Epoxyeicosatrienoic acid-induced relaxation is impaired in insulin resistance.

We assessed the effect of epoxyeicosatrienoic acids (EETs) in intact mesenteric arteries and Ca(2+)-activated K(+) (BK(Ca)) channels of isolated vascular smooth muscle cells from control and insulin-resistant (IR) rats. The response to 11,12-EET and 14,15-EET was assessed in small mesenteric arteries from control and IR rats in vitro. Mechanistic studies were performed in endothelium intact or denuded arteries and in the presence of pharmacological inhibitors. Moreover, EET-induced activation of the BK(Ca) channel was assessed in myocytes in both the cell-attached and the inside-out (I/O) patch-clamp configurations. In control arteries, both EET isomers induced relaxation. Relaxation was impaired by endothelium denudation, N(omega)-nitro-L-arginine, or iberiotoxin (IBTX), whereas it was abolished by IBTX + apamin or charybdotoxin + apamin. In contrast, the EETs did not relax IR arteries. In control myocytes, the EETs increased BK(Ca) activity in both configurations. Conversely, in the cell-attached mode, EETs had no effect on BK(Ca) channel activity in IR myocytes, whereas in the I/O configuration, BK(Ca) channel activity was enhanced. EETs induce relaxation in small mesenteric arteries from control rats through K(Ca) channels. In contrast, arteries from IR rats do not relax to the EETs. Patch-clamp studies suggest impaired relaxation is due to altered regulatory mechanisms of the BK(Ca) channel.

Angiotensin II relaxes microvessels via the AT(2) receptor and Ca(2+)-activated K(+) (BK(Ca)) channels.

Angiotensin II (Ang II) is one of the most potent vasoconstrictor substances, yet paradoxically, Ang II may dilate certain vascular beds via an undefined mechanism. Ang II-induced vasoconstriction is mediated by the AT(1) receptor, whereas the relative expression and functional importance of the AT(2) receptor in regulating vascular resistance and blood pressure are unknown. We now report that Ang II induces relaxation of mesenteric microvessels and that this vasodilatory response was unaffected by losartan, an AT(1) receptor antagonist, but was inhibited by PD123,319, a selective antagonist of AT(2) receptors. In addition, reverse transcriptase-polymerase chain reaction studies revealed high amounts of AT(2) receptor mRNA in smooth muscle from these same microvessels. Ang II-induced relaxation was inhibited by either tetraethylammonium or iberiotoxin, suggesting involvement of the large-conductance, calcium- and voltage-activated potassium (BK(Ca)) channel. Subsequent whole-cell and single-channel patch-clamp studies on single myocytes demonstrated that Ang II increases the activity of BK(Ca) channels. As in our tissue studies, the effect of Ang II on BK(Ca) channels was inhibited by PD123,319, but not by losartan. In light of these consistent findings from tissue physiology, molecular studies, and cellular/molecular physiology, we conclude that Ang II relaxes microvessels via stimulation of the AT(2) receptor with subsequent opening of BK(Ca) channels, leading to membrane repolarization and vasodilation. These findings provide evidence for a novel endothelium-independent vasodilatory effect of Ang II.

cAMP-dependent vasodilators cross-activate the cGMP-dependent protein kinase to stimulate BK(Ca) channel activity in coronary artery smooth muscle cells.

cAMP-dependent vasodilators are used to treat a variety of cardiovascular disorders; however, the signal transduction pathways and effector mechanisms stimulated by these agents are not fully understood. In the present study we demonstrate that cAMP-stimulating agents enhance the activity of the large-conductance, calcium-activated potassium (BK(Ca)) channel in single myocytes from coronary arteries by "cross-activation" of the cGMP-dependent protein kinase (protein kinase G, PKG). Single-channel patch-clamp data revealed that 10 micromol/L isoproterenol, forskolin, or dopamine opens BK(Ca) channels in coronary myocytes and that this effect is attenuated by inhibitors of PKG (KT5823; Rp-8-pCPT-cGMPS), but not by inhibiting the cAMP-dependent protein kinase (protein kinase A, PKA). In addition, a membrane-permeable analog, CPT-cAMP, also opened BK(Ca) channels in these myocytes, and this effect was reversed by KT5823. Direct biochemical measurement confirmed that dopamine or forskolin stimulates PKG activity in coronary arteries but does not elevate cGMP. Finally, the stimulatory effect of cAMP on BK(Ca) channels was reconstituted in a cell-free, inside-out patch by addition of purified PKG activated by either cGMP or cAMP. In contrast, channel gating was unaffected by exposure to the purified catalytic subunit of PKA. In summary, findings from on-cell and cell-free patch-clamp experiments provide direct evidence that cAMP-dependent vasodilators open BK(Ca) channels in coronary myocytes by cross-activation of PKG (but not via PKA). Biochemical assay confirmed this cross-activation mechanism of cAMP action in these arteries. This signaling pathway is a novel mechanism for regulation of potassium channel activity in vascular smooth muscle and other cells.

