séminaire d'agro-écologie

"The evolution of interdependency by neutral evolution in holobionts" - Pr Marc-André Selosse, MNHN - lundi 12 octobre à 11h, salle de conférences de l'Inra




In holobionts the evolution of each partner is partly driven by the other partners (co-evolution). The evolution of holobionts is often viewed as a progressive evolution, based on the emergence of new adaptive properties that enhance the fitness, or diversify the niche. In many cases, such evolution enhances the interdependence between partners, because new functions are only achieved in symbiosis. The present paper focuses on the emergence of interdependence, and emphasizes that it can arise even without emergence of any new property, or any progressive evolution. I propose two examples where such a neutral evolution may have acted.

First, this may explain why microbiotas are acting both in plant and animals as developmental signals for immunity maturation: the existence of the so-called ‘priming effect’ was revealed by observation of germ-free animals and non-mycorrhizal plants. Since no evidence supports that microbial signals are more relevant than endogenous ones, a neutral evolution can account for this dependency: any hypothetic endogenous signal can be lost because microbial colonization, reliably occurring at germination or birth, can substitute for it.  

Second, neutral evolution may explain the extreme genetic reduction in some endosymbionts. Mitochondria and plastids are organelles of endosymbiotic origin, and in some eukaryotic lineages (respectively, anaerobic or achlorophyllous lineages) they fully lost their genomes. Yet, they persisting as bodies surrounded by two membranes dividing in host cytoplasm. This extreme evolution of bacterial dependence to the host is facilitated by the genetic redundancy between the host and the endosymbiotic bacterium.

Generally, when two initially independent partners permanently interact, redundant properties become unstable: a mutation in one of the partner can be complemented by the presence of the other, or even by a complementing mutation in the other. Independency is then lost without any gain of function, nor any positive selection. This can thus been viewed as a neutral evolution. Moreover, the accumulation of such ratchet steps over times makes the reversion to independency more and more unlikely: neutral evolution can be predicted to drive reciprocal dependencies within holobionts.

Pr Marc-André Selosse
Institut de Systématique, Évolution, Biodiversité
Muséum national d’Histoire naturelle,
Sorbonne Universités,
57 rue Cuvier, CP50,
75005, Paris, France