Lentiviral gene therapy has held great promise for treating a wide range of neurological disorders due to its ability to stably integrate into the genome of nondividing cells like neurons, in addition to dividing cells. The nervous system is a complex and highly heterogeneous system, and while a therapeutic intervention may have beneficial effects in one population of cells it may have severe side effects in another. For this reason, specific targeting of lentiviral vectors is crucial for their ultimate utility for research and clinical research use. Two different approaches for focusing the targeting of lentiviral vectors were employed in these studies. The first method involved assessing the effects of vector production strategies on the resulting virus’s tropism both in vivo and in vitro. The changes in vector transduction were determined via flow cytometry on cells in culture and immunohistochemistry following brain injections. Results from these experiments suggest that while the production conditions do impact the vectors efficacy, there is not a distinct effect on their tropism. A unique characteristic of retroviral and lentiviral vectors is their capacity for being pseudotyped, conferring a new tropism on the vector. Native tropisms are generally not specific beyond very broad cell types, which may not be sufficient for all applications. In this case, chimeric targeting molecules can provide an even more refined targeting profile compared to native pseudotypes. The second approach utilizes novel chimeric glycoproteins made from nerve growth factor and the vesicular stomatitis virus glycoprotein. These chimeras are designed to pseudotype lentiviral vectors to target nociceptive sensory neurons for a variety of disorders. While these chimeras were successfully produced as protein, they were misfolded and sequestered in the endoplasmic reticulum and therefore unavailable to produce lentivirus. While neither strategy was completely successful, they do provide interesting information for the design and creation of lentiviral vectors. This research shows that small differences in the steps followed as part of a lentivirus production protocol can greatly impact the resulting vectors efficacy. It also shows that while VSV has been used to create chimeric glycoproteins, not all targeting molecules are suitable for this purpose.

Indiana University-Purdue University Indianapolis (IUPUI) / Lentiviral vector mediated gene therapy has made great strides in recent years with several successful clinical trials. However, adverse events encountered with some early trials have highlighted the necessity to improve upon its safety. Improvements can range from early steps in vector production to evaluation of insertion sites post-transduction. We have evaluated an FDA approved DNase for removal of residual plasmid DNA during vector production, developed novel non-integrating lentiviral vectors and employed modified insertion site analysis post-transduction to improve the safety of lentiviral vector mediated gene therapy. To prevent the exposure of gene therapy patients to HIV-1 DNA it is essential to remove residual plasmid DNA during vector production. We evaluated a recombinant human DNase which has been FDA approved for use in patients as an alternative to a bacterially derived DNase. Our results indicate this DNase is an effective alternative with a potentially safer profile for use in patients. The ability of lentiviral vectors to stably integrate their genome into a host cell’s DNA can have negative side-effects due to the risk of insertional mutagenesis. Non-integrating lentiviral vectors have been developed to alleviate this risk in applications where integration is not necessary. However, a low frequency of illegitimate integration persists when using these vectors. We have developed a novel non-integrating vector mutation and evaluated the efficacy of combining it with other mutations for reducing the frequency of illegitimate integration. We demonstrate that combining mutations that inhibit integration can further reduce the frequency of illegitimate integration. Several methodologies have been developed for evaluating the insertion sites of normal integrating lentiviral vectors. Illegitimate integration by non-integrating vectors demonstrates mechanisms which result in insertions and/or deletions at the vector-genome junction. Current methods lack the sensitivity to account for these variables in a high-throughput manner. We have adapted modifications to current methods to improve the capture of these variable insertion sites for analysis. The results of these studies improve the safety of lentiviral vector mediated gene therapy by improving the purity of the vector product, providing a safer vector for non-integrase mediated applications, and allowing more sensitive analysis of insertion sites post-transduction.