Molecular Medicine Israel

A single nanobody neutralizes multiple epochally evolving human noroviruses by modulating capsid plasticity

Abstract

Acute gastroenteritis caused by human noroviruses (HuNoVs) is a significant global health and economic burden and is without licensed vaccines or antiviral drugs. The GII.4 HuNoV causes most epidemics worldwide. This virus undergoes epochal evolution with periodic emergence of variants with new antigenic profiles and altered specificity for histo-blood group antigens (HBGA), the determinants of cell attachment and susceptibility, hampering the development of immunotherapeutics. Here, we show that a llama-derived nanobody M4 neutralizes multiple GII.4 variants with high potency in human intestinal enteroids. The crystal structure of M4 complexed with the protruding domain of the GII.4 capsid protein VP1 revealed a conserved epitope, away from the HBGA binding site, fully accessible only when VP1 transitions to a “raised” conformation in the capsid. Together with dynamic light scattering and electron microscopy of the GII.4 VLPs, our studies suggest a mechanism in which M4 accesses the epitope by altering the conformational dynamics of the capsid and triggering its disassembly to neutralize GII.4 infection.

Introduction

Human noroviruses (HuNoVs), members of the genus Norovirus in the family Caliciviridae, are the leading causative agents of epidemic and sporadic acute viral gastroenteritis worldwide1,2. While most immunocompetent patients recover without treatment, norovirus infection can be life-threatening in infants, the elderly, and people with underlying diseases3. It is estimated that HuNoVs cause ~684 million illnesses and ~212,000 deaths annually4,5,6,7. The direct health system and societal costs are estimated to be over $60 billion per year8. Despite the substantial societal and economic burdens caused by HuNoVs, no antivirals or norovirus vaccines are available7.

Noroviruses (NoVs) are nonenveloped, positive-sense single-stranded RNA viruses with a genome consisting of three open reading frames (ORFs). ORF2 and ORF3 encode the major capsid protein VP1 and the minor structural protein VP2, respectively9. The amino acid sequence of VP1 is used to classify NoVs into at least ten genogroups (GI-GX), which are further subdivided into 49 genotypes10. Among these genogroups, GI, GII, GIV, GVIII, and GIX infect humans, and the viruses in the GII genogroup and genotype 4 (GII.4) are the most predominant. These viruses exhibit preferential accumulations of mutations within VP1 that have indicated the occurrence of genetic drift and selection with each variant descended from chronological predecessors11. Observed genetic findings and changes in epidemiology indicate population immunity drives the epochal evolution of GII.4 norovirus with the periodic emergence of a variant with new antigenic profiles replacing the previous variant as a means of immune evasion11, similar to H3N2 influenza A virus12.

Despite the initial obstacles to HuNoV cultivation, there has been remarkable progress in using human intestinal enteroid (HIE) systems for virus replication to study the determinants of infectivity, innate immune responses, and antibody-mediated neutralization13,14,15. However, there are still challenges in these systems to successfully propagate and obtain the virus in sufficient quantities for structural and biochemical studies, which still rely on virus-like particles (VLPs) produced by the co-expression VP1 and VP216. These VLPs are structurally and immunologically similar to authentic virions. While there are some considerable drawbacks to using VLPs, such as the lack of genomic RNA, which may play a role in differentially stabilizing the virus capsid, the use of these VLPs has been invaluable in understanding the structural, immunological, and biological aspects of many strains of HuNoVs17.

To date, the structures of several caliciviruses have been determined, including feline calicivirus18, San Miguel sea lion virus19, murine norovirus (MNV)20,21,22, and HuNoV VLPs23,24,25,26,27,28 using X-ray crystallography and high-resolution cryo-electron microscopy (cryo-EM). These structural studies have shown that the capsid of calicivirus virions consists of 90 copies of VP1 dimers assembled with a T = 3 icosahedral symmetry28,29. Each VP1 subunit consists of an internal N-terminal arm (NTA) and two distinct domains, termed shell (S−) and protruding (P-) domain, separated by a flexible hinge29 (see Supplementary Fig. 1). As first observed in MNV22,30,31, recent structural studies21,23,24,25,28 on HuNoV VLPs have shown that VP1 can exist in two distinct conformations, the “resting” conformation in which the P-domain closely interacts with the S-domain, and the “raised” conformation in which the P-domain is rotated and raised above the S-domain, which is driven either by the removal of stabilizing ions, as in the case of GII.4 VLPs28, or with increase in pH, as in case of MNV21. The P-domain is further divided into P1 and P2 subdomains, with the distal P2 subdomain involved in recognition of cell attachment factors, which is in the case of GII.4 HuNoV are the histo-blood group antigens (HBGAs) that are also the susceptibility factors32,33 (see Supplementary Fig. 1) The HuNoV VLPs have been useful in the biochemical epitope mapping and structural characterization of the human-derived neutralizing and non-neutralizing monoclonal antibodies (mAbs)14,34,35,36,37,38,39,40,41,42,43.

In addition to the traditional mAbs, llama-derived single-domain antibodies, also known as ‘nanobodies’, that recognize the HuNoV P-domain have been identified38,42,44. Nanobodies have several advantages over traditional antibodies for their development as immunotherapeutic agents. They are smaller in size (~15 kDa), exhibit higher stability over a wide range of temperatures, and are resistant to protease cleavage45,46,47. There has also been substantial work done on the development of nanobodies against several other viral agents, such as hepatitis B virus48, influenza virus49, human immunodeficiency virus50, poliovirus51, rotavirus52, and respiratory syncytial virus53. We have previously developed a panel of nanobodies against both prototype GI.1 (Norwalk-1968) and the predominant GII.4 (MD2004) VLPs44. Among these nanobodies, we chose M4 as it recognized multiple GII HuNoV strains belonging to genotypes 1, 2, 3, 4, 6, and 7 via ELISA44 and inhibited GII.4 VLP binding to HBGA and saliva, suggesting that it has a strong potential for further development as a therapeutic agent against HuNoVs44. However, whether M4 can inhibit virus replication and how it recognizes the GII HuNoV has remained unclear.

Here, using HIEs, we show that M4 inhibits replication of GII.4 HuNoVs very effectively. To understand the mechanism of the M4-mediated neutralization, we determined the crystal structure M4 in complex with GII.4 P-domain. The structure reveals a conserved epitope among GII HuNoVs, which remarkably overlaps with the epitopes of infection- and vaccine-derived human mAbs34,54. Modeling of M4 onto GII.4 capsid structure with VP1 in “resting” and “raised” conformations indicates that M4 binds to the raised VP1 conformation. Along with negative-stain EM these observations suggest that M4 uses a novel neutralization mechanism by restricting the conformational plasticity of the capsid to induce stress and mediate the disassembly of virus particles. Our study provides a molecular basis for the further development of nanobody as a therapeutic agent against HuNoVs.

Results

M4 neutralized multiple strains of GII.4 HuNoVs

To examine the neutralization potential of M4, which showed binding to multiple GII VLPs in previous studies44, we infected HIE cultures with 10% stool filtrates containing either GII.3 or different variants of GII.4 HuNoV. M4 effectively neutralized the infection of all the GII.4 HuNoV variants were used in our studies including GII.4 Sydney, GII.4 New Orleans, and GII.4 Den Haag with an IC50 of 53 ng/ml, 56 ng/ml, and 379 ng/ml, respectively (Fig. 1). Interestingly, despite binding to GII.3 VLPs44, M4 did not neutralize GII.3 HuNoV infection…

Sign up for our Newsletter