Institut de Pharmacologie Moléculaire et Cellulaire du Centre
National de la Recherche Scientifique (V.B., P.V., C.F.), Valbonne,
France; and Institut National de la Santé et de la Recherche
Médicale U343 (J.P.B.), Hôpital de l'Archet, BP 79, Nice,
France
Derivatives of salicylic acid (SA) and benzoic acid prevent
endothelin-1 (ET-1) binding to ETA receptors. This study analyzed actions of 30 derivatives of benzoic acid and salicylic acid on 125I-ET-1 binding to recombinant rat ETA receptors. The
most active compounds were 3,5-dibromosalicylic acid (Br2SA,
Ki = 0.5 mM) and 3,5-diiodosalicylic
acid (Ki = 0.3 mM). They were
about 50 times more potent than SA and aspirin. Br2SA inhibited
equilibrium 125I-ET-1 binding in an apparently competitive
manner. It accelerated 8-fold the dissociation of 125I-ET-1
receptor complexes and did not modify the second order rate constant of
association of 125I-ET-1 to its receptors. Br2SA also
decreased the affinity of ETA receptors for receptor antagonists BQ-123
and bosentan. Br2SA accelerated dissociation of
125I-ET-1-solubilized ETA receptor complexes and decreased
the apparent molecular size of solubilized receptors. Br2SA and
3,5-diiodosalicylic acid inhibited two cellular actions of ET-1: the
mobilization of intracellular Ca2+ stores in isolated cells
and contractions of rat aortic rings. They accelerated the relaxing
action of BQ-123 and bosentan in ET-1-treated aortic rings. The results
suggest the existence of an allosteric modifier site on ETA receptors
that recognizes selected derivatives of SA. SA derivatives might be of
therapeutic interest to relieve tight ET-1 binding and to favor actions
of receptor antagonists.
 |
Introduction |
Endothelin-1
is a potent vasoconstrictor peptide that recognizes G protein-coupled
receptors that stimulate the phospholipase C-signaling cascade (Van
Renterghem et al., 1988
; Yanagisawa et al., 1988). A unique property of
ET-1 is its capacity to bind almost irreversibly to its receptors
(Waggoner et al., 1992). Quasi-irreversible binding of ET-1 has many
functional and pharmacological consequences that have only recently
been appreciated. The affinity of ET-1 for its receptors is
overestimated in many binding experiments because of time-limited
second order kinetic conditions (Desmarets et al., 1996). Tight binding
has been proposed to be responsible for the lack of action of guanine
nucleotides on ET-1 binding (Nambi et al., 1996) and for the long-term
refractoriness that follows actions of ET-1 (Leite et al., 1994;
Hilal-Dandan et al., 1997). Tight binding imposes conditions in which
short-range (autocrine) actions of ET-1 are favored over long-range
(paracrine) actions. It provides an explanation for the observation
that functional receptors serve as clearance receptors (Frelin and
Guedin, 1994). It has also been suggested that, because of its
irreversible binding, ET-1 is more likely to contribute to long-term
physiological or physiopathological regulations than to short-term
regulations (Hilal-Dandan et al., 1997). Another consequence of tight
ET-1 binding is that circulating levels of ET-1 are not representative of the real amount of ET-1 present in tissues (Ferrari et al., 1998).
Finally, tight binding may limit the access of competitive receptor
antagonists to the receptors. Indeed, if ET-1 already sits on a
receptor, receptor antagonists cannot act until ET-1 leaves the site
and gives the antagonist a chance to compete with ET-1 for the
occupancy of newly accessible sites (Talbodec et al., 2000). This
simple fact may be a reason for the limited usefulness of ET receptor
antagonists against endogenous ET-1.
Several G protein-coupled receptors are regulated by allosteric
ligands. Well studied examples are D2 dopamine receptors (Hoare and
Strange, 1998), muscarinic receptors (Tucek and Proska, 1995), adenosine A1 receptors (Bruns and Fergus, 1990), and 
2-adrenergic receptors (Nunnari et al., 1987). We previously reported that aspirin
and salicylic acid are allosteric inhibitors of ETA receptors (Talbodec
et al., 2000). These actions of salicylates were observed at
concentrations >10 mM, which were too large to carry out a detailed
investigation of their mechanism of action. We therefore screened a
number of derivatives of SA and benzoic acid to find compounds that
would be more potent. This procedure led to the identification of
dihalogenated derivatives of SA that are about 50 times more potent
than aspirin. This paper defines the properties of interaction of Br2SA
with recombinant rat ETA receptors and analyzes some of the
pharmacological properties of Br2SA and I2SA.
| |
Materials and Methods |
Chemicals.
BQ-123 was from Néosystem (Strasbourg,
France). ET-1, indo-1/AM, CHAPS, digitonine, and protease inhibitors
were from the Sigma Chemical Co. (St. Louis, MO).
125I-ET-1 (2200 Ci/mmol) was prepared as
previously described (Desmarets et al., 1996) and stored at 
20°C.
Salicylic acid and benzoic acid derivatives were purchased from Avocado
(Heysham, UK), Acros Organics (Geel, Belgium), or Sigma. Sodium salts
were used. Bosentan was obtained from Dr. M. Clozel (Actelion, Basel, Switzerland).
Cell Cultures.
Stable transfectant CCl39 cells expressing
functional rat ETA receptors were prepared as previously described
(Gresser et al., 1996). Cell membranes were prepared as previously
described (Desmarets et al., 1996). They were resuspended in a buffer
of the following composition: 5 mM EGTA, 250 mM sucrose, 10 mM Tris-Cl, pH 7.4, supplemented with a cocktail of protease inhibitors (1 µM
bacitracin, 0.1 mM phenylmethylsulfonyl fluoride, 1 µM leupeptin). Membranes (4-8 mg of protein/ml) were stored at 20°C until use. Proteins were determined according to the Bradford (1976) method using
bovine serum albumin as standard. Rat brain capillary endothelial cells
of the B7 clone were grown as previously described (Vigne et al.,
1990).
Binding Experiments.
All experiments were performed at room
temperature. Membranes (0.1-20 µg of protein/ml) were incubated in
binding buffer supplemented with 1 to 120 pM
125I-ET-1 and effectors. The binding buffer was
an Earle's salt solution (140 mM NaCl, 5 mM KCl, 0.8 mM
MgSO4, 1.8 mM CaCl2, 25 mM
HEPES, pH 7.4) supplemented with 0.05% bovine serum albumin and
protease inhibitors. After selected times of incubation, aliquots of
the incubation solutions were filtered under reduced pressure onto Sartorius 0.2-µm filters and washed three times with 3 ml of 0.1 M
MgCl2. Filters were then counted. Nonspecific
binding was measured in parallel experiments using 100 nM unlabeled
ET-1. Triplicate experiments were performed.
A first type of binding experiment was competitive binding assays. In
these experiments, the binding of a fixed concentration of
125I-ET-1 was measured in the presence of a range
of concentrations of putative inhibitors. Membranes (20 µg of
protein/ml) were incubated for 4 h with 20 pM
125I-ET-1 and different concentrations of
inhibitors in 0.7 ml of assay buffer. Specific binding represented 5000 to 8000 cpm. Nonspecific binding was 10 to 15% of the total binding component.
Saturation analysis of 125I-ET-1 binding was
carried out using a large assay volume (4 ml) and a very low protein
concentration (0.1 µg of protein/ml) in the assay. Under these
conditions, the total concentration of receptors was about 1 pM.
Membranes were incubated in the presence of a range of concentrations
of freshly prepared 125I-ET-1 (1-120 pM) in the
absence or the presence of Br2SA (0.25 or 0.75 mM). After 16 h of
incubation at room temperature, the whole incubation solution was
filtered. The nonspecific binding component was determined in parallel
incubations. Triplicate experiments were performed.
In dissociation experiments, membranes (20 µg of protein/ml) were
incubated in the presence of 15 to 25 pM
125I-ET-1. The total volume was 14 ml. After
4 h of incubation at room temperature, dissociation kinetics were
initiated by the addition of 100 nM unlabeled ET-1. Duplicate, 200-µl
aliquots were filtered after different times. In experiments using
Br2SA or I2SA, SA derivatives were added at the start of the
association process. Maximum binding was 10,000 to 12,000 cpm.
We checked that, as previously described for SA (Talbodec et al.,
2000), Br2SA did not induce a degradation of ET-1 either in the free form or in a receptor-bound form.
Association experiments were performed as described previously
(Talbodec et al., 2000). Control experiments were performed at 0.5 µg
of membrane protein/ml. This concentration was raised in experiments
using Br2SA to compensate for the decrease in the specific binding.
They were 5 and 10 µg/ml in experiments using 2 and 7 mM Br2SA,
respectively. The concentration of 125I-ET-1 used
was 9 to 20 pM. The total volume was 14 ml.
Reversibility of Actions of Br2SA.
Three samples of
membranes (40 µg of protein/ml in 14 ml of binding buffer) were
processed in parallel. Sample A was treated with 7 mM Br2SA. After 30 min of incubation at room temperature, all samples were centrifuged (10 min, 16,000 rpm) and resuspended into 14 ml of binding buffer. After 40 min at room temperature, all samples were centrifuged. Pellets were
resuspended into 14 ml of binding buffer supplemented with 20 pM
125I-ET-1. Sample B was supplemented with 7 mM
Br2SA. Association was allowed to proceed for 3 h, and the
dissociation kinetics were initiated by the addition of 100 nM
unlabeled ET-1.
Solubilization and Gel Filtration of ETA Receptors.
