Over the past two decades high-throughput screening of protein-ligand interactions has evolved to become a central tool in the process of drug discovery. The very high-throughput in vitro screening (HTS) strategies are generally based on detection of ligand-binding using highly sensitive spectroscopic techniques such as fluorescence and, more recently, mass spectrometry. These types of approaches typically fail to reveal structural context for the protein-ligand complexes that are detected. For this reason, NMR-based screening, though of lower throughput, has emerged as an important contributor especially for post facto validation of HTS hits. Recently, the design of screening libraries has become more sophisticated. This is particularly true with the fragment-based drug discovery libraries developed over the past decade or so. These libraries are designed to take advantage of advances in combinatorial chemistry with the goal of expanding the flexibility of taking a hit to a lead compound. Unfortunately, fragment-type ligands are inherently weak binders. This introduces a considerable cost in material (in order to reach concentrations necessary to detect binding) and an increased likelihood of false positive binding through non-specific interactions. This proposal will develop a strategy that will largely eliminate both of these concerns. It is based on the nanoscale properties of the reverse micelle. Utilizing the reverse micelle as a confined space, we anticipate being able to 1) efficiently detect weak specific binding; 2) reduce or eliminate non-specific binding; 3) significantly reduce the amount of ligand and protein required; and, most importantly, 4) enter a region of chemical space that is highly desired but difficult to access using existing screening strategies (i.e. weak hydrophilic binders). These advances rely on our recently developed ability to create solutions of individual protein molecules encapsulated within the nanoscale water core of reverse micelles. We will develop this screening strategy for high-resolution solution NMR spectroscopy as well as in a HTS context using fluorescence. Preliminary results indicate the feasibility of this approach. Several proteins of biomedical interest will be used to develop and test the use of nanoscale encapsulation to enhance the discovery of low affinity binders from a fragment library. Dihydrofolate reductase, a historical drug target, will be used to develop many of the strategies. A human aldoketoreductase enzyme (AKR1c3), currently a target for prostate cancer, will also be screened to demonstrate the utility of the approach for large soluble proteins. The myristoylated HIV matrix protein will be examined in its lipid extruded membrane-anchored state and will be representative of a class of proteins that have thus far generally eluded satisfactory screening using more standard methods. In summary, the studies proposed here will demonstrate a significant enhancement in screening of fragment libraries using nanoscale encapsulation and may lead to the discovery of initial compounds directed at these targets related to important human diseases.
Screening of protein-ligand interactions is a central tool in the process of drug discovery. Very small molecule (fragment) libraries promise to improve developing 'hits' into viable drugs. This proposal will use nanoscale encapsulation of target proteins to make identification of fragment hits more efficient and reliable.