DEVELOPMENTAL DYNAMICS 207235-252 (1996)
Expression Patterns of I&, Id2, and Id3 Are Highly
Related But Distinct From That of Id4 During
YALE JEN, KATIA MANOVA, AND ROBERT BENEZRA
Cell Biology Program, Memorial Sloan Kettering Cancer Center, New York, New York 10021
helix-loop-helix (dnHLH) proteins inhibit the ac-
tivities of bHLH transcription factors in diverse
cell lineages (Benezra et al. 119901 Cell 61:49-59;
Christy et a1 119911 Proc. Natl. Acad. Sci. U.S.A.
88:1815-1819; Sun et a1 119911 Mol. Cell Biol. 11:
5603-5611; Riechmann et al.  Nucleic Acids
Res. 22749-755). Currently, there are four mem-
bers in the dnHLH family, Idl, Id2, Id3, and Id4. In
this report, we have performed a detailed compar-
ative in situ hybridization analysis to examine
their expression pattern during post-gastrula-
tional mouse development. Idl, 2, and 3 are ex-
pressed in multiple tissues, whereas Id4 expres-
sion can only be detected in neuronal tissues and
in the ventral portion of the epithelium of the de-
veloping stomach. The regions where Id13 genes
are expressed, such as gut, lung, kidney, tooth,
whisker, and several glandular structures, are un-
dergoing active morphogenetic activities. The ex-
pression patterns of Idl, 2, and 3 overlap in many
organs, except in the tissues derived from primi-
tive gut. In the latter, Id1 and Id3 signals are de-
tected in the mesenchyme surrounding the epi-
thelium, whereas Id2 is expressed within the
epithelium. The difference in the patterns of ex-
pressions of Id13 and Id4 suggest that the domi-
nant negative transcriptional activity of these two
subclasses of the Id family may have different
physiological consequences. o 1996 Witey-Liss, Inc.
The murine dominant negative
Key words: Mouse Embryogenesis, Helix-Loop-
Helix, Id, In Situ Hybridization, Dif-
Eukaryotic transcription factors can be divided into
groups defined by shared sequences and presumed
structural similarities. The group which contains the
basic helix-loop-helix (bHLH) domain constitutes an
evolutionarily conserved family that has been impli-
cated in cell type determination (for review see Wein-
traub et al., 1991). The bHLH proteins form either
homo- or hetero-dimers through the HLH dimerization
domain and bind to their target DNA sequences to reg-
One of the mechanisms that regulates the available
0 1996 WILEY-LISS, INC.
pool of bHLH dimers in cells is the sequestration of the
bHLH proteins by the dominant negative HLH
(dnHLH) proteins (for review see Kadesch, 1992). The
dnHLH proteins possess an HLH motif but do not con-
tain the adjacent stretch of basic amino acids which is
required for DNA binding and therefore form non-func-
tional heterodimers with bHLH proteins. The Id pro-
teins from both mammals, and Xenopus (Wilson and
Mohun, 1995; Zhang et al., 1995) and the Drosophila
extramachrochaetae ( e m ) gene (Ellis, 1994) are the
members of the dnHLH class identified thus far and all
are capable of inactivating bHLH proteins by direct
physical interaction. Currently, there are four mem-
bers in the mammalian Id family: Idl, Id2 (Sun et al.,
19911, Id3 (Christy et al., 19911, and Id4 (Riechmann et
al., 1994). The Xenopus homologs of Id2 and Id3 have
also been identified (Wilson and Mohun, 1995; Zhang
et al., 1995). The mammalian Id genes appear to an-
tagonize the activity of bHLH proteins involved in
many developmental processes, such as myogenesis
(Jen et al., 1992; Kurobayashi et al., 1994), myelopoie-
sis (Kreider et al., 1992), lymphopoiesis (Wilson et al.,
1991), bone morphogenesis (Ogata et al., 1993), kidney
glomerular mesangial cell development (Simonsen, et.
al., 1993), trophoblast development (Janatpour et al.,
1994), as well as in cell cycle progression (Hara et al.,
The embryonic expression pattern of Id1 has been
determined by in situ hybridization analysis (Duncan
et al., 1992, 1994; Evans and O'Brien, 1993; Wang et
al., 1992). Id1 expression appears in many tissues and
its expression roughly correlates with the less differ-
entiated state of the cells. Id2 and Id3 expression pat-
terns have been documented during embryonic neuro-
genesis only (Neuman et al., 1993; Ellmeier et al.,
1992; Nagata and Todokoro, 1994). The Id4 embryonic
expression pattern has not been reported.
In this study, we have examined the expression pat-
tern of all four Id genes by in situ hybridization of
adjacent sections in order to determine precise regions
of overlap or differences in the pattern of expression of
this gene family. Our results suggest that Idl, 2, and 3
Received January 5, 1996; accepted May 24, 1996.
Address reprint requestslcorrespondence to Robert Benezra, Cell
Biology Program, Memorial Sloan Kettering Cancer Center, 1275
York Avenue, New York, NY 10021.
JEN ET AL.
are expressed in multiple tissues at the sites which
undergo active morphogenic activities, whereas Id4 ex-
pression is mainly detected in neuronal tissues as well
as in more differentiated regions of several other tissue
types. In addition to providing a useful catalogue of the
patterns of Idl-4 gene expression, these results also
suggest that the consequences of dominant negative
regulation of transcription factor activity may be dif-
ferent for different Id family members.
Probes for Idl-4 were generated outside of the con-
served HLH domain to ensure specificity (see Materials
and Methods). Northern analysis using each of these
probes confirmed that they were specific for the appro-
priate RNA species (data not shown). Furthermore, in
situ analysis indicated that some tissues only hybridize
to one of the four probes consistent with the idea that
the specificity of the probes was maintained under in
situ hybridization conditions. Sense RNA probes were
used in the in situ hybridization analysis as negative
controls. None of the sense probes showed any signals
above background. However, the Id4 probe gave a
lower signalhoise ratio compared to the other three
probes. Note that a summary of the expression pattern
of all four Id genes is presented in Table 1.
During gastrulation, all of the Id genes, except Id4,
are expressed in a specific manner (Jen et al., unpub-
lished data): Id1 and 3 are mainly detected in the inner
cell mass derived tissues, whereas Id2 expression ap-
pears in the trophoectodermal derivatives. The overall
postgastrulational expression pattern of Id genes de-
scribed in this study is as follows: Id1 (Wang et al.,
1992; Duncan et al., 1992; Evans and O'Brien, 1993;
this work), Id2 and Id3 are expressed in multiple tis-
sues, and the expression patterns among them exhibit
extensive overlap (Fig. 1.1-4). For example, the signal
of these three Id genes can be detected in the mandib-
ular arch, somites, and in the mesenchyme surround-
ing the dorsal side of the developing stomach. The Id4
signal can be detected mainly in neuronal tissues and
in the ventral side of the stomach epithelium during
early stages (Fig. 1.4). Later in development, several
other tissues also acquire Id4 expression.
Expression in Developing Cartilage and Bone
Embryonic skeletal development is composed of
three distinct lineages (for a review see Erlebacher et
al., 1995): the craniofacial skeleton is derived from the
neural crest cells; the somitic derived sclerotome gen-
erates both the axial skeleton and the lateral plate
mesoderm. The latter will later form the appendicular
skeleton. Idl, 2, and 3 are expressed during skeletoge-
nesis in all three lineages, and their expression pat-
terns are very similar (Fig. 2A-0). By 8.5 dpc, all three
Id mRNAs can be detected in the ridges of the unclosed
head folds and the surrounding cranial mesenchyme,
due presumably to the migrating neural crest cells
(data not shown). By 9.5 dpc, the expression of the
three Id genes is detected in the branchial arches (data
not shown), and the expression persists (Fig. 1.1-3) as
more craniofacial structures are being developed. The
craniofacial skeletal structures become more distinct
at 14.5 dpc and the messages of the three Id genes can
be found in many sites undergoing chondrogenesis or
cartilage formation, such as the otic capsule, nasal sep-
tum, and many cranal bones (Fig. 2K,L,M, and N). No
significant level of Id4 can be detected in the craniofa-
cia1 skeletal structures (Fig. 2K,O).
During axial skeleton development, Id1 (Wang et al.,
1992; Evans and OBrien, 1993; this work), Id2, and Id3
signals are first observed in the 10.5 dpc somites. The
somitic expression of these Id genes is mainly located in
the sclerotome, and very weak signals appear in the
dermatome. The intervening myotome is devoid of any
Id signals (data not shown) (for Idl, see Wang et al.,
1992; Evans and OBrien, 1993; this work). In the 11.5
dpc sclerotome, the expression of all three Id genes ap-
pears in the rostral half which contains less cells (Fig.
2A-D). Later, the rostral sclerotome will give rise to the
center of the vertebrae, which also expresses the same
Id genes (data not shown). The more densely packed
caudal sclerotome, which later forms the intervertebral
disk, is negative for Id messages. From 12.5 dpc, the
perichondria surrounding the developing ribs also ex-
press Idl, 2, and 3. At later stages, both the ribs and the
vertebrae acquire some weak Id4 expression (data not
shown). No significant Id4 message can be detected in
the sclerotome (Fig. 2A,E) and its derivatives.
Id expression in the appendicular skeletal develop
ment is similar to that of the axial skeletal lineages:
Idl, 2, and 3 are first expressed in the limb buds be-
tween 9.5 to 10.5 dpc (data not shown). By 12.5 dpc, the
expression of the three Id genes can be detected in both
(1) the interdigital mesenchyme, which later under-
goes apoptosis (Hammar and Mottet, 1971) in order to
remove the intervening web between the digits, and (2)
in the blastemal condensations where the chondrifica-
tion has just started (Rugh, 1990) (Fig. 2P-S). Later,
all four Id genes are expressed in the hypertrophic car-
tilage and the newly forming bony diaphysis of many
long bones (Fig. 2FJ) and digits in 16.5 dpc embryos.
However, the level of Id4 expression is just slightly
above background (Fig. 2F,J).
Craniofacial and trunk dermis are derived from neu-
ral crest and the dermatome portion of the somite, re-
spectively. Low levels of Id1 (Evans and OBrien, 1993;
this work), Id2, and Id3 are expressed in the developing
dermis (data not shown).
Expression in the Epithelial-Mesenchymal
Interacting Craniofacial Structures
The developing whisker follicle and tooth are both
regulated by the sequential and reciprocal interactions
between the cranial neural crest-derived mesenchyme
and the facial or oral epithelial placodes (Lumsden,
1988; Hardy, 1992; Panaretto, 1993). The early mor-
phological events of tooth development are the invagi-
EXPRESSION OF Id FAMILY IN MOUSE
TABLE 1. Expression of Each Id Gene i n the 1 1 . 5 - t o 16.5-Day Mouse Embryo"
Id1 Id2 Id3
Septum and valves
Epith. of otic vesicle
sur. mes. +
sur. mes. +
Epith. and sm. mus. +
selected neurons +
stir. mes. +
selected neurons +
Mature neurons -
Mature neurons +
Mature neurons -
Nasal processes 11.5D
Mes. adjacent to
supporting cells 5
supporting cells t
Postmitotic neurons +
supporting cells t
Dental papilla mes.
dermal sheath, and
outer root sheath
Inner root sheath
aEpith. = epithelium; Mes = mesenchyme; Sm.mus. = smooth muscle; Sur. mes. = surrounding mesenchyme. Relative level
of mRNA expression is indicated by: - = no expression; 5 = trace expression; + = mRNA is expressed. ND = not
JEN ET AL.
Fig. 1. The expression of four Id genes at 11.5 dpc mouse embryo.
The brightfield panel (top) shows the sagittal section of 11.5 dpc embryo,
and the darkfield panels (1-4) show the expression pattern of each Id
gene as the number on the top-left corner indicated. The expression of
Idl, 2, and 3 appears in many tissues, whereas Id4 signal can be mainly
detected in the neural tissues. Arrow in the brightfield and in the darkfield
panel which hybridized to Id4 indicates the mantle layer of the spinal cord.
Magnification: x 25. Dr: dorsal root ganglion; Gu: gut; Lu: lung bud; Ma:
mandible; Me: mesencephalon; Mt: metencephalon; My: myelenceph-
alon; Np: nasal process; Oe: olfactory epithelium; 0s: optic stalk; Ra:
Rathke’s pouch; So: somite; Sp: spinal cord; Te: telencephalon.
nation of the local thickened epithelium, formation of
the tooth bud, and the subsequent condensation of the
surrounding mesenchyme. This “bud stage” occurs
around 12-13 dpc. At 12.5 dpc, both Id1 and Id3 ex-
pression are localized to a patch of toothbud epithelium
(data not shown). Id2 as well as some low levels of Id1
and Id3 transcripts are detected in the condensed mes-
enchymal cells (Cdm) (data not shown). No Id4 signals
can be detected in either structures at this stage (data
not shown). At the cap stage (-14.5 dpc), the toothbud
epithelium further invaginates and gives rise both to
the inner (Iee) and outer (Oee) enamel epithelium, and
EXPRESSION OF Id FAMILY IN MOUSE
to the cells of the stellate reticulum (Sr). The latter is
believed to provide space and support for the develop-
ing crown. The majority of the Id1 message is detected
in the Iee and Oee, with some faint signals in the Cdm
and Sr (data not shown). Relatively intense Id3 tran-
scripts are detected in all four structures (data not
shown). The Sr exhibits some faint Id2 and Id4 mes-
sages, whereas Cdm expresses low level of Id2 (data not
shown). By 16.5 dpc, the tooth development progresses
into the bell stage: the Cdm first differentiate into den-
tal papilla mesenchyme (Dpm), and subsequently, its
most superficial cell layer becomes the odontoblast
(Ob) layer; meanwhile, the Iee form the ameloblast
(Ab) layer. Odontoblast and ameloblast secrete dentin
and enamel matrices, respectively. Various levels of all
four Id genes are expressed in the Dpm (Fig. 3.1-4),
whereas no Id signals can be detected in the Ob. High
levels of Id1 and Id3 transcripts are detected in the Ab,
whereas signals of all four Id genes are located in the
region of the Sr adjacent to the Ab. The Ab in the AbISr
border also express Id4.
Mouse whisker development has been described in
detail by Hardy and co-workers (Davidson and Hardy,
1952; for review see Hardy, 1992; Panaretto, 1993):
epidermal placode formation starts from the down-
growth of a small group of basal epithelial cells into the
subjacent condensing mesenchyme, which has induc-
ing activity (stage 1, occurs around 12.5 dpc). At this
stage, Idl, 2, and 3 expression can be detected in the
condensing mesenchyme underneath the epidermal
placodes (data not shown). No Id4 signals are detected.
Later, the condensed mesenchyme responds to the
overlying epidermal signal and becomes dermal papil-
lae. The dermal papillae further induce the adjacent
epithelial cells to form hair matrix cells, which divide
rapidly, and move upward. They subsequently differ-
entiate into inner and the outer root sheaths. The ex-
pression pattern of Idl, 2, and 3 is quite similar during
these later stages: by 14.5 through 16.5 dpc, all three Id
genes are expressed in the outer root sheath, dermal
papillae, and the surrounding dermal sheath region
(Fig. 4A, C, and data not shown), although the signal of
Id3 is much weaker than the other two Id genes (data
not shown). For the first time, Id4 expression now can
be detected at 16.5 dpc only in the inner root sheath
(Fig. 4B,D). Figure 4E illustrates the expression pat-
tern of all four Id genes during embryonic whisker de-
Expression i n Developing Viscera
The expression of the Id genes is found in many or-
gans during their development, Detailed analysis will
be presented here for the digestive tract, lung, kidney,
The mammalian digestive system is derived from the
endodermal structure, the primitive gut: The stomach,
which is the caudal-most part of the foregut, becomes
dilated at 11 dpc, At the same time, a clockwise rota-
tion occurs and locates the stomach on the left side of
the midline (Kaufman, 1992). During and following the
rotation, the dorsal part of the stomach grows faster
than the ventral part and eventually gives rise to the
curvature of the stomach. From 10.5 to 11.5 dpc, vari-
ous levels of Idl, Ida, and Id3 expression can be de-
tected in both the endoderm and surrounding mesen-
chyme in the dorsal part of the stomach (Fig. 5A,B and
data not shown), whereas a low level of Id4 expression
is found in the ventral part of the stomach endoderm
(Fig. 5C,D). The expression pattern of all four Id genes
persists through 14.5 dpc, except the expression of Idl,
2, and 3 is more concentrated on the surrounding mes-
enchyme than in the epithelial cells (data not shown).
