Downregulation of H19 Improves the Differentiation Potential of Mouse Parthenogenetic Embryonic Stem Cells

Abstract

Parthenogenetic embryonic stem cells (P-ESCs) offer an alternative source of pluripotent cells, which hold great promise for autologous transplantation and regenerative medicine. P-ESCs have been successfully derived from blastocysts of several mammalian species. However, compared with biparental embryonic stem cells (B-ESCs), P-ESCs are limited in their ability to fully differentiate into all 3 germ layers. For example, it has been observed that there is a differentiation bias toward ectoderm derivatives at the expense of endoderm and mesoderm derivatives—muscle in particular—in chimeric embryos, teratomas, and embryoid bodies. In the present study we found that H19 expression was highly upregulated in P-ESCs with more than 6-fold overexpression compared with B-ESCs. Thus, we hypothesized that manipulation of the H19 gene in P-ESCs would alleviate their limitations and allow them to function like B-ESCs. To test this hypothesis we employed a small hairpin RNA approach to reduce the amount of H19 transcripts in mouse P-ESCs. We found that downregulation of H19 led to an increase of mesoderm-derived muscle and endoderm in P-ESCs teratomas similar to that observed in B-ESCs teratomas. This phenomenon coincided with upregulation of mesoderm-specific genes such as Myf5, Myf6, and MyoD. Moreover, H19 downregulated P-ESCs differentiated into a higher percentage of beating cardiomyocytes compared with control P-ESCs. Collectively, these results suggest that P-ESCs are amenable to molecular modifications that bring them functionally closer to true ESCs.

Introduction

P arthenogenetic development starts when an oocyte is artificially activated and develops in the absence of a sperm. Parthenogenetic embryos have been used to derive parthenogenetic embryonic stem cells (P-ESCs) in mouse [1,2], monkey [3,4], rabbit [5], and human [6,7]. Importantly, in mammals, parthenogenetic embryos fail to develop past the early to mid-stages of gestation [8 –10].

P-ESCs provide an alternative source of stem cells for generation of autologous tissues and organs. This source of cells is mitochondria and major histocompatibility complex (MHC)-matched to the recipient [6,7,11 –16]. Thus, the application of human P-ESCs in regenerative medicine is feasible and would bypass many of the ethical concerns associated with destroying fertilized embryos for biparental ESC (B-ESC) derivation.

P-ESCs, like B-ESCs derived from fertilized embryos, can differentiate—in vitro as embryoid bodies (EBs) and in vivo as teratomas—into cells from the 3 major germ layers: ecto-, meso-, and endoderm [17,18]. Although these cells have the potential to differentiate in a manner similar to their biparental counterparts, a differentiation bias toward ectoderm derivatives at the expense of endoderm and mesoderm derivatives—and muscle in particular—has been observed in chimeric embryos, teratomas, and EBs [10,19,20]. The major reason for failure of the PG embryos to develop to term, and the limited differentiation potential of P-ESCs in teratomas and EBs, has been attributed to defects in establishing correct expression patterns of the imprinted genes [17,19,21 –24]. For example, the H19 gene has been reported to be highly deregulated in PG embryos and P-ESCs [25,26]. H19 is a maternally expressed (paternally imprinted) gene and codes for an untranslated mRNA. It has been reported that H19 can function as a long antisense RNA and as a micro RNA (mir675) [27,28]. In uniparental mouse fetuses the H19 gene was found to be overexpressed up to 400 times compared with biparental concepti [21,29]. Moreover, deregulation of H19 acts as a major barrier toward proper development of parthenogenetic embryos [21,22,30,31]. Kono and co-workers showed that a 13 kb deletion encompassing the H19 coding region and the upstream differentially methylated region (DMR) allows parthenogenetic embryos to develop to term [22]. The efficiency of full-term development significantly increases when the H19/Igf2 DMR together with the DMR found within a second imprinted gene cluster Dlk1/Dio3 on chromosome 12 are deleted [23,32]. Most recently Hikichi et al. showed that the development of PG embryos can be extended following serial nuclear transfer (NT) and is associated with a decrease in H19 expression [33]. Altogether, these results suggest that the correct expression of H19 is important for normal embryonic development.

The results of the aforementioned studies led us to hypothesize that correct expression of H19 is required for proper differentiation of mouse P-ESCs into the 3 germ layers. To test this we utilized a retroviral-mediated small hairpin RNA (shRNA) approach to silence H19 gene expression in mouse P-ESCs. Further, we developed a method to quantify the abundance of the tissue derivatives of the 3 main germ layers in teratomas. We found that downregulation of H19 led to an increase of mesoderm-derived muscle and endoderm in P-ESCs teratomas similar to that observed in B-ESCs teratomas. This phenomenon coincides with upregulation of the mesoderm-specific genes such as Myf5, Myf6, and MyoD. These results suggest that P-ESCs are amenable to molecular modifications that bring them functionally closer to true ESCs.

