Poster M19 - A single cell, microdroplet-based approach for antigen-agnostic antibody discovery

Introduction: Antibody-based therapies are used to treat a broad range of diseases, including many cancers, but the methods to develop monoclonal antibodies against specific antigen targets remain highly inefficient. Tumor-specific cell surface receptors have not been identified for many solid tumors, and this lack of antigen identification adds an extra difficulty to engineering appropriate antibodies.

However, tumor-bearing subjects naturally produce antibodies against their tumors, albeit in small numbers. Therefore, if we can probe their antibody-producing cell repertoire, antibodies against the tumor cells of interest could be found without identifying the specific antigen on the cells (i.e. antigen-agnostic).

By utilizing advances in antibody capture technology and next-generation sequencing in microdroplet reactors, we developed a single-cell based method to capture the paired variable heavy and light chain gene transcripts (the genes that determine an antibody’s specificity), as well as the specific antibody protein it secretes. This method yields the antibody repertoire matched to the sequence needed to express these antibodies, which drastically reduces the degree of screening typically needed if only one of this information was available.

Materials and

Methods: We are developing a massively parallel microdroplet-based method for parallel barcoding of antibody protein (specifically IgG) with its paired variable heavy and light chain transcripts from single antibody-secreting B cells. We developed barcoding-hydrobeads capable of (1) covalently capturing and barcoding mouse and human IgG antibody proteins and (2) capturing genes of interest. The hydrobeads are co-encapsulated with a single antibody-producing hybridoma or primary B cell inside an agarose based microdroplet system. The antibody-producing primary cells are isolated from Balb/c mice bearing syngeneic, orthotopic 4T1 tumors as a preclinical model of triple negative breast cancer (TNBC). Antibody production, cell lysis, and barcoding of both the antibody and mRNA transcripts occurs inside the isolated microdroplet system. Once out of droplet, the barcoded antibodies are screened against target cells of interest (e.g. the hybridoma antibody target cell or tumor cell) and the barcodes of antibodies with positive targeting are matched with their similarly barcoded and sequenced variable heavy and light chain genes. These sequences enable expression of these antibodies for further affinity characterization. Barcoded antibodies can also be used to immunoprecipitate its bound cell surface antigen for antigen identification.

Results, Conclusions, and Discussions: Figure 1 describes our method for (A) producing hydrobeads capable of barcoding mRNA and antibody protein and co-encapsulating these hydrobeads with single antibody-producing cells in agarose microdroplets, (B) antibody production and capture in microdroplet, (C) triggered cell lysis compatible with maintaining antibody function, (D) mRNA capture and reverse transcription, (E) collection of barcoded cDNA and barcoded-IgG, (F) next-generation sequencing library preparation, (G-H) two antibody target screening methods, and (I) a proposal for antigen discovery from barcoded-antibodies.

The droplet generation method allows nearly 100% input cell capture in a matter of minutes, making this method amenable to small and rare cell populations. The pipeline is capable of capturing antibodies from hybridomas cells that bind to its known cell targets, and variable heavy and light chain genes of the hybridoma, confirming the accuracy of the capture system. TNBC was chosen as the disease model as it notably lacks all three tumor-specific cell surface receptors for which we have current antibody treatment for. Despite a lack of identified tumor-specific antigen on 4T1 tumors, antibodies collected from 4T1-bearing mice can specifically target these tumors, but not healthy tissue. Our pipeline allows us to identify top binding antibody candidates and their variable sequences for expression for further target affinity characterization.

This single cell, droplet-based approach for high-throughput screening of millions of IgG-secreting primary plasma cells is the first to enable immunoglobulin variable sequence, specificity, and target antigen data to be collected from the same cells. A priori knowledge about target antigens is not required. These advances will be transformative in not only deepening our understanding of humoral immunity, by providing a near complete picture of the humoral response to natural, endogenous antigens/pathogens (i.e. not limited to model systems), but also in accelerating the speed in which new antibodies and their targets are discovered.

Acknowledgements (Optional):