The importance of the analysis of signaling pathways has been proven for many years by the elucidation of key signaling molecules. However, in most cases these pathways tend to represent a rather narrow view of the biological state under investigation. Clearly a more detailed understanding of the complexities of cross-talk between signaling pathways is required to further our knowledge of normal and disease processes. The tools that provide the framework for this increased understanding of biology, those that enable identification, characterization, and quantitation of sites of phosphorylation in proteins, have advanced over the past 25 years. This review will present a brief overview of the history of the tools used in phosphorylation analysis and the latest technologies that are being applied in this field, such as mass spectrometry (for broad-based discovery efforts) and flow cytometry (for translation to clinical applications).

Introduction

Phosphorylation-dependent signaling events mediate many cellular processes. As with most aspects of biology, gains in knowledge are led by technological advances; the measurement of protein phosphorylation is no exception. During the last 25 years, a variety of technologies have been developed for determining modification sites of phosphoproteins. Discovery of these sites is being accelerated through the use of mass spectrometric (MS) techniques, which are now being conducted on a proteome-wide scale. In addition, the quantitative measurement of specific phosphorylation sites using immunological reagents and flow cytometric techniques in defined cell subsets is coming of age. These techniques provide an approach that can be applied in the clinic to understand the effects of targeted therapeutics on signaling pathways. To gain a greater understanding of signaling, though functional studies need to be undertaken to determine the cause and effect relationship of specific modifications, these tools enhance our early ability to uncover the complexities of signaling in ways that were not previously possible. This review will provide a brief summary of the history of phosphorylation analysis over the past 25 years, emphasizing the more recent advances in mass spectrometry and flow cytometry. More comprehensive reviews of phosphorylation encompassing a longer time frame can be found elsewhere (1,2,3).

Radiolabeling Methods for Phosphorylation Analysis

Historically, metabolic radiolabeling [³⁵S]-methionine for total protein synthesis, and ³²PO₄ supplemented media (phosphate depleted) to tag in vivo phosphorylation sites, was used for quantitative experiments in cell biology due to its exquisite sensitivity. When enriched or purified components were available, in vitro kinase reactions with [γ³²P]-ATP were used to label products of the reaction. These products were then separated by either one-dimensional or two-dimensional gel electrophoresis (1-DE and 2-DE, respectively) and imaged using autoradiography or fluorography (see (Figure 1)). This approach can be used for analysis of any posttranslational modifications for which radiolabeled precursors are available (4). By combining extracts from the ³⁵S-methionine and ³²PO₄ experiments, separating by 2-DE, and imaging the gel, quantitative analysis of which proteins were phosphorylated is possible. Originally this was accomplished using film-based imaging, but the introduction of storage phosphor imaging in the early 90s allowed more rapid visualization (5). Additionally, through shielding the β-emission of the ³⁵S-label, ³²P-labeled proteins could be identified in the 2-DE image (autoradiographic or storage phosphor) of mixed label samples using appropriate imaging software (5,6,7). However, these radiolabeling studies often required only vanishingly small quantities of total protein for analysis, and so experiments to identify and characterize the protein of interest were repeated with larger quantities of unlabeled protein for subsequent characterization by matching these 2-DE images back to the original experiments.

Figure 1.

For further characterization of the phosphorylated proteins, their identity, nature, and site of phosphorylation needed to be determined. In early studies, phosphoamino acid analysis was used to characterize the nature of the phosphorylation. The isolated metabolically radiolabeled proteins were hydrolyzed, mixed with excess cold phosphoamino acid standards, and separated by thin-layer chromatography (TLC) ((Figure 1)). The locations of the ³²P-phosphoamino acids were revealed by autoradiography and their identity determined by correlation with the migration of ninhydrinstained standards (8). To identify the site (or region) of phosphorylation, a similar strategy was used whereby the gel-separated metabolically labeled protein was subjected to enzymatic digestion and the resulting peptides separated by two-dimensional (2-D) TLC with visualization as before (9). If the protein of interest was known, purified forms of the protein could be subjected to the same methodology to allow sufficient material for the identification of peptides that carried the phosphate. However, confirmatory studies using mutational analysis of the phosphoproteins were required to confirm the phosphorylation site and allow functional analyses to be performed (10,11). These approaches fall into the general schema of phosphorylation site analysis strategies in which 2-D TLC peptide mapping images derived from in vivo and in vitro experiments are compared (12). These approaches were applied in reductionist analysis strategies where limited numbers of proteins were under investigation.