Synthesis and opioid activity of [D-Pro10]Dynorphin A-(1-11) analogues with N-terminal alkyl substitution

Article abstract of DOI:10.1021/jm960747t

Several N-terminal di- and monoalkylated derivatives of [D- Pro¹⁰]dynorphin A-(1-11) were synthesized in order to explore the structure-activity relationships for antagonist vs agonist activity at κ- opioid receptors. N,N-Dialkylated and N-monoalkylated (alkyl = allyl, benzyl, and cyclopropylmethyl (CPM)) tyrosine derivatives were prepared from tyrosine tert-butyl ester and the corresponding alkyl halides. [D-Pro¹⁰]Dyn A-(2- 11) was prepared by solid phase synthesis using Fmoc-protected amino acids, and the tyrosine derivatives were coupled to the peptide with BOP ((benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophosphate). Both the degree of substitution and the identity of the alkyl group affected κ-receptor affinity, selectivity, and efficacy. All of the N-monoalkylated derivatives exhibited much higher affinity (K(i) < 0.05 nM) for κ receptors in the guinea pig cerebellum and greatly enhanced κ-receptor selectivity (K(i) ratio (κ/μ) > 200) compared to the N,N-dialkyl [D-Pro₁₀]Dyn A-(1- 11) analogues, although one disubstituted analogue, N,N-diCPM[D-Pro₁₀]Dyn A-(1-11), retained high affinity (K(i) = 0.19 nM) for κ receptors. Thus the introduction of the second alkyl group at the N-terminus lowered κ-receptor affinity and selectivity. The N-allyl and N-CPM analogues were moderately potent agonists in the guinea pig ileum (GPI) assay, while the N-benzyl derivative was a weak agonist in this assay. In vivo in the phenylqninone abdominal stretching assay the N-CPM analogue exhibited potent antinociceptive activity (ED₅₀ = 1.1 μg/mouse), while N-allyl[D- Pro¹⁰]Dyn A-(l-11) exhibited weak antinociceptive activity (ED₅₀ = 27 μg/mouse). For the N,N-dialkyl derivatives the identity of the N-terminal alkyl group affected the efficacy observed in the smooth muscle assays. The N,N-diCPM analogue exhibited negligible agonist activity, and N,N-diallyl[D- Pro¹⁰]Dyn A-(l-11) showed weak antagonist activity against Dyn A-(1- 13)NH₂ in the GPI. In contrast, the N,N-dibenzyl compound showed appreciable opioid agonist activity in this assay. In vivo the N,N-diallyl analogue exhibited weak antinociceptive activity (ED₅₀ = 26 μg/mouse in the phenylquinone abdominal stretching assay). The N-monoalkylated peptides are among the most κ-selective opioid peptides reported to date, showing comparable or greater selectivity and higher affinity than the κ-selective nonpeptide agonists U-50,488 and U-69,593. The N,N-diCPM and N,N-diallyl peptides are lead compounds in the development of peptide-based κ-receptor antagonists.

Full text of DOI:10.1021/jm960747t

J . Med. Chem. 1997, 40, 2733-2739
2733
Syn th esis a n d Op ioid Activity of [D-P r o10]Dyn or p h in A-(1-11) An a logu es w ith
N-Ter m in a l Alk yl Su bstitu tion †
Heekyung Choi,‡ Thomas F. Murray,§ Gary E. DeLander,§ William K. Schmidt,|, and J ane V. Aldrich*,‡
College of Pharmacy, Oregon State University, Corvallis, Oregon 97331, Department of Pharmaceutical Sciences, School of
Pharmacy, University of Maryland, 20 North Pine Street, Baltimore, Maryland 21201-1180, and Dupont Merck
Pharmaceutical Company, Experimental Station, Wilmington, Delaware 19880
Received October 28, 1996X
Several N-terminal di- and monoalkylated derivatives of [D-Pro10]dynorphin A-(1-11) were
synthesized in order to explore the structure-activity relationships for antagonist vs agonist
activity at κ-opioid receptors. N,N-Dialkylated and N-monoalkylated (alkyl ) allyl, benzyl,
and cyclopropylmethyl (CPM)) tyrosine derivatives were prepared from tyrosine tert-butyl ester
and the corresponding alkyl halides. [D-Pro10]Dyn A-(2-11) was prepared by solid phase
synthesis using Fmoc-protected amino acids, and the tyrosine derivatives were coupled to the
peptide with BOP ((benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophos-
phate). Both the degree of substitution and the identity of the alkyl group affected κ-receptor
affinity, selectivity, and efficacy. All of the N-monoalkylated derivatives exhibited much higher
affinity (Ki < 0.05 nM) for κ receptors in the guinea pig cerebellum and greatly enhanced
κ-receptor selectivity (Ki ratio (κ/µ) > 200) compared to the N,N-dialkyl [D-Pro10]Dyn A-(1-11)
analogues, although one disubstituted analogue, N,N-diCPM[D-Pro10]Dyn A-(1-11), retained
high affinity (Ki ) 0.19 nM) for κ receptors. Thus the introduction of the second alkyl group
at the N-terminus lowered κ-receptor affinity and selectivity. The N-allyl and N-CPM analogues
were moderately potent agonists in the guinea pig ileum (GPI) assay, while the N-benzyl
derivative was a weak agonist in this assay. In vivo in the phenylquinone abdominal stretching
assay the N-CPM analogue exhibited potent antinociceptive activity (ED50 ) 1.1 µg/mouse),
while N-allyl[D-Pro10]Dyn A-(1-11) exhibited weak antinociceptive activity (ED50 ) 27
µg/mouse). For the N,N-dialkyl derivatives the identity of the N-terminal alkyl group affected
the efficacy observed in the smooth muscle assays. The N,N-diCPM analogue exhibited
negligible agonist activity, and N,N-diallyl[D-Pro10]Dyn A-(1-11) showed weak antagonist
activity against Dyn A-(1-13)NH2 in the GPI. In contrast, the N,N-dibenzyl compound showed
appreciable opioid agonist activity in this assay. In vivo the N,N-diallyl analogue exhibited
weak antinociceptive activity (ED50 ) 26 µg/mouse in the phenylquinone abdominal stretching
assay). The N-monoalkylated peptides are among the most κ-selective opioid peptides reported
to date, showing comparable or greater selectivity and higher affinity than the κ-selective non-
peptide agonists U-50,488 and U-69,593. The N,N-diCPM and N,N-diallyl peptides are lead
compounds in the development of peptide-based κ-receptor antagonists.
In tr od u ction
potent and more selective derivatives in order to elicit
the pharmacological mechanisms of κ receptors.2 Ex-
perimental evidence suggests that the N-terminal tet-
rapeptide fragment, referred to as the “message” se-
quence,3 is responsible for opioid activity and that the
N-terminal tyrosine residue is important for both opioid
activity and binding to opioid receptors.4 The C-
terminal extension (“address” sequence) directs the
peptide to κ-opioid receptors.3
Although dynorphin A (Dyn A) has been postulated
to be an endogenous opioid ligand for κ-opioid receptors,1
it has been difficult to elucidate the physiological roles
of Dyn A mediated by κ receptors because of the
existence of multiple opioid receptor types and the low
selectivity of Dyn A for κ receptors.2 Numerous struc-
ture-activity studies involving modification of Dyn
A-(1-13) have been devoted to the development of
N-Terminal alkylation of the tyrosine residue in Dyn
A has been examined in a search for selective antago-
nists, based on the fact that substitution of the N-methyl
group in morphine analogues by an N-allyl or N-
cyclopropylmethyl group can impart antagonist activity
to the classical alkaloid opiates (e.g. naloxone and
naltrexone). Previously this modification was success-
fully applied to [Leu5]enkephalin analogues. N,N-
Diallyl substitution of [Leu5]enkephalin yields a selec-
tive δ antagonist,5 while the introduction of a single allyl
group at the N-terminus of [D-Ala2,Met5]enkephalin
yields a mixed agonist/antagonist.6 Introduction of N,N-
diallyltyrosine at position 1 of [D-Pro10]Dyn A-(1-11)7
and Dyn A-(1-13)8 has been reported to impart antago-
nist activity to these peptides at κ and µ receptors, and
† Abbreviations used for amino acids follow the rules of the IUPAC-
IUB J oint Commission of Biochemical Nomenclature in Biochem. J .
1984, 219, 345-373. Additional abbreviations: Boc, tert-butyloxycar-
bonyl; BOP, (benzotriazol-1-yloxy)tris(dimethylamino)phosphonium
hexafluorophosphate; CHO, Chinese hamster ovary; CPM, cyclopro-
pylmethyl; DAMGO, Tyr-D-Ala-Gly-MePhe-NH(CH2)2OH; DIC, N,N′-
diisopropylcarbodiimide; DCM, dichloromethane; DIEA, N,N-diisopro-
pylethylamine; DMA, N,N-dimethylacetamide; DMF, N,N-dimethyl-
formamide; DPDPE, [D-Pen2,D-Pen5]enkephalin; Dyn A, dynorphin A;
FAB-MS, fast atom bombardment mass spectrometry; Fmoc, 9-fluo-
renylmethoxycarbonyl; GPI, guinea pig ileum; HOBt, 1-hydroxyben-
zotriazole; HPLC, high-performance liquid chromatography; MVD,
mouse vas deferens; NalBzoH, naloxone benzoylhydrazone; Pmc,
2,2,5,7,8-pentamethylchroman-6-sulfonyl; TFA, trifluoroacetic acid.
‡ University of Maryland.
§ Oregon State University.
| Dupont Merck Pharmaceutical Company.
Current address: NorthStar Research & Development Ltd., 1
Bridleshire Circle, Newark, DE 19711.
X Abstract published in Advance ACS Abstracts, J uly 15, 1997.
S0022-2623(96)00747-9 CCC: $14.00 © 1997 American Chemical Society

2734 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 17
Choi et al.
Ta ble 1. Characterization of N-Alkylated [D-Pro10]Dyn
at κ receptors, respectively. N,N-Diallylation of both
peptides, however, leads to decreases in binding affinity
and selectivity for κ-opioid receptors and results in
analogues which show only weak antagonist activity in
the guinea pig ileum (GPI). N,N-Diallyl[D-Pro10]Dyn
A-(1-11) is reported to be devoid of agonist activity,
whereas N,N-diallyl-Dyn A-(1-13) still possesses full
agonist activity in the GPI.7,8 Incorporation of D-Pro at
position 10 in Dyn A-(1-11) is reported to enhance
κ-receptor selectivity and potency compared to Dyn
A-(1-11).9 We therefore chose N,N-diallyl[D-Pro10]Dyn
A-(1-11) as our lead compound for further modification.
The major goals of our study were to explore whether
the second alkyl group in N,N-diallyl[D-Pro10]Dyn A-(1-
11) is required for antagonist activity and how mono-
vs disubstitution of the N-terminus affects opioid recep-
tor selectivity and opioid efficacy. Therefore, we syn-
thesized several N-monosubstituted and N,N-disubsti-
tuted [D-Pro10]Dyn A-(1-11) analogues, and evaluated
their affinity in radioligand binding assays and their
opioid activity in the GPI assay. In a preliminary
communication,10 we described the remarkable κ-recep-
tor selectivity of the N-monoalkylated derivatives. Here
we describe the synthesis of these analogues and
compare the effects of mono- vs disubstitution on opioid
receptor affinity and efficacy. Additional pharmacologi-
cal characterization of both the N-mono- and N,N-
dialkylated [D-Pro10]Dyn A-(1-11) derivatives is also
reported, including determination of κ-receptor subtype
affinities of selected peptides and evaluation of in vivo
antinociceptive activities in the mouse phenylquinone
abdominal stretching assay. The results of these assays
are described below.
A-(1-11) Derivatives by HPLC and Fast Atom Bombardment
Mass Spectrometry (FAB-MS)
HPLC tR
FAB-MS
(M + 1)
peptide
(% purity)a
N,N-diallyl[D-Pro10]Dyn A-(1-11)
19.7 (100)
1443
1543
1471
1403
1453
1417
N,N-dibenzyl[D-Pro10]Dyn A-(1-11) 25.6 (98.4)
N,N-diCPM[D-Pro10]Dyn A-(1-11)
N-allyl[D-Pro10]Dyn A-(1-11)
N-benzyl[D-Pro10]Dyn A-(1-11)
N-CPM[D-Pro10]Dyn A-(1-11)
20.7 (100)
18.4 (98.5)
20.1 (100)
18.9 (98.7)
a
Conditions: Vydac 214 TP C-4 column, 0-60% solvent B (0.1%
TFA in AcCN) over 40 min at 1.5 mL/min, detection at 214 nm.
which were more difficult to separate from the desired
products than the starting material. The N,N-dialky-
lated tyrosine derivatives were prepared in good yields
using excess alkyl halide and an organic base.
[D-Pro10]Dyn A-(2-11) was assembled on a (hydroxy-
methyl)phenoxyacetic acid resin using Fmoc-protected
amino acids with N,N′-diisopropylcarbodiimide (DIC)
and 1-hydroxybenzotriazole (HOBt) as the coupling
agents. Side chain protecting groups used were the
2,2,5,7,8-pentamethylchroman-6-sulfonyl (Pmc) group
for Arg and the Boc group for Lys. The tyrosine
derivatives were incorporated into the resin-bound
protected Dyn A-(2-11) using (benzotriazol-1-yloxy)tris-
(dimethylamino)phosphonium hexafluorophosphate
(BOP),12 HOBt, and N-methylmorpholine in N,N-dim-
ethylacetamide (DMA). The BOP reagent in the pres-
ence of HOBt has been reported to result in more rapid
and efficient coupling as well as a reduction in racem-
ization when compared with DIC plus HOBt.13 The
peptides were cleaved from the support with concen-
trated trifluoroacetic acid (TFA) containing 5% water
as a scavenger for the released carbonium ions. After
purification of the crude peptides using reverse phase
HPLC, the molecular weights were confirmed by FAB-
MS and the purities evaluated by analytical HPLC
(Table 1). The amino acid analyses gave the expected
values for each amino acid (Table 2).
R,R′Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile-Arg-D-Pro-LysOH
R ) allyl, cyclopropylmethyl (CPM), or benzyl; R′ ) H
R ) R′ ) allyl, CPM, or benzyl
N-alkylated [D-Pro10]Dyn A-(1-11) derivatives
Ch em istr y
P h a r m a cologica l Resu lts
The N-alkylated peptides were prepared by Fmoc (9-
fluorenylmethoxycarbonyl) solid phase synthesis in
conjunction with incorporation of an N-alkylated ty-
rosine. Several methods are available for preparation
of optically pure N-monoalkyl-substituted amino acid
derivatives, including alkylation of N-protected amino
acids. This method was developed by Cheung and
Benoiton11 to make various N-methyl amino acid de-
rivatives using NR-Boc (tert-butyloxycarbonyl)samino
acids, sodium hydride as a base, and methyl iodide. The
reaction with the bulkier alkyl halides of interest for
preparation of the N-alkyl Dyn A derivatives, however,
was extremely slow and gave very poor yields. We
therefore examined conditions for alkylation of the free
amine with alkyl halides. The N-monosubstituted and
a significant amount of the N,N-disubstituted deriva-
tives were simultaneously produced when tyrosine tert-
butyl ester was treated with only an equimolar amount
of alkyl bromide, even in the absence of a base. Finally,
the reaction conditions were optimized to obtain a better
yield of N-monosubstituted derivatives by utilizing less
reactive alkyl halides (i.e. alkyl chlorides) and an
organic base. The reaction was terminated before
complete consumption of the starting material in order
to minimize generation of N,N-disubstituted derivatives,
The N-alkylated Dyn A analogues were examined in
several assays to determine their affinity for opioid
receptors and their opioid activity. All of the peptides
were evaluated for κ-, µ-, and δ-receptor affinity in
radioligand binding assays, and selected analogues were
also examined in binding assays for κ-opioid receptor
subtypes. In order to identify compounds to examine
further for possible antagonist activity, the effect of
sodium and Gpp(NH)p on the affinity of peptides for κ
receptors (sodium shift assay) was investigated. Opioid
activity was examined in the GPI assay, and selected
compounds were also evaluated in the mouse vas
deferens (MVD) preparation. The compounds were
tested in vivo for antinociceptive activity in the phe-
nylquinone abdominal stretching assay.
