key: cord-0267585-i0vhb52u authors: Saunders, Nell; Planas, Delphine; Bolland, William; Rodriguez, Christophe; Fourati, Slim; Buchrieser, Julian; Planchais, Cyril; Prot, Matthieu; Staropoli, Isabelle; Guivel-Benhassine, Florence; Porrot, Françoise; Veyer, David; Péré, Hélène; Robillard, Nicolas; Saliba, Madelina; Baidaliuk, Artem; Seve, Aymeric; Hocqueloux, Laurent; Prazuck, Thierry; Mouquet, Hugo; Simon-Lorière, Etienne; Bruel, Timothée; Pawlotsky, Jean-Michel; Schwartz, Olivier title: Fusogenicity and neutralization sensitivity of the SARS-CoV-2 Delta sublineage AY.4.2 date: 2022-01-10 journal: bioRxiv DOI: 10.1101/2022.01.07.475248 sha: 06ba242767e13cc3b9cf61109a56031f66affb9d doc_id: 267585 cord_uid: i0vhb52u SARS-CoV-2 lineages are continuously evolving. As of December 2021, the AY.4.2 Delta sub-lineage represented 20 % of sequenced strains in UK and has been detected in dozens of countries. It has since then been supplanted by the Omicron variant. AY.4.2 displays three additional mutations (T95I, Y145H and A222V) in the N-terminal domain (NTD) of the spike when compared to the original Delta variant (B.1.617.2) and remains poorly characterized. Here, we analyzed the fusogenicity of the AY.4.2 spike and the sensitivity of an authentic AY.4.2 isolate to neutralizing antibodies. The AY.4.2 spike exhibited similar fusogenicity and binding to ACE2 than Delta. The sensitivity of infectious AY.4.2 to a panel of monoclonal neutralizing antibodies was similar to Delta, except for the anti-RBD Imdevimab, which showed incomplete neutralization. Sensitivity of AY.4.2 to sera from individuals having received two or three doses of Pfizer or two doses of AstraZeneca vaccines was reduced by 1.7 to 2.1 fold, when compared to Delta. Our results suggest that mutations in the NTD remotely impair the efficacy of anti-RBD antibodies. The temporary spread of AY.4.2 was not associated with major changes in spike function but rather to a partially reduced neutralization sensitivity. The pandemic circulation of SARS-CoV-2 is associated with emergence of variants with increased interindividual transmission or immune evasion properties. The Delta Variant of Concern (VOC), originally identified in India in 2020, has supplanted pre-existing strains worldwide in less than 6 months 1 2 . The spike protein of Delta contains 9 mutations, when compared to the B.1 ancestral strain (D614G), including five changes in the NTD (T19R, G142D, Δ156, Δ157, R158G), two in the receptor binding domain (RBD) (L452R, T478K), one mutation close to the furin cleavage site (P681R) and one in the S2 region (D950N) 3 . This set of mutations reduces sensitivity to antibody neutralization, enhances the fusogenicity of the spike and improves viral fitness 3 4 5 6 7 8 . The increased transmissibility of VOCs may also be due to mutations in other viral proteins, such as R203N in the nucleocapsid (N) 9 . The Delta lineage is heterogeneous and continues to evolve. It can be divided into sublineages or clades 10 11 12 . Different classifications exist. Next strain has classified the Delta variant into 3 main clades (21A, 21I and 21J). The Pangolin nomenclature is more resolutive and has designed almost 180 sublineages within these clades, all named AY as aliases to the B.1.617.2 lineages 13 . Mutations fixed in one sublineage (e.g. spike: T19R, G142D or D950N) are also present at low frequencies in other sublineages. This may reflect founder effects or similar selective pressures on these variants. One sublineage, termed AY.4.2 (or VUI-21OCT-01) has drawn attention due to its slow but continuous rise in UK between July and December 2021 14 15 . AY.4.2 sequences from 45 countries have been uploaded to the GISAID database. As of Dec 18, 2021 , about 62,000 genomes have been reported in the UK on GISAID, representing about 15% of reported Delta cases in this country between December 1 and 18, 2021. Its occurrence has since then strongly diminished, as the Delta lineages have been replaced by Omicron strains worldwide 16, 17 18 . The AY.4.2 sub-lineage is notably defined by the presence of Y145H and A222V mutations that lie within the NTD. Their impact on spike function is poorly characterized. Through modelling, the Y145H