Id sustains Hes1 expression to inhibit precocious neurogenesis by releasing negative autoregulation of Hes1.

Abstract

Negative bHLH transcription factor Hes1 can inhibit neural stem cells (NSCs) from precocious neurogenesis through repressing proneural gene expression; therefore, sustenance of Hes1 expression is crucial for NSC pool maintenance. Here we find that Ids, the dominant-negative regulators of proneural proteins, are expressed prior to proneural genes and share an overlapping expression pattern with Hes1 in the early neural tube of chick embryos. Overexpression of Id2 in the chick hindbrain upregulates Hes1 expression and inhibits proneural gene expression and neuronal differentiation. By contrast, Hes1 expression decreases, proneural gene expression expands, and neurogenesis occurs precociously in Id1;Id3 double knockout mice and in Id1-3 RNAi-electroporated chick embryos. Mechanistic studies show that Id proteins interact directly with Hes1 and release the negative feedback autoregulation of Hes1 without interfering with its ability to affect other target genes. These results indicate that Id proteins participate in NSC maintenance through sustaining Hes1 expression in early embryos.

Full text

Page 1

Developmental Cell
Article
Id Sustains Hes1 Expression to Inhibit
Precocious Neurogenesis by Releasing
Negative Autoregulation of Hes1
Ge Bai,1Nengyin Sheng,1Zhihui Xie,1Wei Bian,1Yoshifumi Yokota,2Robert Benezra,3Ryoichiro Kageyama,4
Francois Guillemot,5and Naihe Jing1,*
1Laboratory of Molecular Cell Biology, Key Laboratory of Stem Cell Biology, Institute of Biochemistry and Cell Biology,
Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320 Yue Yang Road, Shanghai 200031, China
2Department of Molecular Genetics, School of Medicine, University of Fukui, 23-3 Shimoaizuki, Matsuoka, Fukui 910-1193, Japan
3Department of Cancer Biology and Genetics, Memorial Sloan-Kettering Cancer Center, New York, NY 10021, USA
4Institute for Virus Research, Kyoto University, Kyoto 606-8507, Japan
5Division of Molecular Neurobiology, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA,
United Kingdom
*Correspondence: njing@sibs.ac.cn
DOI 10.1016/j.devcel.2007.05.014
SUMMARY
NegativebHLHtranscriptionfactorHes1canin-
hibit neural stem cells (NSCs) from precocious
neurogenesis through repressing proneural
gene expression; therefore, sustenance of
Hes1 expression is crucial for NSC pool mainte-
nance. Here we find that Ids, the dominant-
negative regulators of proneural proteins, are
expressed prior to proneural genes and share
an overlapping expression pattern with Hes1
in the early neural tube of chick embryos. Over-
expression of Id2 in the chick hindbrain upregu-
lates Hes1 expression and inhibits proneural
gene expression and neuronal differentiation.
By contrast, Hes1 expression decreases, pro-
neural gene expression expands, and neuro-
genesis occurs precociously in Id1;Id3 double
knockout mice and in Id1–3 RNAi-electropo-
rated chick embryos. Mechanistic studies
show that Id proteins interact directly with
Hes1 and release the negative feedback auto-
regulation of Hes1 without interfering with its
ability to affect other target genes. These re-
sults indicate that Id proteins participate in
NSC maintenance through sustaining Hes1 ex-
pression in early embryos.
INTRODUCTION
The vertebrate central nervous system derives from the
neural tube, a tubular structure that arises from the neuro-
ectoderm of the gastrula-stage embryo. Neural stem cells
(NSCs) in the neural tube are self-renewing and multipo-
tent cells which can generate intermediate and mature
cells of both neuronal and glial lineages (Gage, 2000).Dur-
ing neural development, NSCs change their competences
and developmental potential over time. In an early expan-
sion stage, NSCs primarily undergo extensive self-
renewal to enlarge their cell population (Alvarez-Buylla
et al., 2001). As development proceeds, some NSCs pro-
gressively initiate the expression of proneural genes in re-
sponse to region-specific neurogenic signals (Lillien,
1998). Proneural genes such as Mash1, Ngns, and
Math1 can induce pan-neuronal gene expression and de-
termine neuronal fate (Bertrand et al., 2002; Cai et al.,
2000). With the initiation of proneural gene expression,
cells in the neural tube become heterogeneous, including
neuronal progenitors and NSCs which express and do not
express proneural genes, respectively (Bertrand et al.,
2002; Gage, 2000). Subsequently, neuronal progenitors
exit the cell cycle and differentiate into postmitotic neu-
rons, whereas NSCs remain in an undifferentiated state
and give rise to later born cell types (Temple, 2001). Main-
tenance of the NSC pool is therefore essential for normal
neural development, as precocious neurogenesis allows
generation of early born cell types only and disorganizes
the shape and cytoarchitecture of the brain (Kageyama
et al.,2005). Although progress has been made in identify-
ing some components of the genetic program involved in
regulating NSC maintenance (Guillemot, 2005), the
detailed underlying mechanisms still remain largely un-
known.
The Hes genefamily of negative bHLH transcription fac-
tors plays crucial roles in maintaining the NSC population
(Hatakeyama et al., 2004; Hirata et al., 2001; Kageyama
et al., 2005; Sasai et al., 1992). Hes1, the best-character-
ized family member, is highly expressed in NSCs (Ka-
geyama et al., 2005). Through repressing proneural gene
expression,itinhibitsNSCsfromdifferentiatingalongneu-
ronal lineages (Nakamura et al., 2000). As development
advances, Hes1 expression is downregulated in some
NSCs, and proneural gene expression is activated (Sasai
et al., 1992). Subsequently, proneural proteins promote
these NSCs to commit to the neuronal fate and
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 283

Page 2

differentiate into neurons. In Hes gene mutant embryos,
the proneural genes Mash1 and Math3a are upregulated
together with premature neuronal differentiation (Hata-
keyama et al., 2004; Ishibashi et al., 1995; Nakamura
et al., 2000). In contrast, overexpression of Hes1 prevents
cortical progenitors from migrating out of the ventricular
zone and expressing neuronal markers (Ishibashi et al.,
1994). Therefore, sustained Hes1 expression is essential
for normal neural development. In NSCs, Notch signaling
activates Hes1 expression and inhibits premature neuro-
nal differentiation (Kageyama and Ohtsuka, 1999). How-
ever, it is difficult to explain how Notch-induced Hes1 ex-
pression is sustained at a relatively high level, considering
its capacity for autorepression (Hirata et al., 2002; Take-
bayashi et al., 1994).
In contrast with Hes1, Id family members possess
a highly conserved HLH domain but lack a DNA-binding
domain (Perk et al., 2005; Ruzinova and Benezra, 2003).
Id proteins act primarily by sequestering E proteins, pre-
venting them from forming functional heterodimers with
proneural proteins, and act as dominant-negative regula-
tors that interfere with the transcriptional activities of pro-
neural proteins in neuronal progenitors (Yokota, 2001).
Similar to Hes1, Id expression is very important for proper
neural development. Misexpression of Id in cortical pro-
genitors inhibits neuron-specific gene expression (Cai
etal.,2000).Conversely,progenitorsinId1;Id3doublemu-
tant mice withdraw prematurely from the cell cycle and
aberrantly express neuron-specific markers at an early
embryonicstage(Lydenetal.,1999).Moreover,Idproteins
arenotonlyexpressedinneuronalprogenitorsduringneu-
rogenesis but are also highly expressed in the early neural
tubes prior to the onset of proneural gene expression (Kee
and Bronner-Fraser, 2001a, 2001b; Martinsen and Bron-
ner-Fraser, 1998). In contrast with the well-documented
functions of Id proteins in neuronal progenitors, little is
known of their earlier function in the neural tube.
Here we show that Ids have expression patterns that
overlap with that of Hes1 in the early neural tube of chick
embryos. Overexpression of Id2 increases Hes1 expres-
sion, whereas inhibition of Id expression downregulates
Hes1 expression. We further show that Id proteins can di-
rectly interact with Hes1 and release the feedback autoin-
hibition of Hes1. These findings indicate that Id proteins
participate in NSC maintenance through sustaining Hes1
expression in the NSCs of early embryos.
RESULTS
Id Genes Are Expressed prior to Proneural Genes
in Early Chick Embryos
To better understand the function of Id proteins in early
embryos, we first sought to examine the expression of
Id1–3 in the chick neural tube. Consistent with previous
observations (Kee and Bronner-Fraser, 2001a, 2001b;
Martinsen and Bronner-Fraser, 1998), we found that
Id1–3 exhibited broad expression throughout the neural
tube at early stages (data not shown). In situ hybridization
showed that Id1–3 mRNA primarily resided in the dorsal
region of the metencephalon at HH (Hamburger and Ham-
ilton, 1951) stage 11, when neurogenesis has not started
in the metencephalon (Figures 1Aa–1Ac). Weak hybridiza-
tion signals could also be detected in this region for Hes1
(Figure1Ad).Hes5, however,wasonlyexpressedin aven-
tral domain (Figure 1Ae). At this stage, no expression of
proneural genes such as Cash1, c-Ngn1, c-Ngn2, or
Cath1 was detected, and no type III b-tubulin (Tuj1)-posi-
tive cellswere found either (Figures 1Ba–1Bd and data not
shown).
At HH stage 15, after the onset of neurogenesis in the
metencephalon, Id1 mRNA could be detected in the
dorsal-most region of the neural tube on either side of
the midline, and weaker signals were present throughout
the rest of the metencephalon except in the ventral-most
region (Figure 1Aa0). Id2 was much more strongly ex-
pressed than Id1 in the dorsal-most region, and gradually
decreasedalongthedorsal-ventralaxisofthemetenceph-
alon. The ventral-most region was devoid of Id2 signals,
except at the midline (Figure 1Ab0). Id3 was expressed in
a pattern largely similar to that of Id2, except that signifi-
cant signals were also detected in the ventral-most region
(Figure 1Ac0). Hes1 expression was concentrated in the
dorsal- and ventral-most regions of the neural tube, and
only weaker expression was found in the intermediate re-
gion (Figure 1Ad0). Hes5 exhibited a pattern complemen-
tary to that of Hes1, being highly expressed in the interme-
diate region and absent in the dorsal- and ventral-most
regions (Figure 1Ae0). At this stage, the expression of pro-
neural genes except Cath1 was detectable in the inter-
mediate region but not in the dorsal- and ventral-most
regions (Figures 1Ba0–1Bc0). Similarly, Tuj1-positive cells
were only detectable in the lateral aspects of the interme-
diateregion(Figure1Bd0).Weobservedsimilarexpression
patternsforId,Hes,andproneuralgenesinthedeveloping
spinal cord of the chick embryo (data not shown).
Takentogether,thesedatashowthattheonsetofIdand
Hes1 expression precedes that of proneural genes in the
chick neural tube. In the dorsal- and ventral-most regions
of HH stage 15 neural tubes, where both Id and Hes1 are
highly expressed, the expression of proneural genes and
the generation of Tuj1-positive neurons are both sup-
pressed, suggesting that Id and Hes1 might play inhibitory
roles in neuronal differentiation in early neural stem cells.
Ectopic Expression of Id2 or Hes1 Inhibits
Neuronal Differentiation
To test this hypothesis, we expressed Id or Hes1 ectopi-
cally in the intermediate region of the metencephalon at
HH stage 10–11 and assayed for neuronal differentiation
at HH stage 15. Following electroporation of an Id2-
IRES-GFP expression construct, the number of Tuj1-pos-
itive cells decreased significantly in regions of ectopic Id2
expression, and expression of Id2-GFP and type III b-tu-
bulin appeared mutually exclusive (Figures 2A and 2B).
As the reduction in Tuj1-positive cells might have been
caused by cell death, we tested neuronal apoptosis by
the TUNEL assay. No significant difference in level of ap-
optosis was observed between the Id2-electroporated
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube
284 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.

Page 3

side and the control side (data not shown). To assess the
status of the Id2-expressing cells that were inhibited from
neuronal differentiation, we examined the expression of
the proliferative cell nuclear antigen (PCNA) and found
that most of the Id2-GFP-expressing cells were PCNA im-
munoreactive(Figure2C),suggestingthatId2overexpres-
sion maintains two characteristics of neural stem cells:
proliferative capacity and inhibition of differentiation. Sim-
ilar results were obtained when electroporating a Hes1-
IRES-GFP expression construct; the proportion of Tuj1-
positive cells decreased significantly (Figures 2D and 2E)
and most of the Hes1-GFP-expressing cells were PCNA
immunoreactive (Figure 2F). Collectively, these data indi-
cate that ectopic expression of Id2 or Hes1 maintains
the self-renewal and inhibits the neuronal differentiation
of NSCs.
Id2 Upregulates Hes1 Expression
Overlapping expression patterns and similar functions
suggest a potential functional relationship between Id
and Hes1 proteins. To address this possibility, we electro-
porated mouse Id2 (m-Id2) into the intermediate region of
the metencephalon of HH stage 10–11 chick embryos and
examined endogenous c-Hes1 and c-Hes5 expression by
in situ hybridization (Figure 3A). We found that ectopic
expression of m-Id2 upregulated c-Hes1 expression (Fig-
ures 3Aa and 3Ab), whereas no significant change was
observed in c-Hes5 expression (Figure 3Ac). Interestingly,
Hes1 upregulation was restricted to the ventral half of the
m-Id2-electroporated region, suggesting that different
regions of the metencephalon might have different levels
of competence in Id expression. To confirm these in situ
hybridizationresults, wedissectedseparatelythe
Figure 1. Spatiotemporal Expression Pattern of Id, Hes, and Proneural Genes in Early Chick Embryos
(A) In situ hybridization in transverse sections of the metencephalon at HH stage 11 (a–e) and stage 15 (a0–e0). Chick Id1–3,Hes1,and Hes5 were used
as probes. The scale bars represent 150 mm for (a)–(e), and 250 mm for (a0)–(e0).
(B) In situ hybridization in transverse sections of the metencephalon at HH stage 11 (a–c) and stage 15 (a0–c0). Chick Cash1, Ngn1, and Ngn2 were
used asprobes.Immunostaining wasperformed withanti-Tuj1 antibody (d andd0).Thescale barsrepresent 150mmfor(a)–(d) and250mmfor (a0)–(d0).
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 285
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 4

