N-Guanidyl and C-Tetrazole Leu-Enkephalin Derivatives: Efficient Mu and Delta Opioid Receptor Agonists with Improved Pharmacological Properties

Article abstract of DOI:10.1021/acschemneuro.8b00550

Leu-enkephalin and d-Ala₂-Leu-enkephalin were modified at their N- and C-termini with guanidyl and tetrazole groups. The resulting molecules were prepared in solution or by solid phase peptide synthesis. The affinity of the different analogues at mu (MOP) and delta opioid receptors (DOP) was then assessed by competitive binding in stably transfected DOP and MOP HEK293 cells. Inhibition of cAMP production and recruitment of β-arrestin were also investigated. Finally, lipophilicity (logD_7.4) and plasma stability of each compound were measured. Compared to the native ligands, we found that the replacement of the terminal carboxylate by a tetrazole slightly decreased both the affinity at mu and delta opioid receptors as well as the half-life. By contrast, replacing the ammonium at the N-terminus with a guanidyl significantly improved the affinity, the potency, as well as the lipophilicity and the stability of the resulting peptides. Replacing the glycine residue with a d-alanine in position 2 consistently improved the potency as well as the stability of the analogues. The best peptidomimetic of the whole series, guanidyl-Tyr-d-Ala-Gly-Phe-Leu-tetrazole, displayed sub-nanomolar affinity and an increased lipophilicity. Moreover, it proved to be stable in plasma for up to 24 h, suggesting that the modifications are protecting the compound against protease degradation.

Full text of DOI:10.1021/acschemneuro.8b00550

Subscriber access provided by Iowa State University | Library
Article
N-guanidyl and C-tetrazole Leu-Enkephalin Derivatives: Efficient Mu and
Delta Opioid Receptor Agonists with Improved Pharmacological Properties
Beaudeau Jean-Louis, Veronique D. Blais, Brian J. Holleran, Alexandre
Bergeron, Graciela Pineyro, Brigitte Guérin, Louis Gendron, and Yves L. Dory
ACS Chem. Neurosci., Just Accepted Manuscript • Publication Date (Web): 07 Jan 2019
Downloaded from http://pubs.acs.org on January 7, 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 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
N-guanidyl and C-tetrazole Leu-Enkephalin
Derivatives: Efficient Mu and Delta Opioid
Receptor Agonists with Improved
Pharmacological Properties
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Jean-Louis Beaudeau1, Véronique Blais2, Brian J. Holleran2, Alexandre Bergeron2, Graciela Piñeyro4,
Brigitte Guérin3*, Louis Gendron2*, Yves L. Dory1*
1. Département de chimie ; 2. Département de Pharmacologie-Physiologie ; and 3. Département de
Médecine Nucléaire et Radiobiologie, Institut de Pharmacologie, Centre de recherche du CHUS,
Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Québec J1H5N4, Canada.
4. Département de Psychiatrie, Centre de Recherche du CHU Ste-Justine, Université de Montréal
KEYWORDS: Delta Opioid Receptor; Pain; Peptides; Tetrazole; Guanidine; Lipophilicity; Isostere;
Pharmacological properties; Plasma Stability.
ABSTRACT: Leu-enkephalin and D-Ala2-Leu-Enkephalin were modified at their N- and C-termini with
guanidyl and tetrazole groups. The resulting molecules were prepared in solution or by solid phase peptide
synthesis. The affinity of the different analogs at mu (MOP) and delta opioid receptors (DOP) was then
1
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 48
1
2
3
4
5
6
7
8
9
assessed by competitive binding in stably transfected DOP and MOP HEK293 cells. Inhibition of cAMP
production and recruitment of -arrestin were also investigated. Finally, lipophilicity (logD7.4) and plasma
stability of each compound were measured. Compared to the native ligands, we found that the replacement
of the terminal carboxylate by a tetrazole slightly decreased both the affinity at mu and delta opioid
receptors as well as the half-life. By contrast, replacing the ammonium at the N-termini with a guanidyl
significantly improved the affinity, the potency, as well as the lipophilicity and the stability of the resulting
peptides. Replacing the glycine residue with a D-Alanine in position 2 consistently improved the potency
as well as the stability of the analogues. The best peptidomimetic of the whole series, Guanidyl-Tyr-D-
Ala-Gly-Phe-Leu-Tetrazole, displayed sub-nanomolar affinity and an increased lipophilicity. Moreover,
it proved to be stable in plasma for up to 24h, suggesting that the modifications are protecting the
compound against protease degradation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Chronic pain is one of the most frequent causes of medical advice seeking. Each year the direct cost of
chronic pain amounts to about 560 billion USD in the US alone.1 The neuronal networks involved in pain
processing are well known, and their investigation has led to the identification of numerous potential
targets to alleviate pain.2–7 However, despite the huge number of molecules developed, mu opioid agonists
such as Morphine remain the gold standard. Opioid receptors exist at least in three subtypes named mu
(MOP), delta (DOP), and kappa (KOP).8,9 Several studies suggest that selective activation of DOP could
provide good analgesia without the adverse effects linked to traditional MOP-activating opioids.10–13
Based on these studies, the hypothesis that the selective activation of DOP would lead to a better
therapeutic profile is generally acknowledged.14–16 Therefore, great efforts have been made during the last
decades. However, many DOP-selective agonists were shown to produce non-lethal convulsions in
2
ACS Paragon Plus Environment

Page 3 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
rodents and nonhuman primates.17 Interestingly, recent advance revealed that not all DOP agonists induce
convulsions.13,18,19 Indeed, some non-peptide candidates even reached clinical trials for the treatment of
pain or depression. However, these were ultimately discarded, because of a lack of efficiency in humans.19
8
9
Owing to their high affinity for DOP, their lack of MOP-related adverse effects, their structural simplicity,
and their low toxicity, many efforts were also directed toward endogenous peptides such as enkephalins
(Tyr-Gly-Gly-Phe-Leu/Met), deltorphins (Tyr-D-Ala-Phe-Asp/Glu-Val-Val-Gly-NH2), and their
optimization as druggable molecules. Unfortunately, their peptide bonds are easily metabolized by
peptidases (half-life of ~2 min),20 and prevent them from crossing the blood brain barrier (BBB) to reach
the central nervous system (CNS), where the key opioid receptors related to analgesia are located.8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Over the last decades, different strategies aiming to improve the pharmacokinetic profile of enkephalins
were used with limited success. Among them, macrocyclisation21 has led to selective DOP peptides such
as H-Tyr-c[D-Pen-Gly-Phe-D-Pen]-OH (DPDPE),22 and H-Tyr-c[D-Cys-Phe-D-Pen]-OH (JOM-13),23
with improved stability, but still suffering from poor BBB penetration. A useful strategy to increase
peptide stability is to replace dipeptides with isosteres. We have already successfully carried such studies
on the backbone of Leu-enkephalin (LENK) by replacing each amide bond by different dipeptide isosteres
(ester, N-methyl amide, triazole, alkene and fluoroalkene).24–28 Together, these studies have been of a
great help to uncover the interactions necessary for LENK to fold into its bioactive conformation in the
DOP.
Surprisingly, few studies have been conducted on the isosteric replacement of both ammonium and
carboxylate N- and C- termini of enkephalins.29 These positions are particularly interesting since
enzymatic degradation of enkephalins occurs mostly at these sites.20,30 Furthermore, the carboxylate and
the ammonium are the main functions responsible for the low bioavailability of enkephalins. Because of
their important roles for both the biological activity and selectivity,31–33 candidates were chosen
3
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 48
1
2
3
4
5
6
7
8
9
cautiously, according to their ability to mimic the ionic state of the endogenous ligand at physiological
pH (Figure 1). Thus, tetrazole is a well-suited analog of a carboxylic acid. Indeed, their properties such
as size, acidity and ability to form an H-bond are similar. Consequently, tetrazole could easily replace the
free carboxylate group at the C-terminal position, improving at the same time lipophilicity and stability
while keeping the key molecular interactions within the receptor. This carboxylate/tetrazole relationship
has already been investigated by Manturewicz et al,29 but surprisingly the study was limited to
pharmacodynamics. Recently, Weltrowska et al,34 described the synthesis of mixed mu agonist/delta
antagonist compounds, where the free ammonium group was replaced by a guanidinium cation. Such
replacement led to a higher cationic profile (increased pKa) and a higher membrane permeability via
association with BBB capillaries.35,36 In addition, biological switching from DOP antagonism to agonism
was observed. The same strategy was applied to cyclic deltorphins,37 with similar observations concerning
their biological activity and pharmacokinetics. To date, such strategies have not been explored for
enkephalin derivatives. Such modifications could however improve both the BBB permeability and the
stability of the analogs.38 Homologation of the Gly-2 residue with a D amino acid is also known to be a
valid strategy to improve pharmacokinetic and selectivity in compound such as H-Tyr-D-Ala-Gly-Phe-
D-Leu-OH (DADLE),39 H-Tyr-D-Ser-Gly-Phe-Thr-OH (DSLET),40 H-Tyr-D-Thr-Gly-Phe-Thr-OH
(DTLET).41 The latter approach has been combined here to N- and C-terminal modifications of LENK.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. pKa values of ammonium cation, carboxylic acid and their respective isosteres guanidine and
tetrazole.
4
ACS Paragon Plus Environment

Page 5 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
In this report, we present the synthesis, affinity and activity of several enkephalin analogs containing one,
two or three modifications on their structure, and the influence of these combinations on their affinity and
potency at MOP and DOP as well as on their lipophilicity and plasmatic stability.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Leu-Enkephalin (LENK) is one of the endogenous opioid peptides, often considered as the natural ligand
for DOP.42 However, as most peptides it has a low bioavailability when systemically-administered and is
readily degraded by enkephalinases and other peptidases.20,43–45 These limitations have to be addressed to
transform such peptides into druggable molecules. In line with our previous studies, we systematically
replaced each amide bond by dipeptide isosteres, resulting in peptidomimetics with improved stability
profiles.24–27 Another important aspect not previously considered is the neutral/ionic state of both LENK
ammonium and carboxylate termini. Indeed, at physiological pH, LENK is a zwitterion with high
hydrophilicity. Furthermore, the first positions to be degraded are the first, the second and the fourth
amide positions.46 Replacing the N- and C-termini of LENK by unnatural bioisosteric functional groups
should increase its resistance toward degrading enzymes. Tetrazole and guanidyl groups were selected to
respectively replace carboxylate and ammonium ions.
Chemistry. The two monosubstituted peptidomimetics 2, 3 and the disubstituted analog 4 were
synthesized and their affinities and other relevant properties could be compared to LENK 1 (Figure 2).
The tetrazole derivative 2 was found to bind DOP and MOP with a lower affinity than compounds 1, 3
and 4; therefore, additional modifications involving residues Gly2 and Leu5 were later applied to 2. These
two positions are known to impact positively the affinity of LENK derivatives as in DADLE, bearing D-
Ala and D-Leu residues at positions 2 and 5 respectively.39 Thus, two new tetrazole derivatives, 5 and 6,
5
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 48
1
2
3
4
5
6
7
8
9
with D-Ala2 and D-Leu5 respectively, were prepared. Subsequently, the tetrazole mimetic 7 combining
D-Ala2 and D-Leu5 in the same structure was also prepared. When compared to the D-Leu5 substituted
compound 6, the D-Ala2 analog 5 led to the best binding results. Finally, the three optimal modifications
(from 3 and 5) were incorporated in the last structure of this series to give rise to compound 8, a
guanidylated LENK containing a tetrazole mimetic at its N-termini and a D-Ala at position 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
ACS Paragon Plus Environment

Page 7 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Figure 2. Structures, in vitro affinities, in vitro potencies and rat plasma stability of target peptides 1-8.
The binding affinity (Ki) of each compound was determined by their ability to inhibit the binding of [125I]-
deltorphin I and [125I]-DAMGO, used as selective radioligands respectively for DOP and MOP (Kd = 0.13
nM Delt I and Kd = 0.69 nM DAMGO). Membrane preparations of stably transfected HEK cells were
7
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 8 of 48
1
2
3
4
5
6
7
8
9
used. Ki values are the means +/- S.E.M. of three to six independent experiments, each performed in
duplicate. The potency (EC50) and the efficacy (Emax) of each compound was determined by their
respective ability to inhibit the forskolin-induced cAMP production through both DOP and MOP, using
the GPF10-EPAC- RLucII BRET biosensor. The potency of each compound to recruit -arrestin 2 using
the GFP10-arrestin DOP/MOP-RLucII BRET couple. EC50 values are the means +/- S.E.M. of three to
five independent experiments each performed in triplicate. The half-life time (T ½) or rat plasma stability
of each compound was measured at 37 °C (ex-vivo) on an analytical Water H Class Acquity UPLC
coupled with a SQ detector 2 and a PDA e detector (200 to 400 nm) and paired with an Acquity UPLC
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CSH C18 column, 1.7 M, 2.1 Å ~ 50 mm, flow 0.8 mL/min, starting with 0.1% formic acid in water,
then moving to 95% acetonitrile containing 0.1% formic acid in 1.3 min. The degradation fragments were
identified through MS analysis and their first appearance time are shown at the cleavage site.
Synthesis. The tetrazole modified Fmoc-Leucines 12a and 12b were synthesized in a straightforward
fashion as described by Sureshbabu et al.47 (Scheme 1). First, the carboxylic acids 9a and 9b were
efficiently converted to their amides 10a and 10b, before being transformed into their corresponding
nitriles 11a and 11b. The latter were transformed into the tetrazoles 12a and 12b (click chemistry) with
overall yields of 69% and 56% from 9a and 9b respectively.
These modified Fmoc-Leucines 12a and 12b were coupled overnight on 2-chlorotrityl resin through their
tetrazole group,48 with an overall loading of 80%. Solid phase synthesis then processed in an iterative way
until completion of the peptide sequences (Scheme 2), using an Fmoc strategy. For peptides 1, 2, 5-7,
Boc-Tyr-OH was introduced in N-terminal position instead of Fmoc-Tyr-OH to remove simultaneous the
N-terminal protective group during the resin cleavage.
8
ACS Paragon Plus Environment

Page 9 of 48
ACS Chemical Neuroscience
1
2
3
Scheme 1. Synthesis of modified Fmoc-Leucine-Tetrazoles 12a and 12b.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the replacement of the ammonium by a guanidinium (as in peptides 3, 4 and 8), the most popular
procedures usually require cleavage of peptides from the resin, followed by an orthogonal protection in
C-terminal position and the conversion of the amine to the guanidine in one or several steps.37 These
multiple steps make the synthesis workflow somehow cumbersome. On the other hand, solid-phase
protocols described in the literature are often expensive,49 or suffer from a poor efficiency,50,51 and/or side
reactions52,53 in addition to being dependent on solvents or steric hindrance.54,55 Nevertheless, we opted
for the solid phase approach due to its reduced number of steps. We succeeded in applying such
methodology using Mukaiyama’s conditions with N,N-diBoc-Methyl-thio-isourea as guanidylating agent
and HgCl2 as reducing agent (Scheme 2).52,53
Total conversion was observed after 2 h (occasionally, the procedure needed to be repeated twice for
completion). The cytotoxicity of mercury chloride was not an issue since it could easily be removed by
filtration due to its high solubility in DMF. As a result, these optimized conditions for solid phase
synthesis allowed us to rapidly obtain all the compounds of our library in a convergent way.
Scheme 2. Solid Phase Peptide Synthesis protocol.
9
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 10 of 48
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Binding Properties and Activity on DOP and MOP. To evaluate the ability of each compound to bind
DOP and MOP, we performed competitive binding assays using membrane preparations from transfected
HEK cells expressing either DOP (HEK/DOP) or MOP (HEK/MOP). Leu-enkephalin 1 and compounds
2-8 inhibited the binding of 125I-deltorphin I (on DOP) and 125I-DAMGO (on MOP) in a concentration-
dependent manner (Figure 2). The replacement of the carboxylate by a tetrazole in compounds 1 and 3,
to produce the derivatives 2 and 4 respectively, led to a slight decrease in affinity. When a guanidinium
was introduced in LENK instead of the ammonium to get compound 3, affinity was improved with a
significant preference for MOP. This selectivity trend was confirmed when a guanidyl group was included
in addition to tetrazole in compound 4.
10
ACS Paragon Plus Environment

Page 11 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
After the preliminary exploration involving the two ends of LENK, additional modifications were
introduced at its second and fifth residues. Thus, starting from peptide 2, Gly2 was replaced by a D-Ala2
residue in mimetic 5 and the chirality of modified (tetrazole) Leu5 was inverted in 6. The two changes
were then combined in the analog 7. From the Ki values measured for the four compounds 2, 5-7, we
could easily conclude that D-Ala at position 2 is indeed a favorable element for binding to DOP, whereas
inversion of the side chain of the fifth residue is not.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Finally, features perceived as beneficial were combined in analog 8, which can be seen as 4 with added
D-Ala2 or 5 with ammonium-guanidinium replacement. Compound 8 was found to be very potent, being
as good as 3 and 5, and slightly better than LENK 1. Overall, the four molecules 1 (the reference ligand),
3, 5 and 8 have a good affinity for both DOP and MOP.
Following the determination of the binding affinity of the library, the potency of each compound was
assessed. Activation of the opioid receptors by their ligands initiates a cascade of intracellular reactions.56
One of these phenomena is the inhibition of adenylate cyclase through the activation of the Gi protein,57
resulting in a decrease in the cAMP production. To monitor changes in second messenger levels we used
a BRET-based biosensor in which acceptor (GFP10) and donor (RLucII) moieties were respectively
introduced at the N- and C-terminal ends of the protein EPAC (GPF10-EPAC-RLucII), used here as a
cAMP sensor.58 cAMP produced in the presence of forskolin (3 µM) imposes conformational changes to
the biosensor which, in turn causes a reduction of the BRET signal. Once activated,59 opioid receptors are
also phosphorylated, leading to the recruitment of -arrestin.60 The -arrestin pathway is involved, among
other processes, in receptor internalization and downregulation and is thought to be responsible for the
deleterious effects of opioids61–63 (although this has been challenged by recent findings TRV130 and
PZM21).64,65 A second signaling outcome that was studied was therefore the recruitment of -arrestin 2-
GFP10 to DOP or MOP bearing RLucII in their C-terminus. All tested compounds showed full agonist
11
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 12 of 48
1
2
3
4
5
6
7
8
activity for the EPAC and -arrestin pathway (Figure 2; Emax). Potencies followed the same trend as
binding affinities (Figure 2). Almost all compounds are less potent than LENK, except for compound 8,
indicating that, once again, the best results were obtained with the latter compound.
9
As shown in Figure 2, when compared to LENK 1, for most compounds a higher concentration was
required to recruit -arrestin. Interestingly, compounds 3 and 5, and to a lesser extent 7 and 8, behave
much like LENK itself (compound 1). When all the characteristics (guanidyl, tetrazole, D-Ala2) of these
three analogues are included in one structure (compound 8), the -arrestin pathway becomes slightly
disfavored, at least with DOP. Generally speaking, the chosen modifications (apart from D-Leu5) thus
prevent any major change in affinity and potency when compared to LENK.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular modeling. To better understand our binding results, compounds 1-8 were docked into the
Human Delta Opioid Receptor crystallographic structure 4N6H,31 following the Induced Fit Docking
(IFD) algorithm. The ligands were given full mobility, whereas the receptor itself was allowed to freely
adjust the geometry of its side chains to a distance of 15 Å away from the docked ligands. Three regions
are of particular importance, the N- and C-terminal positions of the ligands, as well as the second residue.
(Figure 3).
These three locations were thoroughly scrutinized since the following modifications from LENK 1: amine
→ guanidine (at the N-terminus), Gly → D-Ala (at position 2) and carboxylic acid → tetrazole (at the C-
terminus) led to the best compound (compound 8) of the whole series. With respect to the C-terminus, we
found that the interactions between the carboxylate of 1 (Leu5) and the receptor are essentially retained
in its tetrazole mimetic (compound 2) (compare Figure 3a and b). More specifically, both groups similarly
interact with the side chain of Arg290 through a salt bridge. The negatively charged groups of compounds
12
ACS Paragon Plus Environment

Page 13 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
1 and 2 are further held in place by additional interactions with Trp284 and Lys108 respectively. As for
the N-termini, the essential salt bridge between the ammonium (Tyr1) and Asp128 and the two water
molecules linking the phenol group (Tyr1) and His278 are conserved with the guanidyl replacement
(compare Figure 3a and c). In contrast to previous calculation,34 no significant shifting of the ligand was
observed here. Indeed, 1 and 3 (as well as other guanidylated ligands 4 and 8) were found to adopt a
similar orientation inside DOP. This divergence is most likely the result of docking in a flexible receptor
(present work) as opposed to a frozen receptor.34 In the latter, it can be hypothesized that the lack of
flexibility of the receptor decreased its ability to accommodate the bulkier guanidinium cation. In the IFD
model illustrated here, the guanidyl group develops additional stabilizing interactions with the side chains
of Asp128, Asn131 and Trp274, supporting the high affinity measured for some compounds bearing this
modification (3, 8).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 14 of 48
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Induced Fit Docking snapshots of compounds 1-3, 5 in DOP. Side chains were allowed full
mobility. The distance for the -turn (between second and fifth residues) of the ligands are indicated (Å).
a) LENK 1. b) Compound 2. c) Compound 3. d) Compound 5.
In LENK 1, the -turn between Gly2 and Leu5 appears to be a bit loose, with a C=O---H−N distance of
2.80 Å. This might be the consequence of the relative positions of the terminal ammonium and carboxylate
being pulled away from each other by charged residues within the receptor, with consequential elongation
14
ACS Paragon Plus Environment

Page 15 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
of the -turn H-bond. Upon replacement of the carboxylate by the larger tetrazole anion, the C end of
compound 2 experiences a slight inward movement to facilitate the fit. This results in a shorter H-bond
(2.67 Å). This phenomenon is even further enhanced when the ammonium is replaced by the much bulkier
guanidinium cation in compound 3. In this compound, the -turn hydrogen bond length is significantly
shorter at 2.45 Å.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Replacement of Gly2 by a D-Alanine in compound 5 (Figure 3b, d) does not disrupt the -turn of LENK.
In fact, it rather appeared to stabilize it, as suggested by the very short distance of 2.10 Å for the
corresponding H-bond. This stabilization of the active conformation might be a leading cause for the
slight increased affinity observed in compounds 5 and 8 (Ki = 0.36 and 0.32 nM), when compared to their
respective parent compounds 2 and 4 (Ki = 1.24 and 1.61 nM), lacking the extra methyl substituents. Note
that in this model, the leucine side chain remains within the hydrophobic pocket thought to be responsible
for selectivity between MOP and DOP.
Hydrophobicity and Plasma stability. The next things we were logically concerned about were the
physicochemical properties (i.e. lipophilicity and topological polar surface area) and the metabolic
stability of the mimetics 2-8, by comparison with the natural ligand LENK 1. First, the lipophilicity of the
different mimetics was calculated and evaluated experimentally (Table 1).
The logD7.4 of a compound (C), which is the log of its partition coefficient ([C]octanol/[C]water) at pH 7.4, is
acknowledged to be a good index for lipophilicity and blood-brain barrier (BBB) permeability.66
Measured values (logD7.4) were consistently higher than the calculated clogD,67 occasionally by a small
margin (8), but more generally by a large gap (1-7). Obviously, the chosen model inadequately considers
15
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 16 of 48
1
2
3
the structural factors and the weak intramolecular interactions of the molecules.68,69 Once again,
4
5
compound 8 is the most lipophilic, with a gain of almost 2 log units.
6
7
8
9
Table 1. Measured and calculated physicochemical properties of analogs 1-8.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Compound
logD7.4
-1.37
-0.71
-0.42
0.02
clogDb
-1.86
-1.57
-1.58
-0.13
-1.00
-1.57
-1.00
0.44
tPSAb (Å2)
204.40
215.83
240.40
239.80
215.83
215.83
215.83
239.80
LLEc
11.20
9.25
10.20
9.05
9.25
8.66
9.38
9.47
1
2
3
4
5
6
7
8
0.35
-0.18
-0.07
0.53
alogD7.4 values were measured using the shake-flask method with PBS at pH 7.4. bclogD and tPSA were
calculated with ChemAxon software. cLLE = – logEC50(EPAC) – logD7.4.
Topological polar surface area (tPSA) is the sum of areas of all polar contributions in the molecule, and
is therefore an ideal metric for predicting drug transport across the BBB.67,70 All compounds were found
to have polar surface areas greater than 200 Å2, well above the recommended 90 Å2 superior limit.71,72
However, some peptides with equivalent tPSA are active to the CNS.21,73,74 The increase calculated (9 to
36 Å2) for the synthetic mimetics 2-8, LENK being the reference, can be attributed to the addition of polar
atoms in the isostere functions guanidinium and tetrazolate. The logD7.4 and tPSA values are useful
16
ACS Paragon Plus Environment

Page 17 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
descriptors for passive permeation, but is worth noting that addition of guanidine groups may favor
binding to albumin, thus helping BBB crossing through an active process.35,36 Finally, ligand lipophilicity
efficiency (LLE) was calculated for each compound, providing a way to evaluate their potency without
their lipophilicity contribution.75 Since all LLEs are above 7,76 peptides 1-8 are expected to be highly
potent ligands.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Plasma stability of compounds 1-8 was then monitored (Figure 2) to provide hints on the metabolic
stability of each compound and to study the influence of the four modifications reported here.
On its own, a C-terminal tetrazole did not significantly improve the half-life of compound 2 when
compared to LENK 1. This is not surprising, since the main degradation pathway of enkephalins usually
involves the first peptide bond.20 Accordingly, this sensitive bond is better protected by an N-terminal
guanidyl group in compound 3. When the two terminal modifications are present in a same molecule, as
in mimetics 4 and 8, the half-life is significantly increased, as no degradation was observed even after 6
h of incubation in plasma. From the results collected with compounds 2, 5 and 7, one can infer that a Gly2
→ D-Ala replacement also significantly improved the stability of enkephalins.
CONCLUSION
Novel Leu-enkephalin mimetics have been synthesized and their pharmacological profiles evaluated. The
ammonium and carboxylate ions of Leu-enkephalin have respectively been replaced by the similarly
charged guanidinium and tetrazolate ions. Side chains (of D conformation) have also been introduced on
the alpha carbon atoms at the second (Gly2  D-Ala) and fifth residues (S i-butyl  R i-butyl). While no
significant DOP/MOP selectivity could be achieved, the guanidinium-tetrazolate combination showed
increased plasma stability as well as improved lipophilicity. Among the eight derivatives studied here
(including Leu-enkephalin), the best biological properties were obtained for the peptidomimetic 8, bearing
17
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 18 of 48
1
2
3
4
5
6
7
8
9
the three following modifications: guanidinium and tetrazolate ions plus D-Ala2. The affinity and the
potency of this compound for DOP and MOP remained similar to that of Leu-enkephalin, but the
measured logD7.4 was increased by 1.9 log units when compared to Leu-enkephalin, to reach 0.53.
Similarly, the sensitivity to degradation was significantly decreased as was already observed with the
analogue 4, which also possesses a N-terminal guanidine and a C terminal tetrazole. Altogether, our results
reveal that guanidinium and tetrazole groups can be introduced in opioid peptides to improve their stability
and their lipophilicity, without affecting their ability to bind and activate MOP and DOP.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
METHODS
Chemistry.
For SPPS, commercial grade reagents from Chem Impex were used without further purification. Optical
1
13
rotation measurements were made on a Perkin-Elmer 241 polarimeter. H and C NMR spectra were
recorded in deuterated solvents on a Brucker Ascend 400 MHz (100 MHz for 13C spectra) or a Brucker
Advance 300 Ultrashield 300 MHz (75 MHz for 13C spectra) NMR instrument. The residual solvent peaks
have been used as internal references. The peak multiplicities are described as follows: s (singlet), d
(doublet), t (triplet), q (quartet), quin (quintet), spt (septet), m (multiplet) and br (broad). HRMS were
recorded on a Triple TOF 5600, ABSciex mass spectrometer. HPLC preparative purifications were done
on a VYDAC 218TP C18 column. HPLC semi-preparative purifications were done on a Zorbax Eclipse
XDB-C18, 5 µm, 9.4 × 250 mm column. Analytical HPLCs were done with an Eclipse XDB-C18, 5µm,
4.6 × 250 mm column.
(S)-(9H-fluoren-9-yl) methyl (1-amino-4-methyl-1-oxo-pentan-2-yl) carbamate (10a). Fmoc-Leu-OH
(9a) (2.00 g, 5.6 mmol, 1 eq), Boc anhydride (1.60 g, 7.35 mmol, 1.3 eq) and pyridine (0.28 mL, 3.47
mmol, 0.62 eq) were mixed together in dioxane (7 mL) at room temperature. Ammonium carbonate (0.56
g, 7.13 mmol, 1.26 eq) was added to the mixture that was stirred at room temperature until completion of
18
ACS Paragon Plus Environment

Page 19 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
the reaction (6h). EtOAc (30 mL) was added, and the organic layer was washed with water (30 mL) then
5% aqueous H2SO4 (30 mL), dried over MgSO4 and concentrated to dryness. The resulting solid was
triturated in Et2O, and was isolated via filtration to yield the title compound as a pure white solid (1.71 g,
87%). 1H NMR (300 MHz, DMSO-d6)  : 7.91 (d, J = 7.5 Hz, 2H), 7.74 (d, J = 7.0 Hz, 2H), 7.49-7.23
(m, 4H), 6.96 (br s, 1H), 4.39-4.17 (m, 3H), 3.96 (m, 1H), 1.64 (m, 1H), 1.53-1.37 (m, 2H), 0.89 (d, J =
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
6.8 Hz, 3H), 0.87 (d, J = 6.8 Hz, 3H). C NMR (75 MHz, DMSO-d6)  : 174.9, 156.4, 144.3, 141.2,
128.1, 127.5, 125.8, 120.6, 65.8, 53.4, 47.2, 41.3, 24.8, 23.5, 21.9. Mp 189-190 °C (Et2O). HRMS (ESI+)
m/z calcd for C21H24N2O3 (M + Na)+ : 375.1679, found : 375.1681, =0.2 mDa. [D20 −10 (c = 5 mg.mL-1
in MeOH).
(R)-(9H-fluoren-9-yl) methyl (1-amino-4-methyl-1-oxo-pentan-2-yl) carbamate (10b). The same
1
procedure was applied to prepare the enantiomer from Fmoc-D-Leu-OH (9b) (yield : 91%). H NMR
(300 MHz, DMSO-d6)  : 7.90 (d, J = 7.5 Hz, 2H), 7.73 (d, J = 7.0 Hz, 2H), 7.49-7.21 (m, 4H), 6.98 (br
s, 1H), 4.41-4.14 (m, 3H), 3.95 (m, 1H), 1.64 (m, 1H), 1.54-1.40 (m, 2H), 0.89 (d, J = 6.5 Hz, 3H), 0.87
(d, J = 6.5 Hz, 3H). 13C NMR (75 MHz, DMSO-d6)  : 175.0, 156.4, 144.2, 141.2, 128.1, 127.5, 125.8,
120.6, 65.9, 53.4, 47.2, 41.3, 24.7, 23.6, 21.9. HRMS (ESI+) m/z calcd for C21H24N2O3 (M + Na)+ :
375.1679, found : 375.1688, =0.9 mDa. [D20 +12 (c = 5 mg.mL-1 in MeOH).
(S)-(9H-fluoren-9-yl) methyl (1-cyano-3-methyl-butyl) carbamate (11a). Trifluoroacetic anhydride
(0.67 mL, 4.85 mmol, 1.1 eq) was added dropwise to a solution of the amide 10a (1.56 g, 4.41 mmol, 1
eq) and pyridine (0.71 mL, 8.82 mmol, 2 eq) in dry THF (10 mL) cooled in an ice-bath, in order to keep
the internal temperature below 5 °C. The resulting mixture was stirred at room temperature during 16 h,
then iced-water was added. The white precipitate that formed was isolated by filtration, rinsed three times
with cold water and finally recrystallized in a minimum amount of water to give the title compound as
19
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 20 of 48
1
2
3
4
5
6
7
8
9
white powder (1.32 g, 89%). 1H NMR (300 MHz, DMSO-d6)  : 8.16 (d, J = 7.9 Hz, 1H), 7.90 (d, J =
7.5 Hz, 2H), 7.70 (d, J = 7.5 Hz, 2H), 7.43 (t, J = 6.8 Hz, 2H), 7.35 (dt, J = 1.1, 7.5 Hz, 2H), 4.57-4.36
(m, 3H), 4.26 (m, 1H), 1.74-1.55 (m, 3H), 0.89 (d, J = 6.0 Hz, 3H), 0.88 (d, J = 6.0 Hz, 3H). 13C NMR
(75 MHz, DMSO-d6)  : 155.9, 144.1, 141.3, 128.1, 127.5, 125.5, 120.6, 120.3, 66.3, 47.1, 41.2, 40.7,
24.7, 22.4, 22.0. Mp 115-117 °C (water). Rf 0.8 (DCM). HRMS (ESI+) m/z calcd for C21H22N2O2 (M +
Na)+ : 357.1573, found : 357.1579, =0.6 mDa. [D20 −27 (c = 5 mg.mL-1 in MeOH).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-(9H-fluoren-9-yl) methyl (1-cyano-3- methyl-butyl) carbamate (11b). The same procedure was
applied to prepare the enantiomer from the amide 10b (yield: 85%). 1H NMR (300 MHz, DMSO-d6)  :
8.16 (d, J = 7.9 Hz, 1H), 7.90 (d, J = 7.5 Hz, 2H), 7.70 (d, J = 7.3 Hz, 2H), 7.44 (t, J = 6.8 Hz, 2H), 7.33
(dt, J = 1.1, 7.5 Hz, 2H), 4.58-4.36 (m, 3H), 4.26 (m, 1H), 1.74-1.54 (m, 3H), 0.89 (d, J = 5.9 Hz, 3H),
13
0.88 (d, J = 5.9 Hz, 3H). C NMR (75 MHz, DMSO-d6)  : 155.9, 144.1, 141.3, 128.1, 127.5, 125.5,
120.6, 120.3, 66.3, 47.1, 41.2, 40.7, 24.6, 22.4, 22.0. HRMS (ESI+) m/z calcd for C21H22N2O2 (M + Na)+ :
357.1573, found : 357.1578, =0.5 mDa. [D20 +31 (c = 5 mg.mL-1 in MeOH).
(9H-fluoren-9-yl) methyl (3-methyl-1- (1H-tetrazol-5-yl) butyl) carbamate (12a). Cyano (11a) (0.97
g, 2.90 mmol, 1 eq), sodium azide (0.38 g, 5.80 mmol, 2 eq), ZnBr2 (0.33 g, 1.45 mmol, 0.5 eq) were
mixed together in isopropanol/water (10/25 mL) at room temperature. The mixture was heated at 100 °C
and DMF was added until total dissolution. The resulting solution was refluxed for 72 h and 3N HCl (10
mL) was added. The precipitate formed was dissolved with the minimum amount of EtOAc. The mixture
was extracted with EtOAc (3 × 40 mL). The organic extracts were collected together and washed with
brine (50 mL), before being dried over MgSO4 and concentrated to dryness. The solid was purified by
recrystallization in EtOAc/ Hexane, to give the title compound as a pure white solid (0.75 g, 89 %). 1H
NMR (300 MHz, DMSO-d6)  : 8.07 (d, J = 8.1 Hz, 1H), 7.90 (d, J = 7.5 Hz, 2H), 7.72 (t, J = 7.0 Hz,
20
ACS Paragon Plus Environment

Page 21 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
2H), 7.42 (t, J = 6.8 Hz, 2H), 7.33 (dt, J = 3.0, 7.3 Hz, 2H), 4.95 (m, 1H), 4.35 (m, 2H), 4.24 (m, 1H),
1.91-1.66 (m, 2H), 1.58 (m, 1H), 0.93 (d, J = 6.8 Hz, 3H), 0.90 (d, J = 6.8 Hz, 3H). 13C NMR (75 MHz,
DMSO-d6)  : 156.3, 150.3, 144.2, 141.2, 128.1, 127.5, 125.7, 120.6, 66.1, 47.2, 44.8, 42.2, 24.5, 23.1,
21.9. Mp 191-192 °C (water) HRMS (ESI+) m/z calcd for C21H23N5O2 (M + Na)+ : 400.1744, found :
400.1751, =0.7 mDa. [D20 −30 (c=5 mg.mL-1 in MeOH).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(9H-fluoren-9-yl) methyl (3-methyl-1- (1H-tetrazol-5-yl) butyl) carbamate (12b). The same
procedure was applied to prepare the enantiomer from the cyanide 11b (yield: 73%). 1H NMR (300 MHz,
DMSO-d6)  : 8.10 (d, J = 8.1 Hz, 1H), 7.88 (d, J = 7.3 Hz, 2H), 7.72 (t, J = 6.9 Hz, 2H), 7.41 (t, J = 6.8
Hz, 2H), 7.30 (m, 2H), 5.00 (m, 1H), 4.51-4.16 (m, 3H), 1.90-1.66 (m, 2H), 1.59 (m, 1H), 0.93 (d, J = 6.6
Hz, 3H), 0.90 (d, J = 6.6 Hz, 3H). 13C NMR (75 MHz, DMSO-d6)  : 156.3, 150.4, 144.1, 141.2, 128.1,
127.5, 125.7, 120.6, 66.2, 47.2, 44.9, 42.4, 24.6, 23.1, 21.9. HRMS (ESI+) m/z calcd for C21H23N5O2 (M
+ Na)+ : 400.1744, found : 400.1768, =2.4 mDa. [D20 −27 (c=5 mg.mL-1 in MeOH).
General method for peptide synthesis on solid support (SPPS methodology). Between each step the
following sequence of washes was applied: DMF (3 ×), iPrOH (3 ×) and DCM (3 ×). For the first coupling,
the 2-Cl-trityl resin (624 mg, 0.71 mmol.g-1 factory loading) was placed in a sintered glass peptide
synthesis vessel and swollen with DCM. The Fmoc protected Leu or 12a or 12b (1.2 eq) and DIPEA (4
eq) were added to the resin. The suspension was agitated in a shaker for 16 h. The resin was washed with
DCM / MeOH / DIPEA (17/2/1). The loading was measured via mass difference after drying the resin.
After the initial coupling to the resin, the peptide sequence was then synthetized using conventional SPPS
techniques with Fmoc-Phe-OH (coupling #1), Fmoc-Gly-OH (coupling #2), Fmoc-Gly-OH or Fmoc-D-
Ala-OH (coupling #3) and Boc-Tyr(tBu)-OH or Fmoc-Tyr(tBu)-OH (coupling #4). All four couplings
were performed using 3 eq of protected amino acid, HATU (3 eq) and HOBt-Cl (3 eq), with 9 eq of
DIPEA in a minimum volume of DMF. The resulting suspension was agitated in a shaker for 1 h at a time.
21
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 22 of 48
1
2
3
4
5
6
7
8
9
This step was repeated until Kaiser’s test result was negative. Fmoc deprotections were carried out by
suspending then agitating the resin in 20% piperidine in DMF for 5 min (procedure repeated one more
time). The final peptides were cleaved off the resin, using 3 mL of cleavage solution per 1 g of resin in a
glass vial and with vigorous agitation on a shaker for 3 h. The cleavage solution consisted in 95 % TFA,
2.5 % H2O, 2.5 % TIPS. After cleavage, the mixture was filtered on cotton to remove the residual resin.
The solvents were concentrated under vacuum and in cold Et2O (15 mL) was added at 0 °C. The
precipitates that formed were centrifuged and the supernatant Et2O was removed. The remaining
precipitates were dissolved in water and tBuOH (less than 50%) was added to complete the dissolution if
necessary. The final aqueous solutions were frozen and lyophilized. The crude peptides were purified
using preparative reverse-phase HPLC (detection at 210 nm), with a C18 column (Solvent A: 0.1% TFA
in water and solvent B: 0.1% TFA in Acetonitrile. Gradient: 10 to 70% of B in A over 50 min; flow rate:
1 mL/min). The purity of all fractions was analyzed using an analytical HPLC (detection at 210 nm), with
a C18 column. All pure fractions (over 96% purity) were combined, frozen and lyophilized.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General method for guanidylation on solid support (concerns peptides 3, 4, 8). Fmoc-Tyr(tBu)-OH
was used instead of Boc-Tyr(tBu)-OH. Fmoc was cleaved as described in the SPPS methodology part.
After positive Kaiser test, the resin was swelled in a minimum amount of DCM. A solution of Et3N (9 eq)
and N,N-DiBoc-N-methylisothioisourea (1.5 eq) in DMF (6 mL) was added. The reactor was shaken for
10 min, before adding HgCl2 (3 eq) into the solution. The vessel was then agitated for 2 h, or until
completion of the reaction monitored by Kaiser test. Sometimes, the reaction needed to be repeated twice
to reach completion. The peptide was then deprotected and cleaved from the resin, as described before.
1
Peptide 2. H NMR (300 MHz, DMSO-d6)  : 8.84-8.66 (m, 2H), 8.19-7.95 (m, 5H), 7.23-7.13 (m, 5
H), 7.06 (d, J = 9.5 Hz, 2H), 6.72 (d, J = 9.5 Hz, 2H), 5.22 (m, 1H), 4.54 (m, 1H), 4.01 (m, 1H), 3.90-
3.63 (m, 4H), 3.08-2.93 (m, 2H), 2.87-2.65 (m, 2H), 1.87-1.65 (m, 2H), 1.52 (m, 1H), 0.93 (d, J = 6.7 Hz,
22
ACS Paragon Plus Environment

