Many protein-based medications include attached sugars, or glycans. For these glycoproteins, the types of sugars and where they are located can significantly impact the performance of a drug. Also, any change to the method of producing a glycoprotein can alter the structure and location of the sugars. Consequently, profiling glycans makes up a crucial step in drug discovery through manufacturing. However, this is no easy task.
“In my opinion,” says Biswa Pronab Choudhury, technical director of the glycoanalytics core at the University of California, San Diego, “looking at glycans is complicated in the sense that there are several iso-types of glycans, which can bear the same sugars but differ by their linkage patterns as found in several N-linked glycans and O-linked glycans.”
Scientists once largely ignored the glycan composition and location in biotherapeutics. Now, says Detlev Suckau—global R&D manager biopharma solutions at Bruker Daltonics—“Everyone knows this is very critical.”
In fact, the glycans often determine the value of a medication.
In fact, the glycans often determine the value of a medication. As Rowan Moore, senior manager of Thermo Fisher Scientific’s European pharma and biopharma customer solutions center, puts it: “The level and type of glycosylation on a therapeutic protein can impact the safety and activity of the final drug product.” So, the glycans in a biotherapeutic must be controlled and monitored.
Profiling platforms
Given the diversity of glycoproteins, scientists use a collection of tools to analyze their structures and content. To analyze an N-glycan, for example, Choudhury recommends ultra-high-performance liquid chromatography (UHPLC) or high-performance anion exchange chromatography with fluorescence detection. “This is a fluorophore-assisted technique with high sensitivity and—as most of the glycans in therapeutic proteins are known—this would be a good start and affordable,” Choudhury explains. An alternative is LC followed by mass spectrometry (MS) or matrix assisted laser desorption/ionization (MALDI) MS-based identification of glycans. “For glycoproteomics, you need a high-end mass spec, which not all labs or small startups have access to or expertise on,” Choudhury explains. For O-glycans, he notes, “the common method is by mass-spec analysis of glycans or by LC-MS for glycoproteomics.”
Biocompare’s Mass Spec Search Tool
Find, compare and review mass spec
tools from different suppliers Search
The kind of MS also matters. When asked about the most useful analytical technique for glycan analysis, Suckau says, “You need to consider the context.” In clone screening during the early development of a glycosylated biotherapeutic, for instance, dozens to hundreds of samples must be analyzed quickly. For that, Suckau recommends quadrupole time of flight (Q-TOF) or LC-MALDI MS.
When trying to locate glycans at particular positions on a biotherapeutic, the glycan must remain attached to the protein. For that, Suckau recommends a glycopeptide-centric approach. In 2015, the National Institute of Standards and Technology (NIST) organized an interlaboratory study to measure glycosylation in monoclonal antibodies. “The average method identified 20 glycans,” Suckau says, “and our maXis Q-TOF identified more than 50.” So, it really matters how biotherapeutics get analyzed.
In some cases, scientists analyze free glycans. Here, hydrophilic interaction liquid chromatography (HILIC) can be used with MS/MS.
Really, it all depends on the feature that a scientist needs the most. MALDI provides fast analysis, for instance, but for in-depth analysis a scientist might opt for LC with electrospray ionization (ESI)-MS. When it comes to these features—speed and in-depth analysis—Suckau says, “you don’t get both at the same time.”
The importance of automation
Given the large numbers of glycoproteins that must be analyzed, automation can add speed and accuracy. One example is the AssayMAP Bravo automated protein and peptide sample prep solution from Agilent. “This platform provides superior sample preparation for glycoprotein purification or released glycan digest and labeling,” says David Edwards, senior director of mass spectrometry marketing at Agilent Technologies.
The samples are separated with the Agilent 1290 Infinity II UHPLC, using AdvanceBio LC columns, which were designed for glycan separations. “Then, the Agilent 6545XT AdvanceBio LC/Q-TOF system—designed specifically to handle multiple workflows in biopharmaceutical characterization—acquires the data for both large molecules, such as monoclonal antibodies, and smaller ones, such as released glycans, including use of a fragile ions mode to avoid unwanted fragmentation of glycans,” Edwards explains. Agilent MassHunter BioConfirm software analyzes the results. Edwards adds that this “enables the user to set up the analysis of intact proteins, protein digests, and released glycans using one panel.” Plus, a scientist can review the data in visualizations of, for instance, mass spectra or theoretical isotope distributions.
In some cases, automation can be used to take apart glycoproteins before analyzing them. “The majority of biotherapeutics are large, heterogeneous proteins, such as monoclonal antibodies,” says Moore. “In order to accurately and quantitatively study the attached glycans, a protein’s tertiary structure must first be disrupted, and if this is not performed effectively the subsequent deglycosylation—glycan release—could be biased.”
For unbiased deglycosylation and clean-up, scientists can use Thermo Fisher Scientific’s GlycanAssure HyPerformance kit. This product can be used as an automated magnetic bead-based N-glycan release and labeling technology. It can be used upstream of UHPLC with fluorescence or capillary-electrophoresis detectors. “The GlycanAssure HyPerformance kit does not contain sodium cyanoborohydide, it involves no vacuum drying steps, is high-throughput amenable and makes use of a well-understood, stable glycan dye,” Moore explains. In addition, this method involves just three steps that can be done in one well or tube, and this can be performed manually, semi-automated or fully automated.
The key to an unbiased method is getting a more complete picture of a glycoprotein. Moore says that, with her company’s technique, “no glycan-types are missed.” She adds, “This is particularly important for detection and monitoring of immunogenic moieties, which are a problem when working with particular cell lines—for example, murine cell lines—and high mannose and sialylated glycan species, which have the potential to affect drug clearance and thus impact in vivo efficacy.”
It is clear that for the glycans to get analyzed in biotherapeutics, the process must be effective and reliable. Also, as more glycoproteins get approved as medications, even faster and more accurate methods of glycan profiling must be developed. Suckau mentions seeing this topic coming up at meetings more and more. Consequently, basic and industrial scientists are seeking more tools and newer techniques to take on this challenge.