IL-33eInduced Atopic DermatitiseLike Inflammation in Mice Is Mediated by Group 2 Innate Lymphoid Cells in Concert with Basophils

Prévia do material em texto

IL-33eInduced Atopic Derm
Inflammation in Mice Is M
Innate Lymphoid Cells in Co
Yasutomo Imai1, Koubun Yasuda2, Makoto Nagai1, Mino
and Kiyofumi Yamanishi1
IL-33 is a proinflammatory cytokine that plays a pivot
expressing IL-33 driven by a keratin-14 promoter (IL33t
spontaneously with the activation of group 2 innate lymp
effector cells, such as T helper type 2 cells, ILC2s, an
induced by IL-33. To address the question, we examine
L3
n
m
,
d
a
at
in
id
and/or rhinitis (Weidinger et al., 2018). In developed coun- released rapidly as an endogenous danger signal or alarmin
eous
are
pha chain)
cells, ba-
2s (previ-
010) are a
ce markers
crophage,
þ Thy1.2
ILC2s lack
See related commentary on pg 2077 ORIGINAL ARTICLE
accepted manuscript published online 21 May 2019; corrected proof
IL-33 binds its receptor ST2 (IL-33 receptor al
(Schmitz et al., 2005), which is expressed on Th2
sophils, mast cells, and ILC2s. Of these cells, ILC
ously termed natural helper cells) (Moro et al., 2
population of lineage-negative (Line; lacking surfa
for T cell, B cell, natural killer cell, monocyte, ma
neutrophil, and eosinophil lineages) Sca-1
(CD90.2)þ lymphoid cells. In contrast to Th2 cells,
College of Medicine, 1-1, Mukogawa-cho, Nishinomiya, Hyogo, 663-8501
Japan. E-mail: kyamanis@hyo-med.ac.jp
Abbreviations: Ab, antibody; AD, atopic dermatitis; CFSE, carboxy-
fluorescein succinimidyl ester; DT, diphtheria toxin; ILC2, group 2 innate
lymphoid cells; mMCP-8, mouse mast cell protease 8; Rag2, recombination-
activating gene 2; RORa, RAR-related orphan receptor alpha; Th2, T helper
type 2; TRECK, toxin receptoremediated conditional cell knockout; TSLP,
thymic stromal lymphopoietin
Received 21 February 2019; revised 13 April 2019; accepted 18 April 2019;
4 (Stott et al., 2013). Thus, IL-33 is also involved in cutan
barrier function and neuroimmune interactions, which
impaired in AD (Weidinger et al., 2018).
and 4Laboratory for Cytokine Regulation, RIKEN Center for Integrative
Medical Sciences (IMS), Yokohama, Kanagawa, Japan
Correspondence: Kiyofumi Yamanishi, Department of Dermatology, Hyogo
tries, a lifetime prevalence of AD is estimated to reach
15e20% (Deckers et al., 2012). T helper type 2 (Th2) cells are
believed to mediate inflammation in AD with the induction
of cytokines IL-4 and IL-13, which are responsible for clinical
manifestations of AD. On another front, recent studies on AD
have highlighted a pathogenic role of keratinocyte-derived
cytokines, such as IL-33, thymic stromal lymphopoietin
(TSLP), and IL-25 (IL-17E), that elicit Th2 cytokine responses
by activating group-2 innate lymphoid cells (ILC2s).
to alert and activate the innate immune system (Moussion
et al., 2008). IL-33 has been reported to participate in the
disease processes of allergic disorders such as allergic
conjunctivitis (Matsuba-Kitamura et al., 2010; Matsuda et al.,
2009), asthma (Kondo et al., 2008), and allergic rhinitis
(Haenuki et al., 2012). In AD, IL-33 is abundant in lesional
epidermal keratinocytes (Savinko et al., 2012), and we
demonstrated that IL-33 upregulated in keratinocytes induces
AD-like inflammation in a transgenic mouse line (IL33tg)
expressing mouse IL-33 driven by a keratin-14 promoter (Imai
et al., 2013). In addition, claudin-1 is downregulated in
IL33tg mice (Ryu et al., 2018), and filaggrin expression is
reduced by IL-33 in the skin (Seltmann et al., 2015). IL-31 is
released by leucocytes and directly stimulates neurons to
mediate itch. IL-33 is able to induce IL-31 in synergy with IL-
1Department of Dermatology, Hyogo College of Medicine, Nishinomiya,
Japan; 2Department of Immunology, Hyogo College of Medicine,
Nishinomiya, Japan; 3Division of Molecular Pathology, Research Institute
for Biomedical Science, Tokyo University of Science, Noda, Chiba, Japan;
cells. AD-like inflammation still developed in Rag2KO I
were depleted in IL33tg mice via bone marrow transpla
alphaedeficient mice, the development of AD-like infla
were accumulated in the inflamed skin of IL33tg mice
ditional depletion of basophils using anti-FcεRIa antibo
basophil-depleted IL33tg skins, ILC2s were decreased,
CCL5 were reduced. From these results, we suggest th
innate immune responses that are mediated by ILC2s
Journal of Investigative Dermatology (2019) 139, 2185e2194; doi:10.1016/j.j
INTRODUCTION
Atopic dermatitis (AD) is a common inflammatory skin dis-
ease characterized by chronic eczema with severe itching.
AD frequently occurs with atopic diseases such as asthma
published online 8 August 2019
ª 2019 The Authors. Published by Elsevier, Inc. on behalf of the Society for Inv
atitiseLike
ediated by Group 2
ncert with Basophils
ri Kusakabe1, Masato Kubo3,4, Kenji Nakanishi2
al role in allergic disorders. In a transgenic mouse
g), atopic dermatitis (AD)-like inflammation develops
hoid cells (ILC2s). However, it remains unknown how
d basophils, contribute to the inflammatory process
d the phenotype of IL33tg mice lacking each of these
3tg mice lacking T and B cells; in contrast, when ILC2s
tation from ILC2-lacking, RAR-related orphan receptor
mation was almost completely suppressed. Basophils
and AD-like inflammation was alleviated by the con-
ies or a Bas-TRECK transgenic mouse system. In these
nd cytokines and chemokines such as IL-5, IL-13, and
IL-33einduced AD-like inflammation is dependent on
concert with basophils.
.2019.04.016
IL-33 is a proinflammatory cytokine in the IL-1 family
(Schmitz et al., 2005). In the epithelial or endothelial cells,
IL-33 is constitutively expressed and is normally sequestered
in the cell nucleus. Upon cellular damage or stress, IL-33 is
TCRs but are able to produce large amounts of Th2 cytokines
estigative Dermatology. www.jidonline.org 2185
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
2186
IL-5 and IL-13 in response to IL-33 without antigen stimula-
tion (Moro et al., 2010). In IL33tg skin lesions, ILC2s are
accumulated and activated to produce IL-5 and IL-13,
possibly in response to the IL-33 upregulated in keratino-
cytes (Imai et al., 2013). ILC2s expressing CD25 and ST2 are
present in healthy human skin and are enriched in AD skin
lesions (Kim et al., 2013; Mashiko et al., 2017; Salimi et al.,
2013). It must be noted that an ILC2 population isolated from
human AD lesions is activated to produce these cytokines
predominantly by IL-33, not by TSLP or IL-25 (Salimi et al.,
2013). These findings are highly suggestive of the patho-
genic role of the IL-33eresponsive ILC2s in AD.
