You can evolve there from here. And from here. And here …

Littorina saxatilis

Littorina saxatilis. Photo by Sergey Yeliseev.

If evolutionary history somehow reverted back to the “warm little pond” in which life began, and started over from almost-scratch, would the re-diversification of life end up, four billion years later, pretty much as we see it today? I think most evolutionary biologists would say, after noting that “pretty much as we see it today” is a mighty vague hypothesis statement, that it probably wouldn’t. Especially at the scale of millions of years, world-changing events happen by chance, making the odds pretty slim that a second four-billion year run would go all the way from the origin of life to a planet dominated by ape-descended life-forms who think wireless phones are a pretty neat idea.

On a smaller scale, though, it often does seem that evolutionary history repeats itself. Different populations of the same organism, encountering similar environments or the same natural enemies, adapt similarly—as, for example, in the repeated parallel changes of marine sticklebacks colonizing freshwater, or three different lizard species adapting to the same white sand dune formation. But when the traits that change in the course of adaptation are created by the collective action of many genes, it’s reasonable to think that changes in different subsets of those contributing genes might create similar changes in the visible trait, the phenotype.

As modern sequencing methods let us track genetic changes with greater precision, it’s possible to look for exactly that process—different genetic paths to the same adaptive result. A study just released online ahead of print in Molecular Ecology seems to have found such a case in populations of small snails.

Littorina saxatilis is a thumb-sized snail that lives on rocky shores and in salt marshes along the Atlantic coastlines of Europe and North America. Rough periwinkles, as they’re commonly called, are found in dense populations that are often isolated from each other over quite small distances; unlike some mollusks, the snails’ larvae don’t disperse through the open ocean, and adults travel, at most, less than a dozen meters over the course of a lifetime, unless they hitch a ride on floating debris.

This poor dispersal sets the stage for genetic differentiation, and L. saxatilis comes in a wide array of shapes, and colors. A recurrent theme among these “ecotypes” is that periwinkles from habitats exposed to the force of incoming waves tend to be small, with large shell apertures that allow maximum grip on the rocks. In contrast, periwinkles from calmer habitats, where predatory crabs are a greater risk than washing out to sea, have large, thick shells with smaller apertures. These “wave” and “crab” ecotypes can be found within a few meters of each other, and such pairings turn up everywhere rough periwinkles are found.

Butlin et al (2014), figure 1B.

Examples of the “wave” and “crab” rough periwinkle ecotypes from three different European sites. Photo from Butlin et al (2014), figure 1.

Wave-type periwinkles and crab-type periwinkles prefer to mate with each other. Male periwinkles hunt for mates by following the trails of mucous females leave behind, and, given a choice of trails, will follow the mucous of their own type—and, even when cross-type matings do occur, they’re often so brief that no actual sperm is transferred. Between this prezygotic isolation and the fact that snails crossing into the wrong habitat are at a disadvantage, the two ecotypes are also differentiated at a genetic level, though the differentiation between ecotypes at the same location isn’t as great as differentiation between periwinkles from different locations.

The new study aims to dissect that genetic differentiation more finely, to see whether pairs of ecotypes at different locations differ at the same places across the genome. Anja Westram and her coauthors assembled RNA-sequence data for samples of each ecotype at sites in Spain, Britain, and Sweden. Rather than recover sequence data for every individual periwinkle in their sample, the team created pooled samples, two for each ecotype at each site. Aligning the pooled RNA-seq data to a (very early) draft assembly of a L. saxatilis genome sequence let them estimate the frequency of different genetic variants within each pool, which they could use to calculate genetic differentiation between pools, between ecotypes, and between sampling sites.

Westram et al. then identified loci having the greatest FST metric of genetic differentiation between ecotypes, and compared these “outlier” loci between the replicate ecotype pools within sampling sites, and between different sampling sites. Loci that showed up as FST outliers at more than one sampling site would be cases where the periwinkles’ evolution into two different ecotypes followed the same genetic path—and while there were a number of those loci, they weren’t in the majority.

Westram et al (2014), Figure 3.

Scatterplots of between-ecotype FST at the Spanish (SP, red), Swedish (SW, blue), or British (UK, white) sites. Points that are double-colored (blue-red, white-red, white-blue) are outliers at both sites compared. From Westram et al (2014), Figure 3.

The above plots from the paper’s third figure illustrate this nicely. In the three plots, each point is a genetic locus, placed based on its between-ecotype FST value at two sampling sites. Colored points are outliers within at least on site—so double-colored points are outliers at both of the sites being compared. The proportion of loci that were outliers at more than one sampling site was greater than expected at random, and also greater than the number of outlier loci in comparisons of the same ecotype, which the authors used rather cleverly as a “natural null” expectation.

Though these comparisons of different sampling sites found more shared outliers than expected purely by chance, they were still a relatively small proportion of the outlier list—usually 20% or less, depending on how stringently outliers were defined. That could suggest that the authors’ FST-based “genome scan” has found a lot of false positives, and comparison of different populations serves as a kind of triangulation to reveal those false positives. However, Westram et al. argue that if this were the case, we might expect that the proportion of shared outliers would increase if they set a higher threshold FST value to define outliers—and, when they tried this, the proportion of shared outliers stayed about the same.

It seems probable that at least some portion of the loci with high between-ecotype FST that aren’t shared between populations actually do represent different loci that have contributed to the evolution of separate ecotypes at different sites. As the authors note, this could reflect differences between periwinkles of the same ecotype at different sites—even though they are adapted to the same broad habitat type, they may have evolved to deal with other environmental factors that differ between the three sampling sites. But it’s also what we expect if the traits that separate periwinkle ecotypes have evolved by more than one genetic route.


Butlin R., G Charrier, B Jackson, C André, A Caballero, JA Coyne, J Galindo, JW Grahame, J Hollander, and P Kemppainen. 2014. Parallel evolution of local adaptation and reproductive isolation in the face of gene flow. Evolution. 68(4) 935-949. doi: 10.1111/evo.12329

Johannesson K., P. Kemppainen, C. Andre, E. Rolan-Alvarez and R. K. Butlin. 2010. Repeated evolution of reproductive isolation in a marine snail: unveiling mechanisms of speciation. Phil. Trans. Royal Soc. B 365(1547) 1735-1747. DOI: 10.1098/rstb.2009.0256

Westram A., M Alm Rosenblad, JW Grahame, M Panova and RK Butlin. 2014. Do the same genes underlie parallel phenotypic divergence in different Littorina saxatilis populations? Molecular Ecology. doi: 10.1111/mec.12883


About Jeremy Yoder

Jeremy Yoder is an Assistant Professor of Biology at California State University, Northridge. He also blogs at Denim and Tweed, and tweets under the handle @jbyoder.
This entry was posted in adaptation, genomics, population genetics and tagged , , , . Bookmark the permalink.