Design, synthesis, and biological activities of four angiotensin II receptor ligands with γ-turn mimetics replacing amino acid residues 3-5

Article abstract of DOI:10.1021/jm960553d

Disulfide cyclization is a powerful method for reducing the conformational space of a peptide. This in turn may enable the study of its bioactive conformation. Several analogues of angiotensin II (Ang II) containing a disulfide bridge between amino acids

Full text of DOI:10.1021/jm960553d

J . Med. Chem. 1997, 40, 903-919
903
Design , Syn th esis, a n d Biologica l Activities of F ou r An gioten sin II Recep tor
Liga n d s w ith γ-Tu r n Mim etics Rep la cin g Am in o Acid Resid u es 3-5
Boris Schmidt,†,# Susanna Lindman,† Weimin Tong,† Gunnar Lindeberg,‡ Adolf Gogoll,§ Zhennan Lai,∇
Madeleine Tho¨rnwall,∇ Barbro Synnergren,∇ Annika Nilsson, Christopher J . Welch,† Morgan Sohtell,
Christer Westerlund,| Fred Nyberg,∇ Anders Karle´n,† and Anders Hallberg*,†
Departments of Organic Pharmaceutical Chemistry and Medical and Physiological Chemistry, Biomedical Centre, Uppsala
University, Box 574, SE-751 23 Uppsala, Sweden, Department of Organic Chemistry, Uppsala University, SE-751 21 Uppsala,
Sweden, Department of Biological Research on Drug Dependence, Uppsala University, SE-751 24 Uppsala, Sweden, and
Department of Pharmacology and Medicinal Chemistry II, Astra Ha¨ssle AB, SE-431 83 Mo¨lndal, Sweden
Received J uly 29, 1996X
Disulfide cyclization is a powerful method for reducing the conformational space of a peptide.
This in turn may enable the study of its bioactive conformation. Several analogues of
angiotensin II (Ang II) containing a disulfide bridge between amino acids 3 and 5 have been
reported. Among these the cyclic octapeptides c[Hcy3,5]-Ang II, c[Cys3,5]-Ang II, and c[Pen3,5]-
Ang II showed significant activity at Ang II receptors. We have performed conformational
1
analysis studies using theoretical calculations and H-NMR spectroscopy on tripeptide model
compounds of these cyclic octapeptides which show that the cyclic moieties of c[Cys3,5]-Ang II
and c[Pen3,5]-Ang II preferentially assume an inverse γ-turn conformation. On the basis of
these results, we substituted amino acid residues 3-5 in Ang II with two different γ-turn
mimetics giving four diastereomeric Ang II analogues. Interestingly, two of these are equipotent
to Ang II in binding to AT1 receptors. In the contractile test using rabbit aorta rings, one of
the analogues is an agonist with full contractile activity approximately equipotent to c[Pen3,5]-
Ang II but 300-fold less potent than Ang II. This low potency may suggest that Ang II does
not adopt a γ-turn in the 3-5 region when interacting with the receptor.
In tr od u ction
sis, J oseph et al.7 proposed putative models of the Ang
II receptor-bound conformations, one of which is closely
related to that proposed by Nikiforovich and Marshall.
Samanen et al.8 also recently proposed a refined con-
formational model of Ang II. The models above were
all partly based on the pioneering work of Spear et al.9
and Sugg et al.10 who synthesized and evaluated 3,5-
disulfide-bridged Ang II analogues.
Among the compounds synthesized by Spear et al.9
(Chart 1), c[Hcy3,5]-Ang II (1) exhibited high contractile
activity in isolated rabbit aortic rings (pD2 ) 8.48) and
high binding affinity to rat uterine membrane prepara-
tions (IC50 ) 2.1 nM). The related peptides c[Cys3,5]-
Ang II (2) and c[Pen3,5]-Ang II (3) were reported to be
essentially devoid of, or to have only weak, contractile
activity. Despite the low contractile activity of 2 and
3, they showed excellent binding affinity (43 nM for 2
vs 2.6 nM for 3). The authors concluded that these
results may indicate the presence of some partial
antagonist character, although this was not explicitly
examined.11
Small peptide hormones as well as peptide neuro-
transmitters have become important candidates for the
development of new chemical entities. However, pep-
tides are seldom directly usable as drugs due to low oral
absorption and/or rapid metabolism. Therefore, a major
challenge is to devise strategies for the transformation
of peptides into nonpeptides. A hypothesis of a bioactive
conformation can serve as a starting point for the design
of such analogues. Restriction of the conformational
freedom by cyclization provides one tool with which to
gain valuable information on the bioactive conformation-
(s).1 We have an interest in angiotensin II (Ang II;
Chart 1) which plays an important role in blood pressure
regulation. The structure-activity relationships (SAR)
of Ang II are well established2-4 making this peptide a
suitable and challenging target for the design and
synthesis of nonpeptide analogues.
Several models of the bioactive conformation of Ang
II have been proposed. Among these, a possible recep-
tor-bound conformation was identified by Nikiforovich
and Marshall by searching for a common spatial ar-
rangement of functionally important groups in low-
energy conformers of Ang II and its active analogues.5,6
This model allowed for the design of [D-Tyr4,Pro5]-Ang
II which was found to exhibit good binding affinity.5 On
the basis of conformational analysis and cluster analy-
Our objective is to develop rational and general
strategies for the transformation of peptides to nonpep-
tides by iterative incorporation of well-defined second-
ary-structure mimetics in the target peptides. In this
approach conformational restriction through side chain
disulfide formation constitutes the first step to reduce
the conformational complexity of the peptides. If the
cyclized peptide is active at the given receptor, confor-
mational analysis techniques can be used to study if the
cyclic moiety prefers certain turn geometries (γ-turns,
â-turns, etc.) which at a later stage can be substituted
for relevant secondary-structure mimetics. This is an
iterative process and should optimally lead to receptor
selective and potent nonpeptidic ligands. The apparent
* Author to whom correspondence should be addressed.
† Department of Organic Pharmaceutical Chemistry.
‡ Department of Medical and Physiological Chemistry.
§ Department of Organic Chemistry.
∇ Department of Biological Research on Drug Dependence.
Department of Pharmacology.
| Medicinal Chemistry II.
#
Current address: Institut fu¨r Organische Chemie, Universita¨t
Hannover, Schneiderberg 1B, D-30167 Hannover, Germany.
X Abstract published in Advance ACS Abstracts, February 15, 1997.
S0022-2623(96)00553-5 CCC: $14.00 © 1997 American Chemical Society

904 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
Ch a r t 1
Ch a r t 2
impact of the conformation in the 3-5 region of Ang II
for the biological response encouraged us to synthesize
secondary-structure mimetics based on the anticipated
conformational preferences of the cyclic moiety of 1-3.
We herein describe the conformational characteristics
as deduced by 1H-NMR spectroscopy and theoretical
calculations of tripeptide model compounds of 1-3.
Furthermore, we report the synthesis of four diastereo-
mers of two γ-turn mimetics incorporated into Ang II,
4a ,b and 5a ,b (Chart 2), as well as the affinities of these
compounds for rat pituitary AT1 receptors and the
contractile activity of 4a ,b on rabbit aorta strips. (a and
b denote single diastereomers of unassigned absolute
stereochemistry. We use the convention for graphic
presentation of enantiomerically pure compounds with
unknown absolute configuration, wedge outlines, and
broken lines, as suggested by Maehr.12)
selected as suitable models for the cyclic region in 1-3
and were used in the NMR spectroscopic investigation
and theoretical calculations. The assignment of all 1H-
NMR resonances of peptides 6-8 is based on the use of
the N-methyl or the tyrosine â-protons as unambiguous
starting points and was achieved with standard 2D
NMR techniques (see Supporting Information for 1H-
NMR chemical shifts for 6-8). Peptides 6 and 7 showed
only one set of signals over the monitored temperature
range (25-45 °C). Peptide 8 showed two main sets of
signals (A and B) of relative intensity 1:0.9 and a minor
set (relative intensity 0.2). The two major components
were in equilibrium with each other. Signals of the
minor component could not be assigned, since most of
them were obscured by those of the major components.
Temperature coefficients of the NH proton chemical
shifts varied between 0.0 and 8.0 ppb/K and are
provided in Table 1. In 7 and 8, the low temperature
coefficient for the NH proton of residue 3 indicates its
involvement in an intramolecular hydrogen bond,
whereas this is not the case in 6. Instead, in that
Resu lts
1
Con for m a tion a l An a lysis. (a ) H-NMR Sp ectr os-
cop y of 6-8. The tripeptides 6-8 (Chart 3) were

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 905
Ch a r t 3
and 9.6 Hz (signals for A and B overlap); 7, 5.6 and 9.1
Hz; and 6, 7.8 and 5.9 Hz. NOE data were evaluated
semiquantitatively from cross-peak volumes in ROESY
spectra but could not be assigned to a single conformer
(data not shown).
(b) Con for m a tion a l En er gy Ca lcu la tion s a n d
Tu r n Cla ssifica tion of 6-8. During the course of this
study different theoretical conformational analysis stud-
ies have been presented on tri- and octapeptide ana-
logues of Ang II, i.e., c[Hcy-Ala-Hcy],8,13 c[Cys-Ala-
Cys],8,13 c[Cys-Tyr-Cys],14 c[Cys3,5]-Ang II,7 and c[Hcy3,5]-
Ang II.7 In addition,
a
1H-NMR spectroscopic
investigation of Boc-Cys-Ala-Cys-NHCH3 has been pre-
sented.15 We have performed a conformational analysis,
using the Amber* all atom force field and the GB/SA
water solvation model16 in Macromodel version 4.5,17
on blocked model tripeptides 6-8 (Chart 3) in order to
study their conformational characteristics, especially if
preferred turn geometries exist.18 To help us classify
the conformations into γ-turns and â-turns, we used the
XCluster program19 in Macromodel. Families with
similar conformational characteristics were classified
into clusters based on the torsional rms (root-mean-
square) difference. To cluster γ-turns, the Φi+1 and Ψi+1
angles were used, and to cluster â-turns, the Φi, Ψi,
Φi+1, and Ψi+1 or the Φi+1, Ψi+1, Φi+2, and Ψi+2 angles
were used. By displaying the different clusters in
Macromodel, the conformations forming γ-turns and
â-turns can be studied. Twenty-two out of the 87 low-
energy conformations identified for 6 assumed an
inverse γ-turn conformation (Table 3). The ranges of
Φ
i+1 and Ψi+1 angles in the inverse γ-turn conformations
Ta ble 1. Temperature Coefficients of NH Chemical Shiftsa of
6-8 in DMSO-d6
were -74-(-91)° and 60-125°, respectively. When the
distance from the CR of the first residue to CR of the
fourth residue is less than 7 Å and the tetrapeptide
sequence is not in an R-helical region, it is considered
to be a â-turn.20 Only one of the 87 conformations
identified, a type I â-turn, fulfilled this distance crite-
rion. All the inverse γ-turn conformations of 6 had
energies more than 11.2 kJ /mol above the lowest energy
minimum and are therefore not significantly populated
according to a Boltzmann distribution. In the XCluster
analysis of the 19 conformations identified for 7 using
the Φi+1 and Ψi+1 angles, three clusters were identified.
Two of the three clusters, containing 18 out of the 19
conformations, contained inverse γ-turn conformations,
and the third cluster contained one type I â-turn
conformation. The ranges of Φi+1 and Ψi+1 angles in
the inverse γ-turn conformations were -73-(-124)° and
52-89°, respectively. The XCluster analysis of 8 also
shows a strong preference for the inverse γ-turn con-
formation. Twelve out of the 16 conformations identi-
fied contained an inverse γ-turn conformation. One
conformation corresponded to a type III â-turn. The
ranges of Φi+1 and Ψi+1 angles in the inverse γ-turn
conformations were -75-(-113)° and 44-78°, respec-
tively.
peptide
residue 1
residue 2
residue 3
CH3NH
6
7
4.7
6.4
5.5
6.9
4.0
4.0
3.5
3.5
8.0
0.4
0.0
0.0
0.0
4.8
4.4
4.5
8Ab
8Bb
a
b
∆δ/∆T (ppb/K). Peptide 8 showed three sets of signals, of
which two are assigned to the main conformers 8A,B.
3
Ta ble 2. J NH-CRH Coupling Constants of 6-8 in DMSO-d6
and the Possible Values for the Corresponding Backbone
Angles Φa
peptide
residue 1
residue 2
residue 3
6
7
8.0 (-147,-93) 9.2 (-132,-108)
8.0 (-147,-93)
7.2 (-154,-86) 6.5 (-159,-81,45,75) 9.1 (-133,-107)
8Ab 9.4 (-127,-113) 7.6 (-150,-90)
8.8 (-138,-102)
8.0 (-147,-93)
8Bb 9.4 (-127,-113) 7.9 (-148,-92)
a
In parentheses Φ angles (degrees) obtained from Ludvigsen’s
Karplus equation (Ludvigsen et al.24). J values are given in hertz.
Peptide 8 showed two sets of signals corresponding to the two
b
conformations in equilibrium.
compound the NH proton of the C-terminal aminometh-
yl group has a low temperature coefficient.
Vicinal NH-CRH coupling constants were measured
in the range of 6.5-9.4 Hz (Table 2). The coupling
constants in residue 3 are rather similar for all peptides.
Residue 2 shows a considerably larger value (9.2 Hz)
for peptide 6 as compared to 7 and 8. For residue 1,
peptide 8 has a significantly larger coupling constant
than the other peptides. The two major conformers of
peptide 8 (A and B) show no significant difference. For
the Tyr2 side chain, the coupling constants between
CRH and the two â-protons were determined as 8, 4.0
(c) P ossible Solu tion Con for m a tion s of 6-8. A
temperature coefficient for the amide proton chemical
shift of <3 ppb/K is indicative of intramolecular hydro-
gen bond formation.21 Thus, the temperature coef-
ficients of the residue 3 amide protons in 7 and 8 (0.4
and 0.0, respectively) and the CH3NH proton in peptide
6 (0.0) indicate that these protons are intramolecularly
hydrogen bonded. There are two possible hydrogen-

906 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
Ta ble 3. Comparison of Results Obtained from Theoretical Calculationsa and Experimental 1H-NMR Spectroscopic Data
no. of low-energy
conformationsa
no. of inverse
no. of â-turn
preferred NMR
conformationd
peptide
γ-turn conformationsb
conformationsc
6
7
8
87
19
16
22
18
12
1
1
1
no preferencee
inverse γ-turn
inverse γ-turn
a
b
The number of conformations identified within 20 kJ /mol of the lowest energy conformation. The number of inverse γ-turn
conformations identified within 20 kJ /mol of the lowest energy conformation. c The number of â-turn conformations identified within 20
kJ /mol of the lowest energy conformation. Preferred turn conformation around Tyr2 in solution as determined by 1H-NMR spectroscopy.
d
e Does not have a preferred turn geometry around Tyr2.
bonding arrangements for residue 3 in 7 and 8: a γ-turn
to residue 1 or a â-turn to the acetyl group. For the
CH3NH group in 6, there are at least three hydrogen-
bonding possibilities. Using SYBYL,22 we extracted and
examined those conformations for each peptide (within
20 kJ /mol of the lowest energy minimum) in which the
identified NH group participated as a hydrogen bond
donor.23 The Φ values of these conformations were
systematically compared to those Φ values calculated
from the 3J NH-CRH coupling constants using Ludvigsen’s
Karplus equation (Table 2).24 In trying to determine
the solution conformation, we allowed for differences in
the Φ values of (15°. The conformation with the lowest
energy that fulfilled these criteria is shown in Figure
1. In 7 we had to search for conformations above 20
kJ /mol to identify one that fulfilled the predefined
criteria. Using these principles, one conformation of
each peptide was identified that possessed the required
intramolecular hydrogen bond, had dihedral angles as
3
indicated by J NH-CRH coupling constants, and also fit
reasonably well, but not completely, to the NOE data
(Figure 1). The energy differences between these con-
formers and the lowest energy conformation identified
were ∆E ) 18.9 kJ /mol (6), 32.9 kJ /mol (7), and 8.1 kJ /
mol (8). The relatively high energy of these conforma-
tions may reflect that calculations were conducted in a
water solvent model, while NMR spectroscopy was
performed in DMSO. These conformers, however, most
likely represent an average peptide conformation, due
to the presence of a conformational equilibrium. This
is indicated in the case of 8, where separate subspectra
for at least three conformations were observed, most
probably caused by the bulky methyl groups. The main
conformers A and B are similar and may represent
different conformations of the C(Me)2S-S(Me)2C chain.
In peptide 6, containing the larger ring structure, no
F igu r e 1. Conformations of 6 (top), 7 (middle), and 8 (bottom)
derived from energy calculations and which are consistent with
hydrogen bond in the cyclic part of the molecule seems
to be present.
NMR data. The Φi, Ψi, Φi+1, Ψi+1, Φi+2, and Ψi+2 values in these
Ch em istr y. (a ) Selection a n d Design of Secon d -
conformations are -160°, 127°; -131°, 116°; -86°, 68° (6);
a r y-Str u ctu r e Mim etics to In cor p or a te in An g II.
-163°, 137°; -74°, 76°; -112°, 16° (7); -143°, 134°; -80°, 52°;
Compound 6 is highly flexible with no regular secondary
structure. In contrast, 7 and 8 preferentially adopt the
inverse γ-turn conformation (Table 3). Although these
results may not reflect the actual conformation of the
octapeptides when binding to the receptor, we decided
to focus initially on γ-turn mimetics to replace amino
acids 3-5 in Ang II. Only a few γ-turn mimetics have
been reported in the literature.25-27 We decided to use
the seven-membered azepine system, reported by Huff-
man et al.,25 represented by structure 9 (Scheme 1). In
this mimetic the amide bond between the i and i + 1
residues is replaced with a double bond and the oxygen
and hydrogen involved in the hydrogen bond are each
replaced by an ethylene group. Huffman et al. have
incorporated the template γ-turn mimetic 9 in both
linear25 and cyclic enkephalin analogues,28 in fibrinogen
-153°, 140° (8).
receptor antagonists,29 and in a model HIV-protease
substrate.30 X-ray crystallography of this template
alone revealed torsional angles quite similar to those
of an idealized inverse γ-turn,28 with the R2 substituent
in a pseudoequatorial orientation.
Synthetic features to be incorporated into the γ-turn
mimetic 9, besides Tyr in position i + 1, are the i and i
+ 2 side chains. The lipophilic side chains of Val3 and
Ile5 in Ang II are primarily considered to have a
conformation stabilizing role.31 However, it cannot be
excluded that they contribute to binding by favorable
hydrophobic interactions with the Ang II receptor.
Assuming that the side chains in Val3 and Ile5 have a
conformational stabilizing role only, it would be suf-

