Actin binding of ezrin is activated by specific recognition of PIP2-functionalized lipid bilayers.

In a quantitative manner, we investigated the mechanism of switching ezrin from the dormant to the active, F-actin binding state by recognition of PIP 2. For this purpose, we established a novel in vitro model mimicking ezrin-mediated membrane-cytoskeleton attachment and compared the F-actin binding capability of ezrin that either had been coupled via a His tag to a lipid bilayer displaying Ni-NTA or had been bound to supported membranes containing PIP 2. Epifluorescence and colloidal probe microscopy (CPM) were employed to demonstrate ezrin's conformational switch into an active conformation capable of binding F-actin. Epifluorescence images revealed attachment of fluorescently labeled F-actin solely to PIP 2-bound ezrin. For the first time, colloidal spheres equipped with an artificial cytoskeleton composed of firmly attached F-actin filaments were used to measure quantitatively the maximal adhesion forces and the work of adhesion of the ezrin-F-actin interface. We found that the work of adhesion between PIP 2-bound ezrin and F-actin is substantially larger than that measured between F-actin and ezrin bound to the membrane via the His tag. Collectively, these data indicate that activation of ezrin can occur as a consequence of PIP 2 binding and does not require additional cofactors.

Cooperative adsorption of ezrin on PIP2-containing membranes.

By means of the quartz crystal microbalance (QCM) and scanning force microscopy (SFM), the adsorption of ezrin, a member of the ezrin/radixin/moesin protein family, on l-alpha-phosphatidylinositol-4,5-bisphosphate (PIP(2)) containing solid-supported membranes was investigated. An increase in the PIP(2) content in 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) membranes resulted in an increased amount of bound ezrin strongly supporting the crucial role of PIP(2) for ezrin recruitment to membranes. No ezrin adsorption to membranes composed of pure POPC was detected. To characterize the binding process in more detail, the kinetics and reversibility of ezrin adsorption were investigated by the QCM technique, showing that the protein remains partly bound after rinsing with pure buffer, which we suspected to be a result of lateral interactions between the proteins. SFM images revealed the formation of two-dimensional ezrin clusters on PIP(2)-doped POPC membranes. Time-elapsed SFM images show that the growth of protein domains occurs from a few nucleation sites. The QCM data in conjunction with the results obtained by SFM led us to propose that the binding process of ezrin occurs in a positive cooperative manner. When lateral interactions of the proteins on the membrane were taken into account, we were able to simulate the kinetics obtained from time-resolved QCM readouts by employing a model developed by Minton. On the basis of the kinetic analysis, we were also able to reconstruct the adsorption isotherm.

Scrutiny of annexin A1 mediated membrane-membrane interaction by means of a thickness shear mode resonator and computer simulations.

The dissipational quartz crystal microbalance (D-QCM) technology was applied to monitor the adsorption of vesicles to membrane-bound annexin A1 by simultaneously reading out the shifts in resonance frequency and dissipation. Solid-supported membranes (SSMs) composed of a chemisorbed octanethiol monolayer and a physisorbed 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine/1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoserine monolayer were immobilized on the gold electrode of a 5 MHz quartz plate. Adsorption and desorption of annexin A1 to the SSM was followed by means of the QCM technique. After nonbound annexin A1 was removed from solution, the second membrane binding was monitored by the D-QCM technique, which allowed distinguishing between adsorbed and ruptured vesicles. The results show that vesicles stay always intact independent of the amount of bound annexin and the vesicle and buffer composition. It was shown that the vesicle adsorption process to membrane-bound annexin A1 is fully irreversible and is mediated by two-dimensional annexin clusters. For N-terminally truncated annexin A1, a decrease in the amount of bound vesicles was observed, which might be the result of fewer binding sites presented by the annexin A1 core. Supported by computer simulations, the results demonstrate that the vesicle adsorption process is electrostatically driven, but compared to those of sole electrostatic binding, the rate constants of adsorption are 1-2 orders of magnitude smaller, indicating the presence of a potential barrier.