- Email: [email protected]
Enhancement of primate oocyte maturation and fertilization in vitro by inhibin A and activin A*†
FERTILITY AND STERILITY® Copyright
Vol. 66, No.4, October 1996 Printed on acid-free paper in U. S. A
1996 American Society for Reproductive Medicine
Enhancement of primate oocyte maturation and fertilization in vitro by inhibin A and activin A*t Baha M. Alak, Ph.D.t§ Gary D. Smith, Ph.D.tll Teresa K. Woodruff, Ph.D.~**
Richard L. Stouffer, Ph.D.ttt Don P. Wolf, Ph.D.HHt§§
Oregon Regional Primate Research Center, Beaverton and Oregon Health Sciences University, Portland, Oregon; and Genentech Inc., South San Francisco, California
Objective: A role for inhibin and activin in primate oocyte maturation was investigated. Design: The maturation and fertilization of rhesus monkey oocytes recovered from the excised ovaries of nine regularly cycling animals was compared for untreated germinal vesicle (GV)intact controls versus oocytes cultured in the presence of inhibin, activin, inhibin + activin, or in a combination with follistatin. Setting: Nonhuman primates in a research institute environment. Interventions: Bilateral oophorectomy. Main Outcome Measure: Oocyte maturation from germinal vesicle breakdown (GVBD) to metaphase II (MIl) and fertilization. Results: Germinal vesicle breakdown, progression to MIl and fertilization was monitored in oocytes cultured for 48 hours. Activin alone, at an optimum concentration of 100 ng/mL, stimulated GVBD whereas both GVBD and MIl development was enhanced in the presence of inhibin + activin. The latter also accelerated the rate of maturation to MIl. All treatment groups exhibited a higher incidence of GVBD compared with controls. When follistatin was added, the stimulatory effect of activin or activin + inhibin was abolished. Exposure to medium containing inhibin + activin significantly increased the percentage of MIl oocytes that fertilized compared with controls (68% versus 25%, respectively). Conclusions: Inhibin and activin are potent stimulators of primate oocyte maturation, producing mature oocytes in vitro that fertilize with high efficiency. Fertil Steril® 1996;66:646-53 Key Words: Inhibin, activin, oocyte maturation, in vitro fertilization, Macaca mulatta
In each normal menstrual cycle, a cohort of follicles is recruited to grow and develop through the Received January 22, 1996; revised and accepted May 9, 1996. * This work was presented in part at The American Fertility Society 49th Annual Meeting, October 9 to 14, 1993 and American Society of Reproductive Medicine 51st Annual Meeting, October 7 to 12, 1995. t This work was performed as part ofthe National Cooperative Program on Non-Human In Vitro Fertilization and Preimplantation Development and by the National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland through cooperative agreement HD28484; and also supported by grants HD18185, RR00163 from the National Institutes of Health, Bethesda, Maryland and by Genentech Inc., South San Francisco, California; Oregon Regional Primate Research Center Publication No. 2001. :j: Division of Reproductive Sciences, Oregon Regional Primate Research Center.
646
preantral to antral stage by a highly regulated process involving autocrine-paracrine and endocrine mechanisms. In primates, typically only one antral follicle survives in each cycle to ovulate based on its
§ Present address: Department of Obstetrics and Gynecology, The Ohio State University, Columbus, Ohio. II Present address: Department of Obstetrics and Gynecology, The University of Chicago, Chicago, Illinois. ~ Department of Cell Biology, Genentech Inc. ** Present address: Department of Medicine, Northwestern University, Chicago, Illinois. tt Department of Physiology, Oregon Health Sciences University. :j::j: Department of Obstetrics and Gynecology, Oregon Health Sciences University. §§ Reprint requests: Don P. Wolf, Ph.D., Oregon Regional Primate Research Center, 505 NW 185th Avenue, Beaverton, Oregon 97006 (FAX: 503-690-5563; e-mail: [email protected].
AIak et a1. Role of inhibin and activin in oocyte maturation
Fertility and Sterility®
ability to produce steroids and nonsteroidal regulatory peptides. The latter include the dimeric proteins, inhibin and activin, members of the transforming growth factor-,B (TGF-,B) superfamily that were first identified by their opposing effects on the synthesis and release of FSH from pituitary cells in culture (1). The fully grown oocyte of the preovulatory follicle is arrested in prophase I of meiosis until maturation is reinitiated in vivo by the midcycle LH surge or until the oocyte is released from the follicle and undergoes spontaneous maturation in vitro. The regulation of oocyte maturation is not well understood in mammals. Limited, albeit conflicting, studies in the rat model implicate the inhibin-related proteins as regulators of oocyte maturation. For example, TGF,B reportedly accelerates oocyte maturation in both follicle-enclosed and cumulus-enclosed oocytes obtained from preovulatory follicles of pregnant mare serum gonadotropin (PMSG)-stimulated rats (2). Inhibin, on the other hand, inhibited spontaneous maturation of denuded oocytes obtained from immature, unstimulated rats (3). In the same study, follistatin, a binding protein of both inhibin and activin (4), opposed the effects of inhibin, and activin had no influence on oocyte maturation. Further, Itoh et al. (5) demonstrated acceleration of oocyte maturation by activin offollicle-enclosed, cumulus-enclosed, and denuded oocytes obtained from PMSG-stimulated rats. In another report, however, TGF-,B, inhibin, activin, or follistatin did not influence the meiotic maturation offollicle-enclosed oocytes or the spontaneous maturation of cumulus-enclosed oocytes obtained from PMSG-stimulated rats (6). In these studies, germinal vesicle breakdown (GVBD) was used as an indicator of nuclear maturation, but neither maturation to metaphase II (MIl; as indicated by extrusion of a first polar body) nor cytoplasmic maturation, as indicated by the ability of oocytes to fertilize, was monitored. Outcome discrepancies in these studies could be due to the source of oocytes, the stage of follicular maturation, the culture conditions, or the specific forms of the peptides used. Ovarian stimulation with exogenous gonadotropins, in the context ofthe assisted reproductive technologies, represents an effort to override the follicle selection process, thereby promoting the growth of multiple follicles and the recovery of several oocytes per treatment cycle. An alternative to ovarian stimulation involves the recovery of immature, prophase I oocytes followed by their in vitro maturation and IVF (7). Success of in vitro maturation-IVF in primates is confined to a few reports in monkeys (8, 9) and women (10, 11). The production of live young by in vitro maturation-IVF of oocytes would circumvent the need for Vol. 66, No.4, October 1996
ovarian stimulation, solve many problems associated with oocyte donation, and might lead to knowledge that improves IVF and the treatment of infertility. The present study was undertaken to investigate the roles of inhibin and activin in the regulation of oocyte maturation in a primate model, the rhesus monkey. MATERIALS AND METHODS Animals
Adult female rhesus monkeys (age range 5 to 13 years) exhibiting regular menstrual cycles of approximately 28 days were individually caged in temperature-controlled (24°C), light-regulated (12 hours light: 12 hours dark) rooms at the Oregon Regional Primate Research Center. The animals were subjected to bilateral oophorectomy without regard to the stage of the menstrual cycle. Healthy male rhesus monkeys of proven fertility, caged under similar conditions, were used as sperm donors. Ovary Collection and Oocyte Recovery
Ovaries were recovered from anesthetized animals by paramedian pelvic laparotomy as described previously (8). The ovaries were collected in air-buffered TALP media (TALP-HEPES; (12) with heparin (25 IUlmL) , pH 7.4 at 37°C. Within 10 minutes of oophorectomy, the ovaries were sectioned into quarters. All manipulations were done with ovaries submerged in TALP-HEPES at 37°C. Antral follicles (2:500 f.1m) were individually punctured with a 25gauge needle using a dissecting microscope. The released oocyte-cumulus complexes were collected and pooled into 500-f.1L drops of TALP-HEPES under 6 mL ofTALP-equilibrated sterile mineral oil at 37°C. Only oocyte-cumulus complexes with germinal vesicle-intact (visible or obscured) oocytes > 100 f.1m in vitelline diameter, without degenerative signs (vacuoles or organelle withdrawal), and enclosed with two or more layers of cumulus cells were used in this study (Fig. 1A). Oocyte Culture and In Vitro Maturation Evaluation
Within 60 minutes of collection, oocyte-cumulus complexes were assigned randomly to a culture condition and incubated individually in 50-f.1L drops under 5 mL of TALP-equilibrated, sterile mineral oil in a humidified atmosphere of 5% CO 2 in air at 37°C. The oocyte-cumulus complexes were evaluated in a blinded fashion with regard to treatment conditions using an inverted microscope with Hoffman optics (X400 at 6, 12, 20, 24, 30, 36, and 48 hours) for cumulus expansion and mucification (13), perivitel-
Alak et al. Role of inhibin and activin in oocyte maturation
647
assess the stage of maturation. If cumulus expansion did not occur, this process was postponed for 24 hours. Disappearance of the germinal vesicle was the outcome measure used for GVBD stage (Fig. 1B), whereas germinal vesicle absence and first polar body presence were the markers for nuclear maturation to MIl stage (Fig. 1C). Semen Preparation and IVF
Semen was collected by penile electroejaculation and processed for insemination as described previously (14). Oocytes that completed nuclear maturation to MIl were transferred individually to 100ILL drops of bicarbonated TALP culture medium under oil and inseminated with 200,000 activated-capacitated sperm. At 10 to 16 hours after insemination, oocytes were examined for evidence of fertilization, i.e., two polar bodies and two pronuclei. Fertilized oocytes were cultured in vitro for evidence of timely cleavage and subsequently cryopreserved (15). Peptides, Reagents, and Culture Medium
Figure 1 Photomicrographs of representative rhesus monkey oocytes showing the sequence of events during in vitro maturation. (A), Germinal vesicle-intact oocyte enclosed in a dense layer of cumulus cells. (B), Oocyte after GVBD with expansion and mucification of cumulus cells (12 to 24 hours of culture). (e), A Mil stage oocyte (24 to 30 hours of culture) showing a first polar body. Hoffman optics, x400 optical magnification.
line space dimension, and germinal vesicle and/or first polar body presence or absence. If, during the 12-hour evaluation period, cumulus expansion of +3 or +4 degree (13) was observed, gentle pipetting was used to dissociate oocytes from the cumulus to better 648
The TALP culture medium and air-buffered TALP (TALP-HEPES) were made fresh using reagent grade chemicals including 0.3% bovine serum albumin (BSA) fraction V (A9647, 96% to 99% albumin; Sigma, St. Louis, MO) and gentamicin (Sigma) at pH 7.4 and an osmolality of 285 to 295 mOsm (12). Fetal calf serum (FCS) was obtained from Hyclone (Logan, UT). Recombinant human inhibin A (a(3A) and recombinant human activin A ((3 A(3 A) were provided by Genentech, Inc. (South San Francisco, CA). Recombinant human follistatin (Genentech) was derived from Chinese hamster ovary cells transfected with the full-length mammalian long-form follistatin complementary DNA and, therefore, likely was composed of several isoforms. The hormones were stored at 4°C and diluted fresh for each experiment with 0.15 M NaCl and 0.05 M Tris-HCI, pH 7.0. In the experiments evaluating concentration, activin A was added alone or in combination with inhibin A to TALP medium resulting in final concentrations of 0, 50, 100, and 200 ng of each peptide/mL. Subsequent experiments were conducted in TALP with no peptide (control), TALP containing 100 ng/mL activin A or inhibin A, and TALP containing 100 ng/mL of both activin A and inhibin A. Follistatin, when used, was constituted fresh in TALP medium at a final concentration of 100 ng/mL. Statistical Analysis
Ovaries from each animal formed a replicate and experiments were designed with matched sibling
Alak et a1. Role of inhibin and actiuin in oocyte maturation
Fertility and SterilityR
oocytes assigned to each treatment group, within a comparison, to remove animal-to-animal variation. Because of limited numbers of oocytes per animal placed in each treatment, the influence of increasing doses of either activin or inhibin + activin on the proportion of oocytes undergoing GVBD and MIl maturation were compared using X2 analysis. In addition, limited numbers of MIl oocytes available for insemination necessitated the use of X2 to analyze the influence of peptides on the proportion of MIl oocytes that fertilized. The proportions of oocytes undergoing GVBD or maturing to MIl at each time point from 6 to 48 hours were compared between treatment groups using survival analysis, followed by X2 analysis between individual groups at each time point. The influence of the peptides on the extent of oocyte maturation was determined by comparing percentages using one-way analysis of variance for repeated measures followed by StudentN ewman-Keuls tests to determine significant differences among the group means. Percentages were transformed using the arc sine transformation for unequal variance. P values < 0.05 were considered significant. RESULTS
80
A
60 40
--0~
. a
,. a
------.'
"
,,'
,
0
0
80
:::J
200
b,c
B
+-'
~
.•
Activin (ng/mL)
0
OJ
.........
100
50
':.j:j
OJ
......
20
c
10....
.
60
b ••
--.......... 11', c
,
,,
, ,,
•......
,,
.. ~
d
40
•
20 o+------,------~------,-------,
o 50 100 200 Inhibin + Activin (ng/mL)
Serum-Free In Vitro Maturation
In vitro maturation often is conducted in serumcontaining medium supplemented with gonadotropins in an effort to simulate the intrafollicular environment. To conduct definitive experiments with inhibin and activin, a serum-free culture system was first validated using 77 germinal vesicle oocytes recovered from ovaries offour females (i.e., four experiments conducted for 72 hours of culture). In vitro maturation, in the presence of 0.3% BSA as indicated by GVBD, was 56%, and maturation to MIl was 38%, which was at least as efficient as that observed in serum-containing medium (GVBD 42% and 24% maturation to MIl). Culturing oocytes beyond 48 hours did not result in any changes in oocyte in vitro maturation, therefore, all subsequent experiments were conducted in serum-free medium supplemented with BSA and cultures were terminated at 48 hours.
Figure 2 Influence of media containing increasing concentrations of activin (A) or inhibin + activin (B) on in vitro maturation of rhesus monkey oocytes. Oocyte maturation is expressed as a percentage of the total number of germinal vesicle oocytes that underwent GVBD (--.--) or reached MIl (-e-) after 48 hours of culture with the different treatments (mean). Each treatment contained 30 to 49 oocytes. Symbols with similar letters are significantly different (P < 0.05).
of activin alone or in combination with inhibin, a significant increase in GVBD was observed compared with controls (P < 0.05). The maturation of oocytes to MIl was not altered significantly by any dose of activin alone, whereas a significant stimulation was associated with exposure to inhibin + activin (at 100 ng/mL each compared with controls; P < 0.05). Neither 200 ng/mL activin alone nor in combination with inhibin followed the dose-dependency trends and did not significantly influence GVBD or MIl maturation in comparison to controls.
Peptide Concentrations
To ensure that an effective concentration of activin or inhibin + activin was used, experiments were conducted assessing GVBD and development to MIl of oocytes cultured in increasing concentrations of peptide (Fig. 2). Oocyte culture in 50 ng/mL of activin or of activin + inhibin did not alter significantly GVBD compared with the control. However, at 100 ng/mL Vol. 66, No.4, October 1996
Nuclear Maturation-Kinetics
One of the possible outcomes of in vitro maturation in the presence of inhibin and activin is an alteration in the rate of nuclear maturation. Therefore, the influence of these peptides on maturation kinetics was determined by evaluating GVBD (Fig. 3A) and the extrusion of the first polar body (Fig. 3B) of
Alak et a!. Role of inhibin and activin in oocyte maturation
649
relative to only those oocytes that matured. Over all time points, survival analysis showed enhancement in the rate ofGVBD or MIL Inhibin and activin alone had no significant effect, however, their combination tended to accelerate oocyte GVBD (at 12 hours; P = 0.06) over controls. Also, the maturation rate to MIl (at 12 and 20 hours) was enhanced (P < 0.05) in the presence of inhibin + activin over the control values by 37% and 64%, respectively.
A
80
c m > <-'
60
Extent of Nuclear Maturation
To investigate the influence of inhibin and activin on the extent of oocyte in vitro maturation, the percentages of germinal vesicle oocytes placed in culture that underwent GVBD (Fig. 4A) or matured to MIl
40
'#.
