UC Berkeley

T cells discriminate between foreign agonist- and self-antigen peptides presented on Major Histocompatibility Complexes (MHC) proteins that bind to T-cell Receptors (TCRs). Mimicking this process is difficult in vivo due to the monoclonal activity of pMHC. To overcome this, antibodies are commonly used to activate T-cells via conserved regions of the TCR. However, antibodies bind TCRs strongly and activate T-cells via a different mechanism than pMHC binding. We previously reported the binding capabilities of a Fab’-DNA ligand that mimics strong TCR agonists and only activates T-cells when tethered to a supported lipid bilayer. To mimic native pMHC, we created a panel of a human antiCD3 -fab’ (OKT3-Fab’) that more closely mimic native pMHC binding kinetics, but in a polyclonal manner. Single molecule characterization via total internal reflection fluorescence (TIRF) is used to determine dwell times and measure diffusion characteristics. Potential uses for these ligands include measuring the activation of human primary T-cells. Following activation, adapter proteins including growth factor receptor bound protein 2 (Grb2) and son of sevenless (SOS) are recruited to the lipid bilayer. Two-dimensional protein condensation phase transitions on the membrane driven by tyrosine phosphorylation have been identified in several receptor signaling systems, including T cell receptor via the Linker for the Activation of T-cells (LAT) that contains a tyrosine sites and Epidermal Growth Factor Receptor (EGFR) that contains tyrosine sites on the cytoplasmic tail. In the cases of both LAT and EGFR, we have found that the receptor scaffold and Grb2 are sufficient to form a membrane surface condensate in reconstituted systems using supported bilayer membranes. Furthermore, phosphorylation at the Y160 residue on Grb2 is sufficient to block this condensation. To follow up on these reconstitution results in live cells, we investigated Grb2 mediated phase condensation in epithelial cells. CHO cells, a type of epithelial cell with a low expression of Grb2 and no endogenous EGFR, were transfected with various Grb2 mutants and EGFR-eGFP. Clustering of EGFR was observed with all Grb2 dimer mutants, suggesting EGFR clustering is not mediated solely through the Grb2:Grb2 dimer interaction. However, EGFR clustering was prevented with a mutant that blocks the Grb2:SOS interaction. EGFR clustering was recovered with larger amounts of this mutant, suggesting that the Grb2:Grb2 dimer plays a role in EGFR clustering under certain conditions.