|
|
||||||||
Articles |
Knockout Mice1
a Departments of Animal Sciences and
b Biochemistry and Child Health, University of Missouri at Columbia, Columbia, Missouri 65211
c Department of Biomedical Sciences, University Medical School, Edinburgh EH8 9AG, United Kingdom
| ABSTRACT |
|---|
|
|
|---|
(ER
) knockout (ER
KO) female mice are infertile. Initially, they exhibit normal follicular development, but by 45 wk of age, they begin to develop hemorrhagic ovarian cysts. Follicles in adult ER
KO female mice progress to the graafian stage, but there are no corpora lutea (CL). To test whether ER
is required for ovarian folliculogenesis, ovulation, and CL formation, eCG and hCG were used to ovulate 3- to 5-wk-old ER
KO and wild-type (WT) sibling mice. Gonadotropin administration resulted in ovulation in both ER
KO and WT mice. Gonadotropin-treated ER
KO females that ovulated produced 7.09 ± 0.77 oocytes per mouse, whereas gonadotropin-treated WT female mice had 16.17 ± 0.84 oocytes. Surprisingly, ruptured ER
KO ovarian follicles developed into CL that had normal morphology. Gonadotropin-treated ER
KO mice had 3-fold higher concentrations of serum progesterone than did control ER
KO mice that had been administered saline rather than gonadotropins. Thus, the CL in gonadotropin-treated ER
KO mice appeared to be steroidogenically functional. On the basis of these findings, ovarian folliculogenesis, ovulation, and CL formation can occur in the absence of ER
, although to a lesser extent than in WT mice.
| INTRODUCTION |
|---|
|
|
|---|
Evidence supporting a direct intraovarian effect of estrogen through its cognate receptors has been based on hypophysectomy studies [7], ovarian follicle cultures [1, 2, 4], and administration of either estrogen receptor (ER) antagonists [8, 9] or antiaromatase compounds [10, 11]. Estrogen has been reported to modulate granulosa cell gap junction formation [7], steroidogenesis [1215], FSH and LH receptor expression [16, 17], and ovarian follicular development [18]. It also inhibits granulosa cell apoptosis [19]. In rats [20] and rabbits [21, 22], estrogen seems required for maintenance and function of corpora lutea (CL), even in the absence of gonadotropins.
In contrast, other groups have shown that estrogen does not locally affect ovarian folliculogenesis [5, 6]. For example, addition of the antiestrogenic compound ICI 182,780 and/or antiestrogen antibodies to ovarian follicular cultures of late primary mouse ovarian follicles did not affect the growth and development of the follicles to the preovulatory stage [6].
Estrogen needs to bind to its cognate receptor to exert its effects. Currently, two estrogen receptors, ER
[23] and ERß [24, 25], have been characterized. Estrogen receptor-
knockout (ER
KO) female mice are infertile, as a result of pubertal hemorrhagic ovarian cyst formation [3]. Women who have mutations of the aromatase gene [26] and mice that have targeted disruption of the aromatase gene, cyp19, are infertile, and no CL are present in cyp19-deleted mice [27, 28]. On the basis of naturally occurring human aromatase deficiency cases [26] and targeted gene-disrupted mice [3, 27, 28], it may be postulated that estrogen/ER is required for normal ovarian function. However, disruption of these genes may cause other systemic effects such as elevated serum concentrations of LH [27, 29] that hinder interpretation of the direct effects of estrogen within the ovary. To examine the ovarian function of ER
KO female mice, gonadotropins were used to ovulate prepubescent ER
KO and WT female mice.
| MATERIALS AND METHODS |
|---|
|
|
|---|
ER
KO and wild-type (WT) female mice of a mixed C57BL/6J/129 background were used according to institutional animal care protocols. They were housed at the University of Missouri Animal Sciences Research Center laboratory animal facility and maintained ad libitum on mouse chow formulation 5001 (Purina, St. Louis, MO) and water. They were on a 12D:12L cycle. The genotypes of the mice were determined on the basis of ER
polymerase chain reaction (PCR) analysis [3].
Examination of Ovaries from Adult ER
KO and WT Female Mice
Ovaries from nontreated 6- to 8-wk-old ER
KO (n = 25) and WT (n = 25) mice were fixed in Bouin's fixative (Sigma Chemical Co., St. Louis, MO) and histologically examined, as described below.
Gonadotropin Treatment of Young ER
KO and WT Female Mice
Three- to 5-wk-old ER
KO and WT female mice were given either 5 (Intervet, Cambridge, UK) or 10 IU of eCG (Sigma Chemical Co.) i.p., followed 4854 h later by 5 IU of hCG (Sigma Chemical Co.). Age-matched control WT and ER
KO female mice received 0.9% saline (Sigma Chemical Co.). Mice were anesthetized with CO2 and killed by cervical dislocation. In order to recover ovulated oocytes, ovaries and oviducts from gonadotropin-treated WT (n = 26) and ER
KO (n = 24) female mice 12 h post-hCG were placed in potassium simplex optimized medium (KSOM; Specialty Media, Phillipsburg, NJ) or M2 medium (Sigma Chemical Co.) in the presence of 300 µg/ml of hyaluronidase (Sigma Chemical Co.). The ampullary region of the oviduct was examined under a Nikon SMZ stereomicroscope (Nikon, Melville, NY), and the ovulated oocytes were counted. To further examine CL formation and serum progesterone concentrations in gonadotropin-treated ER
KO and WT mice, sera and ovaries from gonadotropin- and saline-treated WT and ER
KO female mice were collected 48 h post-hCG.
Histology
Ovaries and oviducts from gonadotropin- and saline-treated WT and ER
KO female mice were fixed in either Bouin's fixative (Sigma Chemical Co.) or 4% paraformaldehyde (w:v; Electron Microscopic Sciences, Fort Washington, PA) and embedded in paraffin or glycomethacrylate (Polysciences, Inc., Warrington, PA), respectively. Two- to 4-µm-thick sections were cut and stained with Gill's hematoxylin (Fisher Scientific, St. Louis, MO) and eosin (Fisher Scientific). Ovaries and oviducts were photographed using a Spot 2 digital camera (Diagnostic Instruments, Inc., Sterling Heights, MI), and images were printed with a Fuji Pictography 3000 printer (Fuji, Tokyo, Japan).
Progesterone RIA
Concentrations of progesterone in serum were determined with a Coat-a-Count progesterone kit (Diagnostics Products Corp., Los Angeles, CA), as previously described [30]. To validate the progesterone RIA for mice, serum from ovariectomized WT and ER
KO mice were used as negative controls. Since progesterone has previously been shown to peak in mice at about Day 6 of pregnancy [31], serum from WT 6-day post-coitus mice were used as positive controls. Serum from 3- to 5-wk-old WT and ER
KO female mice three days after either hCG or 0.9% saline were assayed for serum progesterone concentrations. Undiluted and serial dilutions of 1:1, 1:2, and 1:4 of serum were used to determine parallelism. Radioactivity was measured by an LKB Wallac beta counter (Wallac Inc., Gaithersburg, MD).
Statistical Analysis
The number of ER
KO and WT female mice that ovulated in response to gonadotropins was analyzed by chi-square analysis. The numbers of oocytes ovulated in WT and ER
KO mice and serum progesterone concentrations were calculated as the mean ± SEM. The following comparisons were analyzed by Student's t-test: gonadotropin-treated WT (n = 4) versus gonadotropin-treated ER
KO (n = 9) female mice, gonadotropin-treated ER
KO (n = 9) versus saline-treated ER
KO female mice (n = 9) and gonadotropin-treated WT (n = 4) versus saline-treated WT female mice (n = 10).
| RESULTS |
|---|
|
|
|---|
KO and WT Female Mice
Histological analysis confirmed previous results [3] that the ovaries of ER
KO mice were dominated by hemorrhagic cysts (Fig. 1, B and C). These structures were absent in normal adult WT female ovaries (Fig. 1A). However, in contrast to previous results [32], ER
KO female mice developed graafian ovarian follicles (Fig. 1, B and C), although they were abnormal. There was abnormal stratification of granulosa cells, with some areas of the follicle surrounded by multiple layers of cells but other regions having a single layer of squamous-appearing cells (Fig. 1, B and C). As noted earlier [3,32], CL did not form in the ovaries of adult ER
KO female mice (compare Fig. 1, B and C, with Fig. 1A).
