|
Mass Spectrometry for Glycomics
If proteomics is the "next big thing" after genomics, then glycomics – the study of carbohydrates attached to proteins and lipids – may be the next big thing after proteomics.
Cell-surface glycans play key roles in crucial cell-cell interactions, such as sperm-egg binding and immune-system recognition of pathogens. For example, variant glycosylation of blood plasma proteins shows great promise as a source of biomarkers for early diagnosis of cancer.
Working with top glycomics laboratories, PARC researchers have
developed new algorithms and software for profiling and identifying
glycans and glycopeptides by mass spectrometry.
PARC has developed these software programs – with support from the National Institutes of Health (NIH)* – to:
- annotate peaks in single-mass spectrometry profiles (Cartoonist);
- make identifications from tandem-mass spectrometry spectra (Cartoonist2); and
- identify glycopeptides using a combination of single and tandem mass spectrometry (Peptoonist).
At this early stage glycomics relies on a variety of mass spectrometry techniques, which is something of a handicraft when compared to the factory approach now dominant in proteomics.
Example
An example of how PARC's tools have been used was included in a recently published study aimed at elucidating sperm-egg binding at the molecular level. One of the key proteins in this interaction is ZP3 (ZP for zona pellucida, the membrane surrounding an unfertilized egg).
A
manual analysis of mass spectra found one glycosylation site on ZP3 populated with about 50 possible glycoforms – a surprisingly large amount of heterogeneity. However, automatic
analysis of the same spectra using Peptoonist found that the
number of glycoforms had been underestimated by a factor of two.
The spectrum below (shown at two different zooms) shows a portion containing glycoforms that were missed by the manual analysis:
Peptoonist, like PARC's other glycomics software, recalibrates time-of-flight m/z measurements to obtain high mass
accuracy and then uses the predicted abundance of isotopes to avoid false-positives when identifying peaks corresponding to glycopeptides.
Both Cartoonist and Peptoonist then apply detailed, expert-system knowledge of mammalian N-linked glycans to assign likely "cartoons'' (molecular structures) to the identified glycans. Cartoonist2 is designed for O-linked glycans, which are smaller but more variable than N-linked glycans – so it uses less a priori knowledge.
|
