Armor in Freshwater Sticklebacks: Selection Against, or Just No Selection For?

Jan 12 2009 Published by under Anti-Evolutionism, Biology, Science

There are certain organisms that you hear about a lot in evolutionary biology. In some cases, like Drosophila flies or E. coli bacteria, that's because the organisms are easy to use in experimental studies. Other organisms, like Hawaiian silversword plants or Galapagos finches, come up frequently because they're fantastic examples of evolution happening out in the "real world". And then there are those rare cases where an organism is both a fantastic example of evolution in the field, and a convenient organism to work with in more controlled circumstances. The three-spined stickleback (Gasterosteus aculeatus) is one of those doubly-convenient organisms.

There are populations of three-spined sticklebacks in the ocean and in many freshwater streams and ponds. The oceanic populations have been around for a long time, but the freshwater populations are all relatively recent in evolutionary terms - they're found in bodies of freshwater that were formed after the ice sheets retreated about 12,000 years ago. These populations appear to have evolved independently of each other, but they share a number of similar traits.

One of the more notable of the traits concerns the bony "armor" along the sides of the fish. The marine populations typically have a line of over 30 bony plates along their sides. The freshwater populations typically have only 6-9 of these plates. Why this is the case is a classic evolutionary biology question: do the freshwater populations lose the armor because there is a real advantage to losing the plates, or do they just lose them because there's no real disadvantage to losing them.

Casey Luskin cites a news story about a recent scientific paper to support his view that the loss of the armor is just the result of the freshwater populations not facing the selective pressure seen in the oceans:

A scientific study published a few months ago reports that the marine stickleback (the ones with the armor plates) came before freshwater sticklebacks (the ones without armor-plating), meaning that this is not an example of the evolution of a new function, but an example of loss-of-function, or what one might term devolution. As a Science Daily press release on the paper stated, this evolution entailed "[s]hedding some genetically induced excess baggage":

Shedding some genetically induced excess baggage may have helped a tiny fish thrive in freshwater and outsize its marine ancestors. Measuring three to 10 centimetres long, stickleback fish originated in the ocean but began populating freshwater lakes and streams following the last ice age. Over the past 20,000 years - a relatively short time span in evolutionary terms - freshwater sticklebacks have lost their bony lateral plates, or "armour," in these new environments. "Scientists have identified a mutant form of a gene, or allele, that prohibits the growth of armour," says UBC Zoology PhD candidate Rowan Barrett. Found in fewer than one per cent of marine sticklebacks, this allele is very common in freshwater populations.

Alas, Collins' example, which is intended to break down the distinction between macroevolution and microevolution, really only provides evidence that populations of organisms can lose unique and complex features when selection pressure is relaxed. ...

The scientific paper itself, however, reaches a very different conclusion:

Our results highlight the utility of direct measurements of natural selection on genes for understanding the genetic basis of adaptation by enabling us to test a mechanism favoring reduction of lateral plates in freshwater environments. Many of our results are consistent with selection against high plate number, although they do not rule out the possibility that selection is also occurring on genes tightly linked to Eda. Our results also expose opposing selection on Eda early in life similar in magnitude to the measured advantage of the low allele later in life. This demonstrates not only that countervailing selection pressures diminish the advantage of the low allele over the whole life span but also that the overall fitness effects of Eda do not seem to be determined solely by differences in lateral plate number. Along with the fluctuating dominance in fitness at the Eda locus, these results indicate that there may be additional pleiotropic effects of this gene. This work underscores the need for a synthesis of population biology and genomics, to determine the genetic basis of fitness differences in natural populations.

The two different opinions raise a key question: how can we tell the difference between selection in action and a lack of selection in action? The Science paper in question provides us with a simple answer: you do a hell of a lot of hard work to conduct a series of fairly simple experiments.

Dolph Schluter's lab at the University of British Columbia has been working with the sticklebacks for a while, and one of his grad students, Rowan Barrett, took a look at the armor question. The experimental design was very simple: he went out and trapped a large number of sticklebacks in the ocean. He took a genetic sample from each of them, and checked to see if the fish in question had a copy of a gene that is known to cause a drop in armor plates (this gene is found in about 1% of the oceanic fish). He selected wild-caught fish that had one copy of the gene in question (heterozygotes), and placed similar numbers of the fish into each of four experimental freshwater ponds. The fish were allowed to breed, then Barrett sampled the offspring repeatedly during the following year to follow any changes in the gene frequency over time.

The easiest way to show you what Barrett found is just to show you one of the figures from his paper:

322 255 F2-A

(Modified from Figure 2 of Barrett et al 2008)

That figure shows the percentage of alleles on the x-axis and time on the y-axis. The four lines on the graph represent the four separate experimental ponds.

If there's one thing that stands out on that graph, it's the agreement among the lines. At least three of the four are always doing the same thing at the same time. The pattern is not remotely random-looking. If the changes in gene frequency were simply the result of a selective pressure disappearing, we would not at all expect to see a graph like that one. We'd expect to see each of the ponds doing its own thing, and the gene frequency remaining more or less stable over time. It's also clear from the graph that whatever is going on is not simple. There's clearly a strong selective pressure favoring the armor gene early in the lifespan of the organisms, which is then followed by a strong selective pressure against the gene. It's going to be interesting to see what the Schluter lab turns up as they continue to study these organisms.

This experiment does clearly illustrate one thing, though: talking about "devolution" or "loss of function" doesn't make much sense at all when we're talking about evolutionary biology. At least as far as these fish are concerned, not having armor isn't a lack of a trait. It is a trait. A lack of armor has its own set of advantages and disadvantages, and is - as Barrett illustrated - subject to both positive and negative selection. Saying that freshwater sticklebacks simply lack the armor trait doesn't make any more sense than it would to say that the saltwater sticklebacks lack the "lack of armor" trait.

It's just another example of something that's easy to lose sight of: organisms do not exist to manufacture complex traits. Complex traits exist because - and when - they help manufacture more complex organisms.

R. D. H. Barrett, S. M. Rogers, D. Schluter (2008). Natural Selection on a Major Armor Gene in Threespine Stickleback Science, 322 (5899), 255-257 DOI: 10.1126/science.1159978

14 responses so far

  • Frank J says:

    You mixed up X and Y axis. And yes, you caught at least one of us off guard at the Panda's Thumb with "Freshwater" in the title.

  • KP says:

    FYI there is also an endangered freshwater population in S. California that lacks the plates completely: the "unarmored threespine stickleback." Worked as a tech on a project dealing with these when I was first out of undergrad...
    I don't see what Casey's point is supposed to be. The paper isn't about "macroevolution vs. microevolution." It is about evaluating selection and fitness on a variable trait. Armor has different fitness in different environments, just like antibiotic resistance has different fitness inside versus outside hospitals. Furthermore, it is difficult to predict the complete effects of selection if more than one trait might be linked.
    Note to Casey: Changes in allele frequencies are evidence that evolution is occurring. Writing it off as "microevolution" does not change the fact that it is evolution and the authors do not predict that speciation will occur. But thanks for drawing attention to a paper I had missed!

  • KP says:

    ...I thought I'd better be fair and read Casey's whole blog before concluding that it had no point. I stand by my argument.
    Another note to Casey: Some traits have fitness costs as well as benefits. The paper shows evidence that the benefit of armor as protection in one environment is outweighed by the cost of producing it in another.

  • Jim Thomerson says:

    Many years ago, in population genetics class, I learned of the (person's name) effect. Where an unused adapation is lost; for example, eyes in cave species. Can anyone tell me the person's name?
    A while back, in Science Daily, there was a news release about sticklebacks in a lake (Lake Washington?) which has had its pollution cleared up so that the water is clearer and there are fewer plants. The sticklebacks are regaining their armor, presumably as a result of being exposed to stronger predation. Anyone have that reference?

  • Jim Thomerson says:

    Here is the Science Daily news relaease.

  • Jim Platt says:

    As a person who teaches evolution at the undergraduate level, I have been familiar with the stickleback example for some time and use it quite frequently in classes to illustrate the arbitrary difference between "micro" and "macro" evolutionary phenomena. There are, in fact, two traits that appear to rapidly increase in frequency in freshwater populations of sticklebacks. Not surprisingly, Luskin has chosen the loss of body armor to use as an example of "genetic loss" since this the less studied trait. The other trait - loss of pelvic spines - is understood at a greater level of molecular detail (mainly through work in David Kingsley's lab at Stanford). In this case, we know with reasonable certainty we are not dealing with "genetic loss". The relevant gene is this case is Pitx1, a gene coding for a transcription factor that is deployed during development of the stickleback (and in most, if not all, other vertebrates). One of the functions of the gene is to activate the transcription of other genes that are required for hind limb development (pelvic spines in this case). But, the gene also has other critical roles including the normal development of the pituitary and lower jaw skeleton. The point here is that these other functions of the gene must be retained in sticklebacks that lose their pelvic spines. Thus, the gene is not lost - indeed, DNA sequencing has shown that the coding region of Pitx1 is identical in marine and freshwater populations. What has occurred is a mutation in the regulatory region of Pitx1 in freahwater populations that acts to prevent the expression of the gene, but only in the developing pelvic region. Pitx1 continues to be expressed normally in other areas. While there is much more than this to the stickleback story, this should provide enough information to demonstrate that Luskin's "genetic loss" hypothesis has been conclusively falsified in the case of the Pitx1 gene and the loss of pelvic spines in sticklebacks.

  • Mark Duigon says:

    Another example is the loss of eyes in cave critters. A very interesting investigation was reported by Yamamoto & Stock in Nature (2004, v. 431 p. 844-847). They found surface-dwelling fish related to blind cavefish, and diddled with levels of the regulator molecule hedgehog and...lo and behold, produced blind fish. What they found was that the eyeless condition was not so much mere loss of a feature that was unused (Casey's "devolution") but rather the developmental process involved selection for enhanced sensory and feeding features whose development was controlled by genes that, like genes for eye development, were regulated by hedgehog. So while it's true that a little-used feature was lost, the driving force for that loss was selection favoring other features whose embryological development shared a regulator with genes for the lost feature.

  • Bjorn Ostman says:

    Complex traits exist because - and when - they help manufacture more complex organisms.

    That should be "Complex traits exist because - and when - they help manufacture fitter organisms." While some measures of complexity may increase in evolution, it really isn't the yardstick by which the dynamics are determined. Fitness is.

  • gillt says:

    I would suggest doing away with the relative, baggey, error-prone distinction between micro and macro evolution. Speciation, phenotypic plasticity, natural selection, and genetic drift are all more accurate and more common to scientific journals. So why are these inadequate?

  • John Marley says:

    They aren't inadequate, that's the problem. Creationists and other loons who want to deny Modern Evolutionary Theory need ambiguous, poorly specified terms to facilitate goal-post shifting and similar tactics.

  • John Marley says:

    Thank you for pointing out Luskin's nonsense. But I have to say, I really don't read it too closely anymore. If he were to say "My name is Casey Luskin", I would demand a government-issued photo ID and two alternate forms of identification before I believed him. Now when I see his name on something, I just take it as read that it's wrong.

  • John Kwok says:

    Years ago I read some fascinating work done by Michael Bell - whom, I believe has collaborated with Schluter - comparing and contrasting paleontological with neontological data on sticklebacks. Without a doubt, one of the better examples of studying microevolution using both sets of data. I was especially impressed seeing the evidence for substantial morphological variation within this genus.
    As for trying to dissolve the distinctions between microevolution and macroevolution, I strongly beg to differ. There is ample scientific literature which delves into both, and we should not be beholden to creationist IDiots for deciding what is - and what isn't - valid scientific terminology.
    Sincerely yours,

  • dreikin says:

    While I think I've read before that the same effect has been experimentally observed (saltwater sticklebacks transported to freshwater, see if they lose armor), I'm curious about what evidence there is that the saltwater variety came first - do we know this, or is it an assumption? Doesn't help the creationists either way, but it'd still be interesting to know/find out.
    Also, is it me, or does just the translation from a saltwater freshwater environment seem like it could have all sorts of developmental effects without any mutations at all, just by messing with any regulators that depend on salinity/molarity/etc.?

  • KP says:

    Re: Dreikin. The Gasterosteidae are described as "Peripheral marine" by:
    TM Berra - 1981. An atlas of distribution of the freshwater fish families of the world
    That generally means that they were a marine group that invaded freshwater; however, it is difficult to know whether certain individual species within peripheral families arose in freshwater and then re-invaded marine envirionments. With sticklebacks it seems safer to assume a primarily marine, then freshwater sequence. With sculpins (cottidae), I have taken flak from Barry Chernoff (formerly @ Field Museum in Chicago) over this assumption.
    Also, there is a 2001 update of Berra's book and I don't know if the info for stickleback changed...