Novel fluoroalkyl derivatives of selective kappa opioid receptor antagonist JDTic: Design, synthesis, pharmacology and molecular modeling studies

Article abstract of DOI:10.1016/j.ejmech.2014.12.016

Novel N- and O-fluoroalkyl derivatives of the highly potent KOR antagonist JDTic were designed and synthesized. Their opioid receptor properties were compared in both in vitro binding assays and modeling approach. All compounds displayed nanomolar affinit

Full text of DOI:10.1016/j.ejmech.2014.12.016

European Journal of Medicinal Chemistry 90 (2015) 742e750
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Novel fluoroalkyl derivatives of selective kappa opioid receptor
antagonist JDTic: Design, synthesis, pharmacology and molecular
modeling studies
a
,
b
,
,
c
,
d
a, b, c, *
c, Nathalie Colloc'h b
, Cecile Perrio
ꢀ
ꢀ
Sebastien Schmitt
a CNRS, UMR 6301 ISTCT, LDM-TEP, GIP CYCERON, Boulevard Henri Becquerel, 14074 Caen, France
Universite de Caen Basse-Normandie, Normandie Univ., France
b
ꢀ
c CEA, DSV/I2BM, France
d CNRS, UMR 6301 ISTCT, CERVOxy group, GIP CYCERON, Boulevard Henri Becquerel, 14074 Caen, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel N- and O-fluoroalkyl derivatives of the highly potent KOR antagonist JDTic were designed and
synthesized. Their opioid receptor properties were compared in both in vitro binding assays and
modeling approach. All compounds displayed nanomolar affinities for KOR. The fluoropropyl derivatives
were more active than their fluoroethyl analogues. N-Fluoroalkylation was preferable to O-alkylation to
keep a selective KOR binding. Compared to JDTic, the N-fluoropropyl derivative 2 bound to KOR with an
Received 29 October 2014
Received in revised form
8 December 2014
Accepted 10 December 2014
Available online 11 December 2014
only 4-fold lower affinity and a higher selectivity relative to MOR and DOR [Ki(k ¼ 1.6 nM; Ki(m
/
)
)
Ki(k ¼ 12; Ki(d)/Ki(k ¼ 159 for 2 versus Ki(k ¼ 0.42 nM; Ki(m)/Ki(k ¼ 9; Ki(d)/Ki(k ¼ 85 for JDTic]. Modeling
)
)
)
)
)
Keywords:
Kappa opioid receptor
JDTic
studies based on the crystal structure of the JDTic/KOR complex revealed that fluorine atom in ligand 2
was involved in specific KOR binding. Ligand 2 was concluded to merit further development for KOR
exploration.
(3R,4R)-3,4-dimethyl-4-(3-hydroxyphenyl)
© 2014 Elsevier Masson SAS. All rights reserved.
piperidine
Selectivity
Docking
1. Introduction
have been detected in cerebral regions such as the ventral
tegmental area, nucleus accumbens, prefrontal cortex, hippocam-
The kappa opioid receptor (KOR, OPRK or
k
) belongs to the
pus, striatum, amygdala and hypothalamus, thought to be critical in
mood modulation, motivation, stress reactivity, perception,
learning memory, and behavior response to drugs. Growing evi-
dence indicates that changes in DYN/KOR system contribute to
symptom clusters that are shared by various psychiatric and
addictive disorders (i.e., decreased motivation and negative affect),
and that KOR disruption produces anti-stress effects [12e16].
Consequently, KOR is strongly believed to be a molecular key-target
to investigate the mechanisms involved in the psychopathologies
and to elaborate new therapeutic strategies. This finding has
recently stimulated interest in the development of KOR antagonists
as pharmacotherapies to treat depression, anxiety, schizophrenia,
alcoholism, drugs seeking and relapse [17e22]. Specific KOR an-
tagonists also represent valuable candidates as in vivo imaging
agents to map KORs in brain and to examine their functions both in
healthy and pathological conditions. Indeed, the successful devel-
opment of selective radiotracers for KOR imaging by positron
emission tomography (PET) would allow new investigations of
subfamily of G-proteinecoupled receptors (GPCRs) and is closely
associated to the action of dynorphin (DYN) peptides as specific
endogenous ligands [1]. KOR shares extensive homology with mu
(MOR, OPRM or m) and delta (DOR, OPRD or d) opioid receptor sub-
types, but remains unique by its pharmacology and physiological
effects [2]. The three opioid receptors, KOR, MOR and DOR, regulate
major functions including pain, emotional tone, appetite and
reward circuitry. It is well established that KOR agonists produce an
aversive effect, whereas agonists at the MOR and DOR sites are
rewarding and reinforcing [3e6]. KOR is widely expressed in hu-
man throughout the central and peripheral nervous system, and is
the most abundant OR in brain [7e11]. High levels of KOR mRNA
* Corresponding author. CNRS, UMR 6301 ISTCT, LDM-TEP, GIP CYCERON,
Boulevard Henri Becquerel, 14074 Caen, France.
E-mail address: perrio@cyceron.fr (C. Perrio).
http://dx.doi.org/10.1016/j.ejmech.2014.12.016
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
743
neuropsychiatric and addictive disorders, and help the develop-
ment of novel therapeutic agents by correlating dose, in vivo
pharmacokinetic parameters, and receptor occupancy of novel
KOR-targeting drugs.
Very few pure selective KOR antagonists are reported so far
(Chart 1). Classically, the reference agents were nor-BNI (nor-
binaltorphimine) and GNTI (50-guanidinonaltrindole) [23,24]. Both
compounds are derived from naltrexone and depend on the N-
cyclopropylmethyl group for their antagonist properties. First
discovered nor-BNI was commonly used as the standard tool in
opioid pharmacology [23], and subsequently generated GNTI was
found to be slightly more potent than norBNI [norBNI:
Chart 2. JDTic.
Ki(k
¼
0.24 nM, Ki(m)/Ki(k
¼
204, Ki(d)/Ki(k
¼
170; GNTI:
)
)
)
Ki(k ¼ 0.18 nM, Ki(m)/Ki(k ¼ 125; Ki(d)/Ki(k ¼ 257] [24,25]. In vivo,
)
)
)
both morphinan-derivatives were showed to display a very slow
brain uptake and release, and to antagonize the actions of KOR
agonists for a long time (up to several weeks) [26e29]. Recently, a
new class of aminobenzyloxyarylamides has been identified as KOR
antagonists with, however, a lower potency in terms of in vitro
affinity and selectivity compared to norBNI [30]. Among them,
LY2456302 [Ki(k ¼ 0.8 nM, Ki(m)/Ki(k ¼ 30, Ki(d)/Ki(k ¼ 194 versus
hydroxylphenyl)piperidine class of compounds, represents the
best established pure selective KOR antagonist in in vitro experi-
ments [Ki(k ¼ 0.03 nM, Ki(m)/Ki(k ¼ 338, Ki(d)/Ki(k ¼ 4935 to be
)
)
)
compared to the above cited values for norBNI and LY2456302]
[31]. A recent study has reported that there was no non-opioid
target from a broad panel of 43 receptors and transporters for
which JDTic showed a significant affinity [38]. In vivo, JDTic also
demonstrated highly specific KOR antagonist properties; JDTic was
reported to block the KOR agonist U50488-induced antinociception
in mouse, while not antagonizing mu-subtype opioid receptor
(MOR) agonist-induced analgesia [37,39]. Despite having a poor
brain penetration, JDTic was found to produce in mouse, rat and
rhesus monkeys, long-lasting antagonistic effects [28,29], and to
display a robust effectiveness in various rodent models of depres-
sion, anxiety, alcohol seeking, nicotine withdrawal and stress-
induced cocaine relapse [39e42]. JDTic has been evaluated in
phase I clinical trials as drug in the treatment of cocaine addiction
[43]. However adverse ventricular tachycardia effects were noticed,
that were unpredictable based on the absence of cardiotoxicity in
the non-human primate [44].
