International Immunology, Vol. 19, No. 1, pp. 1–10
ª The Japanese Society for Immunology. 2006. All rights reserved.
For permissions, please e-mail: firstname.lastname@example.org
IL9 leads to airway inflammation by inducing
IL13 expression in airway epithelial cells
Ulla-Angela Temann1, Yasmina Laouar1, Elizabeth E. Eynon1,2, Robert Homer3,4and
Richard A. Flavell1,2
1Section of Immunobiology,2Howard Hughes Medical Institute and3Department of Pathology, Yale University School of
Medicine, New Haven, CT 06520, USA
4Pathology and Laboratory Medicine Service, VA Connecticut Healthcare System, 950 Campbell Avenue, West Haven,
CT 06511, USA
Keywords: asthma, cytokines, IL9, IL13
Constitutive expression of IL9 in the lungs of transgenic (Tg) mice resulted in an asthma-like
phenotype consisting of lymphocytic and eosinophilic lung inflammation, mucus hypersecretion and
mast cell hyperplasia. Several Th2 cytokines including IL4, IL5 and IL13 were expressed in the lung
in response to Tg IL9. IL13 was absolutely necessary for the development of lung pathology. To
understand how IL9 induces IL13-dependent lung inflammation and mucus production, we sought the
IL13-producing cells. Surprisingly, we found that the absence of T cells and B cells in recombinase-
activating gene 1 (RAG1)-deficient IL9 Tg mice enhanced lung inflammation and dramatically
enhanced IL13 production. In addition, the lack of mast cells or eosinophils in IL9 Tg mice did not
affect IL13 levels in the lung. In situ hybridization for IL13 on lung sections from RAG1?/? IL9 Tg mice
revealed that airway epithelial cells were the major IL13-producing cell type. Our results implicate the
lung epithelium as a potentially important source of inflammatory cytokines in asthma.
Allergic asthma is recognized as a chronic inflammatory dis-
order of the airways. The complex immune response to anti-
gen in the lung involves various inflammatory cell types
including T cells, B cells, eosinophils, macrophages and mast
cells (1). There is growing evidence that the predominant
cell type involved in the regulation of pathophysiology in
asthma is the CD4 T cell of the Th2 subtype characterized
by the expression of cytokines like IL4, IL5, IL9, IL10 and
IL13 (2). The contribution of these individual cytokines, how-
ever, to the development of pathologic and physiologic
changes seen in the lungs of asthmatics such as eosino-
philia, mucus hypersecretion, mast cell hyperplasia and
bronchial hyperresponsiveness are still not completely under-
stood. Recently, interest has been focused on IL9 and its
implications in the pathology of asthma (3).
IL9 was originally described as a growth factor for a
subset of murine T cell clones (4). Subsequently, more activ-
ities of IL9 on various cell types including mast cells,
hematopoietic progenitors, B cells, eosinophils, neutrophils
and airway epithelial cells have been described (5–10). Most
of these inflammatory cells are involved in the pathology
of asthma and therefore suggested a potentially important
function for IL9. Studies in humans and animals have shown
the involvement of IL9 in lung eosinophilia, mucus hyperse-
cretion, Ig production and pulmonary mastocytosis (7, 9–14).
These findings have been strongly supported by our studies
with transgenic (Tg) mice that constitutively express IL9 se-
lectively within their lungs (15, 16). These mice developed
many features that resembled human asthma, including
eosinophilic and lymphocytic infiltration of the lung, mucus
hypersecretion, sub-epithelial fibrosis, mast cell hyperplasia
and bronchial hyperresponsiveness. Further, we showed that
the expression of several Th2 cytokines besides IL9, includ-
ing IL4, IL5 and IL13, was necessary for the same pathology
in inducible, lung-specific IL9 Tg mice (16). We demon-
strated that IL13, in particular, was essential for the develop-
ment of lung inflammation and mucus production.
To investigate how IL9 induces pathology resulting from
Th2 cytokine expression, we sought to identify potential tar-
get cells for the action of IL9. Surprisingly, the elimination of
T cells did not abolish lung pathology in IL9 Tg mice. We
now show that airway epithelial cells are the major producers
of IL13 in response to the IL9 transgene expression in the
lung, in the absence of inflammatory cells such as T cells
Correspondence to: R. A. Flavell; E-mail: email@example.comReceived 8 March 2006, accepted 4 October 2006
Advance Access publication 13 November 2006
by guest on June 12, 2013
and eosinophils. Therefore, airway epithelial cells might rep-
resent a potential target cell for IL9 in the lung and an impor-
tant source of pathologic cytokines in asthma.
Mice (6- to 10-weeks old) were used for the experiments
described here. IL9 Tg mice expressing constitutive, lung-
specific IL9 under the control of the CC-10 promoter have
been previously described (15). Mice deficient for signal trans-
duction of activated T cells 6 (STAT6) (17), IL5 (18) or recom-
binase activating gene 1 (RAG1) (19) were purchased from
The Jackson Laboratory (Bar Harbor, ME, USA) and main-
tained on C57BL/6 background. IL13-deficient (?/?) mice
(20) were kindly provided by A.N.J. McKenzie (MRC, Cam-
bridge, UK) and back-crossed onto C57BL/6 background.
