Comparison of three γ-turn mimetic scaffolds incorporated into angiotensin II

Article abstract of DOI:10.1016/S0968-0896(00)00169-3

Rigidification of peptides by cyclization and iterative incorporation of well-defined secondary structure mimetics constitutes one approach to the design of non-peptidergic structures with better defined conformations. We herein present the synthesis of a potential γ-turn mimetic scaffold, and its incorporation in the 3-5 position of angiotensin II. Two analogues of angiotensin II (Ang II) incorporating this 1,3,5-trisubstituted benzene γ-turn scaffold were synthesized. Evaluation of the compounds in a radioligand binding assay showed that they lacked affinity to the AT₁ receptor. To rationalize these results a geometrical and electrostatical comparison with Ang II analogues encompassing a bicyclic scaffold that delivered inactive pseudo peptides and an azepine scaffold producing highly active ligands was made. This analysis did not provide a clear rationale for the inactivity of the benzene γ-turn scaffolds. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00169-3

Bioorganic & Medicinal Chemistry 8 (2000) 2375±2383
Comparison of Three ꢀ-Turn Mimetic Sca€olds Incorporated
into Angiotensin II
Susanna Lindman,a Gunnar Lindeberg,a Fred Nyberg,b Anders Karlena
and Anders Hallberga,*
aDepartment of Organic Pharmaceutical Chemistry, Uppsala Biomedical Centre, Uppsala University,
Box 574, SE-751 23 Uppsala, Sweden
bDepartment of Biological Research on Drug Dependence, Uppsala Biomedical Centre, Uppsala University,
Box 591, SE-751 24 Uppsala, Sweden
Received 15 May 2000; accepted 26 June 2000
AbstractÐRigidi®cation of peptides by cyclization and iterative incorporation of well-de®ned secondary structure mimetics con-
stitutes one approach to the design of non-peptidergic structures with better de®ned conformations. We herein present the synthesis
of a potential g-turn mimetic sca€old, and its incorporation in the 3±5 position of angiotensin II. Two analogues of angiotensin II
(Ang II) incorporating this 1,3,5-trisubstituted benzene g-turn sca€old were synthesized. Evaluation of the compounds in a radioligand
binding assay showed that they lacked anity to the AT1 receptor. To rationalize these results a geometrical and electrostatical
comparison with Ang II analogues encompassing a bicyclic sca€old that delivered inactive pseudo peptides and an azepine sca€old
producing highly active ligands was made. This analysis did not provide a clear rationale for the inactivity of the benzene g-turn
sca€olds. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
desired properties, subsequent conformational analyses of
the ring systems will guide the selection of a suitable
organic sca€old to substitute and to mimic the secondary
structure motif.2 6
Determination of the bioactive conformation(s) of
important peptides remains an exciting challenge. In
drug discovery processes where such peptides serve as
starting points for design, the ultimate goal is to identify
non-peptidic molecules with high bioavailability and with
retained biological activity. Since no three-dimensional
structural data are available for peptides interacting with
G-protein coupled receptor targets, crucial insights into
the topological requirements within a peptide±receptor
complex have to be gained in an indirect fashion. One
common strategy is to enforce the recognition elements
into the correct region of space by the iterative introduc-
tion of various conformational constraints at selected sites
of the peptide. Monocyclization by disul®de or amide
bond formation constitutes experimentally facile proce-
dures that eciently restrict the ¯exibility of linear pep-
tides.1 These methods should preferably be exploited prior
to the initiation of often-tedious synthetic programs
aimed at more complex organic sca€olds. Thus, after
the identi®cation of a constrained peptide with the
Cyclization between the i and i+2 residues with peni-
cillamine (Fig. 1) has been of particular interest to us for
two reasons. Firstly, incorporation of c[Pen3,5] in angio-
tensin II (Ang II, Asp-Arg-Val-Tyr-Ile-His-Pro-Phe), see
Figure 2, delivered a full agonist with contractile activity
in isolated rabbit aortic strips and with high binding
anity to the AT1 receptor.7,8 Secondly, theoretical
analyses and NMR studies of the 11-membered c[Pen3,5
]
ring system (Fig. 1) suggested a strong preference for the
adoption of an inverse g-turn-like conformation. There-
fore, it seemed that g-turn mimetics7,9 19 should be the
prime candidates for incorporation into peptides where
1,3-penicillamine cyclization a€ords desirable biological
responses.7
We aim to develop g-turn mimetics that are easy to prepare
and that allow introduction of a large variety of side-
chains, including non-natural, from a common precursor.
Ang II, with a suggested turn centered around Tyr-4,8,20 26
was identi®ed as a suitable target peptide for biological
evaluation of diverse turn mimetics.7,27,28 We previously
*Corresponding author. Fax: +46-18-471-4474;
e-mail: anders.hallberg@bmc.uu.se
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00169-3

2376
S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
and almost equipotent with c[Pen3,5]Ang II. Unfortu-
nately, this azepine g-turn mimetic required considerable
e€orts to synthesize.
When an alternative bicyclic g-turn mimetic was intro-
duced instead of amino acid residues 3±5 in
c[Pen3,5]Ang II to give 3 and 4 (Fig. 2), no receptor bind-
ing anity was observed.7 We speculated that the lack of
anity was due to the extra steric bulk of the bicyclic
ring system, which may give rise to a negative steric
interaction with the receptor. To challenge this hypoth-
esis we have now designed and synthesized the geome-
trically smaller and synthetically more accessible 1,3,5-
trisubstituted benzene g-turn sca€old and incorporated
this into Ang II to give 5 and 6. We herein present the
synthesis of these compounds and a comparison of the
benzene sca€old in 5 and 6, the bicyclic sca€old in 3 and
4 and the azepine sca€old in 1 and 2 as g-turn mimetics.
Figure 1. c[Pen3,5] ring system in Ang II adopting a g-turn conformation.
replaced amino acids 3±5 in c[Pen3,5]Ang II with an
azepine g-turn mimetic, giving 1 and 2 (Fig. 2).7 The
synthesized diastereomers exerted similar anity as
c[Pen3,5]Ang II for the AT1 receptor. Interestingly, one
diastereomer was an agonist with full contractile activity
Figure 2. Angiotensin II and analogues incorporating potential g-turn sca€olds.

