Oligomeric formylpeptide activity on human neutrophils

Article abstract of DOI:10.1016/j.ejmech.2009.08.010

A series of oligomeric formylpeptides were synthesized by cross-linking the prototype fMLP using a Lys residue. They were then investigated for their ability to stimulate chemotaxis, superoxide anion production, and lytic enzyme release in human neutrophi

Full text of DOI:10.1016/j.ejmech.2009.08.010

European Journal of Medicinal Chemistry 44 (2009) 4926–4930
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Oligomeric formylpeptide activity on human neutrophils
,
a
b
b
Giorgio Cavicchioni a , Anna Fraulini , Sofia Falzarano , Susanna Spisani
*
a Department of Pharmaceutical Sciences, University of Ferrara, Via Fossato di Mortara 17/19, Ferrara 44100, Italy
b Department of Biochemistry and Molecular Biology, University of Ferrara, Via L. Borsari 46, Ferrara 44100, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of oligomeric formylpeptides were synthesized by cross-linking the prototype fMLP using a Lys
residue. They were then investigated for their ability to stimulate chemotaxis, superoxide anion
production, and lytic enzyme release in human neutrophils.
Although active in stimulating the different receptor isoforms, leading to the different biological
responses, these analogues showed a lesser potency and affinity than the standard peptide. On the basis of
the results reported here, we can hypothesise that: (i) the increased bulk of these molecules seems to
hinder their correct positioning into the receptor pocket, thereby hindering favourable receptor interac-
tion; and that: (ii) fMLP space positions do not seem to allow the ligand to increase biological responses.
Ó 2009 Elsevier Masson SAS. All rights reserved.
Received 29 September 2008
Received in revised form
9 July 2009
Accepted 11 August 2009
Available online 31 August 2009
Keywords:
N-formylmethionyl peptides
Human neutrophils
Chemotaxis
Superoxide anion generation
Lysozyme release
1. Introduction
concentrations of fMLP required to induce chemotaxis are lower
than those required to cause superoxide anion production or
Human neutrophils are phagocytic cells specialized in innate
host defence and they represent the largest subset of circulating
leukocytes. This cell type also has major associations with acute
inflammatory reactions.
lysosomal enzyme release [1]. fMLP acts by binding to classical
G-protein-coupled receptors, first identified in 1976 and subse-
quently classified as high-affinity (FPR) or low-affinity (FPRL-1,
FPR-like 1) fMLP receptors [4–6].
Neutrophils circulate in the bloodstream, and can be recruited in
high numbers in infected or inflamed tissues along the concentra-
tion gradients of chemoattractants released either by the infecting
microorganisms or arising from inflammatory reactions in the
infected tissue. These cells express a number of different chemo-
attractant receptors which enable them to sense invading microbes
and approach the site of infection by direct migration. Invaders are
subsequently phagocytized then degraded by toxic oxygen radicals
(respiratory burst) and/or hydrolytic enzyme release. The potent
bactericidal compounds produced by neutrophils are non-selective
agents, and their production must therefore be strictly regulated in
order to be efficient and safe for the host organism [1,2].
Various inflammatory mediators, including the N-formyl-
methionyl peptides derived from newly synthesised bacterial (or
mitochondrial) proteins, are capable of activating neutrophils via
specific receptors [3]. A distinct hierarchy of neutrophil responses
to varying concentrations of the prototype N-formyl-methionyl-
leucyl-phenylalanine (fMLP) has been clearly demonstrated; the
Downstream of this interaction, a number of signalling systems
are triggered: the intracellular FPR-cascade includes activation of
phosphoinositide 3-kinases (PI3Ks), phospholipase A (PLA), PLD,
and mitogen-activated protein kinases (MAPKs) [7–9].
Receptor-specific ligands may act as useful tools for studying
such complex signal transduction systems. A bivalent ligand, dis-
playing two copies of the pharmacophores, would be expected to
exhibit a considerably greater potency than the monomer, due to
entropic factors triggered when the linker possesses sufficient
length to bridge proximal recognition sites of different receptor
subtypes. Dimeric ligands of bioactive peptide molecules have
previously been synthesized and demonstrated to be specific and
selective. It has also been established that dimeric enkephalins,
neurokinins, and bradykinin, cross-linked by methylene chains,
display remarkable activities and high receptor subtype selectivity
[10–13]. These results suggest that: (i) dimeric ligands may exhibit
increased affinity, reliant on the length of the cross-linking spacers,
as compared to the corresponding monomeric analogues, and
(ii) the spacer length which produces a peak increase in affinity is
dependent upon the receptor type [10,13].
