Influence of C-terminal modifications of bradykinin antagonists on their activity

Article abstract of DOI:10.1135/cccc19971940

In the present study we have described the synthesis and some pharmacological properties of four new analogues of bradykinin (BK). Two peptides were designed by substitution of positions 7 and 8 of known B₂ antagonists with N-methyl-D-phenylala

Full text of DOI:10.1135/cccc19971940

1940
Prahl et al.:
INFLUENCE OF C-TERMINAL MODIFICATIONS OF BRADYKININ
ANTAGONISTS ON THEIR ACTIVITY
a1
b
b
b
Adam PRAHL , Tomasz WIERZBA , Pawel WINKLEWSKI , Magdalena WSZEDYBYL ,
b
b
a2
Maciej CHEREK , Witold JUZWA and Bernard LAMMEK
a Department of Chemistry, University of Gdansk, Sobieskiego 18, 80-952 Gdansk, Poland;
e-mail: 1 egon@hebe.chem.univ.gda.pl, 2 bernard@hebe.chem.univ.gda.pl
b Institute of Physiology, Medical Academy, Debinki 1, 80-211 Gdansk, Poland
Received March 4, 1997
Accepted October 22, 1997
In the present study we have described the synthesis and some pharmacological properties of four
new analogues of bradykinin (BK). Two peptides were designed by substitution of positions 7 and 8
of known B2 antagonists with N-methyl-L-phenylalanine [Phe(Me)]. The next two analogues were ob-
tained by replacement of D-Phe residue in position 7 of known B2 antagonist with 1-naphthyl-D-alanine
or 2-naphthyl-D-alanine. The antagonistic potency of peptides was assessed by their ability to inhibit
vasodepressor response to exogenous bradykinin in conscious rats. Although our studies demonstrated
disadvantegous influence of Phe(Me)7,8 modification for B2 antagonism, we showed that D-amino
acid residue in position 7 of BK antagonists may be replaced by suitable L-amino acid residue. As re-
gards (D-Nal)7 substitution, we found strikingly different antagonistic potencies of analogues which differ
only in the presence of D-1-Nal and D-2-Nal. We assume that it is due to different conformations of
these peptides, proving the importance of the shape of the C-terminal part of B2 antagonists for their
activity.
Key words: Peptides; Bradykinin; B2 antagonists; Blood pressure.
Modifications currently used for structure–activity relationships study of peptides are
of various types: addition, deletion, substitution of one or more amino acid residues,
cyclization, or use of peptide bond isosters. In this field, N-methylation is considered as
one of the local and subtle modes of conformational constraint. There are many infor-
mation outlining structural perturbations induced by N-methylation: steric constraints,
suppression of a proton donating N–H group capable of hydrogen bonding, reduction of
the predominance of the trans vs cis peptide bond and increased basicity of the carbo-
nyl group. Detailed conformational studies indicate that the influence of N-methylation
on a conformation depends to a large extent on the chirality of the residues surrounding
the modified peptide bond1. The biological significance of this modification still re-
mains a question, there are suggestions that it either provides enhanced resistance
against biodegradation or increases hydrophobicity.
Having all this in mind we decided to check how replacement of amino acid residues
in positions 7 and 8 of the Stewart’s antagonist [D-Arg0,Hyp3,Thi5,8,D-Phe7]BK
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Influence of C-Terminal Modifications
1941
with N-methyl-L-phenylalanine [Phe(Me)]* will influence pharmacological
properties of the resulting analogue 1. We also found it interesting to apply the same
modification for Aaa[D-Arg0,Hyp3,Thi5,8,D-Phe7]BK, one of our previously synthesized
potent B2 antagonists2 (analogue 2). This analogue, in fact, may be considered as a
derivative of 1 with the 1-adamantaneacetyl group (Aaa) on its N-terminus.
It is believed that the critical change conferring bradykinin antagonist activity upon
analogues is the replacement of Pro7 with aromatic D-amino acids3. D-Phe has been
most widely used and appears to be acceptable in analogues with a wide variety of
additional modifications. Later it was shown that only a narrow range of D-amino acid
residues is acceptable for production of antagonistic activity. Naphthylalanine (Nal) is
one of the amino acids which lead to less active compounds3. On the other hand, our
results obtained for modification of position 3 of arginine vasopressin (AVP) or its
analogues with L-2-Nal and L-1-Nal have proved there to be great differences in acti-
vities of compounds, which are distinguished only by the presence of these two
residues. These amino acids differ only because the naphthalene ring is connected by
its position 1 and 2 to the backbone of the molecule, respectively. The hindering effect
caused by bulky naphthalene ring near the peptide bond is in the case of L-1-Nal much
greater than for L-2-Nal. In our opinion this may have a significant impact on confor-
mation of an analogue and can thus influence its interaction with receptors4.
This finding prompted us to investigate how substitution of position 7 of our
previously synthesized antagonist, which we already used as a model to design anal-
ogue 2, with D-1-Nal and D-2-Nal will affect pharmacological properties of resulting
compounds 3 and 4.
All synthesized analogues have the following structure:
X-D-Arg-Arg-Pro-Hyp-Gly-Thi-Ser-Y-Z-Arg-OH
Analogue
X
Y
Z
1
2
3
4
H
Phe(Me)
Phe(Me)
D-1-Nal
D-2-Nal
Phe(Me)
Phe(Me)
Thi
Aaa
Aaa
Aaa
Thi
* Abbreviations: The symbols of the amino acids and peptides are in accordance with 1983
Recommendations of the IUPAC-IUB Joint Commission on Biochemical Nomenclature (European J.
Biochem. 138,
9 (1984)). Other abbreviations: Aaa, 1-adamantaneacetic acid; DCM,
dichloromethane; DMF, dimethylformamide; HOBt, N-hydroxybenzotriazole; D-1-Nal, 1-naphthyl-
D-alanine; D-2-Nal, 2-naphthyl-D-alanine; TBTU, benzotriazol-1-yl-N,N,N′,N′-tetramethyluoronium
tetrafluoroborate; Thi, α-(2-thienyl)alanine.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1942
Prahl et al.:
EXPERIMENTAL
The optical rotations were measured by means of a Perkin–Elmer Model 141 polarimeter. For amino
acid analysis, the peptides (0.5 mg) were hydrolyzed with constantly boiling hydrochloric acid (400 µl),
containing phenol (20 µl), in evacuated sealed ampoules for 18 h at 110 °C. The analyses were per-
formed on a Microtechna type AAA881 analyzer. The elemental analyses were determined on a
Carlo–Erba Model 1106 analyzer. TLC was carried out on silica plates (Merck), and the spots were
visualized by iodine or ninhydrine. The following solvent systems were used: (A) butan-1-ol–acetic
acid–water, 4 : 1 : 5 (v/v), upper phase; (B) ethyl acetate–pyridine–acetic acid–water, 5 : 5 : 1 : 3 (v/v);
(C) butan-1-ol–pyridine–acetic acid–water, 52 : 12 : 12 : 25 (v/v).
The purity of the peptides was also ascertained by HPLC. Analyses of the analogues were performed
on a Gold System Beckman chromatograph with an Ultrasphere ODS column (5 µm, 4.6 × 150 mm).
Solvent system: (1) 0.1% trifluoroacetic acid (TFA), (2) acetonitrile–0.1% TFA, 80 : 20 (v/v), linear
gradient from 30 to 90% of (2) for 20 min, λ = 226 nm, flow rate 1 ml/min. Each analogue gave a
single peak. The purity of all peptides was between 95 and 97% as determined from the integrated
areas recorded at 226 nm.
N-tert-Butoxycarbonyl-N-methylphenylalanine was obtained in 86% yield from N-tert-butoxycar-
bonylphenylalanine (7.95 g, 0.05 mol), methyl iodide (15 ml, 0.24 mol) and sodium hydride disper-
sion (3.96 g, 0.09 mol) according to Cheung5 as an oil, M+ = 279 (m/z); dicyclohexylammonium salt:
[α]2D0 = –24.1° (c 1, methanol). For C27H44N2O4 (480.6) calculated: 70.4% C, 9.6% H, 6.1% N;
found: 70.2% C, 9.7% H, 6.1% N.
Peptide Synthesis
All peptides were prepared by the solid phase synthesis method by stepwise coupling of Boc-amino
acids to the growing peptide chain on a Merrifield resin6. Boc-Arg(Tos)-resin (Sigma; 0.35 mmol of
amino acid per gram; 1.0 g) was converted to the protected decapeptidyl resins (analogue 1) or
acyldecapeptidyl resins (analogues 2, 3, 4) in the nine and ten cycles of standard solid phase syn-
thesis, respectively6. Boc-Arg(Tos)-OH, Boc-D-Arg(Tos)-OH, D-1-Nal-OH and D-2-Nal-OH were dis-
solved prior to coupling in a mixture DMF–DCM (3 : 1). For coupling of Boc-Phe(Me)-OH,
TBTU/HOBt in a mixture DMF–DCM (2 : 1) was used. Boc-Hyp-OH was coupled without protec-
tion of the OH-group. N-1-Adamantaneacetic acid was used in the final coupling steps for analogues
2, 3 and 4. The completion of all coupling reactions was monitored by the Kaiser test7. After syn-
thesis was completed, 1 g of the protected acylpeptidyl resin was treated with 10 ml of liquid hy-
drogen fluoride containing 1 ml of anisole at –70 °C and stirred for 50 min at 0 °C. After the removal
of the HF and anisole in vacuo the mixture was washed with anhydrous diethyl ether (3 × 30 ml) and
then with 30% acetic acid (5 × 40 ml). The acetic acid extracts were combined and lyophilized to
yield the crude product. The material was desalted by gel chromatography on a Sephadex G-15 col-
umn (120 × 2.9 cm) with 50% aqueous acetic acid at a flow rate of 5 ml/h. Fractions comprising the
major peak were pooled and lyophilized, and the residue was further subjected to gel chromatography
on a Sephadex LH-20 column (120 × 1.4 cm) with 30% aqueous acetic acid at a flow rate of 2 ml/h.
The peptide was eluted as a single peak. Lyophilization of the pertinent fractions gave the bradykinin
analogue. Physico-chemical properties of the new analogues (1–4) are presented in Table I.
Effect of Analogues on Rat Blood Pressure
The antagonistic potency of the analogues was assessed by their ability to inhibit the vasodepressor
response to exogenous bradykinin in conscious rats2,8, as follows. Male, intact Wistar albino rats
(300–350 g) were maintained on a regular chow diet, as well as tap water in a room at constant
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Influence of C-Terminal Modifications
1943
temperature (23 ± 1 °C). One day before the experiment, the right carotid and the iliac artery were
catheterized with polyethylene tubing (PE50) under light ether anaesthesia. A “Y” type connection
was attached to the carotid artery for injection of bradykinin and for infusion of the bradykinin anal-
ogues. All catheters were exeriorized subcutaneously at the back of the neck. On the day of the ex-
periment, the rats were conscious and unrestrained in plastic cages. Mean arterial pressure (MAP)
and heart rate (HR) were monitored through a Gould–Statham P23-ID pressure transducer (Gould,
Cleveland, OH, U.S.A.) connected to the iliac catheter and recorded on a paper chart recorder. A 30 min
stabilization period was allowed before initiation of the experiment. Angiotensin-converting enzyme
inhibitor, Enalapril (Merck Sharp and Dohme Research Lab., Rahway, NJ, U.S.A., 1 mg/kg) was in-
jected through the iliac catheter. Thirty to sixty minutes later, when a stable blood pressure was ob-
tained, bradykinin acetate salt (Sigma) (62.5, 125, 250 ng) dissolved in 5% glucose to a concentration
of 2.5 µg/ml, was injected every 4 to 5 min into one branch of the carotid catheter. Each dose was
repeated two or three times until the vasodepressor responses to exogenous bradykinin were stable.
The vasodepressor response to BK was plotted against the logarithm of the bradykinin dose. The
average values of the responses to 125 and 250 ng were calculated from the regression line obtained
from the log dose–effect plot. Both vasodepressor responses to 125 and 250 ng were taken, respec-
tively, as the control responses. The BK analogue dissolved in 5% dextrose solution was infused to
a branch other than the BK of the carotid catheter. A constant rate of infusion, 125 µl/min, was pro-
vided using an infusion pump (F5z Dialyse 15; Dascon BV, Uden, The Netherlands). The bradykinin
analogue administration was initiated with an 8-min infusion at a concentration of 0.5 µg/ml (this gave
a dose of 62.5 ng/min). By the end of the 3rd and the 7th minute of this infusion 250 ng of BK was
injected into the carotid artery. In some experiments, where the vasodepressive responses to BK were
different from each other by more than 3 mm Hg, the infusion was prolonged by 150 s and the third
dose of BK was injected by the end of this infusion. The mean value of the vasodepressive responses
to BK were taken for further data analysis. The dose of bradykinin antagonists was then increased
(0.5, 2, 8, 32, 120 and, if necessary, 400 and 1 000 µg/ml) and the some procedure was repeated
until the vasodepressor response to exogenous bradykinin decreased to less than 10% of the control
response.
The antagonistic potency of the bradykinin analogues was quantitatively expressed as the effective
doses; ED20, ED50 and ED90. The effective doses: ED20, ED50 and ED90 represented doses of brady-
kinin antagonist (µg/min) that inhibit 20%, 50% and 90%, respectively, the vasodepressor response
TABLE I
Physico-chemical characteristics of bradykinin analogues
RF
M+
m/z
[α]2D0
(c = 1, 1M AcOH)
Yielda, %
(crude)
Yieldb, %
(purified)
Peptide
A
B
C
1
2
3
4
1 316
1 492
1 520
1 520
–
0.68
0.66
0.77
0.75
0.44
0.27
–
–85.0
–105.5
–72.1
78
62
75
67
21
12
27
22
–
0.11
0.13
–
–69.7
a
b
Yields were calculated on the basis of the arginine content of the starting resin. All peptides gave
expected amino acid analysis ratios after hydrolysis (±0.05). The purity of analogues determined on HPLC
was between 95 and 97%.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1944
Prahl et al.:
to its agonist (250 ng of bradykinin). For evaluation of the ED values linear regression analysis was
used. For each experiment, the relationship between the dose of the tested bradykinin analogue and
the inhibition of the vasodepressor response to bradykinin was determined by fitting a least-square
regression line. In all cases significant linear regression was found between the dose and the effect
of bradykinin antagonist (correlation coefficient r was never lower than 0.96 and P was never higher
than 0.05). The ED values were finally assessed from the regression line. In some experiments, the
ED90 (bradykinin analogues 2 and 3) exceeded the applied bradykinin analogue dose range and the
values of those ED were evaluated from the regression lines obtained by extrapolation.
Results are reported as mean values of ±S.E. Comparison of the two analogues was accomplished
by Student’s non-paired t-test9. Differences were considered to be significant for P < 0.05.
RESULTS AND DISCUSSION
Four new, C-terminal-related analogues of bradykinin were synthesized using the solid-
phase method on chloromethylated Merrifield resin. The antagonistic potency of pep-
tides was evaluated by their ability to inhibit vasodepressor response to exogeneous
bradykinin in conscious rats2,8. Some pharmacological properties of new analogues 1–4
in comparison with those of some previously synthesized antagonists, which we tested
under the conditions of our assay, are presented in Table II. We assessed relatively high
values of S.E. in our assay. Since bradykinin and also its analogues are known to ex-
hibit potent sensory effects (e.g., itching, pain) the vasomotoric responses evoked are
not usualy uniform in conscious animals and this is probably the reason for the diver-
sity obtained within the experimental subpopulations. On the other hand, bradykinin is
highly involved in the afferent part of the cardiovascular reflexes controlling blood
TABLE II
Pharmacological data on bradykinin analogues
Antagonistic potency, µg/min
Analogue
na
ED20
ED50
ED90
[D-Arg0,Hyp3,Thi5,8,(NMe)Phe7,8]BK
Aaa[D-Arg0,Hyp3,Thi5,8,(NMe)Phe7,8]BK (2)
Aaa[D-Arg0,Hyp3,Thi5,8,D-1-Nal7]BK
Aaa[D-Arg0,Hyp3,Thi5,8,D-2-Nal7]BK
[D-Arg0,Hyp3,Thi5,8,D-Phe7]BKb
(1)
3
4
8
6
7
7
weak agonist
4.58 ± 2.75 49.18 ± 41.17 1 676.36 ± 1 088.78
1.54 ± 0.99 27.72 ± 18.11 1 148.34 ± 867.32
(3)
(4)
(5)
(6)
0.37 ± 0.09
1.49 ± 0.29 10.52c
0.84 ± 0.09
2.80c
1.99 ± 0.42
18.57 ±
142.5
13.94 ±
3.27
26.4
1.69
±
Aaa[D-Arg0,Hyp3,Thi5,8,D-Phe7]BKb
a
b
Number of rats tested. Data from ref.10. This peptide was previously designed by Stewart’s
group3. As we used a different assay for the evaluation of antagonistic properties of our peptides we
tested this analogue in our system as a reference. The ED50 value was evaluated by extrapolation
c
from the ED20 and ED90 values.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Influence of C-Terminal Modifications
1945
pressure. Anaesthesia fades these reflexes considerably and alters cardiovascular re-
sponses to bradykinin.
