Modern Developmental Genetic Discoveries | Evolution | Biology

The eyes of insects and the eyes of vertebrates were, until the early 1990s, considered to be a standard example of “analogous” structures. They perform the same function but have utterly different internal structures, suggesting that they evolved independently from a common ancestor that lacked eyes.

Then the laboratory of Walter Gehring in Switzerland began to research genes that are crucial for eye development in fruitflies and mice. One gene, ey, was known to be needed in fruitflies; another gene, Pax6, was needed in mice.

The sequences of the two genes turned out to be similar, suggesting that they are really the same (that is, homologous) gene. The ey gene could be shown to cause eye development in fruitflies, because if the gene is switched on in inappropriate parts of the body, such as a leg, it induces the development of an “ectopic” eye.

Then genetic tricks were used to introduce the fruitfly ey gene into mice. These mice grew up with fly-type compound eyes. It seems that the same gene is used in both mice and fruitflies to cause eye development. If the insect and vertebrate eyes have evolved independently, we would hardly expect them to have hit on the same gene to act as the master gene of eye development.

Two interpretations are possible. One is that the common ancestor of fruit flies and mice had eyes. The structure of insect and vertebrate eyes are still so different that they probably evolved independently, but perhaps from a common ancestor that had, rather than lacked, eyes.

The eye in that common ancestor might have been a much simpler structure, but there would be an element of homology between the insect and vertebrate eyes. The evolution of eyes in the two taxa would have been easier if they already possessed the developmental genetic machinery for specifying something about eye development.

Alternatively the homology may be more abstract- ey/Pax6, or the ancestral gene from which they evolved, might have specified some activity only in a particular location in the body (the top front of the head). Then the use of the same gene in mice and fruitflies would reflect only the fact that the two animals grow eyes in a similar body region.

The common ancestor of mice and fruitflies had a head, and would have had genes to work in the regions of the head. It would be less remarkable if mice and fruitflies have homologous genes for controlling development in a particular region of the head, than if they have homologous genes for developing eyes. At some level, homology must exist between mice and fruitfly eyes; the question is whether the homology is at the level of eyes, or head regions.

In general, structures that are not homologous at one level will be homologous at another, more abstract level. Ultimately, this reflects the fact that all life on Earth traces back to a common ancestor near the origin of life. Consider the wings of birds and bats. As wings, they are not homologous.

They evolved independently from a common ancestor that lacked wings. But as forelimbs, they are homologous. Bird wings and bat wings are modified forelimbs, descended from a common ancestor that possessed forelimbs.

Since the Gehring lab’s work on eyes, several other structures that had been thought to be analogous rather than homologous in insects and vertebrate have been found to have common genetic control. Some of these structures may turn out to be homologous in a specific sense, others only in an abstract sense. We shall not know which until the actions of the genes concerned are better understood.

Hox Gene Complex:

Are changes in the developmental genes associated with major evolutionary changes in the history of life? The Hox genes are the most hopeful gene set for answering this question at present.

More is known for the Hox genes about which genes are present in which animal taxa than is known for any of the other genes associated with development. We mainly know about the number of Hox genes in different taxa, and can therefore look at when in animal evolution the numbers of Hox genes changed.

The Hox gene complex clearly expanded at two points in the phylogeny. One is near the origin of the triploblastic Bilateria. Cnidaria have radial symmetry and only two cell layers. They are simpler than the other animal groups, which have three-cell layers and bilateral symmetry.

Only two Hox genes have been found in Cnidaria, against a common set of at least seven Hox genes in Bilateria. Probably the number of Hox genes went up by about five some time near the origin of the Bilateria.

A second major expansion occurred near the origin of the vertebrates. Invertebrates have a single set of up to 13 Hox genes. This set is also found, in a single copy, in the closest relative of the vertebrates, the lancelet Amphioxus. Vertebrates, including humans, have four copies of the 13-gene set.

The Hox gene set was increased fourfold, perhaps in a series of duplications, during the origin of vertebrates. Some biologists have explained the fourfold increase in the Hox genes by Ohno’s hypothesis that the genome as a whole was duplicated twice near the origin of the vertebrates.

Ohno’s hypothesis is not well supported, but even if the genome as a whole was not tetraploidised, the Hox gene set itself was. So also were some other sets of genes that operate in development. This increase in gene numbers may have contributed to the evolution of vertebrates.

Vertebrates are arguably more complex life forms than invertebrate animals; for one thing that they have more cell types. Also, many biologists think that the anatomic complexity of vertebrates is greater than for invertebrates. Complexity is difficult to measure objectively, but if vertebrates are more complex than invertebrates, the increase in the number of Hox genes may be part of the explanation.

Once life forms had evolved with extra Hox genes they may have become able to evolve, in the future, increased complexity. For instance, the number of Hox genes concerned with the posterior end of the body seems to have expanded in the origin of the deuterostomes.

The accuracy of inferences about when Hox gene numbers changed depends on the accuracy of the phylogeny. For example, Hox gene numbers appear to have decreased in the nematodes (represented by the worm Caenorhabditis elegans). This may be correct. However, the position of the nematodes in a group with the arthropods is based on recent molecular evidence from a small number of genes.

Traditionally nematodes belonged to a branch nearer the base of the tree, between the Cnidaria and the rest of the Bilateria. Then we should not infer that they have lost genes, but that they are an intermediate stage in the early increase from two to seven Hox genes. The inferences for these early events are uncertain, and in any case we require a well substantiated phylogeny before we can draw confident conclusions.

Changes in the Embryonic Expression of Genes are Associated with Evolutionary Changes in Morphology:

The vertebrae that make up the spine, or backbone, of a mouse differ from head to tail. For instance, the cervical vertebrae in the mouse’s neck differ in form from the thoracic vertebrae down the mouse’s back. The cervical and thoracic vertebrae also differ in other vertebrate animals, such as chicken and geese.

Geese and chickens have more neck vertebrae than mice do, and the division between cervical and thoracic vertebrae occurs further down the spine. The difference between species appears early in the embryo. The position of the boundary between cervical and thoracic vertebrae is further down the developing goose embryo than in a mouse embryo.

The boundary in the embryo between developing cervical and thoracic vertebrae is associated with the anterior boundary of expression of the Hoxc6 gene. The Hoxc6 gene is probably part of the control system that switches on the development of thoracic, rather than cervical, vertebrae.

Thus, an evolutionary change in the morphology of the spine was probably partly produced, at a genetic level, by a change in the spatial expression of the Hoxc6 gene in the embryo. Vertebrates develop in an anterior-posterior direction, with the head being specified first. A delay in switching on hox6c could cause the cervical-thoracic boundary to be shifted to the posterior, down the spine.

Changes in the timing of Hox gene expression can also contribute to morphological evolution. The five-digit limb of tetrapods, for example, has evolved from a fin in fish. Hox genes are expressed in two phases during the development of fish fins. These phases might, for instance, help to cause an outward growth of bones to form the fin.

In tetrapods, the Hox genes are also expressed in a third, later phase during limb development. The third phase is associated with the further growth outwards of the limb bones, to form the limb and hands. Thus, part of the mechanism by which fins may have evolved into limbs may have been for certain Hox genes to be switched on for a third time in the developing limb.

Morphological evolution may be caused by a change in which genes a Hox gene interacts with. For example, insects differ from some other arthropods in lacking legs on their abdomens. An insect has legs on its thorax and not its abdomen, but myriapods and many crustaceans have abdominal legs.

During evolution, leg development came to be switched off in the embryonic insect abdomen. The genetic mechanism, simplified, is that the Hox genes ultrabithorax (Ubx) and Abd-A are expressed down the abdomen of insects, crustaceans, and myriapods.

They are regional controllers of development. In insects, Ubx and Abd-A repress the gene distal-less (Dll); Dll is the gene that directs leg development. In myriapods and crustaceans, Ubx and Abd- A do not repress Dll.

Two hypotheses can explain events such as the loss of limbs from the insect abdomen. One is a change in a transcription factor such as Ubx. In the evolution of insects, Ubx may have changed such that it became able to repress the genes, such as Dll, controlling limb development.

The other hypothesis is that the enhancer of Dll may have changed during insect evolution. The enhancer may have ceased to bind Ubx. Alternatively, the enhancer may have continued to bind Ubx, but has changed its interaction with it such that Ubx now switches off limb development in the abdomen rather than switching it on.

Some evidence supports the first hypothesis. Crustacean Ubx is unable to repress Dll in fruitflies. That result suggests that Ubx itself has changed between crustaceans and insects. If Ubx were unchanged, crustacean Ubx should have the same effect in fruitflies as normal fruitfly Ubx.

Evolution of Genetic Switches:

The examples in the previous section illustrate how evolutionary changes in gene regulatory networks can underlie morphological evolution. In the hoxc6 example, in which the number of cervical vertebrae changed between mice and geese, the change concerned the regulatory relations between the hoxc6 gene and some higher control gene.

The anterior-posterior coordinates of the animal are probably given by a chemical gradient down the body. These chemicals may bind the enhancer of hoxc6, switching it off at some chemical concentrations and on at other concentrations. The hoxc6 gene is then switched on in a certain region of the body.

Morphological change can be produced if the enhancer of hoxc6 changes such that it is switched on and off at somewhat different concentrations of the chemicals that specify the anterior-posterior axis. In the example of insect abdominal legs, the change was in which other genes were regulated by Ubx and Abd-A.

Whether the changes in these examples came about by the exact genetic mechanisms suggested here is not important. Several kinds of change in an enhancer, or the molecules that interact positively and negatively with an enhancer, could produce the same general outcome.

What does matter, and is of broad interest, is that morphology can be altered by adding or subtracting switches that control existing genes. If a gene can cause, or help to cause, a leg to develop, then new legs can be added to (or old legs subtracted from) the body by switching the gene on or off. The gene may gain, or lose, an enhancer that binds to a transcription factor produced by one of the embryo’s regional- specifier genes.

It is instructive to compare evolutionary change produced by gain or loss of regulatory elements with change produced by sequence change in the gene itself. The sequence of a globin gene may change, for example, such that the oxygen-binding attributes of the hemoglobin molecule are altered. This is an obvious way for a molecule to change its function, and much functional change has likely been produced by sequence changes.

The importance of genetic switches may be more in the evolutionary addition of new functions. Brake field el al. and Keys et al. describe how a five-gene regulatory circuit has come to control the development of “eyespots” on the wings of butterflies. The gene circuit is able to produce borders, or boundaries, and is used in all insects to produce a certain boundary in the structure of the wing.

Most insect wings do not have eyespots but some butterfly wings do. The eyespot has a distinct circular shape, with a boundary at the edge. Eyespots probably evolved when this “boundary producing” gene circuit came to be expressed in a new gene network. In a butterfly eyespot, the boundary-producing genes are controlled by certain spatial-specifier genes within the wing, and they in turn control certain pigment- producing genes.

Thus, a pre-existing set of genes came to be expressed in a new circumstance, probably by changes in the enhancers of the genes concerned. The boundary-producing gene circuit had gained a new function.

When a gene adds an enhancer, which switches it on in a new circumstance, it can gain a new function without compromising its existing function. If a molecule, or morphological organ, changes to add a new function, it will usually perform its existing function less well. If a mouth is used for both eating and breathing, it is likely to do each less well than if it did one alone.

A molecule can add a new function by changes in its internal sequence, although this evolutionary process is inherently difficult. However, the molecule is also likely to perform its old function less well as it adds its new function. The difficulty is avoided if the new function is added by a change in gene regulation. The existing, unchanged gene comes to be switched on in new circumstances and the old function need not be compromised at all.