Did radial symmetry evolve twice?

Did radial symmetry evolve twice?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

In the history of animal evolution, both the Cnidarians and the Echinoderms have radial symmetry, according to this cladogram:


Do both groups having this feature mean radial symmetry evolved twice?

Good question! I had never really though about it, so thank you!

Echinodermata have a pentaradial symmetry

Echinodermata actually don't have a radial symmetry like jellyfish do. They have a pentaradial symmetry as they systematically have 5 arms.


Even if Echinodermata were radially symmetric, then it would actually be unlikely that bilateral symmetry evolved, was then lost and evolved again. It is much more parsimonious that bilateral symmetry evolved once and was later lost in the echinodermata lineage.

Developmental and Genetics of pentaradial symmetry

As one would expect modifications of hox genes are involved into the evolution of this peculiar body plan. From Lowe and Wray 1997

Here we report the expression domains in echinoderms of three important developmental regulatory genes ( distal-less, engrailed and orthodenticle ), all of which encode transcription factors that contain a homeodomain. Our findings show that the reorganization of body architecture involved extensive changes in the deployment and roles of homeobox genes. These changes include modifications in the symmetry of expression domains and the evolution of several new developmental roles, as well as the loss of roles conserved between arthropods and chordates. Some of these modifications seem to have evolved very early in the history of echinoderms, whereas others probably evolved during the subsequent diversification of adult and larval morphology.

Here is the whole abstract of Mooi and David 2008

The strangeness of echinoderm pentaradiality results from superposition of radial symmetry onto ancestral deuterostome bilaterality. The Extraxial-Axial Theory shows that echinoderms also have an anterior/posterior (A/P) axis developed independently and ontogenetically before radiality. The A/P axis is first established via coelomic stacking in the extraxial region, with ensuing development of the pentamerous hydrocoel in the axial region. This is strongly correlated with a variety of gene expression patterns. The echinoid Hox cluster is disordered into two different sets of genes. During embryogenesis, members of the posterior class demonstrate temporal, spatial, and genetic colinearity within the extraxial region. We suggest that displacement of genes from the more anterior Hox classes toward the 5' end of the chromosome leads to control of the later-developing, radially symmetric axial region. Genetic disorder is therefore another way of using colinearity to build the unique echinoderm symmetry.

As insinuated above, note that (in most echinodermata species at least) only adults have a radial symmetry. Larvea have a bilateral symmetry. Here are a bunch of picturesand schema.



It is actually hypothesized that radial symmetry did evolve again, but there are homologies between the echinoderm symmetry and bilateral animals:


The exact relationships between the different cnidarian groups are unknown. Among theories proposed on the evolution of the phylum Cnidaria, most treat the radial symmetry and tissue level of organization as evidence that the group is primitive (that is, it evolved before the evolution of bilateral symmetry) and hold that the medusa is the original body form, being the sexually reproductive phase of the life cycle. Another theory is that the original cnidarian was a planula-like organism that preceded both polyp and medusa. In either case, Hydrozoa is considered to be the most ancient of cnidarian classes, and Trachylina is thought to be the most primitive extant order of that group. An alternative view is that anthozoans are the stem of the phylum, which evolved from bilateral flatworms and is secondarily simplified. A corollary to this theory is that the polyp is the ancestral body form.

Speculations about the origin of the phylum are not easily resolved, for preservable skeletal structures developed relatively late in cnidarian evolution. The oldest fossilized cnidarians were soft-bodied. Representatives of all four modern classes have been identified in Ediacaran fauna of the Precambrian Period (that is, those appearing between about 635 million and 541 million years ago) known from more than 20 sites worldwide. As much as 70 percent of Ediacaran species have been considered to be cnidarians. Curiously, there are few fossil cnidarians of the Cambrian Period (541 million to 485.4 million years ago). The Conulariida, which existed from the Cambrian Period to the Triassic Period (251.9 million to 201.3 million years ago), are considered by some scientists to be skeletal remains of scyphopolyps, either ancestral to the coronates or without modern derivatives. Presumed fossil sea anemones are found in the lower Cambrian System. Colonies of Stromatoporoidea, considered to be an order of the class Hydrozoa that extended from the mid-Cambrian Period to the Cretaceous Period (about 145 million to 66 million years ago), produced massive skeletons. Although there were two groups of Paleozoic corals, neither of which has modern descendants, they were not great reef-builders during that era. Scleractinians arose in the mid-Triassic Period blue corals, gorgonians, millepores, and hydrocorals have records from the Jurassic Period (201.3 million to 145.0 million years ago) or the Cretaceous Period to the present. Most other cnidarians are known only from the Holocene Epoch (within the last 11,700 years).

Did internal transport, rather than directed locomotion, favor the evolution of bilateral symmetry in animals?

The standard explanation for the origin of bilateral symmetry is that it conferred an advantage over radial symmetry for directed locomotion. However, recent developmental and phylogenetic studies suggest that bilateral symmetry may have evolved in a sessile benthic animal, predating the origin of directed locomotion. An evolutionarily feasible alternative explanation is that bilateral symmetry evolved to improve the efficiency of internal circulation by affecting the compartmentalization of the gut and the location of major ciliary tracts. This functional design principle is illustrated best by the phylum Cnidaria where symmetry varies from radial to tetraradial, biradial and bilateral. In the Cnidaria, bilateral symmetry is manifest most strongly in the internal anatomy and the disposition of ciliary tracts. Furthermore, the bilaterally symmetrical Cnidaria are typically sessile and, in those bilaterally symmetrical cnidarians that undergo directed locomotion, the secondary body axis does not bear a consistent orientation to the direction of locomotion as it typically does in Bilateria. Within the Cnidaria, the hypothesized advantage of bilateral symmetry for internal circulation can be tested by experimental analysis and computer modeling of fluid mechanics. The developmental evolution of symmetry within the Cnidaria can be further explored through comparative gene expression studies among species whose symmetry varies.

Presentation of the hypothesis

Now we focus on the aquatic environment because bilateral symmetry (and animal life, itself) formed there, and had to be maintained there for millions of years, before bilaterians conquered the land. Let us start with the elementary physical fact that to locomote in a fluid, a body has to overcome drag (the resistance of the medium in which the body moves, acting in opposition to the direction of locomotion).

The magnitude of the drag force is:

where F is the drag force, ρ is the density of the medium, c is the dimensionless drag coefficient dependent on the body shape, A is the area of the maximal section of the body in the direction of motion, and v is the body’s velocity [9, 10]. The negative sign on the right side indicates that drag is opposite to the direction of motion. It is important to note that this equation is valid for situations where the viscous forces are negligible compared to inertial forces, in what is loosely described as the macroscopic world (i.e at high Reynolds numbers). In the microscopic world, the forces are dominated by the viscosity of the fluid rather than by the inertia (i.e. at low Reynolds numbers) [10], however a discussion of the locomotion in the micro scale world is not the concern of this paper (for an in-depth analysis see ref. [11]).

Given the fact that the medium imposes resistance on the body, if resistance forces are unequally distributed around the body, their resultant force will not be zero compared to the rectilinear direction (i.e. movement straight ahead), so the body will not move on a linear path. This is the case when a moving body is asymmetric. Thus, it follows that a directionally locomoting animal has to be symmetric in order to avoid this effect. To be able to move forward, the animal can have any type of symmetry, so the approach outlined here is not sufficient to explain the success of bilateral symmetry. Rectilinear motion is, however, not the only element of locomotion. One other important element is changing direction, the importance of which, in this regard, has been mostly ignored in the literature so far. A slight deviation from the straight trajectory can easily be obtained by flawing one element of symmetry, thus generating asymmetry in the original direction of motion. This can be achieved by any symmetrical body. However, when a quick changeover is required, the situation becomes very different.