Figure 2. Inhibitory Effects of Id2 and Hes1 on Neuronal Differentiation
Immunostaining was performed on transverse sections of the chick metencephalon (HH stage 15), which had previously been electroporated on the
right side of the neural tube with plasmids encoding GFP, Id2-IRES-GFP, or Hes1-IRES-GFP.
286 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 5

electroporated side and control side of the metencepha-
lon,andusedreal-timePCRtodeterminegeneexpression
levels. We found that Id2 increased c-Hes1 expression
and inhibited Cash1 and p27 expression (Figure 3B).
Cash1 and p27 mark neuronal commitment and neuronal
differentiation, respectively, and are both known down-
stream targets of Hes1 (Chen et al., 1997; Murata et al.,
2005; Stanke et al., 2004). These results suggest that up-
regulation of c-Hes1 and the subsequent downregulation
of c-Hes1 target genes might explain the inhibitory effect
of Id2 on the neuronal differentiation of NSCs.
In the m-Hes1-electroporated region, no change was
observed upon expression of c-Id1 and c-Id2, and c-Id3
expression was slightly repressed (see Figure S1 in the
Supplemental Data available with this article online). How-
ever, the ectopically expressed m-Hes1 was found to
repress expression of Cash1, p27, and endogenous
c-Hes1 (Figure S1). This suggests that Hes1 cannot effi-
ciently regulate Id expression. Together, these results
indicatethatIdproteinsactasupstreamregulatorstosus-
tain Hes1 expression in NSCs of the chick embryo.
Id Proteins Are Required for Sustaining
Hes1 Expression
Given that ectopic expression of Id2 can increase Hes1
expression and inhibit neuronal differentiation, we won-
dered whether inhibition of Id1–3 expression could de-
crease Hes1 expression and promote precocious neuro-
genesis. We generated siRNAs directed against chick
Id1, Id2, and Id3 and tested their efficiency and specificity
by cotransfection with c-Id1–3 and c-Hes1 expression
vectors into HEK293 cells. Immunoblot analysis showed
that knockdown of Id1–3 expression was efficient and
genespecific(datanotshown).Inordertoavoid functional
redundancy of different Id proteins, we coelectroporated
siRNAs against Id1, Id2, and Id3 into the dorsal-most re-
gion of the metencephalon at HH stage 10–11, and ana-
lyzed gene expression at HH stage 15 (Figure 3C). We
found that the electroporation of Id1–3 siRNAs could in-
hibit endogenous Id1, Id2, and Id3 expression to moder-
ate levels in the dorsal-most region (Figures 3Ca0–3Cc0),
whereas control siRNAs had no effect (Figures 3Ca–
3Cc). No apoptosis was detected by TUNEL assay (data
not shown). As predicted, Id1–3 siRNAs caused Hes1
downregulation in the dorsal metencephalon, but had
no effect on Hes1 expression in the unelectroporated
ventral-most region (Figures 3Cd and 3Cd0). Thus, inhibi-
tion of Id expression in the dorsal metencephalon de-
creases Hes1 expression.
The dorsal patterning morphogen Wnt1 is almost miss-
ing in Hes1;Hes5 double knockout mice (Hatakeyama
et al., 2004), and we wondered whether reduction of
Hes1 expression by Id knockdown could also affect
Wnt1 expression. In the dorsal-most region, inhibition of
Id expression indeed led to a dramatic reduction in Wnt1
expression (Figure 3Ce0), whereas in control siRNA-
treated embryos, Wnt1 expression could still be observed
(Figure 3Ce). However, unlike Hes1 knockout mice (Hata-
keyama et al., 2004), the transient and partial downregula-
tion of Hes1 expression induced by Id1–3 knockdown did
not result in the ectopic expression of Hes5. Therefore,
Hes5failed tocompensate theHes1functionofsustaining
Wnt1expressioninthisregion(FigureS2).Together,these
data further indicate that Id proteins are responsible for
maintaining the Hes1-dependent expression of patterning
morphogens such as Wnt1.
To confirm the results obtained with siRNAs in chick
embryos, we examined Hes1 expression in Id1;Id3 double
mutant mice. Id1?/?Id3?/?mutant embryos have been
shown to display aberrant neurogenesis, with premature
neuronal differentiation and extensive expression of pro-
neural genes (Lyden et al., 1999). We found that E11.5
Id1?/?Id3?/?embryos present a significant reduction in
Hes1 expression in the dorsal-most region of the meten-
cephalon and developing spinal cord (Figures 4Aa0and
4Ca0). Noticeably, downregulation of Hes1 expression
was also evident in the intermediate region of the hind-
brain and developing spinal cord (Figures 4Ab0and
4Ca0). These results were confirmed by real-time PCR
(Figure 4D). As previously reported (Lyden et al., 1999),
the expression of proneural genes, such as Math1, ex-
panded significantly in regions with reduced Hes1 expres-
sion(Figures4Ac0and4Cb0).MAP2-positiveneuronswere
prematurelygeneratedfromnormallyneuron-freeregions,
suchasthedorsal-mostregionandtheventricularzone,in
Id1;Id3 double knockout embryos (Figures 4B and 4E).
Taken together, these results show that inhibition of Id
expression by siRNA electroporation in chick embryos or
knockout in mice causes Hes1 downregulation, proneural
gene expansion, and precocious neurogenesis, support-
ing the view that Id proteins are required for sustaining
Hes1 expression in NSCs.
(A) GFP fluorescence (green), Tuj1 immunofluorescence (red), and the corresponding merged images (yellow) are shown for typical fields of GFP-
electroporated(a–c)andId2-electroporated(e–g) chickmetencephalon.Boxed regionsin(c)and(g)areenlargedin(d)and(h),respectively.Thescale
bars represent 200 mm for (a)–(c) and (e)–(g), and 80 mm for (d) and (h).
(B) Percentage of Tuj1+cells among GFP+cells in the GFP-electroporated (n = 9) and Id2-electroporated chick metencephalon (n = 12). Data are
presented as the mean ± SD.
(C) GFP fluorescence (green) and PCNA immunofluorescence (red) are shown for typical fields of Id2-electroporated chick metencephalon (a and b).
The corresponding merged image of the boxed region (yellow) is enlarged in (c). The scale bars represent 200 mm for (a) and (b), and 70 mm for (c).
(D) GFP fluorescence (green), Tuj1 immunofluorescence (red), and the corresponding merged images (yellow) are shown for typical fields of Hes1-
electroporated chick metencephalon (a–c). The boxed region in (c) is enlarged in (d). The scale bars represent 200 mm for (a)–(c), and 80 mm for (d).
(E) The percentage of Tuj1+cells among GFP+cells in GFP-electroporated (n = 9) and Hes1-electroporated chick metencephalon (n = 9). Data are
presented as the mean ± SD.
(F)GFPfluorescence(green)andPCNAimmunofluorescence(red)areshownfortypicalfieldsofHes1-electroporatedchickmetencephalon(aandb).
The corresponding merged image of the boxed region (yellow) is enlarged in (c). The scale bars represent 200 mm for (a) and (b), and 70 mm for (c).
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 287
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 6

Hes1 Mediates the Inhibitory Effect of Id Proteins
on Neuronal Differentiation
Having established the roles of Id proteins in sustaining
Hes1 expression, we set out to demonstrate that Hes1 is
involved in mediating the inhibitory effect of Id proteins
on neuronal differentiation. For this, we electroporated
an Id2-IRES-GFP expression construct together with
a control siRNA or a Hes1 siRNA into the intermediate re-
gion of the metencephalon of an HH stage 10–11 chick
embryo, and examined Cash1 expression and Tuj1-posi-
tive neurons at HH stage 15. We reasoned that if the inhib-
itory effect on neuronal differentiation by Id proteins re-
quires upregulation of endogenous Hes1 expression,
coelectroporation of a Hes1 siRNA should rescue the
Id2-induced inhibition of Cash1 expression and neuronal
differentiation. Consistent with this hypothesis, we found
Figure 3. Effects of Id Overexpression and Knockdown in the Chick Metencephalon
(A) Transverse sections through HH stage 15 chick metencephalon, with the right side electroporated with mouse Id2 expression construct. In situ
hybridization was performed with mouse Id2, chick Hes1, and chick Hes5 as probes. Note that the probe for mouse Id2 does not recognize endog-
enous chick Id2 mRNA. The scale bar represents 200 mm.
(B)Real-timePCR analysisof chick Hes1,Cash1,and p27mRNA expressioninId2-electroporated embryonic chick brain (n=25). Dataare presented
as the mean ± SD of triplicate quantifications.
(C) Transverse sections through HH stage 15 chick metencephalon, with the dorsal-most region electroporated with control siRNA or Id1–3 siRNA. In
situ hybridization was performed with chick Id1–3, Hes1, and Wnt1 as the probes. The scale bar represents 300 mm.
288 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 7