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 907
Sch em e 1
Sch em e 2a
ficient to introduce Gly in these positions, but since we
believe that these side chains also contribute to binding,
we desired a lipophilic side chain in either of these
positions. We preferred the lipophilic side chain at-
tached to the 5 rather than to the 3 position, since Spear
and co-workers have shown that c[Cys3,Pen5]-Ang II
exhibited higher binding affinity than c[Pen3,Cys5]-Ang
II.9 In addition, the lipophilic side chain attached to
the imidazole ring of the nonpeptidic Ang II receptor
antagonist DuP 753 has been suggested to correspond
to Ile5 in Ang II.32,33
The γ-turn mimetic represented by 10, which is
structurally similar to compounds currently under
investigation in our laboratory as potential â-turn
mimetics, was used as an alternative to the γ-turn
mimetic 9 (Scheme 1). In the bicyclic mimetic 10 the
amide bond connecting the i and i + 1 residues is
replaced by part of an aromatic ring. The characteristic
feature of this mimetic is the aromatic π-electron system
and the anilide amide. Isoleucine was introduced in the
i + 2 position based on the same criteria that were
described for γ-turn mimetic 9.
a
Reagents: (a) (i) Na, NH3, THF, EtOH, (ii) TBDPSiCl,
imidazole, DMF, 50%; (b) (i) O3, MeOH, Et2O, CH2Cl2, (ii) Me2S,
67%; (c) IleOtBu‚HCl, NaOAc, NaBH3CN, MeOH, 55%; (d) NaBH4,
MeOH, >95%; (e) TsCl, Et3N, DMAP, CH2Cl2, 90%; (f) IleOtBu‚HCl,
THF, 90%.
side reactions. Double lithiation, as described by Huff-
man et al.,25 followed by benzylation with 4-(benzyloxy)-
benzyl bromide allowed regioselective introduction of
the protected tyrosyl side chain to give the diastereo-
mers 19 in a 5:3 ratio. Mitsunobu and Gabriel reac-
tions40 followed by treatment with Fmoc-Cl afforded the
fully protected diastereomers 21. Cleavage of the tert-
butyl ester by TFA yielded the acids 22a ,b in a 3:2 ratio,
which were easily separated, but no attempt was made
to assign their absolute stereochemistry. The Fmoc
derivatives were then submitted to manual SPPS,
coupled to a His(Boc)-Pro-Phe-resin, and elaborated to
the final peptides using Fmoc-Arg(Pmc)-OH41 and Fmoc-
Asp(OtBu)-OH. As debenzylation of tyrosine is particu-
larly troublesome for Ang II analogues,42 a two-step soft
acid/hard acid deprotection was used. It furnished the
peptides 4a ,b, but in a rather poor overall yield.
Catalytic hydrogenolysis43,44 was also unsuccessful.
Classical HF treatment, however, which was applied for
the deprotection of one isomer, afforded 4a in 8.6%
overall yield.
The γ-turn mimetics 32a ,b, required for incorporation
in the pseudopeptides 5a ,b, were prepared from 3-iso-
chromanone, 23 (Schemes 4 and 5). Nitration using
ammonium nitrate and trifluoroacetic anhydride45 gave
a mixture of 6- and 7-nitro-3-isochromanones 24, which
were separable only with great difficulty and loss of
material. The mixture was therefore used without
further purification for the subsequent alkylation reac-
tion where the aromatic portion of the tyrosine was
introduced using 4-(benzyloxy)benzyl bromide and LDA
(b) Syn th esis. The synthetic approach to the ben-
zylated azepine γ-turn mimetic incorporated into 4 is
based on the strategy outlined by Huffman et al.25,29,30
utilizing anis alcohol 11 in a Birch reduction (Schemes
2 and 3).34 Protection of the alcohol35 was followed by
selective ozonolysis36,37 to give 13. The unstable alde-
hyde 13 underwent reductive amination38 with isoleu-
cine tert-butyl ester to afford the diester 16, which after
selective saponification and cyclization with diphenyl
phosphorazidate (DPPA)39 provided the lactam 17. The
reductive amination also resulted in the formation of
some isomerized product which contaminated the de-
sired compound and caused a reduced yield in the
following step. In an alternative approach, the cyclic
trimer 13a , spontaneously formed from the aldehyde 13
regardless of conditions used for storage or separation
of the latter, was reduced to give the homoallylic alcohol
14 in quantitative yield. The same alcohol was also
obtained quantitatively by reduction of 13 or reduction
of a mixture of aldehyde and trimer under the same
reaction conditions. The alcohol function was then
replaced by the amino group of the tert-butyl ester of
isoleucine in a two-step procedure. First, the tosylate
ester 15 was prepared using standard conditions; there-
after the product was used to alkylate the amino acid
ester to provide, after cyclization, the lactam 17 in high
purity. The silyl protecting group was removed with
TBAF prior to the alkylation reaction in order to avoid

908 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
Sch em e 3a
a
Reagents: (a) (i) 1 N NaOH, aq dioxane, 3 N HCl, (ii) DPPA, Na2CO3, DMF, 32%; (b) TBAF‚3H2O, THF/CH2Cl2, 71%; (c) LDA,
4-BnOBnBr, THF, 62%; (d) DEAD, PPh3, phthalimide, THF, 98%; (e) (i) H2NNH2‚H2O, EtOH, (ii) Fmoc-Cl, TBAF, dioxane, Na2CO3, 47%;
(f) TFA, CH2Cl2, anisole, 95%; (g) His(Boc)-Pro-Phe-Wang resin, TBTU, HOBt‚H2O, DIEA, DMF; (h) Fmoc-Arg(Pmc), TBTU, HOBt‚H2O,
DIEA, DMF; (i) Fmoc-Asp(OtBu), TBTU, HOBt‚H2O, DIEA, DMF; (j) TFA; (k) TFMSA, TFA, Me2S, m-cresol; (l) HF.
Sch em e 4a
a
Reagents: (a) NH4NO3, (CF3CO)2O, 81%; (b) LDA, 4-BnOBnBr, THF, 24%; (c) H2SO4, EtOH, 48%; (d) CBr4, PPh3, THF, 58%; (e)
Et3N, IleOtBu, THF, 91%; (f) toluene, AcOH.
to give pure racemic 25 after a relatively easy separa-
tion. Refluxing 25 in ethanolic sulfuric acid opened the
lactone46 to give the hydroxy ester 26, which was readily
converted into the bromide 27.47 This reaction should
be performed as soon as possible in order to avoid the
reformation of the lactone. Substitution of the bromide
with isoleucine tert-butyl ester gave the diastereomeric
mixture of 28, which cyclized to form the δ-lactams
29a ,b upon treatment with acetic acid in refluxing
toluene (Scheme 4). These diastereomers were sepa-
rated by chromatography on silica gel. All subsequent
reactions were then performed as a parallel series on
the pure compounds. Removal of the benzyl group, to
reveal the tyrosine phenolic function, was performed
simultaneously with reduction of the nitro function,
using palladium-catalyzed reduction,48 to give 30a ,b .
Standard reaction conditions were used to introduce the
Fmoc group in 31a ,b and to cleave the tert-butyl ester
giving the partially protected dipeptide analogues 32a ,b
for use in solid phase peptide synthesis (Scheme 5).
P h a r m a cology. (a ) In Vit r o Bin d in g Affin it y.
Compounds 1-5 were analyzed in a radioligand binding

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 909
Sch em e 5a
F igu r e 2. Concentration-response curves in isolated rabbit
aortic strips for Ang II (b), 3 (9), and 4b (2). Values represent
the mean ( SEM (n ) 42 for Ang II, n ) 5 for 3, and n ) 4 for
4b except at lowest and second lowest concentrations where
n ) 1 and 2, respectively).
the binding activity of other Ang II ligands markedly,
we decided to omit it in the binding assays of 4a ,b and
5a ,b.
The IC50 values of compounds 1-3 using displacement
of [125I]Ang II have previously been reported9 using a
rat uterine membrane preparation in which both AT1
and AT2 receptors are present.51 In the present study
the high affinities of 1-3 observed by Spear et al. were
confirmed. Compound 4a was equipotent to Ang II in
binding to AT1 receptors, and a similar observation was
made for compound 4b. In contrast, compounds 5a ,b
lacked any affinity for the AT1 receptor (IC50 >10-4 M).
(b) F u n ction a l Da ta . Compounds 3 and 4a ,b were
evaluated for agonistic and antagonistic properties on
rabbit aorta (Table 4). 3 and 4b behaved as full agonists
(Figure 2). Compound 4b was slightly less potent than
3 and about 300 times less active than Ang II itself. In
both 3 and 4b the concentration-response curve was
potentiated with a maximum effect more than 30%
higher than Ang II. Compound 4a showed no agonistic
effect at 10 µM. In the antagonist model no effect was
observed for the compounds tested. Compounds 5a ,b
were not tested in the functional assay since they lacked
affinity for the AT1 receptor.
a
Reagents: (a) 10% Pd/C, HCO2NH4, HCOOH, CH2Cl2; (b) (i)
Fmoc-Cl, pyridine, (ii) TFA, CH2Cl2; (c) His(Boc)-Pro-Phe-2-
chlorotrityl resin, PyBOP, HOBt‚H2O, DIEA, DMF; (d) Fmoc-
Arg(Pmc), PyBOP, HOBt‚H2O, 4-methylmorpholine, DIEA, DMF;
(e) Fmoc-Asp(OtBu), PyBOP, HOBt‚H2O, 4-methylmorpholine,
DIEA, DMF; (f) TFA/H2O/triethylsilane (90:5:5).
Ta ble 4. Binding Affinities for the AT1 Receptor and Agonistic
Activity
IC50 (nM)a
compd
AT1
AT1
EC50 (nM)d
2.6
b
c
Ang II
DuP 753
2.0
1.8
1.7
2.9
0.9
2.0
2.8
1
2
3
15.4
28
3.8
217
830
4a
4b
5a
5b
>105
>105
Discu ssion
a
Concentration of the analogue needed to produce 50% inhibi-
Introduction of γ-turn mimetic 9 (Scheme 1) in Ang
II resulted in two diastereomeric peptidomimetics, 4a ,b,
which bind to the AT1 receptor with equal or almost
equal affinity compared to Ang II. Interestingly, one
diastereomer is an agonist with full contractile activity
but 300 times less potent than Ang II. The other
diastereomer had no contractile activity, nor did it
antagonize Ang II contraction. A rationale for this
finding is at present unclear to us. When the γ-turn
mimetic 10 was incorporated in Ang II, affording
peptidomimetics 5a ,b, no receptor binding affinity was
observed. At least three reasons for the loss of binding
affinity in compounds 5a ,b as compared to 4a ,b can be
suggested as follows: (a) γ-turn mimetic 10, in contrast
to 9, does not effectively mimic the γ-turn conformation,
(b) the extra steric bulk of the aromatic ring of the
bicyclic mimetics may interact negatively with the
receptor, and (c) the pharmacophore groups after incor-
tion in specific binding of [125I]Ang II to rat pituitary membranes
(AT1). Data represent the mean of duplicate determinations.
BSA (0.2%) included in the assay mixture. c No BSA included in
b
d
the assay mixture. The concentration of the peptide which
produces 50% of the maximal contractile response at rabbit aorta
strips.
assay involving displacement of [125I]Ang II from rat
pituitary membranes, mainly containing AT1 sites
(Table 4). It has been reported that BSA (bovine serum
albumin) may affect the binding affinity of certain Ang
II receptor antagonists.49,50 Therefore, two slightly
different versions of the assay were performed for
compounds 1-3. In one version, the assay mixture
contained 0.2% BSA, while in the other, no BSA was
present. The IC50 values obtained from the assay where
BSA was omitted were 4 -10 times lower as compared
to the assay where BSA was included. Since BSA
affected the binding of 1-3, and since it has influenced

910 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
is also a fairly good geometrical mimetic of the classic
γ-turn conformation (rms ) 0.35 Å). In 34 the CRi-
CRi+2 distance in the classic and inverse γ-turns is
somewhat longer than the experimentally determined
distances. The rms fit is also slightly worse than for
33. Taken together, our theoretical calculations support
the previous findings25,26,28 that the template 9 (Scheme
1) is a good mimetic of a γ-turn. The γ-turn template
10 also shows a reasonable good fit to the experimental
γ-turn.
When 33 and 34 are superimposed using the atoms
described in the legend to Figure 5, the larger ring
skeleton of 34 is evident. In this overlay the extra van
der Waals volume of 34 compared to 33, as depicted in
Figure 5, could be deleterious for binding affinity. The
different orientation of the Ni-CRi bond vector in 33 as
compared to 34 also becomes evident in this superposi-
tion which may also be a reason for the difference in
activity of 33 and 34.
F igu r e 3. (a) Compound 33, a model compound of 4b. (b)
Conformations identified from the conformational search of 33
which adopt a classic γ-turn conformation. Relative steric
energies are as follows: ∆E ) 10.6 kJ /mol (dotted lines) and
∆E ) 31.8 kJ /mol (solid lines). (c) Conformations identified
from the conformational search of 33 which adopt an inverse
γ-turn conformation. Relative steric energies are as follows:
∆E ) 0 kJ /mol (dotted lines) and ∆E ) 19.3 kJ /mol (solid lines).
The pharmacophore elements of Ang II have been
suggested to correspond to Tyr4, His6, Phe8, and the
C-terminal carboxyl group.2 Inspection of Figure 5 and
Chart 2 indicates that residues 4-8 should be able to
assume the same conformations in 4 and 5. In 5, some
Tyr4 side chain rotamers may be energetically disfa-
vored due to unfavorable interactions between the
pseudoequatorial Tyr4 side chain and the aromatic ring
hydrogen. Therefore, the inactivity of 5 could be due
to an incorrect position of Tyr4 in respect to the other
pharmacophore groups. The hydroxyl group in Tyr4 is
necessary for agonistic activity, and a methylated hy-
droxyl group produces antagonists.53 J oseph et al. also
recently implicated the hydroxyl group of the Tyr4 side
chain in a postulated mechanism for Ang II receptor
activation.54
In one of the proposed models of the bioactive con-
formations of Ang II, a γ-turn was suggested to be
centered at Tyr4.55 If Ang II interacts with its receptor
in a γ-turn conformation, an increase in activity of 4
compared to Ang II would be expected. This would be
explained by the fact that the loss of internal rotational
entropy upon binding is smaller in the more rigid
analogue 4. The considerable loss in agonistic activity
of 4b can therefore be interpreted as that another
conformation corresponds to the bioactive conformation.
However, the structural features of γ-turn mimetic 9
are somewhat different from the native γ-turn, which
may also influence the activity, i.e., the substitution of
the peptide bond for an alkene group and/or the
replacement of the hydrogen bond for two methylene
units. Khosla et al.56 introduced an N-methyl group in
the 4 position of the Ang II antagonist sarile ([Sar1,Ile8]-
Ang II) producing antagonists with reduced activities.
Conformational studies suggested that the N-methyla-
tion changed the conformation of the peptide backbone
and limited the rotation of the Tyr4 side chain, which
makes it difficult to draw any conclusions regarding the
importance of the Tyr NH for receptor binding. Regard-
ing the ethylene bridge, Huffman et al.28 concluded that
the steric bulk introduced can change the conforma-
tional preferences of the backbone and side chains on
either side of the mimetic. The fact that [Pro5]-Ang II,57
[Sar1,Dtc5]-Ang II,58 and also the bicyclic analogue
c[Hcy3,MPt5]-Ang II,59 where the amide hydrogen in-
volved in the proposed γ-turn conformation has been
replaced by a methylene group, still show agonistic
F igu r e 4. (a) Compound 34, a model compound of 5b. (b)
Conformations identified from the conformational search of 34
which adopt a classic γ-turn conformation. The relative steric
energy is ∆E ) 0 kJ /mol. (c) Conformations identified from
the conformational search of 34 which adopt an inverse γ-turn
conformation. The relative steric energy is ∆E ) 3.4 kJ /mol.
poration of 10 in Ang II are oriented in an unfavorable
position for interacting with the receptor.
Huffman et al. compared the torsion angles for the i
+ 1 residue of a γ-turn in an ideal conformation,18 the
same angles derived from X-ray crystallography28 and
theoretical calculations26 of γ-turn mimetic 9. The
torsion angles compared favorably, and interestingly the
X-ray structures assumed the inverse γ-turn conforma-
tion. We have studied the conformational preferences
of 33 (Figure 3a), a model compound of 4, and 34 (Figure
4a), a model compound of 5, by use of a Monte Carlo
search using the MM2* force field in Macromodel (no
solvation model was used in these calculations). Four
different conformations within 50 kJ /mol of the global
minimum were identified for 33. Two of these confor-
mations corresponded to the classic γ-turn (Figure 3b)
and the other two to the inverse γ-turn (Figure 3c).
These ring conformations differ mainly in the position
of the two methylene groups; however, the direction of
the incoming and outgoing bonds from the azepine ring
are very similar in the classic and inverse γ-turn
conformation. In 34, two conformations within 50 kJ /
mol of the lowest energy conformation were identified
which correspond to the classic γ-turn and the inverse
γ-turn (Figure 4b,c, respectively).
The conformations identified were compared to ex-
perimentally determined γ-turns in proteins52 in order
to evaluate how well they mimicked the γ-turn confor-
mation (Table 5). In the comparison we focused on two
parameters: the distance between CRi and CRi+2 and
the rms distance between CRi, CRi+1,Câi+1, Ni+2, and
CRi+2 in the crystal γ-turn and the corresponding atoms
in 33 and 34. Ni was not included in the fitting
procedure since the position of this atom is not defined
by the rest of the cyclic system in 33. The CRi-CRi+2
distance is almost exactly the same in the classic and
inverse γ-turn conformations in 33 as found in the
crystal structures. The 5-atom rms deviation in the
inverse γ-turn conformation is 0.15 Å, making 33 a good
geometrical mimic of an inverse γ-turn. Compound 33

