Full text of DOI:10.1016/S1074-7613(00)80215-0

Immunity, Vol. 12, 643–652, June, 2000, Copyright  2000 by Cell Press
Cell-Type-Restricted Binding
of the Transcription Factor NFAT
to a Distal IL-4 Enhancer In Vivo
†
†
Suneet Agarwal, Orly Avni, and Anjana Rao*
Department of Pathology
and is strongly potentiated by IL-4 and the IL-4-induced
transcription factor STAT6 (reviewed in Abbas et al.,
1996; O’Garra, 1998). In contrast, the second step of
acute IL-4 gene transcription is much less dependent
on STAT6 and IL-4 (Lederer et al., 1996; Huang et al.,
1997). The early stages of Th2 differentiation are accom-
panied by long-range changes in the DNase I hypersen-
sitivity pattern and DNA methylation status of the IL-4/
IL-5/ IL-13 gene cluster (Agarwal and Rao, 1998; Bird
et al., 1998; Takemoto et al., 1998). They are also marked
by selective up- or downregulation of transcription fac-
tors and cell surface receptors characteristic of the Th2
lineage (Ho et al., 1996; Szabo et al., 1997a; Zheng and
Flavell, 1997; Sallusto et al., 1998).
The antigen-induced transcription factor NFAT and
its cooperating transcription factor AP-1 (Fos/Jun) are
implicated in the second phase of induced IL-4 gene
transcription by differentiated Th2 cells (reviewed in Rao
et al., 1997). NFAT is a family of proteins that includes
the four calcium-regulated transcription factors NFAT1
Harvard Medical School and
The Center for Blood Research
00 Longwood Avenue
Boston, Massachusetts 02115
2
Summary
By DNase I hypersensitivity analysis, we have identi-
fied an inducible, cyclosporin A–sensitive enhancer
located 3؅ of the interleukin-4 (IL-4) gene. The en-
hancer binds the Th2-specific transcription factor
GATA3 in vivo but is not perceptibly influenced by the
absence of a second Th2-specific factor, cMaf. The
antigen-inducible transcription factor NFAT1 binds the
IL-4 enhancer and the IL-4 promoter only in stimulated
Th2 cells; conversely, NFAT1 binds to the interferon
(
IFN)-␥ promoter only in stimulated Th1 cells. Our re-
(
(
1
NFATp, NFATc2), NFAT2 (NFATc, NFATc1), NFAT3
NFATc4), and NFAT4 (NFATc3) (reviewed in Rao et al.,
997; Crabtree, 1999; Kiani et al., 2000); a fifth NFAT
protein, NFAT5/ TonEBP, is regulated by osmotic shock
Lopez-Rodriguez et al., 1999; Miyakawa et al., 1999).
Activation of the four calcium-regulated NFAT proteins
hereafter abbreviated NFAT) requires the calcium/cal-
sults support a model whereby transcription factors
such as NFAT1, which are nonselectively induced in
antigen-stimulated T cells, gain access to cytokine
regulatory regions only in the appropriate subset of
differentiated T cells in vivo. This restricted access
enables antigen-dependent and subset-specific tran-
scription of cytokine genes.
(
(
modulin-dependent phosphatase calcineurin and is in-
hibited by the immunosuppressive agents cyclosporin A
Introduction
(CsA) and FK506. In transgenic mice, reporter activity
driven by multiple copies of a composite NFAT-AP-1
site was higher in Th2 than in Th1 cells (Rincon and
Flavell, 1997b; Wenner et al., 1997), possibly because
Th2 cells express several-fold higher levels of the AP-1
family member JunB (Li et al., 1999). It is generally as-
sumed, however, that because NFAT itself is nonselec-
tively induced in both Th1 and Th2 cells (reviewed in
Szabo et al., 1997b; Kuo and Leiden, 1999), it cannot
explain the subset specificity of cytokine gene expres-
sion. Rather, transcription factors that are selectively
expressed in Th1 or Th2 lineage cells are likely to be
involved.
The transcription factors cMaf (Maf) and GATA3 show
a striking level of Th2 restriction. Both Maf and GATA3
are expressed at low levels in naive T cells and become
upregulated during Th2 differentiation; both are consti-
tutively expressed in differentiated Th2 cells but not in
differentiated Th1 cells; and both are targets of the IL-
The immunomodulatory cytokine interleukin-4 (IL-4)
plays a central role in the development of effector B and
T cells and is critical for host defense during parasitic
infections. Dysregulation of IL-4 expression results in
uncontrolled allergic inflammation and aberrant immune
responses to pathogens (reviewed in Abbas et al., 1996;
O’Garra, 1998). IL-4 expression is, in general, mutually
exclusive with the expression of interferon (IFN)-␥, a
cytokine with a major influence on the inflammatory re-
sponse. T helper (Th) cells that produce the inflamma-
tory cytokines IFN-␥ and TNF but not IL-4 are classified
in the Th1 subset, while Th cells that produce IL-4 but
not IFN-␥ are classified in the Th2 subset (reviewed in
Abbas et al., 1996; Mosmann and Sad, 1996; O’Garra,
1
998). In addition, Th2 cells show coordinate expression
of the closely linked IL-5 and IL-13 genes.
Transcription of the IL-4 gene by T cells is regulated
in two distinct steps (reviewed in Rincon and Flavell,
4
-dependent transcription factor STAT6 (Ho et al., 1996;
1
997a; Szabo et al., 1997b; Agarwal and Rao, 1998;
Zhang et al., 1997; Zheng and Flavell, 1997; Kurata et
al., 1999). Analysis of Maf-deficient and Maf-transgenic
mice indicates that Maf exerts a selective effect on IL-4
gene expression without directly affecting the expres-
sion of other Th2 cytokines including IL-5, IL-6, IL-10,
and IL-13 (Ho et al., 1998; Kim et al., 1999). In contrast,
GATA3 appears to have a global effect on Th2 differenti-
ation. Differentiated Th1 cells from GATA3-transgenic
mice showed increased expression of mRNAs encoding
IL-4, IL-6, and IL-10 (Zheng and Flavell, 1997); GATA3-
overexpressing B and primary Th1 cell lines showed
Murphy, 1998; O’Garra, 1998; Agarwal et al., 1999). The
first step involves the differentiation of naive T cells into
mature effector Th2 cells, while the second step is the
inducible transcription of the IL-4 gene by the differenti-
ated Th2 cells. Th2 differentiation is triggered when na-
ive T cells initially encounter antigen in the periphery
*
To whom correspondence should be addressed (e-mail: arao@
cbr.med.harvard.edu).
†
These authors contributed equally to this paper.

Immunity
44
6
Figure 1. TCR or Ionomycin Stimulation of Primary Th2 Cells Induces a Novel DNase I HS Site Located 3Ј of the IL-4 Gene
A) Stimulation with anti-CD3. Primary Th2 cells differentiated for 2 weeks were either left unstimulated (panel 1) or stimulated for 6 hr using
(
plate-bound anti-CD3⑀ (panel 2). Cells were subjected to DNase I HS analysis and membranes were hybridized using a probe from the 3Ј end
of the 19 kb BamHI fragment spanning the IL-4 gene (Agarwal and Rao, 1998). The constitutive HS sites IV and V (Agarwal and Rao, 1998)
and the inducible site V
B) Requirement for calcium and calcineurin. DNase I HS analysis of primary Th2 cells differentiated for 1 week and then either left unstimulated
panel 1) or stimulated for 6 hr with PMA and ionomycin (P ϩ I, panel 2), ionomycin alone (panel 3), or PMA, ionomycin, and CsA (P ϩ I ϩ
CsA, panel 4). Membranes were hybridized using the 3Ј probe described in (A). The inducible site V is indicated by an asterisk in panels 2
and 3.
