Using artificial systems to explore the ecology and evolution of symbioses

Abstract

The web of life is weaved from diverse symbiotic interactions between species. Symbioses vary from antagonistic interactions such as competition and predation to beneficial interactions such as mutualism. What are the bases for the origin and persistence of symbiosis? What affects the ecology and evolution of symbioses? How do symbiotic interactions generate ecological patterns? How do symbiotic partners evolve and coevolve? Many of these questions are difficult to address in natural systems. Artificial systems, from abstract to living, have been constructed to capture essential features of natural symbioses and to address these key questions. With reduced complexity and increased controllability, artificial systems can serve as useful models for natural systems. We review how artificial systems have contributed to our understanding of symbioses.

Fig. 1

Artificial systems for studying symbioses demonstrate the trade-off between controllability and complexity of a system. Insets show examples at different scales of controllability: i the Lotka–Volterra mathematical model for studying the dynamics of predator–prey systems [9]; ii Avida digital organisms [14] for studying evolution; iii an engineered bacterial system for studying the predator–prey interactions [26]; and iv Biosphere 2 project, a synthetic ecosystem to study the earth biosphere [41]. Photo by C. Allan Morgan, courtesy of Global Ecotechnics

Fig. 2

Potential origins of mutualism and cooperation. Yellow indicates genetic changes. a In the laboratory environment, mutualism has been observed to arise spontaneously (iv, [68]), through rapid phenotypic adaptation (i, [37]), as the result of genetic engineering (ii, [38]) or evolution (iii, [39]). b In nature, mutualism can arise as the result of initially parasitic (i), competitive (ii), or commensal (iii, iv) interactions. i In an initially antagonistic host–parasite interaction, a parasite (small oval) invades its host (large circle), exploits host nutrients (blue squares), and produces a byproduct (green rectangle) that cannot be utilized by the host. Mutualism results if the host evolves to retain, utilize, and perhaps depend upon the byproduct of the parasite. ii Initially, organisms compete for a resource (blue squares) and convert it into a waste product (red triangle) that inhibits growth by reducing metabolic flux. Natural selection can favor the evolution of a type that is able to utilize the waste product as a primary source of energy, transforming competition into mutualism. iii, iv An initially commensal species (hexagon) can evolve to generate a benefit (green rectangle) for its partner (magenta rod), resulting in mutualism. If production of the released product (green rectangle) is costly, a cooperator variant (yellow hexagon) can rise to high frequency in a spatially structured habitat [39] (iv). c Communication can be co-opted for cooperation. i A single cell (rod) produces an “inexpensive” membrane-permeable small molecule (black circle) that can, if at a high concentration, activate transcription of genes encoding expensive excreted molecules. The small molecule serves to monitor the diffusivity of the environment: in turbulent water (top), the molecule leaves rapidly and its low intracellular concentration is insufficient to activate gene expression. In soil (bottom), the molecule takes much longer to diffuse away, resulting in an intracellular concentration sufficient to induce (red) the production of expensive excreted molecules (green squares), which can be used, for example, to break down otherwise indigestible resources. ii Precisely the same mechanism could be used to allow a group of cells to sense when their local density is high enough to engage in cooperative group behavior such as synchronized release of excreted products (green squares). Group benefits are realized if a higher concentration of the excreted product generates a disproportionally larger benefit

Fig. 3

a A Turing pattern can result from an autocatalytic “activator” that promotes the formation of an “inhibitor” with a greater diffusivity (wavy arrow) than the activator. b Random small disturbances in an initially homogenous mixture of activator and inhibitor can result in the spontaneous formation of highly ordered patterns, such as stripes

Fig. 4

If two interacting species, such as host (grey circles) and parasite (grey squiggles), are propagated separately (a), mutations that allow the host to escape parasite infection (blue) and mutations that allow the parasite to increase infection of host (orange) are not selected for. If the parasite is allowed to evolve against a constant host at the ancestral state (b), mutations in the parasite that increase its ability to infect the host are selected for (light grey, dark grey, and black indicate increasing levels of infectivity). Evolutionary arms race in coevolution (c) results in a faster rate of evolution and increased genetic diversity for both partners, as the host evolves to evade parasite invasion (changing host color) and the parasite evolves to attack the host in a frequency-dependent manner (changing parasite color)