Endothelial vasodilator production by ovine uterine and systemic arteries: ovarian steroid and pregnancy control of ERalpha and ERbeta levels

Abstract

Pregnancy and the follicular phase are physiological states of elevated oestrogen levels and rises in uterine blood flow (UBF). The dramatic increase in utero-placental blood flow during gestation is required for normal fetal growth and development. Oestrogen exerts its vasodilatory effect by binding to its specific oestrogen receptors (ER) in target cells, resulting in increased expression and activity of endothelial nitric oxide synthase (eNOS) to relax vascular smooth muscle (VSM). However, the regulation of endothelial versus VSM ERalpha and ERbeta expression in uterine arteries (UAs) during the ovarian cycle, pregnancy and with exogenous hormone replacement therapy (HRT) are currently unknown. ER mRNA and protein localization was determined by in situ hybridization (ISH) using 35S-labelled riboprobes and immunohistochemistry (IHC), respectively. UA endothelial (UAendo), UA VSM, omental artery endothelium (OA endo), and OA VSM proteins were isolated and ERalpha and ERbeta protein expression was determined by Western analysis. We observed by ISH and IHC that ERalpha and ERbeta mRNA and protein were localized in both UAendo and UA VSM. Immunoblot data demonstrated ovarian hormone specific regulation of ERalpha and ERbeta protein in UAendo and UA VSM. Compared to luteal phase sheep, both ERalpha and ERbeta levels in UAendo were elevated in follicular phase sheep. Whereas ERbeta was elevated by pregnancy in UAendo and UA VSM, ERalpha was not appreciably altered. eNOS was increased in UAendo from follicular and pregnant sheep. Ovariectomized ewes (OVEX) had substantially reduced UAendo ERbeta, but not UAendo ERalpha or OAendo ERalpha and ERbeta. In contrast, OVEX increased UA VSM ERalpha and ERbeta and decreased OA VSM ERalpha and ERbeta. Treatment with oestradiol-17beta (E2beta), but not progesterone or their combination, increased UAendo ERalpha levels. The reduced ERbeta in UAendo from OVEX ewes was reversed by E(2)beta and progesterone treatment. While ERalpha and eNOS were not elevated in any other reproductive or non-reproductive endothelia tested, ERbeta was augmented by pregnancy in uterine, mammary, placenta, and coronary artery endothelia. ERalpha and ERbeta mRNA and protein are expressed in UA endothelium with expression levels depending on the endocrine status of the animal, indicating UA endothelium is a target for oestrogen action in vivo, and that the two receptors appear to be differentially regulated in a spatial and temporal fashion with regard to the reproductive status or HRT.

Figure 1. Localization of ERα (A) and ERβ (B) mRNAs in ovine uterine arteries

Uterine arteries from pregnant sheep were fixed and paraffin-embedded, and whole uterine artery sections (∼5 μm) were cut. Localization of ERα and ERβ mRNAs was determined by in situ hybridization using ³⁵S-labelled sense (Control) or antisense riboprobes synthesized from a specific ovine ERα cDNA or bovine ERβ cDNA. Silver grains shown in darkfield images represent positive mRNA labelling. Images were taken at 10× magnification. Abbreviations: BL, basal lamina; EC, endothelial cell; L, lumen; VSM, vascular smooth muscle.

Figure 2. ERα and ERβ protein expression in ovine UAendo (A) and UA VSM (B) lysates from luteal, follicular and pregnant ewes

Western blot analysis was performed to compare the relative expression levels of ERα and ERβ protein in UAs from luteal (n = 12), follicular (n = 8), and pregnant (n = 12) sheep. A, UAendo samples represent endothelium (i.e. mechanically isolated tunica intima). B, UA VSM samples are denuded vessels that have had their endothelium removed leaving the basal lamina and VSM. Samples from luteal, follicular and pregnant sheep are expressed as fold of the average of all luteal samples run on the same Western blots. Data are means ± s.e.m. Means with different letter superscripts are statistically different (P < 0.05) within a tissue preparation. Representative Western immunoblots are shown above for ERα and ERβ at molecular weights of 67 and 59 kDa, respectively. The UA VSM immunoblots were reprobed for smooth muscle myosin and showed homogeneity of protein loading.

