| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gamete Biology |
Follicle Biology Laboratory,3
Centre for Reproductive Medicine4
Department for Pathology,5 Dutch-Speaking Free University of Brussels (VUB), 1090 Brussels, Belgium
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
follicle, granulosa cells, meiosis, oocyte development, phosphodiesterases
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Supplementing culture medium with compounds that maintain elevated cAMP can prevent spontaneous maturation. This can be achieved by several compounds: cAMP analogues such as dibutyryl cAMP (dbcAMP) [7, 23], pharmacological agents that stimulate cAMP production via adenylate cyclase (forskolin) [24], and inhibitors of the cAMP-degrading isoenzymes phosphodiesterase (PDE), such as 3-isobutyl-1-methylxanthine (IBMX) [11, 17].
PDEs are responsible for the breakdown of cAMP and can cause intracellular cAMP levels to decrease. This is a large group of proteins in which at least 11 different families of genes encoding nucleotide PDEs have been characterized [25, 26]. Some PDE families consist of several subtypes and numerous PDE isoform-splice variants. Nonselective PDE inhibitors have been used to understand the role of cAMP in the resumption of meiosis. Suppression of cAMP catabolism by using different specific PDE inhibitors demonstrated the meiosis arresting action of these chemical compounds [27]. In situ hybridization studies have indicated that PDE 3A mRNA, which was initially named PDE 3B, is expressed in rat and mouse oocytes [12, 20, 28]. PDE 3 has been demonstrated to act directly in the oocyte without interfering with somatic cell functions [12]. In the ovarian follicle, PDE4 is mainly involved in the metabolism of cAMP in granulosa cells. The use of specific PDE 3 inhibitor in rodents proved to block oocyte maturation in vivo and in vitro [12, 20, 29], and consequently prevented pregnancy [29]. Tsafriri et al. [12] showed that the type 3 PDE inhibitors, milrinone and cilostamide, inhibited both spontaneous and LH/hCG-induced maturation of isolated and follicle-enclosed rat oocytes. In the bovine, cumulus-oocyte complexes and denuded oocytes incubated in culture medium supplemented with cilostamide or milrinone resulted in meiotic arrest. In contrast, PDE 4 inhibition by rolipram did not block meiotic resumption of bovine oocytes [21]. Recently, it has been shown that culture of denuded rhesus macaque oocytes, in the presence of PDE 3 inhibitors, blocked resumption of meiosis for 48 h [30]. By using a selective PDE 3 inhibitor, interference with cAMP pathway mechanisms in the somatic cells can be avoided. The increased intraoocyte cAMP results in an increase in type I isozyme protein kinase A (PKA) activation and subsequent phosphorylation of specific proteins, which are inhibitory to nuclear maturation [31].
In the present study, we investigated the expression of PDE 3A mRNA in immature human oocytes. It was examined whether a selective phosphodiesterase isoenzyme (PDE 3) inhibitor temporally suppresses nuclear maturation. Morphological and functional parameters were used to evaluate the impact on oocyte quality during and after exposure.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This research project was approved by the Institutional Review Board of the Academic Hospital of the Free University of Brussels. Patient consent was obtained for using oocytes from immature follicles to study new maturation designs.
Collection of Immature Oocytes
Patients included in this study were undergoing fertility treatment and were primed with variable doses of gonadotropins while being carefully monitored using serum hormone levels and vaginal ultrasound [32]. Cumulus-oocyte complexes (COCs) were collected from small follicles (6–14 mm) in the following clinical conditions: I) during laparoscopic surgery, II) before intrauterine insemination (to avoid multiple gestation) [33], or III) during ovum pick up, small follicles (<15 mm) were aspirated after COC collection for IVF treatment. These aspirations yielded two types of COCs: a first group (pre-hCG) that was not exposed to LH activity (conditions I and II), and a second group (post-hCG) (conditions II and III) that was exposed to hCG or to an endogenous LH peak.
