The structural dynamics of proteins and other macromolecules typically serve crucial roles for their respective biological function. While rigid protein structures are used in classic “lock and key” descriptions of enzymology and receptor-ligand interactions, more and more evidence suggest that the majority of molecular interactions occur on the spectrum between induced-fit binding and conformational selection binding. This model of biomolecular interaction requires, to differing degrees, conformation plasticity and dynamics of the protein itself. To characterize the determinants and implications of protein dynamics, there exists no more suited biophysical technique than nuclear magnetic resonance (NMR) spectroscopy. This method is capable of probing the individual atomic nuclei of proteins in a site-specific manner. Furthermore, NMR spectroscopy is unique in being able to access timescales from picoseconds to seconds, providing information on events from bond vibration and libration to protein folding and ligand binding. The breadth of biophysical information accessible by NMR spectroscopy has led to its widespread use in the study of protein dynamics. The work presented herein involves i) the use of NMR for investigation of structure and dynamics in two separate biological systems that demonstrate a high degree of flexibility for folded proteins and ii) the improvement of pulse sequences and methodology for better characterizing picosecond to nanosecond backbone and side-chain dynamics. The organizing principle of this work, which is best exemplified in the structural studies of the piRNA-pathway protein Gametocyte-specific factor 1, is the unmatched capability of NMR spectroscopy to decipher molecular details within dynamic protein systems.
First, the molecular structure and RNA-binding properties of gametocyte-specific factor 1 (GTSF1) of the piRNA effector pathway were investigated. A partially disordered protein with two Zn finger domains, the work presented here describes the isolation of a GTSF1 protein construct amendable to study by NMR spectroscopy. Chemical shift assignment of GTSF1 allowed site-specific observation of amide correlations, which established the basis for NMR structure calculation of GTSF1 and the evaluation of binding to candidate RNA sequences, with goal of the identification of an in vivo RNA binding partner for GTSF1. The work presents compelling data that indicate GTSF1 Zn finger 1 specifically binds a motif GGUUC(G/A) RNA, which in this study was found in the T-arm loop of transfer RNA. Zn finger 2 is affected by the interaction with RNA, but the available structural and binding data indicate that the second Zn finger is a more dynamic, breathable entity, supported by cysteine chemical shift and structural differences between the two GTSF1 Zn fingers. Although it’s currently speculative, the function of GTSF1 might first require binding of RNA to the more stable Zn finger 1, which then leaves Zn finger 2 poised for binding to another molecular species. tRNA-derived fragments that include the T-arm TC loop have been recently implicated in silencing of transposable elements in mammalian cells. GTSF1, which was identified in a genetic screen for piRNA-pathway proteins as vitally required for gene silencing, might plausibly act as a sensor of transcription of transposable elements and help initiate Piwi-piRISCs-mediated chromatin modification and heterochromatin formation.
Next, NMR spectroscopy is used to investigate protein thermostability in psychrophilic (cold-loving) cytochrome c552. Isolated from the bacterium Colwellia psychrerythraea (Cp), previous work has implicated two conserved Cpcyt c552 methionine residues, which are both conserved across psychrophilic and psychrotolerant cytochromes, as acting in dynamical ligand substitution with a third methionine that is the axial heme ligand. It is proposed that elevated backbone dynamics in these methionine residues and the ability for them swap into the axial ligand position accounts for an uncharacteristically high melting temperature (Tm) compared to meso- and thermophile c-type cytochromes. Progress was made in NMR sample preparation and backbone chemical shift assignment of both redox states of Cpcyt c552, and insight from 1D 1H NMR experiments focused on the heme group bound to Cp cytochrome c552 is discussed. Additionally, chemical shifts are used to predict protein dynamics as a first test of a multiple methionine axial ligand hypothesis. Initial data analysis predicts relatively large measures of Random Coil Index for residues surrounding the native axial heme ligand, and shows the hyperfine shifts localized to the residues surrounding the heme. Future experiments will selectively record methyl group dynamics of methionine residues for elucidation of rate constants of methionine substitution and to determine the structural properties of this minor conformation.
Finally, two NMR methodology studies are presented in this thesis: a novel simultaneous-acquisition TROSY pulse sequence for measurement of backbone spin relaxation rates (R1 and {1H}-15N heteronuclear NOE) and a side-chain 2H spin relaxation method for using multifield experimental datasets for better sampling of the spectral density function. Together, these pulse sequences represent significant advancements in NMR measurement of microscopic rate constants and more nuanced detail of protein dynamics.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/d8-rg5e-gf61 |
| Date | January 2019 |
| Creators | O'Brien, Paul |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0022 seconds