(Picture from here.)
I spoke of it a bit frivolously in the context of anti-evolution conversations. I didn’t give it the airtime it deserved.
So. Today, let’s talk about how drosophila, cephalopods, and human beings can develop from a zygote into vastly different adults using very similar mechanism. We’re going to talk about homeosis, the transformation of one organ into another.
Such a transformation has to happen as a structural change. In animals, this happens in animal development. There are additional structural changes in those animals that undergo metamorphosis. But I’m not down that path this time. Besides, I’m not at all sure how a caterpillar can turn into a sack of goo and then out of that become a butterfly.
The evolutionary introduction of development presents the opportunity for natural selection to effect structural changes. One description of development is to consider genes as musical notes and development the organization of those notes over time into a symphony. This is okay as a metaphor, if there’s no real conductor, the musicians perform different tunes at the same time when in different places, and also perform different tunes in the same place at different times.
All development starts with the sperm and egg. In vertebrates, the sperm donates certain key proteins along with a significant DNA package. But the egg is packed with pre-programming. In the zygote’s first few divisions, the cells are symmetric. Cells have been broken apart in this period and formed identical twins. One experiment I read about in college involved a frog egg that had been separated at the first division and grew into two embryos. In humans—and other mammals—twinning can happen up into the blastocyst stage.
At some point in early embryonic development, mitosis becomes asymmetric. Different cells have different destinies.
The morphological destiny of those cells is determined by the homeobox genes.
Such a gene is a fairly small DNA sequence, usually around 180 base pairs long, that is directly involved in determining structure—large scale anatomical features. There are homeobox genes that determine the head-to-tail organization: put the head here. The tail goes down there. There are homeobox genes that determine where legs grow. Antenna. Eyes. A protein derived from these genes can regulate the expressions of many target genes at different times, often by inducing long genetic cascades resulting in cells differentiating into target cell types: blood, neuron, bone cell, etc.
One of the best known homeobox sequence are the hox genes. These specify regions of the body such as head-to-tail access and locality of anatomical structures. Hox genes specify which appendage occurs where.
One of the interesting things about hox genes—and homeobox genes, in general—is how strongly they are conserved across animal divisions. The gene “eyeless” that creates eyes in Drosophila has been shown to be essentially the same as the gene Pax-6 in mice. (See here.) Modifying the genome, scientists have been able to cause fruit flies to create eyes all over their body—legs, arms, wings—depending on where the eyeless gene was placed. The Drosophila gene has been transplanted in frogs to produce extra eyes—showing that the same gene is causing the same behavior in a species evolutionarily separated by at least 400 million years.
Thus, a change in a homeobox gene either in content or in location can have drastic effects.
This has tremendous effect in evolution of the organisms that contain them. Genetics create the organism but natural selection generally doesn’t directly on the genes. Natural selection only reflects differential reproduction of the entire organism. Consider the protein collagen. It is found everywhere in the mammalian body from blood vessels to muscles to intestines. Any change to the genes that produce it would reflect in all of those components. Thus, a selection for, say, longer legs would never show up as a change in collagen. Such a change might help but not be attributable to it.
But, say, a gene set that expresses variability in leg length would expose that trait to selection. This would allow selection of a trait directly expressed by gene set. Or consider where eyes are located: in the front for primates and predators, on the side for prey. (Think cheetahs vs squirrels.) That location derives from when those homeobox sequences are expressed. Variability there exposes that genetic sequence directly to selection.
My point is that the creation of gene controlled development created the opportunity for selection of those traits we find most interesting: the size scale of dinosaurs, the leg loss of whales, the brain size of human beings.
It’s as close to a direct feedback mechanism from environment to genes as we’re going to find. So far.