Messenger RNA Detection
COCs (n = 2) collected from 14-mm follicles post-hCG were denuded and the cumulus cells of each COC and the individual oocytes were snap frozen in liquid nitrogen. Total RNA was extracted from individual oocytes and cumulus cells using an RNeasy mini kit (Qiagen, Westburg, the Netherlands). Two thirds of the mRNA was used for reverse transcription with the First-Strand synthesis kit (Amersham Pharmacia Biotech, Belgium) using random hexamers. Two pairs of primers were selected to be complementary to both the human and the mouse sequence of PDE 3A (cGMP-inhibited cAMP phosphodiesterase). The PDE 3A1 primer pair F—CAACAGTGACAGCAGTGACAT, R—GAATACGGCCACATTTTCTTCC amplifies a fragment of 245 base pairs (bp). The PDE 3A2 primer pair: F—ACTGACCTGAAGAAACACTTTG, R—GCTGAGTTATTTGGCAGTAGAT amplifies a 511-bp fragment downstream. Both fragments are localized in the Delta 2 coding region [20]. As an endogenous control for the RNA extraction and the reverse transcription (RT) reaction, ribosomal 18S RNA was amplified using the following primer sequences: F—TCAAGAACGAAAGTCGGAGG, R—GACATCTAAGGGCATCACA (sequences were kindly provided by Prof. Dr. R. Einspanier, Technical University of Munich, Freising-Weihenstephan, Germany). PCR amplification (20 cycles for 18S RNA and 30 cycles for PDE 3A primers) was performed using 1:16 of the cDNA reaction with AmpliTaq (Perkin Elmer Applied Biosystems, Belgium). RT-PCR amplification was analyzed on 1.5% agarose gels stained with ethidium bromide.
Classification of COCs
COCs were classified as follows: Type I , compact mass of 3–5 layers of granulosa cells; type II, expanded distal layers of granulosa cells (cumulus), but a compact proximal (corona cells) granulosa cell layer; type III, expanded cumulus and corona cells; and type IV, partially denuded oocytes due to expanded granulosa cells [34].
Culture of COCs
Only the COCs classified as type I or II were placed in culture. The basal culture medium was Dulbecco modified Eagle medium (DMEM; D-5280), with 2 mM Glutamax-I, NaHCO3 (2 g/liter; all from Invitrogen, Merelbeke, Belgium) supplemented with 10 mIU/ml FSH (Gonal-F, kindly donated by Ares Serono, Geneva, Switzerland), 100 ng/ml rIGF-I (R&D, UK), 5 ng/ml insulin, 5 ng/ml selenium, 5 µg/ml transferrin (Invitrogen) and 0.1% human serum albumin (CAF-DCF, Brussels, Belgium), 1 µg/ml 17ß-E2 (Sigma, Bornem, Belgium). The E2 was delivered to the medium through diffusion out of oil, and the final E2 concentration in the culture medium was confirmed using specific radioimmunoassay. A refreshment of 5 µl conditioned medium was performed each day. All cultures were carried out at 37°C in humidified atmosphere in an incubator gassed with 5% CO2 in air.
The inhibitor Org 9935, 4,5-dihydro-6-(5,6-dimethoxybenzo[b]thien-2-yl)-5-methyl-3(2H)-pyridazinone, a specific inhibitor of phosphodiesterase type 3 (PDE 3; kindly provided by Organon, Oss, The Netherlands) was added at a 10 µM final concentration (0.1% dimethyl sulfoxide). Directly after aspiration, single COCs were washed in flushing medium (DMEM with Hepes) and placed into 10-µl droplets under oil. Cumulus-oocyte complexes were randomized per patient over two culture conditions: with and without PDE 3 inhibitor (sibling oocytes).
Evaluation of COC and Reversal of Meiosis
At 24, 48, and 72 h of culture, oocyte stages (GV, GVBD, and PB) and degree of cumulus expansion were recorded and PB-extruded oocytes were removed from culture. In controls, all the remaining oocytes were removed from culture after 48 h.
Oocytes in culture with PDE 3 inhibitor for 48 and 72 h were studied for reversal of meiotic arrest by removing the remaining somatic cells and by placing the oocytes in the control medium for an additional 24–30 h. Oocyte stages were recorded afterward.
