4-Aryl-1-oxa-4,9-diazaspiro[5.5]undecane Derivatives as Dual μ-Opioid Receptor Agonists and σ1 Receptor Antagonists for the Treatment of Pain

Article abstract of DOI:10.1021/acs.jmedchem.9b01256

The synthesis and pharmacological activity of a new series of 1-oxa-4,9-diazaspiro[5.5]undecane derivatives as potent dual ligands for the sigma-1 receptor (σ₁R) and the μ-opioid receptor (MOR) are reported. The different positions of the central scaffold, designed using a merging strategy of both target pharmacophores, were explored using a versatile synthetic approach. Phenethyl derivatives in position 9, substituted pyridyl moieties in position 4 and small alkyl groups in position 2 provided the best profiles. One of the best compounds, 15au, showed a balanced dual profile (i.e., MOR agonism and sigma antagonism) and a potent analgesic activity, comparable to the MOR agonist oxycodone in the paw pressure test in mice. Contrary to oxycodone, as expected from the addition of σ₁R antagonism, 15au showed local, peripheral activity in this test, which was reversed by the σ₁R agonist PRE-084. At equianalgesic doses, 15au showed less constipation than oxycodone, providing evidence that dual MOR agonism and σ₁R antagonism may be a useful strategy for obtaining potent and safer analgesics.

Full text of DOI:10.1021/acs.jmedchem.9b01256

Subscriber access provided by BUPMC - Bibliothèque Universitaire Pierre et Marie Curie
Article
4-Aryl-1-oxa-4,9-diazaspiro[5.5]undecane derivatives as dual µ-opioid
receptor agonists and #1 receptor antagonists for the treatment of pain.
Mónica García, Marina Virgili, Mònica Alonso, Carles Alegret, Begoña Fernández, Adriana Port, Rosalia
Pascual, Xavier Monroy, Alba Vidal-Torres, María-Teresa Serafini, José Miguel Vela, and Carmen Almansa
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01256 • Publication Date (Web): 19 Nov 2019
Downloaded from pubs.acs.org on November 19, 2019
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 68
Journal of Medicinal Chemistry
4
-Aryl-1-oxa-4,9-diazaspiro[5.5]undecane
1
2
3
4
5
6
7
8
9
derivatives as dual µ-opioid receptor agonists and
σ1 receptor antagonists for the treatment of pain.
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
1
2
2
2
1
1
Mónica García, Marina Virgili, Mònica Alonso, Carles Alegret, Begoña Fernández,
1
1
1
1
Adriana Port, Rosalía Pascual, Xavier Monroy, Alba Vidal-Torres, María-Teresa Serafini,
1
1*
José Miguel Vela, and Carmen Almansa.
1ESTEVE Pharmaceuticals S.A., Drug Discovery and Preclinical Development, Carrer Baldiri
Reixac, 4-8. Parc Científic de Barcelona, 08028 Barcelona, Spain.
2Enantia, S.L., Carrer Baldiri Reixac, 10. Parc Científic de Barcelona, 08028 Barcelona, Spain.
KEYWORDS. µ-opioid receptor, sigma-1 receptor, dual ligands, multimodal analgesia, pain.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 68
ABSTRACT
The synthesis and pharmacological activity of
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
new series of 1-oxa-4,9-
diazaspiro[5.5]undecane derivatives as potent dual ligands for the sigma-1 receptor (σ R) and
1
the µ-opioid receptor (MOR) are reported. The different positions of the central scaffold,
designed using a merging strategy of both target pharmacophores, were explored using a
versatile synthetic approach. Phenethyl derivatives in position 9, substituted pyridyl moieties in
position 4 and small alkyl groups in position 2 provided the best profiles. One of the best
compounds, 15au, showed a balanced dual profile (i.e. MOR agonism and sigma antagonism)
and a potent analgesic activity, comparable to the MOR agonist oxycodone in the paw pressure
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
test in mice. Contrary to oxycodone, as expected from the addition of σ R antagonism, 15au
1
showed local, peripheral activity in this test, which was reversed by the σ R agonist PRE-084.
1
At equianalgesic doses, 15au showed less constipation than oxycodone, providing evidence
that dual MOR agonism and σ R antagonism may be a useful strategy for obtaining potent and
1
safer analgesics.
ACS Paragon Plus Environment
2

Page 3 of 68
Journal of Medicinal Chemistry
Introduction
1
2
3
4
5
6
7
8
9
The adequate management of pain constitutes an important challenge, since currently available
treatments provide in many cases only modest improvements, leaving many patients
1
2
unrelieved and leading to important productivity losses and socio-economical burden.
Existing pain therapies include non-steroidal anti-inflammatory drugs (NSAIDs), opioid
agonists, calcium channel modulators and antidepressants, but all of them show limited
efficacy and a range of secondary effects that can preclude their use, especially in chronic
settings.
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
Opioids are among the most effective therapies when addressing severe pain states. They act
through three different types of opioid receptors (µ, κ and δ) which are transmembrane G-
protein coupled receptors (GPCRs), although the main analgesic action is attributed to the
activation of the µ-opioid receptor (MOR). However, the use of MOR agonists is limited by
their important side effects, such as constipation, respiratory depression, emesis, dependence3
4
and tolerance. As a consequence, long-term treatment with MOR agonists usually require
substantial increases in dosing in order to maintain a clinically satisfactory pain relief, which
finally results in either unacceptable side effects, leading to poor patient compliance, or
5
alternatively in addiction, as revealed by the recent opioid crisis affecting the USA. Three of
the most widely used MOR agonists, morphine (1), oxycodone (2) and fentanyl (3) are
depicted in Figure 1.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 68
HO
O
O
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
O
O
N
HO
H
MeO
OH
N
N
N
1
, Morphine
2, Oxycodone
3, Fentanyl
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
F
O
O
N
N N
O
N
HO
4
, Haloperidol
5, E-52862
Cl
Figure 1. Structures of representative MOR agonists (1-3) and σ R antagonists (4, 5).
1
Several approaches have attempted to improve the benefit-risk balance of opioids (i.e., to
enhance or maintain their analgesic efficacy without increasing or, ideally, reducing their
6
adverse effects). These include developing allosteric MOR ligands, targeting different opioid
7
8
receptor subtypes or MOR receptor isoforms and reducing central nervous system (CNS)
9
penetration. More recently, the development of biased ligands has attracted much attention
and the G-protein biased (cAMP versus β-arrestin pathway) MOR agonist (GPB-MOR)
1
0
11
oliceridine was developed by Trevena on the basis of its improved side effects profile.
However, currently prevailing evidence suggests that GPB-MOR will retain opioid-like abuse
potential12 and the US Food and Drug Administration recently declined to approve
oliceridine.13
An alternative approach is the development of multimodal compounds exhibiting activity at
both MOR and additional relevant target/s in the same molecule. This was the strategy behind
the development by Grünenthal of cebranopadol,14 a highly potent dual MOR and
nociceptin/orphanin FQ peptide ligand that reached phase 3 clinical trials, although progression
to the market has not been reported.
ACS Paragon Plus Environment
4

Page 5 of 68
Journal of Medicinal Chemistry
As a new approach to obtain potent but safer opioids, we devised dual compounds targeting
1
2
3
4
5
6
7
8
9
both the MOR and the sigma-1 receptor (σ R). The goal was to enhance the opioid analgesia
1
component by adding unique analgesic properties provided by the σ R antagonist component,
1
while reducing side effects. Accordingly, consistent with a role of the σ R in chronic pain
1
sensitizing conditions15 and modulation of opioid-induced efficacy- and safety-related
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
16
outcomes, we reasoned that a dual MOR agonist and σ R antagonist could provide analgesics
1
with i) similar efficacy to opioids, ii) activity in chronic pain conditions with no/reduced
pharmacodynamic tolerance, iii) activity in pain of neuropathic origin, where opioids are
poorly effective, and iv) a better safety profile. This strategy also addressed the intrinsic
multimodal nature of pain, which is often treated in the clinic by combination of available
1
7
drugs. Combination of two mechanisms of action in a single molecule is a major challenge,
but as compared with drug cocktails or multi-component drugs, multiple (dual) ligands offer
the potential advantages of a better efficacy, superior treatment compliance (one instead of
various pills at potentially different dosage intervals/frequency), lowering the risk of drug-drug
interactions, simpler pharmacokinetics and less variability, both in drug exposure and response
to treatment, among patients.18
The σ R gene encodes a 24-kDa protein of 223 amino acids anchored to the endoplasmic
1
reticulum (ER) and plasma membranes.19 The σ R sequence has no homology with other
1
mammalian proteins and is structurally and functionally different from other target classes,
since it acts as a molecular chaperone that interacts with diverse proteins to modulate their
2
0
functions. The first link of the σ R to analgesia was established by Chien and Pasternak, who
1
described it as an endogenous anti-opioid system, based on the finding that σ R agonists
1
counteracted opioid receptor-mediated analgesia, while σ R antagonists, such as haloperidol
1
(
4), potentiated it. Many additional preclinical studies support a role of the σ R in the treatment
1
1
5
21
of pain. The development of the σ R knockout mice (σ R-KO), which shows no obvious
1
1
phenotype and perceives normally sensory stimuli, was a key milestone in this endeavour. In
physiological conditions the responses of the σ R-KO to mechanical and thermal stimuli were
1
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 68
undistinguishable from wild type mice, but they do not develop attenuated pain behaviours
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
when pain pathways become sensitized following capsaicin or formalin injection or in
neuropathic pain conditions after nerve injury (i.e., sciatic nerve ligation) or secondary to
chemotherapy (i.e., paclitaxel-induced neuropathy). Many of these actions were confirmed by
2
2
the use of σ R antagonists and led to the advancement of one compound, S1RA (E-52862, 5),
1
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
23
into clinical trials for the treatment of different pain states. Interestingly, compound 5, when
administered in combination with morphine, potentiated its analgesia in models of nociceptive
pain, but not its adverse effects, and reduced its addictive/rewarding effects as revealed in the
1
6
conditioned place preference test paradigm. Potentiation of opioid analgesia was reproduced
combining other σ antagonists and different -opioid reference agonists, and has been
1
evidenced in several animal pain models, including nociceptive, visceral, inflammatory and
neuropathic pain.24,25 Clinical data support preclinical findings, as administration of compound
5
produced a clinically relevant reduction of opioid-associated adverse events (nausea,
vomiting, and dizziness) and a meaningful reduction for the need of concomitant antiemetic
medication, meaning that an enhancement of the benefit-risk of the opioid (morphine) overall
26
clinical profile was obtained. All these preclinical and clinical results support the potential
therapeutic value of a dual MOR agonist/σ R antagonist compound for the treatment of pain.
1
The design of dual drugs is a complex process, since achieving good affinity at an appropriate
ratio, the right agonistic/antagonistic pharmacological activity at each molecular target and
selectivity for only two targets, while avoiding many others, is a significant challenge. An
additional issue is getting the good ADME and physicochemical properties required for in vivo
activity as the room for optimization is reduced by the need of keeping both activities.27
Combining in the same molecule a MOR agonist and σ R antagonist profile adds to this
1
complexity also because there is no high throughput in vitro test for unambiguously
establishing the functionality of σ R ligands, often functionally designated only after in vivo
1
testing.28
ACS Paragon Plus Environment
6

Page 7 of 68
Journal of Medicinal Chemistry
Several approaches have been used for the design of dual compounds.18,29 The conceptually
1
2
3
4
5
6
7
8
9
most simple is the mutual prodrug approach, where two active molecules are linked through a
labile bond that cleaves in vivo to give the two unaltered compounds, which then act
3
0
individually on their respective targets. A different approach is the design of bivalent ligands
(or conjugates), where two pharmacophores are linked via a long spacer of variable chemical
nature and are designed to span across homo- or heterodimeric GPCRs and target two
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
1
individual receptors simultaneously. Representative examples have been developed in the
opioid field by the Portoghese group.32,33 This strategy usually provides high in vitro affinities
but compounds not able to achieve oral activity and CNS penetration due to their high
molecular weight and flexibility. Shortening the linker up to a point when the pharmacophores
are touching provides the so called “fused” dual compounds. However, the best scenario in
relation to obtaining the desired drug-like properties, is provided by “merging” the two
pharmacophores by taking advantage of commonalities in their structures, as exemplified by
the aforementioned cebranopadol.14
As a first exploration to a new multimodal approach to potent and safer analgesics, we report
here the synthesis and structure-activity relationships (SAR) studies that were designed using a
merging approach and led to the identification of a new series of 4-aryl-1-oxa-4,9-
3
4
diazaspiro[5.5]undecane derivatives 15 (Scheme 1) with dual affinity towards the MOR and
the σ R.
1
Chemistry
The synthesis of the compounds described herein was accomplished as depicted in Schemes 1
to 5 and indicated in Tables 1 to 6. Starting from N-Boc-piperidone 6, a highly versatile route
was devised to prepare derivatives 15, following a similar strategy previously described from
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 68
3
5
36
N-Cbz-piperidone (Scheme 1). Epoxide 7 was prepared using Corey-Chaykovsky reagent,
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
3
7
followed by thermal ring opening with arylamines 8 leading to aminoalcohols 9 in good
yields. Acylation with an adequate acyl halide, 10, and final intramolecular cyclization by
treatment with potassium tert-butoxide at low temperatures rendered spirocyclic compounds
12. Finally, Boc-deprotection followed by N-alkylation provides the target compounds 15 in
good yields. In particular, this method was used for the preparation of compounds 15ah-aj and
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
5an-aq using the appropriate reagents, aniline 8-1, 2-chloropropionylchloride 10-1 and the
corresponding alkylating agents 14.
Scheme 1a
R2 R2'
R '
X
OH
O
R2
N
2
NH2
X
O
R1
OH
X
O
O
8
H
10
N
R1
R1
N
N
N
N
Boc
a
Boc
b
Boc
c
Boc
6
7
9
11
X
R R '
R R '
R R '
2
2
2
2
2
R3
2
O
O
O
1
4
O
O
O
N
N
N
d
R1
e
R1
f
R1
N
NH
N
Boc
R3
12
13
15
O
O
X
O
Ph
8
NH2
-1
Cl
O
R3
O
Cl 10-1
N
14
N
7
NH
N
d, e
R3
b
c
f
1
1
5ah-aj
13-1
5an-aq
aReagents and conditions: (a) TMSOI, NaH, DMSO, rt to 50 ºC, 1 h; (b) EtOH-water 9:1, 100
t
ºC, 16 h; (c) TEA, DCM, 0 ºC, 2 h; (d) BuOK, THF, -78 to -30 ºC, 1-4 h; (e) TFA, DCM, rt or
reflux; (f) K CO , NaI, acetonitrile, 80 ºC, 16 h.
2
3
ACS Paragon Plus Environment
8

