IL-33, the IL-1-like cytokine ligand for ST2 receptor,
is a chromatin-associated nuclear factor in vivo
Virginie Carriere*, Lucie Roussel*, Nathalie Ortega*, Delphine-Armelle Lacorre*, Laure Americh†, Luc Aguilar†,
Ge ´rard Bouche*, and Jean-Philippe Girard*‡
*Laboratoire de Biologie Vasculaire, Equipe Labellise ´e ‘‘La Ligue 2006,’’ Institut de Pharmacologie et de Biologie Structurale, Centre National
de la Recherche Scientifique, Unite ´ Mixte de Recherche 5089, 205 Route de Narbonne, 31077 Toulouse, France; and†Endocube,
Prologue Biotech, BP700, Rue Pierre et Marie Curie, 31319 Labe `ge Cedex, France
Edited by Charles A. Dinarello, University of Colorado Health Sciences Center, Denver, CO, and approved November 6, 2006 (received for review
August 8, 2006)
Recent studies indicate that IL-1? functions intracellularly in path-
ways independent of its cell surface receptors by translocating to
the nucleus and regulating transcription. Similarly, the chromatin-
associated protein HMGB1 acts as both a nuclear factor and a
secreted proinflammatory cytokine. Here, we show that IL-33, an
IL-1-like cytokine that signals via the IL-1 receptor-related protein
ST2 and induces T helper type 2-associated cytokines, is an endo-
thelium-derived, chromatin-associated nuclear factor with tran-
scriptional repressor properties. We found that IL-33 is identical to
NF-HEV, a nuclear factor preferentially expressed in high endothe-
situ hybridization demonstrated that endothelial cells constitute a
major source of IL-33 mRNA in chronically inflamed tissues from
patients with rheumatoid arthritis and Crohn’s disease. Immuno-
staining with three distinct antisera, directed against the N-termi-
nal part and IL-1-like C-terminal domain, revealed that IL-33 is a
heterochromatin-associated nuclear factor in HEV endothelial cells
in vivo. Association of IL-33 with heterochromatin was also ob-
served in human and mouse cells under living conditions. In
addition, colocalization of IL-33 with mitotic chromatin was noted.
Nuclear localization, heterochromatin-association, and targeting
to mitotic chromosomes were all found to be mediated by an
evolutionarily conserved homeodomain-like helix–turn–helix mo-
tif within the IL-33 N-terminal part. Finally, IL-33 was found to
possess transcriptional repressor properties, associated with the
homeodomain-like helix–turn–helix motif. Together, these data
suggest that, similarly to IL1? and HMGB1, IL-33 is a dual function
protein that may function as both a proinflammatory cytokine and
an intracellular nuclear factor with transcriptional regulatory
heterochromatin ? nucleus ? endothelium ? HMGB1
The three best known members of this family, IL-1?, IL-1?, and
IL-18, are highly inflammatory cytokines, and dysregulation of
(1, 2). IL-33 is the most recent addition to the IL-1 family (3).
IL-33 has been shown to induce T helper (Th) type 2 responses
(3) by signaling through the IL-1 receptor-related protein ST2
(IL-1 R4), an orphan member of the IL-1 receptor family (4, 5).
Treatment of mice with recombinant IL-33 resulted in blood
eosinophilia, splenomegaly, and increased serum levels of IgE,
IgA, IL-5, and IL-13 (3). Exposure to IL-33 also caused severe
pathological changes in the lungs and gastrointestinal tract,
including eosinophilic and mononuclear infiltrates, increased
mucus production, and epithelial cell hyperplasia and hypertro-
IL-33 shares with the other members of the IL-1 family a single
IL-1/FGF ?-trefoil fold (6, 7). Similarly to IL-1? (271 aa, 30.6 kDa)
and IL-1? (269 aa, 30.7 kDa) (1), IL-33 (270 aa, 30.7 kDa) is
ytokines of the IL-1 family play a major role in a wide range
of inflammatory, infectious, and autoimmune diseases (1).
This IL-33 precursor has been shown to be cleaved by caspase-1 in
vitro, and it has been proposed that, similarly to IL-1? and IL-18 (1,
activity (3). However, it remains to be determined whether IL-33 is
processed to a mature active form by caspase-1 in vivo.
The observation that the precursor form of each member of the
IL-1 family (8), with the exception of the IL-1 receptor antagonist
IL-1Ra, lacks a signal peptide suggested persistence of an early
evolutionary role of these proteins as intracellular factors (9).
and growth factors of the IL-1/FGF family. For instance, IL-1?,
which is rarely found in the extracellular compartment but rather is
primarily a cell-associated cytokine (1, 10), has been proposed to
regulate cell migration, proliferation, senescence, and differentia-
tion through intracrine mechanisms and intracellular pathways
independent of its cell-surface membrane receptors (9, 11–16).
Nuclear translocation of the IL-1? precursor, mediated by a con-
sensus nuclear localization sequence (NLS) within its N-terminal
(Nter) part (17), has been shown to be critical for the intracellular
functions of IL-1? (12, 13). Several nuclear targets of the IL-1?
precursor have been identified that interact specifically with the
16). These targets include the histone acetyltransferases p300,
PCAF and Gcn5 (16), and the growth suppressor necdin (15). The
identification of the histone acetyltransferases as nuclear partners
of IL-1? is in agreement with a recent report demonstrating an
important role of the IL-1? precursor as an intracrine proinflam-
matory activator of transcription (9). Together, these observations
indicate that IL-1? is a dual-function protein that acts as both a
nuclear factor and a proinflammatory cytokine (9). Interestingly, a
similar duality of function has been shown for high-mobility group
box 1 (HMGB1) protein, an abundant chromatin-associated pro-
tein involved in transcriptional regulation that is released by ne-
crotic cells and secreted by activated macrophages during inflam-
mation and functions extracellularly as a potent proinflammatory
In this study, we show that IL-33, which we found to be identical
to the NF-HEV protein we characterized (21), is a nuclear factor
associated with heterochromatin in vivo and mitotic chromosomes
Author contributions: V.C. and L.R. contributed equally to this work; J.-P.G. designed
research; V.C., L.R., N.O., D.-A.L., L. Americh, and G.B. performed research; V.C., L.R., L.
Aguilar, and J.-P.G. analyzed data; and J.-P.G. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS direct submission.
Abbreviations: Cter, C-terminal; EC, endothelial cell; HEV, high endothelial venules; HTH,
helix–turn–helix; ISH, in situ hybridization; IHC, immunohistochemistry; NLS, nuclear lo-
calization sequence; Nter, N-terminal; RA, rheumatoid arthritis.
‡To whom correspondence should be addressed. E-mail: email@example.com.
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2006 by The National Academy of Sciences of the USA
January 2, 2007 ?
vol. 104 ?
no. 1 www.pnas.org?cgi?doi?10.1073?pnas.0606854104
in living cells, that possesses potent transcriptional-repressor prop-
erties. Our findings suggest that IL-33, similarly to IL-1?, may
function as both a proinflammatory cytokine and an intracellular
nuclear factor involved in transcriptional regulation.
Endothelial Cells (ECs) Constitute a Major Source of IL-33 mRNA in
Chronically Inflamed Tissues from Patients with Rheumatoid Arthritis
(RA) and Crohn’s Disease. Using in situ hybridization (ISH), we have
shown abundant and preferential expression of NF-HEV/IL-33
patches (21). In contrast, RT-PCR analysis of human cDNA
libraries has revealed expression of IL-33 mRNA in activated
dermal fibroblasts, keratinocytes, and activated bronchial smooth
muscle cells, but expression in ECs was not reported (3). To define
the major cell types responsible for IL-33 mRNA expression in
chronically inflamed human tissues, we further characterized IL-33
mRNA expression by ISH. Prominent and specific dot-like signals
were detected with the antisense IL-33 ISH probe in blood vessels
from human tonsils (Fig. 1A), Crohn’s disease intestine (Fig. 1B),
and RA synovium (Fig. 1C), whereas the sense probe gave only
background signals. Combined ISH for IL-33 and immunohisto-
chemistry (IHC) for postcapillary venule EC marker DARC re-
vealed expression of IL-33 mRNA in both DARC-positive and
-negative ECs (Fig. 1C). Semiquantitative RT-PCR confirmed
abundant expression of IL-33 mRNA in ECs freshly isolated from
human tonsils, RA synovium, and Crohn’s intestine (Fig. 1D). In
contrast, IL-33 mRNA was not detected in HeLa epithelial cancer
cells. Together, these findings indicate that ECs constitute a major
source of IL-33 mRNA in chronically inflamed tissues.
already described NF-HEV/IL-33 as a nuclear factor preferentially
expressed in HEV endothelial cells from human tonsils (21).
Because IL-33 has been rediscovered as a cytokine ligand for a
cell-surface receptor (3), we decided to confirm its nuclear local-
ization using three distinct antisera against the Nter part (Nter,
IL-33, amino acids 1–15) and IL-1-like Cter domain (Cter1, Cter2).
