Novel biosensing concepts for advancing antigen-specific antibody detection

Antibodies are of great interest in the biotech and medical field due to their high specificity and sensitivity obtained after an immune response. They have been extensively employed as tools ranging from affinity ligands in diagnostic devices to therapeutic agents for treatment of diseases. For this reason, the characterization of antigen-specific antibodies has become paramount. To explore the antibody repertoire under pathologic conditions, analytical tools have been designed over the past 40 years for the detection and isolation of antigen-specific antibodies. Current methods are focused on designing immortalized B cells (such as hybridoma cells) or displaying antibodies on the surface of filamentous bacteriophages (phage display). However, these methods do not resemble the natural human antibody response or lack of natural pairing of the different variable regions of the antibody (VH-VL). Therefore, study of the human immunoglobulin repertoire has been described by detecting and sorting live B cells from human samples since those are the only lymphocytes able to produce and secrete antibodies in our body. To access the antibody amino acid sequence for further cloning and expression, DNA sequencing of the variable region of the antibody, containing the antigen binding site, is needed. The current methods for antibody detection, identification and isolation of antigen-specific B cells face problems associated with long workflows, low sensitivity, unspecific binding caused by biological material and/or lack of versatility for detection of multiple targets in a single reaction.

In this dissertation, we aim to address these problems by designing new biosensing approaches for the study, detection and isolation of new antigen-specific antibodies. For such, the work was divided in three major aims: Aim I includes the development of an immunoassay based on fiber-optic surface plasmon resonance (FO-SPR) platform for the detection of autoantibodies in plasma samples. For the validation of this assay, the autoimmune disease – immune-mediated Thrombotic Thrombocytopenic Purpura (iTTP) – was chosen as a model disease. Aim II includes the development of a microfluidic tool for the seeding and analysis of single B cells. The goal was to seed B cells in microwells and retrieve a cell of interest using an optical tweezer (OT) setup. Aim III includes the application of immuno-rolling circle amplification (iRCA) for the multiplex detection of antigen-specific antibody isotypes. Such as in Aim I, here we also validate the assay using iTTP as a model disease. In order to provide biological samples for the development and validation of the new methods, a biobank of plasma and peripheral blood mononuclear cells (PBMCs) was created from healthy donors (HD) and characterized for the presence of anti-ADAMTS13 antibodies. This study allowed us to identify 10 from the 416 HD carrying anti-ADAMTS13 autoantibodies. However, due to the low concentration of antibodies present in those positive samples, only the negative plasma samples and PBMCs without detectable anti-ADAMTS13 antibody were used (as negative controls) in developing the new methods described in aims I, II and III. As positive controls, samples from iTTP patients were used in aims I and III.

In more detail, an immunoassay was developed in an FO-SPR platform for the detection of anti-ADAMTS13 autoantibodies. For such, a hexahistidine tagged recombinant ADAMTS13 (rADAMTS13-His6) was immobilized on FO probes using cobalt-nitrilotriacetic acid (Co(III)-NTA) chemistry that enables stable immobilization of the antigen in an oriented manner. The assay was optimized and validated for its specificity using controls with and without the target antibody. As proof-of-concept, a calibration curve was built using a monoclonal anti-ADAMTS13 autoantibody and iTTP patient samples measured on the FO-SPR platform. Data obtained with the FO-SPR assay were compared with the data obtained using the reference enzyme-linked immunosorbent assay (ELISA). These studies confirmed the required sensitivity and accuracy for measuring clinical plasma samples, and demonstrated the applicability of FO-SPR as a fast and automated detection method of autoantibodies in plasma samples.

Next, off-stoichiometry thiol-ene-epoxy (OSTE+) microwell arrays were designed for the seeding and analysis of human B cells isolated from peripheral blood. Due to unspecific binding of cells outside the microwells, this work focused on optimizing the surface chemistry of the microwell arrays to achieve optimal seeding and cell manipulation. In practice, different treatments based on grafting of polyethylene glycol (PEG) with different molecular weights were tested and evaluated using two cell types - human B cells and yeast cells. A reduction of cell adhesion to the arrays as well as high efficiency in lifting the cells using OT was achieved with PEG 2,000 and PEG Mix (equimolar ratio of PEG 500:2,000). The integration of a microfluidic channel allowed controlled seeding and washing with a syringe pump, and validation the platform for detection of IgG+ B cells and evaluation of antifungal activities. The microwell-based microfluidic tool combined with the OT setup can be applied for the detection of antigen-specific memory B cells and for the selection of these cells for further on- or off-chip analysis.

For the detection of autoantibodies and antigen-specific B cells, previous strategies exclusively focused on detection of IgG isotypes, thereby missing information of other isotypes. For this reason, an assay based on immuno-RCA was designed and optimized for the multiplex detection of IgA, IgG and IgM antibody isotypes using antibody-oligonucleotide conjugates (AOCs). The assay was optimized for the detection of anti-ADAMTS13 antibodies and validated using a plasma sample from an iTTP patient. The sensitivity and specificity of the assay was verified and compared to ELISA. The detection of secreted antibodies using immuno-RCA was also confirmed through capture of secreted antibodies from activated B cells and hybridoma cells using microengraving. The developed assay can be incorporated with previously described methods and allow the simultaneous detection of antibody isotypes specific for different antigens.

This dissertation has focused on addressing today’s challenges for the detection of autoantibodies and the discovery of new target-specific antibodies by providing proof-of-concept for new solutions and setups. The presented strategies focused on biosensing, microfluidics and surface chemistry stressing the potential of these approaches for the rapid advance of the antibody discovery field. Many challenges still remain, but we hope this work will inspire more people to keep on working for a better tomorrow.