| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Division of Molecular Bioscience, John Curtin School of Medical Research, Australian National University, Canberra, Australian Capital Territory, Australia;
School of Molecular Bioscience, Washington State University, Pullman, WA 99164; and
School of Biochemistry and Molecular Genetics, University of New South Wales, Sydney, New South Wales, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,
, and 
subunits (reviewed in Ref. 1). Together they
form a strong autocrine loop for promoting T cell growth and clonal
expansion, and possibly contributing to B cell differentiation,
macrophage, and NK cell activation (2). High levels of
IL-2 are produced when T cells are activated through both the
Ag-specific TCR interacting with peptide:MHC complexes on APCs and
costimulatory interactions such as B7 on APCs with CD28 receptors on T
cells (3). IL-2 is necessary for the generation of correct
immune responses, as indicated by the fact that IL-2-deficient mice
develop a thymic disorder that disrupts thymocyte development
(4). Other conditions such as hemolytic anemia, chronic
inflammatory bowel disease, and severe autoimmunity are also
prevalent in IL-2-deficient mice leading to the conclusion that IL-2
plays a critical role in T cell homeostasis (5).
Transcription from the IL-2 gene is highly inducible following Ag
stimulation of T cells and is further enhanced by cross-linking
costimulatory CD28 receptors with an anti-CD28 Ab
(6, 7, 8). The regulatory regions of the human IL-2 promoter
have been studied in detail and it was initially established that
300 nt upstream from the start of transcription was sufficient for
maximal activity in transient transfection assays in Jurkat T cells
(9, 10). Within this minimal promoter region, two major
TCR responsive regions were identified in the human gene and termed
Ag-responsive elements
(ARRE),4 ARRE-1 and
ARRE-2 (Ref. 11 ; see Fig. 1
). The binding of AP-1 and
NF-AT as well as Oct transcription factors controls both ARRE-1 and
ARRE-2 (12, 13, 14, 15, 16, 17). Another important site within the IL-2
minimal promoter is a region that responds to CD28 costimulation, known
as the CD28 response region (RR) (8, 18, 19) (see Fig. 1).
This region contains a CD28 response element (RE) that is a variant
NF-
B site and binds RelA homodimers and c-Rel-containing complexes
as well as NFAT proteins (8, 18, 19, 20, 21, 22, 23). The CD28RR also
contains an AP-1 transcription factor-binding site that functions
cooperatively with the CD28RE (24). Although controversy
exists concerning the AP-1 proteins that operate on this site,
ATF-2/CREB bZIP proteins and c-fos and c-jun AP-1
family members have been shown to bind (24, 25, 26).
|
B, NF-AT, and AP-1 interactions at this
region (20, 28). For example, HMGA1 selectively promotes
binding of both c-Rel and AP-1-containing complexes to the CD28RE
whereas the binding of NF-AT is inhibited by HMGA1 (23).
These findings have led to the suggestion that HMGA1 plays an important
regulatory role in the formation of transcription factor complexes on
the CD28RR (28). The association of these transcription factors with the IL-2 promoter occurs in the context of chromatin within the nucleus. Alterations in the chromatin structure of the promoter/enhancer regions of inducible genes are thought to play an important role in regulating their transcription (29, 30, 31). A number of mechanisms govern chromatin remodeling events including transient modification and subsequent destabilization of the nucleosomal structure by histone acetylases, nucleosome sliding, dissociation of some core histones, or stable disruption leading to nucleosome-free regions (32, 33, 34, 35, 36). DNase I digestion, restriction enzyme accessibility assays, and in vivo genomic footprinting analyses have been used to probe the chromatin structure of the promoter region of the IL-2 gene before and after its transcriptional activation in human and mouse T lymphocytes. These experiments revealed that in resting mouse and human T cells, when the IL-2 gene is silent, the proximal promoter region is inaccessible to DNase I and restriction enzyme digestion. Following simulation, new DNase I hypersensitive sites appear suggesting changes in chromatin structure (37, 38, 39, 40, 41).
In this study, we report that the IL-2 proximal promoter has the intrinsic capacity to assemble a precisely positioned nucleosome in vitro and present evidence that a similarly positioned nucleosome exists in vivo. This nucleosome prevents transcription factor binding in vitro and appears to be transiently remodeled following T cell activation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant hexahistidine-tagged full-length Fos and Jun proteins, as well as truncated Fos 116–211 and Jun 224–334 proteins, were prepared and purified by affinity chromatography using a Ni-NTA-agarose column (Qiagen, Valencia, CA) as previously described (42). Fos and Jun proteins were co-renatured in vitro into the heterodimer transcription factor AP-1 by step-wise dialysis from 6 M urea with the final buffer containing 25 mM sodium phosphate, pH 7.6, 5% glycerol, and 5 mM DTT (34). Full-length recombinant wild-type human HMGA1 proteins (i.e., HMGA1a and HMGA1b, formerly called HMG-I and HMG-Y, respectively; Ref. 27), and the HMGI-MII, MIII mutant protein, were produced as described elsewhere (43, 44). NF-ATp and c-Rel were prepared as previously described (20, 23). The purity of each recombinant preparation was assessed by SDS-PAGE (44).
