How shells actually form remains unclear. Yet the key, says Vernon Reinhold, director of the UNH Glycomics Center at the University of New Hampshire, could be sugars. In mollusks, there is a thin pocket of liquid, called extrapillial fluid, between the inner shell surface and the exterior of the animal, which drives the process. Chief amongst its components is a single calcium-binding protein called EPG, a homodimer of 213 amino acids decorated with at least three complex polysaccharides.
In that regard, EPG is not unusual. According to Richard Cummings, director of the Harvard Medical School Center for Glycoscience in Boston, many cellular proteins, and most cell-surface and secreted proteins, are modified by glycosylation—a process that alters protein function across a broad range of biological processes. “Transplants, transfusion, cardiovascular biology, infectious diseases, bioengineering, angiogenesis, vascular biology, inflammation—glycans contribute to all of these biological pathways and disorders,” Cummings says.
To understand those processes, “we need to understand the relationship between [glycan] structure and function,” Cummings says. “And yet we know so little about that it’s embarrassing.”
In part, that’s because glycans are particularly tough nuts to crack. Unlike proteins and nucleic acids, glycans are not built via template—their form is determined by the suite of enzymes a given cell expresses. The sugar chains can be linear or branched, with multiple possible linkages, and each sugar monomer can be chemically modified. And any protein may support multiple possible glycan structures. “So the complexity of what we term the ‘glycome’—that is, the collection of all the glycans in the human body or an animal’s body—is really immense, and not known,” Cummings says.
Unlike proteins and nucleic acids, glycans are not built via template—their form is determined by the suite of enzymes a given cell expresses.
Tools for simplifying the complexity of the glycome are slowly falling into place, however.
One key tool for glycomics research is mass spectrometry. As in proteomics, researchers can harness mass spec to break down complex glycan structures into their component parts. But the process is considerably more complex for sugars than for proteins, says Reinhold. For one thing, many sugar monomers are structural isomers, meaning they have the same mass (and thus the same behavior in a mass spectrometer). And because glycans can be linked in multiple ways, it is not enough to know the order of sugars; researchers must get creative.
As in proteomics, researchers can harness mass spec to break down complex glycan structures into their component parts.
For Reinhold, the answer is ion trap mass spectrometry, an instrument configuration that enables researchers to perform so-called MS(n) experiments, in which ions are repeatedly weighed, fragmented and reanalyzed to work out the minutiae of their structure. “Knowing the pathway by which [a glycan] fell apart, we can follow that backwards and put it back together again,” he explains.
Reinhold and his team used that strategy to meticulously work out, over a period of four years, the precise chemical structure of the EPG glycans —information that could potentially provide insights into other biomineralization processes, including bone deposition, he says.
Julian Saba, a senior glycomics specialist at Thermo Fisher Scientific, recommends glycomics researchers opt for a high-resolution mass spectrometer capable of multiple types of fragmentation—such as the company’s Tribrid Orbitrap-based Fusion and Lumos instruments. “In terms of performance, it is quite a leap,” Saba says. The instruments have “a tribrid technology, which allows you to manipulate the ions introduced into the mass spectrometer to really slice and dice your glycans as thoroughly as possible. The multiple fragmentation modes available on the instruments, including collision-induced dissociation (CID), higher-energy collisional dissociation (HCD), electron transfer dissociation (ETD) and a novel combined fragmentation approach called EThcD, enable a researcher to characterize the glycans while they are still attached to the peptide/protein.” Saba recommends ETD or EThcD for peptide sequencing to identify the glycosylation sites of glycopeptides, and HCD or CID for the glycan composition.
For researchers looking to get more in-depth glycan information, Saba recommends releasing glycans from peptides/proteins and examining them separately. To compare glycan profiles over time or conditions, Saba recommends using his company’s multiplexing aminoxyTMT reagents, which are carbonyl-reactive and were specifically designed to label glycans. Researchers can combine up to six samples in a single mass spec run using these reagents.
In any event, most researchers don’t analyze glycans in their native form, Saba says—they derivatize them first to improve ionization, fragmentations and separation. For instance, permethylation of glycans is a powerful strategy. According to Saba, that treatment improves HPLC separation on C18 reverse-phase columns, as well as ionization and fragmentation.
Glycomics reagents also are available. Glycan microarrays are available commercially from RayBiotech, as well as from the Consortium for Functional Glycomics, which sells them at cost. Vector Laboratories offers a selection of lectins (carbohydrate-binding molecules), and a French company called Elicityl sells purified glycan structures.
The enzyme PNGase F (available from New England Biolabs and other vendors) specifically frees most N-linked glycans from their protein scaffolds. But there’s no equivalent reagent for O-linked glycans, which must be released chemically.
Cummings recently reported a method based on household bleach (sodium hypochlorite), which can liberate both N- and O-linked glycans from animal or plant tissue samples in minutes . He uses that strategy to drive a method he calls “shotgun glycomics,” in which glycans are freed from tissue, purified, separated on columns, printed onto microarrays or beads and then interrogated. Among other applications, he says, that strategy can be used to suss out each individual’s unique anti-carbohydrate antibody signature—information that may help explain people’s differential disease susceptibility.
A second paper, published in 2015, provides a method called CORA for amplifying O-glycans in live cells—like a PCR method for sugars. The procedure uses a modified sugar to “prime glycan biosynthesis,” according to the published results , improving the detection of O-linked glycans up to 1,000-fold.
Shotgun glycomics, Cummings says, “provides a way of getting at the human glycome without having to use chemical synthesis.” Chemical synthesis of glycans is difficult, he explains. But there’s something of a chicken-and-egg element to the problem, as well: “Unless you know what the human glycome is, you don’t know what to synthesize.”
 Zhou, H, et al., “Anomalous N‐glycan structures with an internal fucose branched to GlcA and GlcN residues isolated from a mollusk shell-forming fluid,” J Proteome Res, 12:4547-55, 2013. [PMID: 23919883]
 Song, X, et al., “Oxidative release of natural glycans for functional glycomics,” Nat Methods, doi:10.1038/nmeth.3861, published online May 2, 2016. [PMID: 27135973]
 Kudelka, MR, et al., “Cellular O-glycome reporter/amplification to explore O-glycans of living cells,” Nat Methods, doi:10.1038/nmeth.3675, published online November 30, 2015. [PMID: 26619014]