In the specific field of fMLP derivatives, Miyazaki et al. studied
dimeric fMLPs and found that (i) the length of the linker portion has
* Corresponding author. Tel.: þ39 532 455278; fax: þ39 532 455948.
E-mail address: g5z@unife.it (G. Cavicchioni).
0223-5234/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.08.010

G. Cavicchioni et al. / European Journal of Medicinal Chemistry 44 (2009) 4926–4930
4927
a determining influence on its activities and receptor selectivity,
(ii) lower activity may result from linkers with a branched structure,
probably due to an increased rigidity in the overall conformation,
(iii) a short linker triggers maximal activity, (iv) dimeric analogues
are highly selective ligands for chemotactic activity, which is elicited
far more potently than superoxide anion production, and (v) the
length of the dimeric analogue discriminates between the two
different receptor subtypes for chemotaxis and superoxide anion
production [14].
Because a short linker triggers maximal activity, we synthesized
cross-linked di-fMLP analogues by binding the carboxylic terminal
group of fMLP to a functional group from the side chain of a suitable
second residue from an fMLP-OMe analogue. The aim was to
ascertain whether (and in this case to what extent), the cross-linked
fMLP was able to contribute to enhance the biological response.
Our data was found to confirm the common conviction that the
isoform responsible for killing mechanisms is less exactly than that
responsible for chemotaxis, as superoxide anion production and
secretagogue activity, unlike chemotaxis, benefit from the presence
of the cross-linked tripeptide, particularly if sterically hindered.
Even though it is less potent than monomeric fMLP, this dimeric
analogue shows high receptor affinity [15,16].
Fig. 2. Superoxide anion production in human neutrophils provoked by fMLP-OMe
and its analogues. The data are the means of five separate experiments performed in
duplicate. S.E.M. are in 0.1–4 nmol Oꢀ2 range.
to be lower than those of the parent peptide. Specifically, the
potency of the analogues was ten times lower than that of fMLP-
OMe, and their chemotactic index was marginally lower than that
fMLP-OMe (C.I. ¼1.05) at 10ꢀ8 M, although in the same narrow
range (C.I. ¼ 0.90–1.01). Considering the other concentrations, we
can see that the efficacy of analogues was 4 z 1 z 3 (C.I.w0.95 vs
1.15 of fMLP-OMe), while compound 2 showed the lowest efficacy.
Analysing superoxide anion production (Fig. 2), all four deriva-
tives were found to have the same order of efficacy at increasing
concentrations. This was consistently lower than that of fMLP-OMe,
and reached its greatest difference at 10ꢀ6 M.
Lysozyme release activity (Fig. 3) showed a particular trend; the
efficacy was found to be incrementally lower than the prototype in
the concentration range 10ꢀ10 to 10ꢀ6 M, reaching an efficacy of
about 40% (compounds 2 and 4) and 20% (compounds 1 and 3) vs
about 60% of fMLP-OMe at 10ꢀ6 M. The lysozyme release activity of
the derivatives peaked at the concentration of 10ꢀ5 M, once again
a concentration ten times higher than that of the monomer
prototype.
By taking into account the above considerations, we report here
biological studies on human neutrophils of the following newly
synthesized oligo-fMLP cross-linked analogues: for-Met-Leu-Phe-
Lys[
(Phe-Leu-Met-for)-OMe 2, for-Met-Leu-Phe-Lys(Phe-Leu-Met-
for)- Lys(OMe)-Phe-Leu-Met-for 3, and for-Met-Leu-Phe-Lys(-
gAsp(OMe)-Phe-Leu-Met-for]-OMe 1, for-Met-Leu-Phe-Lys
3
Phe-Leu-Met-for)-Lys(Phe-Leu-Met-for)-OMe 4.
2. Results and discussion
The biological activity of the new compounds (1–4) under
examination was determined on human neutrophils and compared
to that of the reference ligand fMLP-OMe. Directed migration
(chemotaxis), superoxide anion production and lysozyme release
were measured as ‘‘efficacy’’ (which corresponds to the maximum
effect of a ligand) and ‘‘potency’’ (which corresponds to the concen-
tration of a ligand at which 50% of its maximum effect is reached).
Chemotactic activity elicited by the reference compound and
ligands 1–4 are reported in Fig. 1. The dose response curves, as
usually found for chemoattractants, all increased to reach a peak
and then decreased rapidly as the concentration of the ligand
increased. The presence of fMLP/fMLP-OMe in the same molecule
failed to enhance the activation of chemotaxis, irrespective of the
number inserted. Moreover, both potency and efficacy were found
3. Conclusions
As a part of a broad strategy in the study of oligomeric for-
mylpeptides, a series of fMLP analogues cross-linked by a Lys residue
were synthesized, in order to specifically investigate their ability to
better stimulate and/or selectively trigger neutrophil functions.
From the data presented here, the following items can be
drawn:
Compounds 1, 2, 3, and 4 are active in stimulating the different
receptor isoforms, leading to the different biological responses,
even though with a lesser potency and affinity than the standard
peptide.
Fig. 1. Chemotactic activity in human neutrophils provoked by fMLP-OMe and its
analogues. The data are the means of five separate experiments performed in
duplicate. S.E.M. are in 0.02–0.09 no % chemotactic index range.
Fig. 3. Release of neutrophil granule enzymes evaluated by determining lysozyme
activity induced by fMLP-OMe and its analogues. The data are the means of five
separate experiments performed in duplicate. S.E.M. are in 1–6% range.

4928
G. Cavicchioni et al. / European Journal of Medicinal Chemistry 44 (2009) 4926–4930
Fig. 4. – Scheme of synthesis and structural formula of for-Met-Leu-Phe-Lys[gAsp(OMe)-Phe-Leu-Met-for]-OMe.
These analogues have a very similar common behaviour with an
Retention times of the peptides were determined using HPLC
conditions in the above solvent system (solvents A and B) pro-
grammed at flow rates of 1 ml/min and using the following linear
gradients: from 0% to 100% B in 25 min. All peptides showed less than
1% impurities when monitored at 220 and 254 nm.
activity peak centered at 10ꢀ8 M and 10ꢀ5 M for chemotaxis and
superoxide anion production, respectively. Concerning lysozyme
release the four derivatives exhibit a profile of activity with
a maximum centered at 10ꢀ5 M and a rank of potency 4 > 2 ¼ 1 >3.
On the basis of the above reported results, we can argue that
(i) our polymeric analogues are unable to stimulate more than just
one receptor pocket, (ii) the increased bulk of these molecules
seems to hinder their correct allocation so avoiding to establish
requirements for optimal ligand–receptor interaction.
4.2. General procedure for the synthesis
for-Met-Leu-Phe-Lys[gAsp(OMe)-Phe-Leu-Met-for]-OMe 1 (Fig. 4)
– Colourless solid (mp 107–110 ꢁC; Rf 0.31; [
a
]D ¼ ꢀ17.3ꢁ, c ¼ 1,
In conclusion, the here described molecules provide new infor-
mations on structure–activity relationship concerning chemotactic
peptides and formyl-peptide receptors, taking into account the
recent identification of novel FPRs agonists which broadens the
spectrum of functional significance of such receptors. Further model
possessing different steric hindrances and different fMLP steric
positions are in progress in our laboratories to explore this topic.
MeOH; Mass: Naþ adduct 1151.34; Kþ adduct 1167.34).
for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)-OMe (Fig. 5) 2 – Colour-
less solid (mp 96–100 ꢁC; Rf 0.45; [
a
]D ¼ ꢀ15.2ꢁ, c ¼ 1, MeOH; Mass:
Naþ adduct 1022.1; Kþ adduct 1038.1).
for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)-3Lys(OMe)-Phe-Leu-Met-for
3 (Fig. 6) 3 – Colourless solid (mp 182–185 ꢁC; Rf 0.48; [
a
]D ¼ ꢀ20.4ꢁ,
c ¼ 1, MeOH; Mass: molecular ion 1547; Naþ adduct 1570; Kþ adduct
1586).
for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)-Lys(Phe-Leu-Met-for)-
OMe (Fig. 7) 4 – Colourless solid (mp 193–197 ꢁC; Rf 0.58;
4. Experimental protocols
[a
]D ¼ ꢀ23.7ꢁ, c ¼ 1, MeOH; Mass: molecular ion 1547; Naþ adduct
4.1. Chemistry
1570; Kþ adduct 1586).