Immunostaining of human tonsil sections with the three indepen-
dent antisera demonstrated abundant expression of endogenous
IL-33 in the nuclei of HEV ECs double stained with DAPI and
HEV-specific antibody MECA-79 (Fig. 2 A, D, and G). Nuclear
tissues. (A–C) ISH analysis of IL-33 mRNA in human tonsils (A), Crohn’s disease
intestine (B), and RA synovium (C). Cryosections were hybridized with IL-33
sense and antisense riboprobes. ISH signals were detected with FITC-
conjugated anti-biotin antibody (A and C) or incubation with Fast-red (B). In
C, ISH (green signal) was combined with IHC for the postcapillary venule
marker DARC (red signal), and nuclei were counterstained with DAPI (blue
signal). (C Inset) Abundant IL-33 signal (green dots) in a DARC-positive post-
capillary venule (red signal). (D) Detection of IL-33 mRNA in ECs freshly
purified from human tonsils, Crohn’s disease intestine, and RA synovium by
using semiquantitative RT-PCR. HeLa epithelial cell line and GAPDH primers
were used as controls.
Expression of IL-33 mRNA in ECs from chronically inflamed human
mAb MECA79 and three distinct IL-33 antisera against the Nter and IL-1-like
Cter domains, preabsorbed with a control peptide, demonstrated abundant
expression of endogenous IL-33 in the nuclei of HEV ECs in vivo. Nuclei were
counterstained with DAPI. (B, E, and H) Nuclear staining of HEV ECs with the
three IL-33 antisera was abrogated by preincubation of the antibodies with
that endogenous IL-33 accumulates in nuclear domains that colocalize with
dense regions of DAPI staining, indicating association with heterochromatin.
IL-33 is a heterochromatin-associated nuclear factor in HEV ECs in
Carriere et al.
January 2, 2007 ?
vol. 104 ?
no. 1 ?
staining of HEV ECs with the anti-IL-33 antibodies was specific
because it was abrogated by preincubating the antibodies with the
corresponding IL-33 peptides (Fig. 2 B, E, and H) but not control
peptides (Fig. 2 A, D, and G). Interestingly, higher magnification
revealed that endogenous IL-33 is enriched in multiple nuclear
domains (Fig. 2 C, F, and I). The IL-33-containing domains
overlapped with high local concentrations of DNA as shown by in
Because DAPI and Hoechst are known to bind preferentially
endogenous IL-33 with heterochromatin. We observed similar
independent antisera, indicating that both Nter and Cter parts of
IL-33 accumulate in the nucleus of HEV ECs and associate with
heterochromatin in vivo.
IL-33 Associates with Heterochromatin and Mitotic Chromosomes in
Living Cells. To characterize association of IL-33 with heterochro-
matin under living conditions, we analyzed the subcellular local-
ization of IL-33 tagged with GFP at the C terminus, in various cell
types. Nuclear accumulation and heterochromatin association of
including human HEK-293T (Fig. 3A) and HeLa (data not shown)
GFP alone localized throughout the cell without accumulation in
the nucleus (Fig. 3C). In all cell types, IL-33 colocalized with dense
regions of Hoechst staining, including the perinucleolar hetero-
chromatin at the nucleolar periphery in human cells and the
pericentromeric heterochromatin in mouse cells (Fig. 3 A and B).
IL-33 was tagged with GFP at the N terminus (Fig. 3D). We then
asked whether IL-33 was still associated with chromatin during
mitosis. As expected, GFP alone was distributed throughout the
whole cell and mainly excluded from the condensed mitotic chro-
mosomes (Fig. 3E). In contrast, we observed a clear association of
IL-33-GFP with mitotic chromatin, as revealed by the bright green
fluorescence on the condensed chromosomal arms counterstained
with Hoechst (Fig. 3F). Unlike HMGB1 (22), association of IL-33
with mitotic chromosomes was retained after permeabilization and
3G). Finally, we confirmed association of IL-33 with chromatin,
using a biochemical approach (Fig. 3H). Chromatin fractions were
prepared from control HEK-293T cells or cells transfected with
IL-33 expression vector by digesting nuclei with DNase and ex-
tracting proteins with 0.4 M or 0.8 M NaCl. Western blot analysis
with the three distinct IL-33 antibodies revealed the presence of
full-length IL-33 (?30 kDa) in both 0.4 M and 0.8 M NaCl
chromatin extracts and residual pellet (corresponding to proteins
not extracted at 0.8 M NaCl), similarly to the internal control
An Evolutionarily Conserved Motif Within the N Terminus Is Necessary
and Sufficient for IL-33 Nuclear Localization, Heterochromatin Asso-
ciation, and Targeting to Mitotic Chromosomes. Sequence alignment
of vertebrate IL-33 sequences revealed two evolutionarily con-
served regions [see supporting information (SI) Fig. 6]: the pre-
dicted homeodomain-like helix–turn–helix (HTH) motif (amino
(amino acids 112–270) (3). We observed that fusion of GFP to the
IL-1-like domain resulted in an even distribution of the fusion
(amino acids 1–111) resulted in nuclear localization and hetero-
chromatin association in living cells (Fig. 4 A and B). Essentially
identical results demonstrating the critical role of the IL-33 Nter
part in nuclear targeting and heterochromatin association were
epithelial cells and mouse 3T3 fibroblasts. To further define, within
the IL-33 Nter part, the molecular determinants responsible for
IL-33 localization, we first investigated the role of the candidate
bipartite NLS (21). Surprisingly, double and quadruple point mu-
tations in the predicted NLS revealed that it is not required for
IL-33 nuclear targeting and heterochromatin association (Fig. 4C).