C) Map of the murine IL-4 locus showing the approximate location of the constitutive DNase I HS sites and the location of the inducible HS
site V . The 3Ј probe used in (A) and (B) is indicated. The expanded view shows the 1.2 kb HindIII–SwaI fragment containing the inducible
site V , with the restriction sites used in Figure 3A indicated.
A are indicated. The use of a 5Ј promoter probe did not reveal any other inducible sites.
(
(
A
(
A
A
striking induction of IL-5 (Ranganath et al., 1998; Zhang
et al., 1998); expression of antisense GATA3 in a differen-
tiated Th2 clone led to decreased expression of IL-4,
IL-5, IL-6, and IL-10 (Zheng and Flavell, 1997); expres-
sion of dominant-negative GATA3 in transgenic mice
led to reduced expression of IL-4, IL-5, and IL-13 (Zhang
et al., 1999); and expression of GATA3, even in STAT6-
deficient T cells, resulted in production of Th2 cytokines
and appearance of the differentiated Th2 pattern of
DNase I hypersensitivity on the IL-4 locus (Ouyang et
al., 2000).
impressive than that of the endogenous IL-4 gene. These
results point to the existence of distal enhancer ele-
ments that cooperate with the IL-4 promoter to modulate
IL-4 gene expression in vivo.
Here we have used DNase I hypersensitivity mapping
and transient transfection assays to identify a distal IL-4
enhancer located 3Ј of the IL-4 gene. The enhancer was
discovered as an inducible DNase I hypersensitive (HS)
site appearing in stimulated Th2 cells but not in either
resting or stimulated Th1 cells. T cells lacking Maf con-
tinued to show appearance of both the constitutive and
the inducible pattern of DNase I HS sites, suggesting
that Maf is not critical for chromatin remodeling of the
IL-4 gene by naive T cells or for the activity of the IL-4
enhancer. In contrast, chromatin immunoprecipitation
assays showed that the enhancer binds GATA3 in differ-
entiated Th2 cells in vivo, consistent with the possibility
that GATA3 is required for enhancer function. Both the
IL-4 enhancer and the IL-4 promoter bind the NFAT
family member NFAT1 only in stimulated Th2 cells; con-
versely, the IFN-␥ promoter binds NFAT1 only in stimu-
lated Th1 cells. These results define an antigen-respon-
sive, NFAT-dependent, GATA3 binding distal regulatory
element in the IL-4 gene, and suggest a mechanism by
which nonselectively induced transcription factors such
as NFAT may regulate subset-specific transcription of
cytokine genes.
The regulation of IL-4 gene transcription has been
studied by many laboratories. With a few exceptions
(
1
Henkel et al., 1992; Kubo et al., 1997; Agarwal and Rao,
998; Bird et al., 1998; Takemoto et al., 1998), essentially
all the focus has been on the proximal IL-4 promoter
reviewed in Brown and Hural, 1997; Rincon and Flavell,
997a; Szabo et al., 1997b; Murphy, 1998). Transgenic
(
1
experiments have documented, however, that the IL-4
promoter does not suffice for optimal IL-4 gene tran-
scription in vivo (Wenner et al., 1997). The levels of induc-
ible transcription from an IL-4 promoter-reporter trans-
gene were several orders of magnitude lower than those
observed for the endogenous IL-4 gene. Furthermore,
although the integrated promoter-reporter construct
achieved a considerable degree of Th2 specificity (ف40-
fold), the level of differential expression was much less

Binding of NFAT to a Novel IL-4 Enhancer In Vivo
45
6
Figure 2. Only Remodeled Cytokine Loci Re-
spond to Acute Stimulation with Further
Changes in Chromatin Structure
(A) DNase I HS analysis of the IL-4 locus in
D10 (Th2) and D5 (Th1) cells either left unstim-
ulated (panels 1 and 3) or stimulated for 6 hr
with PMA and ionomycin (P ϩ I, panels 2 and
4
). Membranes were hybridized using the 3Ј
probe indicated in Figure 1. Constitutive HS
sites are indicated, and the inducible HS site
VA
is marked with an asterisk.
(B) DNase I HS analysis of the IFN-␥ locus in
unstimulated and stimulated (P ϩ I) D10 and
D5 cells. Blots from (A) were stripped and
rehybridized using an IFN-␥ exon 4 probe.
Th1-specific constitutive HS sites (Agarwal
and Rao, 1998) are indicated in panel 3.
Results
of the D5 (Ar-5) Th1 clone did not result in induction of
site VA, nor did it cause any other change in DNase I
Identification of Inducible Hypersensitive Site
VA in the IL-4 Gene
To identify promoter-distal regulatory elements associ-
ated with stimulated transcription of the IL-4 gene, we
compared DNase I hypersensitivity patterns on the IL-4
locus of unstimulated and stimulated Th2 cells (Figure
HS patterns throughout the IL-4 locus (Figure 2A, panels
3 and 4). However, D5 cells displayed clear changes in
the chromatin structure of the IFN-␥ locus following PMA
plus ionomycin stimulation, with the most prominent
effect being disappearance of HS site II (Figure 2B, pan-
els 3 and 4). No marked changes were observed in the
DNase I HS pattern of the IFN-␥ locus in stimulated D10
cells (Figure 2B, panels 1 and 2). Therefore, despite
identical stimulation of the D5 Th1 and D10 Th2 cell
clones, the changes in chromatin structure that corre-
lated with active transcription were restricted to those
cytokine loci that had undergone stable chromatin re-
modeling during T cell differentiation.
1
). CD4 T cells, differentiated under polarizing Th2 condi-
tions for 1 week, showed the expected constitutive pat-
tern of DNase I hypersensitivity (Agarwal and Rao, 1998)
(Figure 1A, left panel). These resting cells did not pro-
duce detectable levels of IL-4 transcripts. In contrast,
cells stimulated with plate-bound anti-CD3 produced
abundant levels of IL-4 mRNA and exhibited a novel,
inducible HS site (Figure 1A, right panel). We termed this
The transcription factor Maf, which is expressed at
elevated levels in Th2 cells relative to Th1 cells, has
been implicated in IL-4 gene transcription (Ho et al.,
1996, 1998; Kim et al., 1999). To determine whether Maf
contributes to the changes in chromatin structure on
the IL-4 locus, either during Th2 differentiation or during
acute stimulation of differentiated Th2 cells, we incu-
bated CD4 T cells from normal and Maf-deficient mice
with IL-4 and anti-IL-12 (Th2 differentiation conditions).
Following this period of differentiation, we compared
DNase I HS patterns in resting and PMA plus ionomycin
stimulated cells (Figure 3). DNase I hypersensitivity map-
ping with a 5Ј probe encompassing the IL-4 promoter
revealed no differences in the resting DNase I HS pattern
of normal versus Maf-deficient T cells (Figure 3A), indi-
cating that Maf is not required for the long-range
changes in chromatin structure that occur in the IL-4
genetic locus during Th2 differentiation. Likewise, the
inducible HS site “V
the Th2-specific constitutive HS site V (“five”) and its
dependence on activation. Site V was also observed
in primary Th2 cells and Th2 clones stimulated either
with phorbol ester (PMA) plus calcium ionophore (iono-
mycin) or with ionomycin alone (Figure 1B, panels 2 and
A” (“five A”) based on its proximity to
A
3
), and its appearance was abrogated by the immuno-
suppressive agent cyclosporin A (CsA) (Figure 1B, panel
4
5
). Site V
A mapped ف12 kb 3Ј of the IL-4 promoter and
kb 3Ј of the end of the IL-4 coding region (Figure 1C).