In quick changes in direction, the body has to exercise a force in the opposite direction to the desired new orientation. This means that it has to have a “pushing” surface in water from which to depart in the new direction. This surface is formed by the water layer against which the body is standing in order to push itself away, and it is produced by creating a great instantaneous drag force. Since ρ in the equation is unchanged, and v is diminishing or constant, the animal has to increase the maximal surface A and/or the drag coefficient c.

We will now overview the main symmetry types in terms of their capacity to create a pushing-off drag force. Given that a swimming body has to minimize the overall drag, its skin friction [9], and thus its wetted area, has to be adequately reduced. Thus, only three main body forms can be considered: spherical (with endless symmetry planes and symmetry axes), cylindrical (with endless symmetry planes and one symmetry axis) and bilateral (with one plane of symmetry). An elongated radial body that shows a star-like section is suboptimal since it has a very large surface that is far from ideal for swimming forwards.

A spherically symmetrical body cannot generate the pushing surface, being of equal shape and drag in every direction. Since the forces – which are different from the one operating in the direction of its motion – acting on this body are all equalized, it will not be able to depart in a new direction. It can only rotate around itself to deviate to a small extent (as soccer players bend the ball), but this is hardly an effective changeover and obviously cannot guarantee accurate manoeuvrability (understood simply as the capacity to perform quick and accurate changeovers). In this context we can disregard how it was able to move directly in the first place.

A cylindrical (Figure 1.A) or approximately cylindrical (or radial) body locomoting with lateral or vertical undulation is able to increase A, which will be generated by a section of its body opposed to the direction in which it wants to move. The area of this surface is given approximately by the product of the diameter and the length of the body portion in question (and of course by its angular orientation to the axis of translation). If its lateral drag coefficient (c) is greater than the frontal, then when the animal turns its body can also increase c in the equation. However, regardless of the relationship between the anterior and lateral c, if the product (c A) from the lateral view is greater than that from the frontal one, this body will be able to move forward as well as to change direction.

Schematic representation of a cylindrical (A) and two bilateral bodies (B and C) generating pushing surfaces while changing direction. Denser grids indicate a greater drag force.

A bilateral body (Figure 1.B and C) can alter both coefficients A and c as well. Since it has only one plane of symmetry (in the main direction of the motion), vertically it can carry structures with an extended surface area. The lateral area (A) of the body will be further increased by these structures (just think of the vertically posed fins of a shark). Furthermore, equipped with these, and with the more or less flattened sides of the body, the animal can also greatly increase c. Since it is streamlined only from the frontal view, its lateral (or vertical if the animal is dorsoventrally flattened) drag coefficient is very high compared to the frontal one.

The flatter its sides – including the appendages – are, the greater its lateral drag coefficient will be as compared to that of a cylindrical body. And knowing that a rectangular plate has an approximately 50 to 70 % higher drag coefficient (depending on the height to length ratio) than a cylinder (at Re = 10 5 ) [9], we can say that bilateral symmetry offers the evolutionary possibility of increasing F by as much as 50 to 70 % compared to cylindrical symmetry, thanks simply to the drag coefficient. In other words, when a hypothetical cylindrical and a bilateral body have the same A (and frontal c), the bilateral body will enjoy a greater advantage in turning because it can produce a pushing force much greater than the cylindrical body because laterally it is less streamlined. Is this condition sufficient to assure a marked evolutionary advantage for bilateral symmetry? Since this capacity offers a very effective locomotion with potentially excellent manoeuvrability, we suggest that it is. Otherwise we would have to argue that effective locomotion is not a great advantage for an organism for whom a basic feature is precisely locomotion. Compared to a bilateral body, the cylindrical form has lower resistance in sideways movement, so the cylindrical body “slides” laterally in changeovers, as we do when we try to change direction on ice.

One could argue that a bilateral body can manoeuvre well only in left-right directions while a cylindrical body can, in theory, turn in every direction away from that of the motion. Bilateral animals, being not rigid objects (like ships or aircraft are), solve the problem simply by twisting the body and the appendages in the desired directions.

Based on the arguments explained so far it could be stated that a symmetry that is streamlined in only one direction, while non-streamlined in other directions, is favourable for manoeuvrable locomotion.

It is important to say that the changeover does not necessarily have to be drag-assisted. Some radially symmetrical animals, such as jellyfish, use asymmetric contractions of the bell, thus generating asymmetric jet flows to steer. However, the accuracy and the speed of this medusan-type manoeuvring [12] are much more modest than the drag-based manoeuvring of bilateral pelagic animals.

Bilateral symmetry has also proved to be succesful both on land and in the air. On land, the force-generating role of the drag in water is replaced by gravitation and so by the necessity of leaning on the land. In this regard, locomotion on land is analogous to that on the fluid–solid interface. This locomotion essentially occurs in two dimensions, thus, direction shift on land requires the body to be capable of turning left or right, and so of being supported from the right as well as from the left. The effectiveness of creeping locomotion has been improved by the evolution of limbs, which, placed on the two sides of the bilateral body, satisfy the above-mentioned condition. (For the sake of simplicity, we will not deal with the limbless evolution of snakes and limbless lizards here.)

Flying, similarly to swimming, requires the animal to create pushing surfaces in the air. The evolution of large-surface wings allowed the animals to locomote in a medium which, compared to water, has a lower density, and as a consequence, is almost completely lacking in the hydrostatic pressure that to a certain extent counterbalances the force of gravity in water.

The combination of bilaterality with the centralisation of the nervous system and cephalisation allowed the evolution of really successful body plans ensuring precise locomotion and rapid information processing.

WHY FIVE? Mysteries in Symmetry!

Thus, pentameral symmetry is a SECONDARY characteristic "on top of" the "basic" bilateral symmetry. It had to acquire this feature during its evolutionary development. And so when we ask WHY do they have pentameral symmetry? We are also asking HOW/WHY did this unusual symmetry evolve in echinoderms. Figuratively speaking, it adds an additional step to its evolution by appearing with this body form. What's the deal with that??

2. Are all adult echinoderms purely pentameral?
You may suddenly realize "AHA! I GOT YOU! SOME echinoderms show kind of BILATERAL SYMMETRY AS ADULTS!" Don't they.

Two notable exceptions: "irregular" urchins and sea cucumbers. Both are unusual in that most are detritivores or process sediment for food. Therefore requiring movement in one direction.

Bilateral symmetry is associated with directed movement and so, its presence is often associated with organisms which show some kind of single-directed motion.

What happens in "irregular urchins is that yet ANOTHER "symmetry" is overlain/"added" over the radial symmetry.. So, these animals go from bilateral (as larvae)

and eventually develop bilateral symmetry IN ADDITION to pentameral symmetry. This is called SECONDARY bilateral symmetry.

Sea cucumbers show bilateral symmetry (right and left sides) along their worm-like bodies. Presumably, again because they have a life mode which requires them to show directed movement in order to feed.

So, all adult modern echinoderms show SOME kind of 5-part or pentameral symmetry. Even if it doesn't always look like it.

