- Email: [email protected]
Embryo biopsy: the fate of abnormal pronuclear embryos
RBMOnline - Vol 17. No 6. 2008 782-788 Reproductive BioMedicine Online; www.rbmonline.com/Article/3511 on web 15 October 2008
Article Embryo biopsy: the fate of abnormal pronuclear embryos Nicole Noyes MD has been practicing reproductive endocrinology since 1990. She is currently a full-time Associate Professor at the New York University (NYU) School of Medicine, USA and is board-certified in obstetrics and gynaecology as well as reproductive endocrinology. She is co-director of the NYU oocyte cryopreservation program. Her clinical work has focused on infertility, fertility preservation and reproductive surgery and research efforts on implantation and early embryo development. Dr Noyes has been involved in the treatment of over 16,000 infertility patients using assisted reproductive technologies and has authored over 40 publications in peer-review journals.
Dr Nicole Noyes Nicole Noyes1, M Elizabeth Fino, Lewis Krey, Caroline McCaffrey, Alexis Adler, James Grifo NYU Fertility Centre, New York University School of Medicine, New York, New York, USA 1 Correspondence: e-mail: [email protected]
Abstract This study assessed 1908 embryos, including those with abnormal numbers of pronuclei, in IVF cycles from July 2001 to December 2006 in which preimplantation genetic screening (PGS) was performed on day 3 post-retrieval and ‘euploid’ embryos transferred the following day. PGS-intracytoplasmic sperm injection and PGS-translocation cycles were excluded. At 18 h post-insemination, the zygote distribution was 19% 0PN, 4% 1PN, 69% 2PN and 8% 3PN. No pregnancy occurred following 0PN or 1PN embryo transfers. A retrospective, blinded morphological ranking of all embryos on day 3 was performed and the results compared with PGS; no 0PN or 1PN embryo would have been chosen for transfer based on morphological superiority alone. Blastocyst formation occurred in 1PN embryos (29%) but not in 0PN embryos when evaluated on day 5. Euploid karyotypes were reported for biopsies of 0PN (3%), 1PN (5%) and 2PN (19%) embryos (P = 0.015, 1PN versus 2PN). A Y chromosome was observed in 0PN (17%) and 1PN (32%) embryos; surprisingly, 91% of these Y chromosome-bearing embryos were aneuploid. Many different meiotic and fertilization errors can result in 0PN or 1PN zygotes; these results indicate that the resultant embryos should not be transferred, especially when normally fertilized embryos are available. Keywords: chromosomes, embryo biopsy, embryo morphology, fertilization, preimplantation genetic screening, pronuclear
Introduction The time it takes to complete meiosis and the first mitotic cell cycle is known to vary greatly from one oocyte to another (Capmany et al., 1996). Following ovulation, human oocytes normally arrest at meiotic metaphase II and remain there until a spermatozoon enters, triggering resumption of the meiotic division and consequent extrusion of chromosomes in a second polar body. The appearance of two haploid pronuclei (2PN; one paternally and one maternally derived) marks the postfertilization transition from zygote into G1 of the first embryonic mitosis. This can occur anywhere from 3 to 20 h post-sperm insemination or injection (Balakier et al., 1993). Approximately 5% of human zygotes exhibit a single pronucleus rather than the expected two pronuclei when evaluated at 16–18 h postinsemination and this abnormality does not appear to be age related (Staessen et al., 1993).
782
In the early years of IVF, zygotes exhibiting only one pronucleus (1PN) were thought to have arisen from
parthenogenetic activation rather than fertilization by a spermatozoon (Kaufman, 1983). However, with the advent of preimplantation genetic screening (PGS), the frequent presence of a Y chromosome within the genetic make-up of these 1PN zygotes suggested that fertilization had taken place (Plachot, 1991, Munné et al., 1993; Staessen et al., 1993). In addition to parthenogenetic activation, other currently proposed mechanisms for 1PN zygote formation include asynchronous PN formation (Munné et al., 1993; Staessen et al., 1993; Payne et al., 1997), failure of either male or female chromatin to form a pronucleus (Flaherty et al., 1995) and abnormal fusion of the male and female pronuclei. The latter includes formation of a common pronuclear envelope (Levron et al., 1995) and fusion of two pronuclei that later form a larger single pronucleus (Tesarik and Mendoza, 1996). Embryos lacking or having only one pronucleus are thought to be suboptimal. Previously published results have shown a relationship between pronuclear morphology and preimplantation embryo development. Balaban
© 2008 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB23 8DB, UK
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. and colleagues demonstrated that pronuclear morphology can predict the risk of developmental arrest, irregular cleavage and even aneuploidy in human embryos, suggesting that pronuclear asynchrony or inequality might be markers for abnormalities in genetic programming (Balaban et al., 2004). The goal of the present work was to assess the morphological competence and chromosomal complement of these atypical zygotes.
Materials and methods All IVF with PGS cycles performed from July 2001 to December 2006 were reviewed (n = 126 cycles, 1908 embryos). Institutional approval for this research review was obtained. All intracytoplasmic sperm injection (ICSI) cycles and cycles performed for translocation were excluded. Women of all ages were included (mean age: 37.6 ± 4 years, range: 25–47 years). Ovarian stimulation protocols were individualized to achieve adequate numbers of mature oocytes at retrieval. Most patients were down-regulated with a gonadotrophinreleasing hormone (GnRH) agonist (leuprolide acetate, Tap, Lake Forest, IL, USA) and then treated with combinations of injectable recombinant FSH (Gonal F®, Serono, Rockland, MA, USA; Follistim®, Organon, Orange, NJ, USA) and/or human menopausal gonadotrophin (Repronex® or Menopure®, Ferring, Suffern, NY). Alternatively, LH suppression was achieved using GnRH antagonists (ganirelix acetate, Organon; cetrorelix acetate, Serono). When lead follicles reached a mean diameter of 17–18 mm, human chorionic gonadotrophin (HCG) was administered, and approximately 34 h later, oocytes were collected by ultrasound-guided transvaginal aspiration and placed in human tubal fluid media (HTF; Irvine Scientific, Irvine, CA, USA) supplemented with 6% Plasmanate (5% USP plasma protein fraction; Bayer, Elkhart, IN, USA) overlaid with Sage mineral oil (Cooper Surgical, Trumbull, CT, USA). The partner’s sperm was collected on the day of oocyte retrieval and prepared via swim-up or isolate. Oocytes were fertilized by routine insemination (4–6 h after retrieval). Following insemination, laboratory assessment of pronuclei was carried out 16–18 h later and recorded. All zygotes with more than 2 pronuclei were excluded from PGS. All other embryos arising from 0PN, 1PN or 2PN zygotes that were suitable for biopsy (≥3 cells with at least one intact blastomere) were subjected to genetic evaluation by PGS on day 3 postretrieval. A total of 1278 embryos underwent PGS. When performing embryo biopsy for PGS, one cell was removed from the cleavage-stage embryos on the morning of day 3 post-oocyte retrieval (Grifo, 1992). The cell was then fixed and analysed according to protocols previously described by Munné and co-workers (Munné and Weier, 1996; Munné et al., 1998; Bahce et al., 1999) at Reprogenetics, Inc. (Livingston, NJ, USA) for genetic assessment. Fluorescence insitu hybridization (FISH) analysis was performed to quantify the number of sex chromosomes (X and Y), and autosomes 13, 15, 16, 17, 18, 21 and 22 (8 and 14 in select cases). Selected euploid embryos were transferred to the uterus on day 4 or 5 post-retrieval. The patient population undergoing prenatal genetic screening included those at an advanced maternal age, with or without recurrent failed IVF (n = 66, 52%), those who had previously RBMOnline®
suffered two or more spontaneous miscarriages (n = 57, 45%) and those who electively chose PGS for other reasons (n = 3, 2%). In a later retrospective review, researchers were blinded to PGS, PN assessment and pregnancy outcomes and the biopsied embryos were ranked for embryo transfer based only on their morphology 3 days post-oocyte retrieval, prior to embryo biopsy. Ranking of these embryos was performed by two examiners using a quality score based on cell number and fragmentation. Good quality embryos had the highest number of cells and least fragmentation; the worst quality embryos developed slowly and exhibited significant fragmentation. The grading system was as follows: Quality Score 1, ≥5 cells and ≤10% fragmentation; Quality Score 2, ≥5 cells with ≥15% fragmentation; Quality Score 3, any cell number with ≥20% fragmentation or any embryo with fewer than 5 total cells. Once ranked by morphology, embryos blindly ‘selected’ for embryo transfer were compared with PGS results and the embryo choices made at the time of uterine replacement. Pregnancy outcomes were noted. A clinical pregnancy was defined as an intrauterine gestation with documentation of fetal cardiac activity. Statistical analysis was performed using the chi-squared and Student’s t-test (Sigmastat, SYSTAT Software Inc., CA, USA) as appropriate, with alpha equal to 0.050. Yates correction for continuity was used when appropriate.