100 6 12 18 24 130 36
*
42 48 80
B
100
A
C
60
In
>
80
==
CJ
*
-;J? 0
40 20
60
'#.
0
40 100 20
80
o~-o~
o
__
-
~.-~~~~~~
6 12 18 24 30 36 42 48
individual oocytes at different time points (6 to 48 hours). To compare maturation rates, the results have been normalized and expressed as a percentage 650
60
::E
Time (hours) Figure 3 Effect ofinhibin(lNH, -.-), activin (ACT, -A-), or INH + ACT (each at 100 ng/mL, - 0 - ) on the time course of rhesus monkey oocyte maturation in vitro (control, -e-). Resumption of meiotic maturation is expressed by the proportion of germinal vesicle-intact oocytes that underwent GVBD (A) or reached MIl (B) with each treatment. Results (from six experiments, 25 to 40 total oocytes per treatment with measurements at each time point) were normalized to the meiotically competent oocytes, i.e., oocytes that underwent GVBD or matured to MIL *Significant difference compared with control (CaNT) values; P < 0.05.
*
tfl.
40 20 0
Figure 4 Effect of inhibin (INH), activin (ACT), or INH + ACT (each at 100 ng/mL) on the extent of rhesus monkey oocyte maturation in vitro. Oocyte maturation is expressed as a percentage of the total number of germinal vesicle oocytes that (A) underwent GVBD or (B) reached MIl after 48 hours of culture with the different treatments (mean ::<:: SE). Results are from nine experiments using 40 to 70 germinal vesicle oocytes per treatment with measurements at each time point. The number of oocytes per treatment group is in brackets. *Significant difference compared with control values; P < 0.01.
Alak et al. Role of inhibin and activin in oocyte maturation
Fertility and Sterility®
l
-(Fig. 4B) after 48 hours of incubation in different culture conditions were compared. All treatment groups exhibited a higher incidence of GVBD (inhibin, 56.9% ± 2.9%; activin, 59.7% ± 4.6%; and inhibin + activin, 70.4% ± 7.2%) compared with controls (35.5% ± 5.2%; mean ± SEM; P < 0.01). Moreover, a twofold increase in the percentage of MIl oocytes was observed after exposure to inhibin + activin (55.3% ± 9.8% versus 27.5% ± 3.7% for the controls, P < 0.01). Specificity of the Response
A convenient means of evaluating specificity in the inhibin + activin stimulation was to include follistatin (100 ng/mL) in the culture medium because this glycoprotein is an inhibin and activin-binding protein (4). Five experiments were conducted each involving 20 to 40 oocytes per treatment and the results are expressed as the mean percentage of the total number of germinal vesicle oocytes that underwent GVBD or matured to MIl after 48 hours of culture. Control values (n = 49) were 38.6% ± 12.7% and 23.4% ± 5.0% for GVBD and MIl, respectively. Follistatin alone (n = 30) gave GVBD and MIl levels that were comparable to the control at 40.4% ± 4.3% and 31.8% ± 2.1%, respectively. Follistatin in the presence of activin (39.0% ± 8.6%; n = 23) negated (P < 0.01) the stimulatory effect of activin alone (75.3% ± 10.4%; n = 19) on GVBD. Similarly, follistatin, in the presence of both activin and inhibin, negated (P < 0.01) the stimulatory effect on GVBD (48.6% ± 7.3% [n = 30] versus 66.3% ± 12.9%; [n = 21]). The results for MIl maturation were less dramatic; however, the trend toward increased maturation in the presence of activin + inhibin (42.0% ± 8.9%) was negated by follistatin (23.6% ± 3.3%), which was similar to the control value. The values for activin alone (35.3% ± 6.4%) and activin + follistatin (34.4% ± 4.8%) were similar in this experimental series. Fertility of In Vitro Matured Oocytes
The ultimate criterion for measuring in vitro maturation success is the ability of oocytes to fertilize in vitro and develop into viable embryos. Our results indicate that the f~rtilization rate of oocytes exposed to the inhibin and activin combination (15/22; 68%) is significantly higher (P < 0.01) than that of control oocytes (4/16; 25%). Although exposure to inhibin alone (5/9; 56%) or activin alone (13/27,48%) tended to double fertilization rates in comparison to controls, the limited number of oocytes did not permit statistical significance. Vol. 66, No.4, October 1996
DISCUSSION
This report summarizes one in a series of studies aimed at understanding the mechanisms underlying oocyte maturation in vitro and its regulation in a primate model, the rhesus monkey. These results demonstrate for the first time that inhibin and activin are effective stimulators of primate oocyte maturation, producing mature oocytes in vitro that fertilize with high efficiency and that inhibin and activin do not antagonize each other in these processes. The present experiments were conducted in serum-free medium, thereby eliminating confounding issues associated with serum-bound factors, which may include steroids, gonadotropins, inhibin, activin, and other growth factors and binding proteins such as follistatin. Bovine serum albumin, as a fraction V preparation, was used and, although it does not provide a chemically defined environment, it does eliminate many serum components. The inhibin, activin, and follistatin preparations were purified recombinant human proteins. The use of sibling oocytes further strengthens the significance of our results as individual animal variability largely is eliminated. In this study, oocyte nuclear maturation (an average of 49% MIl) levels observed for activin-, inhibin-, or activin + inhibin- (Fig. 4B) treated oocytes exceeded those reported previously (37%) in the macaque by Morgan and coworkers (16), or those obtained in this laboratory (10; 40% for oocytes obtained from unstimulated animals and cultured in TALP-FCS medium). The specificity ofthe activin effect on oocyte maturation was corroborated by concentration-dependent effects at the GVBD level and neutralization of the peptides effects on oocyte maturation by follistatin, a binding protein for both inhibin and activin (4). The neutralization effect was evident at the GVBD level. Less evident, however, was follistatin neutralization of the peptide's effect at the MIl level, which could be attributed to artificial error caused by limited numbers of MIl oocytes. Activin and inhibin typically oppose each other's action in diverse gonadal and extragonadal cell types (17, 18). This report demonstrates, for the first time, that these peptides do not oppose each other in oocyte maturation. Two explanations that might account for this effect of inhibin and activin include the existence of a receptor in oocyte-cumulus complexes or oocytes that are specific for only the common f3 subunit in the dimeric proteins or the monomeric f3 subunit that could result from protein dissociation in vitro. Recently, the expression of subtype II activin receptor (ActRII) in the rat oocyte was reported (19). Another possible explanation is that one peptide works at the cumulus
Alak et al. Role of inhibin and activin in oocyte maturation
651
~
I
cell level, whereas the other acts on the oocyte. Such a possibility could be tested by examining the effects of these peptides on denuded versus cumulus enclosed oocytes. Because enhancement of GVBD occurs early in the meiotic resumption process, it seems possible that any influence on the rate of in vitro maturation would be cumulus cell mediated. A third possible explanation is that the exogenous peptides may have altered the dynamics by which the receptors and/or the messenger RNAs (mRNAs) are expressed in the different compartments of the ovarian follicle; i.e., theca, granulosa, and cumulus cells and the oocyte. As such, the normal signaling pathways could be altered with resultant changes in somatic cells and oocyte function including oocyte maturation. Using immunocytochemistry and in situ hybridization, inhibin and activin subunits (a, fJA and fJB) and their mRNAs have been localized in granulosa cells ofthe murine and primate follicles (20-22) as well as in the cytoplasm of one-cell and two-cell embryos (23). Similarly, follistatin mRNA is expressed in granulosa cells, and its protein is present in the follicular fluid of many species (24, 25). This work collectively indicates that the expression and production of inhibin, activin, and follistatin undergoes dynamic changes during follicular maturation and may be species specific. These observations and our finding that inhibin and activin stimulate oocyte maturation also relate to the question of whether the inhibin and activin effect on the oocyte-cumulus complex is mediated via the cumulus cells or is directly on the oocyte. Preliminary results from our laboratories indicate that recombinant human activin A enhances the maturation of denuded monkey oocytes obtained from excised and gonadotropinstimulated ovaries and that this protein binds to ovulated oocyte-cumulus complexes in the rat model (unpublished data). Furthermore, follistatin in the present study showed a tendency, which was not statistically significant, to negate the activin-inhibin enhancement on macaque oocyte attainment of MIl similar to Tsafriri et al. (6), but unlike 0 et al. (3) data in the rat model. Taken together, inhibin and activin may play an important autocrine-paracrine role in oocyte maturation. Also, a role for follistatin in these processes cannot be excluded and warrants further examination. In vitro maturation normally is monitored by nuclear maturation alone because GVBD and the extrusion of a polar body are easily measured events. However, this study suggests a beneficial effect on cytoplasmic maturation as well because fertilization levels of mature oocytes exposed to inhibin, activin, or the combination of inhibin and activin were elevated over controls. The average value obtained here 652
of 50% exceeds that reported previously for the macaque (17%) by Morgan et al. (16) or as obtained in our laboratory previously for oocytes from nonstimulated ovaries (8) (32%) and is within the range reported by Cha and coworkers (10) in women (32% to 81 %). Moreover, in preliminary trials, the developmental potential of embryos produced after in vitro maturation-IVF of inhibin + activin-treated oocytes as measured by the ability to undergo cleavage has been superior to untreated controls. Thus, the acceleration and enhancement of nuclear maturation, coupled with an improvement in fertilization, supports the conclusion that inhibin + activin exposure enhances the in vitro maturation-IVF capability in primates. The ability to conduct in vitro maturation-IVF on a routine basis in unprimed animals carries significant advantages. In rare or endangered species, this approach provides a mechanism to salvage a genetic contribution from a dead or morbid animal. Also, it could be used in women at risk of infertility secondary to radiation or chemotherapy for ovarian cancer. In the context of the assisted reproductive technologies, the need to conduct ovarian stimulation with exogenous gonadotropin preparations, at significant cost and patient inconvenience, could be minimized and alternatives would be available for the gonadotropin-resistant or sensitive patient. Adequate ovarian stimulation for in vitro maturation-IVF could, perhaps, be achieved with a single gonadotropin injection or by the use of clomiphene citrate. Also, coupled with oocyte cryopreservation, some of the ethical issues concerning the freezing and storage of human embryos could be circumvented. In conclusion, inhibin and activin or their combination may significantly improve the success rate of in vitro maturation-IVF in primates. Further studies to clarify the mechanism of action ofthese regulatory peptides should include attempts at receptor localization in the oocyte-cumulus complex and the detection of inhibin and activin mRNA as well as the activin type II receptor mRNA. Consideration is warranted to extend these studies in women.
Acknowledgments. The authors thank Robert Brenner, Ph.D., and Ov Slayden, Ph.D., for cooperating in animal resource management and utilization; Mary Zelinski-Wooten, Ph.D., for manuscript review and discussion; Mr. William Baughman for surgical performance; Mr. Manfred Alexander and Ms. Cindy Christensen for technical assistance; and Mrs. Carol Gibbins for manuscript preparation. All of the above are affiliated with the Oregon Regional Primate Research Center.
Alak et aI. Role of inhibin and activin in oocyte maturation
REFERENCES 1. Ling N, Ying S-Y, Deno N, Shimasaki S, Esch F, Hotta M, et al. Pituitary FSH is released by a heterodimer of the (3-
Fertility and Sterility®
2.
3. 4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
subunits from the two forms of inhibin. Nature 1986; 321:779-82. Feng P, Catt KJ, Knecht M. Transforming growth factor-,B stimulates meiotic maturation of the rat oocyte. Endocrinology 1988;122:181-6. 0 W-S, Robertson DM, de Krester DM. Inhibin as an oocyte meiotic inhibitor. Mol Cell EndocrinoI1989;62:307-11. Shimonaka M, Inoye S, Shimasaki S, Ling N. Follistatin binds to both activin and inhibin through the common betasubunit. Endocrinology 1991; 128:3313-5. Itoh M, Igarashi M, Yamada K, Hasegawa Y, Seki M, Eto Y, et al. Activin A stimulates meiotic maturation of the rat oocyte in vitro. Biochem Biophys Res Commun 1990; 166:147984. Tsafriri A, Vale W, Hsueh AJW. Effects of transforming growth factors and inhibin-related proteins on rat preovulatory Graafian follicles in vitro. Endocrinology 1989; 125: 1857-62. Eppig JJ, O'Brien MJ. Development in vitro of mouse oocytes from primordial follicles. BioI Reprod 1996; 54: 197 - 207. Alak BM, Wolf DP. Rhesus monkey oocyte maturation and fertilization in vitro: roles of the menstrual cycle and of exogenous gonadotropins. BioI Reprod 1994;51:879-87. Schramm RD, Bavister BD. Follicle-stimulating hormone priming of rhesus monkeys enhances meiotic and developmental competence of oocytes matured in vitro. BioI Reprod 1994;51:904-12. Cha KY, Koo JJ, Ko JJ, Choi DH, Han SY, Yoon TK. Pregnancy after in vitro fertilization of human follicular oocytes collected from nonstimulated cycles, their culture in vitro and their transfer in a donor oocyte program. Fertil Steril 1991; 55:109-13. Trounson A, Wood C, Kausche A. In vitro maturation and the fertilization and developmental competence of oocytes recovered from untreated polycystic ovarian patients. Fertil Steril 1994; 62:353-62. Boatman DE. In vitro growth of non-human primate pre- and peri-implantation embryos. In: Bavister BD. The mammalian preimplantation embryo. New York: Plenum Press, 1987: 273-308. Downs SM. Specificity of epidermal growth factor action on maturation of the murine oocyte and cumulus oophorus in vitro. BioI Reprod 1989;41:371-9.
Vol. 66, No.4, October 1996
14. Bavister BD, Boatman DE, Leibfried L, Loose M, Vernon MW. Fertilization and cleavage of rhesus monkey oocytes in vitro. BioI Reprod 1983;28:983-99. 15. Wolf DP, Vandevoort CA, Meyer-Haas GR, Zelinski-Wooten MB, Hess DL, Baughman WL, et al. In vitro fertilization and embryo transfer in the rhesus monkey. BioI Reprod 1989;41: 335-46. 16. Morgan PM, Warikoo PK, Bavister BD. In vitro maturation of ovarian oocytes from unstimulated rhesus monkeys: assessment of cytoplasmic maturity by embryonic development after in vitro fertilization. BioI Reprod 1991;45:89-93. 17. Mather JP, WoodruffTK, Krummen LA. Paracrine regulation of reproductive function by inhibin and activin. Proc Soc Exp BioI Med 1992;201:1-15. 18. Findlay JK. An update of the roles of inhibin, activin and follistatin as local regulators of folliculogenesis. BioI Reprod 1993;48:15-23. 19. Cameron VA, Nishimura E, Mathews LS, Lewis KA, Sawchenko PE, Vale WW. Hybridization histochemicallocalization of activin receptor subunits in rat brain, pituitary, ovary and testis. Endocrinology 1994; 134:799-808. 20. Woodruff TK, D'Agostino JB, Schwartz NB, Mayo KE. Dynamic changes in inhibin messenger RNAs in rat ovarian follicles during the reproductive cycle. Science 1988; 239: 1269-99. 21. Schwall RH, Mason AJ, Wilcox IN, Bassett SG, Zeleznik AJ. Localization ofinhibinlactivin subunit mRNAs within the primate ovary. Mol Endocrinol 1990;4:75-9. 22. Yamoto M, Minami S, Nakano R, Kobayashi M. Immunohistochemical localization of inhibinlactivin subunits in human ovarian follicles during the menstrual cycle. J Clin Endocrinol Metab 1992;74:989-93. 23. Lu R, Shiota K, Tachi C, Takahoshi M. Histochemical demonstration of activinlinhibin ,BA-,BB and/or a subunits in early embryos and oviducts of different strains of mice. J Reprod Dev 1992;38:79-90. 24. Nakatani A, Shimasaki S, Del'aolo LV, Erikson GF, Ling N. Cyclic changes in follistatin messenger ribonucleic acid and its protein in the rat ovary during the estrous cycle. Endocrinology 1991; 129:603-11. 25. Shukovski L, Zhang ZW, Michel U, Findlay JK. FSH-suppressing protein (FSP) mRNA expression in bovine ovarian tissues. J Reprod Fertil 1992;95:861-7.
Alak et a1.
Role of inhibin and activin in oocyte maturation
653
Vol. 66, No.4, October 1996 Printed on acid-free paper in U. S. A
1996 American Society for Reproductive Medicine
Enhancement of primate oocyte maturation and fertilization in vitro by inhibin A and activin A*t Baha M. Alak, Ph.D.t§ Gary D. Smith, Ph.D.tll Teresa K. Woodruff, Ph.D.~**
Richard L. Stouffer, Ph.D.ttt Don P. Wolf, Ph.D.HHt§§
Oregon Regional Primate Research Center, Beaverton and Oregon Health Sciences University, Portland, Oregon; and Genentech Inc., South San Francisco, California
Objective: A role for inhibin and activin in primate oocyte maturation was investigated. Design: The maturation and fertilization of rhesus monkey oocytes recovered from the excised ovaries of nine regularly cycling animals was compared for untreated germinal vesicle (GV)intact controls versus oocytes cultured in the presence of inhibin, activin, inhibin + activin, or in a combination with follistatin. Setting: Nonhuman primates in a research institute environment. Interventions: Bilateral oophorectomy. Main Outcome Measure: Oocyte maturation from germinal vesicle breakdown (GVBD) to metaphase II (MIl) and fertilization. Results: Germinal vesicle breakdown, progression to MIl and fertilization was monitored in oocytes cultured for 48 hours. Activin alone, at an optimum concentration of 100 ng/mL, stimulated GVBD whereas both GVBD and MIl development was enhanced in the presence of inhibin + activin. The latter also accelerated the rate of maturation to MIl. All treatment groups exhibited a higher incidence of GVBD compared with controls. When follistatin was added, the stimulatory effect of activin or activin + inhibin was abolished. Exposure to medium containing inhibin + activin significantly increased the percentage of MIl oocytes that fertilized compared with controls (68% versus 25%, respectively). Conclusions: Inhibin and activin are potent stimulators of primate oocyte maturation, producing mature oocytes in vitro that fertilize with high efficiency. Fertil Steril® 1996;66:646-53 Key Words: Inhibin, activin, oocyte maturation, in vitro fertilization, Macaca mulatta
In each normal menstrual cycle, a cohort of follicles is recruited to grow and develop through the Received January 22, 1996; revised and accepted May 9, 1996. * This work was presented in part at The American Fertility Society 49th Annual Meeting, October 9 to 14, 1993 and American Society of Reproductive Medicine 51st Annual Meeting, October 7 to 12, 1995. t This work was performed as part ofthe National Cooperative Program on Non-Human In Vitro Fertilization and Preimplantation Development and by the National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland through cooperative agreement HD28484; and also supported by grants HD18185, RR00163 from the National Institutes of Health, Bethesda, Maryland and by Genentech Inc., South San Francisco, California; Oregon Regional Primate Research Center Publication No. 2001. :j: Division of Reproductive Sciences, Oregon Regional Primate Research Center.
646
preantral to antral stage by a highly regulated process involving autocrine-paracrine and endocrine mechanisms. In primates, typically only one antral follicle survives in each cycle to ovulate based on its
§ Present address: Department of Obstetrics and Gynecology, The Ohio State University, Columbus, Ohio. II Present address: Department of Obstetrics and Gynecology, The University of Chicago, Chicago, Illinois. ~ Department of Cell Biology, Genentech Inc. ** Present address: Department of Medicine, Northwestern University, Chicago, Illinois. tt Department of Physiology, Oregon Health Sciences University. :j::j: Department of Obstetrics and Gynecology, Oregon Health Sciences University. §§ Reprint requests: Don P. Wolf, Ph.D., Oregon Regional Primate Research Center, 505 NW 185th Avenue, Beaverton, Oregon 97006 (FAX: 503-690-5563; e-mail: [email protected].
AIak et a1. Role of inhibin and activin in oocyte maturation
Fertility and Sterility®
ability to produce steroids and nonsteroidal regulatory peptides. The latter include the dimeric proteins, inhibin and activin, members of the transforming growth factor-,B (TGF-,B) superfamily that were first identified by their opposing effects on the synthesis and release of FSH from pituitary cells in culture (1). The fully grown oocyte of the preovulatory follicle is arrested in prophase I of meiosis until maturation is reinitiated in vivo by the midcycle LH surge or until the oocyte is released from the follicle and undergoes spontaneous maturation in vitro. The regulation of oocyte maturation is not well understood in mammals. Limited, albeit conflicting, studies in the rat model implicate the inhibin-related proteins as regulators of oocyte maturation. For example, TGF,B reportedly accelerates oocyte maturation in both follicle-enclosed and cumulus-enclosed oocytes obtained from preovulatory follicles of pregnant mare serum gonadotropin (PMSG)-stimulated rats (2). Inhibin, on the other hand, inhibited spontaneous maturation of denuded oocytes obtained from immature, unstimulated rats (3). In the same study, follistatin, a binding protein of both inhibin and activin (4), opposed the effects of inhibin, and activin had no influence on oocyte maturation. Further, Itoh et al. (5) demonstrated acceleration of oocyte maturation by activin offollicle-enclosed, cumulus-enclosed, and denuded oocytes obtained from PMSG-stimulated rats. In another report, however, TGF-,B, inhibin, activin, or follistatin did not influence the meiotic maturation offollicle-enclosed oocytes or the spontaneous maturation of cumulus-enclosed oocytes obtained from PMSG-stimulated rats (6). In these studies, germinal vesicle breakdown (GVBD) was used as an indicator of nuclear maturation, but neither maturation to metaphase II (MIl; as indicated by extrusion of a first polar body) nor cytoplasmic maturation, as indicated by the ability of oocytes to fertilize, was monitored. Outcome discrepancies in these studies could be due to the source of oocytes, the stage of follicular maturation, the culture conditions, or the specific forms of the peptides used. Ovarian stimulation with exogenous gonadotropins, in the context ofthe assisted reproductive technologies, represents an effort to override the follicle selection process, thereby promoting the growth of multiple follicles and the recovery of several oocytes per treatment cycle. An alternative to ovarian stimulation involves the recovery of immature, prophase I oocytes followed by their in vitro maturation and IVF (7). Success of in vitro maturation-IVF in primates is confined to a few reports in monkeys (8, 9) and women (10, 11). The production of live young by in vitro maturation-IVF of oocytes would circumvent the need for Vol. 66, No.4, October 1996
ovarian stimulation, solve many problems associated with oocyte donation, and might lead to knowledge that improves IVF and the treatment of infertility. The present study was undertaken to investigate the roles of inhibin and activin in the regulation of oocyte maturation in a primate model, the rhesus monkey. MATERIALS AND METHODS Animals
Adult female rhesus monkeys (age range 5 to 13 years) exhibiting regular menstrual cycles of approximately 28 days were individually caged in temperature-controlled (24°C), light-regulated (12 hours light: 12 hours dark) rooms at the Oregon Regional Primate Research Center. The animals were subjected to bilateral oophorectomy without regard to the stage of the menstrual cycle. Healthy male rhesus monkeys of proven fertility, caged under similar conditions, were used as sperm donors. Ovary Collection and Oocyte Recovery
Ovaries were recovered from anesthetized animals by paramedian pelvic laparotomy as described previously (8). The ovaries were collected in air-buffered TALP media (TALP-HEPES; (12) with heparin (25 IUlmL) , pH 7.4 at 37°C. Within 10 minutes of oophorectomy, the ovaries were sectioned into quarters. All manipulations were done with ovaries submerged in TALP-HEPES at 37°C. Antral follicles (2:500 f.1m) were individually punctured with a 25gauge needle using a dissecting microscope. The released oocyte-cumulus complexes were collected and pooled into 500-f.1L drops of TALP-HEPES under 6 mL ofTALP-equilibrated sterile mineral oil at 37°C. Only oocyte-cumulus complexes with germinal vesicle-intact (visible or obscured) oocytes > 100 f.1m in vitelline diameter, without degenerative signs (vacuoles or organelle withdrawal), and enclosed with two or more layers of cumulus cells were used in this study (Fig. 1A). Oocyte Culture and In Vitro Maturation Evaluation
Within 60 minutes of collection, oocyte-cumulus complexes were assigned randomly to a culture condition and incubated individually in 50-f.1L drops under 5 mL of TALP-equilibrated, sterile mineral oil in a humidified atmosphere of 5% CO 2 in air at 37°C. The oocyte-cumulus complexes were evaluated in a blinded fashion with regard to treatment conditions using an inverted microscope with Hoffman optics (X400 at 6, 12, 20, 24, 30, 36, and 48 hours) for cumulus expansion and mucification (13), perivitel-
Alak et al. Role of inhibin and activin in oocyte maturation
647
assess the stage of maturation. If cumulus expansion did not occur, this process was postponed for 24 hours. Disappearance of the germinal vesicle was the outcome measure used for GVBD stage (Fig. 1B), whereas germinal vesicle absence and first polar body presence were the markers for nuclear maturation to MIl stage (Fig. 1C). Semen Preparation and IVF
Semen was collected by penile electroejaculation and processed for insemination as described previously (14). Oocytes that completed nuclear maturation to MIl were transferred individually to 100ILL drops of bicarbonated TALP culture medium under oil and inseminated with 200,000 activated-capacitated sperm. At 10 to 16 hours after insemination, oocytes were examined for evidence of fertilization, i.e., two polar bodies and two pronuclei. Fertilized oocytes were cultured in vitro for evidence of timely cleavage and subsequently cryopreserved (15). Peptides, Reagents, and Culture Medium
Figure 1 Photomicrographs of representative rhesus monkey oocytes showing the sequence of events during in vitro maturation. (A), Germinal vesicle-intact oocyte enclosed in a dense layer of cumulus cells. (B), Oocyte after GVBD with expansion and mucification of cumulus cells (12 to 24 hours of culture). (e), A Mil stage oocyte (24 to 30 hours of culture) showing a first polar body. Hoffman optics, x400 optical magnification.
line space dimension, and germinal vesicle and/or first polar body presence or absence. If, during the 12-hour evaluation period, cumulus expansion of +3 or +4 degree (13) was observed, gentle pipetting was used to dissociate oocytes from the cumulus to better 648
The TALP culture medium and air-buffered TALP (TALP-HEPES) were made fresh using reagent grade chemicals including 0.3% bovine serum albumin (BSA) fraction V (A9647, 96% to 99% albumin; Sigma, St. Louis, MO) and gentamicin (Sigma) at pH 7.4 and an osmolality of 285 to 295 mOsm (12). Fetal calf serum (FCS) was obtained from Hyclone (Logan, UT). Recombinant human inhibin A (a(3A) and recombinant human activin A ((3 A(3 A) were provided by Genentech, Inc. (South San Francisco, CA). Recombinant human follistatin (Genentech) was derived from Chinese hamster ovary cells transfected with the full-length mammalian long-form follistatin complementary DNA and, therefore, likely was composed of several isoforms. The hormones were stored at 4°C and diluted fresh for each experiment with 0.15 M NaCl and 0.05 M Tris-HCI, pH 7.0. In the experiments evaluating concentration, activin A was added alone or in combination with inhibin A to TALP medium resulting in final concentrations of 0, 50, 100, and 200 ng of each peptide/mL. Subsequent experiments were conducted in TALP with no peptide (control), TALP containing 100 ng/mL activin A or inhibin A, and TALP containing 100 ng/mL of both activin A and inhibin A. Follistatin, when used, was constituted fresh in TALP medium at a final concentration of 100 ng/mL. Statistical Analysis
Ovaries from each animal formed a replicate and experiments were designed with matched sibling
Alak et a1. Role of inhibin and actiuin in oocyte maturation
Fertility and SterilityR
oocytes assigned to each treatment group, within a comparison, to remove animal-to-animal variation. Because of limited numbers of oocytes per animal placed in each treatment, the influence of increasing doses of either activin or inhibin + activin on the proportion of oocytes undergoing GVBD and MIl maturation were compared using X2 analysis. In addition, limited numbers of MIl oocytes available for insemination necessitated the use of X2 to analyze the influence of peptides on the proportion of MIl oocytes that fertilized. The proportions of oocytes undergoing GVBD or maturing to MIl at each time point from 6 to 48 hours were compared between treatment groups using survival analysis, followed by X2 analysis between individual groups at each time point. The influence of the peptides on the extent of oocyte maturation was determined by comparing percentages using one-way analysis of variance for repeated measures followed by StudentN ewman-Keuls tests to determine significant differences among the group means. Percentages were transformed using the arc sine transformation for unequal variance. P values < 0.05 were considered significant. RESULTS
80
A
60 40
--0~
. a
,. a
------.'
"
,,'
,
0
0
80
:::J
200
b,c
B
+-'
~
.•
Activin (ng/mL)
0
OJ
.........
100
50
':.j:j
OJ
......
20
c
10....
.
60
b ••
--.......... 11', c
,
,,
, ,,
•......
,,
.. ~
d
40
•
20 o+------,------~------,-------,
o 50 100 200 Inhibin + Activin (ng/mL)
Serum-Free In Vitro Maturation
In vitro maturation often is conducted in serumcontaining medium supplemented with gonadotropins in an effort to simulate the intrafollicular environment. To conduct definitive experiments with inhibin and activin, a serum-free culture system was first validated using 77 germinal vesicle oocytes recovered from ovaries offour females (i.e., four experiments conducted for 72 hours of culture). In vitro maturation, in the presence of 0.3% BSA as indicated by GVBD, was 56%, and maturation to MIl was 38%, which was at least as efficient as that observed in serum-containing medium (GVBD 42% and 24% maturation to MIl). Culturing oocytes beyond 48 hours did not result in any changes in oocyte in vitro maturation, therefore, all subsequent experiments were conducted in serum-free medium supplemented with BSA and cultures were terminated at 48 hours.
Figure 2 Influence of media containing increasing concentrations of activin (A) or inhibin + activin (B) on in vitro maturation of rhesus monkey oocytes. Oocyte maturation is expressed as a percentage of the total number of germinal vesicle oocytes that underwent GVBD (--.--) or reached MIl (-e-) after 48 hours of culture with the different treatments (mean). Each treatment contained 30 to 49 oocytes. Symbols with similar letters are significantly different (P < 0.05).
of activin alone or in combination with inhibin, a significant increase in GVBD was observed compared with controls (P < 0.05). The maturation of oocytes to MIl was not altered significantly by any dose of activin alone, whereas a significant stimulation was associated with exposure to inhibin + activin (at 100 ng/mL each compared with controls; P < 0.05). Neither 200 ng/mL activin alone nor in combination with inhibin followed the dose-dependency trends and did not significantly influence GVBD or MIl maturation in comparison to controls.
Peptide Concentrations
To ensure that an effective concentration of activin or inhibin + activin was used, experiments were conducted assessing GVBD and development to MIl of oocytes cultured in increasing concentrations of peptide (Fig. 2). Oocyte culture in 50 ng/mL of activin or of activin + inhibin did not alter significantly GVBD compared with the control. However, at 100 ng/mL Vol. 66, No.4, October 1996
Nuclear Maturation-Kinetics
One of the possible outcomes of in vitro maturation in the presence of inhibin and activin is an alteration in the rate of nuclear maturation. Therefore, the influence of these peptides on maturation kinetics was determined by evaluating GVBD (Fig. 3A) and the extrusion of the first polar body (Fig. 3B) of
Alak et a!. Role of inhibin and activin in oocyte maturation
649
relative to only those oocytes that matured. Over all time points, survival analysis showed enhancement in the rate ofGVBD or MIL Inhibin and activin alone had no significant effect, however, their combination tended to accelerate oocyte GVBD (at 12 hours; P = 0.06) over controls. Also, the maturation rate to MIl (at 12 and 20 hours) was enhanced (P < 0.05) in the presence of inhibin + activin over the control values by 37% and 64%, respectively.
A
80
c m > <-'
60
Extent of Nuclear Maturation
To investigate the influence of inhibin and activin on the extent of oocyte in vitro maturation, the percentages of germinal vesicle oocytes placed in culture that underwent GVBD (Fig. 4A) or matured to MIl
40
'#.
100 6 12 18 24 130 36
*
42 48 80
B
100
A
C
60
In
>
80
==
CJ
*
-;J? 0
40 20
60
'#.
0
40 100 20
80
o~-o~
o
__
-
~.-~~~~~~
6 12 18 24 30 36 42 48
individual oocytes at different time points (6 to 48 hours). To compare maturation rates, the results have been normalized and expressed as a percentage 650
60
::E
Time (hours) Figure 3 Effect ofinhibin(lNH, -.-), activin (ACT, -A-), or INH + ACT (each at 100 ng/mL, - 0 - ) on the time course of rhesus monkey oocyte maturation in vitro (control, -e-). Resumption of meiotic maturation is expressed by the proportion of germinal vesicle-intact oocytes that underwent GVBD (A) or reached MIl (B) with each treatment. Results (from six experiments, 25 to 40 total oocytes per treatment with measurements at each time point) were normalized to the meiotically competent oocytes, i.e., oocytes that underwent GVBD or matured to MIL *Significant difference compared with control (CaNT) values; P < 0.05.
*
tfl.
40 20 0
Figure 4 Effect of inhibin (INH), activin (ACT), or INH + ACT (each at 100 ng/mL) on the extent of rhesus monkey oocyte maturation in vitro. Oocyte maturation is expressed as a percentage of the total number of germinal vesicle oocytes that (A) underwent GVBD or (B) reached MIl after 48 hours of culture with the different treatments (mean ::<:: SE). Results are from nine experiments using 40 to 70 germinal vesicle oocytes per treatment with measurements at each time point. The number of oocytes per treatment group is in brackets. *Significant difference compared with control values; P < 0.01.
Alak et al. Role of inhibin and activin in oocyte maturation
Fertility and Sterility®
l
-(Fig. 4B) after 48 hours of incubation in different culture conditions were compared. All treatment groups exhibited a higher incidence of GVBD (inhibin, 56.9% ± 2.9%; activin, 59.7% ± 4.6%; and inhibin + activin, 70.4% ± 7.2%) compared with controls (35.5% ± 5.2%; mean ± SEM; P < 0.01). Moreover, a twofold increase in the percentage of MIl oocytes was observed after exposure to inhibin + activin (55.3% ± 9.8% versus 27.5% ± 3.7% for the controls, P < 0.01). Specificity of the Response
A convenient means of evaluating specificity in the inhibin + activin stimulation was to include follistatin (100 ng/mL) in the culture medium because this glycoprotein is an inhibin and activin-binding protein (4). Five experiments were conducted each involving 20 to 40 oocytes per treatment and the results are expressed as the mean percentage of the total number of germinal vesicle oocytes that underwent GVBD or matured to MIl after 48 hours of culture. Control values (n = 49) were 38.6% ± 12.7% and 23.4% ± 5.0% for GVBD and MIl, respectively. Follistatin alone (n = 30) gave GVBD and MIl levels that were comparable to the control at 40.4% ± 4.3% and 31.8% ± 2.1%, respectively. Follistatin in the presence of activin (39.0% ± 8.6%; n = 23) negated (P < 0.01) the stimulatory effect of activin alone (75.3% ± 10.4%; n = 19) on GVBD. Similarly, follistatin, in the presence of both activin and inhibin, negated (P < 0.01) the stimulatory effect on GVBD (48.6% ± 7.3% [n = 30] versus 66.3% ± 12.9%; [n = 21]). The results for MIl maturation were less dramatic; however, the trend toward increased maturation in the presence of activin + inhibin (42.0% ± 8.9%) was negated by follistatin (23.6% ± 3.3%), which was similar to the control value. The values for activin alone (35.3% ± 6.4%) and activin + follistatin (34.4% ± 4.8%) were similar in this experimental series. Fertility of In Vitro Matured Oocytes
The ultimate criterion for measuring in vitro maturation success is the ability of oocytes to fertilize in vitro and develop into viable embryos. Our results indicate that the f~rtilization rate of oocytes exposed to the inhibin and activin combination (15/22; 68%) is significantly higher (P < 0.01) than that of control oocytes (4/16; 25%). Although exposure to inhibin alone (5/9; 56%) or activin alone (13/27,48%) tended to double fertilization rates in comparison to controls, the limited number of oocytes did not permit statistical significance. Vol. 66, No.4, October 1996
DISCUSSION
This report summarizes one in a series of studies aimed at understanding the mechanisms underlying oocyte maturation in vitro and its regulation in a primate model, the rhesus monkey. These results demonstrate for the first time that inhibin and activin are effective stimulators of primate oocyte maturation, producing mature oocytes in vitro that fertilize with high efficiency and that inhibin and activin do not antagonize each other in these processes. The present experiments were conducted in serum-free medium, thereby eliminating confounding issues associated with serum-bound factors, which may include steroids, gonadotropins, inhibin, activin, and other growth factors and binding proteins such as follistatin. Bovine serum albumin, as a fraction V preparation, was used and, although it does not provide a chemically defined environment, it does eliminate many serum components. The inhibin, activin, and follistatin preparations were purified recombinant human proteins. The use of sibling oocytes further strengthens the significance of our results as individual animal variability largely is eliminated. In this study, oocyte nuclear maturation (an average of 49% MIl) levels observed for activin-, inhibin-, or activin + inhibin- (Fig. 4B) treated oocytes exceeded those reported previously (37%) in the macaque by Morgan and coworkers (16), or those obtained in this laboratory (10; 40% for oocytes obtained from unstimulated animals and cultured in TALP-FCS medium). The specificity ofthe activin effect on oocyte maturation was corroborated by concentration-dependent effects at the GVBD level and neutralization of the peptides effects on oocyte maturation by follistatin, a binding protein for both inhibin and activin (4). The neutralization effect was evident at the GVBD level. Less evident, however, was follistatin neutralization of the peptide's effect at the MIl level, which could be attributed to artificial error caused by limited numbers of MIl oocytes. Activin and inhibin typically oppose each other's action in diverse gonadal and extragonadal cell types (17, 18). This report demonstrates, for the first time, that these peptides do not oppose each other in oocyte maturation. Two explanations that might account for this effect of inhibin and activin include the existence of a receptor in oocyte-cumulus complexes or oocytes that are specific for only the common f3 subunit in the dimeric proteins or the monomeric f3 subunit that could result from protein dissociation in vitro. Recently, the expression of subtype II activin receptor (ActRII) in the rat oocyte was reported (19). Another possible explanation is that one peptide works at the cumulus
Alak et al. Role of inhibin and activin in oocyte maturation
651
~
I
cell level, whereas the other acts on the oocyte. Such a possibility could be tested by examining the effects of these peptides on denuded versus cumulus enclosed oocytes. Because enhancement of GVBD occurs early in the meiotic resumption process, it seems possible that any influence on the rate of in vitro maturation would be cumulus cell mediated. A third possible explanation is that the exogenous peptides may have altered the dynamics by which the receptors and/or the messenger RNAs (mRNAs) are expressed in the different compartments of the ovarian follicle; i.e., theca, granulosa, and cumulus cells and the oocyte. As such, the normal signaling pathways could be altered with resultant changes in somatic cells and oocyte function including oocyte maturation. Using immunocytochemistry and in situ hybridization, inhibin and activin subunits (a, fJA and fJB) and their mRNAs have been localized in granulosa cells ofthe murine and primate follicles (20-22) as well as in the cytoplasm of one-cell and two-cell embryos (23). Similarly, follistatin mRNA is expressed in granulosa cells, and its protein is present in the follicular fluid of many species (24, 25). This work collectively indicates that the expression and production of inhibin, activin, and follistatin undergoes dynamic changes during follicular maturation and may be species specific. These observations and our finding that inhibin and activin stimulate oocyte maturation also relate to the question of whether the inhibin and activin effect on the oocyte-cumulus complex is mediated via the cumulus cells or is directly on the oocyte. Preliminary results from our laboratories indicate that recombinant human activin A enhances the maturation of denuded monkey oocytes obtained from excised and gonadotropinstimulated ovaries and that this protein binds to ovulated oocyte-cumulus complexes in the rat model (unpublished data). Furthermore, follistatin in the present study showed a tendency, which was not statistically significant, to negate the activin-inhibin enhancement on macaque oocyte attainment of MIl similar to Tsafriri et al. (6), but unlike 0 et al. (3) data in the rat model. Taken together, inhibin and activin may play an important autocrine-paracrine role in oocyte maturation. Also, a role for follistatin in these processes cannot be excluded and warrants further examination. In vitro maturation normally is monitored by nuclear maturation alone because GVBD and the extrusion of a polar body are easily measured events. However, this study suggests a beneficial effect on cytoplasmic maturation as well because fertilization levels of mature oocytes exposed to inhibin, activin, or the combination of inhibin and activin were elevated over controls. The average value obtained here 652
of 50% exceeds that reported previously for the macaque (17%) by Morgan et al. (16) or as obtained in our laboratory previously for oocytes from nonstimulated ovaries (8) (32%) and is within the range reported by Cha and coworkers (10) in women (32% to 81 %). Moreover, in preliminary trials, the developmental potential of embryos produced after in vitro maturation-IVF of inhibin + activin-treated oocytes as measured by the ability to undergo cleavage has been superior to untreated controls. Thus, the acceleration and enhancement of nuclear maturation, coupled with an improvement in fertilization, supports the conclusion that inhibin + activin exposure enhances the in vitro maturation-IVF capability in primates. The ability to conduct in vitro maturation-IVF on a routine basis in unprimed animals carries significant advantages. In rare or endangered species, this approach provides a mechanism to salvage a genetic contribution from a dead or morbid animal. Also, it could be used in women at risk of infertility secondary to radiation or chemotherapy for ovarian cancer. In the context of the assisted reproductive technologies, the need to conduct ovarian stimulation with exogenous gonadotropin preparations, at significant cost and patient inconvenience, could be minimized and alternatives would be available for the gonadotropin-resistant or sensitive patient. Adequate ovarian stimulation for in vitro maturation-IVF could, perhaps, be achieved with a single gonadotropin injection or by the use of clomiphene citrate. Also, coupled with oocyte cryopreservation, some of the ethical issues concerning the freezing and storage of human embryos could be circumvented. In conclusion, inhibin and activin or their combination may significantly improve the success rate of in vitro maturation-IVF in primates. Further studies to clarify the mechanism of action ofthese regulatory peptides should include attempts at receptor localization in the oocyte-cumulus complex and the detection of inhibin and activin mRNA as well as the activin type II receptor mRNA. Consideration is warranted to extend these studies in women.
Acknowledgments. The authors thank Robert Brenner, Ph.D., and Ov Slayden, Ph.D., for cooperating in animal resource management and utilization; Mary Zelinski-Wooten, Ph.D., for manuscript review and discussion; Mr. William Baughman for surgical performance; Mr. Manfred Alexander and Ms. Cindy Christensen for technical assistance; and Mrs. Carol Gibbins for manuscript preparation. All of the above are affiliated with the Oregon Regional Primate Research Center.
Alak et aI. Role of inhibin and activin in oocyte maturation
REFERENCES 1. Ling N, Ying S-Y, Deno N, Shimasaki S, Esch F, Hotta M, et al. Pituitary FSH is released by a heterodimer of the (3-
Fertility and Sterility®
2.
3. 4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
subunits from the two forms of inhibin. Nature 1986; 321:779-82. Feng P, Catt KJ, Knecht M. Transforming growth factor-,B stimulates meiotic maturation of the rat oocyte. Endocrinology 1988;122:181-6. 0 W-S, Robertson DM, de Krester DM. Inhibin as an oocyte meiotic inhibitor. Mol Cell EndocrinoI1989;62:307-11. Shimonaka M, Inoye S, Shimasaki S, Ling N. Follistatin binds to both activin and inhibin through the common betasubunit. Endocrinology 1991; 128:3313-5. Itoh M, Igarashi M, Yamada K, Hasegawa Y, Seki M, Eto Y, et al. Activin A stimulates meiotic maturation of the rat oocyte in vitro. Biochem Biophys Res Commun 1990; 166:147984. Tsafriri A, Vale W, Hsueh AJW. Effects of transforming growth factors and inhibin-related proteins on rat preovulatory Graafian follicles in vitro. Endocrinology 1989; 125: 1857-62. Eppig JJ, O'Brien MJ. Development in vitro of mouse oocytes from primordial follicles. BioI Reprod 1996; 54: 197 - 207. Alak BM, Wolf DP. Rhesus monkey oocyte maturation and fertilization in vitro: roles of the menstrual cycle and of exogenous gonadotropins. BioI Reprod 1994;51:879-87. Schramm RD, Bavister BD. Follicle-stimulating hormone priming of rhesus monkeys enhances meiotic and developmental competence of oocytes matured in vitro. BioI Reprod 1994;51:904-12. Cha KY, Koo JJ, Ko JJ, Choi DH, Han SY, Yoon TK. Pregnancy after in vitro fertilization of human follicular oocytes collected from nonstimulated cycles, their culture in vitro and their transfer in a donor oocyte program. Fertil Steril 1991; 55:109-13. Trounson A, Wood C, Kausche A. In vitro maturation and the fertilization and developmental competence of oocytes recovered from untreated polycystic ovarian patients. Fertil Steril 1994; 62:353-62. Boatman DE. In vitro growth of non-human primate pre- and peri-implantation embryos. In: Bavister BD. The mammalian preimplantation embryo. New York: Plenum Press, 1987: 273-308. Downs SM. Specificity of epidermal growth factor action on maturation of the murine oocyte and cumulus oophorus in vitro. BioI Reprod 1989;41:371-9.
Vol. 66, No.4, October 1996
14. Bavister BD, Boatman DE, Leibfried L, Loose M, Vernon MW. Fertilization and cleavage of rhesus monkey oocytes in vitro. BioI Reprod 1983;28:983-99. 15. Wolf DP, Vandevoort CA, Meyer-Haas GR, Zelinski-Wooten MB, Hess DL, Baughman WL, et al. In vitro fertilization and embryo transfer in the rhesus monkey. BioI Reprod 1989;41: 335-46. 16. Morgan PM, Warikoo PK, Bavister BD. In vitro maturation of ovarian oocytes from unstimulated rhesus monkeys: assessment of cytoplasmic maturity by embryonic development after in vitro fertilization. BioI Reprod 1991;45:89-93. 17. Mather JP, WoodruffTK, Krummen LA. Paracrine regulation of reproductive function by inhibin and activin. Proc Soc Exp BioI Med 1992;201:1-15. 18. Findlay JK. An update of the roles of inhibin, activin and follistatin as local regulators of folliculogenesis. BioI Reprod 1993;48:15-23. 19. Cameron VA, Nishimura E, Mathews LS, Lewis KA, Sawchenko PE, Vale WW. Hybridization histochemicallocalization of activin receptor subunits in rat brain, pituitary, ovary and testis. Endocrinology 1994; 134:799-808. 20. Woodruff TK, D'Agostino JB, Schwartz NB, Mayo KE. Dynamic changes in inhibin messenger RNAs in rat ovarian follicles during the reproductive cycle. Science 1988; 239: 1269-99. 21. Schwall RH, Mason AJ, Wilcox IN, Bassett SG, Zeleznik AJ. Localization ofinhibinlactivin subunit mRNAs within the primate ovary. Mol Endocrinol 1990;4:75-9. 22. Yamoto M, Minami S, Nakano R, Kobayashi M. Immunohistochemical localization of inhibinlactivin subunits in human ovarian follicles during the menstrual cycle. J Clin Endocrinol Metab 1992;74:989-93. 23. Lu R, Shiota K, Tachi C, Takahoshi M. Histochemical demonstration of activinlinhibin ,BA-,BB and/or a subunits in early embryos and oviducts of different strains of mice. J Reprod Dev 1992;38:79-90. 24. Nakatani A, Shimasaki S, Del'aolo LV, Erikson GF, Ling N. Cyclic changes in follistatin messenger ribonucleic acid and its protein in the rat ovary during the estrous cycle. Endocrinology 1991; 129:603-11. 25. Shukovski L, Zhang ZW, Michel U, Findlay JK. FSH-suppressing protein (FSP) mRNA expression in bovine ovarian tissues. J Reprod Fertil 1992;95:861-7.
Alak et a1.
Role of inhibin and activin in oocyte maturation
653