SEARCH

SEARCH BY CITATION

Keywords:

  • Parthenogenesis;
  • Nuclear transfer;
  • Reprogramming;
  • Embryonic stem;
  • Regenerative medicine

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Parthenogenesis is the process by which an oocyte develops into an embryo without being fertilized by a spermatozoon. Although such embryos lack the potential to develop to full term, they can be used to establish parthenogenetic embryonic stem (pES) cells for autologous cell therapy in females without needing to destroy normally competent embryos. Unfortunately, the capacity for further differentiation of these pES cells in vivo is very poor. In this study, we succeeded in improving the potential of pES cells using a nuclear transfer (NT) technique. The original pES cell nuclei were transferred into enucleated oocytes, and the resulting NT embryos were used to establish new NT-pES cell lines. We established 84 such lines successfully (78% from blastocysts, 12% from oocytes). All examined cell lines were positive for several ES cell markers and had a normal extent of karyotypes, except for one original pES cell line and its NT-pES cell derivatives, in which all nuclei were triploid. The DNA methylation status of the differentially methylated domain H19 and differentially methylated region IG did not change after NT. However, the in vivo and in vitro differentiation potentials of NT-pES cells were significantly (two to five times) better than the original pES cells, judged by the production of chimeric mice and by in vitro differentiation into neuronal and mesodermal cell lines. Thus, NT could be used to improve the potential of pES cells and may enhance that of otherwise poor-quality ES cells. It also offers a new tool for studying epigenetics.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

There has been much research into the applications of human embryonic stem (ES) cells in regenerative medicine, where ES cells have been envisioned as potential resources for cell or tissue replacement therapy. However, as with any allogeneic material, ES cells derived from fertilized blastocysts—and the progeny of such cells—inevitably face the risk of immunorejection on transplantation. Therefore, ES cells derived from embryos cloned from the host's own cells by somatic cell nuclear transplantation (NT-ES cells) represent a potential solution to the problem of rejection, as any replacement cells would be genetically identical to the recipient's cells [1, [2]3]. It is known that somatically derived NT-ES cells and fertilized embryos derived from ES cells are transcriptionally and functionally indistinguishable [4, [5]6]. However, there is an ongoing ethical controversy surrounding human NT, arising from the need to destroy a reconstructed blastocyst with the potential to form an individual. To avoid this problem, at least in a mouse model, NT-ES cells have now been established from Cdx2-deficient blastocysts [7]. Such embryos are nonviable because they cannot implant, so it is possible in theory to obtain valuable ES cells without destroying live individuals. Alternatively, ES cells can be established by nondestructive biopsy of single blastomeres of embryos for progeny in the future [8, 9]. Such techniques can help reduce ethical concerns, but they could not be applied to humans therapeutically as any cell line would differ genetically from the patient.

One other possibility is to establish ES cell lines from parthenogenetically activated embryos (parthenogenetic embryonic stem [pES] cells), which also lack full developmental potential as the paternal genome is missing [10, 11]. Such pES cells may be more acceptable ethically for the treatment of human patients. For primates, it has already been demonstrated that pES cells can be differentiated into all three embryonic germ cell lines in vitro [12]. However, there are several negative reports of the application of pES cells. When chimeric mouse offspring were generated from parthenogenetic and fertilized embryos, they showed postnatal growth retardation; moreover, the cell lineages of parthenogenetic cells were restricted, particularly in forming mesoderm and endoderm [13, [14], [15]16]. Although the incorporation of pES cells did not produce postnatal growth retardation in chimeric mice, the contribution of the cells was restricted in some organs, and Igf2, which is a paternally expressed gene, was repressed [17]. Moreover, loss of genomic imprinting is observed commonly in human tumors [18], and chimeric mice derived from imprinting-free mouse embryonic stem cells develop multiple tumors within a year [19]. Thus, there may be limitations to the use of pES cells in human regenerative medicine.

