A circulating substance cross-reacting with antiimidazoline antibodies. Detection in serum in relation to essential hypertension.

It has been shown in various mammal species that clonidine, a well known centrally acting hypotensive agent, acts through the activation of imidazoline receptors (IRs) in the nucleus reticularis lateralis (NRL) of the brainstem. Specific binding sites sensitive to imidazolines and insensitive to catecholamines have been detected in rat and bovine, as well as human brains. An endogenous ligand, other than catecholamines, should exist for these IRs. Such a ligand could play a role in the pathophysiology of human essential hypertension. Therefore, we developed two RIAs with polyclonal and monoclonal anticlonidine antibodies. These antibodies presented specificity spectra similar to that of the IRs: they bound imidazolines and not catecholamines at all. These RIAs were used to detect imidazoline-like immunoreactivity in the human serum. Immunoreactive substance was measured in 26 normotensive subjects' sera, and specificity of interaction between antibodies and sera was verified. None of the known endogenous substances tested so far were able to interact with the two antibodies. Immunoreactivity in 32 essential hypertensive patients' sera proved higher in approximately 30% of cases. Values of immunoreactivity positively correlated with the mean arterial pressure values. This study demonstrates the existence of an "imidazoline-like" immunoreactive substance in the human serum with high levels in some hypertensive patients.

Integrin α5ß1 and p53 convergent pathways in the control of anti-apoptotic proteins PEA-15 and survivin in high-grade glioma.

Integrin α5ß1 expression is correlated with a worse prognosis in high-grade glioma. We previously unraveled a negative crosstalk between integrin α5ß1 and p53 pathway, which was proposed to be part of the resistance of glioblastoma to chemotherapies. The restoration of p53 tumor-suppressor function is under intensive investigations for cancer therapy. However, p53-dependent apoptosis is not always achieved by p53-reactivating compounds such as Nutlin-3a, although full transcriptional activity of p53 could be obtained. Here we investigated whether integrin α5ß1 functional inhibition or repression could sensitize glioma cells to Nutlin-3a-induced p53-dependent apoptosis. We discovered that α5ß1 integrin-specific blocking antibodies or small RGD-like antagonists in association with Nutlin-3a triggered a caspase (Casp) 8/Casp 3-dependent strong apoptosis in glioma cells expressing a functional p53. We deciphered the molecular mechanisms involved and we showed the crucial role of two anti-apoptotic proteins, phosphoprotein enriched in astrocytes 15 (PEA-15) and survivin in glioma cell apoptotic outcome. PEA-15 is under α5ß1 integrin/AKT (protein kinase B) control and survivin is a p53-repressed target. Moreover, interconnections between integrin and p53 pathways were revealed. Indeed PEA-15 repression by specific small-interfering RNA (siRNA)-activated p53 pathway to repress survivin and conversely survivin repression by specific siRNA decreased α5ß1 integrin expression. This pro-apoptotic loop could be generalized to several glioma cell lines, whatever their p53 status, inasmuch PEA-15 and survivin protein levels were decreased. Our findings identify a novel mechanism whereby inhibition of α5ß1 integrin and activation of p53 modulates two anti-apoptotic proteins crucially involved in the apoptotic answer of glioma cells. Importantly, our results suggest that high-grade glioma expressing high level of α5ß1 integrin may benefit from associated therapies including integrin antagonists and repressors of survivin expression.

New concepts on the central regulation of blood pressure. Alpha 2-adrenoceptors and "imidazoline receptors".

The most usual hypothesis to explain the central hypotensive effect of clonidine-like substances was to admit that these drugs stimulated alpha 2-adrenoceptors within the brainstem. Now it has been demonstrated that neither the endogenous ligand to the alpha-adrenoceptors, noradrenaline, nor any other catecholamine or phenylethylamine was hypotensive in the medullary nucleus reticularis lateralis, where all imidazolines proved to be such. Recently, a membrane receptor population sensitive to clonidine and insensitive to catecholamines was described within the nucleus reticularis lateralis; this subgroup of receptors represented 20 to 30 percent of the [3H]clonidine binding sites in the bovine nucleus reticularis lateralis and 100 percent within the human nucleus reticularis lateralis region. Thus, the existence of such imidazoline specific receptors was clearly established and the endogenous ligand for those receptors, which is neither a catecholamine nor likely a peptide, is under processing for purification. Therefore, it appeared that the hypotensive effect of substances with an imidazoline or imidazoline-like structure might be due to their action within medullary receptors specific for this endogenous ligand temporarily named "clonidine displacing substance." Rilmenidine, structurally close to imidazolines, also interfered with these receptors. The central component of its hypotensive effect was recently confirmed in rabbits, where its central cardiovascular effects were antagonized by "the clonidine displacing substance." Although exhibiting a lower affinity than the reference substance for these receptors, rilmenidine might have a higher selectivity, thus explaining its restricted side effects. A structure-activity study with this molecule would bring a confirmation to these first observations.