As the stomach becomes further developed, the expres-
sion of Idl, 2, and 3 can be detected in both the endo-
derm-derived glandular mucosal layer and in the out-
most presumptive smooth muscle layer, which is
derived from the surrounding mesenchyme (data not
shown). Id4 expression remains limited to the glandu-
lar mucosal layer of the ventral side (data not shown).
The expression of Idl, 2, and 3 can also be detected
during the development of the mid and hind gut. At
10.5 dpc, both the Id1 and Id3 messages are found in
the mesenchyme surrounding the gut epithelium,
whereas the Id2 signal is located in the gut endoderm
(data not shown). By 14.5 dpc, the mesenchyme be-
comes progressively segregated into two distinct areas:
cells located at the periphery become oriented circu-
larly to give rise to the muscle layers, and the subset of
cells confined close to the epithelium arrange parallel
to it (Simon-Assmann and Kedinger, 1993). The latter
plays a critical role in forming the basement mem-
brane, which is very important in normal intestinal
development. At this stage, the peripheral presumptive
muscle cells exhibit expression of Idl, 2, and 3 (Fig.
5E-H). Signals for Id1 and Id3 also can be found in the
mesenchymal cells located adjacent to the intestinal
epithelium (Fig. 5E,F,H), whereas Id2 expression can
still be detected in the gut epithelium (Fig. 5E,G). No
Id4 signal can be detected in any of these structures at
this stage (data not shown). At 16.5 dpc, as villus for-
mation progresses, very little expression of Idl, 2, and
3 can be detected in the intestine (data not shown).
Embryonic lung development starts from an out-
growth of the foregut endoderm, the so-called primitive
trachea. This structure gives rise to two main bronchi,
and together with the surrounding mesenchyme forms
the lung buds. The epithelial bronchus undergoes
branching morphogenesis from the bronchial tree,
whereas the surrounding mesenchyme later develops
into the connective structure of the lung (Rugh, 1990).
The expression pattern of the Id genes in the lung has
some similarity to that of the mid and hind gut: both
Id1 (Wang et al., 1992; this work), and Id3 expression
are located in the surrounding mesenchyme (Fig. 6B,E
JEN ET AL.
EXPRESSION OF Id FAMILY IN MOUSE
and data not shown) and Id2 signal is detected in the
bronchial epithelium (data not shown). Later, as the
lung develops further, Id2 mRNA is detected only in
the terminal bronchioles, whereas earlier formed bron-
chi lack Id2 signal (Fig. 6A,D). The expression of Id4
can only be found in very limited areas of the main
bronchi after 14.5 dpc (Fig. 6C,F).
Metanephric kidney development starts from 11.5
dpc when the nephric duct derived ureteric bud invades
the blasterna of metanephric mesenchyme (Mm). The
subsequent reciprocal induction causes Mm to con-
dense and the ureteric bud grows and bifurcates. The
branched ureteric bud later gives rise to the collecting
system. The aggregated Mm undergoes epithelializa-
tion and gives rise to the entire filtration system, the
nephron: the distal end forms the primitive tubule
which fuses with the collecting duct, whereas the prox-
imal tip, together with the invaginating capillaries,
form the glomerulus, which is encapsulated in the
Bowman’s capsule (Bard et al., 1994). At 12.5 dpc, Id1
Fig. 2. The expression of Id genes during skeletal and limb develop-
ment. Brightfield (A) and corresponding darkfield (WE) photomicro-
graphs of sagittal sections of sclerotomes through the midtrunk region of
11.5 dpc mouse embryo. Brightfield (F) and corresponding darkfield
(G-J) photomicrographs of longitudinal sections of the radius in the fore-
limb of 16.5 dpc mouse embryos. Brightfield (K) and corresponding dark-
field ( L a ) photomicrographs of transverse sections through the head
region of 14.5 dpc embryos show various craniofacial skeletal structures.
Brightfield (P) and corresponding darkfield (Q-T) photomicrographs of
longitudinal sections through the hind limb of 12.5 dpc embryos. B, G, L,
and 0 hybridized to the Id1 probe; C, H, M, and R hybridized to the Id2
probe; D, I, N, and S hybridized to the Id3 probe; E, J, 0, and T hybridized
to the Id4 probe. Magnification: A-E, x 100; F-T, x 50. The hybridization
signals of Id1 , 2, and 3 can be detected in the rostral half of the sclero-
tomes (A-D) in the 11.5 dpc embryos. This structure later gives rise to the
vertebrae which also expresses same sets of Id genes (data not shown).
The caudal half of the sclerotomes do not exhibit any Id signals. No Id4
signal can be detected in the sclerotome (A,E). During limb development,
hybridization signals of Idl, 2, and 3 can be seen in the developing
skeletal structures, such as the long bones and the digits (data not
shown). By 16.5 dpc, the expression of Id1 , 2, and 3 can be found in the
hypertrophic cartilage region (arrowheads in F-I) and in the bony diaphy-
sis of the radius in the forelimbs. Very low levels of Id4 signals can also
be detected in these structures at this stage (F,J). In the 14.5 dpc em-
bryos, the expression of Id1 ,2, and 3 can also be detected in many neural
crest derived craniofacial skeletal structures (K-N), such as the ossifica-
tion center near the palatal shelf and Meckel’s cartilage, and the cartilage
primordium of the sphenoid bone and hyoid bone. No Id4 signal can be
detected in these structures at this stage (K,O). In the developing limb,
besides skeletogenesis, the expression of Idl, 2, and 3 can also be seen
in the interdigital mesenchymal cells of the 12.5 dpc embryos (P-S).
These cells later undergo apotosis in order to remove the interventing
web between the digits. The open arrows in P-S indicate the blastemal
condensation where chondrification is in progress. No Id4 signal is de-
tected in these structures (P,T). Bod: bony diaphysis; Cst: caudal scle-
rotomes; Dr: dorsal root ganglion; Ey: developing eye; Hc: hypertrophic
cartilage; Hyb: cartilage primordium of hyoid bone; Idm: interdigital mes-
enchyme; Mkc: Meckel’s cartilage; Mtp: upper molar tooth primordium;
Oe: olfactory epithelium; Omk: ossification in the mandible around Meck-
el’s cartilage; Ops: ossification center in the lateal part to the palatal shelf;
Ps: palatal shelf; Sp: spinal cord; Spb: cartilage primordium of sphenoid
bone; Rst: rostral sclerotomes.
(Duncan et al., 1994; this work), 2, 3, and very low
levels of Id4, can be detected in the condensed Mm,
whereas the ureteric bud is devoid of any Id signals
(data not shown). The signals of all four Id genes are
also detected in the renal pelvis. By 14.5 dpc, the prim-
itive glomeruli are seen for the first time, and both Id1
(data not shown) and Id3 (Fig. 6H,K) are expressed in
the capillary derived glomerular tuft, but not Id2 (Fig.