)
)
)
Ki(k ¼ 0.15 nM, Ki(m)/Ki(k ¼ 216, Ki(d)/Ki(k ¼ 43 for norBNI in the
)
)
)
same set of evaluation experiment] was found to exhibit short-
acting pharmacokinetic properties in vivo, and to reduce ethanol
self-administration in alcohol-preferring rats [31]. LY2456302 also
demonstrated activity in mouse model predictive of
antidepressant-like efficacy [31], and has been advanced to phase II
clinical trials for the augmentation of the antidepressant therapy in
treatment-resistant depression [32]. The analogue LY2795050 was
lastly labelled with carbon-11 (bþ emitter, t1/2 ¼ 20.4 min) and
demonstrated favorable pharmacokinetic properties and binding
profile in primate by PET imaging [33,34]. To date, JDTic [35e37]
that
belongs
to
the
trans-(3R,4R)-3,4-dimethyl-4-(3-
Due to its remarkable pharmacological properties, JDTic re-
mains an important lead ligand for KOR exploration. Structural
features of the receptor binding pocket were recently provided by
KOR/JDTic complex co-crystallisation revealing a tight fit of JDTic in
the bottom of the binding cleft forming ionic polar and extensive
hydrophobic interactions with the receptor [45]. The protonated
amines in both piperidine and tetrahydroisoquinoline moieties in
JDTic formed salt bridges with a highly conserved aspartic acid
(Asp138) side chain, probably fixing the ligand in a stabilized V-
shaped conformation (Chart 2). From numerous structureeactivity
relationship studies conducted on JDTic, it has been well estab-
lished that the (3R,4R)-3,4-dimethyl-4-(3-hydroxyphenyl)piperi-
dine moiety provided pure opioid antagonist properties while KOR
selectivity was a consequence of the N-substituent [35,36,46e51].
Only a few structural modifications on JDTic structure were
allowed to retain binding properties. Introduction of a methyl
group at the nitrogen atom and the hydroxyl group of the tetra-
hydroisoquinoline ring to give RTI-5989-97 and RTI-5989-212
respectively (Chart 1), led to minimal alteration of in vitro
intrinsic antagonism activity [JDTic: Ke(k
¼
0.02 nM, Ke(m /
)
)
Ke(k ¼ 1255, Ke(d)/Ke(k ¼ 3830; RTI-5989-97: Ke(k ¼ 0.16 nM,
)
)
)
Ke(m)/Ke(k
¼
1313, Ke(d)/Ke(k
¼
3070; RTI-5989-212:
)
)
Ke(k ¼ 0.06 nM, Ke(m)/Ke(k ¼ 857, Ke(d)/Ke(k ¼ 1970] [36,47].
)
)
)
Surprisingly, the N-methyl derivative RTI-5989-97 was found to be
a long-acting antagonist in vivo like JDTic, whereas the O-methyl
analogue RTI-5989-212 displayed short duration of action [29]. In
previous works, we developed the radiolabelling with carbon-11 of
RTI-5989-97 (named as [11C]Me-JDTic) [52]. Ex vivo evaluation in
mouse revealed a very high specific binding of [11C]MeJDTic for
Chart 1. Structures of known potent KOR antagonists and of target compounds 1e4.

744
S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
Scheme 1. Synthesis of N-fluoroalkyl JDTic derivatives 1 and 2. Reagents and conditions: (a) TsCl, pyridine, 0 ꢀC, 4e7 h; b) TBAF, ACN, reflux, 3 h; c) tosylate 5 or 6, ACN, reflux, 72 h.
KOR but a poor brain penetration of the radioligand. Based on
these promising in vivo results, we were next interested in further
exploring O- and N-substituted JDTic derivatives in order to
improve the overall in vitro and in vivo profile. We chose to deri-
vate JDTic with lipophilic fluoroalkyl chains already known to have
a beneficial impact on efficacy and selectivity in pharmaceuticals,
and also to affect adsorption, distribution, metabolism, and
excretion properties of the parent compound [53,54]. This
approach also allowed a further radiolabeling with fluorine-18 (bþ
emitter, t1/2 ¼ 109.8 min) that possesses the best physical char-
acteristics for imaging purposes, i.e., an optimal physical half-life
for more convenient radiochemistry conditions and in vivo inves-
tigation, a high decay purity (97%), a relatively low positron energy
(0.64 MeV), and relatively high experimental specific activities
(correlated to the rather low abundance of fluorine in nature). Such
an 18F-labelled radioligand for KOR PET imaging is still missing to
date. Here, we report the synthesis, pharmacological evaluation
and molecular modeling studies for the N- and O-fluoroethyl and
fluoropropyl JDTic analogues 1e4 (Chart 1). We demonstrated that
both the alkylation site and the length of the alkyl chain were
crucial for KOR binding and antagonism, and that fluorine atom
positively participated to the ligand-receptor recognition.
2. Chemistry
N-Fluoroalkyl JDTic derivatives 1 and 2 were prepared by
alkylation of JDTic with the corresponding fluoroalkyltosylates 5
and 6 as depicted in Scheme 1. JDTic was synthetized from racemic
1,3-dimethyl-4-piperidinone according to the reported method
[36]. Tosylates 5 and 6 were obtained from ethylene and propylene
glycols 7 and 8 according to a two-step procedure involving a bis-
tosylation followed by a monofluorination reaction with TBAF
[55,56].
The synthesis of O-fluoroalkyl JDTic derivatives 3 and 4 is dis-
played in Scheme 2. Compounds 3 and 4 were obtained from JDTic
precursor 9 [36] and the O-fluoroalkyltetrahydroisoquinolinic acids
10 and 11 issued from N-Boc-D-7-hydroxy-1,2,3,4-tetrahydroi
soquinoline-3-carboxylic acid (N-Boc-D-Tic) 12. N-Boc-D-Tic 12
prepared as reported [57], was converted to the benzyl ester 13 by
esterification with benzyl bromide. Ester 13 was subjected to an O-
alkylation by reaction with the fluoroalkyltosylates 5 and 6 to yield
14 and 15 respectively. Debenzylation of 14 and 15 with KOH
afforded acids 10 and 11. Coupling of acids 10 and 12 with amine 9
using BOP followed by Boc-deprotection with TFA gave the O-flu-
oroalkyl target compounds 3 and 4.
Scheme 2. Synthesis of O-fluoroalkyl JDTic derivatives 3 and 4. Reagents and conditions: (a) BnBr, NaHCO3, DMF, rt, 24 h; b) tosylate 5 or 6, K2CO3, acetone, reflux, 48 h; c) KOH,
EtOH/water, rt, 6 h; d) acid 10 or 11, BOP, Et3N, THF, rt, 3 h; e) TFA, DCM, ꢁ20 ꢀC, 10 min then rt, 1 h.

S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
745
Table 1
Affinity binding of references compounds, JDTic and fluoroalkyl-JDTic derivatives 1e4 at the different subtypes of opioid receptors (KOR, MOR and DOR).a
Compound
Ki (nM)a
Selectivity
MOR/KOR
KORb ([3H]U69593)
MORc ([3H]diprenorphine)
DORd ([3H]DADLE)
DOR/KOR
U50488
Naltrexone
DPDPE
JDTic
N-fluoroethylJDTic 1
N-fluoropropylJDTic 2
O-fluoroethylJDTic 3
O-fluoropropylJDTic 4
0.36 0.09
nde
nde
nde
e
e
e
e
0.45 0.3
nde
e
nde
nde
1.41 0.30
36.11 5.58
72.81 7.82
254.16 13.42
81.33 8.39
72.92 7.93
e
0.42 0.10
5.51 0.92
1.62 0.65
24.01 1.78
8.19 1.77
4.00 0.40
13.23 3.39
19.61 3.19
63.08 5.12
57.52 6.27
9.5
2.4
12.2
2.6
7.0
85.7
13.2
158.9
3.4
8.9
a
Data are shown as the mean values SD from at least three independent experiments.
Rat recombinant KOR CHO cells.
Human recombinant MOR HEK cells.
Human recombinant DOR CHO cells.
Not determined.
b
c
d
e
3. Results and discussion
3.2. Molecular modeling
3.1. OR binding studies
The overall pharmacological results revealed that introduction
of the fluoroalkyl chain on JDTic affected its binding properties. The
degree of perturbation depended on both the alkylation site and
the length of the fluoroalkyl chain. Interestingly, substitution with
the more sterically hindered fluoropropyl chain at the N-position
led to a small reduced KOR affinity and an increased selectivity.