Cell culture and stimulation
Human Jurkat T leukemia cells were maintained in RPMI 1640 medium supplemented with 10% FCS and the antibiotics, gentamicin and penicillin. Approximately 2.5 x 107 Jurkat T cells were stimulated using 20 ng/ml of PMA and 1 µM calcium ionophore. Bristol-Myers Squibb (Seattle, WA) supplied the activating CD28 mAb (WA93165) used at a 1/10,000-fold dilution.
Chromatin reconstitution and characterization of nucleosome rotational settings
Histones were reconstituted onto radiolabeled IL-2 promoter fragments by high-salt to low-salt step-wise exchange with 30- to 60-fold excess of either isolated chicken erythrocyte nucleosome core particles or H1-depleted oligonucleosomes as previously described (41, 45). The in vitro-reconstituted nucleosomes were routinely purified from both contaminating free DNA and unbound proteins by loading onto a pre-equilibrated 5–20% 10 ml sucrose gradient in buffer containing 1 M Tris, pH 7.5, 500 mM EDTA, and 100 mM PMSF and centrifuged at 4°C for 16 h at 35,000 rpm. Nucleosome-containing fractions were determined by PAGE, pooled, and then dialysed in a buffer of 10 mM Tris pH 7.5, 1 mM EDTA, and 1 mM 2-ME for 16 h. Following dialysis, the nucleosome preparation was concentrated to a volume of 400 µl using Amicon-30 MW columns (Beverly, MA).
Determination of the boundaries of translationally positioned nucleosomes
The approximate 5' and 3' boundaries of nucleosomes that are
translationally positioned on a PCR-generated 220-bp fragment of IL-2
promoter DNA reconstituted into chromatin in vitro (see Fig. 2F) were determined by previously described methods
(46). Briefly, the reconstituted chromatin was digested
with micrococcal nuclease (MNase; Boehringer Mannheim,
Indianapolis, IN) to release monomer nucleosome core particles
(containing 146–150 bp of DNA). The quality of the released core
particles was routinely monitored by analyzing their electrophoretic
mobility on native nucleoprotein gels, by SDS-PAGE analysis of their
histone protein content and by analysis of the sizes of the DNA
fragments by electrophoretic separations on 6% denaturing
polyacrylamide sequencing gels. Monomer nucleosome-length DNA was
purified from the core histone proteins and enzymatically
32P-labeled on both its 5' ends using T4
polynucleotide kinase (New England Biolabs, Beverly, MA) and the
resulting fragment digested with restriction endonucleases (e.g.,
MseI, HinfI, SspI; see Fig. 2F) to determine whether their recognition sites were
present within the fragments. The sizes of the released radiolabeled
DNA subfragments indicate the distance from cut site to the 5' and 3'
borders of the core particle (see Fig. 2F).
|
Purified recombinant AP-1, c-Rel, preexisting NF-AT (NF-ATp), and HMGA1 were incubated for 5 min at room temperature (RT) in a buffer with final concentrations of 10 mM Tris, pH 7.5, 10 mM MgCl2, 5 mM EDTA, 10 mM DTT, 0.2% Nonidet P-40, 1% glycerol, 0.4% sucrose, 0.5 mg/ml BSA, and 100 ng of poly(dG:dC) in 20 µl. Either radiolabeled nucleosome-assembled IL-2 (0.1 ng) or free DNA (0.1 ng) was used with indicated amounts of proteins for each experiment and incubated at RT for 20 min, before the reactions were separated on 5% 1x Tris-borate EDTA polyacrylamide gels. Gels were dried and exposed to x-ray film or phosphorimager intensifying screens. Digital images were produced on a Fuji LAS 3000 phosphorimager and analyzed with ImageGauge software (Fuji, Tokyo, Japan).
DNase I and hydroxyl radical footprinting
Promoter fragments were 5'-end radiolabeled by T4 polynucleotide
kinase and [-32P]ATP. Footprinting of HMGA1
recombinant proteins on promoter DNA fragments radiolabeled on only one
5' end using the nuclease DNase I followed detailed published protocols
(43, 47). For each protein and DNA substrate, optimal
conditions for footprinting were empirically determined. ssDNA cleavage
products were then separated by electrophoresis on 6% sequencing gels
with Maxam-Gilbert "G+A" chemical cleavage products of
control DNA fragments serving as a reference standard. Band intensities
were quantified using a Fuji phosporimager and ImageGauge
software (Fuji). Hydroxyl radical cleavage of both free IL-2 promoter
DNA and in vitro reconstituted chromatin substrates followed published
procedures (46, 48).
