Mutual fitness benefits arise during coevolution in a nematode-defensive microbe model

Abstract

Species interactions can shift along the parasitism‐mutualism continuum. However, the consequences of these transitions for coevolutionary interactions remain unclear. We experimentally coevolved a novel species interaction between Caenorhabditis elegans hosts and a mildly parasitic bacterium, Enterococcus faecalis, with host‐protective properties against virulent Staphylococcus aureus. Coinfections drove the evolutionary transition of the C. elegans–E. faecalis relationship toward a reciprocally beneficial interaction. As E. faecalis evolved to protect nematodes against S. aureus infection, hosts adapted by accommodating greater numbers of protective bacteria. The mutualism was strongest in pairings of contemporary coevolved populations. To generally assess the conditions under which these defensive mutualisms can arise and coevolve, we analyzed a model that showed that they are favored when mild parasites confer an intermediate level of protection. Our results reveal that coevolution can shape the transition of animal‐parasite interactions toward defensive symbioses in response to coinfections.

How parasite interaction strategies alter virulence evolution in multi‐parasite communities

“…; Rafaluk‐Mohr et al. ). Thus, analyzing them in biologically relevant scenarios is a step forward toward understanding virulence evolution in the field.…”

“…Ultimately, cross immunity, immunopathology, and spite can all lead to reductions in host exploitation rates, but this potential is only revealed in our asymmetric models. All of the interactions analyzed in this study have been found to take place in multi-parasite, asymmetric scenarios (Graham et al 2005;Ezenwa and Jolles 2011;Shrestha et al 2013;Rafaluk-Mohr et al 2018). Thus, analyzing them in biologically relevant scenarios is a step forward toward understanding virulence evolution in the field.…”

“…Alternatively, one parasite may directly interfere with a coinfecting parasite that does not return the interference, as in bacterial coinfections of Caenorhabditis elegans (Rafaluk‐Mohr et al. ). In these cases, we must consider how an interaction mechanism alters the evolution of both the focal parasite (the one triggering the interaction) and the nonfocal parasite (the one receiving the interaction).…”

“…For instance, one parasite species may suppress the immune system, whereas the other does not, as in the case of many coinfections involving helminths or HIV (Lauer et al 2002;Ezenwa et al 2010). Alternatively, one parasite may directly interfere with a coinfecting parasite that does not return the interference, as in bacterial coinfections of Caenorhabditis elegans (Rafaluk-Mohr et al 2018). In these cases, we must consider how an interaction mechanism alters the evolution of both the focal parasite (the one triggering the interaction) and the nonfocal parasite (the one receiving the interaction).…”

Simplified model assumptions artificially constrain the parameter range in which selection at the holobiont level can occur

“…This simple alteration of VVD's model-assumptions leads to a broad parameter space in which a stable population of helper microbes is maintained despite their costly behavior, suggesting selection at the holobiont-level (Fig 2). Such a non-linear dependency of fitness on an underlying factor that can be influenced by the microbiome is not uncommon in the real world, stemming from physiological, ecological, or behavioral factors (2)(3)(4)(5). VVD's model may thus suggest the opposite conclusion from the one they reach, depending on the choice of fitness function.…”