Identification of a drug binding pocket in TMEM16F calcium-activated ion channel and lipid scramblase

TMEM16F exacerbates tau pathology and mediates phosphatidylserine exposure in phospho-tau-burdened neurons

TMEM16F is a calcium-activated phospholipid scramblase and nonselective ion channel, which allows the movement of lipids bidirectionally across the plasma membrane. While the functions of TMEM16F have been extensively characterized in multiple cell types, the role of TMEM16F in the central nervous system remains largely unknown. Here, we sought to study how TMEM16F in the brain may be involved in neurodegeneration. Using a mouse model that expresses the pathological P301S human tau (PS19 mouse), we found reduced tauopathy and microgliosis in 6- to 7-mo-old PS19 mice lacking TMEM16F. Furthermore, this reduction of pathology can be recapitulated in the PS19 mice with TMEM16F removed from neurons, while removal of TMEM16F from microglia of PS19 mice did not significantly impact tauopathy at this time point. Moreover, TMEM16F mediated aberrant phosphatidylserine exposure in neurons with phospho-tau burden. These studies raise the prospect of targeting TMEM16F in neurons as a potential treatment of neurodegeneration.

In or out of the groove? Mechanisms of lipid scrambling by TMEM16 proteins

Phospholipid scramblases mediate the rapid movement of lipids between membrane leaflets, a key step in establishing and maintaining membrane homeostasis of the membranes of all eukaryotic cells and their organelles. Thus, impairment of lipid scrambling can lead to a variety of pathologies. How scramblases catalyzed the transbilayer movement of lipids remains poorly understood. Despite the availability of direct structural information on three unrelated families of scramblases, the TMEM16s, the Xkrs, and ATG-9, a unifying mechanism has failed to emerge thus far. Among these, the most extensively studied and best understood are the Ca2+ activated TMEM16s, which comprise ion channels and/or scramblases. Early work supported the view that these proteins provided a hydrophilic, membrane-exposed groove through which the lipid headgroups could permeate. However, structural, and functional experiments have since challenged this mechanism, leading to the proposal that the TMEM16s distort and thin the membrane near the groove to facilitate lipid scrambling. Here, we review our understanding of the structural and mechanistic underpinnings of lipid scrambling by the TMEM16s and discuss how the different proposals account for the various experimental observations.

Anoctamin pharmacology

TMEM16 proteins, also known as anoctamins, are a family of ten membrane proteins with various tissue expression and subcellular localization. TMEM16A (anoctamin 1) is a plasma membrane protein that acts as a calcium-activated chloride channel. It is expressed in many types of epithelial cells, smooth muscle cells and some neurons. In airway epithelial cells, TMEM16A expression is particularly enhanced by inflammatory stimuli that also promote goblet cell metaplasia and mucus hypersecretion. Therefore, pharmacological modulation of TMEM16A could be beneficial to improve mucociliary clearance in chronic obstructive respiratory diseases. However, the correct approach to modulate TMEM16A activity (activation or inhibition) is still debated. Pharmacological inhibitors of TMEM16A could also be useful as anti-hypertensive agents given the TMEM16A role in smooth muscle contraction. In contrast to TMEM16A, TMEM16F (anoctamin 6) behaves as a calcium-activated phospholipid scramblase, responsible for the externalization of phosphatidylserine on cell surface. Inhibitors of TMEM16F could be useful as anti-coagulants and anti-viral agents. The role of other anoctamins as therapeutic targets is still unclear since their physiological role is still to be defined.

Lipid osmosis, membrane tension, and other mechanochemical driving forces of lipid flow

Nonvesicular lipid transport among different membranes or membrane domains plays crucial roles in lipid homeostasis and organelle biogenesis. However, the forces that drive such lipid transport are not well understood. We propose that lipids tend to flow towards the membrane area with a higher membrane protein density in a process termed lipid osmosis. This process lowers the membrane tension in the area, resulting in a membrane tension difference called osmotic membrane tension. We examine the thermodynamic basis and experimental evidence of lipid osmosis and osmotic membrane tension. We predict that lipid osmosis can drive bulk lipid flows between different membrane regions through lipid transfer proteins, scramblases, or similar barriers that selectively pass lipids but not membrane proteins. We also speculate on the biological functions of lipid osmosis. Finally, we explore other driving forces for lipid transfer and describe potential methods and systems to further test our theory.

Identification of determinants of lipid and ion transport in TMEM16/anoctamin proteins through a Bayesian statistical analysis

The TMEM16/Anoctamin protein family (TMEM16x) is composed of members with different functions; some members form Ca2+-activated chloride channels, while others are lipid scramblases or combine the two functions. TMEM16x proteins are typically activated in response to agonist-induced rises of intracellular Ca2+; thus, they couple Ca2+-signalling with cell electrical activity or plasmalemmal lipid homeostasis. The structural domains underlying these functions are not fully defined. We used a Naïve Bayes classifier to gain insights into these domains. The method enabled identification of regions involved in either ion or lipid transport, and suggested domains for possible pharmacological exploitation. The method allowed the prediction of the transport property of any given TMEM16x. We envisage this strategy could be exploited to illuminate the structure-function relationship of any protein family composed of members playing different molecular roles.