Basophils, which also express ST2 on the cell surface, are
postulated to participate in Th2 skewing in allergic inflam-
mation by producing IL-4, IL-6, IL-13, and/or histamine
(Nakashima et al., 2018). Basophils from patients with
allergic asthma are able to produce IL-4 and IL-13 in
response to IL-33 (Salter et al., 2016). Some reports have
shown that basophils are accumulated in the lesional skin of
AD (Plager et al., 2006), in which IL-4eproducing basophils
are increased (Ito et al., 2011; Kim et al., 2014). The fre-
quency of basophils in the skin has been shown to correlate
with that of ILC2s in patients with AD (Mashiko et al., 2017).
Furthermore, ILC2s are activated by basophils via IL-4 in a
mouse asthma model (Motomura et al., 2014) and in MC903
(calcipotriol)-induced AD-like inflammation in mice (Kim
et al., 2014).
Regardless of these recent studies focusing on ILC2s and/or
basophils, it is not fully understood how these immune cells
and Th2 cells contribute to the disease process of AD in as-
sociation with IL-33. In this study, we addressed the question
by depleting each of these cells in IL33tg mice and defining
their roles in AD-like cutaneous inflammation induced by
IL-33.
RESULTS
AD-like inflammation occurs in IL33tg mice lacking Th2 cells
It was likely that Th2 cells mediate AD-likeinflammation
developed spontaneously in IL33tg mice because Th2 cells
express IL-33 receptor a (ST2), and IL-33 stimulates Th2
cells to produce IL-5 and IL-13 (Kondo et al., 2008). To
examine the role of Th2 cells in cutaneous inflammation,
we crossed IL33tg mice with recombination-activating
gene 2 (Rag2) knockout (KO) mice to generate Rag2KO
IL33tg mice that lacked Th2 cells. However, as shown in
Figure 1a, AD-like inflammation still developed spontane-
ously between 6 and 8 weeks of age in all Rag2KO IL33tg
mice and remained thereafter, as in IL33tg mice (Imai
et al., 2013). The expressions of the IL-5 and IL-13 genes
were upregulated in Rag2KO IL33tg mice as well as in
IL33tg mice, whereas the expressions of the TSLP and IL-25
genes were unaltered (Figure 1b). The protein levels of
these cytokines and chemokines, MIP-3a/CCL2, RANTES/
CCL5, and Eotaxin 1/CCL11, were still induced in Rag2KO
IL33tg skin, as in IL33tg skin (Imai et al., 2013) (Figure 1c).
The concentrations of psoriasis-related cytokines, such as
IL-23p40, IL-12p70, tumor necrosis factor-a, and IL-17A,
were unaltered (Figure 1c). These results suggest that AD-
like inflammation that developed in IL33tg mice is less
dependent on Th2 cells.
Journal of Investigative Dermatology (2019), Volume 139
ILC2s are essential to the development of AD-like
inflammation in IL33tg mice
ILC2s are markedly induced in the lesional skin of IL33tg
mice (Imai et al., 2013, 2017); therefore, we next assessed
the role of ILC2s in AD-like cutaneous inflammation devel-
oped in IL33tg mice. B6.C3(Cg)-Rorasg/J (RAR-related orphan
receptor alpha [RORa]edeficient) mice lack a population of
ILC2s because RORa is a transcription factor that is critical
for ILC2 development (Halim et al., 2012). To deplete ILC2s
from mice, bone marrow prepared from RORa-deficient mice
was transplanted into irradiated IL33tg mice; 15 weeks later,
ILC2 populations were assessed by draining lymph nodes, as
reported previously (Salimi et al., 2013). ILC2s were suc-
cessfully lost (Figure 2a), and AD-like dermatitis was
improved in ILC2-lacking, RORa-deficient bone
marrowetransplanted IL33tg mice (Figure 2b and c). Protein
array analysis revealed that IL-5, IL-13, MIP-3a/CCL2,
RANTES/CCL5, or Eotaxin 1/CCL11, which are derived
mainly from ILC2s, were almost at normal levels in the
mouse skin (Figure 2d). These results suggest that the devel-
opment of AD-like cutaneous inflammation in IL33tg mice
depends on the presence of ILC2s.
Basophils activate ILC2s via IL-4
We then focused on basophils because IL-33 stimulates ba-
sophils to produce IL-4 (Kondo et al., 2008). Mouse mast cell
protease 8 (mMCP-8) is specifically expressed by basophils,
but not by mast cells (Imai et al., 2017; Tsutsui et al., 2017).
The gene expressions of mMCP-8 and IL-4 were significantly
increased in the skin of IL33tg mice (Figure 3a), and this
prompted us to examine basophils in the skin. Immunofluo-
rescence showed that mMCP-8þ basophils accumulated in
the inflamed skin of IL33tg mice (Figure 3b). Flow cytometry
revealed that basophils from the skin were intracellular IL-4-
positive (Supplementary Figure S1; Figure 3c) and that ILC2s
from IL33tg mouse skin expressed IL-4Ra (Figure 3d). These
results implied that ILC2s might be activated in response to
IL-4 derived from basophils in IL33tg mice. To assess the
possibility, carboxyfluorescein succinimidyl ester (CFSE)-
labeled sort-purified ILC2s were cocultured with bone
marrowederived basophils from wild-type or IL-4edeficient
mice. As measured by CFSE dilution, the proliferation of
ILC2s was induced by IL-33estimulated wild-type basophils,
but not by basophils from IL-4edeficient mice (Figure 3e).
Furthermore, lymph node cells from Rag2KO IL33tg mice,
which are enriched in IL-5eproducing ILC2s (Supplementary
Figure S2), were cultured with wild-type or IL-4edeficient
mouse basophils (ratio 4:1) for 48 hours, and IL-5 production
was examined by ELISA. As shown in Figure 3f, basophils
lacking IL-4 could not promote IL-5 production by the lymph
node cells. These results suggest that basophil-derived IL-4
activates ILC2s to proliferate and produce IL-5 in IL33tg mice.
AD-like inflammation is suppressed in basophil-depleted
IL33tg skin
To assess the role of basophils in AD-like inflammation in
IL33tg mice, basophils were depleted using anti-FcεRI anti-
body (Ab; MAR-1). FcεRI is present not only on basophils but
also on mast cells. MAR-1 is used exclusively for the deple-
tion of basophils because the Ab does not affect mast cells
(Sokol et al., 2008). Mice were injected intraperitoneally with
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
10 mg of MAR-1 or isotype control Ab twice a day for 3
consecutive days every week, then killed at 21 days. Immu-
nofluorescence showed that mMCP-8þ basophils were
depleted in MAR-1etreated Rag2KO IL33tg mice (Figure 4a)
without the depletion of mast cells (Supplementary
Figure 1. AD-like inflammation developed in Rag2KO IL33tg mice. (a) H&E sta
inflammation is not abolished in Rag2KO IL33tg skin. Bars, 100 mm. Panels are re
gene expressions of Th2 cytokines in the skin of 20-week-old WT, Rag2KO, IL33
significantly increased in Rag2KO IL33tg mouse skin, as well as in IL33tg mouse s
(c) Cutaneous cytokine and chemokine profiles of 20-week-old Rag2KO and Ra
measured as described in Materials and Methods. IL-5, IL-13, MIP-3a/CCL2, RA
lesional skin. Data are expressed as means � SEM (n¼9), *P < 0.05 (two-tailed
transgenic; KO, knockout; Rag2, recombination-activating gene 2; SEM, standard
wild-type.
