© 2006 Nature Publishing Group
Card9 controls a non-TLR signalling
pathway for innate anti-fungal immunity
Olaf Gross1, Andreas Gewies1, Katrin Finger1, Martin Scha ¨fer2, Tim Sparwasser2, Christian Peschel1,
Irmgard Fo ¨rster2† & Ju ¨rgen Ruland1
Fungal infections are increasing worldwide due to the marked rise in immunodeficiencies including AIDS; however,
immune responses to fungi are poorly understood. Dectin-1 is the major mammalian pattern recognition receptor for the
fungal component zymosan. Dectin-1 represents the prototype of innate non-Toll-like receptors (TLRs) containing
immunoreceptor tyrosine-based activation motifs (ITAMs) related to those of adaptive antigen receptors. Here we
identify Card9 as a key transducer of Dectin-1 signalling. Although being dispensable for TLR/MyD88-induced
responses, Card9 controls Dectin-1-mediated myeloid cell activation, cytokine production and innate anti-fungal
immunity. Card9 couples to Bcl10 and regulates Bcl10–Malt1-mediated NF-kB activation induced by zymosan. Yet, Card9
is dispensable for antigen receptor signalling that uses Carma1 as a link to Bcl10–Malt1. Thus, our results define a novel
innate immune pathway and indicate that evolutionarily distinct ITAM receptors in innate and adaptive immune cells use
diverse adaptor proteins to engage selectively the conserved Bcl10–Malt1 module.
Pathogen recognition by immune cell receptors is essential for host
defence against microorganisms. Whereas the highly diverse T- and
B-cell receptors (TCRs and BCRs) for adaptive immunity are
generated through somatic rearrangement of antigen receptor
genes, the evolutionarily ancient pattern recognition receptors
(PRRs) of the innate immune system are germline encoded and
control the first line of defence. These PRRs include the archetypical
TLRs as well as non-TLRs like intracellular NOD proteins, lectin
receptors and others1–3. Upon ligand binding, both innate and
adaptive immunoreceptors engage intracellular signalling pathways
that converge on conserved core transcription factors for cell acti-
vation. To mediate specificity, each receptor system uses distinct sets
of signalling molecules that relay proximal events to downstream
effectors. Although substantial progress has beenmade in deciphering
engaged by non-TLR PRRs remain poorly defined.
One key transcription factor for both innate and adaptive immu-
nity is NF-kB5. Almost all TLRs signal via MyD88 for NF-kB
activation and subsequent inflammatory cytokine production1. In
contrast, the T- and B-cell antigen receptors activate proximal Src
and Syk tyrosine kinases and then engage the caspase recruitment
domain (CARD) proteins Carma1 and Bcl10 fordownstream signal-
ling. Together with Malt1, the Carma1–Bcl10 complex mediates
protein kinase C (PKC)-controlled NF-kB induction for lymphocyte
activation and differentiation6. The poorly characterized Card9
protein is structurally related to Carma1 and is expressed in a variety
of tissues including peripheral blood lymphocytes and the spleen7.
Similar to Carma1 and the Carma family members 2 and 3, Card9
coil region. However, Card9 lacks the carboxy-terminal MAGUK
domain that characterizes Carma proteins and the Carma linker
region that regulates PKC-dependent cell activation8,9. As such,
Card9 has been suggested to have a modulatory role in the
Carma1/Bcl10-dependent pathway of lymphocyte differentiation
and activation, possibly by interfering with the Carma1–Bcl10
interaction10. Here we report that Card9 is surprisingly not involved
in TCR or BCR signalling. Instead, Card9 operates as a key adaptor
for non-TLR PRR signal transduction, which is required to link
for innate anti-fungal immunity.