Fluoroethyl substitution and O-alkylation provided significant loss
of affinity and selectivity for KOR. In order to understand the effect
of JDTic substitution, we undertook docking studies using the
crystallographic structure of the KOR/JDTic complex [45]. Fluo-
roethyl and fluoropropyl derivatives 1e4 have been built and
docked in the active site of the crystallographic structure of KOR
showing that the affinity of the four ligands could be in part
explained by the crystal structure. The docking of the four flexible
compounds 1e4 in the binding pocket of rigid KOR has been per-
formed using AutoDock4 [61]. For each of the four compounds 1e4,
the best ranked docking pose with the lower estimated free energy
and the best estimated inhibition constant was chosen and
analyzed.
The four compounds 1e4 were each evaluated for their binding
affinity for KOR, MOR and DOR. Affinities were determined by
radioligand displacement experiments with specific KOR [3H]
U69593, MOR [3H]diprenorphine and DOR [3H]DADLE ligands in rat
recombinant KOR CHO cells [58] and human recombinant MOR [59]
or DOR [60] HEK cells. U50488, naltrexone and DPDPE were used as
reference ligands of KOR, MOR and DOR respectively for validation
of the binding assays. For comparison, parent JDTic prepared by our
hands was also tested. The overall results were presented in Table 1,
as Ki values in nM. JDTic was found to display a subnanomolar af-
finity (Ki ¼ 0.42 nM) for KOR similar to the previously reported
value (0.32 nM) determined in the same assay [35]. All new ligands
1e4 displayed KOR binding potency lower than JDTic but Ki values
remained in the nanomolar range. The highest affinity and selec-
tivity for KOR was found for the N-fluoropropyl derivative 2. Af-
finity of 2 for KOR (Ki(k ¼ 1.6 nM) was about 12-fold higher than for
)
MOR (Ki(m
¼
19.6 nM) and 158-fold higher than for DOR
)
(Ki(d ¼ 254.2 nM). Compared to JDTic (Ki(k ¼ 0.42 nM), compound
In the crystal structure of KOR/JDTic [45], the protonated ni-
trogen atoms in both piperidine (Np) and isoquinoline (Ni) of JDTic
form salt bridges to Asp138 side chain with bond lengths of 2.96
)
)
2 displayed an only 3.8-fold lower affinity for KOR, and a signifi-
cantly higher selectivity with Ki(m)/Ki(k and Ki(d)/Ki(k of 12.2 and
)
)
158.9 respectively for 2 versus 9.5 and 85.7 for JDTic. The N-fluo-
and 2.77 Å for Np-Od1 and Ni-Od2 respectively (Chart 2). The iso-
roethyl analogue 1 (Ki(k ¼ 5.5 nM) was less potent than 2, with a 3-
quinoline nitrogen Ni is solvent-accessible and points toward the
exterior of the pocket (Fig. 1A). The hydroxyl group on isoquinoline
is buried at one end of the ligand binding pocket, and participates in
water-mediated interactions with residues Tyr139, Lys227 and
His291 (Fig. 1B).
Construction of ligand 4 by introduction of a fluoropropyl
chain on the isoquinoline hydroxyl led to a complete different
location of 4 compared to JDTic (Fig. 1C). This change was due to
steric hindrance in the bottom of the binding pocket. The salt
bridge to Asp138 was disrupted with a very high value for the
)
fold lower affinity and a very poor selectivity for KOR. The O-fluo-
roalkyl derivatives 3 (Ki(k ¼ 24.0 nM) and 4 (Ki(k ¼ 8.2 nM) had
)
)
significantly reduced affinity and selectivity for KOR by comparison
with N-substituted analogue 1e2 and JDTic. Thus, the N-substitu-
tion was preferred to the O-alkylation. Whatever the alkylation site,
the longer fluoropropyl chain was better tolerated than the shorter
fluoroethyl group.
Np-Od1 distance (7.46 Å). Ligand 4 displayed the worst scores
(
D
G ¼ ꢁ10.29 kcal/mol; Ki ¼ 28.59 nM; Table 2), a result in
Table 2
accordance with the in vitro pharmacological evaluation (Table 1).
On the opposite, the fluoroethyl chain could be built on the hy-
droxyl of the isoquinoline cycle without disturbing location of
Docking results for compounds 1e4 compared to JDTic.
Compound
D Od Od
G (kcal/mol)a Ki (nM)b 1 e Np (Å)c 2 eNi (Å)d
N-fluoroethylJDTic 1 ꢁ10.43
N-fluoropropylJDTic 2 ꢁ10.86
O-fluoroethylJDTic 3 ꢁ12.19
O-fluoropropylJDTic 4 ꢁ10.29
22.72
10.86
1.16
2.99
2.68
2.74
7.46
3.41
3.24
2.85
2.75
JDTic structure (Fig. 1D). The Np-O
distances were not significantly affected (Table 2). In addition, the
docking procedure showed that the O-fluoroethyl JDTic
d1 (2.74 Å) and Ni-Od2 (2.85 Å)
3
28.59
(
D
G ¼ ꢁ12.19 kcal/mol; Ki ¼ 1.16 nM) was the best ligand
a
Estimated free energy of binding DG.
Estimated inhibition constant Ki.
amongst the four compounds 1e4 (Table 2). This result was quite
surprising on the basis of the experimental in vitro affinities
(Table 1). In the KOR/JDTic analogues recognition, the importance
b
c
Distance between Asp138 O
d
d
1 atom and piperidine nitrogen (Np).
2 atom and isoquinoline nitrogen (Ni).
d
Distance between Asp138 O

746
S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
of direct or indirect interactions between isoquinoline and KOR
was clearly revealed [45,51]. In a recent docking study on JDTic
analogues [51], the authors suggested than removing the water-
mediated interactions between KOR and ligand could result in
enhanced or altered properties of the ligand. As illustration, the
replacement of isoquinoline hydroxyl by hydrogen atom that
prevented water-mediated interactions to take place, resulted in
about a 100-fold reduction of affinity. On the other hand, when
hydroxyl was replaced by a carboxamide function, the carbox-
amide provided a direct hydrogen-bond interaction with Lys227
without requiring an intervening water molecule, and the affinity
remained unchanged. In the case of ligand 3, the docked fluo-
roethyl chain took the place of the structured water molecules.
The fluorine atom was found to directly interact with carboxyl
oxygen of Lys227 and His291 (2.74 and 3.76 Å distances), but
these interactions probably were not as favorable as hydrogen
bonds. Thus, although modeling showed that ligand 3 fit well in
the JDTic binding pocket, the lack of strong hydrogen bonding
could explain the differences between the theoretical and
experimental affinities.
A fluoroalkyl chain could also been easily built on the iso-
quinoline nitrogen Ni of JDTic in the binding cleft with no drastic
steric clash with the KOR atoms. The Np-O
tances for ligands 1 and 2 remained close to that of JDTic
(Table 2). The N-fluoropropylJDTic 2 (
G ¼ ꢁ10.86 kcal/mol:
Ki ¼ 10.86 nM) was bound with a better score than the N-fluo-
roethylJDTic 1 (
G ¼ ꢁ10.43 kcal/mol; Ki ¼ 22.72 nM; Table 2).
d1 and Ni-Od2 dis-
D
D
This result was in agreement with the experimental in vitro af-
finities (Table 1). Ligand 1 with a shorter fluoroethyl chain was
bound slightly more loosely to KOR (Fig. 1E, Table 2) than the
fluoropropyl analogue 2. This difference was due to some steric
hindrance around the fluorine atom. Indeed, introduction of the
fluoroethyl chain led to positioning the fluorine atom too close to
KOR residues, leading to a slight shift toward a more loosely
bound position. With the longer fluoropropyl substituent, the
fluorine atom could nicely interact with the hydroxyl group of
Tyr312 side chain with a 3.84 Å distance between fluorine and
hydroxyl (Fig. 1F). This distance is higher (4.47 Å) with N-fluo-
roethyl JDTic 1 (Fig. 1E). Thus, the FeOH interaction for ligand 2
provided supplementary ligand-receptor interaction which could
Fig. 1. Best Docking Poses of Fluoroalkyl Derivatives 1e4 in the Ligand KOR Binding Pocket. (A) JDTic bound in its KOR binding pocket (KOR is shown with its solvent-accessible
surface). (B) Water-mediated interactions between JDTic isoquinoline hydroxyl and KOR residues (KOR is shown with its solvent-accessible surface in transparency). (C) O-Fluo-
ropropylJDTic 4 and JDTic for comparison. (D) O-FluoroethylJDTic 3 with the fluoroethyl chain taking the place of two water molecules. (E) N-FluoroethylJDTic 1. (F) O-Fluo-
ropropylJDTic 2. In all panels, JDTic, fluoroalkyl derivatives 1e4 and the selected KOR residues in interaction with the ligands are shown as capped sticks colored by atom type and
with carbon atoms colored green (KOR), cyan (JDTic) and purple (derivatives 1e4); the water molecules are shown as red spheres. (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article.)

