Autotrophic adaptive laboratory evolution of the acetogenClostridium autoethanogenumdelivers the gas-fermenting strain LAbrini with superior growth, products, and robustness

Abstract

ABSTRACTMicrobes able to convert gaseous one-carbon (C1) waste feedstocks are of growing importance in transitioning to the biosustainable production of renewable chemicals and fuels. Acetogens are particularly interesting biocatalysts since gas fermentation usingClostridium autoethanogenumhas already been commercialised. Most non-commercial acetogen strains, however, need complex nutrients, display slow growth, and are not sufficiently robust for routine bioreactor fermentations. In this work, we used three adaptive laboratory evolution (ALE) strategies to evolve the wild-type model-acetogenC. autoethanogenumto grow faster, without complex nutrients and to be robust in operation of continuous bioreactor cultures. Seven evolved strains with improved phenotypes were isolated on minimal medium with one strain, named “LAbrini” (LT1), exhibiting superior performance in terms of the maximum specific growth rate, product profile, and robustness in continuous cultures. Differing performance of the strains between bottle batch and continuous cultures shows the importance of testing novel strains in industrially relevant continuous fermentation conditions. Interestingly, a very distinct transcriptome profile linked to a potential CO toxicity phenotype was observed in bioreactor cultures for one evolved strain. Whole-genome sequencing of the seven evolved strains identified 25 mutations with two genomic regions under stronger evolutionary pressure. Our analysis also suggests that the genotypic changes that are potentially responsible for the improved phenotypes may serve as useful candidates for metabolic engineering of cell factories. This work provides the robustC. autoethanogenumstrain LAbrini to the academic community to speed up phenotyping and genetic engineering, improve quantitative characterisation of acetogen metabolism, and facilitate the generation of high-quality steady-state datasets.