Date of Award

Fall 2009

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

Vernon Reinhold


The EP glycoprotein, the major protein component isolated from the EP fluid of the mollusc, Mytilus edulis, is known to bind many metal ions and to contain 14.3% of carbohydrate in weight. To structurally evaluate the sugar function of this glycoprotein, a new mass spectrometry-based strategy was applied with a focus on the comprehensive characterization of its N-glycans. This included: (a) accurate mass measurement and GC/MS to determine the monosaccharide compositions; (b) positive ion CID-MS n to disassemble the deuterated permethylated derivates (CD3 ) to reveal the antenna sequence, and natural methoxyl group location; (c) negative ion CID-MSn on native glycans to define general antennary topology. Three major complex-type N-glycans were characterized with two of them expressing an unique tetrasaccharides antenna sequence: 4-O-methyl-GlcA(1-4)[GlcNAc(1-3)]Fuc(1-4)GlcNAc. The third one is identified as a pentasaccharide antenna: mono-O-methyl-Fuc(1-2)-4- O-methyl-GlcA(1-4)[GlcNAc(1-3)]Fuc(1-4)GlcNAc. This is the first report of these novel N-glycans and their intriguing structures may implicate their involvement in metal ion binding.

The neuraminidase (NA) of the 1918 influenza virus (H1N1) has active (tetramer) and inactive forms (monomer/dimer) with difference residing in their N-glycosylations. A structural analysis was carried out on the released N-glycans from both tetramer and dimer forms. Significant amounts of fucosylated N-linked structures were detected in dimer compared to tetramer (relative ratio: 73.6% vs. 18.1%), while no other structural difference were observed. Thus, it appeared that fucosylated glycans can play a key role to inhibit the formation of the tetramer NA, especially those located in the stalk region. However, enzymatically incorporation of fucose onto the tetramer in vitro didn't alter its enzyme activity, indicating that this regulation must occur at an early biosynthetic stage.

Common lectin methods failed to detect any Sialalpha(2,3)Gal residues, the proposed avian influenza virus receptors, in the human upper respiratory tract, contrasting current reports that avian viruses can grow efficiently in the human upper respiratory tract tissues. To solve this disparity, two mass spectrometry-based strategies were employed to identify the Sial-Gal linkages on two human cell lines: Detroit 562 and A549, which represent the human upper and lower respiratory tract, respectively. One strategy involved MSn fragment ion spectral comparison with standard oligosaccharides. A second approach was to differentiate sial-Gal with a specific cyclization reagent DMTMM. Using this reagent, Sialalpha(2,3)Gal epitope formed a lactone while Sialalpha(2,6)Gal analog provided a methyl ester, easily distinguished by a 32 Da mass difference. Significant amounts of Sialalpha(2,3)Gal were found in Detroit 562 cell lines, confirming the presence of avian virus receptors in the human upper respiratory tract. Thus, the poor human-to-human transmission of avian influenza virus must be attributed to other unknown reasons, rather than the paucity of SA(alpha2,3)Gal. This is the first approach to identify influenza virus human receptors by mass spectrometry.