Emerging Vaccine-Breakthrough SARS-CoV-2 Variants

Abstract

The surge of COVID-19 infections has been fueled by new SARS-CoV-2 variants, namely Alpha, Beta, Gamma, Delta, and so forth. The molecular mechanism underlying such surge is elusive due to the existence of 28 554 unique mutations, including 4 653 non-degenerate mutations on the spike protein. Understanding the molecular mechanism of SARS-CoV-2 transmission and evolution is a prerequisite to foresee the trend of emerging vaccine-breakthrough variants and the design of mutation-proof vaccines and monoclonal antibodies. We integrate the genotyping of 1 489 884 SARS-CoV-2 genomes, a library of 130 human antibodies, tens of thousands of mutational data, topological data analysis, and deep learning to reveal SARS-CoV-2 evolution mechanism and forecast emerging vaccine-breakthrough variants. We show that prevailing variants can be quantitatively explained by infectivity-strengthening and vaccine-escape (co-)mutations on the spike protein RBD due to natural selection and/or vaccination-induced evolutionary pressure. We illustrate that infectivity strengthening mutations were the main mechanism for viral evolution, while vaccine-escape mutations become a dominating viral evolutionary mechanism among highly vaccinated populations. We demonstrate that Lambda is as infectious as Delta but is more vaccine-resistant. We analyze emerging vaccine-breakthrough comutations in highly vaccinated countries, including the United Kingdom, the United States, Denmark, and so forth. Finally, we identify sets of comutations that have a high likelihood of massive growth: [A411S, L452R, T478K], [L452R, T478K, N501Y], [V401L, L452R, T478K], [K417N, L452R, T478K], [L452R, T478K, E484K, N501Y], and [P384L, K417N, E484K, N501Y]. We predict they can escape existing vaccines. We foresee an urgent need to develop new virus combating strategies.

Keywords: COVID-19; SARS-CoV-2; comutations; infectivity; vaccine-breakthrough; vaccine-resistant.

Figure 1:

Most significant RBD mutations. a The 3D structure of SARS-CoV-2 S protein RBD and ACE2 complex (PDB ID: 6M0J). The RBD mutations in ten variants are marked with color. b Illustration of the time evolution of 455 ACE2 binding-strengthening RBD mutations (blue) and 228 ACE2 binding-weakening RBD mutations (red). The $x$-axis represents the date and the $y$-axis represents the natural log of frequency. There has been a surge in the number of infections since early 2021. c BFE changes of RBD complexes with ACE2 and 130 antibodies induced by 75 significant RBD mutations. A positive BFE change (blue) means the mutation strengthens the binding, while a negative BFE change (red) means the mutation weakens the binding. Most mutations, except for vaccine-resistant Y449H and Y449S, strengthen the RBD binding with ACE2. Y449S and K417N are highly disruptive to antibodies.

Figure 2:

Properties of RBD co-mutations. a Illustration of RBD 2 co-mutations with a frequency greater than 90. b Illustration of RBD 3 co-mutations with a frequency greater than 30. c Illustration of RBD 2 co-mutations with a frequency greater than 20. Here, the $x$-axis lists RBD co-mutations and the $y$-axis represents the predicted total BFE change between S RBD and ACE2 of each set of RBD co-mutations. The number on the top of each bar is the AI-predicted number of antibody and RBD complexes that may be significantly disrupted by the set of RBD co-mutations, and the color of each bar represents the natural log of frequency for each set of RBD co-mutations. (Please check the interactive HTML files in the Supporting Information S2.2.4 for a better view of these plots.)

Figure 3:

a 2D histograms of antibody disruption count and total BFE changes for RBD 2 co-mutations (unit: kcal/mol). b 2D histograms of antibody disruption count and total BFE changes (unit: kcal/mol) for RBD 3 co-mutations. c 2D histograms of antibody disruption count and total BFE changes (unit: kcal/mol) for RBD 4 co-mutations. d The histograms of total BFE changes (unit: kcal/mol) for RBD co-mutations. e The histograms of the natural log of frequency for RBD co-mutations. f The histograms of antibody disruption count for RBD co-mutations. In figures a, b, and c, the color bar represents the number of co-mutations that fall into the restriction of $x$-axis and $y$-axis. The reader is referred to the web version of these plots in the Supporting Information S2.2.2 and S2.2.3.

Figure 4:

Illustration of the time evolution of 2, 3, and 4 co-mutations on the S protein RBD of SARS-CoV-2 from January 01, 2021, to July 31, 2021, in 12 COVID-19 devastated countries: the United Kingdom (UK), the United States (US), Denmark (DK), Brazil (BR), Germany (DE), Netherlands (NL), Sweden (SE), Italy (IT), Canada (CA), France (FR), India (IN), and Belgium (BE). The $y$-axis represents the natural log frequency of each RBD co-mutation. The top 5 high-frequency co-mutations in each country are marked by red, blue, green, yellow, and pink lines. The cyan line is for the RBD co-mutation [L452Q, F490S] on the Lambda variant, and the other co-mutations are marked by light grey lines. Notably, there are two blues lines in the panel of FR due to the same frequency of [K417N, E484K, N501Y] and [E484K, N501Y]. (Please check the interactive HTML files in the Supporting Information S2.2.1 for a better view of these plots.)

Figure 5:

a Illustration of genome sequence data pre-processing and BFE change predictions. b Comparison of experimental CT-P59 IC₅₀ fold change (reduction) and predicted BFE changes induced by mutations L452R and T478K. c Comparison of predicted BFE changes and relative luciferase units for pseudovirus infection changes of ACE2 and S protein complex induced by mutations L452R and N501Y.