PIP2 as local second messenger: a critical re-evaluation

Abstract

Phosphatidylinositol 4,5-biphosphate (PIP2) has been proposed to act as a second messenger in the regulation of many cell processes. If so, then PIP2 should fulfill two important criteria's: first PIP2 levels should vary spatially or temporally under physiological conditions, and secondly, these variations should suffice to influence cellular processes. In this thesis we have addressed the issue and provide data that support the idea that PIP2 can function as a second messenger, since PIP2 levels were found to vary significantly over time affecting cell survival, as well as cortical actin dynamics. Interestingly, these results also suggest that PIP2 influences multiple physiological processes within the same cell, apparently with spatial resolution. This view also prevails in the literature: it is widely hypothesized that the plasma membrane contains spatially confined PIP2 pools or domains. Indeed, recent studies that used GFP-tagged pleckstrin homology domains (GFP-PH) as fluorescent PIP2 sensors suggest that this lipid is enriched in membrane patches. However, we report here that this concept needs revision. Using three distinct fluorescent GFP-tagged pleckstrin homology domains, we show that highly mobile GFP-PH patches colocalize perfectly with various lipophilic membrane dyes and, hence, represent increased lipid content rather than PIP2-enriched microdomains. We show that bright patches are caused by submicroscopical folds and ruffles in the membrane that can be directly visualized with a novel numerically enhanced imaging method. Moreover, to test sub-resolutional clustering, we analyzed clustering of PIP2-binding pleckstrin homology domains by electron microscopy as well as by FRET in live cells. These assays suggest that PIP2 is neither clustered in micrometer nor in sub-resolution domains. This can be explained by the fast diffusion of PIP2 which limits the formation of small domains, as established by FRAP experiments. However, at larger scale, and especially at structures where diffusion is limiting, PIP2 gradients can be induces. Thus, our data support a model in which PIP2 has a second messenger role in the regulation of cell survival and the cortical actin cytoskeleton, but they challenge a model in which this regulation exist at a smaller scale.