INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CORTICOSTEROID HORMONES play an important role in the control of vascular smooth muscle tone by their permissive effects in potentiating vasoactive responses to catecholamines through glucocorticoid receptors. Increased cortisol response has been associated with an increase in arterial contractile sensitivity to norepinephrine (NE) and vascular resistance (5, 22, 23, 39-42). Despite the fundamental importance of cortisol in regulating sympathetic-mediated contraction of vascular smooth muscle, little is currently known about the cellular mechanisms of vascular smooth muscle in response to cortisol.

During pregnancy in several species, including humans and sheep, maternal plasma cortisol concentrations approximately double (21, 28). In sheep, cortisol plasma levels were increased from ~5 ng/ml in nonpregnant animals to ~10 ng/ml in pregnant animals (21). Given the importance of precise regulation of uterine blood flow for fetal growth and maternal cardiovascular well being during pregnancy, a study of the effect of cortisol on the regulation of uterine artery contraction is fully warranted. Recently, we (45) demonstrated that cortisol potentiates NE-mediated contractions of ovine uterine artery by decreasing nitric oxide release and increasing NE binding affinity to ₁-adrenoceptors. Comparison of cortisol-mediated responses in the uterine arteries obtained from nonpregnant and near-term pregnant (140 days gestation) sheep indicated that pregnancy attenuated uterine artery sensitivity to cortisol (45), which is likely to be important in maintaining a low vascular reactivity of the uterine artery to NE during pregnancy. Our finding that glucocorticoid receptors were not different between nonpregnant and pregnant uterine arteries (45), suggests that the pregnancy-associated decrease in cortisol sensitivity is not mediated by changes in glucocorticoid receptor numbers. The question arises as to whether, or to what extent, cortisol regulates NE-mediated contractile mechanisms differentially in the nonpregnant and pregnant uterine arteries.

Despite the striking physiological changes in uterine circulation during pregnancy, and the previous studies showing an important role of uterine endothelial nitric oxide (for review, see Ref. 33), little is known about the adaptation of contractile mechanisms of the uterine artery to pregnancy. It has been demonstrated that NE contracts the uterine artery by acting on ₁-adrenoceptors and increasing inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃], which correlates well with the contractile responses in the uterine artery (20, 49). Release of intracellular Ca²⁺ from the sarcoplasmic reticulum by Ins(1,4,5)P₃ is a major mechanism of pharmacomechanical coupling in smooth muscle (35, 48). There are two major components in receptor-mediated pharmacomechanical coupling: 1) an agonist-induced increase in intracellular free Ca²⁺ concentration ([Ca²⁺]_i), and 2) agonist-mediated Ca²⁺ sensitivity of contractile myofilaments.

In the present study, we sought to examine the regulatory effects of cortisol on ₁-adrenoceptor-mediated pharmacomechanical coupling at multiple steps in the signal transduction pathway in nonpregnant and pregnant ovine uterine arteries. The effect of cortisol on the density of ₁-adrenoceptors was measured by a radioligand binding method. Classic pharmacological approaches were employed to evaluate the relations between receptor occupancy and contractile response and thereby determine the coupling efficiency of ₁-adrenoceptors to contractions. To determine the intrinsic activity of ₁-adrenoceptors in coupling to Ins(1,4,5)P₃ synthesis, we analyzed the relations between ₁-adrenoceptors occupied and Ins(1,4,5)P₃ synthesis. In addition, we determined the effect of cortisol on NE-mediated intracellular Ca²⁺ mobilization and Ca²⁺ sensitivity of contractile myofilaments in the uterine arteries. Our results indicate that cortisol regulates ₁-adrenoceptor-mediated pharmacomechanical coupling differentially in nonpregnant and pregnant uterine arteries.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tissue preparation. Nonpregnant and time-dated pregnant (~140-day gestation) sheep were obtained from Nebeker Ranch (Lancaster, CA). Animals were anesthetized with thiamylal (10 mg/kg) administered via the external left jugular vein, and anesthesia was maintained with 1.5-2.0% halothane in oxygen throughout surgery. An abdominal incision was made to expose the uterus, and the uterine arteries were isolated, removed without being stretched, and placed into a modified Krebs solution (pH 7.4) of the following composition (in mM): 115.21 NaCl, 4.70 KCl, 1.80 CaCl₂, 1.16 MgSO₄, 1.18 KH₂PO₄, 22.14 NaHCO₃, and 7.88 dextrose. EDTA (0.03 mM) was added to suppress the oxidation of amines. The Krebs solution was oxygenated with a 95% O₂-5% CO₂ mixture. After the tissues were removed, the animals were killed with euthanasia solution (T-61; Hoechst-Roussel; Somerville, NJ). All procedures and protocols used in this study were approved by the Animal Research Committee of Loma Linda University and followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Contraction studies. After cortisol pretreatment, arterial contractions were quantified in the continuous presence of cortisol in Krebs solution in tissue baths at 37°C, as described previously (45). Isometric tensions were measured. After 60 min of equilibration in the tissue bath, each ring was stretched to the optimal resting tension as determined by the tension developed in response to KCl added at each stretch level. One ring was used for each determination in each animal, and n represents the number of animals. Concentration-response curves were obtained by cumulative addition of NE in approximate one-half log increments. Prism software (GraphPad; San Diego, CA) was used to fit the curve and determine the apparent affinity (pD₂) values (log EC₅₀) and the maximum response. To determine the relation between receptor occupancy and response, the apparent dissociation constant (K_A) of NE to ₁-adrenoceptors determined previously (45) was used. The fractional receptor occupancy was calculated from the equation [RA]/[R_T] = [A]/([A] + K_A), where [RA] is the concentration of the receptor agonist complex, [R_T] is the total concentration of the receptors, and [A] is the concentration of the agonist (15). To estimate the coupling efficiency of ₁-adrenoceptors to Ins(1,4,5)P₃ synthesis, [RA]/[R_T] was converted to the number of the receptors occupied by 1 or 10 µM NE, using the total receptor density determined by [³H]prazosin. NE-elicited Ins(1,4,5)P₃ productions were then expressed as picomoles of Ins(1,4,5)P₃ per femtomole of ₁-adrenoceptors occupied.

Radioligand binding studies. Saturation binding of [³H]prazosin (DuPont-NEN; Boston, MA), an ₁-adrenoceptor antagonist radioligand, was performed by a rapid filtration method, as described previously (18). Briefly, the vessels were homogenized with a homogenizer (speed setting 5.5 × 15 s; model Polytron PT10/35; Brinkman) in ice-cold 50 mM Tris · HCl (pH 7.4) buffer containing 1 mM EGTA. Nuclei and cell debris were removed by low-speed centrifugation at 1,086 g for 10 min. The supernatant was centrifuged at 50,000 g for 60 min. The microsomal pellet was resuspended in the same Tris buffer to yield ~0.2 mg/ml protein, as determined by the method of Bradford (4). Equilibrium binding was carried out at 30°C for 45 min in a 500-µl volume, consisting of 440 µl of membrane suspension, 50 µl of radioligand, and 10 µl of drug or diluent. The concentrations of [³H]prazosin employed were from 0.002 to 4 nM. Nonspecific binding was determined by the addition of 10 µM phentolamine. All determinations were performed in triplicate. Bound and free radioligand were separated by rapid filtration of the membrane suspension over polyethylenimine (0.5%)-pretreated filters (model GF/C; Whatman) with a Brandel cell harvester. Filters were rinsed with two 5-ml aliquots of the ice-cold Tris buffer and counted for radioactivity at 45% efficiency in liquid scintillation analyzer (model 1900CA Tri-Carb, Packard Instrument; Downers Grove, IL).

Measurement of Ins(1,4,5)P₃. After treatments with or without cortisol, the tissues were equilibrated in Krebs solution at 37°C for 30 min and then stimulated with different concentrations of NE for 30 s at its peak level of Ins(1,4,5)P₃ production, as described previously (19). Ins(1,4,5)P₃ was measured by the competitive ligand binding radioreceptor assay (19). Briefly, the tissue reactions were terminated by flash freezing tissues in liquid N₂. The tissues were then homogenized in ice-cold 16.7% trichloroacetic acid. The homogenate was centrifuged at 1,500 g for 10 min at 4°C. The supernatant was extracted with water-saturated diethyl ether to remove trichloroacetic acid, and the pellet was saved for protein determination by using the method of Bradford (4). Ins(1,4,5)P₃ in the supernatant was determined with the use of a radioreceptor assay kit from DuPont-NEN. Values were expressed as picomoles of Ins(1,4,5)P₃ per milligram of protein.

