Characterization of an immune-evading doxycycline-inducible lentiviral vector for gene therapy in the spinal cord.

Gene therapy is a powerful approach to promote spinal cord regeneration. For a clinical application it is important to restrict therapeutic gene expression to the appropriate time window to limit unwanted side effects. The doxycycline (dox)-inducible system is a widely used regulatable gene expression platform, however, this system depends on a bacterial-derived immunogenic transactivator. The foreign origin of this transactivator prevents reliable regulation of therapeutic gene expression and currently limits clinical translation. The glycine-alanine repeat (GAR) of Epstein-Barr virus nuclear antigen-1 protein inhibits its presentation to cytotoxic T cells, allowing virus-infected cells to evade the host immune system. We developed a chimeric transactivator (GARrtTA) and show that GARrtTA has an immune-evading advantage over "classical" rtTA in vivo. Direct comparison of lentiviral vectors expressing rtTA and GARrtTA in the rat spinal cord shows that the GARrtTA system is inducible for 6 doxycycline-cycles over a 47 week period, whereas with the rtTA-based system luciferase reporter expression declines during the 3rd cycle and is no longer re-inducible, indicating that GARrtTA provides an immune-advantage over rtTA. Immunohistochemistry revealed that GARrtTA expressing cells in the spinal cord appear healthier and survive better than rtTA expressing cells. Characterization of the immune response shows that expression of GARrtTA, in contrast to rtTA, does not recruit cytotoxic T-cells to the transduced spinal cord. This study demonstrates that fusion of the GAR domain to rtTA results in a functional doxycycline-inducible transactivator with a clear immune-advantage over the classical rtTA in vivo.

Frameshift proteins in autosomal dominant forms of Alzheimer disease and other tauopathies.

Frameshift (+1) proteins such as APP(+1) and UBB(+1) accumulate in sporadic cases of Alzheimer disease (AD) and in older subjects with Down syndrome (DS). We investigated whether these proteins also accumulate at an early stage of neuropathogenesis in young DS individuals without neuropathology and in early-onset familial forms of AD (FAD), as well as in other tauopathies, such as Pick disease (PiD) or progressive supranuclear palsy (PSP). APP(+1) is present in many neurons and beaded neurites in very young cases of DS, which suggests that it is axonally transported. In older DS patients (>37 years), a mixed pattern of APP(+1) immunoreactivity was observed in healthy looking neurons and neurites, dystrophic neurites, in association with neuritic plaques, as well as neurofibrillary tangles. UBB(+1) immunoreactivity was exclusively present in AD type of neuropathology. A similar pattern of APP(+1) and UBB(+1) immunoreactivity was also observed for FAD and much less explicit in nondemented controls after the age of 51 years. Furthermore, we observed accumulation of +1 proteins in other types of tauopathies, such as PiD, frontotemporal dementia, PSP and argyrophylic grain disease. These data suggest that accumulation of +1 proteins contributes to the early stages of dementia and plays a pathogenic role in a number of diseases that involve the accumulation of tau.

Frame-shifted amyloid precursor protein found in Alzheimer's disease and Down's syndrome increases levels of secreted amyloid beta40.

Frame-shifted amyloid precursor protein (APP(+1)), which has a truncated out-of-frame C-terminus, accumulates in the neuropathological hallmarks of patients with Alzheimer's disease pathology. To study a possible involvement of APP(+1) in the pathogenesis of Alzheimer's disease, we expressed APP695 and APP(+1) in the HEK293 cell-line and studied whether the processing of APP695 was affected. APP(+1) is a secretory protein, but high expression of APP695 and APP(+1) results in the formation of intracellular aggregate-like structures containing both proteins and Fe65, an adaptor protein that interacts with APP695. APP(+1) is shown to interact with APP695, suggesting that these structures consist of functional protein complexes. Such an interaction can also be anticipated in post-mortem brains of young Down's syndrome patients without any sign of neuropathology. Here we observed APP(+1) immunoreactivity in beaded fibres. Additional support for functional consequences on the processing of APP695 comes from a 1.4-fold increase in levels of secreted amyloid beta40 in cells co-expressing APP695 and APP(+1), although APP(+1) itself does not contain the amyloid beta sequence. Taken together, these data show that co-expression of APP695 and APP(+1) affects the processing of APP695 in a pro-amyloidogenic way and this could gradually contribute to Alzheimer's disease pathology, as has been implicated in Down's syndrome patients.