Regulation of the cGMP-cPKG pathway and large-conductance Ca²⁺-activated K⁺ channels in uterine arteries during the ovine ovarian cycle

Abstract

The follicular phase of the ovine ovarian cycle demonstrates parallel increases in ovarian estrogens and uterine blood flow (UBF). Although estrogen and nitric oxide contribute to the rise in UBF, the signaling pathway remains unclear. We examined the relationship between the rise in UBF during the ovarian cycle of nonpregnant sheep and changes in the uterine vascular cGMP-dependent pathway and large-conductance Ca²⁺-activated K⁺ channels (BK_Ca). Nonpregnant ewes (n = 19) were synchronized to either follicular or luteal phase using a vaginal progesterone-releasing device (CIDR), followed by intramuscular PGF_2α, CIDR removal, and treatment with pregnant mare serum gonadotropin. UBF was measured with flow probes before tissue collection, and second-generation uterine artery segments were collected from nine follicular and seven luteal phase ewes. The pore-forming α- and regulatory β-subunits that constitute the BK_Ca, soluble guanylyl cyclase (sGC), and cGMP-dependent protein kinase G (cPKG) isoforms (cPKG_1α and cPKG_1β) were measured by Western analysis and cGMP levels by RIA. BK_Ca subunits were localized by immunohistochemistry. UBF rose >3-fold (P < 0.04) in follicular phase ewes, paralleling a 2.3-fold rise in smooth muscle cGMP and 32% increase in cPKG_1α (P < 0.05). sGC, cPKG_1β, and the BK_Ca α-subunit were unchanged. Notably, expression of β₁- and β₂-regulatory subunits rose 51 and 79% (P ≤ 0.05), respectively. Increases in endogenous ovarian estrogens in follicular-phase ewes result in increases in UBF associated with upregulation of the cGMP- and cPKG-dependent pathway and increased vascular BK_Ca β/α-subunit stoichiometry, suggesting enhanced BK_Ca activation contributes to the follicular phase rise in UBF.

uterine blood flow (UBF) increases during the follicular phase of the ovarian cycle and throughout pregnancy (7, 9, 11, 28, 35). Although the mechanisms responsible for these increases in UBF are incompletely understood, both are associated with increases in circulating endogenous estrogens of ovarian and placental origin, respectively, and the estrogen-to-progesterone ratio (E/P), suggesting similar mechanisms may be involved (1, 9, 18, 20, 35, 43). Acute and chronic exogenous estrogen increase UBF in nonpregnant and pregnant sheep and other species (28). Markee (22) demonstrated that the uterine hyperemia in the follicular phase of the ovarian cycle was reproducible after administration of ovarian estrogens. Subsequently, estradiol-17β (E₂β) was shown to be a potent uterine vasodilator when infused either systemically or directly in the uterine circulation, increasing UBF >10-fold within 90–120 min (15, 19, 28, 32, 34). This was reproducible every 24 h and characterized by a 30-min delay followed by a peak and plateau at 90–120 min and a gradual decline over 8–12 h (15). The delay in the acute rise in UBF after E₂β has been shown to reflect upregulation and activation of uterine artery nitric oxide synthase (NOS) followed by enhanced nitric oxide (NO) synthesis and its downstream effects on cGMP (31, 39, 48, 50). Of note, ovarian estrogens and local NO synthesis also contribute to the rise in UBF in the follicular phase of the ovine ovarian cycle (9, 18, 21, 43); however, the steps in this cascade are unclear.

