|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
Department of Animal Sciences, The Ohio State University, Columbus, Ohio 43210
| ABSTRACT |
|---|
|
|
|---|
-dihydrotestosterone (DHT) on estrogenic actions in the pig uterus and its associations with changes in expression of the estrogen receptor (ER)
and ERß. Postpubertal gilts (120130 kg of body weight; n = 16) were ovariectomized, and 34 weeks later received once-a-day injections (i.m.) of one of the following treatments during four consecutive days: 1) vehicle (corn oil), 2) E (250 µg), 3) E (250 µg) plus 1 mg DHT, or 4) E (250 µg) plus 10 mg DHT. Uterine tissues were collected 24 h after the last treatment. Gilts receiving E or E plus 1 mg DHT had greater uterine wet weight, uterine horn diameter, luminal epithelium thickness, and endometrial gland diameter compared with gilts treated with vehicle or E plus 10 mg DHT. Gilts receiving E or E plus 1 mg DHT were not different in these characteristics. Relative amounts of mRNAs in the endometrium for the cell proliferation marker histone H2a and the E-inducible protein complement component C3 increased in gilts treated with E compared with gilts treated with vehicle. E-induced increases in histone H2a and C3 mRNAs were not altered by cotreatment with E plus 1 mg DHT but were inhibited by E plus 10 mg DHT. Androgen receptor (AR) mRNA in the endometrium increased by treatment with E. Cotreatment of gilts with E and DHT did not alter the E-induced AR mRNA increase. Gilts treated with E plus 10 mg DHT had lesser amounts of immunoreactive ER
in cell nuclei of the myometrium and endometrial stroma and a tendency for a decrease in luminal epithelium compared with gilts treated with E. Amounts of immunoreactive ER
in glandular epithelium were not influenced by the treatments. Relative amounts of ER
and ERß mRNAs decreased in the endometrium of gilts treated with E plus 10 mg DHT compared with gilts treated with E. Downregulation of the ERs, particularly ER
in the myometrium and endometrial stroma, might be a relevant mechanism in the antagonism of estrogenic effects by DHT in the pig uterus.
androgen receptor, estradiol receptor, gene regulation, steroid hormones, uterus
| INTRODUCTION |
|---|
|
|
|---|
The actions of androgens in the uterus depend on the concentrations of these hormones reaching the uterus, which have been shown to be greater than in systemic blood in pigs [12]. Concentrations of androgens in the pig uterus might increase in the presence of elongating blastocysts that have the capacity to synthesize androgens [13, 14]. Furthermore, certain abnormal conditions can significantly increase follicular secretion of androgens, which might enhance (or inhibit) gene transcription, or interactions with other pathways, in the uterus, particularly if parallel increases in the androgen receptor (AR) occur [15].
Results using the nonaromatizable AR agonist 5
-dihydrotestosterone (DHT), and AR antagonists indicated that androgens inhibited estrogenic effects by activating the AR [8, 10, 16]. What occurs after AR activation is not completely clear. Panet-Raymond et al. [17] determined that ligand-activated AR was able to physically interact with the ER
(but not ERß), producing heterodimers with less transactivational activity. Alternatively, the antagonism of estrogen actions by androgens might be through inhibition of ER(s) synthesis. Results regarding this possible mechanism have not been conclusive. For instance, in the mammary gland of rhesus monkeys, testosterone partially blocked the upregulation of ER
mRNA induced by estradiol-17ß (E) [9]. Other investigators [6] have determined, using ligand-binding assays, that cotreatment of rats with E and DHT did not decrease the concentrations of ER (presumably both ER
and ERß) in uterine homogenates. A better understanding of androgenic effects on ER expression in the uterus requires examining different cell types and expression of the more recently discovered ERß. In the present experiment, 1) effects of DHT on physiological, morphological, and biochemical actions induced by E; and 2) simultaneous changes in amounts of immunoreactive ER
protein, and ER
and ERß mRNAs in the uterus of ovariectomized gilts, were determined.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Histology
Tissue sections were stained with hematoxylin and eosin and examined using light microscopy. Thickness of the luminal epithelium at two locations and diameters of 30 circular cross sections of endometrial glands were determined using a micrometer at a magnification of 400x. The means of luminal epithelium thickness and endometrial gland diameter were calculated and used for statistical analysis.
Immunohistochemistry
The antibody against ER
, PA1-309, and the corresponding immunizing peptide (PEP-037) were obtained from Affinity BioReagents (Golden, CO). The sequence (NELEPLNRPQL) of peptide PEP-037 is 100% conserved in several species, including the pig (residues 2132, GenBank accession Z37167).
Immunohistochemistry (IHC) was performed using the Vectastain ABC kit (Vector Laboratories Inc., Burlingame, CA) and was similar to a procedure validated in our laboratory [18]. Briefly, sections of uterine tissue were deparaffinized and hydrated in serial alcohol solutions and then subjected to antigen retrieval by incubation in 0.01 M sodium citrate, pH 6.0, for 30 min at 95°C. Slides remained in the same citrate buffer until the temperature decreased to 38°C. Tissue sections were rinsed in PBS (pH 7.3) for 10 min at room temperature and then treated with 0.3% hydrogen peroxide in PBS for 30 min to block endogenous peroxide. Sections were washed three times (5 min each) with PBS containing 0.1% gelatin (PBS-Gel). Subsequent washings were performed in the same manner. To help decrease background staining, sections were incubated in normal goat serum for 20 min. Thereafter, sections were washed with PBS-Gel and incubated with antibody (4 µg/ml in PBS-Gel) for 18 h at 4°C. Sections were washed in PBS and then incubated in biotinylated goat anti-rabbit IgG (secondary antibody) for 30 min. This was followed by washing in PBS, incubation in Vectastain ABC reagent for 60 min, and washing again in PBS. Sections were incubated for 8 min with diaminobenzidine (DAB) substrate containing nickel chloride to produce gray-black staining. Sections were rinsed in water, dehydrated using increasing alcohol solutions, and covered with a coverslip and Permount. Adjacent tissue sections were processed simultaneously as negative controls (primary antibody preabsorbed with immunizing peptide, no primary antibody, or no secondary antibody). Sections of all animals were processed in a single assay.
Binding of ER
by the PA1-309 antibody was verified by Western blotting following a published procedure [18]. Aliquots of endometrial tissue homogenates (equivalent to 100 µg of protein) were loaded into a 9% polyacrylamide gel for electrophoretic analysis and then transferred by blotting onto a polyvinylidene difluoride membrane. The ER
was detected using reagents provided in the Vectastain ABC kit and DAB substrate for peroxidase. A band of approximately 66 kD, coincidental in size with the porcine ER
, was observed (Fig. 1).
|
Images (8-bit, 256 tones of gray) of two random areas of the myometrium, endometrium (adluminal and basal), and luminal epithelium were obtained using a CCD camera attached to a light microscope. Images were used for densitometric analysis of staining intensity in nuclei of the myometrium, glandular epithelium, luminal epithelium, and adluminal (beneath the luminal epithelium) and basal endometrial stroma using the ImageJ software (National Institutes of Health, available at http://rsb.info.nih.gov/ij/) as previously described [18]. Nuclei were outlined using a minimum threshold, and the mean gray value of outlined nuclei was determined using the particle analysis feature of ImageJ. The mean gray value after background correction was considered immunostaining intensity. This type of analysis was not performed in basal endometrial stroma because in some samples it was not possible to accurately identify and outline stromal cells using a threshold owing to their low staining intensity (similar to background and less than adjacent glandular epithelium). For this reason, staining intensity in stromal cells of the basal endometrium was graded 1 (absent), 2 (low), 3 (medium), or 4 (high) by an observer unaware of sample identification.
Reverse Transcription Polymerase Chain Reaction (RT-PCR)
Total RNA was isolated from endometrial tissue using TRI Reagent (MRC Inc., Cincinnati, OH) and analyzed using formaldehyde-gel electrophoresis to assess RNA integrity (28S to 18S rRNAs ratio). Purity of RNA samples was verified by calculating the ratio of the absorbances at 260 and 280 nm. Concentrations of total RNA were determined using the absorbance at 260 nm. The RNA samples were stored at -80°C until used for semiquantitative RT-PCR determinations of the following mRNAs: ER
; ERß; AR; the estrogen-inducible gene complement component C3 [1921]; histone H2a (cell proliferation marker, S-phase of the cell cycle [22, 23]); and the housekeeping gene ribosomal protein L19.
