Evidence for separate receptors for insulin and insulin-like growth factor-I in choroid plexus of rat brain by quantitative autoradiography

Binding of insulin and insulin-like growth factor-I (IGF-I) to the choroid plexus was quantitatively characterized using autoradiography and computer densitometry. Slide-mounted brain slices were incubated in 0.1 nM [125I]-insulin or [125I]-[Thr59]IGF-I. To determine specificity of the binding sites, the labeled peptides were mixed with unlabeled analogues. Autoradiography was done with LKB Ultrofilm and analyzed with a computer image analysis system and program for densitometry. Results showed that binding was time and temperature dependent and reversible. Binding of the iodinated insulin and IGF-I was inhibited by unlabeled peptides in a dose-dependent manner. The rank order of potency of these peptides in competing for the choroid plexus iodoinsulin binding sites was: chicken insulin greater than porcine insulin greater than desoctapeptide insulin greater than IGF-I. IGF-I was more potent than porcine insulin in competing for the choroid plexus iodolGF-I binding sites. Somatostatin was ineffective. Non-linear regression analysis revealed the presence of high- (Kd 1.3 +/- 0.2 nM) and low-affinity (Kd 36 +/- 1.4 nM) binding sites for insulin and a single high-affinity binding site (Kd 3.1 +/- 0.3 nM) for IGF-I in the choroid plexus. There were approximately 50 times more binding sites (Bmax) for IGF-I than for insulin high-affinity sites, whereas the number of low-affinity sites for insulin was about equal to the number of IGF-I high-affinity sites. The results of these binding studies with iodinated insulin and [Thr59]IGF-I support the conclusion that the rat choroid plexus has separate high-affinity receptors for insulin and IGF-I, and that the IGF-I receptors outnumber the insulin receptors.     

Effect of insulin on glucagon binding to isolated rat epididymal adipocytes

Effect of insulin on glucagon binding to rat epididymal adipocytes was studied in vitro. [125I]iodoglucagon binding to isolated adipocytes was increased by preincubation of the cells with insulin. Maximal increase was observed with 7 X 10(-10) M insulin. In Scatchard analysis, [125I]iodoglucagon competition data generated one binding site with a single affinity for glucagon binding in the cells pretreated with buffer alone. Pretreatment of the cells with insulin increased the affinity without changes in the number of binding sites. [125I]iodoglucagon binding to isolated adipocytes was not affected by pretreatment of the cells with luteinizing hormone, follicle-stimulating hormone, growth hormone, or with prolactin. These results suggest that insulin stimulates glucagon binding to adipocytes.     

α-Bungarotoxin Binds to Low-Affinity Nicotine Binding Sites in Rat Brain

Reported differences in the pharmacology and distribution of [3H]nicotine and [125I]alpha-bungarotoxin binding sites in mammalian brain suggest that these ligands label separate receptor sites. Affinity purification of an alpha-bungarotoxin binding protein from rat brain failed to copurify the high-affinity nicotine binding site, which remained in the nonbound soluble fraction after the affinity chromatography step. This confirms the independence of these putative receptor sites. Nevertheless, the binding of [125I]alpha-bungarotoxin to P2 membranes was inhibited by (-)-nicotine (Ki = 9 X 10(-6) M), and this sensitivity was preserved after affinity purification. It is proposed that alpha-bungarotoxin binds to a population of low-affinity nicotine binding sites. Comparison of the enantiomers of nicotine in competition studies at both radioligand binding sites revealed an 80-fold preference for the (-) form at the high-affinity [3H]nicotine binding site, whereas the site labelled by [125I]alpha-bungarotoxin displayed little stereoselectivity. In this respect, the brain alpha-bungarotoxin binding site resembles the nicotinic acetylcholine receptor from Torpedo electric organ.     

Presence of NK2 binding sites in the rat brain

Attempts were made to label tachykinin NK2 binding sites in the adult rat brain using [125I]neurokinin A (NKA) as ligand in the presence of NK1 and NK3 agonist or antagonist to avoid labelling of NK1 and NK3 binding sites, respectively. A high-affinity, specifically NK2-sensitive, [125I]NKA-binding, temperature-dependent, reversible, sensitive to GTPgammaS and correspondence to a single population of binding sites (K(D) and B(max) values: 2.2 nM and 7.3 fmol/mg protein) was demonstrated on hippocampal membranes. Competition studies performed with tachykinins and tachykinin-related compounds indicated that the pharmacological properties of these NK2-sensitive [125I]NKA binding sites were identical to those identified in the rat urinary bladder and duodenum. NKA, neuropeptide K, and neuropeptide gamma, as well as the potent and selective NK2 antagonists SR 144190, SR 48968 and MEN 10627, presented a nanomolar affinity for these sites. The regional distribution of these NK2-sensitive [125I]NKA binding sites differs markedly from those of NK1 and NK3 binding sites, with the largest labeling being found in the hippocampus, the thalamus and the septum. Binding in other brain structures was low or negligible. A preliminary autoradiographic analysis confirmed [125I]NKA selective binding in hippocampal CA1 and CA3 areas, particularly, and in several thalamic nuclei.     

