Full text of DOI:10.1016/S0896-6273(00)00123-9

Neuron, Vol. 28, 437–447, November, 2000, Copyright  2000 by Cell Press
Olfactory Sensory Axons Expressing
a Dominant–Negative Semaphorin Receptor Enter
the CNS Early and Overshoot Their Target
Michael J. Renzi, Tamara Lee Wexler,
and Jonathan A. Raper*
cones in vitro has suggested that it plays a role as a
chemorepellent for specific axon tracts during develop-
ment in vivo. For instance, sensory neurons from DRGs
send axons into the spinal cord where they form syn-
apses with their appropriate targets. SEMA-3A expres-
sion in the dorsal spinal cord has been hypothesized to
prevent sensory afferents from entering the cord prema-
turely (Shepherd et al., 1997). Sensory axons enter only
after SEMA-3A expression becomes restricted to the
ventral cord. Later, ventral expression of SEMA-3A may
help prevent most sensory axons from invading ventral
regions of the spinal cord (Messersmith et al., 1995).
However, the analysis of Sema3a knockout mice has
provided little or no evidence that SEMA-3A plays a
role in patterning sensory trajectories in the spinal cord
(Behar et al., 1996; Taniguchi et al., 1997). Based on the
distributions of SEMA-3A in the developing CNS, several
other pathways might be expected to be perturbed by
its absence. These include cerebellar mossy fiber pro-
jections, thalamocortical pathways, basal forebrain pro-
jections, and motor nerves. Thus far, these pathways
also appear to project normally in Sema3a knockout
mice (Catalano et al., 1998). One possible explanation
for these observations is that other guidance molecules,
possibly other class 3 secreted semaphorins, have over-
lapping distributions and functional activities that com-
pensate for the loss of SEMA-3A. One possible ap-
proach that might overcome this problem would be to
express a dominant–negative receptor in vivo that simul-
taneously blocks the functions of multiple class 3 sema-
phorins.
1
4
115 BRB II/III
21 Curie Boulevard
Department of Neuroscience
University of Pennsylvania School of Medicine
Philadelphia, Pennsylvania 19104
Summary
Sensory axons extend from the chick olfactory epithe-
lium to the telencephalon well before the maturation
of their target, the olfactory bulb. During a waiting
period of several days, olfactory axons arrive and ac-
cumulate outside the CNS while the bulb differentiates
beneath them. Semephorin-3A is expressed in the tel-
encephalon during this period and has been proposed
to prevent their entry into the CNS. We show that the
misexpression of a dominant–negative neuropilin-1
that blocks SEMA-3A-mediated signaling in olfactory
sensory axons induces many of them to enter the tel-
encephalon prematurely and to overshoot the olfac-
tory bulb. These results suggest that chemorepellents
can prevent the premature innervation of immature
targets.
Introduction
Growing axons are guided to their targets by specific
guidance cues located in their environment. These can
act as either attractants or repellents, and either at a
distance as diffusible molecules or on contact as extra-
cellular matrix or cell surface molecules. Many of these
guidance cues are likely to bind to receptors on the
growth cone surface and stimulate signaling pathways
that affect growth cone motility. The semaphorins are
a large family of signaling molecules, at least some of
which can function during development as axonal guid-
ance cues (for reviews, see Kolodkin, 1998; Raper 1999).
Class 3 semaphorins are secreted proteins of about 120
kDa (for review, see Yu and Kolodkin, 1999). The best-
characterized member of this class is SEMA-3A (seme-
phorin-3A, formerly known as chick collapsin-1, human
sema-III, mouse sema-D), which acts as a repellent for
growth cones from specific neurons in vitro. These in-
clude DRG neurons (Luo et al., 1993), sympathetic neu-
rons (Adams et al., 1997; Koppel et al., 1997), motor
neurons (Shepherd et al., 1996; Varela-Echavarria et al.,
The receptors for class 3 semaphorins are just now
being fully defined. Neuropilin-1 is a SEMA-3A binding
protein that is absolutely required for SEMA-3A function
(
He and Tessier-Lavigne, 1997; Kitsukawa et al., 1997;
Kolodkin et al., 1997). Neuropilin-1 binds other class 3
semaphorins with approximately equal affinities (Feiner
et al., 1997) and is required for SEMA-3C function as
well (Nakamura et al., 1998; Renzi et al., 1999). A second
neuropilin family member, neuropilin-2, binds SEMA-3C
and SEMA-3F with high affinity but binds poorly, if at all,
to SEMA-3A. Neuropilin-2 is required for the repulsive
effects of SEMA-3C and SEMA-3F on sympathetic neu-
rons in vitro (Chen et al., 1998; Giger et al., 1998; Naka-
mura et al., 1998). Neuropilins probably do not mediate
ligand signaling on their own and appear to require addi-
tional receptor components for this purpose. Members
of the plexin family of transmembrane proteins are likely
candidates for these additional receptor components
(Takahashi et al., 1999; Tamagnone et al., 1999). Plexin-1
forms stable complexes with either neuropilin-1 or neu-
ropilin-2. SEMA-3A binds to plexin-1/neuropilin-1 com-
plexes expressed in Cos cells and induces a change in
their cell morphology analogous to growth cone col-
lapse. A dominant–negative plexin-1 can block SEMA-
1
997), sensory neurons from the cranial nerve ganglia
V and VII (Kobayashi et al., 1997), olfactory sensory neu-
rons (Kobayashi et al., 1997), cortical neurons (Bagnard
et al., 1998), and hippocampal neurons (Chedotal et al.,
1
998).
3
A-induced collapse of sensory growth cones in vitro
The ability of SEMA-3A to collapse and repel growth
(
Takahashi et al., 1999). Semaphorin receptors are there-
fore likely to consist of one or both neuropilins that bind
and present specific semaphorins to one or more plexins
thereby initiating a biological response. The information
*
To whom correspondence should be addressed (e-mail: raperj@
mail.med.upenn.edu).

Neuron
38
4
that is currently available suggests that neuropilin-1 pre-
sents SEMA-3A, neuropilin-2 presents SEMA-3F, and
neuropilin-1 and -2 together present SEMA-3C (Chen et
al., 1998; Giger et al., 1998; Nakamura et al., 1998).
We have previously described a dominant–negative
receptor strategy to block the function of class 3 sema-
phorins (Renzi et al., 1999). Sympathetic neurons are
normally repelled by SEMA-3A, SEMA-3C, and SEMA-
3F. When caused to express a truncated neuropilin-1
that is missing a specific extracellular domain, growth
cones from sympathetic neurons are no longer respon-
sive to SEMA-3A and SEMA-3C, but remain responsive
to SEMA-3F. This truncated form of neuropilin-1 (dnNP-1)
therefore acts as a dominant–negative receptor for spe-
cific secreted semaphorins. A dominant–negative ap-
proach to blocking semaphorin function might be ad-
vantageous since multiple family members with similar
biological functions can be blocked all at once. This
could be important if the loss of a single semaphorin is
partially compensated for by other family members with
similar functions.
