| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Reproductive Medicine (B.C., K.H., J.C., G.F.E.), University of California, San Diego, La Jolla, California 92093; and Universidade Federal de São Paulo (G.A.R.M.), São Paulo, Brazil 01311-940
Address all correspondence and requests for reprints to: Gregory F. Erickson, Ph.D., Department of Reproductive Medicine, University of California, San Diego, La Jolla, California 92093-0633. E-mail: gerickson{at}ucsd.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Design: P450c17 protein was visualized in sections of control and anencephalic ovaries using immunohistochemistry.
Subjects: Subjects included control (nonanencephalic) and anencephalic human fetal ovaries during the second and third trimesters.
Results: In second-trimester control ovaries, P450c17 was highly expressed in primary interstitial cells (PIC) located between the ovigerous cords near the cortical-medullary border where meiosis and primordial follicle formation were occurring. Morphometric analysis revealed a progressive decrease in the number of PIC during the second trimester, suggesting that PIC might have a finite lifetime. Between 25 and 32 wk, relatively few cells stained positive for P450c17; however, after 33 wk, P450c17 was strongly expressed in theca interstitial cells (TIC) bordering developing follicles. Surprisingly, the TIC appeared remarkably early during folliculogenesis, e.g. as early as the primary-to-secondary transition, and exhibited notable hyperplasia throughout preantral and early antral follicle growth. Owing to large numbers of developing preantral follicles, the third trimester was characterized by an increased abundance of P450c17-positive TIC. During this time period, P450c17 was strongly expressed in the hilus interstitial cells juxtaposed to the rete ovarii. Studies of ovaries of anencephalic fetuses revealed a similar spatiotemporal pattern of P450c17 expression in the PIC, TIC, and hilus interstitial cells, consistent with the possibility that pituitary hormones may not be involved in P450c17 expression in fetal ovaries.
Conclusion: We identified three different classes of P450c17-expressing interstitial cells in the human fetal ovary, each having a different spatiotemporal pattern of P450c17 expression and, presumably, a different set of physiological functions.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
hydroxylase-17,20-lyase) is a major steroidogenic enzyme that catalyzes the conversion of pregnenolone and progesterone to their corresponding androgen products, dehydroepiandrosterone (DHEA) and androstenedione, respectively (1). The enzyme is a single protein that possesses both the hydroxylase and lyase activities. Recent data indicate that most androstenedione is produced from DHEA, not by the activity of P450c17 on 17-hydroxyprogesterone (2). In adult human ovaries, P450c17 expression is highly expressed in the theca interstitial, theca lutein, and hilar cells (3, 4). Pituitary LH is the primary physiological stimulus for P450c17 expression (5), but other factors, such as insulin, appear to be important regulators as well (6, 7). The regulated expression of P450c17 in ovarian interstitial cells is physiologically important because the androgens generated provide substrate for P450 aromatase and ultimately estrogen production (8). The clinical significance of ovary P450c17 is emphasized by the fact that overexpression of interstitial androgen production is a major component in the pathophysiology of polycystic ovary syndrome (PCOS) (9, 10). In this context, questions concerning the expression of P450c17 in female fetuses are of interest because of evidence linking fetal hyperandrogenism to the programming of PCOS in adult primates, including women (11). Attempts to analyze the androgenic potential of human fetal ovaries go back to a classical experiment by Bloch (12) in 1964. He showed that in contrast to the fetal testes, androgens are not produced when homogenates of fetal ovaries between 9 and 19 wk of gestation are incubated with exogenous progesterone. Ten years later, however, Payne and Jaffe (13) demonstrated that homogenates of ovaries at 12–17 wk gestation have the capacity to convert pregnenolone sulfate to DHEA and androstenedione. Although only three fetuses were studied, it was concluded that human fetal ovaries express P450c17 during the second trimester. Subsequent studies have provided additional evidence to support this conclusion. First is the work by Wilson and Jawad (14), showing that fetal ovaries at 12, 20, and 22 wk gestation secrete DHEA and androstenedione spontaneously during organ culture in serum-free medium. And second are the molecular biology experiments demonstrating that human fetal ovaries between 14.9 and 21.5 wk express low, but detectable, amounts of P450c17 mRNA (15, 16).
