Cleavage of the Wnt Receptor Ryk
Regulates Neuronal Differentiation
during Cortical Neurogenesis
Jungmook Lyu,1Vicky Yamamoto,1and Wange Lu1,*
1Eli and Edythe Broad Center for Regenerative Medicine and Stem Cell Research, Department of Biochemistry and Molecular Biology,
Keck School of Medicine, University of Southern California, Los Angeles, CA 90033
Ryk is a transmembrane receptor tyrosine kinase
(RTK). It functions as a receptor of Wnt proteins re-
quired for cell-fate determination, axon guidance,
and neurite outgrowth in different organisms; how-
ever, the molecular mechanism of Ryk signaling is
unknown. Here, we show that Ryk is cleaved, permit-
ting the intracellular C-terminal fragment of Ryk to
translocate to the nucleus in response to Wnt3 stim-
ulation. We also show that the cleaved intracellular
domain of Ryk is required for Wnt3-induced neuronal
differentiation in vitro and in vivo. These results dem-
onstrate an unexpected mechanism of signal trans-
duction for Ryk as a Wnt receptor, in which the intra-
cellular domain itself functions as the transducing
molecule to bring extracellular signals from the cell
surface into the nucleus, to regulate neural progeni-
tor cell differentiation.
The embryonic neocortex contains multipotent neural stem and
progenitor cells in the ventricular zone (VZ). During cortical de-
velopment, neural progenitor cells (NPCs) can self-renew, prolif-
erate, and differentiate into astrocytes, oligodendrocytes, and
neurons (Temple, 2001). Immature neurons generated from
NPCs undergo radial migration out of the VZ and form the corti-
cal plate (CP). The immature neurons further undergo maturation
side-out’’ manner, and eventually establish the six distinct neo-
cortical layers (Olson and Walsh, 2002; Kriegstein et al., 2006).
factors and extrinsic cell signaling. Wnt signaling is among the
cell signaling events that play an important role in regulating
the growth and differentiation of NPCs. However, the molecular
mechanisms remain poorly characterized.
Several of the 19 wnt genes identified in mammals are ex-
pressed in the developing cortex (Roelink et al., 1990; Parr et al.,
1993; Grove et al., 1998) and play critical roles in the proliferation
and neuronal differentiation of NPCs (Hirabayashi et al., 2004;
Israsena et al., 2004). wnt genes encode secreted glycoproteins
that activate intracellular signaling, including the b-catenin/TCF
pathway, the planar cell polarity pathway, and the Wnt/Ca2+
pathway. Signaling through the canonical b-catenin/TCF path-
way is mediated by the Frizzled (Fz) receptor and low-density
lipoprotein receptor-related protein(Lrp)families (He etal.,2004;
Kuhl et al., 2000).
Recently, many studies have elucidated the role(s) of Ryk as
a receptor for Wnt ligands (Yoshikawa et al., 2003). Ryk receptor
is required for the neurite outgrowth of dorsal root ganglion neu-
rons induced by Wnt3a (Lu et al., 2004), for the inhibition of ex-
tension of cortical axons by Wnt5a (Keeble et al., 2006), and
for repulsive axon guidance by Wnt3 (Schmitt et al., 2006). Ryk
is an atypical member of the receptor tyrosine kinase (RTK)
family (Halford and Stacker, 2001). The RTK family consists of
59 cell-surface receptors with similar structure and functional
characteristics: a ligand-binding extracellular domain, a trans-
tyrosine kinase activity. Their signaling, upon ligand binding, is
mediated by specific kinase-dependent cascades (Schles-
singer, 2000). The structure of Ryk, with a glycosylated extracel-
lular domain and an intracellular kinase domain, is consistent
with members of the RTK family. However, unlike other RTK
members, the intracellular kinase domain of Ryk contains highly
unusual subdomains and lacks tyrosine kinase activity (Katso
et al., 1999; Hovens et al., 1992), thus leaving unresolved the
mechanism whereby Ryk protein transduces signals from the
cell surface to the nucleus in response to Wnt stimulation.
Here, we report an unexpected molecular mechanism of Ryk-
mediated Wnt signaling involved in the neuronal differentiation
RESULTS AND DISCUSSION
Ryk Protein Is Cleaved, and Ryk ICD Translocates
into the Nucleus during Neuronal Differentiation
embryonic day 14.5 (E14.5) mouse embryos to investigate Ryk’s
function in vivo. Immunostaining of the immature neuronal
marker TUJ1 (bIII-tubulin) (Alexander et al., 1991) showed that
the number of TUJ1-positive cells in the CP, which are the newly
generated neurons derived from NPCs in the VZ (McConnell
et al., 1989; Kriegstein et al., 2006), is reduced in Ryk?/?fore-
brain compared to that of wild-type (Figure 1A; see Figure S1A
available online). Reduction in the number of neurons can be
caused by a defect in NPC proliferation (Heins et al., 2002) or
Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc. 773
neuronal apoptotic cell death (Oppenheim, 1991). However,
no significant difference in the number of proliferative NPCs
(as determined by the BrdU incorporation assay) or apoptotic
neurons (as determined by TUNEL staining) was observed be-
tween Ryk+/+and Ryk?/?mice (data not shown). In addition,
RT-PCR analysis demonstrated that expression of the NPC
markers pax6 and nestin is not changed in Ryk?/?forebrain,
whereas expression of the neuronal markers bIII-tubulin and
tbr1 is reduced in the cortex of Ryk?/?forebrain compared to
that of wild-type forebrain (Figure S1C).
To investigate the expression pattern of Ryk protein, coronal
sections of wild-type forebrains were immunostained by using
an anti-Ryk antibody that recognizes the C terminus of Ryk. In
E14.5 brains, Ryk protein was detected at the CP and at the VZ
(Figure 1B; Figure S1D). The immature neuronal marker TUJ1
was strongly expressed in the CP, but not detected in the VZ,
whereas the NPC marker Nestin was restricted within the VZ
(Figure 1B). Expression of Ryk protein is consistent with ryk
gene expression, which we demonstrated by b-gal immuno-
staining in Ryk+/?brain in which the ryk gene is replaced with
the LacZ gene (Figure S1E). Interestingly, higher magnification
of the CP and the VZ (Figure 1B) showed that some Ryk protein
is detected in the nucleus of TUJ1-positive cells, whereas in
Nestin-positive cells Ryk is mainly localized in the membrane.
Quantification of cells exhibiting a nuclear Ryk signal revealed
that nuclear Ryk was detected more frequently in TUJ1-positive
cells than in Nestin-positive cells (Figure 1C). This suggests that
there is a link between the nuclear localization of Ryk and neuro-
Development of the cerebral cortex is initiated by progenitor
cells in the E11.5 mouse neocortex (Lopez-Bendito and Molnar,
2003), and we found that Ryk protein is expressed in the cortex
of the E11.5 mouse forebrain (unpublished data). To examine if
opment and whether the nuclear Ryk protein is a cleavage prod-
uct, we performed western blot on the cell lysates of neocortices
of E11.5, E14.5, and E18.5 mouse embryos. A C-terminal frag-
ment (CTF) of Ryk of ?42 kDa was detected in the E14.5 mouse
cortex (Figure 1D). Moreover, the abundance of this fragment in-
creases with development of the cortex, whereas full-length Ryk
decreases. To examine the cleaved CTF at the cellular level, we
generated a Ryk construct whose C-terminal end is tagged with
a myc epitope, and we transiently transfected HEK293T cells
with the construct. Western blot analysis of whole-cell lysate
by using an anti-myc antibody revealed not only full-length
Ryk, but also a smaller, cleaved Ryk fragment of ?42 kDa
Localized in the Nucleus In Vivo
Ryk?/?mice. Representative immunostaining of
TUJ1 (red) in coronal sections from forebrain of
E14.5 Ryk+/+and Ryk?/?mouse embryos, coun-
terstained with Hoechst 33258. TUJ1 expression
isreduced inthecorticalplate (CP)ofRyk?/?com-
paredwiththat ofRyk+/+mouse cortex. Scale bars
are 50 mm (left) and 10 mm (right).
(B) Expression and localization of Ryk in the E14.5
Ryk+/+mouse forebrain cortex. Ryk proteins are
expressed in both the CP and the ventricular
zone (VZ). Ryk is detectable particularly in the nu-
clei of TUJ1 (green)-positive cells at the region of
the CP and at the membrane of Nestin (green)-
positive cells of the VZ. The nuclear signal is re-
vealed by Hoechst staining. The arrowhead and
arrow indicate the fluorescent signal for Ryk in
the nuclear and membrane, respectively. Scale
bars are 50 mm (left) and 10 mm (right).
(C) Quantitative analysis of nuclear-localized Ryk
in TUJ1-positive cells at the CP and in Nestin-
positive cells at the VZ. Error bars represent the
mean ± SD of triplicate tests.
(D) Western blot analysis of Ryk expression in the
Ryk+/+mouse forebrain cortex at the indicated de-
velopmental stages. Ryk antibody recognizes the
C terminus of the protein. GSK3b expression
was used as a loading control.
(E) Detection of the Ryk cleavage product in the
cytosolic fraction. cDNA encoding wild-type Ryk
with a myc tag at the C terminus was transfected
into HEK293T cells. The whole-cell lysate, mem-
brane, and cytosolic extracts were subjected to
immunoprecipitation, followed by western blot
with an anti-myc antibody. Actin and E-cadherin
confirmed the purity of the cytosolic and mem-
branefractions, respectively.Theasterisk denotes
the heavy chain of the antibody.
1. RykProtein Is Cleavedand
Wnt3/Ryk Signaling in Cortical Neurogenesis
774 Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc.
(Figure 1E, lane 1). Subsequent western blot analysis of subcel-
in the membrane fraction and that the CTF band was localized
exclusively in the cytoplasmic fraction (Figure 1E, lanes 2 and
3). Therefore, we refer to the latter as the Ryk ICD fragment
Cleavage of Ryk Is Important for Neuronal
Differentiation of NPCs
To further investigate the role of cleavage of Ryk in neuronal dif-
ferentiation, we employed an in vitro culture system using NPCs,
which have the ability to differentiate into astrocytes, oligoden-
drocytes, or neurons (Sauvageot and Stiles, 2002; Bani-
Yaghoub et al., 2006). When Ryk+/+and Ryk?/?NPCs were iso-
lated from the neocortex of E11.5 mouse forebrain and cultured
under differentiation conditions, the numbers of TUJ1-positive
cells and MAP2-positive (a marker of mature neurons) cells
were significantly reduced in cultures from Ryk?/?neocortex
compared to cultures from Ryk+/+neocortex, confirming that
the ryk gene may be required for neuronal differentiation
(Figure 2A). As previously demonstrated, the amount of Ryk
cleavage product increases with neuronal development. There-
fore, we sought to test if Ryk cleavage is also regulated during
neuronal differentiation in vitro. When NPCs were cultured under
differentiation conditions without basic fibroblast growth factor
(FGF), Ryk ICD was detected; however, no Ryk ICD was de-
tected in cultures under proliferation conditions in the presence
of basic FGF (Figure 2B). Furthermore, we could detect Ryk ex-
pression in the nucleus and cytoplasm of TUJ1-positive neurons
by immunostaining with fluorescence microscopy (Figure 2C),
Western blot analysis of subcellular extracts from differentiating
NPCs further confirmed that Ryk is cleaved and that Ryk ICD is
localized in the nucleus in differentiating NPCs (Figure 2C, right).