S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
2377
Chemistry
benzoyl chloride the reaction proceeded smoothly. The
carbonylation32,33 of 9 was conducted at 1 atmosphere
with palladium acetate as precatalyst and dppp34 as
ligand and in the presence of an excess of trimethylsilyl
ethanol to furnish the TMSE ester35 10 in 69% yield.
The cyanation was thereafter performed with palladium
tetrakistriphenylphosphine as catalyst and with zinc
cyanide as additive in DMF according to a procedure
developed by Tschaen,36 and 11 was isolated in 87%
yield. By hydrogenation on Pd/C using the classical
conditions for hydrogenolysis of benzylic ketones,37
debenzylation and reduction of the nitrile and carbonyl
groups were performed and 12 was obtained in a single
step. Fmoc protection of the amine and the phenolic
hydroxyl groups preceded a TFA mediated liberation of
the carboxy group and the building block 14 was
obtained. Subjection to standard solid phase chemistry
provided 5 and 6. In the preparation of Ang II analogue
6, His was protected as the less bulky Boc-derivative
instead of Trt in order to diminish the steric hindrance
during coupling.
The g-turn mimetic sca€olds in compounds 1 and 2 were
prepared essentially according to the procedure outlined
by Hu€man11 13 and relied on Birch reduction, ozonoly-
sis, lactam-cyclization and alkylation as key transfor-
mations.7 The synthesis of the bicyclic sca€old in 3 and
4 were smoothly conducted from 3-isochromanone via
alkylation, lactone-ring opening and cyclization.7 The
preparation of the aromatic sca€old of 5 and 6 is out-
lined in Scheme 1. The commercially available 3-bromo-
5-iodobenzoic acid 7 underwent esteri®cation29 with
phenol and N,N0-dicyclohexylcarbodiimide under stan-
dard conditions to deliver 8. Fries reaction30,31 a€orded
under the conditions applied predominantly p-benzoyl-
ation and after benzylation 9 was isolated. Direct Frie-
del±Crafts benzoylation of benzyl phenyl ether and 7
with polyphosphoric acid, or alternatively with AlCl3
starting from the corresponding acyl chloride, was
initially attempted but met with failure, despite the fact
that in a control experiment with benzoic acid or with
Scheme 1. Reagents: (a) Phenol, DCC, DMAP, Et2O, 73%; (b) (i) AlCl3; (ii) BnBr, K2CO3, acetone, 33%; (c) CO, Pd(OAc)2, dppp, Et3N,
(CH3)3SiCH2CH2OH, DMF, 69%; (d) Zn(CN)2, Pd(PPh3)4, DMF, 87%; (e) H2, Pd/C, HCl, EtOH, EtOAc, 37%; (f) Fmoc-Cl, Et3N, CH2Cl2, 64%;
(g) TFA, CH2Cl2, 81%; (h) (i) Ile-His(Trt)-Pro-Phe-Wang resin, PyBOP, HOBt, DIEA, DMF, (ii) piperidine, DMF; (i) (i) His(Boc)-Pro-Phe-2-
chlorotrityl resin, PyBOP, HOBt, DIEA, DMF; (ii) piperidine, DMF; (j) (i) Fmoc-Arg(Pbf), PyBOP, HOBt, DIEA, DMF, (ii) piperidine, DMF;
(k) (i) Fmoc-Asp(OtBu), PyBOP, HOBt, DIEA, DMF, (ii) piperidine, DMF; (1) TFA:H2O:triethylsilane (90:5:5).

2378
S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
Results and Discussion
Furthermore, in accordance with the design and synth-
esis of the analogues 1 and 27 we substituted Val-3 by
Gly in 5 and 6.
g-Turns are de®ned by three adjoining amino acid resi-
dues forming a pseudo-seven-membered ring. Two types
of g-turns have been de®ned, the inverse g-turn which is
the most common in proteins38,39 and the classic g-turn. In
the inverse g-turn conformation the i+1 substituent
assumes a pseudo equatorial position while in the classic
g-turn conformation the i+1 substituent is pseudo-axially
oriented (Fig. 3).40,41
Compounds 5 and 6 were evaluated in a radioligand
binding assay based on displacement of [125I]-Ang II
from AT1 receptors in rat liver membranes. Disappoint-
ingly, both compounds lacked anity for the AT1
receptor (IC50>10 mM). Ang II displayed an anity
with a Ki-value of 0.9 nM. There may be several reasons
for the lack of binding anity of compounds 5 and 6 as
compared to 1 and 2. We decided to focus on the follow-
ing: (a) they are not geometrical, similar and/or (b) the
electronic character of the benzene sca€old di€ers from
that of the azepine sca€old.
The general framework of the 1,3,5-trisubstituted benzene
sca€old ®ts well within the dimensions de®ned for the
inverse and classic g-turn42 (Fig. 3a and b). The overall
®t is better to the inverse g-turn, since in both the 1,3,5-
trisubstituted benzene sca€old and the inverse g-turn
the i+1 substituent is positioned in a pseudo-equatorial
orientation.
It has already been demonstrated by the use of X-ray
crystallography43 and theoretical calculations9 that there
is a good overall geometrical similarity between the aze-
pine g-turn mimetic incorporated in 1 and 2 and the
inverse g-turn. We therefore used the azepine mimetic as
our common frame of reference. The similarity between
the g-turn mimetic sca€olds compared in this study was
assessed by recording the bond distances and angles of
the AM144 minimized truncated model compounds 15±
18 shown in Figure 4. We focused on the distances and
angles between the three Ca-atoms within the g-turn
moiety. The g-turn mimetic sca€olds in compounds 15
and 16 can adopt both the classic and inverse g-turn,
while the sca€old in 17 can only adopt the inverse g-turn.
In these comparisons we considered only the inverse g-
turn conformation. The measured distances are given
outside and angles inside the triangles in Figure 4.
Further inspection of Figure 3 shows that the atom
types in the C-terminal end of the g-turn and the aro-
matic g-turn mimetic are out of phase. The Cai+2 atom
of the g-turn is superimposed on the carbonyl carbon of
the turn mimetic. This creates a problem concerning what
amino sequence should be coupled to the C-terminal end
of the sca€old in order to mimic Ang II. If Cai+2 of the
mimetic is hypothesized to correspond to the atom after
the benzamide moiety then Ile-His-Pro-Phe should be
used. However, if Cai+2 corresponds to the aromatic
carbon of the benzene ring then His-Pro-Phe should be
used. In the latter case the Ile side-chain is assumed to
be mimicked by a part of the benzene ring. We synthe-
sized both of the alternative compounds, i.e. 5 and 6.
In the comparison between model compounds 15 and 16
the largest deviation is found in the distance from Cai to
Cai+1 and Cai to Cai+2 (Cai in 16 is assumed to corre-
spond to the aromatic carbon of the bicyclic system as
depicted in Figure 4). The other distances and angles
compare well and a likely reason for the inactivity of 3
and 4 is thus the extra steric bulk of the bicyclic system.
In the comparison between the model compounds 15 and
17 it is evident that the g-turn moieties are di€erent. For
example, the Cai to Cai+2 and Cai+1 to Cai+2 distances
are signi®cantly shorter, which is also demonstrated in
Figures 4 and 5d. In 18 the ¯exibility of the C-terminal
side-chain prevents the unambiguous measurement of
distances and angles. However, visual inspection of Fig-
ure 4 and preliminary conformational analysis studies
shows that the Cai to Cai+1 and Cai to Cai+2 distances are
longer in 18 than in 15. Thus, due to the unfavorable shift
in the C-terminal end of the 1,3,5-trisubstituted benzene
sca€old, as compared to the g-turn, a sub-optimal geo-
metrical ®t is obtained, making the g-turn moiety either
too wide or too compact.
Figure 3. (a) Stereo image of the rms best ®t of the 1,3,5-substituted
benzene sca€old and an inverse g-turn (i+1 substituent pseudo-equa-
torial).42 (b) Stereo image of the rms best ®t of the 1,3,5-substituted
benzene sca€old and a classic g-turn (i+1 substituent pseudo-axial).42
The electronic properties of the three g-turn mimetic
sca€olds of the model compounds of 1, 3 and 6 were
also compared using the AM1 optimized geometries
shown in Figure 4. The electrostatic potential was mapped
onto the molecular surface of the model compounds 15,
16 and 17 of the mimetics using MOLCAD within
In both Figure (a) and (b), Ci(O), Ni+1, Cai+1, Ci+1(O), Ni+2, Cai+2
,
Ci+2(O) of the g-turn and the corresponding atoms in the benzene ring
were included in the ®tting procedure. For clarity, the hydrogens are
omitted and only the Cb atom of the side-chains is displayed.

S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
2379
SYBYL.45 As is obvious from Figure 5 there are simi-
larities but also di€erences between the sca€olds studied.
For example, there is an electronegative potential in the
center of the ring systems although this is somewhat less
pronounced in the azepine sca€old. However, taken
together we conclude that a comparison of the elec-
trostatic potential of the molecular surface does not
provide a rationale for the inactivity of 3±6.
Figure 4. Distances and angles between Ca atoms of model compounds of 1 (15), 3 (16) and 6 (17) incorporating potential g-turn sca€olds. Com-
pounds were minimized with the AM1 method.44 Distances are given outside the triangle and angles inside the triangle. Ca atoms are highlighted
with a ®lled circle.
Figure 5. (a) Electrostatic potential mapped onto the Connolly surface of 15, 16 and 17. Negative charge on the surface corresponds to blue colors and
positive charge to red colors. (b) Same molecules as in Figure 5(a) but they have been rotated 180 degrees along the Y-axis. (c) Same as in Figure 5(a)
but only the molecules are shown. (d) Superimposition of molecules 15 (red), 16 (green) and 17 (violet). The Ca atoms were used in the molecular ®t.

2380
S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
Conclusion
Materials. SPPS resins and amino acid derivatives were
obtained from Bachem (Bubendorf, Switzerland), Cal-
biochem-Novabiochem (Laufel®ngen, Switzerland), or
Alexis Corporation (Laufel®ngen, Switzerland). DMF
(peptide synthesis grade) was obtained from Perseptive
Biosystems (Hamburg, Germany) and was used without
further puri®cation. 2-(1H-Benzotriazol-1-yl)-1,1,3,3-
tetramethyluronium hexa¯uorophosphate (HBTU) and
1-hydroxybenzotriazole (HOBt) were purchased from
Richelieu Biotechnologies (St Hyacinthe, QC, Canada).
(Benzotriazol-1-yloxy)tripyrrolidinophosphonium hexa-
¯uorophosphate (PyBOP) was purchased from Calbio-
chem-Novabiochem (Laufel®ngen, Switzerland). 3-
Bromo-5-iodobenzoic acid was obtained from Aldrich.
THF was distilled from sodium/benzophenone before
use. All other reagents were obtained from commercial
sources and used as received.
Previous attempts to substitute amino acids 3±5 in Ang
II with an azepine g-turn mimetic delivered analogues
which had high anity and agonistic activity, but were
laborious to synthesize. When an alternative, slightly
more bulky bicyclic mimetic was introduced in Ang II in
the same position no binding anity was observed. It
was assumed that the lack of anity was attributed to
the extra steric bulk and therefore we incorporated a
smaller and synthetically more accessible 1,3,5-sub-
stituted benzene sca€old in the 3±5 position of Ang II
giving two analogues, which also lacked anity for the
AT1 receptor. This may be due to a sub-optimal simi-
larity of the benzene sca€old with the azepine sca€old
and/or to the shift in amino acid sequence that occurs
between c[Pen3,5]Ang II and compounds 5 and 6. An
alternative explanation may be that the di€erent geo-
metry of the benzene sca€olds compared to the azepine
sca€old prevents the pharmacophore groups from
adopting the bioactive conformation. Taken together,
the results presented here and in previous studies7,8,27,28
indicate that modi®cations of Ang II in the 3±5 region
have a dramatic e€ect on its interaction with the AT1
receptor.
