Institute of Experimental and Clinical Pharmacology and Toxicology
Medical University of Lübeck, Lübeck, Germany
 |
Introduction |
Clonidine
and the related compounds moxonidine and rilmenidine that induce
inhibition of noradrenaline release from sympathetic neurons are second
line antihypertensives. Recently, the presynaptic 
2-adrenoceptor mediating an inhibition of
noradrenaline release from sympathetic nerves at which these compounds
act was subclassified as the 2A- and
2C-adrenoceptor (Altman et al., 1999
). In
addition, clonidine and other imidazolines have been suggested to
modulate noradrenaline release via
non-I1-/non-I2-presynaptic
imidazoline binding sites, which were identified on the sympathetic
axon terminals of rabbit, rat, guinea pig, and human cardiovascular
tissue (Göthert et al., 1999; Molderings and Göthert,
1999). In contrast, investigations by other authors revealed that
these drugs inhibit noradrenaline release exclusively by activating
prejunctional 2-adrenoceptors (Bohmann et al.,
1994; Gaiser et al., 1999). Whether imidazoline binding sites are
involved (in addition to 2-adrenoceptors) in noradrenaline release may be dependent on stimulation conditions (Molderings et al., 1999a) and/or species differences (Molderings et
al., 2000).
Such results have been based mainly on in vitro studies appropriate for
the identification of a drug's presynaptic site of action. However, to
what extent do such in vitro results apply in vivo? Studies in
conscious and pithed rabbits have supported the view that central
sympathetic inhibition is mediated only via
2-adrenoceptors (Urban et al., 1995; Bock et
al., 1999; Szabo et al., 1999). In pithed, spontaneously hypertensive
rats (SHR) and rabbits, rilmenidine and moxonidine decreased the
stimulated overflow of noradrenaline (Häuser et al., 1995; Urban
et al., 1995; Szabo et al., 1999). Because imidazoline derivatives were still able to reduce noradrenaline overflow dose dependently after rauwolscine pretreatment, an
2-adrenoceptor-independent mechanism was
suggested (Häuser et al., 1995). However, it seems not unlikely that clonidine, moxonidine, or rilmenidine induces sympathetic inhibitory effects via 2-adrenoceptors even in
the presence of rauwolscine, because this
2-blocker was characterized to be a competitive antagonist (Bock et al., 1999), which would strengthen the
idea of an imidazoline binding site-independent regulation of
noradrenaline release. The best way to show in vivo whether imidazoline
binding sites would have some impact in regulating noradrenaline
release is to use 2-adrenergic receptor
knockout (KO) mice (Hein, 2001). However, the residual
2-mediated effect in the
2A/D-KO mice suggests that the
2C-adrenoceptor also functions as a
presynaptic autoreceptor for inhibiting noradrenaline release (Altman
et al., 1999). Because only the 2A/C-double KO
mouse (an animal model that was not available to us) would serve as a
suitable experimental tool to answer this question, we decided to
overcome the limitation of a competitive blockade of
2-adrenoceptors by performing experiments on
reduced noradrenaline overflow evoked by clonidine-like substances
(e.g., moxonidine) under irreversible 2-adrenoceptor blockade in pithed SHR. This
rat strain is a well established model of sympathetic hyperactivity,
which represents another rationale for using SHR instead of mice.
Using an irreversible blocker of
2-adrenoceptors (phenoxybenzamine) and a
specific ligand for imidazoline binding sites
(2-endo-amino-3-exo-isopropylbicyclo[2.2.1]heptane; AGN192403), the aim of this study was to determine 1) whether imidazoline binding sites contribute to the moxonidine-induced modification of noradrenaline release in vivo and, if so, 2) to specify
the subtype of imidazoline binding sites involved in this effect.
| |
Materials and Methods |
Animal Preparation.
