Lobular architecture of human adipose tissue defines the niche and fate of progenitor cells.

Human adipose tissue (hAT) is constituted of structural units termed lobules, the organization of which remains to be defined. Here we report that lobules are composed of two extracellular matrix compartments, i.e., septa and stroma, delineating niches of CD45-/CD34+/CD31- progenitor subsets characterized by MSCA1 (ALPL) and CD271 (NGFR) expression. MSCA1+ adipogenic subset is enriched in stroma while septa contains mainly MSCA1-/CD271- and MSCA1-/CD271high progenitors. CD271 marks myofibroblast precursors and NGF ligand activation is a molecular relay of TGFß-induced myofibroblast conversion. In human subcutaneous (SC) and visceral (VS) AT, the progenitor subset repartition is different, modulated by obesity and in favor of adipocyte and myofibroblast fate, respectively. Lobules exhibit depot-specific architecture with marked fibrous septa containing mesothelial-like progenitor cells in VSAT. Thus, the human AT lobule organization in specific progenitor subset domains defines the fat depot intrinsic capacity to remodel and may contribute to obesity-associated cardiometabolic risks.

Analysis of the pharmacological and molecular heterogeneity of I(2)-imidazoline-binding proteins using monoamine oxidase-deficient mouse models.

The I(2) subgroup of imidazoline-binding sites was identified as monoamine oxidases (MAOs), but it is unclear whether there are I(2)-binding sites located on proteins distinct from MAOs. To address this issue, we characterized I(2)-binding proteins in liver and brain of wild-type and MAO A- and MAO B-deficient mice. I(2)-binding sites were identified using [(3)H]idazoxan and the photoaffinity adduct 2-[3-azido-4-[(125)I]iodophenoxyl]methylimidazoline ([(125)I]AZIPI). [(3)H]Idazoxan labeled binding sites with ligand recognition properties typical of I(2) sites in both brain and liver of wild-type mice. High-affinity, specific [(3)H]idazoxan binding were not altered in MAO A knockout (KO) mice. In contrast, [(3)H]idazoxan binding was completely abolished in both liver and brain of MAO B KO mice. In wild-type mice, [(125)I]AZIPI photolabeled three proteins with apparent molecular masses of approximately 28 (liver), approximately 61 (brain), and approximately 55 kDa (liver and brain). The photolabeling of each protein was blocked by the imidazoline cirazoline (10 microM). Photolabeling of the approximately 61- and approximately 55-kDa proteins was not observed in MAO A and B KO mice, respectively. In contrast, photolabeling of the liver approximately 28-kDa protein was still observed in MAO-deficient mice, indicating that this protein is unrelated to MAOs. These data indicate that I(2) imidazoline-binding sites identified by [(3)H]idazoxan reside solely on MAO B. The binding sites on MAO A and the liver approximately 28-kDa protein may represent additional subtypes of the family of the imidazoline-binding sites.

High expression of monoamine oxidases in human white adipose tissue: evidence for their involvement in noradrenaline clearance.

The clearance of plasma adrenaline and noradrenaline by human adipose tissue suggests the expression of the catecholamine-degrading enzyme monoamine oxidases and of catecholamine transport systems in adipocytes. In the present study, we identified and characterized the monoamine oxidases and an extraneuronal noradrenaline transporter expressed in human adipocytes. Enzyme assays using the monoamine oxidase A/B substrate [14C]tyramine showed that abdominal and mammary human adipocytes contain one of the highest monoamine oxidase activities in the body. Characterization of the enzyme isoforms by inhibition profiles of [14C]tyramine oxidation and Western and Northern blot analyses showed that mRNAs and proteins related to both monoamine oxidases A and B were expressed in adipocytes. Quantification of each enzyme isoform performed by enzyme assay and Western blot showed that monoamine oxidase A was predominant, representing 70-80% of the total enzyme activity. In uptake experiments, the monoamine oxidase substrate [3H]noradrenaline was transported into white adipocytes (Vmax 0.81+/-0.3 nmol/30 min/100 mg of lipid, Km 235+/-104 microM). The inhibition of [3H]noradrenaline uptake by specific inhibitors indicated that white human adipocytes contain an extraneuronal-type noradrenaline transporter. Competition studies of [14C]tyramine oxidation showed that noradrenaline is metabolized by monoamine oxidases in intact cells. In conclusion, the concomitant expression of monoamine oxidases and of a noradrenaline transporter in human white adipocytes supports the role of the adipose tissue in the clearance of peripheral catecholamines. These results suggest that adipocytes should be considered as a previously unknown potential target of drugs acting on monoamine oxidases and noradrenaline transporters.

