| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Gamete Biology Group, Department of Reproductive Biology, German Primate Center, D-37077 Göttingen, Germany
b Research and Development Group, N.V. Organon, Oss, Netherlands
ABSTRACT
We evaluated the effects of follicular exposure in vitro to either of two mutually exclusive isoforms of FSH (least acidic and acid) on the subsequent capacity of oocytes for embryonic development. The effects of dose and follicle culture duration were examined. At the threshold dose (that required to produce antra) and at one subthreshold dose, the major difference between the two isoform fractions was the timing and effectiveness of acquisition of two-cell embryonic developmental capacity. With the least-acidic fraction, the highest rate of two-cell development (
80%) occurred after 3 days of follicle culture only at the threshold dose (2.5 ng/ml). With the acid fraction, the highest two-cell rate (60%) occurred after 5 days of culture but at equivalent rates over a range of doses between 10 ng/ml and 100 ng/ml (threshold dose was 50 ng/ml). At threshold dose or below, the capacity for two-cell embryo production appeared not to be influenced by antral status for either isoform. At above threshold doses, the least-acidic fraction induced an increasing proportion of antral follicles with increasing dose, but this increase was associated with a progressive decrease in embryo production. This relationship was more extreme after longer culture and was due to degeneration of the cumulus-oocyte complex associated with apparently increased differentiation of the mural granulosa cells. The acid fraction was by comparison less bioactive and insensitive to overdosing. The broader isoform mix of the unfractionated FSH provided a measure of protection against overdosing characteristic of the acid fraction while retaining the capacity of the least-acidic fraction to induce antral formation at a low dose.
early development, follicle, follicle-stimulating hormone, folicular development, oocyte development
INTRODUCTION
FSH, a glycopeptide secreted by the pituitary, is a critical agent in the progressive development of ovarian follicles [1], and it affects many different aspects of follicular differentiation [2, 3]. Within the follicle, the major targets of the actions of FSH are the granulosa and cumulus cells, which in turn regulate the growth and development of the oocyte [4]. FSH, as other gonadotropins, exists as a range of isohormones that differ from each other in their posttranslationally modified carbohydrate composition based on differences in the degree of terminal sialation and in side-chain branching. One hypothesis is that this carbohydrate-based microheterogeneity, the distribution of which varies over the cycle, play a key role in the induction and modulation of the multiple FSH functions [5]. Evidence is accumulating through in vitro and in vivo studies that FSH isoforms induce pleiotropic actions on the target somatic cells of the follicle [6–9] and that this effect extends to influences on the oocyte's capacity for maturation [10].
The results of a previous study [11], in which we used follicle culture to compare three isoform fractions derived from recombinant human FSH, are supportive of and extend these findings. These results represent the first information concerning the effects of FSH isoforms on the development of the intact follicle unit. Our results further indicate that exposure to one or the other of the isoforms with more extreme isoelectric points (pI) during follicular development has a differential effect on the capacity of the oocytes for subsequent maturation [12]. However, information is not yet available on the issue of whether FSH isoforms may ultimately also influence the subsequent developmental potential of the oocytes through the influence of these isoforms on follicular development. To investigate this question, the two FSH isoforms already shown to influence follicle development and oocyte maturability were compared with the unfractionated parent FSH to determine whether embryo developmental capacity would be differentially influenced.
MATERIALS AND METHODS
Animals
Ovaries were collected from 18- to 20-day-old F1 female offspring from C57Bl/6J female and CBA/J male mice (Harlan Winkelmann GmbH, Borhen, Germany). Additionally, 21- to 24-day-old female mice of the same F1 generation were used for control in vitro maturation trials. Adult male F1 mice were used as sperm donors for in vitro fertilization. Mice used for hypogonadotropic (hpg) serum collection were 40- to 45-day-old offspring of hpg/bm mice (Jackson Laboratory, Bar Harbor, ME). The animals were anesthetized with diethyl ether and killed by cervical dislocation. Experiments were conducted according to German animal protection laws and the Guiding Principles for the Care and Use of Research Animals promulgated by the Society for the Study of Reproduction.