S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
747
contribute to a better recognition with KOR. In addition, as
5.1.1. ((R)-2-(2-Fluoroethyl)-7-hydroxy-N-((S)-1-((3R,4R)-4-(3-
hydroxyphenyl)-3,4-dimethylpiperidin-1-yl)-3-methylbutan-2-yl)-
1,2,3,4-tetrahydroisoquinoline-3-carboxamide 1
A solution of JDTic (100 mg, 0.22 mmol) [36] and fluo-
roethyltosylate 5 (48 mg, 0.22 mmol) [55] in ACN (1 mL) was
refluxed for 72 h. The solvent was removed under reduced pressure
and the residue was purified by chromatography on silica gel (DCM/
MeOH from 98:2 to 95:5 v/v) to give the title compound as a yellow
Tyr312 is not conserved in the OR family (Trp in MOR and Leu in
DOR) [45,62], the lack of a hydroxyl group interacting with the
fluorine atom in MOR and DOR could explain the high selectivity
of ligand 2 for KOR.
4. Conclusion
In summary, fluoroalkylation on tetrahydroisoquinoline ring of
JDTic provided compounds with nanomolar affinity for KOR. Po-
tency was strongly dependent on the alkylation site and the length
of the fluoroalkyl chain. Alkylation of the isoquinoline nitrogen was
preferred to O-alkylation of the isoquinoline hydroxyl. The N-flu-
oropropyl derivative 2 was only 4-fold less affine than JDTic. The
fluoroethyl analogue 1 and the O-fluoroalkyl derivatives 3e4 were
found to be significantly less potent. Interestingly, N-fluoropropyl
ligand 2 displayed increased KOR selectivity relative to MOR and
DOR compared to JDTic. Modeling studies highlighted the contri-
bution of the N-fluoropropyl chain in the selective receptor binding
by revealing interaction between fluorine atom in 2 and Tyr312
hydroxyl from KOR. Thus, the overall results provided new insights
onto KOR recognition by JDTic-based compounds. Pharmacological
properties of 2 suggested that this ligand may be considered as a
potential candidate for radiolabelling with fluorine-18 for KOR PET
imaging, and may represent a valuable tool for further KOR
exploration.
solid (38 mg, 35%): m.p.: 116e118 ꢀC; [
a
]
D
25: þ81.2ꢀ (c 0.35, EtOH);
1H NMR (MeOD):
d
0.69 (d, J ¼ 6.9 Hz, 3H), 0.86e0.93 (m, 6H), 1.28
(s, 3H), 1.51e1.56 (m, 1H), 1.81e1.88 (m, 1H), 1.95e1.98 (m, 1H),
2.16e2.24 (m, 1H), 2.28e2.45 (m, 3H), 2.50e2.63 (m, 2H),
2.69e2.73 (m, 1H), 2.83e3.08 (m, 4H), 3.50 (t, J ¼ 6.3 Hz, 1H), 3.72
and 7.00 (AB, d, J ¼ 14.7 Hz, 2H), 3.87e3.93 (m, 1H), 4.50e4.53 (m,
1H), 4.67e4.70 (m, 1H), 6.57e6.64 (m, 3H), 6.72e6.77 (m, 2H), 6.94
(d, J ¼ 8.1 Hz, 1H), 7.11 (t, J ¼ 7.8 Hz, 1H); 13C NMR (MeOD):
d 16.5,
17.7, 19.8, 27.9, 29.5, 32.0, 39.2, 39.9, 51.7, 52.6, 53.4, 55.8 (d,
JCeF ¼ 19.0 Hz), 56.9, 61.9, 64.2, 83.1 (d, JCeF ¼ 165.7 Hz), 113.4, 113.7,
114.1, 115.1, 117.9, 126.0, 129.8, 130.1, 137.3, 154.4, 157.1, 158.3, 173.5;
IR (neat):
n
3336, 1667, 1615, 1505, 1454; MS (ESIþ): m/z 512.4
([MþH]þ); HRMS (ESIþ) Calcd for C30H42FN3O3 (Mþ): 511.3210;
Found: 511.3245. Anal. Calcd. for C30H42FN3O3$H2O: C, 68.03%; H,
8.37%; N, 7.93%; Found: C, 68.46%; H, 8.62%; N, 8.13%.
5.1.2. (3R)-2-(3-Fluoropropyl)-1,2,3,4-tetrahydro-7-hydroxy-N-
((S)-1-((3R,4R)-4-(3-hydroxyphenyl)-3,4-dimethylpiperidin-1-yl)-
3-methylbutan-2-yl)isoquinoline-3-carboxamide 2
5. Experimental
A solution of JDTic (100 mg, 0.22 mmol) [36] and fluo-
ropropyltosylate 6 (48 mg, 0.22 mmol) [56] in ACN (1 mL) was
refluxed for 72 h. The solvent was removed under reduced pressure
and the residue purified by chromatography on silica gel (DCM/
MeOH from 98/2 to 95/5) to give the title compound as a yellow
5.1. Chemistry
All reactions were performed under anhydrous conditions and
an atmosphere of nitrogen. All reagents were purchased from
Acros Organics, Fluka or SigmaeAldrich and were used as
commercially supplied. Anhydrous THF, DCM and ACN were ob-
tained from a Mbraun SPS-800 solvents delivery system. HPLC
solvents were purchased from Merck, SigmaeAldrich or SDS. 1H
and 13C NMR were recorded on Brucker DPX 300 at 300.0 MHz
(1H), 75.5 MHz (13C) and 282.4 MHz (19F) or on Brucker DPX 400 at
400.0 MHz (1H), 100.6 MHz (13C) and 376.4 MHz (19F). Samples
were dissolved in an appropriate deuterated solvent (CDCl3,
solid (42 mg, 37%): m.p.: 124e126 ꢀC; [
a
]
25: þ87.7ꢀ (c 0.40, CH3Cl);
D
1H NMR (MeOD):
d
0.72 (d, J ¼ 7.0 Hz, 3H), 0.88e0.94 (m, 6H), 1.29
(s, 3H), 1.59 (d, J ¼ 13.0 Hz, 1H), 1.87e2.00 (m, 4H), 2.23 (dt, J ¼ 4.3
and 13.0 Hz, 1H), 2.45e2.49 (m, 3H), 2.63e2.80 (m, 5H), 2.97e2.99
(m, 2H), 3.43 (t, J ¼ 6.0 Hz,1H), 3.59 and 3.99 (AB, d, J ¼ 14.8 Hz, 2H),
3.90e3.93 (m, 1H), 4.47 (t, J ¼ 5.6 Hz, 1H), 4.59 (t, J ¼ 5.6 Hz, 1H),
6.60e6.65 (m, 3H), 6.74e6.78 (m, 2H), 6.94 (d, J ¼ 8.1 Hz,1H), 7.11 (t,
J ¼ 7.9 Hz, 1H); 13C NMR (MeOD):
d 16.5, 17.9, 19.8, 27.9, 29.7 (d,
JCeF
¼
19.7 Hz), 30.2, 31.5, 39.2, 39.9, 49.9, 51.7, 52.7 (d,
MeOD). Chemical shifts (
d
) are quoted in parts per million (ppm)
JCeF ¼ 11.2 Hz), 56.9, 61.6, 64.3, 83.2 (d, JCeF ¼ 164.0 Hz), 113.4, 113.7,
and referenced to proton resonances resulting from incomplete
deuteration of the NMR solvent. Coupling constants (J) are given in
Hz. Coupling patterns are abbreviated as follows: s (singlet),
d (doublet), t (triplet), q (quadruplet), m (mutiplet), dt (doublet of
triplet), qd (quadruplet of doublet). MS and high resolution mass
spectra (HRMS) were recorded using a Waters Q-TOF micro spec-
trometer by electrospray ionisation (ESI). Relative intensities are
given in brackets. Infrared spectra were recorded on a Thermo
Nicolet 350 FT-IR ATR spectrometer. Only selected absorbances are
114.0, 115.1, 117.9, 125.8, 129.7, 130.2, 137.4, 152.7, 157.0, 158.3, 175.9;
IR (neat):
n
3249, 1644, 1614, 1503, 1446. MS (ESIþ): m/z 526.4
([MþH]þ); HRMS (ESIþ) Calcd for C31H45FN3O3 ([MþH]þ):
526.3445; Found: 526.3426. Anal. Calcd. for C31H44FN3O3$H2O: C,
68.48%; H, 8.53%; N, 7.73%; Found: C, 68.85%; H, 8.95%; N, 7.77%.