Relationship between I2 imidazoline binding sites and monoamine oxidase B in liver.

Biochemical and pharmacologic studies suggest that I2 imidazoline binding sites (I2BS) represent a heterogeneous family of membrane proteins. Indeed, the imidazoline binding sites located on monoamine oxidases (MAO) A and B display different pharmacologic properties. Recent results suggest that in liver and brain, I2BS may be located on proteins distinct from MAOs. The following observations indicate that in liver and brain, [3H]idazoxan binds exclusively to I2BS located on MAO-B: (1) size exclusion chromatography of digitonin-solubilized preparations from rabbit and human liver showed that [3H]idazoxan-specific binding eluted only in two peaks (approximately 175,000 and approximately 100,000 Da, corresponding to 90% and 10% of the recovered [3H]idazoxan binding) which also contained MAOs as determined by [14C]tyramine oxidation and Western blot analysis; (2) according to previous results obtained in various human and rat tissues, experiments performed in mice liver and brain showed that idazoxan was a potent inhibitor of [125I]-AZIPI photoincorporation to MAO-B but not to MAO-A; (3) in MAO-deficient transgenic mice, [3H]idazoxan binding to liver and brain membranes was completely abolished in MAO-B knockout mice and was not affected in MAO-A knockout mice. Together, these results show that in both liver and brain, I2BS are located exclusively on MAO-B. The imidazoline binding site on MAO-A, which photoincorporates [125I]-AZIPI and displays a low affinity for idazoxan, may not belong to the family of the I2 imidazoline binding sites.

Characterization of monoamine oxidase isoforms in human islets of Langerhans.

In this paper, we describe the characterization of the expression of monoamine oxidase (MAO) in whole pancreas and in isolated islets of Langerhans from human. Classical monamine oxidase activity assays reveal that both isoforms A & B are present in human pancreas. Two complementary approaches indicated that both MAO A and B are expressed in isolated islet: RT-PCR using specific primers revealed amplification products with the expected size for MAO-A and MAO-B: two peptides corresponding to MAO A (approximately 61 kDa) and B (approximately 55 kDa) were detected using a polyclonal anti MAO-A/MAO-B antiserum. Western blotting and subsequent densitometric analysis indicate that whole and endocrine pancreas express the two isoforms with different relative proportions. Islets appear to express almost twice as much MAO protein as whole pancreas, in near equal proportions of the two isoforms, whereas whole pancreas expresses more MAO-A than the B isoform. The expression of MAO A and B in islets could be the first step toward the characterization of the functional properties of these enzymes in the endocrine pancreas.

Characterization of [3H]idazoxan binding proteins in solubilized membranes from rabbit and human liver.

Several studies have shown that I2 imidazoline binding sites are localized on monoamine oxidases. Recent results obtained after solubilization of rat brain membranes and analysis by size-exclusion chromatography suggested the existence of additional I2 imidazoline binding sites located on proteins distinct from monoamine oxidases. In order to define whether such binding sites are expressed in human and rabbit liver, we solubilized I2 imidazoline binding sites and monoamine oxidases and compared their elution profile by size-exclusion chromatography. I2 binding sites were labeled using [3H]idazoxan. Monoamine oxidases were identified by the measure of [14C]tyramine oxidation and Western blot analysis using an anti-MAO-A/MAO-B polyclonal antiserum. After solubilization of rabbit or human liver using 1% digitonin, 90% of [3H]idazoxan binding eluted in a major peak corresponding to a Mr of approximately 175000 Da. A minor peak, (Mr approximately equal to 100000 Da) representing 10% of the recovered [3H]idazoxan binding, was also observed. [14C]tyramine oxidation as well as immunoreactive bands corresponding to MAOs were exclusively detected in fractions containing [3H]idazoxan binding. These results show that solubilized I2 imidazoline binding sites distinct from monoamine oxidases are not detectable in rabbit and human liver.

The renal monoamine oxidases: pathophysiology and targets for therapeutic intervention.

The mitochondrial enzyme monoamine oxidases A and B are among the major metabolic agents for the degradation of the biogenic amines adrenaline, noradrenaline, dopamine and serotonin. The fact that the kidney contains a large amount of monoamine oxidase suggests that the renal effects of biogenic amines might depend in part on the activity of these enzymes.