Normal adaptive immunity without Card9
To investigate the functions of Card9 we generated Card9-deficient
(Card92/2) mice (Supplementary Fig. 1) and first analysed the
requirement of Card9 for adaptive immunity. Flow cytometric
analysis revealed normal development of all tested Card92/2
T- and B-cell subsets including regulatory T cells, recirculating B
cells, mature follicular and marginal zone B cells as well as B1 B cells
(Supplementary Fig. 2 and data not shown). Furthermore, no
significant differences in basal immunoglobulin levels were detected
with either the T-cell-independent antigen 2,4,6-trinitrophenyl
(TNP) conjugated to Ficoll or the T-cell-dependent antigen
4-hydroxy-3-nitrophenylacetyl (NP) conjugated to ovalbumin we
detected normal IgM and IgG responses in Card92/2animals
(Fig. 1b). To study directly antigen-receptor-mediated lymphocyte
activation, we stimulated purified T and B cells through their
ligation or with phorbol ester andcalcium ionophore(Iono,Fig.1c).
Normal viability (data not shown) and normal signal-induced
proliferation was detected in Card92/2lymphocytes. We also
found normal TCR- or BCR-induced NF-kB activation in Card92/2
critically involved in the Carma1–Bcl10 signalling pathway that
mediates T- and B-cell differentiation and activation for adaptive
1III. Medizinische Klinik, Klinikum rechts der Isar, Technische Universita ¨t Mu ¨nchen, Ismaninger Str. 22, 81675 Munich, Germany.2Institut fu ¨r Medizinische Mikrobiologie,
Immunologie und Hygiene, Technische Universita ¨t Mu ¨nchen, Trogerstr. 30, 81675 Munich, Germany. †Present address: Institut fu ¨r Umweltmedizinische Forschung, University of
Du ¨sseldorf, Auf’m Hennekamp 50, 40225 Du ¨sseldorf, Germany.
© 2006 Nature Publishing Group
Card9 controls zymosan-induced cell activation
Because Card9 is also expressed in myeloid cells7(Fig. 2a) we
we generated bone-marrow-derived dendritic cells (BMDCs) from
as ligands for non-TLR PRRs(Fig. 2b) and measured ligand-induced
production of tumour necrosis factor-a (TNF-a). No reproducible
differences were detectedbetweenwild type and Card92/2BMDCs in
response to the TLR ligands Pam3CSK4(TLR2), lipopolysaccharide
(LPS; TLR4), R848 (TLR7), CpG DNA (TLR9) or upon stimulation
with the Nod2 agonist muramyl dipeptide (MDP) alone or in
combination with LPS. In contrast, TNF-a production induced by
zymosan is severely defective in the absence of Card9.
Zymosan is a major yeast cell wall component that is principally
composed of b-glucans and additionally contains a-mannan and
mannoproteins11. The main mammalian PRR for zymosan is the
b-glucan receptor Dectin-1 (ref. 12), yet zymosan also co-stimulates
TLR2 (ref. 13). To characterize further the requirement of Card9 in
zymosan-induced dendritic cell activation, we obtained dose-
response curves for TNF-a, interleukin (IL)-6 and IL-2 production.
A strong Card9 gene dosage dependency was observed for the
production of all three cytokines (Fig. 2c). Comparable results
were obtained with zymosan preparations from two independent
sources (data not shown). In contrast, LPS stimulation induced
regular dose-dependent TNF-a and IL-6 synthesis in Card92/2
BMDCs (Fig. 2d), whereas LPS failed to induce IL-2 production in
both Card92/2and control cells (data not shown).
Itis believed that the recognition of zymosan or b-glucans triggers
immune responses that are primarily designed for the control of
fungal pathogens14. Therefore, we next studied the role of Card9 in
immune responses to whole fungal cells using Candida albicans as a
model15. Consistent with a role of Card9 in zymosan recognition
signalling, Card92/2BMDCs exhibit severe defects in C. albicans-
relevance of these findings in vivo we injected live C. albicans into
Card92/2mice and heterozygous littermates and compared their
Figure 1 | Regular adaptive immunity in Card9-deficient mice. a, Basal
immunoglobulin concentrations. Immunoglobulin isotypes in sera of
8–12-week-old wild type (þ/þ, n ¼ 9) and Card92/2(n ¼ 18) mice are
shown. b, Regular antibody responses. Left: control (þ/2, filled symbols,
n ¼ 8) and Card92/2mice (open symbols, n ¼ 6) were immunized with
TNP-Ficoll for T-cell-independent immune responses. Concentrations of
TNP-specific IgM and IgG3 were determined before and after
immunization. Right: for T-cell-dependent antibody responses, control
(þ/2, filled symbols, n ¼ 4) and Card92/2mice (open symbols, n ¼ 3)
were injected with NP-Ova absorbed on alum. NP-specific IgM and IgG3
concentrations were measured before and after immunization. Results are
representative for two independent experiments. c, Signal-induced
lymphocyte proliferation. CD4þT cells and B220þB cells were stimulated
for 48h with agonistic antibodies (anti-IgM or anti-CD3, as indicated; anti-
CD28, 2mgml21; anti-CD40, 5mgml21) or PMA plus calcium ionophore
(Iono) (10ngml21each) and assessed for proliferation by measuring
[3H]thymidine incorporation. Data are representative of four independent
experiments with a total of n ¼ 18 Card92/2and n ¼ 18 control mice.
