Article Kinetic markers of human embryo quality using time-lapse recordings of IVF/ICSI-fertilized oocytes

Transcription

1 RBMOnline - Vol 17 No Reproductive BioMedicine Online; on web 30 July 2008 Article Kinetic markers of human embryo quality using time-lapse recordings of IVF/ICSI-fertilized oocytes Dr Josephine Lemmen obtained her PhD in 2002 from the Hubrecht Insitute for Developmental Biology and Stem Cell Research, Utrecht, The Netherlands. Her thesis was entitled Estrogens and their receptors in murine embryogenesis. In 2003 she obtained a 2-year Marie Curie Fellowship to study in-utero exposure to oestrogens and hormonal imprinting, using a transgenic mouse model at the Laboratory for Reproductive Biology in Copenhagen, Denmark. Since 2006 she has been working as a biologist at the Fertility Clinic of the Rigshospitalet in Copenhagen. Her main interests are early embryo development and scoring, and fertility treatment of HIV- and hepatitis-positive patients. Dr Josephine Lemmen JG Lemmen 1, I Agerholm 2, S Ziebe 1,3 1 The Fertility Clinic, Rigshospitalet, Copenhagen, Denmark; 2 The Fertility Clinic, Braedstrup Hospital, Denmark 3 Correspondence: sziebe@rh.regionh.dk; Tel: ; Fax: Abstract A time-lapse system was used to study the timing and coordination of events during early development from zygote to cleavage stage embryo. The aim was to identify markers linked to good-quality embryos and implantation. A total of 102 fertilized oocytes were followed for h. Events such as appearance and disappearance of (pro)nuclei and timing and synchronization of cell cleavage were logged as time points after fertilization. Averages for these events and their synchrony were calculated and linked with fertilization method, embryo quality and implantation success. Fertilized oocytes that developed into 4-cell embryos had an earlier pronuclei disappearance and first cleavage than those that developed to 3- or 2-cell embryos. Intracytoplasmic sperm injection-fertilized 4-cell embryos spent a significantly shorter period as 2-cell compared with IVF-fertilized embryos (P = ). Development in the time-lapse system was similar to their siblings cultured in normal incubators, suggesting that the data from the time-lapse system can be extrapolated to the clinic s laboratory setting. Early disappearance of pronuclei and onset of first cleavage after fertilization was correlated with a higher number of blastomeres on day 2 after oocyte retrieval. In addition, synchrony in appearance of nuclei after the first cleavage was significantly associated with pregnancy success (P < 0.05). Keywords: development, kinetics, marker, morphology, synchrony Introduction In order to provide new markers for clinical use, a timelapse system was used to obtain new basic knowledge about morphological events occurring during early embryo development. The time-lapse system enabled a continuous registration of early embryo development. The aim was to identify early morphological markers linked to development of good-quality embryos and implantation. Time-lapse photography of a human embryo has previously been successfully used to observe polar body extrusion and pronuclear formation (Payne et al., 1997) and to show internalization of fragments (Hardarson et al., 2002). In addition there have been time-lapse studies with animal embryos showing the localization of methylated DNA in preimplantation mouse embryos (Yamazaki et al., 2007). Studies of kinetics in human embryos have also included a study of oocyte activation, pronucleus formation and cleavage of intracytoplasmic sperm injection (ICSI)-fertilized human oocytes (Nagy et al., 1994). In the present study, fertilized two-pronuclei (2PN) zygotes were placed in the time-lapse system on the morning of day 1 and a picture was taken every fifth minute until 8.00 on day 2. A number of events such as appearance and disappearance of pronuclei or nuclei and time and synchronization of cell cleavage were logged as time points after fertilization. The average times for these events and the synchrony of them in different blastomeres were calculated. Finally these were linked with fertilization method, embryo quality and implantation success Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK

2 386 Materials and methods Oocytes fertilized with either IVF or ICSI methods were scored for the presence of pronuclei on the morning of day 1 after oocyte retrieval. Patients were recruited from the routine programme in the study clinic and were stimulated with either recombinant FSH (Gonal-F, Serono, Denmark; or Puregon, Organon, Denmark) or Menopur (Ferring, Denmark). Human chorionic gonadotrophin (Profasi, Serono) was given 36 h before oocyte retrieval. As the study involved exposing the oocytes to nothing else but visual examination, patient or Institutional Review Board consent were not required. Culture of oocytes, time of fertilization and embryo scoring In the clinic s routine programme, aspirations take place between 9.00 and and up to three oocyte cumulus-complexes are cultured together per well in a 4-well dish (NUNC, Roskilde, Denmark) in fertilization medium (Cook, Limerick, Ireland) covered with oil (Cook). IVF insemination was performed by adding 150,000 progressive motile sperm cells to each well (0.5 ml) at On the following day between 8.00 and 9.00, the oocytes are denuded and individually transferred to 25 µl drops of cleavage medium under mineral oil (both Cook). For ICSI insemination, the cumulus oocyte complexes were denuded using hyaluronidase (Cook). ICSI was performed between and (actual time was recorded) in 5 µl drops of cleavage medium under mineral oil (both Cook). After ICSI the oocytes were transferred to 25 µl drops of cleavage medium covered with mineral oil. Both IVF and ICSI oocytes were scored for the presence of pronuclei on day 1 between and and for early cleavage between and On the morning of day 2 between and all embryos are scored before transfer. Only one zygote was placed in the time-lapse microscope at a time. Selection was done on the morning of day 1 when fertilization was scored. The zygote was chosen randomly among all 2PN oocytes. For time-lapse recordings the 2PN oocyte was placed in a 25 µl drop of cleavage media under mineral oil, and placed in the time-lapse system between and on day 1. At on day 2, the time-lapse system finished recording and the embryos were returned to their original droplet in the incubator. Another embryologist scored the time-lapse embryos and their sibling embryos and the embryo(s) with the highest quality score was selected for transfer according to the clinic s normal procedure. A scoring system was used as previously described (Ziebe et al., 2003). Briefly, it consists of assessment of cell number, fragmentation (0%, 1 10%, 11 20%, 20 50%, >50%), blastomere size (similar size or more than 25% difference in size), multinuclearity and nuclear appearance (one nucleus in all blastomeres, one nucleus in some blastomeres or all blastomeres without nucleus). Time-lapse system The time-lapse system consists of a Nikon Diaphot 300 microscope with camera in a closed system, where both temperature and CO 2 are controlled. The time-lapse camera is controlled using the Visiopharm Integrator System programme (Visiopharm, Hørsholm, Denmark). After finishing recording, the following episodes/time points were registered: disappearance of pronuclei, first cleavage, appearance of nuclei in each of the two blastomeres, disappearance of nuclei in each of the two blastomeres, second cleavage of each of the two blastomeres and re-appearance of nuclei in the blastomeres. All time recordings were normalized to the exact time of insemination. However, in some cases, not all events could be evaluated in all embryos, because the nuclei might have been out of focus after the first cleavage. Data and statistical analysis Data were registered at time points after insemination: for calculation and statistical purposes these were registered as minutes, but in the figures they are shown as hours to facilitate interpretation of the results. For the statistical analysis of results shown in Figures 2, 3 and 5, the Shapiro Wilk test was used to test for normality using Statistics Package for Social Sciences (SPSS) 13.0 (SPSS Inc., USA). Not all data were normally distributed and therefore non-parametric tests were applied. When multiple groups were compared a Kruskal Wallis test was performed, followed by a Dunn s post-test using GraphPad Prism Linear trends were tested by applying analysis of variance (ANOVA) with a post-test for linear trends (Prism 3.02). When two groups are compared, a Mann Whitney U- test was performed. Results In total, time-lapse recordings were made for 102 2PN oocytes, of which 63 were fertilized with regular IVF and 39 were fertilized with ICSI. On day 2 after oocyte retrieval, 4% of the 2PN oocytes had not cleaved, 26% had developed into a 2-cell embryo, 12% into a 3-cell embryo, 52% into a 4-cell embryo and 6% into a 5-cell embryo. The development of the 2PN oocytes in the time-lapse system was identical to that of their sibling embryos; 5% not divided, 19% 2-cell, 10% 3-cell, 56% 4-cell and 10% 5 cell at 8.00 am on day 2 after oocyte retrieval. Overall, 82% of the fertilized oocytes developed as embryos with a quality sufficient for transfer or cryopreservation. This was equivalent to the development of their sibling embryos in the normal incubators (81%). In total, 29 embryos with timelapse recording were transferred in 22 single-embryo transfer cycles resulting in six pregnancies and seven double-embryo transfer cycles resulting in two pregnancies. Of these, 19 single-embryo transfers were performed with 4-cell embryos, allowing a more detailed analysis. Also 2-, 3- and 5-cell embryos were transferred but these groups were too small for individual analysis. The fate of the 2-cell embryos was mainly cryopreservation (18 embryos), six were discarded and three were transferred, one of which led to a pregnancy. The fate of the 3-cell was as follows: five were discarded, five were frozen, and two were transferred, one of which led to a pregnancy. The fate of the 5-cell embryos was mainly that they were frozen (four embryos), one was transferred and did not lead to a pregnancy, and one was donated for research. Patients were recruited from the clinic s routine programme and had an age range from 23 to 39 years, with a mean of 33 years ± 4.6. Dependent on the number of blastomeres of the embryo on the morning of day 2 after oocyte retrieval, the average time points of the scored events were calculated (Figure 1). In general, there was a

3 relation between the number of cells on day 2 and the onset of the scored events. The higher the cell number, the earlier the events take place. Disappearance of pronuclei Disappearance of pronuclei showed a significant and linear trend correlated to cleavage stage on day 2 (P = , slope = ). Fertilized oocytes that developed into 4r-cell embryos had a significantly earlier disappearance of their pronuclei compared with 2-cell embryos (P < 0.001, Figure 2a). First cleavage and appearance of nuclei A significant linear (P = , slope = 43.69) trend between the number of blastomeres on day 2 and the time of first cleavage was observed (Figure 2b). 2PN oocytes developing into 5 cell embryos had significantly earlier first cleavage than those developing into 2-cell embryos (P < 0.01). The 2PN oocytes developing into 4-cell embryos were significantly earlier in their first cleavage than those 2PN oocytes developing into 2- or 3-cell embryos (P < and P < 0.05). It was found that the appearance of nuclei after the first cleavage was positively correlated to the resulting cell number on day 2 (ANOVA: 4-cell versus 3-cell, P < 0.05; 4-cell versus 2-cell, P < 0.001; and 5 cell versus two-cell, P < 0.01, post test for linear trend P < 0.001, slope 1.226, Figure 2c). IVF versus ICSI Embryo development after ICSI was significantly faster than after IVF (P = 0.035). When the development of IVF-fertilized oocytes was compared to ICSI-fertilized oocytes, it was found that the 2-cell stage in ICSI-inseminated 4-cell embryos was approximately 2.5 h shorter than in IVF-inseminated 4-cell embryos (Figure 3) (P = ). A non-significant trend was found towards ICSI-inseminated embryos spending 1.3 h longer at the 3-cell stage compared with IVF-inseminated embryos. The time from the 2-cell stage to the 4-cell stage was 13.4 h for IVF-inseminated and 12.5 h for the ICSI-inseminated oocytes (Figure 3). Further, the time to initiation of the second cleavage was significantly longer in the IVF-inseminated embryos (P < 0.001) compared with ICSI-inseminated embryos. The time from the second cleavage until the nuclei appeared in the blastomeres was significantly longer in the IVF-inseminated embryos (P < 0.001) compared with ICSI-inseminated embryos (P < 0.001) (data not shown). Synchrony dependent on the cell number on day 2 Synchrony in appearance of nuclei after the first cleavage was not correlated to the resulting cell number on day 2, (data not shown). For the embryos completing the second cleavage (4- and 5 cell embryos), no difference in synchrony was found between second cleavage of blastomere 1 and second cleavage of blastomere 2 and the resulting cell number on day 2 (data not shown). This was, however, with a relatively small sample size in the 5 cell group. The age of the patient did not affect the kinetic parameters of 4-cell embryos. Impact of kinetics on embryo assessment The time-lapse recordings showed that in some embryos there were dramatic differences in fragmentation grade in the timelapse embryos just prior to the time of embryo assessment on day 2; an example is shown in Figure 4. Between the third and fourth picture of Figure 4, 1h 45 min had elapsed, in which the embryo managed to reduce the amount of fragmentation from 20% to 10% and, in addition, the blastomeres became equal in size. Pregnancy data Overall, 29 embryos were transferred after time-lapse recording resulting in eight pregnancies. A subgroup of 19 single-embryo transfers using 4-cell embryos after time-lapse recording resulted in six pregnancies. Comparing the time-lapse recordings between the single, 4-cell embryo transfers resulting in an ongoing pregnancy with those that did not implant showed that the implanted embryos had a faster appearance of nuclei in the first blastomere after first cleavage (Figure 5a). In addition, they had a significantly more synchronous appearance of the nuclei in the two blastomeres after the first cleavage (P < 0.05) (Figure 5b). The embryos resulting in a pregnancy may have a shorter time difference between the second cleavages compared with the embryos that did not result in a pregnancy, although no statistically significant difference was found (Figure 5c). No difference was found between pregnant and non-pregnant groups concerning regular early cleavage scoring (66% versus 58%) or whether all blastomeres contained nuclei on day 2 (100% versus 83%). A short time-lapse video recording showing examples of an embryo with synchronous appearance of nuclei and division is available online [ Representative images are shown in Figure 6. Discussion The development of the 2PN oocytes in the time-lapse system was the same as that of their sibling 2PN oocytes in the regular incubators. This indicates, primarily, that the time-lapse system does not impair embryo development and, secondly, that the knowledge obtained in the time-lapse system can be applied to the rest of the clinic s laboratory setting. This is in accordance with previous studies showing no impairment of embryo development by sequential exposure to light of mouse oocytes (Barlow et al., 1992), rabbit embryos (Bedford et al., 1989) and human oocytes (Payne et al., 1997). In general, these data show that all scored events occur earliest in those embryos that developed with the highest cell number on day 2 after oocyte retrieval. Those 2PN oocytes that develop into a 4-cell embryo on day 2 after oocyte retrieval show earlier disappearance of their pronuclei and first cleavage than those that develop into 2- or 3-cell embryos. When comparing IVF- and ICSI-fertilized oocytes, it was found that ICSI-fertilized 4-cell embryos spent approximately 2.5 h less in the 2-cell stage than IVF-fertilized 4-cell embryos. However, the ICSI- and IVF-fertilized embryos become 4-cell embryos at the same time point after insemination, suggesting that the ICSI embryos spent longer in the 3-cell stage. This could indicate that IVF-fertilized embryos have a higher synchrony 387

4 Figure 1. Time points of the different scored events, e.g. disappearance of pronuclei (PN), first and second cleavage, appearance of nuclei after first and second cleavage, and disappearance of nuclei after first cleavage. Data are shown grouped according to the scoring of the timelapse embryo on the morning of day 2 after oocyte retrieval. Values are mean ± SD; blast = blastomeres. 388 Figure 2. Time points: (a) of disappearance of pronuclei (PN), based on the scoring of the embryo by time-lapse recordings, on the morning of day 2 after oocyte retrieval; (b) of first cleavage, based on scoring of the embryo by time-lapse recordings, on the morning of day 2 after oocyte retrieval; (c) of appearance of nuclei, based on the scoring of the embryo by time-lapse recordings, on the morning of day 2 after oocyte retrieval. The horizontal line indicates mean for each group. The number of embryos in each group was: 2 cells, n = 27; 3 cells, n = 12; 4 cells, n = 53; 5+ cells, n = 6. Statistical differences between the groups were tested with a Kruskal Wallis test in Prism 3.0: * = P < 0.05; ** = P < 0.01; *** = P <

5 Figure 3. Time difference in divisions between IVF- and ICSI-fertilized embryos that contained four blastomeres on day 2 after oocyte retrieval. Different cleavage stages are shown, e.g. zygote, 2-cell, 3-cell, 4-cell, with different shades as indicated. In the 2-cell cleavage stage, the actual number of hours of this period is indicated. The actual time in the laboratory is indicated at the top and the time after fertilization is indicated at the bottom. Periods where laboratory personnel were scoring the embryos are indicated by the stippled boxes. The oval shape indicates the average time point of disappearance of the pronuclei. The number of embryos in the IVF group was 35 and the number of embryos in the ICSI group was 18. Figure 4. Kinetics of cellular division just prior to embryo scoring: an example of the dynamic cellular division, fragment formation and fragment uptake of an embryo that can occur just prior to scoring of the embryos on day 2 after oocyte retrieval. 389

