2-
(dimethylamino)methyl compound 26 also displayed moderate potency and efficacy as an
antagonist.
ACS Paragon Plus Environment
1
8

Page 19 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Functional activity of 5′-aryl compounds at KOR. Only compounds that emerged as promising
MOR agonist/DOR antagonists were evaluated for potential agonist activity at KOR. The
evaluated compounds were found to be devoid of significant agonist efficacy at KOR. Most of the
compounds did not produce detectable activation. These are designated as non-converged (NC) in
Table 1. The 4-bromo compound 9 and the 4-methoxy compound 21 displayed agonist potencies
of 5.52 nM and 2.02 nM, respectively. These compounds, however, displayed <35% efficacy
indicating a partial agonist profile at KOR. Since these compounds displayed strong to moderate
KOR binding affinity but little agonist activity, they likely act as KOR antagonists. However, this
was not explicitly tested.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Binding affinity of 5′-heteroaryl compounds at MOR. Among 5′-heterocyclic analogues, high
affinity binding (K
i
< 10 nM) was displayed by compounds possessing 2-methyl-4-pyridyl (34),
2
-methoxy-4-pyridyl (35), 1-methyl-4-pyrazolyl (40) and 2-(5-methyl-1,3,4-oxadiazolyl) (42)
substituents. All three pyridyl isomers (31–33) displayed modest to weak binding affinity. The
introduction of a methoxy (35) or methyl (34) group at the 2-position of the 4-pyridyl moiety led
to significant (4-fold and 45-fold, respectively) improvements in binding affinity. Whereas the N-
methyl-4-pyrazolyl compound 40 displayed high affinity (K
i
= 3.79 nM), the N-unsubstituted
pyrazole 39 displayed significantly decreased affinity (K = 121 nM). This >30-fold reduction in
i
binding affinity could be due to the hydrogen-bond donor feature present in the latter.
Binding affinity of 5′-heteroaryl compounds at DOR. Among the 5′-heteroaryl compounds, the 1-
pyrrolyl compound 30 displayed subnanomolar binding affinity at DOR. The 2-pyridyl (31) and
4
-pyridyl (33) isomers displayed higher affinity (DOR K
i
<10 nM) than the 3-pyridyl isomer (32,
DOR K = 30.6 nM). Among the 5-membered heterocyclic compounds, the 1-methyl-4-pyrazolyl
i
compound 40 and the 2-(5-methyl-1,3,4-oxadiazolyl) compound 42 displayed DOR binding
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
affinity of <10 nM. Similar to the profile at MOR, the N-unsubstituted pyrazole 39 displayed
significantly decreased affinity in comparison to the N-methyl congener 34 (K
for 39 vs. 7.96 nM for 40).
i
value of 257 nM
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Binding affinity of 5′-heteroaryl compounds at KOR. Most of the compounds displayed moderate
binding affinity with K values in the 10–30 nM range. The 2-methyl-4-pyridyl compound (34)
i
and the 1-methyl-4-pyrazolyl compound (40) that displayed high affinity at MOR also displayed
high affinity at KOR.
Functional activity of 5′-heteroaryl compounds at MOR. More disparate results were seen in the
agonist functional activity in this series of compounds. The 1-pyrrolyl compound 30 displayed
high potency (EC50 1.61 nM) and efficacy (Emax 84.7%). However, some of the compounds, for
example, compounds 32, and 36–38, displayed very weak potency as agonists. The low potency
of these compounds prevented full evaluation of their efficacy with fully saturated concentration–
response curves. For these compounds, the efficacy listed in parenthesis in Table 2 is the maximum
stimulation produced at 10 µM concentration in the GTPγS assay, which was <50%. Among
isomeric pyridyl compounds only the 4-pyridyl isomer 33 displayed moderate potency and
efficacy. Interestingly, introduction of a methyl group (compound 34) or a methoxy group
(compound 35) at the 2-position of the 4-pyridyl ring gave ligands with high potency (EC50 <3
nM) and efficacy (Emax >80%).
