Killer whales. Credit: Robert Pittman
Killer whales. Credit: Robert Pittman

On Dolphins, Big Brains, Shared Genes and Logical Leaps

In 2012, a team of Chinese scientists showed that a gene called ASPM has gone through bouts of accelerated evolution in two very different groups of animals—whales and dolphins, and ourselves.

The discovery made a lot of sense. Many earlier studies had already shown that ASPM is one of several genes that affect brain size in primates. Since our ancestors split apart from chimps, our version of ASPM has changed with incredible speed and shows signs of intense adaptive evolution. And people with faults in the gene develop microcephaly—a developmental disorder characterised by having a very small brain. Perhaps this gene played an important role in the evolution of our big brains.

It seems plausible that it did something similar in whales and dolphins (cetaceans). They’re also very intelligent, and their brains are very big. Compared to a typical animal of the same size, dolphin brains are 4-5 times bigger than expected, and ours are 7 times bigger than expected. The Chinese team, led by Shixia Xu, concluded that “convergent evolution might underlie the observation of similar selective pressures acting on the ASPM gene in the cetaceans and primates”.

It made for a seductive story. I was certainly seduced. In my uncritical coverage of the study, I wrote: “It seems that both primates and cetaceans—the intellectual heavyweights of the animal world—could owe our bulging brains to changes in the same gene.”

Many other scientists were sceptical—check out the comments in my original post—and it seems they were right to be. Three British researchers—Stephen Montgomery, Nicholas Mundy and Robert Barton—have now published a response to Xu’s analysis, and found it wanting. “It’s a completely plausible hypothesis but they didn’t test it very well,” says Montgomery.

In the original paper, Xu’s team looked at how ASPM has changed in 14 species of cetaceans and 18 other mammals, including primates and hippos. ASPM encodes a protein, and some changes in the gene don’t affect the structure of the protein. These “synonymous mutations” are effectively silent. Other “non-synonymous mutations” do change the protein and can lead to dramatic effects (like microcephaly). The Chinese team claimed that a few cetacean families had a high ratio of non-synonymous to synonymous mutations in ASPM—a telltale sign of adaptive evolution.

But Montgomery’s team had two problems with this conclusion. First, it’s statistically weak. Second, it’s not unique to cetaceans. Xu’s team largely looked at brainy groups like cetaceans and primates, but the British trio found exactly the same signature of selection in other mammals, including those with average-sized brains. “It looks like ASPM evolved adaptively in all mammals,” says Montgomery. “It could be that ASPM is a general target of selection in episodes of brain evolution and isn’t specific to large brains.”

Xu’s team also failed to check if the changes they found in ASPM were actually related to differences in cetacean brains. If the gene is changing quickly under the auspices of natural selection, does that translate to equally fast changes in brain size? The Chinese team never explicitly addressed that question. Montgomery’s team did, and their answer was a resounding no.

“We felt a little bad picking on them because it’s quite a common problem,” says Montgomery. “People pick a gene to analyse because it’s linked to something interesting. They find that it’s got this pattern of evolution, and they infer that it’s doing what they thought it was doing. It’s a circular argument. “

“These analyses need to be followed up with experimental work (if that is possible) or treated with caution if not,” says Graham Coop from University of California, Davis. “At best, such studies can only act to generate hypotheses about the role of a particular gene in phenotypic evolution”. That’s because most genes do many jobs, “and we are profoundly ignorant of many of these roles and how they differ across organisms.”

ASPM, for example, isn’t a “brain gene”. It creates molecular structures that help cells to divide evenly. It’s activated in the embryonic cells that make neurons, so if it’s not working properly, fewer neurons are made and individuals end up with small brains. But ASPM is also activated in other parts of the body.

As Vincent Lynch pointed out in a comment to my earlier post, ASPM affects the development of the testes:

“This brain-testis connection was described by Svante Pääbo’s lab. They swapped the mouse and human ASPM genes, I assume hoping to breed a super-intelligent strain of mice, and surprisingly found that nothing happened. Bummer… But rather than uncovering a role for ASPM as a casual agent of increased brain size in the human lineage, these authors found ASPM was required for male fertility (yes, the jokes are obvious) and suggested that the signal of selection observed in humans and other primates is likely related to role in testis. It is on old observation that many testis expressed genes evolve rapidly, many under some form of positive selection.”

So, maybe ASPM’s fast evolution in primates is more a story about nuts than noggins. Then again, Montgomery’s team have indeed found that changes in primate ASPM are related to differences in the size of their brains but not their testes.

These conflicting results illustrate just how important it is to test hypotheses carefully, rather than finding bits of evidence that look nice together, and uniting them through conjecture. It’s a valuable cautionary note to both scientists and journalists alike.

Reference: Montgomery, Mundy & Barton. 2013. ASPM and mammalian brain evolution: a case study in the difficulty in making macroevolutionary inferences about gene–phenotype associations. Proceedings of the Royal Society B