|
|
|
|||||||
|
|||||||||
(Received for publication, December
9, 1994; and in revised form, January 26, 1995) From the
The oncoprotein v-ErbA is a mutated version of thyroid hormone
receptor The avian erythroblastosis virus causes acute erythroleukemia
and fibrosarcomas in chickens. Its genome contains two oncogenes
designated v-erbA and v-erbB(1, 2) .
The v-erbB product is a truncated form of the epidermal growth
factor receptor. The v-ErbA protein is a mutated form of the thyroid
hormone receptor A
comparison of the amino acid sequence of the chicken TR There is evidence to suggest that the DNA recognition sequences of
v-ErbA and TR
Eight rounds of selection were performed in total. In the third EMSA
a v-ErbA-
Co-transfections included 1 µg of a
human growth hormone (GH) expressing vector (pTKGH) per 60-mm Petri
dish to control for transfection efficiency. Cells were transfected in
the presence of 10% charcoal stripped fetal bovine serum and 100 nM dexamethasone. Cells were cultured ± 10 nM T
Figure 1:
EMSA selection of a DNA pool that binds
v-ErbA monomers. DNA pools were end-labeled with
[
A faint slower migrating complex also was seen in the EMSAs ( Fig. 2and Fig. 3). The exact nature of this complex is
not known. The band was supershifted by a specific v-ErbA antibody
(gift of M. Privalsky; data not shown), indicating that it contains
v-ErbA. The band was not affected by an anti-MBP antibody and did not
co-migrate with an MBP-v-ErbA fusion protein-DNA complex, indicating
that the band does not contain MBP or MBP-v-ErbA fusion protein.
Footprinting studies were not possible due to the faint nature of the
protein-DNA complex. This faint band was not present when the probe
size was decreased to 20 bp. Overall the data suggest the faint band
may be a v-ErbA homodimer that requires weak and perhaps relatively
nonspecific interactions with other sequences in the 49-bp probe.
Figure 2:
Affinity of v-ErbA for individual clones
analyzed by competition EMSA.
Figure 3:
Effect of mutations in the sequence
TGAGGTCACG on the binding of v-ErbA analyzed by competition EMSA.
In the case of TR
Figure 4:
Combined guanine methylation interference
and uracil interference footprinting of v-ErbA monomer binding to clone
5. A, the DNA strand of clone 5 containing the sequence
TAAGGTCACG was arbritrarily designated the top strand. F
Figure 5:
A DR+4 of the decamer TAAGGTCACG is a
more potent v-ErbA response element than a DR+4 of the idealized
hexamer AGGTCA in a transient transfection assay. JEG-3 cells were
transfected with the reporter plasmid pUTKAT3 containing a single copy
of either a DR+4 of TAAGGTCACG (RT3) or a DR+4 of AGGTCA
(6DR), along with the internal control plasmid pTKGH. All cells also
received the expression vector for TR
v-ErbA
also caused v-ErbA potentiates the ability of a variety of transmembrane
oncoproteins to transform erythroblasts and
fibroblasts(4, 5) . In animal cells v-ErbA behaves as
a repressor of transcriptional activation regulated by TR and
RAR(9, 10, 22) . v-ErbA arrests the
expression of at least three erythroid genes: carbonic anhydrase II,
erythrocyte anion transporter (band 3), and v-ErbA is closely
related to its cellular counterpart TR It is not completely understood which regions of the v-ErbA protein
account for its specific high affinity interaction with this 10-bp DNA
sequence. It has been suggested for several members of the erbA
superfamily, including v-ErbA, that the P box is responsible for the
recognition of the hexamer AGGTCA. There also is evidence that the A
and T boxes located 3` to the zinc fingers are important in DNA-protein
interaction(31, 32) . In particular, the A box is
critical for recognition of the 5` end of the core DNA motif for the
orphan receptor NGFI-B/nur 77(31) . By analogy the A box in
v-ErbA could determine the specific recognition of the two nucleotides
T(A/G) at the 5` end of the v-ErbA consensus site. It also has been
shown that amino acid sequences outside the DNA binding domain of
v-ErbA play a role in half-site specificity. Thus, the specific
recognition for thymine at position 4 of the hexamer AGGTCA is
determined by two amino-terminal amino acids, His-12 and Cys-32,
interacting with Ser-61 in the P box of the DNA binding
domain(33, 34) . A comparison of the optimal DNA
binding sites of v-ErbA and TR These results suggest that v-ErbA should bind to and repress certain
TREs more than others. Previous data support this concept in that it
was shown that TR In
contrast with the above, the random selection strategy allowed us to
identify two sequences that are essentially equally strong as TREs, yet
v-ErbA differs dramatically in its ability to inhibit the T In
conclusion, our results indicate that the optimal binding site for
v-ErbA is the decamer T(A/G)AGGTCACG. This sequence binds v-ErbA with
higher affinity and also gives stronger suppression of TR
Volume 270,
Number 14,
Issue of April 7, 1995 pp. 7957-7962
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
1 (*)
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
1. Although the basis for the oncogenic action of v-ErbA
is unknown, expression of this protein is known to inhibit thyroid
hormone and retinoic acid induction of target genes. The DNA binding
domain of v-ErbA differs from that of thyroid hormone receptor 1
in two amino acids felt to be crucial for determining the specificity
of DNA binding. However, the DNA binding properties of v-ErbA have not
been examined independent of a comparison of binding to already known
thyroid hormone response elements. In the current studies a non-biased
strategy was used to select from a pool of random DNA those sequences
that bind v-ErbA with high affinity. The highest affinity binding
sequence was identified as the decamer 5`-T(A/G)AGGTCACG, which is
closely related to the optimal thyroid hormone receptor 1 binding
sequence, TAAGGTCA. Transfection studies demonstrate that among equal
thyroid hormone responsive elements, those that contain the optimal
v-ErbA consensus will be repressed by v-ErbA in preference to those
that do not. These studies indicate that v-ErbA and thyroid hormone
receptor 1 regulate overlapping sets of response elements, and
that all sequences that are highly responsive to thyroid hormone are
not necessarily responsive to v-ErbA.
1 (TR1) (
)and belongs to the
large family of zinc finger transcription factors which includes the
receptors for steroids, vitamin D, and retinoic acid
(RAR)(2, 3) . v-ErbA has little to no transforming
capacity of its own in vivo but cooperates with the
transforming properties of a variety of tyrosine kinase-encoding
oncogenes, including v-ErbB and sarcoma oncogenes, as well as with the
Ha-Ras oncogene(4, 5) . v-ErbA blocks spontaneous cell
differentiation and allows tolerance to a wide variation in the pH and
ionic strength of culture media(4, 5) .1 and
v-ErbA proteins reveals that v-ErbA has undergone 13 single amino acid
mutations(6) . Two of these are located in the P and D boxes of
the DNA binding domain, regions that are crucial for determining the
DNA sequence specificity of protein binding(6, 7) .
Additional mutations, including a 9-amino acid deletion, are found in
the carboxyl-terminal domain. As a consequence of these mutations,
v-ErbA expressed in mammalian cells is unable to bind thyroid hormone
(T
)(6, 8) . In mammalian and avian cells
v-ErbA acts as a constitutive dominant repressor of transcription
regulated by TR and RAR, suggesting that these activities may be
central to its oncogenic
activity(9, 10, 11, 12) . However,
the actual mechanisms underlying v-ErbA's oncogenic action are
unknown, as are the key target genes for v-ErbA induced oncogenesis. 1 are related although not necessarily
identical(9, 10) . Therefore the goal of the present
work was to identify the optimal DNA sequence for v-ErbA binding and to
study its function as a v-ErbA response element. A non-biased approach
was taken similar to that employed by Blackwell et
al.(13) , to study the DNA binding of c-Myc. By using
random DNA pool selection, competition DNA binding assays, DNA
footprinting, mutational analysis, and transient transfections, we were
able to identify the optimal v-ErbA binding site as the decamer
5`-T(A/G)AGGTCACG.
