1123
Development 130, 1123-1133
© 2003 The Company of Biologists Ltd
doi:10.1242/dev.00327
Molecular analysis of axon repulsion by the notochord
Christopher N. G. Anderson*, Kunimasa Ohta, Marie M. Quick, Angeleen Fleming, Roger Keynes and
David Tannahill†
Department of Anatomy, University of Cambridge, Downing Street, Cambridge CB2 3DY, UK
*Present address: Centre for Cell and Molecular Dynamics, Department of Anatomy and Developmental Biology, University College London, 5th Floor Rockefeller
Building, Gower Street, London WC1E 6BT, UK
†Author for correspondence (e-mail: [email protected])
Accepted 29 November 2002
SUMMARY
During development of the amniote peripheral nervous
system, the initial trajectory of primary sensory axons is
determined largely by the action of axon repellents. We
have shown previously that tissues flanking dorsal root
ganglia, the notochord lying medially and the
dermamyotomes lying laterally, are sources of secreted
molecules that prevent axons from entering inappropriate
territories. Although there is evidence suggesting that
SEMA3A contributes to the repellent activity of the
dermamyotome, the nature of the activity secreted by the
notochord remains undetermined. We have employed an
expression cloning strategy to search for axon repellents
secreted by the notochord, and have identified SEMA3A as
a candidate repellent. Moreover, using a spectrum of
different axon populations to assay the notochord activity,
together with neuropilin/Fc receptor reagents to block
semaphorin activity in collagen gel assays, we show that
SEMA3A probably contributes to notochord-mediated
repulsion. Sympathetic axons that normally avoid the
midline in vivo are also repelled, in part, by a semaphorinbased notochord activity. Although our results implicate
semaphorin signalling in mediating repulsion by the
notochord, repulsion of early dorsal root ganglion axons is
only partially blocked when using neuropilin/Fc reagents.
Moreover, retinal axons, which are insensitive to SEMA3A,
are also repelled by the notochord. We conclude that
multiple factors act in concert to guide axons in this system,
and that further notochord repellents remain to be
identified.
INTRODUCTION
expressed on posterior half-sclerotome cells that prevent
growing axons from invading the posterior half-somite (Davies
et al., 1990). Although Ephrin-Bs and F-spondin have been
implicated in peripheral spinal nerve segmentation (Koblar et
al., 2000; Tzarfati-Majar et al., 2001; Wang and Anderson,
1997), an unidentified peanut agglutinin-binding molecule is
also likely to play an important role (Davies et al., 1990;
Vermeren et al., 2000).
The initial trajectory of DRG and motor axons within the
somites is further restricted within the dorsoventral plane,
because growing axons avoid the notochord lying medially to
the DRG and the dermamyotome which lies laterally (Fig. 1A).
Insights into the mechanisms that influence DRG axon
trajectory in the periphery have come from experiments
involving tissue explants cultured in three-dimensional
collagen gels (Keynes et al., 1997; Nakamoto and Shiga, 1998).
It was found that the notochord and dermamyotome, and to a
lesser extent ventral neural tube, secrete repellents acting on
DRG axons over considerable distances. It is probable,
therefore, that tissues lying either side of the DRG channel
axon trajectories by repulsion in vivo. It has also been found
that the notochord has axon-repulsive activity at the
appropriate developmental stages for early outgrowing DRG
axons (Keynes et al., 1997; Nakamoto and Shiga, 1998).
During development, growing axons navigate towards their
targets under the influence of multiple guidance cues present
in the embryonic environment. These cues, which are diffusible
or bound to plasma membrane and/or extracellular matrix,
promote or prevent axon growth by attractive or repulsive
mechanisms, respectively (Goodman, 1996; Mueller, 1999).
The overall balance of such cues determines, therefore, the path
that an axon follows. A simple anatomical system for assessing
the role of different axonal guidance cues concerns the
development of the vertebrate peripheral spinal nervous system
(Tannahill et al., 2000). In this system, the initial trajectory of
spinal axons in the periphery is largely imparted by repulsive
cues that surround outgrowing axons in three dimensions.
Insights into axon repulsive mechanisms have come from
considering the influences that establish the initial trajectory of
peripheral spinal nerves in different embryonic axes. For
example, along the anteroposterior axis, peripheral spinal nerve
segmentation is imparted by the neighbouring somites, which
act to limit sensory dorsal root ganglia (DRG) and motor axon
trajectories to the anterior half-sclerotome of each somite
(Keynes and Stern, 1984; Rickmann et al., 1985). This
segmental restriction is mediated by repulsive molecules
Key words: Axon guidance, Neural development, Chick embryo,
Notochord, Spinal nerves, Dorsal root ganglia, Semaphorin
1124 C. N. G. Anderson and others
Fig. 1. (A) Surround repulsion of peripheral DRG axons in the chick
embryo trunk. Schematic diagram through the trunk in transverse
section showing tissues that are sources of secreted repulsive signals
(red arrows). These guidance cues operate in concert to constrain
DRG axon paths in a typical bipolar trajectory (black arrows).
(B) Expression of SEMA3A in the chick embryo notochord.
Following whole-mount in situ hybridisation on stage 19 embryos,
transverse sections (30 µm) were cut and mounted. The section
shown is through the brachial region and SEMA3A expression is
seen in the notochord and dermamyotome. Note that the neural tube
split open during the hybridisation procedure. dm, dermamyotome;
drg, dorsal root ganglion; fp, floor plate of neural tube; nt, neural
tube; n, notochord; sc, sclerotome. Scale bar: ~100 µm.
Notochord removal experiments in ovo have lent further
support for such a role of the notochord, together with its
surrounding matrix, in axon repulsion (Stern et al., 1991;
Tosney and Oakley, 1990). In contrast, surgical removal of
dermamyotome has proved less informative as residual
ectoderm, which is also a source of repulsive cues, regenerates
in situ to maintain a source of repulsion (Keynes et al., 1997;
Tosney, 1987).
Although the molecules mediating axon repulsion by the
notochord and dermamyotome remain to be identified,
dermamyotome-mediated repulsion may be explained, in
part, by SEMA3A (collapsin 1), a secreted member of the
semaphorin family (Castellani and Rougon, 2002; He et al.,
2002; Raper, 2000). SEMA3A activity is mediated through
homodimers of neuropilin 1, forming part of the semaphorin
receptor complex (He and Tessier-Lavigne, 1997; Kolodkin et
al., 1997; Miyazaki et al., 1999b; Nakamura et al., 1998;
Takahashi et al., 1998). SEMA3A protein has the appropriate
repulsive activity on DRG and motor axons, as their growth
cones collapse in response to it (Luo et al., 1993; VarelaEchavarria et al., 1997). SEMA3A mRNA is also present at
high levels in the dermamyotome at the appropriate stages
(Adams et al., 1996; Shepherd et al., 1996; Wright et al., 1995).
Furthermore, in mice lacking Sema3a or neuropilin 1, some
axons are seen to project aberrantly towards the
dermamyotome (Kitsukawa et al., 1997; Taniguchi et al., 1997;
White and Behar, 2000). A role for SEMA3A in contributing
to repulsion from the notochord is less obvious. Although
SEMA3A is expressed in the notochord, the level of expression
is much lower than that in the dermamyotome (Shepherd et al.,
1996). Also, in Sema3a and neuropilin 1 knockout mice, no
defects consistent with semaphorin-mediated repulsion by the
notochord have been described in the initial trajectories of
DRG or motor axons (Kitsukawa et al., 1997; Taniguchi et al.,
1997; White and Behar, 2000).
Two other semaphorin family members are also expressed
at low levels in the chick embryo notochord, namely SEMA3C
(collapsin 3) and SEMA3D (collapsin 2) (Luo et al., 1995).
SEMA3C, whose activity is thought to be mediated through
heterodimers of neuropilin 1 and 2 (Chen et al., 1998; Raper,
2000; Renzi et al., 1999; Takahashi et al., 1998), has been
reported to induce the collapse of sympathetic chain ganglia
(SCG) but not DRG axons (Adams et al., 1997; Chen et al.,
1998; Koppel et al., 1997; Takahashi et al., 1998). Moreover,
SEMA3D does not induce collapse of either DRG or
sympathetic growth cones (Koppel et al., 1997; Raper, 2000).