The synthesized compounds were evaluated in radio-
ligand binding assays in the guinea pig cerebellum, in
which at least 80% of the opioid receptors are κ-binding
sites,14 using [3H]bremazocine as the radioligand, and
in rat brain, which contains significant populations of
µ and δ receptors,15 using [3H]DAMGO ([D-Ala2,-
NMePhe4,glyol]enkephalin) and [3H]DPDPE ([D-Pen2,D-
Pen5]enkephalin), respectively, as the radioligands. As

[D-Pro10]Dynorphin A-(1-11) Analogues
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 17 2735
Ta ble 2. Amino Acid Analysis of N-Alkylated [D-Pro10]Dyn A-(1-11) Derivativesa
peptide
Gly (2)
Phe (1)
Leu (1)
Arg (3)
Ile (1)
Pro (1)
Lys (1)
N,N-diallyl[D-Pro10]Dyn A-(1-11)
N,N-dibenzyl[D-Pro10]Dyn A-(1-11)
N,N-diCPM[D-Pro10]Dyn A-(1-11)
N-allyl[D-Pro10]Dyn A-(1-11)
1.73
1.88
1.92
1.92
1.87
1.95
1.05
1.05
1.04
1.00
1.00
1.03
1.05
1.03
1.04
1.08
1.00
1.00
3.17
3.12
3.10
3.05
3.09
3.07
0.96
0.95
0.97
1.02
0.91
0.96
1.04
0.96
0.90
0.94
1.13
0.98
1.04
1.02
1.03
0.98
1.00
1.01
N-benzyl[D-Pro10]Dyn A-(1-11)
N-CPM[D-Pro10]Dyn A-(1-11)
a
The frequency of the amino acid is in parentheses.
Ta ble 3. Opioid Receptor Binding Affinity of N-Alkylated [D-Pro10]Dyn A-(1-11) Analogues
Ki (nM)a
peptide
[3H]bremazocine (κ)
[3H]DAMGO (µ)
[3H]DPDPE (δ)
Ki ratiob (κ/µ/δ)
[D-Pro10]Dyn A-(1-11)
0.030 ( 0.001
3.60 ( 0.50
0.24 ( 0.002
31.7 ( 3.8
128 ( 1
3.90 ( 0.10
10.9 ( 0.5
31.1 ( 0.2
9.6 ( 0.6
2.1 ( 0.4
149 ( 58
1/8/70
1/8.8/41
1/35/757
1/21/880
1/222/9160
1/1070/6080
1/480/27900
N,N-diallyl[D-Pro10]Dyn A-(1-11)
N,N-dibenzyl[D-Pro10]Dyn A-(1-11)
N,N-diCPM[D-Pro10]Dyn A-(1-11)
N-allyl[D-Pro10]Dyn A-(1-11)
N-benzyl[D-Pro10]Dyn A-(1-11)
N-CPM[D-Pro10]Dyn A-(1-11)
3.66 ( 0.10
2770 ( 130
166 ( 14
0.19 ( 0.002
0.049 ( 0.001
0.029 ( 0.001
0.020 ( 0.001
499 ( 15
176 ( 17
558 ( 8.9
a
Affinity for κ receptors was determined by measuring the inhibition of [3H]bremazocine binding to guinea pig cerebellar membranes.
Affinities for µ and δ receptors were measured in rat forbrain using [3H]DAMGO and [3H]DPDPE, respectively. See the Experimental
Section for details of the binding assays. Ratio of Ki values where the lowest Ki is used in the denominator.
b
Ta ble 4. Binding Affinity of [D-Pro10]Dyn A-(1-11) Analogues at κ Subtypesa
Ki (nM)
b
peptide
[3H]U69,593 (κ1)
[3H]bremazocine (κ2, IC50
)
[3H]NalBzoH (κ3)
Ki ratio (κ1/κ2/κ3)
Dyn A-(1-11)
1.70
21.3
2.61
3.24
13.3
4970
388
23.1
113
1/7.8/14
1/233/5.3
1/150/17
1/132/18
N,N-diallyl[D-Pro10]Dyn A-(1-11)
N,N-diCPM[D-Pro10]Dyn A-(1-11)
N-benzyl[D-Pro10]Dyn A-(1-11)
44.4
427
58.1c
a
b
See Experimental Section for details of the assays. For the κ2 receptors, Ki values are not reported because low Hill coefficients
suggest that there are more than one [3H]bremazocine binding site. c The Hill coefficient ()2.7) for this compound is quite different from
that observed for other compounds (∼1).
shown in Table 3, the parent peptide [D-Pro10]Dyn A-(1-
11) showed affinity for κ receptors similar to that
reported by Gairin et al.9 Compared to their results,
however, the affinities of this peptide for µ and δ
receptors were increased about 10 and 3 times, respec-
tively, resulting in only modest selectivity for κ receptors
(Ki ratio (κ/µ/δ) ) 1/8/70). The introduction of an N,N-
dialkylated tyrosine at position 1 caused a significant
decrease in affinity for κ receptors. N,N-Bis(cyclopro-
pylmethyl) substitution affected affinity for κ receptors
substantially less than did N,N-diallyl or N,N-dibenzyl
substitution. N-Monoalkylation of [D-Pro10]Dyn A-(1-
11), however, had little influence on affinity for κ
receptors. Both N-terminal mono- and dialkylation
considerably decreased affinity for µ receptors (16-530-
fold) and δ receptors (70-1320-fold).
N,N-Diallyl-, N,N-diCPM-, and N-benzyl[D-Pro10]Dyn
A-(1-11) were examined in radioligand binding assays
for κ-opioid receptor subtypes in guinea pig brain
membranes (Table 4). Dyn A-(1-11) showed the high-
est affinity for the κ1 receptors and the lowest affinity
for the κ3 subtype. All of the synthetic derivatives
preferentially interacted with κ1 receptors; the affinities
of the N,N-diCPM and N-benzyl derivatives were simi-
lar to that of Dyn A-(1-11) for this receptor subtype.
N-Alkylation decreased affinity for κ2 receptors by 30-
370-fold so that these analogues had the lowest affinity
for this subtype. Except for the N,N-diallyl analogue,
N-terminal alkylation decreased κ3-receptor affinity by
<2.5-fold. The N,N-diallyl derivative showed much
lower affinity for all three receptor subtypes than did
the other analogues tested.
The selectivity of the N-alkylated peptides for κ
receptors depended on both the identity of the N-
substituent and on the extent of alkylation (N-mono-
vs N,N-dialkylation). N,N-Diallyl[D-Pro10]Dyn A-(1-
11), a putative antagonist for κ receptors, showed
κ-receptor selectivity similar to that found by Gairin et
al. (Ki ratio (κ/µ/δ) ) 1/6.5/17),7 while other N,N-
dialkylated analogues showed slightly improved selec-
tivity for κ vs µ receptors and significantly higher
selectivity for κ vs δ receptors. As described in our
initial report,10 N-monoalkylation of [D-Pro10]Dyn A-(1-
11) resulted in a marked enhancement in κ-receptor
selectivity (200-1000-fold) over µ receptors. These
compounds were also extremely selective (9000-28000
times) for κ over δ receptors. It is of interest to note
that a distinct trend toward greater selectivity was
associated with increasing bulk of the substituents
(benzyl > CPM > allyl).
