Pluripotency is controlled by three master proteins—Oct3/4, Nanog, and Sox2. These are transcription factors, proteins that control genes by binding to the relevant DNA. When these transcription factors themselves are knocked down or eliminated, pluripotency is lost; that is, embryonic stem (ES) cells differentiate into mature cells.

Several genes associated with pluripotency contain enhancer binding sites for Oct3/4 and Sox2. Thus, a plausible model for regulating pluripotency is that Oc3/4 and Sox2 activate these enhancers, turning up their own expression plus the expression of downstream genes in the network. This week, Masui et al report that, surprisingly, Sox 2 is not needed to activate Oct-Sox enhancers in mouse cells.

The researchers created a mouse ES cell line with an inducible null Sox2 gene, which meant Sox2 levels could be reduced on command. Surprisingly, the activity of the Oct-Sox enhancers stayed on even when the Sox2 gene was off. Next, Masui et al made cells with mutated enhancers, such that the enhancers were incapable of binding Sox2. They found that when Sox2 is out of the picture, other Sox factors step in to regulate the Oct-Sox enhancers.

In that case, why is Sox2 essential for pluripotency? The researchers used microarrays to probe for altered gene expression in Sox2 null ES cells and found several genes that regulate Oct3/4. Restoring Oct3/4 levels in the Sox2-null ES cells restored pluripotency. That supports the idea that Sox2 is essential for pluripotency because it regulates genes that affect Oct3/4, not because of its effects on enhancers.

Not surprisingly, many pieces of this intricate puzzle are missing. For instance, the role other Sox family members play in regulating these enhancers is unclear. Also unknown is what Sox2 physically binds to influence Oct3/4 gene expression. However, it appears that Sox2 had been put into the wrong place in terms of its role in regulating pluripotency. Now, additional pieces of this puzzling network should be easier to fit.