Predominant expression of monoamine oxidase B isoform in rabbit renal proximal tubule: regulation by I2 imidazoline ligands in intact cells.

Previous studies have shown that a subpopulation of the catecholamine-degrading enzymes monoamine oxidase (MAO) A and B holds a previously unknown regulatory site, the I2-imidazoline binding site (I2BS). In the present work, we characterized the isoforms of monoamine oxidases expressed in the rabbit renal proximal tubule, defined their relationship with I2BS, and investigated the ability of I2BS ligands to inhibit enzyme activity in intact cells. Two findings indicate that MAO-B is the predominant isoform expressed in the renal proximal tubule cells: 1) Western blot performed with an anti-MAO-A/MAO-B polyclonal antiserum revealed a single 55-kDa band corresponding to MAO-B; 2) enzyme assays showed an elevated MAO-B activity ([14C]beta-phenylethylamine oxidation: Vmax = 1.31 +/- 0.41 nmol/min/mg protein), whereas MAO-A activity was only detectable ([14C]5-HT oxidation: Vmax = 80.3 +/- 19 pmol/min/mg protein). Photoaffinity labeling with the I2BS ligand [125I]2-(3-azido-4-iodophenoxy)-methylimidazoline revealed a single 55-kDa band, which indicates that MAO-B of the renal proximal tubule cells holds the I2 imidazoline binding site. [3H]Idazoxan binding studies and enzyme assays showed that, in intact cells, I2BS ligands bind to and inhibit MAO-B. Indeed, the increase in the accessibility of intracellular compartment by cell permeabilization did not enhance [3H]idazoxan binding, which indicates that, in intact cells, intracellular I2BS are fully occupied by imidazoline ligands. In addition, enzyme assays showed that incubation of proximal tubule cells with imidazoline ligands leads to a complete, dose-dependent inhibition of MAO activity. These data show the predominant expression of MAO-B in rabbit renal proximal tubule and its regulation by imidazoline ligands in intact cells.

I2-imidazoline binding sites and monoamine oxidase activity in human postmortem brain from patients with Parkinson's disease.

I2-imidazoline binding site (I2BS) has been identified with a regulatory site located on a sub-population of monoamine oxidase (MAO)-A and -B. Previous studies showed a modification of MAO and I2BS in the elderly and in neurodegenerative processes such as Alzheimer's disease. In the present study, we studied the potential modification of I2 binding sites and monoamine oxidases in Parkinson's disease. Putamen and cerebral cortex were collected from 17 normal subjects (79 +/- 12 yr) and 16 patients (76 +/- 9 yr) affected by Parkinson's disease. In mitochondrial preparations, radioligand binding studies with [3H]idazoxan showed that putamen and frontal cortex express equivalent amount of I2BS. The density and affinity of I2BS were similar in normal subjects (putamen: Bmax = 207 +/- 58 fmol/mg of protein, Kd = 10.1 +/- 3.4 nM; cerebral cortex: Bmax = 193 +/- 54 fmol/mg of protein, Kd = 12.8 +/- 6.8 nM) and Parkinson's disease patients (putamen: Bmax = 193 +/- 60 fmol/mg of protein, Kd = 9.8 +/- 4.6 nM; cerebral cortex: Bmax = 199 +/- 49 fmol/mg of protein, Kd = 15.9 +/- 8.1 nM). The activity of total monoamine oxidase and monoamine oxidase B, measured by [14C]tyramine and [14C]phenylethylamine oxidation, respectively, were higher in putamen than in cerebral cortex. No differences have been detected in the enzyme activity between normal and pathological subjects. These data suggest that, although MAO and I2BS may play a role in the development of Parkinson's disease, they are not altered in the chronic phase of this disease.

Molecular mechanisms underlying regional variations of catecholamine-induced lipolysis in rat adipocytes.

The mechanisms underlying catecholamine control of lipolysis were studied in rat white adipocytes from epididymal, retroperitoneal, and subcutaneous fat depots. Sensitivity of subcutaneous adipocytes to selective beta 3-adrenoceptor agonists was lower than that of internal adipocytes. beta 3-Adrenoceptor mRNA levels were lower in subcutaneous adipocytes. A decreased beta 1/beta 2-adrenoceptor-mediated lipolysis was also observed in these adipocytes, and the number of beta 1/beta 2-adrenoceptors was lower than in the internal adipocytes. The number of alpha 2-adrenoceptors was higher in subcutaneous adipocytes without a marked difference in alpha 2-adrenoceptor-mediated antilipolysis between the depots. Subcutaneous adipocytes were also characterized by a lower maximal lipolytic response to drugs acting at different levels of the lipolytic cascade, suggesting differences at the postreceptor level. Lower hormone-sensitive lipase activity and mRNA levels in subcutaneous adipocytes were in agreement with the lipolysis data. These results suggest that the pattern of expression of the genes of the lipolytic pathway varies with the anatomic location of the fat depot.

