The reversible phosphorylation of proteins is thought to regulate most cellular processes, but despite the importance of these modifications, scientists lack knowledge about which proteins are modified on which regions. Nonetheless, significant advances in mass spectrometry and protein enrichment have led to large-scale databases such as PHOSIDA, which enumerates roughly 9,000 human phosphorylation sites. This feat, however, still does not encompass all the protein modifications found in various cell types.

Researchers led by Joshua Coon and James Thomson at the University of Wisconsin School of Medicine and Public Health in Madison have taken up this challenge. They have recently published a study1 identifying nearly 12,000 phosphorylation sites in human embryonic stem (hES) cells, including five previously unreported sites on Oct4 and Sox2, the two transcription factors which have, so far, been necessary for all reports of inducing human cells to pluripotency. Coon explains that this study should be looked at as “the first large-scale analysis of the hES cell phosphoproteome, and [it] provides a database — a resource, if you will — for the field”.

The study compares two types of conventional tandem mass spectrometry–based sequencing technology — collision-activated dissociation (CAD) and electron transfer dissociation (ETD), a more recently developed method. CAD has an intrinsic problem with its methodology: the necessary protein processing steps obscure the detection of many phosphorylation sites. ETD peptide-dissociation methodology is based more upon electron capture and transfer than collisions; as a result, those post-translational modifications that are unstable in CAD are far more stable using ETD.

These experiments led to the identification of 11,995 unique phosphopeptides on 4,339 proteins. Of these, 6,182 were found only by ETD, 3,352 found only by CAD and 2,421 found by both techniques. Not only did ETD find many more phosphopeptides, but also it localized the phosphorylation site to a specific amino acid residue about 50% of the time compared with 30% for CAD. ETD was also able to sequence whole classes of phosphopeptides that were previously unobserved. However, the researchers conclude, both techniques should be used to get the broadest view of protein sites that are modified by phosphorylation.

“To fully understand mechanisms governing embryonic stem cell development, these modifications must be systematically analyzed”, says Anthony Whetton of the University of Manchester, UK, whose work has shown that changes in the protein levels of differentiating stem cells often are not well-correlated with changes in gene expression. Though much stem cell research has rightly focused on genes, he says, “the state of proteins has been relatively neglected”. This is a lapse, however, that growing technical expertise is likely to correct. Mass spectrometry is becoming more sensitive and effective, he says, resulting in the removal of “one roadblock in the study of embryonic stem cells at the protein level”.