Insect cell/baculovirus display is a well-established technology, yet unlike bacterial or yeast surface display, this system has not been extensively used for protein engineering applications (Makela and Oker-Blom 2006; Makela and Oker-Blom 2008). Despite the relatively few examples of protein engineering with insect cell/baculo-virus systems, which require greater expertise in handling compared to microbial-based systems, recent studies have shown much potential. This format involves either the infection of insect cells with baculovirus and subsequent expression of displayed proteins on the cell wall, or display of proteins on the virus capsid. Display of het-erologous proteins mainly uses the gp64 major envelope glycoprotein as an anchor (Boublik et al. 1995), but other proteins have also been used (Ernst et al. 1998). This eukaryotic system has the ability to process proteins with complex folds and the capability to perform post-translational modifications. The first demonstration of this system for the display and screening of a library involved the display of an HIV-1 gp41 epitope specific for a neutralizing monoclonal antibody (Ernst et al. 1998). Amino acids adjacent to the epitope were randomized to produce a library of 8000 variants on baculovirus-infected insect cells. Flow cytometric sorting led to a clone with significantly increased binding affinity to the antibody. Baculovirus-infected insect cells have also been used to display peptide libraries bound to class I and class II MHC proteins (Crawford et al. 2006). In this system, each insect cell was able to display a unique MHC-peptide complex on its surface. These libraries were screened with soluble, fluorescent T cell receptors to identify peptide mimotopes for MHC class I-specific (Wang et al. 2005a) and MHC class II-specific (Crawford et al. 2004) T cells. Furthermore, these mimotopes could be used to stimulate T cells expressing the same receptor. Follow-up studies demonstrated the ability to use baculovirus-infected cell-displayed libraries to examine the role of MHC-bound peptides in the interaction of the staphylococcal enterotoxin A superantigen with an MHC class II protein, and to identify peptide mimotopes that bound with high affinity to an MHC class I protein and an antigenic TCR for potential uses in tumor therapy (Crawford et al. 2006). Although there are limited examples of using insect cell/baculovirus technologies for protein engineering, this methodology should be generally applicable for studying other ligand/receptor interactions.
Cell surface display systems using mammalian hosts are also useful for engineering eukaryotic proteins. As in the case of insect cell display, mammalian surface display platforms have not yet found widespread use for protein engineering, but this may also be due to the high level of skill needed on the part of the researcher. Recent reports suggest that their use will become more prominent in the near future. As an initial proof of concept, a peptide library of ~107 members was displayed as fusions to the N-terminus of the CCR5 chemokine receptor in Jurkat-e cells, and screened to isolate an epitope mimetope (Wolkowicz et al. 2005). Beyond this study, most mammalian cell display efforts are motivated by the need to address problems that arise when translating antibodies engineered in nonmammalian hosts to diagnostic and therapeutic applications. With this in mind, human embryonic kidney 293T cells were used to display scFvs tethered to the transmembrane domain of human platelet-derived growth factor receptor (Ho et al. 2006). An scFv with higher affinity was enriched 240-fold by a single pass of cell sorting after being mixed with a large excess of cells expressing wild-type antibody with a slightly lower affinity. This study also reported significant enrichment of a mutant with increased binding affinity for CD22 after a single round of screening from an scFv library with a randomized antibody hotspot. The authors reported that transformation protocols with this system permit library sizes of ~109 variants to be relatively accessible. Antibody libraries of ~108 members of whole IgG molecules were displayed on the surface of a human embryonic kidney-derived cell line, and screened for binding to target by a combination of magnetic bead sorting and flow cytometry (Akamatsu et al. 2007). IgG antibodies that neutralize human and mouse interleukin-12 were successfully isolated from this library. In another report, human scFvs with specificity to virus antigen or to nicotine were isolated from a mammalian display library in BHK cells (Beerli et al. 2008). The therapeutic potential of monoclonal antibodies derived from the nicotine-specific scFvs was validated in a preclinical mouse model by inhibiting nicotine entry into the brain.
Was this article helpful?