Evolution of viral quasispecies during SARS-CoV-2 infection

Abstract

Objectives: Studies are needed to better understand the genomic evolution of the recently emerged severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This study aimed to describe genomic diversity of SARS-CoV-2 by next-generation sequencing (NGS) in a patient with longitudinal follow-up for SARS-CoV-2 infection.

Methods: Sequential samples collected between January 29th and February 4th, 2020, from a patient infected by SARS-CoV-2 were used to perform amplification of two genome fragments-including genes encoding spike, envelope, membrane and nucleocapsid proteins-and NGS was carried out with Illumina® technology. Phylogenetic analysis was performed with PhyML and viral variant identification with VarScan.

Results: Majority consensus sequences were identical in most of the samples (5/7) and differed in one synonymous mutation from the Wuhan reference sequence. We identified 233 variants; each sample harboured in median 38 different minority variants, and only four were shared by different samples. The frequency of mutation was similar between genes and correlated with the length of the gene (r = 0.93, p = 0.0002). Most of mutations were substitution variations (n = 217, 93.1%) and about 50% had moderate or high impact on gene expression. Viral variants also differed between lower and upper respiratory tract samples collected on the same day, suggesting independent sites of replication of SARS-CoV-2.

Conclusions: We report for the first time minority viral populations representing up to 1% during the course of SARS-CoV-2 infection. Quasispecies were different from one day to the next, as well as between anatomical sites, suggesting that in vivo this new coronavirus appears as a complex and dynamic distributions of variants.

Keywords: Infection follow-up; Minority variants; NGS; Quasispecies; SARS-CoV-2.

Fig. 1

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genome diversity during infection. (A) Genome coverage (y axis) according to nucleotide position (x axis). (B) Distribution (x axis) and frequency (y axis) of the 233 intra-sample viral variants identified. Each sample is represented by the same colour in (A) and (B), and the impact of mutations on gene expression is represented by a different symbol (low: a rhombus, moderate: a square, high: a circle).

Fig. 2

Distribution of mutation frequency and correlation with gene length or viral load. (A) HeatMap representing the frequency and distribution of the mutations and their impact on gene expression between the different genes. (B) Linear regression line between the number of viral variant (x axis) and the gene length (y axis). (C) HeatMap representing the frequency and distribution of the mutations and their impact on gene expression between the different samples. (D) Linear regression line between the number of frameshift and stop codons (x axis) and the viral load expressed in cycle threshold (Ct) value (y axis). Viral variants by gene: S, n = 87; N, n = 49; ORF3a, n = 25; M, n = 23; ORF7a, n = 10; E, n = 8; ORF6, n = 6; ORF7b, n = 5; ORF8, n = 5; ORF10, n = 4. Scale on the right of (A) and (C) represents the frequency in percentages, with the largest value in dark and the lowest value in light.

figs1