| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Department of Molecular and Integrative Physiology and Ralph L. Smith Research Center, University of Kansas Medical Center, Kansas City, Kansas 66160-7401
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the hamster, the smallest preantral follicles having 1–4 layers of granulosa cells (GC) and lacking theca cells secrete progesterone in vitro in response to FSH but not LH [3]. After long-term culture of rat GCs [4] and GCs of other species [5], they also produce progesterone after incubation with FSH. One of our objectives was therefore to determine whether intact small preantral rat follicles, in short-term incubations, are more responsive to FSH than to LH.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Immature female Sprague-Dawley rats were obtained at 23 days of age from Sasco, Inc. (Omaha, NE) and housed under controlled conditions of temperature, humidity, and lighting (14L:10D; lights-on at 0500 h). We relied on the age of the rats rather than weight because some animals weighing between 40 and 50 g had already ovulated whereas no rats 25 days old or younger had corpora lutea.
Experimental Design
Prepubertal rats were killed at 0900 h on Day 25 as a control group; two other groups were injected s.c. with 5 IU eCG (Sigma Chemical Co., St. Louis, MO) on Days 23 and 24 and killed 48 or 24 h later, respectively (on Day 25). Ovaries were removed and cleaned.
Isolation of Follicles
Antral follicles (400–700 µm) were manually dissected with watchmaker forceps and saved in ice-cold medium 199 (Gibco, Grand Island, NY). Any damaged antral follicles were discarded. As a control for the preantral follicles enzymatically incubated overnight (see below), follicles 389 µm in diameter were manually dissected from ovaries collected at 24 h post-eCG. This mixture of small antral and preantral follicles was then immediately incubated with LH, and progesterone and estradiol were measured as described below. The ovarian remnants were then incubated with collagenase (type 1, 200 U/pair ovaries; Sigma) and DNase 1 (100 IU; Sigma) and incubated at 37°C for 20 min with gentle shaking in a water bath; the suspension was gently agitated with a Pasteur pipette, and an equal volume of medium 199 was then added.
Contrary to our experience with hamsters [2] and mice [6], the rat preantral follicles were still enmeshed in tightly adherent interstitial cells after 20-min incubation. We therefore adopted the modifications used to completely free human preantral follicles [7]; i.e., the mixture of follicles and collagenase, in Krebs-Ringer bicarbonate (KRB), was incubated overnight at 4°C. The digest was then centrifuged at 54 x g for 3 min at 4°C; the supernate was decanted, and 5 ml of Ca2+- and Mg2+-free KRB containing 0.5% BSA was then added.
The suspension was filtered through Teflon (DuPont, Wilmington, DE) or nylon mesh (70, 110, 150, 248, and 350 µm) to separate the isolated follicles by size. The follicles were then further categorized using precalibrated mouth pipettes (55, 110, 148, 222, and 389 µm in diameter) and saved in ice-cold medium 199.
Incubation of Follicles
The predetermined number of follicles for each size class (see below) were placed in 24-well culture dishes (Costar, Cambridge, MA) containing 30-µm mesh (Transwell; Costar), and 1 ml medium 199 was added; the culture dish was placed in a water bath at 37°C for 1 h. This baseline medium was then aspirated, and fresh medium 199 was added with or without 10 ng ovine LH (S-25, NIDDK; NIH, Bethesda, MD) or 100 ng bovine FSH (B-1, USDA, Beltsville, MD); the follicles were then incubated for an additional hour. The media samples were stored at -20°C for subsequent steroid assays.
The number of follicles needed per replicate to assay the steroids depended on the number required to measure detectable levels of progesterone on Day 25 in the control group after addition of LH. The numbers of replicates for each hormone determination are listed in Figures 2–5.
|
Steroid Assays
Progesterone, androstenedione, and estradiol were measured in each media sample by RIAs using previously described methods [3, 8]. Antisera against the steroids were kindly provided by Dr. Gordon Niswender (progesterone: GDN 337 and estradiol GDN 244; Colorado State University, Fort Collins, CO) and Dr. John Resko (androstenedione: 18-9-15-80; Oregon Health Sciences University, Portland, OR). The results are expressed as pg/follicle per hour.
