by Jeffrey M. Perkel
Not that long ago, analyzing the phosphorylation of a given cellular protein could best be described as painful. If a researcher wanted to know whether a particular protein's phosphorylation status changed in response to drug treatment, he or she would have to load the cells with radioactive phosphate, treat the cells with drug, extract protein, immunoprecipitate with an antibody to the protein of interest, and electrophoretically resolve the resulting material on a gel. Alternatively, if a global view of phosphorylation was required, he could skip the IP step and resolve whole protein extracts, instead.
Both cases present multiple difficulties. For one thing, they employ significant quantities of radioisotopes, presenting handling problems, tying up cell culture incubators, and producing copious radioactive waste, among other things. But more importantly, at the end of all this work, the researcher still wouldn't know just which amino acids' phosphate statuses had changed (if at all), nor to what extent.
And that's important, says Michael Browning, president of PhosphoSolutions, a phospho-specific antibodies vendor in Aurora, CO, "because some [modification] sites can activate a protein and others can deactivate it, and if you don't know which one is phosphorylated you don't know what the functional consequence of the event is."
Today, interpreting those functional consequences is a whole lot easier, thanks to a diverse phosphoprotein toolbox that includes everything from gel stains
and antibodies, to purification resins and mass spectrometers.
The simplest (and coarsest) way to assess protein phosphorylation is to do so en masse, for instance using Life Technologies' Pro-Q® Diamond phosphoprotein gel stain (fluorescent) or Thermo Scientific's Pierce GelCode™ Phosphoprotein Staining Kit (colorimetric). Resolve a protein extract on a one- or two-dimensional polyacrylamide gel, stain with dyes against both total protein and the modification itself, and the result is a 30,000-foot view of the overall phosphoprotein landscape of the sample.
For a finer grained analysis, researchers often use phosphospecific antibodies. These generally detect discrete phosphorylation events, such as calmodulin kinase II phosphorylated on threonine-286, though so-called "pan-phosphoantibodies," such as the 4G10 antibody that detects all phosphotyrosine residues, also exist.
Usable in any assay that requires an antibody, from Western blots and immunohistostaining to ELISAs and flow cytometry, phosphospecific antibodies are, says Browning (whose company offers some 200 such reagents), "designed to allow you to analyze a specific site in a quantitative way."
Millipore has more than 500 different phosphospecific antibodies in its catalog, and releases about 50 every month, according to Product Manager Michele Hatler. With such a catalog, it's not surprising that the company also supports several different antibody-based approaches to protein phosphorylation.
Among those options are Millipore's FlowCellect™ PI3K-mTOR Signaling Cascade kit, which probes cell signaling pathway activation via flow cytometry, and its MILLIPLEX EpiQuant Cell Signaling kit, which couples phosphospecific antibodies with Luminex's xMAP platform to enable multiplexed analysis of up to 40 different phosphoproteins at once.
"The quantitation is the key," says Hatler. "[The EpiQuant system] allows quantitation and multiplexing at the same time."
Unlike traditional Luminex assays (including the company's MILLIPLEX MAP cell signaling kits), which detect intact proteins, EpiQuant kits target digested protein peptides, reducing the cross-reactivity that sometimes complicates phosphospecific antibody work.
A similar concern motivated the development of Olink Bioscience's Duolink assay. Based on the company's proximity ligation assay (PLA) technology, Duolink was designed for enhanced sensitivity and specificity in immunohisto- and immunocyto-chemical applications, because it, like sandwich ELISA and PCR, relies on two distinct binding events to generate a signal, says Chief Technology Officer Mats Gullberg.
"There are a lot of antibodies for phosphoproteins, but they are hampered because cross-reactivity is rather a big problem for those antibodies," Gullberg says, "usually far worse than for antibodies against total protein."
In Duolink, the two detection events are mediated by two antibodies, one to the protein itself and one to its modification. These two reagents are applied to cells on a cover slip or microscope slide, followed by secondary antibodies. But, instead of being conjugated to a fluorophore or colorimetric enzyme, these secondary antibodies are each coupled to a distinct oligonucleotide.
To develop the reaction, two additional oligonucleotides are added, along with DNA ligase. If the protein is phosphorylated, the total protein and phosphoprotein antibodies will be close enough that their oligo tails and the added oligos can form a stable complex and undergo ligation to form a closed circle. That circle is then amplified up to 1,000 times by rolling circle amplification and detected by hybridization with a fluorescent probe.
According to Gullberg, the signal generated by this process is so strong that it produces quantitative, rather than qualitative data on protein abundance and modification. "You get a count of numbers of objects in your image, which is a better way to quantify than to measure, say, overall fluorescence," he says.
Not every phosphorylation assay requires an antibody. Amy Zumwalt, proteomics marketing programs manager at Thermo Fisher Scientific, says phosphorylation analysis is a major application for proteomicists, as well. "A huge percentage of our proteomics customers are looking at phosphorylation," she says.
Zumwalt advises her phosphorylation-focused clients to consider either an LTQ Velos ion trap or LTQ Orbitrap Velos (Orbitrap-ion trap hybrid). Both have the speed and sensitivity required to sequence peptides and detect the 79-odd-dalton mass difference that is the hallmark of phosphorylation. But more importantly, both can fragment (and thus, sequence) peptide ions using electron transfer dissociation (ETD).
ETD fragmentation, Zumwalt says, "is particularly useful for post-translational modification characterization," because it (unlike the more common collision-induced dissociation) tends not to break the bond between the phosphate group and its attachment point. Thus, it "allows you to both characterize where the phosphate came from and do the sequencing of the peptide backbone."
Ion traps and Orbitraps are most commonly used for phosphopeptide discovery work, Zumwalt says. That is, they are used (much like gel stains) to get a 30,000-foot overview of proteins and peptides whose phosphorylation status changes under a given set of conditions. Once a researcher has identified a small set of peptides of interest, however, she advises switching to a triple quadrupole-based instrument, such as Thermo Fisher's TSQ Vantage, as these can be used in more quantitative, multiple reaction monitoring experiments.
Whether you plan to use a mass spectrometer or antibodies for your phosphorylation research, you'll need to think about up-front sample preparation. A number of companies offer products to simplify and improve the efficiency of this process.
Roche Applied Science's PhosSTOP is a tablet phosphatase-inhibitor cocktail that can be used to preserve protein phosphorylation during extraction procedures—a key concern, given the prevalence of phosphatases in cellular lysates.
"If you don't somehow protect [your] protein from phosphatases … you could be making the wrong conclusions because your protein isn't the same as it was before purification," says marketing manager Steve Bye.
At the recent American Society for Mass Spectrometry annual meeting, Thermo Fisher Scientific released two new purification tools to improve phosphopeptide sample preparation following extraction: the Pierce Fe-NTA Phosphopeptide Enrichment Kit and the Pierce Graphite Spin Column.
According to Zumwalt, these two columns should be used in a tandem—Fe-NTA followed by graphite—prior to sample injection onto the mass spectrometer. Both selectively bind phosphopeptides, but the graphite kit also helps remove many of the contaminants that can reduce mass spec efficiency, such as urea. “Our data showed that both [the commonly used purification matrix] TiO2 and Fe-NTA bind singly and multiply phosphorylated peptides simultaneously, but enrich for different populations of unique phosphopeptides, so both have utility to researchers who are studying the phosphoproteome,” Zumwalt says. Fe-NTA binds preferentially to multiply phosphorylated peptides while TiO2 binds preferentially to singly phosphorylated peptides.
If you would rather stick with antibody-based approaches, be advised that all antibodies are not the same. Quality matters, and certainly varies. Browning advises ensuring that your antibody has been properly validated. Ideally, that means detecting a single band on a Western blot using total cell lysates (not purified recombinant protein) and showing somehow that the antibody is phosphospecific—for instance, is the signal blocked by competition with a phosphorylated peptide but not a non-phosphorylated peptide.
"If you cannot find that information," Browning says, "I would call the manufacturer and ask them for that validation. They ought to have it, or else they shouldn't be selling it. If they can't provide that information, I wouldn't buy the antibody."