Our results demonstrate the feasibility and efficacy of DN-TNF gene transfer as a potential therapeutic strategy to achieve nigrostriatal neuroprotection in PD. Specifically, here we show that lentiviral DN-TNF infection results in detectable secretion of DN-TNF both in vitro
in a dopaminergic cell line and in primary mixed cultures, as well as in vivo
in the SNpc. Lentivirus-derived DN-TNF was efficacious in blocking TNF-induced signaling and microglial activation in vitro
and attenuated neurotoxin-induced DA neuron loss in vitro
and in vivo
. Results from this study in which nigral lenti-DN-TNF infection led to an approximate doubling of remaining nigral dopaminergic neurons (42% of nigral DA neurons remaining in lenti-DN-TNF infected animals compared to 24% remaining in lenti-GFP infected animals) are consistent with our previous report of nigral DA neuron protection mediated by the chronic infusion of the recombinant DN-TNF biologic XENP345 (31% of nigral neurons remained with vehicle infusion compared to 62% with DN-TNF infusion) 8
. There are a number of possible explanations for the greater variability in DA neuron survival obtained with lenti-DN-TNF delivery as opposed to infusion of the DN-TNF protein XENP345 . One possibility is that the amount of DN-TNF production in the lentivirus-infected SNpc may have been more variable compared to the constant amount of XENP345 protein delivered by chronic infusion through an osmotic pump; alternatively, the spread of lentivirus-derived DN-TNF protein may have been more restricted. Another difference in experimental design that may account for the greater variability in DA neuron rescue with lenti-DN-TNF versus infusion of DN-TNF protein is the inherent lag time (typically 3–4 days) required for gene expression following transduction with the lentivirus; this is not the case with infusion of DN-TNF inhibitor which is available immediately to block the potent neurotoxic effects of TNF. Thus, injection of the lentivirus days or weeks prior to 6-OHDA lesions (as done in several glial-derived neurotrophic factor (GDNF) studies 18–20
) might have led to an even greater rescue of DA neurons and larger behavioral effect. In addition to doubling the number of remaining nigral DA neurons after a striatal 6-OHDA lesion, lentiviral delivery of DN-TNF also attenuated behavioral deficits both in forelimb placement and drug-induced rotation resulting from 6-OHDA lesions. We did not investigate the neuroprotective effects of intrastriatal delivery of lenti-DN-TNF in these studies because previous experience with intrastriatal delivery of XENP345 afforded no neuroprotection 8
. Taken together, our studies provide proof of concept for the feasibility and efficacy of DN-TNF gene delivery as a means to administer TNF inhibitors into specific regions of the CNS without the need for an invasive chronic infusion device. In the future, it will be of interest to determine the extent to which significantly delayed TNF signaling inhibition still affords neuroprotective effects, a question of therapeutic relevance when one considers a clinical diagnosis of Parkinson’s disease means significant dopamine neuron loss has already occurred.
Although anti-TNF drugs may be effective disease-modifying therapies in other conditions characterized by chronic inflammation 21
, the currently FDA-approved systemic administration of anti-TNF biologics (i.e., large pegylated Fc-fused TNF decoy receptors or TNF antibodies) to treat peripheral autoimmune diseases such as rheumatoid arthritis and Crohn’s disease, is unlikely to provide adequate brain penetration to affect TNF signaling in the brain 22–24
. Therefore, CNS applications may require use of chronic infusion devices or alternative delivery methods such as the one reported here. Gene therapy approaches are attractive for use in PD treatment for several reasons including the spatially-defined and cell type-specific pathology of the disease, the requirement for consistent drug administration to minimize or prevent dose fluctuations, and the difficulty in chronically administering drugs which can not cross the blood brain barrier. Pending questions about the usefulness of a DN-TNF gene-based therapy in the clinic include the need to identify when and where it would need to be administered to achieve efficacy in patients. Because the disease process begins well before degeneration of nigral DA neurons gives rise to clinical symptoms, use of anti-inflammatory therapy is likely to be most effective in the early stages of disease. In addition, it may also be necessary to target several nuclei in order to achieve robust rescue and prevent progression of the disease. Spatial restriction of TNF signaling inhibition may be important as TNF signaling has been demonstrated to regulate synaptic scaling in hippocampal neurons 25
In summary, although the exact mechanisms responsible for degeneration of nigral DA neurons in PD have not been fully delineated, the wealth of data implicating inflammatory processes in the progressive loss of these neurons coupled with the protective effects of NSAIDs against idiopathic PD, provide compelling reasons to accelerate research approaches to selectively target inflammatory factors with demonstrated neurotoxic effects on DA neurons. Our studies demonstrate the feasibility and efficacy of dominant negative Tumor Necrosis Factor gene transfer as a novel neuroprotective strategy to prevent or delay nigrostriatal pathway degeneration in a pre-clinical model of PD.