Indicated values are means ^ s.d. of triplicates.
Figure 2 | Impaired zymosan-induced cytokine production in Card92/2
BMDCs. a, RT–PCR analysis of Card9 expression in BMDCs (DC) and
BMDMs (MP) from Card2/2mice. b, TNF-a production in BMDCs from
Card9þ/2or Card92/2mice that were left unstimulated (Medium) or
stimulated with TLR ligands Pam3CSK4(200ngml21), LPS (100ngml21),
for 24h. ND, not detected. Data are means ^ s.d. of triplicates and
representative of three independent experiments. c, Dose-dependent
TNF-a, IL-6 or IL-2 production from Card9þ/þ, Card9þ/2or Card92/2
BMDCs after 24h of zymosan stimulation. Data are means ^ s.d. of
triplicate samples and representative of three independent experiments.
d, Dose-dependent TNF-a or IL-6 production from Card9þ/2or Card92/2
BMDCs after 24h of LPS stimulation. Data are means ^ s.d. of triplicate
samples and representative of three independent experiments.
© 2006 Nature Publishing Group
susceptibility to infection (Fig. 3b). All Card9-deficient animals died
within 5days after infection whereas more than 50% of the control
mice survived for more than 12days. In another set of experiments
we killed the animals 4days after injection and assessed intravital
C. albicans growth. It was macroscopically apparent that the kidneys
of all Card92/2animals were massively infiltrated by the fungus
load of C. albicans in the kidneys and livers of Card92/2mice and
additionally observed high C. albicans titres in their lungs (Fig. 3d).
In contrast, Card92/2mice regularly cleared the bacterial pathogen
Staphylococcus aureus (Fig. 3e). Thus, Card9 is critically required for
zymosan-induced dendritic cell activation and innate anti-fungal
Dectin-1/Syk signalling depends on Card9
activation12,13,16–18. Dectin-1 represents the prototype of non-TLR
PRRs that contain ITAMs related to those of antigen receptors or Fc
receptors19. Zymosan-induced Dectin-1 signalling depends on an
intact ITAM and downstream activation of Syk, whereas zymosan-
induced TLR signalling is transduced via MyD88 (ref. 17). To
investigate selectively the involvement of Card9 in the Dectin-1-
independent TLR2 pathway, we used two independent pure TLR2
agonists, Pam3CSK4and peptidoglycan (PGN), for BMDC stimu-
lation20,21. Consistent with the normal TLR2-dependent clearance of
S. aureus22in Card92/2mice, both ligands induced normal TNF-a
and IL-6 production (Fig. 4a and data not shown) in the absence of
Because there is no single agonistic Dectin-1 ligand available, it is
more intricate to study Dectin-1/Syk signalling without considering
TLR2 co-stimulation. However, recent work has shown that the
Dectin-1/Syk axis selectively controls zymosan-induced IL-10 pro-
duction entirely independently of TLR/MyD88 signalling17. We thus
usedzymosan-inducedIL-10productionasaspecific readout for the
Dectin-1/Syk pathway. In addition to Card92/2and wild-type
BMDCs we included BMDCs from Myd882/2mice23as well as
wild-type BMDCs that were pre-incubated with the small molecule
Syk inhibitor piceatannol as selective controls for TLR/MyD88 and
Dectin-1/Syk signalling (Fig. 4b). Consistent with published data17
zymosan induces strong IL-10 production in wild-type BMDCs that
is MyD88-independent but robustly blocked by inhibiting Syk.
