Predictive evolution of metabolic phenotypes using model-designed environments

Paula Jouhten, Dimitrios Konstantinidis, Filipa Pereira, Sergej Andrejev, Kristina Grkovska, Sandra Castillo, Payam Ghiachi, Gemma Beltran, Eivind Almaas, Albert Mas, Jonas Warringer, Ramon Gonzalez, Pilar Morales, Kiran R. Patil (Corresponding Author)

Research output: Contribution to journalArticleScientificpeer-review

7 Citations (Scopus)


Adaptive evolution under controlled laboratory conditions has been highly effective in selecting organisms with beneficial phenotypes such as stress tolerance. The evolution route is particularly attractive when the organisms are either difficult to engineer or the genetic basis of the phenotype is complex. However, many desired traits, like metabolite secretion, have been inaccessible to adaptive selection due to their trade-off with cell growth. Here, we utilize genome-scale metabolic models to design nutrient environments for selecting lineages with enhanced metabolite secretion. To overcome the growth-secretion trade-off, we identify environments wherein growth becomes correlated with a secondary trait termed tacking trait. The latter is selected to be coupled with the desired trait in the application environment where the trait manifestation is required. Thus, adaptive evolution in the model-designed selection environment and subsequent return to the application environment is predicted to enhance the desired trait. We experimentally validate this strategy by evolving Saccharomyces cerevisiae for increased secretion of aroma compounds, and confirm the predicted flux-rerouting using genomic, transcriptomic, and proteomic analyses. Overall, model-designed selection environments open new opportunities for predictive evolution.

Original languageEnglish
Pages (from-to)e10980
JournalMolecular Systems Biology
Issue number10
Publication statusPublished - 1 Oct 2022
MoE publication typeA1 Journal article-refereed


  • adaptive evolution
  • genome-scale metabolic model
  • predictive evolution
  • Saccharomyces cerevisiae
  • wine aroma


Dive into the research topics of 'Predictive evolution of metabolic phenotypes using model-designed environments'. Together they form a unique fingerprint.

Cite this