Use of Sanger protocols to identify variants of concern, key mutations and track evolution of SARS-CoV-2

Abstract

Vaccination and the emergence of SARS-CoV-2 variants mark the second year of the pandemic. Variants have amino acid mutations at the spike region, a viral protein central in the understanding of COVID-19 pathogenesis and vaccine response. Variants may dominate local epidemics, as Gamma (P.1) in Brazil, emerging in 2020 and prevailing until mid-2021. Different obstacles hinder a wider use of Next-Generation Sequencing for genomic surveillance. We describe Sanger based sequencing protocols: i) Semi-nested RT-PCR covering up to 3.684 kb (>96 %) spike gene; ii) One-Step RT-PCR for key Receptor Binding Domain (RBD) mutations (codons 417-501); iii) One-Step RT-PCR of partial N region to improve genomic capability. Protocols use leftovers of RNA extracted from nasopharyngeal swabs for quantitative RT-PCR diagnosis; with retro-transcribed DNA sequenced at ABI 3500 using dye termination chemistry. Analyses of sequences from 95 individuals (late 2020/early 2021) identified extensive amino acid variation, 57 % with at least one key mutation at the Receptor Binding Domain, with B.1.1.28 lineage most prevalent, followed by Gamma and Zeta variants, with no Delta variant observed. The relatively low cost and simplicity may provide an accessible tool to improve surveillance of SARS-CoV-2 evolution, monitor new variants and vaccinated breakthroughs.

Keywords: Brazil; COVID-19; Gamma; Genomic surveillance; SARS-CoV-2; Variants.

Fig. 1

Schematic representation of PCR amplified fragments using three protocols described in the methodology. Illustrative sketch of the A) long protocol to obtain most of the spike region of SARS-CoV-2, with primers annealing region and the expected size of the amplified fragment; B) short protocol, to obtain part of the spike region of SARS-CoV-2, covering the Receptor Binding Domain and the; C) protocol to obtain the nucleocapsid (N) region of SARS-CoV-2.

Fig. 2

Variant classification of SARS-CoV-2 sequences obtained in our study, most collected in January and February 2021. Sequences were further classified according to the absence or presence of additional mutations, not associated with the variant definition.

Fig. 3

Bayesian phylogenetic tree inference implemented with BEAST v.1.7.4 under GTR (G + I) with concatenated Spike and Nucleocapsid regions of SARS-CoV-2 partial sequences joined through BioEdit, along with references sequences obtained at the GISAID platform. The sequences were aligned with ClustalW. A coronavirus HKU1 sequence was used as outgroup.

Fig. 4

Frequency of SARS-CoV-2 mutations detected in spike protein according to the variant classification of sequences. Solid Light Blue: Unique mutation; Solid Dark Blue: Existing mutations; Dotted Red: P.1 variant mutations; Dashed Purple: Common mutations to P.1 and P.2 variants; Grid Orange: common mutations to P.1, P.2, and B.1.1.28. Mutations A845S, N925K, G72V, and L229F were present at P.1 variant; T547I, A623S, A899S, and P812S at P.2 variant; P621S, E654V, A771S, T1117I, A67V, T95I, D215G, A222V, Q239R, R273S, Q677H, N658T, S689I and M153T at B.1.1.28 variant.