The three-dimensional (3D) structures of proteins and peptides in vivo largely determine their biological functions. In vitro these native structures and their heterogeneity reflect a subtle balance between noncovalent intramolecular interactions and those with the surrounding solvent molecules. Decoupling the intra- and intermolecular interactions and revealing the intrinsic structures of biomolecules in the gas phase is crucial for understanding protein-peptide (-protein, -membrane) binding processes and protein folding, and can assist in silico drug design. In addition to this, the spectroscopic studies of (bio)molecules in the gas phase provide crucial information that serves as a benchmark for validation of theoretical approaches, used for structural computations. In this thesis report we demonstrate the use of conformer-selective, Cold Ion Spectroscopy (CIS), combined with mass spectrometry and electrospray ionization technique, to reveal spectroscopic fingerprints of closed-shell (bio)molecular species in the gas phase. These fingerprints provide a stringent test for validating calculated structures of the molecules. Electronic and conformer-specific vibrational spectra of charged molecules, collisionally cooled in a linear radio frequency 22-pole ion trap have been measured using photofragment detection schemes that employs UV and/or IR laser light. A detailed set of spectroscopic and structural constraints, obtained using CIS method for an antibiotic peptide, gramicidin S (GS) allows an unambiguous determination of its 3D structure in the gas phase. The calculated gas-phase structure of GS differs substantially from its known structure in the condensed phase and, thus, provides complementary information for modeling biological activity of this antibiotic. At the next step we form hydrogen-bonded complexes of GS with a gradually increasing number of water molecules. Spectroscopy of such peptide-water clusters targets to bridge the gap between the gas phase and the in vitro structures. We have shown that both, UV and IR spectra of large GS-water clusters remain vibrationally resolved, if the species are cooled to cryogenic temperatures. The observed detailed spectroscopic fingerprints of the hydrated peptides provide a benchmark for their structural analysis. While CIS method was developed for biological species, here we demonstrate its successful application for structural determination of highly-reactive catalytic metalorganic species. The experimentally observed lack of a characteristic NH stretch vibration in the IR spectra of such a species allowed an unambiguous selection of its structures from several low-energy calculated candidates. The current approach for solving structures of biomolecules implies a comparison of a measured spectroscopic fingerprint with that, calculated for a candidate structure. Unfortunately, high accuracy structural computations become time consuming and challenging for large peptides. We revealed a specificity in electronic spectra of phosphorylated peptides, which coordinates with the docking position of a phosphogroup in peptides. This observation, if confirmed for a large variety of peptides, may allow a prediction of phosphorylation sites without a call for calculations. Regarding a relative simplicity of UV spectroscopy we see a potential of this approach as an orthogonal analytical method that assists mass spectrometry in some puzzling cases of peptide analysis. At the end of this work we tried to push the limits of CIS approach towards very large species and, for the first time, measured IR spectrum of a bare protein, cytochrome c, in the gas phase.
EPFL2011