6 Figure 6. Kinetic markers of human embryo quality using timelapse recordings of IVF/ICSI fertilized oocytes. (a) The nuclei appear almost simultaneous in the first two blastomeres. (b) The second division is almost simultaneous in both blastomeres. (c) The nuclei appear not simultaneous in the first two blastomeres. (d) The second division is later in one of the blastomeres, resulting in a longer period as 3-cell embryo. Images shown are representative stills from a short film available online at Figure 5. Pregnant versus not-pregnant groups after single embryo transfer of 4-cell embryos. (a) First cleavage: period in minutes from the first cleavage until nuclei appeared in the two blastomeres for 4-cell embryos that developed into an ongoing pregnancy (n = 10) compared with the blastomeres of 4-cell embryos that did not develop into a pregnancy (n = 19). (b) Synchrony in appearance of nuclei: period in minutes from when the nucleus appeared in the first blastomere until it appeared in the second blastomere after the first cleavage. This is shown for 4-cell embryos that developed into an ongoing pregnancy (n = 4) compared with 4-cell embryos that did not develop into a pregnancy (n = 6). (c) Time in minutes between the second cleavage of blastomere 1 until the second cleavage of blastomere 2, for the pregnant (n = 5) and notpregnant group (n = 12). Horizontal bars indicate mean values. Statistical differences between the groups were tested with a Mann Whitney U-test in Prism 3.0; * = P < in their second cleavage than ICSI-fertilized embryos. In accordance with others, these results show that a higher percentage of ICSI-fertilized embryos have early cleavage than IVF-fertilized embryos (Lundin et al., 2001; Giorgetti et al., 2007). From this study s data, it becomes clear that the ICSIfertilized embryos have their first cleavage on average 26 h after injection whereas IVF-fertilized embryos cleave 27 h after insemination. So a more individualized early cleavage scoring time, or at least one that is adjusted for insemination technique, would be essential before early cleavage scoring between IVF- and ICSI-fertilized embryos can be directly compared. There was no difference in early cleavage between the implanted and non-implanted 4-cell embryos transferred. Although it should be noted that the number of embryos in the present study was very limited, the results seem to contradict previous studies describing early cleavage as a strong separate prognostic factor, even though it was inversely correlated to embryo score (Van Montfoort et al., 2004). As all transferred embryos in this study were high quality, this could indicate a smaller importance of early cleavage in high-quality embryos as previously suggested (Lundin et al., 2001; Emiliani et al., 2006). Synchronous development is considered to be a positive parameter for embryo quality. Scott et al. (2007) show that 2- and 4-cell embryos on day 2 implant better than 3-cell or 5-cell embryos. In extension of these data, this study s results show that four-cell embryos that implant are more synchronous in their development, with a significantly shorter time from appearance of nuclei in the first blastomere to appearance of the nucleus in the second blastomere after the first cleavage (P < 0.05). In addition, the four-cell embryos that implant may have a shorter time difference between the second cleavages of the blastomeres and thus have a higher degree of synchrony. Reabsorption of fragments, as observed previously (Antczak and Van Blerkom, 1999; Van Blerkom et al., 2001; Hardarson

7 et al., 2002; Van Blerkom, 2007), was often seen in the timelapse embryos. It is well known that cellular division is a very dynamic process. The data from this study showed that, just prior to the scoring time point on day 2, there were dramatic differences in fragmentation grade and that many embryos finished their second cleavages very close to the scoring time point. This is in accordance with the conclusion by Scott et al. (2007) that the most important point when assessing embryo evaluations is time of scoring. To compensate for this, the fertilization time point in the authors laboratory has been changed to allow fertilization from onward on the day of oocyte retrieval, allowing the IVF-fertilized embryos more time to finish their second cleavages before scoring on the morning of day 2. In a study with ICSI-fertilized oocytes, Nagy et al. (1994) showed that h after fertilization (which roughly corresponds to the scoring-time on day 2, which is h after fertilization) 13% had become 2-cell, 53% had become 3- or 4-cell and 29% were more than 4-cell embryos. So in this study the ICSI-fertilized oocytes were markedly faster in their development than in this study s setting. One explanation for this could be that Nagy et al. used a different culture medium (B2) than the Cook cleavage medium used in the present study. Two studies comparing Cook medium with respectively P1 medium (Ben-Yosef et al., 2004) or G1.2 medium (Van Langendonckt et al., 2001) showed that embryos cultured in Cook medium had a slower development than embryos cultured in P1 or G1.2 medium. However, there is no information on the number of blastomeres on day 2 in these studies. The time-lapse recordings of embryos that are transferred have provided the opportunity to study the episodes not only related to embryo quality but also actually related to implantation. Embryos that implant have a faster appearance of nuclei in the first blastomere after the first cleavage and a more synchronous appearance of the nuclei in the two blastomeres after the first cleavage. This means that the number of visible nuclei in the early cleavage evaluation seems to be a parameter that could be considered to be included in the scoring. In conclusion, this study shows that there is extensive information to be obtained from time-lapse studies. A fast disappearance of pronuclei and first cleavage has been shown to be associated with a higher number of blastomeres on day 2 after oocyte retrieval. ICSI-fertilized 4-cell embryos spent a significantly shorter time as 2-cell embryos than IVF-fertilized embryos and synchrony in appearance of nuclei after the first cleavage was significantly associated with increased pregnancy success. In addition, the continuous monitoring of embryos during and after cleavage revealed a huge difference in two essential parameters in the scoring system: degree of fragmentation and blastomere evenness over time. An obvious risk is that wrong timing of scoring could have the effect that good-quality embryos are scored as poor embryos or even discarded. Acknowledgements The authors would like to acknowledge the help of the laboratory technicians of the Rigshospital Fertility Clinic. References Antczak M, Van Blerkom J 1999 Temporal and spatial aspects of fragmentation in early human embryos: possible effects on developmental competence and association with the differential elimination of regulatory proteins from polarized domains. Human Reproduction 14, Barlow P, Puissant F, Van der Zwalmen P et al In-vitro fertilization, development, and implantation after exposure of mature mouse oocytes to visible light. Molecular Reproduction and Development 33, Bedford JM, Dobrenis A 1989 Light exposure of oocytes and pregnancy rates after their transfer in the rabbit. Journal of Reproduction and Fertility 85, Ben-Yosef D, Amit A, Azem F et al Prospective randomized comparison of two embryo culture systems: P1 medium by Irvine Scientific and the Cook IVF Medium. Journal of Assisted Reproduction and Reproduction and Genetics 21, Emiliani S, Fasano G, Vandamme B et al Impact of the assessment of early cleavage in a single embryo transfer policy. Reproductive BioMedicine Online 13, Giorgetti C, Hans E, Terriou P et al Early cleavage: an additional predictor of high implantation rate following elective single embryo transfer. Reproductive BioMedicine Online 14, Hardarson T, Lofman C, Coull G et al Internalization of cellular fragments in a human embryo: time-lapse recordings. Reproductive BioMedicine Online 5, Lundin K, Bergh C, Hardarson T 2001 Early embryo cleavage is a strong indicator of embryo quality in human IVF. Human Reproduction 16, Nagy ZP, Liu J, Joris H et al Time-course of oocyte activation, pronucleus formation and cleavage in human oocytes fertilized by intracytoplasmic sperm injection. Human Reproduction 9, Payne D, Flaherty SP, Barry MF et al Preliminary observations on polar body extrusion and pronuclear formation in human oocytes using time-lapse video cinematography. Human Reproduction 12, Scott L, Finn A, O Leary T et al Morphologic parameters of early cleavage-stage embryos that correlate with fetal development and delivery: prospective and applied data for increased pregnancy rates. Human Reproduction 22, Van Blerkom J 2007 Translocation of the subplasmalemmal cytoplasm in human blastomeres: possible effects on the distribution and inheritance of regulatory domains. Reproductive BioMedicine Online 14, Van Blerkom J, Davis P, Alexander S 2001 A microscopic and biochemical study of fragmentation phenotypes in stageappropriate human embryos. Human Reproduction 16, Van Langendonckt A, Demylle D, Wyns C et al Comparison of G1.2/G2.2 and Sydney IVF cleavage/blastocyst sequential media for the culture of human embryos: a prospective, randomized, comparative study. Fertility and Sterility 76, Van Montfoort AP, Dumoulin JC, Kester AD et al Early cleavage is a valuable addition to existing embryo selection parameters: a study using single embryo transfers. Human Reproduction 19, Yamazaki T, Yamagata K, Baba T 2007 Time-lapse and retrospective analysis of DNA methylation in mouse preimplantation embryos by live cell imaging. Developmental Biology 304, Ziebe S, Lundin K, Loft A et al FISH analysis for chromosomes 13, 16, 18, 21, 22, X and Y in all blastomeres of IVF pre-embryos from 144 randomly selected donated human oocytes and impact on pre-embryo morphology. Human Reproduction 18, Declaration: The authors report no financial or commercial conflicts of interest. Received 4 December 2007; refereed 21 December 2007; accepted 25 April