Generic bumps and recycled genetic cascades

How do you make a limb? Vertebrate limbs are classic models in organogenesis, and we know a fair bit about the molecular events involved. Limbs are induced at particular boundaries of axial Hox gene expression, and the first recognizable sign of their formation is the appearance of a thickened epithelial bump, the apical ectodermal ridge (AER). The AER is a signaling center that produces, in particular, a set of growth factors such as Fgf4 and Fgf8 that trigger the growth of the underlying tissue, causing the growing limb to protrude. In addition, there’s another signaling center that forms on the posterior side of the growing limb, and which secretes Sonic Hedgehog and defines the polarity of the limb—this center is called the Zone of Polarizing Activity, or ZPA. The activity of these two centers together define two axes of the limb, the proximo-distal and the anterior-posterior. There are other genes involved, of course—this is no simple process—but that’s a very short overview of what’s involved in the early stages of making arms and legs.

Now, gentlemen, examine your torso below the neck. You can probably count five protuberances emerging from it; my description above accounts for four of them. What about that fifth one? (Not to leave the ladies out, of course—you’ve also got the same fifth bump, it’s just not quite as obvious, and it’s usually much more tidily tucked away.)

Compared to legs and arms, of course, that fifth bump isn’t much, even in the guys. Early in development, though, it’s much more impressive in relative scale. Here is a crotch shot of a mouse embryo; the paddle-like things on either side are the hindlimbs, and the thick fireplug of a protuberance in the middle, looking like something Vaughn Bode would have drawn, is the genital tubercle…and at these early stages, it would look pretty much the same in a male or female mouse.

A: SEM photo of an embryonic GT and hindlimbs at E12.0. Arrow: Distal urethral epithelium (DUE) and urethral plate; Arrowheads: apical ectodermal
ridge (AER) in hindlimb. The embryonic tail region was removed.

In this paper by Yamada et al., the investigators examined the patterns of gene expression in that developing genital tubercle, and the place to start, obviously enough, was to use probes against genes that were known to be important in the development of those other four protuberances: genes like the Fgfs, Shh, and other known components of signalling cascades, like Wnt5a and Bmp genes. I trust you won’t be too surprised to learn that there’s quite a bit of correspondence.

Expression patterns of several genes (Fgf8, Shh, Wnt5a, Bmp4, Bmp7) during mouse genital tubercle (GT; B-F) and limb development (G-K). Whole-mount in situ hybridization with digoxigenin-labeled Fgf8, Shh, Wnt5a, Bmp4,
and Bmp7 probes. B-F: Murine embryonic GT at E12.5. Lower (ventral) side of GT is shown. G-J: Embryonic limb at E11.5. K: Murine embryonic limb
at E10.5. Note the “similar” expression of Fgf8 and Wnt5a in the distal epithelia and distal mesenchyme of limbs and GT. Shh is expressed in the
urethral plate (UP) (C) and in ZPA (H). Bmp4 is expressed in ventral-distal mesenchyme and ventral bilateral mesenchyme close to the urethral plate
(E). Bmp4 is expressed in limb mesenchyme (J). Bmp7 is expressed in the DUE (F) and also in AER of limbs (K).

Similar signaling cascades are being used in the assembly of both the limb and the genital tubercle—there’s Wnt5a/Fgf center at the tip, and a site for secretion of Shh localized to the posterior edge. There are also major differences, of course. One significant structural difference between a limb and a penis is that a penis contains a central duct, the urethra, and this has to form by folding of an epithelial sheet. Still, you can see a core developmental module in place in both kinds of structures, with a distal growth center using Wnts and Fgfs and Bmps, and a posterior polarizing center that uses Shh.

Signaling cascades for genital tubercle (GT) formation (A) and for limb formation (B). DUE,
the signaling epithelia of the GT, expresses Fgf8 and Bmp7. Adjacent to DUE, the distal mesenchyme of the GT expresses Bmp4 and Wnt5a. In the GT, the urethral plate expresses Shh in the
ventral midline (red region in A). Bilateral to the urethral plate, Fgf10 is expressed (C). AER is located
in the distal tip of the developing limb (B) and it expresses Fgf8, Fgf9, and Fgf4. Shh is expressed
in ZPA of the posterior limb and Wnt5a is expressed in the distal mesenchyme of limb (B). In the
GT, midline Shh is expressed adjacent to the bilateral mesenchyme, which expresses Fgf10 (D, E).
Shh (D) and Fgf10 (E) gene expression in the E13.5 embryonic mouse GT.

As additional evidence for some kind of coupling between the molecules of the limbs and genitals, the authors cite a number of rare genetic syndromes that share simultaneous defects in limbs and genitals, such as hand-foot-genital syndrome, Robinow syndrome, Pallister-Hall syndrome, and so on (not all of which I find convincing—they have such a broad distribution of effects that tying genital and limb formation together is difficult.)

Seeing these similarities is not a surprising result at all. Evolution reuses what it can; it’s far easier to develop a novel protuberance by switching on an existing ‘protuberance pathway’ in a new place than to generate entirely new molecular mechanisms to do the same thing. Both limbs and mammalian genitals bulge outward, and at least in that superficial element of their organization, it’s interesting to see that the same molecules are involved. What’s also fascinating is that they also have major differences, and sorting out the causal mechanisms behind the variations imposed on the root homologies is going to be extremely informative.

Unfortunately, the paper didn’t address the really pressing question: if limb and genital development are coupled to some degree, is it true what they say about shoe size?

Goodman FR,
Bacchelli C,
Brady AF,
Brueton LA,
Fryns JP,
Mortlock DP,
Innis JW,
Holmes LB,
Donnenfeld AE,
Feingold M,
Beemer FA,
Hennekam RC,
Scambler PJ (2000) Novel HOXA13 mutations and the phenotypic spectrum of hand-foot-genital syndrome. Am J Hum Genet. 67(1):197-202.

Yamada G, Suzuki K, Haraguchi R, Miyagawa S, Satoh Y, Kamimura M, Nakagata N, Kataoka H, Kuroiwa A, Chen Y (2006) Molecular genetic cascades for external genitalia formation: an emerging organogenesis program. Dev Dyn 235(7):1738-52.


  1. Steviepinhead says

    Maybe that, if you’re a lady, jb, you’re gonna have a much tougher time than normal tucking it tidily away…?

  2. CaptainMike says

    Apparently for the most part it is true what they say about shoe size. Bigger men tend to have bigger appendages of all kinds.

    However it should be noted that this is not always a reliable indicator. For example, I myself am of average height, but my feet and hands are quite small and my penis is larger than normal for my frame.

    Sonic Hedgehog is a protein of some kind. It’s because scientists keep giving things names like this that creationists claim that they’re making it all up. That and creationists can’t understand any other way of getting evidence. Oh, and they’re morons (the creationists). But that is just my opinion. Mind you, it is the opinion of someone with grossly oversized genitalia, which should count for something.

  3. says

    I hope that tubercle escapes the lab and solves the population problem.

    Thanks professor Myers (for not kicking me off your blog yet) haha.

  4. jb says

    I ain’t a lady, and tucking it away was never a problem until after the prostate surgery….

  5. G. Tingey says

    Didn’t uncle Tel (Pratchett) call it “The joy of socks” in “Monstrous Regiment” ????

  6. Kamensind says

    “sonic hedgehog”, just goes to show that geneticists are the only scientists with a true sense of humor.
    braniac, cheap date, lush, dreadlocks, grim reaper – yep all names of genes. More can be found here:

    But anyhow, wasn’t it something about the size of the nose that was to be indicative of the size of the fifth bump ?

  7. Kamensind says

    Actually the more pertinent question is this: as men age why are their ears getting bigger, the eyebrows bushier, the prostate larger and the penis smaller ?

  8. says

    I would like to understand more of the vocabulary that you are using PZ. ZPA, AER, Fgf8, Shh… sigh. Anyone has an idea which book I should read as a poor bioinformatician here to catch up on evo-devo basics?

