key: cord-0759219-6lbh8pg8
authors: Iketani, Sho; Liu, Lihong; Guo, Yicheng; Liu, Liyuan; Huang, Yiming; Wang, Maple; Luo, Yang; Yu, Jian; Yin, Michael T.; Sobieszczyk, Magdalena E.; Huang, Yaoxing; Wang, Harris H.; Sheng, Zizhang; Ho, David D.
title: Antibody Evasion Properties of SARS-CoV-2 Omicron Sublineages
date: 2022-02-09
journal: bioRxiv
DOI: 10.1101/2022.02.07.479306
sha: 9bed65cada08ea4c7935152e1080eea47174c0de
doc_id: 759219
cord_uid: 6lbh8pg8

The identification of the Omicron variant (B.1.1.529.1 or BA.1) of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) in Botswana in November 20211 immediately raised alarms due to the sheer number of mutations in the spike glycoprotein that could lead to striking antibody evasion. We2 and others3–6 recently reported results in this Journal confirming such a concern. Continuing surveillance of Omicron evolution has since revealed the rise in prevalence of two sublineages, BA.1 with an R346K mutation (BA.1+R346K) and B.1.1.529.2 (BA.2), with the latter containing 8 unique spike mutations while lacking 13 spike mutations found in BA.1. We therefore extended our studies to include antigenic characterization of these new sublineages. Polyclonal sera from patients infected by wild-type SARS-CoV-2 or recipients of current mRNA vaccines showed a substantial loss in neutralizing activity against both BA.1+R346K and BA.2, with drops comparable to that already reported for BA.12,3,5,6. These findings indicate that these three sublineages of Omicron are antigenically equidistant from the wild-type SARS-CoV-2 and thus similarly threaten the efficacies of current vaccines. BA.2 also exhibited marked resistance to 17 of 19 neutralizing monoclonal antibodies tested, including S309 (sotrovimab)7, which had retained appreciable activity against BA.1 and BA.1+R346K2–4,6. This new finding shows that no presently approved or authorized monoclonal antibody therapy could adequately cover all sublineages of the Omicron variant.

The identification of the Omicron variant (B.1.1.529.1 or BA.1) of SARS-CoV-2 (severe acute 24 respiratory syndrome coronavirus 2) in Botswana in November 2021 1 immediately raised alarms 25 due to the sheer number of mutations in the spike glycoprotein that could lead to striking 26 antibody evasion. We 2 and others 3-6 recently reported results in this Journal confirming such a 27 concern. Continuing surveillance of Omicron evolution has since revealed the rise in prevalence 28 of two sublineages, BA.1 with an R346K mutation (BA.1+R346K) and B.1.1.529.2 (BA.2), with 29 the latter containing 8 unique spike mutations while lacking 13 spike mutations found in BA.1. 30

We therefore extended our studies to include antigenic characterization of these new sublineages. 31

