Integrin Phosphorylation as an Off Switch for Integrin Activation

An individual cell inside the human body is in a dynamic environment: it not only has to anchor itself to its surroundings but also be able to communicate with them and respond as appropriate. One group of proteins--the integrins--play a central role in all of these tasks. The integrins are large (about 200,000 Da) membrane-spanning proteins, and each integrin consists of two subunits (alpha and beta). The vast majority of the integrin is located on the exterior of the cell, where it anchors the cell to the extracellular matrix. Each subunit has a short tail inside of the cell, and the tails on the beta subunit in particular are involved in a variety of interactions with intracellular proteins.

Unlike normal cell surface receptors, which send signals into the cell after binding soluble molecules outside of the cell, the integrins can signal in both directions across the cell membrane. These protein-protein interactions involving the beta cytoplasmic tail are key for signaling in both directions. Integrins only bind the extracellular matrix in their active state, and integrins can be activated by inside-out signaling when a protein called talin binds to the beta cytoplasmic tail, causing the extracellular portion of the integrin to adopt a more open conformation capable of binding to the extracellular matrix. In a cell that is stationary in the body, the integrins remain in an active conformation, holding the cell in place. In cells circulating in the blood stream--such as platelets--however, the integrins exist in an inactive state, allowing such cells to circulate freely. When the body is wounded, though, the platelet integrins are activated, causing the platelets to form a clot.

A more nuanced case occurs when a cell has to migrate through the body (in the formation of a new blood vessel, for example). This cell has to remain adherent to the matrix but also free to crawl along it. This involves a dynamic interplay of signaling molecules, selectively switching individual integrins on and off. How such intricate signaling is accomplished remains a mystery, but our most recent paper (spearheaded by Camilla Oxley) indicates that it is likely accomplished in part by competition between different proteins that bind to beta integrin tails.

In the paper, we explore the structural basis of the competition between two proteins: talin (an integrin activator) and Dok1 (an inhibitor of activation). We show that for unmodified integrins, talin has a higher affinity for the integrin tail than Dok1. However, if a particular tyrosine residue in the beta integrin tail is phosphorylated, the affinities are reversed, and Dok1 has a much higher affinity for Dok1 than talin (and this is due primarily to a large increase in Dok1 affinity as opposed to the much smaller decrease in talin affinity). The bottom line is that phosphorylating the beta integrin tail is a way of deactivating the integrin. This is particularly interesting, because here you have a tiny 80 Da modification acting as an on-off switch for a 200,000 Da protein.

Through both x-ray crystallography and NMR, we were able to explain this switch at the structural level. We are currently doing more work to understand the role of phosphorylation in integrin activation, and we have already collected quite a bit of data beyond what's published in the current paper. An interesting but more subtle point that emerges from the current paper, though, is that integrin phosphorylation inhibits integrin activation by increasing Dok1 affinity for the beta integrin tail, but not necessarily by decreasing talin affinity. Previous scientific work has largely assumed that the effects of integrin phosphorylation are due to changes in talin affinity, but here we demonstrate that this is not a valid assumption.

Cell adhesion and migration are scientifically interesting in their own right, especially since they play a central role in a variety of biological processes, including growth, development, and healing. This work is medically significant as well, especially in regards to cancer. Normally, if a cell loses contact with the extracellular matrix, it kills itself by going into programmed cell death (apoptosis). One of the hallmarks of a cancer cell is that it no longer requires contact with the matrix for its survival. Because of this, such a cell can break away from its environment and invade new tissues. This, of course, is the process of metastasis. Although the work in our current paper is far removed from any of these broader clinical applications, a deeper understanding of cell adhesion and migration should help address conditions--such as cancer--that involve pathological cell adhesion and migration.

Oxley, C.L., Anthis, N.J., Lowe, E.D., Vakonakis, I., Campbell, I.D., Wegener, K.L. (2007). An integrin phosphorylation switch: the effect of beta3 integrin tail phosphorylation on DOK1 and talin binding. Journal of Biological Chemistry DOI: 10.1074/jbc.M709435200