Yeast Yeast 2008; 25: 835–847. Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/yea.1638 Research Article Identification of common traits in improved xylose-growing Saccharomyces cerevisiae for inverse metabolic engineering Oskar Bengtsson 1 , Marie Jeppsson 1 , Marco Sonderegger 2# , Nadia Skorupa Parachin 1 , Uwe Sauer 2 , B¨ arbel Hahn-H¨ agerdal 1 and Marie-F. Gorwa-Grauslund 1 * 1 Department of Applied Microbiology, Lund University, PO Box 124, 221 00 Lund, Sweden 2 Institute for Molecular Systems Biology, ETH Z¨ urich, CH-8093 Z¨ urich, Switzerland *Correspondence to: Marie-F. Gorwa-Grauslund, Department of Applied Microbiology, Lund University, PO Box 124, 221 00 Lund, Sweden. E-mail: Marie-Francoise.Gorwa@ tmb.lth.se # Present address: Hoffmann-La Roche Ltd, Biotech Manufacturing Basel, Pilot Plant B66 (PTBB-E)-Downstream Processing, Building 66/202A, CH-4070 Basel, Switzerland. Received: 29 January 2008 Accepted: 26 September 2008 Abstract Four recombinant Saccharomyces cerevisiae strains with enhanced xylose growth (TMB3400, C1, C5 and BH42) were compared with two control strains (TMB3399, TMB3001) through genome-wide transcription analysis in order to identify novel targets for inverse metabolic engineering. A subset of 13 genes with changed expression levels in all improved strains was selected for further analysis. Thirteen validation strains and two reference strains were constructed to investigate the effect of overexpressing or deleting these genes in xylose-utilizing S. cerevisiae. Improved aerobic growth rates on xylose were observed in five cases. The strains overexpressing SOL3 and TAL1 grew 19% and 24% faster than their reference strain, and the strains carrying deletions of YLR042C, MNI1 or RPA49 grew 173%, 62% and 90% faster than their reference strain. Copyright  2008 John Wiley & Sons, Ltd. Keywords: inverse metabolic engineering; transcriptome; xylose; Saccharomyces cerevisiae; SOL3; TAL1 ; YLR042C; MNI1; RPA49 Introduction The interest for renewable energy sources is increasing due to the uncertainty of future oil availability and increasing greenhouse gas emis- sions. Bio-ethanol produced from lignocellulosic substances is expected to become one of the dom- inating renewable energy sources for the transport sector (Hahn-H¨ agerdal et al., 2006). The microor- ganism most commonly used in industrial ethanol production is the yeast Saccharomyces cerevisiae. Native strains of this species are unable to fer- ment the pentose fractions of lignocellulosic feed- stocks. The most abundant pentose sugar is xylose, xylan constituting up to 23% dry weight of ligno- cellulosic materials (Hayn et al., 1993; Hamelinck et al., 2005). Substantial research efforts have been made to develop yeast strains for commercial ethanolic fermentation of xylose (Jeffries, 2006; Hahn-H¨ agerdal et al., 2007). Heterologous expression of xylose pathways in S. cerevisiae first led to slow xylose utilization (K¨ otter and Ciriacy, 1993; Walfridsson et al., 1996; Kuyper et al., 2003). Strains with improved xylose consumption rates have since been generated by changing the genome of recombinant S. cerevisiae by controlled approaches. Xylose fermentation was improved by overexpression of the endogenous XKS1 gene, encoding xylulokinase (XK) (Ho et al., 1998; Eliasson et al., 2000; Toivari et al., 2001) and was further enhanced when the four genes of the non-oxidative part of the pentose phosphate pathway (PPP) were overexpressed; transaldolase (TAL1 ), transketolase (TKL1 ), ribose 5-phosphate Copyright  2008 John Wiley & Sons, Ltd.