These results further confirm that cleavage of Ryk occurs during
The size of Ryk ICD suggests that Ryk is cleaved in its trans-
membrane or juxtamembrane region. To map the potential
cleavage site, we generated two constructs: Ryk RC, a chimeric
construct in which the transmembrane region of Ryk was re-
placed with that of the EGF receptor; and another deletion mu-
tant of Ryk (Ryk DS/T), in which the juxtamembrane region
including the serine/threonine-rich domain was deleted. Ryk
ICD was used as a control to show the cleaved ICD size. Both
wild-type Ryk and Ryk DS/T were cleaved when expressed in
HEK293T cells (Figure S2). However, no cleavage ICD product
was detected in cells expressing Ryk RC (Figure S2). These re-
sults suggest that the cleavage site of Ryk lies within the trans-
extracts confirmed that Ryk RC is resistant to cleavage. In cells
transfected with the Ryk RC construct, Ryk RC was not cleaved
and was localized only in the membrane (Figure 2D). In cells
transfected with a DNA construct encoding wild-type Ryk, the
full-length Ryk protein localized only in the membrane, and the
cleaved C terminus localized only in the cytoplasm (Figure 2D).
However, in cells with the transfected Ryk ICD construct, Ryk
ICD was localized in both the nucleus and cytoplasm
chemistry of Cos-7 cells transfected with different Ryk con-
structs. The Ryk RC mutant localized at the membrane, whereas
Ryk ICD primarily localized in the nucleus (Figure 2E). Thesedata
are consistent with the following model of Ryk-mediated signal-
ing: (1) Ryk is cleaved in its transmembrane domain; (2) the ICD
fragment of Ryk is released to the cytoplasm and then translo-
cated into the nucleus.
To examine if cleavage ofRyk isimportantfor Ryk’sfunctionin
neuronal differentiation, we introduced genes encoding wild-
type Ryk, Ryk RC (Figure S2), Ryk ICD, or nuclear localization
signal-fused Ryk ICD (Ryk NLS-ICD) into Ryk?/?NPCs by using
lentivirus. Expression of wild-type Ryk in Ryk?/?NPCs rescued
the mutant phenotype, as the percentage of differentiating neu-
rons was significantly increased (from 2.1% ± 0.74% to 16.8% ±
1.24% for TUJ1-positive cells per total cells, and from 4.1% ±
1.4% to 15.5% ± 2.25% for MAP2-positive cells per total cells)
compared to control Ryk?/?NPCs (Figure 2F), and the number
of differentiated neurons in these cultures was comparable to
that of Ryk+/+NPCs (10.06% ± 2.21% and 6.96% ± 1.5% for
TUJ1- and MAP2-positive cell per total cells, respectively). Ryk
ICD in Ryk?/?NPCs also promoted neuronal differentiation
(9.23% ± 1.30% and 8.53% ± 1.58% for TUJ1- and MAP2-
positive cells per total cells, respectively), although it did not res-
cue the phenotype as completely as wild-type Ryk. Interestingly,
Ryk NLS-ICD in Ryk?/?NPCs further increased neuronal differ-
entiation (21.79% ± 1.89% and 17.36% ± 2.13% for TUJ1-
and MAP2-positive cellsper total cells, respectively), whencom-
pared with wild-type Ryk expression. In contrast, the Ryk RC
gene failed to rescue the defect in neuronal differentiation in
Ryk?/?cells. Taken together, results from these experiments
demonstrate that cleavage of Ryk and the nuclear localization
of cleaved ICD are necessary for the protein’s proper functioning
in neuronal differentiation.
Ryk Protein Is Cleaved by g-Secretase
To further investigate how Ryk protein is cleaved, we tested
a few candidate proteases. Other transmembrane proteins,
such as Notch and ErbB4, undergo proteolytic processing that
allows for shedding of the extracellular domain by metallopro-
tease, followed by intramembrane cleavage by g-secretase/pre-
senilin, and then release of the ICD into the cytoplasm (Ni et al.,
2001; Kopan and Ilagan, 2004). To test if Ryk protein is cleaved
by g-secretase/presenilin, HEK293T cells and differentiating
NPCs were transfected with myc-tagged Ryk and treated with
g-secretase inhibitors. In DAPT-treated HEK293T cells, the Ryk
ICD level in the cytoplasm was decreased (Figure 3, lanes 7
and 8), whereas full-length Ryk was increased (Figure 3, lanes
3 and 4), suggesting that Ryk cleavage is dependent on g-secre-
tase. Inhibition of g-secretase also reveals a truncated CTF of
could be an intermediate product resulting from ectodomain
shedding, although the molecular mechanism is unclear. Treat-
ment of differentiating NPCs expressing wild-type Ryk with the
g-secretase inhibitor DAPT or L685,458 also reduces the Ryk
ICD level, whereas it increases the Ryk CTF level (Figure S3A).
