Chronic corticosterone administration down-regulates metabotropic glutamate receptor 5 protein expression in the rat hippocampus.

Several lines of evidence suggest a dysfunctional glutamate system in major depressive disorder (MDD). Recently, we reported reduced levels of metabotropic glutamate receptor subtype 5 (mGluR5) in postmortem brains in MDD, however the neurobiological mechanisms that induce these abnormalities are unclear. In the present study, we examined the effect of chronic corticosterone (CORT) administration on the expression of mGluR5 protein and mRNA in the rat frontal cortex and hippocampus. Rats were injected with CORT (40 mg/kg s.c.) or vehicled once daily for 21 days. The expression of mGluR5 protein and mRNA was assessed by Western blotting and quantitative real-time PCR (qPCR). In addition, mGluR1 protein was measured in the same animals. The results revealed that while there was a significant reduction (-27%, P=0.0006) in mGluR5 protein expression in the hippocampus from CORT treated rats, mRNA levels were unchanged. Also unchanged were mGluR5 mRNA and protein levels in the frontal cortex and mGluR1 protein levels in both brain regions. Our findings provide the first evidence that chronic CORT exposure regulates the expression of mGluR5 and are in line with previous postmortem and imaging studies showing reduced mGluR5 in MDD. Our findings suggest that elevated levels of glucocorticoids may contribute to impairments in glutamate neurotransmission in MDD.

Structure-activity analysis of guanidine group in agmatine for brain agmatinase.

To identify a selective inhibitor of mammalian agmatinase, screening was performed on four analogues of agmatine with modifications directly to the guanidine group, six analogues with modifications to the carbon-amine chain, and one analogue with modifications at both ends of the molecule. Control compounds were aminoguanidine and 7-nitroindazole, known inhibitors of the three isoforms (i, e, n) of nitric oxide synthase (NOS), and arcaine, a known inhibitor of the glutamate NMDA receptor. These compounds were compared for inhibition of rat agmatinase and arginine decarboxylase (ADC) activities. Results were studied by ab initio Hartee-Fock descriptors based on optimized geometries and van der Waals radii. Linear correlations were obtained using various geometric and electronic descriptors of the carbon (C), nitrogen (N), and hydrogen (H) atoms in the guanidine moiety. The best fit equation for percent activity remaining of rat agmatinase was = 0.3225 D + 72.76 D1916 + 64.97 D1920 - 192.58 H21 - 253.09 (r = 0.89), where D is the calculated dipole moment, D1916 and D1920 are the N19-N16 and N19-N20 distances, respectively, and H21 is the charge on H21. This agmatinase equation is distinct from the equations fit for ADC, the three NOS isoforms, and inhibition of NMDA receptor binding.

Localization of agmatine in vasopressin and oxytocin neurons of the rat hypothalamic paraventricular and supraoptic nuclei.

Agmatine (decarboxylated l-arginine), an endogenous ligand of imidazoline and alpha(2) adrenoreceptors, is particularly enriched in the rat hypothalamic paraventricular (PVN) and supraoptic (SON) nuclei. The present study utilized light and electron microscopic immunocytochemical methods to determine the distribution and extent of colocalization of agmatine relative to subpopulations of vasopressin- (VP) and oxytocin- (OT) producing neurons in PVN and SON nuclei. By light microscopy, agmatine-immunoreactive perikarya were found in both the magnocellular and the parvocellular neuronal subdivisions of PVN and SON. Confocal and electron microscopy revealed that agmatine-immunoreactivity (I) within neuronal perikarya was associated with the nuclear membrane as well as mitochondria, Golgi complexes, endoplasmic reticula, and plasmalemma. Additionally, agmatine-I was identified in both axons and axonal terminals, which were enriched in large dense-core vesicles. Dual and triple immunocytochemical labeling experiments also demonstrated that agmatine coexists with VP or OT in most PVN and SON magnocellular neurons. Combinations of iontophoretic injections of Fluorogold into the dorsomedullary complex with immunocytochemical labeling revealed that many retrogradely labeled neurons in the parvocellular region of the PVN contained agmatine-I and either VP or OT. These findings provide evidence that agmatine may function as a modulator of both hypothalamically mediated neuroendocrine and autonomic responses.

The asymptotic closed-loop approach to predictive vector quantizer design with application in video coding.

The basic vector quantization (VQ) technique employed in video coding belongs to the category of predictive vector quantization (PVQ), as it involves quantization of the (motion compensated) frame prediction error. It is well known that the design of PVQ suffers from fundamental difficulties, due to the prediction loop, which have an impact on the convergence and the stability of the design procedure. We propose an approach to PVQ design that enjoys the stability of open-loop design while it ensures ultimate optimization of the closed-loop system. The method is derived for general predictive quantization, and we demonstrate it on video compression at low bit rates, where it provides substantial improvement over standard open and closed loop design techniques. Further, the approach outperforms standard DCT-based video coding.

Toward optimality in scalable predictive coding.

A method is proposed for efficient scalability in predictive coding, which overcomes known fundamental shortcomings of the prediction loop at enhancement layers. The compression efficiency of an enhancement-layer is substantially improved by casting the design of its prediction module within an estimation-theoretic framework, and thereby exploiting all information available at that layer for the prediction of the signal, and encoding of the prediction error. While the most immediately important application is in video compression, the method is derived in a general setting and is applicable to any scalable predictive coder. Thus, the estimation-theoretic approach is first developed for basic DPCM compression and demonstrates the power of the technique in a simple setting that only involves straightforward prediction, scalar quantization, and entropy coding. Results for the scalable compression of first-order Gauss-Markov and Laplace-Markov signals illustrate the performance. A specific estimation algorithm is then developed for standard scalable DCT-based video coding. Simulation results show consistent and substantial performance gains due to optimal estimation at the enhancement-layers.

Agmatine reverses pain induced by inflammation, neuropathy, and spinal cord injury.

Antagonists of glutamate receptors of the N-methyl-d-aspartate subclass (NMDAR) or inhibitors of nitric oxide synthase (NOS) prevent nervous system plasticity. Inflammatory and neuropathic pain rely on plasticity, presenting a clinical opportunity for the use of NMDAR antagonists and NOS inhibitors in chronic pain. Agmatine (AG), an endogenous neuromodulator present in brain and spinal cord, has both NMDAR antagonist and NOS inhibitor activities. We report here that AG, exogenously administered to rodents, decreased hyperalgesia accompanying inflammation, normalized the mechanical hypersensitivity (allodynia/hyperalgesia) produced by chemical or mechanical nerve injury, and reduced autotomy-like behavior and lesion size after excitotoxic spinal cord injury. AG produced these effects in the absence of antinociceptive effects in acute pain tests. Endogenous AG also was detected in rodent lumbosacral spinal cord in concentrations similar to those previously detected in brain. The evidence suggests a unique antiplasticity and neuroprotective role for AG in processes underlying persistent pain and neuronal injury.

Characterization of arginine decarboxylase in rat brain and liver: distinction from ornithine decarboxylase.

We compared the properties of mammalian arginine decarboxylase (ADC) and ornithine decarboxylase (ODC) in rat liver and brain. Mammalian ADC is thermally unstable and associated with mitochondrial membranes. ADC decarboxylates both arginine (Km = 0.75 mM) and ornithine (Km = 0.25 mM), a reaction not inhibited by the specific ODC inhibitor, difluoromethylomithine. ADC activity is inhibited by Ca2+, Co2+, and polyamines, is present in many organs being highest in aorta and lowest in testis, and is not recognized by a specific monoclonal antibody to ODC. In contrast, ODC is thermally stable, cytosolic, and mitochondrial and is expressed at low levels in most organs except testis. Although ADC and ODC are expressed in cultured rat C6 glioma cells, the patterns of expression during growth and confluence are very different. We conclude that mammalian ADC differs from ADC isoforms expressed in plants, bacteria, or Caenorhabditis elegans and is distinct from ODC. ADC serves to synthesize agmatine in proximity to mitochondria, an organelle also harboring agmatine's degradative enzyme, agmatinase, and a class of imidazoline receptor (I2) to which agmatine binds with high affinity.

Is agmatine a novel neurotransmitter in brain?

Recent evidence suggests that agmatine, which is an intermediate in polyamine biosynthesis, might be an important neurotransmitter in mammals. Agmatine is synthesized in the brain, stored in synaptic vesicles in regionally selective neurons, accumulated by uptake, released by depolarization, and inactivated by agmatinase. Agmatine binds to alpha2-adrenoceptors and imidazoline binding sites, and blocks NMDA receptor channels and other ligand-gated cationic channels. Furthermore, agmatine inhibits nitric oxide synthase, and induces the release of some peptide hormones. As a result of its ability to inhibit both hyperalgesia and tolerance to, and withdrawal from, morphine, and its neuroprotective activity, agmatine has potential as a treatment of chronic pain, addictive states and brain injury.

Agmatine: an endogenous ligand at imidazoline receptors is a novel neurotransmitter.

Agmatine, an amine and organic cation, is an endogenous ligand at alpha 2-adrenergic and imidazoline (I-) receptors, to which it binds with high affinity. In addition, agmatine has properties of an endogenous neurotransmitter. Thus, agmatine (a) is locally synthesized in brain by a specific enzyme, arginine decarboxylase; (b) is stored in a large number of neurons with selective distribution in the CNS; (c) is associated with small vesicles in axon terminals that, at least in hippocampus, make synaptic asymmetric (excitatory) synapses on pyramidal cells; (d) is released from synaptosomes in a Ca(2+)-dependent manner; (e) can be enzymatically degraded by agmatinase in synaptosomes; (f) can be inactivated by selective reuptake; (g) blocks the ligand-gated NMDA receptor channel at sites distinct from ligand-binding and polyamine sites; and (h) has systemic actions when administered intraventricularly. Additionally, (i) agmatine is a precursor of brain putrescine and, hence, of higher polyamines, and (j) it competitively inhibits the activity of all isozymes of nitric oxide synthase. Agmatine meets most criteria to establish it as a novel neurotransmitter/neuromodulator in the CNS. However, agmatine differs from forms of clonidine displacing system with respect to distribution, bioactivity, and capacity to interact with antibodies raised to imidazoline-like drugs. Thus, there are multiple endogenous ligands of the imidazoline receptors, one of which is agmatine.

Anti-proliferative and anti-inflammatory actions of imidazoline agents. Are imidazoline receptors involved?

We have shown that cultured vascular smooth muscle cells (VSMC) and brain astroglial cells express I-receptors of the I2 subtype. While imidazoline agents are anti-proliferative in smooth muscle cells, they increase the expression of glial fibrillary acidic protein (GFAP) in astrocytes. Because increases in GFAP suppress the induction of calcium-independent, inducible nitric oxide synthase (NOS-2), we measured whether idazoxan and related imidazolines and agmatine would also suppress the expression of NOS-2. Cultured astrocytes and macrophages, RAW 264.7 cell line, were incubated with lipopolysaccharide (LPS, 1 microgram/ml) or cytokine mixture in the presence of 1-100 microM of idazoxan, agmatine, or other imidazoline agents. Idazoxan potently (IC50 10 microM) decreased the activity of NOS-2 in astrocytes, but was less potent in RAW 264.7 cells. By contrast, agmatine was most potent in RAW 264.7 cells (IC50, 10 microM) but less potent in glial cells and VSMC. Both idazoxan and agmatine decreased the activity of NOS-2 by reducing the levels of enzyme protein as measured by immunoblot and immunocytochemistry. No specific binding of [3H]-idazoxan was observed in RAW 264.7 cell membranes. We conclude that idazoxan, agmatine, and selected imidazoline agents inhibit the expression of NOS-2 and proliferation in primary glial cells and VSMC. While the antiproliferative actions appear mediated by I-receptors of the I2 type, the anti-inflammatory response is probably not mediated by I-receptors but possibly by direct actions on signal transduction enzymes.

Inhibition of astroglial nitric oxide synthase type 2 expression by idazoxan.

Binding of idazoxan (IDA) to imidazoline receptors of the I2 subtype in astrocytes influences astroglial gene expression as evidenced by increased expression of glial fibrillary acidic protein and mRNA. To determine whether IDA affected glial inflammatory gene expression, we tested the effects of IDA on astroglial nitric oxide synthase type-2 (NOS-2) expression. NOS-2 was induced in primary rat astrocytes and C6 glioma cells by incubation with 1 microgram/ml lipopolysaccharide (LPS) plus three cytokines (tumor necrosis factor-alpha, interleukin-1beta, and interferon-gamma) or three cytokines alone. Cells were incubated with 1-100 microM IDA, and at 24 h NOS-2 expression assessed. In astrocytes and C6 cells, preincubation with IDA dose-dependently inhibited nitrite accumulation (IC50 approximately 25 microM), accompanied by a reduction in NOS-2 protein levels and L-citrulline synthesis activity in cell lysates. IDA also inhibited nitrite production in LPS stimulated RAW 264.7 macrophages. In astrocytes, but not C6 cells, longer preincubation times with IDA yielded significantly greater suppression, and maximal suppression (>90%) was achieved after a 8 h preincubation in 100 microM IDA. The degree of inhibition was diminished whether IDA was added after LPS plus cytokine mixture. In contrast to NE, continuous incubation with IDA was required to achieve suppression. IDA reduced induction of NOS-2 protein levels, steady state NOS-2 mRNA levels, and activity of a NOS-2 promoter construct stably transfected in C6 cells. These results show that IDA inhibits NOS-2 activity and protein expression in glial cells and macrophages, and suggest that this occurs by decreasing transcription from the NOS-2 promoter.

Agmatine: an endogenous ligand at imidazoline receptors may be a novel neurotransmitter in brain.

Agmatine, which in other life forms serves as a metabolic intermediate for polyamine biosynthesis, appears to have properties in mammals consistent with its actions as a neurotransmitter/neuromodulator. Thus, agmatine is synthesized unequally in brain by arginine decarboxylase (ADC); is stored in neurons and axon terminals with a heterogeneous distribution; is released from synaptosomes by depolarization; is enzymatically converted by agmatinase to putrescine; interacts not only with alpha2-adrenergic and I-receptors in the CNS, but also may selectively block NMDA receptor channels; and, when administered centrally, has several potent biological actions. Clarification of its role in normal brain function, however, has not yet been fully established, in part because of the absence of agents that selectively affect its biosynthesis or degradation.

Imidazoline receptor proteins are regulated in platelet-precursor MEG-01 cells by agonists and antagonists.

The I1-imidazoline receptor is a novel brainstem modulator of sympathetic outflow that is elevated on platelets and in brains of depressed patients. A positive correlation has been reported (accompanying manuscript) between plasma norepinephrine (NE) concentrations and the densities (Bmax) of platelet I1 binding sites (I1 sites). I1-candidate proteins of 33 kDa and 85 kDa are now identified on Western blots probed with anti-imidazoline receptor antiserum (IRBP antiserum), that correlate with Bmax values for I1 sites. Furthermore, a human megakaryoblastoma cell line (MEG-01) has been used to study the regulation of these proteins on megakaryocytic cells, while bovine adrenal chromaffin cells provide a standard I1 cell type for comparison. Both the 33 kDa and 85 kDa IRBP-immunoreactive bands were enriched in plasma membrane fractions. IRBP antiserum did not cross-react with I2 imidazoline binding sites located on platelet mitochondrial membranes. The 85 kDa band was enhanced under conditions lacking fetal bovine serum (FBS) from the culture medium 6 h prior to harvesting. Conversely, 33 kDa protein was enhanced on MEG-01 cells grown in the presence of 10% FBS; suggesting that a precursor (85 kDa) and product (33 kDa) relationship might be induced by serum. The 85 kDa band was robustly up-regulated in response to imidazoline receptor-sensitive ligands; moxonidine, idazoxan and agmatine (10 microM each for 6 h). NE also up-regulated the 85 kDa IRBP-immunoreactive protein on MEG-01 membranes, but to a lesser extent. Idazoxan, an imidazoline alpha 2-antagonist, off-set its induction of 85 kDa protein by reducing the 33 kDa band. Yohimbine, a non-imidazoline alpha 2-antagonist, was ineffective alone, or in combination with moxonidine (up to 40 microM), but yohimbine blocked NE's induction of the 85 kDa band. Therefore, a rise in either plasma NE and/or endogenous I-site ligands (i.e. agmatine) could explain an elevation of imidazoline receptors observed in depression.

Regional localization of agmatine in the rat brain: an immunocytochemical study.

The distribution of agmatine (decarboxylated arginine) was mapped in the central nervous system (CNS) in the rat. Agmatine-like immunoreactivity was identified by light microscopy, exclusively in the cytoplasm of neuronal perikarya. Immunoreactive neurons were present in the cerebral cortex, predominantly within laminae VI and V and, to a lesser extent, III and mainly in retrosplenial, cingulate, primary somatosensory and auditory cortices, and the subiculum. In the lower brainstem, immunoreactivity was selectively localized to visceral relay nuclei: the nucleus tractus solitarii and pontine parabrachial complex, and periventricular areas including the laterodorsal nucleus, locus coeruleus and dorsal raphe. In the midbrain, immunolabeled cells were concentrated in the ventral tegmental area and periaqueductal gray. In the forebrain, subcortical neurons were labeled predominantly in the preoptic area, amygdala, septum, bed nucleus of the stria terminalis, midline thalamus, and the hypothalamus. Ultrastructural analysis of layer V of the somatosensory cortex demonstrated agmatine-immunoreactivity in neurons, primarily in large dense-core vesicles located in the cytoplasm. Agmatine immunoreactivity was also affiliated with endoplasmic reticulum and the plasmalemma. Cortical neurons and the subiculum were labeled in animals not administered the axonal transport inhibitor, colchicine; thus, may normally contain higher concentrations of the amine than other brain regions. The central distribution of agmatine is consistent with the hypothesis that the amine may be a novel neurotransmitter of neurons involved in behavioral and visceral control.

Metabolism of agmatine in macrophages: modulation by lipopolysaccharide and inhibitory cytokines.

Agmatine is an amine derived from the decarboxylation of arginine by arginine decarboxylase (ADC) and metabolized to putrescine by agmatinase. While prevalent in bacteria and plants, agmatine and its metabolic enzymes have been recently identified in mammalian tissues. In the present study we sought to determine: (a) whether macrophages (cell line RAW 264.7) express ADC and agmatinase, and (b) if the enzymes are regulated by lipopolysaccharide (LPS), and/or by the inhibitory cytokines transforming growth factor-beta (TGF-beta), interleukin-10 (IL-10) and interleukin-4 (IL-4). LPS induced a dose-dependent stimulation of agmatinase, while it decreased ADC, the effect in both cases being maximum at 20 h. As expected, LPS dose-dependently stimulated the inducible nitric oxide synthase activity (iNOS). A strong correlation was observed between the effects of LPS on the agmatine-related enzymes and iNOS. By contrast, exposure to IL-10 and TGF-beta caused a reduction in ADC and agmatinase, whereas IL-4 was ineffective on ADC, but reverted the LPS-induced increase of agmatinase. We conclude that the agmatine pathway may be an alternative metabolic route for arginine in macrophages, suggesting a regulatory role of agmatine during inflammation.

Immunocytochemical localization of an imidazoline receptor protein in the central nervous system.

Imidazoline (I) receptors have been implicated in the regulation of arterial blood pressure and behavior although their distribution in the central nervous system (CNS) remains in question. Presumptive I- receptor sites were detected in the rat central nervous system with a polyclonal antibody to an imidazoline receptor protein (IRP) with binding characteristics of the native receptor. IRP-like immunoreactivity (LI) was detected in neurons and glia by light and electron microscopy. Spinal cord: processes were heavily labeled in superficial laminae I and II of the dorsal horn, lateral-cervical and -spinal nuclei and sympathetic cell column. Medulla: label was concentrated in the area postrema, rostral, subpostremal and central subnuclei of nucleus tractus solitarii, spinal trigeminal nucleus caudalis, and inferior olivary subnuclei. Visceromotor neurons in the dorsal vagal and ambigual nuclei were surrounded by high concentrations of immunoreactive processes. In reticular formation, label was light, though predominant in the intermediate reticular zone and ventrolateral medulla. Pons: label was detected in the neuropil of the periventricular gray, concentrated in the dorsal- and external-lateral subnuclei of lateral parabrachial nucleus, and present intracellularly in the mesencephalic trigeminal nucleus. Midbrain: IRP-LI was most heavily concentrated in the interpeduncular nucleus, nuclei interfascicularis and rostral-linearis, the subcommissural organ, central gray, and in glia surrounding the cerebral aqueduct. Diencephalon: high densities were detected in the medial habenular nucleus, nucleus paraventricularis thalami, other midline-intralaminar thalamic nuclei, the supramammillary and mediobasal hypothalamic nuclei. In the median eminence, immunolabeled processes were restricted to the lamina interna and lateral subependymal zone. Telencephalon: IRP-LI was concentrated in the central amygdaloid nucleus, bed nucleus of stria terminalis and globus pallidus, followed by moderate labeling of the medial amygdaloid nucleus, amygdalostriatal zone and caudoputamen, the hilus of the dentate gyrus, and stratum lacunosum-moleculare of field CA1 of Ammon's horn. The subfornical organ and organum vasculosum lamina terminalis were filled with diffuse granular immunoreactivity. Ultrastructural studies identified IRP-LI within glia and neurons including presynaptic processes. I-receptor(s) localize to a highly restricted network of neurons in the CNS and circumventricular regions lying outside of the blood-brain barrier. Putative imidazoline receptors have a unique distribution pattern, show partial overlap with alpha 2 adrenoreceptors and are heavily represented in sensory processing centers and the visceral nervous system.

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.

Agmatine: a novel neurotransmitter?

Current evidence is consistent with an hypothesis that agmatine meets many criteria for a neurotransmitter-neuromodulator. It is synthesized, stored, and released in brain; is contained in neurons and axon terminals with a heterogeneous distribution; interacts with cell-specific receptors; and elicits biological actions within the central nervous system. Its role in normal brain function, however, has not yet been established, in part because of the absence of agents that selectively affect its biosynthesis or degradation.

Uptake of agmatine into rat brain synaptosomes: possible role of cation channels.

Agmatine (decarboxylated arginine), an endogenous ligand for imidazoline receptors, has been identified in brain where it is synthesized from arginine by arginine decarboxylase. Here we report a mechanism for the transport of agmatine into rat brain synaptosomes. The uptake of agmatine was energy- and temperature-dependent and saturable with a Km of 18.83 +/- 3.31 mM and a Vmax of 4.78 +/- 0.67 nmol/mg of protein/min. Treatment with ouabain (Na+,K+-ATPase inhibitor) or removal of extracellular Na+ did not attenuate the uptake rate. Agmatine transport was not inhibited by amino acids, polyamines, or monoamines, indicating that the uptake is not mediated by any amino acid, polyamine, or monoamine carriers. When we examined the effects of some ion-channel agents on agmatine uptake, only Ca2+-channel blockers inhibited the uptake, whereas a reduction in extracellular Ca2+ increased it. In addition, some imidazoline drugs, such as idazoxan and phentolamine, were strong noncompetitive inhibitors of agmatine uptake. Thus, a selective, Na+-independent uptake system for agmatine exists in brain and may be important in regulating the extracellular concentration of agmatine.

Stimulation of imidazoline receptors inhibits proliferation of human coronary artery vascular smooth muscle cells.

Vascular smooth muscle cells of rat aorta express imidazoline receptors whose stimulation, by drugs or an endogenous ligand, agmatine, inhibits serum-stimulated proliferation. We investigated whether imidazoline receptors are expressed in human vascular smooth muscle cells if their stimulation is antiproliferative. Cultured human coronary artery vascular smooth muscle cells express a nonadrenergic binding site for 3H-idazoxan and an imidazoline receptor-binding protein as revealed by immunocytochemical and immunoblot analyses with a specific antibody. Idazoxan and agmatine dose-dependently inhibited serum-stimulated proliferation of these cells as measured by the incorporation of 3H-thymidine (IC50: 5 and 70 micromol/L, respectively) and serum-stimulated expression of proliferating cell nuclear antigen and cell numbers. The agents inhibited proliferation of human and rat (aorta) smooth muscle cells stimulated by either norepinephrine (6560+/-440 disintegrations per minute norepinephrine versus 3345+/-220 norepinephrine and idazoxan), angiotensin II (7680+/-335 disintegrations per minute angiotensin II versus 3769+/-450 angiotensin II and idazoxan), or platelet-derived growth factor (IC50: 3 micromol/L for idazoxan and 40 micromol/L for agmatine), indicating inhibition of mitosis mediated by G-protein or receptor tyrosine kinase pathways. We conclude that human vascular smooth muscle cells express imidazoline-receptors whose activation inhibits proliferation by interacting at a distal step in an intracellular signal cascade common to G-protein and receptor tyrosine kinase mitogenic pathways. Agmatine, synthesized in endothelium, may act as a paracrine regulator of vascular smooth muscle cell proliferation through imidazoline receptors, and agents acting at this site may be useful in treating vascular hyperplasia.