A team led by researchers at Pacific Northwest National Laboratory has developed a mass spec-based workflow for global analysis of O-linked N-acetylglucosamine, or O-GlcNAc, protein modification.
The researchers used the workflow, detailed in a paper published this week in the Proceedings of the National Academy of Sciences, to analyze O-GlcNAc modifications in mouse brain tissue, identifying 168 previously unknown O-GlcNAcylated proteins and bolstering hypotheses proposing cross-talk between O-GlcNAcylation and phosphorylation, said Richard Smith, director of proteomics research at PNNL and one of the leaders of the study.
O-GlcNAc modification is a reversible protein post-translational modification found on serine and threonine residues. It has been observed most widely in brain and pancreatic tissue and has been linked in previous studies to Alzheimer's disease.
Studying protein O-GlcNAcylation has proven challenging, though, due to its low abundance and high lability. By taking advantage of recent improvements in enrichment techniques and advances in mass spec fragmentation, Smith and his colleagues were able to significantly increase coverage of the modification.
For the enrichment the researchers used a chemical/enzymatic photochemical cleavage, or CEPC, method in which O-GlcNAcylated peptides are enzymatically labeled with azidogalactosamine, which is then conjugated to a photocleavable biotin probe. These biotinylated peptides can then be enriched via avidin affinity chromatography and then released by photochemical cleavage.
The technique was originally developed by Johns Hopkins University researcher Gerald Hart and University of Virginia researcher Donald Hunt – both also authors on the PNAS paper – and was optimized in the recent study to enhance the sensitivity and selectivity of the enrichment. This optimization enabled the researchers to use significantly less tissue than in previous experiments, which, the authors noted, could open the door to applications like "high-resolution spatial mapping of O-GlcNAcylation patterns" using samples collected via laser-capture microdissection.
Also key to the work, Smith told ProteoMonitor, was use of electron transfer dissociation mass spec and higher-energy collisional dissociation in combination with collisional dissociation mass spec, as opposed to CID alone.
"Since O-GlcNAc modification is labile, it falls off during conventional [CID] quite readily, so the site of the modification is conventionally difficult to identify," he said. With ETD and HCD, on the other hand, "that is less of a problem, so you get more information and are better able to pin down the site of the modification."
Running samples of mouse cerebrocortical tissue on a Thermo Scientific LTQ Orbitrap Velos, the researchers identified a total of 274 O-GlcNAcylated proteins, including the aforementioned 168 proteins not previously known to be modified by O-GlcNAc. They identified a total of 458 O-GlcNAc sites across 195 proteins.
Of the 274 O-GlcNAcylated proteins identified, 268 are also known to be phosphorylated, and roughly 24 percent of the identified O-GlcNAc sites have either reciprocal or proximal phosphorylation sites, facts that suggest possible cross-talk between O-GlcNAcylation and phosphorylation, Smith said.
The notion of such cross-talk "is not a proven hypothesis … all the evidence for it is still circumstantial and indirect at this point" he said, but findings like those presented in the PNAS paper help build the case for the idea.
"At some level it almost has to be true, given that, for example, in this case, almost a quarter of the O-GlcNAc sites we identified are also sites that can be phosphorylated," Smith said. "So there is significant overlap. This is another contribution to unraveling what is going on here."
To further unravel possible links between the two modifications, Smith and his team are now working to concurrently measure levels of phosphorylation and O-GlcNAcylation.
"We have a couple of studies in progress where we are making a large number of measurements in the same samples following phosphorylation and O-GlcNAc modification, and I would guess that there are a number of other labs out there that are beginning to work on this," he said. "I think over the next year a lot of this story ought to begin to be revealed."
The PNAS study's results also offer support to past research linking O-GlcNAc modification to Alzheimer's, localizing O-GlcNAc sites on a number of proteins – including synapsin I and II, synaptopodin, α-synuclein, and several microtubule-associated proteins – that have been implicated in the disease. It also identified fewer O-GlcNAcylated proteins in tissue from an Alzheimer's mouse model than a wild-type mouse, a finding consistent with previous observations of decreased levels of brain protein O-GlcNAcylation in Alzheimer's disease.
The researchers performed the study in brain tissue in part because of this previously noted association with Alzheimer's, Smith said. They are now moving forward with work in additional tissues, including one recently launched study examining O-GlcNAcylation in breast cancer and another that they are planning in ovarian cancer.
"I think that, like phosphorylation, O-GlcNAc modifications will turn out to be a vital switching mechanism that is quite broadly important in eukaryotes," he said.
"The proteomics community has developed quite effective methodologies for looking at the phosphoproteome," Smith added. "O-GlcNAc tools aren't at that level of sophistication yet. But I think that they will be increasingly broadly applied as more of the [modification's] story is revealed."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.