Experimental Setup/Design (Fig. 1)
COC preparation for Light (LM) and Transmission Electron Microscopy (TEM) Oocytes were fixed in glutaraldehyde 2.5% in cacodylate buffer, pH 7.3, for at least 2 h at 4°C. All oocytes were embedded individually in SPURR (Taab; Bodson, Liège, Belgium) [35]. Approximately 30 levels (range 15–35) of each oocyte were cut, and each level consisted of four semithin sections of 0.5 µm and three ultrathin sections of 75 nm. The semithin sections were stained with toluidine blue for light microscopic guidance. The ultrathin sections were transferred on wide single-slot copper grids (Agar Scientific LTD, Stansted, UK), coated with a polyvinyl formal film (Formvar, Laborimpex, Brussels, Belgium), stained with lead citrate, and examined using a Zeiss TEM type 9S2 microscope.
|
Oocyte preparation for Laser Scanning Confocal Microscopy (LSCM) Thirty-nine polar body-extruded oocytes (25 oocytes at end of 48 h of culture in control medium and 14 oocytes 24–30 h after inhibitor removal) were fixed with 2.5% glutaraldehyde in cacodylate buffer for at least 2 h, washed with PBS/1% BSA, placed into 0.1% Triton X-100 for 5 min, and rinsed again with PBS. Afterward, oocytes were placed into Texas red-labeled phalloidin (7.5 U/ml), staining f-actin, and Pico green for DNA staining (1:2000 PBS).
Statistical Analysis
Evaluation of cumulus expansion morphology in culture related to follicle treatment, resumption of maturation, and polar body extrusion were analyzed using the 
2 test and Fisher exact test for comparison between control and PDE 3 inhibitor-treated groups. Differences were considered statistically significant when P 
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For the pre-hCG group, a total of 139 follicles were aspirated from 30 patients. The follicles between 6 and 12 mm (45.3%) yielded an oocyte recovery rate of 80.9% (mean ± SD of oocytes per patient: 3.2 ± 2.2; range 1–9). The aspirated follicles larger than 12 mm (54.7%) yielded an oocyte recovery rate of 35.5% (1.7 ± 0.9; range 1–4).
For the post-hCG group, a total of 128 follicles between 6 and 12 mm were aspirated from 27 patients at 36 h after hCG administration. These yielded an oocyte recovery of 50.0% (2.4 ± 1.5; range 1–6). A total of 54 follicles of 13 and 14 mm were aspirated from 9 IUI patients with an endogenous LH peak a day before follicle aspiration. These yielded an oocyte recovery rate of 37.0% (2.2 ± 0.8; range 2–4).
RT-PCR Analysis
As represented in Fig. 2, the GV oocytes show a clear amplification for both the primer pairs, showing that PDE 3A mRNA is present in both human and mouse oocytes. Negative control samples exclude the possibility that the amplification is from a genomic origin. Cumulus cells do not express the PDE 3A mRNA, while the amplification of 18S confirms the integrity of the cumulus cells' RNA and cDNA.
|
Morphological Characteristics of COC at the Moment of Retrieval and During Culture
The expansion pattern of COCs during culture was not modified by the presence of PDE 3 inhibitor in the medium. Although a tendency for retention of cell expansion was observed in the inhibitor group compared with controls, this difference was not significant (Fig. 3).
|
At retrieval (0 h), a higher proportion of compact COCs (type I) was obtained from pre-hCG group compared with post-hCG (P < 0.01; Fig. 4, A and B). The proportion of compact COCs (type I) from pre-hCG group was higher even after 24 h, but only when oocytes were cultured with PDE 3 inhibitor (P < 0.05; Fig. 4B). At 48 h of culture, the oocytes in the pre-hCG group (controls and inhibitor) had a higher proportion of type II compared with the oocytes in the post-hCG group (P < 0.05; Fig. 4, A and B). At this time, the post-hCG group had a higher proportion of partially denuded oocytes (P < 0.05; Fig. 4, A and B).
|
Timed Analysis of Nuclear Maturation
Evaluation at 24 h of culture In the pre-hCG group, a total of 39 COCs at GV stage were cultured in the presence of PDE 3 inhibitor and 37 COCs without (= control). After 24 h of culture with inhibitor, all oocytes remained at the GV stage in contrast with only 24.3% in the control medium (P < 0.001; Fig. 5A). In controls, four oocytes (10.8%) extruded the PB. In a large proportion of oocytes (64.9%), the maturational stage (GVBD or PB) could not be assessed due to the presence of cumulus cells, which impaired the visualization on the perivitelline space (not shown in Fig. 5A).
|
In the post-hCG group, a total of 42 and 40 immature COCs were cultured with and without PDE 3 inhibitor, respectively. After 24 h of culture, 92.9% of oocytes remained at the GV stage in PDE 3 inhibitor medium while only 22.5% in controls (P 0.001). Of those oocytes that did not arrest at the GV stage despite PDE 3 inhibitor, only 2.4% extruded a visible polar body (PB; not shown in graph; Fig. 5B). In controls, 10.0% had extruded a PB. The proportion of oocytes in which it was impossible to visualize the presence of PB in perivitellino space was 67.5% in controls.