Histology of Isolated Follicles
Follicles were fixed and sectioned as previously described [2]. The sections were stained with hematoxylin and eosin.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Figure 1 shows representative follicles sectioned after overnight incubation with collagenase. Note the normal histology, unaffected by the prolonged incubation time.
|
Steroidogenesis by Antral Follicles (700–400 µm) (Figure 2)
The most striking finding was the strict gradation in steroidogenesis, with the 700-µm follicles producing maximal amount of steroids and with drastic declines as one moved from 550- to 400-µm antral follicles. For example, comparison of steroid production at 24 h post-eCG between 700- and 400-µm follicles, incubated with LH, shows that the large follicles produced 13.3-, 4.5-, and 325-fold greater amounts of progesterone, androstenedione, and estradiol, respectively, than the smallest antral follicles. The fall in estradiol secretion in response to LH was especially pronounced between the two extremes of antral follicles.
For the 700-µm follicles, baseline levels of progesterone were extremely low across the 3-day period, ranging from 3 to 4 pg/follicle per hour, whereas there was a progressive increase in androstenedione and estradiol at 48 h post-eCG; basal release of androstenedione and estradiol was 248 and 151 pg/follicle per hour, respectively. After the addition of 10 ng LH, progesterone was elevated in all groups but did not differ significantly between days (Day 25 controls: 110 ± 13 pg/follicle per hour; 24 h post-eCG: 122 ± 43 pg/follicle per hour; 48 h post-eCG: 155 ± 36 pg/follicle per hour).
For the 700-µm follicles, the greatest response to LH for androstenedione and estradiol was at 48 h post-eCG: 1099 ± 134 pg/follicle per hour and 191 ± 37 pg/follicle per hour, respectively. It is curious that the high spontaneous release of estradiol in the baseline hour was not further increased by either LH or FSH. The dominant steroid produced in response to LH was androstenedione for the 400- to 700-µm follicles; progesterone was the next dominant, with estradiol lagging far behind in the 550- and 400-µm follicles. For the latter two sizes of follicles, adding either LH or FSH drastically reduced estradiol accumulation below baseline levels. In general, 100 ng FSH rarely elevated any of the steroids beyond the level for the control incubation except for progesterone at 24 h post-eCG for the 700-µm follicles.
Steroidogenesis by Preantral Follicles (389–55 µm)
The results for progesterone are summarized in Figure 3. Follicles in the 389-µm collections represented a 50:50 mixture of large preantral and small antral follicles (Table 1). Consequently, as a transitional group after addition of LH, except for 48 h post-eCG (for which there were 5 replicates), there were only 2 detectable values on Day 25 in control follicles and at 24 h post-eCG. In the manually dissected 389-µm follicles, immediately incubated with 10 ng LH, the mean value of progesterone (for 3 replicates) was 9.0 ± 0.44 (SEM). This overlapped the progesterone accumulation in the 389-µm follicles that were enzymatically incubated overnight before LH was added in vitro (Fig. 3). However, after addition of LH at 48 h post-eCG, the level of progesterone was approximately 5-fold higher in the 389-µm pool than in the 400-µm follicles (cf. Figs. 2 and 3). The stepwise reduction in steroidogenesis as follicles became smaller and smaller continued down to the 110-µm follicles that had 2–3 layers of GCs and were surrounded by theca cells (Table 1). For these small follicles, it is not known whether the stimulatory effects of LH on progesterone secretion represent a physiological response or are artifactual due to the large number of follicles (n = 500–550) pooled per replicate. However, for the primary follicles (55 µm), the levels of progesterone induced by LH were so minuscule that they are of doubtful value. What is clear, though, is that FSH in only one instance (24 h post-eCG; 110-µm follicles) significantly increased progesterone levels in comparison to the baseline value, whereas LH was much more effective in consistently stimulating progesterone accumulation.
|
Androstenedione secretion by preantral follicles is shown in Figure 4. Androstenedione was still the dominant steroid, still measurable in follicles as small as 110 µm. FSH again failed to stimulate hormone release, whereas 10 ng LH was effective. Note that the primary follicles (55 µm) did not secrete appreciable amounts of androstenedione.
|
For the sake of completeness, Figure 5 summarizes estradiol accumulation by the preantral follicles. Even the 389-µm follicles produced less than 0.6 pg/follicle per hour after addition of LH on Day 25, with further declines evident in the smaller-sized follicles. The 389-µm follicles that were manually dissected and immediately incubated with LH produced negligible amounts of estradiol, 0.13 pg/follicle per hour (n = 3). This was similar to the amount produced by the same-sized follicles incubated overnight before the in vitro addition of LH.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
After exposure for 24 h to eCG, the preantral rat follicles develop hCG receptors, and follicles with 2 or more layers of GCs, even in the untreated animal, have FSH receptors [1], confirming previous observations [10]. Despite the consistent presence of FSH receptors, 100 ng FSH only sporadically increased progesterone production by the intact follicles, especially the smaller ones. This differs from the situation in the hamster, where estrous follicles with 1–4 layers of GCs (and lacking theca cells) spontaneously secrete progesterone in baseline incubations and significantly increase the hormone when FSH, but not LH, is added [3]. Thus, a true species difference seems to exist.
It is possible that the preantral rat follicle might respond to higher doses of FSH and longer incubation times. However, the dosage used—100 ng FSH—is comparable to the amount added in previous studies [11]; 100 ng bovine FSH contains 2 ng LH activity. A more likely explanation is that the extensive positive results reported in the literature stem from the use of cultured rat GCs collected for several days before FSH is added [4, 5, 12, 13]. The cultured GCs of the rhesus monkey and other species undergo spontaneous luteinization evidenced by fine structural changes such as increased cytoplasmic-nuclear ratios and development of smooth endoplasmic reticulum and tubular mitochondria, correlated with increased progesterone secretion [14]. As a result, these cells are often referred to as "granulosa-luteal cells" [4]. The cultured rat cells increase progesterone synthesis in response to FSH, LH, and prolactin [4]. Rat luteal cells recovered on Day 4 of pregnancy also respond in vitro to FSH, LH, and prolactin by increasing progesterone accumulation; simultaneous incubation with all three gonadotropins results in an additive increase in progesterone production [15].
An interesting finding of this study is the strict gradation in steroidogenesis based on the size of the follicles. The largest antral follicles (700 µm) are the group that will ovulate after injection of hCG, and they produce the maximal amounts of steroids over the 2-day period. A drastic decline in all three steroids occurs in the 400- and 550-µm antral follicles, with an especially sharp drop in estradiol secretion. A similar pattern is found in hamster proestrous follicles between large follicles (500 µm +) and medium-sized antral follicles (100–300 µm); the latter usually contain no detectable amounts of estradiol [16].
For the 700-µm follicles at 48 h post-eCG, estradiol accumulation was elevated in the baseline incubation over the 24-h value but was not further increased by 10 ng LH, in contrast to significant increases at 48 h post-eCG in progesterone and androstenedione. In this model, 10 ng LH is perhaps not enough to maximally stimulate estradiol production. For example, using large antral follicles from proestrous rats (collected at 0900 h) and incubating for 4 h (with medium changed every hour), 100 ng LH, but not 1 or 10 ng, stimulated steroid secretion, e.g., estradiol accumulation of 
1.5 ng/follicle per hour [17]. In the present study, the steroids were all in the picogram range with androstenedione as the dominant steroid accumulated, followed by progesterone and, lastly, estradiol. This order was true for follicles of all sizes. The reduced steroid levels reported in this paper could result from the use of a minimal priming dose of 5 IU eCG to stimulate the first wave of follicular development. In most other studies, 20–30 IU eCG was injected.
As previously mentioned, significant accumulation of estradiol was restricted to the mature follicles (700 µm). A curious, paradoxical observation was that the smaller antral follicles (400 and 550 µm) at 48 h post-eCG actually decreased estradiol accumulation below baseline levels after either LH or FSH was added. The only tentative explanation we can offer is that in these immature antral follicles, the androgen precursors may be converted to metabolites not measured by the RIA for estradiol.
We never incubated the dissected antral follicles (400–700 µm) overnight, anticipating that they would be atretic after deprivation of gonadotropin support for at least 24 h.
With regard to the preantral follicles (< 389 µm), estradiol accumulation was negligible, although the intracellular concentration might still be of some physiological consequence. However, addition of LH resulted in production of both progesterone and androstenedione in follicles as small as 110–148 µm, with still appreciable levels of the latter steroid. These are preantral follicles that have 2–4 layers of GCs and that possess theca cells. These follicles show hCG receptors in theca cells and FSH receptors in GCs [1].
In the previous autoradiographic study [1], preantral follicles before eCG injection were surrounded by spindle-shaped theca cells, whereas 24 h after eCG, the theca cells were epithelioid in appearance. It was suggested that this may result in increased steroid production. This is verified in Figure 4: steroidogenesis by preantral follicles (148–389 µm) shows that androstenedione secretion is enhanced by 24 h post-eCG by LH. Follicles 55 µm in diameter, surrounded by 1 layer of GCs and lacking a theca investment, seem incapable of secreting steroids. These primary follicles do not react to eCG by prematurely developing a theca coat or increasing the number of GC layers [1].
Small secondary follicles concurrently produce progesterone and androstenedione, consistent with the localization of 3ß-hydroxysteroid dehydrogenase and c17-C20 lyase in the theca interna (for references, see [18]). The rapid conversion and buildup of androstenedione are consistent with a very efficient c17-C20 lyase system. On the other hand, there is a long lag before estradiol is produced in appreciable amounts by the preovulatory follicles. This agrees with the late appearance of immunocytochemical localization of aromatase in immature rat ovaries, which first occurs 48 h after eCG in the GCs of large graafian follicles [19]. Similarly, aromatase mRNA is not detectable in rat ovaries containing only small immature follicles; it is only after treatment with recombinant human FSH or human menopause gonadotropin (hMG) and development of antral follicles that the signal is expressed solely in GCs [20]. Not all of the antral follicles stimulated by hMG show FSH mRNA. The present results agree in showing that the smaller categories of antral follicles stimulated by eCG produce little estradiol, especially the 400-µm group.
An obvious concern is the viability of the preantral follicles incubated overnight before addition of FSH or LH in vitro. Several lines of evidence suggest that the follicles were functional: 1) the morphology of the follicles was normal (Fig. 1), and numbers of pyknotic GCs were minimal; 2) steroid accumulation dramatically increased in the second hour of incubation in response to LH of preantral follicles ranging in size from 110 µm to 389 µm (Figs. 3 and 4); 3) perhaps most convincingly, preantral follicles 389 µm, manually dissected from the ovary and immediately incubated with LH, produced progesterone in the same range as the follicles maintained overnight at 4°C.
However, the definitive experiment has not yet been performed, i.e., long-term incubation of preantral follicles in the presence of FSH for several days. This has been done with preantral human follicles isolated by the same overnight procedure [7]. The result was the differentiation by 120 h of antral follicles secreting progesterone, androstenedione, and estradiol and significant increase in DNA by 24 h. In the present study, the steroid results were expressed per follicle rather than in terms of DNA. However, a direct correlation exists between DNA content and stage of follicular development as assessed by size [21], which we feel justifies expressing the steroid values per follicle.
In summary, enzymatically dissociated rat follicles—from large graafian follicles (700 µm) to small preantral follicles with 2–4 layers of GCs and theca cells (110–148 µm)—produce steroids in vitro in response to 10 ng LH but not 100 ng FSH. There is a distinct hierarchy in steroid synthesis: the largest antral follicles secrete maximal amounts of progesterone, androstenedione, and estradiol, and there is a precipitous drop in the smaller antral follicles (550 and 400 µm), especially in estradiol accumulation. Preantral follicles were also capable of producing progesterone and minuscule amounts of estradiol after addition of LH, with androstenedione as the dominant steroid. Although the secondary follicles were consistently stimulated by 10 ng LH, 100 ng FSH was ineffective. However, primary follicles with 1 layer of GCs and no theca cells did not synthesize appreciable amounts of any of the steroids. The simple technique described in this paper, enzymatic dissociation of intact preantral rat follicles, may be valuable for a wide variety of investigations.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: G. Greenwald, Department of Molecular and Integrative Physiology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Kansas City, KS 66160-7401. FAX: 913 588 7430; ggreenwa{at}kumc.edu
Accepted: June 7, 1999.
Received: February 4, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Baltsen Gonadotropin-Induced Accumulation of 4,4-Dimethylsterols in Mouse Ovaries and Its Temporal Relation to Meiosis Biol Reprod, December 1, 2001; 65(6): 1743 - 1750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kishi and G. S. Greenwald Autoradiographic Analysis of Follicle-Stimulating Hormone and Human Chorionic Gonadotropin Receptors in the Ovary of Immature Rats Treated with Equine Chorionic Gonadotropin Biol Reprod, November 1, 1999; 61(5): 1171 - 1176. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
