Membrane Biology and Immune Signaling

The cell membrane not only separates the intracellular space from the external environment, but also provides a platform for the processing and transduction of inter- or intra-cellular signals. The two-dimensional geometry, the interaction between proteins and lipids, and the local membrane curvature all converge to generate emergent properties of membrane-proximal signaling whereas the underlying mechanisms are not well understood. The T cell receptor (TCR) pathway represents a good example of this scenario. Although major components in the pathway have been identified, it remains unclear how individual components assemble and coordinate to build up a pathway for effectively receiving and amplifying signals from pathogenic antigens. The goal of our research program is to understand the general principals of membrane-proximal signaling under the biological context of  T cell activation and other immune response processes.

How does phase separation regulate lipid signaling?

Our previous work identified T cell microclusters as phase separated structures driven by multivalent protein-protein interactions. These microclusters display liquid-like properties and contain the capability of organizing molecules to promote biochemical reactions such as actin polymerization (Movie 1). Beyond that, T cell microclusters reside underneath cell membranes. It remains totally unclear how these proteinaceous structures interact with lipids and regulating lipid metabolism. Questions we are interested in addressing includes:

  • How do T cell microclusters regulate PIP2 turnover?

  • How do charged lipids regulate T cell signaling?

  • Mechanisms of leukemia-associated mutations in phospholipase C gamma

How does Chimeric antigen receptor (CAR) signal?


The chimeric antigen receptor (CAR) enables T cells to specifically target and kill cancer cells. Despite of its success in clinical trials, the cellular mechanism of how CAR is activated and how activated CAR triggers downstream signaling pathways remains unclear. The domain structure of CAR is very different from the endogenous T cell receptor (TCR), which raises the question of whether CAR activates T cells in a similar mechanism to TCR or not, and how the T cell signaling network accommodates a synthetic receptor. From the clinical side, so far one of the biggest challenges in CAR-T cell therapy is to minimize severe side effects (cytokine storm and neurotoxicity). Understanding CAR signaling will provide clues to designing improved CARs for cancer therapy.

We started by using a supported lipid bilayer system together with TIRF imaging to investigate the dynamics of CAR signaling during T cell activation (Movie 2). Future questions include: 

  • What factors mediate CAR signal transduction downstream of receptor?

  • How does phase separation affect CAR signaling?

  • How sensitive is CAR in response to tumor antigens? 


Signaling on ER membrane

The endoplasmic reticulum (ER) forms an extensive intracellular membrane system that connects all membrane-bound organelles. Compared to the well-studied signaling pathways on the plasma membrane (PM), very little is known about the signaling events occurring on the ER membrane, though a variety of receptors and ion channels have been identified on the ER membrane. The junction sites where ER membranes make close contacts with plasma membranes are particularly interesting. Many calcium channels and lipid transporters are localized there, indicating an active site for signaling and lipid exchange. Questions we are interested in include:

  • What are the tethers making ER-PM junctions in T cells?

  • How is TCR activation coupled to ER-PM junction formation?

  • How does ER-PM junction regulate lipid signaling?

Join us!

Movie 1. Reconstitution of a TCR signaling pathway showed that T cell microclusters (green) promote actin (red) polymerization on the membrane.

Movie 2. Reorganization of chimeric antigen receptor (CAR) after T cell activation by the ligand CD19 presented on a synthetic membrane.