(672f) Computational Design of Binding Proteins for Sars-Cov-2 Using Aubie

Between December 2019 and April 1, 2020, Coronavirus Disease 2019 (COVID-19), caused by the novel coronavirus Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), has infected nearly one million people and caused tens of thousands of deaths globally. In early 2020, vital information has been discovered about the genome, structural proteins and infection mechanism of SARS-CoV-2. Using this knowledge, a mammoth research effort to develop COVID19 therapeutics has begun but has not yet resulted in an approved treatment or vaccine. Hence, there is an urgent global need for the quick development of effective therapeutics to combat SARS-CoV-2.

Antibodies are a major class of therapeutic proteins which can be used against SARS-CoV-2. Antibodies are Y-shaped proteins used by the vertebrate adaptive immune system to bind to any foreign molecule (i.e. antigen) that enters the body. Currently, the global market for antibody applications such as therapeutics and biosensors is valued at over $100 billion. Over the past two decades, significant progress has been made towards the development of computational tools for the engineering and de novo design of antibodies to bind a target antigen. The use of such computational antibody engineering tools, in combination with experimental procedures, can expedite the development of COVID-19 therapeutics.

Over the past two years, our lab has developed a computational program called the Algorithm for Ultra-rapid Binding Interactions Engineering (AUBIE) for the high-speed design of binding proteins towards a specific region (i.e. epitope) of a target antigen. AUBIE identifies protein loops from the Protein Data Bank (PDB) that can be positioned into the binding regions of the selected scaffold to have strong interactions, such as hydrogen bonds and pi-pi stacking, with the targeted epitope. While AUBIE can design antibodies, the most commonly used binding protein, it can also engineer binding loops onto other smaller sized scaffolds such as the 10^th Type 3 Fibronectin (10Fn3) Domain, nanobodies and anticalins. Computational tools like AUBIE can be used to quickly design binding proteins that can specifically bind a functionally significant virus epitope and hence neutralize the virus.

Like other coronaviruses, SARS-CoV-2 uses the receptor binding domain (RBD) of the spike protein to bind to a host cell surface receptor. The experimental atomic structures of the SARS-CoV-2 RBD, unbound and in complex with its receptor Angiotension Converting Enzyme 2 (ACE2), are now available in the PDB. We have used AUBIE to design neutralizing antibodies, nanobodies and 10Fn3 domains to bind to the ACE2 binding site in the SARS-CoV-2 RBD. Furthermore, we have also used AUBIE’s ability to target residue-specific interactions to design broadly neutralizing binding proteins for mutated strains of SARS-CoV-2. This has been done using two approaches which differ in their epitope residue/atom selection for interaction formation. The first approach targets the epitope backbone atoms and the second approach targets the residues that form strong interactions with ACE2. This presentation will describe the predicted binding metrics, such as binding energies and shape complementarity, of the AUBIE designs. Experimental testing of these designs is currently planned and results will be shared as part of the presentation.