Evaluation at 48 h of culture In COCs where the nuclear stage of the oocyte was not clearly visualized, cumulus cells were mechanically removed from oocytes. The COCs that were fixed for EM analysis had their maturational stage determined afterward (on the semithin sections).
In the pre-hCG group, 88.9% of the oocytes cultured in the presence of the PDE 3 inhibitor remained at the GV stage in contrast with only 19.4% in controls (P < 0.001). In the presence of PDE 3 inhibitor, only 5.5% of oocytes were GVBD, 2.8% were PB, and in 2.8%, the stage could not be determined. In controls, 12.9% were GVBD, more oocytes had a PB extruded (64.5%; P < 0.001), and 3.2% were at an undetermined stage (Fig. 5A).
In the post-hCG group, COCs cultured in the presence of PDE 3 inhibitor showed a significantly greater proportion of oocytes at the GV stage (83.3%) compared with controls (6.0%; P < 0.001). In PDE 3 inhibitor, 8.3% of oocytes were GVBD, only 5.6% were PB, and 2.8% had a nondetermined stage. In controls, 12.1% were GVBD and a significantly greater proportion of oocytes had a PB extruded (81.8%) (P < 0.001; Fig. 5B). Table 1 shows the number of PB-extruded oocytes in relation to follicle diameter.
|
It was also observed (from the relation cumulus typing-nuclear progression) that from the COC classified as type II at the time of retrieval from post-hCG patients (51.2%), 77.6% could remain arrested at the GV stage for 48 h of culture. This shows that the inhibitor was effective even in the oocytes with expanded cumulus morphology at the time of oocyte retrieval.
Evaluation at 72 h of culture After 72 h in the presence of PDE 3 inhibitor, 16 out of 19 (84.2%) oocytes from pre-hCG and 10 out of 15 (66.6%) from post-hCG patients were still arrested at the GV stage.
Reversibility After Arrest for 48 or 72 h
As shown in Table 2, after 48 or 72 h of culture in PDE 3 inhibitor, a higher proportion of oocytes from the pre-hCG group did not reinitiate meiosis compared with oocytes from the post-hCG group 24–30 h after inhibitor removal (P < 0.05).
|
The proportion of oocytes that progressed through meiosis up to PB extrusion after inhibitor removal from the post-hCG group was higher when exposure to PDE 3 inhibitor was performed for 72 h than for 48 h (P < 0.05).
Ultrastructure Analysis of COC (LM and TEM)
In order to analyze the effect of PDE 3 inhibition on nuclear and cytoplasm organization, ultrastructural observation was performed at the different culture times in oocytes arrested or not by PDE 3 inhibitor. The oocytes analyzed originated from follicles (out of groups II and III) with diameters between 8 and 12 mm. Numbers and conditions of oocytes analyzed by TEM are listed in Table 3.
|
Morphology of COC at time of retrieval (0 h) Semithin and ultrathin sections revealed that microvilli extended from the oolemma through the inner cortex of the zona pellucida (ZP). The granulosa cells (GCs) appeared as oval or round cells (Fig. 6A) apposed to the ZP with numerous cellular projections crossing the ZP reaching the oolemma (Fig. 6, B and C). Cells displayed a compact arrangement and were tightly adhered to each other by closely apposed plasma membranes resembling junctional complexes (Fig. 6D). The nuclei of the cells were eccentrically located.
|
The GV of these oocytes consisted of an envelope formed by a two-layer membrane, containing one dense compact nucleolus associated with interchromatin granular complexes and extranucleolar chromatoid bodies on the periphery of nucleoli [36]. The nucleoli were regularly shaped, round or oval, homogeneous and formed by fibrillar granules. In all oocytes analyzed, the formation of nucleolus-associated chromatin was observed as a continuous mass (SN, karyosphere), which normally exists in oocytes obtained from antral follicles of 8 mm and larger [37]. The karyosphere occupied an excentric position in the nucleus and consisted of loosely packed fibrils in contact with the nucleolus (Fig. 6E). At the level of cytoplasm organization, a characteristic of all the oocytes was the presence of areas devoid of organelles in the cortex and intermediate cytoplasm. Clumps of mitochondria were found distributed throughout the cytoplasm (Fig. 6E). Aggregates of mitochondria were intermingled with vesicles of different diameters surrounding the GV. Cortical granules were found dispersed throughout the cytoplasm and some were observed under the oolemma of the oocyte (Fig. 6F).
