Adapting in larger numbers can increase the vulnerability ofEscherichia colipopulations to environmental changes

Abstract

Larger populations generally adapt faster to their existing environment. However, it is unknown if the population size experienced during evolution influences the ability to face sudden environmental changes. To investigate this issue, we subjected replicate Escherichia coli populations of different sizes to experimental evolution in an environment containing a cocktail of three antibiotics. In this environment, the ability to actively efflux molecules outside the cell is expected to be a major fitness‐affecting trait. We found that all the populations eventually reached similar fitness in the antibiotic cocktail despite adapting at different speeds, with the larger populations adapting faster. Surprisingly, although efflux activity (EA) enhanced in the smaller populations, it decayed in the larger ones. The evolution of EA was largely shaped by pleiotropic responses to selection and not by drift. This demonstrates that quantitative differences in population size can lead to qualitative differences (decay/enhancement) in the fate of a character during adaptation to identical environments. Furthermore, the larger populations showed inferior fitness upon sudden exposure to several alternative stressful environments. These observations provide a novel link between population size and vulnerability to environmental changes. Counterintuitively, adapting in larger numbers can render bacterial populations more vulnerable to abrupt environmental changes.

An interplay of population size and environmental heterogeneity explains why fitness costs are rare

“…Since these populations faced only one carbon source throughout the experiment, their evolution was blind to fitness changes in other carbon sources. The pleiotropic disadvantages of beneficial mutations are generally expected to be correlated with their direct effects (Lande 1983;Orr & Coyne 1992;Otto 2004;Chavhan et al 2019a). Since the larger asexual populations adapt primarily via beneficial mutations with relatively greater direct effect sizes (Desai et al 2007;Desai & Fisher 2007b;Sniegowski & Gerrish 2010;Chavhan et al 2019b), adapting to homogeneous environments in larger numbers should lead to heavier costs of adaptation, as observed in our study (Fig.…”

“…This opens up the possibility that factors other than environmental heterogeneity may be important in shaping the emergence of fitness costs. One such factor is populations size, which has been shown to be important in shaping the correlated changes in populations' fitness in alternative environments (Chavhan et al 2019a. For example, a recent study showed that larger populations evolving in a homogeneous environment containing a single carbon source suffer greater fitness costs in alternative environments .…”

Larger bacterial populations evolve heavier fitness trade-offs and undergo greater ecological specialization

“…This relationship between population size and ecological specialization has many important implications. Foremost, owing to their higher extent of specialization, larger populations can become vulnerable to sudden changes in the environment, as predicted by a recent study (Chavhan et al 2019b). Interestingly, if the environment abruptly shifts between two states that show fitness trade-offs with each other, then populations with a history of evolution at larger numbers would be at a greater disadvantage than historically smaller populations.…”

Larger bacterial populations evolve heavier fitness trade-offs and undergo greater ecological specialization

“…Using a randomized complete block design (RCBD), we conducted the fitness measurements over six different days, assaying one replicate population of each type in both the environments on a given day (Milliken and Johnson, 2009). We estimated fitness as the maximum growth rate (R) (Kassen, 2014; Ketola and Saarinen, 2015; Vogwill et al , 2016), which was computed as the maximum slope of the growth curve over a moving window of ten readings (Leiby and Marx, 2014; Karve et al , 2015, 2016, 2018; Chavhan et al , 2019a; Chavhan et al , 2019b).…”