Infection requires that SARS-CoV-2 slip into a cell, unnoticed by multiple alarms, and camouflage itself to resemble cellular genes and hijack the cell’s machinery. To this end, the virus is equipped with numerous means to thwart cellular alarms and to disguise its messenger RNAs to resemble that of the cell. Tripping the cellular alarms leads to the destruction of the virus early on. Failure to successfully mimic viral RNA and the viral message is destroyed before essential viral components can be made.
SARS-CoV-2 genome encodes 30 proteins. Virtually all of them participate in one way or another the intricate dance of deception and subterfuge. Blocking one more of these essential functions with target-specific drugs will strip Covid’s camouflage, preventing virus replication.
A recent study by Park et al. unveils the unexpected details of one such trick, the formation of viral RNA’s five prime (5’) cap. In so doing, the authors provide a road map to guide the development of multiple anti-viral drugs to prevent and treat SARS-CoV-2 infections.
All messenger RNAs require a cap, a structure at the 5 prime termini (the beginning) of the messenger RNAs. The structure of the cap is shown in figure 2. The newly formed messenger RNA is modified first by adding phosphorylated guanosine, then adding methyl groups to the newly added guanosine and the terminal 5′ terminal residue of the messenger RNA. Absent the cap, the messenger is not recognized by the synthetic protein machinery and is degraded. Moreover, RNAs that lack the methyl group on the initial nucleotide of the RNA trip cellular alarms that initiate a signaling cascade to activate the innate immune response.
The cellular messenger RNAs are made and capped in the cell’s nucleus before export to the cytoplasm for translation. The nuclear enzymes include an RNA triphosphatase to remove the terminal phosphate of the nascent messenger RNA), a guanylyltransferase which transfers a guanosine monophosphate from GTP to form the cap core, a guanine-N7-methyltransferase which adds a methyl group to the terminal guanosine at the N7 position and a nucleoside 2-O-methyltransferase which adds a methyl group to the 2’ hydroxyl group of the first nucleotide of the messenger RNA.
As mentioned, all these reactions occur in the cell nucleus. These are not available to the messenger RNAs of SARS-CoV-2, as virus replication takes in the cytoplasm. Viral RNA never enters the nucleus.
Prior work on other coronaviruses and SARS-CoV-2 found three of the viral proteins encoded by the initial long open reading frame orf1ab. Nsp13 encodes an RNA triphosphatase necessary to remove the 5′ most phosphate of the nascent RNA. N7 methylation of the added guanine to yield is accomplished by nsp14. Nsp16 (in a complex stabilized by nsp10) is required to add the final 2’-O-methyl group to the terminal nucleotide.
The process of coronavirus messenger RNA synthesis is unique. The positive-strand genomic RNA is initially copied from a nested set of complementary negative strands. This nested set of negative-strand RNAs serves as a template for synthesizing the positive strand messenger RNAs, all of which contain an identical 5’ leader sequence.
The SARS-CoV-2 replication transcription complex includes several different enzymes. These include nsp12, the RNA-dependent RNA polymerase required for RNA synthesis, a helicase to open double-stranded RNA structures for copying, an error-correcting exonuclease through which nascent RNAs must pass, nsp9, and what are described as stabilizing proteins, nsp7, nsp8, and nsp10. In addition to a structure common to all viral RNA-dependent RNA polymerases, the SARS-CoV-2 polymerase contains an additional amino-terminal domain called NiRAN (Nidovirus RNA dependent RNA Polymerase Associated Nucleotidyltransferase)
Park et al. uncovered an unexpected pathway when used purified viral proteins to reconstruct how the SARS-CoV-2 cap is formed. They expected to find that by analogy with cellular processes, the first step of cap formation would be the removal of the 5’ terminal phosphate of the nascent RNA to yield ppA…mRNA. Note that the initial nucleotide of all SARS-CoV-2 messenger RNAs is adenosine (A).
That is not what they found. Using a short RNA (5’’pppA…n10) as a surrogate messenger RNA, they discovered that the very first step in cap formation is the covalent attachment of the RNA to the nsp9 protein. Further investigation revealed that attachment occurs to the amino-terminal asparagine of nsp9. The phosphate closest to adenosine (the alpha phosphate) attaches to the backbone amino group, the terminal asparagine of nsp9. Park et al. call the process of covalent enzymatic attachment of RNA to a protein RNAylation.
The authors also show that attachment of the nascent RNA to nsp9 is catalyzed by NiRAN, the amino-terminal domain nsp12 protein. Mutations at positions 73 and 218 that inactivate the phosphotransferase activity of NiRAN fail to attach RNA to nsp9. By contrast, a mutation at 760 of nsp12 that inactivates the polymerase activity does not affect RNAylation of nsp9. NiRAN but not polymerase activity is essential for cap formation.
How then is the terminal guanine diphosphate added to the cap? Park et al. find that NiRAN binds guanosine diphosphate (GDP). The addition of guanosine diphosphate occurs via NiRAN catalyzed transfer of the asparagine-RNA complex to guanosine yield the cap’s GppA…mRNA core structure. The subsequent addition of an N7-methyl group to guanosine is followed by O-methylation of the terminal adenosine is mediated by nsp14 and the nsp16/10 to yield the final cap structure.
Park et al. also report the 3-dimensional structure of a complex that contains the nsp12/nsp9 together with the stabilizing proteins nsp7 and nsp8. They find that the amino terminus of nsp9 inserts into the active site of the NiRAN domain of nsp12, providing a structural explanation of how the covalent addition of the nascent RNA to nsp9 occurs.
Park et al. provide two models by which the nascent messenger RNA chain encounters nsp9. In one model, all the reactions occur on a single nsp12/nsp9 complex. This model requires the nascent RNA to twist and bend during synthesis. According to this model, on exit from the polymerase, the growing messenger must first transit the exonuclease, then twist and bend to enter NiRAN. For this reason, they suggest that it is possible and even likely that RNAylation occurs on a second nsp12/nsp9 complex.
This work explains how the SARS cap is formed outside of the nucleus. The authors mention several well-known targets for antiviral drugs, the polymerase domain of nsp12 (the target of remdesivir), as well as nsp14 and nsp16/nsp10. They also identify two new targets, the NiRAN domain of nsp12 and the amino terminus of nsp9. Single drugs or combinations of drugs that inhibit these enzymes will likely prevent and treat SARS-CoV-2 infections effectively.
I favor efforts to discover drugs that inhibit the nsp12/nsp9 reactions over those that interfere with the activity of the two methylases. The viral methylase proteins closely resemble their cellular counterparts. For that reason, it may be difficult to find drugs that inhibit the viral enzymes that are not toxic to cells as well. By contrast, RNAylation of nsp9 is unique to coronaviruses and, therefore, would seem to be an excellent selective target. Park et al. validate this strategy as they find that mutations in NiRAN that inactive RNAyation also prevent replication, as is an asparagine at the amino terminus nsp9.
The paper Park et al. is an elegant demonstration of the power of a novel approach to anti-viral drug development—specifically to the development of drugs that target viral proteins that either camouflage the virus on entry and/or block the innate immune response. Figure 7 highlights the plethora of such targets we identified, which are some among many in the SARS-CoV-2 genome.