Abstract
Broadly neutralizing antibodies (bnAbs) to coronaviruses (CoVs) are valuable in their own right as prophylactic and therapeutic reagents to treat diverse CoVs and, importantly, as templates for rational pan-CoV vaccine design. We recently described a bnAb, CC40.8, from a coronavirus disease 2019 (COVID-19)-convalescent donor that exhibits broad reactivity with human beta-coronaviruses (β-CoVs). Here, we showed that CC40.8 targets the conserved S2 stem-helix region of the coronavirus spike fusion machinery. We determined a crystal structure of CC40.8 Fab with a SARS-CoV-2 S2 stem-peptide at 1.6 Å resolution and found that the peptide adopted a mainly helical structure. Conserved residues in β-CoVs interacted with CC40.8 antibody, thereby providing a molecular basis for its broad reactivity. CC40.8 exhibited in vivo protective efficacy against SARS-CoV-2 challenge in two animal models. In both models, CC40.8-treated animals exhibited less weight loss and reduced lung viral titers compared to controls. Furthermore, we noted CC40.8-like bnAbs are relatively rare in human COVID-19 infection and therefore their elicitation may require rational structure-based vaccine design strategies. Overall, our study describes a target on β-CoV spike proteins for protective antibodies that may facilitate the development of pan-β-CoV vaccines.
INTRODUCTION
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to the current global pandemic (1–3). SARS-CoV-2 is a virus that belongs to the coronaviridae family of which six members have previously crossed into humans from animal reservoirs and established widespread infections (4, 5). These include four endemic human coronaviruses (HCoVs) (HCoV-229E, HCoV-HKU1, HCoV-OC43, HCoV-NL63) responsible for non-severe, seasonal infections (4) as well as SARS-CoV-1 and MERS-CoV (Middle East Respiratory Syndrome CoV) that are associated with high morbidity and mortality in humans (6, 7). Among the seven HCoVs, SARS-CoV-2 closely resembles SARS-CoV-1 and, to lesser degree, MERS-CoV. Together with HCoV-HKU1 and HCoV-OC43, these viruses belong to the β-coronavirus genus (4, 5). SARS-CoV-2 is highly transmissible in humans and causes coronavirus disease-2019 (COVID-19), associated with severe respiratory failure leading to high morbidity and a reported mortality of about 0.7 to 2% of infected individuals worldwide (2, 8, 9). There are considerable concerns that future coronavirus spillovers will trigger new pandemics (10–15).Coronavirus pandemic preparedness may consider responses through establishment of techniques for rapid generation of specific reagents to counter the emerging coronavirus and control spread. An alternative is to seek to identify broadly neutralizing antibodies (bnAbs) to coronaviruses and use molecular information gleaned on their epitopes to rationally design pan-coronavirus vaccines (16–18). Pan-coronavirus vaccines and antibodies could be stockpiled ahead of the emergence of a new coronavirus and used to rapidly contain the virus. BnAbs and pan-coronavirus vaccines that target more conserved regions of the virus may also be more effective against antigenically variant viruses, such as have been described for the variants of concern in the COVID-19 pandemic (19–22).All HCoVs possess a surface envelope spike glycoprotein that mediates interaction with host cell receptors and enables virus fusion (4, 23). SARS-CoV-2 (similar to SARS-CoV-1) utilizes the receptor binding domain (RBD) in the S1 subunit of the spike protein to engage human angiotensin converting enzyme 2 (hACE2) on host cells for cell entry and infection (23–27). The SARS-CoV-2 spike glycoprotein is the primary target of neutralizing antibodies (nAbs) (28–31). On the spike protein, the RBD is highly immunogenic and is recognized by the majority of nAbs (28, 32–43), and thus is a major focus of current nAb-based vaccine design efforts (28, 44, 45). However, due to sequence diversity, cross-reactivity to the RBD region is limited, especially among emerging coronaviruses with pandemic potential (10–13). The most potent nAbs in humans during natural infection are typically raised to epitopes overlapping the ACE2 binding site (32, 33, 42, 45, 46). As the rapid spread of the SARS-CoV-2 virus continues, these epitopes are coming under strong immune selection pressure at the population level, leading to the selection of SARS-CoV-2 neutralization escape variants (19–22, 47–49). The relevant mutations may result in reduced effectiveness of vaccine-induced antibody responses in humans since such responses also tend to target RBD epitopes overlapping the ACE2 binding site, and because all currently approved vaccines are based on the wild-type virus. The most striking example that illustrates the capability of the RBD to mutate without majorly affecting the ability of the virus to engage host receptor is the variability of the RBD across the two families of HCoVs: SARS-CoV-2/1 (β-HCoVs) and HCoV-NL63 (α-HCoV) (23–27, 50). These HCoVs possess divergent RBDs, but all use the ACE2 receptor for viral entry suggesting that SARS-CoV-2, and potentially other emerging sarbecoviruses with human pandemic potential, can tolerate changes in this domain with limited fitness cost. Therefore, we believe that other sites on the spike protein should be explored as targets of bnAbs.We recently isolated a SARS-CoV-1/2 cross-neutralizing antibody from a COVID-19 donor, CC40.8, that exhibits broad cross-reactivity with human β-CoVs (51). Here, we show that the CC40.8 bnAb targets an S2 stem-helix epitope, which is part of the coronavirus fusion machinery. We first identified a long 25-mer S2 peptide from HCoV-HKU1 that bound CC40.8 with high affinity and then determined the crystal structure of CC40.8 with the SARS-CoV-2 S2 peptide. The S2 stem peptide adopts a largely helical structure that is embedded in a groove between the heavy and light chain complementarity determining regions (CDRs) of the antibody. Key epitope contact residues were further validated, by alanine scanning, to be important for peptide binding and for virus neutralization. These contact residues are largely conserved between β-CoVs, consistent with the cross reactivity of CC40.8. In SARS-CoV-2 challenge models, CC40.8 showed in vivo protective efficacy by reducing weight loss and lung tissue viral titers. Although two recent studies have described S2-stem nAbs isolated from mice and mice transgenic for human Ig (52, 53), CC40.8 represents a human HCoV S2-stem directed bnAb isolated from natural infection (51) and may facilitate development of antibody-based interventions and prophylactic pan-sarbecovirus and pan-β-coronavirus vaccine strategies.
RESULTS
CC40.8 binds a conserved peptide from the S2 region of β-coronaviruses.
We recently isolated a bnAb, CC40.8, from a 62-year-old SARS-CoV-2 convalescent donor from peripheral blood mononuclear cell (PBMC) samples collected 32 days post-infection (51). CC40.8 bnAb neutralizes SARS-CoV-1 and SARS-CoV-2 and exhibits broad reactivity against β-coronaviruses, notably the endemic coronavirus HCoV-HKU1 (Fig. 1A and B) (51). Here, we observed that CC40.8 bnAb can effectively neutralize clade 1b and clade 1a ACE2 receptor-utilizing sarbecoviruses (Fig. 1A, fig. S1). In addition, the CC40.8 bnAb was consistently effective against the current SARS-CoV-2 variants of concern (VOCs) (Fig. 1A, fig. S1). The effectiveness of CC40.8 bnAb with SARS-CoV-2 VOCs is consistent with a lack of mutations in the S2 stem helix region in the current VOCs (21). To assess the cell-cell inhibition ability of CC40.8 bnAb, we conducted experiments in HeLa cells expressing SARS-CoV-2 spike protein or hACE2 receptor. We observed that CC40.8 bnAb can prevent cell-cell fusion of HeLa cells expressing SARS-CoV-2 spike protein with HeLa cells expressing the hACE2 receptor (fig. S2)….