Figure S3a), which were as abundant as in the lesional skin
of IL33tg mice (Imai et al., 2013). Moreover, the depletion of
basophils was associated with the reduction of ILC2 re-
sponses in the skin (Supplementary Figure S4a and b) and
with the alleviation of AD-like cutaneous inflammation
ining of Rag2KO and Rag2KO IL33tg mouse skin. Note that cutaneous
presentative of at least seven mice and two independent experiments. (b) The
tg, and Rag2KO IL33tg mice. The expressions of Il5, Il13, and Il33 were
kin. Data are expressed as means � SEM (n ¼ 6), *P < 0.05 (two-tailed t test).
g2KO IL33tg mice. The concentrations of cytokines and chemokines were
NTES/CCL5, and Eotaxin 1/CCL11 were induced in Rag2KO IL33tg mouse
t test). AD, atopic dermatitis; H&E, hematoxylin and eosin; IL33tg, IL-33
error of the mean; Th2, T helper type 2; TNF-a, tumor necrosis factor-a; WT,
www.jidonline.org 2187
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
2188
(Figure 4b). Th2 cytokines and chemokines such as IL-5, IL-
13, CCL2, and CCL5 were significantly reduced in the skin by
MAR-1 treatment (Figure 4c).
To confirm the role of basophils in vivo, we also used Bas-
TRECK mice (Sawaguchi et al., 2012), in which basophils
expressing the diphtheria toxin (DT) receptor are specifically
depleted by DT treatment using a toxin receptoremediated
conditional cell knockout (TRECK) system. IL33tg mice
were crossed with Bas-TRECK mice to generate IL33
Figure 2. AD-like inflammation in IL33tg mice was abrogated by RORa-deficien
(Rorawt/sg) and ILC2-deficient (Rorasg/sg) bone marrowetransplanted IL33tg mice. C
indicate the proportion of ILC2s, which were undetectable in ILC2-deficient (Rora
mice and two independent experiments. (b) Cutaneous manifestations of control (
(24e26 weeks old). Note that skin lesions were improved in ILC2-deficient (Roras
from at least eight mice are shown. Similar results were obtained in two indepen
(Rorasg/sg) bone marrowetransplanted IL33tg mice. Note that epidermal thickenin
marrowetransplanted IL33tg mice. Bars, 100 mm. Panels are representative of at l
chemokine profiles in control (Rorawt/sg) and ILC2-deficient (Rorasg/sg) bone marrow
0.05 (two-tailed t test). AD, atopic dermatitis; H&E, hematoxylin and eosin; IL33tgRORa, RAR-related orphan receptor alpha; SEM, standard error of the mean.
Journal of Investigative Dermatology (2019), Volume 139
Bas-TRECK Tg mice, and basophils were selectively removed
using DT. Immunofluorescence showed that basophils were
depleted by DT (Figure 5a) and that mast cells were unaf-
fected, as reported previously (Sawaguchi et al., 2012)
(Supplementary Figure S3b). Cutaneous inflammation was
milder (Figure 5b), and the concentrations of Th2 cytokines
and chemokines were lower, in the skin of DT-treated IL33
Bas-TRECK Tg mice than in the skin of DT-treated IL33tg
mice (Figure 5c). These results suggest that basophils are
t bone marrow transplantation. (a) Flow cytometry of ILC2s from control
ervical lymph node cells were gated on Line cells for ILC2s. The numbers
sg/sg) bone marrowetransplanted IL33tg mice. Data are representative of three
Rorawt/sg) and ILC2-deficient (Rorasg/sg) bone marrowetransplanted IL33tg mice
g/sg) bone marrowetransplanted IL33tg mice. Representative photographs taken
dent experiments. (c) H&E staining of control (Rorawt/sg) and ILC2-deficient
g and inflammatory infiltrates were restored in ILC2-deficient (Rorasg/sg) bone
east four mice and two independent experiments. (d) Cutaneous cytokine and
etransplanted IL33tg mice. Data are expressed as means� SEM (n¼ 7e9), *P<
, IL-33 transgenic; ILC2, group 2 innate lymphoid cells; Lin, lineage marker;
Figure 3. Basophils accumulated in the inflamed skin of IL33tg mice; basophil-derived IL-4 is necessary for ILC2 proliferation. (a) The expressions of Mcpt8
and Il4 in 20-week-old WT, Rag2KO, IL33tg, and Rag2KO IL33tg mouse skin. Data are expressed as means � SEM (n ¼ 6), *P < 0.05 (two-tailed t test). (b)
Immunofluorescence of mMCP-8 in WT, Rag2KO, IL33tg, and Rag2KO IL33tg mouse skin. Bars, 100 mm. Data are representative of at least five mice. (c)
Intracellular cytokine detection by flow cytometry of cells from 20-week-old IL33tg mouse skin. Cells stimulated with phorbol 12-myristate 13-acetate/
ionomycin were gated on the skin DX5þ basophils. The numbers indicate the proportion of basophils producing IL-4. (d) Expression of IL-4Ra (black line) on
ILC2s from IL33tg mice. Cells were gated on the skin ST2þ ILC2s. (e) CFSE dilution of CFSE-labeled sort-purified ILC2s cultured with WT (solid gray histogram) or
IL-4edeficient mice (black line) basophils. Cells cultured with IL-2, IL-3, and IL-33 were gated on the ILC2 fraction. The numbers indicate the proportion of
proliferated (CFSE-diluted) ILC2s. (f) IL-5 production by lymph node cells from Rag2 KO IL33tg mice (ILC2s) cultured with WTor IL-4edeficient mouse basophils
(ratio 4:1). Representative data from two independent experiments (bef). AD, atopic dermatitis; CFSE, carboxyfluorescein succinimidyl ester; IL33tg, IL-33
transgenic; ILC2, group 2 innate lymphoid cells; KO, knockout; mMCP-8, mouse mast cell protease 8; Rag2, recombination-activating gene 2; SEM, standard
error of the mean; WT, wild-type.
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
www.jidonline.org 2189
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
2190
involved in the exacerbation of ILC2-mediated AD-like
inflammation in IL33tg mice.
