Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
In the world of cellular macromolecules, sugars are arguably the most complex. For all their heterogeneity, proteins and nucleic acids are template-driven and predictable. They are reasonably easy to synthesize and sequence. And the impact of chemical changes in these molecules is often straightforward to monitor.
Sugars, whether coupled to proteins (glycoproteins) or lipids (glycolipids), share none of these features. For example, glycan synthesis is not guided by a template; that is, there is no known glycan analog of DNA that dictates how sugars are joined together. There also is no way to amplify a glycan, so researchers can only study what they actually extract and purify from cells, and many interesting sugar chains are present in low abundance.
More complicated yet, there are multiple ways to build a glycan. Sugar chains can be attached to proteins via nitrogen atoms on asparagine residues (N-linked glycosylation) or via an oxygen linkage off any of nine amino acids (O-linked glycosylation). Those chains can be short or long, linear or branched and comprised of a variety of chemically similar sugars. And, like proteins, these chains are subject to post-synthetic chemical modifications.
But understanding all the sugar-adding and -modifying enzymes a cell makes is of limited use, because the enzymes act in concert, and the rules under which they operate remain unclear, says Richard Cummings, director of the National Center for Functional Glycomics at the Beth Israel Deaconess Medical Center and Harvard Medical School. “You could only know what the cell could potentially make, not what it actually is making.”
Bottom line: A cell can create an enormous number of glycan structures over a wide dynamic range of abundances. Fortunately, the set of tools for studying those structures is constantly growing.
Sweet complexity
According to Cummings, glycobiology encompasses everything from determining glycan structures (whether individually or cell-wide, a subdiscipline termed “glycomics”) to identifying interactions between sugar-binding proteins and their ligands. His lab has developed glycan arrays for functional glycomics research and is now gearing up to develop specific anti-sugar monoclonal antibody reagents to drive development of a “human glycome atlas.”
Yet perhaps the most widely used glycobiology tools enable a straightforward workflow involving glycan-cleaving enzymes called endo- and exo-glycosidases. Endoglycosidases prune N-linked glycan arbors from proteins, cutting either between the first sugar of the oligosaccharide and the protein backbone (e.g., PNGase F or PNGase A) or between the first two sugars (e.g., Endo S). In the former case, the enzyme reaction completely separates a glycan from the protein to which it was attached, enabling analysis of the complete glycan structure but complicating determination of the sugar-attachment site; in the latter case, a single sugar remains on the protein to simplify such analyses. (O-linked glycans usually are removed from glycans via chemical hydrolysis using a strong base, such as NaOH.)
Evolving the cleaving process
A typical PNGase F reaction takes one to 24 hours, says Alicia Bielik, group leader for glycobiology and proteomics production at New England Biolabs (NEB). However, earlier this year the company released a Rapid PNGase F formulation that completes the reaction in just 10 minutes. “Your protocol goes from 24 hours to 10 minutes,” she says. “And it’s mass spectrometry-compatible, and releases the glycans rapidly and without bias.”
Exoglycosidases cut glycans one sugar at a time, working inward from the outside of their sugar antennae. Using a collection of such enzymes—some specific for individual sugars or linkages and others more general—researchers can begin to work out the sequence for the most distal sugars and linkages in a complex glycan chain. NEB, for instance, has released seven exoglycosidases in the past year, says Bielik, and two more are in the pipeline. The company will release the latter as an array for glycan sequencing, likely in 2016.
Tagging and separating
By whatever means they are separated from their molecular hosts, liberated glycans are then typically fluorescently tagged for chromatographic separation and/or permethylated for mass spectrometric (MS) analysis.
Common labeling reagents include 2-aminobenzylmide (2AB) and propanamide, both of which attach fluorophores to the glycans, making them amenable to ionization. Researchers at Sigma-Aldrich have recently favored another reagent called procainamide, says Brian Gau, a senior scientist at Sigma-Aldrich who specializes in glycobiology. “The nice thing about procainamide, as opposed to other fluorophores, is procainamide has a tertiary amine,” Gau says. “That is very basic in the gas phase. So when doing mass spectrometry, you get a very good ion signal for these glycans in positive-ion mode.”
Chromatographic separation commonly is accomplished using a hydrophilic interaction chromatography (HILIC) column. “Glycans present a huge range of hydrophilicities, so these things are very well separated if you can take advantage of that sort of interaction,” says Gau. Sigma-Aldrich, for instance, offers a fused-core pentahydroxyl-HILIC column under its Supelco brand that is ideal for this application, Gau says.
Deeper analysis
The final step of the glycan analysis—and likely the most complicated—is mass spectrometry. According to Gau, a simple ion trap mass spectrometer will suffice for many glycan analyses; in fact, it is well qualified for the job. “It allows you to do your typical MS/MS fragmentation experiment on an LC time scale,” he says. “But an ion trap also allows you to do MS3 experiments.”
That means the instrument is capable of iteratively measuring and fragmenting ions, which enables researchers to drill down into deep structural details. For instance, he says, researchers can permethylate a glycan to label every hydroxide group with methyls, thereby enabling cross-ring fragmentation, can reveal which monosaccharide carbons are linked in glycosidic bonds. “That’s a very laborious task, but that is exactly what the ion trap is good at,” he says.
Cummings, though, does his glycan work on high-end Thermo Scientific Orbitrap instruments. That’s because Cummings and his colleagues are interested in mapping glycan structures with all their modifications exactly—something few researchers are willing to do because of the time and materials requirements. “If you have a glycan with 25 to 30 sugars, you have 25 to 30 linkages you have to identify individually,” he explains. “That’s very difficult.” Many researchers forego that task, but the high mass accuracy and resolving power of the Orbitrap, coupled with its multiple available fragmentation options, should simplify the work, he says.
After researchers identify a glycan of interest, they encounter a whole new set of problems, Cummings adds. It’s difficult to synthesize glycans from scratch and couple them to proteins, or even to purify them from culture. Bacterial and eukaryotic expression systems are notoriously bad at properly recapitulating the microheterogeneity of protein glycosylation, Gau notes – a fact that bedevils biopharmaceuticals development. “In addition to the analytic challenge, this requires a deep understanding of the expression system and a lot of empirical trial-and-error.”
For many years, glycobiology has had the reputation for being an exceptionally challenging discipline of biology. Now, says Chris Taron, scientific director of protein expression and modification research at NEB,that perception is slowly changing. With the availability of new reagents, separation technologies and improved and standardized methods, he says, the science of sugar analysis is more accessible than ever.