(Cell, Aug 2012) New research unveils how cells maintain identity through mitosis, when most genes turn off and transcription factors abandon their target sites in DNA.
Every dividing cell faces a series of challenges beyond the basic requirements of making sure that its DNA is fully and faithfully replicated and that its cellular components are properly divvied up. Lately, scientists have typically been more preoccupied with understanding how progenitor cells lay the ground for proper differentiation of their progeny. But there is a flip side of this coin that remains an often underappreciated challenge for : to remain the same.
The identity of cells is dictated by their gene expression profiles, which is in turn controlled by a barrage of protein factors that interact with DNA and direct to a subset of genes to be expressed. However, RNA polymerases and most of these factors separate from DNA during mitosis, something that researchers have always seen as a threat to the identity of dividing cells. Of course, daughter cells can inherit a ready-to-use collection of and a series of marks that guide those factors to specific places in DNA. That alone could explain to some extent how cells rapidly recreate a specific gene expression landscape after dividing. But scientists have also recently discovered factors that remain associated to when cells divide. In a recent Cell article, Stephan Kadauke and a group of colleagues led by Gerd Blobel at the Childrens Hospital in Philadelphia reported that such binding is critical for the rapid re-expression of key identity genes following mitosis (1).
Using a series of clever tricks, Kadauke and colleagues compared resting vs. mitotic cells with regards to the DNA binding profile of Gata1, a transcription factor required for establishing and maintaining identity. They observed that a very small fraction of sites occupied by Gata1 in non-dividing cells (5.3%) remained bound to Gata1 through mitosis. More importantly, these sites were mostly associated to genes essential for identity, in contrast to other Gata1 targets that have nothing to do with erythro-megakaryocytic differentiation and were not bound by Gata1 in mitotic cells.
The researchers then compared the expression of a selected subset of genes prior and shortly after mitosis. Genes bound by Gata1 through mitosis reached maximum expression following mitosis much more quickly than counterparts that release Gata1 during cell division. Moreover, when Kadauke and colleagues replaced the endogenous Gata1 protein with an engineered construct that ensures the quick degradation of Gata1 during mitosis, they observed that the rapid re-expression of genes “bookmarked” by Gata1 was abrogated.
A burning question remains as to why specific genes ought to be bookmarked through mitosis. Kadauke and colleagues did not report on what happened to the identity of cells in which Gata1 is rapidly degraded during mitosis. Interestingly, some of the targets bookmarked by Gata1 are in fact by this factor, which invites the exciting possibility that a quick re-repression of genes that may confuse a daughter cell’s identity places as much pressure to select for gene bookmarking as ensuring the rapid re-expression of genes that determine a cell’s identity.
Blobel’s team could not yet identify what distinguishes the sites bound by Gata1 during mitosis from the sites that are not. This was not for a lack of trying; they searched for , differential co-binding of Gata1 , differences in and even tried identifying novel sequences that might distinguish both types of sites. But none of the above seemed to explain why Gata1 would bookmark certain sites over others.
Identifying the mechanisms that make transcription factors remain bound to specific targets, and how these aid in maintaining (or otherwise altering) the identity of daughter cells will surely remain an exciting field for a while. Like a bookmark in a book that we can’t wait to go back and read again.
(1) Kadauke S et al. Tissue-specific mitotic bookmarking by hematopoietic transcription factor GATA1. Cell, 2012 Aug 17;150(4):725-37. doi: 10.1016/j.cell.2012.06.038.