In this paper, we report that expressing dnNP-1 in
olfactory sensory axons alters their trajectories when
they reach the telencephalon. Instead of pausing at the
surface of the brain and waiting for their target, the ol-
factory bulb, to mature as they normally would; dnNP-1-
expressing axons enter the brain prematurely and over-
shoot the area that will become their appropriate target.
These results suggest that class 3 semaphorins act as
repellents in vivo and prevent axons from entering their
target prematurely.
Results
Transfection of Olfactory Sensory Neurons
Using In Ovo Electroporation
The expression of dnNP-1 was induced in embryonic
chicks by electroporating an appropriate eukaryotic ex-
pression plasmid in ovo. This method of misexpressing
genes in the chick has several advantages over avian
retroviral vectors. These include: (1) high levels of re-
combinant protein are produced within 8 hr of electro-
poration, (2) large recombinant proteins can be pro-
duced since larger coding inserts are better tolerated
than with viral vectors, (3) there are no reported limits
to the cell types that can be transfected, and (4) expres-
sion of recombinant proteins is restricted to transfected
cells and their progeny. The transfection of olfactory
neurons within the olfactory epithelium is possible since
the olfactory placode that gives rise to the olfactory
epithelium is derived from superficial ectoderm and is
accessible to plasmid DNA delivered from outside the
embryo.
A chimeric protein composed of human placental al-
kaline phosphatase (AP) fused to the transmembrane
and cytoplasmic portions of chick NP-1 was used as a
tracer construct to identify transfected cells and visual-
ize their axonal processes. Expression of this AP-tagged
marker does not affect semaphorin signaling in cul-
tured sympathetic neurons (data not shown). It was
cotransfected in a 1:10 ratio with either a plasmid
carrying the ␤-galactosidase reporter gene as a control
Figure 1. Transfection of the Olfactory Epithelium in the Embryonic
Chick Using In Ovo Electroporation
(
A) A stage 13 embryo marked with blue dye to show the injection
site beneath the amniotic membrane and adjacent to the nasal pit.
The electrodes (ϩ, Ϫ) were placed 5 mm apart and positioned as
shown.
(
B) Distribution of transfected cells in E4 whole-mount embryo. AP-
labeled transfected cells can be seen in and around the nasal pit.
C) A whole-mount preparation of an E6 embryo bisected at the
(
midline and viewed from the medial surface. Labeled olfactory axons
leave the olfactory epithelium at lower left and project within the
olfactory nerve to the nascent olfactory bulb.
lin-1 (APϩNP-1), or a plasmid containing a myc-tagged
dnNP-1 (APϩdnNP-1).
Plasmid DNA was injected into the amniotic sac adja-
cent to the nasal pit of stage 13 embryos (E2) and elec-
troporated with the electrodes oriented to force the plas-
mid toward the embryo (Figure 1A). Visualization of the
(
APϩ␤-gal), a plasmid containing full-length neuropi-

Neuropilin-1 Mediates an Olfactory Axon Waiting Period
39
4
AP marker 2 days later showed transfection of ectoder-
mal cells in and around the nasal pit and in the lens of
the eye (Figure 1B). At later ages, the labeled axons of
olfactory sensory neurons could be seen leaving the
olfactory epithelium to form the olfactory nerve (Figure
1
C). Other structures that are, at least in part, derived
from the ectoderm were also transfected. These in-
cluded the lens of the eye as well as the trigeminal,
vagal, and glossopharyngeal ganglia (data not shown).
Electroporation resulted in a large amount of embryo
mortality. In individual experiments, as many as 50% of
the treated embryos died within 3 days of electropor-
ation. The survival rate decreased further with time, fall-
ing to as low as 20% by 7 days post treatment. This
high lethality is unlikely caused by electroporation itself
or the incorporation of expression plasmids into embry-
onic tissues. Similar rates of lethality are observed with
uninjected embryos electroporated with 0 volts. Surviv-
ing embryos were found to be normal upon gross in-
spection, and the axonal trajectories of olfactory sen-
sory axons in embryos electroporated with control
constructs appeared normal (see below).
Olfactory Axons Expressing dnNP-1 Overshoot
Their Normal Target
The first olfactory axons exit the olfactory epithelium and
reach the telencephalon by E5 where the vast majority of
them halt for several days before entering the CNS (Fig-
ure 2A). As olfactory axons accumulate outside the CNS,
the olfactory bulb evaginates from the telencephalon
and differentiates beneath them (Kobayashi et al., 1997).
Olfactory axons enter the bulb at E8—first projecting
into the superficial olfactory nerve fiber layer (ONL), and
then projecting into and making synaptic connections
within the deeper glomerular layer. To determine if the
expression of dnNP-1 in sensory axons interferes with
their guidance, the trajectories of olfactory axons co-
transfected with APϩ␤gal, APϩNP-1, or APϩdnNP-1
plasmids were compared in E7 embryos.
Olfactory axons normally make contact with the sur-
face of the telencephalon without entering the CNS by
E7 (Figure 2B). Olfactory axons transfected with AP
tracer tag plus ␤-galactosidase (APϩ␤-gal) were seen
to exit the olfactory epithelium, join the olfactory nerve,
and extend to the telencephalon. At E7, the majority of
these axons were found to terminate just outside of
the rostral-most telencephalon where the olfactory bulb
evaginates (Figures 3A, 3C, and 3E). In some of these
control embryos, occasional axons were observed ex-
tending beyond the point where the rest of the olfactory
axons had stopped (see Figure 3E). In embryos where
the bulb had already begun to form, olfactory axons
covered its surface but did not cross the border between
the olfactory bulb and the forebrain (Figure 3C). Migrat-
ing cells transfected with APϩ␤-gal were observed mi-
grating beyond the point at which sensory axons termi-
nate. The majority of these cells were located along a
specific pathway that extended dorsocaudally from the
olfactory nerve for some distance before diving ventrally
toward the midbrain.
Figure 2. Progressive Development of the Olfactory Nerve in the
Embryonic Chick
Coronal sections through the developing olfactory nerve and olfac-
tory sensory axons were visualized using an anti-neurofilament anti-
body and a Cy3-conjugated secondary antibody.
(
A) At E5, sensory axons have grown out of the olfactory epithelium
and crossed the intervening mesenchyme to reach the telencepha-
lon. At this stage, axons stop upon contact with the surface of the
telencephalon and do not enter the CNS.
(B) E7 olfactory axons continue to project to and accumulate on the
surface of the telencephalon.
(
C) By E9, the olfactory bulb has formed and olfactory axons form
the olfactory nerve layer.
Scale bar: 100 ␮m.
Their trajectories within the olfactory nerve appeared
indistinguishable from those transfected with ␤-gal. Al-
though the majority of APϩdnNP-1-transfected axons
were found to terminate on the surface of the telenceph-
Olfactory sensory axons cotransfected with AP tracer
tag plus dnNP-1 (APϩdnNP-1), like controls, entered
the olfactory nerve and projected to the telencephalon.

