Show Me The Science Month, Day 2

How do two populations change genetically when they are subjected to different evolutionary pressures? To answer this question, many intrepid evolutionary biologists have trudged out into the field to painstakingly study wild populations, but in many cases, we can learn more by studying evolution in the lab. In a paper published the February issue of Nature Genetics, a group of Portuguese and US researchers report a study of 28 years of evolution in a set of lab fruit fly populations. Their results are an example of how studying evolution in the lab, even for a short time, can provide insight in to how natural selection shapes the genetic contours of a population.

The researchers use a set of fruit fly populations that have been evolving under lab conditions for 28 years. The experiment started off with an original population of flies, grown under standard lab conditions. Those flies were then split up and placed under different evolutionary pressures: one population was subjected to starvation, another was under selection to breed early, while a final population was selected to breed late. Actually, the researchers didn't just do this once; they did the experiment 4 times in parallel, something you can't do when you study evolution out in the field.

The fly populations evolved under these various conditions for about 50 generations, and then they were split again. This time, evolution was put in reverse: one population from each group was placed back under the normal, standard lab conditions for another 50 generations. The result is a set of fly populations that have been subject to selection for 100 generations (to put this in perspective, note that your ancestors 100 generations back lived about 2500 years ago). Each population adapted to its new environment - the populations were better able to resist starvation, or breed earlier, for example.

But what happens to the genes under these conditions? Each population started out with a gene pool of a particular depth - a range of genetic variation among the flies that provided the raw material for evolution to act on. When the flies were place in a new environment, beneficial genetic variants increased in frequency, while non-beneficial ones decreased. The authors found that the genetic change experienced by these population was not due to new, beneficial mutations showing in the flies; the change was mostly due to alterations in the frequency of existing mutations. This kind of change can happen easily in a sexually-reproducing population, since each generation experiences a mix-n-match as paternal and maternal chromosomes swap pieces (a process called recombination). The result is this: a gene with a beneficial mutation has neighbors who may just be dead weight - they're not especially beneficial or harmful, but recombination allows the gene harboring the beneficial mutation to try out new neighbors. Recombination makes it possible for selective pressure to genetically shape a population without relying on brand-new beneficial mutations.

It's important to note that some of the genetic changes were not due to selection at all. Random sampling plays a big role in evolution; some genetic variants spread through a population not because they are beneficial, but just by sheer chance.

So what happens when you put evolution in reverse? After 50 generations of a return to standard lab conditions, the reverse-evolved flies were fully adapted to their new (but ancestral) environment. Perhaps not surprisingly, evolution reversed itself at the genetic level as well - genetic variants in the fly population returned about half way to their original, ancestral frequencies. In other words, genetically they didn't need to completely retrace their steps in order to become fully adapted to the environment their ancestors had left 100 generations ago.

Why is this important? The authors write that "To our knowledge, this study is the most comprehensive description of the molecular population genetics of adaptation in a sexual species." They suggest that this study "confirms the prediction that evolution resulting from standing variation is more repeatable than evolution resulting from mutational input." In other words, evolution is more reversible if it doesn't involve new adaptive mutations, which I don't think is that surprising. The bigger question seems to be about how a population's evolutionary past impacts it's evolutionary future when placed in a new environment.

While it's not an earth-shattering paper, this work is a solid demonstration of how evolution is studied in the lab, and how evolutionary models can be rigorously tested with quantitative experiments.

Join me tomorrow for day three of Show Me the Science month. Evolution as a science is alive and well. Each day I will blog about a paper related to evolution published in 2009.