Content uploaded by Gunnar Nilsson
Author content
All content in this area was uploaded by Gunnar Nilsson on Mar 02, 2015
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
of June 8, 2011
This information is current as
http://www.jimmunol.org/content/186/4/2523
doi:10.4049/jimmunol.1003383
January 2011;
2011;186;2523-2528; Prepublished online 14J Immunol
Carolina Lunderius-Andersson
Möller-Westerberg, Padraic G. Fallon, Gunnar Nilsson and
Mattias Enoksson, Katarina Lyberg, Christine
IL-33 Recognition
Mast Cells as Sensors of Cell Injury through
References http://www.jimmunol.org/content/186/4/2523.full.html#ref-list-1
, 21 of which can be accessed free at:cites 46 articlesThis article
Subscriptions http://www.jimmunol.org/subscriptions is online atThe Journal of ImmunologyInformation about subscribing to
Permissions http://www.aai.org/ji/copyright.html
Submit copyright permission requests at
Email Alerts http://www.jimmunol.org/etoc/subscriptions.shtml/
Receive free email-alerts when new articles cite this article. Sign up at
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Immunologists, Inc. All rights reserved.
by The American Association ofCopyright ©2011
9650 Rockville Pike, Bethesda, MD 20814-3994.
The American Association of Immunologists, Inc.,
is published twice each month byThe Journal of Immunology
on June 8, 2011www.jimmunol.orgDownloaded from
The Journal of Immunology
Mast Cells as Sensors of Cell Injury through IL-33
Recognition
Mattias Enoksson,* Katarina Lyberg,* Christine Mo
¨ller-Westerberg,* Padraic G. Fallon,
†
Gunnar Nilsson,*
,1
and Carolina Lunderius-Andersson*
,1
In response to cell injury, caused, for example, by trauma, several processes must be initiated simultaneously to achieve an acute
inflammatory response designed to prevent sustained tissue damage and infection and to restore and maintain tissue homeostasis.
Detecting cell injury is facilitated by the fact that damaged cells release intracellular molecules not normally present in the ex-
tracellular space. However, potential underlying mechanisms for the recognition of endogenous danger signals released upon cell
injury have yet to be elucidated. In this study, we demonstrate that mast cells, potent promoters of acute inflammation, play a key
role in responding to cell injury by recognizing IL-33 released from necrotic structural cells. In an in vitro model of cell injury, this
recognition was shown to involve the T1/ST2 receptor and result in the secretion of proinflammatory leukotrienes and cytokines by
mouse mast cells. Remarkably, of all of the components released upon necrosis, our results show that IL-33 alone is a key component
responsible for initiating proinflammatory responses in mast cells reacting to cell injury. Our findings identify IL-33 as a key danger
signal released by necrotic structural cells capable of activating mast cells, thus providing novel insights concerning the role of mast
cells as sensors of cell injury. The Journal of Immunology, 2011, 186: 2523–2528.
The ability to recognize and respond to cell injury is fun-
damental to the survival of all animal species. Upon cell
injury, endogenous danger signals, so-called damage-
associated molecular patterns, are released by necrotic cells, in-
cluding heat shock proteins (1), high-mobility group box 1
(HMGB1) (2, 3), uric acid (4), and cytokines of the IL-1 family:
IL-1a(5, 6) and, as recently suggested, IL-33 (7, 8). Such en-
dogenous danger signals are recognized by various immune cells
that initiate inflammatory processes (9). The requirements made
on the specialized cells that respond to tissue damage are nu-
merous. In the first place, such sentinel cells must be prepositioned
in tissues, allowing a rapid response. Secondly, these cells must
possess the capability to produce and secrete selective mediators
required for the induction of an acute inflammatory response in-
volving vascular changes and the recruitment of leukocytes. Mast
cells possess these important characteristics (10). These long-lived
cells are present in all tissues, especially numerous at sites ex-
posed to the external environment (11, 12), and rapidly produce
and secrete a variety of signal substances upon activation, in-
cluding histamine, proteases, eicosanoids, chemokines, and cyto-
kines (13). Taken together, these properties make mast cells ideal
first-hand responders to tissue damage/cell injury, capable of ini-
tiating and orchestrating an inflammatory response.
As mast cells have previously been shown to recognize and
respond to exogenous danger signals (also called pathogen-
associated molecular patterns) such as LPS, zymosan, and pepti-
doglycan (11, 14), we hypothesized that mast cells also play an
important role in the recognition of endogenous danger signals,
such as IL-33, thereby contributing cell injury responses.
IL-33 is a novel cytokine of the IL-1 family that has previously
been shown to induce Th2-associated cytokines (15), as well as
induce release of proinflammatory mediators in mouse bone
marrow-derived mast cells (BMMCs) (16–20) and human mast
cells (21, 22), in which IL-33 also promoted maturation (21), en-
hanced mast cell survival, and increased mast cell adhesion to fibro-
nectin (22). In addition, IL-33 is involved in tryptase (mouse mast
cell protease-6) regulation in BMMCs (23). Importantly, IL-33
seems to be preferentially released from necrotic cells (24). For
instance, IL-33 has been shown to be released from endothelial cells
following mechanical injury (7). On the contrary, release of bioac-
tive IL-33 from apoptotic cells has not been demonstrated. Instead,
IL-33 has been shown to be inactivated during apoptosis (24).
In this study, we have examined the role of mast cells as sen-
sors of cellular injury. The hypothesis that an important function
for mast cells is to quickly respond to cell injury was proposed al-
ready 50 y ago (25). Despite the fact that mast cells are recog-
nized as important inflammatory cells (10, 13), the mechanism for
this hypothesis has not been deciphered. In this report, by moni-
toring responses of mast cells treated with cell-free supernatant
from necrotic cells, we provide a mechanism for this hypothesis.
Materials and Methods
Experimental animals, isolation, and in vitro stimulation of
cells
Bone marrow was isolated from C57BL/6 wild-type, MyD88
2/2
(26),
TLR1
2/2
(27), TLR2
2/2
(28), TLR4
2/2
(29), TLR5
2/2
(30), TLR6
2/2
(31), TLR7
2/2
(32), TLR8
2/2
(33) and TLR9
2/2
(34), A
2A
R
2/2
(35),
*Clinical Immunology and Allergy Unit, Department of Medicine, Karolinska In-
stitute, SE-171 76 Stockholm, Sweden; and
†
Institute of Molecular Medicine, St.
James’s Hospital, Trinity College Dublin, Dublin 8, Ireland
1
G.N. and C.L.-A. contributed equally to this work.
Received for publication October 12, 2010. Accepted for publication December 10,
2010.
This work was supported by the Swedish Research Council, the Swedish Cancer
Foundation, the Consul Th C Bergh Foundation, the Ellen, Walter, and Lennart
Hesselman Foundation, the Ollie and Elof Ericsson Foundation, the King Gustaf
V’s 80 Years Foundation, the Hans von Kantzow Foundation, the A
˚ke Wiberg Foun-
dation, the Magnus Bergvall Foundation, and the Karolinska Institute.
Address correspondence and reprint requests to Dr. Gunnar Nilsson, Clinical Immu-
nology and Allergy Unit, Department of Medicine, Karolinska Institute, KS L2:04,
SE-171 76 Stockholm, Sweden. E-mail address: Gunnar.P.Nilsson@ki.se
Abbreviations used in this article: BMMC, bone marrow-derived mast cell; HMGB1,
high-mobility group box 1; MEF, mouse embryonal fibroblast; qPCR, quantitative
PCR; siRNA, small interfering RNA.
Copyright Ó2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1003383
on June 8, 2011www.jimmunol.orgDownloaded from
A
3
AR
2/2
(36), and T1/ST2
2/2
(37) mice, and BMMCs were prepared as
described earlier (38). The purity of the BMMCs obtained was routinely
.95%, as assessed by toluidine blue staining. In a typical experiment,
10
6
/ml BMMCs were seeded into each well of 48-well plates and exposed to
various amounts of necrotic cell supernatant (see further below) for 0.5 or
24 h, after which the supernatants from each culture were collected. The
total volume per well was 200 ml. As a positive control for mast cell acti-
vation, ionomycin (Sigma-Aldrich) was used. For inhibition studies, cells
were pretreated with 10 mM SB203580 (a p38 inhibitor) for 30 min prior to
the addition of necrotic cell supernatant or rIL-33 (Alexis Biochemicals).
Preparation of necrotic cell supernatant
Supernatant was collected from mouse embryonal fibroblasts (MEFs),
HMGB1
2/2
MEFs, smooth muscle cells, keratinocytes, monocytes, spleno-
cytes, mixed neuronal cells, and BMMCs rendered necrotic by repeated
freeze-thawing. In brief, the cells were resuspended in PBS at a density
of 20 310
6
, subjected to repeated freeze-thawing (cycling between 280
and 37˚C four times), and then centrifuged at 13,000 rpm for 10 min and
the supernatant collected.
Monitoring the responses of mast cells
Histamine release was determined using an immunoassay kit (Immunotech;
Beckman Coulter). Secreted cys-leukotrienes were quantified using an en-
zyme immunoassay kit (Amersham Biosciences) and cytokines/chemokines
with commercial ELISA kits (Biosource International and R&D Systems)
and Luminex (Millipore).
Knockdown of IL-33 expression by MEFs with small
interfering RNA
To transiently silence IL-33 expression by MEFs, these cells were trans-
fected with IL-33 small interfering RNA (siRNA) or, as a control, non-
specific siRNA (Dharmacon), harvested 24 h later, and rendered necrotic for
the preparation of necrotic supernatant as described above. To validate IL-
33 knockdown, expression of the corresponding mRNA was analyzed by
quantitative PCR (qPCR) and supernatant levels of the protein itself by
ELISA. qPCR was performed in an iCycler (Bio-Rad) using the following
primers: IL-33 sense: 59-TCC TTG CTT GGC AGT ATC CA-39and IL-33
antisense: 59-TGC TCA ATG TGT CAA CAG ACG-39.
Western blotting
Following treatment of BMMCs with 10 ng/ml rIL-33 or necrotic cell
supernatant, levels of phosphorylated p38 and total p38 protein were de-
termined by standard procedures (38), utilizing Abs directed against
phospho-p38 MAPK (Thr
180
/Tyr
182
) and p38 MAPK, respectively (both
from Cell Signaling Technology). IL-33 was detected in necrotic cell su-
pernatant utilizing an anti-mouse IL-33 Ab (AF3626; R&D Systems), fol-
lowed by a donkey anti-goat IgG-HRP secondary Ab (Santa Cruz Biotec-
hnology).
Statistical analysis
The Mann–Whitney and one-way ANOVA (Kruskal-Wallis test) tests were
employed for statistical analyses, with a pvalue ,0.05 being considered
statistically significant.
