Investigating protein-membrane interactions using native reverse micelles constructed from naturally sourced lipids.

Advancing the study of membrane associated proteins and their interactions is dependent on accurate membrane models. While a variety of membrane models for high-resolution membrane protein study exist, most do not reflect the diversity of lipids found within biological membranes. In this work, we have developed native reverse micelles (nRMs) formulated with lipids from multiple eukaryotic sources, which encapsulate proteins and enable them to interact as they would with a biological membrane. Diverse formulations of nRMs using soy lecithin, porcine brain lipids, or bovine heart lipids combined with n-dodecylphosphocholine were developed and characterized by dynamic light scattering and 31 P-NMR. To optimize protein encapsulation, ubiquitin was used as a standard and protein NMR verified minimal changes to its structure. Peripheral membrane proteins, which bind reversibly to membranes, were encapsulated and include glutathione peroxidase 4 (GPx4), phosphatidylethanolamine-binding protein 1 (PEBP1), and fatty acid binding protein 4 (FABP4). All three proteins showed anticipated interactions with the membrane-like inner surface of the nRMs as assessed by protein NMR. The nRM formulations developed here allow for efficient, high-resolution study of membrane interacting proteins up to and beyond ~21 kDa, in a more biologically relevant context compared to other non-native membrane models. The approach outlined here may be applied to a wide range of lipid extracts, allowing study of a variety of membrane associated proteins in their specific biological context.

Membrane-Mimicking Reverse Micelles for High-Resolution Interfacial Study of Proteins and Membranes.

Despite substantial advances, the study of proteins interacting with membranes remains a significant challenge. While integral membrane proteins have been a major focus of recent efforts, peripheral membrane proteins (PMPs) and their interactions with membranes and lipids have far less high-resolution information available. Their small size and the dynamic nature of their interactions have stalled detailed interfacial study using structural methods like cryo-EM and X-ray crystallography. A major roadblock for the structural analysis of PMP interactions is limitations in membrane models to study the membrane recruited state. Commonly used membrane mimics such as liposomes, bicelles, nanodiscs, and micelles are either very large or composed of non-biological detergents, limiting their utility for the NMR study of PMPs. While there have been previous successes with integral and peripheral membrane proteins, currently employed reverse micelle (RM) compositions are optimized for their inertness with proteins rather than their ability to mimic membranes. Applying more native, membrane-like lipids and surfactants promises to be a valuable advancement for the study of interfacial interactions between proteins and membranes. Here, we describe the development of phosphocholine-based RM systems that mimic biological membranes and are compatible with high-resolution protein NMR. We demonstrate new formulations that are able to encapsulate the model soluble protein, ubiquitin, with minimal perturbations of the protein structure. Furthermore, one formula, DLPC:DPC, allowed the encapsulation of the PMPs glutathione peroxidase 4 (GPx4) and phosphatidylethanolamine-binding protein 1 (PEBP1) and enabled the embedment of these proteins, matching the expected interactions with biological membranes. Dynamic light scattering and small-angle X-ray scattering characterization of the RMs reveals small, approximately spherical, and non-aggregated particles, a prerequisite for protein NMR and other avenues of study. The formulations presented here represent a new tool for the study of elusive PMP interactions and other membrane interfacial investigations.

Electrostatic Drivers of GPx4 Interactions with Membrane, Lipids, and DNA.

Glutathione peroxidase 4 (GPx4) serves as the only enzyme that protects membranes through the reduction of lipid hydroperoxides, preventing membrane oxidative damage and cell death through ferroptosis. Recently, GPx4 has gained attention as a therapeutic target for cancer through inhibition and as a target for inflammatory diseases through activation. In addition, GPx4 isoforms perform several distinct moonlighting functions including cysteine cross-linking of protamines during sperm cell chromatin remodeling, a function for which molecular and structural details are undefined. Despite the importance in biology, disease, and potential for drug development, little is known about GPx4 functional interactions at high resolution. This study presents the first NMR assignments of GPx4, and the electrostatic interaction of GPx4 with the membrane is characterized. Mutagenesis reveals the cationic patch residues that are key to membrane binding and stabilization. The cationic patch is observed to be important in binding headgroups of highly anionic cardiolipin. A novel lipid binding site is observed adjacent to the catalytic site and may enable protection of lipid-headgroups from oxidative damage. Arachidonic acid is also found to engage with GPx4, while cholesterol did not display any interaction. The cationic patch residues were also found to enable DNA binding, the first observation of this interaction. Electrostatic DNA binding explains a mechanism for the nuclear isoform of GPx4 to target DNA-bound protamines and to potentially reduce oxidatively damaged DNA. Together, these results highlight the importance of electrostatics in the function of GPx4 and illuminate how the multifunctional enzyme is able to fill multiple biological roles.

Identification of Phenazine-Based MEMO1 Small-Molecule Inhibitors: Virtual Screening, Fluorescence Polarization Validation, and Inhibition of Breast Cancer Migration.

Phosphorylation-dependent protein-protein interactions play a significant role in biological signaling pathways; therefore, small molecules that are capable of influencing these interactions can be valuable research tools and have potential as pharmaceutical agents. MEMO1 (mediator of ErbB2-cell driven motility) is a phosphotyrosine-binding protein that interacts with a variety of protein partners and has been found to be upregulated in breast cancer patients. Herein, we report the first small-molecule inhibitors of MEMO1 interactions identified through a virtual screening platform and validated in a competitive fluorescence polarization assay. Initial structure-activity relationships have been investigated for these phenazine-core inhibitors and the binding sites have been postulated using molecular dynamics simulations. The most potent biochemical inhibitor is capable of disrupting the large protein interface with a KI of 2.7 µm. In addition, the most promising phenazine core compounds slow the migration of breast cancer cell lines in a scratch assay.

Analysis of MEMO1 Binding Specificity for ErbB2 Using Fluorescence Polarization and Molecular Dynamics Simulations.

ErbB2 signaling pathways are linked to breast cancer formation, growth, and aggression; therefore, understanding the behavior of proteins associated with these pathways as well as regulatory factors that influence ErbB2 function is essential. MEMO1 is a redox active protein that is shown to associate with phosphorylated ErbB2 and mediate cell motility. We have developed a fluorescence polarization assay to probe the interaction between MEMO1 and an ErbB2-derived peptide containing a phosphorylated tyrosine residue. This interaction is shown to be pH-dependent and stronger with longer peptides as would be expected for protein-protein interactions. We have quantitatively mapped the binding interface of MEMO1 to the peptide using the fluorescence polarization assay and molecular dynamics simulations. We have confirmed that phosphorylation of the peptide is essential for binding and through mutagenesis have identified residues that contribute to favorable interactions. Our results highlight the importance of the protein-protein interactions of MEMO1 that complement the oxidase activity. In the future, these studies will provide a method for screening for selective modulators of MEMO1, which will allow for additional biological investigations.