We know that cells lead intricate lives of growth, change, and division. We also know that DNA has not only the four letters A, T, C, and G, but also an intricate grammar of modifications on DNA-associated proteins, termed chromatin, that changes over time. We can surmise that there is a connection between the life cycle of a cell, called the cell cycle, and its chromatin. But how does the cell cycle influence chromatin? Yang Xu and colleagues shed new light on this question in a paper in the latest issue of the Cell Cycle.
Before a cell can divide, it must first condense its chromatin into packages called mitotic chromosomes, so that its genome may be evenly divided between its two daughters. One of the chromatin modifications that promotes this condensation is the deubiquitination of histone H2A. It’s been known for six years that a protein called Ubp-M can deubiquitinate histone H2A. Now Xu and colleagues explain what causes Ubp-M to deubiquitinate histone H2A before mitosis and not at other times in the cell cycle.
Xu and colleagues focused on a phosphorylation on the 552nd amino acid, a serine, of Ubp-M. This serine is in a motif that a kinase called CDK1 likes to phosphorylate. CDK1 is to the cell cycle what a conductor is to the symphony orchestra: it coordinates all the events, so that they happen in the right sequence and at the appropriate time. By knocking down CDK1 and using chemical inhibitors, Xu and colleagues established that CDK1 indeed phosphorylates Ubp-M on its serine 552.
Phosphorylation changes interactions between proteins. To find the function of the phosphorylation of serine 552, Xu and colleagues looked at the interaction between Ubp-M and a nuclear exporter called CRM1. This is a particularly interesting interaction because Ubp-M spends most of the cell cycle in the cytoplasm, even though it must go to the nucleus to deubiquitinate histone H2A. Therefore, Ubp-M is actively exported from the nucleus, and Xu and colleagues used an inhibitor of CRM1 to show that CRM1 participates in this export. Interestingly, a mutant version of Ubp-M that cannot be phosphorylated on the 552nd amino acid does not get exported as much. This mutant version also decreases cell proliferation and reduces the number of cells that enter mitosis. However, the mutation has no effect on the ability of Ubp-M to deubiquitinate histone H2A. Since CDK1 becomes more active before mitosis, Xu and colleagues propose that it phosphorylates Ubp-M on serine 552 and increases the fraction of Ubp-M in the nucleus, thus promoting chromatin condensation and mitosis.
Serine 552 of Ubp-M is present in primates but is not conserved in the mouse or rat homolog of Ubp-M. Though this particular example of temporal control using phosphorylation and localization occurs in only a few animal species, the principle is likely more general. Moreover, Ubp-M may contain other more conserved phosphorylation sites that function in the same way. And it is intriguing to speculate what special function this phosphorylation may serve in primates. Regardless, Xu and colleagues flesh out a direct connection between the cell cycle and chromatin modification to a rare level of detail.