From a University of Chicago Press Release.

The genetic toolkit that animals use to build fins and limbs is the same genetic toolkit that controls the development of part of the gill skeleton in sharks, according to research to be published in Proceedings of the National Academy of Sciences on March 23, 2009, by Andrew Gillis and Neil Shubin of the University of Chicago, and Randall Dahn of Mount Desert Island Biological Laboratory.

“In fact, the skeleton of any appendage off the body of an animal is probably patterned by the developmental genetic program that we have traced back to formation of gills in sharks,” said Andrew Gillis, lead author of the paper and a graduate student in the Department of Organismal Biology and Anatomy at the University of Chicago. “We have pushed back the evolutionary origin of the developmental genetic program that patterns fins and limbs.”

This new finding is consistent with an old theory, often discounted in science textbooks, that fins and (later) limbs evolved from the gills of an extinct vertebrate, Gillis added. “A dearth of fossils prevents us from definitely concluding that fins evolved from gills. Nevertheless, this research shows that the genetic architecture of gills, fins and limbs is the same.”

The research builds on the breakthrough discovery of the fossil Tiktaalik, a “fish with legs,” by Neil Shubin and his colleagues in 2006. “This is another example of how evolution uses common developmental programs to pattern different anatomical structures,” said Shubin, who is the senior author on the PNAS paper and Professor and Associate Dean of Organismal and Evolutionary Biology at the University of Chicago. “In this case, shared developmental mechanisms pattern the skeletons of vertebrate gill arches and paired fins.”

The research also showed for the first time that the gill arch skeleton of embryonic skates (a living relative of sharks that has gill rays) responds to treatment with the vitamin A derivative retinoic acid in the same way a limb or fin skeleton does: by making a mirror image duplicate of the structure as the embryo develops. According to the researchers, the genetic circuitry that patterns paired appendages (arms, legs and fins) has a deep evolutionary origin that actually predates the origin of paired appendages themselves.

“These findings suggest that when paired appendages appeared, the mechanism used to pattern the skeleton was co-opted from the gills,” Gillis said. “Perhaps we should think of shark gills as another type of vertebrate appendage–one that’s patterned in essentially the same way as fins and limbs.”

The deep structural, functional, and regulatory similarities between paired appendages and developing gill rays, as well as the antiquity of gills relative to paired appendages, suggest that the signaling network that is induced by retinoic acid had a patterning function in gills before the origin of vertebrate appendages, the research concludes. And this function has been retained in the gill rays of living cartilaginous fishes.

Comments

  1. #1 Jason Thibeault
    April 19, 2009

    As a computer guy who only recently discovered how awesome genetics is (and who in fact occasionally wonders if he missed his calling in life, and who also finds talking in third person to be quite irritating to readers), I’ve been drawing parallels between DNA and code for a computer in my mind that probably aren’t really analogous. Figured I’d try one out on someone who actually knows something about biology.

    This sounds a lot like a subroutine that is called with different variables based on where in the code it’s called from. My understanding is that the DNA around something that codes for a particular feature can affect the outcome, even if it’s not directly involved in the code for that feature; like how one sequence of code can actually code for several different enzymes depending on where the read starts, one chunk of code that creates an appendage can be altered by what’s to either side of that code. Since DNA isn’t like a real computer language, I envision it more like an esoteric programming language like Malbolge or Brainf***, where tiny mutations in the code can cause all sorts of drastic results in the end product.

    What do you think, sir? It’s damn cool either way.

  2. #2 Jason Thibeault
    April 19, 2009

    To further my own analogy, what if we could one day write an interpreter for DNA? We’d basically have to code for physics, chemistry, an environment conducive to reproduction and life itself, so we’d be writing a virtual reality program unparalleled by anything but reality itself, so that would basically make us God for this reality.

    O NOES WHAT IF WE ARE THE SIMULATION!!1one

  3. #3 NewEnglandBob
    April 19, 2009

    Jason, there already is an interpreter for DNA and it is called life.
    I think you are focusing too much on the DNA itself and not at all on the proteins and enzymes.

    DNA is more like a macro language and the numerous proteins are the macros, to be called upon to do the actual work, with timing and concurrency of utmost importance.

  4. #4 RBH
    April 20, 2009

    Tempting though it is, ‘ware of pushing the computer metaphor too far. At that level it’s chemistry, not coding. One sees computer folks get themselves into awful twists pushing the metaphor way past where it’s helpful. GilDodgen at Uncommonly Dense comes immediately to mind.

  5. #5 Jason Thibeault
    April 20, 2009

    Indeed — reality has a lot more margin for error, what with copying fidelity and the timing of the chemical reactions. The closest a computer could come to this is to multithread to the n-th degree to allow simultaneous processing of every chemical reaction. Abstracting genetics to a computer language is helpful in my own understanding, but I do understand it’s only an abstraction. In discussions on genetics with my girlfriend who is most certainly not a computer programmer, any time I pull out a computer analogy, she usually replies with “…what?” Where she “got it” before, my analogy usually just confuses and annoys her.

    Still though, there are some lessons to be learned from the analogy, or there wouldn’t have been so much buzz from the EvoLisa project around January. And it wouldn’t have driven me to rewrite the same program in Python, with an actual population size and varying selection pressures (shameless self plug — not that it’s worth looking at). Nor would better programmers, like cdk007 on Youtube, have come up with a program putting the lie to the “blind watchmaker” stuff from the creationists’ camp.

  6. #6 NewEnglandBob
    April 20, 2009

    Actually, there is more analogy to an analog computer, which I did get to use in college, just before my 30+ year career in software. Tweaking those knobs could show a ‘butterfly’ effect cascade at the output.

  7. #7 AK
    April 22, 2009

    @Jason Thibeault, NewEnglandBob:

    I’ve been working on a post on the subject, too much to put in a comment. See my How Smart is the Cell? Part II: The Gene Activation network as an Analog Computer.

    You may want to start with part I…