Functional control of soluble rhodopsin mimics using high resolution structure-based design and evaluation

Visualizing the microenvironments of protein/small molecule interactions is the missing link in evaluating the structure-function relationship in many "redesigned" protein systems requiring small molecule binding. The primary goal of this thesis is to manage protein/small molecule interactions to achieve new functions through rational protein engineering in a protein scaffold, which is "evolutionarily naive." The snapshots of each engineering step are collected using high resolution protein crystallography, opening doors to the design strategies of future measures. Finally, the mechanism of the system is elucidated by connecting structural information and biochemical assays. The protein scaffolds used in our study are hCRBPII and CRABPII, belonging to the iLBP protein family. By reengineering their binding pockets, we generated a rhodopsin mimic ligating with small molecules with aldehyde functionalities through protonated Schiff base formation. In Chapter I, we employ the aforementioned strategy to create a new model system based on reengineered CRABPII, mimicking the critical steps of microbial rhodopsin isomerization in a single crystal. Using atomic resolution crystal structures, different mechanisms of retinal/protein interactions with light are demonstrated. Specially, a new photoswitchable protein is identified that does not require chromophore isomerization or a conformational change. In Chapter II, the effect of ligand binding on the conformational states of the domain-swapped dimer of hCRBPII is investigated. A new protein conformational switch is created through a designed disulfide bond that can be activated and adopt new conformations in response to retinal/fatty acid binding and/or reduction potential of the environment. A novel allosterically regulated zinc-binding site is engineered, whose binding affinity can be tuned by the conformational states of our protein. Additionally, using merocyanine, a synthetic fluorophore, a new "swap back" domain-swapped dimer is identified in hCRBPII at atomic resolution, leading to the largest conformational change in the protein. This demonstrates the power of our system to adopt new conformations with different small molecules. Through systematic mutational studies and high resolution crystal structures, the role of the hinge loop region in imposing new conformations/functions in the iLBP family is explored. In Chapter III, the discovery of the domain-swapped trimer as an unprecedented fold for the iLBP family is mentioned. Through a designed disulfide bond and metal- binding site, we are able to favor trimer formation. The mechanism of each step is examined using crystal structures and binding and stability assays. Finally, in Chapter IV, the mechanism of a new class of fluorescent protein tags using the hCRBPII rhodopsin mimic bound with synthetic fluorophores is inspected. By exploiting high resolution crystallography, the microenvironments of protein/ligand interactions is visualized in different fluorescent protein tags applications.