DNA Microarrays

The need for rapid, high-quality, comprehensive gene-related data for a variety of purposes has stimulated the development of DNA microarrays or biochips. These provide a radically different approach to largescale characterisation of genes and gene expression. DNA microarrays are a departure from electrophoresis-based approaches and currently represent the extreme end of miniaturisation. A microarray is an ordered
array of many thousands of DNA oligonucleotides, either singlestranded oligonucleotides or double-stranded cDNAs, attached (either by direct printing or in situ synthesis of short oligonucleotides) to a glass or silicon ‘chip’ that is about the size of a microscope coverslip. The bound DNA can then be hybridised with test DNA or RNA which has been labelled with one or two fluorescent dyes depending on the experimental design. Following an incubation step, the unhybridised material is washed away and the result is recorded using a confocal laser scanner. The data are collected and displayed automatically using dedicated computer programs.
Initially microarrays were designed to measure mRNA transcripts from thousands of genes in a single experiment. This enabled the physiological state of cells and overall gene expression pattern to be correlated. For example, transcriptional profiles have been obtained for many types of human cancer and the accumulated data promise to lead the way to a better understanding of neoplasia and new therapeutic targets.
DNA microarrays are not limited to gene expression studies. The first genotyping biochips were devised to identify key mutations in highly variable medically important genes and genomes such as the tumour suppressor gene TP53 (OMIM 191170) and the human immunodeficiency virus (HIV). In both cases, successful genotyping of clinical material is achieved using the microarray approach. More recently, genotyping chips capable of assigning SNP alleles on a ‘whole genome’ basis have been devised, which have applications in a variety of post-genomic projects such as the HapMap initiative to map all human variations in different population groups. Three commercially available microarrays are available that make it possible to analyse simultaneously many thousands of SNPs in an individual’s DNA. Two microarrays from Affymetrix, with roughly 100 000 and 500 000 SNPs, respectively, and a microarray from Illumina, with more than 300 000 SNPs, enable a large sample of the genetic variation of an individual to be assessed in a single experiment. Such tools should improve the opportunities for correlating common diseases with genetic differences in statistically high-powered association studies. Commercially available microarrays have now been developed to monitor SNP profiles in economically or scientifically important non-humans such as cattle and mice. However, the current estimate is that there are more than 10 million SNP sites in the human genome and considerable future developments will probably be required to give the required coverage of SNPs in all human populations sufficient for genome-wide association studies.