Production and Purification of v-ErbA
The v-erbA
cDNA was ligated in-frame into the Escherichia coli expression
vector pMAL (New England Biolabs) to produce a fusion protein
consisting of maltose binding protein (MBP) followed by the recognition
sequence for the protease factor Xa, fused to v-ErbA. Expression and
purification of recombinant v-ErbA were performed according to the
vendor's protocol. Briefly, E. coli strain XL1 was
transformed with the expression plasmid. Expression of the MBP-v-ErbA
fusion protein was induced with
isopropylthio-
-D-galactoside, and the recombinant fusion
protein was purified by amylose affinity chromatography. Purity was
confirmed by SDS-polyacrylamide gel electrophoresis (PAGE) and
Coomassie Blue staining, which revealed only a single band of 
86
kDa, consistent with the expected combined masses of MBP (42 kDa) plus
v-ErbA (44 kDa) (data not shown). Incubation with factor Xa was then
performed to release v-ErbA from MBP. The cleaved MBP was not removed,
since control studies indicated it does not bind DNA. The final
products were shown by SDS-PAGE to be of the appropriate size.Random DNA Pool Construction
Oligonucleotide
primers A (5`-TCCGAATTCCTACAG) and B (5`-AGACGGATCCATTGCA) were
synthesized. A pool of oligonucleotides of 49 nucleotides in length was
synthesized containing the primer A sequence at the 5` end, an internal
random sequence of 18 nucleotides (for each nucleotide addition the
oligonucleotide synthesizer was programmed to add equal amounts of all
4 bases simultaneously), and the reverse complement of primer B at the
3` end. This internally random sequence oligonucleotide pool was
converted to double stranded DNA by annealing with primer B and
filling-in with a Klenow reaction. The double stranded pool was
purified by PAGE.Electrophoretic Mobility Shift Assays
(EMSA)
Protein-DNA binding reactions were performed in 35 µl
of 20 mM HEPES, pH 7.8, 20% glycerol, 1.4 µg of
poly(dI
dC), 1 mM dithiothreitol, 50 mM KCl (in
the later steps of the selection process, up to 400 mM KCl was
employed), 0.1% Nonidet P-40, 
P-labeled DNA, and E.
coli expressed v-ErbA. The amounts of v-ErbA and P-labeled DNA are specified in subsequent sections.
Reactions were incubated at room temperature for 45 min prior to
electrophoresis. Electrophoresis was carried out on 0.25 
TBE
(22 mM Tris base, 22 mM boric acid, 0.5 mM EDTA), 6% polyacrylamide gels (29:1, acrylamide:bisacrylamide) at
room temperature. Wet gels were exposed to film for 24 h at 4 °C
during the selection process. For all other EMSAs, gels were fixed in
30% methanol, 10% acetic acid; dried and exposed to film with an
intensifying screen for 12-36 h at -70 °C.Selection Process
The random pool of double
stranded DNA was end-labeled with [
-P]ATP
by T4 polynucleotide kinase. An EMSA was performed with 40,000 cpm of P-labeled random DNA pool plus 40 ng of non-radiolabeled
random DNA pool and 5 µg of v-ErbA. In the initial round of
selection we expected the DNA-v-ErbA complex to be undectectable.
Therefore we constructed a 52-bp fragment containing the hexamer AGGTCA
as a marker. The location of the marker DNA-v-ErbA complex was used as
a reference to identify the portion of the gel to excise in the random
DNA pool-v-ErbA lane. The DNA was eluted into 0.1% SDS, 0.5 M NH
OAc, 1 mM EDTA, ethanol precipitated, and
amplified by the polymerase chain reaction using primers A and B.
Cycling conditions included 1 min each of denaturation at 94 °C,
annealing at 56 °C, and extension at 72 °C for 30 cycles. The
products were purified by PAGE and the selection process was repeated.
For subsequent selections 40,000 cpm of P-labeled DNA pool
without non-radiolabeled DNA were used in the protein-DNA incubations. P-labeled DNA complex could be visualized. In the
first 2 rounds, the protein-DNA incubations used 50 mM KCl,
and in the first 6 rounds 5 µg of v-ErbA was employed. To select
for higher affinity binding sites, rounds 3-5 and 6-8 were
performed with 150 and 400 mM KCl, respectively, and the
amount of v-ErbA was reduced to 500 ng for rounds 7 and 8. The selected
v-ErbA-binding DNA pool was ligated into the BamHI site of the
plasmid vector pUTKAT3(14) . Individual bacterial clones were
isolated and used to generate plasmid DNA. The sequences of all clones
were determined by the dideoxynucleotide method using vector primers.
Polymerase chain reaction with primers A and B was used to amplify the
contained 49-bp sequences, which were purified by PAGE and used in the
studies described below.Competition Assays
Five clones were randomly
chosen, end-labeled with P and used in an EMSA to test for
v-ErbA monomer binding. The clone showing the strongest protein-DNA
complex (clone 5) was chosen as the standard for further studies. The
sequence of the random 18-mer within clone 5 is CCCAGTCTAAGGTCACGG.
Next the affinity of v-ErbA for all clones was assessed with a
competition EMSA. To accomplish this, 20,000 cpm of P-labeled clone 5 DNA was incubated with 50 ng of v-ErbA
plus graded doses of non-radiolabeled DNA from each clone. The amount
of this competitor DNA required to reduce the intensity of the standard P-labeled clone 5 DNA-v-ErbA complex by 50%
(C
, measured by densitometry of autoradiograms) was taken
as a measure of relative affinity. Competition assays were performed at
least twice for all clones.DNA Footprinting Analysis
Footprinting was
performed with v-ErbA on clone 5 using methylation interference to
identify critical guanine residues and uracil interference to identify
critical thymine residues. Labeling of the DNA and footprinting
protocols were as described previously(15, 16) .Mutational Analysis
The consensus sequence of the
highest affinity clones was subjected to mutagenesis. A series of
mutant oligonucleotides was made changing 1-2 bp at a time
scanning across the consensus sequence. These were used in a C EMSA analysis with clone 5 as a probe. The data were used in
conjuction with the footprinting data to assess which bases are most
important for protein-DNA binding.Transient Transfections
JEG-3 cells were grown in
90% Eagle's minimum essential medium plus 10% fetal bovine serum
and were transfected using standard calcium phosphate
precipitation(17) . TR1 and v-ErbA were expressed from the
vectors pCDM (17) and pRSV(18) , respectively.
Transfections included 100 ng of pCDMTR1, 3 µg of pRSV-v-erbA
(or vector) plus an additional 3 µg of pCDM as ``filler''
plasmid. Potential v-ErbA response elements were ligated into pUTKAT3
at a BamHI site 5` to the basal herpes simplex virus thymidine
kinase promoter driving expression of chloramphenicol acetyltransferase
(CAT). The top strand sequences of the putative v-ErbA response
elements are as follows (top strands excluding BamHI
compatible overhangs): RT3, TAAGGTCACGTAAGGTCAC; and 6DR,
AGCAGGTCATAGCAGGTCAG. Reporter plasmids were transfected at a dose of 4
µg/60-mm Petri dish. for 2 days prior to harvest. CAT and hGH assays were performed as
described previously(17) . v-ErbA suppression of CAT reporter
gene expression was calculated as CAT/hGH for cells cultured with
v-ErbA divided by CAT/hGH for cells cultured without v-ErbA in either
the presence or absence of T. Results are presented as the
mean ± S.E. for four independent transfections per assay
condition.
Selection of High Affinity v-ErbA Monomer-binding DNA
Pool
The random DNA selection strategy lead to the isolation of
a DNA pool that bound v-ErbA with high affinity (Fig. 1). This
v-ErbA monomer-binding DNA pool was subcloned and 17 different clones
were analyzed. Ten of these clones contained the 9-bp sequence
T(A/G)AGGTCAC ( Table 1and Table 2). Also a guanine at
position 10 was found in 5 of these 10 clones. Those clones with the
perfect decamer sequence had the highest affinities with C values of 0.8 ± 0.1 ng. Clones containing the 9-bp
consensus but without a guanine at position 10 showed modestly inferior
binding with C values of 1.3 ± 0.17 ng. All other
clones had substantially lower affinities, with C values
ranging from 1.8 to greater than 4 ng. Neither the position nor
orientation of the decamer within the 18-bp random sequences was
conserved, nor was there conservation of sequence outside the decamer.
