| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Enhancer In Vivo1
* Basel Institute for Immunology, Basel, Switzerland; and
Department of Immunology, Duke University Medical Center, Durham, NC 27710
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
locus,
the TCR enhancer and the TCR enhancer (E), are responsible
for orchestrating the distinct developmental programs for V(D)J
recombination and transcription of the TCR and genes,
respectively. E function depends critically on transcription factors
core binding factor (CBF)/polyoma enhancer-binding protein 2 (PEBP2)
and c-Myb as measured by transcriptional activation of transiently
transfected substrates in Jurkat cells, and by activation of V(D)J
recombination within chromatin-integrated substrates in transgenic
mice. To understand the molecular mechanisms for synergy between these
transcription factors in the context of chromatin, we used in vivo
footprinting to study the requirements for protein binding to E
within wild-type and mutant versions of a human TCR minilocus in
stably transfected Jurkat cells. Our data indicate that CBF/PEBP2 plays
primarily a structural role as it induces a conformational change in
the enhanceosome that is associated with augmented binding of c-Myb. In
contrast, c-Myb has no apparent affect on CBF/PEBP2 binding, but is
critical for transcriptional activation. Thus, our data reveal distinct
functions for c-Myb and CBF/PEBP2 in the assembly and function of an
E enhanceosome in the context of chromatin in
vivo. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, 
, and genes, and at the
CD4+CD8+ (DP) stage for the
TCR gene (1). Enhancers and promoters within the TCR
loci functionally collaborate to modify chromatin structure in a
lineage-specific, locus-specific, and developmental stage-specific
manner, thereby imparting developmental control to the processes of
transcription and V(D)J recombination (1, 2, 3, 4). They do so
by binding unique combinations of ubiquitous and cell- and
stage-specific transcription factors.
The disparate temporal programs of TCR and TCR gene
rearrangement during thymocyte development are particularly notable
because the TCR gene is embedded within the TCR gene
(5). The unique genomic structure of the TCR locus
dictates that the TCR gene is deleted from the chromosome upon V
to J recombination at the DP stage. This process irreversibly
commits DP thymocytes to the -T cell lineage. Because of these
properties, the locus represents an interesting model to study
developmental stage-specific regulation of transcription and V(D)J
recombination. Two enhancers at the TCR locus, TCR enhancer
(E) and TCR enhancer (E), are responsible for orchestrating
the distinct developmental programs for V(D)J recombination and
transcription of the TCR and genes (6, 7, 8). During
thymocyte development, E is "off" and E is "on" at the DN
stage, whereas E is on and E is off at the DP stage
(7). These observations provide the foundation for a model
in which these two enhancers work as a "developmental switch" for
the activation of transcription and V(D)J recombination at the
TCR locus.
E function has been analyzed at the level of transcription within
transiently transfected reporter constructs in Jurkat cells and at the
level of V(D)J recombination within chromatin-integrated substrates in
transgenic mice. These studies have demonstrated a critical synergism
between core binding factor (CBF)/polyoma enhancer-binding protein 2
(PEBP2), a member of the Runx family of transcription factors, and
c-Myb, bound to two precisely spaced binding sites within the E3
element of the human E (9, 10, 11). To understand the
molecular mechanism for functional synergy between CBF/PEBP2 and c-Myb,
we previously performed a detailed analysis of binding of these factors
in vitro, and found that they appear to bind independently, rather than
cooperatively, to DNA (12). These data led to a model in
which synergy between these transcription factors may result from their
ability to form a composite surface that makes simultaneous and
stereospecific contacts with additional components of the
transcriptional machinery.
