Collision cross sections of protein ions and noncovalent complexes of proteins studied by electrospray ionization-mass spectrometry

The advent of gentle ionization sources has made it possible to study the higher-order structure of proteins, and noncovalent interactions between proteins and small molecules or other proteins with mass spectrometry (MS). Electrospray ionization (ESI), which allows formation of intact gas phase protein ions, is one such technique. This work focuses on fundamental aspects of the structure and stability of gas phase proteins formed by ESI. Collision cross sections, a measure of ion "size", provide insight into the conformations of protein ions. The first part of this thesis concerns the experimental and theoretical descriptions of collision cross sections of gas phase protein ions. The energy losses of protein ions (myoglobin and cytochrome c), produced by ESI, in collisions with Ne, Ar, and Kr were measured with a triple quadrupole mass spectrometer. The results were interpreted with a drag coefficient model, in which the thermal motion of the target gas, the scattering angle distribution, and inelastic collisions have been effectively included by introducing a drag coefficient to a previously proposed simple hard sphere model. Comparisons of cross sections obtained with different gases and comparisons to literature cross sections measured by the ion mobility method, suggest that a "diffuse" scattering model, suitable for collisions with a rough surface, gives the best description of collisions between protein ions and neutrals. The drag coefficient model can also applied to interpret mobility experiments. The drag coefficient model suggests the projection areas obtained from the ion mobility should be reduced by a factor of about 0.74. Cross section results obtained from the energy loss method and the ion mobility method agree within 3% when both are interpreted with the diffuse scattering model. This model also shows that, for a given charge state, collision cross sections with Ne, Ar, and Kr are similar but have small differences in the order Ne < Ar < Kr. No evidence is found for substantial contributions to the cross section from ion-induced dipole interactions. This work has successfully unified two methods of cross section measurements, energy loss and ion mobility. The second part of this work develops and evaluates a new collision model which can be used to calculate relative energies transferred to protein ions in tandem mass spectrometry. This collision model considers the collision cross sections and the energy losses of ions in the activation process in the collision cell of a triple quadrupole tandem mass spectrometer system. The model can reduce the ca. 250% change in dissociation voltages over a range of pressures from 0.50 to 1.50 millitorr to a better than 10% spread in calculated internal energy required to cause dissociation. Noncovalent complexes of proteins binding a small molecule or another protein were studied using the approaches developed in this work. In the case of myoglobin, highly charged holomyoglobin ions were observed by ESI-MS with a novel continuous-flow mixing setup. Collision cross section measurements show that the protein has unfolded appreciably in high charge states. However, measurements of the energies needed to dissociate heme show that the heme binding energy decreases only slightly in these more highly charged ions. Thus, much of the heme pocket appears to remain in this protein as it unfolds in the gas phase. Further, noncovalent interactions of bovine liver cytochrome bs and a series of yeast iso-1- cytochrome c mutants (wild type, trimethyl-Lys72Ala, Lys73Ala, Lys79Ala and Lys87Ala) were studied by collision cross section measurements and tandem mass spectrometry. The results show that similar energies are required to dissociate gas phase complexes of these mutants with cytochrome 65. This illustrates that these mutations do not cause substantial perturbation for the formation, structure and stability of cytochrome c-cytochrome b5 noncovalent complexes in the gas phase.Science, Faculty ofChemistry, Department ofGraduat