Differential gene expression patterns in porcine nuclear transfer embryos reconstructed with fetal fibroblasts and mesenchymal stem cells


  • B. Mohana Kumar,

    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    Search for more papers by this author
  • Hai-Feng Jin,

    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    Search for more papers by this author
  • Jung-Gon Kim,

    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    Search for more papers by this author
  • Sun-A Ock,

    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    Search for more papers by this author
  • Yonggeun Hong,

    1. Department of Physical Therapy, Cardiovascular & Metabolic Disease Center, College of Biomedical Science & Engineering, Inje University, Gimhae, Republic of Korea
    Search for more papers by this author
  • S. Balasubramanian,

    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    2. Department of Clinics, Madras Veterinary College, Tamilnadu Veterinary and Animal Sciences University, Chennai, Tamil Nadu, India
    Search for more papers by this author
  • Sang-Yong Choe,

    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    Search for more papers by this author
  • Gyu-Jin Rho

    Corresponding author
    1. Institute of Animal Medicine, College of Veterinary Medicine, Gyeongsang National University, Chinju, Republic of Korea
    • Department of Obstetrics and Theriogenology, College of Veterinary Medicine, Gyeongsang National University, 900 Gazwa, Chinju, Republic of Korea 660-701
    Search for more papers by this author


The present study compared the developmental ability and gene expression pattern at 4-cell, 8-cell, morula, and blastocyst stages of porcine nuclear transfer (NT) embryos from fetal fibroblasts (FFs) and mesenchymal stem cells (MSCs), in vitro fertilized (IVF), and in vivo derived embryos. MSC-NT embryos showed enhanced blastocyst formation, higher total cell number, and a low incidence of apoptosis compared to FF-NT embryos. Alterations in the expression pattern of genes implicated in transcription and pluripotency (Oct4, Stat3, Nanog), DNA methylation (Dnmt1, Dnmt3a), histone deacetylation (Hdac2), growth factor signaling, and imprinting (Igf2, Igf2r) and apoptosis (Bax, Bcl2) regulation were observed in NT embryos. The expression of transcripts in MSC-NT embryos more closely followed that of the in vivo derived embryos compared with FF-NT embryos. In conclusion, MSCs with a relatively undifferentiated genome might serve as suitable donors that could be more efficiently reprogrammed to re-activate expression of early embryonic genes in porcine NT. Developmental Dynamics 236:435–446, 2007. © 2006 Wiley-Liss, Inc.


Despite recent improvements in the cloning of various animal species by nuclear transfer (NT), only a few reconstructed embryos are capable of supporting term development after transfer (Jaenisch et al.,2005). A favored hypothesis for the developmental incompetence of clones is inadequate reprogramming of the transplanted nucleus to a state equivalent to that of an early embryonic nucleus. Nuclear reprogramming of donor cells (within the ooplasm) is a complex process involving the orderly switching off and on of a diverse array of somatic and embryonic genes, to support embryogenesis and concomitant developmental events (Tong et al.,2006). It is likely that inappropriate expression of key developmental genes may contribute to the lethality of cloned embryos (Li,2002; Jaenisch et al.,2005). Examination of the gene expression pattern in embryos with different developmental potential at the preimplantation stage will be valuable, because the dynamics can be observed just after reprogramming of the donor nucleus in enucleated oocytes (Li et al.,2005).

Errors leading to inappropriate expression could arise at any of the levels at which regulation of gene expression occurs, including the organization of the nucleus, chromatin structure, and availability of regulatory molecules (Li,2002). Studies on NT embryos during preimplantation development have revealed striking defects to faithfully recapitulate many of the essential early events including transcription/translation (Kirchhof et al.,2000; Boiani et al.,2002), epigenetic modifications such as DNA methylation (Dean et al.,2001; Mann et al.,2003), histone acetylation (Santos et al.,2003), and chromatin configuration (Vignon et al.,2002), as well as genetic imprinting (Latham et al.,1994). In support of these observations, embryos produced by NT in mice and bovine consistently displayed a myriad of developmental abnormalities caused by the aberrant expression pattern of genes involved in early development (Daniels et al.,2000; Wrenzycki et al.,2001,2005), maintenance of pluripotency (Kirchhof et al.,2000; Boiani et al.,2002; Bortvin et al.,2003), epigenetic reprogramming (Daniels et al.,2001; Wrenzycki et al.,2001; Mann et al.,2003; Vassena et al.,2005) and apoptosis regulation (Lonergan et al.,2003). Altered levels of expression have also been detected in NT embryos when compared to their in vitro and in vivo counterparts (Wrenzycki et al.,2001). However, the amount of information gathered on the profile of gene expression in porcine NT preimplantation embryos is still scarce and limited to a handful of genes (Zhu et al.,2004; Miyazaki et al.,2005).

The initial molecular mechanisms of reprogramming in NT embryos are still unclear and furthermore, the ease of reprogramming has been linked to the differentiation status of donor cell genome (Rideout et al.,2001; Zhu et al.,2004; Inoue et al.,2005; Jaenisch et al.,2005). In domestic animals, the technical difficulty in obtaining and maintaining an undifferentiated ES cell line has constrained the comparative studies on cloning efficiency using ES and somatic cells. However, attempts have been made to determine the efficiency of adult stem cells as nuclear donors, and have successfully showed the ability of mesenchymal stem cells (MSCs) in reprogramming and improving the developmental potential of bovine and porcine NT embryos (Colleoni et al.,2005; Bosch et al.,2006). Similarly, we also have previously demonstrated an enhanced potential of multipotent bone marrow MSCs as donor cells in achieving increased production of porcine NT embryos compared to fetal fibroblasts (FFs) (Jin et al.,2006).

