Recently, Sarah Kopac and Jonathan Klassen described a new model to understand the evolution of host-microbe symbioses in their paper “Can They Make It on Their Own? Hosts, Microbes, and the Holobiont Niche”. In this model, hosts and microbes can change the shape of their respective niches, and thereby the context in which selection can act. This permits a test of holobiont theory by identifying instances where an organism’s phenotype depends on its symbiotic partners. Jonathan answered the following questions, incorporating feedback from Sarah. Be sure to read all the way through.
1. What motivated the theory reported in the paper?
It feels a bit murky now, but looking back I think there were three main motivations. First, we were quite struck by how host-microbe relationships can be thought of as a community ecology problem, i.e., where and how do interspecific relationships form and what is their outcome? So because much of the holobiont/hologenome literature has started from a population-genetics perspective (e.g., http://journal.frontiersin.org/article/10.3389/fmicb.2014.00046/full), we thought that drawing from a different intellectual tradition might be a way to get beyond current sticking points in the field's discussion. Another interesting aspect of the community ecology perspective is that it defines the context in which selection might act, leaving whether or not selection does act as a separate question. I think that it is too common to assume selection without rigorously testing against alternative hypotheses, e.g., drift or dispersal limitation, to our detriment. Second, we wanted to strongly highlight how any selection that did occur likely impacted hosts and microbes differently, e.g., because microbes have shorter lifespans and higher population sizes and dispersal rates than their hosts (http://doi.wiley.com/10.1002/bies.201500074). Third, one major take-away from my years of reading the lovely Dynamic Ecology blog (https://dynamicecology.wordpress.com/) is that mathematical models have a rigor and testability that verbal models cannot match, and thereby overcome the potential for verbal models to become misconstrued. Expressing our ideas mathematically so that they would be unambiguously testable therefore became an important goal.
2. How did you come up with the title?
Truthfully, this was a revision suggested by one of the reviewers. (And they were right - thanks!) Sarah came up with this version, and I think it’s great because it really captures the key question of our model: do host-microbe interactions matter for the evolution of either partner? I think the answer is that it depends on the context, i.e., does the interaction change the shape of either partner’s niche in a way that changes how selection might act? I also love how (in my mind at least) there’s a U2 reference - https://www.youtube.com/watch?v=CuDqHtAR6L8. I’ve long been a fan.
3. When and how did you two come together and agree on this paper? Were there varied opinions about how to approach the problem?
We agreed on the central idea of explaining holobionts in terms of niche theory early on, and so were unified on that front. There was a bit of divide and conquer after that, with Sarah focusing more on the examples of different niche shifts and I focusing more on the models themselves. Then we came back together and made sure that the examples we saw in the literature could be explained using our models. I found it very helpful to get someone else’s view of the literature – accurately synthesizing everything that’s out there alone would have been a massive challenge.
4. What are the most salient findings?
I think that it is striking how our models are agnostic towards any particular mechanism of host-microbe interaction. For example, a similar shift in a host’s niche could be result from that host selecting microbes with a particular trait from the environment each generation, or it could be due to tight host-microbe co-evolution and vertical transmission. The important thing is that in either case the interaction changes the shape of a host’s niche in a way that selection might act upon. Both partners are therefore needed to completely describe host evolution. I am also struck by how our models accommodate the full diversity of symbiotic relationships, including the various forms of conflict that change the shape of a host’s niche but in a different way than mutualism. Finally, it seems that the symbiosis community has already been at least implicitly thinking in terms of our model because it was easy to find studies that had already tested it by changing one parameter at a time, e.g., by swapping microbiomes between constant host genotypes and vice versa. Hopefully our work can guide future experiments that continue to explicitly test these ideas.
5. What are the most crucial questions moving forward in your mind? What do you want to do next?
Perhaps one problem with our model is that precisely defining the shape of an organism’s niche is fiendishly difficult, mostly because there are a potentially infinite number of niche axes to consider. I think it will be interesting to understand how many of these are actually needed to come up with an accurate niche description. For example, to what extent to do microbe-microbe interactions need to be considered to describe how microbes alter a host’s niche? I also think it will be critical to learn more about microbes in non-host environments so that we can determine if hosts modify the niches of their microbial symbionts outside of obvious cases like intracellular mutualists and pathogens. Lastly and as described above, I think it will be crucial to explicitly test whether any host-microbe interaction phenotype actually arises due to selection, e.g., vs drift or dispersal limitation.
6. Anything else you want to add?
Thanks to our colleagues in the symbiosis field for being so collegial, even in disagreement! I look forward to increasing the rigor of our studies, and to a deeper understanding of how ecology and evolution both shape host-microbe relationships.