In contrast, deletions targeting the homeodomain-like HTH motif
(amino acids 1–65) abrogated nuclear accumulation and hetero-
chromatin association of the IL-33-GFP fusion proteins (Fig. 4 A
and D), and this motif alone was sufficient for targeting GFP to the
nucleus and heterochromatin (Fig. 4D). Interestingly, the home-
cells. (A–C) Association of IL-33-GFP protein with heterochromatin was ob-
mouse 3T3 fibroblasts (B). Colocalization of IL-33-GFP with dense regions of
Hoechst staining was found at the perinucleolar heterochromatin in human
cells and the pericentromeric heterochromatin in mouse cells (A and B, ar-
rowheads), whereas GFP alone localized throughout the cell (C). (D) Hetero-
chromatin association of IL-33 in living cells was also observed with an Nter
GFP tag. (E–G) During mitosis, IL-33-GFP was found associated with mitotic
chromatin (F), whereas GFP alone was distributed throughout the whole cell
(E). Association of IL-33-GFP with mitotic chromosomes was retained in fixed
cells (G). Staining of mitotic chromosomes with IL-33 Nter antibody was
with the three IL-33 antisera (Nter, Cter1, and Cter2) and control histone H3
antibody. Chromatin fractions were prepared by digestion of nuclei with
DNase and extraction with 0.4 M NaCl and 0.8 M NaCl. Residual pellet
corresponds to proteins not extracted at 0.8 M NaCl.
www.pnas.org?cgi?doi?10.1073?pnas.0606854104Carriere et al.
odomain-like HTH motif was also found to be sufficient for
association of IL-33-GFP fusion proteins with mitotic chromo-
somes (Fig. 4E). We concluded that the evolutionarily conserved
homeodomain-like HTH motif within the Nter part is necessary
and sufficient for IL-33 targeting to the nucleus, heterochromatin,
and mitotic chromosomes.
IL-33 Has Transcriptional Repressor Properties Associated with the
Homeodomain-Like HTH Motif. Because one of the hallmarks of
heterochromatin is that it constitutes a transcriptionally repressive
environment in the nucleus, we next wished to determine whether
IL-33 may possess transcriptional-repressor properties. For that
purpose, full-length IL-33 or deletion constructs were fused to the
Gal4-DNA-binding domain (Gal4-DB) and used in gene reporter
assays with a GAL4-responsive luciferase reporter. The full-length
repressor activity (Fig. 5A) that was not observed with a reporter
lacking GAL4-responsive elements (Fig. 5B). This effect required
tethering of IL-33 to DNA through the GAL4-DB, and it was
similar to that of the potent heterochromatin-associated transcrip-
tional repressor histone lysine methyltransferase SUV39H1 (Fig. 5
A and C). Analysis of IL-33 deletion mutants revealed a good
Nter part of IL-33 (amino acids 1–111) and the homeodomain-like
HTH motif (amino acids 1–65), whereas the Cter IL-1-like domain
(amino acid 112–270) did not exhibit any significant transcriptional
regulatory properties when fused to the GAL4-DB (Fig. 5D).
Together, these data indicate that IL-33 possesses potent transcrip-
tional-repressor activity associated with the evolutionarily con-
served Nter homeodomain-like HTH motif.
IL-33 is a recently described member of the IL-1 family that
signals through the IL-1 receptor-related protein ST2 (3). IL-33
was found to be a potent inducer of Th2 responses and Th2-
associated cytokines IL-4, IL-5, and IL-13, suggesting that IL-33
may play an important role in asthma and other allergic-type
diseases (3, 23). In this study, we show that IL-33 localizes to the
nucleus, associates with heterochromatin and mitotic chromo-
tin association, and targeting to mitotic chromosomes. (A) Diagram of IL-33 and
its deletion- or point-mutant derivatives. The capacity of each mutant to accu-
was determined. (B–E) The different IL-33 mutants fused to GFP were expressed
Hoechst 33342 in living cells and with DAPI in fixed cells.