Materials and Methods

Mouse ESCs derivation and culturing

Mouse P-ESCs and B-ESCs cells were derived from parthenogenetically activated and fertilized oocytes, respectively. P-ESCs and B-ESCs were derived from TgR (ROSA26)26Sor strain of mice (The Jackson Laboratory) expressing a constitutive β-galactosidase gene in all tissues and organs of the animal [34]. P-ESCs cells were produced by artificial activation of metaphase II eggs preventing the extrusion of second polar body [4,35]. Briefly, for P-ESCs derivation, matured MII oocytes were collected around 16 h post-human chorionic gonadotrophin injection. The matured oocytes were activated by 5 mM SrCl₂ in 2 mM Ca²⁺-free mCZB medium in the presence of 5 μg/mL cytochalasin B for 6 h and cultured at 37°C in 5% CO₂ balance air up to 4 days in Ca²⁺ included mCZB medium [4]. Blastocysts were then treated with Tyrode's solution to remove the zona pellucida and placed in 96-multi well dishes coated with 90% confluent mytomycin C-inactivated mouse embryonic fibroblast (MEFs) until attachment. From the plated embryos that exhibited sufficient inner cell mass out-growth and colony formation, the colonies were manually cut, pipetted gently, and replated onto fresh mitomycin C-inactivated MEFs. The cells were sent for karyotyping to Cell Line Genetics. The P-ESCs line 1 and line 2 and 1 B-ESCs line used in this study have been karyotyped. Twenty cells at metaphase state have been karyotyped from each of the 3 cell lines mentioned above. The karyotype analysis is as follows: The P-ESCs line1 (also called mPGESR26-7) was karyotyped at passage 5 (p5) and found to contain 40XX normal female karyotype with a single cell 39, X, -X, +3, −14, and a single cell 41, XX, +marker chromosome. The P-ESCs line 2 (also called mPGESR26-1) was karyotyped at p4 and found to contain a normal 40XX female karyotype and no abnormal cells with trisomy 12 and/or 17 were detected. The B-ESC (also called mBPESR26-9) was karyotyped at p7 and found to posses normal 40,XY karyotype with a single cell 39,XY, −12. The cells were grown with complete medium consisting of KO DMEM (Invitrogen), knockout serum, 10% (Invitrogen), l-glutamine 2 mM (Invitrogen), nonessential amino acids 1×solution (Invitrogen), FGF2, 4 ng/mL, 2-mercaptoethanol (Sigma), penicillin–streptomycin, 1×final volume, and 10³ U of leukemia inhibitory factor (LIF) (Millipore). Normal fertilized B-ESCs controls were derived from fertilized blastocysts at 3.5 dpc [36] and cultured under the same conditions as the P-ESCs described above.

RNA isolation and cDNA synthesis from ESCs

Total mRNA from mouse P-ESCs and B-ESCs was isolated using the RNeasy Isolation Kit (Qiagen) following the manufacturer's instructions. Total mRNA amount was measured using Nanodrop. One microgram of total RNA with OD260/280>2.1 was used for first-strand cDNA synthesis. cDNA was synthesized with Superscript II (Invitrogen) using anchored Oligo (dT_12-18) primers (Invitrogen) and following the manufacturer's instructions.

RNA isolation and cDNA synthesis from formalin-fixed, paraffin-embedded teratomas

RNA was isolated from ectoderm and endoderm derivatives and from whole teratoma sections following strict RNAse and DNAse free techniques. Ectoderm and endoderm derivatives were isolated using laser capture microdissection system (LCM) (Supplementary Fig. S1A, B; Supplementary Data are available online at www.liebertonline.com/scd). Total mRNA was extracted using High Pure RNA paraffin Kit (Roche) and after each RNA extraction procedure, the RNA was further purified by using NucleoSpin RNA II protocol for RNA purification (Machinery-Nagel). A validation of this method has been described by Hoshida et al. [37]. Total RNA (tRNA) was measured by RNA picochip on Agilent 2100 Bioanalyzer. For the LCM obtained samples, 1 ng of total RNA was converted to cDNA using SuperScript II first strand cDNA kit (Invitrogen) using random oligo primers. For the tRNA extracted from whole sections, 1 μg was used for cDNA synthesis following the same protocol as mentioned above.

Quantitative real-time RT-PCR

The quantification of all gene transcripts was performed by reverse transcription of total RNA followed by absolute real-time quantitative RT-PCR using SYBR Green PCR Master Mix (Applied Biosystems). Absolute quantification using this method is described elsewhere [38,39]. Primers for absolute real-time PCR were designed using Primer Express program (Applied Biosystems) and derived from mouse sequences found in GeneBank (Supplementary Table S1). A primer matrix was performed for each gene tested to determine optimal concentrations. Each reaction mixture consisted of 2 μL of cDNA, optimum concentration of each forward and reverse primer, nuclease-free water, and 12.5 μL of SYBR Green PCR Master Mix in a total reaction volume of 25 μL (96-well plates). Reactions were performed in triplicate for each sample using an ABI Prism 7000 Sequence Detection System (Applied Biosystems). The thermal cycle consisted of 40 cycles of 95°C for 15 s and 60°C for 1 min. Standard curves for each gene and controls were constructed using tenfold serial dilution and run on sample plates as standards. For gene expression normalization (except for the expression analysis done on the LCM obtained endoderm and ectoderm derivatives from fixed tumors) expression levels of β-actin mRNA was used. For the normalization of gene expression in the LCM obtained endoderm and ectoderm derivatives from fixed tumors, Gapdh expression levels were used. Copies of β-actin and Gapdh RNA in each pool were determined using standard curves constructed from the plasmid PCR2.1. Partial cDNA sequences for β-actin, Gapdh, H19, IGF2, and other genes tested were amplified from mouse P-ESCs, cloned into pCR2.1 Topo vector (Invitrogen), and subjected to fluorescent dye primer sequencing to confirm identity. Resulting plasmids were used to construct standard curves. Representative R² for β-actin Gapdh, H19, and the rest of the genes were estimated and only the one ≥0.98 used. For each measurement, threshold lines were adjusted to intersect amplification lines in the exponential portion of amplification curve [40].

Bisulfite sequencing analysis

Bisulfite sequencing of genomic DNA was performed using EZ DNA Methylation-Direct™ (ZymoResearch) and following the manufacturer's recommendations. The primers used for the bisulfite sequencing and the amplification conditions were as described elsewhere [41].

