Nucleic Acids Research, 1994, Vol. 22, No. 5
The expression pattern of 1d4, a novel dominant negative
helix-loop-helix protein, is distinct from Idl, Id2 and 1d3
Veit Riechmann, Ingeborg van Cruchten and Fred Sablitzky*
Max-DelbrOck-Laboratorium in der Max-Planck-Gesellschaft Carl-von-Linne-Weg 10,
50829 Koln, Germany
Received December 20, 1993; Revised and Accepted February 3, 1994
Molecular interaction between transcription factors
containing an basic-helix-loop-helix (bHLH) domain is
known to regulate differentiation in several cellular
systems including myogenesis, neurogenesis and
haematopoiesis. DNA-binding activity of the bHLH
proteins is mediated via the basic region and
dependent upon formation of homo- and/or hetero-
dimers of these transcription factors. Dominant nega-
tive (dn) HLH proteins (Id1, 1d2, 1d3 and emc) also
contain the HLH-dimerization domain but lack the DNA-
binding basic region. Formation
between dnHLH and bHLH proteins abolishes the DNA-
binding activity of the latter. Concordantly,
shown that the dnHLH protein Idl inhibits different-
iation of muscle and myeloid cells in vitro. Therefore,
it was postulated that dnHLH proteins serve as general
antagonists of cell differentiation. We have isolated and
characterized a novel mouse dnHLH gene, designated
1d4. The 1d4 protein contains a HLH domain highly
conserved among the dnHLH proteins from mouse and
drosophila. Outside of the HLH domain, three additional
short regions of the dnHLH proteins show some degree
of homology. DNA-binding of E47 homo- as well as
E47/MyoD heterodimers is inhibited by 1d4. Transcript-
ion of the Id4 gene results in three RNA molecules of
3.7, 2.0 and 1.7 kb which are presumably a result of
differential splicing and/or alternatively used poly-
adenylation sites within the 3' untranslated region.
During embryogenesis, 1d4 expression is up-regulated
between day 9.5 and 13.5 of gestation. The highest
expression in adult tissues was detected in testis, brain
and kidney. Comparison of the expression patterns of
the four mouse dnHLH genes revealed that Id4 expres-
sion differs from the more restricted expression of 1d2
as well as from the widespread expression of Idl
The helix-loop-helix (HLH) domain, was initially identified as
a structural motifby comparing the amino acid sequences of the
drosophila neurogenic proteins encoded by the achaete-scute
EMBL accession no. X75018
(ac-sc) complex (1), daughterless (2) and the mammalian c-
myc (3) and MyoD (4, 5) proteins. Murre et al. (6) proposed
that amino acids conserved within the region of homology
between the various proteins adopted a common secondary
structure: two amphipathic et-helices separated by an intervening
loop. They and others demonstrated that the HLH domain
mediates protein dimerization (6) and that dimeric proteins bind
DNA via a group of positively charged amino acids [the basic
(b) region] directly adjacent to the HLH domain (7, 8). The DNA
sequences that these bHLH proteins bind are the so called E-
boxes which are represented by the core sequence CANNTG (9,
10). bHLH proteins are transcription factors which have been
differentiation in several cellular systems including myogenesis
[MyoD (3), myf-5 (11), MRF4 (13-14) and myogenin (15-17)],
neurogenesis [ac-sc (1), MASH] (18)] and haematopoiesis
[SCL/TCL-5/tal-] (19-23) and lyl-1 (24-26)]. It was shown
(4, 25) that these cell-type-restricted bHLH proteins form
heterodimers with the ubiquitously expressed bHLH proteins
which are encoded by the E2A gene [E12, E47 (6) and E2-5/ITFJ
(26)] and the E2-2 gene [ITF2 (26)] in mouse and by the
daughterless gene (2) in drosophila. Furthermore, formation of
Transcriptional activity of E-box containing promoters and/or
enhancers might therefore exclusively be dependent upon the
concentration of different bHLH proteins within the cells.
However, subgroups of HLH proteins were described which
antagonize the function of the above bHLH proteins. The
drosophila hairy (28) and Enhancer of split (29) gene products
and the rat protein, HES-1 (30), are distinct from the bHLH
proteins in that their basic regions contain a proline residue
altering their DNA-binding capacity. Proteins of this subgroup
do not bind the E-box efficiently but they do bind the so called
N-box motif [CACNAG (29-31)]. Another subgroup of HLH
proteins was described [emc (32, 33), Id] (34), HLH 462 (35),
and Id2 (36)] which lack the basic domain. These proteins still
dimerize but they cannot bind DNA. Moreover, dimerization with
bHLH proteins block DNA-binding of the resulting heterodimers
(32 -34). Therefore, these proteins are referred to as dominant
negative (dn) HLH proteins (37). It was shown in drosophilafor
example, that emc represses proneural gene expression of the
ac-sc complex by sequestering the latter proteins in complexes
*To whom correspondence should be addressed
Nucleic Acids Research, 1994, Vol. 22, No. 5
unable to interact with DNA (38, 39). In controlling the spatial
pattern ofac-sc transcription, emc is involved in the regulation
of pattern formation of functional proneural clusters and the
subsequent bristle formation in the wing imaginal disc (38). For
the mammalian dnHLH gene products, Idl and/or Id2, it was
demonstrated that their expression is inversely correlated with
cell differentiation in myogenesis (42) and neurogenesis (58) as
well as myelopoiesis (43) and lymphopoiesis (36, 40, 44).
