Endless Forms Most Beautiful

Carroll, Sean B.

Evo Devo can trace the modifications of structures through vast periods of evolutionary time—to see how fish fins were modified into limbs in terrestrial vertebrates, how successive rounds of innovation and modification crafted mouthparts, poison claws, swimming and feeding appendages, gills, and wings from a simple tubelike walking leg, and how many kinds of eyes have been constructed


HUXLEY, MAN’S PLACE IN NATURE (1863) Symmetry and Polarity In


HUXLEY, MAN’S PLACE IN NATURE (1863) Symmetry and Polarity In addition to the repetition of modular parts, animal bodies and body parts usually display two additional features—symmetry and polarity. Most familiar animals are bilaterally symmetrical in that they have matching right and left sides with a central axis of symmetry running down the middle of the long axis of the body. This design also imposes a front/rear orientation to animals and has enabled the evolution of many efficient modes of locomotion. Some animals exhibit other symmetries, such as the pentaradial (five-fold) echinoderms,


The axes of symmetry in an animal are clues to how the animal is built.


test whether the first two cells of a newt embryo had similar or different properties. Spemann used a fine baby’s hair, taken from his own daughter, to tie off and separate individual halves of the embryo. The cells on each side of the knot gave rise to normal newt tadpoles, demonstrating that the two halves of early amphibian embryos could give rise to two entirely identical animals.


When Spemann divided the egg differently by tying it perpendicular to the furrow between the two cells of the embryo, he obtained a dramatically different result. Only one side made a normal tadpole, while the other made a disorganized mess of belly tissue.


Polydactyly is known widely throughout vertebrates, especially in cats, mice, and chickens. It is striking that similar digit patterns can occur in different animals, including humans, and they can be induced by experimental manipulations or inherited. This suggested that there could be some mechanisms in common for generating extra human fingers and chicken digits.


The relationship between DNA, RNA, and protein is that DNA is a template for the making of RNA, and RNA is in turn the template for the making of proteins.


Each very long molecule of DNA in a cell is a chromosome.


Each gene occupies its own interval along a DNA molecule.


DNA is made of two strands of building blocks called nucleotides.


Each nucleotide contains one of four distinct bases; these are abbreviated A, C, G, and T.


The first step in decoding the information in a gene is called transcription, which involves making a single-stranded “messenger RNA” (mRNA) transcript of the gene that is complementary to one strand of the DNA molecule.


In the second step, the mRNA is then directly decoded into a protein sequence in a process called translation (figure 3.2). This involves a universal genetic code to translate RNA sequences into protein sequences.


Proteins are made up of building blocks of amino acids that are linked together in long chains. There is a direct correspondence between the sequence of bases in DNA and the sequence of amino acids in proteins.


The sequence of amino acids in each protein determines their individual shape and chemical properties, whether they carry oxygen, form muscle fibers, or break down lactose and so forth.


A genetic switch controls beta-galactosidase production and lactose metabolism in E. coli. In the absence of lactose, the lac repressor binds to the switch and represses gene transcription. In the presence of lactose, the repressor falls off the switch, transcription and translation occur, and the enzyme is produced.


Interestingly, the genes sat close together in two clusters. One cluster, the Bithorax Complex, contained three genes that affected the back half of the fly; the other, the Antennapedia Complex, contained five genes that affected the front half of the fly. Even more provocative, the relative order of the genes in these two clusters corresponded to the relative order of the body parts they affected


No biologist had even the foggiest notion that such similarities could exist between genes of such different animals. These Hox genes were so important that their sequences had been preserved throughout this enormous span of animal evolution.


Aniridia is the same gene known as Small eye in mice that when mutated caused the reduction or elimination of eye formation. This discovery was intriguing and provocative because our camera-type eyes and the fly’s compound eyes are very different in structure and adapted for very different needs. Why would the same gene be involved in the formation of such different kinds of eyes? Was this a fluke or a hint of something deeper?


The second experiment was to introduce the mouse Small eye gene into flies so that it was also turned on in weird places in the fly. What do you think happened? The result was the same as the experiment with the fly gene—fly tissues were induced to form eye structures. However, it is important to emphasize that the tissues formed were fly eye structures, not mouse eye structures.


Since these animals are all members of one phylum and share a common jointed limb design, the use of Dll in all these species made sense.


several mammalian versions of tinman were discovered, with the not quite so magical name


We found that the formation of all sorts of things that stuck out of animal bodies were associated with use of the Dll gene. These included chicken legs, fish fins, the appendages of marine worms (called “parapodia”), the ampullae and siphons on sea squirts, and even the tube feet on sea urchins. This was another example, like Pax-6, of a tool kit gene involved in building vastly different structures that only share, at most, the common feature of projecting away from the main body.


The four-limbed pattern of vertebrates is ancient and there are many general similarities to limb development among all vertebrates


The forelimbs arise at different somite numbers in different vertebrates, but are always found at the cervical/thoracic boundary.


The buds, while initially very small, are three-dimensional and possess three axes along which the top (back) and bottom (palm), front (thumb) and back (pinkie), and proximal (e.g., shoulder) and distal elements (digits) of the limb will form as the bud grows dramatically.


In a cookie-cutter-like fashion, the interdigital tissue is carved away, leaving the digits.


The generation of regular spacing patterns. In an initially uniform field of cells (first panel), two cells begin to differentiate (black circles, second panel) and inhibit cells in contact with them from doing so. Cells in other regions begin to differentiate and inhibit their nearest neighbors


All of these patterns are generated locally by cell interactions, not specified by global coordinates.


the central idea is that animal forms evolve through changes in embryo geography. We will learn the specifics of how geography and form evolve by changing the way the tool kit genes are used. Evolution of form is very much a matter of teaching very old genes new tricks!


Several kingdoms preceded our own—bacteria, archaea, protists, and fungi (land plants came later than animals although their forerunners, green algae, predate the animals).


Insects and vertebrates are representatives of the two different main branches of the animal tree.


(in sea urchins and other echinoderms, larvae are bilaterally organized, even though the adults show various radial symmetries).


It turns out that the different numbers of Hox clusters in different vertebrates reflects differences in the overall size of the whole genetic tool kit.


Not only were Hox gene clusters duplicated in vertebrate evolution, but many other kinds of genes in the tool kit were as well.


in vertebrates the transition from one vertebral type to another—cervical/thoracic, thoracic/lumbar, lumbar/sacral, sacral-caudal—corresponds to transitions between the zones of expression of particular Hox genes.


the evolution of body forms in two of the most successful and diverse groups of animals—arthropods and vertebrates—has been shaped by similar mechanisms of shifting Hox genes up and down the main body axis.


We can begin to think of individual groups—insects, spiders, and centipedes, or birds, mammals, and reptiles, as well as their long extinct fossil relatives—not so much in terms of their uniqueness, but as variations on a common theme.


Different vertebrates have different numbers of neck vertebrae,


The boundary between neck and trunk vertebrae is marked by expression of the Hoxc6 gene in all cases, but the position differs in each animal relative to the overall body. The forelimb arises at this boundary in all four-legged vertebrates; in snakes this boundary is shifted far forward to the base of the skull and no limbs develop.


Changing the sequence of switches allows for changes in embryo geography without disrupting the functional integrity of a tool kit protein.


Vertebrates came to land on the modified pectoral and pelvic fins of fish ancestors. Although they had just two pairs of limbs to work with, they have taken to the air three separate times to give rise to new types of animals (pterosaurs, birds, and bats), returned to the water many times (whales and dolphins, seals,


Vertebrates came to land on the modified pectoral and pelvic fins of fish ancestors. Although they had just two pairs of limbs to work with, they have taken to the air three separate times to give rise to new types of animals (pterosaurs, birds, and bats), returned to the water many times (whales and dolphins, seals, etc.), and evolved all sorts of limbs for making way on land.


The importance of serially repetitive body design is the ability to shift the burden of some task from two or more pairs of structures onto fewer structures, then to specialize the freed-up structures for new purposes.


the same organ often performs wholly distinct functions at the same time, and that two distinct organs may also simultaneously contribute to the same function. This multifunctionality and redundancy create the opportunity for the evolution of specialization through the division of labor. The availability of duplicate structures enables animals to “have their cake and eat it too” or, more accurately, to “have their limbs while learning to eat with them too.”


Many different appendages co-occur in individual species that provide both tool-like dexterity as well as savage weaponry. Take a look at all of the different implements a humble crayfish carries (figure 7.2)—there are more gizmos on this one animal than on a deluxe Swiss Army knife.


The most probable scenario is that Apterous and Nubbin were used in making respiratory lobes in an aquatic crustacean ancestor of insects and have stayed on the job ever since, as this branch evolved into wings (and, as we will see later, other structures in different animals).


The most likely scenario for wing origins is that the wings of adult terrestrial insects evolved in animals that also had gills in their larval stages. Wings could then evolve as adult structures modified from gills, without dispensing with gills at all.


Mayflies and dragonflies develop from immature aquatic nymphs that have gills on their abdomens, and these are the most primitive winged insects.


adult mayflies and dragonflies are entirely different animals than their aquatic young, living in entirely different environments. The adaptation to these different environments has taken place simultaneously in one genome, by separating developmental programs for building the nymph from those for building the adult.


The evolution of radically different larval and adult morphologies is a pervasive theme in animals (think of caterpillars and butterflies, or bilateral echinoderm larva and pentaradial adults).


Three separate times—in pterosaurs, birds, and bats—the tetrapod forelimb has been remodeled into a wing for powered flight.


In pythons and boas, a vestigial remnant of the hindlimb still forms, but the forelimb does not.


Pythons and boas do develop a small hindlimb spur near the cloaca; more recent families of snakes do not. It is likely that full limblessness in these snakes is due to failure of hindlimb formation at an even earlier stage.


If you have ever grabbed a moth in your hand or between your fingers, you have noticed the “dust” residue—these are scales. The easy detachment of these scales is an advantage for these large-winged animals in freeing themselves from sticky places, such as spiderwebs.


Our logic in tackling the making of butterfly wings was to rely on the evolutionary relationships among insects. Since insect wings evolved just once, then what we knew about the making of fruit fly wings should apply, in general, to the making of butterfly wings.


The Heliconius butterflies of Central and South America display warning colors, especially reds and yellows, that advertise they are unpalatable. Mimicry occurs among different geographic populations of Heliconius butterflies. Different species in a given geographic region converge on a similar wing pattern, but different geographic populations of these same species may display different patterns.


how much of an advantage is sufficient in order that a selection will favor these individuals?


The short answer to the question “How much of a difference matters?” is that natural selection requires a surprisingly small difference in relative success between two forms to work. This difference may often be imperceptible or unmeasurable in the field, but is sufficient to favor the evolution of one form over another. Population geneticists have developed


The short answer to the question “How much of a difference matters?” is that natural selection requires a surprisingly small difference in relative success between two forms to work. This difference may often be imperceptible or unmeasurable in the field, but is sufficient to favor the evolution of one form over another.


Let’s close this chapter by returning to the zebra debate using the same logic I just applied to spotted rock pocket mice. In deciding the value of stripes, isn’t it worth thinking about why all those zebras we see are striped? If it didn’t matter, wouldn’t we see lots of zebras without stripes?


Fur coloration mutants are sufficiently common in mammals that dramatic mutants (e.g., white tigers or spotted zebras) occur at some rare frequency in the wild.


It is crucial to realize that our own species has been around for only a tiny fraction (about 3 percent) of the total time span of hominin evolution. Most of the physical evolution of interest predated the origin of H. sapiens.