Inhibition of amine oxidase activity by derivatives that recognize imidazoline I2 sites.

Nonadrenergic imidazoline binding sites (imidazoline I2 sites) have been described to be colocated with monoamine oxidase (MAO) in the mitochondrial fraction of various cell types. In the present work, the authors considered whether this colocation could be associated with a functional interplay. In rat liver membranes, [3H]-idazoxan binding to I2 receptors was competed for by naphazoline and idazoxan, which also shared a high affinity for alpha-2 adrenoceptors (alpha-2 ARs). The chemicals 2-n-heptylimidazoline (S 15430), 1-methyl-5-n-heptylimidazole (S 15674), 2-benzofuran-2-yl-imidazoline (RX 801077) and 2-(1,3-benzodioxanyl)-2-imidazoline (RX 821029) exhibited higher affinity for I2 receptors than for alpha-2 ARs. The most selective agent was S 15430 with a 150-fold higher affinity for liver I2 receptors than for adipocyte alpha-2 ARs. Moreover, [3H]-idazoxan binding was also competed for by several MAO inhibitors (MAOI) that are not imidazoline or guanidinium derivatives such as tranylcypromine, harmaline, clorgiline and pargyline. Rat liver MAO activity was not only inhibited by MAOIs but also by some imidazoline derivatives: cirazoline, naphazoline, S 15674, RX 801077 and RX 821029. Idazoxan had no effect on MAO activity; it neither inhibited MAO nor prevented the inhibition induced by other imidazolines or MAOIs. This suggested that the ligand recognition site of I2 receptors was distinct from the MAOI target site. Furthermore, some imidazolines inhibited the activity of bovine plasma amine oxidase, an enzyme that does not possess the same cofactor as MAO and is insensitive to harmaline or pargyline.(ABSTRACT TRUNCATED AT 250 WORDS)

Expression of hormone-sensitive lipase in the human colon adenocarcinoma cell line HT29.

Hormone-sensitive lipase expression was studied in the human colon adenocarcinoma cell line, HT29. Diacylglycerol lipase and cholesterol esterase [corrected] activities in HT29 cells were inhibited by known inhibitors of hormone-sensitive lipase (diethyl-p-nitrophenyl phosphate, NaF and HgCl2) to the same extent as in human adipocytes. A polyclonal antiserum directed against rat hormone-sensitive lipase inhibited 89% of HT29 cell lipase activity. HT29 hormone-sensitive lipase was the same size as the adipocyte enzyme as was its mRNA. Complete homology between mRNA sequences in HT29 and adipocyte was demonstrated using ribonuclease protection assay. These data are consistent with the expression of a protein closely related, if not identical, to the enzyme expressed in human adipose tissue. HT29 is the first human cell line where hormone-sensitive lipase expression has been shown.

Coupling of inhibitory receptors with Gi-proteins in hamster adipocytes: comparison between adenosine A1 receptor and alpha 2-adrenoceptor.

Adenosine A1 receptors and alpha 2-adrenoceptors are both potent inhibitors of adipocyte lipolysis when activated by their agonists. The aim of this work was to compare the coupling of these receptors to the Gi-proteins in hamster adipocytes. The adenosine A1 receptor was characterized with the antagonist [3H]dipropyl-cyclopentyl-xanthine ([3H]DPCPX) and the agonist [3H](-)-phenylisopropyladenosine ([3H]PIA). It was demonstrated by [32P]ADP-ribosylation with pertussis toxin and immunoblotting that Gi1, Gi2 and Gi3 are expressed in hamster adipocytes. Partial ADP-ribosylation of Gi-proteins by pertussis toxin, acting on the intact cells or on the adipocyte membranes, demonstrated that the adenosine A1 receptor was less sensitive to the disappearance of functional Gi-proteins than the alpha 2-adrenoceptor. These results are in accordance with the weak sensitivity of the binding of the agonist [3H]PIA to guanine nucleotides and seem to confirm that the adenosine A1 receptor is strongly and differently coupled than the alpha 2-adrenoceptor to the Gi-proteins in hamster adipocyte membranes.

