Clusters of integrin (green) grow and adhesions (yellow) are redistributed in fibroblasts lacking ICAP-1 (bottom).
The tail region of β1 integrin is tiny—only 47 amino acids long. That doesn't leave much room for the binding of cytoplasmic accessory proteins that help control the integrin's activation state. When talin holds this position, it flips open integrin as a first step toward full activation. Active integrins then cluster into patches called focal adhesions, which stick a cell to its surface. Signals from the adhesions then relay matrix information to the rest of the cell.
In the new paper, the authors show that talin's competitor for integrin binding—a protein called ICAP-1—ensures that integrin activation does not occur prematurely. They found that fibroblasts and osteoblasts that lack ICAP-1 had overzealous focal adhesions: the adhesions were larger, more widespread, and formed more rapidly than those of normal cells grown on the β1 integrin substrate fibronectin.
In the mutant cells, talin was recruited to integrins prematurely. Their integrins were thus more adhesive, fooling cells into sensing a denser matrix than was actually present. The ICAP-1 mutant cells spread and migrated on lower densities of fibronectin and collagen (another β1 integrin substrate). The authors reproduced these effects by artificially activating β1 integrins in wild-type cells.
Matrix sensing is important for determining migration speed, cell spreading, and lineage specification in progenitor cell lines. The authors previously showed that mice lacking ICAP-1 have structural bone defects but are viable. Evidently, some cell types—such as bone forming osteoblasts, which start out on a smooth, flexible substrate that later hardens—are more sensitive to their environment than others.
Although delayed by ICAP-1, talin must eventually make its way onto the integrin tail for adhesions to form. The authors are now searching for signals that trigger the dismissal of ICAP-1, allowing talin to take its place. 
Reference:
