Conserved C-Terminal Tail Is Responsible for Membrane Localization and Function of Pseudomonas aeruginosa Hemerythrin.

Many bacteria have hemerythrin (Hr) proteins that bind O2, including Pseudomonas aeruginosa, in which microoxia-induced Hr (Mhr) provide fitness advantages under microoxic conditions. Mhr has a 23 amino-acid extension at its C-terminus relative to a well-characterized Hr from Methylococcus capsulatus, and similar extensions are also found in Hrs from other bacteria. The last 11 amino acids of this extended, C-terminal tail are highly conserved in gammaproteobacteria and predicted to form a helix with positively charged and hydrophobic faces. In cellular fractionation assays, wild-type (WT) Mhr was found in both membrane and cytosolic fractions, while a MhrW143* variant lacking the last 11 residues was largely in the cytosol and did not complement Mhr function in competition assays. MhrL112Y, a variant that has a much longer-lived O2-bound form, was fully functional and had a similar localization pattern to that of WT Mhr. Both MhrW143* and MhrL112Y had secondary structures, stabilities, and O2-binding kinetics similar to those of WT Mhr. Fluorescence studies revealed that the C-terminal tail, and particularly the fragment corresponding to its last 11 residues, was sufficient and necessary for association with lipid vesicles. Molecular dynamics simulations and subsequent cellular analysis of Mhr variants have demonstrated that conserved, positively charged residues in the tail are important for Mhr interactions with negatively charged membranes and the contribution of this protein to competitive fitness. Together, these data suggest that peripheral interactions of Mhr with membranes are guided by the C-terminal tail and are independent of O2-binding.

A Comparative Analysis of the Membrane Binding and Remodeling Properties of Two Related Sorting Nexin Complexes Involved in Autophagy.

The sorting nexin (SNX) proteins, Atg20 and Atg24, are involved in nonselective autophagy, are necessary for efficient selective autophagy, and are required for the cytoplasm-to-vacuole transport pathway. However, the specific roles of these proteins in autophagy are not well understood. Atg20 and Atg24 each contain a Phox homology domain that facilitates phosphoinositide binding. They also each contain an SNX-Bin/Amphiphysin/Rvs domain that forms a cup-shaped dimer, capable of binding to curved membranes and remodeling those membranes in some cases. Atg20 and Atg24 form two distinct complexes, an Atg24/Atg24 homodimer and an Atg20/Atg24 heterodimer. Despite the presence of Atg24 in both complexes, it is currently unclear if these complexes have different membrane binding and remodeling properties. Therefore, in this study, we explored the membrane binding and shaping properties of these two dimeric complexes. We found that Atg24/Atg24 and Atg20/Atg24 have distinct membrane binding preferences. Both dimers recognized membranes containing phosphatidylinositol 3-phosphate [PI(3)P] and phosphatidylinositol 3,5-bisphosphate, but Atg20/Atg24 bound to a broader array of liposomes, including those lacking phosphorylated phosphatidylinositol. In addition, we discovered that while both complexes bound to autophagosomal-like liposomes containing at least 5% PI(3)P, Atg20/Atg24 was capable of binding to autophagosomal-like liposomes lacking PI(3)P. Lastly, we observed that the Atg20/Atg24 heterodimer tubulates PI(3)P-containing and autophagosomal-like liposomes, but the Atg24/Atg24 homodimer could not tubulate these liposomes. Our findings suggest that these two dimers contain distinct membrane binding and shaping properties.

A new GUV-based assay to reconstitute membrane tethering in vitro demonstrates the vesicle tethering ability of the selective autophagy scaffold Atg11.

Membrane tethering is essential for the generation of organelle contact sites and to anchor incoming vesicles to their target membranes before vesicle fusion. During autophagy in yeast, tethering of 30 nm vesicles to cargo is one of the first steps in the generation of the isolation membrane that engulfs the cargo to be degraded. While membrane tethering is critical for cellular function, many of the current biochemical techniques to assay for membrane tethering rely on indirect readouts and are limited in their ability to monitor protein localization at sites of tethering. As such, we developed a fluorescence-microscopy-based GUV liposome tethering assay (GLT) to directly visualize membrane tethering and monitor protein localization at tethering sites simultaneously. We initially used GLT with engineered membrane tethers to demonstrate the ease of use, versatility and sensitivity of the assay. We also demonstrated that the selective autophagy scaffolding protein Atg11 can bind, and tether negatively charged membranes but is unable to bind GUVs mimicking the charge of the outer mitochondrial membrane. Atg11 instead requires the selective autophagy receptor Atg32 to be recruited to mitochondrial membranes. Lastly, we demonstrate that Atg11 bound to Atg32 on GUVs can tether negatively charged vesicles to GUVs. Collectively, our results reconstitute one of the first steps during the initiation of mitochondrial autophagy and highlight the versatility of GLT to study a range of membrane tethering events biochemically.

Atg23 is a vesicle-tethering protein.

Small 30-nm vesicles containing the integral membrane protein Atg9 provide the initial membrane source for autophagy in yeast. Atg23 is an Atg9 binding protein that is required for Atg9 vesicle trafficking but whose exact function is unknown. In our recent paper, we explored the function of Atg23 using an approach combining cellular biology and biochemistry on purified protein. We determined that Atg23 is an elongated dimer spanning 320 Å in length. We also demonstrated that Atg23 is a membrane-binding and -tethering protein. Furthermore, we identified a series of amino acids residing in a putative coiled-coil region that when mutated prevent Atg23 dimer formation resulting in a stable Atg23 monomer. Last, we demonstrated that when monomeric Atg23 is expressed in yeast lacking Atg23, this leads to a loss of Atg23 puncta, a reduction in Atg9 puncta, a reduction in nonselective autophagy and a complete block in the cytoplasm-to-vacuole targeting (Cvt) pathway.

Dimerization-dependent membrane tethering by Atg23 is essential for yeast autophagy.

Eukaryotes maintain cellular health through the engulfment and subsequent degradation of intracellular cargo using macroautophagy. The function of Atg23, despite being critical to the efficiency of this process, is unclear due to a lack of biochemical investigations and an absence of any structural information. In this study, we use a combination of in vitro and in vivo methods to show that Atg23 exists primarily as a homodimer, a conformation facilitated by a putative amphipathic helix. We utilize small-angle X-ray scattering to monitor the overall shape of Atg23, revealing that it contains an extended rod-like structure spanning approximately 320 Å. We also demonstrate that Atg23 interacts with membranes directly, primarily through electrostatic interactions, and that these interactions lead to vesicle tethering. Finally, mutation of the hydrophobic face of the putative amphipathic helix completely precludes dimer formation, leading to severely impaired subcellular localization, vesicle tethering, Atg9 binding, and autophagic efficiency.

Protein-Protein Interactions on Membrane Surfaces Analysed Using Pull-Downs with Supported Bilayers on Silica Beads.

Discovery-based proteomics workflows that identify novel interactors rely on immunoprecipitations or pull-downs with genetically tagged bait proteins immobilized on appropriate matrices. But strategies to analyse protein interactions on a diffusible-membrane surface combined with the practical ease of pull-downs remain unavailable. Such strategies are important to analyse protein complexes that mature in composition and stability because of diffusion-based encounter between participant proteins. Here, we describe a generic pull-down strategy to analyse such complexes using chelating lipid-containing supported bilayers formed on silica beads. These templates can display desired His-tagged bait proteins on a diffusible-membrane surface. Using clathrin-mediated endocytosis as a paradigm, we find that the clathrin-binding adaptor protein epsin1 displayed as bait on these templates pulls down significantly higher amounts of clathrin from brain lysates than when immobilized on conventional matrices. Together, our results establish the potential of such templates as superior matrices for analysing protein-protein interactions and resultant complexes formed on membrane surfaces.

SMrT Assay for Real-Time Visualization and Analysis of Clathrin Assembly Reactions.

Clathrin-mediated endocytosis manages the vesicular transport of the bulk of membrane proteins from the plasma membrane and the trans-Golgi network. During this process, discrete sets of adaptor proteins recognize specific classes of membrane proteins, which recruit and assemble clathrin lattices on the membrane. An important determinant to the success of this vesicular transport reaction is the intrinsic ability of adaptors to polymerize clathrin on a membrane surface. Adaptor-induced clathrin assembly has traditionally been analyzed using static electron microscopy-based approaches. Here, we describe a methodology to follow adaptor-induced clathrin assembly in real-time using fluorescence microscopy on a facile model membrane assay system of supported membrane tubes (SMrT). Results from such assays can be conveniently run through routine image analysis procedures to extract kinetic parameters of the clathrin assembly reaction.