6G,J). Both the developing collecting tubules and con-
densed Mm exhibit the expression of Id1 (Duncan et al.,
1994; data not shown), Id2, and Id3 (Fig. 6G,H,J,K). No
Id4 hybridization signal can be detected in any struc-
tures in the developing kidney at this stage (Fig. 61,L).
The less condensed mesenchyme, which will give rise
to the connective tissues of the kidney, does not exhibit
any expression of the Id genes (Fig. 6G-L). The expres-
sion pattern of all four Id genes is essentially the same
at 16.5 dpc, except only Id1 and Id2 are expressed in the
convoluted tubules (data not shown).
The development of the four-chambered heart from
the much simpler tubular structure requires many cel-
lular changes. The separation of different chambers is
achieved by the formation of endocardial cushions,
which later develop into septa and valves. The majority
of these structures are formed from the endocardial
cushion tissue (Ect) . Ect are mesenchymal cells derived
from myocardial endothelium, which originates from
the splanchnic mesoderm (Noden, 1991). In developing
chick, the caudal cranial neural crest cells are known
to migrate into the outflow tract of the developing
heart and participate in closure of the outflow septum
(Kirby et al., 1983). Id1 (Wang et al., 1992; Evans and
OBrien, 1993; this work), 2, and 3 expression can be
detected in the developing heart, in overlapping ex-
pression patterns (Fig. 7.1-3). No Id4 signal is found
during heart development (Fig. 7.4). At 11.5 dpc, when
the transformation of myocardial endothelium into Ect
has occurred in both the atrioventricular canal and the
bulbous arteriosus, moderate levels of signals of the
three Id genes can be found in those regions, and more
intense signals are detected in the endocardial cushion.
Furthermore, moderate levels of the three Id messages
are also found in the outflow tract. The same pattern of
Id expression in the heart valves persists through 16.5
dpc, which is the last stage of our analysis (data not
Expression During Yascularization
Murine blood vessel development is achieved via two
different mechanisms: vasculogenesis and angiogenesis
(Coffin et al., 1991; reviewed by Noden, 1989). Vascu-
logenesis is defined as de novo formation of blood vessels
from free angioblasts, which arise from mesoderm. The
development of the larger vascular network, such as the
dorsal aorta, is accomplished by vasculogenesis. Angio-
genesis means formation of blood vessels by sprouting
from the pre-existing vessels. This process occurs dur-
JEN ET AL.
Fig. 3. The expression of Id genes in the developing teeth. Brightfield
(top) and the corresponding darkfield (14) photomicrographs of the
transverse sections through the developing tooth of 16.5 dpc mouse
embryos. The number in the darkfield photomicrographs represents the
different Id genes. Magnification: x 200. The arrowheads indicate the
ameloblast layer (Ab), whereas the open arrows indicate the odontoblast
layer (Ob). The hybridization signals of all four Id genes can be seen in
the dental papilla mesenchyme (Dpm) and stellate reticulum (Sr). Ex-
pression of Id1 and Id3 can be seen in the ameloblast layer (see the
arrowheads in 1 and 3). Odontoblast layers (Ob) do not exhibit any sig-
nificant expression of all four Id genes. Ab: ameloblast layer; Dpm: dental
papilla mesenchyme; Ob: odontoblast layer; Sr: stellate reticulum.
ing formation of the blood vessel network of many or-
gans, such as brain, kidney, and limb buds. At 10.5 and
11.5 dpc, the expression ofboth Id1 (Evans and O’Brien,
1993; this work) (Fig. 8A,E) and Id3 (data not shown)
can be detected in the blood vessels of meninges sur-
rounding the developing brain whereas neither Id2
(Fig. 8B,F) nor Id4 (data not shown) expression can be
detected in the these structures. Later, the expression
of Id1 (Duncan et al., 1992; Evans and O’Brien, 1993;
this work) and Id3 can also be found in the vasculature
of choroid plexus in both third and fourth ventricles of
the brain (Jen et al., unpublished data). The choroid
plexi also exhibit low level of Id2 expression. By 16.5
dpc, both Id1 (data not shown) and Id3 (Fig. 8C,G) sig-
nals can be found in the developing blood vessels of the
brain, and in the hyaloid plexus of the developing eye.
EXPFtESSION OF Id FAMILY IN MOUSE
Fig. 4. Expression of Id genes in the developing whisker follicle.
Phase contrast brightfield (A,B) and corresponding darkfield (C,D) pho-
tomicrographs of oblique sections through the snout of 16.5 dpc mouse
embryos. C hybridized to Id2 probe and D hybridized to Id4 probe. Mag-
nification: x 200. Hybridization signal of Id2 can be detected in the dermal
sheath (Ds) (double arrows) surrounding the maturing follicle, in the der-
ma1 papilla (Dp), and in the outer root sheath (Or) (A,C). The Id4
expression can be only seen in the inner root sheath (Ir) (B,D). E is a
schematic illustration of the expression of each Id gene in the developing
whisker follicle. Expression of Id1 and Id3 (data not shown) can be seen
in the similar structures as Id2 (A,C) and the expression of Id3 is relatively
weaker than the other Id genes. The Id4 signal is only detected in the
inner root sheath (Ir). Dp: dermal papilla; 0s: dermal sheath; Hr: hair; Ir:
inner root sheath; Or: outer root sheath.
Again, no Id2 (data not shown) and Id4 (Fig. 8D, H)
signals are found in these structures. These results, as
well as the data in the previous section describing Id
gene expression in the capillary-derived glomerular
tuft in the developing kidney, are consistent with the
notion that Id1 and Id3, but not Id2 and Id4, may be the
major Id genes involved in angiogenesis.
By 9.5 dpc, intense signals of Idl, 2 and 3 can be
detected in the mesenchyme surrounding the dorsal
aorta, and the expression persists for the next few days
(data not shown). However, the involvement of Id gene
expression in the development of larger vascular net-
works, such as the dorsal aortae, needs further inves-
Fig. 5. The expression of Id genes in the developing digestive tract.
Brightfield (A,E) and the corresponding darkfield (B-H) photomicro-
graphs of both the sagittal sections of developing stomach (A-D) of 11.5
dpc mouse embryos, and the transverse sections of the hindguts (E-H)
o f the 14.5 dpc mouse embryos. B and G hybridized to Id2 probe; C and
H nywdiZ@l t b Id3 pt'ctbe; D hyblidited to Id4 probe and F hybridized to
Id1 probe. Magnification: x 200. In A-0 the dorsal side of the embryo is
located on the right, whereas the ventral side is on the left. The arrow-
heads in A and D indicate the ventral part of the stomach endoderm. le:
intestinal endoderm; Mae: mesenchyme adjacent to endoderm; Mnt: met-
anephric tubules; Pml: presumptive muscle layer; St: stomach.
EXPRESSION OF Id FAMILY IN MOUSE
Fig. 6. The expression of Id genes in the developing lung and kidney.
and the corresponding darkfield (LF,J-L) photo-
micrographs of transverse sections of both the developing lung of 16.5
dpc embryos (A-F), and the developing kidney of 14.5 dpc embryos
(G-L). Magnification: x200. D and J hybridized to Id2 probe; E and K
hybridized to Id3 probe; F and L hybridized to Id4 probe. The arrows in
A-F indicate the bronchioles. During lung development, the expression of
Id1 (data not shown) and Id3 (B,E) can be detected in the mesenchymal
cells surrounding the main bronchi and the smaller terminal bronchi/
bronchioles, whereas the Id2 signal is only concentrated in the newly
emerging terminal bronchi/bronchioles, but not the main bronchus (A,D).
Lower level of Id4 signal (open arrows in C and F) can be detected in the
main bronchus at this stage. In the developing kidney of 14.5 dpc em-
bryos, the expression of Id1 (data not shown), Id2 (G,J), and Id3 (H,K)
can be seen both in the condensed rnesenchymal cells of the cortex, and
in the developing collecting tubules. The expression of Id1 (data not
shown) and Id3 are also detected in the glomerular tuft of the glomerulus
(arrowheads in H and K), whereas no Id2 signal can be found in these
structures (arrowhead in G and J). No significant level of Id4 signal can be
detected in the developing kidney at this stage (data not shown). Bc:
Bowman’s capsule; Bco: bronchiole; Crn: cortical condensed mesen-
chyme; Ct: collecting tubule: Mb: main bronchus.