Genotyping of mice during back-crossing onto IL9 Tg mice
was performed by PCR analysis of tail DNA using specific
primers for wild-type (WT) and mutant knockout (KO) allele.
STAT6: 5#-ATG TCT CTG TGG GGC CTA ATT TCC-3# and
5#GCC TCA GGC CAG GAC TTC TC-3# (WT) and 5#-GAT
CTG GAC GAA GAG CAT CAG G-3# and 5#-GCC TCA
GGC CAG GAC TTC TC-3# (KO). IL5: 5#-CTG GCC TTC
AAC TCC TGA TCC T-3# and 5#-GAA CTC TTG CAG GTA
ATC CAG G-3# (WT) and 5#-CTT GGG TGG AGA GGC TAT
TC-3# and 5#-AGG TGA GAT GAC AGG AGA TC-3# (KO).
IL13: 5#-GGG TGA CTG CAG TCC TGG CT-3# and 5#-GTT
GCT CAG CTC CTC AAT AAG C-3# (WT) and 5#-CTT GGG
TGG AGA GGC TAT TC-3# and 5#-AGG TGA GAT GAC
AGG AGA TC-3# (KO). RAG1: 5#-CAA TCT TGC GCG GGA
CAC TTG ACC-3# and 5#-ATC GAC GTG AAG GCA GAT
GTT GAC-3# (WT) and 5#-CTT GGG TGG AGA GGC TAT
TC-3# and 5#-AGG TGA GAT GAC AGG AGA TC-3# (KO).
For the generation of mast cell-deficient mice, KitWvmice
on C57BL/6 background and KitWmice on WB/ReJ back-
ground (21) were obtained from The Jackson Laboratory.
Mast cell-deficient mice were identified by their white fur color.
Lung lavage, tissue fixation and staining
Mice were anesthetized by methoxyflurane inhalation and then
sacrificed by carbon dioxide inhalation. Lung lavage and
cell enumeration were performed as described previously
(15). The lungs were excised completely from the chest,
inflated with 1 ml of 10% formalin and immersed in 10%
formalin. Tissue processing and histological stainings were
performed by the Yale Medical School Research Histology,
Department of Pathology.
Quantitation of cytokine levels in lung lavage fluid was per-
formed by ELISA according to the manufacturer’s instruc-
tions. The detection limit for the IL5 ELISA (BD Bioscience,
San Diego, CA, USA) was 40 pg ml?1and for the IL13 ELISA
(R&D Systems, Minneapolis, MN, USA) was 39 pg ml?1.
RNase protection assay
RNase protection assay was performed as described pre-
viously (16) using the RiboQuant Multi-Probe RNase Pro-
tection Assay System according to the manufacturer’s
instructions (BD Bioscience).
In situ hybridization
Lung tissue was fixed in formaldehyde and processed into
paraffin. Sections (5 l) were cut, deparaffinized, and then
treated with proteinase K (10 lg ml?1) for 20 min at 37?C.
Subsequently, sections were treated with 0.1 M triethylnol-
amine/0.25% acetic anhydride, pH 8.0, for 10 min at room
temperature, and then rinsed in PBS. The IL13 probe, kindly
provided by Zhou Zhu (Yale University School of Medicine),
was generated as a 615-bp fragment from IL13 cDNA. Re-
striction sites for XhoI and XbaI were incorporated in the
primers (sense: 5#-CCT GTG CTC GAG CAG CAT GGT ATG
GAG TGT GG-3# and anti-sense: 5#-AGC CCC TCT AGA
ATA GGC AGC AAA CCA TGT CC-3#) and the fragment
Fig. 1. Characterization of inflammatory cells in lung lavage fluid from
IL9 Tg mice. IL9 transgene-negative (IL9 Tg–) or transgene-positive
(IL9 Tg+) mice were back-crossed onto IL13?/? or STAT6?/?
background. (A) Total lung lavage cell counts. (B) Differential cell
counts on lung lavage fluid. Results are expressed as total cell
numbers and were obtained from at least 200 cell counts. Data are
expressed as mean 6 SD. Significant differences in total cell numbers
(A) or in the number of eosinophils (B) compared with that of IL9 Tg
mice (IL9 Tg+) are indicated with an asterisk.
2IL9 induces IL13 in airway epithelial cells
by guest on June 12, 2013
was sub-cloned into plasmid pBluescript KS (Stratagene, La
Jolla, CA, USA). Sense and anti-sense RNA probes were gen-
erated using T7 and T3 promoter sites, labeled with a digoxi-
genin RNA labeling kit (Roche, Indianapolis, IN, USA), and
then denatured at 65?C for 10 min. Sense and anti-sense
probes were diluted in hybridization buffer (Ambion, Austin,
TX, USA) at 30 ng ml?1and added to sections for overnight
incubation at 55?C. The tissues were then washed twice with
43 standard saline citrate (SSC) for 5 min at room tempera-
ture, twice with 23 SSC for 10 min at 37?C and incubated
with RNAse A (10 lg ml?1) for 45 min at 37?C. This was fol-
lowed by two washes (10 min) in 23 SSC at room tempera-
ture and three washes (20 min) in 0.23 SSC at 56?C. After
blocking non-specific binding, sections were incubated
overnight at 4?C with an anti-digoxygenin antibody conju-
gated with HRP (Roche). For signal amplification, slides
were incubated with biotinyl tyramide (DAKO GenPoint Kit)
in the dark for 8 min. The probe was detected by incubation
with alkaline phosphatase-conjugated streptavidin (DAKO
GenPoint Kit) followed by 4-nitroblue tetrazolium chloride/
Values are expressed as means 6 SDs. The data were nor-
mally distributed, and group means were compared with the
Student two-tailed, unpaired t-test using Excel 5 for Apple
Macintosh (Microsoft Corporation, Redmond, WA, USA).