In differentiating NPCs, the cleavage of endogenous Ryk is
also inhibited by DAPT (Figure S3B). Presenilin is one of the im-
portant components in the g-secretase complex. In presenilin1-
deficient cells, the Ryk ICD level is decreased, whereas levels of
full-length and Ryk CTF are both increased (Figure 3B). There-
fore, these results suggest that Ryk undergoes a two-step
Wnt3/Ryk Signaling in Cortical Neurogenesis
Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc. 775
Figure 2. Cleavage of Ryk Is Required for Neuronal Differentiation
(A)NPCsderivedfromRyk+/+orRyk?/?mousecortexwereculturedunderdifferentiatingconditions, andthensubjectedtoimmunostaining forneuronalmarkers.
There are fewer TUJ1- and MAP2-positive cells in Ryk?/?mice. Scale bars are 20 mm.
(B) Cleavage of Ryk protein in NPCs cultured under undifferentiation (UD) or differentiation (D) conditions. Ryk ICD protein levels increase under D conditions.
(C) Localization of Ryk protein in the nucleus of TUJ1-positive cells differentiated from NPCs is demonstrated by immunocytochemistry (left panels) and by west-
and membrane fractions, respectively. The scale bar is 10 mm.
(D) Western blot analysis of subcellular extracts of HEK293T cells confirms the presence of Ryk RC only in the membrane fraction and Ryk ICD in both the
to determine the expression of Ryk, lamin A/C, actin, and E-cadherin. Lamin A/C, actin, and E-cadherin confirmed the purity of the nuclear, cytosolic, and
membrane fractions, respectively.
(E) Expression of Ryk and its mutants in Cos-7 cells was determined by immunostaining with an anti-myc antibody. Nuclei were visualized by Hoechst staining.
Ryk ICD is primarily localized in the nucleus, whereas wild-type Ryk and Ryk RC are localized in the membrane. The scale bar is 20 mm.
wild-type Ryk, Ryk RC, Ryk ICD, or Ryk NLSICD. The cells were cultured under D conditions prior to immunostaining for TUJ1 or MAP2. The percentage of dif-
ferentiating cells was evaluated by quantifying the number of TUJ1- or MAP2-positive cells per total cells. Hoechst dye was used for counterstaining. Each error
bar represents the mean ± SD of four independent experiments; each assay was performed in duplicate. The scale bar is 20 mm.
Wnt3/Ryk Signaling in Cortical Neurogenesis
776 Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc.
cleavage that is mediated by an unknown protease and g-secre-
tase/presenilin, and that cleavage mediated by g-secretase/
presenilin leads to production of Ryk ICD.
Wnt3 Stimulates the Nuclear Localization of Ryk ICD
The extracellular domain of Ryk contains a conserved Wnt-bind-
ing sequence (Patthy, 2000) that has been shown to bind Wnt1,
Wnt3, Wnt3a, and Wnt5a (Figure S4A) (Lu et al., 2004; Keeble
et al., 2006; Liu et al., 2005; Schmitt et al., 2006). Ryk plays
very important roles in axon guidance and neurite outgrowth.
The ICD of Ryk is essential for axon guidance by Ryk-mediated
Wnt signaling in Drosophila and the chick (Yoshikawa et al.,
2003; Liu et al., 2005). Although these data suggest an important
role for the ICD of Ryk in Wnt signal transduction, the question
remains as to how Wnt proteins regulate Ryk intracellular signal-
ing. Do they regulate the production and translocation of Ryk
struct expressing Ryk-GFP together with Wnt3 or empty vector.
Wnt3 stimulated the nuclear localization of Ryk-GFP (Figure 3C).
The number of nuclear Ryk-GFP-positive cells was increased by
6-fold upon Wnt3 expression compared to the control
(Figure 3C, bottom). In contrast, Ryk-GFP nuclear localization
was decreased in DAPT-treated cells even in the presence of
Wnt3 stimulation. Using subcellular extracts from HEK293T
cells, we also confirmed that the level of Ryk ICD in the nucleus,
but not other forms of Ryk protein, was increased by stimulation
of Wnt3-conditioned medium (CM) (Figure 3D). The level of total
Ryk ICD was not increased by Wnt3, nor was the level of full-
length Ryk decreased (Figure 3E). Wnt3 does not seem to have
any effect on the amount of Ryk CTF either (data not shown).
These results indicate that Wnt3, when applied from outside of
the cells, does not regulate the cleavage of Ryk; instead, it con-
trols thenuclear translocation ofRyk ICD.Wepreviously showed
that Ryk, acting as a coreceptor for Wnt1 and Wnt3a, can en-
hance Tcf-driven gene transcription (Lu et al., 2004). However,
that mechanism does not appear to involve Ryk ICD because
Ryk ICD neither activated a Tcf-luciferase reporter nor stabilized
cytoplasmic b-catenin (Figures S4C and S4D), suggesting that
Ryk ICD is mediating the b-catenin-independent Wnt3 signaling
Ryk Cleavage Is Required for Wnt3-Induced Neuronal
From our rescue study, introducing Ryk ICD alone was not suffi-
cient to completely rescue the defect in neuronal differentiation
of Ryk?/?NPCs, whereas full-length wild-type Ryk could fully
rescue. This suggests that binding of ligand to the extracellular
domain of Ryk may be required for full Ryk activity (Figure 2F).