Appearance of Inducible HS Site V
but Does Not Require the Transcription Factor cMaf
The formation of site V was strictly Th2 restricted in
A Is Th2 Restricted
A
vivo (Figure 2). Stimulation of the D10 Th2 clone with
PMA plus ionomycin resulted in strong induction of site
VA (Figure 2A, panels 1 and 2). In contrast, stimulation

Immunity
46
6
Figure 3. The Th2-Restricted Transcription Factor Maf Is Not Required for Formation of Constitutive or Inducible DNase I HS Sites in the IL-4
Locus
CD4 T cells were purified from the lymph nodes and spleens of wild-type or Maf-deficient mice and differentiated under strongly polarizing
Th2 conditions for one week.
(
A) DNase I HS analysis was performed on resting cultures of wild-type and MafϪ/Ϫ Th2 cells. Genomic DNA was digested to completion with
BamHI and subjected to Southern analysis using a probe corresponding to the 5Ј end of the IL-4 locus.
B) Wild-type and MafϪ Th2 cells were either left unstimulated (panels 1 and 2) or stimulated for 6 hr with 20 nM PMA and 2 ␮M ionomycin
/
Ϫ
(
(panels 3 and 4). Cells were harvested and subjected to DNase I HS analysis. Membranes were hybridized using a probe corresponding to
the 3Ј end of the IL-4 locus (as in Figure 1).
use of a probe from the 3Ј end of the IL-4 locus revealed
no differences in DNase I HS patterns in either resting
or stimulated cells (Figure 3B), indicating that Maf also
does not play a role in the stimulation-dependent ap-
driven by a minimal SV40 promoter (Figure 5A; pGL3-
438). A shorter (300 bp) PstI–SwaI fragment retained
essentially all the enhancer activity of the 438 bp frag-
ment (Figure 5A; pGL3-V 300). When tested with the
VA
A
pearance of the inducible HS site V
A
. These results are
IL-4 promoter, the fragment conferred 60-fold enhance-
ment of induced transcription in stimulated D10 cells
compared to 30-fold enhancement with the IL-4 pro-
moter alone (Figure 5B). The lower effect of the enhancer
in combination with the IL-4 promoter reflects the fact
that the IL-4 promoter itself is highly active in stimulated
cells. The enhancer conferred no increase in basal tran-
scription from either the SV40 promoter or the IL-4 pro-
moter in unstimulated D10 T cells, indicating that stimu-
lation was absolutely required for enhancer activity
(Figures 5A and 5B).
consistent with the finding that Maf regulates expression
of only the IL-4 gene in the IL-4/IL-5/IL-13 gene cluster
by virtue of its ability to bind a site in the IL-4 promoter
(Ho et al., 1996, 1998; Kim et al., 1999).
Hypersensitive Site V Exhibits Inducible,
A
CsA-Sensitive Enhancer Activity
in Th2 Cells
To determine whether the inducible HS site VA plays a
functional role in IL-4 gene transcription, we localized
it to a short restriction fragment and tested its function
in transient transfection assays (Figure 4). First, we used
Subset-Restricted Binding of NFAT to Cytokine
Regulatory Regions In Vivo
HS analysis to map site VA to a 438 bp EcoRI–SwaI
fragment, hereafter termed the core IL-4 enhancer (Fig-
ure 4A). The restriction map of this region is shown in
Figure 1C and its sequence in Figure 4B. Restriction
enzyme accessibility experiments confirmed that the
SwaI site was localized to the vicinity of the inducible
hypersensitive site (Figure 4C). When SwaI was diffused
into intact nuclei, its ability to cleave the IL-4 locus in-
creased after stimulation of D10 but not D5 cells (Figure
The requirements for appearance of site V
A (Figure 1)
and the properties of the site V enhancer (Figure 5)
A
paralleled the requirements for nuclear translocation
and activation of NFAT (Rao et al., 1997). Inspection
of the nucleotide sequence of the enhancer fragment
(Figure 4B) revealed four potential NFAT binding sites
(T/AGGAA; Rao et al., 1997), which bound NFAT when
tested in electrophoretic mobility shift assays (EMSA)
in vitro. Binding was nonselective, since it was observed
with nuclear extracts from stimulated D5 as well as D10
T cells (data not shown). These results confirm previous
work demonstrating equivalent induction of NFAT in nu-
clear extracts from stimulated Th1 and Th2 cells (Rooney
et al., 1994; Li-Weber et al., 1997; Rincon and Flavell,
1997b).
4
C, compare lanes 2 and 4 with 6 and 8), indicating that
the “accessibility” of this genomic region to both DNase
I and SwaI increased in a Th2-specific manner upon
stimulation.
DNA fragments containing HS site VA behaved as in-
ducible, CsA-sensitive enhancer elements in transient
transfection assays in D10 cells (Figure 5). A single copy
of a 438 bp EcoRI–SwaI fragment spanning the core
enhancer conferred both PMA/ionomycin responsive-
ness and CsA sensitivity on a luciferase reporter plasmid
To reconcile the nonselective binding of NFAT to site
VA
in vitro with the Th2-specific appearance of site V
A in
vivo (Figure 2), we used chromatin immunoprecipitation

Binding of NFAT to a Novel IL-4 Enhancer In Vivo
47
6
Figure 4. Localization and Sequence of Inducible HS Site V
A
(
A) Mapping of DNase I HS site V to a 438 bp fragment EcoRI–SwaI fragment 3Ј of the IL-4 gene. Lanes 1–5, D10 nuclei were subjected to
A
DNase I titration, and the genomic DNA was digested to completion with HindIII and analyzed on a Southern blot. Lanes 6–10, markers were
provided by D10 genomic DNA digested to completion with Hind III and the indicated restriction enzymes. Membranes were hybridized with
a probe derived from the 5Ј end of a 5.5 kb HindIII fragment containing both sites V and VA (see Figure 1C). The top half of this autoradiogram
and the lanes containing the markers were exposed for a shorter time than the bottom half, necessitated by the fact that the intensity of the
parent band and of the markers was much greater than that of the hypersensitive band.
(B) Restriction enzyme accessibility analysis of the SwaI site in resting and stimulated D5 and D10 cells. Cells were left unstimulated (Ϫ) or
were stimulated (ϩ) for 6 hr with PMA and ionomycin (P ϩ I). Nuclei were prepared and left untreated (Ϫ) or treated (ϩ) for 1 hr with SwaI,
following which DNA was purified and digested to completion with HindIII. Southern blots were hybridized as described in (A). The relative
intensities of the 1.2 kb band in lanes 2, 4, 6, and 8 (expressed as a percentage of the total intensity in the lane) were 2.5% , 3.0% , 7.7% , and
4
(
0.6% , respectively.
C) Nucleotide sequence of the 438 bp EcoRI–SwaI fragment encompassing the IL-4 enhancer. Relevant restriction enzyme sites are underlined
and labeled. Sequences that match consensus binding motifs (on either DNA strand) for NFAT (A/TGGAA) (Rao et al., 1997), GATA (A/TGATAA/G)
Ko and Engel, 1993), and STAT [TTC(X)3–4GAA] (Schindler et al., 1995) transcription factors are shown in bold.