BUT was that ALWAYS the case.

3. Not all echinoderms were pentameral.
I've mentioned in past blogs that the echinoderms of the Paleozoic were mostly NOTHING like they
are today.

One of the lesssons from paleontology though: symmetry in echinoderms might be part of a changing/evolving body form through time rather than some discrete, adaptive event.

4. A crystalographic/developmental explanation?

from Nichols 1967

So, basically during development, Nichols arguments that the arrangement as seen above in "b"the theoretical development of these plates that this is essentially the strongest arrangement of these plates. Four or six plate arrangements (a or c) presents a clear breakage plane whereas the 5-plate arrangement does not.

He goes on to apply this structural explanation to various living echinoderms, but unfortunately, even Nichols admits, that this idea was experimentally untestable.

5. Some insight from Evo-Devo!
Some of the more intriguing clues into the "How did pentameral symmetry evolve?" are almost certainly going to be found from the field of "Evo-Devo", which is short for "Evolution & Development". A multidisciplinary field which integrates genetics and developmental biology. Which genes "turn on" or express certain characters??

One paper by Arenas-Mena et al. (2000) from Andy Cameron's lab at the California Institute of Technology in Development shows expression of the Hox cluster of genes in the purple sea urchin, Strongylocentrotus purpuratus.

There has undoubtedly been more work on this topic, but honestly, this was about all I was going to gather in the time I had and its a VERY involved field!


Most animals belong to Bilateria (see Glossary, Box 1), a group encompassing organisms with three germ layers (ectoderm, endoderm and mesoderm) and two body axes, i.e. an anterior-posterior axis and a dorsal-ventral (D-V) axis. Body axes can be thought of as systems of molecular coordinates (Niehrs, 2010), allowing different parts of the body to develop differently. For example, the central nervous system develops at the dorsal side of the vertebrate body, but ventrally in insects and many other animals. The anterior end is usually characterized by a concentration of sensory organs, such as eyes and the olfactory system. Bilaterality also favours the formation of left-right asymmetry in many animals, including vertebrates. However, among the non-bilaterian Metazoa (see Glossary, Box 1), other types of symmetry exist (Fig. 1). For example, sponges (Porifera), although missing a clear body symmetry in their modular, sessile adult state, have an obvious radial symmetry as larvae. Comb jellies (Ctenophora) are bi-radially symmetric, with an oral-aboral axis and two other planes of symmetry, one going through the bases of the tentacles and the other through the slit-like mouth. Placozoans are irregularly shaped, crawling animals that exhibit a dorsal and ventral surface, although how these surfaces arise is unclear as placozoan embryogenesis is unknown. These various types of symmetry and body axes raise the question of how and when bilaterality – a trait that likely contributed to the diversification of body plans (see Box 2) – might have arisen.

Acoela. A group of animals with a single gut opening previously thought to be members of flatworms, but now usually placed within the earliest branching bilaterian lineage Xenacoelomorpha.

Ambulacraria. Besides chordates, one of the two major clades of Deuterostomia. Ambulacraria includes echinoderms (sea urchins, starfish, etc.) and hemichordates. In contrast to chordates, ambulacrarians do not have a centralized nervous system and, similar to non-deuterostome Bilateria, possess a ventral BMP signalling minimum.

Amphistomy. The mode of gastrulation in which the lateral lips of the blastopore fuse in a slit-like fashion leaving two openings: an anterior mouth and a posterior anus connected by a U-shaped gut.

Bilateria. The phylogenetic group of bilaterally symmetric animals, consisting of three germ layers. Bilateria are subdivided into Xenacoelomorpha, Deuterostomia and Protostomia.

Chordata. The second major clade of Deuterostomia, including cephalochordates (amphioxus), tunicates (ascidians, larvaceans, etc.) and vertebrates.

Deuterostomia. An animal group consisting of Ambulacraria and Chordata. The name comes from the fact that their mouth forms separately from the blastopore.

Ecdysozoa. An animal clade uniting moulting animals (nematodes, priapulids, arthopods, etc.).

GLWamide-positive neurons. Neurons expressing neuropeptides carrying GLWamide on the C terminus.

Lophotrochozoa. An animal clade uniting groups with trochophore-like larvae (molluscs, annelids, ribbon worms, etc.) and lophophorate animals (bryozoans, brachiopods, etc.). Currently considered as a subclade within Spiralia, which include also Gnathifera (gnathostomulids, rotifers, etc.) and Rouphozoa (flatworms, gastrotrichs), and uniting animals with spiral cleavage.

Mesenteries. Endodermal folds of anthozoans harbouring longitudinal muscles and gonads.

Metazoa. The clade uniting all animal phyla.

Planula. A type of diploblastic ciliated larva typical for all cnidarian clades.

Primary polyp. A developmental stage following metamorphosis of the cnidarian planula. A Nematostella primary polyp has four tentacles. As it develops, further tentacles will intercalate between the first four.

Protostomia. An animal group well supported by molecular phylogenies and containing Spiralia and Ecdysozoa. The name comes from the notion that, in protostomes, the mouth forms directly from the blastopore, which is not always the case.

Urbilaterian. The last common ancestor of all Bilateria.

The distribution of different body symmetries among animals. Alternative scenarios that can explain the emergence of bilaterality are depicted. An-Veg, animal-vegetal DS-VS, dorsal surface-ventral surface O-A, oral-aboral A-P, anterior-posterior D-V, dorsal-ventral.

The distribution of different body symmetries among animals. Alternative scenarios that can explain the emergence of bilaterality are depicted. An-Veg, animal-vegetal DS-VS, dorsal surface-ventral surface O-A, oral-aboral A-P, anterior-posterior D-V, dorsal-ventral.

Figure 7. Shown are the planes of a quadruped goat and a bipedal human. The midsagittal plane divides the body exactly in half, into right and left portions. The frontal plane divides the front and back, and the transverse plane divides the body into upper and lower portions.

Vertebrate animals have a number of defined body cavities, as illustrated in Figure 8. Two of these are major cavities that contain smaller cavities within them. The dorsal cavity contains the cranial and the vertebral (or spinal) cavities. The ventral cavity contains the thoracic cavity, which in turn contains the pleural cavity around the lungs and the pericardial cavity, which surrounds the heart. The ventral cavity also contains the abdominopelvic cavity, which can be separated into the abdominal and the pelvic cavities.

Figure 8. Vertebrate animals have two major body cavities. The dorsal cavity contains the cranial and the spinal cavity. The ventral cavity contains the thoracic cavity and the abdominopelvic cavity. The thoracic cavity is separated from the abdominopelvic cavity by the diaphragm. The abdominopelvic cavity is separated into the abdominal cavity and the pelvic cavity by an imaginary line parallel to the pelvis bones. (credit: modification of work by NCI)

Did radial symmetry evolve twice? - Biology

Unit Five. Evolution of Animal Life

19.3. Six Key Transitions in Body Plan

The evolution of animals is marked by six key transitions: the evolution of tissues, bilateral symmetry, a body cavity, segmentation, molting, and deuterostome development. These six body transitions are indicated at the branchpoints of the animal evolutionary tree in figure 19.3.

Figure 19.3. Evolutionary trends among the animals.