In the sodium shift assay (Table 5) the ability of the
compounds to displace [3H]diprenorphine binding to
guinea pig cerebellar membranes was determined in the
absence and presence of 120 mM sodium chloride and
50 µM Gpp(NH)p. In this assay, low sensitivity to
sodium ion and guanine nucleotide is expected for
compounds lacking agonist activity. Except for the
N-benzyl derivative, the rank order potency of the Dyn
A derivatives in the absence of sodium and Gpp(NH)p
was similar to that seen in the radioligand binding
assay using [3H]bremazocine. [D-Pro10]Dyn A-(1-11)
showed a large shift in the binding affinity ratio,
consistent with the potent agonist activity observed in
the GPI assay, but the ratio was somewhat smaller than
that reported by Gairin et al. (IC50(+)/IC50(-) ) 140).7
Gairin et al. reported that N,N-diallylation of [D-Pro10]-
Dyn A-(1-11) caused complete loss of sensitivity to

2736 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 17
Choi et al.
Ta ble 5. Effects of 120 mM NaCl + 50 µM Gpp(NH)p on Binding of [D-Pro10]Dyn A-(1-11) Analogues at κ Sites in the Guinea Pig
Cerebellum
IC50 (nM)a
peptide
(-)
(+)
ratio IC50(+)/IC50(-)
[D-Pro10]Dyn A-(1-11)
0.26 ( 0.002
21.3 ( 2.2
16.5 ( 2.0
64.6 ( 4.3
391 ( 47
63.8
3.03
9.05
8.79
99.4
12.8
66.9
N,N-diallyl[D-Pro10]Dyn A-(1-11)
N,N-dibenzyl[D-Pro10]Dyn A-(1-11)
N,N-diCPM[D-Pro10]Dyn A-(1-11)
N-allyl[D-Pro10]Dyn A-(1-11)
N-benzyl[D-Pro10]Dyn A-(1-11)
N-CPM[D-Pro10]Dyn A-(1-11)
43.2 ( 2.8
3.31 ( 0.21
2.30 ( 0.25
7.32 ( 0.58
0.41 ( 0.05
29.1 ( 1.7
229 ( 20
93.7 ( 9.4
27.1 ( 2.0
a
(-) indicates the absence of 120 mM NaCl + 50 µM Gpp(NH)p and (+) indicates the presence of the 120 mM NaCl + 50 µM Gpp(NH)p.
Ta ble 6. Opioid Activity of [D-Pro10]Dyn A-(1-11) Analogues
Among the tested compounds, only the monosubsti-
tuted N-allyl and the N-CPM analogues exhibited
potent opioid activity in the GPI. Low pA2 values for
antagonism of these analogues by naloxone were con-
sistent with the κ-receptor selectivity observed in the
radioligand binding assays. In spite of its high affinity
for κ receptors in the binding assays, the N-benzyl
derivative behaved as a very weak agonist in the GPI.
The intermediate value of the N-benzyl analogue in the
sodium shift assay is in agreement with the weak
agonist activity found in the GPI assay. This compound
did not, however, exhibit any significant antagonist
activity against Dyn A-(1-13)NH2 at doses up to 1 µM.
In the MVD assay, the N-benzyl compound exhibited
negligible agonist activity (<10% inhibition at a con-
centration of 2 µM) and no µ- or κ-receptor antagonism.
All of the analogues except the N,N-dibenzyl analogue
were examined in vivo in the phenylquinone abdominal
stretching assay following icv administration. Among
the tested compounds, the N-CPM derivative showed
the most potent antinociceptive activity with an ED50
of 1.1 µg/mouse, while the lead compound N,N-diallyl-
[D-Pro10]Dyn A-(1-11) and the N-monoallyl analogue
showed similar weak analgesic activity, with ED50
values of 26 and 27 µg/mouse, respectively. Neither of
these two peptides produced antagonist activity against
0.1 µg/mouse of [D-Ala2]Dyn A-(1-13)NH2 at a dose of
10 µg/mouse. The N,N-diCPM and the N-benzyl deriva-
tives did not exhibit significant agonist or antagonist
activity at doses up to 10 µg/mouse.
in the Guinea Pig Ileum
IC50 (nM)a
naloxone pA2
b
peptide
[D-Pro10]Dyn A-(1-11)
0.22 (0.11-0.49) 7.2 (6.6-7.7)
>3000c
N,N-diallyl[D-Pro10]Dyn A-(1-11)
N,N-dibenzyl[D-Pro10]Dyn A-(1-11) 228 (128-562)
N,N-diCPM[D-Pro10]Dyn A-(1-11) >10000
N-allyl[D-Pro10]Dyn A-(1-11)
N-benzyl[D-Pro10]Dyn A-(1-11)
N-CPM[D-Pro10]Dyn A-(1-11)
18.3 (13.2-22.4) 7.6 (7.0-8.2)
990 (657-1500)
2.16 (1.6-2.8)
7.3 (6.8-7.7)
a
Ninety-five percent confidence intervals are given in paren-
theses. The pA2 value for morphine was 8.2 (7.9-8.5). c After
b
treatment with N,N-diallyl[D-Pro10]Dyn A-(1-11), the twitch
height of the ileum declined gradually with time.
sodium ion and Gpp(NH)p (IC50(+)/IC50(-) ) 1.5),7 but
in our assay this compound displayed a slightly larger
shift (ratio ) 3.0). The other two N,N-dialkylated
analogues gave intermediate values (8.8-9.0) for the
sodium shift. Both the N-monoallyl and N-CPM deriva-
tives showed large sodium shifts, which is consistent
with their potent agonist activity seen in the GPI assay
(see below). The N-benzyl derivative exhibited a mod-
erate shift (12.8), which is consonant with its weak
agonist activity determined in the GPI assay (see below).
Opioid activity of all of the compounds was deter-
mined in the GPI assay (Table 6), and selected ana-
logues were examined in the mouse vas deferens.
Incorporation of substituents at the N-terminus reduced
agonist potency relative to the parent peptide. For N,N-
diallyl[D-Pro10]Dyn A-(1-11), typical agonist activity
was not observed following treatment with this deriva-
tive, but the twitch height of the ileum declined gradu-
ally with time. Because of this effect on the tissues, the
potential antagonist activity of this compound was
determined by performing individual dose rather than
a cumulative dose-response curve for Dyn A-(1-13)-
NH2 (see the Experimental Section). At a concentration
of 10 µM this compound caused a 5-fold rightward shift
in the dose-response curve of Dyn A-(1-13)NH2 in the
GPI, which is weaker antagonist activity than reported
by Gairin et al. (Ke ) 130 nM against [D-Pro10]Dyn
A-(1-11)).7 The differences in results for this compound
may be partially due to strain differences of guinea pigs
used in our study (Hartley) vs in the experiments by
Gairin et al. (tricolor). Among the N,N-dialkylated
compounds only the N,N-dibenzyl derivative exhibited
significant agonist activity, while the N,N-diCPM ana-
logue showed negligible agonist activity. At doses up
to 1 µM the N,N-diCPM derivative did not display any
significant antagonist activity against Dyn A-(1-13)-
NH2 in the GPI assay. The N,N-diallyl and N,N-diCPM
analogues were also examined in the MVD; no agonist
activity was detected in this assay for either compound.
As in the GPI, the N,N-diallyl derivative was an
antagonist (pA2 ∼ 6.9) in the MVD.
Con clu sion s
Various N-alkylated tyrosine derivatives were intro-
duced at position 1 in [D-Pro10]Dyn A-(1-11) in order
to explore the structure-activity relationships (SAR) for
the N-terminus of Dyn A. Both the degree of substitu-
tion and identity of the alkyl group affected κ-receptor
affinity, selectivity, and efficacy. In particular, N-
monoalkylation led to marked increases in κ-receptor
selectivity with retention of high affinity for this recep-
tor. The N-allyl and N-CPM derivatives were potent
agonists in the GPI, while the N-benzyl analogue
exhibited weak agonist activity in this assay, in spite
of high affinity for κ receptors in the binding assay. The
N-benzyl derivative exhibited a decreased IC50 ratio in
the sodium shift assay compared to the parent peptide
or other N-monosubstituted derivatives, consistent with
the decreased potency observed in the GPI.
The identity of the alkyl group in the N,N-disubsti-
tuted analogues affected both receptor affinity and
efficacy. The disubstituted peptides had much lower
affinity for κ receptors compared to the N-monosubsti-
tuted derivatives, although one analogue, the N,N-
diCPM peptide, retained sub-nanomolar affinity for κ
receptors. The significant decrease in affinity for κ

[D-Pro10]Dynorphin A-(1-11) Analogues
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 17 2737
mass spectra (FAB-MS) were obtained on a Kratos MS50RF
(Department of Agricultural Chemistry, Oregon State Uni-
versity) at +8 kV in the positive or negative mode. Amino
acid analysis was carried out on a Beckman System Gold
HPLC amino acid analyzer system at Oregon State University
as previously reported.18
receptors upon disubstitution may be due to steric
hindrance caused by incorporation of a second bulky
alkyl group at the N-terminus. All of the N-alkylated
analogues tested for binding affinity to κ receptor
subtypes preferentially interacted with κ1 followed by
κ3 receptors; compared to the unsubstituted peptide
these derivatives exhibited greatly reduced affinity for
κ2 receptors. In the sodium shift assay the N,N-
disubstituted peptides exhibited significantly lower
sodium shifts than the parent peptide, suggesting that
these peptides warranted further investigation as pos-
sible antagonists. In the GPI the N,N-diallyl derivative
exhibited antagonist activity, consistent with the previ-
ous report by Gairin et al.,7 although only at high
concentrations. In contrast, the N,N-dibenzyl analogue
was a full agonist in the GPI. The N,N-diCPM dis-
played neither agonist nor antagonist activity in the
smooth muscle assays at the concentrations examined.
Functional assays utilizing smooth muscle prepara-
tions and in vivo assays are complicated by the presence
of multiple opioid receptors. With the cloning of the
opioid receptors it is now possible to examine the effects
of opioid ligands on second messenger systems in cells
expressing a single type of receptor. Using Chinese
hamster ovary (CHO) cells stably expressing rat κ
receptors, we have examined the efficacies of selected
peptides as inhibitors of adenylyl cyclase production of
cAMP.16 In this assay both N,N-diCPM- and N,N-
diallyl[D-Pro10]Dyn A-(1-11) exhibit greatly reduced
efficacy (maximum inhibition of adenylyl cyclase <25%)
and were able to reverse the inhibition of adenylyl
cyclase caused by 10 nM Dyn A-(1-13)NH2 in a dose-
dependent manner.16
Syn t h esis of N,N-Disu b st it u t ed Tyr osin e Der iva -
tives: Gen er a l P r oced u r e. Tyrosine tert-butyl ester (0.5-
1.0 g, Sigma Chemical Co., St. Louis, MO) was dissolved in
DMF (2 mL/mmol TyrOtBu), and N,N-diisopropylethylamine
(DIEA, 2.5 equiv) was added to the solution dropwise, followed
by alkyl bromide (6-7 equiv). The reaction mixture was
stirred either at room temperature or at 60 °C under N2. The
progress of the reaction was followed by TLC (EtOAc), and
after completion of the reaction (generally 5-6 h), the mixture
was diluted with water and extracted with Et2O (3 × 20 mL).
The combined extracts were washed with saturated NaCl,
dried over Na2SO4, filtered, and concentrated under reduced
pressure to yield the N,N-disubstituted tyrosine tert-butyl ester
as an oil. The tert-butyl ester was then cleaved by 90% TFA/
8% anisole/2% DCM at room temperature under N2 for 4 h.
The mixture was then concentrated in vacuo and dried under
vacuum overnight. Crystallization and recrystallization from
MeOH/Et2O gave the desired tyrosine derivatives.
N,N-Dia llyltyr osin e. Tyrosine tert-butyl ester (0.51 g, 2.2
mmol) was reacted with DIEA (0.68 g, 5.4 mmol) and allyl
bromide (1.8 g, 15.1 mmol) in DMF (4.5 mL) at 60 °C under
N2 for 5 h according to the general procdure. The product was
isolated and the tert-butyl group removed as described above.
Recrystallization from MeOH/Et2O provided the desired prod-
uct as the free acid (361 mg, 63%): mp 196-198 °C; [R]25
)
+15.4° (c 2.0, AcOH); HPLC tR ) 8.8 min (100% purity); FAB-
1
MS 260 (M - 1); H NMR (DMSO-d6) δ 9.0 (s, 1H), 6.95 (d, J
) 8.3 Hz, 2H), 6.64 (d, J ) 8.3 Hz, 2H), 5.65 (m, 2H), 5.09 (m,
4H), 3.46 (t, J ) 7.5 Hz, 1H), 3.29 (dd, J ) 6.9 and 14.7 Hz,
2H), 3.09 (dd, J ) 6.9 and 14.7 Hz, 2H), 2.85 (dd, J ) 13.8
and 7.5 Hz, 1H), 2.70 (dd, J ) 13.8 and 7.5 Hz, 1H). Anal.
(C15H19NO3) C, H, N.
N,N-Diben zyltyr osin e. Tyrosine tert-butyl ester (1.0 g,
4.22 mmol) was alkylated with DIEA (1.37 g, 10.6 mmol) and
benzyl bromide (4.44 g, 25.3 mmol) in 8.5 mL of DMF under
N2 at room temperature for 6 h and then extracted and
concentrated as described above. The oily residue was purified
by medium-pressure chromatography (DCM/3% EtOAc) to
afford 1.67 g (94.3%) of N,N-dibenzyltyrosine tert-butyl ester
as an oil. The residue was deprotected and then crystallized
from MeOH/Et2O to yield the desired product as the free acid
(0.90 g, 59%): mp 181-183 °C; [R]25 ) -0.174° (c 0.5, MeOH);
The N-substituted Dyn A analogues described here
provide valuable SAR information in our search for
potent and selective agonists and antagonists for
κ-opioid receptors. The N-monoalkylated dynorphin
derivatives have high affinity for κ receptors and are
among the most κ-selective opioid peptides reported to
date,17 showing comparable or greater selectivity and
higher affinity than those of the κ-selective non-peptide
agonists U-50,488 and U-69,593.10 Therefore these
peptides will be useful tools for future studies of the
structure and function of κ-opioid receptors. The N,N-
diCPM analogue represents a new lead compound
which, along with the N,N-diallyl peptide, can be
modified further in an attempt to identify potent and
selective κ-receptor antagonists.
1
HPLC tR ) 22.0 min (98% purity); FAB-MS 360 (M - 1); H
NMR (MeOH-d4) δ 7.31 (m, 10H), 6.91 (d, J ) 8.3 Hz, 2H),
6.68 (d, J ) 8.3 Hz, 2H), 4.18 (d, J ) 13.6 Hz, 2H), 4.10 (d, J
) 13.6 Hz, 2H), 3.81 (t, J ) 7.3 Hz, 1H), 3.18 (dd, J ) 8.0 and
14.1 Hz, 1H), 3.10 (dd, J ) 8.0 and 14.1 Hz, 1H). Anal.
(C23H23NO3‚CF3CO2H) C, H, N.
N,N-Bis(cyclop r op ylm eth yl)tyr osin e. Tyrosine tert-bu-
tyl ester (0.76 g, 3.2 mmol) was treated with DIEA (1.03 g,
8.0 mmol) and cyclopropylmethyl bromide (2.6 g, 19.2 mmol)
in 7 mL of DMF under N2. The remainder of the procedure,
including deprotection, followed the general procedure de-
scribed above. Crystallization from MeOH/Et2O gave the pure
product (561 mg, 61%): mp 219-220 °C; [R]25 ) -0.482° (c
0.5, DMA); HPLC tR ) 11.4 min (100% purity); FAB-MS 290
Exp er im en ta l Section
Ma ter ia ls. The sources of Fmoc-amino acids, reagents, and
solvents for syntheses and purification were the same as
previously reported.18 Melting points were determined on a
Thomas-Hoover capillary apparatus and are uncorrected.
Thin-layer chromatography was performed on silica gel plates
(Kieselgel 60 F254, 0.20 mm thickness, E. Merck) and visual-
ized by UV or by staining with 0.2% ninhydrin in isobutanol
and heating. Medium-pressure column chromatography was
performed with 230-400 mesh, 60 Å silica gel (E. Merck).
Proton nuclear magnetic resonance (1H NMR) spectra of N,N-
dialkylated tyrosine derivatives were recorded on a Bruker
300A (300 MHz) spectrometer in the Department of Chemistry
at Oregon State University. Elemental analysis was per-
formed by M-H-W Laboratories, Phoenix, AZ.
1
(M + 1); H NMR (MeOH-d4) δ 7.14 (d, J ) 8.5 Hz, 2H), 6.70
(d, J ) 8.5 Hz, 2H), 4.21 (t, J ) 7.2 Hz, 1H), 3.37 (d, J ) 7.6
Hz, 1H), 3.31 (d, J ) 7.5 Hz, 1H), 3.11 (m, 4H), 1.12 (m, 2H),
0.70 (m, 4H), 0.39 (m, 4H). Anal. (C17H23NO3) C, H, N.
Syn t h esis of N-Mon osu b st it u t ed Tyr osin e Der iva -
tives: Gen er a l P r oced u r e. Tyrosine tert-butyl ester (0.5-
0.8 g) was dissolved in DMF (2 mL/mmol TyrOtBu) under N2,
and the solution was cooled to 0 °C. DIEA (1 equiv) and alkyl
halide (1 equiv) were added slowly. The progress of the
reaction was monitored by TLC (EtOAc). The reaction mixture
was diluted with water and extracted with EtOAc (3 × 20 mL).
The EtOAc extract was washed with saturated NaCl, dried
(Na2SO4), filtered, and concentrated under reduced pressure
to dryness. Following purification of the N-alkylated tyrosine
tert-butyl ester derivatives, deprotection and crystallization
The equipment used for solid phase peptide syntheses and
for analytical and preparative HPLC systems was described
previously.18 The solvents used for analytical and preparative
HPLC were aqueous 0.1% TFA for solvent A and 0.1% TFA in
acetonitrile (AcCN) for solvent B. Fast atom bombardment

2738 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 17
Choi et al.
were performed as described in the general procedure for N,N-
disubstituted tyrosine derivatives above.
bestatin, and nonspecific binding was determined in the
presence of 10 µM Dyn A-(1-13)NH2. The [3H]DAMGO and
[3H]DPDPE binding assays were performed at 4 °C for 5 h with
rat forebrains prepared in 50 mM Tris, pH 7.7. Each assay
tube also contained bestatin, captopril, and L-leucyl-L-leucine
at final concentrations of 10, 30, and 50 µM, respectively.
Nonspecific binding was determined in the presence of 10 µM
levorphanol for the [3H]DAMGO binding assay and in the
presence of 10 µM unlabeled DPDPE for the [3H]DPDPE
binding assay. Competition data were fitted by nonlinear
regression analysis using GraphPad. The IC50 values derived
from competition analyses were converted to dissociation
constants (Ki) using the Cheng and Prusoff equation;22 the KD
values used for the conversion of IC50 to Ki values for [3H]-
labeled bremazocine, DAMGO, and DPDPE were 0.0547,
0.314, and 7.63 nM, respectively.
Selected analogues were evaluated for affinity for κ-receptor
subtypes (κ1, κ2, κ3) by Stanford Research Institute (NIDA
contract no. 271-89-8159) according to standard procedures:
the brains from Hartley guinea pigs were homogenized in 50
mM Tris-HCl buffer, pH 7.7 (25 mL/brain), using a Polytron,
and the homogenate centrifuged at 40000g for 15 min, reho-
mogenized, and centrifuged. The final pellet was resuspended
in Tris-HCl, pH 7.7, at a final concentration of 6.67 mg original
wet weight of tissue/mL, except for tissues prepared for [3H]-
NalBzoH (naloxone benzoylhydrazone) binding which were
resuspended in buffer containing 5 mM EDTA. The following
radioligands (∼1 nM) were used to label the specific receptor
binding sites: [3H]U69,593 (κ1), [3H]bremazocine in the pres-
ence of 100 nM DAMGO, DSLET and U69,593 (κ2), and [3H]-
NalBzoH in the presence of 100 nM U69,593 (κ3). The guinea
pig brain suspension (1.8 mL) was incubated in 50 mM Tris-
HCl, pH 7.7, for 1 h at 25 °C with 100 µL of radioligand and
100 µL of test compound (10-5-10-11 M). Nonspecific binding
was determined in the presence of 1 µM unlabeled U69,593
for κ1 assays and 10 µM bremazocine and NalBzoH for the κ2
and κ3 assays, respectively. Each tube contained a peptidase
inhibitor cocktail (6.25 µg/mL bacitracin, 10 µM bestatin, and
0.3 µM thiorphan). The suspensions were then filtered
through glass fiber filters on a 48-well Brandel cell harvester
and the filters washed with 3 × 3 mL of buffer. Filters were
incubated overnight with 5 mL of scintillation cocktail before
counting. The Ki values were determined using the Cheng and
Prusoff equation.22 The KD values were obtained by computer
analysis of detailed self-inhibition curves for each of the labeled
ligands (L), using the curve-fitting program LIGAND.
The sensitivity of the binding of the prepared compounds
to sodium and Gpp(NH)p was measured with [3H]diprenor-
phine in guinea pig cerebellar membranes. Guinea pig
cerebellar membranes were prepared as previously described,21
except that 50 mM Tris, pH 7.4, was used as the buffer.
Equilibrium binding experiments were performed at 25 °C for
60 min with peptidase inhibitors (10 µM bestatin and 1 mM
Leu-Leu) using a final [3H]diprenorphine concentration of 0.3
nM in either the absence or presence of 120 mM NaCl and 50
µM Gpp(NH)p, as reported by Gairin et al.23 Final incubation
volume was 1 mL, and each assay tube also contained 100 nM
DAMGO. Nonspecific binding was determined in the presence
of 1 µM Dyn A-(1-13)NH2. The reaction was terminated by
rapid filtration over glass filters (Whatman GF/B) using a
Brandel M24-R cell harvester; the filters were presoaked for
2 h in 0.5% polyethyleneimine to decrease nonspecific filter
binding. The filter disks were then placed in minivials with
4 mL of Cytocint (ICN Radiochemicals) and allowed to elute
for 6 h before counting in a Beckman LS6800 scintillation
counter.
Sm ooth Mu scle Assa ys. Opioid activity in the GPI was
determined as described previously.21 Agonist activity was
determined following addition of cumulative doses of the
dynorphin analogues to the bath. Mean IC50 values were
determined from three to six replications using tissues from
different guniea pigs. In order to examine antagonist activity,
the tested peptides, except for N,N-diallyl[D-Pro10]Dyn A-(1-
11), were added to the bath in the presence of peptidase
inhibitors (100 nM bestatin (200 µL) and 3 nM thiorphan (200
µL)) 10 min prior to the determination of the IC50 of Dyn A-(1-
13)NH2. For N,N-diallyl[D-Pro10]Dyn A-(1-11) following de-
N-Allyltyr osin e. Tyrosine tert-butyl ester (0.75 g, 3.2
mmol) in 7 mL of DMF was treated with DIEA (0.41 g, 3.2
mmol) and allyl chloride (0.24 g, 3.2 mmol) according to the
general procedure described above. After 40 h, the crude
product was isolated as described above. The residue was
purified by medium-pressure chromatography (91% CHCl3/8%
MeOH/1% HOAc/3% 2-propanol) to separate the monosubsti-
tuted compound from the starting material and disubstituted
product, yielding 358 mg (41%) of pure N-allyltyrosine tert-
butyl ester. The tert-butyl group was cleaved as described
above and the free acid crystallized from Et2O (142 mg, 20%):
mp 243-244 °C; FAB-MS 222 (M + 1); HPLC tR ) 3.6 min
(99% purity).19 Anal. (C12H15NO3) C, H, N.
N-Ben zyltyr osin e. Tyrosine tert-butyl ester (0.5 g, 2.1
mmol) in 4.5 mL of DMF was reacted with DIEA (0.27 g, 2.1
mmol) and benzyl chloride (0.36 g, 2.1 mmol) for 45 h according
to the general procedure, the reaction mixture was diluted with
water and extracted with Et2O (3 × 20 mL), and the crude
product was isolated as described above. The residue was
purified by medium-pressure chromatography (90% DCM/10%
EtOAc/2.5% 2-propanol) to afford 390 mg (57%) of N-benzyl-
tyrosine tert-butyl ester. After deprotection N-benzyltyrosine
was recrystallized from Et2O, yielding 298 mg (52.3%): mp
232-234 °C; FAB-MS 272 (M + 1); HPLC tR ) 10.7 min (100%
purity).19 Anal. (C16H15NO3) C, H, N.
N-(Cyclop r op ylm eth yl)tyr osin e. Tyrosine tert-butyl es-
ter (0.8 g, 3.37 mmol) in 9 mL of DMF was alkylated with
DIEA (0.44 g, 3.37 mmol) and cyclopropylmethyl bromide (0.46
g, 3.37 mmol) according to the general procedure. The reaction
was terminated after 37 h, diluted with water, and extracted
with Et2O (3 × 20 mL). The crude product was isolated as
described above. Three hundred milligrams of the residue (out
of 824 mg) was purified by medium-pressure chromatography
(91% CHCl3/8% MeOH/1% HOAc/5% 2-propanol), yielding 280
mg (28%) of pure N-(cyclopropylmethyl)tyrosine tert-butyl
ester. The tert-butyl group was cleaved by the deprotection
method described above and the product recrystallized from
Et2O to give 125 mg (16%) of the pure free acid: mp 227-229
°C; FAB-MS 236 (M + 1); HPLC tR ) 5.0 min (100% purity).19
Anal. (C13H17NO3) C, H, N.
P ep tid e Syn th esis. Resin-bound [D-Pro10]Dyn A-(2-11)
was assembled on a (hydroxymethyl)phenoxyacetic acid resin
(PAC resin, 0.5 g, 0.21 mmol substitution) by Fmoc solid phase
synthesis using DIC couplings as described previously.18 The
side chains of Arg and Lys were protected by Pmc and Boc,
respectively. After removal of the final Fmoc protecting group,
the N-alkylated tyrosine derivatives were coupled to the
assembled peptide chain using BOP reagent.12 Prior to
addition to the resin, the N-alkylated tyrosine (2 equiv) was
activated using BOP:HOBt:N-methylmorpholine (2:2:3 equiv)
in DMA (0.2 M final concentration). The progress of the
coupling reaction was monitored by ninhydrin.20 After comple-
tion of the coupling reaction, the peptide-resin was washed
with 10 × 5 mL of DCM/DMA (1:1), 7 × 5 mL of DCM, and 4
× 5 mL of MeOH and dried in vacuo.
The crude peptides were released from the support using 5
mL of a TFA/H2O (95%/5%) mixture under N2 for 4 h. The
resin was filtered and rinsed with an additional 2 mL of TFA.
The filtrate was concentrated under reduced pressure and
dried in vacuo. The crude peptide was then triturated with
cold Et2O and collected by filtration.
The crude peptides were purified by preparative reverse
phase HPLC employing a linear gradient of 0-50% solvent B
over 50 min at a flow rate of 20 mL/min. The purity of each
fraction was evaluated by analytical reverse phase HPLC, and
pure fractions were combined and lyophilized. Analytical data
for the purified peptides are listed in Tables 1 and 2.
Op ioid Recep tor Bin d in g Assa ys. The inhibitory effects
of the peptides on the binding of [3H]bremazocine (κ) to guinea
pig cerebellar membranes and of [3H]DAMGO (µ) and [3H]-
DPDPE (δ) to rat forebrain membranes were determined as
previously described.21 Briefly, κ-receptor binding assays were
carried out at 4 °C for 3 h with [3H]bremazocine and guinea
pig cerebellar membranes in 20 mM HEPES, pH 7.4. Each
assay tube also contained 100 nM DAMGO and 100 nM

[D-Pro10]Dynorphin A-(1-11) Analogues
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 17 2739
(4) Turcotte, A.; Lalonde, J .-M.; St.-Pierre, S.; Lemaire, S. Dynor-
phin-(1-13). I. Structure-Function Relationships of Ala-
Containing Analogs. Int. J . Pept. Protein Res. 1984, 23, 361-
367.
(5) Shaw, J . S.; Miller, L.; Turnbull, M. J .; Gormley, J . J .; Morley,
J . S. Selective Antagonists at the Opiate Delta-Receptor. Life
Sci. 1982, 31, 1259-1262.
(6) Pert, C. B.; Bowie, D. L.; Pert, A.; Morell, J . L.; Gross, E. Agonist-
Antagonist Properties of N-Allyl-[D-Ala]2-Met-enkephalin. Na-
ture (London) 1977, 269, 73-75.
(7) Gairin, J . E.; Mazarguil, H.; Alvinerie, P.; Botanch, C.; Cros, J .;
Meunier, J .-C. N,N-Diallyl-tyrosyl Substitution Confers Antago-
nist Properties on the κ-Selective Opioid Peptide [D-Pro10]-
Dynorphin A(1-11). Br. J . Pharmacol. 1988, 95, 1023-1030.
(8) Lemaire, S.; Parent, P.; Lapierre, C.; Michelot, R. N,N-Dially-
lated Analogs of Dynorphin A-(1-13) as Potent Antagonists for
the Endogenous Peptide and Selective κ Opioid Analgesics. In
International Narcotics Research Conference (INRC) ‘89; Quirion,
R., J hamandas, K., Gianoulakis, C., Eds.; Alan R. Liss, Inc.: New
York, 1990; Vol. 328, pp 77-80.
(9) Gairin, J . E.; Gouarderes, C.; Mazarguil, H.; Alvinerie, P.; Cros,
J . [D-Pro10]Dynorphin-(1-11) is a Highly Potent and Selective
Ligand for κ Opioid Receptors. Eur. J . Pharmacol. 1985, 106,
457-458.
(10) Choi, H.; Murray, T. F.; DeLander, G. E.; Caldwell, V.; Aldrich,
J . V. N-Terminal Alkylated Derivatives of [D-Pro10]dynorphin
A-(1-11) are Highly Selective for κ-Opioid Receptors. J . Med.
Chem. 1992, 35, 4638-4639.
(11) Cheung, S. T.; Benoiton, L. N. N-Methylamino Acids in Peptide
Synthesis. V. The Synthesis of N-tert-Butyloxycarbonyl,N-me-
thylamino Acids by N-Methylation. Can. J . Chem. 1977, 55,
906-910.
termination of the IC50 of Dyn A-(1-13)NH2 in the absence of
the N,N-diallyl derivative, each tissue was incubated with 10
µM of the N,N-diallyl peptide for 10 min followed by treatment
with a single dose of Dyn A-(1-13)NH2. This method was used
due to the gradual decline in twitch height with time following
treatment with the N,N-diallyl derivative. Naloxone antago-
nism (naloxone pA2) was measured by the standard procedure.
Various concentrations of naloxone were incubated in the bath
for 20 min before redetermination of the IC50 of the test
peptides. All Schild analyses24 were performed using a
computerized pharmacologic data analysis system.25
The MVD assay was performed by Stanford Research
Institute as follows: the vas deferens from Swiss-Webster mice
(30-35 g) were prepared as described by Hughes et al.26 The
tissues were mounted in an 8 mL, 31 °C organ bath containing
a magnesium-free Krebs solution, which was bubbled with a
mixture of oxygen and carbon dioxide (95:5). An initial tension
of 150-200 mg was used. The field stimulation parameters
were modified slightly from those described by Ro´nai et al.;27
paired shocks of 100 ms delay between supramaximal rect-
angular pulses of 1 ms duration were delivered at a rate of
0.1 Hz. A Grass S-88 electrostimulator was used for stimula-
tion. The concentrations were recorded using an isometric
transducer (Metrigram) coupled to a Grass 7D multichannel
polygraph. The agonist potency was determined as described
in the GPI assay.
In Vivo An tin ocicep tive a n d An ta gon ist Deter m in a -
tion . Selected compounds were investigated for in vivo
antinociceptive and possible antagonist activity by observing
the phenylquinone abdominal stretching response. Male CF1
mice (Charles River Breeding Laboratories, Wilmington, MA),
weighing 18-22 g at the time of testing, were housed a
minimum of 6 days under carefully controlled environmental
conditions (22.2 ( 1.1 °C; 50% average humidity; 12 h lighting
cycle/24 h). Mice were fasted overnight (16-22 h) prior to
testing (unfasted mice were used for antagonist activity
determinations). For observation of stretching responses, mice
were placed in individual clear plastic cages (13 cm long × 9
cm wide × 12.5 cm deep) with hinged clear plastic lids and
wire screen bottoms arranged in units of 30 on a 20° incline.
Randomized and coded doses of test compounds were admin-
istered by intracerebroventricular (icv) injection in a volume
of 5 µL/mouse. Observation times represent the period from
drug administration to the start of the observation period; the
phenylquinone challenge dose (1.25 mg/kg ip phenyl-p-benzo-
quinone) was injected in a volume of 0.25 mL/20 g at 5 min
prior to the specified observation time. The concentration of
phenylquinone solution was 0.1 mg/mL in 5% aqueous ethanol.
For scoring purposes a “stretch” was indicated by whole body
stretching or full contraction of the abdomen. Mice were
observed for 10 min for the presence or absence of the
characteristic abdominal contraction and stretching response,
beginning 5 min after the phenylquinone injection. Antinoci-
ception was indicated by a complete blockade of the stretching
response. Greater than 95% of the control (vehicle-treated)
mice are expected to exhibit a stretching response. Antinoci-
ceptive activity was calculated as the percentage of mice failing
to respond to the phenylquinone challenge dose. ED50 values
were determined by the moving averages method28 and
statistical significance determined by ø2 analysis.
(12) Castro, B.; Dormoy, J .-R.; Evin, G.; Selve, C. Reactifs de
Couplage Peptidique IV (1)-L’hexafluorophosphate de Benzot-
riazolyl N-Oxytrisdimethylamino Phosphonium (B.O.P.). Tetra-
hedron Lett. 1975, 1219-1222.
(13) Hudson, D. Methodological Implications of Simultaneous Solid-
Phase Peptide Synthesis. 1. Comparison of Different Coupling
Procedures. J . Org. Chem. 1988, 53, 617-624.
(14) Robson, L. E.; Foote, R. W.; Maurer, R.; Kosterlitz, H. W. Opioid
Binding Sites of the κ-Type in Guinea Pig Cerebellum. Neuro-
science 1984, 12, 621-627.
(15) Gillan, M. G. C.; Kosterlitz, H. W. Spectrum of the µ-, δ- and
κ-Binding Sites in Homogenates of Rat Brain. Br. J . Pharmacol.
1982, 77, 461-469.
(16) Soderstrom, K.; Choi, H.; Aldrich, J . V.; Murray, T. F. N-
Alkylated Derivatives of [D-Pro10]Dynorphin A-(1-11) are High
Affinity Partial Agonists at the Cloned Rat κ-Opioid Receptor,
submitted.
(17) Lung, F.-D. T.; Meyer, J .-P.; Li, G.; Lou, B.-S.; Stropova, D.;
Davis, P.; Yamamura, H. I.; Porreca, F.; Hruby, V. J . Highly κ
Receptor-Selective Dynorphin A Analogues with Modifications
in Position 3 of Dynorphin A(1-11)-NH2. J . Med. Chem. 1995,
38, 585-586.
(18) Choi, H.; Aldrich, J . V. Comparison of Methods for the Fmoc
Solid-Phase Synthesis and Cleavage of a Peptide Containing
Both Tryptophan and Arginine. Int. J . Pept. Protein Res. 1993,
42, 58-63.
(19) NMR data for N-monoalkylated tyrosine derivatives could not
be obtained due to their insolubility in any solvent except DMA.
(20) Stewart, J . M.; Young, J . D. Solid Phase Peptide Synthesis, 2nd
ed.; Pierce Chemical Co.: Rockford, IL, 1984; p 176.
(21) Story, S. C.; Murray, T. F.; DeLander, G. E.; Aldrich, J . V.
Synthesis and Opioid Activity of 2-Substituted Dynorphin A-(1-
13) Amide Analogues. Int. J . Pept. Protein Res. 1992, 40, 89-
96.
(22) Cheng, Y. C.; Prusoff, W. H. Relationship Between the Inhibition
Constant (Ki) and the Concentration of Inhibitor Which Causes
50 Percent Inhibition (IC50) of an Enzymatic Reaction. Biochem.
Pharmacol. 1973, 22, 3099-3108.
(23) Gairin, J . E.; Botanch, C.; Cros, J .; Meunier, J . C. Binding of
Dynorphin-A and Related Peptides to Kappa-Opioid and Mu-
Opioid ReceptorssSensitivity to Na+ Ions and Gpp(NH)p. Eur.
J . Pharmacol.sMol. Pharmacol. Sect. 1989, 172, 381-384.
(24) Schild, H. O. Br. J . Pharmacol. 1947, 2, 189-206.
(25) Tallarida, R. J .; Murray, R. B. Pharmacolgical Calculations, 2nd
ed.; Springer-Verlag: New York, 1987.
Ack n ow led gm en t. This research was supported by
NIDA Grant R01 DA05195. The authors thank Dr.
Valerie Caldwell for performing the radioligand binding
assays, Martin Knittel and Elizabeth Olenchek for
performing the GPI assays, and Lisa Colona for per-
forming the in vivo assays.
(26) Hughes, J .; Kosterlitz, H. W.; Leslie, F. M. Assessment of the
Agonist and Antagonist Activities of Narcotic Analgesic Drugs
by Means of the Mouse vas deferens. Br. J . Pharmacol. 1974,
51, 139P-140P.
(27) Ro´nai, A. Z.; Gra´f, L.; Sze´kely, J . I.; Dunai-Kova´cs, Z.; Bajusz,
S. Differential Behaviour of LPH-(61-91)-Peptide in Different
Model Systems: Comparison of the Opioid Activities of LPH-
(61-91)-Peptides and its Fragments. FEBS Lett. 1977, 74, 182-
184.
(28) Thompson, W. R. Use of Moving Averages and Interpolation to
Estimate Median Effective Dose. I. Fundamental Formulas,
Estimation of Error, and Relation to Other Methods. Bacterio-
logical Rev. 1947, 11, 115-145.
Refer en ces
(1) Chavkin, C.; J ames, I. F.; Goldstein, A. Dynorphin is a Specific
Endogenous Ligand of the κ Opioid Receptor. Science 1982, 215,
413-415.
(2) Aldrich, J . V. Analgesics. In Burger’s Medicinal Chemistry and
Drug Discovery; Wolff, M. E., Ed.; J ohn Wiley & Sons, Inc.: New
York, 1996; Vol. 3: Therapeutic Agents, pp 321-441.
(3) Chavkin, C.; Goldstein, A. Specific Receptor for the Opioid
Peptide Dynorphin: Structure-Activity Relationships. Proc.
Natl. Acad. Sci. U.S.A. 1981, 78, 6543-6547.
J M960747T