In the present study, to gain more insights into the relation between the state of differentiation and cloning efficiency, we have assessed nuclear transfer embryos reconstructed with FFs and MSCs for their in vitro developmental ability and for gene expression patterns, and compared these with that of preimplantation in vivo and in vitro produced embryos as controls.


Cell-Surface Antigen Profile and Mesenchymal Lineage Differentiation of Porcine MSCs

The isolated MSCs from adult porcine bone marrow appeared as single, stretched cells or spindle shaped with long processes leading to large clusters of stellate cells as they multiplied by exhibiting their characteristic property of attaching to plastic culture dishes (Fig. 1A). Cell-surface antigen profile was ascertained by staining with specific monoclonal antibodies, and MSCs were found to be positive for integrins CD29 (β1-integrin) and CD49b (α1-integrin); matrix receptors CD44 (hyaluronate receptor) and CD105 (endoglin) are shown in Figure 1B–E. Furthermore, MSCs were negative for CD45 (leukocyte common antigen) and CD133 (prominin), indicating these cells are not of hematopoietic origin (data not shown).

Figure 1.

Morphology, cell-surface antigen profile, and mesenchymal lineage potential of porcine MSCs. In growth media (ADMEM+10% FBS), MSCs appeared with stable fibroblastic morphology (A). Cell-surface antigen profile of MSCs analyzed by immunofluorescence assay showed membrane localization (arrows) for integrins CD29 (β1-integrin) and CD49b (α1-integrin); matrix receptors CD44 (hyaluronate receptor) and CD105 (endoglin) (B–E, respectively). MSCs cultured in osteogenic medium formed colonies of mineralized matrix shown by alkaline phosphatase (AP) activity (F) and von Kossa staining (G). MSCs cultured in adipogenic medium displayed a characteristic phenotype of adipocytes with formation of lipid vacuoles demonstrated by Oil red O staining (H). MSCs induced for chondrogenic differentiation showed homogeneous Alcian blue 8GX stainable proteoglycans and mucopolysaccharides (I). Original magnification, ×100 (A), ×400 (B–E), ×200 (F–I).

Under osteogenic conditions, the spindle shape of MSCs flattened and broadened with the expression of alkaline phosphatase (AP) activity detected by day 12 leading to a steadily increasing number of AP-positive cells as the induction time progressed (Fig. 1F). Formation of mineralized matrix was evidenced by staining with von Kossa (Fig. 1G). The formations of lipid droplets filling the whole cytoplasm of single cells were observed when MSCs were grown in adipocyte-specific induction media. The lipid droplets were noticeable as early as 1 week after induction and visualized by staining with oil red O (Fig. 1H). After 3 weeks of MSCs differentiation in the presence of TGF-β1, chondrogenic potential was observed by the accumulation of sulfated proteoglycans as visualized by Alcian blue 8GX staining (Fig. 1I).

Developmental Rate, Total Cell Number, and Apoptosis

The developmental potential of FF-NT, MSC-NT, and IVF embryos assessed in terms of cleavage and blastocyst rate, total cell number, and apoptosis is presented in Table 1. The cleavage rate was significantly (P < 0.05) higher in IVF (82.6 ± 4.8%) than in NT embryos from FF and MSC (53.8 ± 3.4% and 64.7 ± 3.6%, respectively). However, blastocyst rates in IVF and NT embryos derived from MSCs did not differ (21.3 ± 2.7% and 20.1 ± 2.9%, respectively) but were significantly (P < 0.05) higher than NT embryos from FFs (9.9 ± 1.8%).

Table 1. Developmental Rate, Total Cell Number, and Apoptosis in Porcine In Vivo, IVF and NT Embryos
GroupsDonor cellsOocytes usedMean ± SEM
Cleavage (%)Blastocyst (%)Total cell numberApoptosis (%)
  • a, b, c

    Different superscripts in the same column denote a significant difference (P < 0.05), 5 replicates.

  • —, not applicable.

In vivo49.3 ± 4.21.3 ± 0.4a
IVF328271 (82.6 ± 4.8)b70 (21.3 ± 2.7)b39.8 ± 4.22.9 ± 1.3a
NTFFs312202 (64.7 ± 3.6)a31 (9.9 ± 1.8)a23.2 ± 3.77.8 ± 1.6c
 MSCs318171 (53.8 ± 3.4)a64 (20.1 ± 2.9)b34.5 ± 4.14.7 ± 1.5b

Total cell number did not show significant (P < 0.05) differences between IVF and NT embryos derived from MSCs (39.8 ± 4.2 and 34.5 ± 4.1, respectively) but were higher than from FFs (23.2 ± 3.7). However, a significantly (P < 0.05) higher number of total cells (49.3 ± 4.2) was observed in embryos derived from in vivo.

Proportional data (mean ± SEM) of TUNEL-positive cells in NT embryos from FFs (7.8 ± 1.6%) were significantly (P < 0.05) higher than from MSCs (4.7 ± 1.5%), IVF (2.9 ± 1.3%) and in vivo (1.3 ± 0.4%) derived embryos.

Comparison of Gene Expression Patterns in Embryos Derived From Different Origins

The relative abundance (RA) of the gene transcripts at the 4-cell, 8-cell, morula, and blastocyst stages in FF-NT and MSC-NT embryos, in vivo and IVF embryos is shown in Figure 2A–J.

Figure 2.

Relative abundance of Oct4 (A), Stat3 (B), Nanog (C), Dnmt1 (D), Dnmt3a (E), Hdac2 (F), Ifg2 (G), Igf2r (H), Bax (I), and Bcl2 (J) transcripts in porcine preimplantation embryos derived from in vivo, IVF, and NT (FF and MSC). mRNA from pools (triplicates) of 5 embryos each at the 4-cell, 8-cell, morulae, and blastocyst stages were reverse transcribed, and subjected to real-time quantitative PCR using transcript-specific primers (Table 2). All reactions were conducted in triplicate, and normalized for histone H2a mRNA expression. Data are presented as mean ± SEM (bars) of triplicate determinations. Values with superscripts “abcd” refers to significant (P < 0.05) differences in relative transcript abundance between in vivo, IVF, and NT embryos at a given stage of development.