In this chapter, we examine a series of key evolutionary innovations in the animal body plan, shown here along the branches. Some of the major animal phyla are shown on this tree. Lophophorates exhibit a mix of protostome and deuterostome characteristics. The traditional tree shown here assumes segmentation arose only once among the invertebrates, while molting arose independently in nematodes and arthropods. The newly proposed molecular phylogenies assume molting arose only once, while segmentation arose independently in annelids, arthropods,and chordates.

The simplest animals, the Parazoa, lack both defined tissues and organs. Characterized by the sponges, these animals exist as aggregates of cells with minimal intercellular coordination. All other animals, the Eumetazoa, have distinct tissues with highly specialized cells.

2. Evolution of Bilateral Symmetry

Sponges also lack any definite symmetry, growing asymmetrically as irregular masses. Virtually all other animals have a definite shape and symmetry that can be defined along an imaginary axis drawn through the animal’s body.

Radial Symmetry. Symmetrical bodies first evolved in marine animals exhibiting radial symmetry. The parts of their bodies are arranged around a central axis in such a way that any plane passing through the central axis divides the organism into halves that are approximate mirror images.

Bilateral Symmetry. The bodies of all other animals are marked by a fundamental bilateral symmetry, a body design in which the body has a right and a left half that are mirror images of each other. This unique form of organization allows parts of the body to evolve in different ways, permitting different organs to be located in different parts of the body. Also, bilaterally symmetrical animals move from place to place more efficiently than radially symmetrical ones, which, in general, lead a sessile or passively floating existence. Due to their increased mobility, bilaterally symmetrical animals are efficient in seeking food, locating mates, and avoiding predators.

3. Evolution of a Body Cavity

A third key transition in the evolution of the animal body plan was the evolution of the body cavity. The evolution of efficient organ systems within the animal body was not possible until a body cavity evolved for supporting organs, distributing materials, and fostering complex developmental interactions.

The presence of a body cavity allows the digestive tract to be larger and longer. This longer passage allows for storage of undigested food and longer exposure to enzymes for more complete digestion. Such an arrangement allows an animal to eat a great deal when it is safe to do so and then to hide during the digestive process, thus limiting the animal’s exposure to predators.

An internal body cavity also provides space within which the gonads (ovaries and testes) can expand, allowing the accumulation of large numbers of eggs and sperm. Such storage capacity allows the diverse modifications of breeding strategy that characterize the more advanced phyla of animals. Furthermore, large numbers of gametes can be stored and released when the conditions are as favorable as possible for the survival of the young animals.

4. The Evolution of Segmentation

The fourth key transition in the animal body plan involved the subdivision of the body into segments. Just as it is efficient for workers to construct a tunnel from a series of identical prefabricated parts, so segmented animals are assembled from a succession of identical segments. Segmentation was assumed to have evolved only once among the invertebrates in the traditional taxonomy, as it seemed such a significant alteration of body plan.

5. The Evolution of Molting

Most coelomate animals grow by gradually adding mass to their body. However, this creates a serious problem for animals with a hard exoskeleton, which can hold only so much tissue. To grow further, the individual must shed its hard exoskeleton, a process called molting or, more formally, ecdysis.

Ecdysis occurs among both nematodes and arthropods. In the traditional taxonomy these are treated as two independent evolutionary events. The new phylogenies suggest ecdysis evolved only once. This would imply that arthropods and nematodes, both of which have hard exoskeletons and molt, are sister groups, and that segmentation rather than ecdysis must have evolved several times among the invertebrates, rather than once.

6. The Evolution of Deuterostome Development

Bilateral animals can be divided into two groups based on differences in the basic pattern of development. One group is called the protostomes (from the Greek words protos, first, and stoma, mouth) and includes the flatworms, nematodes, mollusks, annelids, and arthropods. Two outwardly dissimilar groups, the echinoderms and the chordates, together with a few other smaller related phyla, comprise the second group, the deuterostomes (Greek, deuteros, second, and stoma, mouth). Protostomes and deuterostomes differ in several aspects of embryo growth and will be discussed later in the chapter.

Deuterostomes evolved from protostomes more than 630 million years ago, and the consistency of deuterostome development, and its distinctiveness from that of the proto-stomes suggests that it evolved once, in a common ancestor to all of the phyla that exhibit it.

Characteristics of the major animal phyla are described in table 19.2.


Key Learning Outcome 19.3. Six key transitions in body design are responsible for most of the differences we see among the major animal phyla.

Art Connection

Animal phylogeny. The phylogenetic tree of animals is based on morphological, fossil, and genetic evidence. The Ctenophora and Porifera are both considered to be basal because of the absence of Hox genes in this group, but how they are related to the “Parahoxozoa” (Placozoa + Eumetazoa) or to each other, continues to be a matter of debate.

Which of the following statements is false?

  1. Eumetazoans have specialized tissues and parazoans don’t.
  2. Lophotrochozoa and Ecdysozoa are both Bilataria.
  3. Acoela and Cnidaria both possess radial symmetry.
  4. Arthropods are more closely related to nematodes than they are to annelids.

Trends in flower symmetry evolution revealed through phylogenetic and developmental genetic advances

A striking aspect of flowering plant (angiosperm) diversity is variation in flower symmetry. From an ancestral form of radial symmetry (polysymmetry, actinomorphy), multiple evolutionary transitions have contributed to instances of non-radial forms, including bilateral symmetry (monosymmetry, zygomorphy) and asymmetry. Advances in flowering plant molecular phylogenetic research and studies of character evolution as well as detailed flower developmental genetic studies in a few model species (e.g. Antirrhinum majus, snapdragon) have provided a foundation for deep insights into flower symmetry evolution. From phylogenetic studies, we have a better understanding of where during flowering plant diversification transitions from radial to bilateral flower symmetry (and back to radial symmetry) have occurred. From developmental studies, we know that a genetic programme largely dependent on the functional action of the CYCLOIDEA gene is necessary for differentiation along the snapdragon dorsoventral flower axis. Bringing these two lines of inquiry together has provided surprising insights into both the parallel recruitment of a CYC-dependent developmental programme during independent transitions to bilateral flower symmetry, and the modifications to this programme in transitions back to radial flower symmetry, during flowering plant evolution.

1. Introduction

Variation in flower symmetry has attracted the attention of botanists for more than a century [1–4]. Research has centred on understanding the developmental mechanisms that establish patterns of symmetry, the ecological contexts in which alternative patterns of symmetry are favoured, and the evolutionary history of transitions between different forms. This research has provided key insights into how, when and why transitions in floral symmetry evolve.

During the diversification of flowering plants (angiosperms), there have been numerous evolutionary transitions between radial flower symmetry (polysymmetry, actinomorphy figure 1a) and bilateral flower symmetry (monosymmetry, zygomorphy figure 1d), or in more extreme cases, flower asymmetry (figure 1c) [5,6]. Bilateral symmetry is predominant in a number of species-rich lineages—for example, Lamiales (mints and allies) and Fabaceae (legumes) in eudicots, and Orchidaceae in the monocots. Bilateral symmetry in these lineages is not only common, but also highly elaborate. However, a survey of flowering plant lineages demonstrates that both elaborate and subtle forms of bilateral flower symmetry have evolved from radially symmetrical ancestors many times, and that reversals from bilateral to radial, or approximately radial symmetry are not uncommon (reviewed in [6]).

Figure 1. Flower symmetry diversity and bilateral flower symmetry developmental genetics. The range of floral symmetries include radial symmetry with multiple planes of mirror image symmetry (a, Potentilla sp.), disymmetry with two planes of mirror image symmetry (b, Cardaminopsis arenosa), asymmetry with zero planes of mirror image symmetry (c, Pedicularis racemosa) and bilateral symmetry with just a single plane of mirror image symmetry (d, Antirrhinum majus). At the developmental level, one or more genetic signals must differentiate the dorsal (adaxial) from the ventral (abaxial) domains of the developing flower, for example a genetic programme that distinguishes dorsal identity (e, dorsal shading in cartoon of early developing flower). In the model species A. majus, the genetic programme that establishes dorsoventral flower identity from early stages of development includes the dorsal identity genes and protein products CYCLOIDEA (CYC), DICHOTOMA (DICH) and RADIALIS (RAD) as well DIVARICATA (DIV) which specifies ventral flower development (f). DIV is excluded from the dorsal domain of the developing A. majus flower through post-translational negative regulation by RAD. C. arenosa flower is taken from image for which copyright is held by Meneerke bloem ( This image is used under a Creative Commons Attribution-Share Alike 3.0 Unported, 2.5 Generic, 2.0 Generic and 1.0 Generic licence. P. racemosa flower taken from image for which copyright is held by Jerry Friedman ( This image is used under the Creative Commons Attribution-Share Alike 3.0 Unported licence. (Online version in colour.)

It is generally accepted that these transitions in flower symmetry are associated with pollination syndromes. For example, transitions from radial to bilateral flower symmetry appear to be linked to the evolution of specialized plant–pollinator interactions. Bilateral symmetry is most often evident in the petal and stamen whorls and may promote pollinator approach and legitimate (pollen transferring) landings, and may increase the specificity of pollen deposition during pollinator visits [7–9]. In part because of the relationship between symmetry and specialized pollination biology, transitions to bilateral flower symmetry are hypothesized to represent key innovations associated with diversification of species-rich flowering plant lineages [10,11].

More recently, much attention has turned to the developmental programmes that specify bilateral flower symmetry. Now, many genes and genetic interactions necessary for the development of bilateral flowers are understood from the model species Antirrhinum majus (snapdragon, reviewed in [12]). Because of the historical interest in floral symmetry pollination ecology and evolution, these newer insights from A. majus provide the jumping board for comparative studies. The primary comparative question has been whether the developmental programme identified in A. majus contributes to the establishment of bilateral symmetry in other flowering plant lineages. Strikingly, current evidence suggests that a similar developmental programme, first identified in A. majus, has been recruited many times independently during the parallel evolution of bilateral flower symmetry (reviewed in [12–16]).

Linking model system findings (e.g. in A. majus flower development) to comparative developmental questions is not a new concept. Leslie Gottlieb (1936–2012) was an early and strong proponent of the idea that developmental genetic studies in model plant species can inform our understanding of natural variation in flower form [17,18]. For example, he furthered the hypothesis that induced mutations, identified early in the establishment of A. majus as a model species [19], may provide genetic information about floral traits that distinguish species or genera. With respect to floral symmetry, he recognized that A. majus mutants with increased petal and/or stamen number and radial flower symmetry may be significant for understanding Verbascum flower development. Likewise, he pointed out that A. majus mutants with a reduced corolla limb and tubular configuration show similarities to Rhinanthus flowers [18]. Similar to Gottlieb's early examples, many current comparative flower developmental genetic (evo-devo) studies aim to test how variation in genetic pathways, identified largely through loss or gain of function mutations in model plant species, may explain natural variation in flower form.

Extrapolating from model system studies of flower development to the genetic basis of interspecific variation in flower form is best approached in a phylogenetic framework. A well-resolved hypothesis of phylogenetic relationships among flowering plant species allows assessment of ancestral character states, and pinpoints evolutionary transitions towards or away from character states of interest. The past 40 plus years of molecular phylogenetic studies in flowering plants have provided and continue to provide this critical framework (reviewed in [20]). Early on, Gottlieb embraced molecular tools for plant phylogenetic and evolutionary studies [21–25]. His contributions made an impact that provided momentum to the field, and this momentum has not waned. We now have a clearer understanding of relationships among major lineages of flowering plants (figure 2a and see [26–28]) likewise, molecular phylogenetic studies have contributed to resolution of relationships within many key lineages (e.g. figure 2b and [29–32]). With respect to flower symmetry development and evolution, the products of molecular phylogenetic studies allow researchers to determine how often and in which lineages transition from radial flower symmetry to bilateral symmetry (and back to radial flower symmetry) have occurred [16,31,33–37], thus providing the framework for informed choice of species when addressing comparative developmental questions.

Figure 2. Evolutionary transitions in floral symmetry in a phylogenetic framework. (a) Phylogeny of major angiosperm lineages (from [26]). Lineages in which elaborate bilateral flower symmetry can be found (from [6]) are in red text (grey text in print version). Stars indicate lineages containing species for which CYCLOIDEA homologues have been implicated in transitions to bilateral flower symmetry. (b) Phylogeny of representative Lamiales lineages (from [31]). One possible parsimonious history of floral symmetry evolution is shown suggesting multiple transitions from radial to bilateral flower symmetry early in Lamiales diversification, followed by multiple transitions from bilateral to radial (or approximately radial) flower symmetry. Lineages with radial flower symmetry are in black/bold those with bilateral flower symmetry are in red or grey/not bold. Taxa were scored at the species level (see Schaferhoff et al. [31] for complete taxon list). Species exhibiting elaborate bilateral symmetry in the corolla and/or androecium were scored as having bilaterally symmetrical flowers. (Online version in colour.)

Here, I review some recent advances in understanding flower symmetry evolution. I address multiple important contributions of molecular phylogenies to the field. Additionally, I demonstrate how the past 10 years of linking model system findings to comparative developmental questions has shed light on the extensive developmental parallelism in independent transitions between flower symmetry forms.

2. Diversity in floral symmetry

Although the focus of this review is evolutionary transitions between radial and bilateral flower symmetry (figure 1a,d), it is important to recognize that these two symmetry forms represent only part of the diversity in symmetry found across flowering plants. Flower symmetry is generally assessed via the face-on view of a flower at the time of anthesis, and is usually expressed most strongly in the petal and stamen whorls of the flower. Radially symmetrical flowers (figure 1a) display several planes of symmetry that bisect the flower into mirror images, and bilaterally symmetrical flowers (figure 1d) display just a single plane of mirror image symmetry. However, flowers may be disymmetrical (figure 1b), having just two planes of mirror image symmetry, or asymmetrical (figure 1c), lacking altogether a plane of symmetry that bisects the flower into mirror images. Interestingly, bilateral symmetry may often be an intermediate state between radial symmetry and asymmetry. For example, asymmetric Pedicularis (figure 1c) and asymmetric Phaseolus and Lathyrus are nested within Lamiales (figure 2b) and Faboideae, respectively, two flowering plant lineages in which bilateral flower symmetry is predominant. Likewise, multiple forms of bilateral flower symmetry are derived from disymmetry. For example, bilaterally symmetrical Iberis (Brassicaceae) and Corydalis (Papaveraceae) are derived from ancestors with disymmetrical flowers [38–41].

