Optimal Cellular Phenotypic Adaptation in Fluctuating Environments
Institutional Repository Document
Overview
Research
Identity
Other
View All
Overview
abstract
AbstractPhenotypic adaptation is a core design feature of bacterial populations and multicellular systems navigating highly variable environments. Recent empirical data implicates the role of memory-driven decision-making in cellular systems navigating uncertain future nutrient environments based on prior experience, wherein a distinct growth phenotype emerges in fluctuating conditions. We develop a simple stochastic mathematical model to describe the cellular decision-making required for systems to optimally navigate such uncertainty. We demonstrate that adaptive populations capable of sensing their environment and estimating the nutrient landscape more efficiently traverse changing environments. We find during environmental transitions that larger memory capacities strike a trade-off between inertia of past environmental memory and higher resolution for estimating the optimal phenotype whenever the underlying landscape is close to a critical break-even point. Moreover, systems that tune their memory capacity avoid growth penalties resulting from maladaptive phenotypes following changes to the metabolic landscape. Our model predicts that the nutrient availability of adaptive cells is universally reduced in fluctuating nutritional environments relative to those in constant ones, which recapitulates empirical observations in bacterial systems. Our findings demonstrate that this deviation is a consequence of environmental mis-estimation together with bet-hedging in uncertain adaptive landscapes, and suggests that this deviation is fully determined by cellular memory capacity and the proximity of the environmental landscape to the systems critical break-even environment. We anticipate that our mathematical framework will be more broadly useful for studying memory-driven cellular decision-making in biological contexts where there is a trade-off for cells selecting from multiple phenotypic states. Such a tool can be used for predicting the response of complex systems to environmental alterations and for testing therapeutically relevant policies.