Solid-phase peptide synthesis (SPPS). The starting resins
were synthesized on a 150 mmol scale with a Symphony
instrument (Protein Technologies Inc., Tucson, AZ) using
Fmoc/tert-butyl protection. The side-chain protecting
groups were as follows: Asp(OtBu), Arg(Pbf) and
His(Trt) or His(Boc). Removal of the Fmoc group was
achieved by reaction with 20% piperidine in DMF for
5+10 min. Coupling of the amino acids (200 mmol) was
done in DMF (2.5 mL) using HBTU (200 mmol) in the
presence of N-methylmorpholine, NMM (500 mmol).
Double couplings (2Â30 min) were used for all amino
acids. After the introduction of each amino acid,
remaining amino groups were capped by addition of
20% acetic anhydride in DMF (2.5 mL) to the coupling
mixture and allowing the reaction to proceed for 5 min.
The couplings of the aromatic building block and the
subsequent amino acids (3 equiv) were done manually
with PyBOP (3 equiv) in the presence of HOBt (3 equiv)
and diisopropylethylamine, DIEA (6 equiv), in DMF
(0.7 mL). After completion of the synthesis, the Fmoc
group was removed and the partially protected peptide
resin was washed with several portions of DMF and
CH2Cl2 and dried in a stream of nitrogen and in vacuo.
The peptides were cleaved and deprotected by TFA:H2O:
triethylsilane (90:5:5), precipitated and puri®ed by pre-
parative RP-HPLC. Yields for the puri®ed Ang II ana-
logues were corrected for peptide content.
Experimental
Chemistry
General comments. 1H and 13C NMR spectra were
recorded on a Jeol JNM EX270 spectrometer at 270 and
67.8 MHz respectively. Spectra were recorded at ambi-
ent temperature. Chemical shifts are reported as d
values (ppm), referenced to Me4Si. Low-resolution
electron-impact MS spectra were measured at an ioni-
zation potential of 70 eV. The mass detector was inter-
faced with a gas chromatograph equipped with an HP-1
(25 mÂ0.20 mm) column. Infrared spectra were recor-
ded as solids on a Perkin±Elmer Model 1605 FT-IR
instrument mounted with a Microfocus Beam Con-
denser with ZnSe lenses in a Diasqueeze Plus Diamond
Compressor Cell (Graseby Specac Inc., Smyrna, USA)
and are reported as nmax (cm 1). Melting points
(uncorrected) were determined in open glass capillaries
in a melting point microscope. Elemental analyses were
performed by Mikro Kemi AB, Uppsala, Sweden and
were within 0.4% of the calculated values. Column
chromatography was performed using Merck silica gel
60 (40±63 mm). Thin-layer chromatography (TLC) was
performed 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. Mass spec-
troscopy was carried out on an Applied Biosystems
(Uppsala, Sweden) BIOION 20 plasma desorption mass
spectrometer. Amino acid analyses and peptide content
determinations were performed at the Department of
Biochemistry, Biomedical Centre, Uppsala, Sweden, on
24-h hydrolysates with an LKB 4151 alpha plus analyzer,
using ninhydrin detection.
Phenyl(3-bromo-5-iodo)benzoate (8). Phenol (2.26 g,
24mmol), N,N0-dicyclohexylcarbodiimide (DCC) (4.95g,
24mmol) and 4-dimethylaminopyridine (0.24 g, 2 mmol)
were added to a solution of 3-bromo-5-iodobenzoic acid
7 (6.54 g, 20 mmol) in diethyl ether (250 mL) at 0 ꢀC.
The reaction mixture was allowed to reach rt overnight.
The N,N0-dicyclohexyl urea was ®ltered o€ and the ®l-
trate was washed with water (3Â180 mL), 5% aqueous
acetic acid (3Â180 mL), water (3Â180 mL), dried
(MgSO4), ®ltered, concentrated and puri®ed by column
chromatography (isohexane:EtOAc, 9:1, to EtOAc) to
give 8 (5.86 g, 73%) TLC Rf 0.61 (isohexane:EtOAc,
ꢀ
1
9:1); mp 79±80 C; H NMR (CDCl3) d 7.15±7.23 (m,
2H, Ar), 7.25±7.34 (m, 1H, Ar), 7.39±7.49 (m, 2H, Ar),
8.11 (dd, 1H, Ar), 8.29 (dd, 1H, Ar), 8.45 (dd, 1H, Ar);
13C NMR (CDCl3) d 94.1, 121.4, 123.2, 126.3, 129.6,
132.4, 132.7, 137.6, 144.4, 150.4, 162.4; IR (solid)

S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
2381
1725 cm 1; MS [IP 70 eV; m/z (% rel int)] 404 (M++2,
18), 402 (M+, 18), 311 (100), 309 (100). Anal. (C13H8
BrIO2) C, H.
[2-(Trimethylsilyl)ethyl]3-(40-benzyloxy)benzoyl-5-cyano-
benzoate (11). Zn(CN)2 (31.3 mg, 0.267 mmol) and
Pd(PPh3)4 (26.4 mg, 0.023 mmol) were mixed in DMF
(1 mL) under a stream of nitrogen. Compound 10
(195 mg, 0.381 mmol) dissolved in DMF (1 mL) was
added to the mixture. The heavy-walled Pyrex tube was
sealed with a screw cap ®tted with a Te¯on gasket, and the
reaction mixture was stirred at 80 ꢀC for 6 h. After cool-
ing, the black mixture was diluted with EtOAc (20 mL),
washed with aqueous saturated Na2CO3 (3Â15mL) and
brine (3Â15 mL), dried (MgSO4), ®ltered, concentrated
and puri®ed by column chromatography (isohexane:
EtOAc, 9:1) to give 11 as a solid (138 mg, 87%). TLC Rf
0.28 (isohexane:EtOAc, 9:1); mp 73±75 ꢀC; 1H NMR
(CDCl3) d 0.09 (s, 9H, Me3), 1.16 (m, 2H, CH2Si), 4.47
(m, 2H, OCH2), 5.18 (s, 2H, OCH2Ph), 7.08 (dm,
J=9.1 Hz, 2H, 30), 7.25±7.66 (m, 5H, Bn), 7.79 (dm,
J=9.1 Hz, 2H, 20), 8.18 (dd, 1H, Ar), 8.48 (dd, 1H, Ar),
8.58 (dd, 1H, Ar); 13C NMR (CDCl3) d 1.5, 17.5, 64.6,
70.3, 113.2, 114.9, 117.2, 127.4, 128.4, 128.68, 128.71,
132.2, 132.6, 134.0, 135.5, 135.8, 136.4, 139.8, 163.2,
164.1, 192.1; IR (solid) 2238, 1713, 1655 cm 1. Anal.
(C27H27NO4Si) C, H, N.
3-Bromo-5-iodo-40-benzyloxybenzophenone (9). Phenyl
(3-bromo-5-iodo)benzoate 8 (3.19 g, 7.91 mmol) and
AlCl3 (2.11 g, 15.8 mmol) were vigorously stirred and
heated in a preheated oil bath (145 ꢀC). The reaction
mixture melted and a white fume developed. When a red
fume started to develop after about 30 min the reaction
was stopped and the reaction mixture was allowed to
reach rt. The solid was dissolved in EtOAc (50 mL) and
1 M HCl (50 mL). The layers were separated and the
aqueous layer was extracted with EtOAc (2Â30 mL).
The combined organic layers were dried (MgSO4), ®l-
tered and concentrated. The crude product was ®ltered
through silica (isohexane:EtOAc, 9:1, to EtOAc) to give
a residue (1.05 g), which was dissolved with acetone
(25 mL) and K2CO3 (0.38 g, 2.8 mmol) and benzyl
bromide (0.52 g, 3.04 mmol) was added. The reaction
mixture was heated to re¯ux. After 5 h the reaction mix-
ture was cooled and diluted with ether (50 mL). The
organic layer was washed with H2O (50 mL), 2 M NaOH
(3Â50 mL), dried (MgSO4), ®ltered, concentrated and
puri®ed by column chromatography (isohexane:EtOAc,
9:1, to EtOAc) to give 9 as a solid (1.10 g, 28%). TLC Rf
[2 - (Trimethylsilyl)ethyl]3 - aminomethyl - 5 - (40 - hydroxy-
benzyl)benzoate (12). A mixture of compound 11
(150mg, 0.546 mmol) and 10% Pd(C) (50 mg, 0.047 mmol
Pd) in absolute ethanol (10 mL), EtOAc (4 mL) and 3 M
HCl (10 mL) was stirred under H2 (1 atm) overnight.
The mixture was adjusted to pH 8 using aqueous satu-
rated NaHCO3 and concentrated. The residue was dis-
solved in CH2Cl2:EtOH (95:5), ®ltered through a pad of
Celite and concentrated. The residue was dissolved in
water (10 mL) and extracted with EtOAc (3Â10 mL).
The combined organic layers were dried (Na2SO4), ®l-
tered, concentrated and puri®ed by column chromato-
graphy (CHCl3:MeOH, 9:1, to CHCl3:MeOH, 1:1) to
give 12 as a solid (73 mg,ꢀ37%). TLC Rf 0.31 (CHCl3:
0.50 (isohexane:EtOAc, 9:1); mp 104±106 ꢀC; H NMR
1
(CDCl3) d 5.16 (s, 2H, OCH2Ph), 7.06 (dm, J=8.9 Hz,
2H, 30), 7.34±7.47 (m, 5H, BnAr), 7.76 (dd, 1H, Ar),
7.80 (dm, J=8.9 Hz, 2H, 20), 7.97 (dd, 1H, Ar), 8.04
(dd, 1H, Ar); 13C NMR (CDCl3) d 70.2, 94.2, 114.7,
122.9, 127.4, 128.3, 128.7, 129.0, 131.7, 132.5, 135.9,
1
136.8, 141.4, 142.5, 162.9, 192.0; IR (solid) 1648 cm
;
MS [IP 70 eV; m/z (% rel int)] 494 (M++2, 4), 492
(M+, 4), 91 (100). Anal. (C20H14BrIO2) C, H.
[2-(Trimethylsilyl)ethyl] 3-(40 -benzyloxy)benzoyl-5-bro-
mobenzoate (10). Pd(OAc)2 (14.6 mg, 0.065 mmol) and
dppp (26.9 mg, 0.065 mmol) were mixed in DMF (2 mL)
under a stream of nitrogen. The yellow mixture was stir-
red for 10min before the addition of a solution of 9
(1.07 g, 2.17mmol) in DMF (8 mL), triethylamine
(0.36 mL, 2.6 mmol) and trimethylsilylethanol (2.56 g,
21.7 mmol). The reaction mixture was stirred and CO
was bubbled through the solution at rt for 20 min before
the dark brown±red solution was heated to 85 ꢀC. After
5 h CO-bubbling, the mixture was placed under 1 atm of
CO. After 2 days the reaction mixture was cooled and
concentrated. The residue was dissolved in ether
(50 mL), washed with brine (2Â40 mL), dried (MgSO4),
®ltered, concentrated and puri®ed by column chroma-
tography (isohexane:EtOAc, 9:1) to give 10 (0.75 g,
1
MeOH, 1:1); mp 111±113 C; H NMR (CDCl3) d 0.07
(s, 9H, Me3), 1.11 (m, 2H, CH2Si), 3.75±3.89 (br s, 2H,
NH2), 3.85 (s, 2H, CH2NH2), 3.87 (s, 2H, CH2Ph), 4.38
(m, 2H, OCH2), 6.68 (dm, J=8.5 Hz, 2H, 30), 6.96 (dm,
J=8.5 Hz, 2H, 20), 7.29 (dd, 1H, Ar), 7.77 (m, 2H,
2ÂAr); 13C NMR (CDCl3) d 1.5, 17.4, 40.8, 45.6, 63.4,
115.7, 126.1, 128.7, 129.9, 131.0, 131.6, 132.3, 141.9,
142.7, 155.0, 166.8; IR (solid) 3354, 3286, 3500±2300,
1713 cm 1. Anal. (C20H27NO3Si) C, H, N.
