Quantification of protein phosphorylation by liquid chromatography-mass spectrometry

Abstract

The identification and quantification of specific phosphorylation sites within a protein by mass spectrometry has proved challenging when measured from peptides after protein digestion because each peptide has a unique ionization efficiency that alters with modification, such as phosphorylation, and because phosphorylation can alter cleavage by trypsin, shifting peptide distribution. In addition, some phosphorylated peptides generated by tryptic digest are small and hydrophilic and, thus, are not retained well on commonly used C18 columns. We have developed a novel C-terminal peptide (2)H-labeling derivatization strategy and a mass balance approach to quantify phosphorylation. We illustrate the application of our method using electrospray ionization liquid chromatography-mass spectrometry by quantifying phosphorylation of troponin I with protein kinase A and protein kinase C. The method also improves the retention and elution of hydrophilic peptides. The method defines phosphorylation without having to measure the phosphorylated peptides directly or being affected by variable miscleavage. Measurement of phosphorylation is shown to be linear (relative standard error <5%) with a detection limit of <10%.

Figure 1

Cleavage of a hypothetical protein XYZ into peptides before measurement by mass spectrometry. Protein XYZ contains two cleavage sites that should generate peptides X, Y, and Z unless there is miscleavage producing peptides XY and YZ. C₁ and C₂ define the probability of cleavage at each site. The terms k_i represent the intrinsic ionization efficiency of each peptide (i), N represents the amount of XYZ prior to digestion, and I_i represent the ion current measured by the mass spectrometer for each peptide.

Scheme 1. Analytic Scheme Used for Phosphorylation Analysis

The protein sample is split into two fractions. One protein fraction is treated with phosphatase and serves as an internal standard. Each fraction is digested with trypsin. The peptides in each fraction are differentially labeled with propylation reagent-d₀ or -d₄, and then the fractions are recombined. Ion current ratios are measured for each pair of labeled isotopologues.

Scheme 2. Scheme Used to Create Protein Samples with an Increased Mole Fraction of Phosphorylation

Figure 2

ESI-LC−MS chromatogram of underivatized SSANYR (m/z = 697.3) and derivatized propyl-SSANYR (m/z = 739.3) peptides. The addition of a single propyl group increased the retention time away from the void volume (∼5 min) by ∼15 min.

Figure 3

Peptide elution profiles of of d₀- and d₄-isotopologues with varying degrees of propylation. Shown are the [M + 2H]²⁺ ion currents versus time for peptides: (A) NITEIADLNQK (m/z = 692.8 and 698.8), (B) ISADAMMQALLGAR (m/z = 766.4 and 770.4), and (C) FKRPTLR (m/z = 480.6 and 482.6). These peptides contain 3, 2, and 1 propyl groups and 12, 8, and 4 deuteriums, respectively. The total separation at the peak apexes are 5.5, 2.1, and 2.0 s, respectively, with an average peak width at ¹/₂ height of 9.6 s.

Figure 4

Linearity and precision of isotopologue ratio measurements. Data shown are the measured ion current (I₀/I₄) ratios for each peptide listed in Table 1 for samples prepared with defined mole d₀/d₄ ratios ranging from 0:1 to 1:1 d₀- to d₄-labeled peptides. A linear response in the measured ion current ratio, I₀/I₄ = (0.956 ± 0.005) (d₀/d₄) + (0.025 ± 0.003) with r² = 0.999 was observed as a function of mole ratio of unlabeled (d₀) to deuterated (d₄) peptides.

Figure 5

Measurement of d₀/d₄ isotope ratios as a function of dilution of PKA-treated peptides with dephosphorylated peptides. Dilution of PKA-treated peptides correlates to an increase in the isotopologue ratio for phosphorylated peptides. The x-axis lists the peptides in Table 1 by numbers, and the y-axis plots ion current ratios (I₀/I₄) measured for each peptide across a dilution range of PKA treatment. Symbols indicate the mole fraction of PKA treatment in each sample ranging from 0 to 100%. The measured I₀/I₄ ratios of peptides SSANYR and ISASR decrease with increased PKA treatment and resulting phosphorylation. The reference peptide ratio used was the mean ratio of the following five peptides: EAEER, CQPLELAGLGFAELQDLCR, IFDLR, KLQLK, and TLMLQIAK.

Figure 6

Demonstration of linearity of phosphorylation measurement. The degree of phosphorylation was determined for peptides SSANYR (●) and ISASR (○) from the samples shown in Figure 5. A linear response was observed for SSANYR, p(%) = (0.937 ± 0.017) p_Rx(%) + (5.6 ± 6.4); r² = 0.998 and for ISASR, p(%)= (0.438 ± 0.036) p_Rx(%) + (8.7 ± 9.5), r² = 0.961, where p is the fraction of peptide that is phosphorylated and p_Rx is the degree of PKA treatment.