SEARCH

SEARCH BY CITATION

Keywords:

  • Oocytes;
  • Mouse;
  • Embryonic stem cells;
  • Butyrolactone I

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

ESCs are most commonly derived from embryos originating from oocytes that reached metaphase II. We describe here a novel approach where ESCs with all pluripotency parameters were established from oocytes in which metaphase I was converted, from the cell cycle perspective, directly into metaphase II-like stage without the intervening anaphase to telophase I transition. The resulting embryos initiate development and reach the blastocyst stage from which the ESC lines are then established. Thus, our approach could represent an ethically acceptable method that can exploit oocytes that are typically discarded in in vitro fertilization clinics. Moreover, our results also indicate that the meiotic cell cycle can be converted into mitosis by modulating chromosomal contacts that are typical for meiosis with subsequent licensing of chromatin for DNA replication. STEM CELLS 2011;29:517–527


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Even with the advent of induced pluripotent stem cell technologies embryonic stem cells (ESCs) still represent the gold standard that is used for comparison of different characteristics of pluripotent cell lines. This has been confirmed very recently when NIH approved 13 human ESC (hESC) lines for federal funding [1]. The reason for this is that many essential questions concerning their derivation, pluripotency regulation, and differentiation are not yet satisfactorily answered. Moreover, for unknown reasons, some species, that is, cattle or pig, are refractory to ESC line derivation. Thus, it is highly desirable to develop new approaches of ESC production.

ESCs are most commonly derived from blastocyst inner cell mass (ICM) originating from fertilized metaphase II oocytes; less commonly from parthenogenetic embryos, epiblast, primordial germ cells, or cleaving embryos [2, 3]. In humans, the derivation of ESC lines from spare blastocysts originating from fertilized oocytes faces an ethical dilemma as it is often argued that this approach destroys the embryo and thus, possibly, a new life. For this reason, the search for new and ethically more sound approaches is desirable. The solution to this problem might be the use of parthenogenetic ESC lines. Parthenogenetic ESCs are most often derived from activated oocytes at the metaphase II stage [4, 5], these cells are, however, rather sparse especially in humans where the vast majority of mature oocytes (metaphase II) are used in assisted reproduction schemes. Thus, a method that would use oocytes that are typically discarded would represent a clear benefit. Although parthenogenetic ESCs are typically viewed as having a lower developmental potential when compared with ESCs containing both parental genomes, recent discoveries have shown a far greater developmental potential and plasticity than expected [6]. The only exception is their extremely low ability to penetrate the germ line in chimera experiments [5, 7].

In our experiments, we have studied the effect of butyrolactone I (BL1) on maturing oocytes as a mean of parthenogenetic activation [8]. BL1 was first described in detail by Kitagawa et al. who discovered this substance while searching for a specific cyclin-dependent kinase 1 (Cdk1; Cdc2) inhibitor. These authors showed that BLI indeed effectively inhibits the protein products of Cdc2 and Cdk2 genes but has a little effect on several other kinases including mitogen-activated protein kinase, protein kinase C, and several other kinases [9]. Furthermore, BL1 has been shown to function well in both in vitro assays and in cell lines thus opening the possibility to affect the cell cycle in cultured somatic cells [9, 10].

Here, we show that this drug can convert maturing oocytes into embryos from which ESC lines can be subsequently derived. Moreover, from developmental point of view, we demonstrate that meiosis can be converted into mitosis simply by abolishing the typical meiotic chromosome contacts along with inducing replication licensing of the converted oocyte chromatin.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Source, In Vitro Maturation, and BL1 Treatment of Maturing Mouse Oocytes

Immature mouse oocytes were obtained from large antral follicles of B6D2F1/Crl (BDF1) or 129S2/SvPasCrl (129S2) females stimulated with 7.5 IU pregnant mare serum gonadotropin (PMSG; Calbiochem, Merck, Prague, Czech Republic, www.merck-chemicals.com) intraperitoneally. After about 44 hours post PMSG injection, the oocytes were released from follicles into human tubal fluid (HTF)-Hepes medium (Cambrex, Charles City, IA, USA, www.cambrex.com) and only those oocytes surrounded with compact cumulus were selected for further experiments. The oocytes were then cultured in Minimum Essential Medium (MEM) supplemented with gentamicin (50 μg/ml), sodium pyruvate (0.22 mM), and bovine serum albumin (BSA; 4 mg/ml) at 37°C and atmosphere 5% CO2 in air for 90 minutes. Then cumulus cells were removed from oocytes by vigorous pipetting and only those oocytes undergoing germinal vesicle breakdown were cultured further for another 6 hours before transferred into medium supplemented with BL1 (100 μM; Biomol, Enzo Life Sciences GmbH, Lörrach, Germany, www.enzolifesciences. com). The oocytes were cultured in this medium for another 14–16 hours and then evaluated under the stereomicroscope. The oocytes with visible nuclei (nucleoli) and polar bodies were washed several times in HTF medium and then cultured in potassium simplex optimized medium (KSOM) (Millipore, Billerica, MA, USA, www.millipore.com).

Alternatively, immature mouse oocytes of both BDF1 and 129S2 mice were collected and cultured in MEM as described above and allowed to reach the metaphase II stage (approximately 15–16 hours later) where they were parthenogenetically activated.

Source of Ovulated Mouse Oocytes

For generation of parthenogenetic ESC lines derived from metaphase II oocytes, BDF1 or 129S2 females were induced to superovulate by treatment with 7.5 IU PMSG, followed 48 hours later by 7.5 I.U. human chorionic gonadotropin (hCG; Puberogen, Tokyo, Japan or Intervet, Prague, Czech Republic, www.intervet.com). Metaphase II oocytes were then collected from ampulae of sacrificed females.

Parthenogenetic Activation of Oocytes

The oocytes were artificially activated as described in [11]. Briefly, oocytes of both BDF1 and 129S2 mice were treated with 5 mM SrCl2 in EDTA supplemented KSOM supplemented with 5 μg/ml cytochalasin B to suppress polar body extrusion for 6 hours. After this, activated oocytes were washed several times in KSOM and further cultured in the same medium until the blastocyst stage.

ESC Line Derivation

The blastocysts were allowed to hatch in the KSOM or alternatively zonae pellucidae were removed by Acid Tyrode's solution. Zona-free blastocysts were placed onto gelatin-coated dishes (0.1% gelatin in water) and allowed to attach to the bottom of the well and cultured in ESC culture media: Dulbecco's modified Eagle's medium (DMEM; SLM-220-B, Millipore), 15% KnockOut Serum Replacement (Invitrogen), 1× EmbryoMax penicillin/streptomycin (Millipore), 1× EmbryoMax glutamine (Millipore), 1× EmbryoMax nucleosides (Millipore), 1 mM sodium pyruvate (Millipore), 1× EmbryoMax nonessential amino acids (Millipore), 0,1 mM 2-mercaptoethanol (M7522, Sigma), supplemented with mitogen-activated protein kinase kinase 1 (MEK1) inhibitor at a final concentration of 50 μM (PD98059, Cell Signaling, Danvers, MA, USA, www.cellsignal.com) and 1,000 U/ml ESGRO (leukemia inhibitory factor, Millipore) [12]. For the derivation of 129S2 ESC lines, MEK1 inhibitor was omitted from the ESC media as we have found it to negatively affect the ICM expansion in this strain. After several days, the attached blastocysts were observed and expanded ICMs were removed and disaggregated by incubating them in Accumax dissociation solution (Millipore) and aspirating the cells by a glass pipette. The cell suspension was transferred into wells coated with Mitomycin C inactivated feeder cells (Millipore). Primary ESC colonies were either further cultured in ESC culture media (with or without MEK1 inhibitor) or were taken for analysis (Alkaline phosphatase staining, antibody labeling, etc.).

Control ESC Line Generation

For generation of fertilization-derived BDF1 ESC lines, in vitro fertilization (IVF) was conducted using oocytes collected from superovulated C57BL/6NCrSlc females and epididymal spermatozoa from DBA/2CrSlc males according to the method previously reported [13]. In brief, spermatozoa from epididymis were resuspended in 450 μl HTF medium that contained 0.5% BSA, covered with silicone oil (Sigma), and preincubated at 37°C under 5% CO2 in air for 1–2 hours. Cumulus-oocyte complexes were collected from the oviducts and placed into 80 μl drops of HTF under silicone oil. Insemination was carried out by adding a small drop of sperm suspension to these HTF drops. Four to six hours after insemination, the eggs were removed from the fertilization drops, washed in PB1 medium, and cultured in KSOM for 96 hours.

The ESC line derivation was carried out by the same procedure as described above.

In Vitro Differentiation

Embryoid bodies (EBs) were prepared as described in “Manipulating the Mouse Embryo—A Laboratory Manual” [12] on Ultra-low attachment dishes (Corning). Briefly, approximately 1 × 106 cells were disaggregated into single-cell suspension and resuspended in medium containing 10% fetal calf serum (FCS). Medium composition: DMEM (SLM-220-B, Millipore), 10% FCS (ES qualified, Millipore), 1× penicillin/streptomycin (Millipore), 1×glutamine (Millipore), and 0.1 mM 2-mercaptoethanol (M7522, Sigma). After 7 days, EBs were transferred onto tissue culture-grade dished and processed for immunofluorescence or collected for RNA isolation.