Simultaneous measurement of [Ca²+]_i and tension. Simultaneous recordings of contractile tension and free [Ca²⁺]_i in the same tissue were conducted as described previously (50). Briefly, the arterial rings were attached to an isometric force transducer in a 5-ml tissue bath mounted on a intracellular Ca²⁺ analyzer (model CAF-110, Jasco; Tokyo, Japan). The tissues were equilibrated in Krebs buffer under a resting tension of 0.5 g for 40 min and loaded under the same tension with 5 µM fura 2-acetoxymethyl ester (Molecular Probes; Eugene, OR) for 4 h in the presence of 0.02% cremophor EL at 25°C. The tissues were then washed with Krebs solution at 37°C for 30 min to allow for hydrolysis of fura 2 ester groups by endogenous esterase. Contractile tension and fura 2 fluorescence were measured simultaneously at 37°C in the same tissue. The tissues were illuminated alternatively (125 Hz) at excitation wavelengths of 340 and 380 nm, respectively, by means of two monochromators in the light path of a 75-W xenon lamp. Fluorescence emission from the tissue was measured at 510 nm with a photomultiplier. The fluorescence intensity at each excitation wavelength (F₃₄₀ and F₃₈₀, respectively) and their ratio (R_f340/f380) were recorded with a time constant of 250 ms and stored with the force signal on a computer.

Data analysis. Saturation binding and concentration response curves were analyzed by computer-assisted nonlinear regression to fit the data and to determine the dissociation constant (K_D), receptor density, and pD₂ with the use of Prism software. Results were expressed as means ± SE, and the differences were evaluated for statistical significance (P < 0.05) by Student's t-test and analysis of variance.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NE-induced contractions. We (45) showed that cortisol (1-30 ng/ml) treatment for 24 h produces a dose-dependent increase in NE-mediated contractions in the uterine arteries. Figure 1 shows that cortisol (10 ng/ml for 24 h) significantly increased NE pD₂ values (5.61 ± 0.02  6.36 ± 0.07, n = 6, P < 0.05) and the maximal response (5.46 ± 0.05 g  7.06 ± 0.18 g, P < 0.05) in nonpregnant uterine arteries (Fig. 1A). In pregnant uterine arteries, cortisol increased NE pD₂ values (6.22 ± 0.11  6.55 ± 0.06, n = 7, P < 0.05) without affecting the maximal response (Fig. 1B). As reported previously, the degree of cortisol-mediated potentiation of NE pD₂ was significantly decreased in pregnant uterine arteries (45).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effect of cortisol on norepinephrine (NE)-induced contraction in the uterine artery. Arterial rings were pretreated in the absence or presence of 10 ng/ml cortisol for 24 h at 37°C and then subjected to the cumulative addition of NE in the tissue bath. Data are the means ± SE of tissues from 5 to 7 animals: pregnant (A) and nonpregnant (B). The apparent affinity (pD2) (log EC50) values were presented in the text.

Radioligand binding studies. The effects of cortisol on the density of ₁-adrenoceptors in the uterine arteries were determined by evaluating the saturation binding of [³H]prazosin, a selective ₁-adrenoceptor antagonist radioligand. As shown in Fig. 2, the binding of [³H]prazosin to ₁-adrenoceptors was specific and saturable and was best described by an interaction of the radioligand with a single class of high-affinity binding sites in both nonpregnant and pregnant uterine arteries. There was no difference in the K_D of [³H]prazosin to ₁-adrenoceptors between the nonpregnant (0.35 ± 0.10 nM, n = 5) and pregnant (0.23 ± 0.06 nM, n = 5) arteries. In contrast, the density of ₁-adrenoceptors was significantly higher in pregnant (75.7 ± 9.6 fmol/mg protein, n = 5) than nonpregnant (30.0 ± 6.6 fmol/mg protein, n = 5) uterine arteries (P < 0.05). Cortisol did not affect the K_D of [³H]prazosin in either nonpregnant (0.35 ± 0.10  0.19 ± 0.05 nM, n = 5, P > 0.05) or pregnant (0.23 ± 0.06  0.54 ± 0.14 nM, n = 5, P > 0.05) uterine arteries, but significantly increased the density of ₁-adrenoceptors in pregnant uterine arteries (75.7 ± 9.6  136.2 ± 17.9 fmol/mg protein, n = 5, P < 0.05). In contrast, cortisol did not change ₁-adrenoceptor density in the nonpregnant arteries (30.0 ± 6.6  31.8 ± 5.8 fmol/mg protein, n = 5, P > 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Saturation binding of [3H]prazosin. Nonpregnant (A) and pregnant (B) uterine arteries were treated in the absence or presence of cortisol (10 ng/ml) for 24 h, and membranes were then prepared. Specific [3H]prazosin binding was defined as the arithmetic difference between total binding and nonspecific binding obtained in presence of 10 µM phentolamine. Analysis of specific binding data by nonlinear computer-based methods (fit to a rectangular hyperbola) confirmed that [3H]prazosin bound to a single class of binding sites in the vessels. Data are the means ± SE of tissues from 5 animals.

Receptor occupancy-contraction relation. We (45) demonstrated in sheep of similar weight and gestational age that cortisol decreased the K_A of NE to ₁-adrenoceptors in nonpregnant (24.6 ± 6.5  6.3 ± 1.4 µM, P < 0.05), but not in pregnant (5.2 ± 2.0  2.2 ± 0.3 µM, P > 0.05), uterine arteries. To examine the effect of cortisol on the postreceptor mechanisms (i.e., beyond the change in receptor numbers), the K_A values determined previously were used to calculate the fraction of ₁-adrenoceptors occupied ([RA]/[R_T]) at each NE concentration used in construction of the respective concentration-contraction curves. The respective occupancy-response relations constructed for NE-mediated contractions are presented in Fig. 3. Cortisol treatment significantly increased the NE-mediated contractions by 25% at the maximal receptor occupancy in nonpregnant uterine arteries and significantly decreased the receptor occupancy required to produce 50% of the maximal response from 0.113 ± 0.007 to 0.076 ± 0.012 (P < 0.05). In contrast, cortisol did not affect the ₁-adrenoceptor occupancy-contraction relation in pregnant uterine arteries (Fig. 3).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   NE receptor occupancy-response relation. Nonpregnant (A) and pregnant (B) uterine arteries were treated in the absence or presence of cortisol (10 ng/ml) for 24 h, and contractions were then induced by NE. Fraction of 1-adrenoceptors occupied at each NE concentration was calculated as described in METHODS. Data are the means ± SE of tissues from 5 to 7 animals.

Ins(1,4,5)P₃ synthesis. NE produced a concentration-dependent increase of Ins(1,4,5)P₃ in both nonpregnant and pregnant uterine arteries with pD₂ values of 6.49 ± 0.17 and 6.83 ± 0.10 (n = 5, P > 0.05), respectively (Fig. 4). To examine the effect of cortisol on the NE-mediated Ins(1,4,5)P₃ synthesis in the uterine artery, we quantified Ins(1,4,5)P₃ production induced by 0.1 µM NE in the tissues pretreated with different concentrations of cortisol (0-30 ng/ml for 24 h). Figure 5 shows that cortisol produces a dose-dependent potentiation of NE-induced Ins(1,4,5)P₃ synthesis in both nonpregnant and pregnant uterine arteries. Given that the effect of cortisol is regulated by 11-hydroxysteroid dehydrogenase (11-HSD), and that our previous findings suggested an increase in type-2 11-HSD activity in the pregnant arteries (45), we examined the effect of 11-HSD on the cortisol-potentiated Ins(1,4,5)P₃ synthesis in the uterine artery. Tissues were pretreated with 10 ng/ml cortisol in the absence or presence of the 11-HSD inhibitor carbenoxolone (3 µM) for 24 h, and NE (0.1 µM)-stimulated Ins(1,4,5)P₃ production was then measured. As shown in Fig. 6, cortisol significantly potentiated NE-induced Ins(1,4,5)P₃ synthesis in both nonpregnant and pregnant uterine arteries. Carbenoxolone alone did not affect NE-mediated Ins(1,4,5)P₃ synthesis, but significantly enhanced cortisol-mediated potentiation of NE-stimulated Ins(1,4,5)P₃ synthesis in the pregnant arteries. In contrast, cortisol-mediated potentiation of NE-induced Ins(1,4,5)P₃ synthesis in the nonpregnant arteries was not affected by carbenoxolone (Fig. 6). To estimate the coupling efficiency of ₁-adrenoceptors to Ins(1,4,5)P₃ synthesis, NE-elicited Ins(1,4,5)P₃ productions were expressed as picomoles of Ins(1,4,5)P₃ per femtomole of ₁-adrenoceptors occupied, as described in METHODS. As shown in Fig. 7, cortisol significantly increased the coupling [pmol Ins(1,4,5)P₃/fmol receptor] in nonpregnant (P < 0.05), but not pregnant, arteries.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   NE induced inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] production. Nonpregnant and pregnant uterine arteries were incubated with different concentrations of NE for 30 s. Data are the means ± SE of tissues from 5 animals.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Effect of cortisol on NE-induced Ins(1,4,5)P3 production. Nonpregnant and pregnant uterine arteries were treated in the absence or presence of cortisol (0.3 to 30 ng/ml) for 24 h, and Ins(1,4,5)P3 production was then induced by NE (0.1 µM for 30 s). Nonpregnant and pregnant data overlap at the point of 0 ng/ml cortisol. Data are the means ± SE of tissues from 5 animals.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Effect of carbenoxolone on NE-induced Ins(1,4,5)P3 production. Nonpregnant and pregnant uterine arteries were treated in the absence or presence of cortisol (10 ng/ml), carbenoxolone (Carb; 3 µM), or cortisol + Carb for 24 h, and Ins(1,4,5)P3 production was induced by NE (0.1 µM for 30 s). Data are means ± SE of tissues from 5 to 10 animals. aP < 0.05 vs. nonpregnant; bP < 0.05 vs. control; cP < 0.05 vs. cortisol alone.

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Relation of 1-adrenoceptor (AR) occupancy and Ins(1,4,5)P3 production. Nonpregnant and pregnant uterine arteries were treated in the absence or presence of cortisol (10 ng/ml) for 24 h. Production of Ins(1,4,5)P3 stimulated by 1 or 10 µM NE for 30 s was plotted to show picomoles of Ins(1,4,5)P3 per femtomole of 1-adrenoceptors occupied by NE, calculated as described in METHODS. Data are the means ± SE of tissues from 5 animals.

[Ca²+]_i and [Ca²+]_i-tension relation. NE produced a dose-dependent increase of free [Ca²⁺]_i in both nonpregnant and pregnant uterine arteries with pD₂ values of 5.78 ± 0.11 and 5.24 ± 0.06 (n = 4, P < 0.05), respectively (Fig. 8). Cortisol treatment did not significantly affect NE-induced [Ca²⁺]_i in nonpregnant arteries (pD₂: 5.78 ± 0.11  6.05 ± 0.10, n = 4, P > 0.05) but significantly increased pD₂ of NE-stimulated [Ca²⁺]_i in pregnant (5.24 ± 0.06  5.96 ± 0.16, n = 4, P < 0.05) uterine arteries. To examine whether cortisol-mediated potentiation of [Ca²⁺]_i responses in pregnant uterine arteries was due to increased coupling efficiency of Ins(1,4,5)P₃ and Ca²⁺ release, we evaluated the relation between Ins(1,4,5)P₃ production and [Ca²⁺]_i responses. As shown in Fig. 9, NE-evoked Ins(1,4,5)P₃ production correlated significantly with increased [Ca²⁺]_i in the uterine arteries, implicating a key role of Ins(1,4,5)P₃ in Ca²⁺ mobilization. There was no significant difference between the slopes of the Ins(1,4,5)P₃-[Ca²⁺]_i relation {R_f340/f380 [Ca²⁺]_i/Ins(1,4,5)P₃} determined in control and cortisol-treated pregnant arteries (0.00098 ± 0.00017  0.00062 ± 0.00013, P > 0.05; Fig. 9), suggesting that cortisol did not affect the apparent coupling efficiency of Ins(1,4,5)P₃ to Ca²⁺ mobilization in the pregnant arteries. In contrast, cortisol caused a significant decrease in the slope in the nonpregnant arteries (0.00091 ± 0.00018  0.00017 ± 0.00002, P < 0.05; Fig. 9).

View larger version (16K): [in this window] [in a new window]   Fig. 8.   NE-induced intracellular Ca2+ mobilization. Nonpregnant (A) and pregnant (B) uterine arteries were treated in the absence or presence of cortisol (10 ng/ml) for 24 h, and intracellular Ca2+ concentration ([Ca2+]i; fura 2 signal, Rf340/f380) was then stimulated by NE. Data are the means ± SE of tissues from 4 animals.

View larger version (16K): [in this window] [in a new window]   Fig. 9.   Relation of Ins(1,4,5)P3 production and [Ca2+]i. Nonpregnant (A) and pregnant (B) uterine arteries were treated in the absence or presence of cortisol (10 ng/ml) for 24 h. Increases of [Ca2+]i (fura 2 signal, Rf340/f380) stimulated by different concentrations of NE were plotted to show responses as a function of Ins(1,4,5)P3 production at each corresponding concentration of NE. Data are the means ± SE of tissues from 4 animals.

