Structure-Activity Relationships of (+)-Naltrexone-Inspired Toll-like Receptor 4 (TLR4) Antagonists

Article abstract of DOI:10.1021/acs.jmedchem.5b00426

Activation of Toll-like receptors has been linked to neuropathic pain and opioid dependence. (+)-Naltrexone acts as a Toll-like receptor 4 (TLR4) antagonist and has been shown to reverse neuropathic pain in rat studies. We designed and synthesized compoun

Full text of DOI:10.1021/acs.jmedchem.5b00426

Subscriber access provided by NEW YORK UNIV
Article
Structure-Activity Relationships of (+)-Naltrexone-
Inspired Toll-Like Receptor 4 (TLR4) Antagonists
Brandon R Selfridge, Xiaohui Wang, Yingning Zhang, Hang Yin, Peter
M Grace, Linda R. Watkins, Arthur E. Jacobson, and Kenner C. Rice
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00426 • Publication Date (Web): 26 May 2015
Downloaded from http://pubs.acs.org on May 27, 2015
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.
Journal of Medicinal Chemistry 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 59
Journal of Medicinal Chemistry
1
2
Structure-Activity Relationships of (+)-Naltrexone-Inspired Toll-Like Receptor 4 (TLR4)
Antagonists
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
†
,§
‡,§,◊
‡
◊,
ω
‡
Brandon R. Selfridge, Xiaohui Wang,
Yingning Zhang, Hang Yin, Peter M. Grace,
‡
†
†,*
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
Linda R. Watkins, Arthur E. Jacobson, and Kenner C. Rice
†
Drug Design and Synthesis Section, Molecular Targets and Medications Discovery Branch,
Intramural Research Program, National Institute on Drug Abuse and the National Institute on Alcohol
Abuse and Alcoholism, National Institutes of Health, Department of Health and Human Services,
9
800 Medical Center Drive, Bethesda, MD 20892ꢀ3373, USA.
‡
Department of Psychology and Neuroscience, University of Colorado at Boulder, Boulder, CO
8
0309, United States
◊
Department of Chemistry and Biochemistry the BioFrontiers Institute, University of Colorado
at Boulder, Boulder, CO 80309, United States
ωCenter of Basic Molecular Science and Department of Chemistry, Tsinghua University, Beijing,
China, 100084.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 59
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: Activation of Tollꢀlike receptors has been linked to neuropathic pain and opioid
dependence. (+)ꢀNaltrexone acts as a Tollꢀlike receptor 4 (TLR4) antagonist and has been shown to
reverse neuropathic pain in rat studies. We designed and synthesized compounds based on (+)ꢀ
naltrexone and (+)ꢀnoroxymorphone and evaluated their TLR4 antagonist activities by their effects on
inhibiting lipopolysaccharide (LPS) induced TLR4 downstream nitric oxide (NO) production in
microglia BVꢀ2 cells. Alteration of the Nꢀsubstituent in (+)ꢀnoroxymorphone gave us a potent TLR4
antagonist. The most promising analog, (+)ꢀNꢀphenethylnoroxymorphone ((4S,4aR,7aS,12bR)ꢀ4a,9ꢀ
dihydroxyꢀ3ꢀphenethylꢀ2,3,4,4a,5,6ꢀhexahydroꢀ1Hꢀ4,12ꢀmethanobenzofuro[3,2ꢀe]isoquinolinꢀ
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
7
(7aH)ꢀone, 1j) showed ~ 75 times better TLRꢀ4 antagonist activity than (+)ꢀnaltrexone, and the ratio
of its cell viability IC , a measure of its toxicity, to TLRꢀ4 antagonist activity (140 µM /1.4 µM) was
5
0
among the best of the new analogs. The 1j was active in vivo; it significantly increased and
prolonged morphine analgesia.
INTRODUCTION
1
Tollꢀlike receptor 4 (TLR4) detects pathogenꢀassociated molecular patterns, damageꢀassociated
2
, 3
4ꢀ6
molecular patterns, and xenobioticꢀassociated molecular patterns, triggering signal
1
transduction cascades culminating in the production of proꢀinflammatory mediators. These proꢀ
inflammatory factors exaggerate neuronal excitability, thereby contributing to neuropathic pain
7
ꢀ10
and drug dependence. Because of these actions, TLR4 has become an important nonꢀ
3
neuronal therapeutic target.
We recently identified naltrexone as a TLR4 antagonist by a variety of biophysical, in vitro,
11ꢀ13
in silico, and in vivo assays.
Naltrexone was found to nonꢀstereoselectively reverse
ACS Paragon Plus Environment

Page 3 of 59
Journal of Medicinal Chemistry
1
2
1
4
neuropathic pain in a rat model, which indicates that this action of naltrexone is independent
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
1
5, 16
of opioid receptors. Unlike the pharmacologically active (ꢀ)ꢀisomer, (+)ꢀnaltrexone
((4aR,7aS,12bR)ꢀ3ꢀ(cyclopropylmethyl)ꢀ4a,9ꢀdihydroxyꢀ2,3,4,4a,5,6ꢀhexahydroꢀ1Hꢀ4,12ꢀ
methanobenzofuro[3,2ꢀe]isoquinolinꢀ7(7aH)ꢀone 1a, Figure 1 does not interact with opioid
,
)
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
receptors. Since it cannot act as an opioid antagonist, it does not prevent the beneficial analgesic
effect of opioids. Further studies have shown that (+)ꢀnaltrexone can reduce opioidꢀ and
cocaineꢀinduced conditioned place preference, selfꢀadministration, drugꢀprimed reinstatement
4
, 17, 18
(relapse), and incubation of craving.
However, (+)ꢀnaltrexone itself does not appear to be
the best ligand for treating chronic pain and drug abuse as it has a relatively poor cell viability
to TLR4 antagonist ratio (>400/105.5) and is not longꢀacting in vivo, nor is it particularly potent
in its action. Because of this, we have explored the design and synthesis of new ligands based
on the (+)ꢀnaltrexone molecular template in order to obtain a more promising TLR4 antagonist
with an improved cell viability to TLR4 antagonist ratio. In this initial study we evaluated the
effect of modification or removal of oxygen atoms in (+)ꢀnaltrexone, as well as alteration of the
Nꢀsubstituent. We found that the removal of the furan ring and the ketone from (+)ꢀnaltrexone
afforded an analog that was ~15 times more potent as a TLR4 antagonist. Most gratifying, the
replacement of the Nꢀcyclopropylmethyl substituent in (+)ꢀnaltrexone with an Nꢀphenethyl
moiety led to analog 1j with ~75 fold better TLR4 antagonist activity than (+)ꢀnaltrexone. Nꢀ
Phenethylnoroxymorphone (4S,4aR,7aS,12bR)ꢀ4a,9ꢀdihydroxyꢀ3ꢀphenethylꢀ2,3,4,4a,5,6ꢀ
hexahydroꢀ1Hꢀ4,12ꢀmethanobenzofuro[3,2ꢀe]isoquinolinꢀ7(7aH)ꢀone (1j), also had relatively
low cell cytotoxicity and in vivo efficacy.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 59
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
RESULTS AND DISCUSSION
Several logical disconnections and modifications to the structure of (+)ꢀnaltrexone 1a (Figure 1)
were prepared as structural analogs to establish an initial qualitative SAR. Stereoselective
reduction of (+)ꢀnaltrexone 1a afforded (+)ꢀβꢀnaltrexol 2 and (+)ꢀαꢀnaltrexol 3. The oxide
bridge in (+)ꢀnaltrexone 1a was removed to provide (+)ꢀdesoxynaltrexone ((4bS,8aR)ꢀ11ꢀ
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
(cyclopropylmethyl)ꢀ3,8aꢀdihydroxyꢀ8,8a,9,10ꢀtetrahydroꢀ5Hꢀ9,4bꢀ
(epiminoethano)phenanthrenꢀ6(7H)ꢀone, 4), and this structure was further simplified by
removing the ketone as well, to afford (+)ꢀdidesoxynaltrexone (4bR,8aR)ꢀ11ꢀ
cyclopropylmethyl)ꢀ5,6,7,8,9,10ꢀhexahydroꢀ8aHꢀ9,4bꢀ(epiminoethano)phenanthreneꢀ3,8aꢀdiol,
). Further, we examined the effect of the Nꢀsubstituent in (+)ꢀnaltrexone by removing the
(
5
cyclopropylmethyl group and adding various alkyl (e.g., compound 6, Figure 1) and
phenylalkyl moieties to the secondary amine. Removal of the Cꢀ10 carbon from ring B in (+)ꢀ
naltrexone, a pentacyclic compound, provided the structural analog (+)ꢀ10ꢀnornaltrexone
(4aR,7aS,12bR)ꢀ3ꢀ(cyclopropylmethyl)ꢀ4a,9ꢀdihydroxyꢀ2,3,4,4a,5,6ꢀhexahydroꢀ1Hꢀ
19
benzofuro[3,2ꢀe]isoquinolinꢀ7(7aH)ꢀone, 7), a somewhat less strained tetracyclic compound.
1
9
The synthesis of 7 was described previously.
ACS Paragon Plus Environment

Page 5 of 59
Journal of Medicinal Chemistry
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 1. Synthetic targets 2 – 7
The 12 step synthesis of (+)ꢀnaltrexone 1a (Scheme 1) began with hydrogenation using Pd/C
to saturate the doubleꢀbond in the enol ether present in sinomenine 8, followed by treatment of 9
in a CHCl ꢀ methanesulfonic acid twoꢀphase system to afford (+)ꢀhydrocodone 10. Formation
3
2
0
of the Cꢀ6 methyl enol ether in 11 was conducted using the Rapoport et al. procedure on our
+)ꢀenantiomer with our critical modification of changing the acid from H SO to 5ꢀ
(
2
4
1
5
sulfosalicylic acid for reasons previously described. Bromination of 11 was improved using
20
Nꢀbromosuccinimide (NBS) rather than the reported Nꢀbromoacetamide (NBA) in MeOH to
give 12. Transformation of bromide 12 into one of our needed intermediates, (+)ꢀ
2
1
noroxymorphone 19, was performed in 7 steps following known procedures, with the
modification of 3M HCl instead of H SO in the transformation of Nꢀcyano 18 to (+)ꢀ
2
4
noroxymorphone 19. Finally, Nꢀalkylation with cyclopropylmethyl bromide as we have
1
5
previously described afforded our starting material, (+)ꢀnaltrexone 1a.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 59
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 1. Synthesis of (+)ꢀNaltrexone 1a from Commercially Available Sinomenine 8
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

Page 7 of 59
Journal of Medicinal Chemistry
1
2
Stereospecific reduction of (+)-naltrexone to (+)-naltrexols 2 and 3
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
We initially attempted to increase the potency of (+)ꢀnaltrexone 1a by determining whether
the Cꢀ6 keto moiety was necessary for its action; whether a Cꢀ6 hydroxyl group might be better.
Since it was unpredictable whether that hydroxyl group would be more effective above or
below the plane of the ring, we decided to prepare and test both compounds. The stereoselective
reduction of (+)ꢀnaltrexone 1a to obtain analogs 2 and 3 was performed using procedures that
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
22
were known to provide the corresponding (ꢀ)ꢀisomers. The use of formamidine sulfinic acid in
aqueous sodium hydroxide on (+)ꢀnaltrexone 1a provided (+)ꢀβꢀnaltrexol 2, while reaction with
KH and KꢀSelectride afforded (+)ꢀαꢀnaltrexol 3 (Scheme 2).
Scheme 2. Stereoselective Reduction of (+)ꢀNaltrexone 1a
The effect of the (+)ꢀenantiomers 2 and 3 on TLR4 was examined in cell assays. Microglia are
23
the resident cells of the innate immune system in the central nervous system. TLR4 is
expressed in microglia and its activation is one of the main mechanisms for
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 59
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
2
4
neuroinflammation. We used the microglia BVꢀ2 cell line as the model system of microglia,
2
5
as they reproduce the primary microglia with high fidelity. TLR4 activation induces the
production of the downstream inflammatory factor nitric oxide (NO), which is one of the factors
26
27, 28
contributing to the development of neuropathic pain and drug addiction.
Therefore, NO
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
3
was used as the readout of lipopolysaccharide (LPS) induced TLR4 activation. As shown in
Figure 2, (+)ꢀβꢀnaltrexol 2 and (+)ꢀαꢀnaltrexol 3 each showed reduced TLRꢀ4 antagonism in
comparison to (+)ꢀnaltrexone 1a. (+)ꢀNaltrexone 1a had an IC = 105.5 ± 10.1 µM in that
5
0
assay, whereas (+)ꢀβꢀnaltrexol 2 and (+)ꢀαꢀnaltrexol 3 were less potent with IC ’s = 242.8 ±
50
1
9
2
1.3 and 143.5 ± 20.8 µM, respectively (Table 1). In contrast, (+)ꢀ10ꢀnornaltrexone 7 had an
IC = 76.2 ± 27.2 µM. Each of these compounds showed low cell cytotoxicity, their IC > 400
5
0
50
µM in a cell viability assay. However, the ratio of cytotoxicity to TLR4 antagonism was low
<10) for these analogs, and neither appeared to have great promise.
(
Figure 2. (+)ꢀNaltrexone 1a, (+)ꢀβꢀnaltrexol 2 and (+)ꢀαꢀnaltrexol 3 inhibit LPS induced NO
overꢀproduction in a doseꢀdependent manner. BVꢀ2 cells were treated with 200 ng/mL LPS and
various concentrations of compounds for 24 h. NO in the supernatant was measured by the 2,3ꢀ
ACS Paragon Plus Environment

Page 9 of 59
Journal of Medicinal Chemistry
1
2
6
, 29
diaminonaphthalene method.
as 100%.
The NO in the LPS (200 ng/mL) control group was normalized
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
Table 1. IC Values of (+)ꢀNaltrexone 1a, (+)ꢀβꢀNaltrexol 2, (+)ꢀαꢀNaltrexol 3, and (+)ꢀ10ꢀ
5
0
Nornaltrexone 7
Compound
IC50 of NO
IC50 of
Compound
IC of NO (µM) IC50 of
50
(
µM)
viability
viability
(µM)
(
µM)
1
05.5 ± 10.1 > 400
143.5 ± 20.
> 400
8
2
42.8 ± 21.3 > 400
76.2 ± 27.2
> 400
Removal of Oxygen Functionality
+)ꢀDesoxynaltrexone 4 (Scheme 3) was prepared using the procedures of Nagase et al.,
30, 31
(
in their synthesis of the (ꢀ)ꢀenantiomer of 24 (1a ꢀ 24, Scheme 3). This synthesis began with a
methylation of the Cꢀ3 hydroxy of (+)ꢀnaltrexone 1a to arrive at the 3ꢀmethoxyether 20, which
was subsequently treated with activated zinc and ammonium chloride to open the oxide bridge
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 59
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
and afford the Cꢀ4 hydroxy compound 21 (Scheme 3). The methoxyphenol 21 was converted to
the Cꢀ4 phenyl ether 22. Protection of the Cꢀ6 ketone in 22 as the cyclic ketal 23 followed by
dissolving metal reduction conditions and deprotection of the ketone under acidic conditions
provided (+)ꢀOꢀmethyldesoxynaltrexone 25. Finally, Oꢀdemethylation of 25 with BBr gave
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
(+)ꢀdesoxynaltrexone 4.
Scheme 3. Synthesis of (+)ꢀDesoxynaltrexone 4 from (+)ꢀNaltrexone 1a
As shown in Figure 3, in the NO assay (+)ꢀdesoxynaltrexone 4 had an IC = 118.1 ± 22.5 µM
50
with an IC of cell viability of > 400 µM. (+)ꢀDesoxynaltrexone 4 was a more effective TLR4
5
0
antagonist than (+)ꢀnaltrexone 1a at concentrations above 100 µM; at lower concentrations,
ACS Paragon Plus Environment

Page 11 of 59
Journal of Medicinal Chemistry
1
2
(+)ꢀnaltrexone 1a was somewhat more potent. These assays indicated that (+)ꢀ
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
desoxynaltrexone 4 would not be a better TLR4 antagonist than (+)ꢀnaltrexone 1a.
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. (+)ꢀDesoxynaltrexone 4 inhibits LPS induced NO overꢀproduction in a doseꢀ
dependent manner. (+)ꢀNaltrexone 1a served as a control. BVꢀ2 cells were treated with 200
ng/mL LPS and various concentrations of the test compounds for 24 h. NO in the supernatant
6
, 29
was measured by the 2,3ꢀdiaminonaphthalene method.
control group was normalized to 100%.
The NO in the LPS (200 ng/mL)
To determine if the Cꢀ6 keto moiety in (+)ꢀdesoxynaltrexone 4 had an effect on potency, as
measured by the NO assay, we prepared (+)ꢀdidesoxynaltrexone 5. Synthesis of 5 began with a
WolffꢀKishner reduction of the Oꢀmethyl ether 20 to give alkenyl phenol 26 as described by Wu
32
et al., which was reduced under H atmosphere with PtO to afford 27 (Scheme 4). The same
2
2
synthetic approach utilized in synthesizing 4 (21 ꢀ 25, Scheme 3) was then applied to 27 with
formation of phenyl ether 28 followed by dissolving metal reduction conditions to provide 29.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 59
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
Finally, an Oꢀdemethylation using BBr afforded (+)ꢀdidesoxynaltrexone 5, the Nꢀ
3
cyclopropylmethyl analog of (+)ꢀbutorphanol.
Scheme 4. Synthesis of (+)ꢀDidesoxynaltrexone 5 from (+)ꢀOꢀMethylnaltrexone 20
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
(+)ꢀDidesoxynaltrexone 5 performed better as a TLRꢀ4 antagonist than (+)ꢀnaltrexone 1a
(Figure 4). The increased antagonism difference became more apparent at the 20 µM testing
dose. (+)ꢀDidesoxynaltrexone 5 had an IC value in the NO assay of 48.9 ± 6.8 µM with an
5
0
IC of cell viability = 366 ± 78.8 µM. Removal of the Cꢀ6 keto moiety from (+)ꢀ
5
0
desoxynaltrexone 4 gave a slightly more potent compound. However, although more potent
than both (+)ꢀnaltrexone 1a and (+)ꢀdesoxynaltrexone 4, the (+)ꢀdidesoxynaltrexone 5 was not
sufficiently different from either to warrant further work with the compound.
ACS Paragon Plus Environment

Page 13 of 59
Journal of Medicinal Chemistry
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. (+)ꢀDidesoxynaltrexone 5 inhibits LPS induced NO overꢀproduction in a doseꢀ
dependent manner. (+)ꢀNaltrexone 1a served as the control compound for comparison. BVꢀ2
cells were treated with 200 ng/mL of LPS and various concentrations of test compounds for 24
6
, 29
h. NO in the supernatant was measured by the 2,3ꢀdiaminonaphthalene method.
the LPS (200 ng/mL) group was normalized as 100%.
The NO in
Modification of the N-Substituent
The known strong TLRꢀ4 agonist lipopolysaccharide (LPS) contains a lipid A portion which
has 6 unsubstituted linear alkanes (11 carbons and greater). The crystal structure of LPS bound
to the TLR4/MD2 complex, where MDꢀ2 is the myeloid differentiation factor 2, reported by
3
3
Park et al., shows 5 of these lipid chains interacting with MDꢀ2 while one of them interacts
with TLR4. We thought we would try to mimic the lipid chain structure by introducing a longer
or bulkier alkyl chain on the secondary nitrogen in (+)ꢀnoroxymorphone 19 (Scheme 1), with
the hope that it might have an effect on the potency of these compounds as TLR4 antagonists.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 59
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
To that end, a fourꢀcarbon chain was introduced to obtain (+)ꢀNꢀbutylnoroxymorphone 1b, and
five to eightꢀcarbon chains were also synthesized (Scheme 5). The compound with the longest
chain was (+)ꢀNꢀoctylnoroxymorphone 1f. In addition, to determine the effect of
hydrophobicity and bulk on the alkyl chain, (+)ꢀNꢀ(5ꢀphenyl)pentylnoroxymorphone 1g through
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
(+)ꢀNꢀphenethylnoroxymorphone 1j were prepared. Also, aryl fluoride 1k was synthesized to
see if adding an electronꢀwithdrawing substituent on the arene would provide any added benefit.
Scheme 5. NꢀAlkyl and NꢀArylalkyl Substituents in (+)ꢀNoroxymorphone 1bꢀ1k
The (+)ꢀNꢀalkylꢀ and (+)ꢀNꢀarylalkylꢀnoroxymorphone analogs 1bꢀ1k showed increased
TLR4 antagonism versus LPS in the BV2 microglial cell assay ranging from 4ꢀ88 times greater
than (+)ꢀnaltrexone 1b (Table 2). A representative example from this group of compounds was
that of (+)ꢀNꢀhexylnoroxymorphone 1d (Figure 5). An increase in the length of the carbon
chain from 4 carbons (1b) through 8 carbons (1f) results in an increase in TLRꢀ4 antagonism
ACS Paragon Plus Environment

Page 15 of 59
Journal of Medicinal Chemistry
1
2
versus LPS with 1b showing IC = 24.8 ± 3.4 µM, and 1f with IC = 1.7 ± 0.1 µM (Table 2).
5
0
50
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
This increase in TLRꢀ4 antagonism, however, came at the expense of cell viability where IC50
of viability values > 200 µM and 33.4 ± 0.1 µM for 1b and 1f, respectively. The Nꢀphenylalkyl
containing compounds 1gꢀ1k were generally more potent TLRꢀ4 antagonists than the Nꢀalkyl
chain compounds 1bꢀ1f (e.g., 1j, Figure 6). The same trend in cell cytotoxicity showing a direct
relationship to chain length was observed in this series as well. Although the TLRꢀ4
antagonism remained roughly the same (~ 1.4 µM) for compounds 1gꢀ1j, the ratio of cell
toxicity to TLR4 antagonism degenerated as the arylakyl chain length increased. Also, the
addition of fluorine to the aromatic ring on the 2ꢀposition (1k) resulted in an increase in the cell
cytotoxicity by approximately twoꢀfold.
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 5. (+)ꢀNꢀHexylnoroxymorphone 1d inhibits LPS induced NO overꢀproduction in a
doseꢀdependent manner. (+)ꢀNaltrexone 1a served as the control compound for comparison.
BVꢀ2 cells were treated with 200 ng/mL LPS and various concentrations of test compounds for
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 59
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
6
, 29
2
4 h. NO in the supernatant was measured by the 2,3ꢀdiaminonaphthalene method.
The NO
in the LPS (200 ng/mL) control group was normalized to 100%.
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 6. (+)ꢀNꢀPhenethylnoroxymorphone 1j inhibits LPS induced NO overꢀproduction in a
doseꢀdependent manner. (+)ꢀNaltrexone 1a served as the control compound for comparison.
BVꢀ2 cells were treated with 200 ng/mL LPS and various concentrations of test compounds for
6
, 29
24 h. NO in the supernatant was measured by the 2,3ꢀdiaminonaphthalene method.
in the LPS (200 ng/mL) group was normalized to 100%.
The NO
ACS Paragon Plus Environment

Page 17 of 59
Journal of Medicinal Chemistry
1
2
Table 2. IC Values of (+)ꢀNꢀAlkylnoroxymorphone Analogs 1bꢀ1k.
5
0
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
IC of IC of
IC of IC of
50
50
50
50
NO
µM)
viability (µM)
NO
viability (µM)
(
(µM)
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
2
4.8 ±
> 200
1.7 ±
0.4
26.8 ± 0.4
37.1 ± 1.9
69.3 ± 3.4
140.8 ± 5.5
65.7 ± 3.3
3.4
7.5 ±
> 200
1.2 ±
0.2
1
.7
6
0
.7 ±
.4
121.4 ± 7.3
68.5 ± 2.8
33.4 ± 1.0
1.4 ±
0.2
3
0
.0 ±
.2
1.4 ±
0.3
1
0
.7 ±
.1
1.8 ±
0.2
Combination of Oxygen Removal and N-Substitution
We incorporated in our designed ligand a combination of the benefits of an extended alkyl
chain (e.g., as in (+)ꢀNꢀhexylnoroxymorphone 1d) with the benefits obtained by removal of the
oxide bridge and ketone from (+)ꢀnaltrexone (that also increased TLRꢀ4 antagonism, as in (+)ꢀ
didesoxynaltrexone 5), with the hope of obtaining a compound with amplified antagonist
potency. Thus, the phenol of (+)ꢀNꢀhexylnoroxymorphone 1d was methylated (Scheme 6) and
34
the resultant diether 30 was subjected to modified WolffꢀKishner reduction conditions to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 59
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
afford monoether 31. Attempts to prepare a phenyl ether from the phenolic hydroxyl in 31 like
that of 27 ꢀ 28 (Scheme 4) were unsuccessful. Instead, the phosphonate ester 32 (Scheme 6)
was synthesized and subsequently treated under dissolving metal reduction conditions to afford
the didesoxy compound 33. An Oꢀdemethylation of 33 followed by hydrogenation of the
alkene with Pd/C provided (+)ꢀdidesoxyꢀNꢀhexylnoroxymorphone ((4bR,8aR)ꢀ11ꢀhexylꢀ
5,6,7,8,9,10ꢀhexahydroꢀ8aHꢀ9,4bꢀ(epiminoethano)phenanthreneꢀ3,8aꢀdiol, 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 6. Synthesis of (+)ꢀDidesoxyꢀNꢀhexylnoroxymorphone 6 from (+)ꢀNꢀ
Hexylnoroxymorphone 1d
ACS Paragon Plus Environment

Page 19 of 59
Journal of Medicinal Chemistry
1
2
3
4
(
+)ꢀDidesoxyꢀNꢀhexylnoroxymorphone 6 did have slightly increased TLRꢀ4 antagonism
activity (Figure 7) in comparison to both (+)ꢀNꢀhexylnoroxymorphone 1d and (+)ꢀNꢀ
didesoxynaltrexone 5, with an IC = 5.2 ± 1.3 µM. However, the increased activity is within
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
5
0
the experimental error of the activity of (+)ꢀNꢀhexylnoroxymorphone 1d itself at 6.7 ± 0.4 µM.
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
In addition, the IC of viability assay for (+)ꢀdidesoxyꢀNꢀhexylnoroxymorphone 6 was found to
50
be 73.4 ± 4.0 µM, while that of (+)ꢀNꢀhexylnoroxymorphone 1d was 121.4 ± 7.3 µM. We also
examined the Nꢀhexyl compound 34 in the BVꢀ2 microglial cell assay, and it was more potent,
with an IC = 22.5 ± 3.2 µM, and less toxic with an IC of viability = 154.7 ± 5.0 µM,
50
50
although the ratio (~7) was not as good as desired.
Figure 7. (+)ꢀDidesoxyꢀNꢀhexylnoroxymorphone 6 inhibits LPS induced NO overꢀproduction
in a doseꢀdependent manner. (+)ꢀNaltrexone 1a served as the control compound for comparison.
BVꢀ2 cells were treated with 200 ng/mL LPS and various concentrations of test compounds for
6
, 29
2
4 h. NO in the supernatant was measured by the 2,3ꢀdiaminonaphthalene method.
The NO
in the LPS (200 ng/mL) group was normalized to 100%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 59
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
Comparison of Phenol Versus Methyl Ether
All the compounds synthesized and tested up to this point contained a phenol as found in
classic agonist and antagonist opiates such as oxymorphone, morphine, and naltrexone. In
order to determine if the phenol functionality was necessary for TLR4 activity or whether it had
an effect on toxicity, methyl ether 35 ((4aR,7aS,12bR)ꢀ4aꢀhydroxyꢀ9ꢀmethoxyꢀ3ꢀphenethylꢀ
2,3,4,4a,5,6ꢀhexahydroꢀ1Hꢀ4,12ꢀmethanobenzofuro[3,2ꢀe]isoquinolinꢀ7(7aH)ꢀone) was
synthesized (Scheme 7) for comparison with the phenol 1j in the BVꢀ2 microglia cell assay.
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 TLR4 antagonist activity of methyl ether 35 was found to be ~4 fold less (IC = 5.7 ± 1.3
50
µM) than that of phenol 1j (IC = 1.4 ± 1.3 µM). In addition, the cell viability assay showed
5
0
greater cell cytotoxicity for 35 (IC of viability = 74.7 ± 3.5 µM) than 1j (IC of viability =
5
0
50
140.8 ± 5.5 µM).
Scheme 7. Synthesis of (+)ꢀNꢀPhenethylnoroxycodone 35
Finally, the in vivo efficacy of (+)ꢀNꢀphenethylnoroxymorphone (1j) was tested on the
3
5
Hargreaves thermal hyperalgesia assay. Potentiation of morphine analgesia was tested as a
clinically relevant outcome, since morphine signaling at TLR4 opposes analgesia in a proꢀ
5, 36
inflammatory fashion.
Here, intrathecal morphine (15 ꢀg plus vehicle; open circle) produced
significant analgesia to radiant heat, compared to baseline (Fig. 8; P < 0.05). Coadministration
of 1j (60 ꢀg, intrathecal; closed circle) significantly increased and prolonged intrathecal
ACS Paragon Plus Environment

Page 21 of 59
Journal of Medicinal Chemistry
1
2
morphine analgesia (Figure 8; Drug x Time: F18,180 = 0.83, P = 0.6; Drug: F1,10 = 16.7, P < 0.01;
Time: F18,180 = 3.73, P < 0.001).
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
Morphine + vehicle
Morphine + 1j
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
100
*
** *
*
*
*
*
50
#
0
-50
0
20 40 60 80 100 120 140 160 180
Time post injection (mins)
Figure 8. Intrathecal coadministration of morphine (15 ꢀg) with 1j (60 ꢀg) produced a significant
#
potentiation of morphine tailflick analgesia. n = 6/group *P < 0.05, **P < 0.01: relative to vehicle; P
<
0.05: relative to baseline.
CONCLUSION
Stereoselective reduction of the ketone of (+)ꢀnaltrexone 1a to the (+)ꢀnaltrexols 2 and 3
resulted in a slight loss of TLR4 antagonist activity, while removal of the oxide bridge and
ketone to give (+)ꢀdidesoxynaltrexone 5 afforded an ~15 fold increase in activity relative to (+)ꢀ
naltrexone 1a. Removal of the bridging benzylic Cꢀ10 carbon of (+)ꢀnaltrexone resulted in an
analog, (+)ꢀ10ꢀnornaltrexone 7, with only a slight improvement in TLR4 antagonism.
Interestingly, the replacement of the Nꢀcyclopropylmethyl substituent in (+)ꢀnaltrexone with a
linear alkyl chain (4ꢀ8 carbons) resulted in greatly increased activity (~4ꢀ62 fold). There was,
however, increased cell cytotoxicity, and a poorer toxicity/activity ratio, associated with longer
chain length. The Nꢀphenylalkyl analogs were found to be among the most potent TLRꢀ4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 59
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
antagonists (up to 88 fold compared to (+)ꢀnaltrexone). NꢀPhenethyl through Nꢀphenylpentyl
analogs were found to have similar TLR4 antagonism activity (IC ≈ 1.4 µM), while the cell
5
0
cytotoxicity increased with chain length. Finally, it was found that a methyl ether on the
aromatic ring, as in 35, decreases TLR4 antagonism and increases toxicity compared with the
corresponding phenolic compound. The Nꢀphenethylꢀsubstituted noroxymorphone (1j) was
among the most potent compounds found and had the best ratio (~ 100) of TLR4 potent
antagonist activity (IC = 1.4 ± 0.3 ꢀM), and cell cytotoxicity (IC of viability = 140.8 ± 5.5).
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
50
This analog 1j also showed in vivo efficacy in potentiating morphine analgesia.
EXPERIMENTAL
Cell based characterization.
6
NO assay. NO assays were performed as described previously. BVꢀ2 microglia cells were
grown in DMEM medium supplemented with 10% FBS. BVꢀ2 cells were detached from the
flask by trypsin digestion when ~80% confluence was reached. Cells were seeded at a density of
4
4
×10 cells/well in 96ꢀwell plates. After overnight incubation, the medium was aspirated and
changed to DMEM (without FBS). LPS (200 ng/mL) and indicated concentration of compounds
were added. Following an additional 24 h treatment, 100 µL of media was removed and added
to flat black 96ꢀwell microfluor plates (Thermo Scientific, MA, USA). Subsequently, 10 µL of
2,3ꢀdiaminonaphthalene (0.05 mg/mL in 0.62 M HCl) was added to each well and incubated for
1
5 min. The reaction was quenched by addition of 5 µL of 3 M NaOH and the plate was read on
a Beckman Coulter DTX880 reader (Fullerton, CA, USA) with excitation at 360 nm and
emission at 430 nm. Assay of NO in BVꢀ2 microglia cells, rather than assay of HEK293 TLR4
overꢀexpressing cells, was chosen as the HEK293 TLR4 cells can only report activation of
ACS Paragon Plus Environment

Page 23 of 59
Journal of Medicinal Chemistry
1
2
NFkappaB signaling pathways so do not appropriately represent the complexity of TLR4
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
3
7, 38
signaling.
Cell viability assay. Crystal violet staining was used to determine cell viability as described
5, 6
previously. After drug treatment, cells were fixed by 3.7% paraformaldehyde (PFA) for 5
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
min. and then stained by 0.05% crystal violet for 30 min. The plates were subsequently washed
2 times with tap water and dried for 30 min at room temperature. 200 µL of ethanol was added
to each well and the plates were shaken for 15 min at room temperature to dissolve the dye.
Absorbance at 540 nm was measured with a Beckman Coulter DTX880 reader (Fullerton, CA,
USA)
Data analysis. Origin 7.5 (OriginLab Corporation, Northampton, MA, USA) was used for
plotting of the data and statistical analysis. Nonꢀlinear logistic regression was used to plot and
analyze concentrationꢀresponse curves and to obtain IC values. Each experiment was
5
0
independently repeated at least 3 times with 3ꢀ4 biological replicates
In vivo characterization.
Animals. Pathogenꢀfree, adult male Sprague Dawley rats were used (300ꢀ325 g on arrival;
Harlan Labs, Madison, WI), and housed in temperature (23 ± 3 °C) and light (12 h:12 h
light:dark cycle; lights on at 0700) controlled rooms with standard rodent chow and water
available ad libitum. All procedures were approved by the Institutional Animal Care and Use
Committee of the University of Colorado Boulder.
Intrathecal catheter implantation and drug delivery. The method of acute intrathecal drug
administration and the construction and implantation of the indwelling intrathecal catheters was
3
9
based on work described previously. In brief, the day prior to experimentation, intrathecal
operations were conducted under isoflurane anesthesia by threading sterile polyethyleneꢀ10
tubing (PEꢀ10 Intramedic Tubing; Becton Dickinson Primary Care Diagnostics, Sparks, MD,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 59
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
USA) guided by an 18ꢀgauge needle between the L5 and L6 vertebrae. The catheter was
inserted such that the proximal catheter tip lay over the lumbosacral enlargement. The needle
was removed and the catheter was sutured to the superficial musculature of the lower back. The
catheters were preꢀloaded with 60 ꢀg of 1j in 1.5% DMSO in 0.9% saline + 15 ꢀg morphine in
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
0.9% saline or vehicle (1.5% DMSO in 0.9% saline + 15 ꢀg morphine in 0.9% saline) at the
distal end in a total volume of 10 ꢀL. The catheters were 90 cm in length, allowing remote drug
delivery without touching or otherwise disturbing the rats during the testing. On the day of
experimentation, drugs were delivered over 20ꢀ30 s.
Hargreaves test for analgesia. Rats received at least three 60 min habituations over
successive days to the test environment prior to behavioral testing. Latencies for behavioral
response to radiant heat stimuli applied to the tail were assessed using a modified Hargreaves
3
5
test. All testing was conducted blind with respect to group assignment. Briefly, baseline
withdrawal values were calculated from an average of three consecutive withdrawal latencies of
the tail, measured at 10 min intervals. A short baseline latency (2ꢀ3 s) was used to allow
quantification of analgesia (lengthening of the latency, relative to baseline, in response to
analgesia). A cutꢀoff time of 10 s was imposed to avoid tissue damage. Nociceptive assessments
for acute administration experiments were then made at 0 (immediately following remote drug
delivery) and every 10 min thereafter for 180 mins.
Statistics and data analysis. The analgesic responses were calculated, as is standard in the
pain field, as the % of maximal possible effect (%MPE) using the following equation %MPE =
test latency ꢀ baseline latency/ cutoff – baseline latency x 100. Repeated measures twoꢀway
ANOVA with a Holm Sidak posthoc test was used to determine group differences. Repeated
measures oneꢀway ANOVA with a Dunnett posthoc test was used to determine changes from
baseline. Data are presented as mean ± SEM, and significance was set at P < 0.05.
ACS Paragon Plus Environment

Page 25 of 59
Journal of Medicinal Chemistry
1
2
Chemistry
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
General Methods. Melting points were determined on a Buchi Bꢀ545 instrument and are
1
13
uncorrected. Proton and carbon nuclear magnetic resonance ( H and C NMR) spectra were
recorded on a Varian Geminiꢀ400 spectrometer in CDCl (unless otherwise noted) with the
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
1
values given in ppm (TMS as internal standard) and J (Hz) assignments of H resonance
coupling. Mass spectra (HRMS) were recorded on a VG 7070E spectrometer or a JEOL
SX102a mass spectrometer. The optical rotation data were obtained on a PerkinElmer
polarimeter model 341. Thin layer chromatography (TLC) analyses were carried out on
Analtech silica gel GHLF 0.25 mm plates using various gradients of CHCl /MeOH containing
3
1
% NH OH or gradients of EtOAc:nꢀhexane. Visualization was accomplished under UV light
4
or by staining in an iodine chamber. Flash column chromatography was performed with Fluka
silica gel 60 (mesh 220ꢀ400). Atlantic Microlabs, Inc., Norcross, GA, or MicroꢀAnalysis, Inc.,
Wilmington, DE, performed elemental analyses, and the results were within ±0.4% of the
theoretical values.
(4S,4aR,7aS,12bR)-3-Butyl-4a,9-dihydroxy-2,3,4,4a,5,6-hexahydro-1H-4,12-
methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1b). To a stirred solution of (+)ꢀNꢀnoroxymorphone
dihydrate 19 (0.500 g, 1.54 mmol) in DMF (10 mL) in a screw cap vial was added NaHCO (0.502 g,
3
5
.98 mmol) and 1ꢀbromobutane (0.365 mL, 3.39 mmol). The vial was capped (the cap was teflonꢀlined)
and heated at 90 °C for 3 h. The reaction mixture was cooled to room temperature, filtered through
Celite, and concentrated in vacuo. To the crude residue was added H O (10 mL), concentrated aq
2
NH OH (2 mL) and CHCl (20 mL). The organics were separated and the aqueous layer was extracted
4
3
with CHCl (20 mL). The combined organics were washed with H O (5 x 20 mL), dried over Na SO ,
3
2
2
4
filtered through Celite, and concentrated in vacuo to afford (+)ꢀNꢀbutylnoroxymorphone 1b as a crude
white solid. Purification of this solid by SiO column chromatography with 10% NH OH in
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 59
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
MeOH/CHCl (gradient, 0 ꢀ4%) afforded 1b (0.151 g, 29%) as a white solid. All spectra reported on
3
2
0
ꢀ1 1
the free base. [α] = + 179.5 (1.1, CHCl ); mp 150ꢀ151 °C; IR (thin film) 3184 cm ; H NMR (400
D
3
MHz, CDCl ) δ 6.69 (d, J = 8.0 Hz, 1H), 6.54 (d, J = 8.4 Hz, 1H), 6.08 (br s, 2H), 4.73 (s, 1H), 3.08ꢀ
3
2
.94 (m, 3H), 2.57ꢀ2.38 (m, 5H), 2.26 (m, 1H), 2.13 (td, J = 12.0, 3.6 Hz, 1H), 1.86 (m, 1H), 1.61 (td, J
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
3
= 14.0, 3.2 Hz, 1H), 1.52 (m, 1H), 1.48ꢀ1.29 (m, 4H), 0.90 (t, J = 7.2 Hz, 3H); C NMR (100 MHz,
CDCl ) δ 209.8, 143.5, 138.8, 128.9, 123.9, 119.6, 117.9, 90.3, 70.5, 62.6, 53.9, 50.8, 43.4, 36.0, 31.1,
3
+
+
3
3
0.5, 29.6, 22.7, 20.3, 13.9; HRMS (TOF MS ES ) calcd for C H NO (M + H ) 344.1856, found
20 26 4
44.1857. An HCl salt was prepared by dissolving 1b free base in hot iꢀPrOH (1.0 mL) followed by the
addition of concentrated aq HCl (0.10 mL, 3 equiv) and cooling to 5 °C. The crystals were filtered,
rinsed with cold iꢀPrOH, and airꢀdried to give 1b as its HClꢁ2H O salt. Anal. Calcd for C H ClNO
6
2
20 30
(
1bꢁHClꢁ2H O): CHN.
2
(4S,4aR,7aS,12bR)-4a,9-Dihydroxy-3-pentyl-2,3,4,4a,5,6-hexahydro-1H-4,12-
methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1c). The same procedure used for 1b was
also used for 1c, with 19 (0.500 g, 1.54 mmol), pentyl bromide (0.421 mL, 3.39 mmol),
NaHCO (0.502 g, 5.98 mmol), and DMF (10 mL). Purification of the resulting solid by SiO
3
2
column chromatography with 10% NH OH in MeOH/CHCl (gradient, 0 ꢀ4%) afforded Nꢀ
4
3
pentylnoroxymorphone 1c (0.506 g, 92%) as a white solid. All spectra reported on the free
2
0
ꢀ1 1
base. [α] = + 173.1 (1.2, CHCl ); mp 168ꢀ169 °C; IR (thin film) 3181 cm ; H NMR (400
D
3
MHz, CDCl ) δ 6.71 (d, J = 8.4 Hz, 1H), 6.58 (d, J = 8.0 Hz, 1H), 4.70 (s, 1H), 3.11ꢀ2.88 (m,
3
3H), 2.59ꢀ2.28 (m, 5H), 2.15 (td, J = 12.0, 4.0 Hz, 1H), 1.87 (m, 1H), 1.67ꢀ1.45 (m, 3H), 1.37ꢀ
1
3
1
1
.27 (m, 4H), 0.91 (t, J = 7.0 Hz, 3H); C NMR (100 MHz, CDCl ) δ 209.7, 143.5, 138.8,
3
29.0, 124.0, 119.7, 117.8, 90.4, 70.5, 62.7, 54.2, 50.9, 43.4, 36.1, 31.2, 30.5, 29.4, 27.2, 22.7,
+
+
22.5, 14.0; HRMS (TOF MS ES ) calcd for C H NO (M + H ) 358.2013, found 358.2013.
2
1
28
4
An HCl salt was prepared by dissolving 1c free base in hot iꢀPrOH (3.5 mL) followed by
ACS Paragon Plus Environment

Page 27 of 59
Journal of Medicinal Chemistry
1
2
concentrated HCl (0.34 mL, 3 equiv) and cooling to 5 °C. The crystals were filtered, rinsed
with cold iꢀPrOH, and airꢀdried to give 1c as the HCl salt. Anal. Calcd for C H ClNO
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
21 28
4
(1cꢁHCl): CHN.
(4S,4aR,7aS,12bR)-3-Hexyl-4a,9-dihydroxy-2,3,4,4a,5,6-hexahydro-1H-4,12-
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
methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1d). The same procedure used for 1b was
also used for 1d, with 19 (3.00 g, 9.25 mmol), 1ꢀbromohexane (2.90 mL, 20.4 mmol), NaHCO3
(3.1 g, 37 mmol), and DMF (60 mL) to afford (+)ꢀNꢀhexylnoroxymorphone 1d (3.0 g, 87%) as
an offꢀwhite solid that was used without further purification. All spectra reported on the free
20
ꢀ1 1
base. [α] = + 187.3 (1.1, CHCl ); mp 160ꢀ162 °C; IR (thin film) 3190 cm ; H NMR (400
D
3
MHz, CDCl ) δ 6.70 (d, J = 8.4 Hz, 1H), 6.55 (d, J = 8.0 Hz, 1H), 6.24 (br s, 1H), 4.75 (s, 1H),
3
3
1
.09ꢀ2.95 (m, 3H), 2.58ꢀ2.39 (m, 5H), 2.29 (d, J = 14.8 Hz, 1H), 2.13 (t, J = 11.6, 3.2 Hz, 1H),
.86 (m, 1H), 1.62 (td, J = 14.0, 3.2 Hz, 1H), 1.53 (m, 1H), 1.48ꢀ1.40 (m, 2H), 1.39ꢀ1.21 (m,
1
3
6H), 0.87 (t, J = 6.4 Hz, 3H); C NMR (100 MHz, CDCl ) δ 209.7, 143.5, 138.8, 128.9, 123.9,
3
1
1
19.6, 117.8, 90.3, 70.5, 62.6, 54.2, 50.8, 43.3, 36.0, 31.6, 31.1, 30.5, 27.4, 26.8, 22.7, 22.5,
+
+
4.0; HRMS (TOF MS ES ) calcd for C H NO (M + H ) 372.2169, found 372.2169. An HCl
22 30
4
salt was prepared by dissolving 1d free base in hot iꢀPrOH (21 mL) followed by concentrated
HCl (1.9 mL, 3 equiv) and cooling to 5 °C to give 1d as the HCl salt. Anal. Calcd for
C H ClNO (1dꢁHCl): CHN.
22
30
4
(4S,4aR,7aS,12bR)-3-Heptyl-4a,9-dihydroxy-2,3,4,4a,5,6-hexahydro-1H-4,12-
methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1e). The same procedure used for 1b was
also used for 1e, with 19 (0.500 g, 1.54 mmol), 1ꢀbromoheptane (0.533 mL, 3.39 mmol),
NaHCO (0.502 g, 5.98 mmol), and DMF (10 mL). Purification of this solid by SiO column
3
2
chromatography with 10% NH OH in MeOH/CHCl (gradient, 0 ꢀ4%) afforded 1e (0.319 g)
4
3
that still contained some DMF. To the impure 1e was added iꢀPrOH (1.9 mL) followed by
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 59
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
concentrated aq HCl (0.20 mL). The suspension was cooled to 5 °C, filtered, rinsed with cold iꢀ
PrOH (2 x 2 mL), and airꢀdried to afford pure (+)ꢀNꢀheptylnoroxymorphone 1e (0.291 g, 45%)
2
0
as a white HCl•0.3H O salt. All spectra reported on the free base. [α] = + 174.7 (1.2,
2
D
ꢀ1 1
CHCl ); mp 143ꢀ145 °C; IR (thin film) 3170 cm ; H NMR (400 MHz, CDCl ) δ 6.70 (d, J =
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
3
8
2
.0 Hz, 1H), 6.54 (d, J = 8.0 Hz, 1H), 6.03 (br s, 2H), 4.73 (s, 1H), 3.08ꢀ2.94 (m, 3H), 2.57ꢀ
.38 (M, 4H), 2.28 (d, J = 14.4 Hz, 1H), 2.13 (m, 1H), 1.65ꢀ1.44 (m, 4H), 1.40ꢀ1.20 (m, 6H),
1
3
0.86 (t, J = 6.4 Hz, 3H); C NMR (100 MHz, CDCl ) δ 209.8, 143.5, 138.8, 128.9, 124.0,
3
1
2
19.7, 117.9, 90.4, 70.5, 62.6, 54.2, 50.9, 43.4, 36.0, 31.7, 31.2, 30.5, 29.1, 27.5, 27.2, 22.7,
+
+
2.5, 14.0; HRMS (TOF MS ES ) calcd for C H NO (M + H ) 386.2326, found 386.2327.
23 32
4
Anal. Calcd for C H ClNO (1e•HCl•0.3H O): CHN.
2
3
32.6
4.3
2
(4S,4aR,7aS,12bR)-4a,9-Dihydroxy-3-octyl-2,3,4,4a,5,6-hexahydro-1H-4,12-
methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1f). The same procedure used for 1b was
also used for 1f, with 19 (0.500 g, 1.54 mmol), 1ꢀbromooctane (0.589 mL, 3.39 mmol),
NaHCO (0.502 g, 5.98 mmol), and DMF (10 mL). Purification of this solid by SiO column
3
2
chromatography with 10% NH OH in MeOH/CHCl (gradient, 0 ꢀ4%) afforded 1f (0.414 g)
4
3
still containing some DMF. To the impure 1f was added iꢀPrOH (2.5 mL) followed by
concentrated aqueous HCl (0.25 mL). The suspension was cooled to 5 °C, filtered, rinsed with
cold iꢀPrOH (2 x 2 mL), and airꢀdried to afford pure (+)ꢀNꢀoctylnoroxymorphone 1f (0.391 g,
2
0
5
8%) as a white HCl•0.15H O salt. All spectra reported on the free base. [α] = + 156.8 (1.2,
2
D
ꢀ1 1
CHCl ); mp 139ꢀ141 °C; IR (thin film) 3184 cm ; H NMR (400 MHz, CDCl ) δ 6.71 (d, J =
3
3
8
.4 Hz, 1H), 6.57 (d, J = 8.0 Hz, 1H), 4.71 (s, 1H), 3.10ꢀ2.94 (m, 3H), 2.59ꢀ2.37 (m, 4H), 2.27
(m, 1H), 2.15 (td, J = 12.0, 3.6 Hz, 1H), 1.87 (m, 1H), 1.66ꢀ1.44 (m, 3H), 1.30ꢀ1.25 (m, 8H),
13
0
.88 (t, J = 6.6 Hz, 3H); C NMR (100 MHz, CDCl ) δ 209.8, 143.5, 138.8, 128.9, 123.9,
3
119.6, 117.9, 90.3, 70.5, 62.6, 54.2, 50.8, 43.3, 36.0, 31.7, 31.2, 30.5, 29.4, 29.1, 27.5, 27.2,
ACS Paragon Plus Environment