A novel transduction mechanism mediating dopamine-induced vascular relaxation: opening of BKCa channels by cyclic AMP-induced stimulation of the cyclic GMP-dependent protein kinase.

Dopamine dilates the coronary, renal and other vascular beds; however, the signaling pathway underlying this effect is unclear. In this study the signal-transduction process mediating dopamine-induced relaxation of porcine coronary arteries was investigated in isolated vessels and single arterial myocytes. Dopamine-induced relaxation of arteries was mediated through the DA- receptor and involved K+ efflux, and subsequent patch-clamp studies demonstrated that either dopamine or fenoldopam, a selective DA-1 agonist, increased the opening probability of the large-conductance, calcium- and voltage-activated K+ (BKCa) channel in coronary myocytes. Moreover, blockade of this channel by iberiotoxin prevented dopamine-induced coronary relaxation. Dopamine stimulation of BKCa channels was completely prevented by a DA-1-receptor antagonist, but was unaffected by propranolol. Furthermore, inhibiting adenylyl cyclase activity prevented stimulation of BKCa channel activity, whereas chlorophenylthio (CPT)-cyclic adenosine monophosphate (AMP), a membrane-permeable analog of cyclic AMP, mimicked the effects of dopamine. Interestingly, inhibiting the cyclic AMP-dependent protein kinase (PKA) did not affect the response to dopamine, whereas dopamine-induced channel activity was completely blocked by inhibiting the activity of the cyclic guanosine monophosphate (GMP)-dependent protein kinase (PKG). These findings demonstrate that activation of DA-1 receptors causes stimulation of BKCa channel activity by a mechanism involving cyclic AMP-dependent stimulation of PKG, but not PKA, and further suggest that this cross-reactivity mediates dopamine-induced coronary vasodilation.

Nitrovasodilators relax mesenteric microvessels by cGMP-induced stimulation of Ca-activated K channels.

Nitric oxide (NO) released from endothelial cells or exogenous nitrates is a potent dilator of arterial smooth muscle; however, the molecular mechanisms mediating relaxation to NO in the microcirculation have not been characterized. The present study investigated the relaxant effect of nitrovasodilators on microvessels obtained from the rat mesentery and also employed whole cell and single-channel patch-clamp techniques to identify the molecular target of NO action in myocytes from these vessels. Both sodium nitroprusside (SNP) and S-nitroso-N-acetylpenicillamine (SNAP) relaxed phenylephrine-induced contractions by approximately 80% but were significantly less effective in relaxing contractions induced by 40 mM KCl. Relaxation to SNP was also inhibited by the K(+)-channel blocker tetraethylammonium or by inhibition of the activity of the guanosine 3',5'-cyclic monophosphate (cGMP)-dependent protein kinase (PKG). These results suggest that SNP stimulated K+ efflux by opening K+ channels via PKG-mediated phosphorylation. Perforated-patch experiments revealed that both SNP and SNAP increased outward currents in microvascular myocytes, and single-channel studies identified the high-conductance Ca(2+)- and voltage-activated K+ (BKCa) channel as the target of nitrovasodilator action. The effects of nitrovasodilators on BKCa channels were mimicked by cGMP and inhibited by blocking the activity of PKG. We conclude that stimulation of BKCa-channel activity via cGMP-dependent phosphorylation contributes to the vasodilatory effect of NO on microvessels and that a direct effect of NO on BKCa channels does not play a major role in this process. We propose that this mechanism is important for the therapeutic effect of nitrovasodilators on peripheral resistance and arterial blood pressure.

Whole-cell and perforated patch recordings of four distinct K+ currents in acutely dispersed coeliac-superior mesenteric ganglia neurons of adult rats.