Morphology of COC after 24 h of culture Semithin sections showed that cumulus cells of the controls were more elongated, retracted from the zona pellucida, and partially detached from the oocyte. Mostly, the nuclei of the cells were displaced to the periphery at the opposite side of the elongation (Fig. 7A). In inhibitor-treated oocytes, the cumulus cells started to expand and elongate, although some cells were still round or oval and the nuclei of some cells did not yet migrate to the periphery (Fig. 7C).
|
TEM analysis of oocytes cultured in controls demonstrated that in vitro maturation was associated with reorganization of mitochondria and smooth endoplasmic reticulum (SER), which appear as round vesicles into cytoplasm (Fig. 7B). This reorganization was not different for oocytes retrieved pre- or post-hCG stimulation. Mitochondria were spherical or oval and were dispersed throughout the cytoplasm or around aggregates of SER. These aggregates formed characteristic complexes with peripheral mitochondria in the maturing human oocytes [36] and were predominantly present in the PB-extruded ones (Fig. 7B). The oocyte that remained at the GV stage after 24 h in control medium had a SN constitution and organelle distribution as described at 0 h (figure not shown).
In the PDE 3 inhibitor culture, the GV of almost all oocytes had condensed chromatin around the nucleoli (SN constitution). The nuclear membranes of all oocytes had remained intact (Fig. 7D). All oocytes showed signs of immature cytoplasm similar to the ones fixed at the time of retrieval. In the surroundings of the GV, aggregates of mitochondria were intermingled with vesicles of different diameter (Fig. 7, D and E). Mostly, the mitochondria were found distributed in clumps throughout the cytoplasm (Fig. 7E). A single oocyte (retrieved from a 9-mm post-hCG follicle aspiration) had a more homogeneous nucleoplasm of less condensed chromatin containing chromatin foci at the GV (NSN configuration) and its cytoplasm had a more homogeneous distribution of organelles (figure not shown).
Cortical granules (CG) of oocytes cultured in controls showed a discontinuous distribution at the cell surface of the GVBD oocytes and appeared in 1–2 layers at the cortex of PB-extruded oocytes (Fig. 7B). In PDE 3 inhibitor-treated oocytes, a few cortical granules were found dispersed throughout the cytoplasm and beneath the oolemma of the oocyte (Fig. 7, D and E). The cytoplasm of the GVBD oocyte in inhibitor group showed an inhibition of the distribution of organelles, which were aggregated as in the cytoplasm of a GV-stage oocyte (figure not shown). The oocyte that underwent PB extrusion in inhibitor had the same cytoplasmic morphology as the control oocytes (figure not shown).
Morphology of COC after 48 h of culture Most oocytes cultured in controls and in inhibitor had their surrounding cells more dispersed or completely detached from the zona and fewer transzonal processes appeared from cumulus cells into the ZP (Fig. 8, A and D).
|
The PB-extruded oocytes analyzed after 48 h of culture in controls formed the meiotic spindle perpendicular to the oolemma (Fig. 8B). The organelles, mitochondria, and SER (as vesicles or as aggregates of tubular elements) were dispersed throughout the ooplasm.
All the GV-stage oocytes analyzed after culture in the inhibitor showed a SN constitution (Fig. 8E). The nuclear membranes of all oocytes had remained intact. The organelles in the inhibitor-treated oocytes were located around the nucleus but started to dissociate from their aggregates (Fig. 8E). Although in some oocytes, the organelles were dispersed throughout the ooplasm (Fig. 8F).
In control oocytes, the CG formed 1–3 layers under the oolemma, and in one matured oocyte (pre-hCG), numerous CG were grouped beneath the oolemma at the place of the meiotic spindle. Some swollen SER vesicles were observed perhaps due to oocyte aging (Fig. 8C). In all oocytes, a few CG could also be observed throughout the cytoplasm.
