Structural study on two tandem helix bundles of the ROD domain of talin, an integrin activator

Talin, as the activator of integrin and the adaptor between the cytoskeleton and integrin, plays a key role in a series of processes such as cell adhesion and migration. The activation of integrin involves F3 subdomain of Talin-FERMdomain binding the cytoplasmic tail of integrin β-subunit. Talin has two states: auto-inhibited and activated. We previously reported the auto-inhibition complex structure of Talin F2F3/R9, in which the integrin binding site F3 interacts with R9(1654∼1822 a.a.) of Talin-ROD, such that integrin cannot be activated. However, besides F3 and R9, it remains unclear what structural or functional roles the other domains of the 270 ku Talin play in the regulation of its activation. Here we solved the crystal structures of Talin R9-R10 (1654∼1973 a.a.) and R10-R11 (1815∼2140 a.a.), respectively. R9, R10 and R11 are all 5-helix bundles. R9 and R10 is joined together by a long α-helix instead of a flexible loop, and the two bundles are located at the opposite sides of the long helix with an angle of about 150°. The linker between R10 and R11 is stabilized by neighboring hydrogen bonds, forming an angle of about 120° between the two bundles. These angles observed in our crystal structures are consistent with the previously reported SAXS and EMresults. After superimposition of R9-10, R10-11 with previously reported structures of R7-8 and R11-12, a model of R7-12 was acquired, which adopts an elongated linear conformation, except that R8 protrudes from the ROD. According to this model, R10-12 does not intrude the interaction between F3 and R9, whereas R8 not only masks the F3 binding site of R9, but also might electrostatically hinders F2F3 approaching via its unique positively charged surface. This hypothesis was further verified by the results of size exclusion chromatography. Our work provides a new structural basis for studying the mechanismof Talin auto-inhibition.