5.1.3. (R)-2-tert-Butyl-3-benzyl-3,4-dihydro-7-
hydroxyisoquinoline-2,3(1H)-dicarboxylate 13
To
a
solution
of
N-Boc-D-7-hydroxy-1,2,3,4-
reported (
n
in cmꢁ1). Melting points were determined on a Barn-
tetrahydroisoquinoline-3-carboxylic acid (N-Boc-D-Tic) 12 (1.90 g,
9.8 mmol) [57] and NaHCO3 (940 mg, 19.6 mmol) in DMF (80 mL)
was added dropwise benzyl bromide (5.80 mL, 49.2 mmol). The
reaction mixture was stirred at room temperature for 24 h and ice
cooled water was added. After extraction with EtOAc, the organic
fraction was washed with water, dried over MgSO4, filtered and
concentrated under reduced pressure. The residue was purified by
chromatography on silica gel (pentane/diethyl ether from 100:0 to
70:30 v/v) to give the title compound as a white solid (1.10 g, 38%):
stead Electrothermal IA 9100 Mp apparatus and are uncorrected.
Optical rotations were measured with a PerkineElmer 341 polar-
imeter. Elemental analyses were performed on a ThermoQuest
analyser CHNS and were within 0.4% of the calculated values.
Flash column chromatography was carried out on silica gel (Merck
Kieselgel 60 F254, 40e63
raphy (TLC) was performed on Merck aluminium-backed plates
pre-coated with silica (0.2 mm, 60 F254) which were visualized by
mm). Analytical thin layer chromatog-
quenching of ultraviolet fluorescence (
l
¼ 254 and 366 nm) or by
m.p.: 115e116 ꢀC; 1H NMR (MeOD) (rotamers):
d 1.39 (s, 4.5H), 1.50
ninhydrin, KMnO4, vanillin or phosphomolybdic acid hydrate spray
reagent. High Performance Liquid Chromatography (HPLC) was
carried out by a Merck L-6200 pump and a Merck L-4250 UVevi-
sible detector.
(s, 4.5H), 3.02e3.19 (m, 2H), 4.38e4.57 (m, 2H), 4.83 (t, J ¼ 5.0 Hz,
0.5H), 4.98e5.07 (m, 2.5H), 6.52e6.53 (m, 0.5H), 6.58e6.62 (m,
1.5H), 6.90e6.94 (m, 1H), 7.08e7.15 (m, 2H), 7.25e7.29 (m, 3H); 13
C
NMR (MeOD) (rotamers):
d 28.5, 28.7, 31.7, 31.9, 45.4, 46.1, 54.9,

748
S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
56.3, 67.8, 82.0, 113.6, 113.7, 115.0, 115.2, 123.8, 124.2, 128.7, 129.1,
129.2, 129.4, 129.5, 129.9, 130.4, 135.2, 135.8, 137.2, 156.8, 157.3,
pressure. The residue was purified by chromatography on silica gel
(hexane/diethyl ether from 100:0 to 0:100) to give the title com-
157.5, 157.6, 172.9, 173.3; IR (neat):
n
2978, 1722, 1687, 1502, 1475;
pound as a white solid (110 mg, 60%): m.p.: 60e62 ꢀC; [
a
]
25: ꢁ4.2ꢀ
D
MS (ESIþ): m/z 406.2 ([MþNa]þ); HRMS (ESIþ) Calcd for
(c 1.00, CHCl3); 1H NMR (CDCl3) (rotamers):
d
1.41e1.51 (m, 9H),
C
22H25NNaO5 ([MþNa]þ): 406.1630; Found: 406.1632. Anal. Calcd.
3.03e3.22 (m, 2H), 4.11e4.14 (m, 1H), 4.21e4.23 (m, 1H), 4.43 (d,
J ¼ 16.8 Hz, 1H), 4.60e4.67 (m, 2H), 4.75e4.82 (m, 1.5H), 5.09 (m,
0.5H), 6.66e6.77 (m, 2H), 7.06 (d, J ¼ 8.2 Hz, 1H); 13C NMR (CDCl3)
for C22H25NO5: C, 68.91%; H, 6.57%; N, 3.65%; Found: C, 68.98%; H,
7.07%; N, 3.82%.
(rotamers):
d 28.2, 28.4, 30.1, 30.6, 44.1, 44.7, 52.5, 54.3, 67.1 (d,
5.1.4. (R)-3-Benzyl-2-tert-butyl-7-(2-fluoroethoxy)-3,4-
dihydroisoquinoline-2,3(1H)-dicarboxylate 14
JCeF ¼ 20.4 Hz), 81.0, 81.9 (d, JCeF ¼ 169.9 Hz), 111.9, 112.1, 113.7,
124.4,124.5,128.9,129.6,133.9,135.0,154.8,155.7,157.3,157.4,177.1,
A
solution of phenol 13 (300 mg, 0.78 mmol), fluo-
177.8; IR (neat): n
2976, 1695, 1615, 1504, 1475; MS (ESIꢁ): m/z 338.1
roethyltosylate 5 (205 mg, 0.94 mmol) [55] and K2CO3 (204 mg,
1.48 mmol) in acetone (6 mL) was refluxed for 48 h. The solution
was cooled to room temperature and a saturated aqueous solution
of NH4Cl was added. After extraction with EtOAc, the organic
fraction was dried over MgSO4, filtered and concentrated under
reduced pressure. The residue was purified by chromatography on
silica gel (hexane/diethyl ether from 100:0 to 60:40) to give the title
compound as a white solid (260 mg, 78%): m.p.: 64e65 ꢀC; 1H NMR
([MꢁH]ꢁ).
5.1.7. (R)-2-(tert-Butoxycarbonyl)-7-(3-fluoropropoxy)-1,2,3,4-
tetrahydroisoquinoline-3-carboxylic acid 11
To a solution of benzyl ester 15 (220 mg, 0.50 mmol) in EtOH
(5 mL) was added dropwise a solution of KOH (114 mg, 2.03 mmol)
in water (5 mL). The solution was vigorously stirred for 6 h at room
temperature, and EtOH was removed under reduced pressure. The
aqueous solution was acidified to pH 6 with dilute HCl. After
extraction with DCM, the organic fraction was washed with water,
dried over MgSO4, and concentrated under reduced pressure. The
residue was purified by chromatography on silica gel (hexane/
diethyl ether from 100:0 to 60:40) to give the title compound as a
(CDCl3) (rotamers): d 1.40 (s, 4.5H),1.51 (s, 4.5H), 3.05e3.26 (m, 2H),
4.11e4.14 (m, 1H), 4.21e4.23 (m, 1H), 4.47 (AB, d, J ¼ 16.5 Hz, 1H),
4.61e4.68 (m, 2H), 4.80e4.83 (m, 1.5H), 4.98e5.09 (m, 2H), 5.18 (m,
0.5H), 6.66 (m, 1H), 6.74 (m, 1H), 7.02 (d, J ¼ 8.4 Hz, 1H), 7.14 (m,
2H), 7.27e7.30 (m, 3H); 13C NMR (CDCl3) (rotamers):
d 28.3, 28.4,
30.6, 31.0, 44.4, 44.9, 52.9, 54.6, 66.7, 67.2 (d, JCeF ¼ 20.3 Hz), 80.8,
81.8 (d, JCeF ¼ 180.5 Hz), 111.8, 112.1, 113.7, 124.6, 124.7, 127.7, 128.0,
128.2, 128.4, 128.5, 129.0, 129.6, 130.0, 134.1, 135.1, 135.5, 135.7,
white solid (120 mg, 70%): m.p. 70e73 ꢀC; [
a
]
25: ꢁ13.1ꢀ (c 0.82,
D
CHCl3); 1H NMR (CDCl3) (rotamers):
d
1.41 (s, 4.5H), 1.50 (s, 4.5H),
2.14 (qd, J ¼ 26.1 and 6.0, 2H), 3.02e3.21 (m, 2H), 4.05 (t, J ¼ 6.0 Hz,
2H), 4.39e4.47 (m, 1H), 4.62e4.75 (m, 3.5H), 5.07e5.09 (m, 0.5H),
6.64e6.75 (m, 2H), 7.04 (d, J ¼ 8.1 Hz, 1H); 13C NMR (CDCl3)
154.8, 155.4, 157.4, 171.4, 171.8; IR (neat):
n 2978, 1722, 1687, 1502,
1475; MS (ESIþ) m/z 452.2 ([MþNa]þ); HRMS (ESIþ) Calcd for
C
24H28FNNaO5 ([MþNa]þ): 452.1849; Found: 452.1837.
(rotamers):
d
28.2, 28.4, 30.1, 30.3 (d, JCeF ¼ 11.0 Hz), 44.1, 44.8, 63.5,
63.6, 68.4, 80.7 (d, JCeF ¼ 163.4 Hz), 80.9, 111.7, 112.0, 113.6, 124.1,
5.1.5. (R)-3-Benzyl-2-tert-butyl-7-(3-fluoropropoxy)-3,4-
dihydroisoquinoline-2,3(1H)-dicarboxylate 15
128.9, 129.5, 134.9, 154.8, 157.7, 177.2; IR (neat):
n 2974, 1695, 1615,
1504, 1475; MS (ESIꢁ): m/z 352.1 ([MꢁH]ꢁ).
A
solution of phenol 13 (250 mg, 0.65 mmol), fluo-
ropropyltosylate 6 (180 mg, 0.77 mmol) [56] and K2CO3 (172 mg,
1.24 mmol) in acetone (5 mL) was refluxed for 48 h, cooled to room
temperature and a saturated aqueous solution of NH4Cl was added.
After extraction with EtOAc, the organic fraction was dried over
MgSO4, filtered and concentrated under reduced pressure. The
residue was purified by chromatography on silica gel (hexane/
diethyl ether from 100:0 to 70:30) to give the title compound as a
5.1.8. (3R)-7-(2-Fluoroethoxy)-1,2,3,4-tetrahydro-N-((S)-1-
((3R,4R)-4-(3-hydroxyphenyl)-3,4-dimethylpiperidin-1-yl)-3-
methylbutan-2-yl)isoquinoline-3-carboxamide 3
To a solution of carboxylic acid 10 (110 mg, 0.32 mmol), amine
9 (94 mg, 0.32 mmol) [36] and Et3N (151 mL, 1.12 mmol) in THF
(20 mL) was added BOP (166 mg, 0.37 mmol). The mixture was
stirred at room temperature for 3 h and diluted in diethyl ether.
The solution was washed with a saturated aqueous solution of
NaHCO3 and water, dried over MgSO4, filtered and concentrated
under reduced pressure. The residue was purified by chromatog-
raphy on silica gel (DCM/MeOH from 100:0 to 80:20). The residue
was dissolved in DCM (2 mL) and TFA (0.75 mL) was added
dropwise at ꢁ20 ꢀC. The reaction mixture was stirred at ꢁ20 ꢀC
for 10 min, and then at room temperature for 1 h. The mixture
was washed with a saturated aqueous solution of NaHCO3 and
water, dried over MgSO4, filtered and concentrated under reduced
pressure. The residue was purified by chromatography on silica gel
(DCM/MeOH from 100:0 to 80:20) to give the title compound as a
white solid (245 mg, 85%); m.p.: 70e73 ꢀC; [
a
]
25: þ6.5ꢀ (c 1.13,
D
CHCl3); 1H NMR (CDCl3) (rotamers):
d
1.40 (s, 4.5H), 1.51 (s, 4.5H),
2.14 (m, 2H), 3.04e3.24 (m, 2H), 4.06 (t, J ¼ 6.0 Hz, 3H), 4.48 and
4.62 (AB, d, J ¼ 16.2 Hz, 2H), 4.65 (dt, J ¼ 41.3 and 6.0 Hz, 2H), 4.82 (t,
J ¼ 5.1 Hz, 0.5H), 4.99e5.09 (m, 2H), 5.19 (m, 0.5H), 6.62e6.74 (m,
2H), 7.00 (d, J ¼ 8.4 Hz, 1H), 7.14 (m, 2H), 7.28 (m, 4H); 13C NMR
(CDCl3) (rotamers):
d
28.3, 28.4, 30.3, 30.6 (d, JCeF ¼ 20.3 Hz), 44.4,
44.9, 52.9, 54.6, 63.5 (d, JCeF ¼ 4.8 Hz), 66.7, 80.6, 80.7 (d,
JCeF ¼ 163.4 Hz), 111.6, 112.1, 113.4, 113.5, 124.1, 124.3, 127.7, 127.9,
128.0, 128.2, 128.4, 128.5, 128.9, 129.5, 134.0, 135.1, 135.6, 135.7,
154.8, 155.5, 157.6, 157.7, 171.4, 171.8; IR (neat):
n 2978, 1723, 1689,
1585, 1502; MS (ESIþ) m/z 466.2 ([MþNa]þ); Anal. Calcd. for
C
25H30FNO5: C, 67.70%; H, 6.82%; N, 3.16%; Found: C, 67.97%; H,
yellow solid (11 mg, 6%); m.p. 113e116 ꢀC; [
a
]
25: þ79.2ꢀ (c 0.25,
D
7.16%; N, 3.39%.
EtOH); 1H NMR (CDCl3):
d
0.77 (d, J ¼ 7.1 Hz, 3H), 0.95e0.99 (m,
6H), 1.34 (s, 3H), 1.59e1.71 (m, 2H), 1.86e1.93 (m, 1H), 2.06e2.16
(m, 1H), 2.32e2.39 (m, 1H), 2.76e3.07 (m, 7H), 3.65e3.69 (m, 2H),
4.02e4.07 (m, 4H), 4.52e4.55 (m, 1H), 4.63e4.66 (m, 1H),
5.1.6. (R)-2-(tert-Butoxycarbonyl)-7-(2-fluoroethoxy)-1,2,3,4-
tetrahydroisoquinoline-3-carboxylic acid 10
To a solution of benzyl ester 14 (250 mg, 0.58 mmol) in EtOH
(5 mL) was added dropwise a solution of KOH (134 mg, 2.39 mmol)
in water (5 mL). The reaction mixture was vigorously stirred at
room temperature for 6 h, and EtOH was removed under reduced
pressure. The aqueous solution was acidified to pH 6 with dilute
HCl. After extraction with DCM, the organic fraction was washed
with water, dried over MgSO4, and concentrated under reduced
6.59e6.77 (m, 5H), 7.04e7.11 (m, 2H); 13C NMR (CDCl3):
d
17.9,
19.2, 30.3, 30.9, 38.2, 38.5, 55.9, 56.8, 59.2, 67.1, 80.8, 111.7, 112.8,
113.4, 117.2, 126.9, 129.2, 130.3, 156.7, 175.0; IR (neat):
n 3341, 1677,
1618, 1503, 1433; MS (ESIþ): m/z 512.4 ([MþH]þ); HRMS (ESIþ)
Calcd for C30H43FN3O3 ([MþH]þ): 512.3288; Found: 512.3281;
Anal. Calcd. for C30H42FN3O3$H2O: C, 68.03%; H, 8.37%; N, 7.93%;
Found: C, 68.41%; H, 8.58%; N, 8.26%.

S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
749
5.1.9. (3R)-7-(3-Fluoropropoxy)-1,2,3,4-tetrahydro-N-((S)-1-
Acknowledgments
((3R,4R)-4-(3-hydroxyphenyl)-3,4-dimethylpiperidin-1-yl)-3-
methylbutan-2-yl)isoquinoline-3-carboxamide 4
ꢀ
This research was supported by the “Region Basse-Normandie”
(“IMADEP” project 12P03303) and the European Regional Devel-
opment Fund (ERDF, ISCE-Chem & INTERREG Iva program 4061
grant). The authors thank the CRIHAN (Centre de Ressources
Informatiques de Haute Normandie), the European Community
(FEDER) for the molecular modeling software Discovery Studio
(Accelrys, San Diego, CA, USA) and Jana Sopkova-de Oliveira Santos
(CERMN, Caen, France) for her help in building compounds with
Discovery Studio.