Page 9 of 68
Journal of Medicinal Chemistry
Although Scheme 1 is of general application, some variations were used in particular cases. In
1
2
3
4
5
6
7
8
9
the case of simple R groups, such as benzyl and phenethyl, the N-substituted piperidones (16)
3
were commercially available and were used to prepare compounds 15 bearing the substituent
R right from the start. This allowed diversification in substituents R -R (Scheme 2), by
3
1
2
building the spirobicyclic core following a similar strategy to that described in Scheme 1. 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
related spiro compound 20 was prepared from aminoalcohol 18-1 (where R is H and n is 1)
1
upon treatment with carbonyldiimidazole (CDI).38
In an analogous way to that described in Scheme 2, the pyrrolidinyl derivative 21 (Table 1) was
prepared starting from N-benzylpyrrolidin-3-one with final exchange of R group to install the
3
phenethyl chain present in compound 21 by conventional N-debenzylation followed by N-
alkylation, as described in the experimental part.
Scheme 2a
O
n=0-1
n
N
16
a
R2 R2'
X
R '
2
R '
1
NH2
O
R2
X
OH
O
OH
O
8
X
10
H
R '
1
R '
1
N
N
n
b
n
c
n
N
N
N
17
18
19
d
e
R R '
O
2
2
O
O
O
N
R '
1
n=0-1
n
N
N
N
20
15a-f
1
1
5h-i
5l,15r,15t
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 68
aReagents and conditions: (a) TMSOI, NaH, DMSO, rt to 50 ºC, 1 h; (b) EtOH-water 9:1, 100
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
t
ºC, 16 h; (c) TEA, DCM, 0 ºC, 2 h; (d) BuOK, THF, -78 to -30 ºC, 1-4 h; (e) CDI, THF,
reflux, 16 h.
Alternatively, the aryl or heteroaryl group in R was installed in the last step of the sequence
1
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 3). Epoxides 17 were treated with methanolic ammonia to render aminoalcohols 22.
Subsequent acylation and cyclization were conducted under modified reaction conditions
regarding those shown in Schemes 1 and 2, to minimize formation of by-products derived from
the amide NH present in compounds 24. Thus, acylation was best performed under a biphasic
system (ethyl acetate-water) and final cyclization to spirocyclic compounds 26 was carried out
at higher temperatures and with inverse addition (acyl compound 24 was added over a solution
of tert-butoxide). Finally, arylation with aryl halides 28 under Ullmann or Buchwald-Hartwig
coupling conditions rendered target compounds 15n-p, 15v-z and 15aa-ag. Compounds 15ar-
1
5at (Table 6) were synthesized following a similar approach, but starting from N-Boc
protected epoxide 7. It was transformed to key intermediate 27, that was either converted into
6 or alternatively it was first arylated to render key precursor 12 and then, substituent R was
2
3
incorporated as described in Scheme 1.
ACS Paragon Plus Environment
1
0

Page 11 of 68
Journal of Medicinal Chemistry
Scheme 3a
1
2
3
4
5
6
7
8
9
O
Cl
OH
Cl
0-1
O
OH
O
NH3
a
1
Cl
H N
2
HN
N
N
b
N
Z
Z
Z
1
7 Z=R3
Z=Boc
22 Z=R3
23 Z=Boc
24 Z=R3
25 Z=Boc
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
R X
1
O
O
28
O
O
N
HN
d or e
c
R1
N
N
Z
Z
2
2
6 Z=R3
7 Z=Boc
15 Z=R3
12 Z=Boc
15n-p
15v-z
f, g
f, g
15aa-15ag
15ar-15at (from 27)
a
Reagents and conditions: (a) NH MeOH, rt, 16 h; (b) K CO , EtOAc-water, 0 ºC, 30 min; (c)
3
,
2
3
t
BuOK, THF, rt, 16 h; (d) CuI, K PO , trans-1,2-cyclohexanediamine, 1,4-dioxane, 110 ºC, 16
3
4
h; (e) For compounds 15ab, 15as, 15ar and 15at: Pd (dba) , XantPhos, Cs CO , 1,4-dioxane,
2
3
2
3
1
10 ºC, 16 h; (f) TFA, DCM, rt, 3 h; (g) K CO , NaI, acetonitrile, 80 ºC, 16 h.
2 3
Compounds bearing a spirocyclopropyl group in position 2 were synthesized (Scheme 4) using
-methoxybenzyl (PMB) as N-protecting group. The key step was the acylation of
4
3
9
aminoalcohol 30 with a conveniently functionalized acid chloride 31, followed by treatment
with excess of potassium tert-butoxide, which provided two concomitant cyclizations to yield
the spirocyclopropyl compound 33 in 48% yield (2 steps). Boc deprotection and introduction of
the phenethyl chain was performed as described in Scheme 1 to render 36. Finally, PMB group
removal using cerium ammonium nitrate (CAN) and arylation under Ullmann or Buchwald-
Hartwig coupling conditions provided compounds 15m and 15au.
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 68
Scheme 4a
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
NH2
Cl
Cl
Br
31
Br
O
O
29
OH
O
HO
H
OMe
Cl
b
N
N
PMB
PMB
N
N
N
Boc
a
Boc
Boc
c
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
30
32
Br
R '
O
O
3
O
35
O
O
O
N
N
N
PMB
d
PMB
e
PMB
N
NH
N
Boc
R '
3
33
34
36
CF3
Br
8-2
O
O
N
O
CF
O
3
2
HN
N
F
N
N
h
N
f
g
R '=2-F
R '
3
3
7
15au
3
I
2
8-1 R '=H
3
O
O
N
N
15m
aReagents and conditions: (a) EtOH-water 9:1, 100 ºC, 16 h; (b) TEA, DCM, 0 ºC, 2 h; (c)
t
BuOK, THF, 0 ºC, 2 h; (d) TFA, DCM, rt, 3 h; (e) K CO , NaI, acetonitrile, 80 ºC, 16 h; (f)
2
3
CAN, acetonitrile-water, rt, 16 h; (g) Pd (dba) , XantPhos, Cs CO , 1,4-dioxane, 110 ºC, 16 h;
2
3
2
3
(h) CuI, K PO , trans-1,2-cyclohexanediamine, 1,4-dioxane, 110 ºC, 16 h.
3
4
Additionally, some compounds were obtained by functional group transformations over final
derivatives. Compound 15g was obtained by debenzylation of 15h, while hydrolysis of 15i
afforded the acid 15j, which was coupled with dimethylamine to provide 15k (Table 2).
ACS Paragon Plus Environment
1
2

Page 13 of 68
Journal of Medicinal Chemistry
Compounds 15q, 15s and 15u, (Table 3), were obtained by demethylation of their methoxy
1
2
3
4
5
6
7
8
9
precursors with boron tribromide, as described for 15q in the experimental part. Compounds
1
5ak, 15al and 15am (Table 5, Scheme 5) were obtained by acetylation of the corresponding
anilines 40, that in turn were prepared by reduction of their nitrophenyl precursors, 39,
synthesized using the procedure describe in Scheme 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
Scheme 5a
O
Br
O
O
O
NO2
N
38
N
NH
N
a
NO2
39
13-1
O
O
O
O
N
N
N
N
15ak-15am
b
c
NH2
NHAc
40
a
Reagents and conditions: (a) K CO , NaI, acetonitrile, 80 ºC, 16 h; (b) Pd/C, MeOH, 3 bar H ,
2
3
2
rt, 16 h; (c) AcCl, TEA, DCM, rt, 16 h.
Enantiopure compounds 15 were obtained by chiral preparative HPLC separation of the
corresponding racemic mixtures, as indicated in the supplementary information. The absolute
configuration of the eutomers was assigned as (R) as established by chiral synthesis of
compounds 15cR and 15asR. Starting from (S)-2-chloropropionyl chloride (10-1S, 95% ee)
and following the procedure described in Scheme 2, with modified acylation conditions, the
corresponding (S)-acyl derivative 19-1S was obtained. This was cyclized assuming inversion
of configuration to give compound 15cR with 91% ee, as determined by chiral HPLC. In turn,
compound 15asR was prepared from 10-1S using the procedure described in Scheme 3.
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 68
However, it was only obtained with a 67% ee as determined by chiral HPLC, due to partial
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
epimerization in the metal catalyzed arylation step under the standard coupling conditions.
Results and discussion
For the design of the dual MOR and σ R ligands we envisaged a merging approach, i.e.,
1
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
combining the two target pharmacophores in the same central core. As at the time we initiated
the project there were no crystal structures available of any of the two receptors, the design was
based on pharmacophore models generated from known available ligands for each of them.
The MOR pharmacophore was generated with the ‘Auto Pharmacophore Generation’ protocol
4
0
within Discovery Studio, using morphine (1) as input ligand, chosen because of its structural
rigidity. The protocol enumerates all pharmacophores compatible with the input molecule
considering the following features: Hydrogen Bond Acceptor (HBA), Hydrogen Bond Donor
(HBD), Hydrophobic (HYD), Negative Ionisable (NI), Positive Ionisable (PI) and Aromatic
ring (AR). It then chooses those pharmacophores with the highest selectivity as predicted by a
Genetic Function Approximation built from a training set of 500 pharmacophore models. Of
the resulting selected pharmacophores, we prioritized the one having two HBA features (HBA1
and HBA2). The HBD generated on account of the two hydroxyl groups of morphine were
discarded since other MOR agonists lacked these groups. As several opioid agonists, i.e.
fentanyl (3), contain a HYD at the other site of the basic amine, a HYD was added (Figure 2A),
not as a requirement but as a potentially tolerated feature. In the case of the σ R a new model
1
4
1
42
was built using Langer pharmacophore as starting point. Based on our in house SAR data,
one of its four hydrophobic features, HYD1, was omitted as it was not deemed necessary for
σ1R binding, and HYD3 and HYD4 were merged into one single HYD (Figure 2B).
With these results in hand a merging approach seemed indeed feasible: both pharmacophores
require a basic amine, i.e. a PI, and the AR of MOR may as well match a HYD present in the
σ R pharmacophore. Accordingly, both σ R and MOR models were aligned by the PI, the
1
1
HYD and the HYD-AR pair. Their superposition is shown in Figure 2C. Based on this model
ACS Paragon Plus Environment
1
4

Page 15 of 68
Journal of Medicinal Chemistry
several bicyclic scaffolds were proposed (15a,b, 20 and 21, Table 1) and the new series of
1
2
3
4
5
6
7
8
9
dually active 4-aryl-1-oxa-4,9-diazaspiro[5.5]undecane derivatives 15 was identified.
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. (A) MOR pharmacophore with aligned morphine (1); (B) σ R pharmacophore with
1
aligned E-52862 (5); (C) superposition of both models.
3
The compounds synthesized were evaluated in a primary human σ R binding assay using [ H]-
1
2
2
3
43
(
+)-pentazocine as radioligand and in a human MOR binding assay using [ H]-DAMGO as
radioligand. The MOR functional assay was carried out by measuring cyclic AMP (cAMP) on
4
4
CHO-K1 cells providing EC as well as Emax data (Tables 1-6). Generally speaking, a full
5
0
agonist profile is suggested for compounds active at the MOR with Emax above 80%.
4
4
However, a cell line overexpressing the MOR with a high receptor reserve is used in this
assay, and thus no unequivocal determination of the functional nature (i.e., full vs partial
agonism) can be established.
The active compounds were also evaluated at 1 µM in a selectivity panel constituted by the
guinea-pig sigma-2 receptor (σ R) and the following human protein targets: 5-HT receptor,
2
1A
5
-HT2B receptor, α A adrenoreceptor (α AR), α A adrenoreceptor, dopamine transporter, δ-
1 1 2
opioid receptor, κ-opioid receptor, histamine H receptor, norepinephrine transporter and
1
serotonin transporter. This panel includes targets validated for pain and others associated with
undesirable side effects, in order to rule out off target activities that could interfere with the
analgesic response of the compounds. In addition, we assayed from the beginning the blockade
of the human Ether-a-go-go-Related Gene (hERG), the well-known potassium channel linked
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 68
to cardiac toxicity,45 since hERG inhibition was a recurrent issue in our previous σ R
1
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
2
programs. The results obtained in a patch-clamp assay are shown in Tables 1-6.
The initial compounds prepared contained the naked central scaffold and a benzyl (15a) or
phenethyl (15b) tail (Table 1). Both of them showed an interesting σ R affinity, but 15b was
1
superior regarding MOR agonism. Two modifications of the central scaffold, the 1-oxa-3,8-
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
diazaspiro[4.5]decan-2-one derivative 20 and the 6-oxa-2,9-diazaspiro[4.5]decan-8-one
derivative 21 were also synthesized. Both of them showed good σ R affinity but were poorly
1
active at MOR. This could be explained by a worse compliance of 20 and 21 with the MOR
pharmacophore, since the ether oxygen in both scaffolds was not able to match HBA2 (Figure
3
). On these bases, we decided to focus our SAR study on compound 15b, which also showed
affinity for two additional receptors of the selectivity panel (inhibitions at 1µM of 85 % for the
α AR and 69% for the 5-HT receptor). Since the α AR activity was a constant throughout the
1
2B
1
series its binding results (% at 1 µM) are also reported in Tables 2-6.
Figure 3. Fitting of compounds 15b (A), 20 (B) and 21 (C) into the overlapped MOR and σ R
1
pharmacophores. 20 and 21 do not match HBA2, as shown by the un-highlighted green
spheres, automatically set by the Discovery Studio Ligand-Pharmacophore Mapping protocol.
Exploration of substitution in position 2 of the central scaffold (R , Table 2) showed that
2
methylation afforded a substantial increase in affinity on both primary targets. However,
compound 15c retained activity for the α AR and the hERG channel. Hence, efforts in this
1
program were mainly directed to reduce these off-target activities. Interestingly, while the σ R
1
affinity of the two enantiomers of 15c was similar, the R enantiomer (15cR) showed an
ACS Paragon Plus Environment
1
6

Page 17 of 68
Journal of Medicinal Chemistry
improvement on MOR agonism of one order of magnitude over the S counterpart. This
1
2
3
4
5
6
7
8
9
enantiospecific increase in MOR affinity obtained with alkyl substitution in position 2 might be
due to a favourable lipophilic interaction with the receptor near the HBA2 feature, not covered
by the MOR pharmacophore model. A four-fold difference in hERG inhibition also favoured
15cR, while α AR affinity remained similar. Elongation of the alkyl chain to ethyl (15d) or
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
isopropyl (15e) increased MOR activity, but the higher affinity for MOR versus the σ R was
1
not considered a desirable balance. The desired ratio of in vitro affinities for the MOR and σ R
1
receptors is complex to establish and in fact this is one of the main challenges in the design of
this particular class of dual compounds. However, a predominant affinity for MOR would not
make evident the enhancement of the opioid analgesia provided by the σ R antagonist
1
component. Note that enhancement of opioid analgesia in combination studies was obtained
using lower doses of opioids than of σ R ligands (example morphine at 2.5-5 mg/kg and E-
1
5
2862 at 40 mg/kg) and that maximal enhancement occurs when opioids are combined at low
1
6
doses (around ED if administered alone).
2
5
Introduction of a polar group, such as methoxy, provided a more balanced derivative (15f).
However, its clogP decrease did not afford a significant improvement in hERG inhibition, and
the same was observed for the hydroxymethyl derivative 15g. Benzylation of this group (15h)
was tolerated by both targets, but it was clearly detrimental regarding hERG inhibition,
probably due to the its increased lipophilicity. Introduction of carboxyl derivatives (15i-k)
decreased affinity for both primary targets. Altogether these results indicate that the
introduction of carboxylic acid or polar groups to decrease cLogP, strategies that have been
4
6
used to reduce hERG inhibition, are not applicable in this position. Finally, dimethylation
provided a highly active compound, 15l, but with high hERG inhibition, somehow reduced in
the spirocyclopropyl analogue 15m.
Maintaining a methyl group in position 2, the introduction of substituents in the phenyl ring of
position 4 was explored (R , Table 3). Halogen atoms in ortho (15n,o) provided good affinities,
1
but not the desired improvement in hERG inhibition that could have been achieved by
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 68
generation of steric bulk through planarity reduction.46 A methoxy group in ortho (15p)
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
reduced MOR affinity, which worsened with a hydroxyl group (15q). Similarly, these groups
were not tolerated in the meta or para positions (15r-u). Introduction of other polar groups
(
CN, NHCOMe or CONMe ) provided also a loss of affinity (results not shown).
2
Introduction of heteroaryl groups in R was also explored (Table 4). The three possible pyridyl
1
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
isomers (15v-x) provided a decrease in MOR affinity in relation to the parent 15c. A similar
result was obtained with the pyrazinyl derivative 15y, which in addition showed a high hERG
inhibition. However, substitution by fluoro or trifluoromethyl groups in adequate positions of
the 2 and 3-pyridyl derivatives (15z-ag) improved affinities. Particularly interesting were
derivatives containing a trifluoromethyl group in ortho (15ab, 15ae, 15af), which showed
4
6
better hERG inhibition results, probably due to increase in steric bulk. Other five- and six-
membered heterocycles were also introduced (results not shown), but all of them showed
diminished MOR affinity. The only exception was a 2-thiazolyl group that showed acceptable
affinities for the primary targets but an exceptionally high hERG inhibition, in spite of its
reduced lipophilicity.
Finally, position 9 (R ) was modified (Table 5). The phenethyl group of 15c was substituted
3
with fluorine atoms, which maintained potency but only in the case of the ortho derivative
(15ah) a slight improvement in hERG inhibition was observed. Many other small groups were
introduced (halogen, alkyl, alkoxy, hydroxyl, results not shown), but only the 3-acetylamino
derivative 15al, more active than its ortho (15ak) and para (15am) isomers, showed a
favourable hERG inhibition result, not reproduced by its cyano analogue, 15an. Introduction of
five-membered rings provided a loss of MOR affinity, except for the thiophene analogue that
showed a similar profile to the parent phenyl derivative 15c (results not shown). In the case of
six-membered heterocycles, the 3 and 4-pyridyl derivatives (15ap, 15aq) showed an impaired
MOR agonism, while the 2-pyridyl isomer (15ao), with a more hindered nitrogen atom, was
tolerated, providing as well a reduction in hERG and α AR inhibition. Unfortunately, pyridyl
1
substitution of this isomer was detrimental either in terms of affinity or hERG inhibition
ACS Paragon Plus Environment
1
8

Page 19 of 68
Journal of Medicinal Chemistry
(results not shown). In addition to the phenethyl type tails, more than 30 different substituents
1
2
3
4
5
6
7
8
9
were introduced in R . These went from elongation to phenylpropyl or phenoxyethyl to change
3
to alkyl substituents, such as isopentyl, or introduction of polarity by preparing pyranylethyl,
pyranylmethyl, morpholinoethyl, 1-piperidinylcarbonylmethyl or carboxyethyl derivatives
(results not shown). However, these modifications provided an impairment in MOR affinity
that precluded any further consideration.
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
With the results of the individual modifications in hand, combinations of the best substituents
were devised. More than 40 compounds were prepared, but only some of them offered the
desired properties. As often happens in Medicinal Chemistry, assumptions that might seem
logical did not work. Particularly, combining polar substituents at both ends of the central
scaffold was detrimental for the primary activities. One of such examples is compound 15ar
(Table 6), that contains some of the best substituents in R (4-trifluoromethyl-3-pyridyl) and R
1
3
(2-pyridylethyl). We hoped that this combination could improve hERG inhibition, since 15ae
and 15ao were among the best individual modifications in this respect, but unfortunately, and
probably due to an overly reduction in lipophilicity, compound 15ar showed reduced primary
affinities.
In the end, the four most interesting derivatives identified are depicted in Table 6. Compounds
1
5asR and 15atR showed a balanced activity for the primary targets and reduced α AR
1
binding, but yet retained some hERG inhibition. The R-enantiomer of compound 15al, 15alR,
confirmed the reduced hERG inhibition, but, in addition to retaining α AR binding, the
1
presence of a functionality that could render an aniline moiety was considered as a downside in
4
7
view of aniline reported link to idiosincratic toxicity. Overall, 15au emerged as the best
derivative, since it showed a good balance of primary targets affinity, a moderate α AR binding
1
(
K = 470 nM), and was selective (less than 50% inhibition at 1 M) in the remaining
i
selectivity panel, except for the 5HT2B receptor where it showed a K = 754 nM. Compound
i
1
5au behaved as a potentially full MOR agonist (Emax = 100%) and was devoid of hERG
inhibition. It also showed a good physicochemical profile, heralded by a good solubility
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 68
(thermodynamic solubility of 2.63 mg/mL at pH 2 and 0.16 mg/mL at pH 7.4) and moderate
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
basicity (pK 7.82) and lipophilicity (experimental logP 3.41). In addition it showed adequate
a
48
in vitro metabolic stability in human liver microsomes (Clint = 20 µl/min/mg protein) and no
potential for drug-drug interactions based on the low inhibition (< 50% at 10 µM) of
recombinant human cytochrome P450 (rhCYP) isoforms (1A2, 2C9, 2C19, 2D6, 2E1 and
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
A4).49
3
Compound 15au was tested in vivo in the mouse paw pressure pain test in comparison to
oxycodone (2). Mechanical antinociception was measured as the increase in struggle response
latency in drug-treated mice with respect to control mice, when constant pressure (450 g) was
applied to the hindpaw. As shown in Figure 4A,B, compound 15au exerted a dose-dependent
antinociceptive effect with an ED of 15 mg/kg after ip administration (our standard route for
50
in vivo testing of new compounds in order to ensure exposure), a reduced potency versus
oxycodone (in this case administered orally, since we have historical data validating this route).
The effect of 15au on intestinal transit reduction (constipation) in mice was also evaluated as a
measure of the potential of the compound in showing a reduced opioid-induced side effect
profile. The percentage of distance travelled by charcoal meal versus the total length of the
small intestine was measured after drug administration in relation to the control group that
received vehicle. Systemic administration of compound 15au elicited a reduced intestinal
transit inhibition in mice in comparison with oxycodone at equianalgesic doses (20 mg/kg ip
and 10 mg/kg po, respectively, Figure 4C).
ACS Paragon Plus Environment
2
0

Page 21 of 68
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. Dose–response antinociception curves of compound 15au (A) and oxycodone (2, B)
in the paw pressure test in mice. Each point and vertical line represents the mean ± SEM
percentage of antinociception (n = 8–12 per group); C. Comparative effect of 15au and 2 on
the intestinal transit inhibition in the charcoal test in mice at equianalgesic doses (20 mg/kg ip
and 10 mg/kg po, respectively) in the paw pressure test in mice. The bars and vertical lines
represent the mean ± SEM percentage of antinociception or intestinal transit inhibition (n = 5-
7
); D. Antinociception induced by ipl administration of 15au (25 µg, ipl) in the paw pressure
test in mice in comparison to the σ R agonist PRE-084 (32 mg/kg, sc) and the concomitant
1
administration of both 15au and PRE-084. The bars and vertical lines represent the mean ±
SEM struggle response latency during stimulation with 450 g pressure on the hindpaw of mice.
In addition, compound 15au locally administered (25 µg ipl) produced a clear antinociceptive
effect in the paw pressure test in mice. This is a relevant fact, since selective MOR agonists,
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 68
5
0
such as oxycodone, fail to provide local analgesia, probably due to the fact that peripheral
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
MOR antinociception in response to mechanical stimulation is limited by σ R tonic inhibitory
1
effect.51 Interestingly, the mechanical antinociception induced by the local administration
5
2
compound 15au was abolished by concomitant administration of the σ R agonist PRE-084
1
(32 mg/kg, s.c.), indicating that 15au is recruiting σ R (in addition to MOR) and acting as a
1
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
σ1R antagonist to exert its in vivo antinociceptive effect (Figure 4D).
Conclusion
In summary, the synthesis and pharmacological activity of a new series of 1-oxa-4,9-
diazaspiro[5.5]undecane derivatives, 15, as potent dual ligands for σ R and MOR is reported
1
herein. This family, which was designed by using a merging strategy of both target
phamacophores, also displayed high hERG inhibition and suboptimal selectivity for the α AR.
1
Attempts to reduce hERG inhibition by increasing polarity were not successful, since both
parameters poorly correlated in this case. In addition, marked reduction in lipophilicity led to a
loss of potency for the primary targets. Significant improvements in the affinity for σ R and
1
MOR were achieved by substitution in position 2 of the central scaffold with small alkyl
groups and the 2R enantiomer was shown to be the eutomer for MOR affinity. Phenethyl
derivatives in position 9 and substituted aryl and pyridyl moieties in position 4 were the
optimal residues for dual affinity. The best compound identified (15au) showed a balanced
dual profile, low hERG inhibition and good ADME properties, as well as a promising in vivo
behavior. Thus, compound 15au showed analgesic activity in the mice paw pressure test
comparable to oxycodone, but less opioid induced constipation at equianalgesic doses.
Contrary to oxycodone, 15au provided local analgesic activity, reversed by the σ R agonist
1
PRE-084. Altogether these results indicate that dual MOR-agonism/σ R-antagonism could be a
1
useful strategy for obtaining potent analgesics with a safer profile. Compound 15au was
selected for a lead optimization program that led to the discovery of a clinical candidate, whose
identification will be reported soon.
ACS Paragon Plus Environment
2
2

Page 23 of 68
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Experimental Section
Unless otherwise noted, all materials were obtained from commercial suppliers and used
without further purification. Reference compounds used in biological assays were obtained
from the following vendors: oxycodone hydrochloride (2) from Macfarlan Smith Ltd (UK); E-
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
2862 (5) was obtained as previously described,22 PRE-084 ([2-(4-morpholinethyl) 1-
phenylcyclohexanecarboxylate) hydrochloride was acquired from Tocris Cookson Ltd., Bristol,
United Kingdom. Flash chromatography was performed on a Teledyne Isco CombiFlash RF
1
system with disposable columns. H spectra were recorded on a Varian Mercury 400 MHz
(spectrometer fitted with a 5 mm ATB 1H/19F/X probe) with 2 H lock in deuterated solvents.
Chemical shifts (δ) are in parts per million. Commercially available reagents and solvents
(HPLC grade) were used without further purification for all the analytical tests. Analytical
HPLC−MS for all final compounds and intermediates except 15cR, 15cS and 15alR were
performed on an Agilent HP1200-MS 6110 system. A reverse phase XBridge C18 XP column
was used (4.6 x 30 mm, 2.5 µm), gradient 5−100% B (A = 10 mM ammonium bicarbonate, B =
acetonitrile) over 8 min, injection volume: 1 μL, flow: 2.0 mL/min, temperature: 40 ºC. UV
spectra were recorded at 210 nm using an Agilent HP1200 VWD detector. Mass spectra were
obtained over the range m/z 50-1000 by electron spray ionization (ESI) in positive mode using
Agilent MS6110 Simple Quadrupole. Data were integrated and reported using Agilent
ChemStation software. Analytical HPLC−MS for compounds 15cR, 15cS and 15alR were
performed on a Waters Acquity H-Class system. A reverse phase Acquity UPLC BEH C18
column was used (2.1 x 50 mm, 1.7µm), gradient 2-95% B (A = 10 mM ammonium
bicarbonate, B = acetonitrile) over 4.75 min, injection volume: 2 μL, flow: 0.61 mL/min,
temperature: 35 ºC. PDA spectra were recorded at 195-315 nm range using a Waters PDA
detector. Mass spectra were obtained over the range m/z 100-1000 by electron spray ionization
(ESI) in positive mode using Waters ZQ Simple Quadrupole. Data were integrated and
reported using Waters Masslynx software. All final compounds displayed purity equal or
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 68
higher than 95 % as determined by said methods, with the exception of compounds 15m and
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
15w, whose purity was 93 % and 94 %, respectively. Accurate mass measurements were
carried out using an Agilent 6540 UHD Accurate-Mass QTOF system and obtained by electron
spray ionization (ESI) in positive mode. Chiral analytical and preparative HPLC were
performed in Agilent 1100 analytical and preparative HPLC systems, respectively. The coated
or immobilized polysaccharide columns were acquired from Chiral Technologies Europe;
heptane and ethanol or IPA with or without DEA at variable proportions were used as mobile
phases. All compounds active in biological assays were electronically filtered for structural
attributes common to pan assay interference compounds (PAINS) and were found to be
negative.53
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
Determination of physicochemical properties. pKa was calculated using ACDLABS 9.0.3
and clogP using ChemDraw Ultra 10.0.3. Thermodynamic solubility was measured by HPLC
after stirring solid compound for 24 h in phosphate buffer at pH = 7.4. LogP and pKa were
5
4
determined by using a pHmetric technique in a GlpKa or a Sirius-T3 Analytical instrument.
General procedures for the synthesis of 4-aryl-1-oxa-4,9-diazaspiro[5.5]undecane
derivatives 15.
(2R)-2-Methyl-9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-3-one
(15cR)
(Representative example of Scheme 2). Step 1: To a suspension of trimethylsulfoxonium
iodide (13.00 g, 59.07 mmol) and NaH (2.36 g, 60 wt% in mineral oil, 59.00 mmol) in dry
DMSO (70 mL), a solution of 1-phenethylpiperidin-4-one (16-1, 10.00 g, 49.19 mmol) in dry
DMSO (70 mL) was added dropwise. The reaction mixture was stirred at rt for 30 min, then it
was heated at 50 ºC for 1 h. After cooling to rt, ice was slowly added, and the reaction mixture
was extracted three times with ethyl acetate. The organic phases were combined, washed with
water, dried over MgSO and concentrated under vacuum to give 6-phenethyl-1-oxa-6-
4
1
azaspiro[2.5]octane (17-1) as an orange oil (8.24 g, 77%). H NMR (400 MHz, CDCl ) δ 7.31
3
–
1
7.25 (m, 2H), 7.23 – 7.17 (m, 3H), 2.87 – 2.79 (m, 2H), 2.74 – 2.58 (m, 8H), 1.88 (ddd, J =
3.2, 9.0, 4.3 Hz, 2H), 1.63 – 1.53 (m, 2H). HPLC-MS: Purity 99%, m/z=218.1 [M+H]+.
ACS Paragon Plus Environment
2
4

Page 25 of 68
Journal of Medicinal Chemistry
Step 2: To a solution of 17-1 (2.00 g, 9.20 mmol) in a mixture of EtOH-water 9:1 (40 mL),
1
2
3
4
5
6
7
8
9
aniline (0.84 mL, 9.20 mmol) was added. The reaction mixture was heated to 100 ºC overnight
in an autoclave reactor. The solvent was removed under vacuum and the residue was purified
by flash chromatography (SiO , DCM/MeOH up to 20%) to give 1-phenethyl-4-
2
1
((phenylamino)methyl)piperidin-4-ol (18-1) as a cream solid (2.18 g, 76%). H NMR (400
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
MHz, CDCl ) δ 7.32 – 7.24 (m, 2H), 7.23 – 7.14 (m, 5H), 6.72 (t, J = 7.3 Hz, 1H), 6.68 (d, J =
3
7
.7 Hz, 2H), 3.95 (t, J = 6.3 Hz, 1H), 3.13 (d, J = 6.5 Hz, 2H), 2.85 – 2.80 (m, 2H), 2.80 – 2.72
(m, 2H), 2.67 – 2.59 (m, 2H), 2.48 – 2.40 (m, 1H), 1.81 – 1.69 (m, 4H). HPLC-MS: Purity
9
3%, m/z=311.2 [M+H]+.
Step 3: To a solution of 18-1 (578 mg, 1.86 mmol) in DCM (22 mL), TEA (0.62 mL, 4.47
mmol) was added and the mixture was cooled to 0 ºC under a nitrogen atmosphere. Then, a
solution of 2-chloropropionyl chloride (10-1, 0.27 mL, 2.79 mmol) in DCM (22 mL), was
added dropwise and the reaction mixture was stirred at 0 ºC for 2 h. Saturated aqueous
NaHCO was then added and it was extracted with DCM. The combined organic phases were
3
washed with brine, dried over MgSO4, filtered and concentrated to dryness to give 2-chloro-N-
((4-hydroxy-1-phenethylpiperidin-4-yl)methyl)-N-phenylpropanamide (19-1) as an oil used in
1
the next step without further purification (670 mg, 89%). H NMR (400 MHz, CDCl ) δ 7.49 –
3
7
1
.36 (m, 4H), 7.30 – 7.23 (m, 3H), 7.22 – 7.15 (m, 3H), 4.31 (q, J = 6.6 Hz, 1H), 3.89 (d, J =
4.2 Hz, 1H), 3.77 (d, J = 14.2 Hz, 1H), 2.84 – 2.77 (m, 2H), 2.77 – 2.67 (m, 2H), 2.67 – 2.57
(m, 2H), 2.54 – 2.40 (m, 2H), 1.76 – 1.67 (m, 2H), 1.67 – 1.60 (m, 2H), 1.58 (d, J = 6.6 Hz,
3
H).
Step 4: To a solution of 19-1 (670 mg, 1.67 mmol) in THF (28 mL), cooled to -78 ºC using a
dry ice/acetone bath, potassium tert-butoxide solution (2.5 mL, 1 M in THF, 2.50 mmol) was
added under a nitrogen atmosphere. The reaction mixture was then stirred at -30 ºC for 3 h.
Water was added and it was extracted with ethyl acetate. The combined organic phases were
washed with brine, dried over MgSO , filtered and concentrated to dryness. The residue was
4
purified by flash chromatography (SiO , DCM/MeOH up to 10%) to give 2-methyl-9-
2
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 68
phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-3-one (15c) as a cream solid (511 mg,
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
1
84%). H NMR (400 MHz, CDCl ) δ 7.44 – 7.37 (m, 2H), 7.32 – 7.26 (m, 5H), 7.23 – 7.17 (m,
3
3
H), 4.38 (q, J = 6.8 Hz, 1H), 3.81 (d, J = 12.2 Hz, 1H), 3.41 (d, J = 12.2 Hz, 1H), 2.89 – 2.78
m, 2H), 2.78 – 2.51 (m, 5H), 2.44 (t, J = 10.3 Hz, 1H), 2.28 (d, J = 14.1 Hz, 1H), 1.82 – 1.69
(m, 3H), 1.53 (d, J = 6.8 Hz, 3H). HPLC-MS: Purity 100%. HRMS [M] (diff ppm) 364.2227 (-
.93).
(
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
Starting from racemic 15c (241 mg), a chiral preparative HPLC separation (column: Chiralpak
IA; temperature: ambient; flow: 12 mL/min; eluent: n-Heptane/(IPA + 2% DEA) 95/5 v/v) was
carried out to obtain (2R)-2-methyl-9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-3-
one (15cR) as a white solid (83 mg, recovery yield 69%) with ee 97% (chiral HPLC) and [α]D20
3
5 (c 0.09, CHCl ), and enantiomer (2S)-2-methyl-9-phenethyl-4-phenyl-1-oxa-4,9-
3
diazaspiro[5.5]undecan-3-one (15cS) as a white solid (92 mg, recovery yield 77%) with ee
00% (chiral HPLC) and [α]D20 -35 (c 0.15, CHCl ). The absolute configuration was
1
3
established by chiral synthesis of compound 15cR: starting from (S)-2-chloropropionyl
chloride (10-1S) (ee 95%) and following Scheme 2 but using modified acylation conditions
(
analogous to the ones described in Scheme 3 Step 2, i.e. potassium carbonate as base in a
biphasic system of ethyl acetate-water), the corresponding (S)-acyl derivative (19-1S) was
obtained, which assuming inversion of configuration in cyclization step, rendered compound
1
2
5cR with ee 91%, as determined by chiral HPLC.
-(Hydroxymethyl)-9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-3-one (15g).
A mixture of 2-(methoxymethyl)-9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-3-
one (15h, 1.53 g, 3.26 mmol) and palladium hydroxide (306 mg, 20%wt on charcoal) in a
mixture of THF/AcOH 9:1 (15 mL) was heated overnight at 50 ºC under 3 bars of hydrogen.
After inertizing with nitrogen, the catalyst was filtered off and the solvent was evaporated to
dryness. This hydrogenation cycle was repeated 3 times, adding fresh catalyst each time until
significant conversion was achieved. Then, the reaction mixture was finally filtered over a pad
of celite to remove the catalyst and the filtrate was concentrated to dryness. The residue was
ACS Paragon Plus Environment
2
6

Page 27 of 68
Journal of Medicinal Chemistry
purified by flash chromatography (SiO , DCM/MeOH up to 10%) to give 2-(hydroxymethyl)-
2
1
2
3
4
5
6
7
8
9
9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro [5.5]undecan-3-one (15g) as a white solid (0.66 g,
1
5
3
3%). H NMR (400 MHz, CDCl ) δ 7.45 – 7.39 (m, 2H), 7.33 – 7.26 (m, 5H), 7.24 – 7.17 (m,
3
H), 4.32 (t, J = 4.4 Hz, 1H), 4.07 – 3.89 (m, 2H), 3.81 (d, J = 12.3 Hz, 1H), 3.43 (d, J = 12.4
Hz, 1H), 2.93 – 2.53 (m, 6H), 2.53 – 2.38 (m, 1H), 2.30 (d, J = 14.9 Hz, 1H), 1.80 (br s, 2H),
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
1
.57 (br s, 2H). HPLC-MS: Purity 100%. HRMS [M] (diff ppm) 380.2168 (1.4).
2
-(3-Oxo-9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-2-yl)acetic
acid
hydrochloride (15j). To a solution of methyl 2-(3-oxo-9-phenethyl-4-phenyl-1-oxa-4,9-
diazaspiro[5.5]undecan-2-yl)acetate (15i, 250 mg, 0.58 mmol) in a mixture of MeOH/water 7:1
(7 mL), potassium carbonate (317 mg, 2.29 mmol) was added and the reaction mixture was
stirred at rt for 6 days. Water was added, the pH was adjusted to 3-4 by addition of 1 M HCl
aqueous solution and it was extracted with DCM. The combined organic phases were dried
over MgSO , filtered and concentrated to dryness to give 2-(3-oxo-9-phenethyl-4-phenyl-1-
4
oxa-4,9-diazaspiro[5.5]undecan-2-yl)acetic acid (15j free base) as a white solid (234 mg,
quantitative). To a slurry of the previous free base (50 mg, 0.12 mmol) in anhydrous diethyl
ether (1 mL), HCl (2 M solution in diethyl ether, 0.06 mL, 0.12 mmol) was added and the
mixture was stirred at rt for 1 h. The solids were filtered and dried in vacuo to give 15j HCl salt
1
as a white solid (53 mg, 97%). H NMR (400 MHz, CD OD) δ 7.51 – 7.40 (m, 2H), 7.40 –
3
7.22 (m, 8H), 4.60 (t, J = 4.7 Hz, 1H), 3.98 (d, J = 12.5 Hz, 1H), 3.73 – 3.52 (m, 3H), 3.47 –
3.24 (m, 4H), 3.20 – 3.08 (m, 2H), 2.97 (qd, J = 17.1, 4.7 Hz, 2H), 2.72 (d, J = 14.2 Hz, 1H),
2.19 – 1.90 (m, 3H). HPLC-MS: Purity 98%. HRMS [M] (diff ppm) 408.2126 (-1).
N,N-Dimethyl-2-(3-oxo-9-phenethyl-4-phenyl-1-oxa-4,9-diazaspiro[5.5]undecan-2-
yl)acetamide hydrochloride (15k). To a solution of 2-(3-oxo-9-phenethyl-4-phenyl-1-oxa-
4
,9-diazaspiro[5.5]undecan-2-yl)acetic acid (15j, 50 mg, 0.12 mmol) in DCM (4 mL),
dimethylamine solution (0.18 mL, 2 M in THF, 0.36 mmol), 1-hydroxybenzotriazole (18 mg,
.13 mmol), N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide hydrochloride (24 mg, 0.12
0
mmol) and TEA (0.06 mL, 0.40 mmol) were added under a nitrogen atmosphere and the
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 68
reaction mixture was stirred at rt overnight. Water was added and it was extracted with DCM.
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
The combined organic phases were washed with brine, dried over MgSO , filtered and
4
concentrated to dryness. The residue was purified by flash chromatography (SiO2,
DCM/MeOH up to 10%) to give N,N-dimethyl-2-(3-oxo-9-phenethyl-4-phenyl-1-oxa-4,9-
diazaspiro[5.5]undecan-2-yl)acetamide (15k free base) as an oil (17 mg, 32%). To a solution of
the previous free base (17 mg, 0.04 mmol) in anhydrous diethyl ether (1 mL), HCl (2 M
solution in diethyl ether, 0.02 mL, 0.04 mmol) was added and the mixture was stirred at rt for 2
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
h. Then it was evaporated to dryness and dried in vacuo to give 15k HCl salt as a white solid
1
(
14 mg, 76%). H NMR (400 MHz, CD OD) δ 7.48 – 7.41 (m, 2H), 7.40 – 7.24 (m, 8H), 4.63
3
(t, J = 4.7 Hz, 1H), 4.04 (d, J = 12.4 Hz, 1H), 3.70 – 3.56 (m, 2H), 3.53 (d, J = 12.4 Hz, 1H),
3
.51 – 3.33 (m, 4H), 3.15 – 3.09 (m, 2H), 3.08 (s, 3H), 3.06 (d, J = 4.6 Hz, 2H), 2.98 (s, 3H),
2.77 (d, J = 15.4 Hz, 1H), 2.07 – 1.91 (m, 3H). HPLC-MS: Purity 97%. HRMS [M] (diff ppm)
4
35.2605 (-2.1).
8
-Phenethyl-12-phenyl-4-oxa-8,12-diazadispiro[2.1.5.3]tridecan-13-one
(15m)
(Representative example of Scheme 4 using copper catalyst in final step). Step 1: To a
suspension of trimethylsulfoxonium iodide (24.30 g, 110.42 mmol) and NaH (4.40 g, 60 wt%
in mineral oil, 110.00 mmol) in DMSO (140 mL), a solution of tert-butyl 4-oxopiperidine-1-
carboxylate (6, 20.00 g, 100.38 mmol) in DMSO (140 mL) was added dropwise. The reaction
mixture was stirred at rt for 30 min, then it was heated at 50 ºC for 1 h. After cooling to rt, ice
was slowly added, and the reaction mixture was extracted with ethyl acetate. The organic
phases were combined, washed with water, dried over MgSO and concentrated under vacuum
4
to give tert-butyl 1-oxa-6-azaspiro[2.5]octane-6-carboxylate (7) as a white solid (17.60 g,
1
8
2%). H NMR (400 MHz, CDCl ) δ 3.77 – 3.63 (m, 2H), 3.42 (ddd, J = 13.3, 9.5, 3.7 Hz,
3
2
H), 2.68 (s, 2H), 1.79 (ddd, J = 13.7, 9.5, 4.4 Hz, 2H), 1.46 (s, 9H), 1.44 – 1.40 (m, 2H).
+
HPLC-MS: Purity 95%, m/z=158.1 [M+H-56] (-tBu, Boc fragmentation).
Step 2: To a solution of 7 (10.00 g, 46.89 mmol) in a mixture of EtOH-water 9:1 (200 mL), 4-
methoxybenzylamine (6.43 g, 46.90 mmol) was added. The reaction mixture was heated at 100
ACS Paragon Plus Environment
2
8

Page 29 of 68
Journal of Medicinal Chemistry
ºC overnight in a sealed vessel. The solvent was removed under vacuum and the residue was
1
2
3
4
5
6
7
8
9
purified by flash chromatography, (SiO , DCM/MeOH up to 20%) to give tert-butyl 4-
2
hydroxy-4-(((4-methoxybenzyl)amino)methyl) piperidine-1-carboxylate (30) as a yellow oil
1
(
13.55 g, 82%). H NMR (400 MHz, CDCl ) δ 7.25 – 7.20 (m, 2H), 6.89 – 6.84 (m, 2H), 3.80
3
(s, 3H), 3.89 – 3.70 (br s, 2H), 3.78 (s, 2H), 3.15 (t, J = 11.7 Hz, 2H), 2.55 (s, 2H), 1.54 – 1.47
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
(m, 2H), 1.45 (s, J = 9.5 Hz, 9H), 1.43 – 1.35 (m, 2H). HPLC-MS: Purity 98%, m/z=351.2
[
M+H]+.
Step 3: To a solution of compound 30 (9.94 g, 28.36 mmol) and TEA (9.5 mL, 68.16 mmol) in
3
9
DCM (500 mL), a solution of 2-bromo-4-chlorobutanoyl chloride (31) (9.35 g, 42.53 mmol)
in DCM (200 mL) was added dropwise at 0 ºC under a nitrogen atmosphere. The reaction
mixture was stirred at 0 ºC for 2 h. Saturated aqueous NaHCO was added and it was extracted
3
with DCM. The combined organic phases were dried over MgSO , filtered and concentrated to
4
dryness
to
give
crude
tert-butyl
4-((2-bromo-4-chloro-N-(4-
methoxybenzyl)butanamido)methyl)-4-hydroxypiperidine-1-carboxylate (32) as an orange oil
(15.10 g, quantitative). To a solution of crude 32 (14.80 g, 27.72 mmol) in THF (185 mL),
cooled to 0 ºC under a nitrogen atmosphere, potassium tert-butoxide solution (111 mL, 1 M in
THF, 111.00 mmol) was slowly added and the reaction mixture was stirred at 0 ºC for 2 h.
Saturated aqueous NH Cl was then added and it was extracted with ethyl acetate. The organic
4
phases were combined, dried over MgSO , filtered and concentrated to dryness. The residue
4
was purified by flash chromatography (SiO , DCM/MeOH up to 10%) to give tert-butyl 12-(4-
2
methoxybenzyl)-13-oxo-4-oxa-8,12-diazadispiro[2.1.5.3]tridecane-8-carboxylate (33) as a
1
yellow oil (5.51 g, 48%). H NMR (400 MHz,CDCl ) δ 7.18 (d, J = 8.6 Hz, 2H), 6.87 (d, J =
3
8
1
.7 Hz, 2H), 4.54 (s, 2H), 3.81 (s, 3H), 3.63 (br s, 2H), 3.18 (s, 2H), 3.06 (t, J = 11.4 Hz, 2H),
.86 (d, J = 13.5 Hz, 2H), 1.43 (s, 9H), 1.48 – 1.34 (m, 4H), 0.98 (q, J = 4.5 Hz, 2H). HPLC-
MS: Purity 95%, m/z=417.2 [M+H]+.
Step 4: To a solution of 33 (1.50 g, 3.60 mmol) in DCM (36 mL), trifluoroacetic acid (2.8 mL,
3
6.34 mmol) was added, and the reaction mixture was stirred at rt for 4 h. The solvent was
ACS Paragon Plus Environment
2
9