Converting non-neutralizing SARS-CoV-2 antibodies targeting conserved epitopes into broad-spectrum inhibitors through receptor blockade

All but one of the authorized monoclonal antibody-based treatments for SARS-CoV-2 are ineffective against Omicron, highlighting the critical need for biologics capable of overcoming SARS-CoV-2 evolution. These ineffective therapeutic antibodies target epitopes that are not highly conserved. Here we describe broad-spectrum SARS-CoV-2 inhibitors developed by tethering the receptor-binding, angiotensin-converting enzyme 2 (ACE2) domain to antibodies that were originally defined as non-neutralizing, but that also target highly conserved epitopes in the viral spike protein. These inhibitors, called Receptor engaged conserved non-neutralizing Antibodies (ReconnAbs), potently neutralize all SARS-CoV-2 variants of concern (VOC), including Omicron. Neutralization potency is lost when the linker joining the two ReconnAb components is severed. A bispecific ReconnAb, made by linking ACE2 to two non-neutralizing antibodies with non-overlapping conserved epitopes, defined here, shows sub-nanomolar neutralizing activity against all VOCs, including Omicron. Given their conserved targets and modular nature, ReconnAbs have potential to act as broad-spectrum therapeutics against SARS-CoV-2 and other emerging pandemic diseases.

Methods
A library of antibodies directed against SARS-CoV-2 Spike (S) protein was developed using paired antibody sequences, meaning antibody sequences for which the heavy and light chain are both known, from the Coronavirus Antibody Database, CoV-AbDab. All antibody sequences from convalescent COVID-19 donors which had been deposited before July 9th, 2021 (http://opig.stats.ox.ac.uk/webapps/covabdab/static/downloads/CoV-AbDab_090721.csv), were inserted into a table, categorized by their binding to the SARS-CoV-2 RBD portion of the spike protein or to a non-RBD portion of SARS-CoV-2 spike. Following this analysis, antibodies that were cataloged for non-RBD binding were preferentially identified, resulting in a total of 385 paired antibody sequences. For these non-RBD binding antibodies, the amino acid sequences of the corresponding heavy chain and light chain V-genes and CDR3 regions, already compiled from the Coronavirus Antibody Database, were imported into Geneious Prime v2021.1.1 (a bioinformatics software; geneious.com). Using Geneious Prime, the heavy chain sequences and light chain sequences were separately analyzed to produce phylogenetic trees. For these phylogenetic trees, RBD binding antibodies were also included to ensure selection of antibody sequences that were both non-RBD binding and clearly distinct from RBD-binding sequences. 371 RBD binding antibody, 59 germline antibody, and 325 non-RBD binding antibody nucleic acid sequences of the corresponding heavy chain and light chain genes were imported, for a total of 755 heavy chain and light chain sequences (696 excluding germline antibodies). The sequences were first aligned using the MUSCLE algorithm, and then two phylogenetic trees were made, both using PhyML 3.3.20180621.

目前获批的抗新型冠状病毒（SARS-CoV-2）单克隆抗体治疗方案中，仅有一种对奥密克戎毒株有效，这凸显了研发可应对新冠病毒进化的生物制剂的迫切需求。此类无效治疗性抗体所靶向的表位并非高度保守。

本研究开发出广谱新冠病毒抑制剂，其构建策略为将血管紧张素转换酶2（ACE2）受体结合域与最初被定义为非中和性、但同时靶向新冠病毒刺突（S）蛋白高度保守表位的抗体进行拴系连接。此类抑制剂被称为受体结合型保守非中和抗体（ReconnAbs），可强效中和包括奥密克戎在内的所有新冠病毒关切变异株（Variants of Concern, VOC）。若切断连接ReconnAb两个组分的接头序列，则会丧失其中和活性。

本研究构建的双特异性ReconnAb，通过将ACE2与两个靶向非重叠保守表位的非中和性抗体相连而成，对包括奥密克戎在内的所有VOC均展现出亚纳摩尔级的中和活性。鉴于其靶向保守表位且具备模块化特性，ReconnAbs有望成为对抗新冠病毒及其他新发大流行疾病的广谱治疗药物。

## 方法

本研究从冠状病毒抗体数据库（Coronavirus Antibody Database, CoV-AbDab）中获取成对抗体序列（即重链与轻链序列均已知的抗体序列），构建了针对新冠病毒刺突（S）蛋白的抗体文库。将2021年7月9日前提交的所有康复期新冠感染者抗体序列（来源：http://opig.stats.ox.ac.uk/webapps/covabdab/static/downloads/CoV-AbDab_090721.csv）导入表格，并根据其结合新冠病毒刺突蛋白的受体结合域（Receptor Binding Domain, RBD）还是非RBD区域进行分类。经此分析后，优先筛选出被归类为结合非RBD区域的抗体，最终获得385条成对抗体序列。

针对这些结合非RBD区域的抗体，将已从冠状病毒抗体数据库中整理得到的对应重链、轻链可变区基因（V-genes）及互补决定区3（CDR3）区域的氨基酸序列导入Geneious Prime v2021.1.1（一款生物信息学软件；geneious.com）。借助Geneious Prime，分别对重链序列与轻链序列进行分析以构建系统发育树。构建此类系统发育树时，同时纳入结合RBD区域的抗体，以确保筛选出的抗体既结合非RBD区域，又与结合RBD区域的抗体序列存在显著差异。

本研究共导入371条结合RBD区域的抗体、59条种系抗体以及325条结合非RBD区域的抗体的对应重链与轻链基因核酸序列，总计755条重链和轻链序列（不含种系抗体的话为696条）。首先使用MUSCLE算法对序列进行比对，随后分别使用PhyML 3.3.20180621构建两棵系统发育树。