Developmental biology

A mouse is not a cow

Early cell-lineage decisions during embryonic development differ between mice and cows. This finding calls for a re-examination of developmental variations across mammals, but does not undermine use of the mouse as a model organism.

The mammalian blastocyst is a thing of beauty. Over a period of a few days after the union of an egg with sperm, the fertilized egg divides to generate this tiny hollow sphere of cells, which has a cluster of enclosed cells at one end of the fluid-filled cavity. The outer cells are called the trophectoderm and the inner cells are, inventively, named the inner cell mass. But when do cells commit to becoming one or the other, and how? Writing in Developmental Cell, Berg et al.1 show that the answers to these questions are not the same for mice and cattle.

Pluripotency — a cell's ability to differentiate into all cell types of the body — is a common property of the inner cell mass (ICM) of all mammalian blastocysts and is always associated with the expression and function of the transcription factor Oct4. The trophectoderm, which later generates all of the specialized layers of the placenta, also expresses a number of lineage-restricted transcription factors, most notably Cdx2.

In mice, deletion of either the Oct4 gene (also known as Pou5f1) or the Cdx2 gene leads to the formation of abnormal blastocysts: ICM cells of Oct4-mutant blastocysts express trophectoderm markers and lose pluripotency2, whereas the outer cells of Cdx2-mutant blastocysts express pluripotency markers such as Oct4 ectopically and fail to differentiate further down the trophectoderm lineage3. This suggests a model — albeit an overly simplistic one — whereby restricted expression of Oct4 and Cdx2 leads to reciprocal repression of the opposing lineage and establishes cell fate.

Berg et al.1 asked whether this model applies to cell-fate decisions in cows. They find that, unlike in mice, Oct4 expression is not restricted only to the ICM during the early stages of cow blastocyst development. Instead, Oct4 is co-expressed with Cdx2 in the trophectoderm for some time after the beginning of blastocyst formation. This observation is consistent with previous reports and has also been made for pig and human embryos (for example, see refs 4, 5). Even in the mouse, Oct4 expression overlaps with Cdx2 expression during the late cleavage and early blastocyst stages of embryonic development, and is restricted to the ICM only by the fully expanded blastocyst stage3.

So why is Oct4 expression maintained for longer in the cow trophectoderm than in its mouse equivalent? Through experiments involving cow blastocysts engineered to express a fluorescently tagged version of mouse Oct4 (the mouse Oct4GFP transgene), Berg and co-workers show that the factors that restrict Oct4 expression to the ICM are not available, or not functional, in the cow blastocyst (Fig. 1a). Cdx2 could be one such factor, but the authors' data suggest that this protein has a role only later during cow embryonic development. However, Berg and colleagues do not investigate whether the role of Cdx2 in restricting Oct4 expression is simply delayed in the cow embryo, nor whether Oct4 is ectopically expressed later during development in embryos treated to express reduced levels of Cdx2.

Figure 1: Oct4 regulation in mouse and cow blastocysts.

a, Berg et al.1 find that the expression of a GFP fluorescent reporter transgene controlled by the regulatory elements of the mouse Oct4 gene (blue) is restricted to the inner cell mass (ICM) in mouse blastocysts but not cow blastocysts. b, The same transgene, but containing the bovine Oct4 regulatory elements (red), is not restricted to the ICM in either cow or mouse blastocysts. c, The authors narrow down this effect of bovine regulatory elements to the CR4 region.

The paper1 shows that a mouse Oct4GFP transgene containing the bovine Oct4 regulatory elements is expressed in both the ICM and trophectoderm in fully expanded blastocysts of both the cow and the mouse (Fig. 1b). This suggests that Cdx2, which is active in mouse blastocysts, is not the only factor that affects the timing of Oct4 repression. It also indicates that bovine regulatory elements do not respond to the factors that downregulate Oct4 in mouse blastocysts.

Of the four evolutionarily conserved regulatory regions around the Oct4 locus, CR4 shows the most sequence divergence between the mouse and the cow. When Berg et al. replaced mouse CR4 with the cow version in the mouse Oct4GFP construct, it behaved like the cow gene in the mouse blastocysts (Fig. 1c). Thus changes in both DNA regulatory regions and the factors that bind to such sequences drive differences in the regulation of Oct4 expression between mouse and cow blastocysts.

It would be interesting to test, in transgenic mice, whether regulatory elements of the human OCT4 gene behave like the mouse or the cow sequences. Although human blastocysts, like those of domestic animals, express Oct4 in the trophectoderm for an extended period compared with mice, the period of overlap of Cdx2 and Oct4 expression is only slightly longer than in the mouse. Human OCT4 is clearly restricted to the ICM by day 6 before embryo implantation6.

But why do these regulatory differences exist among the blastocysts of different mammals? Evolutionarily, the placenta is a recent invention, and still seems to be a work in progress. There is huge variation in trophectoderm and placental morphology across different mammalian species, accompanied by recent evolutionary divergence in placenta-specific gene families7. For example, a mouse blastocyst attaches and implants in the uterus by embryonic day 5 (E5); a human blastocyst grows a little larger but then implants by E7–9 with highly invasive trophoblast outgrowth; and in cows, pigs and sheep the blastocyst floats in the uterus for 2–3 weeks before attaching.

Berg et al. propose that such differences lead to earlier restriction of trophectoderm cell fate in the mouse than in the cow. Indeed, results of their experiments — involving chimaeric blastocysts generated by mixing trophectoderm cells from different stages of development with host embryos — support this proposal.

In a remarkable technical tour de force, they also transferred the chimaeric cow blastocysts to recipient cows and recovered them later in development to show that early trophectoderm cells can contribute to developing ICM derivatives. This is one of the first attempts to test the timing of lineage restriction in a species other than the mouse.

This study emphasizes the need to explore the timing and mechanism of functional lineage restriction in blastocysts of different mammals, including humans. Differences in these parameters may underlie the known difficulty in deriving validated pluripotent embryonic stem cells and trophoblast stem cells from many mammalian species. Although fibroblasts have been reprogrammed into induced pluripotent stem cells in several domestic species, including the cow, these lines often depend on continued expression of exogenous reprogramming factors. Clearly, we need a better understanding of the control of pluripotency in all these species.

As we learn more about the precise details of mouse blastocyst development, we must be constantly evaluating similarities and differences between them and those of humans and other species. This will help us to truly understand mammalian embryo diversity.


  1. 1

    Berg, D. K. et al. Dev. Cell 20, 244–255 (2011).

    CAS  Article  Google Scholar 

  2. 2

    Nichols, J. Cell 95, 379–391 (1998).

    CAS  Article  Google Scholar 

  3. 3

    Strumpf, D. et al. Development 132, 2093–2102 (2005).

    CAS  Article  Google Scholar 

  4. 4

    Rossant, J. Reprod. Fertil. Dev. 19, 111–118 (2007).

    CAS  Article  Google Scholar 

  5. 5

    Blomberg, L., Hashizume, K. & Viebahn, C. Reproduction 135, 181–195 (2008).

    CAS  Article  Google Scholar 

  6. 6

    Chen, A. E. et al. Cell Stem Cell 4, 103–106 (2009).

    CAS  Article  Google Scholar 

  7. 7

    Wildman, D. E. Placenta 32, 142–145 (2011).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Janet Rossant.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Rossant, J. A mouse is not a cow. Nature 471, 457–458 (2011).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing