Exploration of the fifth position of leu-enkephalin and its role in binding and activating delta (DOP) and mu (MOP) opioid receptors

Article abstract of DOI:10.1002/pep2.24070

Enkephalins are pentapeptidic endogenous ligands that regulate nociception by binding to mu (MOP) and delta (DOP) opioid receptors. To further explore the role of the leucine residue of Leu-enkephalin, 12 peptidomimetic analogs were synthesized by systematically replacing this residue with non-natural amino acids. The analogs were tested for their ability to bind DOP and MOP. We also investigated the potency of these analogs to inhibit cAMP production and to recruit β-arrestin 2 via both receptors. We found that replacement of the leucine residue by substituted non-natural amino acid derivatives of alanine, cycloleucine, or isoleucine was generally well tolerated. By contrast, substituting leucine with homoproline greatly reduced the affinity for DOP and, to a lesser extent, for MOP. Interestingly, when compared to Leu-enkephalin, analogs containing either aza-β-homoleucine or cycloleucine showed a bias toward inhibition of cAMP production through the activation of DOP but not MOP. By contrast, derivatives containing 4,5-dehydroleucine or_D-allo-isoleucine conferred a bias toward β-arrestin 2 at MOP, but not DOP. Our results suggest that position 5 in Leu-enkephalin analogs can be further exploited to develop compounds with the potential to produce bias toward G protein or β-arrestin 2.

Full text of DOI:10.1002/pep2.24070

Received: 7 December 2017
DOI: 10.1002/pep2.24070
Revised: 14 March 2018
Accepted: 23 March 2018
|
|
O R I G I N A L A R T I C L E
Exploration of the fifth position of leu-enkephalin and its role in
binding and activating delta (DOP) and mu (MOP) opioid
receptors
Dominique Bella Ndong1,2
V eꢀ ronique Blais1,3
Isabelle Cantin-Savoie1,4
Sophie Beauchemin1,4
Brigitte Gu eꢀ rin1,5,7
Louis Gendron1,3,5
Brian J. Holleran1,3
William Robert1,3
Richard Leduc1,3,5
|
|
|
Arnaud Proteau-Gagn eꢀ 1,4
|
|
|
Jean-François Nadon1,4
Graciela Pi n~ eyro6
|
|
|
Yves L. Dory1,4,5
|
|
|
1Institut de Pharmacologie, Universit eꢀ de Sherbrooke, 3001 12e av. Nord, Sherbrooke, Qu eꢀ bec, J1H 5N4, Canada
2D eꢀ partement de Biochimie, Facult eꢀ de M eꢀ decine et des Sciences de la Sant eꢀ , Universit eꢀ de Sherbrooke, Sherbrooke, Qu eꢀ bec, Canada
3D eꢀ partement de Pharmacologie-Physiologie, Facult eꢀ de M eꢀ decine et des Sciences de la Sant eꢀ , Universit eꢀ de Sherbrooke, Sherbrooke, Qu eꢀ bec, Canada
4D eꢀ partement de Chimie, Facult des Sciences, Universit eꢀ de Sherbrooke, Sherbrooke, Qu eꢀ bec, Canada
5Centre de recherche du CHU de Sherbrooke, Sherbrooke, Qu eꢀ bec, Canada
6D eꢀ partement de Psychiatrie, Centre de Recherche du CHU Ste-Justine, Universit eꢀ de Montr eꢀ al, Montreal, Qu eꢀ bec, Canada
7D eꢀ partement de M eꢀ decine Nucl eꢀ aire et de Radiobiologie, Facult eꢀ de M eꢀ decine et des Sciences de la Sant eꢀ , Universit eꢀ de Sherbrooke, Sherbrooke, Qu eꢀ bec, Canada
Correspondence
Abstract
Yves L. Dory D eꢀ partement de Chimie,
Facult eꢀ des Sciences, Universit eꢀ de
Enkephalins are pentapeptidic endogenous ligands that regulate nociception by binding to mu
Sherbrooke, Sherbrooke, Qu eꢀ bec, Canada
(MOP) and delta (DOP) opioid receptors. To further explore the role of the leucine residue of Leu-
Email: Yves.Dory@usherbrooke.ca,
Louis Gendron D eꢀ partement de
Pharmacologie-Physiologie, Facult eꢀ de
m eꢀ decine et des sciences de la sant eꢀ ,
Universit eꢀ de Sherbrooke, Sherbrooke,
Qu eꢀ bec, Canada
enkephalin, 12 peptidomimetic analogs were synthesized by systematically replacing this residue
with non-natural amino acids. The analogs were tested for their ability to bind DOP and MOP. We
also investigated the potency of these analogs to inhibit cAMP production and to recruit b-arrestin
2
via both receptors. We found that replacement of the leucine residue by substituted non-natural
amino acid derivatives of alanine, cycloleucine, or isoleucine was generally well tolerated. By con-
trast, substituting leucine with homoproline greatly reduced the affinity for DOP and, to a lesser
extent, for MOP. Interestingly, when compared to Leu-enkephalin, analogs containing either aza-
b-homoleucine or cycloleucine showed a bias toward inhibition of cAMP production through the
activation of DOP but not MOP. By contrast, derivatives containing 4,5-dehydroleucine or D-allo-
isoleucine conferred a bias toward b-arrestin 2 at MOP, but not DOP. Our results suggest that
position 5 in Leu-enkephalin analogs can be further exploited to develop compounds with the
potential to produce bias toward G protein or b-arrestin 2.
Email: Louis.Gendron@usherbrooke.ca,
Brigitte Gu eꢀ rin, D eꢀ partement de M eꢀ decine
Nucl eꢀ aire et de Radiobiologie, Facult eꢀ de
M eꢀ decine et des Sciences de la Sant eꢀ ,
Universit eꢀ de Sherbrooke, Sherbrooke,
Qu eꢀ bec, Canada
Email: Brigitte.Guerin2@usherbrooke.ca
Funding information
Canadian Institutes of Health Research,
Grant/Award Number: MOP-136871; LG is
the recipient of a Senior-salary support
from the Fonds de recherche Qu eꢀ bec -
Sant eꢀ (FRQS)
K E Y W O R D S
biased signaling, enkephalins, opioid receptors, peptidomimetics, solid-phase synthesis
1
|
INTRODUCTION
receptors (GPCRs) named mu (MOP), delta (DOP), and kappa (KOP)
[
3]
opioid receptors. Commonly used opioids mainly produce analgesia
[
3,4]
Opioids are among the most effective drugs used for the treatment of
through the activation of MOP.
However, morphine, the golden
[
1,2]
chronic pain.
They bind and activate three G protein-coupled
standard in pain management, produces multiple adverse effects such
Peptide Science. 2018;e24070.
https://doi.org/10.1002/pep2.24070
wileyonlinelibrary.com/journal/pep2
V
C
2018 Wiley Periodicals, Inc.
|
1 of 10

2
of 10
|
BELLA NDONG ET AL.
as sedation, myosis, euphoria, nausea, constipation, respiratory depres-
sion, and a high risk of addiction. As these adverse effects are mostly
[4]
produced by the activation of MOP, the other opioid receptor-
subtypes (ie DOP and KOP) represent interesting alternatives for the
[5]
development of novel analgesics. If activation of KOP is accompanied
by nausea, dysphoria, stress, and hallucinations, recent studies have
shown that DOP agonists have the potential to induce analgesia with
[5–7]
less or no important adverse effects.
FIGURE 1 The structural component of Leu-enkephalin
Endogenous opioids include endomorphins, endorphins, dynor-
[8,9]
phins, and enkephalins.
It is known that endomorphins are selective
washed. Amine loading was determined by UV quantization of Fmoc
release: An aliquot (10 mg) of resin was dried under vacuum; a mixture
of piperidine (5 mL) and DMF (5 mL) was added, and the suspension
was agitated in a shaker for 30 min; a portion of the solution (0.5 mL)
was diluted in DCM (4.5 mL) and its absorbance was read with a UV
to MOP and dynorphins to KOP. However, enkephalins and endor-
phins bind both DOP and MOP with a similar affinity. Despite this fact,
enkephalins are generally considered as the endogenous DOP ligands.
