A new acid mix enhances phosphopeptide enrichment on titanium- and zirconium dioxide for mapping of phosphorylation sites on protein complexes

https://doi.org/10.1016/j.jchromb.2009.12.017Get rights and content

Abstract

The selective enrichment of phosphorylated peptides prior to reversed-phase separation and mass spectrometric detection significantly improves the analytical results in terms of higher number of detected phosphorylation sites and spectra of higher quality. Metal oxide chromatography (MOC) has been recently described for selective phosphopeptide enrichment (Pinkse et al., 2004 [1]; Larsen et al., 2005 [2]; Kweon and Hakansson, 2006 [3]; Cantin et al., 2007 [4]; Collins et al., 2007 [5]). In the present work we have tested the effect of a modified loading solvent containing a novel acid mix and optimized wash conditions on the efficiency of TiO2-based phosphopeptide enrichment in order to improve our previously published method (Mazanek et al., 2007 [6]). Applied to a test mixture of synthetic and BSA-derived peptides, the new method showed improved selectivity for phosphopeptides, whilst retaining a high recovery rate. Application of the new enrichment method to digested purified protein complexes resulted in the identification of a significantly higher number of phosphopeptides as compared to the previous method. Additionally, we have compared the performance of TiO2 and ZrO2 columns for the isolation and identification of phosphopeptides from purified protein complexes and found that for our test set, both media performed comparably well. In summary, our improved method is highly effective for the enrichment of phosphopeptides from purified protein complexes prior to mass spectrometry, and is suitable for large-scale phosphoproteomic projects that aim to elucidate phosphorylation-dependent cellular processes.

Introduction

The phosphorylation of proteins on Tyr, Thr, and Ser residues is an essential and frequently occurring mechanism for controlling biochemical processes. The high importance of protein phosphorylation makes the comprehensive identification of phosphorylation sites an important task. Mass spectrometry has been proven to be a suitable and reliable technique for the detection and measurement of peptides in proteomic samples. However, detection of phosphopeptides in complex proteomics samples is still challenging despite the recent major improvements in mass spectrometry, due to their low abundance relative to other proteins in a biological sample. The signals of phosphorylated peptides are often suppressed by signals of higher-abundant non-phosphorylated peptides. Further, the ionization efficiency for phosphorylated peptides by electrospray is lower compared to other peptides. Therefore, to detect phosphorylated peptides it is necessary to reduce the complexity of a sample prior to its introduction into the mass spectrometer for analysis. This can be achieved by performing an additional separation step on the protein level, e.g. affinity purification of a protein complex of interest. Another important technique which can boost the detection of phosphosites is the selective enrichment of phosphopeptides from peptide mixtures prior to analysis.

Several strategies to achieve this aim exist [5]; these can be grouped into chemical derivatization and affinity-based methods. The chemical methods use β-elimination and Michael addition [7], [8] or phosphoramidate chemistry [9], [10]. Affinity-based methods for phosphopeptide enrichment include immobilized metal affinity chromatography (IMAC) [11], [12], [13], [14], [15], [16], [17] and metal oxide chromatography (MOC) [1], both of which are based on the affinity of the negatively charged phosphate groups for positively charged metal ions. IMAC uses Fe3+, Ga3+ or other metal ions immobilized via a linker to a solid support, whereas MOC uses solid metal beads or beads coated with titanium dioxide [1], [4], zirconium dioxide [3] or aluminium oxide [18]. In another chromatographic approach, phosphopeptides are enriched by employing a strong cation exchange (SCX) column, where phosphopeptides tend to elute in the early fractions [19]. SCX is effectively used in combination with prior separation by SDS-PAGE [19], [20] or with subsequent application of IMAC [21] or MOC [22]. The phosphopeptide-enriched fractions are then subjected to RP-chromatography coupled to an appropriate mass spectrometer. One limitation of both IMAC and MOC is that non-phosphorylated peptides containing residues with acidic side chains copurify with the phosphorylated peptides. To prevent this unwanted binding all peptides can be derivatized in a methyl esterification reaction, which converts acidic side chains to methyl esters [16], although this procedure can lead to increased sample complexity due and to substantial sample loss. Therefore, methyl esterification is only applicable if sufficient sample material is available.

Protein complexes isolated from cells are typically of low concentration and give low (silver-stainable) yields. For such samples, phosphopeptide enrichment techniques which require methyl esterification are not suitable. Recently, an improved method for phosphopeptide enrichment using TiO2 microcolumns was published [2], which does not involve any prior chemical modification step. Here it was shown that addition of 2,5-dihydroxybenzoic acid (DHB) or phthalic acid at high concentration during sample loading led to a reduction of the binding of unphosphorylated peptides to TiO2, and thus to an improved selectivity for phosphopeptides. The major limitation of this protocol is that it is not directly applicable to LC–MS/MS analysis, because residual amounts of DHB in the sample contaminate both the LC system and the mass spectrometer. Therefore, we have searched for alternative reagents for the exclusion of non-phosphorylated peptides, and have developed and described an offline TiO2 chromatography approach which is directly compatible with subsequent analysis by online nano-LC–MS/MS [6]. In this method we used a loading buffer consisting of a combination of high concentrations of 1-octanesulfonic acid (OSA) and low concentrations of DHB. This mixture helped to reduce the binding of non-phosphorylated peptides, thereby increasing the trapping selectivity of phosphopeptides, but at the same time did not cause any noticeable contamination of the system. Recently, the successful use of other reagents as non-phosphopeptide excluders such as lactic acid and β-hydroxypropanoic acid [23], glutamic acid [24] and glycolic acid [25] were reported.

We wished to further improve the MOC technique in order to successfully apply it to affinity-purified protein complexes and enhance our ability to identify phosphosites. In the present study we have investigated the effect of an additional ion-pairing agent, namely heptafluorobutyric acid (HFBA) on the phosphopeptide trapping selectivity, in order to improve our phosphopeptide enrichment method. Using HFBA, we have optimized our loading and wash procedure, with the aim of further reducing the retention of unphosphorylated peptides by MOC. Finally, we have compared TiO2 chromatography with ZrO2 chromatography with regard to the enrichment efficiency and selectivity for phosphopeptides. The effect of the different composition in the loading and washing buffer and the column material on the phosphopeptide trapping efficiency and selectivity was tested with a set of 12 synthetic peptides (10 phosphorylated and 2 unphosphorylated) mixed with a tryptic digest of BSA. The optimized enrichment protocol using either TiO2 or ZrO2 chromatography was then applied to the analysis of two affinity-purified protein complexes, which are known to be phosphorylated during mitosis. Both chromatographic media performed similarly well and their application to the samples resulted in the detection and identification of a considerably higher number of phosphopeptides as compared to the previous enrichment protocol, or the analysis of an unenriched sample.

Section snippets

General reagents

All reagents were used as purchased without further purification. Acetic acid (AA) 100% Suprapur, formic acid (FA) 98–100% Suprapur, ACN HPLC grade and ammonia 25% p.A. were purchased from VWR International (Vienna, Austria). Ammonium bicarbonate Ultra (ABC) and HFBA GC grade >99.0% were purchased from Fluka (Buchs, Switzerland). 1-Octanesulfonic acid p.A. (OSA), iodoacetamide (IAA), ammonium dihydrogen phosphate 99.99%, 2,5-dihydroxybenzoic acid (DHB), and bovine serum albumin (minimum 99%)

Results and discussion

We have previously described a reliable protocol for the selective enrichment of phosphopeptides using TiO2-packed tips [6]. Although reliable, this method clearly had significant room for improvement in terms of phosphopeptide selectivity and recovery. Thus, we wished to further improve this procedure by assessing the contributions of two key variables, the mobile and the stationary phases. Firstly, by trying innovative combinations of acids in the loading and washing steps, we aimed to

Acknowledgements

We are grateful to our colleagues from IMP-IMBA Vienna: Ines Steinmacher for support in carrying out mass spectrometry measurements; Mathias Madalinski and Gabriela Krssakova for peptide synthesis; and Andreas Schmidt and Thomas Köcher for useful discussions and a critical reading of the manuscript. This work is supported by the Austrian Proteomics Platform (APP) within the Austrian Genome Research Programme (GEN-AU) and by the MitoCheck project within the Sixth Framework Programme of the

References (47)

  • M.R. Larsen et al.

    Mol. Cell Proteomics

    (2005)
  • L. Andersson et al.

    Anal. Biochem.

    (1986)
  • S. Li et al.

    Anal. Biochem.

    (1999)
  • T.S. Nuhse et al.

    Mol. Cell Proteomics

    (2003)
  • B.A. Ballif et al.

    Mol. Cell Proteomics

    (2004)
  • A. Gruhler et al.

    Mol. Cell Proteomics

    (2005)
  • J.V. Olsen et al.

    Cell

    (2006)
  • N. Sugiyama et al.

    Mol. Cell Proteomics

    (2007)
  • M. Shibue et al.

    J. Chromatogr. A

    (2005)
  • F. Herzog et al.

    Methods Enzymol.

    (2005)
  • T. Hirano

    Curr. Biol.

    (2005)
  • D. Gerlich et al.

    Curr. Biol.

    (2006)
  • P. Cohen

    Trends Biochem. Sci.

    (2000)
  • M.W. Pinkse et al.

    Anal. Chem.

    (2004)
  • H.K. Kweon et al.

    Anal. Chem.

    (2006)
  • G.T. Cantin et al.

    Anal. Chem.

    (2007)
  • M.O. Collins et al.

    Proteomics

    (2007)
  • M. Mazanek et al.

    Nat. Protoc.

    (2007)
  • M.B. Goshe et al.

    Anal. Chem.

    (2001)
  • Y. Oda et al.

    Nat. Biotechnol.

    (2001)
  • H. Zhou et al.

    Nat. Biotechnol.

    (2001)
  • W.A. Tao et al.

    Nat. Methods

    (2005)
  • D.C. Neville et al.

    Protein Sci.

    (1997)
  • Cited by (0)

    View full text