[2 - (Trimethylsilyl)ethyl]3 - [(9 - ¯uorenylmethyloxycarbo-
nyl)amino]methyl-5-[40(9-¯uorenylmethyloxycarbonyloxy)-
benzyl]benzoate (13). Compound 12 (50 mg, 0.140 mmol)
and Et3N (21.2 mg, 0.210 mmol) were dissolved in
CH2Cl2 (3 mL) and cooled to 0 ꢀC. Fmoc-Cl (90.5 mg,
0.350 mmol), dissolved in CH2Cl2 (2 mL), was added
dropwise, whereafter the reaction mixture was allowed
to reach rt under continued stirring overnight. The mix-
ture was concentrated and puri®ed by column chroma-
tography (isohexane:EtOAc, 3:1, to isohexane:EtOAc,
2:1) to give 13 as amorphous crystals (72 mg, 64%). TLC
Rf 0.45 (isohexane:EtOAc, 2:1); 1H NMR (CDCl3) d 0.06
(s, 9H, Me3), 1.10 (m, 2H, CH2Si), 3.99 (s, 2H, CH2Ph),
4.22 (t, J=7.0 Hz, 1H, CH Fmoc-N), 4.27±4.46 (m, 7H,
CH2O, CH2 Fmoc, CH2N, CH Fmoc-O), 4.47±4.52 (m,
1
69%). TLC Rf 0.36 (isohexane:EtOAc, 9:1); H NMR
(CDCl3) d 0.08 (s, 9H, Me3), 1.14 (m, 2H, CH2Si), 4.44
(m, 2H, OCH2), 5.17 (s, 2H, OCH2Ph), 7.06 (dm,
J=8.9 Hz, 2H, 30), 7.32±7.48 (m, 5H, Bn), 7.80 (dm,
J=8.9 Hz, 2H, 20), 8.06 (dd, 1H, Ar), 8.28 (dd, 1H, Ar),
8.35 (dd, 1H, Ar); 13C NMR (CDCl3) d 1.5, 17.4, 64.1,
70.2, 114.7, 122.6, 127.4, 128.3, 128.7, 129.0, 129.2,
132.55, 132.56, 135.4, 136.0, 136.2, 140.3, 162.9, 164.7,
192.9; IR (solid) 1722, 1658 cm 1; MS [IP 70 eV; m/z (%
rel int)] (M+ missing) 484/482 (M+ 28, 1:1), 91 (100),
73 (8). Anal. (C26H27BrO4Si) C, H.

2382
S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
2H, CH2 Fmoc), 5.13 (br t, 1H, NH), 7.06±7.11 (m, 2H, 30),
7.15±7.20 (m, 2H, 20), 7.25±7.46 (m, 9H, CH Ar Fmoc-O,
CH Ar Fmoc-N, Ar(1H)), 7.55±7.65 (m, 4H, CH Ar Fmoc-
O, CH Ar Fmoc-N), 7.72±7.84 (m, 6H, CH Ar Fmoc-O,
CH Ar Fmoc-N, Ar(2H)); 13C NMR (CDCl3) d 1.5,
17.4, 41.0, 44.6, 46.7, 47.2, 63.4, 66.8, 70.4, 119.9, 120.1,
121.1, 125.0, 125.1, 126.4, 127.0, 127.1, 127.6, 127.9,
129.1, 129.8, 131.4, 132.4, 138.2, 139.1, 141.26, 141.29,
141.5, 143.1, 143.8, 149.6, 153.6, 156.3, 166.4; IR (solid)
3354, 1761, 1715 cm 1. Anal. (C50 H47NO7Si) C, H, N.
was monitored at 230 nm and selected fractions were
analyzed by analytical RP-HPLC (220 nm) and by
PDMS. One compound of the expected mass was iso-
lated. The ®nal yield of 5 was 24.6 mg (59%). Amino
acid analysis: Asp 1.00, Arg 1.01, Ile 1.01, His 1.02, Pro
0.94, Phe 1.03 (73% peptide); PDMS (MW 1023.2):
1024.7 (M+H+).
Ang II analogue 6. The peptide was synthesized accord-
ing to the procedure used for 5, but with His(Boc)-Pro-
Phe-2-chlorotrityl-resin (123 mg, 38 mmol) as the start-
ing polymer. The partially protected peptide resin
(133.5 mg) was cleaved as described above to yield 38 mg
of crude peptide. The peptide was dissolved in 0.1%
aqueous TFA (12 mL) and puri®ed in two runs by pre-
parative RP-HPLC on a Vydac 10-mm C18 column
(2.2Â25 cm) as described above. One compound of the
expected mass was isolated. The ®nal yield of 6 was
21.5 mg (57%). Amino acid analysis: Asp 1.00, Arg
1.01, His 1.01, Pro 0.98, Phe 1.01 (72% peptide); PDMS
(MW 910.0): 910.3 (M+H+).
3-[(9-Fluorenylmethyloxycarbonyl)amino]methyl-5-[40(9-
¯uorenylmethyloxycarbonyloxy)benzyl]benzoic acid (14).
Compound 13 (59 mg, 0.073 mmol) was treated with
50% TFA in CH2Cl2 (4 mL) at room temperature for
2 h. The reaction mixture was concentrated and puri®ed
by column chromatography (CHCl3 to CHCl3:MeOH,
9:1) to give compound 14 as a solid (41.8 mg, 81%). TLC
ꢀ
1
Rf 0.43 (CHCl3:MeOH, 9:1); mp 108±110 C; H NMR
(CDCl3) d 3.97 (s, 2H, CH2Ph), 4.19 (br t, 1H, CH Fmoc-
N), 4.24±4.52 (m, 7H, CH Fmoc-O, 2ÂCH2 Fmoc,
CH2N), 5.17 (br s, 1H, NH), 7.02±7.09 (m, 2H, 30), 7.11±
7.18 (m, 2H, 20), 7.25±7.44 (m, 9H, CH Ar Fmoc-O, CH
Ar Fmoc-N, Ar(1H)), 7.54±7.63 (m, 4H, CH Ar Fmoc-
O, CH Ar Fmoc-N), 7.71±7.78 (m, 4H, CH Ar Fmoc-O,
CH Ar Fmoc-N), 7.85 (m, 2H, 2ÂAr); 13C NMR
(CDCl3) d 40.9, 44.5, 46.7, 47.1, 66.7, 70.4, 119.2, 120.1,
121.2, 125.0, 125.1, 126.6, 127.0, 127.2, 127.7, 127.9,
129.79, 129.81, 131.0, 133.2, 137.9, 139.4, 141.29,
141.31, 141.6, 143.1, 144.1, 149.7, 153.6, 156.6, 170.7;
IR (solid) 3600±2400, 3448, 1757, 1720 cm 1. Anal.