The present study was conducted
according to the declaration of Helsinki, following the guidelines for
the care and use of laboratory animals as adopted by the Ministerium
für Natur und Umwelt des Landes Schleswig Holstein, Deutschland,
animal protocol no. 9/A4/91. Male, spontaneously hypertensive rats
(Charles River, Sulzfeld, Germany), weighing 200 to 250 g, were
pithed under ether anesthesia using a steel rod (1.5 mm diameter)
coated with enamel except for the length of the thoracolumbar spinal
cord (Th4-Th12 segments) as described by Gillespie and Muir (1967). A
steel cannula was inserted as an indifferent electrode into the dorsal
subcutis located near the lumbar vertebral column. Both vagal nerves
were cut at the neck, and neuromuscular junctions were blocked by
d-tubocurarine (3 mg/kg). Polyethylene catheters were
inserted into both femoral veins (PE-10) for drug administration and
into both carotid arteries (PE-50) for measuring blood pressure and
collecting blood samples. The polyethylene catheter (PE-50), which was
inserted into the left carotid artery, was connected to a Statham P23
Db pressure transducer (Hellige, Freiburg, Germany). Blood pressure and
heart rate were recorded continuously and sampled digitally.
Influence of Moxonidine, Efaroxan, and Idazoxan on Blood Pressure
Response.
Dose-response curves for blood pressure were generated
for moxonidine, efaroxan, and idazoxan over concentrations ranging between 0.1 and 10,000 µg/kg by cumulative bolus injections. The ED50 values for moxonidine's dose-response
curves were calculated. All subsequent stimulation experiments were
performed using 3, 10, or 30 times the ED50 value
of moxonidine.
Stimulation Experiments.
Electrical stimulation of the
thoracolumbar portion of the spinal cord was performed at 10 V (1-ms
pulse duration at 0.5 Hz for 3 min), applied via the pithing rod. After
preparation, pithed SHR were allowed to recover until blood pressure
and heart rate were constant. Before any stimulation, all rats were
pretreated with desipramine (0.5 mg/kg) to inhibit neuronal
catecholamine uptake and, when appropriate, phenoxybenzamine (3 or 10 mg kg
1) to block
2-adrenoceptors. Depending on the protocol,
pretreatment of SHR also included a short-lasting infusion within
30 s of either efaroxan (0.1 or 1.0 mg/kg) or idazoxan (0.01 or
0.1 mg/kg). Thereafter, either moxonidine or solvent (as a control) was
infused over 10 min. Seven minutes after starting the infusion, rats
were stimulated for 3 min as described above. During the last 30 s
of stimulation, blood (1 ml) was sampled from the carotid artery. The
loss in volume was compensated for by the infusion of hydroxyethyl
starch (6%; Fresenius, Homburg, Germany) (1 ml) over 5 min.
Before infusing a higher dose of moxonidine, animals were allowed to
recover for at least 10 min.
In another set of experiments, the influence of agmatine on
noradrenaline overflow was tested. Wherever appropriate, animals were
pretreated 10 min before agmatine infusion (6, 33, or 60 mg/kg) either
with phenoxybenzamine (10 mg/kg, infusion within 30 s) or
AGN192403 (10 mg/kg, infusion within 30 s) to block
2-adrenoceptors or
I1-binding sites, respectively. Within the last
30 s of the 2-min agmatine infusion, global sympathetic outflow
(C7-L3) was induced by preganglionic electrical stimulation (20 V, 1-ms
pulse duration at 0.5 Hz for 0.5 min) via the steel rod before blood (1 ml) was sampled for noradrenaline analysis.
Noradrenaline Determination.
Blood samples were centrifuged
at 4°C for preparing plasma. Plasma noradrenaline as an index of
sympathetic overflow was measured by high-performance liquid
chromatography and electrochemical detection after plasma (350 µl) was adsorbed onto alumina in a Tris-buffer system (700 µl,
consisting of 1.5 M tris[hydroxy-methyl]aminomethane hydrochloride,
68 nM EDTA, and 3.6 mM glutathione) and eluted with 100 µl of 0.1 M
perchloric acid using 500 pg of dihydroxybenzylamine as an internal standard.
Substances.