Figure 4. Decreased Hes1 Expression in Id1?/?Id3?/?Mouse Embryos
(A) In situ hybridization in transverse sections of E11.5 mouse hindbrain. In Id1?/?Id3?/?embryos, Hes1 mRNA expression decreased in the dorsal-
most region of the metencephalon (a0) and the intermediate region of the hindbrain (b0), compared with control Id1?/?Id3+/?embryos. Math1 expres-
sionwasexpandedintheintermediateregionofId1?/?Id3?/?hindbrain(c0).Thescalebarsrepresent100mmfor(a)and(a0),and200mmfor(b),(b0),(c),
and (c0).
(B)ImmunostainingofMAP2intransversesectionsofE11.5metencephalon.Theboxedregionisenlargedintheinsets.Thescalebarrepresents200mm.
(C) In situ hybridization in transverse sections of E11.5 spinal cord. In Id1?/?Id3?/?embryos, Hes1 mRNA expression was barely detectable in the dorsal
and intermediate regions of the spinal cord (a0), compared with control Id1?/?Id3+/?embryos (a). Math1 was ectopically expressed in the dorsal-most
region of Id1?/?Id3?/?spinal cord (b0). The scale bar represents 100 mm.
(D)TheneuraltubewasdissectedfromE11.5Id1?/?Id3+/?orId1?/?Id3?/?embryos,andtotalRNAwasextracted.Hes1mRNAexpressionwasanalyzed
by real-time PCR. Data are presented as the mean ± SD of triplicate quantifications.
(E) Immunostaining of MAP2 in transverse sections of E11.5 spinal cord. Note that MAP2-positive neurons were prematurely generated from the dorsal-
most region and the ventricular zone of Id1?/?Id3?/?embryos (arrows). The scale bar represents 100 mm.
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 289
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 8

thattheHes1siRNAcouldnotonlyinhibitId-inducedHes1
upregulation but also partially rescue both Cash1 repres-
sion and neuronal inhibition imposed by ectopic expres-
sion of Id2 (Figures 5Aa00–5Ad00and 5B), whereas the con-
trol siRNA was not effective (Figures 5Aa0–5Ad0). These
results suggest that Hes1 is a downstream effector of Id
proteins in the inhibition of proneural gene expression
and neuronal differentiation. However, the observation
that Hes1 knockdown failed to fully rescue the inhibitory
effectofIdonneuronaldifferentiation(Figure5B)suggests
that there might be an additional Hes1-independent
mechanism(s) involved in mediating the functions of Id
proteins.
Taken together, gain- andloss-of-functionanalysisof Id
proteinshasallowedustoidentifyafunctionalrelationship
between Id and Hes1; Id proteins act as upstream regula-
tors of Hes1 to sustain Hes1 expression, and Hes1 medi-
ates, at least partially, Id’s inhibitory effect on neuronal
differentiation through repressing its target genes, such
as Cash1 and p27.
Figure 5. Hes1 Mediates the Inhibitory
Effect of Id Proteins on Neuronal Differ-
entiation
(A) Transverse sections through HH stage 15
chick metencephalon, with the right side elec-
troporated with a mouse Id2-IRES-GFP ex-
pression construct and Hes1 or control siRNA.
Insituhybridizationwasperformedwithmouse
Id2 (a–a00), chick Hes1 (b–b00), and chick Cash1
(c–c00) as probes. Immunostaining was per-
formed with anti-Tuj1 antibody (d–d00). The
scale bar represents 250 mm.
(B) The percentage of Tuj1+cells among GFP+
cells in GFP-electroporated (n = 6), Id2-GFP-
plus control siRNA-electroporated (n = 6), or
plus Hes1 siRNA-electroporated chick meten-
cephalon (n = 6). Data are presented as the
mean ± SD.
290 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 9

Id Proteins Activate Hes1 Expression by Releasing
Its Negative Autoregulation
Previous studies have shown that a dominant-negative
form of Hes1 (DN-Hes1), in which the DNA-binding do-
main was mutated, could form a non-DNA-binding heter-
odimer with wild-type Hes1 protein and increase Hes1 ex-
pression through suppressing the negative autoregulation
oftheHes1gene(Hirataetal.,2002;Strometal.,1997).As
Id proteins also lack the DNA-binding domain (Benezra
et al., 1990), we speculated that they might use a similar
mechanism to regulate Hes1 expression. To validate this
hypothesis, we performed a luciferase assay with the
2.5 kb mouse Hes1 promoter (Takebayashi et al., 1994),
cotransfecting it with increasing amounts of either DN-
Hes1 or Id2 expression constructs into P19 cells. We
found that Id2 was indeed as efficient as DN-Hes1 at in-
creasingreporter constructactivity(Figure6A).Toconfirm
that Id2 promotes Hes1 transcription through the Hes1-
binding N box in the Hes1 promoter (Takebayashi et al.,
1994), we generated a luciferase reporter construct con-
taining a mutant Hes1 promoter in which all three N boxes
were inactivated (Takebayashi et al., 1994). We cotrans-
fected this construct with DN-Hes1 or Id2 expression
plasmids, and found that the basal transcriptional activity
of the mutant Hes1 promoter was higher than that of wild-
type in P19 cells, which might be due to the mutant Hes1
promoter failing to respond to the inhibition of endoge-
nous Hes1 in P19 cells (Figure 6A). And, increasing
amounts of DN-Hes1 or Id2 had no effect on reporter ac-
tivity, suggesting thatthe N box was required for DN-Hes1
or Id2 to increase Hes1 gene transcription. To further con-
firm this issue, we generated a luciferase reporter plasmid
containing six repeats of the N box and cotransfected the
construct with DN-Hes1 or Id2. Similarly, DN-Hes1 or Id2
alsoincreasedluciferasereporteractivity(Figure6B).Sim-
ilar results were obtained with Id1 and Id3 (Figure S3). As
P19 cells express endogenous Hes1 protein (Sasai et al.,
1992),wecouldnotdeterminewhetherIdproteinsactually
act through Hes1. Therefore, HeLa cells which do not ex-
pressendogenous Hes1(Murata etal.,2005)were chosen
to perform the same luciferase assay. Cotransfection of
DN-Hes1 or Id2 with the N box-luciferase reporter in
HeLa cells did not increase reporter gene expression
(data not shown). However, a Hes1 expression vector
could repress luciferase reporter construct activity, and
both DN-Hes1 and Id2 could rescue such repression (Fig-
ure 6C). Taken together, these results support our model
whereby Id proteins sustain Hes1 expression by releasing
the inhibitory effect of Hes1 on its own promoter.
Id Proteins Interact with Hes1 and Suppress
Its N Box Binding Activity
To determine whether Id proteins release the inhibitory ef-
fect of Hes1 by directly interacting with Hes1 proteins,
GST pull-down and coimmunoprecipitation assays were
performed. In vitro translated Hes1 protein was pulled
down by a GST-Id2 fusion protein, but not by GST alone
(data not shown). To clarify whether Id2 interacted with
Hes1 through direct physical binding, Myc-tagged Id2
and Flag-tagged Hes1 were cotransfected into HEK293T
cells. A coimmunoprecipitation assay showed that Hes1
could be coprecipitated with Id2 and vice versa (Fig-
ure 6D). Id1 and Id3 proteins could also be coprecipitated
withHes1(Figure S4).Similarresults havebeen previously
reported (Jogi et al., 2002). To confirm these results ob-
tained in transfected cells, coimmunoprecipitation exper-
iments were repeated in P19 cells and E10.5 mouse brain
tissue. Endogenous Hes1 protein was found in immuno-
precipitates with anti-Id2 antibody but not with control
IgG (anti-Hes1 antibody was not available for immunopre-
cipitation) (Figure 6E), indicating that endogenous Id2 and
Hes1 proteins indeed form a complex in P19 cells and in
the developing mouse brain.
To determine whether Id proteins could interfere with
the DNA binding activity of Hes1, in vitro purified Hes1
protein was subjected to an electrophoresis mobility-shift
assay (EMSA) using a32P-labeled N box-containing oligo-
nucleotide as a probe. A prominent band appeared when
the probe was mixed with Hes1 protein (Figure 6F, lane 2)
but not with BSA (Figure 6F, lane 1), and the band disap-
peared in the presence of an excess of unlabeled wild-
type (WT) but not N box-mutated (Mut) oligonucleotide
(Figure 6F, lanes 3 and 4). To examine the effect of Id2
on the DNA binding activity of Hes1, we preincubated
thesame amountof Hes1protein withincreasingamounts
(50, 150, and 450 ng) of GST-Id2 fusion protein or GST
protein as a control. The binding activity of Hes1 to the
N box probe decreased gradually with increasing concen-
trations of Id2protein (Figure6F,lanes5–7)but notof con-
trol GST protein (Figure 6F, lanes 8–10). These results
show that Id proteins can interact with Hes1 and suppress
its N box binding activity.
To determine which domain of Id2 protein was required
to interact with Hes1, we generated GFP-Id2 fusion pro-
teins with deletions of the N-terminal (DN), HLH (DHLH),
or C-terminal (DC) domains and cotransfected these con-
structs with Flag-Hes1 into HEK293T cells. Coimmuno-
precipitation experiments showed that Hes1 could inter-
act with GFP-Id2 fusion proteins with the deletion of the
N-terminal or C-terminal domain, but not with the deletion
of the HLH domain. The luciferase reporter assay further
showed that the GFP-Id2-DHLH fusion protein could not
release the inhibitory effect of Hes1 on the N box (Fig-
ure S5). These results indicate that the HLH domain of
Id2 protein is the region that interacts with Hes1, and
that Id2 requires this interaction to activate the Hes1 pro-
moter.
Taken together, these results show that Id proteins
interact directly with Hes1 through their HLH domain
and suppress the DNA binding activity of Hes1, thereby
releasing the negative feedback loop of Hes1 on its own
promoter.
Id2 Releases the Negative Autoregulation of Hes1
in the Developing Chick Brain
In the previous section, we described in vitro experiments
to establish the mechanism through which Id2 proteins in-
teract with Hes1 and release its negative feedback loop.
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 291
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 10

