Direct conversion of fibroblasts traces the way back to our first organ—the placenta

Cell fate determination has always fascinated scientists predominantly embryologists and challenged them to decipher master genes or networks of transcriptional factors which together with cell signaling cascades act as key regulators for the cell lineage allocation. With the rediscovery of the pluripotent precursor cells in an adult body and the possibility to induce pluripotency in somatic cells (iPSCs), as sources for alternative therapeutic options in translational medicine, this field has attracted high interest in medical research and application (1). iPSC have several advantages over embryonic stem cells such as avoiding tumor formation and to circumvent ethical concerns. Meanwhile the understanding of pluripotency has broadened thanks to abundant studies on converting ESC and somatic cells into another cell fate using predominantly transfection of a bunch of lineage specific transcription factors. While many studies focused on the regulation of ESC conversion less is known about the hierarchically organized network of transcriptional factors which determine the trophoblast lineage for placenta development. The first and drastic lineage segregation to occur is the differentiation of the blastocyst into the inner cell mass (ICM) which give rise to the embryo and the trophectoderm (TE). The trophectodermal lineage establishes the placenta for a nutrition route to the fetus which guarantees the development the future embryonic life (2). After fertilization a series of cell divisions lead to eight seemingly identical cells, called blastomeres, which are defined as totipotent because of being capable to giving rise to both lineages either to ICM or TE. Once commitment to TE or ICM occurs, however, it is generally considered irreversible (3) and a stable genetic barrier between the embryonic and extraembryonic lineage is established.

Like embryonic stem cells (ESC) derived from the ICM of the blastocysts, trophoblast stem cells (TSC) can be established from the blastocysts or the postimplantation ectoplacental cone of the mouse characterized by self-renewal and the possibilities to differentiate into the trophoblast subpopulations of placental tissue with the support of special culture treatment (4). With the help of a number of targeted deletions in early mouse embryos a set of transcription factors, including Cdx2, Eomes, Elf5, Ets2, Gata3, Tcfap2c, Esrrb, Sox2 and Tead4, have been identified to be critical for the establishment and/or maintenance of the multipotent state of TSCs with Tead4 on the top of this hierarchically ordered network (5).

Several attempts have been made to overcome the established cell lineages from ESC into TSC by converting ESC directly with a cocktail of transcription factors mentioned above, but these iTSCs stop at an intermediate state and are not completely transformed into a stable TS cell phenotype and differ in respect to incomplete reprogramming and a distinct epigenetic memory. In a previous study Cambuli et al. (6) could give evidence that this incomplete reprogramming is due to a group of TSC genes, so called gate keeper genes, such as Elf5, Tead4 and Hand1 which did not change their methylation profile after conversion. To achieve their function, these lineage genes need to be hypomethylated, but even the experimentally induced rapid demethylation approaches failed because of a rapid re-methylation in the culture medium. The authors suspected that the epigenetic memory is retained as ES cell memory and represent the key to stabilize the barrier between this two cell lineages. Thus not only the appropriate set of transcription factors but the epigenetic marks which are laid down during lineage allocation must be deleted for complete reprogramming. The present commentary focus on the study of Kubaczka et al. recently published in Cell Stem Cell, 2015 (7) which further corroborated and extended these findings by using a bright strategy. Instead using ESC, the group of Schorle started from mouse embryonic fibroblasts (MEFs) and postnatal tail tip fibroblast and converted these somatic cells into TSC. This direct conversion approach from somatic cells without a pluripotent intermediate state has been already successfully performed by Tanabe et al. transforming fibroblasts into neurons (8) and recently by Benchetrit et al. (9) into iTSC.

Kubaczka et al. (7) started their strategy by transforming MEFs cultured in TS cell medium using a transfection cocktail from 12 candidate genes for trophoblast conversion and induced these genes for 10 days. This approach resulted in iTSC colonies expressing Cdx2, Elf5, Eomes and Tfap2c, protein and mRNA, in a similar level as TSCs combined with the disappearance of the fibroblast specific Fsp1 and the lack of ESC markers such as Pou5f1(Oct4) and Nanog.

The authors improved their experimental approach elegantly and narrowed down their candidate genes by selecting clones which resembled most likely TSC but had only four integrated candidate genes, Eomes, Tfap2c, Ets2 and Gata3.These four transcription factors were able to induce endogenous Cdx2 and Elf5 in iTSC. Moreover, these clones re-expressed epigenetically regulated gatekeeper genes such as Tead4, Cdx2, Eomes Tfap2c, Gata3, ETs2 and Elf5 in a pattern indistinguishable from non-induced TSC. This gene independency was confirmed by gene array expression profiles which reveal that iTSC and control TS-EGFP shared the gene clustering pattern in great quantities but was distinct from the parental MEFs. In addition, these clones regain self-renewal capacity similar to parental TSC. The use of a previously published new serum free defined TSC culture medium (10) seem to improve this conversion and iTSCs gained gene stability for 40–50 passages. A further reduction of the four transcription factors, however, did not induce reprogramming of MEFs into iTSC which indicated that a set of interacting transcription factors is needed for TSC induction already at the beginning of lineage allocation. Using the same experimental approach they successfully converted fibroblasts from the tail of newborn mice with less pluripotency state compared to MEFs, but failed to generate stable iTSC from ESCs.

The appropriate functionality of these iTSCs was tested according to well established tests: iTCSs were able to differentiate along their lineages in vitro as indicated by marker gene profiles, they contributed to the placenta when injected into blastocysts and formed the typical transient tumors with hemorrhagic lesions under the skin of nude mice (11). These iTSC tumors gave proof that the iTSC were able to differentiate into a specific type of giant cells, the spiral arteries lining trophoblast (SpA-TGC), which is able to invade into host vessels.

This study of Kubaczka et al. (7) gave further evidence that Cdx2 does not set up the TE lineage but has the role to keep the undifferentiated state of TSC by suppressing Oct4. In fact Tfaf2c and Eomes bind to Tead4, Elf5 and Hand1 promoter which put them both into the upper hierarchical position. Moreover, unexpectedly Elf5 does not belong to the four lineage determiners but instead Ets2.

The authors asked further, why this trans-differentiation into TSC is more effective starting from somatic cells than from ESC and suggested that the recovery from the epigenetic signature is one of the most important mechanisms. Kubaczka et al. (7) found that the global methylation pattern in all iTSC clones resembled each other but distinguishes from parental MEF pattern. However to note, MEFs were generally less methylated than ESCs and epigenetic signature of MEFS was closer to iTSC than to ESC. Moreover, all MEF derived iTSC display a more authentic epigenetic signature when compared to ESC based reprogrammed TSCs.

In MEFs derived iTSC all important lineage genes which are repressed by methylation in ESC were demethylated but Oct4 remained unchanged. Moreover, hypomethylated CG islands in MEFs, which are enriched in Homebobox genes linked to the embryonic lineage, were re-methylated in TSC. Thus the resetting of the epigenetic memory seems to be more uncomplicated from MEFS into iTSC than from ESC into iTSC.

These new strategies may offer important advantages with respect to a direct conversion of somatic cells as an excellent tool to detect the appropriate hierarchical organization of the molecular network for trophoblast lineage differentiation as the first embryonic organ for establishing and to maintain pregnancy. Furthermore it gives the chance to decipher the most important methylation signatures to keep the gate between TE and ICM.

Because placental insufficiency is the most common identifiable cause of intrauterine growth restriction, this study represents a knowledge base to establish human iTSC, which would provide a tool to unravel the mechanisms of human placental pathogenesis in a trophoblast in vitro system.

While it is possible to establish human embryonic stem cells, reports on stable human TSCs are hampered by the problem to compare them with appropriate controls and test systems to prove convincingly their trophoblast character.

To date TS like cells have been already derived from conventional hESCs (12), from human blastomeres ESC (13) or from human fibroblastic iPSC (14) but nothing is reported of long term gene independency as well as of a conversion of the complete methylation signature or the methylations status of defined key genes such as Elf5.

It is worthwhile now to re-investigate the possibility to establish stable hTSC by the use of different human fibroblast such as human foreskin fibroblasts, or placental fibroblast with the strategy described by Kubaczka et al. (7). In particular, it has been recently shown that mesenchymal stem cells (MSC) from bone marrow-and adipose tissue-derived MSCs represented an optimal stem cell source for tissue engineering and regenerative medicine (15).

To summarize the discovery presented in the study of Kubaczka et al. (7) demonstrated that a small combination of TSC specific transcription factors can directly re-program somatic cells, here fibroblasts, and recover not only the appropriate genetic but also the epigenetic signature of TSC. These findings support the view that these lineage barriers can be overcome easier than assumed when the optimal somatic cell in combination with the appropriate set of transcription factors is employed.

Acknowledgements

Funding: This research was supported by DFG grant Wi 774/27-1.

Footnote

Provenance: This is a Guest Commentary commissioned by Editor-in-Chief Zhizhuang Joe Zhao (Pathology Graduate Program, University of Oklahoma Health Sciences Center, Oklahoma City, USA).

Conflicts of Interest: The author has no conflicts of interest to declare.

Comment on: Kubaczka C, Senner CE, Cierlitza M, et al. Direct induction of trophoblast stem cells from murine fibroblasts. Cell Stem Cell 2015;17:557-68.

References

1. Takahashi K, Yamanaka S. A decade of transcription factor-mediated reprogramming to pluripotency. Nat Rev Mol Cell Biol 2016;17:183-93. [Crossref] [PubMed]
2. Maltepe E, Bakardjiev AI, Fisher SJ. The placenta: transcriptional, epigenetic, and physiological integration during development. J Clin Invest 2010;120:1016-25. [Crossref] [PubMed]
3. Dyce J, George M, Goodall H, et al. Do trophectoderm and inner cell mass cells in the mouse blastocyst maintain discrete lineages? Development 1987;100:685-98. [PubMed]
4. Tanaka S, Kunath T, Hadjantonakis AK, et al. Promotion of trophoblast stem cell proliferation by FGF4. Science 1998;282:2072-5. [Crossref] [PubMed]
5. Latos PA, Hemberger M. Review: the transcriptional and signalling networks of mouse trophoblast stem cells. Placenta 2014;35 Suppl:S81-5. [Crossref] [PubMed]
6. Cambuli F, Murray A, Dean W, et al. Epigenetic memory of the first cell fate decision prevents complete ES cell reprogramming into trophoblast. Nat Commun 2014;5:5538. [Crossref] [PubMed]
7. Kubaczka C, Senner CE, Cierlitza M, et al. Direct Induction of Trophoblast Stem Cells from Murine Fibroblasts. Cell Stem Cell 2015;17:557-68. [Crossref] [PubMed]
8. Tanabe K, Haag D, Wernig M. Direct somatic lineage conversion. Philos Trans R Soc Lond B Biol Sci 2015;370:20140368. [Crossref] [PubMed]
9. Benchetrit H, Herman S, van Wietmarschen N, et al. Extensive Nuclear Reprogramming Underlies Lineage Conversion into Functional Trophoblast Stem-like Cells. Cell Stem Cell 2015;17:543-56. [Crossref] [PubMed]
10. Kubaczka C, Senner C, Araúzo-Bravo MJ, et al. Derivation and maintenance of murine trophoblast stem cells under defined conditions. Stem Cell Reports 2014;2:232-42. [Crossref] [PubMed]
11. Kibschull M, Nassiry M, Dunk C, et al. Connexin31-deficient trophoblast stem cells: a model to analyze the role of gap junction communication in mouse placental development. Dev Biol 2004;273:63-75. [Crossref] [PubMed]
12. Chung Y, Klimanskaya I, Becker S, et al. Human embryonic stem cell lines generated without embryo destruction. Cell Stem Cell 2008;2:113-7. [Crossref] [PubMed]
13. Zdravkovic T, Nazor KL, Larocque N, et al. Human stem cells from single blastomeres reveal pathways of embryonic or trophoblast fate specification. Development 2015;142:4010-25. [Crossref] [PubMed]
14. Chen MJ, Lu Y, Hamazaki T, et al. Reprogramming adipose tissue-derived mesenchymal stem cells into pluripotent stem cells by a mutant adeno-associated viral vector. Hum Gene Ther Methods 2014;25:72-82. [Crossref] [PubMed]
15. Heo JS, Choi Y, Kim HS, et al. Comparison of molecular profiles of human mesenchymal stem cells derived from bone marrow, umbilical cord blood, placenta and adipose tissue. Int J Mol Med 2016;37:115-25. [PubMed]