An evolutionarily conserved homeodomain-like HTH motif within the
Human HEK-293T epithelial cells were transiently transfected with increasing
(Gal4-IL33) or Suv39H1 (Gal4-Suv), along with reporter vector pLex-Gal4 in
which luciferase gene transcription is under the control of five GAL4-DB-
binding sites (A) or with the same reporter vector deleted of the GAL4-DB-
and expressed relative to the values obtained with empty expression vector
pCMV-2N3T (control). Results are means and standard deviations of three
independent transfection experiments. (C) Fold repression was calculated by
dividing the normalized luciferase activity of cells expressing GAL4-DB alone
(Gal4) by the activity of the IL-33 (Gal4-IL33) and Suv39H1 (Gal4-Suv) fusion
proteins (500-ng expression vector). (D) The IL-33 Nter part and homeodo-
main-like HTH motif, but not the IL-1-like Cter part, exhibit transcriptional
repressor activity. IL-33 full-length (1–270) or deletion mutants (1–65, 1–111,
and 112–270) were expressed as GAL4-DB fusions (10-, 100-, 500-, or 750-ng
expression vectors) and tested for their ability to repress the pLex-Gal4 lucif-
erase reporter. Normalized luciferase activities were determined as described
IL-33 has potent transcriptional repressor properties. (A and B)
Carriere et al.
January 2, 2007 ?
vol. 104 ?
no. 1 ?
somes, and exhibits potent transcriptional-repressor properties.
To the best of our knowledge, interleukins have not been shown
previously to exhibit such a localization profile in living cells
and/or human tissues in vivo. In addition, we identify an evolu-
tionarily conserved domain within the IL-33 Nter part that is
necessary and sufficient for IL-33 nuclear localization, associa-
tion with heterochromatin and mitotic chromosomes, and tran-
scriptional-repression activity. This domain is predicted to ex-
hibit structural homology with the homeodomain and other
HTH DNA-binding domains (21), but has no similarity with the
Nter part of other IL-1 family cytokines. Together, our data
provide strong evidence that IL-33 is a ‘‘dual-function’’ cytokine
that may function as both an intracellular nuclear factor and a
potent proinflammatory cytokine. A similar duality of function
has been shown for IL-1? and chromatin-associated factor
HMGB1 (9, 11–16, 18–20).
By acting as a dual-function cytokine, IL-33 may be more similar
to IL-1? than to IL-1?. Indeed, IL-33 (270 aa) and IL-1? (271 aa)
have a similar size, share the IL-1/FGF fold at their C terminus,
translocate to the nucleus through signals located within their Nter
parts, and exhibit transcriptional regulatory properties. IL-33 also
shares similarity with HMGB1 by its capacity to associate with
passively released by necrotic cells (19) or secreted by activated
macrophages after hyperacetylation of lysine residues (24). It
remains to be seen whether similar mechanisms may contribute to
IL-33 release during inflammation. The mature form of IL-33 has
been proposed to be secreted after maturation by caspase-1 (3).
Surprisingly, the predicted cleavage site for caspase-1 was not
conserved in the canine, bovine, and porcine IL-33 orthologues (SI
by caspase-1 in vivo (3). In agreement with this observation, we
found no evidence for IL-33 processing in vivo, i.e., differential
localization of N- and Cter parts, for both endogenous IL-33 in
HEV ECs (Fig. 2 A, D, and G) and ectopic IL-33 in HEK-293T
epithelial cells (Fig. 3 A, D, and G). Membrane-associated IL1?
biological activity has been demonstrated in many studies (1, 10,
25), and it remains possible that IL-33 may similarly function as a
that IL-33 is a heterochromatin-associated nuclear factor in vivo,
and future studies will therefore be required to determine how the
IL-1-like cytokine domain is released and/or presented to the ST2
receptor expressed on target cells.
Our data show that ECs constitute a major source of IL-33
mRNA and protein in vivo. ISH analysis in human chronically
inflamed tissues revealed abundant expression of IL-33 mRNA in
blood vessel ECs from RA synovium and Crohn’s disease intestine,
suggesting that IL-33 may play an important role at the level of the
endothelium during chronic inflammation. IHC with three distinct
antisera directed against the Nter part and IL-1-like Cter domain,
indicate that IL-33 is abundantly expressed by HEVs in human
tonsils. Analyses have shown that IL-33 mRNA constitutes one of
Together, these expression data indicate that IL-33 is likely to play
important roles in HEV ECs. Like HMGB1, IL-33 may act at the
chromatin level and function as a nonhistone chromosomal protein
involved in the assembly of nucleoprotein complexes on DNA and
the maintenance or establishment of chromatin structure (27, 28).
