The development and manufacturing of biopharmaceutical products are complex endeavors, where the most minor details can profoundly impact efficacy, safety, and production efficiency. While much attention often centers on the protein backbone, there's a third layer of complexity that has gained increasing recognition for its critical role: glycan composition. 

These intricate sugar structures, attached to proteins as post-translational modifications, are not mere decorations. Their precise arrangement and presence are now understood as critical quality attributes (CQAs) that significantly influence a drug's efficacy and stability.

This concept is discussed in greater detail with Henning von Horsten in an episode of the Smart Biotech Scientist Podcast, hosted by David Brühlmann, founder of Brühlmann Consulting.

The Future of Biomanufacturing: A Paradigm Shift in Production Hosts

A compelling, even disruptive, vision for the future of bioprocess development centers on the potential of yeast cells to fundamentally alter drug manufacturing. While legacy products will continue to be made in animal cell culture, the trend in the pharmaceutical industry is moving towards obtaining completely defined, homogeneous molecules.

Achieving this level of homogeneity, it is argued, can be more reliably accomplished through chemical rather than biological means. In this context, yeast emerges as a powerhouse with immense potential to replace traditional CHO cells for future drugs eventually.

“Achieving this level of homogeneity, it is argued, can be more reliably accomplished through defined, controlled processes. In this context, yeast emerges as a powerhouse with immense potential to eventually replace traditional CHO cells for future drugs, offering a potentially more consistent and scalable biological production system.”

The rationale is compelling: yeast cells offer significant cost savings in manufacturing. They can double every 90 minutes, generating substantial biomass and recombinant protein in a properly folded fashion. This inherent efficiency begs the question of why expensive cell culture media and animal cell culture, which are far too costly for producing drugs intended for global access, should remain the dominant platform. 

If the goal is to shift drug manufacturing to regions like Africa and other countries, bringing down the cost of goods manufactured is paramount. Within 10 to 20 years, it is envisioned that yeast could largely supersede CHO culture. This outlook, although potentially contentious for those deeply invested in current animal cell culture practices, highlights a powerful economic and logistical driver for the evolution of biomanufacturing.

A Journey into the Chemistry of Life: The Path to Glycosylation Expertise

The journey into this specialized field often begins with an innate curiosity for chemistry, particularly the elegant and specific nature of biochemistry. For many experts, such as Henning, a background in both chemistry and biology, which leads to a passion for the "chemistry of life," serves as a strong foundation. 

This academic pursuit transitions into industry through varied career paths, often involving roles in business development, project management, and technology management within contract development and manufacturing organizations (CDMOs). The dynamic environment of such organizations can provide fertile ground for innovation, particularly in areas like glycosylation technologies.

The career trajectory can take unexpected turns, from industry back to academia, driven by a desire to teach, conduct research, and mentor the next generation of scientists. Collaborating with companies on conjoined projects, especially for small and medium enterprises, becomes a crucial avenue for research funding, demonstrating the interconnectedness of academic inquiry and industrial application. These non-linear paths often lead to specialized expertise, like that found in glycan composition.

Glycan Composition: The Third Layer of Complexity as a Critical Quality Attribute

The understanding of biologics has evolved significantly since the early 1990s, when the therapeutic potential of the body's molecules—such as interferons and monoclonal antibodies—was first recognized. Initially, the focus adhered to the classic dogma of molecular biology: gene to mRNA to protein.

Post-translational modifications, especially glycans, were overlooked or considered secondary. While glycans were early identified as modifications of increasing complexity, featuring branched structures, diverse anomericity, and complex terminal sugars, the prevailing view was that the protein's function was independent of its glycan involvement; an enzyme would work without glycans, an antibody would bind its target regardless.

However, the pharmaceutical industry, with its stringent focus on drug efficacy, safety, and uniformity, began to scrutinize glycan heterogeneity. The central question became: Does this lack of uniformity have an impact on a drug's efficacy or safety profile? This inquiry led to the pivotal discovery of the critical importance of glycans.

Sialylation, for example, was recognized early on as a key attribute to monitor. Sialic acids, which are anionic and reside at the terminal ends of glycan structures (likened to leaves on a tree), impact a molecule's pharmacokinetic (PK) properties. If these sialic acids detach, the protein can be rapidly bound to the liver and removed from circulation, thereby reducing its therapeutic action. A prominent case in point is epoetin, an extensively glycosylated molecule. 

Initially carrying three N-glycans, additional N-glycosylation sites were later added to darbepoetin alpha to enhance its beneficial PK behavior further. Ideally, these sites should be fully sialylated; however, the inherent complexity of animal cell processing often works against achieving complete, homogeneous sialylation.

Another crucial discovery involved fucosylation, which was found to influence not only pharmacokinetics but also the pharmacodynamic action of an antibody. These insights solidified the understanding that glycosylation is not merely an optional modification but a critical determinant of a therapeutic protein's functional attributes, making its monitoring and control essential.