To a solution of carboxylic acid 11 (120 mg, 0.34 mmol), amine 9
(100 mg, 0.34 mmol) [36] and Et3N (161
mL, 1.16 mmol) in THF
(20 mL) was added BOP (176 mg, 0.39 mmol). The mixture was
stirred at room temperature for 3 h and diluted with diethyl ether.
The mixture was washed with a saturated aqueous solution of
NaHCO3 and water, dried over MgSO4, filtered and concentrated
under reduced pressure. The residue was purified by chromatog-
raphy on silica gel (DCM/MeOH from 100:0 to 80:20). The residue
was dissolved in DCM (2 mL) and TFA (0.75 mL) was added drop-
wise at ꢁ20 ꢀC. The reaction mixture was stirred at ꢁ20 ꢀC for
10 min, and then at room temperature for 1 h. The mixture was
washed with a saturated aqueous solution of NaHCO3 and water,
dried over MgSO4, filtered and concentrated under reduced pres-
sure. The residue was purified by chromatography on silica gel
(DCM/MeOH from 100:0 to 80:20) to give the title compound as a
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.12.016.
References
yellow solid (12 mg, 7%): m.p. 105e107 ꢀC; [
a
]
25: þ82.7ꢀ (c 0.25,
D
EtOH); 1H NMR (CDCl3):
d
0.75e0.80 (m, 3H), 0.94e1.08 (m, 6H),
[1] B.N. Dhawan, F. Cesselin, R. Raghubir, T. Reisine, P.B. Bradley, P.S. Porthoghese,
M. Hamon, Pharmacol. Rev. 48 (1996) 567e592.
[2] M. Minami, M. Satoh, Neurosci. Res. 23 (1995) 121e145.
[3] M. Waldhoer, S.E. Bartlett, J.L. Whistler, Annu. Rev. Biochem. 73 (2004)
953e990.
[4] J.M. van Ree, M.A. Gerrits, L.J. Vanderschuren, Pharmacol. Rev. 51 (1999)
341e396.
[5] P.E. Lutz, B.L. Kieffer, Trends Neurosci. 36 (2013) 195e206.
[6] J. Le Merrer, J.A.J. Becker, K. Befort, B.L. Kieffer, Physiol. Rev. 89 (2009)
1379e1412.
1.33 (s, 3H), 1.43e1.47 (m, 1H), 1.61e2.10 (m, 5H), 2.28e2.46 (m,
2H), 2.59e3.16 (m, 6H), 3.58e3.78 (m, 3H), 4.01e4.20 (m, 4H),
4.61e4.62 (m, 1H), 4.73e4.74 (m, 1H), 6.58e6.60 (m, 1H), 6.66e6.77
(m, 4H), 7.04e7.13 (m, 2H); 13C NMR (CDCl3):
d 16.1, 16.6, 17.9, 19.9,
27.9, 28.8, 31.4, 32.3, 39.2, 39.8, 46.1, 46.7, 51.5, 52.1, 56.3, 61.2, 64.8,
64.9, 80.6, 113.0, 113.5, 114.7, 117.9, 126.3, 130.1, 136.2, 151.9, 157.2,
158.4, 159.3, 174.5; IR (neat):
n 3339, 1675, 1618, 1510, 1431. MS
(ESIþ) m/z 526.4 ([MþH]þ); HRMS (ESIþ) Calcd for C31H45FN3O3
[7] J.M. Hiller, L.Q. Fan, Neurochem. Res. 21 (1996) 1333e1345.
[8] D. Peckys, G.B. Landwehrmeyer, Neuroscience 88 (1999) 1093e1135.
[9] F. Simonin, C. Gaveriaux-Ruff, K. Befort, H. Matthes, B. Lannes, G. Micheletti,
([MþH]þ): 526.3445; Found: 526.3437. Anal. Calcd. for
ꢀ
M.G. Mattei, G. Charron, B. Bloch, B. Kieffer, Proc. Natl. Acad. Sci. U. S. A. 92
C
31H44FN3O3$H2O: C, 68.48%; H, 8.53%; N, 7.73%; Found: C, 68.88%;
(1995) 7006e7010.
H, 8.97%; N, 7.98%.
[10] J. Peng, S. Sarkar, L. Chang, Drug Alcohol Depend. 124 (2012) 223e228.
[11] A.M. Mathieu-Kia, L.Q. Fan, M.J. Kreek, E.J. Simon, J.M. Hiller, Brain Res. 893
(2001) 121e134.
[12] H.A. Tejeda, T.S. Shippenberg, R. Henriksson, Cell Mol. Life Sci. 69 (2012)
857e896.
5.2. OR binding assays
[13] A. Van't Veer, W.A. Carlezon, Psychopharmacology 229 (2013) 435e452.
[14] Y. Shirayama, H. Ishida, M. Iwata, G.I. Hazama, R. Kawahara, R.S.J. Duman, J.
Neurochem. 90 (2004) 1258e1268.
[15] B.B. Land, M.R. Bruchas, J.C. Lemos, M. Xu, E.J. Melief, C.J. Chavkin, Neurosci-
ence 28 (2008) 407e414.
[16] R. Przewłocki, B. Przewłocka, Eur. J. Pharmacol. 429 (2001) 79e91.
[17] A.T. Knoll, E.G. Meloni, J.B. Thomas, F.I. Caroll, W.A. Carlezon Jr., J. Pharmacol.
Exp. Ther. 323 (2007) 838e845.
Binding affinities for KOR, MOR and DOR were determined by
radioligand displacement experiments with specific KOR [3H]
U69593 (1 nM), MOR [3H]diprenorphine (0.4 nM) and DOR [3H]
DADLE (0.5 nM) ligands in rat recombinant KOR CHO cells [58]
and human recombinant MOR [59] or DOR [60] HEK cells ac-
cording to published procedures. Non-specific binding was
defined as the binding of radioligand in the presence of naltrexone
[18] J.V. Aldrich, J.P. McLaughlin, AAPS J. 11 (2009) 312e322.
ꢀ
[19] W.A. Carlezon, C. Beguin, A.T. Knoll, B.M. Cohen, Pharmacol. Ther. 123 (2009)
(10 mM for KOR and DOR, 1 mM for MOR). Assays were first vali-
334e343.
dated with U50488, naltrexone and DPDPE used as reference li-
gands of KOR, MOR and DOR respectively. Eight different
concentrations of each compounds 1e4, JDTic and reference li-
gands were tested. All assays were conducted in 50 mM Tris buffer
(pH 7.4) at room temperature for 60 min for KOR binding and for
120 min for MOR and DOR binding. All experiments were per-
formed in triplicate.
[20] B.M. Walker, G.F. Koob, Neuropsychopharmacology 33 (2008) 643e652.
[21] F.I. Carroll, W.A. Carlezon Jr., J. Med. Chem. 56 (2013) 2178e2195.
[22] M. Urbano, M. Guerrero, H. Rosen, E. Roberts, Bioorg. Med. Chem. Lett. 24
(2014) 2021e2032.
[23] P.S. Portoghese, A.W. Lipkowski, A.E. Takemorif, J. Med. Chem. 30 (1987)
238e239.
[24] W.C. Stevens, R.M. Jones Jr., G. Subramanian, T.G. Metzger, D.M. Ferguson,
P.S. Portoghese, J. Med. Chem. 43 (2000) 2759e2769.
[25] R.M. Jones, P.S. Portoghese, Eur. J. Pharmacol. 396 (2000) 49e52.
[26] H. Zhang, Y.-G. Shi, J.H. Woods, S.J. Watson, M.-C. Ko, Eur. J. Pharmacol. 570
(2007) 89e96.
5.3. Docking studies
[27] S. Kishioka, N. Kiguchi, Y. Kobayashi, C. Yamamoto, F. Saika, N. Wakida, M.-
C. Ko, J.H. Woods, Neurosci. Lett. 552 (2013) 98e102.
[28] T.A. Munro, L.M. Berry, A. Van't Veer, C. Beguin, F.I. Carroll, Z. Zhao,
ꢀ
The molecular modeling studies were performed using the
published crystal structure of KOR in complex with JDTic (Protein
Data Bank code 4DJH) [45]. Discovery Studio 3.5 suite (Accelrys, San
Diego, CA, USA) was used to add hydrogen atoms and to build the
fluoroalkyl compounds 1e4 on the JDTic structure. An automatic
docking procedure has been performed for each of the four fluo-
roalkyl compounds, with AutoDock4 [61], using a flexible ligand, a
rigid receptor and a Lamarckian genetic algorithm with the default
setting parameters. For each compound, the best scoring ligand
with the lower estimated free energy and the best estimated in-
hibition constant was chosen. Fig. 1 was prepared using PYMOL
(DeLano Scientific, CA, USA).