(
(
ChIP) assays (Figure 6). The ChIP technique can estab-
(“chromatin”) were immunoprecipitated from stimulated
D10 (Th2) but not D5 (Th1) cells (Figure 6A, compare
lanes 3 and 4). Conversely, NFAT1 bound to the endoge-
nous IFN-␥ promoter, which contains a putative NFAT
site (Rao et al., 1997), only in stimulated D5 cells (Figure
6B, compare lanes 3 and 4). Control experiments
showed that the IL-4 promoter and enhancer sequences
were present at equivalent concentrations in template
DNA from D5 and D10 cells (Figure 6C). ChIP analysis
using primary differentiated Th1 and Th2 cells confirmed
that NFAT1 binding to the IL-4 enhancer was dependent
on stimulation, as expected from the fact that NFAT1 is
nuclear only in stimulated cells (Rao et al., 1997) (Figure
6D). We conclude that the IL-4 promoter and enhancer
are selectively accessible to NFAT1 binding in Th2 cells,
while the IFN-␥ promoter is selectively accessible to
NFAT1 binding in Th1 cells.
lish whether a known transcription factor truly binds in
the vicinity of a known regulatory element in living cells
(Parekh and Maniatis, 1999). Three of the four calcium-
regulated NFAT proteins, NFAT1, NFAT2 and NFAT4,
are expressed in T cells (Timmerman et al., 1996; Lyakh
et al., 1997); of these, we chose to examine NFAT1 since
it is the most abundant NFAT protein in T cells, constitut-
ing 80% –90% of total NFAT (Xanthoudakis et al., 1996).
Moreover, NFAT1 can potently transactivate the IL-4
promoter, especially in combination with Maf (Ho et al.,
1
996; Hodge et al., 1996a). ChIP assays using anti-
NFAT1 antibodies showed that NFAT1 bound to both
the IL-4 promoter and IL-4 enhancer regions in an induc-
ible and Th2-specific manner in vivo: that is, PCR prod-
ucts were generated with the IL-4 promoter and en-
hancer primers only when the DNA–protein complexes

Immunity
648
Figure 5. The Inducible HS Site VA Contains
an Inducible, CsA-Sensitive Transcriptional
Enhancer
(A) SV40 minimal promoter: D10 cells were
transfected with an SV40 promoter-luciferase
plasmid (pGL3) or the same plasmid con-
taining either the 438 bp EcoRI–SwaI frag-
ment (pGL3-V
fragment (pGL3-V
A
438) or the 300 bp PstI–SwaI
300). Twenty-four hours
A
after transfection, cells were left unstimulated
or stimulated for 8 hr with PMA and ionomycin
(P ϩ I) or PMA, ionomycin, and CsA (P ϩ I ϩ
CsA). Results are represented as fold activa-
tion relative to the luciferase activity of pGL3
in unstimulated cells (set at 1; range 30,000–
1
to be visible.
B) IL-4 promoter: D10 cells were transfected with an IL-4 promoter-luciferase construct (pIL4-Luc) or the same construct containing the 438
bp EcoRI–SwaI fragment containing HS site V (pIL4-Luc VA). Twenty-four hours after transfection, cells were left unstimulated or stimulated
00,000 relative luciferase units). Bars indicate standard deviations of three independent experiments, which in several cases are too small
(
A
for 8 hr with PMA and ionomycin (P ϩ I) or PMA, ionomycin, and CsA (P ϩ I ϩ CsA). Results are represented as fold activation relative to the
luciferase activity of pIL-4-Luc in unstimulated cells (set at 1; range 150,000–200,000 relative luciferase units). Bars indicate standard deviations
of three independent experiments, which in several cases are too small to be visible.
The Th2-Restricted Transcription Factor GATA3
Binds to the IL-4 Enhancer In Vivo
The core IL-4 enhancer also contained several putative
binding sites for the Th2-restricted transcription factor
GATA3, identified by visual inspection as related to the
consensus sequence A/TGATAA/G (Ko and Engel, 1993)
regions are implicated. Second, the inducible, CsA-sen-
sitive appearance of site V in primary Th2 cells mirrors
the features of endogenous IL-4 transcription, as well
as the properties of the site V enhancer in transient
transfection assays. Third, the site V enhancer is the
only inducible HS site in the immediate vicinity of the
IL-4 gene. The locus is bounded at its 5Ј and 3Ј ends by
the IL-13 gene and the CNS-specific gene KIF3 (Frazer et
al., 1997). The DNase I HS pattern across this 31 kb
A
A
A
(Figure 4B). To determine whether GATA3 indeed bound
to the IL-4 enhancer region in vivo, we again used ChIP
assays. The results showed that GATA3 binds in vivo
to the IL-4 enhancer region in D10 Th2 cells but not in
D5 Th1 cells that do not express GATA3 (Figure 7A,
compare lanes 3 and 4). GATA3 was not observed to
bind to the IL-4 promoter in Th2 cells (Figure 7B, lane
interval has been mapped, and HS site VA is the only
HS site that is completely restricted to activated Th2
cells (Takemoto et al., 1998; this report; S. A., unpub-
lished data).
Our results are consistent with a large body of other
evidence, suggesting that the Th2-specific transcription
factor Maf is involved in regulating the inducible phase
of IL-4 gene transcription rather than in broadly imple-
menting a Th2 genetic program (Ho et al., 1996, 1998;
Kim et al., 1999). Maf-deficient T cells show a relative
defect in IL-4 but not IL-5 or IL-13 production (Kim et al.,
4
), despite the presence of several reasonably good
GATA binding sites (Ranganath et al., 1998; Zhang et al.,
998). In control experiments, the IL-4 promoter region
1
could be detected with the same primers after immuno-
precipitation with anti-NFAT1 (Figure 7C, lane 4). Pre-
sumably, GATA3 binds with higher affinity to the putative
sites in the IL-4 enhancer, relative to the sites present
in the IL-4 promoter (Ko and Engel, 1993). However,
because of the intrinsic limitations of the ChIP technique
1
999); conversely, Th2 cells from Maf-transgenic mice
show a tendency to overexpress IL-4 but not the other
Th2 cytokines (Ho et al., 1998). Most importantly, Maf
overexpression in transgenic mice does not result in
IL-4 production by differentiated Th1 cells, indicating
that Maf does not force a Th2 program when ectopically
expressed (Ho et al., 1998). In complete accordance
with these findings, we have shown that Maf-deficient
T cells are as competent as wild-type T cells in acquiring
the differentiated Th2 pattern of DNase I hypersensitivity
during the first stimulation of primary T cells with anti-
gen. Thus, despite the fact that Maf is a direct or indirect
target of STAT6 and GATA3 (Kurata et al., 1999; Ouyang
et al., 2000), it is unlikely to play a major role in regulating
the acquisition of transcriptional competence on the
IL-4 locus during Th2 differentiation. Differentiated Maf-
deficient Th2 cells also showed no impairment, relative
to wild-type Th2 cells, in the appearance of the inducible
(the immunoprecipitated chromatin contains DNA frag-
ments of average size ف1 kb), we cannot definitively
identify the specific GATA sites to which GATA3 is
bound, although they are clearly located in the vicinity
of the IL-4 enhancer region and are likely to include
those identified as residing within the core IL-4 enhancer
in Figure 4C.
Discussion
In this work, we have defined an inducible DNase I HS
site, site V
ments support the hypothesis that site V
a distal IL-4 enhancer in vivo, cooperating with the IL-4
promoter to enable optimal, antigen-dependent and
Th2-restricted expression of IL-4. First, the IL-4 pro-
moter is clearly insufficient to support high-level and
fully Th2-specific transcription of a linked reporter gene
in vivo (Wenner et al., 1997), and thus distal regulatory
A, located 3Ј of the IL-4 gene. Several argu-
A
functions as
DNase I HS site VA upon stimulation. This result indicates
that Maf does not play a critical role in the function of
the IL-4 enhancer. Rather, the effect of Maf may be
localized to its known binding site in the proximal pro-
moter (Ho et al., 1996; Cron et al., 1999).

Binding of NFAT to a Novel IL-4 Enhancer In Vivo
49
6
Figure 7. GATA3 Binds In Vivo to the IL-4 Enhancer but Not the
Promoter
In vivo binding of GATA3 to the IL-4 promoter and the IL-4 enhancer
in stimulated D5 (Th1) and D10 (Th2) cells was assessed by ChIP.
