Failure of Meiotic Competence in Human Oocytes'

BIOLOGY OF REPRODUCTION 50, 1100-1107 (1994)

Failure of Meiotic Competence in Human Oocytes' 3 '4 34 SHALOM BAR-AMI,2' 3 4 EFRAT ZLOTKIN, 3 JOSEPH M. BRANDES, and JOSEPH ITSKOVITZ-ELDOR

The Bruce Rappaport Faculty of Medicine,3 Technion-lsrael Institute of Technology, 31096 Haifa, Israel Department of Obstetrics and Gynecology,4 Rambam Medical Center, 31096 Haifa, Israel ABSTRACT The administration of hCG to women undergoing in vitro fertilization and embryo transfer (IVF/ET) results in the meiotic maturation of cumulus-oocyte complexes (COC). Sometimes oocytes being aspirated for lVF/ET fail to resume meiosis in vivo and even after a subsequent 20-h incubation in vitro and are thus defined as meiotic competence failure (MCF) oocytes. The relationship between the proportion of MCF oocytes and other IVF/ET outcomes was studied over 3 years in 703 tested cycles of 487 women. Women yielding one or more MCF oocytes in at least one menstrual cycle represented 8.6% of this population and were defined as MCF women. Cumulus state in the MCF oocyte population was characterized as mature in 57.4 ± 6.7%, intermediate in 13.9 ± 4.0%, immature in 24.1 ± 8.7%, and atretic in 4.6 ± 2.7%. These values differed significantly, by 0.6-, 2.9-, 7.1-, and 4.6-fold, respectively, as compared to the corresponding COC aspirated from women yielding only meiotically competent (MC) oocytes. In a menstrual cycle yielding both MC and MCF oocytes, the IVF/ET variables were evaluated in the MC oocytes. Thus, in such cases the incidence of fertilization or cleavage and the number of blastomeres per embryo were significantly reduced concomitant with the increase in percentage of MCF oocytes. When the percentage of MCF oocytes was 25% or more, no pregnancy was achieved. Various follicular parameters and serum 17p-estradiol (E,) and progesterone (P4 ) levels were compared in MC and MCF women over the four days preceding day of aspiration. In women yielding 25% or more MCF oocytes, the normal increase in follicular number was attenuated, and the serum E,/ number of follicles remained significantly high on day of aspiration, as compared to the corresponding values in women yielding only MC oocytes. This study suggests that the increase in percentage of MCF oocytes represents an overall decrease in IVF/ET variables in terms of oocyte and follicular functionality.

INTRODUCTION The meiotic division of the mammalian oocyte is a protracted process subject to several stop/go controls. It starts shortly before or after birth and is arrested at the diplotene stage, when the nucleus and nuclear membrane reappear. After the ovulatory surge of LH, the meiotic process resumes. Meiotic resumption can also be induced by removing the oocytes from the Graafian follicles of sexually mature animals and culturing them in vitro. However, failure of the oocyte to resume meiosis spontaneously was described in the mouse about 20 years ago [1]. This study was followed by numerous other studies reporting the same phenomenon in many mammalian species [2-4]. It appeared that meiotic competence was acquired concurrently with increase in oocyte size [3, 5, 6], formation of perinucleolar encapsulation [7], formation of the antral cavity [5], and increase in follicle size [8]. Furthermore, in the rat, oocytes failed to acquire meiotic competence in hypophysectomized animals, but the effect of hypophysectomy was abolished by treatment with FSH and/or 17,3-estradiol (E2), suggesting hormonal regulation of this process [2,6,9]. In a different series of studies, it was observed that in certain strains of mice such as the random-bred NMRI/Han, Accepted December 30, 1993. Received June 24, 1993. 'This research was supported by Grant 184-211 (to S.B.-A, J.M.B., and J.I.-E.) from the Technion V.P.R Fund-Hedson Fund for Medical Research. This paper is submitted in partial fulfillment of the requirements for the D.Sc. degree of E.Z. at the Technion-lsrael Institute of Technology. ZCorrespondence: Dr. Shalom Bar-Ami, Bruce Rappaport Faculty of Medicine, Technion-lsrael Institute of Technology, P.O.B. 9649, 31096 Haifa, Israel. FAX: 9724-532102.

about 1% of the ovulated oocytes are diploid. Usually these oocytes are arrested at metaphase of the first meiotic division [10, 11]. Furthermore, at induction of superovulation, the rate of diploid oocytes in the NMRI/Han mouse increases up to 2-4% of the total ovulated oocytes [11-13]. Women undergoing in vitro fertilization and embryo transfer (IVF/ET) are subjected to various treatment protocols aimed at inducing multiple follicular growth. Oocyte meiotic maturation is induced by hCG. Usually the resumption of meiosis and achievement of second metaphase take place within 18 and 28-38 h, respectively, following the LH surge or in vitro culture [14, 15]. Although the women are exposed to hCG for 36 h, a length of time sufficient to induce the ovulatory processes [16], sometimes the aspirated oocytes fail to resume meiosis in vivo and even after a subsequent 20-h incubation in vitro. Thus, these oocytes have been called meiotic competence failure (MCF) oocytes. In view of the MCF that has been observed in women, the present study was initiated to correlate other follicular and IVF/ET parameters in women yielding germinal vesicle (GV) oocytes, i.e., to evaluate whether an increase in the percentage of MCF oocytes would still indicate only the failure of these specific oocytes or might manifest a general failure of ovarian functions. MATERIALS AND METHODS Patients Participants in this study were regularly menstruating women who experienced infertility in the majority of cases 1100

FAILURE OF MEIOTIC COMPETENCE IN HUMAN OOCYTES

due to absent or blocked fallopian tubes and who were undergoing IVF/ET treatment at the Rambam Medical Center, Haifa, Israel, during the years 1984-1987. Multiple follicular growth was induced by various combinations of daily treatment with human FSH (Metrodin; Teva, Petah Tiqva, Israel) and human menopausal gonadotropin, 150 IU/day (Pergonal; Teva). The ovulation processes were induced by administration of 10 000 IU hCG (Chorigon; Teva) when two or more follicles had achieved a diameter of 16-17 mm or more. About 36 h after administration of hCG, the follicular contents were collected by laparoscopy, under general anesthesia, and sometimes by transvaginal ultrasonic probe. The day of hCG administration was designated Day 0. In this study, 703 menstrual cycles of 487 women aged 25-38 yr were investigated. Of this group, 58 menstrual cycles of 42 women yielded one or more MCF oocytes. In this cohort of women, the spermatogram values of the husband were compatible with values designated as normal by the World Health Organization [17]. Monitoring FollicularGrowth and Serum Steroid Level

Follicular growth was monitored by ultrasound. Serum E2 and progesterone (P4) levels were measured by ZER Science Based Industries Ltd., Jerusalem, Israel. The E2 RIA was conducted with a kit from Isodan Diagnostic Laboratories (Jerusalem, Israel) that included 12 5I-labeled E2 as the tracer hormone. The antibodies directed against E2 are very specific and have very low cross-reactivity with other steroids measured at 50% displacement compared to E2. For the various testicular androgens the cross-reactivity did not exceed 0.005%; for estriol and estrone it was < 0.5% and < 11.5%, respectively, and for P4 it was < 0.001%. P4 was measured by means of a kit from ICN Biomedicals (Carson, CA), with 12 5I-labeled P4 as the tracer hormone. The antibodies directed against P4 are also very specific and have very low cross-reactivity with other steroids. For testosterone the crossreactivity was 0.08%; for cortisol, 0.03%; and for pregnenolone, 0.12%. The sensitivities of the assays for E2 and P4 were as low as 12.5 and 150 pg/ml, respectively. MorphologicalAssessment of Cumulus-Oocyte Complexes (COC)

The degree of maturation of COC was defined as described elsewhere [18]. Briefly, COC were assigned to the following categories: immature COC, with tight cumulus mass adherent to the zona pellucida; intermediate COC, with clusters of cumulus cells scattered in the cumulus hyaluronic acid and with partially dispersed corona radiata; and mature COC, with full homogeneous dispersal of each component (corona and cumulus mass) producing a large expanded cumulus mass. Morphology of COC was examined under a light microscope (Diavert, Leitz, Wezlar, Germany) 4 h after aspiration,

1101

and COC were graded for degree of maturation. COC were usually characterized as atretic when they were marked by a compact corona together with various degenerative criteria such as dark and grainy cytoplasm, misshapen oolemma empty or fractured zona pellucida, very large perivitelline space, etc. Medium and Serum Preparation The culture medium consisted of Ham's F-10 powder (Gibco, Grand Island, NY), 9.8 g/L, buffered with sodium bicarbonate (Gibco), 2.1 g/L, achieving a pH of 7.4. This medium was enriched with penicillin G (50 pLg/ml; Sigma Chemical Corp., St. Louis, MO), streptomycin (50 [g/ml; Sigma), calcium lactate (2 mmol/L; Calbiochem-Behring, La Jolla, CA), and human fetal cord serum (10%, v/v). These ingredients were dissolved in purified water (BDH Chemicals, Poole, UK) to achieve 280-290 mOsm/kg. After sterilization through 0.22-,im filters, 0.5-L size (Nalge Co., Rochester, NY), medium was incubated at least overnight at 370C in 5% CO2:95% air with 95% relative humidity before being subjected to routine quality-control evaluation. Human fetal cord serum was prepared as described elsewhere [18]. Oocyte and Sperm Coculture Human COC were collected about 36 h after hCG administration. The complexes were placed immediately in 1 ml Ham's F-10 medium, each single COC in one organculture dish (Becton, Dickinson, Lincoln Park, NJ), and incubated at 37 C in 5% CO2:95% air with 95% relative humidity. Fresh semen obtained by masturbation was divided into 2-ml aliquots. Semen (2 ml) was mixed with Ham's F-10 medium (8 ml) and centrifuged at 200 x g for 10 min. The supernatant was decanted, and the pellet was resuspended in 10 ml Ham's F-10 medium and recentrifuged. After removal of the supernatant, 1 ml of culture medium was added, and the pellet was gently detached from the test tube surface. About 1 h later, the medium containing the motile sperm was collected and counted in a Makler chamber (Sefi Medical Instruments, Haifa, Israel). One hundred thousand motile sperm were added to each culture dish within 4-6 h after recovery of the COC. In cases of an immature or intermediate COC, insemination was performed 30 or 14 h, respectively, after aspiration. Freshly ejaculated sperm were subjected to these preparatory steps to inseminate each type of COC. Seventeen to 20 h after addition of spermatozoa, the oocytes were mechanically denuded of their corona cells for viewing of fertilization, which was marked by the appearance of two pronuclei in the ooplasm. The oocyte or zygote was transferred to a new dish filled with fresh medium. Embryonic cleavage of 2-8-cell stages was observed by 40-48 h postinsemination. The incidence of fertilization and of cleavage was calculated only within the population of meiotically compe-