Bilateral flower symmetry itself can range from elaborate to subtle patterns of low complexity (reviewed in [6]). Most familiar forms of complex bilateral flower symmetry are the bilabiate (lipped and keeled) flower forms. In bilabiate flowers, reproductive organs (stamens and carpels) are found inside an elaborate corolla that is differentiated along the dorsal/ventral (adaxial/abaxial) floral axis. In lip flowers, the reproductive organs are held in the upper side of the corolla resulting in pollen transfer on the backs of visiting pollinators in keel flowers, the reproductive organs are held in the lower (keel) side of the corolla resulting commonly in pollen transfer on the underside of pollinators. Bilaterally symmetrical flowers of the lip form are extremely prevalent in Lamiales (e.g. A. majus, figure 1d), but are found in other lineages, including Campanulales and Orchidaceae. Those of the keel form are well known from Fabaceae, but can also be found in Polygalaceae. Less elaborate forms of bilateral flower symmetry also result from organ differentiation primarily in the petal and/or stamen whorls, and may be due to displacement of organ initiation, size or shape variation in organs along the dorsoventral axis of the flower, or sigmoidal curvature of organs (reviewed in [6,16]).

3. A phylogenetic context for floral symmetry evolution

From assessments of taxonomic distribution of bilateral flower symmetry [1], and variation in the form of bilaterally symmetrical flower (e.g. lip versus keel bilabiate flowers), it has historically been quite clear that transitions from radial to bilateral flower symmetry were probably frequent during flowering plant diversification. However, it is only in the context of robust phylogenetic hypotheses for the relationships among flowering plant lineages that we can determine along which lineages evolutionary transitions from radial to bilateral (and back to radial) flower symmetry have occurred [42]. And it is primarily advances in molecular phylogenetics that provide the context for studies of floral symmetry character evolution.

Studies that have used molecular phylogenies to reconstruct the ancestral flower conclude that it was radially symmetrical [43]. A clear understanding of the ancestral form of symmetry is an excellent starting point for determining where bilateral flower symmetry has been gained or lost in flowering plants. A number of recent molecular phylogenies that sample taxa at approximately the family level are now being used to assess patterns of floral character evolution, including symmetry [26–28]. Figure 2a shows the ordinal-level backbone phylogeny from Soltis et al. [26] on which orders containing species with more or less elaborate bilateral flower symmetry [6] are indicated. This is by no means a critical evaluation of floral symmetry evolution, but illustrates the widely dispersed nature of transitions to bilateral flower symmetry. Citerne et al. [16] undertook an excellent analysis of floral symmetry evolution on the estimate of flowering plant family relationships presented in Bremer et al. [27]. Using a parsimony approach, and scoring for flower symmetry at the family level (which is likely to underestimate the number of transitions to bilateral flower symmetry), they identified a single transition to bilateral flower symmetry among the basal angiosperms, 23 transitions in monocots, and 46 independent transitions in the eudicots. Therefore, using a well resolved and densely sampled (at the family level) estimate of flowering plant phylogeny, Citerne et al. [16] suggest at least 70 transitions to bilateral flower symmetry—twice as many as previously reported.

Studies of character evolution on large-scale phylogenies, such as the one undertaken by Citerne et al., represent important advances in our understanding of floral evolution. Ideally, as advances are made in molecular phylogenetics, we will have at our disposal estimates of the flowering plant phylogeny that are densely sampled at the genus (or even species) level, and for which phylogenetic branch length estimates are available. It will be in this context that floral symmetry evolution will be most critically evaluated using statistical methods for ancestral state reconstruction [44]. That bilateral flower symmetry is a key innovation leading to increased diversification rates has been hypothesized, and to a limited extent tested [10,11]. As with studies of character state evolution, it will be in this context of densely sampled phylogenies that the relationships between shifts in flower symmetry and clade diversification will best be investigated [45]. Excitingly, researchers are anticipating these large datasets. For example, both the National Evolutionary Synthesis Center (NESCent)-supported working group ‘Floral assembly: quantifying the composition of a complex adaptive structure’ ( and eFLOWER ( are developing massive data matrices of floral traits, including floral symmetry, scored at the species level.

If we move our focus from the entire clade to specific lineages within flowering plants, then we find that more fully resolved assessments of floral symmetry evolution are possible. This more focused view will certainly suggest additional transitions to and from bilateral symmetry to those that would be seen on an ordinal- or family-level sampled phylogeny of flowering plants. For example, Schaferhoff et al. [31] generated a densely sampled, well-resolved phylogeny of Lamiales. Scoring for corolla symmetry at just the family level of their backbone phylogeny, they recover one transition from radial to bilateral symmetry, and one transition back to radial symmetry. Using the same phylogeny, but scoring for corolla and stamen whorl symmetry at the species level, based on the species sampled in the Schaferhoff et al. [31] phylogeny, I recover possibly two transitions from radial to bilateral flower symmetry early in Lamiales diversification (considering the bilateral symmetry in the stamen whorl of many Oleaceae species), and multiple transitions from bilateral to radial flower symmetry (figure 2b). Others have undertaken similar analyses of floral symmetry evolution in large flowering plant lineages, scoring symmetry for genera or species. Some key findings are multiple transitions from radial to bilateral symmetry inferred during Solanaceae [29,35], Brassicaceae [38] and Ranunculales [46] diversification. By contrast, in Malpighiales, a single transition to bilateral flower symmetry is recovered, followed by multiple transitions from bilateral to radial symmetry [37,47].

4. Developmental genetics of floral symmetry

As described above, bilateral flower symmetry has evolved multiple times and its form varies in complexity. Research in the model species A. majus, with its elaborate bilabiate form (figure 1d), provided the first ground-breaking insights into the genetic control of bilateral flower symmetry. At the foundation of this control is a programme that differentiates dorsal (adaxial) and ventral (abaxial) flower identity from very early stages of floral organ initiation and differentiation (figure 1e).

Two recently duplicated TCP (Teosinte branched 1/Cycloidea/proliferating cell factors) family transcription factors [48–51], CYCLOIDEA (CYC) and DICHOTOMA (DICH), function partially redundantly to specify dorsal flower identity (figure 1f) [52,53]. These paralogues represent the upstream extent of our knowledge of dorsal flower specification. In other words, we do not yet know what gene products control the regulation of CYC and DICH. Expression of CYC and DICH corresponds with their function in specifying dorsal flower identity. Both are expressed in the dorsal region of the floral meristem from initiation, and their dorsal-restricted expression is maintained throughout petal and stamen development [52–54]. CYC and DICH expression and function in the dorsal flower domain are necessary for establishing the distinct shape of dorsal petals (figure 1d), abortion of the dorsal (medial) stamen, as well as petal and stamen merosity. In an A. majus cycdich double mutant background, flowers completely lack dorsal identity, are radially symmetrical and develop with ventral identity in the ventral, lateral and dorsal domains [52,53]. CYC and DICH appear to determine the distinct shape of dorsal petals and the formation of the dorsal staminode by affecting patterns of cell growth and proliferation. This is in line with the widely recognized function of TCP transcription factors in promoting and/or repressing tissue growth (reviewed in [12,51]).

While CYC and DICH are necessary to differentiate dorsal floral identity, a single MYB family transcription factor, DIVARICATA (DIV), functions to specify ventral identity (figure 1f) [54,55]. DIV expression and function in the ventral flower domain are necessary for establishing the shape of the ventral (medial) petal, which distinctly contributes to the lower lip of the bilabiate A. majus flower (figure 1d). Interestingly, in early flower development, DIV is expressed in both the dorsal and ventral domains of the flower, but its expression becomes somewhat restricted to the developing ventral petal at later stages of development [54]. The effects of CYC and DICH on dorsal flower development and of DIV on ventral flower development are in part mediated through an additional MYB transcription factor, RADIALIS (RAD). RAD expression is positively regulated by CYC and DICH. Therefore, RAD expression and function are primarily restricted to the dorsal domain of developing flowers (figure 1f) [56,57]. It is RAD protein in the dorsal flower domain that post-translationally restricts DIV function to the ventral domain (figure 1f) [55,56,58,59].

