This research aims to understand how to design and control molecular hinges. The molecular hinges of interest are nano-sized equivalents of door hinges. Such hinges could find applications in new materials or the design of new drugs.

The foundation for this research was the observation that a large, ring-shaped molecule - a so-called macrocycle – prepared by a colleague folded and unfolded rapidly at room temperature. Two research questions arose from this observation: was the hinge behavior unique to this molecule, and could the hinging rate be controlled?

Addressing these questions required the three-step synthesis of a related macrocycle. This new molecule had groups equivalent to putting grit around the hinge's pin. The difference in the rate of hinging motion due to the addition of these groups was observed using a technique called variable temperature NMR spectroscopy.

The results of this work revealed that hinging is a general phenomenon for some of these macrocycles. Second, the 'molecular dirt' designed into this new hinge reduced the rate of hinge motion from 2000 times per second to 20 times per second.

This work is being written up for communication to the Journal of the American Chemical Society based on the novelty of this molecular device and the scientific community's interest in molecular machines.

In the world of drugs, the chemical property that is most important is logP, the predictor of whether a drug can be taken orally and cross the cell membrane. Pharmaceutical companies will not explore molecules with logPs that are outside the ideal range. But what if predictions are wrong? The rules for predicting logP are based on small molecules, but the industry is moving towards large molecule drugs. This poster looks at synthesizing models of large molecule drugs (ring-shaped molecules called macrocycles) to determine if the logP of large molecules can be predicted. Synthesis of a hydrophobic macrocycle shows that the industry predicted logP failed. New prediction methods are needed. To develop these methods, additional macrocycles were made to serve as models for prediction. These molecules also allow us to explore another avenue in drug design challenge another paradigm in drug discovery. Pharmaceutical companies avoid hydrophilic functional groups because of ill predictions about logP. Combining these hydrophilic groups with predictable hydrophobic groups will make the molecule's logP acceptable. That is, by design, the undesirable hydrophilic group is balanced with the desirable hydrophobic group to bring polar groups through the membrane. Overall, the work will allow for a wider range of molecules to be considered for potential drug design.

Oxidative stress occurs when there is an imbalance between free radical activities, including those of reactive oxygen species (ROS), and the body’s natural antioxidant mechanism. To help restore this balance, the Green research group at TCU has developed tetradentate pyridine-containing cyclen macrocycles capable of simultaneously carrying out various modes of antioxidant activities. As drug candidates , these molecules need to be further modified with different functional groups to fine-tune their activities and pharmacological properties, resulting in a large library of up to hundreds of derivative structures. Isoelectric point (pI) and acidity (pKa) play a vital role in assessing the membrane permeability of these molecules. Given the size of the library, experimental determination of these values is an unnecessarily time-consuming endeavor. Using the state-of-the-art Density Functional Theory (DFT), this project aims to 1) show how pI values of any molecules in this library can be predicted with reference to a desired value and 2) predict the pKa of different acidic sites on these multifunctional molecules. This can potentially shed light on the effects of covalent modifications on pI and pKa values, and with further optimizations, can be applied to a virtual screening protocol for any libraries of drug candidates.

Protein crystallization is regarded as a more economically sustainable strategy for achieving protein purification compared to traditional downstream processing chromatography. However, protein crystallization is not a well understood process and still relies on empirical protocols. This work examines the rational design of protein crystallization for lysozyme, a model protein, by exploiting the formation of metastable protein-rich droplets by liquid-liquid phase separation (LLPS). Specifically, sodium chloride, which is a salting-out agent, is used to induce LLPS, while 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) is a salting-in agent used to modulate LLPS conditions. It was found that HEPES enhances protein crystallization from protein-rich droplets. This effect can be explained by examining the relative shift of the LLPS boundary with respect to crystal solubility in the temperature-composition phase diagram. This work suggests that LLPS-mediated protein crystallization may be enhanced in the presence of salting-in agents.

There are a wide variety of unnatural amino acids whose properties could be used to study the structure and function of proteins and create proteins with enhanced or novel functions. The purpose of this research is to develop a method to add unnatural amino acids to proteins via site-specific modification. This is done through aminoacyl tRNA synthetases (aaRSs) which are proteins that attach the correct amino acid to its corresponding tRNA. The loaded tRNA then transports the amino acid to the ribosome where it is incorporated into an elongating protein. Usually, aaRSs have editing domains that remove any amino acids that the synthetase is not specific to. To solve this problem, we have paired Methanobacterium thermoautotrophicum leucyl tRNA synthetase (MLRS) with a removed editing domain with Halobacterium sp. NRC-1 leucyl tRNA to incorporate unnatural amino acids into proteins in Escherichia coli. The binding site of MLRS has been identified, and we have created millions of MLRS variants by randomizing the five amino acids in the binding sites. Using genetic screening procedures, we have identified variants with larger binding sites, and we are currently testing for successful incorporation of unnatural amino acids like dansyl-DAP into the z-domain model protein.