Functional activity of 5′-heteroaryl compounds at DOR. All of the compounds displayed high
antagonist potency (IC50 <1.0 nM – 2.0 nM) and efficacy (Imax >85%) except for the 4-pyridyl
compound 33 and the N-methylpyrazolyl compound 40 that displayed moderate potencies of 18.3
nM and 19.6 nM, respectively.
ACS Paragon Plus Environment
2
0

Page 21 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Functional activity of 5′-heteroaryl compounds at KOR. Similar to the 5′-aryl compounds, all of
the evaluated 5′-heteroaryl compounds displayed non-converged (NC) dose-response curves
indicating their non-agonist profile at KOR. Similar to the 5′-aryl series, these compounds likely
act as KOR antagonists.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Binding affinity and functional activity of 14-alkoxy compounds. Conversion of the pendant phenyl
group in the 14-O-substituent of 6 to a 4-pyridyl group gave a compound 43 that retained high
affinity binding at MOR, DOR, and KOR. However, the functional activity at MOR was drastically
altered to a non-agonist profile. Moreover, the antagonist efficacy at DOR was also greatly
diminished (Imax 16.7%). Exchanging an oxygen atom for the benzylic methylene group of the
propyl chain of 6 led to compound 44 that displayed greatly diminished binding at MOR and DOR.
The drastic reduction of binding affinity at MOR could be attributed to favorable hydrophobic
interactions sustained by the methylene group of the C
6
H
5
CH
2
substituent and unfavorable
O group. The fused benzene
repulsion encountered by the electron rich oxygen atom of the of C
6
H
5
ring of the quinolinylmethyl substituent present in 45 could potentially mimic the phenyl group of
the phenylpropyl substituent of 6. However, this compound while retaining moderate binding
potency at DOR and KOR, lost affinity at MOR. Interestingly, in the functional assay, these three
compounds 43–45 failed to function as agonists at MOR. On the 5′-phenylpyridomorphinan
template, the 14-O-3-(3-pyridylpropyl) compound 46 and the 14-O-3-(4-pyridylpropyl) compound
4
7 displayed moderate to high binding affinity at all three receptors. These two compounds also
lost agonist potency and efficacy at MOR while retaining potent and efficacious antagonist activity
at DOR. The introduction of a fluorine at the para-position on the phenyl ring of 8 gave a
compound 48 that displayed diminished MOR binding and agonist potency. Saturating the pendant
phenyl to a cyclclohexyl group (compound 49) also led to a reduction in binding affinity and
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
agonist activity at MOR. Interestingly, replacement of the cyclohexyl group with a 4-pyranyl group
gave a compound (50) that displayed single digit nanomolar binding and agonist potency but
moderate efficacy at MOR. Despite its weak binding affinity at DOR, in the functional assay, the
compound displayed potent DOR antagonist activity. Since in principle antagonist activity should
rely on intrinsic efficacy, this disparity may reflect allosteric binding for this ligand, producing
strong functional activity but weak orthosteric competition. Extension of the linker chain length
from propyl to butyl (compound 51) also had a deleterious effect on MOR binding and MOR
agonist activity. Removal of the pendant aryl group altogether (compound 52) also resulted in poor
binding affinity at both MOR and KOR. Thus, the nature of the alkoxy substituent has a significant
influence on binding and functional activity, and subtle changes in the structure lead to drastic
changes in binding and activation of MOR.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ADME Properties. In vitro ADME screens were used to profile lead compounds for drug- like
properties and to help prioritize potent compounds for in vivo proof-of-concept studies. Our
strategy focused on profiling compounds for aqueous solubility and lipophilicity (Log D) as these
properties are important predictors of drug bioavailability.34,35 As our ideal lead candidate profile
includes CNS drug exposure, we sought leads with physicochemical properties indicative of good
brain permeability/CNS exposure.36,37 Additionally, we screened compounds for metabolic
stability using mouse and human liver microsomes to help predict drug exposure to rank the lead
compounds. For the majority of compounds screened, aqueous solubility was moderate to high
(Table S1). In an effort to identify leads with good permeability, we prioritized compounds with
Log D in the range of 0–3.38-40 With only a few exceptions, all compounds screened fell within
this range (Tables 1-3).
ACS Paragon Plus Environment
2
2

Page 23 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Most of the compounds screened were rapidly metabolized in mouse liver microsome preparations,
with t values falling less than 30 minutes, with the exception of compounds 44 (t½ = 160 min)
½
possessing an oxygen in place of benzylic methylene group and 48 (t½ = 81 min) possessing a
fluorine at the para position of the phenyl ring of the phenylpropyl group. Oxidative N-
dealkylation of the cyclopropylmethyl group at N17 as well as the metabolic susceptibility of
benzylic methylene and the aryl ring of the 14-O-substituent could potentially be contributing to
the observed poor MLM stability of most of the compounds. Rapid metabolism in MLM suggests
poor exposure in vivo and influenced our decision to dose iv or sc for in vivo studies. Overall,
compounds were significantly more stable in HLM preparations (Table S1). Species differences
in microsomal stability are not uncommon; additional studies are needed to determine the
mechanism of this apparent species difference.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Molecular Modeling. In an effort to gain insights into the SAR, we performed molecular docking
4
1
studies using the X-ray crystal structures of the active state of MOR (PDB ID 5C1M), and
inactive state of DOR and KOR (PDB IDs 4N6H and 4DJH, respectively).42,43 Since the binding
affinities of most of the ligands spanned a relatively narrow range, the differences in the docking
scores in general were subtle. Nevertheless, the type and complementarity of interactions between
the ligand and the receptor residues seen in the docking poses gave useful insights. The general
binding mode of the ligands is illustrated using the docking pose of compound 8 and 46 at MOR
(Figure 2A and 2B, respectively). The epoxymorphinan core binds to the bottom of the orthosteric
pocket defined by residues W2936.48, H2976.52, D1473.32, and M1513.36. The cyclopropylmethyl
group forms hydrophobic contact with W2936.48 while the basic nitrogen N17, which is protonated
at physiological pH, forms a salt bridge interaction with D1473.32. The phenolic hydroxyl forms a
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hydrogen bond with H2976.52 either directly or through a network of two water molecules. These
interactions are shown by all ligands and are conserved in all three receptor subtypes.
The aryl/heteroaryl substituent at the 5′-position on the pyridine ring, on the other hand, plays a
role in influencing binding selectivity at MOR, DOR, and KOR. This is due to the extracellular
region that these substituents occupy, which varies in volume and the nature of the residues among
the three receptor subtypes. As shown in the docking pose of compound 8 at MOR (Figure 2A),
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the phenyl group at the 5′-position forms π-π stacking interaction with W3187.34 and a potential π-
cation interaction with K2335.40 and with the subtype variant K3036.58 residue.41,44,45 Besides
selectivity, such interactions may also influence conformational changes that play a role in the
functional state of the receptor. At DOR, the volume of this extracellular region is even larger than
in MOR and is capable of accommodating a phenyl ring in its favorable orientation. Indeed, in the
docking pose of compound 8 at DOR (Figure 3A), the phenyl group adopts an orientation that
positions it near W2846.58 and K2145.40 which could form π-π stacking and π-cation interaction. In
contrast, when docked to KOR (Figure 3B) in which the volume of this region is smaller than that
of MOR, the phenyl group comes in close unfavorable contact with glutamate E2976.58
(
corresponding to K3036.58 in MOR), thus providing a possible explanation for the 5 to 7-fold
lower affinity at KOR compared to its affinity at MOR and DOR.
