AI Designed Mini Proteins to Block Middle Eastern Respiratory Syndrome Infection

Middle Eastern Respiratory Syndrome coronavirus (MERS-CoV), a zoonotic pathogen with a 36% fatality rate, lacks any approved vaccines or targeted treatments. MERS-CoV infects host cells via its spike (S) protein, which binds the DPP4 receptor. Mini proteins, AI-designed alpha-helical bundles, offer a promising strategy to combat coronaviruses as cost-effective, scalable, intranasal therapeutics targeting the receptor binding domain (RBD) of the S protein.Existing mini binders for MERS S protein are limited to a small hydrophobic pocket (HP) on the RBD, making them vulnerable to escape through minimal viral mutations. Additionally, designs using only alpha helices cannot target polar residues critical for receptor binding. Recent advances show that incorporating beta sheets into mini binders enables targeting of polar regions, expanding the range of residues available for binding.We propose designing mini binders that use both alpha helices and beta sheets to target the HP and proximal polar regions essential for MERS-CoV infection. Using RFdiffusion, we will generate mini binder backbones optimized for interactions with conserved hydrophobic and polar residues. ProteinMPNN will optimize binder sequences for stability and affinity, and AlphaFold3 will screen designs in silico. Promising candidates will be tested for binding affinity using biolayer interferometry and structurally characterized using cryo-EM.By targeting a broader range of residues critical for receptor binding, we aim to to create a therapeutic mini binder with increased affinity and resilience to MERS S RBD mutations. Even prior to a final therapeutic, the work will provide insights into MERS S protein structure, dynamics, and binding, enhancing pandemic preparedness for a potential MERS outbreak.