Comparative studies of protein dynamics in coronaviruses

Abstract

In coronavirus (CoV) infection, polyproteins (pp1a/pp1ab) are processed into non-structural proteins (nsps), which largely form the replication/transcription complex (RTC). The polyprotein processing and complex formation is critical and offers potential therapeutic targets. However, the interplay of polyprotein processing and RTC-assembly remains poorly understood. In this work, two key aspects were studied: The order of polyprotein processing by viral main protease Mpro and its influence on complex formation with the methyltransferase nsp16. Central to this investigation was the establishment of an approach to determine rate constants k from cleavage sites in structured CoV polyprotein based on native mass spectrometry (MS). We used this approach for a comprehensive analysis of polyprotein processing in four human CoVs: Severe acute respiratory syndrome CoV 1 and 2 (SARS-CoV-1 and -2), middle east respiratory syndrome CoV (MERS-CoV), and human CoV-229E (HCoV-229E). Our sensitive and precise native MS approach provided novel insights into polyprotein processing of nsp7-11 region revealing both conserved features and species-specific variations. The experimentally determined rate constants are put into perspective with a comprehensive analysis of primary sequences and structural models. Kinetic rate constants were determined for the four cleavage sites, CS7/8, CS8/9, CS9/10, and CS10/11, in all four viruses. Based on the presence of intact cleavage sites, processing species were assigned to the cleavage sites, which simplified the multi-reaction to a first-order reaction. This approach allowed us to extract cleavage site kinetics for each site and compare them between CoV species. The kinetics of multi-cleavage reaction revealed that the order and rate of processing are not conserved across species. Conversion rates at CS7/8 in all four CoVs were substantially slowed down compared to other cleavage sites. The primary structure that influences Mpro cleavage efficiency could not explain the different rates alone, suggesting a structural hindrance at CS7/8. AlphaFold prediction models indicated an α‑helical fold at this location, which reduces the cleavage efficiency of Mpro. In general, the AlphaFold predictions confirmed the experimental data and could provide structural rationale, though local confidence scores at the cleavage sites were low, potentially due to flexible loop regions. Notably, species-specific differences indicated that cleavage at CS10/11 is not essential. Binding experiments with SARS-CoV-2 nsp16, SARS-CoV-2 nsp10, and the nsp10-11 processing intermediate of MERS-CoV confirmed that cleavage at CS10/11 is not required for nsp16+10 complex formation. However, full cleavage at CS9/10 appears necessary for forming an active methyltransferase complex. A key advantage of our native MS approach is its ability to capture structural context while directly detecting processing intermediates and protein-protein interactions. This provides significant benefits over traditional peptide-based assays. Moreover, cleavage site kinetics were extracted including protein folding. Our findings offer new mechanistic insights into CoV polyprotein processing and complex assembly, which may inform future antiviral drug development strategies targeting these essential viral processes.