Deletion of Card9 also completely abolishes zymosan-induced
IL-10 synthesis, indicating a function for Card9 in Dectin-1/Syk-
dependent, TLR/MyD88-independent events. Compared to zymo-
of IL-10 in Card92/2BMDCs (Fig. 4c). We next studied zymosan-
induced production of TNF-a, IL-6, IL-2 and IL-12 in wild-type
BMDCs with or without Syk inhibitor and in Card92/2BMDCs
(Fig. 4d). Zymosan-induced production of TNF-a and IL-2 is highly
dependent on Syk and Card9. However, consistent with previous
data the production of IL-12 (ref. 17) and IL-6 is at least in part
independent of Syk and also independent of Card9 function. Thus,
the effects of Syk inhibition or Card9 deletion on zymosan-induced
cytokine production are in each case comparable, further suggesting
an essential function of Card9 in Dectin-1/Syk signalling.
Card9 relays Dectin-1/Syk signals to NF-kB
To identify the molecular role of Card9 in Dectin-1/Syk-mediated
cell activation we determined Dectin-1 expression and Dectin-1-
triggered Syk phosphorylation in Card92/2BMDCs. The receptor is
regularly expressed (data not shown) and normally activates Syk
also regulates zymosan phagocytosis17. To quantify zymosan internal-
ization we incubated BMDCs with or without Syk inhibitor together
with fluorescently labelled zymosan (Fig. 4f). Using fluorescence-
activated cell sorting (FACS), we observed normal Syk-dependent
ways controlled by Dectin-1/Syk do not depend on Card9.
Because overexpression studies have indicated that Card9 can
induce NF-kB activation7, we probed NF-kB activation by monitor-
ing RelA nuclear translocation in zymosan- or LPS-stimulated
BMDCs (Fig. 4g). TLR4 stimulation induced normal RelA nuclear
translocation in Card92/2BMDCs. Microscopy also demonstrated
that Card92/2BMDCs internalized zymosan normally. However,
these cells exhibit a substantial impairment in zymosan-induced
TLR2 could be observed. In addition, we performed electromobility
shift assays using nuclear extracts from bone-marrow-derived
macrophages (BMDMs) that were stimulated with zymosan or
PGN, LPS or R848 (Fig. 4h). Whereas TLR ligation induced normal
NF-kB DNA binding, zymosan-induced NF-kB activation was
diminished in the absence of Card9. Thus, Card9 is specifically
required to transduce zymosan-activated Dectin-1 signals to NF-kB.
Card9 signals via Bcl10 and Malt1
As Card9 can interact with Bcl10 (ref.7), we studied whether the two
proteins functionally cooperate (Fig. 5). Consistent with previous
results7single overexpression of Card9 or Bcl10 weakly activates
NF-kB. However, co-overexpression of both proteins induced more
than additive NF-kB induction, indicating synergism (Fig. 5a). To
establish a hierarchy between Card9 and Bcl10 we introduced Card9
into wild type or Bcl10-deficient murine embryonic fibroblasts24
(Fig. 5b). In parallel we overexpressed the Bcl10 upstream activator
Figure 3 | Impaired immune responses to Candida albicans. a, C. albicans-
induced TNF-a, IL-6 or IL-2 production in Card92/2BMDCs. Cells were
stimulated for 24h. Results are means ^ s.d. of triplicates. b, Card9þ/2and
Card92/2mice were infected intravenously with 1.5 £ 105colony-forming
units (c.f.u.) C. albicans and monitored daily for survival. One out of three
infected intravenously with 3 £ 104c.f.u. C. albicans. Four days later the left
kidneys were inspected (c) and C. albicans titres were determined in one
kidney, the liver and the lung (d). e, Card9þ/2and Card92/2mice were
infected intravenously with 5 £ 105c.f.u. S. aureus. Five days later S. aureus
titres were determined in the spleen.
© 2006 Nature Publishing Group
Carma3 (ref. 25) and the Bcl10-independent NF-kB inducer Nod1.
Nod1 induced equal NF-kB activation in wild type and Bcl102/2
cells, whereas Carma3 activated NF-kB only in Bcl10þ/2cells.