Role of STAT6 and IL13 in inflammation in IL9 Tg mice
We have recently shown that the development of lung path-
ology observed in inducible IL9 Tg mice was dependent on
IL13 expression (16). Blockade of IL13 completely abolished
lung inflammation, airway epithelial cell hypertrophy and mu-
cus overproduction. We obtained similar results after cross-
ingconstitutive IL9Tg mice
background. Total lung lavage cell counts in IL13?/? IL9 Tg
mice were significantly decreased compared with IL13+/+
IL9 Tg mice and were comparable to those from non-Tg
(WT) control mice (Fig. 1A). Differential cell counts on lung
lavage cells confirmed that the increased numbers of macro-
phages, lymphocytes and especially eosinophils observed
in the airways of IL13+/+ IL9 Tg mice were not found in
IL13?/? IL9 Tg mice (Fig. 1B). Histologic assessment of
lung tissue demonstrated that the accumulation of inflamma-
tory cells around blood vessels and airways and airway epi-
thelial cell hypertrophy with mucus overproduction seen in
lung sections of IL13+/+ IL9 Tg mice was absent in IL13?/?
IL9 Tg mice (Fig. 2).
To determine if IL13R signals through the STAT pathway
were required STAT6?/? mice were examined. The effects
of IL13 on the development of lung pathology were depen-
dent on STAT6 signaling since airway inflammation was
also totally abolished in IL9 Tg mice which had been
Fig. 2. Lung histology of conducting airways and parenchyma. Lung sections from IL9 transgene-negative (A) and transgene-positive (B and D)
mice were stained with alcian blue/periodic acid Schiff for the detection of mucin by light microscopy. Airway epithelial cells from IL9 transgene-
positive mice were hypertrophied (arrow) and stained strongly (purple) for mucin (B). This was not observed in lung sections from IL9 transgene-
positive mice after back-crossing onto IL13?/?(C) or STAT6 ?/?background (D) or from IL9 transgene-negative mice (A). Original
IL9 induces IL13 in airway epithelial cells3
by guest on June 12, 2013
back-crossed onto the STAT6?/? background (Fig. 1A and
B). No increased numbers of inflammatory cells were found
in lung lavage fluid from STAT6?/? IL9 Tg mice (Fig. 1A
and B). Further, lung sections did not show the epithelial cell
hypertrophy with mucus overproduction that was seen in
STAT6+/+ IL9 Tg mice (Fig. 2).
The role of lymphocytes in lung pathology of IL9 Tg mice
CD4+ T cells are a major source of IL13. To investigate if
T cells are important producers of IL13 in the lungs of IL9
Tg mice, we back-crossed these mice onto a RAG1?/?
background eliminating mature T and B cells. Surprisingly,
in the absence of Tand B cells, lung inflammation was dramat-
ically increased in RAG1?/? IL9 Tg mice compared with
RAG1+/+ IL9 Tg mice. Total lung lavage cell counts were
significantly increased in RAG1?/? IL9 Tg mice compared
with RAG1+/+ IL9 Tg mice. This increase was mostly due to
eosinophils (Fig. 3A and B). Also, examination of the lung
tissue revealed that epithelial cell hypertrophy with mucus
overproduction was still present in the absence of T and
B cells in RAG1?/? IL9 Tg mice (Fig. 4). We found that mice
on this background developed disease earlier than RAG+/+
mice. Therefore in the data shown, tissue was harvested at
6–7 weeks of age as opposed to the 8- to 10-week old mice
shown in Fig. 1.
ELISA performed on lung lavage fluid from RAG1?/? IL9
Tg mice detected high levels of IL13 (84.9 6 28.4 pg ml?1)
and IL5 (57.6 6 14.5 pg ml?1) while both cytokines were
below the detection limit (39 pg ml?1for IL13 and 40 pg ml?1
for IL5) in lung lavage fluid from RAG1+/+ IL9 Tg mice (data
RNase protection assay on total RNA from lung tissue
confirmed that mRNA levels of IL13, as well as IL5, were
clearly elevated in RAG1?/? IL9 Tg mice compared with
the mRNA levels in RAG1+/+ IL9 Tg mice (Fig. 5). The
mRNA levels of other cytokine genes like IL15, IL2, IL3, IFN-c,
the transgene IL9 and the housekeeping genes were at
comparable levels (Fig. 5). Therefore, in the absence of
T cells another source of IL13 must exist in the lungs of IL9
The role of eosinophils and mast cells in pathology of
IL9 Tg mice
Eosinophils and mast cells have been shown to be potent
producers of several different kinds of cytokines including
IL13. Both cell types are represented in the inflammatory re-
sponse caused by the IL9 transgene expression in the lung
(15). To investigate if one of these cell types might be in-
volved in IL13 production, we back-crossed IL9 Tg mice
with IL5?/? mice which are not able to generate mature
eosinophils, or mast cell-deficient (KitW/KitWv) mice. The lack
of eosinophils in IL5?/? IL9 Tg mice resulted in a decreased
total lung lavage cell count (Fig. 6A) compared with IL5+/+
IL9 Tg mice but the numbers of macrophages and lympho-
cytes were still elevated compared with non-Tg control mice
(Fig. 6B). Similar results were obtained from IL9 Tg mice
which were back-crossed onto the mast cell-deficient (KitW/
KitWv) background. Total lung lavage cell counts were de-
creased in KitW/KitWvIL9 Tg mice compared with WT IL9 Tg
mice (Fig. 6A). Surprisingly, the lack of mast cells resulted
in the reduction of eosinophils retrieved from the airways
(Fig. 6C) which could be an effect of the mixed genetic
background of mast cell-deficient (KitW/KitWv) mice or IL5
production by these cells. However, the absence of IL5 and
eosinophils (Fig. 7C) or mast cells (Fig. 7D) did not affect
the development of epithelial cell hypertrophy and mucus
overproduction in IL9 Tg mice as shown in lung sections
stained with alcian blue/periodic acid Schiff.
RNase protection assay on total lung RNA from IL5?/?
and IL5+/+ IL9 Tg mice revealed that levels of IL13 mRNA,
as well as those of all other cytokines mRNAs tested, were
Fig. 3. Inflammatory cells in lung lavage fluid from RAG1 ?/? mice.
IL9 transgene-negative (Tg–) or IL9 transgene-positive (IL9 Tg+) mice
were crossed with RAG1 ?/? mice. (A) Total lung lavage cell counts.
(B) Differential cell counts were derived from at least 200 cell counts
and are presented as total cell numbers. Data shown are the mean 6
SD of four mice per group. Significant differences between the total
cell numbers (A) or the number of eosinophils (B) compared with
that from IL9 transgene-positive (IL9 Tg+) mice are indicated with
an asterisk. Mice in these experiments were 6–7 weeks of age as
RAG ?/? mice develop disease earlier than RAG+/+ mice.
4IL9 induces IL13 in airway epithelial cells
by guest on June 12, 2013
similar (Fig. 8). Eosinophils were, therefore, excluded as
a major source of IL13 in IL9 Tg mice. Also, RNase protec-
tion assay using total lung RNA from KitW/KitWvIL9 Tg mice
did not show any difference in IL13 mRNA levels (data not
shown). This result suggests that mast cells are also not the
major source of IL13 in the lungs of IL9 Tg mice.
Role for IL5 in eosinophilia of RAG1?/? mice but not
RAG1?/? IL9 Tg mice had extremely high numbers of eosi-
nophils in their airways which correlated with the elevated
IL5 and IL13 levels. Since T cells and T cell-derived cytokines
might suppress expression of cytokines by eosinophils, we
investigated if the absence of T cells in RAG1?/? IL9 Tg
mice resulted in cytokine expression, in particular IL13, by
eosinophils. We back-crossed RAG1?/? IL9 Tg mice with
IL5?/? mice. Even in the absence of eosinophils, IL5?/?
RAG1?/? IL9 Tg mice still showed increased total lung
lavage cell counts (Fig. 9Aa) which was mostly due to
elevated numbers of macrophages compared with IL5?/?
RAG1?/? non-Tg mice (Fig. 9Ab). Further, lung sections
from IL5?/? RAG1?/? IL9 Tg mice still showed epithelial
cell hypertrophy with positive staining for mucus (Fig. 9B).
RNase protection assay using total lung RNA demon-
strated that even in the absence of eosinophils and T cells
Fig. 4. Histologic staining for mucin. Lung sections from IL9 transgene-negative (A and C) or IL9 transgene-positive (B and D) mice were stained
with alcian blue/periodic acid Schiff for light microscopy. Intense, positive (magenta) staining for mucin was observed not only in hypertrophied
airway epithelial cells from IL9 transgene-positive mice (B) but also in IL9 transgene-positive mice that had been back-crossed onto RAG1 ?/?
background (D). Mucin-positive epithelial cells were not detected in IL9 transgene-negative mice (A) or IL9 transgene-negative mice on RAG1 ?/?
background (C). Original magnification 3 300.
Fig. 5. Analysis of cytokine expression in total lung tissue. Levels of
various mRNAs encoding predominantly Th2 cytokines from individual
IL9 transgene-negative (lanes 1–4) and -positive (lanes 5–8) mice
were compared by RNase protection assay. Representative mice in
lanes 3, 4, 7 and 8 had been back-crossed onto RAG1 ?/? back-
ground. The P32-labeled multiprobe template set mCK-1b was used
as a size marker. Levels of L32 and GAPDH mRNAs encoding house-
keeping genes were used to compare loading of samples.
IL9 induces IL13 in airway epithelial cells5
by guest on June 12, 2013
in IL5?/? RAG1?/? IL9 Tg mice, there were still elevated
levels of IL13 mRNA detectable (Fig. 10A, lanes 1 and 2).
IL13 mRNA was not detected in lung tissue from IL5?/?