Biochemical and Pharmacological Characterization of Serotonin-O-Carboxymethylglycyl[125I]Iodotyrosinamide5 a New Radioiodinated Probe for 5-HT1B and 5-HT1D Binding Sites

There is a lack of radioactive probes, particularly radioiodinated probes, for the direct labeling of serotonin-1B (5-HT1B) and serotonin-1D (5-HT1D) binding sites. Serotonin-O-carboxymethylglycyltyrosinamide (S-CM-GTNH2) was shown previously to be specific for these two subtypes; we, therefore, linked a 125I to its tyrosine residue. Biochemical and pharmacological properties of S-CM-G[125I]TNH2-binding sites were studied by quantitative autoradiography on rat and guinea pig brain sections. S-CM-G[125I]TNH2 binding is saturable and reversible with a KD value of 1.3 nM in the rat and 6.4 nM in the guinea pig. Binding is heterogeneous, paralleling the anatomical distribution of 5-HT1B sites in the rat and of 5-HT1D sites in the guinea pig. The binding of 0.02 nM S-CM-G[125I]TNH2 was inhibited by low concentrations of 5-HT, S-CM-GTNH2, CGS 12066 B, 5-methoxytryptamine, and tryptamine in both species. Propranolol inhibited the radioligand binding with a greater affinity in the rat than in the guinea pig. Conversely, 8-hydroxy-2-(di-n-propylamino)tetralin inhibited S-CM-G[125I]TNH2 binding with a greater affinity in the guinea pig than in the rat. Other competitors, specific for 5-HT1C, 5-HT2, 5-HT3, and adrenergic receptors, inhibited S-CM-G[125I]TNH2 binding in rat and guinea pig substantia nigra and in other labeled structures known to contain these receptors, but only at high concentrations. S-CM-G[125I]TNH2 is then a useful new probe for the direct study of 5-HT1B and 5-HT1D binding sites.     

Pharmacological characterization of tachykinin septide-sensitive binding sites in the rat submaxillary gland

Binding studies have shown that [125I]NKA is a selective ligand of tachykinin septide-sensitive binding sites from membranes of the rat submaxillary gland. Indeed, this ligand bound with high affinity to a single population of sites. In addition, competition studies indicated that natural tachykinins and tachykinin-related compounds had a similar affinity for these sites than for those labeled with [3H]ALIE-124, a selective ligand of septide-sensitive binding sites. Moreover, selective tachykinin NK2, or NK3 agonists or antagonists exhibited weak or no affinity for [125I]NKA binding sites. As indicated by Ki values of several compounds, the pharmacological characteristics of the septide-sensitive binding sites (labeled with [125I]NKA) largely differ from those of classic NK1 binding sites, as determined on crude synaptosomes from the rat brain using [125I]Bolton-Hunter substance P (SP) as ligand. Indeed, several tachykinins including neurokinin A (NKA), neuropeptide K (NPK), neuropeptide gamma (NKgamma), and neurokinin B, as well as some SP and NKA analogues or C-terminal fragments such as septide, ALIE-124, SP(6-11), NKA(4-10), which have a weak affinity for classic tachykinin NK1 binding sites exhibited a high affinity for the septide-sensitive binding sites. In contrast, SP, classic selective NK1 agonists, and antagonists had a high affinity for both types of binding sites. The presence of a large population of tachykinin septide-sensitive binding sites in the rat submaxillary gland may thus explain why NPK and NPgamma induce salivary secretion and may potentiate the SP-evoked response in spite of the absence of tachykinin NK2 receptors in this tissue.     

Demonstration of specific binding sites for pituitary adenylate cyclase activating polypeptide (PACAP) in rat astrocytes

The high and low affinity binding sites for PACAP were identified in rat astrocytes using [125I]PACAP27 as the labeled ligand. Scatchard analysis of displacement of the bound tracer by unlabeled PACAP27 indicated the existence of two classes of binding sites, with the dissociation constant (Kd) = 1.22 +/- 0.4 nM, the binding maximal capacity (Bmax) = 821 +/- 218 fmols/mg protein for the high affinity binding site, and Kd = 0.59 +/- 0.06 microM, Bmax = 563 +/- 12 pmols/mg protein for the low affinity binding site, respectively. The specificity of [125I]PACAP27 binding was tested using PACAP38 and peptides structurally related to PACAP, such as VIP, GHRF, PHI, secretin and glucagon. PACAP38 completely displaced the binding of [125I]PACAP27 and Scatchard analysis also indicated the presence of two classes of binding sites with similar Kd and Bmax to those for PACAP27. VIP and GHRF competed with [125I]PACAP27, but to a much lesser extent than unlabeled PACAP27 in binding. Other peptides tested did not displace the binding of [125I]PACAP27 at 10(-6) M.     

Effect of pH and buffers on insulin binding to normal and neoplastic mammary cells, fat cells and membrane preparations.

As a function of buffer pH, [125I]-insulin binding to rat mammary cells, rat adipocytes, or membranes prepared therefrom, at 4 degrees or 20 degrees C, showed 2 peaks in different buffers. Specific insulin binding at the pH 7.7. peak (100 +/- 11%) was lower than at pH 8.8 (140 +/- 17%) with no change in nonspecific binding. Although insulin stimulation of glucose uptake into fat cells was highest at pH 7.5, this response was also seen at pH 8.6. Scatchard affinity profiles, or in the kinetics of dissociation. Insulin degradation (< 10%) and binding to insulin antibody were similar over the pH range of 7 to 9.     

Comparison of Insulin Binding Proteins In Plasma and Nuclear Membranes of Obese and Lean Mouse Liver

Abstract

Plasma membranes obtained from obese (ob/ob) and lean (+/+ or +/ob) mouse livers were chemically crosslinked to [¹²⁵I] -insulin and examined by electrophoresis and autoradiography. The pattern of crosslinked hormone was qualitatively similar in obese and lean plasma membranes. A major insulin binding protein of approximately M 120,000 was observed. Two additional bands were apparent, one which remained near the top of the gel and one about M 90,000. A minor band at approximately M 50,000 was also detected. For each of the insulin binding proteins a reduction in the amount of [¹²⁵I]-insulin bound was observed with obese plasma membranes as compared with lean. For all proteins the insulin binding was specific as determined by competition with unlabeled hormone. In addition to plasma membrane receptors, insulin has also been reported to bind to nuclear membranes. The autoradiographic patterns of gels of [¹²⁵]-insulin bound and crosslinked to nuclear membranes from obese and lean mouse livers indicated the presence of proteins of the same M as those described for plasma membranes. Nuclear membrane proteins bound less insulin than plasma membranes and, again, the obese was decreased relative to the lean. Contamination of the nuclear membrane fraction by plasma membranes was ruled out. Scatchard analyses of [¹²⁵]-insul in bound to plasma and nuclear membranes indicated that the decrease in hormone binding in the obese mouse is a result of a reduction in the absolute number of receptors. The findings presented in this study provide additional support for this conclusion by demonstrating that membranes from obese mice are comprised of the same set of apparently unaltered insulin binding proteins. Further, the presence of similar insulin binding proteins in both nuclear and plasma membranes suggests a physiological relationship between these structures with respect to hormone binding and/or in the mechanism of action of insulin.     

Autoradiographic localization of peptide YY and neuropeptide Y binding sites in the medulla oblongata

Peptide YY is a highly potent emetic when given intravenously in dogs. We hypothesized that the area postrema, a small brain stem nucleus that acts as a chemoreceptive trigger zone for vomiting and lies outside the blood-brain barrier, might have receptors that PYY would bind to, in order to mediate the emetic response. We prepared [125I]PYY and used autoradiography to show that high affinity binding sites for this ligand were highly localized in the area postrema and related nuclei of the dog medulla oblongata. Furthermore, the distribution of [125I]PYY binding sites in the rat medulla oblongata was very similar to that in the dog; the distribution of [125I]PYY binding sites throughout the rat brain was seen to be similar to the distribution of [125I]NPY binding sites.     

Autoradiographic reevaluation of the binding properties of 125I-[Leu31,Pro34]peptide YY and 125I-peptide YY3-36 to neuropeptide Y receptor subtypes in rat forebrain

125I-[Leu31,Pro34]peptide YY (PYY) and 125I-PYY3-36, initially described as selective neuropeptide Y Y1 and Y2 receptor ligands, respectively, were recently shown to label also Y4 and Y5 receptors. We used receptor autoradiography to assess whether these ligands can be reliably used to investigate the various neuropeptide Y receptors in rat forebrain. In most of the brain regions examined (in coronal sections at the level of dorsal hippocampus), specific 125I-[Leu31,Pro34]PYY binding was completely inhibited by 1 microM BIBP-3226, a selective Y1 receptor ligand, but unaffected by 10 nM rat pancreatic polypeptide, selectively inhibiting Y4 receptors, suggesting that Y4 receptors are present in negligible numbers compared with Y1 receptors in the areas examined. Significant numbers of BIBP-3226-insensitive 125I-[Leu31,Pro34]PYY binding sites were measured in the CA3 subfield of the hippocampus only, possibly representing Y5 receptors. 125I-PYY3-36 binding was unchanged by 1 microM BIBP-3226, whereas a population of 125I-PYY3-36 binding sites was sensitive to 100 nM [Leu31,Pro34]neuropeptide Y, likely representing Y5 receptors. The possibility of distinguishing between Y2 and Y5 receptors using 125I-PYY3-36 as radioligand was validated by their different regional distribution and their distinct changes 24 h after kainate seizures, i.e., binding to Y5 receptors was selectively decreased in the outer cortex, whereas binding to Y2 receptors was enhanced in the hippocampus. Thus, the use of selective unlabeled compounds is required for distinguishing the various receptor subtypes labeled by 125I-[Leu31,Pro34]PYY and 125I-PYY3-36 in rat brain tissue.     

beta-Adrenergic receptors in rat brain.

125I-Iodohydroxybenzylpindolol ([125I] IHYP), a potent beta-adrenergic receptor antagonist, has been used to study beta-adrenergic receptors in rat brain. Binding of [125I] IHYP (30 pM) to a membrane fraction min and dissociation took place with a half time of about 16 min. Phentolamine (10(-4) M) decreased non-receptor binding but it had no effect on the binding of [125I] IHYP to beta-adrenergic receptors in cortex, cerebellum or caudate. In the presence of phentolamine specific binding (defined as binding which was blocked by 0.3 muM dl-propranolol) represented 70-85% of total binding. The binding of [125I] IHYP was inhibited by beta-adrenergic agonists and antagonists. d-Stereoisomers were 2-3 orders of magnitude less potent than the corresponding 1-isomers. The denstiy of [125I] IHYP binding sites was studied in membrane fractions from cerebral cortex, cerebellum, and caudate nucleus by means of Scatchard analysis. The K(D) of [125I] IHYP was similar in the three regions studied, and the density of [125I] IHYP binding sites was approximately 50% greater in the cortex and caudate than in the cerebellum. The Hill coefficient for the binding of [125I] IHYP to membranes from cerebral cortex was 1.02. The properties of the binding of [125I] IHYP are similar to those which would be expected of binding to beta-adrenergic receptors in vitro.     

Characterization of functional melanotropin receptors in lacrimal glands of the rat

The specific melanotropin (MSH) binding sites of rat lacrimal glands were characterized with respect to anatomic distribution, peptide specificity and selectivity, and coupling to a biological response. Tissue distribution of MSH binding sites was determined by autoradiography following in situ binding of a radiolabeled, biologically active preparation of a superpotent alpha-MSH analog, [125I]-[Nle4,D-Phe7]-alpha-MSH ([125I]-NDP-MSH). Intense, specific (i.e., alpha-MSH-displaceable) [125I]-NDP-MSH binding was observed throughout lacrimal acinar tissue, but not in ducts or stroma. In freshly isolated lacrimal acinar cells, specific binding of [125I]-NDP-MSH was maximal within 30 min and rapidly reversible, with a dissociation half-time of about 15 min. A number of melanotropins [alpha-MSH, [N,O-diacetyl-Ser1]-alpha-MSH, [des-acetyl-Ser1]-alpha-MSH, beta-MSH, ACTH(1-24) and ACTH(1-39)] were recognized by these binding sites, as assessed by their inhibition of [125I]-NDP-MSH binding; NDP-MSH was the most potent (IC50 = 1.3 x 10(-9) M). In contrast, other peptides, including ACTH(4-10) and the nonmelanotropic peptides VIP, substance P, somatostatin, and ACTH(18-39) (CLIP), had no effects on tracer binding. In isolated lacrimal acinar cells, alpha-MSH and NDP-MSH stimulated intracellular cyclic AMP accumulation. We conclude that lacrimal acinar cells express functional receptors recognizing melanotropins, suggesting that the lacrimal gland may be a target for physiological regulation by endogenous melanotropins.     

Characterization and Regulation of Insulin Receptors in Rat Brain

An in vitro receptor binding assay, using filtration to separate bound from free [125I]insulin, was developed and used to characterize insulin receptors on membranes isolated from specific areas of rat brain. The kinetic and equilibrium binding properties of central receptors were similar to those of hepatic receptors. The binding profiles in all tissues were complex and were consistent with binding in multiple steps or to multiple sites. Similar binding properties were found among receptors in olfactory tubercle/bulb, cerebral cortex, hippocampus, striatum, hypothalamus, and cerebellum. High affinity [125I]insulin binding sites (KD = 3-11 nM) were distributed evenly between membranes isolated from P1 and P2 fractions of these brain areas, with the exception of the olfactory tubercle in which binding to P2 membranes was four-fold greater (Bmax = 150 fmol/mg protein). One difference between insulin receptors in brain and peripheral target tissues, however, was observed. Following exposure to 0.17 microM insulin for 3 h at 37 degrees C, the number of specific [125I]insulin binding sites on adipocytes decreased by 40%, while the number of binding sites on minces of cerebral cortex/olfactory tubercle remained constant. The results suggest that although the binding characteristics of central and peripheral insulin receptors are similar, these receptors do not appear to be regulated in the same manner.     

Receptor binding and biological activity of [SerB24]-insulin, an abnormal mutant insulin

[SerB24]-insulin, the second structurally abnormal mutant insulin, and [SerB25]-insulin were semisynthesized and were studied for receptor binding and biological activity. Receptor binding and biological activity determined by its ability to increase 2-deoxy-glucose uptake in rat adipocytes were 0.7-3% of native insulin for [SerB24]-insulin and 3-8% for [SerB25]-insulin. Negative cooperative effect of these analogues was also markedly decreased. Immunoreactivity of [SerB24]-insulin was decreased whereas that of [SerB25]-insulin was normal. Markedly decreased receptor binding of [SerB24]-insulin appeared to be due to substitution of hydrophobic amino acid, Phe, with a polar amino acid, Ser, at B24.     

Methyllycaconitine and (+)-anatoxin-a differentiate between nicotinic receptors in vertebrate and invertebrate nervous systems

Specific high-affinity binding sites for 125I-alpha-bungarotoxin and (-)-[3H]nicotine have been measured in rat brain and locust (Schistocerca gregaria) ganglia. The binding sites for 125I-alpha-bungarotoxin had similar Kd values of 1.5 x 10(-9) and 0.8 x 10(-9) M for rat and locust preparations, respectively; the corresponding values for the (-)-[3H]nicotine-binding site were 9.3 x 10(-9) and 1.7 x 10(-7) M. Methyllycaconitine (MLA) potently inhibited 125I-alpha-bungarotoxin binding in both rat and locust. MLA was a less effective inhibitor of (-)-[3H]nicotine binding whereas (+)-anatoxin-a was a very potent inhibitor at this site in the rat but not in the locust. These data suggest that (+)-anatoxin-a is a useful probe for the high-affinity nicotine-binding receptor in vertebrate brain, whereas MLA is a preferential probe for the subclass of receptor that binds alpha-bungarotoxin.     

Interactions between dendrotoxin, a blocker of voltage-dependent potassium channels, and charybdotoxin, a blocker of calcium-activated potassium channels, at binding sites on neuronal membranes

Dendrotoxin I (DpI) from black mamba venom (Dendroaspis polylepis) has high affinity binding sites on rat brain synaptic membranes. Native DpI displaced [125I]-DpI binding with a Ki of 1 x 10(-10) M, and over 90% of specific binding was displaceable. Charybdotoxin isolated from the Israeli scorpion venom (Leiurus quinquestriatus hebraeus), also displaced [125I]-DpI binding, with a Ki of approximately 3 x 10(-9) M, although the displacement curve was shallower than with native DpI. Both toxins are thought to be high affinity blockers of specific K+ currents. Charybdotoxin selectively blocks some types of Ca2+-activated K+ channels, whereas dendrotoxins only block certain voltage-dependent K+ channels. The interaction between the two types of toxin at the DpI binding site is unexpected and may suggest the presence of related binding sites on different K+ channel proteins.     