Quantification of endoderm, ectoderm, and mesoderm derivatives in teratomas

Teratomas from B-ESCs, P-ESCs, and P-ESCs infected with H19shRNA-expressing retroviral vector (H19shRNA cells) and P-ESCs infected with the control viral vector (Con.vector cells), were derived by subcutaneous injection into immune-deficient Nude mice, CD-1 (Charles River Laboratories). The teratomas were formalin-fixed, sectioned, and embedded in paraffin blocks; each was entirely submitted for histological analysis. Initially, 1–3 gross tissue samples per tumor were obtained for this study. The College of American Pathologists guidelines for gross sampling of solid tumors states that 1 sample for each centimeter of tumor or 3 samples, whichever is greater, should be used for tumor grading. Histological grading was performed utilizing the modified Thurlbeck-Scully histological grading system for solid ovarian teratomas proposed by Norris et al. [42] and successfully implemented by O'Connor and Norris [43] and Steeper and Mukai [44]. This system, which is widely employed clinically, estimates percentage of neuroepithelium in the total tumor mass by counting the number of microscopic 10×low-power fields (LPFs) to differentiate between tumors grades I, II, and III. We estimated that from our teratoma samples about 12–16 LPFs could be evaluated per full tissue section on 1 glass microscopic slide and that 3 tissue sections would yield ∼36–48 LPFs for examination, and thus comply with the modified Thurlbeck-Scully histological grading system. However, to increase our quantification precision in an inherently heterogeneous tumor and adjust for variability in tumor size (8–10 mm in diameter), we obtained consecutive additional deep level sections 15, 30, and 45 μm apart from each tumor sample paraffin block. Thus, the number of sections was increased to 10–18 per tumor, the number of LPFs evaluated was increased to 120–288 because the tumors were different size, and sections reflected whole tumor. The sections identity was hidden and the sections were also given to an independent histopathologist for evaluation.

Generation pMSCV-Hygro-H19shRNA and pMSCV-Hygro-EGFP-expressing viral vectors and infection of mouse P-ESCs

For generation of H19 shRNA-expressing plasmid, a fragment containing H1/T0 promoter and PolII termination site was excised from pENTR/H1/TO plasmid (Invitrogen) and was sub-cloned into a unique Xho1 site of pMSCV-Hygro viral vector (Clontech). Control viral vector was designed by inserting the β-actin-CMV-IE Enhancer-EGFP fragment derived from pCX-EGFP plasmid into a unique Xho1 site. Both viruses were packaged and infected into B-ESCs and P-ESC1 according to the manufacturer's recommendation (Clontech). Cells infected with the shRNA-expressing and control viral vectors were subjected to 200 μg/mL hygromycin selection for 10–14 days. Single transgenic clones were selected and expanding on fresh mitomyosin C (Invitrogen)-treated MEFs.

Statistics

qRT-PCR experimental data and teratoma quantification data were analyzed by analysis of variance procedure using the mixed procedure of SAS. Differences of P<0.05 were considered statistically significant.

Results

Generation of mouse P-ESCs and B-ESCs

Mouse P-ESC and B-ESC lines were derived from parthenogenetic and biparental blastocysts of Rosa 26 strain of mice (see Materials and Methods section). All lines (P-ESC1, P-ESC2, and 1 B-ESC) were found to possess a normal karyotype when cultured under normal growth conditions (see Materials and Methods section). Morphologically, the P-ESCs and B-ESCs displayed small cytoplasmic/nuclear ratio and grew as small compact colonies with prominent nucleoli (Supplementary Fig. S2A, B). These cells were extensively propagated in vitro and maintained their ESC morphology. The P-ESCs and B-ESCs stained positive for β-galactosidase (Supplementary Fig. S2C, D) and the pluripotency markers OCT4 and SSEA1 (Supplementary Fig. S3). Moreover, after injection into immunodeficient Nude mice-CD-1 the P-ESC lines were capable of forming teratomas (Supplementary Fig. S4).

Alteration of key imprinted genes in P-ESCs

In vitro culture of B-ESCs and P-ESCs can lead to alterations in DNA methylation and covalent modifications of histones such as histone acetylation and methylation [19,33,45 –48]. Moreover, in P-ESCs the expression of several imprinted genes is altered due to absence of a paternal chromosomal complement [19,46,47]. We analyzed the expression profile of the paternally imprinted gene H19 and the maternally imprinted gene Igf2. Both H19 and Igf2 share the same DMR as the main regulator of their expression (54). We found that expression levels of the paternally imprinted gene (i.e., maternally expressed gene) H19 was upregulated greater than 6-fold in P-ESCs compared with control B-ESCs (P<0.05; Fig. 1A). Contrary to its maternal imprinted status, there were significantly higher levels of Igf2 in P-ESCs compared with B-ESCs (Fig. 1A). These results demonstrate that the expression of key imprinted genes such as H19 and Igf2 is deregulated in P-ESCs.

FIG. 1.

Alterations in H19/Igf2 expression and DMR methylation in P-ESCs. (A) Quantification of mRNA abundance of H19 and Igf2 imprinted gene expression in mouse B-ESCs and P-ESCs by qRT-PCR. (B) Methylation profile of H19/Igf2 DMR in Rosa 26 strain of mice, on the bottom are indicated the CpG sites in nucleotides (nt), on the top is indicated the number of the CpG site. (C) Percent (%) methylation of the 15th CpG islands of the DMR in Rosa26 strain of mice. For all qRT-PCR experiments 3 biological replicates in triplicate were used in each run. P<0.05 has been considered as statistical significant, unless stated otherwise. a and b indicate statistical significant difference (P<0.05). B-ESCs, biparental embryonic stem cells; DMR, differentially methylated region; P-ESCs, parthenogenetic embryonic stem cells.

