Unlocking the puzzle: non-defining mutations in SARS-CoV-2 proteome may affect vaccine effectiveness

Abstract

Introduction: SARS-CoV-2 variants are defined by specific genome-wide mutations compared to the Wuhan genome. However, non-clade-defining mutations may also impact protein structure and function, potentially leading to reduced vaccine effectiveness. Our objective is to identify mutations across the entire viral genome rather than focus on individual mutations that may be associated with vaccine failure and to examine the physicochemical properties of the resulting amino acid changes.

Materials and methods: Whole-genome consensus sequences of SARS-CoV-2 from COVID-19 patients were retrieved from the GISAID database. Analysis focused on Dataset_1 (7,154 genomes from Italy) and Dataset_2 (8,819 sequences from Spain). Bioinformatic tools identified amino acid changes resulting from codon mutations with frequencies of 10% or higher, and sequences were organized into sets based on identical amino acid combinations.

Results: Non-defining mutations in SARS-CoV-2 genomes belonging to clades 21 L (Omicron), 22B/22E (Omicron), 22F/23A (Omicron) and 21J (Delta) were associated with vaccine failure. Four sets of sequences from Dataset_1 were significantly linked to low vaccine coverage: one from clade 21L with mutations L3201F (ORF1a), A27- (S) and G30- (N); two sets shared by clades 22B and 22E with changes A27- (S), I68- (S), R346T (S) and G30- (N); and one set shared by clades 22F and 23A containing changes A27- (S), F486P (S) and G30- (N). Booster doses showed a slight improvement in protection against Omicron clades. Regarding 21J (Delta) two sets of sequences from Dataset_2 exhibited the combination of non-clade mutations P2046L (ORF1a), P2287S (ORF1a), L829I (ORF1b), T95I (S), Y145H (S), R158- (S) and Q9L (N), that was associated with vaccine failure.

Discussion: Vaccine coverage associations appear to be influenced by the mutations harbored by marketed vaccines. An analysis of the physicochemical properties of amino acid revealed that primarily hydrophobic and polar amino acid substitutions occurred. Our results suggest that non-defining mutations across the proteome of SARS-CoV-2 variants could affect the extent of protection of the COVID-19 vaccine. In addition, alteration of the physicochemical characteristics of viral amino acids could potentially disrupt protein structure or function or both.

Keywords: SARS-CoV-2; conservation; mutations; proteome; vaccine.

Figure 1

Flowchart of the methodology used in this study. 1. Sequences and metadata of vaccinated and unvaccinated COVID-19 patients from Italy and Spain were downloaded from GISAID. 2. The SARS-CoV-2 genome sequences of the COVID-19 patients were aligned to the Wuhan reference genome. 3. The mutations of interest (MOIs) are then defined. 4. Next, the proteomes are obtained and those sequences that have the same combination of MOIs are grouped together. 5 and 6. Statistical analysis is now performed using a Chi-square test to identify sets of sequences between fully vaccinated and unvaccinated patients that are associated with vaccine failure. 7. The role of non-clade-defining mutations in the risk of vaccine effectiveness is then investigated by comparing populations. 8. Finally, the physicochemical properties of the MOIs in the sequence sets associated with vaccine failure are analyzed.

Figure 2

Mutations of interest (MOIs) identified for Dataset_1.

Figure 3

Mutations of interest (MOIs) identified for Dataset_2.

Figure 4

Genetic distances between amino acids per site within each sequence set of Dataset 1 and 2 obtained by averaging all sequence pairs, along with the standard error estimates The average distance was calculated using the Bootstrap method for variance estimation, with 1,000 bootstrap replicates. The p-distance model was used for the amino acid substitution type. Ambiguous positions were removed for each pair of sequences using the pairwise deletion option. The bars are organized based on the datasets and are color-coded according to sets of sequences that share the same MOIs.