Expression in Glandular Structures
Different Id signals can also be detected in several
developing endo- and exocrine glands. At 14.5 dpc, both
Id1 (Fig. 9A,E) and Id3 (Fig. 9C,G) signals can be de-
tected in the rnesenchymal cells surrounding the epi-
thelium in the developing salivary gland. The mes-
sages of Id2 (Fig. 9B,F) and Id4 are located in the
epithelial structures, with Id4 being expressed in a
more restricted fashion (Fig. 9D,H). The expression of
all four Id genes can be detected in the developing thy-
roid gland at 14.5 dpc (Fig. 9J-L and data not shown)
although the levels of Id2 and Id4 are lower than that
of Id1 and Id3. The parathyroid gland only exhibits the
expression of Id4 (Fig. 9K,L). In the nose, both Id2 and
JEN ET AL.
Fig. 7. The expression of Id genes in the developing heart. Brightfield
(top) and the corresponding darkfield (1-4) photomicrographs of sagittal
sections through the developing heart of 11.5 dpc mouse embryos. The
number in each darkfield photomicrograph represents the different Id
gene. Magnification: x 100. Both the endocardial cushion and outflow
track express Idl, 2, and 3 (1, 2, 3). No Id4 signal is seen in the devel-
oping heart (4). Ecc: endocardial cushion: Ot: outflow tract.
Id4 signals are detected in the developing nasal glands
(data not shown). At 12.5 dpc, the expression of Idl, 2,
and 3 are also detected in the developing pancreas
(data not shown). In the developing pituitary gland, the
expression of all four Id genes can be detected in the
infundibulum (Jen et al., unpublished data), which
give rise to the pars neuralis, whereas the signals of
Idl, 2 and 3 are found in Rathke’s pouch (see Fig. 11,
from which the adenohyphysis will be derived.
We have examined in detail the temporal and spatial
expression pattern of the four mouse Id genes. The ex-
pression of the Idl, 2,3 genes are highly related to one
Fig. 8. The expression of id genes during vascularization. The bright-
field (A-D) and the corresponding darkfield (E-H) photomicrographs of
both the sagittal sections through the developing hindbrain (A,B,E,F) of
11.5 dpc mouse embryo, and the transverse sections through the devel-
oping eye (C,D.G.H) of 14.5 dpc mouse embryo. Magnification: x 200. E
hybridized to Id1 probe; F hybridized to Id2 probe; G hybridized to Id3
probe; H hybridized to probe 164. The arrows in all the panels indicate the
developing blood vessels. Both Id1 (A, E and data not shown) and Id3 (G,
G and data not shown) are expressed in the blood vessels of the devel-
oping brain and in the developing vasculature at the hyaloid plexus of the
developing eye. Neither Id2 (B, F, and data not shown) nor 164 (D, H, and
data not shown) is expressed in those structures. The bright curve line in
0 and H represents an artifact from the pigment of the retina. The signals
on the right side of E and F come from different neuronal cells of the
brain. Fv: fourth ventricle: Hp: hyaloid plexus; Inl: inner neural layer; Ln:
lens; Onl: outer neural layer; 0s: optic stalk.
JEN ET AL.
EXPRESSION OF Id FAMILY IN MOUSE
another and appear in many sites associated with ac-
tive mesenchymal-epithelial interactions during orga-
nogenesis. Id4 expression, on the other hand, only ap-
pears in very limited areas (e.g., neuronal tissues and
the ventral side of the developing stomach) during
early development, and its expression later can be de-
tected in several tissues which are in more advanced
stages of differentiation. These results suggest that the
activity of Id4 may be distinct from the other members
of the Id family (see below). Table 1 catalogues the
expression of the different Id genes in various tissues/
We can divide the developing organs which express
different Id genes into three categories based on their
origin: (1) mesoderm-derived organs, such as the devel-
oping kidney, heart, blood vessels, and bone. In most of
these structures, the expression pattern of Idl, 2, and 3
is quite similar. The expression of the Id genes is par-
ticularly prominent in many sites where morphogenic
activity occurs, such as epithelization of the mesen-
chyme (kidney, blood vessels) (Ekblom, 1989), as well
as mesenchymalization of the epithelium (heart) (No-
den, 1991); (2) organs developed from ectoderm-meso-
derm interactions: tooth, whisker, and nasal gland are
examples in this category. All four Id genes are ex-
pressed in different stages of tooth and whisker devel-
opment, whereas only Id2 and Id4 are expressed in the
developing nasal gland. The tooth (Hardy, 19921, whis-
ker (Lumsden, 1988), craniofacial skeletal structures
(Erlebacher et al., 1995), thyroid gland (Bockman and
Kirby 19841, and part of the heart (Kirby et al., 1983)
are derived from and influenced by the cranial neural
crest cells; (3) organs formed from mesoderm-endoderm
interactions: most of the organs in this category arise
from the primitive gut, such as gut, lung, salivary
gland, thyroid, parathyroid, and pancreas. The signals
of Id1 and Id3 are generally detected in the mesenchy-
ma1 cells which surround the endodermal epithelium
(i.e., gut, lung, and salivary gland), whereas Id2 mes-
sage is located in the epithelium. Id4 signal can also be
detected in the epithelium of several glandular struc-
tures (salivary, thyroid, and parathyroid). The expres-
sion of Id4 in the ventral epithelial cells of the stomach
Fig. 9. The expression of Id genes in the developing salivary gland
and in thyroidlparathyroid gland. Brightfield (A-D,I,K) and the mrre-
sponding darkfield (E-H,J,L) photomicrographs of transverse sections
through both the salivary gland (A-H) of 14.5 dpc mouse embryos, and
thyroid/parathyroid gland (CL) of 16.5 dpc mouse embryos. E hybridized
to Id1 probe: F hybridized to Id2 probe; G and J hybridized to Id3 probe;
H and L hybridized to Id4 probe. Magnification: x200. Arrows in A-H
indicate the epithelial structure of the salivary gland. The arrowheads in
A. C, E, and G indicate the hybridization signals of either Id1 or Id3 which
is located in the mesenchymal cells surrounding the epithelium in the
developing salivary gland. Expression of Id2 can be seen ubiquitously in
the epithelial structure (B and F), whereas Id4 signal can only be detected
in a more limited region of the epithelium (D and H, where the open
arrows indicate the epithelial structures which are devoid of any Id4 sig-
nals). All four Id genes (I-L and data not shown) hybridize to the thyroid
gland, but only Id4 message can be detected in the parathyroid gland (K
and L). Pt: parathyroid gland: Ty: thyroid gland.
is of particular interest, because only a few genes are
expressed in restricted pattern along the dorsal-ven-
tral axis during the gut development (Lyons et al.,
1995; Echelard et al., 1993). Furthermore, the onset of
Id4 expression in the gut coincides with the rotation of
the stomach during its development. Lung, kidney, and
salivary glands undergo a process termed branching
morphogenesis, as the tubular epithelia grow and
branch through the surrounding mesenchymal cells. It
is well documented that both the surrounding mesen-
chyme, as well as the epithelial cells, play important
roles in this morphogenetic process (Nogawa and Mi-
zuno, 1981; Bard et al., 1994; Roman et al., 1991). The
differential expression pattern of the Id genes in either
the epithelial or the mesenchymal component of these
developing structures suggests they may play some
role in this process.
The expression pattern of different members of the Id
gene family shares some similarity with that of mem-
bers of other multi-gene families, such as PAX (re-
viewed by Stuart et al., 1994) and Bone Morphogenetic
Protein (BMP) (Fukagawa et al., 1994; Jones et al.,
1991; Lyons et al., 1989,1990, 1995; Wall et al., 1993).
BMP-2 is worth special mention: BMP-2 is a member of
the bone morphogenetic proteins (BMP) family which
is capable of inducing cartilage and bone formation in
bioassays. It has been demonstrated that BMP-2 can
enhance the expression of Id1 in osteoblastic cells
(Ogata et al., 1993). Furthermore, ectopic expression of
BMP-2 can convert the C2C12 myoblasts, which are
commonly used as a myogenesis model system, into
osteoblasts (Katagiri et al., 1994). The level of Id1 mes-
sage is also elevated during this process. Comparison of
the expression pattern of different Id genes and BMP-2
during embryogenesis shows that the expression of Id1
and Id3 overlaps with BMP-2 in many tissues: during
bone and cartilage formation (Lyons et al., 1989), whis-
ker and tooth development (Lyons et al., 19901, and gut
development (Lyons et al., 1995). This suggests that
BMP-2 may influence the expression of Id1 (and pos-
sibly Id2 and Id3) during embryonic skeletogenesis.
Recently, Tournay and Benezra (Tournay and Ben-
ezra, 1996) have determined that Id1 activation in re-
sponse to serum is mediated in part by the early growth
response gene Egr-1 in tissue culture cells. The tran-
scription of Egr-1, which is a zinc finger-containing
transcription factor, is itself induced rapidly after se-
rum induction (Sukhatme et al., 1988) and newly syn-
thelized Egr-1 targets the Id1 promoter for activation
in response to mitogen stimulation. In the periosteal
region of developing bone and cartilage, the embryonic
expression pattern of Egr-1 (McMahon et al., 1990) and
Id1 (as well as Id2 and Id3) overlaps. This latter result
suggests that Egr-1 may affect Id1 expression during
embryogenesis consistent with the tissue culture anal-
Based on our current understanding, the dnHLH
proteins exert their activity by sequesting bHLH pro-
teins. Among the known bHLH proteins, only a few
JEN ET AL.
exhibit an expression pattern which partially overlaps
that of the Id genes. Examples are E2A (Roberts et al.,
19931, M-twist (Wolf et al., 1991), Purdzis (Burgess et
al., 19951, Sclerazis (Cserjesi et al., 19951, and HES-1
(Sasai et al., 1992). M-twist is noteworthy since like
Idl, M-twist can inhibit muscle differentiation in tissue
culture (Hebrok et al., 1995). M-twist expression can be
detected in the sclerotorne, mesenchymal cells in the
head, branchial arches, and limb bud, and in cardiac
cushions. The expression pattern of M-twist in the
somite, limb bud, and craniofacial structures is mutu-
ally exclusive with respect to the expression pattern of
myogenic factor, myfs (Hebrok et al., 1995; Ott et al.,
1991; Wolf et al., 1991) analogous to the relationship
between Id1 and myE in the same structures (Wang et
al., 1992; Evans and O'Brien, 1993). Based on these
results, it will be of interest to further investigate the
possible relationship between Id and M-twist. Interest-
ingly, the expression of Myf5, as well as Idl, Id2, and
Id3 in branchial arches, occurs early and precedes myo-
genesis. This observation suggests that Idl, which is
also expressed in the brachial arches, can prevent the
myogenic regulatory factors (MRFs) from activating
downstream differentiation genes as well as antago-
nize the expression of MRFs in the somites (see Jen et
al., 1992; Wang et al., 1992).
In neuronal tissues, Id4 expression appears in differ-
entiated neurons, unlike Id1 and Id3 which are found
in mitotically active neuroblasts (Jen et al., unpub-
lished data). In addition, Id4 expression in other tissues
is more limited than that of Idl-3, and is present in
more differentiated regions. If we assume that all of the
Id genes can inhibit other bHLH proteins from binding
DNA, these results would suggest that Id4 may target
a distinct set of proteins leading to different physiolog-
ical consequences. The close similarity in the expres-
sion pattern of Id3 and Idl, on the other hand, suggests
a redundancy between these two genes. The existence
of such a redundancy could be due to the selective ad-
vantage of cumulative function and/or of higher fidel-
ity (Thomas, 1993). The actual functional relationship
among the different members of Id gene family will
need to be resolved by loss-of-function genetic manip-
MATERIALS AND METHODS
Generation of Each Id Specific Clones
All Id gene specific clones were generated by PCR
amplifying the region of each Id gene, which was de-
void of the HLH domain, followed by cloning each Id
specific fragment into pBluescriptKS (Stratagene, La
Jolla, CA). The templates used for generating Idl, 2,
and 3 fragments are clone pMH18AR (Benezra et al.,
1990), Id2 genomic clone (Sun et al., 1991), and
HLH462 phagmid (Christy et al., 19911, respectively.
The total genomic DNA from the tail of the C57BL/
CBA mice was used as template to generate Id4 specific
fragments. The oligonucleotide sequences, which were
used for generating Id4 specific fragments, are derived
from the available Id4 cDNA sequences in Genbank.
The Id1 specific clone, pBS/KS-Idlsp, contains an in-
sert of 350 bp which corresponds to the nucleotides 390
to 742 of the Id1 cDNA. The Id2 specific clone, pBS/
KS-Idasp, contains an insert of 195 bp which corre-
sponds to the nucleotides 1 to 195 of the Id2 cDNA. The
Id3 specific clone, pBS/KS-Id3sp, contains an insert of
270 bp which corresponds to the nucleotides 365 to 635
of the Id3 cDNA. The Id4 specific clone, pBS/KS-Id4sp,
contains an insert of 260 bp which corresponds to the
nucleotides 665 to 945 of the Id4 cDNA.
(-y-35S)UTP labeled riboprobes were used for all in
situ hybridization analysis. Antisense Id1 specific ribo-
probes were generated by linearizing pBS/KSXdlsp
with XbaI and transcribing by T3 RNA polymerase.
The control sense probes can be generated by lineariz-
ing with XhoI and transcribing by T7 RNA poly-
merase. The Id2 specific antisense riboprobes were gen-
erated by linearizing pBS/KSId2sp with EcoRI and
transcribing by T7 RNA polymerase. The control sense
probes were generated by linearizing with B a d 1 and
transcribing by T3 RNA polymerase. The Id2 full
length cDNA antisense riboprobes which generated
from linearizing Id2 cDNA clones, pId2k (Sun et al.,
19911, with XhoI, and transcribing with SP6 RNA poly-
merase, were also used in in situ hybridization analy-
sis. Both Id2 specific and full length probes gave an
identical specific signal in the in situ hybridization
analysis. However, compared to the Id2 specific probes,
the full length Id2 riboprobes gave much stronger sig-
nals and have been used for most of the in situ hybrid-
ization analysis. The Id3 specific probes were gener-
ated by linearizing pBS/KS-Id3sp with Sac1 and
transcribing by T3 RNA polymerase. The control sense
probes were generated by linearizing with EcoRI and
transcribing by T7 RNA polymerase. The Id4 specific
probes were generated by linearizing pBS/KS-Id4sp
with XbaI and transcribing by T3 RNA polymerase.
The control sense probes can be generated by lineariz-
ing with EcoRI and transcribing by T7 RNA poly-
Staging of Mouse Embryos
Embryos obtained from the mating between C57BL/
CBA F1 mice were used for all the in situ hybridization
analysis. The day when the vaginal plugs was detected
was counted as 0.5 days postcoitum (dpc).