Role of imidazoline receptors in cardiovascular regulation.

The involvement of nonadrenergic imidazoline specific receptors in the central control of the vasomotor tone and in the mechanism of action of drugs bearing an imidazoline structure, or analogs, is now well documented. Imidazoline-specific binding sites were found in many tissues and species. Moreover, until now, it is only in the brainstem that such binding sites are associated with a function: the hypotensive effect of imidazoline-like drugs. Rilmenidine, which is an oxazoline structurally related to the reference imidazolines, exerts a central hypotensive effect of central origin involving imidazoline receptors. The selectivity of rilmenidine for the imidazoline receptors compared to alpha 2-adrenergic receptors could explain the low incidence of sedative side effects observed with this antihypertensive drug. A specific anti-imidazoline radioimmunoassay allowed us to detect the presence of an immunoreactive imidazoline-like substance in human sera. High levels of this immunoreactive substance are associated with high blood pressure in 20-30% of the hypertensive patients. This observation indicates that high levels of this immunoreactive substance in the serum can be associated with some kinds of primary hypertension. The cause-and-effect relation between these 2 phenomena has not yet been determined. This substance is in process of purification; it could be a candidate to be an endogenous ligand of the imidazoline receptors.

Evidence for the involvement of imidazoline receptors in the central hypotensive effect of rilmenidine in the rabbit.

1. Rilmenidine has recently been introduced as a new centrally-acting antihypertensive agent. We examined its cardiovascular effects after intracerebral injection to anaesthetized rabbits. Cumulative doses of rilmenidine injected intracisternally (1 to 300 micrograms kg-1) led to dose-dependent decreases in arterial blood pressure and heart rate. The effective doses of rilmenidine were lower when injected centrally than when injected intravenously. 2. Pretreatment with the same dose of yohimbine or idazoxan shifted the rilmenidine dose-response curves for its hypotensive and bradycardic effects to the right. Idazoxan, which has an imidazoline structure, proved to be a more active antagonist than yohimbine of rilmenidine centrally-mediated cardiovascular effects. 3. The dose-response curve for the central hypotensive effect of rilmenidine was also shifted to the right after pretreatment with a bovine brain extract. This extract contains the endogenous ligand of the imidazoline-preferring receptors which is not a catecholamine. 4. Rilmenidine, like clonidine, proved to be active when micro-injected into the rabbit nucleus reticularis lateralis region. 5. In conclusion, rilmenidine exhibited in the rabbit a central hypotensive effect which originated in the same area as where clonidine acts. Specific imidazoline-preferring receptors appear to be involved in this hypotensive effect.

Relevance of the use of [3H]-clonidine to identify imidazoline receptors in the rabbit brainstem.

1. [3H]-clonidine binding was investigated in membranes isolated from the ventral medulla oblongata of the rabbit, where clonidine produced a hypotensive effect which was not mediated by adrenoceptors. [3H]-clonidine specific binding, as defined by the difference between the binding of [3H]-clonidine in the presence and in the absence of 10 microM cirazoline, occurred at two sites: a high affinity site with a KD = 2.9 +/- 0.7 nM and a Bmax of 40 +/- 8 fmol mg-1 protein and a low affinity site with a KD = 18.2 +/- 0.4 nM and a Bmax of 66 +/- 14 fmol mg-1 protein. 2. The high affinity sites being catecholamine-sensitive were identified as alpha 2-adrenoceptors. The low affinity binding of [3H]-clonidine was insensitive to catecholamines, as well as to other alpha 2-adrenoceptor specific probes, and could be inhibited with high affinity only by compounds which lowered blood pressure when directly injected in the nucleus reticularis lateralis of the ventral brainstem, or by antagonists. 3. It was concluded that in the ventral medulla of the rabbit, [3H]-clonidine labelled alpha 2-adrenoceptors and imidazoline receptors (IRs). Only the latter were related to the hypotensive effects of clonidine and rilmenidine directly injected into the rostroventrolateral medulla oblongata (RVLM) of the rabbit. The methodological problems regarding the study of IRs with [3H]-clonidine are discussed.