1102

BAR-AMI ET AL. TABLE 1. Degree of cumulus maturation in relation to meiotic resumption in the corresponding oocytes.a MCF women (%oocytes) Menstrual cycles

Cumulus state

MC women (%oocytes)

Menstrual cycles yielding MC oocytes

Mature Intermediate Immature Atretic

90.7 + 0.7 4.8 0.6 3.4 ± 0.4 1.0 + 0.2

91.2 ±+ 2.1 3.8 ± 1.5 3.3 ± 1.2 1.7 1.0

yielding both MC and MCF oocytes MC oocytesb MCF oocytes 86.0 3.7 6.5 2.9

+ 2.7 + 1.3 + 1.9 + 1.0

57.4 13.9 24.1 4.6

+ 6.7c + 4.6 + 5.7c + 2.7

aDegree of cumulus maturation was designated between 4 and 6 h after aspiration of oocytes. Following 20 h of incubation, oocytes were denuded from the residual cumulus cells and their cytoplasm was tested for MCF, namely, the presence of GV in the ooplasm. bThe relative distribution of the oocytes among the various degrees of cumulus maturation as well as atretic oocytes was calculated for each individual group of MCF or MC oocytes. Cp < 0.01 vs. the corresponding values in MC women.

tent (MC) oocytes. Thus, in a menstrual cycle yielding both

spondence between degree of cumulus maturation and capability of the oocyte to resume meiosis. The percentage of MCF oocytes enclosed in either mature, intermediate, or immature cumulus mass differed by 0.6-, 2.9-, and 7.1-fold, respectively, from the percentage of the corresponding oocytes in the group of women yielding only MC oocytes (Table 1). In the MCF group, 4.6 + 2.7% of the oocytes were designated as atretic, vs. 1.0 + 0.2% in the MC group. In menstrual cycles yielding both MCF and MC oocytes, the distribution of MC oocytes among the different categories was 86.0 + 2.7 for mature, 3.7 + 1.3 for intermediate, 6.5 ± 1.9 for immature, and 2.9 + 1.0 for atretic. These values were not significantly different from the corresponding values observed in menstrual cycles yielding only MC oocytes. Women yielding MCF oocytes (MCF women) were divided into two groups according to the percentage of MCF oocytes out of the total aspirated oocytes in each individual menstrual cycle (Table 2). The mean number of oocytes aspirated per menstrual cycle was lowest in MCF women yielding 25% or more MCF oocytes and highest in MCF women yielding fewer than 25% MCF oocytes (Table 2). In menstrual cycles yielding both MC and MCF oocytes, the

MC and MCF oocytes, the percentage of fertilized or cleaved oocytes was equal to (number of fertilized or cleaved oocytes x 100)/ number of MC oocytes. Only embryos con-

sisting of defined blastomeres and a small number of fragments that did not exceed a reasonable part of the volume within the zona pellucida were transferred to the woman. StatisticalAnalysis Results are expressed as means + SEM. Data were sub-

jected to one-way ANOVA followed by unpaired Student's t-test for individual comparisons between means. P < 0.05 was considered statistically significant. RESULTS Correlationbetween Meiotic Competence and COC Activity

Of the 703 scored menstrual cycles of 487 women, only 58 menstrual cycles of 42 women (8.6%) were found to yield at least one MCF oocyte. Table 1 shows the corre-

TABLE 2. Relationship between percentage of MCF oocytes and outcomes of IVF/ET. a MCF women

IVF/ET parameters

MC women

Mean number of aspirated oocytes/menstrual cycle %of oocytes containing pronuclei %of cleaved embryos No. of blastomeres/embryo Mean no. of transferred embryos %of clinical pregnancies

5.9 76.1 56.7 3.3 2.6

+ 0.16 + 1.0 + 1.2 + 0.05 + 0.3 11

Menstrual cycles yielding MC oocytes 6.5 - 0.5 72.3 3.3 54.8 + 3.9 3.4 + 0.2 3.4 + 0.3 8.6

Menstrual cycles yielding both MC and MCF oocytes (%MCF) 0.05.

