Appl Microbiol Biotechnol (2006) 70: 237–246 DOI 10.1007/s00253-005-0070-2 APPLIED MICROBIAL AND CELL PHYSIOLOGY Sung A. Schoondermark-Stolk . Michael Jansen . Janine H. Veurink . Arie J. Verkleij . C. Theo Verrips . Gert-Jan W. Euverink . Johannes Boonstra . Lubbert Dijkhuizen Rapid identification of target genes for 3-methyl-1-butanol production in Saccharomyces cerevisiae Received: 17 April 2005 / Revised: 16 June 2005 / Accepted: 16 June 2005 / Published online: 23 July 2005 # Springer-Verlag 2005 Abstract Extracellular conditions determine the taste of fermented foods by affecting metabolite formation by the micro-organisms involved. To identify targets for improve- ment of metabolite formation in food fermentation pro- cesses, automated high-throughput screening and cDNA microarray approaches were applied. Saccharomyces cer- evisiae was cultivated in 96-well microtiter plates, and the effects of salt concentration and pH on the growth and synthesis of the fusel alcohol-flavoured substance, 3-meth- yl-1-butanol, was evaluated. Optimal fermentation con- ditions for 3-methyl-1-butanol concentration were found at pH 3.0 and 0% NaCl. To identify genes encoding enzymes with major influence on product formation, a genome-wide gene expression analysis was carried out with S. cerevisiae cells grown at pH 3.0 (optimal for 3-methyl-1-butanol for- mation) and pH 5.0 (yeast cultivated under standard condi- tions). A subset of 747 genes was significantly induced or repressed when the pH was changed from pH 5.0 to 3.0. Expression of seven genes related to the 3-methyl-1- butanol pathway, i.e. LAT1, PDX1, THI3, ALD4, ILV3, ILV5 and LEU4, strongly changed in response to this switch in pH of the growth medium. In addition, genes involved in NAD metabolism, i.e. BNA2, BNA3, BNA4 and BNA6, or those involved in the TCA cycle and glutamate metabolism, i.e. MEU1, CIT1, CIT2, KDG1 and KDG2, displayed significant changes in expression. The results indicate that this is a rapid and valuable approach for iden- tification of interesting target genes for improvement of yeast strains used in industrial processes. Introduction Food fermentation covers a wide range of microbial pro- cessing. Changes in extracellular conditions during the food fermentation have great impact on the final product quality. They influence the microbial metabolic fluxes, growth and the production of flavour compounds (Bisson 1999; Ribéreau-Gayon et al. 2000). Important issues in industrial food fermentation, i.e. quality, taste and process time, concern the efficiency of the process. Improvements in efficiency are usually obtained via strain selection or strain improvement (Volschenk et al. 2001). Consequently, up-to-date knowledge is required of the physiology, me- tabolism and genetic properties of the fermenting micro- organisms. High-throughput techniques for metabolite screening and gene expression analysis provide suitable tools for de- tailed analysis of microbial physiology. We applied a pow- erful combination of a microtiter plate (MTP) metabolite screening assay with high-density cDNA filters (micro- array gene expression analysis) to investigate the produc- tion of fusel alcohols. These alcohols are important flavour substances in fermented foods. Since a pH- and salt-step is included in many food fermentations such as soy sauce, we investigate the effects of changes in important extracellular parameters such as pH and salt concentration on the for- mation of fusel alcohols in Saccharomyces cerevisiae. The MTP metabolite screening assay allows rapid screen- ing of metabolite formation under different fermentation conditions. Cells are cultivated in 96 deep-well MTPs, each well representing an individual extracellular condition with minor variations in pH and salt concentration. Using a ded- icated facility with several MTP handling and pipetting robots (http://www.bioexplore.com), large numbers of con- ditions can be screened in a relatively short time. The com- pounds of interest produced, e.g. amino acids and alcohols, can be analysed using GC or HPLC-MS methods (Jansen et al. 2003). cDNA microarrays visualize the genome-wide expression profile of an organism. New sets of induced and repressed genes have been identified by the use of micro- arrays, in response to gene deletions or to various extra- S. A. Schoondermark-Stolk (*) . M. Jansen . J. H. Veurink . A. J. Verkleij . C. T. Verrips . G.-J. W. Euverink . J. Boonstra . L. Dijkhuizen Department of Molecular Cell Biology and the Institute of Biomembranes (IB), Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands e-mail: S.Schoondermark@nioo.knaw.nl