Moxonidine was a generous gift from Solvay
(Hanover, Germany). Efaroxan, idazoxan, and phenoxybenzamine were
obtained from Sigma/RBI (Natick, MA); AGN192403 was from Tocris
(Bristol, UK); and agmatine, D-tubocurarine, and
desipramine were from Sigma (Deisenhofen, Germany). All other chemicals
(HPLC or analytical grade) were purchased either from Sigma Chemie
(Deisenhofen, Germany) or Merck (Darmstadt, Germany).
Statistical Analysis.
Data shown in tables and figures are
expressed as means ± S.E.M. ED50 values
were calculated by nonlinear curve fitting (GraphPad Prism; GraphPad
Software, Inc., San Diego, CA). Statistical analysis was performed by
one-way analysis of variance followed by appropriate post hoc tests
(Bonferroni's multiple comparison test). A significance level of 0.05 or less was considered to represent a statistically significant difference.
| |
Results |
Effects of Drugs on the Blood Pressure and Heart Rate of Pithed
SHR.
In the following experiments, both diastolic and systolic
blood pressure were monitored via a carotid catheter. Changes induced by various drugs were similar for diastolic and systolic blood pressure. For reasons of clarity and because changes in blood pressure
of pithed SHR are related more to peripheral resistance (which is
reflected more by the diastolic blood pressure), only this parameter is
depicted in the following figures. Dose-response curves of moxonidine,
efaroxan, and idazoxan were determined for their increasing effect on
diastolic blood pressure (Fig. 1A): the
ED50 values were 60 ± 15, 28 ± 9, and
16 ± 9 µg/kg, respectively. The maximal increase
(Emax) of diastolic blood pressure was
136 ± 5 mm Hg using moxonidine, but considerably less in the case of efaroxan (24 ± 2 mm Hg) or idazoxan (30 ± 8 mm Hg). The
heart rate decreased slightly under moxonidine infusion, whereas
efaraxon and idazoxan actually increased the heart rate slightly (Fig. 1B). All subsequent stimulation experiments were performed using moxonidine doses 3, 10, and 30 times the ED50
values leading to maximal or submaximal effects on blood pressure
[slope of the moxonidine dose-response curve
(nH): 1.53 ± 0.16)].
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1.
Dose-response curves for moxonidine ( ), efaroxan
( ), and idazoxan ( ) on diastolic blood pressure (DBP; A) and
heart rate (B) in pithed spontaneously hypertensive rats. Values are
expressed as means ± S.E.M. of 8 to 10 experiments. *,
p < 0.05, versus controls.
|
|
Effectiveness of the Phenoxybenzamine-Induced
2-Adrenoceptor Blockade.
In a further set of
experiments in pithed SHR, the plasma noradrenaline concentration after
electrical stimulation of preganglionic sympathetic nerves (designated
as noradrenaline overflow in this and subsequent sections) was
determined. Plasma noradrenaline concentrations after electrical
stimulation without phenoxybenzamine pretreatment were 95 ± 8.4 pg/ml (Fig. 2A). These plasma
noradrenaline levels were increased 5- to 6-fold after blocking
2-adrenoceptors with phenoxybenzamine (3 or 10 mg/kg). Higher phenoxybenzamine doses did not further increase
noradrenaline overflow (Fig. 2A). Evoked noradrenaline overflow
remained unchanged at the four stimulation periods (Fig. 2C). The
stimulation-evoked elevation of diastolic blood pressure without
phenoxybenzamine pretreatment was 16.5 ± 2.7 mm Hg and tended to
decrease over the four stimulation periods (Fig. 2D). No
stimulation-evoked elevation of diastolic blood pressure was observed
in the experiments with phenoxybenzamine (3, 10, and 30 mg/kg; Fig. 2,
B and D). Both findings, namely, the lack of any further increase in
noradrenaline overflow or blood pressure response when using
phenoxybenzamine doses >10 mg/kg, clearly rule out the presence of any
active 2-adrenoceptors after phenoxybenzamine
treatment.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 2.
Influence of the phenoxybenzamine dosage on plasma
noradrenaline overflow (A) and stimulation-dependent increase in
diastolic blood pressure (DBP; B) in pithed SHR. C, influence of four
stimulation periods (10 V, 0.5 Hz, 1 ms, and 3 min) on plasma
noradrenaline overflow (C) and DBP (D) in the absence of
phenoxybenzamine () or after pretreatment with 3 ( ) or 10 ( )
mg/kg of the drug. Values are expressed as means ± S.E.M. of 8 to
10 experiments. *, p < 0.05, compared with
phenoxybenzamine-free controls;  , p < 0.05, compared with 3 mg/kg phenoxybenzamine.
|
|
Influence of Moxonidine on Stimulated Noradrenaline Overflow.
The electrically stimulated plasma noradrenaline concentration without
phenoxybenzamine treatment was 87.6 ± 8.2 pg/ml (Fig. 3A), consistent with values (95.4 ± 8.8 pg/ml) obtained from experiments where the dependence of plasma
noradrenaline concentration on phenoxybenzamine dose was demonstrated
(Fig. 2A). This level was almost halved by moxonidine under all dose
regimes. The elevation of electrically stimulated plasma noradrenaline
levels under 2-adrenoceptor blockade was
reduced by moxonidine dose dependently to values (96.4 pg/ml) similar
to those observed in phenoxybenzamine- and moxonidine-free
animals (87.6 ± 8.2 pg/ml; Fig. 3A). Complete 2-adrenoceptor blockade was confirmed by
the lack of blood pressure response, because the moxonidine- and
stimulation-dependent increase in diastolic blood pressure was
completely attenuated in the presence of phenoxybenzamine (Fig. 3B).
Under control conditions, moxonidine induced a dose-dependent increase
of diastolic blood pressure in pithed SHR compared with controls. The
maximal increase was 86 ± 15 mm Hg after 1.8 mg/kg moxonidine.
This increase was abolished by phenoxybenzamine.
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 3.
Influence of moxonidine on stimulated plasma
noradrenaline concentrations (PNA; 10 V, 0.5 Hz, 1 ms, and 3 min; A)
and stimulation-evoked elevation of diastolic blood pressure (DBP; B)
in the absence () or presence of 10 mg/kg () phenoxybenzamine.
Values are expressed as means ± S.E.M. of 8 to 10 experiments.
*, p < 0.05, compared with moxonidine-free
controls; , p < 0.05, compared with
phenoxybenzamine-treated animals.
|
|
Influence of Efaroxan and Idazoxan on the Stimulated Noradrenaline
Overflow.
In the presence of phenoxybenzamine (10 mg/kg), the
noradrenaline concentration after preganglionic electrical stimulation was 454 ± 40 pg/ml; this was increased 3.7-, 4.7-, and 6.1-fold after efaroxan at 0.1, 1, and 3 mg/kg, respectively (Fig.
4). Idazoxan at 0.01 and 0.1 mg/kg had
similar effects on stimulated noradrenaline overflow because
electrically stimulated noradrenaline plasma concentrations were
increased by respective factors of 2 and 3 compared with
phenoxybenzamine controls (Fig. 4). The stimulation-evoked increase in
diastolic blood pressure in the presence of phenoxybenzamine was
4.4 ± 2.7 mm Hg, and did not change in the presence of efaroxan
or idazoxan. Heart rate (417 beats/min in the absence of efaroxan or
idazoxan) also remained unchanged after efaroxan or idazoxan treatment.
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 4.
Influence of efaroxan () and idazoxan () on the
stimulation-dependent (10 V, 0.5 Hz, 1 ms, and 3 min) plasma
noradrenaline concentrations (PNA) in pithed SHR that were pretreated
with desipramine (0.5 mg/kg) and phenoxybenzamine (10 mg/kg). Values
are expressed as means ± S.E.M. of 5 to 10 experiments. *,
p < 0.05, versus control.
|
|
Influence of Moxonidine on Noradrenaline Overflow under Blockade of
2-Adrenoceptors and I1-Binding Sites.