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 911
Ta ble 5. Comparison of Distances and rms Values in the Model γ-Turn Mimetics 33a and 34 to the Classicb and Inversec γ-Turns
Found in the Crystal Structure
classic γ-turn
CRi- CRi+2 (Å)d
inverse γ-turn
CRi-CRi+2 (Å)d
system/conformer
33
rms fit (Å)e
rms fit (Å)e
5.7
0.35
0.36
5.6
5.9
5.5c
0.15
0.21
34
6.0
experimental γ-turn
5.6b
a
b
The classic and inverse γ-turn conformations of 33 with lowest energy were used in the comparisons. From hemotaxis protein CheY-
Escherichia coli (PDB code name 3CHY). c From rat oncomodulin (PDB code name 1RRO). The distance (Å) between CRi and CRi+2
.
d
e Best fit between calculated and experimental γ-turn geometry (atoms fitted are CRi, CRi+1, Câi+1, Ni+2, CRi+2).
F igu r e 5. Stereorepresentation of the best fit of 33 (white) and 34 (green) in the inverse γ-turn conformation. CRi, CRi+1, Câi+1
,
Ni+2, and CRi+2 were included in the fitting procedure. Mean distance between fitted atoms ) 0.16 Å. The excess van der Waals
volume of 34 when compared to 33 is shown in yellow. The excess volume was generated in SYBYL using the MVOLUME command.
Ni was added to 33 in order to depict the Ni-CRi bond vector.
NMR spectra were recorded on a J EOL J NM-EX270 spec-
trometer at 270 (67.8) MHz or a Varian Unity 400 spectrom-
eter at 400 (100.58) MHz (final products). Chemical shifts are
reported as δ values (ppm) downfield from Me4Si. IR spectra
were recorded on a Perkin-Elmer 298 or 1600 FT-IR instru-
ment and are recorded in νmax (cm-1). Mass spectroscopy was
performed on an Applied Biosystems BIOION 20 plasma
desorption mass spectrometer. Samples were applied on
aluminized mylar foils, coated with electrosprayed nitrocel-
lulose, by drying from EtOH/H2O/TFA mixtures and removal
of excess liquid by nitrogen. Elemental analyses were per-
formed by Mikro Kemi AB, Uppsala, Sweden. Flash column
chromatography was carried out using Merck silica gel 60 (40-
63 and 15-40 µm) and 60G (5-40 µm, compunds 25-32).
Thin-layer chromatography (TLC) was carried out using
aluminum sheets precoated with silica gel 60 F254 (0.2 mm; E.
Merck). Chromatographic spots were visualized by UV and/
or spraying with an acidic, ethanolic solution of p-anisaldehyde
or an ethanolic solution of ninhydrin followed by heating. For
preparative TLC, plates precoated with silica gel 60 F254 (2.0
mm; E. Merck) were used. Optical rotations were obtained
activity does not support the γ-turn as a probable
bioactive conformation of Ang II.
Spear et al.9 measured the binding affinity and
contractile activity of 3 and showed that it had high
affinity for rat uterine membrane receptors, but a
maximal response plateau was not always obtained in
contraction studies using isolated rabbit aortic rings.
In our hands the high affinity for AT1 receptors (rat
pituitary) was confirmed. However, in the rabbit aorta
strips it had full contractile activity (n ) 5). In fact, 3
and 4b seem to have a similar pharmacological profile
in the rabbit aorta strip test (Figure 2) which may
indicate that 3 interacts with the Ang II receptor in a
γ-turn conformation.
Con clu sion
We have substituted amino acid residues 3-5 in Ang
II with two different γ-turn mimetics giving four dia-
stereomeric Ang II analogues. Introduction of γ-turn
mimetic 9 (Scheme 1) in Ang II resulted in two diaster-
eomeric peptidomimetics (4a ,b), which bind to the AT1
receptor with equal or almost equal affinity compared
to Ang II. The finding that one of the diastereomers
(4b) is an agonist with 300-fold reduced potency as
compared to Ang II indicates that Ang II is not adopting
a γ-turn around Tyr4 when interacting with the recep-
tor. The suggestion by Nikiforovich et al.6 of an open
turn in this region seems to us very likely. The reason
for the inactivity of Ang II analogues 5a ,b containing
γ-turn mimetic 10 may be due to a negative steric
interaction with the receptor of the bicyclic moiety and/
or that γ-turn mimetic 10 is not equally effective as
mimetic 9 in mimicking a γ-turn.
on
a Perkin-Elmer Model 241 polarimeter. Amino acid
analyses and peptide content measurements were performed
by Dr. M. Sundquist on 24 h hydrolysates with an LKB 4151
alpha plus analyzer using ninhydrin detection.
Ma ter ia ls. SPPS resins and amino acid derivatives were
bought from Fluka Chemie (Switzerland), NovaBiochem (Swit-
zerland), Millipore, Saxon Biochemicals (FRG), or Bachem
(Switzerland). THF was dried and distilled from sodium and
benzophenone. Diisopropylamine was distilled and stored over
NaOH. DMF was stored over 3 Å molecular sieves. All other
commercial chemicals were used without further purification.
Solid -P h a se P ep tid e Syn th esis (SP P S). Most peptides
were synthesized by manual SPPS using Fmoc/tert-butyl
protection. The starting polymer was 2-chlorotrityl chloride
resin derivatized with Fmoc-Phe-OH60 or preloaded Fmoc-Phe-
Wang resin. For the Fmoc amino acids the side chain
protecting groups used are as follows: His(Boc), Arg(Pmc), and
Asp(OtBu). Removal of the Fmoc group was achieved by
reaction with 20% piperidine in DMF for 5 + 15 min. Coupling
of the amino acids (3 equiv) was normally performed with
TBTU or PyBOP (3 equiv) in the presence of HOBt (3 equiv)
and DIEA (6 equiv) in DMF for 30-60 min. Histidine was
Exp er im en ta l Section
Ch em istr y. Gen er a l Com m en ts. Melting points (uncor-
rected) were determined in open glass capillaries on an Axel
Kistner or an Electrothermal apparatus. The 1H- and 13C-

912 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
occasionally coupled as the pentafluorophenyl ester. At the
end of each coupling cycle a resin sample (1-3 mg) was
transferred to an Eppendorf tube. The Fmoc group was
removed by treatment with 20% piperidine/DMF (200 µL) for
10 min. Ether (1 mL) was added, and the resin was carefully
washed by decantation with additional portions of ether and
allowed to dry. The peptide was released by reaction with 95%
aqueous TFA (100 µL) for 15-30 min. An aliquot (10 µL) was
added to 50% aqueous EtOH (50 µL) for PDMS analysis.
Syn th esis of Cyclic P ep tid es 1-3 a n d 6-8. The peptides
1-3 and 6-8 were obtained from the Biomolecular Resource
Facility, University of Lund, Sweden. In short, the solid-phase
synthesis of these compounds was carried out on Merrifield
or Pam resins using Boc/Bzl protection. The amino acids were
coupled as HOBt esters or as preformed symmetrical anhy-
drides. The fully deprotected octapeptides 1-3 were obtained
by cleavage with liquid HF (HF/DMS/p-cresol/p-thiocresol, 10:
1:0.75:0.25; 1.5-2.5 h at 0-5 °C), whereas the linear tripep-
tides 6-8 were cleaved from the resin by reaction with
methylamine (33% in EtOH; 3 d at room temperature) before
removal of the thiol protecting groups with HF as above. In
all cases the disulfide bridges were formed by oxidation with
iodine (20% AcOH/MeOH; 5 min). The products were purified
by RP-HPLC on a Kromasil C8 support using a 0.1% TFA-
MeCN gradient. The purity was established by HPLC and
PDMS analysis.
(E)-Meth yl 4-[[(ter t-Bu tyld ip h en ylsilyl)oxy]m eth yl]-6-
oxo-3-h exen oa te (13). 1-[[(tert-Butyldiphenylsilyl)oxy]meth-
yl]-4-methoxy-1,4-cyclohexadiene (12) (30.0 g, 79.2 mmol),
prepared according to literature,28,29 was dissolved in MeOH
(75 mL), ether (50 mL), and CH2Cl2 (20 mL), cooled to -58
°C, ozonized for 2.5 h at an O3 flow rate of 2.5 g/h (130 mmol),
quenched by addition of dimethyl sulfide (Me2S; 10 mL), and
allowed to reach room temperature overnight. The concen-
trated mixture was purified by gradient flash chromatography
(ether/pentane, 1:2 to 1:1) to yield 13 (53 mmol, 20.8 g, 67%)
as a colorless oil, which trimerized on prolonged chromatog-
raphy or standing: TLC Rf 0.35 (ether/pentane, 1:2); 1H-NMR
(CDCl3) δ 1.05 (s, 9H, t-Bu), 3.08 (d, J ) 7.3 Hz, 2H, 2), 3.17
(d, J ) 2.0 Hz, 2H, 5), 3.68 (s, 3H, OMe), 4.15 (s, 2H, CH2O),
5.90 (t, J ) 7.3 Hz, 1H, 3), 7.36-7.43 (m, 6H, Ar), 7.63-7.67
(m, 4H, Ar), 9.56 (t, J ) 2.1 Hz, 1H, 6); 13C-NMR (CDCl3) δ
19.2 (SiCMe3), 26.8 (t-Bu), 33.2 (2), 43.5 (5), 52.0 (OMe), 68.0
(CH2O), 121.8 (3), 127.8, 129.8 (Ar), 133.1 (4), 133.5, 135.6 (Ar),
mmol) was added carefully, and the reaction mixture was then
allowed to reach room temperature overnight. The mixture
was poured into aqueous NaHCO3 (200 mL), and the organic
phase was washed with NaHCO3 (1 × 100 mL) and 10% citric
acid (3 × 130 mL), dried (Na2SO4), filtered, concentrated, and
purified by flash chromatography (CH2Cl2) to give 15 (10.14
g, 90%): TLC Rf 0.24 (CH2Cl2); 1H-NMR (CDCl3) δ 1.00 (s, 9H,
t-Bu), 2.40-2.47 (m, 5H, Ar-CH3, 5), 3.05 (d, J ) 7.3 Hz, 2H,
2), 3.68 (s, 3H, OMe), 3.98-4.03 (m, 4H, 6, CH2O), 5.71 (t, J
) 7.3 Hz, 1H, 3), 7.25-7.29 (m, 2H, Ar), 7.33-7.43 (m, 6H,
Ar), 7.59-7.62 (m, 4H, Ar), 7.72 (d, J ) 8.2 Hz, 2H, Ar); 13C-
NMR (CDCl3) δ 19.1 (SiCMe3), 21.6 (Ar-CH3), 26.7 (t-Bu), 27.9
(5), 33.0 (2), 51.8 (OMe), 67.3 (CH2O), 68.4 (6), 121.0 (3), 127.7,
127.8, 129.7 (2C), 132.9 (Ar), 133.1 (4), 135.4, 136.0, 144.6 (Ar),
171.8 (1); IRCHCl 1733s, 1360s, 1175s. Anal. (C31H38O6SSi)
3
C, H.
(E)-Meth yl 6-[[1(S),2(S)-1-(ter t-Bu toxyca r bon yl)-2-m e-
th ylbu tyl]am in o]-4-[[(ter t-bu tyldiph en ylsilyl)oxy]m eth yl]-
3-h exen oa te (16). The isoleucine tert-butyl ester (3.3 g, 17.6
mmol) was dissolved in THF (15 mL), and the tosylate 15 (2.0
g, 3.53 mmol, dissolved in THF (10 mL)) was added dropwise
in 30 min at room temperature. The reaction temperature was
then raised to 80 °C. After 5 days of reflux the reaction was
quenched with aqueous NH4Cl (75 mL) and the mixture
extracted with EtOAc (3 × 60 mL), dried (Na2SO4), filtered,
concentrated, and purified by flash chromatography (ether/
pentane, 1:2) to give 16 (1.61 g, 78%): TLC Rf 0.44 (ether/
pentane, 1:2); 1H-NMR (CDCl3) δ 0.84-0.89 (m, 6H, 2′Me, 4′),
1.06 (s, 9H, t-BuSi), 1.09-1.2 (m, 1H, 3′), 1.36-1.58 (m, 12H,
t-BuO, NH, 2′, 3′), 2.22 (m, 2H, 5), 2.4 (m, 1H, 6), 2.6 (m, 1H,
6), 2.86 (d, J ) 5.9 Hz, 1H, 1′), 3.15 (d, J ) 7.3 Hz, 2H, 2),
3.69 (s, 3H, OMe), 4.12 (s, 2H, CH2O), 5.76 (t, J ) 7.3 Hz, 1H,
3), 7.34-7.42 (m, 6H, Ar), 7.65-7.68 (m, 4H, Ar); 13C-NMR
(CDCl3) δ 11.5 (4′), 15.4 (2′Me), 19.2 (SiCMe3), 25.8 (3′), 26.5,
26.8 (t-BuSi), 28.1 (t-BuO), 29.0 (5), 33.1 (2), 38.4 (2′), 46.8 (6),
51.7 (OMe), 66.4 (1′), 67.0 (CH2O), 80.7 (OCMe3), 117.5 (3),
127.6, 129.6, 133.6, 135.5 (Ar), 139.5 (4), 172.4 (CO), 174.3
(CO); IRCHCl 3324w 1745s, 1730s. Anal. (C34H51NO5Si) C, H,
3
N.
(E)-Meth yl 6-[[1(S),2(S)-1-(ter t-Bu toxyca r bon yl)-2-m e-
th ylbu tyl]am in o]-4-[[(ter t-bu tyldiph en ylsilyl)oxy]m eth yl]-
3-h exen oa te (16). Isoleucine tert-butyl ester hydrochloride
(3.87 g, 17.3 mmol) and the aldehyde 13 (4.10 g, 10 mmol) were
mixed in dry MeOH (30 mL) with NaOAc (1.43 g, 17.5 mmol)
for 10 min and treated with freshly ground 3 Å molecular
sieves (14 g) prior to the slow addition of NaBH3CN (1.07 g,
17 mmol). The mixture was stirred at room temperature for
2 h, filtered, and concentrated. The residue was dissolved in
H2O (100 mL), extracted with EtOAc, dried (Na2SO4), concen-
trated, and purified by flash chromatography (ether/petroleum
ether, 1:1) to give 16 (3.20 g, 55%) contaminated with less than
10% of an isomer. Spectroscopic data were consistent with
data described above. Anal. (C34H51NO5Si) C, H, N.
171.5 (1), 198.5 (6); IRCHCl 2725m, 1740s, 1730s, 1110s.
3
Tr im er : TLC Rf 0.27 (ether/pentane, 1:2); 1H-NMR (CDCl3)
δ 1.01 (s, 27H, t-Bu), 2.36 (d, J ) 5.6 Hz, 6H, 5), 3.18 (d, J )
7.3 Hz, 6H, 2), 4.16 (s, 6H, CH2O), 4.36 (t, J ) 5.6 Hz, 3H, 6),
5.83 (t, J ) 7.3 Hz, 3H, 3), 7.37-7.44 (m, 18H, Ar), 7.65-7.71
(m, 12H, Ar); 13C-NMR (CDCl3) δ 19.3 (SiCMe3), 26.8 (t-Bu),
32.4 (5), 33.2 (2), 53.7 (OMe), 67.7 (CH2O), 104.2 (6), 119.1
(3), 127.7, 129.6 (Ar), 133.5 (4), 135.6, 136.8 (Ar), 172.4 (1);
IRCHCl 1730s.
3
(E)-Meth yl 4-[[(ter t-Bu tyld ip h en ylsilyl)oxy]m eth yl]-6-
h yd r oxy-3-h exen oa te (14). The aldehyde and its trimer 13,
13a (8.67 g, 21.1 mmol), were dissolved in MeOH, and sodium
borohydride (0.397 g, 10.5 mmol) was added in small portions.
After 3 h the reaction was quenched (NH4Cl) and the mixture
filtered and concentrated to yield a white residue, which was
dissolved in aqueous NaHCO3 (150 mL) and extracted with
ether (4 × 130 mL). The combined organic phases were dried
(Na2SO4), filtered, concentrated, and purified by flash chro-
matography (ether/pentane, 2:1) to give 14 (8.26 g, 95%): TLC
Rf 0.45 (ether/pentane, 2:1); 1H-NMR (CDCl3) δ 1.07 (s, 9H,
t-Bu), 1.86 (t, exchangeable by D2O, 1H, OH), 2.33-2.47 (m,
2H, 5), 3.15 (d, J ) 7.6 Hz, 2H, 2), 3.63 (m, 2H, 6), 3.69 (s, 3H,
OMe), 4.11 (s, 2H, CH2O), 5.74 (m, 1H, 3), 7.36-7.47 (m, 6H,
Ar), 7.65-7.69 (m, 4H, Ar); 13C-NMR (CDCl3) δ 19.1 (SiCMe3),
26.8 (t-Bu), 31.9 (5), 33.1 (2), 51.9 (OMe), 60.8 (6), 67.8 (CH2O),
120.1 (3), 127.7, 129.66, 129.74 (Ar), 133.1 (4), 135.5 (Ar), 172.5
1-[1(S),2(S)-1-(ter t-Bu toxyca r bon yl)-2-m eth ylbu tyl]-5-
[[(ter t-bu tyldiph en ylsilyl)oxy]m eth yl]-2,3,6,7-tetr ah ydr o-
1H-a zep in -2-on e (17). The methyl tert-butyl diester 16 (33
mmol) was dissolved in dioxane (75 mL) and treated with 1
M aqueous NaOH (66 mL, 2 equiv) at room temperature for 6
h. The mixture was brought to pH 3 by 3 M aqueous HCl and
extracted with EtOAc. The combined acidic extracts were
dried (Na2SO4) and concentrated. The residual oil was filtered
through silica gel using (I) ether/pentane (1:2) and (II) EtOAc.
The ethyl acetate fractions yielded the desired acid on con-
centration sufficiently pure for immediate cyclization. The
acid (10 mmol) was dissolved in dry DMF (200 mL) and cooled
to 0 °C prior to the addition of DPPA (28 mmol, 7.7 g, 6 mL)
and Na2CO3 (100 mmol, 90 g). The mixture was stirred at 0
°C for 3 days under N2 atmosphere, then filtered (CH2Cl2), and
concentrated below 30 °C. The residue was extracted with
ether and filtered through silica gel. The concentrated filtrates
were purified by flash chromatography (ether/pentane, 1:2) to
give the lactam as colorless oil (5.80 g, 32% over two steps):
TLC Rf 0.27 (ether/pentane, 1:2); 1H-NMR (CDCl3) δ 0.85 (t, J
) 7.3 Hz, 3H, 4′), 0.94 (d, J ) 6.5 Hz, 3H, 2′Me), 1.02-1.14
(m, 10H, t-BuSi, 3′), 1.35-1.46 (m, 10H, t-BuO, 3′), 1.92 (m,
1H, 2′), 2.22 (m, 2H, 6), 3.28 (m, 2H, 3), 3.72 (t, J ) 5.8 Hz,
(1); IRCHCl 3450bm, 1730s. Anal. (C24H32O4Si) C, H.
3
(E)-Meth yl 4-[[(ter t-Bu tyld ip h en ylsilyl)oxy]m eth yl]-6-
(tosyloxy)-3-h exen oa te (15). The alcohol 14 (8.2 g, 19.4
mmol), triethylamine (6.03 g, 59.6 mmol, 8.3 mL), and (dim-
ethylamino)pyridine (0.125 g, 0.99 mmol) were mixed together
with CH2Cl2 and cooled to 0 °C. Tosyl chloride (5.68 g, 29.8

