Using covalent modifications to distinguish protein electrospray mechanisms: Charged residue model (CRM) vs. chain ejection model (CEM)

Abstract

Different mechanisms have been proposed for the formation of gaseous protein ions during electrospray ionization (ESI). In the charged residue model (CRM) ions are produced upon nanodroplet evaporation to dryness. This mechanism is thought to dominate in native ESI, where proteins retain compact conformations, with charge states close to the Rayleigh charge of protein-sized aqueous droplets. Much higher charge states are generated from proteins that are unfolded in solution. The chain ejection model (CEM) has been proposed for ESI under such denaturing conditions. In the CEM proteins are gradually expelled, while mobile H+ equilibrate between the droplet and its protruding tail. Providing clear-cut evidence for these scenarios remains difficult, because electrosprayed ions do not usually retain any features that reveal their formation mechanism. In this work we propose that the stepwise elimination of basic sites can serve to distinguish between the CRM and CEM. Using cytochrome c as a model system, we studied proteins that had between zero and 19 Lys blocked by acetylation. In native ESI (pH 7) the same low charge states were observed regardless of acetylation. This behavior is consistent with the CRM, where charge states are governed by protein size, rather than protein surface chemistry. Denaturing (pH 2) conditions resulted in much higher ESI charge states. Intriguingly, spectra acquired under these pH 2 conditions gradually shifted to lower charge states when the number of acetylated Lys was increased. This charge reduction is attributed to the fact that lowering the number of basic sites compromises the ability of the protein to compete with the droplet for mobile H+ during the CEM. In conclusion, we illustrate that simple covalent modifications can help distinguish between protein ion formation via the CRM or the CEM.