Genetic re-targeting of adenoviral vectors for gene therapy applications

Abstract

The adenovirus (Ad) is a frequent cause of upper respiratory tract infections, enteritis and conjunctivitis. Since its first isolation in 1953, the study of Ad has contributed greatly to the understanding of intracellular events and has also served as a model system for many viruses. Today, Ad is one of the most commonly used vectors for gene therapy and is considered to have considerable potential within this field. The main aim of this thesis was to develop a technology, based on genetic engineering of the virus fiber, to create genetically re-targeted adenovirus type 5 (Ad5) constructs that could be used as in vivo gene therapy vectors. To achieve this aim, the knob and the 15 terminal shaft repeats of the virus fiber gene were deleted and replaced with an external trimerization motif and a new cell-binding ligand. Different polypeptide ligands for cell-surface molecules, such as single-chain antibodies, epidermal growth factor and affibody-like molecules (see below) were cloned into the recombinant fibers. Phenotypic analysis of the fiber constructs was performed, and the ability of Ad5 to encapsidate the new ligand-bound fibers to yield functional re-targeted viruses was studied. Recombinant viruses were characterized with regard to cell specificity, fiber content, growth rate and infectivity. Recombinant fibers retained the basic biological functions of the native fiber, i.e., trimerization, nuclear import, penton formation, and ligand binding, on the condition that the inserted ligand folds correctly within the cellular cytoplasm. Correct cytoplasmic folding of the ligand is the major limiting factor in the ability of recombinant fibers to be rescued into viable recombinant Ad5. Therefore, ligands that depend on disulphide bond formation for stable folding are not functional within this context, since the reducing milieu in the cytoplasm is not conducive to the formation of disulphide bonds. We used a new class of cell-surface-binding ligands, which is based on a triple a-helix framework from one of the Ig-binding domains of staphylococcal protein A. This Ig-binding domain folds correctly in the cytosol and can be mutated at certain positions so that it retains its a-helical framework. Ligands that are generated by this technique have new specificities so-called affibodies, and are well suited to the construction of Ads with novel tropisms. It has been demonstrated that recombinant viruses expressing affibody-like ligands or the RGD motif infected cells through binding of the new ligand to cellular receptors and had lost the binding properties of the native Ad5 that were associated with the fiber knob. It was also shown that viruses with manipulated fibers had diminished growth rates and infectivity and had reduced fiber content. The reasons for these defects have not been fully elucidated.The results of the work described in this thesis show that it is possible to generate viable, recombinant, knobless Ad5 with new cellular specificities by deleting the fiber knob and replacing it with a new cell-binding ligand and a non-native trimerization motif. Previously, a major constraint on the ligands that could be used for genetic re-targeting of Ad was the ability to fold correctly in the cellular cytoplasm. We have identified a new class of ligands that meet this criterion. Fundamental techniques for the construction of genetically re-targeted Ads for human gene therapy are described. However, recombinant viruses that have been generated according to the above strategy are less infectious and contain fewer fibers per virion compared to wild-type Ad5. It will be necessary to optimize the technology to overcome these limitations before recombinant targeted viruses can be applied to human gene therapy.