Chimera Production and Teratoma Formation Assay

Chimeras were produced essentially as described in “Manipulating the Mouse Embryo—A Laboratory Manual” [12]. BL1 ESCs were injected into ICR mouse blastocysts and transferred to pseudopregnant ICR recipient females (Jcl:ICR). Teratomas were produced by injecting BL1 ESCs under the kidney capsule and into hind paws of severe combined immunodeficiency (SCID) mice (C.B-17/lcr-scid/scidJcl). Isolated teratomas were fixed in formaldehyde and embedded in paraffin. Tissue sections were stained by H&E.

Unless stated otherwise all chemicals were purchased from Sigma (Prague, Czech Republic, www.sigmaaldrich. com). Plastics were purchased from Corning (Amsterdam, The Netherlands, www.corning.com).

Additional Materials and Methods can be found in the Supporting Information Materials and Methods.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

The Effect of BL1 on Maturing Mouse Oocytes

Metaphase I BDF1 or 129S2 mouse oocytes treated with BL1 expel the first polar body but the typical second metaphase plate is never formed [8]. Instead, chromosomes decondense and form pseudopronuclei (82% in BDF1, 93% in 129S2 strain; Fig. 1A and Supporting Information Table S1), and these can be detected as early as 1.5 hours of BL1 treatment and their size increases with the BL1 incubation time. Their presence has been confirmed by labeling the oocytes against the nuclear lamina (Lamin A/C) and nuclear pore complex (Supporting Information Fig. S1). When BL1 is washed out, the oocytes cleave and form two-cell stage-like embryos, that is, after approximately 24 hours post the beginning of BL1 treatment (Fig. 1D). We speculated that DNA replication must occur in BL1 pseudopronuclei. To analyze in more detail what happens in oocytes after BL1 treatment, we incubated them with bromodeoxyuridine throughout the BL1 treatment. We detected the replication in pseudopronuclei approximately 16–17 hours after the beginning of treatment (Fig. 1B). Interestingly, the replication occurs irrespectively of the presence of BL1. However, we estimated that the critical incubation time with BL1 is at least 8 hours and thereafter the formation of pseudopronuclei was irreversible. The analysis of the first cleavage metaphase showed the presence of perfect spindles with well-arranged mitotic chromosomes (Fig. 1C). This and the absence of condensed chromosomes during the treatment, indicates that in the presence of BL1 metaphase I oocytes exit from this stage but they never reach true metaphase II instead they are converted directly into embryos without any obvious or typical activation stimulus. In contrast, the metaphase II oocytes do not respond to BL1 treatment and remain arrested at this stage (not shown). This clearly indicates the different response of metaphase I and metaphase II oocytes to BL1.

thumbnail image

Figure 1. Production of butyrolactone I (BL1) embryos and ESCs. (A): Metaphase I oocytes incubated with BL1 give off the first polar body and in their cytoplasm a prominent pseudopronucleus containing the nucleolus can be detected. (B): DNA replication in pseudopronuclei occurs even in the presence of BL1 (bromodeoxyuridine incorporation). (C, D): After the completion of DNA replication, the first mitotic metaphase can be detected (C) and is followed by cleavage into two-cell stage (D). (E–G): Cleavage stages of BL1 embryos–four-cell stage (E), compacting eight-cell stage (F) and blastocyst stage with prominent inner cell mass (ICM; [G]). (H): ICMs in these blastocysts show positive labeling for Oct4, red; 4',6 diamidino-2-phenylindole (DAPI), blue. (I–L): When seeded on mouse embryonic fibroblasts, typical primary ESC colonies were effectively produced (I). Primary ESC colony positive for alkaline phosphatase activity (J), Oct4 (K); DAPI (L). Scale bar = 50 μm.

Download figure to PowerPoint

Derivation and Analysis of ESCs from BL1-Treated Oocytes

The BL1 embryos cleaved well in culture and after approximately 4 days 20%–30% of them reached blastocyst stage with well-developed ICM that was positive for Oct4 and Nanog (Fig. 1E–1H; Supporting Information Fig. S2). As expected, the trophoblast cells were Cdx2 positive (Supporting Information Fig. S2). To determine the extent to which our culture conditions might negatively influence the developmental potential of BL1 embryos, we have parthenogenetically activated both in vitro matured (IVM) and ovulated metaphase II oocytes in both BDF1 and 129S2 mice (Supporting Information Tables S2 and S3). The results indicate about 10% difference in the developmental rate to the blastocyst stage between BL1 embryos and parthenogenetic embryos obtained from IVM oocytes (27% and 33% for BL1 treatment vs. 40% and 42% for IVM oocytes in BDF1 and 129S2 mice, respectively). However, when the blastocysts obtained by different methods were transferred to ESC media and cultured, many of the parthenogenetic embryos failed to expand their ICM. This phenomenon was especially pronounced in parthenogenetic embryos originating from IVM oocytes of both BDF1 and 129S2 mice (Supporting Information Table S4). Next, we asked whether ESCs can be derived from BL1 blastocysts.

When BL1 blastocysts were disaggregated and seeded on mitomycin-treated feeders typical primary ESC colonies could be obtained (Fig. 1I–1L). These colonies were continuously propagated without losing their ESC characteristics and the overall success rate of ESC line derivation was 70%–80% (Supporting Information Fig. S3). Next, we have analyzed the karyotype of these ESC lines to investigate whether the BL1 treatment has some negative effects on proper chromosome segregation. To separate the influence of BL1 treatment and a possible negative effect of culture conditions, all lines were analyzed within the first five passages. The majority of lines (14/16 BDF1 lines tested, 87.5% and 5/5 129S2 lines tested, 100%; Fig. 2A) showed a normal karyotype with 40 chromosomes, this also indicates a direct transition of oocytes from meiosis I straight to mitosis and shows that the BL1 treatment does not induce chromosome segregation abnormalities.

thumbnail image

Figure 2. A detailed analysis of karyotype and karyotypic abnormalities in butyrolactone I (BL1) ESC lines. (A): A normal karyotype with 40 chromosomes was detected in most lines. (B): However, at advanced passage numbers, karyotype abnormalities were sometimes observed. In this case, the most frequent abnormality was the presence of chromosome fusion. (C, E, G): One line was scored as abnormal from a very early passage again exhibiting a chromosome fusion. To characterize this abnormality, G-banding was employed and Chromosome 19 was identified as a possible candidate; metacentric chromosome (arrow; [C]). (E, G): Chromosome 19 involvement was verified by fluorescence in situ hybridization (FISH)–4',6 diamidino-2-phenylindole (DAPI) (blue); metacentric chromosome (arrow; [E]), Chromosome 19 (red; [G]). (B): As noted before, culture of BL1 ESC lines also induced some chromosomal fusions. These typically appeared around passage 20. (D, F, H): Here, the most affected chromosome was the Chromosome 8 as identified by G-banding (D) and confirmed by FISH; DAPI (blue), chromosome fusion (arrow; [F]). This was also often accompanied by a loss of one X Chromosome; Chromosome 8 (green), X Chromosome (red; [H]). Thus, a typical picture of a culture induced abnormality was trisomy of Chromosome 8 (two copies in the form of a metacentric chromosome, one copy separate) and a loss of one X Chromosome. Scale bar = 50 μm.

Download figure to PowerPoint

One of the two BDF1 lines that were scored as abnormal in the initial karyotype screening had lost one chromosome (39 chromosomes), the other line showed the presence of a fused chromosome (corresponding to a total number of 41 chromosomes; Fig. 2C). As expected, karyotype aberrations eventually appeared in some ESC lines at advanced passages. Because all our lines should theoretically be female, we expected that the loss of one X Chromosome will be the most frequent chromosomal abnormality as previously described for female pluripotent cell lines [14]. To our surprise, this was not the case and the most often observed abnormality was the presence of a fused (metacentric) chromosome (Fig. 2B, 2D). Next, we wanted to determine if the chromosome fusion involves always the same chromosomes or different chromosomes and whether it is identical to the metacentric chromosome observed in the initial screening of ESC lines. Thus, we carried out G-banding experiments followed by fluorescence in situ hybridization to identify the fused chromosomes. This analysis showed that Chromosome 8 fusion (sometimes accompanied by loss of X Chromosome) leading to trisomy was prevailing (Fig. 2B, 2D, 2F, 2H). In contrast, in the line that was scored as abnormal in initial screening the fused chromosome was composed of two Chromosome 19 (Fig. 2C, 2E, 2G). Our finding is in agreement with a previous study that described the Chromosome 8 trisomy as the most frequent abnormality in ESC lines. This study, however, focused only on ESC lines derived from the 129 genetic background [15]. Thus, our results demonstrate that abnormalities involving Chromosome 8 are a general phenomenon in ESC lines irrespectively of their genetic background. Moreover, it is clear that abnormalities involving Chromosome 8 seem to be typical culture-induced karyotype aberrations.

