4 RA Signaling in Limb Development and Regeneration …
105
mesoderm and later in the proximal region of the limb bud. In contrast, there is no
meis expression in the distal end of the bud.
Interestingly, meis2 is a RA responsive gene (Oulad-Abdelghani et al. 1997).
Accordingly, insertion of beads embedded with RAR antagonists in the proximal
region or in the lateral plate, led to the repression of meis genes. In contrast, insertion
of RA beads in distal regions of developing limb buds, up-regulated the expression
of meis1 and meis2 in the distal limb. Remarkably, after RA exposure, the meisexpressing cells showed a proximal relocation.
Then, the question was how meis expression is restricted to proximal limb cells
during limb growth? Several studies showed that the AER restricts meis expression
via fgf8. Indeed, meis expression domains expanded distally after AER removal or
insertion of beads soaked with an inhibitor of fgf signaling. It has been suggested
that FGFs promote limb distalization by counteracting the RA pathway, which is
essential to maintain the proximalizating activity of meis. Thus, the proximodistal
development of the limb appears to be controlled by opposing RA and FGF signals
though the regulation of meis genes (Rosello-Diez et al. 2011; Cooper et al. 2011).
Figure 4.4 summarizes the distribution of the RA enzymatic machinery and the
effect of an excess or a deficiency of RA on the chick limb bud. A summary of the
RA-dependent gene cascades is shown in Fig. 4.5.
Mouse Limb Development
Is RA Required?
The role of RA in mouse limb development has been explored using gene knockouts
since mouse limb bud cannot be surgically manipulated in vivo. However, solely
deleting genes encoding enzymes that synthesize RA failed to reveal the effects of
RA on limb development because the knockout embryos died soon after limb bud
formation.
Just before death, raldh2
−/− embryos invalidated for retinaldehyde dehydrogenase
2, which oxidizes retinaldehyde into RA, showed hypoplastic forelimb buds and no
sign of hindlimb buds due to a severely shortened posterior region (Niederreither
et al. 1999). Raldh2
−/− mutants crossed with RARE-lacZ reporter mice showed a
complete absence of RA production in the lateral plate mesoderm adjacent to where
the limb buds will appear (Mic et al. 2004). Tbx5, the gene involved in limb initiation,
meis2, fgf-8 and dHand expression were all eliminated in the presumptive forelimb
region of these embryos.
Upon rescuing the RA-deficient embryos with maternal RA administration, development continued further, the forelimbs depicting a range of defects whereas the
hindlimbs were normal. The forelimbs exhibited a range of AP and PD axis defects
that were lessened by increasing the dose of RA administered or by varying the length
of time of dosing. Over a low to high RA concentration/exposure range, the forelimb
defects ranged from (a) no humerus, ulna and single digit present; to (b) a partial
105
mesoderm and later in the proximal region of the limb bud. In contrast, there is no
meis expression in the distal end of the bud.
Interestingly, meis2 is a RA responsive gene (Oulad-Abdelghani et al. 1997).
Accordingly, insertion of beads embedded with RAR antagonists in the proximal
region or in the lateral plate, led to the repression of meis genes. In contrast, insertion
of RA beads in distal regions of developing limb buds, up-regulated the expression
of meis1 and meis2 in the distal limb. Remarkably, after RA exposure, the meisexpressing cells showed a proximal relocation.
Then, the question was how meis expression is restricted to proximal limb cells
during limb growth? Several studies showed that the AER restricts meis expression
via fgf8. Indeed, meis expression domains expanded distally after AER removal or
insertion of beads soaked with an inhibitor of fgf signaling. It has been suggested
that FGFs promote limb distalization by counteracting the RA pathway, which is
essential to maintain the proximalizating activity of meis. Thus, the proximodistal
development of the limb appears to be controlled by opposing RA and FGF signals
though the regulation of meis genes (Rosello-Diez et al. 2011; Cooper et al. 2011).
Figure 4.4 summarizes the distribution of the RA enzymatic machinery and the
effect of an excess or a deficiency of RA on the chick limb bud. A summary of the
RA-dependent gene cascades is shown in Fig. 4.5.
Mouse Limb Development
Is RA Required?
The role of RA in mouse limb development has been explored using gene knockouts
since mouse limb bud cannot be surgically manipulated in vivo. However, solely
deleting genes encoding enzymes that synthesize RA failed to reveal the effects of
RA on limb development because the knockout embryos died soon after limb bud
formation.
Just before death, raldh2
−/− embryos invalidated for retinaldehyde dehydrogenase
2, which oxidizes retinaldehyde into RA, showed hypoplastic forelimb buds and no
sign of hindlimb buds due to a severely shortened posterior region (Niederreither
et al. 1999). Raldh2
−/− mutants crossed with RARE-lacZ reporter mice showed a
complete absence of RA production in the lateral plate mesoderm adjacent to where
the limb buds will appear (Mic et al. 2004). Tbx5, the gene involved in limb initiation,
meis2, fgf-8 and dHand expression were all eliminated in the presumptive forelimb
region of these embryos.
Upon rescuing the RA-deficient embryos with maternal RA administration, development continued further, the forelimbs depicting a range of defects whereas the
hindlimbs were normal. The forelimbs exhibited a range of AP and PD axis defects
that were lessened by increasing the dose of RA administered or by varying the length
of time of dosing. Over a low to high RA concentration/exposure range, the forelimb
defects ranged from (a) no humerus, ulna and single digit present; to (b) a partial