|
Gonadotropin Treatment of Young ER
KO and WT Female Mice
Gonadotropin treatment of young 3 to 5-wk-old WT (Fig. 2, A and B; Fig. 3, A and B) and ER
KO (Fig. 2, C and D; Fig. 3, C and D) mice resulted in ovulation of oocytes with expanded cumulus cells into the ampulla of the oviduct. Fewer gonadotropin-treated ER
KO female mice ovulated than gonadotropin-treated WT mice (11 of 24 versus 23 of 26; P < 0.01). In addition, ER
KO mice ovulated fewer oocytes than gonadotropin-treated WT mice (7.09 ± 0.77 versus 16.17 ± 0.84; P < 0.01). As shown in Figures 2C and 3C, the ER
KO female mice were only just beginning to develop hemorrhagic ovarian follicular cysts at this age. The hemorrhagic ovarian cysts could be distinguished from corpora hemorrhagica (Fig. 2A) by their larger size and the presence of nonluteinized granulosa cells surrounding them. CL were present in gonadotropin-treated ER
KO mice (Fig. 3, E and F). Multiple cell types were evident within the CL and some cells contained lipid droplets (Fig. 3F) suggesting that they were steroidogenically functional.
|
|
Progesterone Results
Serum concentrations of progesterone were low in both WT (0.41 ± 0.09 ng/ml) (n = 4) and ER
KO (1.56 ± 0.60) (n = 4) female mice that had been previously ovariectomized (Table 1). Serum concentrations of progesterone in 6-day post-coitus WT mice were, as expected, elevated (24.13 ± 1.19 ng/ml; n = 4), and corresponded to the peak circulating concentrations of progesterone noted previously for pregnant mice [31]. Together, these negative and positive controls support the validity of the assay. Three days after gonadotropin treatments, ER
KO females had statistically significant lower serum progesterone concentrations (P < 0.01) than WT mice (6.41 ± 1.5 ng/ml versus 17.68 ± 2.51 ng/ml). However, gonadotropin-treated ER
KO females had 3-fold higher concentrations (P < 0.01) of serum progesterone than ER
KO age-matched female mice that had received only saline (6.41 ± 1.5 ng/ml versus 1.98 ± 0.59 ng/ml). This rise in progesterone correlates with the appearance of CL in gonadotropin-treated ER
KO females. As expected, gonadotropin-treated WT mice also had higher progesterone concentrations (P < 0.01) than control WT mice (17.68 ± 2.51 ng/ml versus 1.28 ± 0.26 ng/ml). The elevated progesterone concentrations in gonadotropin-treated WT versus ER
KO female mice is consistent with the increased number of CL present in WT compared to ER
KO female mice.
|
| DISCUSSION |
|---|
|
|
|---|
, although to a lesser extent than in WT mice.
As there was a decrease in the percentage of gonadotropin-treated ER
KO female mice that ovulated as well as in the number of oocytes ovulated per mouse, ER
could have an auxiliary ovarian role. Ovarian ER
might, for example, facilitate ovarian follicular development and maturation, so that there are fewer antral follicles induced to undergo ovulation in ER
KO mice. Estrogen is known to promote FSH-induced ovarian follicular growth in other species [1, 4], although it is unclear which ER(s) mediates the effect. Possibly, ERß rather than ER
is involved.
ERß mRNA and its protein have been detected in granulosa cells at various stages of ovarian follicular development in the rat [24, 3339], human [40], and cow [41]. Multiple alternative spliced variants of ERß have been identified within the ovary [42]. Therefore, to determine whether estrogen has a direct role within the ovary, all of the currently described ER and splice variants need to be considered. Furthermore, other novel estrogen receptors may exist in the ovary [4345] and the uterus [46]. Presently, ERß appears to be the predominant ER in the ovary.
Unlike the ER
KO mouse, the ERßKO female mouse is fertile, although there are fewer ovarian follicles and subsequently fewer ovulations [47]. Consequently, there are fewer CL and smaller litter sizes than in WT mice [47]. Gonadotropin-treated ERßKO [47], like ER
KO female mice, also respond subnormally to gonadotropins. In both instances, fewer oocytes are released than in WT counterparts. Therefore, neither known ER is essential, but each may be needed to provide full ovarian function.