Results Table 1 lists the demographics and outcomes of all evaluated IVF-PGS cycles. Of 1933 retrieved oocytes, 1908 (99%) developed to zygote stage by 18 h; 69% (1316/1908) of the zygotes exhibited 2PN. The remaining 31% (592/1908) of the zygotes displayed an abnormal pronuclei number, with significantly more 0PN (368) zygotes than 1PN (78) or 3PN (146) zygotes (P < 0.001). Of the embryos deemed developmentally competent and suitable for biopsy and PGS, 1188/1316 (90%) developed from 2PN zygotes, 32/368 (9%) from 0PN zygotes (P < 0.001) and 58/78 (74%) from 1PN zygotes. There was no FISH result for 36/1278 (3%) of the embryos. The mean number of embryos transferred per treatment cycle was 1.5 ± 0.1. The clinical pregnancy rate was 34% (37/109). There was no transfer in 17 cycles because all the embryos were reported as chromosomally abnormal. The embryo implantation rate (sacs per embryo transferred) was 28% (54/190); however, the implantation rate was 0% (0/4) for embryos that developed from either 0PN or 1PN zygotes, and 29% (54/186) for embryos that developed from 2PN zygotes. All transfers of embryos derived from either 0PN or 1PN zygotes were single-embryo transfers due to the lack of other available chromosomally normal embryos in the patient’s cohort. This allowed for adequate assessment of the embryo implantation rate for all pronuclear subgroups. No patients received a combination of a 2PN with a 0PN- or 1PN-derived embryo. Overall, 222/1242 (18%) of the biopsies were reported to have a normal number of chromosomes (Table 2, Figure 1). A euploid chromosome complement was reported in only 3% (1/30) of the biopsies from embryos from 0PN zygotes and 5% (3/57) of those from embryos from 1PN zygotes; in contrast 19% (218/1155) of the biopsies from embryos from 2PN zygotes were reported to be chromosomally normal (P = 0.015 for 1PN
783
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al.
Table 1. Demographic characteristics of IVF-preimplantation genetic screening cycles (PGS). Characteristic
Value
Number of cycles Age (years) No. of oocytes/cycle No. of embryos (%) Total 2PN 3PN 1PN 0PN Biopsied embryos/cycle Embryos–no PGS result (%) Aneuploid embryos (%) Aneuploid embryos/cycle Euploid embryos transferred/cycle Clinical pregnancy rate (fetal heart) per cycle (%) per transfer (%) Embryo implantation rate (%)
126 37.6 ± 0.4 (range 25–47) 15 ± 0.7 (total 1933) 1908 (100) 1316 (69) 146 (8) 78 (4) 368 (19) 10 ± 0.5 (total 1278) 36 (3) 1020/1242 (82) 8 ± 0.4 1.5 ± 0.1 37/126 (29) 37/109 (34) 54/190 (28)
Data are expressed as mean ± SEM unless otherwise stated.
Table 2. Distribution of chromosomal abnormalities (in embryos with FISH result) in all pronuclear subgroups. Pronuclear Total Euploid Haploid Polyploid Complex Monosomy/ subgroup abnormal Trisomy 0PN 1PN 2PN
30 1 (3) 4 (13) 4 (13) 57 3 (5) 13 (23) 2 (4) 1155 218 (19) 44 (4) 77 (7)
20 (68) 20 (35) 383 (33)
1 (3) 19 (33) 433 (37)
Data expressed as n (%).
Percentage
80 70
0PN n = 30
60
1PN n = 57 2PN n = 1155
50 40 30 20 10
Monosomy/ trisomy
Complex abnormal
Polyploid
Haploid
Euploid
0
Figure 1. Distribution of chromosomal abnormalities in all pronuclear subgroups.
784 RBMOnline®
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. versus 2PN; not statistically significant for 0PN versus 2PN). Moreover, although only 23% (13/57) of the biopsies from embryos from 1PN zygotes were haploid, this incidence was significantly higher than for embryos from 2PN zygotes 4% (44/1155) (P < 0.001). Curiously, 4% (2/57) of the embryos that developed from a 1PN zygote had a biopsy that was polyploid. Biopsied cells from embryos that developed from a 0PN zygote routinely presented a complex abnormal karyotype of multiple chromosome anomalies 67% (20/30) (Table 2, Figure 1). Eighteen (32%) embryos that developed from 1PN zygotes and five (17%) from 0PN zygotes tested positive for a Y chromosome, indicating that sperm had penetrated the oocyte; however, a euploid karyotype was noted in biopsies from only two of the 18 1PN and none of the five 0PN Y-bearing embryos. More than half of the 0PN and a third of 1PN embryos had a complex abnormal karyotype (Table 3). The results of the blinded morphological rankings of embryos on day 3 are shown in Table 4. Most (88%) of the embryos from 0PN zygotes (28/32) were given a quality score of 3 and were unsuitable for transfer; in contrast >40% of the embryos from 1PN (25/58) and 2PN (570/1188) zygotes were scored top quality (P < 0.001). Despite the structural integrity of the
embryos given a top quality score (Score = 1), biopsies of these morphologically superior embryos revealed aneuploidy in 88% (22/25) and 77% (437/570) of embryos from 1PN and 2PN zygotes, respectively. When embryos were assessed retrospectively for developmental competence, only 38% (12/32) of embryos that developed from 0PN zygotes had advanced to 6 cells at day 3; in contrast 55% (32/58) and 60% (713/1188) of the embryos from 1PN and 2PN zygotes, respectively, did so (P = 0.017 for 0PN versus 2PN). The cell number achieved by day 3 post-retrieval in each of the groups is shown in Table 5 and Figure 2. The developmental disparity broadened among the groups on day 5; of those assessed on day 5, no embryo from a 0PN zygote developed to the blastocyst stage whereas 29% (11/38) and 45% (322/719) of the embryos from 1PN and 2PN zygotes, respectively, reached this stage (P < 0.005; 0PN versus 1PN and 0PN versus 2PN zygotes).
Discussion Normal fertilization involves an integrated series of temporal events (Collas and Poccia, 1998; Marieb, 2001; Moore and Persaud, 2003; Feenan and Herbert, 2006). After a spermatozoon
Table 3. Genetic assessment of 0PN and 1PN zygotes biopsied 3 days post-oocyte retrieval. Genetic characteristic
0PNa (n = 32)
1PNa (n = 58)
Euploid 1 (3) 3 (5) Haploid 4 (13) 13 (23) Polyploid 4 (13) 2 (4) Complex abnormalb 20 (67) 20 (35) Monosomy 1 (3) 8 (14) Trisomy 0 (0) 6 (110) Monosomy and trisomy 0 (0) 5 (9) No result/no nucleusa 2 (6) 1 (2) Sex chromosome (n) Abnormal X0 (12), Y0 (1), XXX (5), X0 (22), XXX (3), XXXX (2), XXXX (2), XXY (3), 00 (1) XYY (1) Normal XX (5), XY (1) XX (12), XY (17) Values are n (%), unless otherwise stated. a Percentages are expressed as the portion of those with results except for the ‘no nucleus’ category, which is expressed as the portion of the total biopsied. b The category ‘complex abnormal’ includes embryos with two or more chromosomes having abnormal counts, but not completely polyploid or haploid, and includes embryos found to have multiple monosomies (Munné et al., 1998).