Results
BMMCs exhibit a potent proinflammatory response to
treatment with necrotic cell supernatant
When cells die in a necrotic fashion following cell injury, cell
content is released into the extracellular space. Certain intracellular
molecules released in this way function as endogenous danger
signals, alerting the immune system, causing it to mount an in-
flammatory response. We hypothesized that mast cells can sense
such warning signals, thereby participating in the initiation of an
inflammatory response through release of proinflammatory medi-
ators. To test this hypothesis, we rendered fibroblasts necrotic by
repeated freeze-thawing cycles. The cell-free necrotic supernatant
isolated from MEFs was subsequently transferred to a culture of
C57BL/6 BMMCs and the responses of the latter monitored. When
incubated in this manner with supernatant from cultures containing
as many or twice as many necrotic MEFs, the BMMCs did not
degranulate (i.e., did not release histamine) (Fig. 1A), but released
leukotrienes (Fig. 1B) and secreted the proinflammatory cytokines
IL-6 (Fig. 1C)andTNF-a(Fig. 1D), thus exhibiting a potent in-
flammatory response. In contrast, supernatants from MEFs rendered
apoptotic through gamma-radiation did not generate IL-6 secretion
by BMMCs (data not shown). Similar results were obtained with
BMMCs prepared from BALB/c mice (data not shown). IL-4, IL-10,
IL-13, and MCP-1 were not secreted in this system (data not shown).
No IL-6 secretion was detected in supernatants from cells treated for
0.5 h with necrotic cell supernatant (data not shown). As a positive
control, cells were treated with 1 mM ionomycin, which induced
histamine release (45.5 68.2% of total, n= 5, measured 30 min
post-treatment) and IL-6 release (39,980 62,836 pg/ml, n=5,
measured 24 h post-treatment) (data not shown).
Our next attempt aimed at identifying the molecular signals
responsible for activating the mast cells. Our initial results led us to
suspect the involvement of HMGB1, a well-characterized danger
signal known to be released from necrotic cells (3) and activate
macrophages (39). To investigate the role of HMGB1 in our
system, BMMCs were subsequently exposed to the necrotic su-
pernatant from MEFs isolated from HMGB1
2/2
mice (40). Under
these conditions, the BMMCs secreted levels of IL-6 that did not
differ significantly from the response of wild-type MEFs to ne-
crotic cell supernatant (Fig. 2A). Next, to examine the possible
involvement of uric acid, another danger signal released from
damaged cells (4), the MEFs were rendered necrotic after culture
in the presence of allopurinol (4), an inhibitor of uric acid pro-
duction. Once again, no difference in BMMC activation compared
with untreated necrotic supernatant was observed (Fig. 2A).
Moreover, the use of BMMCs from A
2A
R
2/2
and A
3
R
2/2
mice,
deficient in the adenosine A
2A
and A
3
receptors, respectively, still
gave the same levels of IL-6 secretion (Fig. 2B).
Together, these findings revealed that the inflammatory response
of mast cells to factors released by necrotic fibroblasts does not
involve the known danger signals HMGB1, uric acid, or adenosine.
Mast cell responses to necrotic cell supernatant are mediated
through the adaptor protein MyD88
Because several exogenous danger signals (including LPS and
zymosan), as well as certain endogenous danger signals (41, 42),
FIGURE 1. Mast cells initiate proinflammatory responses upon expo-
sure to the supernatant of necrotic cell cultures. BMMCs exposed to the
supernatant from necrotic cultures containing as many (1:1) or twice as
many (2:1) cells did not release histamine (A) but released cysteinyl leu-
kotrienes (B) during 0.5 h and IL-6 (C) and TNF-a(D) during 24 h. In all
experiments, unexposed BMMCs (Unstim) served as controls. During
BMMC treatment, 10
6
cells/ml were used in a total volume of 200 ml. The
values presented are means 6SEM (n= 5). For statistical analysis, one-
way ANOVA test was applied. The experiments were repeated at least
three times. *p,0.05, **p,0.01 compared with unexposed cells.
2524 MAST CELLS AS SENSORS OF CELL INJURY
on June 8, 2011www.jimmunol.orgDownloaded from
act through TLRs, for which MyD88 acts as an adaptor protein,
we reasoned that responses to necrotic supernatant perhaps also
could be mediated through a receptor using MyD88. To investigate
this, BMMCs were prepared from MyD88
2/2
mice and treated
with necrotic supernatant. MyD88
2/2
BMMCs were found to lack
the normal IL-6 response to necrotic cell supernatant completely
(Fig. 2C), indicating that the observed IL-6 production was MyD88
dependent. To pinpoint the upstream receptor of MyD88 that re-
sponds to necrotic cell supernatant, BMMCs were derived from
TLR1
2/2
, TLR2
2/2
, TLR4
2/2
, TLR5
2/2
, TLR6
2/2
, TLR7
2/2
,
TLR8
2/2
, and TLR9
2/2
mice. When exposed to necrotic cell su-
pernatant, these cells secreted IL-6 at similar levels as wild-type
cells (Fig. 2D), thus demonstrating that none of these receptors
alone are necessary for the observed response.
IL-33 present in necrotic cell supernatant activates mast cells
through the T1/ST2 receptor
Because members of the IL-1R family share MyD88-dependent
signaling pathways with TLRs, the T1/ST2 receptor (43), a mem-
ber of the IL-1 family of receptors, was examined next.
Interestingly, we found that mast cells lacking the T1/ST2 re-
ceptor failed to secrete IL-6 and released considerably lower lev-
els of TNF-acompared with wild-type cells upon exposure to ne-
crotic cell supernatant (Fig. 3A). In addition, T1/ST2
2/2
BMMCs
also produced substantially lower levels of cysteinyl leukotrienes
and LTB
4
upon exposure to necrotic supernatant (Fig. 3B). As all
observed responses were T1/ST2 dependent, we investigated IL-
33 protein levels in necrotic cell supernatant by Western blotting
(Fig. 3C) and found full-length pro–IL-33 to be present as a band
at ∼30 kDa, in line with a previous publication (19). In addition,
we observed a band migrating identically to rIL-33, likely repre-
senting cleaved IL-33. We next investigated responses of BMMCs
treated with the T1/ST2 ligand IL-33. As shown in Fig. 3D,
treating BMMCs with IL-33 resulted not only in a dose-dependent
release of IL-6 and TNF-a, but also in the release of cysteinyl
leukotrienes. Thus, treating BMMCs with IL-33 generated similar
responses as BMMCs treated with necrotic supernatant. Taken
together, the observed results imply that IL-33 released by ne-
crotic cells induce the production of IL-6, TNF-a, and leuko-
trienes by mast cells, as production of these mediators was absent
or limited in T1/ST2
2/2
mast cells.
In an attempt to detect further similarities in responses of
BMMCs treated with necrotic supernatant and IL-33, we next in-
vestigated signaling pathways downstream of T1/ST2 and MyD88.
This was investigated by monitoring IL-6 secretion by mast
cells preincubated with SB203580, an inhibitor of p38, prior to
treatment with necrotic cell supernatant or IL-33. Preincubation
with SB203580 reduced IL-6 secretion markedly, both in BMMCs
treated with necrotic supernatant and also in BMMCs treated
with IL-33 (Fig. 4A). In agreement with these findings, phos-
phorylation of p38 was detected 5 min after exposure of BMMCs
to necrotic supernatant or IL-33 (Fig. 4B).
To provide definite evidence that IL-33 is the molecular com-
ponent of necrotic cell supernatant that activated the mast cells,
expression of this IL by MEFs was silenced with siRNA, and the
MEFs were rendered necrotic. The effectiveness of this silencing
was validated by qPCR (Fig. 5A) and by determining the levels of
IL-33 in the necrotic supernatant (Fig. 5B). When BMMCs were
exposed to the supernatant from necrotic MEFs lacking IL-33,
secretion of IL-6 was potently attenuated (Fig. 5C), demonstrat-
ing that the observed responses of BMMCs treated with necrotic
supernatant are dependent almost exclusively on IL-33. In con-
clusion, our results demonstrate that IL-33 released during ne-
crosis is an important mast cell activator.
IL-33 is released from necrotic cells of structural but not
hematopoietic origin
We next investigated the possibility that necrotic structural cells
other than fibroblasts, such as keratinocytes, neuronal cells (as-
trocytes), and smooth muscle cells, some of which express IL-33 at
relatively high levels, also release factors that activate mast cells.
The supernatant from cultures of above-mentioned necrotic struc-
tural cells could indeed induce IL-6 secretion by BMMCs (Fig. 5D),
whereas corresponding necrotic supernatant prepared from cells
of hematopoietic origin (splenocytes, monocytes, and BMMCs)
had no effect. Moreover, the necrotic supernatant from structural
cells contained higher levels of IL-33 (Fig. 5E). Taken together, our
data suggest that IL-33 released upon cell injury mainly derives
from damaged structural cells.
Discussion
In this study, we investigated the hypothesis that mast cells are
important sensors of cell injury, as they display qualities important
FIGURE 2. The activation of mast cells by necrotic cell supernatant involves MyD88. Following 24 h of exposure to supernatant from the same number
(1:1) or twice as many (2:1) necrotic MEFs, secretion of IL-6 by BMMCs exposed to the supernatant of MEFs treated with allopurinol or isolated from
HMGB1
2/2
mice (A); A
2A
R
2/2
and A
3
AR
2/2
BMMCs (2:1 ratio) (B); MyD88
2/2
BMMCs (C); and TLR1
2/2
, TLR2
2/2
, TLR4
2/2
, TLR5
2/2
, TLR6
2/2
,
TLR7
2/2
, TLR8
2/2
, and TLR9
2/2
BMMCs (2:1 ratio) (D) was determined. During BMMC treatment, 10
6
cells/ml were used in a total volume of 200 ml.
The values presented are means 6SEM (n= 3–5). The experiments were repeated at least three times. *p,0.05, **p,0.01 in comparison with untreated
control cells (Unstim) (A) or wild-type cells (C). For statistical analyses, one-way ANOVA (A) and Mann–Whitney (B–D) test were applied.
The Journal of Immunology 2525
on June 8, 2011www.jimmunol.orgDownloaded from
to first-hand responders to tissue damage. By monitoring the
responses of mast cells treated with cell-free supernatant from
necrotic cells, we demonstrate that IL-33 is a key alarmin released
by necrotic structural cells. We show that mast cells are activated by
IL-33 released from necrotic cells to initiate a potent proin-
flammatory response, suggesting that mast cells make important
contributions to early cell injury responses.
During necrosis, cell integrity is compromised, resulting in the
release of several different danger signals. To investigate whether
mast cells could respond to endogenous danger signals released
from necrotic cells, we rendered MEFs necrotic and treated mast
cells with the supernatant from these cells. The use of necrotic
supernatant from lysed cells rather than purified individual danger
signals better represents physiological conditions, allowing for
a more credible in vitro model of cell injury. Our initial experi-
ments revealed that BMMCs treated with supernatant from necrotic
MEFs exhibited a potent inflammatory response, manifested by IL-
6 and TNF-asecretion and leukotriene release. However, this
response was not accompanied by degranulation, indicating solely
de novo production of the released mediators. It has been shown
earlier that HMGB1 is an important danger signal released by dying
cells and that HMGB1
2/2
cells have a greatly reduced capacity to
induce inflammation compared with wild-type necrotic cells (3).
However, in our hands, BMMCs treated with necrotic supernatant
obtained from HMGB1
2/2
MEFs secreted IL-6 at similar levels as
BMMCs treated with necrotic supernatant from wild-type cells.
This is in agreement with a study performed by Chen et al. (5), in
FIGURE 4. Mast cell responses to necrotic cell supernatant and IL-33
are mediated through a p38-dependent pathway. A, IL-6 release was de-
termined in wild-type BMMCs pretreated with or without 10 mM selective
p38 inhibitor SB203580 for 30 min prior to treatment with necrotic su-
pernatant (2:1 ratio) or 10 ng/ml IL-33. B, Levels of phosphorylated p38
(p-p38) in wild-type BMMCs following treatment with necrotic cell su-
pernatant (2:1 ratio) or 10 ng/ml IL-33 for the time periods indicated were
examined by Western blotting. During BMMC treatment, 10
6
cells/ml
were used in a total volume of 200 ml. The values shown are means 6
SEM (n= 5). For statistical analysis, the Mann–Whitney Utest was ap-
plied. The experiments were repeated at least three times. **p,0.01 in
comparison with cells not exposed to SB203580.
FIGURE 5. IL-33 released from necrotic cells is responsible for the
activation of mast cells. siRNA-mediated silencing of the expression of IL-
33 in MEFs was validated by qPCR (A) and an ELISA assay (B). C,
Following exposure of BMMCs for 24 h to supernatant from twice as many
(2:1) MEFs rendered necrotic following treatment with IL-33 or non-
targeting siRNA, IL-6 secretion was assayed. D, Following exposure of
BMMCs for 24 h to supernatant from twice as many (2:1) cells of the types
indicated, IL-6 secretion was determined. E, The levels of IL-33 in the
same necrotic cell supernatants were measured by ELISA. During BMMC
treatment, 10
6
cells/ml were used in a total volume of 200 ml. In A,
a representative experiment is shown. In B–D, the values presented are
means 6SEM (n= 4–5). In E, the values are means 6SEM (n= 2–3). The
experiments were repeated at least three times. *p,0.05 in comparison
with the cells treated with nontargeting siRNA. For statistical analysis, the
Mann–Whitney Utest was applied (B,C). SMC, smooth muscle cell.
FIGURE 3. The proinflammatory responses activated by mast cells upon
exposure to necrotic cell supernatant are mediated by IL-33. After expo-
sure for 24 h to supernatant from the same number (1:1) or twice as many
(2:1) necrotic MEFs, secretion of cytokines IL-6 and TNF-a(A) by wild-
type and T1/ST2
2/2
BMMCs were measured. After exposure for 30 min,
release of cysteinyl leukotriene and LTB
4
(B) by these same BMMCs was
determined. IL-33 protein was investigated in necrotic cell supernatant and
compared with rIL-33 using Western blotting (C). D, Release of IL-6,
TNF-a, and cysteinyl leukotrienes were measured in BMMCs treated with
0, 10, or 100 ng/ml IL-33. During BMMC treatment, 10
6
cells/ml were
used in a total volume of 200 ml. The values shown are means 6SEM (n=
5–6). For statistical analyses, Mann–Whitney (A,B) and one-way ANOVA
(D) were applied. The experiments were repeated at least three times. *p,
0.05, **p,0.01, ***p,0.001 in comparison with wild-type cells (A,B)
or untreated cells (D).
2526 MAST CELLS AS SENSORS OF CELL INJURY
on June 8, 2011www.jimmunol.orgDownloaded from
which it is was demonstrated that necrotic HMGB1
2/2
cells initi-
ated inflammation similarly to necrotic HMGB1
+/+
cells in mice.
Additionally, in our experimental system, necrotic supernatant
from MEFs with inhibited uric acid formation did not significantly
reduce the proinflammatory responses in BMMCs, suggesting
other components than HMGB1 or uric acid to be responsible for
the observed mast cell activation.
Supported by the finding that MyD88
2/2
BMMCs were unable
to secrete IL-6 in response to necrotic supernatant, we also in-
vestigated responses to necrotic supernatant by TLR-deficient
mast cells. Although TLRs are best known for recognizing ex-
ogenous danger signals like LPS, some members of the TLR
family have previously been reported to respond to endogenous
ligands. For instance, recognition of heat shock protein 60 has
been suggested to be TLR4 dependent (44). However, TLR-
deficient BMMCs secreted IL-6 at similar levels as wild-type
BMMCs when exposed to necrotic cell supernatant, indicating
another receptor upstream of MyD88 to be involved. This does
not, however, exclude the possibility that the necrotic cell super-
natant affects multiple TLR signaling pathways. However, as
responses to necrotic supernatant were completely abrogated in
T1/ST2
2/2
BMMCs, we consider it unlikely that necrotic super-
natant might affect mast cells by activating a combination of
TLRs.
The discovery that T1/ST2
2/2
BMMCs displayed abrogated
IL-6, TNF-a, cysteinyl leukotrienes, and LTB
4
secretion in re-
sponse to necrotic supernatant (Fig. 3A,3B) led us to speculate
that IL-33, the only known T1/ST2 ligand, might be involved in
the mast cell’s ability to respond to necrotic supernatant. In agree-
ment with our hypothesis that IL-33 is a principal danger signal
released from necrotic cells, it has previously been shown that
IL-33 is released upon necrosis rather than during apoptosis (7,
24). Importantly, IL-33 does not seem to require proteolysis for
activation, a favorable feature indeed for an alarmin. Lu
¨thi et al.
(24) also showed that IL-33 was proteolytically cleaved during
apoptosis, thus diminishing its bioactivity. Most importantly, only
small amounts of IL-33 were released during apoptosis, whereas
induction of necrosis led to IL-33 release (24). In line with this,
we show in Fig. 3Cthat necrotic MEFs release full-length pro–
IL-33 as well as cleaved IL-33. The fact that a much higher
concentration of rIL-33 was required to elicit IL-6 secretion
by BMMCs compared with the concentration we detected in
necrotic supernatant by ELISA might thus suggest that pro–IL-33
is more active compared with rIL-33.