The substituent extending from the oxygen at 14-position binds to a hydrophobic region in all three
receptor subtypes. A group such as the phenylpropoxy at the C-14 has the appropriate size and
hydrophobicity to fit into this hydrophobic region and interact with residues such as I1443.29 in
MOR. As discussed, compounds with some of the substituent variations at C-14 did not effectively
stimulate MOR. For example, in contrast to the MOR agonist profile of compound 8, its 3-pyridyl
analogue 46 displayed neither efficacious agonism nor antagonism at MOR. A plausible
ACS Paragon Plus Environment
2
4

Page 25 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
explanation is provided by the docking result with 46, which indicates that the pendant pyridyl
ring of 46 binds in a different orientation than the phenyl ring of 8. Instead of the hydrophobic
contact that the phenyl ring of 8 has with residues such as I1443.29 (Figure 2A), the pyridyl ring
adopts an orientation that forms a polar hydrogen bond interaction with N1272.63 (Figure 2B).
Thus, the new pattern of interaction displayed by 46 in this middle region of TM2/TM3 might
affect the conformational change in a manner that is different from that induced by a typical agonist
of MOR. Moreover, the specific interaction brought about by the pendant group of the substituent
on C-14 might prevent the ligand from going deeper into the pocket, thus, reducing its ability to
interact with residues at the bottom of the orthosteric site that play a key role in the conformational
switch needed for the activation of the receptor.41,46
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Docking poses at MOR (PDB ID 5C1M). (A) Compound 8 and (B) Compound 46.
Compounds are colored green. Hydrogen bond, salt bridge, hydrophobic contact, and π-π stacking
are indicated by black, cyan, yellow, and dark green dashed lines, respectively. All residue
numbers are based on MOR crystal structure 5C1M. Extracellular side is facing top.
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Docking poses of compound 8 at (A) DOR (PDB ID 4N6H) and (B) KOR (PDB ID
4
DJH). Compounds are colored green. Hydrogen bond, salt bridge, hydrophobic contact, π-π
stacking, and π-cation interaction are indicated by black, cyan, yellow, dark green, and purple
dashed lines, respectively. Extracellular side is facing top. W2746.48, H2786.52, D1283.32 in DOR
and W2876.48, H2916.52, and D1383.32 in KOR correspond to conserved residues W2936.48, H2976.52
,
and D1473.32 in MOR.
Antinociceptive Studies in Mice. The antinociceptive potency and efficacy of select compounds
was evaluated in male CD1 mice using the 55°C warm water tail-flick assay. Previous studies with
compound 6 were performed via icv administration. In the current study, we selected 16
compounds that displayed promising in vitro binding and functional activity profiles and evaluated
them following icv or intravenous (iv) administration. Tail withdrawal latencies were assessed at
various time-points for up to five hours post-dose and were compared to vehicle. Due to variability
in aqueous solubility of the test compounds, three different vehicles were used: saline (icv and iv),
100% DMSO (icv), and 10:10:80 DMSO:Tween 80:saline (icv, iv). When 100% DMSO was used
as a vehicle for icv injections, the tail-flick latency was significantly increased over saline or
1
0:10:80 for the first hour of evaluation (p=0.001=0.05); data collected with 100% DMSO was
ACS Paragon Plus Environment
2
6

Page 27 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
excluded based upon solubility limitations and confounding effects on behavior. No statistically
significant differences were seen among vehicles for iv administration (p=0.97), therefore, data
points were combined. The observed antinociceptive profile of the evaluated compounds is
presented in Table 4. Compounds 9, 16, 21, 48, and 50 showed poor antinociceptive activity by
the icv route of administration and were not advanced to systemic evaluation. All other compounds