In the inhibitor-treated oocytes, cortical granules were located under the oolemma and formed a single layer. Some were found dispersed throughout the cytoplasm (Fig. 8F). The inhibitor-treated oocytes that underwent maturation (GVBD and PB) had the same cytoplasmic morphology as the controls.
Morphology of COCs after 72 h of culture with inhibitor All the oocytes had a SN constitution at the GV (Fig. 9A). An increased vesiculation was present in the ooplasm, resulting from swelling of SER. Microvilli were less evident and ZP was more homogeneous with no transzonal processes from cumulus cells, indicating a complete loss of connections with the oocyte (Fig. 9B).
|
Microfilaments and Chromosome Analysis
A total of 20 oocytes (7 out of 8 pre-hCG; and 13 out of 17 post-hCG) that were stained after 48 h of culture in control medium presented a metaphase II (MII) plate, of which 80.0% had well-aligned chromosomes and were positioned perpendicular to the oolemma. One oocyte presented chromosomes dislocated from the MII plate (Fig. 10A). Four oocytes had nuclei that formed a clump of chromosomes at the location of the metaphase plate, probably a sign of aging. Microfilaments were observed in the scanned levels of oocytes with a more prominent staining at the optical sections close to the site of first polar body described as a generalized pattern [38]. No superposed microfilaments of actin were observed at the site of polar body extrusion; instead, the whole oolemma had a homogeneous staining at this plane (Fig. 10B).
|
Twelve oocytes presented the MII plate 24–30 h after inhibitor removal (3 out of 3 pre-hCG and 9 out of 11 post-hCG), in which 83.3% had well-aligned chromosomes and were positioned perpendicular to the oolemma (Fig. 10C). Two oocytes had nuclei forming a clump of chromosomes at the place of metaphase plate. The generalized aspect of microfilaments was also observed in those oocytes, which had homogeneous prominent staining at the whole oolemma at the plane of the PB extrusion (Fig. 10D).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Most studies focusing on the cAMP pathway have been carried out in rodents in which meiosis is particularly sensitive to this nucleotide [12, 17, 28, 29], although comparable results were found in other mammalian species as well. In rhesus monkey oocytes, dbcAMP and hypoxanthine significantly reduced spontaneous maturation rate [39, 40]. Jensen et al. [30] reported that culture of primate oocytes, collected following ovarian stimulation (pre-hCG), in the presence of 1 µM Org 9935 blocked meiosis in 100.0% of denuded oocytes for 48 h. Org 9935 blocked meiosis more efficiently than the same concentration of cilostamide (91.0%) and milrinone (59.0%). Meiosis reversal was not investigated in previous studies.
Other substances independent of the cAMP transduction pathway have been used to maintain oocytes at the GV stage in large mammalians. Making use of an inhibitor of protein synthesis (cycloheximide) or an inhibition of protein phosphorylation (DMAP) allowed the blocking of meiosis in cattle and human oocytes for 24 h [41–43]. However, the kinetics of nuclear maturation was accelerated after DMAP treatment, and subsequent developmental competence was compromised [41, 42]. Anderiesz et al. [43] reported that H1 kinase, which is related to chromatin condensation, was activated after gonadotropin stimulation and DMAP did not inactivate this process. Another inhibitor, roscovitine, a purine known to inhibit specifically MPF activity of the cells, was investigated for its effectiveness in maintaining bovine oocytes GV arrested for a period of only 24 h with no signs of subsequent deleterious embryo development [44]. Roscovitine also had an effect on granulosa cells: it caused inhibition of cumulus expansion even in the presence of epidermal growth factor. It is as yet unclear whether the roscovitine inhibitory effect on cumulus cells after a prolonged culture might have consequences for oocyte developmental competence. Bovine oocytes, treated with butyrolactone I (100 µM) for 21 h to arrest meiotic resumption, had a higher developmental potential upon maturation compared with untreated controls when IVM arrest was performed under low oxygen tension in combination with fetal bovine serum [45]. It must also to be taken into account that, by using kinase and phosphorylation inhibitors, there might be interference with different kinase activities that modify the function of proteins related to embryogenesis.