The enkephalins (Leu-enkephalin and Met-enkephalin) are five amino
acid peptides diverging only in their last residue (Tyr–Gly–Gly–Phe–
Leu and Tyr–Gly–Gly–Phe–Met). They are produced by proteolysis of
the pro-peptide precursor proenkephalin and were first isolated in
spectrometer. Loading of the resin (mmol/g) 5 Absorbance (301 nm) 3
3
1
0 3 20/(7800 3 weight of the aliquot in mg).
(
B) After the initial loading, the remaining free sites were protected
[10]
1
975 by Hughes and colleagues.
Similar to most peptides naturally
using equal amounts of benzoyl chloride and pyridine. The solution was
agitated in a shaker for 3 h.
produced in the central nervous system, the enkephalins have a poor
pharmacokinetic profile and cannot be used as analgesics. When
injected in the peripheral system, hydrophilic opioid peptides are
unable to cross the blood–brain barrier (BBB) to reach the receptors in
the central nervous system and they are particularly sensitive to enzy-
(
C) For Fmoc deprotection, the resin was treated with 50% piperi-
dine in DMF, and the suspension was agitated in a shaker for 30 min.
D) All couplings were performed using 6 eq of protected amino
(
acid, and 6 eq of HATU (or COMU to couple Fmoc-Phe during the syn-
theses of 5 and 10) with 12 eq of NMM in a minimum volume of DMF
and the suspension was agitated in a shaker. When other coupling pro-
cedures were used, they are described in the “Experimental” section.
All coupling procedures were stopped after 16 h or after negativity of
the Kaiser’s test.
[
9,11,12]
matic degradation.
In the message-address concept first introduced by Robert
[
13]
Schwyzer in 1977,
the “message” portion of the peptide is responsi-
ble for the pharmacological activity, whereas the “address” is responsi-
ble for the selectivity of the molecule for its target. In Leu- and Met-
enkephalin the “message” contains the pharmacophore residues tyro-
sine and phenylalanine linked together by two glycine residues. The
remaining leucine is viewed as the residue responsible for the specific-
(
E) All the final peptides were cleaved from their resin in a glass
vial, and the suspension was stirred for 1.5 h with a magnetic stirrer. All
cleavage solutions were done with TFA/TIPS/H O (95:2.5:2.5) with a
2
[9]
ity (ie the “address”; Figure 1). We have shown previously that a sys-
tematic replacement of the amide bonds contained in Leu-enkephalin
was useful to explore and understand their role and involvement in the
total amount of 10 mL/g of resin. After cleavage, the mixtures were fil-
tered on cotton and dropped in a large amount of water (20 mL). The
remaining solvents were concentrated under vacuum, and the aqueous
solution was frozen and lyophilized.
[
12,14–16]
interaction of Leu-enkephalin with DOP.
In the present study,
we investigated the role of the fifth position of Leu-enkephalin in bind-
ing and activating DOP and MOP. The systematic replacement of the
fifth residue of Leu-enkephalin with non-natural amino acids was
therefore used to further explore the lipophilic pocket of MOP and
DOP receiving the “address” portion of Leu-enkephalin.
(F) Purification and purity requirements: All crude peptides were
purified using prepara1tive reverse-phase HPLC, detected at 280 nm,
with a VYDAC 218TP C18 column and using ACN gradient in a 0.1%
TFA aq solution (from 1:9 to 2:3), with a flow rate of 5 mL/min over
1 h. The purity of all fractions was analyzed using an Agilent 1100
series analytical HPLC, detection was performed at 214, 254, and
275 nm, with an Agilent 5 mm 3.0 3 50 mm C-18 column using ACN
gradient in a 0.1% TFA aq solution (from 0:1 to 1:0) over 40 min, with
a flow rate of 1 mL/min. The fractions with purities of 95% or higher
were combined, frozen and lyophilized.
2
2
|
EXPERIMENTAL PROCEDURES
.1
|
General procedure for peptide synthesis on solid
support
Between each step, the resin was washed using DMF (33), iPrOH (33)
and DCM (33), unless stated otherwise.
2
.2
|
Cell culture
(
A) For each synthesis, the loading of the Wang resin was deter-
HEK293 (human embryonic kidney) cells were grown at 378C in a
humidified atmosphere of 95% air and 5% CO2 in DMEM supple-
mented with 10% fetal bovine serum and 10 IU penicillin, 100 mg/L
streptomycin and 2 mM glutamine. Cells stably expressing the mouse
Flag-DOP obtained from Dr. Richard Howells (New Jersey Medical
mined as follows: Resin (1 g, 0.07 mmol/g supplier loading) was placed
in a sintered glass peptide synthesis vessel. 2 eq of the first amino acid,
2
eq of 2,6-dichlorobenzoyl chloride, and 4 eq of pyridine were added.
The suspension was agitated in a shaker for 16 h. The resin was

BELLA NDONG ET AL.
|
3 of 10
School, Newark, NJ) or human Flag-MOP obtained from Dr. Mark von
Zastrow (University of California, San Francisco, CA) were grown in the
presence of 250 mg/L geneticin.
4.2 mM KCl, 146 mM NaCl, 5.5 mM glucose, pH 7.4). Coelenterazine
400A (an Rluc2 substrate) was added to a final concentration of 5 lM,
10 min before stimulation. For the EPAC assay, cells were first stimu-
lated with 3 lM forskolin to increase cAMP. Cells were stimulated with
ligand concentrations ranging from 1 pM to 10 mM (DOP) or 100 pM
to 1 mM (MOP) and incubated for 10 min prior to the signal acquisi-
tion. BRET2 signals were measured using a TECAN M1000 fluores-
cence reader (TECAN). RLuc2 and GFP10 emissions were collected in
the 400–450 nm window (RLuc2) and 500–550 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 RLuc2. All data
were analyzed using the nonlinear curve fitting equation in GraphPad
Prism 7.0 to estimate the pEC50 values of the curves for the different
pathways. Results are expressed as the mean 6 SEM (nM or mM) of
four to five independent experiments, each performed in triplicate.
2
.3
|
Radioligand competition binding studies
Stably transfected HEK293 cells grown to confluence in 150-mm Petri
dishes were frozen at 2808C until use. On the day of the experiment,
the cells were submitted to a heat shock by placing the Petri dishes at
3
78C for 60 s before returning on ice. The cells were then harvested in
ice-cold buffer A (20 mM Tris-HCl, 5 mM MgCl , 150 mM NaCl, pH
.4) using a cell scraper and centrifuged at 3200g for 15 min at 48C.
2
7
The pellet containing the membrane extract was re-suspended in 1 mL
of buffer A. The protein concentration was determined with Bio-Rad
TM
DC Protein Assay reagents (Bio-Rad Laboratories, Mississauga, ON,
Canada) on a TECAN M200 plate reader (TECAN, Gr o€ dig, Austria) and
the pellet was further diluted in buffer A containing 0.1% BSA. Experi-
ments were performed using 10 mg of the cell preparation and 150 000
2
.5
|
Transduction coefficients and bias factors
determination
1
25
cpm of the radiolabeled ligand (we used [ I]-Deltorphin I (specific
1
25
Transduction coefficients and bias factors (BF) were derived using the
activity 472 Ci/mmol) for DOP or [ I]-DAMGO (specific activity
200 Ci/mmol) for MOP) in a total volume of 200 mL per well in 96-
[
17]
method developed by Kenakin et al.,
as detailed by van der West-
Briefly, transduction coefficients (log(s/K )),
where s (tau) represents ligand efficacy to induce a response of interest
and K its “functional affinity,” were first derived using the operational
model equation. To this end, the equation allowing calculation of con-
2
[
18]
huizen and colleagues.
A
well plates. Non-specific binding was determined using 10 lM non-
radioactive Deltorphin II or DAMGO. Incubation was performed for 60
min at room temperature with increasing concentrations of Leu-
enkephalin or compound 1–12 (1 pM to 10 lM). After the incubation
period, 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
A
[
18]
solidated log(s/K
was introduced in GraphPad Prism 7.0. Leu-enkephalin, the natural
DOP agonist, was taken as the reference compound to yield Dlog(s/K
A
) ratios
(here, all compounds were full agonists)
A
)
values and were used to calculate differential ligand efficiency to
engage cAMP signaling versus b-arrestin 2 recruitment by subtracting
using a nonlinear fitting analysis, and the K
i
values were determined
is
Dlog(s/K
A
) values of one signaling pathway from the Dlog(s/K
A
) of the
using GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA). K
i
other. BF correspond to antilogarithm base 10 of DDlog(s/K
A
) values.
expressed as the mean6 SEM (nM) from three independent experi-
ments, each performed in duplicate.
2
.6
|
Homoleucine analog (1)
2
.4
|
Rluc2-GFP10 BRET2-based biosensor assays
1.00 g of resin with a loading of 0.669 mmol/g was used. The title pep-
1
tide was obtained as a white solid (115 mg, 30%). H NMR (300 MHz,
For BRET assays, HEK293 cells were transfected with the receptor and
the adequate biosensor. Briefly, one day before transfection cells were
washed with PBS at room temperature, trypsinized, and seeded at 3
3
CD OD) d (ppm) 8.50 (d, 2H, J 5 8.0 Hz), 7.90 (d, 2H, J 5 8.0 Hz),
7
.29–7.12 (m, 5H), 7.11 (d, 2H, J 5 3.0 Hz), 6.82 (d, 2H, J 5 3.0 Hz),
.76–4.71 (m, 1H), 4.36–4.30 (m, 1H), 4.09 (dd, 2H, J 5 6.5 and 8.5
4
0
00 000 cells in a 100-mm Petri dish. For Rluc2-EPAC-GFP10 trans-
Hz), 3.95–3.72 (m, 4H), 3.17–3.14 (m, 2H), 3.00 (dd, 2H, J 5 9.0 and
fection, 6000 ng of pcDNA3-Flag-ratDOP with 600 ng of Rluc2-EPAC-
GPF10 or 3000 ng of pcDNA3-3HA-ratMOP with 150 ng of Rluc2-
EPAC-GPF10 was added to 600 lL of 150 mM NaCl containing 27 lg
of PEI. ssDNA was added to complement to 9 lg the total amount of
DNA used in each transfection. For Arr2-GFP10/receptor-Rluc2 trans-
fection, 600 ng of pcDNA3-Flag-ratDOP-Rluc2 or pcDNA3-3HA-
ratMOP-Rluc2 with 12 000 ng Arr2-GFP10 was added to 600 mL of
1
2.0 Hz), 1.89–1.83 (m, 1H), 1.75–1.70 (m, 1H), 1.59–1.52 (m, 1H),
1
3
1
.31–1.23 (m, 2H), 0.93 (d, 6H, J 5 6.5 Hz). C NMR (75 MHz,
OD) d (ppm) 173.8, 172.3, 170.0, 169.6, 156.9, 136.9, 130.1,
29.0, 128.0, 126.3, 124.5, 115.5, 54.7, 54.2, 52.6, 42.5, 41.7, 37.5,
CD
3
1
3
1
5
21
6.2, 34.5, 29.1, 27.5, 21.5, 21.2. IR (NaCl) m (cm ) 3600–3100 (br),
660, 1518. HRMS: calculated for C29 : 570.2922, found:
39 7 5
H O N
20
70.2922. [a] D 133.6 (c 5 5.99, MeOH).
1
50 mM NaCl containing 37.8 mg PEI. The mixture was incubated for
2
0 min before to being added to the cultured cells. At 24 h post-
2
.7
|
Cyclopropyl alanine analog (2)
transfection, cells were washed with PBS, trypsinized and plated in 96-
well white plates (75 000 cells/well) and grown for another 24 h. The
cells were then equilibrated at room temperature for at least 1 h with
1.00 g of resin with a loading of 0.651 mmol/g was used. The title pep-
1
tide was obtained as a white solid (100 mg, 28%). H NMR (300 MHz,
7
0 mL stimulation buffer (10 mM Hepes, 1 mM CaCl
2
, 0.5 mM MgCl
2
,
3
CD OD) d (ppm) 7.28–7.16 (m, 5H), 7.07 (d, 2H, J 5 8.5 Hz), 6.76 (d,

4
of 10
|
BELLA NDONG ET AL.
1
3
2
H, J 5 8.5 Hz), 4.70 (dd, 1H, J 5 5.0 and 10.0 Hz), 4.40 (dd, 1H,
J 5 7.0 Hz), 1.52–1.49 (m, 2H), 1.02 (s, 9H). C NMR (75 MHz, CDCl
3
)
J 5 5.5 and 8.0 Hz), 4.04 (dd, 2H, J 5 6.5 and 8.5 Hz), 3.97–3.68 (m,
d (ppm) 178.5, 155.9, 143.6, 141.3, 127.7, 127.1, 125.0, 120.0, 67.1,
2
1
4
0
H), 3.18–3.11 (m, 2H), 2.99–2.90 (m, 2H), 1.69–1.62 (m, 2H), 0.85–
51.6, 47.2, 45.8, 30.8, 17.6. IR (NaCl) m (cm ) 3500–3300 (br), 3069,
1
3
.70 (m, 1H), 0.46–0.41 (m, 2H), 0.13–0.05 (m, 2H). C NMR (75
OD) d (ppm)173.6, 172.1, 170.1, 169.6, 156.9, 136.9, 130.1,
28.0, 127.9, 126.3, 124.5, 115.4, 54.7, 54.3, 52.9, 47.3, 42.4, 41.7,
2958, 1720, 1447. HRMS: calculated for C22H25O N: 367.1784, found:
4
20
MHz, CD
3
367.1775. [a] D -17.0 (c 5 8.80, CHCl3).
1
3
1
21
7.5, 36.3, 7.1, 3.7, 3.2. IR (NaCl) m (cm ) 3500–3100 (br), 1653,
2.11
|
Neopentylglycine analog (5)
518. HRMS: calculated for C28 : 554.2609, found: 554.2610.
H O N
35 7 5
2
0
[a] D 121.8 (c 5 5.67, MeOH).
1
.00 g of resin with a loading of 0.656 mmol/g was used. The title pep-
1
tide was obtained as a white solid (143 mg, 42%). H NMR (300 MHz,
CD OD) d (ppm), 8.60 (d, 2H, J 5 8.0 Hz), 7.80 (d, 2H, J 5 8.0 Hz),
2
.8
|
Cyclohexylalanine analog (3)
3
1
.00 g of resin with a loading of 0.143 mmol/g was used. The title pep-
7.33–7.26 (m, 5H), 7.12 (d, 2H, J 5 8.0 Hz), 6.77 (d, 2H, J 5 8.5 Hz),
4.72–4.59 (m, 1H), 4.40–4.35 (m, 1H), 4.06 (dd, 2H, J 5 6.5 and 8.0
Hz), 3.97–3.67 (m, 4H), 3.20–3.07 (m, 2H), 2.99–2.85 (m, 2H), 1.83 (dd,
1H, J 5 3.0 and 14.5 Hz), 1.65 (dd, 1H, J 5 9.0 and 14.5 Hz), 0.92 (s,
1
tide was obtained as a white solid (163 mg, 100%). H NMR (300 MHz,
CD
3
OD) d (ppm) 8.52 (d, 2H, J 5 8.0 Hz), 7.84 (d, 2H, J 5 8.0 Hz),
.27–7.10 (m, 5H), 7.08 (d, 2H, J 5 2.0 Hz), 6.76 (d, 2H, J 5 2.0 Hz),
.81–4.67 (m, 1H), 4.42–4.39 (m, 1H), 4.05 (dd, 1H, J 5 6.5 and 8.5
7
4
1
3
9H), 0.46–0.41 (m, 2H), 0.13–0.05 (m, 2H). C NMR (75 MHz,
CD OD) d (ppm)174.6, 171.9, 170.0, 169.6, 156.9, 136.9, 130.1, 128.9,
127.9, 126.3, 124.5, 115.4, 54.7, 54.2, 48.1, 47.6, 47.0, 46.7, 44.4,
Hz), 3.96–3.67 (m, 5H), 3.18–3.12 (m, 2H), 2.94 (dd, 2H, J 5 9.5 and
3
1
3
1
4.0 Hz), 1.79–1.58 (m, 9H), 1.32–1.14 (m, 2H), 1.01–0.83 (m, 1H).
NMR (75 MHz, CD OD) d (ppm) 174.3, 172.3, 170.0, 169.7, 169.6,
56.9, 136.9, 130.1, 129.0, 128.0, 126.3, 124.5, 115.4, 54.7, 54.2,
0.1, 42.5, 41.7, 38.6, 37.5, 36.2, 33.9, 33.3, 31.8, 25.9. IR (NaCl) m
C
21
3
42.5, 41.2, 37.4, 36.3, 30.1, 28.5. IR (NaCl) m (cm ) 3500–3100 (br),
1
5
1669, 1517, 1372. HRMS: calculated for C29H39O N : 570.2922,
7
5
2
0
found: 570.2935. [a] D 115.3 (c 5 5.30, MeOH).