Similarly, Card9 induced NF-kB activation only in wild-type cells,
indicating a requirement for Bcl10 in Card9-mediated NF-kB signal-
ling. To investigate directly the role of a Card9–Bcl10 complex in
Dectin-1 signalling, we introduced Dectin-1 and a combination of
low levels of endogenous Syk26) and stimulated these with zymosan
(Fig. 5c). Overexpression of Dectin-1 with or without zymosan
treatment did not induce NF-kB activity. In addition, zymosan
treatment did not increase the Card9–Bcl10-induced NF-kB lucifer-
ase signal without Dectin-1 expression. However, co-expression of
Dectin-1 and Card9–Bcl10 conferred 293 cells with the ability to
respond to zymosan, resulting in a substantial additional increase in
NF-kB reporter activity upon zymosan treatment.
Syk pathway, we stimulated BMDCs from Bcl102/2mice24
with zymosan or TLR ligands. Because signals downstream of
Bcl10 can involve Malt1 we also examined BMDCs from Malt12/2
animals27. Bcl10- and Malt1-deficient BMDCs produced normal
amounts of TNF-a in response to PGN, LPS and CpG (Fig. 5d). In
addition, zymosan phagocytosis was not affected by Bcl10 or Malt1
deficiency (data not shown). However, zymosan-induced cytokine
production was severely defective in both Bcl102/2and Malt12/2
BMDCs (Fig. 5e). Next, we stimulated Bcl102/2and Malt12/2
BMDCs with whole fungal cells of C. albicans (Supplementary
Fig. 3a). Severe defects in TNF-a, IL-2 and IL-10 production were
detected. Finally, we investigated the roles of Bcl10 and Malt1 in
zymosan-induced NF-kB signalling (Supplementary Fig. 3b). As
before, zymosan induced robust nuclear translocation of RelA in
wild-type cells. However, similar to Card92/2BMDCs, both Bcl10-
and Malt1-deficient BMDCs showed an impairment in zymosan-
induced NF-kB activation, whereas they exhibited normal nuclear
translocation of RelA upon LPS stimulation.
Bcl10–Malt1 signalling in innate and adaptive immunity
Collectively,our results identify anovel signalling pathway for innate
immunity. Whereas Card9 is dispensable for adaptive antigen recep-
tor signalling as well as for Bcl10-dependent Fc1RI signalling28(not
shown), our data support the hypothesis that Card9 operates
upstream of Bcl10 to transduce, together with Bcl10 and Malt1,
non-TLR signals for NF-kB activation and cytokine production. For
a model see Fig. 5f. In contrast to immunity directed against viruses
and bacteria, far less is known about innate responses to fungal
infections. Yet, the incidence of these infections is rapidly growing
with theworldwide increaseinthenumberofimmunocompromised
patients suffering from AIDS or undergoing immunosuppressive
therapies. To potentially manipulate anti-fungal immunity it is
necessary to understand the pathways controlling these responses.
Dectin-1 was recently discovered as the key non-TLR PRR for fungal
b-glucan detection19. This receptor uses an intracellular YxxL ITAM
Figure 4 | Impaired Dectin-1/Syk signalling in Card92/2cells. a, Normal
TLR2-induced TNF-a production. BMDCs from Card9þ/2or Card92/2
mice were stimulated for 24h with the selective TLR2 ligands Pam3CSK4or
PGN and supernatants assayed for TNF-a concentration. b, Defective
Dectin-1/Syk-dependent TLR/MyD88-independent IL-10 production in
Card92/2BMDCs. Wild-type BMDCs (WT), Card92/2BMDCs (2/2),
wild-type BMDCs pre-treated with the Syk inhibitor piceatannol (Syk-inh.)
or Myd882/2BMDCs (Myd882/2) were stimulated with zymosan and 24h
later assayed for IL-10 production. c, Normal IL-10 production in BMDCs
from Card9þ/2and Card92/2mice that were left unstimulated (Medium)
or stimulated with TLR ligands LPS (100ngml21) or CpG (1mM) for 24h.
