**3. Embryonic germ cells**

Embryonic germ cells (EGCs) are pluripotent cells derived *in vitro* from primordial germ cells (PGCs). They were first derived in the mouse by Matsui et al. [16] and Resnick et al. [17]. Mouse EGCs are morphologically similar to mouse ESCs, and express similar pluripotency markers such as AP, SSEA1, Oct4, Sox2, and Nanog. The differentiation potential of these cells is also similar to that of mouse ESCs, as evidenced by their ability to differentiate *in vitro* into different tissues from the three primary germ layers [18], as well as to re-enter the germline when injected into blastocyst embryos [19]. Mouse EGCs may thus be considered equivalent to ESCs.

Similar to the conditions required to derive mouse ESCs, mouse EGCs need two important growth factors in addition to leukemia inhibitory factor (LIF) for their successful derivation from PGCs. The first is stem cell factor (SCF), a ligand for c-Kit, which supports PGC survival *in vivo* and *in vitro* by suppressing apoptosis [20]. The second is bFGF, which plays a major role in the reprogramming of mouse PGCs into pluripotent cells *in vitro* [21]. Although the mechanism of this reprogramming is still poorly understood, bFGF is known to downregulate the expression of basic lymphocyte maturation protein 1 (Blimp1) *in vitro*, which in turn causes the upregulation of c-Myc and Klf4 [21]. Because mouse PGCs also express Oct4 and Sox2, a similar mechanism may be involved in their reprogramming to ESCs.

Porcine EGCs were first derived by Shim et al. [22], who used embryos from domestic breeds on day 25 of gestation under conditions similar to those used to culture mouse EGCs. The established cell lines had an ESC-like morphology, expressed AP, and were able to differentiate into different cell types *in vitro* and *in vivo*. Other groups have derived porcine EGC lines using day 25-28 embryos from domestic breeds [23] [24] [25] [26] [27], and from the Chinese minipig [28]. Recently, putative porcine EGC lines derived from the PGCs of day 20-24 embryos from Danish Landrace crosses and Yucatan minipigs were reported [29]. Notably, these findings suggest that EGC lines can be established from any porcine breed, unlike mouse ESCs, which are restricted to certain strains.

Immunocytochemical analysis of pluripotency marker expression showed that, in addition to AP activity, EGCs express Oct4, SSEA1, SSEA3, and SSEA4 at variable levels. Analysis of the porcine EGC gene expression suggested that they are similar to human EGCs in expressing AP, Oct4, SSEA1, and SSEA4 [30]. The pig EGCs typically have long cell cycles, and proliferate slowly over many passages.

The tissue culture conditions used by all groups for the derivation and propagation of porcine EGCs are similar to those used for human and mouse ESCs. The cells are grown on feeder layers of mitotically inactivated embryonic mouse fibroblasts, most often on immortalized mouse fibroblast cells from a cell line known as STO. The initial establishment of cell lines is hampered by the fact that large numbers of porcine PGCs undergo apoptosis and die in culture within 6 hours of their incubation [25]. However, by using protease inhibitors and antioxidants to reduce the level of PGC apoptosis *in vitro*, Lee et al. were able to increase the number of EGC colonies in the primary culture. In addition, alpha 2-macroglobulin together with N-acetyl‐ cystein and butylated hydroxyanisole (both antioxidants) increased the number of AP-positive colonies in primary cultures at least twofold. Another approach is to use growth factors and various feeder cells to increase the survival and prevent the differentiation of PGCs in longterm culture. Lee et al. [26] showed that supplementing the culture medium with three growth factors (LIF, SCF, and bFGF) increased the number of colonies obtained in primary culture, and improved the quality of the colonies in subsequent passages compared to cells treated with only two of these growth factors. The use of feeder layers other than STO cells did not have a significant effect on PGC survival or on the quality of the derived EGCs. In another study, membrane-bound SCF and soluble LIF were sufficient to increase the number of surviving PGCs on days 3 and 5 of primary culture, while adding bFGF did not affect the results significantly [23]. It is possible that the membrane-bound form of SCF is more potent than the soluble form, which could explain these differing results. However, it has also been reported that established PGCs are capable of generating chimeras without added growth factors, while adding LIF to the culture does not improve the efficiency of establishing EGC lines [22]. Thus, further study is needed to elucidate the effects of these supplements in culture.

