Move over, genomics and proteomics. The up-and-coming “omics” field that will soon be playing an increasingly important role in bioscience—and not just in basic research, but in the development of therapeutics as well—is glycomics.
Glycomics refers to the study of the entire carbohydrate complement of living organisms, whether uncombined (free) or combined in more complex molecules. Chains of sugars, known as glycans, are key building blocks of all living cells, and the majority of human proteins are modified by the attachment of glycans through glycosylation. Because they are involved in many molecular processes, such as protein folding, cell adhesion, and signal transduction, glycans play critical roles in cellular function both in health and disease. They are implicated in pathogen infection (such as influenza viruses and rotavirus), autoimmune and inflammatory diseases, and tumor proliferation. Glycans have also been recognized as important targets for therapeutics; for example, fine-tuning glycosylation in monoclonal antibodies has been shown to improve the effectiveness of cancer drugs.
It’s about 20 years behind its sister fields, but thanks to new structural and analytic tools, glycomics is finally beginning to reach its potential for both therapeutics and diagnostics, experts say. “Analyzing the structure of glycans has historically been very difficult, because there is no standard,” says Richard Cummings, director of the National Center for Functional Glycomics and professor in the department of surgery at Beth Israel Deaconess Medical Center and Harvard Medical School. “Almost no de novo sequencing of proteins is done anymore—it’s all done using databases. But with glycans, it’s all new. Your glycan may have a similar molecular weight to something that’s already in the database, but that doesn’t help you.”
Natural glycans are difficult to isolate, making it challenging to characterize them or take advantage of them as drug targets.
To address this gap, major efforts to develop bioinformatics glycan databases are finally bearing fruit. One of the largest is GlyTouCan (www.glytoucan.org), an international, freely available, uncurated registry for glycan structures. Any glycan structure, ranging in resolution from monosaccharide composition to fully defined structures can be registered as long as there are no inconsistencies in the structure.
“This is really what the field has needed,” says Cummings, who is involved in GlyTouCan. “All glycans deposited there are given unique accession numbers. Eventually, we will be able to match these unique glycans to the enzymes that made the glycan, and the protein or lipid to which the glycan is linked.”
A common challenge of glycomics is the mystery carbohydrate. “You have a glycan and you don’t know what it is. You know that it binds to this but not to that, and that it reacts to reagent A but not reagent B,” says Cummings. “As our databases become more robust about the specificity of reagents, we can come to clearer conclusions about what the structure must have in it.”
To that end, New England Biolabs (NEB) is developing a kit of highly characterized recombinant exoglycosidases that cleave very specific monosaccharide modifications, which can assist in identifying a glycan. “We have recently released new endoglycosidases, and an improved protein deglycosylation mix, tailored for high-throughput analyses particularly in the biopharma setting,” says Paula Magnelli, , a glycobiology research scientist at NEB. “And we have new proteases to study not just the glycan, but the protein as well—if you have a glycoprotein, it’s a two-way street and you need to analyze both components,” adds Beth McLeod, a research associate at NEB.
In Glorious Array
Just as the widespread availability of gene microarrays jump-started genomics, so too is the emergence of glycan arrays expected to take glycomics to the next level. “Many people in discovery mode have proteins that they think might recognize carbohydrates, but they don’t know what the structure is that is being recognized. By using a broad array with many different carbohydrate structures or epitopes, you can overlay a labeled protein and see what lights up. From that, you can infer what structures are recognized biologically,” says James Paulson, professor of molecular biology and molecular and experimental medicine at the Scripps Research Institute in La Jolla, California.
The most widely used array at present is an academic-derived, 600-glycan array from the National Center for Functional Glycomics (NCFG) .Commercial glycan arrays have taken longer to reach maturity, but Paulson predicts that is about to change. “It involves a huge investment of resources to synthesize these glycans, but the good news right now is that arrays of a couple hundred glycans are becoming available commercially, and I expect that to grow pretty rapidly.”
Biocompare’s Proteomics Search Tool
Find, compare and review proteomics
tools from different suppliers Search
The two leading companies offering commercial glycan arrays are Z Biotech and the cancer biopharmaceutical company Siamab Therapeutics. “These are still fledgling operations, but their arrays are getting to the point where they’re useful enough that hundreds of publications have appeared using them,” Paulson says.
A significant challenge of glycan arrays is that they require specialized printing, as well as systems to analyze them and collate the data. One service provided by the NCFG is helping investigators analyze findings from glycan arrays. “New formats are just around the corner that will take advantage of multiplexing, which I envision will be more amenable to use in diagnostic tests,” Paulson explains.
A Glycan of One’s Own
Natural glycans are difficult to isolate, making it challenging to characterize them or take advantage of them as drug targets. Thus, synthesizing glycans is key to advancement. At least a dozen companies have now formed, many with support from the NIH’s Small Business Innovation Research and Small Business Technology Transfer programs, to synthesize glycans needed in the field. “Ten years ago, there were very few commercially available glycans, while today there are hundreds and hundreds. We expect that to grow exponentially, with thousands available in the next five to ten years,” Cummings states.
All of these efforts are finally beginning to produce “tangible fruit,” says Rachel King, co-founder and CEO of GlycoMimetics, which has a pipeline of drug candidates that are rationally designed as “mimetics” of naturally occurring carbohydrates, rather than natural carbohydrates themselves. The company has two drug candidates that have reached the clinical trial phase: its glycomimetic compound for sickle cell disease, GMI-1070 (rivipansel), a pan-selectin antagonist, has moved on to a Phase III trial after strong Phase II data, while GMI-1271, an E-selectin antagonist that completed a Phase I/II trial in acute myeloid leukemia, recently received U.S. FDA Breakthrough Therapy Designation.
“Glycomics is very different from genomics. With the genome, if you discover a gene, you can make a protein, and often the protein itself can be the drug,” says King. “That has not been the case with naturally occurring carbohydrates. They are often metabolized very quickly and don’t have drug-like properties, so the ability to go from the knowledge of the native structure, to the glycomimetic which has a drug-like property, requires the knowledge base that we’re building here. We model structures based on the native carbohydrates, then do the organic chemistry to make the novel structures and see how they bind. We then optimize them for drug-like properties and develop synthetic methods that can be scaled up.
In September 2016, Harvard’s Radcliffe Institute for Advanced Study hosted an international workshop on mapping the human glycome. “We announced the quest to go after the human glycome, defining all the glycan structures that humans can and do make—a major international effort, working with colleagues in Europe and Japan and involving hundreds of laboratories,” says Cummings. The ultimate deliverable of the Human Glycome Project, or HuGlyP, would be a 4-dimensional map of glycans and their expression, achieved by isolating glycans from human cells and tissues, determining their structures, and defining their spatiotemporal expression. Now that would be sweet.
Image: Dreamstime Images