Biotherapeutics are claiming an ever-increasing share of the pharmaceutical pie, with monoclonal antibody (mAb)-based therapeutics accounting for more than half of the potential therapeutics being developed and characterized.
Biotherapeutic manufacturing relies not only on template-driven processes to mass-produce the exact same amino acid chain, but also on nontemplate-driven processes to modify the resultant protein (or peptide), affording a broad diversity of product depending upon, among other things, the host cell and culture conditions. Meanwhile, post-translational modifications (PTMs) such as glycosylations, for example, are known to have crucial effects on stability, activity, immunogenicity, binding affinity and other properties crucial to the safety and efficacy of mAb-based drugs.
The composition, length, branching and linkage of glycans on any given mAb species can vary tremendously.
This microheterogeneity is essential to mAb function, and it is important not only to characterize it in the final product but also to monitor (and perhaps reproduce) it during process development—scale-up or lot-to-lot manufacturing, for example—as well as when creating biosimilars. Here we look at the use of mass spectrometry (MS) to monitor the glycan structure of mAbs.
Know what to look for
To track a mAb’s glycosylation, it’s important to have a prior analysis of it. This is generally accomplished by “an initial characterization phase (or at the very least some form of literature search on your protein of interest) in order to know what to expect,” notes Guillaume Tremintin, Bruker Daltonics’ market area manager, biopharma. Such characterizations are performed twice in a biopharmaceutical’s life cycle: First in the discovery phase, to know what’s there, whether that’s a new entity or one to be replicated as a biosimilar. Then the exercise is repeated on the few lead candidates being prepared for clinical trials.
“Between those stages, there’s a lot of development. You have to select the best possible cell line, the best possible growth conditions and the best formulation. One of the attributes you’re going to look for is whether they have an attractive glycan profile, in the case of a new molecule, or a profile that is highly similar to the originator, in the case of a biosimilar project,” Tremintin explains. “And when the protein is being expressed, and they’re doing scale-up of those expression conditions, people will look very closely at the glycan profile. These would be the monitoring steps—you have an idea of what you’re looking for, and you don’t need to do quite such complex assays.”
There are at least three ways to monitor glycosylation by MS in such cases: at the (perhaps reduced) intact protein level, typically using straight liquid chromatography (LC)/MS; at the glycopeptide level, following enzymatic digestion of the protein; and at the released glycan level, following a glycosylase step. “I’m not sure people will do all three—they would have their validated methods to do the monitoring,” says Ying Qing Yu, a senior scientist on the biopharmaceutical team at Waters Corp. “I don’t think there is a set standard for analysis.”
Because only a single purified product is being monitored, and there is only a single available glycosylation site in the mAb’s conserved region, monitoring glycosylation is more a matter of knowing the identities of glycans and their relative quantities than of knowing on which residue they are located.
An accurate mass instrument will be able to tease out even small differences in molecular weight among the various species of glycan (except for isomers), allowing for their relatively facile identification when compared against a database. “Once you get into monitoring aspects, you may not be as motivated to use a lot of MS/MS,” remarks Tremintin.
In fact, the use of MS at all may be overkill for mAbs. “Given the limited complexity of glycans coming out of monoclonals, it’s a question of whether you’re going to need the power of the MS for monitoring, like you might do for some other, more complex glycoproteins,” notes Michael Betenbaugh, professor of chemical and biomolecular engineering at Johns Hopkins University. “But of course, if you can get the turnaround time as good or better for MS as you can for some of these other tools [such as LC and fluorescence labeling of the released glycans], then I think people would rather use MS.”
Take it easy
There is a trend in the industry to make things easier for the researcher. In terms of the front end, “I think there’s a significant amount of effort going into making the sample prep easier,” notes Tremintin. This ranges from simpler digestion and glycan-release strategies to “robotic front ends or things you can hook up directly to bioreactors—that’s something we’re working on at the moment.”
Another growing trend is for vendors to package “workflows that build in a complete system—meaning that you have the LC part and the MS part and the informatics all together to do a type of analysis,” points out Yu. This may include compliance aspects—something that a proteomics discovery workflow doesn’t typically get into.
Robert Cole, director of the Mass Spectrometry and Proteomics Facility at Johns Hopkins, thinks that for monitoring purposes the industry will shift toward using capillary electrophoresis (CE) coupled with high-resolution MS. He cites “the fact that these CE instrument interfaces are now getting very easy to use and robust. That becomes much easier—we’re starting to do that.”
Make it easy
It’s one thing to know the amino acid sequence and even the glycosylation profile of an originator compound (which may be published), but “companies don’t reveal their magic sauce that is used to generate it,” points out Gurmil Gendeh, pharmaceutical marketing manager at Shimadzu. But there is an expanding body of knowledge about how culture conditions—from temperature and oxygen partial pressure to glycerol and nucleoside sugar concentrations—influence the type and quantity of PTMs found on the mAbs.
“We are using LC/MS/MS, based on a triple-quad MS, to monitor the substrates in the cell-culture media that drive glycosylation in a certain direction,” Gendeh shares. Previously, this was performed with LC and sensors; however, that type of information was “very limited” compared with the nearly 100 vitamins, nucleic acids and other primary metabolites queried by the current solution, which was developed in the past few years in collaboration with the company’s customers. “It’s a complete package … including a software component,” Gendeh says, adding that all of the solution’s components are “critical for you to be able to measure all those compounds in a single analysis in a short time.”
This exemplifies another trend in the industry. “With advancements in informatics and LC/MS instrumentation, some companies are starting to monitor multiple attributes, meaning multiple PTMs, in a single analysis,” Yu says. “Like glycosylation and methionine oxidation and glycation … at the same time.”
As the technology of and around MS continues to improve—including speed, sensitivity and accuracy of instrumentation; integration; and informatics—its use to monitor the glycosylation profiles of mAbs will only get easier.
Image: Shutterstock Images