Main

Peptide display, the screening of random peptide libraries to find a motif that interacts with a biomolecule of interest, is a powerful technique in the molecular biologist's toolbox. By far the most popular scaffolds for display experiments are prokaryotic phages. Yet for some applications, the absence of eukaryotic post-translational modification on phage-displayed peptides is a serious drawback. Researchers have thus experimented with alternative eukaryotic systems, such as displaying peptides on the surface of yeast or of baculovirus, which is produced in insect cells.

But when one is looking at peptides that are normally expressed on the surface of mammalian cells, the context of a real mammalian membrane is irreplaceable. As Garry Nolan of Stanford University puts it, “There are situations where you are coming at an epitope, and you want it to be as much as possible in physiologic context, where it got the appropriate secondary modifications of sugar or other things in the normal surface of the cell, such as lipid components, that you are just not going to find in yeast”. In a recent paper in the Journal of Biological Chemistry, Roland Wolkowicz, a research associate in Nolan's lab, presents a new system to display peptide libraries on the surface of mammalian cells.

For any display system, the key to success is a scaffold that will be expressed at a high level and will display the peptide in a position that maximizes exposure. For this purpose, Wolkowicz and Nolan were inspired by the results of Warner Greene's group at the University of California, San Francisco, who showed that a specific epitope could be recognized in the context of the chemokine receptor CCR5. Wolkowicz thus adopted this seven-transmembrane-domain protein as a scaffold and engineered libraries in which peptides are displayed in its amino-terminal region, which is exposed on the outside of the cells. Having learned from the lab's experience with peptide libraries, Wolkowicz made sure to constrain the peptides' structure to a loop conformation to increase the chances that peptides would retain their binding affinity once cleaved from the display scaffold.

The CCR5-based library, once stably expressed on the surface of human T cells through retroviral transduction, can be screened for binding to a 'bait' by flow cytometry. A proof-of-principle experiment showed that a peptide diluted in the library to a ratio of 1 in 10 million random clones could be recovered after only three rounds of FACS sorting with a high-affinity antibody.

Nolan explains that the choice of CCR5 was also influenced by the fact that it displays peptides extremely close to the membrane. This feature opens the possibility of using this system for finding reagents that have to be active in the context of the cell membrane, such as ligands, receptor decoys and antibodies that block viral infection.

As Nolan points out, it is also a way of making 'epitope mimics', or mimetopes, of surface proteins, and this is actually one of the reasons why the technique was developed in the first place. Nolan describes a particular monoclonal antibody directed against the HIV envelope that is broadly neutralizing, but whose epitope has been impossible to map. Mammalian surface display offers a way of using this existing antibody to screen a library of peptides in search of a mimetope that recapitulates the shape and immunoreactivity of the original epitope and that could be used to make a vaccine. “That was sort of the genesis of the original technique,” says Nolan, “and now we are moving it forward. The next stage will actually be to put peptide libraries on and then go back to that original antibody.”