And this is a long one Oooooooops, so under the fold it goes.
Fig. 17. (Color online) Reprinted from reference . A mutant mouse (gli3-/-) having 9 digits in its digital arch (“M” in the original caption). The hand shape reﬂects the shape of the early limb bud (“H”), and the number of digital rays equates (in the background) the number of somites at the connection between ﬂanks and bud. The ﬁgure to the right shows an overlap of the hand, at the stage “M” of the growth, over the initial limb bud along the ﬂank, showing that the digital arch spans exactly the early form of the limb bud (image obtained by centering, scaling and averaging M and H), with the distal-most phalanges following exactly the apical ectodermal ridge, where nails form. [...]
This is from reference , which is: Hoxd and Gli3 interactions modulate digit number in the amniote limb, Rushikesh Sheth, M. Félix Bastida, Marian Ros, Developmental Biology 310 (2007) 430 – 441, doi:10.1016/j.ydbio.2007.07.023
Panel “M”, at the left, is from fig. 1 of , and corresponds to Gli3XtJ/XtJ; HoxdDel(11-13)/+.
The Gli3XtJ/XtJ;+/+ mouse forelimb is from fig. 1:
Panel “H”, at the middle, is from fig. 3 of , and corresponds to Gli3XtJ/XtJ; HoxdDel(11-13)/Del(11-13)
And one more bud, the Gli3XtJ/XtJ one, from fig. 3
Nobody during the review process noticed that Fleury use stuff from two different mutants?!
Does it matter? Probably not, but just try to play the tricky little game Fleury played superposing the developed forelimb to the limb bud. Enjoy!
In the text one can read [p18, col2, §1]:
For example the forelimb pattern of the gli3-/- mouse which has 9 ﬁngers is that of a regular fan of digital rays forming an arch. This arch overlaps exactly the shape of the early limb bud, at the moment when it outgrows from the ﬂanks, and this number corresponds to the number of vertebrae precursors located parallel to it.
Well, not as regular as he would like it to be. And how many vertebrae precursors located parallel to the gli3-/- bud?
Here, let’s go for a reality check; the vertebrae region is contrasted to help you count:
On the left the wild type and on the right the double mutant Gli3XtJ/XtJ; HoxdDel(11-13)/Del(11-13). Well, OK, no correlation between the vertebrae and fingers numbers, who would think otherwise and don’t check?
Now, as the comparison is made easy, check the arc covered by Fgf4 between wt and mutant [where the authors point the arrowhead]. Much larger in the mutant, no ? Check also the Grem1 expression domain, a slightly larger arc then the one of Fgf4 cover and also much larger in the mutant.
Let me help you with that:
And a nice gradient of Grem1 expression with that:
Bottom lines list:
- check your mutants when you want to superimpose stuff, don’t do it the Fleury way,
- always check the control when you imagine some relation, say between vertebrae and fingers
- check the correlations the authors point to: some make good observations (Sheth et al., Fgf4/Grem1 arc), other are bogus.
Filed under: Oops!