One difference between the two mutant mouse strains is that adult ER
KO mice cannot be induced to ovulate [32]. Possibly, the young ER
KO mice can be induced to ovulate because they have not been exposed to prolonged and elevated levels of LH and subsequently have not developed hemorrhagic ovarian cysts. If the LH induction of these cysts could be blocked, adult ER
KO females might be capable of undergoing normal ovulation and CL formation. There seems little doubt that such cysts arise in response to elevated levels of LH [48,;th49], but the basis of the pathology is unclear. There may be weakening of the follicular basement membrane, which ruptures before ovulation, allowing entry of blood into the follicle.
Mice that are unable to synthesize estrogen because of targeted disruption of the P450 aromatase gene are able to progress up through the antral stage of ovarian follicular development, but they do not form CL [27, 28]. This phenotype is consistent with the hypothesis that estrogen is not necessary for ovarian folliculogenesis but that it might be required for ovulation and CL formation. However, deletion of the P450 aromatase gene results in other systemic effects. Testosterone, FSH, and LH are all elevated in these mutant female mice [27] and could thus underlie the failure of these mice to ovulate.
In conclusion, we have shown that contrary to previous data [32] and implied expectations [2022], a proportion of ER
KO mice can be induced to ovulate and develop what appear to be functionally normal CL. To further understand the differences in response to gonadotropin treatment within ER
KO female mice, studies are underway to quantitate endogenous gonadotropins in these mice. Additionally, future studies include quantitating the number of ovarian follicles and CL in gonadotropin-treated ER
KO mice. It remains to be determined whether ovulated ER
KO oocytes can be fertilized and undergo normal development if transferred to recipient WT female mice. The data presented in this paper combined with the ERßKO studies [47] suggest both ER
and ERß are needed for full ovarian function or that alternative mechanisms exist, such as a novel estrogen receptor.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 This work was supported by a USDA National Needs Fellowship to C.S.R. Part of this work was presented at the Serono Ovarian Workshop, Houston, TX, 1998. ![]()
2 Correspondence: Dennis B. Lubahn, University of Missouri, 163 ASRC, 920 East Campus Drive, Columbia, MO 65211. FAX: 573 882 6827; lubahnd{at}missouri.edu ![]()
Accepted: October 14, 1999.
Received: May 17, 1999.
| REFERENCES |
|---|
|
|
|---|
with mutations near the carboxyl terminus on luteal lifespan in sheep. Biol Reprod 1997; 56:214220.[Abstract]
in the rat ovary. Endocrinology 1999; 140:963971.
and ß in rat ovary and uterus. J Mol Endocrinol 1999; 22:3744.[Abstract]This article has been cited by other articles:
![]() |
T. da Silva Faria, F. de Bittencourt Brasil, F. J B Sampaio, and C. da Fonte Ramos Maternal malnutrition during lactation alters the folliculogenesis and gonadotropins and estrogen isoforms ovarian receptors in the offspring at puberty J. Endocrinol., September 1, 2008; 198(3): 625 - 634. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. A. Emmen, J. F. Couse, S. A. Elmore, M. M. Yates, G. E. Kissling, and K. S. Korach In Vitro Growth and Ovulation of Follicles from Ovaries of Estrogen Receptor (ER){alpha} and ER{beta} Null Mice Indicate a Role for ER{beta} in Follicular Maturation Endocrinology, June 1, 2005; 146(6): 2817 - 2826. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Hegele-Hartung, P. Siebel, O. Peters, D. Kosemund, G. Muller, A. Hillisch, A. Walter, J. Kraetzschmar, and K.-H. Fritzemeier Impact of isotype-selective estrogen receptor agonists on ovarian function PNAS, April 6, 2004; 101(14): 5129 - 5134. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. A. EMMEN and K. S. KORACH Developing Animal Models for Analyzing SERM Activity Ann. N.Y. Acad. Sci., December 1, 2001; 949(1): 36 - 43. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. F. Palter, A. B. Tavares, A. Hourvitz, J. D. Veldhuis, and E. Y. Adashi Are Estrogens of Import to Primate/Human Ovarian Folliculogenesis? Endocr. Rev., June 1, 2001; 22(3): 389 - 424. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. F. Roby Alterations in Follicle Development, Steroidogenesis, and Gonadotropin Receptor Binding in a Model of Ovulatory Blockade Endocrinology, June 1, 2001; 142(6): 2328 - 2335. [Abstract] [Full Text] [PDF] |
||||
![]() |
S Dupont, A Krust, A Gansmuller, A Dierich, P Chambon, and M Mark Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes Development, January 10, 2000; 127(19): 4277 - 4291. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |