June 15, 2002;
GABA and development of the Xenopus optic projection.
In the developing visual system of Xenopus laevis retinal ganglion
) axons extend through the brain
towards their major target in the midbrain
, the optic tectum
. Enroute, the axons are guided along their pathway by cues in the environment. In vitro, neurotransmitters have been shown to act chemotropically to influence the trajectory of extending axons and regulate the outgrowth of developing neurites, suggesting that they may act to guide or modulate the growth of axons in vivo. Previous work by Roberts and colleagues (1987) showed that populations of cells within the developing Xenopus diencephalon
express the neurotransmitter gamma amino butyric acid (GABA). Here we show that Xenopus RGC
axons in the midoptic tract grow alongside the GABAergic cells and cross their GABA immunopositive nerve
processes. Moreover, RGC
axons and growth cones express GABA-A and GABA-B receptors, and GABA and the GABA-B receptor agonist baclofen both stimulate RGC neurite
outgrowth in culture. Finally, the GABA-B receptor antagonist CGP54626 applied to the developing optic projection
in vivo causes a dose-dependent shortening of the optic projection
. These data indicate that GABA may act in vivo to stimulate the outgrowth of Xenopus RGC
axons along the optic tract.
[+] show captions
Figure 1 Developmental expression of GABA in the optic tract coincides spatially and temporally
with the developing optic projection. Whole-mount double immunohistochemistry for GABA
(green) and the HRP-labeled optic projection (red) at stages 35/36 (A) and 37/38 (B). (A1–B1).
Lateral view of double-immunolabeled wholemount Xenopus brains double-immunolabeled for
GABA and HRP. (A2–B2). GABA immunocytochemistry. GABA positive cells are found in the
mid-optic tract, tectum as well as in the hindbrain. (A3–B3): HRP-labeled optic projection. (C–D).
Double immunocytochemistry for GABA (green) and HRP-labeled RGC axons (red) in a transverse
cryostat section through the brain and eyes of a stage 35/36 embryo. RGC axons grow alongside
GABAergic cells in the mid-diencephalon. Amacrine cells in the inner nuclear layer (INL) of the
retina are also immunopositive for GABA. (D) is a 1-m confocal section of RGC axons growing in the
mid-diencephalon. RGC axons grow either in close proximity to GABAergic cells in the diencephalon
(arrow), or alongside GABA-positive processes in the neuropil (arrowhead). Di, diencephalon; Np,
neuropil; Tec, tectum; Tel, Telencephalon; Hb, hindbrain; A, anterior; V, ventral; Mot, mid optic tract.
Scale bar in (B3) is 100 m for (A–B) Scale bar in (C) is 30 m, and bar in (D) is 25 m.
Figure 2 The GABA-A and GABA-B receptor antibodies
recognize Xenopus proteins of the appropriate size. Western
blots of stage 35/36–37/38 Xenopus head tissue showing
GABA-A (A) and GABA-B (B) receptor protein expression.
Molecular weights based on standards run at the same
time as the sample tissue are indicated. (A) A prominent
protein band with an approximate molecular weight of 58
kDa was observed, which corresponds to the weight of
known GABA-A receptor -subunits. (B) A single band
with an approximate molecular weight of 100 kDa was
observed using a polyclonal antibody (Santa Cruz) against
the GABA-B receptor.
Figure 3 GABA receptors are expressed by developing Xenopus RGCs. GABA-B receptor
expression as revealed by immunocytochemistry with an antibody (Santa Cruz) against the
GABA-B receptor. (A–B) Transverse sections through a stage 35/36 Xenopus eye and brain.
GABA-B receptor immunoreactivity was found in the diencephalic neuropil (A), the optic nerve
[see arrowheads in (B)], and the retina. (C) No retinal labeling was seen in a similar section when
the primary antibody was preincubated with a blocking peptide (Santa Cruz). (D) GABA-B receptor
immunolabeling is seen in the stage 37/38 embryonic Xenopus retina shown at higher power.
Immunoreactivity was observed in the inner nuclear layer (INL), retinal ganglion cell layer (RGCL),
and the inner (IPL) and outer (OPL) plexiform layers. The optic nerve head (onh) was also
immunopositive (arrow). Di, diencephalon; PE, pigment epithelium; L, lens; Np, neuropil; D,
dorsal; V, ventral. Scale bar in (D) is 200 m for (A)–(C), and 100 m for (D).
Figure 4 GABA receptors are expressed by developing
Xenopus RGC growth cones. Stage 24 eye buds were cultured
as explants for 40–48 h and processed for GABA
receptor immunocytochemistry. (A) The GABA-A receptor
antibody revealed relatively uniform labeling of RGC axons
and growth cones. (B) Intense immunolabeling for GABAB
receptors was found in the axons and body of RGC
growth cones, with lower level labeling in the filopodia and
lamellopodia. (C–D) GABA-B receptor immunoreactivity
in a growth cone (C) could be eliminated by incubation with
a blocking peptide for the GABA-B receptor antibody (D).
Scale bar in (A) is 20 m for (A) and (B), and 25 m for
(C) and (D).
Figure 6 CGP54626 applied to the exposed brain preparation inhibits the outgrowth of RGC
axons in vivo: the GABA-B receptor antagonists phaclofen (200–500 M) and CGP54626 (100–
500 M) were applied for 20 h to the developing optic projection starting at stage 33/34 when the
RGC axons are first entering the contralateral brain. Shown are lateral views of the HRP-labeled
optic projection in wholemount stage 40 brains (A–B) exposed to either control solution (A) or 500
M phaclofen (B). No obvious effect on the optic projection was observed. (C) Optic projection
exposed to a control solution containing the same DMSO concentration as the 500 M CGP54626
exposed brain in (F). (D–F) Optic projections exposed to either 100 M CGP54626 (D), 250 M
CGP54626 (E), or 500 M CGP54626 (F). Scale bar in (F) is 150 m. Tec, tectum; Tel,
telencephalon; ot, optic tract.
Figure 7 (A–F) Immunolabeling with various neuroepithelial cell markers indicates that the
neuroepithelium was not obviously affected by treatment with 250 and 500 M CGP54626. (A–B)
Immunocytochemistry with an antibody against islet-1 shows ventrally located cells in both control
(0.5% DMSO)- (A) and 500 M CGP54626- (B) treated brains. (C–D) Zn-12, a general neural
marker, exhibits comparable labeling in control (1% DMSO)- (C) and 250 M CGP54626- (D)
treated brains. (E) Unexposed (UnEx) and exposed (Ex) sides of a 250 M CGP54626-treated brain
show comparable immunolabeling with an antibody against -tubulin. (F) GABA immunoreactivity
is unaffected by treatment with 500 M CGP54626. Bar in F is 100 m. (G) Mean normalized optic tract
length in control and GABA-B receptor antagonist treated brains. For each CGP54626
concentration, control brains were exposed to equivalent levels of the DMSO carrier. CGP54626
caused a dose-dependent shortening of the optic projection, whereas phaclofen had no obvious effect
on the optic projection. *p .05, **p .01 (unpaired, two-tailed Student t test). Numbers in
brackets are the number of embryos from three to five separate experiments.