Interestingly, when the same lines were cultured again from earlier passages and reanalyzed, they showed a different spectrum of karyotype abnormalities or no abnormalities at all. This indicates that chromosomal aberrations arise rather randomly but become rapidly fixed in the given cell population and also that individual lines are not predisposed to a particular chromosomal abnormality.

Because of the above-mentioned facts, we have performed the same karyotype analysis in BL1 ESC lines derived from the 129S2 strain. Thus, the five ESC lines that initially had a normal karyotype were cultured further and reanalyzed. Four lines showed a normal karyotype after 17 passages and one line lost one X Chromosome (39 chromosomes). No Chromosome 8 abnormality was observed (not shown).

In addition, the absence of paternal genome contribution was verified by bisulfite sequencing of several imprinted genes (Igf2r, Snrpn, H19, Dlk1/Gtl2). As expected, the methylation pattern of these genes was essentially similar to metaphase I oocytes from which these lines were derived (Fig. 3A). In contrary to previous reports showing epigenetic instability in ESC lines [16–18], repeated analysis of one line at passage 43 showed that imprinting is faithfully maintained even at advanced passages.

thumbnail image

Figure 3. Characterization of ESC lines. (A): The methylation pattern is similar between the derived butyrolactone I ESC lines (line BL8, BL7) and metaphase I oocytes. Moreover, the methylation pattern does not change even with advanced passage number (BL7 passage 9 vs. BL7 passage 43). (B): Hierarchical clustering analysis indicates that BL1 ESC lines (line BL1, BL11, BL9, and BL7–BDF1 genetic background) are more similar to control ESC line obtained from fertilized oocytes (“Fert ES,” BDF1 genotype) than to parthenogenetic ESC lines (“IVM partheno ES”–ESC line obtained by parthenogenetic activation of metaphase II IVM BDF1 oocytes, “MII partheno ES”–ESC line obtained after parthenogenetic activation of ovulated BDF1 oocytes). Note that in vitro maturation of oocytes prior to their activation does not seem to grossly affect the overall gene expression. Also note the outlying BL7 ESC line that performed poorly in the chimera experiment and few passages later exhibited altered colony morphology (albeit normal karyotype and the presence of pluripotency markers). The clustering was performed by Ward's method and the distance corresponds to Euclidean distance. Abbreviations: BL1, butyrolactone I; EB, embryoid body; IVM, in vitro matured; MI, metaphase I.

Download figure to PowerPoint

We have also analyzed the gene expression in four BDF1 BL1 ESC lines by low density arrays and compared it with the gene expression of control BDF1 female ESC line and parthenogenetic ESC lines generated from either ovulated or IVM oocytes. This showed that our lines are highly similar to control BDF1 ESCs obtained from fertilized oocytes and thus containing both female and male genomes (Fig. 3B). The analysis of significantly upregulated or downregulated genes showed that many of the downregulated genes can be classified as markers of the trophectodermal or hematopoietic lineage (Supporting Information Table S5). Interestingly, alteration of the overall gene expression seems to be line specific as few upregulated and/or downregulated genes were shared among two or more BL1 ESC lines.

BL1 ESC Lines Give Rise to Cells of All Three Germ Layers

It has been suggested that the differentiation potential of uniparental ESC lines can differ from that of biparental ESC lines containing both maternal and paternal genome [4, 19]. Early studies showed a specific elimination of parthenogenetic ESC from mesodermal and endodermal tissues when the cells were subjected to in vivo differentiation (chimera production assays). One plausible explanation for this observation might be an inadequate expression of imprinted genes. On the other hand, recent experiments demonstrate that the expression level of several imprinted genes in parthenogenetic ESCs is practically identical to the levels observed in conventional ESC lines [18]. Moreover, it has been shown that parthenogenetic ESCs are able to form well-developed fetuses of far more advanced stages than those described for gynogenetic embryos [6, 20, 21]. Thus, the latter study [6] clearly demonstrates that ESC lines lacking the paternal genome have better developmental potential than expected based on previous studies.

Because of the lack of paternal genome, the BL1 ESC lines could thus potentially show a limited differentiation. The induction of in vitro differentiation in six independent BL1 ESC lines of both BDF1 and 129S2 genotype (including the four lines that were also analyzed by low density arrays) resulted in formation of embryoid bodies that gave rise to cells of all three germ layers, including mesoderm and endoderm (Supporting Information Fig. S4). Next, we have selected three BL1 ESC lines with characterized in vitro differentiation potential and gene expression pattern for in vivo analysis. All these lines were of BDF1 origin. When these BL1 ESC lines were injected into SCID mice, teratomas were formed and when evaluated histologically, they contained a wide range of somatic tissues, again originating from all three embryonic layers (Fig. 4B–4F). We have also injected cells of the same BL1 ESC lines into ICR blastocysts that were transferred to recipient females, and we obtained several high-percentage chimeric mice from two of these lines but no germ line transmission was observed (Fig. 4A). We cannot, however, exclude the possibility that the production of more chimeric mice and their extensive breeding would eventually lead to the birth of such offspring. However, taking into account previously published data on parthenogenetic ESC and their germline transmission competence, the birth of such offspring is rather improbable [5, 7].