Figure 6. Id2 Interacts with Hes1 and Suppresses Its Negative Autoregulation
(A and B) The wild-type/mutant Hes1 promoter-luc (A) or pN6-bA-luc plasmid (B) was transfected into P19 cells with increasing amounts of DN-Hes1
or Id2 expression construct. Each experiment was repeated at least three times. The results are presented as mean ± SD.
(C) pN6-bA-luc plasmidwastransfectedintoHeLacells withHes1 expression vector and increasing amountsofDN-Hes1orId2 expression construct
as indicated. Each experiment was repeated at least three times. The results are presented as mean ± SD.
(D) Interaction of Hes1 and Id2 proteins in transfected cells. HEK293T cells were transfected with plasmids encoding Flag-Hes1, Id2-Myc, or the cor-
responding empty vectors as indicated. The cell lysates were subjected to immunoprecipitation (IP) with antibodies to Myc or Flag, and the resultant
immunoprecipitates were analyzed by immunoblot (IB).
292 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 11

To confirm that this mechanism operates in vivo, an EGFP
reporterplasmiddrivenbythe2.5kbHes1promoter(Hes1
promoter-EGFP) was electroporated into the intermediate
region of the metencephalon of HHstage 10–11chick em-
bryos. EGFP expression was detected at HH stage 15
(Figures 7Aa and 7Ab), probably due to endogenous
Notch signaling, which is known to activate the Hes1 pro-
moter. When the reporter plasmid was coelectroporated
with a Hes1 expression construct, EGFP expression was
abolished (Figures 7Ac–7Ae). Moreover, when the re-
porter plasmid was coelectroporated with both Hes1
and Id2-myc expression plasmids, EGFP expression
was rescued (Figures 7Af–7Ai). These results indicate
that Id2 protein can release the autorepression of Hes1
in the developing chick brain.
Id2 Protein Fails to Release the Inhibitory Effect
of Hes1 from the Class C Site of the Hash1 Promoter
Hes1proteincanbindtothepromotersofproneuralgenes
to inhibit proneural gene expression (Chen et al., 1997).
Therefore, it is possible that Id proteins can also release
the inhibition of Hes1 on proneural genes. To validate
this possibility, the proneural gene Hash1 promoter, to
which Hes1 protein bound and suppressed its expression
(Chen et al.,1997),was chosen to perform the reporter as-
say. We cotransfected a luciferase reporter plasmid con-
taining five repeats of the class C site of the Hash1 pro-
moter with the Hes1 expression vector in HeLa cells and
found that Hes1 could repress the expression of the re-
porter gene, whereas DN-Hes1 could release this repres-
sion. Interestingly, Id2 expression could notrelease the in-
hibition of Hes1 to the class C site of the Hash1 promoter
(Figure 7B). Similareffects were alsoobserved in P19cells
(Figure7C).TheseresultssuggestthatIdproteinscanonly
release the inhibitory effect of Hes1protein from the N box
of its own promoter (Figures 6B and 6C) but not from the
class C site of the proneural gene promoter.
Next, we wondered whether this was caused by the
different abilities of Id2 to interfere with the Hes1 DNA
binding activity between the N box and the C site. As pre-
viously reported (Jogi etal., 2002),Hes1could also bindto
a32P-labeled class C site probe (Figure 7D, lanes 1 and 2),
andthebindingwasspecificfortheclassCsite(Figure7D,
lanes 3 and 4). The high concentration of Id2 protein (300
and450ng)couldefficientlycompetewithHes1bindingto
this probe (Figure 7D, lanes 7 and 8), whereas the effect of
lower concentrations (50 and 150 ng) was not as signifi-
cant (Figure 7D, lanes 5 and 6). This was different from
ourprevious datathatevenrelatively lowdosesofId2pro-
tein (150 ng) could significantly interfere with Hes1 binding
to the N box (Figure 6F, lane 6). Moreover, the high-dose
Id2 protein (300 ng) could completely block the binding
activity of Hes1 to the N box (Figure 7D, lane 14), whereas
this concentration was insufficient to completely block
Hes1 binding to the class C site (Figure 7D, lane 7). To-
gether, these results show that Id2 has a lower threshold
to inhibit the N box binding activity of Hes1 than to inhibit
that of the class C site. This might provide an explanation
for the differing abilities of Id2 to release the inhibition of
Hes1 on itself and on its target genes.
DISCUSSION
Inthis study, weshowed that in the neural tubeof the early
chick embryo, Id genes shared an overlapping expression
pattern with Hes1 and that ectopic expression of Id2 in-
ducedHes1 expressionandrepressedproneuralgeneex-
pression and normal neuronal differentiation. Conversely,
inhibition of Id expression led to decreased Hes1 expres-
sion, expanded proneural gene expression, and prema-
ture neuronal differentiation in many regions of the central
nervous system. As a result, RNAi-electroporated chick
embryos and Id1;Id3 double mutant mice had a reduced
hindbrain size (data not shown). Taken together, these
results strongly support the notion that sustained Hes1
expression by Id proteins is a critical mechanism for main-
tenance of the NSC pool in early embryos.
Previous studies showed that Id proteins could act as
dominant-negative regulators to interfere with the tran-
scriptional activities of proneural proteins and inhibit pre-
mature differentiation of neuronal progenitors (Yokota,
2001). In this study, we found a novel function of Id pro-
teins to sustain Hes1 expression and prevent precocious
neuronal differentiation of NSCs in the early embryo.
How do the two activities of Id proteins relate to each
other? We showed that Id2 overexpression could strongly
inhibit neuronal differentiation of NSCs, and that Hes1
knockdown could not fully rescue the inhibitory effect of
Id2 protein. Based on this observation and the fact that
the neural tube at this stage is composed of heteroge-
neous cell types including NSCs and neuronal progeni-
tors, we propose that Id proteins inhibit neuronal differen-
tiation possibly through two different mechanisms: they
repress proneural gene expression through sustaining
Hes1 expression in NSCs, or act as dominant-negative
regulators to block the function of proneural proteins in
neuronal progenitors. Therefore, the observed inhibition
of neuronal differentiation of Id2 protein in early embryos
should be the synergistic effect of these two mechanisms.
The Hes1 expression inhibition by siRNA could only inter-
fere with the former mechanism and leave the latter intact.
Thus, Hes1 knockdown could only partially rescue the
(E) Interaction of endogenous Hes1 and Id2 proteins. P19 cells or E10.5 mouse brains were lysed and subjected to immunoprecipitation with anti-Id2
antibody. The resultant immunoprecipitates were analyzed by immunoblot.
(F) Electrophoretic mobility-shift assay (EMSA) of N box binding activity. GST, GST-Id2, and Hes1 proteins were expressed and purified from Escher-
ichia coli and subjected to EMSA. The32P-labeled N box probe was incubated with BSA (lane 1) or Hes1 proteins (lane 2–10). As a competitor, an
excess of wild-type (WT) or N box-mutated (Mut) oligonucleotides was added (lanes 3 and 4). Increasing amounts (50, 150, and 450 ng, lanes 5–7)
of GST-Id2 fusion protein were mixed with Hes1 protein, after which the labeled N box probe was added. The same amounts of GST protein (lanes
8–10) were used as a negative control.
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 293
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 12

