University of Gothenburg researchers have completed a study of Aβ glycopeptides in human cerebrospinal fluid that identified tyrosine10 glycosylated Aβ peptides as potential biomarkers for Alzheimer's disease.
Published online last month in the Proceedings of the National Academy of Sciences, the study suggests that glycosylation may play a key role in the proteolytic processing of the amyloid precursor protein, implicating it in the generation of Aβ peptide fragments like Aβ1-42 that have been linked to Alzheimer's.
According to Goran Larson, a University of Gothenburg professor and an author on the paper, while there has been previous research looking at glycosylation of APP, little investigation has been done looking at glycosylation of the Aβ peptides derived from that protein.
"Usually protein people are most interested in looking at the proteins and the glycan people are interested in looking at the glycans," he told ProteoMonitor. "So the glycobiologist purifies the proteins and looks at them, and the protein people characterize the peptides but get rid of the glycans so they don't interfere with the mass spectra and so on."
The technical difficulties involved in glycopeptide analysis have also limited their study, Larson said.
"You have to be aware of glycan structure and fragmentation [patterns]. If you're not used to looking at glycosylation you won't recognize the sequential loss of glycan," he said. "Secondly you have to manipulate the Mascot [protein-identification software] in such a way that you can specifically look for the loss of glycans or additional sugars on peptide sequences."
Glycoprotein researchers also run into the issue of spotty annotation data in databases like UniProt, Larson said, particularly for O-glycosylations.
"There are good techniques for recognizing the attachment sites of N-glycans, and these are often annotated in the UniProt database," he said. "But with O-glycosylation, because there is no consensus, unless you really get the glycopeptides out you won't be able to find out what glycan is attached to what amino acid. Surprisingly few studies have addressed this question."
For their Aβ research the Gothenburg scientists pulled Aβ peptides out of cerebrospinal fluid using the 6E10 antibody and then isolated the glycopeptides with an enrichment strategy they first described in a paper published in Nature Methods in October 2009.
The technique uses selective periodate oxidation of sialic acids followed by their capture on hydrazide beads, enabling the enrichment of glycoproteins. The captured glycoproteins are then trypsinized such that only the glycopeptides remain attached to the beads. Following that, these remaining glycopeptides are released by acid hydrolysis and analyzed via mass spectrometry – in the case of the PNAS study on a Thermo Scientific LTQ FT machine.
With this method the researchers identified 33 unglycosylated Aβ peptides as well as 37 Aβ glycopeptides with sialyated core-1-like O-glycans attached to threonine residues. The researchers also found a series of 27 glycopeptides glycosylated on Tyr10, which Larson said marks the first ever identification of a sialyated glycan on tyrosine in a mammalian protein.
Using a sample set comprising six Alzheimer's patients and seven subjects suffering from non-Alzheimer's dementia, the researchers also linked an increase of Tyr10 glycosylated Aβ peptides in CSF to Alzheimer's disease, observing a nearly 2.5-fold increase of Tyr10 glycosylated peptides in the Alzheimer's sufferers.
As Larson noted, the small sample size limits the conclusions that can be drawn regarding the usefulness of these peptides as Alzheimer's biomarkers, but, he added, the large number of Tyr10 glycosylations present "speaks in favor of an accumulation of the glycopeptides" playing a role in the development of the disease.
The study also provides an example of the potential usefulness of the more established Alzheimer's protein biomarker panel developed by scientists including University of Pennsylvania researchers Les Shaw and John Trojanowski of Aβ1-42, phosphorylated tau protein, and total tau in research into the disease (PM 06/11/2010). The Gothenburg researchers used this panel – which has been shown in past studies to identify Alzheimer's disease with 90 percent specificity and sensitivity – to distinguish Alzheimer's from non-Alzheimer's subjects in their investigation of Tyr10 glycosylation as a biomarker.
He suggested that adding the Try10 glycopeptides to this existing three-protein panel could further improve its effectiveness.
Alzheimer's biomarkers are a major area of protein biomarker research and are expected to represent a cumulative $9 billion market over the next ten years according to a recent report commissioned by proteomics firm Proteome Sciences (PM 06/03/2011). In addition to their promise for diagnosis and predicting progression of the disease, early detection and progression biomarkers are expected to play an increasing role in pharmaceutical research into drugs for the disease as well.
The next step in exploring Try10 glycopeptides as biomarkers, Larson said, will be to put together a larger study with better characterized subjects. The researchers are currently working to line up funding for such a project, he said.
He added that they also plan to develop an antibody to the Tyr10 glycopeptides in hopes of making it easier for other researchers to follow up on their findings. It would also allow for the sort of higher-throughput analyses necessary to establish the biomarkers' clinical utility, he added.
"The data that we have so far are encouraging," Larson said, "but we would also like to make the analysis simpler. It's good to use mass spectrometry for research purposes, but if we could develop a simpler [antibody-based] technique, it could be [investigated] by other groups more easily."
"There are many more studies to be done. This is just the first report that [Tyr10 glycopeptides] are there and that there is a connection to Alzheimer's disease," he said. "But we don't have the full picture yet."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.