Why life is hot

TL;DR

The paper investigates why life dissipates heat and shows that coupling reaction networks to strongly out-of-equilibrium reservoirs provides a versatile mechanism to optimize multiple fitness functions. By theory and numerical modeling, it demonstrates that adding cycles to biochemical networks creates non-equilibrium steady-state currents ($J^{s}$) and enables tuning of concentrations and fluxes to enhance robustness, precision, or responsiveness. The work quantifies heat production in actin treadmilling and kinetic proofreading, shows that the dominant currency hydrolysis yields large $\Delta G$ values (approximately $50$--$70$ kJ/mol) and saturation of both cycle current and error reduction, and argues that natural selection drives systems to operate near this limit. Together, the results imply substantial heat dissipation is an inherent consequence of maintaining a flexible, saturating mechanism for diverse cellular functions.

Abstract

The process of evolution by natural selection leads to fitness-maximising phenotypes. On the level of cellular chemical reaction networks, maximising fitness can mean optimising a variety of fitness functions such as robustness, precision, or sensitivity to external stimuli. Using theory and numerics, we show that these diverse goals can be achieved by a versatile, generic mechanism: coupling chemical reaction networks to reservoirs that are strongly out of equilibrium. Moreover, we demonstrate that the degree of optimality achievable by this mechanism saturates, and that nature appears to operate near saturation. We find that the amount of heat generated by this mechanism constitutes a significant fraction of the total heat produced by living organisms, so that 'life is hot' largely because of the need for a versatile mechanism to optimise a variety of fitness functions.

Figures (3)

• Figure 1: (a): Tree-like chemical reaction network. (b) Chemical reaction network with cycles. Dashed ellipses denote reservoirs.
• Figure 2: Reaction network with a heat bath and three chemiostats, which control the concentrations of the species $A$, $D$ and $P$.
• Figure 3: Kinetic proof reading model: summed cycle flux (red) and error rate (blue), expressed as a function of ${K/{\cal{Q}}}$ for the GTP hydrolysis equilibrium, as the GTP : GDP ratio is varied, constraining $\text{[GTP]}+\text{[GDP]}$ and $\text{[Pi]}$ to $1\,\text{mM}$. (Square brackets indicate concentrations.) Elongation factor complex concentrations are $0.1\,\text{mM}$ for both 'C' (correct) and 'D' (wrong) amino acid types. A copy number of $5000$ ribosomes is assumed in a cell volume $2µm^3$ (concentration $\simeq4.2µ\molar$). Detachment rates of the complexes are enhanced by a factor $f=100$ for the 'wrong' amino acid type. For the saturated total cycle flux $J^s/[\ce{R}]_{\text{tot}}=k^s\simeq10\,\mathrm{s}^{-1}$, similar to the range quoted in the main text above \ref{['eq:ecoliq']}. For more details see Appendix.