Partial CpG methylation of the H19/Igf2 DMR in P-ESCs

The methylation status of the H19/Igf2 DMR has been viewed as the main regulator of H19 and Igf2 gene expression [49]. On the maternal allele the CpG dinucleotides are not methylated, which allows for a CTCF insulator protein to bind, preventing the enhancer downstream of H19 to reach Igf2 gene [49]. This leads to H19 being expressed on the maternal allele and Igf2 silenced. On the paternal allele the DMR is methylated, which inhibits binding of the CTCF insulator protein and allows for the expression of Igf2 from the paternal allele while H19 remains silenced. Since we were able to detect expression of Igf2 in our P-ESCs, and the expression of H19 and Igf2 is coordinately regulated, we decided to analyze the methylation status of the H19/IGF2 DMR. It has been previously reported that this region in CD-1 mice contains 16 critical CpG sites important for CTCF binding [41]. After bisulfite sequencing analysis we found that in the Rosa26 strain of mice there are 15 CpG sites, the fourth of which seems not to be methylated in normal fertilized B-ESCs (Fig. 1B, C). The DMR methylation status in both P-ESC lines (P-ESC1 and P-ESC2) revealed a partial methylation of the CpG sites analyzed with P-ESC1 and P-ESC2 having 7.3% and 19.3% methylation, respectively (Fig. 1B, C). These results demonstrate that the DMR of the H19/Igf2 genes is not completely demethylated and that this may account for the elevated levels of Igf2 expression in mouse P-ESCs.

Elevated levels of H19 persist throughout P-ESCs differentiation in vivo

Next we wanted to determine the expression status of the H19 gene throughout differentiation in vivo. We hypothesized that expression levels of H19 remained higher in teratomas derived from P-ESCs compared with B-ESCs. To test this we injected P-ESCs and B-ESCs subcutaneously into immune-deficient Nude mice-CD1. Teratomas were recovered and stained for β-galactosidase expression, confirming that the teratomas originated from the injected cells (Supplementary Fig. S5A). The ectoderm, endoderm, and mesoderm derivatives were also stained for α-fetoprotein, β-tubulin, and α-smooth muscle actin indicating the identity of the morphologically observed structures (Supplementary Fig. S5B). qRT-PCR data indicated that the expression levels of H19 in ectoderm and endoderm derivatives and in whole teratoma sections from P-ESCs teratomas were significantly higher than those in controls (P<0.05; Fig. 2A, B, respectively). Therefore, upon in vivo differentiation, the expression pattern of the imprinted gene H19 remains faithful to the origin of the P-ESCs from which the teratomas were derived.

FIG. 2.

Elevated levels of H19 expression persist during P-ESCs differentiation in vivo. (A) Quantification of H19 from laser capture microdissection-isolated primitive glandular structures (endoderm) and neural rosettes (ectoderm) by qRT-PCR. (B) Quantification of H19 mRNA abundance in whole teratoma sections by qRT-PCR (P<0.05). a and b indicate statistical signifcant difference (P<0.05).

Teratomas derived from P-ESCs give rise to a lower proportion of mesoderm-derived muscle compared with teratomas derived from B-ESCs

Results of previous studies suggest that there is a differentiation bias of P-ESCs toward ectoderm derivatives at the expense of endoderm and mesoderm derivatives and muscle in chimeric embryos, teratomas, and EBs [19,20,50]. We hypothesized that correct expression of H19 is required for normal germ layer formation in vivo. We quantified the percentage of ecto-, endo-, and mesoderm derivatives in teratomas derived from our P-ESCs and B-ESCs lines using our modified Thurlbeck-Scully grading system (see materials and Methods). We found that teratomas derived from P-ESCs consisted of predominantly ectoderm derivatives (70%) (Fig. 3A) compared with B-ESCs (50%). There was less endoderm (∼20%) in P-ESC-derived teratomas compared with B-ESCs ones (∼30%). Interestingly, we found that P-ESCs give rise to less mesoderm-derived muscle (∼5%) compared with B-ESCs cell teratomas (∼20%) (Fig. 3A).

FIG. 3.

Lack of mesoderm muscle derivatives and downregulation of mesoderm-specific transcripts in P-ESCs upon differentiation in vivo. (A) Quantification (%) of germ layer derivatives: ectoderm, endoderm, and mesoderm (the letters above indicate a statistical significant difference between the B-ESCs and P-ESCs in their ability to give rise to the specific germ layer lineage, P<0.05) (B) qRT-PCR measuring the mRNA abundance of germ layer-specific transcripts in P-ESC- and B-ESC-derived teratomas (P<0.05).

We also performed qRT-PCR to analyze the expression pattern of genes specific to ectoderm (Vimentin and Nestin), endoderm [α-feto protein (Afp) and Gata4], and mesoderm (Myf5, Myf6, and MyoD) germ layers in these teratomas (Fig. 3B). In addition, we analyzed the expression pattern of the Mier2 gene. Mier2 is an early mesoderm-promoting gene that we identified as a target of mir675 by performing miRNA targets using TargetScan (Supplementary Table S2). qRT-PCR data indicated that there was no difference in expression of ectoderm and endoderm-specific transcript in P-ESC- and B-ESC-derived teratomas. In contrast, there was a significant decrease in the expression of Myf5, Myf6, and MyoD in P-ESCs teratomas compared with B-ESCs ones (P<0.05; Fig. 3B). The expression of Mier2 showed the same pattern (Fig. 3B) and coincided with high levels of H19 gene expression (P<0.05; see Fig. 2A, B). We also analyzed the expression pattern of deltaNp63. DeltaNp63 is a product of the Tumor Protein 63 gene (TP63) and is involved in stem cell maintenance and commitment to ectoderm lineages [51,52]. Disruption of deltaNp63 in mice results in failure of proper ectoderm development [53]. Overexpression of deltaNp63 may induce the P-ESCs, when differentiated in vivo, to give rise to mostly ectodermal derivatives. We found the expression of deltaNp63 to be 250-fold overexpressed in P-ESCs compared with B-ESCs control cells (P<0.05; Supplementary Fig. S6A). These data suggest that upon differentiation in vivo P-ESCs exhibit limited developmental potential toward endoderm and mesoderm germ layer, muscle in particular. This phenotype coincides with overexpression of H19 and downregulation of the muscle and mesoderm promoting genes—Myf5, Myf6, MyoD, and Mier2.