Figure 7. Effect of Id2 on the Negative Autoregulation of Hes1 in the Developing Chick Brain
(A)Hes1promoter-EGFPandtracerconstructpHcRed1werecoelectroporatedintochickmetencephalonwithplasmidsencodingHes1,Myc-tagged
Id2, or the corresponding empty vector. Tracer (a, c, and f), EGFP (b, d, and g), Hes1 (anti-Hes1; [e and h]), and Id2 (anti-Myc; [i]) expressions were
detected 20 hr later. The scale bar represents 500 mm.
294 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 13

inhibitory effect of Id2 protein. Of course, we cannot rule
out another unknown Hes1-independent mechanism(s)
involved in this process.
Feedback inhibition is an important mechanism to reg-
ulate Hes1 expression during embryonic development
(Baek et al., 2006; Hirata et al., 2002; Takebayashi et al.,
1994). In the neural tube, for example, Hes1 exhibits two
different modes of expression. In compartment regions
such as the intermediate aspects of the neural tube,
Hes1isexpressedatvariablelevelsindifferentcells,prob-
ably due to the activity of the Hes1 negative feedback
loop. In boundary cells, such as the dorsal-most region
of the neural tube, however, Hes1 has a persistent and
high-level expression,anditseemsthatthenegativefeed-
back loop does not work there (Baek et al., 2006). More-
over, expression of proneural genes is inhibited and neu-
ronal differentiation is delayed or does not occur in these
boundary regions (Guthrie and Lumsden, 1991; Kahane
and Kalcheim, 1998). Cells in the dorsal-most region of
the neural tube express Wnt and BMP family members
and act as an organizing center to regulate the dorsal-
ventral patterning of the neural tube. Persistent, high-level
Hes1 expression in these regions is crucial for mainte-
nance of these NSC populations as well as maintenance
of organizer activities (Baek et al., 2006). However, it re-
mains unclear how this high and persistent Hes1 expres-
sion is sustained in these boundary regions. We found
that the inhibition of Id expression reduced Hes1 expres-
sion in the dorsal-most region of the metencephalon and
the developing spinal cord in Id1–3 RNAi-electroporated
chick and Id1;Id3 double mutant mouse embryos. We
also found that Id proteins can interact directly with
Hes1 and inhibit its DNA binding activity, thereby sup-
pressing the feedback repression of its own promoter.
These results suggest that in these boundary regions,
the highly expressed Id proteins act as suppressors of
the Hes1 negative feedback loop and sustain the persis-
tent and high expression of Hes1.
Lateral inhibition is a well-characterized Notch-medi-
ated mechanism involved in NSC maintenance (Bertrand
et al., 2002). Newly formed neurons upregulate the Notch
ligand Delta, thereupon activating Notch signaling in adja-
cent NSCs. Upon activation, the intracellular domain of
Notch is translocated to the nucleus and forms a complex
with the intracellular molecule RBP-J. This complex di-
rectly binds to the Hes1 promoter and activates Hes1 ex-
pression, thereby inhibiting the precocious neuronal
differentiation of NSCs (Selkoe and Kopan, 2003). How-
ever, in Id1;Id3 double knockout embryos, Hes1 expres-
sion decreases to a very low level and premature neuro-
genesis occurs, suggesting that Notch signaling alone is
not sufficient to sustain Hes1 expression and prevent pre-
matureneuronaldifferentiationofNSCs.Therefore,Idpro-
teins are required to repress the negative feedback of
Hes1 andmaintain Notchsignal-induced Hes1 expression
(Figure7E).Moreover,giventhatsecretedfactors,suchas
Wnts and BMPs, can regulate Id expression in different
tissues(datanotshown;Hollnagel etal.,1999;Nakashima
etal.,2001;Rockmanetal.,2001),thisnovel functionalre-
lationshipbetweenIdandHes1proteinsprovidesamolec-
ular basis to integrate extrinsic signals with the intrinsic
program of NSC maintenance (Figure 7E).
The function of Hes1 is to inhibit neuronal differentiation
ofNSCsthroughrepressingexpressionofitstargetgenes,
such as Mash1 and p27. In contrast with the self-inhibition
that involves binding to N boxes (CACNAG) in its own pro-
moter, Hes1 represses target gene transcription through
binding to the class C site (CACGCA) in the promoter of
thetargetgene(Chenetal.,1997).However,littleisknown
about the effects of different binding activities of Hes1 on
these different binding sites. It is also unclear whether Id
proteins interfere differentially with Hes1 binding activities
to the N box and C site. We found that Id2 could not liber-
ate the inhibition of Hes1 on C site-driven reporter gene
expression in P19 and HeLa cells. We also found that
only high concentrations of Id2 fusion protein could effec-
tively compete with Hes1 binding to a C site probe, and
that the effect of lower concentrations was not as signifi-
cant. These results were different from our observations
on the N box, in which Id expression could release the in-
hibition of Hes1 on N box-driven reporter gene expres-
sion, and that even relatively low doses of Id2 protein
could efficiently interfere with Hes1 binding to the N box.
One possible explanation for this discrepancy is that
Hes1 protein has different binding affinities for the N box
and C site, with Hes1 binding to the C site with higher af-
finity than to the N box (data not shown; Ohsako et al.,
1994; Van Doren et al., 1994). Due to rapid degradation
or interaction with other proteins (Bounpheng et al.,
1999; Jogi et al., 2002), it is difficult for Id proteins to accu-
mulate to a very high concentration in the developing
nervous system. Thus, Id proteins could only release
(B) Effect of Id2 on transcriptional activity of the class C site-driven reporter. pC5-bA-luciferase plasmid was transfected into HeLa cells with Hes1
expression vector and increasing amounts (125, 250, and 500 ng) of DN-Hes1 or Id2 expression construct. Each experiment was repeated at least
three times. The results are presented as mean ± SD.
(C) pC5-bA-luciferaseplasmidwastransfectedintoP19cells withincreasing amounts(125, 250,and 500ng) ofDN-Hes1 orId2expressionconstruct.
Each experiment was repeated at least three times. The results are presented as mean ± SD.
(D) Electrophoretic mobility-shift assay of the class C site binding activity of Hes1 protein. The32P-labeled class C site probe was mixed with 50 ng
BSA (lane 1) or Hes1 protein (lanes 2–12). The competition experiments were performed with an excess of unlabeled wild-type (WT) or class C site-
mutated (Mut) oligonucleotides (lanes 3 and 4). Increasing amounts (50, 150, 300, and 450 ng; lanes 5–8) of GST-Id2 fusion protein or, as a negative
control, GST protein (lanes 9–12), were mixed with Hes1 protein before the probe was added. As a positive control, Id2 fusion protein (300 ng) was
mixedwithHes1protein(50ng)beforethe32P-labeledNboxprobewasadded.NotethatId2couldeffectivelyinterferewithHes1bindingtotheNbox
(lanes 13 and 14).
(E) Mechanism of NSC maintenance in early embryos. In NSCs, Id proteins function to sustain Notch-induced Hes1 expression through suppression
of the negative autoregulation of Hes1, thereby inhibiting premature expression of proneural genes and neuronal differentiation.
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 295
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 14

