Margarita P. Bracamonte

Vascular effects of estrogens: arterial protection versus venous thrombotic risk

Mechanisms by which estrogen reduces the risk of arterial disease, while simultaneously increasing the risk of venous thrombosis in postmenopausal women, are not clearly understood. In addition to providing beneficial arterial effects on the lipid profile, estrogen both increases production of nitric oxide and decreases production of endothelin-1 from arterial endothelium, decreases intracellular calcium in arterial smooth muscle and might favor fibrinolysis. All of these effects could act in concert to protect against development of arterial occlusive disease. However, comparable effects on venous endothelium and smooth muscle have not been studied systematically, and although blood elements such as platelets and leukocytes contain estrogen receptors, much remains to be learned about the effect that dose and duration of estrogen-treatment might have upon these cells. An integrative approach to understanding the actions of estrogen on the venous system and the interaction of blood elements with the vascular wall is necessary before new therapeutic interventions will provide arterial protection with no risk of venous thrombosis.

Mechanism of relaxations to dendroaspis natriuretic peptide in canine coronary arteries.

Experiments were designed to determine mechanisms by which Dendroaspis natriuretic peptide (DNP) causes relaxations in coronary arteries. Rings of canine left circumflex artery with and without endothelium were suspended in organ chambers filled with Krebs-Ringer bicarbonate solution (37 degrees C, bubbled with 95% O2/5% CO2). Concentration-response curves to DNP (10(-10) to 3 x 10(-7) M) were obtained in arteries contracted with prostaglandin (PGF(2alpha), 2 x 10(-6) M), either in the absence or the presence of C-ANP (10(-6) M) to inhibit natriuretic clearance receptors; indomethacin to inhibit cyclooxygenase (INDO, 10(-5) M), N(G)-monomethyl-L-arginine to inhibit production of nitric oxide (L-NMMA; 10(-4) M), HS-142-1 to inhibit particulate guanylate cyclase (10(-5) M); 1H-[1,2,4]oxadiazolo-[4,3-alpha]quinoxalin-1-one to inhibit soluble guanylate cyclase (ODQ; 10(-5) M), or tetraethylammonium to inhibit potassium channels (TEA; 10(-3) or 10(-2) M). Relaxations to DNP were greater in rings with than in those without endothelium. C-ANP significantly attenuated relaxations to DNP only in rings with endothelium. HS-146-1 but not INDO, L-NMMA, ODQ, and TEA significantly reduced relaxations to DNP in rings with and without endothelium contracted with PGF(2alpha). These results suggest that the endothelium augments inhibitory effects of DNP and that natriuretic clearance receptors mediate this component of the response to DNP in canine coronary arteries. In addition, relaxations to DNP in canine arterial smooth muscle involve activation of particulate guanylate cyclase but not hyperpolarization.

Erbin and the NF2 Tumor Suppressor Merlin Cooperatively Regulate Cell-Type-Specific Activation of PAK2 by TGF-β

Transforming growth factor beta (TGF-beta) family ligands are pleotropic proteins with diverse cell-type-specific effects on growth and differentiation. For example, PAK2 activation is critical for the proliferative/profibrotic action of TGF-beta on mesenchymal cells, and yet it is not responsive to TGF-beta in epithelial cells. We therefore investigated the regulatory constraints that prevent inappropriate PAK2 activation in epithelial cultures. The results show that the epithelial-enriched protein Erbin controls the function of the NF2 tumor suppressor Merlin by determining the output of Merlins physical interactions with active PAK2. Whereas mesenchymal TGF-beta signaling induces PAK2-mediated inhibition of Merlin function in the absence of Erbin, Erbin/Merlin complexes bind and inactivate GTPase-bound PAK2 in epithelia. These results not only identify Erbin as a key determinant of epithelial resistance to TGF-beta signaling, they also show that Erbin controls Merlin tumor suppressor function by switching the functional valence of PAK2 binding.

Mechanism of raloxifene-induced relaxation in femoral veins depends on ovarian hormonal status

Experiments were designed to study effects of raloxifene, a selective estrogen receptor modulator, on venous endothelium and smooth muscle. Rings of femoral veins with and without endothelium from adult gonadally intact, and ovariectomized female pigs were suspended for measurement of isometric force in organ chambers. Concentration–response curves to raloxifene (10−9–10−5M) were obtained in rings at baseline tension or following contraction with prostaglandin (2 × 10−6M) in the absence or presence of NG-monomethyl-l-arginine (l-NMMA) (nitric oxide synthase inhibitor), 1H-(1.2.4) oxadiazolo (4,3-A) quinoxalin-1-one (ODQ, soluble guanylate cyclase inhibitor), tetraethylammonium acetate (TEA; potassium channel blocker), or indomethacin (cyclooxygenase inhibitor). Raloxifene caused acute, concentration-dependent relaxations that were greater in rings with than in rings without endothelium from both groups. The l-NMMA significantly inhibited relaxations to raloxifene in rings with endothelium from ovariectomized females whereas TEA only inhibited relaxations in rings with endothelium from intact female pigs. ODQ and indomethacin significantly inhibited relaxations in rings with endothelium from both groups. These results suggest that raloxifene acutely relaxes femoral veins through release of endothelium-derived factors and by direct stimulation of vascular smooth muscle cells. Whether nitric oxide or potassium channel activation contributes to relaxations by raloxifene may depend on ovarian hormonal status of the animal.

Treatment with raloxifene and 17beta-estradiol differentially modulates nitric oxide and prostanoids in venous endothelium and platelets of ovariectomized pigs.

Oral treatment with raloxifene, a synthetic estrogen receptor modulator (SERM), or 17β-estradiol (E2) increases risk for venous thrombosis in women. Acute application of either substance releases endothelium-derived factors from isolated femoral veins but it is not known how their chronic use affects venous functions or the interaction of platelets with veins. This study tested the hypothesis that treatment of ovariectomized animals with oral raloxifene or E2 would increase release of proaggregatory factors from venous endothelium and platelets. Ovariectomized (OVX) pigs were either untreated or treated with oral raloxifene (60 mg/day) or E2 (2 mg/day) for 4 weeks. Plasma concentrations of nitric oxide were comparable in both treatment groups and greater than in OVX pigs. Ratio of plasma thromboxane to prostacyclin was twofold greater in raloxifene compared to E2-treated pigs. In isolated femoral veins, NG-monomethyl-L-arginine (L-NMMA; 10−4 M) augmented endothelium-dependent relaxations to adenosine diphosphate in veins from E2-treated pigs but inhibited relaxations in veins from raloxifene-treated pigs. Addition of indomethacin (10−5 M) reversed these effects. Endothelium-dependent relaxations to thrombin were inhibited by L-NMMA only in OVX and raloxifene-treated pigs. Autologous platelets contracted veins in all groups; the magnitude of contractions depended upon the number of platelets and existing tone. Basal release of thromboxane from platelets was greatest in raloxifene compared to OVX or E2-treated pigs. Raloxifene treatment compared to E2 increased production of contractile and proaggregatory prostanoids from venous endothelium and platelets. These differences, if found in humans, may contribute to varying degrees of thrombotic risk with the SERM compared to the natural hormone.

Activation of soluble guanylate cyclase and potassium channels contribute to relaxations to nitric oxide in smooth muscle derived from canine femoral veins.

Experiments were designed to examine mechanisms of relaxations to nitric oxide (NO) in venous smooth muscle. Rings of canine femoral veins without endothelium were suspended for measurement of isometric force in organ chambers. Concentration-response curves to NO and 8-Br-cyclic guanosine monophosphate (cGMP) were obtained in veins contracted with KCl (60 mM) or prostaglandin F2alpha (PGF2alpha; 2x10(-6) M) in the absence and presence of inhibitors of soluble or particulate guanylate cyclase or K+ channel antagonists. In rings contracted with PGF2alpha, relaxations to NO were reduced significantly by inhibition of soluble but not particulate guanylate cyclase. Relaxations to NO were reduced in rings contracted with KCI. Tetraethylammonium (10(-2) M) and glibenclamide (10(-7) M) + charybdotoxin (10(-7) M) significantly reduced relaxations to NO in rings contracted with PGF2alpha. Relaxations to 8-Br-cGMP were decreased significantly only by charybdotoxin. These results suggest that relaxations to NO in canine femoral veins involve at least two intracellular processes: activation of soluble guanylate cyclase and activation of adenosine triphosphate (ATP)-sensitive and large-conductance, Ca+2-activated K+ channels. The large-conductance, Ca+2-activated K+ channels seem to be activated by cGMP-dependent mechanisms. Therefore relaxations to NO in venous smooth muscle involve intracellular processes similar to those in arterial smooth muscle.