In vitro maturation of human preovulatory oocytes reconstructed by germinal vesicle transfer

Transcription

1 FERTILITY AND STERILITY VOL. 71, NO. 4, APRIL 1999 Copyright 1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. In vitro maturation of human preovulatory oocytes reconstructed by germinal vesicle transfer John Zhang, M.D., Ph.D.,* Chia-Woei Wang, M.D.,* Lewis Krey, Ph.D.,* Hui Liu, M.D.,* Li Meng, Ph.D., Anna Blaszczyk, M.S.,* Alexus Adler, B.Sc.,* and Jamie Grifo, M.D., Ph.D.* *Program for In Vitro Fertilization, New York University Medical Center, New York, New York, and Oregon Regional Primate Research Center, Beaverton, Oregon Received October 28, 1998; revised and accepted October 29, General Program Prize Paper at the 53rd Annual Meeting of the American Society for Reproductive Medicine, Cincinnati, Ohio, October 18 22, Reprint requests: John Zhang, M.D., Ph.D., Program for In Vitro Fertilization, Reproductive Surgery, and Infertility, New York University Medical Center, 660 First Avenue, 5th Floor, New York, New York (FAX: ) /99/$20.00 PII S (98) Objective: To describe a micromanipulation-electrofusion procedure for transferring germinal vesicles (GVs) between immature human oocytes. Design: Pilot study to assess oocyte maturation after an invasive micromanipulation procedure. Setting: Research laboratory at a university medical center. Patient(s): Immature oocytes were discarded from intracytoplasmic sperm injection (ICSI)-IVF cycles of patients years of age. Intervention(s): Initially, GV removal and transfer were performed on the same oocyte; these selfreconstructed oocytes were then cultured in vitro for up to 50 hours and examined periodically for maturation as judged by the extrusion of the first polar body. In a second study, GVs from oocytes of old patients ( 38 years old) were successfully transferred into enucleated immature oocytes of young patients ( 31 years old). Main Outcome Measure(s): Extrusion of the first polar body was monitored in reconstructed and control oocytes; karyotypes also were analyzed at meiosis II. Result(s): From 48 oocytes from old patients, 12 GVs were successfully removed, transferred, and fused into previously enucleated oocytes from young patients. After in vitro culture, 7 of these reconstructed oocytes matured to meiosis II, a maturation rate not significantly different from that observed in nonmanipulated controls. A normal, second meiotic metaphase chromosome complement was observed in 4 of 5 reconstructed oocytes. Conclusion(s): Normal meiosis can occur after the transfer of a GV into an enucleated host oocyte. Germinal vesicle transfer may be a valuable research procedure that generates cell models to characterize the cytoplasmic-nuclear interplay for cell cycle regulation, maturation, and fertilization in the human oocyte; it also may be a potentially attractive alternative to oocyte donation. (Fertil Steril 1999;71: by American Society for Reproductive Medicine.) Key Words: Germinal vesicle transfer, in vitro maturation, oocyte reconstruction As in the natural human population, birth rate is related closely to maternal age in infertile patients. In vitro fertilization clinics throughout the United States have reported that the clinical pregnancy rate drops precipitously when the age of the female partner is 40 years; this is paralleled closely by a fall in embryo implantation rate (1). These declines in clinical pregnancy and implantation rates appear to be related primarily to an increase in the incidence of chromosomal aneuploidy in the oocytes and embryos of older women (2 7). This anomaly in chromosome segregation arises from a dysfunctional first and/or second meiosis during final oocyte maturation (8). In the oocytes of laboratory animals, interactions between the nuclear genome and numerous factors in the cytoplasm influence meiosis the reduction in chromosome number from 4N (germinal vesicle [GV] stage) to 2N (metaphase II stage) that is necessary if fertilization is to occur normally (9 12). These interactions ensure that normal meiotic spindles are constructed from microtubules at each meiotic division. The findings of recent studies 726