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 913
2H, 7), 3.97 (s, 2H, CH2O), 4.84 (d, J ) 10.5 Hz, 1H, 1′), 5.67
(bt, J ) 5.9 Hz, 1H, 4), 7.32-7.44 (m, 6H, Ar), 7.62-7.66 (m,
4H, Ar); 13C-NMR (CDCl3) δ 10.9 (4′), 15.4 (2′Me), 19.1
(SiCMe3), 25.0 (3′), 26.7 (SiCMe3), 27.8 (CMe3), 29.7 (6), 34.5
(2′), 35.3 (3), 41.2 (7), 60.8 (1′), 68.1 (CH2O), 81.2 (CMe3), 114.9
(4), 127.6, 129.6, 133.2, 135.4 (Ar), 138.6 (5), 170.5 (CO), 173.6
(Ar), 122.8 (4), 127.5, 127.9, 128.6, 130.1, 132.3, 137.1 (Ar),
138.8 (5), 157.2 (Ar), 171.0 (CO), 174.7 (CO); IRCHCl 3600m,
3432bm, 1720s, 1650s. Anal. (C31H41NO5‚1/2H2O) C3, H, N.
3(R,S)-3-[4-(Ben zyloxy)ben zyl]-1-[1(S),2(S)-1-(ter t-bu -
t oxyca r b on yl)-2-m et h ylb u t yl]-5-(p h t h a lim id om et h yl)-
2,3,6,7-tetr a h yd r o-1H-a zep in -2-on e (20). Am in a tion : The
diastereomeric mixture of 19 (2.37 mmol, 1.20 g), triphen-
ylphosphine (2.4 mmol, 0.65 g), and phthalimide (2.6 mmol,
0.39 g) was dissolved in dry THF (25 mL) and treated with
diethyl azodicarboxylate (DEAD; 2.8 mmol, 0.46 mL) under
N2 atmosphere for 16 h. The mixture was concentrated to a
yellow oil and triturated with ether/pentane (1:1; 50 mL). The
extracts were filtered, concentrated, and purified by gradient
flash chromatography (ether/pentane, 1:2, to ether) to give 20
as a inseparable set of diastereomers (1.48 g, 98%); TLC Rf
0.59 (ether/pentane, 2:1).
(CO); IRCDCl 3420bm, 1717s, 1647s. Anal. (C33H47NO4Si) C,
3
H, N.
1-[1(S),2(S)-(ter t-Bu t oxyca r b on yl)-2-m et h ylb u t yl]-5-
(h ydr oxym eth yl)-2,3,6,7-tetr ah ydr o-1H-azepin -2-on e (18).
Dep r otection : 17 (1.90 g, 3.45 mmol) was dissolved in CH2-
Cl2 (15 mL) and treated with tetrabutylammonium fluoride
trihydrate (TBAF‚3H2O; 1.18 g, 3.73 mmol) dissolved in dry
THF (15 mL) at room temperature for 6 h. The solution was
filtered through silica gel using (I) ether/pentane (1:1) and (II)
EtOAc. Solvent removal yielded the alcohol as thin needles
(765 mg, 71%): TLC Rf 0.34 (EtOAc); mp 92 °C; [R]23 -81.9
Ma jor isom er : 1H-NMR (CDCl3) δ 0.75 (t, J ) 7.4 Hz, 3H,
4′), 0.90 (d, J ) 6.4 Hz, 3H, 2′Me), 0.89-1.00 (m, 1H, 3′), 1.21-
1.37 (m, 1H, 3′), 1.39 (s, 9H, t-Bu), 1.82 (m, 1H, 2′), 2.23 (m,
2H, 6), 2.70 (dd, J ) 14.2, 8.6 Hz, 1H, 3-CHaAr), 3.24 (dd, J )
14.2, 6.4 Hz, 1H, 3-CHbAr), 3.63-3.90 (m, 3H, 3, 7), 4.06 (m,
2H, CH2N), 4.84 (d, J ) 10.4 Hz, 1H, 1′), 4.99 (s, 2H, PhCH2),
5.43 (m, 1H, 4), 6.77-6.86 (m, 2H, BnOC6H4), 7.05-7.14 (m,
2H, BnOC6H4), 7.31-7.46 (m, 5H, Bn), 7.68-7.76 (m, 2H,
phthal), 7.79-7.87 (m, 2H, phthal); 13C-NMR (CDCl3) δ 10.9
(4′), 15.7 (2′Me), 25.4 (3′), 28.0 (OCMe3), 31.3 (6), 33.7 (2′), 36.3
(3-CH2), 41.0 (3), 43.2 (7), 44.1 (CH2N), 61.6 (1′), 70.0 (PhCH2),
81.4 (CMe3), 114.7, 123.3 (Ar), 125.8 (4), 127.4, 127.8, 128.5,
130.1, 131.9, 132.4, 133.5, 134.0 (Ar), 137.5 (5), 157.2 (Ar),
167.8 (CO), 170.5 (CO), 174.2 (CO).
D
1
( 0.5° (c ) 1.031, CHCl3); H-NMR (CDCl3) δ 0.88 (t, J ) 7.3
Hz, 3H, 4′), 0.95 (d, J ) 6.6 Hz, 3H, 2′Me), 1.0-1.14 (m, 1H,
3′), 1.34-1.43 (m, 1H, 3′), 1.45 (s, 9H, t-Bu), 1.92 (m, 1H, 2′),
2.32 (m, 2H, 6), 3.30 (m, 2H, 3), 3.74 (t, J ) 5.8 Hz, 2H, 7),
3.95 (s, 2H, CH2O), 4.85 (d, J ) 10.6 Hz, 1H, 1′), 5.67 (m, 1H,
4); 13C-NMR (CDCl3) δ 10.9 (4′), 15.5 (2′Me), 25.1 (3′), 27.8
(OCMe3), 29.8 (6), 34.5 (2′), 35.3 (3), 41.3 (7), 60.9 (1′), 67.5
(CH2O), 81.5 (OCMe3), 115.9 (4), 139.7 (5), 170.5 (CO), 173.6
(CO); IRCHCl 3616s, 3428bm, 1722s, 1636s; GC-MS (HP-1)
25.89 min 3311 (0.1, M+), 282 (1), 210 (100, -CO2t-Bu), 192
(30, -CO2t-Bu, -H2O). Anal. (C17H29NO4) C, H, N.
3(R,S)-3-[4-(Ben zyloxy)ben zyl]-1-[1(S),2(S)-1-(ter t-bu -
toxyca r bon yl)-2-m eth ylbu tyl]-5-(h yd r oxym eth yl)-2,3,6,7-
tetr a h yd r o-1H-a zep in -2-on e (19). Ben zyla tion : A round
bottom flask (25 mL) was charged under N2 atmosphere with
dry THF (4 mL) and freshly generated LDA (3.09 mmol) and
cooled to -78 °C. A solution of 16 (320 mg, 1.03 mmol) in dry
THF (4 mL) was added over 12 min. The solution was rapidly
warmed to -40 °C to give a stable gel and after 20 min cooled
to -78 °C prior to the slow addition of 4-(benzyloxy)benzyl
bromide (512 mg, 1.85 mmol) in dry THF (2 mL). The clear
solution was stirred at -78 °C for 10 min and then rapidly
warmed to -25 °C, kept at this temperature for 30 min, and
finally warmed to 0 °C, and the reaction was quenched by
saturated NH4Cl solution (10 mL). The mixture was sus-
pended in EtOAc (15 mL), washed with H2O (3 × 15 mL), dried
(Na2SO4), filtered through silica gel, and concentrated to a
yellow oil. The oil was purified by flash chromatography
(ether/pentane, 2:1; Rf 0.24) to obtain a 9:5 mixture of the
diastereomers (322 mg, 62%), which can be separated by
repeated preparative TLC using CHCl3/EtOH (16:1).
Min or isom er : 1H-NMR (CDCl3) δ 0.70 (t, J ) 7.4 Hz, 3H,
4′), 0.88 (d, J ) 6.4 Hz, 3H, 2′Me), 0.89-1.00 (m, 1H, 3′), 1.21-
1.37 (m, 1H, 3′), 1.37 (s, 9H, t-Bu), 1.85 (m, 1H, 2′), 2.23 (m,
2H, 6), 2.69 (dd, J ) 14.2, 8.1 Hz, 1H, 3-CHaAr), 3.24 (dd, J )
14.2, 6.4 Hz, 1H, 3-CHbAr), 3.63-3.90 (m, 3H, 3, 7), 4.06 (m,
2H, CH2N), 4.83 (d, J ) 11.0 Hz, 1H, 1′), 4.99 (s, 2H, PhCH2),
5.33 (m, 1H, 4), 6.77-6.86 (m, 2H, BnOC6H4), 7.05-7.14 (m,
2H, BnOC6H4), 7.31-7.46 (m, 5H, Bn), 7.68-7.76 (m, 2H,
phthal), 7.79-7.87 (m, 2H, phthal); 13C-NMR (CDCl3) δ 10.5
(4′), 15.4 (2′Me), 24.7 (3′), 27.8 (OCMe3), 30.8 (6), 33.7 (2′), 35.4
(3-CH2), 39.7 (3), 43.0 (7), 43.9 (CH2N), 60.5 (1′), 70.0 (PhCH2),
81.3 (CMe3), 114.6, 123.3 (Ar), 124.3 (4), 127.4, 127.8, 128.5,
130.1, 131.9, 132.4, 133.6, 134.0 (Ar), 137.2 (5), 157.2 (Ar),
167.8 (CO), 170.9 (CO), 173.7 (CO); IRCHCl 3530bw, 1773m,
3
1716s, 1651m; PDMS (MW 636.8) 637.6 (M+ + H+), 660.6 (M+
+ Na). Anal. (C39H44N2O6) C, H, N.
3(R,S)-3-[4-(Ben zyloxy)ben zyl]-1-[1(S),2(S)-1-(ter t-bu -
toxyca r bon yl)-2-m eth ylbu tyl]-5-[[(9-flu or en ylm eth yloxy-
ca r bon yl)a m in o]m eth yl]-2,3,6,7-tetr a h yd r o-1H-a zep in -2-
on e (21). The diastereomeric phthalimide 20 (0.67 mmol, 427
mg) was dissolved in dry EtOH (6 mL) and treated with
hydrazine hydrate (H2NN2H‚H2O; 1.4 mmol, 68 µL) under N2
atmosphere for 24 h. The mixture was filtered through silica
gel (CHCl3/EtOH, 3:1) and concentrated to give the crude
amine. Partial purification of the residue by flash chroma-
tography (CHCl3/EtOH, 8:1) gave 192 mg, 57%; TLC Rf 0.51
(ether/pentane, 2:1). The diastereomeric amine, TBAF‚3H2O
(0.63 µmol, 20 mg), and Na2CO3 (1.14 mmol, 120 mg) were
suspended in dry dioxane (4 mL) and treated with Fmoc-Cl
(0.53 mmol, 137 mg) under N2 atmosphere for 20 h. The
mixture was filtered through silica gel (EtOAc) and concen-
trated to leave a yellow foam which was purified by flash
chromatography on silica gel 60, 15-40 µm, using EtOAc/
petroleum ether (1:3) to give 227.7 mg, 83%; TLC Rf 0.42
Ma jor isom er : TLC Rf 0.16 (CHCl3/EtOH, 100:1); [R]26
D
-31.5 ( 0.5° (c ) 0.729, CHCl3); 1H-NMR (CDCl3) δ 0.88 (t, J
) 7.3 Hz, 3H, 4′), 0.97 (d, J ) 6.6 Hz, 3H, 2′Me), 1.00-1.12
(m, 1H, 3′), 1.28-1.45 (m, 1H, 3′), 1.43 (s, 9H, t-Bu), 1.62 (bs,
1H, OH), 1.85 (m, 1H, 2′), 2.32 (m, 2H, 6), 2.71 (dd, J ) 14.2,
9.2 Hz, 1H, 3-CHaAr), 3.27 (dd, J ) 14.2, 5.6 Hz, 1H, 3-CHb-
Ar), 3.74-3.93 (m, 5H, CH2O, 7, 3), 4.88 (d, J ) 10.2 Hz, 1H,
1′), 5.03 (s, 2H, PhCH2), 5.44 (bs, 1H, 4), 6.90 (d, J ) 8.6 Hz,
2H, BnOC6H4), 7.17 (d, J ) 8.6 Hz, 2H, BnOC6H4), 7.32-7.45
(m, 5H, Bn); 13C-NMR (CDCl3) δ 11.2 (4′), 15.8 (2′Me), 25.5
(3′), 28.0 (OCMe3), 30.1 (6), 33.9 (2′), 35.8 (3-CH2Ar), 41.3 (3),
43.0 (7), 61.8 (1′), 68.1 (OCH2Ar), 70.0 (CH2OH), 81.5 (CMe3),
114.8 (Ar), 123.6 (4), 127.5, 127.9, 128.6, 130.1, 132.3, 137.0
(Ar), 139.1 (5), 158.2 (Ar), 170.5 (CO), 174.8 (CO); IRCHCl
3600m, 3430bm, 1723s, 1650s. Anal. (C31H41NO5‚1/2H2O) C3,
H, N.
Min or isom er : TLC Rf 0.15 (CHCl3/EtOH, 100:1); [R]26
(EtOAc/petroleum ether, 1:3); IRCHCl 3460m, 1780s, 1730bs,
D
3
-41.0 ( 0.5° (c ) 0.6, CHCl3); 1H-NMR (CDCl3) δ 0.74 (t, J )
7.3 Hz, 3H, 4′), 0.92 (d, J ) 6.6 Hz, 3H, 2′Me), 1.00-1.12 (m,
1H, 3′), 1.28-1.45 (m, 1H, 3′) 1.45 (s, 9H, t-Bu), 1.67 (bs, 1H,
OH), 1.85 (m, 1H, 2′), 2.32 (bs, 2H, 6), 2.72 (dd, J ) 14.2, 9.2
Hz, 1H, 3-CHaAr), 3.27 (dd, J ) 14.2, 5.3 Hz, 1H, 3-CHbAr),
3.74-3.93 (m, 5H, CH2O, 7, 3), 4.89 (d, J ) 10.5 Hz, 1H, 1′),
5.03 (s, 2H, PhCH2), 5.42 (bs, 1H, 4), 6.89 (d, J ) 8.6 Hz, 2H,
BnOC6H4), 7.17 (d, J ) 8.6 Hz, 2H, BnOC6H4), 7.30-7.45 (m,
5H, Bn); 13C-NMR (CDCl3) δ 10.6 (4′), 15.5 (2′Me), 24.8 (3′),
27.9 (OCMe3), 29.6 (6), 33.9 (2′), 36.3 (3-CH2Ar), 39.9 (3), 43.0
(7), 60.5 (1′), 68.1 (OCH2Ar), 70.0 (CH2OH), 81.5 (CMe3), 114.7
1650s.
3(R,S)-3-[4-(Ben zyloxy)ben zyl]-1-[1(S),2(S)-1-ca r boxy-
2-m eth ylbu tyl]-5-[[(9-flu or en ylm eth yloxycar bon yl)am in o]-
m eth yl]-2,3,6,7-tetr a h yd r o-1H-a zep in -2-on e (22a ,b). The
Fmoc protected diastereomeric tert-butyl esters 21 (227.7 mg,
312 µmol) were dissolved in CH2Cl2 (3 mL) and treated with
a mixture of CH2Cl2/TFA/anisole (10/3/1; 7 mL) at 0 °C for 14
h. The mixture was filtered through silica gel (CHCl3) to leave
the crude product as a white foam upon concentration. The
diastereomers were separated by flash chromatography on
silica gel 60, 15-40 µm (CHCl3/2-propanol, 11:1), to yield the