After administration of efaroxan (0.1 µg/kg) and phenoxybenzamine (10 mg/kg), the noradrenaline plasma concentration after electrical
stimulation was 1728 ± 65 pg/ml, and this was markedly reduced by
moxonidine compared with controls (Fig.
5A). SHR serving as time controls
(pretreated with phenoxybenzamine and efaroxan, but not with
moxonidine) showed almost constant noradrenaline plasma concentrations
at control levels over the four stimulation periods (data not shown).
When the efaroxan dosage was enhanced by a factor of 10 to 1 mg/kg, the
curve depicting the moxonidine-evoked reduction of plasma noradrenaline
concentration was shifted upward because values differed significantly
from those achieved after low-dosage efaroxan pretreatment (Fig. 5A).
Blood pressure in all animals included in this protocol was not
affected by efaroxan and moxonidine (Fig. 5B).
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5.
Influence of moxonidine on electrically (10 V, 0.5 Hz, 1 ms, and 3 min) stimulated plasma noradrenaline concentrations
(PNA; A) and the stimulation-evoked elevation of diastolic blood
pressure (DBP; B) in pithed SHR after treatment with efaroxan (, 0.1 mg/kg;  , 1 mg/kg) and phenoxybenzamine (10 mg/kg). Values are
expressed as means ± S.E.M. of 5 to 10 experiments. *,
p < 0.05, compared with moxonidine-free controls;
, p < 0.05, compared with low-dose (0.1 mg/kg)
efaroxan.
|
|
Noradrenaline plasma concentrations after pretreatment with idazoxan
(0.01 µg/kg) and phenoxybenzamine (10 mg/kg) and electrical stimulation were 905 ± 80 pg/ml (Fig.
6A). These plasma noradrenaline levels
were not altered in time-matched controls during four stimulation periods within a total period of 80 min (data not shown). However, noradrenaline plasma concentrations were reduced by moxonidine to below
200 pg/ml (Fig. 6A). Similar to efaroxan experiments, the curves for
moxonidine-induced reduction of noradrenaline plasma levels were
shifted upwards when the idazoxan dosage (0.1 mg/kg) was increased
(Fig. 6A). The blood pressure response also remained unaffected by
idazoxan and/or moxonidine in these experiments (Fig. 6B).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 6.
Influence of moxonidine on electrically (10 V, 0.5 Hz, 1 ms, and 3 min)-stimulated plasma noradrenaline concentrations
(PNA; A) and the stimulation-evoked elevation of diastolic blood
pressure (DBP; B) in pithed SHR after treatment with idazoxan (,
0.01 mg/kg; or , 0.1 mg/kg) and phenoxybenzamine (10 mg/kg). Values
are expressed as means ± S.E.M. of 8 to 10 experiments. *,
p < 0.05, compared with moxonidine-free controls;
, p < 0.05, compared with low-dose idazoxan
(0.01 mg/kg).
|
|
Influence of Agmatine on the Stimulated Noradrenaline
Overflow.
The last set of experiments on electrically stimulated
noradrenaline release in pithed SHR was performed using agmatine. In desipramine-free animals, agmatine (33 mg/kg) did not influence noradrenaline plasma concentrations on its own (187 ± 35 pg/ml; data not shown), but did abolish the stimulated blood pressure increase
(59.1 ± 6.3 mm Hg). In contrast, with
desipramine/phenoxybenzamine (0.5/10 mg/kg) pretreated pithed SHR, the
electrically stimulated plasma noradrenaline concentrations were
elevated by a factor of 4.8 compared with basal levels (585 ± 47 pg/ml). Under the same conditions, additional agmatine (6 and 60 mg/kg)
significantly reduced the stimulated noradrenaline plasma
concentrations by 36 and 51%, respectively (Fig.