Oct4 expression steadily increased and exhibited higher levels at the blastocyst stage of in vivo embryos. Though IVF embryos did not differ significantly (P < 0.05) when compared to NT embryos, the blastocyst, however, had a higher expression. In NT embryos, the blastocysts had a higher expression than other stages, but it was lower than in vivo blastocysts. The difference in Oct4 transcript between the NT embryos was significant (P < 0.05) only at the morula stage.

For Stat3 transcript, the level amongst in vivo derived embryos was lower at the 4-cell stage and gradually increased with a maximum abundance at the blastocyst stage, whereas the level was higher at the 4-cell stage than at other stages in IVF embryos. A significant (P < 0.05) variation between FF and MSC-NT embryos was observed at all stages, where, with an exception at the blastocyst stage, the mRNA for Stat3 was higher in FF-NT embryos than in MSC-NT embryos.

The RA of Nanog transcript was significantly (P < 0.05) higher at the 4-cell stage in FF-NT embryos, whereas no difference was observed between IVF and MSC-NT embryos. At the 8-cell stage, levels did not vary between FF and MSC-NT embryos, but were significantly (P < 0.05) higher than IVF embryos. No significant (P < 0.05) difference in the RA between IVF and FF-NT embryos was observed at the morula stage, while a minimal abundance was detected in MSC-NT embryos. Differences between IVF and NT embryos were significant (P < 0.05) at the blastocyst stage, where the levels were higher in MSC-NT embryos. In vivo embryos displayed a relatively higher abundance than IVF and NT embryos at all stages.

The Dnmt1 transcript level was significantly (P < 0.05) higher at all stages of NT and IVF than in in vivo embryos. In in vivo and IVF embryos, the level increased gradually and reached the maximum at the morula stage, then was down-regulated at the blastocyst stage, whereas in NT embryos it was higher at the 8-cell stage than at the morula and blastocyst stages. Dnmt1 levels at the 4-cell, 8-cell, and morula stages were significantly (P < 0.05) higher in FF-NT embryos followed by MSC-NT and IVF embryos, but there were no differences at the blastocyst stage.

The RA of Dnmt3a was low at the 4-cell stage, but increased markedly at the 8-cell, morula, and blastocyst stages in NT, in vivo, and IVF embryos. A significantly (P < 0.05) higher level was detected at the blastocyst stage of in vivo embryos, while a relatively lower transcript was observed at the 4-cell stage than in NT embryos. Among IVF and NT embryos, except at the blastocyst stage, a significantly (P < 0.05) higher abundance was observed at other developmental stages in FF-NT embryos.

Levels of Hdac2 increased gradually from the 4-cell to blastocyst stages in in vivo embryos. At the morula and blastocyst stages, it was significantly (P < 0.05) higher than IVF and NT embryos. At the 8-cell stage, no significant (P < 0.05) variation was observed between in vivo and IVF embryos. In contrast, the levels in NT embryos were down-regulated from the 4-cell to the morula stage with a slight increase at the blastocyst stage. But there were no significant (P < 0.05) differences observed at the 8-cell and morula stages. No differences were found between IVF and MSC-NT embryos at the blastocyst stage, whereas lower levels were detected in FF-NT embryos.

The Igf2 transcript level was similar for MSC-NT and in vivo embryos at all developmental stages, except at the 8-cell stage. It was higher at all stages in IVF and FF-NT than in in vivo and MSC-NT embryos. Moreover, it was lower at the 4-cell stage and further up-regulated at the 8-cell, morula, and blastocyst stages in NT, in vivo, and IVF embryos. Less variation was observed between NT and IVF embryos at all stages, whereas no significant (P < 0.05) differences were detected between FF and MSC-NT embryos at the morula and blastocyst stages. A minimal abundance was observed at all stages of in vivo embryos.

Transcription levels of Bax mRNA in in vivo, IVF, and NT embryos increased gradually and reached a maximum at the blastocyst stage. NT and IVF embryos showed higher levels at all stages than in vivo counterparts. Significantly (P < 0.05) higher RA of Bax was observed in FF-NT embryos followed by MSC-NT and IVF embryos.

Transcript abundance of Bcl2 was significantly (P < 0.05) higher in in vivo embryos at all stages compared to NT and IVF embryos. In FF-NT embryos, the levels were low from the 4-cell to the blastocyst stages, but at the 4-cell and 8-cell stages they were similar to MSC-NT embryos. Except at the 8-cell stage, no significant (P < 0.05) differences in the RA of Bcl2 were observed between MSC-NT and IVF embryos.


Numerous investigators have now demonstrated that MSCs can be isolated from a variety of adult tissues and differentiated into various tissue lineages (Pittenger et al.,1999; Jiang et al.,2002). In this study, fibroblast-like adherent MSCs isolated from porcine adult bone marrow resembled the morphological features of MSCs from earlier observations (Ringe et al.,2002; Vacanti et al.,2005; Jin et al.,2006). These cells were negative for lineage markers such as CD45 and CD133, but positive for MSC-specific markers including integrins and matrix receptors. Furthermore, multilineage differentiation along the three classical mesenchymal pathways—osteogenic, adipocytic, and chondrocytic lineages—was observed. Expression of cell-surface markers and lineage-specific differentiation demonstrated in this study confirm them as MSCs (Pittenger et al.,1999).

