Design, synthesis and biological evaluation of cyclic angiotensin II analogues with 3,5 side-chain bridges: Role of C-terminal aromatic residue and ring cluster for activity and implications in the drug design of AT1 non-peptide antagonists

Article abstract of DOI:10.1016/S0960-894X(02)00474-2

The novel amide linked Angiotensin II (ANG II) cyclic analogues: γ,ε-cyclo(3, 5)-[Sar¹-Glu³-Lys⁵-Ile⁸] ANG II (I) and γ, ε-cyclo(3, 5)-[Sar¹-Glu³-Lys⁵-Phe⁸] ANG II (II) have been designed, synthesized and bioassayed in anesthetized rabbits in order to unravel structural ring cluster characteristics important for receptor activation. Analogue I with Ile at position 8 was an inhibitor of Angiotensin II while analogue II with Phe at position 8 was found to be an agonist. Similar results were reported for cyclic compounds that have reversed the linking between positions 3 and 5. The overall results show that positions 3 and 5 do not govern the biological activity of the synthetic analogues. It also appears that the aromatic ring cluster (Tyr-His-Phe) in agonist peptides is an essential stereo-electronic feature for Angiotensin II to exert its biological activity. A non-peptide mimetic of ANG II, 1-[2′-[(N-benzyl)tetrazol-5-yl]biphenyl-4-yl]methyl]-2- hydroxymethylbenzimidazole (BZI8) has been designed and synthesized. This molecule is more rigid and much less active than AT₁ non-peptide mimetic losartan probably because it lacks to mimic the orientation of tetrazole and the pharmacophore segments of butyl chain and imidazole ring.

Full text of DOI:10.1016/S0960-894X(02)00474-2

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2627–2633
Design, Synthesis and Biological Evaluation of Cyclic Angiotensin
II Analogues with 3,5 Side-Chain Bridges:
Role of C-Terminal Aromatic Residue and Ring Cluster for
Activity and Implications in the Drug Design of AT1
Non-peptide Antagonists
Panagiota Roumelioti,a Ludmila Polevaya,b Panagiotis Zoumpoulakis,c
Nektarios Giatas,a Ilze Mutule,b Tatjana Keivish,b Anastasia Zoga,c
Demetrios Vlahakos,d E. Iliodromitis,d Demetrios Kremastinos,d
Simona Golic Grdadolnik,e Thomas Mavromoustakos,b,*
and John Matsoukasa
aDepartment of Chemistry, University of Patras, 26500 Patras, Greece
bLaboratory of Peptide Chemistry, Latvian Institute of Organic Synthesis, Riga, LV-1006, Latvia
cInstitute of Organic and Pharmaceutical Chemistry, National Hellenic Research Foundation, Athens, 11635, Greece
dOnassis Cardiac Surgery Center, 17674 Athens, Greece
eNational Institute of Chemistry, Hajdrihova 19, POB 30 SI-1115 Ljubljana, Slovenia
Received 29 January 2002; revised 17 May 2002; accepted 17 June 2002
Abstract—The novel amide linked Angiotensin II (ANG II) cyclic analogues: g,e-cyclo(3, 5)-[Sar1-Glu3-Lys5-Ile8] ANG II (I) and g,
e-cyclo(3, 5)-[Sar1-Glu3-Lys5-Phe8] ANG II (II) have been designed, synthesized and bioassayed in anesthetized rabbits in order to
unravel structural ring cluster characteristics important for receptor activation. Analogue I with Ile at position 8 was an inhibitor of
Angiotensin II while analogue II with Phe at position 8 was found to be an agonist. Similar results were reported for cyclic com-
pounds that have reversed the linking between positions 3 and 5. The overall results show that positions 3 and 5 do not govern the
biological activity of the synthetic analogues. It also appears that the aromatic ring cluster (Tyr-His-Phe) in agonist peptides is an
essential stereo-electronic feature for Angiotensin II to exert its biological activity. A non-peptide mimetic of ANG II, 1-[20-[(N-
benzyl)tetrazol-5-yl]biphenyl-4-yl]methyl]-2-hydroxymethylbenzimidazole (BZI8) has been designed and synthesized. This molecule
is more rigid and much less active than AT1 non-peptide mimetic losartan probably because it lacks to mimic the orientation of
tetrazole and the pharmacophore segments of butyl chain and imidazole ring. # 2002 Published by Elsevier Science Ltd.
Introduction
carboxylate, analogous to that found in serine pro-
teases,3 as well as a ring cluster of the triad key ami-
noacids Tyr4-His6- Phe8 which appears to be responsible
for agonist activity.4,5 Thus, conformational analysis
using modern 2D NMR techniques in receptor-simulat-
ing environments has shown proximity of the three key
aminoacids sidechains and the formation of tyrosinate
has been demonstrated by nanosecond time resolved
fluorescence studies.6,7 Comparative nuclear magnetic
resonance studies of the backbone structure between
peptide agonists and antagonists have shown that only
agonists display ring clustering and form a charge relay
The octapeptide Angiotensin II (ANG II, Asp-Arg-Val-
Tyr-Ile-His-Pro-Phe) is the main pressor component of
the Renin-Angiotensin System (RAS).1 Accumulated
experimental evidence for Angiotensin II supports a
bioactive conformation characterized by a charge relay
system between Tyr hydroxyl, His imidazole and Phe
*Corresponding author. Tel.: +30-1-727-3869; fax: +30-1-727-3831;
e-mail: tmavro@eie.gr
0960-894X/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.
PII: S0960-894X(02)00474-2

2628
P. Roumelioti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2627–2633
system. In addition, the proposed conformation of pep-
tide antagonists overlay the recently discovered non-
peptide ANG II receptor antagonist losartan and ana-
logues when molecular modeling techniques and super-
imposition studies are applied.8 Furthermore, the ring
cluster conformation was recently supported by the
design and synthesis of a novel constrained ANG II
cyclic analogue, [Sar1-Lys3-Glu5] ANG II, which pos-
sesses biological activity when tested in the rat uterus
assay and in anesthesized rabbits.5 This potent cyclic
analogue was designed to have as a major molecular
feature the integrity of the ring cluster. Other structure–
activity studies have illustrated the importance of the
C-terminal aromatic residue Phe for agonist activity.
Replacement of residue Phe at position 8 with an ali-
phatic one, as Ile, results in an antagonist [Sar1-Ile8]
ANG II (Sarilesin).1,2
active peptides, the rigid geometry of the cyclic peptides
enhances the binding affinity towards a selected target
molecule compared to their linear counterparts. Fur-
thermore, cyclic analogues are important intermediates
in the design and synthesis of non-peptide mimetics with
the potential to be used as drugs.9ꢀ11 With this aim, our
group has been involved for several years in the design
and synthesis of cyclic analogues of important peptides
such as, Angiotensin II, Thrombin Receptor Peptides
and Myelin Basic Protein12ꢀ15 involved respectively in
hypertension, cancer and multiple sclerosis.
So far, a limited number of conformationally restricted
Angiotensin II analogues via cyclization have been
reported. In these studies cyclization was achieved either
by the disulfide method using cysteine moieties at var-
ious locations of the peptide molecule or by the amide-
linkage method.16ꢀ27
In this work, two cyclic ANG II analogues g,e-cyclo(3,
5)-[Sar1-Glu3-Lys5-Ile8] ANG II (I) and g,e-cyclo(3,5) -
[Sar1-Glu3-Lys5-Phe8] ANG II (II) have been synthe-
sized with Glu, Lys residues at positions 3 and 5 and
with Ile or Phe respectively at position 8 (Fig. 1). The
aim of this work was to investigate furthermore the role
of the 3,5-positions and the ring cluster receptor con-
formation in agonist activity and shed light to stereo-
electronic differences in activity and conformation upon
replacement of aromatic residue Phe with aliphatic Ile.
While replacement of Phe with Ile at position 8 pro-
duces an antagonist, the reverse of the Lys3-Glu5 of the
earlier reported cyclic peptides to Glu3-Lys5 has a mini-
mal effect in the agonist or antagonist activity of cyclic
ANG II analogues. A 15-membered ring in the central
core X-Oyr-Y allows a ring clustering and activity
regardless of the Glu-Tyr-Lys or Lys-Tyr-Glu sequence
in the agonist cyclic peptide. Based on these structure
activity relationships which demand the presence of
Phe, Tyr and His residues in ANG II as well as in linear
and constrained ANG II analogues to possess biological
activity, it can be inferred that the ability to form a ring
cluster and consequently a charge relay system may be
the key stereoelectronic molecular features of ANG II
for exerting biological activity.
Based on this information and modeling comparisons
between ring cluster and pharmacophoric groups of
AT1 antagonists we have synthesized an AT1 non-pep-
tide antagonist 1-[20-[(N-benzyl)tetrazol-5-yl]biphenyl-4-
yl]methyl]-2-hydroxymethylbenzimidazole.
Detailed
synthetic procedure for this analogue will be reported
elsewhere. This molecule was found to be much less
active inhibitor of ANG II than losartan in anesthetized
rabbits. For this reason we sought to compare its
superimposition ability with losartan (Fig. 1).
Rationale for the Synthesis of Cyclic Peptide Analogues
The limited stability of peptides often severely restricts
their medical and industrial application. Therefore, the
engineering of stable proteins is of great technological
and economical importance. If designed carefully with-
out causing drastic changes in the conformation of
Figure 1. Chemical structures of: BZI8 (top), Sar1-Arg2-[-cyclo(Glu3-
Tyr4-Lys5)]-His6-Pro7-Ile8 (middle), and Sar1-Arg2-[-cyclo(Glu3-Tyr4-
Lys5)]-His6-Pro7-Phe8 (bottom).

P. Roumelioti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2627–2633
2629
Scheme 1. Synthetic routes for the peptides I and II.

2630
P. Roumelioti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2627–2633
Our interest in the conformational model of Angio-
tensin II, which could be used as a basis for the synth-
esis of non-peptide receptor antagonists, prompted us to
design and synthesize two novel cyclic amide linked
Angiotensin II analogues, cyclo(3,5)-[Sar1-Glu3-Lys5-
Ile8] ANG II (I) and cyclo(3,5)-[Sar1-Glu3-Lys5-Phe8]
ANG II (II) (Fig. 1) that differ only in residue at posi-
tion 8. Analogues I and II as it was stated above differ
from cyclo(3,5)-[Sar1-Lys3-Glu5-Phe8] ANG II and
cyclo(3,5)-[Sar1-Lys3-Glu5-Ile8] ANG II recently repor-
ted5 only in the positioning of the Lys-Glu bridging
residues.
ANG II and g,e-cyclo(3,5) [Sar1-Glu3-Lys5-Ile8] ANG
II in 5% dextrose at final concentration of 5 and 50
mg/mL, respectively. Each solution and dose was
given in random sequence after a wash-out period of
30 min.
The hypertensive responses to angiotensin II infusion at
1, 2 or 3 mg/min and to g,e-cyclo(3,5) ANG II analogues
at 10, 20 or 30 mg/min were recorded continuously.
Angiotensin II-dependent hypertension was then
induced and maintained by a constant infusion of
angiotensin II via a syringe pump at a rate of 0.2 mL/
min (1 mg/min). Five min after the establishment of
hypertension, boluses of different doses of losartan
(0.05, 0.1 and 0.2 mg), BZ18 (1 and 2 mg) and the g,e-
cyclo(3,5) ANG II analogues (75, 150 and 300 mg) were
given via an ear vein in random sequence and the drop
in mean blood pressure was recorded.
Strategy of the synthesis of two cyclic analogues. Cycli-
zation was achieved by forming an amide-linkage
between the –NH2 and –COOH side chain groups of
Glu and Lys residues at positions 3 and 5, respectively,
which are the least important for activity.
Scheme 1 shows the synthesis of the new cyclic analo-
gues I and II.
The results presented in Figure 2 represent the mean of
three experiments. The cyclic ANGII analogue [Sar1,
Glu3, Lys5] ANG-II was capable to induce a dose-
related hypertensive response, which was of less
magnitude compared with that elicited by an isovolemic
and equimolar solution of angiotensin II (ANG II). In
contrast, the cyclic analogue [Sar1, Glu3, Lys5, Ile8]
ANG II, in which an isoleucin was substituted for
phenylalanine at position 8, did not preserve any
hypertensive action and in fact, it was transformed in a
potent angiotensin II antagonist reducing the
angiotensin II-dependent hypertension in a dose-related
manner.
Biological data. Adult normotensive male New Zealand
White rabbits weighing between 2.5 and 3.3 kg were
used in the study. All animals were anesthetized by
pentobarbitone (30 mg/kg), intubated and mechanically
ventilated with 100% oxygen using a respirator for
small. The tidal volume used was 15 mL and the rate
was adjusted to keep blood gases within normal range.
Two polyethylene catheters were inserted, one in the
carotid artery for continuous blood pressure monitoring
via a transducer attached to a multichannel recorder
and the other one in the jugular vein for the adminis-
trations of solutions made by diluting ANG II and its
cyclic analogues g,e-cyclo(3,5) [Sar1-Glu3-Lys5-Phe8]
As expected, losartan produced a significant dose-
dependent hypotensive response, when it was given as
Figure 2. Dose-dependent hypertensive responses were observed when solutions of ANG II or its cyclic analogue [Sar1, Glu3, Lys5] ANG II were
infused in anesthetized rabbits. In contrast, dose-related reduction of the mean blood pressure was seen when boluses of [Sar1, Glu3, Lys5, Ile8] ANG
II, losartan or BZ18 were given in anesthetized rabbits with angiotensin-II-induced hypertension.

P. Roumelioti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2627–2633
2631
Figure 3. Low energy conformers of cyclic agonist I and antagonist II peptides derived using conformational search methods.
a
bolus in our model of ANG II-dependent
Structure elucidation and conformational analysis of
BZI8. BZI8 was structurally elucidated using 2D DQF
COSY, ROESY and literature reported data on losar-
tan and eprosartan.28 ROESY and NOESY experiments
gave valuable information on the spatial proximity
between protons of the flexible part of the molecule. The
most critical observed NOE was between H4 and H29.
This suggests a clustering between the phenyl rings A
and F. The two rings that constitute the biphenyl system
are not linear to each other as it is also observed with
Losartan and Eprosartan.8,28 Absence of NOE between
–OH and H11 suggests that –OH group is oriented
away from the two protons of H-11. Based on this
information distance constraints were used along with
dynamics experiments and minimization procedures to
generate low energy conformers that are consistent with
NMR data. A representative low energy conformer for
BZI8 is shown in Figure 4. Computational details about
the methods used for the conformational analysis are
reported in our previous publication.8 Superimposition
hypertension in anesthetized rabbits. A similar but less
potent hypotensive response was observed when
boluses of the novel non-peptide ANG II receptor
antagonist BZI8 were given in the same experimental
model.
Conformational analysis of cyclic analogues I and II.
The new cyclic analogues were considered as mutated
structures of previously reported synthetic cyclic analo-
gues5 differing only in the order of aminoacids 3 and 5.
Figure 3 shows low energy structures for c-[Sar1-Lys3-
Glu5] ANG II and c-[Sar1-Lys3-Glu5-Ile8] ANG II gen-
erated using a combination of minimization algorithms
and molecular dynamics. The important stereoelec-
tronic characteristics of the generated models are the
following: the cyclic analogue c-[Sar1-Lys3-Glu5-Phe8]
ANG II adopts a Tyr4-Ile5-His6 bend, a trans amide
His6-Pro7 configuration and a side chain aromatic clus-
ter of the three key aminoacids Tyr4, His6 and Phe8. In
cyclic antagonist peptide c-[Sar1-Lys3-Glu5-Ile8] ANG II
with Ile at position 8, the ring cluster stereoelectronic
feature is missing.
The cyclic agonist and cyclic antagonist peptides adopt
the same conformation at the central part of the mole-
cule in which the three residue (Lys-Tyr-Glu) bridge is
the same for both agonist and antagonist peptides.
Therefore the difference in activity originates exclusively
from the C-terminal segment. The difference in the
magnitude of the biological activity between the cyclic
agonist analogue and the parent ANG II peptide hor-
mone is probably due to the different conformations
that each molecule assumes as it approaches the recep-
tor. Angiotensin II, like other linear peptide hormones,
exists in several different structural forms in water, but
as it approaches the receptor it adopts a predominant
conformation with much reduced flexibility. Cyclization
of Angiotensin II restricts the number of possible con-
formations, thereby presumably preventing the cyclic
peptide from assuming optimal conformation when it
binds to the receptor.
Figure 4. A representative low energy conformer of BZI8 derived
using a combination of 2D NMR spectroscopy and computational
analysis.

2632
P. Roumelioti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2627–2633
between losartan and BZI8 is shown in Figure 5. As it
can be seen the two tetrazole rings are oriented in the
opposite site. The biphenyl system of both antagonists
occupy the same space as well as the imidazole rings.
The butyl chain of losartan has not a satisfactory
counterpart even the phenyl ring attached to the tetra-
zole is located towards it. Superimposition between II
and BZI8 is shown in Figure 6. As it can be observed
benzyl tetrazole is in a close vicinity with C-terminal
carboxylate of Ile8 and its alkyl chain. However,
imidazole of His6 and phenolic hydroxyl group of Tyr4
are not in a spatial vicinity with 2-hydroxymethylbenz-
imidazole.