Polyclonal sera from patients infected by wild-type SARS-CoV-2 or recipients of current mRNA 32 vaccines showed a substantial loss in neutralizing activity against both BA.1+R346K and BA.2, 33 with drops comparable to that already reported for BA.1 2,3,5,6 . These findings indicate that these 34 three sublineages of Omicron are antigenically equidistant from the wild-type SARS-CoV-2 and 35 thus similarly threaten the efficacies of current vaccines. BA.2 also exhibited marked resistance 36 to 17 of 19 neutralizing monoclonal antibodies tested, including S309 (sotrovimab) 7 , which had 37 retained appreciable activity against BA.1 and BA.1+R346K 2-4,6 . This new finding shows that 38 no presently approved or authorized monoclonal antibody therapy could adequately cover all 39 sublineages of the Omicron variant. globally has been truly remarkable 8 . Continuing surveillance of its evolution in the population 45 over the past six weeks has revealed that the proportion of the original form, BA.1, has been 46 decreasing steadily while the proportions of two other sublineages have increased noticeably 47 (Fig. 1a) . In fact, the BA.1+R346K sublineage now accounts for ~30% of Omicron sequences 48 globally, and ~30-45% in South Africa, United Kingdom, and United States. On the other hand, 49 the BA.2 sublineage accounts for only ~13% of Omicron sequences globally, but it is not only on 50 the rise but also the dominant form in countries such as Denmark and India. These three 51 sublineages of Omicron share 21 mutations in the spike protein, wherein BA.2 contains 8 unique 52 mutations and BA.1 contains 13 unique mutations (Fig. 1b) . Of course, BA.1+R346K has one 53 mutation more than BA.1. Given these differences, their antigenic properties cannot be assumed 54 to be the same or similar. 55 56 Therefore, we first investigated the neutralization sensitivity of the Omicron sublineages by 57 polyclonal sera from convalescent patients or individuals given mRNA vaccines, with or without 58 a booster shot. These serum samples, as well as the pseudovirus neutralization assay used, were 59 identical to ones previously reported 2 . The wild-type D614G pseudovirus was included as a 60 comparator. As was observed and reported for BA.1 2,3,5,6 , a marked and significant loss of serum 61 neutralizing activity against BA.1+R346K and BA.2 relative to D614G was noted, with 62 neutralizing titers for numerous samples dropping below the limit of detection (Fig. 1c) . The loss 63 of neutralizing activity against BA.1+R346K or BA.2 sublineages was less prominent for sera 64 obtained from individuals who received a booster vaccination (Fig. 1c , right panel), consistent 65 with reported findings for BA.1 2,3,6 . Among these samples, the mean serum neutralizing titers 66 against Omicron sublineages were significantly lower than the mean titer for D614G; although 67 the mean titer was slightly lower for BA.2, the difference from BA.1 sublineages did not reach 68 statistical significance (P = 0.242). Overall, 17 of 19 monoclonal antibodies were either totally inactive or severely impaired in 79 neutralizing BA.2 (Fig. 2a) , somewhat like previous findings for BA.1 and BA.1+R346K 2 but 80 with important differences (Fig. 2b) . All class 4 antibodies tested lost greater neutralizing 81 potency against BA.2 versus BA.1 sublineages. Two class 3 antibodies, COV2-2130 and 2-7, 82 retained decent activity against BA.2 while having no activity against BA.1 viruses. S309 or sotrovimab lost 27-fold neutralizing activity against BA.2; this is particularly important because 84 it was found to be the only clinically approved or authorized monoclonal antibody to retain 85 activity against the original form of Omicron 2-4 . LY-CoV1404, another class 3 antibody in 86 development, remained potent in neutralizing all Omicron sublineages, suggesting that there is 87 still a patch within this antibody-binding region that is unaffected by all spike mutations found in 88 SARS-CoV-2 variants to date. Although there was a lack of an observable difference among the 89 Omicron sublineages in neutralization by polyclonal sera (Fig. 1c) and tested them using the same panel of 19 monoclonal antibodies (Fig. 2b) . S371F broadly 100 affected most of the RBD-directed antibodies, similar to what was observed for S371L in BA.1 2 101 but with a greater negative impact, perhaps due to the bulkier side chain of phenylalanine. 102

Intriguingly but importantly, S371F appears to be majorly responsible for the loss in potency of 103 S309, although this mutation was not observed previously as a marker for clinical resistance to 104 sotrovimab 20 . CB6 was adversely affected by the D405N mutation, likely due to its position 105 within the epitope of this antibody 12 . It is not clear how T19I and L24S mutations in the NTD 106 subtly impaired the neutralizing activity of class 1 antibodies to RBD. implications. First, polyclonal sera showed a substantial loss in neutralizing activity against both 111 sublineages, with drops comparable to that of BA.1 (Fig. 1c) . These three sublineages of 112

Omicron, therefore, seem to be antigenically equidistant from the wild-type SARS-CoV-2, likely 113 threatening the efficacies of current COVID-19 vaccines to a similar extent. The present study, however, does not address the antigenic distance between BA.1 and BA.2, which will require 115 cross-neutralization experiments using sublineage-specific sera to determine. Second, 116 monoclonal antibodies were affected in a disparate manner for the different Omicron sublineages. 117

For clinically approved or authorized antibodies, only S309 (sotrovimab) retained activity 118 against both BA.1 and BA.1+R346K, but its activity against BA.2 has dropped 27-fold (Fig. 2b)  119 to a 50% inhibitory concentration (IC 50 ) of ~1 μg/mL (Fig. 2a) . Only COV2-2130 (cilgavimab) 120 and its combination with COV2-2196 (tixagevimab) retained activity against BA.2, but this 121 antibody combination is only authorized for preventive use. Presently, no authorized therapeutic 122 monoclonal antibody could adequately treat all sublineages of the Omicron variant. This finding 123 poses a therapeutic dilemma in geographic regions where all three sublineages are present in 124 sufficient numbers. As COVID-19 treatment options are narrowed by the emergence of more and 125 more variants, it is imperative that we continue to devise novel strategies to contain this ever-126 evolving pathogen. 

Rapid epidemic expansion of the SARS-CoV-2 Omicron variant in 153 southern Africa

Striking Antibody Evasion Manifested by the Omicron Variant of SARS-155

Broadly neutralizing antibodies overcome SARS-CoV-2 Omicron 157 antigenic shift

Omicron escapes the majority of existing SARS-CoV-2 neutralizing 159 antibodies

Omicron extensively but incompletely escapes Pfizer BNT162b2 161 neutralization

Considerable escape of SARS-CoV-2 Omicron to antibody neutralization

Cross-neutralization of SARS-CoV-2 by a human monoclonal SARS-CoV 165 antibody

Global initiative on sharing all influenza data -from 167 vision to reality

Studies in humanized mice and convalescent humans yield a SARS-169

CoV-2 antibody cocktail

Potently neutralizing and protective human antibodies against SARS-171

The neutralizing antibody, LY-CoV555, protects against SARS-CoV-2 173 infection in nonhuman primates

A human neutralizing antibody targets the receptor-binding site of SARS-176

Human neutralizing antibodies elicited by SARS-CoV-2 infection

LY-CoV1404 (bebtelovimab) potently neutralizes SARS-CoV-2 180 variants. bioRxiv

Broad and potent activity against SARS-like viruses by an 182 engineered human monoclonal antibody

In vitro and in vivo functions of SARS-CoV-2 infection-enhancing and 185 neutralizing antibodies

Broad sarbecovirus neutralization by a human monoclonal 188 antibody

Potent neutralizing antibodies against multiple epitopes on SARS

Isolation and comparative analysis of antibodies that broadly neutralize 192 sarbecoviruses. bioRxiv

No statistical methods were used to predetermine sample size. The experiments were not 201 randomized and the investigators were not blinded to allocation during experiments and outcome

Identical samples from a previous study were utilized 2 . All collections were conducted under 206 protocols reviewed and approved by the Institutional Review Board of Columbia University

Antibodies and pseudovirus neutralization

The expression of antibodies, construction of variant SARS-CoV-2 spike plasmids, production 210 and neutralization of pseudoviruses

This study was supported by funding from the Gates Foundation, JPB Foundation

the National Science Foundation (MCB-2032259), and the NIH SARS-CoV-2 Assessment of

Viral Evolution (SAVE) Program

conceived this project. S.I. and Lihong Liu conducted pseudovirus neutralization 220 experiments. Y.G. and Z.Z. conducted bioinformatic analyses. Liyuan Liu and Yiming Huang 221 constructed the spike expression plasmids. M.W. aided sample collections. Y.L. managed the 222 project

Yaoxing Huang contributed to discussions