Emergence and phenotypic characterization of the global SARS-CoV-2 C.1.2 lineage

Abstract

Global genomic surveillance of SARS-CoV-2 has identified variants associated with increased transmissibility, neutralization resistance and disease severity. Here we report the emergence of the PANGO lineage C.1.2, detected at low prevalence in South Africa and eleven other countries. The initial C.1.2 detection is associated with a high substitution rate, and includes changes within the spike protein that have been associated with increased transmissibility or reduced neutralization sensitivity in SARS-CoV-2 variants of concern or variants of interest. Like Beta and Delta, C.1.2 shows significantly reduced neutralization sensitivity to plasma from vaccinees and individuals infected with the ancestral D614G virus. In contrast, convalescent donors infected with either Beta or Delta show high plasma neutralization against C.1.2. These functional data suggest that vaccine efficacy against C.1.2 will be equivalent to Beta and Delta, and that prior infection with either Beta or Delta will likely offer protection against C.1.2.

Fig. 1. C.1.2 a highly mutated SARS-CoV-2 variant.

a Phylogenetic tree of 6,192 global sequences (1,991 sequences from South Africa), including variants of concern (VOC), variants of interest (VOI), and the C.1.2 lineage, colored by Nextstrain clade (shown in the key; VOCs and VOIs are in color, other clades in greyscale) and scaled by divergence (number of mutations). The C.1.2 lineage (purple) forms a distinct, highly mutated cluster within clade 20D. b Regression of root-to-tip genetic distances against sampling dates for sequences belonging to lineage C.1.2 sampled either in South Africa (solid purple) or in other countries (white) indicating that the C.1.2 sequences evolved in a clock-like manner (correlation coefficient = 0.43, R² = 0.18). The regression gradient is an estimate of the rate of sequence evolution, which is 3.04 × 10⁻³ nucleotide substitutions/site/year. Regression lines are shown with error buffers (shaded area) representing 90% confidence intervals. c Full genome representation of C.1.2 showing lineage defining mutations (seen in ≥50% of C.1.2 assigned sequences), with those in the spike (green) colored according to functional regions, including the N-terminal domain (NTD, blue), receptor binding domain (RBD, red), receptor binding motif (RBM, purple), subdomain 1 and 2 (SD1 or SD2, gray) and the cleavage site (S1/S2, yellow). Figure generated by covdb.stanford.edu. d A phylogenetic tree highlighting the introduction of spike mutations in the different sub-clades of the C.1.2 lineage. The tree is colored by month of collection, with tip symbols indicating country of collection, as indicated in the key (Figure generated from a Nextstrain build of global C.1.2 sequences with ≥90% coverage). Mutations in bold represent those observed in ≥50% of the C.1.2 sequences.

Fig. 2. Shared spike mutation results in similar immune responses.

a Visualization of C.1.2 spike mutations, highlighting those shared with VOCs and VOIs (colored by the Nextstrain clade). All C.1.2 mutations are shown, and colored according to prevalence within the C.1.2 sequences (shown in key). For VOCs and VOIs only mutations present in at least 50% of sequences are shown (as determined by frequency information at outbreak.info, accessed September 13, 2021). b Schematic showing C.1.2 mutations on the RBD-down conformation of SARS-CoV-2 spike, with domains of a single protomer shown in cartoon view and colored cyan (N-terminal domain, NTD), red (C-terminal domain/receptor binding domain, CTD/RBD), gray (subdomain 1 and 2, SD1 and SD2), and green (S2). The adjacent protomers are shown in translucent surface view and colored shades of gray. Lineage-defining mutations (found in >50% of sequences) are colored dark purple, with additional mutations (present in <50% of sequences) colored light purple. Key mutations known/predicted to influence neutralization sensitivity (C136F and P25L, Δ144Y, Δ242L/243A, and E484K), or furin cleavage (H655Y and N679K) are indicated. Image was created using the PyMOL molecular graphics program. c, d Neutralization activity of biologically independent plasma samples taken from donors previously vaccinated with either AZD1222 (N = 11), Ad26.COV.2.S (N = 10 for D614G, Beta and Delta and N = 9 for C.1.2) or BNT162b2 (N = 7 for C.1.2 and N = 6 for D614G, Beta, and Delta) (c) and patients previously infected during the first (N = 10 for D614G, N = 7 for C.1.2 and N = 5 for Beta and Delta), second (N = 10 for D614G, Beta, and Delta and N = 7 for C.1.2) or third (N = 9 for D614G, Beta, and Delta and N = 7 for C.1.2) waves in South Africa (d) against the wild-type (D614G), Beta, Delta, and C.1.2 variants. Bar graphs represent the geometric mean titer (GMT) for each group with the error bars representing the 95% confidence intervals, dots represent individual sample titers. Statistical significance based on the Wilcoxon two-tailed matched-pairs signed rank test are shown above graphs. p-values are denoted with “*” symbols: *p < 0.05, **p < 0.01, and ***p < 0.001. GMT and fold-change (FC) differences relative to D614G are given below the graph, with red representing decreased titer and green representing increased titer. No adjustments for multiple testing were made.