Large Molecule Characterization Based upon Individual Ion Detection with Electrospray Ionization-FTICR Mass Spectrometry.

We report a new method for mass spectrometric measurements of high-molecular-weight species based on the summation of sequential Fourier transform ion cyclotron resonance (FTICR) spectra of individual multiply charged ions. This approach produces statistically useful mass spectra for large multiply charged molecular species formed by electrospray ionization and circumvents conventional limitations upon achievable resolving power and precision for high-molecular-weight species which arise due to Coulombic constraints. For very large molecules with tens to thousands of charges each, the total number of charges required to define the charge-state distribution, and thus provide accurate mass information, greatly exceeds the useful charge capacity of the FTICR cell. As trapped ion populations approach or exceed this capacity, FTICR performance degrades due to large frequency shifts, peak coalescence phenomena, and rapid loss of ion packet coherence, which effectively precludes high-resolution and precision measurements for molecules above ∼80-kDa size for a 7-T magnetic field strength. The present approach is based on the summation of many spectra having moderate populations of individual ions and relies on sensitivity sufficient for individual ion detection. While the number of trapped ions contributing to each mass spectrum may generally be insufficient to define the isotopic or charge-state distributions (and thus produce accurate information on the molecular weight distribution in a conventional fashion), the present data processing and summation approach suppresses the noise component (as well as smaller signals) that would otherwise be problematic. Importantly, this approach circumvents natural limitations for very high molecular weight species due to Coulombic interactions and thus provides a basis for much greater resolution and mass measurement accuracy than otherwise possible. This paper presents the details of this approach and its demonstration for the 66-kDa protein bovine serum albumin (where the conventional approach is also feasible) and discusses important aspects of the data manipulation.