Expression of Hoxc6 at rostral thoracic levels failed to suppress
Hoxc8 and conversely, ectopic expression of Hoxc8 at rostral
brachial levels did not suppress Hoxc6 or RALDH2 expression
(Supplementary Fig. S6f–i). In contrast, expression of Hoxc8 at
rostral brachial levels suppressed Hoxc5 expression (Supplementary
Fig. S6j), providing evidence that the exclusive domains of
expression of Hoxc5 and Hoxc8 within the LMC are also established
by selective cross-repressive interactions.
Hox6 and Hox9 expression and regulation in neural progenitors
Does the profile of Hox expression in post-mitotic MNs emerge
from an earlier rostrocaudal difference in Hox6 and Hox9
expression by neural progenitors? Expression of
was absent from brachial progenitor cells between stages 10 to 14,
whereas expression of
messenger RNA was detected in
thoracic progenitors (Supplementary Fig. S5e–g). Hoxc6 protein,
however, was not detected in thoracic progenitors (Fig. 4f).
mRNA and Hoxc9 protein
was detected in thoracic neural progenitors at stage 15, just before
the generation of post-mitotic MNs (Fig. 4a–c, e). Each
was expressed over the same general domain, with a rostral limit
of expression at somites 20/21 (Fig. 4a–c). Taken together, these
findings indicate that brachial progenitors lack expression of Hox6
and Hox9 proteins, whereas thoracic progenitors express Hox9 but
not Hox6 proteins (Fig. 4j).
We next examined whether FGF8 signalling changes the profile of
gene expression in neural progenitors. After brachial
Rapid repatterning of neural progenitor Hox profile after brachial
, Rostral expansion in domains of expression of
paralogues and of
, 8 h after unilateral
electroporation. Red arrowhead indicates normal rostral
expression boundary, green arrowhead indicates rostral boundary after exposure to
, Expression of Hoxc9 and absence of Hoxc6 protein in stage 15 thoracic neural tube.
, Ectopic brachial expression of Hoxc9 protein, but not Hoxc6 protein, 8 h after
electroporation in stage 12 neural tube. The restriction of the repressive activity of Hox9
proteins to post-mitotic neurons could reflect gating of Hox activities by co-factors such as
, a possibility supported by the restricted expression of Meis1 and
Pbx3 in MNs (J.S.D and T.M.J., unpublished observations).