Friday 28 March 2014

Quick thoughts on bacterial individuality

Are bacteria individuals? Does it matter? I'm planning to pose (and answer) these questions for my upcoming talk at PBUK2014 in Cambridge.

A biofilm is a structured, surface attached colony of millions of bacterial cells. Recent biology conferences have been full of people detailing the ways in which biofilms are analagous to multicellular organisms. They demonstrate complex systems of intercellular signalling, describe intricate and variable colony architectures and propose various adaptive hypotheses at the whole biofilm level. It is clear that the analogy has captured the biological imagination (one paper is titled 'Biofilm: City of microbes') and fuelled much research.

But what turns on it? I have found three decent arguments presented in the literature.


One argues from adaptation as follows;
Premise: Biofilms exhibit biofilm-level adaptations. Examples -
Premise: Only individuals bear adaptations. This is from GC Williams - adaptations at a level require selection at that level.
Conclusion: Biofilms are individuals.

I'm not inclined to go along with this argument, because adaptive hypotheses are just too easy to come up with. They are limited only by our imagination.

Another argument is from cooperation. I take this from Nadell et al 2009;
Premise: Biofilms exhibit complexity/organisation, the existence of which requires cooperation among biofilm parts. (optional reading: complexity requires signalling, which requires cooperation).
Premise: Only individuals have cooperation among their parts.
Conclusion: Biofilms are individuals.

I think that there are merits to this argument, but applying it requires some careful thinking about the exact spatial scale at which we expect cooperation to occur - it will be bigger than bacteria, but smaller than whole biofilms i imagine. We also need to be careful in thinking about what properties are required to support cooperation - it isn't only about relatedness!


Finally, there is an argument from evolvability, mostly found in Ereshefsky & Pedroso 2013;
Premise: Individuals require some mechanism that makes them evolvable. The paradigm example is a reproductive bottleneck, which makes a lineage of multicellulars evolvable by helping mutations to spread through whole organisms.
Premise: Lateral Gene transfer makes biofilms evolvable.
Conclusion: Biofilms are indivdiuals.

My concern with this argument is that it licenses a conclusion for the individuality of bacteria, not of biofilms. In the multicellular case, we say that evolvability applies at the level of lineages. If biofilms have evolvability, isn't that reason to say they are lineages?

Some issues that need to be settled:
Are we convinced that identification of complex or irreducibly complex higher-level phenotypes offers justification for the historical action of higher-level natural selection?

What is the correct scale at which we should relativise our questions about the individuality of microbial communities?

How much lateral gene transfer really goes on in microbial communities and to what extent can we keep track of it?

I still need to do some thinking before my slot on tuesday. Fingers crossed I get some sleep.......... 

Nadall, Xavier & Foster 2009 “The sociobiology of biofilms”
 
Ereshefsky & Pedroso 2013 "Biological Individuality: The case of biofilms"

No comments: