-
A Tangled Web
Submit responseBuckner et al. (2009) are to be commended for their illuminating investigation of the spatial congruence between beta-amyloid (β- amyloid) deposition, as revealed by Pennsylvania Compound B (PiB)-positron emission tomography (PET), and elements of the default mode network (DMN), as revealed by functional magnetic resonance imaging (fMRI). The cortical “hubs” they have identified are crucial neuroanatomical substrates of clinical dementia (Royall et al., 2002), and not just Alzheimer’s disease (AD)-related dementias (Zekry et al., 2003).
Furthermore, it seems very plausible that those specific regions of interest (ROI) succumb to AD pathology as a result of their exceptional connectivity within the DMN. However, the authors’ speculation that this follows from an increased metabolic demand, which might unleash a β- amyloid “cascade” is not consistent with the literature, and there are alternative explanations for the vulnerability of these ROI, which could also arise from their central positions within the network.
First, it is not true that these ROI are exceptionally vulnerable to AD lesions. In fact, they appear moderately vulnerable to paired helical filament (PHF) tauopathy (Royall et al., 2002) and neurofibrillary tangle (NFT) formation (Arnold et al., 1991), relative to CA1, for example, which is not centrally located in the DMN. Second, these ROI are not exceptionally metabolically active. While the posterior cingulate is relatively highly active at rest in non-demented persons, so are frontal system nuclei that are less important to the DMN, while the other DMN hubs are intermediately active structures (Szabo et al., 1992). Third, these same ROI are actually hypoactive in non-demented persons at risk for AD, even in the absence of psychometric impairment (Reiman et al., 1996). At that point in the disease, there may be β-amyloid in those ROI, but there are no NFT (Royall et al., 2002).
One alternative explanation for Buckner’s finding may be that these ROI share a relatively late myelogenesis (Brodmann, 1910). I cannot speculate what effects this may have on the voxelwise correlations used to produce Buckner’s network, but Braak and Braak (1996) have previously noted that Alzheimer-related neurofibrillary changes in the neocortex inversely recapitulate cortical myelogenesis.
A better studied alternative explanation is that AD lesions may propagate retrogradely and transynaptically throughout neural networks, from an origin in the olfactory cortex (Pearson & Powell, 1989). This propagation can explain the elegant laminar distribution of AD lesions and is at once inconsistent with the β-amyloid cascade hypothesis [because “senile” plaque (SP) and NFT do not coincide in space (Armstrong, Meyers & Smith, 1993)] and consistent with Buckner’s findings (because the transynaptic propagation will inevitably lead to involvement of the most highly interconnected regions).
A few of Pearson & Powell’s predictions from 1989 bear reconsideration in light of Buckner’s work:
“… it is predicted that the degree of involvement of an area in terms of NFT and SP formation would depend not only upon the projection to a single affected area, but upon the number of efferent targets of that area which are also involved in the disease process, and on the degree of interconnection between NFT containing neurons and other adjacent, initially unaffected cells within the same area (Pearson & Powell, 1989; pg. 108).”
“One example of such differential involvement of areas according to their connections might be predicted to be the areas of the association cortex in the temporal and frontal lobes having heavy connections with the amygdala as well as with the parahippocampal gyrus. Such areas should be more severely affected than similar areas of the neocortex connecting with only one of these… (Pearson & Powell, 1989; pg. 108).”
Finally, transneuronal propagation of AD lesions through the DVM would help explain the most obvious weakness of the β-amyloid cascade hypothesis, i.e., early involvement of the neocortex by SP /Pib PET β -amyloid in the absence of neocortical NFT, but late involvement of the hippocampus by SP /Pib PET β-amyloid in contrast to NFT.
“…it follows…that SP should be more widely distributed than NFT’s….for example, a cell in the locus coeruleus with widely divergent axon collaterals will contain an NFT if one of its projection sites becomes involved in the disease; degeneration of all its collaterals would lead to plaques in all the sites to which the neuron projects, whether or not the neuronal cell bodies in all the projection areas contain NFTs; in the case of a nucleus such as the locus coeruleus which does not receive axons from the cortex, this plaque formation would not lead to subsequent tangle formation by retrograde transport of the causative agent. Orthograde transneuronal transfer would lead to tangle formation over a longer time course… Powell, 1989; pg. 107).”
References: Armstrong RA, Myers D, Smith CU. The spatial patterns of plaques and tangles in Alzheimer's disease do not support the 'cascade hypothesis'. Dementia. 1993;4:16-20.
Arnold SE, Hyman BT, Flory J, Damasio AR, Van Hoesen GW. The topographical and neuroanatonical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cerebral Cortex 1991;1:103-116.
Braak, H., & Braak, E. (1996). Development of Alzheimer-related neurofibrillary changes in the neocortex inversely recapitulates cortical myelogenesis. Acta Neuropathologica, 92, 197–201.
Brodmann, K. (1910). Feinere Anatomie des Großirns. In M. Lewandowsky (Ed), Handbuch der Neurology (Erster Band, p. 241). Berlin: Springer.
Buckner RL, Sepulcre J, Talukdar T, Krienen FM, Liu H, Hedden T, Andrews-Hanna JR, Sperling RA, Johnson KA. Cortical Hubs Revealed by Intrinsic Functional Connectivity: Mapping, Assessment of Stability, and Relation to Alzheimer’s Disease. The Journal of Neuroscience, 2009; 29:1860–1873.
Pearson RCA, Powell TPS. The neuroanatomy of Alzheimer’s disease. Rev Neurosci, 1989;2:101-122.
Reiman EM, Caselli RJ, Yun LS et al. Preclinical evidence of Alzheimer’s disease in persons homozygous for the e4 allele for apolipoprotein E. N Engl J Med 1996;334:752–758. Royall DR, Palmer R, Mulroy A, Polk, MJ, Román GC, David J-P, Delacourte A. Pathological determinants of clinical dementia in Alzheimer’s disease. Experimental Aging Research 2002;28:143-162.
Szabo Z, Camargo EE, Sostre S, Shafique I, Sadzot B, Links JM, Dannals RF, Wagner HN Jr. Factor analysis of regional cerebral glucose metabolic rates in healthy men. Eur J Nucl Med. 1992;19:469-475.
Zekry D, Duyckaerts C, Belmin J, Geoffre C, Herrmann F, Moulias R, Hauw JJ. The vascular lesions in vascular and mixed dementia: the weight of functional neuroanatomy. Neurobiol Aging. 2003;24:213-219.
- Print ISSN: 0270-6474
- Online ISSN: 1529-2401