914 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
pure compounds in a ratio of 5:3 at a total yield of 95% (185 +
111 µmol, 124.2 + 74.7 mg).
removal of HF in vacuo, the resin was transferred to a glass
filter funnel and washed several times with ether, CHCl3 and
finally ether. The fully deprotected peptide 4a was then
extracted into 1 M AcOH (3 × 2.5 mL) and lyophilized. PDMS
showed, in addition to the expected compound, the presence
of a benzylated product (small amount), the tetramethylguan-
ylated tripeptide, and the peptide Asp-Arg-His-Pro-Phe.
P u r ifica t ion : The crude peptide was dissolved in 0.1 M
NH4OAc, pH 8.5, and purified by gel filtration on a Sephadex
G15 column (1 × 100 cm) in the same buffer followed by
preparative RP-HPLC on a Waters Radial Pak 10 µm µBonda-
pak C18 column (2.5 × 10 cm) with a 40 min gradient of 10-
40% MeCN or a 30 min gradient of 15-30% MeCN in 0.1%
aqueous TFA at a flow rate of 8 mL/min. The separation was
monitored at 220 and 254 nm and by PDMS. The total yield
of 4a was 11.4 mg. Amino acid analysis: Asp 0.97, Arg 0.96,
His 1.03, Pro 1.01, Phe 1.02 (54% peptide). PDMS (MW
1013.4): 1014.2 (M + H+), 1035.0 (M + Na+), 1052.3 (M +
K+).
4b: This peptide was prepared from 22b by the same
procedure as described above for 4a . Cleavage and debenzy-
lation were accomplished by treatments with TFA and TFM-
SA, respectively (procedure I above). Additional chromato-
graphic steps were included in the purification of 4b. The
material recovered after RP-HPLC was dissolved in 20 mM
Na phosphate, 0.1 M NaCl, pH 7.0 (buffer A), and applied onto
a Cu2+-loaded column of Chelating Superose (1 × 2 cm) at a
flow rate of 1 mL/min. After washing with the starting buffer
(12 mL), the peptide was desorbed by gradient elution (buffer
B: 100 mM Na phosphate, 0.1 M NaCl, pH 3.8) using a 6 min
gradient of 0-80% B followed by a 42 min gradient of 80-
100% B. The separation was monitored at 254 nm and by
analytical RP-HPLC. The fractions containing the product
were desalted and further purified by rechromatography on
the C18 support. The final yield of 4b was 2.6 mg. Amino
acid analysis: Asp 1.00, Arg 0.98, Ile 0.12, His 0.99, Pro 1.01,
Phe 1.02 (59% peptide). PDMS: 1014.4 (M + H+), 1036.3 (M
+ Na+).
7-Nitr o-3-isoch r om a n on e (24). Trifluoroacetic anhydride
(60 mL) was added to a solution of 3-isochromanone (23) (7.4
g, 50 mmol)46,61,62 and ammonium nitrate (7.9 g, 100 mmol) in
CH2Cl2 (20 mL), and the reaction mixture was stirred at room
temperature for 3 h. The mixture was concentrated, and the
residue was purified by flash chromatography to give a mixture
of 6-nitro and 7-nitro isomers in the ratio of 1:5 (7.8 g, 81%):
TLC Rf 0.24 (petroleum ether/EtOAc, 2:1); 1H-NMR (CDCl3) δ
3.85 (s, 2H, 4), 5.42 (s, 2H, 1), 7.44 (d, J ) 8.1 Hz, 1H, 5), 8.16
(d, J ) 2.2 Hz, 1H, 8), 8.23 (dd, J ) 8.1, 2.2 Hz, 1H, 6); 13C-
NMR (CDCl3) δ 36.2 (4), 69.1 (1), 120.0, 124.0, 128.2, 133.1,
138.3, 147.2 (7), 168.6 (3); IRcap.film 1757, 1529, 1348. Anal.
(C9H7NO4) C, H, N.
4(R ,S )-4-[4-(B e n zy lo x y )b e n zy l]-7-n it r o -3-is o c h r o -
m a n on e (25). Compound 24 (7.1 g, 30 mmol) was dissolved
in THF (250 mL), cooled to -78 °C, and treated with LDA (20
mL, 2 M, 40 mmol). The mixture was warmed to -30 °C and
stirred for 30 min. 4-(Benzyloxy)benzyl bromide (11.08 g, 40
mmol) was added, and the reaction mixture was stirred at 0
°C overnight. The solution was poured into aqueous citric acid
(5%, 500 mL) and extracted by CH2Cl2 (3 × 200 mL). The
organic phase was concentrated and purified by flash chro-
matography to give compound 25 (2.8 g, 24%): TLC Rf 0.50
(petroleum ether/EtOAc, 2:1), Rf 0.90 (1% MeOH in CH2Cl2);
1H-NMR (CDCl3) 3.23 (dd, J app ) 12.0, 5.2 Hz, 1H, C6H4CH2a),
3.32 (dd, J app ) 12.0, 6.8 Hz, 1H, C6H4CH2b), 4.08 (dd, J )
6.8, 5.2 Hz, 1H, 4), 4.60 (d, J ) 14.7 Hz, 1H, 1′), 5.04 (s, 2H,
C6H5CH2), 5.16 (d, J ) 14.7 Hz, 1H, 1′′), 6.83 (m, 4H,
BnOC6H4), 7.13 (d, J ) 8.3 Hz, 1H, 5), 7.32-7.43 (m, 5H, C6H5-
CH2), 7.97 (d, J ) 2.2 Hz, 1H, 8), 8.14 (dd, J ) 8.3, 2.2 Hz,
1H, 6); 13C-NMR (CDCl3) δ 38.3 (C6H4CH2), 47.6 (4), 68.7 (1),
70.0 (PhCH2), 115.1, 119.5, 123.4, 127.5, 127.9, 128.0, 128.5,
128.7, 130.2, 132.7, 136.6, 140.9, 147.0 (7), 158.9, 170.7 (3);
IRcap.film 1745, 1531, 1347. Anal. (C23H19NO5) C, H, N.
Eth yl (R,S)-2-[2-(Hyd r oxym eth yl)-4-n itr op h en yl]-3-[4-
(ben zyloxy)p h en yl]p r op a n oa te (26). Compound 25 (3.9 g,
10 mmol) was treated with H2SO4 (10 mL) in EtOH (200 mL)
at reflux overnight. The solution was poured into ice/water,
Ma jor isom er 22a : TLC Rf 0.43 (CHCl3/i-PrOH, 11:1);
[R]24D -21.4 ( 0.3° (c ) 2.218, CHCl3); 1H-NMR (CDCl3) δ 0.85
(m, 3H, 4′), 1.02 (d, J ) 6.1 Hz, 3H, 2′Me), 0.90-1.10 (bs, 1H,
3′), 1.24-1.45 (bs, 1H, 3′), 2.01 (bs, 2H, 2′), 2.19 (bs, 1H, 6),
2.71 (m, 1H, 3-CHaAr), 3.22 (m, 1H, 3-CHbAr), 3.42 (bs, 1H,
3), 3.61 (bs, 2H, 7), 3.80-4.10 (m, 2H, 5-CH2N), 4.19 (m, 1H,
Fmoc), 4.41 (d, J ) 6.1 Hz, 2H, Fmoc), 4.68 (d, J ) 8.2 Hz,
1H, 1′), 4.94 (s, 2H, PhCH2), 5.30 (s, 1H, 4), 5.65 (bs, 1H, NH),
6.84 (d, J ) 7.9 Hz, 2H, BnOC6H4), 7.11 (d, J ) 7.9 Hz, 2H,
BnOC6H4), 7.22-7.45 (m, 9H, Fmoc, Ph), 7.59 (m, 2H, Fmoc),
7.76 (d, J ) 7.4 Hz, 2H, Fmoc), 9.12 (bs, 1H, COOH); 13C-NMR
(CDCl3) δ 11.1 (4′), 16.0 (2′Me), 25.5 (3′), 30.2 (6), 34.8 (2′),
36.2 (3-CH2N), 43.1 (7), 43.5 (3), 47.3 (5-CH2N), 47.7 (Fmoc),
64.3 (1′), 66.7 (Fmoc), 69.9 (PhCH2), 114.8, 120.0 (Ar), 123.3
(4), 125.0, 127.1, 127.4, 127.8, 127.9, 128.6, 130.2, 131.7, 136.6
(Ar), 137.1 (5), 141.3, 143.9 (Ar), 156.5 (OCON), 157.3 (Ar),
174.3 (CO), 175.5 (CO); IRCHCl3 3610m, 3500bw, 3460s, 1780s,
1730bs, 1650s; PDMS (MW 672.8) 674.2 (M+ + H+), 695.8 (M+
+ Na), 712.6 (M+ + K). Anal. (C42H44N2O6) N; C: calcd, 74.98;
found, 66.2; H: calcd, 6.59; found, 5.9.
Min or isom er 22b: TLC Rf 0.37 (CHCl3/i-PrOH, 11:1) [R]24
D
1
-10.9 ( 0.3° (c ) 1.195, CHCl3); H-NMR (CDCl3) δ 0.83 (m,
3H, 4′), 0.89-1.05 (m, 1H, 3′), 1.00 (d, J ) 6.4 Hz, 3H, 2′Me),
1.19-1.39 (m, 1H, 3′), 2.08 (m, 1H, 2′), 2.35 (m, 2H, 6), 2.71
(m, 1H, 3-CHaAr), 3.22 (m, 1H, 3-CHbAr), 3.40-3.70 (m, 3H,
7, 3), 3.84 (m, 2H, 5-CH2N), 4.18 (t, J ) 6.0 Hz, 1H, Fmoc),
4.38 (d, J ) 6.0 Hz, 2H, Fmoc), 4.80 (d, J ) 8.0 Hz, 1H, 1′),
4.92 (s, 2H, PhCH2), 5.30 (s, 1H, 4), 6.10 (bs, 1H, NH), 6.82 (d,
J ) 7.8 Hz, 2H, BnOC6H4), 7.14 (d, J ) 7.8 Hz, 2H, BnOC6H4),
7.20-7.34 (m, 9H, Fmoc, Ph), 7.55-7.65 (m, 2H, Fmoc), 7.70-
7.80 (m, 2H, Fmoc), 8.6 (bs, 1H, COOH); 13C-NMR (CDCl3) δ
10.6 (4′), 15.8 (2′Me), 25.0 (3′), 29.8 (6), 33.0 (2′), 36.2 (3-CH2),
41.5 (3), 42.8 (7), 47.2 (5-CH2N), 47.4 (Fmoc), 61.9 (1′), 66.8
(Fmoc), 70.0 (PhCH2), 114.8 (Ar), 120.0 (Fmoc), 122.3 (4), 125.1,
127.2, 127.5, 127.8, 127.9, 128.6, 130.2, 132.0, 136.3 (Ar), 136.8
(5), 141.3 (Fmoc), 143.9 (Ar), 156.6 (OCON), 157.3 (Ar), 174.1
(CO), 175.3 (CO); IRCHCl3 3600w, 3440m, 1780s, 1720bs, 1650s.
Anal. (C42H44N2O6) H; C: calcd, 74.98; found, 68.1. N: calcd,
4.10; found, 3.65.
Azep in e-Ba sed An g II Mim etics 4a ,b. 4a , Syn th esis:
The Fmoc-protected template 22a (83 mg, 123 µmol), TBTU
(39.6 mg, 123 µmol), HOBt‚H2O (16.7 mg, 123 µmol), and DIEA
(43 µL, 247 µmol) were dissolved in DMF (1.8 mL) and added
to preswollen His(Boc)-Pro-Phe-Wang resin (250 mg, 123 µmol)
in a 5 mL disposable syringe equipped with a porous polyeth-
ylene filter. After mixing by slow rotation at room tempera-
ture for 3 h the resin was filtered off and washed with DMF
(6 × 3 mL). PDMS analysis of a deprotected resin sample
showed, in addition to the expected compound, a peak corre-
sponding to the tetramethylguanylated tripeptide. The free
tripeptide could not be detected. After deprotection with 20%
piperidine/DMF (2 × 3 mL, 5 + 15 min) and washing with
DMF (6 × 3 mL), the peptide was further elongated by reaction
with, in turn, Fmoc-Arg(Pmc)-OH and Fmoc-Asp(OtBu)-OH as
described above (general procedures). Fmoc deprotection,
washing with DMF and CH2Cl2, and drying in air and in vacuo
afforded the partially protected peptide resin (313 mg, 66%
yield according to the weight increase).
Clea va ge: I. TF MSA. Part of the resin (150 mg, 59 µmol)
was reacted with 95% aqueous TFA (4 mL) for 1 h. The
mixture was filtered through a small plug of glass wool in a
Pasteur pipet and the resin washed with TFA (2 × 1 mL). The
combined filtrates were evaporated in a stream of dry nitrogen
to ca. 1 mL, and the product was precipitated by the addition
of cold anhydrous ether (5 mL). The precipitate was collected
by centrifugation, washed with ether (4 × 4 mL), and dried to
furnish 48.5 mg (74%) of crude benzylated 4a . Finally, the
benzyl group was removed by reaction with TFMSA/TFA/DMS/
m-cresol (1:5:3:1), 0.1 mL/g, at 0 °C for 15-30 min. The crude
final product 4a was precipitated and isolated as above and
immediately submitted to gel filtration.
II. HF . A second batch of the partially protected peptide
resin (187 mg) was treated with HF (5 mL) in the presence of
m-cresol (300 µL) and DMS (200 µL) at 0 °C for 1 h. After

