Determination Of Pseudocontact Shifts Of Excited States By NMR

This thesis contains two parts. Chapter 1 to 3 describe the determination of pseudocontact shifts (PCSs) of low-populated excited-states using chemical exchange saturation transfer (CEST), and Chapter 4 presents preliminary studies of weak interactions between proteins and ligands using nuclear magnetic resonance (NMR) spectroscopy.Proteins’ excited-states play essential roles in protein folding, molecular recognition and enzyme catalysis. During the past several decades, crystallography and NMR have proven fruitful to determine the ground-state structure of proteins. However, the excited-states remain ’invisible’ for most techniques because their populations are often too low, eg., less than 5%. NMR relaxation dispersion detects the excited-state conformations exchanging with the ground state in the timescale of microsecond to millisecond. Recently, CEST experiments have been developed to extract chemical shifts of the excited-states exchanging in a slower time scale of ～2 to 100 ms. However, building atomic structure models of excited-states solely based on chemical shifts pose anenormous challenge. Hence it urges the development of new methods, which carry more enriched structural information.PCSs carry both distance and orientation information, and in principle can be used to describe conformations of excited-states. Here, we develop a method named PCS-CEST to observe PCSs of low-populated excited states. We measured PCSs of excited-states using 1D/2D CEST. We first validate this method in the slow-exchanging system of Abplp SH3 domain and Arklp peptide, where PCSs of the minor states agree with those of the holo-form ground state as expected. We hence applied this method on HYPA/FBP 11 FF domain with preexisting folding transitional states, and observed that PCSs of the excited states were remarkably smaller than those of the ground state, which indicated the minor states were at least partially unfolded. We also designed a 1D selectively CEST experiment, which time-efficiently scanned chemical shifts of the excited-states at a high resolution. Therefore, PCS-CEST method can be used to extract PCSs of excited-states in slow exchanging with the ground-state. Chapter 4 describes the weak protein-ligand interactions studied by NMR spectroscopy. In the past two decades, fragment-based lead discovery (FBLD) has proven fruitful in drug discovery. Because of the low affinity between targets and fragment hits, it is difficult to obtain complex crystals. Limited NMR restraints are useful to build complex structure models, NMR is hence highly complementary to crystallography. We first applied a fragment-based screening (FBS) for LARG PDZ domain using ligand-observed NMR spectra, then exploited various NMR methods to generate protein-ligand binding models using NMR approaches, e.g., residual dipolar couplings (RDCs), paramagnetic relaxation enhancement (PRE), and PCSs.