Chromosome Mosaicism in Human Embryos

BIOLOGY OF REPRODUCTION 51, 373-379 (1994)

Chromosome Mosaicism in Human Embryos S. MUNNE, 1'2 H.U.G. WEIER, 3'4 J. GRIFO, 2 and J. COHEN2 Centerfor Reproductive Medicine and Infertility,2 Department of Obstetrics and Gynecology New York Hospital-Cornell University Medical Center, New York, New York 10021 Division of Molecular Cytometry,3 Department of Laboratory Medicine, School of Medicine University of California, San Francisco, California94143-0808 Division of Life Sciences,4 University of California, Lawrence Berkeley Laboratory, Berkeley, California92720 ABSTRACT In the human, mosaicism may occur before implantation; but, to determine when it first occurs, it is necessary to study the chromosomal complement of all blastomeres. Full karyotypes of blastomeres from 2- to 8-cell human embryos by conventional kary6typing of metaphase spreads are difficult to obtain. The aim of this study was to assess the stage at which mosaicism occurred in preimplantation human embryos through use of fluorescence in situ hybridization (FISH) with multiple probes. All or most blastomeres from 2- to 12-cell human embryos were analyzed by FISH using probes for gonosomes and chromosome 18. FISH was performed on blastomeres from 117 morphologically normal monospermic embryos that were not transferred after preimplantation diagnosis because of their risk of carrying X-linked disease; 20 (17.1%) of these embryos were mosaic. Another group of 163 arrested or morphologically abnormal monospermic embryos were also analyzed by FISH; 47 (28.8%) of these embryos were mosaic. In addition, 37 dispermic embryos were analyzed, and 28 (75.7%) of these were found to be mosaic. Mosaicism first occurred at the second cleavage division when the monospermic embryo was mostly diploid and at the first cleavage division when the embryo was mostly haploid, polyploid, or dispermic.

INTRODUCTION

percussions on culture of embryos and preimplantation genetic diagnosis. For example, mosaicism arising at the first embryonic division will be easily detected by single-cell embryo biopsy followed by FISH or karyotype diagnosis of that cell, but mosaicism occurring at later stages will not. In addition, mosaicism may be caused by in vitro culture conditions and its onset may indicate when a disturbance in the embryonic development has occurred.

Karyotype analysis has been performed in non-arrested human preimplantation embryos demonstrating a high proportion of mosaicism [1-4]. It is, however, necessary to study the chromosomal complement of all blastomeres, dividing and non-dividing, in order to determine the onset and mechanism of mosaicism. A complete analysis of blastomeres from 2- to 8-cell human embryos by conventional karyotyping of metaphase spreads is difficult to obtain, since non-dividing human embryos cannot be analyzed. For interphase chromosome enumeration, fluorescence in situ hybridization (FISH) is the method of choice. FISH chromosome studies on human blastomeres have been performed using single probes, probes for X and Y chromosomes, or probes binding to chromosomes X, Y, 18, 13, and 21 simultaneously [5-9]. Cells that are aneusomic for one particular chromosome can be differentiated from truly haploid or triploid cells by this method, assuming a low frequency of double aneusomy comparable to numerical abnormalities found in human sperm complements [10-12]. The purpose of the present study was to determine the onset of mosaicism in 2- to 8-cell human embryos and to shed light on the mechanisms of formation of mosaicism at that stage. The onset and mechanism of mosaicism in preimplantation human embryos may have important re-

MATERIAL AND METHODS

Source of Embryos, Blastomere Biopsy, and Fixation Embryos were obtained from patients undergoing IVF treatment for infertility at The Center for Reproductive Medicine and Infertility, Cornell University Medical Center. The number of pronuclei per zygote was scored after insemination. Embryos in which a second polar body was clearly extruded and in which there were two (monospermic) or three pronuclei (dispermic) were allowed to cleave for three days. Embryos without a clear second polar body could possibly have been digynic and were excluded. Embryos from two sources were used for the current study. One source comprised embryos that had not yet reached the 8-cell stage on the fourth day of development and had not cleaved during a 24-h period ("arrested") [13]. The majority of these embryos exhibited dysmorphisms such as embryonic fragmentation exceeding 20-25%, granular cytoplasm, multinucleation, and uneven cells. These embryos are normally discarded. Arrested embryos may be investigated by FISH according to a protocol endowed by the Human Investigation Committee of the New York Hospital-

Accepted April 15, 1994. Received December 8, 1993. 'Correspondence: Dr. S. Munne, The Gamete and Embryo Research Laboratory, Center for Reproductive Medicine and Infertility, Department of Obstetrics and Gynecology, New York Hospital-Cornell University Medical Center, P.O. Box 30, 1300 York Avenue, New York, NY 10021. FAX: (212) 746-8589.

373

374

MUNNE ET AL.

Cornell University Medical College (protocol #0890-701). The second source of embryos consisted of normally developing embryos considered at risk of being affected by an X-linked disease after preimplantation genetic diagnosis [14]. Patients from whom the embryos were obtained had given informed consent to these procedures. Blastomeres were removed from each embryo by micromanipulation [15] and fixed individually in accordance with a previously described protocol, which was not modified [8]. FISH A FISH technique developed for two and three probes was used [6,16]. Briefly, DNA probes for alpha-satellite repeat clusters in the centromeric region of chromosomes X and 18 and for the satellite III DNA on the long arm of the Y chromosome were purchased from Imagenetics (Naperville, IL). These probes were directly labeled with fluorescent haptens: the Y chromosome-specific probe with a red fluorochrome (CEP Spectrum Orange, Imagenetics) and the 18 chromosome-specific probe with a green fluorochrome (CEP Spectrum Green, Imagenetics). The X chromosome probe was a 1:1 mixture of probes labeled with red and green fluorochromes. With the use of a dual-wavelength filter (ChromaTechnology, Brattleboro, VT), the Y chromosome appeared in red, the 18 in green, and the X in yellow. One microliter of each probe was added to 8 l of hybridization mix II solution (component of the SpectrumCEP system, Imagenetics) [17]. No blocking DNA was added to the hybridization solution since it was already present in the probe solutions purchased from Imagenetics. Hybridization solution (12 l) was added to an area no more than 18 x 18 mm 2 of the slide containing the fixed blastomeres; the slide was then coverslipped and denatured for 8 min in a slide warmer at 80°C. The coverslip was sealed with rubber cement and the slide was left for 3-6 h at 37°C to hybridize in a humidified chamber. The coverslips were then removed and slides were immersed in 50% formamide in double-strength saline-sodium citrate (SSC), pH 7.0, at 420C for 15 min; this was followed by two washes of 15 min each in PN buffer (0.1 M sodium phosphate buffer, pH 8, 1% Nonidet P-40) at room temperature. The DNA was counterstained with 4',6-diamino-2-phenylindole (DAPI) mixed in antifade solution (Imagenetics). Fluorescence microscopy was performed on a Nikon Optiphot microscope (Nikon Inc., Melville, NY) using a dual filter set for simultaneous observation of fluorescein isothiocyanate and Texas Red (ChromaTechnology). This allowed viewing of Spectrum Orange signals in combination with Spectrum Green signals. A Nikon filter was used for DAPI observation. Scoring Criteria Scoring criteria from Hopman et al. [18] were followed. Minor hybridization spots with lower fluorescent intensity

were not scored, as they are considered to be cross-hybridization signals to nontarget chromosomes. Spots found in close vicinity to one another, interconnected, or in paired arrangements were counted as one signal. They most probably represent sister chromatids or split signals. FISH Failure Criteria FISH was considered to have failed in particular blastomeres when: 1) blastomeres with a clear nucleus detected at phase-contrast were not found, 2) blastomeres did not have signals or were covered by debris, and 3) blastomere had less than two gonosomes or chromosome 18specific signals. These were considered false-negative FISH errors unless sibling blastomeres had extra signals compensating for the ones missing in a given blastomere. We consider such embryos to be compensated mosaics. Other observations often confused with false negatives can be seen in embryos in which the majority of blastomeres have the same abnormality, for example, monosomy or haploidy. The three criteria specified above were also used for polynucleated blastomeres, with all the nuclei present in the same blastomere being counted as a unity. In addition, according to a fourth criterion, false-positive FISH failure was considered to have occurred when blastomeres had more than two gonosomes or chromosome 18-specific signals. Exceptions to this can occur when sibling blastomeres have missing signals compensating for the extra ones in other blastomeres. Such embryos were also considered to be compensated mosaics. A similar rule applied to embryos in which all blastomeres had the same abnormality (trisomy and polyploidy). Finally, blastomeres having four signals per chromosome were considered tetraploid within a diploid embryo. Cells with more than two gonosomes or chromosome 18-specific signals per cell were considered chromosomally abnormal unless any of these signals were cross-hybridization signals or split signals, as stated above. Mosaic Embryos and the Origin of Mosaicism The cell division generating chromosome mosaicism was determined by assessing the number of blastomeres of each cell type. This could be accomplished when the majority of embryonic cells had been analyzed. Blastomeres, regardless of their karyotype, were assumed to have the same chance to develop, because previous studies of arrested embryos suggest that abnormal cells can cleave (unpublished observations). All blastomeres of monospermic embryos were abnormal when the chromosome abnormality occurred during the first embryonic division (Fig. 1). When one half or one fourth of the blastomeres were abnormal, mosaicism arose at the second and third division, respectively. RESULTS Only embryos in which half or more of the cells were analyzed were considered for this study. FISH was per-

MOSAICISM IN HUMAN EMBRYOS

375

37 dispermic embryos were analyzed, and 28 (75.7%) of these were found to be mosaics. The gonosome and 18chromosome constitution of some of these embryos has been described previously [6-8, 14,16]. Sixty-seven monospermic embryos and 18 dispermic

embryos in which the majority of cells were analyzed were found to be mosaic for one or both chromosome pairs. Of the monospermic mosaic embryos, 35 were mostly diploid, 27 were polyploid, and 5 were haploid. Aneuploid mosaicism in the diploid embryos was found in 19 instances; 17 embryos were partially tetraploid, and 5 were partially haploid. Six of the embryos had two types of mosaicism: 1 was aneuploid mosaic and partially haploid, and 5 were aneuploid mosaics with tetraploid cells (Table 1). Figure 2 shows normal diploid, trisomic, tetraploid, and haploid cells. The mechanism of aneuploid mosaicism was due mostly to mitotic non-disjunction, since a combination of monosomic, normal, and trisomic cell types was detected in 11 embryos. In 4 instances, random mitotic divisions occurred in some cells. For example, an embryo showed three normal XY1818 cells in addition to one XX181818, one XY181818, and two cells with just one Y in each and no X or 18 signals. These last four cells combined had-on average-an XY1818 complement when they were counted as three cells. Another embryo with nine cells was probably created after anaphase lag during the second division, since five of its cells were XX1818 and four were X01818. These could also have been four false negatives, but the FISH failure rate with this protocol (49 of 982; 5%) was well below the 45% of X01818 cells observed in that embryo. The remaining (n = 4) aneuploid mosaics were more complex and were of unknown origin (Table 1). Mosaicism in diploid embryos was generated almost invariably during the second or third division (Table 2), and in only 1 (2.9%) embryo were all the cells abnormal, four of these being X01818 and four XXX1818. In addition, this embryo was hybridized with chromosome 16-specific probes and all the cells showed two signals. FIG. 1. Onset of mosaicism in monospermic human embryos. The onset of chromosome mosaicism can be determined in early embryos by the ratio of normal to chromosomally abnormal cells. This is exemplified in the figure by a non-disjunction of sex chromosomes occurring at different cleavage stages. Mosaicism occurring at the first cleavage will result in an embryo in which all the cells are abnormal (a). Mosaicism occurring at the second cleavage in one of the two cells will result in an embryo in which half of the cells are abnormal (b). Mosaicism occurring at the third cleavage in one of the four cells will result in one fourth of the cells being abnormal (c). This model does not represent mechanisms of division in dispermic embryos, since a considerable proportion of those may divide in a tripolar fashion 119, 201.