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 915
neutralized with Na2CO3, extracted with CH2Cl2 (3 × 100 mL),
concentrated, and purified by flash chromatography to give
compound 26 (2.1 g, 48%) and starting material (1.3 g, 33%):
TLC Rf 0.50 (petroleum ether/EtOAc, 2:1), Rf 0.71 (1% MeOH
in CH2Cl2); 1H-NMR (CDCl3) δ 1.16 (t, J ) 7.1 Hz, 3H,
CH2CH3), 2.99 (dd, J ) 13.7, 7.7 Hz, 1H, 3a), 3.42 (dd, J )
13.7, 7.7 Hz, 1H, 3b), 4.11 (m, 2H, CH2CH3), 4.23 (t, J ) 7.7
Hz, 1H, 2), 4.56 (d, J ) 13.2 Hz, 1H, CH2aOH), 4.68 (d, J )
13.2 Hz, 1H, CH2bOH), 5.01 (s, 2H, C6H5CH2), 6.83 (d, J ) 8.8
Hz, 2H, BnOC6H4), 7.00 (d, J ) 8.8 Hz, 2H, BnOC6H4), 7.32-
7.42 (m, 5H, C6H5), 7.69 (d, J ) 8.5 Hz, 1H, 6′), 8.15 (dd, J )
8.5, 2.6 Hz, 1H, 5′), 8.23 (d, J ) 2.6 Hz, 1H, 3′); 13C-NMR
(CDCl3) δ 13.9 (Me), 38.5 (C6H4CH2), 48.6 (2), 61.5 (CH3CH2),
62.3 (CH2OH), 69.9 (3), 114.9, 123.9, 123.3, 127.4, 127.9, 128.5,
128.6, 129.9, 130.3, 136.8, 140.6, 144.2, 147.0, 157.5, 172.3;
C25H25NO6, MW 435.4811; HRMS calcd 435.168 172 8, found
435.1707.
E t h yl (R,S)-2-[2-(Br om om et h yl)-4-n it r op h en yl]-3-[4-
(ben zyloxy)p h en yl]p r op a n oa te (27). The compound 26
(870 mg, 2 mmol) was treated with CBr4 (1.33 g, 4 mmol) and
triphenylphosphine (1.05 g, 4 mmol) in THF at room temper-
ature for 3 h. After concentration, the residue was purified
by flash chromatography to give compound 27 (580 mg, 58%):
TLC Rf 0.82 (petroleum ether/EtOAc, 2:1); 1H-NMR (CDCl3) δ
1.14 (t, J ) 7.1 Hz, 3H, CH2CH3), 3.00 (dd, J ) 13.7, 6.6 Hz,
1H, 3a), 3.42 (dd, J ) 13.7, 8.8 Hz, 1H, 3b), 4.11 (m, 2H, CH2-
CH3), 4.22 (t, J app ) 6.4 Hz, 1H, 2), 4.28 (d, J app ) 10.7 Hz,
1H, CH2aBr), 4.55 (d, J ) 10.9 Hz, 1H, CH2bBr), 5.01 (s, 2H,
C6H5CH2O), 6.87 (d, J ) 8.8 Hz, 2H, C6H4CH2), 7.09 (d, J )
8.8 Hz, 2H, C6H4CH2), 7.33-7.43 (m, 5H, C6H5), 7.75 (dd, J )
7.3, 1.5 Hz, 1H, 6′), 8.15-8.21 (m, 2H, 3′, 5′); 13C-NMR (CDCl3)
δ 14.0 (Me), 29.4 (CH2Br), 38.6 (3), 49.0 (2), 61.5 (CH3CH2),
70.0 (C6H5CH2), 115.0, 123.9, 125.2, 127.4, 127.9, 128.6, 129.5,
8.9 Hz, 1H, 5), 7.35-7.45 (m, 5H, BnO), 7.99 (dd, J ) 8.9, 2.2
Hz, 1H, 6), 8.04 (d, J ) 2.2 Hz, 1H, 8); 13C-NMR (CDCl3) δ
11.0 (Me-Ile), 15.6 (Me-Ile), 25.9 (CH2-Ile), 27.9 (t-Bu), 34.5
(CH-Ile), 38.2 (CH2Bn), 46.0 (4), 50.7 (CH-Ile), 60.4 (1), 69.2
(PhCH2), 81.9 (t-BuO), 114.8, 120.4, 122.2, 127.4, 128.0, 128.6,
128.9, 130.0, 130.3, 133.6, 136.8, 142.9, 146.7, 157.7, 170.0
(CO), 170.6 (CO); IRcap.film 1730, 1655, 1523, 1345. Anal.
(C33H38N2O6) C, H, N.
4(R/S)-7-Am in o-2-[1(S),2(S)-1-(ter t-bu toxyca r bon yl)-2-
m eth ylbu tyl]-4-(4-h yd r oxyben zyl)-1,4-d ih yd r o-3(2H)-iso-
qu in olin on e (30a ). The compound 29a (560 mg, 1 mmol) was
reduced by treatment with ammonium formate (630 mg, 10
mmol), formic acid (460 mg, 10 mmol), and 10% Pd/C (100 mg)
in CH2Cl2 (50 mL) at room temperature for 1 h. The mixture
was filtered, concentrated, and purified by flash chromatog-
raphy to give compound 30a (382 mg, 87%): TLC Rf 0.10
(petroleum ether/EtOAc, 2:1); 1H-NMR (CDCl3) δ 0.76 (t, J )
7.3 Hz, 3H, Me-Ile), 0.90-0.94 (m, 1H, CH2a-Ile), 0.98 (d, J )
6.4 Hz, 3H, Me-Ile), 1.15-1.22 (m, 1H, CH2b-Ile), 1.50 (s, 9H,
t-Bu), 2.06 (m, 1H, CH-Ile), 2.79 (dd, J ) 13.2, 8.5 Hz, 1H,
CHaBn), 3.04 (dd, J ) 13.2, 5.3 Hz, 1H, CHbBn), 3.70 (dd, J )
8.3, 5.3 Hz, 1H, 4), 4.09 (d, J ) 15.8 Hz, 1H, 1′), 4.21 (d, J )
15.8 Hz, 1H, 1′′), 5.02 (d, J ) 10.7 Hz, 1H, RH-Ile), 6.42-6.50
(m, 3H, 5, 6, 8), 6.64 (d, J ) 8.5 Hz, 2H, Bn), 6.73 (d, J ) 8.5
Hz, 2H, Bn); 13C-NMR (CDCl3) δ 10.6 (Me-Ile), 15.7 (Me-Ile),
24.6 (CH2-Ile), 28.1 (t-Bu), 32.9 (CH-Ile), 38.5 (CH2-Bn), 45.8
(4), 50.1 (CH-Ile), 60.8 (1), 81.8 (t-Bu), 111.6, 114.6, 115.2,
125.2, 128.8, 129.0, 130.5, 132.5, 144.7, 155.1, 170.0 (CO), 173.3
(CO); IRCHCl 1726, 1651. Anal. (C26H34N2O4) C, H, N.
3
4(R/S)-7-Am in o-2-[1(S),2(S)-1-(ter t-bu toxyca r bon yl)-2-
m eth ylbu tyl]-4-(4-h yd r oxyben zyl)-1,4-d ih yd r o-3(2H)-iso-
qu in olin on e (30b). The compound 29b (535 mg, 0.95 mmol)
gave 30b by using the conditions described for compound
30a : yield 345 mg, 83%; TLC Rf 0.10 (petroleum ether/EtOAc,
2:1); 1H-NMR (CDCl3) δ 0.90 (t, J ) 7.3 Hz, 3H, Me-Ile), 0.99
(d, J ) 6.6 Hz, 3H, Me-Ile), 1.03-1.12 (m, 1H, CH2a-Ile), 1.26-
1.36 (m, 1H, CH2b-Ile), 1.34 (s, 9H, t-Bu), 1.97 (m, 1H, CH-
Ile), 2.86 (dd, J ) 13.2, 8.5 Hz, 1H, CHaBn), 3.07 (dd, J ) 13.2,
4.5 Hz, 1H, CHbBn), 3.74 (dd, J ) 8.3, 4.5 Hz, 1H, 4), 4.05 (d,
J ) 15.5 Hz, 1H, 1′), 4.48 (d, J ) 15.6 Hz, 1H, 1′′), 4.98 (d, J
) 10 Hz, 1H, RH-Ile), 6.40-6.50 (m, 3H, 5, 6, 8), 6.64 (d, J )
8.5 Hz, 2H, Bn), 6.70 (d, J ) 8.5 Hz, 2H, Bn); 13C-NMR (CDCl3)
δ 10.9 (Me-Ile), 15.7 (Me-Ile), 25.6 (CH2-Ile), 27.9 (t-Bu), 34.0
(CH-Ile), 38.8 (CH2Bn), 46.3 (4), 59.7 (CH-Ile), 60.6 (1), 81.6
(t-Bu), 111.5, 114.6, 115.1, 125.1, 128.9, 129.0, 130.5, 132.5,
144.7, 155.2, 170.2 (CO), 173.0 (CO); IRCHCl3 1724, 1648. Anal.
(C26H34N2O4) C, H, N.
130.0, 130.3, 136.9, 137.6, 145.2, 146.8, 157.7, 171.9; IRCHCl
3
1730, 1524, 1348; C25H24NO5Br, MW 498.38; HRMS calcd
(C25H24NO5Br79) 497.0838, found 497.0840; calcd (C25H24NO5-
Br81) 499.0817, found 499.0815.
Eth yl 2(R,S)-3-[4-(Ben zyloxy)p h en yl]-2-[2-[[[(1(S),2(S)-
1-(ter t-bu toxyca r bon yl)-2-m eth ylbu tyl]a m in o]m eth yl]-4-
n itr op h en yl]p r op a n oa te (28). The solution of compound 27
(747 mg, 1.5 mmol), triethylamine (303 mg, 3 mmol), and
isoleucine tert-butyl ester (842 mg, 4.5 mmol) in THF (50 mL)
was heated to 45 °C and stirred for 2 h. After concentration,
the residue was purified by flash chromatography to give a
mixture of diastereomers 28 (825 mg, 91%): TLC Rf 0.88
(petroleum ether/EtOAc, 2:1).
4(R/S)-4-[4-(Ben zyloxy)ben zyl]-2-[1(S),2(S)-1-(ter t-bu -
toxyca r bon yl)-2-m eth ylbu tyl]-7-n itr o-1,4-d ih yd r o-3(2H)-
isoqu in olin on e (29a ,b). The compound 28 (800 mg, 1.56
mmol) was treated with acetic acid (2 drops) in toluene at
reflux overnight. After concentration, the residue was sepa-
rated by flash chromatography to give compounds 29a (340
mg, 47%) and 29b (330 mg, 45%). 29a : TLC Rf 0.60 (petro-
leum ether/EtOAc, 2:1); 1H-NMR (CDCl3) δ 0.78 (t, J ) 7.3
Hz, 3H, Me-Ile), 0.88-1.00 (m, 1H, CH2a-Ile), 1.01 (d, J ) 6.6
Hz, 3H, Me-Ile), 1.15-1.23 (m, 1H, CH2b-Ile), 1.52 (s, 9H, t-Bu),
2.08 (m, 1H, CH-Ile), 2.98 (dd, J ) 13.5, 8.8 Hz, 1H, CHaBn),
3.22 (dd, J ) 13.5, 5.1 Hz, 1H, CHbBn), 3.91 (dd, J ) 8.8, 5.1
Hz, 1H, 4), 4.33 (m, 2H, 1), 5.02 (s, 2H, C6H4CH2), 5.03 (d, J
) 10.7 Hz, 1H, RH-Ile), 6.78 (m, 4H, Bn), 6.86 (d, J ) 8.3 Hz,
1H, 5), 7.30-7.43 (m, 5H, BnO), 7.98 (dd, J ) 8.3, 2.3 Hz, 1H,
6), 8.03 (d, J ) 2.2 Hz, 1H, 8); 13C-NMR (CDCl3) δ 10.5 (Me-
Ile), 15.6 (Me-Ile), 24.9 (CH2-Ile), 28.1 (t-Bu), 33.0 (CH-Ile),
37.9 (CH2Bn), 45.5 (4), 50.7 (CH-Ile), 60.6 (1), 69.9 (PhCH2),
81.9 (t-Bu), 114.8, 120.3, 122.3, 127.4, 127.9, 128.5, 128.9,
128.9, 130.2, 133.6, 136.8, 142.8, 146.6, 157.6, 169.8 (CO), 170.6
(CO); IRcap.film 1728, 1655, 1526m, 1346. Anal. (C33H38N2O6‚
1/4H2O) C, H, N.
4(R/S)-2-[1(S),2(S)-1-(ter t-Bu t oxyca r b on yl)-2-m et h yl-
bu tyl]-7-[(9-flu or en ylm eth yloxyca r bon yl)a m in o]-4-(4-h y-
dr oxyben zyl)-1,4-dih ydr o-3(2H)-isoqu in olin on e (31a). The
compound 30a (220 mg, 0.5 mmol) was treated with Fmoc-Cl
(143 mg, 0.55 mmol) in pyridine at room temperature over-
night. The reaction mixture was concentrated and purified
by flash chromatography to give compound 31a (282 mg,
86%): TLC Rf 0.21 (petroleum ether/EtOAc, 2:1); 1H-NMR
(CDCl3) δ 0.70 (t, J ) 7.3 Hz, 3H, Me-Ile), 0.83-0.91 (m, 1H,
CH2a-Ile), 0.95 (d, J ) 6.6 Hz, 3H, Me-Ile), 1.11 (m, 1H, CH2b
-
Ile), 1.46 (s, 9H, t-Bu), 2.00 (m, 1H, CH-Ile), 2.77 (dd, J ) 13.2,
8.8 Hz, 1H, CHaBn), 3.05 (dd, J ) 13.2, 4.9 Hz, 1H, CHbBn),
3.71 (dd, J ) 8.8, 4.9 Hz, 1H, 4), 4.15-4.24 (m, 3H, 1, CH-
Fmoc), 4.50 (d, J ) 6.0 Hz, 2H, CH2-Fmoc), 4.99 (d, J ) 10.5
Hz, 1H, RH-Ile), 6.50 (d, J ) 8.1 Hz, 1H, 5), 6.64 (m, 4H, Bn),
6.81-6.92 (m, 2H, 6, 8), 7.23-7.40 (m, 4H, Fmoc), 7.57 (d, 2H,
Fmoc), 7.75 (d, 2H, Fmoc); 13C-NMR (CDCl3) δ 10.5 (Me-Ile),
15.7 (Me-Ile), 24.7 (CH2-Ile), 28.1 (t-Bu), 32.8 (CH-Ile), 38.3
(CH2-Bn), 45.9 (CH-Fmoc), 47.0 (4), 53.4 (CH-Ile), 60.9 (1), 66.8
(CH2-Fmoc), 81.8 (t-BuO), 115.2, 117.6, 120.0, 124.8, 127.1,
127.8, 128.5, 128.8, 130.1, 130.4, 132.5, 136.3, 141.3, 143.6,
153.5 (CO-Fmoc), 155.1, 169.7 (CO), 172.8 (CO); IRcap.film 3311,
1727, 1629. Anal. (C41H44N2O6‚1/4H2O) C, H, N.
4(R/S)-2-[1(S),2(S)-1-(ter t-Bu t oxyca r b on yl)-2-m et h yl-
bu tyl]-7-[(9-flu or en ylm eth yloxyca r bon yl)a m in o]-4-(4-h y-
dr oxyben zyl)-1,4-dih ydr o-3(2H)-isoqu in olin on e (31b). The
compound 30b (220 mg, 0.5 mmol) gave 31b by using the
conditions described for compound 31a : yield 289 mg, 88%;
TLC Rf 0.21 (petroleum ether/EtOAc, 2:1); 1H-NMR (CDCl3) δ
29b: TLC Rf 0.54 (petroleum ether/EtOAc, 2:1); 1H-NMR
(CDCl3) δ 0.93 (t, J ) 7.2 Hz, 3H, Me-Ile), 1.01 (d, J ) 6.7 Hz,
3H, Me-Ile), 1.10-1.20 (m, 1H, CH2a-Ile), 1.37 (s, m, 9H, t-Bu),
1.37-1.42 (m, 1H, CH2b-Ile), 2.00 (m, 1H, CH-Ile), 2.99 (dd, J
) 13.9, 9.2 Hz, 1H, CHaBn), 3.23 (dd, J ) 13.9, 4.8 Hz, 1H,
CHb-Bn), 3.92 (dd, J ) 9.2, 4.8 Hz, 1H, 4), 4.11 (d, J ) 16.1
Hz, 1H, 1′), 4.77 (d, J ) 16.1 Hz, 1H, 1′′), 5.02 (s, 2H, C6H4CH2),
5.04 (d, J ) 10 Hz, 1H, RH-Ile), 6.76 (m, 4H, Bn), 6.86 (d, J )

916 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
0.91 (t, J ) 7.3 Hz, 3H, Me-Ile), 1.00 (d, J ) 6.6 Hz, 3H, Me-
Ile), 1.06-1.15 (m, 1H, CH2a-Ile), 1.24-1.32 (m, 1H, CH2b-Ile),
1.34 (s, 9H, t-Bu), 1.97 (m, 1H, CH-Ile), 2.88 (dd, J ) 13.2, 8.8
Hz, 1H, CHaBn), 3.11 (dd, J ) 13.2, 4.3 Hz, 1H, CHbBn), 3.79
(dd, J ) 8.4, 4.3 Hz, 1H, 4), 4.10 (d, J ) 15.4 Hz, 1H, 1′), 4.25
(t, J ) 6.5 Hz, 1H, CH-Fmoc), 4.52 (d, 2H, CH2-Fmoc), 4.56
(d, 1H, 1′′), 5.01 (d, J ) 9.8 Hz, 1H, RH-Ile), 6.55 (d, J ) 8.3
Hz, 1H, 5), 6.67 (m, 4H, Bn), 6.93 (bd, 1H, 6), 7.04 (s, 1H, 8),
7.30 (t, 2H, Fmoc), 7.41 (t, 2H, Fmoc), 7.61 (d, 2H, Fmoc), 7.78
(d, 2H, Fmoc); 13C-NMR (CDCl3) δ 10.9 (Me-Ile), 15.8 (Me-Ile),
25.7 (CH2-Ile), 27.8 (t-Bu), 34.1 (CH-Ile), 38.6 (CH2Bn), 46.5
(CH-Fmoc), 47.0 (4), 53.4 (CH-Ile), 60.8 (1), 66.9 (CH2-Fmoc),
81.7 (t-BuO), 115.1, 117.6, 120.0, 124.9, 127.1, 127.7, 128.4,
128.7, 130.0, 130.5, 132.4, 136.3, 141.6, 143.6, 153.5 (CO-
Fmoc), 155.2, 169.9 (CO), 172.5 (CO); IRcap.film 3317bm, 1727,
1624. Anal. (C41H44N2O6‚1/4H2O) C, H, N.
DIEA (8 equiv) in DMF (500 µL) for 1 h. After washing with
DMF (6 × 1 mL), deprotection (3 × 1 mL, 5 + 5 + 20 min),
and washing with DMF (6 × 1 mL), Fmoc-Asp(OtBu)-OH was
introduced by the same double-coupling procedure to yield 65.9
mg of the partially protected peptide resin. MS analysis now
indicated complete reaction.
Clea va ge: The resin (65.9 mg, 18.3 µmol) was reacted in
the syringe with TFA/H2O/triethylsilane (90:5:5; 1 mL) for 1.5
h, filtered, and rinsed with TFA (3 × 500 µL). The combined
filtrates were evaporated in a stream of dry nitrogen to ca.
500 µL, and the product was precipitated by addition of cold
ether (12 mL). The precipitate was collected by centrifugation,
washed with ether (3 × 4 mL), and dried to yield 19.7 mg of
crude product.
P u r ifica tion : The crude peptide was dissolved in 0.1%
aqueous TFA (6 mL) and purified by RP-HPLC on a 10 µm
Vydac C18 column (2.2 × 25 cm) using a 60 min gradient of
0-60% MeCN in 0.1% TFA at a flow rate of 4 mL/min. The
separation was monitored at 230 nm and by PDMS. Some of
the fractions were submitted to rechromatography using the
same conditions. The fractions containing the pure product
were combined and lyophilized to give 6.2 mg (33% total yield)
of 5a . Amino acid analysis: Asp 1.04, Arg 1.01, His 0.98, Pro
0.96, Phe 1.01 (69% peptide). PDMS (MW 1034.7): 1035.9 (M
+ H+), 1057.5 (M + Na+).
5b: This peptide was prepared from 32b by the same
sequence of reactions. The final yield of 5b was 3.7 mg (22%).
Amino acid analysis: Asp 1.04, Arg 0.91, His 1.02, Pro 0.98,
Phe 1.05 (66% peptide). PDMS: 1035.5 (M + H+), 1057.6 (M
+ Na+).
Ra t P itu ita r y AT1 Recep tor Bin d in g Assa y. Pituitary
membranes containing Ang II AT1 receptors were prepared
from tissues collected from male Sprague-Dawley rats (Alab
AB, Sollentuna, Sweden) in a weight range of 200-220 g. The
animals were adapted to the environment as previously
described.63 They were decapitated without any prior drug
treatments, and the pituitaries were removed and kept on dry
ice. Subsequently all glands were transferred to a vessel and
homogenized by ultrasonication at 0 °C, in 25 mM Tris-HCl
buffer (pH 7.4) containing 10% (w/v) sucrose (50 mL of buffer/g
of tissue). The homogenate was centrifuged at 1000g for 15
min, and the pellet was collected and suspended in 25 mM
Tris-HCl (pH 7.4). All procedures were carried out at 5 °C or
on ice, and the suspensions were immediately used or stored
at -80 °C.
4(R/S)-2-[1(S),2(S)-1-Ca r boxy-2-m eth ylbu tyl]-7-[(9-flu o-
r en ylm et h yloxyca r b on yl)a m in o]-4-(4-h yd r oxyb en zyl)-
1,4-d ih yd r o-3(2H)-isoqu in olin on e (32a ). The compound
31a (220 mg, 0.33 mmol) was treated with TFA in CH2Cl2 at
room temperature for 1 h. The reaction mixture was concen-
trated and purified by flash chromatography to give compound
32a (178 mg, 89%): TLC Rf 0.66 (EtOH/EtOAc, 1:2); 1H-NMR
(CDCl3 + CD3OD) δ 0.83 (t, J ) 7.1 Hz, 3H, Me-Ile), 0.95-
1.05 (m, 1H, CH2a-Ile), 0.99 (d, J ) 6.6 Hz, 3H, Me-Ile), 1.26
(m, 1H, CH2b-Ile), 2.15 (m, 1H, CH-Ile), 3.04 (m, 2H, CH2-Bn),
3.80 (dd, J ) 7.6, 4.8 Hz, 1H, 4), 3.96 (d, J ) 15.8 Hz, 1H, 1′),
4.28 (t, J ) 6.3 Hz, 1H, CH-Fmoc), 4.39 (d, 1H, 1′′), 4.62 (m,
2H, CH2-Fmoc), 4.89 (d, J ) 10.9 Hz, 1H, RH-Ile), 6.52 (s, 1H,
8), 6.60 (m, 5H, 6, Bn), 6.63 (d, J ) 7.0 Hz, 1H, 5), 7.28 (t, 2H,
Fmoc), 7.38 (t, 2H, Fmoc), 7.57 (d, 2H, Fmoc), 7.75 (d, 2H,
Fmoc); 13C-NMR (CDCl3) δ 10.4 (Me-Ile), 15.8 (Me-Ile), 24.5
(CH2-Ile), 32.3 (CH-Ile), 40.2 (CH2Bn), 47.0 (CH-Fmoc), 48.6
(4), 53.4 (CH-Ile), 61.7 (1), 67.0 (CH2-Fmoc), 114.9, 115.2, 118.2,
120.1, 124.9, 127.8, 128.2, 129.4, 130.6, 132.4, 136.5, 141.3 and
143.6, 155.5 (CO-Fmoc), 173.6 (CO), 174.4 (CO); IRcap.film 3319,
1718, 1613. Anal. (C37H36N2O6‚1/4CF3COOH) C, H, N.
4(R/S)-2-[1(S),2(S)-1-Ca r boxy-2-m eth ylbu tyl]-7-[(9-flu o-
r en ylm et h yloxyca r b on yl)a m in o]-4-(4-h yd r oxyb en zyl)-
1,4-d ih yd r o-3(2H)-isoqu in olin on e (32b). The compound
31b (220 mg, 0.33 mmol) gave 32b by using the conditions
described for compound 32a : yield 168 mg, 84%; TLC Rf 0.66
(EtOH/EtOAc, 1:2); 1H-NMR (CDCl3 + CD3OD) δ 0.83 (t, J )
7.1 Hz, 3H, Me-Ile), 0.95-1.02 (m, 1H, CH2a-Ile), 0.99 (d, J )
6.4 Hz, 3H, Me-Ile), 1.27 (m, 1H, CH2b-Ile), 2.16 (m, 1H, CH-
Ile), 3.01 (m, 2H, CH2-Bn), 3.84 (m, 2H, 1′, 4), 4.24 (m, 2H, 1′′,
CH-Fmoc), 4.54 (m, 2H, CH2-Fmoc), 4.99 (d, J ) 10.9 Hz, 1H,
RH-Ile), 6.53-6.62 (m, 6H, 6, 8, Bn), 6.63 (d, J ) 7.0 Hz, 1H,
5), 7.28 (t, 2H, Fmoc), 7.39 (t, 2H, Fmoc), 7.57 (d, 2H, Fmoc),
7.76 (d, 2H, Fmoc); 13C-NMR (CDCl3) δ 10.6 (Me-Ile), 15.7 (Me-
Ile), 25.4 (CH2-Ile), 33.6 (CH-Ile), 39.5 (CH2Bn), 47.0 (CH-
Fmoc), 49.0 (4), 53.4 (CH-Ile, 1), 67.1 (CH2-Fmoc), 115.2, 120.0,
124.8, 127.8, 128.2, 129.0, 130.6, 131.5, 136.5, 141.3, 143.5,
154.9 (CO-Fmoc), 174.2 (CO), 174.4 (CO); IRCHCl3 3430, 3302bm,
1728, 1634, 1614. Anal. (C37H36N2O6‚1/4CF3COOH) C, H, N.
Isoqu in olin on e-Ba sed An g II Mim etics 5a ,b. 5a , Syn -
th esis: The Fmoc-protected template 32a (29.6 mg, 49 µmol),
PyBOP (51.0 mg, 98 µmol), HOBt‚H2O (33.1 mg, 245 µmol),
and DIEA (40 µL, 234 µmol) dissolved in DMF (500 µL) were
reacted with His(Boc)-Pro-Phe-2-chlorotrityl resin (121 mg,
47.2 µmol) in a 2 mL syringe vessel as described above for the
synthesis of 4a . The coupling was allowed to proceed for 20
h, at which point PDMS analysis of a deprotected resin sample
indicated that the reaction had gone to completion. The resin,
after washing with DMF (6 × 1 mL), was deprotected with
20% piperidine/DMF (2 × 1 mL, 10 + 20 min) and washed
with DMF (6 × 1 mL), CH2Cl2 (3 × 1 mL), and ether (3 × 1
mL). After drying in vacuo 129 mg of the partially protected
peptide resin were recovered. One-half of this material (64.4
mg, 23.6 µmol) was submitted to coupling with Fmoc-Arg-
(Pmc)-OH (5 equiv) using PyBOP (5 equiv), HOBt‚H2O (10
equiv), and 4-methylmorpholine (NMM; 15 equiv) for activa-
tion. The reaction which was carried out in DMF (500 µL)
was allowed to proceed for 19 h. PDMS analysis showed that
the coupling was still incomplete. Therefore, the arginine
derivative (4 equiv) was recoupled with PyBroP (4 equiv) and
In the binding assay the incubation mixture contained [125I]-
Ang II (105 cpm), pituitary membranes, 100 µM bacitracin, 100
mM NaCl, 10 mM MgCl2, 1 mM EGTA, and 0.2% BSA buffered
at pH 7.4 with 50 mM Tris-HCl in a final volume of 0.5 mL. A
presumable interference by binding to AT2 receptors was
blocked by the addition of 1 µM CGP-42211. Binding analysis
was also performed in the absence of BSA. Samples were
incubated at 25 °C for 1 h, and the reaction was terminated
by the addition of 0.5 mL of cold 50 mM Tris-HCl buffer (pH
7.4). The precipitates were collected by centrifugation (10000g
and 5 °C for 10 min) and counted in a γ-counter. Compounds
to be tested were dissolved in the same buffer as above.
Ca lcu la tion a n d Bin d in g Con sta n ts. After correction for
nonspecific binding, the bound radioactivity in the presence
of a given concentration of the test compound was compared
to that of a control to determine the percent inhibition. The
concentration of the test compound required to inhibit the
specific binding of [125I]Ang II by 50% (IC50) was calculated
using the EBDA-LIGAND computer program (Biosoft Cam-
bridge, U.K.). In the case of binding to pituitary AT1 sites,
the reported IC50 values represent an average of at least two
determinations from separate assays.
Va scu la r Con tr a ctility Stu d ies. Male New Zealand
white rabbits (Sollentuna, Sweden) in a weight range of 2.5-
3.5 kg were anesthetized by intravenous injection of pento-
barbitone (Mebumal vet, 30 mg/kg of body weight). The
thoracic aorta was dissected and placed into prewarmed (38.5
°C) and oxygenated Kreb’s bicarbonate buffer. Excessive fat
and connective tissue were removed, and the aorta was cut
into rings of 2 mm. The segments obtained were mounted in
3 mL organic baths containing Krebs’ buffer (120 mM NaCl,

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 917
4.75 mM KCl, 2.54 mM CaCl2, 1.2 mM MgSO4, 1.19 mM
KH2PO4, 25 mM NaHCO3, and 11 mM D-(+)-glucose). The
buffer was kept at 38.5 °C and pH 7.4 and aerated with 95%
O2 and 5% CO2. Isometric contraction was recorded using
Radnoti force transducers. Initial resting tension was set to
2 g, and the strips were equilibrated for about 90 min. During
this time the strips were stimulated once by the addition of
Ang II into the baths to give a final concentration of 300 nM.
After the maximal contractile response to this peptide was
reached, the strips were rinsed and allowed to relax to base-
line tension. After equilibration a control cumulative concen-
tration-contractile response curve for Ang II (3 × 10-10-10-7
M) was recorded. Thereafter the aorta strips were repeatedly
washed and allowed to return to base-line tension. Subse-
quent studies were divided in two groups of experiments: one
for studies of agonistic properties and the other for studies of
antagonistic properties of the test compounds.
program (command SPMC); 1000 step runs were performed,
and those conformations within 50 kJ /mol of the global
minimum were kept. Torsional memory and geometrical
preoptimization were used. Truncated Newton conjugated
gradient (TNCG) minimization with a maximum of 2000
iterations was used in the conformational search with the
default derivate convergence set at a value of 0.001 (kJ /mol)/
Å.
Ack n ow led gm en t. We thank The Swedish Re-
search Council for Engineering Sciences (TFR), Swedish
National Board for Industrial and Technical Develop-
ment (NUTEC), Astra Ha¨ssle AB, Deutsche Akademis-
cher Austauschdienst (DAAD), and Svenska Institutet,
Stockholm, for financial support. The authors express
their gratitude to Dr. J . Callahan (SK&B) for access to
unpublished results and additional experimental infor-
mation.
In the agonist experiments the contractile response of the
test compound was recorded at concentrations of 3 × 10-10
-
10-5 M to produce a cumulative concentration-response curve
in analogy with that of Ang II.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails describing the pharmacological investigation of antago-
nism of Ang II-induced pressor response in spontaneously
In the antagonist experiments the aorta strips were prein-
cubated (20 min) separately at three fixed concentrations (0.1,
1.0, and 10 µM) of the test compound before a second
cumulative concentration-contractile response curve for Ang
II was recorded. The results of the agonist/antagonist experi-
ments are expressed as a percentage of maximum control
contractile force obtained from the first cumulative concentra-
tion-response curve for Ang II.
1
hypertensive rats and a table of H-NMR chemical shifts for
tripeptides 6-8 (2 pages). Ordering information is given on
any current masthead page.
Refer en ces
1H-NMR Sp ectr oscop y of 6-8. Spectra were recorded at
400 MHz with 12-27 mM solutions in DMSO-d6 at 25-35 °C
using a Varian Unity spectrometer. Chemical shifts are
indirectly referenced to the residual solvent signal at 2.49 ppm.
Coupling constants were extracted from lD proton spectra.
Signal assignments were made from DQ-COSY,64 TOCSY,65
and ROESY65 or NOESY66,67 spectra, which all were run in
phase-sensitive mode using the method of States et al.66 The
spectral width was 3200-3700 Hz. Between 16 and 48
transients for 2150f1 increments with a relaxation delay of 1.5
s following the acquisition time were acquired. Mixing times
were set to 60 ms for the TOCSY experiments, and a series of
mixing times varying from 0.1 to 0.6 s for each sample was
used in the ROESY and NOESY experiments. A time-shared
spin-lock field > 4000 Hz was used in the ROESY experi-
ments. The data matrices were zero-filled to 2 × 2K before
Fourier transformation.
Con for m a tion a l En er gy Ca lcu la tion s. The calculations
of 6-8 were performed using the Amber* all atom force field
and the all atom charge set as implemented in the program
Macromodel version 4.5.17 The general born solvent accessible
surface area (GB/SA) method for water developed by Still16
was used in all calculations unless otherwise stated. The
number of torsion angles allowed to vary simultaneously
during each Monte Carlo step ranged from 1 to n - 1, where
n equals the total number of rotatable bonds (n ) 12 in 1 and
n ) 10 in 2 and 3). All torsion angles except the amide bonds,
which were set in the trans configuration, were defined as
rotatable. Conformational searches was conducted by use of
the systematic unbound multiple minimum search (SUMM)
method68 in the batchmin program (command SPMC); 10 000
step runs were performed, and those conformations within 50
kJ /mol of the global minimum were kept. The ring closure
bond was defined as the bond between the sulfur atoms.
Torsional memory and geometrical preoptimization were used.
Truncated Newton conjugated gradient (TNCG) minimization
with a maximum of 500 iterations was used in the conforma-
tional search with the default derivate convergence set at a
value of 0.05 (kJ /mol)/Å. In the subsequent minimization to
fully converged structures a maximum of 5000 steps of TNCG
minimization was followed by a maximum of 5000 steps of full
matrix Newton Raphson (FMNR) minimization with the
convergence criteria being 0.001 (kJ /mol)/Å in both runs.
The calculations of model compounds 33 and 34 were
performed using the MM2* force as implemented in the
program Macromodel version 4.5.17 Conformational searches
was conducted by use of the SUMM method68 in the batchmin
(1) Hruby, V. J . Conformational Restriction of Biologically Active
Peptides via Amino Acid Side Chain Groups. Life Sci. 1982, 31,
189-199.
(2) Regoli, D.; Park, W. K.; Rioux, F. Pharmacology of Angiotensin.
Pharmacol. Rev. 1974, 26, 69-123.
(3) Moore, G. J . Angiotensin and Bradykinin Receptors. Membranes
and Receptors; Pergamon Press: Oxford, 1990; Vol. III, pp 961-
980.
(4) Hodges, J . C.; Hamby, J . M.; Blankley, C. J . Angiotensin II
Receptor Binding Inhibitors. Drugs Future 1992, 17, 575-593.
(5) Nikiforovich, G. V.; Marshall, G. R. Three-Dimensional Recogni-
tion Requirements for Angiotensin Agonists: A Novel Solution
for an Old Problem. Biochem. Biophys. Res. Commun. 1993, 195,
222-228.
(6) Nikiforovich, G. V.; Kao, J . L.-F.; Plucinska, K.; Zhang, W. J .;
Marshall, G. R. Conformational Analysis of Two Cyclic Analogs
of Angiotensin: Implication for the Biologically Active Confor-
mation. Biochemistry 1994, 33, 3591-3598.
(7) J oseph, M.-P.; Maigret, B.; Scheraga, H. A. Proposals for the
Angiotensin II Receptor-Bound Conformation by Comparative
Computer Modeling of AII and Cyclic Analogs. Int. J . Pept.
Protein Res. 1995, 46, 514-526.
(8) Samanen, J . M.; Peishoff, C. E.; Keenan, R. M.; Weinstock, J .
Refinement of a Molecular Model of Angiotensin II (AII) Em-
ployed in the Discovery of Potent Nonpeptide Antagonists.
Bioorg. Med. Chem. Lett. 1993, 3, 909-914.
(9) Spear, K. L.; Brown, M. S.; Reinhard, E. J .; McMahon, E. G.;
Olins, G. M.; Palomo, M. A.; Patton, D. R. Conformational
Restriction of Angiotensin II: Cyclic Analogues Having High
Potency. J . Med. Chem. 1990, 33, 1935-1940.
(10) Sugg, E. E.; Dolan, C. A.; Patchett, A. A.; Chang, R. S. L.; Faust,
K. A.; Lotti, V. J . Cyclic Disulfide Analogs of [Sar1,Ile8]-
angiotensin II. In Peptides: Chemistry, Structure and Biology:
Proceedings of the Eleventh American Peptide Symposium;
Rivier, J . E., Marshall, G. R., Eds.; ESCOM Publishers: Leiden,
1990; pp 305-306.
(11) To further evaluate compounds
2 and 3, we studied their
antagonistic activity in two pharmacological in vivo models (for
experimental details, see Supporting Information). The first
model investigated the antihypertensive response in conscious
spontaneously hypertensive rats (SHR), using DuP 753 (losar-
tan) as a reference compound. Compound 2 displayed weak
antihypertensive effects compared to DuP 753. Compound 3,
however, showed a pronounced hypotensive effect even at very
low concentrations (0.001 µmol/kg). Higher concentrations than
0.1 µmol/kg compound 3 could not be given due to adverse
cardiac effects. These effects did not seem to be related to
parasympatic activity as similar results were obtained in rats
pretreated with atropine. In the second model, SHR were
constantly infused with Ang I. No dose dependent hypotensive
effect was seen for compound 2 or 3, even at a dose of 10 µmol/
kg compound 2 and 1 µmol/kg compound 3. DuP 753 investigated
in the latter in vivo model showed a dose dependent hypotensive
effect on the mean arterial blood pressure. The antagonistic/
hypotensive effect that compound 3 displays in one of the in vivo

