Detection of Multiple Protein Conformational Ensembles in Solution via Deconvolution of Charge-State Distributions in ESI MS
2001-09-15T00:00:00Z (GMT) by
Monitoring the changes in charge-state distributions of protein ions in electrospray ionization (ESI) mass spectra has become one of the commonly accepted tools to detect large-scale conformational changes of proteins in solution. However, these experiments produce only qualitative, low-resolution information. Our goal is to develop a procedure that would produce quantitative data on protein conformational isomers coexisting in solution at equilibrium. To that end, we have examined the evolution of positive ion charge-state distributions in the ESI spectra of two model proteins, α-helical myoglobin (Mb) and β-sheet cellular retinoic acid binding protein I (CRABP I), as a function of solution pH. A detailed analysis of the charge-state distributions over a wide range of pH (2.6−8.5) suggests that each spectrum (i.e., relative ion abundance <i>I</i> vs its charge state <i>n</i>) can be approximated as a linear combination of a limited number of basis functions <i>B</i><i><sub>i</sub></i>(<i>n</i>), i.e. <i>I</i>(<i>n</i>) = ∑<i>b</i><i><sub>i</sub></i><i>B</i><i><sub>i</sub></i>(<i>n</i>). These basis functions (approximated as normal, or Gaussian, distributions) are not significantly affected by the pH variations; however, their relative intensities (coefficients <i>b</i><i><sub>i</sub></i>) exhibit strong pH dependence giving rise to complicated overall charge-state distributions. Analysis of the experimental data, aided by the vast existing body of knowledge of Mb and CRABP I conformational properties (both structure and dynamics) leads to a conclusion that each basis function in fact represents a single conformational isomer. Average charge state corresponding to each basis function (e.g., position of the maximum of <i>B</i><i><sub>i</sub></i>(<i>n</i>) on the protein ion charge scale <i>n</i>) characterizes the conformer‘s overall shape (most likely, projected surface area). The width of each basis function (i.e., standard deviation of the normal distribution) represents the conformer's heterogeneity. Overall, this technique is suitable for analysis of complex mixtures of protein conformational isomers in solution and complements existing experimental methods that are used to study macromolecular dynamics by characterizing protein shape in solution (e.g., scattering techniques).