Untersuchung der Bedeutung von Stonin1 für die Dynamik fokaler Adhäsionen und die Unterdrückung von Tumoren
Zusammenfassung der Projektergebnisse
Adhesion structures are critical for anchoring cells in their environment, as signaling platforms and for cell migration. In line with these diverse functionalities, there are different types of cellmatrix adhesions. The best-studied type are the canonical integrin-based focal adhesions. However, non-canonical integrin adhesions which lack focal adhesion proteins have also been described. The most prominent example are the reticular adhesions (RAs), which are characterized by the presence of endocytic factors and are also known as clathrin plaques or flat clathrin lattices. So far, we still lack a deeper understanding of the dynamic relationship between the different adhesion types sharing a common integrin backbone. At the onset of this DFG project we set out to understand the role of stonin1 in the dynamics of focal adhesions since we had observed stonin1 to accumulate at sites of focal adhesion disassembly. However, in the course of our studies we discovered that sites of focal adhesion disassembly are at the same time locations where non-canonical cell-matrix adhesions form and that stonin1 localizes quantitatively to these non-canonical integrin-based adhesions. In fact, stonin1 is the only known protein to date that resides exclusively at non-canonical avb5 integrin adhesions and thus can be used as unambiguous marker. Using multi-colour live cell imaging and employing the stonin1 as bona fide marker for non-canonical avb5 integrin-based adhesions, we discovered that canonical and non-canonical adhesions can reciprocally interconvert by selective loss and recruitment of components to a stable avb5 integrin scaffold. Thereby, non-canonical adhesions can serve as points of origin for the generation of focal adhesions. Based on our observations we propose a novel mode of focal adhesion assembly which bypasses the phase of gradual integrin clustering during lamellipodial protrusion by making use of pre-existing avb5 integrin scaffolds. Conventionally, focal adhesions start out as small, shortlived adhesions, so-called nascent adhesions, forming right behind the leading edge of a migrating cell by the clustering of integrins which is promoted by retrograde actin flow. These initial dot-like adhesions can mature into the larger elongated focal adhesions when successfully connecting to contractile actomyosin containing stress fibers and experiencing tension. This model inherently limits the formation of focal adhesions to the cell periphery, more specifically to the protruding leading edge of moving cells and does not provide a mechanism for the formation of focal adhesions at internal cellular locations or within stationary confluent cells. We demonstrate that the use of pre-existing avb5 integrin scaffolds that were previously "hosting" non-canonical adhesion proteins can circumvent this limitation of the conventional focal adhesion assembly mechanism, thereby enlarging the options for focal adhesion assembly in different cellular contexts and allowing for the rapid adaptation of adhesion structures to changing cellular needs. In summary, we show that depending on cellular conditions integrins can serve as scaffolds for diverse adhesion types comprising canonical focal adhesions as well as noncanonical RAs/plaques, migratory retraction fibers and extensive adhesion networks. We propose a model where a stable non-canonical integrin scaffold structure that is positiv for stonin1 can either be converted into a focal adhesion by the reciprocal loss of stonin1 and recruitment of canonical focal adhesion proteins or can be turned into an RA/plaque by the association of endocytic proteins. Consequently, focal adhesions and RAs/plaques can interconvert via this intermediate avb5 integrin- and stonin1-positive adhesion structure. Thus, our results expand the emerging paradigm that integrin-based adhesions should not be viewed so much as discrete structures that form and disassemble independently of each other, but rather as an interconvertible and dynamic continuum of adhesion types that can rapidly adapt to the changing (extra-)cellular environment by the selective loss and/or recruitment of molecular components.
