Team Develops Method for Glycosylating Proteins

By Jason Socrates Bardi

Alfred Hitchcock, who is famous for meticulously planning every shot of some of his movies before he exposed even a single frame of film, once remarked that actors should be treated as cattle—presumably so that they can be herded through their scenes the way he wanted.

Like making movies, producing glycoproteins—proteins that have specific carbohydrates (sugar groups) covalently attached to them—is a process that is sometimes easier to envision than enact. Molecules, like actors, can be difficult to work with.

Nature excels at attaching specific sugars to specific amino acid residues within proteins through posttranslational modification, and, by some estimates, more than half of all the proteins in the human body have carbohydrate molecules attached. Posttranslational modification is widely used by nature to modify the folding, stability, activity, and interactions of proteins, playing a central role in cell signaling, development, immunology, and many other important biological processes.

Scientists would likewise like to have a way to create these same molecules in the laboratory in order to have an easy way to examine the effect of glycosylation on the structure or function of proteins and to develop glycosylated therapeutics, such as certain antibodies, cytokines, erythropoietin, and other medicines. These goals have been hindered, however, by the lack of a convenient and reliable laboratory method for producing selectively glycosylated proteins.

Now a team of scientists from The Scripps Research Institute (TSRI) has developed a powerful method that takes a huge step in the direction of developing such a method. The team was led by Professor Chi-Huey Wong, who holds the Ernest W. Hahn Chair in Chemistry, and Peter G. Schultz, who holds the Scripps Family Chair in Chemistry. Wong and Schultz are also both members of The Skaggs Institute for Chemical Biology at TSRI.

Using a technology developed in Schultz's laboratory to site-specifically introduce unnatural amino acids into proteins, the team has demonstrated the ability to create glycoprotein "mimetics"—proteins that contain unnatural glycosylated residues that mimic naturally glycosylated proteins.

In the study, Schultz, Wong, and their colleagues engineered a codon within a gene that codes for a small domain of the staphylococcal protein A. In response to this engineered codonan unnatural amino acid—called p-acetyl-L-phenylalanine is inserted into the protein. This unnatural amino acid resembles the natural amino acid phenylalanine but with a "keto" group attached. The keto group is like a chemical handle to which the sugars can be attached.

Schultz and Wong found that when they incubated the keto-containing protein with a sugar solution containing the sugar aminooxy saccharide, the sugar was covalently attached to the protein at the end of the day. They further found that they could sequentially attach a second sugar to the first, and a third sugar to the second through enzymatic manipulations.

This technology is general enough to synthesize glycoprotein mimetics of any protein that can be expressed in E. coli, say the authors.

To read the article, "A Method for the Generation of Glycoprotein Mimetics" by Haitian Liu, Lei Wang, Ansgar Brock, Chi-Huey Wong, and Peter G. Schultz, please see J. Am. Chem. Soc., 125 (7), 1702 -1703, 2003.




Two routes to sweetening proteins. Schultz, Wong, and colleagues glycosylated the unnatural amino acid-containing protein with three sugars. They were able to attach the sugars both one at a time (clockwise steps) and all at once. Image by Haitian Liu. Click to enlarge.