5. Parallel recruitment of a CYC-dependent pathway in bilateral symmetry evolution

From extensive molecular phylogenetic work and studies of character evolution in flowering plants, we have a clearer understanding of the history of flower symmetry evolution. Additionally, from research on flower development in A. majus, we know at least one way by which flower symmetry can be established at the molecular level. Together, these provide a foundation for comparative developmental studies. Bilateral flower symmetry evolved early in the diversification of Lamiales (figure 2b) therefore, bilateral symmetry in A. majus is homologous to bilateral flower symmetry found in other Lamiid lineages (with the possible exception of Oleaceae figure 2b). A reasonable, testable hypothesis is that the A. majus CYC/RAD/DIV developmental programme (figure 1f) evolved early in Lamiales and is conserved among relatives of A. majus with bilateral flower symmetry. In addition, either similar or divergent genetic programmes may have been recruited to specify independent origins of bilateral flower symmetry elsewhere in flowering plants (figure 2a). Possibilities include independent recruitment of a CYC-dependent programme to specify dorsal or ventral identity (figure 3a,b), or novel recruitment of a CYC-independent developmental programme to specify either dorsal or ventral flower identity (figure 3c,d). Results from many comparative studies now demonstrate that there is striking parallelism in the independent evolution of bilateral symmetry with a CYC-dependent programme frequently recruited to specify dorsal identity, and in some cases ventral identity (reviewed in [13,15,16]).

Figure 3. Hypothesized CYC-dependent and CYC-independent pathways for recurrent evolutionary transitions from radial to bilateral, and bilateral to radial flower symmetry. (a,b) A CYC-dependent programme is necessary for the development of bilateral flower symmetry through the specification of dorsal or ventral identity, respectively. (c,d) A CYC-independent programme is necessary for the development of bilateral flower symmetry through the specification of dorsal or ventral identity, respectively. (e,f) Radial flower symmetry is derived from CYC-dependent bilateral symmetry through loss of the dorsoventral restricted CYC-dependent programme. (g,h) Radial flower symmetry is derived from CYC-dependent bilateral symmetry through an independent programme that compensates for the effects of the CYC-dependent programme. (Online version in colour.)

(a) Asterids

Limited data support the hypothesis that the CYC/RAD/DIV programme is conserved across Lamiales. In bilaterally symmetrical flowers of Veronica and Gratiola (belonging to the same family as A. majus, Plantaginaceae), CYC and RAD homologues are expressed in the dorsal regions of flowers and with nearly identical spatial distributions, suggesting conservation of positive regulation of the RAD gene by CYC protein [60]. This is also true in Chirita and Bournea from the early diverging Lamiales lineage Gesneriaceae [61,62]. Whether post-translational negative regulation of DIV by RAD protein is conserved in Lamiales is not clear and difficult to test, because analyses of DIV transcript localization will not reflect where in the flower DIV protein is present and functional (although see [62]). In addition to limited information on the conservation of the CYC/RAD/DIV programme within Lamiales, few studies have investigated the regulatory interactions among these genes/gene products in other asterid lineages (but see [63,64] summarized below), especially those most closely related to Lamiales (e.g. Boraginaceae, Solanales, Gentianales figure 2a). In the distantly related model species Arabidopsis thaliana (rosid lineage), CYC- and RAD-like genes and gene products do not seem to be directly regulated by one another [57,65], but the phylogenetic distance makes it difficult to draw conclusions about when the CYC/RAD network interactions evolved.

Elsewhere within asterids, the role of CYC-like genes in independent transitions to bilateral flower symmetry has been investigated in Dipsacales and Asterales (figures 2a and 4). In the bilaterally symmetrical flower of Lonicera (Caprifoliaceae, Dipsicales), duplicate CYC-like genes are expressed in the dorsal or dorsal plus lateral petals (figure 4). This is in striking contrast to the radially symmetrical flowers of related Viburnum (Adoxaceae, Dipsacales) where these CYC-like orthologues show no pattern of differential expression across the floral axis. Interestingly, a Lonicera RAD orthologue is expressed similarly to one of the Lonicera CYC paralogues, providing some indication that the CYC/RAD regulatory interaction may have been established early in asterid evolution and retained in both Dipsacales and Lamiales, but this hypothesis requires extensive further testing.

Figure 4. Summarized expression of CYC-like genes from comparative developmental genetic studies. Blue (grey in print version) shading indicates approximate pattern of CYC-like gene expression in the corolla of representative taxa. Phylogeny as in figure 2a, but for only the subset of orders with bilaterally symmetrical species. (Online version in colour.)

Multiple lines of evidence demonstrate that CYC-like genes play a role in establishing the developmental differentiation of ray flowers (bilaterally symmetrical) from disc flowers (radially symmetrical) in Asteraceae inflorescences (capitula). CYC homologues in Helianthus, Senecio and Gerbera are preferentially expressed around the periphery of the capitulum where ray flowers are expected to develop, but either at low levels, or not at all in the region of disc flower development (figure 4) [66–68]. In Helianthus (sunflower), naturally occurring mutations transform disc flowers to ray flowers (double-flowered mutants), and ray flowers to disc-like tube flowers (tubular-rayed mutants). Double-flowered mutants are due to mutations that cause overexpression of a CYC-like gene in the region of disc flower development, thus causing their transformation to ray identity [66]. Similarly, transgenic overexpression of a CYC homologue in Gerbera leads to transformation of disc flowers to ray identity [68]. Tubular-rayed mutants are due to loss-of-function mutations in a CYC-like gene resulting in conversion of ray flowers to disc-like tubular flowers [66,69]. Interestingly, it appears to be different CYC paralogues in different Asteraceae lineages that are responsible for differentiation of ray flowers [66]. Although this is somewhat surprising, this is consistent with ray flowers having evolved multiple times in the family [70].

(b) Rosids

Similar to asterids, bilateral flower symmetry has evolved multiple times in rosids, and in at least three instances is associated with independent recruitment of a CYC-dependent developmental programme (Fabaceae, Brassicaceae and Malpighiaceae figure 4). Developmental genetic studies of bilateral symmetry in Lotus and Pisum (Fabaceae) are extensive and second only to the work on floral symmetry in A. majus. In both Lotus and Pisum, there are three CYC-like paralogues. Through analysis of naturally occurring mutants, as well as gene silencing and overexpression transgenic studies, it is clear that two of these paralogues, CYC1 and CYC2 (LST in Pisum, and SQU1 in Lotus) function redundantly to establish dorsal petal identity [71–74]. Unlike in A. majus where the ground state for development seems to be lateral petal identity (cycdichdiv mutant background [75]), in Pisum and Lotus the ground state appears to be ventral petal identity because in addition to dorsal identity controlled by CYC1 and CYC2, the third paralogue, CYC3 (K in Pisum, KEW in Lotus), directs the development of lateral petals (figure 4) [71–74].

In Brassicaceae, the role of CYC-like genes for directing development of bilateral flower symmetry has been investigated in Iberis, a close relative to the model species Arabidopsis. In Iberis, the two ventral petals are expanded relative to the two dorsal petals (figure 4). This difference is established late during Iberis flower development, and is associated with relatively late expression of IaTCP1, a CYC homologue, in the smaller dorsal petals (figure 4) [38,39]. Because Iberis is closely related to Arabidopsis, heterologous functional studies of IaTCP1 in Arabidopsis provided meaningful assessment of IaTCP1 function. Overexpression of IaTCP1 in Arabidopsis resulted in reduced cell proliferation in both vegetative organs and petals [39], consistent with reduced dorsal petal size where IaTCP1 is expressed in Iberis. Lastly for the rosids, and similar to Fabaceae and Brassicaceae, CYC-like genes have been implicated in the evolution of bilateral symmetry in Malpighiaceae, with expression of CYC-like genes restricted to dorsal and dorsal/lateral petals (figure 4) [47].

(c) Early diverging eudicots and monocots

While most of the comparative work of flower symmetry developmental genetics has been undertaken in core eudicot lineages, a handful of studies have tested whether the extensive parallel recruitment of a CYC-dependent programme for bilateral flower symmetry extends to non-core eudicot taxa. And again, we find evidence supporting a role for CYC-like genes in the development of bilaterally symmetrical flowers from early diverging eudicot and monocot lineages. Bilaterally symmetrical flowers of Capnoides in the Fumariodeae lineage of Papaveraceae (Ranunculales) are derived from disymmetric flowers [40,41]. The plane of bilateral symmetry in Fumariodeae flowers is transverse (figure 4), although partial resupination ultimately brings the transverse plane into dorsoventral orientation. In Capnoides, expression of two CYC-lineage paralogues [40,41] is asymmetric, with slightly stronger expression at the base of the outer petal that forms a nectary [76].

In monocots, transitions from radial to bilateral flower symmetry are pervasive [16], yet are quite under studied. In bilaterally symmetrical flowers of Costus and Heliconia (Zingerberales), as well as bilaterally symmetrical flowers of Commelina (Commelinales), expression of at least one CYC-like gene is asymmetric across the dorsoventral flower axis. In both monocot lineages studied, asymmetric CYC-like gene expression in the perianth is restricted to the ventral side of flowers (figure 4) [77,78]. This is in striking contrast to the general pattern of a CYC-dependent programme independently recruited to specify dorsal flower development across eudicots (figure 4). It is interesting to note, however, that a CYC-like gene from rice, RETARDED PALEA1, functions to specify palea development [79], an organ that develops on the dorsal side of grass florets. Whether the emerging pattern of dorsal flower expression in eudicots and ventral expression in monocots is a general pattern, perhaps reflecting developmental constraints, awaits further comparative work in monocots, as well as a clearer understanding of how CYC homologue expression is regulated during monocot and eudicot flower development.

6. Evolutionary transitions from bilateral to radial flower symmetry

Given the frequent association of bilateral symmetry with restricted expression of CYC-like genes to either the dorsal (most dicots), or ventral (most monocots) side of developing flowers, it is expected that reversals from CYC-dependent bilateral symmetry to radial symmetry will involve functional or regulatory changes to CYC homologues or their upstream regulators. There are, however, multiple hypothesized ways by which CYC-dependent bilateral flower symmetry might be lost in derived species with radial flower symmetry. One possibility is complete loss of CYC-like gene expression in flowers, through either regulatory evolution or gene loss (figure 3e). By contrast, regulatory evolution may result in expansion of CYC-like gene expression across the dorsoventral axis of developing flowers (figure 4f). Alternatively, the evolution of radial symmetry from CYC-dependent bilateral symmetry could arise through mechanisms independent of functional or regulatory evolution of CYC-like genes. For instance, compensatory changes might evolve in genes/genetic pathways downstream of CYC, or in developmental pathways non-overlapping with a CYC-like programme (figure 4g,h). Results from multiple comparative studies suggest that evolutionary changes at or upstream of CYC-like genes frequently underlie transitions from CYC-dependent bilateral to radial flower symmetry. However, results from some studies are not inconsistent with a hypothesis of compensatory evolution.

Examples of derived radial symmetry (from CYC-dependent bilateral symmetry) for which the expression of CYC homologues has been studied include Plantago (Plantaginaceae, Lamiales), Cadia (Fabaceae, Fabales), two independent transitions to radial from bilateral symmetry in Gesneriaceae (Lamiales)—Bournea and Tengia, and four independent transitions to radial from bilateral symmetry in Malpighiaceae (Malpighiales)—Psychopterys, Sphedamnocarpus, Microsteria and Lasiiocarpus (figure 4) [62,80–83]. For each of these, two or more paralogous CYC-like genes are dorsally expressed in close relatives. Therefore, expression of all paralogues was investigated in these derived radially symmetrical lineages.

The most common pattern observed is a paralogue-specific combination of CYC loss of expression (figure 3e) with expanded CYC expression (figure 3f). In Plantago, Tengia, Cadia and Microsteria, one CYC-like paralogue (or set of closely related paralogues in the case of Tengia) is expressed across the dorsoventral flower axis, owing to regulatory evolution either at or upstream of that paralogue. The other CYC-like paralogue has been lost (Plantago and Microsteria), or is no longer expressed in flowers (Cadia and Tengia) [80–83]. Alternatively, both CYC-like paralogues are expressed across the dorsoventral flower axis (Psychopterys), or neither is expressed in flowers (Sphedamnocarpus) [82]. For two studied lineages with derived radially symmetrical flowers, one CYC-like paralogue has either expanded or lost floral expression, but the other paralogue retains dorsal-specific expression (Lasiocarpus and Bournea, respectively) [81,82]. In these cases, dorsal-specific CYC-like gene expression should be interpreted with caution. Dorsally restricted expression may be transient, or only occur early in development, and therefore may not specify a dorsal-specific developmental programme. Alternatively, there may indeed be functional consequences to retention of dorsal-specific expression, and these developmental consequences may be compensated by evolutionary changes in downstream or independent developmental programmes (figure 3g,h).

7. Prospects

Our current knowledge of the repeated recruitment of a CYC-dependent developmental programme during independent transitions to bilateral flower symmetry is staggering. These insights are possible through a combination of advances in flowering plant molecular phylogenetic research and studies of character evolution, as well as detailed flower developmental genetic studies in a few model species (namely A. majus and L. japonica). Likewise, it is daunting to imagine how little we would know about these evolutionary developmental processes had evolution not proceeded with such extensive parallelism. I believe we are now justified in stating that parallel recruitment of a CYC-dependent developmental programme for bilateral flower symmetry is extensive. However, we have yet to determine the depth of this parallelism: does independent recruitment occur through regulatory changes at CYC-like loci, or through evolutionary changes to one or multiple upstream regulators of CYC, or through a combination of these possibilities that is taxon specific? Additionally, paralogues belonging to different CYC-like gene lineages are implicated in the evolution of bilateral flower symmetry in the core eudicots, early diverging eudicots and monocots. Also, CYC-like genes are generally regulators of cell proliferation [51]. Other than this broad recruitment from different CYC paralogue groups, and a possible general role in regulating cell proliferation, we know little about the specific function or regulation of CYC-dependent genetic pathways that might shed light on why they have so frequently been recruited to flower symmetry developmental programmes.

Watch the video: Zbrush - Radial Symmetry (December 2022).