Neurobiology of Aging 24 (2003) 397–413 Gene expression analysis in a transgenic Caenorhabditis elegans Alzheimer’s disease model Christopher D. Link a,∗ , Andrew Taft a , Vadim Kapulkin a , Kyle Duke b , Stuart Kim b , Qing Fei c , Douglas E. Wood c , Barbara G. Sahagan c a Institute for Behavioral Genetics, University of Colorado, Campus Box 447, Boulder, CO 80309, USA b Department of Developmental Biology, Stanford University School of Medicine, Stanford, CA 94305, USA c CNS Discovery, Pfizer Global R&D, Groton, CT 06340, USA Received 19 June 2002; received in revised form 8 October 2002; accepted 20 November 2002 Abstract We have engineered transgenic Caenorhabditis elegans animals to inducibly express the human beta amyloid peptide (A). Gene expression changes resulting from A induction have been monitored by cDNA hybridization to glass slide microarrays containing probes for almost all known or predicted C. elegans genes. Using statistical criteria, we have identified 67 up-regulated and 240 down-regulated genes. Subsets of these regulated genes have been tested and confirmed by quantitative RT-PCR. To investigate whether genes identified in this model system also show gene expression changes in Alzheimer’s disease (AD) brain, we have also used quantitative RT-PCR to examine in post-mortem AD brain tissue transcript levels of B-crystallin (CRYAB) and tumor necrosis factor-induced protein 1 (TNFAIP1), human homologs of genes found to be robustly induced in the transgenic C. elegans model. Both CRYAB and TNFAIP1 show increased transcript levels in AD brains, supporting the validity of this approach. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Alzheimer; Transgenic; Caenorhabditis elegans; Microarray; Quantitative RT-PCR; Inducible expression; A42 1. Introduction Many lines of evidence support the view that the beta amyloid peptide (A) is centrally involved in Alzheimer’s disease pathogenesis (reviewed in [30]). However, the mech- anism(s) of A toxicity remain unresolved. Establishing the cellular response to A exposure should provide clues to the mechanism of A toxicity. This cellular response is likely to be reflected in gene expression changes, and thus the pow- erful technique of DNA microarray hybridization has been applied to post-mortem AD brain tissue [10,13,22,26] or post-mortem microglia exposed to synthetic A 1–42 [33]. Interpretation of these studies is complicated by numerous factors, including the genetic and environmental heterogene- ity of the samples, use of post-mortem tissue, and the likely long time lag between the onset of pathology and when tis- sue was recovered for the analysis. We have developed a simplified model based on inducible transgenic expression of human A 1–42 peptide in the nematode Caenorhabdi- tis elegans that has allowed us to examine A-dependent ∗ Corresponding author. Tel.: +1-303-735-5112; fax: +1-303-492-8063. E-mail address: linkc@colorado.edu (C.D. Link). gene expression independent of the confounding factors de- scribed above. We have previously demonstrated that in transgenic C. el- egans animals expression of a human A minigene driven by a constitutive muscle-specific promoter leads to the accu- mulation of intracellular A-immunoreactive deposits and intracellular amyloid, leading to a progressive paralysis phe- notype [19,21]. Intracellular A dimers have been detected in primary human neurons and in neuronal cell lines [34], and intraneuronal A42 has also been demonstrated in hu- man brain [11]. Recent studies have also implicated intracel- lular A 1–42 in apoptosis of transfected rat cortical neurons [15]. Here we describe a new transgenic model that allows temperature-dependent induction of A, which results in a rapid and dramatic paralysis of induced transgenic animals. This inducibility has allowed us to look at gene expression changes occurring both before and after gross pathological changes (paralysis). Given that we are using a simplified in- vertebrate model, the goal of these studies was not to gen- erate a detailed global description of the gene expression response to intracellular A accumulation, but instead to identify candidate specific A-responsive genes that could subsequently be investigated in other AD models and human 0197-4580/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. doi:10.1016/S0197-4580(02)00224-5