| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
* Prince Henry’s Institute of Medical Research, Clayton, Victoria;
Department of Biochemistry and Molecular Biology,
Department of Anatomy and Cell Biology, Monash University, Clayton, Victoria; and
The Jean Hailes Foundation, Clayton, Victoria, 3168, Australia
1Correspondence: Prince Henry’s Institute of Medical Research, PO Box 5152, Clayton, Victoria, 3168, Australia. E-mail: kara.britt{at}med.monash.edu.au
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: estrogen deficiency • Sertoli cell • granulosa cell • sex steroids
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. Sry is expressed in the supporting cells of the XY indifferent gonad (1
, 3
, 4)
and directs their differentiation into Sertoli cells as well as inducing Sox9 expression. Sox9 is responsible for the induction of Müllerian inhibiting substance (MIS) secretion, which in turn suppresses the development of the female Müllerian ducts (5
, 6)
. The Leydig cells of the testis differentiate from an interstitial, steroidogenic cell lineage under the influence of the Sertoli cells; once formed, they secrete testosterone (7)
, which directs the development of male secondary sex characteristics. It is thus Sry and its direct downstream targets Sox9 and MIS that mediate sex differentiation in male testicular somatic cells.
In the absence of Sry and under the direction of an XX genotype, the mammalian gonad develops an ovarian phenotype. The supporting somatic cells form pregranulosa cells that attach to female primordial germ cells to form primordial follicles (8)
. The presence of XX primordial germ cells is believed to be essential for differentiation of female somatic cells in the indifferent gonad (8)
. Growing follicles subsequently recruit thecal cells from the surrounding interstitial cells.
The role of estrogen in gonadal differentiation and maintenance of the ovarian phenotype of somatic cells in eutherian mammals has not been clearly established (9
, 10)
. Accounts of natural and experimental differentiation of male somatic cells have been reported in XX gonads under conditions of variable estrogen deficiency. Seminiferous tubule-like structures have been observed in freemartin cattle (11)
, after premature oocyte death (12)
, and in XX gonads of W/Wv mutant mice and busulphan-treated rats (13
, 14)
. They have also been observed in fetal rodent ovaries transplanted into male hosts (15)
and aging rodents (16
, 17)
. The transdifferentiation of ovarian somatic cells to a testicular phenotype in these examples has been ascribed to the absence of viable oocytes (8
, 12)
rather than the involvement of steroids. A role for estrogen in the phenotype of the somatic cell is suggested in mice deficient in both types of functional estrogen receptors, 
and ß (18
, 19)
. In this model, seminiferous tubule-like structures develop in the gonads with concomitant oocyte loss. The inability of estrogen to transduce a genomic signal in these animals, however, precludes replacement studies to investigate the role of estrogen in the development and maintenance of female somatic cells within XX ovaries.
The aromatase knockout (ArKO) mouse, an alternate model of estrogen deficiency that remains estrogen responsive, is produced by a null mutation in the aromatase (Cyp19) gene (20
, 21)
. Preliminary analysis of ArKO ovaries suggested the presence of seminiferous tubule-like structures containing Sertoli cells, as well as interstitial Leydig cells (22)
.
The aims of this study were to 1) identify, structurally and biochemically, the phenotype of the somatic cells in the ArKO ovaries, and 2) test the effects of estrogen replacement, either as phytoestrogen in the diet or by estradiol pellet, on the phenotype of the ArKO ovary. The results provide definitive evidence that estrogens play a critical role in the phenotype of somatic cells in the mammalian ovary.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were maintained under SPF conditions and fed ad libidum either 1) standard pelleted mouse chow containing 14% soy meal (containing 0.146 mg/g isoflavones: genistin (0.084 mg/g)/, genistein (0.006 mg/g), daidzin (0.053 mg/g), and daidzein (0.003 mg/g), 14% lupins, and 60% wheat (soy+) or 2) soy-free mouse chow (Glen Forrest Stockfeeders, Western Australia) with undetectable levels of isoflavones. Diets were corrected for amino acids, vitamins, and minerals per kilogram of feed. Slight differences in the dietary constituents were present, most notably the percentage of crude fiber (5% vs. 3.7% for soy+ and soy-free diets, respectively) and fat (5.2% vs. 6.1% for soy+ and soy-free diets, respectively). Subcutaneous pellets containing either 17ß-estradiol (0.05 mg) or placebo (Innovative Research of America, Sarasota, FL) were administered for 3 wk to 7-wk-old Wt and ArKO mice (23)
. These pellets were estimated to restore serum estradiol to 50–100 pg/ml (peak estrous levels).
Tissue collection, histology, and ultrastructure
Animals were killed at 6 or 16 wk by cervical dislocation. One ovary from each animal (n=17/group) was immersion-fixed in Bouin’s fluid for 24 h, embedded in paraffin, and serially sectioned (3 µm). Every fifth section was stained using a modified Masson’s trichrome. The contralateral ovary was fixed for 12 h in 3% glutaraldehyde with 2% formaldehyde and 0.01% picric acid in 0.05M cacodylate buffer, postfixed in cacodylate buffered 2% osmium tetroxide, stained en bloc with maleate-buffered uranyl acetate, dehydrated, cleared in propylene oxide and embedded in epon-araldite. Semi-thin (1 µm) sections were stained with toluidine blue for light microscopy and ultra-thin sections were stained with lead citrate and uranyl acetate for electron microscopy. Follicle types were defined as described previously (24)
. The ultrastructural characteristics of ovaries of Wt and ArKO mice were analyzed in comparison to testes from Wt males, prepared by the same methods.
Immunohistochemical localization of Espin
Immunohistochemical analysis of Wt testis and Wt and ArKO ovaries for Espin (antibody kindly provided by James Bartles, Northwestern University, Chicago) was performed with slight modifications to previously described experiments (25)
. Testes from Wt littermates were used as positive controls and preimmune serum as negative controls.
Serum gonadotropins
An established RIA (24)
was used to measure LH and FSH levels in serum of ArKO mice maintained on either a soy-free or soy+ diet. The lower limits of detection were 1.05 ng/ml (FSH) and 0.08 ng/ml (LH). The intra-assay % coefficient of variations in the FSH and LH assays (n=3) were 5.3–6.5 and 8.7–11.7%, respectively. The interassay % CVs for LH and FSH assays (n=3) were 1.4 and 12.1%, respectively, calculated using a pool of normal mouse serum.
mRNA analyses in gonads
RNA was extracted from ovary and testis (2 pools of 4 animals each) using a phenol chloroform based method (Ultraspec, Fisher Biotec, Pittsburgh, PA). Sox9 mRNA was analyzed by real-time PCR (Lightcycler, Roche, Nutley, NJ) using in-house generated Sox9 primers. Primers were designed according to the human cDNA (accession no. Z46629) directed against base pairs 1607–1620 and 2006–1683 respectively (Forward 5' ACA GAT CTC CTA CAG CCC CTT CAA 3'; Reverse 5' GCC GGA GTT CTG ATG GTC AGC GTA 3'). The PCR product was 100 bp and within the PQS domain of Sox9. Expression of Sox9 was corrected for levels of expression of the housekeeping gene, 18S ribosomal RNA.
Statistical analyses
Data are presented as mean ± SE. Statistical analysis was performed using Sigmastat statistical software V2.0 (Jandel, Corte Madera, CA). Comparisons within each age group across genotypes and for both soy+ and soy-free diets were performed using ANOVA in conjunction with Tukey’s test. If normality or equal variance failed, a Kruskal-Wallis in conjunction with Dunn’s multiple comparisons test was performed. The Sox9 data were log transformed, then analyzed by one-way ANOVA and Tukey’s post hoc pairwise comparison. Values of P < 0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were similar. A full complement of follicle types from primordial to antral (healthy and atretic), as well as corpora lutea, was observed (Fig. 1
A, B). Healthy granulosa cells with the characteristic low cytoplasmic-to-nuclear ratio (Fig. 1B, C
) were arranged in concentric layers around the oocytes. Granulosa cells in secondary and antral follicles were separated from the surrounding theca cells by a basement membrane.
|
Ultrastructural analysis confirmed the low cytoplasmic-to-nuclear ratio (Fig. 2
A) of the Wt granulosa cells and revealed heterogeneous nuclear chromatin. The interstitial cells were plentiful (Fig. 1B, C
) and contained modest levels of smooth (sER) and rough endoplasmic reticulum (rER), lipid droplets, an irregularly shaped nucleus and heterogeneous chromatin distribution (Fig. 3
A). Mitochondria with lamellar cristae were observed within the interstitial cells (Fig. 3A
).
|
|
In Wt testis from animals on a soy-free diet, Sertoli cells displayed a voluminous cytoplasm with finger-like cytoplasmic projections and an ovoid, often irregularly shaped, basally located nucleus with a prominent (often tripartite) nucleolus (Fig. 2C
). Testicular Leydig cells contained well-developed tubular sER, characteristically arranged into whorls, and mitochondria with tubulovesicular cristae (Fig. 2C
).
ArKO, 6-wk-old, soy-free diet
These ovaries lacked corpora lutea, although follicles of all stages of maturation were observed (Fig. 1D
). The more mature follicles appeared atretic as evidenced by the uneven granulosa cell layers and cellular debris within antral compartments. The medullary (central) region of the ovary was sparsely populated with cells. Apart from fibers of connective tissue, nests of abnormal follicles housing somatic cells of altered morphology were frequently observed (Fig. 1E
). We likened and subsequently proved these somatic cells to be Sertoli cells on the basis of their 1) high cytoplasmic-to-nuclear ratio (Fig. 1F
) with luminally extending, veil-like cytoplasmic extensions, 2) basally located (Fig. 1F
), irregularly shaped nucleus, often exhibiting crypt-like indentations, and 3) homogeneous nuclear chromatin distribution with defined, and frequently tripartite, nucleolus (Fig. 1F
). These clusters of Sertoli-like cells were germ cell deplete and limited by a basement membrane (Fig. 1F
). The interstitial cells surrounding these tubules (Fig. 1E, F
) appear quite dense with the deep pink cytoplasmic staining presumably a result of their altered cytoplasmic constituents.
Ultrastructurally, the ovaries of 6-wk-old ArKO animals maintained on a soy-free diet possessed Sertoli cells identified on the basis of the aforementioned criteria (Fig. 2B
). The Sertoli cells are similar to those observed in Wt testis (Fig. 2C
). Sertoli cell-only structures, enclosed by a basement membrane (Fig. 2E
) complete with ectoplasmic specializations (ES) (Fig. 2F
) were observed. ES are ultrastructural markers of Sertoli cells. They consist of hexagonally packed actin bundles (26)
sandwiched between a layer of rER and the adjacent Sertoli cell membrane (27)
(Fig. 2F
). They are found exclusively between adjacent mature Sertoli cells and between Sertoli cells and spermatids (28)
. The presence of ES in the ArKO ovaries was confirmed by the localization of Espin, an actin bundling protein found within these structures in the testis (see below).
The interstitial cells surrounding tubule-like structures within the ArKO ovary were identified ultrastructurally as Leydig cells on the basis of the tubulovesicular arrangement of the cristae within their ‘steroidogenic’ mitochondria (Fig. 3A vs. B
), the presence of an annular nucleolus within the nucleus and especially the abundance of sER arranged in whorl-like formations (Fig. 3B
). The latter feature defines mouse Leydig cells (Fig. 3C
) and is absent from the interstitial cells of the Wt ovary (Fig. 3C
), as are the tubulovesicular cristae of the mitochondria, which instead contain multilamellar cristae.
ArKO, 16–18 wk, soy-free diet
The ovaries displayed a near-complete block in folliculogenesis at the primordial or primary stage, concurrent with an increased incidence of hemorrhagic cysts (Fig. 1G
). An age-dependent increase in collagen deposition was noted, extending from the interstitial region (Fig. 1G
). Abnormal follicles/seminiferous tubule-type structures containing both granulosa and Sertoli-like cells, or only the latter (Fig. 1H
), were observed with their predominance in the cortical and medullary regions, respectively. Occasionally follicles develop beyond the primordial stage and possess typical granulosa cells (Fig. 1I
). Seminiferous tubule-like structures were present in 
80% of the ovary; interstitial cells of Leydig cell morphology were present, adjacent to well-developed tubules and in especially high numbers toward the medullary region (Fig. 1I
). Also within the medullary region were empty oval (lacunae-like) structures previously observed in heterozygous and ArKO ovaries (24)
, resembling the remnants of atretic growing follicles. In some instances, these remnants possessed internal structures resembling zona pellucidae. Many are surrounded by transformed somatic cells (data not shown). The ultrastructural analysis confirmed the presence of ES between adjacent mature Sertoli cells (Fig. 2D
).
ArKO, 16–18 wk, soy+ diet
In contrast to their soy-free counterparts, the ovaries of these animals displayed several secondary and antral stage follicles, and their interstitium appeared to maintain a modest population of stromal/interstitial cells (Fig. 1J
). The ovaries displayed several antral-sized hemorrhagic cysts in addition to some smaller cysts, all of which were enclosed by one or two layers of granulosa cells (Fig. 1J
). Once more empty oval structures, reminiscent of atretic primary follicles were found in high numbers. The remnants of zona pellucida were often noted within these structures (Fig. 1K
), supporting their derivation from preexisting follicles. In contrast to their soy-free counterparts, the majority of follicles within these ovaries housed somatic cells of typical granulosa cell structure (Fig. 1J, K
). Although some follicles devoid of oocytes presented with cells resembling Sertoli cells (Fig. 1L
), ovaries from soy+ animals did not show extensive Sertoli-filled tubules.
Immunohistochemical localization of Espin
Espin was immunolocalized to the ES complexes between adult-type Sertoli cells in mouse testes (Fig. 4
B) and between Sertoli-like cells within the tubule-type structures of the ArKO ovaries (Fig. 4A
). Espin immunopositive staining was undetectable in granulosa cell-containing normal follicles in ArKO (Fig. 4C
) and Wt ovaries. Espin was detected in the tubular structures of medullary location (containing exclusively Sertoli cells) in ovaries from 6-wk-old ArKO mice maintained on a soy-free diet. At 16 wk of age, staining was detected in tubules located cortically. By contrast, in ArKO mice maintained on a soy + diet, Espin-positive staining was not observed in the ovaries at 6 or 16 wk of age (data not shown).
|
Serum gonadotropins
There were no significant effects of diet (soy+ or soy-free) or age on serum LH and FSH in ArKO mice at 10–12, 21–26, or 52 wk of age (Table 1
).
|
Sox9 expression in Wt and ArKO gonads
The relative levels of Sox9 mRNA (Fig. 5
) in the soy-free ArKO ovary were similar to Wt testis (P>0.05) but significantly elevated compared to Wt ovary (37-fold increase, P<0.05). Sox9 transcript levels were significantly elevated in Wt testis compared to Wt ovary (41-fold increase, P<0.05). The levels of transcripts for the housekeeping 18S were reduced in the ArKO ovaries compared to Wt controls, presumably as a consequence of reduced RNA quality. This phenomenon appeared to be specific to the ArKO ovary because a reduction in housekeeping transcripts was not observed in RNA extracted from the ArKO liver. Electrophoretic separation of RNA revealed a single band of 18S of reduced intensity in samples from ArKO ovaries compared to Wt controls, with no evidence of degraded RNA (data not shown). The identity of the Sox 9 PCR amplicon was confirmed by sequencing (ABI Prism, 377 DNA Sequencer, Applied Biosystems, Foster City, CA) products amplified from Wt and ArKO ovaries.
|
Estradiol treatment
The uterine weights of placebo-treated ArKO mice were 18% of age-matched Wt controls (Table 2
). Estradiol treatment significantly increased the uterine weight of Wt and ArKO mice, the latter of which demonstrated weights comparable to placebo-treated Wt controls (Table 2)
. The ovaries of Wt animals treated with estradiol did not appear grossly different from their placebo-treated controls, with both containing follicles at all stages of maturation and corpora lutea (Fig. 6
A). The typical granulosa cell morphology was noted in all Wt follicles (Fig. 6B
). The ovaries of 10-wk-old ArKO placebo-treated mice possessed few follicles, no corpora lutea (Fig. 6C
), and medullary located Sertoli and Leydig cell clusters (Fig. 6D
). Normal follicular morphology was partly restored in estradiol-treated ArKO mice (Fig. 6E
) as evidenced by an increase in the number of growing (primary, secondary, and antral) follicles and increased cell density in the stromal compartment. A decrease in atretic follicles (pyknotic nuclei and uneven granulosa cells layers) was noted (data not shown). ArKO mice treated with estradiol did not display Sertoli-like cell tubules or Leydig-like cells (Fig. 6F
). The somatic cells within the ovaries of estradiol-treated ArKO mice (Fig. 6E, F
) resembled those in Wt ovaries (Fig. 6A, B
).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 24)
and levels of testosterone typically observed in males (20)
. We have shown that as the animals age, they display abnormal follicles reminiscent of seminiferous tubules containing Sertoli-like cells and Leydig-like cells in the interstitial areas. The occurrence of mixed follicles (Fig. 1H
) suggested the transformation of granulosa to Sertoli cells within the cortically located follicles. Estrogen replacement, given either as dietary phytoestrogens or as a 17ß estradiol pellet, delayed the onset of ovarian degeneration and prevented or reversed the appearance of the male somatic cell phenotype in the ArKO ovary.
Molecular confirmation of the somatic cell transdifferentiation in ovaries of ArKO mice was obtained by measuring Sox9 expression. Sox9 is an HMG box transcription factor with sexually dimorphic expression in the gonads of many eutherian species, including mice (29
, 30)
. Weak Sox9 expression occurs in the morphologically indifferent murine gonad of either sex (up to day 12.5 pc), after which time levels are up-regulated in the differentiating Sertoli cells of the developing testis and down-regulated in the ovary (30
, 31)
. We have shown that consistent with the appearance of Sertoli cells, ovaries of ArKO mice expressed levels of Sox9 mRNA similar to those observed in Wt testis, though significantly higher than in Wt ovaries. This extends recent observations (18)
of increased Sox9 expression in ovaries of estrogen receptor ß knockout mice with similar Sertoli cell-like structures. Sox9 expression has also been associated with Sertoli cell development in ‘oddsex’ mice (32)
and in female intersex chick gonads (33)
. XX mice overexpressing Sox9 contain Sertoli-like cells in their ovaries (34)
. Moreover, the increased collagen deposition observed in the ArKO ovaries (demonstrated by Masson’s trichrome connective tissue stain) correlates with the regulation of the collagen type 1 encoding col1a gene by Sox9 (35)
. Ultrastructural and immunohistochemical analysis likewise confirmed the identity of the Sertoli cells in ArKO ovaries. The development of Sertoli cell-specific intercellular junctions (ES junctions) was greater in the medullary Sertoli cell-only tubules than in the cortically located mixed granulosa/Sertoli cell follicles, suggesting the former nests housed more mature Sertoli cells. The presence of ES within the latter would, however, support the development of these Sertoli cells from preexisting granulosa cells.
The role of estrogen in the differentiation and maintenance of ovarian somatic cells of eutherians has not been unequivocally established before (9
, 10
, 36)
. Our findings do, however, parallel studies performed within lower vertebrates and pigs. Estrogen has been shown to override sex determination in chickens (33
, 37)
, pigs (38)
, and turtles (39)
, which exhibit alternate mechanisms of sex determination that are dependent on the external biological environment, such as temperature dependent switches, which have been shown to be mediated via aromatase activity (40
, 41)
.
There have been several examples, both natural and experimental, suggesting that estrogens might be involved in the ovarian phenotype. Recent investigations by Coveney and colleagues (42)
provide additional support for the critical role of estrogen in sexual differentiation and development in marsupials. Estradiol benzoate inhibits testicular development in the tammar wallaby when administered orally to XY male pouch young from the day of birth to day 25 postpartum. The testes of treated wallabies display partial or complete transformation into ovaries complete with meiotic germ cells. In eutherians, there are rodent models of parabiosis (43)
, testis transplantation (44
45
46
47)
and sex steroid administration (12
, 48)
where somatic cells of the ovary undergo testicular differentiation. This phenomenon occurs in nature with female freemartin cattle (49
, 50)
, aging female mice (17)
and postmenopausal women (51)
. Recently, transformation from granulosa to Sertoli cells was described in the ovarian follicles of compound estrogen receptor ß knockout mice, which lack the capacity to respond to estrogen (18
, 19)
; however, the phenotype of the interstitial cells was not described. The ovaries of ArKO mice deficient in estrogen, on the other hand, display aberrant granulosa cell differentiation (Sertoli cell development) and interstitial cells exhibiting a phenotype characteristic of Leydig cells. We have extended previous findings by demonstrating that the masculinization is associated with molecular (Sox9) and protein (espin) markers of Sertoli cell differentiation.
Estradiol replacement in 7-wk-old ArKO female mice for 3 wk prevented the onset of the male phenotype of somatic cells. Toda et al. (21)
reported that ArKO mice treated with estrogen from 4 wk of age reversed the accumulation of sER in luteinized interstitial cells. However, these cells were not definitively identified and the authors did not comment on the appearance of Sertoli-like cells in the ArKO ovary. We have shown that these luteinized interstitial cells are ultrastructurally similar to Leydig cells and that estrogen replacement delayed their appearance, in addition to suppression/reversal of the appearance of the tubular structures containing Sertoli-like cells.
We have provided further evidence for a role for estrogen in murine gonadal development by demonstrating that phytoestrogens present in the soy of mouse chow act as estrogen agonists in terms of ovarian morphology, delaying ovarian degeneration and somatic cell reversal in the ArKO mouse. In the absence of soy, the somatic cells of the ArKO ovary are transformed to their male counterparts as early as 6 wk of age. When soy was present in the diet, ArKO mice did not possess transformed somatic cells before 18 wks of age. These effects were not mediated through serum gonadotropins, as levels of FSH and LH were comparable in ArKO mice maintained on either diet. The majority of biological effects of estrogens are mediated via the classical nuclear estradiol receptors and ß, which have been localized to the ovary and testis (52
, 53)
. Phytoestrogens such as genistein, present in soy meal, can act as agonists via these receptors, in particular through estrogen receptor ß, which is highly expressed in the granulosa cells of the ovary (52)
. We have preliminary data to suggest that estrogen receptor ß is present in the ArKO ovary, albeit at reduced levels, whereas estrogen receptor is undetectable (54)
.
All reported cases of XX sex reversal are preceded by the loss of germ cells (8
, 55)
, indicating that germ cells in the ovary prevent the differentiation of Sertoli cells from the supporting cell lineage. There were no germ cells observed in any of the tubule structures containing Sertoli-like cells in sections of the ArKO ovaries. Estrogen replacement prevented or reversed the somatic cell phenotype in this ArKO model, suggesting that the effects of estrogen on somatic cells in the mouse ovary may be independent of the presence of germ cells. This raises the intriguing hypothesis that the influence of female germ cells on ovarian somatic cells after follicle formation might be mediated in part by intrafollicular estradiol.
Wingless transcription factor 4 (Wnt4) and fibroblast growth factor 9 (Fgf9) deficient mice exhibit either partial or complete sex reversal (56
, 57)
. Wnt 4 is a member of the Wnt family of intercellular growth and differentiation factors, which regulate several key developmental steps. Deficiency of Wnt4 leads to partial female to male sex reversal and a marked reduction in oocyte number (10% of Wt), the latter possibly mediating the former. These mice possess cells that express 17-hydroxylase/C17–20 lyase and 3ß-hydroxysteroid dehydrogenase, markers of Leydig cells. Within this model (Wnt4-/-) Sox9, MIS, and Dhh (Sertoli cell markers) are present only after birth, and so this is not an example of primary sex reversal. (57)
XY Fgf9-/- gonads (E18.5) possessed predominantly female histology with occasional testicular cords and reduced mesenchyme (56)
. These mice presented with oviducts and fused uteri. Of the cords present in sex-reversed XY Fgf 9-/-, impaired Sertoli cell differentiation and gene and protein expression were evident. The lack or incomplete maturation of these structures in these mice reflects the importance of Fgf9 in testicular development.
Compared with these models, ArKO XX mice exhibit a hybrid gonad in terms of somatic cells, possessing Sertoli (with defined nuclear, cytoplasmic and intercellular connections) and Leydig cells as well as basal lamina. We see a change in Sox9 expression and are currently conducting experiments to determine whether any alterations in Sox9 occur during embryonic and postnatal development. ArKO mice do not possess alternate reproductive duct development, which would indicate that the underlying mechanisms of this phenotype arose before secondary sexual differentiation. Furthermore, in line with studies in Wnt4-/-, we have not seen aberrant Sertoli cell development in XX gonads before birth, indicating that this is not a case of primary sex reversal.
In summary, our study demonstrates that estrogen insufficiency results in transdifferentiation of somatic cells in the ArKO ovary to cells characteristic of a male phenotype. Estrogen replacement, either in the form of dietary phytoestrogens or as an estradiol pellet, can prevent this transdifferentiation. Our data constitute definitive evidence for a role for estrogen in maintaining female somatic interstitial and granulosa cells in the eutherian ovary.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 19, 2002.
Revision received May 21, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M Seneda, M. Godmann, B. D Murphy, S. Kimmins, and V. Bordignon Developmental regulation of histone H3 methylation at lysine 4 in the porcine ovary Reproduction, June 1, 2008; 135(6): 829 - 838. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Drummond, M. Tellbach, M. Dyson, and J. K. Findlay Fibroblast Growth Factor-9, a Local Regulator of Ovarian Function Endocrinology, August 1, 2007; 148(8): 3711 - 3721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Wang and S. K. Roy Development of Primordial Follicles in the Hamster: Role of Estradiol-17{beta} Endocrinology, April 1, 2007; 148(4): 1707 - 1716. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Lin, O. Ercan, J. Raza, C. P. Burren, S. M. Creighton, R. J. Auchus, M. T. Dattani, and J. C. Achermann Variable Phenotypes Associated with Aromatase (CYP19) Insufficiency in Humans J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 982 - 990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Poutanen Toward understanding the endocrine regulation of gonadal somatic cells. Endocrinology, August 1, 2006; 147(8): 3662 - 3665. [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Guigon and S. Magre Contribution of Germ Cells to the Differentiation and Maturation of the Ovary: Insights from Models of Germ Cell Depletion Biol Reprod, March 1, 2006; 74(3): 450 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. F. Sneddon, N. Walther, and P. T. K. Saunders Expression of Androgen and Estrogen Receptors in Sertoli Cells: Studies Using the Mouse SK11 Cell Line Endocrinology, December 1, 2005; 146(12): 5304 - 5312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Guigon, N. Coudouel, S. Mazaud-Guittot, M. G. Forest, and S. Magre Follicular Cells Acquire Sertoli Cell Characteristics after Oocyte Loss Endocrinology, July 1, 2005; 146(7): 2992 - 3004. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Britt, P. K. Saunders, S. J. McPherson, M. L. Misso, E. R. Simpson, and J. K. Findlay Estrogen Actions on Follicle Formation and Early Follicle Development Biol Reprod, November 1, 2004; 71(5): 1712 - 1723. [Abstract] [Full Text] [PDF]
|
|
|
