Noradrenergic stimulation of BDNF synthesis in astrocytes: mediation via alpha1- and beta1/beta2-adrenergic receptors

Brain-derived neurotrophic factor (BDNF) synthesis in astrocytes induced by noradrenaline (NA) is a receptor-mediated process utilizing two parallel adrenergic pathways: beta1/beta2-adrenergic/cAMP and the novel alpha1-adrenergic/PKC pathway. BDNF is produced by astrocytes, in addition to neurons, and the noradrenergic system plays a role in controlling BDNF synthesis. Since astrocytes express various subtypes of alpha- and beta-adrenergic receptors that have the potential to be activated by synaptically released NA, we focused our present study on the mediatory role of adrenergic receptors in the noradrenergic up-regulation of BDNF synthesis in cultured neonatal rat cortical astrocytes. NA (1 microM) elevates BDNF levels by four-fold after 6 h of incubation. Its stimulation was partly inhibited by either the beta1-adrenergic antagonist atenolol, the beta2-adrenergic antagonist ICI 118,551, or by the alpha1-adrenergic antagonist prazosin, while the alpha2-adrenergic antagonist yohimbine showed no effect. BDNF levels in astrocytes were increased by the specific beta1-adrenergic agonist dobutamine and the beta2-adrenergic agonist salbutamol, as well as by adenylate cyclase activation (by forskolin) and PKA activation (by dBcAMP). However, none of the tested agonists or mediators of the intracellular beta-adrenergic pathways were able to reach the level of NA's stimulatory effect. BDNF cellular levels were also elevated by the alpha1-adrenergic agonist methoxamine, but not by the alpha2-adrenergic agonist clonidine. The increase in intracellular Ca2+ by ionophore A23187 showed no effect, whereas PKC activation by phorbol 12-myristate 13-acetate (TPA) potently stimulated BDNF levels in the cells. The methoxamine-stimulated BDNF synthesis was inhibited by desensitizing pretreatment with TPA, indicating that the alpha1-stimulation was mediated via PKC activation. In conclusion, the synthesis of astrocytic BDNF stimulated by noradrenergic neuronal activity is an adaptable process using multiple types (alpha1 and beta1/beta2) of adrenergic receptor activation.     

Subtypes of alpha 1- and alpha 2-adrenergic receptors.

The adrenergic receptors are members of the superfamily of G protein-coupled receptors. There are three major types of adrenergic receptors: alpha 1, alpha 2, and beta. Each of these three major types can be divided into three subtypes. Within the alpha 1-adrenergic receptors, alpha 1A and alpha 1B subtypes have been defined pharmacologically on the basis of reversible antagonists, such as WB4101 and phentolamine, and the irreversible antagonist chloroethylclonidine. In at least some tissues the mechanism of action of the alpha 1A subtype is related to activation of a calcium channel, whereas the alpha 1B receptor exerts its effect through the second messenger inositol trisphosphate. Both of these receptor subtypes as well as a third, the alpha 1C, have been identified by molecular cloning. Three pharmacological subtypes of the alpha 2-adrenergic receptor have also been identified. Prototypic tissues and cell lines in continuous culture have been developed for each of these subtypes, which facilitated their study. The definition of the alpha 2 subtypes has been based on radioligand binding data and more limited functional data. All three subtypes have been shown to inhibit the activation of adenylate cyclase and thus reduce the levels of cAMP. Three alpha 2-adrenergic receptor subtypes have been identified by molecular cloning in both the human and rat species. There is reasonable agreement between the pharmacological identified subtypes and those identified by molecular cloning.     

Loss of rat ventricular alpha 1-adrenergic receptors during aging

Ventricular alpha 1-adrenergic receptor concentration, measured by specific binding of [3H]-prazosin, decreases by 33% as rats age from 3 to 24 months. No age changes occur in binding affinity for [3H]-prazosin or potency of various alpha-adrenergic agonists and antagonists to displace [3H]-prazosin. The ratio of membrane protein to ventricular wet weight also does not change significantly with age. These results suggest a possible mechanism for loss of cardiovascular alpha-adrenergic responsiveness during aging.     