Determinations of relative amounts of mRNA by RT-PCR were performed using a kit (GenAmp, Applied Biosystems, Foster City, CA). Primers and product sizes are described in Table 1. Briefly, RNA was reverse transcribed using Maloney murine leukemia virus reverse transcriptase primed with random hexamers. Aliquots (5 µl) of the reverse transcription reaction, equivalent to 250 ng of total RNA, plus AmpliTaq DNA polymerase (1.25 U), 1.5 mM MgCl2, and 0.3 µM of primers (total volume of 25 µl) were used for the amplification of ER
, ERß, AR, C3, histone H2a, and L19. Samples were incubated at 95°C for 1 min and then subjected to cycles of 95°C for 1 min, 55°C (57°C for histone H2a) for 45 sec, and 72°C for 2 min. The numbers of cycles for ER
, ERß, AR, C3, histone H2a, and L19 were 28, 33, 25, 28, 28, and 23, respectively. These numbers of cycles were within the exponential portion of the corresponding amplification curves as determined during the validation processes. Immediately after the last cycle, reaction tubes were incubated at 72°C for 7 min, cooled to 5°C, and then stored at -20°C until analysis by agarose gel electrophoresis. Bands were visualized using ethidium bromide staining and UV illumination and photographed using a gel documentation system. Staining intensity of each cDNA band was determined using densitometric analysis (ImageJ). The relative content of mRNA for each gene was expressed as a ratio of band intensity relative to L19. No products were observed in negative controls in which reverse transcriptase was replaced by water. Authenticity of RT-PCR products was verified by sequencing.
|
Statistical Analyses
Data were analyzed by one-way ANOVA except for grades of staining intensity in basal stroma, which were analyzed using the Kruskal-Wallis test. Data having heterogeneity of variance were transformed into logarithms for analysis. In these cases, means and SEM of original data are presented. Pairwise comparisons of means were performed using protected least significant difference tests, or the method described by Dunn [24] applicable after the Kruskal-Wallis procedure. Analyses were performed using the SYSTAT software (SPSS, Chicago, IL)
| RESULTS |
|---|
|
|
|---|
|
Immunohistochemical staining of ER
was observed in nuclei of various cell types of the uterus including epithelial (luminal and glandular), stromal, and myometrial (Fig. 2). Gilts treated with E plus 10 mg DHT had lesser amounts of immunoreactive ER
in the myometrium than gilts treated with vehicle, E, or E plus 1 mg DHT. In luminal epithelium, ER
immunostaining in gilts treated with E plus 1 or 10 mg of DHT tended (P < 0.10) to decrease relative to gilts treated with E and was less (P < 0.05) compared with gilts treated with vehicle. ER
immunostaining in adluminal and basal endometrial stroma of gilts treated with combinations of E and DHT was less (P < 0.05) than in gilts treated with E or vehicle. No differences were observed in ER
immunostaining between gilts treated with vehicle and E in all cell types examined. Treatments did not affect (P = 0.98) ER
immunostaining in nuclei of glandular epithelium (Table 3).
|
|
Cotreatment of gilts with E and 10 mg DHT decreased relative amounts of ER
mRNA in the endometrium compared with gilts receiving the other hormonal treatments (Table 4). Relative amounts of ER
mRNA were not different between gilts treated with vehicle, E, and E plus 1 mg DHT. ERß mRNA in the endometrium was less (P < 0.05) in gilts treated with E compared with vehicle-treated gilts, and was further decreased (P
0.06) in gilts treated with E plus 10 mg DHT relative to the other treatment groups (Table 4). Amounts of AR mRNA in the endometrium of gilts treated with E and E plus 1 or 10 mg DHT were not different, but were greater (P < 0.05) than in gilts treated with vehicle (Table 4).
|
Relative amounts of complement component C3 mRNA in whole endometrium were greater (P < 0.05) in gilts treated with E or E plus 1 mg DHT and was not different in gilts treated with E plus 10 mg DHT relative to those receiving vehicle. Relative amounts of histone H2a mRNA in whole endometrium were approximately 9- to 10-fold greater in gilts treated with E or E plus 1 mg DHT compared with gilts treated with vehicle. Gilts administered E plus 10 mg DHT had less (P = 0.01) histone H2a mRNA relative to gilts treated with E; however, values were greater than in gilts treated with vehicle (Table 4).
| DISCUSSION |
|---|
|
|
|---|
Antagonism of estrogenic effects by DHT in the pig uterus occurred simultaneously with decreases in relative amounts of immunoreactive ER
and ER's mRNAs. Interestingly, significant decreases in ER
immunoreactivity were observed in the myometrium and endometrial stroma, whereas no changes were detected in glandular epithelium. Cells of the endometrial stroma appeared to be more sensitive to DHT antagonism than other uterine cell types because cotreatment of gilts with E plus the smallest dose of DHT (1 mg) produced a significant decrease in ER
immunoreactivity, which was not observed in other cell types. This is the first description of inhibition of ER
and ERß mRNAs and cell-type specific inhibition of immunoreactive ER
in the uterus of a mammalian species cotreated with estrogen and androgen. Downregulation of the ERs by DHT, and subsequent inhibition of E-responsive genes (as suggested by the results on C3 mRNA), could mediate, at least in part, the attenuation of estrogenic effects produced by DHT. In an experiment [6] in which immature rats were treated with E plus DHT, no changes were observed in concentrations of estrogen receptors in nuclear and cytosolic fractions of uterine homogenates. Perhaps the cell-type specific nature of DHT effects on ER
could have decreased the ability to detect changes in whole uterine preparations.
Our attempts to detect ERß by IHC or Western blotting were unsuccessful and likely due to the presence of small cellular amounts of ERß. Differences in PCR amplification (numbers of cycles and ratios to L19) indicated more than 100-fold greater abundance of ER
mRNA compared with ERß, suggesting that ER
might be more relevant than ERß in mediating estrogen actions in the pig uterus as proposed in rodents [26, 27]. However, ERß has been detected by IHC in luminal and glandular epithelia of pregnant pigs [28], indicating influences of methodologies or effects of physiological stage on ERß gene expression.
Ovariectomized gilts treated with vehicle had relatively large amounts of immunoreactive ER
and ER
mRNA in the uterus. Therefore, ovarian factors did not appear to be necessary for maintaining ERs in the pig uterus. This assertion was supported by results of relatively greater amounts of ER
in the uterus of ovariectomized neonatal pigs [29, 30]. Moreover, treatment with E in the present experiment produced no change in the contents of immunoreactive ER
protein and ER
mRNA. These effects of E agree, in part, with results in rats examined 24 h after treatment with E (similar to the present experiment) in which it was observed that ER
immunostaining decreased in glandular epithelium but did not change in other uterine cell types [26]. Although not examined, it is possible that changes in ER
could have occurred before 24 h because downregulation of ER
beginning 46 h after E administration, and a return to pretreatment levels by 24 h, have been described in pigs [31] and rats [3234]. We have observed (present experiment) that E decreased ERß mRNA in endometrial samples obtained 24 h after E administration.
Endometrial epithelia appear to depend on interactions with stromal cells to undergo certain functions. In this regard, it has been proposed that ER
in stromal cells is necessary for stimulation of epithelial cell proliferation, and that ER
in both stromal and epithelial cells is necessary for secretory functions by epithelial cells in response to E [21, 35]. Our results support the importance of ER
in stromal cells because these were the only endometrial cells that clearly exhibited a decrease in amounts of ER
associated with a remarkable inhibition of estrogenic effects by DHT. Furthermore, downregulation of C3 mRNA in the present experiment, simultaneously with a decrease in ER
in the stroma and a tendency for a decrease in the luminal epithelium, is consistent with the above hypothesis. C3 is secreted by the luminal epithelium [19], and the C3 gene promoter contains an estrogen response element [20, 36]. The specific factors from stromal cells necessary for regulation of endometrial epithelia are not known and might vary depending on particular cellular functions.
How activation of the AR downregulates ERs and how it would occur in the majority of cell types and not in glandular epithelium is not clear. In the present experiment, ER
was detected in the myometrium, endometrial stroma, and the glandular and luminal epithelia, similar to previous descriptions in intact (nonovariectomized) gilts [37] and sows [38]. The AR is primarily present in endometrial epithelia, and in lesser amounts in myometrium and endometrial stroma of the pig uterus [18]. Therefore, AR and ER
appear to be coexpressed within the same uterine cell types, and these receptors [17], or their pathways, might be able to interact. These interactions might be supported by the upregulation of AR by E [18, present results], which was not inhibited by cotreatment with E plus DHT. The presence of androgen response element(s) in the promoter of the ER genes has not been reported, so a direct action of AR on transcription of the ER genes is not supported. Alternatively, DHT might have inhibited secondary transcriptional regulators [39] or factors controlling ER mRNA stability. The latter has been considered the mechanism by which E treatment increased ER
mRNA in sheep endometrium [40]. The absence of one or more factors mediating the inhibitory effects of DHT on ER
in glandular epithelium might explain the lack of effects observed in these cells. Biologically, this might represent a protection against androgenic effects on ER
in glandular epithelium, and in some degree in luminal epithelium, because these cell types of the pig uterus have the greatest amounts of AR [18]. These relationships from the present study suggest that an adequate balance between androgenic and estrogenic effects seem to be needed for normal uterine functions as has been proposed for certain male reproductive processes [41].
In conclusion, simultaneous administration of DHT and E to ovariectomized gilts inhibited typical effects observed in the uterus after administration of E. These effects of DHT were associated with decreases in ER
immunostaining mainly in the myometrium and endometrial stroma but not in glandular epithelium, and with downregulation of the ER
and ERß mRNAs in whole endometrial preparations.
| FOOTNOTES |
|---|
2 Correspondence: Horacio Cárdenas, 2027 Coffey Road, Columbus, OH 43210. FAX: 614 292 7116; cardenas-seijas.2{at}osu.edu ![]()
Received: 18 August 2003.
First decision: 1 September 2003.
Accepted: 24 September 2003.
| REFERENCES |
|---|
|
|
|---|
and ß in the rat uterus. Biol Reprod 1999 61:955-964
) and the proliferative marker Ki-67 in the sow uterus at different stages of the estrous cycle. Reprod Dom Anim 2003 38:5-12
messenger ribonucleic acid in sheep endometrium by increasing its stability. Biol Reprod 1999 60:134-139This article has been cited by other articles:
![]() |
M. L. Rhoads, J. P. Meyer, W. R. Lamberson, D. H. Keisler, and M. C. Lucy Uterine and Hepatic Gene Expression in Relation to Days Postpartum, Estrus, and Pregnancy in Postpartum Dairy Cows J Dairy Sci, January 1, 2008; 91(1): 140 - 150. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. M. Joyce, R. C. Burghardt, R. D. Geisert, J. R. Burghardt, R. N. Hooper, J. W. Ross, M. D. Ashworth, and G. A. Johnson Pig Conceptuses Secrete Estrogen and Interferons to Differentially Regulate Uterine STAT1 in a Temporal and Cell Type-Specific Manner Endocrinology, September 1, 2007; 148(9): 4420 - 4431. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Ka, S. Al-Ramadan, D. W. Erikson, G. A. Johnson, R. C. Burghardt, T. E. Spencer, L. A. Jaeger, and F. W. Bazer Regulation of Expression of Fibroblast Growth Factor 7 in the Pig Uterus by Progesterone and Estradiol Biol Reprod, July 1, 2007; 77(1): 172 - 180. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Zang, L. Sahlin, B. Masironi, E. Eriksson, and A. Linden Hirschberg Effects of Testosterone Treatment on Endometrial Proliferation in Postmenopausal Women J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2169 - 2175. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ivanga, Y. Labrie, E. Calvo, P. Belleau, C. Martel, V. Luu-The, J. Morissette, F. Labrie, and F. Durocher Temporal analysis of E2 transcriptional induction of PTP and MKP and downregulation of IGF-I pathway key components in the mouse uterus Physiol Genomics, March 14, 2007; 29(1): 13 - 23. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. S. Bhattacharyya, A. V. Krishnan, S. Swami, and D. Feldman Fulvestrant (ICI 182,780) down-regulates androgen receptor expression and diminishes androgenic responses in LNCaP human prostate cancer cells. Mol. Cancer Ther., June 1, 2006; 5(6): 1539 - 1549. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Yan, P. L Ryan, F. F Bartol, and C. A Bagnell Uterotrophic effects of relaxin related to age and estrogen receptor activation in neonatal pigs. Reproduction, May 1, 2006; 131(5): 943 - 950. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. V. Nantermet, P. Masarachia, M. A. Gentile, B. Pennypacker, J. Xu, D. Holder, D. Gerhold, D. Towler, A. Schmidt, D. B. Kimmel, et al. Androgenic Induction of Growth and Differentiation in the Rodent Uterus Involves the Modulation of Estrogen-Regulated Genetic Pathways Endocrinology, February 1, 2005; 146(2): 564 - 578. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. M. Miller, D. J. Tindall, and P. Y. Liu Of Mice, Men, and Hormones Arterioscler. Thromb. Vasc. Biol., June 1, 2004; 24(6): 995 - 997. [Full Text] [PDF] |
||||
![]() |
A. A. Kowalski, D. S. Vale-Cruz, F. A. Simmen, and R. C.M. Simmen Uterine Androgen Receptors: Roles in Estrogen-Mediated Gene Expressionand DNA Synthesis Biol Reprod, May 1, 2004; 70(5): 1349 - 1357. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |