Altered neuronal gene expression in brain regions differentially affected by Alzheimer's disease: a reference data set

Abstract

Alzheimer's Disease (AD) is the most widespread form of dementia during the later stages of life. If improved therapeutics are not developed, the prevalence of AD will drastically increase in the coming years as the world's population ages. By identifying differences in neuronal gene expression profiles between healthy elderly persons and individuals diagnosed with AD, we may be able to better understand the molecular mechanisms that drive AD pathogenesis, including the formation of amyloid plaques and neurofibrillary tangles. In this study, we expression profiled histopathologically normal cortical neurons collected with laser capture microdissection (LCM) from six anatomically and functionally discrete postmortem brain regions in 34 AD-afflicted individuals, using Affymetrix Human Genome U133 Plus 2.0 microarrays. These regions include the entorhinal cortex, hippocampus, middle temporal gyrus, posterior cingulate cortex, superior frontal gyrus, and primary visual cortex. This study is predicated on previous parallel research on the postmortem brains of the same six regions in 14 healthy elderly individuals, for which LCM neurons were similarly processed for expression analysis. We identified significant regional differential expression in AD brains compared with control brains including expression changes of genes previously implicated in AD pathogenesis, particularly with regard to tangle and plaque formation. Pinpointing the expression of factors that may play a role in AD pathogenesis provides a foundation for future identification of new targets for improved AD therapeutics. We provide this carefully phenotyped, laser capture microdissected intraindividual brain region expression data set to the community as a public resource.

Fig. 1

Heat map of differentially expressed genes in the entorhinal cortex (EC). A heat map for EC was created to display those statistically significant (P < 0.01 with multiple testing corrections applied) genes with the greatest region-specific fold changes. Normalized expression signals are represented on a log scale for which colder colors correspond to lower levels of expression and warmer colors correspond to higher levels of expression. Heat maps were generated with Genecluster 2. AD, Alzheimer’s disease.

Fig. 2

Heat map of differentially expressed genes in the hippocampus (HIP). A heat map for HIP was created to display those statistically significant (P < 0.01 with multiple testing corrections applied) genes with the greatest region-specific fold changes.

Fig. 3

Heat map of differentially expressed genes in the middle temporal gyrus (MTG). A heat map for MTG was created to display those statistically significant (P < 0.01 with multiple testing corrections applied) genes with the greatest region-specific fold changes.

Fig. 4

Heat map of differentially expressed genes in the posterior cingulate cortex (PC). A heat map for PC was created to display those statistically significant (P < 0.01 with multiple testing corrections applied) genes with the greatest region-specific fold changes.

Fig. 5

Heat map of differentially expressed genes in the superior frontal gyrus (SFG). A heat map for SFG was created to display those statistically significant (P < 0.01 with multiple testing corrections applied) genes with the greatest region-specific fold changes.

Fig. 6

Heat map of differentially expressed genes in the primary visual cortex (VCX). A heat map for VCX was created to display those statistically significant (P < 0.01 with multiple testing corrections applied) genes with the greatest region-specific fold changes.

Fig. 7

AD pathogenesis: relationships in neurofibrillary tangle (NFT) and plaque formation. This figure shows a summary of the relationships of factors encoded by statistically significant genes (α = 0.01, corrected) and which displayed altered expression across profiled regions. Factors shown were selected (through literature searches) based on previous research implicating/showing a role of respective factors in pathways surrounding formation of tangle and plaque pathologies. The statistically significant genes shown in this figure were only identified in EC, HIP, MTG, and PC analyses. Increased and decreased expression are identified with up and down arrows, respectively. Arrows with 2 heads indicate that multiple probes, isoforms, or subunits demonstrating different directions in expression changes were identified (separate fold changes for all factors are located in the supplemental material for this article). Each brain region is represented with a color code. For those factors that are encoded by multiple subunits (e.g., multiple genes), an average of the fold changes was calculated to evaluate the expression change. UBE, ubiquitin factors (ubiquitin-activating and -conjugating enzymes, ubiquitin ligases); PSM, proteasomal subunits; PHF, paired helical fragment.

Fig. 8

RT-PCR validation of selected genes. RT-PCR was used to independently validate unprofiled control subjects and AD cases in MTG (control: n = 9, AD cases: n = 6) and PC (control: n = 8, AD cases: n = 8). Genes for validation were selected based on relevance to AD pathophysiology and demonstration of significant expression changes from array analysis. Normalized folds are shown on the y-axis, and genes are listed on the x-axis. Quantitative results are listed in Table 2.