Dengue virus NS4A protein dynamics drive membrane deformation

PROJECT ABSTRACT/SUMMARY Mosquito-borne flaviviruses impose a significant global health burden by causing diseases such as hemorrhagic fever, encephalitis, and congenital birth defects. Dengue virus (DENV) poses the largest risk among these viruses, with nearly 400 million infections estimated to occur annually. Secondary DENV infections are especially severe and are increasing in frequency as the endemic range expands due to climate change. Despite this impact, no antiviral drugs are currently approved for the treatment or prevention of DENV infection. A critical barrier to therapeutic development is our limited understanding of the molecular mechanisms governing DENV replication. RNA viruses, including DENV, create specialized replication factories to concentrate enzymes and substrates required for genome replication. The formation of these compartments involves the physical bending of the endoplasmic reticulum (ER) membrane to generate invaginations that hide viral replication intermediates. DENV non-structural protein 4A (NS4A) provides a major driving force for these membrane rearrangements. However, the precise mechanism by which NS4A generates the force to bend the ER membrane remains unclear despite its potentials as a therapeutic target. Our long-term goal is to elucidate the molecular mechanisms by which NS4A promotes the formation of virus replication compartments. The long-standing model in the field is that the oligomerization of ER-resident NS4A accelerates the dynamics of NS4A, which facilitates membrane bending and catalyzes the formation of replication factories. Testing this model computationally or experimentally has been limited by 1) computational cost associated with large systems and long simulation times, and 2) limitations in producing small hydrophobic membrane proteins in large quantities for biophysical assays. To overcome these challenges, we will employ novel molecular dynamics (MD) simulation techniques and coarse- grained (CG) models to study this system over timescales required to observe membrane bending. Simulation predictions will be tested experimentally using cell-free synthesis of NS4A and in vitro membrane bending assays. In Aim 1, we will establish the impacts of NS4A on ER membrane dynamics using a combined simulation and experimental approach. Aim 1a focuses on dissecting the effects of a single NS4A monomer on ER membrane dynamics. We will simulate a lipid bilayer that reflects the composition and temperature of the ER in human and mosquito membranes. NS4A monomers will be oriented in the membrane based on previous biochemical assays. Aim 1b extends these investigations to include NS4A dimers and trimers. These simulations will pinpoint specific NS4A domains and amino acids that have the greatest impact on membrane bending and their interactions with lipids. Aim 1c will focus on experimentally validating the computational predictions derived from these studies using nanodisc technology and giant unilamellar vesicles (GUVs). Ultimately, our work will catalyze therapeutic development by identifying small molecules capable of disrupting NS4A-mediated membrane dynamics and engineering proteins to bend membranes on demand.