Designing a Universal Coronavirus Vaccine

Part II: SARS-CoV-2 Variants

Photo by Fusion Medical Animation on Unsplash

Welcome back to my segment on designing a universal coronavirus vaccine. You can read Part I below, where I discuss the need for a universal coronavirus vaccine and review the spike protein of human coronaviruses.

Designing a Universal Coronavirus Vaccine

I left off feeling good about the S2 subunit of the spike protein, given that it was reasonably conserved among human coronaviruses — meaning a new coronavirus is likely to maintain that conservation in its S2 subunit. However, what wasn’t conserved during the initial analysis was the spike protein S1 subunit, which contains the receptor-binding domain (RBD). For SARS-CoV-2, the receptor binding domain attaches to the ACE2 receptor in humans, but other coronaviruses can use different receptors to infect humans. So, to attack the spike protein S1 subunit problem, let’s look at the SARS-CoV-2 variants of concern (VOC) and variants of interest (VOI) since they provide real-world evidence of viral adaptation and evolution. You can see the WHO SARS-CoV-2 variant tracking site here.

Table 1. Notable mutations for SARS-CoV-2 variants of concern/interest. Not a comprehensive list of mutations, but those with some potential clinical relevance. delta = deletion, ins = insertion.

Figure 1. Selected mutations in the spike protein from SARS-CoV-2 variants of concern/interest. Created with BioRender.com

Looking at Table 1 and Figures 1 & 2, it’s apparent there are several mutations that most variants of concern/interest have, such as D614G (not listed above), N501Y, P681H, and E484K. When thinking about a universal coronavirus vaccine, these kinds of common mutations across different variants suggest an important viral adaptation, so I would include those in the S1 subunit part of the vaccine. I would also stick with the concept of choosing to insert a particular mutation into the vaccine if multiple variants have that mutation in the S1 subunit, but revert to the original SARS-CoV-2 reference sequence if there’s any doubt on which amino acid to use in a particular location.

For all of the MSAs, I used Unipro UGENE version 41. “Unipro UGENE: a unified bioinformatics toolkit” Okonechnikov; Golosova; Fursov; the UGENE team Bioinformatics 2012 28: 1166–1167

Figure 2. Conserved regions of the SARS-CoV-2 variants of concern/interest spike protein S1 subunit. In descending order, the MSA covers regions 1–98, 130–226, 234–310 and 570–663 of the S protein. Each colored block represents an individual amino acid. The more in agreement the alignment is, the better. Note: after residue 144, the alignment numbers may be off by +1 or +4 from other references thanks to insertions from the Mu and Omicron BA.1 variants.

So, while there are numerous mutations in the spike protein when looking at variants of concern/interest, most of them are “one-offs” and might not be too important. However, when mutations start to show up in multiple variants, it usually means a selective advantage of some sort and something worth keeping (from a virus perspective). When considering the spike protein analysis of all HCoVs (Part I) and the variants of concern/interest in Part II, I would argue this is a solid start to designing a universal coronavirus vaccine — using the conserved regions of all HCoVs for the S2 subunit and the conserved regions of notable variants for the S1 subunit. Now, I don’t expect that strategy to result in a perfect vaccine. Still, it is a starting point that can be fine-tuned through various experiments and looking at antibody response and neutralization of HCoVs. Finally, in Part III, I’ll look at adding additional coronavirus surface proteins to the vaccine to hopefully generate an additional immune response separate from the spike protein — more shots on goal might give an increased chance at universal coronavirus protection.

Some websites of interest are included below.

https://covariants.org/variants

https://covdb.stanford.edu/page/mutation-viewer/#alpha

https://viralzone.expasy.org/9556