Downregulation of H19 expression in P-ESCs promotes mesoderm differentiation in vivo

Proper H19 regulation is central to normal development of parthenogenic embryos [22,54]. Moreover, H19 may be important for proper muscle cell differentiation [55,56]. For example, chimeras between wild-type and P-ESCs result in restricted contribution of P-ESCs to muscle tissues in chimeric embryos [56]. Another study by Milligan et al. found a connection between H19 and muscle differentiation in vitro [55]. In this study it was elegantly shown that H19 ncRNA accumulates in the cytoplasm of C2C12 myoblastic cells during muscle differentiation and is protected from degradation by association with polysome complexes. The association with polysomes leads to stabilization of the H19 ncRNA, which is important for proper muscle differentiation and is controlled by post-transcriptional, yet unknown, mechanism [55]. Because we found that teratomas derived from P-ESCs are composed of a very limited amount of muscle tissue (Fig. 3A), we hypothesized that deregulation of H19 expression in P-ESCs is responsible for the decreased differentiation potential of these cells into muscle. To test this hypothesis we employed an H19 shRNA-expressing retroviral vector system to chronically suppress H19 gene expression in P-ESC1 cells, given that they had the highest expression of H19 gene at the start of the experiments. P-ESC1 cells were infected with an H19 shRNA retroviral vector or control vector (Fig. 4). qRT-PCR analysis revealed that H19 was stably and efficiently downregulated in the H19shRNA cell lines compared with P-ESC1 cells infected with the control vector (Con.vector) cell lines (P<0.05; Fig. 4). Moreover, low levels of H19 transcripts were faithfully maintained in teratomas derived from H19 shRNA P-ESC1 cells (Fig. 5A).

FIG. 4.

Stable and efficient downregulation of H19 mRNA levels in mouse P-ESCs. Quantification of H19 mRNA abundance by qRT-PCR in H19 downregulated P-ESCs (H19shRNA) and in P-ESCs infected with the control retroviral vector (Con.vector) (P<0.05). a and b indicate statistical significant difference (P<0.05). shRNA, small hairpin RNA.

FIG. 5.

Downregulation of H19 expression in P-ESCs promotes normal endoderm and mesoderm differentiation in vivo. (A) Quantification of H19 by qRT-PCR in samples derived from whole teratoma sections (P<0.05). (B) Quantification (%) of germ layer derivatives: ectoderm, endoderm, and mesoderm in teratomas derived from B-ESCs, H19 downregulated P-ESCs (H19shRNA) and in P-ESCs infected with the control retroviral vector (Con.vector). (C) Quantification of mesoderm-specific transcripts: MyoD, Myf6, and Mier2 by qRT-PCR in samples derived from whole teratoma sections (P<0.05). a and b indicate statistical significant difference (P<0.05).

To determine whether downregulation of H19 expression improved the differentiation potential of P-ESCs in vivo, we used our modified Thurlbeck-Scully grading system to quantify the percentage of ecto-, endo-, and mesoderm derivatives in teratomas. We found that there was a significant increase in endoderm (15%) and mesoderm-muscle derivatives (11%) in H19 shRNA-derived teratomas compared with Con.vector P-ESC1 cells (P<0.05; Fig. 5B). This increase was similar to the percentage of endoderm and muscle derivatives in teratomas derived from B-ESCs (P<0.05).

To further confirm our findings, we performed qRT-PCR analysis of mesoderm-specific transcripts—MyoD, Myf6, and Mier2, whose transcription levels were significantly altered in teratomas derived from wild-type P-ESC1 cells (see Fig. 3B). We found that upon downregulation of H19, the levels of MyoD, Myf6, and Mier2 were significantly upregulated in H19 shRNA-derived teratomas compared with control ones (P<0.05; Fig. 5C). Moreover, we found that suppression of H19 gene expression in P-ESC1 cells results in significant downregulation of the deltaNp63, which coincides with less propensity of these cells to differentiate to ectoderm lineage (P<0.05; Supplementary Fig. S6B). Altogether, these results suggest that H19 is required for proper mesoderm differentiation in vivo.

Induction of beating EBs from H19 downregulated P-ESCs

To further compare and analyze the differential potential of B-ESCs, P-ESCs, Con.vector P-ESC1 cells, and H19 shRNA P-ESC 1 lines, we employed an EB assay to examine cardiomyocyte formation. We adopted a protocol for derivation of beating cardiomyocytes by Yuasa et al. [57]. The incidence of beating EBs was recorded at days 8, 10, and 12 of EB formation. Beating EBs stained positive for cardiomyocyte-specific markers α-sarcomeric actin, α-sarcomeric myosin, and Myf5 (Fig. 6A). Cardyomyocyte contractions were inhibited upon addition of 10 μM Verapamil, which acts by specifically blocking L-type Ca²⁺ channels in smooth and cardiac muscle (Supplementary Movie S1). Our data indicated that P-ESCs have a lower potential to give rise to beating EBs compared with control B-ESCs. Remarkably, upon downregulation of H19, this phenotype was completely reversed (Fig. 6B). The incidence of beating cardiomyocytes was similar between EBs derived from H19shRNA P-ESCs and B-ESCs (P>0.05). In summary, downregulation of H19 gene expression increases the differentiation potential of P-ESCs into cardiomyocytes and beating EBs in a manner similar to that of normal B-ESCs.

FIG. 6.

Derivation of beating EBs from B-ESCs, Control P-ESCs, and Control vector (Con.vector), and H19shRNA P-ESCs. (A) Immunostaining for cardiomyocyte-specific markers: α-sarcomeric actin (αSM), α-sarcomeric myosin (αSM), and Myf5. (B) Incidence of derivation of beating EBs in percentage (%) on y-axis vs. days in culture on the x-axes. These experiments were repeated 3 times and total of 100 B-ESCs, 176 Control P-ESCs, 140 H19shRNA P-ESCs, and 224 Control vector (Con.vector) P-ESC-derived EB were analyzed. EB, embryoid body. Color images available online at www.liebertonline.com/scd

Discussion

Parthenogenetic embryonic stem cells: an alternative source of pluripotent stem cells in regenerative medicine

In light of an increased need for cell and tissue replacement therapy, a number of different strategies have been implemented to obtain an MHC-matched pluripotent source of stem cells. P-ESCs are unique in that they can potentially overcome the current limitations with the destruction of viable embryos, produced by fertilization or NT, during B-ESCs derivation. Also, the efficiency of derivation of P-ESCs is higher than the efficiency for derivation of B-ESCs from NT [14,15,58,59].

Moreover, P-ESCs with MHC antigens that are a complete match to the oocyte donor have been derived with high efficiency in mice, humans, and primates [6,14,15,59]. These cells are not only compatible with the oocyte donors' antigens, but also with the mother donor siblings.

The results reported in this study strongly suggest that H19 plays an important role for the P-ESCs development toward mesoderm germ layer, muscle in particular. We used teratomas as an experimental model for evaluating the differentiation potential of the P-ESCs instead of chimeras in an effort to avoid the potential effect of the host B-ESCs. During embryo development the P-ESCs in the chimeras are in constant cross-talk with the B-ESCs of the host embryo. These cell–cell interactions could, in turn, influence the P-ESCs developmental potential in the chimeras. Thus, to evaluate the intrinsic differentiation potential of our mouse P-ESCs, we used teratoma formation and cardiomyocyte differentiation assays to analyze the sole ability of these cells to give rise to derivatives of all 3 germ layers, ectoderm, endoderm, and mesoderm, and to beating EBs. The results reported in this study demonstrate for the first time, to our knowledge, that downregulation of a single imprinted gene can increase the differentiation potential of P-ESCs toward mesoderm development and beating EBs and that this phenotype coincides with an up-regulation of the expression of mesoderm-specific genes—MyoD, Myf6, and Mier2. These findings are consistent with reported studies by Surani and Kono's groups revealing the significance of H19 in development and differentiation of parthenogenetic embryos and P-ESCs [30,31,56].

In an earlier study McKarney et al. used parthenogenetic ESCs in an EB differentiation assay to evaluate the expression of H19 and Igf2 along with the muscle-specific genes Myogenin, Myf5, Myf6, and MyoD1 [60]. The authors did not observe an association between expression of muscle-specific genes and the level of H19 [60]. The apparent differences between our experimental findings and those by McKarney et al. may be due to the expression of H19 and/or the differentiation assay used to assess muscle formation. Unlike McKarney et al.—where the levels of H19 were not reported in the undifferentiated P-ESCs—we monitored the expression of H19 before and after differentiation and were able to make an association with H19 levels, the expression of muscle-specific genes, and muscle formation. We utilized an in vivo assay where we estimated the percentage of muscle contribution to whole tumor mass before and after down regulation of H19. We extended the findings of McKarney et al. by using a functional assay where the P-ESCs were induced to differentiate to beating cardiomyocytes after specific inhibition of BMP signaling; we demonstrated that P-ESCs with suppressed H19 expression were capable of giving rise to a higher percentage of beating EBs [60]. Clearly, H19 is not the only gene responsible for the observed phenotype. However, we have shown evidence indicating that H19 participates in the cascade of molecular events leading to muscle and endoderm formation.

The exact mechanism through which H19 exerts its role in the P-ESCs differentiation potential is currently not known. One suggestive mechanism could be that H19 through its product mir675 regulates the expression of MyoD, Myf6, and Mier2. We already have identified Mier2 as a potential target of mir675 based on miRNA TargetScan (Supplementary Table S2) [61]. In turn, mir675 can target Mier2 either on transcriptional level by inducing Mier2 mRNA degradation and/or on post-transcriptional level by repressing Mier2 mRNA translation. Indeed, the potential of mir675 to induce translational repression has been recently shown in human cancer cell lines where RB protein was found to be a direct target of mir675 [62]. We are currently investigating the effect of mir675 on Mier2 expression. This will help elucidate the mechanism that H19 exerts on mesoderm layer development.

We also found the expression of deltaNp63 to be 250-fold overexpressed in P-ESCs compared with B-ESC control cells. It is well known that long noncoding RNAs (ncRNA), such as H19, are involved in transcriptional regulation of multiple gene targets [63]. Therefore, one possible mechanism for overexpression of deltaNp63 in P-ESCs may involve the H19ncRNA. H19ncRNA may act as transcriptional activator of deltaNp63 expression, and therefore suppression of H19 may result in decrease of the deltaNp63 transcript. However, further evaluation of these proposed mechanisms is required to support these hypotheses.

In conclusion, these data demonstrate that elevated expression of H19 in P-ESCs has a deleterious effect on the ability of these cells to give rise to mesoderm muscle derivatives. By suppressing H19 expression P-ESC can indeed differentiate into muscle at similar rate when compared with B-ESC. Our data, together with others, support the rationale for using P-ESCs in the clinic. Studies in P-ESCs of primates, the closest species to human, have already shown the differentiation plasticity of these cells to give rise to all 3 germ layers in teratomas and to functional neurons when differentiated in vitro and transplanted into animal models [11 –13]. Further studies of the role of the imprinted genes in modulating the developmental potential of human P-ESCs are needed to assess the safety and efficacy of these cells to replace or regenerate the needed tissue in vivo.

Footnotes

Acknowledgments

We appreciate the support from the Michigan State University (MSU) Foundation, MSU's Vice President of Research, and the Michigan Agricultural Experiment Station (MAES), as well as the Naylor Family Foundation. This study was funded in part by U.S. National Institutes of Health grant DE13513 (B.C.S.). We would like to acknowledge the HistoPathology Laboratory at Michigan State University and especially Amy S. Porter and Kathy A. Joseph. We would like to acknowledge Dr. Hasan H. Out, Harvard University, Cambridge, MA, for his assistance with predicting mir675 target genes. We would like to acknowledge Dr. Juan Pedro Steibel, Animal Science Department, Michigan State University, for his assistance with the statistical data. We would like to acknowledge all past and present members of the Cellular Reprogramming Laboratory, Michigan State University. We would also like to acknowledge the “Van Andel Institute,” Grand Rapids, MI, for their cooperation.

Author Disclosure Statement

No competing financial interests exist.

References

Kaufman

et al. 1983. Establishment of pluripotential cell lines from haploid mouse embryos. J Embryol Exp Morphol, 73:249–261.

Robertson

, Evans

, Kaufman

. 1983. X-chromosome instability in pluripotential stem cell lines derived from parthenogenetic embryos. J Embryol Exp Morphol, 74:297–309.

Cibelli

et al. 2002. Parthenogenetic stem cells in nonhuman primates. Science, 295:819.

Mitalipov

, Nusser

, Wolf

. 2001. Parthenogenetic activation of rhesus monkey oocytes and reconstructed embryos. Biol Reprod, 65:253–259.

Fang

et al. 2006. Rabbit embryonic stem cell lines derived from fertilized, parthenogenetic or somatic cell nuclear transfer embryos. Exp Cell Res, 312:3669–3682.

Dighe

et al. 2008. Heterozygous embryonic stem cell lines derived from nonhuman primate parthenotes. Stem Cells, 26:756–766.

Revazova

et al. 2007. Patient-specific stem cell lines derived from human parthenogenetic blastocysts. Cloning Stem Cells, 9:432–449.

Kono

et al. 1996. Epigenetic modifications during oocyte growth correlates with extended parthenogenetic development in the mouse. Nat Genet, 13:91–94.

Barton

, Surani

MAH

, Norris

. 1984. Role of paternal and maternal genomes in mouse development. Nature, 311:374–376.

10.

Sturm

, Flannery

, Pedersen

. 1994. Abnormal development of embryonic and extraembryonic cell lineages in parthenogenetic mouse embryos. Dev Dyn, 201:11–28.

11.

Ferrari

et al. 2006. Transplanted dopamine neurons derived from primate ES cells preferentially innervate DARPP-32 striatal progenitors within the graft. Eur J Neurosci, 24:1885–1896.

12.

Sanchez-Pernaute

et al. 2008. Parthenogenetic dopamine neurons from primate embryonic stem cells restore function in experimental Parkinson's disease. Brain, 131:2127–2139.

13.

Sanchez-Pernaute

et al. 2005. Long-term survival of dopamine neurons derived from parthenogenetic primate embryonic stem cells (cyno-1) after transplantation. Stem Cells, 23:914–922.

14.

Kim

et al. 2007. Histocompatible embryonic stem cells by parthenogenesis. Science, 315:482–486.

15.

Lengerke

et al. 2007. Differentiation potential of histocompatible parthenogenetic embryonic stem cells. Ann N Y Acad Sci, 1106:209–218.

16.

Eckardt

et al. 2007. Hematopoietic reconstitution with androgenetic and gynogenetic stem cells. Genes Dev, 21:409–419.

17.

Park

et al. 1998. Differentiative potential of a mouse parthenogenetic embryonic stem cell line revealed by embryoid body formation in vitro. Jpn J Vet Res, 46:19–28.

18.

Sturm

et al. 1997. Unrestricted lineage differentiation of parthenogenetic ES cells. Dev Genes Evol, 206:377–388.

19.

Allen

et al. 1994. A functional analysis of imprinting in parthenogenetic embryonic stem cells. Development, 120:1473–1482.

20.

Onodera

et al. 2010. Differentiation diversity of mouse parthenogenetic embryonic stem cells in chimeric mice. Theriogenology, 74:135–145.

21.

Sotomaru

et al. 2002. Unregulated expression of the imprinted genes H19 and Igf2r in mouse uniparental fetuses. J Biol Chem, 277:12474–12478.

22.

Kono

et al. 2004. Birth of parthenogenetic mice that can develop to adulthood. Nature, 428:860–864.

23.

Kawahara

et al. 2007. High-frequency generation of viable mice from engineered bi-maternal embryos. Nat Biotechnol, 25:1045–1050.

24.

Ogawa

et al. 2006. Disruption of parental-specific expression of imprinted genes in uniparental fetuses. FEBS Lett, 580:5377–5384.

25.

Liu

J-H

et al. 2008. Diploid parthenogenetic embryos adopt a maternal-type methylation pattern on both sets of maternal chromosomes. Genomics, 91:121–128.

26.

Liu

et al. 2010. Genome-wide gene expression profiling reveals aberrant MAPK and Wnt signaling pathways associated with early parthenogenesis. J Mol Cell Biol, 2:333–344.

27.

Schoenfelder

et al. 2007. Non-coding transcripts in the H19 imprinting control region mediate gene silencing in transgenic Drosophila. EMBO Rep, 8:1068–1073.

28.

Berteaux

et al. 2008. A novel H19 antisense RNA overexpressed in breast cancer contributes to paternal IGF2 expression. Mol Cell Biol, 28:6731–6745.

29.

Platonov

, Penkov

, New

. 2002. [Effects of growth factors FGF2 and IGF2 on the development of parthenogenetic mouse embryos in utero and in vitro] Ontogenez, 33:60–67.

30.

Kono

. 2006. Genomic imprinting is a barrier to parthenogenesis in mammals. Cytogenet Genome Res, 113:31–35.

31.