21
(
cm ) 3600-3300 (br), 1655, 1517, 1206, 803. HRMS: calculated for
2
0
C
31 41 7 5
H O N : 596.3079, found: 596.3091. [a] D 117.3 (c 5 6.54,
MeOH).
2.12
|
Azabeta-leucine analog (6)
1
.00 g of resin with a loading of 0.691 mmol/g was used. The title pep-
2
.9
|
4,5-Dehydro leucine analog (4)
1
tide was obtained as a white solid (171 mg, 43%). H NMR (300 MHz,
1
.00 g of resin with a loading of 0.7mmol/g was used. The title peptide
3
CD OD) d (ppm) 7.25–7.18 (m, 5H), 7.05 (d, 2H, J 5 8.0 Hz), 6.75 (d,
1
was obtained as a white solid (58.6 mg, 15%). H NMR (300 MHz,
2H, J 5 8.0 Hz), 4.55–4.40 (m, 1H), 4.05 (dd, 2H, J 5 6.5 and 8.0 Hz),
3.86–3.79 (m, 4H), 3.24–3.08 (m, 2H), 3.07–2.96 (m, 3H), 2.48–2.39
CD
3
OD) d (ppm) 7.30–7.15 (m, 5H), 7.09 (d, 2H, J 5 8.5 Hz), 6.77 (d,
H, J 5 8.5 Hz), 4.74–4.59 (m, 1H), 4.58–4.42 (m, 1H), 4.11–4.06 (m,
H), 4.07 (s, 3H), 3.95 (d, 2H, J 5 9.5 Hz), 3.72–3.59 (m, 4H), 3.21–3.08
1
3
2
1
(m, 3H), 1.40–1.25 (m, 1H), 0.96–0.76 (m, 6H). C NMR (75 MHz,
CD OD) d (ppm) 172.0, 170.1, 169.6, 156.9, 136.3, 130.1, 129.0,
3
(
m, 2H), 3.01–2.87 (m, 2H), 2.63–2.540 (m, 1H), 2.44–2.30 (m, 1H),
128.2, 126.5, 124.5, 115.4, 64.1, 58.1, 54.7, 53.4, 48.1, 47.7, 46.9,
21
1
3
2
.12–1.99 (m, 1H), 1.78 (s, 2H), 1.49 (s, 1H), 1.41 (s, 1H). C NMR (75
OD) d (ppm) 156.9, 140.6, 136.8, 130.1, 128.9, 128.1, 127.9,
26.4, 126.3, 116.7, 115.4, 112.9, 82.3, 54.3, 50.8, 46.7, 41.9, 39.2,
42.3, 41.7, 37.4, 36.3, 2537, 19.5. IR (NaCl) m (cm ) 3500–3100 (br),
MHz, CD
3
38 7 6
1668, 1518, 1370, 996. HRMS: calculated for C28H O N : 571.2875,
2
0
1
3
1
5
found: 571.2877. [a] D 115.1 (c 5 6.78, MeOH).
21
7.5, 36.3, 27.6, 25.8, 20.8. IR (NaCl) m (cm ) 3500–3100 (br), 1664,
515, 836. HRMS: calculated for C28 : 554.2209, found:
35 7 5
H O N
2
0
|
54.2615. [a] D 118.4 (c 5 7.24, MeOH).
2.13 Isoleucine analog (7)
1
.00 g of resin with a loading of 0.849 mmol/g was used. The peptide
2
.10
|
Fmoc carbamate of L-neopentylglycine
1
was obtained as a white solid (16.8 mg, 3%). H NMR (300 MHz,
3
CD OD) d (ppm) 8.35 (d, 1H, J 5 8.0 Hz), 7.87 (d, 1H, J 5 8.0 Hz),
The unnatural amino acid L-neopentylglycine (692 mg, 4.77 mmol) and
NaHCO (s) were dissolved in an aqueous solution of H O/THF (7:13,
0 mL) at room temperature. Fmoc-Cl (1.85 g, 7.15 mmol) was added
7
4
.21–7.13 (m,5H), 7.04 (d, 2H, J 5 6.0 Hz), 6.74 (d, 2H, J 5 6.0 Hz),
.76 (dd, 1H, J 5 5.0 and 9.0 Hz), 4.32–4.28 (m, 1H), 4.08 (dd, 1H,
3
2
7
to the previous solution slowly under stirring at 08C. The reaction was
J 5 6.5 and 8.0 Hz), 3.96–3.68 (m, 4H), 3.17–3.10 (m, 2H), 2.98–2.90
(m, 2H), 1.89–1.81 (m, 1H), 1.53–1.45 (m, 1H), 1.28–1.21 (m, 1H),
allowed to reach rt and stirred for 5 h. The aqueous phase was acidified
1
3
2
with 1 N HCl to reach pH 2 after the addition of H O (100 mL) and
0.95–0.86 (m, 6H). C NMR (75 MHz, CD
3
OD) d (ppm) 189.5, 170.1,
extracted with EtOAc (3 3 30 mL). The organic phases were collected
169.9, 169.7, 152.8, 136.9, 130.1, 128.9, 128.0, 126.3, 115.4, 54.8,
21
4
together, dried (MgSO ) and concentrated under vacuum. The title
54.5, 42.7, 41.8, 37.4, 37.1, 36.3, 24.8, 14.7, 10.5. IR (NaCl) m (cm
)
1
product was obtained as a white solid (1.06 g, 60%). H NMR (300
3706–2822 (br), 1664, 1534, 1197. HRMS: calculated for
2
0
3
MHz, CDCl ) d (ppm) 7.76 (d, 2H, J 5 7.0 Hz), 7.58 (d, 2H, J 5 7.0 Hz),
28 37 5 7
C H N O : 578.2585, found: 578.2570. [a] D 123.9 (c 5 3.89,
7
.43–7.26 (m, 4H), 5.15 (br, 2H), 4.43 (d, 1H, J 5 2.0 Hz), 4.24 (t, 1H,
MeOH).

BELLA NDONG ET AL.
|
5 of 10
2
.14
|
D-isoleucine analog (8)
1
.00 g of resin with a loading of 0.665 mmol/g was used. The title pep-
1
tide was obtained as a white solid (31.9 mg, 9%). H NMR (300 MHz,
CD
3
OD) d (ppm) 7.30–7.26 (m, 5H), 7.08 (d, 2H, J 5 8.5 Hz), 6.76 (d,
H, J 5 8.5 Hz), 4.22–4.18 (m, 1H), 4.15–4.01 (m, 1H), 3.99–3.92 (m,
H), 3.20–3.03 (m, 2H), 2.99–2.85 (m, 2H), 1.77–1.60 (m, 1H), 0.84–
2
5
0
1
4
1
3
3
.77 (m, 8H). C NMR (75 MHz, CD OD) d (ppm) 169.5, 130.1, 129.0,
28.1, 126.4, 124.5, 115.5, 54.7, 54.3, 48.2, 48.0, 47.9, 47.7, 47.6,
7.5, 47.3, 47.1, 46.8, 42.5, 37.9, 36.7, 36.2, 24.6, 14.4. IR (NaCl) m
21
(
cm ) 3500–3100 (br), 1666, 1520, 1443. HRMS: calculated for
2
0
C
28 37 7 5
H O N : 556.2766, found: 556.2767. [a] D 116.6 (c 5 5.60,
MeOH).
2
.15
|
D-alloisoleucine analog (9)
1
.00 g of resin with a loading of 0.7 mmol/g was used. The title peptide
1
was obtained as a white solid (107 mg, 28%). H NMR (300 MHz,
3
CD OD) d (ppm) 7.40–7.17 (m, 5H), 7.10 (d, 2H, J 5 8.0 Hz), 6.77 (d,
2
1
2
H, J 5 8.0 Hz), 4.87–4.72 (m, 1H), 4.40 (d, 1H, J 5 5.0 Hz), 4.10 (dd,
H, J 5 6.5 and 8.0 Hz), 3.99–3.78 (m, 4H), 3.28–3.04 (m, 2H), 3.01–
.87 (m, 2H), 1.95–1.79 (m, 1H), 1.19–0.93 (m, 2H) 0.87–0.72 (m, 6H).
FIGURE 2 Non-natural amino acids replacing the 5th residue of
the Leu-enkephalin. They are classified as families (d, g, b, a, N)
according to their branching positions
1
3
3
C NMR (75 MHz, CD OD) d (ppm) 173.4, 172.1, 170.2, 169.6, 156.9,
1
4
3
5
36.7, 130.1, 128.9, 128.1, 126.4, 124.5, 115.4, 55.4, 54.7, 54.3, 47.2,
2
1
m (cm ) 3500–3200 (br), 1665, 1517, 1452. HRMS: calculated for
21
5.2, 41.9, 37.9, 36.7, 36.2, 25.6, 13.5, 10.6. IR (NaCl) m (cm ) 3500–
2
0
29 37 7 5
C H O N : 568.2766, found: 568.2762. [a] D 18.55 (c 5 6.49,
37 7 5
100 (br), 1660, 1514, 1443. HRMS: calculated for C28H O N :
MeOH).
2
0
56.2766, found: 556.2770. [a] D 117.1 (c 5 5.75, MeOH).
2
.18
|
Homoproline analog (12)
2
.16
|
Cycloleucine analog (10)
1
.00 g of resin with a loading of 0.59 mmol/g were used. The title pep-
5
00 mg of resin (1.5 mmol/gr factory loading) with an arbitrary loading
1
tide was obtained as a white solid (13.0 mg, 4%). H NMR (300 MHz,
of 1.5 mmol/g was used. The title peptide was obtained as a white
3
CD OD) d (ppm) 7.28–7.19 (m, 5H), 7.09 (d, 2H, J 5 8.5 Hz), 6.76 (d,
1
solid (6.80 mg, 2%). H NMR (300 MHz, CD
3
OD) d (ppm) 8.45 (s, 2H),
2
H, J 5 8.5 Hz), 4.59–4.51 (m, 1H), 4.06 (dd, 1H, J 5 6.5 and 8.0 Hz),
.99–3.70 (m, 5H), 3.46–3.33 (m, 3H), 3.18–3.04 (m, 3H), 2.98–2.88
7
6
8
2
.82 (d, 2H, J 5 8.0 Hz), 7.30–7.28 (m, 5H), 7.14 (d, 2H, J 5 8.0 Hz),
3
.81 (d, 2H, J 5 8.0 Hz), 4.69–4.61 (m, 1H), 4.11 (dd, 1H, J 5 6.5 and
.0 Hz), 3.95–3.72 (m, 4H), 3.18–3.10 (m, 2H), 3.01–2.93 (m, 2H),
13
(
m, 3H), 2.37 (td, 2H, J 5 6.5 and 2.5 Hz), 1.50–0.80 (m, 1H). C NMR
75 MHz, CD OD) d (ppm) 174.1, 171.9, 170.2, 169.6, 156.9, 136.8,
30.1, 128.9, 128.0, 126.4, 124.5, 115.5, 54.7, 42.4, 41.9, 39.1, 37.6,
(
3
1
3
.20–2.02 (m, 2H), 1.98–1.79 (m, 2H), 1.74–1.57 (m, 4H). C NMR (75
OD) d (ppm) 174.1, 173.5, 140.8, 134.1, 133.0, 131.9, 130.3,
28.4, 119.4, 58.7, 57.9, 52.1, 51.8, 51.5, 51.3, 50.9, 50.7, 46.5, 45.8,
1
MHz, CD
3
21
3
6.3, 34.9, 32.9. IR (NaCl) m (cm ) 3500–3100 (br), 1665, 1443, 1213.
1
4
8
2
0
35 7 5
HRMS: calculated for C28H O N : 553.2537, found: 553.2543. [a] D
21
0.2, 40.1, 27.9. IR (NaCl) m (cm ) 3600–3300 (br), 1655, 1517, 1206,
1
2.33 (c 5 1.13, MeOH).
03. HRMS: calculated for C28 : 554.2609, found: 554.2607.
35 7 5
H O N
2
0
[
a] D 124.8 (c 5 2.55, MeOH).
3
|
RESULTS AND DISCUSSION
2
.17
|
Homocycloleucine analog (11)
As already stated, Leu-enkephalin binds to both DOP and MOP with
similar affinity. Aiming to further explore the “address” portion of Leu-
enkephalin, this is to say the fifth amino acid residue, a total of 12 ana-
logs were synthesized and tested for their potency and efficacy at
DOP and MOP. These pentapeptides share the same Tyr–Gly–Gly–
Phe “message” as natural enkephalins, whereas the amino acid Leu
forming the “address” region has been replaced by similar lipophilic res-
idues that is non-natural amino acid derivatives of leucine, alanine, or
proline (Figure 2). The analogs can be classified into five distinct fami-
lies, namely d, g, b, a, and N according to their branching positions. The
1
.00 g of resin with a loading of 0.258 mmol/g was used. The title pep-
1
tide was obtained as a white solid (19.0 mg, 13%). H NMR (300 MHz,
CD
3
OD) d (ppm) 8.20 (s, 2H), 7.75 (d, 2H, J 5 8.0 Hz), 7.31–7.15 (m,
H), 7.12 (d, 2H, J 5 8.0 Hz), 6.81 (d, 2H, J 5 8.0 Hz), 4.80–4.75 (m,
H), 4.05 (dd, 1H, J 5 6.0 and 8.5 Hz), 3.95–3.60 (m, 4H), 3.16–3.01
5
1
(
m, 2H), 2.99–2.80 (m, 2H), 2.03–1.81 (m, 2H), 1.80–1.65 (m, 2H),
1
3
1
1
5
3
.60–1.23 (m, 6H). C NMR (75 MHz, CD OD) d (ppm) 176.2, 171.7,
70.1, 169.5, 156.9, 136.9, 130.1, 129.0, 128.0, 126.3, 124.5, 115.5,
8.9, 54.7, 54.0, 42.6, 41.7, 37.6, 36.2, 31.8, 31.6, 24.9, 20.9. IR (NaCl)

6
of 10
|
BELLA NDONG ET AL.
3
.2
|
Synthesis
The classical methodology of solid phase peptide synthesis (SPPS) was
used unless otherwise stated. For the analogs 5 and 10, (1-Cyano-2-
ethoxy-2-oxoethylidenaminooxy)dimethylamino-morpholino-carbenium
hexafluorophosphate (COMU), an excellent alternative to 1-[Bis(dime-
thylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexa-
fluorophosphate (HATU) in the case of sterically demanding peptide
[
19,20]
bonds,
was used for the coupling between phenylalanine and the
non-natural residue.
3
.3
|
Binding affinity of leu-enkephalin analogs
To evaluate the affinity of the synthesized analogs for DOP and MOP,
we performed competitive binding assays using HEK293/DOP and
1
25
HEK293/MOP cell membrane preparations with radiolabeled
I-Del-
1
25
torphin I (DOP-selective ligand) and
ligand).
I-DAMGO (MOP-selective
As shown in Table 1, most compounds, with the exception of ana-
logs 8 and 9, bind MOP with a slightly better affinity than DOP. Analog
12, containing a homoproline in position 5 displays very low affinity for
both receptors. This low affinity is likely the consequence of detrimen-
tal conformational constraints or unfavorable interactions imposed by
the ring since we have previously show that the N-methylated analog
Tyr–Gly–Gly–Phe–NMe–Leu has a comparable affinity for DOP when
[
12]
compared to Leu-enkephalin.
When compared to the other com-
pounds, analog 6 was found to have a slightly lower affinity for both
receptors. The geometry of the modified residue substantially differs
from Leu and bears a positive charge that might induce unfavorable
SCHEME 1 SPPS of compounds 1–12 using the Fmoc strategy
with the Wang resin. Reaction conditions: (a) Initial coupling, (b)
protection of the remaining free sites of the resin, (c) Fmoc
deprotection (Pip/DMF 1:1), 30 min at room temperature (r.t.) (d)
protected amino acid (AA) (6 eq); HATU or COMU (6 eq); NMM
TABLE 1 In Vitro binding affinity of compounds 1–12 at DOP and
(
12 eq) DMF, 16 h at r.t., and (e) TFA/TIPS/H2O (38:1:1), 1.5 h at
MOP
r.t
Family
Analog
K
i
d (nM)
i
K l (nM)
branching pattern of the most numerous g group (2–6) is the same as
Leu in Leu-enkephalin and the fifth residue in 1 and 3 are, respectively,
as long or slightly longer than a Met residue. For the b, a and N fami-
lies, residues were chosen because the length of their side chains is
either equal (7–9 and 11) or slightly shorter (10 and 12) than that of a
Leu residue. Thus, all amino acid residues were selected in order to
explore the steric requirements of the hydrophobic pocket naturally
filled by the side chains of Leu and Met. Additionally, the effect of a
positive charge in the address was studied in the case of the aza-
b-homoleucine derivative 6.
g
d
g
LENK
0.64 6 0.29
0.71 6 0.20
0.95 6 0.18
0.43 6 0.08
2.71 6 0.42
2.53 6 0.57
16.4 6 3.5
0.20 6 0.09
0.30 6 0.08
0.22 6 0.07
0.15 6 0.06
0.67 6 0.18
0.57 6 0.21
4.6 6 1.6
Homoleucine 1
Cyclopropylalanine 2
Cyclohexylalanine 3
4
,5-dehydroleucine 4
Neopentylglycine 5
Aza-b-homoleucine 6
Isoleucine 7
b
0.58 6 0.13
0.63 6 0.17
1.19 6 0.33
1.86 6 0.49
1.62 6 0.52
2220 6 220
0.44 6 0.05
0.95 6 0.22
1.17 6 0.98
0.82 6 0.53
0.90 6 0.37
69 6 17
D-isoleucine 8
D-allo-isoleucine 9
Cycloleucine 10
Homocycloleucine 11
Homoproline 12
3
.1
|
Chemistry
a
The twelve analogs (1–12) and Leu-enkephalin were synthesized on
solid support using the Fmoc strategy on Wang resin (Scheme 1).
Neopentylglycine was purchased and Fmoc protected before being
used to obtain the corresponding analog (5). Fmoc-aza-b-leucine
was synthesized and attached to the Wang resin to produce analog
N
125
The K
: 0.6 nM; specific activity: 472 Ci/mmol), [ I]-DAMGO (K
specific activity: 2200 Ci/mmol) and stable HEK293 cell lines expressing
mouse DOP or human MOP. K are the means 6 SEM (nM) of three inde-
pendent experiments each performed in duplicate.
i
value of each compound was determined using [ I]-Deltorphin I
125
(
K
d
d
: 0.12 nM;
(
6). All other residues were purchased N-protected with an Fmoc
i
group.

BELLA NDONG ET AL.
|
7 of 10
TABLE 2 In Vitro potency of analogs 1–12 on DOP and MOP
DOP
MOP
EPAC
bArr2
EPAC
bArr2
Analog
(EC50; nM)
Emax (% ref)
100 6 0
98 6 1
96 6 6
95 6 4
89 64
93 6 2
90 6 8
86 6 4
98 6 3
956 1
103 6 4
89 6 3
69 6 6
(EC50; nM)
Emax (% ref)
100 6 0
92 6 7
(EC50; nM)
Emax (% ref)
100 6 0
95 6 3
98 6 9
96 6 10
95 6 6
98 6 2
95 6 3
88 6 2
95 6 4
92 6 2
93 6 3
96 6 6
86 6 4
(EC50; mM)
Emax (% ref)
100 6 0
93 6 3
LENK
0.24 6 0.09
1.54 6 0.03
0.62 6 0.09
0.14 6 0.06
1.71 6 0.57
0.39 6 0.14
4.0 6 1.6
13.2 6 4.1
8.2 6 1.8
1.2 6 0.2
1.4 6 0.6
1.6 6 0.7
1.9 6 0.7
3.7 6 1.7
19.9 6 7.1
39.4 6 10.3
2.6 6 0.7
19.4 6 8.3
13.2 6 4.3
9.9 6 4.0
9.9 6 2.9
324 6 179
0.7 6 0.2
0.8 6 0.2
0.8 6 0.3
0.8 6 0.4
1.0 6 0.1
5.9 6 1.5
15.1 6 7.3
0.6 6 0.2
6.5 6 2.9
3.3 6 0.7
3.7 6 0.9
3.4 6 1.7
No signal
Homoleucine 1
Cyclopropylalanine 2
Cyclohexylalanine 3
27.8 6 3.6
4.8 6 1.7
77 6 5
102 6 5
95 6 4
73 6 5
4
,5-d eꢀ hydroleucine 4
61 6 18
90 6 7
84 6 8
Neopentylglycine 5
Aza-b-homoleucine 6
Isoleucine 7
50.5 6 6.8
814 6 64
15.3 6 4.3
22.2 6 4.6
41.6 6 10.9
95.2 6 10.2
48.7 6 10.8
No signal
95 6 4
87 6 8
64 6 6
81 6 9
0.51 6 0.19
0.42 6 0.11
1.1 6 0.7
90 6 5
88 6 6
D-isoleucine 8
82 6 4
101 6 16
84 6 7
D-allo-isoleucine 9
Cycloleucine 10
Homocycloleucine 11
Homoproline 12
79 6 7
0.57 6 0.13
1.34 6 0.41
276 6 111
85 6 6
84 6 3
78 6 2
97 6 6
No signal
No signal
The potency (EC50; nM or mM) of compounds 1–12 to inhibit the forskolin-induced cAMP production or to recruit b-Arrestin 2 following activation of
DOP or MOP was respectively assayed using Rluc2-EPAC-GPF10 biosensor or bArr2-GPF10 biosensor with a Rluc2 tagged version of DOP or MOP.
Values are the means 6 SEM (nM or mM) of at least three independent experiments each performed in triplicate. The efficacy of each analog (Emax)
was also determined as the percentage of the maximum activation achieved by the reference compound, LENK (% ref).
7
.35
interactions with the receptor. Indeed, the natural residues Leu and
Met at position 5 are less hydrophilic than aza-b-homoleucine. The
d-branched analog 1 retains a very good affinity for the opioid recep-
tors, so does the cyclohexylalanine analog 3, containing a g-branched
residue. These peptides confirm that there is a fairly large available
space in the corresponding region of DOP and MOP. This observation
is in agreement with the slightly higher affinity of Met-enkephalin, for
which the side chain has a similar length as homoleucine and cyclohex-
ylalanine. The three remaining g-branched residues cyclopropylalanine,
Leu300 . The exploration was pursued with two D isomers of Ile,
namely D-isoleucine and D-allo-isoleucine, incorporated in analogs 8 and
9, respectively. These analogs proved to be very effective with respective
i
binding affinity (K) values of 0.63 and 1.19 nM for DOP and 0.95 and
1.17 nM for MOP. Finally, the two a-branched analogs 10 and 11 bind
DOP and MOP with a slightly decreased affinity when compared to Leu-
enkephalin. It is worth mentioning that, under our experimental condi-
125
125
tions (using
I-Deltorphin I,
I-DAMGO and stable HEK293 cell lines
expressing DOP or MOP), Leu-enkephalin was found to bind MOP with
a slightly better affinity than DOP.
4
,5-dehydroleucine and neopentylglycine are highly similar to the leu-
cine residue. However, only analog 2, containing a cyclopropyl side
chain, mimics a locked i-propyl group as found in Leu-enkephalin. Inter-
estingly, the affinity and the potency of this analog for DOP and MOP
are similar to those of Leu-enkephalin (Tables 1, 2). The slightly less bulky
analog 4 sees its affinity decreased by ꢀ 3- to 4-fold for both DOP and
MOP when compared to Leu-enkephalin. This is presumably the conse-
quence of suboptimal van der Waals interactions with some residues in
the receptor pocket. The same remark applies to analog 5 containing a t-
butyl side chain. In this particular case, the decreased affinity is likely due
to excessive bulk and steric clashes. As for the b-branched residues, it
was shown that replacing leucine with a valine residue at position 5 of
We then evaluated the potency of all the synthesized analogs
using EPAC and bArr2 BRET biosensors. When an agonist binds to an
[
22,23]
opioid receptor, complex signaling pathways are activated.
