Quantitative physiology of bacterial survival under carbon starvation and temperature stress

A large number of the bacteria on Earth live for long periods in states of very low metabolic activity and little or no growth due to starvation and other environmental stresses. Within millions of years, bacteria have developed several strategies to adapt to many different environments, where they survive and evolve to optimize their fitness and to undergo rapid division cycles when conditions become favourable. However, many of these survival strategies are still a puzzle and relatively little is known about the mechanisms that underpin the dominant modes of bacterial existence. This is particularly alarming, as the growth-arrest phase has become crucial to understand the contribution of microorganisms to human physiology and predisposition to disease as well as microbial tolerance and resistance to antibiotics. The dearth of information is mainly due to the difficulties in defining, reproducing and measuring bacterial behaviours in growth-arrest states, which may often seem erratic and unpredictable, while cell physiology is similarly diverse and often specific to the particular environmental conditions. Thus, determining how molecular contributions affect survival is challenging. This explains why, in the last century, bacteria have been mainly studied during the exponential growth phase, which is, on the contrary, a well-defined and reproducible steady state of constant growth, gene expression and molecular compositions. As a result, an increasing combined use of experiments and predictive models focused on this phase has provided a deep understanding of bacterial physiology and gene regulation during growth. A similar quantitative approach that focuses on the growth-arrest phase is largely missing. In this thesis, we contribute to fill this gap by developing new quantitative approaches to investigate bacterial physiology in hostile environments where stresses, such as lack of nutrients and additional environmental perturbations, like temperature increase, force the cells to activate strategies of survival. To do so, we choose to work with the bacterium Escherichia coli (E. coli) that, among the estimated 10^12 microbial species living in our planet, is one of the most studied thanks to its hardiness, versatility and ease of handling. In Chapter 1, we provide an overview of the physiology of E. coli life cycle and of the main quantitative methods so far used to study it, especially focusing on its behaviour during the growth-arrest phase. In Chapter 2, we establish the missing quantitative approach to study E. coli physiology in the death phase. We show that in carbon starvation, an exponential decay of viability emerges as a collective phenomenon, with viable cells recycling nutrients from dead cells to maintain viability. The observed collective death rate is determined by the maintenance rate of viable cells and the amount of nutrients recovered from dead cells, the yield. Using this relation, we study the cost of a wasteful enzyme during starvation and the benefit of the stress response sigma factor RpoS. While the enzyme activity increases maintenance and thereby the death rate, RpoS improves biomass recycling, decreasing the death rate. Our approach thus enables quantitative analyses of how cellular components affect the survival of non-growing cells. In Chapter 3, we use the quantitative approach developed in the previous chapter to study how survival of E. coli in carbon starvation depends on the previous culture conditions. We show that environments that support only slow growth lead to longer survival in starvation because of a decrease of maintenance rate, meaning that slower growing cells need less energy to survive. Our results suggest a physiological trade-off between the ability to proliferate fast and the ability to survive long that could shed light on the long-standing question of why bacteria outside of laboratory environments are not optimized for fast growth. In Chapter 4, we study E. coli physiology under the combined stresses of carbon starvation and high temperatures, characterizing a thermal fuse that leads to a dormant and antibiotic persistent sub-population. This fuse is implemented by a thermally unstable enzyme, MetA, in the methionine synthesis pathway. The combination of a positive feed-back in the methionine system and a dual-use of methionine for protein synthesis and as a methyl-donor results in the bacterial population splitting into two distinct states at elevated temperatures, growing and dormant. We then reveal that these dormant bacteria not only survive antibiotic treatment, but also heat shocks, suggesting that the thermal fuse has originally evolved as a ''bet-hedging'' strategy to ensure survival in heat shocks. Our findings, summarized in Chapter 5, pave the way for the development of a new theoretical framework and experimental approach to understand bacterial physiology in the growth-arrest phase, by linking phenomenological modeling to molecular mechanisms.