2 FIGURE 1 Removal of germinal vesicle from a human oocyte. Original magnification, 400. suggest that similar interactions take place in the human oocyte (13 15); cytoplasmic factors initiate and organize the construction of the meiotic spindles during final maturation and after sperm penetration. Moreover, structural abnormalities in spindle formation can be identified by immunocytochemical staining procedures for tubulin and are thought to be largely responsible for an aneuploid result (15, 16). Our understanding of the nuclear cytoplasmic interactions necessary for normal meiosis in oocytes is limited currently by a lack of appropriate cell models to examine the relative contributions of each intracellular compartment. The influences of maternal aging on these mechanisms is impacted similarly. In this report, we describe a cell model to study nuclear cytoplasmic interrelationships and meiosis in human oocytes. This model is generated by a micromanipulation-electrofusion procedure to transfer a GV between immature oocytes retrieved from patients of different ages. We present results on the success rate of this procedure and preliminary observations that such reconstructed oocytes undergo a normal final maturation in vitro. MATERIALS AND METHODS Source of Oocytes Germinal vesicle stage oocytes were obtained from patients (25 42 years old) who were undergoing intracytoplasmic sperm injection (ICSI) after controlled ovarian hyperstimulation; all patients were treated with hcg (5,000 or 10,000 IU) 36 hours before transvaginal oocyte retrieval. These immature oocytes were unsuitable for ICSI; if patient consent was obtained, they were discarded routinely for use in this approved project (NYUMC-IBRA Protocol H 6902). The discarded oocytes were divided into two groups according to patient age: group I, oocytes from young patients ( 31 years of age); and group II, oocytes from old patients ( 38 years of age). Oocyte manipulation, GV transfer, and in vitro maturation usually were initiated within 6 hours after oocyte retrieval. Germinal Vesicle Removal and Transfer In preparation for ICSI, all oocytes were denuded of cumulus cells by repeated pipetting in human tubal fluid (HTF) medium (Irvine Scientific, Santa Ana, CA) containing 80 IU/mL hyaluronidase (Sigma Chemical Co., St. Louis, MO). Oocytes that had a GV were collected in HTF supplemented with 10% (vol/vol) fetal bovine serum (HyClone, Road Logan, UT). The technique for GV removal was modified from that previously described by Meng and coworkers (17). The oocytes were incubated in HTF containing 7.5 g/ml of cytochalasin B (Sigma Chemical Co.) for approximately 1 hour before the GV was removed directly with a small amount of local cytoplasm (karyoplast) with an enucleation pipette (inner diameter, approximately 20 m) inserted through a slit made in the overlying zona pellucida with a sharp needle developed by Tsunoda et al. (18). Alternatively, the enucleation pipette was used to lance the zona pellucida immediately overlying the GV; the pipette FERTILITY & STERILITY 727

3 FIGURE 2 Transfer of the germinal vesicle karyoplast into the perivitelline space of the same oocyte after enucleation. Original magnification, 400. was then withdrawn, and the pressure inside the holding pipette was increased gently to expel the GV-karyoplast through the hole (Fig. 1). With both procedures, the removed GV was surrounded by a thin corona of cytoplasm, which was in turn membrane encapsulated. No effort was made to vary the amount of cytoplasm associated with the GV as has been done with mouse oocytes (19). The GV-karyoplast was inserted subzonally into an enucleated oocyte that was then washed in HTF medium (Fig. 2). Membrane fusion between oocyte and karyoplast was performed by electrofusion at room temperature. Each oocyte was placed in a microfusion chamber consisting of platinum electrodes approximately 5 mm apart (BTX, San Diego, CA) containing a solution of 0.3 M mannitol, 0.1 mm magnesium sulfate, 0.1 mm calcium chloride, and 0.05 mg/ml bovine serum albumin Fraction V (BSA; Sigma Chemical Co.). The oocyte was aligned by exposure to an alternating current pulse of 8 V for 6 seconds; fusion was accomplished by a direct current pulse of 1.36 kv/cm for microseconds with use of an ElectroCell Fusion instrument (BTX 2001, BTX). Routinely, fusion of oocytes and karyoplasts was observed within 15 minutes. After fusion, the reconstructed oocytes were cultured in vitro. In Vitro Maturation Granulosa cells routinely discarded during oocyte donation retrievals were harvested if patient consent was obtained, and they were dispersed by repeated pipetting in HTF medium containing 0.3% trypsin (wt/vol, Sigma Chemical Co.) and 10% fetal calf serum (vol/vol, HyClone). They were washed twice in HTF and then suspended at a concentration of cells/ml Medium 199 (Sigma Chemical Co.) in a central well dish. Oocytes (n 1 5) were cocultured with the granulosa cells for hours at 37 C in 5% CO 2. The medium was supplemented with 10% fetal bovine serum, IU/mL FSH (Metrodin, Serono, Oakville, Ontario, Canada), 35 ng/ml insulin (Sigma Chemical Co.), and 20% pooled human follicular fluid (vol/vol) obtained from oocyte donation cycles with consent. After incubation, oocytes were examined for maturation under an inverted microscope at 24, 36, 42, and 50 hours. Oocytes were considered mature if the first polar body was observed (MII stage). Analysis of Chromosome Number in MII Oocytes Some reconstructed oocytes were analyzed to assess chromosome number after in vitro maturation procedure. These oocytes were incubated in a hypotonic solution of 0.8% sodium citrate (wt/vol) solution supplemented with 0.3 mg/ml of BSA (Sigma Chemical Co.) for 10 minutes and fixed sequentially in methanol/acetic acid/water (5:1:4) and methanol/acetic acid (3:1) for 5 minutes each (20). After gradual air drying, the fixed chromosomes were stained with Giemsa, and their distribution in the metaphase spread was 728 Zhang et al. Human oocytes reconstructed by GV transfer Vol. 71, No. 4, April 1999

4 analyzed with use of a Cytovision 2.21 Imaging System (Applied Imaging, Pittsburgh, PA). Experiment Design In experiment 1, GV removal and transfer were performed on the same oocytes to determine whether this procedure itself compromised oocyte maturation. Additional oocytes were manipulated as for GV removal, but no microsurgery was performed; these oocytes were also matured in vitro to serve as controls. In experiment 2, the GV was removed from oocytes from the old patients and transferred into enucleated oocytes of young patients. After electrofusion, these reconstructed oocytes were cultured in vitro, and nuclear maturation was monitored. Additional old oocytes (group II) that were not manipulated also were matured in vitro to serve as controls. Statistics The 2 test was used to determine statistically significant differences (P.05) between experimental groups. Statistics were performed with SigmaStat (Jandel Scientic, San Rafael, CA). RESULTS Experiment 1 A total of 32 immature oocytes were manipulated. A GV-karyoplast was removed successfully without lysis in 21 oocytes and replaced subzonally in 16 of them. Karyoplast fusion was accomplished in 14 oocytes for a final success rate of 44%. After in vitro coculture with granulosa cells, 9 self-reconstructed oocytes extruded the first polar body within 50 hours a maturation rate of 64%. Similar results were noted for the control oocytes that were not manipulated; 8 of 14 (57%) reached metaphase II stage when cultured under similar conditions. The maturation rate between these two groups was not statistically significant. Experiment 2 The GV was removed successfully in 28 of 47 (60%) of the host oocytes from the young patients and 33 of 60 (55%) of the GV donor oocytes from the old patients. Intact GV karyoplasts were inserted successfully subzonally in 19 of 28 (68%) of the host oocytes; of these, 12 (63%) fused to form a reconstructed oocyte. Thus, the success rate for the procedure was 12 of 60 (20%). After in vitro coculture with granulosa cells, 7 of 12 (58%) of the reconstructed oocytes extruded a polar body. By comparison, the maturation rate (9 of 13 [69%]) for immature, nonmanipulated control oocytes from old patients was not statistically significantly different. Karyotypes were analyzed successfully in five reconstructed oocytes at meiosis II, and four displayed a normal, second meiotic chromosome complement (Fig. 3). The karyotype of seven control oocytes also were analyzed; only two had a normal second meiotic complement, with the FIGURE 3 A chromosome spread showing a normal karyotype from a reconstructed oocyte that matured through the first meiotic division. Original magnification, 400. remainder displaying various abnormalities in the metaphase plate, including no segregation with predivision and hyperploidy with monodyads or dyads (4). DISCUSSION In this study we demonstrate that micromanipulation and electrofusion can be used to transfer germinal vesicles between preovulatory human oocytes in the same developmental stage. Furthermore, these reconstructed oocytes subsequently mature in vitro and appear to experience a morphologically normal first meiotic division to the 2N stage. Such observations provide preliminary support that oocytes reconstructed by GV transfer may be appropriate cell models to investigate the relative contributions of cytoplasmic and nuclear factors for normal meiosis. These cell models also might be useful to identify the intracellular factors and/or mechanisms that underlie the increase in aneuploidy reported in the oocytes of older women. Tarin (13) suggested that the cytoplasm of oocytes from older women may be incompetent to regulate normal meiotic division because of compromised mitochondrial function and a resultant increase in oxidative stress. Battaglia and coworkers (15) and Plachot and Crozet (16) observed abnormalities in the construction of a normal meiotic spindle from cytoplasmic microtubules in aged oocytes. Clearly, important biologic support for both these hypotheses would be provided by observations of a decrease in aneuploidy frequency when the nuclei of oocytes of older women are transferred at the GV stage to enucleated oocytes of young women. In this regard, it is significant that, in experiment 2, we noted a normal second meiotic chromosome complement in 80% of such reconstructed oocytes. However, our data are FERTILITY & STERILITY 729

5 still limited in numbers and prohibit us from concluding that aneuploidy frequency decreases when the GVs from oocytes of older patients mature within enucleated oocytes of younger ones. Future studies are directed at this question. In addition, we plan to analyze the spindle structure in reconstructed oocytes with immunocytochemical probes for tubulin. Such analyses would provide further confirmation that the meiotic division is normal after GV transfer. In the present study, the success rate of each procedure in GV transfer was 40% 60%. These rates can be improved. The micromanipulation procedure for GV-karyoplast removal and transfer is complicated and uses a relatively large bore pipette. This is the most difficult step in the procedure; clearly, there is a learning curve for this micromanipulation technique as there has been for others such as ICSI (21). As a result of recent trials, we have learned to identify the potential problems associated with each procedural step and have improved to a 80% success rate. In these trials we removed the GV by inserting the enucleation pipette into the oocyte through a hole lanced through the zona pellucida. Technical improvements in the micromanipulation tools, such as a double holder to micromanipulate the two oocytes in the same microscopic field, also may increase our proficiency. Electrofusion is still a relatively new procedure in human reproduction, and different pretreatments and electrical parameters may result in higher rates of fusion with fewer potential side effects. In vitro maturation of human oocytes is currently under study in many research and clinical laboratories (22 24). Clearly, the maturation rate of reconstructed oocytes depends on several factors, and it is important for this and other clinical and research studies to identify the optimal in vitro conditions for maturing the human oocyte. Media and metabolic, gonadotropic, and growth factor supplements need to be tested; the role of cumulus granulosa cells also needs to be assessed. However, it is important to note that, in experiments 1 and 2, the maturation rates of reconstructed GV oocytes were similar to those of control oocytes that were handled simply but not otherwise disturbed. These results suggest that this micromanipulation procedure, when successfully completed, has little or no residual effect on maturation to meiosis II. Such results also suggest that culture conditions and/or an oocyte s biologic background are the primary factors that determine its ability to mature in vitro. Thus, it will also be important to study the impact of several treatment parameters, e.g., follicle size at retrieval, presence or absence of hcg treatment, and timing of retrieval after hcg, that may regulate the biologic development of the oocyte before and after the onset of meiosis I. Although devised for research purposes, GV transfer also offers the potential for a unique form of oocyte donation, the donation of an enucleated cytoplast. Although oocyte donation has been used widely to treat female infertility, the oocyte recipient is the biologic, but not the genetic, mother of her child. She could be both, however, if a GV from her oocyte can be normally and reliably matured in vitro within an enucleated oocyte from a young donor. Such a goal also was achieved recently by another micromanipulation procedure, cytoplasmic transfer (25). However, with this technique only a small volume of cytoplasm is transferred, and there are no procedures to verify that the injected cytoplasm has in fact altered the recipient oocytes biochemistry and/or physiology. Recently, cryopreservation banking of ovarian tissue was performed as a future treatment option for patients about to undergo chemotherapy or irradiation (26). Considering that nuclei of primordial follicles in the ovary are in the GV stage, they can be used for transfer as has been done successfully in mice (27). Clearly, such an approach would bypass the long period required to grow primordial follicles to the antral stage. However, before GV transfer can even be considered for donation purposes, additional research must be conducted in four areas. First, additional work is necessary to establish that oocytes reconstructed by GV transfer mature through all stages of meiosis normally. Studies on meiotic spindle morphology and more precise assessments of chromosomal anomalies must be performed after both the first and second meiotic division. Second, the success rate of each aspect of this procedure must be improved. Considering that low numbers of oocytes are retrieved from older women after ovarian hyperstimulation, a procedural success rate of 12% as reported for study 2 is grossly inadequate. As discussed above, these rates should increase to acceptable levels with practice. Third, in vitro maturation of human oocytes must be optimized; currently, the pregnancy rates with in vitro matured eggs are significantly lower than that of conventional IVF with controlled ovarian stimulation (22 24). Finally, we need to obtain information about the interplay of cytoplasmic and nuclear factors that determine cell cycle regulation, maturation, and fertilization of the human oocyte. The reconstructed cell models generated by GV transfer will be valuable in our attempts to obtain this information. References 1. Society for Assisted Reproductive Technology, American Society for Reproductive Medicine. Assisted reproductive technology in the United States and Canada: 1992 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology registry. Fertil Steril 1995;64: Hook E. Rates of chromosomal abnormalities at different maternal ages. Obstet Gynecol 1981;58: Hollander D, Breen JL. Pregnancy in the older gravida: how old is old? Obstet Gynecol Surv 1991;45: Angell RR, Xian J, Keith J, Ledger W, Baird DT. First meiotic division abnormalities in human oocytes: mechanism of trisomy formation. Cytogenet Cell Genet 1994;65: Munne S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates and maternal age are correlated with chromosome abnormalities. Fertil Steril 1995;64: Benavida CA, Kligman I, Munne S. Aneuploidy 16 in human embryos increases significantly with maternal age. Fertil Steril 1996;66: Verlinsky Y, Kuliev A. Preimplantation diagnosis of common aneuploidies in fertile couples of advanced maternal age. Hum Reprod 1996;11: Zhang et al. Human oocytes reconstructed by GV transfer Vol. 71, No. 4, April 1999

6 8. Moore DP, Orr-Weaver TL. Chromosome segregation during meiosis: building an unambivalent bivalent. Curr Top Dev Biol 1998;37: Masui Y, Markert CL. Cytoplasmic control of nuclear behavior during meiotic matuation of frog oocytes. J Exp Zool 1971;177: Hashimoto N, Kishimoto T. Regulation of meiotic metaphase by a cytoplasmic maturation-promoting factor during mouse oocyte maturation. Dev Biol 1988;126: Flood JT, Chillik CF, van Uem JFHM, Iritani A, Hodgen G. Ooplasmic transfusion: prophase germinal vesicle oocytes made developmentally competent by microinjection of metaphase II egg cytoplasm. Fertil Steril 1990;53: Plancha CE, Albertini DF. Hormonal regulation of meiotic maturation in the hamster oocyte involves a cytoskeleton-mediated process. Biol Reprod 1994;51: Tarin JJ. Aetiology of age-associated aneuploidy: a mechanism based on the free radical theory of ageing. Mol Hum Reprod 1995;10: Van Blerkom J, Davis PW, Lee J. ATP content of human oocytes and developmental potential and outcome after in vitro fertilization and embryo transfer. Hum Reprod 1995;10: Battaglia DE, Goodwin P, Klein NA, Soules MR. Influence of maternal age on meiotic spindle in oocytes from naturally cycling women. Hum Reprod 1996;11: Plachot M, Crozet N. Fertilization abnormalities in human in vitro fertilization. Hum Reprod 1992;7: Meng L, Rutledge J, Kidder G, Khamsi F, Armstrong DT. Influence of germinal vesicle on the variance of patterns of protein synthesis of rat oocytes during maturation in vitro. Mol Reprod Devel 1996;43: Tsunoda Y, Tokunaga T, Imai H, Uchida T. Nuclear transplantation of male primordial germ cells in the mouse. Development 1989;107: Takeuchi T, Ergun B, Rosenwaks Z, Palermo GD. The bearing of the karyo-cytoplasmic ratio on the maturation of germinal vesicles cultured in vitro (abstract P-114). Hum Reprod 1998;13: Kamiguchi Y, Rosenbusch B, Sterzik K, Mikamo K. Chromosomal analysis of unfertilized human oocytes prepared by a gradual fixationair drying method. Hum Genet 1993;90: Grimbizis G, Vandervorst M, Camus M, Tournaye H, Van Steirteghem A, Devroey P. Intracytoplasmic sperm injection, results in women older than 39, according to age and the number of embryos replaced in selective or non-selective transfers. Hum Reprod 1998;13: Cha KY, Chung HM, Han SY, Yoon TK, Oum KB, Chung MK. Successful in vitro maturation, fertilization and pregnancy by using immature follicular oocytes collected from unstimulated polycystic ovarian syndrome patients (abstract O-044) ASRM annual meeting, Boston, MA. S Russell JB, Knezevich KM, Fabian KF, Dickson JA. Unstimulated immature oocyte retrieval: early versus midfollicular endometrial priming. Fertil Steril 1997;67: Barnes FL, Crombie A, Gardner DK, Kausche A, Lacham-Kaplan O, Suikkari A, et al. Blastocyst development and birth after in vitro maturation of human primary oocytes, intracytoplasmic sperm injection and assisted hatching. Hum Reprod 1995;10: Cohen J, Scott R, Alikani M, Schimmel T, Munne S, Levron J, et al. Ooplasmic transfer in mature human oocytes. Mol Hum Reprod 1998; 4: Gosden RG, Rutherford AJ, Norfolk DR. Ovarian banking for cancer patients: transmission of malignant cells in ovarian grafts. Hum Reprod 1997;12: Czolowska R, Tarkowski AK. First meiosis of early dictyate nuclei from primordial oocytes in mature and activated mouse oocytes. Zygote 1996;4: FERTILITY & STERILITY 731