https://www.cell.com/cell-reports-physical-science/fulltext/S2666-3864(23)00152-2
Attractor dynamics drives self-reproduction in protobiological catalytic networks
Amit Kahana, Lior Segev, and Doron LancetSummary
The origin of life must have involved an unlikely transition from chaotic chemistry to self-reproducing supramolecular structures. Previous quantitative analyses of self-reproducing mutually catalytic networks made of simple molecules have led to increasing popularity of this pre-RNA scenario for life’s origin. Here, we investigate in detail the reproduction characteristic of the graded autocatalysis replication domain (GARD) computer-simulated physicochemically rigorous lipid-based model. This model displays compatibility with heterogeneous environments, addresses the network’s spatial demarcation, and portrays trans-generational compositional information transfer. However, we find that compositionally reproducing states are extremely rare, suggesting that random roaming would be a vastly inefficient path toward reproduction. Rewardingly, the present study shows that all self-reproducing states are also dynamic attractors of the catalytic network. This suggests a greatly enhanced propensity for the spontaneous emergence of reproduction and primal evolution, augmenting the likelihood of protolife appearance.
Another paper that argues a kind of evolution and informatin transfer is possible in chemical systems not based on genetic polymer sequences:
In the realm of network-governed supramolecular assemblies, the criterion for reproduction capacity is the faithful generation of assembly copies. Since such assemblies do not possess a sequence as do self-replicating polynucleotides, it is necessary to use a different criterion for assembly copying success. The criterion used for a classical CAS [Collectively Autocatalytic Set] network is catalytic closure and the appearance of autocatalytic and cross-catalytic cycles.31 This is defined as a state in which every molecular component within the network is formed in endogenous reactions catalyzed by at least one of the network members.13 Catalytically closed CAS networks are proposed to be capable of reproduction.13,17 Indeed, making sure that every component of the network is endogenously generated appears to guarantee the propagation of the gamut of molecule types within the assembly (i.e., compositional copying).
When assessing the success of polynucleotide self-replication, one uses a quantitative measure: the percentage sequence identity between parent molecule and its offspring. This serves as a measure of sequence information transmission along generations. One may ask how a non-covalent assembly may propagate information of some kind.3 As in previous studies, 16,23,32,49,53,54 we employ here the assembly composition as a useful descriptor of the state of a molecular assembly as well as the basis for evaluation of reproduction. Composition is defined as a vector of the normalized amounts or concentrations of every compound type within the demarcated volume of a molecular network. The compositional dynamics is then translated to a trajectory in multidimensional space.55 Preservation of molecular composition along the growth of a mutually catalytic network (homeostatic growth16,23) is a fitting criterion for the quality of assembly reproduction. Compositional information is thus a substitute for sequence information,16,56 and is similar to certain types of epigenetic information in present-day cells23,49 (e.g., proteome and transcriptome).57,58
We note arguments that autocatalytic networks evolve not only all at once but in a piecemeal manner.4 This indeed is the case for GARD dynamics, whereby self-copying of lipid assemblies happens via a graded molecular entry, culminating in reaching homeostatically growing composomes.16 At the same time, it was pointed out that this distinction is not critically important, and what matters is whether the system can evolve, exhibiting cumulative, adaptive changes. Later, when genes emerged and horizontal gene transfer became reality, the system could reach a more advanced evolutionary Darwinian threshold as defined by Vetsigian et al. and Woese.6,7