were evaluated after iv dosing.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 4. Antinociceptive Activity of Selected Compounds in Warm Water Tail-Flick
Assay.a
compd
Route
icv
iv
Dose
MPE ± SEM (n) Time of Effect Onset (min) Time of Peak Effect (min) Duration of Effect (min)
9
10 nmol
10 mg/kg
10 mg/kg
10 nmol
10 mg/kg
10 mg/kg
10 nmol
10 mg/kg
10 mg/kg
10 mg/kg
10 mg/kg
10 mg/kg
10mg/kg
10 nmol
10 mg/kg
10 nmol
75 ± 22 (5)
100 ± 0 (4)
100 ± 0 (5)
65 ± 18 (5)
100 ± 0 (4)
100 ± 0 (5)
86 ± 14 (5)
100 ± 0 (4)
99 ± 0.5 (5)
100 ± 0 (5)
75 ± 14 (5)
100 ± 0 (4)
100 ± 0 (5)
80 ± 13 (5)
100 ± 0 (3)
77 ± 22 (4)
10
15
15
10
15
15
10
15
15
15
15
15
15
15
30
10
30
45
10-45
15-180
15-240
10
14
15
iv
15
16
icv
iv
10
19
15
15-240
15-300
15-300
15-300
15-180
15-120
15-60
20
iv
30-180
10
21
icv
iv
25
15
30
iv
45
34
iv
30
39
iv
15
41
iv
15
15-120
15-180
15-45
42
iv
30
48
icv
iv
15
49
60
30-180
10-60
50
icv
45
aTail-flick latencies were determined as described in the experimental section. Time of effect onset
is listed when the antinociceptive effect was first observed to be >30%. Duration of effect is given
as the total time antinociception remained above 30% maximum possible effect (MPE).
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 117
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Based on an MPE duration of >2 hrs, and lack of respiratory depression, aggression, and related
side effects via qualitative observation, two compounds 20 and 42 were chosen for further
evaluation. These two compounds produced dose-dependent antinociception following iv
administration of 1 mg/kg, 3 mg/kg, and 10 mg/kg doses. Maximal effects were observed
following all three doses of 20 starting 15 min after injections that persisted through 90 min. The
highest dose tested, 10 mg/kg, retained >50% antinociceptive effects that were significantly higher
than vehicle-treated values for four hours after administration (Figure 4A). Administration of 42
dose-dependently mitigated heat evoked nociception beginning 15 min after injection. In contrast
to 20, which displayed a threshold antinociceptive effect (100 ± 0%) for 15-180 min,
the maximal peak effect (80 ± 20%) of 42 occurred at 45 min (Figure 4B). Relative to morphine
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
10 mg/kg, iv),47 compound 20 produced potent, maximal antinociception despite modest
MOR/DOR binding affinity (91 and 48 nM, respectively, Table 1). One potential explanation is
that while 20 had modest binding affinity, it showed a strong in vitro functional potency and
efficacy (MOR agonist EC50 = 13 nM and DOR antagonist IC50 = 2.4 nM, Table 1). This difference
suggests that 20 has strong intrinsic efficacy, whereby modest binding could still produce potent,
efficient receptor activation.48 In addition, blood brain barrier permeability, P-gp efflux, and
favorable pharmacokinetic properties could underlie the potency and efficacy of 20 in producing
antinociception as well as for its ability to produce antinociception by systemic administration
compared with compound 6.
ACS Paragon Plus Environment
2
8

Page 29 of 117
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Antinociceptive dose– and time–response curves for 20 (panel A) and 42 (panel B) in
the 55°C warm water tail-flick assay following iv administration to naïve male CD-1 mice.
Compounds were evaluated in a minimum of two groups with the tester blinded to treatment
(sample sizes per group noted within the graph). Two-way repeated measure (RM) ANOVA,
Bonferroni post-hoc was used for statistical analysis.
Since compound 20 showed a longer duration and maximal antinociception compared to 42 and
morphine (10 mg/kg) following iv administration, additional in vivo characterization was pursued
with compound 20 (designated SRI-39067). We evaluated compound 20 by subcutaneous (sc)
administration of 3.2, 10, 18, and 32 mg/kg doses as compared to morphine. Importantly,
compound 20 produced significant dose-dependent antinociception following subcutaneous
administration (F (36,190) = 4.436, p<0.0001) with a longer duration than morphine49,50 (Figure
ACS Paragon Plus Environment
2
9