Similarly to the heterochromatin-associated lymphocyte-specific
transcription factor Ikaros, which plays a key role in the transcrip-
tional regulation of genes required for lymphocyte development
(29), IL-33 may regulate gene-expression programs required for
development and/or maintenance of HEV ECs in vivo. We found
that IL-33 has transcriptional repression properties and localizes to
heterochromatin, a nuclear domain linked to gene silencing (30).
IL-33 may therefore repress gene expression in HEV ECs. How-
ever, as proposed for Ikaros (31), IL-33 could also function as a
sors at heterochromatin, thus decreasing their local concentrations
on specific promoters and allowing activators to bind more effi-
ciently. Ikaros has been shown to associate with histone deacety-
lases and chromatin remodeling complexes (32, 33), and it will
be important in future studies to determine whether IL-33 also
associates with histone-modifying and chromatin-remodeling
So far, HEV ECs constitute the only cell type that has been
they rapidly dedifferentiate outside the lymphoid tissue microen-
vironment (34). We looked for another cell culture model, but we
were not able to detect endogenous IL-33 protein expression in any
cultured cell lines, even after stimulation with proinflammatory
mediators. These later observations suggest that in vivo studies will
be required to further define the biological roles of IL-33 and the
molecular mechanisms regulating its production.
Materials and Methods
Plasmid Constructions. Plasmid pEGFP.N3 (BD Biosciences Clon-
tech, Mountain View, CA) encoding EGFP was modified by
deletion of Kozak and ATG sequences of EGFP. IL-33 and
IL-33 deletion mutants were amplified by PCR using the human
NF-HEV/IL-33 cDNA (21) as a template; all sense primers con-
tained an EcoRI site, a Kozak sequence (CCACC) and an ATG
start codon, whereas the stop codon in antisense primers was
replaced with a BamHI site. The PCR fragments, thus obtained,
were digested with EcoRI and BamHI and cloned in-frame up-
stream of EGFP in pEGFP.N3 modified vector. Point mutants of
IL-33 were generated by two successive rounds of PCR using
pEGFP.N3-IL-33 as a template. IL-33 cDNA was also cloned in
plasmid pEGFP.C2 (BD Biosciences Clontech) to express IL-33
with an Nter GFP tag and in plasmid pEGFP.N3 with a stop codon
before GFP to express IL-33 with no tag. All primer sequences are
available upon request.
Mammalian Cell Culture and Fluorescence Microscopy. Human HeLa
and HEK293T epithelial cell lines and mouse 3T3 fibroblasts were
grown in DMEM supplemented with 10% FCS and 1% penicillin-
streptomycin (all from Invitrogen, Carlsbad, CA). For live cell
imaging, cells seeded (150 ? 103cells per well) in Chamber Slide
System 2 wells (Lab-Tek; Nunc, Roskilde, Denmark) 24 h before
transfection were transfected with 2 ?g of plasmid DNA by using
a phosphate calcium precipitation method. Two days after trans-
fection, DNA was stained for 10 min at 37°C in culture medium
were observed by fluorescence microscopy on an inverted fluores-
cence microscope (Eclipse TE300; Nikon Corp, Tokyo, Japan).
In Situ Hybridization. Tissue samples for ISH and IHC experiments
have been described (35). ISH was performed as described (34) by
NM?033439, IL-33 cDNA, nucleotides 216–828), labeled with the
DIG RNA Labeling Kit (Roche, Indianapolis, IN). Hybridized
(1/15; DAKO, Carpenteria, CA) by using the biotin-tyramide
(Sigma, St. Louis, MO). For combined ISH/IHC, ISH signal was
revealed with FITC-conjugated goat anti-biotin antibody (1/100;
Vector Laboratories, Burlingame, CA) and sections were further
incubated with DARC mAb [10 ?g/ml, mAb Fy6; kindly provided
by Y. Colin (Institut National de la Transfusion Sanguine, Paris,
France)]and Cy3-conjugated goat anti-mouse IgG (1/1,000; GE
Healthcare, Nottinghamshire, U.K.). Sections were counterstained
with DAPI and viewed on a Nikon Eclipse TE300 fluorescence
www.pnas.org?cgi?doi?10.1073?pnas.0606854104 Carriere et al.