There's competition between glycosylation and the synthesis activity and growth activity of animal cells, which makes it really complex to get all in. Once you can get a high titer of the desired target molecule, but then the glycan attribute will go down in quality or you get a low output but high quality glycan composition and vice versa. This is a conundrum that needed to be tackled and it's really, really difficult.

Modulating Glycans: Strategies for Enhancing Therapeutic Proteins

Once the critical importance of glycans for drug efficacy was established, biopharmaceutical engineers naturally sought ways to design processes that would yield products with ideal glycosylation attributes. However, this proved to be a formidable challenge due to the immense complexity of the cellular machinery involved.

Within a eukaryotic cell, the sugar nucleotide metabolism generates activated sugars, which are the building blocks for the large glycan trees. These sugars are synthesized in the cytosolic compartment and then transported through unique membrane transporters into the lumen of membrane-enclosed compartments, such as the endoplasmic reticulum (ER) and Golgi apparatusER, where enzymes utilize them to assemble glycans. This compartmentalization and intricate transport system create significant hurdles.

Further complicating matters, in a typical mammalian manufacturing process where cells double every 24 hours, there is intense competition for energy resources (ATP) and incoming sugars. The primary catabolism that degrades glucose not only supplies energy but also serves as the source for the sugar nucleotides needed for glycosylation. 

This competition between glycosylation, biomass synthesis, and cell growth activities in animal cells creates a conundrum: achieving a high titer of the desired target molecule often comes at the expense of glycan quality, and vice versa. This difficult balance, exacerbated by the metabolic uncoupling observed in the glycolysis and TCA cycle of some cancer cell lines, limits their efficiency in generating ATP from glucose.

Given this immense complexity, various strategies have been explored to engineer glycans:

• Media Composition Engineering: Early approaches, particularly at companies like Genentech, involved adding specific nutrients such as manganese, nucleotides, and galactose to the media. The aim was to feed into different parts of the cell's metabolism to elevate the pool of available sugar nucleotides at specific times. While this method showed some success, it often did not achieve the complete conversion levels desired by chemists, illustrating the inherent challenges of biological systems compared to precise chemical reactions.
• Cell Line Engineering (Deletion): A more direct strategy involves genetically engineering cell lines to remove specific glycan-attaching enzymes. For example, deleting the fucosyltransferase eight enzyme effectively eliminates fucose from the growing glycan structure. This "knockout" strategy can be highly effective in eliminating a particular sugar, but it also carries its risks, such as potential off-target effects on cell growth or viability. While effective for deletions, this approach has proven less successful for increasing the complexity of glycans, as simply elevating the amount of a glycosyltransferase (like galactosyltransferase for enhancing galactosylation) isn't enough; it also requires boosting transporter activity and synthetic activity in the cytosolic compartment.
• Transglycosylation with Chemically Synthesized Glycans: A more recent and auspicious advancement involves the synthesis of glycans as oxazolines and then using them as substrates for transglycosylation. This innovative method enables the production of a protein with a suboptimal glycan profile, followed by the addition of a desired, highly purified glycan in excess, which is then enzymatically linked to the protein's structure. If commercially feasible methods for synthesizing N-glycans in large quantities can be achieved (e.g., via enzyme-assisted synthesis in continuous flow reactors), this could be a "total game changer" for the industry, enabling optimized glycans to be hooked up to proteins, potentially even those synthesized in yeast. This represents an ambitious future outlook for achieving highly defined, optimized glycosylation.

Ensuring Consistency and Scalability in Production

A significant challenge in biopharmaceutical manufacturing is achieving consistency—producing the same molecule reliably over and over, not just in the lab but at large-scale production facilities. Scaling up a bioprocess is far from a simple linear transfer from the laboratory to large-scale production. As experts in the field understand, lab-scale experiments, often conducted in shaker flasks, do not directly translate to thousand-liter bioreactors.

Scaling up introduces complex considerations such as:

• Oxygen transfer: Ensuring cells receive enough oxygen (kLa values).
• Shear forces: Managing the mechanical stress on cells from impellers.
• Impeller design: Selecting appropriate types (e.g., Marine vs. Rushton) for optimal mixing and power input.
• Fluid dynamics: Achieving desired axial or radial transport.

These factors can create conflicting demands. For example, reducing mixing time to ensure rapid nutrient distribution often requires increasing power input, which in turn increases impeller tip speed, potentially leading to increased cell shearing. This delicate balance highlights the intricate challenge of process engineering, which can be as complex as engineering the cell itself. Mastering this transition from lab to large-scale facility is where specialized consulting becomes invaluable, guiding scientists through the highly complex bioprocess journey.

The Analytical Frontier: Measuring and Modulating Glycan Attributes

To ensure consistent production and to actively modulate glycan profiles, robust analytical capabilities are indispensable. Determining critical quality attributes (CQAs) for glycans presents its own set of challenges and advancements.

For N-glycans, the process is relatively well-established:

• Enzymatic Release: Most human N-glycans can be easily cleaved by the ubiquitous enzyme PNGase F, releasing intact glycan structures.
• Labeling and Separation: The released glycans can then be labeled with a fluorescent tag and separated by HPLC.
• Quantification: Peak area and relative fluorescence intensity can be used for direct quantification of the sample.