On the other hand, it has been demonstrated that somatic cell nuclear transfer (SCNT) techniques can reprogram donor cell nuclei from a differentiated state to an undifferentiated one [20, 21]. Although the mechanism is not clear, epigenetic modifications to imprinted genes are often observed after NT [22, [23], [24], [25], [26]27]. Here, we hypothesized that successive rounds of NT might also reprogram the epigenetic status of parthenogenetically derived ES cell nuclei. Following the reestablishment of NT-pES cells, these might improve the potential for differentiation over that of the original pES cells. To test this hypothesis, we established several second- and third-generation NT-pES cell lines from original pES cell lines by NT. These were examined for their potential to differentiate in vivo and in vitro and were compared with the original pES cell lines. We also examined the DNA methylation status of the differentially methylated domain (DMD) H19 and the differentially methylated region (DMR) IG, regions upstream of Dlk1, which defines a cluster of imprinted genes.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Reconstruction of Hybrid Oocytes and Induction of Parthenogenetic Embryos Expressing Green Fluorescent Protein

In this study, we established original pES cell lines by combining two genetically different mouse oocytes instead of the usual method of producing parthenogenetic embryos from single oocytes (Fig. 1). Genetically, the resulting constructs are gynogenetic F1 hybrids, which may help avoid any inbred degeneracy or genetic imbalance from an F2 background. Moreover, an F1 genetic background is known to enhance the SCNT cloning success rate [28, 29]. This approach also allowed us to introduce the green fluorescence protein (GFP) marker into pES cells at the same time (Fig. 2A, 2B). Oocytes from C57BL/6-GFP Tg [30] and 129/Sv mature females (2–3 months old) were collected following superovulation [4]. A group of 129/Sv oocytes was transferred to a droplet (∼10 μl) of HEPES-CZB medium [31] containing 5 μg/ml cytochalasin B under mineral oil on the micromanipulation microscope stage at 27°C. A normal oocyte was selected [32] and immobilized at the tip of a holding pipette by gentle suction. The zona pellucida of the oocyte was then penetrated by a piezo-actuated enucleation pipette (Prime Tech, Ibaraki, Japan, http://www.primetech-jp.com) [21]. The metaphase II chromosome-spindle complex, distinguished as a translucent disc in the ooplasm, was drawn into the pipette with minimal accompanying ooplasm. As the pipette tip was withdrawn through the zona pellucida entry site, the contents were pinched off. Each chromosome-spindle complex was then inserted into the perivitelline space of C57BL/6 oocytes through the zona pellucida and then transferred to a fusion chamber with 0.5-mm-separated electrodes covered with calcium-free fusion medium (0.3 M mannitol, 0.1 mM MgSO4, and 0.1% polyvinyl alcohol). A pulse of 20 V at 1 MHz alternating current was applied for 2 seconds to align the oocyte and karyoplast. Immediately after this, 100 V of direct current was applied for 20 microseconds to induce fusion. The oocyte-karyoplast mosaics were kept for 5 minutes in the chamber and then transferred to CZB medium and incubated at 37°C under 5% CO2 in air for at least 1 hour to recover from any fusion damage. The fused cells with two metaphase II spindles were transferred to oocyte activation medium (10 mM SrCl2 in Ca2+-free CZB medium) [33] for 1 hour. Reconstructed oocytes that extruded two second polar bodies were cultured 4 days further in CZB medium. When the parthenotes developed to blastocysts, they were used for the establishment of pES cells.

thumbnail image

Figure Figure 1.. Diagram of hybrid parthenogenetic/gynogenetic mouse embryo. The metaphase II chromosome-spindle complex of a 129/Sv strain mouse oocyte was removed and fused with a C56BL/6-GFP oocyte. Those oocytes become tetraploid and possess two MII spindles. When reconstructed oocytes are activated by SrCl2, those oocytes extrude two second polar bodies and became diploid embryos. These embryonic cells express GFP but also have the potential to enhance nuclear transfer cloning efficiency because of their F1 genetic heterogeneity. Abbreviation: GFP, green fluorescence protein.

Download figure to PowerPoint

thumbnail image

Figure Figure 2.. In vitro differentiation of parthenogenetic embryonic stem (pES)-nuclear transfer (NT) and pES cells. (A–D): The NT-pES cells showed pluripotency and potential for differentiation. (A): Phase contrast imaging of PG1(III)-1 NT-pES cells. (B): Expression of green fluorescence protein (GFP) under UV light. (C, D): Cells positive for the expression of the pluripotency marker Oct3/4 (C) and embryoid body formation (D). (E–G): Chimeric mice derived from PG1(I) pES cells. (H–J): Chimeric mice derived from PG1(III)-1 NT-pES cells. Brain (F, I) and heart (G, J) tissues were extracted from those chimeric mice. The contributions of pES/NT-pES cells were detected from GFP fluorescence signals, which were expressed only in the pES or NT-pES cells (E′–J′). PG1(III)-1 cells contributed more to chimeric mice than PG1 cells. (K, L): Spectral karyotyping with fluorescence in situ hybridization painting of long-passaged pES cells showing abnormal karyotypes. (K): All cells had lost one X chromosome and often showed trisomy of chromosome 8. (L): PG3 and PG3(III)-1, the third generation of cell line PG3(I) showing triploid karyotypes in all cells. (M): Comparison of contribution rate of pES and NT-pES in each organ of the chimeric mice studied. We estimated the contribution rate by fluorescence-activated cell sorting analysis using the GFP signal as a marker. The rates of contribution of some NT-pES cells—PG1(II)-1, PG1(II)-3, and PG2(II)-1—were higher than those of the pES cells PG1(I) and PG2(I). The contribution rates of PG1(I) long-passaged cell lines were not higher than those of the original PG1 cell line. However, the contribution rates of all NT-pES cell lines were not higher than control embryonic stem cells. Details are shown in Table 3.

Download figure to PowerPoint

Establishment of Parthenogenetic ES Cells

Parthenogenetic or cloned embryos at the morula or blastocyst stages were used to establish ES cell lines as described [1, 34, [35]36], with a slight modification in that 20% Knockout Serum Replacement (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) and 0.1 mg/ml adrenocorticotropic hormone (American Peptide Company, Sunnyvale, CA, http://www.americanpeptide.com) were added to the multiES cell medium instead of fetal calf serum [34]. Briefly, morulae/blastocysts were treated with acid Tyrode's solution to remove the zona pellucida and placed in 96-multiwell dishes coated with mouse embryonic fibroblasts (ICR origin) for at least 10 days. Proliferating outgrowths were dissociated using trypsin digestion and replated on fibroblasts until stable cell lines grew out. The culture medium was replaced with the cryopreservation medium Bambanker (Nippon Genetics, Tokyo, http://www.n-genetics.com), and the cell lines were stored in a −80°C freezer soon after being established, as described previously [37].

Nuclear Transfer and Establishment of NT-pES Cell Lines

Nuclear transfer was performed as described [21, 35]. Briefly, pES cells were thawed and cultured for 2 days. Individual pES cells were collected from flasks by trypsinization for 5 minutes. To remove the trypsin, cells were washed at least twice in 5 ml of phosphate-buffered saline (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com), and finally 2 μl of condensed cell suspension was transferred to Hepes-CZB containing 12% (w/v) polyvinylpyrrolidone (Mr 360,000) (Sigma-Aldrich). Oocytes were collected from BDF1 females (2–3 months old), and metaphase II spindles were removed as described above. For nuclear injection, enucleated oocytes were transferred to the micromanipulation chamber, and then each pES cell was drawn in and out of the injection pipette until the cell membrane was broken. In some cases, a few piezo pulses were applied to break the plasma membrane. Several nuclei were individually drawn into a pipette and injected individually into separate enucleated oocytes at room temperature. After nuclear transfer, the reconstructed oocytes were activated by exposure to 10 mM SrCl2 in Ca2+-free CZB medium in the presence of 5 μg/ml cytochalasin B and 1% dimethyl sulfoxide and cultured for 4 days in CZB medium [21, 35, 38]. Cloned parthenotes that reached the blastocyst stage were used to establish NT-pES cell lines as described [36]. All established cell lines were examined for the production of GFP to demonstrate the success of nuclear transfer. Even if some parthenogenetic embryos had been produced accidentally, they could have been detected because they would not express the gene for GFP. As a control, 129/Sv oocytes were activated parthenogenetically with cytochalasin B, which prevented second polar body extrusion and allowed the establishment of diploid parthenogenetic ES cell lines.

Immunohistochemistry

We used the appropriate manufacturers' staining procedures throughout. Alkaline phosphatase staining was according to the manufacturer's protocol (Sigma-Aldrich). Immunohistochemistry was performed using the following antibodies: anti-Oct3/4 (monoclonal 1:100; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), anti-Nanog (monoclonal 1:200; ReproCELL Inc., Tokyo, http://www.reprocell.com/en); anti-stage-specific embryonic antigen (AAEA)-1 and -3 (monoclonal 1:100; Chemicon, Temecula, CA, http://www.chemicon.com). Alexa Fluor 568-labeled secondary antibodies (Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com) were used for detection as appropriate.

Karyotype Analysis by Giemsa and Spectral Karyotyping with Fluorescence In Situ Hybridization Painting

Chromosomes from NT-pES cells were stained using Giemsa and spectral karyotyping with fluorescence in situ hybridization (SKY-FISH) chromosome painting techniques (Spectral Imaging Ltd., Vista, CA, http://www.spectral-imaging.com) according to the manufacturer's protocols. All NT-pES cells were used within five passages except for the PG1 long-passaged cells (described below), for which 30 passages were used. More than 50 metaphase nuclei (for Giemsa staining) or 15–20 metaphase nuclei (for SKY-FISH staining) were examined for each examined cell line.

Production of Chimeric Mice

Blastocyst injections were performed as described [1, 39, 40]. Briefly, ICR females were mated with ICR males, and eight-cell embryos were collected at 3 days post coitum (dpc). In some cases, BDF1 female mice mated with BDF1 male mice were also used to collect embryos. The next day, 15–20 of the pES or NT-pES cells were injected into each blastocyst, and those injected embryos were transferred into 2.5-dpc pseudopregnant ICR females. At 18.5–19.5 dpc, some of the pregnant females were euthanized by cervical dislocation, and chimeric offspring were obtained by cesarean section. Others were delivered naturally, and the chimeric offspring were examined for coat color contribution and germline transmission.

In Vivo Differentiation Analysis

The rates of contribution of pES or NT-pES cells to each chimeric mouse were analyzed. Six organs (brain, heart, liver, kidney, intestine, and skin) from the chimeric offspring were collected at 18.5 or 19.5 dpc, and small pieces of each tissue were homogenized with 0.25% trypsin. After filtration and washing, single cells were resuspended, and GFP-positive (parthenogenetic origin) or -negative (host embryo origin) cell numbers were counted using fluorescence-activated cell sorting (FACSAria cell sorter; BD Biosciences, San Diego, http://www.bdbiosciences.com), or collected into a separate tube. Nonviable cells were eliminated by 7-amino actinomycin D staining (BD Biosciences). The number of nonviable cells was 3% or less of the total in all internal organs. It is known that the GFP gene promoter is not subject to gene silencing and that GFP-positive cells can contribute to all internal tissues of adult chimeric mice, except for erythrocytes and hair [30].

In Vitro Differentiation

Induction of ES cell differentiation into neuronal and mesodermal cell lines was performed as described [41, [42]43]. To test for neuronal differentiation, we cultured pES and NT-pES cells on PA6 feeder cells and induced neural differentiation by the stromal cell-derived-inducing activity method [41, 42]; cultured cells were stained with three neuronal markers and 4,6-diamidino-2-phenylindole (DAPI). The numbers of colonies were counted, and the same examinations were repeated twice independently. We used nestin (BD Biosciences), class III β-tubulin (Covance, Princeton, NJ, http://www.covance.com), and tyrosine hydroxylase (Chemicon) as primary, middle, and terminal differentiation markers, respectively. Regardless of size, the numbers of DAPI-positive colonies were calculated as a proportion of the total colony number; we observed 60–90 colonies in each examination. For mesodermal cell differentiation, pES and NT-pES cells were cultured in α-minimal essential medium supplemented with 10% fetal calf serum and 5 × 10−5 M 2-mercaptoethanol for 4 days. Platelet-derived growth factor receptor-α (PDGFR-α) and FLK1/KDR were used as mesoderm markers [43], and the rate of differentiation was analyzed using a FACSCalibur cell sorter (BD Biosciences) at 4 days after induction. Two normally fertilized embryo-derived ES cell lines, E14tg2a (unknown passage number) and 129B6F1 (established in our laboratory and used within five passages), were used as controls.

Allele-Specific DNA Methylation Analysis

Genomic DNA was extracted from each set of cultured ES cells using DNeasy Tissue Kits (Qiagen, Hilden, Germany, http://www1.qiagen.com). The DNA was subjected to bisulfite modification, polymerase chain reaction (PCR) amplification, subcloning, and sequencing as described [44] with some modifications. The probes used were derived from PCR products, using the following primer sets: H19-bi forward (F), 5′-GGT TGA GGA TTT GTT AAG GTG TTA TTG-3′; H19-bi reverse (R), 5′-TAA TAA CTA ATT TAA ACA CTC CTC ACC-3′; IG-bi-F, 5′-AAG GTA TTT TTT ATT GAT AAA ATA ATG TAG-3′; and IG-bi-R, 5′-CCT ACT CTA TAA TAC CCT ATA TAA TTA TAC-3′.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Establishment of pES and NT-pES Cells

To establish original pES cell lines, 20 reconstructed oocytes were activated parthenogenetically, and six of these developed into blastocysts. When these were plated onto ES cell establishment medium, four first-generation (I) pES cell lines were established and labeled PG1(I) to PG4(I) (Table 1). Next, newly established pES cell nuclei were transferred into enucleated B6D2F1 oocytes, and the resulting NT blastocysts were used for new ES cell derivation [1]. As a result, totals of 18, 5, 7, and 6 second-generation NT-pES cell lines were obtained from the PG1(I), PG2(I), PG3(I), and PG4(I) pES cell lines (75%–88%), respectively (Table 1). We labeled the second-generation (II) NT-pES cell lines PG1(II)-1 to PG1(II)-18, derived from PG1(I) pES cell nuclei (Table 1). We repeated this process and obtained totals of 9, 15, 12, and 12 third-generation (III) NT-pES cell lines (71%–83%), respectively, which were labeled PG1(III)-1 to PG1(III)-9 and so forth (Table 1).

Table Table 1.. Establishment of pES and NT-pES cell lines
Thumbnail image of

All of the established pES/NT-pES cell lines were positive for alkaline phosphatase, Oct3/4 (Fig. 2C), Nanog, and SSEA-1 (ES cell marker) and formed embryoid bodies (evidence of totipotency; Fig. 2D). By microscopy (×20), no difference could be observed between pES and NT-pES cells.

Karyotyping

Seventeen cell lines (four original pES cell lines, one PG1(I) long-passaged cell line, four second-generation NT-pES cell lines, and eight third-generation NT-pES cell lines) were karyotyped by Giemsa staining and SKY-FISH painting. These showed that a normal extent of karyotypes was maintained, even to the third generation (58%–88% of cells). The only exception was the PG1(I) long-passaged cell, PG3, and its later generations. PG3(I), one of the original pES cell lines, showed trisomy in most cells (87%), and this phenotype was inherited to the third generation as shown in PG3(III)-1 (Fig. 2L).

In Vivo Differentiation Potential of NT-pES Cells

Using these four lines of pES and 10 lines of NT-pES cell lines, we compared the potential for in vivo differentiation in chimeric mice produced by cell injection into the blastocoels of ICR or B6D2F2 strain blastocysts. Because these cells were GFP-positive, we could distinguish the injected from the host cells. Chimeric offspring were collected at 19.5 dpc, and the rate of contribution of pES/NT-pES cells was determined by GFP expression under fluorescent microscopy (Fig. 2E–2H). The birth rates of chimeric offspring increased with successive generations (48%–78%; Table 2), but offspring often died just after birth, irrespective of the generation. We found that the contribution rate of the pES cells was very low; only two pups (18%) were born with high GFP expression in the skin, but they were cannibalized the next day. However, when NT-pES cell lines were used, more than 60% of the chimeric mice showed high contribution rates, as determined by GFP expression throughout the body, irrespective of the cell line. Note that in this experiment, some of the chimeric mice showed growth retardation when either the pES or the NT-pES cell contribution rate was high. This was also reported previously when chimeric mice were constructed by aggregation between parthenogenetic embryos and normally fertilized embryos [15, 16], but this phenomenon was not observed when pES cells were used [17]. As there were several differences between these published studies and those reported here, such as the genetic background of the pES cell lines used, it is difficult to know the reason, but it is possible that our established pES cell lines retained the lethal genetic defects of parthenogenetic embryos, even after NT.

Table Table 2.. Contribution of parthenogenetic embryonic stem (pES)/nuclear transfer-pES cells to chimeras following cell injection into normal fertilization-derived blastocysts
Thumbnail image of

To further examine the contribution rates of pES and NT-pES cells, three to six offspring were collected randomly from each chimeric line, and six organs (brain, kidney, liver, heart, intestine, and skin) were examined by fluorescence-activated cell sorting (FACS) analysis (Fig. 2F, 2G, 2I; Table 3). The average contribution rate of pES cells in the chimeric mice was only 2%–8% in all tissues, as in previous reports [17], and thus these results did not confirm germline transmission, the most important criterion for genetic normality. However, when chimeric mice were produced from the second generation of NT-pES cells, germline transmission was confirmed in the PG1(II)-1 cell line by natural mating. Moreover, in some organs, especially in the brain and intestine, 7 of 10 examined NT-pES cell lines showed that contribution rates were increased up to fivefold compared with pES cell lines (Fig. 2I, 2I′; Table 3). However, the contribution rates were lower than in control ES cells, and in the heart and liver, the contribution rate did not increase even in the third-generation NT-pES cell line (3% in NT-pES vs. 2% in pES; Fig. 2J, 2J′). On the other hand, such significant improvement in the potential for differentiation in vivo was not observed in long-passaged PG1 cell lines (Fig. 2M). The contribution rate was not improved in successive NT generations, as shown in the PG(III)-1 cell line (Table 3).

Table Table 3.. Contribution of embryonic stem, pES, and NT-pES cells to chimeric offspring following cell injection into normal fertilization-derived blastocysts
Thumbnail image of

In Vitro Differentiation Potential of NT-pES Cells

We also examined the in vitro differentiation potential of pES and NT-pES cells into neuronal and mesodermal cells. Anti-nestin, anti-TuJ, and anti-tyrosine hydroxylase antibodies were used as markers for primary, intermediate, and terminal differentiation of neuronal cells, respectively [41, 42], and the numbers and percentages of positive colonies were calculated among the total. The primary differentiation of pES/NT-pES cells was similar to that of fertilized ES cells at 7 days after differentiation (Fig. 3A). However, the rate of differentiation of the pES/NT-pES cells was lower than control ES cells at 7 and 10 days (Fig. 3B). Although the rate of terminal differentiation of pES/NT-pES cells was also lower than that of control ES cells at 7 days after differentiation, the rate in NT-pES cells was significantly higher than that in pES cells following 3 days of culture (23.4% for PG1(I) vs. 53.5% for PG2(I) and 54.8% for PG3(I); Fig. 3C). When TH-positive cell colonies were counted separately from TuJ-positive colonies, the TH-positive rate of NT-pES cells was almost the same as that of the control ES line (Fig. 3D). Next, PDGFR-α and FLK1/KDR were used as mesodermal markers [43], and the rate of differentiation of PG1(I) and PG1(III)-1 cell lines was analyzed by FACS at 4 days (Fig. 3E–3G). The rate of differentiation into mesoderm was higher in the NT-pES cells than the pES cells, but it was still lower than in the control ES cells and was the same as that seen for neuronal differentiation. Thus, although parthenogenetic ES cells have a lower potential for differentiation in vivo and in vitro than ES cells, this potential can be improved by NT and by the reestablishment of pES cells as NT-pES cells.

thumbnail image

Figure Figure 3.. Differentiation of parthenogenetic embryonic stem (pES)/nuclear transfer (NT)-pES cells. (A–D): In vitro potential to differentiate into dopaminergic neural cells. We cultured pES and NT-pES cells on PA6 feeder cells and induced neural differentiation using the SDAI method. Percentages of nestin-positive (A), TuJ-positive (B), and TH-positive (C) colonies, indicating primary, intermediate, and terminal differentiation, respectively, are shown at 5, 7, and 10 days after differentiation. The rate was calculated on the basis of the numbers of 4,6-diamidino-2-phenylindole-positive colonies. (D): The potential for terminal differentiation to neural cells from TuJ-positive cells. Although neither pES nor NT-pES cells showed lower potential for differentiation into neuronal cells, the potential to differentiate into terminal stages from intermediate stages was increased in NT-pES cells, almost to the level of fertilized ES cells. (E): In vitro potential of pES, NT-pES, and E14 cells to differentiate into mesodermal cells. PDGFR-α and FLK1/KDR were used as mesodermal markers, and the rate of differentiation was analyzed by fluorescence-activated cell sorting 4 days after differentiation. In differentiated mesodermal cells, the surface markers FLK1 and PDGFR-α were detected as higher signals. The rates of differentiation into mesoderm were higher in PG1(III)-1 NT-pES cells (F) than in PG1 pES cells (E) but still lower than control E14tg2a ES cells (G), as with the case of neuronal differentiation. Abbreviations: FLK-1, fms-related tyrosine kinase 1; PDGFRa, platelet-derived growth factor receptor-α; SDIA, stromal cell-derived-inducing activity; TH, tyrosine hydrozylase.

Download figure to PowerPoint

DNA Methylation Status of H19-DMD and IG-DMR

To investigate whether nuclear transfer affects the typical epigenetic errors of parthenogenetic cells, we checked the DNA methylation status of H19-DMD and IG-DMR, which should show only hypomethylated alleles in parthenogenetic cells [45, 46]. As shown in Figure 4, we confirmed that the original pES cell lines (PG1(I) and PG2(I)) showed such a hypomethylated status. Interestingly, although the in vivo differentiation potential of NT-pES cell was improved in chimeric mice, the hypomethylation status of all alleles was maintained.

thumbnail image

Figure Figure 4.. Differential methylation status of the H19-DMD and IG-DMR alleles. The methylation patterns of the DMRs of H19 (A) and IG (B), which are upstream of Dlk1, were detected using bisulfite sequencing (filled circles, methylated; open circles, unmethylated). Cytosines are shown for a number of independently sequenced templates (horizontal lines). These DMRs are methylated in paternal gametes only. In the normally fertilized cell, approximately half the sequenced templates were highly methylated, presumably the paternal allele. By contrast, in the parthenogenetic cells, all sequenced templates were unmethylated because they possessed only maternal alleles. In the NT-pES cells, almost all sequenced templates were unmethylated, and significant changes were not detected in H19 or IG DMRs, even after nuclear transfer. Abbreviations: DMD, differentially methylated domain; DMR, differentially methylated region; ES, embryonic stem.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

The established NT-pES cells showed greatly enhanced differentiation potential in vivo and in vitro. When we examined the chimeric mice, the rate of contribution of NT-pES cell varied between organs: for example, brain and intestine showed contribution rates two to five times higher than the original pES cells. This remarkable change occurred in the first round of NT, and successive nuclear transfers produced no significant change in the cell line characteristics. This suggests that the most important genes for embryonic development are reprogrammed by the first NT. This is supported by our previous results, showing that the cloning success rate did not increase when NT was repeated successively several times [1, 2, 47, 48]. In general, it is thought that the SCNT procedure causes epigenetic modifications to the donor nuclei, transforming them from a somatic state to an embryonic state and including imprinted genes [24, [25], [26]27]. Recently, parthenogenetic mice were born following NT using a combination of H19 gene knockout mouse oocytes and nongrowing oocytes [49]. Most offspring died just after birth; two healthy pups were obtained, but this may have been the result of reprogramming of imprinted genes after NT. Astonishingly, the parthenogenetic embryos had almost normal developmental potential. This study found remarkable epigenetic modification of an imprinted gene, Dlk1. Although we could not detect similar alternations in the differentially methylated sequences of either H19-DMD or IG-DMR, it may be that some epigenetic modifications in the NT-pES cell nuclei occurred after nuclear transfer. Interestingly, in the heart and liver, the NT-pES cells contributed only approximately 4%, which is consistent with previous work [15, [16]17]. It is possible that the effects of NT differ between internal organs or that paternal gene expression is required in particular organs, and this could not be reversed by the NT procedure.

It is known that there is epigenetic instability between ES cells and their subclones [27], so the improvement of NT-pES cell performance might be caused by NT-based subcloning selection. This is because each NT-pES cell line was established from a single donor cell; only those nuclei with normal gene expression that is compatible with early development will permit blastocyst formation and the subsequent derivation of competent ES cells. Previously, the instability of the XX karyotype and DNA methylation has been characterized in a series of diploid pES cell lines, and the resulting ES cells possessed only one X chromosome (XO genotype) [50, 51]. Therefore, if subcloning is performed by NT-based selection, the rate of XO NT-pES cell lines will reflect the proportion of XO cells in the original pES cell line. In our experiments, all cells of the PG1 long-passaged line lacked one X chromosome (passage 30; Fig. 2K). This cell line was cultured for more than 2 months (usually one passage every 2 days), which was approximately the same culture period for the establishment of the second-generation of NT-pES lines. However, all our NT-pES cell lines except for the PG3 series were judged to be normal XX cell lines, irrespective of generation. Moreover, although allelic methylation differences between subclones have often been observed for the H19 gene [27], all the NT-pES cell lines examined here maintained the same methylation status as the original pES cell line for H19. Thus, although we cannot exclude the possibility of NT-based selection, it appears likely that NT techniques could be applied to improve some particular epigenetic disorders of cell lines.

On the other hand, one pES cell line (PG3) and its following NT-pES cell lines were triploid. Probably, when the original reconstructed oocyte was activated, it extruded only a single second polar body. We may have mistaken the first polar body as a second one. Interestingly, this triploid pES cell line could still contribute to chimeric mice, and the karyotypes were maintained after NT, but the rate of contribution of this abnormal cell line was not improved after NT. This suggests that NT can improve only epigenetic disorders, not genetic abnormalities per se.

In this study, we established NT-pES cell lines from all original pES cell lines, with a very high success rate. Previously, NT-ES cells have been established from somatic cells or ES cells, and the rate of establishment is approximately 30%–50% from cloned blastocysts [1, 2, 4, 47, 52]. However, in the present experiments, the rates of establishment of all NT-pES cell lines were over 70%, irrespective of cell line or NT generation. The reason is not clear, but probably it was because we used very early passaged cells (within five passages) and the B6129F1 genetic background. This strain exhibits the best F1 genetic combination for successful mouse cloning [29]. It seems unlikely to be associated with parthenogenesis per se, because the control ES and pES cells were almost identical except for the imprinted alleles.

Embryonic stem cell derivation from parthenogenetically activated oocytes has been envisioned for autologous cell therapy without the need to destroy normally competent embryos or to require donation of another woman's oocytes [12]. Although parthenogenetic ES cells have some abnormalities in their gene regulation systems [17], their capacity for differentiation is clearly improved by successive nuclear transfer procedures. So far, NT techniques required additional oocytes; however, in this study, only eight oocytes per line were required for establishment (a mean of 12.4% from oocytes; Table 1) because of the high success rate of growing NT-pES cell lines. Thus, we have shown that incompetent ES cells can be improved through in vitro manipulation. Although these nuclei maintained parthenogenetic phenotypes (Fig. 4), those cloned embryos are still destined to die, so ES cell derivation could be done without destroying potentially live individuals. This procedure would help to circumvent the ethical objections to the sacrifice of normal embryos. We have also established a useful tool for investigating the effect of epigenetic modifications, because these NT-pES cells permit the possibility of investigating the roles of imprinted genes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

We thank Dr. Sasai, Dr. Watanabe, Dr. J. Cummins, and D. Sipp for discussion and critical reading of the manuscript. We are grateful to the Laboratory for Animal Resources and Genetic Engineering for the housing of mice. T.W. was supported by Grant 15080211, and T.W. and S. N. were supported by a Project for the Realization of Regenerative Medicine from the Ministry of Education, Science, Sports, Culture and Technology of Japan.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References