It seems unlikely, therefore, that SEMA3C or SEMA3D have
a direct influence on DRG axon trajectory. In mice, expression
of the neuropilin 1-dependent semaphorin, Sema3e, has been
found in the perinotochordal mesenchyme (Miyazaki et al.,
1999a; Miyazaki et al., 1999b). Although the biological
activity of chick SEMA3E protein has yet to be reported,
mouse Sema3E protein appears to repel DRG axons (Miyazaki
et al., 1999a; Miyazaki et al., 1999b). As described above, the
phenotype of neuropilin 1 knockout mice, which has brought
into question the role of neuropilin 1-dependent semaphorins,
also raises questions concerning the role of other semaphorins,
such as SEMA3C, which are thought to act through neuropilin
1 together with neuropilin 2. The role of secreted semaphorins
acting through neuropilin 2, such as SEMA3F (Chen et al.,
1998; Raper, 2000; Renzi et al., 1999), also remains unclear.
Although defasciculation of peripheral spinal nerves has been
reported in neuropilin 2 knockout mice, defects in initial axon
trajectory consistent with neuropilin 2-dependent repulsion by
the notochord and dermamyotome have not been described
(Giger et al., 2000).
Recently, it has been shown that slit genes are expressed in
the notochord, raising the possibility that they could contribute
to repulsion of peripheral spinal axons. Consistent with this is
the observation that slit2 expression precedes, and continues
beyond, the initial outgrowth of DRG axons (Holmes and
Niswander, 2001; Li et al., 1999). Expression of slit3 has also
been reported in the notochord at similar stages (Holmes and
Niswander, 2001). Slit2 protein has been shown to repel many
different axon types, including olfactory bulb, hippocampal,
spinal motor column and retinal axons (Brose et al., 1999;
Erskine et al., 2000; Li et al., 1999; Nguyen Ba-Charvet et al.,
1999; Niclou et al., 2000; Ringstedt et al., 2000; Yuan et al.,
1999). DRG axons, however, are not repelled by Slit2 in
collagen gel assays, and DRG growth cones do not collapse in
response to Slit2 (Nguyen Ba-Charvet et al., 1999; Niclou et
al., 2000). Further evidence against a role for Slit2 in motor
axon repulsion has come from an examination of the repulsive
properties of the floor plate, which expresses high levels of
Slit2 (Holmes and Niswander, 2001; Holmes et al., 1998; Yuan
et al., 1999). When soluble robo/Fc receptor fusion proteins are
used to inhibit Slit2, motor axon repulsion by the floor plate is
Axon repulsion by the notochord 1125
unaffected (Patel et al., 2001). At first sight, this argues against
a role for Slit2 in peripheral spinal axon guidance, but recent
studies have suggested that a membrane-bound N-terminal
Slit2 fragment is not a preferred substratum for DRG axon
growth (Nguyen-Ba-Charvet et al., 2001). Thus, the role of Slit
proteins in peripheral spinal axon guidance remains unclear.
In order to investigate the molecular nature of peripheral
spinal axon repulsion by the notochord, we undertook a screen
for axon repellents secreted by the notochord and identified
SEMA3A as a candidate repellent. Furthermore, experiments
in which notochord-derived SEMA3A repulsion is inhibited
indicate that, although SEMA3A contributes to the notochord
activity, other unidentified factors are also critical in
determining spinal sensory axon trajectories in the periphery.
MATERIALS AND METHODS
Tissue dissection
Chick embryos were staged according to Hamburger and Hamilton
(Hamburger and Hamilton, 1992). Stage 26-27 DRGs, stage 36 DRGs
and stage 35-36 SCG were dissected between thoracic and upper
lumbar regions. Olfactory bulb and retinal explants were dissected
from stage 36 and stage 26-29 embryos, respectively. Notochords
were dissected from stage 17-21 embryos.
Collagen gel cultures
Tissue explants were co-cultured in collagen gels essentially as
described previously (Keynes et al., 1997; Ohta et al., 1999). Tissues
were cultured in 15 mm four-well dishes (Nunc) and rat-tail type VII
collagen was obtained from Sigma or Roche. Unless otherwise stated,
DRG and SCG explants were positioned 250-500 µm from notochord
or COS cell explants, whereas retinal and olfactory bulb explants were
positioned 200-400 µm away. DRG, SCG and olfactory bulb cultures
were incubated for 24-30 hours and retinal explants for 48-72 hours.
All tissues were cultured in DMEM (Sigma) containing gentamycin
(50 µg/ml, Sigma) with either 50 ng/ml NGF (Alomone) or 50 ng/ml
NT3 (a gift from Regeneron Pharmaceuticals) for DRG and SCG
cultures; ITS (Sigma) for olfactory bulb cultures, or 3% chick embryo
extract (Ohta et al., 1999) for retinal explants. Neuropilin 1/Fc and
neuropilin 2/Fc fusion proteins were obtained from R&D Systems.
After incubation, cultures were fixed in 4% paraformaldehyde, 15%
sucrose, phosphate-buffered saline pH 7.4.
Evaluation of axon growth
Axon outgrowth in the quadrant proximal to the notochord or COS
cell co-explant was compared with the distal quadrant and scored as
described previously (Keynes et al., 1997). A score of zero represents
no outgrowth towards the co-explant; two, a weak and/or patchy
outgrowth; four, more consistent outgrowth which does not extend
closer to the co-explant than ~50 µm; six, growth approaches the coexplant closer than ~50 µm, but does not contact it; eight, axons
contact the co-explant, but some asymmetry between proximal and
distal remains; ten, no difference in outgrowth in the proximal versus
distal quadrants, with axons growing into and/or past the co-explant.
At least three independent experiments were performed for each
assay. Photography was performed using an Axiovert microscope with
dark-field or phase-contrast optics (Zeiss). Statistics were performed
using a Kruskal-Wallis non-parametric test followed by a
Bonferroni/Dunn post-hoc ANOVA test at the 1% significance level.
COS cell transfection
COS-7 cells were transfected with a pCAGGS (Niwa et al., 1991)
vector containing SEMA3A, SEMA3C or vector control using
LipofectAMINE (Invitrogen) as described previously (Ohta et al.,
1999). Full-length chick SEMA3C was cloned from a chick notochord
cDNA expression library based on published sequences (accession
number AF022946). COS cell aggregates were made by the hanging
drop method (Kennedy et al., 1994). SEMA3A-expressing COS cells
were mixed with vector-transfected cells to moderate the amount of
SEMA3A secreted by the aggregate. Aggregates were sub-divided
before placing in collagen gels. To control for SEMA3A expression
levels, the same COS cell aggregate was used in comparing the effects
of neuropilin 1/Fc with controls.
Whole-mount in situ hybridisation
Whole-mount in situ hybridisation was performed as described
previously with minor variations (Nieto et al., 1996). Probes were
generated by PCR from a notochord cDNA library and sub-cloned
into pCR2.1 (Invitrogen). Hybridisation was performed at 65˚C. Postin situ hybridisation, embryos were embedded in TissueTek OCT
compound (Sakura) and 30 µm sections cut using a cryostat. The
following chicken sequences were used: SEMA3A, nucleotides 22212869 (Accession number, GGU02528); SEMA3C, nucleotides 17862197 (Accession number, AF022946); and neuropilin 2 nucleotides
1170-1785 (Accession number, AF417236).
RESULTS
SEMA3A as a candidate for mediating axonal
repulsion by the notochord
In order to identify candidate axon repellents secreted by the
notochord, we screened supernatants from pools of a notochord
cDNA library expressed in COS cells for growth cone collapseinducing activity. After screening ~1000 pools containing 5001000 clones each, we identified one positive pool that was
subsequently found to contain SEMA3A. Consistent with
previous studies (Shepherd et al., 1996), SEMA3A expression
in the notochord was confirmed by in situ hybridisation (Fig.
1B). Expression of other class 3 semaphorins has also been
reported at low levels in the notochord (Luo et al., 1995). By
PCR, we confirmed the presence of SEMA3C, SEMA3D and,
to a lesser extent, SEMA3E in the notochord library (data not
shown). Given the low levels of SEMA3E, and that SEMA3C
and SEMA3D are not known to repel DRG axons (Adams et al.,
1997; Chen et al., 1998; Koppel et al., 1997; Takahashi et al.,
1998), SEMA3A represents the best candidate for a notochordderived semaphorin mediating peripheral spinal axon guidance.
To explore the potential contribution of SEMA3A to
notochord-mediated repulsion, we exploited the sensitivity of
different DRG axon populations, selected by growth medium
containing different neurotrophins (Farinas et al., 1998;
Martin-Zanca et al., 1990; Phillips and Armanini, 1996), to
repulsion by SEMA3A. Older DRG axons grown in NGF, but
not NT3, are repelled by SEMA3A in vitro (Messersmith et
al., 1995; Pond et al., 2002; Puschel et al., 1996; Shepherd et
al., 1997). This is thought to limit the extension of NGF-, but
not NT3-, responsive axons in vivo to the dorsal spinal cord,
as they are repelled by SEMA3A expressed in the ventral
spinal cord (Messersmith et al., 1995; Pond et al., 2002;
Puschel et al., 1996; Shepherd et al., 1997). The differential
SEMA3A sensitivity also correlates with the expression of
neuropilin 1, which is required for SEMA3A-mediated
repulsion. NGF-responsive axons maintain neuropilin 1
expression, whereas NT3-responsive axons lose neuropilin 1
expression around the time that they project into the spinal cord
(Fu et al., 2000; Pond et al., 2002).
1126 C. N. G. Anderson and others
We therefore employed collagen gel co-cultures to assay
repulsion by the notochord of different populations of stage 36
(day 10) DRG axons. Our results indicate that stage 36 DRG
axons cultured in the presence of NGF, but not NT3, are
repelled by stage 17-21 (day 3) notochords (Fig. 2A,B,F; mean
score ± standard error of the mean is 1.2±0.1 versus 7.9±0.2).
Although this is consistent with the action of SEMA3A, it does
not exclude the action of other semaphorins acting through
neuropilin 1, or of other unidentified repellents. In contrast to
the differential responsiveness of stage 36 DRG axons to
SEMA3A when cultured in NGF versus NT3, it has been noted
that earlier DRG axons are responsive to SEMA3A when
cultured in either neurotrophin (Pond et al., 2002; Puschel et
al., 1996; Shepherd et al., 1997). We therefore examined
repulsion by the notochord of earlier DRG axons, grown from
stages closer to the times of initial axon outgrowth in vivo.
Stage 26-27 (day 5) DRGs were used, being the earliest stages
at which DRGs can be isolated and cultured reproducibly
(Masuda et al., 2000). Similar to the response of stage 29-31
DRG axons to SEMA3A (Pond et al., 2002; Shepherd et al.,
1997), earlier DRG axons (stage 26-27) cultured in either
neurotrophin are strongly repelled by the notochord, consistent
with a role for SEMA3A (Fig. 2C,D,F; mean scores are 0.8±0.2
and 1.6±0.2 for NGF and NT3 cultures, respectively).
Similar to peripheral DRG axons, axons from SCGs, lying
ventral to the DRGs, avoid the notochord and do not cross into
the contralateral side. SCG axons are known to be sensitive to
repulsion by several semaphorins, including SEMA3A,
SEMA3C and SEMA3F (Adams et al., 1997; Chen et al., 1998;
Koppel et al., 1997; Takahashi et al., 1998). We therefore tested
whether SCG axons are sensitive to notochord repulsion. Stage
36 SCG axons were strongly repelled by the notochord (mean
score 0.3±0.1), and were significantly more sensitive
to notochord repulsion than similar stage DRG axons
cultured in NGF (Fig. 2E,F; P<0.001). To examine
whether the differential sensitivity of SCG and DRG
axons to notochord repulsion could be accounted for
by differences in SEMA3A responsiveness, we
compared the repulsion of SCG and DRG axons to
SEMA3A-expressing cells within the same collagen
gel. Qualitatively, SCG and DRG axons of both stages
showed a similar extent of repulsion to SEMA3A
secreted from COS cell aggregates (Fig. 3). The
finding that SCG axons are more repelled by the
notochord than stage 36 DRG axons (Fig. 2F) is
consistent with the secretion of a repellent additional
to SEMA3A (see below). Taken together, these results
implicate a neuropilin 1-binding semaphorin,
probably SEMA3A, as a candidate molecule that
contributes to notochord-derived axon repulsion.
Fig. 2. Growth of different axon populations in response to the notochord in
collagen gel co-cultures. (A-E) Dark-field photomicrographs of representative
co-cultures. (A,B) Stage 36 (St36) DRG explants (n=90 and 36, respectively);
(C,D) Stage 26-27 (St26) DRG explants (n=55 and 34, respectively); (E) Stage
36 SCG explants (n=28). Explants were cultured with NGF (A,C,E) or with
NT3 (B,D). (F) Histogram showing the mean repulsion score±s.e.m. for neural
explants co-cultured with stage 17-21 notochord. Asterisk indicates no
significant repulsion. Scale bar: ~100 µm.
Fig. 3. Growth of different axon populations in response to
SEMA3A-expressing COS cells in collagen gel cocultures. (A) Left, stage 36 (St36) DRG; right, stage 26-27
(St26) DRG; (B) Left, stage 36 DRG; right, stage 36 SCG.
Scale bar: ~100 µm.
Axon repulsion by the notochord 1127
Inhibition of notochord repulsion by
soluble neuropilins
To test whether neuropilin 1-binding
semaphorins, including SEMA3A, are secreted
by the notochord, we employed soluble
neuropilin receptor proteins, comprising the
extracellular domains of rat neuropilin 1 or
neuropilin 2 fused to the human IgG
complement-binding fragment (Fc), to titrate out
semaphorin activity in collagen gel assays. To
control for the effectiveness of the neuropilin
1/Fc receptor reagent, we first used neuropilin
1/Fc to inhibit SEMA3A repulsion derived from
COS cells expressing SEMA3A. In these
experiments we used SEMA3A-expressing cell
aggregates that matched the extent of repulsion
obtained with notochord explants. Using vectortransfected cells, with or without neuropilin 1/Fc,
no significant change in the growth of stage 36
DRG axons was observed (Fig. 4A,E). However,
in co-cultures of stage 36 DRGs and SEMA3Aexpressing cells, axon repulsion was significantly
reduced in the presence of the neuropilin 1/Fc
(P<0.0001), and many more axons approached
and occasionally contacted the cell aggregate
(from a mean score of 1.4±0.2 to 5.0±0.8;
compare Fig. 4B,C,E). As a control, the addition
of 10 µg/ml neuropilin 2/Fc did not significantly
reduce repulsion (Fig. 4D,E), consistent with the
finding that SEMA3A binds neuropilin 2 only
with low affinity (Chen et al., 1997).
When stage 36 DRGs were cultured in NGF
with stage 17-21 notochords in the presence of
neuropilin 1/Fc, repulsion of DRG axons was
Fig. 4. Effects of soluble neuropilin/Fc reagent in collagen gel co-culture
significantly reduced in a dose-dependent
experiments. (A-D) Stage 36 DRG explants cultured with COS cell aggregates
manner (P<0.0001). In the presence of neuropilin
transfected with vector (VEC) alone (A) (n=25) or SEMA3A (SEMA3A) (B-D).
