last updated May 7th, 2015

Protein switches
funded by NIH/NIGMS

The incredible complexity of biological systems derives from the high degree of interactions amongst their constituent components. As such, the cell is often described as a complex circuit consisting of an interacting network of molecules. A key component of these networks are protein switches that serve to couple cellular functions. A switch changes its activity state (the output) in the presence of a biomolecular signal (the input). Examples of natural switches include allosteric enzymes which couple effector levels to enzymatic activity and ligand-dependent transcription factors that couple ligand concentration to gene expression. The ability to create switches comprised of any desired input and output functions would enable the rewiring of the cellular circuitry to our own design and has tremendous potential for developing novel molecular sensors, selective protein therapeutics, and as a tool for elucidating molecular and cellular function.

Our directed evolution strategy for switch construction involving the recombination of genes encoding the prerequisite input and output functions for the switch has proven very successful. We have created ligand-activated enzymes with up to 600-fold changes in enzyme activity in response to the presence of a ligand and developed switches in which a cancer marker triggers prodrug activation. We have begun to develop design rules for switch construction and have proposed potential mechanisms of how allosteric functions evolve. Our current research efforts are divided between seeking an understanding of switch mechanisms and applying our techniques to create biosensors and selective therapeutic proteins.

Protein evolution
funded by NSF-DEB and NSF-CBET

The fitness landscape model, as first conceptualized by John Maynard Smith in 1970, is a landmark concept for the fields of molecular evolution and macromolecular structure and function. The model imagines evolution as a process by which a sequence moves by mutations across the fitness landscape. The nature of the fitness landscape fundamentally shapes evolution. Using TEM-1 β-lactamase as a model gene, we are building a comprehensive and detailed map of the effects of mutation on gene/protein/cell function, effects that substantially determine the fitness landscape. We are using this map to test hypotheses on the origin of the genetic code, to evaluate theories for the effects of synonymous mutations, to establish a new framework for understanding mutational tolerance, to understand the nature and origins of non-additive effects of mutation, and to establish relationships between biological fitness and thermodynamic stability of mRNA and proteins. In addition we are using these results to formulate new approaches to applied directed evolution.


Our research is currently funded by grants from NIH/NIGMS, NSF-DEB, and NSF-CBET.


  Targeted DNA methylation using bifurcated enzymes.


  Mutational landscape of TEM-1 β-lactamase (click image to enlarge).