RAG?/? IL9 transgene-negative mice (Fig. 10A, lanes
3 and 4).
Cellular source of IL13 mRNA in lung sections from
IL5?/? RAG?/? IL9 Tg mice
Our results indicated that IL13 was still produced even in
the absence of major inflammatory cells like T cells, mast
cells and eosinophils in IL9 Tg mice. Therefore, we used
Fig. 6. Lavage cell counts from IL9 Tg mice on IL5 ?/? or mast cell-deficient background. IL9 transgene-negative (IL9 Tg–) or IL9 transgene-
positive (IL9 Tg+) mice had been crossed with IL5 ?/? or mast cell-deficient (MC ?/?) mice. (A) Total lung lavage cell counts. (B and C)
Differential lung lavage cell counts represented as total cell numbers were retrieved from at least 200 cell counts. Data are expressed as mean 6
SD. Significant differences in total cell numbers (A) or in the number of eosinophils (B and C) compared with that of IL9 Tg+ mice are indicated
with an asterisk.
Fig. 7. Histologic analysis of lung sections stained with alcian blue/periodic acid Schiff for the detection of mucin. Airway epithelial cells from IL9
transgene-positive mice (B) were hypertrophied and stained positive (magenta) for mucin which was also found in lung sections from IL9
transgene-positive mice on IL5?/?(C) or mast cell-deficient (D) background. Epithelial cell hypertrophy or positive staining for mucin was not
seen in lung sections from IL9 transgene-negative mice (A). Original magnification 3300.
6 IL9 induces IL13 in airway epithelial cells
by guest on June 12, 2013
lung tissue from IL5?/? RAG1?/? IL9 Tg mice for in situ hy-
bridization to detect the source of IL13 in these mice. Using
an IL13-specific anti-sense probe for hybridization, strong
positive staining was observed in the airway epithelium of
IL5?/? RAG1?/? IL9 Tg mice (Fig. 10Bc and d). In con-
trast, hybridization with the sense probe produced no stain-
ing in the lung tissue from IL5?/? RAG1?/? IL9 Tg mice
(Fig. 10Ba and b).
The expression of IL9 selectively within the lungs of Tg mice
resulted in pathologic changes similar to that seen in the
lungs of human asthmatics including lymphocytic and eosin-
ophilic airway inflammation, mucus hypersecretion and mast
cell hyperplasia (15). Previously, we had shown that various
Th2 cytokines including IL4, IL5 and IL13 were produced in
the lung in response to inducible IL9 expression (16). Al-
though one of them, IL13, was absolutely necessary for the
induction of lung inflammation and mucus production, the
combined action of all three cytokines was necessary for full
pathology to develop. In order to understand how IL9 in-
duces IL13-dependent lung inflammation, we tried to identify
potential target cells for the action of IL9 by focusing on lung
inflammation, mucus production and, in particular, the ex-
pression of IL13.
By crossing constitutive IL9-expressing mice onto the
IL13?/? background, we confirmed that IL13 is a critical
mediator for the development of pathology in the lung. The
effects of IL13 in IL9 Tg mice were STAT6 dependent co-
inciding with previously published data that this signaling
pathway is critical to the contribution of IL13 to phenotypic
features of allergic asthma (22, 23), just as in our Tg model.
A major source of IL13 is the CD4+ T cell of the Th2 sub-
type (24). The importance of these cells for the development
of lung pathology has been demonstrated using a murine
model of antigen-induced lung inflammation wherein the ab-
sence of CD4+ T cells prevents the development of asthmatic
features including eosinophilic infiltration of the lung and air-
way hyperreactivity (25). Assuming that in IL9 Tg mice
CD4+ T cells also play a central role in the immune response
through the production of cytokines, especially IL13, we in-
vestigated if T cells are important target cells for the action
of IL9. We crossed IL9 Tg mice with RAG1?/? mice which
lack mature T and B cells (19). Studies with RAG1?/? mice
had shown before that these mice failed to develop allergen-
induced lung inflammation, mucus hypersecretion and
airway hyperresponsiveness in murine models of asthma
(26). However, reconstitution of RAG1?/? mice with CD4
T cells alone could restore the asthmatic phenotype in re-
sponse to allergen challenge of the lung (26). Surprisingly,
the lack of T and B cells in IL9 Tg mice did not result in ab-
rogation of the asthma-like phenotype but quite the reverse
response occurred. Lung inflammation, characterized by ex-
treme high numbers of eosinophils, epithelial cell hypertro-
phy with mucus accumulation and Th2 cytokine production,
in particular IL13, was dramatically increased. We con-
cluded that IL13 production was independent of T and
B cells and was sufficient to induce a strong asthma-like phe-
notype in the lung. Results that IL13 could cause pathology
independent of T and B cells by administering IL13 intra-
nasally to RAG1?/? mice had been published previously (27).
Our data suggested to us that T cells had more of a regula-
tory function in our model by keeping lung inflammation un-
der control rather than being the major IL13 producer. We
predicted that some other cell type in the lung must there-
fore be responsible for producing the high levels of IL13 in
response to IL9 expression.
Two inflammatory cell types that are present in the lungs
of IL9 Tg mice at high numbers are eosinophils and mast
cells. For both it has been described that they are able to
express IL13 (28–30). The development of mast cells within
the airway epithelium is a unique feature of IL9 Tg mice (15).