We have yet to observe *bona fide* ESCs produced from the pig. Nevertheless, continued research toward this goal is important, as the pig is a particularly useful biomedical model for studying human disease, and ESCs are a unique cell type that is especially useful for studying human

Embryonic germ cells (EGCs) are pluripotent cells derived *in vitro* from primordial germ cells (PGCs). They were first derived in the mouse by Matsui et al. [16] and Resnick et al. [17]. Mouse EGCs are morphologically similar to mouse ESCs, and express similar pluripotency markers such as AP, SSEA1, Oct4, Sox2, and Nanog. The differentiation potential of these cells is also similar to that of mouse ESCs, as evidenced by their ability to differentiate *in vitro* into different tissues from the three primary germ layers [18], as well as to re-enter the germline when injected into blastocyst embryos [19]. Mouse EGCs may thus be considered equivalent to ESCs.

Similar to the conditions required to derive mouse ESCs, mouse EGCs need two important growth factors in addition to leukemia inhibitory factor (LIF) for their successful derivation from PGCs. The first is stem cell factor (SCF), a ligand for c-Kit, which supports PGC survival *in vivo* and *in vitro* by suppressing apoptosis [20]. The second is bFGF, which plays a major role in the reprogramming of mouse PGCs into pluripotent cells *in vitro* [21]. Although the mechanism of this reprogramming is still poorly understood, bFGF is known to downregulate the expression of basic lymphocyte maturation protein 1 (Blimp1) *in vitro*, which in turn causes the upregulation of c-Myc and Klf4 [21]. Because mouse PGCs also express Oct4 and Sox2, a

Porcine EGCs were first derived by Shim et al. [22], who used embryos from domestic breeds on day 25 of gestation under conditions similar to those used to culture mouse EGCs. The established cell lines had an ESC-like morphology, expressed AP, and were able to differentiate into different cell types *in vitro* and *in vivo*. Other groups have derived porcine EGC lines using day 25-28 embryos from domestic breeds [23] [24] [25] [26] [27], and from the Chinese minipig [28]. Recently, putative porcine EGC lines derived from the PGCs of day 20-24 embryos from Danish Landrace crosses and Yucatan minipigs were reported [29]. Notably, these findings suggest that EGC lines can be established from any porcine breed, unlike mouse ESCs, which

Immunocytochemical analysis of pluripotency marker expression showed that, in addition to AP activity, EGCs express Oct4, SSEA1, SSEA3, and SSEA4 at variable levels. Analysis of the porcine EGC gene expression suggested that they are similar to human EGCs in expressing AP, Oct4, SSEA1, and SSEA4 [30]. The pig EGCs typically have long cell cycles, and proliferate

The tissue culture conditions used by all groups for the derivation and propagation of porcine EGCs are similar to those used for human and mouse ESCs. The cells are grown on feeder layers of mitotically inactivated embryonic mouse fibroblasts, most often on immortalized

similar mechanism may be involved in their reprogramming to ESCs.

disease.

**3. Embryonic germ cells**

200 Pluripotent Stem Cell Biology - Advances in Mechanisms, Methods and Models

are restricted to certain strains.

slowly over many passages.

Pluripotent stem cells are able to form EBs, spherical aggregates containing differentiated cells, when cultured in suspension. Porcine EGCs cultured in "hanging drops" can also form simple EBs, consisting of large epithelial-like cells on the periphery surrounding mesenchy‐ mal-like cells in the center. When allowed to attach to gelatin-treated plastic dishes, the EB cells proliferate and spread on the surface of the dish, giving rise to several different types of cells [27].

Pluripotent cells can proliferate and differentiate *in vivo*, forming tumors containing differen‐ tiated tissues called teratomas, when injected into immunodeficient mice. To date, only one group has reported teratoma testing for porcine EGCs, using cell lines derived from the Chinese minipig [28]. The authors reported that the teratomas contained cells from the three primary germ layers: epithelial, neuroepithelial, and adipose tissue. In contrast to these results, two studies showed that the injection of human EGCs into immunodeficient mice failed to generate teratomas. [31] [32]. However, more recently, the formation of teratomas from human EGCs cultured under serum-free conditions was reported [33].

Another way of testing the differentiation potential of pluripotent cells *in vivo* is by chimera formation, in which the cells injected into early embryos contribute to the three germ layers and potentially to the germline. Unlike mouse ESCs and EGCs, which have been shown to reenter the germline of chimeras, pig EGC chimeras have not displayed a proven germline contribution; furthermore, the somatic tissues of these porcine chimeras contained only a low percentage of donor-derived cells [22] [34] [24]. Other researchers demonstrated a similarly low chimeric contribution after the injection of somatic cells into sheep blastocysts and 8-cell mouse embryos [35, 36]. This is troubling because germline contribution is considered to be the ultimate proof of pluripotency, and has been demonstrated only in the mouse and chicken to date. Human ESCs are presumed to be true stem cells as well, even though their germline potential has not been tested for ethical reasons. Caution is needed when interpreting chimera experimental results. Proper controls are necessary to distinguish the "true" stem cells from somatic cells that can be integrated into the embryos after partial reprogramming by the surrounding embryonic cells.

human and mouse iPSCs, three different groups reported the establishment of iPSCs in the pig. One of these groups generated porcine iPSCs using fetal fibroblasts from Danish Landrace pigs, by the lentiviral transduction of six human transcription factors (OCT4, SOX2, c-Myc, KLF4, LIN28, and NANOG) under the control of a doxycycline-inducible promoter [41]. Simultaneously, another Chinese group reported the derivation of iPSCs from fibroblasts from the Tibetan minipig, using constitutively expressed lentiviral vectors carrying the mouse cDNA sequences of Oct4, Sox2, c-myc, and Klf4 [42]. The third report published shortly thereafter by Ezashi et al. [43] also described the production of porcine iPSCs through the use of 4 human transcription factors (OCT4, SOX2, c-Myc, and KLF4). In addition, these researchers claimed that the porcine iPSCs were positive for SSEA-1 but negative for SSEA3 and SSEA-4,

The Minipig — A New Tool in Stem Cell Research

http://dx.doi.org/10.5772/57603

203

Transcriptional profiling of the cell line by Affymetrix microarray confirmed that the cells were indeed reprogrammed, and expressed a variety of ESC markers endogenously. However, the continued expression of the exogenous transcription factors was detected in the iPSCs generated by all three groups. This problem is not unique to the pig, as it has been reported in other species as well [39]. In any case, the continued expression of pluripotency genes did not pose any problems for the differentiation of the cells, since all three laboratories demonstrated that their cell lines were able to differentiate *in vitro* (including EB formation) and in vivo by forming teratomas containing cells of all three germ layers. More recently, it was demonstrated that porcine iPSCs can form chimeras with high efficiency (85.3%), and contribute to all three germ layers [44]. Because germline chimerism was not confirmed in this study, it remains to

One of the major advantages iPSCs may offer in the future is the potential for custom derivation of pluripotent cells from individual patients to use for regenerative therapies without the risk of immune rejection. However, since the epigenetic reprogramming necessary to produce them requires prolonged expression of the transgenes (2-3 weeks), most of the iPSCs described to date have been generated with the use of lentiviral vectors that integrate known oncogenes (such as c-myc and Klf4) into the cell genome. The dangers of using these genes became evident in a study in which mouse chimeras generated with iPSCs developed tumors following reactivation of the initially silenced transgene c-myc [38]. In the only report to date in which porcine chimeras were produced from iPSCs [44], no tumor formation was reported in the newborn piglets. However, because long-term testing and monitoring were not conducted prior to the publication of these results, at present there is no guarantee that these iPSCs would

To avoid problems associated with viral integration and the use of oncogenes, many groups have developed strategies for iPSC production in which the disadvantages discussed above have been minimized. For example, it has been shown that reprogramming can be achieved in both human and mouse cells without the use of c-myc, albeit at the expense of efficiency [45]. In addition, the use of certain transcription factors can be omitted when using cell types already expressing them; for example, neural progenitor cells, which already express endogenous Sox2 and c-myc were reprogrammed using only the induced expression of OCT4 and Klf4 [46] [47].

indicating that they were more similar to mouse than human iPSCs.

be determined whether the porcine iPSCs are fully equivalent to mouse iPSCs.

be safe for clinical applications.

The reasons for the limited pluripotency of the porcine EGCs are currently unclear. It is possible that the tissue culture conditions are not sufficient for maintaining the pluripotency of the PGCs *in vitro*, similar to the problems that exist in the cultures of inner cell mass and epiblast cells. Another possibility is that the pig PGCs do not undergo full epigenetic reprogramming into pluripotency like the mouse PGCs do. One indication that there might be differences in the biology of PGCs between the pig and mouse is that porcine PGCs survive and proliferate *in vitro* in the absence of externally added growth factors, regardless of whether they are cultured in serum-supplemented or serum-free conditions, while mouse PGCs fail to survive in the absence of any of the three growth factors LIF, SCF, or bFGF. In addition, mouse PGCs have not been shown to form chimeras, whereas porcine PGCs can contribute to somatic tissues after their injection into early blastocysts [24]. Thus, some of the molecular mechanisms that are important for reprogramming mouse PGCs to pluripotent EGCs in culture may be different in the pig.