DISCUSSION
Immune responses are based on two major branches, adap-
tive and innate immunity in disease processes. Previously, we
reported that an antigen in the presence of IL-33 stimulates
Th2-polarized CD4þ T cells to produce IL-5 and IL-13 (Kondo
et al., 2008). In an epicutaneously ovalbumin-sensitized AD
mouse model, IL-33 and ST2 expressions are upregulated in
the skin (Savinko et al., 2012), whereas an MC903 (calci-
potriol)-induced AD model, which is known as an innate
immune AD model, is dependent on IL-33 at a later stage of
Figure 4. AD-like inflammation was alleviated by basophil depletion using
MAR-1 Ab. (a) Immunofluorescence of mMCP-8 in control IgG1 or MAR-1
Ab-injected Rag2KO IL33tg mouse skin (8 week old). Note that mMCP-8þ
basophils were depleted by MAR-1. Bars, 100 mm. Data are representative of
at least four mice and two independent experiments. (b) H&E staining of
control IgG1 or MAR-1 Ab-injected Rag2KO IL33tg mouse skin. Note that
epidermal thickening and inflammatory infiltrates were improved by MAR-1.
Bars, 100 mm. Panels are representative of at least four mice and two
independent experiments. (c) Cutaneous cytokine and chemokine profiles in
control IgG1 and MAR-1 Ab-injected Rag2KO IL33tg mice. Data are
expressed as means � SEM (n ¼ 4e5), *P < 0.05 (two-tailed t test). Ab,
antibody; AD, atopic dermatitis; H&E, hematoxylin and eosin; IL33tg, IL-33
transgenic; ILC2, group 2 innate lymphoid cells; KO, knockout; mMCP-8,
mouse mast cell protease 8; Rag2, recombination-activating gene 2; SEM,
standard error of the mean.
Journal of Investigative Dermatology (2019), Volume 139
inflammation (Li et al., 2017). These reports suggest that IL-33
is involved in both adaptive and innate immune responses in
the inflammation of AD. However, as shown in this study,
AD-like inflammation developed in IL33tg mice was not
abrogated in Rag2KO IL33tg mice that lack both Tand B cells
(Figure 1). Hence, innate immunity—rather than adaptive
immunity—may contribute predominantly to the IL-33-
induced AD-like inflammation.
Furthermore, we demonstrated that the inflammation is
almost abolished by ILC2 depletion (Figure 2). Thus, the IL-
33einduced dermatitis is dependent on ILC2s, which are
actually increased in human AD skin lesions (Kim et al., 2013;
Mashiko et al., 2017; Salimi et al., 2013). A RORa-dependent
Figure 5. AD-like inflammation was suppressed in DT-treated IL33 Bas-
TRECK Tg mice. (a) Immunofluorescence of mMCP-8 in DT-treated IL33tg
and IL33 Bas-TRECK Tg mouse skin (8 week old). Note that mMCP-8þ
basophils were depleted in DT-treated IL33 Bas-TRECK Tg mice. Bars, 100
mm. Data are representative of at least four mice and two independent
experiments. (b) H&E staining of DT-treated IL33tg and IL33 Bas-TRECK Tg
mouse skin. Note that epidermal thickening and inflammatory infiltrates were
almost absent in DT-treated IL33 Bas-TRECK Tg mice. Bars, 100 mm. Panels
are representative of at least four mice and two independent experiments. (c)
Cutaneous cytokine and chemokine profiles in DT-treated IL33tg and IL33
Bas-TRECK Tg mouse skin. Data are expressed as means � SEM (n ¼ 3e7), *P
< 0.05 (two-tailed t test). AD, atopic dermatitis; DT, diphtheria toxin; IL33tg,
IL-33 transgenic; mMCP-8, mouse mast cell protease 8; SEM, standard error
of the mean; TRECK, toxin receptoremediated conditional cell knockout.
A
on
era
itis
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
ILC2-depletion model may affect cytokine production from
Figure 6. A possible mechanism of IL-33-induced cutaneous inflammation in
mast cells in the skin, independently on Th2 cells. IL-33 stimulates the producti
lesions. IL-33 also stimulates basophils to produce IL-4, which promotes prolif
with down-regulation of claudin-1 or filaggrin expression. AD, atopic dermat
group 3 innate lymphoid cells and T helper type 17 cells (Lo
et al., 2016), but a role of these cells could not be excluded
in this study. IgE to any allergens are not detected in 20% of
patients with AD (Cox and Calderon, 2016). An egg and milk
exclusion diet is not effective for patients with AD (Neild et al.,
1986), and allergen immunotherapy, which works well for
allergic rhinitis and asthma, is controversial in AD (Cox and
Calderon, 2016). In contrast to these therapies for acquired
immunity, etokimab, a humanized IgG1 Ab that simply targets
IL-33, proved efficacious for AD in a phase 2a trial (Londei
et al., 2017; Weidinger et al., 2018). We speculate that AD
may have an aspect of antigen-independent innate inflam-
mation in which IL-33eactivated ILC2s play a centralrole in
the pathophysiology of the disease.
In this study, we found that basophils were accumulated in
the inflamed skin of IL33tg mice (Figure 3aec). This is compa-
rable to the reports that indicate basophils are increased in AD
skin lesions (Ito et al., 2011; Kim et al., 2014; Mashiko et al.,
2017). IL-33 is able to induce the expansion of basophils
in vivo (Schneider et al., 2009). In addition, the ablation of ba-
sophils attenuates AD-like inflammation in IL33tg mice
(Figures 4 and 5). These findings suggest that basophils exacer-
bate IL-33einduced dermatitis. Previously, we reported that
basophils are able to produce IL-4 in various conditions
(Yoshimoto et al., 2009) and even in response to IL-33 (Kondo
et al., 2008). Motomura et al. (2014) reported that basophil-
derived IL-4 promotes the activation of ILC2s in papain-
induced eosinophilic lung inflammation. In humans, IL-4 also
acts in synergy with IL-33 to induce the expansion of ILC2s and
the production of IL-5 and IL-13 by the cells (Bal et al., 2016).
D. IL-33 produced by epidermal keratinocytes activates ILC2s, basophils, and
of IL-5 and IL-13 by ILC2s, which lead to eosinophil accumulation in the skin
tion of ILC2s and production of IL-5 and IL-13 and exacerbates the dermatitis
; ILC2, ILC2, group 2 innate lymphoid cells; Th2, T helper type 2.
Therefore, it is assumed that IL-33 activates basophils toproduce
IL-4, which also potentiates the activity of ILC2s in the skin.
Recently, ILC2s have been classified in two subtypes—
inflammatory ILC2s and natural ILC2s (Huang and Paul,
2016)—which are mainly responsive to IL-25/TSLP and
IL-33, respectively. According to the classification, ILC2s
induced in IL33tg mice belong to natural ILC2s. Kim et al.
(2013) reported that TSLP activates IL-33eindependent
inflammatory ILC2s in MC903 (calcipotriol)-induced AD-
like inflammation in mice and that TSLP-stimulated baso-
phil-derived IL-4 promotes the proliferation of inflammatory
ILC2s (Kim et al., 2014). On the other hand, we speculate
that IL-33estimulated basophil-derived IL-4 promotes natural
ILC2 proliferation in IL33tg mice (Figure 3def). The depletion
of basophils resulted in the reduction of ILC2s in the lesional
skin of IL33tg mice (Supplementary Figure S4), and it is
therefore likely that basophils promote the activation of nat-
ural ILC2s via IL-4 in IL33tg mice. Dupilumab, a human mAb
against the IL-4 receptor, is used for therapy of AD (Simpson
et al., 2016), and its efficacy on AD might be partly because it
blocks the activation of ILC2s via IL-4 (Bal et al., 2016).