25%, F =

various IVF/ET outcomes obviously were evaluated only in the MC oocytes. Thus, increased percentage of MCF oocytes was associated with a significant reduction in percentage of fertilized oocytes (p < 0.05), percentage of cleavage (p < 0.05), and average number of transferred embryos. The ratio of percentage of cleavage to percentage of fertilization represents the percentage of fertilized oocytes in which cleavage took place. This value did not manifest a significant difference between the various groups of women and was in the range of 75-80% (data not shown). In menstrual cycles yielding 25% or more MCF oocytes, in the corresponding MC oocytes that did fertilize and cleave, the number of blastomeres per embryo was significantly lower than in women who yielded only MC oocytes in all their menstrual cycles (Table 2). No pregnancy was recorded in MCF women yielding 25% or more MCF oocytes. In MCF women yielding less than 25% MCF oocytes, the percentage of clinical pregnancies was slightly reduced as compared to those in MC women (i.e., women yielding only MC oocytes in all tested menstrual cycles) (Table 2). When pregnancy was evaluated on a per-embryo basis, zero pregnancy incidence in women yielding 25% or more MCF oocytes was still indicative of failure in the performance of these menstrual cycles. In women whose menstrual cycles yielded either MC oocytes only or less than 25% MCF oocytes, about 4.3% of the transferred embryos developed into fetuses. However, in women whose menstrual cycles yielded 25% or more MCF oocytes, none of the transferred embryos was implanted. To test whether the reduction in IVF/ET outcomes in the MCF women was chronic, other menstrual cycles in which these women yielded only MC oocytes were evaluated. It appeared that the number of aspirated oocytes per menstrual cycle, proportions of fertilization and cleavage,

1104 N

BAR-AMI ET AL. eclg Ou.I r

MCF (%)

MC .

A U-

_

A

N

M~U)

E

Qw

500'

0

L Z j o Wn0 W

cm

350200- F

IL z to D Z cm W,

on -au-

II

I

I

2 40

E

T

_o 200o0o

M LI 0 160IcnW W =m 1204

I I

I

(n

DISCUSSION

?

W z

80

I

I

-1

0

+1

+2

D

Day of hCG administration FIG. 2. Changes in E2 and P4 ratios over total number of follicles during 4 days before oocyte aspiration. Data represent mean + SEM. Each curve was subjected to one-way ANOVA, as follows: Top: In these curves, F was in the range of 15.4-20.5; thus in all curves p < 0.001. Bottom: In MC women, F = 8.2, p < 0.001; in MCF women in menstrual cycles in which the percentage of MCF oocytes was equal to 0, F = 8.1, p < 0.001; < 25%, F = 4.7, p < 0.01;

2

women yielding 25% or more MCF oocytes, serum E2 levels were not significantly changed after hCG administration, whereas in all other groups a significant increase on Day + 1 and a significant decrease on Day +2 were noted (Table 3). Changes in follicular growth and activity were analyzed only in follicles having diameters - 11 mm. The average follicular diameter as well as the number of follicles increased concomitantly with progress of the menstrual cycle from Day -1 to Day +2 (Fig. 1). However, in women yielding 25% or more MCF oocytes, the number of follicles leveled off with progression of the menstrual cycle, and no significant difference was noted by ANOVA between Day -1 and Day +2. Furthermore, serum E2 level per follicle with diameter of - 11 mm significantly decreased on Day +2 in all groups of women analyzed except those yielding 25% or more MCF oocytes (Fig. 2). Thus, no significant difference was noted when this value was analyzed by ANOVA, in contrast to the high significance noted in all other groups (Fig. 2).

25%, F = 0.2, p > 0.05.

and number of blastomeres per embryo were not significantly different from the corresponding values in MC women (Table 2). Correlationbetween Meiotic Competence and Follicular Activity in Terms of Growth and Serum E2 and P4 Levels In the MC women, a maximum level of serum E2 was detected about 10 h after hCG administration (Day + 1); but about 34 h after hCG administration, the serum E2 level was reduced by 40% (Table 3). Serum P4 level increased progressively with the menstrual cycle from Day -1 to Day + 2 in the MC women (Table 3). A similar pattern of changes in E2 and P4 was observed in MCF women at menstrual cycles yielding either MC oocytes only or both MC and MCF oocytes (Table 3). There were no significant differences in absolute serum E2 or serum P4 level between MC women and MCF women yielding 25% or more MCF oocytes (Table 3). In MCF women yielding fewer than 25% MCF oocytes, the serum E2 levels over Days -1 to +2 were significantly higher than in MC women (Table 3). However, in MCF