Properties and modulation of outward membrane currents in sympathetic neurons acutely dispersed from coeliac-superior mesenteric ganglia (C-SMG) of adult rats were examined using both the whole-cell variant of the patch-clamp technique and the perforated patch approach. Under voltage-clamp, four distinct outward currents were observed: a transient outward current (IA), a voltage-dependent sustained outward current consisting of a Ca(2+)-dependent component (IKCa) and a Ca(2+)-insensitive component (IKV), and a muscarinic agonist-sensitive outward current (IM). IA was isolated by digital subtraction, and characterized by very rapid activation at potentials more positive than -60 mV and by fast and complete voltage-dependent inactivation. Half inactivation potential (Vh) and slope factor (K) were -76 mV and 8.3 mV, respectively. IA was not affected by removal of external Ca2+, 1 mM tetraethylammonium ions, muscarinic agonists, or 8-bromo-cyclic AMP, but was suppressed by 4-aminopyridine (1 mM). Depolarizing pulses from of a holding potential of -50 or -60 mV to potentials more positive than -25 mV concomitantly activated two, independent sustained outward currents which decayed slowly; one exhibited voltage-dependent activation similar to the delayed rectifier current (IKV) and the other being triggered by Ca2+ influx into the cell (IKCa). The addition of tetraethylammonium ions (1 mM) strongly reduced the amplitude of the sustained outward currents. IM was characterized as a noninactivating time- and voltage-dependent outward current which activated at membrane potentials more positive than -60 mV and slowly turned off when the membrane was hyperpolarized back to -60 mV, and was suppressed by muscarinic agonists. The rank order of potency of the agonists tested was: oxotremorine > muscarine > bethanechol.

Interactions of a novel catecholamine, GP-2-128, with adrenoceptors.

GP-2-128 is a novel catecholamine designed for transdermal iontophoretic delivery in patients with limited mobility to prevent deconditioning and muscular wasting. We characterized the interactions of this agent with alpha- and beta-adrenoceptors in vitro. In electrically stimulated rat left atria, GP-2-128 produced a concentration-dependent increase in contractile force. pD2 values for GP-2-128, isoproterenol (ISO), and dobutamine (DOB) were 10.6 +/- 0.12, 8.55 +/- 0.02, and 7.0 +/- 0.20, respectively. Metoprolol caused a shift in the concentration-effect curves for the three agonists. In spontaneously beating rat right atria, pD2 values of GP-2-128, ISO, and DOB are 10.4 +/- 0.24, 8.82 +/- 0.18, and 6.92 +/- 0.18, respectively. The affinity constant (KA) of GP-2-128, ISO, and DOB for cardiac beta 1-adrenoceptors was determined by competition binding assays to be 8.09, 6.04 and 4.49, respectively. In guinea pig trachea precontracted with histamine, GP-2-128 and ISO produced a concentration-dependent relaxation. pD2 values were 10.0 +/- 0.1 and 8.2 +/- 0.1, respectively. DOB was more potent than GP-2-128 in contracting isolated rat aortic rings (alpha 1 effect) and in displacing [3H]rauwolscine (alpha 2 effect). We also studied the interactions of GP-2-128 and ISO with the atypical beta-adrenoceptors (beta 3) in guinea pig ilea and rat and hamster adipocytes. Both agents inhibited twitches produced by transmural nerve stimulation in the presence of 10(-5) M nadolol. The EC30 for GP-2-128 and ISO at this atypical receptor site were 4.25 x 10(-10) and 5.05 x 10(-8) M, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

TTX-sensitive Na+ channels and Ca2+ channels of the L- and N-type underlie the inward current in acutely dispersed coeliac-mesenteric ganglia neurons of adult rats.

