Sonic hedgehog signaling is decoded by calcium spike
activity in the developing spinal cord
Yesser H. Belgacem and Laura N. Borodinsky1
Department of Physiology and Membrane Biology and Institute for Pediatric Regenerative Medicine, Shriners Hospital for Children and University of
California Davis School of Medicine, Sacramento, CA 95817
Edited* by Lynn T. Landmesser, Case Western Reserve University, Cleveland, OH, and approved February 9, 2011 (received for review December 5, 2010)
Evolutionarily conserved hedgehog proteins orchestrate the pat-
terning of embryonic tissues, and dysfunctions in their signaling
can lead to tumorigenesis. In vertebrates, Sonic hedgehog (Shh) is
essential for nervous system development, but the mechanisms
underlying its action remain unclear. Early electrical activity is an-
other developmental cue important for proliferation, migration,
and differentiation of neurons. Here we demonstrate the interplay
between Shh signaling and Ca2+dynamics in the developing spinal
cord. Ca2+imaging of embryonicspinal cells shows that Shh acutely
increases Ca2+spike activity through activation of the Shh corecep-
tor Smoothened (Smo) in neurons. Smo recruits a heterotrimeric
GTP-binding protein-dependent pathway and engages both intra-
cellular Ca2+stores and Ca2+influx. The dynamics of this signaling
are manifested in synchronous Ca2+spikes and inositol triphos-
phate transients apparent at the neuronal primary cilium. Interac-
tion of Shh and electrical activity modulates neurotransmitter
phenotype expression in spinal neurons. These results indicate that
electrical activity and second-messenger signaling mediate Shh
action in embryonic spinal neurons.
G protein|neuronal specification
mediated, has an impact on several aspects of nervous system
development. Neurotransmitter-mediated signaling regulates pro-
liferation of cortical neuroblasts (1) and neuronal migration in the
developing cerebellum (2), hippocampus (3), and subventricular
zone (4). Motor neuron axons are guided by Ca2+-mediated
spontaneous patterned activity (5). The acquisition of neurotrans-
mitter phenotype is regulated by Ca2+spike activity in the de-
dependent events have a great impact on the establishment of
connections among neurons and target cells (10, 11).
During early embryogenesis a gradient of Sonic hedgehog
(Shh) establishes the dorsoventral patterning of the spinal cord
(12–17). Shh persists after this process is finished and guides
commissural spinal axons during midline crossing (18–22). In
vertebrates, Shh binds to its receptor, Patched, allowing the re-
cruitment and activation of Smoothened (Smo) in the primary
cilium that triggers the regulation of targeted gene expression
(23, 24). Smo is a seven-transmembrane receptor, and although
its interaction with GTP-binding protein α-i (Gαi) has been
demonstrated in vivo in Drosophila (25) and in vitro (25, 26), the
functional relevance of this coupling in vertebrates has remained
elusive (27–29). Activation of G protein-coupled receptors often
engages second messengers such as Ca2+, and cilia are structures
especially suitable for coordinating second-messenger dynamics
(30, 31). We investigated the interplay between electrical activity
and Shh signaling in embryonic spinal neurons.
lectrical activity is present in the developing nervous system
before synapse formation. This activity, which is largely Ca2+
Shh Signaling Acutely Regulates Levels of Ca2+Spike Activity in the
Developing Spinal Cord. Spontaneous Ca2+spike activity spans
10 h of Xenopus laevis spinal cord development after neural tube
closure (8, 32). We imaged cells of the ventral and dorsal sur-
faces of the developing neural tube and found that spiking cells
are embryonic neurons (283 of 290 spiking cells expressed the
neuronal marker N-β-tubulin). On the other hand, neural pro-
genitors, identified by sex-determining region Y-box 2 (Sox2)
expression, do not exhibit Ca2+spikes (none of 148 Sox2+cells
spiked) (Fig. 1). Both the incidence of spiking cells and the fre-
quency of Ca2+spikes are higher in ventral cells than in their
dorsal counterparts (Fig. 2 A–C), demonstrating a ventral-to-dorsal
gradient that parallels the Shh gradient (12–14, 16) in the embry-
onic spinal cord. To probe this correspondence, we first confirmed
the presence of Shh and its coreceptor Smo in the developing
spinal cord (Fig. S1 A and B) and then investigated the influence of
modulators of Shh signaling on Ca2+spike activity. We imaged
Ca2+dynamicsin cells of the ventral surface of the Xenopus spinal
cord andfound that recombinantN-terminal Shh peptide (N-Shh)
acutely increases Ca2+spike activity in a dose-dependent manner
(Fig. 2 D–F). This effect is mimicked by an agonist for Smo (SAG)
and is prevented by cyclopamine, a Smo antagonist (Fig. 2G).
Moreover, overexpression of SmoM2, a constitutively active form
of Smo (33), in developing embryos after neural tube closure
increases Ca2+spike activity (Fig. 2 H–K and Fig. S1C).
To determine whether an endogenous gradient of Shh is able
to imprint a gradient of Ca2+spike activity on neurons posi-
tioned across it, we first cultured embryonic spinal cells and
demonstrated that the effects of exogenous Shh and cyclopamine
on Ca2+spike activity observed in vivo also are evident in vitro
(Fig. S2). We then designed an in vitro system of neuron/noto-
chord explant coculture and found that the incidence of spiking
cells is higher in the half of the field containing the explant than
in the other half (Fig. 2L). This differential distribution of spik-
ing cells is prevented by the addition of cyclopamine, suggesting
that Shh secreted from the notochord explant is responsible for
the increased Ca2+spike activity.
These results identify a signaling pathway for Shh involving the
activation of its canonical coreceptor Smo that, in turn, induces
an increase in Ca2+spike activity in developing spinal neurons.
This dose-dependent effect suggests that Ca2+spike activity may
serve as a readout of the Shh gradient.
To elucidate further the molecular mechanisms underlying Shh-
induced increase in Ca2+spike activity, we investigated the par-
ticipation of Ca2+influx and Ca2+release from intracellular
stores, both required for the generation of Ca2+spikes (34). Shh
fails to increase Ca2+spike activity when voltage-gated Ca2+
channels are blocked or extracellular Ca2+is removed, indicating
that Shh-induced Ca2+spikes depend on extracellular Ca2+entry
(Fig. 3A). Smo is a seven-pass transmembrane protein capable of
Author contributions: Y.H.B. and L.N.B. designed research; Y.H.B. performed research;
Y.H.B. and L.N.B. analyzed data; and Y.H.B. and L.N.B. wrote the paper.
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
Freely available online through the PNAS open access option.
1To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| March 15, 2011
| vol. 108
| no. 11www.pnas.org/cgi/doi/10.1073/pnas.1018217108
recruiting heterotrimeric Gαiβγ protein (25, 26). Gαi activation
inhibits adenylate cyclase, decreasing cAMP levels and hence
inhibiting protein kinase A (PKA). The presence of pertussis
toxin (PTX) or overexpression of a constitutively active form of
PKA (a mutated catalytic subunit, CQR) (35) immediately fol-
lowing neural tube closure blocks the Shh-induced increase in
Ca2+spike activity (Fig. 3B and Fig. S3). In contrast, Shh
increases Ca2+spike activity following overexpression of
a dominant negative form of PKA (a mutated regulatory sub-
unit, RAB) (Fig. 3B and Fig. S3) (36), indicating that the Shh-
induced effect depends on PTX-sensitive Gαβγ protein and
subsequent PKA inhibition.
Recruitment of G protein also can activate phospholipase C
(PLC) that in turn increases inositol triphosphate (IP3) levels
and induces Ca2+release from internal stores. Pharmacological
blockade of either PLC or IP3 receptors (IP3R) prevents Shh-
induced Ca2+spike activity, suggesting that Ca2+stores are re-
quired for the Shh-mediated effect (Fig. 3C). In turn, emptying of
Ca2+stores has been proposed as a trigger for transient receptor
potential cation channel (TRPC) activation leading to Ca2+influx
(37). Xenopus TRPC1 (xTRPC1) has been cloned (38) and is
expressed in embryonic spinal neurons (39, 40). Pharmacological
inhibition of TRPCs or molecular knockdown of xTRPC1 (Fig.
S4) blocks the Shh-induced increase in Ca2+spike activity (Fig.
3D). These results identify second messengers and channels in-
volved in Shh signaling in embryonic spinal neurons.
Shh Signaling Induces Synchronous Ca2+Spikes and IP3 Transients at
the Neuronal Primary Cilium. In vertebrates the Shh signaling
machinery clusters and functions in primary cilia (Fig. S1B, Inset)
(23, 24). We find that IP3R localize at the base of the neuronal
primary cilium (Fig. 4A), whereas Gαi and TRPC1 are distrib-
uted along its tip and shaft, respectively (Fig. 4 B and C). In-
cubation of neuronal cultures with SAG expands Gαi localization
to the full extent of the cilium (Fig. 4D), resembling the change in
Smo distribution in NIH 3T3 cells after stimulation with Shh (23,
24). To assess the interplay between Shh and Ca2+spike activity
dynamically, we microinjected mRNA encoding an RFP-tagged
pleckstrin homology (PH) domain from phospholipase C-δ1
[mRFP-PH(PLCδ)], a molecular probe for IP3 levels (41), in de-
veloping embryos to visualize simultaneously Shh-induced Ca2+
spikes and IP3 transients. When Smo is overexpressed (Fig. S1C)
and in the presence of SAG, we observed localized IP3 transients
at the primary cilium synchronized with Ca2+spikes (Fig. 4 E–H).
The onset of IP3 transients precedes the onset of Ca2+spikes by
8 ± 2 s (mean ± SEM; n = 17) (Fig. 4H), suggesting that Shh-
induced Ca2+spikes depend on IP3-induced Ca2+release from
intracellular stores. The incidence of synchronized Ca2+spikes
and IP3 transients is highest in the presence of SAG and is abol-
ished by cyclopamine (Fig. S5). The restricted visualization of
Shh-induced IP3 transients at the primary cilium probably is
caused by localized signaling (23) and not by the inability of the
probe to reveal cytosolic changes in IP3. Indeed, this bioprobe is
able to report global increases in IP3 levels elicited by dihy-
droxyphenylglycol (DHPG), a metabotropic glutamate receptor
(mGluR) agonist (Fig. S6). These results suggest that Shh sig-
naling is able to elicit fast and localized responses at the primary
cilium by recruiting second messengers.
Regulation of Neurotransmitter Specification by Shh Signaling Relies
on the Interplay with Ca2+Spike Activity. Neurotransmitter speci-
fication is a crucial event of neuronal differentiation that enables
the establishment of functional circuits in the developing nervous
system. Ca2+spike activity modulates expression of neuro-
transmitter phenotype in developing neurons, and activity-
dependent components that participate in the transcriptional
regulation of the GABAergic phenotype have been identified
recently (9). Therefore, we investigated whether Shh signaling
acts on electrical activity to modulate GABAergic phenotype
specification (Fig. S7A). We find that enhancement or inhibition
of Shh signaling mimics the effect on GABAergic phenotype
expression observed when Ca2+spike activity is enhanced or
suppressed, respectively (Fig. 5A). The numbers of spinal cells or
ventral progenitors do not change in manipulated embryos (Fig.
S7B), in agreement with the constancy in domains of specified
ventral progenitors observed in chicken embryos in which Shh
signaling had been perturbed at late developmental stages (42).
Imposition of changes in Ca2+spike activity (Fig. S8) occludes
SAG- or cyclopamine-induced phenotypes (Fig. 5A). These
results suggest that Shh signals to engage Ca2+spike activity in
the process of neurotransmitter specification revealing a function
of Shh in postmitotic neuron differentiation.
We propose a model (Fig. 5B) in which Shh activates Smo at the
primary cilium, resulting in the recruitment of PTX-sensitive
Gαβγ protein that leads to activation of PLC and increases in IP3
levels. Opening of IP3R-operated stores and activation of
TRPC1 and voltage-gated channels result in an increase of Ca2+
spike activity. Activated Smo also inhibits PKA, which can inhibit
IP3-induced Ca2+release (43). Hence, modulation of Ca2+spike
activity may occur by pathways that are parallel or convergent to
the one operated by PLC. The precise sequence in which Ca2+
influx and stores operate for the generation of Shh-induced Ca2+
spikes remains to be addressed. For instance, activation of
TRPC1 by emptiness of Ca2+stores may depolarize the mem-
brane, leading to the activation of voltage-gated channels. The
effect on spinal neuron differentiation of this signal transduction
pathway, connecting Shh with Ca2+spike activity, demonstrates
integration of genetically driven and electrical activity-dependent
mechanisms. This interplay allows greater plasticity and more
efficient proofreading of nervous system development (44).
Shh drives the dorsoventral patterning of the neural tube early
in development by regulating gene expression. Dorsoventral
(A) Ca2+imaging of the ventral spinal cord of a stage-24 (26-h-postfertiliza-
tion) embryo for 20 min. Circles identify cells spiking during 20-min recording.
Inset shows Ca2+spike activity for the cell outlined in yellow. (B) (Left) After
imaging, the same preparation was whole-mount immunostained for Sox2
and N-β-tubulin. (Right) Immunostaining of a transverse section of the spinal
cord from a stage-24 embryo. (C) Ca2+imaging of an open-book spinal cord
preparation. (D) Whole-mount immunostaining of the same preparation for
Sox2 and N-β-tubulin. (E) Diagram of the open-book spinal cord preparation
shown in C and D. D, dorsal; V, ventral. (Scale bars, 20 μm.)
Spiking cells in the developing neural tube are postmitotic neurons.
Belgacem and BorodinskyPNAS
| March 15, 2011
| vol. 108
| no. 11
excitability of the developing spinal cord may be set by this
transcription-dependent mechanism. Later in development Shh
may contribute to maintaining and modulating the dorsoventral
gradient of electrical activity in spinal neurons independently of
transcription, as suggested in the present study. Specific patterns
of electrical excitability along the dorsoventral axis of the de-
veloping spinal cord have been identified in several species. In
Xenopus embryos, dorsal spinal cells comprising Rohon–Beard
sensory neurons and a minority of interneurons generate sparse
numbers of action potentials upon sustained depolarization; in
contrast, ventral cells, including motor neurons and the majority
of interneurons, fire repetitively (45). Similarly, in larval zebra-
fish, recruitment of excitatory dorsal spinal neurons requires
higher levels of stimulation than do their ventral counterparts,
whereas inhibitory spinal neurons show the opposite pattern; this
topography underlies distinctive swimming behaviors (46). This
theventral thaninthedorsalneural tube(stage24).(B)Afterimaging,thesame preparationwas whole-mountimmunostainedforhomeodomainproteinHb9,
a ventrally expressed neuronal marker, to indicate its dorsoventral orientation. Circles identify cells spiking during 20-min recording, and Insets in A show Ca2+
Ventral view of stage-24 developing spinal cord in the absence (D) or presence (E) of N-Shh. Insets show Ca2+spike activity during 15-min recording from the
same cell (outlined in yellow). (F) Dose–response curve for N-Shh–induced Ca2+spike activity. Data are mean ± SEM percent of spiking cells in the presence of
N-Shh compared to number of cells spiking before addition of N-Shh (0). (G) Dose–response curve for cyclopamine blockade of Ca2+spike activity induced by
SAG. Data are mean ± SEM percent of spiking cells inthepresence ofSAG andcyclopaminecompared tonumber ofcells spiking before addition ofcyclopamine
(0). (H–K) Expression of SmoM2 increases Ca2+spike activity. (H) Electroporation of a stage-19 embryo with SmoM2 demonstrates a higher incidence of Ca2+
spike activity 6 h after electroporation (stage 24) in electroporated cells (red) than in nonelectroporated cells (black). (I) Effective overexpression of SmoM2 was
verified by whole-mount immunostaining against Smo after Ca2+imaging. Circles identify cells spiking during recording. (J) Ca2+spike activity during 20-min
recording for immunonegative and immunopositive cells outlined in yellow in H and I. (K) Bar graphs show mean ± SEM percent incidence of spiking cells and
spike frequency for electroporated (SmoM2) and nonelectroporated (Control) cells. n = 5 stage-24 (26-h postfertilization) embryos per experimental group (C–
K). (L) Endogenous Shh released by the notochord increases Ca2+spike activity of neurons. (Upper) Dissociated neuron/notochord explant (Not) coculture.
(Lower) The imaged field was divided in halves proximal and distal to the notochord explant. Values are mean ± SEM percent of spiking cells in proximal and
distal regions in the absence or presence of cyclopamine (Cyclo). n = 5 independent cultures; *P < 0.05. (Scale bars, 20 μm.)
Shh increases Ca2+spike activity of developing spinal neurons. (A) Lateral view of a developing spinal cord showing higher levels of Ca2+spike activity in
| www.pnas.org/cgi/doi/10.1073/pnas.1018217108Belgacem and Borodinsky
profile appears to be rooted in early development through an
orderly addition of neurons to the developing network (47). In
chicken embryos, neuronal activity is higher in the ventral two
thirds of the spinal cord than in the dorsal region (48). Taken
together these studies suggest that the patterning of excitability
along the dorsoventral axis of the developing spinal cord is highly
conserved and relevant for proper spinal cord development.
The necessity of a subcellular compartment such as the primary
cilium for Shh signaling (12, 23, 24, 49, 50) allows the spatio-
temporal integration of two second-messenger codes generated by
Ca2+and IP3 transients. The universal character of second-mes-
senger signaling predicts that this pathway is common to different
cell types, although different classes of cells may exhibit distinctive
second-messenger dynamics (51, 52). It will be of interest to in-
vestigate how these steps of decoding Shh signaling are connected
to other elements of its canonical pathway and to determine the
mechanisms by which the interactions between Shh, IP3, and Ca2+
are interpreted and translated into expression of specific genes.
Materials and Methods
Cell Cultures. Cell cultures were grown as previously described (8). For neuron–
were prepared aspreviously described (8),grown for 5 h, and loaded with 1 μM
fluo4-AM (Invitrogen). A 0.004-mm3piece of notochord from a stage-24 em-
conjugate (Invitrogen) was placed in the neuronal culture.
Ca2+Imaging. Ca2+imaging was performed as described previously (8). Stage-
23 to -26 (24.75- to 29.5-h-old) neural tubes were exposed and loaded with 1
spike activity. (A–D) Ca2+imaging of the ventral spinal cord. (A) Ca2+influx
was blocked by a mixture of Na+and Ca2+voltage-gated channel blockers
(VGC block) or by perfusion with a Ca2+-free medium (Ca2+-free). (B) Gαi was
inhibited by 10 mM PTX. Perturbations of PKA activity were implemented
by electroporating constitutively active (CQR) or dominant negative (RAB)
forms of PKA in stage-19 embryos. Ca2+imaging was performed 6 h after
electroporation. (C) PLC was inhibited by 10 μM U73122, and IP3R were
inhibited by 20 μM 2-aminoethoxydiphenyl borate (2-APB) or 20 μM xesto-
spongin C (XeC). (D) TRPC channels were blocked by 50 μM SKF96365 or by
molecular knockdown with xTRPC1 morpholino (MO). Control morpholino
(CMO). Values are mean ± SEM percent incidence of spiking cells in the
ventral surface of neural tubes compared with control (30-min recording
before addition of 100 nM SAG). n = 5 stage-24 (26-h postfertilization)
embryos per experimental group; *P < 0.05 (A–D).
Molecular identification of the components linking Shh and Ca2+
mary cilium. (A–D) Immunostaining of immature spinal neurons grown in
vitro for 7 h. Acetylated tubulin staining is shown in green, and DAPI staining
is shown in blue. (A) IP3R (red) localize at the base of the primary cilium. (B)
TRPC1 (red) localizes to the primary cilium. (C and D) Gαi protein (red) local-
ization at the primary cilium expands when Shh signaling is enhanced.