Figure 3. Ryk Is Cleaved by g-Secretase,
and Wnt3 Induces Nuclear Localization of
(A) g-secretase inhibitor inhibits the cleavage of
Ryk. HEK293T cells expressing myc-tagged Ryk
were treated with 0.5 mM DAPT, an inhibitor of g-
secretase, or DMSO as a control. The cell lysates
were fractionated and then subjected to western
blot with an anti-myc antibody. Detection of E-
cadherin and actin at different fractions confirmed
(B) Cleavage of Ryk is reduced in presenilin1-
deficient cells. DNA constructs for wild-type Ryk
with a myc tag together with GFP plasmid were
transfected into presenilin1+/+or presenilin1?/?
cells. The levels of full-length Ryk, Ryk CTF, and
Ryk ICD were determined by western blot.
cDNA encoding a Ryk-GFP fusion protein was
transfected into Cos-7 cells together with Wnt3 or
were treated with the indicated chemicals for 6 hr.
Quantification of nuclear localization of Ryk was
evaluated by the percentage of cells with anuclear
represent the mean ± SD of triplicate tests.
tracts from myc-tagged Ryk-expressing HEK293T
dium (CM) and with or without 0.5 mM DAPT. Wnt3
CM increased the level of nuclear Ryk ICD. Treat-
ment with DAPT inhibited the increase of nuclear
Ryk ICD induced by Wnt3. Antibodies against
of the nuclear and the cytosolic fractions.
(E) Western blot analysis of the whole-cell lysate
shows that Wnt3 CM did not affect the cleavage
Wnt3/Ryk Signaling in Cortical Neurogenesis
Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc. 777
Interestingly, although the expression of Ryk mRNA remains
constant during development, expression of Wnt3 mRNA in-
creases in the developing cortex (Figure 4A) (Roelink et al.,
1990) and in NPCs cultured under differentiation conditions
(Figure 4B). Since expression of wnt3 correlates with the cleav-
age of Ryk, we hypothesized that Wnt3 is a key factor in Ryk sig-
naling during neuronal differentiation of NPCs.
To test this hypothesis, we infected NPCs with lentivirus ex-
pressing Wnt3 and then cultured the cells under differentiation
conditions. Ryk expression in NPC cultures under these
Figure 4. Cleavage of Ryk and Ryk ICD Is Required for Wnt3-Induced Neuronal Differentiation
(B) NPCs cultured under UD or D conditions. Expression of gapdh was used as an internal control. wnt3 expression correlates with cleavage of Ryk protein in the
developing cortex and during neuronal differentiation from NPCs. Each error bar represents the mean ± SD of four independent experiments.
(C) Wnt3 stimulates neuronal differentiation, and its promotion of differentiation requires the ICD of Ryk. NPCs infected with lentivirus expressing Wnt3, wild-type
Ryk, Ryk DICD, or both Wnt3 and wild-type Ryk or Ryk DICD were cultured under D conditions. The cells were analyzed by immunostaining for TUJ1 (red) and
MAP2 (data not shown). Hoechst dye was used for counterstaining. The percentage of TUJ1- or MAP2-positive cells per total cells is shown in the bottom panel.
The scale bar is 40 mm.
(D) Ryk DICD inhibits Wnt3-induced nuclear localization of Ryk ICD. Ryk?/?NPCs stably expressing wild-type Ryk-myc together with Ryk DICD or control vector
to western blot with anti-myc, lamin A/C, and actin antibodies. Treatment with Wnt3 CM induces the increase of the nuclear Ryk ICD level. Ryk DICD expression
inhibits the increase of the nuclear Ryk ICD level. Lamin A/C and actin confirmed the purity of the nuclear and cytosolic fractions, respectively.
(E) Ryk RC inhibits the neuronal differentiation induced by Wnt3, whereas Ryk ICD can further stimulate neuronal differentiation. NPCs expressing Wnt3 or empty
vector were infected with lentivirus expressing wild-type Ryk, Ryk RC, Ryk ICD, Ryk NLSICD, or control vector. The percentages of TUJ1- and MAP2-positive
cells were determined as described above.
Each error bar represents the mean ± SD of (C) three or (E) five independent experiments; each assay was performed in triplicate.
Wnt3/Ryk Signaling in Cortical Neurogenesis
778 Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc.