Currently, we know nothing about the cellular sites of P450c17 expression in the human fetal ovary during development. In this study, we have characterized the spatiotemporal pattern of expression of P450c17 protein in ovaries of control fetuses during the second and third trimesters of pregnancy and determined whether the pattern of P450c17 expression is modified in anencephaly.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Specimens were obtained from 36 human fetuses at 17–42 wk of gestational age from the Department of Pathology at University of California, San Diego (UCSD), Medical Center and from Children’s Hospital. Both ovaries from most fetuses were available for study. The estimated gestational age was determined by the date of the last menstrual period and confirmed by direct measurement of foot length. All fetuses studied had a normal karyotype. The clinical data of the fetuses are presented in Table 1
. In this study, we analyzed a total of 48 ovaries from control fetuses and 18 ovaries from anencephalic fetuses. The term control indicates fetuses that did not have anencephaly. We cannot assume the controls had normal physiology, only that they were not anencephalic. Pathology reports from all of the anencephalic fetuses indicated hypoplastic or atrophied adrenals, which is consistent with the fact that anencephaly leads to hypoplasia of the anterior pituitary (17). We have no direct evidence that our anencephalic fetuses were not producing other pituitary hormones including some LH and/or FSH; however, a previous publication has shown that pituitary gonadotropes are almost absent in anencephalic fetuses during development (18). Ovary samples were classified as follows: second trimester (14–26 wk; n = 20), and third trimester (27–42 wk; n = 16). Normal ovaries from regular ovulatory women were obtained from the National Institute of Child Health and Human Development Reproductive Tissue Sample and Repository. The UCSD Human Research Protection Program Institutional Review Board approved this study.
|
The ovaries were fixed in either Bouin’s solution or 10% neutral buffered formalin. They were embedded in paraffin, sectioned randomly at 10 µm, and stained with hematoxylin and eosin. Sections were full slices of the ovaries, thereby profiling the cortex, medulla, and hilum. Follicles were classified as primordial, unilaminar nongrowing follicles with an oocyte arrested in the dictyate stage of meiosis and a single layer of flattened or squamous granulosa cells; transitional primary, growing follicles with a dictyate oocyte enveloped by a single layer of mixed flattened and cuboidal granulose cells; classic primary, growing follicles with an oocyte surrounded by a single layer of cuboidal granulosa cells; secondary, preantral follicles with an oocyte surrounded by two to eight layers of granulosa cells; and Graafian follicles containing a fluid-filled antrum. The interstitial cells were classified as primary interstitial cells (PIC), large eosinophilic glandular cells in the interstitial tissue between the ovigerous cords; theca interstitial cells (TIC), eosinophilic cells juxtaposed to the basal lamina of growing follicles; and hilus interstitial cells (HIC), eosinophilic glandular cells juxtaposed to the rete ovarii.
Antibody generation and immunohistochemistry
Polyclonal antibody was raised in rabbits against recombinant full-length human cytochrome P450c17 protein expressed in Escherichia coli as described previously (19, 20). P450c17 protein was examined in 10-µm sections of tissue essentially as described previously (21), using a streptavidin-horseradish peroxidase kit according the manufacturer’s instructions (Dako, Carpinteria, CA; no. K0673). Sections were mounted onto poly-L-lysine-coated slides, deparaffinized in xylene, hydrated, and digested with proteinase K (5 µg/ml for 20 min at 37 C). After washing (3 min), sections were incubated with 3% H202 for 5 min at room temperature. Sections were washed for 3 min in Tris-buffered saline and then incubated (30–120 min at room temperature) with the primary antibody diluted 1:2000 with 1% normal goat serum. After washing, the slides were incubated (10 min at 23 C) with biotin-labeled affinity isolated goat antirabbit Ig in PBS containing stabilizing protein and 0.015 mol/liter sodium azide. After washing, the P450c17 antibody complexes were visualized using streptavidin-conjugated horseradish peroxidase in PBS containing stabilizing protein and antimicrobial agents. Reaction products were developed during 2–3 min incubation in a 3,3'-diaminobenzene tetrahydrochloride (DAB Substrate Chromogen Solution, Dako). Sections were counterstained with hematoxylin. The negative control sections were incubated with preimmune serum in the absence of primary antibody.
Cell counts
To compare the abundance of PIC present at the different stages of fetal development, a cell count was performed. The number of immunoreactive PIC was counted in full sections of control (n = 13) and anencephalic (n = 4) subjects during the second trimester (14–26 wk gestation). For each subject examined, we scored a total of six random sections from each ovary. The cell counts for each subject were normalized to area and expressed as means per square millimeter. The areas of the ovary sections were measured using Image J and the area determined as pixels per square millimeter. To obtain a representative density of PIC, the total area of the sections scored was divided by the total number of cells scored in all six sections.
Statistics
The statistical analysis was performed using Student’s t test. A P value < 0.05 was considered to be significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We first validated the anti-P450c17 antibody in human ovary tissue, using sections of ovaries of normal cycling women. In these ovaries, immunoreactive P450c17 was detected in the TIC (Fig. 1A), theca lutein (Fig. 1B), and HIC (Fig. 1C). No staining was observed in any other cell type (Fig. 1). In control experiments, no staining was observed when sections were incubated with preimmune serum (data not shown). Regarding cross-reactions, it is well known that granulosa lutein cells strongly express P450c21 and P450 aromatase. The absence of immunoreactive signal above background in the granulosa lutein cells (Fig. 1B) demonstrates that the antihuman P450c17 antibody does not cross-react with P450c21 and P450 aromatase.
|
To determine whether P450c17 is expressed in human fetal ovaries during the third trimester, morphological and immunohistochemistry studies were performed with control (nonanencephalic) ovaries.
Morphologically, the third-trimester ovaries had a thick cortex that surrounded the medulla and hilum (Fig. 2A). A prominent feature of these ovaries was the presence of large numbers of primordial follicles in the cortex. Between these follicles were blood vessels. Sometimes ovigerous cords were seen in the outer areas of the cortex near the surface epithelium. Growing follicles were evident in all third-trimester ovaries, located along the cortical-medullary border (CMB) (see for example Fig. 2A). Between 27 and 32 wk, the vast majority of growing follicles were transitional and classic primary follicles, although a few small secondary follicles were occasionally seen. Beginning at about 33 wk, typical small Graafian follicles (
0. 5 mm in diameter) were present (Fig. 2A). These follicles contained a fully grown oocyte surrounded by multiple layers of granulosa cells and a very prominent theca interna composed of numerous eosinophilic TIC and large blood vessels. At term, the largest Graafian follicles measured 1.0 mm in diameter. The medulla was composed of loose connective tissues and large blood vessels. In the hilar region were cords of dark-staining cells, the rete ovarii (Fig. 2A). The rete ovarii themselves sometimes contained HIC, but in most cases, these cells were confined to the periphery, appearing as clusters of large eosinophilic glandular cells.
|
). A rather remarkable feature of the early Graafian follicles was the presence of an unusually well-developed theca interna containing large numbers of intensely stained TIC and unusually large blood vessels (Fig. 2C). It is interesting that small clusters of strongly positive interstitial cells, presumably TIC, were present at the periphery of small preantral follicles, being evident as early as the primary-to-secondary transition stage (Fig. 2D). As growing follicles increased in size, the number of P450c17-positive cells in the theca compartment increased steadily, such that when a secondary follicle had attained full size, three to five layers of intensely stained TIC surrounded it (Fig. 2E). After 34 wk, clusters of P450c17-positive HIC were seen in and around the rete ovarii (Fig. 2F). Positive-staining hilus cells were rarely seen in ovaries before 33 wk. Second-trimester ovaries: morphology and P450c17 immunostaining
We next determined the morphology and spatiotemporal pattern of P450c17 protein expression in control (nonanencephalic) ovaries during the second trimester, e.g. between 14 and 26 wk. During this period, the cortex was the major component of the fetal ovary (Fig. 3A). It was composed primarily of irregular ovigerous cords that extended from the surface epithelium to the CMB (Fig. 3A). The cords were limited by a basement membrane and were densely packed with germ cells (oogonia and oocytes) and pregranulosa cells (Fig. 3B). The bulk of the mitotic oogonia appeared at the outer surface of the cortex (zone 1), whereas the bulk of oocytes at various stages of meiosis (leptotene, zygotene, pachytene, and diplotene) were in the deeper regions (zone 2) of the cortex. Apoptotic germ cells were found throughout the cords. At 19 wk, a few primordial and transitional primary follicles were present in small groups in the deepest regions of zone 2 (Fig. 3B). By the end of the second trimester, primordial and primary follicles were very abundant, and occasionally, a rare early secondary follicle with two layers of granulosa cells was evident. Between the ovigerous cords was a loose network of interstitial tissue composed of fibroblast-like cells and abundant blood vessels (Fig. 3B). Scattered in the interstitial tissue of zone 2 were the PIC (Fig. 3C). They were unusually large cells with a highly eosinophilic cytoplasm and a large spherical nucleus that had abundant euchromatin and a prominent nucleolus (Fig. 3C). PIC were frequently in close association with capillaries (Fig. 3C). In the hilum, rete ovarii were present, but few if any glandular HIC were apparent in the second-trimester ovaries.
|
, D–F). Toward the end of the second trimester, P450c17-positive cells were present along the CMB where primordial follicle formation and recruitment were occurring (Fig. 3G). P450c17-stained cells were rarely found in the cortex after 23 wk. The shape of the stained PIC varied considerably. As a rule, they appeared in small clusters of tightly packed polyhedral cells (Fig. 3H). However, they also appeared singly as fusiform cells with long and tapering processes; such isolated PIC were often intimately associated with the basement membrane of the ovigerous cords (Fig. 3I). No P450c17-staining cells were detected in the hilum of the second-trimester ovaries. P450c17 expression in ovaries of anencephalic fetuses
Given that primary Leydig cells fail to develop in anencephaly (17, 22), we asked whether anencephaly was associated with a modification in the expression of P450c17 protein in fetal ovaries. To investigate this possibility, we analyzed ovaries of anencephalic fetuses between 17 and 42 wk.
In agreement with previous work (17, 22), anencephalic ovaries appeared to develop normally up to the early third trimester; however, after that, growing preantral and antral follicles, which are a characteristic feature of normal ovaries, were rare or absent in anencephalic ovaries. As shown in Fig. 4A, the second-trimester anencephalic ovaries contained strongly positive PIC. These cells were present in the interstitial tissue between the ovigerous cords (Fig. 4A), where they appeared singly or in clusters of a few cells, often in intimate association with capillaries and the basement membrane of the ovigerous cords (Fig. 4B).
|
), P450c17 expression in the TIC and HIC appeared typical. That is, clusters of stained cells were found juxtaposed to primordial and growing primary follicles (Fig. 4C); a well-developed theca interna containing abundant P450c17-positive TIC was evident in secondary follicles (Fig. 4D); multiple layers of strongly positive TIC were present in the theca interna of small Graafian follicles (Fig. 4E); and immunoreactive HIC were present in and around the rete ovarii of third-trimester anencephalic ovaries (Fig. 4F). Number of primary interstitial cells
To better confirm the temporal pattern of PIC, we performed a morphometric analysis of P450c17-stained cells in control and anencephalic ovaries during the second trimester. As shown in Fig. 5, control ovaries showed a time-dependent decrease in the number of immunoreactive PIC, decreasing from peak levels at 19 wk and becoming very low or undetectable at 24 wk. By comparison, no statistically significant change was observed in the number of stained PIC in the anencephalic ovaries.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The finding that PIC express P450c17 demonstrates that PIC are differentiated androgenic cells. This conclusion is reinforced by previous reports. First, Payne and Jaffe (13) demonstrated that homogenates of fetal ovaries between 12 and 18 wk gestation have the capacity to convert pregnenolone sulfate to DHEA and androstenedione but not testosterone. Notably, the level of P450c17 activity in the fetal ovary was comparable to that of human fetal testes. Second, Wilson and Jawad (14) showed that fetal ovaries at 12, 20, and 22 wk secrete steroid hormones (DHEA > androstenedione > progesterone > estrone > estradiol > testosterone) de novo in serum-free medium. Third, PIC exhibit the fine structure typical of active steroid-secreting cells (23, 24, 25, 26). And fourth, Voutilainen and Miller (16) reported that human fetal ovaries at 15–22 wk have low, but detectable, amounts of mRNA for P450c17 and another steroidogenic enzyme, P450ssc.
Our morphometric analysis indicates large numbers of P450c17-positive PIC at 19 wk with their number decreasing steadily to essentially zero at 24 wk. Consistent with this finding is a previous structural study indicating that PIC first appear at 12 wk, become very numerous from wk 15–18, and then are difficult to find after 21 wk (23). The view emerging from these results is that P450c17-expressing PIC appear to be produced with a finite lifetime, being at their greatest concentration between 15 and 19 wk gestation. An interesting question is what ultimately happens to the PIC? Do they dedifferentiate and redifferentiate into some other cell type such as a TIC and/or HIC, or do they only die? Additional work will be needed to investigate these possibilities. It should be noted that our results contrast with a recent immunohistochemistry study that failed to detect P450c17 in human fetal ovaries (3). This difference is perhaps because Dharia studied ovaries at midgestation when P450c17-expressing interstitial cells are relatively scarce.
Regarding the regulation of P450c17 expression in the PIC, there is evidence that dibutyryl cAMP, but not human chorionic gonadotropin or LH plus FSH, can stimulate androgen production by second-trimester fetal ovaries (14). In keeping with these findings, LH/human chorionic gonadotropin receptors are not detectable in second-trimester ovaries (27), and we found a near-normal spatiotemporal pattern of P450c17 expression in PIC of anencephalic fetuses. Thus, the regulated expression of P450c17 in PIC would appear to be a result of a gonadotropin-independent cAMP-signaling pathway. It is interesting to note that the anterior pituitary is obligatory for the formation of primary Leydig cells in the fetal testes during the second trimester (17). It seems, therefore, that the development of ovarian PIC and testicular primary Leydig cells is regulated by different mechanisms. We have reported that insulin is a potent stimulator of androgen production by postnatal ovary interstitial cells (7). Consistent with a possible role of insulin in regulating fetal ovary androgen production is the evidence that: 1) insulin receptors are expressed in human fetal ovaries (28), and 2) a functional link between insulin and fetal ovary androgen production is suggested by the development of enlarged polycystic ovaries, hirsutism, and clitoromegaly in newborn girls with severe insulin resistance and hyperinsulinemia (29, 30, 31, 32).
More experiments will be needed to determine whether the PIC produce androgens in utero. Nonetheless, it is worthy to consider structure/function relationships. First, the PIC are restricted to zone 2 of the cortex where granulosa development, meiosis, and primordial follicle formation are occurring. Thus, it is reasonable to hypothesize that PIC androgens may play an autocrine/paracrine role in the morphogenesis of the ovigerous cords, perhaps acting directly at the level of the fetal ovary androgen receptor (33) and/or indirectly via their aromatization to estrogen (34). The present findings and those of others (23, 24, 25, 26) demonstrate that clusters of PIC and accompanying blood vessels form glandular-like units similar to those found in the human testis (26). This structure led Nottola et al. (26) to propose that the clusters of PIC are endocrine units whose organization could facilitate the exchange of substances, such as hormones, between blood and the PIC.
We found that the third trimester is characterized by the increasing accumulation of differentiated TIC and HIC. With respect to the TIC, there are several aspects to consider. One is our observation that P450c17-positive TIC develop remarkably early during preantral follicle development, e.g. at or around the primary-to-secondary transition stage. Thus, the onset of secondary follicle development seems to be a major event leading to the development of differentiated TIC in the fetal ovary. Whether these differentiated TIC originate from a precursor stem cell in the surrounding stroma or are recruited from a pool of P450c17-positive cells in the vicinity of the early secondary follicles remains unknown. Another striking aspect is that fetal preantral folliculogenesis is associated with a great increase in the number of P450c17-positive TIC. This is surprising because in postnatal ovaries, P450c17-positive TIC are not found until the preantral follicle undergoes antrum formation (3, 4). Therefore, the premature and overexpression of P450c17-positive TIC during the very early phases of follicle development is a most unusual feature of the human fetal ovary during the third trimester.
Questions concerning the regulation of fetal TIC formation are of interest because these cells, as a group, have the potential of being a significant source of fetal androgens. Data available supporting the potential androgen activity include the present finding of P450c17 expression and a report that fetal monkey ovaries containing small Graafian follicles synthesize large quantities of DHEA and androstenedione (35). We know from our study that fetal TIC express P450c17 during the very early stages of preantral folliculogenesis. According to dogma, this is the gonadotropin-independent period of folliculogenesis. But, the situation could be more complicated. Gulyas and co-workers (36) have shown that fetal hypophysectomy results in the cessation of folliculogenesis at the classic primary stage in monkey fetal ovaries. This is relevant because we have confirmed the finding (17, 22) that few if any recruited primordial follicles grow beyond the classic primary stage in ovaries of anencephalic fetuses. Future studies should build on these findings by exploring experimentally the functional consequences of the pituitary gonadotropins in the formation of TIC in the human fetal ovary as it relates to the initiation of the first wave of folliculogenesis.
A role for androgens in promoting primary and secondary follicle development through mechanisms involving androgen-receptor-mediated increases in mitosis and FSH receptor expression in granulosa cells has been presented (37, 38, 39). And the greatest growth activity of primordial, primary, and secondary follicles occurs during fetal/postnatal life (40). It will be of interest to investigate a possible connection between the precocious expression of P450c17 in the TIC and the high-growth activity of the pool of preantral follicles. We have shown here that the theca interna of developing preantral follicles is unusually rich in differentiated TIC and blood vessels. This structure/function relationship is consistent with the hypothesis that fetal TIC might secrete androgens that could evoke differentiated responses in target tissues, such as the fetal hypothalamus and pituitary, via endocrine mechanisms.
A third class of P450c17-expressing cells, the HIC, was identified in the fetal ovary during the third trimester. This is in good agreement with previous histological studies indicating that HIC are sparse before the third trimester but increase significantly in number between 33 and 42 wk (41). Our finding that P450c17 expression is definable in HIC in anencephaly is consistent with a previous structural study (42). In adult women, HIC are extraglandular testicular Leydig cells involved in the production of testosterone (4, 43, 44, 45, 46). It remains to be determined whether fetal HIC reflect a similar function.
Finally, there is interest in the concept that prenatal androgen excess may be causally connected to the development of PCOS in adult women (47). Given the data presented here, it will be important to determine whether this clinical concept in any way relates to the spatiotemporal overexpression of P450c17 in the human fetal ovary. In this context, it should be noted that the embryonic lethality of P450c17 knockout mice (48), in contrast to the viability of P450ssc knockouts (49), suggests that P450c17 protein may have other functions in morphogenesis besides 17-hydroxylase and 17,20-lyase activities. If this concept is true for humans, then it is possible that novel activities might exist for P450c17 in fetal interstitial cells during ovary development.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure statement: The authors have nothing to disclose.
First Published Online July 5, 2006
Abbreviations: CMB, Cortical-medullary border; DHEA, dehydroepiandrosterone; HIC, hilus interstitial cells; PCOS, polycystic ovary syndrome; PIC, primary interstitial cells; TIC, theca interstitial cells.
Received March 23, 2006.
Accepted June 23, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/17,20-lyase), and P450c21 (21-hydroxylase) in the human fetus. J Clin Endocrinol Metab 63:1145–1150This article has been cited by other articles:
 |
|
 |
|
  S. Franks Do Animal Models of Polycystic Ovary Syndrome Help to Understand Its Pathogenesis and Management? Yes, but Their Limitations Should be Recognized Endocrinology, September 1, 2009; 150(9): 3983 - 3985. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Homburg Androgen circle of polycystic ovary syndrome Hum. Reprod., July 1, 2009; 24(7): 1548 - 1555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. H Abbott, D. K Barnett, J. E Levine, V. Padmanabhan, D. A Dumesic, S. Jacoris, and A. F Tarantal Endocrine Antecedents of Polycystic Ovary Syndrome in Fetal and Infant Prenatally Androgenized Female Rhesus Monkeys Biol Reprod, July 1, 2008; 79(1): 154 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Rosenfield Identifying Children at Risk for Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, Control (nonanencephalic); 
, anencephalic.