Read on Twitter
If you need a quick distraction from stress scrolling (and let’s face it, no significant election news is coming out for a few more hours!)- check out our new @eLife paper out on the evolution of cellular differentiation. https://elifesciences.org/articles/54348
This work was a collaboration with @YunkerLab (the leader of this project!) and @joshuasweitz, and the work was primarily done by our students @dayan1406, @PedroM_Z, and Shane Jacobeen. You can read Corina Tarnita’s excellent summary here: https://elifesciences.org/articles/63328 .
Cellular differentiation is central to multicellularity, cementing the transition to a new level of individuality and underpinning the evolution of increased complexity. It’s widely assumed specialization only evolves when it ‘pays off’ with higher returns than generalism.
Mathematically, this means the ROI needs to be accelerating with increased specialization, or convex. This is certainly true for organisms that live in unstructured, mass-action environments (like microbes in a water droplet), but cells often have limited trading partners.
Things get interesting when we model trade among cells in simple multicellular organisms (specified as topologically-defined graphs, which are actually a great approximation for early multicellular life), restricting cells to trade only with their neighbors.
In these cases, fitness doesn't only depend on the total ROI generated from trade, but also in how those resources are distributed between compatible specialists (i.e., trade between like-specialists has no value).
We found that most topologies still favor specialization when the return from trade is decelerating (concave)! Remember, these are conditions that classical theory predicts should favor generalism, not specialization.
This is because it’s easy for these cells to interact preferentially with an appropriate trading partner: somatic cells can preferentially direct their ‘viability-supporting’ resources towards adjacent germ cells, which can make trade adaptive when it otherwise wouldn’t be.
In particular, the more bipartite a cellular topology is, the more concave the ROI curve can be while still favoring specialization. Bipartite-like graphs are common in early multicellular organisms, because it’s easy for cells to grow by staying together after division.
This is a neat result because it expands our understanding of the conditions under which cellular specialization can evolve, which has been heavily shaped by thinking about trade in unstructured conditions.
The spatial structure created by early multicellular bodies matters, stacking the deck in favor of simple multicellular organisms evolving reproductive specialization.