Rat liver membrane AT1 receptor binding assay. Rat
liver membranes were prepared according to the method
of Dudley et al.46 Binding of [125I]-Ang II to membranes
was conducted in a ®nal volume of 0.5 mL of 50 mM
Tris±HCl (pH 7.4), supplemented with 100 mM NaCl,
10 mM MgCl2, 1 mM EDTA, 0.025% bacitracin and
0.2% BSA, and containing liver homogenate corre-
sponding to 5 mg of the original tissue weight, [125I]-
Ang II (70 000 cpm, 0.03 nM), and variable concentra-
tions of test substance. Samples were incubated at
25 ꢀC for 1 h, and binding was terminated by ®ltration
through Whatman GF/B glass-®ber ®lter sheets, using
a Brandel cell harvester. The ®lters were washed with
4Â2 mL of Tris±HCl (pH 7.4) and transferred to tubes.
The radioactivity was measured in a gamma counter.
Non-speci®c binding was determined in the presence of
10 mM saralasin. All experiments were performed in
duplicate.
(C45H35NO7 1H2O) C, H, N.
.
2
Ang II analogue 5. The Fmoc-protected building block
14 (20.9 mg, 29.8 mmol), PyBOP (45.4 mg, 87.2 mmol),
HOBt (129.5 mg, 218 mmol) and DIEA (37 mL,
218 mmol) were dissolved in DMF (0.5 mL) and added
to preswollen Ile-His(Trt)-Pro-Phe-Wang resin (75 mg,
38 mmol) (see SPPS) in a 2 mL disposable syringe
equipped with a porous polyethylene ®lter. After mixing
by slow rotation at rt for 20 h the resin was ®ltered o€
and washed with DMF (5Â1 mL), deprotected with
20% piperidine in DMF (3Â1 mL, 1+5+10 min) and
washed with DMF (6Â1 mL). The peptide was further
elongated by reaction with, in turn, Fmoc-Arg(Pbf)-OH
(19 h) and Fmoc-Asp(OtBu)-OH (1.5 h) as described
above (SPPS). Fmoc deprotection, washing with DMF,
and drying in air and in vacuo a€orded the partially
protected peptide resin (100.9 mg). The resin was treated
with TFA:H2O:triethylsilane (90:5:5) (1 mL) for 1.5 h
and the mixture was ®ltered through a small plug of
glass wool in a Pasteur pipette. After washing with TFA
(3Â0.3 mL), the product was precipitated by the addi-
tion of cold, anhydrous ether (13.5 mL). The precipitate
was collected by centrifugation, washed with ether
(4Â4.5 mL) and dried to furnish 50.5 mg of crude
peptide.
Theoretical calculations. The minimizations of 15±18
were performed using the AM144 method as imple-
mented in SYBYL45 using the keywords PRECISE,
XYZ and NOMM. Graphical manipulations were
carried out within SYBYL 6.6. The electrostatic poten-
tial was mapped onto the calculated Connolly surfaces
(MOLCAD module). Default parameters were used. To
force the electrostatic potential colors of each molecule
into common borders, the option global was selected.
Acknowledgements
We would like to thank Barbro Synnergren for skillful in
vitro binding experiments, and the Swedish Foundation
for Strategic Research (SSF) and AstraZeneca Molndal
for ®nancial support.
The peptide was dissolved in 0.1% aqueous TFA
(12 mL) and puri®ed in two runs by preparative RP-
HPLC on a Vydac 10-mm C18 column (2.2Â25 cm)
using a 120 min gradient of 20±50% CH3CN in 0.1%
aqueous TFA at a ¯ow rate of 4 mL/min. The separation
References and Notes
1. Hruby, V. J. Life Sci. 1982, 31, 189.

S. Lindman et al. / Bioorg. Med. Chem. 8 (2000) 2375±2383
2383
2. Giannis, A.; Kolter, T. Angew. Chem. Int. Ed. Engl. 1993,
32, 1244.
26. Sugg, E. E.; Dolan, C. A.; Patchett, A. A.; Chang, R. S.
L.; Faust, K. A.; Lotti, V. J. In Peptides: Chemistry, Structure
and Biology: Proceedings of the Eleventh American Peptide
Symposium; Rivier, J. E.; Marshall, G. R., Eds.; ESCOM
Science Publishers: Leiden, 1990, pp 305±306.
27. Johannesson, P.; Lindeberg, G.; Tong, W.; Gogoll, A.;
Karlen, A.; Hallberg, A. J. Med. Chem. 1999, 42, 601.
28. Johannesson, P.; Lindeberg, G.; Tong, W.; Gogoll, A.;
Synnergren, B.; Nyberg, F.; Karlen, A.; Hallberg, A. J. Med.
Chem. 1999, 42, 4524.
3. Holzemann, G. Kontakte (Darmstadt) 1991, 3.
4. Holzemann, G. Kontakte (Darmstadt) 1991, 55.
5. Liskamp, R. M. J. Recl. Trav. Chim. Pays-Bas 1994, 113, 1.
6. Fairlie, D. P.; West, M. L.; Wong, A. K. Curr. Med. Chem.
1998, 5, 29.
7. Schmidt, B.; Lindman, S.; Tong, W.; Lindeberg, G.;
Gogoll, A.; Lai, Z.; Thornwall, M.; Synnergren, B.; Nilsson,
A.; Welch, C. J.; Sohtell, M.; Westerlund, C.; Nyberg, F.;
Karlen, A.; Hallberg, A. J. Med. Chem. 1997, 40, 903.