The large-conductance, voltage- and Ca²⁺-activated K⁺ channel (BK_Ca) is a heterotetramer of its pore-forming α-subunit encoded by slo1 gene and one or more isoforms of tissue specific regulatory β-subunits (8). It is ubiquitously expressed among excitable and nonexcitable mammalian tissues, including neurons and myocytes (3). Its open state probability is increased by membrane depolarization or increases in cytosolic Ca²⁺ (14). Depending on the cell type, BK_Ca activity may differ in Ca²⁺ sensitivity and macroscopic kinetics. Enhanced BK_Ca activity in smooth muscle cells (SMC) produces membrane hyperpolarization, inhibits Ca²⁺ entry through L-type Ca²⁺ channels, decreases cytosolic Ca²⁺, and consequently produces SMC relaxation (17). BK_Ca has substantial functional diversity stemming from alternative splicing of the α-subunit, interaction with an array of regulatory β-subunits, and metabolic regulation, including phosphorylation by protein kinases (3, 17, 24, 40). Several variants of the β-subunit exist, but β₁- and β₂-subunits are specific to SMC (24, 40). Altered β/α stoichiometry through increases in the β₁-subunit enhances Ca²⁺ and voltage sensitivity (24). In nonpregnant and pregnant ewes, BK_Ca contribute to acute E₂β-mediated increases in UBF (30, 37), as well as the effects of prolonged E₂β exposure on baseline UBF in ovariectomized nonpregnant sheep and baseline uteroplacental blood flow in ovine pregnancy (23, 36). Moreover, prolonged E₂β exposure upregulates β₁-subunit expression, altering channel stoichiometry and sensitivity (23, 39). The role of BK_Ca in the rise in UBF during the ovarian cycle and increases in endogenous ovarian estrogens has not been studied.

In the present report, we studied uterine arteries obtained from nonpregnant ewes synchronized into the follicular or luteal phases of the ovarian cycle (9). In this model, there is evidence of endogenous estrogen-mediated increases in uterine artery NO, resulting in the rise in UBF during the follicular phase (9, 18, 21, 43). Therefore, we sought to determine if the rise in UBF was related to alterations in the NO-cGMP-protein kinase G pathway as well as BK_Ca expression and stoichiometry.

METHODS

Animal preparation and experimental protocol.

Tissues obtained from 19 nonpregnant, intact multiparous ewes (50–65 kg) of mixed Western breed were used in the present studies. Animals were studied between September and May; those studied between September and March had evidence of estrous cycles, whereas those studied April to May previously exhibited cycles. Animals were randomly assigned to either follicular (n = 10) or luteal (n = 9) phases of the ovarian cycle; all were implanted with a vaginal progesterone (P₄)-controlled internal drug-releasing device (CIDR, 0.3 g; Latinagro de Mexico, Monterrey, Mexico) to clamp systemic P₄ levels at luteal phase values (9, 43). After at least 7 days of P₄ treatment, ewes received two 7.5-mg im injections of PGF_2α (Dinoprost Tromethamine-Lutalyse; Upjohn, Kalamazoo, MI) 4 h apart to lyse existing corpora lutea. On the 10th day following CIDR placement, the CIDR was removed, and each animal received one dose of 500 IU pregnant mare serum gonadotropin intramuscularly. This regimen causes a rapid fall in P₄; a rise in plasma estrogens and the estrogen-to-P₄ ratio (E/P), with levels peaking at 36–48 h; a surge in luteinizing hormone; and ovulation within 48–56 h (9, 43). The above procedures were performed at the breeding farm under natural lighting conditions, and animals were transported to the laboratory before termination where similar lighting conditions were maintained. Follicular phase ewes were studied ∼48 h (±30 min) after CIDR removal, designated as day 0 of the ovarian cycle. Luteal phase ewes were studied 12–13 days after CIDR removal, designated as day 10–11 of the cycle. Ovaries were collected from all animals at the time of termination and mapped to confirm the status within the cycle (Table 1). Ovaries from follicular phase ewes had more total follicles and large follicles 4–6 mm than luteal phase ovaries as well as avascular regressing corpora luteum (CL) vs. large vascular CL in luteal phase ewes. Measurements of P₄ in jugular venous blood were available in four follicular (0.5 ± 0.2 ng/ml) and two luteal (5.6 and 6.8 ng/ml) phase animals, demonstrating an ∼10-fold rise in circulating P₄ in the latter (P < 0.025). More detailed values for E₂β, P₄ and, the E/P have been reported (9, 43).

Table 1.

Ovarian structures in nonpregnant ewes synchronized into either the follicular or luteal phases of the ovarian cycle as described in methods

Diameter, mm	Follicular Phase (n = 9)	Luteal Phase (n = 7)
Corpora lutea	6.9±0.6(avascular)*	10.1±0.3(vascular)
Medium follicles (3–5 mm)	3.8±0.2(28)	4.0±0.3(8)
Large follicles (≥6 mm)	8.2±0.4(14)*	6.9±0.6(4)

Values are means ± SE for the no. of specific ovarian structures for all ovaries from each ewe; n, no. of ewes in each group. The total no. of structures is in parentheses.

^* P < 0.05 by nonpaired t-test.

On the day of study, a percutaneous jugular venous catheter was inserted for intravenous administration of ketamine (10 mg/ml) in 0.9% saline and 5% dextrose with supplemental pentobarbital sodium (Nembutal; 50 mg/ml) as needed for surgical anesthesia as previously described (9, 43). The abdomen was shaved and aseptically scrubbed, following which, the uterus was exposed through a midventral laparotomy, and Transonic flow probes (3–4 mm ID; Transonic Systems, Ithaca, NY) were placed around the middle uterine artery of each uterine horn. UBF in each uterine horn was recorded for 4–6 min after stabilization. Animals were then killed with intravenous pentobarbital sodium (75 mg/kg), and tissues were collected rapidly. Animal handling and protocols for these experiments were approved by the University of Wisconsin-Madison Research and Animal Care and Use Committees of both the Medical School and the College of Agriculture and Life Sciences.

Tissue preparation.

The intact uterus was removed in block, and second-generation uterine arteries were dissected from both uterine horns and placed in sterile chilled physiological saline solution; the adventitia and intraluminal blood were removed. Samples were frozen in liquid nitrogen and stored at −80°C until the time of assay (21, 29, 39). Additional uterine artery samples were prepared for immunohistochemistry as described below. Tissues from both groups of animals were used in other studies; therefore, samples were not available from each animal for each assay, resulting in a difference in the n values noted in the results.

Western immunoblots.

Samples of second-generation uterine arteries (30 mg) were weighed and homogenized in 40× volumes of SDS buffer as previously described (23, 29, 33, 38). Homogenates were centrifuged at 10,000 g for 2 min, the supernatant was removed, and an aliquot from each animal was used to measure cellular or soluble protein by bicinchoninic acid reagent (Pierce, Rockford, IL) as previously reported (23, 33, 38). Bromphenol blue and 2-mercaptoethanol were added to the aliquots representing each animal studied, and equal amounts of soluble protein were loaded (20 μg) on a single 7.5–10% polyacrylamide minigel and subjected to SDS-PAGE followed by transfer to nitrocellulose paper at 100 volts for 1 h (23, 29, 33, 38). Running all samples on a single gel takes into account differences in protein transfer; because similar amounts of soluble protein are loaded, comparisons can be made between samples. The immunoblots were incubated overnight with antisera to BK_Ca α-subunit (1:300; AB5228–200UL; Chemicon International, Temecula, CA), β₁-subunit (1:400; ab3587, Abcam, Cambridge, MA), β₂-subunit (1:200; APC-034; Alomane Labs, Jerusalem, Israel), soluble guanylyl cyclase (sGC; 1:1,000; G 4405; Sigma, St. Louis, MO), cGMP-dependent protein kinase G_1α (cPKG_1α; 1:1,000; PK10; Calbiochem, San Diego, CA), or cPKG_1β (1:750; a gift from HK Surks, Tufts-New England Medical Center). To determine what protein could be used as a loading control, we performed preliminary studies in which we measured uterine artery smooth muscle actin contents. These studies demonstrated that actin was unaffected (P > 0.1; data not shown) after 7 days of daily E₂β exposure (1 μg/kg iv) in nonpregnant castrated ewes; thus, we chose to use α-actin as a loading control (1:8,000; monoclonal anti-α actin; Sigma) on immunoblots performed with cPKG_1α and the BK_Ca β₁-subunit in luteal (n = 4) and follicular (n = 4) phase arteries and reprobed for α-actin at 42 kDa. There was no difference in α-actin on either immunoblot (P > 0.8; Fig. 1). The repeat immunoblots performed to demonstrate equal loading plus the two proteins of interest are available as Supplemental Fig. 1 (Supplemental data for this article can be found on the American Journal of Physiology: Endocrinology and Metabolism website.). After 1 h of incubation with anti-rabbit immunoglobulin G conjugated with horseradish peroxidase (1:2,000), immunoreactive protein was visualized by enhanced chemiluminescence. Samples were compared by scanning densitometry in arbitrary units (TotalLab Software Package; Biosystematica, Wales, UK) as reported (23, 29, 33, 38). Ovine cerebellum served as a positive internal control for BK_Ca α-subunit.

