Gabriela Alvarado

Alvarado_Gabriela headshot

Molecular determinants of human antibody-mediated inhibition of human norovirus

Project Abstract

The overall goal of this project is to provide a detailed understanding of the human humoral response to Norovirus (NoV) infection and to define the molecular and structural basis for inhibition of NoV by human antibodies. NoV is the leading cause of sporadic and epidemic gastroenteritis in humans. There are currently no vaccines, therapeutics, or prophylactics available to prevent or treat NoV infection. Vaccine design also has been difficult due to the antigenic variation within and between NoV genogroups. To design a vaccine that elicits broad protective immunity, we must have a solid understanding of the NoV-mediated human antibody response to infection and antigenic sites recognized by these antibodies, so a critical area of HuNoV research has become the identification of type specific and cross-genotype and genogroup epitopes. The general hypothesis of this study is that inter- and intragenogroup I and II cross-reactive broadly blocking human antibodies exist and that the blocking function of such antibodies is principally mediated by IgA isotype molecules. The approach to this project will include high-efficiency isolation of human mAbs from patients previously infected with GI and GII strains of human NoV. I then will use this panel of human antibodies to determine if antibody isotype influences binding to NoV virus-like particles (VLPs) and blocking of binding of VLPs to host attachment factors. Finally, using site-directed mutagenesis and three-dimensional structural analysis, I will map immunoreactive epitopes on current circulating strains of NoV and determine whether cross-reactive broadly blocking epitopes exist.

Project Update

As was done below for Norwalk virus, we intend to characterize three-dimensionally where broadly binding and blocking human monoclonal antibodies bind to current circulating strains of norovirus.

Picture1

Progress Report March 2017

We hypothesize that human monoclonal antibodies (mAbs) specific to common circulating strains of HuNoV can be isolated from patients previously infected with HuNoV, and that there is enough sequence conservation between circulating genogroups and genotypes to isolate cross-reactive broadly blocking mAbs. We also predict that Ab isotype plays a role in antibody binding and neutralization or blockade functions. Part of my research is based on preliminary studies that suggest that human monoclonal antibody binding affinity and blockade response to genogroup I, genotype 1 (GI.1) NoV VLPs is influenced by antibody isotype. Specifically, even though immunoglobulin A’s (IgA) have lower binding affinities than immunoglobulin G’s (IgG) to GI.1 VLPs, that they block more potently. To determine the effects of antibody isotype, specifically the constant domain, on binding and blocking efficiency, I engineered and expressed IgG, monomeric IgA (mIgA) and dIgA isotype switch variant recombinant Igs for 10 GI.1 VLP specific hybridoma mAbs and assessed mAb binding and blocking of GI.1 VLPs from binding to host cell attachment factors. To analyze disruption of interaction between the VLPs and host cell attachment factors in vitro, I used a surrogate neutralization assay or blocking assay. With these recombinant mAbs I could test whether the differences in binding and blocking were attributable to general differences between the isotypes themselves such as avidity or molecular size, or to distinctions in antibody specificity, as a results of variances in their variable domain sequences.

Table1

Table 1. Blocking potencies of various isotypes of anti-NV mAbs

 

For our publication, Frequent Use of the IgA Isotype in Human B Cells Encoding Potent Norovirus-Specific Monoclonal Antibodies That Block HBGA Binding in PLoS Pathogens, we incorporated 5 of the 10 initial hybridomas isolated in addition to their IgG, mIgA, and dIgA recombinant constructs as seen above in Table 1. Listed are binding EC50 and blocking IC50 values corresponding to each antibody tested. From these studies we were able to conclude that in fact the IgA forms of each antibody block GI.1 VLPs from binding to host cell attachment factors in vitro more potently. In September of 2016, in collaboration with Dr. Venkataram Prasad’s laboratory we also published crystallographic studies of the GI.1 protruding domain in complex with the Fab fragment of 5I2 IgA, one of the 5 original hybridomas in our study. The three-dimensional structure revealed that the 5I2 Fab blocks the histo-blood group antigen (HBGA) binding site and predominantly using its complementarity-determining region (CDR) light chain 1.

Cross-reactive inter- and intra-genogroup murine Abs to several strains of NoV have been previously isolated by other groups, but the blocking abilities of these Abs were never tested. A murine Ab response to a human infection offers little information about the actual human response. Furthermore, antigenic blockade epitopes on GII.4 Sydney, GII.6, GII.17, GII.3, or GI.2 NoV strains have yet to be identified. My next main objective is to isolate cross-reactive human broadly blocking monoclonal antibodies to GI.1, GI.2, GII.3, GII.4, GII.6 and GII.17 and to map their corresponding epitopes. The information we gather also will be useful when reformulating multivalent VLPs for HuNoV vaccine trials, since the goal of a vaccine is to elicit a protective response against more than one circulating strain of HuNoV. To validate blockade function of GI.1, GII.3, GII.4, and GII.17 specific mAbs, I will work with our collaborator Dr. Mary Estes to test reduction of replication of live virus using human intestinal enteroids (HIE). Unfortunately, GI.2 and GII.6 cannot be tested using the HIE system because they have yet to be grown robustly in vitro, so I will test their function using a surrogate neutralization assay. Using a baculovirus-insect cell expression system, I have successfully expressed and purified all 5 VLPs. I also will be making VLPs composed of just shell (S) subdomain sequences and protruding (P) subdomain dimers to be used for epitope mapping. To date I have isolated over 40 GII.4 Sydney, GII.6, GII.17, GII.3, or GI.2 specific or cross-reactive human monoclonal antibodies. I am currently working on characterizing their specificity, function and structure.

Publications

Shanker S, Czakó R, Sapparapu G, Alvarado G, Viskovska M, Sankaran B, Atmar RL, Crowe JE Jr, Estes MK, Prasad BV. Structural basis for norovirus neutralization by an HBGA blocking human IgA antibody. Proc Natl Acad Sci U S A. 2016 Oct 4;113(40):E5830-E5837. PMID: 27647885

Sapparapu, G.*, Czakó, R.*, Alvarado, G.*, Shanker, S., Prasad, B.V.V., Atmar, R., Estes, M., Crowe, J. Jr. (2016). Frequent use of the IgA isotype in human B cells encoding potent norovirus-specific monoclonal antibodies that block HBGA binding. PLoS Pathog. 2016 Jun 29. Doing:10.1371/journal.ppat.1005719. eCollection 2016.

[* co-first authors]

Mentors

Primary: James E. Crowe, Jr.

Secondary: Ivelin Georgiev

Type of Trainee

Graduate student