To activate transcription in vivo, transcription factors must be able
to overcome the repressive effects of chromatin to bind to their
specific DNA-binding sites. To establish the molecular basis for
synergy between CBF/PEBP2 and c-Myb at E in vivo, we have analyzed
the requirements for factor binding to this enhancer by genomic
footprinting and DNase I hypersensitivity. In contrast to previous in
vitro studies (12), we found that these transcription
factors do not independently occupy their sites in E in a native
chromatin context. Rather, protein binding to the CBF/PEBP2 site
appears to induce a conformational change in the E3,4 enhanceosome
that is associated with elevated c-Myb binding. CBF/PEBP2 appears to
play primarily a structural role, as it appears not to be required for
transcriptional activation per se. In contrast, c-Myb has no apparent
effect on CBF/PEBP2 binding, but is critical for the activation of
enhancer-dependent transcription. Thus, c-Myb and CBF/PEBP2 appear to
play distinct roles in the assembly and function of an E
enhanceosome within chromatin in vivo.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human T leukemia cell line Jurkat, the human T
leukemia cell line Molt-13 and the human B cell leukemia cell line Raji
were grown in RPMI 1640 medium (Life Technologies, Gaithersburg, MD)
supplemented with 10% FCS (Atlanta Biologicals, Norcross, GA) and
penicillin-streptomycin (Cellgro-Mediatech, Herndon, VA).
Stable transfections
A total of 5–20 x 106 Jurkat cells
were resuspended in 400 µl of complete medium, and kept on ice for 15
min. Cells were then electroporated at 400 µF and 230 V using 10–20
µg of KpnI-linearized and ethanol-precipitated TCR
minilocus DNA, and 1–2 µg of XbaI-linearized and
ethanol-precipitated pFNeo plasmid. Cells were kept for 15 min on ice
after electroporation, diluted in 6 ml of complete medium in
30-mm2 plates, and incubated for 24–48 h before
selection. Transfected cells were selected by adding 1–1.5 mg/ml of
gentamicin. After a week of selection, surviving cells were isolated by
Ficoll filtration and cloned at 1 cell/well in 96-well plates. Positive
clones were detected on slot blots using 5 µg of purified DNA, and
transgene copy number was determined by genomic DNA analysis, comparing
with tail DNAs from previously identified single-copy transgenic mice.
Slot blots were probed with a radiolabeled C fragment and the
resultant hybridization signals were quantified using a
PhosphorImager (Molecular Dynamics, Sunnyvale, CA). Independent
clones were identified by Southern blot of BamHI digested
genomic DNA and a -32P-labeled V1 probe
(13). Structure of the transgenes were examinated by
sequential hybridizations of these same Southern blots with
-32P-labeled V2 and J3 probes, and by
PCR amplifications of J3, J1, and C-containing minilocus
regions.
Genomic footprinting
Genomic footprinting was performed as described
(14). Oligonucleotides for E top-strand analysis were
GCTGAGAAGCTCAACTAAAAGACTG, CTGATTCTGTTTCAGTCACTCAGGGC, and
CTGTTTCAGTCACTCAGGGCAGGAAAC. Those for E bottom-strand analysis
were TATCCAACTAAATCAGACCAGGATTAAG,
TAACTTGTAACTCCCTTGAAAGTCAGCC, and
CCCTTGAAAGTCAGCCAGAGTATGTCTC.
DNase I hypersensitivity assays
DNase I hypersensitivity assays were performed as previously
described with minor modifications (7, 15). Briefly, cells
were washed twice in PBS and permeabilized by incubation with 0.067
mg/ml of lysolecithin in buffer C (0.15 M sucrose, 80 mM KCl, 5 mM
MgCl2, 0.5 mM CaCl2, and 30
mM HEPES pH 7.4) for 4 min at room temperature. Cell suspensions were
incubated in a water-ice bath for 5 min before transferring them to an
ice bath for DNase I (Worthington Biochemical, Freehold, NJ)
treatments. DNase I was then added and digestions were incubated for 10
min on ice. Reactions were stopped by adding EDTA (pH 8) to a final
concentration of 10 mM. Cells were lysed by incubating them at 37°C
for 16 h in the presence of 0.4 mg/ml of proteinase K and 0.4% of
SDS. Purified DNA was digested with SacI, electrophoresed
through a 0.9% agarose gel, transferred to a nylon membrane (Micron
Separations, Westboro, MA), and hybridized with a
-32P-labeled 1.1-kb J3 genomic fragment as
previously described (15).
