SARS-CoV-2: from its discovery to genome structure, transcription, and replication

Abstract

SARS-CoV-2 is an extremely contagious respiratory virus causing adult atypical pneumonia COVID-19 with severe acute respiratory syndrome (SARS). SARS-CoV-2 has a single-stranded, positive-sense RNA (+RNA) genome of ~ 29.9 kb and exhibits significant genetic shift from different isolates. After entering the susceptible cells expressing both ACE2 and TMPRSS2, the SARS-CoV-2 genome directly functions as an mRNA to translate two polyproteins from the ORF1a and ORF1b region, which are cleaved by two viral proteases into sixteen non-structural proteins (nsp1-16) to initiate viral genome replication and transcription. The SARS-CoV-2 genome also encodes four structural (S, E, M and N) and up to six accessory (3a, 6, 7a, 7b, 8, and 9b) proteins, but their translation requires newly synthesized individual subgenomic RNAs (sgRNA) in the infected cells. Synthesis of the full-length viral genomic RNA (gRNA) and sgRNAs are conducted inside double-membrane vesicles (DMVs) by the viral replication and transcription complex (RTC), which comprises nsp7, nsp8, nsp9, nsp12, nsp13 and a short RNA primer. To produce sgRNAs, RTC starts RNA synthesis from the highly structured gRNA 3' end and switches template at various transcription regulatory sequence (TRS_B) sites along the gRNA body probably mediated by a long-distance RNA-RNA interaction. The TRS motif in the gRNA 5' leader (TRS_L) is responsible for the RNA-RNA interaction with the TRS_B upstream of each ORF and skipping of the viral genome in between them to produce individual sgRNAs. Abundance of individual sgRNAs and viral gRNA synthesized in the infected cells depend on the location and read-through efficiency of each TRS_B. Although more studies are needed, the unprecedented COVID-19 pandemic has taught the world a painful lesson that is to invest and proactively prepare future emergence of other types of coronaviruses and any other possible biological horrors.

Fig. 1

Genome structure and coding potentials of human coronaviruses. The viral genome is a single-stranded, positive-sense RNA with a cap (grey circle) at the 5′-end and a poly-A tail (A30-60) at the 3′ end. The genome encodes 16 non-structural proteins (ORF1a → nsp1-11 and ORF1b → nsp12-16) from the left three-fourth of the genome, and 4–5 structural proteins (S, spike; E, envelope; M, membrane; N, nucleocapsid; HE, hemagglutinin esterase) and various number of accessory proteins (numbered boxes) from the right one-fourth of the genome

Fig. 2

Coronavirus genome replication and transcription. Diagram showing the key steps in coronavirus entry (1), initial translation of incoming viral +gRNA to express viral non-structural proteins (nsp1-16) (2), genome replication in double-membrane vesicles (DMVs), continuous transcription of gRNA through a −gRNA-intermediate by viral replication and transcription complex (RTC) (3), generation of sgRNA by discontinuous transcription RTC (4), the expression of structural and accessory proteins from +sgRNA (S, spike; M, membrane; E, envelope; N proteins) (5), and virion assembly and release (6)

Fig. 3

A proposed model of viral RNA transcription and template switch during SARS-CoV-2 infection. A Continuous 5′–3′ transcription of viral genomic +gRNA leads to synthesis of the full-length, negative-sense viral genomic RNA (−gRNA) (left). Because RTC-mediated RNA transcription starts from the highly structured viral gRNA 3′ end, this transcription often leads to discontinuous 5′–3′ transcription by proposed template switch (right). Through interactions between transcription regulatory sequences (TRS) located in the leader (TRS_L) and the genome body (TRS_B), the template switch results in the production of viral subgenomic RNAs (−sgRNAs). B Diagram of SARS-CoV-2 genome with predicted ORFs (colored boxes) and TRS (smaller red boxes) upstream of individual ORFs. Above are the canonical TRS_L-dependent junctions detected in the individual sgRNAs from SARS-CoV-2-infected cells by RNA-seq, with the junction reads corresponding to the sgRNA encoding N protein being the most abundant. Below are the TRS_B-independent interactions of TRS_L (red) and non-TRS dependent (blue) junctions detected by RNA-seq with unknown function [35, 38]

Fig. 4

Structures of SARS-CoV-2 replication and transcription complex (RTC). A A composite structure of RTC from three PDB coordinates, 7CXM (architecture of nsp7, nsp8 X2, nsp12 and nsp13 X2 bound to RNA template and primer), 6XEZ (the ADP·AlF3, bound in the nsp13 helicase active site), and 7CYQ (nsp9 associated with nsp12 and GDP in the active site of RNA capping). The RNA template pieces bound to nsp13 and nsp12 are not connected and would be pulled by the two enzymes in opposite directions as indicated by the yellow double arrowheads. B, C Zoom-in views of RTC bound to inhibitors, Favipiravir (PDB: 7AAP), Remedisivir (RMP) (PDB: 7B3B), and Suramin (PDB: 7D4F). RdRP (nsp12) is shown in grey in B, C, the three inhibitors are in distinct colors. With several SO₄ groups mimicking the phosphate backbone of RNA, two Sumarin molecules (cyan) compete for the RNA template and primer binding. Remedisivir (blue) is already incorporated in the RNA primer strand at − 3 position. Favipiravir RTD (magenta) occupies the incoming nucleotide position, but the phosphates are in a non-productive conformation. The active site residues are shown in pink-red sticks and Mg²⁺ ions are shown as green spheres

Fig. 5

The expression of sgRNAs during human coronavirus infection. On the left are the diagrams of SARS-CoV-2 (A), hCoV-OC43 (B) and hCoV-NL63 (C) genomes and their coding potentials. Individual sgRNAs (lines) with a 5′ leader (small red box) obtained through the template switch are illustrated below and named by their corresponding proteins encoded. The viral gRNA (vgRNA) is generated by continuous transcription of the entire viral genome. On the right are the sgRNA expression profiles in African monkey kidney Vero E6 cells infected for 24 h with SARS-CoV-2 (A), 189 h with hCoV-NL63 (C), or human colorectal adenocarcinoma HCT-8 cells infected for 48 h with hCoV-OC43 (B). The sgRNAs detected by Northern blot analysis of total RNA extracted from infected cells using the individual antisense probes specific to each viral N gene. The Northern blot gel of SARS-CoV-2 sgRNA in A was modified with permission from a reference [38]