Early eukaryogenesis by host-initiated obligate ectosymbiosis and metabolic inhibition

Microbial symbiosis is extremely common among metabolically coupled cells, and, presumably, gave rise to mitochondria. How such symbioses emerge, evolve and stabilize are unknown, especially in the prokaryotic domain, where endosymbioses are virtually nonexistent. Yet there is growing evidence suggesting that the eukaryotic cell emerged from such a prokaryotic partnership, where integration was not the result of phagocytotic inclusion (as in case of plastids) but of metabolic cooperation. While prokaryotes almost ubiquitously engage in metabolic syntrophy, it is unknown if such cooperation alone can enable stable, dependent ectosymbioses that could pave the road toward physical integration of parties. Here, we tested the hypothesis that free-living syntrophy can lead to stable ectosymbiosis between microbial partners, using an ecological mathematical model. Assuming an already syntrophic and asymmetric partnership of free-living hosts and symbionts, we investigated under what conditions obligate ectosymbiosis evolve. Our results show that reduced inhibition (of self-inhibiting metabolic products) over the contact surface of partners can stabilize the ectosymbiotic consortia against free-living forms. Furthermore, strong metabolic activity between the host and their ectosymbionts could facilitate obligacy in their physical association. The model points to the significance of the contact surface in the evolution of crucial endosymbiotic features. Our results support the hypothesis that cooperative, syntrophic microbes (especially prokaryotes) are capable of coevolving to form species-specific ectosymbiosis by means of reducing the inhibition of accumulating products, a necessary first step towards endosymbiotic integration. The model provides a plausible explanation on how common metabolic syntrophy can lead to physical integration of parties through gradual ectosymbiosis. Moreover, our work fills a gap between microbial cooperation models (assuming free-living species) and those that are concerned only with already concluded physical integration under a multilevel selection paradigm.