SARS-CoV-2 Variants are Selecting for Spike Protein Mutations that Increase Protein Stability

Highlighted by 

Soni Mohapatra¹ and Alexander Solis^{*, 1}

*Undergraduate Researcher, (1) Johns Hopkins University, Baltimore, USA

Background 

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused huge disruption across the globe, resulting in over 4 million deaths as of August 2021 (1). The highly contagious nature of SARS-CoV-2 can be attributed to its molecular structure. In particular, the structural spike proteins which radiate from the protein center bind with a high affinity to ACE2 (human angiotensin-converting enzyme-2), which acts as a receptor for the virus and is expressed in a variety of human organs (2). Errors during RNA replication of the virus as well as evolutionary pressures has led to the rise of mutant strains as the pandemic spreads. Mutations in spike proteins could lead to increased ACE2 receptor binding, changes to cleavage/glycosylation sites (3), increased protein stability, and/or better infectivity. A thorough understanding of the mutational landscape of the spike proteins, as well as the ability to predict deadly mutations, could therefore be paramount for vaccine and therapeutic drug production, as well as to inform public health measures.

Changes in Gibbs free energy of unfolding (ΔΔG) between the mutant and the original protein, is a thermodynamic parameter for predicting how mutations in a protein affect the stability of the protein. ΔΔG = ΔG_mutant − ΔG_wild‑type, where ΔG is the difference between the free energies of the folded and unfolded states. Typically, ΔΔG < 0 indicates that the mutations are stabilizing whereas a ΔΔG > 0 suggests that mutations are destabilizing. As such, it can be used to obtain insights regarding possible future mutations and their effects on the protein as a whole. Mutations of SARS-CoV-2 spike proteins that increase protein stability could lead to the survival of the protein in different environmental conditions and therefore result in a longer lifespan of the protein.

In this preprint, the authors calculated the respective ΔΔG for all possible missense mutations for the SARS-CoV-2 spike protein as a means of understanding how spike protein stability has and will evolve over time.

Key findings

Evolutionary enrichment of mutations that stabilize spike proteins  

The authors calculated the ΔΔG for all possible 19440 missense mutations of the spike protein using the most recently published structure (PDBID: 6XVV). Mutations with ΔΔG < -1 kcal/mol and ΔΔG > 2.5 kcal/mol were defined as strongly stabilizing and strongly destabilizing mutants, respectively. Only 3.9% of these ‘background’ missense mutations were found to be strongly stabilizing. They compared these ‘background’ mutations to those that are found in the WHO “Variants of Concern” (VOC) and “Variants of Interest” (VOI) categories. Interestingly, ΔΔG of mutations occurring in the above two categories are significantly lower than the overall ΔΔG of the ‘background’ mutations. 4 of the 17 mutations observed in the VOC had a ΔΔG <= ~ -1 kcal/mol. Additionally, it was noted that none of the mutations present in the VOC category were strongly destabilizing. In stark contrast, 34% of the ‘background’ mutations are strongly destabilizing. Taken together, their data suggested that stabilizing mutations in the spike proteins are enriched in the WHO variants category compared to all ‘background’ missense mutations.

They also calculated the ΔΔG distribution of all the different mutations present in 10 variants belonging to VOC and VOI categories. 7 out of the 10 variants had a mean ΔΔG distribution below that of the mutational background (Fig. 1), which provided further support to the claim that stabilizing mutations are enriched in these categories compared to mutational background.

Fig. 1: Distribution of ΔΔG for mutations observed in the WHO VOC and VOI categories vs. background mutations (Taken from Fig. 3 of the preprint, made available by a CC-BY 4.0 license)

Predicting the potential evolutionary order of variants and gaining foresight into traits of future variants

The authors also used this calculation of ΔΔG to gather insight into the evolutionary order in which the mutations appeared in the variants. They calculated the ΔΔG for every possible combination of mutations in the different variants. For the Alpha variant, the gradual addition of mutations to the original spike protein resulted in lowering of ΔΔG and therefore increased stabilization of the variant compared to the original protein. In contrast, the resulting ΔΔG from the combination of the mutations in the Beta variant is destabilizing. The authors noted that the calculated ΔΔG is however still lower than that derived from a combination of mutations that could lead to the most destabilizing variant.

This analysis provided insight into evolutionary mechanisms for selective mutations that are employed by the variants. In their analysis, the authors find that mutations which lead to a more stable variant are generally favored over others. However, it was also alluded to that some combinations of mutations, which include those that are destabilizing, may ultimately create an evolutionarily superior product than the stabilizing mutations This is due to the fact that mutations have varying effects when expressed and can provide advantageous qualities, such as increased affinity to binding sites, better survival in different environmental conditions, amongst others. They therefore conclude that variant evolution does not only favor stabilizing mutations, but also those which carry properties that can positively manipulate a protein’s function.

Why we like the preprint 

In this article, the authors make use of an accessible and easy-to-understand concept of ΔΔG as a means of comprehending the complex mechanics of protein stability in SARS-CoV-2. This was done through performing calculations of ΔΔG to measure the stability of present variants as well as using ΔΔG to understand the evolutionary order of different mutations in the variants. What we particularly enjoyed about the article is that the calculation of a simple biophysical parameter, ΔΔG makes understanding the evolution of SARS-CoV-2 accessible to a wider audience, but also allows one to predict the uprising of future deadly variants.

Questions for authors

1. Is it possible to use the ΔΔG calculations to guide vaccine development?
2. Are there mutations in the ‘background’ mutational scan that have a ΔΔG lower than the mutations occurring in the WHO categories? Are those the mutations we should watch out for in the future?

References

1.  “WHO Coronavirus (COVID-19) Dashboard.” World Health Organization, World Health Organization, 16 July 2021, covid19.who.int/.
2.  Zheng, Jun. “SARS-CoV-2: an Emerging Coronavirus that Causes a Global Threat.” International journal of biological sciences vol. 16,10 1678-1685. 15 Mar. 2020, doi:10.7150/ijbs.45053
3. Tortorici, M. A.; Veesler, D. Structural Insights into Coronavirus Entry. In Advances in Virus Research; 2019; Vol. 105, pp 93–116. ISSN 0065-3527. ISBN 9780128184561. https://doi.org/10.1016/bs.aivir.2019.08.002.

 

doi: https://doi.org/10.1242/prelights.30208

Read preprint (No Ratings Yet)