In our study on human oocytes from immature follicles, PDE 3 inhibitor efficiently blocked meiosis in almost all oocytes (100.0% from follicles retrieved pre-hCG and 92.9% post-hCG). A previous study by Tornell et al. [23] had shown only transient arrest for maximally 24 h by using dbcAMP, an analogue of cAMP (77.0%). This transient inhibition could probably be because dbcAMP needs to split off one of their butyryl groups and the resulted active monobutyryl cAMP is hydrolyzed by PDE, while a PDE inhibitor is rather stable under these culture conditions. Another possible reason for this limited inhibition period when using dbcAMP in human COCs could be the limited transport of cAMP or other downstream inhibitor molecules from cumulus into oocyte as a result of the rapid loss of connections during culture. Perhaps cAMP needs patent gap junctions to maintain high cAMP influx levels within the oocyte [7, 11].
It was also observed from the present results that, using a serum-free culture medium, cumulus cells started to loose connections to the oocyte after 24 h. In oocytes collected after hCG exposure, breakdown of connections was more rapid and transzonal processes could be maintained only for a maximum period of 48 h. The PDE 3 inhibitor itself had no statistically significant influence on preventing this progressive loss of coupling between somatic cells and oocyte, although there was a tendency by PDE 3 inhibitor to slow down the expansion process. This emphasizes the advantage of using a selective PDE 3 inhibitor acting directly into the oocyte cAMP metabolism via the intraoocyte PDE 3A bypassing the needs for somatic cell-dependent cAMP generation to maintain nuclear maturation arrest.
In the present study, it was deliberately chosen to collect COCs from small antral follicles before and after hCG administration (or in vivo LH rise). A certain effect of LH/hCG action on the follicle mural granulosa cell was observed. Some of the COCs retrieved from the small follicles after hCG administration showed more expansion in the outer cumulus layers at the time of retrieval than COCs retrieved in the early follicular phase before any hCG effect. This moderate cell expansion is most probably the reflection of a beginning sensitivity of the mural cells in vivo to LH/hCG signaling. Surprisingly, the PDE 3 inhibitor was equally effective in prolonging meiotic arrest in oocytes from complexes despite an initially slightly expanded cumulus. These findings suggest that oocytes retrieved from antral follicles ranging from 6- to 12-mm diameter, which are normally not punctured during an IVF-ovum pick up, could constitute a source of oocytes for setting up in vitro maturation studies. These oocytes are not used for IVF as yet but could be an interesting additional source for the patient.
The human oocyte has a follicle size-dependent ability to resume meiosis in vitro. Oocytes retrieved in the follicular phase of the menstrual cycle from follicles of 9- to 15-mm diameter complete meiotic maturation to metaphase II at a higher rate than oocytes from follicles of 3–4 mm [6, 46]. Oocytes retrieved from follicles in the luteal phase of the menstrual cycle can complete maturation at the same rate irrespective of follicle size [46]. In this study, oocytes were retrieved during the follicular phase at different days of the cycle. We observed an overall lower incidence of in vitro maturation in GV-stage oocytes derived from follicles smaller than 12 mm (Table 1), confirming the relation between follicle size and potential to complete meiosis. This relation has already been well documented in several species [47, 48]. In bovine and goat, embryo development is compromised when oocytes originate from smaller follicles [49, 50].
We observed a higher maturation rate in oocytes retrieved from follicles exposed to hCG pretreatment compared with oocytes punctured during early follicular phases. The hCG administration in vivo apparently exerts an influence on oocyte maturation in vitro. Granulosa cells express the LH receptor at a fairly progressed stage of follicle maturation [51, 52]. In theca cells, the LH receptor is expressed from the earliest stages of follicle development in most mammalian species examined [53]. Exposure of the immature follicles to hCG or endogenous LH could as such have activated mechanisms in the theca cell compartment, which were than further propagated to the granulosa cells. Positive effects of hCG administration during the follicular phase on oocyte maturation in vitro had already been demonstrated in polycystic ovary (PCO) patients. Chian et al. [54, 55] showed that in vitro maturation and fertilization rates from PCO patients were improved with hCG priming before oocyte retrieval. PCO patients have a very particular endocrine microenvironment leading to growth arrest of follicles reaching 10–12 mm. These follicles often demonstrate hypertrophied theca cell layers with an increased steroidogenesis [56]. The PCO follicle might therefore be particular in the way it responds to hCG. This study in endocrinologically normal women also suggests that the positive effects of hCG on maturation of oocytes from small follicles is not solely related to special endocrine conditions such as PCO.
