SARS-CoV-2 exhibits intra-host genomic plasticity and low-frequency polymorphic quasispecies

Abstract

In December 2019, an outbreak of atypical pneumonia (Coronavirus disease 2019 -COVID-19) associated with a novel coronavirus (SARS-CoV-2) was reported in Wuhan city, Hubei province, China. The outbreak was traced to a seafood wholesale market and human to human transmission was confirmed. The rapid spread and the death toll of the new epidemic warrants immediate intervention. The intra-host genomic variability of SARS-CoV-2 plays a pivotal role in the development of effective antiviral agents and vaccines, as well as in the design of accurate diagnostics. We analyzed NGS data derived from clinical samples of three Chinese patients infected with SARS-CoV-2, in order to identify small- and large-scale intra-host variations in the viral genome. We identified tens of low- or higher- frequency single nucleotide variations (SNVs) with variable density across the viral genome, affecting 7 out of 10 protein-coding viral genes. The majority of these SNVs (72/104) corresponded to missense changes. The annotation of the identified SNVs but also of all currently circulating strain variations revealed colocalization of intra-host as well as strain specific SNVs with primers and probes currently used in molecular diagnostics assays. Moreover, we de-novo assembled the viral genome, in order to isolate and validate intra-host structural variations and recombination breakpoints. The bioinformatics analysis disclosed genomic rearrangements over poly-A / poly-U regions located in ORF1ab and spike (S) gene, including a potential recombination hot-spot within S gene. Our results highlight the intra-host genomic diversity and plasticity of SARS-CoV-2, pointing out genomic regions that are prone to alterations. The isolated SNVs and genomic rearrangements reflect the intra-patient capacity of the polymorphic quasispecies, which may arise rapidly during the outbreak, allowing immunological escape of the virus, offering resistance to anti-viral drugs and affecting the sensitivity of the molecular diagnostics assays.

Keywords: COVID-19 epidemic; Genomic rearrangements; Intra-host variability; Molecular diagnostics; Quasispecies; SARS-CoV-2.

Fig. 1

Intra – host SNVs: (A) Intra host SNV frequency vs sequencing read depth (X coverage) in the corresponding alignment position. (B) Venn diagram representing unique and common SNVs isolated from the three patients (C) Boxplot of intra-host SNVs frequency vs. SNV type – synonymous, missense, nonsense (stop gained) (low, moderate and high impact respectively). Average values are in red rhombs. (D) Intra-host SNVs frequency vs. all seven genes affected (ORF1ab, S, ORF3a, ORF6, ORF7a, ORF8, N). Average values are in red rhombs. (E) Density histogram of intra-host SNVs isolated from all patients (total number of SNVs / 100 bp - blue bars) and average sequencing read depth (X coverage – green line), across the SARS-CoV-2 genome map (genes in orange, 5′ and 3′ untranslated regions in light blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Fig. 2

Truncated map of SARS-CoV-2 genome illustrating a subset of intra-host (blue lines) and globally collected, isolate-specific SNVs (orange lines) with respect to the genomic targets of molecular diagnostics assays (red arrows – primers, red bars - probes). Three intra-host variants (orange triangles), and two strain specific variants (Wuhan/IVD-HB-04/2020 and Chongqing/YC01/2020 - red triangles), are colocalized with the RdRP_SARSr probe (15,474 T > G), the 2019-nCoV_N1 forward primer (28,291 C > T), the HKU-N reverse primer (28,971 A > G) and the 2019-nCoV-N2 probe (29,188 T > C and 29,200 C > T). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Fig. 3

Alignment of the de novo assembled contigs on the genomic map (bottom). Concordantly aligned contigs (correct or gapped) are in green, while discordantly aligned contigs are in red. Sequencing read depth (X coverage) across the genome (blue histograms) and relative % GC content (green line) is presented for each sample. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Fig. 4

Recombination events in S gene. Samples (A) SRR10903401 and (B) SRR10903402. Alignments of the de novo assembled contigs with respect to the reference genome (MN 975262). Donor – acceptor palindrome sequences are indicated in green bars. Raw, non-duplicated NGS reads, validating the recombination event, are represented below the corresponding contig. (C): Prediction of the secondary structure of the genomic region spanning the rearrangement breakpoint (100 bases upstream and 100 bases downstream). The corresponding donor- acceptor sequences, exposed in internal loops, are indicated in green bars. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).