David Calderwood, PhD

Integrins are essential heterodimeric adhesion receptors formed by the non-covalent association of a and ß subunits. Each subunit is a type I transmembrane glycoprotein that has relatively large extracellular domains and a generally short cytoplasmic tail. Humans contain 18 a and 8 ß subunits that combine to produce at least 24 different heterodimers, each of which can bind to a specific repertoire of cell surface, ECM or soluble protein ligands. Cell-cell and cell-substratum adhesion is mediated by the binding of integrin extracellular domains to diverse protein ligands, however, cellular control of these adhesive interactions and their translation into dynamic cellular responses, such as cell spreading or migration, requires the integrin cytoplasmic tails. These short tails bind to intracellular ligands that connect the receptors to signaling pathways and cytoskeletal networks. Hence, by binding both extracellular and intracellular ligands, integrins provide a transmembrane link for the bidirectional transmission of mechanical force and biochemical signals across the plasma membrane.

We have found that talin binds to integrin ß subunit cytoplasmic tails through the FERM domain within the talin head. Integrin binding occurs via a variant of the classical PTB domain-NPxY interaction and, in addition to linking integrins to actin stress fibers, this interaction induces conformational changes in the integrin ectodomains that regulate integrin ligand-binding affinity (integrin activation). Tight regulation of integrin activation is essential because it controls cell adhesion, migration, and assembly of an extracellular matrix. Hence integrin activation is a critical step in angiogenesis, tumor cell metastasis, embryonic development, cardiac function and the immune response, and cellular control of integrin activation plays important roles in health and disease throughout development and during the course of adult life.

More recently we have investigated the binding of another class of FERM domain
proteins, the kindlins. Kindlins, like talins, contain an atypical FERM domain and we predict them to be structurally closely related to the integrin-activating talin head domain. We find that kindlins bind integrin ß tails and regulate integrin activation and signaling. The molecular and structural basis for kindlins effects on integrins are the subject of ongoing work.

We have also found that actin cross-linking proteins of the filamin family (filamin A, B and C) bind integrin ß tails, that tight association of filamin with integrin ß tails inhibits cell migration and that filamins can inhibit integrin activation. The regulated binding of filamin to integrin ß tails may therefore provide a control point for regulation of cell migration. Our structural analyses of filamin-integrin interactions revealed the basis for integrin binding and identified regulatory mechanisms including, competition with talin (which provides a mechanism by which filamin binding can suppress integrin activation), auto-inhibition by adjacent filamin domains (which may be released by mechanical stretching) and competition with other filamin-binding proteins. Recently we have shown that filamins are important for initiation of cell migration and that during cell differentiation filamin levels can be controlled by poly-ubiquitinylation and proteasomal degradation. Investigations of the molecular bases for these effects are underway.

Finally we have initiated structural and functional studies on the integrin-linked kinase (ILK); a signaling protein implicated in both integrin activation, possibly through interactions with kindlins, and in signaling downstream of integrins. In collaboration with the Boggon Lab (Pharmacology, Yale) we have solved the structures of a complex between the ILK ankyrin-repeat domain and the 1st LIM domain of the ILK-binding protein PINCH1 or PINCH2. These structures revealed the molecular basis of ILK-PINCH interactions, which are essential for targeting of ILK-PINCH-parvin heterotrimeric complexes to adhesion sites where they act as key signaling nodes. Ongoing structural and functional studies focus on other ILK domains as well as larger multi-domain complexes.