Inward membrane currents of sympathetic neurons acutely dispersed from coeliac-superior mesenteric ganglia (C-SMG) of adult rats were characterized using the whole-cell variant of the patch-clamp technique. Current-clamp studies indicated that C-SMG neurons retained electrical properties similar to intact ganglia. Voltage-clamp studies designed to isolate Na+ currents revealed that tetrodotoxin (TTX, 1 microM) completely inhibited the large transient inward current. Half activation potential (Vh) and slope factor (K) were -26.8 mV and 6.1 mV, respectively. Inactivation parameters for Vh and K were -65 mV and 8.2 mV, respectively. Voltage-clamp studies also revealed a high-voltage-activated sustained inward Ca2+ current which was blocked by the removal of external Ca2+ or the presence of Cd2+ (0.1 mM). The dihydropyridine agonist, (+)202-791 (1 microM), caused a small increase (20%) in the amplitude of the Ca2+ current at more negative potentials and markedly prolonged the tail currents. omega-Conotoxin GIVA (omega, CgTX, 15 microM) caused a 66% inhibition of the high-voltage-activated Ca2+ current amplitude. Norepinephrine (1 microM) caused a 49% reduction in the peak Ca2+ current. This study is the first demonstration that dispersed C-SMG neurons from adult rats retain electrical characteristics similar to intact ganglia. A TTX-sensitive Na+ current as well as a high voltage-activated sustained Ca2+ current underlie the inward current in C-SMG neurons. The macroscopic Ca2+ current is composed of a small dihydropyridine-sensitive (L-type current) and a large omega-CgTx-sensitive (N-type current) component.(ABSTRACT TRUNCATED AT 250 WORDS)

Increased muscarinic responsiveness and decreased muscarinic receptor content in ileal smooth muscle in diabetes.

Longitudinal muscles from the ileum of rats rendered diabetic for 10 to 12 weeks by injection of streptozotocin (STZ) developed greater contractile force in response to muscarinic agonists, (acetylcholine, carbamylcholine and bethanechol) than muscles from age-matched control rats. There was no change, however, in the sensitivity of longitudinal smooth muscles to muscarinic agonists as reflected by the EC50 values for stimulation of muscle contraction. This increased responsiveness of the muscles was accompanied by a 32% reduction in the density of muscarinic receptors (381 +/- 93 vs. 560 +/- 74 fmol/mg of membrane protein) in muscles from STZ-diabetic rats. There was no change in agonist or antagonist binding affinities in muscles from diabetic rats, and there was no alteration in the distribution of receptors between the states characterized by high and low affinity agonist binding. There was also no change in acetylcholinesterase activity in muscle membranes isolated from STZ-diabetic rats. The origin of the increased responsiveness of ileal smooth muscles in this model of diabetes is not clear, but may involve changes in the muscarinic signal transduction pathway beyond the receptor level, or in the contractile apparatus itself.

Vascular contraction induced by activation of membrane calcium ion channels is enhanced in streptozotocin-diabetes.

Previous results from our laboratory (White and Carrier, Enhanced Vascular Alpha-Adrenergic Neuroeffector System in Diabetes: Importance of Calcium. Am. J. Physiol. 255: H1036-1042, 1988) demonstrated that mesenteric arteries from streptozotocin (STZ)-diabetic rats exhibit an enhanced responsiveness to alpha adrenergic agonists. The present study demonstrates that this enhanced responsiveness is dependent upon the presence of extracellular calcium. Arteries from STZ-diabetic (10-12 weeks) rats developed greater contractile force in response to norepinephrine or KCl. Development of these effects was prevented by daily insulin treatment, indicating these alterations are related to the diabetic state. Similarly, the contractile response to extracellular calcium in the presence of norepinephrine (3 x 10(-6) M) or KCl (60 mM) was greater in arteries from STZ-diabetic animals. BAY K 8644, a calcium channel agonist, induced greater contraction in arteries from STZ-diabetic animals, as did activation of protein kinase C by phorbol dibutyrate. In contrast, contraction induced by release of calcium from intracellular sources (alpha-1 adrenoceptor-mediated or caffeine-induced) was unaltered by diabetes. These findings indicate that enhanced vascular contraction in STZ-diabetes is of a nonspecific nature, i.e., the contractile response to any agent which induces extracellular calcium-dependent contraction should be enhanced in diabetes. We propose that STZ-diabetes enhances the activity and/or number of calcium ion channels in vascular smooth muscle.

Insulin prevention of altered muscarinic receptor-G protein coupling in diabetic rat atria.

Right atria from rats rendered diabetic by injection of streptozocin (STZ-D) for 8-10 wk are supersensitive to the negative chronotropic effects of muscarinic agonists but have decreased levels of muscarinic receptors and acetylcholinesterase activity. Insulin treatment completely prevents the development of these changes. The proportion of atrial muscarinic receptors displaying high-affinity agonist binding is lower in STZ-D rats; however, the sensitivity of high-affinity agonist binding to regulation by a guanine nucleotide (5'-guanylylimidodiphosphate) is greater in atria from diabetic rats. Again, insulin treatment eliminates these differences. These findings indicate that alterations in atrial muscarinic systems in STZ-D rats are a consequence of the elaboration of the diabetic state and suggest that an alteration of functional muscarinic receptor-G protein coupling contributes to the altered physiological responsiveness of the heart in diabetes.

Enhanced vascular alpha-adrenergic neuroeffector system in diabetes: importance of calcium.

Mesenteric arteries from streptozotocin (STZ)-diabetic rats developed greater contractile force in response to norepinephrine and related alpha-agonists than arteries from age-matched controls. Subsequent experiments attempted to define the mechanisms underlying these findings. Transmural nerve stimulation of mesenteric arteries from both groups of animals revealed a similar optimal frequency and voltage of stimulation; however, arteries from STZ-diabetic rats developed greater contractile force than controls. Second, determination of selective alpha-adrenergic antagonist affinities (pA2 values) revealed qualitatively similar postjunctional alpha 1-adrenoceptors in both groups of arteries. Third, disruption of the endothelium did not abolish the enhanced responsiveness of arteries from STZ-diabetic rats. In contrast, the increased vascular responsiveness in STZ-diabetes was associated with a greater dependency on extracellular calcium, with no change in the response to alpha-agonist-induced release of calcium from cellular stores. Thus the enhanced responsiveness of mesenteric arteries from STZ-diabetic rats to alpha-adrenergic agonists cannot be attributed to neuronal deterioration, altered postjunctional alpha-adrenoceptor subtypes, endothelium degeneration, or enhanced release of intracellular calcium but is associated with a greater dependency on extracellular calcium.

Altered muscarinic receptor properties and function in the heart in diabetes.

The cardiac cholinergic system was studied in streptozotocin (STZ)-diabetic and age-matched control rats. STZ-diabetic rats (8-10 weeks) were supersensitive to the negative chronotropic effects of acetylcholine, carbamylcholine and bethanechol; inotropic responses to these muscarinic agonists were unaltered. This phenomenon was associated with a decrease in acetylcholinesterase activity but no change in the rate and extent of neuronal choline uptake. [3H]N-methylscopolamine bound to muscarinic receptors in atria from both groups of rats with the same high affinity. The density of [3H]N-methylscopolamine binding sites, however, was 34% lower in atria from STZ-diabetic rats. Agonist binding affinity was lower in diabetes; carbamylcholine had a lower affinity for both the high- and low-affinity receptors. These results indicate that cardiac cholinergic supersensitivity in right atria in diabetes occurs before the development of autonomic neuropathy insofar as neuronal [3H]choline uptake is unaltered at this stage of STZ diabetes. Changes in agonist binding conformation, without a concomitant change in antagonist binding affinity, suggest that supersensitivity of right atria to muscarinic agonist may be a consequence of altered coupling of muscarinic receptor to transduction mechanisms involved in chronotropism in diabetes.

Alpha 1- and alpha 2-adrenoceptor agonist-induced contraction in rat mesenteric artery upon removal of endothelium.

Contractile response to norepinephrine, methoxamine and clonidine were determined in rat mesenteric arteries with and without an intact endothelium. Removal of the endothelial cell layer markedly enhanced the maximum contractile effect of norepinephrine (2.8 fold), methoxamine (4.0 fold) and clonidine (13.0 fold). Furthermore, there was a significant decrease in the EC50 values for these agonists. These findings indicate that both alpha 1- and alpha 2-adrenoceptor agonists can induce contraction of mesenteric arteries which is modulated by the endothelial cell layer.

Supersensitivity and endothelium dependency of histamine-induced relaxation in mesenteric arteries isolated from diabetic rats.

Diabetes mellitus is known to produce alterations in vascular reactivity. In the present study we have examined the effects of short-term diabetes on histamine-induced relaxation of isolated mesenteric arteries, and the role of the endothelial cell layer in this response. Removal of the endothelium completely abolished the histamine relaxation effect in both diabetic and age-matched control rats. In contrast, vessels isolated from streptozotocin-diabetic rats were supersensitive to histamine, and this relaxation was mediated only through the H1-receptors. The present findings suggest that histamine-induced relaxation of rat mesenteric arteries is dependent upon endothelial cell processes which are enhanced in arteries from STZ-diabetic rats.

Effects of repetitive stimulation by norepinephrine, histamine or potassium chloride on contractile responses of pulmonary vascular smooth muscle.