In our assay analogue 4 is a very potent antagonist whereas peptides 2 and 3 exhibit
rather low potency. Analogue 1 is a weak agonist. In lower doses (ED20) our most
active peptide 4 is about 4 and 2 times more potent, respectively, than antagonists
synthesized previously in Stewart’s (compound 5) and our (compound 6) laboratory. At
higher doses (ED90), peptide 4 is approximately 9 times more potent than compound 5
and equipotent with compound 6. Activities of new antagonists 2 and 3 are of similar
range as those of 5 and 6 only for ED20. In higher doses both new analogues are much
less active than 5 and 6.
From the results presented it is clear that (NMe)Phe7,8 modification results in dra-
matic decrease of antagonistic activity (analogue 2) or even conversion it into weak
agonist (analogue 1). On the other hand, analogue 2 is the first example of B2 antagonist
having L-amino acid residue in position 7. The presence of D-amino acid residue, until
now, was considered to be necessary for B2 antagonism. The present data seem also to
support our previously raised thesis concerning interaction between analogues acylated
with bulky substituent and receptor. It is evident that peptides 1 and 2, which differ
only by the presence of 1-adamantaneacetyl group on the N-terminus of 2, showing
opposite types of activities, e.g. agonism and antagonism, are good examples of the
influence of bulky acyl substituent on antagonistic potencies of analogues.
Passing on to compounds 3 and 4, regardless of the previous knowledge about disad-
vantageous effect of D-Nal7 substitution3, we decided to substitute D-Phe7 in our potent
B2 antagonist, Aaa[D-Arg0,Hyp3,Thi5,8,D-Phe7]BK, with D-1-Nal and D-2-Nal. As men-
tioned in the introduction, the hindering effect caused by the bulky naphthalene ring
near the peptide bond is in the case of L-1-Nal much greater than for L-2-Nal. Indeed,
analogues 3 and 4 having strikingly different antagonistic potencies proved the import-
ance of conformation of the C-terminal part of the peptide for its antagonistic proper-
ties. It should be emphasized, that such great difference of activities is due to minor
change in the structure of analogues, which differ only because the naphthalene ring is
connected by its position 1 or 2 to the backbone of the molecule. We can assume that
this factor, e.g. hindering effect of naphthalene ring, has a significant impact on the
bioactive conformations of molecules which contain these amino acids and thus can
influence their interaction with B2 receptors.
Summing up, although our studies demonstrated that the Phe(Me)7,8 modification is
disadvantageous for B2 antagonism, we showed that D-amino acid residue in position 7
of BK antagonists which, until now, was considered to be necessary for B2 antagonism,
may be replaced by suitable L-amino acid residue. As regards D-Nal7 substitution, the
great difference of activities due to minor change in the structure caused by the
presence in position 7 of analogues D-1-Nal or D-2-Nal support the known thesis about
importance of conformation of C-terminal part of B2 antagonists for their activity. Our
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

1946
Prahl et al.:
study besides providing new information about structure–activity relationship of brady-
kinin antagonists, resulted in the highly active analogue 4 which may be useful in clari-
fying the role of BK in various physiological effects.
This work was partially supported by the Polish State Committee for Scientific Research, Grant No.
PB 0209/P3/94/06.
REFERENCES
1. Vitoux B., Aubry A., Cung M. T., Marrand M.: Int. J. Pept. Protein Res. 27, 617 (1986).
2. Lammek B., Wang Y. X., Gavras I., Gavras H.: Peptides 11, 1041 (1990).
3. Stewart J. M., Vavrek R. J. in: Bradykinin Antagonists: Basic and Clinical Research (M. Bürch,
Ed.), p. 51. Dekker, New York-Basel-Hong Kong 1991.
4. Lammek B., Czaja M., Derdowska I., Rekowski P., Trzeciak H., Sikora P., Szkrobka W., Stojko
R., Kupryszewski G.: Int. J. Pept. Protein Res., in press.
5. Cheung S. T., Benoiton N. L.: Can. J. Chem. 55, 906 (1977).
6. Stewart J. M., Young J. D.: Solid Phase Peptide Synthesis. Pierce Chemicals, Rockford (IL)
1984.
7. Kaiser E., Colescott R. L., Bossinger C. D., Cook P. J.: Anal. Biochem. 34, 595 (1970).
8. Lammek B., Kazmierkiewicz R., Ito Y., Gavras H.: Pol. J. Chem. 67, 1053 (1993).
9. Tallarida R. J., Murray R. B.: Manual of Pharmacologic Calculations. Springer,
Berlin-Heidelberg-New York 1987.
10. Lammek B.: Pol. J. Chem. 68, 913 (1994).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)