Northern blot
Total RNA samples were isolated from transfected cells as
described (16). Purified RNA (10 µg/sample) was
electrophoresed through a 1.5% agarose gel containing 2.2 M
formaldehyde and transferred to a nylon membrane (Micron Separations),
and sequentially hybridized with -32P-labeled
C and GAPDH DNA probes.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
occupancy in vivo by genomic
footprinting
To elucidate the molecular basis for functional synergism between
the proteins that bind to E, we decided to study the requirements
for factor binding to this enhancer in vivo by genomic footprinting.
This technique allows one to detect protein-DNA contacts in vivo at
single nucleotide resolution. For this purpose, genomic DNA was treated
with dimethylsulfate (DMS) either as naked DNA in vitro or as chromatin
in live cells. DMS methylates accessible guanines at the N7 position
and, less efficiently, adenines at the N3 position. Sites of
methylation were mapped by ligation-mediated PCR and, by comparing the
methylation patterns in the two samples, footprints indicating occupied
DNA binding sites in vivo were identified.
We first analyzed in vivo transcription factor binding to the
endogenous human E in the human T cell line Molt-13 (Fig. 1
A). This analysis revealed
clear and strong occupancy of all previously identified and
functionally relevant binding sites for transcription factors within
the enhancer. Protein binding to the CBF/PEBP2 and c-Myb binding sites
within E3 was comparable to that previously observed in the analysis
of the endogenous murine E in DN thymocytes (7).
Occupancy of the CBF/PEBP2 binding site was visualized as two protected
guanines on the top strand. Occupancy of the c-Myb binding site was
detected as one protected guanine and one lightly hypersensitive
adenine on the top strand, and one protected guanine and one
hypersensitive guanine on the bottom strand. In addition to the E3
binding sites, occupancy of the two GATA-3 binding sites present in the
human E4 element was also clearly detected (17, 18, 19).
The upstream GATA-3 binding site was visualized as one protected
guanine and one hypersensitive guanine on the bottom strand, whereas
the downstream GATA-3 site was detected as one protected guanine on the
bottom strand. A summary of the footprints obtained from the analysis
of E occupancy in Molt-13 cells is presented in Fig. 1B.
As expected from the T cell-specific expression of E binding
proteins (9, 17, 18, 19, 20, 21), in vivo E occupancy was T
cell-specific and no footprints were detected in the human B cell line
Raji (data not shown).
|
in a homogenous cell
population, we studied the occupancy of wild-type and mutant forms of
E in chromosomally integrated templates in stably transfected Jurkat
cells. This human T cell line has been used extensively for the
study of E structure and function (9, 20, 21). Of note,
because the endogenous TCR gene has been deleted by TCR gene
rearrangement, there is no endogenous enhancer to interfere with the
analysis of a transfected E in these cells. We generated stably
transfected Jurkat cell clones using our previously characterized
unrearranged human TCR minilocus containing either wild-type or
mutant Es (10, 11, 22) (Fig. 1
C). This 22-kb
construct consists of unrearranged V1, V2, D3, J1, and
J3 gene segments, C, and E in its natural intronic location.
The generation of fully rearranged VDJ products within this construct
in transgenic thymocytes is E-driven and depends on the presence of
an intact E, because introduction of point mutations in either the
c-Myb or the CBF/PEBP2 binding site abrogates the ability of E to
activate V(D)J recombination (10, 11). We obtained four
different transfected clones carrying a TCR minilocus with a
wild-type E. These clones were denoted W2, W6, W7, and W70. W2
carries a single copy of the transgene that is truncated at the 5' end,
W7 and W70 carry two copies of the transgene, and W6 carries four
copies of the transgene. Analysis E occupancy in these clones
revealed footprints indistinguishable from those obtained in occupancy
analysis of the endogenous E in Molt-13 cells (see Figs. 1A, 2, 3, and 4). Thus, transgene integration into
ectopic sites does not significantly perturb E occupancy.
