Not Exactly Rocket Science

i-9d4dd18a4560f6f5bd65c4baf31b1add-Echolocation.jpg

Millions of years before humans invented sonar, bats and toothed whales had mastered the biological version of the same trick – echolocation. By timing the echoes of their calls, one group effortlessly flies through the darkest of skies and the other swims through the murkiest of waters. It’s amazing enough that two such different groups of mammals should have evolved the same trick but that similarity isn’t just skin deep.

The echolocation abilities of bats and whales, though different in their details, rely on the same changes to the same gene – Prestin. These changes have produced such similar proteins that if you drew a family tree based on their amino acid sequences, bats and toothed whales would end up in the same tight-knit group, to the exclusion of other bats and whales that don’t use sonar.

This is one of the most dramatic examples yet of ‘convergent evolution’, where different groups of living things have independently evolved similar behaviours or body parts in response to similar evolutionary pressures.

It is one of a growing number of studies have shown that convergence on the surface – like having venom, being intelligent or lacking enamel – is borne of deeper genetic resemblance. But this discovery is special in a deliciously ironic way. It was made by two groups of scientists, who independently arrived at the same result. The first authors even have virtually identical names. These are people who take convergence seriously!

Yang Liu from the East China Normal University had previously shown that echolocating bats share very similar versions of Prestin, even species that were only distantly related. This time, he sequenced the gene in even more bats as well as a wide range of whales. These included toothed species (dolphins, porpoises, orcas and sperm whales) that use sonar, and baleen species that don’t.

Based on the DNA sequences of these Prestin versions, Liu drew a mammal family tree (a ‘phylogeny’). It looked much like what you would expect, with the whales and bats clustering in separate family groups. But convert the sequences into amino acids and the picture changes dramatically. Suddenly, the family tree becomes utterly misleading. The echolocating mammals, be they bats or whales, are united as close relatives, to the exclusion of their rightful evolutionary kin.

i-00682c5ce79908a8819b08d9563d4add-Prestin-tree.jpg

Ying Li (see what I mean?) from the University of Michigan found a similar result. She sequenced the Prestin gene in the bottlenosed dolphin and compared it to sequences from other mammals. Again, she found that Prestin sequences place the dolphin as a close cousin of echolocating bats rather than species that it’s actually more closely related to, such as cows. 

At first, it might seem strange to see such strong convergence at the genetic level. After all, bats and toothed whales echolocate very differently. Bats create their sonar pulses using their voicebox while whales pass air through their nasal bones. Bats send their calls through air and whales send their through water. A single gene can’t have accounted for these differences in production.

Instead, Prestin‘s role is in detecting the rebounding echoes. It is activated in the “outer hair cells” of the ear, which allow mammals to hear high frequencies. In echolocating species, these cells are shorter and stiffer than normal, making them exquisitely sensitive to the ultrasonic frequencies used in echolocation. Li thinks that the Prestin changes might have helped to tune the outer hair cells of echolocators to high-pitched noises.

Liu used his sequences to reconstruct what Prestin would have looked like the ancestor of all bats and the ancestor of all whales. Compared to these original versions, echolocating species have accrued the same set of 14 amino acid changes, whether they have wings or flippers. It seems that there are only very few ways, if not only one, for mammals to hear the ultrasonic sounds needed for biological sonar.

Exactly what these amino acid changes did to the Prestin protein, and how they led to the evolution of echolocation, is a mystery for another time. It will also be interesting to see if these changes have started cropping up in the Prestins of other animals with cruder forms of sonar, like oilbirds, swiftlets, shrews and tenrecs.

References:

Liu et al. Convergent sequence evolution between echolocating bats and dolphins. Current Biology in press.

Li et al. The Hearing Gene Prestin Unites Echolocating Bats and Whales. Current Biology in press.

More on convergent evolution:

i-77217d2c5311c2be408065c3c076b83e-Twitter.jpg&nbspi-988017b08cce458f49765389f9af0675-Facebook.jpg&nbspi-6f3b46114afd5e1e9660f1f502bf6836-Feed.jpg&nbspi-deec675bab6f2b978e687ca6294b41a5-Book.jpg


Comments

  1. #1 sjburnt
    January 25, 2010

    This is quickly becoming my favorite science blog. the articles are well written and informative. Thanks, Ed!

  2. #2 Starry Night
    January 25, 2010

    The genetic basis of echolocation is fascinating.

    Also of particular interest is the “echo” relationship between bats and some species of moths that bats hunt. It turns out that some moths can hear a bat’s sonar pulses and avoid becoming dinner. These moths, using just four nerve cells, can tell how close a bat is and whether it is over, under, left, or right.That information can tell a moth to dive rapidly and increase its chances of escaping capture.

  3. #3 Sven DIMilo
    January 25, 2010

    Clearly another example of not convergence, but horizontal gene transfer. Long ago there were whales that didn’t really hear that well and so they struck up a symbiotic relationship with a bat (the bat’s angle was to compensate for its legendary blindness). The bats used to kind of drive the whales around until the fateful viral-mediated gene transfer that allowed whales to finally hear what they’d been missing. Whales could now dive deep enough to drive off the annoying and now redundant bats and so that’s where manta rays came from.

    You could look it up.

  4. #4 Jack
    January 25, 2010

    “Again, he found that Prestin sequences place the dolphin as a close cousin of echolocating bats rather than species that it’s actually more closely related to, such as cows.”

    But a cladogram is just a way of presenting similarities of characteristics. A way of showing the matrix. There is no concept of “actually more closely related to”.
    Is that not right?

  5. #5 Jacob
    January 25, 2010

    A very cool result!

    Jack, you are correct that that dolphin isn’t really more closely related to bats in any way; it’s an illusion of the cladogram. Usually a cladogram does reflect evolutionary relationships well, but in this case convergence is so strong that it breaks the connection between relatedness and sequence similarity.

    Also, I just want to point out that Ying Li is female (the post says “he”).

  6. #6 Jack
    January 25, 2010

    What I mean is that the phrase “actually more closely related to” does not have any meaning. No cladogram shows how creatures are actually related.
    It just shows how creatures can be sorted depending on which characteristics you choose.
    Correct?

  7. #7 Jacob
    January 25, 2010

    The goal of most cladograms is to show how species are actually related. Whether they do or not depends on what data go into it and what methods are used to generate it, but that is the idea. Phentics is grouping species by overall similarity, while cladistics is grouping species according to traits they share due to common heritage. Most cladograms these days are generated with cladistic goals, even though some methods are rooted in phenetics.

    Of course, the cladogram above is not supposed to show relationships, it’s supposed to show convergence, but most other cladograms are different.

  8. #8 Lab Rat
    January 26, 2010

    Jack: The cladogram is not *showing* that the cows are more closely related to bats. It is generally *known* that cows are more closely related, and they are just being used for comparison in the cladogram.

    And if you put the right kind of information into cladograms, they can give an approximation of evolutionary relatedness, particularly if you use highly conserved RNA sequences.

  9. #9 Lab Rat
    January 26, 2010

    Jack: The cladogram is not *showing* that the cows are more closely related to bats. It is generally *known* that cows are more closely related, and they are just being used for comparison in the cladogram.

    And if you put the right kind of information into cladograms, they can give an approximation of evolutionary relatedness, particularly if you use highly conserved RNA sequences.

  10. #10 Jake Cameron
    January 26, 2010

    I’m with the first commentor!

    As luck would have it, I am finally tackling Dawkin’s Blind Watchmaker this week, such a lovely illustration would well serve the next print edition! Thanks for your passion and clarity Ed!

  11. #11 Jason
    January 26, 2010

    Very, very cool. And butterflies using the earth’s geomagnetic field, and ants using the sun and counting their steps (http://thoughtfulanimal.wordpress.com/2010/01/26/path-integration-in-the-desert-ant/) – seems like a big week in the blogosphere for animal navigation.

  12. #12 Ed Yong
    January 27, 2010

    Jacob – thanks for the correction on Li’s gender! Fixed.

  13. #13 Blackbird
    January 27, 2010

    Cool post! A beautiful science story and I think this will become a textbook example of molecular convergence. There is a third group of echolocating vertebrates, many species of cave swiftlet – of fame due to the Chinese birds nest soup – also echolocate. It would be cool to look at their prestins. As for the cladogram discussion, a cladogram based on one gene shows just the similarities of that gene/protein. In many cases, this is informative on the genetic relationships of organisms, in many others, not.

  14. #14 LKL
    January 27, 2010

    This is so, so cool. I love this universe.

  15. #15 Comrade PhysioProf
    January 30, 2010

    Fascinating post, Ed!

  16. #16 kai
    February 1, 2010

    But convert the sequences into amino acids and the picture changes dramatically.

    So, does this mean that the DNA sequences actually were different, but code for the same amino acid sequence? That’s pretty cool and definitely shows that it’s not horizontal gene transfer but convergence.