Discovery and Genomic Characterization of a 382-Nucleotide Deletion in ORF7b and ORF8 during the Early Evolution of SARS-CoV-2

Abstract

To date, limited genetic changes in the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genome have been described. Here, we report a 382-nucleotide (nt) deletion in SARS-CoV-2 that truncates open reading frame 7b (ORF7b) and ORF8, removing the ORF8 transcription regulatory sequence (TRS) and eliminating ORF8 transcription. The earliest 382-nt deletion variant was detected in Singapore on 29 January 2020, with the deletion viruses circulating in the country and accounting for 23.6% (45/191) of SARS-CoV-2 samples screened in this study. SARS-CoV-2 with the same deletion has since been detected in Taiwan, and other ORF7b/8 deletions of various lengths, ranging from 62 nt to 345 nt, have been observed in other geographic locations, including Australia, Bangladesh, and Spain. Mutations or deletions in ORF8 of SARS-CoV have been associated with reduced replicative fitness and virus attenuation. In contrast, the SARS-CoV-2 382-nt deletion viruses showed significantly higher replicative fitness in vitro than the wild type, while no difference was observed in patient viral load, indicating that the deletion variant viruses retained their replicative fitness. A robust antibody response to ORF8 has been observed in SARS-CoV-2 infection, suggesting that the emergence of ORF8 deletions may be due to immune-driven selection and that further deletion variants may emerge during the sustained transmission of SARS-CoV-2 in humans.IMPORTANCE During the SARS epidemic in 2003/2004, a number of deletions were observed in ORF8 of SARS-CoV, and eventually deletion variants became predominant, leading to the hypothesis that ORF8 was an evolutionary hot spot for adaptation of SARS-CoV to humans. However, due to the successful control of the SARS epidemic, the importance of these deletions for the epidemiological fitness of SARS-CoV in humans could not be established. The emergence of multiple SARS-CoV-2 strains with ORF8 deletions, combined with evidence of a robust immune response to ORF8, suggests that the lack of ORF8 may assist with host immune evasion. In addition to providing a key insight into the evolutionary behavior of SARS-CoV-2 as the virus adapts to its new human hosts, the emergence of ORF8 deletion variants may also impact vaccination strategies.

Keywords: COVID-19; ORF8; natural selection; phylogeny; vaccines.

FIG 1

Genomic organization and evolutionary relationships of human SARS-CoV-2 and SARS-CoV-2 Δ382. (A) Schematic diagram of the genomes of SARS-CoV-2 isolates Wuhan-Hu-1 (GenBank accession no. MN908947) and human CoV-19/Singapore/12/2020 (hCoV-19/Singapore/12/2020) Δ382 (GISAID: EPI_ISL_414378). (B) Magnification of genomic region (pink box in panel A) showing the 382-nt deletion in ORF7b and ORF8 (indicated by red line) in hCoV-19/Singapore/12/2020. Other ORF7b/8 deletions are indicated by blue lines as follows: a 138-nt deletion in hCoV-19/Australia/VIC671/2020 (EPI_ISL_426967), a 345-nt deletion in hCoV-19/Bangladesh/BARJ_CVASU_CTG_517/2020 (EPI_ISL_450344), and a 62-nt deletion in hCoV-19/Spain/COV001404/2020 (EPI_ISL_452497). Horizontal axes indicate the nucleotide position relative to Wuhan-Hu-1; open reading frames (ORFs) are indicated by solid colored arrows. Transcription regulatory sequences (TRSs) are indicated by yellow arrows. (C) Temporal phylogeny of 419 complete genomes inferred using an uncorrelated lognormal relaxed clock model in BEAST. Colored circles at the tips represent geographic locations of virus sampling. Colored triangles represent ORF7b/8 deletion variants. A fully labeled tree with Bayesian posterior probabilities indicated is presented in Fig. S4. Red isolate names indicate SARS-CoV-2 from Singapore with a 382-nt deletion as described in this study. Blue isolate names indicate a 382-nt deletion from Taiwan, whereas green isolate names indicate a 138-nt deletion from Australia. Node A represents the time to most common ancestor (TMRCA) for the lineage containing Δ382 viruses from Singapore and Taiwan, while node B represents the TMRCA of the clade containing Δ382 viruses from Singapore. Bayesian posterior probabilities of ≥0.95 are indicated at nodes. Scale bar represents time in years. Abbreviations: AUS, Australia; SP, Spain; BGD, Bangladesh.

FIG 2

Prevalence, viral antigen staining, and growth kinetics of human SARS-CoV-2 and SARS-CoV-2 Δ382. (A) Daily number of wild-type and Δ382 deletion variants of SARS-CoV-2 detected in Singapore based on PCR screening of 191 patient specimens. (B) Cellular and viral fluorescence staining of SARS-CoV-2 wild-type and Δ382 strains in Vero-E6 cells. Bright-field images were captured to visualize cytopathic effect. SARS-CoV-2 antigen was stained with COVID-19 convalescent-phase serum. Red, viral proteins; blue, DAPI (nuclei). (C) Growth curves of SARS-CoV-2 wild-type and Δ382 strains at a multiplicity of infection (MOI) of 0.01 in Vero-E6 cells. Virus titers were expressed as 50% tissue culture infectious doses (TCID₅₀)/ml and were plotted as means of results from three independent replicates and standard deviations. Error bars indicate standard errors of means. Significance was calculated by two-way ANOVA with Tukey’s multiple-comparison test.