conditions was detected mostly in oligodendrocytes and neu-
rons (Figure S5). Interestingly, Wnt3 stimulated the generation
of TUJ1-positive neurons, but not O4-positive oligodendrocytes
or GFAP-positive astrocytes (Figure S6). Furthermore, Wnt3
stimulation clearly induced more nuclear localization of Ryk in
neurons compared to control cells (Figure S7). In adult hippo-
campal progenitor cells, Wnt3/b-catenin signaling can increase
the population of neurons by inducing the proliferation of neuro-
nally committed precursor cells (Lie etal., 2005).Given thatthere
was no significant difference in the numbers of the BrdU-incor-
porated cells between wild-type and Ryk knockout cell cultures
(data not shown), Ryk signaling does not affect the proliferation
of NPCs. Therefore, our finding emphasizes that Ryk receptor
is an essential component of Wnt3 signaling in the neuronal dif-
ferentiation process. To investigate if Wnt3-induced neuronal
differentiation requires cleavage of Ryk, we first examined
whether Ryk-mediated Wnt signaling is necessary for neuronal
differentiation. NPCs were seeded and then infected with lentivi-
rus expressing dominant-negative Ryk (DICD) (Figure S4B)
(Schmitt et al., 2006), which only has the extracellular and trans-
membrane domains, to inhibit the binding of Wnt3 to endoge-
nous Ryk. After incubating for 4 days under differentiation condi-
tions, neuronal differentiation was quantified by determining the
percentage of TUJ1- and MAP2-positive neurons. Wnt3 in-
creased the percentage of TUJ1- and MAP2-positive cells by
2-fold comparedto thecontrol(Figure4C). However, areduction
of Wnt3-induced neuronal differentiation was observed in NPCs
expressing Ryk DICD. This reduction might be caused by inhibi-
tion of Wnt3-induced nuclear localization of Ryk ICD. Treatment
with Wnt3 CM in Ryk?/?NPCs stably expressing wild-type Ryk-
myc, which mimics wild-type NPCs, increased the nuclear Ryk
ICD level and decreased the cytosolic Ryk ICD level, whereas
thetotal full-length Rykand RykICD levels showednosignificant
difference between the cells treated with control medium and
Wnt3 CM (Figure 4D). Interestingly, expression of Ryk DICD re-
duced the nuclear Ryk ICD level, whereas the cytoplasmic Ryk
ICD level increased (Figure 4D). This result demonstrates that
Ryk DICD inhibits nuclear localization of Ryk ICD induced by
Wnt3. To determine whether Ryk cleavage and its nuclear trans-
location are involved in the neuronal differentiation by Ryk-medi-
ated Wnt3 signaling, we next introduced Ryk RC or Ryk ICD into
NPCs cultured under differentiation conditions. As we described
above, Ryk ICD is translocated into the nucleus when it is over-
expressed (Figures 2D and 2E). Overexpression of Ryk ICD or
cells among the NPCs (Figure 4E; Figure S8), suggesting that the
nuclear translocation of Ryk ICD induces neuronal differentia-
tion. In contrast, in NPCs expressing both Ryk RC and Wnt3,
the percentages of differentiated neurons were significantly re-
duced, as were the numbers of TUJ1- and MAP2-positive cells,
by 61.8% ± 4.4% and 82.2% ± 4.9%, respectively, compared to
NPCs that received the control vector and Wnt3 (Figure 4E).
These data provide strong evidence that Ryk cleavage and
subsequent nuclear localization are required to mediate Wnt3-
induced neuronal differentiation.
Ryk protein is a receptor for Wnt proteins. Our discovery has
revealed an unexpected mechanism of Wnt signaling in which
Ryk undergoes specific proteolytic cleavage to fulfill its role in
signal transduction. We demonstrated here that cleavage of
Ryk can be regulated by g-secretase, and that this cleavage is
critical for Ryk’s role in neuronal differentiation of cortical
NPCs. It is clear from our data that nuclear signaling of Ryk plays
a critical role in regulating the neuronal differentiation of NPCs. In
our unpublished microarray analysis, expression of Ryk ICD
fused with a nuclear localization signal (NLS) was found to regu-
late expression of some of the same downstream target genes
as wild-type Ryk, when compared with Ryk?/?cells (unpub-
lished data). This strongly supports a model in which cleaved
Ryk ICD can immediately impact transcriptional activity. Future
research will focus on identifying the target genes to understand
the Ryk-mediated Wnt3 signaling cascade that regulates the ex-
pression of key genes in cortical neurogenesis.
Neural Progenitor Cell Culture
Neocortices were dissected from embryonic day 11.5 (E11.5) brain of Ryk+/+
and Ryk?/?mice in Hank’s balanced salt solution (HBSS; Cellgro). Embryos
of Ryk?/?mice were generated as described (Halford et al., 2000) and were
kindly provided by Dr. Steven A. Stacker (Ludwig Institute of Royal Melbourne
Hospital). Neocorticesweremechanicallydissociated intosinglecellsbyusing
a flame-polished Pasture pipet. Dissociated cells (1 3 106cells/dish) were
seeded onto polyornithine- (15 mg/ml; Sigma) and fibronectin (2 mg/ml; Invitro-
gen)-coated 10 cm dishes in DMEM/F12 medium containing B27 supplement
(GIBCO-BRL) and were cultured in the presence of fibroblast growth factor 2
(FGF2, 20 ng/ml) to expand the neural progenitor cell (NPC) population (prolif-
eration condition; ‘‘UD’’). To induce the differentiation of NPCs, cells were
seeded and further cultured in the absence of FGF2 (differentiation condition;
‘‘D’’). Expression of Wnt3, wild-type Ryk, or Ryk mutants in NPCs was trans-
duced by lentiviral infection (Supplemental Data). For western blot and
qPCR, 1.5 3 106cells were seeded onto coated 10 cm dishes and further
ing, cells were plated onto polyornithine- and fibronectin-coated coverslips
(5 3 104cells/well in 24-well dishes).
Immunoprecipitation, Western Blotting, and Immunostaining
For immunoprecipitation, 500 mg of protein samples were incubated with
a specific antibody for 2 hr at 4?C and then with Protein A/G agarose beads
(Pierce) overnight. The beads were precipitated by centrifugation and washed
extensively. The immune complexes were eluted by SDS sample buffer. Each
protein sample was separated by 10% SDS-PAGE and transferred onto PVDF
membranes. After blocking, the blots were incubated with a primary antibody
as indicated and subsequently with a peroxidase-conjugated secondary
antibody. The bound secondary antibody was then detected by enhanced
chemiluminescence (ECL) reagent (Santa Cruz Biotechnology). To detect the
full-length form of Ryk in this study, lysates were subjected to immunoprecip-
itation, followed by western blot. For immunostaining, cells growing on cover-
slips were fixed with 4% paraformaldehyde for 15 min and permeabilized with
0.2% Triton X-100. Fixed cells were incubated with blocking solution contain-
ing 2% BSA for 1 hr and were incubated with antibodiesovernight at 4?C. After
rinses with PBS, cells were incubated with secondary antibodies at room tem-
perature for 1 hr and were counterstained with Hoechst dye. Images were ob-
tained by using a fluorescence microscope with an AxioCam camera (Zeiss) or
confocal microscope (LSM5 PASCAL, Zeiss). The percentage of antibody-
labeled cells was evaluated by quantifying a minimum of 1000 cells in 10 ran-
domly chosen microscopic fields, and the values were obtained from at least 3
Quantitative Analysis of Nuclear Ryk
To determine the nuclear localization of Ryk in differentiated and undifferenti-
ated neuronal cells in vitro and in vivo, mouse brain sections or cultured neu-
ronal cells were immunolabeled with anti-Ryk and TUJ1 for the detection of
differentiating neurons or Nestin for detection of undifferentiated neural pro-
genitors. Hoechst 33258 was used for nuclear staining. Fluorescent images
were obtained with a fluorescent microscope and an AxioCam camera (Zeiss).
Wnt3/Ryk Signaling in Cortical Neurogenesis
Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc. 779
Within each nuclear area, fluorescent intensity of the secondary antibody for
anti-Ryk antibody was measured by using Axio Imager software. A minimum
of 100 cells randomly chosen from TUJ1-positive cells at the CP and Nestin-
positive cells at the VZ, or from differentiated cells (TUJ1-positive cells) cul-
tured in vitro, were examined. The values obtained from at least three samples
were averaged and presented as means ± SD.
Generation of Ryk Antibody
To detect the Ryk C-terminal fragment (CTF), peptides corresponding to res-
idues 314–562 of mouse Ryk in the form of a GST fusion were immunized in
rabbits, and the collected antiserum was affinity purified. The specificity to
the C terminus was confirmed by western blot analysis with lysates from
both wild-type and Ryk knockout mouse brain, NPCs, embryonic stem cells,
and HEK293T cells transfected with Ryk plasmids (Figure S9).
Supplemental References, and nine figures are available at http://www.
We thank Michael Stallcup and Martin Pera for critically reading this manu-
script. We are grateful to Steven Stacker for providing the Ryk knockout
mouse. This work was partially supported by pilot grant IRG-58-007048
from the American Cancer Society. J.L. was supported by an award from
the Korea Research Foundation (KRF-2006-214-C00076) and by a fellowship
from the California Institute of Regenerative Medicine (CIRM). V.Y. was
supported by a fellowship from the CIRM.
Received: May 6, 2008
Revised: September 10, 2008
Accepted: October 8, 2008
Published: November 10, 2008
Alexander, J.E., Hunt, D.F., Lee, M.K., Shabanowitz, J., Michel, H., Berlin,
S.C., MacDonald, T.L., Sundberg, R.J., Rebhun, L.I., and Frankfurter, A.
(1991). Characterization of posttranslational modifications in neuron-specific
class III b-tubulin by mass spectrometry. Proc. Natl. Acad. Sci. USA 88,
Bani-Yaghoub, M., Tremblay, R.G., Lei, J.X., Zhang, D., Zurakowski, B.,
Sandhu, J.K., Smith, B., Ribecco-Lutkiewicz, M., Kennedy, J., Walker, P.R.,
and Sikorska, M. (2006). Role of Sox2 in the development of the mouse neo-
cortex. Dev. Biol. 295, 52–66.
Grove, E.A., Tole, S., Limon, J., Yip, L., and Ragsdale, C.W. (1998). The hem
of the embryonic cerebral cortex is defined by the expression of multiple
Wnt genes and is compromised in Gli3-deficient mice. Development 125,
Halford, M.M., and Stacker, S.A. (2001). Revelations of the RYK receptor. Bio-
essays 23, 34–45.
Halford, M.M., Armes, J., Buchert, M., Meskenaite, V., Grail, D., Hibbs, M.L.,
Wilks, A.F., Farlie, P.G., Newgreen, D.F., Hovens, C.M., et al. (2000). Ryk-
deficient mice exhibit craniofacial defects associated with perturbed Eph
receptor crosstalk. Nat. Genet. 25, 414–418.
teins 5 and 6 in Wnt/b-catenin signaling: arrows point the way. Development
Heins, N., Malatesta, P., Cecconi, F., Nakafuku, M., Tucker, K.L., Hack, M.A.,
Chapouton, P., Barde, Y.A., and Gotz, M. (2002). Glial cells generate neurons:
the role of the transcription factor Pax6. Nat. Neurosci. 5, 308–315.
Hirabayashi, Y., Itoh, Y., Tabata, H., Nakajima, K., Akiyama, T., Masuyama, N.,
and Gotoh, Y. (2004). The Wnt/b-catenin pathway directs neuronal differentia-
tion of cortical neural precursor cells. Development 131, 2791–2801.
Hovens, C.M., Stacker, S.A., Andres, A.C., Harpur, A.G., Ziemieki, A., and
Wilks, A.F. (1992). Ryk, a receptor tyrosine kinase-related molecule with un-
usual kinase domain motifs. Proc. Natl. Acad. Sci. USA 89, 11818–11822.
Israsena, N., Hu, M., Fu, F., Kan, L., and Kessler, J.A. (2004). The presence of
FGF2 signaling determines whether b-catenin exerts effects on proliferation or
neuronal differentiation of neural stem cells. Dev. Biol. 268, 220–231.
Katso, R.M., Russell, R.B., and Ganesan, T.S. (1999). Functional analysis of
H-Ryk, an atypical member of the receptor tyrosine kinase family. Mol. Cell.
Biol. 19, 6427–6440.
Keeble, T.R., Halford, M.M., Seaman, C., Kee, N., Macheda, M., Anderson,
R.B.,Stacker,S.A., and Cooper, H.M. (2006).The WntreceptorRykisrequired
for Wnt5a-mediated axon guidance on the contralateral side of the corpus cal-
losum. J. Neurosci. 26, 5840–5848.
Kopan, R., and Ilagan, M.X. (2004). g-Secretase: proteasome of the mem-
brane? Nat. Rev. Mol. Cell Biol. 5, 499–504.
Kriegstein, A., Noctor, S., and Martinez-Cerdeno, V. (2006). Patterns of neural
stem and progenitor cell division may underlie evolutionary cortical expansion.
Nat. Rev. Neurosci. 7, 883–890.
Kuhl, M., Sheldahl, L.C., Park, M., Miller, J.R., and Moon, R.T. (2000). The Wnt/
Ca2+pathway: a new vertebrate Wnt signaling pathway takes shape. Trends
Genet. 16, 279–283.
Lie, D.C., Colamarino, S.A., Song, H.J., Desire, L., Mira, H., Consiglio, A., Lein,
E.S., Jessberger, S., Lansford, H., Dearie, A.R., and Gage, F.H. (2005). Wnt
signaling regulates adult hippocampal neurogenesis. Nature 437, 1370–1375.
Liu, Y., Shi, J., Wu, C., Zhang, H., Zheng, Z., Geiger, T.L., Nuovo, G.L., Liu, Y.,
and Zheng, P. (2005). Ryk-mediated Wnt repulsion regulates posterior-
directed growth of corticospinal tract. Nat. Neurosci. 8, 1151–1159.
Lopez-Bendito, G., and Molnar, Z. (2003). Thalamocortical development: how
are we going to get there? Nat. Rev. Neurosci. 4, 276–289.
Lu, W., Yamamoto, V., Ortega, B., and Baltimore, D. (2004). Mammalian Ryk
is a Wnt coreceptor required for stimulation of neurite outgrowth. Cell 119,
McConnell, S.K., Ghosh, A., and Shatz, C.J. (1989). Subplate neurons pioneer
the first axon pathway from the cerebral cortex. Science 245, 978–982.
Ni, C.Y., Murphy, M.P., Golde, T.E., and Carpenter, G. (2001). g-Secretase
cleavage and nuclear localization of ErbB-4 receptor tyrosine kinase. Science
Olson, E.C., and Walsh, C.A. (2002). Smooth, rough and upside-down neocor-
tical development. Curr. Opin. Genet. Dev. 12, 320–327.
Oppenheim, R.W. (1991). Cell death during development of the nervous sys-
tem. Annu. Rev. Neurosci. 14, 453–501.
Parr, B.A., Shea, M.J., Vassileva, G., and McMahon, A.P. (1993). Mouse Wnt
genes exhibit discrete domains of expression in the early embryonic CNS
and limb buds. Development 119, 247–261.
Patthy, L. (2000). The WIF module. Trends Biochem. Sci. 25, 12–13.
Roelink, H., Wagnaar, E., Lopes da Silva, S., and Nusse, R. (1990). Wnt-3,
a gene activated by proviral insertion in mouse mammary tumors, is homolo-
gous to int-1/Wnt-1 and is normally expressed in mouse embryos and adult
brain. Proc. Natl. Acad. Sci. USA 87, 4519–4523.
Sauvageot, C.M., and Stiles, C.D. (2002). Molecular mechanisms controlling
cortical gliogenesis. Curr. Opin. Neurobiol. 12, 244–249.
Schlessinger, J. (2000). Cell signaling by receptor tyrosine kinases. Cell 103,
Schmitt, A.M., Shi, J., Wolf, A.M., Lu, C.C., King, L.A., and Zou, Y. (2006).
Wnt-Ryk signaling mediates medial-lateral retinotectal topographic mapping.
Nature 439, 31–37.
Yoshikawa, S., McKinnon, R.D., Kokel, M., and Thomas, J.B. (2003). Wnt-
mediated axon guidance via the Drosophila Derailed receptor. Nature 422,
Wnt3/Ryk Signaling in Cortical Neurogenesis
780 Developmental Cell 15, 773–780, November 11, 2008 ª2008 Elsevier Inc.