Occupancy of the CBF/PEBP2 binding site is required for a distinct
E3,4 enhanceosome conformation and enhanced c-Myb binding
To investigate the molecular basis for functional synergy between
CBF/PEBP2 and c-Myb, we compared the in vivo occupancy of wild-type and
mutant versions of E by genomic footprinting. Previous studies of
enhancer mutants mMyb and mCore (Fig. 1C) revealed the
mutations to abrogate the binding of c-Myb and CBF/PEBP2, respectively,
as determined by EMSA, and to eliminate both enhancer-dependent
activation of transcription in transiently transfected Jurkat cells and
V(D)J recombination in transgenic mice (9, 10, 11, 21). We
obtained five different stably transfected cell clones containing the
human TCR minilocus with an EmMyb and four different clones with
an EmCore. Transfectants containing EmMyb were denoted M3, M20,
M36, M53, and M62, and those containing EmCore were denoted C63,
C64, C74, and C81. C81 carries a single copy of the transgene that is
truncated at the 5' end. C63, C64, M3, M20, M36, and M53 carry two
copies of the transgene, C74 carries four copies of the transgene, and
M62 carries eight copies of the transgene.
We first analyzed in vivo enhancer occupancy in the transfectants
carrying EmMyb (see Fig. 3). Taking into account loading differences
among the various lanes, we found no evidence of occupancy of the
mutated c-Myb binding site. However, the CBF/PEBP2 binding site was
clearly occupied in the presence of the c-Myb site mutation (see Fig. 3). These results are consistent with our previous in vitro binding
studies, which indicated independent binding of c-Myb and CBF/PEBP2 to
E (12). Similarly, protected guanines over the two
GATA-3 sites were indistinguishable from those detected within
wild-type E-containing clones (Figs. 2 and 3). Although the hypersensitive
guanine on the bottom strand of the upstream GATA-3 binding site was
less obvious in M53 and M62 in the experiment shown in Fig. 3, hypersensitivity at this site was confirmed in other analyses of the
same clones (data not shown). Thus, our data indicates that binding of
c-Myb to E is not required for CBF/PEBP2 and GATA-3 to gain access
to and stably occupy their DNA binding sites within the enhancer. In
agreement with in vitro experiments (Ref. 12 and data not
shown), both CBF/PEBP2 and GATA-3 can bind independently of c-Myb to
the enhancer in vivo.
|
|
mCore
transfectants revealed unexpected results. In the absence of CBF/PEBP2
binding, footprints over the c-Myb binding site were partially
inhibited, as judged by the diminished protection of top- and
bottom-strand guanines (Fig. 4).
Moreover, although the hypersensitive adenine on the top strand of the
c-Myb site was present in all EmCore clones, the bottom-strand
hypersensitive guanine was absent. In addition, although protection of
the bottom-strand guanine over the upstream GATA-3 sites was present in
all the EmCore clones, the hypersensitive bottom-strand guanine over
this site was absent (Fig. 4). No change in footprinting was detected
at the downstream GATA-3 site.
|
The hypersensitive guanines at the edges of the c-Myb and
upstream GATA-3 sites should reflect a unique DNA conformation that
renders these nucleotides particularly accessible to DMS. Therefore, we
interpret the loss of these hypersensitive nucleotides in the absence
of CBF/PEBP2 to mean that this conformation has been disrupted. Because
the GATA-3 site hypersensitivity is retained in EmMyb but not in
EmCore, it is not a consequence of GATA-3 binding per se, but rather
of binding of GATA-3 together with CBF/PEBP2. Similarly, because
EmCore displays residual c-Myb binding but lacks the c-Myb site
hypersensitivity, the c-Myb site hypersensitivity must
require simultaneous binding of c-Myb and CBF/PEBP2, rather than
binding of c-Myb alone. Thus, CBF/PEBP2 appears to play a unique and
central role in the assembly of a multiprotein complex on the enhancer
by coordinating a distinct conformation for the E3,4 enhanceosome,
and by promoting elevated occupancy of the c-Myb site.
DNase I hypersensitive sites (HSs) result from local distortions of the
canonical nucleosomal structure due to binding of transcription factors
(23). To confirm the results of genomic footprinting
analysis, we analyzed chromatin structure over the enhancer and
surrounding areas by DNase I digestion. For this purpose, transfected
clones containing wild-type or mutant versions of E were
permeabilized with lysolecithin and treated with increasing amounts of
DNase I. DNA was then purified and digested with SacI for
Southern blot analysis, as previously described (15). A
1.1-kb J3 probe detected the predicted 10-kb SacI DNA
fragment in all clones (Fig. 5). Clones
W6, M53, M62, and C63 also displayed one or two strong clone-specific
bands that are presumably a consequence of truncations at the 3' end of
one transgene copy (Fig. 5). These truncations do not affect detection
of HSs because the probe detects the 5' end of the SacI
fragment and the truncations occur 3' of all HSs. Several clear DNase I
HSs were found in the analysis of the transfected clones containing a
wild-type enhancer (Fig. 5). The strongest mapped to E, presumably
reflecting chromatin distortion due to protein binding at the E3 and
E4 elements (denoted E3/4 HS). This HS is lineage-specific
because it was detected at the endogenous locus in Molt-13 but not in
Raji cells (data not shown). Two additional very clear HSs (Fig. 5, arrows) seem to map to two described matrix-attachment regions (MARs)
that flank E (24). In addition, we detected an HS
located upstream of the J3 gene segment, which may identify a
promoter associated with transcription and recombination of this
segment (denoted 5' J3). Identical results were obtained for all
transfectants containing the wild-type E (Fig. 5 and data not
shown).
|
mMyb transfectants were identical with those
in wild-type transfectants (Fig. 5). The E3/4 HS is presumably a
consequence of normal binding of CBF/PEBP2 and GATA-3. These results
are consistent with those obtained by genomic footprinting analysis of
the mutant enhancer, and suggest that similar local distortions of
chromatin are present in both E and EmMyb. However, different
results were obtained from analysis of transfectants containing
EmCore (Fig. 5). A diminished E3/4 HS signal was detected in
EmCore clones C63 and C74 (Fig. 5), and no signal was detected in
clones C64 and C81 (data not shown). Diminution of the E3/4 HS in
EmCore as compared with EmMyb or E clones is highly
reproducible and can be readily judged with reference to the HSs
associated with flanking MARs (Fig. 5 and data not shown). These data
are consistent with the diminished protections and altered
conformational properties of the mutant enhancer detected by in vivo
footprinting, and indicate that a mutation in the CBF/PEBP2 binding
site results in a chromatin structure that is distinct from that
conferred by either a wild-type E or EmMyb. The CBF/PEBP2 binding site is dispensable for enhancer-dependent transcriptional activation in Jurkat transfectants
To analyze the functional effects of mutations in the c-Myb and
CBF/PEBP2 binding sites in this system, we measured transcription in
the transfected clones by Northern blotting (Fig. 6). Previous functional analysis of TCR
minilocus transgenic mice demonstrated the presence of two
independently regulated domains within the construct (22).
The 5' domain contains the V and D gene segments and rearranges
independently of E, whereas the 3' domain contains the J gene
segments and C region and rearranges only in the presence of an
intact E. Hence, as a measurement of E-dependent transcription,
we analyzed C transcripts which are dependent on the presence of an
active enhancer within the minilocus (15). As controls, we
analyzed C transcripts derived from the endogenous TCR gene
in Molt-13 and Jurkat cells, and from the TCR minilocus in cells
from the mouse transgenic lines A and recombination-activating gene
2-/- x A (R x A) (3, 22) (Fig. 6). As expected, high levels of C transcripts were
found in Molt-13 cells, whereas no C transcripts were detected in
Jurkat cells. Previous data demonstrated that E function is
down-regulated at the transition from DN to DP thymocytes
(7). Consistent with this, high levels of transcription
were detected in thymocytes of line R x A, which contains only DN
cells (3), low levels of transcription were detected in
total thymocytes from line A, which contains mostly DP cells, and no
transcription was detected T cells from line A.
|
transcripts detected in minilocus
transfectants of Jurkat initiate upstream of J3 (J.
Carabaña and M. S. Krangel, unpublished observations),
probably driven by a promoter located 5' of the J3 gene segment
(Fig. 5). Three of three transfectants carrying a wild-type E
minilocus with an intact J3 5' region displayed clear evidence for
C transcripts (Fig. 6). Similar results were obtained for four
additional wild-type transfectants (J. Carabaña and M.
S. Krangel, unpublished observations). However, C transcripts were
not detected in four of five EmMyb clones (Fig. 6). The outlier,
M62, may have a minilocus integration site that is near an active
enhancer. These results confirm that c-Myb binding is critical for E
function, and that CBF/PEBP2 and GATA-3 binding is not sufficient for
activity.
In contrast to the results obtained from the analysis of EmMyb
clones, C transcripts were clearly detected in all EmCore clones
containing an intact J3 5' region (Fig. 6). Furthermore, the levels
of C transcription detected in these clones were roughly similar to
those detected in wild-type clones. Thus, binding of c-Myb to E is
required for transcriptional activation, whereas binding of CBF/PEBP2
is not. Although average occupancy of the c-Myb site is somewhat
reduced in the absence of CBF/PEBP2, these results indicate that
transcriptional activation occurs normally in those cells in which
c-Myb is bound and occurs independent of the CBF/PEBP2-dependent
conformational change in the E3,4 enhanceosome. Thus, the data
reveal distinct functions for CBF/PEBP2 and c-Myb, with CBF/PEBP2
playing primarily a structural role, and c-Myb required for
transcriptional activation.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
within
chromatin does not occur independently, and that CBF/PEBP2 both
stimulates the binding of c-Myb and plays a central structural role in
the assembly of an appropriately configured E3,4 enhanceosome. We
suggest that a similar interplay between CBF/PEBP2 and c-Myb might
occur at other elements regulated by this pair of factors, such as E
(33, 34), the long-terminal repeats of murine leukemia
viruses (35), and the myeloperoxidase promoter
(32), and might occur between CBF/PEBP2 and other
transcription factors, as recently suggested for CBF/PEBP2 and a
vitamin D-responsive transcription factor at the rat osteocalcin
promoter (36). Transcription factors use several molecular mechanisms to stably occupy their binding sites within chromatin. Previous in vitro analysis has demonstrated that binding of disparate transcription factors to nucleosomal DNA is inherently cooperative (37, 38). Polach and Widom (38) proposed a theoretical model that explains how regulatory proteins that bind independently to naked DNA might bind with reciprocal cooperativity to nucleosomal target sites. They suggest that nucleosomes are dynamic structures that transiently expose factor binding sites with the exposed configuration stabilized by the binding of multiple factors to the same nucleosome. Because our data suggest that CBF/PEBP2 promotes c-Myb occupancy but that c-Myb does not impact CBF/PEBP2 occupancy, cooperativity in this case does not appear to be reciprocal.
In some cases, a specific factor plays a critical role in initiating the assembly of a functional multiprotein complex (39, 40). This is the case for inducible transcription factors such as nuclear receptors. The hierarchy of factor binding depends on the position of binding sites with respect to the nucleosome surface (37, 41, 42, 43). Thus, binding of one transcription factor to a peripheral site on a nucleosome may induce a structural change that stimulates the binding of additional factors to internal sites on the same nucleosome. Therefore, CBF/PEBP2 might bind at the edge of a precisely positioned nucleosome and stimulate access to the c-Myb site in this manner.
A specific factor may also play a critical role in coordinating the
assembly of a functional multiprotein complex. This is the case for
structural proteins such as HMG-I at the -IFN enhancer and T
cell-specific factor-1/lymphoid enhancer-binding factor-1 at
E (26, 44, 45). Recent crystal structure analysis of
the CBF/PEBP2-DNA complex revealed an interaction of CBF/PEBP2 with
the minor groove that induces a DNA bend of 
20° around the last
base of the consensus binding-site TGTGGTT (46). This is
consistent with an inherent structural role for CBF/PEBP2 and provides
another possible mechanism for the ability of CBF/PEBP2 to potentiate
c-Myb binding.
CBF/PEBP2 might also promote binding of other factors in vivo by
recruiting specific chromatin modifying activities, such as histone
acetyl transferases and ATP-dependent remodeling complexes, that
themselves may facilitate factor binding to nucleosomal DNA
(47, 48, 49, 50, 51, 52). In this regard, it has been demonstrated that
CBF/PEBP2 recruits the histone acetyl transferases p300 and CBP
(53). Moreover, an intact E CBF/PEBP2 site is required
to promote histone H3 acetylation across the TCR minilocus in
transgenic mice (3).
CBF/PEBP2 not only stimulates the binding of c-Myb, but also stimulates the appearance of hypersensitive guanines that flank the c-Myb and upstream GATA-3 sites. A perturbed DNA conformation at the edges of the c-Myb and GATA-3 binding sites could depend on CBF/PEBP2-induced DNA bending, on direct interactions between CBF/PEBP2 and GATA-3, or on coactivator-mediated interactions among CBF/PEBP2, GATA-3, and c-Myb. The unique enhanceosome structure defined by these hypersensitive guanines might in turn play a direct role in stabilizing the binding of c-Myb.
Our data suggest that transcription requires c-Myb but not CBF/PEBP2, and occurs normally on those alleles that bind c-Myb in the absence of CBF/PEBP2. The fact that the CBF/PEBP2 site mutation has no apparent effect on transcription even though it results in diminished c-Myb binding probably reflects an effect on c-Myb binding that is less dramatic than variations in transcriptional activity due to copy number and integration site heterogeneity.
The fact that the c-Myb site but not the CBF/PEBP2 site mutation
inhibits transcription in stably transfected Jurkat contrasts with our
results in transgenic mice in which both mutations have dramatic
effects on V(D)J recombination within the same set of constructs. This
might suggest different requirements for transcription and V(D)J
recombination, but the results must be interpreted cautiously as they
may reflect differences in the milieus of transcription factors,
coactivators, and chromatin modifying activities in Jurkat cells and DN
thymocytes. Finally, we note an apparent discrepancy between our
transcription data in transiently (9, 21) vs stably
transfected Jurkat cells. CBF/PEBP2 binding seems important for
transcription in the former, but not the latter. The distinct chromatin
states in transient and stable transfectants might contribute to this
difference (54, 55). Alternatively, transcription may be
influenced by other cis-acting elements with which E
interacts and that differ between the two test constructs (i.e.,
promoters, MARs). Additional experiments in which the chromatin
structural environments or associated cis-acting elements
are systematically varied will be required to address this issue in
greater detail.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Cristina Hernández-Munain, Edificio de Biologicas, Lab C-210, Centro de Biología Molecular, Universidad Autónoma de Madrid, Cantoblanco, 28007 Madrid, Spain. E-mail address: chmunain{at}cbm.uam.es
3 Abbreviations used in this paper: DN, CD4-CD8- thymocytes; DP, CD4+CD8+ thymocytes; CBF, core binding factor; DMS, dimethylsulfate; E, TCR enhancer; E, TCR enhancer; HS, hypersensitive site; MAR, matrix-attachment region; PEBP2, polyoma enhancer-binding protein 2; R x A, recombination-activating gene 2-/- x A.
Received for publication April 11, 2002. Accepted for publication August 5, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ locus. Immunol. Rev. 165:131.[Medline]
enhancer in and T cells. Immunity 7:505.[Medline]
enhancer to TCR enhancer function during thymocyte maturation. Immunity 10:723.[Medline]
locus accessibility and expression by the TCR enhancer. Immunity 10:503.[Medline]
enhancer by functional cooperation between c-Myb and core-binding factors. Mol. Cell Biol. 14:473. gene rearrangement by c-Myb. J. Exp. Med. 183:289. gene rearrangement by CBF/PEBP2. J. Exp. Med. 185:1193. enhancer. Mol. Cell Biol. 15:3090.[Abstract]
complementary DNA clones. Science 238:678.-enhancer chromatin in vivo. Mol. Cell Biol. 18:3223. DNA binding protein is a member of the human GATA family. EMBO J. 10:1809.[Medline]
gene enhancer. Mol. Cell Biol. 11:2778. locus. Science 247:1225. enhancer. Mol. Cell Biol. 11:5671. transgene. J. Exp. Med. 179:43. enhancer to promote V(D)J recombination. Proc. Natl. Acad. Sci. USA 96:11970. enhancer complex is dependent on LEF-1-induced DNA bending and multiple protein-protein interactions. Genes Dev. 9:995.2) synergistically activate the macrophage colony-stimulating factor receptor promoter. Mol. Cell Biol. 16:1231.[Abstract]
synergize with different regions of AML1. Mol. Cell Biol. 18:3915. promoter. J. Biol. Chem. 274:31577. and - gene expression. J. Immunol. 150:3905.[Abstract]
gene enhancer. J. Immunol. 154:5195.[Abstract]
enhanceosome. Mol. Cell 1:119.[Medline]
enhancer assembly and activation on chromatin templates in vitro. Genes Dev. 11:887.. Cell 104:755.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Tani-ichi, M. Satake, and K. Ikuta Activation of the mouse TCR{gamma} enhancers by STAT5 Int. Immunol., September 1, 2009; 21(9): 1079 - 1088. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. del Blanco, J. L. Roberts, N. Zamarreno, N. Balmelle-Devaux, and C. Hernandez-Munain Flexible Stereospecific Interactions and Composition within Nucleoprotein Complexes Assembled on the TCR{alpha} Gene Enhancer J. Immunol., August 1, 2009; 183(3): 1871 - 1883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Chayka, J. Kintscher, D. Braas, and K.-H. Klempnauer v-Myb Mediates Cooperation of a Cell-Specific Enhancer with the mim-1 Promoter Mol. Cell. Biol., January 1, 2005; 25(1): 499 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. V. Johnson, A. G. Bert, G. R. Ryan, A. Condina, and P. N. Cockerill Granulocyte-Macrophage Colony-Stimulating Factor Enhancer Activation Requires Cooperation between NFAT and AP-1 Elements and Is Associated with Extensive Nucleosome Reorganization Mol. Cell. Biol., September 15, 2004; 24(18): 7914 - 7930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Arman, J. Calvo, M. E. Trojanowska, P. N. Cockerill, M. Santana, M. Lopez-Cabrera, J. Vives, and F. Lozano Transcriptional Regulation of Human CD5: Important Role of Ets Transcription Factors in CD5 Expression in T Cells J. Immunol., June 15, 2004; 172(12): 7519 - 7529. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
,
Exons; 
, protein binding sites.
