FERTILITY AND STERILITY Copyright
©
Vol. 58, No.4, October 1992
Printed on acid-free paper in U.S.A.
1992 The American Fertility Society
Nuclear degeneration and meiotic aberrations observed in human oocytes matured in vitro: analysis by light microscopyt'
Catherine Racowsky, Ph.D.t Milissa L. Kaufman, B.S. Department of Obstetrics and Gynecology, University of Arizona, Tucson, Arizona
Objective: To apply an improved air-dry procedure for the light microscopic identification of both degenerate and aberrant meiotic configurations in cultured human oocytes. Material and Design: Meiotically immature, normal appearing human oocytes retrieved after oophorectomy were placed into culture for 9 to 46 hours. Subsequently, oocytes were assessed morphologically and then air dried for light microscopic examination of chromatin configurations. Main Outcome Measures: Oocyte chromatin configurations were identified as normal, degenerate, or aberrant and then classified according to meiotic stage. Retrospective analyses were conducted to determine if [1] normal meiotic configurations were associated with morphologically viable oocytes and [2] degenerate or aberrant meiotic configurations were always associated with degenerate oocytes. In addition, for each meiotic stage, the proportion of oocytes exhibiting either degenerate or aberrant chromatin configurations was calculated. Results: Of 101 oocyte chromatin configurations analyzed, 71.3% were normal, 11.9% were degenerate, and 16.8% displayed meiotic aberrations. Retrospective analyses revealed that the majority of both normal and aberrant chromatin configurations were associated with morphologically viable oocytes (93.1 % and 88.2%, respectively), whereas all of the degenerate chromatin configurations were associated with morphologically degenerate oocytes. When assessed by stage, nuclear degeneration was observed exclusively at the germinal vesicle and diakinesis stages, whereas meiotic aberrations occurred most frequently after chromosome condensation. These aberrations were manifested either as clumped metaphase I configurations or as two distinct groups of bivalents that appeared to result from bivalent migration along the meiotic spindle without homologue segregation. Conclusions: Slightly >25% of human oocytes recovered after oophorectomy were incapable of undergoing normal meiotic maturation in culture. The majority of these abnormal oocytes appeared morphologically normal and yet possessed meiotic aberrations. These observations indicate that caution should be taken when using oocytes matured in vitro for application in assisted reproductive technology programs. Fertil Steril 1992;58:750-5 Key Words: Human oocyte, human chromosomes, degenerate chromatin, meiotic aberrations, light microscopy
Received April 20, 1992; revised and accepted June 30, 1992. * Supported in part by grant 15-0803 from the March of Dimes Birth Defects Foundation, White Plains, New York. t Reprint requests: Catherine Racowsky, Ph.D., Department of Obstetrics and Gynecology, College of Medicine, University of Arizona, Tucson, Arizona 85724.
meiosis to metaphase II. However, it since has been shown that as many as 70% of such oocytes recovered after oophorectomy fail to reach this stage of meiotic maturation (2), thereby excluding them from any possibility of normal fertilization. Based on light microscopic analysis of the chromosomes in oocytes from several animal species (3, 4), this incompetence in human oocytes may relate either to degeneration of the nuclear material or to aberration in the meiotic process. In the present in vitro study, we provide evidence for each of these two possibilities. This evidence has
Human oocyte anomalies
Fertility and Sterility
As first demonstrated by Edwards in 1965 (1), meiotically immature human oocytes, once released from the preovulatory follicular environment, will spontaneously undergo germinal vesicle breakdown in vitro and progress through the various stages of
750
Racowsky and Kaufman
been made possible through use of our improved airdry technique for preparing human oocytes for chromosomal analysis by light microscopy. As previously reported (5), this technique allowed for discrete visualization of normal chromosomal configurations and has allowed us to identify further those oocytes that exhibit either degenerate or aberrant meiotic forms. Such identification provides a new dimension to the field of human oocyte chromosomal analysis that, previously, has been dominated by cytogenetic studies of aneuploid mature human oocytes (6) with little regard being given to gross meiotic anomalies.
MATERIALS AND METHODS Collection of Ovarian Specimens and Harvesting, Selection, and Culture of Oocytes
The procedures used for the collection of ovaries after oophorectomy, as well as for the harvesting, selection, and culture of oocytes, were identical to those previously described (5). Thus, oocytes were released from antral follicles ranging in diameter from 2 to 5 mm. Follicles> 5 mm were not punctured to avoid the possible collection of maturing oocytes. Oocytes that appeared morphologically viable with intact cumuli were incubated for various time intervals from 9 to 46 hours under 5% CO 2 in humidified air at 37°C. Ham's F-10 medium was supplemented with 50% heat-inactivated follicular fluid that had been obtained from women undergoing follicular hyperstimulation before in vitro fertilization or gamete intrafallopian transfer procedures, as previously described by Cha et al. (7). Morphological Assessment of Oocytes After Culture
Before preparation of meiotic chromosomal configurations, oocytes were examined morphologically using a Bausch and Lomb dissecting microscope (model no. 620; Bausch and Lomb). Oocytes were classified as viable when the cytoplasm was light brown and uniformly granular, the oolemma was smooth, and the cell had an even, spherical configuration (8). In contrast, oocytes were classified as degenerate when the cytoplasm was darkened and appeared dense, and the oolemma was ruffled and shrunken away from the zona pellucida (ZP) (9). Some nonviable oocytes were also mis-shaped (9) and, in the extreme, had collapsed as discs. Vol. 58, No.4, October 1992
Preparation of Oocytes for Light Microscopy of Meiotic Configurations
Oocytes were prepared for chromosomal analysis as previously described (5). Briefly, cumulus-free oocytes were exposed to protease for 20 to 40 seconds to digest partially the proteins within the ZP and then fixed on a microscope slide with 3:1 (vol/vol) ethanol:acetic acid. After air drying, oocyte chromatin preparations were stained with Wright's stain (5) and subsequently analyzed for meiotic status under bright field illumination using a Zeiss standard compound microscope (model no. RA-38; Carl Zeiss, Inc., New York, NY). Classification of Meiotic Configurations
Oocyte meiotic configurations were classified as normal, degenerating, or aberrant based on previous observations of typical chromatin characteristics both in numerous animal studies (3, 4, 10, 11) and in two previous studies involving human oocytes matured in vitro (1, 5). Subsequently, within each of these classes, the meiotic configurations were subclassified into groups according to the stages of germinal vesicle, diakinesis, metaphase I, and anaphase I to metaphase II. Data Analyses
Retrospective analyses were carried out to determine if normal meiotic configurations always were associated with oocytes classified as morphologically viable after culture. In addition, similar analyses were performed to identify if degenerate and aberrant meiotic forms were associated exclusively with oocytes classified as degenerate after culture. Subsequently, further analyses were conducted to delineate the proportion of oocytes within a specific meiotic group that exhibited either degenerating nuclear chromatin or a meiotic aberration. Photographic Preparation of Meiotic Configurations
Preparations found to display nuclear degeneration or meiotically aberrant forms were photographed using a Nikon Fluophot compound microscope and Kodak Technical Pan 2415 film (Eastman Kodak Company, Rochester, NY) rated at an exposure of 64. The exposed film was developed at 68°C with a working solution of TD-3 (Photographer's Formulary, Missoula, MT). Racowsky and Kaufman
Human oocyte anomalies
751
Table 1 Retrospective Analyses Correlating the Incidence of Normal, Degenerate, and Aberrant Chromosomal Configurations With Oocyte Morphological Appearance After Culture Morphological appearance Chromosomal configuration
Viable
Degenerate
Normal Degenerate Aberrant
93.1 (67)* 0.0 (0) 88.2 (15)
6.9 (5) 100.0 (12) 11.8 (2)
* Values are percents with the number of oocytes in each group in parentheses.
RESULTS Correlation of Meiotic Classification and Morphological Appearance
Of the 101 air-dried preparations analyzed in this study, 71.3% displayed normal chromosomal configurations, 11.9% were degenerate, and 16.8% exhibited meiotic aberrations. Retrospective analysis revealed that most of those oocytes displaying normal chromosomal configurations (93.1 %) had been classified as morphologically viable after culture (Table 1). Conversely, all of those oocytes displaying nuclear degeneration were considered morphologically degenerate after culture. Of interest, however, the majority of oocytes containing aberrant meiotic configurations (88.2%) had been classified as morphologically normal when examined at the end of culture.
Incidence of Nuclear Degeneration and Meiotic Aberrations by Stage Germinal Vesicle Stage
In contrast to the normal germinal vesicle nucleus with a prominent nucleolus and homogeneously dispersed fibrous chromatin strands (5), a minority of immature oocytes (21.9%) contained degenerate chromatin configurations similar to that depicted in Figure 1A. As shown, this degenerate condition was typified by small fragmented chromatin pieces (arrows in Fig. 1A), numerous cytoplasmic vacuoles (small arrowheads) and a more-or-less spherical structure that stained heterogeneously and contained a vacuole (large arrowhead in Fig. 1A). Figure 1B shows an aberrant immature oocyte possessing two germinal vesicles, a condition observed in only 1 (3.1%) ofthe 32 oocytes scored at the germinal vesicle stage. The two nuclei are 23.5 and 25.0 Jl-m in diameter, respectively, which is approximately half the diameter of a normal, single vesicular nucleus (50.0 Jl-m) (5). Interestingly, however, each contains a nucleolus equal in diameter to that contained in a typical germinal vesicle stage oocyte (10 Jl-m). Diakinesis
Figure 2A to C shows a series of atypical meiotic configurations representing progressive stages of degeneration during diakinesis. In our study population, 36.8% of oocytes in diakinesis exhibited some form of nuclear degeneration. One of these oocytes
Figure 1 Immature oocytes exhibiting (A) degenerate chromatin, X667 and (B) two germinal vesicles (arrows), each containing a darkly staining spherical nucleolus, X529. Note that the degenerate germinal vesicle is characterized by small chromatin fragments (arrows) and a heterogenously staining nucleolus (large arrowhead) set against a background of cytoplasmic vacuoles (small arrowheads). 752
Racowsky and Kaufman Human oocyte anomalies
Fertility and Sterility
Figure 2 Progressive stages of degeneration during diakinesis. Note (A) lack of a nucleolus in the presence of stringy, degenerating chromatin, X500; (B) uncharacteristically clumped chromatin aggregates (arrows), some of which are adherent to the nucleolus (arrowhead), X975; and (C) condensed chromatin clusters lacking clear definition and periodically interconnected via chromatin bridges (arrow), X1,300.