Conclusion
This research was aiming at establishing differences
between Angiotensin II agonist and antagonist peptides
and at confirming the aromatic ring clustering con-
formational model for Angiotensin II which has been
recently proposed on the basis of structure–activity
relationships, NMR and fluorescence life time studies.
A strong Tyr4-Ile5-His6 bend exists in cyclic agonist and
antagonist cyclic analogues which both bear the same
central Glu-Tyr-Lys or (Lys-Tyr-Glu as it is reported in
earlier study) ring moiety, indicating that the difference
in activity is resulting mainly from the nature of the C-
terminal residue. An aromatic residue such as Phe
allowing the formation of a ring cluster is necessary for
exertion of agonist activity in ANG II. The constrained
cyclic amide linked ANG II analogue c-[Sar1-Glu3-Lys5-
Phe8] and earlier reported c-[Sar1-Lys3-Glu5] ANG II
were designed to keep intact the clustering and back-
bone bend characteristics of the peptide hormone ANG
II. On the contrary, the cyclic antagonists [Sar1-Glu3-
Lys5-Ile8] ANG II and [Sar1-Lys3-Glu5-Ile8] ANG II
without an aromatic residue at position 8, lack the ring
cluster conformation. The molecules were designed with
the hypothesis that residues 3 and 5 do not govern the
biological activity and exist on the other side of the
molecule from the functionally important aromatic side
chains and this structure can be accommodated in the
charge relay conformation proposed for Angiotensin II.
The lower potency of both agonist and antagonist cyclic
analogues compared to ANG II and Sarilesin points out
that the achieved conformation by cyclic analogues is
not the optimal in inducing or suppressing ANG II
caused hypertension. This supports the hypothesis and
SAR studies that positions 3 and 5 are not responsible
for agonist or antagonist activity even though they
contribute to the optimum activity. In the potent con-
strained agonist analogues the three rings are indeed
closely spaced at the same side of the cyclic ring as it has
been shown by NOE interactions and molecular mod-
eling. Activity is similar regardless of Lys-Tyr-Glu or
Glu-Tyr-Lys order in cyclic agonist or antagonist
peptides.
Figure 5. Superimposition of low energy conformer of BZI8 with
Losartan.
The obtained data confirm our hypothesis that the aro-
matic side chains together with the C-terminal carbox-
ylate are the essential pharmacophoric groups for
receptor activation. These data also emphasize the role
of the closely spaced residues 4, 6 and 8 to form a pos-
sible relay system, which is not possible in Sarilesin and
cyclic ANG II antagonists, lacking a C-terminal aro-
matic residue.
Superimposition studies between the weak antagonist
BZI8 and losartan showed not a good mimicking
between their pharmacophore segments. For example,
their tetrazole moieties were located in the opposite site.
However, their biphenyl rings showed a good matching.
The lipophilic butyl chain of losartan was mimicked,
although not perfectly, by two rings. These structural
similarities and differences may give a plausible expla-
nation of the weak antagonist activity of BZI8. It is of
special interest that while the orientation of tetrazole in
Figure 6. Superimposition of low energy conformers of BZI8 with
peptide antagonist I.

P. Roumelioti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2627–2633
2633
the two molecules is in the opposite site the hydro-
phobic parts are located in spatial proximity.
11. Ono, S.; Hirano, T.; Yasutake, H.; Matsumoto, T.;
Yamaura, I.; Kato, T.; Morita, H.; Fujii, T.; Yamazaki, I.;
Shimasaki, C.; Yoshimura, T. Biosci. Biotechnol. Biochem.
1998, 62, 1621.
12. Alexopoulos, K.; Fatseas, P.; Melissari, E.; Vlahakos, D.;
Smith, J.; Mavromoustakos, T.; Saifeddine, M.; Moore, G.;
Hollenberg, M.; Matsoukas, J. Bioorg. Med. Chem. 1999, 7,
1033.
13. Matsoukas, J.; Hondrelis, J.; Agelis, G.; Barlos, K.; Gan-
ter, R.; Moore, D.; Moore, G. J. J. Med. Chem. 1994, 37,
2958.
14. Matsoukas, J.; Panagiotopoulos, D.; Keramida, M.;
Mavromoustakos, T.; Yamdagni, R.; Wu, Q.; Moore, G. J.;
Saifeddine, M.; Hollenberg, M. D. J. Med. Chem. 1996, 39,
3585.
Acknowledgements
This work was supported by the Ministry of Develop-
ment, General Secretariat of Research and Technology
of Greece (EPET II/115, PENED 1999, Joint Research
and Technology Program between Slovenian and
Greece 2001), a NATO Linkage Grant (974548), the
Latvian Science Council grant (96.0726) and the Lat-
vian National Scientific Programme (ZP-10).
15. Tselios, T.; Probert, L.; Daliani, I.; Matsoukas, E.; Tro-
ganis, A.; Gerothanassis, I.; Mavromoustakos, T.; Moore, G.;
Matsoukas, J. J. Med. Chem. 1999, 42, 170.
16. Miranda, A.; Juliano, L. Braz. J. Med. Biol. Res. 1988, 21,
903.
References and Notes
17. 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.
18. Matsoukas, J.; Scanlon, M.; Moore, G. J. J. Med. Chem.
1984, 27, 404.
19. Nikiforovich, G. V.; Marshall, G. R. Biochem. Biophys.
Res. Commun. 1993, 195, 222.
20. 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.; Skeen, R.; Marshall, G. R. J. Med.
Chem. 1990, 36, 1902.
1. Gavras, H.; Brunner, H. R.; Turini, G. A.; Kershaw, G. R.;
Tifft, C. P.; Cuttelod, S.; Gavras, I.; Vukovich, R. A.;
McKinstry, D. N. N. Engl. J. Med. 1978, 298, 991.
2. Blow, D. M.; Birktoft, J. J.; Hartley, B. S. Nature 1969,
221, 377.
3. Matsoukas, J.; Hondrelis, J.; Keramida, M.; Mavromous-
takos, T.; Makriyannis, A.; Yamdagni, R.; Wu, Q.; Moore,
G. J. J. Biol. Chem. 1994, 269, 5303.
4. Matsoukas, J.; Ancans, J.; Mavromoustakos, T.; Roume-
lioti, P.; Vlahakos, D. V.; Yamdagni, R.; Wu, Q.; Moore, G. J.
Bioorg. Med. Chem. 2000, 8, 1.
5. Turner, R. J.; Matsoukas, J. M.; Moore, G. J. Bioch. Bioph.
Res. Commun. 1990, 171, 996.
21. Zhang, W. J.; Nikiforovich, G. V.; Perodin, J.; Richard,
D. E.; Escher, E.; Marshall, G. R. J. Med. Chem. 1996, 39,
2738.
22. Jorgensen, E.; Patton, W. J. Med. Chem. 1969, 12, 935.
23. Nikiforovich, G. V.; Kao, J. L.-F.; Plucinska, K.; Zhang,
W. J.; Marshall, G. R. Biochemistry 1994, 33, 3591.
24. Ancans, J.; Biseniece, D.; Myshliakova, N.; Porunkevich,
E. Bioorg. Khim. 1986, 12, 118.
6. Turner, R. J.; Matsoukas, J. M.; Moore, G. J. Biochimica
et Biophysica Acta 1991, 21, 21.
7. Matsoukas, J.; Agelis, G.; Wahhab, A.; Hondrelis, J.;
Panagiotopoulos, D.; Yamdagni, R.; Wu, Q.; Mavromousta-
kos, T.; Maia, H.; Ganter, R.; Moore, G. J. J. Med. Chem.
1995, 38, 4660.
8. Mavromoustakos, T.; Kolocouris, A.; Zervou, M.; Rou-
melioti, P.; Matsoukas, J.; Weisemann, R. J. Med. Chem.
1999, 42, 1714.
9. Brugghe, F.; Timmermans, M.; Van Unen, A.; Ten Hove,
J.; Van De Werken, G.; Poolman, T.; Hoogerhout, P. Int. J.
Pept. Sci. 1999, 43, 166.
25. Ancans, J.; Biseniece, D.; Myshliakova, N.; Chipens, G.
Bioorg. Khim. 1990, 16, 358.
26. Biseniece, D.; Ancans, J.; Myshliakova, N.; Kublis, G.;
Porunkevich, E. Bioorg. Khim. 1990, 13, 149.
27. Matsoukas, J.; Hondrelis, J.; Agelis, G.; Barlos, K.;
Gatos, D.; Ganter, R.; Moore, D.; Moore, G. J. J. Med.
Chem. 1994, 37, 2958.
28. Zoumpoulakis, P., Grdadolnik, S. G., Matsoukas, J.,
Mavromoustakos, T. J. Pharmaceut. Biomed. 2002, 28, 125.
10. Mezo, G.; Majer, Z.; Valero, L.; Andreu, D. J. Pept. Sci.
1999, 5, 272.