Protein phosphorylations—and increasingly other post-translational modifications (PTMs) such as acetylation and ubiquitination—are thought to be the principle means of intracellular communication. When genes are knocked in, out, or down, a receptor is bound, or the cell is perturbed in any other way, it responds with cascades of phosphorylations on different sites on a host of proteins in a variety of signaling pathways, nodes, and networks. Learning and tracking exactly when and where those changes occur goes a long way to understanding, mapping, and manipulating how those pathways, nodes, and networks interact and function, and with them health, disease, and mechanisms of drug action.
Forest White’s lab at MIT focuses on therapeutic resistance—asking what are the pathways by which a tumor continues to grow in the presence of, say, EGFR inhibitor? This can be done by sequencing the genome to look for mutations that emerge upon drug treatment, or by looking at transcriptomic changes in naïve versus treated tumors. “But both of those are basically proxies where you’re trying to identify activated pathways,” he says. Liquid chromatography coupled to mass spectrometry (LC-MS) lets you “quantify the phosphorylation sites … and therefore actually read out pathway activation.”
“The first pass should really be ‘discovery mode’ analysis, where you’re basically just doing a data-dependent analysis and letting the instrument tell you which peptides are present,” he says. (It’s called “data-dependent” because the MS uses the information about which are the most prevalent peptides to make real-time decisions about which to fragment and analyze.) Look at those in, say, ten samples.
“Now you want to go back and analyze 1,000 more tumors, and you want to make sure you see exactly those same peptides that you saw in the first ten,” he says. At that point the instrument can be programmed to follow only the peptides of greatest interest, using a targeted approach. It’s considered data-independent in that spectra of fragmented peptides are collected based on how the instrument is set, not on what it sees during the run. Data is then analyzed post-acquisition.
In a targeted method “we’re not trying to see everything in the phosphoproteome. Instead, we have phosphosites that we either already know or that we think might be important, and we want to study them in different conditions,” explains Amanda Paulovich, member, clinical research division at the Fred Hutchinson Cancer Research Center.
Enrichment
Roughly speaking, maybe 1 in 1,000 tryptic peptides will be a phosphopeptide. A cell lysate may contain about 10-15,000 proteins, with about 100,000 phosphosites. Some of these may be present at one copy per cell, while many orders of magnitude more can be found for others. And if you work on phospho-tyrosine, like White’s lab does, “that’s 1/1,000th of the phosphoproteome—so it’s effectively one millionth less than the total proteome.”
With all the recent furor about antibodies failing to live up to their claims, much effort is being put into determining and documenting what they actually detect.
All this is to say that it’s important to have a very fast, sensitive, high mass-accuracy MS, and to “get the enrichment right.”
White’s lab likes to multiplex samples. So after lysates are digested, peptides are labeled with an isobaric multiplex tagging reagent (like iTRAQ® or Tandem Mass Tag® (TMT®)) and mixed together. “Then we would take these samples, do phospho-tyrosine peptide immunoprecipitation [IP] where we pull out phospho-tyrosine peptides. Then we have a second stage of enrichment where we actually take that and pass it over an IMAC [immobilized metal ion affinity chromatography] column. This helps to decrease the nonspecific binding that occurs during the IP step.”
IMAC and TiO2 are common ways to enrich for phosphopeptides—mostly phospho-serine (which is about 90% of the phosphoproteome) and phospho-tyrosine (about 10%). They each enrich different but overlapping spectrums of phosphopeptides. Thermo Fisher Scientific introduced a protocol at the American Society for Mass Spectrometry meeting in early June called SMOAC (serial metal oxide affinity chromatography) that takes the washes and flowthrough from titanium enrichment as the input for IMAC enrichment. “Using that serial approach we almost doubled the number of phosphopeptides,” explains John Rogers, the company’s senior R&D manager, mass spectrometry reagents.
Motif and Specific IP
Sometimes it’s desirable to further decrease the number of peptides in the sample. Selective antibodies to phosphoserine or phosphothreonine resides per se are not available, but antibodies are often used to immunoprecipitate phosphopeptides containing extended motifs. For example, antibodies against a kinase substrate motif will enrich for phosphopeptides containing that motif.
Cell Signaling Technology, for example, makes motif antibodies that work really well for things like MAP kinases or Akt, points out White. But “they don’t work so well for the Src-family kinases because they don’t really have a motif, it’s basically acidic residues anywhere near the phosphorylation site.”
“That sort of global enrichment with motifs or IMAC, people use in their early-stage discovery where they’re trying to find out what proteins get phosphorylated, which phosphorylations change with these perturbations,” points out Christie Hunter, director, omics applications at Sciex.
“You can develop antibodies for phosphopeptides that will be reasonably specific—you’re going to be enriching exactly that peptide from exactly that protein, and then analyzing it by MS,” she adds.
Paulovich’s lab looks at tumor suppressor proteins and the cellular response to DNA damage. “Our cells get a lot of DNA damage every day, and they repair that damage using a whole ensemble of proteins that carry out a phospho-based signal transduction network that the cell uses to respond to that damage,” Paulovich relates. “We’ve built a targeted assay panel that can let folks measure these tumor suppressor proteins in their phosphorylated and unmodified forms … and Sciex has picked that up and has been working to convert that into a commercial kit to make it widely available to the community. They’re beta testing it right now.”
Targeted mass spectrometry techniques, such as MRM (for the MS modality of multiple reaction monitoring), have been widely adopted by the MS community, she adds. Individual assays are available from open-source portals such as assays.cancer.gov (funded by the NCI’s Clinical Proteomic Tumor Analysis Consortium [CPTAC]), for example. Kits—allowing the technique to be more broadly adopted among biologists—will likely include antibodies and standards as well as sample prep and MS protocols.
How Specific?
With all the recent furor about antibodies failing to live up to their claims, much effort is being put into determining and documenting what they actually detect. One of the ways Rogers’ team addresses the issue is to use an antibody to immunocapture phosphoproteins, and then use MS to ask if it identifies the right target, and what else it identifies. “MS is really the only tool that allows you to figure out what else an antibody might bind,” he says.
Biocompare’s Proteomic Supplies Search Tool
Find, compare and review proteomics
tools from different suppliers Search
An antibody that recognizes multiple targets may yield a smear, or fail to recognize that both a phosphorylated and nonphosphorylated protein migrated together, on a Western blot. Similarly, other immunoassays such as immunohistochemistry and ELISA may give ambiguous or erroneous results.
But with MS, if the protein you care about binds to the antibody along with 100 other proteins, that’s okay with Paulovich: “We’re going to pull it back off the antibody and put it into an MS, and we’re going to get a mass spectrum of exactly what we’re measuring.”
Image: Dreamstime Images