1/Fc, the mean score increased from 1.6±0.2 in
(B) Repulsion by SEMA3A-expressing cells in the absence of neuropilin/Fc (n=14).
control co-cultures to 3.2±0.2 and 5.7±0.4 for 3
(C) Neuropilin 1/Fc at 10 µg/ml [NP1(10)] reduces repulsion by SEMA3A (n=13);
µg/ml and 10 µg/ml neuropilin 1/Fc, respectively
whereas (D) neuropilin 2/Fc at 10 µg/ml [NP2(10)] has little or no effect on
(Fig. 4F,G,J). In these experiments, the reduction
SEMA3A-induced repulsion (n=8). (E) Histogram showing the mean repulsion
of repulsion was equivalent to that seen using
score±s.e.m. Neuropilin 1 or 2/Fc does not affect axon outgrowth in co-cultures of
neuropilin 1/Fc to block SEMA3-mediated
notochord and vector-expressing cells (n=12 and 7, respectively). (F,G) Stage 36
repulsion (see above). The observation that
DRG explants co-cultured with stage 17-21 notochords in the presence of (F) 3
µg/ml neuropilin 1 [NP1(3)] or (G) 10 µg/ml neuropilin 1/Fc [NP1(10)]; (n=12 for
neuropilin 2/Fc does not cause a similar
both). (H,I) Stage 36 DRG explants co-cultured with stage 17-21 notochords in the
reduction in repulsion (Fig. 4H-J; mean score
presence of (H) 3 µg/ml neuropilin 2 [NP2(3)] or (I) 10 µg/ml neuropilin 2/Fc
1.6±0.2 and 2.1±0.2 for 3 µg/ml and 10 µg/ml
[NP2(10)]; (n=12 and 16, respectively). (J) Histogram showing the mean repulsion
neuropilin 2/Fc, respectively) indicates that the
score±s.e.m. (n=25 for control cultures in the absence of neuropilin/Fc). Asterisks
Fc domain is not responsible for reducing
indicate no significant repulsion, or a significant reduction in repulsion, compared to
notochord-induced repulsion, and also suggests
positive controls (SEMA). Scale bar: ~100 µm.
that neuropilin 2-binding semaphorins are
unlikely to be involved. Taken together, these
to 2.3±0.3, and from 0.2±0.1 to 2.2±0.2, respectively (Fig.
results are in keeping with the proposition that the repulsive
5A,D,G). At 10 µg/ml neuropilin 2, an apparent reduction in
activity secreted by the notochord has a major component
repulsion was observed for DRG and SCG axons, but this was
based on SEMA3A.
not statistically significant at the 1% level (Fig. 5B,E,G; the
We next tested whether the neuropilin/Fc receptor reagents
mean score increased from 0.6±0.2 to 1.0±0.3, and from
could reduce the repulsion by the notochord of earlier (stage
0.2±0.1 to 1.2±0.4, respectively). When combinations of
26-27) DRG axons as well as SCG axons. Overall, in these
neuropilin 1/Fc and neuropilin 2/Fc were used, the reduction
experiments a reduction in repulsion was evident, but this was
in repulsion was not significantly different from that for
significantly less than for stage 36 DRG axons (P<0.0001) and
neuropilin 1/Fc alone (Fig. 5C,F,G; at 5 µg/ml of each
was more variable (Fig. 5). At 10 µg/ml neuropilin 1, the mean
neuropilin/Fc the mean score increased from 0.6±0.2 to
score for DRG and SCG axon repulsion increased from 0.6±0.2
1128 C. N. G. Anderson and others
Fig. 5. Effects of soluble neuropilin/Fc reagents on repulsion by the
notochord of stage 26-27 DRG and stage 36 SCG axons in collagen
gel co-cultures. (A-C) Stage 26-27 DRG explants cultured with (A)
10 µg/ml neuropilin 1/Fc (NP1), (B) 10 µg/ml neuropilin 2/Fc (NP2)
and (C) 5 µg/ml each of neuropilin 1 and 2/Fc (NP1/2); (n=27, 20
and 11, respectively). (D-F) As for A-C, but with stage 36 SCG
explants (n=24, 16 and 10, respectively). (G) Histogram showing the
mean repulsion score±s.e.m. (n=24 and 27 for DRG and SCG
explants, respectively, in the absence of neuropilin/Fc reagents. The
neuropilin/Fc reagent dose in µg/ml is indicated in brackets).
Asterisks indicate a significant reduction in repulsion compared to
controls. Scale bar: ~100 µm.
2.4±0.5, and from 0.2±0.1 to 2.0±0.3, for DRG and SCG
axons, respectively; at 10 µg/ml of each neuropilin/Fc the
respective values were 2.7±0.5 and 1.7±0.3).
Because neuropilin 1/Fc receptor reagents were
significantly less effective in inhibiting notochord repulsion
of early DRG and SCG axons compared with later DRG
axons, even though they appear equally sensitive to SEMA3A
(see above), it is probable that multiple repellents are secreted
by the notochord. Although much of the repulsive activity on
early DRG axons appears not to be mediated by neuropilinbinding proteins, our results support a role for SEMA3A
nonetheless. Neuropilin 1-binding molecules are secreted by
the notochord, as there is a significant reduction in repulsion
of DRG and SCG axons with neuropilin 1/Fc. A role for
neuropilin 2 is not revealed by these experiments, but the
expression of neuropilin 2 in mouse (Chen et al., 1997) and
chick DRGs (Fig. 5) could indicate a potential contribution
of neuropilin 2.
Expression of SEMA3C and neuropilin 2
in the chick embryo trunk
SCG neurons express both neuropilin 1 and
neuropilin 2, and respond to semaphorins that
bind either receptor (Chen et al., 1998; Giger et
al., 1998). The finding that SCG axons are more
repelled by the notochord than stage 36 DRG
axons raises the possibility that SCG axons also
respond to a neuropilin 2-binding semaphorin,
such as SEMA3C. We therefore confirmed the
expression of SEMA3C in the notochord by
wholemount in situ hybridisation (Fig. 6A,B), in
keeping with previous data (Luo et al., 1995).
SEMA3C expression was also noted in the trunk
at sites similar to SEMA3A, such as the posterior
half-sclerotome of more anterior somites (Fig.
6B), consistent with expression of the mouse
homologue (Puschel et al., 1995).
Although it has been reported that DRG axons are
insensitive to SEMA3C (Adams et al., 1997; Chen et al., 1998;
Koppel et al., 1997; Takahashi et al., 1998), very early DRG
axons have not been tested. If these axons are SEMA3Csensitive, they should express neuropilin 2. We examined the
expression of neuropilin 2 in the trunk and found strong
expression in the anterior but not posterior half-sclerotome of
the earliest forming somites, with expression being maintained
in more anterior somites in regions where neural crest cells
coalesce to form DRGs (Fig. 6C-E). By stage 26, neuropilin 2
expression had decreased in the DRGs, but was detectable in
both DRGs and their axons (Fig. 6F). The spinal motor
columns of the ventral neural tube were also seen to express
neuropilin 2 (Fig. 6D). This expression pattern is reminiscent
of mouse neuropilin 2 (Chen et al., 1997).
These results are consistent with a possible role for a
neuropilin 2-binding semaphorin in notochord-mediated
repulsion, but the inability of neuropilin 2/Fc to prevent
repulsion (Fig. 5G) argues against this. As a further test we
examined whether DRG axons taken from early stages are able
to respond to a source of SEMA3C (Fig. 7). As a control, we
found that SCG axons were strongly repelled by cells
expressing SEMA3C (Fig. 7B,E; mean score 1.7±0.3);
however, DRG axons taken from stage 26-27 were not repelled
(Fig. 7D,E; mean score 8.8±0.2). Therefore, although the
notochord may secrete SEMA3C, it is probable that SEMA3C
does not account for the repulsion of these DRG axons, as they
do not respond to SEMA3C and express neuropilin 2 only
weakly.