Optical rotations were determined in MeOH at 20 ꢁC with
a Perkin–Elmer Model 241 polarimeter. Melting points were
determined on a Reichert–Kofler block, and are uncorrected.
Thin-layer chromatography (TLC) was performed on pre-coated
silica gel F254 plates (Merck) with the solvent system methylene
chloride/methanol 5:1.
Satisfactory C, H, N, S microanalyses were obtained for all
compounds, analytical results being within 0.4% of the theoretical
values.
4.3. Biological activity
4.3.1. Peptide dilution
A 10ꢀ2 M stock solution of fMLP-OMe (Sigma Chemical Co. St.
Louis MO, U.S.A.) and the new analogues were prepared in dimethyl-
sulfoxide (DMSO, Sigma) and diluted in Krebs–Ringer-phosphate
containing 0.1% w/v glucose [KRPG, pH 7.4] before use. KRPG was
made up as a five-times-working-strength stock solution with the
Amino acids were purchased from Fluka. Peptides were synthe-
sized following standard procedures in solution [16]. Removal of the
Boc group was performed by the treatment with a 1:1 mixture of
TFA/CHCl3. Peptide coupling was achieved by the 1-hydroxy-1,2,3-
benzotriazole (HOBt)/N-(3-dimethylaminopropyl)-N0-ethylcarbodii
mide (EDC) method [17], whereas the formyl group was introduced
according the N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline
(EEDQ) method [18].
Molecular weight of compounds was determined by an ESI
Micromass, ZMD2000 mass spectrometer.
Crude peptides were purified by preparative reverse-phase HPLC
using a Waters Delta Prep 4000 system with a Waters PrepLC 40 mm
Assembly column (30 ꢂ 4 cm, 300 Å, 15
mm spherical particle size
column). The column was perfused at a flow rate of 40 ml/min with
a mobile phase containing solvent A (10%, v/v, acetonitrile in 0.1%
TFA), and a linear gradient from 0 to 100% of solvent B (60%, v/v,
acetonitrile in 0.1% TFA) in 25 min was adopted for elution of the
compounds. HPLC analysis was performed by a Beckman System
Fig. 5. – Scheme of synthesis and structural formula of for-Met-Leu-Phe-Lys
(Phe-Leu-Met-for)-OMe. X, for-Met-Leu-Phe-OMe; , 1 for-Met-Leu-Phe-Lys[
Phe-Leu-Met-for]-OMe; A 2 for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)-OMe; C 3 for-Met-
Leu-Phe-Lys(Phe-Leu-Met-for)- Lys(OMe)-Phe-Leu-Met-for; for-Met-Leu-Phe-Lys
(Phe-Leu-Met-for)-Lys(Phe-Leu-Met-for)-OMe.
gAsp(OMe)-
3
6 4
Gold with a Beckman ultrasphere ODS column (5
mm, 4.6 ꢂ 250 mm).

G. Cavicchioni et al. / European Journal of Medicinal Chemistry 44 (2009) 4926–4930
4929
4.3.3. Random locomotion and chemotaxis
Random locomotion and chemotaxis studies were performed
with a 48-well microchemotaxis chamber (BioProbe, Milan, Italy),
and migration into the filter was evaluated by the leading-front
method, according to Zigmond and Hirsch [19]. The random
movement, used as control, was 32
mm ꢄ 3 SE of ten separate
experiments performed in duplicate. Chemotaxis was studied by
adding each peptide to the lower compartment of the chemotaxis
chamber. Peptides were diluted from a stock solution (10ꢀ2 M in
DMSO) with KRPG containing 1 mg/mL of bovine serum albumin,
and used at concentrations ranging from 10ꢀ12 to 10ꢀ5 M. Data are
expressed in terms of the chemotactic index (C.I.) ratio as follows:
(migration toward test attractant minus migration toward the
buffer)/(migration toward the buffer).
