Effect of IVF and laser zona dissection on DNA methylation pattern of mouse zygotes
- First Online:
- Cite this article as:
- Peters, D.D., Lepikhov, K., Rodenacker, K. et al. Mamm Genome (2009) 20: 664. doi:10.1007/s00335-009-9227-0
- 207 Downloads
In vitro fertilization (IVF) and zona pellucida laser microdissection-facilitated IVF (Laser-IVF) are presently routine procedures in human assisted reproduction. The safety of these methods at the epigenetic level is not fully understood. Studies on mouse Laser-IVF embryos provide evidence that the use of Laser-IVF leads to reduced birth rate, indicating a potential harm of this technique for the embryo. Hence, the aim of this study was to examine the difference in DNA methylation pattern between IVF- and Laser-IVF-derived mouse zygotes. We examined two experimental groups of C3HeB/FeJ oocytes: (1) zona-intact and (2) laser-microdissected oocytes that were fertilized in vitro with freshly collected spermatozoa. Zygotes were fixed 5, 8, and 12 h after fertilization, and indirect immunofluorescence staining was studied using an anti-5-methylcytidine (5-MeC) antibody. The fluorescence intensities of paternal and maternal pronuclei were evaluated using the computer-assisted analysis of digital images. In addition, we performed a semiquantitative RT-PCR analysis of the presence of transcripts of three developmental marker genes, Oct4, Dab2, and Dnmt3b, in IVF- and Laser-IVF-derived blastocysts. We observed no significant differences in methylation status of the paternal genome and in the transcripts of the developmental marker genes after IVF and Laser-IVF. In conclusion, epigenetic patterns and early embryonic development are not altered by laser-assisted IVF techniques and another explanation must be sought for the poor implantation rates observed in mice.
Since the initiation of large-scale mouse mutagenesis projects (Hrabe de Angelis and Balling 1998; Hrabe de Angelis et al. 2000; Nolan et al. 2000), numerous mouse lines have been generated and cryopreserved (Marschall and Hrabe de Angelis 1999; Nakagata 2000; Thornton et al. 1999). Cryopreservation of mouse germ cells generally provides a means for long-term storage of valuable mouse lines and facilitates the management of animal facilities. Subsequently, the cryopreserved mouse lines have to be recovered using thawed spermatozoa and in vitro fertilization or thawed embryos and embryo transfer (Marschall et al. 1999).
In vitro fertilization has been frequently used to overcome defects in the reproductive ability of mutant or aged mice (Marschall et al. 1999; Nakagata 2000; Sztein et al. 2000; Thornton et al. 1999). Inbred mouse strains have defined reproductive parameters, which are characteristic of the strain. Also, differences in the ability of spermatozoa to survive cryopreservation and to fertilize oocytes of the same genetic background were described in several studies (Nakagata 2000; Nakagata and Takeshima 1993; Sztein et al. 2000). For example, for mutant mouse lines on a C57BL/6 J background, embryo freezing after in vitro fertilization using fresh sperm of a mutant male and wild-type oocytes (Cryo-IVF) is the routine cryopreservation method, because frozen-thawed sperm of these strains has poor fertilizing ability (Landel 2005).
To increase the efficiency of fertilization the routinely used rederivation methods can be enhanced by additional assisted reproductive technologies (ART) such as zona pellucida removal (Ribas et al. 2006), partial zona dissection (PZD) (Depypere et al. 1988; Nakagata et al. 1997; Payne 1995), zona incision by a piezomanipulator (ZIP) (Kawase et al. 2002, 2004), subzonal insemination (SUZI) (Enginsu et al. 1995), or intracytoplasmic sperm injection (ICSI) (Szczygiel et al. 2002; Yanagimachi 1998). Since the 1990s laser-assisted systems have been used for zona microdissection of murine and human embryos (Enginsu et al. 1995; Germond et al. 1995, 1996; Hollis et al. 1997; Liow et al. 1996; Peters et al. 2006). The purpose of these laser systems was to improve the fertilization rate (el-Danasouri et al. 1993) or the hatching rate of the blastocysts (Obruca et al. 1994). The safety of the procedure was tested for some systems with respect to the potential damage at the ultrastructural (scanning and transmission electron microscopy) and biological (number of mice born, morphologic examination, and reproductive capacity of offspring) levels (Germond et al. 1995, 1996; Liow et al. 1996), but epigenetic safety of the procedure has not been studied yet. One study that used the same laser system (Octax laser) as in this study reported a significantly higher cleavage rate but, at the same time, a reduced birth rate of laser-dissected compared to normal IVF embryos (Boersma et al. 2007).
