Dissociation: Heterochrony and allometry

Not all parts of the embryo are connected to one another. One can dissect out the limb field of a salamander neurula, for example, and the eyes are not affected. By means of mutation or environmental perturbation, one part of the embryo can change without the other parts changing. This modularity of development can allow changes that are either spatial or temporal.

Heterochrony is a shift in the relative timing of two developmental processes from one generation to the next. In other words, one module can change its time of expression relative to the other modules of the embryo. We have come across this concept in our discussion of neoteny and progenesis in salamanders (see Chapter 18). Heterochrony can be caused in different ways. In salamander heterochronies in which the larval stage is retained, heterochrony is caused by gene mutations in the ability to induce or respond to the hormones initiating metamorphosis. Other heterochronic phenotypes, however, are caused by the heterochronic expression of certain genes. The direct development of some sea urchins involves the early activation of adult genes and the suppression of larval gene expression (Raff and Wray 1989). Thus, heterochrony can “return” an organism to a larval state, free from the specialized adaptations of the adult. Heterochrony can also give larval characteristics to an adult organism, as in the small size and webbed feet of arboreal salamanders (Figure 22.17) or the fetal growth rate of human newborn brain tissue (see Chapter 12).

Figure 22.17

Salamander heterochrony. The progenesis of limb development in Bolitoglossa can create a tree-climbing salamander. (A) Foot of an adult B. rostratus, a terrestrial salamander. (B) Foot of an adult B. occidentalis, an arboreal salamander. The feet, skull, (more...)

Another consequence of modularity is allometry. Allometry occurs when different parts of an organism grow at different rates (see Chapter 1). Allometry can be very important in forming variant body plans within a phylum. Such differential growth changes can involve altering a target cell's sensitivity to growth factors or altering the amounts of growth factors produced. Again, the vertebrate limb can provide a useful illustration. Local differences among chondrocytes cause the central toe of the horse to grow at a rate 1.4 times that of the lateral toes (Wolpert 1983). This means that as the horse grew larger during evolution, this regional difference caused the five-toed horse to become a one-toed horse. A particularly dramatic example of allometry in evolution comes from skull development. In the very young (4- to 5-mm) whale embryo, the nose is in the usual mammalian position. However, the enormous growth of the maxilla and premaxilla (upper jaw) pushes over the frontal bone and forces the nose to the top of the skull (Figure 22.18). This new position of the nose (blowhole) allows the whale to have a large and highly specialized jaw apparatus and to breathe while parallel to the water's surface (Slijper 1962).

Figure 22.18

Allometric growth in the whale head. An adult human skull is shown for comparison. The whale's upper jaw (maxilla) has pushed forward, causing the nose to move to the top of the skull. (The premaxilla is present in the early human fetus, but it fuses (more...)

Allometry can also generate evolutionary novelty by small, incremental changes that eventually cross some developmental threshold (sometimes called a bifurcation point). A change in quantity eventually becomes a change in quality when such a threshold is crossed. It has been postulated that this type of mechanism produced the external fur-lined “neck” pouches of pocket gophers and kangaroo rats that live in deserts. External pouches differ from internal ones in that they are fur-lined and they have no internal connection to the mouth. They allow these animals to store seeds without running the risk of desiccation. Brylski and Hall (1988) have dissected the heads of pocket gopher and kangaroo rat embryos and have looked at the way the external cheek pouch is constructed. When data from these animals were compared with data from animals that form internal cheek pouches (such as hamsters), the investigators found that the pouches are formed in very similar manners. In both cases, the pouches are formed within the embryonic cheeks by outpocketings of the cheek (buccal) epithelium into the facial mesenchyme (Figure 22.19). In animals with internal cheek pouches, these evaginations stay within the cheek. However, in animals that form external pouches, the elongation of the snout draws up the outpocketings into the region of the lip. As the lip epithelium rolls out of the oral cavity, so do the outpockets. What had been internal becomes external. The fur lining is probably derived from the external pouches' coming into contact with dermal mesenchyme, which can induce hair to form in epithelia (see Chapter 12). Such a pouch has no internal opening to the mouth. Indeed, the transition from internal to external pouch is one of threshold. The placement of the evaginations anteriorly or posteriorly determines whether the pouch is internal or not. There is no “transitional stage” having two openings, one internal and one external.^* One could envision this externalization occurring by a chance mutation or set of alleles that shifted the outpocketing to a slightly more anterior location. Such a trait would be selected for in desert environments, where dessication is a constant risk. As Van Valen reflected in 1976, evolution can be defined as “the control of development by ecology.”

Figure 22.19

Transverse section through the anterior region of a pocket gopher (Thomomys) embryo, showing the anterior opening of the pouch (AP) and the continuity at this stage between the pouch and the buccal cavity (BC) across the developing lip area. (EP, epithelial (more...)

Duplication and divergence

Modularity also allows duplication and divergence. The duplication part of this process allows the formation of redundant structures, and the divergence part allows these structures to assume new roles. One of the copies can maintain the original role while the others are free to mutate and diverge functionally. This can happen at numerous levels. The Hox genes, TGF-β family genes, MyoD family genes, and globin genes each probably started as a single gene that duplicated several times. After the duplication, mutations caused the divergences that gave the members of each family new functions. At the tissue level, one sees duplication and divergence in the somites that give rise to the cervical, thoracic, and lumbar vertebrae.

Co-option

No one structure is destined for any particular purpose. A pencil can be used for writing, but it can also be used as a toothpick, a dagger, a hole-puncher, or a drumstick. On the molecular level, the gene engrailed is used for segmentation in the Drosophila embryo, is used later to specify its neurons, and is used in the larval stages to provide an anterior-posterior axis to imaginal discs. Similarly, a protein that functions as an enolase or alcohol dehydrogenase enzyme in the liver can function as a structural crystallin protein in the lens (Piatigorsky and Wistow 1991). In other words, preexisting units can be co-opted (recruited) for new functions. Sometimes, whole pathways are co-opted from one system to another. For instance, the pathway by which Hedgehog protein induces Engrailed protein to pattern and extend the insect wing is later used within the wing blades to make the eyespots of butterflies and moths. Distal-less, another protein used to extend the wing imaginal disc, is later used to form the center of such eyespots (Figure 22.20). Co-option can also be seen on the morphological level. Wings have evolved three times during vertebrate evolution, and in each case, different forearm structures were modified for an entirely new function. A structure originally used for walking has been recruited into a structure suitable for flying.

Figure 22.20

Recruitment of pathways in the formation of butterfly eyespots. (A) The interaction between Hedgehog (green) and Engrailed (blue) is first seen in the insect blastoderm, and is then modified during the creation of a new structure, the wing. In the wing (more...)

A famous case of co-option is the use of embryonic jaw parts in the creation of the mammalian middle ear, as explicated in Chapter 1 (see Figure 1.14; Gould 1990). First, the gill arches of jawless fishes became the jaws of their descendants; then, millions of years later, the upper elements of the reptilian jawbone became the malleus and incus (hammer and stirrup) bones of the mammalian middle ear (Figure 22.21).

Figure 22.21

Evolution of the mammalian middle ear bones from the reptilian jaw. The quadrate and articular bones of reptiles were part of the lower jaw. Sound could be transmitted from these bones via the large stapes. When the dentary bone grew and took over the (more...)

Scientists are looking at co-option in the formation of novel evolutionary structures. For instance, the carapace (dorsal shell) of the turtle is an evolutionary novelty that appears to form in a manner reminiscent of limbs. There is even a carapacial ridge that organizes the mesenchyme much like the apical ectodermal ridge of the limb bud (Figure 22.22; Burke 1989a). The bones themselves appear to form in the manner of skull bones. It is possible that certain developmental pathways (those used to form the limbs and those used to form the skull bones) have been recruited to form this new structure. The existence of discrete developmental modules allows the principles of dissociation, duplication and divergence, and co-option to form new types of organisms.

Figure 22.22

Midtrunk cross section through the embryo of the turtle Chelydra serpentina. (A) The carapacial ridge (arrowhead) is formed at the boundary of the somitic and lateral plate mesoderm and now represents the dorsal-ventral boundary. The thickened mesodermal (more...)

Footnotes

*

The lack of such transitional forms is often cited by creationists as evidence against evolution. For instance, in the transition from reptiles to mammals, three of the bones of the reptilian jaw became the incus and malleus, leaving only one bone (the dentary) in the lower jaw (see Chapter 1 and below). Gish (1973), a creationist, says that this is an impossible situation, since no fossil has been discovered showing two or three jaw bones and two or three ear ossicles. Such an animal, he claims, would have dragged its jaw on the ground. However, such a specific transitional form (and there are over a dozen documented transitional forms between reptilian and mammalian skulls) need never have existed. Hopson (1966) has shown on embryological grounds how the bones of the jaw could have divided and been used for different functions, and Romer (1970) has found reptilian fossils wherein the new jaw articulation was already functional while the older bones were becoming useless. There are several species of therapsid reptiles that had two jaw articulations, with the stapes brought into close proximity with the upper portion of the quadrate bone (which would become the incus).