Phenotypic heterogeneity in lag reflects an evolutionarily stable bet-hedging strategy

A. George, Ned S. Wingreen, Gautam ReddybioRxiv 2025 · DOI

TL;DR

This paper explores why yeast populations split into "arresters" (non-growing) and "recoverers" (delayed growth) when faced with changing nutrient environments. They show this heterogeneity is an evolutionarily stable bet-hedging strategy, maximizing long-term survival by balancing growth in one environment with rapid adaptation to another.

What Problem Does This Paper Address?

Microbes, like yeast, constantly face fluctuating environments (e.g., changes in nutrient availability). To survive and thrive, they need strategies to cope with these unpredictable shifts. One observed phenomenon in yeast is that when moved to a less-preferred nutrient source, a single clonal population splits into two distinct groups: "recoverers" that eventually resume growth after a delay (lag phase), and "arresters" that remain dormant for extended periods. While this heterogeneity is consistently observed, it's unclear why this split occurs and if it's an optimal strategy for long-term population survival. The authors aim to understand the evolutionary advantage of this phenotypic diversification and the underlying physiological mechanisms.

Methods

The core of the study is a theoretical framework based on dynamic programming. This approach is used to identify the "optimal phenotype distribution" that maximizes a population's long-term growth rate in environments that switch stochastically between a "preferred" (P) and a "non-preferred" (NP) state. They model a trade-off where faster growth in the preferred environment comes at the cost of a longer lag phase in the non-preferred environment.

To explore the physiological basis of this trade-off, they developed a minimal metabolic model. This model simulates how enzyme allocation and metabolic fluxes influence growth and lag times when yeast switches between carbon sources.

Datasets

The authors analyzed existing experimental data from single-cell measurements of yeast populations undergoing glucose-to-maltose transitions. This data included:

  • Lag phase measurements for various wild yeast strains.
  • Lag phase measurements for evolved yeast isolates.
  • Lag phase measurements for populations with different pre-shift growth durations.

Key Insights and Ideas

  • Bet-hedging is key: The split into arresters and recoverers isn't random; it's an evolutionarily stable strategy that maximizes long-term population fitness by hedging bets across different environmental conditions.
  • Two phenotypes are often enough: Even when considering a continuous range of possible phenotypes, the optimal strategy typically involves at most two distinct types: a "specialist" arrester for the preferred environment and a "specialist" recoverer for the non-preferred environment, or a single "generalist" that balances both.
  • Physiological trade-offs drive heterogeneity: A minimal metabolic model reveals that the way cells allocate enzymes to process nutrients can inherently create a growth-lag trade-off, explaining why this phenotypic split arises.
  • An "evolutionary rheostat": The relationship between the fraction of recoverers and their lag time acts like a "rheostat," allowing populations to tune their strategy based on the statistics of environmental changes.

Key Results

  • The dynamic programming model predicts that optimal strategies for fluctuating environments consist of at most two phenotypes.
  • These optimal strategies are either a combination of an arrester (infinite lag in NP) and a recoverer, or a single generalist phenotype.
  • The minimal metabolic model reproduces a power-law relationship between pre-shift growth rate and post-shift lag duration, a key trade-off that favors phenotypic heterogeneity.
  • A predicted relationship between the fraction of recoverers and their lag duration was found to align well with experimental data from various yeast strains and conditions.

Limitations and Future Work

  • Model Simplifications: The theoretical framework makes simplifying assumptions, such as exponentially distributed dwell times in NP and a specific form for the growth-lag trade-off. More complex dwell-time distributions could potentially lead to optimal strategies with more than two phenotypes.
  • Experimental Validation: The authors suggest future experimental work to:
  • Test the predicted "re-entrant" generalist-specialist-generalist behavior by manipulating environmental switching rates (w and w') in microfluidic setups.
  • Investigate the coupling between growth and lag by genetically altering pre-shift enzyme allocation.
  • Test the "perfect arrester" hypothesis by measuring the survival of long-lag arresters.
  • Further explore the "evolutionary rheostat" concept by examining how this relationship functions across different yeast strains.

Summary updated

Abstract

Single-cell experiments in yeast reveal two distinct heritable phenotypes-'arresters' and 'recoverers'-when a clonal population experiences a negative shift in its growth environment. Recoverers exhibit a variable yet finite lag before resuming growth in the new environment, whereas arresters remain in a non-growing, arrested state until more favorable conditions return. Although the diversification of individual cells into arresters and recoverers is a robust phenomenon, it remains unclear whether this coexistence constitutes an evolutionarily stable strategy. Here, we demonstrate that a heterogeneous strategy composed of both arrester and recoverer phenotypes maximizes long-term population fitness across a broad spectrum of growth-lag trade-offs. Our analysis employs a dynamic programming framework to identify the fitness-maximizing distribution of phenotypes for populations that stochastically switch between preferred and non-preferred environments. We propose a minimal model incorporating metabolism, growth, and enzyme allocation to explain the physiological origin of a power-law growth-lag trade-off that favors phenotypic heterogeneity. The theory predicts a nontrivial relationship between the fraction of recoverers and their lag time, which aligns with existing data from wild yeast strains, evolved isolates, and variations in pre-shift growth conditions. This relationship suggests an evolutionary 'rheostat'-like mechanism that enables populations to rapidly adapt to changing environmental conditions.
BibTeX
@misc{george2025phenotypic,
  author = {A. George and Ned S. Wingreen and Gautam Reddy},
  title = {{Phenotypic heterogeneity in lag reflects an evolutionarily stable bet-hedging strategy}},
  year = {2025},
  doi = {10.1101/2025.11.02.686100},
  howpublished = {bioRxiv},
}
Figure