FSH Fractions
The recombinant human FSH preparations used for this study were produced and supplied by the Research and Development Group of Organon and were the same as those used in our previous studies [11, 12]. Unfractionated FSH had a purity of more than 95% and a biologic potency of 9380 IU/mg and comprised a pI range of 3.6–5.6. The fractionation of FSH was performed by ion exchange chromatography (EIX). The two mutually exclusive isoform fractions derived from unfractionated FSH used in this study were referred to as acid (pI: 3.6–4.6, main peak at 3.8 = 48%) and least acidic (pI: 5.0–5.6, main peak at 5.5 = 63%). The doses used were chosen according to results obtained from the previous study [11] (suboptimal, optimal, and two higher doses for each fraction).
Follicle Culture
The follicle culture system applied was the same as described previously [11]. Follicles isolated by needle dissection within a size range of 150–180 µm were cultured in alpha minimal essential medium (90044; Gibco Life Technologies, Invitrogen, Karlsruhe, Germany) supplemented with 5 µg/ml insulin (I 1882; Sigma-Aldrich, Deisenhofen, Germany), 1 µM L-glutamine (15039; Gibco), 0.01 mg/ml transferrin (T 5391; Sigma), 50 µg/ml L-ascorbic acid (Sigma A 4544; Sigma), and 5% hpg mouse serum, which was collected and stored as described previously [11]. The culture medium was supplemented with either unfractionated, least-acidic or acid FSH fractions at concentrations ranging from 0.5 to 250 ng/ml. The culture was carried out in humidified incubators at 37°C with 5.5% CO2 in air. The follicle size was assessed by measurement of two perpendicular diameters between the outer theca layers. Both follicular size and antral formation were observed with a stereomicroscope (Leica, Bensheim, Germany) at a magnification of 50x. The detection of antral formation by stereomicroscopy during culture was compared with histologic assessment and was considered accurate.
In Vitro Maturation
After 3, 4, or 5 days of culture, the follicles of each treatment group were punctured for oocyte retrieval. After follicle puncture, the proportion of oocytes with clearly visible GV stages and the proportion of oocytes completely enclosed by cumulus were assessed for each treatment group. As a quality control of the in vitro maturation (IVM)/in vitro fertilization (IVF) system, cumulus-enclosed oocytes were also collected from antral follicles obtained from mice 48 h after injection of eCG (5 IU/mouse; Intergonan; Vemie Veterinar Chemie, Kempen, Germany). The released oocytes were aspirated, pooled, and matured for 16 h in M-199 (Gibco) medium supplemented with 10% fetal bovine serum (FBS) (Seromed S0775S; Biochrom, Berlin, Germany), 25 µg/ml pyruvate (Sigma P-8574), 50 µg/ml gentamycin sulfate (Sigma G-3632), 100 ng/ml unfractionated recombinant human FSH (9380 IU/mg; Organon, Oss, Netherlands), and 20 ng/ml epidermal growth factor (855731; Boehringer Mannheim, Mannheim, Germany). Unfractionated FSH was used without LH to provide a standardized and simplified maturation stimulus unrelated to treatment during follicle growth. In an earlier study, >90% mature oocytes were produced from in vivo antral follicles and >80% mature oocytes were produced from follicles grown with unfractionated FSH, indicating that the oocyte quality produced by the follicle culture and the IVM system were satisfactory.