Kono

et al. 2006. Paternal dual barrier by Ifg2-H19 and Dlk1-Gtl2 to parthenogenesis in mice. Ernst Schering Res Found Workshop, 60:23–33.

32.

et al. 2006. Regulated expression of two sets of paternally imprinted genes is necessary for mouse parthenogenetic development to term. Reproduction, 131:481–488.

33.

Hikichi

et al. 2010. Functional full-term placentas formed from parthenogenetic embryos using serial nuclear transfer. Development, 137:2841–2847.

34.

Zambrowicz

et al. 1997. Disruption of overlapping transcripts in the ROSA beta geo 26 gene trap strain leads to widespread expression of beta-galactosidase in mouse embryos and hematopoietic cells. Proc Natl Acad Sci U S A, 94:3789–3794.

35.

Kono

et al. 1996. A cell cycle-associated change in Ca2+ releasing activity leads to the generation of Ca2+ transients in mouse embryos during the first mitotic division. J Cell Biol, 132:915–923.

36.

Abbondanzo

, Gadi

, Stewart

. 1993. Derivation of embryonic stem cell lines. Methods Enzymol, 225:803–823.

37.

Hoshida

et al. 2008. Gene expression in fixed tissues and outcome in hepatocellular carcinoma. N Engl J Med, 359:1995–2004.

38.

, Wang

. 2000. Application of real-time polymerase chain reaction for the quantitation of interleukin-1[beta] mRNA upregulation in brain ischemic tolerance. Brain Res Protoc, 5:211–217.

39.

Whelan

, Russell

, Whelan

. 2003. A method for the absolute quantification of cDNA using real-time PCR. J Immunol Methods, 278:261–269.

40.

Bettegowda

et al. 2006. Quantitative analysis of messenger RNA abundance for ribosomal protein L-15, cyclophilin-A, phosphoglycerokinase, β-glucuronidase, glyceraldehyde 3-phosphate dehydrogenase, β-actin, and histone H2A during bovine oocyte maturation and early embryogenesis in vitro. Mol Reprod Dev, 73:267–278.

41.

Lucifero

et al. 2002. Methylation dynamics of imprinted genes in mouse germ cells. Genomics, 79:530–538.

42.

Norris

, Zirkin

, Benson

. 1976. Immature (malignant) teratoma of the ovary: a clinical and pathologic study of 58 cases. Cancer, 37:2359–2372.

43.

O'Connor

, Norris

. 1994. The influence of grade on the outcome of stage I ovarian immature (malignant) teratomas and the reproducibility of grading. Int J Gynecol Pathol, 13:283–289.

44.

Steeper

, Mukai

. 1984. Solid ovarian teratomas: an immunocytochemical study of thirteen cases with clinicopathologic correlation. Pathol Annu, 19,Pt 1:81–92.

45.

Dean

et al. 1998. Altered imprinted gene methylation and expression in completely ES cell-derived mouse fetuses: association with aberrant phenotypes. Development, 125:2273–2282.

46.

Horii

et al. 2008. Loss of genomic imprinting in mouse parthenogenetic embryonic stem cells. Stem Cells, 26:79–88.

47.

Jiang

et al. 2007. Activation of paternally expressed imprinted genes in newly derived germline-competent mouse parthenogenetic embryonic stem cell lines. Cell Res., 17:792–803.

48.

Bibikova

et al. 2006. Human embryonic stem cells have a unique epigenetic signature. Genome Res, 16:1075–1083.

49.

Sasaki

, Ishihara

, Kato

. 2000. Mechanisms of Igf2/H19 imprinting: DNA methylation, chromatin and long-distance gene regulation. J Biochem, 127:711–715.

50.

Jagerbauer

et al. 1992. Parthenogenetic stem cells in postnatal mouse chimeras. Development, 116:95–102.

51.

Honeycutt

, Koster

, Roop

. 2004. Genes involved in stem cell fate decisions and commitment to differentiation play a role in skin disease. J Investig Dermatol Symp Proc, 9:261–268.

52.

Senoo

et al. 2007. p63 is essential for the proliferative potential of stem cells in stratified epithelia. Cell, 129:523–536.

53.

Yang

et al. 1999. p63 is essential for regenerative proliferation in limb, craniofacial and epithelial development. Nature, 398:714–718.

54.

Kono

et al. 2002. Mouse parthenogenetic embryos with monoallelic H19 expression can develop to day 17.5 of gestation. Dev Biol, 243:294–300.

55.

Milligan

et al. 2000. H19 gene expression is up-regulated exclusively by stabilization of the RNA during muscle cell differentiation. Oncogene, 19:5810–5816.

56.

Fundele

et al. 1990. Temporal and spatial selection against parthenogenetic cells during development of fetal chimeras. Development, 108:203–211.

57.

Yuasa

et al. 2005. Transient inhibition of BMP signaling by Noggin induces cardiomyocyte differentiation of mouse embryonic stem cells. Nat Biotechnol, 23:607–611.

58.

Thuan

, Kishigami

, Wakayama

. 2010. How to improve the success rate of mouse cloning technology. J Reprod Dev, 56:20–30.

59.

Hao

et al. 2009. Human parthenogenetic embryonic stem cells: one potential resource for cell therapy. Sci China C Life Sci, 52:599–602.

60.

McKarney

, Overall

, Dziadek

. 1997. Myogenesis in cultures of uniparental mouse embryonic stem cells: differing patterns of expression of myogenic regulatory factors. Int J Dev Biol, 41:485–490.

61.

Friedman

et al. 2009. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res, 19:92–105.

62.

Tsang

et al. 2010. Oncofetal H19-derived miR-675 regulates tumor suppressor RB in human colorectal cancer. Carcinogenesis, 31:350–358.

63.

Barrandon

, Spiluttini

, Bensaude

. 2008. Non-coding RNAs regulating the transcriptional machinery. Biol Cell, 100:83–95.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

4.75 MB

1.17 MB