(A) DNA–protein complexes were immunoprecipitated without anti-
bodies (“No Ab”) or with anti-GATA3 antibodies. The PCR primers
were specific for the IL-4 enhancer (239 bp).
(
B) The same DNA purified in (A) was amplified using primers specific
for the IL-4 promoter.
C) As a control, the IL-4 promoter region could be detected with
(
the same primers as in (B) after immunoprecipitation of DNA–protein
complexes with anti-NFAT1 antibodies.
lines and in mice results in impaired Th2 cytokine pro-
duction (Zheng and Flavell, 1997; Ranganath et al., 1998;
Zhang et al., 1998, 1999; Ouyang et al., 2000). While it
is clear that GATA3 acts directly on the IL-5 promoter
Figure 6. Specific In Vivo Binding of NFAT to the IL-4 Promoter and
Enhancer in Th2 Cells, and the IFN-␥ Promoter in Th1 Cells
(A) In vivo binding of NFAT1 to the IL-4 promoter and the IL-4
(Zhang et al., 1997; Lee et al., 1998), the mechanism
enhancer in stimulated D5 (Th1) and D10 (Th2) cells was assessed
by chromatin immunoprecipitation (ChIP). DNA–protein complexes
were immunoprecipitated without antibodies (“No Ab”) or with a
cocktail of antibodies specific for NFAT1. PCR primers specific for
through which it promotes IL-4 gene expression has
been controversial (Zheng and Flavell, 1997; Ranganath
et al., 1998; Zhang et al., 1998, 1999). Our results show-
the IL-4 enhancer (“site V
of 239 bp and 182 bp, respectively.
B) The same DNA purified in (A) was amplified using primers specific
for the IFN-␥ promoter, which amplify a product of 250 bp.
C) As a control for parts (A) and (B), to show that DNA–protein
complexes from D5 Th1 and D10 Th2 cells contained equivalent
levels of IL-4 promoter/enhancer target sequences, PCR was per-
formed directly on dilutions of DNA from complexes not subjected
to immunoprecipitation.
A
”) and the IL-4 promoter amplify products
ing that GATA3 binds to the site VA enhancer provide a
site of action for this key transcription factor in regulating
IL-4 expression. In addition, GATA3 may act through
other GATA binding elements dispersed throughout the
IL-4/IL-13 locus, which have been shown to possess
enhancer function in transient reporter assays (Ranga-
nath et al., 1998).
(
(
We have shown that the distal site VA enhancer con-
(
D) ChIP analysis of CD4 T cells differentiated under Th1 or Th2
conditions for 1 week and either left unstimulated (Ϫ) or stimulated
ϩ) for 6 hr with PMA plus ionomycin. DNA was amplified with
primers corresponding to the IL-4 enhancer.
E) As a control for part (D), to show that DNA–protein complexes
tains binding sites for the antigen-induced transcription
factor NFAT and binds the family member NFAT1 in vivo.
The ChIP assay only detects binding and cannot by itself
establish whether a given protein acts as a transcrip-
tional activator or repressor when bound to a specific
genomic region in vivo. Nevertheless, careful analysis
of the phenotype of NFAT1-deficient mice (Kiani et al.,
(
(
from stimulated Th1 and unstimulated or stimulated Th2 cells con-
tained equivalent levels of IL-4 enhancer target sequences, PCR
was performed directly on dilutions of DNA from complexes not
subjected to immunoprecipitation.
1
997) suggests that NFAT1 activates transcription when
bound to the IL-4 promoter and enhancer sites. T cells
from NFAT1-deficient mice show a mild degree of Th2
skewing and hyperproduce Th2 cytokines (Hodge et al.,
In contrast to Maf, GATA3 appears to play a major
role in specifying the Th2 phenotype (Zheng and Flavell,
1
996b; Xanthoudakis et al., 1996), a phenotype that is
1
997; Zhang et al., 1999; Ouyang et al., 2000). GATA3
greatly exaggerated in mice lacking both NFAT1 and
NFAT4 (Ranger et al., 1998a); in contrast, T cells lacking
NFAT2 show a relative impairment of IL-4 production
(Ranger et al., 1998b; Yoshida et al., 1998). These obser-
vations have led to the hypothesis that NFAT2 is a posi-
tive regulator of IL-4 gene expression, while NFAT1 and
NFAT4 act solely as negative regulators. This interpreta-
tion cannot be strictly accurate: NFAT1-deficient T cells
is essential for T cell development (Hattori et al., 1996;
Ting et al., 1996); thus, analyses of GATA3-deficient T
cells await the development of a conditional GATA3
knockout mouse. Ectopic overexpression of GATA3 in
cell lines and in mice results in increased expression of
a spectrum of Th2 cytokines; conversely, overexpres-
sion of antisense and dominant-negative GATA3 in cell

Immunity
50
6
show no difference in peak levels of IL-4 transcripts
relative to wild-type T cells but rather a prolongation of
the late phase of IL-4 transcription (Kiani et al., 1997).
If NFAT1 acted solely as a transcriptional repressor, a
A prediction from our data is that distal enhancer
elements may be involved in regulating the expression of
many if not all NFAT-dependent genes. This is especially
likely in the case of the IFN-␥ gene and other genes
that might possess similarly weak and poorly cell-type-
restricted proximal promoter regions (Penix et al., 1993,
1996; Campbell et al., 1996; Young, 1996; Aune et al.,
1997; Sweetser et al., 1998). The array of NFAT sites in
the IL-4 enhancer is reminiscent of the multiple NFAT
sites present in the IL-4 and IL-2 promoters and the IL-3
and GM-CSF enhancers (reviewed in Rao et al., 1997).
The apparent redundancy of NFAT binding to these cy-
tokine promoter and enhancer regions resembles the
situation at the ␤-globin locus, where related transcrip-
tional activators bind at different stages of erythroid
development to recurring sequence motifs in the locus
control region and in the globin gene promoters (Orkin,
5
- to 10-fold increase in peak levels would be expected,
since NFAT1 accounts for 80% –90% of total NFAT and
is present in wild-type cells at 5- to 10-fold higher levels
than in NFAT2 (Xanthoudakis et al., 1996). Thus, it is
likely that NFAT1, NFAT2, and NFAT4 all activate the
early phase of IL-4 gene transcription, but only NFAT1
and NFAT4 are able to induce negative feedback mecha-
nisms that blunt the late phase of IL-4 gene transcription
(Kiani et al., 1997; Miaw et al., 2000) (reviewed in Agarwal
et al., 1999; Kiani et al., 2000).
The most surprising aspect of our studies was that
NFAT1 binding to cytokine regulatory regions was highly
subset-specific in vivo. NFAT proteins are known to be
expressed at equivalent levels in Th1 and Th2 cells and
to bind equivalently in vitro to naked DNA containing
the IL-4 enhancer and the IL-4 promoter (Rooney et al.,
1
995). The use of widely dispersed regulatory regions,
containing redundant binding sites for the same family
of transcription factors, may be a general strategy to
limit the inappropriate expression of cell-type-specific
genes.
1
994; Li-Weber et al., 1997; Rincon and Flavell, 1997b;
O. A., unpublished data). Nevertheless, NFAT1 bound
to the IL-4 promoter/enhancer regions in vivo only in
Th2 cells and to the IFN-␥ promoter region only in differ-
entiated Th1 cells. These results predict that mecha-
nisms operating in the chromatin context restrict the
access of NFAT to cytokine regulatory regions in vivo.
