News and Publications

Design, Discovery, and Understanding of the Complex Properties of Self-Organized Chemical Networks

Understanding the molecular basis of life has been the central goal of chemical and biological sciences. The complete or partial genome sequences of more than 20 organisms have been reported recently, and according to estimates, within the next 3 years, the entire human genome will be sequenced. These milestones coupled with the new and emerging technologies aimed at functional analyses of gene products are expected to provide the basic chemical map of living systems. However, although this information is necessary for a complete understanding of the molecular basis of life, it is not sufficient to answer the central question of how inanimate chemical transformations manifest into a living being. This fundamental level of understanding is essential for the continued ability to discover and produce novel therapeutic agents, determine essential targets for genetic interventions, provide mechanisms for early disease diagnostics, and formulate the basis of aging--essentially, every imaginable process associated with living organisms.

The main goal of our research program is to discover and understand the mechanisms that transform inanimate chemical transformations into the animate chemical characteristics of living systems. To reach this goal, we rationally design and re-create in the laboratory various basic forms of autocatalytic chemical networks--the postulated first step in the transition from inanimate to animate--and study how the interplay of molecular information and nonlinear catalysis can lead to self-organization of molecular systems and the expression of emergent properties.

The required basic method, an informational nonlinear chemical system, was recently established through our rational design of synthetic amide bond­forming catalysts (ligases and replicases; Fig. 1). These catalysts efficiently promote sequence-specific condensation of peptide fragments under neutral aqueous solution conditions and are useful in the design and study of the primary forms of self-organized chemical networks (Fig. 2). In one study, we showed that a small number of reactants can combine to spontaneously form an "autocratic" network that has dynamic error correction as its emergent property. The system can sense the formation of mistakes (mutant peptides) and respond by upregulating the production of the native replicator sequence.

In another study, a "mutualistic" network was formed in which 2 replicators joined to enhance each other's production and in turn each individual replicator's chance of survival. This system is the first and only known example of symbiosis at the molecular level. More recently, we showed formation of a pure "reciprocal" network in which each species in the autocatalytic set acts as a specific catalyst for the formation of the other species (reciprocation). Although none of the species alone is capable of self-replication, the overall network effectively reproduces itself. The reciprocal network is the simplest example of how an emergent property (in this case, self-reproduction) can arise out of coupled catalytic cycles.

Through the studies described here, we determined the necessary and sufficient conditions for effecting self-organization of molecular systems and the production of emergent properties. Our simple constructs have some of the basic properties often associated with living systems, such as selection, reciprocation, symbiosis, and error correction. Current efforts focus on the design, discovery, and characterization of more complex networks and molecular ecosystems. We hope that by studying such synthetic constructs, a blueprint for better understanding of living chemistry will emerge.