Fig. 1.

Representative immunoblot for the 42-kDa protein α-actin that served as a loading control for studies of cGMP-dependent protein kinase G (cPKG_1α) and the large-conductance Ca²⁺-activated K⁺ channel (BK_Ca) β₁-subunit. There were no differences between luteal (n = 4) and follicular (n = 4) phase uterine arteries, P > 0.8 by nonpaired t-test on each immunoblot. Supplemental Fig. 1 shows the relationship with the proteins of interest.

Measurement of cGMP.

The cGMP content of uterine arteries harvested from follicular and luteal phase animals was measured by radioimmunoassay (cGMP [¹²⁵I] RIA KIT; NEX-133; Perkin-Elmer Life Sciences, Boston, MA) as previously reported (33, 39). Briefly, 50 mg of frozen uterine artery were homogenized in 1 ml of 6% TCA and centrifuged at 3,000 g to remove bulk proteins. TCA was added to the homogenate and extracted from the supernatant with water-saturated dimethyl-ether three times. The supernatant was precipitated by lyophilization, dissolved in 0.05 M sodium acetate, and used in the assay. All samples were measured in a single assay.

Immunohistochemistry.

At the time of tissue collection, intact segments of second-generation uterine arteries were washed in PBS, fixed in 4% paraformaldehyde for 6 h at room temperature, and subsequently embedded in paraffin as previously described (29, 39). Sections were mounted on slides, deparaffinized, hydrated, incubated with avidin-biotin blocking agent for 30 min, and incubated overnight at room temperature with polyclonal antibodies to the BK_Ca α-, β₁-, or β₂-subunits (1:30). After endogenous peroxidases were quenched with 3% H₂O₂ in H₂O for 30 min, immunostaining was detected with standard strepavidin-biotin-horseradish peroxidase and hematoxylin counterstaining. Samples were randomly selected from three follicular and three luteal phase animals and embedded in a single block. Nonimmune rabbit serum was used to show antibody specificity.

Statistical analysis.

Differences between follicular and luteal phase measurements were determined with Student's nonpaired t-test. Data are presented as means ± SE.

RESULTS

UBF.

UBF measurements did not differ in the two uterine horns (P > 0.1); therefore, we used the average of the two values for each ewe. UBF was 3.2-fold higher in the follicular phase animals (n = 5) at the time of tissue collection compared with those allowed to progress to the luteal phase (n = 3) (Fig. 2; P < 0.04).

Fig. 2.

Change in basal uterine blood flow during the luteal and follicular phases of the ovine ovarian cycle. Data are means ± SE; *P < 0.04, nonpaired t-test.

