31 May 2010

Laying down boundaries for brain evolution in cichlids

ResearchBlogging.orgWalter Garstang famously said that ontogeny creates phylogeny: you need to understand the development of a structure to understand the diversity of that structure across species.

There are a few different ways to change the way a structure is put together. Research on the development of limbs has tended to view morphological changes as being caused by changing boundaries that delineate different regions of the embryo. If you want a bigger forebrain, shift the boundary between the forebrain and the midbrain backward.

Neurobiologists, however, have sometimes suggested that brains differ due to the amount and duration of neurogenesis that goes on during development. If you want a bigger forebrain, make neurons for a longer time in the forebrain than in the midbrain.

A new paper by Sylvester and colleagues tries to look at the relative importance of these two developmental mechanisms in brain evolution. To do this, they looked at the brains of cichlid fishes. Cichlids are renowned among evolutionary biologists for their diversity and the speed at which it was achieved, and this diversity extends to their brain morphology. They chose six species to study; Cynotilapia afra and Copadichromis borleyi are shown here (click to enlarge). The former is a mbuna, a rock-dwelling fish, and the latter is a sand-dweller; the authors had three species of each.

The brains of the mbuna and non-mbuna differ in their proportions: the mbuna have smaller forebrains (strictly speaking, purists would say the telencephalon) than the non-mbuna.

Sylvester and colleagues visualized the cichlid brains using several that are well-known to be involved in development, like sonic hedgehog (shh), and Wnt and found that the pattern of expression of these did indeed differ in these species, typically following the mbuna / non-mbuna split.

To test whether these genes caused the differences in brain sizes, the authors exposed embryos of the mbuna species Labeotropheus fuelleborni (right) to lithium chloride, which affects the Wnt pathway. When they did this, they found that the L. fuelleborni brains were looking rather more like non-mbuna brains, which proportionately larger telencephalons.

Sylvester and company argue that this suggests the very early expression of genes that lay out patterns in the nervous system are key regulators of brain evolution. Considering that they sort of set up timing of neurogenesis as an alternate hypothesis in their introduction, however, it’s worth pointing out that they didn’t actually measure neurogenesis or try to manipulate it in a manner similar to what they did for these patterning genes and proteins. It’s possible that neurogenesis is playing a role in shaping diversity of brains in other ways than tested here.

Additional: The journal is publishing a commentary on this paper. While the original article is open access, the commentary, alas, is not.


Sylvester, J., Rich, C., Loh, Y., van Staaden, M., Fraser, G., & Streelman, J. (2010). Brain diversity evolves via differences in patterning Proceedings of the National Academy of Sciences 107(21): 9718-9723. DOI: 10.1073/pnas.1000395107

Photo of Cynotilapia afra by Calwhiz  on FLickr; photo of Labeotropheus fuelleborni by Lee Nachtigal on Flickr; photo of Copadichromis borleyi by Petrichor on Flickr. All used under a Creative Commons license.

1 comment:

Mike Mike said...

Like you said, after reading this article, it worried me that they didn't actually quantify neurogenesis at all. It may be that the techniques that work in mammals aren't as effective in fish (I'm not that familiar with the field), but it's hard to believe that there isn't something they could do to make a more direct assay of their proposed mechanism of the observed effect of lithium. I was also sort of irked that they skimmed over an explanation of the possibility of Lithium treatment causing changes in brain growth outside of the WNT pathway. It seems like a big deal to me when you don't know quite what your experimental manipulation is doing (or it at least sounds suspicious when you act like that doesn't bother you.)