-P]ATP, incubated with recombinant v-ErbA
or buffer, and subjected to PAGE. The v-ErbA-DNA complex is invisible
following incubation with the random pool DNA (lane 2), but is
clearly seen following 6 rounds of selection (lanes 3 and 4, indicated by the arrow). v-ErbA-DNA binding was
performed in the presence of 50 mM KCl except in lane
4, where 400 mM was used to select for higher affinity
protein-DNA interactions.
P-Labeled clone 5 was
incubated with 50 ng of recombinant v-ErbA plus various amounts of
non-radiolabeled competitor DNAs, and protein-DNA complexes were
analyzed by EMSA. The v-ErbA-DNA monomer complexes are indicated by the arrow. The dose of competitor DNA that competes 50% of the
v-ErbA-P-labeled clone 5 complex (C) was
determined by densitometry and used as a measure of relative affinity.
A competition assay using both a high affinity clone (clone 5; C of 0.7 ng) and a low affinity clone (clone 35; C of
3.4 ng) is shown. A faint slower migrating complex (indicated by the asterisk) was seen in these EMSA studies, and may represent a
v-ErbA homodimer interacting weakly with other sequences in the 49-bp
probe.
P-Labeled clone 5 was incubated with 50 ng of recombinant
v-ErbA plus the indicated amounts of non-radiolabeled competitor DNAs.
Competitors were synthetic 20-bp oligonucleotides; the top strand
sequences are shown, except that all also contain GATCT at the 5` end
and CGATC at the 3` end. The first competitor contains the optimal
decamer TGAGGTCACG, and it shows 50% competition at 0.1 ng (A). All other competitors contain mutations, shown with lower case letters in the sequence (B-D). Mutant
competitors are weaker than the consensus, thus confirming assignment
of the optimal sequence. The asterisk denotes a faint slower
migrating complex as described in Fig. 2.
Mutational Analysis
To confirm that the
nucleotides immediately 5` and 3` to the idealized AGGTCA hexamer are
important in v-ErbA binding, a series of mutant oligonucleotides was
synthesized and their affinities were tested by EMSA competition (Fig. 3). The oligonucleotide containing the optimal decamer
sequence TGAGGTCACG had a C of 0.1 ng. Mutation of the
first, second, or fourth plus fifth base pairs decreased the affinity
of v-ErbA binding significantly, resulting in C values of
0.45, 0.9, and >16 ng, respectively. Furthermore, mutating the
dinucleotide at positions 9 and 10 from CG to AT also decreased the
affinity, resulting in a C of 0.35 ng. In contrast,
changing the second nucleotide from guanine to adenine did not affect
the affinity, confirming that either purine is optimal at that
position.1, mutating the two nucleotides
immediately 5` to the hexamer AGGTCA from TA to GC decreased the
affinity of the site for TR1 5-fold(15) . However,
mutations of the two nucleotides immediately 3` of the AGGTCA did not
affect the affinity for TR1 (data not shown).DNA Footprinting
Footprinting was performed using
clone 5 containing the conserved sequence TAAGGTCACG. Methylation
interference was employed to identify critical guanine residues and
uracil interference was used to identify critical thymine residues. As
shown in Fig. 4, when the guanine at position 4, 5, 7, or 9 from
either strand in the decamer sequence was methylated, v-ErbA binding
was impaired. In a similar way, loss of the thymine 5-methyl group by
changing the bottom strand thymine to uracil at position 8 within the
decamer interfered with v-ErbA binding. No evidence of methylation
sensitivity was noted outside the conserved decamer.
and F
represent the free DNAs cleaved either at guanines alone or
guanines plus uracils, respectively. B represents v-ErbA-bound DNA cleaved at guanines plus
uracils. Footprinted bases are indicated by closed circles. B,
the DNA sequence of clone 5 is indicated, with footprinted bases
designated by closed circles.
Transient Transfection Studies
The natural ligand
for v-ErbA has not been identified (or does not exist) and this
oncoprotein behaves as a transcriptional repressor in mammalian cells.
Therefore, to test the function of potential v-ErbA response elements
we examined their ability to direct v-ErbA repression of
T-dependent CAT expression mediated by TR1. In
particular we were interested in determining whether the decamer
TAAGGTCACG is a stronger v-ErbA response element than the idealized
hexamer AGGTCA. Since it had previously been shown that the optimal
spacing for T response elements (TREs) is a direct repeat
with a 4-bp spacer (DR+4) (19, 20) , we
constructed oligonucleotides that contained a DR+4 of AGGTCA (6DR)
or of TAAGGTCACG (RT3) (in the latter case the last 2 nucleotides, CG,
of the 5` half-site and the first 2 nucleotides, TA, of the 3`
half-site function as the 4-bp spacer, since the sequence is a decamer
rather than an hexamer). These oligonucleotides were ligated into
pUTKAT3 and tested for T induction. (Cotransfection with
TR1 was required to generate a T response, consistent
with prior studies and the known very low level of endogenous TRs in
JEG-3 cells(21) ). As shown in Fig. 5, these two
sequences are similarly strong TREs. However, as predicted by the DNA
binding data, v-ErbA was a much stronger repressor of T induction from RT3 than from 6DR. Specifically, v-ErbA suppressed
T mediated CAT activity on RT3 to 58 ± 2% of that
observed in the absence of v-ErbA, but was without effect on 6DR (95
± 7% of that observed in the absence of v-ErbA; n = 4 for both). In addition, in the absence of T v-ErbA suppressed base-line CAT activity a modest but
reproducible 15% on RT3 but did not suppress CAT activity on 6DR. Taken
together these data demonstrate that not only does v-ErbA bind to the
sequence T(A/G)AGGTCACG with higher affinity than to AGGTCA, but also
the decamer is more potent as a functional v-ErbA half-site.
1 ± v-ErbA (or empty
vector). Cells were cultured for 2 days in the presence or absence of
10 nM T and then cell lysates were analyzed for
CAT activity and medium for hGH. v-ErbA suppression is defined as
CAT/hGH for cells cultured with v-ErbA divided by CAT/hGH for cells
cultured without v-ErbA.
60% suppression of T-mediated CAT activity
when a single copy of the optimal decamer was used as a monomer T response element (data not shown). However, the hexamer AGGTCA
used as a single site is not T responsive (15) ,
and therefore cannot be tested for v-ErbA suppression.
-aminolevulinate
synthase. However, the block in erythroid differentiation is not
related to the suppression of any of these three genes(23) .
Although evidence suggests that the oncogenic function of v-ErbA
correlates with its ability to interfere with RAR action on a synthetic
palindromic retinoic acid response element
(RARE)(22, 24) , naturally occurring RAREs appear to
be direct repeats, not
palindromes(25, 26, 27, 28) .
Despite major advances in our understanding of v-ErbA function, the
genetic targets that account for its oncogenic action remain unknown.
Thus, it is not clear if this oncogenic function is based on
suppression of RAR or TR function, or if it is mediated through
uncharacterized v-ErbA specific responsive genes.1. In the DNA binding
domain, these proteins differ in two amino acids located in the P and D
boxes, regions known to be crucial for DNA specific
recognition(6, 7) . The importance of the unique P box
serine in v-ErbA has been addressed by mutating it to glycine as is
found in TR1 (S61G)(22, 29) . v-ErbA carrying
this S61G mutation is unable to effect erythroid neoplasia. This result
is consistent with the notion that this serine is crucial to the
oncogenic function of v-ErbA, and suggests that this oncogenic function
requires different DNA binding properties than are found in TR1.
