Bacterial fitness is determined both by how fast cells grow in nutrient-rich environments and by how well they survive when nutrients are depleted. However, these behaviors are not independent, since the molecular composition of non-growing cells is affected by their prior growth history. For instance, recent work observed that the death rates of Escherichia coli cultures that rapidly entered carbon starvation depend on their prior growth rates, with faster growth leading to exponentially faster death. On the other hand, it is well known that cells adapt their molecular composition as they slow down growth and enter stationary phase, which is generally believed to improve their chance of survival. Hence, the question arises to what extent this adaptation process reduces the subsequent death rate. And how does the duration of the time window during which cells are allowed to adapt determine the reduction in death rate, and thus the fitness benefit of adaptation? Here, we study these quantitative questions by probing the adaptation of E. coli during gradual transitions from exponential growth to carbon starvation. We monitor such transitions in cultures with different initial growth conditions and measure the resulting rates of cell death after the transition. Our experiments demonstrate that cells with the opportunity to adapt their proteome composition before entering a state of starvation exhibit lower death rates compared to those that cannot, across various substrate conditions. The quantitative data is consistent with a theoretical model built on the assumption that before starvation, cells up-regulate a specific sector of the proteome, the effect of which is to decrease the death rate in energy-limiting conditions. This work highlights the influence of the non-genetic memory of a cell, specifically in the form of inherited proteome composition, on bacterial fitness. Our results emphasize that a comprehensive understanding of bacterial fitness requires quantitative characterization of bacterial physiology in all phases of their life cycle, including growth, stationary phase, and death, as well as the transitions between them. AUTHOR SUMMARY Bacteria inhabit dynamic environments and are frequently challenged by scarcity of nutrients. A recent study uncovered a curious link – faster bacterial growth leads to more rapid death when resources run out. We find that bacteria that gradually enter starvation exhibit significantly enhanced survival compared to those that do not have the chance to adapt. We interpret the observed quantitative behavior with the help of a theoretical model, which shows that our data is not compatible with a passive adaptation process, which would rely only on the general remodeling of the cellular proteome that is associated with growth transitions. Instead, our data are consistent with an active adaptation via up-regulation of genes that enhance survival during starvation. These results provide a novel perspective on bacterial survival strategies and underscore the importance of quantitatively investigating all phases of bacterial life cycles. ### Competing Interest Statement The authors have declared no competing interest.

Adaptive stress resistance in microbes is mostly attributed to the expression of stress response genes, including heat-shock proteins. Here, we report a response of E. coli to heat stress caused by degradation of an enzyme in the methionine biosynthesis pathway (MetA). While MetA degradation can inhibit growth, which by itself is detrimental for fitness, we show that it directly benefits survival at temperatures exceeding 50°C, increasing survival chances by more than 1,000-fold. Using both experiments and mathematical modeling, we show quantitatively how protein expression, degradation rates, and environmental stressors cause long-term growth inhibition in otherwise habitable conditions. Because growth inhibition can be abolished with simple mutations, namely point mutations of MetA and protease knockouts, we interpret the breakdown of methionine synthesis as a system that has evolved to halt growth at high temperatures, analogous to “thermal fuses” in engineering that shut off electricity to prevent overheating.