Direct quantitation of peptide mixtures without standards using clusters formed by electrospray ionization mass spectrometry

Abstract

In electrospray ionization mass spectrometry, ion abundances depend on a number of different factors, including analyte surface activity, competition between analytes for charge, analyte concentration, as well as instrumental factors, including mass-dependent ion transmission and detection. Here, a novel method for obtaining quantitative information about solution-phase concentrations of peptide mixtures is described and demonstrated for five different peptide mixtures with relative concentrations ranging from 0.05% to 50%. In this method, the abundances of large clusters containing anywhere from 0 to 13 impurity molecules are measured and directly related to the relative solution-phase concentration of the peptides. For clusters containing approximately 15 or more peptides, the composition of the clusters approaches the statistical value indicating that these clusters are formed nonspecifically and that any differences in ion detection or ionization efficiency are negligible at these large cluster sizes. This method is accurate to within approximately 20% or better, even when the relative ion intensities of the protonated monomers can differ by over an order of magnitude compared to their solution-phase concentrations. Although less accurate than other quantitation methods that employ internal standards, this method does have the key advantages of speed, simplicity, and the ability to quantitate components in solution even when the identities of the components are unknown.

Figure 1

ESI mass spectra of a solution containing 1% molar fraction of [ala]₃-leucine enkephalin in leucine enkephalin (2.4 mM total peptide concentration) measured at external accumulation hexapole offset potentials of: a) 2.8 b) 3.6 and c) 4.0 V. Regions of the spectra showing molecular clusters are inset.

Figure 2

a) Normalized abundances of protonated A₃LE (triangles) and LE (squares) from ESI mass spectra as a function of A₃LE molar fraction for solutions with 2.4 mM total peptide concentration. b) Molar fractions of A₃LE calculated from the abundances of the protonated molecular ions from an equimolar solution with LE as a function of total peptide concentration. Error bars correspond to the standard deviation of three replicate mass spectra obtained from three different nanoelectrospray tips at each solution concentration. Dashed lines represent the ideal trend based on solution concentration.

Figure 3

Expanded region of ESI mass spectrum in Figure 1a showing homogeneous and heterogeneous clusters with 19 constituent molecules corresponding to [19LE + 5H]⁵⁺ and [18LE + A₃LE + 5H]⁵⁺. Minor peaks correspond to adducts, such as Na⁺, K⁺ and acetate, and other cluster sizes.

Figure 4

The % molar fraction calculated from the cluster ion intensities assuming statistical incorporation and identical ionization efficiencies obtained from ESI mass spectra of a 1% molar fraction solution of A₃LE in LE (2.4 mM total peptide) as a function of cluster size, n. A dashed line represents the ideal trend based on solution concentration.

Figure 5

Stickplots representing three theoretical distributions of heterogeneous cluster ions of a given size as a function of increasing cluster size or increasing molar fraction (a–c). Each distribution corresponds to a single cluster size, n, in which each line corresponds to the intensity of a cluster ion that contains some number of impurity molecules, h. The dashed line, N, represents the noise level.

Figure 6

The % molar fraction of A₃LE in LE obtained from the protonated molecular ions (circles) and from the cluster abundances for n ≥ 17 as a function of the solution % molar fraction. The cluster data is linear with a slope of 0.92 whereas the protonated molecular ion data has a slope of 14.2 and is not well fit to a line. Error bars correspond to a standard deviation of three replicate spectra at different DC offset voltages of 2.8, 3.6 and 4.0 V.

Figure 7

The % molar fraction calculated from cluster ion intensities obtained from ESI mass spectra of a 1% molar fraction BK1-8 with BK (2.4 mM total peptide concentration) as a function of cluster size, n. A dashed line represents the ideal trend based on solute concentration, assuming statistical incorporation of subunits into clusters.

Figure 8

The % molar fractions calculated from the protonated molecular ions (squares) and cluster ions (triangles) as a function of the % molar fraction in solution for two component mixtures consisting of a) ME with LE, b) BK2-9 with BK, c) BK1-8 with BK and d) KLE with LE. Data from clusters with n ≥ 12, 15, 15 and 20 subunits were used for these respective mixtures. Best-fit slopes of 0.95, 1.19, 1.08 and 1.10 are obtained for the cluster data a–d, respectively (see text).