Biosensors and Bioelectronics 20 (2004) 956–966 Cohort analysis of a single nucleotide polymorphism on DNA chips Susanne Schwonbeck a , Andrea Krause-Griep a , Nenad Gajovic-Eichelmann a , Eva Ehrentreich-Förster a , Walter Meinl b , Hansrüdi Glatt b , Frank F. Bier a,∗ a Department of Molecular Bioanalytics and Bioelectronics, Fraunhofer Institute for Biomedical Engineering, Arthur-Scheunert-Allee 114-116, D-14558 Nuthetal, Germany b Department of Nutritional Toxicology, German Institute of Human Nutrition, Nuthetal, Germany Available online 6 August 2004 Abstract A method has been developed to determine SNPs on DNA chips by applying a flow-through bioscanner. As a practical application we demonstrated the fast and simple SNP analysis of 24 genotypes in an array of 96 spots with a single hybridisation and dissociation experiment. The main advantage of this methodical concept is the parallel and fast analysis without any need of enzymatic digestion. Additionally, the DNA chip format used is appropriate for parallel analysis up to 400 spots. The polymorphism in the gene of the human phenol sulfotransferase SULT1A1 was studied as a model SNP. Biotinylated PCR products containing the SNP (The SNP summary web site: http://www.ncbi.nlm.nih.gov/SNP/snp summary.cgi) (mutant) and those containing no mutation (wild-type) were brought onto the chips coated with NeutrAvidin using non-contact spotting. This was followed by an analysis which was carried out in a flow-through biochip scanner while constantly rinsing with buffer. After removing the non-biotinylated strand a fluorescent probe was hybridised, which is complementary to the wild-type sequence. If this probe binds to a mutant sequence, then one single base is not fully matching. Thereby, the mismatched hybrid (mutant) is less stable than the full-matched hybrid (wild-type). The final step after hybridisation on the chip involves rinsing with a buffer to start dissociation of the fluorescent probe from the immobilised DNA strand. The online measurement of the fluorescence intensity by the biochip scanner provides the possibility to follow the kinetics of the hybridisation and dissociation processes. According to the different stability of the full-match and the mismatch, either visual discrimination or kinetic analysis is possible to distinguish SNP-containing sequence from the wild-type sequence. © 2004 Elsevier B.V. All rights reserved. Keywords: Cohort analysis; Single nucleotide polymorphism; DNA chips 1. Introduction The study of possible associations between symptoms of diseases and of specific genotypes might provide new facts for a better understanding of different aspects of nutrition, physiology or toxicology. In this respect, besides the de- termination of unknown single nucleotide polymorphisms (SNPs), it is also important to analyse known SNPs in a high number of individuals. Single nucleotide polymorphisms or SNPs occur with a frequency of more than 1% in a given population and are thought to contribute substantially to the phenotype varia- tion among individuals. It has been estimated that there is approximately one SNP per 1000 base pairs (Wang et al., ∗ Corresponding author. Tel.: +49-33200-88378; fax: +49-33200-88452. E-mail address: frank.bier@ibmt.fhg.de (F.F. Bier). 1998). More than 6 million SNPs have been submitted to the NCBI database dbSNP (see the SNP summary web site). Three groups of SNPs have been classified: coding-region, perigenic and intergenic SNPs (Nebert, 1999). Depending on where an SNP occurs, it might lead to different con- sequences at the phenotypic level. Intergenic SNPs occur between genes throughout the genome and have been used for analysis of ethnic differences and tribal migrations (Kaessmann et al., 1999) or for identification of cryptic species (Belfiore et al., 2003). SNPs in the coding region of a gene may alter the function or structure of the encoded protein due to changes in the amino acid sequences. There- fore, cSNPs can be used as a genetic marker for mapping the disease gene loci and for candidate gene association studies. Furthermore, cSNPs are highly relevant in phar- macogenetics, for instance, in the magnitude of anti-cancer drug response. Although anti-cancer drugs are effective in cancer treatment, therapy is often limited to a cumulative 0956-5663/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2004.06.012