Normal embryonic development in mammals requires differential imprinting in both male and female genomes (McGrath and Solter 1984). After the oocyte is activated by sperm and completes meiosis, haploid oocyte chromosomes transform into a maternal pronucleus and the haploid sperm nucleus transforms into a paternal pronucleus. When sperm chromatin protamines are replaced with oocyte cytoplasmic histones, the extensive demethylation of paternal DNA occurs; this has been described as “active demethylation” (Mayer et al. 2000; Oswald et al. 2000). After completion of the first cell cycle the demethylation of the maternal genome proceeds and is referred to as the “passive phase” of demethylation (Howlett and Reik 1991). This DNA methylation pattern is important in the sex-specific germline “marking” of parental alleles and contributes to the appropriate expression of many imprinted and developmentally important genes (Reik and Walter 2001; Reik et al. 2001).
Previous studies have shown that preimplantation embryos are sensitive to the environment and manipulation can effect developmental potential and fetal growth. For example, the methylation pattern of imprinted genes has been shown to change in response to embryo culture conditions (Doherty et al. 2000; Khosla et al. 2001a, b). In addition, zona-free embryos in which the zona pellucida was removed between 1 and 8 h post fertilization show a significant reduction in the DNA methylation level at two-cell and four-cell stages (Ribas et al. 2006). These results indicate that changes in the environment of the embryo, such as in vitro culture and mechanical or chemical manipulation, can cause some epigenetic modifications. On the other hand, Fulka and Fulka (2006) found no differences in the DNA methylation pattern in mouse zygotes produced in vivo, in vitro, or by intracytoplasmic sperm injection. For this reason the analysis of very early changes occurring just after fertilization is an essential step in the assessment of the safety of given procedures. Thus, the safety of all ART methods with respect to epigenetic patterns should be evaluated. This is especially important because these techniques are widely applied for the archiving and rederivation of valuable mice mutant strains in international repositories such as the European Mouse Mutant Archive (EMMA; www.emmanet.org). Furthermore, the mouse serves as a model organism to estimate the safety of procedures before their application in human assisted reproduction.
The analysis of the expression of marker genes for some developmentally important genes using RT-PCR provides additional valuable information about correct embryonic development. Several genes have been shown to be indispensable for embryonic development, and the properly orchestrated gene expression profile accompanies the formation of two distinct cell lineages in blastocyst: trophectoderm and inner cell mass (ICM). For example, the pluripotency state of ICM cells is characterized by the expression of specific genes, including Oct4 (Scholer et al. 1990), Dab2 (Yang et al. 2002), and Dnmt3b (Hirasawa and Sasaki 2009).
In this study we investigated the effect of IVF and Laser-IVF on mouse zygotes. The formation of pronucleus (PN) stages, the DNA methylation pattern, the methylation status of the paternal genome at three different time points, and the expression of three developmentally important genes (Oct4, Dab2, Dnmt3b) after IVF and Laser-IVF are also examined.
Materials and methods
Inbred C3HeB/FeJ mice served as sperm (12-week-old) and oocyte (56-day-old) donors. Mice were bred at the Helmholtz Zentrum Munich. The animal facility was tested for microorganisms according to the FELASA recommendations for mouse health monitoring (Nicklas et al. 2002). The husbandry conditions were as follows: room temperature at 20–24°C, air humidity was 50–60%, there were 20 air changes per hour, and the lighting regimen was on a 12-h light/dark cycle. Wood shavings (Altromin GmbH, Lage; Germany) were used as bedding material. Mice were fed a standardized mouse diet (1314 Altromin) and drinking water was available ad libitum.
Collection of sperm samples
Mice were killed by cervical dislocation. Both caudae epididymides and vasa deferentia were dissected and placed in 220 μl of human tubal fluid (HTF) medium which was prepared from reagent-grade chemicals (Sigma-Aldrich, Taufkirchen, Germany) as Quinn et al. (1985) reported. The tissues were cut with a sterile injection needle (B. Braun, Melsungen, Germany) and the sperm were allowed to swim out and disperse for 10 min at 37°C and 5% CO2. The dish was gently shaken until the sperm suspension was homogeneous and 6 μl of the sperm suspension was placed in 500 μl of HTF medium under mineral oil (embryo tested, Sigma-Aldrich, Taufkirchen, Germany). The HTF drops with spermatozoa were incubated at 37°C and 5% CO2 and used for in vitro fertilization. The sperm were capacitated for 1.0 h. The sperm quality was determined using a Hamilton Thorn IVOS computerized semen analyzer (Hamilton Thorn, Beverly, MA) as previously reported by Sztein et al. (2000).
In vitro fertilization
Female mice were superovulated by intraperitoneal (i.p.) injection of 5 IU of eCG (Intergonan 1000, Intervet, Unterschleißheim, Germany) followed by 5 IU of hCG (Ovogest 1500, Intervet) 48 h later. The oocytes were collected 14 h after hCG injection according to standard procedures (Nagy et al. 2003). Mice were killed by cervical dislocation and the oviducts were dissected. The cumulus-oocytes complexes were isolated from oviducts in a 500-μl HTF drop under mineral oil. The oocytes were denuded of cumulus cells by washing in M2 medium containing 600 μg/ml of bovine testes hyaluronidase (Sigma-Aldrich, Taufkirchen, Germany). Four aliquots of 100 μl of M2 medium (Sigma-Aldrich) were placed into petri dishes under mineral oil and denuded oocytes were distributed into these drops. Oocytes from two drops of M2 medium were used for IVF without laser microdissection. Oocytes from two other drops of M2 medium were used for Laser-IVF. After laser microdissection all oocytes were added immediately to the appropriate fertilization dishes with capacitated sperm samples. After 5 h of incubation (5% CO2, 37°C, and 95% humidity), the oocytes were washed four times through four medium changes (100 μl each) of KSOM (Powdered Media Kit, Metachem Diagnostics Ltd., Northampton, UK) and incubated in a CO2 incubator in a fifth drop of 100 μl of KSOM until 8 or 12 h post fertilization.
Laser microdissection with Octax Laser
The Octax Laser Shot System with a 1.48-μm infrared diode laser (MTG Medical Technology Vertriebs-GmbH, Altdorf, Germany) was used for the microdissection of the zona pellucida. The laser system was adapted to an inverted microscope (Leica DM IRB, Leica Microsystems GmbH, Wetzlar, Germany), and images of the oocytes were transmitted by video camera (Octax Eye USB Camera) to the monitor system. Petri dishes were fixed on the microscope stage and holes of about 7-μm diameter were cut through the zona pellucida. Under visual control, the zona pellucida was partly, but not completely, penetrated. The success of the microdissection procedure was verified on the computer screen and oocytes with completely disrupted zona pellucida were discarded. After laser microdissection, the oocytes were placed into fertilization dishes with capacitated spermatozoa.
Fixation of mouse zygotes
All procedures were performed on microscope slides with cavities (Carl Roth, Karlsruhe, Germany), siliconized with 5% dimethyldichlorosilane in toluene (Sigma-Aldrich, Taufkirchen, Germany). The zona pellucida was removed by incubation for 1–2 min in 0.1 ml of acidic Tyrode’s solution (Sigma-Aldrich). Directly after zona removal zygotes were fixed in 0.1 ml of 3.7% paraformaldehyde in 1× PBS for 30 min at room temperature (RT).
All procedures were carried out at RT in a humidified chamber. The immunostainings were performed in a blocking solution (BS) which consisted of 1% BSA and 0.1% Triton X100 in 1× PBS. After fixation with 3.7% PFA the zygotes were permeabilized with 0.2% Triton X100 in 1× PBS for 20 min and DNA was denatured for 10 min with 4 M HCl. After denaturation embryos were neutralized for 10 min with 0.1 M Tris in 1× PBS (pH 8.0) and post-fixed for 30 min with 3.7% PFA. After permeabilization with 0.2% Triton X100 in 1× PBS for 10 min, embryos were blocked for 3 h in BS. Subsequently, embryos were incubated with the primary anti-5-methylcytidine antibody (Eurogentec, Köln, Germany, Cat. No. BI-MECY-0100) for 3 h in BS. Washing was carried out three times with 100 μl of BS for 1 h. Incubation with the secondary antibody Alexa Fluor 488 goat anti-mouse IgG (Invitrogen GmbH, Karlsruhe, Germany, Cat. No. A11001) was performed for 2 h in BS followed by subsequent washes with BS. In addition, propidium iodide (PI) was added to BS for 15 min after a second wash with BS. Afterward embryos were equilibrated for 5 min in a small drop of mounting medium (Vectashield, Linaris, Germany). In the meantime clean microscope slides with a small drop (1–2 μl) of the mounting medium were prepared and up to 20 zygotes were put into one drop. A tiny drop of mounting solution was placed on the coverslip and gently (using fine forceps) put over the drop with the embryos. After 5 min the coverslips were sealed with nail polish.
Digital image microscopy and analysis
The slides were analyzed on a Zeiss Axiovert 200 M fluorescence microscope equipped with a 40 ×/0.75 Plan-Neofluar lens and the filter sets for Alexa Fluor 488 (with excitation 450–490 and emission 515–565) and PI (with excitation 540–552 and emission 575–640) and the digital black/white camera Axiocam HRm for imaging. The final images were generated using Zeiss Axiovision software 4.5.
The statistical package SAS v9.1 (SAS Institute, Cary, NC, USA) was used to analyze the difference between the anti-5-MeC fluorescence intensity patterns of paternal and maternal pronuclei seen in IVF- and Laser-IVF-derived mouse oocytes in order to assess the difference in DNA methylation patterns. Simple data descriptions and comparisons were set up by procedures UNIVARIATE and NPAR1WAY. The SAS procedures REG and GLM were used to develop synoptic interactive models controlling for treatment of relative fluorescence intensities on time, paternal intensities on maternal intensities, and paternal area on maternal area of the pronuclei. In these models, gender of genome and treatment are dummy variables coding 0 for the male and 1 for the female genome, and coding 0 for IVF and 1 for Laser-IVF treatment. These models fit the data reasonably well. The SAS procedure CATMOD was used to develop a log linear model for the time pattern of the methylation states of pronuclei by treatment.
Semiquantitative RT-PCR analysis
Primers used for PCR
Primer sequences 5′ → 3′
Annealing temperature (°C)
Fragment size (bp)
Formation of pronucleus stages after IVF and Laser-IVF
Five hours after fertilization most of the IVF and Laser-IVF zygotes reached PN stage 2 or 3 (86 and 91%, respectively). The majority of zygotes in the IVF group were already at PN3 (50% at PN3 vs. 36% at PN2), while for Laser-IVF most of the zygotes were at PN2 (53% at PN2 vs. 38% at PN3). Eight hours after fertilization the highest percentage of IVF and Laser-IVF zygotes demonstrated PN3 (53 and 48%, respectively). Twelve hours after fertilization the obtained zygotes were mostly at PN5 in the IVF group (56%) and at PN4 in the Laser-IVF group (65%).
Maximum-likelihood analysis of variance showed that the second-order effect pronuclear state * time is dependent on the type of treatment (IVF vs. Laser-IVF, P = 0.0187).
DNA methylation pattern after IVF and Laser-IVF
After 5 h of fertilization, most of IVF zygotes reached PN3 (50%) and DNA methylation was observed in both pronuclei in 100% of cases. At 8 h post fertilization, 19% of IVF zygotes showed demethylation of paternal pronuclei and 53% of the zygotes reached PN3. At 12 h post fertilization, the IVF zygotes demonstrated 100% demethylation of paternal pronuclei and 56% of the zygotes were at PN5.
The Laser-IVF zygotes, fixed 5 h after fertilization, presented the demethylation of paternal pronuclei in 12% of the cases and the majority of zygotes were at PN2 (53%). At 8 h post fertilization, 7% of Laser-IVF zygotes showed demethylation of paternal pronuclei and most of the zygotes reached PN3 (48%). At 12 h post fertilization, Laser-IVF zygotes demonstrated 100% demethylation of paternal pronuclei and reached PN4 in 65% of the cases.