IVF and Embryo Culture
For each trial, a sperm suspension was prepared from each cauda epididymis and vas deferens of at least two mature F1 males. The spermatozoa were collected and kept in modified Tyrode solution T6 [13] supplemented with 0.6% BSA (05477; Fluka, Sigma-Aldrich) and 50 µg/ml gentamycin sulfate. The matured oocytes were placed into fertilization drops and immediately mixed with the prepared sperm sample (final sperm count, 1.5 Mio/ml). The fertilization medium was M16 [14] supplemented with 5% FBS, 0.4% BSA (A3311; Sigma), and 50 µg/ml gentamycin sulfate. After 5 h of oocyte/sperm coincubation, the oocytes were placed into embryo culture medium, M16 supplemented with 0.4% BSA and 50 µg/ml gentamycin sulfate. The proportion of two- to four-cell embryos (two, three, or four clearly visible blastomeres) was assessed 24 h after onset of embryo culture. The proportions of morulae (more than 16 cells/embryo) and blastocysts (clearly visible blastocoel) were recorded at 120 h after IVF.
Histologic Methods
The histologic methods used were described in detail in a previous report [15]. Follicles were fixed in 3% paraformaldehyde and 2.5% glutaraldehyde in phosphate buffer, dehydrated, and embedded in Technovit 7100 (Heraeus Kulzer, Lehrheim, Germany) in beem capsules. Sections (0.5 µm) were stained with undiluted Löffler methylene blue solution (Merck, Darmstadt, Germany).
Data Analysis
Only follicles that remained intact for the whole culture period were considered for analysis. The distributions of follicle starting sizes between the treatment groups of the main experiment were not significantly different as assessed by one-way ANOVA. The mean follicle size obtained after culture was assessed by calculation of the mean (±SEM) sizes of the follicles in each treatment group. Comparison of follicle sizes between the treatment groups was analyzed using one-way ANOVAs. Differences between treatment groups in the number of follicles showing antral formation, the number of GV-stage and cumulus-enclosed oocytes, and the number of two-cell embryos and blastocysts were assessed with the chi-square test and, where total numbers were low, with the Fisher exact test. P-values of <0.05 were considered significant.
RESULTS
Higher Doses of Both FSH Fractions Have a Detrimental Effect on Follicle Size
Four concentrations of unfractionated, least-acidic, and acid FSH were tested for their effects on follicle development and oocyte functionality after 5 days of follicle culture. The range was individualized for each, based on the previous study. Included for each was one dose below antral threshold, one at threshold, and two above the threshold dose. The first dose at which follicle growth surpassed the threshold size for antral formation (340-µm diameter) was 2.5 ng/ml for both unfractionated and least-acidic and 50 ng/ml for acid FSH. For both least-acidic and acid fractions, the highest dose tested resulted in a smaller mean follicle diameter than did lower doses. In contrast, the end follicle size was similar at all higher doses of unfractionated FSH (Fig. 1A).
|
The proportion of follicles showing antral formation (Fig. 1B) was similar and stable for both acid and unfractionated FSH with threshold and higher (range, 54%–60%) doses, in spite of the reduced end size with the highest dose of acid FSH. Only least-acidic FSH produced increasing and high rates of antral formation with increasing dose (starting at 72% and increasing to 95%). Examples of follicles cultured at threshold doses of least-acidic and acid FSH are shown in Figure 2, A and D, respectively.
|
Differential Effects of FSH Isoform Fractions on Oocyte Quality
Higher doses of least-acidic FSH were associated with a reduced oocyte recovery; no oocytes could be recovered in 15%–20% of the follicles. There was also a reduction in the proportion of oocytes that were completely enclosed by cumulus (<50% compared with >70% in all other treatments), and in 25% of these oocytes a GV nucleus was not visible at follicle puncture. A less severe reduction in oocyte quality occurred at the highest dose of acid FSH, where only 8% of oocytes did not have a GV nucleus immediately after release from the follicle.