However, from our previous studies on electron-microscopic
appearance and histamine release of these cells, we con-
cluded that mast cells are not degranulated or releasing
any mediators in our model. We speculated that the effect
of IL9 on mast cells was more related to differentiation and
proliferation in the airway epithelium rather than to mediator
release (15). As predicted, back-crossing of IL9 Tg mice
onto a mast cell-deficient background did not abolish lung
inflammation or mucus production even though eosinophilia
was decreased, suggesting no critical involvement of mast
cells in IL13 production. On the other hand, eosinophils
seemed to be a major candidate for IL13 production be-
cause increased numbers of eosinophils in the lungs of
RAG1?/? IL9 Tg mice correlated in our study with increased
IL13 expression levels. Evidence that IL9 might directly act
on eosinophils was provided before by the detection of
IL9Ra expression on human peripheral blood eosinophils
and by the IL9-stimulated induction of IL5Ra on eosinophils
(11). However, an effect of IL9 on lung eosinophils resulting
in IL13 expression could not be confirmed in our study since
Fig. 8. Cytokine expression in the lung. Total lung RNA isolated from
individual IL9 transgene-negative (lanes 2–7) and IL9 transgene-
positive (lanes 8–13) mice was used to detect levels of different
cytokine mRNAs by RNase protection assay. Mice representative for
lanes 5–7 and lanes 11–13 had been back-crossed onto the IL5 ?/?
background. The P32-labeled multiprobe template set mCK-1b was
used as a size marker. Levels of L32 and GAPDH mRNAs encoding
housekeeping genes were used to demonstrate equal loading of
IL9 induces IL13 in airway epithelial cells7
by guest on June 12, 2013
the lack of eosinophils in IL5?/? IL9 Tg mice did not de-
crease IL13 expression levels or lung pathology in particular
epithelial cell hypertrophy. Interestingly, we never detected
any basophils in the bronchial infiltrates of the IL9 Tg mice
and therefore we did not pursue any role for these cells in
the associated disease.
To exclude that the presence of T cells might have nega-
tively influenced potential IL13 production by eosinophils,
we investigated IL13 production and lung pathology in IL5/
RAG1 double-deficient IL9 Tg mice. However, also no role
for eosinophils in IL13 production in IL9 Tg mice could be
confirmed in the absence of T and B cells. RAG1/IL5 double-
deficient IL9 Tg mice had extremely high levels of IL13
expression and lung pathology except eosinophilia was
completely preserved. In conclusion, major inflammatory cell
types like T cells, mast cells and eosinophils could not be
verified as a major source for IL13 in IL9 Tg mice.
We finally decided to use in situ hybridization to detect the
IL13-expressing cells in IL5/RAG1 double-deficient IL9 Tg
mice. These mice had the highest IL13 expression levels
and we were able to identify airway epithelial cells as the
main IL13 producer. The airway epithelium is believed to
play an important role during asthmatic inflammation in the
lung by not only secreting mucus but also by producing a
variety of chemokines and cytokines. Several recent obser-
vations have demonstrated that IL9 might directly influence
airway epithelial cell function since IL9Ra had been detected
on human and murine epithelial cells (31, 32). An involve-
ment of IL9 in regulating mucus expression in the airways
had been shown in vivo and in vitro (10, 15, 16, 31). Further,
the IL9-induced release of chemokines in vitro by human
and murine lung epithelial cells was previously reported (13,
32). The IL9 stimulation of human bronchial epithelial cells
resulted in release of two T cell chemotactic factors, IL16
and regulated on activation, normal T cells expressed and
secreted (RANTES) (13). In mice, addition of IL9 to primary
lung epithelial cultures and cell lines induced the expres-
sion of the CC-chemokines eotaxin and MCP-5 (27). The direct
Fig. 9. (A) Inflammatory cells in IL9 transgene-negative (IL9 Tg –) and IL9 transgene-positive (IL9 Tg+) mice on the RAG1 ?/? and IL5 double-
deficient (RAG1?/? IL5?/?) background. (a) Total lung lavage cell counts. (b) Differential lung lavage cell counts presented as total cell
numbers. Data are expressed as mean 6 SD from four mice per group. Significant differences in total cell numbers (a) or the number of
macrophages (b) are indicated with an asterisk. (B) Histologic staining for mucin. Lung sections from IL9 transgene-positive (b) mice, but not IL9
transgene-negative mice (a), back-crossed onto RAG1 and IL5 double-deficient background still showed airway epithelial cell hypertrophy and
positive (magenta) staining for mucin (arrow). Original magnification 3300.
8 IL9 induces IL13 in airway epithelial cells
by guest on June 12, 2013
expression of IL13 by epithelial cells has not been demon-
strated so far. One study has suggested the production of
IL4 by immunohistochemistry (33). Therefore, our observa-
tion, that IL9 might induce directly or indirectly the expres-
sion of IL13 in airway epithelial cells represents a novel and
important finding. There is evidence in humans for expres-
sion of IL13 in airway epithelium using immunohistochemis-
try on biopsies of bronchial epithelium (34). However, IL13
expression has never been demonstrated via in situ hybrid-
ization to show cellular source of message for IL13. Further-
more, there is no previous evidence for epithelial production
of IL13 in a well-defined murine system. Production of IL13
by human alveolar macrophages has been reported (35).
We cannot exclude, therefore, that macrophages contribute
to IL13 production in addition to the epithelial cells de-
scribed herein. In conclusion, we have shown that IL13 pro-
duction in IL9 Tg mice occurred independent of major
inflammatory cell types like T cells, B cells, mast cells and
eosinophils. In IL5/RAG1 double-deficient IL9 Tg mice, we
have identified airway epithelial cells as the major location
of IL13 expression and lymphocytes as inhibitory of its ex-
pression. In future experiments, we will have to elucidate if
epithelial cells represent a direct target cell for IL9 in the
lung. However, our findings presented here might implicate
a new role for IL9 in airway epithelial cell function by induc-
ing cytokine production. In addition, our data add new inter-
esting aspects to the contribution of airway epithelial cells to
bronchial inflammation in asthma.
We would like to thank Z. Zhu for helpful discussions on the
manuscript and Fran Manzo for assistance with manuscript prepa-
ration. This work was supported by an award to R.A.F. from the
Sandler Program for Asthma Research and NIH HL 56389. R.A.F. is an
investigator of the Howard Hughes Medical Institute.
standard saline citrate
signal transduction of activated T cell
1 Lukacs, N. W., Strieter, R. M. and Kunkel, S. L. 1995. Leukocyte
infiltration in allergic airway inflammation. Am. J. Respir. Cell Mol.
2 Robinson, D. S., Hamid, Q., Ying, S. et al. 1992. Predominant TH2-
like bronchoalveolar T-lymphocyte population in atopic asthma.
N. Engl. J. Med. 326:298.
3 Nicolaides, N. C., Holroyd, K. J., Ewart, S. L. et al. 1997.
Interleukin 9: a candidate gene for asthma. Proc. Natl Acad. Sci.
4 Uyttenhove, C., Simpson, R. J. and Van Snick, J. 1988. Functional
and structural characterization of P40, a mouse glycoprotein with
T-cell growth factor activity. Proc. Natl Acad. Sci. USA 85:6934.
5 Hultner, L. and Moeller, J. 1990. Mast cell growth-enhancing
activity (MEA) stimulates interleukin 6 production in a mouse bone
marrow-derived mast cell line and a malignant subline. Exp.
6 Williams, D. E., Morrissey, P. J., Mochizuki, D. Y. et al. 1990. T-cell
growth factor P40 promotes the proliferation of myeloid cell lines
and enhances erythroid burst formation by normal murine bone
marrow cells in vitro. Blood 76:906.
7 Dugas, B., Renauld, J. C., Pene, J. et al. 1993. Interleukin-9
potentiates the interleukin-4-induced immunoglobulin (IgG, IgM
and IgE) production by normal human B lymphocytes. Eur. J.
8 Donahue, R. E., Yang, Y. C. and Clark, S. C. 1990. Human P40
T-cell growth factor (interleukin-9) supports erythroid colony
formation. Blood 75:2271.
9 Abdelilah, S., Latifa, K., Esra, N. et al. 2001. Functional expression
of IL9 receptor by human neutrophils from asthmatic donors: role
in IL8 release. J. Immunol. 166:2768.
10 Longphre, M., Li, D., Gallup, M. et al. 1999. Allergen-induced IL9
directly stimulates mucin transcription in respiratory epithelial
cells. J. Clin. Invest. 104:1375.
Fig. 10. (A) Analysis of cytokine expression in lung tissue. Levels of
different cytokine mRNAs from individual IL9 transgene-positive
(lanes 1 and 2) and -negative (lanes 3 and 4) mice back-crossed
onto IL5 ?/? and RAG1 ?/? background and IL9 transgene-positive,
IL5 ?/? , Rag+/? (lanes 5 and 6) and finally IL9 transgene-negative,
IL5 ?/? Rag+/? (lanes 7 and 8) were compared by the RNAse
protection assay. Lane 9 is the RNA control from the kit. The
P32-labeled multiprobe template mCK-1b was used as a size marker.
The bands corresponding to IL9 (the transgene) and IL13 are indicated
by the arrows. (B) Expression of IL13 in airway epithelial cells. In situ
hybridization of lung tissue from IL5 ?/? RAG1 ?/? IL9 Tg mice with
an IL13-specific sense probe as a control did not show any staining
(a and b). In contrast, using an IL13-specific anti-sense probe for
hybridization with the same tissue resulted in strong positive staining
of the airway epithelial cells (c and d arrows). Original magnification 3
150 (a and c) and 3300 (b and d).
IL9 induces IL13 in airway epithelial cells9
by guest on June 12, 2013
11 Gounni, A. S., Gregory, B., Nutku, E. et al. 2000. Interleukin-9 Download full-text
enhances interleukin-5 receptor expression, differentiation, and
survival of human eosinophils. Blood 96:2163.
12 Louahed, J., Zhou, Y., Maloy, W. L. et al. 2001. Interleukin 9
promotes influx and local maturation of eosinophils. Blood
13 Little, F. F., Cruikshank, W. W. and Center, D. M. 2001. IL9
stimulates release of chemotactic factors from human bronchial
epithelial cells. Am. J. Respir. Cell Mol. Biol. 25:347.
14 Townsend, J. M., Fallon, G. P., Matthews, J. D., Smith, P., Jolin,
E. H. and McKenzie, N. A. 2000. IL9-deficient mice establish
fundamental roles for IL9 in pulmonary mastocytosis and goblet
cell hyperplasia but not T cell development. Immunity 13:573.
15 Temann, U. A., Geba, G. P., Rankin, J. A. and Flavell, R. A. 1998.
Expression of interleukin 9 in the lungs of Tg mice causes airway
inflammation, mast cell hyperplasia, and bronchial hyperrespon-
siveness. J. Exp. Med. 188:1307.
16 Temann, U. A., Ray, P. and Flavell, R. A. 2002. Pulmonary
overexpression of IL9 induces Th2 cytokine expression, leading to
immune pathology. J. Clin. Invest. 109:29.
17 Kaplan, M. H., Schindler, U., Smiley, S. T. and Grusby, M. J. 1996.
Stat6 is required for mediating responses to IL4 and for
development of Th2 cells. Immunity 4:313.
18 Kopf, M., Brombacher, F., Hodgkin, P. D. et al. 1996. IL5-deficient
mice have a developmental defect in CD5+ B-1 cells and lack
eosinophilia but have normal antibody and cytotoxic T cell
responses. Immunity 4:15.
19 Mombaerts, P., Iacomini, J., Johnson, R. S., Herrup, K., Tonegawa,
S. and Papaioannou, V. E. 1992. RAG-1-deficient mice have no
mature B and T lymphocytes. Cell 68:869.
20 McKenzie, G. J., Emson, C. L., Bell, S. E. et al. 1998. Impaired
development of Th2 cells in IL13-deficient mice. Immunity 9:423.
21 Kitamura, Y., Go, S. and Hatanaka, K. 1978. Decrease of mast
cells in W/Wv mice and their increase by bone marrow trans-
plantation. Blood 52:447.
22 Kuperman, D., Schofield, B., Wills-Karp, M. and Grusby, M. J.
1998. Signal transducer and activator of transcription factor 6
(Stat6)-deficient mice are protected from antigen-induced airway
hyperresponsiveness and mucus production. J. Exp. Med.
23 Akimoto, T., Numata, F., Tamura, M. et al. 1998. Abrogation of
bronchial eosinophilic inflammation and airway hyperreactivity in
signal transducers and activators of transcription (STAT)6-
deficient mice. J. Exp. Med. 187:1537.
24 de Vries, J. E. 1998. The role of IL13 and its receptor in allergy and
inflammatory responses. J. Allergy Clin. Immunol. 102:165.
25 Gavett, S. H., Chen, X., Finkelman, F. and Wills-Karp, M. 1994.
Depletion of murine CD4+ T lymphocytes prevents antigen-
induced airway hyperreactivity and pulmonary eosinophilia.
Am. J. Respir. Cell Mol. Biol. 10:587.
26 Corry, D. B., Grunig, G., Hadeiba, H. et al. 1998. Requirements for
allergen-induced airway hyperreactivity in T and B cell-deficient
mice. Mol. Med. 4:344.
27 Grunig, G., Warnock, M., Wakil, A. E. et al. 1998. Requirement for
IL13 independently of IL4 in experimental asthma. Science
28 Krzesicki, R. F., Winterrowd, G. E., Brashler, J. R. et al. 1997.
Identification of cytokine and adhesion molecule mRNA in murine
lung tissue and isolated T cells and eosinophils by semi-
quantitative reverse transcriptase-polymerase chain reaction.
Am. J. Respir. Cell Mol. Biol. 16:693.
29 Burd, P. R., Thompson, W. C., Max, E. E. and Mills, F. C. 1995.
Activated mast cells produce interleukin 13. J. Exp. Med.
30 Schmid-Grendelmeier, P., Altznauer, F., Fischer, B. et al. 2002.
Eosinophils express functional IL13 in eosinophilic inflammatory
diseases. J. Immunol. 169:1021.
31 Louahed, J., Toda, M., Jen, J. et al. 2000. Interleukin-9
upregulates mucus expression in the airways. Am. J. Respir. Cell
Mol. Biol. 22:649.
32 Dong, Q., Louahed, J., Vink, A. et al. 1999. IL9 induces chemokine
expression in lung epithelial cells and baseline airway eosinophilia
in Tg mice. Eur. J. Immunol. 29:2130.
33 Ando, M., Miyazaki, E., Fukami, T., Kumamoto, T. and Tsuda, T.
1999. Interleukin-4-producing cells in idiopathic pulmonary
fibrosis: an immunohistochemical study. Respirology 4:383.
34 Pathmanathan, S., Krishna, M. T., Blomberg, A. et al. 2003.
Repeated daily exposure to 2 ppm nitrogen dioxide upregulates
the expression of IL5, IL10, IL13, and ICAM-1 in the bronchial
epithelium of healthy human airways. Occup. Environ. Med.
35 Hancock, A., Armstrong, L., Gama, R. and Millar, A. 1998.
Production of interleukin 13 by alveolar macrophages from normal
and fibrotic lung. Am. J. Respir. Cell Mol. Biol. 18:60.
10 IL9 induces IL13 in airway epithelial cells
by guest on June 12, 2013