formed on blastomeres of 117 monospermic developing embryos rejected after preimplantation diagnosis due to X-linked disease and on 163 arrested or morphologically abnormal monospermic embryos; 20 (17.1%) and 47 (28.8%) mosaic embryos were detected, respectively. In addition,

Polyploid mosaicism occurred often (33%, 9 of 27) or predominantly (61%, 17 of 28) at the first cleavage division when the embryos were monospermic and dispermic, respectively (Table 2). However, monospermic and dispermic polyploid mosaic embryos showed different mechanisms of formation. Monospermic polyploid embryos were generated in 52% (14 of 27, Table 3) of the cases by endoreduplication or cytokinesis failure in a polyploid cell. These embryos showed polyploid mosaics such as 4N/8N, 3N/6N, 8N/16N (Table 3). The rest of the monospermic polyploid embryos showed complex mosaicism with most cells having different karyotypes, indicative of the mosaicism having originated at the first embryonic division. In contrast, most dispermic embryos showed polyploid mosaicism that was of complex origin, that arose at the first division, or that was generated by random divisions (Table 3). The first cleavage of dispermic embryos may produce

376

MUNNE ET AL.

o

,

E,

.6

.

E

0E. o0

0

Egg E E a)og

0

E

oa)

o .r O'A

.m -o· " E o E EE ,OE -

E 'A. o

c E3 o

E o

'a

a

v E0

(D

C

xE '

E

E too

Yi

.

c

un E

i5 °

c

MOSAICISM IN HUMAN EMBRYOS TABLE 1. Types of mosaicism and mechanisms in monospermic diploid embryos. 19ab 10 4 1 4 17' 5b 35 b

Aneuploid mosaics by non-disjunction by random division by anaphase lag by unknown mechanism 2N/4N mosaics 2N/N mosaics Total

'Five embryos with aneuploid, tetraploid, and disomic cells are counted only once in the total. bOne embryo with aneuploid, haploid, and disomic cells is counted only once in the total.

two or three cells [19, 20] indicating that a proportion divide in a tripolar pattern. The dispermic embryos studied here were divided according to the number of cells resulting from their first cleavage. Those dividing into two or three cells showed similar rates of mosaicism. Mosaicism started at the first division in 13 of 21 (62%) and in 4 of 7 (58%) of these embryos, respectively. Five haploid mosaic embryos were also found. Mosaicism in these originated mostly at the first embryonic division. If only two chromosome pairs are analyzed, however, haploidy may be mistaken for double monosomy. This problem was circumvented by re-hybridization with a probe for chromosome 8. Results confirmed mosaicism for chromosome 8 in all haploid embryos, with an average of one autosome for each kind per cell. DISCUSSION The cell division generating chromosome mosaicism was determined by FISH analysis of chromosomes X, Y, and 18 on all or most blastomeres from human embryos. By this method the number of cell types was obtained and the onset of mosaicism determined. Mosaicism occurred at the first embryonic division when two different cells, each chromosomally abnormal, were detected; and mosaicism arose at the second and third division when half or one fourth of the blastomeres were abnormal (Fig. 1). The frequency of different cells provided information on the mechanism of mosaic formation; for instance, normal, monosomic, and trisomic cells suggested a mitotic non-disjunction event. Striking differences in the mechanisms and onset of mosaicism appeared when monospermic embryos with varying ploidy were compared to dispermic embryos. Polyploid and haploid mosaic embryos were usually created at the first division, whereas monospermic diploid mosaics were generated in 97% of cases at the second or later divisions. These differences may be explained by events taking place during the first and second embryonic divisions. During the first cleavage, two pronuclei with replicated DNA enter syngamy and then continue to karyokinesis, but the first time that a diploid nucleus undergoes a mitotic division is

377

TABLE 2. Number of observed mosaic embryos according to their onset of mosaicism and ploidy. Monospermic Cleavage 1st 2d 3d or later Total

Diploid

Polyploid

Haploid

Dispermic

1 17 17 35

9 15 3 27

3 1 1 5

17 11 0 28

during the second division. The results suggest that in diploid embryos, pronuclear syngamy occurs correctly, whereas mosaicism is generated by ensuing mitotic aberrations. Triploid dispermic zygotes, in contrast, may have problems distributing their chromosomes equally in two nuclei due to the presence of several paternal centrosomes with a resulting disarrangement of chromosomes [21]. The nucleolar alignment subsequent to pronuclear apposition that is frequently observed prior to syngamy [22] suggests that the chromosomes are organized in a bipolar fashion. The typical Y-shaped chromosome alignment of tripronucleate zygotes may contribute to a disorganized tripolar first division. Differences between second- and third-division mosaicism may be less important. Most polyploid monospermic mosaic embryos, for example, arrested at the 4cell stage. Cleavage patterns of dispermic embryos may be abnormal because many of these embryos divide directly into three cells [19]. This has been associated with the presence of multiple paternal centrosomes in such zygotes [20]. The incidence and onset of mosaicism in zygotes dividing into two or into three cells are the same, according to our results. This suggests that the presence of two paternal centrosomes, and not the number of cells into which the zygote divides, is the mechanism responsible for immediate

TABLE 3. Types of mosaicism and mechanisms in Monospermic and dispermic polyploid embryos. Monospermic: Aneuploid mosaics complex origin by non-disjunction random division 4N/8N Mosaics 3N/6N Mosaics 8N/16N Mosaics 3N/N Mosaics 5N/6N Mosaics 6N/2N Mosaics Total monospermic Dispermic: Aneuploid mosaics complex origin random division by non-disjunction 3N/6N Mosaics Total dispermic

13 2 2 2 5 3 2 2 1 1 27 17 8 7 2 1 18

378

MUNNE ET AL.

mosaicism. More embryos should be studied, since only seven zygotes cleaving into three cells were analyzed. Mosaicism in monospermic diploid embryos was mostly due to mitotic non-disjunction or to tetraploidization of one or two cells. Giant polyploid cells are a frequent phenomenon in trophoblastic tissue of many animals, including humans [23-26]. The presence of similar cells on Day 3 of development may indicate early differentiation, although the present data do not prove that there is a direct link between the two events. Dinucleated cells are also a frequent phenomenon that after FISH analysis suggests failure of cytokinesis, which may lead to tetraploidy [7]. More than half of the monospermic polyploid embryos were a combination of two cell types of different ploidy levels, most of these embryos having one cell type with twice the ploidy of the other. This suggests endoreduplication or karyokinesis failure. The remainder were mostly complex mosaics whose origin was unknown. This occurred in the majority of analyzed polyploid dispermic embryos and in those in which the nuclei may have split at random. In these mosaics, either there were abnormal complementing karyotypes involving more than two cells, or two cells were needed to obtain a single diploid karyotype, thus excluding non-disjunction as a cause. Similar phenomena were observed in a fraction of dinucleated blastomeres [7]. The fact that the complex mosaics observed in both polyploid groups arose at the first embryonic division suggests that a fraction of the monospermic embryos were polyspermic. Asynchrous pronuclear formation may explain why polypenetrated zygotes with two pronuclei become classified as monospermic. Such a hypothesis could be tested by re-hybridizing these embryos with probes for polymorphic markers of the Y chromosome. The present study raises questions about the use of preimplantation genetic diagnosis, which involves removal of one or two cells from 8-cell stage embryos and subsequent rapid analysis by polymerase chain reaction or FISH. Only embryos with a presumably normal blastomere are being replaced into the uterus. Although successfully applied to carriers of X-linked diseases and cystic fibrosis [15, 27], this approach does not consider the possible occurrence of mosaicism. The chances of detecting mosaicism in a diploid 8-cell embryo by biopsy of a single cell are 25 to 50% depending on the onset of the mosaicism. The present findings indicate that more than one cell should be biopsied for correct diagnosis; however, survival of the embryo decreases when too many cells are removed [28]. The relevance for normal embryonic development in mosaics involving one fourth to one eighth abnormal cells is unknown and should be determined. Mosaicism may be a minor problem for preimplantation genetic diagnosis if the abnormal cells do not participate in inner cell mass formation.

ACKNOWLEDGMENTS Embryologists Mina Alikani, Alexis Adler, Adrienne Reing, Toni Ferrara, and Elena Kissin are acknowledged for technical assistance. We are grateful to Drs. Rosenwaks, Bedford, and Ledger for their support of this work. Giles Tomkin is acknowledged for assistance during preparation of this manuscript. We wish to thank the Laboratory for Cell Analysis, UCSF, for providing access to the microscope equipment. H.-U. Weier was supported by the Office of Health and Environmental Research, Department of Energy, under contract DE-AC-03-76SF00098 with additional support from Amoco Technology Corp., Naperville, IL.

REFERENCES 1. Plachot M, Junca AM, Mandelbaum J, De Grouchi J, Salat-Baroux J, Cohen J. Chromosome investigations in early life. II. Human preimplantation embryos. Hum Reprod 1987; 2:29-35. 2. Plachot M, Veiga A, Montagut J, De Grouchy, Calderon G, Lepretre JS, Junca AM, Santalo J, Carles E, Mandelbaum J, Barri P, Degoy J, Cohen J, Egozcue J, Sabatier J, Salat-Baroux J. Are clinical and biological IVF parameters correlated with chromosomal disorders in early life: a multicentric study. Hum Reprod 1988; 3:627635. 3. Zenzes MT, Casper RF. Cytogenetics of human oocytes, zygotes, and embryos after in vitro fertilization. Hum Genet 1992; 88:367-375. 4. Bongso A, Ng SC, Lim J, Foung CY, Ratman S. Preimplantation genetics: chromosomes of fragmented human embryos. Fertil Steril 1991; 56:66-70. 5. Griffin DK, Wilton LJ, Handyside AH, Wiston RML, DelhantyJDA. Dual fluorescent in situ hybridization for simultaneous detection of X and Y chromosome-specific probes preimplantation embryonic nuclei. Hum Genet 1992; 89:18-22. 6. Munne S, Weier HUG, Stein J, Grifo J, Cohen J. A fast and efficient method for simultaneous X and Y in situ hybridization of human blastomeres. J Assisted Reprod Genet 1993; 10:82-90. 7. Munne S, CohenJ. Unsuitability of multinucleated human blastomeres for preimplantation genetic diagnosis. Hum Reprod 1993; 8:1120-1125. 8. Munne S, Lee A, Rosenwaks Z, Grifo J, Cohen J. Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum Reprod 1993; 8:21852191. 9. Schrurs BM, Winston RML, Handyside AH. Preimplantation diagnosis of aneuploidy using fluorescent in-situ hybridization: evaluation using a chromosome 18-specific probe. Hum Reprod 1993; 8:296-301. 10. Brandriff B, Gordon L. Human sperm cytogenetics and the one-cell zygote. In: Allen JW, Bridges BA, Lyon MF, Moses MJ, Russell LB (eds.), Biology of Mammalian Germ Cell Mutagenesis. Branbury #34. Cold Spring Harbor: Cold Spring Harbor Press; 1990: 183-194. 11. Martin RH, Rademaker A. The frequency of aneuploidy among individual chromosomes in 6821 human sperm complements. Cytogenet Cell Genet 1990: 53:103107. 12. Estop AM, Cieply K, Van Kirk V, Munn6 S, Garver K. Cytogenetic studies in human sperm. Hum Genet 1991; 87:447-451. 13. Cohen J, Alikani M, Trowbridge J, Rosenwaks Z. Implantation enhancement by selective assisted hatching using zona drilling of embryos with poor prognosis. Hum Reprod 1992; 7:685-691. 14. Munne S, Tang YX, Grifo J, Rosenwaks Z, Cohen J. Sex determination of human embryos using the polymerase chain reaction and confirmation by fluorescence in situ hybridization. Fertil Steril 1994; 61:111-117. 15. Grifo JA, Tang YX, Cohen J, Gilbert F, Sanyal MK, Rosenwaks Z. Ongoing pregnancy in a hemophilia carrier by embryo biopsy and simultaneous amplification of X and Y chromosome specific DNA from single blastomeres. JAMA 1992; 6:727729. 16. Munn6 S, Grifo J, Cohen J, Weier HUG. Chromosome abnormalities in arrested human preimplantation embryos: a multiple probe fluorescence in situ hybridization (FISH) study. Am J Hum Genet 1994; 55:150-159. 17. Pinkel D, GrayJW, Trask B, van den Engh G, FuscoeJ, van Dekken H. Cytogenetic hybridization with fluorescently labeled nucleic acid probes. In: Cold Spring Harbor Symposia on Quantitative Biology, vol. LI. Cold Spring Harbor: Cold Spring Harbor Laboratory; 1986: 151-157. 18. Hopman AHN, Ramaekers FCS, Raap AK, Beck JLM, Devilee P, van der Ploeg M, Vooijs GP. In situ hybridization as a tool to study numerical chromosome aberrations. Histochemistry 1988; 89:307-316. 19. Kola I, Trounson A, Dawson G, Rogers P. Tripronuclear human oocytes: altered cleavage patterns and subsequent karyotypic analysis of embryos. Biol Reprod 1987; 37:395-401.

MOSAICISM IN HUMAN EMBRYOS 20. Sathananthan Al, Kola I, Osborne J, Trounson AO, Ng SC, Bongso A, Ratnam SS. Centrioles in the beginning of human development. Proc Natl Acad Sci USA 1991; 88:4806-4810. 21. Long CR, Pinto-correia C, Duby RT, Ponce-de-Leon FB, Boland MP, Roche JF, Robl JM. Chromatin and microtubule morphology during the first cell cycle in bovine zygotes. Mol Reprod Dev 1993; 36:23-32. 22. Wright G, Wiker S, Elsner C, Kort H, Massey J, Mitchell D, Toledo A, Cohen J. Observations on the morphology of human zygotes, pronuclei and nucleoli and implications for cryopreservation. Hum Reprod 1990; 5:109-115. 23. Murray JD, Moran C, Boland MP, Nancarrow CD, Sutton R, Hoskinson RM, Scaramuzzi RJ.Polyploid cells in blastocysts and early fetuses from Australian Merino sheep. J Reprod Fertil 1986; 78:439-446. 24. Long SE, Williams CV. A comparison of the chromosome complement of inner cell mass and trophoblast cells in day-10 pig embryos. J Reprod Fertil 1982; 66:645648.

379

25. Hare WCD, Singh EL, Betteridge KJ, Eaglesome MD, Randall GCB, Mitchell D, Bilton RJ, Trouson AO. chromosomal analysis of 159 bovine embryos collected 12 to 18 days after estrus. Can J Genet Cytol 1980; 22:615-626. 26. Benkhalifa M, Janny L,Vye P, Malet P, Boucher D, Menezo Y.Assessment of ploidy in human morulae and blastocysts using co-culture and fluorescent in-situ hybridization. Hum Reprod 1993; 8:895-902. 27. Handyside AH, Kontogianni EH, Hardy K, Winston RML. Pregnancies from biopsled human pre-implantation embryos sexed by Y-specific DNA amplification. Nature 1990; 344:768-770. 28. Tarin j., Conaghan J, Wiston RML, Handyside AH. Human embryo biopsy on the 2nd day after insemination for preimplantation diagnosis: removal of a quarter of embryo. Fertil Steril 1992; 58:970-976.