Background The cellular proteins Pat1p, Lsm1p, and Dhh1p are required for the replication of some positive-strand viruses and therefore are potential targets for new antiviral drugs. To prioritize host targets for antiviral drug screening a comparative metabolome analysis in Saccharomyces cerevisiae reference strain BY4742 Matα his3Δ1 leu2Δ0 lys2Δ0 ura3Δ0 and deletion strains pat1Δ, lsm1Δ and dhh1Δ was performed./nResults GC/MS analysis permitted the quantification of 47 polar metabolites and the identification of 41 of them. Metabolites with significant variation between the strains were identified using partial least squares to latent structures discriminate analysis (PLS-DA). The analysis revealed least differences of pat1Δ to the reference strain as characterized by Euclidian distance of normalized peak areas. The growth rate and specific production rates of ethanol and glycerol were also most similar with this strain./nConclusion From these results we hypothesize that the human analog of yeast Pat1p is most likely the best drug target candidate.

This work was financially supported by the DFG (Deutsche Forschungsgemeinschaft) through the Center for Bioinformatics, Saarland University and grants from the Spanish Ministerio de Educación y Ciencia BFU2007-66933/BMC and the Saarland University (HOMFOR). IAR was supported by the Fundaçao para a Ciencia e Tecnologia (SFRH/BD/9630/2002) through the Gabba programme from the University of Porto, Portugal.