The N protein of human beta coronaviruses (hCoV) is responsible for nucleocapsid assembly and other essential regulatory functions. Its N-terminal domain (N-NTD) interacts and melts the double-stranded transcriptional regulatory sequences (dsTRS), regulating the discontinuous subgenome transcription process, an essential feature for the virus cycle. There is little structural information on the specific binding of dsTRS and single-stranded RNA (TRS+ and TRS-) to the N-NTD [1]. To better describe the dsRNA melting and its role in the template switching, we measured the binding and the melting activity of TRSs of different ORF of the SARS-CoV-2 and non-specific RNA sequence. We used 3 different constructs, N-NTD, N-NTD plus the serine and arginine-rich motif (N-NTD-SR), and the N-NTD flanked by the two intrinsically disordered regions, the N-terminal, and the linker (N-NTR). We showed that the melting activity was only observed for the specific TRSs, in which the specificity depends on the sequence. The specificity is not determined by the differences in affinity of the different RNAs but by the structure and dynamics of the protein binding cleft. Moreover, we determined the importance of the IDR for the melting activity and binding affinity. Theoretical studies by our group showed the importance of a tweezer-like motion at the binding cleft for melting activity [2]. To approach this experimentally, we measured the CPMG relaxation dispersion of N-NTD of SARS-CoV-2, MERS-CoV, and HKU1-CoV and showed that they all present conformational equilibrium at the binding cleft. We determined that the exposure of hydrophobic surfaces at the open conformation is probably correlated to the binding and melting events. Altogether, our data contribute to the understanding of how promiscuous RNA binding proteins such as N protein can exert specific functions.
Acknowledgments: FAPERJ, CNPq, CAPES.