The relative paucity of presentations utilizing array technologies to study alterations of gene expression in Alzheimer's disease indicates that these powerful technologies are yet to be fully utilized in the study of AD. Three laboratories submitted abstracts using either cDNA or oligonucleotide arrays to examine message expression and one laboratory submitted an abstract using the recently developed protein chip array technology in combination with mass spectroscopy to examine a selected aspect of expression of Aß species..
Walter J Lukiw et al. from the LSU Neuroscience Center in New Orleans, LA, used Clontech cDNA arrays to compare the expression of 1184 genes in a pooled sample of message extracted from CA1 from five AD brains with a pooled sample of message from five control brains. The two samples were said to be matched for age (about 69), postmortem delay (about two hours), drug history, and agonal states. None of the AD brains sampled had a known history of familial neurodegenerative disease. Quantification utilized "control and alignment markers" that included α-tubulin, β-actin, HLAC1, HPRT, GAPDH, 60S ribosomal protein 13A, 40S ribosomal protein 9 and ubiquitin (many of which have in fact been shown to be up- or down-regulated in AD). Of the 22 largest differences detected, the AD sample showed significant decreases in the expression of genes encoding seven brain transcription factors and four messages involved in synaptic transmission, such as synaptophysin and ChAT. There were significant increases in message for six potentially proinflammatory genes, including β-amyloid precursor protein (bAPP), interleukin-1 (IL-1) precursors, and cytoplasmic phospholipase A2 (cPLA2). Lukiw suggested that these increases support the hypothesis of highly active neuroinflammatory processes operating in terminal AD hippocampal CA1.
In an additional presentation, Lukiw and Bazan utilized what appeared to be similar Clontech cDNA arrays of 1184 human cDNAs to evaluate recombinant IL-1ß induction of message expression by normal human neural progenitor (NHNP) cells in stable primary culture. They reported that IL-1β-induced cells showed 5.1-fold upregulation of message for cytoplasmic phospholipase and 9.1-fold induction of vascular endothelial growth factor. A 30-minute preincubation with the hetrazepine BN50730 suppressed these increases in cPLA2 and VEGF dramatically, and induced an independent set of genes including the gene encoding human anti-apoptotic factor Bcl-2.
Guilio M Pasinetti and his coworkers at Mount Sinai School of Medicine, New York, examined the expression of over 8,734 genes in brains of mild AD compared to cases with normal cognitive status. Twenty-five differentially regulated genes were identified and related to clusters of specific biological variations. Two such clusters were found to have patterns that correlated with variations in signal transduction pathways and cytoskeleton integrity. Pasinetti suggests that these early molecular markers of AD may correlate more closely with neuropsychological indicators of early AD than traditional neuropathologic measures of plaques and tangles.
Coleman et al. from the University of Rochester utilized home-made cDNA arrays, Clontech cDNA arrays and Affymetrix oligonucleotide arrays to examine message induction of homogenates and of single cells in AD and control brains. Ante-mortem blood samples were also examined. Brain regions examined were CA1 and superior frontal gyrus. Rather than utilize single messages as loading controls they compared the expression level of each gene in the arrays to every other gene in the arrays. They also analyzed the data using multivariate canonical analyses. The data showed: 1) that more genes are up-regulated than are down-regulated in AD. However, total message level is decreased in AD because some genes are down-regulated many-fold more than is the case for up-regulated genes; 2) a large number of genes related to the cell cycle were induced in AD, as were genes related to inflammatory responses. Down-regulated genes included ones related to neuronal plasticity (e.g., GAP-43), synaptic function (e.g., synaptophysin) and structural (e.g., actin) and other "housekeeping" activities (e.g., GAPDH); 3) multivariate analysis of expression data distinguished AD from control samples on the basis of homogenates of superior frontal gyrus, single cells from hippocampal CA1 and blood samples.
Message induction is one step in the progression from DNA to the delivery of a posttranslationally modified protein to its site of action. A single application of a relatively new protein chip technology was presented by Brian M. Austen et al. from St. George's Hospital Medical School, and Ciphergen, all of United Kingdom. This technology utilizes the binding of proteins to aluminum chips. The chips can be prepared to capture molecules on the basis of molecular weight, hydrophobicity, antibody binding or a variety of other properties. The bound species are then subjected to mass spect analysis to detect the m.w. of species captured on the chip. This method allows the discrimination of several hundred protein species, whose identity must be established by further analyses. This technology was used by Austen et al. to examine the effect of cholesterol on the production of Aß species by APP transfected HEK cells. In this study, antibody to the N-terminal 10 residues of Aβ was covalently linked on the ProteinChips. Mass analyses permitted detection of 12 discreet Aβ fragments, with 1-40 being the major species. The ProteinChip was also used to show that Aβ1-42 is the major variant in AD brain homogenates. Aβ1-40 secreted from cells was increased by preincubation with exogenous cholesterol (200mg/ml), and decreased by preincubation with lovastatin (50mg/ml).
In summary, array technologies are still being sparsely utilized. They are providing data that are consistent with data in the literature that have been obtained by other means. New molecules involved in the molecular pathology of AD are also being detected, but their identity is often either not yet determined or not revealed. There are also cases in which results from array technology are in conflict with data in the literature, a situation that emphasizes the need for confirmation of selected findings from arrays by other quantitative means, including quantitative in situ hybridization or real-time quantitative RT-PCR.—Paul Coleman
No Available References
No Available Further Reading