  9. says

    For a nice molecular development introduction try this one:
    Coming to Life: How Genes Drive Development
    Christiane Nusslein-Volhard

    It’s a reasonably gentle introduction to the molecular mechanisms of pattern development in embryos that covers many of the important signalling proteins.

  10. Frumious B says

    How do you SEM a mouse embryo? I mean, aren’t they kind of damp from whatever they are preserved in? Wouldn’t that be hard to put under vacuum? Do you have to sputter them with gold?

  11. fnxtr says

    “sonic hedgehog”, just goes to show that geneticists are the only scientists with a true sense of humor.

    What, as opposed to the astronomers who named an asteroid “Zappafrank”?? :-)

  12. Don Culberson says

    Frumious – Yep.. they are dried first and then coated with gold.
    Uncle Don

  13. Gary says

    “Anyone has an idea which book I should read as a poor bioinformatician here to catch up on evo-devo basics?”
    Sean Carroll’s “Endless Forms Most Beautiful” or Carroll, Grenier, and Weatherbee’s “From DNA to Diversity” are both good places to start.

  14. Mike Fox says

    Remi –

    A great source for cell biology-type information on the web is
    this textbook from NCBI. Use the search box thingy on the left. It’s a good introduction to cell biology/cellular development. And it’s at no charge/no registration from this link. Yay!

    -Mike Fox

  15. TDG says

    So if there are five growth centers for later appendages, are the hox genes which control the initial growth the same as those which initiate the radial (five-fold) symmetry of echinoderm appendages; i.e., is our bilateral symmetry nothing more than a variation on the same theme used by crinoids, sea stars and their ilk? If that’s the case, then would tetrapod (chordate) evolution have a direct antecedent in echinodermata? And does that mean that fish would also have similar, five-fold symmetry controls despite their absence of the tetrapod body plan?


  16. says

    No, it’s not a case of five-fold symmetry — limbs and genitals are basically bolted-on additions to a bilaterally symmetric body plan.

  17. says

    Evolution loves to recycle: why invent something from scratch when you can simply reuse or retool? In morphological development, Sonic Hedgehog (Shh) is a great example of how proteins tend to get used again and again. Shh typically initiates the development of epithelial protrusions, scale buds, feather buds, hair follicle development, lung cilia, and photoreceptors. Evolution does something similar with organs. For example, the tripartite limb of early arthropods included a gill which later evolved into book lung and spinnerets of spiders as the wings of flying insects. And as the gills to spinnerets show, evolution doesn’t see that much of a difference between an innie and an outie.

    PZ had a piece on the gill to wing a while back:

    Flap those gills and fly!
    Monday, October 24, 2005

  18. Michael Buratovich says

    The formation of the tracheal system in Drosophila relies upon FGF signaling to not only initiate it, but also to maintain its branching. Maybe the same signal transduction pathway is used to make the internal duct, the urethra, through the genital tubercle to form the mature penis.

  19. fnxtr says

    Great stuff, PZ.
    On PT there’s an explanation that a whole series of genes are named after famous hedgehogs, Sonic being the most prominent.

    Just wondering where one would find the Spiny Norman gene. :-)

  20. Lizard Face says

    Why does your blog disable the freaking back button? What the hell. Horrible “feature”.

  21. vincent says

    how horrifying. Is this science?
    PZ open your eyes, the entire rationale of this is rubbish.
    Evolution does not “reuse” things etc.
    Development is a matter of extension and fold of a mass of cells, with whatever gene content it may have. The simple fact that each time you search of shh, fgf’s or wnt or whatever, in one place or another place you find them, shows that it is not genetics that determines animal shapes and appendages forms, it is the physical context.
    With exactly the same genes, you make this (a limb), or that (a tail, a penis), depending on the boundary conditions and the previous developmental history. All what you say is obsolete and misleading.The phrase “switching on a protuberance pathway in a new place” has a scientific content equal to ZERO.

    It is neither true that limbs grow at locations defined by an axial hox code.

  22. Owlmirror says

    John, I do believe that you have been tricked into following up a spammer.

    More interesting is vincent @#28. Could that possibly be vincent ( swirly !! ) fleury?