7A). Blood pressure in controls was
moderately decreased by phenoxybenzamine pretreatment and was not
affected by agmatine at either dose applied (Fig. 7B). Heart rate was
increased and not affected by agmatine (Fig. 7C). Using desipramine
(0.5 mg/kg) and AGN192403 (10 mg/kg), the basal noradrenaline plasma
concentrations (151 ± 21 pg/ml) were approximately one-fourth of
those seen with phenoxybenzamine pretreatment (Fig. 7A). Preganglionic
stimulation increased noradrenaline overflow by a factor of 3, resulting in stimulated noradrenaline plasma levels that were only 15 to 20% of those observed with phenoxybenzamine pretreatment. In
contrast to the phenoxybenzamine experiments, stimulated plasma
noradrenaline concentrations were unaffected by agmatine at both
applied doses (Fig. 7A). Stimulation-evoked blood pressure increased
dramatically in response to AGN192403 pretreatment (ca. 100 mm Hg),
which was reduced tendentiously by low-dose, but significantly (73%)
by high-dose agmatine (Fig. 7B). Similar to phenoxybenzamine treatment,
the stimulation-dependent increase in heart rate was not affected by
AGN192403 (Fig. 7C).
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 7.
Influence of agmatine on plasma noradrenaline
concentration (PNA; A), diastolic blood pressure (DBP; B), and heart
rate (C) in the presence (filled symbols) or absence (open symbols) of
preganglionic stimulation (20 V, 0.5 Hz, 1 ms, and 0.5 min) in pithed
SHR, which were pretreated either with desipramine/phenoxybenzamine
(0.5/10 mg/kg; circles) or desipramine/AGN192403 (0.5/10 mg/kg;
squares). Values are expressed as means ± S.E.M. of 8 to 10 experiments. *, p < 0.05, compared with
unstimulated agmatine-free controls; , p < 0.05, compared with stimulated agmatine-free controls.
|
|
| |
Discussion |
Verification of an Effective 2-Blockade.
To
support the concept that imidazoline binding sites play an important
role in addition to 2-adrenoceptor in the
control of sympathetic neurotransmission, experiments on noradrenaline overflow were performed in pithed SHR. From a methodological point of
view, the efficacy of presynaptic adrenoceptor blockade with phenoxybenzamine was established in preliminary studies. The adequacy of 2-blockade was clearly demonstrated in our
study, because noradrenaline overflow was not enhanced any further when
phenoxybenzamine concentrations >10 mg/kg were applied. In addition,
the absence of any stimulation-evoked elevation of blood pressure after
10 mg/kg phenoxybenzamine pretreatment is considered as evidence for a
complete blockade of all 2-adrenoceptors, even
though an increase in blood pressure due to noradrenaline release is
attributed primarily to 1-adrenoceptors and
postsynaptically located 2-adrenoceptors. Finally, moxonidine as an 2-agonist did not
increase blood pressure in the presence of phenoxybenzamine in pithed
rats (Fig. 4), which confirms the assumption that all
2-adrenoceptors were blocked by 10 mg/kg
phenoxybenzamine. A limitation associated with using phenoxybenzamine
is that some consider it as an inhibitor for imidazoline binding sites
(Molderings et al., 1991). However, we feel convinced that
phenoxybenzamine does not occupy all imidazoline binding sites, because
the mixed
2-/I1-antagonists
efaroxan and idazoxan are still able to increase noradrenaline release
in its presence.
Identification of an I1-Dependent Mechanism Regulating
the Noradrenaline Overflow.
Even though all presynaptic
2-adrenoceptors were irreversibly blocked by
phenoxybenzamine, moxonidine was still able to reduce noradrenaline
overflow (Fig. 3A). Due to moxonidine's affinity toward imidazoline
binding sites (Ernsberger et al., 1993) and its lack of affinity toward
other receptors involved in regulating noradrenaline release (Wethmar
et al., 2001), moxonidine's diminishing effect on noradrenaline
overflow must arise via interaction with imidazoline binding sites, as
other in vitro studies have suggested (Göthert et al., 1999;
Molderings and Göthert, 1999). However, this conclusion conflicts
sharply with in vitro findings, i.e., that the
I1-agonist moxonidine (unlike clonidine) inhibits
noradrenaline release from sympathetic nerve terminals via presynaptic
located 2-adrenaoceptors and not via the
non-I1-/non I2-binding
sites (Likungu et al., 1996).