The present study suggests that in menstrual cycles yielding MC and MCF oocytes, the increase in percentage of MCF oocytes is associated with a significant drop in various VF/ET parameters in the corresponding MC oocytes of these particular menstrual cycles. This evidence may imply that, in those women yielding a high rate of MCF oocytes, the rest of the oocytes, namely, those unmarked by a visible GV, fail to present normal values of IVF/ET parameters. Thus, fertilization, cleavage, number of blastomeres per embryo, and number of transferred embryos and successful pregnancies are reduced. The chance for pregnancy is increased with increasing number of transferred embryos. Nevertheless, transfer of one embryo yielded half of the maximal chance of pregnancy in either natural or stimulated cycles [19]. Therefore, the reduction in percentage of clinical pregnancies in women yielding 25% or more MCF oocytes could be associated at most partially with the decrease in number of transferred embryos. The potential of an oocyte to undergo meiotic maturation, fertilization, and subsequent cleavage is derived from certain biochemical and morphological changes that have occurred in the various compartments of the oocyte and the follicle. In primates and rodents, it has been demonstrated that oocytes may become competent to resume meiosis only when they have acquired a certain size and that further progress in the meiotic process up to the second metaphase stage is also dependent on oocyte growth and rimming [5-7,9]. In primates, oocytes presenting a perinucleolar encapsulation of the GV stage have a greater chance to resume meiosis and progress to second metaphase [7]. Oocyte competence for successful fertilization and the proper organization of the male and female pronuclei

FAILURE OF MEIOTIC COMPETENCE IN HUMAN OOCYTES

in metaphase are both dependent on the oocyte's capacity to proceed with meiotic maturation [20-22] and on oocyte competence for adequate cytoplasmic maturation [23]. In the present study, no difference of this kind was investigated. The oocyte zona pellucida was equally penetrable at all stages of meiosis (from GV to metaphase II). However, there is a significant increase in penetrability of the vitellus with progression of meiotic maturity [20]. In the oocytes that do penetrate at the various steps in meiosis before metaphase II, the completion of meiosis is distorted. Furthermore, embryonic cleavage in such oocytes does not proceed well, and polyploidy and mosaic embryos develop [24]. In menstrual cycles yielding both MC and MCF oocytes, the increase in percentage of MCF oocytes may be associated with failure of the MC oocytes to proceed to metaphase II. If this is the case, it could provide an explanation for the observed reduction in oocyte activity in terms of fertilization and cleavage in the MCF women. Resumption of meiotic maturation and completion of meiotic division after entry of the fertilizing spermatozoon involves several processes, including reduction in cAMP level of the oocyte cytosol, reduction in level of cytostatic factor (CSF), and an increase in meiosis-promoting factor (MPF). These changes are evidenced in amphibians [24-26] as well as in various mammalian species [27-31]. Recently several laboratories have identified CSF as the oncogene C-mos [26, 28] and MPF, which promotes the transition from growth arrest into meiosis and mitosis, as containing two components, p3 4 cdc2 kinase and cyclin [32-34]. Whether MCF oo-

cytes manifest a failure to undergo these changes in cytoplasm or whether incompetence is associated with a more fundamental failure is yet to be determined. Some of the developmental changes that occur in the cumulus after hCG/LH stimulation may in some way be regulated by the oocytes. Thus, an increase in the level of cytochrome c P-450 side-chain cleavage (P-4505cc) enzyme in the cumulus cells was noted only when the enclosed oocyte resumed meiosis [35]. Furthermore, cumulus cell proliferation is also suggested as an oocyte-regulating process [36-38], and cumulus expansion and mucification are induced only when the oocyte GV is dissolved [39]. The finding that in MCF oocytes there was an obvious decrease in the percentage of COC undergoing expansion and mucification (Table 1) may have been associated with oocyte failure to resume meiosis. Nevertheless, in the MCF oocyte population, a fraction of COC (57.4 -+ 6.7) underwent cumulus expansion and mucification, although GV was preserved. This may indicate some asynchrony in processes that occur in the various follicular compartments of women yielding MCF oocytes. We must recall that asynchrony in various ovarian processes are well documented in women subjected to various protocols aimed at inducing multiple follicular growth and superovulation as part of IVF/ET treatment [40, 41].

1105

It has been demonstrated that oocytes resuming meiosis in vitro are less fertilizable if they have been denuded of their corresponding cumulus cells [42,43] or if they have been exposed to anticumulus or antizonal antibodies [44,45]. Coculture of the oocytes with cumulus or granulosa cells when meiosis takes place improves their fertilizability and development [46,47]. Thus, follicular cells are a necessary factor in an adequate meiotic maturation for subsequent fertilization and embryonic growth and development. In women yielding MCF oocytes, the possibility exists that the reduction in fertilizability of the MC oocyte is associated not only with incompleteness of oocyte meiotic maturation and/or cytoplasmic maturation, but also with the inability of the follicular cells to endow the oocytes with an adequate environment for later successful fertilization and production of a viable embryo. The reduction in oocyte activity could be associated with a failure in induction of normal follicular growth, and as a consequence the resulting oocytes are not capable of either resuming meiosis or proceeding with meiotic maturation to metaphase II and of giving a normal viable embryo following fertilization. Studying this possibility as a mechanism for various oocyte failures has been carried out previously [23,41]. It has been demonstrated that oocyte capacity for resuming meiosis in either nonstimulated human oocytes [8] or primate oocytes [7] increases with progress in follicular growth. A correlation between follicular growth and acquisition of meiotic competence has been described in other mammals as well [2,3]. A correlation between follicular growth and the maturation and health of the COC has also been shown in stimulated cycles in humans [48]. Schramm et al. [7] recently demonstrated that meiotic competence in primate oocytes increases with both follicular growth and progress in GV perinucleolar encapsulation. The importance of perinucleolar encapsulation as a step toward acquisition of meiotic competence has also been demonstrated in other mammals [7]. Whether the increase in percentage of MCF oocytes represents incomplete perinucleolar encapsulation in the GV or some degree of immaturity in the oocyte should be investigated. Serum E2 and P4 primarily manifest the ovarian products. Thus, changes in pattern of these steroid levels indicate alteration in ovarian steroidogenesis. Table 3 indicates that the pattern of P4 levels within the 4 days until aspiration did not differ significantly between MC and MCF women. Nevertheless, in women yielding < 25% MCF oocytes, the levels of E2 measured in that period were significantly higher than in the corresponding days in the MC women (Table 3). In fact, the number of growing follicles in this selected group was the highest (Fig. 1). Figure 2 indicates that the serum E2 level per number of growing follicles did not manifest any significant difference between the MC women and women yielding less than 25% MCF oocytes. Thus, the higher steroid concentration in serum reflected the presence of more steroid-secreting follicles.

1106

BAR-AMI ET AL.

In earlier observations it appeared that during Days -2 to +2 of the cycle, follicles 11 mm in size could be grouped into one population, reflecting the follicular population, which correlates with other ovarian variables (Oded Kesler, Joseph Itskovitz, and Shalom Bar-Ami, unpublished data). Thus, the number of aspirated COC and follicular performance in terms of serum E2 and P4 levels correlates well with the number of follicles 11 mm. With use of these criteria, it appeared that the increase in average follicular diameter with progression of the menstrual cycle was similar in both MC and MCF women (Fig. 1). However, the increasing number of follicles in this population that was observed in MC women was attenuated in MCF women in menstrual cycles in which the percentage of MCF oocytes was 25% or more. This reduction may explain the decrease in number of aspirated oocytes (Table 2). Interestingly, the change in pattern of increasing number of growing follicles was observed just after administration of hCG. This observation may indicate that, although sizes of follicles prospectively were in the normal range in MC and MCF women, hCG administration resulted in their segregation into different groups. The changes in the steroidogenic pattern during the periovulatory state apparently are due to a change in the activity of various enzymes. Thus, hCG induction of corpus luteum formation is associated with a tremendous increase in the activity of P450scc [49] and 3-hydroxysteroid dehydrogenase [50], which results in a significant increase in the metabolism of cholesterol to P4. On the other hand, there is a significant decrease in the activity of 17o-hydroxylase/ C-17-C-20 lyase, which results in the blockage of progestin metabolism into androgens [51]. Taken together, these processes lead to both a significant increase in P4 and a significant decrease in E2 secretion. In women yielding 25% or more MCF oocytes, there was a significant increase in serum P4 level, which manifested the same pattern as in the MC women. However, in this group of women the serum E2 level did not present a pattern similar to that in the MC women. Thus, serum E2 level was not significantly reduced after hCG administration (Table 3). In addition, Figure 2 (bottom panel) indicates that the reduction in serum E2 level per follicle, which is usually observed 30-36 h after hCG administration, was also attenuated in menstrual cycles in which the percentage of MCF oocytes was 25% or more. This evidence, together with the asynchrony between oocyte meiotic status and cumulus mucification, indicates that luteinization of the ovary was not taking place properly as in the MC women. Whether hCG was administered prematurely or if in these particular menstrual cycles formation of corpora lutea was distorted for other reasons is worth further investigation. Collectively, the findings of the present study indicate that in women yielding 25% or more MCF oocytes there was a significant reduction in follicular growth and activity, i.e., 1) attenuation of the normal increase in follicular num-

ber, 2) insignificant change in serum E2 levels after hCG administration, and 3) lack of expected alteration in the ratio of E2 to number of follicles after hCG administration. We may suggest that in menstrual cycles yielding a high percentage of MCF oocytes there is a dramatic reduction in oocyte activity in terms of 1) percentage of fertilized oocytes, 2) percentage of cleaved oocytes, 3) number of blastomeres per embryo, and 4) number of transferred embryos. This may suggest that the change in percentage of MCF oocytes should be considered an important factor in evaluating other IVF/ET variables. ACKNOWLEDGMENT We would like to thank Ruth Singer for typing and editing the manuscript.

REFERENCES 1. Szybek K In vitro maturation of oocytes from sexually immature mice. J Endocrinol 1972; 54:527-528. 2. Tsafriri A, Bar-Ami S, Lindner HR. Control of the development of meiotic competence and of oocyte maturation in mammals. In: Meier HM, Lindner HR (eds.), Fertilization of the Human Egg In Vitro. Berlin: Springer-Verlag; 1983: 3-17. 3. Wassarman PM. The mammalian ovum. In: Knobil E, Neill JP, Ewing LL,Greenwald GS, Markert CL, Pfatt DW (eds.), The Physiology of Reproduction. New York: Raven Press; 1988: 69-102. 4. Buccione R, Schroeder AC, EppigJJ. Interactions between somatic cells and germ cells throughout mammalian oogenesis. Biol Reprod 1990; 43:543-547. 5. Sorensen RA, Wassarman PM. Relationship between growth and meiotic maturation of mouse oocyte. Dev Biol 1976; 50:531-536. 6. Bar-Ami S, Tsafriri A Acquisition of meiotic competence in the rat: role of gonadotropin and estrogen. Gamete Res 1981; 4:463-472. 7. Schramm RD, Tennier MT, Boatman DE, Bavister BD. Chromatin configuration and meiotic competence of oocytes are related to follicular diameter in nonstimulated Rhesus monkeys. Biol Reprod 1993; 48:349-356. 8. Tsuji K, Sowa M, Nakano R. Relationship between human oocyte maturation and different follicular sizes. Biol Reprod 1985; 32:413-417. 9. Bar-Ami S, Nimrod A, Brodie AMH, Tsafriri A. Role of FSH and estradiol-173 in the development of meiotic competence in rat oocytes. J Steroid Biochem 1983; 19:965-971. 10. Hansmann I, El-Nahass E. Incidence of nondisjunction in mouse oocytes. Cytogenet Cell Genet 1979; 24:115-121. 11. Beermann F, Jenderny J, Hummler E, Hansmann I. Inheritance of a meiosis I error expressed in mouse oocytes and modulated by a maternal factor. Genet Res 1987; 49:239-243. 12. Bartels I, Jenderny J, Hansmann I. Control of meiosis by somatic cells in mice: inheritance of meiosis I error "diploidy" and nondisjunction in sensitive NMRI/ Han oocytes ovulated from chimaeras. Cell Differ 1984; 15:189-194. 13. Beermann F, Hansmann I. Follicular maturation, luteinization and first meiotic division in oocytes after inhibiting mitochondrial function in mice with chloramphenicol. Mutat Res 1986; 160:47-54. 14. Lindner HR, Bar-Ami S, Tsafriri A. Model system for studying oocyte maturation. In: Serio M, Martini L (eds.), Animal Models in Human Reproduction. New York: Raven Press; 1980: 65-85. 15. Seibel MM, Smith DM, Levesque L, Borten M, Taymor ML. The temporal relationship between the luteinizing hormone surge and human oocyte maturation. Am J Obstet Gynecol 1982; 142:568-572. 16. Testart J, Frydman R, Feinstein MC, Thebault A, Roger M, Scholler R Interpretation of plasma luteinizing hormone assay for the collection of mature oocytes from women: definition of a luteinizing hormone surge-initiating rise. Fertil Steril 1981; 36:50-54. 17. World Health Organization. Semen Analysis Techniques. Cambridge, UK: Cambridge University Press; 1987: 27. 18. Bar-Ami S, Gitay-Goren H, Brandes JM. Different morphological and steroidogenic patterns in oocyte/cumulus-corona cell complexes aspirated at in vitro fertilization. Biol Reprod 1989; 41:761-770. 19. Edwards RG, Fishel SB, Cohen J, Fehilly CB, Purdy JM, Slater JM, Steptoe PC, Webster JM. Factors influencing the success of in vitro fertilization for alleviating human infertility. J In Vitro Fert Embryo Transfer 1984; 1:3-22.

FAILURE OF MEIOTIC COMPETENCE IN HUMAN OOCYTES 20. Lopata A, Leung PC. The fertilizability of human oocytes at different stages of meiotic maturation. Ann NY Acad Sci 1988; 541:324-336. 21. Kubiak JZ. Mouse oocytes gradually develop the capacity for activation during the metaphase II arrest. Acta Anat Nippon 1989; 136:537-545. 22. Cuasnicu PS, Bedford JM. Hamster oocyte penetrability during preovulatory maturation. Mol Reprod Dev 1991; 29:72-76. 23. First NL, Leibfried Rutledge ML, Sirad Mk Cytoplasmic control of oocyte maturation and species differences in the development of maturational competence. Prog Clin Biol Res 1988; 267:1-46. 24. Zenzes MT, Belkien L, Bordt J, Kan J, Schneider HPG, Nieschlag E. Cytologic investigation of human in vitro fertilization failures. Fertil Steril 1985; 43:883891. 25. Sadler ES, Mailer JL. Progesterone inhibits adenylate cyclase in Xenopus oocytes: action on the guanine nucleotide regulatory proteins. J Biol Chem 1981; 256:63686373. 26. Sagata N, Oskarsson M, Copeland T, Brumbaugh J, Vande Woude GF. Function of C-mos proto-oncogene product in meiotic maturation in Xenopus oocytes. Nature (Lond) 1988; 335:519-525. 27. Hashimoto N, Kishimoto T. Regulation of meiotic metaphase by a cytoplasmic maturation promoting factor during mouse oocyte maturation. Dev Biol 1988; 126:242-252. 28. Mutter GL, Grills GS, Wolgemuth DJ. Evidence for the involvement of the protooncogene C-mos in mammalian meiotic maturation and possibly very early embryogenesis. EMBO 1988; 7:683-689. 29. Sagata N, Watanabe N, Vande Woude GF, Ikawa Y. The C-mos proto-oncogene product is a cytostatic factor responsible for meiotic arrest in vertebrate eggs. Nature (Lond) 1989; 342:512-518. 30. Wariko PK, Bavister BD. Hypoxanthine and cyclic adenosine 5'-monophosphate maintain meiotic arrest of rhesus monkey oocytes in vitro. Fertil Steril 1989; 151:886-889. 31. Mattioli M, Galeati G, Bacci ML, Barboni B. Changes in maturation-promoting activity in the cytoplasm of pig oocytes throughout maturation. Mol Reprod Dev 1991; 30:119-125. 32. Murray AW. The cell cycle as a cdc2 cycle. Nature (Lond) 1989; 342:14-15. 33. Pines J, Hunter T. Isolation of a human cyclin cDNA: evidence for cyclin mRNA c c2 and protein regulation in the cell cycle and for interaction with p34 d .Cell 1989; 58:833-846. 34. Westendorf ZM, Swenson K, Ruderman JV. The role of cyclin B in meiosis 1.J Cell Biol 1989; 108:1431-1444. 35. Goldschmit D, Kraicer P, Orly J. Periovulatory expression of cholesterol sidechain cleavage cytochrome P-450 in cumulus cells. Endocrinology 1989; 124:369378. 36. Moriya S, Tanaka T, Okada Y, Kikuchi T, Fujimoto S, Ichinoe K Influence of the oocyte upon proliferation, differentiation and steroidogenesis of cultured granulosa cells. Serono Symp Publ Raven Press 1988; 48:146-161.

1107

37. Satoh H, Tanaka T, Takaoka H, Makinoda S, Moriya S, Rujimoto S, Ichinoe K Oocyte factors may be involved in the formation of cumulus oophorus. Serono Symp Publ Raven Press 1988; 48:216-223. 38. Vanderhyden BC, Caron PJ, Buccione R, Eppig . Developmental pattern of the secretion of cumulus expansion enabling factor by mouse oocytes and the role of oocytes in promoting granulosa cell differentiation. Dev Biol 1990; 140:307317. 39. Salustri A, Yanagishita M, Hascall VC. Mouse oocytes regulate hyaluronic acid synthesis and mucification by FSH-stimulated cumulus cells. Dev Biol 1990; 138:2632. 40. Laufer N, Tarlatzis BC, DeCherney AH, Masters JT, Haseltine FP, Maclusky N, Naftolin F. Asynchrony between human cumulus-corona cell complex and oocyte maturation after human menopausal gonadotropin treatment for in vitro fertilization. Fertil Steril 1984; 42:366-372. 41. Sundstrom P, Nilsson BD. Meiotic and cytoplasmic maturation of oocytes collected in stimulated cycles is asynchronous. Hum Reprod 1988; 3:613-619. 42. Sirard MA, Parrish J, Ware CB, Leibfried-Rutledge ML, First NL. The culture of bovine oocytes to obtain developmentally competent embryos. Biol Reprod 1988; 39:546-552. 43. Sirard MA, Bilodeau S. Effect of granulosa cell co-culture on in-vitro meiotic resumption of bovine oocytes. J Reprod Fertil 1990; 89:459-465. 44. Sakai Y, Araki Y, Yamashita T, Kurata S, Oikawa T, Hiroi M, Sendo F. Inhibition of in vitro fertilization by a monoclonal antibody reacting with the zona pellucida of the oviductal egg but not with that of the ovarian egg of the golden hamster. J Reprod Immunol 1988; 14:177-189. 45. TesarikJ, Testart J, Home F. Effects of prolonged administration of anti-cumulus oophorus antibody on reproduction in mice. J Reprod Fertil 1990; 90:605-610. 46. MotlikJ, FulkaJJr. Fertilization of rabbit oocytes co-cultured with granulosa cells. J Reprod Fertil 1981; 63:425-429. 47. Dandekar PV, Martin MC, Glass RH. Maturation of immature oocytes by coculture with granulosa cells. Fertil Steril 1991; 55:95-99. 48. Simonetti S, Veeck LL, Jones HW. Correlation of follicular fluid volume with oocyte morphology from follicles stimulated by human menopausal gonadotropin. Fertil Steril 1985; 44:177-180. 49. Sasano H, Okamoto M, Mason JI, Simpson ER, Mendelson CR, Sasano N, Silverberg SG. Immunolocalization of aromatase 17a-hydroxylase and side-chain cleavage cytochromes P-450 in the human ovary. J Reprod Fertil 1989; 85:163-169. 50. Doody KJ, Lorence MC, Mason JI, Simpson ER. Expression of messenger ribonucleic acid species encoding steroidogenic enzymes in human follicles and corpora lutea throughout the menstrual cycle. J Clin Endocrinol & Metab 1990; 70:1041-1045. 51. Sano Y, Suzuki K, Arai K, Okinaga S, Tamaoki B-I. Changes in enzyme activities related to steroidogenesis in human ovaries during the menstrual cycle. J Clin Endocrinol & Metab 1981; 52:994-1001.