Metabolomics-based systematic prediction of yeast lifespan and its application for semi-rational screening of ageing-related mutants.

Abstract

Metabolomics - the comprehensive analysis of metabolites - was recently used to classify yeast mutants with no overt phenotype using raw data as metabolic fingerprints or footprints. In this study, we demonstrate the estimation of a complicated phenotype, longevity, and semi-rational screening for relevant mutants using metabolic profiles as strain-specific fingerprints. The fingerprints used in our experiments are profiled data consisting of individually identified and quantified metabolites rather than raw spectrum data. We chose yeast replicative lifespan as a model phenotype. Several yeast mutants that affect lifespan were selected for analysis, and they were subjected to metabolic profiling using mass spectrometry. Fingerprinting based on the profiles revealed a correlation between lifespan and metabolic profile. Amino acids and nucleotide derivatives were the main contributors to this correlation. Furthermore, we established a multivariate model to predict lifespan from a metabolic profile. The model facilitated the identification of putative longevity mutants. This work represents a novel approach to evaluate and screen complicated and quantitative phenotype by means of metabolomics.