Signaling pathway.

sGC was detected in all samples of uterine artery smooth muscle examined, but there were no differences in protein expression in the follicular and luteal phases of the ovarian cycle (P = 0.9, data not shown). In contrast, uterine arteries obtained from follicular phase ewes had cGMP contents 2.3-fold greater than vessels collected from luteal phase ewes (P < 0.02; Fig. 3). To further define the NO-cGMP-cPKG pathway, we examined cPKG₁ isoform expression using immunoblot analysis. cPKG_1β expression did not differ in follicular and luteal phase uterine arteries (P = 0.22; Fig. 4B2); however, cPKG_1α protein was 32% higher in follicular vs. luteal phase tissues (P = 0.047; Fig. 4B1). In the secondary analysis to assess protein loading, there was no difference in α-actin (P = 0.9), but cPKG_1α increased 21% (P = 0.008; see Supplemental Fig. 1).

Fig. 3.

Comparison of cGMP contents in luteal (n = 3) and follicular (n = 3) phase uterine arteries during the ovine ovarian cycle. Data are means ± SE; *P < 0.02, nonpaired t-test.

Fig. 4.

Effect of the ovarian cycle on cPKG_1α and cPKG_1β expression in uterine arteries from luteal (n = 4) and follicular (n = 6) phase sheep. A: immunoblot analyses. B: densitometry in arbitrary units from the immunoblots for cPKG_1α (B1) and cPKG_1β (B2). Data are means ± SE; *P = 0.047, nonpaired t-test.

BK_Ca expression.

The density of BK_Ca channels is determined by measuring the expression of the pore-forming α-subunit (17, 23, 33, 38). There was no difference in α-subunit protein in uterine artery smooth muscle from follicular and luteal phase ewes (P = 0.2; Fig. 5B1). In contrast, β₁- and β₂-subunit protein expression was 51% (P = 0.009) and 79% (P = 0.05), respectively, greater in arteries obtained from follicular phase animals vs. luteal phase ewes (Fig. 5, B2 and B3). As above, α-actin did not differ in luteal (n = 4) and follicular (n = 4) phase arteries (P = 0.8), and β₁-subunit increased 37% (P = 0.1; see Supplemental Fig. 1).

Fig. 5.

Effect of the ovarian cycle on BK_Ca subunit expression in the uterine artery during the luteal (n = 4) and follicular (n = 6) phases. A: immunoblot analyses for α-, β₁-, and β₂-subunits. B: densitometry in arbitrary units from the immunoblots for α (B1)-, β₁ (B2)-, and β₂ (B3)-subunits. Data are means ± SE; *P = 0.009 and **P = 0.05, nonpaired t-test.

Immunohistochemistry.

To determine the tissue localization of the α- and β-subunits comprising the BK_Ca channel within the intact uterine artery and to further assess the pattern of expression during the ovarian cycle, we performed immunohistochemistry on randomly selected samples of uterine arteries from three follicular and three luteal phase animals. Nonimmune rabbit serum did not show immunostaining in any vessel examined (Fig. 6, A and E). Although α-subunit immunostaining was observed throughout the uterine smooth muscle in each sample examined, this was not seen in the endothelium of either follicular or luteal phase arteries (Fig. 6, B and F). There also was diffuse immunostaining for the regulatory β₁- and β₂-subunits in the smooth muscle of all arteries assessed, but not in the endothelium. Although densitometric measurements were not made, β₁ (Fig. 6G)- and β₂ (Fig. 6H)-subunit immunostaining appears to be greater than α-subunit immunostaining in the media of follicular phase uterine arteries.

Fig. 6.

Representative immunohistochemistry of BK_Ca α-, β₁-, and β₂-subunits in randomly selected uterine arteries during the luteal (A–D) and follicular (E–H) phases of the ovine ovarian cycle. Figures are at ×40 magnification. NIRS, control nonimmune rabbit serum; L, vessel lumen.

DISCUSSION

Propagation of any species depends on the successful progression through each portion of the reproductive cycle, which in eutherian mammals includes the ovarian cycle, implantation or attachment, placentation, and growth of the maternal uteroplacental and fetal umbilicoplacental vascular beds (28). UBF changes during the reproductive cycle and is responsible for maintaining uterine and placental oxygen and nutrient delivery (28). Although our understanding of UBF regulation in pregnancy has expanded, only modest attention has been given to its regulation in the ovarian cycle and, in particular, the rise in UBF in the follicular phase when the uterus is being prepared for implantation (7, 11, 26). This rise is believed to reflect receptor-mediated actions of ovarian estrogens and local NOS activation (9, 18, 21); however, the remainder of the signaling pathway is incompletely described. In the present study, the rise in UBF in follicular phase ewes was paralleled by upregulation of the cGMP-cPKG pathway and increases in expression of β-regulatory subunits of the BK_Ca channel, which contributes to E₂β-mediated uterine vasodilation and the maintenance of uteroplacental blood flow in pregnancy (23, 30, 33, 36, 37). Thus we now provide evidence that the rise in endogenous ovarian estrogens and UBF in follicular phase ewes, previously shown to be associated with activation of vascular NOS (9, 18, 21), is also associated with upregulation of the cGMP-PKG pathway and BK_Ca.

UBF increases in follicular phase ewes and is associated with increased circulating levels of ovarian estrogens, potent uterine vasodilators, and an elevated E/P (9, 20, 27, 28, 43). During the luteal phase of the cycle, UBF falls, paralleling decreases in the E/P ratio and predominance of P₄, which attenuates estrogen-mediated vasodilation (10, 12, 27, 28). Studies of the interaction of these endocrine changes and UBF in sheep have been difficult because of the seasonality of the cycle and variability in the cyclical pattern of ovarian hormones and UBF (6, 7, 9, 11, 27). Gibson et al. (9) developed a chronic ovine model that allows the study of these interactions independent of the season. They observed a fivefold rise in UBF during the follicular phase that paralleled increases in plasma estrogens and the E/P. In the luteal phase, UBF fell in concert with decreases in the E/P and increases in P₄. Sprague et al. (43) reported similar observations using this model. In our study, UBF also rose three- to fivefold in follicular phase ewes and was associated with low P₄ values. The rise in UBF is also associated with activation of an estrogen- and NO-mediated pathway, since the estrogen receptor (ER) antagonist ICI-182,780 and the NOS inhibitor N^G-nitro-l-arginine methyl ester attenuate the rise in UBF (9, 18, 21). Similar changes occur in ovariectomized nonpregnant ewes after exposure to exogenous E₂β (18, 29, 31, 39, 48, 50), demonstrating that responses to endogenous and exogenous estrogens are very similar. However, ER and NOS inhibition only partially decrease UBF in follicular phase ewes; thus, alternative pathways may also contribute to the rise in UBF. These pathways are unclear as are the changes in vascular cGMP and other aspects of the cascade that contribute to NO-mediated increases in UBF.

E₂β enhances type I and III NOS expression in uterine arteries from nonpregnant ewes, and this is associated with increased cGMP synthesis and uterine vasodilation (29, 31, 39, 48, 50). Although NO contributes to the rise in follicular phase UBF (9), it is unclear if guanylyl cyclase and cGMP are altered. Using the model of Gibson et al. (9), we observed equal fold increases in UBF and uterine artery cGMP in follicular phase animals. However, sGC was unchanged, suggesting enzyme activity increases after exposure to endogenous estrogen. Similar changes are also seen in uterine arteries from late pregnant ewes, which are exposed to endogenous estrogens of placental origin (33). Although endothelial NOS mRNA and protein increase in follicular phase uterine arteries (21, 48), type 1 NOS has not been examined. This is of interest, since daily E₂β increases type I NOS in uterine arteries of ovariectomized ewes (29, 39). Moreover, we have seen increases in uterine artery type I NOS during ovine pregnancy (Rosenfeld, unpublished observations). Future studies should address this.

Smooth muscle contraction and relaxation are regulated by the relative rates of myosin light chain (MLC) phosphorylation by Ca²⁺/calmodulin-dependent MLC kinase and dephosphorylation by myosin phosphatase (5, 13, 42). The NO-cGMP pathway activates relaxation via cPKG and Ca²⁺-dependent and -independent mechanisms (44). The former involves decreases in intracellular Ca²⁺ and the latter activation of MLC phosphatase, which dephosphorylates MLC resulting in SMC relaxation (44, 46). Two cPKG isoforms, cPKG₁ and cPKG₂, exist (25, 44); however, cPKG₁, expressed as cPKG_1α and cPKG_1β, predominates in SMC (46). Both cPKG₁ isoforms are coexpressed in SMC, but their relative levels and specific roles are poorly understood (44, 45). Both modify intracellular Ca²⁺, whereas cPKG_1α also activates MLC phosphatase (44). Notably, the sensitivity of cPKG_1α to cGMP is 10-fold greater than cPKG_1β (2). We found both isoforms in uterine artery SMC, but only cPKG_1α increased in the follicular phase, suggesting endogenous estrogens selectively increase transcription and/or translation. Similar changes are seen in uterine arteries from pregnant sheep, which are exposed to increases in endogenous placental estrogens (1, 33). If cPKG_1α is selectively upregulated by estrogen and is 10-fold more sensitive to cGMP than cPKG_1β, amplification of the NO-cGMP-cPKG pathway is likely to occur after estrogen exposure. This will need to be examined, as well as the mechanisms that regulate cPKG_1α during the reproductive cycle.

BK_Ca contribute to the vasodilatory effects of exogenous estrogens via the NO-cGMP-cPKG pathway (4, 16, 37, 51) and/or directly through β₁-subunit activation (49). β₁- and β₂-regulatory subunits regulate Ca²⁺ and voltage sensitivity and modify BK_Ca activation kinetics (41, 47). The pore-forming α-subunit and both SMC specific β-regulatory subunits were present in uterine artery SMC of cycling nonpregnant sheep, confirming observations in nonpregnant women and ovariectomized nonpregnant and intact pregnant sheep (23, 30, 33, 38). Channel density was unchanged during the ovarian cycle, but β-subunit expression increased at the time of maximum UBF, thereby increasing β/α stoichiometry. Similar alterations occur in E₂β-treated nonpregnant castrated ewes in whom basal UBF increases and responses to acute E₂β exposure are enhanced (23, 39). The change in BK_Ca stoichiometry occurred in concert with increases in uterine vascular NOS expression (29, 31, 39, 48, 50) and cGMP synthesis. Thus upregulation of the NO-cGMP-cPKG pathway in the follicular phase could enhance BK_Ca phosphorylation and activation, resulting in uterine vasodilation. It is notable that nonspecific ER and NOS inhibition do not completely reverse the rise in follicular phase UBF (9, 18). Thus estrogen might also directly activate BK_Ca via the β₁-subunit and contribute to the rise in UBF (37). The role of P₄ in the luteal phase reversal of UBF and attenuation of the NO-cGMP-cPKG-BK_Ca pathway requires further study.

This is the first report showing that increases in endogenous estrogens during the follicular phase of the ovine ovarian cycle contribute to changes in uterine artery BK_Ca expression and possibly function by increasing β/α stoichiometry and thus channel sensitivity and kinetics without altering channel density. This is paralleled by increased uterine artery cGMP contents and cPKG_1α expression; thus, the entire NO-cGMP-cPKG-BK_Ca pathway is upregulated by endogenous estrogen. It is unclear how each segment of the pathway is regulated and if the concertmaster is indeed estrogen or the E/P. If BK_Ca regulate uterine vascular tone during the follicular phase of the ovarian cycle and if increases in β/α stoichiometry modify channel function, the BK_Ca may be a pharmacologic target in conditions associated with impaired uterine vascular function, e.g., diabetes and chronic hypertension.

GRANTS

These studies were supported by National Institutes of Health Grants HD-008783 (C. R. Rosenfeld), HL-49210 (R. R. Magness), and HL87144 (R. R. Magness) and the George L. MacGregor Professorship in Pediatrics (C. R. Rosenfeld).

DISCLOSURES

There are no known conflicts of interest or disclosures regarding these studies.