Figure 3. ERα (A) and ERβ (B) protein expression in ovine uterine artery endothelium of intact luteal phase versus OVEX ovarian steroid hormone treated ewes

Western blot analysis was performed to compare the relative levels of ERα and ERβ protein in luteal (n = 12) versus OVEX sheep treated with vehicle (n = 4), E2β (n = 7), P4 (n = 4) or E2β+ P4 (n = 5). OVEX samples are expressed as fold of the average of the luteal samples run on the same Western blot. Data are means ± s.e.m. Means with different letters are statistically different (P < 0.05). Representative Western immunoblots are shown for ERα and ERβ at molecular weights of 67 and 59 kDa, respectively.

Figure 4. ERα (A) and ERβ (B) protein expression in ovine uterine artery vascular smooth muscle (VSM) of intact luteal phase versus OVEX ovarian steroid hormone treated ewes

Western blot analysis was performed to compare the relative expression of ERα and ERβ protein in luteal (n = 12) versus OVEX sheep treated with vehicle (n = 4), E2β (n = 7), P4 (n = 4) or E2β+ P4 (n = 5). OVEX samples are expressed as a fold of the average of all the luteal samples runs on the same Western blot. Data are means ± s.e.m. Means with different letters are statistically different (P < 0.05). Representative Western immunoblots are shown for ERα and ERβ at molecular weights of 67 and 59 kDa, respectively. These UA VSM immunoblots were reprobed for smooth muscle myosin (SMM) and show homogeneity of protein loading.

Figure 5. ERα (A) and ERβ (B) protein expression in reproductive versus non-reproductive arterial endothelia

Western blot analysis was performed to evaluate the relative levels of ERα and ERβ protein in luteal, follicular, and pregnant sheep reproductive and non-reproductive endothelia. Samples from luteal, follicular and pregnant sheep are expressed as fold of the average of all luteal samples within an artery type, run on the same Western blots. Expression of PAendo is given as fold of the average luteal UAendo also run on the same blot. Treatment groups: luteal: uterine (n = 12), mammary (n = 5), omental (n = 7), renal (n = 6), coronary (n = 7). Follicular: uterine (n = 8), mammary (n = 5), omental (n = 8), renal (n = 6), coronary (n = 8). Pregnant: uterine (n = 12), mammary (n = 6), placental (n = 8) omental (n = 8), renal (n = 8), coronary (n = 8). Data are means ± s.e.m. Means with different letters are statistically different (P < 0.05) within a tissue preparation. ERα UAendo: Lut < Fol (P < 0.05), For ERβ UAendo: Lut < Fol (P < 0.05) and Lut < Preg (P < 0.001); MAendo: Lut < Preg (P < 0.01); CAendo: Lut < Preg (P < 0.05).*For ERα: PAendo < Luteal UAendo but for ERβ PAendo > Luteal UAendo (P < 0.05).

Figure 6. Expression of eNOS in reproductive versus non-reproductive arterial endothelia

Western blot analysis was performed to compare the relative expression of eNOS protein in luteal, follicular and pregnant sheep in reproductive and non-reproductive endothelium. Samples from luteal, follicular and pregnant sheep are expressed as fold of the average of all luteal samples within an artery type, run on the same Western blots. Expression of PAendo is given as fold of the average luteal UAendo also run on the same blot. Treatment groups: luteal: uterine (n = 12), mammary (n = 5), omental (n = 7), renal (n = 6), coronary (n = 7). Follicular: uterine (n = 8), mammary (n = 5), omental (n = 8), renal (n = 6), coronary (n = 8). Pregnant: uterine (n = 12), mammary (n = 6), placental (n = 8), omental (n = 8), renal (n = 8), coronary (n = 8). Data are means ± s.e.m. Means with different letters are statistically different (P < 0.05) within a tissue preparation. UAendo: Lut < Fol (P < 0.001), Lut < Preg (P < 0.001). A representative Western immunoblot is shown above for eNOS at a molecular weight of 140 kDa.