Numbers correspond to the mean ± SEM percent of acetylated tubulin la-
beling that overlaps with Gαi staining at the primary cilium in the absence (C)
or presence (D) of 100 nM SAG for 4 h. n = 10 cells per condition; *P < 0.005.
(E and F) Simultaneous Ca2+and IP3 imaging reveals synchronous transients.
(E) Images correspond to a time before (Left), during (Center), and after
(Right) the spike indicated in the trace in F. (F) Traces represent the changes in
fluorescence intensity for IP3 and Ca2+probes in regions of interest (ROI) in-
dicated in E, Right. (G) IP3 transients are apparent at the primary cilium. The
cell is the same shown in E, stained with DAPI (blue) and anti-acetylated tu-
bulin (green) and overlapped with IP3 frame (red) corresponding to the peak
of the transient shown in E, Center. (Scale bars, 10 μm.) (H) Synchronicity of
Ca2+and IP3 transients. Graph represents onset time of Ca2+spikes vs. onset
time of IP3 transients during simultaneous recordings. Inset represents the
histogram of the difference between onset times; Δt = tIP3 − tCa2+.
Shh and second-messenger signaling converge at the neuronal pri-
Belgacem and BorodinskyPNAS
| March 15, 2011
| vol. 108
| no. 11
μM fluo4-AM. Ca2+imaging was performed at an acquisition rate of 0.2 Hz
with a Nikon Swept-field confocal microscope. The effects of proteins and
drugs were assessed by recording for 30 min before and after addition of
each agent. Changes in Ca2+spike activity were assessed by comparisons of
the two recordings (paired t test).
Drugs were incubated for 30 min with the exception of pertussis toxin
(PTX) (Tocris) for 1 h and N-terminal Sonic hedgehog (Shh) peptide (N-Shh)
(R&D Systems) and Shh agonist (SAG) (Calbiochem) for 10 min. The con-
centrations of drugs used were N-Shh: 0.1–100 nM; cyclopamine: 0.2–10
μM (LC Laboratories), SAG: 100 nM; Na+and Ca2+voltage-gated chan-
nel blockers (VGC block): 20 nM calcicludine (Calbiochem), 1 μM GVIA
ω-conotoxin, 1 μM flunarizine, and 1 μg/mL tetrodotoxin (Sigma); voltage-
gated Na+channel agonist: 1 μM veratridine (Sigma); GTP-binding protein
alpha-i (Gαi) inhibitor: 10 mM PTX; phospholipase C (PLC) inhibitor: 10 μM
U73122 (Tocris); inositol triphosphate 3 (IP3) receptor (IP3R) inhibitors:
20 μM 2-aminoethoxydiphenyl borate (2-APB) (Tocris) and 20 μM Xesto-
spongin C (XeC) (Calbiochem); and transient receptor potential cation
channel (TRPC) inhibitor: 50 μM SKF96365 (Tocris).
IP3 Imaging. RFP-PH(PLCδ), an RFP-tagged pleckstrin homology (PH) domain
from phospholipase C–δ1 serving as a PI(4,5)P2/IP3biosensor, was used to
monitor IP3 levels in cultured cells (41). RFP-PH(PLCδ) was subcloned in the
pCS2+. The RFP-PH(PLCδ) sequence was amplified by PCR. A BamH1 re-
striction site was added to the sense primer (sense: AGTTACAGGATCC-
GCTGGTTTAGTGAACCGTCAG; antisense: AAAACCTCTACAAATGTGGTATGG-
CTGATT). PCR product and plasmid were then digested by BamH1 and
EcoR1. After ligation and purification, mRNA was synthesized as previously
described (8). Neuron-enriched cultures prepared from neural plates of
embryos microinjected at the two-cell stage with 700 pg of mRNA encoding
RFP-PH(PLCδ) and 300 pg of human Smoothened (Smo) were incubated for
10 min with 100 nM SAG and confocally imaged at 0.2 Hz for 20 min.
In Vivo Gene Misexpression. For SmoM2 overexpression, mRNAs were syn-
thesized as previously described (8). Four nanoliters of 400 ng/μL mRNA
encoding SmoM2 along with 20 mg/mL Alexa Fluor 594 dextran were
microinjected in the neural tube lumen of stage-19 embryos (20.75 h post-
fertilization) followed by electroporation (10 pulses of 70 V and 90-ms du-
ration). For protein kinase A (PKA) misexpression, PCR products containing
the T7 RNA polymerase promoter were used to synthesize mRNAs encoding
a dominant negative form (RAB) or a constitutively active (CQR) form of
PKA. The primers used were sense, TAATACGACTCACTATAGGGACTC-
CGTAGCTCCAGCTTCAC and antisense, GTGAAACCCCGTCTCTACCA. Four
nanoliters of 250 ng/μL mRNA encoding either of these two constructs along
with 20 mg/mL Alexa Fluor 594 dextran were microinjected in the neural
tube lumen of stage-19 embryos followed by electroporation (10 pulses of
70 V and 90-ms duration). Controls were electroporated with Alexa Fluor
594 dextran only. For Xenopus TRPC (xTRPC1) knockdown, embryos at the
two-cell stage were microinjected with 100 pg xTRPC1 morpholino or 5-
mispaired control morpholino (MO and CMO, respectively; Genetools) along
with 20 mg/mL Alexa Fluor 594 dextran. Morpholino oligonucleotides were
designed as in Wang and Poo, 2005 (40). For Smo overexpression, mRNA
was synthesized as previously described (8). Three hundred picograms
mRNA encoding human Smo were microinjected bilaterally in both blasto-
meres of embryos at the two-cell stage.
