Response: Role of mouse maternal Cdx2: what's the debate all about?

Here I would like to respond to the commentary ‘Role of mouse maternal Cdx2: what’s the debate all about?’ (Wu and Scholer, 2011) that was written as a response to my original commentary, ‘What is the role of maternally provided Cdx2 mRNA in early mouse embryogenesis?’ (Bruce, 2011). Firstly, I would stress that the arrest phenotypes reported in Jedrusik et al. (2010) were observed in two independent laboratories (Zernicka-Goetz laboratory in Cambridge and Yao laboratory in Stanford) and the experiments were performed by different researchers and by different methods. In both laboratories the functional depletion of both maternal and zygotically provided Cdx2 from the zygote stage resulted in earlier arrest phenotypes than the zygotic Cdx2 knockout. Given this independent verification, it is highly improbable that both sets of results were ‘actually an artefact of the experimental manipulation’ as Wu and Scholer suggest. Additionally, the arrest phenotypes reported by Jedrusik et al. (2010), that were always judged relative to injection controls that successfully developed, were consistent using embryos derived from two distinct genetic crosses, suggesting that genetic background is not a confounding factor; notwithstanding any uncharacterized genetic difference in the embryos employed by Wu and colleagues. Furthermore, Jedrusik et al. (2010) reported Cdx2-specific RNAi-induced arrest phenotypes in 88.6% of embryos before the blastocyst stage, the earliest phenotypes being centred on the 8to 16-cell transition. In the molecular characterization of these arrests, embryos were not preselected in any way to concentrate only on the earlier phenotype, as implied by Wu and Scholer. Consequently, even embryos that could