Glycosylation, the most common posttranslational modification, is highly involved in many biological processes. The concentration, glycosylation site occupancy, and the glycoform profile is known to vary between different biological states. Lin et al.1 hypothesized that, “[c]omparative study of glycoproteins under different biological states may yield important insights into their cellular position and functionality.”
Previous comparative studies typically involved an enzymatic release of glycans from glycoproteins for evaluation. This has been done previously using employ hydrazide chemistry2 and permethylation and stable-isotope glycan labeling.3
The Lin group developed a method of a differentiating between variations in protein concentration, glycosylation site occupancy, and glycoform profile using intact proteins. To illustrate this method, intact proteins from two glycosylation states were hydrolyzed by trypsin. These samples were then split into two groups, one derivatized via reductive methylation with formaldehyde-H2 and the other derivatized via a reaction with formaldehyde-D2. The D2 and H2 derivatized proteins were mixed together, and peptides were separated and purified from the glycopeptides using microcrystalline cellulose. Glycopeptides were purified by eluting from the cellulose-binding fraction, while peptides and the excess derivatization reagents were present in the flow-through fraction This prevented signal suppression from nonglycoslytated peptides during the final stage, ESI-MS.
MS data yielded two spectra — one from peptides and the other from glycopeptides. Variation between the two samples could result from changes occurring in glycoprotein concentration, the occupancy site, the glycoform profile, or a combination of these three factors. These differences were then quantified based on mathematical calculations. Variations due to a change in glycoprotein concentration were obtained by the ration of the relative intensity of the nonglycosylated peptides (PD/PH). The variation due to the change of glycoform profile and change of glycosylation site occupancy were quantified and differentiated based on equations developed by Lin et al.
These calculations were validated by preparing RNase B samples in dilutions with known glycoform profiles. Intact RNase B was treated with α-mannosidase to cleave its single glycosylation site, thereby altering the glycoform profile.4 The modified RNase B samples, as well as the native sample, were prepared under the same conditions as the intact glycopeptides. The change of site occupancy was found in the range of 0.94-1.03, with a mean value of 0.99. A value of 0.99 was close to the expected ratio of 1.
RNase B samples were also prepared with variations in protein concentration, glycosylation site occupancy, and glycoform profile, including a 1.43-fold higher protein concentration, a 50% lower occupancy for glycosylation sites, and a modified glycoform profile. The modified sample was then labeled with formaldehyde-D2 and mixed with the formaldehyde-H2-labeled control sample before ESI-MS analysis. The calculated result based on the D/H ratio and the calibration curves were in close agreement with the expected values, and the change of glycoprotein concentration calculated was 1.45. Likewise, the variation due to glycosylation site occupancy was calculated (0.45-0.51) with a mean value of 0.48.
Glycobiology is an emerging field of study, and the ability to accurately quantify glycoproteins is vital for comparative studies. Successful methods of quantification will likely be a great asset in to understanding disease interactions involving glycoproteins.
1. Lin, C.Y., et al., (2012) ‘A comparative study of glycoprotein concentration, glycoform profile and glycosylation site occupancy using isotope labeling and electrospray linear ion trap mass spectrometry‘, Analytica Chimica Acta, May 30 (728), (pp. 49-56)
2. H. Zhang, X.J., et al. ‘Identification and quantification of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry‘, Nature Biotechnology, 21 (6), (pp. 660-666)
3. P. Kang, Y., et al. (2007), ‘Comparative glycomic mapping through quantitative permethylation and stable-isotope labeling‘, Analytical Chemistry, 79 (16), (pp. 6064-6073)
4. Toumi, M.L., Go, E.P., and Desaire, H., (2009) ‘Development of fully functional proteins with novel glycosylation via enzymatic glycan trimming‘, Journal of Pharmaceutical Sciences’, 8 (8), (pp. 2581-2591)