(500f) Development of a Quantitative Framework for Tuning the Performance of Cell-Based Device

Mammalian cell-based devices represent a promising emerging technology that leverages the natural capabilities of cells â?? including sensing, secretion, and biosynthesis â?? for the treatment of disease. There now exists a substantial set extracellular and intracellular sensor â??partsâ? suitable for engineering mammalian cells to produce a defined output in response to a physiological input. We recently developed one such tool, termed MESA, which is a self-contained receptor-signal transduction platform that enables detection of soluble extracellular cues (via scFv antibody fragment-based ligand binding domains) and in response, modulates the activity of endogenous genes (via a Cas9-based transcription factor). As a proof-of-concept experiment, we utilized these receptors to functionally â??rewireâ? T cells to secrete an immune-potentiating factor when these cells are exposed to an immunosuppressive cue â?? a functionality that is not observed in nature. Thus, this work demonstrated the possibility of engineering or custom-rewiring cellular input/output for a range of applications. However, how one would best implement such a technology to enable robust performance across various contexts, which of particular important for translational applications, remains an open and important question that applies to MESA as well as many other cellular engineering technologies.

Here, we report the development of a systematic framework for evaluating MESA implementation to facilitate tuning biosensor performance in new contexts and for new applications. We first evaluate the effect of genetic context by utilizing Cas9 to stably integrate MESA constructs at specific genomic locations known to be safe harbor loci, comparing biosensor performance between integration sites. Beyond enabling an examination of the effects of integration location on biosensor function, this approach also enables us to systematically examine biosensor performance characteristics within a genetically identical population of cells. We next investigate how relative expression of MESA chains influences biosensor performance. Utilizing synthetic upstream open reading frames to systematically vary translation rates, we evaluate how variations in protein expression influence receptor function. Collection of this quantitative data could in turn enable dynamic modeling of the MESA system, facilitating future analysis and design of MESA systems. Overall, this study addresses the general goal of predictably linking implementation methodologies to the performance of engineered cell-based devices, ultimately enabling the development of safer and more robust therapies.