All
experiments were performed at 4°C. Membranes were diluted in the same
volume of solubilization buffer (300 mM NaCl, 10 mM EDTA, 4 mM EGTA,
1% digitonin, 0.76% CHAPS, 2 µM bacitracin, 0.2 mM
phenylmethylsulfonyl fluoride, 2 µM leupeptin, and 40 mM Tris-Cl, pH
7.4). After 1 h under agitation, the mixture was centrifuged at
100,000g for 30 min. The supernatant was harvested and
loaded onto an Ultrogel ACA34 gel filtration column (Sigma Chemical
Co.). The column (100 × 1.5 cm) was equilibrated in
solubilization buffer containing 0.025% digitonine and 0.1% CHAPS and
eluted with the same buffer. The column was calibrated with the
MW-GF200 kit (Sigma Chemical Co.). Two-milliliter fractions were
collected. 125I-ET-1 binding was assessed on
700-µl aliquots using 100 pM 125I-ET-1. After
90 min of incubation at room temperature, three aliquots (200 µl) of
the incubation mixture were filtered onto polyethyleneimine-treated
(0.3%) Sartorius filters. Nonspecific binding was assessed for each
fraction in parallel incubations using 100 nM ET-1.
In a second series of experiments, solubilized receptors were incubated
for 1 h with 1 to 3 nM 125I-ET-1. Bound and
free radioactivities were separated on a Sephadex G50 column. The bound
radioactivity was loaded onto the ACA34 column. Two-milliliter
fractions were collected and counted.
In experiments using Br2SA, Br2SA (2 mM) was added to the elution buffer.
Intracellular Ca2+ Measurements.
B7 cells
express ETA receptors. Their activation leads to large increases in the
intracellular Ca2+ concentration that have
previously been documented (Vigne et al., 1990, 1993). Changes in
indo-1 fluorescence ratio are conveniently monitored by flow cytometry
analysis of indo-1-loaded cells (Vigne et al., 1990, 1993). Suspended
cells were incubated for 30 min in the presence of 5 µM indo-1/AM,
centrifuged at 1000g, and resuspended into an Earle's salt
solution at a density of 106 cells/ml. ET-1 was
added to the cell suspension. After mild vortexing, tubes were inserted
into a FACS Vantage SE cytometer (Becton Dickinson). Mean
fluorescence ratios were determined for 1000 cells after different
times of exposure to ET-1. Acquisition time was <2 s. Concentration-response curves were defined from indo-1 fluorescence ratios sampled between 8 and 10 s after the addition of ET-1. This
time corresponded to the peak of the intracellular
Ca2+ transients. Fluorescence ratios were
calculated in arbitrary units set to a value of 100 for unstimulated cells.
Contraction Experiments.
Thoracic aorta from 200-g Wistar
rats were cleaned of adherent fat and cut into rings, and the
endothelium was removed. Rings were mounted under 2 g of resting
tension in organ baths (3 ml, 37°C, bubbled with a 5%
CO2 and 95% O2 gas
mixture) containing a Krebs' bicarbonate solution. The composition of
the solution was 118 mM NaCl, 4.7 mM KCl, 1.2 mM
KH2PO4, 1.2 mM
MgSO4, 25 mM NaHCO3, 2 mM
CaCl2, and 5.8 mM glucose, pH 7.4. After a 60-min equilibration during which the buffer was changed at 15-min intervals, three contraction/relaxation cycles were performed using 40 mM KCl.
Rings were then exposed to 100 nM ET-1, and contractions were allowed
to develop for 30 min. Salicylates or BQ-123 were then added without
washing the preparation. Changes in tension were recorded on a TA4000
recorder (Gould, Cleveland, OH).
Data Presentation and Statistical Analysis.
Data are given
as means ± S.E. and the number of independent experiments
performed. Equilibrium binding data were analyzed using the Ligand
software (Jandel Scientific, Corte Madera, CA). Dissociation kinetics
were linearized to yield kd, the
first order rate constant of dissociation of
125I-ET-1 receptor complexes. Association
kinetics were linearized according to a pseudo first order process to
yield k'a, the pseudo first order rate
constant of association of 125I-ET-1 to its
receptors. The second order rate constant of association (ka) was then calculated from the following
relationship: k'a = ka × [ET-1] kd. Curve fitting was performed using a
logistic equation and the Sigma Plot software (Jandel Scientific).
| |
Results |
Derivatives of Benzoic Acid and SA Inhibit 125I-ET-1
Binding to ETA Receptors.
We first used competitive binding assays
to define actions of 30 different derivatives of SA and benzoic acid.
Figure 1 shows the structures of the
major compounds used in this study. Membranes, isolated from ETA
receptor-expressing fibroblasts, were incubated in the presence of
125I-ET-1 and of different concentrations of SA
or benzoic acid derivatives, and 125I-ET-1
binding was measured after 4 h of incubation. The concentrations of derivatives that produced a 50% inhibition of the specific 125I-ET-1 binding (IC50)
were calculated and are listed in Table 1. Table 1 shows that the potency of
benzoic or SA derivatives was markedly increased by substitutions of
the aromatic ring with halogen atoms. Monohalogenated compounds were
less potent than dihalogenated compounds. Introduction of a third
heteroatom did not improve activity of the compounds. The rank order of
potency of different halogens was Br > Cl > F in the
benzoic acid series. 3,5-Dichlorosalicylic acid
(IC50 = 0.6 mM) and Br2SA
(IC50 = 0.5 mM) were equally potent. I2SA
(IC50 = 0.3 mM) was slightly more potent.
