Putting the Pieces Together Again: Characterizing Trisaccharides by the Energetics of Their Primary Fragmentation Pathways and Their Ion Mobility

Abstract

Identification of polysaccharides is not a straightforward task due to the high degree of stereochemistry present in their isobaric monomers. Their isobaric nature causes traditional mass spectrometry to fall short when trying to differentiate not only the conformation of the monomers but the position of the glycosidic bonds that bind them. This structural information is important for biochemists as they study the role of different glycans in biological processes. Tandem mass spectrometry (MS/MS) allows the study of the fragment ions formed during collision induced dissociation (CID), the fragments formed depend on the structure and stability of the precursor molecule and can be used to identify the compounds. These fragmentation pathways will be as complex as the species that form them. To date, typical saccharide fragments are separated into three groups that represent the major fragments: Cross-ring cleavages (A/X), and those resulting from cleaving different sides of the glycosidic bond (B/Y) and (C/Z). Ion mobility separation (IMS) has shown to have some success at discerning polysaccharide conformers and those of other biopolymers such as proteins and polynucleotides. Ion mobility separates gas-phase ions by colliding them with non-reactive gases and relating respective increase in flight time to their collision cross-section (CCS). In this study, the relative energetics of the first steps of the cross-ring cleavage and both glycosidic bond cleavage channels for isomaltotriose [glc(α1-6)glc(α1-6)glc] as well as a minor water loss channel were explored using density functional theory (DFT) calculations at the B3LYP/6-31+g(d) level of theory. It was demonstrated that charge-remote mechanisms are a viable alternative to charge-directed mechanisms when under the high energy short time scale conditions present during an ESI-MS/MS experiment. To verify the efficiency of ion mobility for isomeric separation, the relative experimental CCS of isomaltotriose [glc(α1-6)glc(α1-6)glc], maltotriose [glc(α1-4)glc(α1-4)glc], panose [glc(α1-6)glc(α1-4)glc] and raffinose [gal(α1-6)glc(α1-2)fru] were determined by comparison with literature CCS values for dextran, a variable-length oligomer of α1-6 linked glucose was used as an external calibrant. The experimental CCS of the precursor ions were compared to literature values when available as well as the calculated effective values of the optimized DFT geometries using the trajectory method of the MOBCAL computational suite. As phosphate is often used as an adducting agent to increase the intensity of the precursor ion when running an IMS experiment, the effect of its presence on the fragmentation of isomaltotriose and large isomaltooligosaccharides was studied. It was seen that depending on the location of the phosphate ion, it will preferentially dissociate leaving behind a neutral glycan. This explains the low abundance of fragment ions observed when selecting a phosphate-adducted precursor ion during an MS/MS experiment. IMS and MS-MS are complementary methods that can be used to identify monomers within a polysaccharide and how they are bound.