Long-term phenotypic evolution of bacteria

• Published in: Nature (London) Vol. 517; no. 7534; pp. 369 - 372
• Main Authors: Plata, Germán, Henry, Christopher S., Vitkup, Dennis
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 15-01-2015; Nature Publishing Group
• Subjects: 45; 631/181/2474; Analysis; Bacteria; Bacteria - classification; Bacteria - genetics; Bacteria - growth & development; Bacteria - metabolism; Biological Evolution; Carbon; Evolution & development; Evolutionary biology; Gene Deletion; Genes; Genetic aspects; Genetic research; Genome, Bacterial - genetics; Genomes; Genotype & phenotype; Growth conditions; Humanities and Social Sciences; letter; Metabolism; multidisciplinary; Natural history; Natural selection; Phenotype; Physiological aspects; Science; Selection, Genetic; Studies; Taxonomy; United States
• ISSN: 0028-0836; 1476-4687
• DOI: 10.1038/nature13827

A comparative analysis of bacterial growth and genetic phenotypes using hundreds of genome-scale metabolic models reveals a two-stage evolutionary process that consists of a rapid initial phenotypic diversification followed by a slow long-term divergence.
The long game of bacterial evolution
Various studies have followed the evolution of animal phenotypes on a scale of millions of years, with the evolution of beaks in Darwin's finches as the classic example. This study goes further or at least longer, with a comparative analysis of bacterial growth that spans billions of years of species divergence. The metabolic diversity of bacteria makes it possible to investigate the phenotypic evolution essentially across the whole bacterial world. Dennis Vitkup and colleagues obtained 322 genome-scale metabolic models of species using a recently developed protocol for automatic network reconstruction and observe a two-stage evolutionary process that consists of a rapid initial phenotypic diversification followed by a slow long-term divergence. They confirm this pattern experimentally with 40 diverse bacterial species under 60 different growth conditions and interpret their results in terms of a 'phenotypic clock' akin to the molecular clock in protein evolution.
For many decades comparative analyses of protein sequences and structures have been used to investigate fundamental principles of molecular evolution
1
,
2
. In contrast, relatively little is known about the long-term evolution of species’ phenotypic and genetic properties. This represents an important gap in our understanding of evolution, as exactly these proprieties play key roles in natural selection and adaptation to diverse environments. Here we perform a comparative analysis of bacterial growth and gene deletion phenotypes using hundreds of genome-scale metabolic models. Overall, bacterial phenotypic evolution can be described by a two-stage process with a rapid initial phenotypic diversification followed by a slow long-term exponential divergence. The observed average divergence trend, with approximately similar fractions of phenotypic properties changing per unit time, continues for billions of years. We experimentally confirm the predicted divergence trend using the phenotypic profiles of 40 diverse bacterial species across more than 60 growth conditions. Our analysis suggests that, at long evolutionary distances, gene essentiality is significantly more conserved than the ability to utilize different nutrients, while synthetic lethality is significantly less conserved. We also find that although a rapid phenotypic evolution is sometimes observed within the same species, a transition from high to low phenotypic similarity occurs primarily at the genus level.