substitution has been predicted to decrease spike stability, but this has not been experimentally demonstrated 19 . The mutation is located in close proximity to residue 144, which is deleted in the Alpha variant. A 141-144 deletion has also been reported in several chronically SARS-CoV-2 infected immunocompromised individuals 20 . Furthermore, a 143-145 deletion is also observed in the Omicron variant 21 . Deletions of aa 144 and adjacent residues may drive antibody escape 22, 23 . The A222V mutation was noted in the B.1.177 (or 20A.EU1) lineage that emerged in Spain and spread throughout Europe in summer 2020 24 . This lineage did not have obvious transmission advantage and its spread was mostly explained by epidemiological factors such as travelling 24 . When introduced into the D614G spike, the A222V substitution slightly but not significantly impacted neutralization of pseudoviruses by human convalescent sera 25 . The effect of combined Y145H and A222V mutations on the Delta background remains unknown. Of note, most AY.4.2 sequences (93%) now include the T95I mutation in the NTD of the spike, a substitution that was rarely observed in the original Delta B1.617.2 lineage, but which gradually appeared and is now present in 40% of Delta sequences on GISAID. The T95I substitution was previously detected in the close B.1.617.1 lineage (also termed Kappa) 26 . It was also present in the B.1.526 lineage (also termed Iota) that accounted for up to 30% of sequenced cases in New York City in early 2021 27 . It is also present in the Omicron variant 21 . This substitution was found in two vaccinated individuals with breakthrough infection and selected in an immunocompromised individuals with chronic COVID-19 treated with convalescent plasma and monoclonal antibodies 28 29 . The T95 residue is located outside the NTD antigenic supersite and its contribution to immune evasion is poorly characterized 26 . Here, we studied the AY.4.2 spike by assessing its fusogenic activity, affinity to ACE2 and recognition by antibodies. We also isolated an infectious AY.4.2 strain and examined its sensitivity to a panel of monoclonal antibodies and sera from individuals having received two or three vaccine doses. No statistical methods were used to predetermine sample size. The experiments were not randomized and the investigators were not blinded to allocation during experiments and outcome assessment. Our research complies with all relevant ethical regulation. Orléans Cohort of convalescent and vaccinated individuals. Since August 27, 2020, a prospective, monocentric, longitudinal, interventional cohort clinical study enrolling 170 SARS-CoV-2-infected individuals with different disease severities, and 30 non-infected healthy controls is on-going, aiming to describe the persistence of specific and neutralizing antibodies over a 24-months period. This study was approved by the ILE DE FRANCE IV ethical committee. At enrolment, written informed consent was collected and participants completed a questionnaire which covered sociodemographic characteristics, virological findings (SARS-CoV-2 RT-PCR results, including date of testing), clinical data (date of symptom onset, type of symptoms, hospitalization), and data related to anti-SARS-CoV-2 vaccination if ever (brand product, date of first and second vaccination). Serological status of participants was assessed every 3 months. Those who underwent anti-SARS-CoV-2 vaccination had weekly blood and nasal sampling after first dose of vaccine for a 2 months period (ClinicalTrials.gov Identifier: NCT04750720). For the present study, we selected 56 convalescent and 28 vaccinated participants (16 with Pfizer and 12 with AstraZeneca). Study participants did not receive any compensation. A codon-optimized version of the reference Wuhan SARS-CoV-2 Spike (GenBank: QHD43416.1) was ordered as a synthetic gene (GeneArt, Thermo Fisher Scientific) and was cloned into a phCMV backbone (GeneBank: AJ318514), by replacing the VSV-G gene. The mutations for Alpha and Delta were added in silico to the codon-optimized Wuhan strain and ordered as synthetic genes (GeneArt, Thermo Fisher Scientific) and cloned into the same backbone. The D614G spike plasmid was generated by