Two years ago, Schenk et al. showed that immunization with amyloid-β42 peptide reduced the accumulation of amyloid plaques in a transgenic model of Alzheimer's disease. The excitement generated by this finding was revived last December, when two independent groups reported that this vaccination procedure could also reduce memory loss in two different murine models of the disease. By using modified versions of the Morris water maze, Janus et al. and Morgan et al. obtained evidence that monthly immunization with amyloid-β reduced the behavioural deficits and plaque accumulation observed in the transgenic animals.

Although there are methodological differences between the two papers, both of them favour the idea that a reduction in amyloid accumulation may suffice to prevent memory loss. The findings also support the hypothesis that amyloid plaques, and not other histopathological hallmarks of Alzheimer's such as neurofibrillary tangles, are responsible for the cognitive impairments that characterize the disease.

Also, the two studies are clear examples of the increasing sophistication of behavioural testing in mouse models of Alzheimer's and other neurodegenerative diseases, as the tasks used by these researchers aimed to tap into specific forms of memory. But despite this sophistication, these reports also remind us of the need to develop a more extensive battery of tests to increase the validity of the transgenic models. These papers, together with the three recent articles ( 1, 2, 3)on the evidence for genetic linkage of late-onset Alzheimer's disease to chromosome 10, tell us that the Alzheimer's story is far from over.