3,5-Diisopropylsalicylic acid (IC50 = 0.8 mM) was
as potent as dihalogenated derivatives of SA. These compounds were 25 to 60 times more potent than SA (IC50 = 15 mM)
and aspirin (IC50 = 20 mM). Figure
2A shows typical concentration-response
curves for the inhibition of 125I-ET-1 binding by
Br2SA, I2SA, and SA. Actions of the two most potent compounds, Br2SA
and I2SA, were analyzed in more detail.
View this table:
[in this window]
[in a new window]
|
TABLE 1
Inhibition of 125I-ET-1 binding by derivatives of benzoic acid
and SA
Concentration-response curves were established for each compound using
competitive binding assays. Each concentration-response curve was
performed using triplicates and repeated several times. Means ± S.E. are indicated wherever n was  3. Otherwise means of
two determinations are indicated.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2.
Salicylates inhibit 125I-ET-1 binding to
ETA receptors. A, competitive binding assays were performed in the
presence of 20 pM 125I-ET-1 and the indicated
concentrations of SA ( ), I2SA ( ), and Br2SA ( ). Means ± S.E. from three determinations in representative experiments are shown.
B, Scatchard plots for the specific 125I-ET-1 binding to
ETA receptors. Membranes (0.1 µg of protein/ml) were incubated in the
presence of different concentrations of 125I-ET-1 in the
absence () or the presence () of 0.25 mM Br2SA for 16 h at
room temperature, and the specific binding component was determined.
Means of triplicates are shown. Bound-to-free ratios are expressed in
pmol/mg/pM.
|
|
Br2SA Decreases the Apparent Affinity of ETA Receptors for
ET-1.
Equilibrium binding experiments were used to define the
mechanism of action of Br2SA. Figure 2B shows typical Scatchard plots for the specific 125I-ET-1 binding to ETA
receptors. 125I-ET-1 recognized a single family
of binding sites with a Kd value of 16 pM
and a maximum binding capacity of 11 pmol/mg of proteins in the absence
of Br2SA. In four independent experiments using the same membrane
preparation, the mean Kd value of
125I-ET-1 receptor complexes was 8.9 ± 2.4 pM. The maximum binding density was 11.2 ± 0.5 pmol/mg of
protein. Figure 2B further shows that Br2SA (0.25 mM) decreased the
apparent affinity of ETA receptors for 125I-ET-1
and did not change the maximum number of binding sites. In two
experiments, Br2SA (0.25 mM) increased the
Kd value of 125I-ET-1
receptor complexes 2.6- and 4.2-fold compared with the respective
controls. In two other experiments, Br2SA (0.75 mM) increased the
Kd value for
125I-ET-1 receptor complexes 7.1- and 10.0-fold.
Thus, Br2SA behaves as an apparent competitive antagonist of ET-1 binding.
Br2SA Accelerated Dissociation of 125I-ET-1 Receptor
Complexes.
Figure 3A shows typical
dissociation kinetics of 125I-ET-1 receptor
complexes. Complexes were first allowed to form for 4 h, and the
dissociation kinetics were initiated by the addition of a large excess
of unlabeled ET-1 (100 nM). Figure 3A shows that 125I-ET-1 receptor complexes dissociated very
slowly. In 10 independent experiments, the half-life of the complexes
was 5.9 ± 0.1 h. It corresponded to a
kd of 1.95 × 10
3 min1. Figure 3A
further shows that Br2SA accelerated the dissociation of
125I-ET-1 receptor complexes. In all cases,
dissociation kinetics could be fitted by monoexponentials. Figure 3B
shows the influence of different concentrations of Br2SA on the
half-life of 125I-ET-1 receptor complexes. It
shows that the action of Br2SA was saturable. The half-maximum action
of Br2SA was observed at 0.4 mM, similar to the value obtained in
competitive binding studies (0.5 mM, Table 1). The mean value of the
rate constant of dissociation of 125I-ET-1
receptor complexes obtained at near saturating concentrations of Br2SA
(3, 5, and 7 mM) was 15.3 ± 1.4 × 103 min1. It is
important to note that about 30% of 125I-ET-1
receptor complexes dissociated during a 3-h experiment under control
conditions. More than 90% of the complexes dissociated during the same
time period in the presence of 3 to 7 mM Br2SA. Finally, Fig.
4 shows that actions of Br2SA were fully
reversible.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 3.
Br2SA promoted dissociation of 125I-ET-1
receptor complexes. A, first order plots of the
dissociation of 125I-ET-1 receptor complexes. Complexes
were allowed to form for 4 h in the absence () or the presence
() of 7 mM Br2SA. Dissociation kinetics were then initiated by the
addition of 100 nM unlabeled ET-1. The means of duplicates are shown.
B, concentration-response curve for the action of Br2SA on the
dissociation of 125I-ET-1 receptor complexes. Half-lives of
125I-ET-1 receptor complexes were determined by fitting
dissociation kinetics such as those presented in A according to a first
order process.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Reversibility of actions of Br2SA. Dissociation of
125I-ET-1 receptor complexes was followed in parallel using
three pools of membrane (see Materials and Methods):
, controls without Br2SA; , controls with 7 mM Br2SA
during the association and dissociation process; and , membranes
first reacted with 7 mM Br2SA and washed.
|
|
Four additional experiments were performed using I2SA. In this series
of experiments, the half-lives of 125I-ET-1
receptor complexes were 7.3 ± 0.8 h (n = 4)
in the absence of I2SA and 2.0 ± 0.7 h (n = 4) in the presence of 0.5 mM I2SA. The same 3.6-fold increase in
dissociation rate was obtained with 1 to 2 mM Br2SA (Fig. 3B). A
2.9-fold increase in the dissociation rate of
125I-ET-1 receptor complexes is produced by 20 mM
SA (Talbodec et al., 2000). Therefore, the rank order of potency for
the action of SA derivatives is I2SA Br2SA 
SA.
Association experiments were performed in the absence or the presence
of Br2SA. The second order rate constant of association of
125I-ET-1 to its receptors
(ka) was calculated and found to be
unaffected by Br2SA at concentrations between 0.1 and 7 mM. The pooled
ka value was 11.2 × 108 ± 1.7 M1
min1 (n = 11). The equilibrium
dissociation constant of 125I-ET-1 receptor
complexes was estimated from the ka and
kd values given above
(Kd = kd/ka). It was
1.7 pM in the absence of Br2SA and 13.6 pM in the presence of a near
saturating concentration of Br2SA. The observation that Br2SA
accelerated dissociation of 125I-ET-1 receptor
complexes and had no action on the association of
125I-ET-1 to its receptor indicated that Br2SA
did not act as a simple competitive antagonist. It recognized a site
that was distinct from the ET-1-binding site.
Br2SA Modified the Properties of Interaction of ETA Receptors with
Receptor Antagonists.
The properties of the interaction of ETA
receptor with two competitive antagonists, BQ-123 and bosentan, were
defined using competitive binding assays. Figure
5A shows that BQ-123 prevented 125I-ET-1 binding to ETA receptors with an
IC50 value of 8.6 ± 1.5 nM
(n = 3). Br2SA shifted the dose-response curve to
larger concentrations. IC50 values for BQ-123
were 34 ± 12 and 130 ± 20 nM in the presence of 1 and 3 mM
Br2SA, respectively. Similar results were obtained with bosentan (Fig.
5B). Bosentan inhibited 125I-ET-1 binding with an
IC50 value of 4.6 ± 1.5 nM
(n = 5). This value increased to 190 ± 25 nM
(n = 4) in the presence of 3 mM Br2SA. Thus, Br2SA
decreased the apparent affinity of ETA receptors for BQ-123 and
bosentan. This action was expected from an allosteric type of
mechanism.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
Br2SA decreased the apparent affinity of ETA
receptors for receptor antagonists. Inhibition by BQ-123 (A) or
bosentan (B) of 125I-ET-1 binding to ETA receptors.
Experiments were performed in the absence () or the presence ()
of 3 mM Br2SA. Means of triplicates of a representative experiment are
shown.
|
|
Solubilization and Gel Filtration of ETA Receptors.
ETA
receptors were solubilized using CHAPS and digitonine. Conditions were
chosen so that tight binding could be retained. The kinetics of
dissociation of 125I-ET-1-solubilized receptor
complexes was defined. The half-life of
125I-ET-1-solubilized receptor complexes was
6.5 ± 1.1 h (n = 5). This value was similar
to that obtained for the membrane-bound receptor (5.9 ± 0.1 h). Thus, solubilized receptors retain their native properties.
Solubilized receptors were loaded onto an ACA34 gel filtration column
which fractionates globular proteins in the 20,000 to 350,000 molecular
weight range. Fractions were collected, and the specific
125I-ET-1 binding was assessed on each fraction.
Figure 6A compares the elution profiles
of ETA receptors in the absence or the presence of Br2SA. ETA receptors
eluted as a large peak that followed the void volume. Br2SA (2 mM)
shifted the profile to lower molecular weights. The molecular weight at
the peak was larger than the largest molecular weight standard used to
calibrate the column (200,000). It could not be defined with more
precision. An identical result was obtained in experiments in which
receptors were exposed to Br2SA during or after solubilization.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 6.
Gel filtration of ETA receptors. Solubilized ETA
receptors were loaded onto an ACA34 gel filtration column. A, elution
profiles of free receptors in the absence () or the presence ()
of 2 mM Br2SA. The specific 125I-ET-1 binding was assessed
on each fraction. Means of triplicates are shown. Nonspecific binding
represented less than 10% of the total binding component. B, elution
profiles of preformed 125I-ET-1 receptor complexes in the
absence () or the presence () of 2 mM Br2SA. Calibration of the
column is shown on the top of the figures. The abscissa shows elution
volumes (Ve) divided by the void volume of the column (Vo). The void
volume was determined using trypan blue.
|
|
In a second series of experiments, receptors were first solubilized,
and preformed 125I-ET-1 receptor complexes were
loaded onto the column. Figure 6B shows that the label eluted as two
distinct peaks. The second peak had the same retention time as free
125I-ET-1. The first peak eluted with an apparent
molecular weight of 169,000 ± 6,000 (n = 7),
which was smaller than that observed in the previous experiments.
