The history of antibody drug discovery reflects the advancement of antibody screening technology. The invention of hybridoma technology in 1975 led to the first paradigm in antibody drug discovery. The first monoclonal antibody (mAb), muromonab-CD3 (OKT3), approved by FDA in 1986 was derived from hybridoma screening. The invention of recombinant DNA technology in the early 1970’s allowed gene cloning and subsequently protein engineering for desired properties. The invention of polymerase chain reaction (PCR) in 1983 and phage display technology in 1985 further enabled batch amplification of antibody genes and isolation of target-specific binding antibodies from combinatorial antibody libraries in a high-throughput manner, which created a paradigm shift in antibody drug discovery. In 2002, adalimumab became the first phage display-derived antibody approved by FDA. It was also the first approved human mAb and is currently the best-selling antibody drug on the market. Phage display is now one of the most widely used technologies for antibody discovery and engineering. Following the invention of phage display technology, various other display technologies, such as yeast display and ribosome display, emerged, which further enhanced rapid discovery of antibodies. Compared to hybridoma technology that retains the natural paring of the heavy and light chains, display technologies allow random pairing of the heavy and light chains, which generates additional antibody diversity. Display technologies also facilitate generating antibodies against low immunogenic targets and engineering antibodies for desired properties. However, hybridoma screening and these display technologies are generally based on antibody binding to recombinant antigens. Following hybridoma screening and phage (or yeast, or ribosome) display, expression, and purification of isolated clones and characterization of purified antibodies for biofunction are mandatory. The workload for antibody production and characterization of isolated clones could be enormous, but the chance to obtain functional antibodies (antibody antagonists and agonists) may be slim.
G protein-coupled receptors (GPCRs) play important roles in many (patho)physiological processes and represent the largest family of drug targets. However, the conformation flexibility and low immunogenicity and antigenicity of GPCRs pose challenges to generating target-specific antibody binders, let alone functional antibodies against GPCRs. If the immunization was successful, hybridoma screening or phage-displayed library panning and screening may lead to the identification of neutral binders and, sometimes, antibody antagonists, but rarely antibody agonists. GPCR activation is a rather complex process. It involves a series of sequential conformational changes that are initiated by ligand engagement and coordinated by the transmembrane domains to eventually expose the G protein binding site in the intracellular side of the receptor. One-third of approved drugs target GPCRs, among which, only two are mAbs, Erenumab, and Mogamulizumab, and both were approved in 2018. Erenumab is a Calcitonin gene-related peptide (CGRP) receptor antibody antagonist derived from XenoMouse immunization and hybridoma screening. Mogamulizumab is a humanized, defucosylated CCR4-binding mAb derived from mouse immunization and hybridoma screening. Mogamulizumab is a neutral binder and functions only in vivo through antibody-dependent cellular cytotoxicity for the treatment of adult T-cell leukemia-lymphoma. Ten GPCR-targeting mAbs are currently in clinical trials, among which, GMA102 is a humanized GLP-1R-binding mAb fused to GLP-1 peptide, and the rest are antibody antagonists or neutral binders conjugated to isotopes. None of these antibodies are antibody agonists. Due to the challenges to identifying antibody agonists to membrane receptors, it has been a common practice to use natural peptide ligands or ligand mimetics fused to Fc or other half-life extending moieties to agonize the receptors for the treatment of diseases. This approach may have the risk that the endogenous peptide ligand is wiped out by anti-drug neutralizing antibodies that may be induced by repeated administration of peptide ligand mimetics. Compared to natural peptide ligands and ligand mimetics, antibody agonist is preferred for its predictable pharmacokinetics and well-developed manufacturing process. Therefore, developing a function-based antibody high-throughput screening method would be of great value for identifying mAbs with desired biofunction to complex membrane protein targets.
We aim to develop a universal high-throughput screening method for direct identification of antibody antagonists and agonists to any GPCR targets of interest. We took advantage of glycosylphosphatidylinositol (GPI) anchoring system and β-arrestin recruitment-based reporter assay. GPI allows proteins, including antibodies, to anchor on the cell surface and keep the anchored proteins functional. More importantly, GPI anchoring sequences can be recognized by GPI transamidase for the attachment to the lipid rafts. GPCRs also preferentially diffuse to the lipid rafts for trafficking and signaling. The co-localization of GPCR targets and GPI-anchored antibodies in the lipid rafts would facilitate the interaction between the target and the anchored antibody on the same cell surface. β-arrestin recruitment upon ligand stimulation is a common phenomenon among GPCRs regardless of G protein-mediated downstream signaling pathways (Gs, or Gi, or G 12/13, or Gq). Therefore, GPI-anchored antibody library cell display in combination with β-arrestin recruitment-based reporter assay may be applied to isolating functional antibodies for any GPCR targets.
We validated this new method using human apelin receptor (APJ), a Gαi-coupled class A GPCR, as a model GPCR target. APJ is involved in cardiovascular function and agonizing APJ is beneficial for the treatment of chronic heart failure. APJ is also involved in angiogenesis and elevation of APJ level was observed in tumors.
