Reviewer 1: Gáspár Jékely
This is an interesting analysis providing a physical explanation for the maintenance of bilateral symmetry in animal evolution. I find the paper well written, the arguments convincing, and only have a few comments to clarify the discussion.
The authors refrain from discussing locomotion in the microscopic world. However, I think that they miss an opportunity here. We know that in the micro world many organisms can navigate very efficiently. They achieved this not by being bilaterally symmetrical, but by using helical swimming and the adjustment of the helical trajectories. This happens very often in diverse phototactic protist (e.g. Chlamydomonas, dinoflagellates) and in the close-to spherical ciliated larvae of bilaterians (e.g. annelids, hemichordates). One reason why this is an effective strategy for small organisms but not large ones is, as discussed in the paper, the different Reynolds numbers. This could imply that bilaterality only evolved once the early metazoans had attained a sufficiently large size. This interesting physical threshold could be discussed in more detail.
Authors’ response: First of all we thank Dr. Jékely for his invaluable work. The idea that bilaterality evolved when the animals at hand had reached a certain size (e.g. Valentine Proc Natl Acad Sci USA 1994,91:6751–6757) cannot be proved by convincing evidence at present (see for example Chen et al. Science,305:218–222 – although this is controversial [e.g. Chapter 1 by Budd GE in Animal Evolution: Genomes, Fossils, and Trees edited by Telford MJ, Littlewood DTJ, 2009]). The fact that several small animals, living in the realm of low Reynolds numbers, have bilateral symmetry probably indicates that bilaterality could have evolved in the microscopic world, but most likely it does not offer the kind of advantage over other symmetries there as it does in the macro world. Until it can be excluded that certain factors could favour bilateral symmetry in the micro world, we would rather avoid taking a stand on this – otherwise exciting – evolutionary problem. Please see also response Nr. 1 to Dr. Aravind.
The model implies that the first bilaterians were not burrowing but either freely swimming or crawling on the sediment. It would be interesting to see a discussion of this in the context of the earliest putatively bilaterian trace fossils. The interpretation that these are traces of burrowing animals may be slightly at odds with the hypothesis (e.g. Jensen Integr. Comp. Biol., 43:219–228).
Authors’ response: Given that the precise origin of the first bilaterians is unproven we would not like to take sides on their lifestyle; nevertheless, the topic of the trace fossils, still highly controversial, is very interesting. The following paragraph has been added to the “Apparent problems with the association of symmetry and locomotion” section:
“Another potential question may emerge if one examines the earliest trace fossils from the Precambrian. These traces are retained horizontal burrowings in the upper layer of the sediment [
23,
24] and are also attributed to bilaterian animals [
24]. However, this view has been challenged by the discovery of trace maker giant protists [
25], put forward as candidates for the producers of those ancient trails. Now, according to our hypothesis, it seems easy to reconcile the putative burrowing behaviour and bilaterality in the precambrian animals mentioned above (if they really existed) considering that the upper layer of the sediment is likely to have a loose structure with low density, hence it does not necessarily require the body burrowing in it to be cylindrical.”
Accordingly, the cited references also have been added to the paper.
I suggest to also include in Figure 1. a bilaterian with a cylindrical body but with lateral appendages. A laterally flattened, fish-like body is not general for actively moving bilaterians. For example, errant annelid polychaetes have a body that is roughly cylindrical, but they have lateral appendages that can provide the necessary drag during active locomotion.
Authors’ response: The figure has been added as Figure 1.C.
The authors use the term ‘aerodynamic’ when writing about locomotion primarily in water. Using ‘drag’ or ‘resistance’ may be more fortunate.
Authors’ response: The word “aerodynamic” has been replaced by more appropriate ones.
The sentences in question now read: “… 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].”
“A spherically symmetrical body cannot generate the pushing surface, being of equal shape and drag in every direction.”
“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.”
“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.”
The statement that “Bilateral symmetry with two body axes arose … in slow, flatworm-like organisms” together with Ref. [
7] seems to imply that the first bilaterians were phylogenetically related to acoel flatworms. The latest careful phylogenetic analyses show that acoels are deuterostomes (Nature 470, 255–258), so they do not represent the earliest extant bilaterian metazoans. I suggest to write ‘worm-like organisms’.
Authors’ response: It has been changed to “flat, worm-like organisms”; remaining, at the same time, preferably faithful to the cited reference.
A reference to jellyfish navigation (e.g. Garm et al. The Journal of Experimental Biology 210, 3616–3623) would make the discussion about the medusa-type manoeuvering more convincing.
Authors’ response: Thank you, the reference has been included.
Reviewer 2: L. Aravind
Since Beklemishev it has been generally accepted among students of zoology that the advantages of directed movements are the driving force for the origin and maintenance of bilateral symmetry, the dominant form of symmetry among metazoans. It is usually imagined that such this symmetry emerged in benthic contexts – creeping on a substratum enable favored dorso-ventral differentiation, which coupled with selection for effective directed movement resulted in a bilateral form. However, this has been questioned on the basis of the observations of the asymmetric expression of TGF-beta family members and Short gastrulation orthologs along the directive axis in cnidarians. This implies that even if the ancestral metazoan was outwardly radial symmetric, there might have been a pre-adaptation or pre-disposition for bilaterality as suggested by the situation in cnidarians. This has been used by Finnerty to argue for a role for internal circulation within the gut lumen as a major factor in the origin of bilateral symmetry. The molecular evidence on the whole favors a single major origin for bilaterality in animals, but is subsequent strong maintenance remains less explained.
Here the authors present a simple physical explanation as to why bilaterality is a more stable strategy than any other symmetry once directed locomotion emerges. Central to their explanation is its role in maneuverability that has apparently not been used as done by the authors in this article. The physical arguments by the authors point to bilaterality being an apparently stable strategy at large Reynolds number where the equation used by them for drag forces is appropriate.
However, the main question that arises how large should be the Reynolds number be for this argument to hold. Looking up these values it appears that an unicellular free-swimming eukaryotes might have Re
=
10^-1. Here the viscous forces are probably dominant, allowing for the asymmetric morphology of such forms, e.g. ciliates and dinoflagellates. The smallest vertebrate is said to have Re
~
1 and bilateral metazoans like chaetognatha and rotifer have Re in between those figures. It should be noted that they have strongly bilateral forms. Further, the estimate sizes for the basal bilateralians do not place them much higher than this range in terms of Re. So the key question that arises is whether at these sizes the argument based on negligible viscous forces is entirely valid. It would be good for the authors to consider this issue and present potential tests for their hypothesis.
Authors’ response: We thank Dr. Aravind for his valuable work. The fact that bilateral symmetry is also present in the environment of low Reynolds numbers does not necessarily contradict our hypothesis. The very different pattern of main body symmetries between low and high Re environments probably emerges from their different relations to viscous and inertial forces. But this difference does not necessarily exclude bilateral symmetry from the small-scale world – although manifestly it does not enjoy the advantage over other symmetries which it does in the large-scale world. Our hypothesis implies that bilateral symmetry is advantageous in the high Re-world but this does not mean bilaterality could not have been favoured by certain factors in the low Re-world. However, these factors – to our knowledge – have not been clarified since Beklemishev. Please see also response Nr. 1 to Dr. Jékely.
Minor points
<<Bilateral symmetry with two body axes arose early in animal evolution, probably in slow flatworm-like organisms locomoting on a substrate [
2], likely prior to the Cnidarian–Bilaterian split [
3-
6] in the Precambrian [
7,
8].>>
This sentence is potentially confusing, because it might present a contradiction within it. The authors need to clarify as to what they mean by origin of bilateral symmetry prior to the origin of Bilateria (i.e. the flatworm like organisms). Are they meaning the situation in cnidarians or reconstructing some ancestral form?
Authors’ response: Thank you, this part has hopefully been clarified, and it reads now:
“Bilateral symmetry with two body axes arose early in animal evolution, probably in slow, flat, worm-like organisms locomoting on a substrate [
2]. Genetic analyses have concluded that the genes responsible for bilateral symmetry most likely appeared prior to the cnidarian–bilaterian split [
3-
6], in the Precambrian [
7,
8].” <<“a symmetry that is streamlined in only one direction, while non-aerodynamic in other directions, is favourable for locomotion.” >>
May be the last word should be replaced with maneuverable locomotion.
Authors’ response: The word “manoeuvrable” has been inserted in the sentence, thank you:
“a symmetry that is streamlined in only one direction, while non-aerodynamic in other directions, is favourable for manoeuvrable locomotion.”
Reviewer 3: Eugene Koonin
The authors of this manuscript strive to provide an explanation for the domination of bilateral symmetry in animals. The come up with the idea that bilateral symmetry provides for by far greater ability to swiftly change the direction of movement than any other body plan, hence a substantial advantage for free moving animals. The authors submit that this is a major factor behind the near ubiquity of bilateral symmetry but they are careful in indicating that other factors could be important as well. In my view, this is an interesting and sensible hypothesis although I think that it would gain in strength should the authors coach their hypothesis in specific equations of mechanics.
Authors’ response: We thank Dr. Koonin for taking charge of the review of the manuscript. While formulating the hypothesis we constantly endeavoured to provide the simplest explanation for the problem in the clearest way. We think the basic statements (e.g. the body has to overcome drag; to push itself in a new direction it has to exercise a force in the opposite direction) and the equation of drag with some other minor considerations are necessary and sufficient arguments for the explanation of the theory, but should it require a more detailed rationale we would be grateful for more specific instructions. Please also consider that this is the main hypothesis that may serve as a basis for more specific future analyses for particular scenarios.
1 and Mihály Novák2,3