Western Blots. Western blots were performed as previously described (10).
Protein extracts were obtained from 10 dissected neural tubes from stage-25
embryos for each experimental group using the following antibodies: anti-
Smo, 1:1,000 (Sigma); anti-GAPDH, 1:1,000, anti-PKAIα reg, 1:100, and anti-
PKAα cat, 1:100 (Santa Cruz Biotechnology); and anti-TRPC1, 1:300, Osense
and gift from G. J. Barritt (Flinders University, Adelaide, Australia). Secondary
peroxidase-conjugated antibodies (Jackson ImmunoResearch) were used at
In Vivo Drug Delivery. In vivo drug delivery was performed as previously
described (8, 10). Agarose beads (80 μm; BioRad) were loaded for at least 1 h
with 1 mM veratridine and the Ca2+spike blockers 200 nM calcicludine, 10
μM GVIA ω-conotoxin, 10 μM flunarizine, 10 μg/mL tetrodotoxin, 1 μM SAG,
and 200 μM cyclopamine, or a combination of these agents and implanted
in stage-19 embryos (20-h postfertilization). Stage-34 (2-d postfertilization)
larvae were sectioned for immunostaining.
Immunostaining. Samples were fixed with 4% paraformaldehyde (PFA) and
processed for immunostaining as previously described (8). Incubations with
primary and secondary antibodies were carried out overnight at 4 °C and for
2 h at 23 °C, respectively. Primary antibodies used were directed to acetylated
tubulin, 1:1,000 (Sigma); GαI, 1:50, and IP3R 1:50, (Santa Cruz Biotechnology);
TRPC1, 1:50 (Developmental Studies Hybridoma Bank); sex-determining re-
gion Y-box 2 (Sox2), 1:50 (R&D Systems); GABA, 1:100 (Millipore); N-tubulin,
1:1,000 (Sigma); Shh, 1:100 (Developmental Studies Hybridoma Bank); Smo,
used for detecting endogenous Smo, 1:20 (Abcam); Smo, used for detecting
overexpressed human Smo, does not recognize endogenous Smo 1:300
(Sigma). Immunoreactive cells were counted in at least 20 consecutive 12-μm
sections per embryo.
Data Collection and Statistics. Regions of interest for detection of IP3 levels at
the primary cilium were defined by the area labeled by anti-acetylated tu-
bulin, a marker of primary cilium.
At least five samples were analyzed for each group from at least three
independent clutches of embryos. Statistical tests used were paired or un-
paired t test or ANOVA when multiple experimental groups were compared
simultaneously; P < 0.05.
ACKNOWLEDGMENTS. We thank N. C. Spitzer for critical comments on the
manuscript. We thank F. De Sauvage for the Smo and SmoM2 construct, G. S.
McKnight for PKA constructs, G. J. Barritt for the xTRPC1 antibody, and
T. Meyer for the RFP-PH(PLCδ) construct. This work was supported by an
award to L.N.B. from The Esther A. and Joseph Klingenstein Fund, a grant
to L.N.B. from The Shriners Hospital for Children, and a postdoctoral fellow-
ship to Y.H.B. from The Shriners Hospital for Children.
from embryos treated with agents indicated in the figure. Cyclo, cyclopamine; d, dorsal; Verat, veratridine; VGC block and VGCbl, voltage-gated Na+and Ca2+
channel blockers. (Lower) Graph shows mean ± SEM GABA-immunopositive cells/100 μm of spinal cord. n ≥ 5 stage-34 (45-h postfertilization) embryos per
experimental group; *P < 0.05. (Scale bar, 20 μm.) (B) Model of the molecular mechanisms underlying Shh-induced Ca2+spikes. α, β, γ, subunits of the
heterotrimeric G protein; AC, adenylate cyclase; Cav, voltage-gated Ca2+channels; ER, endoplasmic reticulum; Ptc1, Patched1. Details are given in Discussion.
Ca2+spike activity is necessary for Shh-induced spinal neuron differentiation. (A) (Upper) Immunostaining of transverse sections of the spinal cord
| www.pnas.org/cgi/doi/10.1073/pnas.1018217108 Belgacem and Borodinsky
1. LoTurco JJ, Owens DF, Heath MJ, Davis MB, Kriegstein AR (1995) GABA and glutamate
depolarize cortical progenitor cells and inhibit DNA synthesis. Neuron 15:1287–1298.