It has been shown that endothelial cells subjected to mechanical
wounding by cell scraping or freeze-thawing also release IL-33 (7).
To confirm that IL-33 was the chief component of the necrotic
supernatant activating mast cells in our system, we generated
necrotic supernatant from MEFs with siRNA-silenced IL-33 ex-
pression. In line with our earlier results suggesting IL-33 to be
responsible for activating the BMMCs, necrotic supernatant gen-
erated from MEFs with silenced IL-33 expression failed to induce
IL-6 secretion by BMMCs. Amazingly, IL-33 alone, in a soup of
components released by necrotic cells, thus has the capability to
alone potently activate mast cells. In support of these results, as
shown by us in this paper and also by several earlier studies, IL-33
can induce secretion of IL-6 as well as other cytokines in mast
cells (17–20). In addition to confirming that IL-33 can induce
cytokine release by BMMCs, we also show, to our knowledge, for
the first time that IL-33 can induce release of cysteinyl leuko-
trienes by mast cells.
Taken together, our findings provide an important link between
studies revealing that IL-33 is released upon necrosis and studies
demonstrating that mast cells are activated by IL-33 to secrete
cytokines. Hence, to our knowledge, our findings provide, for the
first time, a plausible mechanism for how mast cells might function
as sensors of cell injury. Most remarkably, of all potential acti-
vators present in the necrotic cell supernatant, our results show IL-
33 to be the sole inducer of cytokine secretion by mast cells, in-
disputably demonstrating that IL-33 is an important alarmin.
Expression of IL-33 has been described in a variety of tissues
(15) and cell types; for instance, in fibroblasts (45), epithelial
cells, and endothelial cells (8). Therefore, we addressed the ques-
tion whether necrotic supernatant generated from other cell types
than MEFs would elicit similar responses in BMMCs. These
experiments revealed that necrotic supernatant generated from
structural cell types (keratinocytes, smooth muscle cells, and
mixed neuronal cells) induced IL-6 secretion in BMMCs, whereas
necrotic supernatant from cells of hematopoietic origin (BMMCs,
monocytes or splenocytes) did not. Moreover, we show that ne-
crotic supernatant from the former cell types correspondingly also
contain higher IL-33 levels, in agreement with previous reports
(15, 46). Theoretically, should a structural cell die by necrosis,
adjacent mast cells will respond to released IL-33 and initiate an
inflammatory response.
In summary, using an in vitro model of cell injury, we provide
a mechanism for the hypothesis that mast cells are important
sensors of cell injury. The present findings indicate that structural
cells, such as fibroblasts and keratinocytes, release IL-33 upon
injury and that adjacent mast cells respond by producing proin-
flammatory factors, including IL-6, TNF-a, and leukotrienes. Sub-
sequently, these signals can induce vascular changes, includ-
ing vasodilatation, increased permeability of the microvasculature,
and recruitment of inflammatory cells to the site of injury. Thus,
our investigation provides support for the relatively early hy-
pothesis that an important physiological function for mast cells is
to act as key sensors of endogenous danger signals released upon
cell injury. Moreover, the present demonstration that this response
involves recognition of IL-33 by T1/ST2 receptors on the surface
of mast cells describes a novel mechanism underlying the role of
mast cells as sensors. At the same time, new evidence com-
plementing earlier studies suggesting that IL-33 is a key alarmin
released by dying cells is also provided. The responses of mast
cells to cell injury are likely to play a highly important role in the
initiation of acute inflammation and subsequent healing and,
thereby, in the maintenance of tissue homeostasis.
Acknowledgments
We thank Drs. Shizuo Akira (Department of Host Defense, Osaka Univer-
sity, Japan), Andrew McKenzie (Medical Research Council, Laboratory of
Molecular Biology, Cambridge, U.K.), Marco Bianchi (Department of Bio-
technology, San Raffaele Scientific Institute, Milano, Italy), Bertil Fred-
holm (Department of Physiology and Pharmacology, Karolinska Institute,
Stockholm, Sweden), Tobias Bergstro
¨m (Department of Medical Biochem-
istry and Microbiology, Uppsala University, Sweden), and Mikael Adner
(Department of Environmental Medicine, Karolinska Institute) for provid-
ing reagents necessary for these experiments. We also thank Dr. Sara Lind
(Department of Medicine, Karolinska Institute) for technical assistance.
Disclosures
The authors have no financial conflicts of interest.
References
1. Basu, S., R. J. Binder, R. Suto, K. M. Anderson, and P. K. Srivastava. 2000.
Necrotic but not apoptotic cell death releases heat shock proteins, which deliver
a partial maturation signal to dendritic cells and activate the NF-kappa B
pathway. Int. Immunol. 12: 1539–1546.
2. Bianchi, M. E. 2007. DAMPs, PAMPs and alarmins: all we need to know about
danger. J. Leukoc. Biol. 81: 1–5.
The Journal of Immunology 2527
on June 8, 2011www.jimmunol.orgDownloaded from
3. Scaffidi, P., T. Misteli, and M. E. Bianchi. 2002. Release of chromatin protein
HMGB1 by necrotic cells triggers inflammation. Nature 418: 191–195.
4. Shi, Y., J. E. Evans, and K. L. Rock. 2003. Molecular identification of a danger
signal that alerts the immune system to dying cells. Nature 425: 516–521.
5. Chen, C. J., H. Kono, D. Golenbock, G. Reed, S. Akira, and K. L. Rock. 2007.
Identification of a key pathway required for the sterile inflammatory response
triggered by dying cells. Nat. Med. 13: 851–856.
6. Eigenbrod, T., J. H. Park, J. Harder, Y. Iwakura, and G. Nu
´n
˜ez. 2008. Cutting
edge: critical role for mesothelial cells in necrosis-induced inflammation through
the recognition of IL-1 alpha released from dying cells. J. Immunol. 181: 8194–
8198.
7. Cayrol, C., and J. P. Girard. 2009. The IL-1-like cytokine IL-33 is inactivated
after maturation by caspase-1. Proc. Natl. Acad. Sci. USA. 106: 9021–9026.
8. Moussion, C., N. Ortega, and J. P. Girard. 2008. 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 3: e3331.
9. Kono, H., and K. L. Rock. 2008. How dying cells alert the immune system to
danger. Nat. Rev. Immunol. 8: 279–289.
10. Nathan, C. 2002. Points of control in inflammation. Nature 420: 846–852.
11. Marshall, J. S. 2004. Mast-cell responses to pathogens. Nat. Rev. Immunol. 4:
787–799.
12. Metcalfe, D. D., D. Baram, and Y. A. Mekori. 1997. Mast cells. Physiol. Rev. 77:
1033–1079.
13. Galli, S. J., J. Kalesnikoff, M. A. Grimbaldeston, A. M. Piliponsky,
C. M. Williams, and M. Tsai. 2005. Mast cells as “tunable” effector and im-
munoregulatory cells: recent advances. Annu. Rev. Immunol. 23: 749–786.
14. Abraham, S. N., and A. L. St John. 2010. Mast cell-orchestrated immunity to
pathogens. Nat. Rev. Immunol. 10: 440–452.
15. Schmitz, J., A. Owyang, E. Oldham, Y. Song, E. Murphy, T. K. McClanahan,
G. Zurawski, M. Moshrefi, J. Qin, X. Li, et al. 2005. 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 23: 479–490.
16. Ali, S., M. Huber, C. Kollewe, S. C. Bischoff, W. Falk, and M. U. Martin. 2007.
IL-1 receptor accessory protein is essential for IL-33-induced activation of
T lymphocytes and mast cells. Proc. Natl. Acad. Sci. USA 104: 18660–18665.
17. Ho, L. H., T. Ohno, K. Oboki, N. Kajiwara, H. Suto, M. Iikura, Y. Okayama,
S. Akira, H. Saito, S. J. Galli, and S. Nakae. 2007. IL-33 induces IL-13 pro-
duction by mouse mast cells independently of IgE-FcepsilonRI signals. J. Leu-
koc. Biol. 82: 1481–1490.
18. Moulin, D., O. Donze
´, D. Talabot-Ayer, F. Me
´zin, G. Palmer, and C. Gabay.
2007. Interleukin (IL)-33 induces the release of pro-inflammatory mediators by
mast cells. Cytokine 40: 216–225.
19. Talabot-Ayer, D., C. Lamacchia, C. Gabay, and G. Palmer. 2009. Interleukin-33
is biologically active independently of caspase-1 cleavage. J. Biol. Chem. 284:
19420–19426.
20. Xu, D., H. R. Jiang, P. Kewin, Y. Li, R. Mu, A. R. Fraser, N. Pitman,
M. Kurowska-Stolarska, A. N. McKenzie, I. B. McInnes, and F. Y. Liew. 2008.
IL-33 exacerbates antigen-induced arthritis by activating mast cells. Proc. Natl.
Acad. Sci. USA 105: 10913–10918.
21. Allakhverdi, Z., D. E. Smith, M. R. Comeau, and G. Delespesse. 2007. Cutting
edge: The ST2 ligand IL-33 potently activates and drives maturation of human
mast cells. J. Immunol. 179: 2051–2054.
22. Iikura, M., H. Suto, N. Kajiwara, K. Oboki, T. Ohno, Y. Okayama, H. Saito,
S. J. Galli, and S. Nakae. 2007. IL-33 can promote survival, adhesion and cy-
tokine production in human mast cells. Lab. Invest. 87: 971–978.
23. Kaieda, S., K. Shin, P. A. Nigrovic, K. Seki, R. T. Lee, R. L. Stevens, and
D. M. Lee. 2010. Synovial fibroblasts promote the expression and granule ac-
cumulation of tryptase via interleukin-33 and its receptor ST-2 (IL1RL1). J. Biol.
Chem. 285: 21478–21486.
24. Lu
¨thi, A. U., S. P. Cullen, E. A. McNeela, P. J. Duriez, I. S. Afonina,
C. Sheridan, G. Brumatti, R. C. Taylor, K. Kersse, P. Vandenabeele, et al. 2009.
Suppression of interleukin-33 bioactivity through proteolysis by apoptotic cas-
pases. Immunity 31: 84–98.
25. West, G. B. 1958. Pharmacology of the tissue mast cell. Br. J. Dermatol. 70:
409–417.
26. Adachi, O., T. Kawai, K. Takeda, M. Matsumoto, H. Tsutsui, M. Sakagami,
K. Nakanishi, and S. Akira. 1998. Targeted disruption of the MyD88 gene results
in loss of IL-1- and IL-18-mediated function. Immunity 9: 143–150.
27. Takeuchi, O., S. Sato, T. Horiuchi, K. Hoshino, K. Takeda, Z. Dong,
R. L. Modlin, and S. Akira. 2002. Cutting edge: role of Toll-like receptor 1 in
mediating immune response to microbial lipoproteins. J. Immunol. 169: 10–14.
28. Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda,
and S. Akira. 1999. Differential roles of TLR2 and TLR4 in recognition of gram-
negative and gram-positive bacterial cell wall components. Immunity 11: 443–
451.
29. Hoshino, K., O. Takeuchi, T. Kawai, H. Sanjo, T. Ogawa, Y. Takeda, K. Takeda,
and S. Akira. 1999. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are
hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene
product. J. Immunol. 162: 3749–3752.
30. Uematsu, S., M. H. Jang, N. Chevrier, Z. Guo, Y. Kumagai, M. Yamamoto,
H. Kato, N. Sougawa, H. Matsui, H. Kuwata, et al. 2006. Detection of patho-
genic intestinal bacteria by Toll-like receptor 5 on intestinal CD11c+ lamina
propria cells. Nat. Immunol. 7: 868–874.
31. Takeuchi, O., T. Kawai, P. F. Mu
¨hlradt, M. Morr, J. D. Radolf, A. Zychlinsky,
K. Takeda, and S. Akira. 2001. Discrimination of bacterial lipoproteins by Toll-
like receptor 6. Int. Immunol. 13: 933–940.
32. Hemmi, H., T. Kaisho, O. Takeuchi, S. Sato, H. Sanjo, K. Hoshino, T. Horiuchi,
H. Tomizawa, K. Takeda, and S. Akira. 2002. Small anti-viral compounds ac-
tivate immune cells via the TLR7 MyD88-dependent signaling pathway. Nat.
Immunol. 3: 196–200.
33. Valenzuela, D. M., A. J. Murphy, D. Frendewey, N. W. Gale, A. N. Economides,
W. Auerbach, W. T. Poueymirou, N. C. Adams, J. Rojas, J. Yasenchak, et al.
2003. High-throughput engineering of the mouse genome coupled with high-
resolution expression analysis. Nat. Biotechnol. 21: 652–659.
34. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto,
K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptor
recognizes bacterial DNA. Nature 408: 740–745.
35. Ledent, C., J. M. Vaugeois, S. N. Schiffmann, T. Pedrazzini, M. El Yacoubi,
J. J. Vanderhaeghen, J. Costentin, J. K. Heath, G. Vassart, and M. Parmentier.
1997. Aggressiveness, hypoalgesia and high blood pressure in mice lacking the
adenosine A2a receptor. Nature 388: 674–678.
36. Salvatore, C. A., S. L. Tilley, A. M. Latour, D. S. Fletcher, B. H. Koller, and
M. A. Jacobson. 2000. Disruption of the A(3) adenosine receptor gene in mice
and its effect on stimulated inflammatory cells. J. Biol. Chem. 275: 4429–4434.
37. Townsend, M. J., P. G. Fallon, D. J. Matthews, H. E. Jolin, and A. N. McKenzie.
2000. T1/ST2-deficient mice demonstrate the importance of T1/ST2 in de-
veloping primary T helper cell type 2 responses. J. Exp. Med. 191: 1069–1076.
38. Selander, C., C. Engblom, G. Nilsson, A. Scheynius, and C. L. Andersson. 2009.
TLR2/MyD88-dependent and -independent activation of mast cell IgE responses
by the skin commensal yeast Malassezia sympodialis. J. Immunol. 182: 4208–
4216.
39. Andersson, U., H. Wang, K. Palmblad, A. C. Aveberger, O. Bloom,
H. Erlandsson-Harris, A. Janson, R. Kokkola, M. Zhang, H. Yang, and
K. J. Tracey. 2000. High mobility group 1 protein (HMG-1) stimulates proin-
flammatory cytokine synthesis in human monocytes. J. Exp. Med. 192: 565–570.
40. Calogero, S., F. Grassi, A. Aguzzi, T. Voigtla
¨nder, P. Ferrier, S. Ferrari, and
M. E. Bianchi. 1999. The lack of chromosomal protein Hmg1 does not disrupt
cell growth but causes lethal hypoglycaemia in newborn mice. Nat. Genet. 22:
276–280.
41. Midwood, K., S. Sacre, A. M. Piccinini, J. Inglis, A. Trebaul, E. Chan,
S. Drexler, N. Sofat, M. Kashiwagi, G. Orend, et al. 2009. Tenascin-C is an
endogenous activator of Toll-like receptor 4 that is essential for maintaining
inflammation in arthritic joint disease. Nat. Med. 15: 774–780.
42. Okamura, Y., M. Watari, E. S. Jerud, D. W. Young, S. T. Ishizaka, J. Rose,
J. C. Chow, and J. F. Strauss, III. 2001. The extra domain A of fibronectin
activates Toll-like receptor 4. J. Biol. Chem. 276: 10229–10233.
43. Yanagisawa, K., T. Takagi, T. Tsukamoto, T. Tetsuka, and S. Tominaga. 1993.
Presence of a novel primary response gene ST2L, encoding a product highly
similar to the interleukin 1 receptor type 1. FEBS Lett. 318: 83–87.
44. Ohashi, K., V. Burkart, S. Flohe
´, and H. Kolb. 2000. Cutting edge: heat shock
protein 60 is a putative endogenous ligand of the toll-like receptor-4 complex. J.
Immunol. 164: 558–561.
45. Sanada, S., D. Hakuno, L. J. Higgins, E. R. Schreiter, A. N. McKenzie, and
R. T. Lee. 2007. IL-33 and ST2 comprise a critical biomechanically induced and
cardioprotective signaling system. J. Clin. Invest. 117: 1538–1549.
46. Hudson, C. A., G. P. Christophi, R. C. Gruber, J. R. Wilmore, D. A. Lawrence,
and P. T. Massa. 2008. Induction of IL-33 expression and activity in central
nervous system glia. J. Leukoc. Biol. 84: 631–643.
2528 MAST CELLS AS SENSORS OF CELL INJURY
on June 8, 2011www.jimmunol.orgDownloaded from
