Possible cell death mechanisms for pars compacta nigro-striatal dopamine neurons in Parkinson's disease include oxidative stress, inflammatory processes, nitric oxide iron accumulation, glutamate toxicity and diminished neurotrophic factor responses. There is a notion that Parkinson's disease is not a single disorder but a syndrome that can be initiated by several factors. Because of limitations of biochemical methods in the global analysis of neuronal death, a full picture of events has not been established. However, recently developed cDNA microarray or microchips, in which the global expression of thousands of genes can be assessed simultaneously, is changing the prospect for understanding the disease process, its progression, response to drugs, etc. The neurotoxin N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) is considered the most valid model of Parkinson's disease. We employed the technique of cDNA microarray gene expression to determine the mechanism of action of MPTP in mouse substantia nigra. Also, we studied neuroprotective processes induced by several compounds, including R-apomorphine and the green tea polyphenol epigallo-catechin-3-gallate (EGCG). This was done in two ways: (1) the time-dependent acute effect of MPTP, for determining which of the initial genes might lead to dopamine neuron death and (2) gene expression at the time of MPTP-induced dopamine neuron death. We observed that early (acute MPTP) gene expression differs from effects seen at the time of death (chronic MPTP), and that early gene changes are crucial for setting into action genes that eventually cause dopamine neuron death. Furthermore, this process is a cascade of "domino" effects, some of which were previously established by biochemical means. However, our findings show an additional large number of events previously unknown. The neuroprotective drugs reversed some but not all of the gene expression, suggesting involvement of these genes in the neurodegenerative process. Because of the profound complexity of "domino" effect it is now reasonable to understand why a single neuroprotective drug has not shown clinical neuroprotective efficacy. Future multi neuroprotective drugs may be necessary for treatment of not only Parkinson's disease, but other neurodegenerative diseases (e.g. Alzheimer's disease) and detrimental states (e.g. ischaemia).