2 September 2005. The devastation that occurs by late-stage Alzheimer disease (AD) is well documented—almost every area of the cerebral cortex is decimated by neuronal loss. But where, exactly, does the damage begin? This has been a harder question to answer. One area of the brain that might succumb first is the medial temporal lobe (MTL). Pathological evidence shows that in this region of the cortex, neurofibrillary tangles and amyloid plaques, the two hallmarks of AD, appear early in the disease. But there have been some niggling doubts that this is really where the pathology starts. Because plaques are more widely distributed in early AD, for example, there may be areas that are affected even prior to the MTL. Evidence for this includes positron emission tomography (PET) imaging, which has shown that in early AD, things are not quite right in more posterior regions of the cortex (see Friedland et al., 1983). Now, using a combination of five in vivo imaging techniques, Randy Buckner and colleagues confirm that the areas first affected in AD are more posterior to the MTL. The regions affected are, in fact, the very same ones used when young adults daydream. The findings raise the intriguing possibility that AD not only starts farther back in the brain, but also much further back in time.
The data, which are based on both new and previous imaging studies, are reported in last week’s Journal of Neuroscience. Buckner and colleagues at Washington University, St. Louis, Missouri, including Mark Mintun and John Morris, teamed up with Bill Klunk and Chet Mathis from the University of Pittsburgh, Pennsylvania, to obtain images of tissue atrophy as revealed by structural magnetic resonance imaging (MRI), and amyloid deposition, detected using PIB-PET. PIB, or Pittsburgh compound B binds tightly to amyloid plaques and was developed by Klunk and colleagues to image these deposits in living tissue (see ARF related news story). They compared these images with previous data on default brain activity (meta analysis of H2O PET data), brain metabolism [multicenter data of 18-fluoro-deoxy-glucose (FDG) PET as reported previously, see Herholz et al., 2002], and memory (meta analysis of functional MRI data).
Buckner and colleagues found that the default H2O PET images of young adults closely matched the amyloid deposition maps generated by PIB-PET analysis of elderly volunteers with dementia. Furthermore, the PIB-PET analysis also revealed amyloid deposition in elderly volunteers who were deemed free of dementia. The areas of the brain where both PIB and H2O PET signals were highest included the precuneus—a section of the parietal lobe of the cortex—and the posterior cingulate and retrosplenial cortices, all regions that are posterior to the MTL. MRI analysis and FDG-PET analysis also showed that tissue atrophy and reduced metabolism, respectively, were evident in these three regions in AD patients. Significantly, MRI images from normal individuals, those with very mild and mild dementia, and those who had “converted” from having no dementia at the beginning to mild dementia during the study, revealed that the MTL and the precuneus were the two areas where atrophy occurred earliest.
Though all imaging modalities revealed changes in areas other than the posterior cortex, it was in the posterior regions near the precuneus and stretching into the posterior cingulate and retrosplenial cortex where all the images converged. The fMRI data also showed that in young adults given a simple memory test, successful recollection was associated with activity in these three regions.
The data hint that there may be a connection between areas in the brain where amyloid is deposited during AD and those areas that operate by “default” in the young adult brain. Default activity would include things like daydreaming, musing, or planning, write the authors, who speculate that: “The specific possibility raised by the present data is that memory networks are modulated, in some manner, during default cognitive states and, as a result, through as yet unspecified activity- or metabolism-dependent mechanisms, may cause preferential accumulation of amyloid.” However, they admit that this interpretation is fairly speculative and they raise several caveats, perhaps not least being that these studies were all carried out on different groups of patients or volunteers. Verification of the data and the interpretation would require comparisons within rather than among individuals, suggest the authors.—Tom Fagan.
Buckner RL, Snyder AZ, Shannon BJ, LaRossa G, Sachs R, Fotenos AF, Sheline YI, Klunk WE, Mathis CA, Morris JC, Mintun MA. Molecular, structural, and functional characterization of Alzheimer’s disease: evidence for a relationship between default activity, amyloid, and memory. J. Neurosci. 2005;25:7709-7717. Abstract