Interactions of SARS-CoV-2 N-protein

National Institute Of Biomedical Imaging And Bioengineering, Bethesda

Investigators

Linked publications & trials

Abstract

Our first goal was to clarify conflicting reports by other laboratories about the oligomeric state of SARS-CoV-2 N-protein in solution. As gold standard methods we have applied sedimentation velocity analytical ultracentrifugation and dynamic light scattering, and found the solution state of nucleic-acid free protein to be a dimer with only ultra-weak higher oligomerization. Next, we have examined the impact of binding of oligonucleotides of different lengths on the N-protein state. We found that oligonucleotides with six or more bases in length are sufficient for high-affinity binding. While hexanucleotides bind N-protein dimers leading to a well-defined 2:2 protein/oligonucleotide complexes, decanucleotides induce significant further reversible protein self-association, and 20mers additionally allow cross-linking of N-protein dimers. The latter leads to strong indefinite self-association promoting liquid-liquid phase separation (LLPS) of mixed N-protein/nucleic acid complexes. Based on published results by other laboratories studying LLPS of N-proteins, it is likely that such phase-separated droplets are part of the mechanism of viral packaging. Based on circular dichroism spectroscopy, we have detected a significant reduction of disorder in the N-protein conformation after nucleic acid binding. In extensive studies of the temperature-dependence of the N-protein size distribution and conformation by dynamic light scattering and circular dichroism, we have observed temperature stabilized intermediate sized clusters prior to the formation of phase separated droplets. These are accompanied by further conformational changes in the protein chain that appear to promote helical structures. These data provide a foundation for further studies of the nature of the assembly intermediates, and to examine to what extent different N-protein domains, residues, and potentially their post-translational modifications contribute to viral assembly. While our initial experiments have utilized a commercial protein source, we have now set up bacterial expression in our laboratory that we are currently optimizing for yield and purity. This will allow us to compare the assembly behavior of different protein constructs. We have already embarked on the study of N-terminal and C-terminal domains using protein preparations from collaborating laboratories. In addition, to better understand the formation of phase-separated droplets, we have initiated light microscopy experiments to observe the morphology and dynamics of supramolecular objects as a function of solution condition, protein construct, and nucleic acid ligands. The goal is to connect the macroscopic behavior with molecular-scale self-association properties that we can measure by sedimentation velocity analytical ultracentrifugation and other biophysical tools.