Upon receptor recognition, the viral envelope will fuse with the host cell membrane or by receptor mediated endocytosis by the spike protein. Once inside the host cytoplasm, the viral RNA is released with ORF1a and ORF1ab translated to make 2 polyproteins: PP1a and PP1ab. In order to make 2 distinct polyproteins, the viral RNA has a slippery sequence (5’-UUUAAAC-3’) and a RNA pseudoknot secondary structure that will cause frameshift activity in the ribosomes, analogous to attenuation in bacterial Tryptophan operon. When the ribosome fails to unwind the psuedoknot in time, it will pause at the slippery sequence and move back by one nucleotide, causing a -1 frameshift before resolving the secondary structure. This upstream frameshift is what gives rise to the second polyprotein PP1ab. Interestingly, this delay in psuedoknot resolution occurs 25% of the time to create a stoichiometric 3 to 1 ratio of PP1a and PP1ab. This ratio might be the ideal stoichiometric balance for efficient viral replication and assembly (Baranov et al. 2005).

After translation of PP1a and PP1ab, these precursors are then cleaved by proteases to form 16 nsps that will associate into the RNA Replicase-transcriptase complex localized in bilayer vesicles derived from the ER to produce the genomic (–) RNAs. The (¬–) RNAs are transcribed and replicated discontinuously to give a diverse pool of subgenomic (+) mRNAs/RNAs. During this process, there are many cis-regulatory elements that strictly regulate different stages of RNA synthesis. For instance, the 5’UTR of the genome consists of 7 stem-loop structures that invade into the replicase 1a gene (Raman et al., 2003). Meanwhile, the 3’UTR end consists of the bulged stem-loop and the pseudoknot sequences that overlap. In other words, both secondary structures cannot form simultaneously, implying that different structures are formed in a time-dependent manner to modulate different stages of replication and translation (Liu et al., 2001).