In porcine, the lower incidence of nuclear abnormalities and enhanced in vitro development of preimplantation NT embryos from fetal skin–derived stem cells suggested that these embryos were more competent to undergo appropriate remodeling during early development (Zhu et al.,2004). The highest formation of morulae/blastocysts per cleaved porcine embryos reconstructed with olfactory bulb progenitor cells implied that undifferentiated cells as nuclear donors may increase the efficiency of NT (Lee et al.,2006). NT embryos derived from bovine and porcine undifferentiated MSCs and their derivatives along the osteogenic lineage gave rise to consistently high preimplantation development (Colleoni et al.,2005). In our previous study, the cleavage rate was significantly higher in IVF embryos than in NT embryos derived from MSCs and FFs. However, blastocyst rates in IVF and NT embryos derived from MSCs did not differ but these rates were significantly higher than that for NT embryos derived from FFs (Jin et al.,2006). Our observations in this study further confirmed the results of the above investigations.

Adequate cell numbers within the preimplantation embryo have been demonstrated to be crucial for the progression of normal development (Machaty et al.,1998). In the present study, total cell numbers in NT embryos from MSCs were significantly higher than for those from FFs. Reduced preimplantation developmental competence in FF-NT embryos would have been manifested by low cell counts of blastocysts. Similarly, a significantly higher total cell number in the porcine fetal stem cells and MSC-NT embryos was observed than in those of the fibroblast NT embryos (Zhu et al.,2004; Jin et al.,2006), suggesting that stem cell–derived NT embryos are of higher quality than fibroblast NT embryos.

It is well known that the postimplantation developmental potential or embryo quality is likely to be affected by apoptotic incidence in the preimplantation stages. During in vitro culture, the NT porcine embryos exhibited higher rates of cytoplasmic fragmentation and developmental arrest as well as higher levels of apoptotic cells than the IVF embryos (Hao et al.,2003). In the present study, proportions of TUNEL-positive cells in NT embryos from FFs were significantly higher than those from MSCs, IVF, and in in vivo embryos. This finding supports a relationship between incidence of cell death and developmental potential as indicated by the lower developmental rate in NT embryos derived from FF. Aberrant gene expression is frequently observed in NT embryos of mouse, bovine, and porcine, and is one of the suggested causes of the low success rates of this approach (Boiani et al.,2002; Bortvin et al.,2003; Zhu et al.,2004; Miyazaki et al.,2005; Li et al.,2005; Tong et al.,2006). Examination of the gene expression pattern in NT embryos would aid in determining the causes of such abnormal development. To the best of our knowledge, this is the first report comparing mRNA expression patterns of development-related genes among preimplantation porcine NT embryos with different developmental potentials.

Oct4, a POU (Pit-Oct-Unc)-domain transcription factor, acts as a unique master regulator of pluripotent cells across mammalian species and is presumed to be a critical factor controlling murine, bovine, and porcine preimplantation embryonic development (Kirchhof et al.,2000). However, similar patterns of transcription were detected for Oct4 in bovine IVF and granulosa cell-NT embryos during the preimplantation stages of development (Daniels et al.,2000). Interestingly, no significant difference in expression of Oct4 at the 8-cell stage was observed between porcine NT embryos derived from fetal skin cells and fibroblasts, and in vivo produced embryos (Zhu et al.,2004). The present study demonstrated that Oct4 expression in NT embryos at the blastocyst stage was significantly decreased when compared with IVF and in vivo embryos. It has been reported that either an increase 150% or a decrease 50% of Oct4 expression serves as a trigger for the differentiation of pluripotent stem cells (Niwa et al.,2000). It has also been opined that a low Oct4 expression level at the blastocyst stage is associated with the low developmental potential of NT embryos (Li et al.,2005). These data indicate that an abnormal pattern of Oct4 expression in NT embryos might lead to their compromised developmental competence.

The Stat family of signaling molecules has been implicated in growth, differentiation, survival, and apoptosis (Chapman et al.,2000). Stat3, a member of the Stat family, is a cytoplasmic transcription factor important in cytokine signaling, the activation of which leads to activation of transcription of genes with specific Stat3 recognition sites (Levy and Lee,2002) and, thus, is essential for embryonic development. In the present study, except at the blastocyst stage, a higher abundance of Stat3 was observed in NT embryos derived from FFs than from MSCs, whereas a steady increase through all stages was observed in in vivo derived embryos. This is consistent with the observed up-regulation of Stat3 mRNA expression at the 8-cell stage in fibroblast-NT embryos compared with both fetal skin cells and in vivo produced embryos (Zhu et al.,2004). Up-regulation of Stat3 may influence cell proliferation (Zhu et al.,2004) and, interestingly, our results showed an increase in the number of cells at the blastocyst stage in IVF and also to some extent in MSC-NT embryos in contrast to its down-regulation and low cell number in FF-NT blastocysts.

Nanog is a homeobox-containing transcription factor with an essential function in maintaining the pluripotent cells of the inner cell mass and in the derivation of ES cells (Mitsui et al.,2003). In the present study, different transcription patterns of Nanog were observed among embryos with a higher abundance in in vivo derived embryos. A low level of Nanog expression induces instability in the pluripotency of ES cells, suggesting that Nanog functions to sustain pluripotency in a dose-dependent manner (Hatano et al.,2005). In addition, the expression of Nanog precisely corresponds to the pluripotent phenotype and, thus, is apparently present in the initial formation of the mouse blastocyst (Chambers et al.,2003). However, our results may be an indication that the molecular mechanisms activating the initial transcription of Nanog during development are different from those regulating maintenance of its expression in the pluripotent cell. We reasoned that a molecular understanding of the regulation of this gene in porcine preimplantation embryos would provide further insights into the maintenance of pluripotency.

Of the DNA (cytosine-5)-methyltransferases (Dnmt's), Dnmt1 is the most extensively studied and abundant DNA methyltransferase and is thought to be responsible for copying methylation patterns following DNA synthesis (Dean et al.,1998), whereas Dnmt3a and -b have been implicated in de novo methylation (Okano et al.,1999). Previous studies have shown species-dependent expression patterns of Dnmt genes in mammalian oocytes and preimplantation embryos (Vassena et al.,2005), and also a correlation between incomplete DNA methylation and the lack of NT success in mammals (Dean et al.,2001; Bortvin et al.,2003). Dnmt1 and Dnmt3a mRNA in bovine were detected continuously from the 2-cell stage to the blastocyst stage produced in vitro (Golding and Westhusin,2003), but the RA of the Dnmt1 transcript significantly varied between in vivo and in vitro produced bovine embryos, with in vivo produced embryos expressing significantly less Dnmt1 (Wrenzycki et al.,2001). Similarly, a significant increase in Dnmt1 expression at the 8-cell stage was reported in porcine FF-NT embryos compared with either in vivo produced or fetal porcine skin originated sphere stem cell NT embryos (Zhu et al.,2004). In this study, compared to IVF and in vivo embryos, the RA of Dnmt1 transcript was significantly increased in NT embryos. Interestingly, only a low level of Dnmt1 transcription was found for in vivo derived embryos. Additionally, Dnmt3a mRNA was expressed at all stages analyzed with a significant increase in abundance between the 8-cell and morula to blastocyst stages. Consistent with the above observations of aberrant DNA methylation in embryos, our data suggested that the reprogramming events that occur during the development of the in vivo produced embryos are less likely to occur in in vitro and NT produced embryos. Whether the consistently low efficiency of NT is related to the inability of a somatic nucleus to undergo the normal changes in methylation as indicated by increased levels of Dnmt1 or to the lack of de novo methylation triggered by low Dnmt3a expression remains unclear. In addition, Dnmt's participate in the regulation of genomic imprinting and modifications of chromatin (Li,2002). Hence, if these alterations remain persistent, this could lead to epigenetic changes suspected to be associated with the abnormalities seen after in vitro production and NT procedures (Niemann and Wrenzycki,2000).

Histone acetylation is a reversible process that is regulated by two classes of enzymes, the histone acetyltransferases (HAT's) and histone deacetylases (Hdac's). Earlier studies in mouse and bovine have shown the expression of Hdac's from the oocyte through the blastocyst stage (Schultz et al.,1999; McGraw et al.,2003) and from the oocyte to the 8-cell embryos at varying levels (Segev et al.,2001). In this study, Hdac2 mRNA demonstrated a low level at the 4-cell stage with a steady increase reaching a maximum at the blastocyst stage in IVF and in vivo derived embryos, whereas a reverse trend was observed in NT embryos. In mouse, the pattern for Hdac2 exhibited a high level of mRNA in the oocyte, followed by degradation until the 4-cell stage and then a major increase in the blastocyst (Schultz et al.,1999), whereas in bovine, an increase in mRNA expression from the 8-cell stage to the blastocyst was quite important for Hdac2 (McGraw et al.,2003). A possible explanation for this expression pattern may be variation in the time of maternal zygotic transition across species. Hdac's are most closely linked with transcriptional regulation owing to deacetylation of the core histones of the chromatin. Increased histone acetylation levels result in a transcriptionally active state by increasing the accessibility of transcriptional factors to DNA (Li,2002). However, repression could be associated with structural changes in the chromatin, and the discovery that histone deacetylation can be associated with chromatin offers enlightenment regarding how histone deacetylation relates to repression of gene expression (Schultz et al.,1999; McGraw et al.,2003). Taken together, our results indicate that the donor genome was unlikely reprogrammed to switch over the pattern of gene expression at the appropriate time from the somatic to the embryonic type following NT.

The Igf2 and Igf2r genes are among the best studied imprinted genes involved in fetal growth regulation, and are essential for normal development (Latham et al.,1994). Aberrant expression patterns of imprinted genes may be responsible for the deviated growth characteristics seen in fetuses and offspring originated from embryos produced using in vitro systems, including NT (Niemann et al.,2002; Reik et al.,2003). In this study, the expression of Igf2 and Igf2r transcripts was significantly increased in IVF and NT embryos compared to in vivo counterparts. Igf2 is a classic imprinted gene as it appears to be biallelically transcribed up to the morula stage, but in the blastocyst stage the maternal Igf2 allele is silenced (Mizuki et al.,2001). Thus, it appears that the NT embryos suffer a loss of imprinting and, therefore, overexpress Igf2. Furthermore, in vitro manipulation might induce epigenetic alterations in various developmentally important genes (Niemann et al.,2002) and studies have shown that such epigenetic mutations particularly affect the expression of genes that are regulated by genomic imprinting (Feil,2001). The Igf2 ligand is imprinted when inherited maternally and Igf2r is imprinted when inherited paternally (Latham et al.,1994), consistent with the hypothesis of Haig and Graham (1991), which predicts that imprinting of growth factors such as Igf2 and Igf2r regulates embryonic growth in the mammals. However, information with regard to imprinting of these genes in porcine is scarce.

Gene expression analysis during preimplantation embryo development revealed that oocytes and embryos express transcripts encoded by many genes involved in programmed cell death activation and execution (Juriscova et al.,2003). Bax is an apoptotic regulatory gene, which induces a caspase-dependent DNA degradation after inducing the release of cytochrome C from mitochondrial membranes (Robertson et al.,2002; Hao et al.,2003). In this study, Bax expression showed a steady increase through the developmental stages and the levels were significantly higher in NT embryos from FFs than in embryos from MSCs and IVF, suggesting that overexpression of the Bax gene accelerates apoptotic cell death (Oltvai et al.,1993). The elimination of unwanted cells is essential in development, but, conversely, apoptosis has the ability to eliminate viable cells, leading to the death of the organism; hence, it must be carefully regulated. This non-regulation of Bax mRNA production could be a cause for the low quality of cloned embryos.