One hypothesis is that although NFAT1 can bind inde-
pendently to the regulatory regions in vitro, it is only
capable of binding them in the context of chromatin if
subset-specific transcription factors such as GATA3 are
also present in the nucleus. Another, not mutually exclu-
sive, hypothesis is that the chromatin structure of the
IL-4 enhancer and promoter is selectively altered during
Th2 differentiation so that these regions become acces-
sible to NFAT1 only in differentiated (and stimulated)
Th2 cells. For instance, transcription factors such as
STAT6 and NFAT, which are acutely induced during the
first antigen plus cytokine stimulation of naive T cells,
might bind transiently to regulatory regions of the IL-4
gene; this binding might then promote stable alterations
in chromatin structure, histone acetylation, or DNA
methylation status of the adjacent regions that persist
in the differentiated Th2 cells after the initial stimulus
has died away. In this context, it is noteworthy that both
Experimental Procedures
Mice
C57BL/6J and CAF1/J mice (Jackson Laboratories, Bar Harbor,
Maine) were maintained in pathogen-free conditions in barrier facili-
ties at the Center for Animal Resources and Comparative Medicine,
Harvard Medical School. c-Maf mice (Kim et al., 1999) were kindly
provided by L. H. Glimcher (Harvard School of Public Health).
–
/–
DNase I Hypersensitivity Analysis
The murine T cell clones D5 (Ar-5; Rao et al., 1984) and D10
(D10.G4.1; Kaye et al., 1983) were maintained as previously de-
scribed (Agarwal and Rao, 1998). CD4 T cells were purified from
C57BL/6J mice and differentiated in the Th1 or Th2 direction as
previously described (Agarwal and Rao, 1998). Isolation and DNase
I digestion of nuclei and purification of genomic DNA were per-
formed as previously described (Agarwal and Rao, 1998; Cockerill,
2
000) using primary Th2 cells differentiated for 2 weeks, which were
either left unstimulated or stimulated for 6 hr with immobilized anti-
CD3⑀ or with 20 nM PMA plus 2 ␮M ionomycin in the presence or
absence of 2 ␮M CsA.
Probe locations are depicted in locus diagrams. The probe from
the 5Ј end of a 19 kb BamHI fragment spanning the IL-4 locus
has been described (Agarwal and Rao, 1998). A similar probe was
generated from the 3Ј end of the same fragment (Figures 1–3; loca-
the IL-4 promoter (Lederer et al., 1996) and the site V
A
enhancer (Figure 4B) contain binding sites for STAT and
NFAT. The functions of NFAT and STAT6 would diverge
in the secondary stimulation of differentiated T cells,
since acute transcription of the IL-4 gene requires anti-
gen stimulation and NFAT but is known not to be as
dependent on IL-4 stimulation and STAT6 (Huang et al.,
tion indicated in Figure 1C). For fine mapping of HS site V
from the 5Ј end of a 5.5 kb HindIII fragment spanning HS sites V
and V was used. The IL-4 probes (5Ј BamHI 19 kb, 3Ј BamHI 19
kb, and 5Ј HS V/V ) were created by digestion of plasmids containing
these probes and gel purification of the fragments. The exon 4 probe
from the IFN-␥ gene (Agarwal and Rao, 1998) was generated by
PCR and gel purified. The template for PCR was a murine IFN-␥
genomic clone (kindly provided by H. A. Young).
A, a probe
A
A
1
997). A precedent for such a model is provided by
studies of the HO endonuclease gene in yeast (Cosma
et al., 1999). In this system, transient binding of the SWI5
transcription factor results in recruitment of the SWI/
SNF chromatin remodeling complex and the SAGA his-
tone acetyltransferase complex to the HO gene; these
enzyme complexes induce chromatin structural changes
that promote the stable binding of a cell-type-specific
transcription factor, SBF. We are currently investigating
whether analogous mechanisms are involved in chroma-
tin remodeling and acquisition of transcriptional compe-
tence by the IL-4 genetic locus.
Restriction Enzyme Accessibility
Accessibility of the IL-4 locus to the restriction enzyme SwaI was
assessed as described (Boyes and Felsenfeld, 1996). Briefly, SwaI
(150 U) was incubated for 1 hr at room temperature with intact nuclei
from unstimulated and stimulated T cells (20 nM PMA plus 2 ␮M
ionomycin for 6 hr). DNA was purified and digested to completion
with HindIII. The probe is the same as that used for the fine mapping
of site V
sites V and V
using the program NIH Image 1.61.
A
(5Ј end of a 5.5 kb HindIII fragment encompassing both
A). Relative hybridization intensities were assessed

Binding of NFAT to a Novel IL-4 Enhancer In Vivo
51
6
Transient Transfection Assays
A single copy of the 438 bp EcoRI–SwaI fragment or the 300 bp
PstI–SwaI fragment spanning HS site VA was cloned 5Ј of the minimal
SV40 promoter in the luciferase reporter plasmid pGL3 (Promega).
D10 cells were electroporated (960 ␮F/270 V) and rested overnight.
Transfected cells were divided and either left unstimulated or stimu-
lated for 8 hr with PMA (20 nM) and ionomycin (2 ␮M) or PMA,
ionomycin, and CsA (2 ␮M). CsA was added 20 min before the
stimuli. Cell lysates were harvested and luciferase activity was as-
sessed using an automated luminometer (Berthold).
Campbell, P., Pimm, J., Ramassar, V., and Halloran, P. (1996). Identi-
fication of a calcium-inducible, cyclosporine sensitive element in
the IFN-gamma promoter that is a potential NFAT binding site.
Transplantation 61, 933–939.
Cockerill, P.N. (2000). Identification of DNase I hypersensitive sites
within nuclei. Methods Mol. Biol. 130, 29–46.
Cosma, M., Tanaka, T., and Nasmyth, K. (1999). Ordered recruitment
of transcription and chromatin remodeling factors to a cell cycle-
and developmentally regulated promoter. Cell 97, 299–311.
Crabtree, G. (1999). Generic signals and specific outcomes: signal-
2
ϩ
ing through Ca , calcineurin, and NF-AT. Cell 96, 611–614.
Chromatin Immunoprecipitation (ChIP) Assays
ChIP analysis was carried out essentially as described (Parekh and
Maniatis, 1999). Unstimulated or stimulated (6 hr, 20 nM PMA plus
Cron, R., Bort, S., Wang, Y., Brunvand, M., and Lewis, D. (1999). T
cell priming enhances IL-4 gene expression by increasing nuclear
factor of activated T cells. J. Immunol. 162, 860–870.
7
2
␮M ionomycin) D5 Th1 cells and D10 Th2 cells (5 to 15 ϫ 10 ) or
CD4 T cells differentiated under Th1 or Th2 conditions for 1 week
Agarwal and Rao, 1998) were cross-linked using formaldehyde.
Frazer, K.A., Ueda, Y., Zhu, Y., Gifford, V.R., Garofalo, M.R., Mohan-
das, N., Martin, C.H., Palazzolo, M.J., Cheng, J., and Rubin, E.M.
(1997). Computational and biological analysis of 680 kb of DNA
sequence from the human 5q31 cytokine gene cluster region. Ge-
nome Res. 7, 495–512.
(
Nuclei were isolated and sonicated, and DNA–protein complexes
were purified by CsCl gradient centrifugation. The chromatin was
immunoprecipitated using a cocktail (5 ␮g) of anti-67.1 and anti-
NFAT1-C antibodies specific for the family member NFAT1 (Wang
et al., 1995) or with 5 ␮g of anti-GATA3 antibodies (Santa Cruz).
Following deproteination and reversal of cross-links, the presence
of selected DNA sequences was assessed by PCR. The primers
Hattori, N., Kawamoto, H., Fujimoto, S., Kuno, K., and Katsura, Y.
(1996). Involvement of transcription factors TCF-1 and GATA-3 in
the initiation of the earliest step of T cell development in the thymus.