overexpression of the Idl
differentiation ofmuscle cells (42) as well as myeloid cells (43).
In addition, ectopic expression of the Idl protein in B cells
represses the activity of the immunoglobulin intron enhancers
of heavy and x-light chains (40) as well as the 3' x enhancer
(44). Hence, a model was proposed in which the Idl protein
serves as a general antagonist of cell differentiation by inhibiting
bHLH proteins specifically required for developmental programs
(34, 37, 45). Although not yet shown directly, it is likely that
in analogy to Idl, mouse (36) and human (46) Id2 as well as
mouse [HLH462 (35)] and human [HLHIR21/Heir-1 (47, 48)]
Id3 are important regulators of differentiation in various cellular
Here we report the isolation, functional characterization and
expression pattern of a novel mouse dnHLH protein, designated
Id4. The ld4 protein contains a HLH motif which is highly
conserved among the dnHLH proteins. Three additional regions
of the Id4 protein are also similar to the other members of the
dnHLH protein family, but conservation is less pronounced. Id4,
which lacks a basic region, forms heterodimers with E47 and
inhibits DNA-binding ofE47homodimers as well as E47/MyoD
heterodimers in vitro. Expression of Id4 is up-regulated during
embryonic development and is highestintestis, brain and kidney
of adult mice. This expression pattern of Id4 differs from the
more restricted expression of1d2 as well as from the widespread
expression of Id] and Id3.
MATERIALS AND METHODS
Isolation and sequence analysis of 1d4 cDNA clones
sponding to the reverse complementary sequence of the second
helix ofIdl (34), human and mouse Id2 (36, 46), HLH462 (35)
and emc (32, 33) was used to screen a Xgtl 1 cDNA library made
from mouse bone marrow cells (Clontech). The filters were
hybridized for 43 hr at 44°C with the 43-mer, which had been
radiolabeled by T4 polynucleotide kinase (Boehringer) in the
presence of -y-32P-ATP (Amersham) in
albumin/i mM EDTA/0.5 M NaHPO4 (pH 7.2)/7 % SDS.
Washing was performed two times with 1 mM EDTA/40 mM
NaHPO4 (pH 7.2)/5 % SDS and eight times with
EDTA/40 mM NaHPO4 (pH 7.2)/1 % SDS at RT for 5 min
each (49). Two cDNA clones (VR4, VR18) were subcloned in
pGEM 7 (Promega) and sequenced (sequenase kit; United States
Biochemical). In order to obtain full length cDNA clones, VR4
was used to screen a XgtlO cDNA library from RNA of 12.5
days old mouse embryos (generously supplied by Drs M.Hanks
and A.Joyner, Toronto). Three overlapping clones (B, F and I;
see Fig. 1) were isolated, subcloned in pGEM 7 and sequenced.
Sequence analysis of GC rich regions of the Id4 cDNA were
performed in parallel reactions using dITP instead of dGTP.
1 % bovine serum
In vitro transcription and translation
In vitro transcription of Id4 (clone I linearized with XbaI which
cuts in the 3' untranslated region), MyoD [plasmid pEMCIIs (4)
generously provided by Dr H.Weintraub, Seattle] and E47
(6) generously provided by Dr
C.Murre, La Jolla] was performed under standard conditions
using 100 u of SP6 or T3 polymerases (Boehringer) in a total
volume of 100 ,l. RNA was purified, ethanol precipitated and
stored at -70 °C in diethylpyrocarbonate treated destilled water.
In vitro translation of the RNAs was performed in 50 td rabbit
reticulocyte lysate (Promega) using 5-45 % of the synthesized
RNA as a substrate. To generate radioactively labeled proteins,
parallel reactions were performed, using [35S] methionine and
the resulting proteins were analyzed on a 12 % discontinuous
SDS-PAGE gel (data not shown).
Electrophoretic mobility shift assay
A 25 bp double stranded oligonucleotide (34) containing the E47
and/or E47/MyoD binding site from the MCK enhancer (5) was
labeled as described (34). 20 dl of in vitro translation reaction
was heated to 37 °C for 20 min. DNA-binding reactions with
20 ,l DNA-binding cocktail including 0.1 ng double stranded
oligonucleotide probe and analysis on a 5 % PAGE were
performed as described (34).
Expression of Id4
polyA+ RNA was isolated from various mouse tissues and cell
lines using the Fast Track kit (Invitrogen). Northern blot analysis
were done under standard conditions (50). Probe a corresponds
to VR4 and probe b corresponds to F2 (Fig. 1).
First strand cDNA was synthesized using oligo(dT)18 and M-
MLV Reverse Transcriptase for 60 min at 37 °C in a total volume
of 50Al(according to the protocol of GIBCO BRL).
First strand cDNA was amplified by PCR (51) using the
primers depicted below:
nucleotide position 84-560 (34) resulting in the amplification
of a 476 bp cDNA fragment
nucleotide position 39-603 (36) resulting in the amplification
of a 564 bp cDNA fragment
nucleotide position 61-441 (35) resulting in the amplification
of a 380 bp cDNA fragment
nucleotide position 268-538 of Id4 sequence resulting in the
amplification of a 270 bp cDNA fragment
nucleotide position 728-1076 (52) resulting in the amplification
of a 348 bp cDNA fragment
Amplification of first strand cDNAs was performed as
described (53) using 1.5 mM MgCl2 and 3 u of Taq polymerase
(GIBCO BRL). In the cases of Idl, Id3 and Id4 9 % DMSO
was added. The amount of input cDNA was adjusted using the
,3-actin amplification for standardization. To assure linear
Nucleic Acids Research, 1994, Vol. 22, No. 5
amplification the optimal number of cycles was determined for
each prime pair (33 cycles Idl, 36 cycles Id2, 23 cycles Id3,
29 cycles 1d4 and 19 cycles ,3-actin). All PCR amplifications were
carried out using the following conditions for denaturation (92°C
for 85 s), annealing (59°C for 85 s) and polymerization (72°C
for 150 s). Whenever possible (Idl, Id2 and ,B-actin) primers were
designed to amplify sequences including introns to be able to
discriminate between genomic and cDNA.
Southern blots ofthe amplified PCR products were hybridized
(50) with probes specific for Idl, Id2 [subclones of amplified
genomic DNA (I.van Criichten and F.Sablitzky, unpublished)],
1d3 [a 0.8 kb EcoRI-cDNA fragment (V.Riechmann and
F.Sablitzky, unpublished)] and Id4 (probe a in Fig. 1).
Nucleotide sequence accession number
Isolation and sequence analysis of Id4 cDNA
Sequence comparison of the dnHLH genes Idl (34), human (46)
and mouse Id2 (36), HLH462 [(35); hereafter called mouse Id3]
and drosophila extramacrochaetae [emc (32, 33)] revealed a
strong homology within the HLH region ofthese genes (see Fig.
2). Helix 2 is especially strongly conserved among dnHLH genes
but distinct from all members of the DNA-binding bHLH gene
family (6). In order to clone novel members ofthe dnHLH gene
family, a degenerate oligonucleotide complementary to the helix
2 region of the dnHLH genes was designed (see Materials and
Methods) and used to screen a cDNA library made from mouse
bone marrow cells. Twenty out of 22 clones initially obtained
were identified by hybridization as either Idl or Id3, respectively.
The two remaining clones (VR4 and VR18) which did not
hybridize to any of the dnHLH genes turned out to be identical.
Sequence analysis revealed that these two cDNA clones encode
for a novel member ofthe dnHLH gene family (Fig. 1). To obtain
full length cDNA, VR4 was used to screen a cDNA library
prepared from 12.5 days old mouse embryos. The sequence
analysis of three overlapping clones (B, F and 1) is summarized
in Fig. 1. Clone F contained two EcoRI fragments (1.5 and 0.7
PPSII X HH
oligonucleotide and VR4 (data not shown). However, the 0.7 kb
fragment hybridizes with two of three Md4 mRNAs ( see below
and Fig. 3) indicating that it is most likely part of the 3'
untranslated regions of two Md4 mRNA species. The 5' end of
clone B started with 51 A residues. Most likely, a short, unrelated
cDNA containing a polyA tail was ligated to the 5' end of the
Md4 cDNA reflecting a cloning artifact (data not shown).
Therefore the composite nucleotide sequence of VR4, B, Fl and
I starts at the 5' end of clone I. The sequenced 1.659 kb Md4
cDNA fragment contains an open reading frame (nucleotide
position 6 to 554) with a putative AUG start codon at nucleotide
position 72, predicting an Id4 protein of 161 amino acids (data
the larger one hybridizing
Id4 is a member of the dnHLH protein family
Comparison of the amino acid sequences of Md4 with Idl (34),
human (46) and mouse (36) Md2, human [Heir-11HLH 1R21 (47,
48)] and mouse (35) Id3 and drosophila emc (32, 33) indicated
that Md4 is a new member of the dnHLH protein family (Fig.
2). The members share a highly conserved HLH motif (box 3
in Fig. 2) with the highest degree of identity in helix 2 [11/16
amino acids are identical including a valine which substitutes for
an alanine residue conserved among bHLH proteins (6)]. The
length of the loop (10 amino acids) and proline residues at two
positions within the loop are conserved. Md4, like the other
TSQVIRYDRVTTAKILLKERKAI I SR.
Figure 1. Structure of the partial 1d4 cDNA. A restriction map ofthe Id4 cDNA
is shown at the top (S: SacI; P: PstI; SII: SacdI; X: XbaI; H: HindIII; C: Clal;
E: EcoRI). The open box represents the putative Id4 coding region. The dnHLH
region is shown as a black box. Solid lines indicate the four cDNA clones isolated.
Arrows depict the sequence strategy used.
Figure 2. Comparison ofthe deduced amino acid sequences ofmouse (Id4, Idl,
Id2, HLH462), human (Id2, Heir-l/HLHIR21) and drosophila (emc) dnHLH
encoding genes, respectively. Sequences were aligned to maximize homology.
Identical amino acids found in at least two different Id genes are shown as white
letters on black background. Four regions ofhomology are underlined and marked
as box 1 to 4. Deduced amino acid sequences 5' of the putative N-terminal ends
of the Id proteins are shown at the top. * indicates stop codons. References are
as indicated in the text.
Nucleic Acids Research, 1994, Vol. 22, No. S
1 5 10!
Figure 3. Inhibition of binding of E47 homodimers (A) and E47/MyoD
heterodimers (B) to the E-box of the MCK enhancer by 1d4. Electrophoretic
mobility shift assay were performed using unlabeled in vitro translated proteins
mixed with a double stranded, 32p labeled MCK enhancer oligonucleotide. The
numbers at the top indicate the relative amounts of RNAs added to translation
reactions. Reticulocyte lysate alone did not give rise to a band shift (lanes 6 and 7).
members of the dnHLH family, does not contain a basic amino
acid region upstream of the HLH motif which is conserved in
DNA-binding bHLH proteins (6). Outside ofthe HLH motifthe
dnHLH proteins are distinct except for three regions which show
some degree of homology (boxes 1, 2 and 4 in Fig. 2). Box 1
comprises the first 8 amino acids of the dnHLH proteins.
Upstream of the putative AUG start codons the amino acid
sequences are diverged. This is also true for the otherwise highly
conserved homologs of human and mouse
respectively. We therefore predict that 1d4, which has a stop
codon 22 residues upstream of the putative AUG codon, shares
a conserved N-terminus with the other dnHLH proteins. It has
been outlined before (36) that the stretches of homology in the
N-terminal (box 2) and C-terminal region (box 4) are rich in
serine and threonine which could be targets forphosphorylation.
In addition, the N-terminus of 1d4 is rich in alanine and the C-
terminus is rich in proline residues. The latter feature is shared
by the human and mouse 1d3 proteins.
1d4 inhibits DNA-binding ofE47homodimers and E47/MyoD
To test whether Id4 is also functionally related to the dnHLH
proteins we asked whether 1d4 can inhibit binding of E47
homodimers or E47/MyoD heterodimers to the E-box in theMCK
(muscle creatine kinase) enhancer oligonucleotide (34). In vitro
transcripts ofId4, E47 (6) andMyoD (4) expression vectors were
cotranslated as described (34, 35) and DNA-binding of the
resulting protein complexes were determined in an electrophoretic
mobility shift assay (Fig. 3). DNA-binding ofE47homodimers
(Fig. 3A) as well as E47/MyoD heterodimers (Fig. 3B) was
inhibited by the Id4 protein and this inhibition was dependent
upon the amount of Id4 RNA added to the in vitro translation
reactions (Fig. 3 lanes 2-4 and 9-12). As expected, 1d4 itself
did not generate a band shift indicating that it does not bind the
MCKenhancer oligonucleotide (Fig. 3 lanes 5 and 8). In parallel
reactions, appropriate in vitro translation of radiolabeled HLH
proteins was monitored by SDS-PAGE analysis (data not shown).
Figure 4. Tissue distribution of Id4 mRNA. Northern blot analysis of polyA+
RNA isolated from adult and fetal organs as indicated. (A) Probe a (corresponding
to VR4; see Figure 1) hybridizes with three mRNAs of 3.7, 2.0 and 1.7 kb,
respectively. (B) Probe b (corresponding to F2; see Figure 1) hybridizes only
with the two large mRNAs. (C) To control the amount ofpolyA+ RNA in each
lane, the same filter was hybridized with a GAPDH probe (55).
1d4 is differentially expressed in adult organs and is up-
regulated during embryogenesis
Northern blot analysis, shown in Fig. 4, demonstrates that
expression of1d4 in adult organs ofthe mouse is highest in brain,
testis and kidney and moderate in thymus. Using probe a which
contains most of the 1d4 coding region (see Fig. 1), three
transcripts (3.7, 2.0 and 1.7 kb, respectively) are present (Fig.
4A). Probe b, however, hybridizes only with the two larger
transcripts (Fig. 4B). We therefore believe that probe b
corresponds to the 3' untranslated region of the two larger 1d4
mRNAs and that the three 1d4 transcripts differ in their 3' ends
due to alternatively used polyadenylation sites and/or alternatively
spliced untranslated exons. In brain and kidney the three
transcripts seemed to be equally abundant contrasting the situation
in testis and thymus where much more of the 1.7 kb transcript
is present. These differences in the abundance of the three
transcripts among the organs suggest a tissue specific regulation
of 1d4 expression.
Although not obvious from the Northern blot analysis,
amplification of cDNA by RT-PCR revealed, that 1d4 is
expressed in bone marrow and spleen, albeit much lower than
in testis, brain, and kidney (Fig. 5; lanes 8 to 13). 1d4 RNA is
hardly detectable in liver and absent in an endothelial cell line
During embryogenesis 1d4 expression is rather low at day 9.5
of gestation. A dramatic increase of1d4 expression occurs within
the following 4 days of development to reach a plateau at day
Nucleic Acids Research, 1994, Vol. 22, No. 5
ITrTF ..wr i.*
Figure 5. Expression patterns ofthe mouse dnHLH genes Idi, 1d2, 1d3 and 1d4.
cDNA from various stages ofembryogenesis (whole embryos from day 9.5, 11,
13.5 and 15.5 of gestation), fetal liver (day 12.5 13.5 and 16.5 of gestation),
adult tissues as indicated [bone marrow (b. m.)] and cell lines [endothelial (end.)
cell line sEnd. 1 (54), embryonic stem cell line D3 (56) and carcinoma cell line
F9 (57)] were amplified using primers specific for the dnHLH genes Idl,
1d3 and 1d4. The PCR products were blotted and hybridized with appropriate
probes. Amplification of (-actin was performed to control the amount of input
cDNA (for details see Material and Methods).
13.5 (Fig. 5; lanes 1 to 4). This up-regulation of Id4 is not a
consequence of the rapidly growing fetal liver during this time
of development since expression of Id4 in the fetal liver is
moderate and rather decreases between day 12.5 and 16.5 of
gestation (Fig. 5; lanes 5 to 6).
The expression pattern of Id4 differs from the restricted
expression of Id2 as well as from the widespread expression
of Idi and Id3
The RT-PCR analysis shown in Fig. 5 revealed that the four
development. The widespread expression pattern ofId3 and Id]
is in contrast to the restricted expression pattern of Id4 and Id2.
In addition, the relative abundance of the transcripts for each
dnHLH gene varies in the analyzed tissues and cell lines.
During embryogenesis (day 9.5 to 15.5 of gestation) the level
ofId3 expression is high and stays constant. Id] expression seems
to increase slightly whereas the expression ofId4 is dramatically
up-regulated. In contrast, Id2 transcripts could only be detected
at day 11 of gestation (Fig. 5; lanes
cells (D3) or carcinoma cells (F9) expression of Id] and 1d4 is
abundant, of Id3 is moderate and of Id2 is not detectable (Fig.
5; lanes 15 and 16). In the developing fetal liver (day 12.5 to
16 of gestation) Id] and 1d3 are constantly expressed. In contrast,
Id2 is up-regulated and 1d4 is down-regulated suggesting a specific
1 to 4). In embryonic stem
regulatory role of these dnHLH proteins in the development of
the fetal liver (Fig. 5; lanes 5-6).
We also determined the expression pattern of the dnHLH genes
in adult organs (Fig. 5; lanes 8- 13). In brain, testis and bone
expression of Id2 is weak compared to the one of Idl, Id3 or
1d4. In liver and spleen, however, only Id3 expression is high.
Idi expression is moderate in liver and barely detectable in spleen
whereas Id4 shows the opposite expression pattern in these two
organs. Id2 transcripts are neither detectable in liver nor in spleen.
In the endothelial cell line sEnd.] (Fig. 5; lane 14) expression
of Idl and Id3 is high, of Id2 low and of Id4 undetectable.
all four dnHLH genes are expressed. In kidney
We have isolated and functionally characterized a fourth member
(Id4) of the mammalian dnHLH gene family and compared its
expression pattern with the ones of the other mouse dnHLH
genes. Structurally, the dnHLH proteins are highly conserved
within the HLH domain resulting in their ability to form
heterodimers in vitro with ubiquitously expressed bHLH proteins
(like E47) sequestering the cell-type restricted bHLH proteins
(like MyoD) and, subsequently, to block DNA-binding. Outside
of the HLH domain the dnHLH proteins are less conserved.
Three short regions (boxes 1, 2 and 4 in Fig. 2) however, show
some degree of homology. Box 2 and 4 are rich in serines and
threonines suggesting a common post-translational regulation via
Expression of the 1d4 gene results in three transcripts (3.7,
2.0 and 1.7kb) which are presumably a result of differential
splicing or alternatively used polyadenylation sites within the 3'
untranslated region. This transcriptional modification seemed to
be regulated in a tissue specific manner since the abundance of
each transcript varies between the organs. Such complex
regulation of expression has not been described so far for the
other members of the dnHLH gene family. Idl, Id2 and Id3
seemed to be expressed as single transcripts of 1.3 kb (58), 1.6
kb (36) and about 1.0 kb (35), respectively. During mouse
development the 1d4 gene expression is also differentially
regulated. In undifferentiated embryonic stem cells (D3) and
carcinoma cells (F9) 1d4 is highly expressed. In the course of
embryogenesis, however, Id4 expression is low at day 9.5 and
increases dramatically within the next 4 days of development.
During the same time ofembryonic development, expression of
Idl is slightly increasing, Id3 is constantly expressed whereas
1d2 transcripts were only detectable by PCR at day 11 of
The latter result is somewhat surprising since Id2 transcripts
are detectable during embryonic development by Northern Blot
analysis (I.v.Cruchten and F.Sablitzky; unpublished). One
possible explanation for this discrepancy could be that the Id2
gene is differentially spliced in the 5' region with the consequence
that the sequence complementary to the Id2-sense-primer is
missing in those transcripts in the embryos detected by Northern
Blot analysis. Alternatively spliced mRNA could also explain the
presence of additional PCR products for Id2 seen in fetal liver
and bone marrow (Fig. 5; lanes 6 and 8) and for 1d4 seen in
some tissues including embryo, bone marrow, testis, kidney and
brain (Fig. 5; lanes 3,4 and 8-11).
In situ hybridization indicated that Id] is highly expressed in
almost all regions of the mouse embryo upon gastrulation (59).
754 Nucleic Acids Research, 1994, Vol. 22, No. S
Between day 9.5 and 16.5 of gestation Idl was specifically
expressed by undifferentiated neuronal precursors of the
ventricular zone, but Idl transcripts could not be detected in their
differentiated derivatives (58). We are currently analyzing the
expression pattern of 1d4 during embryogenesis by in situ
hybridization to determine whether 1d4 expression is similarly
restricted to undifferentiated cells.
In adult organs 1d4 is highly expressed in testis, brain and
kidney and moderately in thymus, bone marrow and spleen. This
expression pattern of1d4 is again different from the other dnHLH
genes. All organs tested contain similar amounts of 1d3
transcripts. This is in agreement with previous data (35) except
that we detected 1d3 transcripts also in testis. As was shown
before (58), Idi is also expressed in all tissues, but Idl transcripts
are less abundant in liver and in spleen. Id2 transcripts on the
other hand, could be amplified in bone marrow, testis and brain
and to a much lesser extent in kidney.
As mentioned before, all four mammalian dnHLH proteins
form heterodimers with bHLH transcription factors in vitro and
block their DNA-binding capacity. Our comparative expression
data indicate that in many tissues two or more dnHLH genes are
expressed. So far,
expression is true at the single cell level. In addition, we note
that the presence of RNA is not necessarily indicative of the
presence of functional protein. However, assuming that some
differentiated cells express functional dnHLH proteins the model
outlined above that dnHLH proteins serve as general antagonists
of cell differentiation should be extended. It could be that in
particular differentiation pathways bHLH transcription factors
are required only transiently. It could also be that mature cells
which maintain a particular differentiated state independent of
the ubiquitous bHLH transcription factors express dnHLH
proteins to inhibit their function. Finally, dnHLH and bHLH
proteins could be co-expressed resulting in a balanced network
ofpositive and negative regulators. Hence, dnHLH proteins might
also be involved in the manifestation of a differentiated cell.
it is unknown whether such overlap of
We thank G.Heimeroth for expert technical help, E.Riipping,
M.Hotfilder and Dr R.Dildrop for reagents and/or valuable
advice and U.Ringeisen for excellent photographic work. We
are grateful to Drs H.Weintraub and C.Murre for generously
providing expression vectors ofE47and MyoD and Drs M.Hanks
and A.Joyner and for generously providing a cDNA library of
mouse embryo. We also thank Dr J.Dangl for critically reading
thismanuscript. This work was
Bundesministerium fiir Forschung und Technologie.
1. Villares, R. and Cabrera, C.V. (1987) Cell, 50, 415-424.
2. Caudy, M., Vassin, H., Brand, M., Tuma, R., Jan, L.Y. and Jan, Y.N.
(1988) Cell, 55, 1061-1067.
3. Battey, J., Moulding, C., Taub, R., Murphy, W., Stewart, T., Potter, H.,
Lenoir, G. and Leder, P. (1983) Cell, 34, 779-787.
4. Davis, R.L., Weintraub, H. and Lassar, A.B. (1987) Cell, 51,987-1000.
5. Lasser, A.B., Buskin, J.N., Lockshon, D., Davis, R.L., Apone, S.,
Hauschka, S.D. and Weintraub, H. (1989) Cell, 58, 823-831.
6. Murre, C., Schonleber McCaw, P. and Baltimore, D. (1989) Cell, 56,
7. Davis, L., Cheng, P.-F., Lassar, A.B. and Weintraub, H. (1990) Cell,66,
8. Voronova, A. and Baltimore, D. (1990) Proc. Natl. Acad. Sci. USA, 87,
9. Ephrussi, A, Church, G.M., Tonegawa, S. and Gilbert, W. (1985) Science,
10. Church, G.M., Ephrussi, A., Gilbert, W. and Tonegawa, S. (1985) Nature
(London), 313, 798-801.
11. Braun, T., Buschhausen-Denker, G., Bober, E., Tennich, E. and Arnold,
H.H. (1991) EMBO J., 8, 701-709.
12. Braun, T., Bober, E., Winter, B.; Rosenthal, N. and Arnold, H.H. (1990)
EMBO J., 9, 821-831.
13. Rhodes, S.J. and Konieczny, S.F. (1989) Genes Dev., 3, 2050-2061.
14. Miner, J.H. and Wold, B. (1991) Proc. Nati. Acad.
15. Braun, T., Bober, E., Buschhausen-Denker, G., Kotz, S., Grzeschik, K.
and Arnold, H.H. (1989) EMBO J., 8, 3617-3625.
16. Edmondson, D.G. and Olson, E.N. (1989) Genes Dev., 3, 628-640.
17. Wright, W.E., Sassoon, D.A. and Lin, V.K. (1989) Cell, 56, 607-617.
18. Johnson, J.E., Birren, S.J. and Anderson, D.J. (1990) Nature (London),
19. Finger, L.R., Kagan,, J., Christopher, G., Kurtzberg, J., Hershfield, M.S.,
Nowell, P.C. and Croce, C.M. (1989) Proc. Natl. Acad. Sci. USA, 86,
20. Begley, C.G., Aplan, P.D., Denning, S.M., Haynes, B.F., Waldmann, T.A.
and Kirsch, I.R. (1989) Proc. Nat. Acad. Sci. USA, 86, 10128-10132.
21. Chen, Q., Cheng, J.-T., Tsai, L.-H., Schneider, N., Buchanan, G., Carroll,
A.; Crist, W., Ozanne, B., Siciliano, M.J. and Baer, R. (1990) EMBO J.,
22. Green, A.R., DeLuca, E. and Begley, C.G. (1991) EMBO J.,
23. Aplan, P.D., Nakahara, K., Orkin, S.H. and Kirsch I.R. EMBO J., 11,
24. Mellentin, J.D., Smith, S.D. and Cleary, M.L. (1989), Cell, 58, 77-83.
25. Visvader, J., Begley, C.G. and Adams, J.M. (1991) Oncogene, 6, 187-194
26. Kuo, S.S., Mellentin, J.M., Copeland, N.G., Gilbert, D.J., Jenkins, N.A.
and Cleary, M.L. (1991) Oncogene, 6, 961-968.
25. Murre, C., Schonleber McCaw, P., Vissin, H., Caudy, M., Jan, L.Y., Jan,
Y.N., Cabrera, C.V., Buskin, J.N., Hauschka, S.D., Lassar, A.B.,
Weintraub, H. and Baltimore, D. (1989) Cell, 58, 537-544.
26. Henthorn, P., Kiledjian, M. and Kadesch, T. (1990) Science, 247,467-470.
27. Lasser, A.B., Davis, R.L., Wright, W.E., Kadesch, T., Murre, C.,
Voronova, A., Baltimore, D. and Weintraub, H. (1991) Cell, 66, 305-315.
