Quantification of the in-vivo affinity of PI(4,5)P2 effectors and sensors with the phosphoinositide phosphatase Ci-VSP

Summary Quantification of the in-vivo affinity of PI(4,5)P2 effectors and sensors with the phosphoinositide phosphatase Ci-VSP Phosphoinositides (PIs) constitute a family of membrane phospholipids which play important roles in many eukaryotic signal transduction pathways. The PI phosphatidylinositol(4,5)bisphosphate (PI(4,5)P2) is predominantly located in the plasma membrane and regulates many cellular functions including protein activation (e.g. ion channels). PI(4,5)P2-mediated signalling usually occurs via dynamic changes in the PI(4,5)P2 concentration. Given the multitude of cellular functions that depend on PI(4,5)P2, it is not fully understood how specificity of signalling is achieved, i.e. how only a subset of potential effector proteins can respond to a given change in the PI(4,5)P2 concentration. One major factor determining signalling specificity may be the PI(4,5)P2 affinity of these effectors. Thus, to understand the impact of dynamic PI(4,5)P2 concentration changes on effector proteins it is necessary to know their PI(4,5)P2 affinity. However, methods to define PI(4,5)P2 affinities used so far have serious limitations in their potential to assess the affinities of PI(4,5)P2 effectors (e.g. ion channels) under physiological conditions, i.e. in the living cell. Therefore, the PI(4,5)P2 affinity of many proteins has yet to be established. In this study, the voltage sensor-containing PI(4,5)P2 5’ phosphatase of Ciona intestinalis (Ci-VSP) was established as a novel tool to quantify the PI(4,5)P2 affinity of effector and sensor proteins. Ion channels are a physiologically relevant class of cellular PI(4,5)P2 effectors with various PI(4,5)P2 affinities, and their regulation by PIs has been the subject of wide-ranging investigation. The activity of inwardly rectifying potassium (Kir) channels, for example, is controlled by PI(4,5)P2 binding. Thus, Kir channels were used as a model system to establish the potential of Ci-VSP for estimating PI(4,5)P2 affinities. Kir currents were recorded during Ci-VSP activity using whole-cell patch-clamp recordings on CHO cells or two-electrode voltage clamp recordings on Xenopus laevis oocytes. Kir currents declined upon depolarization due to PI(4,5)P2 depletion by Ci-VSP activity, and recovered upon repolarization as a result of endogenous PI(4,5)P2 resynthesis. Steady state channel deactivation upon gradual Ci-VSP activation served as a readout for the level of channel-PI(4,5)P2 ii affinity. Different degrees of Ci-VSP activity were required for channel deactivation, consistent with their different PI(4,5)P2 affinities. Kir2.1WT channels were found to have a high PI(4,5)P2 affinity, Kir1.1WT channels an intermediate one, and Kir3 channels a low one. The reduced channel-PI(4,5)P2 affinity occurring as a result of amino acid mutations (e.g. R228Q in Kir2.1) that lead to severe diseases such as Andersen’s syndrome was reliably detected with Ci-VSP. Similarly, using Ci-VSP allowed to track successfully a change in the PI(4,5)P2 affinity of Kir3 channels modulated by Gβγ. These data revealed that Ci-VSP allowed, in contrast to earlier established approaches, the reversible and gradual titration of endogenous PI(4,5)P2 in intact cells. The reliability of Ci-VSP was confirmed by comparison of these results with those obtained with an independent approach to deplete PI(4,5)P2. PI(4,5)P2 depletion was achieved by chemically triggered recruitment of the PI(4,5)P2 5’ phosphatase, Inp54p, to the plasma membrane. Kir currents of all channels investigated declined upon Inp54p-induced PI(4,5)P2 depletion similar to the results obtained with Ci-VSP. Finally, Ci-VSP was used to resolve the controversial PI(4,5)P2 affinities of two PI(4,5)P2 sensor proteins, PLCδ1-PH domain and tubby-CT, with the purpose of assessing their usefulness for monitoring PI(4,5)P2 dynamics in intact cells in future studies. Total internal reflection fluorescence (TIRF) microscopy on CHO cells showed that the PI(4,5)P2 affinity of tubby-CT is lower than that of PLCδ1-PH domain. The dynamic range of the tubby-CT sensor corresponded to the physiological range of PI(4,5)P2 concentration, as indicated by KCNQ2 channel activity recorded simultaneously by patch-clamp. In conclusion, the results establish Ci-VSP as a novel experimental tool for rapid, reversible and gradual manipulation of PI(4,5)P2 in living cells. Ci-VSP allows for the quantitative examination of PI(4,5)P2 sensitivities of cellular effectors and genetically encoded fluorescent sensors of PI(4,5)P2 signaling.