(559g) Rewiring Cellular Input-Output to Engineer Cell-Based Therapies That Interface Robustly with Human Physiology

Cell-based therapies have proven to be useful for treating a wide variety of diseases including autoimmune disease, infectious diseases, and cancer.  A particularly promising frontier is the use of engineered cell therapies, wherein cells are programmed to carry out custom functions such as harnessing the immune system to find and destroy cancerous cells.  Such therapies have now achieved robust clinical successes for some applications. However, controlling or modulating the activity of these therapies post-implantation remains both attractive, to enhance both safety and therapeutic efficacy, and challenging, since the tools necessary to engineer custom functional programs are limited. In particular, engineering cells to sense and recognize specific combinations of environmental cues would be desirable, but no such technology has yet been reported.  Toward this goal, our lab has developed a platform for engineering novel protein biosensors, termed modular extracellular sensor architecture (MESA), for detecting exclusively extracellular cues.  Upon binding extracellular ligand, MESA receptors release sequestered transcription factors from the cytoplasmic face of the plasma membrane, freeing the factor to regulate expression of an “output” gene or genes within the cell. Here we present three areas in which we have recently expanded upon the capabilities of the core MESA technology: (1) achieving novel ligand recognition by integrating modular ligand-binding domains, (2) integrating MESA with intracellular gene circuits that enable the cell to “process” sensory information by logical evaluation, and (3) rewiring cellular input-output by coupling extracellular sensing to intracellular modulation of endogenous gene expression. 

First, as modular ligand binding domains, we used camelid single domain antibody fragments termed “nanobodies”. Each nanobody’s small size enables multiple nanobodies to recognize distinct, non-overlapping epitopes on a ligand of interest. Thus nanobody MESA are able to recognize monomeric or asymmetric ligand inputs. After demonstrating the feasibility of nanobody MESA, we integrated multiple such receptors into a logical processing circuit that enables the multiplexed logical evaluation of distinct protein ligand “inputs”. Next, we demonstrated that human antibody-derived scFv could also be used as ligand-binding domains in MESA, and VEGF-MESA were thus designed to sense the tumor-associated cytokine, VEGF. In order to regulate endogenous genes, the MESA transcription factor domain was replaced with the dCas9 protein, which is able to act as a ‘pseudo-transcription factor’ by tethering transactivation domains to the dCas9 protein and providing small guide RNAs in trans to direct the Cas9 protein to a specific genomic locus.  As proof of principle, we developed VEGF-MESA that induce expression of IL-2, a cytokine important for T-cell growth and proliferation. In principle, the dCas9 protein can be guided to any genomic locus of interest by simply delivering a different small guide RNA. Cells expressing VEGF-MESA exhibited minimal background signaling in the absence of ligand, and such cells showed significant increases in IL-2 gene expression upon exposure to ligand. Thus, this initial study demonstrates that MESA may be harnessed to effectively rewire cellular input-output behavior. This work lays the foundation for leveraging MESA receptors to boost the safety and efficacy of cancer immunotherapies as well as cell-based therapies for a range of clinical applications.