Table 4. Results of blinded embryo grading (best = score 1, worst = score 3). Pronuclear subgroup
Embryo Quality Score (%) 1 2
0PN 1PN 2PN
1/32 (3%) 3/32(9%) 28/32 (88%) 25/58 (43%) 9/58 (16%) 24/58 (41%) 570/1188 (48%) 216/1188 (18%) 402/1188 (34%)
3
785 RBMOnline®
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al.
No. of 0PN 1PN 2PN Euploid only cells (n = 32) (n = 58) (n = 1188) (0PN+1PN+ 2PN) (n = 222) <5 5 6 7 8 9 ≥10
15 (47) 4 (13) 7 (22) 3 (9) 2 (6) 1 (3) 0 (0)
13 (22) 8 (14) 11 (19) 10 (17) 11 (19) 1 (2) 4 (7)
211 (18) 158 (13) 189 (16) 199 (17) 325 (27) 58 (5) 48 (4)
32 (14) 18 (8) 30 (14) 46 (21) 88 (40) 5 (2) 3 (1)
Data expressed as n (%).
passes through the zona pellucida, its plasmalemma fuses with the oolemma of the egg prior to its incorporation into the egg. This invaginatory process results in the triggering of waves of Ca2+ release from local cytoplasmic stores that activate the egg (Evans and Florman, 2002). Activation involves a series of responses by the egg including the release of cortical granules (Abbott and Ducibella, 2001) to modify the molecular structure of the zona pellucida’s inner layer to prevent the penetration of additional spermatozoa (Bleil and Wassarman, 1981) into the egg. Resumption of the second meiotic division culminates in the extrusion of the second polar body containing a haploid set of female chromosomes (Williams, 2002). After the maternal and paternal chromosomes decondense and the paternal chromosomes are selectively demethylated (Xu et al., 2005), pronucleus formation then proceeds. The male pronucleus forms near the site of sperm entry into the egg and becomes enveloped by a nuclear membrane of maternal origin (Collas and Poccia, 1998); the female pronucleus forms concurrently near the site of second polar body extrusion (Payne et al., 1997). Microfilaments radiating from the sperm centrosome then move the two pronuclei towards each other in the centre of the cell (Simerly et al., 1995; Sathananthan et al., 1996). Eventually the two pronuclear membranes disassemble and the parental chromosomes come together in a process known as syngamy to establish the diploid embryonic genome (Nagy et al., 1998, Sathananthan, 1998, Tesarik and Greco, 1999). Studies monitoring fertilization outcome following insemination have revealed that the majority of normally fertilized zygotes display two pronuclei after a 16–18 h period (Nagy et al., 1998; Rienzi et al., 2005), as was observed in the present study. Such a conclusion is based on the assumption that all the eggs were mature and completed the first meiotic division at the time of insemination. However, the possibility exists that some of these 0PN and 1PN zygotes may have resulted from a ‘normal’ fertilization that was delayed until the maturation process was completed. Unfortunately, in the present study, pronuclear status was not routinely rechecked unless there was an abundance of 0PN or 1PN zygotes within the patient’s cohort of zygotes. Whether this delay in fertilization results in good quality blastocysts or euploid embryos remains to be determined.
786
Despite the above reservations, there is little doubt that for many
100%
≥10 cell 9 cell 8 cell
Relative incidence (%)
Table 5. Rate of embryo cleavage 3 days post-oocyte retrieval for 0PN, 1PN and 2PN zygotes.
7 cell 6 cell 5 cell <5 cell 0% 0PN
1PN
2PN
Euploid only
Figure 2. Rate of embryo cleavage 3 days post-oocyte retrieval for 0, 1 and 2 pronuclei (PN) zygotes.
zygotes the observation of zero or one pronucleus at 16–18 h post-insemination reflects an anomalous fertilization. Current explanations for 0PN zygotes include accelerated breakdown of pronuclear membranes representing biological variation in the timing of intracellular events (Manor et al., 1996), or lack of visualization due to granularity (Munné and Cohen, 1998). Suggested mechanisms for the appearance of a 1PN zygote include parthenogenetic activation, asynchronous development of the male and female pronuclei, pronuclear fusion, and failure of either male or female chromatin to form a pronucleus (Munné et al., 1993; Staessen et al., 1993; Flaherty et al., 1995; Levron et al., 1995; Tesarik and Mendoza, 1996; Payne et al., 1997). The latter includes failure of sperm DNA to decondense properly and abnormalities in sperm centrosome function. In time-lapse recordings of early embryos, 38% of the monitored zygotes developed pronuclei asynchronously (Payne et al., 1997). In another study, 25% of the 1PN zygotes at 16–18 h were found to have developed a second pronucleus when rechecked 4–6 h later (Staessen et al., 1993). Abnormalities in pronuclear membrane organization may explain some 1PN zygotes, particularly those where no additional pronuclei were seen on re-evaluation several hours later. Levron and co-workers suggested that in some cases of abnormal pronuclear formation, a common membranous envelope may form to surround both pronuclei (Levron et al., 1995). It is conceivable that this may occur when a spermatozoon enters or is deposited very close to the oocyte spindle, although studies that determined the site of sperm binding to the oolemma with polar body location do not support this hypothesis. PolScope reports that meiotic spindles are often located at considerable distances from the polar body suggest that sperm invagination by the egg may frequently occur near the spindle (Keefe et al., 2003; Konc et al., 2004). Prior karyotyping of 1PN embryos reported that 36–43% exhibited Y chromosome hybridization signals (Munné et al., 1993; Staessen and Van Steirteghem, 1997). These authors concluded that in approximately double the number of cases, X and Y chromosome-bearing sperm had penetrated the oocyte making an argument against parthenogenetic activation as the primary mechanism of uni-pronuclear formation. Sultan and RBMOnline®
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. colleagues compared the chromosome status of 1PN zygotes fertilized either by ICSI or normal sperm insemination (Sultan et al., 1995). Only 9.5% of the ICSI embryos were diploid and 9.5% contained a Y chromosome; in contrast 62% of the inseminated embryos were diploid with 48% showing evidence of a Y chromosome. The presence of the Y chromosome in many of the embryos that developed from 0PN and 1PN zygotes was also documented here, confirming that these embryos were not the product of parthenogenesis. In the IVF laboratory, pronuclear status and the subsequent development rate and morphology are commonly used parameters to select embryos for uterine transfer. In 2005, Rienzi reviewed the morphological attributes of the early embryo and stressed the importance of developmental inferiority reflected in an abnormal pronuclei pattern (Rienzi et al., 2005). In the present study, a strong relationship between abnormal pronuclear development and aneuploidy was found, despite the fact that many of the abnormal pronuclear embryos displayed developmental competence. The importance of these early morphological attributes should not be overlooked and may lead to a more accurate method of embryo selection. Recent studies using PGS to monitor chromosome number suggest that another parameter, ploidy, must also be taken into consideration when choosing the embryo with the highest implantation potential. In embryos, the ploidy of each individual blastomere is determined by events that occur during egg maturation, at the time of pronuclear formation, or during spindle formation and chromosomal alignment on the spindle during any of the first five post-fertilization mitotic divisions. In the present study, early embryo development to day 3 progressed similarly in embryos derived from 1PN and 2PN zygotes, but was significantly slower in 0PN-derived embryos. Significantly, no 0PN embryo developed to blastocyst stage by day 5. However, the blastocyst development rate of 29% observed for embryos developing from the 1PN zygotes was not significantly lower than was noted for those that developed from 2PN zygotes. Despite the developmental competence described above, many of the morphologically sound embryos carried a blastomere with one or more trisomies or worse. Even when graded morphologically suitable for uterine transfer on day 3, significantly more embryos from 1PN zygotes were chromosomally abnormal when compared with those that developed from 2PN zygotes. Moreover, despite the fact that blastocysts developed from 1PN zygotes, the majority of the biopsied blastomeres from these embryos were chromosomally abnormal. Even when a Y chromosome signified paternal fertilization, the majority of the embryos that developed from 1PN zygotes were aneuploid. Most aneuploid embryos lack the capacity to implant after transfer to the womb. A small proportion will initially adhere to the uterus but will subsequently miscarry within the first trimester of pregnancy. Few aneuploid embryos result in live born offspring, a clinical situation better avoided. Importantly, there was no embryo implantation when 0PN- or 1PN-derived embryos were chosen for uterine replacement even though they were all top quality. Thus, even though normal patterns of development can be observed over the first 5 days of embryonic life, the presence of an abnormal pronuclear number RBMOnline®
at 16–18 h post-sperm insemination portends chromosomal incompetence. It is the authors’ opinion that a clinic policy requiring PGS to screen 0-PN or 1-PN derived embryos prior to transfer should await the completion of more definitive PGS or other genetic studies to determine their chromosomal constitution and the cellular origin(s) of their zygote antecedents. Only when this data is in hand, can informed decisions be made about the safety of transferring these embryos.