Hes1feedback repressionon its ownpromoter, but noton
Hes1’sdownstreamtargets.Ofcourse,wecannotruleout
the possibility that unknown cofactors might exist, and
that different complexes between Id proteins and cofac-
tors might determine their interaction with Hes1 and affect
the DNA binding activity of Hes1 to the N box or C site in
NSCs.
EXPERIMENTAL PROCEDURES
Animals
The generation ofId1;Id3 doublemutantmicehas beendescribed pre-
viously (Lyden et al., 1999). All experiments were carried out in accor-
dance with the United States National Institutes of Health Guide for the
Care and Use of Laboratory Animals.
In Ovo Chick Embryo Electroporation
Fertilized eggs were obtained from the Shanghai Academy of Agricul-
tural Sciences. The chick embryo in ovo electroporation was per-
formed as previously described (Yan et al., 2004).
In Situ Hybridization
Section in situ hybridization was performed as described previously
(Birren et al., 1993). Detailed protocols are available upon request.
The following chick in situ probes were used: Id1, Id2, Id3, Hes1,
Hes5, Cash1, Ngn1, Ngn2, and Wnt1. The following mouse in situ
probes were used: Id1, Id2, Id3, Hes1, and Math1.
Coimmunoprecipitation Assay
Immunoprecipitations were performed as described previously (Jogi
et al., 2002). The following antibodies were used: anti-Flag (Sigma),
anti-Myc (Covance), anti-GFP (Covance), anti-Id2 (Santa Cruz), and
anti-Hes1 (kindly provided by T. Sudo).
Luciferase Assay
TheluciferaseplasmidcontainingtheNboxmutantversionoftheHes1
promoterwasconstructedasdescribedpreviously(Takebayashietal.,
1994). Luciferase assays were performed as described previously
(Cheng et al., 2004).
Electrophoretic Mobility-Shift Assay
Protein-DNA complexes were examined by EMSA as previously de-
scribed (Cheng et al., 2004). Detailed information is available in the
Supplemental Data.
Statistics
Each experiment was repeated at least three times, and similar results
were obtained. Data were expressed as mean ± SD. Student’s t tests
were used to compare the effects of all treatments. Differences were
considered statistically significant at p < 0.05.
Supplemental Data
Supplemental Data include five figures, Supplemental Experimental
Procedures, and Supplemental References and are available at
http://www.developmentalcell.com/cgi/content/full/13/2/283/DC1/.
ACKNOWLEDGMENTS
We thank Drs. M. Bronner-Fraser, C. Kalcheim, D. Henrique, and C.
Stern for in situ probes and cDNAs and T. Sudo for the Hes1 antibody.
This work was supported in part by the National Natural Science Foun-
dation of China (90208011, 30470856, 30421005, 30623003), National
KeyBasicResearch andDevelopment
(2005CB522704, 2006CB943902), Ministry of Science and Technol-
ogy of China (2007CB947100), National High-Tech Research and De-
velopment Program of China (2006AA02Z186), Shanghai Key Project
of Basic Science Research (04DZ14005, 06DJ14001, 06DZ22032),
Program ofChina
and the Council of Shanghai Municipal Government for Science and
Technology (05814578).
Received: January 11, 2007
Revised: April 24, 2007
Accepted: May 25, 2007
Published: August 6, 2007
REFERENCES
Alvarez-Buylla, A., Garcia-Verdugo, J.M., and Tramontin, A.D. (2001).
A unified hypothesis on the lineage of neural stem cells. Nat. Rev. Neu-
rosci. 2, 287–293.
Baek, J.H., Hatakeyama, J., Sakamoto, S., Ohtsuka, T., and Ka-
geyama, R. (2006). Persistent and high levels of Hes1 expression reg-
ulate boundary formation in the developing central nervous system.
Development 133, 2467–2476.
Benezra, R., Davis, R.L., Lockshon, D., Turner, D.L., and Weintraub, H.
(1990). The protein Id: a negative regulator of helix-loop-helix DNA
binding proteins. Cell 61, 49–59.
Bertrand, N., Castro, D.S., and Guillemot, F. (2002). Proneural genes
and the specification of neural cell types. Nat. Rev. Neurosci. 3, 517–
530.
Birren, S.J., Lo, L., and Anderson, D.J. (1993). Sympathetic neuro-
blasts undergo a developmental switch in trophic dependence. Devel-
opment 119, 597–610.
Bounpheng, M.A., Dimas, J.J., Dodds, S.G., and Christy, B.A. (1999).
Degradation of Id proteins by the ubiquitin-proteasome pathway.
FASEB J. 13, 2257–2264.
Cai, L., Morrow, E.M., and Cepko, C.L. (2000). Misexpression of basic
helix-loop-helix genes in the murine cerebral cortex affects cell fate
choices and neuronal survival. Development 127, 3021–3030.
Chen, H., Thiagalingam, A., Chopra, H., Borges, M.W., Feder, J.N.,
Nelkin, B.D., Baylin, S.B., and Ball, D.W. (1997). Conservation of the
Drosophila lateral inhibition pathway in human lung cancer: a hairy-
related protein (HES-1) directly represses achaete-scute homolog-1
expression. Proc. Natl. Acad. Sci. USA 94, 5355–5360.
Cheng, L., Jin, Z., Liu, L., Yan, Y., Li, T., Zhu, X., and Jing, N. (2004).
Characterization and promoter analysis of the mouse nestin gene.
FEBS Lett. 565, 195–202.
Gage, F.H. (2000). Mammalian neural stem cells. Science 287, 1433–
1438.
Guillemot, F. (2005). Cellular and molecular control of neurogenesis in
the mammalian telencephalon. Curr. Opin. Cell Biol. 17, 639–647.
Guthrie, S., and Lumsden, A. (1991). Formation and regeneration of
rhombomere boundaries in the developing chick hindbrain. Develop-
ment 112, 221–229.
Hamburger, V., and Hamilton, H.L. (1951). A series of normal stages in
the development of the chick embryo. J. Morphol. 88, 49–92.
Hatakeyama, J., Bessho, Y., Katoh, K., Ookawara, S., Fujioka, M.,
Guillemot, F., and Kageyama, R. (2004). Hes genes regulate size,
shape and histogenesis of the nervous system by control of the timing
of neural stem cell differentiation. Development 131, 5539–5550.
Hirata, H., Tomita, K., Bessho, Y., and Kageyama, R. (2001). Hes1 and
Hes3 regulate maintenance of the isthmic organizer and development
of the mid/hindbrain. EMBO J. 20, 4454–4466.
Hirata, H., Yoshiura, S., Ohtsuka, T., Bessho, Y., Harada, T., Yoshi-
kawa, K., and Kageyama, R. (2002). Oscillatory expression of the
bHLH factor Hes1 regulated by a negative feedback loop. Science
298, 840–843.
Hollnagel, A., Oehlmann, V., Heymer, J., Ruther, U., and Nordheim, A.
(1999). Id genes are direct targets of bone morphogenetic protein in-
duction in embryonic stem cells. J. Biol. Chem. 274, 19838–19845.
296 Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc.
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube

Page 15

Ishibashi, M., Moriyoshi, K., Sasai, Y., Shiota, K., Nakanishi, S., and
Kageyama, R. (1994). Persistent expression of helix-loop-helix factor
HES-1 prevents mammalian neural differentiation in the central ner-
vous system. EMBO J. 13, 1799–1805.
Ishibashi, M., Ang, S.L., Shiota, K., Nakanishi, S., Kageyama, R., and
Guillemot, F. (1995). Targeted disruption of mammalian hairy and En-
hancer of split homolog-1 (HES-1) leads to up-regulation of neural he-
lix-loop-helix factors, premature neurogenesis, and severe neural tube
defects. Genes Dev. 9, 3136–3148.
Jogi,A.,Persson,P.,Grynfeld,A.,Pahlman,S.,andAxelson,H.(2002).
Modulation of basic helix-loop-helix transcription complex formation
by Id proteins during neuronal differentiation. J. Biol. Chem. 277,
9118–9126.
Kageyama, R., and Ohtsuka, T. (1999). The Notch-Hes pathway in
mammalian neural development. Cell Res. 9, 179–188.
Kageyama, R., Ohtsuka, T., Hatakeyama, J., and Ohsawa, R. (2005).
Roles of bHLH genes in neural stem cell differentiation. Exp. Cell
Res. 306, 343–348.
Kahane, N., and Kalcheim, C. (1998). Identification of early postmitotic
cells in distinct embryonic sites and their possible roles in morphogen-
esis. Cell Tissue Res. 294, 297–307.
Kee, Y., and Bronner-Fraser, M. (2001a). Temporally and spatially re-
stricted expression of the helix-loop-helix transcriptional regulator
Id1 during avian embryogenesis. Mech. Dev. 109, 331–335.
Kee, Y., and Bronner-Fraser, M. (2001b). The transcriptional regulator
Id3 is expressed in cranial sensory placodes during early avian embry-
onic development. Mech. Dev. 109, 337–340.
Lillien, L. (1998). Neural progenitors and stem cells: mechanisms of
progenitor heterogeneity. Curr. Opin. Neurobiol. 8, 37–44.
Lyden, D., Young, A.Z., Zagzag, D., Yan, W., Gerald, W., O’Reilly, R.,
Bader, B.L., Hynes, R.O., Zhuang, Y., Manova, K., and Benezra, R.
(1999). Id1 and Id3 are required for neurogenesis, angiogenesis and
vascularization of tumour xenografts. Nature 401, 670–677.
Martinsen,B.J., and Bronner-Fraser, M. (1998). Neural crest specifica-
tion regulated by the helix-loop-helix repressor Id2. Science 281, 988–
991.
Murata, K., Hattori, M., Hirai, N., Shinozuka, Y., Hirata, H., Kageyama,
R., Sakai, T., and Minato, N. (2005). Hes1 directly controls cell prolifer-
ation through the transcriptional repression of p27Kip1. Mol. Cell. Biol.
25, 4262–4271.
Nakamura, Y., Sakakibara, S., Miyata, T., Ogawa, M., Shimazaki, T.,
Weiss, S., Kageyama, R., and Okano, H. (2000). The bHLH gene
hes1 as a repressor of the neuronal commitment of CNS stem cells.
J. Neurosci. 20, 283–293.
Nakashima, K., Takizawa, T., Ochiai, W., Yanagisawa, M., Hisatsune,
T., Nakafuku, M., Miyazono, K., Kishimoto, T., Kageyama, R., and
Taga, T. (2001). BMP2-mediated alteration in the developmental path-
way of fetal mouse brain cells from neurogenesis to astrocytogenesis.
Proc. Natl. Acad. Sci. USA 98, 5868–5873.
Ohsako, S., Hyer, J., Panganiban, G., Oliver, I., and Caudy, M. (1994).
Hairy function as a DNA-binding helix-loop-helix repressor of Dro-
sophila sensory organ formation. Genes Dev. 8, 2743–2755.
Perk, J., Iavarone, A., and Benezra, R. (2005). Id family of helix-loop-
helix proteins in cancer. Nat. Rev. Cancer 5, 603–614.
Rockman, S.P., Currie, S.A., Ciavarella, M., Vincan, E., Dow, C.,
Thomas, R.J., and Phillips, W.A. (2001). Id2 is a target of the b-cate-
nin/T cell factor pathway in colon carcinoma. J. Biol. Chem. 276,
45113–45119.
Ruzinova, M.B., and Benezra, R. (2003). Id proteins in development,
cell cycle and cancer. Trends Cell Biol. 13, 410–418.
Sasai, Y.,Kageyama,R.,Tagawa,Y.,Shigemoto,R.,andNakanishi,S.
(1992). Two mammalian helix-loop-helix factors structurally related to
Drosophila hairy and Enhancer of split. Genes Dev. 6, 2620–2634.
Selkoe,D.,andKopan,R.(2003).NotchandPresenilin:regulatedintra-
membrane proteolysis links development and degeneration. Annu.
Rev. Neurosci. 26, 565–597.
Stanke, M., Stubbusch, J., and Rohrer, H. (2004). Interaction of Mash1
and Phox2b in sympathetic neuron development. Mol. Cell. Neurosci.
25, 374–382.
Strom, A., Castella, P., Rockwood, J., Wagner, J., and Caudy, M.
(1997). Mediation of NGF signaling by post-translational inhibition of
HES-1, a basic helix-loop-helix repressor of neuronal differentiation.
Genes Dev. 11, 3168–3181.
Takebayashi, K., Sasai, Y., Sakai, Y., Watanabe, T., Nakanishi, S., and
Kageyama, R. (1994). Structure, chromosomal locus, and promoter
analysis of the gene encoding the mouse helix-loop-helix factor
HES-1. Negative autoregulation through the multiple N box elements.
J. Biol. Chem. 269, 5150–5156.
Temple, S. (2001). The development of neural stem cells. Nature 414,
112–117.
Van Doren, M., Bailey, A.M., Esnayra, J., Ede, K., and Posakony, J.W.
(1994). Negative regulation of proneural gene activity: hairy is a direct
transcriptional repressor of achaete. Genes Dev. 8, 2729–2742.
Yan, Y., Bian, W., Xie, Z., Cao, X., Le Roux, I., Guillemot, F., and Jing,
N. (2004). Stat3 signaling is present and active during development of
the central nervous system and eye of vertebrates. Dev. Dyn. 231,
248–257.
Yokota, Y. (2001). Id and development. Oncogene 20, 8290–8298.
Developmental Cell 13, 283–297, August 2007 ª2007 Elsevier Inc. 297
Developmental Cell
Id Sustains Hes1 Expression in Early Neural Tube