Structural basis of closed groove scrambling by a TMEM16 protein

Activation of Ca2+-dependent TMEM16 scramblases induces phosphatidylserine externalization, a key step in multiple signaling processes. Current models suggest that the TMEM16s scramble lipids by deforming the membrane near a hydrophilic groove and that Ca2+ dependence arises from the different association of lipids with an open or closed groove. However, the molecular rearrangements underlying groove opening and how lipids reorganize outside the closed groove remain unknown. Here we directly visualize how lipids associate at the closed groove of Ca2+-bound fungal nhTMEM16 in nanodiscs using cryo-EM. Functional experiments pinpoint lipid-protein interaction sites critical for closed groove scrambling. Structural and functional analyses suggest groove opening entails the sequential appearance of two π-helical turns in the groove-lining TM6 helix and identify critical rearrangements. Finally, we show that the choice of scaffold protein and lipids affects the conformations of nhTMEM16 and their distribution, highlighting a key role of these factors in cryo-EM structure determination.

Lipid osmosis, membrane tension, and other mechanochemical driving forces of lipid flow

Nonvesicular lipid transport among different membranes or membrane domains plays crucial roles in lipid homeostasis and organelle biogenesis. However, the forces that drive such lipid transport are not well understood. We propose that lipids tend to flow towards the membrane area with a higher membrane protein density in a process termed lipid osmosis. This process lowers the membrane tension in the area, resulting in a membrane tension difference called osmotic membrane tension. We examine the thermodynamic basis and experimental evidence of lipid osmosis and osmotic membrane tension. We predict that lipid osmosis can drive bulk lipid flows between different membrane regions through lipid transfer proteins, scramblases, or other similar barriers that selectively pass lipids but not membrane proteins. We also speculate on the biological functions of lipid osmosis. Finally, we explore other driving forces for lipid transfer and describe potential methods and systems to further test our theory.

Inhibition of mucus secretion by niclosamide and benzbromarone in airways and intestine

The Ca2+ activated Cl- channel TMEM16A (anoctamin 1; ANO1) is expressed in secretory epithelial cells of airways and intestine. Previous studies provided evidence for a role of ANO1 in mucus secretion. In the present study we investigated the effects of the two ANO1-inhibitors niclosamide (Niclo) and benzbromarone (Benz) in vitro and in vivo in mouse models for cystic fibrosis (CF) and asthma. In human CF airway epithelial cells (CFBE), Ca2+ increase and activation of ANO1 by adenosine triphosphate (ATP) or ionomycin was strongly inhibited by 200 nM Niclo and 1 µM Benz. In asthmatic mice airway mucus secretion was inhibited by intratracheal instillation of Niclo or Benz. In homozygous F508del-cftr mice, intestinal mucus secretion and infiltration by CD45-positive cells was inhibited by intraperitoneal injection of Niclo (13 mg/kg/day for 7 days). In homozygous F508del-cftr rats intestinal mucus secretion was inhibited by oral application of Benz (5 mg/kg/day for 60 days). Taken together, well tolerated therapeutic concentrations of niclosamide and benzbromarone corresponding to plasma levels of treated patients, inhibit ANO1 and intracellular Ca2+ signals and may therefore be useful in inhibiting mucus hypersecretion and mucus obstruction in airways and intestine of patients suffering from asthma and CF, respectively.

Generation of human TMEM16F-specific affibodies using purified TMEM16F

Introduction: TMEM16 family proteins are involved in a variety of functions, including ion transport, phospholipid scrambling, and the regulation of membrane proteins. Among them, TMEM16F has dual functions as a phospholipid scramblase and a nonselective ion channel. TMEM16F is widely expressed and functions in platelet activation during blood clotting, bone formation, and T cell activation. Despite the functional importance of TMEM16F, the modulators of TMEM16F function have not been sufficiently studied. Method: In this study, we generated TMEM16F-specific affibodies by performing phage display with brain-specific TMEM16F (hTMEM16F) variant 1 purified from GnTi- cells expressing this variant in the presence of digitonin as a detergent. Purified human TMEM16F protein, which was proficient in transporting phospholipids in a Ca2+-dependent manner in proteoliposomes, was coated onto plates and then the phage library was added to fish out TMEM16F-binding affibodies. For the validation of interaction between affibodies and TMEM16F proteins, ELISA, bio-layer interferometry, and size exclusion chromatography were conducted. Results and Discussion: As a result, the full sequences of 38 candidates were acquired from 98 binding candidates. Then, we selected 10 candidates and purified seven of them from E. coli expressing these candidates. Using various assays, we confirmed that two affibodies bound to human TMEM16F with high affinity. These affibodies can be useful for therapeutical and diagnostic applications of TMEM16F-related cancer and neurodegenerative diseases. Future studies will be required to investigate the effects of these affibodies on TMEM16F function.

Structural heterogeneity of the ion and lipid channel TMEM16F

Transmembrane protein 16 F (TMEM16F) is a Ca2+-activated homodimer which functions as an ion channel and a phospholipid scramblase. Despite the availability of several TMEM16F cryogenic electron microscopy (cryo-EM) structures, the mechanism of activation and substrate translocation remains controversial, possibly due to restrictions in the accessible protein conformational space. In this study, we use atomic force microscopy under physiological conditions to reveal a range of structurally and mechanically diverse TMEM16F assemblies, characterized by variable inter-subunit dimerization interfaces and protomer orientations, which have escaped prior cryo-EM studies. Furthermore, we find that Ca2+-induced activation is associated to stepwise changes in the pore region that affect the mechanical properties of transmembrane helices TM3, TM4 and TM6. Our direct observation of membrane remodelling in response to Ca2+ binding along with additional electrophysiological analysis, relate this structural multiplicity of TMEM16F to lipid and ion permeation processes. These results thus demonstrate how conformational heterogeneity of TMEM16F directly contributes to its diverse physiological functions.