J. Exp. Med. 184, 1137–1147.
used were as follows: site V
5
5
GGC-3Ј (182 bp product) or 5Ј-ATCAATAGCTCTGTGCCG-3Ј (240
bp product); IFN-␥ promoter, 5Ј-GCTCTGTGGATGAGAAAT-3Ј and
A
, 5Ј-AGGGCACTTAAACATTGC-3Ј and
Ј-ACGCCTAAGCACAATTCC-3Ј (239 bp product); IL-4 promoter,
Ј-TTGGTCTGATTTCACAGG-3Ј and either 5Ј-AACAATGCAATGCT
Henkel, G., Weiss, D.L., McCoy, R., Deloughery, T., Tara, D., and
Brown, M.A. (1992). A DNase I-hypersensitive site in the second
intron of the murine IL-4 gene defines a mast cell-specific enhancer.
J. Immunol. 149, 3239–3246.
Ho, I.-C., Hodge, M.R., Rooney, J.W., and Glimcher, L.H. (1996). The
proto-oncogene c-maf is responsible for tissue-specific expression
of interleukin-4. Cell 85, 973–983.
5
Ј-AAGATGGTGACAGATAGG-3Ј (250 bp product).
Acknowledgments
Ho, I.-C., Lo, D., and Glimcher, L. (1998). C-maf promotes Th2 and
attenuates Th1 differentiation by both IL-4 dependent and indepen-
dent mechanisms. J. Exp. Med. 188, 1859–1866.
We thank M. J. Pazin, W. C. Forrester, and members of the laboratory
for valuable discussions; B. S. Parekh and T. Maniatis for the ChIP
protocol; H. A. Young for IFN-␥ genomic clones; and J. I. Kim and
Hodge, M., Chun, H., Rengarajan, J., Alt, A., Lieberson, R., and
Glimcher, L. (1996a). NF-AT-driven interleukin-4 transcription po-
tentiated by NIP45. Science 274, 1903–1905.
L. H. Glimcher for cMafϪ mice. We also thank Genetics Institute
/
Ϫ
for their generous gift of recombinant IL-12 and anti-IL-12 antibodies
and the National Cancer Institute–Biological Response Modifiers
Program for anti-IL-4 antibodies. This work was supported by Na-
tional Institutes of Health grants R01 CA42471, R01 AI44432, and
P01 AI35297 (to A. R.). S. A. is a Howard Hughes Medical Institute
Predoctoral Fellow and a Ryan Foundation Fellow. O. A. is a Post-
doctoral Fellow of the Cancer Research Institute. A. R. was a Stohl-
man scholar of the Leukemia Society of America.
Hodge, M., Ranger, A., Charles de la Brousse, F., Hoey, T., Grusby,
M., and Glimcher, L. (1996b). Hyperproliferation and dysregulation
of IL-4 expression in NFATp-deficient mice. Immunity 4, 397–405.
Huang, H., Hu-Li, J., Chen, H., Ben-Sasson, S.Z.B., and Paul, W.E.
(1997). IL-4 and IL-13 production in differentiated T helper type 2
cells is not IL-4 dependent. J. Immunol. 159, 3731–3738.
Kaye, J., Porcelli, S., Tite, J., Jones, B., and Janeway, C.A.J. (1983).
Both a monoclonal antibody and antisera specific for determinants
unique to individual cloned helper T cell lines can substitute for
antigen and antigen-presenting cells in the activation of T cells. J.
Exp. Med. 158, 836–856.
Received February 9, 2000; revised April 27, 2000.
References
Kiani, A., Rao, A., and Jose, A. (2000). Manipulating immune re-
sponses with immunosuppressive agents that target NFAT. Immu-
nity 12, 359–372.
Abbas, A.K., Murphy, K.M., and Sher, A. (1996). Functional diversity
of helper T lymphocytes. Nature 383, 787–793.
Agarwal, S., and Rao, A. (1998). Modulation of chromatin structure
regulates cytokine gene expression during T cell differentiation. Im-
munity 9, 765–775.
Kiani, A., Viola, J., Lichtman, A., and Rao, A. (1997). Down-regulation
of IL-4 gene transcription and control of Th2 cell differentiation by
a mechanism involving NFAT1. Immunity 7, 849–860.
Agarwal, S., Viola, J., and Rao, A. (1999). Chromatin-based regula-
tory mechanisms governing cytokine gene transcription. J. Allergy
Clin. Immunol. 103, 990–999.
Kim, J., Ho, I.-C., Grusby, M., and Glimcher, L. (1999). The transcrip-
tion factor c-Maf controls the production of interleukin-4 but not
other Th2 cytokines. Immunity 10, 745–751.
Aune, T., Penix, L., Rincon, M., and Flavell, R. (1997). Differential
transcription directed by discrete gamma interferon promoter ele-
ments in naive and memory (effector) CD4 T cells and CD8 T cells.
Mol. Cell. Biol. 17, 199–208.
Ko, L., and Engel, J. (1993). DNA-binding specificities of the GATA
transcription factor family. Mol. Cell. Biol. 13, 4011–4016.
Kubo, M., Ransom, J., Webb, D., Hashimoto, Y., Tada, T., and Nakay-
ama, T. (1997). T-cell subset-specific expression of the IL-4 gene is
regulated by a silencer element and STAT6. EMBO J. 16, 4007–4020.
Bird, J.J., Brown, D.R., Mullen, A.C., Moskowitz, N.H., Mahowald,
M.A., Sider, J.R., Gajewski, T.F., Wang, C., and Reiner, S.L. (1998).
Helper T cell differentiation is controlled by the cell cycle. Immunity
Kuo, C.T., and Leiden, J.M. (1999). Transcriptional regulation of T
lymphocyte development and function. Annu. Rev. Immunol. 17,
149–187.
9
, 229–237.
Boyes, J., and Felsenfeld, G. (1996). Tissue-specific factors addi-
tively increase the probability of the all-or-none formation of a hyper-
sensitive site. EMBO J. 15, 2496–2507.
Kurata, H., Lee, H., O’Garra, A., and Arai, N. (1999). Ectopic expres-
sion of activated Stat6 induces the expression of Th2-specific cyto-
kines and transcription factors in developing Th1 cells. Immunity
1
1, 677–688.
Brown, M.A., and Hural, J. (1997). Functions of IL-4 and control of
its expression. Crit. Rev. Immunol. 17, 1–32.
Lederer, J.A., Perez, V.L., DesRoches, L., Kim, S.M., Abbas, A.K.,

Immunity
52
6
and Lichtman, A.H. (1996). Cytokine transcriptional events during
helper T cell subset differentiation. J. Exp. Med. 184, 397–406.
is highly inducible in effector CD4ϩ T helper 2 (Th2) cells but not
in Th1 cells. Mol. Cell. Biol. 17, 1522–1534.
Lee, H., O’Garra, A., Arai, K., and Arai, N. (1998). Characterization of
cis-regulatory elements and nuclear factors conferring Th2-specific
expression of the IL-5 gene: a role for GATA-binding protein. J.
Immunol. 160, 2343–2352.
Rooney, J.W., Hodge, M.R., McCaffrey, P.G., Rao, A., and Glimcher,
L.H. (1994). A common factor regulates both Th1- and Th2-specific
cytokine gene expression. EMBO J. 13, 625–633.
Sallusto, F., Lanzavecchia, A., and Mackay, C. (1998). Chemokines
and chemokine receptors in T-cell priming and Th1/Th2-mediated
responses. Immunol. Today 19, 568–574.
Li, B., Tournier, C., Davis, R., and Flavell, R. (1999). Regulation of
IL-4 expression by the transcription factor JunB during T helper cell
differentiation. EMBO J. 18, 420–432.
Schindler, U., Wu, P., Rothe, M., Brasseur, M., and McKnight, S.L.