References Abbott AL, Ducibella T 2001 Calcium and the control of mammalian cortical granule exocytosis. Frontiers in Bioscience 6, D792–D806. Bahçe M, Cohen J, Munné S 1999 Preimplantation genetic diagnosis of aneuploidy: were we looking at the wrong chromosomes? Journal of Assisted Reproduction and Genetics 16, 176–181. Balaban B, Kayhan Y, Bulent U et al. 2004 Pronuclear morphology predicts embryo development and chromosome constitution. Reproductive BioMedicine Online 8, 695–700. Balakier H, MacLusky N, Casper R 1993 Characterization of the first cell cycle in human zygotes implications for cryopreservation. Fertility and Sterility 59, 359–365. Bleil JD, Wassarman PM 1981 Mammalian sperm-egg interaction: Fertilization of mouse eggs triggers modifications of the major zona pellucida glycoprotein, ZP2. Developmental Biology 86, 189–197. Capmany G, Taylor A, Braude P et al. 1996 Cell cycle regulations: The timing of pronuclear formation, DNA synthesis and cleavage in the human 1-cell embryo. Molecular Human Reproduction 2, 299–306. Collas P, Poccia D 1998 Remodeling the sperm nucleus into a male pronucleus at fertilization. Theriogenology 43, 129–140. Evans JP, Florman HM 2002 The state of the union: the cell biology of fertilization. Nature Cell Biology 4, S57-S63. Feenan K, Herbert M 2006 Can ‘abnormally’ fertilized zygotes give rise to viable embryos? Human Fertility (Cambridge) 9, 157-169. Flaherty SP, Payne D, Swann NJ et al. 1995 Aetiology of failed and abnormal fertilization after intracytoplasmic sperm injection. Human Reproduction 10, 2623–2629. Grifo JA 1992 Preconception and preimplantation genetic diagnosis: Polar body, blastomere, and trophectoderm biopsy. In: Cohen J, Malter HE, Talansky BE, Grifo J (eds) Micromanipulation of Gametes and Embryos. Raven Press, New York, pp. 223–249. Kaufman M 1983 Early Mammalian Development: Parthenogenetic Studies. Cambridge University Press, Cambridge, UK. Keefe D, Liu L, Wang W et al. 2003 Imaging meiotic spindles by polarization light microscopy: principles and applications to IVF. Reproductive BioMedicine Online 7, 24–29. Konc J, Kanyó K, Cseh S 2004 Visualization and examination of the meiotic spindle in human oocytes with PolScope. Journal of Assisted Reproduction and Genetics 21, 349–353. Levron J, Munné S, Willadsen S et al. 1995 Male and female genomes associated in a single pronucleus in human zygotes. Biology of Reproduction 52, 653–657. Manor D, Kol S, Lewit N et al. 1996 Undocumented embryos: do not trash them, FISH them. Human Reproduction 11, 2502–2506. Marieb E 2001 Human Anatomy and Physiology 5th edn. Benjamin Cummings, San Francisco, pp. 1119–1122. Moore K, Persaud TVN 2003 The Developing Human: Clinically Oriented Embryology 7th edn. Saunders, Philadelphia. Munné S, Cohen J 1998 Chromosome abnormalities in human embryos. Human Reproduction Update 4, 842–855. Munné S, Weier U 1996 Simultaneous enumeration of chromosomes 13, 18, 21, X and Y in interphase cells for preimplantation genetic diagnosis of aneuploidy. Cytogenetics and Cell Genetics 75, 263–270.
787
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. Munné S, Tang Y, Grifo J et al. 1993 Origin of single pronucleated zygotes. Journal of Assisted Reproduction and Genetics 10, 276–279. Munné S, Magli C, Bahçe M et al. 1998 Preimplantation diagnosis of the aneuploidies most commonly found in spontaneous abortions and live births: XY, 13, 14, 15, 16, 18, 21, 22. Prenatal Diagnosis 18, 1459–1466. Nagy Z, Janssenswillen C, Janssens R et al. 1998 Timing of oocyte activation, pronucleus formation and cleavage in humans after intracytoplasmic sperm injection (ICSI) with testicular spermatozoa and after ICSI or in-vitro fertilization on sibling oocytes with ejaculated spermatozoa. Human Reproduction 13, 1606-1612. Payne D, Flaherty S, Barry M et al. 1997 Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Human Reproduction 12, 532–541. Plachot M 1991 Chromosome analysis of oocytes and embryos. In: Verlinsky Y, Kuliev A (eds) Preimplantation Genetics Plenum Press, New York, pp. 103–112. Rienzi L, Ubaldi F, Iacobelli M et al. 2005 Significance of morphological attributes of the early embryo. Reproductive BioMedicine Online 10, 669–681. Sathananthan A, Ratnam S, Ng S et al. 1996 The sperm centriole: its inheritance, replication and perpetuation in early human embryos. Human Reproduction 11, 345–356. Simerly C, Wu GJ, Zoran S et al., 1995 The paternal inheritance of the centrosome, the cell’s microtubule-organizing center, in humans, and the implications for infertility. Nature Medicine 1, 47–52.
Staessen C, Van Steirteghem A 1997 The chromosomal constitution of embryos developing from abnormally fertilized oocytes after intracytoplasmic sperm injection and conventional in-vitro fertilization. Human Reproduction 12, 321–327. Staessen C, Janssenwillen C, Devroey P et al. 1993 Cytogenetic and morphological observations of single pronucleated human oocytes after in-vitro fertilization. Human Reproduction 8, 221–223. Sultan K, Munné S, Palermo G et al. 1995 Chromosomal status of uni-pronuclear human zygotes following IVF and intracytoplasmic sperm injection. Human Reproduction 10, 132–136. Tesarik J, Greco E 1999 The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Human Reproduction 14, 13181323. Tesarik J, Mendoza C 1996 Spermatid injection into human oocytes. I. Laboratory techniques and special features of zygote development. Human Reproduction 11, 772–779. Williams CJ 2002 Signalling mechanisms of mammalian oocyte activation. Human Reproduction Update 8, 313-321. Xu Y, Zhang J, Grifo J et al. 2005 DNA methylation patterns in human tripronucleate zygotes. Molecular Human Reproduction 11, 167–171.
Declaration: The authors report no financial or commercial conflicts of interest. Received 3 March 2008; refereed 23 April 2008; accepted 25 June 2008.
788 RBMOnline®
Article Embryo biopsy: the fate of abnormal pronuclear embryos Nicole Noyes MD has been practicing reproductive endocrinology since 1990. She is currently a full-time Associate Professor at the New York University (NYU) School of Medicine, USA and is board-certified in obstetrics and gynaecology as well as reproductive endocrinology. She is co-director of the NYU oocyte cryopreservation program. Her clinical work has focused on infertility, fertility preservation and reproductive surgery and research efforts on implantation and early embryo development. Dr Noyes has been involved in the treatment of over 16,000 infertility patients using assisted reproductive technologies and has authored over 40 publications in peer-review journals.
Dr Nicole Noyes Nicole Noyes1, M Elizabeth Fino, Lewis Krey, Caroline McCaffrey, Alexis Adler, James Grifo NYU Fertility Centre, New York University School of Medicine, New York, New York, USA 1 Correspondence: e-mail: [email protected]
Abstract This study assessed 1908 embryos, including those with abnormal numbers of pronuclei, in IVF cycles from July 2001 to December 2006 in which preimplantation genetic screening (PGS) was performed on day 3 post-retrieval and ‘euploid’ embryos transferred the following day. PGS-intracytoplasmic sperm injection and PGS-translocation cycles were excluded. At 18 h post-insemination, the zygote distribution was 19% 0PN, 4% 1PN, 69% 2PN and 8% 3PN. No pregnancy occurred following 0PN or 1PN embryo transfers. A retrospective, blinded morphological ranking of all embryos on day 3 was performed and the results compared with PGS; no 0PN or 1PN embryo would have been chosen for transfer based on morphological superiority alone. Blastocyst formation occurred in 1PN embryos (29%) but not in 0PN embryos when evaluated on day 5. Euploid karyotypes were reported for biopsies of 0PN (3%), 1PN (5%) and 2PN (19%) embryos (P = 0.015, 1PN versus 2PN). A Y chromosome was observed in 0PN (17%) and 1PN (32%) embryos; surprisingly, 91% of these Y chromosome-bearing embryos were aneuploid. Many different meiotic and fertilization errors can result in 0PN or 1PN zygotes; these results indicate that the resultant embryos should not be transferred, especially when normally fertilized embryos are available. Keywords: chromosomes, embryo biopsy, embryo morphology, fertilization, preimplantation genetic screening, pronuclear
Introduction The time it takes to complete meiosis and the first mitotic cell cycle is known to vary greatly from one oocyte to another (Capmany et al., 1996). Following ovulation, human oocytes normally arrest at meiotic metaphase II and remain there until a spermatozoon enters, triggering resumption of the meiotic division and consequent extrusion of chromosomes in a second polar body. The appearance of two haploid pronuclei (2PN; one paternally and one maternally derived) marks the postfertilization transition from zygote into G1 of the first embryonic mitosis. This can occur anywhere from 3 to 20 h post-sperm insemination or injection (Balakier et al., 1993). Approximately 5% of human zygotes exhibit a single pronucleus rather than the expected two pronuclei when evaluated at 16–18 h postinsemination and this abnormality does not appear to be age related (Staessen et al., 1993).
782
In the early years of IVF, zygotes exhibiting only one pronucleus (1PN) were thought to have arisen from
parthenogenetic activation rather than fertilization by a spermatozoon (Kaufman, 1983). However, with the advent of preimplantation genetic screening (PGS), the frequent presence of a Y chromosome within the genetic make-up of these 1PN zygotes suggested that fertilization had taken place (Plachot, 1991, Munné et al., 1993; Staessen et al., 1993). In addition to parthenogenetic activation, other currently proposed mechanisms for 1PN zygote formation include asynchronous PN formation (Munné et al., 1993; Staessen et al., 1993; Payne et al., 1997), failure of either male or female chromatin to form a pronucleus (Flaherty et al., 1995) and abnormal fusion of the male and female pronuclei. The latter includes formation of a common pronuclear envelope (Levron et al., 1995) and fusion of two pronuclei that later form a larger single pronucleus (Tesarik and Mendoza, 1996). Embryos lacking or having only one pronucleus are thought to be suboptimal. Previously published results have shown a relationship between pronuclear morphology and preimplantation embryo development. Balaban
© 2008 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB23 8DB, UK
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. and colleagues demonstrated that pronuclear morphology can predict the risk of developmental arrest, irregular cleavage and even aneuploidy in human embryos, suggesting that pronuclear asynchrony or inequality might be markers for abnormalities in genetic programming (Balaban et al., 2004). The goal of the present work was to assess the morphological competence and chromosomal complement of these atypical zygotes.
Materials and methods All IVF with PGS cycles performed from July 2001 to December 2006 were reviewed (n = 126 cycles, 1908 embryos). Institutional approval for this research review was obtained. All intracytoplasmic sperm injection (ICSI) cycles and cycles performed for translocation were excluded. Women of all ages were included (mean age: 37.6 ± 4 years, range: 25–47 years). Ovarian stimulation protocols were individualized to achieve adequate numbers of mature oocytes at retrieval. Most patients were down-regulated with a gonadotrophinreleasing hormone (GnRH) agonist (leuprolide acetate, Tap, Lake Forest, IL, USA) and then treated with combinations of injectable recombinant FSH (Gonal F®, Serono, Rockland, MA, USA; Follistim®, Organon, Orange, NJ, USA) and/or human menopausal gonadotrophin (Repronex® or Menopure®, Ferring, Suffern, NY). Alternatively, LH suppression was achieved using GnRH antagonists (ganirelix acetate, Organon; cetrorelix acetate, Serono). When lead follicles reached a mean diameter of 17–18 mm, human chorionic gonadotrophin (HCG) was administered, and approximately 34 h later, oocytes were collected by ultrasound-guided transvaginal aspiration and placed in human tubal fluid media (HTF; Irvine Scientific, Irvine, CA, USA) supplemented with 6% Plasmanate (5% USP plasma protein fraction; Bayer, Elkhart, IN, USA) overlaid with Sage mineral oil (Cooper Surgical, Trumbull, CT, USA). The partner’s sperm was collected on the day of oocyte retrieval and prepared via swim-up or isolate. Oocytes were fertilized by routine insemination (4–6 h after retrieval). Following insemination, laboratory assessment of pronuclei was carried out 16–18 h later and recorded. All zygotes with more than 2 pronuclei were excluded from PGS. All other embryos arising from 0PN, 1PN or 2PN zygotes that were suitable for biopsy (≥3 cells with at least one intact blastomere) were subjected to genetic evaluation by PGS on day 3 postretrieval. A total of 1278 embryos underwent PGS. When performing embryo biopsy for PGS, one cell was removed from the cleavage-stage embryos on the morning of day 3 post-oocyte retrieval (Grifo, 1992). The cell was then fixed and analysed according to protocols previously described by Munné and co-workers (Munné and Weier, 1996; Munné et al., 1998; Bahce et al., 1999) at Reprogenetics, Inc. (Livingston, NJ, USA) for genetic assessment. Fluorescence insitu hybridization (FISH) analysis was performed to quantify the number of sex chromosomes (X and Y), and autosomes 13, 15, 16, 17, 18, 21 and 22 (8 and 14 in select cases). Selected euploid embryos were transferred to the uterus on day 4 or 5 post-retrieval. The patient population undergoing prenatal genetic screening included those at an advanced maternal age, with or without recurrent failed IVF (n = 66, 52%), those who had previously RBMOnline®
suffered two or more spontaneous miscarriages (n = 57, 45%) and those who electively chose PGS for other reasons (n = 3, 2%). In a later retrospective review, researchers were blinded to PGS, PN assessment and pregnancy outcomes and the biopsied embryos were ranked for embryo transfer based only on their morphology 3 days post-oocyte retrieval, prior to embryo biopsy. Ranking of these embryos was performed by two examiners using a quality score based on cell number and fragmentation. Good quality embryos had the highest number of cells and least fragmentation; the worst quality embryos developed slowly and exhibited significant fragmentation. The grading system was as follows: Quality Score 1, ≥5 cells and ≤10% fragmentation; Quality Score 2, ≥5 cells with ≥15% fragmentation; Quality Score 3, any cell number with ≥20% fragmentation or any embryo with fewer than 5 total cells. Once ranked by morphology, embryos blindly ‘selected’ for embryo transfer were compared with PGS results and the embryo choices made at the time of uterine replacement. Pregnancy outcomes were noted. A clinical pregnancy was defined as an intrauterine gestation with documentation of fetal cardiac activity. Statistical analysis was performed using the chi-squared and Student’s t-test (Sigmastat, SYSTAT Software Inc., CA, USA) as appropriate, with alpha equal to 0.050. Yates correction for continuity was used when appropriate.
Results Table 1 lists the demographics and outcomes of all evaluated IVF-PGS cycles. Of 1933 retrieved oocytes, 1908 (99%) developed to zygote stage by 18 h; 69% (1316/1908) of the zygotes exhibited 2PN. The remaining 31% (592/1908) of the zygotes displayed an abnormal pronuclei number, with significantly more 0PN (368) zygotes than 1PN (78) or 3PN (146) zygotes (P < 0.001). Of the embryos deemed developmentally competent and suitable for biopsy and PGS, 1188/1316 (90%) developed from 2PN zygotes, 32/368 (9%) from 0PN zygotes (P < 0.001) and 58/78 (74%) from 1PN zygotes. There was no FISH result for 36/1278 (3%) of the embryos. The mean number of embryos transferred per treatment cycle was 1.5 ± 0.1. The clinical pregnancy rate was 34% (37/109). There was no transfer in 17 cycles because all the embryos were reported as chromosomally abnormal. The embryo implantation rate (sacs per embryo transferred) was 28% (54/190); however, the implantation rate was 0% (0/4) for embryos that developed from either 0PN or 1PN zygotes, and 29% (54/186) for embryos that developed from 2PN zygotes. All transfers of embryos derived from either 0PN or 1PN zygotes were single-embryo transfers due to the lack of other available chromosomally normal embryos in the patient’s cohort. This allowed for adequate assessment of the embryo implantation rate for all pronuclear subgroups. No patients received a combination of a 2PN with a 0PN- or 1PN-derived embryo. Overall, 222/1242 (18%) of the biopsies were reported to have a normal number of chromosomes (Table 2, Figure 1). A euploid chromosome complement was reported in only 3% (1/30) of the biopsies from embryos from 0PN zygotes and 5% (3/57) of those from embryos from 1PN zygotes; in contrast 19% (218/1155) of the biopsies from embryos from 2PN zygotes were reported to be chromosomally normal (P = 0.015 for 1PN
783
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al.
Table 1. Demographic characteristics of IVF-preimplantation genetic screening cycles (PGS). Characteristic
Value
Number of cycles Age (years) No. of oocytes/cycle No. of embryos (%) Total 2PN 3PN 1PN 0PN Biopsied embryos/cycle Embryos–no PGS result (%) Aneuploid embryos (%) Aneuploid embryos/cycle Euploid embryos transferred/cycle Clinical pregnancy rate (fetal heart) per cycle (%) per transfer (%) Embryo implantation rate (%)
126 37.6 ± 0.4 (range 25–47) 15 ± 0.7 (total 1933) 1908 (100) 1316 (69) 146 (8) 78 (4) 368 (19) 10 ± 0.5 (total 1278) 36 (3) 1020/1242 (82) 8 ± 0.4 1.5 ± 0.1 37/126 (29) 37/109 (34) 54/190 (28)
Data are expressed as mean ± SEM unless otherwise stated.
Table 2. Distribution of chromosomal abnormalities (in embryos with FISH result) in all pronuclear subgroups. Pronuclear Total Euploid Haploid Polyploid Complex Monosomy/ subgroup abnormal Trisomy 0PN 1PN 2PN
30 1 (3) 4 (13) 4 (13) 57 3 (5) 13 (23) 2 (4) 1155 218 (19) 44 (4) 77 (7)
20 (68) 20 (35) 383 (33)
1 (3) 19 (33) 433 (37)
Data expressed as n (%).
Percentage
80 70
0PN n = 30
60
1PN n = 57 2PN n = 1155
50 40 30 20 10
Monosomy/ trisomy
Complex abnormal
Polyploid
Haploid
Euploid
0
Figure 1. Distribution of chromosomal abnormalities in all pronuclear subgroups.
784 RBMOnline®
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. versus 2PN; not statistically significant for 0PN versus 2PN). Moreover, although only 23% (13/57) of the biopsies from embryos from 1PN zygotes were haploid, this incidence was significantly higher than for embryos from 2PN zygotes 4% (44/1155) (P < 0.001). Curiously, 4% (2/57) of the embryos that developed from a 1PN zygote had a biopsy that was polyploid. Biopsied cells from embryos that developed from a 0PN zygote routinely presented a complex abnormal karyotype of multiple chromosome anomalies 67% (20/30) (Table 2, Figure 1). Eighteen (32%) embryos that developed from 1PN zygotes and five (17%) from 0PN zygotes tested positive for a Y chromosome, indicating that sperm had penetrated the oocyte; however, a euploid karyotype was noted in biopsies from only two of the 18 1PN and none of the five 0PN Y-bearing embryos. More than half of the 0PN and a third of 1PN embryos had a complex abnormal karyotype (Table 3). The results of the blinded morphological rankings of embryos on day 3 are shown in Table 4. Most (88%) of the embryos from 0PN zygotes (28/32) were given a quality score of 3 and were unsuitable for transfer; in contrast >40% of the embryos from 1PN (25/58) and 2PN (570/1188) zygotes were scored top quality (P < 0.001). Despite the structural integrity of the
embryos given a top quality score (Score = 1), biopsies of these morphologically superior embryos revealed aneuploidy in 88% (22/25) and 77% (437/570) of embryos from 1PN and 2PN zygotes, respectively. When embryos were assessed retrospectively for developmental competence, only 38% (12/32) of embryos that developed from 0PN zygotes had advanced to 6 cells at day 3; in contrast 55% (32/58) and 60% (713/1188) of the embryos from 1PN and 2PN zygotes, respectively, did so (P = 0.017 for 0PN versus 2PN). The cell number achieved by day 3 post-retrieval in each of the groups is shown in Table 5 and Figure 2. The developmental disparity broadened among the groups on day 5; of those assessed on day 5, no embryo from a 0PN zygote developed to the blastocyst stage whereas 29% (11/38) and 45% (322/719) of the embryos from 1PN and 2PN zygotes, respectively, reached this stage (P < 0.005; 0PN versus 1PN and 0PN versus 2PN zygotes).
Discussion Normal fertilization involves an integrated series of temporal events (Collas and Poccia, 1998; Marieb, 2001; Moore and Persaud, 2003; Feenan and Herbert, 2006). After a spermatozoon
Table 3. Genetic assessment of 0PN and 1PN zygotes biopsied 3 days post-oocyte retrieval. Genetic characteristic
0PNa (n = 32)
1PNa (n = 58)
Euploid 1 (3) 3 (5) Haploid 4 (13) 13 (23) Polyploid 4 (13) 2 (4) Complex abnormalb 20 (67) 20 (35) Monosomy 1 (3) 8 (14) Trisomy 0 (0) 6 (110) Monosomy and trisomy 0 (0) 5 (9) No result/no nucleusa 2 (6) 1 (2) Sex chromosome (n) Abnormal X0 (12), Y0 (1), XXX (5), X0 (22), XXX (3), XXXX (2), XXXX (2), XXY (3), 00 (1) XYY (1) Normal XX (5), XY (1) XX (12), XY (17) Values are n (%), unless otherwise stated. a Percentages are expressed as the portion of those with results except for the ‘no nucleus’ category, which is expressed as the portion of the total biopsied. b The category ‘complex abnormal’ includes embryos with two or more chromosomes having abnormal counts, but not completely polyploid or haploid, and includes embryos found to have multiple monosomies (Munné et al., 1998).
Table 4. Results of blinded embryo grading (best = score 1, worst = score 3). Pronuclear subgroup
Embryo Quality Score (%) 1 2
0PN 1PN 2PN
1/32 (3%) 3/32(9%) 28/32 (88%) 25/58 (43%) 9/58 (16%) 24/58 (41%) 570/1188 (48%) 216/1188 (18%) 402/1188 (34%)
3
785 RBMOnline®
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al.
No. of 0PN 1PN 2PN Euploid only cells (n = 32) (n = 58) (n = 1188) (0PN+1PN+ 2PN) (n = 222) <5 5 6 7 8 9 ≥10
15 (47) 4 (13) 7 (22) 3 (9) 2 (6) 1 (3) 0 (0)
13 (22) 8 (14) 11 (19) 10 (17) 11 (19) 1 (2) 4 (7)
211 (18) 158 (13) 189 (16) 199 (17) 325 (27) 58 (5) 48 (4)
32 (14) 18 (8) 30 (14) 46 (21) 88 (40) 5 (2) 3 (1)
Data expressed as n (%).
passes through the zona pellucida, its plasmalemma fuses with the oolemma of the egg prior to its incorporation into the egg. This invaginatory process results in the triggering of waves of Ca2+ release from local cytoplasmic stores that activate the egg (Evans and Florman, 2002). Activation involves a series of responses by the egg including the release of cortical granules (Abbott and Ducibella, 2001) to modify the molecular structure of the zona pellucida’s inner layer to prevent the penetration of additional spermatozoa (Bleil and Wassarman, 1981) into the egg. Resumption of the second meiotic division culminates in the extrusion of the second polar body containing a haploid set of female chromosomes (Williams, 2002). After the maternal and paternal chromosomes decondense and the paternal chromosomes are selectively demethylated (Xu et al., 2005), pronucleus formation then proceeds. The male pronucleus forms near the site of sperm entry into the egg and becomes enveloped by a nuclear membrane of maternal origin (Collas and Poccia, 1998); the female pronucleus forms concurrently near the site of second polar body extrusion (Payne et al., 1997). Microfilaments radiating from the sperm centrosome then move the two pronuclei towards each other in the centre of the cell (Simerly et al., 1995; Sathananthan et al., 1996). Eventually the two pronuclear membranes disassemble and the parental chromosomes come together in a process known as syngamy to establish the diploid embryonic genome (Nagy et al., 1998, Sathananthan, 1998, Tesarik and Greco, 1999). Studies monitoring fertilization outcome following insemination have revealed that the majority of normally fertilized zygotes display two pronuclei after a 16–18 h period (Nagy et al., 1998; Rienzi et al., 2005), as was observed in the present study. Such a conclusion is based on the assumption that all the eggs were mature and completed the first meiotic division at the time of insemination. However, the possibility exists that some of these 0PN and 1PN zygotes may have resulted from a ‘normal’ fertilization that was delayed until the maturation process was completed. Unfortunately, in the present study, pronuclear status was not routinely rechecked unless there was an abundance of 0PN or 1PN zygotes within the patient’s cohort of zygotes. Whether this delay in fertilization results in good quality blastocysts or euploid embryos remains to be determined.
786
Despite the above reservations, there is little doubt that for many
100%
≥10 cell 9 cell 8 cell
Relative incidence (%)
Table 5. Rate of embryo cleavage 3 days post-oocyte retrieval for 0PN, 1PN and 2PN zygotes.
7 cell 6 cell 5 cell <5 cell 0% 0PN
1PN
2PN
Euploid only
Figure 2. Rate of embryo cleavage 3 days post-oocyte retrieval for 0, 1 and 2 pronuclei (PN) zygotes.
zygotes the observation of zero or one pronucleus at 16–18 h post-insemination reflects an anomalous fertilization. Current explanations for 0PN zygotes include accelerated breakdown of pronuclear membranes representing biological variation in the timing of intracellular events (Manor et al., 1996), or lack of visualization due to granularity (Munné and Cohen, 1998). Suggested mechanisms for the appearance of a 1PN zygote include parthenogenetic activation, asynchronous development of the male and female pronuclei, pronuclear fusion, and failure of either male or female chromatin to form a pronucleus (Munné et al., 1993; Staessen et al., 1993; Flaherty et al., 1995; Levron et al., 1995; Tesarik and Mendoza, 1996; Payne et al., 1997). The latter includes failure of sperm DNA to decondense properly and abnormalities in sperm centrosome function. In time-lapse recordings of early embryos, 38% of the monitored zygotes developed pronuclei asynchronously (Payne et al., 1997). In another study, 25% of the 1PN zygotes at 16–18 h were found to have developed a second pronucleus when rechecked 4–6 h later (Staessen et al., 1993). Abnormalities in pronuclear membrane organization may explain some 1PN zygotes, particularly those where no additional pronuclei were seen on re-evaluation several hours later. Levron and co-workers suggested that in some cases of abnormal pronuclear formation, a common membranous envelope may form to surround both pronuclei (Levron et al., 1995). It is conceivable that this may occur when a spermatozoon enters or is deposited very close to the oocyte spindle, although studies that determined the site of sperm binding to the oolemma with polar body location do not support this hypothesis. PolScope reports that meiotic spindles are often located at considerable distances from the polar body suggest that sperm invagination by the egg may frequently occur near the spindle (Keefe et al., 2003; Konc et al., 2004). Prior karyotyping of 1PN embryos reported that 36–43% exhibited Y chromosome hybridization signals (Munné et al., 1993; Staessen and Van Steirteghem, 1997). These authors concluded that in approximately double the number of cases, X and Y chromosome-bearing sperm had penetrated the oocyte making an argument against parthenogenetic activation as the primary mechanism of uni-pronuclear formation. Sultan and RBMOnline®
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. colleagues compared the chromosome status of 1PN zygotes fertilized either by ICSI or normal sperm insemination (Sultan et al., 1995). Only 9.5% of the ICSI embryos were diploid and 9.5% contained a Y chromosome; in contrast 62% of the inseminated embryos were diploid with 48% showing evidence of a Y chromosome. The presence of the Y chromosome in many of the embryos that developed from 0PN and 1PN zygotes was also documented here, confirming that these embryos were not the product of parthenogenesis. In the IVF laboratory, pronuclear status and the subsequent development rate and morphology are commonly used parameters to select embryos for uterine transfer. In 2005, Rienzi reviewed the morphological attributes of the early embryo and stressed the importance of developmental inferiority reflected in an abnormal pronuclei pattern (Rienzi et al., 2005). In the present study, a strong relationship between abnormal pronuclear development and aneuploidy was found, despite the fact that many of the abnormal pronuclear embryos displayed developmental competence. The importance of these early morphological attributes should not be overlooked and may lead to a more accurate method of embryo selection. Recent studies using PGS to monitor chromosome number suggest that another parameter, ploidy, must also be taken into consideration when choosing the embryo with the highest implantation potential. In embryos, the ploidy of each individual blastomere is determined by events that occur during egg maturation, at the time of pronuclear formation, or during spindle formation and chromosomal alignment on the spindle during any of the first five post-fertilization mitotic divisions. In the present study, early embryo development to day 3 progressed similarly in embryos derived from 1PN and 2PN zygotes, but was significantly slower in 0PN-derived embryos. Significantly, no 0PN embryo developed to blastocyst stage by day 5. However, the blastocyst development rate of 29% observed for embryos developing from the 1PN zygotes was not significantly lower than was noted for those that developed from 2PN zygotes. Despite the developmental competence described above, many of the morphologically sound embryos carried a blastomere with one or more trisomies or worse. Even when graded morphologically suitable for uterine transfer on day 3, significantly more embryos from 1PN zygotes were chromosomally abnormal when compared with those that developed from 2PN zygotes. Moreover, despite the fact that blastocysts developed from 1PN zygotes, the majority of the biopsied blastomeres from these embryos were chromosomally abnormal. Even when a Y chromosome signified paternal fertilization, the majority of the embryos that developed from 1PN zygotes were aneuploid. Most aneuploid embryos lack the capacity to implant after transfer to the womb. A small proportion will initially adhere to the uterus but will subsequently miscarry within the first trimester of pregnancy. Few aneuploid embryos result in live born offspring, a clinical situation better avoided. Importantly, there was no embryo implantation when 0PN- or 1PN-derived embryos were chosen for uterine replacement even though they were all top quality. Thus, even though normal patterns of development can be observed over the first 5 days of embryonic life, the presence of an abnormal pronuclear number RBMOnline®
at 16–18 h post-sperm insemination portends chromosomal incompetence. It is the authors’ opinion that a clinic policy requiring PGS to screen 0-PN or 1-PN derived embryos prior to transfer should await the completion of more definitive PGS or other genetic studies to determine their chromosomal constitution and the cellular origin(s) of their zygote antecedents. Only when this data is in hand, can informed decisions be made about the safety of transferring these embryos.
References Abbott AL, Ducibella T 2001 Calcium and the control of mammalian cortical granule exocytosis. Frontiers in Bioscience 6, D792–D806. Bahçe M, Cohen J, Munné S 1999 Preimplantation genetic diagnosis of aneuploidy: were we looking at the wrong chromosomes? Journal of Assisted Reproduction and Genetics 16, 176–181. Balaban B, Kayhan Y, Bulent U et al. 2004 Pronuclear morphology predicts embryo development and chromosome constitution. Reproductive BioMedicine Online 8, 695–700. Balakier H, MacLusky N, Casper R 1993 Characterization of the first cell cycle in human zygotes implications for cryopreservation. Fertility and Sterility 59, 359–365. Bleil JD, Wassarman PM 1981 Mammalian sperm-egg interaction: Fertilization of mouse eggs triggers modifications of the major zona pellucida glycoprotein, ZP2. Developmental Biology 86, 189–197. Capmany G, Taylor A, Braude P et al. 1996 Cell cycle regulations: The timing of pronuclear formation, DNA synthesis and cleavage in the human 1-cell embryo. Molecular Human Reproduction 2, 299–306. Collas P, Poccia D 1998 Remodeling the sperm nucleus into a male pronucleus at fertilization. Theriogenology 43, 129–140. Evans JP, Florman HM 2002 The state of the union: the cell biology of fertilization. Nature Cell Biology 4, S57-S63. Feenan K, Herbert M 2006 Can ‘abnormally’ fertilized zygotes give rise to viable embryos? Human Fertility (Cambridge) 9, 157-169. Flaherty SP, Payne D, Swann NJ et al. 1995 Aetiology of failed and abnormal fertilization after intracytoplasmic sperm injection. Human Reproduction 10, 2623–2629. Grifo JA 1992 Preconception and preimplantation genetic diagnosis: Polar body, blastomere, and trophectoderm biopsy. In: Cohen J, Malter HE, Talansky BE, Grifo J (eds) Micromanipulation of Gametes and Embryos. Raven Press, New York, pp. 223–249. Kaufman M 1983 Early Mammalian Development: Parthenogenetic Studies. Cambridge University Press, Cambridge, UK. Keefe D, Liu L, Wang W et al. 2003 Imaging meiotic spindles by polarization light microscopy: principles and applications to IVF. Reproductive BioMedicine Online 7, 24–29. Konc J, Kanyó K, Cseh S 2004 Visualization and examination of the meiotic spindle in human oocytes with PolScope. Journal of Assisted Reproduction and Genetics 21, 349–353. Levron J, Munné S, Willadsen S et al. 1995 Male and female genomes associated in a single pronucleus in human zygotes. Biology of Reproduction 52, 653–657. Manor D, Kol S, Lewit N et al. 1996 Undocumented embryos: do not trash them, FISH them. Human Reproduction 11, 2502–2506. Marieb E 2001 Human Anatomy and Physiology 5th edn. Benjamin Cummings, San Francisco, pp. 1119–1122. Moore K, Persaud TVN 2003 The Developing Human: Clinically Oriented Embryology 7th edn. Saunders, Philadelphia. Munné S, Cohen J 1998 Chromosome abnormalities in human embryos. Human Reproduction Update 4, 842–855. Munné S, Weier U 1996 Simultaneous enumeration of chromosomes 13, 18, 21, X and Y in interphase cells for preimplantation genetic diagnosis of aneuploidy. Cytogenetics and Cell Genetics 75, 263–270.
787
Article - Fate of abnormal pronuclear embryos in IVF - N Noyes et al. Munné S, Tang Y, Grifo J et al. 1993 Origin of single pronucleated zygotes. Journal of Assisted Reproduction and Genetics 10, 276–279. Munné S, Magli C, Bahçe M et al. 1998 Preimplantation diagnosis of the aneuploidies most commonly found in spontaneous abortions and live births: XY, 13, 14, 15, 16, 18, 21, 22. Prenatal Diagnosis 18, 1459–1466. Nagy Z, Janssenswillen C, Janssens R et al. 1998 Timing of oocyte activation, pronucleus formation and cleavage in humans after intracytoplasmic sperm injection (ICSI) with testicular spermatozoa and after ICSI or in-vitro fertilization on sibling oocytes with ejaculated spermatozoa. Human Reproduction 13, 1606-1612. Payne D, Flaherty S, Barry M et al. 1997 Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Human Reproduction 12, 532–541. Plachot M 1991 Chromosome analysis of oocytes and embryos. In: Verlinsky Y, Kuliev A (eds) Preimplantation Genetics Plenum Press, New York, pp. 103–112. Rienzi L, Ubaldi F, Iacobelli M et al. 2005 Significance of morphological attributes of the early embryo. Reproductive BioMedicine Online 10, 669–681. Sathananthan A, Ratnam S, Ng S et al. 1996 The sperm centriole: its inheritance, replication and perpetuation in early human embryos. Human Reproduction 11, 345–356. Simerly C, Wu GJ, Zoran S et al., 1995 The paternal inheritance of the centrosome, the cell’s microtubule-organizing center, in humans, and the implications for infertility. Nature Medicine 1, 47–52.
Staessen C, Van Steirteghem A 1997 The chromosomal constitution of embryos developing from abnormally fertilized oocytes after intracytoplasmic sperm injection and conventional in-vitro fertilization. Human Reproduction 12, 321–327. Staessen C, Janssenwillen C, Devroey P et al. 1993 Cytogenetic and morphological observations of single pronucleated human oocytes after in-vitro fertilization. Human Reproduction 8, 221–223. Sultan K, Munné S, Palermo G et al. 1995 Chromosomal status of uni-pronuclear human zygotes following IVF and intracytoplasmic sperm injection. Human Reproduction 10, 132–136. Tesarik J, Greco E 1999 The probability of abnormal preimplantation development can be predicted by a single static observation on pronuclear stage morphology. Human Reproduction 14, 13181323. Tesarik J, Mendoza C 1996 Spermatid injection into human oocytes. I. Laboratory techniques and special features of zygote development. Human Reproduction 11, 772–779. Williams CJ 2002 Signalling mechanisms of mammalian oocyte activation. Human Reproduction Update 8, 313-321. Xu Y, Zhang J, Grifo J et al. 2005 DNA methylation patterns in human tripronucleate zygotes. Molecular Human Reproduction 11, 167–171.
Declaration: The authors report no financial or commercial conflicts of interest. Received 3 March 2008; refereed 23 April 2008; accepted 25 June 2008.
788 RBMOnline®