A novel cyclic enkephalin analogue with potent opioid antagonist activity

Article abstract of DOI:10.1016/j.bmcl.2004.06.077

The synthesis and in vitro activity profile of a potent opioid peptide antagonist lacking an N-terminal amino group are described. 2′,6′- Dimethyl substitution of the Tyr¹ residue in opioid agonist peptides and deletion of the N-terminal amino

Full text of DOI:10.1016/j.bmcl.2004.06.077

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4731–4733
A novel cyclic enkephalin analogue with potent opioid
antagonist activity
Grazyna Weltrowska,a Yixin Lu,b Carole Lemieux,a Nga N. Chunga
and Peter W. Schillera,
*
aLaboratory of Chemical Biology and Peptide Research, Clinical Research Institute of Montreal,
110 Pine Avenue West, Montreal, Quebec, Canada H2W 1R7
bDepartment of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore
Received 7 May 2004; accepted 23 June 2004
Available online 20 July 2004
Abstract—20,60-Dimethyl substitution of the Tyr1 residue in opioid agonist peptides and deletion of the N-terminal amino group, as
achieved by replacement of Tyr1 with 3-(2,6-dimethyl-4-hydroxyphenyl)propanoic acid (Dhp), have been shown to produce opioid
antagonists. To examine the effect of b-methylation of Dhp1 in opioid peptides on the activity profile, stereoselective syntheses of
(3S)- and (3R)-3-methyl-3-(2,6-dimethyl-4-hydroxyphenyl)propanoic acid [(3S)- and (3R)-Mdp] were carried out. In comparison
with the cyclic parent antagonist peptide Dhp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2, the methylated analogue (3S)-Mdp-c[D-Cys-
Gly-Phe(pNO2)-D-Cys]NH2 showed higher l, d and j antagonist potencies in functional assays and higher binding affinities
for l, d and j opioid receptors (Kil =2.03nM; Kid =2.34nM; Kij =49.5nM), whereas the corresponding (3R)-Mdp1-analogue was
less potent by 1–2 orders of magnitude.
Ó 2004 Elsevier Ltd. All rights reserved.
20,60-Dimethyl substitution of the Tyr1 residue of opioid
agonist peptides and deletion of the positively charged
N-terminal amino group have recently been shown to
represent a general structural modification to convert
opioid peptide agonists into antagonists.1–3 This conver-
sion requires the synthesis of opioid peptide analogues
containing 3-(2,6-dimethyl-4-hydroxyphenyl)propanoic
acid (Dhp) in place of Tyr1.1 The cyclic enkephalin ana-
logue Dhp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2 (8)
turned out to be a quite potent l and d opioid receptor
antagonist and a somewhat less potent j antagonist2
(Tables 1 and 2). In this study, we examine the effect
of b-methylation of Dhp1 in Dhp-c[D-Cys-Gly-Phe(p-
NO2)-D-Cys]NH2 (8) on the in vitro opioid activity pro-
file. This requires replacement of Dhp in the cyclic
peptide with (3S)- or (3R)-3-methyl-3-(2,6-dimethyl-4-
hydroxyphenyl)propanoic acid [(3S)- or (3R)-Mdp (6a
or 6b)]. The presence of a b-methyl group might either
enhance or decrease opioid receptor binding affinity
and it will be of interest to determine the stereochemical
requirements for receptor binding. Therefore, both (3S)-
and (3R)-Mdp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2 (7a
and 7b) were prepared. These compounds will also allow
for an interesting comparison with b-methylated
Tyr(20,60-Me2)1-containing opioid peptide agonists4 in
terms of potency changes and stereochemical
requirements.
The stereoselective synthesis of (3S)-Mdp (6a) is out-
lined in Scheme 1, which is based on a published ap-
proach to the synthesis of chiral b-branched carboxylic
acids.5 Basic hydrolysis of methyl-3-(4-tert-butoxycar-
bonyloxy-2,6-dimethylphenyl)propenoate 16 afforded
acid 2. Incorporation of the chiral auxiliary (S)-(+)-4-
phenyl-2-oxazolidinone5 was carried out in the standard
manner7 to yield 3a.8 Asymmetric Michael addition was
performed using the organocuprate prepared from
methylmagnesium bromide in THF/(CH3)2S to furnish
4a8 with a diastereomeric excess of 48%. It should be
noted that the diastereoselectivity of the Michael addi-
tion to 3a was less than that reported in the literature,5
even though the reaction was carried out in an analo-
gous manner. The diastereomers were easily separated
by flash column chromatography (silica gel) to yield chi-
rally pure 4a in 55% yield. Removal of the chiral auxil-
iary9 and Boc group then gave 6a.8 Compound 6b8 was
*
Corresponding author. Tel.: +1-514-987-5576; fax: +1-514-987-
5513; e-mail: schillp@ircm.qc.ca
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.077

4732
G. Weltrowska et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4731–4733
Table 1. Receptor binding affinities of opioid peptide analogues
Compound
Kil [nM]a
Kid [nM]a
Kij [nM]a
7a (3S)-Mdp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
7b (3R)-Mdp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
8 Dhp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
9 H-Dmt-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
2.10 0.21
94.4 13.6
4.79 0.39
0.247 0.026
2.03 0.09
497 112
11.6 1.1
49.5 2.1
6970 820
299 57
0.704 0.042
3.77 0.72
a Mean of 3–4 determinations SEM.
Table 2. GPI and MVD assay of opioid peptide analogues
Compound
GPI
MVD
Ke [nM]a,b
l
Ke [nM]a,c
j
Ke [nM]a,d
d
7a (3S)-Mdp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
7b (3R)-Mdp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
8 Dhp-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
1.40 0.25
845 135
3.68 0.45
5.81 1.20
1630 290
22.6 3.0
55.0 5.4
3280 430
63.3 10.5
e
9 H-Dmt-c[D-Cys-Gly-Phe(pNO2)-D-Cys]NH2
IC50 =0.541 0.125nM
IC50 =0.182 0.030nM
a Mean of 3–5 determinations SEM.
b Determined with TAPP (H-Tyr-D-Ala-Phe-Phe-NH2) as agonist.
c Determined with U50,488 as agonist.
d Determined with DPDPE as agonist.
e Agonist.
OBoc
OBoc
OBoc
i
ii
COXc
3a
iii
COOH
COOMe
1
2
OH
OBoc
OBoc
iv
v
COOH
COXc
COOH
6a
5a
4a
Scheme 1. Reagents and conditions: (i) 1N aq NaOH/THF, 81%; (ii) Et3N, PvCl, THF, À78 to 0°C, then treated with n-BuLi, Xc, THF, À78 to 0°C,
70%; (iii) CuBr–Me2S complex, THF, CH3MgBr, ether, 0°C, 55%; (iv) LiOH, H2O2, THF/H2O, 90%; (v) 95% TFA/H2O, 0°C, 98%. Xc=(S)-(+)-4-
phenyl-2-oxazolidinone.
prepared in an analogous manner, using (R)-(À)-4-phen-
yl-2-oxazolidinone as chiral auxiliary and following the
same sequence of reactions.
tive reversed-phase HPLC. (3S)-Mdp or (3R)-Mdp were
attached to the N-terminal amino group of H-c[D-Cys-
Gly-Phe(pNO2)-D-Cys]NH2 using 2-(1H-benzotriazole-
1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
(HBTU) as coupling agent. The final peptide products
7a and 7b were purified by preparative reversed-phase
HPLC and their purity and structural identity were
established by TLC, analytical HPLC and FAB-MS.10
The target peptides (3S)- and (3R)-Mdp-c[D-Cys-Gly-
Phe(pNO2)-D-Cys]NH2 (7a,b) were prepared by a com-
bination of manual solid-phase and solution techniques.
The linear precursor peptide of H-c[D-Cys-Gly-Phe(p-
NO2)-D-Cys]NH2 was assembled on a p-methylbenz-
hydrylamine resin using Boc protection of the a-amino
function and acetamidomethyl (Acm) protection of the
Cys side chain. After cleavage from the resin by HF/an-
isole treatment, disulfide bond formation was carried
out in MeOH/H2O using iodine as oxidation agent
and the resulting cyclic peptide was purified by prepara-
In comparison with the Dhp-c[D-Cys-Gly-Phe(pNO2)-
D-Cys]NH2 parent peptide (8), the (3S)-Mdp1-analogue
(7a) showed significantly higher l-, d- and j-receptor
binding affinities (Table 1). This result is in contrast to
the observation made with the (2S,3R)-Tmt1-analogue
of the d agonist DPDPE (H-(2S,3R)-Tmt-c[D-Pen-Gly-

G. Weltrowska et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4731–4733
4733
Phe(pNO2)-D-Pen]OH; Tmt=20,60-dimethyl-b-methyl-
tyrosine),4 which has the same stereochemistry at the
methylated b-carbon of the N-terminal residue as 7a.
[(2S,3R)-Tmt1]DPDPE showed significantly lower l
and d receptor binding affinities and lower d agonist
potency in the mouse vas deferens (MVD) assay as
compared to the parent peptide H-Dmt-c[D-Pen-Gly-
Phe(pNO2)-D-Pen]OH; Dmt=20,60-dimethyltyrosine).11
Obviously, the b-methyl group of (3S)-Mdp in 7a is able
to effectively interact with a lipophilic binding site at all
three opioid receptors to strengthen binding, whereas
the b-methyl group of (2S,3R)-Tmt in [2S,3R)-Tmt1]
DPDPE decreases binding affinity, most likely due to
some steric interference. The ability of the b-methyl
group of (3S)-Mdp to enhance receptor binding affinity
may be due to the greater conformational flexibility of
the (3S)-Mdp residue as compared to the (2S,3R)-Tmt
residue. It is also possible that the conformational
requirements of the active and inactive receptor confor-
mations differ from one another with regard to the inter-
action of the N-terminal residue of the agonist peptide
and the antagonist peptide.
C. In Peptides: The Wave of the Future (Proceedings of the
Second International Peptide Symposium/17th American
Peptide Symposium); Lebl, M., Houghten, R. A., Eds.;
American Peptide Society: San Diego, CA, 2001, pp
676–678.
3. Lu, Y.; Nguyen, T. M.-D.; Weltrowska, G.; Berezowska,
I.; Lemieux, C.; Chung, N. N.; Schiller, P. W. J. Med.
Chem. 2001, 44, 3048.
4. Qian, X.; Shenderovich, M. D.; Kover, K. E.; Davis, P.;
Horvath, R.; Zawelska, T.; Yamamura, H. I.; Porreca, F.;
Hruby, V. J. J. Am. Chem. Soc. 1996, 118,
7280.
´
5. Nicolas, E.; Russell, K. C.; Hruby, V. J. J. Org. Chem.
1993, 58, 766.
6. Lu, Y.; Weltrowska, G.; Lemieux, C.; Chung, N. N.;
Schiller, P. W. Bioorg. Med. Chem. Lett. 2001, 11, 323.
7. Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem.
Soc. 1982, 104, 1737.
8. All new compounds were fully characterized by optical
rotation measurements, 1H and 13C NMR spectra and
HRMS. Data of selected compounds:
20
3a. ½aꢀD +30.5 (c 0.98, CHCl3); 1H NMR (400MHz,
CDCl3) d 7.92–7.88 (d, 1H, J=16.1Hz), 7.56–7.52 (d, 1H,
J=16.1Hz), 7.44–7.36 (m, 5H), 6.89 (s, 2H), 5.56–5.53
(dd, 1H, J=3.84, 8.8Hz), 4.76–4.72 (m, 1H), 4.34–4.31
(dd, 1H, J=3.84, 8.8Hz), 2.37 (s, 6H), 1.59 (s, 9H); 13C
NMR (100.6MHz, CDCl3) d 164.9, 154.0, 152.0, 150.8,
143.9, 139.4, 139.3, 131.7, 129.4, 129.0, 126.3, 122.6, 121.3,
83.8, 70.2, 58.1, 27.9, 21.7; HRMS (FAB) m/e calcd for
C25H27NO6 [M+K]+ 476.1475, found: 476.1477.
In agreement with the receptor binding data, the
(3S)-Mdp1-analogue 7a also showed higher l-, d- and
j-antagonist potencies than the Dhp1-parent 8 in the
functional guinea pig ileum (GPI) and MVD assays
(Table 2). In comparison with the (3S)-Mdp1-analogue
7a, the diastereomeric (3R)-Mdp1-analogue 7b displayed
drastically lower binding affinities and antagonist poten-
cies at all three receptors. This result is in agreement
with observations that (2S,3R)-Tmt1-analogues of opi-
oid agonist peptides generally showed higher opioid
receptor binding affinities and higher agonist potencies
in functional assays than their corresponding (2S,3S)-
Tmt1-analogues.4 It thus appears that the stereochemi-
cal requirements at the b-carbon of the 1-position side
chain of b-methylated Dhp1-antagonist peptides and b-
methylated Dmt1-agonist peptides for opioid receptor
binding are the same. Furthermore, these results indi-
cate that the overall mode of opioid receptor binding
of these agonists and antagonists is similar but not
identical.
20
4a. ½aꢀD +80.5 (c 1.03, CHCl3); 1H NMR (400MHz,
CDCl3) d 7.40–7.25 (m, 5H), 6.76 (s, 2H), 5.37–5.34 (dd,
1H, J=3.48, 8.8Hz), 4.56–4.61 (m, 1H), 4.26–4.22 (dd,
1H, J=3.48, 8.8Hz), 3.94–3.84 (m, 1H), 3.50–3.44 (dd,
1H, J=6.7, 17.1Hz), 3.38–3.32 (dd, 1H, J=7.8, 17.1Hz),
2.45–2.34 (d, 6H, J=44.6Hz), 1.55 (s, 9H), 1.25–1.23 (d,
3H, J=7.3Hz); 13C NMR (100.6MHz, CDCl3) d 171.9,
152.5, 148.7, 139.3, 129.4, 128.8, 126.0, 122.7, 120.9, 83.5,
70.1, 57.9, 40.6, 30.1, 27.9, 21.8, 19.0; HRMS (FAB) m/e,
calcd for C26H31NO6 [M+K]+ 492.1788, found: 492.1790.
20
6a. ½aꢀD +35.3 (c 1, MeOH); 1H NMR (400MHz,
CD3COCD3) d 10.46 (s, 1H), 7.83 (s, 1H), 6.47 (s, 2H),
3.81–3.72 (m, 1H), 2.7–2.6 (m, 2H), 2.33 (s, 6H), 1.32–1.30
(d, 3H, J=7.32Hz); 13C NMR (100.6MHz, CD3COCD3)
d 206.1, 173.9, 155.0, 137.5, 132.7, 117.0, 115.3, 39.8,
30.94, 30.90, 21.0, 18.9; HRMS (EI) m/e calcd for
C12H16O3 [M+] 208.1099, found: 208.1104.
20
6b. ½aꢀD À35.2 (c 1, MeOH); 1H NMR (400MHz,
CD3COCD3) d 10.46 (s, 1H), 7.84 (s, 1H), 6.46 (s, 2H),
3.80–3.72 (m, 1H), 2.75–2.64 (m, 2H), 2.33 (s, 6H), 1.31–
1.29 (d, 3H, J=7.1Hz); 13C NMR (100.6MHz,
CD3COCD3) d 206.1, 173.9, 155.0, 137.5, 132.7, 117.0,
115.3, 39.8, 30.9, 30.8, 21.1, 18.9; HRMS (EI) m/e calcd
for C12H16O3 [M]+ 208.1099, found: 208.1110.
9. Evans, D. A.; Britton, D. C.; Ellman, J. A. Tetrahedron
Lett. 1987, 28, 6141.
Acknowledgements
This work was supported by operating grants from the
U.S. National Institute on Drug Abuse (DA-04443)
and the Canadian Institutes of Health Research
(MOP-5655).
10. Analytical data of peptides 7a and 7b:
References and notes
7a. TLC Rf 0.90 (n-BuOH/AcOH/H2O, 4:1:1), Rf 0.22
(CHCl3/MeOH/AcOH, 85:10:5); FAB-MS [M+H]+ 661.
7b. TLC Rf 0.92 (n-BuOH/AcOH/H2O, 4:1:1), Rf 0.22
(CHCl3/MeOH/AcOH, 85:10:5); FAB-MS [M+H]+
661.
1. Schiller, P. W.; Berezowska, I.; Nguyen, T. M.-D.;
Schmidt, R.; Lemieux, C.; Chung, N. N.; Falcone-Hind-
ley, M. L.; Yao, W.; Liu, J.; Iwama, S.; Smith, A. B., III;
Hirschmann, R. J. Med. Chem. 2000, 43, 551.
2. Schiller, P. W.; Lu, Y.; Weltrowska, G.; Berezowska, I.;
Wilkes, B. C.; Nguyen, T. M.-D.; Chung, N. N.; Lemieux,
11. Hansen, D. W., Jr.; Stapelfeld, A.; Savage, M. A.;
Reichman, M.; Hammond, D. L.; Haaseth, R. C.;
Mosberg, H. I. J. Med. Chem. 1992, 35, 684.