Figure 7.

Models for Three Mechanisms of Talin Action

The top diagrams show models for the way talin functions in muscle and wing and during germband retraction. At muscle attachment sites (left), talin dimers are bound to integrins or membrane with their heads and to actin directly with the C-terminal ABD and indirectly with vinculin. Actomyosin activity and vinculin are important to separate IBS2 and integrins, likely by exerting force on the rod of talin that pulls it toward the inside of the cell (white arrows). In the wing (middle), talin is oriented parallel to the membrane, with each talin dimer using all IBSs. This talin dimer binds actin, directly with the C-terminal ABD and indirectly with vinculin. During germband retraction (right), talin dimers are bound to actin or membrane directly with the head and indirectly with vinculin (only membrane binding is shown), to integrins with IBS2s, and to actin with the C-terminal ABD. In wing and germband retraction (see below), we suggest IBS2 binds integrins but could interact with other functional binding partners. The logic that generated these models from the mutant phenotypes is demonstrated by depicting the phenotypic effect of the mutations in each model. In muscle, talin head is essential to bind integrins and membrane, as its absence (Δhead) resulted in a null phenotype (red background). The absence of ABD (talin²¹²⁰) resulted in a hypomorphic phenotype (orange background), and this remaining function requires vinculin but it is not clear how vinculin is recruited. The additional loss of IBS2 (talin⁶⁴⁶) did not enhance the phenotype, but talin caused a null phenotype when all VBSs were deleted (talin⁵¹¹). In the wing, talin head is important but integrins can still be linked together through dimerized IBS2s. The absence of ABD resulted in a hypomorphic phenotype, and this remaining function requires vinculin. The additional loss of IBS2 resulted in a null phenotype, as integrins cannot be linked together by talin. During germband retraction, the head was not required (green background) but vinculin was essential in this context. It is not clear how vinculin is recruited to Δhead. All truncations deleting ABD activity resulted in a null phenotype.