This story originally ran on July 29.
Researchers at the California Institute of Technology have developed a polyethylene glycol-based mass-tagging strategy for the analysis of glycosylated proteins that does not require protein purification, advanced instrumentation, or expensive radiolabels, according to the Caltech team.
The technique, which is described in a paper in this month's edition of Nature Chemical Biology, simplifies the quantification of glycosylation stoichiometries and allows for the study of the interplay between protein glycosylation and other post-translational modifications, Linda Hsieh-Wilson, associate professor of chemistry at Caltech and one of the paper's authors, told ProteoMonitor.
The method applies specifically to O-GlcNAc glycosylation, which is the addition of N-acetyl-D-glucosamine to serine or threonine residues on a protein. O-GlcNAc glycosylation is thought to play a role in processes like glucose homeostasis, cardiac muscle survival, cell cycle progression, and synaptic transmission.
Study of this post-translational modification has been hampered by the difficulty of determining O-GlcNAc stoichiometries – whether proteins containing the modification exist in singly, doubly, or multiply glycosylated states – knowledge that is key to understanding O-GlcNAc function and regulation.
Conventional methods of stoichiometry quantification typically require time-consuming radiolabeling of large amounts of purified protein. Hsieh-Wilson's team's approach is able to tag glycosylated protein in cell lysate and quantify the stoichiometry simply by running the sample on an SDS-PAGE gel and then Western blotting.
The technique uses a chemoenzymatic reaction to attach a sugar containing a ketone handle to the O-GlcNAc groups of glycosylated proteins. PEG moieties of defined molecular mass are then attached to the ketone handle, which allows for distinguishing between glycosylation states based on the shift in molecular mass by running the sample on a gel. The particular protein of interest is then isolated via Western blotting.
"It allows us to read out the glycosylation state of proteins from cell lysate to look at in vivo glycosylation levels and stoichiometries because we can quantify the amount of protein that's there," Hsieh-Wilson said. Using the example of CREB, she explained, "We can see CREB in the lysate. We can see the unglycosylated form of CREB, the monoglycosylated form, and we can see diglyscosylated CREB. By quantifying the amounts of each of the [SDS-PAGE] bands we can determine that CREB [exists in], say, 30 percent monoglycosylated form and 10 percent diglycosylated form in vivo."
As a proof of concept, the researchers used the PEG-labeling approach to tag and quantify glycosylation sites on ten different proteins. They also used it to compare protein glycosylation states across various tissues and monitor changes in glycosylation in response to the blocking or activation of various pathways.
Beyond simplifying the quantification of glycosylation stoichiometries, the technique also lets researchers investigate the relationship between glycosylation and other post-translational modifications, something, Hsieh-Wilson said, scientists were previously unable to do.
"You can use it to study the interplay between glycosylation and other post-translational modifications, in our case phosphorylation," she said. "One of the things about O-GlcNAc is that it's an intracellular form of glycosylation, and it's dynamic and inducible. So people are very interested in whether and how O-GlcNAc relates back to phosphorylation."
By using phosphospecific antibodies during Western blotting, Hsieh-Wilson and colleagues were able to isolate the phosphorylated population of their protein of interest, allowing them to examine the relationship between phosphorylation and glycosylation levels. In experiments involving the protein MeCP2, the researchers made the unexpected discovery that glycosylation is more common on phosphorylated versions of the protein, rather than unphosphorylated versions as previously assumed based on conventional labeling methods.
"We discovered this interplay that you wouldn't have seen using existing methods because people could only look at the total population [of glycosylated proteins], not the subpopulation. The ability to really hone in on these specific subpopulations and really [determine whether it's] the phosphorylated population that's being glycosylated or the unphosphorylated population that's being glycosylated — that is very important for understanding post-translational modifications and their interplay," Hsieh-Wilson said.
"The point is we really need better tools to be able to dissect on a molecular level what's happening on a given protein, not just on a population level where you're averaging across the population," she added.
Currently, the researchers are looking to develop the technique for use with other types of modifications, in particular with another form of glycosylation. What's needed, Hsieh-Wilson said, are enzymes capable of attaching the PEG-accepting molecule – a sugar with a ketone handle in the case of O-GlcNAc – to the modification of interest.
"In principle it can be applied to other post-translational modifications if the enzymes are available," she said. "You can engineer them to accept a ketone group or an azide group or click chemistry. It's a versatile approach. You just need the enzyme that will selectively tag that modification."
Researchers have used the enzyme in the Nature Chemical Biology study – an engineered β-1,4-galactosyltransferase – to label O-GlcNAc glycosylations for years, Hsieh-Wilson said, noting that her team first demonstrated its usefulness for this purpose in 2004.
"We showed that we could engineer the enzyme and get it to attach to the sugar and attach things like biotin to it," she said. "And we went on to show that you can use it for proteomics studies, that you could pull down the biotin-labeled proteins and capture glycopeptides, sequence them, identify glycosylation sites. So we've been doing a lot with this chemoenzymatic approach."
The new PEG-labeling method has the potential to provide complementary information to the proteomics work enabled by the previously developed techniques, Hsieh-Wilson suggested.
"You get information that's complementary to mass spec," she said. "With mass spec you can identify multiple sites of glycosylation within a protein, but you never know the occupancy of those sites. So you can have a protein that is known to have five or more glycosylation sites, and yet it exists in the monoglycosylated form in vivo."
The original GalT-tagging technology was licensed by Invitrogen, which has products based on the method, including its Click-iT O-GlcNAc Enzymatic Labeling System. Additional products could be developed based on the new PEG-labeling technique, Hsieh-Wilson said, although she had no knowledge of any present plans by Invitrogen to do so.
Invitrogen representatives couldn't be reached for comment on possible plans to commercialize the technology.