thumbnail image

Figure 4. Characterization of pluripotency parameters of butyrolactone I (BL1) ESC lines. (A): High-percentage chimeras were obtained after injection of BL1 ESCs into ICR blastocysts. (B–F): Formation of teratomas by BL1 ESCs in severe combined immunodeficiency mice. These teratomas contain tissues originating from all three germ layers including trachea (B), bone (C), muscle (D), epidermis (E), and brain (F). Scale bar = 50 μm.

Download figure to PowerPoint

Furthermore, we evaluated the degree of contribution of the BL1 ESCs in tissues of produced chimeric mice. To quantify the BL1 ESC contribution, we have taken the advantage of different coat color of recipient blastocyst mouse stock (ICR, albino) and injected BL1 ESCs that were derived from BDF1 mice (black). It has been described that a single-point mutation in tyrosinase when present in a homozygous state is able to cause the albino phenotype and the same mutation is found in most albino mouse strains and stocks [22, 23]. Thus, first we have sequenced the first exon of tyrosinase in both the BDF1 and ICR mice where the mutation should be located (Fig. 5A). This analysis confirmed the presence of the described mutation in the ICR mice and its absence in the BDF1 mice. Because ICR is an outbred mouse stock, we have analyzed the sequences of more than 20 individual mice. Next, we have devised a real-time polymerase chain reaction assay that would allow us to quantify the occurrence of the wild-type (BDF1) versus mutated (ICR) tyrosinase variants. The results show that the specific contribution of BL1 ESCs to internal organs does not generally correspond to the coat color of chimeric mice (Fig. 5B, 5C). Furthermore, BL1 ESCs were clearly absent from some tissues (e.g., spleen) but there does not seem to be a clear correlation between the absence of BL1 ESCs in specific tissues and their germ layer origin as proposed in earlier studies [19, 24].

thumbnail image

Figure 5. Quantification of BL1 ESCs contribution in chimeric mice. (A): The mutation in tyrosinase exon 1 was verified by sequencing; whereas the ICR mice (albino) exhibit the presence of AGA (TCT) sequence, in BDF1 mice ACA (TGT) is found. (B): Several high-percentage chimeric mice were obtained after injection of BL1 ESCs to ICR recipient blastocysts. These chimeric mice were used to investigate the contribution of BL1 ESCs in individual tissues. (C): Quantification of BL1 ESCs contribution to individual tissues of chimeric mice. The values represent the percentage of the wild-type tyrosinase allele (BDF1 genotype) ± SD in tissues of chimeric mice. Note that the coat color does not correlate with contribution to internal organs; values representing more than 20% of the wild-type tyrosinase variant are highlighted in yellow. Abbreviations: BL1, butyrolactone I; ICR, ICR mouse recipient stock.

Download figure to PowerPoint

The Effect of BL1 Treatment on Chromosome Segregation During Meiosis I

The modulation of the exit from metaphase I directly into mitosis is, per se, also very interesting from the basic cell biology point of view. The derivation of ESC lines enabled a more detailed description of the effect of BL1 treatment on meiosis I and chromosome segregation. Simple visualization of chromosomes in BL1-treated oocytes by 4',6 diamidino-2-phenylindole (DAPI) labeling indicated a clear alteration in the chromosome morphology prior to actual chromosome segregation. Therefore, we have decided to investigate the presence and distribution of key structural component of meiotic chromosomes (the meiotic cohesin Rec8) and the position of sister kinetochores. A deeper analysis of metaphase I after the BL1 treatment showed an evident chromosome decondensation and possible premature sister chromatid separation which is indicated by separation of sister kinetochores and loss of Rec8 along the chromosome arms (Fig. 6A–6L). This chromosome morphology is typical for chromosomes at the metaphase II stage [25, 26]. Surprisingly, the treatment does not impair the correct chromosome segregation during anaphase I as shown by prevailing normal karyotypes of the ESC lines and global single nucleotide polymorphism genotyping of ESC lines (Supporting Information Fig. S5). Because of the observed chromosome decondensation, we have also analyzed the level of phosphorylated H3/S10 (not shown). However, no difference was observed between control metaphase I group and BL1-treated oocytes. Thus, the chromosome decondensation can be probably attributed to the loss of some structural proteins, e.g., Rec8. When we used a long exposure for capturing the Rec8 images, we have observed a residual signal at the centromeric part between sister chromatids in BL1-treated oocytes, and this residual Rec8 is probably sufficient to ensure proper chromosome segregation (Fig. 6A, inset).

