Title: Expanding the Toolbox for Membrane Protein Engineering and Synthetic Biology
Advisors: Michelle O'Malley and Patrick Daugherty
G protein-coupled receptors (GPCRs) act as cells’ sensors through the detection of extracellular ligands and subsequent activation of targeted cellular responses. In humans, GPCRs play a prominent role in regulating physiological processes and are implicated in a range of diseases. Therefore, these proteins are heavily targeted by pharmaceutical drugs and represent emerging targets for gene therapy. Further, their natural capacity as sensors offers a promising platform for inexpensive, robust biosensors capable of detecting molecules with diverse chemical and structural properties. To advance GPCRs towards applications in gene therapy and as sensors, their ligand binding properties must be engineered to bind molecules of interest. However, previous efforts to engineer GPCR ligand binding properties have seen limited success due to the use of conventional directed evolution techniques.
Combining directed evolution, next-generation sequencing (NGS), and a novel in silico analysis pipeline, we extend high-throughput protein engineering methods to distal residues within a GPCR ligand binding pocket. This method was used to engineer a human GPCR, the adenosine A2a receptor (A2aR), with a 15-fold improvement in affinity towards a target ligand. Notably, this method can be used to interrogate distal sites in any protein while directly accounting for high-order epistasis, the interdependence between residues that yields unpredictable, non-additive changes to activity.
Further, we present a novel platform to overcome GPCR screening throughput limitations by coupling receptor signaling to cellular viability in the presence of fluoride. This platform enables stringent selection of functional GPCRs in an engineered, fluoride-sensitive yeast strain using selection pressure that is tunable through exogenous NaF concentration. Accordingly, a GPCR protein library can be depleted of undesirable variants en masse, significantly increasing the number of variants that can be screened in search of those with improved functions.
We leverage the fluoride-sensitive yeast strain to construct a biocontainment platform, which inhibits cell growth upon escape into the environment, where fluoride is abundant. Biocontainment limits the ecological risks associated with synthetic biology efforts, which seek to engineer increasingly complex functions in microbes. Additionally, we develop a novel set of DNA vector selection markers that confer fluoride tolerance to otherwise sensitive cells. The vectors replace those containing conventional auxotrophic and antibiotic markers, which utilize nutrient-limiting growth media and introduce risk of generating antibiotic resistance, respectively. Together, the tools and methodologies presented here improve current capacities to engineer GPCR function and provide systems envisioned to benefit the field of synthetic biology.