Evolution and neutralization escape of the SARS-CoV-2 BA.2.86 subvariant

Informed consent and ethical statement

Sex and/or gender were not considered in the study design as enrollment was of all SARS-CoV-2 infected participants. Blood samples and nasopharyngeal swabs for ancestral D614G SARS-CoV-2 isolation, as well as all blood samples used in the neutralization experiments, were obtained after written informed consent from adults with PCR-confirmed SARS-CoV-2 infection who were enrolled in a prospective cohort study at the Africa Health Research Institute approved by the Biomedical Research Ethics Committee at the University of KwaZulu–Natal (reference BREC/00001275/2020). Blood samples were collected from 21 participants with Omicron XBB-derived infection (13 female, 8 male), age range 28–83. For samples used in the serosurvey analysis, blood samples were collected from 40 participants (33 female, 7 male) age range 18–61. For pre-Omicron vaccinated participants, blood samples were collected from 19 participants (12 females, 7 males) age range of 22–75. For the BA.1 infected participants, blood samples were collected from 19 participants (14 female, 5 male) with age range 26-81. Participants received compensation for each study visit as approved by the Biomedical Research Ethics Committee at the University of KwaZulu–Natal. The Omicron/BA.1 and BA.2.86 were isolated from a residual swab sample with SARS-CoV-2 isolation from the sample approved by the University of the Witwatersrand Human Research Ethics Committee (HREC) (ref. M210752). The sample to isolate XBB.1.5 was collected after written informed consent as part of the COVID-19 transmission and natural history in KwaZulu–Natal, South Africa: Epidemiological Investigation to Guide Prevention and Clinical Care in the Center for the AIDS Program of Research in South Africa (CAPRISA) study and approved by the Biomedical Research Ethics Committee at the University of KwaZulu–Natal (reference BREC/00001195/2020, BREC/00003106/2021).

Whole-genome sequencing and genome assembly

For the BA.2.86 swab sample, RNA was extracted on an automated Chemagic 360 instrument, using the CMG-1049 kit (Perkin Elmer, Hamburg, Germany). Libraries for whole-genome sequencing were prepared using the Illumina COVIDseq Assay (Illumina Inc, San Diego, CA) and version 4 SARS-CoV-2 primer pools. Pooled PCR products were fragmented and tagged to adapter sequences. The adapter-tagged amplicons were purified and indexed using sets 1–4 of PCR indexes (Illumina). Libraries were quantified using a Qubit 4.0 fluorometer (ThermoFisher Scientific, Oregon, USA) using the Qubit dsDNA High Sensitivity assay according to the manufacturer’s instructions. Fragment sizes were analyzed using the TapeStation 4200 system (Agilent Technologies, Santa Clara, CA). Libraries were pooled and normalized to 4 nM sample library with a 2% PhiX spike-in. Libraries were loaded onto a 300-cycle NextSeq P2 Reagent Kit v2 and run on the Illumina NextSeq 1000/2000 instrument (Illumina). Sequencing data was analyzed using Exatype v4.1.5 (Hyrax Biosciences, Cape Town, South Africa) with default parameters (10% minimum prevalence to report variants, 80% minimum prevalence to include a variant in consensus sequence). Nextclade (v2.14.1) and Pangolin (v4.3, Pangolin-data v1.21) were used for clade and lineage assignments. Additionally, Nextclade was used for the visualization of the sequences and the identification of frameshifts. Unknown frameshifts were manually corrected using Aliview (v1.24). Outbreak.info was used to determine the prevalence of mutations.

For the BA.2.86 outgrowth sample, Oxford Nanopore sequencing was performed. RNA was manually extracted from either 200 µL input volume using either the MagMAX™ Viral/Pathogen II Nucleic Acid Isolation Kit (Thermo Scientific, A42352) or from 140 µL using the QIAamp Viral RNA Kit (Qiagen, 52906) as per the manufacturer’s protocols. All RNA extractions were measured using Qubit fluorimeter kits (Thermo Scientific, Q32852). The cDNA synthesis was performed using LunaScript RT mastermix (New England BioLabs) followed by whole-genome multiplex PCR using the Midnight Primer pools v3 (EXP-MRT001, Oxford Nanopore) that produce 1200-base-pair amplicons. The amplified products for each pool were combined and used for library preparation procedures using the Oxford Nanopore Rapid Barcoding kit (SQK-RBK110.96, Oxford Nanopore). The barcoded samples were pooled and cleaned up using magnetic beads and loaded on an R9.4.1 flow cell for 8-h sequencing on a MinION device. The raw data was processed using Guppy basecaller and Guppy barcoder (Oxford Nanopore) for basecalling and demultiplexing. The final consensus sequences were obtained using the Genome Detective v2.64. The lineage assignment was determined using Nextclade.

Phylogenetic analysis

We assembled a set of 280 BA.2 (Nextstrain clade 21 L) sequences collected between November 2021 and June 2022 from data deposited on GISAID⁴⁸. BA.2.86 sequences were downloaded on September 7 2023 directly from GISAID. We excluded sequences with reversion mutations relative to BA.2, sequences flagged as poor quality by Nextclade⁴⁹, or sequences with less than 90% coverage of the reference. Sequences were pairwise aligned against Wuhan-Hu-1 using Nextclade. Terminals and gaps were masked as well as all suspected artefactual reversions to reference in BA.2.86 sequences. A tree was built using IQ-tree 2⁵⁰ and post processed using a custom script to correct for incomplete merging of branches in large polytomies.

A time tree was inferred using TreeTime⁵¹ using a clock rate of 0.0005 per site and year²⁶. The rate of the long branch between BA.2 and BA.2.86 was set to be 2 times the rate of the rest of the tree in line with a previous observation that evolution is 2-fold accelerated along many long branches leading to distinct clades²⁶. This acceleration is consistent with the dramatic enrichment of amino acid substitutions in the spike protein along the long branch leading to BA.2.86.

Cells

The VeroE6 cells expressing TMPRSS2 and ACE2 (VeroE6-TMPRSS2), originally BEI Resources, NR-54970 were used for virus expansion and all live-virus assays excluding replication. The Vero-TMPRSS2 cell line was propagated in growth medium consisting of Dulbecco’s Modified Eagle Medium (DMEM, Gibco 41965-039) with 10% fetal bovine serum (Hyclone, SV30160.03) containing 10 mM of hydroxyethylpiperazine ethanesulfonic acid (HEPES, Lonza, 17-737E), 1 mM sodium pyruvate (Gibco, 11360-039), 2mM L-glutamine (Lonza BE17-605E) and 0.1 mM nonessential amino acids (Lonza 13-114E). The H1299-E3 (H1299-ACE2, clone E3) cell line used in the replication assay was derived from H1299 (CRL-5803) and propagated in a growth medium consisting of complete Roswell Park Memorial Institute (RPMI, Gibco, 21875-034) 1640 with 10% fetal bovine serum containing 10 mM of HEPES, 1 mM sodium pyruvate, 2mM l-glutamine and 0.1 mM nonessential amino acids.

Virus expansion

All work with live virus was performed in Biosafety Level 3 containment using protocols for SARS-CoV-2 approved by the Africa Health Research Institute Biosafety Committee. VeroE6-TMPRSS2 cells were seeded at 4.5 × 10⁵ cells in a 6-well plate well and incubated for 18–20 h pre-infection. After one Dulbecco’s phosphate-buffered saline (DPBS) wash, the sub-confluent cell monolayer was inoculated with 500 μL with universal transport medium which contained the swab, diluted 1:2 with growth medium filtered through a 0.45 μm and 0.22 μm filters. Cells were incubated for 2 h. Wells were then filled with 3 mL complete growth medium. After 3 days of infection (completion of passage 1 (P1)), the supernatant was collected, cells were trypsinized, centrifuged at 300 × g for 3 min, and resuspended in 3 mL growth medium. All infected cells and supernatant were added to VeroE6-TMPRSS2 cells that had been seeded at 1.5 × 10⁵ cells per mL, 20 mL total, 18–20 h earlier in a T75 flask for cell-to-cell infection. The coculture was incubated for 1 h and the flask was filled with 20 mL of complete growth medium and incubated for 3 days. The viral supernatant from this culture (passage 2 (P2) stock) was used for experiments.

Live-virus focus-forming assay and neutralization assay

For all neutralization assays, viral input was 100 focus-forming units per well of a 96-well plate. VeroE6-TMPRSS2 cells were plated in a 96-well plate (Corning) at 30,000 cells per well 1-day pre-infection. Plasma was separated from EDTA-anticoagulated blood by centrifugation at 500 × g for 10 min and stored at −80 °C. Aliquots of plasma samples were heat-inactivated at 56 °C for 30 min and clarified by centrifugation at 10,000 × g for 5 min. Virus stocks were used at approximately 50–100 focus-forming units per microwell and added to diluted plasma in neutralization assays. Antibody–virus mixtures were incubated for 1 h at 37 °C, 5% CO₂. Cells were infected with 100 μL of the virus–antibody mixtures for 1 h, then 100 μL of a 1X RPMI 1640 (Sigma-Aldrich, R6504), 1.5% carboxymethylcellulose (Sigma-Aldrich, C4888) overlay was added without removing the inoculum. Cells were fixed 20 h post-infection using 4% PFA (Sigma-Aldrich, P6148) for 20 min. Foci were stained with a rabbit anti-spike monoclonal antibody (BS-R2B12, GenScript A02058) at 0.5 μg/mL in a permeabilization buffer containing 0.1% saponin (Sigma-Aldrich, S7900), 0.1% BSA (Biowest, P6154) and 0.05% Tween-20 (Sigma-Aldrich, P9416) in PBS for 2 h at room temperature with shaking, then washed with wash buffer containing 0.05% Tween-20 in PBS. A secondary goat anti-rabbit HRP conjugated antibody (Abcam ab205718) was added at 1 μg/mL and incubated for 2 h at room temperature with shaking. TrueBlue peroxidase substrate (SeraCare 5510-0030) was then added at 50 μL per well and incubated for 20 min at room temperature. Plates were imaged in an ImmunoSpot Ultra-V S6-02-6140 Analyzer ELISPOT instrument with BioSpot Professional built-in image analysis (C.T.L) which was also used to quantify areas of individual foci.

Statistics and fitting

All statistics were performed in GraphPad Prism version 9.4.1. All fitting to determine FRNT₅₀ and linear regression was performed using custom code in MATLAB v.2019b (FRNT₅₀) or the fitlm function for linear regression, which was also used to determine goodness-of-fit (R²) as well as p-value by F-test of the linear model.

Neutralization data were fit to:

Here, Tx is the number of foci at plasma dilution D normalized to the number of foci in the absence of plasma on the same plate. ID₅₀ is the plasma dilution giving 50% neutralization. FRNT₅₀ = 1/ID₅₀. Values of FRNT₅₀ < 1 are set to 1 (undiluted), the lowest measurable value. We note that the most concentrated plasma dilution was 1:25 and therefore FRNT₅₀ < 25 was extrapolated.

Plaque assay

VeroE6-TMPRSS2 cells were plated in a 96-well plate (Corning) at 30,000 cells per well 1-day pre-infection. Virus stocks (used at the focus-forming units per microwell shown in Fig. 2) were added to cells, and incubated for 1 h at 37 °C, 5% CO2. Following incubation, 100 μL of a 1X RPMI 1640 (Sigma-Aldrich, R6504), 1.5% carboxymethylcellulose (Sigma-Aldrich, C4888) overlay was added without removing the inoculum. Cells were fixed 72 h post-infection using 4% PFA (Sigma-Aldrich, P6148) for 20 min. The fixed cells were washed with distilled water and stained with 30 μL/well of a 0.5% crystal violet solution (Sigma-Aldrich, 61135).

Replication assay

H1299-E3 cells were seeded at 1 × 10⁶ cells in a 5 mL growth medium in a Corning T25 flask 18–20 h pre-infection. Cells were infected with 5 focus-forming units of either ancestral B.1, Omicron XBB.1.5, or Omicron BA.2.86. 300 µL of supernatant was collected at the input (day 0) and on days 1–4 post-infection.

Cycle threshold values for SARS-CoV-2 RNA copies

Samples were diluted 1:3 with PBS and sent to an accredited diagnostic laboratory (Molecular Diagnostic Services, Durban, South Africa) to determine SARS-CoV-2 cycle threshold (Ct) values. At Molecular Diagnostic Services, samples were extracted using a guanidine isothiocyanate/magnetic bead-based method with the NucliSense (Biomerieux) extractor of the KingFisher Flex 96 (Thermo Fisher). Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was performed using the Seegene Allplex 2019 nCoV assay with the Bio-Rad CFX96 real-time PCR instrument as per the kit instructions. RNase P is used as the internal housekeeping gene to monitor extraction and assay efficiency. The kit targets the E, N, and R genes of SARS-CoV-2. Run calls and interpretation were performed by the Seegene Viewer software. Fold-change was calculated as FC = 2^{((mean(Ct input) – Ct sample)} in the replication experiment and FC = 2^{(Ct most dilute sample – Ct sample)}.

Cycle threshold linearity and infectivity assay

To determine whether Ct values were a good correlate for infectious viruses and to determine infectivity (number of viral genomes per focus-forming unit), 2-fold serial dilutions of a viral stock starting at approximately 100 focus-forming units were sent to determine Ct as above and in parallel plated to determine focus-forming units. For infectivity, fold-change was determined from the Ct as above, and linear regression was performed using MATLAB v.2019b against the focus-forming units obtained for the same dilution. To determine the number of viral genomes required per focus-forming unit, Ct values were converted to viral genomes using the approximation from ref. ⁵². Regression was then performed using MATLAB v.2019b against the focus-forming units obtained for the same dilution.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.