Analysis of chromatin by ligation-mediated (LM)-PCR
Cells (2.5 x 107) were washed in PBS
and resuspended at a final concentration of 5 x
106 cells/ml containing 0.5 to 2 mM bleomycin
sulfate (BLM) (Blenoxane; Bristol-Myers Squibb). After a 30-min
incubation at 37°C, the suspension was centrifuged to pellet the
cells, the supernatant was removed, and the cell pellet was resuspended
in 200 µl of PBS. Genomic DNA was prepared using the Qiagen Genomic
DNA purification columns. LM-PCR was performed as previously described
(49, 50) using IL-2-specific oligonucleotides (Table I). Following LM-PCR, 4 µl of each
sample was combined with 4 µl of sequencing loading dye and
electrophoresed on 6% (w/v) polyacrylamide sequencing gels. Gels were
vacuum-dried on DEAE-3 chromatography paper, exposed to phosphorimager
screens (Fuji), and scanned using a Fuji phosphorimager. Image analysis
was performed using ImageGauge software.
|
MNase and restriction enzyme accessibility assays were performed as recently described (51). Briefly, Jurkat T cell nuclei were resuspended in either MNase digestion buffer (10 mM Tris, pH 7.5, 15 mM NaCl, 60 mM KCl, 1 mM CaCl2) or restriction enzyme buffer (1x buffer supplied by New England Biolabs (1 µl of 2 mg/ml leupeptin, 1 µl of 1.8 mg/ml aprotinin, and 1 µl of 100 mM PMSF)) and incubated with the appropriate endonuclease. Reactions were stopped with the addition of EDTA to 20 mM final concentration. A control without the accessibility agent was included from both the stimulated and unstimulated cells to monitor for endonuclease activity. Isolated genomic DNA (50 ng) was used to perform SYBR green real-time PCR as recently described (51). The amplicon length was kept between 100 and 200 bp. Identical thermocycler conditions were used for all primer sets: stage 1, 50°C/2 min for 1 cycle; stage 2, 95°C/10 min for 1 cycle; stage 3, 95°C/15 s, 60°C/1 min for 40 cycles. To correlate the Ct values (threshold values) from the IL-2 CHART-PCR amplification plots to percent accessibility, a standard curve was generated using serial dilutions of genomic DNA and the appropriate primer sets.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine whether the establishment of a precise chromatin
architecture, consisting of positioned nucleosomes, plays a role in the
inducible expression of the human IL-2 gene, 5' end-labeled, IL-2
promoter DNA fragments (Fig. 1
) were
reconstituted into nucleosomes in vitro (52). A 180-bp
fragment corresponding to the region of the IL-2 promoter from nt -55
to -235 (see Fig. 1) formed a distinct shifted complex upon addition
of increasing amounts of chicken donor chromatin (Fig. 2A,
lanes 2–4) suggesting that this portion of the IL-2
promoter can assemble into a discrete mononucleosome. A longer IL-2
promoter fragment from nt -320 to +40 (360 bp; Fig. 1) also assembled
into one major band (Fig. 2A) suggesting discreet nucleosome
positions. In contrast to the IL-2 promoter, a 240-bp fragment (-200
to +40) from the GM-CSF promoter gave rise to several shifted complexes
following nucleosome assembly (Fig. 2B, lanes
2–4). The fine structure of the in vitro nucleosome reconstitutes
was analyzed by a variety of high-resolution mapping techniques to
determine whether the in vitro assembled core particles had assumed
distinct translational and/or rotational positions with respect to the
underlying DNA sequences. DNase I cleavage of the 180-bp (nt -55 to
-235) IL-2 mononucleosome complex revealed a pronounced periodic
pattern of hypersensitive bands distinct from that of naked DNA (Fig. 2C, compare lanes 2–5 with 6–9)
suggesting that a rotationally positioned nucleosome is assembled on
IL-2 promoter DNA. In contrast, digestion of the nucleosome-assembled
GM-CSF gene promoter did not yield a periodic DNase I digestion pattern
(Fig. 2D, compare lanes 8–15 with
1–6) suggesting that this rotational positioning seen for
IL-2 was not a general property of all DNA fragments.
To further define the rotational position of the nucleosome(s) on the
IL-2 promoter, hydroxyl radical cleavage was performed. Fig. 2E shows the results of hydroxyl radical cleavage analysis
of a 220-bp fragment of the IL-2 promoter between nt -1 to -220 (Fig. 1) reconstituted into a mononucleosomal particle. Whereas a nearly
uniform pattern of cleavages is observed for the naked DNA fragment
(Fig. 2E, lane 5), the DNA in the reconstituted
nucleosome exhibits a pronounced pattern of bands spaced at 10-bp
intervals (indicated by dots adjacent to lane 6 in Fig. 2E). This pattern of bands ends at approximately nt 61 where
it reverts to the pattern for naked DNA (Fig. 2E,
lane 6). Consistent with the DNase I cleavage
analysis, this periodic pattern of hydroxyl radical cleavage indicates
that a rotationally positioned nucleosome is present on IL-2 promoter
DNA and that one end of the core particle may be located at
approximately nt 61. Interestingly, this nucleosome may function as a
boundary nucleosome because using the longer 360-bp fragment of the
IL-2 promoter we have found that additional closely packed nucleosomes
are positioned upstream of the one located at -61 (data not
shown).
To further investigate the stability of the rotationally positioned
nucleosome ending at approximately nt 60, restriction enzyme mapping
experiments were performed to determine whether it also exhibited a
strong translational setting on the promoter DNA. Digestion of
MNase-released nucleosomal DNA with MseI releases a distinct
106/107-bp doublet of labeled fragments (Fig. 2F, lane
3) whereas digestion with HinfI releases a 91/92-bp
doublet (Fig. 2F, lane 3). In the lower part of
this gel are additional distinct doublet cleavage bands of 42/43 bp for
the MseI digest and 57/58 bp for the HinfI digest
(data not shown). The sizes of the restriction fragments released by
these two enzymes indicate that the positioned nucleosome adopts a
single highly preferred translational setting on the proximal IL-2
promoter between nt -60/-61 and -208/-209 (Fig. 2F). In
contrast, the SspI enzyme does not cleave the MNase-released
nucleosomal DNA (Fig. 2F, lane 5) because its cut
site at nt -52 lies outside of the positioned core particle, thus
confirming the results of hydroxyl radical cleavage. Because the
prominent DNA fragments released by both restriction enzymes are
doublet bands that differ by only a single base pair in size, we
attribute these size differences to variations in the extent of MNase
trimming of the original chromatin reconstitute.
These data clearly indicate that the IL-2 promoter DNA has the
intrinsic ability to assemble a very stable nucleosome core particle
that is both rotationally and translationally positioned between nt
-60 and -210 encompassing promoter regions that are critical for
inducible expression of the IL-2 gene (Fig. 2F).
The positioned IL-2 nucleosome blocks transcription factor, but not HMGA1, binding to the CD28RR in vitro
The CD28RR of the IL-2 promoter lies close to the dyad of the
positioned nucleosome and has been shown to bind the transcription
factors, c-Rel, RelA, AP-1 (c-fos/c-jun), and
NF-ATp, as well as the architectural protein HMGA1 (20, 21, 23, 25, 28). To determine whether the presence of the
nucleosome affected the ability of these transcription factors to gain
access to their binding sites, increasing amounts of recombinant
transcription factors were added alone or in combination to IL-2
nucleosomes or naked DNA and the resultant complexes were analyzed by
EMSA. The addition of increasing amounts of recombinant full-length
c-fos/c-jun heterodimers (AP-1) (Fig. 3A, lanes 9–14),
rNF-ATp (Fig. 3A, lanes 17 and
18), or recombinant c-Rel (Fig. 3B, lanes
6–10) to the naked 180-bp fragment of the IL-2 promoter (nt -55
to -235) lead to the formation of specific DNA:protein complexes as
would be expected from previous studies (20, 23, 28).
Neither AP-1 (Fig. 3A, lanes 3–8), NFATp (Fig. 3A, lanes 15 and 16), nor c-Rel (Fig. 3B, lanes 1–5) was capable of binding to the
nucleosome-assembled DNA, even with the addition of up to 10 times the
amount of protein required to form specific complexes on naked DNA. The
complexes detected in these binding reactions are due to the presence
of some free DNA in the nucleosome preparations (Fig. 3, A
and B). The reduction in intensity of the nucleosome complex
at the highest concentration of AP-1 is most likely due to nonspecific
interactions as is also evident in the free DNA lanes (Fig. 3A, lanes 3 and 9). Combinations of
these transcription factors were tested for their ability to bind to
nucleosome-assembled DNA but no combinations were found that could
successfully bind (data not shown). DNase I footprinting experiments
confirmed the inability of the transcription factors either alone or in
combination to bind to nucleosome-assembled DNA (data not shown).
|
A, lanes 1–6).
The specificity of HMGA1 binding was tested by using the non-DNA
binding HMGA1a mutant, HMGI-mII,III, which has mutations in the BD II
and III domains that are critical for specific DNA interactions
(26). As expected, HMGI-mII,III did not bind to the IL-2
nucleosome when used at the same concentrations as HMGA1a and HMGA1b
(Fig. 4A, lanes 7–9).
|
B, lanes 3–6) as previously
described (28). HMGA1a footprints are also detected at the
NF-B site and CD28RE when the opposite strand is analyzed (Fig. 4C, lanes 2–5). When HMGA1a was titrated onto
IL-2 nucleosomal DNA, footprints were observed specifically at the
ARRE-1, NF-IL2B regions, CD28RR and the NF-B site (Fig. 4, B, lanes 8–11, and C, lanes
9–11, refer to lines). At these regions, the presence of HMGA1a
also resulted in the reappearance of specific DNase I hypersensitive
bands (Fig. 4, B, lane 8, and C,
lane 9, refer to dots) seen in naked DNA and may represent
local disruption of the nucleosome:DNA contacts at the specific regions
where HMGA1a binds. However, HMGA1 proteins did not appear to alter the
rotational setting of the nucleosome nor did they promote the binding
of transcription factors such as c-Rel or AP-1 to the
nucleosome-assembled IL-2 promoter (data not shown). In summary, all of
the transcription factors tested, with the exception of HMGA1, were
occluded from binding to the nucleosome-assembled IL-2 promoter. HMGA1
bound to the nucleosome-assembled DNA and lead to local but not
complete disruption of nucleosome:DNA contacts. Evidence for a positioned nucleosome at the IL-2 proximal promoter in vivo
To address whether the IL-2 gene promoter in its native
environment in T cell nuclei assembles a positioned nucleosome similar
to the in vitro-assembled IL-2 nucleosome, we examined nucleosome
organization in vivo using two independent, but complementary, methods:
LM-PCR (41) and CHART-PCR (51). The analysis
of nucleosome positioning was first performed using LM-PCR and
the chemical nuclease, BLM (49). BLM preferentially
cleaves linker DNA between nucleosome core particles (49, 50) and has sequence specificity in order of decreasing
preference for GC, GT, GA, and AT base pairs (53). Human
Jurkat T cells were treated with increasing concentrations of BLM,
genomic DNA was isolated, and was then analyzed by LM-PCR using IL-2
promoter-specific primer sets (Fig. 1, set A: 11, 12, 13 and set B: 56,
57, 58).
Densitometric comparisons of cellular (Fig. 5A, lanes 4–7) and
naked DNA (Fig. 5A, lanes 2 and 3)
damage with bleomycin using a coding strand-specific primer "set A"
extending from nt -330 downstream (Fig. 1) indicated that the region
of DNA between -190 and +10 was less digested in the cellular DNA than
in the naked DNA (Fig. 5A, lanes 2 and
6). Bleomycin-specific bands from cellular DNA occurring at
-260 and -270 bp were also significantly reduced in intensity
relative to naked DNA (Fig. 5A), indicating that there may
be a nucleosome downstream of -190 bp and also at -260 to -270
bp. However, the bleomycin hypersensitive bands at -190 and
-235 bp were equally represented in both cellular and naked DNA
samples indicating an apparent lack of protection in these regions. In
addition, clusters of hypersensitive bands of approximately equal
intensity occurred from +10 bp toward the coding region of the IL-2
gene in both naked and genomic DNA indicating a possible lack of
nucleosomes in this region (Fig. 5A, lanes 2 and
6).
|
). This set of primers shows
protected bands from -10 extending upstream from the transcription
start site (Fig. 5B, compare lanes 3 and
4 with 6–8). Again, the region downstream from
the transcription start site was not protected in cellular DNA (Fig. 5B).
The ability of the Hinf1 and Dra1 enzymes
as well as MNase to digest the IL-2 proximal promoter in Jurkat T cell
nuclei was next examined using the CHART-PCR assay (51).
In this assay, the amount of digestion is monitored by real-time PCR
using primer sets that amplify 100 bp and the amount of PCR product
is inversely proportional to the amount of digestion. Real-time
PCR-using primers encompassing the distal promoter region (set A, nt
-595 to -465), the region of the CD28RR (set B, nt -220 to -120)
and downstream of the transcription start site (set C, nt +16 to +110)
(see Fig. 1), were used to perform the CHART-PCR assay across the IL-2
gene. DraI digests the IL-2 promoter at nt -167 close to
the CD28RR whereas HinfI cleaves at nt -151 between the
CD28RE and AP-1 sites within the fragment amplified by primer set B
(Fig. 1). Purified genomic DNA used as a control is almost completely
digested (95%) within 60 min of exposure to either Hinf1 or
Dra1 (Fig. 6A). The
IL-2 promoter in nuclear DNA was highly protected from digestion with
only 15% digestion observed in nuclei from unstimulated Jurkat T
cells compared with undigested DNA (Fig. 6A). This is not
simply an artifact of digestion within the nucleus as the amount of
digestion increases with stimulation (see next section and Fig. 7). In addition, the in vitro-assembled nucleosome also protects
the promoter from restriction enzyme digestion (data not shown). Primer
set A was used to monitor nonspecific accessibility because the region
amplified by this primer set did not contain a Dra1 or
Hinf1 site. Only 3–6% of the DNA was measured as
accessible using this primer set (Fig. 6A) implying that
only a small amount of nonspecific endonuclease digestion of the DNA is
occurring in the experiments.
|
|
B). Taken together, these bleomycin, restriction enzyme, and MNase protection data provide evidence for the presence of a nucleosome(s) across the proximal promoter of the IL-2 gene in unstimulated Jurkat T cells. However, regions outside the proximal promoter may not be as strongly protected by nucleosomes.
Chromatin across the IL-2 proximal promoter is remodeled upon T cell activation
Because the IL-2 gene is a highly inducible gene, becoming rapidly
expressed upon T cell activation, we next asked whether alterations
occur in chromatin structure across the proximal promoter upon
activation of the cell. Chromatin remodeling of the IL-2 proximal
promoter was examined in vivo using restriction enzyme and MNase
digestion with the CHART-PCR assay (51). In unstimulated
Jurkat T cells, we observed a low level of accessibility to
DraI and HinfI (15%) when compared with
undigested DNA from unstimulated cells as described above (Fig. 6A). Increases in DraI accessibility were first
observed in cells stimulated for 1 h and continued to increase
with a maximum reached at 3 h of stimulation (Fig. 7, A and B, 50%
in PMA/I-treated cells and 70% in cells treated with PMA/I +
-CD28). By 4 and 8 h poststimulation, the amount of
DraI accessibility decreased to 25% relative to
unstimulated cells. As a control, a region of the promoter lacking a
DraI site (set A) was analyzed in unstimulated and
stimulated cells over the time period and as anticipated, there was no
change in accessibility in all samples (Fig. 6A; data not
shown). The time course of HinfI accessibility followed that
of DraI with the exception that accessibility was observed
as early as 30 min following activation (Fig. 7, A and
B). When the time course of chromatin accessibility with
Dra1 or Hinf1 was plotted with the time course of
IL-2 mRNA production measured by real-time PCR, it is clear that
chromatin accessibility reaches a maximum 1 h before the maximum
accumulation of IL-2 mRNA (Fig. 7, C and D).
These data suggest that chromatin remodeling is a prerequisite for
maximal expression of the IL-2 gene.
Changes in MNase accessibility were also measured across the IL-2 gene
with the CHART-PCR assay and the three sets of primers shown in Fig. 1.
PMA/I activation resulted in only a small change in MNase accessibility
across the primer set B region (Fig. 7E) but PMA/I + CD28
activation led to almost a 2-fold increase in accessibility across this
region (Fig. 7E). The regions of the gene assayed by primer
sets A (-595 to -465) and C (+15 to +110) showed little change in
MNase accessibility following activation (Fig. 7E).
These results suggest that chromatin remodeling is occurring across the proximal promoter region of the human IL-2 gene following activation. The region of remodeling corresponds to the region that can assemble a positioned nucleosome in vitro and where a nucleosome is also likely to exist in vivo.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
65%) compared with other inducible cytokine
genes (for example, the GM-CSF promoter has 48% A/T content).
Interestingly, there is a notable 10–20 bp periodicity of A/T
stretches throughout much of the IL-2 minimal promoter region, which
could potentially place this sequence in the nucleosome positioning
class of DNA fragments containing "TATA-tetrads"
(57).
The fact that the in vitro-assembled nucleosome inhibited the binding
of CD28RR-specific transcription factors to their DNA binding sites
supports a role for this nucleosome in transcriptional repression of
the IL-2 gene in resting Jurkat T cells. Interestingly, HMGA1 proteins
were found to specifically bind to the ARRE-1, NF-IL2B, CD28RR, and
NF-B sites in the presence of the nucleosome causing local
disruption of core particle DNA interactions but not complete
nucleosome disruption. However, the binding of HMGA1 did not alter
accessibility of restriction enzyme recognition sequences occluded by
the positioned nucleosome nor did it promote transcription factor
binding to these core particles in vitro (data not shown). Therefore,
these data suggest that additional processes, besides HMGA1 binding to
the IL-2 nucleosome, are needed for the subsequent chromatin remodeling
which takes place when cells are activated.
Interestingly, the chromatin structure of the promoter of the human
IL-2R gene shares many similarities to that of the IL-2 promoter
reported in this study. In vitro IL-2R promoter DNA assembles a
stably positioned nucleosome and there is evidence for nucleosome
remodeling of this region of the promoter during transcriptional
activation of the gene (62). In addition, HMGA1 has
recently been shown to bind directionally to the IL-2R promoter when
it is assembled into nucleosomes (62). Reeves et al.
(62) proposed that HMGA1 may form part of a targeting
signal to recruit chromatin remodeling complexes to the IL-2R
promoter. It is intriguing that two genes whose products are critical
for the correct activation of T cells can assemble highly positioned
nucleosomes across their promoter regions which bind HMGA1, and
presumably function to prevent inadvertent transcription from these
genes unless the correct activation signals are present.
Chromatin accessibility data derived from in vivo genomic footprinting
by LM-PCR as well as restriction enzyme and MNase accessibility by
CHART-PCR provides evidence for the presence of a nucleosome on the
IL-2 promoter in resting Jurkat T cells. We have recently examined the
chromatin structure of the mouse IL-2 promoter in EL-4 T cells and
CD4+ primary T cells and find a complete lack of
MNase or RE accessibility across the proximal promoter in these cells
before stimulation (51), again supporting the idea that
the IL-2 promoter is assembled into nucleosomes. Bleomycin-LM-PCR
footprinting data suggested that a nucleosome plus perhaps additional
promoter binding factors may be located in the region from -15 to
-190 bp. The location of the in vitro-positioned nucleosome (-60 to
-210) lies approximately within the bleomycin-footprinted region
suggesting that the bleomycin footprint may in part represent a
positioned nucleosome. It has previously been shown that a constitutive
DNase I footprint occurs across the TATA box region of the IL-2
promoter in Jurkat T cells and this has been interpreted as
representing a paused polymerase complex (40, 41).
Such a complex together with a nucleosome could explain the size of the
observed bleomycin footprint.
In this study, we show, using the CHART-PCR assay with restriction enzyme or MNase accessibility, that the chromatin structure of the proximal promoter region, but not other regions of the IL-2 gene, is remodeled following T cell activation. Time course experiments of the accessibility changes within the IL-2 proximal promoter showed a gradual rise in RE accessibility from 30 min to 3 h following activation. This gradual change in accessibility is in contrast to our experiments with the mouse IL-2 gene where accessibility is not observed until 1.5 h following activation in either EL-4 or primary CD4+ T cells from mouse spleen and at this time point almost complete accessibility is observed (51). It is possible that the nucleosomes across the proximal promoter region in Jurkat T cells are already modified in some manner. This possibility is consistent with the observation that the mouse gene appears to be completely inaccessible to MNase or RE digestion in resting cells (51), while the human gene, at least in Jurkat T cells, has a higher level of intrinsic accessibility especially to MNase digestion. The distinct time course of chromatin accessibility change is reflected in distinct time courses of IL-2 mRNA synthesis for Jurkat and EL-4 cells. However, in both cases, the time of chromatin remodeling precedes the time of mRNA accumulation implying that chromatin remodeling may be a prerequisite for high levels of gene transcription.
There also appears to be some differences in the signals required for chromatin remodeling across the IL-2 promoter in Jurkat T cells and EL-4 cells. In EL-4 cells, P/I leads to almost complete accessibility with a small contribution from CD28 for either restriction enzyme accessibility or MNase accessibility (51). In addition, PMA alone is almost as effective as P/I (51). However, in Jurkat cells while P/I increases restriction enzyme accessibility, it has little effect on MNase accessibility and CD28 signaling is required to generate MNase accessibility. It is unclear why these differences exist but may relate to the higher level of MNase accessibility seen in unstimulated Jurkat T cells compared with EL-4s as discussed below.
The higher constitutive accessibility of the human gene in Jurkat T cells compared with the mouse gene in EL-4 and primary T cells and the faster kinetics of chromatin remodeling and mRNA accumulation may relate to the recent finding that the D3 phosphoinositide phosphatase and tensin homolog is deficient in Jurkat T cells (63). As a consequence, there are elevated levels of the phosphoinositol 3-kinase products, PI-3,4-P2 and PI-3,4,5-P3, in the plasma membrane leading to constitutive recruitment of pleckstrin homology domain proteins such as Itk (63). Thus, constitutive signaling through this pathway could contribute to the more accessible chromatin structure on the IL-2 gene and the faster response to activation. Indeed, it has been shown that phospholipids are involved in the targeting of the BAF chromatin remodeling complex to chromatin following T cell activation (64).
In summary, we suggest that chromatin is a necessary component of IL-2 gene regulation. A highly positioned nucleosome most likely exists across the proximal promoter region of the IL-2 gene in vivo preventing transcription factor binding in the absence of a stimulus that triggers chromatin remodeling. Although HMGA1 can bind to its recognition sites in the presence of a nucleosome it is not sufficient to promote transcription factor binding or to lead to complete disruption of the nucleosome. An understanding of the signals and molecules required for disruption of this nucleosome in vivo will help to further elucidate the role of chromatin in IL-2 gene transcription.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: School of Biochemistry, LaTrobe University, Melbourne, Victoria, Australia.
3 Address correspondence and reprint requests to Dr. M. Frances Shannon, Division of Molecular Bioscience, John Curtin School of Medical Research, Australian National University, Canberra, Australian Capital Territory, 2601, Australia. E-mail address: frances.shannon{at}anu.edu.au
4 Abbreviations used in this paper: ARRE, Ag-responsive element; RR, response region; RE, response element; MNase, micrococcal nuclease; RT, room temperature; LM, ligation-mediated; BLM, bleomycin sulfate; CHART-PCR, chromatin accessibility using real-time PCR; NF-ATp, preexisting NF-AT.
Received for publication April 25, 2002. Accepted for publication June 20, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B-like response element. J. Biol. Chem. 266:14179.-globin gene cluster. Biochemistry 35:8753.[Medline]
gene. Mol. Cell. Biol. 20:4666.This article has been cited by other articles:
|
|
 |
|
  O. Desy, D. Carignan, M. Caruso, and P. O. de Campos-Lima Immunosuppressive Effect of Isopropanol: Down-Regulation of Cytokine Production Results from the Alteration of Discrete Transcriptional Pathways in Activated Lymphocytes J. Immunol., August 15, 2008; 181(4): 2348 - 2355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. McKarns and R. H. Schwartz Biphasic Regulation of Il2 Transcription in CD4+ T Cells: Roles for TNF-{alpha} Receptor Signaling and Chromatin Structure J. Immunol., July 15, 2008; 181(2): 1272 - 1281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. H. Brettingham-Moore, O. R. Sprod, X. Chen, P. Oakford, M. F. Shannon, and A. F. Holloway Determinants of a transcriptionally competent environment at the GM-CSF promoter Nucleic Acids Res., May 1, 2008; 36(8): 2639 - 2653. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Thomas, N. Chunder, C. Chen, S. E. Umetsu, S. Winandy, and A. D. Wells Ikaros Enforces the Costimulatory Requirement for IL2 Gene Expression and Is Required for Anergy Induction in CD4+ T Lymphocytes J. Immunol., December 1, 2007; 179(11): 7305 - 7315. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Bostik, E. S. Noble, S. T. Stephenson, F. Villinger, and A. A. Ansari CD4+ T Cells from Simian Immunodeficiency Virus Disease-Resistant Sooty Mangabeys Produce More IL-2 Than Cells from Disease-Susceptible Species: Involvement of p300 and CREB at the Proximal IL-2 Promoter in IL-2 Up-Regulation J. Immunol., June 15, 2007; 178(12): 7720 - 7729. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Webster, Y. Rodriguez, W. T. Klimecki, and D. Vercelli The Human IL-13 Locus in Neonatal CD4+ T Cells Is Refractory to the Acquisition of a Repressive Chromatin Architecture J. Biol. Chem., January 5, 2007; 282(1): 700 - 709. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Castellano, B. Vire, M. Pion, V. Quivy, D. Olive, I. Hirsch, C. Van Lint, and Y. Collette Active Transcription of the Human FASL/CD95L/TNFSF6 Promoter Region in T Lymphocytes Involves Chromatin Remodeling: ROLE OF DNA METHYLATION AND PROTEIN ACETYLATION SUGGEST DISTINCT MECHANISMS OF TRANSCRIPTIONAL REPRESSION J. Biol. Chem., May 26, 2006; 281(21): 14719 - 14728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Hwang, J.-H. Hong, and L. H. Glimcher IL-2 production in developing Th1 cells is regulated by heterodimerization of RelA and T-bet and requires T-bet serine residue 508 J. Exp. Med., November 7, 2005; 202(9): 1289 - 1300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Adachi and E. V. Rothenberg Cell-type-specific epigenetic marking of the IL2 gene at a distal cis-regulatory region in competent, nontranscribing T-cells Nucleic Acids Res., June 3, 2005; 33(10): 3200 - 3210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Chen, J. Wang, D. Woltring, S. Gerondakis, and M. F. Shannon Histone Dynamics on the Interleukin-2 Gene in Response to T-Cell Activation Mol. Cell. Biol., April 15, 2005; 25(8): 3209 - 3219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Thomas, L. Gao, and A. D. Wells Signals from CD28 Induce Stable Epigenetic Modification of the IL-2 Promoter J. Immunol., April 15, 2005; 174(8): 4639 - 4646. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. C. McKarns, R. H. Schwartz, and N. E. Kaminski Smad3 Is Essential for TGF-{beta}1 to Suppress IL-2 Production and TCR-Induced Proliferation, but Not IL-2-Induced Proliferation J. Immunol., April 1, 2004; 172(7): 4275 - 4284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Smith, I. Collins, G. V. R. Chandramouli, W. G. Butscher, E. Zaitseva, W. J. Freebern, C. M. Haggerty, V. Doseeva, and K. Gardner Targeting Combinatorial Transcriptional Complex Assembly at Specific Modules within the Interleukin-2 Promoter by the Immunosuppressant SB203580 J. Biol. Chem., October 17, 2003; 278(42): 41034 - 41046. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Rao, S. Gerondakis, D. Woltring, and M. F. Shannon c-Rel Is Required for Chromatin Remodeling Across the IL-2 Gene Promoter J. Immunol., April 1, 2003; 170(7): 3724 - 3731. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. F. Holloway, S. Rao, X. Chen, and M. F. Shannon Changes in Chromatin Accessibility Across the GM-CSF Promoter upon T Cell Activation Are Dependent on Nuclear Factor {kappa}B Proteins J. Exp. Med., February 17, 2003; 197(4): 413 - 423. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
