4 RA Signaling in Limb Development and Regeneration …
91
limb, the result is identical to grafting a ZPA (Tickle et al. 1982; Summerbell 1983;
Fig. 4.1d). Also identical to grafting a ZPA, a RA bead placed on the posterior side of
the limb has no effect and a RA bead placed in the center of the limb is considerably
less effective. Increasing concentrations of RA, from 1 to 25 nM, produces increasing
numbers of extra digits in a precise, bilaterally, symmetrical pattern, such that low
concentrations yield a digital formula of 2, 2, 3, 4; mid concentrations yield 3, 2, 2, 3,
4; and high concentrations yield 4, 3, 2, 2, 3, 4. These dose-dependent RA responses
were equivalent to grafting increasing numbers of ZPA cells (Tickle 1981).
Then, it was demonstrated that application of beads soaked with radiolabeled RA
to developing limbs, produces an exponential gradient of RA across the AP axis
(Tickle et al. 1985). Following this discovery, Eichele’s lab mounted a herculean
effort to measure endogenous RA levels across the AP axis of the limb bud, using
high pressure liquid chromatography (HPLC). He found that endogenous RA is
distributed across the AP axis of the limb bud with the same graded form as that
which results from exogenous application of radiolabeled RA (Thaller and Eichele
1987). These striking results led to the statement ‘We have a morphogen’ (Slack
1987), which, in turn, was strong evidence that the mechanism for ZPA effects was
solved!
Interestingly, in the regenerating amphibian limb, RA showed the same graded
concentration effects, but in the PD axis rather than the AP axis (Maden 1982, 1983a).
After amputation of the amphibian limb through the middle of the forearm, the lower
arm and hand regenerate (Fig. 4.2a). When RA is applied either by intraperitoneal
injection or by application directly to the blastema (analogous to application of the
RA beads in the chick limb bud), a complete limb regenerates, starting from the
proximal end of the humerus (Fig. 4.2d). If a lower concentration of RA is used, the
phenomenon is less extreme and the regeneration occurs from the distal end of the
humerus, rather than the proximal end (Fig. 4.2c). If the concentration is even lower,
then the regenerated radius and ulna are longer than normal (Fig. 4.2b). Strikingly,
the difference in concentration between a minor effect and the maximal effect is only
about 2.5 fold (lower panel of Fig. 4.2). The most extreme effect is seen when the
limb is amputated through the hand, a relatively high concentration of RA is added,
and a complete limb is regenerated from the level of the carpals.
Development of the Field
Differences between model systems and methods for studying sites of RA action
have fueled controversies on the exact function of RA in the limb. The pioneering
studies carried out in the 1980s made it clear that RA had different effects on the
developing chick limb (morphogen of the AP axis) and the regenerating salamander limb (morphogen of the PD axis), but the work did not clarify whether these
differences were the consequence of developmental stage (embryonic vs postnatal)
or perhaps, animal model (chick vs salamander). New approaches and studies were
needed to address the fundamental questions raised by the early findings:
91
limb, the result is identical to grafting a ZPA (Tickle et al. 1982; Summerbell 1983;
Fig. 4.1d). Also identical to grafting a ZPA, a RA bead placed on the posterior side of
the limb has no effect and a RA bead placed in the center of the limb is considerably
less effective. Increasing concentrations of RA, from 1 to 25 nM, produces increasing
numbers of extra digits in a precise, bilaterally, symmetrical pattern, such that low
concentrations yield a digital formula of 2, 2, 3, 4; mid concentrations yield 3, 2, 2, 3,
4; and high concentrations yield 4, 3, 2, 2, 3, 4. These dose-dependent RA responses
were equivalent to grafting increasing numbers of ZPA cells (Tickle 1981).
Then, it was demonstrated that application of beads soaked with radiolabeled RA
to developing limbs, produces an exponential gradient of RA across the AP axis
(Tickle et al. 1985). Following this discovery, Eichele’s lab mounted a herculean
effort to measure endogenous RA levels across the AP axis of the limb bud, using
high pressure liquid chromatography (HPLC). He found that endogenous RA is
distributed across the AP axis of the limb bud with the same graded form as that
which results from exogenous application of radiolabeled RA (Thaller and Eichele
1987). These striking results led to the statement ‘We have a morphogen’ (Slack
1987), which, in turn, was strong evidence that the mechanism for ZPA effects was
solved!
Interestingly, in the regenerating amphibian limb, RA showed the same graded
concentration effects, but in the PD axis rather than the AP axis (Maden 1982, 1983a).
After amputation of the amphibian limb through the middle of the forearm, the lower
arm and hand regenerate (Fig. 4.2a). When RA is applied either by intraperitoneal
injection or by application directly to the blastema (analogous to application of the
RA beads in the chick limb bud), a complete limb regenerates, starting from the
proximal end of the humerus (Fig. 4.2d). If a lower concentration of RA is used, the
phenomenon is less extreme and the regeneration occurs from the distal end of the
humerus, rather than the proximal end (Fig. 4.2c). If the concentration is even lower,
then the regenerated radius and ulna are longer than normal (Fig. 4.2b). Strikingly,
the difference in concentration between a minor effect and the maximal effect is only
about 2.5 fold (lower panel of Fig. 4.2). The most extreme effect is seen when the
limb is amputated through the hand, a relatively high concentration of RA is added,
and a complete limb is regenerated from the level of the carpals.
Development of the Field
Differences between model systems and methods for studying sites of RA action
have fueled controversies on the exact function of RA in the limb. The pioneering
studies carried out in the 1980s made it clear that RA had different effects on the
developing chick limb (morphogen of the AP axis) and the regenerating salamander limb (morphogen of the PD axis), but the work did not clarify whether these
differences were the consequence of developmental stage (embryonic vs postnatal)
or perhaps, animal model (chick vs salamander). New approaches and studies were
needed to address the fundamental questions raised by the early findings: