Toward a New Concept of Alzheimer's Disease Models: A Perspective from Neuroinflammation.

The continuing failure to develop an effective treatment for Alzheimer's disease urges a better understanding of the pathogenic mechanisms and the improvement of current animal models to facilitate success for clinical interventions. The transgenic models have been so far designed to recapitulate one, or both, protein lesions found in the brain of patients, the extracellular amyloid plaques and the intraneuronal neurofibrillary tangles. However, in recent years, a third pathogenic component is gaining strength in the onset and progression of this disease, the neuroinflammatory response mediated primarily by the brain's resident immune cells, microglia. This has been highlighted by the identification of genes involved in innate immunity as risk factors to develop this neurodegenerative disease. Our current concept, mostly derived from amyloid-β producing models which show a robust microglial activation, supports an initial beneficial role of these glial cells followed by a pro-inflammatory cytotoxic function later on. This view is now challenged by emerging data in human postmortem samples. We have recently demonstrated that in the hippocampus of Braak V-VI individuals there is a prominent degenerative process of the microglial population, driven by phospho-tau, that might compromise neuronal homeostasis. This scenario of microglial dysfunction/degeneration should be taken into account for developing more reliable animal models of this disease and improve their predictive value for human drug efficacy testing. Finally, correcting dysregulated brain inflammatory responses might be a promising avenue to restore cognitive function.