Regulation by Gonadal Steroids of Estrogen and Progesterone Receptors Along the Reproductive Tract in Female Lambs

The regulation of estrogen and progesterone receptor (ER, PR) expression by estradiol (E2) and progesterone (P4) in the oviduct, uterus and cervix of female lambs was studied. The animals received three intramuscular injections of E2, P4 or vehicle with an interval of 24 h and they were slaugthered 24 h after the third injection. Determinations of ER and PR were performed by binding assays and mRNAs of ERα and PR by solution hybridization. High levels of ER and PR in both cervix and oviduct were found in the female lamb, differing from other mammalian species. No significant effects by either E2 or P4 treatment on ER and PR levels in the cervix and oviduct could be observed. E2 treatment increased the mRNA levels of ERa and PR more than 3-fold in the cervix, while P4 treatment increased the mRNA levels of ERa and PR in the uterus. The results show differential effects of gonadal steroids on sex steroid receptor expression along the reproductive tract in female lambs, suggesting that steroid target tissues can modulate responses to the same circulating levels of steroid hormones.


Introduction
Estrogens and progesterone are the main hormones modulating the function of the female reproductive tract by operating through specific intracellular receptors (ER and PR, respectively). The importance of estrogens for normal growth and differentiation of the reproductive tract has been reported, thus the presence of ER for this action is also needed (Greco et al. 1993, Gorski & Hou 1995. Of the two isoforms of ER described -ERα and the recently discovered ERß (Kuiper et al. 1996) -ERα is predominant in the oviduct, uterus and cervix of the rat (Wang et al. 1999(Wang et al. , 2000. It has been demonstrated in ERα knock-out mice that although they have a normal appearance, abnormalities exist in the reproductive tract (Lubahn et al. 1993). Yamashita et al. (1989) demonstrated in 1-day-old mice the presence of epithelial ER in the oviduct and cervix, but the receptors could not be detected in uterine epithelia until day 4. Regarding sheep, previous studies showed high concentrations of ER and PR in the uterus of prepubertal ewes (Garófalo & Tasende 1996) differing from some species in which PR expression is low or undetectable at this stage of development (guinea pig: Pasqualini et al. 1980, dog: Lessey et al. 1981. A recent study has described the uterine changes in ER and PR expression during the early postnatal period in the lamb and the results support the hypothesis that ERα is also necessary for normal uterine growth and development in ovines (Taylor et al. 2000). Zhao et al (1999) detected few ER positive nuclei in cervical cells of prepubertal ewes, but no quantitative studies were performed. Although sex steroid hormones are known to act in unity with their respective receptor molecule to induce cellular effects, little or no information exists concerning the presence of ER and PR in the cervix and oviduct of immature ewes. In the ovariectomized adult ewe, sex steroid receptor levels in the isthmic oviduct are equal to or slightly higher than in the uterus (Stone & Miller 1978). Similar findings were reported for receptor levels in the cervix vs uterus of the postpartum ewe (Rodriguez-Piñon et al. 2000). The regulation of sex steroid receptors in the oviduct and cervix is similar to that reported for the uterus, i.e., E2 stimulates the expression while P4 downregulates it; although in primates the changes in cervical ER and PR throughout the menstrual cycle are more subtle than in the endometrium and myometrium (Gorodeski 1996, Brenner & Slayden 1996. Miller et al. (1977) have found high rates of protein synthesis and increased RNA:DNA ratios in the ovine endometrium and oviduct at or shortly after estrus, associated with an increased E2 level in plasma. Progesterone action on RNA and protein synthesis differs in the ovine endometrium and isthmic oviduct, with highly significant increases in the former but no effect in the latter (Miller 1976, Miller et al. 1977. Similar findings were described in the immature ewe (Meikle et al. 1997), since the administration of P4 increased uterine weight but had no effect on the oviductal and cervical weights, differing from E2 treatment after which all parts of the reproductive tract increased in weight. Antagonistic effects of progesterone on estrogen action, like inhibition of E2-induced uterine growth, have been demonstrated in rodents (Clark et al. 1977), but P4 does not decrease the weight of the uterus in E2-treated ovariectomized ewes (Miller et al. 1979, Stone et al. 1982). Furthermore, it was shown that P4 alone could increase uterine weight in 2 months old lambs without the need of previous estrogen priming (Meikle et al. 1997). The different actions of the steroid hormones along the reproductive tract are dependent on the presence of the hormone and its specific receptor concentration in the target tissue, which is also regulated by E2 and P4 (Clark et al. 1992). In this study, we focused on the regulation of ER and PR expression by gonadal steroids in the cervix, uterus and oviduct of female lambs.

