Altered gene expression profiles reveal similarities and differences between Parkinson disease and model systems.

Parkinson disease (PD) targets dopaminergic neurons in the substantia nigra, resulting in motor disturbances such as resting tremor, bradykinesia, and rigidity. Pathogenic processes likely occur over several decades, in that an overwhelming percentage of neurons are already dead at the time of clinical diagnosis. For this reason, the usage of animal model systems to discover the early steps in the pathologic cascade is required. These include exposure to the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), which selectively kills dopamine neurons in the substantia nigra, and genetic models incorporating mutations in the alpha-synuclein gene that cause disease in human patients. Through the evaluation of these models at multiple time points, it is possible to discover novel gene expression changes that may underlie disease pathogenesis. Specifically, the authors hypothesize that animal models of PD and human PD brains share a gene expression profile that signifies certain aspects of pathogenesis and/or recovery-resistance. To test this and similar hypotheses, the authors and others have utilized new microarray technology that enables the sampling of thousands of genes' expression level in one assay. Because the technology is fairly new and results can vary depending on methods used, results must be evaluated with care. Multiple array and data-mining options can be used to make the most accurate inferences as to differentially expressed genes in each set of samples. The authors developed a fusion classifier approach whereby individual data-mining algorithms generate lists of significant genes. The lists are subsequently queried, and only genes unanimously called significant are retained for further validation. Although the authors' approach identified hundreds of differentially expressed genes in each of three PD systems, only a few were common between the human and animal substantia nigra. These were related to dopamine phenotype, synaptic function, and the mitochondrial metabolism, implicating the presynaptic terminal as a primary site of injury. The time course of the authors' experiments indicates that if the synaptic changes could be prevented, this may alleviate some cell death, in that these changes precede neuronal loss.