28. Ingham, P.W., Howard, K.R. and Ish-Horowicz, D. (1985) Nature, 318,
29. Kliimbt, C., Knust, E., Tietze, K and Campos-Ortega, J.A. (1989) EMBO
J., 8, 203-210.
30. Sasai, Y., Kageyama, R., Tagawa, Y., Shigemoto, R. and Nakanishi, S.
(1992) Genes Dev., 6, 2620-2634.
31. Tietze, K., Oellers, N. and Knust, E. (1992) Proc. Natl. Acad. Sci. USA,
32. Ellis, H.M., Spann, D.R. and Posakony, J.W. (1990) Cell, 61, 27-38.
33. Garrell, J. and Modolell, J. (1990) Cell, 61, 39-48.
34. Benezra, R., Davis, R.L., Lockshon, D., Turner, D.L. and Weintraub, H.
(1990) Cell, 61, 49-59.
35. Christy, B.A., Sanders, L.K., Lau, L.F., Copeland, N.G., Jenkins, N.A.
and Nathans, D. (1991) Proc. Natl. Acad. Sci. USA, 88, 1815-1819.
36. Sun, X.-H., Copeland, N.G., Jenkins, N.A. and Baltimore, D. (1991) Mol
Cell. Biol., 11, 5603-5611.
37. Kadesch, T. (1992) Immunol. Today, 13, 31-36.
38. Van Doren, M., Powell, P.A., Pastemak, D., Singson, A. and Posakony,
J.W. (1992) Genes Dev., 6, 2592-2605.
39. Martinez, C., Modolell, J., and Garrell, J. (1993) Mo. Cell. Biol., 13,
40. Wilson, R.B., Kiledjian, M., Shen, C.-P., Benezra, R., Zwollo, P., Dymecki,
S.M., Desiderio, S.V. and Kadesch, T. (1991) Moll.
41. Kawaguchi, N., DeLuca, H.F.and Noda, M. (1992) Proc. Natl. Acad. Sci.
USA, 89, 4569-4572.
42. Jen, Y. Weintraub, H. and Benezra, R. (1992) Genes Dev., 6, 1466-1479.
43. Kreider, B.L., Benezra, R., Rovera, G. and Kadesch, T. (1992) Science,
44. Pongubala, J.M.R. and Atchison, M.L. (1991) Moll.
45. Kadesch, T. (1993) Cell Growth Differ., 4, 49-55.
46. Biggs, J., Murphy, E.V.and Israel, M.A. (1992) Proc. Natl. Acad. Sci. USA,
Sci. USA, 87,
Cell. Biol., 11,
Cell. Biol., 11,
Nucleic Acids Research, 1994, Vol. 22, No. 5 Download full-text
47. Ellmeier, W., Aguzzi, A., Kleiner, E., Kurzbauer, R. and Weith, A. (1992)
EMBO J., 11, 2563-2571.
48. Deed, R.W., Bianchi, S.M., Atherton, G.T., Johnston, D., Santibanez-Korf,
M., Murphy, J.J. and Norton, J.D. (1993) Oncogene, 8, 599-607.
49. Church, G.M. and Gilbert, W. (1984) Proc. Natl. Acad. Sci. USA, 81,
50. Sambrook, J., Fritsch, E.F and Maniatis, T. (1989) Molecular Cloning, Cold
Spring Harbor Laboratory Press
51. Saiki, R.K., Faloona, F.A., Mullis, K.B., Horn, C.T., Erlich, H.A. and
Arnheim, N. (1985) Science, 230, 1350-1354.
52. Nudel, U., Zakut, R., Shani, M., Neuman, S. Levy, Z., Yaffe, D. (1983)
Nucl. Acids Res., 3, 1759-1771.
53. Brady, G., Barbara, M. and Iscove, N.N. (1990) Methods Mol. Cel. Biol.,
54. Williams, R.L., Courtneidge, S.A. and Wagner, E.F. (1988) Cell, 52,
55. Piechaczyk, M., Blanchard, J.M., Marty, L., Dani, C., Panabieres, F.,
Sabouty, S.E., Fort, P. and Jeanteur, P. (1984) Nucl. Acids Res., 12,
56. Doetschman, T.C., Eistetter, M., Katz, M., Schmidt, W. and Kemler, R.
(1985) J. Embryol. Exp. Morph., 87, 27-45.
57. Bernstine, E.G., Hooper, M.L., Grandchamp, S. and Ephrussi, B. (1973)
Proc. Nat. Acad. Sci., USA 70, 3899-3903.
58. Duncan, M., DiCicco-Bloom,E.M., Xiang, X., Benezra, R. and Chada,
K. (1992) Dev. Biol., 154, 1-10.
59. Wang, Y, Benezera, R. and Sassoon, D.A. (1992) Dev. Dynamics, 194,