| Popis: |
Evolution via template-based replication 1–4 was probably preceded by a more rudimentary form of evolution based on networks of autocatalytic reactions 5–8 . However, reaction networks possessing the Darwinian properties of variation (in composition, the catalysts present and their relative amounts), differential reproduction (accumulation of products), and heredity (persistence of composition), have so far not been identified. Here we show that networks of catalytic RNAs can possess certain properties of Darwinian systems, that these properties are controlled by network topology, and characterize important trade-offs between them. By combining barcoded sequencing with droplet microfluidics, we screened ̴ 20,000 reactions corresponding to more than 1,800 distinct networks of ribozymes that catalyse their own formation from RNA fragments. We found that more highly connected networks tend to reproduce more quickly (accumulate more ribozymes) and be more robust to perturbations, indicating a trade-off between variation and reproduction. Variations are strongest when adding upstream ribozymes with novel reaction specificities (innovations) which target weakly connected networks. In turn, innovations increase connectivity, thus buffer against further perturbations, highlighting a second trade-off between robustness and variation. Despite this, phases of compositional robustness can alternate with sudden variations of the composition across trajectories of network growth by accretion of novel species. Finally, heredity is found to be limited by the activity of self-assembled, non-covalent ribozymes. Our findings show that connectivity in reaction networks not only determines the probability of autocatalytic sets in chemistries 8 , but also their potential for evolution. They are directly relevant to scenarios where early evolution is driven by environmental heterogeneity 9,10 , as it depends both on robustness (persistence being necessary for selection to act) and susceptibility to changes (in order to explore novel states). More broadly, they provide guiding rules for chemistries capable of Darwinian evolution and constrain scenarios of the origins of life. |
