The invertebrate morphogenesis
The second section of Fleury’s paper is about The invertebrate morphogenesis and the example chosen is Drosophila the fly, which is the most studied of the invertebrates but far from describing the “invertabrates” morphogenesis in general, just take a look at a sponge. Two subsections, The fly bauplan and The homeobox genes.
2.1 The ﬂy bauplan
The use of the term “bauplan” is dangerous and Fleury is aware of it, but he consider its use safe if global animal patterns (such as bilateral, radiate, etc.) are determined by general principles of symmetry [p5, footnote 8].
By deciding to use it (safely) he put himself in charge to prove that global animal patterns are determined by general principles of symmetry. Not described, determined.
That’s one of the main objects of the paper (if not the main) hidden in a footnote! Genetic specification/determinism of biological constituents seems beyond Fleury’s understanding and this lead him to some quite interesting assertions.
Is it true that “The initial “formless” animal is either round, as in vertebrates, or oblate (cigar-like), as in the fruit ﬂy (and also in C. Elegans1).”. No, unless you consider formless (or “formless”) this, which is a Drosophila egg, from a dorsal point of view, with its respiratory appendages positioned anteriorly.
I’ve being a cigar amateur for over 20 years and never came across a cigar looking like a Drosophila egg.
Is it true that “[…] the ﬂy ovocyte (syncitium), is roughly a one dimensional tube?” No, certainly not a tube (not hollow you see).
Concerning the definition of the antero-posterior axis Fleury got it almost right. Except that the gradient is not really the one of bicoid which is concentrated anterio-dorsaly (see fig. 3, upper panel), around the oocyte’s nucleus, but of bicoid, the protein, which diffuse in an anterio-posterior direction, probably not freely but transported along the microtubules, and producing a gradient. And bicoid is far from being the only gradient along the AP axis (just a flavor):
He seems unaware of the dorso-ventral polarization of the oocyte, or that both axes are defined during oogenesis.
He rightly indicates that bicoid is missing in other insects, but wrongly states that “some other gene has to take over this role in other species .“, the innovation being bicoid in Diptera.2
Let’s give a closer look at the cigar3:
Fig. 24. A diagram of a stage 10a Drosophila egg chamber showing localized signals that polarize the AP and DV axes of the embryo. bicoid mRNA (blue), oskar mRNA (red), gurken mRNA (green); pipe expression (dark green); and torsolike expression (magenta).
OK, that’s just a diagram, let’s go for the real thing, always much better:
This image is from the website of the Institute Jacques Monod, Antoine Guichet’s lab website; click on the thumbnail to visit. From Fleury’ lab to the IJM the distance is short. Maybe he will find somebody available to show him some Drosophila eggs.
So, Fleury’s description is at least partial if not simply wrong and I hope nobody else then Fleury will use it. It is dangerous to present to a naive reader such a nasty description of an object, as it may lead him to wrong conclusions. We have a case of SISO and one may wonder what will come out of Fleury’s description.
I will not repeat here the problems of figure 3, you can check that here.
Even the description of the bicoid protein gradient along the embryo is simplistic; it corresponds at the time where the steady-state is achieved, but lacks an interesting part of the developmental process, between the start of translation of bicoid and the reach of the steady stady-state of the gradient (if we assume that such a steady-stae is achieved).
Why is bicoid considered as a morphogenic gene? The author don’t mention it but it’s an important part of the story. That bicoid specify head’s morphogenesis.
Let’s see three of the assertions of the author and comment on them:
- In the same spirit, inside each stripe, cells are seen to align by local rearrangements.
- The stripes of expression correspond to actual cellular lines, which are congruent with concentration level lines.
- Current genetic networks do not take into account this cellular reorganization, likely linked to compression forces.
- So, “cells are seen to align by local rearrangements“, now that would be news. In fact stripes are produced by cells already in place by differential genes’ expression.
- The second phrase is correct (however a little bit fuzzy) but contradicts the first one.
- “likely linked to compression forces“. I wonder where this “likely” comes from. No reference and above all no explanation of where the “compression forces” come from and what specifies them.
We have here one more example of the rhetoric approach of Fleury, not willing to accept genetic determinism, and trying to cover his stance.
2.2 The homeobox genes
How many homeobox genes for the Drosophila melanogaster? Thirty. Check that, don’t believe me and certainly not Fleury, who presents just eight of them, belonging to “a single cluster of 8 genes“, which was probably the case in some ancestral animal but was divided in Drosophila in two complexes, Antp-C, controling specification of anterior regions (head and first thoracic segment) (labial, proboscipedia, deformed, sex combs reduced and antennapedia) and BX-C, controling specificationn of posterior regions (second and third thoracic, and abdominal segments) (ultrabithorax, abdominal-A and abdominal-B).
Domains of expression may overlap, or not, depending initially on the activity of selector genes, during the first hours of embryonic development, then on the activity of homeobox genes themselves.
The author have another way to present it:
This means physically that there is a spatial arrow at the scale of an organ or body part, which is somehow converted into a spatial arrow at the scale of the DNA, and/or vice-versa.
This is probably the weirdest phrase of the paper.
I do see clearly how gene’s expression can specify identity of body parts along the antero-posterior axis ; I suppose that’s what Fleury calls “spacial arrow at the scale of an organ or body part“.
“somehow converted into a spatial arrow at the scale of the DNA“?! Well… If you do have an explanation for this part the comments are open.
It is not just assumed that chemical gradients are chemical cues driving differential gene expression, it is proven that this is so.
It was also proven that mechanical cues participates to differential gene expression. And this is not the only physical factor influencing developmental stages,e.g. temperature is another one.
What seems to be of particular interest for Fleury are the genetically controlled endogenous morphogenetic movements described by Emmanuel Farge , which set’s unambiguously the information flow from the genome to the endogenous morphogenetic movements.
The specification of body segments by the expression of the Antp-C and Bx-C genes is presented in a rather weird way and certainly containing a huge mistake which is presented extensively here. The author not only fails to make his point that topological differences impairs homeotic transformations but demonstrates both a lack of knowledge of the subject he is dealing with and inability to use the scientific literature for a reality-check of his “points of view” before submitting the for publication. By the same, we have a glimpse of the papers reviewers inability to judge the quality of the author’s work.
Following his idea of mechanical forces as morphogenetic factors, without considering the genetic determinism of processes specifying these mechanical forces, the author attacks paired appendages morphogenesis. One can certainly define a basic appendage and trace it back to the first bilaterian which possessed it, and may even call this last common ancestor an archetype (in Darwin’s sense of the term, certainly not as a platonic archetype).
There are errors also in this part of the text but I hope some referee will see to correct them.
The last paragraph of this section is probably the most interesting and I’ll focus on it.
As the author explains, considerable efforts are dedicated to understand the specification of appendages growth and morphology in terms of genetic information required for. This will certainly lead to understand the specification of genetically controlled endogenous morphogenetic movements and their role in development as positional information cues driving cellular differentiation.
One expect from physicists interested by the subject to closely collaborate with evo-devo specialists and geneticists to help understand what mechanical elements are important and what are the genetic elements specifying them through the development of particular “physical phenotypes” and mechanosensitivity, at the molecular level.
Reminders of possibilities unconnected to the subject via experimental evidence may be used to built hypothetical theoretical models but at some point they should be connected with specific experimental data to be useful.
1. species names in italics, please; and eventually the full name of the species for the first occurrence in a text, especially if the public isn’t accustomed with biology (e.g. usual readers of a physical journal)
2. Patterns of conservation and change in honey bee developmental genes, Dearden PK, Wilson MJ, Sablan L, Osborne PW, Havler M, McNaughton E, Kimura K, Milshina NV, Hasselmann M, Gempe T, Schioett M, Brown SJ, Elsik CG, Holland PW, Kadowaki T, Beye M., Genome Res. 2006 Nov;16(11):1376-84
bcd is missing from the honey bee genome, a not unexpected finding given previous studies indicating that this gene evolved its anterior role in Diptera (Stauber et al. 2000, 2002). Its absence in the honey bee implies that other genes must carry out its function, possibly orthodenticle and Hunchback as has been shown in Tribolium (Schroder 2003) and Nasonia (Pultz et al. 2005).
3. So you have the complete solution for my quizz
4. Daniel St Johnston, The beginning of the end, The EMBO Journal Vol. 20 No. 22 pp. 6169±6179, 2001
Next time I meet someone who seems to have an arse where their head should be I’m gonna call them a son-of-a-bicoid!
Maybe I’ll add a son-of-a-bicoid category at coffeandsci.