No abstract

Many protein investigations by electrospray ionization (ESI) mass spectrometry (MS) strive to ensure a "native" solvent environment, i.e., nondenaturing conditions up to the point of gas-phase ion formation. Ideally, these studies would employ a volatile pH buffer to mitigate changes in H(+) concentration that can occur during ESI. Ammonium acetate is a commonly used additive, despite its low buffering capacity at pH 7. Ammonium bicarbonate provides greatly improved pH stabilization, thus offering an interesting alternative. Surprisingly, protein analyses in bicarbonate at pH 7 tend to result in the formation of very high charge states, similar to those obtained when electrospraying unfolded proteins in a denaturing solvent. This effect has been reported previously (Sterling, H. J.; Cassou, C. A.; Susa, A. C.; Williams, E. R. Anal. Chem. 2012, 84, 3795), but its exact mechanistic origin remains unclear. ESI-mediated unfolding does not take place in acetate under otherwise identical conditions. We demonstrate that heating of protein-containing bicarbonate solutions results in extensive foaming, caused by CO2 outgassing. In contrast, acetate solutions do not generate foam. Protein denaturation caused by gas bubbles is a well-known phenomenon. Adsorption to the gas/liquid interface is accompanied by major conformational changes that allow the protein to act as a surfactant. The foaming of beer is a manifestation of this effect. Bubble formation in bicarbonate during ESI is facilitated by collisional and blackbody droplet heating. Our data imply that heat and bubbles act synergistically to cause unfolding during the electrospray process, while proteins reside in ESI droplets. Because of this effect we advise against the use of ammonium bicarbonate for native ESI-MS. Ammonium acetate represents a gentler droplet environment, despite its low buffering capacity.

The mechanisms whereby protein ions are liberated from charged droplets during electrospray ionization (ESI) remain under investigation. Compact conformers electrosprayed from aqueous solution in positive ion mode likely follow the charged residue model (CRM), which envisions analyte release after solvent evaporation to dryness. The concentration of nonvolatile salts such as NaCl increases sharply within vanishing CRM droplets, promoting nonspecific pairing of Cl(-) and Na(+) with charged groups on the protein surface. For unfolded proteins, it has been proposed that ion formation occurs via the chain ejection model (CEM). During the CEM proteins are expelled from the droplet long before complete solvent evaporation has taken place. Here we examine whether salt adduction levels support the view that folded and unfolded proteins follow different ESI mechanisms. Solvent evaporation during the CEM is expected to be less extensive and, hence, the salt concentration at the point of protein release should be substantially lower than for the CRM. CEM ions should therefore exhibit lower adduction levels than CRM species. We explore the adduction behavior of several proteins that were chosen to allow comparative studies on folded and unfolded structures in the same solution. In-source activation eliminates chloride adducts via HCl release, generating protein ions that are heterogeneously charged because of sodiation and protonation. Sodiation levels measured under such conditions provide estimates of the salt adduction behavior experienced by the "nascent" analyte ions. Sodiation levels are significantly reduced for unfolded proteins, supporting the view that these species are indeed formed via the CEM.

No abstract