A thorny parallelism

Featuring: the 'Three-spine' stickleback, Gasterosteus aculeatus

This small fish (barely a few inches long) lives in marine or fresh-water environments. They happen to be very common and, as it seems, easy to maintain in the lab (like most of the fish smaller than 10 cm). And there a number of research groups "lobbying" for stablishing it as a model system, especially for evo-devo studies.

What makes it be so especial? Something really curious: sea populations, like the japanese (JAMA) in the figure, have an "armour" in the posterior part of the body but fresh-water populations, like the Paxton Lake (PAXB) in the figure, don't.

(Alizarin-red tinction, staining bone elements; bar length is 1 cm, fish in real size.)

Don't you find something strange here? Fresh-water populations are isolated from each other, so, how is it possible that all of them have the same phenotype? Two explanations can be proposed: either the difference is due to an environmental cue in the development (what would make the delights of eco-devo'ers) or that there has been massive parallel evolution.

We know that the reason is the latter because the phenotype is inherited even if there is a change in the environment, but now we also know which gene is responsible for the parallelism. Pamela Colosimo and coworkers describe in a Science article the identification of the allele that causes the phenotypic differences. Yes, that's right, all (but one) the fresh-water populations with 'low' morph share the same allelic variant of the gene Ectodysplasin (eda) whereas sea-water populations present the normal variant (Eda). And in the article they not only describe the identification of the gene, no. They also make functional studies that effectively prove that the gene found is necessary and sufficient for triggering the phenotypic change.

This figure is particularly revealing. In the northern hemisphere map you can find the locations of the samples used for the phylogenetic analysis. Blue for 'low' morphs, red for 'complete' morphs. The tree in C is obtained using 193 SNP's (single nucleotide polymorphisms) found in 25 different loci randomly chosen; it should reveal the actual phylogenetic tree of the fish populations. The tree in B is obtained using the respective sequences for the Eda locus. The correlation between morph and allele is almost 100%, only one japanese population (NAKA) has the normal allele, showing an example of convergent evolution, rather than parallel.

What can we deduce from all this? First, that the eda allele has to be present in a very small frequency in the sea-water populations because all fresh-water populations have the same allele; such a high sequence identity is impossible to explain by independent mutations. That, and that 'something' in common must be in all fresh-water environments, not an epigenetic cue but a selective pressure that promotes the fixation of the low morph allele (otherwise, the fixation by genetic drift in all those populations would mean that the frequency of the allele in sea-water populations had to be huge, which is not the case; this is even more evident since there is at least a case of the same phenotype occuring via a different mutant genotype).

What do you think? Population genetics and developmental genetics, all together thanks to genomics. It seems true that this fish has everything!

---

1. P. F. Colosimo et al. (2005), Widespread Parallel Evolution in Sticklebacks by Repeated Fixation of Ectodysplasin Alleles Science 307:1928-1933
2. G. Gibson (2005), The Synthesis and Evolution of a Supermodel Science 307:1890-1891