However, further study is needed to clarify the possibility that
any basophil-derived factors or cytokines other than IL-4
participate in the process.
In this study, we show that ILC2s, in concert with baso-
phils, play a central role in IL-33einduced AD-like inflam-
mation in mice. According to a series of studies using IL33tg
mice and other reports, we propose a mechanism of IL-
33einduced innate inflammation in AD (Figure 6). It is still
www.jidonline.org 2191
Flow Cytometry
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
2192
Homogenized skin specimens were incubated in RPMI 1640 con-
taining 1% fetal calf serum, 85 mg/ml Liberase (Roche, Basel,
Switzerland), and 0.01% DNase I (Roche) at 37 �C for 60 minutes.
Dead cells were excluded by staining with fixable viability dye APC-
eFluor 780 (eBioscience). Cells were preincubated with anti-CD16/
32 Ab for blocking and were then stained with each Ab. Stained cells
were analyzed using a flow cytometer LE-SP6800Z (Sony, Tokyo,
Japan), and data were analyzed using FlowJo software, version 10.5
(Tree Star, Ashland, OR). The classification of cells is as follows:
ILC2s, Lin (B220, CD3, CD4, CD8, Gr-1, FcεRI, NK1.1, Siglec-F)e
CD45þ Sca-1þ ST2þ cells; basophils, Lin (B220, CD3, CD4, CD8,
Gr-1, NK1.1, Siglec-F, Thy-1.2)e c-kite CD45þ FcεRIþ DX5þ cells.
Precise gating strategies to identify ILC2s or basophils were
described previously (Imai et al., 2017) and can be viewed in
Supplementary Figures S1 and S4a, respectively.
unknown whether this mechanism is applicable to patients
with AD, but it is notable that these immune responses eli-
cited by IL-33 may be consistent with the immunological
properties of AD. ILC2s infiltrating in the lesional skin of
patients with AD are able to respond to IL-33 but not to IL-25
or TSLP (Salimi et al., 2013). IL-33 might represent a central
mediator for human AD—with a single administration of 300
mg of etokimab, an IL-33 Ab, 9 of 12 patients with AD (75%)
achieved Eczema Area and Severity Index-50 at day 15 in a
phase 2 trial (Weidinger et al., 2018). Thus, studies using
IL33tg mice help to illuminate the mechanism of AD and may
contribute to the discovery of novel therapeutic means tar-
geting the innate immune response involving ILC2s and
basophils.
MATERIALS AND METHODS
Mice
The mouse line hK14mIL33tg (IL33tg) expressing IL-33 driven by a
keratin-14 promoter was generated as described previously (Imai
et al., 2013). Rag2KO mice were from Taconic Farms (German-
town, PA); B6.C3(Cg)-Rorasg/J mice were from Jackson Laboratories
(Bar Harbor, ME). Bas-TRECK mice, established with a DT-based
conditional basophil deletion system, have been described previ-
ously (Motomura et al., 2014; Sawaguchi et al., 2012). IL33tg mice
were crossed with Bas-TRECK mice to generate IL33 Bas-TRECK Tg
mice. All animal studies were designed in accordance with the In-
ternational Guiding Principles for Biomedical Research Involving
Animals published by the Council for the International Organization
of Medical Science and were reviewed and approved by the Animal
Use and Care Committee of the Hyogo College of Medicine (Permit
Number: B90-251; 13-001; 15-067; 18-002). All mice used in this
study were maintained under specific pathogen-free conditions.
Antibodies
Fluorescence-labeled Abs for mouse B220, CD3, CD4, CD8, CD45,
CD49b (DX5), CD90.2 (Thy1.2), CD278 (ICOS), CD124 (IL-4Ra),
Gr-1, IL-4 (11B11), IL-5 (TRFK5), NK1.1, and Sca-1; anti-CD16/32
Ab (93); and Brilliant Violet 421 streptavidin were purchased from
BioLegend (San Diego, CA). Anti-Siglec-F Ab was from BD Bio-
sciences (San Jose, CA); anti-FcεRI (MAR-1) Ab was from e-Bio-
sciences (San Diego, CA); biotinylated anti-ST2 Ab (DJ8) was from
MD Biosciences, Inc. (Oakdale, MN).
Journal of Investigative Dermatology (2019), Volume 139
Intracellular cytokine staining
Intracellular cytokine staining was performed as described previ-
ously (Imai et al., 2017). In brief, cells from mouse lymph nodes or
skin were incubated in culture medium for 2e3 hours with IL-2
(lymph nodes) or IL-3 (skin) (PeproTech, Rocky Hill, NJ) under the
stimulation of phorbol 12-myristate 13-acetate/ionomycin in the
presence of monensin. Surface antigens were then stained in the
presence of monensin. Cells were fixed with BD Cytofix/Cytoperm
solution (BD Biosciences), permeabilized with BD Perm/Wash buffer
(BD Biosciences), and then stained with antieIL-4, IL-5, or control
IgG Ab.
ILC2 depletion
ILC2s were depleted in mice via bone marrow transplantation from
B6.C3(Cg)-Rorasg/J mice that lacked ILC2s, as described previously
(Halim et al., 2012). IL33tg mice (9e11 weeks old) were sublethally
irradiated (11 Gy) and subsequently received intravenous trans-
plantation of 5 � 106 whole bone marrow cells from 3-week-old
littermate controls (Rorawt/sg) or ILC2-depleted (Rorasg/sg) mice.
These mice were used for analysis 15 weeks after transplantation
(24e26 weeks old).
Basophil depletion
For basophil depletion by Ab, anti-FcεRI mAb (MAR-1) (e-Bio-
sciences) was used (Muto et al., 2014). MAR-1 treatment does not
activate basophils (Sokol et al., 2008). For depletion with MAR-1
mAb, mice (5e6 weeks old) were injectedintraperitoneally with
10 mg of MAR-1 or isotype control IgG1 Abs (eBio299Arm) (e-Bio-
sciences). Abs were injected twice a day for 3 consecutive days
every week (days 0 to 2, 7 to 10, and 14 to 16), and then mice were
killed at 21 days.
For basophil depletion by DT treatment, IL33 Bas-TRECK Tg mice
(5e6 weeks old) were injected intraperitoneally with 500 ng of DT in
250 ml of phosphate buffered saline (PBS). DT were injected every 3
days (days 0, 3, 6, 9, 12, 15, and 18), and then mice were killed at 21
days for analysis.
Tissue staining and immunofluorescence
Excised skin specimens were fixed with 4% (wt/vol) para-
formaldehyde and then embedded in paraffin. The skin tissues were
sectioned at 4-mm thickness, and deparaffinized sections were sub-
jected to hematoxylin and eosin or toluidine blue staining. MCP-8
was stained by immunofluorescence as described previously (Imai
et al., 2014). Sections were incubated with antiemMCP-8 Ab
(clone TUG8, BioLegend), and bound Abs were visualized with a
Cy3-conjugated donkey anti-rat IgG Ab (WAKO, Osaka, Japan).
Following mounting with a ProLong Diamond Antifade with DAPI
(Life Technologies, Gaithersburg, MD), fluorescence images were
recorded using a confocal laser scanning microscope LSM780 (Carl
Zeiss MicroImaging, Thornwood, NY).
Quantitative real-time PCR
A TaqMan RNA-to-Ct Kit (Thermo Fisher Scientific, Waltham, MA)
and an ABI7900HT sequence detection system (Applied Biosystems,
Foster City, CA) were used for quantitative real-time PCR. Total RNA
of mouse skin tissues was prepared using an RNeasy Micro Kit
(Qiagen, Hilden, Germany) according to the manufacturer’s in-
structions. The probes for quantitative real-time PCR were obtained
from Applied Biosystems Assays-on-Demand (Thermo Fisher Scien-
tific, Waltham, MA). The product numbers of the probes for Il4, Il5,
Il13, Il25, Il33, Tslp, Mcpt8, and Gapdh were Mm00445259_m1,
Mm00439646_m1, Mm00434204_m1, Mm00499822_m1,
Hu-Li J, Pannetier C, Guo L, Lo¨hning M, Gu H, Watson C, et al. Regulation of
response to peptide antigen through cell-to-cell contact. J Dermatol Sci
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
Mm00505403_m1, Mm01157588_m1, Mm00484933_m1, and
Mm99999915_g1, respectively. The abundance of each target tran-
script relative to the internal control (the expression of Gapdh) was
assessed according to the manufacturer’s instructions.
Preparation of CFSE-labeled ILC2
Lymph node cells from Rag2KO IL33tg mice were washed twice in
PBS and resuspended in PBS at a concentration of 2 � 107 cells/ml.
An equal volume of 1.25 mM CFSE in PBS (Molecular Probes,
Eugene, OR) was added to the cell suspension and incubated at
room temperature for 4 minutes as described previously (Imai et al.,
2008). The reaction was stopped by fetal calf serum. The cells were
washed with PBS and then incubated with Abs for lineage markers
(see the main text), Sca-1, and ST2. ILC2s (CFSEþ, Line, Sca-1þ,
ST2þ) were purified by electronic cell sorting using a FACSAria III
Cell Sorter (BD Biosciences).
Preparation of basophils
Bone marrowederived basophils were prepared as described
(Yoshimoto et al., 2009). In brief, bone marrow cells were cultured
for 2 weeks with IL-3 (PeproTech) in an RPMI 1640 medium sup-
plemented with 10% fetal calf serum. Samples were separated into
FcεRIþc-Kite cells (basophils) or FcεRIþc-Kitþ cells (mast cells).
Basophils were purified by electronic cell sorting using a FACSAria
III Cell Sorter (BD Biosciences). IL-4edeficient basophils were
derived from GFP/IL-4 knock-in (G4) mice (Hu-Li et al., 2001) pro-
vided by Dr William Paul (the Laboratory of Immunology, National
Institute of Asthma and Infectious Disease, NIH). The G4/IL-4 mice
were bred, and IL-4-deficient homozygous G4/G4 progeny mice
were used for the preparation of IL-4edeficient basophils.
In vitro cell division assay
CFSE-labeled ILC2s (105 cells/ml) were incubated with marrow-
derived basophils (ILC2:basophil ratio, 1:10) in the presence of hu-
man recombinant IL-2 (100 U/ml) (PeproTech), mouse recombinant
IL-3 (10 U/ml) (PeproTech), and mouse recombinant IL-33 (100 ng/
ml) (Hokudo, Sapporo, Japan) in a 96-well plate with flat bottoms.
After 5 days’ culture, cells were collected and analyzed by flow
cytometry.
Enzyme-linked immunosorbent assay
Skin specimens excised from mice were homogenized using a
BeadSmash homogenizer (WAKEN B TECH, Co Ltd, Kyoto, Japan)
and a Bio-Plex Cell Lysis Kit (Bio-Rad, Hercules, CA). The concen-
trations of cytokines and chemokines in skin homogenate samples or
culture supernatants were measured using a Bio-Plex Protein Array
System (Bio-Rad).
Statistical analyses
Statistical analyses were performed using GraphPad Prism 8
(GraphPad Software, San Diego, CA). P-values < 0.05 were
considered as significant differences.
Data availability statement
No datasets were generated or analyzed during the current study.
ORCIDs
Yasutomo Imai: https://orcid.org/0000-0003-3169-5717
Koubun Yasuda: https://orcid.org/0000-0002-3533-1702
Makoto Nagai: https://orcid.org/0000-0003-3638-706X
Minori Kusakabe: https://orcid.org/0000-0003-4422-4471
Masato Kubo: https://orcid.org/0000-0001-6706-4066
Kenji Nakanishi: https://orcid.org/0000-0002-0223-9088
Kiyofumi Yamanishi: http://orcid.org/0000-0003-0484-2320
2008;51:19e29.
Imai Y, Hosotani Y, Ishikawa H, Yasuda K, Nagai M, Jitsukawa O, et al.
Expression of IL-33 in ocular surface epithelium induces atopic kerato-
conjunctivitis with activation of group 2 innate lymphoid cells in mice. Sci
Rep 2017;7:10053.
Imai Y, Yasuda K, Sakaguchi Y, Futatsugi-Yumikura S, Yoshimoto T,
Nakanishi K, et al. Immediate-type contact hypersensitivity is reduced in
interleukin-33 knockout mice. J Dermatol Sci 2014;74:159e61.
Imai Y, Yasuda K, Sakaguchi Y, Haneda T, Mizutani H, Yoshimoto T, et al.
Skin-specific expression of IL-33 activates group 2 innate lymphoid cells
and elicits atopic dermatitis-like inflammation in mice. Proc Natl Acad Sci
USA 2013;110:13921e6.
Ito Y, Satoh T, Takayama K, Miyagishi C, Walls AF, Yokozeki H. Basophil
recruitment and activation in inflammatory skin diseases. Allergy 2011;66:
1107e13.
Kim BS, Siracusa MC, Saenz SA, Noti M, Monticelli LA, Sonnenberg GF, et al.
TSLP elicits IL-33-independent innate lymphoid cell responses to promote
skin inflammation. Sci Transl Med 2013;5:170ra16.
expression of IL-4 alleles: analysis using a chimeric GFP/IL-4 gene. Im-
munity 2001;14:1e11.
Imai Y, Hayashi N, Yasuda K, Tsutsui H, Mizutani H, Nakanishi K. Freshly
isolated Langerhans cells negatively regulate naive T cell activation in
CONFLICT OF INTEREST
The authors state no conflict of interest.
ACKNOWLEDGMENTS
The authors thank Orie Jitsukawa, Saori Kamiyama, Yoshiko Sakaguchi, and
members of the Joint-Use Research Facilities and the Center for Comparative
Medicine, Hyogo College of Medicine, for their help and/or technical assis-
tance. This work was supported in part by JSPS KAKENHI (17K10219 for KY,
17K16356 for MN, 18K08284 for YI) and by a research grant from the Takeda
Science Foundation for YI.
AUTHOR CONTRIBUTIONS
Conceptualization: KN, K Yamanishi, YI; Data Curation: M Kusakabe, MN, YI;
Formal Analysis: YI; Funding Acquisition: K Yamanishi, MN, YI; Investigation:
K Yasuda, M Kusakabe, MN, YI; Methodology: K Yasuda, M Kubo, YI; Project
Administration: K Yamanishi, YI; Resources: KN, K Yamanishi, K Yasuda, M
Kubo, YI; Supervision: KN, K Yamanishi, M Kubo; Validation: K Yasuda, YI;
Visualization: K Yamanishi, YI; Writing - Original Draft Preparation: YI;
Writing - Review and Editing: K Yamanishi, K Yasuda, YI
SUPPLEMENTARY MATERIAL
Supplementary material is linked to the online version of the paper at www.
jidonline.org, and at https://doi.org/10.1016/j.jid.2019.04.016.REFERENCES
Bal SM, Bernink JH, Nagasawa M, Groot J, Shikhagaie MM, Golebski K,
et al. IL-1beta, IL-4 and IL-12 control the fate of group 2 innate lymphoid
cells in human airway inflammation in the lungs. Nat Immunol 2016;17:
636e45.
Cox L, Calderon MA. Allergen immunotherapy for atopic dermatitis: is there
room for debate? J Allergy Clin Immunol Pract 2016;4:435e44.
Deckers IAG, McLean S, Linssen S, Mommers M, van Schayck CP, Sheikh A.
Investigating international time trends in the incidence and prevalence of
atopic eczema 1990e2010: a systematic review of epidemiological
studies. PLoS One 2012;7:e39803.
Haenuki Y, Matsushita K, Futatsugi-Yumikura S, Ishii KJ, Kawagoe T, Imoto Y,
et al. A critical role of IL-33 in experimental allergic rhinitis. J Allergy Clin
Immunol 2012;130:184e94.e11.
Halim TY, MacLaren A, Romanish MT, Gold MJ, McNagny KM, Takei F.
Retinoic-acid-receptor-related orphan nuclear receptor alpha is required
for natural helper cell development and allergic inflammation. Immunity
2012;37:463e74.
Huang Y, Paul WE. Inflammatory group 2 innate lymphoid cells. Int Immunol
2016;28:23e8.
www.jidonline.org 2193
Kim BS, Wang K, Siracusa MC, Saenz SA, Brestoff JR, Monticelli LA, et al.
Basophils promote innate lymphoid cell responses in inflamed skin.
J Immunol 2014;193:3717e25.
Kondo Y, Yoshimoto T, Yasuda K, Futatsugi-Yumikura S, Morimoto M,
Hayashi N, et al. Administration of IL-33 induces airway hyper-
responsiveness and goblet cell hyperplasia in the lungs in the absence of
adaptive immune system. Int Immunol 2008;20:791e800.
Li C, Maillet I, Mackowiak C, Viala C, Di Padova F, Li M, et al. Experimental
atopic dermatitis depends on IL-33R signaling via MyD88 in dendritic
cells. Cell Death Dis 2017;8:e2735.
Lo BC, Gold MJ, Hughes MR, Antignano F, Valdez Y, Zaph C, et al. The
orphan nuclear receptor ROR alpha and group 3 innate lymphoid cells
drive fibrosis in a mouse model of Crohn’s disease. Sci Immunol 2016;1:
eaaf8864.
Londei M, Kenney B, Los G, Marino M. A phase 1 study of ANB020, an anti-
IL-33 monoclonal antibody, in healthy volunteers. J Am Acad Dermatol
2017;76:AB20.
Mashiko S, Mehta H, Bissonnette R, Sarfati M. Increased frequencies of ba-
sophils, type 2 innate lymphoid cells and Th2 cells in skin of patients with
atopic dermatitis but not psoriasis. J Dermatol Sci 2017;88:167e74.
Matsuba-Kitamura S, Yoshimoto T, Yasuda K, Futatsugi-Yumikura S, Taki Y,
Muto T, et al. Contribution of IL-33 to induction and augmentation of
Ryu WI, Lee H, Bae HC, Jeon J, Ryu HJ, Kim J, et al. IL-33 down-regulates
CLDN1 expression through the ERK/STAT3 pathway in keratinocytes.
J Dermatol Sci 2018;90:313e22.
Salimi M, Barlow JL, Saunders SP, Xue L, Gutowska-Owsiak D, Wang X, et al.
A role for IL-25 and IL-33-driven type-2 innate lymphoid cells in atopic
dermatitis. J Exp Med 2013;210:2939e50.
Salter BM, Oliveria JP, Nusca G, Smith SG, Tworek D, Mitchell PD, et al. IL-25
and IL-33 induce type 2 inflammation in basophils from subjects with
allergic asthma. Respir Res 2016;17:5.
Savinko T, Matikainen S, Saarialho-Kere U, Lehto M, Wang G, Lehtima¨ki S,
et al. IL-33 and ST2 in atopic dermatitis: expression profiles and modula-
tion by triggering factors. J Invest Dermatol 2012;132:1392e400.
Sawaguchi M, Tanaka S, Nakatani Y, Harada Y, Mukai K, Matsunaga Y, et al.
Role of mast cells and basophils in IgE responses and in allergic airway
hyperresponsiveness. J Immunol 2012;188:1809e18.
Schmitz J, Owyang A, Oldham E, Song Y, Murphy E, McClanahan TK, et al.
IL-33, an interleukin-1-like cytokine that signals via the IL-1 receptor-
related protein ST2 and induces T helper type 2-associated cytokines.
Immunity 2005;23:479e90.
Schneider E, Petit-Bertron AF, Bricard R, Levasseur M, Ramadan A,
Girard JP, et al. IL-33 activates unprimed murine basophils directly
in vitro and induces their in vivo expansion indirectly by promoting
hematopoietic growth factor production. J Immunol 2009;183:
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
2194
Matsuda A, Okayama Y, Terai N, Yokoi N, Ebihara N, Tanioka H, et al. The
role of interleukin-33 in chronic allergic conjunctivitis. Invest Ophthalmol
Vis Sci 2009;50:4646e52.
Moro K, Yamada T, Tanabe M, Takeuchi T, Ikawa T, Kawamoto H, et al. Innate
production of T(H)2 cytokines by adipose tissue-associated c-Kit(þ)Sca-
1(þ) lymphoid cells. Nature 2010;463:540e4.
Motomura Y, Morita H, Moro K, Nakae S, Artis D, Endo TA, et al. Basophil-
derived interleukin-4 controls the function of natural helper cells, a
member of ILC2s, in lung inflammation. Immunity 2014;40:758e71.
Moussion C, Ortega N, Girard JP. The IL-1-like cytokine IL-33 is constitutively
expressed in the nucleus of endothelial cells and epithelial cells in vivo: a
novel ‘alarmin’? PLoS One 2008;3:e3331.
Muto T, Fukuoka A, Kabashima K, Ziegler SF, Nakanishi K, Matsushita K, et al.
The role of basophils and proallergic cytokines, TSLP and IL-33, in cuta-
neously sensitized food allergy. Int Immunol 2014;26:539e49.
Nakashima C, Otsuka A, Kabashima K. Recent advancement in the mecha-
nism of basophil activation. J Dermatol Sci 2018;91:3e8.
Neild VS, Marsden RA, Bailes JA, Bland JM. Egg and milk exclusion diets in
atopic eczema. Br J Dermatol 1986;114:117e23.
Plager DA, Weiss EA, Kephart GM, Mocharla RM, Matsumoto R, Checkel JL,
et al. Identification of basophils by a mAb directed against pro-major basic
protein 1. J Allergy Clin Immunol 2006;117:626e34.
Journal of Investigative Dermatology (2019), Volume 139
3591e7.
Seltmann J, Roesner LM, von Hesler FW, Wittmann M, Werfel T. IL-33 im-
pacts on the skin barrier by downregulating the expression of filaggrin.
J Allergy Clin Immunol 2015;135:1659e61.e4.
Simpson EL, Bieber T, Guttman-Yassky E, Beck LA, Blauvelt A, Cork MJ, et al.
Two phase 3 trials of dupilumab versus placebo in atopic dermatitis. N Engl
J Med 2016;375:2335e48.
Sokol CL, Barton GM, Farr AG, Medzhitov R. A mechanism for the initiation of
allergen-induced T helper type 2 responses. Nat Immunol 2008;9:310e8.
Stott B, Lavender P, Lehmann S, Pennino D, Durham S, Schmidt-Weber CB.
Human IL-31 is induced by IL-4 and promotes TH2-driven inflammation.
J Allergy Clin Immunol 2013;132:446e54.e5.
Tsutsui H, Yamanishi Y, Ohtsuka H, Sato S, Yoshikawa S, Karasuyama H. The
basophil-specific protease mMCP-8 provokes an inflammatory response in
the skin with microvascular hyperpermeability and leukocyte infiltration.
J Biol Chem 2017;292:1061e7.
Weidinger S, Beck LA, Bieber T, Kabashima K, Irvine AD. Atopic dermatitis.
Nat Rev Dis Primers 2018;4:1.
Yoshimoto T, Yasuda K, Tanaka H, Nakahira M, Imai Y, Fujimori Y, et al.
Basophils contribute to T(H)2-IgE responses in vivo via IL-4 production and
presentation of peptide-MHC class II complexes to CD4þ T cells. Nat
Immunol 2009;10:706e12.
experimental allergic conjunctivitis. Int Immunol 2010;22:479e89.
Supplementary Figure S1. Gating strategy of basophils. Basophils in IL33tg mouse skin were detected as Lin (B220, CD3, CD4, CD8, Gr-1, NK1.1, Siglec-F,
Thy-1.2)e c-kite CD45þ FcεRIþ DX5þ cells by flow cytometry. The numbers indicate the proportion of cells in each gate. IL33tg, IL-33 transgenic; Lin, lineage
markers.
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
Supplementary Figure S2. Lymph node cells from Rag2KO IL33tg mice are enriched in IL-5-producing ILC2s. (a) Flow cytometry of cells from the lymph node
of 8-week-old Rag2KO IL33tg mice. (b) Intracellular IL-5 detection by flow cytometry of cellsfrom the lymph node of Rag2KO IL33tg mice. Cells were
stimulated with phorbol 12-myristate 13-acetate/ionomycin for 3 hours. Note that IL-5 is produced only by ST2þ ILC2s in the lymph node cells. The numbers
indicate the proportion of cells in each gate. Data are representative of at least two independent experiments. IL33tg, IL-33 transgenic; ILC2, group 2 innate
lymphoid cells; KO, knockout; Lin, lineage markers; Rag2, recombination-activating gene 2.
www.jidonline.org 2194.e1
Supplementary Figure S3. Depletion of basophils did not reduce skin mast cells. (a) Toluidine blue staining of mast cells in isotype control IgG-treated and anti-
FcεRI mAb (MAR-1)-treated skins. Rag2KO mice were used as a control. Panels are representative of at least four mice and two independent experiments. Bars,
100 mm. (b) Toluidine blue staining of mast cells in the skin of DT-treated IL33tg and IL33 Bas-TRECK Tg mice. WT mice were used as a control. Panels are
representative of at least four mice and two independent experiments. Bars, 100 mm. Arrowheads indicate toluidine blue-positive mast cells in (a) and (b). DT,
diphtheria toxin; IL33tg, IL-33 transgenic; KO, knockout; mAb, monoclonal antibody; Rag2, recombination-activating gene 2; TRECK, toxin receptoremediated
conditional cell knockout; WT, wild-type.
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
Journal of Investigative Dermatology (2019), Volume 1392194.e2
Supplementary Figure S4. Depletion of basophils reduced ILC2s in Rag2KO IL33tg mice. (a) ICOSþ cells correspond to ILC2s (Sca-1þ ST2þ Thy1.2þ IL-4Raþ)
in Rag2KO IL33tg mouse skin. Both ILC2s and T cells express ICOS, but Rag2KO mice lack T cells. Therefore, ICOSþ cells can be defined as ILC2s by flow
cytometry. The numbers indicate the proportion of cells in each gate or quadrant. (b) Double immunofluorescence of mMCP-8þ basophils and ICOSþ ILC2s in
the skin of isotype control and anti-FcεRI mAb (MAR-1)-Ab-injected Rag2KO IL33tg mice. Sections were incubated with rat antiemMCP-8 Ab (clone TUG8)
(BioLegend, San Diego, CA) and rabbit anti-ICOS Ab (clone EPR20560) (Abcam, Cambridge, MA), then bound Abs were visualized with goat anti-rat IgG Alexa
Flour 488 (A-11006) (Invitrogen) and Cy3-conjugated goat anti-rabbit IgG Ab (WAKO, Tokyo, Japan). Untreated Rag2KO mice were used as a control. Note that
the numbers of ICOSþ ILC2s were reduced in MAR-1 injected Rag2KO IL33tg mouse skin. Bars, 50 mm. Data are representative of at least four mice. Ab,
antibody; IL33tg, IL-33 transgenic; ILC2, ILC2, group 2 innate lymphoid cells; KO, knockout; mAb, monoclonal antibody; mMCP-8, mouse mast cell protease 8;
Rag2, recombination-activating gene 2.
Y Imai et al.
ILC2s and Basophils Mediate IL-33eInduced AD
www.jidonline.org 2194.e3
	IL-33–Induced Atopic Dermatitis–Like Inflammation in Mice Is Mediated by Group 2 Innate Lymphoid Cells in Concert with Baso ...
	Introduction
	Results
	AD-like inflammation occurs in IL33tg mice lacking Th2 cells
	ILC2s are essential to the development of AD-like inflammation in IL33tg mice
	Basophils activate ILC2s via IL-4
	AD-like inflammation is suppressed in basophil-depleted IL33tg skin
	Discussion
	Materials and Methods
	Mice
	Antibodies
	Flow Cytometry
	Intracellular cytokine staining
	ILC2 depletion
	Basophil depletion
	Tissue staining and immunofluorescence
	Quantitative real-time PCR
	Preparation of CFSE-labeled ILC2
	Preparation of basophils
	In vitro cell division assay
	Enzyme-linked immunosorbent assay
	Statistical analyses
	Data availability statement
	ORCIDs
	Conflict of Interest
	Acknowledgments
	Author Contributions
	Supplementary Material
	References