Nucleoside Modifications in Novel Coronavirus SARS-CoV-2

Abstract

RNA viruses are rapidly evolving threats to human health and society on a global scale. In recent history the most notable RNA virus is the positive-sense single-stranded coronavirus SARS-CoV-2, which is responsible for the COVID-19 pandemic. While the urgency of the pandemic has subsided, continued research into the molecular biology of SARS-CoV-2 will provide insight into a family of viruses that continues to impact human health, which may be leveraged for antiviral therapeutics. Chemical modifications to RNA nucleosides have been found in SARS-CoV-2 RNA as a direct interaction between host proteins and viral RNA. These modifications can impact RNA stability, structure, function, and innate immune sensing, but their full effect on viral RNAs has not yet been elucidated. RNA viruses encode proteins that are essential for their lifecycle and not found in host cells, such as the viral RNA replication machinery. RNA-dependent RNA polymerases are responsible for replicating the viral RNA quickly and faithfully, yet they have high error rates that lead to rapid evolution of RNA viruses. SARS-CoV-2 transcriptional activity and regulation is therefore an important target for improving our understanding how coronaviruses evolve and interact with host organisms.

The SARS-CoV-2 replication machinery must encounter nucleoside modifications as it synthesizes new RNA using the viral RNA as a template. I developed a SARS-CoV-2 transcription assay that is capable of pre-steady state rates, which is essential to resolving how each modification impacts viral transcription at the level of single-nucleotide incorporation. Using this experimental design, I found that modification of adenosine with N6-methylation (m6A) and 2’O-methylation (Am) slows the rate of canonical nucleotide incorporation when present on the template RNA, with Am having a much higher rate of slowing. However, neither m6A nor Am completely inhibited processive transcription, as moving the transcription start site downstream of the modification was enough to overcome polymerase stalling. My results indicate that these modifications have greater impact when they are located close to regulatory sites such as the transcription start site or sites of transcription pausing and proofreading. I also found that the SARS-CoV-2 polymerase incorporates the nucleotide analog inhibitor 2’O-methyl UTP into product RNA with a similar reaction speed as nucleotide misinincorporation. This demonstrates a potential route for non-native modifications to enter viral RNA via human-designed antiviral intervention. Some such nucleotide analogs use such a mechanism to increase error rates of the viral polymerase during RNA replication. I further examined the effect of the nucleoside modifications m6A, inosine, and pseudouridine on SARS-CoV-2 polymerase accuracy, finding that all modifications mildly decreased the rate of nucleotide misincorporation when located in the RNA template. Notably, while characterizing the rate of canonical nucleotide incorporation of ATP vs pseudouridine, I discovered a defect compared to unmodified uridine, suggesting that nucleotide binding to the enzyme active site is rate-limiting at low concentrations of ATP. Considering pseudouridine specifically has been found at SARS-CoV-2 transcription regulatory sites, this suggests a possible role for regulation of transcription via slowing at pseudouridine sites. The research presented here focuses on basic principles of RNA viruses, the RNA-dependent RNA polymerases of coronaviruses specifically, and the impact of nucleoside modifications on the transcription activity of RNA viruses. All of these contribute to the growing body of knowledge around the coronavirus lifecycle, which is essential for preparation against future virus outbreaks.