In Situ Hybridization
Preparations of embryonic tissue sections, slide hy-
bridization, and the stringency of washes were the
same as described before (Wang et al., 1992). Briefly,
embryos from 4.5 to 8.5 dpc were retained inside the
surrounding decidua, and the embryos of older ages
were removed from the decidua and fixed in freshly
prepared cold 4% paraformaldehyde in phosphate-
buffer saline overnight. For larger embryos, the fixa-
EXPRESSION OF Id FAMILY IN MOUSE
tion were carried out under vacuum to ensure that the
tissues were properly fixed. Tissues were then slowly
dehydrated and embedded into paraffin blocks. Serial
sections (5-7 krn) were collected on glass microscope
"subbed" slides. Slides with attached sections were de-
paraffinized, rehydrated, Proteinase K treated, acety-
lated, and dehydrated. The hybridizations were carried
out at 52°C for > 16 hr in 50% deionized formamide, 0.3
M NaC1,20 mM Tris-HC1 (pH7.4),5 mM EDTA, 10 mM
NaH,PO, (pH8.01, 10% dextran sulfate, 1 X Den-
hardt's, 500 kg/ml yeast RNA, 0.1 M DTT with 50,000
dpm/ml 35S-labeled RNA probe. After hybridization,
the slides were washed in the following order: (1) 5 x
SSC, 10 mM D!L" at 50°C for 30 min; (2) 50% forma-
mide, 2 x SSC, 0.1 M DlT at 65°C for 20 min; (3) twice
with washing solution (0.1 M Tris-HC1 tpH7.51, 0.4 M
NaC1, 50 mM EDTA) for 10 min at 37°C; (4) washing
solution with 20 kg/ml of RNaseA at 37°C for 1 hr; (5)
washing solution for 5 min at 37°C; (6) repeat the 50%
formamide wash once; (7) 2 x SSC at 37°C for 15 min;
and finally (8) 0.1 x SSC at 37°C for 15 min. Then, the
slides were dehydrated rapidly and processed for stan-
dard autoradiography using Kodak emulsion NTB-2
and exposed for 5 to 28 days at 4°C. After developing in
Kodak D-19 developer and fixer, the sections were
stained with hematoxylin and eosin and mounted with
Cytoseal mounting media 60. Analysis were carried
out using both light and darkfields, or Nomarski optics
on a Zeiss Axiophot microscope (Zeiss, Thornwood,
Y. J. is grateful to Dr. David Sassoon for help with
the in situ analysis. Y. J. and K. M. also thank Dr.
David Lyden for his encouragement. R. B. would like to
dedicate this manuscript to Hal. This work was sup-
ported by grants from the National Science Foundation
(IBN-9118977) and the National Cancer Institute (P30-
NOTE ADDED IN PROOF:
During the preparation of this manuscript, two arti-
cles describing the expression patterns of Id3 and Id4
during mouse embryogenesis were published. The re-
sults are consistent with those presented here. These
articles are (1) Ellmeier and Weith (1995) Dev. Dyn.
203:163-173; and (2) Riechman and Sablitzky (1995)
Cell Growth Differ. 62337-843.
Bard, J. B., McConnell, J. E., and Davies, J. A. (1994) Toward a
genetic basis for kidney development. Mech. Dev. 483-11.
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.
Bockman, D. E., and Kirby, M. L. (1984) Dependence of thymus de-
velopment on derivatives of the neural crest. Science 223:498-500.
Burgess, R., Cserjesi, P., Ligon, K. L., and Olson, E. N. (1995) Paraxis:
A basic helix-loop-helix protein expressed in paraxial mesoderm
and developing somites. Dev. Biol. 168:296-306.
Christy, B. A,, Sanders, L. K., Lau, L. F., Copeland, N. G., Jenkins, N.
A., and Nathans, D. (1991) An Id-related helix-loop-helix protein
encoded by a growth factor-inducible gene. Roc. Natl. Acad. Sci.
Coffin, J. D., Harrison, J., Schwartz, S., and Heimark, R. (1991) An-
gioblast differentiation and morphogenesis of the vascular endothe-
lium in the mouse embryo. Dev. Biol. 148:51-62.
Cserjesi, P., Brown, D., Ligon, K. L., Lyons, G. E., Copeland, N. G.,
Gilbert, D. J., Jenkins, N. A., and Olson, E. N. (1995) Sclemxis: A
basic helix-loop-helix protein that prefigures skeletal formation
during mouse embryogenesis. Development 121:1099-1110.
Davidson, P., and Hardy, M. H. (1952) The development of mouse
vibrissae in vivo and in vitro. J. Anat. 86:342-356.
Duncan, M., DiCicco-Bloom, M., Xiang, X., Benezra, B., and Chada,
K. (1992) The gene for the helix-loop-helix protein, Id, is specifically
expressed in neural precursors. Dev. Biol. 154:l-10.
Duncan, M. K., Shimamura, T., and Chada, K. (1994) Expression of
the helix-loop-helix protein, Id, during branching morphogenesis in
the kidney. Kidney Int. 46~324-332.
Echelard, Y., Epstein, D. J., stJacques, B., Shen, L., Mohler, J., Mc-
Mahon, J. A., and McMahon, A. P. (1993) Sonic Hedgehog, a mem-
ber of a family of putative signaling molecules, is implicated in the
regulation of CNS polarity. Cell 75:1417-1430.
Ekblom, P. (1989) Developmentally regulated conversion of mesen-
chyme to epithelium. FASEB. J. 3:2141-2150.
Ellis, H. M. (1994) Embryonic expression and function of the Drosoph-
ilu helix-loop-helix gene, eztramacrochaetae. Mech. Dev. 47:65-72.
Ellmeier, W., Aguzzi, A., Kleiner, E., Kurzbauer, R., and Weith, A.
(1992) Mutually exclusive expression of a helix-loop-helix gene and
N-myc in human neuroblastomas and in normal development.
EMBO J. 11:2563-2571.
Erlebacher, A., Filvaroff, E. H., Gitelman, S. E., and Deryneck, R.
(1995) Toward a molecular understanding of skeletal development.
Evans, S. M., and OBrien, T. X. (1993) Expression of the helix-loop-
helix factor Id during mouse embryonic development. Dev. Biol.
Fukagawa, M., Suzuki, N., Hogan, B. L. M., and Jones, C. M. (1994)
Embryonic expression of mouse Bone Morphogenetic Protein-1
(BMP-l), which is related to the Drosophila Dorsoventral gene tol-
loid and encodes a putative astacin metalloendopeptidase. Dev.
Hammar, S. P., and Mottet, N. K. (1971) Tetrazolium salt and elec-
tron-microscopic studies of cellular degeneration and necrosis in the
interdigital areas of the developing chick limb. J. Cell Sci. 8:229-
Hara, E., Yamaguchi, T., Nojima, H., Ide, T., Campisi, J., Okayama,
H., and Oda, K. (1994) Id-related genes encoding HLH proteins are
required for G, progression and are repressed in senescent human
fibroblasts. J. Biol. Chem. 269:2139-2145.
Hardy, M. H. (1992) The secret life of the hair follicle. Trends Genet.
Hebrok, M., Wertz, K., and Fuchtbauer, E.-M. (1995) M-twist is an
inhibitor of muscle differentiation. Dev. Biol. 165:537-544.
Janatpour, M. J., McMaster, M. T., Israel, M. A., and Fisher, S. J.
(1994) Expression of the transcriptional negative regulator Id2 is
modulated during human cytotrophoblast differentiation. Mol. Biol.
Cell 5(Suppl.) 453a.
Jen, Y., Weintraub, H., and Benezra, R. (1992) Overexpression of Id
protein inhibits the muscle differentiation program: In vivo associ-
ation of Id with E2A proteins. Genes Dev. 61466-1479.
Jones, C. M., Lyons, K. M., and Hogan, B. L. M. (1991) Involvement of
Bone Morphogenetic Protein4 (BMP4) and Vgr-1 in morphogene-
sis and neurogenesis in the mouse. Development 111:531-542.
Kadesch, T. (1992) Helix-loop-helix proteins in the regulation of im-
munoglobulin gene transcription. Immunol. Today 13:31-36.
Katagiri, T., Yamaguchi, A., Komaki, M., Ah, E., Takahashi, N.,
Ikeda, T., Rosen, V., Wozney, J. M., Fujisawa-Sehara, A., and Suda,
T. (1994) Bone Morphogenetic Protein-2 converts the differentation
pathway of C2C12 myoblasts into the osteoblast lineage. J. Cell
JEN ET AL.
