Multiple Novel Traits without Immediate Benefits Originate in Bacteria Evolving on Single Antibiotics

Abstract

How new traits originate in evolution is a fundamental question of evolutionary biology. When such traits arise, they can either be immediately beneficial in their environment of origin, or they may become beneficial only in a future environment. Compared to immediately beneficial novel traits, novel traits without immediate benefits remain poorly studied. Here we use experimental evolution to study novel traits that are not immediately beneficial but that allow bacteria to survive in new environments. Specifically, we evolved multiple E. coli populations in five antibiotics with different mechanisms of action, and then determined their ability to grow in more than 200 environments that are different from the environment in which they evolved. Our populations evolved viability in multiple environments that contain not just clinically relevant antibiotics, but a broad range of antimicrobial molecules, such as surfactants, organic and inorganic salts, nucleotide analogues and pyridine derivatives. Genome sequencing of multiple evolved clones shows that pleiotropic mutations are important for the origin of these novel traits. Our experiments, which lasted fewer than 250 generations, demonstrate that evolution can readily create an enormous reservoir of latent traits in microbial populations. These traits can facilitate adaptive evolution in a changing world.

Environmental complexity is more important than mutation in driving the evolution of latent novel traits in E. coli

“…A Before conducting any evolution experiments, we determined the viability of both our ancestral wild-type and ancestral mutator strain in 236 different environments with Biolog phenotyping microarrays. The mutator strain was inviable in 58 environments, whereas the wild-type was inviable in 95 environments, as previously reported 5 . In 52 environments both the wild-type and mutator ancestors were inviable.…”

“…By this criterion, our wild-type strain was inviable in 95 of the 236 environments, as reported previously 5 . The mutator strain was inviable in 58 of the 236 environments (Fig.…”

“…Most experimental and theoretical work in evolutionary biology focuses on adaptive traits. However, in recent years experimental evidence has been mounting that such traits may be outnumbered by latent traits without immediate benefits, i.e., traits that are not immediately adaptive in their environment of origin, but that may become adaptive in the right kind of future environment [1][2][3][4][5] . Such potentially adaptive traits are important in evolution, because they can create new morphological structures or physiological abilities, and they can give rise to new ecological niches 6,7 .…”

“…For example, we recently evolved E. coli populations in environments harbouring single antibiotics and showed that these populations evolve the ability to survive in multiple novel environments that inhibit bacterial growth through mechanisms different from the antibiotic they evolved in ref. 5. We refer to the newly acquired (and non-adaptive) viability in any one such environment as a latent novel trait.…”

“…Specifically, we evolve a wild-type and a mutator strain with a twenty-two-fold higher mutation rate than the wild-type 18 . We evolve the mutator strain in the same five simple environments as in a previously reported experiment, in which we had evolved the wild-type strain in five environments containing one antibiotic each 5 . In addition, we evolve both the wild-type and the mutator strain in three complex environments with either three or five antibiotics.…”