Here's a prediction for you: the image below is going to appear in a lot of textbooks in the near future.
That's a technical tour-de-force: it's a confocal image of a Drosophila embryo, stained with 7 fluorescent probes against different Hox genes. You can clearly see how they are laid out in order from the head end (at the left) to the tail end (which extends to the right, and then jackknifes over the top). Canonically, that order of expression along the body axis corresponds to the order of the genes in a cluster on the DNA, a property called colinearity. I've recently described work that shows that, in some organisms, colinearity breaks down. That colinearity seems to be a consequence of a primitive pattern of regulation that coupled the timing of development to the spatial arrangements of the tissues, and many organisms have evolved more sophisticated control of these patterning genes, making the old regulators obsolete…and allowing the clusters to break up without extreme consequences to the animal. A new review in Science by Lemons and McGinnis that surveys Hox gene clusters in different lineages shows that the control of the Hox genes is much, much more complicated than previously thought.
Here, for instance, is a diagram of Hox arrangements in various animals.
Look first at the cephalochordate (on the right, about in the middle). It has the purest, simplest, cleanest arrangement: there are 14 Hox genes, all in one string, all in order, and all with the same orientation. This is also a perfect example of colinearity, with those genes expressed in a tidy front-to-back order in the animal's body. As you can see looking around the diagram, though, it's also an exception.
Take the fly, at the top left, for example. It's Hox cluster is broken in two pieces. Or look at the sea urchin—what a mess. It looks like part of the cluster has been swapped around. Vertebrates, represented by the mammals here, have kept everything in order, but they've duplicated whole clusters, and individual clusters have gaps. Most dramatically, look at the urochordate (this is Oikopleura, which I wrote about before: the cluster is completely broken up, the individual genes scattered throughout the genome with no apparent relationship to one another anymore.
Gene arrangements clearly tell us about the evolutionary history of these organisms, but if we want to understand how gene expression is translated into form, it's going to be a red herring in most cases. There has to be more going on.
One of those things going on, and also one of the current big buzzwords in molecular and developmental biology, is non-coding RNA. Scattered around and between the genes are sequences that are transcribed into RNA, but are not translated into protein—the RNA itself modulates the activity of other genes. In particular, there are microRNAs (miRNAs) interlaced throughout the Hox cluster, that in some cases have been shown to bind to and modify the expression of adjacent Hox genes. Our cartoons of ranks of colored arrows are going to have to be expanded to include squiggles of non-coding RNA, and more arrows showing which genes these RNAs modulate.
The summary of this article speaks for itself.
Genomic analyses have revealed surprising diversity in Hox gene number, organization, and expression patterns in different animals. There are still many animal groups about which little genomic sequence is known, and it remains to be seen how much more variation in Hox gene organization and function will emerge, including the numbers and functions of non-protein-coding RNAs. The property of HOX proteins working as a loosely coordinated system, often with overlapping patterns of expression and function, has apparently fostered their abilities to contribute to morphological change during the evolution of animals. Their colinear arrangement and coordinated regulation in many animals may assist in the maintenance of their overlapping expression patterns. This may have allowed some members of the clusters to subtly and slowly alter their expression patterns and functions to drive groups of cells toward novel structures. But Hox genes still can work as an axial patterning system even when partially dispersed in the genome, and dispersal may foster their rate of functional evolution.
I think the key concept is that Hox genes form a loose network, with multiple factors — the temporal sequencing of colinearity, transcription factors, and miRNAs — that all work together to generate form from genes, and that while this may look like a daunting mess, the complexity of regulation actually facilitates evolutionary change.
Lemons D, McGinnis W (2006) Genomic evolution of Hox gene clusters. Science 313:1918-1922.
Wow! Just WOW! Beautiful!
Who did the septuple in situ?
Whoa. I am utterly amazed, inspired and flummoxed by the sheer complexity and complicatedness that is a developing organism. What a trip!
Someone needs to turn that in situ hybridization image into a poster. My office is crying out to have it.
McGinnis never ceases to amaze.
Crikey, that is a beautiful picture. I think in a million years I could never have developed the technical skill to produce something like that.
Re Dave Carlson's comment: I wish the people who tote their little gods and little medieval universes around with them could somehow be gotten to understand how fascinating, beautiful, immense, bizarre and awe-inspiring the actual universe is...
PZ, can I tell you again how I love your posts?
You can explain the evolutionary waltz in ways that are beautiful and transparent. Though I have a degree in journalism, I took enough biology, etc. courses that if, at the time I took them, U. Kan. gave minors I'd have gotten one. And I try and keep up as well as I can (I'm more interested in natural history and the evolutionary waltz species do among themselvesy to survive and expand).
You make me want to come up, pay and take a course. Though right now that's pretty much impossible (home, debts, day job that pays off these things....)
I am also really enjoying the zebrafish embryo photography.
Keep it up.
Is there a means of determining the possibility of how Hox genes may be influenced by horizontal gene transfer [HGT] as postulated by Carl Woese?
A PNAS paper 'Endogenous retroviruses regulate periimplantation placental growth and differentiation' by KA Dunlap, et al, appears to provide supporting evidence for HGT.
"... Circumstantial evidence suggests that ERVs [Endogenous retroviruses] play a role in mammalian reproduction, particularly placental morphogenesis, because intact ERV envelope genes were found to be expressed in the syncytiotrophoblasts of human and mouse placenta and to elicit fusion of cells in vitro ..." from the abstract.
Wonderful post. I look forward to reading the Science article.
The spatial and temporal co-linearity of mammalian HOX gene expression is too complete to be purely an evolutionary relict. There appears to be an LCR driving the expression of a subset of HOX-D genes in the appendicular segments. I wonder if there is an analogous LCR for each HOX cluster driving expression in the axial segments. This could serve to enforce linkage.
That. Is. Amazing.
the control of the Hox genes is much, much more complicated than previously thought
Why would anyone think it wouldn't be complicated???