Large-scale microarray studies of gene expression in multiple regions of the brain in schizophrenia and Alzheimer's disease

It took about 13 years to complete the working draft map of the human genome. More than 1 million expression sequence tags have been catalogued, corresponding to an estimated 31,000 protein-encoding human genes (Baltimore, 2001). However, the function, regional expression, and regulation of most of these genes have yet to be determined. The assignment of the molecular and cellular functions of these genes will likely be elaborated in the continuation of the Human Genome Project. Two major approaches to determine functions of gene products at the nucleic acid level are used. The structural approach is based on ascertainment of homologies of sequence-specific motifs encoding structural domains, thus providing clues to gene function. The second approach for exploring the functions of gene products is by determining the expression patterns of the genes of interest and attributing functions based on known expression patterns. Although numerous studies have been performed to establish expression patterns for individual genes or groups of genes, analysis of available expression data has not yet reached the point to permit assignment of specific gene expression signatures to specific functions in most cases. In addition to classic molecular biology methods for establishing expression profiles from single gene to several more genes, such as Northern blots, RNAse protection assays, and so on, a major advance in the past decade has been the development of high-throughput technologies that permit simultaneous expression profiling of thousand or tens of thousands of genes. These methods include differential display, serial analysis of gene expression (SAGE), and array-based technologies, consisting of complementary DNA (cDNA) and oligonucleotide microarrays. The last two methods allow simultaneous analysis of the expression of thousands of transcripts providing static information about gene expression (tissue, cell, and time point) and dynamic information (relationship of the expression pattern of a single gene, or a set of genes, to the expression patterns of other genes). These technologies provide powerful tools for neuroscientists that offer the potential to elevate the molecular genetic approach to the systems level for studying the human brain in normal and pathological states. A quick overview of publications related to the use of microarrays in neuroscience shows that from the initial array-related publication year 1999, the number of published reports increased 10-fold by 2001 and 20-fold by 2003 (Fig. 1). Knowledge of regional gene expression patterns of the human brain is essential to understanding the molecular biology of normal and pathological brain function and physiology. Differences in gene expression can reflect morphological and phenotypic differences while indicating cellular responses to signaling elicited by environmental stimuli. Unlike the genome, transcription profiles are highly dynamic, changing relatively rapidly in response to internal and external environmental stimuli or even to pre-programmed events such as cell cycle programs and apoptosis. These alterations of gene expression patterns can help identify "candidate" genes associated with pathophysiological processes in the human brain and can provide an exciting new avenue for research on psychiatric and neurodegenerative disorders. In this chapter, we assess the status of microarray technology and data-mining strategies as they relate to the analysis of postmortem brain with a focus on schizophrenia (SZ), Alzheimer's disease (AD), and tissue- and donor-quality requirements. © 2005 Elsevier Inc. All rights reserved.