The coronavirus (CoV) spike (S) glycoprotein is a trimeric class I fusion protein in which each protomer is cleaved into an S1 subunit, responsible for receptor binding, and an S2 subunit, which mediates membrane fusion. Because of its key role in viral entry, the S glycoprotein remains the primary target for vaccines and antibody therapeutics. Within S2, conserved structural epitopes—including the stem helix, fusion peptide, and central helix—represent attractive targets for broad CoV protection. Recent works have successfully engineered prefusion-stabilized S2 constructs for betacoronavirus S proteins, including for SARS-CoV-2 and MERS-CoV. However, it remains unclear whether similar strategies can be applied across other CoV genera. To expand the scope of S2 immunogen design, we targeted the S2 domains of two human endemic alphacoronaviruses (NL63 and 229E) and the porcine epidemic diarrhea virus (PEDV). Using a suite of computational protein design tools—including protein repair one-stop shop (PROSS) for global stability and expression optimization, rosetta void identification and packing (VIP) for targeted cavity-filling mutations, and DisulfidebyDesign 2.0 for disulfide engineering—we generated prefusion-stabilized S2 constructs predicted to retain the native trimeric architecture while presenting highly conserved S2 epitopes. Preliminary analysis has identified two constructs—PEDV S2-PD6 and PEDV S2-pan1—which have demonstrated improved expression yield or monodispersity, respectively, compared to the unstabilized S2 domain. Our results demonstrate that stabilization strategies initially developed to betacoronavirus S proteins can be extended to alphacoronaviruses . By integrating structure-guided design with systematic in silico mutational scanning, this work establishes a framework for probing antibody recognition of conserved S2 epitopes across divergent coronaviruses and a workflow to generate immunogens for pan-coronavirus vaccine development.
Rubio et al. (Sun,) studied this question.