However, interpretation of these data is somewhat complicated by the
observation that a v-ErbA in which both the P and D box amino acids are
mutated to those found in TR1 (S61G, T78K) has partial restoration
of its transforming properties relative to the S61G
mutant(30) . Nevertheless, these studies suggest that v-ErbA
and TR1 have distinct DNA binding properties, and that these
differences play a role in the function of v-ErbA. Thus,
characterization of the optimal binding sites for v-ErbA should provide
insight into the mechanism of its oncogenic action, and might suggest
potential target genes for this oncoprotein. With this as a rationale,
we utilized a non-biased random DNA selection strategy to identify the
optimal DNA binding site for v-ErbA as the decamer T(A/G)AGGTCACG. 1 (15) shows that they are
highly related but not identical (T(A/G)AGGTCACG versus TAAGGTCA, respectively). In general v-ErbA appears to have a
stricter specificity for binding at the 3` end of the recognition
sequence, and TR1 at the 5` end. This is true not only based upon
the sequences selected from the random pools and their mutagenesis, but
is supported by the footprinting data. Thus, v-ErbA but not TR1
shows footprints on the bottom strand opposite the top strand eight and
ninth nucleotides (AC) of the recognition sequence. In contrast,
TR1 (15) but not v-ErbA shows footprints of the top strand
thymine at position 1 and the bottom strand thymine at position 2. 1 can bind to the hexamer AGGACA whereas v-ErbA
cannot(33) . Consistent with this binding data, v-ErbA was not
able to repress T induction from a TRE comprised of two
copies of the AGGACA hexamer(33) . However, TREs containing the
sequence AGGACA are functionally weak relative to those containing
AGGTCA. We find that TR1 is 6 times more potent in activating CAT
expression on a reporter containing a thymine than on a reporter
containing an adenine at position 4 of the hexamer (TRE's,
TAAGGTCACGTAAGGTCACG versus TAAGG ACACGTAAGGACACG; data not
shown). Thus, TREs comprised of the hexamer AGGACA are weak T inducers but even weaker at allowing v-ErbA suppression. response on these two elements. Furthermore, this function of
v-ErbA correlates directly with its relative binding affinities for
these DNA sequences. This suggests that among equally T responsive genes, some will be repressed by v-ErbA substantially
more than others. Thus, one can use DNA sequence to begin to predict
which genes are likely to be repressed by v-ErbA to the greatest
degree. In addition, one may speculate that cells may contain an
endogenous repressor with activity similar to v-ErbA, and that this,
too, might regulate a specific subset of TREs based upon sequence. In
this regard it is interesting to note that the orphan receptors COUP-TF (35) and ARP-1 (36) share identical P boxes with
v-ErbA. Furthermore, COUP-TF has been shown to repress T action(37) , although a detailed analysis of its DNA
binding specificity and function has not been reported.1
mediated activation than does the hexamer AGGTCA. Despite the amino
acid differences in the P and D boxes, the optimal binding sites for
TR1 and v-ErbA are very similar. The identification of new,
naturally occurring v-ErbA binding sites will be an essential component
of understanding its oncogenic action.
), thyroid
hormone (3,5,3`-triiodothyronine); bp, base pair(s).
We thank M. Privalsky for providing the v-ErbA
antiserum and R. Evans for providing the v-erbA cDNA and the Rous
sarcoma virus expression vector. We thank David Olson, Ron Katz, and
Eric Beninghof for advice and technical assistance.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L. J. DeLong, G. M. C. Bonamy, E. N. Fink, and L. A. Allison Nuclear Export of the Oncoprotein v-ErbA Is Mediated by Acquisition of a Viral Nuclear Export Sequence J. Biol. Chem., April 9, 2004; 279(15): 15356 - 15367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Subauste and R. J. Koenig Characterization of the DNA-Binding and Dominant Negative Activity of v-erbA Homodimers Mol. Endocrinol., September 1, 1998; 12(9): 1380 - 1392. [Abstract] [Full Text]
|
|
|
|
|
|
Y.-Z. Yang, M. Burgos-Trinidad, Y. Wu, and R. J. Koenig Thyroid Hormone Receptor Variant alpha 2. ROLE OF THE NINTH HEPTAD IN DNA BINDING, HETERODIMERIZATION WITH RETINOID X RECEPTORS, AND DOMINANT NEGATIVE ACTIVITY J. Biol. Chem., November 8, 1996; 271(45): 28235 - 28242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Judelson and M. L. Privalsky DNA Recognition by Normal and Oncogenic Thyroid Hormone Receptors J. Biol. Chem., May 3, 1996; 271(18): 10800 - 10805. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Shen and J. S. Subauste Dimerization Interfaces of v-ErbA Homodimers and Heterodimers with Retinoid X Receptor alpha J. Biol. Chem., December 22, 2000; 275(52): 41018 - 41027. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |