In the journal Nature, a fascinating paper appeared from the laboratory of Samuel Pfaff at the Howard Hughes Medical Institute at the Salk Institute for Biological Studies in La Jolla, California near San Diego. In this paper, Todd Macfarlan and his colleagues show that embryonic stem cells cycle between a very primitive developmental stage and a later stage. This cycling is also due to gene expression that is linked to transposable DNA elements.
First, we need some background. The term “totipotent” means that a cell can form any structure in the embryonic or adult body. For example, when the egg undergoes fertilization, it becomes a zygote, which has the capacity to grow into the embryo and all the extraembryonic membranes (amnion, chorion, allantois, placenta, and so on). Another example is a sponge. When a small piece of the sponge is cut from it, the cells in that small piece can de-differentiate and grow into an entire new sponge. Sponge cells are, therefore, totipotent.
Secondly, there is a term “pluripotent,” which means that the cells can form all the adult cell types. Embryonic stem cells are generally thought to be pluripotent and not totipotent. Once the embryo forms the two-cell stage, these two blastomeres are totipotent. However, when the blastocyst stage forms, the inner cell mass cells become pluripotent and lose the ability to form the placenta.
Many years ago, Beddington and Robertson, (1989) implanted mouse embryonic stem cells into the outer layer of cells (trophoblast) to determine if the inner cell mass cells could form the placenta (Development 105, 733–737). The embryonic stem cells were incorporated into the placenta at a very low rate. These data led to an intriguing question: Was the low incorporation due to contamination with trophoblast cells, or could a small proportion of the embryonic stem cells actually become placenta? When gene expression studies examined embryonic stem cells, gene expression was stable in the majority of the cells, but unstable in a small minority of cells (a condition called metastable). It was not surprising that embryonic stem cells were a mixed bag of different cells, but some cells expressed genes that were normally found at earlier developmental stages or were normally expressed in cells that make placenta:
1. Niakan, K. K. et al. Sox17 promotes differentiation in mouse embryonic stem cells by directly regulating extraembryonic gene expression and indirectly antagonizing self-renewal. Genes Dev. 24, 312–326 (2010).
2. Hayashi, K., Lopes, S. M., Tang, F. & Surani, M. A. Dynamic equilibrium and heterogeneity of mouse pluripotent stem cells with distinct functional and epigenetic states. Cell Stem Cell 3, 391–401 (2008).
3. Singh, A. M., Hamazaki, T., Hankowski, K. E. & Terada, N. A heterogeneous expression pattern for Nanog in embryonic stem cells. Stem Cells 25, 2534–2542 (2007).
4. Chambers, I. et al. Nanog safeguards pluripotency and mediates germline development. Nature 450, 1230–1234 (2007)
5. Zalzman, M. et al. Zscan4 regulates telomere elongation and genomic stability in ES cells. Nature 464, 858–863 (2010).
Into the fray comes Macfarlan and company to save (or explain) the day. It turns out that our genomes are loaded with DNA from transposable elements. These DNA elements either have or had at one time, the ability to jump from one location in the genome to another. There are large numbers of these transposable elements in our genomes and almost 50% of the human genome is composed of the remains of these elements. Current transposable elements include Long INterspersed Elements (LINEs), Short INterspersed Elements (SINEs) and SVA (SINE/VNTR/Alu) elements. Others include elements such as Mariner, MIR, HERV-K, and others. The significance of all this is that during development, when the embryo gets to the two-cell stages, in the mouse, particular transposable elements are expressed at very high levels (they produce 3% of the transcribed messenger RNAs, see Peaston, A. E. et al., Dev. Cell 7, 597–606 (2004); Evsikov, A. V. et al., Cytogenet. Genome Res. 105, 240–250 (2004); Kigami, D., et al., Biol. Reprod. 68, 651–654 (2003)), and after two-cell stage, the expression of these transposable elements is silenced (Svoboda, P. et al. Dev. Biol. 269, 276–285 (2004); Ribet, D. et al. J. Virol. 82, 1622–1625 (2008)).
Since these transposons are characteristic of gene expression at the two-cell stage, they can be used as a marker for embryonic stem cells that have reverted back to the two-cell stage. MacFarlan and his co-workers made a reporter gene and placed it into embryonic stem cells that was controlled by the same sequences as the transposons that are activated at the two-cell stage. After growing these engineered embryonic stem cells in culture, they discovered that a small minority of cells expressed this reporter gene.
Did these reporter-expressing cells have characteristics like unto those of the two-cell stage embryos? Yes they did. When Macfarlan and his buddies examined the genes expressed in the cells that expressed the reporter, they found that the traditional genes that are so characteristic of inner cell mass cells (Oct4, Nanog, Sox2, etc.) were not expressed and other genes normally expressed in two-cell stage embryos, such as Zscan4, were expressed. Other features that are found in two-cell-stage embryos were also found in these cells that expressed the reporter gene. (methylation of histone 3 lysine 4 (H3K4) and acetylation of H3 and H4 for those who are interested).
Finally, the reporter-expressing cells were able to contribute to the formation of the placenta when transplanted into a mouse embryo. This shows that these cells not only express the genes of the totipotent stage of development, but they also are totipotent.
These experiments show that most, maybe all embryonic stem cells pass through a short-lived state during which they display features that are characteristic of the totipotent two-cell stage: unlike the vast majority of the ES cells in the culture. During this transition, they lack expression of the pluripotency-promoting proteins Oct4, Sox2 and Nanog, and have the ability to form cells of both the placenta and the fetus.
There is also a moral implication of these experiments. In his book Challenging Nature and on the book’s web site, Lee Silver argues that embryonic stem cells are essentially embryos, and if we don’t object to using and discarding embryonic stem cells, then we should not have any problem with using and discarding embryos. His reasons for asserting that embryonic stem cells are embryos is that in the mouse, embryonic stem cells can be inserted into the inner cell mass of an embryo that has four copies of each chromosome. The tetraploid embryos can form the placenta, but they cannot form the embryo that is attached to the embryo. Inserting embryonic stem cells into the inner cell mass of these embryos will rescue them from dying because the embryonic stem cells with make the embryo and the tetraploid embryos will form the placenta. This experiment is called “tetraploid rescue” and Silver uses it to argue that embryonic cells are essentially embryos.
I find this argument unconvincing for several reasons. First of all, these embryonic stem cells are being manipulated by being inserted into an embryo. Granted this embryo is abnormal, but it is an embryo all the same and it provides a vital function that the embryonic stem cells cannot supply – the making of the placenta. This manipulation helps the embryonic stem cells make the embryo, but not everything else. In this case the embryonic stem cells are only doing part of the job and they are also receiving the structure and inductive signals from the tetraploid embryo to form the embryo proper. This is something that embryonic stem cells do not do in culture.
Secondly, embryos undergo development, a process that we understand rather well. This process of development has a goal toward which the embryo proceeds during development. Embryonic stem cells are not in the process of development. They can be induced to form particular cell types or even tissues, but this is part of the embryo or fetus and forming part of the fetus does not constitute embryonic development but only a small part of it. Embryos do not go backwards during development. Cells that do go backwards are usually cancer cells that grow uncontrollably and cannot move to a more differentiated state that puts the brakes on cell division. The fact that embryonic stem cells do move developmentally backward is another indication that they are not embryos. They do something that embryos do not do and this disqualifies them from being embryos.
Thus another argument against the humanity of the early embryo falls into the pit of very bad arguments.