Among
these pathways, agonist-stimulated opioid receptors couple to Gi het-
erotrimeric proteins inhibiting adenylyl cyclase and reducing cAMP lev-
els. Moreover, opioid receptors also interact with b-arrestins, which
[
22,23]
leads to receptor internalization.
The EPAC BRET biosensor was
used to measure the inhibition of cAMP production, while the investi-
gation of the b-arrestin pathway was assessed with bArr2 BRET bio-
sensor (Table 2). As compared to peptides 1–11, compound 12 shows
a much lower potency in inhibiting cAMP production through DOP and
MOP. Despite having different potencies, all analogs but 12 were con-
sidered full agonists in this pathway as they all induced a maximal
effect (Emax) similar to that of Leu-enkephalin (Table 2). As for the
b-arrestin pathway, all compounds but 12 were found to recruit
b-arrestin 2 upon DOP and MOP activation. Again, all compounds but
12 (for which no BRET signal was detected) produced a near maximal
[21]
the enkephalin leads to very low affinity analogs.
The poor affinity of
7.35
such analogs has been attributed to their inability to reach Leu300
residue of DOP and develop a stabilizing van der Waals interaction. On
the other hand, the longer b-branched residue Ile, found in analog 7, has
a similar affinity for DOP as does Leu-enkephalin. The extra methyl group
(vs. Val) is likely to reach the bottom of the pocket partially defined by

8
of 10
|
BELLA NDONG ET AL.
TABLE 3 Transduction coefficient and bias factor determination for compounds 1–12 following activation of DOP
EPAC
b
2
arr
2
EPAC/b arr
log(s/K
A
)
Dlog(s/K
A
)
log(s/K
A
)
Dlog(s/K
A
)
DDlog(s/K
A
)
BF
LENK
10.02 6 0.34
9.92 6 0.12
9.36 6 0.05
9.83 6 0.17
8.91 6 0.14
9.13 6 0.15
8.27 6 0.17
9.58 6 0.08
9.34 6 0.19
9.13 6 0.21
9.19 6 0.13
9.19 6 0.17
5.80 6 0.15
0.00 6 0.48
20.10 6 0.36
20.67 6 0.34
20.19 6 0.38
21.12 6 0.36
20.90 6 0.37
21.75 6 0.38
20.45 6 0.35
20.68 6 0.39
20.90 6 0.40
20.84 6 0.36
20.83 6 0.38
24.23 6 0.37
8.15 6 0.11
8.13 6 0.10
7.46 6 0.09
8.40 6 0.09
7.09 6 0.10
7.29 6 0.14
5.69 6 0.08
7.79 6 0.09
7.33 6 0.12
7.11 6 0.10
7.02 6 0.11
7.23 6 0.08
0.00 6 0.16
20.03 6 0.15
20.70 6 0.14
0.24 6 0.15
21.06 6 0.15
20.87 6 0.18
22.46 6 0.14
20.37 6 0.15
20.82 6 0.17
21.05 6 0.15
21.14 6 0.16
20.92 6 0.14
nd
0.00 6 0.49
20.08 6 0.39
0.03 6 0.38
20.43 6 0.39
20.06 6 0.39
20.03 6 0.41
0.71 6 0.39
20.08 6 0.38
0.14 6 0.42
0.15 6 0.42
0.30 6 0.39
0.09 6 0.41
nd
1.00
0.84
1.07
0.37
0.87
0.93
5.16
0.83
1.38
1.41
2.00
1.24
nd
Homoleucine 1
Cyclopropylalanine 2
Cyclohexylalanine 3
4
,5-d eꢀ hydroleucine 4
Neopentylglycine 5
Aza-b-homoleucine 6
Isoleucine 7
D-isoleucine 8
D-allo-isoleucine 9
Cycloleucine 10
Homocycloleucine 11
Homoproline 12
nd
Data presented in Table 2 were further analyzed by nonlinear regression using the Operational Model equation as described in the methods to deter-
mine the transduction coefficients (log(s/K )) and bias factors (BF) with LENK as the reference compound. DDlog(s/K ) and BF values were calculated
as described in the methods. Data are the mean 6 SEM of 3–6 independent experiments each performed in triplicate.
A
A
b-arrestin 2 response and were therefore considered full agonists in
this pathway as well.
b-arrestin 2 signaling bias. In particular ligand efficiency to evoke
response at each of these pathways was evaluated using the opera-
[
24]
Interestingly, although there was no significant gain in DOP selec-
tivity by replacing the fifth residue of Leu-enkephalin with various non-
natural amino acids, functional assays allowed us to reveal that several
of the new compounds displayed receptor-specific cAMP versus
tional model of Black and Leff
A
to calculate log(s/K ) transduction
coefficients (refer to the Experimental procedures for further details).
Transduction coefficients for each ligand at either DOP or MOP are
respectively reported in Tables 3 and 4 along with (Dlog(s/K
A
)) values
TABLE 4 Transduction coefficient and bias factor determination for compounds 1–12 following activation of MOP
EPAC
b
2
arr
2
EPAC/b arr
log(s/K
A
)
Dlog(s/K
A
)
log(s/K
A
)
Dlog(s/K
A
)
DDlog(s/K
A
)
BF
LENK
9.00 6 0.06
8.80 6 0.19
8.64 6 0.13
8.77 6 0.27
7.99 6 0.32
8.05 6 0.20
7.44 6 0.10
8.54 6 0.06
7.99 6 0.24
7.78 6 0.19
7.92 6 0.21
8.17 6 0.20
6.29 6 0.35
0.00 6 0.09
20.20 6 0.20
20.36 6 0.15
20.23 6 0.28
21.00 6 0.32
20.95 6 0.21
21.56 6 0.12
20.46 6 0.09
21.01 6 0.25
21.22 6 0.20
21.08 6 0.22
20.83 6 0.21
22.71 6 0.35
6.21 6 0.15
6.20 6 0.19
6.12 6 0.18
6.00 6 0.16
5.76 6 0.16
5.11 6 0.09
4.59 6 0.14
5.72 6 0.20
5.29 6 0.21
5.42 6 0.12
5.33 6 0.12
5.34 6 0.25
0.00 6 0.22
20.01 6 0.24
20.09 6 0.24
20.21 6 0.23
20.45 6 0.22
21.10 6 0.18
21.62 6 0.21
20.49 6 0.25
20.92 6 0.26
20.79 6 0.20
20.88 6 0.20
20.87 6 0.30
nd
0.00 6 0.24
20.19 6 0.32
20.27 6 0.28
20.02 6 0.36
20.56 6 0.39
0.15 6 0.27
0.06 6 0.24
0.02 6 0.27
20.09 6 0.36
20.44 6 0.28
20.20 6 0.29
0.03 6 0.36
nd
1.00
0.65
0.54
0.95
0.28
1.42
1.15
1.05
0.80
0.11
0.63
1.08
nd
Homoleucine 1
Cyclopropylalanine 2
Cyclohexylalanine 3
4
,5-d eꢀ hydroleucine 4
Neopentylglycine 5
Aza-b-homoleucine 6
Isoleucine 7
D-isoleucine 8
D-allo-isoleucine 9
Cycloleucine 10
Homocycloleucine 11
Homoproline 12
nd
Data presented in Table 2 were further analyzed by nonlinear regression using the Operational Model equation as described in the methods to deter-
mine the transduction coefficients (log(s/K )) and bias factors (BF) with LENK as the reference compound. DDlog(s/K ) and BF values were calculated
as described in the methods. Data are the mean 6 SEM of 3–6 independent experiments each performed in triplicate.
A
A

BELLA NDONG ET AL.
|
9 of 10
normalized to Leu-enkephalin. The extent to which each ligand prefer-
entially evoked G protein-dependent modulation of cAMP production
versus b-arrestin 2 recruitment was obtained by subtracting corre-
analog 9. DBN and APG synthesized the analogs 1–8 and 10–12. RL
and GP provided the expertise on the BRET biosensors. BG, LG, and
YLD designed the study. DBN, VB, WR, and BH performed the
experiments and wrote the manuscript. All authors have read and
approved the final version of the manuscript.
A A
sponding Dlog(s/K ) values to yield a bias factor (DDlog(s/K )). For the
purpose of this study, we considered a compound biased when the
bias factor (BF) was ꢁ2.00 (bias toward G protein-dependent cAMP
inhibition) or ꢂ0.50 (biased toward b-arrestin 2 recruitment). As shown
in Table 3, only compounds 6 (BF 5 5.16) and 10 (BF 5 2.00) were
biased toward preferential modulation of cAMP production following
DOP activation while only compound 3 (BF5 0.37) showed b-arrestin
ORCID
Louis Gendron
Yves L. Dory
http://orcid.org/0000-0002-2058-8863
http://orcid.org/0000-0003-4758-1589
http://orcid.org/0000-0001-6854-8003
http://orcid.org/0000-0002-0319-4512
Richard Leduc
Brigitte Gu ꢀe rin
2
recruitment bias at this receptor. This is of a particular interest when
one considers that opioid ligands are thought to produce analgesia
through activation of the G protein dependent-signals and unwanted
effects such as constipation and respiratory depression have been
REFERENCES
[
1] American Academy of Pain Medicine and the American Pain Soci-
[25–28]
associated to b-arrestin recruitement.
Compounds 6 and 10
ety, Clin J Pain 1997, 13, 6.
therefore suggest that slight modifications within the side chain of the
fifth amino acid residue has the potential to induce bias toward G pro-
tein coupling. By contrast, results from Table 4 reveal that none of the
compounds are biased toward the modulation of cAMP production
when MOP activation is considered. In fact, two compounds are rather
biased toward b-arrestin 2 signaling following activation of MOP.
Indeed, compounds 4 and 9 have BF values ꢂ0.50 and therefore their
activation of MOP can be considered biased toward b-arrestin 2 versus
G protein-mediated signaling.
[
[
[
2] Y. Olsen, G. L. Daumit, D. E. Ford, J Pain 2006, 7, 225.
3] B. L. Kieffer, Trends Pharmacol Sci 1999, 20, 19.
4] H. W. Matthes, R. Maldonado, F. Simonin, O. Valverde, S. Slowe, I.
Kitchen, K. Befort, A. Dierich, M. Le Meur, P. Dolle, E. Tzavara, J.
Hanoune, B. P. Roques, B. L. Kieffer, Nature 1996, 383, 819.
[
[
[
5] T. W. Vanderah, Clin J Pain 2010, 26 Suppl 10, S10.
6] K. Abdallah, L. Gendron, Handb Exp Pharmacol 2017.
7] A. A. Pradhan, K. Befort, C. Nozaki, C. Gaveriaux-Ruff, B. L. Kieffer,
Trends Pharmacol Sci 2011, 32, 581.
[
8] R. J. Bodnar, Peptides 2017, 88, 126.
4
|
CONCLUSION
[9] A. Janecka, J. Fichna, T. Janecki, Curr Top Med Chem 2004, 4, 1.
[
10] J. Hughes, T. W. Smith, H. W. Kosterlitz, L. A. Fothergill, B. A. Mor-
gan, H. R. Morris, Nature 1975, 258, 577.
The current study on the optimal shape of the fifth residue of enkepha-
lins revealed that its side chain has to be long enough to be able to
develop positive interactions with the receptors. However, even when
this condition is met, we observed that selectivity toward MOP or
DOP remains unachieved with the chosen non-natural amino acids.
Interestingly, the analogs displaying a high affinity for DOP also
potently inhibited the cAMP production. The replacement of the leu-
cine residue by aza-b-homoleucine or cycloleucine in position 5
resulted in peptidomimetics with significantly lower ability to recruit
b-arrestin 2 toward DOP suggesting that this position has the potential
to produce G protein biased signaling at this receptor. By contrast,
derivatives containing 4,5-dehydroleucine or D-allo-isoleucine were
more efficient at recruiting b-arrestin 2 than inhibiting cAMP produc-
tion following MOP activation.
[11] L. B. Hersh, Mol Cell Biochem 1982, 47, 35.
[12] K. Rochon, A. Proteau-Gagne, P. Bourassa, J. F. Nadon, J. Cote, V.
Bournival, F. Gobeil, Jr., B. Guerin, Y. L. Dory, L. Gendron, ACS
Chem Neurosci 2013, 4, 1204.
[
13] R. Schwyzer, Ann N Y Acad Sci 1977, 297, 3.
[
14] J. F. Nadon, K. Rochon, S. Grastilleur, G. Langlois, T. T. Dao, V.
Blais, B. Guerin, L. Gendron, Y. L. Dory, ACS Chem Neurosci 2017,
8
, 40.
[
15] A. Proteau-Gagne, V. Bournival, K. Rochon, Y. Dory, L. Gendron,
ACS Chem Neurosci 2010, 1, 757.
[
16] A. Proteau-Gagn eꢀ , K. Rochon, M. Roy, P.-J. Albert, B. Gu eꢀ rin, L.
Gendron, Y. L. Dory, Bioorg Med Chem Lett 2013, 23, 5267.
[17] T. Kenakin, C. Watson, V. Muniz-Medina, A. Christopoulos, S. Nov-
ick, ACS Chem Neurosci 2012, 3, 193.
[
18] E. T. van der Westhuizen, B. Breton, A. Christopoulos, M. Bouvier,
Mol Pharmacol 2014, 85, 492.
ACKNOWLEDGMENTS
[
[
19] A. El-Faham, F. Albericio, J Pept Sci 2010, 16, 6.
The authors thank Dr. Richard Howells (Rutgers University-New Jer-
sey Medical School, Newark, NJ, USA) for providing the HEK293
cells expressing the mouse Flag-DOP and Dr. Mark von Zastrow
~
3
20] A. El-Faham, R. SA S. Funosas, R. Prohens, F. Albericio, Chemistry
009, 15, 9404.
2
[
21] C. R. Beddell, R. B. Clark, G. W. Hardy, L. A. Lowe, F. B. Uba-
(
University of California, San Francisco, CA, USA) for HEK293 cells
tuba, J. R. Vane, S. Wilkinson, Proc R Soc Lond B Biol Sci 1977,
1
98, 249.
expressing the human Flag-MOP.
[
[
22] L. Gendron, C. M. Cahill, M. von Zastrow, P. W. Schiller, G. Pineyro,
Pharmacol Rev 2016, 68, 631.
AUTHOR CONTRIBUTIONS
23] J. T. Williams, S. L. Ingram, G. Henderson, C. Chavkin, M. von Zas-
trow, S. Schulz, T. Koch, C. J. Evans, M. J. Christie, Pharmacol Rev
2013, 65, 223.
DBN and VB have contributed equally to this work. SB synthesized
the non-natural amino acid in 8 while ICS and JFN synthesized the

1
0 of 10
|
BELLA NDONG ET AL.
[
[
[
[
24] J. W. Black, P. Leff, N. P. Shankley, J. Wood, Br J Pharmacol 1985,
4, 561.
SUPPORTING INFORMATION
8
Additional Supporting Information may be found online in the sup-
porting information tab for this article.
25] L. M. Bohn, R. R. Gainetdinov, F. T. Lin, R. J. Lefkowitz, M. G.
Caron, Nature 2000, 408, 720.
26] L. M. Bohn, R. J. Lefkowitz, R. R. Gainetdinov, K. Peppel, M. G.
Caron, F. T. Lin, Science 1999, 286, 2495.
How to cite this article: Bella Ndong D, Blais V, Holleran BJ,
et al. Exploration of the fifth position of leu-enkephalin and its
role in binding and activating delta (DOP) and mu (MOP) opioid
receptors. Peptide Science. 2018;e24070. https://doi.org/10.
27] S. M. DeWire, D. S. Yamashita, D. H. Rominger, G. Liu, C. L.
Cowan, T. M. Graczyk, X. T. Chen, P. M. Pitis, D. Gotchev, C.
Yuan, M. Koblish, M. W. Lark, J. D. Violin, J Pharmacol Exp Ther
2
013, 344, 708.
28] K. M. Raehal, J. K. Walker, L. M. Bohn, J Pharmacol Exp Ther 2005,
14, 1195.
1002/bip.24070
[
3