Neuron
40
4
Figure 3. Olfactory Axons Expressing dnNP-1
Are More Likely to Overshoot their Target
Whole-mount images of E7 chick brains
showing AP-labeled axons (seen in purple)
extending in the olfactory nerve to the telen-
cephalon. The majority of olfactory axons ex-
pressing APϩ␤-gal (A, C, and E) stop outside
of the telencephalon and do not enter the
CNS. Occasionally, a single axon expressing
APϩ␤-gal grew past the surface of the telen-
cephalon and into the CNS. Olfactory axons
expressing dnNP-1 (B, D, and F) show an in-
creased tendency to enter and extend within
the telencephalon.
alon in their normal position, a greatly increased number
as compared to controls was seen to overshoot this
stopping point (Figures 3B, 3D, and 3F). When examined
in whole mount, the overshooting axons appeared to
extend on, or just below, the surface of the telencepha-
lon. The trajectories of these escaping axons varied con-
siderably, but the overwhelming majority grew on the
medial side of the nascent olfactory bulb and projected
caudally along the medial surface of the forebrain. In
contrast, two embryos from a total of 16 each had a
large fascicle of APϩdnNP-1-transfected axons that
projected some distance outside of the lateral surface
of the telencephalon. These fascicles of axons appeared
to break off from the main olfactory nerve prior to con-
tact with the telencephalon. Cells transfected with
APϩdnNP-1 were found migrating in the telencephalon
along a route similar to that seen in controls (Figure 3F).
Olfactory sensory axons cotransfected with AP tracer
tag plus full-length neuropilin-1 (APϩNP-1) behaved just
as control (APϩ␤-gal) axons did. The overexpression
of full-length neuropilin-1 does not enhance or reduce
the responsiveness of cultured sympathetic axons to
SEMA-3A (Renzi et al., 1999). Since the dnNP-1 is a
simple deletion construct made from full-length neuro-
pilin-1, the overexpression of NP-1 on olfactory sensory
axons should reproduce any effects that dnNP-1 ex-
pression might have beyond its ability to neutralize
semaphorin signaling. The aberrant trajectories of olfac-
tory sensory axons expressing dnNP-1 therefore most
likely result from the blockade of semaphorin signaling.
To quantify dnNP-1-induced overshooting of the tar-
get, we compared the number of labeled axons over-
shooting in APϩ␤gal-, APϩNP-1-, and APϩdnNP-1-
transfected embryos. Since more axons were observed
to overshoot when many labeled axons arrived at the
target, data are expressed as the number of overshoot-
ing axons in relation to the total number of labeled olfac-
tory axons in each preparation (Figure 4). Transfected
axons could be counted accurately due to their small
numbers; however, it is possible that the number of
arriving axons might occasionally have been underesti-
mated when they fasciculated closely. In APϩ␤-gal-
transfected embryos, no overshooting axons were ob-
served when fewer than four transfected axons reached
the telencephalon. The number of overshooting axons
did not exceed two in any of the APϩ␤-gal-transfected
embryos analyzed. Identical outcomes were observed
for axons transfected with the full-length neuropilin-1
construct, although as many as three or four aberrantly
projecting axons were observed in embryos with an
unusually large number of transfected axons. In con-
trast, dnNP-1-expressing olfactory axons were ob-
served to overshoot their target when as few as two of
them reached the telencephalon. The number of over-
shooting axons increased dramatically as more trans-
fected olfactory axons reached the telencephalon. An

Neuropilin-1 Mediates an Olfactory Axon Waiting Period
41
4
dnNP-1-Induced Misprojection of Olfactory Axons
Persists in E9 Embryos
We next examined the trajectories of olfactory axons
in E9 embryos transfected with either APϩ␤-gal or
APϩdnNP-1 to see if overshooting axons survived to
later ages and/or converged on inappropriate second-
ary targets. Because survival to this late age was rare
after transfection, only four embryos in each treatment
group were analyzed. Olfactory axons transfected with
APϩ␤-gal extended to the nascent olfactory bulb and
terminated on its surface. No labeled axons were seen
to extend past the caudal margin of the olfactory bulb
and into the forebrain (Figures 6A and 7A). In contrast,
all of the four E9 embryos transfected with APϩdnNP-1
contained at least one labeled axon that had overshot
the olfactory bulb and extended into the forebrain (Fig-
ures 6B and 7D). One particularly dramatic aberrant pro-
jection appeared to extend the entire length of the fore-
brain, turning and branching multiple times.
Figure 4. Quantification of Axon Guidance Errors in Transfected
Olfactory Axons
dnNP-1-Expressing Axons Are Present
in the Olfactory Nerve Fiber Layer
The total number of misprojecting olfactory axons (y axis) is com-
pared to the total number of AP-labeled axons (x axis) in embryos
cotransfected with APϩ␤-gal (open squares), APϩNP-1 (open cir-
cles), or APϩdnNP-1 (closed squares). Regardless of overall trans-
fection levels, axons transfected with dnNP-1 showed a substan-
tially greater number of errors than did control axons. An ANOVA
analysis indicates that dnNP-1-expressing axons are significantly
misguided as compared to the combined controls at a level of signifi-
cance of p Յ 0.0001.
Once the olfactory bulb evaginates from the telencepha-
lon, olfactory axons ramify to form the olfactory nerve
fiber layer (ONL) (Figure 2C). To determine if olfactory
axons-expressing dnNP-1 remain in this layer or instead
enter inappropriate deeper layers, olfactory bulbs from
two E9 embryos cotransfected with APϩ␤-gal and two
E9 embryos cotransfected with APϩdnNP-1 were sec-
tioned and costained with an anti-neurofilament anti-
body. Olfactory axons transfected with APϩ␤-gal were
located within the ONL (Figures 7B and 7C). Serial sec-
tions through a single bulb showed that these axons
were distributed throughout the ONL (data not shown).
Olfactory axons transfected with APϩdnNP-1 were also
located in the ONL. A section containing an overshoot-
ing olfactory axon shows that it extended in the deepest
portion of the ONL (Figures 7E and 7F). In this limited
sample size, we did not detect dnNP-1-expressing ax-
ons that projected abnormally into inappropriate deeper
layers of the bulb.
ANOVA analysis of the data indicates that the frequen-
cies with which ␤-gal-expressing and full-length neuro-
pilin-1-expressing axons overshoot their target are not
significantly different, but that overshooting of domi-
nant–negative expressing axons is greater as compared
to the two control conditions taken together at a signifi-
cance level of p Յ 0.0001.
Overshooting Olfactory Axons Enter
the Telencephalon Prematurely
SEMA-3A Is the Most Likely Candidate for a
Repellent that Prevents Olfactory Axons
from Entering the Telencephalon
Olfactory axons express neuropilin-1, and SEMA-3A in-
duces the collapse of their growth cones in vitro. SEMA-
APϩdnNP-1-transfected embryos were sectioned and
counterstained with an anti-neurofilament antibody to
determine if overshooting olfactory axons entered into
the telencephalon or grew upon its surface. Three em-
bryos were selected representing the usual experimen-
tal result in which AP-labeled, dnNP-1-expressing over-
shooting axons were defasciculated and extended
largely on the medial side of the telencephalon. Labeled
axons were found to extend within the CNS in these
embryos. They grew in the most superficial layers of the
telencephalon just beneath the pial surface (Figures 5B
and 5C). A fourth embryo was selected to represent the
two experimental cases in which AP-labeled, dnNP-1-
expressing, overshooting axons were highly fascicu-
lated and extended on the lateral surface of the telen-
cephalon. The labeled axons in this embryo were found
to extend outside the pial membrane on the surface of
the brain. Labeled axons were bundled together with
additional unlabeled axons that may have originated in
the olfactory epithelium (Figures 5E and 5F).
3
A expression in the telencephalon makes it an attrac-
tive candidate for a repellent that keeps these axons
from entering the telencephalon prematurely (Kobayashi
et al., 1997). To investigate the possibility that other
class 3 semaphorins might have comparable repellent
functions for olfactory axons, we determined whether
SEMA-3C, SEMA-3D, or SEMA-3E is expressed in the
telencephalon during the time period that olfactory ax-
ons are halted at its surface. Along with SEMA-3A, these
represent all of the currently identified class 3 chick
semaphorins.
When olfactory axons are just beginning to reach the
telencephalon at E5, SEMA-3A expression is seen in the
olfactory epithelium and in cells of the most superficial
layer of the telencephalon. SEMA-3C and SEMA-3E ex-
pression are seen in deep layers of the telencephalon.

Neuron
42
4
Figure 5. dnNP-1-Expressing Axons that
Overshoot Their Target Grow into the Telen-
cephalon
The trajectories of olfactory axons expressing
dnNP-1 were reacted with AP histochemistry
and examined in whole mounts at E7 (A and
D). The brains were then sectioned and
probed with anti-neurofilament antibodies to
visualize axons within the CNS (C and F). The
majority of overshooting axons extended on
the medial side of the forebrain. In rare in-
stances (2 of 16 embryos), a large bundle of
axons overshot on the lateral side (D). Bright
field (B), and a composite section of bright
field and fluorescence (C), show axons over-
shooting on the medial side extend within the
telencephalon just below the pial surface. A
bright field (E) and a composite section (F)
show axons overshooting on the lateral side
extended outside the CNS. Scale bar for (B),
(
C), (E), and (F): 20 ␮m.
SEMA-3D expression is not detectable in the E5 telen-
cephalon (data not shown). By E7, SEMA-3A expression
is observed in superficial layers throughout the telen-
cephalon including its point of contact with olfactory
axons (Figure 8). SEMA-3C, SEMA-3D, and SEMA-3E
are no longer expressed in the telencephalon, but all
are expressed very weakly by cells located within the
olfactory nerve itself (Figure 8). SEMA-3D is also ex-
pressed by a subset of cells in the olfactory epithelium
(data not shown). By E9, olfactory axons ramify within
the olfactory nerve layer in the bulb. At this age, SEMA-
3A is no longer expressed on the surface of the bulb
but is restricted to deeper layers (Kobayashi et al., 1997).
SEMA-3C is expressed in a cluster of cells located adja-
cent to the olfactory nerve entry point, and SEMA-3D is
expressed within or just below the developing ONL. No
SEMA-3E expression was detected in the E9 bulb (data
not shown). These results indicate that of these class 3
semaphorins, only SEMA-3A is expressed in the correct
position at the appropriate time to provide a repellent
signal that would prevent olfactory axons from ex-
tending into the telencephalon.
Discussion
The developing olfactory system is an ideal system in
which to study the mechanisms that control axon guid-
ance. It is made up of a relatively homogeneous popula-
tion of sensory cells that project to a recognizable target,
and its development has been extensively characterized
in rats (Santacana et al., 1992), mice (Doucette, 1989),
frogs (Byrd and Burd, 1993a, 1993b), and chickens
(Drapkin and Silverman, 1999).
Primary sensory neurons are located in the olfactory
epithelium that is derived from the olfactory placode.
The olfactory placode invaginates from the surface of
the chick embryo to form the nasal pit beginning at stage
18. The first olfactory axons begin to grow out of the
Figure 6. Overshooting Olfactory Axons Persist in E9 Embryos
The trajectories of olfactory axons from four E9 embryos cotrans-
fected with APϩ␤-gal compared to those of four E9 embryos co-
transfected with APϩdnNP-1.
(
A) Composite sketch of labeled axon trajectories from four embryos
transfected with APϩ␤-gal demonstrate that they project to the
superficial layers of the nascent olfactory bulb. No labeled axons
were found extending beyond the olfactory bulb/forebrain border.
(
B) Axons expressing dnNP-1 covered the surface of the nascent
olfactory bulb, and some of these axons overshot the bulb/forebrain
border and grew extensively over the surface of the brain.

Neuropilin-1 Mediates an Olfactory Axon Waiting Period
43
4
Figure 7. dnNP-1-Expressing Axons Are Confined to the Olfactory Nerve Fiber Layer
Whole-mount views of E9 olfactory axon projections in embryos cotransfected with APϩ␤-gal (A) and APϩdnNP-1 (D). Bright-field images (B
and E) and composites of bright-field and fluorescent images (C and F) showing that both control and dnNP-1-expressing axons extended
in the olfactory nerve layer. Scale bar for (B), (C), (E), and (F): 20 ␮m.
olfactory epithelium and into the adjacent mesenchyme
by late stage 19, These axons have reached the surface
of the telencephalon by E5 (Kobayashi et al., 1997; Drap-
kin and Silverman, 1999). The vast majority of olfactory
sensory axons do not enter the CNS at this time but,
instead, halt at the outside surface of the telencephalon
where the olfactory bulb will form. A small number of
axons do penetrate the telencephalon transiently, ac-
companied by cells that originate in the olfactory epithe-
lium and migrate along the olfactory nerve. Olfactory
axons continue to project from the olfactory epithelium
and accumulate on the surface of the telencephalon.
The bulb forms beneath them over the next several days
in chicks and in other species (Doucette, 1989; San-
tacana et al., 1992; Byrd and Burd, 1993a, 1993b). Olfac-
tory axons cover the surface of the nascent olfactory
bulb to form the olfactory nerve fiber layer (ONL) by E9.
They then leave the ONL to make connections in deeper
layers of the bulb.
What keeps olfactory axons out of the CNS until their
target, the olfactory bulb, has formed? There are several
possible mechanisms: (1) a physical barrier might pre-
vent olfactory axons from penetrating the telencepha-
lon, (2) the telencephalon might not express molecules
permissive for olfactory axon growth, or (3) the telen-
cephalon might contain or secrete a repellent that pre-
vents olfactory axons from entering the CNS.
electron microscopy. These axons project transiently
into the nascent olfactory bulb, sometimes reaching as
far as the ventricular surface before retracting and end-
ing in their appropriate layers. The appearance of these
axons coincides with the appearance of mitral cells, the
major output cell of the olfactory bulb (Byrd and Burd,
1991). This observation has led to the suggestion that
these early, transient incursions may be responsible for
inducing the development of cells within the bulb (Byrd
and Burd, 1993a, 1993b; Gong and Shipley, 1995). It has
also been hypothesized that early invading axons may
play a role in defining the boundary between the olfac-
tory bulb and the forebrain in the rat (Santacana et al.,
1992).