IHC. IHC experiments with IL-33 antibodies were performed on
Antigen retrieval was done in a microwave oven in citrate buffer,
incubated overnight at 4°C with IL-33 Nter rabbit polyclonal
antibody (1/200; Eurogentec, Seraing, Belgium) adsorbed with
IL-33 Nter peptide (amino acid 1–15) or control peptide for 1 h at
37°C. Two other IL-33 antisera were also used, IL-33 Cter1 and
IL-33 Cter2 rabbit polyclonal antibodies (1/200, nos. 210–447 and
210–933; Alexis Biochemicals, Lausen, Switzerland), together with
their specific IL-33 Cter peptide (no. 522–098; Alexis Biochemi-
cals). The sections were successively incubated with biotinylated
goat anti-rabbit (1/200; Jackson ImmunoResearch, West Grove,
PA) and Cy3-coupled streptavidin (1/200; Zymed, San Francisco,
CA), followed by MECA79 mAb (20 ?g/ml; Pharmingen, San
search). Sections were finally counterstained with DAPI and
mounted with Moviol.
Semiquantitative RT-PCR. Isolation of microvascular ECs from the
different tissues and RNA analysis by semiquantitative RT-PCR
5?-CACCCCTCAAATGAATCAGG-3? and 5?-GGAGCTCCA-
CAGAGTGTTCC-3? for IL-33, 5?-ACCACAGTCCATGCCAT-
CAC-3? and 5?-TCCACCACCCTGTTGCTGTA-3? for the inter-
nal control GAPDH.
Preparation of Chromatin Fractions. For preparation of nuclear
chromatin extracts, 3 ? 107control 293T cells or cells transfected
with IL-33 (with no tag) expression vector were washed in PBS and
resuspended in 4 ml of chromatin fractionation buffer (0.15 M
NaCl/10 mM MgCl2/10 mM CaCl2/1 mM PMSF/15 mM Tris, pH
7.5/0.1% Tween 20). Cells were ruptured by using Ultra-Turrax
(Labortechnik; Staufen, Germany) in the presence of NP-10 to a
at 4°C), nuclei were digested with DNase 1 (0.2 ?g/?l) for 10 min
at 30°C and pelleted by a brief centrifugation. Chromatin fractions
were prepared by adding NaCl, to a final concentration of 400 mM,
to the nuclear pellets resuspended in chromatin fractionation
buffer. After 30 min at 4°C, the nuclei were centrifuged at 21,000?
g for 10 min, and the supernatant (chromatin fraction 0.4 M) was
saved. Chromatin fraction 0.8 M was similarly prepared by adding
NaCl to a final concentration of 0.8 M NaCl. The final pellet was
saved as residual pellet. Anti-histone H3 antibody [kindly provided
by D. Trouche (Laboratoire de Biologie Mole ´culaire des Eucaryo-
tes–Centre National de la Recherche Scientifique, Toulouse,
France) was used to validate the chromatin fractions.
Reporter Assay. Gal4-IL-33 expression vectors were generated by
inserting the corresponding IL33 fragments, generated by PCR,
into pCMVGT vector downstream of the Gal4-DB (amino acids
1–147). Cotransfection of HEK-293T cells with 10, 100, 500, or 750
ng of pCMVGT constructs, 700 ng of firefly luciferase reporter
vector (pLex-Gal4), and 50 ng of Renilla luciferase construct
(pRL-CMV; Promega, Madison, WI) was performed by using
JetPEI (Polyplus-transfection, San Marcos, CA). The amount of
CMV promoter was kept constant (750 ng) by using pCMV-2N3T
empty vector. pCMVGT, pCMVGT-SUV39H1 (Gal4-Suv), and
pLex-Gal4 vectors were provided by D. Trouche (37). After 24 h,
firefly and Renilla luciferase activities were assayed with the Dual
Luciferase Assay System (Promega). All transfections were nor-
malized to Renilla luciferase activity and repeated at least three
We thank D. Trouche for plasmid vectors and histone H3 antibody; Y.
Colin for DARC mAb; P. Brousset (Institut National de la Sante ´ et de
la Recherche Me ´dicale U563, Toulouse, France) for paraffin-embedded
tissues; F. Viala for iconography; and Prof. F. Amalric for stimulating
discussions. This work was supported by Ligue Nationale Contre le
Cancer (Equipe Labellise ´e), Agence Nationale de la Recherches Pro-
gramme Blanc ‘‘Cuboı ¨dale,’’ Re ´gion Midi-Pyre ´ne ´es, and Migration and
Inflammation European Network of Excellence Grant FP6-502935. V.C.
was supported by the Association pour la Recherche sur le Cancer.
1. Dinarello CA (1996) Blood 87:2095–2147.
2. Dinarello CA (2000) Eur Cytokine Netw 11:483–486.
3. Schmitz J, Owyang A, Oldham E, Song Y, Murphy E, McClanahan TK,
Zurawski G, Moshrefi M, Qin J, Li X, et al. (2005) Immunity 23:479–490.
4. Coyle AJ, Lloyd C, Tian J, Nguyen T, Erikkson C, Wang L, Ottoson P, Persson
P, Delaney T, Lehar S, et al. (1999) J Exp Med 190:895–902.
5. Townsend MJ, Fallon PG, Matthews DJ, Jolin HE, McKenzie AN (2000) J Exp
6. Priestle JP, Schar HP, Grutter MG (1988) EMBO J 7:339–343.
7. Zhang JD, Cousens LS, Barr PJ, Sprang SR (1991) Proc Natl Acad Sci USA
8. Dunn E, Sims JE, Nicklin MJ, O’Neill LA (2001) Trends Immunol 22:533–536.
9. Werman A, Werman-Venkert R, White R, Lee JK, Werman B, Krelin Y,
Voronov E, Dinarello CA, Apte RN (2004) Proc Natl Acad Sci USA 101:2434–
10. Kaplanski G, Farnarier C, Kaplanski S, Porat R, Shapiro L, Bongrand P,
Dinarello CA (1994) Blood 84:4242–4248.
11. Maier JA, Voulalas P, Roeder D, Maciag T (1990) Science 249:1570–1574.
12. Maier JA, Statuto M, Ragnotti G (1994) Mol Cell Biol 14:1845–1851.
13. McMahon GA, Garfinkel S, Prudovsky I, Hu X, Maciag T (1997) J Biol Chem
14. Stevenson FT, Turck J, Locksley RM, Lovett DH (1997) Proc Natl Acad Sci
15. Hu B, Wang S, Zhang Y, Feghali CA, Dingman JR, Wright TM (2003) Proc
Natl Acad Sci USA 100:10008–10013.
17. Wessendorf JH, Garfinkel S, Zhan X, Brown S, Maciag T (1993) J Biol Chem
18. Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, Frazier
A, Yang H, Ivanova S, Borovikova L, et al. (1999) Science 285:248–251.
19. Scaffidi P, Misteli T, Bianchi ME (2002) Nature 418:191–195.
20. Lotze MT, Tracey KJ (2005) Nat Rev Immunol 5:331–342.
21. Baekkevold ES, Roussigne M, Yamanaka T, Johansen FE, Jahnsen FL,
Amalric F, Brandtzaeg P, Erard M, Haraldsen G, Girard JP (2003) Am J Pathol
Marechal V (2003) Mol Biol Cell 14:3414–3426.
23. Dinarello CA (2005) Immunity 23:461–462.
24. Bonaldi T, Talamo F, Scaffidi P, Ferrera D, Porto A, Bachi A, Rubartelli A,
Agresti A, Bianchi ME (2003) EMBO J 22:5551–5560.
25. Kurt-Jones EA, Fiers W, Pober JS (1987) J Immunol 139:2317–2324.
26. Palmeri D, Zuo FR, Rosen SD, Hemmerich S (2004) J Leukoc Biol 75:910–927.
27. Thomas JO, Travers AA (2001) Trends Biochem Sci 26:167–174.
28. Agresti A, Bianchi ME (2003) Curr Opin Genet Dev 13:170–178.
29. Georgopoulos K (2002) Nat Rev Immunol 2:162–174.
30. Brown KE, Guest SS, Smale ST, Hahm K, Merkenschlager M, Fisher AG
(1997) Cell 91:845–854.
31. Koipally J, Heller EJ, Seavitt JR, Georgopoulos K (2002) J Biol Chem
32. Kim J, Sif S, Jones B, Jackson A, Koipally J, Heller E, Winandy S, Viel A,
Sawyer A, Ikeda T, et al. (1999) Immunity 10:345–355.
33. Koipally J, Renold A, Kim J, Georgopoulos K (1999) EMBO J 18:3090–3100.
34. Lacorre DA, Baekkevold ES, Garrido I, Brandtzaeg P, Haraldsen G, Amalric
F, Girard JP (2004) Blood 103:4164–4172.
35. Middleton J, Americh L, Gayon R, Julien D, Mansat M, Mansat P, Anract P,
Cantagrel A, Cattan P, Reimund JM, et al. (2005) J Pathol 206:260–268.
36. Patterson AM, Gardner L, Shaw J, David G, Loreau E, Aguilar L, Ashton BA,
Middleton J (2005) Arthritis Rheum 52:2331–2342.
37. Vandel L, Nicolas E, Vaute O, Ferreira R, Ait-Si-Ali S, Trouche D (2001) Mol
Cell Biol 21:6484–6494.
Carriere et al.
January 2, 2007 ?
vol. 104 ?
no. 1 ?