We examined consecutive contractile responses of isolated rabbit pulmonary arteries during repeated exposures to either norepinephrine, histamine or KC1, with washout and relaxation between trials. For each agonist, EC50 values remained constant during consecutive determinations, but the maximum force increased after the first determination. A maximum concentration of norepinephrine or histamine produced a biphasic contraction: the fast phase increased subsequent to the first determination and was retained in a calcium-free medium, while the slow phase was unaltered during consecutive determinations and was absent in a calcium-free medium. We conclude that in the rabbit pulmonary artery: the initial contractile response is a useful control for studies of sensitivity but not of maximum activity; there may be a nonspecific increase in the availability of intracellular calcium after the initial contraction and relaxation, and desensitization or tachyphylaxis to these agonists does not occur.

Influence of magnesium on norepinephrine-and histamine-induced contractions of pulmonary vascular smooth muscle.

We examined the effects of magnesium on contractile responses of the rabbit pulmonary artery. The contractile force was determined, after applications of norepinephrine or histamine, in a normal or Ca++-free solution containing 0 mM Mg++ or 1.2 mM Mg++. In a normal solution, Mg++ increased the EC50 value for histamine, but did not alter the EC50 value for norepinephrine or the maximum force induced by norepinephrine or histamine. Contractile responses to norepinephrine and histamine were equally reduced by a Ca++-free, Mg++-free solution, and were further reduced by a Ca++-free solution containing Mg++, but with a greater reduction in the response to histamine than in the response to norepinephrine. The results indicate that in the pulmonary artery, Mg++ alters the sensitivity to histamine but not to norepinephrine, and may differentially inhibit bound Ca++ release by norepinephrine and histamine.

Three phase contractile response of rabbit pulmonary artery to norepinephrine: influence of ruthenium red and calcium.

We recorded isometric contractile force of rabbit pulmonary artery after application of norepinephrine (50 microM), plotted semilogarithmic graphs of force development over time, and calculated rates of force development. The normal contractile response contained three phases: an initial fast, a short intermediate and a final slow. Correlation coefficients for each phase and differences between rates of force development of each phase were significant (P less than 0.05). Ruthenium red (1 mM) removed only the slow phase and significantly reduced the rates of the fast and intermediate phases. A calcium-free solution removed both the slow and intermediate phases and significantly reduced the rate of the fast phase.

Enhanced relaxation response of canine coronary artery to isoproterenol and salbutamol after removal of endothelial cells.

The importance of the endothelium for isoproterenol- and salbutamol-induced relaxation of canine coronary artery was examined. The relaxation effect of isoproterenol and salbutamol was significantly enhanced (P less than 0.05) upon removal of the coronary artery endothelium. In contrast, the relaxation effect of acetylcholine was completely abolished. These results indicate that canine coronary endothelium modulates beta-adrenoceptor-mediated relaxation, and that the role of the endothelium in agonist-induced relaxation of vascular smooth muscle appears to be heterogeneous.

Enhancement of alpha-1 and alpha-2 adrenergic agonist-induced vasoconstriction by removal of endothelium in rat aorta.

It is well known that the vascular endothelial cell layer plays an essential role in the vasodilatory response of several agents. In this study we have investigated the possibility that the endothelium may also modulate alpha adrenergic agonist-induced vasoconstriction. The responses of rat aortae to selective alpha-1 and alpha-2 adrenergic agonists were studied. Removal of the endothelium did not significantly alter the maximum contractile response to norepinephrine. However, the maximum responses to selective alpha-1 agonists (phenylephrine and methoxamine) were increased 2-fold. The vasoconstrictor effects of both clonidine and B-HT920 (selective alpha-2 agonists) were enhanced 5- to 7-fold after removal of the endothelial cell layer. The sensitivity of the tissue, as reflected by the EC50 value, to each alpha adrenergic agonist was enhanced in the absence of endothelium. An explanation for the present results is that alpha-1 and alpha-2 adrenergic agonists activate adrenoceptors in the endothelial cells and thereby may promote the release of a relaxing factor to inhibit vascular smooth muscle contraction. Removal of the endothelium would abolish release of this putative inhibitory substance and adrenergic agonist would activate only adrenoceptors in the muscle to cause vasoconstriction. On the other hand, endothelial cells may function as an uptake site for the various adrenergic agonists. Ablation of this uptake process could conceivably result in a greater effective concentration of the agonist in the receptor area and thus promote a stronger vasoconstrictor effect.(ABSTRACT TRUNCATED AT 250 WORDS)