Lifting the veils on TMEM16E function.

Abstract

The TMEM16E protein (synonymous to anocta min 5) has been attracting a great deal of interest, since mutations in the human TMEM16E gene were implicated in two different types of hereditary diseases: in gnathodiaphyseal dysplasia (GDD), a rare skeletal syndrome [1], and in muscular dystrophies, limb-girdle muscular dystrophy-2L (LGMD2L) and distal Miyoshi myopathy (MMD3) [2]. Yet for many years, it was not even known which may be the very basic function carried out by this membrane protein, let alone how this function may contribute to physiological and pathophysiological settings. Several facets have contributed to this uncertainty surrounding TMEM16E function. While proteins of the TMEM16 family were initially considered Ca-activated chloride channels, it became clear later that many of them are in reality Ca-activated lipid scramblases mediating the stimulus-induced passive transport of phospholipids, in particular phosphatidylserine (PtdSer), between the leaflets of the membrane bilayer (for review [3]). Moreover, early localization studies indicated that native or heterologously expressed TMEM16E protein was restricted to intracellular membranes and therefore inaccessible to classical approaches like whole-cell patchclamp and scrambling assays [4]. This uncertainty has now been dispelled. Several studies published in recent years concur on the fact that TMEM16E belongs to the group of family members with Ca-activated phospholipid scrambling (PLS) activity. A first hint in favour of the “scramblase” option came from a chimeric approach in which a 35-aa-long stretch connecting trans-membrane domains 4 and 5 (designated “scrambling domain” in the TMEM16F sister protein [5]) was swapped between TMEM16E and the plasma membrane-localized TMEM16A. Introduction of the short TMEM16E stretch was sufficient to endow the Ca-activated chloride channel TMEM16A with lipid scrambling activity [6]. Work from our group provided direct demonstration of Ca dependent PLS for the human TMEM16E wild-type protein exploiting its partial plasma membrane (PM) localization following transient overexpression in HEK293 cells [7]. Targeting of a TMEM16E-EGFP fusion to the cell surface was shown by colocalization with the PM marker FM4-64. Additional independent evidence for partial PM localization came from surface biotinylation assays on HEK293 cells stably overexpressing a codon-optimized hTMEM16E version [8]. It is not yet clear if the PM localization of TMEM16E has relevance in its diverse physiological contexts or if it is simply a consequence of protein overexpression. Data from isolated mouse muscle cells indicate that TMEM16E PLS activity may indeed contribute to extracellular PtdSer exposure [8] (see below). Detection of PLS typically relies on annexin-V binding to PtdSer accumulating in the outer leaflet of the membrane as a consequence of scrambling activity. In both HEK293 cell models [7,8], scrambling assays concurrently revealed annexin-V binding at the cell surface of TMEM16E-expressing cells, which