Histologic investigation confirmed that the higher doses of least-acidic FSH were associated with variable degrees of oocyte morphologic degeneration. There was no visible degeneration at threshold dose, and a balance of inner (more loosely packed) mural granulosa cells and outer (tightly packed) cells was observed (Fig. 3A). An extreme example of a follicle exposed to a high dose of least-acidic FSH showed a degenerated oocyte and cumulus cells. This effect was further characterized by the mural granulosa cells being composed only of the presumptive outer tightly packed granulosa cell type (Fig. 3B). This effect contrasted with that of acid FSH, where no obvious degeneration was observed at higher doses (Fig. 3, C and D) and the granulosa cells retained a balance between inner loosely packed and the outer tightly packed cells.
|
Functional confirmation of reduced oocyte quality was provided by the significant suppression of two-cell embryo production (Fig. 1C) by higher doses of least-acidic FSH. Figure 2, B and E, show examples of two-cell embryos, of apparently comparable appearance, from least-acidic and acid FSH at their respective threshold doses. Although the threshold doses were identical for both least-acidic and unfractionated FSH, higher doses of unfractionated FSH did not have a detrimental effect on two-cell formation. However, a drastic reduction or a complete absence of blastocysts was produced at suprathreshold doses by both subfractions and by the unfractionated FSH. Figure 2, C and F, show examples of blastocysts from least-acidic and acid FSH at their respective threshold doses. Larger cell sizes and apparently lower cell numbers are evident in the blastocyst from the follicles grown in the acid fraction.
Influence of Follicle Culture Period on Oocyte Quality
To more closely characterize the narrow dose tolerance of the least-acidic fraction, a time series was conducted. The early rapid growth preferentially induced by least-acidic FSH produced follicles that were larger after 3 days of culture than those growing in response to acid FSH (Fig. 4, A and B). With least-acidic FSH, the highest two-cell embryo growth rates were seen after 3 and 4 days at the threshold dose (
70%), with the highest blastocyst rate occurring after 4 days. One additional day of culture reduced blastocyst production. In contrast, acid FSH induced similar proportions of two-cell cleavage after all culture periods. At threshold dose, approximately 20% of two-cell embryos formed morulae after 3 and 4 days of follicle culture, but blastocysts were formed only by oocytes collected after 5 days, indicating that complete oocyte development took 1 day longer with the acid fraction than with the least-acidic fraction.
|
Relationship Between Antral Formation and Oocyte Quality
We investigated the relationship between follicular antrum development and oocyte development with least-acidic and acid FSH by grouping the follicles according to their antral status (Fig. 5, A and B). Antral follicles induced by least-acidic FSH were significantly larger than nonantral follicles after all culture periods. However, after 3 and 4 days of culture, no differences were seen in the potential of oocytes from antral and nonantral follicles to form two-cell embryos. After 5 days of culture, oocytes from antral follicles had higher blastocyst rates than did those from nonantral follicles (29% vs. 11%). In cultures with acid FSH, antral follicles were only obtained after 4 and 5 days of culture. Similar to oocytes in least-acidic FSH, acid FSH oocytes from both antral and nonantral follicles had the same capacity for formation of two-cell embryos after 4 days of culture. However, after 5 days the proportion of two-cell cleavage was significantly greater in oocytes collected from the nonantral (83%) than in oocytes from antral (29%) follicles.
|
DISCUSSION
In a previous study [11], FSH fractions purified from recombinant human FSH and differing in pI range had differential effects on the pattern of follicle development in vitro. The main characteristic of the least-acidic fraction (pI of 5.0–5.6) compared with the acid fraction (pI of 3.6–4.6) was that it specifically stimulated rapid preantral follicular growth. This efficiency also extended to the early and highly effective induction of oocyte meiotic maturation capacity in both antral and preantral follicles [12]. This result is in agreement with that of a previous study showing that FSH isoform range can also have an effect when the exposure is only at the terminal event of meiotic maturation itself [10]. The present study extended these results to reveal the differential effects of optimal and supraoptimal doses of FSH fractions on oocyte quality and subsequent embryo development. Least-acidic FSH had a promoting effect on both. However, this effect could only be achieved when the culture period and dose were restricted; longer culture periods and higher doses had a detrimental effect on embryo development. This finding suggested that continuing exposure to least-acidic FSH without the buffering effect of other isoforms was detrimental.
To examine whether a relationship exists between antral formation and two-cell embryo potential, we separately analyzed nonantral and antral follicles. For up to and including 4 days of culture, there was no significant difference between antral and nonantral follicles in two-cell embryo production for both FSH fractions. This finding indicated that the initial stages of cytoplasmic maturation are not necessarily dependent on antral formation. However, the relatively high rate of blastocyst production specifically from antral follicles after 4 days of culture suggested that for least-acidic FSH, optimal oocyte development is in some way correlated with or dependent upon antral formation. In contrast, the lower rate of two-cell production, with the acid isoform and low formation rate or lack of blastocysts, indicated an overall less effective induction of oocyte functionality by this isoform, regardless of antral status.
After a further day of exposure to the least-acidic fraction (5 days in total), the two-cell and blastocyst production from antral follicles tended to be supressed below what they were after 4 days of culture, but those follicles that were still nonantral after 5 days of exposure to the strong antrus-inducing stimulus of the least-acidic fraction had a significantly lower rate of two-cell embryo production than did their antral counterparts after the same amount of time in culture. Further, there was an absense of blastocysts. This effect can only partially be accounted for by the slightly lower maturation rate of the oocytes [12]. However, the negative effect of an additional day of culture was more marked for the acid fraction after 5 days in culture. Both the capacity for meiotic maturation [12] and embryo production decreased significantly for antral follicles between 4 and 5 days of culture, whereas both capacities for nonantral follicles improved dramatically. These findings, the lower tendency for antral formation, and the lower level of estrogen production [11] indicate that acid FSH might have a maintenance function for the preantral follicle and may act as a counterbalance for the strong proliferation and differentiation stimulus of less acidic isoforms.
Culture period was a very important factor affecting embryo development after exposure to least-acidic FSH. Longer times in culture progressively magnified the detrimental effect of least-acidic FSH overdosing on oocyte quality in spite of the large size and high proportion of antral follicles produced. The negative effect of culture period was not observed as drastically in unfractionated and acid FSH. The strong detrimental impact of least-acidic FSH after longer culture periods suggests that there is a cumulative effect or that later stages are specifically affected. In contrast, the effect of overdosing with acid FSH was apparent even after shorter culture periods, which indicated that early stages of follicle development are more strongly affected by this isoform. The combination of the two isoform fractions in unfractionated FSH seems to provide protection against the negative effects.
The histologic demonstration of oocyte and the cumulus cell degeneration or premature resumption of meiosis provided an explanation for the reduced oocyte recovery and lower embryo development rates found with least-acidic FSH at supraoptimal doses. This effect was particularly striking because other researchers using intact follicle culture have reported that oocytes remained intact and in GV stage in vitro as long as they resided in the follicle [11, 16–18]. The oocyte degeneration in response to least-acidic FSH appeared to be related to a higher number of outer mural granulosa cell layers. It is not yet clear, however, how the degeneration of the inner follicular compartment is triggered.
The differences between the isoforms in dosage sensitivity, which were also shown in our previous study [11], may in part be explained by the differences in binding affinities between FSH fractions. More-acidic FSH fractions have a lower binding affinity than do less-acidic forms, so that higher doses are needed to obtain similar cellular responses [19]. Thus, the least-acidic fraction with its higher binding affinity may induce cellular responses more effectively, possibly leading to a faster change in specific cell differentiation patterns, which cannot be induced by acid FSH during the chosen time frame. It is also conceivable that the cumulative effects of the higher doses of single isoforms over time may include progressive downregulation or desensitization of the receptors and that these effects contribute to the negative effects seen. However, the biologic differences among these isoforms in terms of receptor binding dynamics and signal transduction pathways still need to be explored.
In addition to the quantitative differences, the present study revealed that the isoforms have qualitatively different effects on oocyte development that are likely to be modulated through more complex differences in cellular response than can be explained by differences in binding affinity. These qualitative differences in follicle development induced by the isoform fractions may play an important role in modulating the progress of folliculogenesis in vivo. In humans, for example, the isoform range shifts towards the less acidic end as the follicular phase progresses [20]; this shift might be responsible for the quick follicle growth just prior to ovulation and for potentiation of the oocyte's capacity for maturation [10]. In contrast, during the luteal phase of the cycle, acidic isoforms dominate, which might contribute to slower follicle growth.