The notochord repels retinal axons
Several other axon populations have well-characterised
responses to known axon repellents. These can therefore be
employed in assays of notochord repulsion to explore the
nature of the repellents secreted by the notochord. For example,
retinal axons are known to be insensitive to SEMA3A (Luo et
al., 1993; Vermeren et al., 2000). When stage 26-29 retinal
explants were cultured together with notochord, retinal axons
were repelled (Fig. 8; mean score 2.4±0.3). Furthermore, when
retinal and notochord explants were cultured together in the
presence of soluble neuropilins, there was no significant
reduction in repulsion (Fig. 8; mean scores 2.1±0.5 and 0.9±0.3
in the presence of 10 µg/ml neuropilin 1 or neuropilin 2,
Axon repulsion by the notochord 1129
respectively), again suggesting that the notochord secretes
axon repellents that act independently of neuropilins.
slit2 and slit3 are expressed by the notochord in vivo
(Holmes and Niswander, 2001; Li et al., 1999), and Slit2 can
act as a repellent for retinal axons (Erskine et al., 2000; Niclou
et al., 2000; Ringstedt et al., 2000), raising the possibility that
Slit2 is secreted by the notochord to repel retinal axons. We
examined this possibility by exploiting the observation that
olfactory bulb axons are repelled by Slit2 (Li et al., 1999;
Nguyen Ba-Charvet et al., 1999; Yuan et al., 1999). When stage
36 olfactory bulb explants were cultured together with
notochord, we found that axons were not repelled (Fig. 8; mean
score 7.4±0.4), suggesting that Slit2 does not mediate
notochord repulsion of retinal axons. As olfactory bulb axons
can be repelled by SEMA3F and not SEMA3A (de Castro et
al., 1999), this further suggests that neuropilin 2-binding
semaphorins do not contribute to the repulsive activity of the
notochord. Overall, our results strongly suggest that
unidentified axon repellents, additional to semaphorins,
mediate peripheral spinal axon repulsion by the notochord.
DISCUSSION
Fig. 6. Expression of SEMA3C and neuropilin 2 in the developing
chick embryo trunk. (A,C) Lateral views of wholemount in situ
hybridisation for SEMA3C at stage 18 (A) and for neuropilin 2 at
stage 20 (C). Somites anterior to the wing bud are shown (anterior is
up, dorsal is right). SEMA3C expression in (A) is seen in the
notochord (white arrows) and posterior halves of consecutive
sclerotomes (white arrowheads). Neuropilin 2 expression in (C) is
seen in the DRGs within the anterior half-sclerotomes (white
arrowheads); black arrowheads indicate somite boundaries. Scale
bars: ~100 µm. (B,D) Transverse sections (30 µm) through the trunk
of stage 19 embryos after whole-mount in situ hybridisation.
(B) SEMA3C expression at the level of a posterior half-sclerotome.
Expression is seen in the notochord, posterior half-sclerotome and
ventral half of the neural tube. (D) Neuropilin 2 expression at the
level of the anterior half-sclerotome. Expression is seen in the
anterior half sclerotome in the region of DRG formation, and in the
ventral neural tube. Scale bar: ~100 µm. (E) Lateral view of wholemount in situ hybridisation for neuropilin 2 at stage 24. Expression is
seen in the anterior half-sclerotomes of recently formed somites
(arrow) and in the DRGs forming more anteriorly (arrowheads).
Scale bar: ~1 mm. (F) Whole-mount in situ hybridisation for
neuropilin 2 on a stage 26 embryo trunk. After in situ hybridisation,
the trunk was hemisected and viewed from the neural tube side to
help visualise expression. Weak neuropilin 2 expression is seen in the
DRGs (arrows) and axons leaving the DRGs (arrowheads). Scale bar:
~100 µm. as, anterior half-sclerotome; drg, dorsal root ganglion; nt,
neural tube; n, notochord; ps, posterior half-sclerotome.
SEMA3A as a notochord-derived axon repellent
Using an expression cloning strategy to isolate axon repellents
secreted by the notochord, we have identified SEMA3A as a
likely candidate. First, it is expressed by the notochord at the
appropriate stages. Second, DRG and motor axon populations
express the SEMA3A receptor component, neuropilin 1 (Chen
et al., 1997), and are repelled by SEMA3A (Luo et al., 1993;
Messersmith et al., 1995; Varela-Echavarria et al., 1997;
Vermeren et al., 2000) and by the notochord (Keynes et al.,
1997; Nakamoto and Shiga, 1998). Our results using stage 36
DRG axons are also in keeping with the maintenance of
neuropilin 1 expression in NGF-selected axon populations (Fu
et al., 2000; Pond et al., 2002). Finally, neuropilin 1/Fc but not
neuropilin 2/Fc treatment reduces repulsion by the notochord.
This reduction was statistically significant, and the addition of
higher concentrations of neuropilin 1/Fc did not reduce
repulsion further (data not shown). Using stage 36 DRGs, the
repulsion score with SEMA3A (or notochord) was reduced to
levels equivalent to that of a weak repulsive source.
Several factors may explain why blocking repulsion from
SEMA3A-expressing cells did not reduce the score to that of
control cells. There was individual variability in the scores;
some cultures were completely disinhibited whereas others
were only partially so. This may reflect the difficulty in
standardising the amount of SEMA3A being secreted by the
cell aggregates. Also, the semi-quantitative scoring system that
we use is particularly sensitive to low levels of repulsion and
scores of five and above (which represents half the numerical
scale) are in the realms of weak to no repulsion. Finally, as the
neuropilin receptor reagents are from rat sequences, they may
have less efficacy. Despite these concerns, we conclude that
neuropilin 1/Fc can be used as an effective tool in collagen gel
assays to detect the presence of repulsive molecules such as
SEMA3A.
Taken together, our results strongly suggest that SEMA3A
is secreted by the notochord and repels DRG axons. However,
although it is probable that a major repellent for stage 36 axons
1130 C. N. G. Anderson and others
Fig. 7. SEMA3C repels SCG but
not early DRG axons in collagen
gel co-cultures. (A,B) Stage 35
SCG explants cultured with COS
cell aggregates transfected with
(A) vector (VEC) alone (n=31) or
(B) SEMA3C (SEMA3C) (n=24).
(C,D) Stage 26-27 DRG explants
cultured with COS cell aggregates
transfected with (C) vector (VEC)
alone (n=22) or (D) SEMA3C (SEMA3C) (n=18). (E) Histogram showing the
mean repulsion score±s.e.m. Scores were assessed in cultures in which the
explants lay between 150-450 µm of the COS cell aggregate. Asterisks
indicate no significant repulsion. Scale bar: ~100 µm.
is SEMA3A, our results indicate that other repellents
contribute to the activity of the notochord on stage 26-27 DRG
and stage 36 SCG axons. In particular, we noted that neuropilin
1/Fc treatment reduces notochord repulsion of stage 26-27
DRG axons to a much lesser extent than that of stage 36 DRG
axons.
Other semaphorins as notochord-derived axon
repellents
Although our results are consistent with SEMA3A being
secreted by the notochord, they do not exclude the possibility
that other semaphorins act in concert, because in our inhibition
experiments neuropilin 1/Fc could bind to semaphorins other
than SEMA3A. Several other semaphorins, including
SEMA3C and SEMA3D have been shown to be expressed by
the notochord in chick embryos (Luo et al., 1995; Shepherd et
al., 1996), and it remains to be determined whether any further
semaphorins are also expressed here. It is unlikely that
SEMA3D directly repels DRG axons because it does not
induce collapse of their growth cones (Koppel et al., 1997).
SEMA3E is also expressed at low levels in the perinotochordal
region, and it may contribute to repulsion by the notochord as
it has the appropriate repulsive activity for DRG axons
(Miyazaki et al., 1999a; Miyazaki et al., 1999b).
Our results raise the possibility of a role for SEMA3C,
because it is expressed at moderate levels by the notochord.
SEMA3C is thought to operate through a combination of
Fig. 8. Axon repulsion of other neuronal populations by the
notochord. (A) Axons from stage 26-29 retinal explants are repelled
by the notochord (n=52), but axons from stage 36 olfactory bulb
explants are not (B) (n=20). (A,B) Co-cultures viewed by phasecontrast microscopy. Scale bar: ~100 µm. (C) Histogram showing
the mean repulsion score±s.e.m. Soluble neuropilins do not reduce
the repulsion of retinal axons by the notochord; n=11 for both
neuropilin 1/Fc (NP1) and neuropilin 2/Fc (NP2), respectively.
Asterisk indicates no significant repulsion.
neuropilin 1 and 2 (Chen et al., 1998). Although we find that
DRGs express neuropilin 2, our inhibition experiments suggest
that signalling through this receptor is not functionally
significant. This is also supported by our direct tests of early
DRG axon repulsion by SEMA3C, in which we find no
evidence to support a role for this semaphorin. As we find
higher expression of neuropilin 2 in stage 17-20 DRGs, it
remains possible that SEMA3C plays a role in the earliest
stages of axon outgrowth.
SEMA3C, in addition to SEMA3A, may also contribute to
repulsion of SCG axons by the notochord, as it has the
appropriate activity on SCG axons and is expressed in the
notochord. We also found a diminution of repulsion when
notochord-SCG co-cultures were treated with neuropilin 2/Fc
Axon repulsion by the notochord 1131
(repulsion score increased from 0.2±0.1 to 1.2±0.4), although
this was not statistically significant at the 1% level (P<0.012).
This may reflect the difficulty in scoring small differences in
repulsion in cultures showing strong repulsion.
It has also been reported that SEMA3F, a neuropilin 2binding semaphorin, is expressed in the perinotochordal region
(Eckhardt and Meyerhans, 1998). However, olfactory bulb
axons are not repelled by the notochord but are repelled by
SEMA3F (de Castro et al., 1999), further suggesting that
neuropilin 2 does not mediate notochord repulsion. Again, this
is consistent with the absence of defects in the early guidance
of peripheral spinal nerves in mice lacking neuropilin 2 (Giger
et al., 2000). The role of SEMA3F in vivo remains somewhat
unclear as it has been recently reported that mouse DRG axons
are repelled by Sema3F in vitro (Dionne et al., 2002). It is also
of particular note that both SEMA3A and SEMA3C are
expressed in posterior half-somites (Eickholt et al., 1999;
Shepherd et al., 1996), a well-characterised repulsive barrier to
peripheral spinal nerves (Tannahill et al., 2000). As it is thought
that peripheral nerve segmentation is based on contact
repellents, rather than secreted repellents, any contribution of
semaphorins to this repulsive mechanism will probably be
short- rather than long-range.
The role of other molecules in axon repulsion by the
notochord
The apparently normal pathfinding of peripheral spinal axons
with respect to the notochord in mice lacking Sema3a,
neuropilin 1 or neuropilin 2 would also suggest that factors
other than semaphorins mediate axon repulsion by the
notochord. Consistent with this, our results show that neuropilin
1/Fc is unable to inhibit a large proportion of the repulsion of
stage 26-27 DRG axons by the notochord. In these experiments,
it is probable that the neuropilin 1/Fc reagent is not saturated
as it can almost completely inhibit repulsion by the notochord
of stage 36 DRG axons. Furthermore, retinal axons, which do
not express neuropilins (Takagi et al., 1995; Takahashi et al.,
1998), are still repelled by the notochord. This latter result
might be explained by secretion of molecules from the
notochord that do not normally function as DRG axon
repellents. For example, although we have previously shown
that sonic hedgehog, which is strongly expressed by the
notochord, is unable to repel DRG axons in collagen gel
experiments (Keynes et al., 1997), in some situations it can
interfere with the growth of retinal axons (Trousse et al., 2001).
Similarly, Slit proteins can repel retinal but not DRG axons
(Erskine et al., 2000; Niclou et al., 2000; Ringstedt et al., 2000),
and they are also expressed by the notochord (Holmes and
Niswander, 2001; Li et al., 1999). Our finding that retinal but
not olfactory bulb axons are repelled by the notochord suggests,
however, that notochord repulsion is not mediated by Slit2.
Clearly, the simplicity of axon repulsion by the notochord at
the anatomical level is underlain by a considerable complexity
at the molecular level. Although our results implicate
SEMA3A, the precise identity and relative contributions of the
relevant molecules remain to be elucidated. If many different
molecules are involved, the use of multiple knockouts may be
required to assess the individual contribution of each molecule
in the absence of others. In addition to long-range axon
repellents described in this paper, a full characterisation of
midline repulsion by the notochord will also need to consider
molecules that act at short-range, in the immediate vicinity of
the notochord. For example, the perinotochordal matrix is rich
in extracellular matrix components, such as chondroitin
sulphate proteoglycans (Landolt et al., 1995; Newgreen et al.,
1986; Pettway et al., 1990; Ring et al., 1996), which are known
to inhibit axon growth in other systems (Anderson et al., 1998;
Becker and Becker, 2002; Chung et al., 2000; Dou and Levine,
1994; Fichard et al., 1991; Silver, 1994).
We thank the Wellcome Trust and MRC for support. D. T. is a Royal
Society University Research Fellow.
REFERENCES
Adams, R. H., Betz, H. and Puschel, A. W. (1996). A novel class of murine
semaphorins with homology to thrombospondin is differentially expressed
during early embryogenesis. Mech. Dev. 57, 33-45.
Adams, R. H., Lohrum, M., Klostermann, A., Betz, H. and Puschel, A. W.
(1997). The chemorepulsive activity of secreted semaphorins is regulated by
furin-dependent proteolytic processing. EMBO J. 16, 6077-6086.
Anderson, R. B., Walz, A., Holt, C. E. and Key, B. (1998). Chondroitin
sulfates modulate axon guidance in embryonic Xenopus brain. Dev. Biol.
202, 235-243.
Becker, C. G. and Becker, T. (2002). Repellent guidance of regenerating optic
axons by chondroitin sulfate glycosaminoglycans in zebrafish. J. Neurosci.
22, 842-853.
Brose, K., Bland, K. S., Wang, K. H., Arnott, D., Henzel, W., Goodman,
C. S., Tessier-Lavigne, M. and Kidd, T. (1999). Slit proteins bind Robo
receptors and have an evolutionarily conserved role in repulsive axon
guidance. Cell 96, 795-806.
Castellani, V. and Rougon, G. (2002). Control of semaphorin signaling. Curr.
Opin. Neurobiol. 12, 532-541.
Chen, H., Chedotal, A., He, Z., Goodman, C. S. and Tessier-Lavigne, M.
(1997). Neuropilin-2, a novel member of the neuropilin family, is a high
affinity receptor for the semaphorins Sema E and Sema IV but not Sema III.
Neuron 19, 547-559.
Chen, H., He, Z., Bagri, A. and Tessier-Lavigne, M. (1998). Semaphorinneuropilin interactions underlying sympathetic axon responses to class III
semaphorins. Neuron 21, 1283-1290.
Chung, K. Y., Taylor, J. S., Shum, D. K. and Chan, S. O. (2000). Axon
routing at the optic chiasm after enzymatic removal of chondroitin sulfate
in mouse embryos. Development 127, 2673-2683.
Davies, J. A., Cook, G. M., Stern, C. D. and Keynes, R. J. (1990). Isolation
from chick somites of a glycoprotein fraction that causes collapse of dorsal
root ganglion growth cones. Neuron 2, 11-20.
de Castro, F., Hu, L., Drabkin, H., Sotelo, C. and Chedotal, A. (1999).
Chemoattraction and chemorepulsion of olfactory bulb axons by different
secreted semaphorins. J. Neurosci. 19, 4428-4436.
Dionne, M. S., Brunet, L. J., Eimon, P. M. and Harland, R. M. (2002).
Noggin is required for correct guidance of dorsal root ganglion axons. Dev.
Biol. 251, 283-293.
Dou, C. L. and Levine, J. M. (1994). Inhibition of neurite growth by the NG2
chondroitin sulfate proteoglycan. J. Neurosci. 14, 7616-7628.
Eckhardt, F. and Meyerhans, A. (1998). Cloning and expression pattern of
a murine semaphorin homologous to H-sema IV. NeuroReport 9, 3975-3979.
Eickholt, B. J., Mackenzie, S. L., Graham, A., Walsh, F. S. and Doherty,
P. (1999). Evidence for collapsin-1 functioning in the control of neural crest
migration in both trunk and hindbrain regions. Development 126, 21812189.
Erskine, L., Williams, S. E., Brose, K., Kidd, T., Rachel, R. A., Goodman,
C. S., Tessier-Lavigne, M. and Mason, C. A. (2000). Retinal ganglion cell
axon guidance in the mouse optic chiasm: expression and function of robos
and slits. J. Neurosci. 20, 4975-4982.
Farinas, I., Wilkinson, G. A., Backus, C., Reichardt, L. F. and
Patapoutian, A. (1998). Characterization of neurotrophin and Trk receptor
functions in developing sensory ganglia: direct NT-3 activation of TrkB
neurons in vivo. Neuron 21, 325-334.
Fichard, A., Verna, J. M., Olivares, J. and Saxod, R. (1991). Involvement
of a chondroitin sulfate proteoglycan in the avoidance of chick epidermis
by dorsal root ganglia fibers: a study using beta-D-xyloside. Dev. Biol. 148,
1-9.
1132 C. N. G. Anderson and others
Fu, S. Y., Sharma, K., Luo, Y., Raper, J. A. and Frank, E. (2000). SEMA3A
regulates developing sensory projections in the chicken spinal cord. J.
Neurobiol. 45, 227-236.
Giger, R. J., Cloutier, J. F., Sahay, A., Prinjha, R. K., Levengood, D. V.,
Moore, S. E., Pickering, S., Simmons, D., Rastan, S., Walsh, F. S. et al.
(2000). Neuropilin-2 is required in vivo for selective axon guidance
responses to secreted semaphorins. Neuron 25, 29-41.
Giger, R. J., Urquhart, E. R., Gillespie, S. K., Levengood, D. V., Ginty, D.
D. and Kolodkin, A. L. (1998). Neuropilin-2 is a receptor for semaphorin
IV: insight into the structural basis of receptor function and specificity.
Neuron 21, 1079-1092.
Goodman, C. S. (1996). Mechanisms and molecules that control growth cone
guidance. Annu. Rev. Neurosci. 19, 341-377.
Hamburger, V. and Hamilton, H. L. (1992). A series of normal stages in the
development of the chick embryo. 1951. Dev. Dyn. 195, 231-272.
He, Z. and Tessier-Lavigne, M. (1997). Neuropilin is a receptor for the axonal
chemorepellent Semaphorin III. Cell 90, 739-751.
He, Z., Wang, K. C., Koprivica, V., Ming, G. and Song, H. J. (2002).
Knowing how to navigate: mechanisms of semaphorin signaling in the
nervous system. Sci STKE http://stke.sciencemag.org/cgi/content/full/
sigtrans;2002/119/re1.
Holmes, G. and Niswander, L. (2001). Expression of slit-2 and slit-3 during
chick development. Dev. Dyn. 222, 301-307.
Holmes, G. P., Negus, K., Burridge, L., Raman, S., Algar, E., Yamada, T.
and Little, M. H. (1998). Distinct but overlapping expression patterns of
two vertebrate slit homologs implies functional roles in CNS development
and organogenesis. Mech. Dev. 79, 57-72.
Kennedy, T. E., Serafini, T., de la Torre, J. R. and Tessier-Lavigne, M.
(1994). Netrins are diffusible chemotropic factors for commissural axons in
the embryonic spinal cord. Cell 78, 425-435.
Keynes, R., Tannahill, D., Morgenstern, D. A., Johnson, A. R., Cook, G.
M. and Pini, A. (1997). Surround repulsion of spinal sensory axons in
higher vertebrate embryos. Neuron 18, 889-897.
Keynes, R. J. and Stern, C. D. (1984). Segmentation in the vertebrate nervous
system. Nature 310, 786-789.
Kitsukawa, T., Shimizu, M., Sanbo, M., Hirata, T., Taniguchi, M., Bekku,
Y., Yagi, T. and Fujisawa, H. (1997). Neuropilin-semaphorin III/Dmediated chemorepulsive signals play a crucial role in peripheral nerve
projection in mice. Neuron 19, 995-1005.
Koblar, S. A., Krull, C. E., Pasquale, E. B., McLennan, R., Peale, F. D.,
Cerretti, D. P. and Bothwell, M. (2000). Spinal motor axons and neural
crest cells use different molecular guides for segmental migration through
the rostral half-somite. J. Neurobiol. 42, 437-447.
Kolodkin, A. L., Levengood, D. V., Rowe, E. G., Tai, Y. T., Giger, R. J. and
Ginty, D. D. (1997). Neuropilin is a semaphorin III receptor. Cell 90, 753762.
Koppel, A. M., Feiner, L., Kobayashi, H. and Raper, J. A. (1997). A 70
amino acid region within the semaphorin domain activates specific cellular
response of semaphorin family members. Neuron 19, 531-537.
Landolt, R. M., Vaughan, L., Winterhalter, K. H. and Zimmermann, D.
R. (1995). Versican is selectively expressed in embryonic tissues that act as
barriers to neural crest cell migration and axon outgrowth. Development 121,
2303-2312.
Li, H. S., Chen, J. H., Wu, W., Fagaly, T., Zhou, L., Yuan, W., Dupuis, S.,
Jiang, Z. H., Nash, W., Gick, C. et al. (1999). Vertebrate slit, a secreted
ligand for the transmembrane protein roundabout, is a repellent for olfactory
bulb axons. Cell 96, 807-818.
Luo, Y., Raible, D. and Raper, J. A. (1993). Collapsin: a protein in brain that
induces the collapse and paralysis of neuronal growth cones. Cell 75, 217227.
Luo, Y., Shepherd, I., Li, J., Renzi, M. J., Chang, S. and Raper, J. A. (1995).
A family of molecules related to collapsin in the embryonic chick nervous
system. Neuron 14, 1131-1140. Erratum in Neuron 15 following p. 1218.
Martin-Zanca, D., Barbacid, M. and Parada, L. F. (1990). Expression of
the trk proto-oncogene is restricted to the sensory cranial and spinal ganglia
of neural crest origin in mouse development. Genes Dev. 4, 683-694.
Masuda, T., Okado, N. and Shiga, T. (2000). The involvement of axonin1/SC2 in mediating notochord-derived chemorepulsive activities for dorsal
root ganglion neurites. Dev. Biol. 224, 112-121.
Messersmith, E. K., Leonardo, E. D., Shatz, C. J., Tessier-Lavigne, M.,
Goodman, C. S. and Kolodkin, A. L. (1995). Semaphorin III can function
as a selective chemorepellent to pattern sensory projections in the spinal
cord. Neuron 14, 949-959.
Miyazaki, N., Furuyama, T., Amasaki, M., Sugimoto, H., Sakai, T.,
Takeda, N., Kubo, T. and Inagaki, S. (1999a). Mouse semaphorin H
inhibits neurite outgrowth from sensory neurons. Neurosci. Res. 33, 269274.
Miyazaki, N., Furuyama, T., Sakai, T., Fujioka, S., Mori, T., Ohoka, Y.,
Takeda, N., Kubo, T. and Inagaki, S. (1999b). Developmental localization
of semaphorin H messenger RNA acting as a collapsing factor on sensory
axons in the mouse brain. Neuroscience 93, 401-408.
Mueller, B. K. (1999). Growth cone guidance: first steps towards a deeper
understanding. Annu. Rev. Neurosci. 22, 351-388.
Nakamoto, K. and Shiga, T. (1998). Tissues exhibiting inhibitory [correction
of inhibitory] and repulsive activities during the initial stages of neurite
outgrowth from the dorsal root ganglion in the chick embryo. Dev. Biol. 202,
304-314.
Nakamura, F., Tanaka, M., Takahashi, T., Kalb, R. G. and Strittmatter,
S. M. (1998). Neuropilin-1 extracellular domains mediate semaphorin D/IIIinduced growth cone collapse. Neuron 21, 1093-1100.
Newgreen, D. F., Scheel, M. and Kastner, V. (1986). Morphogenesis of
sclerotome and neural crest in avian embryos. In vivo and in vitro studies
on the role of notochordal extracellular material. Cell Tissue Res. 244, 299313.
Nguyen Ba-Charvet, K. T., Brose, K., Marillat, V., Kidd, T., Goodman, C.
S., Tessier-Lavigne, M., Sotelo, C. and Chedotal, A. (1999). Slit2mediated chemorepulsion and collapse of developing forebrain axons.
Neuron 22, 463-473.
Nguyen-Ba-Charvet, K. T., Brose, K., Marillat, V., Sotelo, C., TessierLavigne, M. and Chedotal, A. (2001). Sensory axon response to substratebound Slit2 is modulated by laminin and cyclic GMP. Mol. Cell. Neurosci.
17, 1048-1058.
Niclou, S. P., Jia, L. and Raper, J. A. (2000). Slit2 is a repellent for retinal
ganglion cell axons. J. Neurosci. 20, 4962-4974.
Nieto, M. A., Patel, K. and Wilkinson, D. G. (1996). In situ hybridization
analysis of chick embryos in wholemount and tissue sections. Methods Cell
Biol. 51, 219-235.
Niwa, H., Yamamura, K. and Miyazaki, J. (1991). Efficient selection for
high-expression transfectants with a novel eukaryotic vector. Gene 108, 193199.
Ohta, K., Tannahill, D., Yoshida, K., Johnson, A. R., Cook, G. M. and
Keynes, R. J. (1999). Embryonic lens repels retinal ganglion cell axons.
Dev. Biol. 211, 124-132.
Patel, K., Nash, J. A., Itoh, A., Liu, Z., Sundaresan, V. and Pini, A. (2001).
Slit proteins are not dominant chemorepellents for olfactory tract and spinal
motor axons. Development 128, 5031-5037.
Pettway, Z., Guillory, G. and Bronner-Fraser, M. (1990). Absence of neural
crest cells from the region surrounding implanted notochords in situ. Dev.
Biol. 142, 335-345.
Phillips, H. S. and Armanini, M. P. (1996). Expression of the trk family of
neurotrophin receptors in developing and adult dorsal root ganglion neurons.
Philos. Trans. R. Soc. London B Biol. Sci. 351, 413-416.
Pond, A., Roche, F. K. and Letourneau, P. C. (2002). Temporal regulation
of neuropilin-1 expression and sensitivity to semaphorin 3A in NGF- and
NT3-responsive chick sensory neurons. J. Neurobiol. 51, 43-53.
Puschel, A. W., Adams, R. H. and Betz, H. (1995). Murine semaphorin
D/collapsin is a member of a diverse gene family and creates domains
inhibitory for axonal extension. Neuron 14, 941-948.
Puschel, A. W., Adams, R. H. and Betz, H. (1996). The sensory innervation
of the mouse spinal cord may be patterned by differential expression of and
differential responsiveness to semaphorins. Mol. Cell. Neurosci. 7, 419431.
Raper, J. A. (2000). Semaphorins and their receptors in vertebrates and
invertebrates. Curr. Opin. Neurobiol. 10, 88-94.
Renzi, M. J., Feiner, L., Koppel, A. M. and Raper, J. A. (1999). A dominant
negative receptor for specific secreted semaphorins is generated by deleting
an extracellular domain from neuropilin-1. J. Neurosci. 19, 7870-7880.
Rickmann, M., Fawcett, J. W. and Keynes, R. J. (1985). The migration of
neural crest cells and the growth of motor axons through the rostral half of
the chick somite. J. Embryol. Exp. Morphol. 90, 437-455.
Ring, C., Hassell, J. and Halfter, W. (1996). Expression pattern of collagen
IX and potential role in the segmentation of the peripheral nervous system.
Dev. Biol. 180, 41-53.
Ringstedt, T., Braisted, J. E., Brose, K., Kidd, T., Goodman, C., TessierLavigne, M. and O’Leary, D. D. (2000). Slit inhibition of retinal axon
growth and its role in retinal axon pathfinding and innervation patterns in
the diencephalon. J. Neurosci. 20, 4983-4991.
Shepherd, I., Luo, Y., Raper, J. A. and Chang, S. (1996). The distribution
Axon repulsion by the notochord 1133
of collapsin-1 mRNA in the developing chick nervous system. Dev. Biol.
173, 185-199.
Shepherd, I. T., Luo, Y., Lefcort, F., Reichardt, L. F. and Raper, J. A.
(1997). A sensory axon repellent secreted from ventral spinal cord explants
is neutralized by antibodies raised against collapsin-1. Development 124,
1377-1385.
Silver, J. (1994). Inhibitory molecules in development and regeneration. J.
Neurol. 242, S22-24.
Stern, C. D., Artinger, K. B. and Bronner-Fraser, M. (1991). Tissue
interactions affecting the migration and differentiation of neural crest cells
in the chick embryo. Development 113, 207-216.
Takagi, S., Kasuya, Y., Shimizu, M., Matsuura, T., Tsuboi, M.,
Kawakami, A. and Fujisawa, H. (1995). Expression of a cell adhesion
molecule, neuropilin, in the developing chick nervous system. Dev. Biol.
170, 207-222.
Takahashi, T., Nakamura, F., Jin, Z., Kalb, R. G. and Strittmatter, S. M.
(1998). Semaphorins A and E act as antagonists of neuropilin-1 and agonists
of neuropilin-2 receptors. Nat. Neurosci. 1, 487-493.
Taniguchi, M., Yuasa, S., Fujisawa, H., Naruse, I., Saga, S., Mishina, M.
and Yagi, T. (1997). Disruption of semaphorin III/D gene causes severe
abnormality in peripheral nerve projection. Neuron 19, 519-530.
Tannahill, D., Britto, J. M., Vermeren, M. M., Ohta, K., Cook, G. M. and
Keynes, R. J. (2000). Orienting axon growth: spinal nerve segmentation and
surround-repulsion. Int. J. Dev. Biol. 44, 119-127.
Tosney, K. W. (1987). Proximal tissues and patterned neurite outgrowth at the
lumbosacral level of the chick embryo: deletion of the dermamyotome. Dev.
Biol. 122, 540-558.
Tosney, K. W. and Oakley, R. A. (1990). The perinotochordal mesenchyme
acts as a barrier to axon advance in the chick embryo: implications for a
general mechanism of axonal guidance. Exp. Neurol. 109, 75-89.
Trousse, F., Marti, E., Gruss, P., Torres, M. and Bovolenta, P. (2001).
Control of retinal ganglion cell axon growth: a new role for Sonic hedgehog.
Development 128, 3927-3936.
Tzarfati-Majar, V., Burstyn-Cohen, T. and Klar, A. (2001). F-spondin is a
contact-repellent molecule for embryonic motor neurons. Proc. Natl. Acad.
Sci. USA 98, 4722-4727.
Varela-Echavarria, A., Tucker, A., Puschel, A. W. and Guthrie, S. (1997).
Motor axon subpopulations respond differentially to the chemorepellents
netrin-1 and semaphorin D. Neuron 18, 193-207.
Vermeren, M. M., Cook, G. M., Johnson, A. R., Keynes, R. J. and
Tannahill, D. (2000). Spinal nerve segmentation in the chick embryo:
analysis of distinct axon-repulsive systems. Dev. Biol. 225, 241-252.
Wang, H. U. and Anderson, D. J. (1997). Eph family transmembrane ligands
can mediate repulsive guidance of trunk neural crest migration and motor
axon outgrowth. Neuron 18, 383-396.
White, F. A. and Behar, O. (2000). The development and subsequent
elimination of aberrant peripheral axon projections in Semaphorin3A null
mutant mice. Dev. Biol. 225, 79-86.
Wright, D. E., White, F. A., Gerfen, R. W., Silos-Santiago, I. and Snider,
W. D. (1995). The guidance molecule semaphorin III is expressed in regions
of spinal cord and periphery avoided by growing sensory axons. J. Comp.
Neurol. 361, 321-333.
Yuan, W., Zhou, L., Chen, J. H., Wu, J. Y., Rao, Y. and Ornitz, D. M.
(1999). The mouse SLIT family: secreted ligands for ROBO expressed in
patterns that suggest a role in morphogenesis and axon guidance. Dev. Biol.
212, 290-306.