Kaufman, M. H. (1992) “The Atlas of Mouse Development.” New
York: Academic Press.
Kirby, M. L., Gale, T. F., and Stewart, D. E. (1983) Neural crest cells
contribute to aorticopulmonary separation. Science 220:1059-1061.
Kreider, B., Benezra, R., Roverea, G., and Kadesch, T. (1992) Inhibi-
tion of myeloid differentiation by the helix-loop-helix protein Id.
Kurobayashi, M., Dutta, S., and Kedes, L. (1994) Serum inducible
factors binding to an activating transcription factors motif regulate
transcription of the Id2A promoter during myogenic differentiation.
J. Biol. Chem. 269:31162-31170,
Lumsden, A. G. S. (1988) Spatial organization of the epithelium and
the role of neural crest cells in the initiation of the mammalian
tooth germ. Developpment 103 (Suppl.):155-169.
Lyons, K. M., Pelton, R. W., and Hogan, B. L. M. (1989) Pattern of
expression of murine Vgr-1 and BMP-2a RNA suggest that trans-
forming growth factor-p-like genes coordinately regulate aspects of
embryonic development. Genes Dev. 3:1657-1668.
Lyons, K. M., Pelton, R. W., and Hogan, B. L. M. (1990) Organogen-
esis and pattern formation in the mouse: RNA distribution patterns
suggest a role for Bone Morphogenetic Protein-2A (BMP-2A). De-
Lyons, K. M., Hogan, B. L. M. and Robertson, E. J. (1995) Colocaliza-
tion of BMP7 and BMP2 RNAs suggests that these factors cooper-
atively mediate tissue interactions during murine development.
Mech. Dev. 50:71-83.
McMahon, A. P., Champion, J. E., McMahon, J. A., and Sukhatme, V.
P. (1990) Developmental expression of the putative transcription
factor Egr-1 suggests that Egr-1 and c-fos are coregulated in some
tissues. Development 108:281-287.
Nagata, Y., and Todokoro, K. (1994) Activation of helix-loop-helix
proteins Idl, Id2 and Id3 during neural differentiation. Biochem.
Biophys. Res. Commun. 1991355-1362.
Neuman, T., Keen, A., Zuber, M. X., Kristjansson, G. I., Gruss, P., and
Nornes, H. 0. (1993) Neuronal expression of regulatory helix-loop-
helix factor Id2 gene in mouse. Dev. Biol. 160:186-195.
Noden, D. M. (1989) Embryonic origins and assembly of blood vessels.
Am. Rev. Respir. Dis. 140:1097-1103.
Noden, D. M. (1991) Origins and patterning of avian outflow tract
endocardium. Development 111:867-876.
and Mizuno, T. (1981) Mesenchymal control over elon-
gating and branching morphogenesis in salivary gland develop-
ment. J Embryol. Exp. Morphol. 66209-221.
Ogata, T., Wozney, J. M., Benezra, R., and Noda, M. (1993) Bone
morphogenetic protein 2 enhances expression of a gene Id (inhibitor
of differentiation), encoding a helix-loop-helix molecule in osteo-
blast-like cells. Proc. Natl. Acad. Sci. U.S.A. 90:9219-9222.
Ott, M.-O., Bober, E., Lyons, G., Arnold, H., and Buckingham, M.
(1991) Early expression of the myogenic regulatory gene, myf-5, in
precursor cells of skeletal muscle in the mouse embryo. Develop-
Panaretto, B. A. (1993) Gene expression of potential morphogens dur-
ing hair follicle and tooth formation: A Review. Reprod. Fertil. Dev.
Riechmann, V., Van Cruchten, I., and Sablizky, F. (1994) The expres-
sion patten of Id4, a novel dominant negative helix-loop-helix pro-
tein, is distinct from Idl, Id2 and Id3. Nucleic Acids Res. 22749-
Roberts, V. J., Steenbergen, R., and Murre, C. (1993) Localization of
E2A mRNA expression in developing and adult rat tissues. Roc.
Natl. Acad. Sci. U.S.A. 90:7583-7587.
Roman, J., Little, C. W., and McDonald J. A. (1991) Potential role of
RGD-binding integrins in mammalian lung branching morphogen-
esis. Development 112:551-558.
Rugh, R. (1990) “The Mouse: Its Reproduction and Development.”
Oxford Univ. Press.
Sasai, Y., Kageyama, R., Tagawa, Y., Shigemoto, R., and Nakanishi,
S. (1992) Two mammalian helix-loop-helix factors structurally re-
lated to Drosophila hairy and Enhancer of split. Genes Dev. 6:2320-
Simon-Assmann, P., and Kedinger, M. (1993) Heterotypic cellular CQ-
operation in gut morphogenesis and differentiation. Semin. Cell
Simonsen, M. S., Rooney, A., and Herman, W. H. (1993) Expression
and differential regulation of Idl, a dominant negative regulator of
basic helix-loop-helix transcription factors, in glomerular mesan-
gial cells. Nucleic Acids Res. 215767-5774.
Stuart, E. T., Kioussi, C., and Gruss, P. (1994) Mammalian PAX
genes. Ann. Rev. Genet. 28:117-140.
Sukhatme, V. P., Cao, X., Chang, L. L., Tsai-Morris, C. H., Stamen-
kovich, D., Ferreira, P. C. P., Cohen, D. R., Edward, s. A., Shows, T.
B., Curran, T., LeBeau, M. M., and Adamson, E. D. (1988) A zinc
finger-encoding gene coregulated with c-fos during growth and dif-
ferentiation and after cellular depolarization. Cell 53:37-43.
Sun, X.-H., Copeland, N. G., Jenkins, N. A,, and Baltimore, D. (1991)
Id proteins Id1 and Id2 selectively inhibit DNA binding by one class
of helix-loop-helix proteins. Mol. Cell Biol. 11:5603-5611.
Thomas, J. H. (1993) Thinking about genetic redundancy. Trends
Tournay, O., and Benezra, R. (1996) Transcription of the dominant
negative helix-loop-helix protein Id1 is regulated by a protein com-
plex containing the early growth response gene EGR-1. Mol. Cell
Wall, N. A., Blessing, M., Wright, C. V. E., and Hogan, B. L. M. (1993)
Biosynthesis and in vivo localization of the Decapentaplegic-Vg-
related protein, DVR-6 (Bone Morphogenetic Protein-6). J. Cell
Wang, Y., Benezra, R., and Sassoon, D. (1992) Id expression during
mouse development: A role in morphogenesis. Dev. Dyn. 194:222-
Weintraub, H., David, R., Tapscott, S., Thayer, M., Krause, M., Ben-
ezra, R., Blackwell, T. K., Turner, D., Rupp, R., Hollenberg, S.,
Zhuang, Y., and Lassar, A. (1991) The MyoD gene family, nodal
point during specification of the cell linage. Science 251:761-766.
Wilson, R. B., Kiledjian, M., Shen, C. P., Benezra, R., Zwollo, P.,
Dymecki, S. M., Desiderio, S. V., and Kadesch, T. (1991) Repression
of immunoglobulin enhancers by the helix-loop-helix protein Id
Implication by B lymphoid development. Mol. Cell Biol. 11:6185-
Wilson, R., and Mohun, T. (1995) XIdx, a dominant negative regulator
of bHLH function in early Xenopw embryos. Mech. Dev. 49:211-
Wolf, C., Thisse, C., Stoetzel, C., Thisse, B., Gerlinger, P., and Perrin-
Schmitt, F. (1991) The M-twist gene of Mus is expressed in subsets
of mesodermal cells and is closely related to the Xenopus X-twi and
Drosophila twist genes. Dev. Biol. 143:363-373.
Zhang, H., Reynaud, S., Kloc, M., Etkin, L. D., and Spohr, G. (1995) Id
gene activity during Xenopus embryogenesis. Mech. Dev. 50119-