Quantitative estimation of the methylation status of paternal genome after IVF and Laser-IVF
At 5 h after IVF, the paternal genome in 62% of zygotes was found to be hypomethylated, similar to the paternal genome after Laser-IVF (55%). At 8 h after fertilization, the paternal genome was hypomethylated in 79% of zygotes of both IVF and Laser-VF groups. At 12 h after IVF and Laser-IVF, hypomethylation of the paternal genome was observed in 83 and 88% of zygotes, respectively.
Maximum-likelihood analysis of variance showed no statistical significance of all tested variables dependent on time and treatment (IVF vs. Laser-IVF, P = 0.9602).
RT-PCR analysis of some developmentally important gene transcripts
In this study we have demonstrated the quantitative analysis of the DNA methylation pattern of IVF- and Laser-IVF-derived mouse embryos. Using indirect immune fluorescence staining with anti-5-MeC antibodies, we compared the methylation status of the paternal genome in mouse zygotes 5, 8, and 12 h post fertilization. In addition, we calculated the PN stage distribution of zygotes in the IVF and Laser-IVF groups, collected at 5, 8, and 12 h post fertilization.
We observed slower development of Laser-IVF zygotes to PN2 5 h after fertilization. Most IVF zygotes reached PN3 during the same time. This delay of development was compensated for by 8 h post fertilization at which time we observed similar development of most zygotes to PN3 in the IVF and Laser-IVF groups. However, at 12 h post fertilization we observed again slower development of the Laser-IVF zygotes to PN4. At this time after fertilization most IVF zygotes reached PN5.
Our experiment’s setup (hyaluronidase treatment of both IVF and Laser-IVF oocytes; fertilization in both groups at the same time) allowed us to attribute the observed delay in development of Laser-IVF zygotes to the laser manipulation of the zona pellucida. In a previous study (Peters et al. 2006) we reported a significantly higher cleavage rate of inbred C3HeB/FeJ oocytes after Laser-IVF; however, the birth rate after Laser-IVF was significantly lower compared to IVF. Boersma et al. (2007) reported a similarly significantly higher cleavage rate of Laser-IVF inbred C57BL/6 J oocytes and significantly lower birth rate after Laser-IVF compared to IVF. When culturing two-cell-stage inbred C3HeB/FeJ embryos through to the blastocyst stage after IVF and Laser-IVF, 84% of the Laser-IVF but only 67% of the IVF embryos developed into blastocysts (unpublished data). These results indicate that further in vitro development of embryos is not delayed after laser zona dissection, but the birth rate after Laser-IVF is significantly lower compared to that after IVF.
Second, we followed the course of paternal pronuclei demethylation in IVF and Laser-IVF zygotes. The active demethylation of paternal pronuclei was observed at 8 h post fertilization in 19% of IVF zygotes (50% at PN3). At 5 and 8 h post fertilization, 12 and 7% of Laser-IVF zygotes, respectively, demonstrated demethylation of paternal pronuclei (53% at PN2 and 48% at PN3, respectively). At 12 h post fertilization, 100% of the IVF and Laser-IVF zygotes demonstrated demethylation of paternal pronuclei (56% at PN5 and 65% at PN4, respectively). Our findings were compared with previous results obtained in mice (Fulka and Fulka 2006; Santos et al. 2002). Santos et al. (2002) demonstrated methylation of both pronuclei in 94–100% of zygotes at PN1 and PN2 during the whole development after fertilization. Zygotes at PN3, PN4, and PN5 presented methylation of both pronuclei in 27, 15, and 8% of cases, respectively. Fulka and Fulka (2006) observed that 12 h after fertilization 91% of PN4 zygotes had undergone demethylation of paternal pronuclei. Our IVF and Laser-IVF zygotes showed methylation of both pronuclei at PN3 in 81 and 93% of cases, respectively. Zygotes at PN4 and PN5 demonstrated 100% demethylation of paternal pronuclei. The relatively high number of IVF and Laser-IVF zygotes presenting methylation of both pronuclei at PN3 at 8 h post fertilization could result from hyaluronidase treatment of zygotes before fertilization in contrast to cumulus-oocyte complexes routinely used for IVF. Also, the time between preparation of oocytes and fertilization was lengthened due to the laser microdissection of zona pellucida compared to the routinely performed IVF. A similar study was performed by Ribas et al. (2006) for zona pellucida removal in early mouse embryos. They observed that the time of zona removal influenced the levels of DNA methylation. When zona removal was delayed for 8 h, there was no difference in DNA methylation levels between zona-free and zona-intact two-cell embryos, indicating that the critical time is early on, between 1 and 8 h post fertilization. They suggested that the period immediately after the fertilization is the most sensitive. At this time the male DNA is even very close to the oolemma, which may make it more vulnerable to changes during zona pellucida removal and may possibly affect DNA methylation as well.
Another reason for the relatively high number of IVF and Laser-IVF zygotes presenting methylation of both pronuclei at PN3 could be the difference in mouse strain and IVF protocol used in our study compared with those used in previously published studies. In our study we used hyaluronidase denuded oocytes from inbred C3HeB/FeJ mice and HTF medium, while Santos et al. (2002) used cumulus-enclosed oocytes from hybrid B6CBAF1 mice and performed fertilization in modified KSOM medium. According to their protocol, the oocytes were incubated with spermatozoa for only 15–30 min, and after extensive washes with KSOM, oocytes were cultured for an additional 4, 6.5, or 9.5 h without spermatozoa. In our IVF procedure, the oocytes were incubated with spermatozoa for 5 h in HTF medium and then washed with KSOM and incubated for another 3 or 7 h. After incubation with spermatozoa for 5 h, we could potentially get more zygotes at different PN stages. This effect was, however, partly compensated by the use of cumulus-denuded oocytes which allowed a more simultaneous fertilization. Fulka and Fulka (2006) used inbred ICR oocytes and after 3–4 h of fertilization in HTF only viable-looking zygotes with a second polar body were chosen for further incubation without spermatozoa. In our study all oocytes were incubated until 12 h post fertilization and no selection in any step of preparation or immunocytochemistry was performed.
Finally, we showed that there were no significant differences in methylation status of the paternal genome after IVF and Laser-IVF. We measured the relative ratio of paternal to maternal methylation using Image J and IDL software and calculated the percentage of zygotes with hyper- (ratio >1.0) or hypomethylated (ratio <1.0) paternal genome at 5, 8, and 12 h post fertilization. At 5, 8, and 12 h post fertilization, the paternal genome was found to be similarly hypomethylated after IVF and Laser-IVF. Similar quantitative analysis of pronuclear methylation levels was performed by Kishigami et al. (2006). Using Image J the authors investigated the differences in DNA methylation of paternal genomes after intracytoplasmic sperm injection (ICSI) and round spermatid injection (ROSI). At 10 h after ROSI, the sperm- and spermatid-derived paternal genomes exhibited different methylation states. Methylation of the spermatid-derived genome was higher than that of the sperm-derived paternal genome. This difference in DNA methylation patterns after ICSI and ROSI was evident in 90% of the zygotes. These results indicated that the DNA hypermethylation of spermatid-derived paternal genomes was caused by remethylation occurring irrespective of demethylation after fertilization.
Our data show no significant differences in methylation status of the paternal genome after IVF and Laser-IVF. Analysis of Oct4, Dab2, and Dnmt3b gene expression also suggests that laser drilling of oocytes’ zona pellucida does not alter the expression profiles of some developmentally important genes in derived blastocysts. In conclusion, epigenetic and early embryonic development patterns are not altered by laser-assisted IVF techniques and another explanation must be found for the poor implantation rates observed in mice. The study of other genetic, epigenetic, and developmental characteristics of the embryo to verify the safety of new artificial reproduction techniques and to clarify, for example, the reason for the reduced birth rate after Laser-IVF should be pursued in further studies.
This work was supported by the European Commission under FP6, by EMMA inf (RII3-CT-2004-506455) to MHdA and by a grant from the National Genome Research Network, NGFN + (01GS0850), to MHdA. We thank Stefanie Dunst, Monika Beschorner, Andrea Bäßler, Alexander Huber, and Bernhard Rey for excellent technical assistance. We are grateful to John Favor for valuable comments on this manuscript.