8. Spear, K. L.; Brown, M. S.; Reinhard, E. J.; McMahon, E.
G.; Olins, G. M.; Palomo, M. A.; Patton, D. R. J. Med. Chem.
1990, 33, 1935.
9. Alkorta, I.; Suarez, M. L.; Herranz, R.; Gonzalez-Muniz,
R.; Garcõa-Lopez, M. T. J. Mol. Model. 1996, 2, 16.
10. Brickmann, K.; Yuan, Z.; Sethson, I.; Somfai, P.; Kihl-
berg, J. Chem. Eur. J. 1999, 5, 2241.
11. Callahan, J. F.; Bean, J. W.; Burgess, J. L.; Eggleston, D.
S.; Hwang, S. M.; Kopple, K. D.; Koster, P. F.; Nichols, A.;
Peisho€, C. E.; Samanen, J. M.; Vasko, J. A.; Wong, A.;
Hu€man, W. F. J. Med. Chem. 1992, 35, 3970.
12. Callahan, J. F.; Newlander, K. A.; Burgess, J. L.; Eggle-
ston, D. S.; Nichols, A.; Wong, A.; Hu€man, W. F. Tetra-
hedron 1993, 49, 3479.
13. Newlander, K. A.; Callahan, J. F.; Moore, M. L.; Tomas-
zek, T. A.; Hu€man, W. F. J. Med. Chem. 1993, 36, 2321.
14. Ferguson, M. D.; Meara, J. P.; Nakanishi, H.; Lee, M. S.;
Kahn, M. Tetrahedron Lett. 1997, 38, 6961.
15. Graf von Roedern, E.; Lohof, E.; Hessler, G.; Ho€mann,
M.; Kessler, H. J. Am. Chem. Soc. 1996, 118, 10156.
16. Kemp, D. S.; Carter, J. S. J. Org. Chem. 1989, 54, 109.
17. Olson, G. L.; Cheung, H.-C.; Chiang, E.; Madison, V. S.;
Sepinwall, J.; Vincent, G. P.; Winokur, A.; Gary, K. A. J.
Med. Chem. 1995, 38, 2866.
29. Hassner, A.; Alexanian, V. Tetrahedron Lett. 1978, 46,
4475.
30. Blatt, A. H. In Organic Reactions; Adams, R., Ed.; Wiley
& Sons, Inc.: New York, 1942; Vol. 1, pp 342±369.
31. Gerecs, A. In Friedel Crafts and Related Reactions; Olah,
G. A., Ed.; Wiley & Sons, Inc.: New York, 1964; Vol. 3, pp
499±533.
32. Hidai, M.; Hikita, T.; Wada, Y.; Fujikura, Y.; Uchida, Y.
Bull. Chem. Soc. Jpn. 1975, 48, 2075.
33. Schoenberg, A.; Bartoletti, I.; Heck, R. F. J. Org. Chem.
1974, 39, 3318.
34. dppp=1,3-bis(Diphenylphosphino)propane.
35. Sieber, P. Helv. Chim. Acta 1977, 60, 2711.
36. Tschaen, D. M.; Desmond, R.; King, A. O.; Fortin, M.
C.; Pipik, B.; King, S.; Verhoeven, T. R. Synth. Commun.
1994, 24, 887.
37. Augustine, R. L. Catalytic Hydrogenation. Marcel Dekker:
New York, 1965, p 83.
38. Milner-White, E. J.; Ross, B. M.; Ismail, R.; Belhadj-
Mostefa, K.; Poet, R. J. Mol. Biol. 1988, 204, 777.
39. Milner-White, E. J. J. Mol. Biol. 1990, 216, 385.
40. Rose, G. D.; Gierasch, L. M.; Smith, J. A. Adv. Protein
Chem. 1985, 37, 1.
41. Smith, J. A.; Pease, L. G. CRC Crit. Rev. Biochem. 1980,
8, 315.
18. Sato, M.; Lee, J. Y. H.; Nakanishi, H.; Johnson, M. E.;
Chrusciel, R. A.; Kahn, M. Biochem. Biophys. Res. Commun.
1992, 187, 999.
19. Schmidt, B.; Kuhn, C. Synlett 1998, 1240
20. Printz, M. P.; Nemethy, G.; Bleich, H. Nature (London),
New Biol. 1972, 237, 135.
42. The inverse g-turn was obtained from the crystal structure
of rat oncomodulin (PDB code name 1RRO) between Ile-50
and Asn-52. The classic g-turn was obtained from chemotaxis
protein Che-Y-Escherichia coli (PDB code name 3CHY)
between residues Pro-61 and Met-63.
43. Hu€man, 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. In Synthetic Peptides: Approaches to
Biological Problems; Tam, J. P.; Kaiser, E. T., Eds.; Alan R.
Liss, Inc.: New York, 1989; Vol. 86, pp 257±266.
44. Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J.
J. P. J. Am. Chem. Soc. 1985, 107, 3902.
21. Nikiforovich, G. V.; Marshall, G. R. Biochem. Biophys.
Res. Commun. 1993, 195, 222.
22. Nikiforovich, G. V.; Kao, J. L.-F.; Plucinska, K.; Zhang,
W. J.; Marshall, G. R. Biochemistry 1994, 33, 3591.
23. 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. J.
Med. Chem. 1993, 36, 1902.
45. SYBYL Molecular Modeling Software, Version 6.6; Tripos
Associates Inc.: St Louis, MO 63144.
24. Samanen, J. M.; Peisho€, C. E.; Keenan, R. M.; Wein-
stock, J. Bioorg. Med. Chem. Lett. 1993, 3, 309.
25. Joseph, M.-P.; Maigret, B.; Scheraga, H. A. Int. J. Pept.
Protein Res. 1995, 46, 514.
46. Dudley, D. T.; Panek, R. L.; Major, T. C.; Lu, G. H.;
Bruns, R. F.; Klinkefus, B. A.; Hodges, J. C.; Weishaar, R. E.
Mol. Pharmacol. 1990, 38, 370.