Although human FSH is bioactive in mouse follicle culture, it is unclear whether isoforms of human FSH function comparably in a homologous system. In support of the the functional relevance of the present results are the facts that human FSH isoforms have been shown to affect in vivo plasminogen activator activity in the rat [9], and that mouse oocyte maturation [10] and cAMP production by the cumulus cells [21] are influenced differentially by human FSH isoforms. The shift in distribution of FSH isoforms during the rat estrous cycle is principally similar to that observed in humans [22], which is a further indication that the functional differences shown here are not species specific.
A potential complication of the system that cannot be ignored is the possible differential accumulation of substances in the medium that could influence the oocyte. However, because the medium was replaced completely every 24 h, any accumulation of substances must have been limited to this time period. One of the potentially important substances to consider is cAMP, because its production is preferentially induced by less acidic FSH fractions in cumulus-oocyte complexes [21]. Because cAMP levels in the medium have not yet been measured in our system, and natural cAMP moves through the membrane less easily than synthetic analogs, the importance of cAMP is unclear. However, abnormally high cAMP levels produced in response to an overdose of the least-acidic fraction could be involved in, and lead to, the oocyte degeneration observed in cultures with least-acidic FSH.
Our results indicate that particular isoform ranges may serve very specific functions and that the effects of each isoform may differ depending on the developmental stage of the follicle and the concentration of the isoform. The rapid early follicle development induced by least-acidic FSH was linked with early and effective acquisition of the oocytes' capacity for embryonic development. Although higher doses of the least-acidic fraction promoted follicle differentiation, they were also associated with reduction in oocyte quality and embryo production. The negative effects on the oocyte were increased with longer culture periods, which indicated that later stages of antral development were particularly vulnerable or that the effects were cumulative over time. In contrast, acid FSH was mostly ineffective in producing antral formation, and the embryo developmental capacity increased gradually with longer culture times. The isoform mixture present in the unfractionated source FSH provided some protection against the negative effects of overdosing while retaining the capacity of the least-acidic fraction to induce antral formation. These findings suggest that the in vivo progressive shift towards more basic isoforms in the late follicular phase [19] may have important modulating effects on follicle and oocyte development in vivo, and that attention should be paid to the isoform distribution of exogenously applied gonadotropins.
ACKNOWLEDGMENTS
We acknowledge the able technical assistance of S. Bete, who was supported by the Organon Contract for this project. We thank P.S. Kiesel for her assistance with the follicle culture and histologic methods and N. Umland for her initial assistance in the establishment of the in vitro fertilization system.
FOOTNOTES
First decision: 20 March 2001.
1 U.A.V. was supported by a research contract from N.V. Organon, Oss, Netherlands, to P.L.N. 
2 Correspondence: Penelope L. Nayudu, Gamete Biology Group, Department of Reproductive Biology, German Primate Center, Kellnerweg 4, D-37077 Göttingen, Germany. FAX: 49 551 3851 288;pnayudu{at}gwdg.de
3 Current address: Ursula A. Vitt, Stanford University Medical Center, Gynecology and Obstetrics, 300 Pasteur Drive, Palo Alto, CA 94305.
Accepted: June 26, 2001.
Received: February 19, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. Ulloa-Aguirre, C. Timossi, J. Barrios-de-Tomasi, A. Maldonado, and P. Nayudu Impact of Carbohydrate Heterogeneity in Function of Follicle-Stimulating Hormone: Studies Derived from in Vitro and in Vivo Models Biol Reprod, August 1, 2003; 69(2): 379 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T. Y. Chiu, M. S. Rogers, C. Briton-Jones, and C. Haines Effects of myo-inositol on the in-vitro maturation and subsequent development of mouse oocytes Hum. Reprod., February 1, 2003; 18(2): 408 - 416. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
