Proteins are the engines of the cell, responsible for most of an organism's structure and function. Protein interactions with nucleic acids, small-molecule metabolites, and other proteins determine the blueprint of development and physiology. Since the advent of recombinant DNA cloning and x-ray crystallography, scientists have sought to determine the principles of molecular function, in order to understand how evolution has brought us to the present day. Others have supplemented nature's protein repertoire with variants improved for a particular function for which there is no selective pressure, e.g., visual display of neuronal activity for a human observer. In recent years, the thermodynamic driving forces of macromolecular structure have been fleshed out; indeed, the field is close to a general solution of the protein-folding problem for average-sized proteins.
In a complementary development, many new techniques have been discovered for the generation and screening of protein variant libraries, to select for beneficial mutations. Numerous protein properties have been improved relative to the naturally evolved sequences; indeed, functions unseen in the natural world have been discovered, either by rational or empirical methods, or increasingly by both. Our lab will use these tools to create reagents to elucidate the function of neural circuits. We will form many collaborations with other researchers to design and test these tools. Of the innumerable possible design goals, I highlight five that we will pursue.
Neurotransmission via signaling molecules and electric potential is the primary means by which neural cells communicate. Reagents that can track these molecules and potentials with high spatiotemporal resolution will facilitate the dissection of complex neurophysiology into discrete signaling events. With Karel Svoboda (HHMI, Janelia Farm Research Campus), we will optimize existing neurotransmitter detectors to improve signal-to-noise ratio, environmental robustness, target specificity, resistance to in vivo malfunction, and temporal response, ideally to the resolution of individual signaling events. We will diversify the available sensor outputs to follow the development of novel imaging modalities and to select the optimal output for a given target. We will also expand the set of detectable ligands to include all neurotransmitters and relevant secondary messenger molecules.
Next-Generation Protein Fluorophores and Imaging Labels
Recent advances in protein engineering and selection, in conjunction with a renewed search for naturally evolved fluorescent proteins, have yielded a new generation of imaging probes, filling the spectrum of possible colors, decreasing sensitivity to environmental parameters, improving folding kinetics and thermodynamics, allowing color changes, and diversifying the attachment-point geometry for protein fusions. We will combine rational design and in vitro and in vivo selection schemes to produce new proteins with defined properties, with an emphasis on those that advance pioneering imaging methods, such as the PALM and optical lattice microscopy developed by Eric Betzig (HHMI, JFRC).
Reengineered Receptor/Ligand Pairs
Binding events between proteins and small molecules dictate biology over short timescales. Organisms have evolved a precise set of neurotransmitters, receptors, and allosteric effectors to optimize communication. Conversely, a number of organisms have evolved toxins to specifically disrupt the signaling events of their competitors. Many neuroactive compounds have been discovered, for some of which structure-function relationships, mode of action, and protein target structures are known. In collaboration with Scott Sternson (HHMI, JFRC), we will design "bumps-and-holes" versions of these neuromodulatory compounds and the receptors to which they bind, thus making new receptor/ligand combinations that function independently. This will allow the activation of particular neural circuits in animals expressing the mutant receptors, without interference from endogenous molecules or other receptors.
Temperature-Sensitive Mutant Library for Flies and Worms
The use of temperature-sensitive gene alleles can facilitate the manipulation of individual genetic circuits while bypassing early lethality in organisms that can survive a thermal shock (fly and worm, and perhaps mouse). Animals are grown at a low, permissive temperature, at which all proteins are functional. Once the animal has grown to an appropriate developmental state, the temperature is increased to a high, selective temperature, at which one or more proteins fail due to their decreased thermodynamic stability. After sufficient time at the restrictive temperature, the function of this protein or proteins is neutralized (a genetic knockout would have failed in development). With Julie Simpson (HHMI, JFRC), we propose the large-scale production of designed temperature-sensitive versions of several critical components of neural signaling pathways, via molecular modeling, site-directed mutagenesis, and screening in vitro and in vivo. In conjunction with targeted expression, this will allow the construction of mutant organisms in which the function of subsets of neurons can be manipulated to define their contribution to behavior.
Structures of Membrane Proteins
Much of biology occurs at the lipid bilayer interface between a cell's "inside" and "outside." This is particularly true for the components of neural signaling circuits, almost all of which are located at or near the plasma membrane. Unfortunately, the complex biophysical properties of this environment are difficult to replicate in an ex vivo setting. This has dramatically hampered the utility of x-ray crystallography and nuclear magnetic resonance, the two principal means of protein structure determination, for membrane protein analysis. With several collaborators in the field, we propose to engineer membrane proteins that are more prone to crystallization than the wild-type versions. We will take as our model system the dopamine transporter, and engineer variants with stabilizing interactions, designed sites, decreased interdomain flexibility, and membrane-simulating derivatizations, in an attempt to discover mutants that form crystals of sufficient quality for structure determination. Knowledge of the structures of these membrane proteins will advance understanding of the molecular basis of neurotransmission. These proteins are also among the most important drug targets in biology.
These "neuroengineering" projects, combined with the pioneering work of other investigators, will help to establish the rules of neural circuit design, facilitating their targeted manipulation. The unifying theme of my research is the discovery of the organizing principles of molecular structure and function, and their extension to complex living systems. I believe that there is no more exciting field than neuroscience, the final frontier of physiology, in which to pursue this goal. We must be judged on the basis of our contributions to the understanding of the brain, but I hope that along the way we learn more about the tiny molecules and machines that compose this amazing organ.