Animals and treatments
Eleven 2 months old Corriedale female lambs (body weight, mean ± SEM: 11.4 ± 0.3 kg) born in September were used. The experiment was carried out in November which corresponds to the nonbreeding season for the Corriedale breed in the experimental field of Veterinary Faculty, Montevideo, Uruguay (35°SL, spring). The lambs were maintained under natural environmental conditions and they were allowed to nurse freely during the experiment. The animals received daily injections of 17 ß-estradiol (E2) (1 µg/kg, group E, n=4), progesterone (P) (0.3 mg/kg, group P, n=4) or corn oil vehicle (0.1 ml/kg group C, n=3) i.m. on Days 0, 1 and 2. Blood samples for P4 and E2 determinations were collected daily by jugular venipuncture before, during and after the treatment, and on Day 3 all lambs were slaughtered. The endocrine data as well as the uterine receptor determinations by binding assays have been published previously (Meikle et al. 1997). Serum levels of E2 and P4 were significantly different in accordance with the treatments used in the different groups. Uteri, cervices and oviducts were dissected at 0-4 °C and weighed. Treatment with E2 significantly increased uterine, cervical and oviductal weights while P4 only affected uterine weight (Meikle et al. 1997). The tissues were frozen in liquid nitrogen and stored at -80°C until assayed. ER and PR determinations by binding assays were performed in cervices and oviducts, as well as mRNA determinations of ERa and PR by solution hybridization. Uterine samples from the lower uterine zone, defined as the third portion next to the cervix, were taken to perform mRNA determinations for ERα and PR.

Assays of steroid receptors
Ligand-binding assays were performed on the cytosolic fractions from the cervix and oviduct of each animal as described previously (Garófalo & Tasende 1996, Meikle et al. 1997). In the ligand-binding assay both ERa and ERb are determined (i.e., the sum of their binding activity is measured). The term cytosolic receptors is used in this study to indicate receptors found in the supernatant fraction of a tissue homogenate after a high-speed centrifugation. Briefly, the cytosolic fraction was incubated with 5 to 6 increasing concentrations of [2,4,6,7-3 H]-estradiol-17ß 86 Ci/mmol (0.3-15 nM), or 3 H-ORG-2058, (16α-ethyl-21-hydroxy-19-nor[6,7-3 H] pregn-4en-3,20-dione 40 Ci/mmol (0.5-30 nM) for 18 h with or without 200-fold molar excess of either unlabeled diethylstilbestrol or unlabeled ORG-2058, respectively. The separation of free hormone was by dextran-coated charcoal and radioactivity was measured by liquid scintillation counting. Protein concentrations were determined by the method of Lowry et al. (1951), using BSA as the standard. A linear regression test of the inverse Scatchard model (Braunsberg 1984) analysis of the data was performed to obtain the dissociation constant (Kd, nM), and the concentration of receptor sites at the intercept, B max, expressed in fmol/mg pro-tein. Dissociation constants (Kd) for ER were 0.57 ± 0.07 nM for the oviduct and 0.30 ± 0.22 nM for the cervix (n=11), while PR Kd for the oviduct was 1.15 ± 0.16 nM and for the cervix 1.31 ± 0.22 nM (n=11). All Kds were well within the range reported for uterine receptors (Garófalo & Tasende 1996, Meikle et al. 1997.

