Baculoviruses surface display of cellular attachment ligands as a new tool for transducing eukaryotic cells

Research project P 14538 Baculoviruses as Delivery Vehicle for Mammalian Cells Wolfgang ERNST 26.6.2000 Baculovirus, primarily the Autographa californica nuclear polyhedrosis virus, has been used for has been many years to over-express a variety of recombinant proteins in insect cells. A few years ago, baculoviruses have been discovered as a new tool for gene transfer into mammalian cells. Viral entry into human hepatocytes, Cos-7 CHO and many other cell lines could be demonstrated as well as high expression cells offers the advantage of a gene cassettes. Using bacalovirus for transduction of mammalian cells offers the advantage of a safe vector and authentic processing of recombinant proteins. The main objective of this project is to increase the efficiency of virus entry into mammalian cells, so that lower multiplicities of infection (mois) will result in high transduction rates. At present, infectious titers need to be as high as moi 200, however for medium or large scale expression, this is the main obstacle. During the past years we have established a baculovirus surface display system, which allows to specifically alter the viral surface by linking recombinant proteins major coat protein gp64. We also have demonstrated the successful modification of the native baculoviral envelope protein by inserting different epitopes. Based on these tools and the fact surface properties are responsible for virus entry, we suggest to introduce cellular attachments ligands, that are known to facilitate the entry into mammalian cells, thereby enhancing the efficiency of virus internalization. The amino acid motif RGD for example is a specific ligand to many integrins, present on the cellular surface. Adenovirus and Coxsackie virus use this strategy for entering their host cells. Also peptides such as the HIV-1 encoded Tat-protein or a Drosophila encoded transciption factor (Ant-p) are able to cross cell membranes and can be used for translocation. By introducing such peptides into suitable positions (antigenic structure) of gp64 and generating recombinant baculovirus particles with altered surface structure, we are hoping to increase viral infectivity, thus mammalian cells. (e.g. CHO) will be established, as well as baculoviral transduction of so far unsusceptible mammalian cells should be accomplished.

Recently, baculoviruses have been discovered as an alternative vehicle for gene delivery into mammalian cells, offering the advantage of high safety and no cytotoxicity on target cells compared to conventional viral vector systems. The aim of this project was the implementation and improvement of this baculoviral delivery system in order to enhance the gene transfer efficiency into mammalian cells. The idea to create such improved vectors is based on specific alteration of the surface properties of the wild type baculovirus. Surface modification of baculovirus particles has been previously used for the display of heterologous proteins, surface expression libraries aimed at selecting for specific ligands and to study protein functions. We have developed a technology which is able to alter the baculoviral surface by direct modification of the viral envelope protein gp64. Our approach allows to genetically modify every single copy of the envelope protein, which results in high copy numbers of the target sequence to be displayed. Based on this tool and the fact that surface properties are responsible for virus entry, we introduced several foreign attachment ligands, that are known to facilitate virus receptor binding and subsequent entry into the target cell. At first we tried to optimize the position of the ligand within the gp64 protein structure. Thereby we had to achieve optimal presentation features for the ligand without interfering with the functional integrity of the essential baculoviral envelope protein. Based on mathematical prediction models we selected several positions with high surface probability which where tested for their suitability by practical experiments. The binding site (6 amino acids) of a model antibody was inserted into these positions, and we used antigen- antibody interaction to measure signal intensity of the different mutants. The result of these analysis suggested amino acid position 274 and 283 in the gp64 baculoviral envelope protein for best presentation of desired ligands. Therefore we engineered two different attachment ligands, a 23 amino acid epitope from the foot-and-mouth disease virus (FMDV) and a synthetic peptide also containing the amino acid motif Arginine-Glycine-Aspartic acid (RGD) as well as the functional part (11 amino acids) of the protein transduction domain of the HIV tat protein into these positions. Functionality of the inserted sequences was shown by specific binding of RGD-viruses to their target integrin receptor in a solid phase receptor-ligand binding assay. The example of tat peptide was chosen to examine the properties of a transduction domain, acting by a receptor-independent mechanism. The transduction efficiencies of constructs AcFMDV283 and AcTAT283 was investigated in several mammalian cell lines by determining the level of reporter gene expression. For BHK-21 hamster cells we could observe specific targeting of viral mutant AcFMDV283, resulting in more than two-fold increase in reporter gene expression. In transduced Chinese hamster (CHO) cells the viral mutant AcFMDV283 caused a 4-fold enhanced, and AcTAT283 induced a ten-fold higher expression level of the reporter gene. Our results demonstrate that engineering of the FMDV- peptide and the HIV tat peptide into the baculoviral envelope protein gp64 results in changes of the viral phenotype with desired surface properties which can be utilized for enhanced uptake of baculoviral particles into different mammalian cell lines.