Page 29 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
+
+
2
2.7, 22.5, 14.0; HRMS (TOF MS ES ) calcd for C H NO (M + H ) 400.2482, found
24
34
4
4
00.2481. Anal. Calcd for C H34.3ClNO4.15 (1f•HCl•0.15H O): CHN.
24 2
6
7
8
(4S,4aR,7aS,12bR)-4a,9-Dihydroxy-3-(5-phenylpentyl)-2,3,4,4a,5,6-hexahydro-1H-
,12-methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1g). The same procedure used for 1b
was also used for 1g, with 19 (0.400 g, 1.24 mmol), 1ꢀbromoꢀ5ꢀphenylpentane (0.504 mL, 2.73
mmol), NaHCO (0.402 g, 4.79 mmol), and DMF (8 mL). Purification of this solid by SiO
9
4
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
3
2
column chromatography with 10% NH OH in MeOH/CHCl (gradient, 0 ꢀ4%) afforded (+)ꢀNꢀ
4
3
phenylpentylnoroxymorphone 1g (0.500 g, 89% corrected for DMF) as an offꢀwhite solid. All
2
0
spectra reported on the free base unless otherwise noted. [α] = + 144.3 (1.2, MeOH, HCl
D
ꢀ1 1
salt); mp of HCl salt dec > 276 °C; IR (thin film) 3286, 1723 cm ; H NMR (400 MHz, CDCl )
3
δ 7.29ꢀ7.25 (m, 2H), 7.19ꢀ7.16 (m, 3H), 6.73 (d, J = 8.4 Hz, 1H), 6.57 (d, J = 8.0 Hz, 1H), 6.01
(br s, 2H), 4.70 (s, 1H), 3.09ꢀ2.99 (m, 3H), 2.62ꢀ2.38 (m, 7H), 2.29 (dt, J = 14.0, 2.8 Hz, 1H),
1
3
2.16 (td, 12.0, 3.6 Hz, 1H), 1.88 (m, 1H), 1.67ꢀ1.49 (m, 6H), 1.41ꢀ1.35 (m, 2H); C NMR (100
MHz, CDCl ) δ 210.0, 143.6, 142.5, 139.1, 128.9, 128.4, 128.3, 125.7, 123.8, 119.8, 118.1,
3
90.3, 70.5, 62.6, 54.2, 50.9, 43.6, 36.6, 36.1, 35.8, 31.3, 30.5, 27.3, 26.9, 22.9; HRMS (TOF MS
+
+
ES ) calcd for C H NO (M + H ) 434.23258, found 434.23233. An HCl salt was prepared by
2
7
32
4
dissolving 1g free base in hot iꢀPrOH (3.5 mL) followed by concentrated aq HCl (0.28 mL, 3
equiv) and cooling to 5 °C. The crystals were filtered, rinsed with cold iꢀPrOH, and airꢀdried to
give the HCl salt. Anal. Calcd for C H ClNO (1g•HCl): CHN.
2
7
32
4
(4S,4aR,7aS,12bR)-4a,9-Dihydroxy-3-(4-phenylbutyl)-2,3,4,4a,5,6-hexahydro-1H-
,12-methanobenzofuro[3,2-e]isoquinolin-7(7aH)-one (1h). The same procedure used for 1b
4
was also used for 1h, with 19 (0.400 g, 1.24 mmol), 1ꢀbromoꢀ4ꢀphenylbutane (0.479 mL, 2.73
mmol), NaHCO (0.402 g, 4.79 mmol), and DMF (8 mL). Purification of this solid by SiO
3
2
column chromatography with 10% NH OH in MeOH/CHCl (gradient, 0 ꢀ4%) afforded (+)ꢀNꢀ
4
3
ACS Paragon Plus Environment