Researchers at the Max Planck Institute of Biochemistry in Martinsried, Germany, have demonstrated a novel technique for identifying N-glycosylated protein sites using mass spectrometry.
The method, which the scientists detailed in the May 28 issue of Cell, enabled them to identify via high-resolution mass spec 6,367 N-glycosylation sites on 2,352 proteins, and will improve mass spec analysis of a wide range of post-translational modifications, Matthias Mann, director of the Department of Proteomics and Signal Transduction at the Max Planck Institute and the study's lead author, told ProteoMonitor.
Involved in a number of important cellular functions and the pathogenesis of many diseases, N-glycoproteins are the subject of significant scientific and clinical interest. However, the low expression levels of many N-glycoproteins and the complexity of the attached sugar molecules have made identification and characterization of these proteins via mass spectrometry difficult.
Using a filter-aided sample-preparation technique that they developed last year (PM 04/23/2009), the Max Planck researchers enriched glycopeptides from mouse tissue and plasma samples by adding lectins as affinity reagents after on-filter protein digestion.
Then, using LC-MS/MS on a Thermo Fisher LTQ Orbitrap Velos instrument, they employed what Mann called a novel "high-high" mass spec strategy, analyzing the peptides in high-resolution precursor mode and the fragments in high-accuracy fragment mode.
Fragmentation was done via higher-energy dissociation, or HCD, which enables readout of the peptide fragments in the high-resolution orbitrap portion of the instrument, as opposed to the ion trap portion. Typically, this higher resolution comes at the price of reduced sensitivity, but, the study noted, the Velos's improved HCD performance allowed for use of high-accuracy mode without significant sensitivity loss.
"Basically, the mass accuracy of each fragment improves an average of 50 fold," Mann said. "So that's a huge jump in certainty. We really know that these are the peptides and that that is where the [glycosylation] site was."
With this method researchers were able identify 6,367 N-glycosylation sites, adding 5,753 new sites to the N-glycosite library, while also identifying roughly 70 percent of already known mouse N-glycosites.
"We increased the known glycoproteome six-fold," Mann said, adding that the team was able to identify 70 percent of all previously discovered glycosylation sites.
"It's quite remarkable that in a single study you could rediscover 70 percent of what people have put in the database over basically the last thirty years," he said.
The study also helped shed light on where within a cell N-glycosylated proteins exist. Generally it has been thought that N-glycoproteins existed in the endoplasmic reticulum and Golgi apparatus. Some research, however, has suggested the existence of N-glycoproteins in the mitochondria, as well. Mann's team found this not to be the case.
"We didn't find any instances of N-glycosylated proteins that were mitochondrial," he said. "So we can more or less put that notion to rest."
The study also has significant clinical implications, Mann said.
"It's well known that during cancer development the glycosylation pattern of proteins changes. So that's of great clinical interest," he said.
Additionally, the FASP-based enrichment technique employed in the study could prove useful in biomarker research.
"The known cancer markers are all glycoproteins by and large, so now you can use this method to very efficiently enrich glycoproteins to serve as biomarkers," he said.
Although the study focused only on N-glycosylations, the method described is broadly applicable to any class of post-translational modification for which there is a good antibody or affinity agent, Mann said.
"Where you have a good antibody against a peptide or an affinity agent like the lectins you can use the same principle where we retain the modified peptides by the antibody on top of the filter while all the other peptides go through," he said. "And then you just elute [the modified peptide] from the antibody, and you can elute them with very high efficiency. It's very cheap and it's very easy."
In addition to N-glycoproteins, Mann's lab has had success using the technique to identify phosphotyrosine peptides, he said. The lab has also tried applying it to O-glycosylations but has thus far been unable to find an appropriate affinity agent.
"We don't have a good lectin that will bind specifically, so that hasn't been successful so far," Mann said.
Analysis of post-translational modifications has traditionally been a challenge for mass spectrometry, but, Mann said, strides have been taken – his lab's new strategy for N-glycoproteins among them.
"Big improvements have been made," he said. "For example, phosphoproteomics is a very active area – people can now get 10,000 sites, so that's advanced very far. There's lysine acetylation, which is very important for drug development. And now we're adding N-glycosylation. So, more and more [post-translational modifications] are coming into the realm of being easily analyzable by mass spectrometry."