2. Komuro H, Rakic P (1996) Intracellular Ca2+ fluctuations modulate the rate of
neuronal migration. Neuron 17:275–285.
3. Manent JB, et al. (2005) A noncanonical release of GABA and glutamate modulates
neuronal migration. J Neurosci 25:4755–4765.
4. Bolteus AJ, Bordey A (2004) GABA release and uptake regulate neuronal precursor
migration in the postnatal subventricular zone. J Neurosci 24:7623–7631.
5. Hanson MG, Landmesser LT (2004) Normal patterns of spontaneous activity are
required for correct motor axon guidance and the expression of specific guidance
molecules. Neuron 43:687–701.
6. Demarque M, Spitzer NC (2010) Activity-dependent expression of Lmx1b regulates
specification of serotonergic neurons modulating swimming behavior. Neuron 67:
7. Dulcis D, Spitzer NC (2008) Illumination controls differentiation of dopamine neurons
regulating behaviour. Nature 456:195–201.
8. Borodinsky LN, et al. (2004) Activity-dependent homeostatic specification of
transmitter expression in embryonic neurons. Nature 429:523–530.
9. Marek KW, Kurtz LM, Spitzer NC (2010) cJun integrates calcium activity and tlx3
expression to regulate neurotransmitter specification. Nat Neurosci 13:944–950.
10. Borodinsky LN, Spitzer NC (2007) Activity-dependent neurotransmitter-receptor
matching at the neuromuscular junction. Proc Natl Acad Sci USA 104:335–340.
11. Catalano SM, Shatz CJ (1998) Activity-dependent cortical target selection by thalamic
axons. Science 281:559–562.
12. Chamberlain CE, Jeong J, Guo C, Allen BL, McMahon AP (2008) Notochord-derived
Shh concentrates in close association with the apically positioned basal body in neural
target cells and forms a dynamic gradient during neural patterning. Development
13. Chen MH, Li YJ, Kawakami T, Xu SM, Chuang PT (2004) Palmitoylation is required for
the production of a soluble multimeric Hedgehog protein complex and long-range
signaling in vertebrates. Genes Dev 18:641–659.
14. Echelard Y, et al. (1993) Sonic hedgehog, a member of a family of putative signaling
molecules, is implicated in the regulation of CNS polarity. Cell 75:1417–1430.
15. Ericson J, Morton S, Kawakami A, Roelink H, Jessell TM (1996) Two critical periods of
Sonic Hedgehog signaling required for the specification of motor neuron identity.
16. Lum L, Beachy PA (2004) The Hedgehog response network: Sensors, switches, and
routers. Science 304:1755–1759.
17. Roelink H, et al. (1995) Floor plate and motor neuron induction by different
concentrations of the amino-terminal cleavage product of sonic hedgehog auto-
proteolysis. Cell 81:445–455.
18. Bourikas D, et al. (2005) Sonic hedgehog guides commissural axons along the
longitudinal axis of the spinal cord. Nat Neurosci 8:297–304.
19. Charron F, Stein E, Jeong J, McMahon AP, Tessier-Lavigne M (2003) The morphogen
sonic hedgehog is an axonal chemoattractant that collaborates with netrin-1 in
midline axon guidance. Cell 113:11–23.
20. Okada A, et al. (2006) Boc is a receptor for sonic hedgehog in the guidance of
commissural axons. Nature 444:369–373.
21. Parra LM, Zou Y (2010) Sonic hedgehog induces response of commissural axons to
Semaphorin repulsion during midline crossing. Nat Neurosci 13:29–35.
22. Yam PT, Langlois SD, Morin S, Charron F (2009) Sonic hedgehog guides axons through
a noncanonical, Src-family-kinase-dependent signaling pathway. Neuron 62:349–362.
23. Corbit KC, et al. (2005) Vertebrate Smoothened functions at the primary cilium.
24. Rohatgi R, Milenkovic L, Scott MP (2007) Patched1 regulates hedgehog signaling at
the primary cilium. Science 317:372–376.
25. Ogden SK, et al. (2008) G protein Galphai functions immediately downstream of
Smoothened in Hedgehog signalling. Nature 456:967–970.
26. Riobo NA, Saucy B, Dilizio C, Manning DR (2006) Activation of heterotrimeric G
proteins by Smoothened. Proc Natl Acad Sci USA 103:12607–12612.
27. DeCamp DL, Thompson TM, de Sauvage FJ, Lerner MR (2000) Smoothened activates
Galphai-mediated signaling in frog melanophores. J Biol Chem 275:26322–26327.
28. Hammerschmidt M, McMahon AP (1998) The effect of pertussis toxin on zebrafish
development: A possible role for inhibitory G-proteins in hedgehog signaling. Dev
29. Low WC, et al. (2008) The decoupling of Smoothened from Galphai proteins has little
effect on Gli3 protein processing and Hedgehog-regulated chick neural tube
patterning. Dev Biol 321:188–196.
30. Hengl T, et al. (2010) Molecular components of signal amplification in olfactory
sensory cilia. Proc Natl Acad Sci USA 107:6052–6057.
31. Whitfield JF (2008) The solitary (primary) cilium—a mechanosensory toggle switch in
bone and cartilage cells. Cell Signal 20:1019–1024.
32. Gu X, Olson EC, Spitzer NC (1994) Spontaneous neuronal calcium spikes and waves
during early differentiation. J Neurosci 14:6325–6335.
33. Zhang J, Rosenthal A, de Sauvage FJ, Shivdasani RA (2001) Downregulation of
Hedgehog signaling is required for organogenesis of the small intestine in Xenopus.
Dev Biol 229:188–202.
34. Holliday J, Adams RJ, Sejnowski TJ, Spitzer NC (1991) Calcium-induced release of
calcium regulates differentiation of cultured spinal neurons. Neuron 7:787–796.
35. Orellana SA, McKnight GS (1992) Mutations in the catalytic subunit of cAMP-
dependent protein kinase result in unregulated biological activity. Proc Natl Acad Sci
36. Rogers KV, Goldman PS, Frizzell RA, McKnight GS (1990) Regulation of Cl- transport in
T84 cell clones expressing a mutant regulatory subunit of cAMP-dependent protein
kinase. Proc Natl Acad Sci USA 87:8975–8979.
37. Boulay G, et al. (1999) Modulation of Ca(2+) entry by polypeptides of the inositol 1,4,
5-trisphosphate receptor (IP3R) that bind transient receptor potential (TRP): Evidence
for roles of TRP and IP3R in store depletion-activated Ca(2+) entry. Proc Natl Acad Sci
38. Bobanovic LK, et al. (1999) Molecular cloning and immunolocalization of a novel
vertebrate trp homologue from Xenopus. Biochem J 340:593–599.
39. Shim S, et al. (2005) XTRPC1-dependent chemotropic guidance of neuronal growth
cones. Nat Neurosci 8:730–735.
40. Wang GX, Poo MM (2005) Requirement of TRPC channels in netrin-1-induced
chemotropic turning of nerve growth cones. Nature 434:898–904.
41. Suh BC, Inoue T, Meyer T, Hille B (2006) Rapid chemically induced changes of PtdIns
(4,5)P2 gate KCNQ ion channels. Science 314:1454–1457.
42. Briscoe J, Pierani A, Jessell TM, Ericson J (2000) A homeodomain protein code specifies
progenitor cell identity and neuronal fate in the ventral neural tube. Cell 101:
43. Tertyshnikova S, Fein A (1998) Inhibition of inositol 1,4,5-trisphosphate-induced Ca2+
release by cAMP-dependent protein kinase in a living cell. Proc Natl Acad Sci USA 95:
44. Ben-Ari Y, Spitzer NC (2010) Phenotypic checkpoints regulate neuronal development.
Trends Neurosci 33:485–492.
45. Pineda RH, Ribera AB (2008) Dorsal-ventral gradient for neuronal plasticity in the
embryonic spinal cord. J Neurosci 28:3824–3834.
46. McLean DL, Fan J, Higashijima S, Hale ME, Fetcho JR (2007) A topographic map of
recruitment in spinal cord. Nature 446:71–75.
47. McLean DL, Fetcho JR (2009) Spinal interneurons differentiate sequentially from
those driving the fastest swimming movements in larval zebrafish to those driving the
slowest ones. J Neurosci 29:13566–13577.
48. Provine RR, Sharma SC, Sandel TT, Hamburger V (1970) Electrical activity in the spinal
cord of the chick embryo, in situ. Proc Natl Acad Sci USA 65:508–515.
49. Breunig JJ, et al. (2008) Primary cilia regulate hippocampal neurogenesis by
mediating sonic hedgehog signaling. Proc Natl Acad Sci USA 105:13127–13132.
50. Han YG, et al. (2008) Hedgehog signaling and primary cilia are required for the
formation of adult neural stem cells. Nat Neurosci 11:277–284.
51. Heo JS, Lee MY, Han HJ (2007) Sonic hedgehog stimulates mouse embryonic stem cell
proliferation by cooperation of Ca2+/protein kinase C and epidermal growth factor
receptor as well as Gli1 activation. Stem Cells 25:3069–3080.
52. Osawa H, et al. (2006) Sonic hedgehog stimulates the proliferation of rat gastric
mucosal cells through ERK activation by elevating intracellular calcium concentration.
Biochem Biophys Res Commun 344:680–687.
Belgacem and BorodinskyPNAS
| March 15, 2011
| vol. 108
| no. 11
Correction for “Sonic hedgehog signaling is decoded by calcium
spike activity in the developing spinal cord,” by Yesser H. Belgacem
and Laura N. Borodinsky, which appeared in issue 11, March 15,
2011, of Proc Natl Acad Sci USA (108:4482–4487; first published
February 28, 2011; 10.1073/pnas.1018217108).
The authors note that the following statement should be
added to the Acknowledgments: “This work was supported by
a grant to L.N.B. from the National Science Foundation.”
| September 13, 2011
| vol. 108
| no. 37www.pnas.org/cgi/doi/10.1073/pnas.1112641108