W.A. Carlezon Jr., B.M. Cohen, BMC Pharmacol. 12 (2012) 5.
[29] E.J. Melief, M. Miyatake, F.I. Carroll, C. Beguin, W.A. Carlezon Jr., B.M. Cohen,
ꢀ
S. Grimwood, C.H. Mitch, L. Rorick-Kehn, C. Chavkin, Mol. Pharmacol. 80
(2011) 920e929.
[30] C.H. Mitch, S.J. Quimby, N. Diaz, C. Pedregal, M.G. de la Torre, A. Jimenez,
Q. Shi, E.J. Canada, S.D. Kahl, M.A. Statnick, D.L. McKinzie, D.R. Benesh,
K.S. Rash, V.N. Barth, J. Med. Chem. 54 (2011) 8000e8012.
[31] L.M. Rorick-Kehn, J.M. Witkin, M.A. Statnick, E.L. Eberle, J.H. McKinzie,
S.D. Kahl, B.M. Forster, C.J. Wong, X. Li, R.S. Crile, D.B. Shaw, A.E. Sahr,
B.L. Adams, S.J. Quimby, N. Diaz, A. Jimenez, C. Pedregal, C.H. Mitch,
K.L. Knopp, W.H. Anderson, J.W. Cramer, D.L. McKinzie, Neuropharmacology
77 (2014) 131e144.
[32] http://clinicaltrials.gov/ct2/show/NCT01913535.
[33] M.-Q. Zheng, N. Nabulsi, S.J. Kim, G. Tomasi, S.-F. Lin, C. Mitch, S. Quimby,
V. Barth, K. Rash, J. Masters, A. Navarro, E. Seest, E.D. Morris, R.E. Carson,

750
S. Schmitt et al. / European Journal of Medicinal Chemistry 90 (2015) 742e750
Y. Huang, J. Nucl. Med. 54 (2013) 455e463.
[48] J.B. Thomas, S.E. Fix, R.B. Rothman, S.W. Mascarella, C.M. Dersch, B.E. Cantrell,
D.M. Zimmerman, F.I. Carroll, J. Med. Chem. 47 (2004) 1070e1073.
[49] S.P. Runyon, L.E. Brieaddy, S.W. Mascarella, J.B. Thomas, H.A. Navarro,
J.L. Howard, G.T. Pollard, F.I. Carroll, J. Med. Chem. 53 (2010) 5290e5301.
[50] T.B. Cai, Z. Zou, J.B. Thomas, L. Brieaddy, H.A. Navarro, F.I. Carroll, J. Med. Chem.
51 (2008) 1849e1860.
[34] S.J. Kim, M.-Q. Zheng, N. Nabulsi, D. Labaree, J. Ropchan, S. Najafzadeh,
R.E. Carson, Y. Huang, E.D. Morris, J. Nucl. Med. 54 (2013) 1668e1674.
[35] J.B. Thomas, R.N. Atkinson, R.B. Rothman, S.E. Fix, S.W. Mascarella, N.A. Vinson,
H. Xu, C.M. Dersch, Y. Lu, B.E. Cantrell, D.M. Zimmerman, F.I. Carroll, J. Med.
Chem. 44 (2001) 2687e2690.
[36] J.B. Thomas, R.N. Atkinson, N.A. Vinson, J.L. Catanzaro, C.L. Perretta, S.E. Fix,
S.W. Mascarella, R.B. Rothman, H. Xu, C.M. Dersch, B.E. Cantrell,
D.M. Zimmerman, F.I. Carroll, J. Med. Chem. 46 (2003), 3127e3131-37.
[37] I. Carroll, J.B. Thomas, L.A. Dykstra, A.L. Granger, R.M. Allen, J.L. Howard,
G.T. Pollard, M.D. Aceto, L.S. Harris, Eur. J. Pharmacol. 501 (2004) 111e119.
[38] T.A. Munro, X.-P. Huang, C. Inglese, M.G. Perrone, A. Van’t Veer, F.I. Carroll,
C. Beguin, W.A. Carlezon Jr., N.A. Colabufo, B.M. Cohen, B.L. Roth, PLoS One 8
(2013) e70701.
[39] P.M. Beardsley, J.L. Howard, K.L. Shelton, F.I. Carroll, Psychopharmacology 183
(2005) 118e126.
[51] C.M. Kormos, M.G. Gichinga, R. Maitra, S.P. Runyon, J.B. Thomas, L.E. Brieaddy,
S.W. Mascarella, H.A. Navarro, F.I. Carroll, J. Med. Chem. 57 (2014) 7367e7381.
[52] G. Poisnel, F. Oueslati, M. Dhilly, J. Delamare, C. Perrio, D. Debruyne, L. Barre,
Nucl. Med. Biol. 35 (2008) 561e569.
[53] S. Purser, P.R. Moore, S. Swallow, V. Gouverneur, Chem. Soc. Rev. 37 (2008)
320e330.
[54] K. Müller, C. Faeh, F. Diederich, Science 317 (2007) 1881e1886.
[55] Y.Z. Ponce, S. Mavel, T. Assaad, S.E. Kruse, S.M. Parsons, P. Emond, S. Chalon,
N. Giboureau, M. Kassiou, D. Guilloteau, Bioorg. Med. Chem. 13 (2005)
745e753.
ꢀ
ꢀ
[40] J.R. Schank, A.L. Goldstein, K.E. Rowe1, C.E. King, J.A. Marusich, J.L. Wiley,
F.I. Carroll, A. Thorsell, M. Heilig, Addict. Biol. 17 (2012) 634e647.
[41] F.I. Carroll, L.S. Harris, M.D. Aceto, Eur. J. Pharmacol. 524 (2005) 89e94.
[42] K.J. Jackson, F.I. Carroll, S.S. Negus, M.I. Damaj, Psychopharmacology 210
(2010) 285e294.
[56] T.W. Bell, H.-J. Choi, W. Harte, M.G.B. Drew, J. Am. Chem. Soc. 125 (2003)
12196e12210.
[57] C. Liu, J.B. Thomas, L. Brieaddy, B. Berrang, F.I. Carroll, Synthesis 6 (2008)
856e858.
[58] F. Meng, G.-X. Xie, R.C. Thompson, R.C. Thompson, A. Mansour, A. Goldstein,
S.J. Watson, H. Akil, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 9954e9958.
[59] S. Zhang, Y. Tong, M. Tian, R.N. Dehaven, L. Cortesburgos, E. Mansson,
F. Simonin, B. Kieffer, L. Yu, J. Pharmacol. Exp. Ther. 286 (1998) 136e141.
[43] http://clinicaltrials.gov/ct2/results?term¼JDTic.
[44] http://www.mclean.harvard.edu/kappa-therapeutics-2013/pdf/Kappa-
Therapeutics-2013-Complete-Program.pdf.
ꢀ
[45] H. Wu, D. Wacker, M. Mileni, V. Katritch, G.W. Han, E. Vardy, W. Liu,
A.A. Thompson, X.-P. Huang, F.I. Carroll, S.W. Mascarella, R.B. Westkaemper,
P.D. Mosier, B.L. Roth, V. Cherezov, R.C. Stevens, Nature 485 (2012) 327e332.
[46] D.M. Zimmerman, R. Nickander, J.S. Horng, D.T. Wong, Nature 275 (1978)
332e334.
[60] F. Simonin, K. Befort, C. Gaveriaux-Ruff, H. Matthes, V. Nappey, B. Lannes,
G. Micheletti, B. Kieffer, Mol. Pharmacol. 46 (1994) 1015e1021.
[61] G.M. Moris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
A.J. Olson, J. Comput. Chem. 16 (2009) 2785e2791.
[62] K. Martinez-Mayorga, K.G. Byler, A.B. Yongye, M.A. Giulianotti, C.T. Dooley,
R.A. Houghten, Eur. J. Med. Chem. 66 (2013) 114e121.
[47] J.P. Cueva, T.B. Cai, W.S. Mascarella, J.B. Thomas, H.A. Navarro, F.I. Carroll,
J. Med. Chem. 52 (2009) 7463e7472.