The presence of a repellent in the telencephalon is
an attractive explanation for the failure of early arriving
olfactory axons to enter the CNS. SEMA-3A is a likely
candidate for this repellent. Olfactory sensory axons
express neuropilin-1 and are sensitive to SEMA-3A-
induced growth cone collapse (Kobayashi et al., 1997).
SEMA-3A expression is detectable in superficial cells
of the telencephalon as early as E5 when the olfactory
axons first reach the CNS. High levels of SEMA-3A ex-
pression are maintained in the superficial half of the
telencephalon through E7, the waiting period during
which olfactory axons fail to enter the CNS (Kobayashi
et al., 1997; this study). The invasion of olfactory axons
into the olfactory bulb coincides with the restriction of
SEMA-3A to deeper layers of the bulb. Thus, early ex-
pression of SEMA-3A in the telencephalon may act to
keep olfactory axons from entering the CNS prema-
turely, and its later expression in deep layers of the bulb
may function to prevent them from growing past their
target layers. SEMA-3A has been hypothesized to play
a similar role in preventing sensory axons from entering
the spinal cord early (Shepherd et al., 1997) and, once
Arguing against the presence of a physical barrier or
the idea that the telencephalon is nonpermissive for
olfactory axon growth is the observation that, during
olfactory development in the chick, a small number of
processes can be observed entering the telencephalon
through small breaks in the basal lamina of the radial
glial boundary (Drapkin and Silverman, 1999). Similar
processes have been described in the mouse (Hinds,
1
972; Doucette, 1989) and confirmed to be axons with

Neuron
44
4
Figure 8. Expression of Class 3 Semaphorins in the Developing Olfactory System
Horizontal sections through the telencephalon and developing olfactory nerve of E7 embryos were incubated with DIG-labeled antisense RNA
probes for chick SEMA-3A, SEMA-3C, SEMA-3D, and SEMA-3E. Arrowheads mark the outside edge of the CNS. SEMA-3A is expressed in
superficial layers of the telencephalon, while SEMA-3C, SEMA-3D, and SEMA-3E are expressed weakly in the olfactory nerve. Scale bar:
100 ␮m.
they do enter, in preventing them from projecting into
inappropriate layers of the cord (Messersmith et al.,
related ligands. The 165 amino acid (aa) isoform of vas-
cular endothelial growth factor (VEGF165) and placenta
growth factor 2 (PlGF) are members of the VEGF family
that bind to neuropilin-1. VEGF promotes cell migration,
cell proliferation, and angiogenesis by signaling through
one of the two receptor tyrosine kinases KDR/flk-1 or
flt-1 (Waltenberger et al., 1994; Clauss et al., 1996). Neu-
ropilin-1 is not required for VEGF signaling, but its pres-
ence in the receptor complex increases VEGF165 binding
to KDR and potentiates VEGF165-mediated cell migration
(Soker et al., 1998). PlGF also induces the migration of
specific cell types (Clauss et al., 1996) and may also
promote cell proliferation and angiogenesis (Park et al.,
1994). Neuropilin-1 binds PlGF but the role it plays in
PlGF signaling is unclear (Migdal et al., 1998). At present
neither VEGF nor PlGF have been shown to function
as repellents or as axon guidance cues, and they are
therefore less attractive candidates than the sema-
phorins for an olfactory repellent produced by the telen-
cephalon. Neuropilin-1 can also act as a cell adhesion
molecule that binds unknown heterophilic partners
(Shimizu et al., 2000). Alterations in olfactory axon guid-
ance induced by the expression of dnNP-1 are unlikely
to be caused by an overabundance of adhesive sites
since the expression of full-length NP-1 does not affect
olfactory axon guidance.
1995).
We therefore expressed dnNP-1 in developing olfac-
tory axons to test the hypothesis that a repellent sema-
phorin is responsible for preventing them from entering
the telencephalon early. We found that olfactory axons
expressing dnNP-1 are more likely to overshoot their
target area than normal axons. These overshooting ax-
ons generally enter the CNS and extend in the most
superficial layers of the telencephalon just below the
pial surface. The dnNP-1 construct we used is known
to block the function of some class 3 semaphorins but
not others. The expression patterns of those we exam-
ined in the developing olfactory system suggests that
SEMA-3A, but not SEMA-3B, -3C, or -3D, is expressed
in the appropriate time and place to exclude olfactory
axons from the telencephalon. These results support
the hypothesis that an active repellent, expressed by the
telencephalon, is responsible for preventing olfactory
axons from entering the telencephalon early and that
the most likely candidate for this repellent is SEMA-3A.
Although SEMA-3A is a very attractive candidate for
this repellent activity, the possibility remains that the
errors resulting from dnNP-1 expression could in princi-
ple reflect neuropilin-1’s role in signaling events medi-
ated by other semaphorins or even non-semaphorin-
It may be functionally significant that SEMA-3A can

Neuropilin-1 Mediates an Olfactory Axon Waiting Period
45
4
not only repel, but also paralyze, responsive growth
cones. Under normal conditions, olfactory axons not
only fail to enter the telencephalon, but accumulate out-
side it without crawling away. Olfactory sensory axons
expressing dnNP-1 continue to extend when they would
normally remain stationary outside the CNS. The para-
lytic activity of SEMA-3A may hold them in place near
their target during the waiting period that precedes their
entry.
We have seen some hints that dnNP-1-induced errors
in olfactory axon pathfinding may be amplified by non-
cell-autonomous effects. As previously mentioned, in
two embryos transfected with dnNP-1, aberrant fasci-
cles coursing outside the lateral surface of the olfactory
bulb were found to contain AP tracer–tagged dnNP-1-
expressing axons and a larger number of untagged neu-
rofilament positive axons. Similar axon bundles were
not observed in control embryos. One interpretation of
these observations is that a small number of dnNP-1-
expressing axons failed to halt at their normal location
outside the nascent bulb and, instead, continued ex-
tending abnormally. These axons appear to have acted
as an attractive pathway upon which additional, normal
olfactory axons fasciculated and extended.
Olfactory axons overshoot their target at a low fre-
quency in normal embryos. As described previously, a
small number of transient projections have been ob-
served entering the telencephalon prior to the formation
of the olfactory bulb during the development of the olfac-
tory system (Hinds, 1972; Doucette, 1989; Drapkin and
Silverman, 1999). These axons then retract to end in their
appropriate layers. The overshooting axons observed in
our control experiments probably represent this popula-
tion of axons because they are rare and, since they are
absent in the older embryos, appear to be transient.
The dnNP-1-expressing overshooting axons can persist
through E9 and, therefore, may not be equivalent to the
early entering, transient population of olfactory axons
present in normal embryos.
anterior regions of the hypothalamus; however, they
have been reported to pause before entering the telen-
cephalon just as olfactory axons do (Mulrenin et al.,
1999). If true, the migrating cells that enter the CNS
without delay may represent some other cell type that
is insensitive to the repellent activities of SEMA-3A.