A radioimmunoassay for human insulin receptor correlation between insulin binding and receptor mass.

We have developed a radioimmunoassay for human insulin receptor. Serum from a patient with Type B severe insulin resistance was used as anti-insulin receptor antiserum. Pure human placental insulin receptor was used as reference preparation and 125I labeled pure insulin receptor as trace. The radioimmunoassay was sensitive (limit of detection less than 17 fmol), reproducible (inter and intra-assay coefficients of variation 12.5% and 1.6% respectively) and specific (no crossreactivity with pure placental IGF-1 receptor, insulin and glucagon). The anti-insulin receptor antibody was, however, able to differentiate between insulin receptor from human placenta and from rat liver. To determine the number of insulin binding sites per receptor, we measured insulin binding (by insulin binding assay) and insulin receptor mass (by radioimmunoassay) in solubilized aliquots from 5 human placentas. The molar ratio of insulin binding to receptor mass was 0.86 +/- 0.12 when binding was determined with monoiodinated 125I-Tyr A 14-insulin. It was 1.94 +/- 0.27 when randomly iodinated 125I-insulin was used. In conclusion, using a sensitive, reproducible and specific radioimmunoassay, we have measured insulin receptor mass independent of insulin binding. Our data are most compatible with binding of one insulin molecule per human placental insulin receptor.     

Characterization of a new radioiodinated probe for the alpha2C adrenoceptor in the mouse brain

[125I]17alpha-hydroxy-20alpha-yohimban-16beta-(N-4-p6 hydroxyphenethyl)carboxamide or [125I]rauwolscine-OHPC, a new radioiodinated probe derived from rauwolscine was synthesized and its binding characteristics investigated on sections of the mouse caudate putamen. [125I]rauwolscine-OHPC binding was saturable and revealed interaction with a single class of binding sites (KD= 0.171 nM, Bmax = 3082 pCi/mg of tissue). The kinetically derived affinity was in close agreement with the affinity evaluated by saturation experiments: k(-1)/k(+1)(0.0403 min(-1)/114 10(6) M(-1) min(-1))=0.35 nM. Competition studies revealed interaction with one single class of binding sites for each of the twelve compounds tested. The rank of potency suggested an interaction with alpha2 adrenoceptors (atipamezole > or = RX 821002 > yohimbine > (-)epinephrine). Moreover, the good affinity of [125I] rauwolscine-OHPC binding sites for spiroxatrine, yohimbine, WB 4101, the relatively good affinity for prazosin (Ki =37.4 nM) and the affinity ratio prazosin/oxymetazoline (37.4/43.4=0.86) were consistent with an alpha2C selective labelling of [125I]rauwolscine-OHPC. The distribution of [125I]rauwolscine-OHPC binding sites in mouse brain was characterized by autoradiography. The density of binding sites was high in the islands of Calleja, accumbens nucleus, caudate putamen and olfactory tubercles, moderate in the hippocampus, amygdala and anterodorsal nucleus of the thalamus. These findings demonstrated that [125I]rauwolscine-OHPC is a useful radioiodinated probe to label alpha2C adrenoceptors in mouse brain.     

Characterization, distribution and plasticity of atrial natriuretic factor binding sites in brain

The [125I]iodotyrosyl derivative of atrial natriuretic factor [( 125I])ANF) apparently binds to a single class of high affinity sites in guinea pig brain membrane preparations. Ligand selectivity pattern reveals that the structural requirements of brain [125I]ANF binding sites are similar to those reported in most peripheral tissues. In vitro receptor autoradiographic studies demonstrate that the brain distribution of [125I]ANF binding sites is species dependent. In rat, high levels of binding are found in olfactory bulb, subfornical organ, area postrema, choroid plexus, and ependyma. In guinea pig, these regions are also enriched with [125I]ANF binding in addition to various thalamic nucleic, amygdala, hippocampus, and cerebellum. In monkey, high densities of sites are seen in the cerebellar cortex. This suggests that brain ANF receptor sites could mediate ANF effects related to the central integration of cardiovascular parameters, as well as other actions not associated with these systems. As in the periphery, it appears that brain [125I]ANF binding sites are associated with guanylate cyclase. Moreover, the density of [125I]ANF receptor binding sites is altered in certain brain regions in spontaneously hypertensive rats and in cardiomyopathic hamsters, demonstrating the plasticity of brain ANF receptors. Thus, ANF and ANF receptors are complementary facets of a new neurotransmitter-neuromodulator system present in mammalian brain.