Hybridization analysis of mRNA
A solution hybridization assay of specific mRNAs for ovine ERa and PR was performed on samples of cervices, uteri and oviducts. The method was essentially performed as in Persson et al. (1997). In short, total nucleic acids were prepared and the concentration of DNA in the TNA samples was measured fluorometrically. The probes used for ERα mRNA and PR mRNA determinations were derived from plasmids containing 360 and 314 bp cDNAs from the ovine ERα and PR, respectively, supplied by Dr. N. Ing, Texas A & M University, TX, USA (Ing et al. 1996). Probes were synthesized in vitro and radiolabeled with 35 S-UTP as described by Melton et al. (1984). Overnight incubation was performed at two different concentrations, and samples were then treated with RNase to digest unhybridized RNA. Labeled hybrids were precipitated with trichloroacetic acid and collected on filters, whereafter radioactivity was determined in a liquid scintillation counter. All the samples from the experiment were determined in the same assay. Receptor mRNA levels were expressed as cpm in relation to DNA content (cpm/µg DNA). In Figure 2, the mRNA levels of ERa and PR are expressed as percentages of the control group average to show changes induced by the treatment.

Statistical analysis
Statistical analyses were carried out using the procedures of the Statistical Analysis Systems Institute Inc. (1994). Differences between groups concerning receptor and mRNA deter-minations were analyzed by the General Linear Model procedure for analysis of variance according to a model including the treatment (groups C, E and P), region of the reproductive tract (cervix, oviduct and uterus), and the interaction between treatment and region. Data are presented as least-square means ± standard errors for each treatment group and P<0.05 is considered as significant. Normality of distribution of the residuals were analyzed using the Univariate Procedure.

ER and PR
Concentrations of ER and PR (fmol/mg protein) in the oviduct and cervix of control and treated lambs are shown in Figure 1A and 1B respectively. There was a significant effect of region of the reproductive tract on both ER and PR levels (P=0.03 and P<0.005, respectively). In control lambs, ER concentrations in the cervix and oviduct were similar (365 ± 65 and 433 ± 142), but significantly lower than in the mid- dle zone of the uterus of the same lambs, (682 ± 53) (Meikle et al. 1997). The same occurred with PR of which the cervix and oviduct had lower concentrations (167 ± 71 and 308 ± 117) than the uterus (1479 ± 153) (Meikle et al. 1997). No significant effects by either E2 or P4 treatment on ER and PR levels in the cervix and oviduct could be observed. No significant interaction between treatment and region was found.

mRNA of ER α and PR
Concentrations of mRNA of ERα and PR in the cervix, oviduct and uterus of the lambs are presented as percentage of controls in Figure 2A and 2B respectively. There was a significant effect of region of the reproductive tract on both mRNA levels (P<0.0005 and P<0.005, respectively). The cervix of control lambs had a lower ERα mRNA level (cpm/mg DNA) than the oviduct and uterus (34.7 ± 3.7, 79.7 ± 11.5 and 80.7±3.2 respectively). PR mRNA concentra- tions (cpm/µg DNA) were lower in the cervix and oviduct than in the uterus (63.0 ± 8.6, 64.0 ± 11.7 and 161 ± 6.7 respectively). Significant interaction between treatment and region was found for ERα mRNA (P=0.0001) and PR mRNA (P=0.0016). The ERα mRNA level increased more than 3fold in the cervix after E2 treatment as compared to both the control group and P4 treatment ( Figure 2A). No differences were found in the ERα mRNA levels in the oviduct, although the levels in the P4 treated group tended to be lower than in the controls. The ERα mRNA level in the uterus was higher after P4 treatment as compared to the control group. The cervical PR mRNA level increased 4-fold after E2 treatment as compared to the control group ( Figure 2B). There were no significant changes seen in the oviduct. In the uterus, P4 treatment increased the PR mRNA level 2-fold, whereas E2 showed no effect.