While effective and well-established, this technology demands significant skill and training. The workflow is intricate, involving the handling of toxic chemicals (though less toxic alternatives, such as picolimborane, are emerging), the removal of excess dyes, and multiple clean-up steps to eliminate impurities and minimize sample loss. 

These factors contribute to a complex and expensive process that requires enzymes, labeling techniques, specialized chemicals, and clean-up kits. Research is ongoing to develop more streamlined and straightforward workflows to address these issues.

For smaller companies without in-house expertise or the desire to acquire all the intricate skill sets required for advanced glycan analysis, the question arises: Are such complicated assays always necessary? The answer lies in the application. 

In academic fields like cancer research, where glycans are emerging as biomarkers, the need for the most intricate and highly resolved analytics is undeniable. Advanced technologies, such as capillary electrophoresis, which offer excellent separations and peak recognition without diffusion limitations, are vital in these contexts.

However, in biomanufacturing, where the complexity of glycan structures on, for instance, an antibody is somewhat confined and known, lower resolution but higher throughput might suffice for identifying primary quality attributes. This has led to the development of high-throughput methods, such as gel-based systems that separate glycans simultaneously on a polyacrylamide gel. These systems can process hundreds of samples within a short runtime, generating high-throughput data. Such patterns are particularly interesting for artificial intelligence approaches, which can learn from these patterns rather than relying solely on detailed analysis of sporadic peaks.

There is a clear need for both approaches: highly accurate and resolved analytics for academic research and biomarker discovery, as well as high-throughput, AI-driven data generation for biomanufacturing. The latter enables more frequent sampling in production processes to monitor and potentially control manufacturing, allowing for real-time adjustments to glycosylation profiles by adding additives or modifying culture parameters. The goal is to obtain readouts from hundreds of experiments simultaneously with high run precision, minimizing operator influence.

It's my conviction that the biggest grail in biomanufacturing is homogeneity of the glycans to get to a homogeneous product. Because with a homogenous product we can control an attribute that has influence on efficacy.

Remaining Challenges and the Future of Glycan Engineering

The pursuit of homogeneity in glycans remains the biggest grail in biomanufacturing. A homogeneous product enables better control over an attribute that directly influences drug efficacy, driving the industry toward its achievement. 

The primary challenges in this quest lie on two fronts:

1. Chemical Synthesis of Glycans: The ability to chemically synthesize complex glycans in large quantities is crucial. Ongoing work, leveraging enzyme-assisted synthesis and consecutive enzyme reactions in flow chemistry, aims to assemble glycans in vitro on a large scale. Once these become commercially available, methods like transglycosylation could enable the precise hooking of optimized glycans onto protein material, even if synthesized in hosts like yeast.
2. Yeast Cell Line Engineering: A significant hurdle for yeast as a production host has been its tendency to skip specific glycan sites, resulting in lower occupancy of N-glycan sites. Addressing this cellular challenge through advanced cell line engineering for yeast, combined with the availability of synthetic glycans, could propel the industry forward and significantly reduce the cost of goods for homogeneously glycosylated biologic modalities.

These two challenges, one at the cellular engineering level for yeast and the other at the chemical synthesis level for complex glycans, represent the major frontiers in glycan engineering.

Final Remarks

The most important takeaway from this discussion is unequivocally: glycans matter. Their importance is pervasive and extends beyond therapeutic proteins to critical areas, such as vaccines and advanced therapy medicinal products (ATMPs). 

For example, recent research has demonstrated the vital role of glycosylation sites in making vaccine constructs soluble, accessible, and capable of eliciting strong immune responses. Glycans contribute to vaccines for challenging targets, like the respiratory syncytial virus G protein, where constructs designed to self-assemble into virus-like particles require glycosylation to function effectively as immunogenic vaccines.

Glycan research is vital because glycans have a profound impact on many contexts. They are essential and should not be overlooked, as they will undoubtedly "come back to bite those who are not aware of them."

About Henning von Horsten

Henning von Horsten is a full professor of Industrial Bioengineering at HTW-Berlin, specializing in the GMP-compliant manufacturing of biopharmaceuticals from animal cell culture. As a Principal Investigator, he leads industry research projects with a focus on glycan engineering and analytics. 

His background includes various roles at ProBioGen AG, a biopharmaceutical CDMO. Dr. von Horsten is a co-inventor on several patents and has multiple publications in biomanufacturing. 

Connect with Henning von Horsten on LinkedIn.

David Brühlmann is a strategic advisor who helps C-level biotech leaders reduce development and manufacturing costs to make life-saving therapies accessible to more patients worldwide.

He is also a biotech technology innovation coach, technology transfer leader, and host of the Smart Biotech Scientist podcast—the go-to podcast for biotech scientists who want to master biopharma CMC development and biomanufacturing.  

Hear It From The Horse’s Mouth

Want to listen to the full interview? Go to Smart Biotech Scientist Podcast. 

Want to hear more? Do visit the podcast page and check out other episodes. 
Do you wish to simplify your biologics drug development project? Contact Us