Research in the Fitzgerald Group is focused on studies of protein folding and function. A combination of covalent labeling strategies (e.g. protein amide H/D exchange and methionine oxidiation) and mass spectrometry techniques are used to investigate the thermodynamic properties of protein folding and ligand binding reactions. Current research efforts involve: (1) the development of new biophysical methods that enable protein folding and stability measurements to be performed on the proteomic scale; and (2) the application of these new methods in the areas of disease detection, diagnosis, and therapy. Outlined below are the analytical methodologies and ongoing applications under investigation in the Fitzgerald Group.
A major part of our research has focused on development of the experimental protocol outlined in Figure 1. The protocol begins with the distribution of a protein or protein-ligand complex into a series of reaction buffers that contain increasing concentrations of a chemical denaturant such as guanidinium chloride (GdmCl) or urea. The protein sample in each reaction buffer is subject to a covalent labeling reaction (e.g., an oxidation reaction with H2O2 or an amide H/D exchange reaction with D2O). After a specified reaction time (a time that is the same for the labeling reactions in each denaturant-containing buffer in the series), the mass of each deuterated protein sample is determined using a mass spectrometry-based readout involving MALDI and/or ESI. Ultimately, the extent of covalent modification (i.e., a Dmass value) is determined for the protein in each denaturant-containing reaction buffer, and these values are used to generate a chemical denaturation curve (i.e., a plot of Dmass versus [denaturant]) at a specific exchange time (see Figure 2). Chemical denaturation curves obtained on protein samples in the absence and in the presence of ligands (e.g., metal ions, small molecules, oligonucleotides, peptides, or other proteins) can be used for the detection and quantitation of protein-ligand binding interactions. Different variations of the basic protocol outlined in Figure 1 have been developed that involve, for example, the use of temperature instead of a chemical denaturant. Recently, we have also interfaced the protocol with a quantitative mass spectrometry-based proteomics platforms. This has provided an avenue by which to study protein-ligand binding interactions on the proteomic scale and opened up the possibility of using protein folding and stability measurements in a number of new applications including drug-mode-of-action studies, disease diagnosis, and the analysis of protein-interaction networks.
Our most mature methodology is SUPREX, which involves the use of amide H/D exchange for the covalent labeling reaction in Figure 1. SUPREX has been firmly established on a number of model protein systems and is currently being used in a number of different interdisciplinary and collaborative projects that require quantitative measurements of protein-protein, protein-metal, and protein-small molecule interactions. On-going collaborative projects in the Fitzgerald group currently include: studies of the protein-protein interactions involved in FbpA-mediated iron-transport in gram-negative bacteria with Professor Crumbliss’ Group (Duke University), studies of small molecule molecular chaperones that rescue the folding of disease related AGT-mutants with Professor Chandra Tucker (Duke University), studies to understand the substrate interactions and molecular mechanism of the protein folding chaperone protein, Hsp33, with Professor Ursula Jakob (University of Michigan), studies to identify novel CypA ligands that might be useful for the development of molecular imaging agents for Lung cancer with Professor Edward Patz (Duke University Medical Center), and more recently studies to understand the detailed mechanism of fibrinectin assembly with Professor Harold Erickson (Duke University Medical Center).
Our newest methodology is SPROX, which involves the selective oxidation of methionine residues in proteins using H2O2. While SPROX is not as fully developed as SUPREX, it has shown great promise for the analysis of protein-ligand binding interactions on the proteomic scale. Following the success of a recent proof-of-principle study, in which SPROX was used in combination with a quantitative mass spectrometry-based proteomics platform to identify yeast protein targets of the immunosuppressive drug cyclosporin A, we have initiated a drug mode-of-action study to identify yeast protein targets of the breast cancer drug tamoxifen. In humans the therapeutic effects of tamoxifen are mediated by the drug’s ability to competitively inhibit estradiol binding to the estrogen receptor. However, a number of nonesterogen-mediated biological activities of the drug have been reported. Interestingly, the drug is cytotoxic to yeast, which do not have an estrogen receptor. While the above nonesterogen-mediated activities of tamoxifen and its yeast toxicity have been identified in cell-based assays, the specific proteins that interact (directly and/or indirectly) with tamoxifen to cause these effects are not known. One goal of our work is to identify these protein targets in order to better understand the molecular basis of tamoxifen side effects in humans.
Also planned and/or under investigation are other proteomic scale applications of SPROX that involve use of the methodology to characterize protein interaction networks, diagnose disease states and to understand the molecular basis of disease. Some of these studies are being conducted in the context of our lung cancer collaboration with the Patz Group (Duke University). One goal of this work will be to determine if protein stability measurements can be used to differentiate lung cancer cells lines from normal lung cells. We also seek to use protein stability measurements to dissect the role of CypA in oncogenesis by comparing protein stability measurements made on proteins in cancer cell lines in which this protein biomarker of lung cancer is over-expressed to protein stability measurements made on the same proteins in cancer cell lines in which CypA expression has been knocked down using stable RNA interference. We expect these studies to elucidate the proteins and biological pathways that are affected by CypA overexpression in lung cancer.