Inflammatory Changes and Apoptosis in Parkinson’s Disease. In: Mizuno Y, Fisher A, Hanin I, editors. Mapping the Progress of Alzheimer’s and Parkinson’s Disease. Boston: Springer US; 2002. pp. 259-63. [DOI: 10.1007/978-0-306-47593-1_43]

Role of cytokines in inflammatory process in Parkinson's disease

We investigated whether the cytokines produced in activated microglia in the substantia nigra (SN) and putamen in sporadic Parkinson's disease (PD) are neuroprotective or neurotoxic. In autopsy brains of PD, the number of MHC class II (CR3/43)-positive activated microglia, which were also ICAM-1 (CD 54)-, LFA-1 (CD 11a)-, TNF-alpha-, and IL-6-positive, increased in the SN and putamen during progress of PD. At the early stage activated microglia were mainly associated with tyrosine hydroxylase (TH)-positive neurites in the putamen, and at the advanced stage with damaged TH-positive neurons in the SN. The activated microglia in PD were observed not only in the nigro-striatal region, but also in various brain regions such as the hippocampus and cerebral cortex. We examined the distribution of activated microglia and the expression of cytokines and neurotrophins in the hippocampus of PD and Lewy body disease (LBD). The levels of IL-6 and TNF-alpha mRNAs increased both in PD and LBD, but those of BDNF mRNA and protein drastically decreased specifically in LBD, in which neuronal loss was observed not only in the nigro-striatum but also in the hippocampus. The results suggest activated microglia in the hippocampus to be probably neuroprotective in PD, but those to be neurotoxic in LBD. As an evidence supporting this hypothesis, two subsets of microglia were isolated from mouse brain by cell sorting: one subset with high production of reactive oxygen species (ROS) and the other with no production of ROS. When co-cultured with neuronal cells, one microglia clone with high ROS production was neurotoxic, but another clone with no ROS production neuroprotective. On the other hand, Sawada with coworkers found that a neuroprotective microglial clone in a culture experiment converted to a toxic microglial clone by transduction of the HIV-1 Nef protein with increasing NADPH oxidase activity. Taken together, all these results suggest that activated microglia may change in vivo from neuroprotective to neurotoxic subtsets as degeneration of dopamine neurons in the SN progresses in PD. We conclude that the cytokines from activated microglia in the SN and putamen may be initially neuroprotective, but may later become neurotoxic during the progress of PD. Toxic change of activated microglia may also occur in Alzheimer's disease and other neurodegenerative diseases in which inflammatory process is found.