SARS-CoV-2 as Wolverine: Mutant Take Over

As COVID-19 envelops the globe, researchers have been sequencing the genome of SARS-CoV-2, the virus behind the pandemic. As of the end of June, more than 60,000 viral genomic sequences have been deposited with GISAID, a multi-national initiative to track and share genetic information from influenza viruses and now SARS-CoV-2. The speed and efficiency with which scientists have collected these sequences is unparalleled, and reflects not just the intensity of focus across the broader health communities on COVID-19 but also the increased availability of genome sequencing technologies. In this post, we will learn a little more about what these sequences tell us about SARS-CoV-2.

The good

First, some good news: unlike Influenza and HIV, SARS-CoV-2 is relatively slow to mutate. Its genome includes an enzyme that helps correct errors when the virus’s RNA polymerase is copying its genome, resulting in high fidelity replication within the host cell. This is good for us; the more stable the virus, the more likely we are to develop long-lasting immunity, and the more likely science is to develop a vaccine or treatment that will work across large populations.

This relatively slow rate of mutation also means that SARS-CoV-2 is similar in many ways to the 2003 SARS. The two viruses share about 70% of their genomes, with higher rates of similarity in some of the protein-coding regions of the genome. This genomic similarity (called sequence homology) has allowed us to much more rapidly understand and respond to COVID-19 than we otherwise would have, as scientists were able to work from what we already knew about 2003 SARS, giving us a headstart in identifying key characteristics of SARS-CoV-2, such as the fact that it binds to the host receptor ACE2.

A mutant spike

Even though the rate of mutation is slow, SARS-CoV-2 has indeed been mutating. One of the most prominent mutant strains of SARS-CoV-2 has been labeled “G614” or “the G clade,” in which the viral RNA has mutated from base A to G at position 22,403. This point mutation in the RNA has an effect on the proteins that it encodes: one amino acid in the SARS-CoV-2 spike has mutated from the original D (Aspartic acid) to G (Glycine). As scientists track these sequences globally, they can see that up through February, only 10% of the ~1K viral sequences had the G clade mutation. During March, in contrast, 67% of viral sequences had this mutation, and by mid-May, 78%.

Mapping this transition over time, researchers show that the G clade sequence became the dominant strain with surprising efficiency and consistency, indicating that the strain is more fit in an evolutionary sense. But why? What functional differences does the mutation convey? The answer here-- and even whether there is an answer-- is not yet clear. Korber et al. find that there is no significant difference in disease severity for the G clade virus when comparing outcomes of hospitalization. They don’t have data for the rate at which each virus strain leads to hospitalization, so that is left as an open question, but they do present evidence that the G clade viruses might be more infectious: the amount of viral particles collected per patient seems to be a higher with the G clade virus, but the evidence here is thin and limited by the available data on spread and patient outcomes dependent on viral strain.

The original viral sequence (orange) was rapidly replaced with the G clade strain (blue) over the course of months.

The original viral sequence (orange) was rapidly replaced with the G clade strain (blue) over the course of months.

Lab studies seem to back up the notion that the G clade virus is more infectious. Korber et al. looked at the effect of the mutation in pseudoviruses that were made to have the SARS-CoV-2 spike, and they found that the G614 mutation increased the number of viral particles they could recover in vitro. A preprint from Zhang et al. backs this theory up, showing in lab studies that the G clade spike is more easily incorporated into viral particles during assembly of a SARS-CoV-2-like pseudovirus, resulting in more efficient creation of viral particles that leave the cell to infect more host cells.

The mutation in the spike doesn’t seem to change binding to ACE2 significantly, but Zhang et al. argues it makes the assembly of the spike into viral particles more efficient.

The mutation in the spike doesn’t seem to change binding to ACE2 significantly, but Zhang et al. argues it makes the assembly of the spike into viral particles more efficient.

Mutations in the SARS-CoV-2 spike could have profound implications; the spike is critical for the virus’s ability to enter cells, and it is also one of the key targets of much of the ongoing vaccine work. Luckily for now, the researchers show that neutralizing antibodies against the original SARS-CoV-2 spike still identify the mutant spike, with slightly greater affinity even. The scientific community will have to continue to keep a close eye on the mutating spike to make sure it doesn’t leverage evolutionary pressure to escape both recognition by the immune system and our vaccine efforts.


Studies and articles talk a lot about the SARS-CoV-2 spike, and the work being done on mutation of the virus is no exception: the strain is named “the G clade” because of the D-to-G mutation in the spike. However, that’s only part of the story-- mutations tend to travel in groups, and the G clade has at least three other mutations that we understand even less well. One mutation is in a region that doesn’t encode proteins and one is a mutation that doesn’t change the amino acid produced, but the last is a mutation in the RNA polymerase.

You may remember that RNA viruses use the host cell’s ribosomes and other machinery to turn RNA into amino acids, but RNA viruses often carry their own nucleotide-chain-former called a polymerase to make copies of the RNA itself. The SARS-CoV-2 RNA polymerase is thus an important part of the virus’s ability to replicate and produce infectious viral particles, even though it is inside the virus, and therefore less likely to produce an antibody response or to have been studied like the spike has. Nonetheless, we know even less about what this mutation does than the mutation in the spike, so this polymerase mutant would definitely bear more study in the lab to understand whether there are functional differences.

All of these types of studies, including those discussed today, need to be heavily caveated, though: we know from the spread of the G clade strain that it is “winning” the evolutionary game. If scientists start with the question, “What makes this virus more fit to spread?” they will find plausible explanations, but they are also biased toward finding these explanations because the conclusion (that the G clade sequence is spreading faster) is already known. So while it’s easy to view the viral evolution with either optimism (less deadly viruses tend to spread faster, as they don’t kill off their host-factories; this must be a less deadly version of SARS-CoV-2 taking over!) or pessimism (the virus is mutating; it will become more deadly and thwart our ability to vaccinate against it!), we should continue to collect evidence both in the lab and in the world before drawing any conclusions.



Popular Posts