918 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6
Schmidt et al.
models could be argued as being an effect on the renin angio-
tensin system as compound 3 did not show any dose dependent
hypotensive effect during Ang I infusion. This would rather
support the agonistic profile seen in the rabbit aorta strips
model.
(31) Marshall, G. R. Definition of the Receptor-Bound Conformation
of Angiotensin II. Dtsch. Apoth. Ztg. 1986, 126, 2783-2786.
(32) Duncia, J . V.; Chiu, A. T.; Carini, D. J .; Gregory, G. B.; J ohnson,
A. L.; Price, W. A.; Wells, G. J .; Wong, P. C.; Calabrese, J . C.;
Timmermans, P. B. M. W. M. The Discovery of Potent Nonpep-
tide Angiotensin II Receptor Antagonists: A New Class of Potent
Antihypertensives. J . Med. Chem. 1990, 33, 1312-1329.
(33) Weinstock, J .; Keenan, R. M.; Samanen, J .; Hempel, J .; Finkel-
stein, J . A.; Franz, R. G.; Gaitanopoulos, D. E.; Girard, G. R.;
Gleason, J . G.; Hill, D. T.; Morgan, T. M.; Peishoff, C. E.; Aiyar,
N.; Brooks, D. P.; Fredrickson, T. A.; Ohlstein, E. H.; Ruffolo,
R. R., J r.; Stack, E. J .; Sulpizio, A. C.; Weidley, E. F.; Edwards,
R. M. 1-(Carboxybenzyl)imidazole-5-acrylic Acids: Potent and
Selective Angiotensin II Receptor Antagonists. J . Med. Chem.
1991, 34, 1514-1517.
(12) Maehr, H. A Proposed New Convention for Graphic Presentation
of Molecular Geometry and Topography. J . Chem. Educ. 1985,
62, 114-120.
(13) Kataoka, T.; Beusen, D. D.; Clark, J . D.; Yodo, M.; Marshall, G.
R. The Utility of Side-Chain Cyclization in Determining the
Receptor-Bound Conformation of Peptides: Cyclic Tripeptides
and Angiotensin II. Biopolymers 1992, 32, 1519-1533.
(14) Pramanik, B.; Tsarbopoulos, A.; Labdon, J . E.; Czarniecki, M.;
Nagabhushan, T. L.; Trotta, P. P. Demonstration of a 1-3
Disulfide Bond in a Synthetic Nonapeptide Derived from the
Signal Sequence and N-Terminus of Human γ-Interferon. Bio-
chem. Biophys. Res. Commun. 1988, 157, 836-843.
(34) Iio, H.; Isobe, M.; Kawai, T.; Goto, T. Total Synthesis of
Vernolepin. I. Synthesis of the Key Intermediate. Tetrahedron
1979, 35, 941-948.
(15) Kishore, R.; Balaram, P. Stabilization of γ-turn Conformations
in Peptides by Disulfide Bridging. Biopolymers 1985, 24, 2041-
2043.
(35) Hanessian, S.; Lavallee, P. The Preparation and Synthetic Utility
of tert-Butyldiphenylsilyl Ethers. Can. J . Chem. 1975, 53, 2975-
2977.
(36) Corey, E. J .; Katzenellenbogen, J . A.; Gilman, N. W.; Roman, S.
A.; Erickson, B. W. Stereospecific Total Synthesis of the dl-C18
Cecropia J uvenile Hormone. J . Am. Chem. Soc. 1968, 90, 5618-
5620.
(37) Pappas, J . J .; Keaveney, W. P.; Gancher, E.; Berger, M. A New
and Convenient Method for Converting Olefins to Aldehydes.
Tetrahedron Lett. 1966, 7, 4273-4278.
(38) Borch, R. F.; Bernstein, M. D.; Durst, H. D. The Cyanohydri-
doborate Anion as a Selective Reducing Agent. J . Am. Chem.
Soc. 1971, 93, 2897-2904.
(39) Shioiri, T.; Ninomiya, K.; Yamada, S. Diphenylphosphoryl Azide.
A New Convenient Reagent for a Modified Curtius Reaction and
for the Peptide Synthesis. J . Am. Chem. Soc. 1972, 94, 6203-
6205.
(40) Roush, W. R.; Straub, J . A.; Brown, R. J . Stereochemistry of the
Epoxidations of Acyclic Allylic Amides. Applications Toward the
Synthesis of 2,3,6-Trideoxy-3-aminohexoses. J . Org. Chem. 1987,
52, 5127-5136.
(41) Ramage, R.; Green, J .; Blake, A. J . An Acid Labile Arginine
Derivative for Peptide Synthesis: NG-2,2,5,7,8-Pentamethyl-
chroman-6-sulphonyl-L-arginine. Tetrahedron 1991, 47, 6353-
6370.
(16) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.
Semianalytical Treatment of Solvation for Molecular Mechanics
and Dynamics. J . Am. Chem. Soc. 1990, 112, 6127-6129.
(17) Mohamadi, F.; Richards, N. G. J .; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W.
C. MacroModel - An Integrated Software System for Modeling
Organic and Bioorganic Molecules Using Molecular Mechanics.
J . Comput. Chem. 1990, 11, 440-467.
(18) Two types of γ-turns are possible which we distinguish by the
names classic γ-turn and inverse γ-turn (the mirror image). In
the classic γ-turn conformation, the i + 1 substituent assumes
a pseudoaxial position, while in the inverse γ-turn conformation,
the
i + 1 substituent is pseudoequatorially oriented. The
characteristic torsion angles for Φi+1, Ψi+1 in classic γ-turns are
70-85°, -60-(-70)° and in inverse γ-turns -70-(-85)°, 60-
70°. The characteristic torsion angles for the Φi+1, Ψi+1 and Φi+2
,
Ψi+2 angles of â-turns are as follows: type I, -60°, -30°, -90°,
0°; type II, -60°, 120°, 80°, 0°; type III, -60°, -30°, -60°, -30°;
type I′, 60°, 30°, 90°, 0°; etc.; see: Smith, J . A.; Pease, L. G.
Reverse Turns in Peptides and Proteins. CRC Crit. Rev. Bio-
chem. 1980, 8, 315-399.
(19) Schenkin, P. S.; McDonald, D. Q. Cluster Analysis of Molecular
Conformations. J . Comput. Chem. 1994, 15, 899-916.
(20) Lewis, P. N.; Momany, F. A.; Scheraga, H. A. Chain Reversals
in Proteins. Biochim. Biophys. Acta 1973, 303, 211-229.
(21) Dyson, H. J .; Rance, M.; Houghten, R. A.; Lerner, R. A.; Wright,
P. E. Folding of Immunogenic Peptide Fragments of Proteins in
Water Solution. J . Mol. Biol. 1988, 201, 161-200.
(22) Sybyl: Molecular Modeling System. Tripos Associates, Inc., St.
Louis, MO.
(23) In Sybyl, hydrogen bonds are identified if the minimum and
maximum distance between a hydrogen bonded to a H-bond
donor and acceptor is between 0.8 and 2.8 Å. Another restriction
is that the angle formed by the donor, the hydrogen, and the
acceptor atoms must be larger than 120°.
(24) Ludvigsen, S.; Andersen, K. V.; Poulsen, F. M. Accurate Mea-
surements of Coupling Constants from Two-Dimensional Nuclear
Magnetic Resonance Spectra of Proteins and Determinations of
Φ-Angles. J . Mol. Biol. 1991, 217, 731-736.
(25) Huffman, W. F.; Callahan, J . F.; Eggleston, D. S.; Newlander,
K. A.; Takata, D. T.; Codd, E. E.; Walker, R. F.; Schiller, P. W.;
Lemieux, C.; Wire, W. S.; Burks, T. F. Reverse Turn Mimics. In
Peptides: Chemistry and Biology: Proceedings of the Tenth
American Peptide Symposium; Marshall, G. R., Ed.; ESCOM
Publishers: Leiden, 1988; pp 105-108.
(26) Callahan, J . F.; Newlander, K. A.; Burgess, J . L.; Eggleston, D.
S.; Nichols, A.; Wong, A.; Huffman, W. F. The Use of γ-Turn
Mimetics to Define Peptide Secondary Structure. Tetrahedron
1993, 49, 3479-3488.
(27) Sato, M.; Lee, J . Y. H.; Nakanishi, H.; J ohnson, M. E.; Chrusciel,
R. A.; Kahn, M. Design, Synthesis and Conformational Analysis
of γ-Turn Peptide Mimetics of Bradykinin. Biochem. Biophys.
Res. Commun. 1992, 187, 999-1006.
(28) Huffman, W. F.; Callahan, J . F.; Codd, E. E.; Eggleston, D. S.;
Lemieux, C.; Newlander, K. A.; Schiller, P. W.; Takata, D. T.;
Walker, R. F. Mimics of Secondary Structural Elements of
Peptides and Proteins. In Synthetic Peptides: Approaches to
Biological Problems; Tam, J . P., Kaiser, E. T., Eds.; A. R. Liss:
New York, 1989; Vol. 86, pp 257-266.
(29) Callahan, J . F.; Bean, J . W.; Burgess, J . L.; Eggleston, D. S.;
Hwang, S. M.; Kopple, K. D.; Koster, P. F.; Nichols, A.; Peishoff,
C. E.; Samanen, J . M.; Vasko, J . A.; Wong, A.; Huffman, W. F.
Design and Synthesis of a C7 mimetic for the Predicted γ-Turn
Conformation Found in Several Constrained RGD Antagonists.
J . Med. Chem. 1992, 35, 3970-3972.
(30) Newlander, K. A.; Callahan, J . F.; Moore, M. L.; Tomaszek, T.
A.; Huffman, W. F. A Novel Constrained Reduced-Amide Inhibi-
tor of HIV-1 protease Derived from the Sequential Incorporation
of γ-Turn Mimetics into a Model Substrate. J . Med. Chem. 1993,
36, 2321-2331.
(42) Tam, J . P.; Merrifield, R. B. Strong Acid Deprotection of
Synthetic Peptides: Mechanisms and Methods. In The Pep-
tides: Analysis, Synthesis, Biology. Special Methods in Peptide
Synthesis, Part C; Udenfriend, S., Meienhofer, J ., Eds.; Academic
Press: San Diego, 1987; Vol. 9, pp 185-248.
(43) Anwer, M. K.; Porter, R. A.; Spatola, A. F. Applications of
Ammonium Formate-Catalytic Transfer Hydrogenation. Int. J .
Pept. Protein Res. 1987, 30, 489-497.
(44) Sivanandaiah, K. M.; Gurusiddappa, S. Catalytic Transfer
Hydrogenation of Protected Peptides at Room Temperature. J .
Chem. Res., Synop. 1979, 108-109.
(45) Crivello, J . V. Nitrations and Oxidations with Inorganic Nitrate
Salts in Trifluoroacetic Anhydride. J . Org. Chem. 1981, 46,
3056-3060.
(46) Philippe, N.; Levacher, V.; Dupas, G.; Duflos, J .; Que´guiner, G.;
Bourguignon, J . Diastereoselective Alkylation of Homochiral
1,2,3,4-Tetrahydroisoquinolin-3-one. A Potential Route to Enan-
tiomerically Pure 4-Substituted Tetrahydroisoquinolines. Tetra-
hedron: Asymmetry 1996, 7, 417-420.
(47) Lan, A. J . Y.; Heuckeroth, R. O.; Mariano, P. S. Electron-
Transfer-Induced Photocyclization Reactions of Arene-Iminium
Salt Systems. Effects of Cation Diradical Deprotonation and
Desilylation on the Nature and Efficiencies of Reaction Pathways
Followed. J . Am. Chem. Soc. 1987, 109, 2738-2745.
(48) Ram, S.; Ehrenkaufer, R. E. A General Procedure for Mild and
Rapid Reduction of Aliphatic and Aromatic Nitro Compunds
Using Ammonium Formate as a Catalytic Hydrogen Transfer
Agent. Tetrahedron Lett. 1984, 25, 3415-3418.
(49) Chiu, A. T.; Carini, D. J .; Duncia, J . V.; Leung, K. H.; McCall,
D. E.; Price, W. A., J r.; Wong, P. C.; Smith, R. D.; Wexler, R. R.;
Timmermans, P. B. M. W. M.; Chang, R. S. L.; Lotti, V. J . DuP
532: A Second Generation of Nonpeptide Angiotensin II Recep-
tor Antagonists. Biochem. Biophys. Res. Commun. 1991, 177,
209-217.
(50) Kim, K. S.; Qian, L.; Bird, J . E.; Dickinson, K. E. J .; Moreland,
S.; Schaeffer, T. R.; Waldron, T. L.; Delaney, C. L.; Weller, H.
N.; Miller, A. V. Quinoxaline N-Oxide Containing Potent An-
giotensin II Receptor Antagonists: Synthesis, Biological Proper-
ties, and Structure-Activity Relationships. J . Med. Chem. 1993,
36, 2335-2342.
(51) Whitebread, S.; Mele, M.; Kamber, B.; de Gasparo, M. Prelimi-
nary Biochemical Characterization of Two Angiotensin II Recep-
tor Subtypes. Biochem. Biophys. Res. Commun. 1989, 163, 284-
291.
(52) We searched the Protein Data Bank (PDB) for structures refined
at high resolution containing a classic γ-turn or an inverse γ-turn
conformation. In the crystal structure of rat oncomodulin (PDB

Ang II Receptor Ligands with γ-Turn Mimetics
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 6 919
code name 1RRO) an inverse γ-turn was found between Ile50
and Asn52 and a classic γ-turn was identified in chemotaxis
protein CheY-Escherichia coli (PDB code name 3CHY) between
(60) Barlos, K.; Gatos, D.; Kallitsis, J .; Papaphotiu, G.; Sotiriu, P.;
Wenqing, Y.; Scha¨fer, W. Darstellung Geschu¨tzter Peptid-
Fragmente unter Einsatz Substituierter Triphenylmethyl-Harze.
(Preparation of protected peptide fragments using triphenyl-
methyl resins.) Tetrahedron Lett. 1989, 30, 3943-3946.
(61) Cottet, F.; Cottier, L.; Descotes, G. New Syntheses of Isochro-
mene. Synthesis 1987, 497.
residues Pro61 and Met63
.
(53) Scanlon, M. N.; Matsoukas, J . M.; Franklin, K. J .; Moore, G. J .
A New Approach to Angiotensin Antagonists: Methylation of
the Tyrosine Hydroxyl in Angiotensin II. Life Sci. 1984, 34, 317-
321.
(62) Spangler, R. J .; Kim, J . H. The Pyrolysis of Isochroman-3-one.
A Convenient Synthesis of Benzocyclobutene. Synthesis 1973,
107-108.
(63) Lai, Z.; Zhai, Q.; Yukhananov, R.; Larsson, C.; Roos, P.; Kreek,
M. J .; Nyberg, F. Effects of Morphine on Prolactin Receptors in
the Rat Brain. FEBS Lett. 1994, 338, 207-211.
(64) Piantini, U.; Sørensen, O. W.; Ernst, R. R. Multiple Quantum
Filters for Elucidating NMR Coupling Networks. J . Am. Chem.
Soc. 1982, 104, 6800-6801.
(65) Bothner-By, A. A.; Stephens, R. L.; Lee, J .; Warren, C. D.;
J eanloz, R. W. Structure Determination of a Tetrasaccharide:
Transient Nuclear Overhauser Effects in the Rotating Frame.
J . Am. Chem. Soc. 1984, 106, 811-813.
(66) States, D. J .; Haberkorn, R. A.; Ruben, D. J . A Two-Dimensional
Nuclear Overhauser Experiment with Pure Absorption Phase
in Four Quadrants. J . Magn. Reson. 1983, 48, 286-292.
(67) Wider, G.; Macura, S.; Kumar, A.; Ernst, R. R.; Wu¨thrich, K.
Homonuclear Two-Dimensional 1H NMR of Proteins. Experi-
mental Procedures. J . Magn. Reson. 1984, 56, 207-234.
(68) Goodman, J . M.; Still, W. C. An Unbounded Systematic Search
of Conformational Space. J . Comput. Chem. 1991, 12, 1110-1117.
(54) J oseph, M.-P.; Maigret, B.; Bonnafous, J .-C.; Marie, J .; Scheraga,
H. A. A Computer Modeling Postulated Mechanism for Angio-
tensin II Receptor Activation. J . Protein Chem. 1995, 14, 381-
398.
(55) Printz, M. P.; Nemethy, G.; Bleich, H. Proposed Models for
Angiotensin II in Aqueous Solution and Conclusion about
Receptor Topography. Nature (London), New Biol. 1972, 237,
135-140.
(56) Khosla, M. C.; Smeby, R. R.; Bumpus, F. M. Structure Activity
Relationship in Angiotensin II Analogs. In Handbook of Experi-
mental Pharmacology; Page, I. H., Bumpus, F. M., Eds.; Springer
Verlag: New York, 1974; Vol. 37, pp 126-161.
(57) J orgensen, E. C.; Rapaka, S. R.; Windridge, G. C.; Lee, T. C.
Angiotensin II Analogs. 7. Stereochemical Factors in the 5
Position Influencing Pressor Activity. 2. J . Med. Chem. 1971,
14, 904-906.
(58) Samanen, J .; Cash, T.; Narindray, D.; Brandeis, E.; Adams, W.,
J r.; Weideman, H.; Yellin, T. An Investigation of Angiotensin II
Agonist and Antagonist Analogues with 5,5-Dimethylthiazoli-
dine-4-carboxylic Acid and Other Constrained Amino Acids. J .
Med. Chem. 1991, 34, 3036-3043.
(59) Plucinska, K.; Kataoka, T.; Yodo, M.; Cody, W. L.; He, J . X.;
Humblet, C.; Lu, G. H.; Lunney, E.; Major, T. C.; Panek, R. L.;
Schelkun, P.; Skeean, R.; Marshall, G. R. Multiple Binding
Modes for the Receptor-Bound Conformations of Cyclic AII
Agonists. J . Med. Chem. 1993, 36, 1902-1913.
J M960553D