A Novel In Vitro Glycosylation Approach For Difficult PTMs
A conversation with Gabriel Cook, Oklahoma State University
As drug pipelines shift toward increasingly complex targets, a persistent challenge has emerged that defies standard processing logic. Temperamental membrane proteins make achieving precise post-translational modifications far more difficult than for the soluble proteins that defined the first wave of bioprocessing.
While mAbs are relatively robust and well-behaved in solution, membrane proteins are notoriously fickle. They are more prone to aggregation, often falling out of solution the moment they are removed from their native lipid environment. The instability creates a significant bottleneck for CMC teams, particularly when it comes to glycosylation.
We met Gabriel Cook, Ph.D., at the CHI PepTalk conference in San Diego, where he described the work his lab is doing on membrane protein glycosylation to address these exact hurdles. The assistant professor at Oklahoma State University and his lab are using an in vitro approach to bring a level of predictability and stability to membrane proteins long enjoyed by those working with simpler biologics.
The core of the challenge lies in the environment. Traditional methods often rely on detergent micelles to keep these proteins soluble, but these lack the structural integrity of a human cell. Cook’s transition toward nanodisc technology represents a significant leap forward, offering a bilayer that mimics the dimensions of a native cell membrane. This ensures that the proteins are not just soluble, but correctly folded and functional.
For process development scientists and engineers, the ultimate goal is scalability at any phase. The ability to control glycosylation patterns is paramount. Cook's research into a generalizable protocol offers a potential roadmap for the industry, suggesting a future where difficult membrane proteins can be modified with the precision required for large-scale pharmaceutical manufacturing. The transcript below is edited for clarity.
It seems like in vitro glycosylation should be well understood by now. What are the barriers to the approach?
Cook: First of all, we're looking at membrane proteins, which are difficult to study on their own. They fall out of solution, they aggregate, they're not quite as easy to work with as soluble protein. It's an understudied area just because it's been difficult to get samples to do experiments like we do in our lab. We're really trying to get a system that allows us to first solubilize those proteins in an environment that's similar to what they see in a cell.
The glycosylation aspect of it is another area that's been very difficult to study because of the heterogeneity of the samples. There are lots of different sugars that can be put on these proteins. If you've got the cell introducing all these different types of sugars lengths and sizes, it can lead to a really big mess in these samples. We're making it as simple as we can. We take a membrane protein — for example, a native human protein associated with a disease, such as gamma-sarcoglycan for muscular dystrophy — and express and purify it in bacterial cells to remove those sugars. And then we're building those sugars up in a very controlled way by putting specific sugars on those specific sites. And so we don't have that heterogeneity problem, and we have a homogeneous sample that we can then work with to do these experiments on those samples.
You started doing this research with micelles and recently transitioned to nanodiscs. What led to that move, and what benefits do nanodiscs provide over the detergent aggregate environment?
Cook: Micelles are a very simple sample mode. They have just one or maybe just a couple of types of detergents in there that help solubilize the protein.
But the dimensions of those lipid assemblies and detergent micelles are not the same dimensions as a human cell. You'll see a lot of people who are working on membrane proteins now using nanodiscs because they have very similar dimensions to the membranes in human cells. That way we can, when we're looking at those proteins in these nanodiscs, we can say that those proteins are behaving and are folded in a very similar way to what they are in a cell because the membrane, the bilayer that we have, is very similar to that membrane in a cell.
Our readers are often transitioning from preclinical to clinical manufacturing models, or from clinical to commercial. Is there a design-of-experiments approach to help determine whether in vitro glycosylation grows with the process through each phase of development?
Cook: Right now, we're working on several proteins — I think the list is up to six or seven. We're attempting to make a general protocol for anybody who's working on membrane proteins and want to see what they look like with sugars attached.
Once they purify their protein like they would normally and incorporate it into these lipids and detergents, we can then glycosylate those proteins. We're aiming to make it so that, no matter what protein system you're working on, if it's a membrane protein that gets glycosylated, you could use these methods to put sugars on your protein and determine whether they change the structure, dynamics, or protein-protein interactions.
This way, it would offer broad applicability. And that also goes for large-scale pharmaceutical companies that want to look at peptides that get glycosylated. Can we give them a method for them to be able to glycosylate those things and look at them in their samples? Yes.
About The Expert:
Gabriel Cook, Ph.D., is an associate professor at Oklahoma State University where he studies the structure, dynamics, and interactions of membrane glycoproteins. He received his bachelor's degree at Concordia University in biology and a Ph.D. in biochemistry from Kansas State University. He performed postdoctoral research at the University of California-San Diego.