introducing the mutation into the Wuhan reference strain via Q5 site-directed mutagenesis (NEB). The T95I, Y145H and A222V were successively introduced into the Delta spike by the same process. Plasmids were sequenced prior to use. The primers used for sequencing and the site-directed mutagenesis are listed in the tables S3A and S3B. The GFP area and the number of nuclei were quantified 18h post-transfection using Harmony High-Content Imaging and Analysis Software, as previously described 37, 38 . For surface staining they were seeded at a confluency of 6*10 4 cells per well, and stained as described below using mAb 129. S-Fuse neutralization assay. U2OS-ACE2 GFP1-10 or GFP 11 cells, also termed S-Fuse cells, become GFP+ when they are productively infected by SARS-CoV-2 37 38 . Cells were tested negative for mycoplasma. Cells were mixed (ratio 1:1) and plated at 8x10 3 per well in a μClear 96-well plate (Greiner Bio-One). The indicated SARS-CoV-2 strains were incubated with mAb, sera or nasal swabs at the indicated concentrations or dilutions for 15 minutes at room temperature and added to S-Fuse cells. The nasal swabs and sera were heat-inactivated 30 min at 56°C before use. 18 hours later, cells were fixed with 2% PFA, washed and stained with Hoechst (dilution 1:1,000, Invitrogen). Images were acquired with an Opera Phenix high content confocal microscope (PerkinElmer). The GFP area and the number of nuclei were quantified using the Harmony software (PerkinElmer). The percentage of neutralization was calculated using the number of syncytia as value with the following formula: 100 x (1 -(value with serum -value in "non-infected")/(value in "no serum" -value in "non-infected")). Neutralizing activity of each serum was expressed as the half maximal effective dilution (ED50). ED50 values (in µg/ml for mAbs and in dilution values for sera) were calculated with a reconstructed curve using the percentage of the neutralization at the different concentrations. We previously reported a correlation between neutralization titres obtained with the S-Fuse assay and a pseudovirus neutralization assay 45 . The full-length SARS-CoV-2 genome from the patient was sequenced by means of next-generation sequencing. Briefly, viral RNA was extracted from nasopharyngeal swabs in viral transport medium We previously assessed the ability of most of these antibodies to recognize the spikes of the Alpha, Beta and Delta variants 3, 30 . To study their activity against AY.4.2, we first transfected the plasmids expressing Delta and AY.4.2 into 293T cells and analyzed antibody binding by flow cytometry (Fig. 1a) . In line with our previous results, the Delta variant was recognized by 9 of the 16 antibodies 3, 30 . AY.4.2 displayed the same binding profile as Delta (Fig. 1a) . Since the three mutations lie in the NTD, we extended our analysis to nine additional monoclonal antibodies targeting this domain. These antibodies were also cloned from SARS-CoV-2 infected individuals and bind to uncharacterized epitopes (Planchais et al, in preparation). As a control we used mAb10, a pan-coronavirus antibody that targets an unknown but conserved epitope within the S2 region 21 (Planchais, manuscript in preparation). They do not display any neutralizing activity against the ancestral Wuhan SARS-CoV-2 (not shown). Six out of the nine antibodies bound to the Delta and AY.4.2 spikes expressed at the cell surface, with various intensities (Fig. 1b) . There was no major difference in their binding to Delta and AY.4.2 spikes, except for NTD-53 which bound slightly more to AY.4.2 than to Delta and, conversely, NTD-105 which bound slightly more to Delta than to AY.4.2 (Fig. 1b) . We next examined the binding of anti-spike antibodies present in the sera of vaccines to Delta and AY.4.2. We selected individuals that received either two doses of Pfizer vaccine, sampled 7 months post second dose (n=10), or three doses, sampled at least one month after the third dose (n = 9) (Table S1A ). We also studied individuals immunized with two doses of AstraZeneca vaccine, sampled at 5 months post second dose (n=18) (Table S1B). Sera were tested at a 1:300 dilution, which allows a quantitative assessment of the antibody levels by flow cytometry 35 36 . Overall antibody levels were similar after two doses of Pfizer or AstraZeneca vaccines, and increased by 8 fold after the boost of Pfizer vaccine (Fig. 1c) . There was no major difference in the binding to the Delta and the AY.4.2 spikes (Fig. 1c) . We then performed a titration of the antibody levels in a subset of 8 sera by serial dilutions and obtained similar binding titres for the two spikes (Fig. S1a) , confirming the results obtained at the 1:300 dilution. Altogether, these results indicate that the T95I, Y145H and A222V mutations are not associated with significant changes in recognition of the spike by a panel of 24 monoclonal antibodies and by sera from vaccine recipients. We previously established a quantitative GFP-Split based cell-cell fusion assay to compare the fusogenic potential of mutant or variant spike proteins 30, 37 . In this assay, 293-T cells expressing part of the GFP protein (GFP1-10) are transfected with the spike plasmid. The transfected donor cells are then co-cultured with acceptor Vero cells expressing the other part of GFP (GFP11) 30 . Upon cell-cell fusion, the syncytia become GFP+ and the fluorescent signal is scored with an automated confocal microscope 30, 37 . Of note, 293T cells were chosen as donors because they lack ACE2 and do not fuse with each other upon spike expression. Vero cells were selected as targets because they endogenously express ACE2 and are naturally sensitive to SARS-CoV-2. We thus analyzed the fusogenic activity of the AY.4.2 spike and compared it to D614G, Alpha and Delta variants. As previously reported, the D614G and Alpha spike variants were less fusogenic than Delta (Fig. 2a) . With Delta, the area of syncytia was higher (Fig. 2a ) despite similar levels of spike expressed at the surface of 293T donor cells (Fig. 2b, Fig S2a,b) . The combination of T95I, Y145H and A222V substitutions did not modify the fusogenic activity of the Delta spike (Fig. 2a) . We next explored AY.4.2 spike binding to the ACE2 receptor. To this aim, we transiently expressed the Delta and AY.4.2 proteins in 293T cells. Cells were then stained with a serial dilution of soluble biotinylated ACE2 and revealed with fluorescent streptavidin before analysis by flow cytometry (Fig. 2c ). We previously reported using this assay that the spike protein of Alpha had the highest affinity to ACE2, followed by Delta and then by D614G, 3, 30 . Titration binding curves were generated with Delta and AY.4.2, showing no difference between the spikes' affinity for ACE2 (Fig. 2c) . Therefore, the fusogenicity and ACE2 binding of the AY.4.2 spike are similar to the ones of the parental Delta variant. We isolated the AY. We asked whether the spike present at the surface of infected cells displays the same characteristics as upon expression by transfection. We examined by flow cytometry the binding of neutralizing and non-neutralizing monoclonal antibodies to Vero cells infected with the Delta and AY.4.2 isolates. We observed the same profile of binding (Fig. 3a,b) for the two strains, and no noticeable difference with transfected 293T cells. to the panel of monoclonal antibodies we tested. We next compared the sensitivity of Delta and AY.4.2 strains to the previously described panel of neutralizing mAbs using the S-Fuse assay (Fig. 4a) . 8 out of 14 antibodies neutralized both strains. With most of the neutralizing antibodies, we observed a slightly increased IC50s against AY.4.2 (average 2.2 fold increase when compared to Delta, Fig. 4a and Table S2 ). Bamlanivimab was inactive against AY.4.2, in agreement with previous results with Delta 3 5 39 . Imdevimab displayed an incomplete neutralization. The maximum neutralization plateaued at 60% against AY.4.2, even at high antibody concentrations (1 µg/mL), whereas it reached almost 100 % against Delta (Fig. 4a) . We obtained similar results with two different batches of Imdevimab (not shown). Therefore, AY.4.2 displays a slightly more elevated resistance to neutralization by the monoclonal antibodies tested than Delta. This resistance is more marked for Imdevimab. We next asked whether vaccine-elicited antibodies neutralized AY.4.2. We used the same set of sera that were characterized by flow cytometry in Fig. 1 and compared their neutralizing activity against Delta and AY.4.2. With the Pfizer vaccine, seven months after the second dose, the levels of neutralizing antibodies were relatively low against Delta (median ED50 of neutralization of 47), reflecting the waning of the humoral response at this time point 3 (Fig. 5a) . These titres were lower against AY.4.2 (ED50 of 28, corresponding to a 1.7 fold decrease compared to Delta). One month after the booster dose (administrated at M7 post vaccination), titres strongly increased (25-50 fold), reaching 2716 and 1260 for Delta and AY.4.2 strains, respectively (Fig. 5a) . We observed a 2.1 fold reduction in the neutralization titres against AY.4.2 when compared to Delta (Fig. 2b) . Therefore, by using a set of sera with either low or high neutralizing antibody titres, we consistently observed a slight but significant decrease (1.7 to 2.1 fold reduction) of their activity against AY.4.2, when compared to the parental Delta variant. Diversification of the Delta variant is regularly reported. The AY.4.2 sublineage was first identified in July 2021 and accounted for 15% and 20% sequenced Delta cases in UK, during the first and third weeks of November, respectively 14 . This corresponds to an AY.4.2 logistic growth rate of 15% per week in this country 14 . AY.4.2 has also been detected in dozens of countries. AY.4.2 was slowly but continuously rising and may thus display a slight selective advantage compared to the parental Delta strain. An increase of the growth rate may depend on the context and should not be necessary interpreted as a change in biological transmissibility 14 to that of Delta. By using a panel of 24 monoclonal antibodies targeting either the RBD or the NTD, we did not detect major differences in antibody recognition, when the spikes are expressed by transient transfection in 293-T cells. Polyclonal sera from individuals having received either Pfizer or AstraZeneca vaccines similarly recognized the two spikes. Their fusogenic activity, when measured in a syncytia formation assay 30 and the binding affinity to ACE2 were also similar for AY.4.2 and Delta. We isolated an authentic AY.4.2 strain from an infected patient and examined its sensitivity to antibody neutralization. Future work in more relevant models, such as primary human bronchial epithelium 40 or viral competition experiments will help determining whether AY.4.2 is more fit than the parental lineage in culture systems. We analyzed the profile of binding of a panel of monoclonal antibodies to infected cells. We did not observe major differences between the two strains. We then studied the neutralization of the two viral isolates by a panel of monoclonal antibodies. Imdevimab, a therapeutic antibody used in combination with Casirivimab in the commercially approved REGN-COV2 cocktail from Regeneron and Roche (Ronapreve TM ), incompletely neutralized AY.4.2. Even at high concentration, the neutralizing activity plateaued at 60%. Incomplete neutralization and deviation from sigmoidal neutralization curves have been for instance previously observed with some HIV broadly neutralizing antibodies (bNAbs) 41 . This process has been attributed to heterogeneity in glycosylation of the HIV gp120/gp41 Env complex 41 . Our results strongly suggest that the various conformations or glycosylation of the AY.4.2 spike at the virion surface may display different sensitivities to antibody neutralization. As AY4.2 does not harbour mutations within the epitope of Imdevimab, our results also indicate that mutations in the NTD of the spike may remotely impact the accessibility of anti-RBD antibodies. The 3D structure of the spike shows that some regions of the NTD are in close proximity to the RBD 34 42 . Imdevimab binds to a lateral region of the RBD and (Class 3 antibody) and may thus be more affected by changes in the NTD than other anti-RBD antibodies binding to the apex of the spike. Furthermore, the other neutralizing anti-RBD antibodies that we tested displayed a slight decrease in sensitivity to AY4.2, when compared to Delta (1. Infectious Diseases" (grant n°ANR-10-LABX-62-IBEID). The funders of this study had no role in study design, data collection, analysis and interpretation, or writing of the article. All data supporting the findings of this study are available within the paper and are available from the corresponding author upon request. Statistical analysis: One-way ANOVA, each strain is compared to D614G or delta. ns: non-significant The biological and clinical significance of emerging SARS-CoV-2 variants Estimated transmissibility and impact of SARS-CoV-2 lineage B.1.1.7 in England Reduced sensitivity of SARS-CoV-2 variant Delta to antibody neutralization SARS-CoV-2 B.1.617.2 Delta variant replication and immune evasion BNT162b2-elicited neutralization of B.1.617 and other SARS-CoV-2 variants Reduced neutralization of SARS-CoV-2 B.1.617 by vaccine and convalescent serum Neutralising antibody activity against SARS-CoV-2 VOCs B.1.617.2 and B.1.351 by BNT162b2 vaccination Membrane fusion and immune evasion by the spike protein of SARS-CoV-2 Delta variant Rapid assessment of SARS-CoV-2 evolved variants using virus-like particles Covid-19 genomic analysis reveals clusters of emerging sublineages within the delta variant SARS-CoV-2 variants of concern and variants under investigation in England The unique evolutionary dynamics of the SARS-CoV-2 Delta variant. medRxiv SARS-CoV-2 variants of concern and variants under investigation in England Neutralisation of SARS-CoV-2 Delta sub-lineage AY.4.2 and B.1.617.2+E484K by BNT162b2 mRNA vaccine-elicited sera. medRxiv Omicron SARS-CoV-2 variant: a new chapter in the COVID-19 pandemic Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in southern Africa. medRxiv Investigation of nonsynonymous mutations in the spike protein of SARS-CoV-2 and its interaction with the ACE2 receptor by molecular docking and MM/GBSA approach Long-Term Evolution of SARS-CoV-2 in an Immunocompromised Patient with Non-Hodgkin Lymphoma Considerable escape of SARS-CoV-2 Omicron to antibody neutralization Recurrent deletions in the SARS-CoV-2 spike glycoprotein drive antibody escape N-terminal domain antigenic mapping reveals a site of vulnerability for SARS-CoV-2 Spread of a SARS-CoV-2 variant through Europe in the summer of 2020 The Antigenicity of Epidemic SARS-CoV-2 Variants in the United Kingdom Molecular basis of immune evasion by the Delta and Kappa SARS-CoV-2 variants Detection and characterization of the SARS-CoV-2 lineage B.1.526 in New York Persistent Coronavirus Disease 2019 (COVID-19) in an Immunocompromised Host Treated by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)-Specific Monoclonal Antibodies Vaccine Breakthrough Infections with SARS-CoV-2 Variants SARS-CoV-2 Alpha, Beta, and Delta variants display enhanced Spike-mediated syncytia formation. The EMBO Journal n/a, e108944 Neutralizing monoclonal antibodies for treatment of COVID-19 Prospective mapping of viral mutations that escape antibodies used to treat COVID-19 Potent neutralizing antibodies against multiple epitopes on SARS-CoV-2 spike SARS-CoV-2 neutralizing antibody structures inform therapeutic strategies A comparison of four serological assays for detecting anti-SARS-CoV-2 antibodies in human serum samples from different populations Asymptomatic and symptomatic SARS-CoV-2 infections elicit polyfunctional antibodies Syncytia formation by SARS-CoV-2 infected cells Sensitivity of infectious SARS-CoV-2 B.1.1.7 and B.1.351 variants to neutralizing antibodies Impact of circulating SARS-CoV-2 variants on mRNA vaccine-induced immunity SARS-CoV-2 infection induces the dedifferentiation of multiciliated cells and impairs mucociliary clearance Incomplete Neutralization and Deviation from Sigmoidal Neutralization Curves for HIV Broadly Neutralizing Monoclonal Antibodies Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation Complete map of SARS-CoV-2 RBD mutations that escape the monoclonal antibody LY-CoV555 and its cocktail with LY-CoV016 Public Health England SARS-CoV-2 variants of concern and variants under investigation in England IgA dominates the early neutralizing antibody response to SARS-CoV-2 Global initiative on sharing all influenza data -from vision to reality Efficient generation of human IgA monoclonal antibodies We thank patients who participated to this study, members of the Virus and Immunity Unit for