4.3.4. Superoxide anion (Oꢀ2 ) production
Superoxide anion production was measured by the superoxide
dismutase-inhibitable reduction of ferricytochrome c modified for
microplate-based assays [15]. Tests were carried out in a final volume
Fig. 6. – Scheme of synthesis and structural formula of for-Met-Leu-Phe-Lys(Phe-Leu-Met-
for)- Lys(OMe)-Phe-Leu-Met-for. X, for-Met-Leu-Phe-OMe; for-Met-Leu-Phe-Lys
Asp(OMe)-Phe-Leu-Met-for]-OMe; A 2 for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)-OMe;
C 3 for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)- Lys(OMe)-Phe-Leu-Met-for; 4 for-Met-
Leu-Phe-Lys(Phe-Leu-Met-for)-Lys(Phe-Leu-Met-for)-OMe.
3
, 1
[g
3
6
of 200
m
l containing 4 ꢂ105 neutrophils,100 nmol cytochrome c and
KRPG. At zero time, different amounts (10ꢀ10 to 10ꢀ4 M) of each
peptide were added and the plates were incubated in a microplate
reader (Ceres 900, Bio-Tek Instruments, Inc) with T set to 37 ꢁC.
Absorbance was recorded at wavelengths of 550 and 468 nm.
Differences in absorbance at the two wavelengths were used to
calculate the nmoles of Oꢀ2 produced, using a molar extinction
coefficient for cytochrome c of 18.5 mMꢀ1 cmꢀ1. Neutrophils were
following composition: NaCl, 40 g/l; KCl, 1.875 g/l; Na2HPO4.2H2O,
0.6 g/l; KH2PO4, 0.125 g/l; NaHCO3, 1.25 g/l; glucose, 10 g/l. 1 mM
MgCl2 and CaCl2 supplemented the buffer before biological tests. All
reagents were of the purest grade commercially available.
4.3.2. Purification of human neutrophils
Cells were obtained from heparinized (10 U/ml) peripheral blood
from fasting healthy subjects. 20 ml of blood was supplemented
with 12 ml of a solution consisting of 6% (by weight) Dextran T70
(Pharmacia, Upsala, Sweden) and was settled in a 50 ml poly-
propylene tube at room temperature for 45 min. The leukocyte-
containing turbid upper supernatant was carefully removed and
layered onto 10 ml of Ficoll-Paque (Pharmacia, density gradient for
lymphocyte isolation), and centrifuged at 250g for 20 min at room
temperature. The pellet containing the neutrophils was further
purified by hypotonic lysis of erythrocytes (0.86% NH4Cl for 10 min).
The cells were then washed twice and resuspended using KRPG, pH
7.4, at a final concentration of 50 ꢂ 106 cells/ml and used immedi-
ately. The neutrophils were 98–100% pure and ꢃ99% viable, as
determined by the Trypan blue exclusion test. The study was
approved by the local Ethics Committee, and informed consent was
obtained from all participants. All experiments were carried out
according to the guidelines set out by local and regional ethics
committees.
pre-incubated with 5
activation by peptides.
mg/ml cytochalasin B for 5 min prior to the
4.3.5. Granule enzyme assay
Neutrophil granule enzyme release was evaluated by deter-
mining the modified lysozyme activity in microplate-based assays
[15]. Cells were incubated in microplate wells in the presence of
each peptide at a final concentration of 10ꢀ10 to 10ꢀ4 M for 15 min
at 37 ꢁC. The plates were then centrifuged for 5 min at 400g, and
the amount of lysozyme was quantified nephelometrically by the
rate of lysis of a cell wall suspension of Micrococcus lysodeikticus.
Neutrophils were pre-incubated with 5 mg/ml cytochalasin B for
15 min at 37 ꢁC prior to the activation by peptides. The reaction rate
was measured using a microplate reader at 465 nm. Enzyme release
is expressed as a net percentage of the total enzyme content
released by 0.1% Triton X-100. Total enzyme activity was 85 ꢄ1
mg/
1 ꢂ107 cells/min.
4.4. Statistical analysis
Data are given as mean ꢄ S.E.M. The significance of the differ-
ences between oligo-fMLP analogues and the reference fMLP-OMe
was assessed by the non-parametric Wilcoxon test. Differences
between groups were judged to be statistically significant at P ꢅ 0.05.
Acknowledgments
This work was supported by MURST [Research Funds ex 60%]
and the Fondazione Cassa di Risparmio di Ferrara, Italy. We are
grateful to the Banca del Sangue of Ferrara for providing fresh blood
and Anna Forster for the English revision of the text.
References
Fig. 7. – Scheme of synthesis and structural formula of for-Met-Leu-Phe-Lys(Phe-Leu-Met-
[1] O. Iizawa, H. Akamatsu, Y. Niwa, Biol. Signals 4 (1995) 14–18.
[2] S. Spisani, R. Selvatici, in: D.E. Caplin (Ed.), Trends in Cell Signal, Nova Science
Publishers, New York, 2006, pp. 1–40 (Chapter 1).
[3] V.L. Katanaev, Biochemistry (Moscow) 66 (2001) 351–368.
[4] J. Hartt, G. Barish, P.M. Murphy, J. Gao, J. Exp. Med. 190 (1999) 741–747.
for)-Lys(Phe-Leu-Met-for)-OMe. X, for-Met-Leu-Phe-OMe;
Asp(OMe)-Phe-Leu-Met-for]-OMe; A 2 for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)-OMe;
C 3 for-Met-Leu-Phe-Lys(Phe-Leu-Met-for)- Lys(OMe)-Phe-Leu-Met-for; 4 for-Met-
Leu-Phe-Lys(Phe-Leu-Met-for)-Lys(Phe-Leu-Met-for)-OMe.
, 1 for-Met-Leu-Phe-Lys
[g
3
6

4930
G. Cavicchioni et al. / European Journal of Medicinal Chemistry 44 (2009) 4926–4930
[5] Y. Le, J.J. Oppenheim, J.M. Wang, Cytokine Growth Factor Rev. 12 (2001) 91–105.
[6] A. Dalpiaz, M.E. Ferretti, G. Vertuani, S. Traniello, A. Scatturin, S. Spisani, Eur. J.
Pharmacol. 436 (2002) 187–196.
[7] S. Spisani, M.C. Pareschi, M. Buzzi, M.L. Colamussi, C. Biondi, S. Traniello,
G. Pagani Zecchini, M. Paglialunga Paradisi, I. Torrini, M.E. Ferretti, Cell. Signal.
8 (1996) 269–277.
[12] H. Kodama, Y. Shimohigashi, K. Sakaguci, M. Waki, Y. Takano, A. Yamada,
Y. Hatae, H. Kamiya, Eur. J. Pharmacol. 151 (1988) 317–320.
[13] J.C. Cheronis, E.T. Whalley, K.T. Nguyen, S.R. Eubanks, L.G. Allen, M.J. Duggan,
S.D. Loy, K.A. Bonharn, J.K. Blodgett, J. Med. Chem. 35 (1992) 1563–1572.
[14] M. Miyazaki, H. Kodama, I. Fujita, Y. Hamasaki, S. Miyazaki, M. Kondo, J. Bio-
chem. 117 (1995) 489–494.
[8] R. Selvatici, S. Falzarano, S. Traniello, G. Pagani Zecchini, S. Spisani, Cell. Signal.
15 (2003) 377–383.
[15] G. Cavicchioni, M. Turchetti, K. Varani, S. Falzarano, S. Spisani, Bioorg. Chem. 31
(2003) 322–330.
[9] R. Selvatici, S. Falzarano, A. Mollica, S. Spisani, Eur. J. Pharmacol. 534 (2006)
1–11.
[16] S. Spisani, A. Fraulini, K. Varani, S. Falzarano, G. Cavicchioni, Eur. J. Pharmacol.
567 (2007) 171–176.
[10] Y. Shimohigashi, T. Costa, H.C. Chen, D. Rodbard, Nature 297 (1982) 333–335.
[11] M. Kondo, H. Kodama, T. Costa, Y. Shimohigashi, Int. J. Pept. Protein Res. 27
(1986) 153–159.
[17] W. Ko¨nig, R. Geiger, Chem. Ber. 103 (1970) 788–798.
[18] G. Lajoie, J.L. Kraus, Peptides 5 (1984) 653–654.
[19] S.H. Zigmond, J.G. Hirsch, J. Exp. Med. 137 (1973) 387–410.