Yeast Study Shows That Cheaters Never Prosper - Except in Unchanging Environments

Yeast research. Source.

That old adage "cheaters never prosper" may actually turn out to be quite true when it comes to the evolution of the population.  This is especially the case in cooperative species, such as microbes (whoever thought microbes were so social?).  A recent open access study looking at colonies of yeast found that populations dominated by cheaters were more likely to face extinction than those dominated by non-cheaters.  That means that in highly social and cooperative species, evolutionary dynamics are influenced by selfish behavior. 

Cooperative behavior helps populations thrive by producing common goods - by secreting molecules to the surrounding media - and not surprisingly, they have higher fitness at higher population densities.  But cooperative populations can be challenged by "cheaters" - or members of the population that use the common goods but do not contribute to its production.  This decreases the population's overall fitness, in a real-life tragedy of the commons; except in this case, the cheaters have higher fitness and proliferate throughout the population, ultimately leading to population collapse.

The authors used two strains of yeast, one wild-type (the cooperators), and one lacking the SUC2 gene (the cheaters).  SUC2 plays an important role in sugar metabolism: the resulting protein is localized to the space between the cell wall and the cell membrane, so when it breaks down sucrose into glucose and fructose, 99% of the products are diffused into the medium.  The yeast strain lacking this protein become the cheaters because they can't break down sucrose into usable products, but they can use what is produced by other individuals.  The proliferation of each population was measured via cell density - ultimately populations with high numbers of cheaters would have slower growth rates.  There's a mathematical model involved to measure this, but I'm not going to get into that because it's pretty complicated (for anyone interested, you can find their description of the model here).



Population density and SUC2 frequency
in yeast populations. 


The cheaters and cooperators were grown together at different starting frequencies of each.  They were also both labeled with a different color fluorescent protein so they could be differentiated during flow cytometry (which is a method for counting cells).  The authors found that depending on the starting frequencies of cooperators and cheaters, there was either an extinction of the population (right, red line), or a cycle, with the ratio of cheaters to cooperators oscillating based on the density of each (right, black line).  That means that when the density of cheaters is too high, environmental glucose levels are reduced (due to the decreased density of cooperators), leading to slower growth of cheaters and thus allowing cooperators to make a comeback. 

Population density and survival
probability after exposure to
a harsh environment (in black)

Once the yeast cultures were exposed to a harsher environment (left, top panel), the strictly cooperative populations fared better than the mixed populations.  The presence of cheaters in a population decreases the resilience of the population, making the population as a whole unable to cope with a rapidly changing environment.

We normally like to think of the evolution of a population as happening on such a large time scale that it is difficult to observe.  But this type of study is particularly interesting because it points to the connection between evolutionary dynamics and population dynamics in the form of an ecological-evolutionary feedback loop (which was the mathematical model used for this study).  This field of study is still in its infancy, but another interesting (and frequently studied) eco-evolutionary feedback loop is the study of predator-prey dynamics.  Nevertheless, studies like this one on intraspecific variations in cooperation clearly show that the demographic fate of a given cooperative population depends on the level of cheaters, and that too many cheaters may lead to collapse in highly variable environments.