Not all axons marked with the AP tracer tag project
aberrantly. In fact, the majority behave normally even
when they reach the telencephalon. There are several
possible explanations for this. First, it is likely that AP-
labeled axons express dnNP-1 at different levels, and
some may express none at all. For this reason, the full
null phenotype, corresponding to the total blockade of
semaphorin function, can never be revealed by this ap-
proach. To address this issue, we constructed an AP-
tagged dnNP-1 that combines the tracer and dominant–
negative functions in a single construct. However, its
relatively low level of expression did not permit us to
trace dnNP-1-expressing axons with the precision and
confidence provided by the AP tracer construct used in
this series of experiments.
Second, dnNP-1 is unlikely to block the functions of
all of the guidance cues that determine olfactory axon
trajectories. Although dnNP-1 can block the function of
more than one class 3 semaphorin, it is possible that
other semaphorins with overlapping functions could
compensate for their loss. For example, some class 3
semaphorin family members have been shown to act
exclusively through neuropilin-2 (Chen et al., 1998; Giger
et al., 1998), and they may play a role in this system.
Although in situ hybridization experiments suggest that
SEMA-3A is the most likely candidate for a telencephalic
repellent activity, other semaphorins we did not exam-
ine, or other altogether unrelated guidance molecules,
may help keep olfactory axons out of the CNS. Loss-
of-function mutations in C. elegans and Drosophila often
show partial penetrance and variable phenotypes due to
the presence of additional guidance cues with partially
redundant functions (Hedgecock et al., 1990; Wills et
al., 1999).
It is instructive to consider olfactory pathfinding
events that are not perturbed by the expression of dnNP-1.
Expressing axons exit from the olfactory epithelium, en-
ter the olfactory nerve, and project along a normal
course to the telencephalon. SEMA-3A is expressed in
the olfactory epithelium when olfactory axons exit (Ko-
bayashi et al., 1997). SEMA-3A might have been thought
to influence the initial direction in which these axons
extend by repelling them out of, and away from, the
olfactory epithelium. Our finding that axons expressing
dnNP-1 exit the olfactory epithelium and enter the olfac-
tory nerve as they normally would does not support this
function for SEMA-3A. Our results also suggest that
those class 3 semaphorins whose function is inhibited
by NP-1 are not likely to play a very important role in
the guidance of olfactory sensory axons while they grow
in the olfactory nerve. It should also be noted that the
expression of dnNP-1 in olfactory sensory neurons
probably does not affect their cell fate or differentiated
state, since their axonal processes behave normally until
they reach the telencephalon.
A third possible explanation for the apparently normal
behavior of the majority of dnNP-1-expressing olfactory
axons is that only the earliest of them may be affected.
Very early arriving axons that grow past the location
where the olfactory bulb will form never have the oppor-
tunity to contact, recognize, and terminate in their ap-
propriate target. But later arriving axons, even those
expressing dnNP-1, could have that opportunity once
the olfactory bulb has started to differentiate. The differ-
entiated bulb may provide appropriate synaptic sites or
other cues that actively encourage olfactory axons to
stop growing and begin to make synapses.
In conclusion, our results suggest that an active
chemorepellent is responsible for preventing olfactory
axons from entering the telencephalon prematurely and
further suggest that SEMA-3A is the most likely candi-
date for mediating this response. The establishment of
long axonal projections is facilitated if neurons make
their appropriate connections early while distances are
short. The obvious disadvantage of this strategy is that
axons may arrive at their destinations well before their
appropriate targets are ready to be innervated. Chemo-
repellents may provide an active mechanism by which
early arriving axons are prevented from entering a target
that is not yet ready to receive them.
A population of cells born in the olfactory epithelium
migrate along the olfactory nerve and enter the telen-
cephalon without any apparent pause in both our control
and dnNP-1-expressing embryos. GnRH-expressing
cells migrate along this route and ultimately populate

Neuron
46
4
Experimental Procedures
Analysis of Sections
Forebrains that had been dissected from transfected embryos were
cryoprotected in 20% sucrose in PBS and imbedded in O.C.T. em-
bedding compound. Those brains that had been cleared in glycerol
were rehydrated in PBS overnight at 4ЊC prior to cryoprotection.
Sections of 30 ␮m were cut on a cryostat (Leica) and collected on
Superfrost Plus slides (Fischer). Sections were then washed in PBS,
incubated in blocker (2% powdered milk in PBS) for 1 hr, and then
incubated with anti-neurofilament antibody (4H6; Developmental
Hybridoma Bank) diluted in blocker for 3 hr at room temperature.
Neurofilament staining was visualized with a Cy3-conjugated sec-
ondary antibody.
DNA Preparation
PCR was used to generate truncated forms of neuropilin-1. A trun-
cated form of neuropilin-1 that is missing its C domain and acts
as a dominant–negative receptor component (dnNP-1) has been
described previously (Renzi et al., 1999). A more severely truncated
form of neuropilin-1 missing its entire extracellular domain (abc
deletion) was made as a control construct and for tracing axonal
trajectories. Standard PCR amplification was performed between
ACCATCATAGCCATGAGTGCA and CAGAATTCTTACTCGGAAGCA
TGA using oligonucleotide primers that placed a BglII restriction
site 5Ј and a NotI restriction site 3Ј of the amplified sequence. The
resulting fragment was cloned into the AP-PAG vector (Kobayashi
et al., 1997), which added a signal sequence and a human placental
alkaline phosphatase tag at the 5Ј end of the clone. An expression
plasmid containing the ␤-galactosidase reporter gene was received
as a gift from Jeff Golden and was used as a control construct in
expression experiments.
In Situ Hybridization
Chick embryos were staged according to Hamburger and Hamilton
(
1951). Brain sections from E5, E7, and E9 embryos were prepared
for in situ hybridization as follows. Embryos were sacrificed, and
their heads were fixed in 4% paraformaldehyde in PBS at 4ЊC over-
night. The following day, the heads from E5 and E7 embryos were
cryoprotected in 20% sucrose in PBS at 4ЊC overnight. To section
an E9 embryo, the forebrain and olfactory bulb were first dissected
out of the E9 embryo then cryoprotected as described above. Tissue
was then frozen in O.C.T. embedding media compound. Sections
of 35 ␮m were cut on a cryostat (Leica) and collected on Superfrost
Plus slides. Sections were washed in PBS, incubated in acetylation
buffer (3.5 ml triethanolamine, 0.75 ml acetic anhydride in 300 ml
sterile water), then permeabilized in PBT (PBS, 0.1% Triton X-100)
and washed again in PBS, all at room temperature. Sections were
prehybridized in hybridization buffer (50% formamide, 4ϫ SSC, 1ϫ
Denhardt’s reagent, 10% Dextran sulfate, and 0.5 mg/ml fish sperm
DNA) for at least 1 hr at room temperature, then incubated overnight
with the DIG-labeled probe, diluted to 400 ng/ml in hybridization
buffer at 72ЊC. The next day, sections were washed in 0.2ϫ SSC at
72ЊC, rinsed briefly in 0.2ϫ SSC, and rinsed in PBS. Sections were
blocked in blocker (2% powdered milk in PBS) for 1 hr and then
incubated in an alkaline phosphatase anti-DIG antibody diluted
1:2500 in blocker for 3 hr at room temperature. Alkaline phosphatase
was visualized by incubating the sections in AP reaction buffer
without substrate for 5 min and then incubated overnight in AP
reaction buffer containing 0.33 mg/ml NBT and 0.17 mg/ml BCIP.
Electroporation
Eggs from white Leghorn chickens were incubated at 37ЊC until
stage 13–14 (48–52 hr). The eggs were then windowed, and visualiza-
tion of the embryo was aided by injecting a 1:10 dilution of pelican
ink in PBS beneath the embryo for contrast. A glass microcapillary
tube was pulled on a Flamin-Brown Electrode puller, attached to a
100 ␮l Hamilton glass syringe and filled with heavy mineral oil. The
capillary was then loaded with 20 ␮l of DNA (1.5 ␮g/␮l final concen-
tration), suspended in TE (10 mM Tris-HCl [pH 8.0], 1 mm EDTA [pH
8.0]). Plasmid containing AP-abc-del neuropilin-1 (AP tracer tag)
was diluted 1:10 with either plasmid containing the ␤-galactosidase
reporter gene (APϩ␤-gal) for control transfections, plasmid con-
taining full-length neuropilin-1 (APϩNP-1), or with plasmid con-
taining dnNP-1 (APϩdnNP-1). The DNA mixture was then injected
into the amniotic sac just rostral to the nasal pit (Figure 1A).
The electroporation apparatus consisted of a circuit designed to
generate the electric field, a DC power source to supply the voltage,
and a function generator to control the frequency of the pulses. The
electric field was applied to the surface of the egg through platinum
genetrodes (BTX industries). For the electroporation of the olfactory
placode, the electrodes were placed on the surface of the egg as
illustrated in Figure 1A. The electrodes were lowered to form a slight
depression in the vitelline membrane that was then filled with 200
Acknowledgments
␮
l of sterile PBS. Three pulses of 25V at 20 Hz followed by three
The authors would like to thank Hiroaki Kobayashi for reading the
manuscript, and Andrea Missias, Simone Niclou, and Andy Webber
for their advice during the course of the project. We would also like
to thank Len Feiner and Jeff Golden for invaluable assistance with
the electroporation technique. We thank Thomas Kreibich for per-
forming the control experiments demonstrating lethality of mock
electroporated eggs and Jeff Golden for his gift of the ␤-galactosi-
dase expression plasmid. We also thank Jeff Aguire for his statistical
analysis of the data. This work was supported by grants: RO1-
NS26527 from NIH (J. A. R.) and T32HD07516 (M. J. R. and T. L. W.).
pulses of 25V at 10 Hz were then applied to the surface of the
embryo. After electroporation, 200 ␮l of 10ϫ Penicillin/streptomycin
(
Life Technologies) was added to the surface of the egg, the egg
was sealed with tape, and it was placed back in the incubator.
Analysis of Whole Mounts
Chick embryos were sacrificed on the appropriate day and fixed in
4% paraformaldehyde in PBS for 2–4 hr at 4ЊC. Embryos were rinsed
with PBS and then incubated in PBS at 65ЊC for 3 hr to inactivate
endogenous alkaline phosphatase. After inactivation, the embryos
were rinsed in AP reaction mix (0.5 mg MgCl
2
, 0.3 mg NaCl, 5 ml
Received November 29, 1999; revised October 6, 2000.
1M Tris-HCl [pH 9.5], 50 ␮l Tween 20) without substrate for 15 min–1
hr at 4ЊC, then incubated in AP reaction buffer with 0.33 mg/ml NBT
and 0.17 mg/ml BCIP for 1–3 days at 4ЊC in the dark. The reaction
was stopped by washing the embryos in acidic PBS (pH 5.0) con-
taining 0.1% Tween 20 overnight at 4ЊC.
References
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.
Embryos with AP-labeled cells in and around the nostril were
identified (Figure 1B). To better visualize the olfactory nerve, these
embryos were bisected sagitally through the midline of the head so
that they could be viewed from the medial surface (Figure 1C).
Embryos that had AP staining in the olfactory nerve were cleared
in 80% glycerol overnight. The olfactory bulb was dissected out
from the remaining half of the head. Cleared whole mounts were
examined at high power through a dissecting microscope, and indi-
vidual axons in the olfactory nerve near the telencephalon were
counted, as were axons that had grown past the normal olfactory
nerve stopping point (defined as the rostral-most end of the telen-
cephalon in E5 and E7 embryos and the olfactory bulb/forebrain
border in E9 embryos).
16, 6077–6086.
Bagnard, D., Lohrum, M., Uziel, D., Puschel, A.W., and Bolz, J. (1998).
Semaphorins act as attractive and repulsive guidance signals during
the development of cortical projections. Development 125, 5043–
5053.
Behar, O., Golden, J.A., Mashimo, H., Schoen, F.J., and Fishman,
M.C. (1996). Semaphorin III is needed for normal patterning and
growth of nerves, bones and heart. Nature 383, 525–528.
Byrd, C.A., and Burd, G.D. (1991). Development of the olfactory bulb
in the clawed frog, Xenopus laevis: a morphological and quantitative
analysis. J. Comp. Neurol. 314, 79–90.

Neuropilin-1 Mediates an Olfactory Axon Waiting Period
47
4
Byrd, C.A., and Burd, G.D. (1993a). Morphological and quantitative
evaluation of olfactory bulb development in Xenopus after olfactory
placode transplantation. J. Comp. Neurol. 331, 551–563.
tion as a selective chemorepellent to pattern sensory projections in
the spinal cord. Neuron 14, 949–959.
Mulrenin, E.M., Witkin, J.W., and Silverman, A.-J. (1999). Embryonic
development of the gonadotrophin-releasing hormone (GnRH) sys-
tem in the chick: a spatio-temporal analysis of GnRH neuronal gener-
ation, site of origin, and migration. Endocrinology 140, 422–433.
Byrd, C.A., and Burd, G.D. (1993b). The quantitative relationship
between olfactory axons and mitral/tufted cells in developing Xeno-
pus with partially deafferented olfactory bulbs. J. Neurobiol. 24,
1
229–1242.
Migdal, M., Huppertz, B., Tessler, S., Comforti, A., Shibuya, M.,
Reich, R., Baumann, H., and Neufeld, G. (1998). Neuropilin-1 is a
placenta growth factor-2 receptor. J. Biol. Chem. 273, 22272–22278.
Catalano, S.M., Messersmith, E.K., Goodman, C.S., Shatz, C.J., and
Chedotal, A. (1998). Many major CNS axon projections develop nor-
mally in the absence of semaphorin III. Mol. Cell. Neurosci. 11,
Nakamura, F., Tanaka, M., Takahashi, T., Kalb, R.G., and Strittmatter,
S.M. (1998). Neuropilin-1 extracellular domains mediate semaphorin
D/III-induced growth cone collapse. Neuron 21, 1093–1100.
1
73–182.
Chedotal, A., Del Rio, J.A., Ruiz, M., He, Z., Borrell, V., de Castro,
F., Ezan, F., Goodman, C.S., Tessier-Lavigne, M., Sotelo, C., and
Soriano, E. (1998). Semaphorins III and IV repel hippocampal axons
via two distinct receptors. Development 125, 4313–4323.
Park, J.E., Chen, H.H., Winer, J., Houck, K.A., and Ferrara, N. (1994).
Placenta growth factor. Potentiation of vascular endothelial growth
factor bioactivity, in vitro and in vivo, and high affinity binding to
Flt-1 but not to Flk-1/KDR. J. Biol. Chem. 269, 25646–25654.
Chen, H., He, Z., Bagri, A., and Tessier-Lavigne, M. (1998). Sema-
phorin-neuropilin interactions underlying sympathetic axon re-
sponses to class III semaphorins. Neuron 21, 1283–1290.
Raper, J.A. (1999). Semaphorins and their receptors in vertebrates
and invertebrates. Curr. Opin. Neurobiol. 10, 88–94.
Clauss, M., Weich, H., Breier, G., Knies, U., Rockl, W., Waltenberger,
J., and Risau, W. (1996). The vascular endothelial growth factor
receptor Flt-1 mediates biological activities. Implications for a func-
tional role of placenta growth factor in monocyte activation and
chemotaxis. J. Biol. Chem. 271, 17629–17634.
Renzi, M.J., Feiner, L., Koppel, A.M., and Raper, J.A. (1999). A domi-
nant negative receptor for specific secreted semaphorins is gener-
ated by deleting an extracellular domain from neuropilin-1. J. Neu-
rosci. 19, 7870–7880.
Santacana, M., Heredia, M., and Valverde, F. (1992). Transient pat-
tern of exuberant projections of olfactory axons during development
in the rat. Brain Res. Dev Brain Res. 70, 213–222.
Doucette, R. (1989). Development of the nerve fiber layer in the
olfactory bulb of mouse embryos. J. Comp. Neurol. 285, 514–527.
Drapkin, P.T., and Silverman, A.J. (1999). Development of the chick
olfactory nerve. Dev. Dyn. 214, 349–360.
Shepherd, I., Luo, Y., Raper, J.A., and Chang, S. (1996). The distribu-
tion of collapsin-1 mRNA in the developing chick nervous system.
Dev. Biol. 173, 185–199.
Feiner, L., Koppel, A.M., Kobayashi, H., and Raper, J.A. (1997). Se-
creted chick semaphorins bind recombinant neuropilin with similar
affinities but bind different subsets of neurons in situ. Neuron 19,
Shepherd, I.T., Luo, Y., Lefcort, F., Reichardt, L.F., and Raper, J.A.
(
1997). A sensory axon repellent secreted from ventral spinal cord
5
39–545.
explants is neutralized by antibodies raised against collapsin-1. De-
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 sema-
phorin IV: insight into the structural basis of receptor function and
specificity. Neuron 21, 1079–1092.
velopment 124, 1377–1385.
Shimizu, M., Murakami, Y., Suto, F., and Fujisawa, H. (2000). Deter-
mination of cell adhesion sites of neuropilin-1. J. Cell Biol. 148,
1283–1293.
Gong, Q., and Shipley, M.T. (1995). Evidence that pioneer olfactory
axons regulate telencephalon cell cycle kinetics to induce the forma-
tion of the olfactory bulb. Neuron 14, 91–101.
Soker, S., Takashima, S., Miao, H.Q., Neufeld, G., and Klagsbrun,
M. (1998). Neuropilin-1 is expressed by endothelial and tumor cells
as an isoform-specific receptor for vascular endothelial growth fac-
tor. Cell 92, 735–745.
Hamburger, V., and Hamilton, H.L. (1951). A series of normal stages
in the development of the chick embryo. J. Morphol. 88, 49–92.
Takahashi, T., Fournier, A., Nakamura, F., Wang, L., Murakami, Y.,
Kalb, R.G., Fujisawa, H., and Strittmatter, S.M. (1999). Plexin-neu-
ropilin-1 complexes form functional semaphorin-3A receptors. Cell
He, Z., and Tessier-Lavigne, M. (1997). Neuropilin is a receptor for
the axonal chemorepellent Semaphorin III. Cell 90, 739–751.
9
9, 59–69.
Hedgecock, E.M., Culotti, J.G., and Hall, D.H. (1990). The unc-5,
unc-6, and unc-40 genes guide circumferential migrations of pioneer
axons and mesodermal cells on the epidermis in C. elegans. Neuron
Tamagnone, L., Artigiani, S., Chen, H., He, Z., Ming, G., Song, H.,
Chedotal, A., Winberg, M.L., Goodman, C.S., Poo, M., et al. (1999).
Plexins are a large family of receptors for transmembrane, secreted,
and GPI-anchored semaphorins in vertebrates. Cell 99, 71–80.
4
, 61–85.
Hinds, J.W. (1972). Early neuron differentiation in the mouse olfac-
tory bulb. II. Electron microscopy. J. Comp. Neurol. 146, 253–276.
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,
Kitsukawa, T., Shimizu, M., Sanbo, M., Hirata, T., Taniguchi, M.,
Bekku, Y., Yagi, T., and Fujisawa, H. (1997). Neuropilin-semaphorin
III/D-mediated chemorepulsive signals play a crucial role in periph-
eral nerve projection in mice. Neuron 19, 995–1005.
5
19–530.
Varela-Echavarria, A., Tucker, A., Puschel, A.W., and Guthrie, S.
1997). Motor axon subpopulations respond differentially to the
(
Kobayashi, H., Koppel, A.M., Luo, Y., and Raper, J.A. (1997). A role
for collapsin-1 in olfactory and cranial sensory axon guidance. J.
Neurosci. 17, 8339–8352.
chemorepellents netrin-1 and semaphorin D. Neuron 18, 193–207.
Waltenberger, J., Claesson-Welsh, L., Siegbahn, A., Shibuya, M.,
and Heldin, C.H. (1994). Different signal transduction properties of
KDR and Flt1, two receptors for vascular endothelial growth factor.
J. Biol. Chem. 269, 26988–26995.
Kolodkin, A.L. (1998). Semaphorin-mediated neuronal growth cone
guidance. Prog. Brain Res. 117, 115–132.
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
Wills, Z., Bateman, J., Korey, C.A., Comer, A., and Van Vactor, D.
(
1999). The tyrosine kinase Abl and its substrate enabled collaborate
9
0, 753–762.
with the receptor phosphatase Dlar to control motor axon guidance.
Neuron 22, 301–312.
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,
Yu, H., and Kolodkin, A.L. (1999). Semaphorin signaling: a little less
per-plexin. Neuron 22, 11–14.
5
31–537.
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, 217–227.
Messersmith, E.K., Leonardo, E.D., Shatz, C.J., Tessier-Lavigne, M.,
Goodman, C.S., and Kolodkin, A.L. (1995). Semaphorin III can func-