These astounding prints were made by at least two individuals, one large, one small, who were walking through a fresh ash fall 3.6 million years ago.


While bipedalism and its associated features evolved early in our lineage, our large brains did not.


Why did our brains get so much bigger during these periods? There are many theories. I’ll mention just one, the adaptation to climatic change, because I think it reflects a view that is becoming more widely accepted about the role of external forces in driving the pace of evolution.


The brain is a very expensive organ in terms of energy consumption, drawing up to 25 percent of an adult human’s energy (and 60 percent of an infant’s).


The tenrec, an insect-eating mammal, has a much smaller cerebral cortex than the marmoset, a primate. Relative shifts in the size of brain regions are a common feature of specializations.


The stuff of our evolution is more likely to be found in the “microanatomy” of our brain, including the interconnections between cortical regions, the architecture of local wiring circuits, or the arrangement of neurons in the cortex.


Human babies have less mature skulls in terms of their shape than do young chimpanzees, even though human skulls and brains are much larger. In humans, the maturation of the skull is slowed dramatically compared with the chimp, which allows for its greater initial size. Chimpanzee and human skulls eventually grow to the same size, but attain very different face sizes and brain case volumes. The relative shift in skull maturation rates indicates that the timing of similar developmental processes has been shifted.


Paleontologists can tell from enamel patterns on fossil teeth that tooth formation times were shorter in Australopithecines and early Homo species than in modern humans. The stages of dental development are reliable indicators of stages of juvenile development and the relative age of sexual maturation.


The human MYH16 gene is expressed in the human temporalis muscle, but the mutation in the gene has inactivated the protein’s function. The muscle fibers in the human temporalis are only about one-eighth the size of those of the macaque.


Based upon the number of changes in the human gene relative to other species, the Pennsylvania group has estimated that the inactivating mutation occurred somewhere between 2.1 and 2.7 million years ago. This is tantalizingly close to the period of the origin of the genus Homo.


The significance of the evolutionary reduction in jaw musculature extends beyond how hominins chewed their food. Muscle anatomy has a large influence on bone growth, and experimental studies have shown that jaw muscle growth has a significant impact on the size and shape of the craniofacial skeleton. Reduction in the jaw musculature, and the force imposed on the mandible, would reduce the stress on bones in the skull. This could have allowed the braincase to become thinner and larger.


Furthermore, the reduction in jaw musculature may have facilitated the eventual evolution of finer control of the mandible, as is required for speech.


All of these connections and associations are intriguing, but we must be cautious not to attribute all of this anatomical change to a single mutation.


we cannot say whether this inactivating mutation was the initial genetic change toward reduction of the temporalis, or one of many sequential or parallel changes, or the last change that occurred once the role of the MYH16 protein in the temporalis became dispensable.


FOXP2 is not at all unique to humans. The gene has been identified in a bunch of primates, rodents, and a bird.


the human FOXP2 protein differs at only 4 out of 716 positions from that of the mouse, at 3 positions from the FOXP2 of the orangutan, and at just 2 positions from those of the gorilla and chimpanzee. This is a smaller amount of change in sequence than most proteins show, indicating that there has been a lot of pressure to conserve the FOXP2 protein sequence throughout mammalian evolution.


what is called a “selective sweep.” The action of natural selection can leave a trail of evidence in the form of the pattern of DNA sequence variation that arises after a favorable mutation is selected for. Variation in a length of DNA sequence accumulates as a function of time unless or until selection acts to favor a particular variant. Selection for a variant causes a “sweep” that reduces overall variation.


From the pattern of reduced variation at a gene relative to its neighbors, geneticists can tell if a gene has experienced a selective sweep.


The signal of a selective sweep at the human FOXP2 locus is one of the strongest at any human gene. This is a good indication that for some period of the past 200,000 years, during the evolution of our species, mutations in the FOXP2 gene were favored and spread throughout H. sapiens.


The evolution of form is the main drama of life’s story, both as found in the fossil record and in the diversity of living species.