In contrast, the Bcl gene family is known to include anti-apoptotic genes that protect a variety of cell types against apoptotic cell death. In this study, Bcl2 transcription in in vivo embryos was higher than in IVF and NT embryos. Transcript abundance of cell death regulatory genes may be altered in embryos undergoing fragmentation, so that expression of the gene involved in cell death (Bax) is elevated, and expression of the gene involved in cell survival (Bcl2) may be reduced. However, it is unlikely that the presence or absence of a single transcript (protein) will determine the survival or death outcome in preimplantation embryos because most cell death gene products work through a complex network of homo- and hetero-dimerization (Jurisicova et al.,2003). It has been reported that chromatin structure changes during early development and in porcine embryos, an in vitro developmental block is frequently observed at the 4-cell stage, a time coincident with major activation of the embryonic genome (Hao et al.,2003). The propensity to apoptosis is continuously counterbalanced in the cell by genes stimulating cell survival and proliferation. In this study, the expression pattern of Bcl2 in NT embryos reflected the results observed with respect to total cell number, number of apoptotic cells, and developmental potential. Our findings support the concept proposed by Oltvai et al. (1993) that the ratio of Bcl2 to Bax determines whether a cell lives or dies. However, it is also known that gene expression can be dramatically affected by the in vitro culture environment and can undergo considerable alteration without arresting development (Wrenzycki et al.,2001). Despite the importance of mRNA expression profiles of apoptosis-related genes in the porcine NT embryos, little information is available on this aspect.

The present study demonstrates the differential expression pattern of genes involved in transcription and pluripotency, DNA methylation, histone deacetylation, growth factor signaling and imprinting, pro-apoptosis and anti-apoptosis during preimplantation development of NT embryos compared to their IVF and in vivo derived counterparts. Moreover, the variations in the RA of target transcripts between IVF and in vivo derived embryos may support the fact that post-fertilization culture of porcine embryos has considerable influence on the transcriptional levels of the resulting embryos at different stages of development as reported earlier in bovine (Niemann and Wrenzycki,2000; Lonergan et al.,2003). In our results, the RA of target transcripts in MSC-NT embryos more closely followed that of the in vivo derived embryos compared with FF-NT embryos. This was further supported by an enhanced preimplantation development with increased cell number and low apoptotic positive cells. The low efficiency of preimplantation development of FF-NT embryos may, at least in part, be due to inadequate reprogramming leading to defects in early embryonic gene expression. Further, it is probable that the aberrant expression of target transcripts had a cumulative effect and was responsible, at least to some extent, for the lower developmental potential of FF-NT embryos compared with MSC-NT and normally fertilized IVF embryos. Our findings, therefore, suggest that MSCs with a relatively undifferentiated genome might serve as efficient donors for NT and may be more efficiently reprogrammed to re-activate expression of early embryonic genes.


All experiments were carried out according to Institutional guidelines for the care and use of research animals.

Chemicals and Media

All chemicals were purchased from Sigma-Aldrich Chemical Company (St. Louis, MO) and media were from Gibco (Invitrogen Corporation, Grand Island, NY), unless otherwise specified. The media used for culture of FFs and MSCs were Dulbecco's modified eagle medium (DMEM, high glucose) and advanced-DMEM (ADMEM) supplemented with 1 mM Na-pyruvate, 0.05 mM pyridoxine hydrochloride, 100 U/mL penicillin G, 100 μg/mL streptomycin sulfate, and 10% fetal bovine serum (FBS). The medium used for in vitro maturation (IVM) of cumulus-oocyte-complexes (COCs) was M-199 containing 5% FBS, 0.57 mM cysteine, 10 ng/mL epidermal growth factor (EGF), 25 mM HEPES, 2.5 mM Na- pyruvate, 1 mM L-glutamine, 1.0% penicillin-streptomycin (10,000 IU and 10,000 μg/mL, respectively), 0.5 μg/mL LH, and 0.5 μg/mL FSH. In vitro fertilization (IVF) was carried out in modified tris-buffered medium (mTBM) consisting of 113.1 mM NaCl, 3.0 mM KCl, 7.5 mM CaCl2¤2H2O, 20.0 mM Tris crystallized free base, 11.0 mM glucose, and 5.0 mM Na-pyruvate, supplemented with 2 mM caffeine and 4 mg/mL bovine serum albumin (BSA, fatty acid free, Fraction V). Embryo culture medium was from North Carolina State University (NCSU) 23 (IVC-PyrLac) supplemented with 4 mg/mL BSA, 0.17 mM Na-pyruvate, 2.73 mM Na-lactate, 20 μL/mL eagle amino acids in basal medium (BME), and 10 μL/mL nonessential amino acids in minimum essential medium (NEAA) for 2 days, and further cultured for 5 days in the same medium (NCSU-23, IVC-Glu) supplemented with 5.55 mM glucose instead of Na-pyruvate and Na-lactate.

Preparation of Donor Cells

Porcine female fetus was obtained via hysterectomy from pregnant gilt on day ∼30 of gestation. After removal of head, limbs, and visceral organs, the remaining tissues were washed in Dulbecco's phosphate buffered saline (DPBS) supplemented with 10% FBS and transferred into 0.05% (w/v) trypsin-elthylenediamine tetra acetic acid (trypsin-EDTA) solution for 5 min. Trypsinized cells were washed once in DMEM by centrifugation at 300g for 10 min to take cell pellet. FFs at a final concentration of 2 × 105 cell/mL were then cultured at 38.5°C in a humidified atmosphere of 5% CO2 in air.

Under sterile conditions, gelatinous bone marrow was extracted from a femur of ∼6-month-old female pig. MSCs were isolated according to the method of Ringe et al. (2002) with minor modifications. Briefly, 3–4 g of gelatinous bone marrow was resuspended in phosphate buffered saline (PBS) and dispersed mechanically and centrifuged at 400g for 10 min. Cells were resuspended and layered upon a Ficoll gradient (Amersham Biosciences, Uppsala, Sweden) and centrifuged at 400g for 40 min at 20°C. The interface buffy layer was collected and washed twice with PBS, and twice with ADMEM supplemented with 10% FBS and cultured at 38.5°C in a humidified atmosphere of 5% CO2 in air. Non-adherent cells were gently rinsed off 2 days after plating. Attached MSCs reached confluency in 12–14 days after plating (passage 0). Once confluent, cells were dissociated using 0.25% (w/v) trypsin-EDTA solution and pelleted at 300g for 5 min. Cells were then regrown and passaged for in vitro differentiation studies. FFs and MSCs were frozen in DMEM and ADMEM, respectively, containing 10% FBS and 10% dimethyl sulfoxide (DMSO), and stored in liquid nitrogen. For NT, cells at passage 3 were thawed, cultured to reach confluency to synchronize the majority into G0/G1 stage, and subsequently used.

In Vitro Differentiation of Porcine MSCs

Multilineage potential of porcine MSCs was evaluated by culturing first- to third-passage cells (2 × 105 cells/35 mm dish) in ADMEM containing 10% FBS under conditions conducive for osteogenic, adipogenic, and chondrogenic differentiation for 3 weeks adopting previously published protocols with minor modifications (Pittenger et al.,1999; Vacanti et al.,2005). The cells in the control group were cultured in ADMEM supplemented with 10% FBS. Cells were stained with BCIP/NBT (Promega, Madison, WI) to detect AP activity. The presence of mineralized matrix was analyzed using von Kossa stain. Cytoplasmic fat globules were identified by oil red O staining and mucosubstances by alcian blue 8GX solution.

Cell-Surface Antigen Profile

MSCs were characterized by cell-surface antigen profile using specific markers. The culture medium was removed and cells were washed twice with PBS before fixation in 3.7% paraformaldehyde in PBS for 10 min. Cells were rinsed three times with PBS and incubated in PBS solution containing 0.3% Triton X-100 and 4 mg/mL BSA for 20 min. Solution was removed and replaced by primary CD29, CD44, CD45, CD49b, CD105, and CD133 antibodies (Santa Cruz Biotechnology, Inc., CA) at 5 μg/mL in PBS containing 3 mg/mL BSA (PBS-3% BSA) for 30 min. After three PBS washes, samples were incubated for 30 min in fluorescein isothiocyanate (FITC) conjugated (Santa Cruz Biotechnology, Inc.) secondary antibodies solution (1/100) in PBS-3 mg/mL BSA. Genomic DNA was then labelled with propidium iodide (PI) solution in PBS for an additional 15 min before final washes and mounting in Vectashield medium (Vector Lab, Burlingame, CA). Samples were viewed using a fluorescence microscope at ×400 magnification (Nikon, Tokyo, Japan).

Oocyte Collection and In Vitro Maturation (IVM)

Porcine ovaries were obtained from prepubertal gilts at a local slaughterhouse and transported to the laboratory in PBS at 35–39°C. COCs were collected from antral follicles (3–6 mm diameter) with an 18-G needle attached to a 10-mL syringe. COCs were cultured (50 COCs/500 μL drop) in IVM medium with hormones for 22 h and further cultured for an additional 20 h in the fresh IVM medium without hormone supplements at 38.5°C in a humidified atmosphere of 5% CO2 in air.

After IVM, oocytes were freed off their cumulus cells by vortexing in DPBS medium supplemented with 0.1% (W/V) hyaluronidase for 1 min. Oocytes with a polar body (PB) and even cytoplasm were selected for the production of IVF and NT embryos.

In Vitro Fertilization (IVF)

Cumulus-free oocytes (20 oocytes/50-μL drops) were transferred into IVF media and inseminated with frozen-thawed sperm prepared by Percoll (Pharmacia, Uppsala, Sweden) density gradient. The final sperm concentration was adjusted to 1 × 105 sperm/mL. Coincubation was carried out at 38.5°C in a humidified atmosphere of 5% CO2 in air for 5 h.

In Vitro Culture (IVC)

Presumptive zygotes were cultured in NCSU-23 (IVC-PyrLac) for 2 days, and further cultured in the same medium (IVC-Glu) at 38.5°C in a humidified atmosphere of 5% O2, 5% CO2, and 90% N2 for 5 days. Cleavage and blastocyst rates were assessed on day 2 and day 7, respectively.

Nuclear Transfer (NT)

NT was carried out with minor modifications of previously described protocol (Kim et al.,2005). Denuded MII-stage oocytes were enucleated by micromanipulation technique in Hepes-buffered M-199 supplemented with 10% FBS, 7.5 μg/mL cytochalasin B (CCB), and 12 mM sorbitol. Briefly, the first polar body and metaphase plate with a small volume of cytoplasm were removed together using a 15-μm beveled micropipette. Single donor cell (FF or MSC) of approximately 10-μm diameter was used for NT. For fusion, the reconstructed eggs were oriented in BTX Electro chamber (BTX, Inc., San Diego, CA) filled with 0.28 M mannitol solution containing 0.1 mM MgSO4, 0.05 mM CaCl2, and 0.1 mg/mL BSA and pulsed twice with 2.0 KV/cm DC for 30 μsec using a BTX Electro Square Porator (ECM 830, BTX, Inc., San Diego, CA). After fusion, eggs were cultured in 50-μL drops of NCSU-23 (IVC-PyrLac) medium supplemented with 7.5 μg/mL CCB at 38.5°C in a humidified atmosphere of 5% CO2 in air for 3 h. NT embryos were cultured as described above.

For gene analysis, IVF and NT embryos at 4-cell, 8-cell, morula, and blastocyst stages were collected at 48, 72, 120, and 168 hr, respectively.

Cytological Analysis

To count total cell number, day-7 IVF and NT blastocysts fixed in methanol-acetic acid (3:1) overnight were stained with 10 μg/mL bisbenzimide (Hoechst 33342) in HEPES-TALP for 10 min and the nuclei were counted under an epifluorescence microscope (Nikon).

The apoptosis of embryos was detected by the Terminal Deoxynucleotidyl Transferase-Mediated dUTP Nick-End Labeling (TUNEL) method using the in situ cell death detection kit (Roche, Penzberg, Germany) following the manufacturer's protocol. Briefly, blastocysts were fixed in 3.7% formaldehyde for 4 hr at room temperature (rt), washed in PBS, permeabilized by incubation in 0.5% Triton X-100 for 1 hr, washed in PBS, incubated with fluorescein-conjugated dUTP and terminal deoxynucleotidyl transferase enzyme for 1 hr at 37°C in dark room, and counterstained with 40 μg/mL PI for 1 hr at 37°C after treatment of 50 μg/mL RNase at rt for 1 hr. Samples were examined under an epifluorescence microscope and those stained red were nucleus and those stained green and yellow were apoptotic body.

Collection of In Vivo Embryos

In vivo embryos for gene expression analysis were recovered from eight natural estrus gilts that were artificially inseminated (AI) twice on the day following onset of estrus. Gilts were slaughtered, reproductive tracts recovered, and embryos were collected by retrograde uterine flushing with 50 mL of PBS supplemented with 10% FBS. Embryos at 4-cell, 8-cell, morula, and blastocyst stages were collected at 36–42, 48–54, 90–96, and 140–146 hr post AI, respectively.

RNA Extraction and Reverse Transcription (RT)

The set of genes analyzed in the present study characterizes important functions including transcription and pluripotency (POU domain transcription factor Oct4, signal transducer and activator of transcription Stat3, homeobox containing transcription factor Nanog), DNA methylation (DNA methyltransferase 1 Dnmt1, DNA methyltransferase 3 alpha Dnmt3a), histone deacetylation (histone deacetylase 2 Hdac2), growth factor signaling and imprinting (insulin-like growth factor 2 Igf2, insulin-like growth factor 2 receptor Igf2r), and apoptosis (pro-apoptotic Bax, anti-apoptotic Bcl2) regulation.

Total RNA was extracted from pools (triplicates) of 5 embryos each at 4-cell, 8-cell, morulae, and blastocyst stages from NT, in vivo and in vitro using Dynabeads® mRNA DIRECT™ kit (Dynal, Oslo, Norway). Briefly, samples were lysed in 50 μL lysis/binding buffer (Dynal). After vortexing and brief centrifugation, the samples were incubated at rt for 10 min. To each sample, 100 μL prewashed Dynabeads oligo (dT)20 was added and after 5 min of hybridization, the beads were separated from the binding buffer using the Dynal magnetic separator. The beads were then washed in buffer A and B (Dynal), and the poly (A) RNA was eluted from the beads by adding 11 μL of tris-HCl. A total of 1.0 μg RNA was converted to cDNA using oligo (dT)20 primers and superscript™ III reverse transcriptase (Invitrogen Corporation, La Jolla, CA) in a 20-μL reaction for 50 min at 50°C. PCR was performed with 1–2 μL of cDNA in 100 μL using hotstart taq DNA polymerase (Qiagen GmbH, Germany) to confirm product size. Amplification consisted of 25 or 30 cycles of 94°C for 30 sec, 60°C for 30 sec, and 72°C for 60 sec performed in a PTC-200 peltier thermal cycler (MJ Research, Inc.).

Real-Time RT-PCR

Real-time PCR was performed on a LightCycler® using FastStart DNA Master SYBR Green I (Roche Diagnostics, Mannheim, Germany) containing MgCl2, dNTP, and FastStart Taq DNA polymerase. Primer sequences, the size of amplified products, and the GenBank accession numbers are shown in Table 2. DNase I treated cDNA (2.0 μL) from embryos was added to a total volume of 20 μL containing 2 μL SYBR Green mix and 0.3 mM each of forward and reverse porcine specific primers. The program used for all genes consisted of a denaturing cycle of 10 min at 95°C; 45–50 cycles of PCR (95°C for 10 sec, 57°C for 5 sec, and 72°C for 10 sec); a melting cycle consisting of 95°C for 0 sec, 70°C for 15 sec, and a step cycle starting at 70°C until 95°C with a 0.1°C/sec transition rate; and, finally, a cooling cycle of 40°C for 30 sec. The quantification of gene transcripts such as Oct4, Stat3, Nanog, Dnmt1, Dnmt3a, Hdac2, Igf2, Igf2r, Bax, and Bcl2 was carried out in 3 replicates. Histone (H2a) was amplified for every sample to confirm the presence of RNA as a housekeeping gene for normalization. Melt curve analysis was conducted to confirm the specificity of each product. As negative controls, tubes were always prepared in which RNA or reverse transcriptase was omitted during the real time RT-reaction.

Table 2. Sequence-Specific Primers Used for Real-Time RT-PCR
GenePrimer sequence (5′–3′)Length(bp)GenBank accession no.

Statistical Analysis

Differences were analyzed among treatments using one-way ANOVA after arc-sine transformation of the proportional data. Data were expressed as mean ± SEM. Comparisons of mean values among treatments were performed using Duncan's and Tukey's multiple comparisons test. Differences were considered significant at P < 0.05.