View larger version (16K): [in this window] [in a new window]   Fig. 10.   Relation of [Ca2+]i and tension. Nonpregnant (A) and pregnant (B) uterine arteries were treated in the absence or presence of cortisol (10 ng/ml) for 24 h. The [Ca2+]i-tension relationship was obtained by the simultaneous measurement of [Ca2+]i (fura 2 signal, Rf340/f380) and tension induced by different concentrations of NE, as described in METHODS. Data are the means ± SE of tissues from 4 animals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study demonstrated clearly that cortisol regulated ₁-adrenoceptor-mediated pharmacomechanical coupling differentially in nonpregnant and pregnant uterine arteries. In the absence of cortisol, the basal levels of Ins(1,4,5)P₃ and NE-induced contractions were increased in pregnant uterine arteries. Although the attenuation of cortisol-potentiated contractions of pregnant uterine arteries may be due in part to elevated endogenous cortisol in pregnant animals, there are striking differences in signaling pathways between nonpregnant and pregnant uterine arteries in response to cortisol. In nonpregnant arteries, cortisol did not affect ₁-adrenoceptor numbers, but significantly enhanced its coupling efficiency by increasing both Ins(1,4,5)P₃ and agonist-mediated Ca²⁺ sensitivity of contractile myofilaments. Pregnancy abolished the effects of cortisol on ₁-adrenoceptor coupling efficiency, and, instead, cortisol upregulated ₁-adrenoceptor numbers, leading to increased Ins(1,4,5)P₃ and Ca²⁺ mobilization.

Coupling of ₁-adrenoceptors to contractile responses can be modulated at several steps in the signal transduction pathway, including receptor density, agonist affinity, and coupling efficiency of the receptor. The finding that cortisol upregulated ₁-adrenoceptor density in pregnant arteries is consistent with previous studies (16) showing that adrenalectomy caused a 40% decrease in ₁-adrenoceptor density in the rat aorta, which was restored by dexamethasone replacement. Whereas the present study did not examine ₁-adrenoceptor subtypes, previous studies (29) showed that glucocorticoids upregulated expression of _1B-adrenoceptors in vascular smooth muscle cells by increasing the rate of gene transcription. These studies suggest that glucocorticoids play a key role in the regulation of ₁-adrenoceptor density in vascular smooth muscle. In contrast to the pregnant uterine artery, cortisol did not affect ₁-adrenoceptor density, but instead increased NE efficiency in contracting nonpregnant uterine arteries, indicating that mechanisms beyond the agonist-receptor interaction are also regulated by cortisol. This increased coupling efficiency may be mediated by multiple mechanisms, including receptor coupling to Ins(1,4,5)P₃ synthesis, Ins(1,4,5)P₃ efficiency in Ca²⁺ mobilization, and Ca²⁺ sensitivity of contractile myofilaments.

In the present study, basal Ins(1,4,5)P₃ levels were elevated in pregnant uterine arteries. Although it is possible that elevated cortisol during pregnancy may play a role, we found that NE sensitivity (pD₂ values) in stimulating Ins(1,4,5)P₃ production was not significantly different between nonpregnant and pregnant uterine arteries. Cortisol potentiated NE-induced Ins(1,4,5)P₃ synthesis in both nonpregnant and pregnant uterine arteries. Glucocorticoid-mediated potentiation of Ins(1,4,5)P₃ production has been reported in vascular smooth muscle for angiotensin II, arginine vasopressin, endothelin-1, and catecholamines (24, 30, 43). The role of Ins(1,4,5)P₃ as the messenger of pharmacomechanical Ca²⁺ mobilization in smooth muscle has been firmly established (35). We have demonstrated that Ins(1,4,5)P₃ is the messenger of pharmacomechanical coupling for ₁-adrenoceptor-mediated contractions in the uterine artery (49). Although ₁-adrenoceptor-mediated increase in intracellular Ca²⁺ may result from both Ca²⁺ release from intracellular stores and Ca²⁺ influx through receptor- and voltage-operated Ca²⁺ channels, the initial signal is dependent on Ins(1,4,5)P₃-mediated Ca²⁺ release from intracellular stores. In the present study, we measured Ca²⁺ by its peak, rather than the area under the curve, which provided a reasonable estimation of Ca²⁺ release from intracellular stores. It has been suggested that both the mobilization of Ca²⁺ from internal stores, as well as the entry of external Ca²⁺, are critically dependent on the formation of Ins(1,4,5)P₃ (3, 48). By using permeabilized vascular smooth muscle preparations, Somlyo et al. (34) demonstrated that photolytically released Ins(1,4,5)P₃ from the caged Ins(1,4,5)P₃ stimulated a rise in free [Ca²⁺]_i that correlated closely with the force development.

The finding that the 11-HSD inhibitor carbenoxolone selectively enhanced cortisol-mediated potentiation of Ins(1,4,5)P₃ synthesis in the pregnant uterine artery is intriguing and suggests an increase in type 2 11-HSD activity in this vessel. This is in agreement with our previous finding that carbenoxolone selectively potentiated NE-induced contraction in the pregnant, but not in nonpregnant, uterine arteries (45). The effect of glucocorticoids on vascular reactivity is regulated with the use of 11-HSD (39). Two 11-HSD isozymes catalyze the interconversion of cortisol and cortisone. The type 1 11-HSD has bidirectional activity, whereas the type 2 enzyme mainly converts cortisol into cortisone, the biologically inactive form. Both type 1 and 2 11-HSD have been found in vascular smooth muscle (6, 42). Several studies (5, 22, 39, 42) have demonstrated that inhibition of 11-HSD with inhibitors such as carbenoxolone increases cortisol-mediated potentiation of vascular response to NE. Although under normal conditions, the type 1 isoform dominates functioning in the oxo-reductase mode that converts cortisone to cortisol in vascular smooth muscle, the two major isoforms are compartmentalized discretely and regulated differentially by steroids such as estrogen and progesterone (36). In human pregnancy, placental type 2 11-HSD activity increases markedly in the third trimester of pregnancy, at a time when maternal circulating levels of glucocorticoid are rising, which serves as a protective mechanism for the fetus (32). Our results suggest an increase in the activity of type 2 11-HSD in pregnant uterine arteries, which is likely to play an important role in the local regulation of cortisol concentrations by limiting cortisol effects on the uterine artery, and protecting it from elevated cortisol levels during pregnancy.

The present study demonstrated that for a given number of ₁-adrenoceptors occupied, cortisol increased Ins(1,4,5)P₃ production in nonpregnant uterine arteries, suggesting that the intrinsic activity of the receptor was enhanced. The mechanisms underlying this enhanced coupling efficiency of ₁-adrenoceptors to Ins(1,4,5)P₃ synthesis are not clear at present, but can occur at multiple levels. For example, heterotrimeric guanine nucleotide-binding proteins (G proteins) are physiological targets of glucocorticoids in vivo (31). It has been shown that glucocorticoids increase G_q/11-protein expression and phospholipase C activity in rat osteoblastic cells (10). In addition, glucocorticoids have been shown to play a crucial role in maintaining coupling of ₁-adrenoceptors to G proteins, by regulating the amounts of G proteins in the rat aorta (16, 17). Given the finding that pregnancy increased inibitory G protein activation/coupling in uterine arteries to certain agonists (37), it is speculated that the increased adrenoceptor binding induced by cortisol treatment in this study may be due to increased G protein/receptor coupling, which augments ligand/receptor binding. Taken together, these studies suggest an important mechanism by which glucocorticoids regulate receptor-G protein coupling, and hence transmembrane signaling pathways, in vascular smooth muscle. Future studies are needed to determine whether cortisol treatment for 24 h increases G proteins expression and activity in the uterine artery. Alternatively, cortisol may enhance the coupling of ₁-adrenoceptors to Ins(1,4,5)P₃ synthesis by increasing phosphoinositide-specific phospholipase C activity. It has been demonstrated that dexamethasone increases phospholipase C activity and mRNA/protein expression of the phospholipase C-₁ isozyme in the rat brain (12).

The finding that cortisol did not affect the coupling efficiency of ₁-adrenoceptors to Ins(1,4,5)P₃ synthesis in the pregnant arteries is consistent with the results that cortisol did not change the intrinsic efficiency of NE in contracting these vessels. In addition, cortisol did not affect Ins(1,4,5)P₃ efficiency in Ca²⁺ mobilization in the pregnant artery. These results suggest that cortisol-induced potentiation of NE-stimulated Ca²⁺ mobilization is mediated predominantly by the upregulation of ₁-adrenoceptor numbers in the pregnant artery. The apparent loss of the ability of cortisol in regulating ₁-adrenoceptor coupling efficiency in the pregnant uterine artery may be due in part to pregnancy-mediated alterations in G protein levels and GTPase activity (7-9, 11, 37). In uterine arteries, pregnancy inhibited stimulatory G_s GTPase activity and the decreased G_s cycling rate, but increased inhibitory G protein activation/coupling (7, 37).

The finding that cortisol did not increase NE-induced Ca²⁺ mobilization in nonpregnant uterine arteries is somewhat surprising, given that cortisol potentiated ₁-adrenoceptor-mediated Ins(1,4,5)P₃ synthesis in the nonpregnant arteries. Nevertheless, cortisol significantly decreased the coupling efficiency of Ins(1,4,5)P₃ to Ca²⁺ mobilization in the nonpregnant arteries, which may counteract the effect of increased Ins(1,4,5)P₃. The coupling of Ins(1,4,5)P₃ to Ca²⁺ mobilization involves the binding of Ins(1,4,5)P₃ to Ins(1,4,5)P₃ receptors. Our previous studies (19, 51) demonstrated that hypoxic stress altered Ins(1,4,5)P₃ binding affinity and Ins(1,4,5)P₃ receptor density in the uterine and cerebral arteries, respectively. Other studies (26) suggested that dexamethasone caused a decrease in Ins(1,4,5)P₃ affinity to Ins(1,4,5)P₃ receptors in NIH3T3 cells. Given the previous finding that glucocorticoids did not alter any of three isoforms of Ins(1,4,5)P₃ receptors in the rat brain (13), it is speculated that cortisol-mediated decrease in the coupling of Ins(1,4,5)P₃ to Ca²⁺ mobilization in the uterine artery is due to decreased Ins(1,4,5)P₃ binding affinity.

Not only does [Ca²⁺] play an important role in the regulation of smooth muscle contraction, Ca²⁺ sensitivity also provides a key determinant of smooth muscle contraction, which is modulated physiologically and pathophysiologically in the uterine arteries (46, 50). In the present study, we have shown that NE-induced Ca²⁺ mobilization is decreased by pregnancy. In contrast, NE-mediated Ca²⁺ sensitivity was increased. To our knowledge, this is the first study to demonstrate the differential adaptation of Ca²⁺ homeostasis in the uterine artery to pregnancy. Although few studies examined the effects of pregnancy and/or steroid hormones on contractile mechanisms in the uterine artery, studies in human myometrium demonstrated that adaptation to pregnancy included 1) cellular mechanisms that preclude the development of high levels of myosin light chain phosphorylation during contraction; and 2) an increase in the stress-generating capacity for any given level of myosin light chain phosphorylation, suggesting a decrease in Ca²⁺ mobilization and an increase in Ca²⁺ sensitivity (44). Although the mechanisms for this differential adaptation of Ca²⁺ mobilization and Ca²⁺ sensitivity to pregnancy are not entirely clear at present, it has been shown that progesterone decreases Ca²⁺ mobilization in myometrial smooth muscle cells (14). On the other hand, an increase in RhoA/Rho kinase (27) and a decrease in myosin light chain phosphatase (38) may play an important role in pregnancy-mediated increase in Ca²⁺ sensitivity. In addition, an increase in contractile proteins of actin and myosin in the pregnant uterine artery (2) may also contribute to the increased Ca²⁺ sensitivity.

In the present study, despite a decrease in NE-induced Ca²⁺ mobilization in pregnant uterine arteries, NE-mediated contractions were increased in pregnant compared with nonpregnant uterine arteries (1, 46). In pregnant uterine arteries, NE had ~10 times lower sensitivity in Ca²⁺ mobilization (pD₂: 5.24) than tension generation (pD₂: 6.22). This suggests that changes in Ca²⁺ sensitivity play a predominant role in the regulation of uterine artery contractility during pregnancy. The finding that cortisol increased Ca²⁺ sensitivity in the nonpregnant uterine artery is intriguing and suggests that cortisol may play an important role in the pregnancy-induced increase in Ca²⁺ sensitivity in the uterine artery, given that maternal plasma cortisol concentrations significantly increase during pregnancy (21, 28). Because progesterone and/or estrogen treatment inhibits agonist- and GTPS-induced Ca²⁺ sensitization of smooth muscle by increasing Rnd1 expression, which inhibits the RhoA-dependent pathways (25), and progesterone has antiglucocorticoid effects and binds to glucocorticoid receptors at a physiological concentration, we propose that cortisol counteracts with progesterone and/or estrogen in regulating Ca²⁺ sensitivity of the uterine artery during pregnancy.

In summary, the present results indicate that cortisol enhances ₁-adrenoceptor coupling efficiency and agonist-mediated myofilament Ca²⁺ sensitivity in the nonpregnant uterine artery, whereas it increases ₁-adrenoceptor density, leading to an increase in Ca²⁺ mobilization in the pregnant artery. To our knowledge, this is the first study of the effect of cortisol on agonist-mediated pharmacomechanical coupling in vascular smooth muscle in general and on the regulation of Ca²⁺ homeostasis in the uterine artery in particular. Although the mechanisms underlying the differential regulatory effects of cortisol on Ca²⁺ mobilization and Ca²⁺ sensitivity in pregnant and nonpregnant uterine arteries remain to be elucidated, the present study suggests an important role of cortisol in the regulation of Ca²⁺ homeostasis in the uterine artery during pregnancy. Our recent study (46) demonstrated that pregnancy altered the ERK/protein kinase C pathway in Ca²⁺ handling in the uterine artery. The potential interaction of glucocorticoids with the ERK/protein kinase C pathway in the regulation of uterine artery contractility presents an intriguing avenue for future investigation.