Figure 6B also shows that Br2SA (2 mM) did not modify the elution of
125I-ET-1 receptor complexes. Identical elution
profiles were observed in experiments in which
125I-ET-1 receptor complexes were formed before
or after solubilization and in experiments performed at different ionic
strengths (0.15 or 0.5 mM NaCl) of the solubilization buffer. Guanosine
5'-[
-thio]triphosphate, which did not prevent
125I-ET-1 binding to ETA receptors, did not
modify the elution profile of ETA receptors. Finally, Fig. 6B shows
that the relative heights of the two peaks differed in the two
experiments. A larger fraction of the label eluted as free
125I-ET-1 when the experiments were performed in
the presence of Br2SA. This was an indication that a larger fraction of
125I-ET-1 receptor complexes dissociated on the
column in the presence of Br2SA, i.e., that Br2SA accelerated
dissociation of solubilized 125I-ET-1 receptor complexes.
Thus, free ETA receptors, Br2SA receptor complexes, and ET-1 receptor
complexes have different mobilities on a gel filtration column. The
results further suggest that Br2SA and ET-1 partially dissociate ETA
receptors from associated proteins. Actions of Br2SA and ET-1 were not additive.
Br2SA Inhibited ET-1-Induced Intracellular Ca2+
Mobilization.
A major action of ET-1 acting via ETA receptors is
to activate phospholipase C. Actions of Br2SA and I2SA on ET-1-induced intracellular Ca2+ mobilization were analyzed to
define functional consequences of the allosteric regulation of ETA
receptors. Experiments were performed using B7 cells that express
endogenous ETA receptors. Figure 7A shows
that ET-1 (100 nM) induced a transient increase in the indo-1
fluorescence ratio that peaked at 10 to 20 s and then declined. It
also shows that Br2SA (1 mM) almost completely abolished this action of
ET-1. An identical result was obtained with I2SA or with SA. The
concentration-response curves for the inhibitions by Br2SA, I2SA, and
SA of the Ca2+ mobilizing action of 30 nM ET-1
are shown in Fig. 7B. Half-maximum inhibitions were observed at
0.19 ± 0.04 mM (n = 3) Br2SA and 0.14 ± 0.05 (n = 3) I2SA. SA (IC50 = 6.7 ± 1.6 mM) was 35 times less potent than Br2SA.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 7.
Br2SA inhibited ET-1-induced intracellular
Ca2+ mobilization. A, intracellular
Ca2+ transiently induced by 100 nM ET-1 in the absence
() or the presence () of 1 mM Br2SA. Indo-1 fluorescence ratios
are expressed in arbitrary units. B, concentration-response curve for
the inhibitory actions of I2SA (), Br2SA (), and SA ( ) on the
peak Ca2+ level. The concentration of ET-1 used was 30 nM.
Means ± S.E. (n = 3) are shown.
|
|
Actions of Br2SA on the Isolated Rat Aorta.
ET-1 induces
long-lasting contractions of isolated rat aortic rings. This action of
ET-1 is mediated by ETA receptors (Marsault et al., 1993). Figure
8A presents a typical recording of the
contractile action of ET-1 on an isolated rat aortic ring. Once maximum
tension had been reached, increasing doses of Br2SA were added at
10-min intervals. Figure 8A shows that Br2SA reversed the contractile action of ET-1 in a concentration-dependent manner. The cumulative concentration-response curve is presented in Fig. 8B. Half-maximum relaxations were observed at 1 mM Br2SA. The concentration-response curves for the relaxing actions of SA or of I2SA were determined using
the same cumulative protocol. Figure 8B shows that SA was 15 times less
potent than Br2SA. I2SA was 2 times more potent than Br2SA.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 8.
Br2SA reversed the contractile action of ET-1 in rat
aortic rings. A, representative trace showing the contractile action of
100 nM ET-1 and the relaxing actions of the indicated concentrations of
Br2SA. B, cumulative concentration-response curves for the relaxing
actions of I2SA (), Br2SA (), and SA (). Means ± S.E.
(n = 6) are shown.
|
|
Figure 9 compares the relaxing actions of
Br2SA and a competitive receptor antagonist, BQ-123. Rings were
precontracted with ET-1. BQ-123 or Br2SA was then added without washing
the preparation. Figure 9 shows that BQ-123 produced complete
relaxations that developed slowly. The half-time for the relaxations
induced by BQ-123 (10 µM) was 42 ± 6 min (n = 6). Relaxations induced by 0.2 mM Br2SA were slower. To quantitate the
difference, we compared the two types of relaxations in aortic rings
prepared from the same animals. At the time at which relaxations
induced by BQ-123 were complete, Br2SA produced only a 50.2 ± 8.0% relaxation (n = 5). At the time at which BQ-123
produced a 50% reduction of tension, Br2SA produced a 21.8 ± 7.5% (n = 5) decrease in tension. Thus, Br2SA-induced
relaxations were about 2 times slower than BQ-123-induced relaxations.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 9.
Comparison of relaxing actions of BQ-123 and Br2SA.
Representative traces showing the contractile action of 100 nM ET-1 and
the relaxing actions of 10 µM BQ-123 and of 0.2 mM Br2SA are shown.
|
|
Br2SA Accelerated Relaxing Actions of BQ-123.
A major interest
of an allosteric inhibitor of ETA receptors is the possibility to
accelerate actions of receptor antagonists (Talbodec et al., 2000).
Figure 10 shows the results of an
experiment in which a rat aortic ring was first contracted with ET-1
and then exposed to 0.2 mM Br2SA for 30 min. Under these conditions, Br2SA induced a 20.4 ± 5.4% (n = 6) relaxation.
Figure 10 shows that further addition of 10 µM BQ-123 induced a rapid
and complete relaxation. Data were quantitated by measuring the
half-time of the relaxations induced by BQ-123 and Br2SA. Table
2 shows that bosentan and BQ-123 relaxed
aortic rings with identical kinetics. Br2SA acted in a
concentration-dependent manner. It was inactive at 0.1 mM and
accelerated relaxations 5-fold at 0.3 mM. Note that a mixture of 0.3 mM
Br2SA and 10 µM BQ-123 reversed half of ET-1 contractions in 7 min
only. Table 2 also shows that I2SA (0.1 mM) and SA (10 mM) potentiated
actions of BQ-123 to the same extent as 0.2 mM Br2SA. Thus, the rank
order of potency for the potentiation by SA derivatives of the relaxing
action of BQ-123 was I2SA Br2SA SA. It was identical to
that obtained in other functional experiments and in binding
experiments. Table 2 also shows that I2SA (0.1 mM) and SA (10 mM)
potentiated actions of BQ-123 and bosentan to similar extents. Their
action was thus independent of the nature of the antagonist used.
Finally, Table 2 shows that relaxations induced by mixtures of I2SA and
bosentan were independent of the order of application of the two drugs.
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 10.
Br2SA and I2SA potentiated relaxing actions of
BQ-123. A representative trace showing the effect of the combined
addition of BQ-123 and Br2SA on ET-1-induced contractions is shown. An
aortic ring was first exposed to 100 nM ET-1. Once maximum tension had
been reached, 0.2 mM Br2SA was added for 30 min. BQ-123 (10 µM) was
then added, and the relaxation was followed until completion.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2
Br2SA, I2SA, and SA potentiated relaxing actions of BQ-123 and bosentan
Rat aortic rings were exposed to 100 nM ET-1. After 30 min, maximum
tension was reached and rings were exposed to a first drug for 30 min.
The second drug was then added. Relaxations were followed until
completion, and the half-times of the relaxations were determined
graphically. Three aortic rings were prepared and tested for each
animal. Means ± S.E. and the number of animals used are
indicated.
|
|
| |
Discussion |
Inhibition by Salicylates of 125I-ET-1 Binding.
The results (Table 1) show that the hydroxyl group of salicylates
contributes little to the effect for related molecules in the benzoic
acid, and SA series are equally potent. Substitution of the aromatic
ring with halogens dramatically improves activity. Actions of halogens
in the benzoic acid series follow the following rank order of potency:
Br > Cl > F. All dihalogenated derivatives of SA are
equipotent. 3,5-Diisopropylsalicylic acid is almost as potent as
dihalogenated derivatives. These indicate that bulky groups at
positions 3 and 5 of the aromatic ring of benzoic acid or of SA favor
activity of the compounds.
Anti-platelet and anti-inflammatory actions of aspirin involve
different molecular targets. The anti-platelet action of aspirin is due
to the inhibition of cyclo-oxygenases (Patrono, 1994). Anti-inflammatory actions of salicylates are due to an inhibition of
tumor necrosis factor-induced nuclear factor-
B activation. The mechanism of this inhibition is not yet clear. It may involve an
inhibition of I-B kinase (Yin et al., 1998) or an activation of p38
mitogen-activated protein kinase (Schwenger et al., 1997, 1998;
Alpert et al., 1999). Salicylates have also been reported to inhibit or
activate c-Jun N-terminal kinases depending on the cell type (Schwenger
et al., 1997, 1999). The inhibition of cyclo-oxygenases by aspirin is
due to the acetylation of a critical serine residue (Patrono, 1994). It
is not observed with SA. The structure activity relationship for the
inhibition of I-B kinase by salicylates has not been defined in
detail. It was noticed however that 5-aminosalicylic acid, which is an
important anti-inflammatory agent used for the management of
inflammatory bowel disease, is a potent inhibitor of I-B kinase (Yan
and Polk, 1999). 5-Aminosalicylic acid was included in our screening,
but it was less active than SA or aspirin (Table 1). Thus, structure
activity relationships for the actions of SA derivatives on
cyclo-oxygenase, I-B kinase, and ETA receptors are different.
Br2SA Is an Allosteric Inhibitor of ETA Receptors.
All the
results of binding experiments are consistent with a simple type of
allosteric mechanism in which Br2SA binds to a site that is distinct
from the ET-1-binding site. First, Br2SA does not modify the
association of 125I-ET-1 to its receptors.
Second, Br2SA accelerates the dissociation of
125I-ET-1 receptor complexes. Finally, The
dose-response curve for the action of Br2SA on the dissociation of
125I-ET-1 receptor complexes is saturable.
Dissociation kinetics were monoexponential as expected if the
dissociation kinetics of the allosteric ligand are fast relative to
those of 125I-ET-1 (Lazareno and Birdsall, 1995).
The equilibrium dissociation constant of ET-1 receptor complexes,
defined from kinetic experiments, is 1.7 pM in the absence of Br2SA. It
is 13.6 pM in the presence of a near-saturating concentration of Br2SA.
A decrease in the apparent affinity of 125I-ET-1
for its receptors is also documented by a Scatchard analysis of
equilibrium binding data. The ratio of the
Kd values, measured in the presence and
absence of a near saturating concentration of Br2SA, is an estimate of
the allosteric constant ( = 8). As expected for an allosteric
inhibitor, Br2SA also decreases the apparent affinities of ETA receptor
for bosentan and BQ-123, two receptor antagonists that bind to the same
binding site as ET-1 (Clozel et al., 1993; Sakamoto et al., 1993).
Anti-ET-1 Actions of Br2SA.
SA derivatives antagonize actions
of ET-1 in B7 cells and in isolated aortic rings. The rank order of
their potency (I2SA = Br2SA SA) is similar to that obtained in
binding experiments. This suggests that anti-ET-1 properties of SA
derivatives are related to the allosteric inhibition documented in
binding experiments.
Actions of ET-1 develop at large nanomolar concentrations. The
EC50 value for ET-1-induced intracellular
Ca2+ mobilization in B7 cells is 10 nM (Vigne et
al., 1993). It is 15 nM for ET-1-induced contractions of isolated
aortic rings (Marsault et al., 1991). These are 3 orders of magnitude
larger than the Kd value of ET-1 receptor
complexes. It was therefore surprising that Br2SA inhibited actions of
large nanomolar concentrations of ET-1, whereas it decreased the
affinity of ETA receptors for ET-1 from 1.7 pM to only 13.6 pM.
One reason for this discrepancy is provided by the modeling work of
Ehlert (1988), which showed that a negative allosteric ligand can be
effective against large concentrations of a highly efficacious agonist
if the allosteric drug exhibits a large cooperativity factor and
reduces the intrinsic efficacy of the agonist receptor complex by a
large factor. Thus, anti-ET-1 actions of salicylates are consistent
with the proposed allosteric mechanism.
Influence of ET-1 and Br2SA on the Molecular Form of Solubilized
Receptors.
Affinity labeling experiments have previously been used
to estimate the molecular mass of ET receptors. The results show
molecular masses in the 32- to 70-kDa range (Sokolovsky, 1995). The
results of gel filtration experiments indicated i) that solubilized ETA receptors form complexes of a much larger mass (>200 kDa) and ii) that
binding of ET-1 alters the apparent receptor size. Similar observations
have been made for other G protein-coupled receptors such as
-adrenergic receptors (Limbird and Lefkowitz, 1978). This study
further shows that Br2SA decreased the apparent molecular weight of ETA
receptors, which could suggest a partial dissociation of the receptor
from associated proteins. Such an action would be consistent with the
possible decreased efficacy of the receptors discussed above. More
importantly, these results indicate that an action of Br2SA on ETA
receptors does not require the presence of ET-1.
SA Derivatives Accelerated Relaxing Actions of BQ-123 and
Bosentan.
We described previously that SA accelerates relaxing
actions of bosentan (Talbodec et al., 2000). This study extends these observations to I2SA, Br2SA, and BQ-123 (Table 2). This acceleration is
another consequence of the allosteric type of mechanism. Binding of
ET-1 to its receptors is almost irreversible. SA derivatives relieve
this irreversibility. They increase the probability that a bound ET-1
molecule leaves its binding site and allows antagonists to bind to the
site. These results fully support the hypothesis that actions of ET
receptor antagonists are limited by the slow rate of dissociation of
ET-1 receptor complexes. They suggest that relieving tight ET-1 binding
by allosteric inhibitors may be of therapeutic interest.
Taken together, these results provide strong evidence for the existence
of an allosteric modifier site on ETA receptors that recognizes
selected derivatives of SA. The main interest of these compounds is to
potentiate actions of receptor antagonists.
We are grateful to J. Kervella and N. Boyer for technical
assistance and to Dr. M. Clozel for the generous gift of bosentan.
This work was supported by the Centre National de la Recherche
Scientifique and the Fondation de France.
ET-1, endothelin-1;
SA, salicylic acid;
Br2SA, 3,5-dibromosalicylic acid;
I2SA, 3,5-diiodosalicylic acid;
BQ-123, cyclo[D-Trp-D-Asp-Pro-D-Val-Leu];
CHAPS, 3-[(3-cholamidopropyl)dimethyl-ammonio]-1-propane sulfonate.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
