Crabtree's study of antigen-stimulated lymphocyte activation has previously shown that phosphatidylinositol 4,5-bisphosphate (PIP2) signaling helps retain a greater proportion of a mammalian chromatin remodeling complex called BAF in the nucleus. Rando and Crabtree have now determined that BAF binds PIP2 directly, but only if the complex retains two particular subunits: actin and the actin-related protein BAF53.
PIP2 does not trigger exchange of BAF's actin subunit, but PIP2 plus BAF does increase the extent of in vitro actin polymerization. In vitro, BAF binds PIP2 vesicles and the ends and junctions of actin filaments.
Crabtree found that actin contacted two separate domains of the Brg1 subunit of BAF, but only one of these contacts was regulated by PIP2. He suggests that PIP2 may free up a domain of actin so that it can interact with other actin subunits, thus anchoring the complex in the nucleus.
The interaction is unlikely to involve long actin polymers. These have not been seen in the nucleus, despite extensive searches. Instead, Crabtree suggests that BAF might interact with a short actin polymer, immediately capped off, that then binds to a membranous structure. “It's a speculative model,” he says, “but it's consistent with all the data we have at the moment.”
Crabtree is not the only one interested in inositol derivatives. Carl Wu (NIH, Bethesda, MD) and Erin O'Shea (University of California, San Francisco, CA) have recently found in vitro and in vivo evidence for regulation of chromatin remodeling by certain water-soluble inositol polyphosphates. The connection with Crabtree's work remains unclear, however, as the inositol polyphosphates directly regulate remodeling activity, rather than affecting nuclear tethering or a nuclear matrix. ▪