Distribution of PYY receptors in human fat cells: an antilipolytic system alongside the alpha 2-adrenergic system.

The antilipolytic effect of peptide YY (PYY) and neuropeptide Y has recently been shown in human adipocytes. PYY receptors were investigated in three human adipose deposits. A greater number of 125I-labeled PYY binding sites was found in femoral adipocyte membranes (maximal binding = 40 +/- 4 fmol/mg protein; dissociation constant = 0.3 +/- 0.1 nM) when compared with mammary and pericolonic adipose tissue. PYY receptors, like alpha 2-adrenergic sites, were largely expressed in femoral fat cells. Such a distribution was not specific either to inhibitory or to stimulating adenylyl cyclase systems since adenosine A1 and beta-adrenergic receptors were more numerous in pericolonic adipocytes. On isolated adipocytes, PYY (10(-7) M) inhibited lipolysis by 58 +/- 2% in femoral and 14 +/- 4% in pericolonic fat cells; epinephrine had the following similar response: 62 +/- 5 and 26 +/- 8%, respectively. A close relationship between the number of alpha 2-sites and PYY sites and the antilipolytic effects initiated by PYY and an alpha 2-agonist was observed. No significant differences were noted in the amount of Gi proteins in femoral and pericolonic adipocyte membranes.

Coupling of the alpha 2-adrenergic receptor to the inhibitory G-protein Gi and adenylate cyclase in HT29 cells.

Previous studies have established that the human colon carcinoma cell line HT29 expresses an alpha 2-adrenergic receptor of the alpha 2A subtype, which is negatively coupled to adenylate cyclase. The purpose of the present study was to examine the mechanisms of alpha 2-adrenergic signal transduction in these cells. [32P]ADP-ribosylation with pertussis toxin and immunoblots using antibodies specific for the Gi alpha-subunits indicated that two distinct Gi-proteins (Gi2 and Gi3) were present in HT29-cell membranes. Treatment of intact cells with pertussis toxin resulted in a time-dependent decrease in the amount of [32P]ADP-ribosylatable Gi2 and Gi3, which coincided with a diminution in the number of alpha 2-adrenergic receptors in high-affinity state for agonists and with a progressive loss of ability of UK14304 to inhibit forskolin-stimulated accumulation of cyclic AMP. When membranes were [32P]ADP-ribosylated with cholera toxin in the absence of exogenous added guanine nucleotides, radioactivity was incorporated into a 45 kDa polypeptide representing Gs, as well as into 40-41 kDa polypeptides corresponding to Gi3 and Gi2. The amount of radioactivity incorporated into the two GiS under basal conditions was decreased by addition of the alpha 2-antagonist RX821002. It was not significantly affected by addition of clonidine (partial alpha 2-agonist), but was doubled by the addition of UK14304 (full alpha 2-agonist). This effect was blocked by RX821002. Study of adenylate cyclase activity indicated that preincubation of HT29 membranes with the antibody AS/7 (anti-alpha i1/alpha i2), but not with the antibody EC/2 (anti-alpha i3), attenuated the inhibitory effect of UK14304 on forskolin-stimulated adenylate cyclase. These data demonstrate that the alpha 2A-adrenergic receptor is coupled to both Gi2 and Gi3, and identify Gi2 as the major mediator of inhibition of adenylate cyclase in HT29 cells.

Vasoactive intestinal peptide and forskolin regulate proliferation of the HT29 human colon adenocarcinoma cell line.

Although several lines of evidence implicate cAMP in the regulation of intestinal cell proliferation, the precise role of this second messenger in the control of the human colon cancer cell cycle is still unclear. In order to investigate the role of cAMP in HT29 cell proliferation, we have tested the effect of vasoactive intestinal peptide (VIP) and forskolin on DNA synthesis and cell number, focusing on the time-dependent efficacy of the treatment. The cells were arrested in G0/G1 phase by incubation for 24 h in serum-free medium and proliferation was re-initiated by addition of either 85 nM insulin or 0.5% fetal calf serum. In the presence of fetal calf serum, G1/S transition was found to occur earlier than with insulin. Exposure of the HT29 cells to 10(-5) M forskolin in the early stages of growth induction (within 12 h from FCS addition or within 14 h from insulin treatment) resulted in a significant inhibition of DNA synthesis and a delayed entry in the S phase. By contrast, VIP (10(-7) M) was inhibitory only when added within a narrow window (10 to 12 h or 12 to 14 h following FCS or insulin addition, respectively). The difference in efficiency of forskolin and VIP to inhibit cell proliferation may be correlated with their own potency to promote long-lasting cAMP accumulation. The combination of VIP plus forskolin had synergistic effects on both cAMP accumulation and cell-growth inhibition. Taken together, our data indicate that cAMP may act at a step in the late G1 or G1/S transition.

The insulin-secreting cell line, RINm5F, expresses an alpha-2D adrenoceptor and nonadrenergic idazoxan-binding sites.

The pharmacological properties of alpha-2 adrenoceptors and the existence of nonadrenergic idazoxan-binding sites (NAIBS) were investigated in the insulin-secreting cell-line, RINm5F, using [3H]RX821002 and [3H]idazoxan. Analysis of [3H]RX821002 saturation isotherms revealed the presence of a single class of binding sites (Bmax = 47.5 +/- 3.5 fmol/mg protein) having high affinity (Kd = 1.26 +/- 0.18 nM). Inhibition of [3H]RX821002 binding by adrenergic compounds showed that the labeled sites displayed the properties expected for an alpha-2 adrenoceptor. Based on competition data with drugs having alpha-2 adrenoceptor subtype selectivity, the receptor from RINm5F is neither an alpha-2B nor an alpha-2C. It resembles the alpha-2A, but deviates from this subtype because of a weak affinity for yohimbine and rauwolscine. In this respect, RINm5F alpha-2 adrenoceptor is identical to the receptor previously described in rat intestinal mucosa and corresponds to a fourth subtype: alpha-2D. Agonist inhibition curves were better fitted by a two-site model and indicated that about half of the receptor population was under a high-affinity state corresponding to G protein-coupled receptors. [32P]ADP-ribosylation with pertussis toxin and immunodetection with specific antibodies permitted the identification of three distinct G proteins: Gi2, Gi3 and G0. Binding experiments with [3H]idazoxan showed that this imidazoline labeled two types of sites corresponding to alpha-2 adrenoceptors and NAIBS. Analysis of saturation isotherms under binding conditions allowing to discriminate between the two site populations indicated that the density of NAIBS (44 +/- 2 fmol/mg protein) was fairly identical to that of alpha-2 adrenoceptors. The pharmacological properties of NAIBS, as assessed by determining the relative affinity of imidazolinic and nonimidazolinic compounds, reasonably matched that reported in other tissues. Taken together, these data make the RINm5F cell-line 1) the first model in permanent culture known as expressing an alpha-2 adrenoceptor of the alpha-2D subtype; 2) a good system for studying in vitro the respective role of alpha-2 adrenoceptors and NAIBS in the regulation of insulin secretion by beta cells.

Alpha-2 adrenoceptor in rat jejunum epithelial cells: characterization with [3H]RX821002 and distribution along the villus-crypt axis.

Alpha-2 adrenergic receptivity of rat jejunum epithelial cells was studied using the new antagonist radioligand, [3H]RX821002 [( 3H]-2-(2-methoxy-1,4-benzodioxan-2-yl)-2-imidazoline). All the parameters of [3H]RX821002 binding were consistent with the labeling of an alpha-2 adrenoceptor. The use of this probe was moreover extremely convenient, because contrarily to [3H]yohimbine and [3H]rauwolscine, [3H]RX821002 displayed in this tissue a very high affinity (Kd = 0.54 +/- 0.12 nM) and a low level of nonspecific binding (5% at 1 nM [3H]RX821002). Competition studies with various antagonists and agonists showed that the labeled sites were alpha-2-selective and stereospecific. Oxymetazoline was much more potent than chlorpromazine or prazosin suggesting that the receptor is of the alpha-2-subtype. Yohimbine and rauwolscine were equipotent, which is also in agreement with the pharmacological definition of this subtype. These two compounds displayed, however, a rather weak affinity (Ki approximately 40 nM), which is somewhat different with what one should expect for a true alpha-2A adrenoceptor. Altogether the competition data indicated that the alpha-2 adrenoceptor from rat jejunal epithelium is neither an alpha-2A, nor an alpha-2B, nor an alpha-2c adrenoceptor and may belong to a fourth subtype.(ABSTRACT TRUNCATED AT 250 WORDS)