The covalent modification of proteins by phosphorylation of specific amino acids has evolved as a powerful and rapid way for cells to respond to a multitude of internal and external stimuli. Traditionally, the study of phosphorylation has been restricted to a "one experiment-one protein" approach. A paper in this week’s early online PNAS shows how the advent of the mass spectrometer (MS)-which can simultaneously separate and characterize thousands of peptides in a complex sample-has made it feasible to measure the phosphorylation and dephosphorylation of proteins en masse.

Researchers at the Novartis Research Foundation and Scripps Research Institute have put this idea to the test, specifically investigating the modification of tyrosine residues. Led by Eric Peters at Novartis, joint first authors Arthur Salomon and Scott Ficarro first used antibodies to enrich cell extracts for phosphotyrosine proteins. This key treatment was essential to reduce background from phosphoserine/threonine peptides that can carry through the experimental procedure and interfere with the analysis. The authors then examined protein phosphorylation/dephosphorylation in response to two types of stimuli-the activation of human T cells by CD3 antibodies, and the response of leukemia cells to the anticancer agent Gleevec.

T cell activation has been well studied and served as a yardstick by which to measure the MS technique. The authors easily detected modification of the16 tyrosine residues that are known to be phosphorylated upon activation of these cells. Moreover, they found five additional sites on four different proteins, CAS-L, CD3, HS1, and LIM.

Next Salomon et al. turned their attention to the effect of Gleevec on chronic myeloid leukemia (CML) cells. Gleevec is an inhibitor of the fusion protein BCR-ABL, itself a tyrosine kinase; the constitutive activation of BCR-ABL leads to this type of leukemia. Activation of BCR-ABL is caused by phosphorylation of tyrosine 177 on BCR, and tyrosine 393 on the fusion protein. When Salomon et al. treated K562 human CML cells with the anticancer agent, these two sites were dephosphorylated within three hours. In total, 19 sites on nine different proteins lost their phosphate moiety while eleven new phosphotyrosines, each on a different protein, were detected.

The power of this type of global analysis is evident by the fact that the data ties together proteins and pathways previously linked by weak correlative evidence. For example, in K562 cells Gleevec is known to upregulate hemoglobin and the protein CD11b, as does activation of the granulocyte-colony stimulating factor (G-CSF) pathway. Salomon et al. were able to show that three members of the G-CSF pathway, Syk, cortactin, and SHC, are phosphorylated in response to Gleevec, thus providing hard evidence that the inhibitor and pathway are inextricably linked.—Tom Fagan


  1. This study demonstrates the power of current state-of-the-art technology in "phosphoproteomics." The authors study the tyrosine phosphorylation events that occur downsteam of T cell activation. They also apply their method to the inhibition of the signal by Novartis' new drug Gleevec, the first FDA-approved protein kinase inhibitor. It inhibits the BCR-ABL tyrosine kinase oncogene that occurs in certain leukemias. The paper covers much of what was previously known about the two systems under study, and among the 64 (!) identified tyrosine phosphorylation sites are several novel ones. Interestingly, the authors repeatedly claim that their method is rapid. However, given the rather large number of authors and the very elaborate chromatography and MS systems used, this may be a slight overstatement. Yet the future should see more signaling pathways and Drugs "profiled" by phosphoproteomic analysis.

  2. This paper is indeed a very interesting publication that once more demonstrates the power of mass spectrometry in biochemical study when combined with other molecular approaches.

    This paper reported a new method to study tyrosine phosphorylation profiles directly from cells, and it reflected the current cutting-edge methodology in the study of phosphoproteomics. Mass spectrometry has long been used in assisting identification of protein phosphorylation (Fenselau et al., 1985) as well as other posttranslational modifications in biochemical research. In the post-genome era, studying systems biology has become possible. Several new strategies and methodologies for the characterization of the cellular protein phosphorylation profile—or phosphoproteome—have been developed. However, all of the methodologies wrestle with the extremely low copy numbers and the wide concentration range of proteins involved in signal transduction pathways.

    To solve these problems, the authors developed a new method, in which an antiphosphotyrosine antibody was used to selectively isolate proteins with tyrosine phosphorylation from other highly abundant phosphorylated proteins (serine and threonine phosphorylation) by immunoprecipitation. The precipitated proteins were digested with trypsin and the phosphorylated tryptic peptides enriched by methyl esterification and immobilized metal affinity chromatography. The phosphorylation was characterized by reversed-phase HPLC and tandem mass spectrometry. The authors demonstrated the beauty of this new method using examples of T cell activation and chemical inhibition of the BCR-ABL signal pathway. In each case, 12 phosphorylated proteins (22 phosphorylated tryptic peptides containing 23 tyrosine phosphorylations) and 20 phosphorylated proteins (42 phosphorylated tryptic peptides containing 40 tyrosine phosphorylations), respectively, were found. Among these identified tyrosine phosphorylations, 32 are novel phosphorylation sites. These newly discovered tyrosine phosphorylation sites provided new clues for the study of potential protein-protein interactions in the signal transduction pathways.

    This method greatly reduced the background level of serine and threonine phosphorylation and identified more than 60 tyrosine phosphorylation sites in four measurements. However, as the authors pointed out in the paper, "We did not detect all expected phosphorylation sites." In addition, 2-D gel electrophoresis study of signal transduction pathways in mouse fibroblasts following stimulation with PDGF showed that about 260 phosphorylated proteins were detected with an antiphosphotyrosine antibody by Western blot (Soskic et al., 1999). In contrast, the current report only detected a fraction of total cellular tyrosine phosphorylation and most likely the most abundant candidates.

    Another challenging issue in proteomics and phosphoproteomics is the quantification. The authors attempted to measure the changes in tyrosine phosphorylation levels in response to T cell activation by using an external standard (m/z 686). However, figure 3 showed that the peak intensity of such a standard can vary by more than 30 percent {100x(2.2x108 – 1.57x108)/[(2.2x108 + 1.57x108)/2]}. It is very difficult to conduct quantitative measurements with such a big variation.

    In conclusion, this work is a wonderful effort in developing state-of-art methodology for phosphoproteomics. Even there are still have some unsolved problems, this method shows great potential.


    . Phosphorylation sites in riboflavin-binding protein characterized by fast atom bombardment mass spectrometry. Anal Biochem. 1985 Nov 1;150(2):309-14. PubMed.

    . Functional proteomics analysis of signal transduction pathways of the platelet-derived growth factor beta receptor. Biochemistry. 1999 Feb 9;38(6):1757-64. PubMed.

  3. Phospho-state-specific antibodies have played a major role in molecular neuropathology for many years. Some of these antibodies have been monoclonals, generated by immunizing with tissue extracts and then screening for certain histochemical properties (e.g., Alz50 against phospho-tau). Other reagents have been developed as polyclonals, generated by synthesizing chemically phosphorylated peptides that mimic the sites of interest and their neighboring amino acids (e.g., see Czernik et al., 1991). Each of these has its advantages and its drawbacks.

    The new protocol described by Salomon et al. adds mass spectrometry as another tool to the armamentarium of those studying protein phosphorylation. In this particular example, immunoprecipitation-mass spectrometry (IP-MS) was applied to a study of the effect of the antineoplastic agent Gleevec on tyrosine phosphorylation status in a cell line. Anti-phosphotyrosine antibodies were used to recover (P)-Tyr proteins from untreated cells prepared in parallel with similar extracts from Gleevec-treated cells. The respective immunoprecipitates were digested with trypsin and the trypsin digest peptides were subsequently separated by HPLC and subjected to MS.

    The precision of the MS technique permitted the exact identification of 64 unique (P)-Tyr sites on 32 different proteins, some of them unexpected and unpredictable.

    Similar results might well be forthcoming from application of this technology to extracts from primary neuronal cultures, with or without treatment with a particular drug of interest, leading to the discovery of novel signaling pathways. Comparisons of the "tyrosine-phospho-profiles" derived from extracts of normal brain might also be revealing when compared side-by-side against the patterns that appear when brains from humans suffering from neurodegenerative illness are studied by IP-MS.

    The complexity of signaling pathways has become so byzantine that several high-impact journals have developed web environments to help us all keep track. (For example, see Science.) These environments will be particularly useful as technologies such as this point us to new and unexpected interactions between signal transduction molecules.


    . Production of phosphorylation state-specific antibodies. Methods Enzymol. 1991;201:264-83. PubMed.

Make a Comment

To make a comment you must login or register.


No Available References

Further Reading

No Available Further Reading

Primary Papers

  1. . Profiling of tyrosine phosphorylation pathways in human cells using mass spectrometry. Proc Natl Acad Sci U S A. 2003 Jan 21;100(2):443-8. PubMed.