For excluding nonspecific effects that may have been involved,
confirming our moxonidine observations, and manifesting the likelihood
of an I1-dependent mechanism in regulating
noradrenaline overflow, we tested whether the mixed
2-/I1-antagonists
efaroxan and idazoxan (Bock et al., 1999), which both counteract the
hypotensive effects of moxonidine and rilmenidine (Ernsberger et al.,
1990; Haxhiu et al., 1994), could reveal effects opposite to those of moxonidine. Continuing on with the moxonidine results, efaroxan as well
as idazoxan (Fig. 4) increased noradrenaline overflow dose dependently
during 2-adrenoceptor blockade, strengthening the idea of antagonism between efaroxan and moxonidine at
I1-binding sites regarding noradrenaline release.
Such a picture regarding noradrenaline release is paralleled when
focusing on presynaptic 2-adrenoceptors: a
stimulation for example by the agonist moxonidine causes a decrease,
whereas an inhibition by rauwolscine or phenoxybenzamine increases
noradrenaline release (Häuser et al., 1995). However, our
findings differ from in vitro findings in that the electrically evoked
[3H]noradrenaline overflow in the presence of
phenoxybenzamine or rauwolscine was either diminished (Göthert
and Molderings, 1991; Molderings et al., 1997) or remained unchanged
(Molderings and Göthert, 1998), a fact suggesting different
modes/sites of action in in vivo and in vitro situations (see above for discussion).
Consistent with the perplexing findings with moxonidine and efaroxan
regarding noradrenaline overflow, moxonidine markedly diminished the
efaroxan- or idazoxan-evoked increases in stimulated plasma
noradrenaline concentration even under
2-blockade (Figs. 5A and 6A). The specificity
of this observation is emphasized by the fact that the
moxonidine-induced reduction of stimulated noradrenaline overflow was
affected differently by using two doses of efaroxan and idazoxan. This
reconfirms the proposed antagonism between moxonidine and idazoxan or
efaroxan concerning their actions at I1-binding
sites (as suggested by others) and its regulating influence regarding
noradrenaline release (Chu et al., 1997) or central blood pressure
control (Ernsberger et al., 1990; Haxhiu et al., 1994).
To strengthen the putative role of imidazoline binding sites in
noradrenaline release in our in vivo model, agmatine, an endogenous ligand for imidazoline binding sites (Li et al., 1994; Raasch et al.,
2001), was investigated regarding its potency in regulating noradrenaline overflow. Electrical stimulation caused an elevation of
plasma noradrenaline in phenoxybenzamine- or AGN192403-pretreated animals, and which was reflected in both experiments as an increase in
heart rate. The magnitude of noradrenaline overflow in the presence of
AGN192403 was clearly lower than that induced by
2-blockade. However, this is consistent with
other findings (Munk et al., 1996), where AGN192403 was first
characterized as an I1 ligand without any
intrinsic activity. Noradrenaline overflow was markedly decreased by
agmatine when 2-adrenoceptors were blocked,
clearly emphasizing the relevancy of imidazoline binding sites.
Confirming this mode of action, noradrenaline overflow was not affected
by agmatine in the presence of AGN192403, because imidazoline binding sites were blocked and agmatine itself only had a low affinity toward
2-adrenoceptors (Li et al., 1994). After
injection of AGN192403, a pronounced stimulation-evoked blood pressure
increase could be observed, which was blunted by phenoxybenzamine due
to 1/2-blockade. In AGN192403-treated animals,
agmatine reduced the stimulation-dependent blood pressure increase only
at a high concentration, probably as a consequence of an interaction
with a vascular I3-binding site that was recently
shown to regulate vasoconstriction (Minyan et al., 2001). Our findings
regarding agmatine's influence on noradrenaline overflow reinforce the
I1-binding site-mediated mechanism.
Probable Mechanisms Underlying the Changes in Stimulated
Noradrenaline Overflow.
