The long-term goal of this research is to understand at the molecular level the factors involved in the tissue-specific expression of neuronal genes and to use this information in conjunction with the enormous advances in molecular genetics to create animal models in which gene expression is spatially and temporally controlled. Based on our observation that a -2.5 kb tyrosine hydroxylase promoter fragment will target midbrain dopaminergic neurons, whereas a -2.4 kb promoter fragment will not, we have hypothesized that sequences critical for cell type-specific expression are located within this region. To test this hypothesis we will determine the consequences of further deletions as well as whether nuclear factors will bind identified sequences. Once identified, these sequences would serve as the starting point for isolating associated regulatory factors that presumably play a role in the development of a dopaminergic phenotype. A second major conceptual goal is to develop animal models in which proteins thought to contribute to the demise of the midbrain dopaminergic neurons are either conditionally induced or deleted. Midbrain-specific targeting fragments defined in the last grant period will be combined with an inducible system to create a genetic switch allowing for spatially and temporally controlled gene expression. This inducible targeted system will allow testing of the hypothesis that overexpression of cell death proteins such as caspase-3 (CPP32) in adult animals will trigger the death of the dopamine neurons and may perhaps serve as a model of Parkinson's disease. Once a spatially and temporally regulated system is achieved, efforts will be made to combine it with the Cre/loxp binary system, which is based on DNA recombination for manipulating gene expression in vivo. In these experiments an inducible Cre recombinase is only activated within dopaminergic neurons. Crossing these animals with mice carrying conditional mutations (genes flanked by loxp sites) will make it possible to examine the consequences of a loss of function phenotype within a few cells at a given time. Initially, transgenic reporter mice will be characterized as to the specificity and inducibility of the Cre/loxp system. In the future loxp-tagged glutamate receptors and/or VMAT2 will be crossed with the TH/Cre animals to test hypotheses related to neurodegenerative disorders such as Parkinson's disease. Taken together, both the gain-of-function and loss-of-function approaches described here will allow a detailed analysis of the underlying mechanisms of a specific animal model for a human disease. Conceivably, knowledge gained from these studies may offer new routes for therapeutic intervention.
