Proteins are one of the most important biomolecules and are involved in a vast number of vital processes inside living organisms. Their structure is closely related to their function and therefore protein misfolding can not only impede crucial biological processes but may also trigger a variety of human disorders such as Alzheimer´s and Parkinson´s disease or type II diabetes. The involved proteins assemble and undergo a characteristic transition from usually unordered conformations into mainly β-sheet rich, insoluble amyloid fibrils. Recent investigations, however, indicate that not the mature fibrils but rather transient, polydisperse and polymorph folded intermediates represent the toxic species in the afore-mentioned diseases. Thus, understanding key factors such as inter- and intramolecular interactions that stabilize the native protein structure or either manipulate or hinder the amyloid formation process is essential for future drug development. Traditional condensed-phase methods are of limited use for the structural characterization of amyloid intermediates, because they only provide ensemble-averages rather than information on individual oligomeric states. In the last decade, ion mobility-mass spectrometry (IM-MS) has emerged as a powerful alternative to investigate amyloid forming systems. IM-MS separates ions not only based on mass-to-charge (m/z) but also on differences in size, shape and charge. More importantly, it can be combined with other orthogonal techniques such as gas-phase infrared (IR) spectroscopy, providing complementary and highly diagnostic information on inter- and intramolecular interactions of m/z- and shape-selected ions. As a result, this combination yields crucial insights on secondary- as well as tertiary-structural transitions of individual oligomeric states. In this thesis, a variety of biomolecular assemblies, ranging from amino acids over peptides to proteins, have been studied using IM-MS with gas-phase IR spectroscopy. First, a novel hydrophobicity scale for amino acids has been developed that in contrast to previous scales now excludes entropic effects of the solvent and therefore more accurately represents the true nature of the side-chain hydrophobicity. In addition, the new scale can be readily extended for non-natural versions, as exemplarily shown for fluorinated amino acid analogues. Subsequently, other effects crucial for peptide and protein folding, in particular charge-dipole interactions and backbone as well as side-chain hydrogen bonding have been studied on a helical model peptide. It is shown that the charge-dipole interaction is most critical for stabilizing the helical structure in the gas phase, while the deletion of a single hydrogen bond only marginally alters the fine structure. Furthermore, it is demonstrated in how far the native structure of two prototypical helical and β-sheet-rich proteins can be maintained in the absence of solvent, and which influence charge has on protein folding in the gas phase. When the proteins are carefully transferred form buffered solution into the solvent-free environment of the gas phase, usually lowly-charged species with comparable sizes to the respective native-structure are observed. In addition, their gas-phase IR spectra show an incredible agreement with their condensed-phase Fourier-transform infrared (FT-IR) and circular dichroism (CD) spectra. Thus, this study unambiguously demonstrates that aspects of the native secondary- as well as tertiary-structure of medium-sized proteins can be preserved in the absence of solvent. When the charges on the protein, however, increase, the native fold is destabilized and the protein refolds into artificial, gas-phase helices. At very high charge states these helices then unravel to form string-like structures that are governed by Coulomb-repulsion and C5-hydrogen bonds. The formation of these string-like structures is independent of the initial protein conformation and therefore can be seen as a new and universal secondary-structure that all proteins can access at very high charge states in the gas phase. Finally, for the first time secondary structural transitions for individual, transient and toxic amyloid intermediates formed by the two peptide sequences VEALYL and NFGAIL have been successfully identified. The data show, that oligomers with as little as four to nine subunits can already contain a significant amount of β-sheet structure. Thus, the characteristic transition from unordered to β-sheet structures already occurs very early in the assembly process of amyloidogenic peptides and proteins. These studies represent the first direct secondary structure analysis of individual amyloid intermediates and highlight the potential of MS-based techniques for the amyloid research.