thumbnail image

Figure 6. The effect of BL1 treatment on the metaphase I chromosome morphology. (A–L): BL1 treatment causes chromosome decondensation accompanied by premature Rec8 loss from sister chromatids (A). However, when overexposed, Rec8 can still be detected at centromeric regions in BL1-treated oocytes ([A], inset); Rec8 is clearly present in control metaphase I (D), but becomes lost during anaphase I/telophase I transition (G) and is essentially undetectable in metaphase II (J). Loss of Rec8 is accompanied by sister kinetochore separation (B, C); kinetochores (green/yellow), pericentric heterochromatin (trimethylated H4/Lys 20; red), 4',6 diamidino-2-phenylindole (blue). Again, sister kinetochores are tightly joined during normal metaphase I (E, F), but become separated during anaphase I/telophase I transition (H, I); metaphase II (K, L). Median filter was applied to images (C, F, I, L). Scale bar = 50 μm.

Download figure to PowerPoint

Replication Licensing in BL1-Treated Oocytes

As noted above, BL1-treated oocytes never reach metaphase II but are DNA replication competent. This represent a certain paradox because a key regulator of DNA replication, Cdc6 protein, is not present in maturing oocytes until metaphase II although its mRNA is stored in the oocyte even at the germinal vesicle stage [27–29]. Generally, the role of Cdc6 in replication is such that it facilitates the loading of the Mcm protein complex (composed of Mcm2-7) onto replication origins, this in turn is necessary for possible replication initiation. Thus, we have analyzed the presence and localization of Cdc6 and several Mcm proteins throughout the BL1 treatment to clarify when these oocytes become replication competent (Fig. 7A–7L; Supporting Information Fig. S6). Different time points after the initiation of BL1 treatment were investigated, as controls, the corresponding phases of normal meiosis and parallel time points were used (metaphase I, anaphase I/telophase I, and metaphase II oocytes). In BL1-treated oocytes, Mcm7 and Cdc6 localize to chromosomes very soon after the exit from metaphase I (35 minutes and 1.5 hours after the beginning of BL1 treatment, respectively). In contrast, in control oocytes Cdc6 cannot be detected on chromosomes before meiosis II (Fig. 7A–7C). In other words, BL1 treatment causes a rapid translation of Cdc6 mRNA that is already present in oocytes and facilitates the association of this protein with chromosomes. This in turn enables the loading of Mcm proteins onto chromosomes. Thus, the BL1 treatment causes replication licensing of chromatin that normally does not occur during meiosis I.

thumbnail image

Figure 7. The influence of BL1 treatment on replication licensing. (A–L): During normal meiosis, Cdc6 can be detected on chromosomes neither at metaphase I (A) nor anaphase I/telophase I (B). However, it becomes associated with chromosomes at metaphase II (C). BL1 treatment, however, causes the association of Cdc6 with oocyte chromosomes and replication licensing. After 35 minutes of BL1 treatment, the chromosomes still do not show the presence of this protein; DAPI (blue), Cdc6 (red; [D, E, F]). However, at 1.5 hours, Cdc6 becomes associated with chromosomes (G, H, I). At 3 hours, when well-developed pseudopronuclei are formed, Cdc6 is again mostly excluded from pseudopronuclei (J, K, L). Scale bar = 50 μm. Abbreviation: BL1, butyrolactone I.

Download figure to PowerPoint

BL1 Treatment of Oocytes of Different Species

Taken together, our results demonstrate a new and very efficient production of ESC lines that may be ethically acceptable for human medicine. As we have no access to human maturing oocytes, we verified if this approach would be applicable to other mammals. The BL1 treatment of bovine metaphase I oocytes showed that even in this species the oocytes respond similarly to BL1 treatment and the resulting embryos frequently reach the blastocyst stage with prominent ICM (Supporting Information Fig. S7). Unfortunately, in this species, OCT4 and CDX2 labeling is not informative as it is not confined specifically to the ICM or trophectoderm, respectively [30]. However, when these blastocysts are placed onto feeders a clear OCT4 positive population of cells diverges from the rest of embryonic cells (Supporting Information Fig. S7H). This clearly shows that our approach is well applicable to other species and produces viable blastocysts, and thus can potentially work even in humans.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We have studied the effect of BL1 on maturing mouse and bovine oocytes as a mean of parthenogenetic activation and embryo production with subsequent ESC line derivation. In a broad sense, parthenogenesis is defined as “the production of an embryo from a female gamete without any genetic contribution from a male gamete, and with or without eventual development into an adult” [31]. In the mouse, parthenogenetic ESCs are typically obtained from matured metaphase II oocytes, in other species, these can be, however, largely unavailable (especially in humans where the vast majority of metaphase II oocytes is used in assisted reproduction schemes). In contrast, our method focuses on maturing oocytes that are, to our knowledge, commonly discarded in IVF clinics. The potential use of such material was also considered by others. Recently, Sung et al. have described a very efficient method of parthenogenetic activation of immature cryopreserved oocytes and proposed its use in human medicine [32].

Furthermore, our (or alternative) approach can also be valuable in young ovarian cancer patients where, after the surgical removal of ovaries, the oocytes would be collected, then matured in vitro and BL1 treated. The resulting ESC lines could be then potentially used for the production of their own gametes [33, 34]. The high efficiency of pseudopronuclei formation after BL1 treatment indicates that this inhibitor induces the exit from metaphase I even in those oocytes that are not yet fully competent to complete maturation. Eppig et al. [35] demonstrated that such oocytes can reach only metaphase I stage where they can be fertilized with the formation of regular pronuclei, as expected, the resulting embryos are, however, triploid. The formation of pseudopronuclei in metaphase I oocytes was first described by Clarke and Masui who treated them with cycloheximide [36]. However, when this inhibitor was washed out DNA replication did not occur, the chromosomes recondensed and formed metaphase II-like stage. In contrast, BL1 treatment stimulates the replication in pseudopronuclei that is followed by regular embryonic development. Thus, our approach of blastocyst production and subsequent ESC line derivation clearly differs from previously described methods [4, 5]. From the cell cycle point of view, in those experiments, the oocytes had to reach the metaphase II stage (irrespectively of first polar body extrusion block) and only at this stage the oocytes could be parthenogenetically activated. In contrast, our approach allows the use of potentially less competent oocytes that never reach the metaphase II stage.

Furthermore, we can expect that some characteristics of the BL1 ESC lines will be shared between these ESCs and biparental/parthenogenetic ESC lines. Thus, for example we have observed that prolonged culture of the ESC lines often leads to chromosomal abnormalities involving the chromosome 8. This is in good correlation with a previously published study [15]. On the other hand, the chromosome fusion might be rather specific to BL1 ESC lines. We can only speculate about the relation of chromosome fusion occurrence and the genetic background of BL1 ESC lines as most used ESC lines are from the 129 genetic background and a detailed analysis of karyotype abnormalities of ESC of different genetic backgrounds is largely unavailable. However, because during the course of this study the 129S2 BL1 ESC lines did not exhibit a similar chromosomal aberration, observed chromosome fusion might be rather specific for BDF1 ESC lines.

Moreover, contrasting results were published on developmental potential of ESCs lacking the paternal genome [4, 6, 19]. Thus, we have devised a strategy that allowed us to quantify the extent of BL1 ESCs contribution in chimeric mice. Our results indicate that a clear correlation between the BL1 ESC contribution to chimeric tissues and a particular embryonic layer cannot be easily established. Moreover, we clearly demonstrate that the coat color is a poor indicator of the overall level of BL1 ESC contribution. The very same can be expected for parthenogenetic ESCs.

Finally, our approach is important and interesting also from the basic biology point of view as the described treatment directly converted meiosis I to mitosis by modulating the chromosome contacts that are characteristic for final stages of oocyte maturation and inducing replication licensing of chromatin.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

In our experiments, we have shown that BL1 causes a precocious exit from meiosis and entry into mitotic cell cycle in maturing oocytes. Moreover, this drug influences the morphology of chromosomes, as it converts meiotic metaphase I chromosomes into chromosomes with mitotic-like morphology, that are subsequently licensed for DNA replication. Finally, we show that unique ESCs can be derived from BL1-treated oocytes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We thank P. Simecek (Institute of Animal Science) for a very helpful consultation on array results analysis and the Molecular Biology department staff for great help with sequencing. We also thank Dr. Jibak Lee for kindly providing the anti-Rec8 antibody. This work was supported by the grant MZE0002701404 from the Ministry of Agriculture, Czech Republic (H.F., T.M., and J.F.) and the grant (no.20062012) from the Japanese Ministry of Education, Culture, Sports, Science and Technology (M.H., K.I., N.O., N.W., S.M., and A.O.).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional supporting information available online.

FilenameFormatSizeDescription
STEM_585_sm_suppinfo1.doc73KSupporting Information 1
STEM_585_sm_suppinfo2.doc2651KSupporting Information 2

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.