March 15, 2001;
Nerve-independence of limb regeneration in larval Xenopus laevis is correlated to the level of fgf-2 mRNA expression in limb tissues.
In both larval and adult urodele amphibians, limb blastema
formation requires the presence of an adequate nerve
supply. In previous research, we demonstrated that the hindlimb
of early Xenopus laevis larvae formed a regeneration blastema
even when denervated, while the denervated limb
of late larvae did not. We hypothesized that the nerve
-independence was due to the autonomous synthesis of a mitogenic neurotrophic-like factor by undifferentiated limb bud
cells. In this paper, we demonstrate that fgf-2
mRNA is present in larval limb
tissues and that its level is correlated to the extent of mesenchymal cells populating the limb
: in early limbs, fgf-2
mRNA is present at high levels all over the limb
, while, in late limbs, the fgf-2
expression is low and detectable only in the distal autopodium
. After denervation, fgf-2
mRNA synthesis increases in amputated early limbs but not in amputated late limbs. The implantation of anti-FGF-2
beads into amputated early limbs hardly lowers the mitotic activity of blastema
cells. However, FGF-2
beads implanted into the blastema
of late limbs prevent the denervation-induced inhibition of mitosis and oppose blastema
regression. Our data indicate that FGF-2
is a good candidate for the endogenous mitogenic factor responsible for blastema
formation and growth in amputated and denervated early limbs. However, in amputated late limbs, the very limited fgf-2
expression is not sufficient to promote blastema
formation in the absence of nerves.
[+] show captions
FIG. 1. Whole-mount in situ hybridization with fgf-2 mRNA (A) and fgf-8 mRNA antisense probes (L, M). (A) Normal hindlimbs from larvae at stages 52, 55, and 57, respectively. (A and B) At stage 52 and 55, fgf-2 mRNA expression is detectable in whole the limb, but a distal to proximal gradient of the fgf-2 expression pattern is visible. (C) At stage 57, a low level of fgf-2 mRNA is observed only at level of the distal autopodium (toes). (D) Regenerating hindlimbs amputated through the tarsalia and denervated, stage 52. (D and E) Three and five days after operation, respectively. A high level of fgf-2 mRNA expression is observed both in the stump and in the regeneration blastema. Red lines indicate the approximate amputation level. (F) Longitudinal section of the same blastema in E. fgf-2 mRNA is confined to mesenchymal cells in the stump and in the regeneration blastema (arrowheads). In the procartilaginous skeleton of the stump (arrow), fgf-2 mRNA is undetectable. The line marks the approximate boundary between the stump and regeneration blastema. (G) Hindlimbs amputated through the tarsalia, stage 57. (G) Three days after amputation and sham-denervation: fgf-2 mRNA expression is evident only in the most distal part of the stump in close proximity to the amputation surface. (H) Nine days after amputation and sham-denervation: fgf-2 mRNA is undetectable both in the stump and in the blastema (arrow). (I) Nine days after amputation and denervation. No blastema is present and fgf-2 mRNA is undetectable in the stumped limb. (L and M) fgf-8 expression in hindlimbs of stage 52 larvae. In the normal limb (L), fgf-8 is expressed in the distal epidermis. In the regenerating limb (M), fixed 4 days after amputation, fgf-8 is expressed in the apical epithelial cap covering the regeneration blastema. rb, regeneration blastema; st, stump. Scale bar, 500 ﰁm.
FIG. 2. Quantitative radioactive RT-PCR analysis: graphic representation
of fgf-2 mRNA levels, expressed as percentage of
ornithine decarboxylase (ODC) mRNA, in normal limbs, amputated
and sham denervated limbs, and amputated and denervated
limbs of stage 52 and 57 X. laevis larvae sacrificed 3 days after
operation. After amputation and denervation, fgf-2 mRNA level
increases in early limbs but not in late limbs. Each column
represents the mean value 6 SD obtained from RT-PCRs performed
on three different pools of ten limbs.
FIG. 3. Polysomes/mRNPs distribution of fgf-2 mRNA. (A) Absorbance
profile of a sucrose gradient showing the polysomal
pattern of cytoplasmic extracts from stage 52 limbs. (B) Quantitative
radioactive RT-PCR analysis of the nine gradient fractions
with primers specific for fgf-2 mRNA from cytoplasmic extracts of
amputated and sham denervated limbs and of amputated and
denervated limbs. In amputated and denervated limb extracts, fgf-2
mRNA is mainly associated with polysomes.
FIG. 4. Effect of implantation of anti-FGF2- or FGF-2-containing
beads on blastema cell proliferation. (a—c) Mitotic indices in the
blastemas of limbs amputated at stage 52 and fixed 4 days after
bead implantation and 5 days after amputation. The MI of denervated
blastemas implanted with anti-FGF2-soaked beads (c) is
significantly lower (P , 0.001) than the MI of both denervated (b)
and sham-denervated (a) blastemas implanted with IgG-soaked
beads. (d—f) Mitotic indices in the blastemas of limbs amputated at
stage 57 and fixed 4 days after bead implantation and 10 days after
amputation. The MI of denervated blastemas implanted with
FGF-2-soaked beads (f) is significantly higher (P , 0.001) than the
MI of denervated blastemas implanted with BSA-soaked beads (e)
and superimposable to the MI of sham-denervated blastemas implanted
with BSA-soaked beads (d). Each bar expresses the mean 6
SD of at least five limbs.
FIG. 5. Longitudinal sections of limbs amputated at stage 57 and
fixed 4 days after bead implantation and 10 days after amputation.
(A) Sham-denervated blastema implanted with BSA-soaked beads.
(B) Denervated blastema implanted with FGF-2-soaked beads. (C)
Denervated blastema implanted with BSA-soaked beads. While the
blastema in C is regressing, the blastema in B is still well developed.
The section in B shows two implanted beads (arrow). Broken lines
mark the approximate amputation level. H&E. Scale bar, 100 mm.