Subcellular distribution of alpha 1-adrenergic receptors in rat liver

The distribution of alpha 1-adrenergic receptors in rat liver subcellular fractions was studied using the alpha 1-adrenergic receptor ligand [3H]prazosin. The highest number of [3H]prazosin binding sites was found in a plasma membrane fraction followed by 2 Golgi and a residual microsomal fraction, the numbers of binding sites were 1145, 845, 629 and 223 fmol/mg protein, respectively. When the binding in these fractions was compared with the activity of plasma membrane 'marker' enzymes in the same fractions a relative enrichment of [3H]prazosin binding sites was found in the residual microsomes and one of the Golgi fractions. Photoaffinity labelling with 125I-arylazidoprazosin in combination with SDS-polyacrylamide gel electrophoresis revealed the specific binding to 40 and 23 kDa entities in a Golgi fraction, while in plasma membranes the binders had an apparent molecular mass of 36 and 23 kDa. When [3H]prazosin was injected in vivo into rat portal blood followed by subcellular fractionation of liver, a pattern of an initial rapid decline and thereafter a slow decline of radioactivity was noted in all fractions. Additionally, in the two Golgi fractions a transient accumulation of radioactivity occurred between 5 and 10 min after the injection. The ED50 values for displacement of [3H]prazosin with adrenaline was lowest in the plasma membrane fraction, followed by the residual microsomes and Golgi fractions, the values were 10(-6), 10(-5) and 10(-4) mol/l, respectively. On the basis of lack of correlation between distribution of alpha 1-adrenergic antagonist binding and adenylate cyclase activity, differences in the molecular mass of alpha 1-adrenergic antagonist binders, differences in the kinetics of in vivo binding and accumulation of [3H]prazosin and also differences in agonist affinity between plasma membrane and Golgi fractions, it is concluded that alpha 1-adrenergic receptors are localized to low-density intracellular membranes involved in receptor biosynthesis and endocytosis.     

Cachectin/TNF as well as interleukin-1 induces prostacyclin synthesis in cultured vascular endothelial cells

Cachectin/tumor necrosis factor (cachectin/TNF) has been shown to be capable of stimulating prostacyclin (PGI2) production by vascular endothelial cells in vitro. The stimulation of PGI2 by cachectin/TNF is comparable to that observed with interleukin-1, the monokine previously suggested to be the principal mediator of this effect. The ability of cachectin/TNF to stimulate PGI2 production suggests that it may play a role in producing depressed blood pressure or shock. If so, it might be possible to prevent such adverse effects with the aid of anti-inflammatory agents.     

Heterodimerization of alpha 2A- and beta 1-adrenergic receptors

beta- and alpha(2)-adrenergic receptors are known to exhibit substantial cross-talk and mutual regulation in tissues where they are expressed together. We have found that the beta(1)-adrenergic receptor (beta(1)AR) and alpha(2A)-adrenergic receptor (alpha(2A)AR) heterodimerize when coexpressed in cells. Immunoprecipitation studies with differentially tagged beta(1)AR and alpha(2A)AR expressed in HEK-293 cells revealed robust co-immunoprecipitation of the two receptors. Moreover, agonist stimulation of alpha(2A)AR was found to induce substantial internalization of coexpressed beta(1)AR, providing further evidence for a physical association between the two receptors in a cellular environment. Ligand binding assays examining displacement of [(3)H]dihydroalprenolol binding to the beta(1)AR by various ligands revealed that beta(1)AR pharmacological properties were significantly altered when the receptor was coexpressed with alpha(2A)AR. Finally, beta(1)AR/alpha(2A)AR heterodimerization was found to be markedly enhanced by a beta(1)AR point mutation (N15A) that blocks N-linked glycosylation of the beta(1)AR as well as by point mutations (N10A/N14A) that block N-linked glycosylation of the alpha(2A)AR. These data reveal an interaction between beta(1)AR and alpha(2A)AR that is regulated by glycosylation and that may play a key role in cross-talk and mutual regulation between these receptors.     

Photoaffinity labeling of alpha 1-adrenergic receptors of rat heart

The photoaffinity probe [125I]aryl azidoprazosin was used to examine structural aspects of rat left ventricular alpha 1-adrenergic receptor. Autoradiography of sodium dodecyl sulfate-polyacrylamide gel electrophoresis-resolved proteins from photoaffinity-labeled membranes revealed a specifically labeled protein of mass 77 kDa. Adrenergic drugs competed with the photoaffinity probe for binding to the receptor in a manner expected of an alpha 1-adrenergic antagonist. Because the autoradiographic pattern was unaltered by incubating labeled membranes in gel sample buffer containing high concentrations of reducing agents, the binding component of the cardiac alpha 1-adrenergic receptor appears to be a single polypeptide chain. The photoaffinity probe specifically labeled a single protein of approximately 68 kDa in membranes of cardiac myocytes prepared from rat left ventricles. The role played by sulfhydryls in receptor structure and function was also studied. Dithiothreitol (DTT) inhibited [3H]prazosin binding to left ventricular membranes and altered both the equilibrium dissociation constant and maximal number of [3H]prazosin-binding sites but not the ability of the guanine nucleotide guanyl-5'-yl imidodiphosphate to decrease agonist affinity for the receptors. When photoaffinity-labeled membranes were incubated with 40 mM DTT for 30 min at room temperature, two specifically labeled proteins of 77 and 68 kDa were identified. The DTT-induced conversion of the 77-kDa protein to 68 kDa was irreversible with washing, but the effect of DTT on [3H]prazosin binding was reversible. Both 77- and 68-kDa proteins were observed with liver membranes even in the absence of reducing agent. We suggest that the DTT-induced conversion of the 77-kDa protein to 68 kDa is due to enhancement in protease activity by the reductant. These results document that the cardiac alpha 1-adrenergic receptor is a 77-kDa protein, similar in mass to the receptor in liver and other sites. Proteolysis likely accounts for lower Mr forms of this receptor found in cardiac myocytes and in previous publications on hepatic alpha 1-receptors.     

The identification and characteristics of subpopulations of alpha 1-adrenergic receptors

Rat liver and brain alpha 1-adrenergic receptors were purified 500 fold by successive chromatographic steps using heparin- and wheat germ agglutinin-agarose; an affinity matrix constructed by coupling CP85.224 (a derivative of prazosin) to affigel 102. It is shown that the existence in brain of an alpha 1-adrenergic receptor subpopulation, which is structurally distinct from that previously characterized. Chlorethylclonidine, irreversibly inactivates [3H] prazosin binding sites in partially purified membrane preparations of rat liver. Under identical conditions, only 50% of receptors are irreversibly inactivated. Computer modelling of data obtained from the competition by the alpha-antagonists, WB 4101 and phentolamine, for [3H] prazosin binding to partially purified preparations of rat liver is best fit by assuming a single low-affinity site for both ligands. However, the partially purified brain preparations indicates the presence of two affinity binding sites for these antagonists. Prior alkylation of brain receptors with chlorethylclonydyne results in the loss of the low-affinity phentolamine and WB4101 binding sites. These data provide evidence for the existence of a single receptor subpopulation (alpha 1b) in rat liver and for two subpopulations (alpha 1a and alpha 1b) in rat brain. The significance of these results in understanding the signal mechanisms which allow cellular responsiveness to alpha 1-adrenergic receptor agonists is discussed.     

α1-肾上腺素受体激活大鼠心脏腺苷酸活化蛋白激酶

为了验证心脏腺苷酸活化蛋白激酶（AMP-activated protein kinase,AMPK）是否为肾上腺素受体（adrenergic receptor,AR）的下游信号分子,本实验在大鼠心室肌源细胞和大鼠心脏中观察了α-AR对AMPK的激活作用,利用Western blot检测了AMPK-α总蛋白表达量及其172位苏氨酸磷酸化水平。雄性Sprague-Dawley大鼠皮下植入去甲肾上腺素（norepinephrine,NE）,苯肾上腺素（phenylephrine,PE）或者溶剂载体[0．01％（W／V）维生素C]的缓释微泵（osmotic minipump）。NE或PE以每小时0．2 mg／kg的速率持续输注,7 d后用AMPK-α抗体免疫沉淀处理样本并测定AMPK的活性。结果显示,在细胞水平,NE引起的AMPK磷酸化水平增高具有时间依赖和剂量依赖特点, NE处理细胞10 min后AMPK磷酸化水平达到最高峰;NE引起的这种效应对β-AR的拮抗剂普萘洛尔（propranolol）不敏感,但是可以被α1-AR拮抗剂哌唑嗪（prazosin）所阻断。结果提示,α1-AR介导AMPK的磷酸化,但β-AR无此作用。AR激动剂持续灌注7 d后,AMPK的活性在NE（7．4倍）和PE（6．0倍）灌注组较对照组显著增高（P〈0．05,H=6）。PE持续灌注组大鼠与对照组相比无明显的心肌肥厚和组织纤维化变化。本文证明α1-AR激动剂可以增强AMPK的活性,揭示了心脏中α1-AR激动在调控AMPK活性方面的重要作用。深入了解α1-AR介导的AMPK激活可能在心衰治疗中具有重要的临床意义。     

Cross-regulation between G-protein-coupled receptors. Activation of beta 2-adrenergic receptors increases alpha 1-adrenergic receptor mRNA levels.

Stimulation of DDT1 MF-2 vas deferens cells with epinephrine resulted in a time- and dose-dependent loss of alpha 1-adrenergic receptor-specific ligand binding. Regulation of alpha 1-adrenergic receptor mRNA was characterized. In monolayer culture, cells displayed 0.7 +/- 0.05 amol of alpha 1-adrenergic receptor mRNA/microgram of total cellular RNA. Epinephrine, which acts at both alpha 1- and beta 2-adrenergic receptors of DDT1 MF-2 cells, induced a short term (2-8 h) increase (50-70%) in the abundance of alpha 1-adrenergic receptor mRNA. Propranolol, a beta 2-adrenergic receptor antagonist, attenuated the epinephrine-mediated increase in alpha 1-adrenergic receptor mRNA but did not affect the decrease in alpha 1-adrenergic receptor-specific ligand binding. Phentolamine, an alpha 1-adrenergic receptor antagonist, did not attenuate the epinephrine-mediated increase in alpha 1-adrenergic receptor mRNA at 4 h but did block the decrease in alpha 1-adrenergic receptor-specific ligand binding. The half-life of the alpha 1-adrenergic receptor mRNA was approximately 7 h in untreated cells as well as in cells challenged with epinephrine. The epinephrine-promoted increase in alpha 1-adrenergic receptor mRNA was found to result from cross-regulation via beta 2-adrenergic receptors. Cholera toxin, forskolin, as well as the cyclic AMP analog CPT cAMP (8-(4-chlorophenylthio)adenosine 3':5'-cyclic monophosphate) increased the alpha 1-adrenergic receptor mRNA at 4 h, as did epinephrine in the presence of alpha 1-antagonists but not in the presence of a beta-adrenergic antagonist. This is the first report of heterologous up-regulation of mRNA levels of adrenergic receptors. Cross-regulation between alpha 1- and beta 2-adrenergic receptor-mediated pathways at 4 h occurs at the level of mRNA whereas later down-regulation of alpha 1-receptor mRNA and binding proceed via agonist activation of alpha 1-adrenergic receptors.     

Cholinergic stimulation of vascular prostacyclin synthesis

The production of prostacyclin by rings of rabbit aorta was assessed by the radioimmunoassay of 6-K-PGF_1α. In steady-state conditions, the rings released 11 ng 6-K-PGF_1α per 100 mg tissue in 30 min. Acetylcholine increased this output: a significant effect was detected at 1 μM and at 10 μM the amplitude of stimulation was 10-fold. The production of PGE₂ and PGF_2α was also increased, but to a lesser extent. The stimulatory action of acetylcholine was mimicked by carbamylcholine and inhibited by atropine; it was abolished in a calcium-free medium. Dog and rat aorta also produced more 6-K-PGF_1α in response to cholinergic agonists. A short rubbing of the intimal surface of the aorta removed the layer of endothelial cells and completely abolished the cholinergic effect. It is concluded that in the aorta, cholinergic agonists, acting on a muscarinic receptor, stimulate the production of prostacyclin by endothelial cells.     

Age-dependent changes in expression of alpha 1-adrenergic receptors in rat myocardium

The expression of alpha 1-adrenergic receptors within ventricular myocardium of rats ranging in age from 21 days of fetal life to 24 months after birth was measured from [125I] 2-(beta hydroxy phenyl) ethylaminomethyl tetralone binding isotherms. No difference was observed in binding affinity between any of the age groups studied. The number of alpha 1-adrenergic receptors was found to be 60-120% higher in membranes from fetal or immature rats up to 25 days of age when compared with adult animals. The increased expression of alpha 1-adrenergic receptors in the developing heart relative to that observed in adult heart is consistent with the hypothesis that alpha 1-adrenergic receptor stimulation may modulate protein synthesis and growth in mammalian myocardium.     

Phorbol esters inhibit alpha 1-adrenergic effects and decrease the affinity of liver cell alpha 1-adrenergic receptors for (-)-epinephrine

4 beta-Phorbol 12-myristate 13-acetate (PMA) modified the metabolic actions of three calcium-dependent hormones in different ways. The stimulations of glycogenolysis ureogenesis and phosphatidylinositol labeling produced by alpha 1-adrenergic agonist was blocked by the phorbol ester. In contrast, PMA slightly increased the stimulation of ureogenesis produced by low concentration of angiotensin II without modifying the maximal response. No effect of PMA was observed on the stimulation of ureogenesis induced by vasopressin. The stimulation of phosphatidylinositol labeling induced by vasopressin was decreased by PMA, whereas that induced by angiotensin II was not affected. In intact freshly isolated hepatocytes, [3H]prazosin binds with high affinity to a site which displays the characteristics of alpha 1-adrenergic receptor. Competitive inhibition studies with (-)-epinephrine reveal two different sites for this agonist: a high affinity site (Kd 9 nM) and a low affinity site (Kd 2 microM). In the presence of phorbol esters, (-)-epinephrine binding data now show the presence of a single class of low affinity sites, with similar affinity to those present in control cells. Thus, the inhibition of hepatocyte alpha 1-adrenergic action by PMA may be related to the loss of high affinity binding sites caused by the tumor promoter.     

Identification of structurally distinct alpha 2-adrenergic receptors

Recent studies involving a variety of membrane receptors and ion channels indicate that diversity exists among these proteins as evidenced by tissue-specific and developmentally related expression of different isoforms. Alpha 2-Adrenergic receptors, plasma membrane proteins involved in sympathetic neurotransmission, may similarly represent a nonhomogeneous class of binding sites based on the following observations. First, their activation can elicit a wide variety of effector cell responses, which are apparently triggered by at least three different signal transduction mechanisms. Second, alpha 2-adrenergic receptors in various tissues and species exhibit marked differences in their ligand recognition properties. To determine if heterogeneity of the receptor protein itself is involved in generating this diversity, we structurally characterized the alpha 2-adrenergic receptor in two tissues that exhibit the greatest differences in ligand recognition properties, neonatal rat lung and human platelet. We report here that these differences in ligand recognition are maintained after partial receptor purification (50-100-fold) and are associated with distinct differences in the physical and structural properties of the receptor protein. The human platelet and neonatal rat lung receptor differ in the apparent molecular weight of their hormone-binding subunits (human platelet, Mr approximately 64,000 versus neonatal rat lung, Mr approximately 44,000) as well as in the number or type of their associated oligosaccharide moieties. The observed diversity is consistent with expression of isoforms of the alpha 2-adrenergic receptor and suggests the presence of more than one gene encoding similar but distinct receptor proteins.     

Subtype-selective noncompetitive or competitive inhibition of human alpha1-adrenergic receptors by rho-TIA

The 19-amino acid conopeptide (rho-TIA) was shown previously to antagonize noncompetitively alpha(1B)-adrenergic receptors (ARs). Because this is the first peptide ligand for these receptors, we compared its interactions with the three recombinant human alpha(1)-AR subtypes (alpha(1A), alpha(1B), and alpha(1D)). Radioligand binding assays showed that rho-TIA was 10-fold selective for human alpha(1B)-over alpha(1A)- and alpha(1D)-ARs. As observed with hamster alpha(1B)-ARs, rho-TIA decreased the number of binding sites (B(max)) for human alpha(1B)-ARs without changing affinity (K(D)), and this inhibition was unaffected by the length of incubation but was reversed by washing. However, rho-TIA had opposite effects at human alpha(1A)-ARs and alpha(1D)-ARs, decreasing K(D) without changing B(max), suggesting it acts competitively at these subtypes. rho-TIA reduced maximal NE-stimulated [(3)H]inositol phosphate formation in HEK293 cells expressing human alpha(1B)-ARs but competitively inhibited responses in cells expressing alpha(1A)- or alpha(1D)-ARs. Truncation mutants showed that the amino-terminal domains of alpha(1B)- or alpha(1D)-ARs are not involved in interaction with rho-TIA. Alanine-scanning mutagenesis of rho-TIA showed F18A had an increased selectivity for alpha(1B)-ARs, and F18N also increased subtype selectivity. I8A had a slightly reduced potency at alpha(1B)-ARs and was found to be a competitive, rather than noncompetitive, inhibitor in both radioligand and functional assays. Thus rho-TIA noncompetitively inhibits alpha(1B)-ARs but competitively inhibits the other two subtypes, and this selectivity can be increased by mutation. These differential interactions do not involve the receptor amino termini and are not because of the charged nature of the peptide, and isoleucine 8 is critical for its noncompetitive inhibition at alpha(1B)-ARs.     

Baroreceptors, alpha1-adrenergic receptors, and regulation of mesenteric blood flow

In the mesenteric circulation of the rat a myogenic autoregulatory system operates at 0.1-0.15 Hz. Negative admittance phase in the region above 0.2 Hz suggested operation of an arterial baroreflex. The present study was designed to test this interpretation and to identify the neurotransmitter involved. In rats anesthetized with isoflurane, blood pressure and mesenteric blood flow (transit time ultrasound) were measured with central mechanisms intact, after sinoaortic denervation, and after denervation of the mesenteric bed. Sinoaortic denervation abrogated the negative phase in the band from 0.3 to 0.6 Hz and increased admittance gain in this region. Subsequent mesenteric denervation had no further effect on the pressure-flow transfer function. In a separate experiment, alpha1-adrenergic blockade reduced, but did not remove, the negative admittance phase in the 0.2- to 0.5-Hz band without altering admittance gain. It is concluded that the baroreflex acting on the mesenteric circulation can be identified by admittance phase, but that admittance gain is uninformative. Part of the response is mediated by alpha1-adrenergic transmission.     

Autoradiographic visualization of alpha 1-adrenergic receptors in cervix of early pregnant rat

alpha 1-Adrenergic receptors were identified, characterized, and localized in rat cervix on Day 6 of pregnancy by autoradiography. Autoradiographic study was performed in slide-mounted rat cervix sections using [3H]-prazosin ([3H]-PRAZ) as ligand. Binding was time dependent and specific. Pharmacological study indicated that specific [3H]-PRAZ binding was inhibited with high affinity by prazosin and phenylephrine and low affinity by yohimbine and clonidine. In cervix, the alpha 1-adrenergic receptors were localized mainly to the inner circular layer of the myometrium. Binding to the outer longitudinal layer of myometrium was moderate, and binding was absent in the endometrium. The regional distribution of alpha 1-adrenergic receptors strongly suggests that the circular layer of myometrium may function as an important modulator of contractile response of the cervix, probably involved in the retention of blastocysts at the utero-cervical end of the horn.     

The prostacyclin receptor induces human vascular smooth muscle cell differentiation via the protein kinase A pathway

Recent studies of cyclooxygenase-2 (COX-2) inhibitors suggest that the balance between thromboxane and prostacyclin is a critical factor in cardiovascular homeostasis. Disruption of prostacyclin signaling by genetic deletion of the receptor or by pharmacological inhibition of COX-2 is associated with increased atherosclerosis and restenosis after injury in animal models and adverse cardiovascular events in clinical trials (Vioxx). Human vascular smooth muscle cells (VSMC) in culture exhibit a dedifferentiated, migratory, proliferative phenotype, similar to what occurs after arterial injury. We report that the prostacyclin analog iloprost induces differentiation of VSMC from this synthetic, proliferative phenotype to a quiescent, contractile phenotype. Iloprost induced expression of smooth muscle (SM)-specific differentiation markers, including SM-myosin heavy chain, calponin, h-caldesmon, and SM alpha-actin, as determined by Western blotting and RT-PCR analysis. Iloprost activated cAMP/protein kinase A (PKA) signaling in human VSMC, and the cell-permeable cAMP analog 8-bromo-cAMP mimicked the iloprost-induced differentiation. Both myristoylated PKA inhibitor amide-(14-22) (PKI, specific PKA inhibitor), as well as ablation of the catalytic subunits of PKA by small interfering RNA, opposed the upregulation of contractile markers induced by iloprost. These data suggest that iloprost modulates VSMC phenotype via G(s) activation of the cAMP/PKA pathway. These studies reveal regulation of VSMC differentiation as a potential mechanism for the cardiovascular protective effects of prostacyclin. This provides important mechanistic insights into the induction of cardiovascular events with the use of selective COX-2 inhibitors.