Discussion
This is the first report of expression of sex steroid receptors in the oviduct of the immature ewe. Zhao et al. (1999), reported few ER immunopositive cells in the cervix of two prepubertal ewes. In contrast, in this study, high levels of ER and PR were found in both cervix and oviduct, although less than in the uterus of the same animals (Meikle et al. 1997). The receptor proteins are functional since treatment induced changes in receptor and mRNA expression. The presence of ER was expected since it is known that ER are needed for normal development of the reproductive tract (Lubahn et al. 1993), but it is not clear why PR levels were so high at this stage of development. Similarly, no clear biological role could be attributed to the presence of high affinity PR in the oviduct in concentrations equal to or sligthly higher than those found in endometrium or whole uterus in the adult ewe (Stone & Miller 1978). In the imma-ture dog, both the uterus and oviduct contain relatively low levels of ER while PR is absent (Lessey et al. 1981). In contrast, the present study and other findings in sheep (Garófalo & Tasende 1996) show that ruminants differ from other mammals in sex steroid receptor expression along the reproductive tract.
No effect of the hormone treatment on receptor concentrations in the cervix and oviduct could be demonstrated and this contrasts with the downregulation of ER and PR expression after E2 and P4 treatment seen in uteri of the same animals (Meikle et al. 1997). Scharl et al. (1988) found that the endocervix -unlike endometrium and myometrium -undergoes minimal changes with regard to nuclear receptor content during the human menstrual cycle. In the oviduct, a clear steroid hormone dependence of ER and PR expression has been found in primates and rodents (Brenner & Slayden 1996). On the other hand, no differences in ER content were found after E2 treatment in the oviduct of immature dogs (Lessey et al. 1981) and PR downregulation in the rabbit oviduct was delayed and not so pronounced as in the uterus (Muechler et al. 1976). In the present study, the tissues were sampled 3 days after the first E2 injection and the dynamics of the receptor contents could not be observed during the period of treatment. This study shows that E2 and P4 have different effects on the regulation of mRNA expression of sex steroid receptors along the reproductive tract in prepubertal ewes. E2 treatment induced an increase in mRNA levels of ERα and PR in the cervix, but no such effect was observed in the oviduct and uterus. In uteri of ovariectomized ewes, E2 increased mRNA levels of ERα and PR with maximal concentrations 24 h after the E2 injection (Ing et al. 1996). We have recently showed that mRNA levels of ERα and PR increase 2-fold 12 h after E2 injection and levels remain high during daily injections of E2 (Meikle et al. 2000a). As for the receptor deter-minations, mRNAs were measured at only one point of the treatment period, and an earlier increase in mRNA concentrations could have been missed. Nevertheless, the 3 to 4-fold increase in mRNA levels of ERα and PR after E2 treatment found only in the cervix, suggests that E2 in this organ maintains the stimulus on mRNA expression for a longer period or that the cervix is able to respond to repeated hormone treatments. An interesting finding was the P4-induced increase in mRNA levels in the uterus, different from the action on the cervix and oviduct. These results agree with previous reports in sheep, which showed that P4 increases RNA and protein synthesis in the endometrium but have no effect in the isthmic oviduct (Miller 1976, Miller et al. 1977. The inability to influence nucleic acid synthesis by P4 in the oviduct was also reported for the mouse (Bronson & Hamilton 1972). In the oviduct of macaques, E2 stimulated differentiation of a fully ciliatedsecretory epithelium primed for gamete transport, whereas P4 induced cell atrophy (Brenner et al. 1990). In the primate endometrium, E2 stimulates epithelial and stromal proliferation while P4 suppresses proliferation and induces differentiation by converting the tissue to a hypertrophied and secretory state able to support implantation (Slayden et al. 1993). This difference in the action by P4, i.e., suppressive effects in the oviduct and inductive in the endometrium, agrees with the tendency to lower levels of ERα mRNA found in the oviduct and the increased ERα mRNA and PR mRNA expression in the uterus observed after P4 treatment as seen in the present study. This is consistent with the effect of P4 on the weight of the oviduct and uterus found in these animals, with no change in the former and a significant increase in the latter (Meikle et al. 1997).
The overall response of the uterus to steroid stimulation will be the product of the combined responses through sex steroid receptor levels in various cell types. Immunohistochemical studies on ER and PR expression during the ovine estrous cycle (Cherny et al. 1991;Spencer and Bazer 1995), showed that in the majority -but not all -of the uterine compartments, receptor levels were high shortly after estrus in the different compartments and then declined to negligible levels at mid luteal phase. For example, ER levels in the innermost region of the basalis zone are not suppressed by the end of the luteal phase (Cherny et al. 1991