Several researchers have studied immature human oocytes collected after gonadotropin stimulation but unexposed to hCG. Culture time varied from 24 to 52 h and polar body extrusion rates ranged from 44% to 81%, probably depending on follicle stage and medium composition [6, 55, 57]. The percentage of in vitro-matured oocytes retrieved from follicles after hCG administration ranged from 30% to 86% [55, 57].
In the present study, the polar body extrusion rate of oocytes retrieved from immature follicles without prior hCG administration was 64.5% after 48 h of culture. Oocytes retrieved from small follicles (<12 mm) at the occasion of an oocyte retrieval for IVF (36 h of hCG administration) yielded 81.8% polar body extrusion rate when cultured under similar conditions. The fact that follicle sizes were equally distributed in pre- and post-hCG groups suggests that hCG exposure prior to oocyte retrieval might have a positive influence on subsequent oocyte maturation in vitro.
Using a selective PDE inhibitor to block cAMP degradation in the oocyte produced normal morphological signs of nuclear stagnation of GV oocytes for 72 h. The oocytes had a SN constitution (karyosphere) at the GV, and the nuclear membrane remained intact and unfolded. It has been reported that karyosphere formation reflects the state of oocyte nucleus being prepared for ovulation with extinction of transcriptional processes [37, 58]. According to previous studies, this nuclear organization should persist for a period of no less than 2 wk, taking into account the different antral follicle sizes analyzed (8–30 mm). In the present study, PDE 3 inhibitor was efficacious in arresting oocytes for 0–72 h, and the presence of chromatin surrounding the nucleolus as a continuous mass (SN, karyosphere) was observed in almost all arrested oocytes analyzed. Parfenov et al. [58] stated that oocytes with karyosphere may be considered preovulatory, as this is supported by other morphological parameters such as a homogeneous ZP without follicular processes and oocyte microvilli that are poorly expressed. In our study, the karyosphere was detectable at the GV of the oocytes when they were still surrounded by a compact cumulus cell mass. Several transzonal processes onto the oocytes were observed with a heterogeneous zona and microvilli were still present and abundant. A recent study from De La Fuente and Eppig [59] reported that, after gonadotropin stimulation during in vitro culture of murine oocytes, there was an increased proportion of tightly enclosed cumulus-cell oocytes with SN configuration at the GV. In the human oocytes analyzed in this study, similar changes in nuclear conformation might have occurred due to the gonadotropin stimulation.
Prolonged nuclear arrest by PDE 3 inhibitor in our medium seemed to affect the migration of organelles in the GV-stage oocytes. The organelles start to dissociate from the aggregates after 48 h in PDE 3 inhibitor culture. Even in GV-intact oocytes and inhibitor-arrested oocytes, cortical granules formed a characteristic single layer under oolemma, Oocyte maturation could be arrested by Org 9935 administration in vivo without affecting ovarian function and the reproductive cycle in rats [29]. The present study unexpectedly revealed that, after pharmacological arrest of human oocytes for up to 72 h, the meiotic cycle can be reinitiated within the normal time frame for human denuded oocytes (24–30 h) [60] by washing out the PDE 3 inhibitor. Depending on follicle pretreatment conditions, 46.0% to 100.0% of oocytes could normally progress through meiosis with a high proportion of a well-aligned metaphase II plate formation.
In conclusion, this study is unique in that it demonstrates the presence and involvement of PDE 3 in the regulation of meiosis in human oocytes. The evidence presented supports the model in which decreased intraoocyte cAMP is an important trigger for reinitiation of meiosis. We demonstrate that a reversible inhibition of nuclear maturation for up to 72 h can be imposed on human oocytes from immature antral follicles even after hCG administration. Culture-medium conditions applied during nuclear arrest supported oocyte maturation as shown by ultrastructure analysis. Cytoplasmic changes occurred during this period of inhibition while the nuclear membrane remained unfolded. Meiosis reinitiation occurred normally after a nuclear arrest period of 72 h of culture. Further adaptations of culture medium during the period of Org9935-induced arrest might constitute the basis for techniques to improve the developmental competence of the human immature oocytes for infertility treatment.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: D. Nogueira, Follicle Biology Laboratory, Dutch-Speaking Free University of Brussels (Vrije Universiteit Brussel), Laarbeeklaan, 1001, 1090, Brussels Belgium. FAX: 32 2 477 50 60; lrianad{at}az.vub.ac.be
Received: 31 January 2003.
First decision: 18 February 2003.
Accepted: 8 May 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