(
1995). Components of a Stat recognition code: evidence for two
Li-Weber, M., Salgame, P., Hu, C., Davydov, I.V., and Krammer,
P.H. (1997). Differential interaction of nuclear factors with the PRE-I
enhancer element of the human IL-4 promoter in different T cell
subsets. J. Immunol. 158, 1194–1200.
layers of molecular selectivity. Immunity 2, 689–697.
Sweetser, M.T., Hoey, T., Sun, Y.L., Weaver, W.M., Price, G.A., and
Wilson, C.B. (1998). The roles of nuclear factor of activated T cells
and ying-yang 1 in activation-induced expression of the interferon-
gamma promoter in T cells. J. Biol. Chem. 273, 34775–34783.
Lopez-Rodriguez, C., Aramburu, J., Rakeman, A., and Rao, A. (1999).
NFAT5, a constitutively nuclear NFAT protein that does not cooper-
ate with Fos and Jun. Proc. Natl. Acad. Sci. USA 96, 7214–7219.
Szabo, S., Dighe, A., Gubler, U., and Murphy, K. (1997a). Regulation
of the interleukin (IL)-12R b2 subunit expression in developing T
helper 1 (Th1) and Th2 cells. J. Exp. Med. 185, 817–824.
Lyakh, L., Ghosh, P., and Rice, N.R. (1997). Expression of NFAT-
family proteins in normal human T cells. Mol. Cell. Biol. 17, 2475–
2
484.
Szabo, S.J., Glimcher, L.H., and Ho, I. (1997b). Genes that regulate
interleukin-4 expression in T cells. Curr. Opin. Immunol. 9, 776–781.
Miaw, S.C., Choi, A., Yu, E., Kishikawa, H., and Ho, I.C. (2000). ROG,
repressor of GATA, regulates the expression of cytokine genes.
Immunity 12, 323–333.
Takemoto, N., Koyano-Nakagawa, N., Yokota, T., Arai, N., Miyatake,
S., and Arai, K.-I. (1998). Th2-specific DNAse I-hypersensitive sites
in the murine IL-13 and IL-4 intergenic region. Intl. Immunol. 10,
Miyakawa, H., Woo, S., Dahl, S., Handler, J., and Kwon, H.M. (1999).
Tonicity-responsive enhancer binding protein, a rel-like protein that
stimulates transcription in response to hypertonicity. Proc. Natl.
Acad. Sci. USA 96, 2538–2542.
1
981–1985.
Timmerman, L., Healy, J., Ho, S., Chen, L., Goodnow, C., and
Crabtree, G. (1997). Redundant expression but selective utilization
of NFAT family members. J. Immunol. 159, 2735–2740.
Mosmann, T., and Sad, S. (1996). The expanding universe of T-cell
subsets: Th1, Th2 and more. Immunol. Today 17, 138–146.
Ting, C.-N., Olson, M., Barton, K., and Leiden, J. (1996). Transcription
factor GATA-3 is required for development of the T-cell lineage.
Nature 384, 474–478.
Murphy, K. (1998). T lymphocyte differentiation in the periphery.
Curr. Opin. Immunol. 10, 226–232.
Wang, D., McCaffrey, P., and Rao, A. (1995). The cyclosporin-sensi-
tive transcription factor NFATp is expressed in several classes of
cells in the immune system. Ann. NY Acad. Sci. 766, 182–194.
O’Garra, A. (1998). Cytokines induce the development of functionally
heterogenous T helper cell subsets. Immunity 8, 275–283.
Orkin, S. (1995). Transcription factors and hematopoietic develop-
ment. J. Biol. Chem. 270, 4955–4958.
Wenner, C.A., Szabo, S.J., and Murphy, K.M. (1997). Identification
of IL-4 promoter elements conferring Th2-restricted expression dur-
ing T helper cell subset development. J. Immunol. 158, 765–773.
Ouyang, W., Lohning, M., Gao, Z., Assenmacher, M., Ranganath,
S., Radbruch, A., and Murphy, K. (2000). Stat6-independent GATA-3
autoactivation directs IL-4-independent Th2 development and com-
mitment. Immunity 12, 27–37.
Xanthoudakis, S., Viola, J., Shaw, K., Luo, C., Wallace, J., Bozza,
P., Curran, T., and Rao, A. (1996). An enhanced immune response in
mice lacking the transcription factor NFAT1. Science 272, 892–895.
Parekh, B.S., and Maniatis, T. (1999). Virus infection leads to local-
ized hyperacetylation of histones H3 and H4 at the IFN-␤ promoter.
Mol. Cell 3, 125–129.
Yoshida, H., Nishina, H., Takimoto, H., Marengere, L., Wakeham, A.,
Bouchard, D., Kong, Y.-Y., Ohteki, T., Shahinian, A., Bachmann, M.,
et al. (1998). The transcription factor NFATc1 regulates lymphoid
proliferation and Th2 cytokine production. Immunity 8, 115–124.
Penix, L., Weaver, W., Pang Y., Young, H., and Wilson, C. (1993).
Two essential regulatory elements in the human interferon gamma
promoter confer activation specific expression in T cells. J. Exp.
Med. 178, 1483–1496.
Young, H. (1996). Regulation of interferon-gamma gene expression.
J. Interferon Cytokine Res. 16, 563–568.
Zhang, D., Cohn, L., Ray, P., Bottomly, K., and Ray, A. (1997). Tran-
scription factor GATA-3 is differentially expressed in murine Th1 and
Th2 cells and controls Th2-specific expression of the interleukin-5
gene. J. Biol. Chem. 272, 21597–21603.
Penix, L., Sweetser, M., Weaver, W., Hoeffler, J., Kerppola, T., and
Wilson, C. (1996). The proximal regulatory element of the interferon-
gamma promoter mediates selective expression in T cells. J. Biol.
Chem. 271, 31964–31972.
Zhang, D.-H., Yang, L., and Ray, A. (1998). Differential respon-
siveness of the IL-5 and IL-4 genes to transcription factor GATA-3.
J. Immunol. 161, 3817–3821.
Ranganath, S., Ouyang, W., Bhattacharya, D., Sha, W., Grupe, A.,
Peltz, G., and Murphy, K. (1998). GATA-3-dependent enhancer activ-
ity in IL-4 gene regulation. J. Immunol. 161, 3822–3826.
Zhang, D., Yang, L., Cohn, L., Parkyn, L., Homer, R., Ray, P., and
Ray, A. (1999). Inhibition of allergic inflammation in a murine model
of asthma by expression of a dominant-negative mutant of GATA-3.
Immunity 11, 473–482.
Ranger, A., Oukka, M., Rengarajan, J., and Glimcher, L. (1998a).
Inhibitory function of two NFAT family members in lymphoid homeo-
stasis and Th2 development. Immunity 9, 627–635.
Zheng, W., and Flavell, R.A. (1997). The transcription factor GATA-3
is necessary and sufficient for Th2 cytokine gene expression in CD4
T cells. Cell 89, 587–596.
Ranger, A., Hodge, M., Gravallese, E., Oukka, M., Davidson, L., Alt,
F., Charles de la Brousse, F., Hoey, T., Grusby, M., and Glimcher, L.
(1998b). Delayed lymphoid repopulation with defects in IL-4-driven
responses produced by inactivation of NFATc. Immunity 8, 125–134.
Rao, A., Faas, S.J., and Cantor, H. (1984). Activation specificity of
arsonate-reactive T cell clones. J. Exp. Med. 159, 479–494.
Rao, A., Luo, C., and Hogan, P.G. (1997). Transcription factors of
the NFAT family: regulation and function. Annu. Rev. Immunol. 15,
7
07–747.
Rincon, M., and Flavell, R.A. (1997a). T-cell subsets: transcriptional
control in the Th1/Th2 decision. Curr. Biol. 7, R729–R732.
Rincon, M., and Flavell, R. (1997b). Transcription mediated by NFAT