The value of added sugar
Roche experts have developed a patented method for enzyme production. It enhances the production process for therapeutic antibodies.
Therapeutic antibodies and proteins are the fastest growing segment of the global pharmaceutical market, and for a good reason. Their unique ability to zero in on specific receptor proteins on the surface of a target cell and thereby trigger processes such as cancer cell death has heralded a new era of targeted therapies.
However, receptor targeting is just one factor behind the efficacy of these compounds. In more than 70 percent of all currently approved protein medicines, it is branched chains of sugar molecules attached to the protein that determine how efficiently these drugs can kill cells, how long they stay in circulation, and how they modulate the immune system. No wonder that biopharma researchers around the globe are trying to optimize the composition of the sugar chains (known as glycans) to make the medicines even more effective.
Feverish search for robust processes
Optimization remains an uphill struggle because the cells used to make the proteins do not always deliver the protein with the desired sugar structure. Instead, production cell lines, even optimized ones grown under ideal production conditions, produce a heterogeneous mix of proteins with up to 15 different—though very similar—sugar structures. Even tiny variations in production conditions or in the growth medium for the cells increase that heterogeneity. They change the amount of enzymes, which the cells use to string together the different sugar building blocks in the right order to form the branched sugar chain that determines protein function.
If only one enzyme or sugar building block is lacking, even for a moment, the cell will build something other than the desired sugar chain.“Alongside the complete product you often get proteins that are not fully glycosylated,” says Harald Sobek from Roche Diagnostics in Penzberg, Germany. “If you have a mixture of many different sugar structures, this will inevitably affect the yield of the active pharmaceutical ingredient.” Since a high yield is crucial to the economic viability of the production process, researchers are feverishly searching for robust processes that reduce the unwanted product heterogeneity.
Lego with sugar “bricks”
Together with a team under Rainer Müller at Roche Custom Biotech, Harald recently presented a method that enables the sugar chains to be completed in a highly efficient manner.“It is very hard to reduce inhomogeneity in the fermenter,” says Rainer. “Although efforts are being made to improve product homogeneity by optimizing the growth medium for the cells, you never get 100 percent.” Harald adds: “For that reason, we started looking for a way to perfect the sugar chains subsequent to protein production.” Akin to playing with lego, the aim is for highly specific enzymes, dubbed glycosyl transferases, to attach the different sugar units one-by-one at exactly the right position in the sugar chain.
It’s an elegant solution, and more complex than it might appear at first sight. After completing the fermentation, the researchers add chemically activated sugars plus the enzymes to the pre-purified product. “Whereas the core glycan is usually in place, quite often the two sugars at the terminus of the chain, galactose and sialic acid, are not,” says Harald.
That’s why the sugar optimizers started their market launch with alpha-2,6 sialyl transferase, which adds the terminal sugar unit (sialic acid) to the sugar chain, making the protein stable. They are currently launching the 1,4-galactosyltransferase enzyme, which attaches the second-last unit. Experimental data obtained so far have proven them right. “In tests, 100 percent of the sugar chains were correctly completed by our galactosyl transferase,” says Rainer. Using sialyl transferase, efficacy was at least 90 percent, which represents a major improvement over the 50 percent or less of correctly glycosylated proteins that most fermentations deliver.
Up until recently, enzyme production was an expensive, laborious and inefficient process in which only experts can produce the required enzymes. “So far, production was only possible using insect cells, which delivered minute yields,” says Harald. The Roche researchers, on the other hand, have developed and patented a new method. It’s the very first process in the world that enables sufficient enzyme amounts to be produced in mammalian cells that industry uses commercially to manufacture therapeutic proteins.
Harald and Rainer are not about to divulge any technical details, but they are happy to talk about how they want to build the market for their novel products. “First, we are going to supply the R&D departments of the pharmaceutical industry,” explains Rainer. At the same time, the market roll-out of a second wave of products will be prepared. These products are compliant with the strict GMP (good manufacturing practice) rules that regulatory agencies require industry to follow for commercial production.