Page 23 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
13
3H), 0.85 (d, J = 6.7 Hz, 3H). C NMR (75 MHz, D2O)  : 172.5, 171.2, 170.7, 169.9, 156.8, 155.2,
135.4, 130.8, 128.9, 128.5, 127.1, 125.4, 115.8, 55.1, 54.5, 42.4, 42.1, 42.0, 40.4, 37.0, 35.9, 24.0, 21.7,
20.7. HRMS (ESI+) m/z calcd for C28H37N9O5 (M + H)+ : 602.2812, found: 602.2809, =0.3 mDa.
9
Peptide 3. 1H NMR (400 MHz, D2O)  : 7.40-7.23 (m, 5H), 7.15 (d, J = 9.5 Hz, 2H), 6.85 (d, J = 9.5 Hz,
2H), 4.63 (m, 1H), 4.44 (m, 1H), 4.33 (m, 1H), 3.96 & 3.91 (AB, J = 15.8 Hz, 2H), 3.86 (s, 2H), 3.23-
3.10 (m, 2H), 3.05-2.93 (m, 2H), 1.68-1.50 (m, 3H), 0.90 (d, J = 6.2 Hz, 3H), 0.84 (d, J = 6.2 Hz, 3H).
13C NMR (100 MHz, D2O)  : 176.2, 172.7, 172.2, 171.4, 170.8, 156.5, 154.7, 136.2, 130.7, 129.2, 128.7,
127.3, 127.1, 115.5, 56.5, 54.8, 51.6, 42.5, 42.2, 39.5, 37.2, 36.9, 24.3, 22.1, 20.6. HRMS (ESI) m/z calcd
for C29H39N7O7 (M + H)+ : 598.29837, found : 598,3256, =1.5 mDa.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Peptide 4. 1H NMR (400 MHz, D2O  : 7.18-7.09 (m, 5H), 7.07-6.98 (m, 2H), 6.83 (d, J = 8.5 Hz, 2H),
5.24 (dd, J = 6.3, 9.6 Hz, 1H), 4.50 (dd, J = 7.1, 8.8 Hz, 1H), 4.44 (dd, J = 5.6, 7.9 Hz, 1H), 3.96 & 3.94
(AB, J = 15.9 Hz, 2H), 3.90 (s, 2H), 3.17 (m, 1H), 3.04-2.84 (m, 3H), 1.81 (m, 1H), 1.64 (m, 1H), 1.45
(m, 1H), 0.86 (d, J = 6.6 Hz, 3H), 0.81 (d, J = 6.6 Hz, 3H). 13C NMR (100 MHz, D2O)  : 172.5, 172.2,
171.5, 170.8, 156.8, 156.5, 154.7, 135.4, 130.7, 128.9, 128.5, 127.2, 127.1, 115.5, 56.5, 55.1, 42.6, 42.2,
42.0, 40.3, 37.1, 36.9, 23.9, 21.7, 20.7. HRMS (ESI+) m/z calcd for C29H39N11O5 (M + H)+ : 622.32084,
found : 622,3305, =9.7 mDa.
Peptide 5. 1H NMR (300 MHz, D2O)  : 7.10-6.98 (m, 5H), 6.97-6.86 (m, 2H), 6.80 (d, J = 8.8 Hz, 2H),
5.14 (m, 1H), 4.41 (m, 1H), 4.16-4.03 (m, 2H), 3.78 (s, 2H), 3.10 (m, 1H), 2.97-2.77 (m, 3H), 1.73 (m,
1H), 1.56 (m, 1H), 1.34 (m, 1H), 1.08 (d, J = 7.3 Hz, 3H), 0.78 (d, J = 6.6 Hz, 3H), 0.73 (d, J = 6.6 Hz,
3H). 13C NMR (75 MHz, D2O)  : 174.9, 172.3, 170.7, 169.2, 157.8, 155.1, 135.4, 130.7, 128.9, 128.4,
127.0, 125.5, 115.8, 55.0, 54.5, 49.8, 42.3, 42.1, 40.7, 37.0, 36.0, 24.0, 21.6, 20.8, 16.1. HRMS (ESI+)
m/z calcd for C29H39N9O5 (M + Na)+ : 616.2966, found : 616.2975, =0.9 mDa.
23
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 24 of 48
1
2
3
4
5
6
7
8
9
Peptide 6. 1H NMR (300 MHz, D2O)  : 7.34-7.20 (m, 3H), 7.19-7.10 (m, 2H), 7.07 (d, J = 8.6 Hz, 2H),
6.81 (d, J = 8.6 Hz, 2H), 5.06 (dd, J = 7.4, 7.9 Hz, 1H), 4.52 (dd, J = 7.5, 7.8 Hz, 1H), 4.18 (t, J = 7.3 Hz,
1H), 3.85 (s, 2H), 3.81 (s, 2H), 3.16-2.84 (m, 4H), 1.53 (m, 2H), 0.96 (m, 1H), 0.73 (d, J = 7.5 Hz, 1H),
0.68 (d, J = 7.5 Hz, 3H). 13C NMR (75 MHz, D2O)  : 173.1, 171.2, 170.7, 169.9, 157.7, 155.1, 135.7,
130.8, 129.1, 128.9, 127.3, 125.4, 115.8, 55.4, 54.5, 43.0, 42.3, 42.1, 40.7, 37.1, 35.9, 23.5, 21.9, 20.5.
HRMS (ESI+) m/z calcd for C28H37N9O5 (M + Na)+ : 602.2810, found: 602.2819, =0.9 mDa.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Peptide 7. 1H NMR (300 MHz, D2O)  : 7.33-7.19 (m, 3H), 7.18-7.11 (m, 2H), 7.08 (d, J = 8.5 Hz, 2H),
6.80 (d, J = 8.6 Hz, 2H), 5.06 (m, 1H), 4.49 (m, 1H), 4.15-4.00 (m, 2H), 3.77 (s, 2H), 3.16-2.84 (m, 4H),
1.63-1.41 (m, 2H), 1.07 (d, J = 7.3 Hz, 3H), 0.94 (m, 1H), 0.73 (d, J = 6.6 Hz, 3H), 0.67 (d, J = 6.6 Hz,
3H). 13C NMR (75 MHz, D2O)  : 174.9, 173.0, 170.8, 169.2, 157.6, 155.1, 135.7, 130.8, 129.1, 128.8,
127.3, 125.5, 115.8, 55.3, 54.5, 49.8, 43.0, 42.1, 40.6, 37.1, 36.0, 23.5, 21.9, 20.5, 16.1. HRMS (ESI+)
m/z calcd for C29H39N9O5 (M + H)+ : 616.2966, found : 616.2981, =1.5 mDa.
Peptide 8. 1H NMR (400 MHz, D2O)  : 7.17-7.07 (m, 5H), 7.05-6.98 (m, 2H), 6.84 (d, J = 8.5 Hz, 2H),
5.23 (dd, J = 6.3, 9.6 Hz, 1H), 4.49 (dd, J = 7.7, 8.5 Hz, 1H), 4.40 (t, J = 6.8 Hz, 1H), 4.21 (q, J = 7.3 Hz,
1H), 3.88 & 3.85 (AB, J = 17.4 Hz, 2H), 3.12-2.85 (m, 4H), 1.81 (m, 1H), 1.64 (m, 1H), 1.45 (m, 1H),
1.28 (d, J = 7.3 Hz, 3H), 0.86 (d, J = 6.6 Hz, 3H), 0.82 (d, J = 6.6 Hz, 3H). 13C NMR (400 MHz, D2O)
 : 175.2, 172.4, 171.3, 170.8, 156.9, 156.2, 154.7, 135.5, 130.7, 128.9, 128.5, 127.0, 117.7, 115.5, 56.1,
55.0, 50.1, 42.2, 42.0, 40.3, 37.4, 36.8, 23.9, 21.7, 20.6, 16.1. HRMS (ESI+) m/z calcd for C30H41N11O5
(M + H)+ : 636.33649, found : 636,3409, =4.4 mDa.
Cell culture. HEK293 (Human embryonic kidney) cells stably expressing the mouse flag-DOP (obtained
from Dr Richards Howells, New Jersey Medical School, Newark, NJ, USA) or the human flag-MOP
(obtained from Dr. Mark von Zastrow, University of California, San Francisco, CA, USA) were grown at
24
ACS Paragon Plus Environment

Page 25 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
37°C in DMEM supplemented with 10% fetal bovine serum and 10 IU penicillin, 100 mg/L streptomycin
and 2 mM glutamine. Geneticin (G418) was added to the medium to maintain the expression of the vector.
Cells were grown in humidified atmosphere of 95% air and 5% CO2.
8
9
Competitive Binding Assays. We evaluated the affinity (Ki) of the different compounds for DOP or
MOP using membrane extracts from stable HEK293 cells. HEK293 cells grown to confluence in 150 mm
Petri dishes were frozen at -80 °C until use. On the day of the experiment, the cells were submitted to a
heat shock by placing the Petri dishes at 37 °C for 60 sec before returning to ice. The cells were then
harvested in ice-cold buffer A (tris-HCl 20 mM, MgCl2 5 mM, NaCl 150 mM, pH 7.4) using a cell scraper
and centrifuged at 3200 g for 15 min at 4 °C. The pellet containing the membrane extract was resuspended
in 1 mL of buffer A. The protein concentration was determined with Bio-Rad DCTM Protein Assay
reagents (Bio-Rad Laboratories, Mississauga, ON, Canada) and the pellet was further diluted in buffer A
containing 0.01% BSA and distributed in 96-well plates. [125I]-Deltorphin I (specific activity:~1139
Ci/mmol) or [125I]-DAMGO (specific activity:~2173 Ci/mmol) was used to determine the binding affinity
of the compounds in a competitive binding assay on DOP or MOP, respectively. Experiments were
performed using a membrane concentration of 10-20 µg of proteins/mL and 150 000 cpm of the
radiolabeled ligand. Nonspecific binding was determined using 10 µM non–radioactive deltorphin II or
DAMGO. Incubations were performed for 60 min at room temperature with increasing concentration of
compound ranging from 10 µM to 1 pM, and the reaction was stopped by filtration using ice-cold buffer
A on filtered 96-well plates. Filters were placed in 5 mL tubes, and the radioactivity was determined using
a Wizard2 Automatic Gamma Counter (PerkinElmer, Woodbridge, ON, Canada). Data were analyzed
using a non-linear fitting analysis, and the Ki values were calculated using GraphPad Prism 7.0. Ki are
expressed as the mean +/- S.E.M. from 3-6 independent experiments, each performed in duplicate.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 26 of 48
1
2
3
4
5
6
7
8
9
BRET signaling. HEK293 cells were maintained in DMEM supplemented with 10% FBS, 100 IU/mL
penicillin, and 100 μg/mL streptomycin at 37 °C in a humidified atmosphere containing 95% air and 5%
CO2. The day prior transfection, cultured cells were washed with PBS at room temperature, trypsinized,
and seeded at 3 000 000 cells in a 10-cm Petri dish. For GPF10-EPAC-RLucII transfection,58 3000 ng of
pcDNA3-3HA-ratMOP with 150 ng of GPF10-EPAC-RLucII or 6000 ng of pcDNA3-Flag-ratDOP with
600 ng of GPF10-EPAC-RLucII was added to 600 μL of 150 mM NaCl containing 27 μg of PEI
(polyethylenimine) and ssDNA was added to complement to 9 μg the total amount of DNA. For Arr2-
GFP10 / receptor-RLucII transfection, 600 ng of pcDNA3-3HA-ratMOP-RLucII or pcDNA3-Flag-
ratDOP-RLucII with 12000 ng Arr2-GFP10 was added to 600 µl of 150 mM NaCl containing 37.8 µg
PEI. The mixture was incubated for 20 min before being added to the cultured cells. At 24 h post-
transfection, cells were washed with PBS, trypsinized and plated in 96-well white plates (75 000
cells/well) and left for another 24 h. The cells were then equilibrated at room temperature for at least 1 h
with 70 µL stimulation buffer (10 mM Hepes, 1 mM CaCl2, 0.5 mM MgCl2, 4.2 mM KCl, 146 mM NaCl,
5.5 mM glucose, pH 7.4). Coelenterazine 400A (an RLucII substrate) was added to a final concentration
of 5 μM, 10 min before stimulation. For the EPAC assay, cells were first stimulated with 3 μM forskolin
to increase cAMP. Cells were stimulated with ligand ranging from 10µM to 1pM and incubated for 10
min prior to signal acquisition. BRET2 signals were measured using a TECAN M1000 fluorescence
reader (TECAN, Grödig, Austria). RLucII and GFP10 emissions were collected in the 400–450 nm
window (RLucII) and 500 to 550 nm window (GFP10). The BRET2 signal was calculated as the ratio of
light emitted by the acceptor GFP10 over the light emitted by the donor RLucII. For each assay, data were
normalized as percentage of the maximal deltorphin II (DOP) or DAMGO (MOP) response. All data were
analyzed using the nonlinear curve fitting equations in GraphPad Prism (v7.0; GraphPad Software, La
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26
ACS Paragon Plus Environment

Page 27 of 48
ACS Chemical Neuroscience
1
2
3
4
5
Jolla, CA) to estimate the pEC50 values of the curves for the different pathways. Results are expressed as
the mean ± SEM from 3-6 independent experiments, each performed in triplicate.
6
7
8
9
Molecular modelling. Calculations were carried out with the Molecular Operating Environment
(MOE),77 using our model of Leu-enkephalin into the structure of the 1.8 Å human -(PDB 4EJ4) opioid
7TM receptor as template. The different modifications were included on LENK with the help of the
“builder” tool and the resulting structures were minimized using the OPLS-AA force field. Each resulting
ligand (Conformation: “None”) was then docked into the receptor, using the general docking protocol of
MOE (Receptor: Receptor + Solvent; Site: Ligand Atoms; Ligand: Ligand Atoms). The “Triangle
Matcher” routine (Timeout: 300 s; Returned poses: 1000) was used as placement method, each pose being
scored with the “London dG” algorithm (30 retained poses). The different poses were refined with the
Induced Fit protocol (Refinement > Induced Fit; Cutoff: 15 Å; Side Chains: Free; Termination Criterion:
Gradient 0.01; Iterations: 500; Pharmacophore Restraint: Force Constant 100; Radius offset: 0.4) then
scored with the “GBVI/WSA dG” algorithm (5 retained poses). For each peptide, the best pose was then
selected based on the best docking score, and visually by the presence of the essential interactions between
the phenolic ring and His278 (via two water bridges), and the ammonium (or guanidinium) with Asp128.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
logD7.4 Determination. The determination of the distribution coefficient (logD) was performed using a
modified version of the shake flask method.27 Before the experiment, octanol and phosphate buffer (PBS
pH 7.4) were mixed together for 24 h to allow saturation of each solution. The mixture was allowed to
rest, and the 2 phases were separated and used as solvents in the coefficient measurement. Each peptide
(0.1 mg) was placed in a vial to which saturated PBS (1 mL) and octanol (0.5 mL) were added. The vial
was then shaken mechanically for 10 min. The mixture was allowed to rest for 30 min or until complete
phase separation. Aliquots of both phases were taken and injected in an HPLC instrument (10 L of each
aliquot was injected in an Agilent 1100 series HPLC. Column: Agilent Eclipse Plus C-18 column, 50 mm
27
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 28 of 48
1
2
3
4
5
6
7
8
9
X 3.0 mm, 1.8 m. Solvent A: 0.1% TFA in water and solvent B: 0.1% TFA in Acetonitrile. Gradient: 10
to 90% of B in A over 60 min; flow rate: 0.4 mL/min; UV detection at 210 nm). The retention time of
each peptide was already known from the HPLC purity analysis of each peptide. The octanol peak did not
interfere with the experiment. The area under the curve (AUC) of the corresponding peak was integrated
for each phase injected. The logD for each peptide was calculated as follows: logD7.4=log10 (AUC octanol
phase/AUC PBS phase).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Plasma stability. Plasma was prepared from two male Sprague−Dawley rats (300−350 g; Charles River
Laboratories, St-Constant, QC, Canada). All animal procedures were approved by the Ethical Committee
for Animal Care and Experimentation of the Université de Sherbrooke (protocol #234-14) and were
performed in accordance to the regulations of the Canadian Council on Animal Care (CCAC). Briefly,
rats were anesthetized with isoflurane (3% isoflurane with 97% medical air) and exsanguinated using an
18G-1½ needle connected to a vacutainer containing EDTA. The blood was rapidly transferred to 1.5 mL
tubes and centrifuged at 1600 g for 15 min at 4 °C. The plasma was then stored at −80 °C in 500 μL
aliquots until use. The stability of Leu-enkephalin and its analogues was determined in plasma diluted to
50% with saline. Therefore, plasma (25 μL) was incubated at 37 °C for 15 min before the addition of the
peptide solution (25 μL of a 100 μM isotonic sodium chloride solution; 0.9% w/v). At each indicated time
point (0, 1, 2, 5, 10, 15, 20, 30, 60, 120, 180, 240, 300 and 360 min), the aliquot was quenched with 100
µL of a methanol solution containing 20 µM Fmoc-Leu-OH as an internal standard. The resulting
solutions were centrifuged at 13 000 rpm for 15 min at 4°C on tubes equipped with 0.2 µm PVDF
centrifugal filters (Canadian Life Science, Peterborough, ON, Canada). A volume of 5 µL of the resulting
supernatant was analyzed on an analytical Water H Class Acquity UPLC coupled with a SQ detector 2
and a PDA e detector (200 to 400 nm) and paired with an Acquity UPLC CSH C18 column, 1.7 µM, 2.1
Å ~ 50 mm, flow 0.8 mL/min, starting with 0.1% formic acid in water, then to 95% acetonitrile containing
28
ACS Paragon Plus Environment

Page 29 of 48
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
0.1% formic acid in 1.3 min. Area under the curve for the peak corresponding to the extracted mass
chromatogram was then measured for both the peptides and the internal standard on the total ion current
(T.I.C.) chromatogram and were used for quantification (using the ratio peptide/standard). Values were
reported over the 0 min incubation time point to determine the half-life (t1/2) using the GraphPad Prism
one-phase decay tool. The results are the mean +/- SEM of 3 independent experiments.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information
All remaining experimental details, spectral characterization, HPLC analysis, and 1H, 13C NMR spectra
for the precursors 10-12 and peptidomimetics 2-8 are in a pdf file. This material can be obtained, free of
charge, via the internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
* E-mail: Brigitte.Guerin2@USherbrooke.ca; Louis.Gendron@USherbrooke.ca;
Yves.Dory@USherbrooke.ca.
Author Contributions
JLB and VB have contributed equally to this work and wrote the original manuscript. BG, GP, LG and
YLD designed the study and finalized the manuscript through contributions of all authors. All authors
have given approval to the final version of the manuscript. JLB, VB, BH and AB performed the
experiments. JLB carried out the virtual docking experiments.
29
ACS Paragon Plus Environment