also revealed an unusual sequence of chromatin condensation followed by nucleolus dissolution since the nucleolar material had disappeared in this oocyte before any dramatic condensation of the chromatin (Fig. 2A). The other oocytes in this degenerate group contained atypical, degenerating meiotic configurations at later diakinesis. For example, in Figure 2B, although the nucleolus (arrowhead) appears to be functioning normally as an organizing center for chromatin condensation (12), the chromatin appears clumped and is condensed into aggregates much shorter than the strands of normal condensing bivalents (arrows). Figure 2C reveals an oocyte at an even later stage of diakinesis in which extremely condensed chromatin clusters were frequently in-
terconnected via chromatin "bridges" (arrow). These bridges probably represent regions along the chromosomes in which chiasmata formed but subsequently failed to terminalize after completion of synapsis. Metaphase I
Figure 3A shows an aberrant metaphase I configuration characterized by the sticky adherence of bivalents into a tightly compacted clump. Thus, in contrast to the discrete distribution of chromosomes in the typical metaphase I configuration, only a few bivalents can be discerned (arrows). This aberration of metaphase I was observed in 12 of 28 oocytes
Figure 3 Meiotic aberrations after chromatin condensation. (A), Clumped metaphase I configuration showing a sticky appearance and exhibiting only a few clearly defined bivalents (arrows), X3,288. (B), Two segregated groups of chromosomes in which the homologues failed to disjoin. Note the indistinct appearance of one of the groups (arrow) that is apparently undergoing degenerative processes typically characterized by a normal, haploid polar body, X3,143. Vol. 58, No.4, October 1992
Racowsky and Kaufman
Human oocyte anomalies
753
(42.9%) that were at this stage of meiotic maturation. Anaphase I to Metaphase II
Adherence of bivalents during anaphase I resulted in an aberrant chromosomal configuration as shown in Figure 3B, in which the homologues failed to disjoin, thereby resulting in the bivalents themselves migrating to opposite ends of the spindle during telophase I. This phenomenon was observed in 3 of the 22 oocytes (13.6%) that had progressed beyond metaphase I. Interestingly, in 2 of these oocytes, one of the groups of bivalents was undergoing degeneration (arrow in Fig. 3B) in a manner similar to that which occurs in the haploid chromosomes of a normal polar body (5). DISCUSSION
The results of this study show that degenerate chromatin is associated exclusively either with the germinal vesicle stage or with diakinesis, the first overt stage of maturation. Consistent with previous electron microscope analyses (13, 14), the present degenerate immature oocytes revealed cytoplasmic vacuoles, and the nucleus appeared pyknotic, exhibiting heterogenous staining and chromatin fragmentation (Fig. 1A). Such abnormal nuclear and cytoplasmic characteristics clearly are inconsistent with chromatin condensation so that the germinal vesicle degenerates before any overt signs of meiotic resumption. However, several oocytes did exhibit degenerate chromatin configurations at diakinesis (Fig. 2). Although such degeneration may have occurred in vitro, it is possible that abnormal follicular development may have triggered this process in vivo because this degenerate meiotic stage has been associated with follicular atresia (15). Nevertheless, our inclusion of only those oocytes that appeared morphologically normal at the onset of culture argues against this possibility. Although the present results indicate that meiotic aberrations occurred predominantly in maturing oocytes that exhibited full chromosomal condensation' one immature oocyte was found to contain two germinal vesicles. Two previous reports have described binucleate human oocytes (16, 17). This rare aberration has been associated with elevated levels of gonadotropin (16) that, in turn, may cause the formation of multiovular follicles with ensuing oocyte fusion to form binucleate or multinucleate oocytes. There are two reasons why we believe that the present binuclear oocyte did not arise from 00754
Racowsky and Kaufman
Human oocyte anomalies
cyte fusion. First, this oocyte was retrieved from a patient diagnosed with endometriosis, a condition not associated with elevated gonadotropin levels. Second, each of the two nuclei in this oocyte was half the size of a normal, diploid germinal vesicle. Based on these facts, we suggest that this condition resulted from an aberration in the last mitotic division of the oogonium from which this primary oocyte arose so that the S-phase of the cycle was omitted and the cell failed to undergo cytokinesis. Thus, a binucleate oogonium entered prophase I of meiosis to result in two semihaploid nuclei. That this oocyte appeared morphologically normal after 46 hours in culture and yet had failed to resume meiosis suggests that this binucleate condition, although compatible with cellular viability, was incompatible with meiotic maturation. The present findings revealed that the majority of meiotic aberrations occurred either at metaphase I or at some stage subsequent to alignment of homologues on the first metaphase plate. These aberrations fall essentially into two related classes of meiotic configuration: the clumped metaphase I (Fig. 3A) and the two groups of bivalents (Fig. 3B). The clumped metaphase I condition, characterized by adhesion of the chromosomes into a compacted mass (Fig. 3A), accounted for almost one half ofthe metaphase configurations. Interestingly, the sticky appearance of such chromosomes has been noted in previous reports investigating in vitro maturation of both pig (3) and human (9) oocytes, although the cause of this phenomenon has yet to be elucidated. Based on the present observations of clumped metaphase I chromosomes, we speculate that this stickiness may relate to an alteration in the surface structure of the chromatin that either is induced by some intrinsic property of the oocyte itself or after an interaction of the oocyte or its surrounding cumulus cells with some component(s) in the culture medium. Numerous studies have shown that follicle cells provide both nutritional (18) and regulatory (19,20) modulation of oocyte metabolism. The possibility exists, therefore, that a subclass of cumulus complexes may respond abnormally to the culture medium, thereby exhibiting metabolic aberrations that impact on the heterologous supply of nutritional and/or regulatory factors critical for normal oocyte metabolism and maintenance of chromosomal integrity. Although earlier studies documented the degeneration of clumped metaphase I chromosomes in cultured pig (3) and human (9) oocytes, a recent study (4) noted the viable persistence of this meiotic Fertility and Sterility
form in cultured hamster oocytes. In the latter study, it was concluded that the clumped metaphase I configuration may progress to an irregular meiotic form similar to that depicted in Figure 3B, in which the bivalents themselves have migrated to opposite poles of the spindle. Such a meiotic form has been attributed to perturbation in assembly of the meiotic spindle and/or to a disturbance in the interaction between the kinetochore-spindle apparatus during anaphase I (4). An alternative explanation, however, may be that chromosomal stickiness compromised either bivalent alignment along the equatorial plate and/or chiasmatic splitting during anaphase 1. Interestingly, the results of the present study show that segregation of chromosomes into two groups of bivalents occurs in > 10% of oocytes progressing beyond anaphase 1. Yet, to our knowledge, no previous study has documented such a meiotic aberration in the human oocyte. This lack of documentation may be due to the inadequacy of past procedures for preparing human chromosomes for light microscopy. Our own experience has shown that two groups of bivalents easily may be mistaken for normal telophase I or metaphase II stages in oocyte air-dry preparations in which overlapping chromosomes may result from poor spreading during fixation. It remains to be determined whether the relatively high incidence of meiotic aberrations observed in the present study reflects the fact that the majority of oocytes were collected from nondominant and thus from incompletely mature follicles. In addition, it is unknown whether the observed incidence is higher than that which exists in oocytes matured in vivo. Ideally, a study parallel to the present study should be performed using oocytes retrieved from dominant follicles both before and after induction of meiotic resumption in vivo. In reality, however, such oocytes are rarely, if ever, available for experimentation. Acknowledgments. The authors thank Rachelle A. Dermer, B.F.A., Arizona State University, Tempe, Arizona, for preparing the photographs and Sujatha Gunnala, M.D., Southwest Fertility Center, Phoenix, Arizona, for assisting in the procurement of research material.
REFERENCES 1. Edwards RG. Maturation in vitro of human ovarian oocytes.
Lancet 1965;1:926-9.
Vol. 58, No.4, October 1992
2. Lopata A, Leung PC. The fertilizability of human oocytes at different stages of meiotic maturation. Ann NY Acad Sci USA 1988;541:324-36. 3. McGaughey RW, Polge C. Cytogenetic analysis of pig oocytes matured in vitro. J Exp Zool 1971;176:383-91. 4. Racowsky C, Hendricks RC, Baldwin KV. Direct effects of nicotine on the meiotic maturation of hamster oocytes. Reprod ToxicoI1989;3:13-21. 5. Racowsky C, Kaufman ML, Dermer RA, Homa ST, Gunnala S. Chromosomal analysis of meiotic stages of human oocytes matured in vitro: benefits of protease treatment before fixation. Fertil Steril1992;57:1026-33. 6. Zenzes MT, Casper RF. Cytogenetics of human oocytes, zygotes, and embryos after in vitro fertilization. Hum Genet 1992;88:367-75. 7. 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. 8. Veeck L. Atlas of the human oocyte and early conceptus. Baltimore: Williams & Wilkins, 1991. 9. Uebele-Kallhardt BM. Human oocytes and their chromosomes. Berlin: Springer-Verlag, 1978. 10. Homa ST. Effects of cyclic AMP on the spontaneous meiotic maturation of cumulus-free bovine oocytes cultured in chemically defined medium. J Exp Zool 1988;248:222-31. 11. Racowsky C, Baldwin KV. In vitro and in vivo studies reveal that hamster oocyte meiotic arrest is maintained only transiently by follicular fluid, but persistently by membrana/cumulus cell contact. Dev BioI 1989;134:297-306. 12. Parfenov V, Potchukalina G, Dudina L, Kostyuchek D, Gruzova M. Human antral follicles: oocyte nucleus and the karyosphere formation (electron microscopic and autoradiographic data). Gamete Res 1989;22:219-31. 13. Zamboni L, Thompson RS, Moore Smith D. Fine morphology of human oocyte maturation in vitro. BioI Reprod 1972;7: 425-57. 14. Sathananthan AH. Maturation of the human oocyte in vitro: nuclear events during meiosis (an ultrastructural study). Gamete Res 1985;12:237-54. 15. Magnusson C, Bar Ami S, Braw R, Tsafriri A. Oxygen consumption by rat oocytes and cumulus cells during induced atresia. J Reprod FertiI1983;68:97-103. 16. Kennedy JF, Donahue RP. Binucleate human oocytes from large follicles. Lancet 1969;1:754-5. 17. Shea BF, Baker RD, Latour JPA. Human follicular oocytes and their maturation in vitro. Fertil Steril1975;26:1075-82. 18. Biggers JD, Whittingham DC, Donahue RP. The pattern of energy metabolism in the mouse oocyte and zygote. Proc Natl Acad Sci USA 1967;58:560-7. 19. Racowsky C. Gamete resources: origin and production of 00cytes. In: Pederson RA, McLaren A, First NL, editors. Animal applications of research in mammalian development. New York: Cold Spring Harbor Press, 1991:23-81. 20. Racowsky C, Satterlie RA. Metabolic and fluorescent dye and electrical coupling between hamster oocytes and cumulus cells during meiotic maturation in vivo and in vitro. Dev BioI 1985;108:191-202.
Racowsky and Kaufman Human oocyte anomalies
755
©
Vol. 58, No.4, October 1992
Printed on acid-free paper in U.S.A.
1992 The American Fertility Society
Nuclear degeneration and meiotic aberrations observed in human oocytes matured in vitro: analysis by light microscopyt'
Catherine Racowsky, Ph.D.t Milissa L. Kaufman, B.S. Department of Obstetrics and Gynecology, University of Arizona, Tucson, Arizona
Objective: To apply an improved air-dry procedure for the light microscopic identification of both degenerate and aberrant meiotic configurations in cultured human oocytes. Material and Design: Meiotically immature, normal appearing human oocytes retrieved after oophorectomy were placed into culture for 9 to 46 hours. Subsequently, oocytes were assessed morphologically and then air dried for light microscopic examination of chromatin configurations. Main Outcome Measures: Oocyte chromatin configurations were identified as normal, degenerate, or aberrant and then classified according to meiotic stage. Retrospective analyses were conducted to determine if [1] normal meiotic configurations were associated with morphologically viable oocytes and [2] degenerate or aberrant meiotic configurations were always associated with degenerate oocytes. In addition, for each meiotic stage, the proportion of oocytes exhibiting either degenerate or aberrant chromatin configurations was calculated. Results: Of 101 oocyte chromatin configurations analyzed, 71.3% were normal, 11.9% were degenerate, and 16.8% displayed meiotic aberrations. Retrospective analyses revealed that the majority of both normal and aberrant chromatin configurations were associated with morphologically viable oocytes (93.1 % and 88.2%, respectively), whereas all of the degenerate chromatin configurations were associated with morphologically degenerate oocytes. When assessed by stage, nuclear degeneration was observed exclusively at the germinal vesicle and diakinesis stages, whereas meiotic aberrations occurred most frequently after chromosome condensation. These aberrations were manifested either as clumped metaphase I configurations or as two distinct groups of bivalents that appeared to result from bivalent migration along the meiotic spindle without homologue segregation. Conclusions: Slightly >25% of human oocytes recovered after oophorectomy were incapable of undergoing normal meiotic maturation in culture. The majority of these abnormal oocytes appeared morphologically normal and yet possessed meiotic aberrations. These observations indicate that caution should be taken when using oocytes matured in vitro for application in assisted reproductive technology programs. Fertil Steril 1992;58:750-5 Key Words: Human oocyte, human chromosomes, degenerate chromatin, meiotic aberrations, light microscopy
Received April 20, 1992; revised and accepted June 30, 1992. * Supported in part by grant 15-0803 from the March of Dimes Birth Defects Foundation, White Plains, New York. t Reprint requests: Catherine Racowsky, Ph.D., Department of Obstetrics and Gynecology, College of Medicine, University of Arizona, Tucson, Arizona 85724.
meiosis to metaphase II. However, it since has been shown that as many as 70% of such oocytes recovered after oophorectomy fail to reach this stage of meiotic maturation (2), thereby excluding them from any possibility of normal fertilization. Based on light microscopic analysis of the chromosomes in oocytes from several animal species (3, 4), this incompetence in human oocytes may relate either to degeneration of the nuclear material or to aberration in the meiotic process. In the present in vitro study, we provide evidence for each of these two possibilities. This evidence has
Human oocyte anomalies
Fertility and Sterility
As first demonstrated by Edwards in 1965 (1), meiotically immature human oocytes, once released from the preovulatory follicular environment, will spontaneously undergo germinal vesicle breakdown in vitro and progress through the various stages of
750
Racowsky and Kaufman
been made possible through use of our improved airdry technique for preparing human oocytes for chromosomal analysis by light microscopy. As previously reported (5), this technique allowed for discrete visualization of normal chromosomal configurations and has allowed us to identify further those oocytes that exhibit either degenerate or aberrant meiotic forms. Such identification provides a new dimension to the field of human oocyte chromosomal analysis that, previously, has been dominated by cytogenetic studies of aneuploid mature human oocytes (6) with little regard being given to gross meiotic anomalies.
MATERIALS AND METHODS Collection of Ovarian Specimens and Harvesting, Selection, and Culture of Oocytes
The procedures used for the collection of ovaries after oophorectomy, as well as for the harvesting, selection, and culture of oocytes, were identical to those previously described (5). Thus, oocytes were released from antral follicles ranging in diameter from 2 to 5 mm. Follicles> 5 mm were not punctured to avoid the possible collection of maturing oocytes. Oocytes that appeared morphologically viable with intact cumuli were incubated for various time intervals from 9 to 46 hours under 5% CO 2 in humidified air at 37°C. Ham's F-10 medium was supplemented with 50% heat-inactivated follicular fluid that had been obtained from women undergoing follicular hyperstimulation before in vitro fertilization or gamete intrafallopian transfer procedures, as previously described by Cha et al. (7). Morphological Assessment of Oocytes After Culture
Before preparation of meiotic chromosomal configurations, oocytes were examined morphologically using a Bausch and Lomb dissecting microscope (model no. 620; Bausch and Lomb). Oocytes were classified as viable when the cytoplasm was light brown and uniformly granular, the oolemma was smooth, and the cell had an even, spherical configuration (8). In contrast, oocytes were classified as degenerate when the cytoplasm was darkened and appeared dense, and the oolemma was ruffled and shrunken away from the zona pellucida (ZP) (9). Some nonviable oocytes were also mis-shaped (9) and, in the extreme, had collapsed as discs. Vol. 58, No.4, October 1992
Preparation of Oocytes for Light Microscopy of Meiotic Configurations
Oocytes were prepared for chromosomal analysis as previously described (5). Briefly, cumulus-free oocytes were exposed to protease for 20 to 40 seconds to digest partially the proteins within the ZP and then fixed on a microscope slide with 3:1 (vol/vol) ethanol:acetic acid. After air drying, oocyte chromatin preparations were stained with Wright's stain (5) and subsequently analyzed for meiotic status under bright field illumination using a Zeiss standard compound microscope (model no. RA-38; Carl Zeiss, Inc., New York, NY). Classification of Meiotic Configurations
Oocyte meiotic configurations were classified as normal, degenerating, or aberrant based on previous observations of typical chromatin characteristics both in numerous animal studies (3, 4, 10, 11) and in two previous studies involving human oocytes matured in vitro (1, 5). Subsequently, within each of these classes, the meiotic configurations were subclassified into groups according to the stages of germinal vesicle, diakinesis, metaphase I, and anaphase I to metaphase II. Data Analyses
Retrospective analyses were carried out to determine if normal meiotic configurations always were associated with oocytes classified as morphologically viable after culture. In addition, similar analyses were performed to identify if degenerate and aberrant meiotic forms were associated exclusively with oocytes classified as degenerate after culture. Subsequently, further analyses were conducted to delineate the proportion of oocytes within a specific meiotic group that exhibited either degenerating nuclear chromatin or a meiotic aberration. Photographic Preparation of Meiotic Configurations
Preparations found to display nuclear degeneration or meiotically aberrant forms were photographed using a Nikon Fluophot compound microscope and Kodak Technical Pan 2415 film (Eastman Kodak Company, Rochester, NY) rated at an exposure of 64. The exposed film was developed at 68°C with a working solution of TD-3 (Photographer's Formulary, Missoula, MT). Racowsky and Kaufman
Human oocyte anomalies
751
Table 1 Retrospective Analyses Correlating the Incidence of Normal, Degenerate, and Aberrant Chromosomal Configurations With Oocyte Morphological Appearance After Culture Morphological appearance Chromosomal configuration
Viable
Degenerate
Normal Degenerate Aberrant
93.1 (67)* 0.0 (0) 88.2 (15)
6.9 (5) 100.0 (12) 11.8 (2)
* Values are percents with the number of oocytes in each group in parentheses.
RESULTS Correlation of Meiotic Classification and Morphological Appearance
Of the 101 air-dried preparations analyzed in this study, 71.3% displayed normal chromosomal configurations, 11.9% were degenerate, and 16.8% exhibited meiotic aberrations. Retrospective analysis revealed that most of those oocytes displaying normal chromosomal configurations (93.1 %) had been classified as morphologically viable after culture (Table 1). Conversely, all of those oocytes displaying nuclear degeneration were considered morphologically degenerate after culture. Of interest, however, the majority of oocytes containing aberrant meiotic configurations (88.2%) had been classified as morphologically normal when examined at the end of culture.
Incidence of Nuclear Degeneration and Meiotic Aberrations by Stage Germinal Vesicle Stage
In contrast to the normal germinal vesicle nucleus with a prominent nucleolus and homogeneously dispersed fibrous chromatin strands (5), a minority of immature oocytes (21.9%) contained degenerate chromatin configurations similar to that depicted in Figure 1A. As shown, this degenerate condition was typified by small fragmented chromatin pieces (arrows in Fig. 1A), numerous cytoplasmic vacuoles (small arrowheads) and a more-or-less spherical structure that stained heterogeneously and contained a vacuole (large arrowhead in Fig. 1A). Figure 1B shows an aberrant immature oocyte possessing two germinal vesicles, a condition observed in only 1 (3.1%) ofthe 32 oocytes scored at the germinal vesicle stage. The two nuclei are 23.5 and 25.0 Jl-m in diameter, respectively, which is approximately half the diameter of a normal, single vesicular nucleus (50.0 Jl-m) (5). Interestingly, however, each contains a nucleolus equal in diameter to that contained in a typical germinal vesicle stage oocyte (10 Jl-m). Diakinesis
Figure 2A to C shows a series of atypical meiotic configurations representing progressive stages of degeneration during diakinesis. In our study population, 36.8% of oocytes in diakinesis exhibited some form of nuclear degeneration. One of these oocytes
Figure 1 Immature oocytes exhibiting (A) degenerate chromatin, X667 and (B) two germinal vesicles (arrows), each containing a darkly staining spherical nucleolus, X529. Note that the degenerate germinal vesicle is characterized by small chromatin fragments (arrows) and a heterogenously staining nucleolus (large arrowhead) set against a background of cytoplasmic vacuoles (small arrowheads). 752
Racowsky and Kaufman Human oocyte anomalies
Fertility and Sterility
Figure 2 Progressive stages of degeneration during diakinesis. Note (A) lack of a nucleolus in the presence of stringy, degenerating chromatin, X500; (B) uncharacteristically clumped chromatin aggregates (arrows), some of which are adherent to the nucleolus (arrowhead), X975; and (C) condensed chromatin clusters lacking clear definition and periodically interconnected via chromatin bridges (arrow), X1,300.
also revealed an unusual sequence of chromatin condensation followed by nucleolus dissolution since the nucleolar material had disappeared in this oocyte before any dramatic condensation of the chromatin (Fig. 2A). The other oocytes in this degenerate group contained atypical, degenerating meiotic configurations at later diakinesis. For example, in Figure 2B, although the nucleolus (arrowhead) appears to be functioning normally as an organizing center for chromatin condensation (12), the chromatin appears clumped and is condensed into aggregates much shorter than the strands of normal condensing bivalents (arrows). Figure 2C reveals an oocyte at an even later stage of diakinesis in which extremely condensed chromatin clusters were frequently in-
terconnected via chromatin "bridges" (arrow). These bridges probably represent regions along the chromosomes in which chiasmata formed but subsequently failed to terminalize after completion of synapsis. Metaphase I
Figure 3A shows an aberrant metaphase I configuration characterized by the sticky adherence of bivalents into a tightly compacted clump. Thus, in contrast to the discrete distribution of chromosomes in the typical metaphase I configuration, only a few bivalents can be discerned (arrows). This aberration of metaphase I was observed in 12 of 28 oocytes
Figure 3 Meiotic aberrations after chromatin condensation. (A), Clumped metaphase I configuration showing a sticky appearance and exhibiting only a few clearly defined bivalents (arrows), X3,288. (B), Two segregated groups of chromosomes in which the homologues failed to disjoin. Note the indistinct appearance of one of the groups (arrow) that is apparently undergoing degenerative processes typically characterized by a normal, haploid polar body, X3,143. Vol. 58, No.4, October 1992
Racowsky and Kaufman
Human oocyte anomalies
753
(42.9%) that were at this stage of meiotic maturation. Anaphase I to Metaphase II
Adherence of bivalents during anaphase I resulted in an aberrant chromosomal configuration as shown in Figure 3B, in which the homologues failed to disjoin, thereby resulting in the bivalents themselves migrating to opposite ends of the spindle during telophase I. This phenomenon was observed in 3 of the 22 oocytes (13.6%) that had progressed beyond metaphase I. Interestingly, in 2 of these oocytes, one of the groups of bivalents was undergoing degeneration (arrow in Fig. 3B) in a manner similar to that which occurs in the haploid chromosomes of a normal polar body (5). DISCUSSION
The results of this study show that degenerate chromatin is associated exclusively either with the germinal vesicle stage or with diakinesis, the first overt stage of maturation. Consistent with previous electron microscope analyses (13, 14), the present degenerate immature oocytes revealed cytoplasmic vacuoles, and the nucleus appeared pyknotic, exhibiting heterogenous staining and chromatin fragmentation (Fig. 1A). Such abnormal nuclear and cytoplasmic characteristics clearly are inconsistent with chromatin condensation so that the germinal vesicle degenerates before any overt signs of meiotic resumption. However, several oocytes did exhibit degenerate chromatin configurations at diakinesis (Fig. 2). Although such degeneration may have occurred in vitro, it is possible that abnormal follicular development may have triggered this process in vivo because this degenerate meiotic stage has been associated with follicular atresia (15). Nevertheless, our inclusion of only those oocytes that appeared morphologically normal at the onset of culture argues against this possibility. Although the present results indicate that meiotic aberrations occurred predominantly in maturing oocytes that exhibited full chromosomal condensation' one immature oocyte was found to contain two germinal vesicles. Two previous reports have described binucleate human oocytes (16, 17). This rare aberration has been associated with elevated levels of gonadotropin (16) that, in turn, may cause the formation of multiovular follicles with ensuing oocyte fusion to form binucleate or multinucleate oocytes. There are two reasons why we believe that the present binuclear oocyte did not arise from 00754
Racowsky and Kaufman
Human oocyte anomalies
cyte fusion. First, this oocyte was retrieved from a patient diagnosed with endometriosis, a condition not associated with elevated gonadotropin levels. Second, each of the two nuclei in this oocyte was half the size of a normal, diploid germinal vesicle. Based on these facts, we suggest that this condition resulted from an aberration in the last mitotic division of the oogonium from which this primary oocyte arose so that the S-phase of the cycle was omitted and the cell failed to undergo cytokinesis. Thus, a binucleate oogonium entered prophase I of meiosis to result in two semihaploid nuclei. That this oocyte appeared morphologically normal after 46 hours in culture and yet had failed to resume meiosis suggests that this binucleate condition, although compatible with cellular viability, was incompatible with meiotic maturation. The present findings revealed that the majority of meiotic aberrations occurred either at metaphase I or at some stage subsequent to alignment of homologues on the first metaphase plate. These aberrations fall essentially into two related classes of meiotic configuration: the clumped metaphase I (Fig. 3A) and the two groups of bivalents (Fig. 3B). The clumped metaphase I condition, characterized by adhesion of the chromosomes into a compacted mass (Fig. 3A), accounted for almost one half ofthe metaphase configurations. Interestingly, the sticky appearance of such chromosomes has been noted in previous reports investigating in vitro maturation of both pig (3) and human (9) oocytes, although the cause of this phenomenon has yet to be elucidated. Based on the present observations of clumped metaphase I chromosomes, we speculate that this stickiness may relate to an alteration in the surface structure of the chromatin that either is induced by some intrinsic property of the oocyte itself or after an interaction of the oocyte or its surrounding cumulus cells with some component(s) in the culture medium. Numerous studies have shown that follicle cells provide both nutritional (18) and regulatory (19,20) modulation of oocyte metabolism. The possibility exists, therefore, that a subclass of cumulus complexes may respond abnormally to the culture medium, thereby exhibiting metabolic aberrations that impact on the heterologous supply of nutritional and/or regulatory factors critical for normal oocyte metabolism and maintenance of chromosomal integrity. Although earlier studies documented the degeneration of clumped metaphase I chromosomes in cultured pig (3) and human (9) oocytes, a recent study (4) noted the viable persistence of this meiotic Fertility and Sterility
form in cultured hamster oocytes. In the latter study, it was concluded that the clumped metaphase I configuration may progress to an irregular meiotic form similar to that depicted in Figure 3B, in which the bivalents themselves have migrated to opposite poles of the spindle. Such a meiotic form has been attributed to perturbation in assembly of the meiotic spindle and/or to a disturbance in the interaction between the kinetochore-spindle apparatus during anaphase I (4). An alternative explanation, however, may be that chromosomal stickiness compromised either bivalent alignment along the equatorial plate and/or chiasmatic splitting during anaphase 1. Interestingly, the results of the present study show that segregation of chromosomes into two groups of bivalents occurs in > 10% of oocytes progressing beyond anaphase 1. Yet, to our knowledge, no previous study has documented such a meiotic aberration in the human oocyte. This lack of documentation may be due to the inadequacy of past procedures for preparing human chromosomes for light microscopy. Our own experience has shown that two groups of bivalents easily may be mistaken for normal telophase I or metaphase II stages in oocyte air-dry preparations in which overlapping chromosomes may result from poor spreading during fixation. It remains to be determined whether the relatively high incidence of meiotic aberrations observed in the present study reflects the fact that the majority of oocytes were collected from nondominant and thus from incompletely mature follicles. In addition, it is unknown whether the observed incidence is higher than that which exists in oocytes matured in vivo. Ideally, a study parallel to the present study should be performed using oocytes retrieved from dominant follicles both before and after induction of meiotic resumption in vivo. In reality, however, such oocytes are rarely, if ever, available for experimentation. Acknowledgments. The authors thank Rachelle A. Dermer, B.F.A., Arizona State University, Tempe, Arizona, for preparing the photographs and Sujatha Gunnala, M.D., Southwest Fertility Center, Phoenix, Arizona, for assisting in the procurement of research material.
REFERENCES 1. Edwards RG. Maturation in vitro of human ovarian oocytes.
Lancet 1965;1:926-9.
Vol. 58, No.4, October 1992
2. Lopata A, Leung PC. The fertilizability of human oocytes at different stages of meiotic maturation. Ann NY Acad Sci USA 1988;541:324-36. 3. McGaughey RW, Polge C. Cytogenetic analysis of pig oocytes matured in vitro. J Exp Zool 1971;176:383-91. 4. Racowsky C, Hendricks RC, Baldwin KV. Direct effects of nicotine on the meiotic maturation of hamster oocytes. Reprod ToxicoI1989;3:13-21. 5. Racowsky C, Kaufman ML, Dermer RA, Homa ST, Gunnala S. Chromosomal analysis of meiotic stages of human oocytes matured in vitro: benefits of protease treatment before fixation. Fertil Steril1992;57:1026-33. 6. Zenzes MT, Casper RF. Cytogenetics of human oocytes, zygotes, and embryos after in vitro fertilization. Hum Genet 1992;88:367-75. 7. 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. 8. Veeck L. Atlas of the human oocyte and early conceptus. Baltimore: Williams & Wilkins, 1991. 9. Uebele-Kallhardt BM. Human oocytes and their chromosomes. Berlin: Springer-Verlag, 1978. 10. Homa ST. Effects of cyclic AMP on the spontaneous meiotic maturation of cumulus-free bovine oocytes cultured in chemically defined medium. J Exp Zool 1988;248:222-31. 11. Racowsky C, Baldwin KV. In vitro and in vivo studies reveal that hamster oocyte meiotic arrest is maintained only transiently by follicular fluid, but persistently by membrana/cumulus cell contact. Dev BioI 1989;134:297-306. 12. Parfenov V, Potchukalina G, Dudina L, Kostyuchek D, Gruzova M. Human antral follicles: oocyte nucleus and the karyosphere formation (electron microscopic and autoradiographic data). Gamete Res 1989;22:219-31. 13. Zamboni L, Thompson RS, Moore Smith D. Fine morphology of human oocyte maturation in vitro. BioI Reprod 1972;7: 425-57. 14. Sathananthan AH. Maturation of the human oocyte in vitro: nuclear events during meiosis (an ultrastructural study). Gamete Res 1985;12:237-54. 15. Magnusson C, Bar Ami S, Braw R, Tsafriri A. Oxygen consumption by rat oocytes and cumulus cells during induced atresia. J Reprod FertiI1983;68:97-103. 16. Kennedy JF, Donahue RP. Binucleate human oocytes from large follicles. Lancet 1969;1:754-5. 17. Shea BF, Baker RD, Latour JPA. Human follicular oocytes and their maturation in vitro. Fertil Steril1975;26:1075-82. 18. Biggers JD, Whittingham DC, Donahue RP. The pattern of energy metabolism in the mouse oocyte and zygote. Proc Natl Acad Sci USA 1967;58:560-7. 19. Racowsky C. Gamete resources: origin and production of 00cytes. In: Pederson RA, McLaren A, First NL, editors. Animal applications of research in mammalian development. New York: Cold Spring Harbor Press, 1991:23-81. 20. Racowsky C, Satterlie RA. Metabolic and fluorescent dye and electrical coupling between hamster oocytes and cumulus cells during meiotic maturation in vivo and in vitro. Dev BioI 1985;108:191-202.
Racowsky and Kaufman Human oocyte anomalies
755