Respective contributions of alpha-adrenergic and non-adrenergic mechanisms in the hypotensive effect of imidazoline-like drugs.

The hypotensive effect of imidazoline-like drugs, such as clonidine, was first attributed to the exclusive stimulation of central alpha2-adrenoceptors (alpha2ARs). However, a body of evidence suggests that non-adrenergic mechanisms may also account for this hypotension. This work aims (i) to check whether imidazoline-like drugs with no alpha2-adrenergic agonist activity may alter blood pressure (BP) and (ii) to seek a possible interaction between such a drug and an alpha2ARs agonist alpha-methylnoradrenaline (alpha-MNA). We selected S23515 and S23757, two imidazoline-like drugs with negligible affinities and activities at alpha2ARs but with high affinities for non-adrenergic imidazoline binding sites (IBS). S23515 decreased BP dose-dependently (-27+/-5% maximal effect) when administered intracisternally (i.c.) to anaesthetized rabbits. The hypotension induced by S23515 (100 microg kg(-1) i.c.) was prevented by S23757 (1 mg kg(-1) i.c.) and efaroxan (10 microg kg(-1) i.c.), while these compounds, devoid of haemodynamic action by themselves, did not alter the hypotensive effect of alpha-MNA (3 and 30 microg kg(-1) i.c.). Moreover, the alpha2ARs antagonist rauwolscine (3 microg kg(-1) i.c.) did not prevent the effect of S23515. Finally, whilst 3 microg kg(-1) of S23515 or 0.5 microg kg(-1) of alpha-MNA had weak hypotensive effects, the sequential i.c. administration of these two drugs induced a marked hypotension (-23+/-2%). These results indicate that an imidazoline-like drug with no alpha2-adrenergic properties lowers BP and interacts synergistically with an alpha(ARs agonist.

Co-detection by two imidazoline receptor protein antisera of a novel 85 kilodalton protein.

Imidazoline receptors (I-receptors) are considered as potential therapeutic targets for a spectrum of stress-induced illnesses. Yet, I-receptors remain poorly defined at the molecular level. In this study, candidate imidazoline receptor proteins were compared using two imidazoline receptor-selective antisera of diverse origins. One antiserum was derived from affinity-purified imidazoline-binding protein. The second antiserum was produced as an anti-idiotypic antiserum, from purified IgG selective for imidazolines. Despite such diverse origins, both antisera co-identified an 85 kDa band on western blots from a variety of tissues. The integrity of the 85 kDa band was dependent on protection by eight different protease inhibitors. Other proteolytic breakdown products (obtained after homogenization with only one protease inhibitor) were comparable in size to previously reported smaller immunoreactive bands. The full-size 85 kDa band was also enriched in plasma membrane fractions and abundant in rat PC12 cells and brain regions known to be abundant in I1 binding sites. Furthermore, the immunodensity of the 85 kDa band, against anti-idiotypic antiserum, was linearly correlated with reported I1 site radioligand Bmax values (r2 = 0.8736, P = 0.0002) across nine rat tissues. Therefore, a possible candidate for the full-length imidazoline receptor(s) appears to be an 85 kDa protein.

Imidazoline receptors. A new concept in central regulation of the arterial blood pressure.

Clonidine-like antihypertensive substances bind to nonadrenergic imidazoline specific sites within the nucleus reticularis lateralis (NRL), their medullary privileged site of action. Rilmenidine, a new central antihypertensive agent, was tested in the anesthetized rabbit. For the same hypotensive effect, cumulative doses given intracisternally proved 50 times more active than of those given systemically. Idazoxan, an alpha 2-adrenergic antagonist structurally related to the imidazolines, given centrally as pretreatment proved more potent in preventing the hypotensive effects than the same molar dose of yohimbine. When injected within the NRL area of the anesthetized rabbit, rilmenidine, like clonidine, always exhibited a hypotensive effect. The influence of clonidine and rilmenidine upon the NRL noradrenergic neurons involved in the blood pressure regulation, and upon those of the locus coeruleus (LC) involved in the sedative effect was studied by differential voltammetry. It was observed that rilmenidine was two times more selective than clonidine in inhibiting the NRL, as opposed to LC, neuronal activity. In addition, binding experiments of tritiated clonidine to human cortical and NRL membrane preparations showed that rilmenidine, as compared to clonidine, has a two to three times higher selectivity for the imidazoline receptors. In conclusion, we are now able to discriminate between the mechanism of the hypotensive effect of imidazoline-like drugs and that of their sedative action. Rilmenidine is the first example of an hypotensive drug more selective for imidazolin preferring receptors than for classical alpha 2-adrenoceptors.

Characterization of imidazoline binding protein(s) solubilized from human brainstem: studies with [3H]idazoxan and [3H]clonidine.

Imidazoline binding sites from the human brainstem were solubilized with 3-[(3-cholamido-propyl)-dimethylammonio]-1-propane-sulfonate (CHAPS). [3H]idazoxan and [3H]clonidine were used as ligands to characterize the solubilized binding sites. In both the soluble and membrane fractions, [3H]idazoxan binding was saturable, stereoselective, sensitive to imidazolines and insensitive to (-)norepinephrine and to amiloride. The affinities of [3H]idazoxan for the soluble and membrane sites were similar (KD = 25 +/- 11 nM and 20 +/- 3 nM). In both soluble and membrane fractions, the alpha 2-adrenoceptor binding being masked with (-)norepinephrine, [3H]clonidine bound to a low affinity site which was insensitive to (-)norepinephrine and which exhibited the same selectivity for various drugs as the [3H]idazoxan binding site. alpha 2-adrenoceptor binding was present in the membrane and the soluble fractions although it was difficult to detect in the soluble fraction because of inhibition of [3H]rauwolscine binding by the CHAPS detergent.

Binding of new cirazoline derivative to imidazoline receptors from human brain.

Imidazoline compounds are known to interact with alpha 2-adrenoceptors as well as with specific non-adrenergic binding sites. Such binding sites are present in the brain and in peripheral tissues. Hypotensive effects of imidazolines were shown to be related to specific interaction with imidazoline binding sites within the brainstem. Heterogeneity of these sites based on differences in selectivities was reported. In order to facilitate the characterization of human brain imidazoline receptors, we synthetized new ligands by substitutions on the cirazoline phenyl ring. Affinities of these cirazoline derivatives were determined in two imidazoline binding site models, namely the human brain and the rabbit kidney. Interaction of these compounds with imidazoline binding sites from the human brain appeared more sensitive to structural variations of the imidazoline than those with rabbit kidney sites. Moreover, no correlation was found between affinities for imidazoline binding sites and those for alpha 2-adrenoceptors of the rat brain. Arylazide derivative of 2-(5-amino-2-methyl-phenoxymethyl)-imidazoline exhibited a higher affinity for human brain imidazoline binding sites than for human brain alpha 2-adrenoceptors. Photoincorporation of this azido-compound in human brain imidazoline binding sites was achieved and blockade of [3H]idazoxan imidazoline specific binding observed. These new tools may allow fine characterization of the different subtypes of imidazoline binding proteins.

Identification of human I1 receptors and their relationship to alpha 2-adrenoceptors.

I1 imidazoline receptors (I1R) were defined as receptors insensitive to catecholamines and highly sensitive to [3H]clonidine and analogs. By contrast, the I2R subtype is more sensitive to [3H]idazoxan. [3H]clonidine and [3H]idazoxan imidazoline specific binding sites (IBS) have been detected in crude human membranes. Pharmacologic characterization by binding assays clearly differentiates IBS from alpha 2-adrenoceptors, whereas differences between [3H]clonidine and [3H]idazoxan IBS are less clear in crude preparations. In fact, only moderate affinity for [3H]clonidine was detectable in such preparations. However, purification procedures allowed detection of high affinity [3H]clonidine IBS in the human brain, corresponding to the I1R. Difficulties in the characterization of the I1R in crude membranes are due to multiple factors including heterogeneity of IBS, their low Bmax value, the existence of allosteric modulation, and possibly the presence of natural binding inhibitors. Immunologic studies with specific anti-idiotypic antibodies revealed a 43-kD protein as the best candidate for I1R as binding activity coincides with immunodetection. No cross-reaction was found with anti-monoamine oxidase (MAO) A/B antibodies and the 43-kD protein, ruling out the possibility of this protein being an MAO-associated I2R. Neither anti-alpha 2A- nor anti-alpha 2B-specific antibodies were able to immunodetect the 43-kD protein in crude membrane preparations or in purified fractions. These results and further biochemical characterization (pHi, N-glycosylation) of the 43-kD protein definitely assessed that human brain I1R and alpha 2-adrenoceptors clearly differ physically. However, coexpression of I1R and alpha 2-adrenoceptors in synaptic plasma membranes of the bovine brainstem reinforce the possibility of a functional relationship between the two types of receptor.

Participation of imidazoline receptors and alpha(2-)-adrenoceptors in the central hypotensive effects of imidazoline-like drugs.

The central hypotensive effect of imidazoline-like drugs (IMs) involves non-adrenergic imidazoline receptors (IRs). IMs cause hypotension irrespective of their affinity and selectivity for one or the other alpha-adrenoceptor subtypes. LNP 509, which binds to I1Rs (Ki = 5.10(-7) M) but roughly not to alpha 2-adrenoceptors (A2Rs) (Ki > 10(-5) M), causes hypotension when injected alone into the brainstem. As far as hybrid drugs, that is, those with mixed binding profiles (I1/alpha 2), are concerned, a significant correlation was reported between their central hypotensive effect and their affinity for IRs. Imidazoline antagonists such as idazoxan competitively antagonized the centrally induced hypotensive effect of IMs. Yohimbine, an A2Rs antagonist, blocks the hypotensive effect of hybrids but usually in a noncompetitive manner. Mutation of A2Rs prevented the hypotensive effects of drugs highly selective for A2Rs, but also that of hybrids such as clonidine. These data indicate that triggering of the hypotensive effects of IMs (1) needs implication of IRs; (2) appears to be facilitated by additional activation of A2Rs; and (3) requires integrity of A2Rs along the sympathetic pathways.

Postnatal development of the endothelial system of the rat cerebellum using immunohistochemical techniques.

The development of the endothelial system of the rat cerebellum was studied by immunohistochemical techniques with two antibodies specific of two membrane antigens characteristic of 'mature' endothelial cells. It appears that the maturation of endothelial cells is very late in the rat cerebellum. This absence of a fully endothelial barrier in cerebella of young rats could explain the extreme sensitivity of this part of the central nervous system to various agents.

Immunohistochemical localization of a lectin-like molecule, R1, during the postnatal development of the rat cerebellum.

The immunohistochemical localization of an endogenous lectin R1 isolated from the rat cerebellum was studied during its postnatal development. The lectin is present in the cerebellum from birth to adulthood, essentially in lysosomes, multivesicular bodies, and parts of the endoplasmic reticulum, principally of large and intermediate size neurons. During the period of massive synaptogenesis in the molecular layer, there is a sprouting of R1 in some distal dendrites of Purkinje cells. The lectin appears to be particularly concentrated on their plasma membranes, in coated pits, in coated vesicles, multivesicular bodies and lysosomes. At the same period, in cerebella of rats treated with chloroquine (an inhibitor of lysosomal function), both the lectin and mannose-rich glycoproteins of newly formed parallel fibres (able to bind specifically this lectin) are found in the same non-functional lysosomes of Purkinje cells. It is thus suggested that both this lectin (with a high-affinity for the glycans of the mannose-rich glycoproteins of the membrane of the newly formed parallel fibres) and these glycoproteins could be the recognition molecules allowing a specific contact between parallel fibres and Purkinje cells at the period of synaptogenesis.

Isolation and immunohistochemical localization of a lectin-like molecule from the rat cerebellum.

A lectin with a mannose specificity was isolated from the cerebellum of young rats. The method of purification was based on the observation that during homogenization of the tissue, the lectin binds to a class of mannose-rich glycoproteins highly insoluble in Triton X-100. Sequential extractions in saline buffer devoid of, then containing, 0.5% Triton X-100 allowed the elimination of a great part of other proteins. Using the same buffer containing 0.5 M mannose, a specific class of protein can be solubilized. This fraction was enriched by affinity adsorption on insolubilized mannose-rich glycoproteins followed by specific detachment with mannose. One of the protein subunits, of molecular weight (MW) 130,000, was isolated by preparative gel electrophoresis. Upon re-electrophoresis, this compound gives two bands of MW 65,000 and 130,000, which appear to be a monomer and a dimer of a molecule called R1. Antibodies were raised against R1 which react with the monomer and the dimer and not against other proteins of the rat cerebellum. The immunohistochemical localization of this lectin was performed in cerebella of 20-day-old rats. The antigen is concentrated in endothelial cells and in large and intermediate size neurons (Purkinje, Golgi, basket and deep nuclei neurons). Granule cell bodies are lightly stained and no label at all was found in glial cells. At the level of electron microscopy, the antigen was found to be very concentrated in multivesicular bodies and lysosomes of large neurons, on parts of the endoplasmic reticulum, on some mitochondrial outer membranes and on the plasma membrane of the dendrites. The possible role of this lectin in cerebella of young rats is discussed in relation to its interaction with a specific class of mannose-rich glycoproteins.

Con A-binding glycoproteins in the developing cerebellum of control and hypothyroid rats.

Concanavalin A (Con A)-binding glycoproteins were studied during the postnatal development of the cerebellum of control and hypothyroid rats. Only 4 glycoprotein bands have a transient behavior in control animals. They progressively increase until the 13th day and markedly decline between the 15th and the 18th postnatal day. In the cerebellum of hypothyroid rats, the level of these compounds is greatly reduced and the previous decrease observed in controls is not found again. This defect of Con A-binding glycoproteins mainly localized on the plasma membrane of parallel fibers might be related to the reduced synaptogenesis observed in the molecular layer of hypothyroid rats between parallel fibers and Purkinje cell dendritic spines.

Postnatal variations of endo-beta-D-N-acetyl-glucosaminidase in the developing rat cerebellum.

Some properties and the level of endo-beta-N-acetylglucosaminidase were studied during the postnatal development of the rat cerebellum. No isoenzymes were detected. The enzyme (essentially cytosolic) showed a peak of total and specific activity at the 24th postnatal day. The hypothesis was formulated that glycoprotein glycans could be degraded in two steps: one lysosomal, the other cytosolic.

Arguments in favour of endocytosis of glycoprotein components of the membranes of parallel fibers by Purkinje cells during the development of the rat cerebellum.

Chloroquine (a drug known to induce a dysfunction of lysosomes) was used to study the behavior of Concanavalin A binding glycoproteins located on the axolemma of parallel fibers in young rat cerebella, and abundant on these membranes at a period preceding synaptogenesis with the dendrites of Purkinje cells. Chloroquine induces in Purkinje cells a large accumulation of grains consisting of membrane whorls in lysosomes. These grains stain for Concanavalin A, and do not stain either for a mitochondrial marker (aspartate aminotransferase mitochondrial isoenzyme) or for a marker of the Purkinje cell internal membrane (PSG). It is suggested that the material accumulating in the Purkinje cells under the effect of chloroquine comes from the parallel fibers. Together with the observation that alpha-D-mannosidase (involved in the degradation of these glycoproteins) is exclusively located inside Purkinje cells, these results provide a firm indication that this material enters the Purkinje cells through pinocytosis. The absence of ATPase activity (ATPase is a glycoprotein plasma membrane marker highly concentrated on parallel fibers) within these grains suggested that not all the components of these membranes are pinocytosed, but that the process is specific for certain molecules. These results are compatible with the ultrastructural observations of others, and support the arguments in favour of the pinocytosis phenomenon being one of the first steps of synapse formation. The observed specificity of pinocytosis for certain molecules suggests that a receptor-mediated recognition of some glycans of glycoproteins is the preliminary event in the establishment of synapses.

Heterogeneity of imidazoline binding sites revealed by a cirazoline derivative.

The affinity of AMPI (2-[3-aminophenoxy]methyl imidazoline) for [3H]clonidine and [3H]idazoxan imidazoline binding sites was determined in various rabbit and human tissues. Although cirazoline showed a high affinity (nM range) in all the tested tissues, its derivative, AMPI, had a high affinity (nM range) in rabbit brain and kidney but a low affinity (microM range) in the human brain. These differences in affinities were very similar to those obtained with amiloride. The same results were obtained when considering [3H]clonidine or [3H]idazoxan specific imidazoline binding sites.