Because our results clearly reveal an
imidazoline binding site-dependent mechanism for regulating the
stimulated noradrenaline overflow, it remained to be seen what
imidazoline binding site subtype was involved in noradrenaline release.
From our data on moxonidine, efaroxan, and AGN192403, we have clear
evidence that I1-binding sites might be involved.
This conflicts sharply with in vitro findings, whereby the
presynaptically located
non-I1-/non-I2-imidazoline binding site was characterized as regulating the release of
noradrenaline as an autoreceptor (Molderings et al., 1991; Molderings
and Göthert, 1995). This discrepancy indicates that the
non-I1-/non-I2-binding sites are presumably not involved in mediating the in vivo observations seen in this study, a hypothesis which is also confirmed by effects seen with efaroxan and idazoxan, which were classified as agonists for
the non-I1-/non-I2-binding
site by Göthert et al. (Göthert and Molderings, 1991;
Molderings et al., 1997), but which showed antagonistic features in our
experiments and elsewhere (Chu et al., 1997).
So where are the I1-binding sites that seem to be
involved located? First, a ganglionic effect can be hypothesized,
because I1-binding sites are present at the cell
bodies of sympathetic ganglia and adrenal medulla (Molderings et al.,
1993). In addition, the observed effects of agmatine may also be due to
a ganglionic effect, because 1) agmatine suppresses the
nicotinic-cholinergic transmission in sympathetic ganglia due to a
blockade of nicotine at the nicotinic acetylcholine (nAch) receptor
(Quik, 1985; Loring, 1990) and because 2) it was shown that ganglionic
excitation contributes markedly to catecholamine release in a whole
animal model (Dendorfer et al., 2002).
However, some results have characterized the
I1-binding sites as an excitatory binding site,
because it releases atrial natriuretic peptide (Gutkowska et al., 1997;
Mukaddam-Daher et al., 1997) and prostaglandins (Ernsberger et al.,
1995). If this is right, activation of such receptors would evoke an
increase in noradrenaline release rather than a decrease as was found
in this study. In view of this, another hypothetical mechanism may
involve an effect on noradrenaline release mediated via
I1-binding sites. Because stimulation of
ganglionic I1-binding sites releases
prostaglandins (Ernsberger et al., 1995) and histamine (Molderings et
al., 1999b), which were both shown to diminish noradrenaline release
from the sympathetic nerves of cardiovascular and other tissues (Starke and Montel, 1973; Levi and Smith, 2000), it is likely that agmatine reduces noradrenaline overflow via such an indirect mechanism. Finally,
nicotine was shown to increase catecholamine release by activating nAch
receptors localized on peripheral postganglionic sympathetic nerve
endings and in the adrenal medulla (Anden et al., 1986). It therefore
seems feasible that agmatine reduces ganglionic, cholinergically evoked
noradrenaline release (Yokotani et al., 2000), because agmatine
inhibits nicotinic-cholinergic transmission in sympathetic ganglia
(Loring, 1990). Clonidine's potency at inhibiting an
I1-binding site-mediated activation of nicotinic
channels confirms this hypothesis (Musgrave et al., 1995). However, all
these indirect mechanisms must be confirmed/excluded by further studies
involving H3 and nAch receptor antagonists or
cyclooxygenase inhibitors.
In summary, we have demonstrated in vivo that imidazoline binding sites
participate in noradrenaline release whereby our data indicate an
I1-binding site-mediated mechanism. Whether these I1-binding sites are located presynaptically on
the sympathetic nerve terminals seems doubtful considering findings
from isolated preparation experiments. Whether an interaction with the
ganglionic cholinergic, the histaminergic, or the prostaglandin system
contributes to the I1-binding site-mediated
noradrenaline release needs to be clarified in future studies.
We thank Dr. J. P. Keogh for linguistic assistance in
preparing the manuscript and A. Kaiser for technical assistance.
SHR, spontaneously hypertensive rats;
KO, knockout;
PE, polyethylene;
nAch, nicotinic acetylcholine.
![[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}
2-Adrenoceptors
Top
Abstract
Introduction
36 or 53%), even under