The indicated values in a–c are means ^ s.d. of triplicates and
representative of three independent experiments. d, Zymosan-induced
TNF-a, IL-6, IL-2 and IL12p40 production in wild-type BMDCs (WT),
wild-type BMDCs pre-treated with Syk inhibitor (Syk-inh.) or Card92/2
BMDCs (2/2) stimulated for 24h as in b. The indicated values are
means ^ s.d. of triplicates. e, Regular Syk activation in Card92/2BMDCs.
Cells were stimulated with zymosan for 45min. Syk activation was
determined by immunoblot with anti-phospho-Syk antibodies. f, Normal
Syk-dependent zymosan phagocytosis. Internalization of FITC-labelled
zymosan in control (þ/2) and Card92/2BMDCs that were pre-incubated
with or without Syk inhibitor (Syk-inh.) was quantified by FACS.
Percentages of cells containing FITC-labelled zymosan before (control) and
after incubation are indicated. g, h, Specific defect in zymosan-induced
NF-kB activation in Card92/2cells. BMDCs from Card9þ/2or Card92/2
mice were stimulated with zymosan or LPS. RelA (red) translocation into
the nucleus (DAPI stained, blue; co-localization, pink) was monitored by
immunofluorescence and quantified by determining the frequency of
RelA-positive nuclei in at least 100 individual cells (g) or NF-kB activation
was determined by EMSA in BMDMs left unstimulated (medium, MED) or
stimulated with zymosan (top, h), PGN, LPS or R848 (bottom, h).
© 2006 Nature Publishing Group
motif to initiate signallingviaSyk throughnovelMyD88-independent
into these pathways that can cooperate with TLR signalling for full
innate immune cell activation and for the induction of pathogen-
specific responses. Our findings also indicate that Bcl10–Malt1
signalling can be engaged by an ITAM-containing PRR through the
adaptor Card9, presumably in an analogous manner to ITAM-
containing lymphocyte antigen receptors that use Carma1 to couple
to Bcl10 and Malt1. This hypothesis suggests an evolutionarily
conserved mechanism for immune cell signalling and puts the
cells can differentially connect to the conserved Bcl10–Malt1 complex
via distinct CARD-coiled-coil adaptor proteins.
Mice, immunizations and infections. Card9-deficient mice were generated
using standard embryonic stem cell technology24. To challenge the immune
system, animals were either immunized intraperitoneally with TNP-Ficoll or
with NP-15-Ova (Biosearch Technologies) absorbed on alum (Sigma)27or
infected intravenously with C. albicans (strain SC5314)15or S. aureus (ATCC
Lymphocyte, BMDC and BMDM culture and cell stimulation. CD4þT cells
and B220þB cells were purified using magnetic beads and cultured in standard
lymphocyte medium24. BMDCs and BMDMs were derived from bone marrow
cells as described29,30. Lymphocytes were stimulated with PMA and Iono
(Sigma), anti-CD3, anti-CD28, anti-IgM and anti-CD40 (Becton Dickinson).
PGN, LPS, CpG, MDP (all from Invivogen), zymosan (from Sigma or Invivogen)
or C. albicans strain SC5314. Cytokine concentrations in the culture supernatants
were quantified using ELISA (Opt-EIA, Becton Dickinson).
Signal transduction. Western blots using antibodies against Syk, phospho-Syk
(NEB), or b-actin (Sigma), electrophoretic mobility shift assays (EMSAs) with
NF-kB-binding-site-containing oligonucleotides and NF-kB luciferase reporter
assays were performed as described25,27. For immunofluorescence, BMDCs were
(Santa Cruz) and Alexa Fluor 594-conjugated secondary antibody, and sub-
sequently analysed using confocal microscopy.
Received 2 April; accepted 17 May 2006.
Published online 12 July 2006.
1. Akira, S. & Takeda, K. Toll-like receptor signalling. Nature Rev. Immunol. 4,
499– -511 (2004).
Gordon, S. Pattern recognition receptors: doubling up for the innate immune
response. Cell 111, 927– -930 (2002).
Inohara, N. & Nunez, G. NODs: intracellular proteins involved in inflammation
and apoptosis. Nature Rev. Immunol. 3, 371– -382 (2003).
Lin, X. & Wang, D. The roles of CARMA1, Bcl10, and MALT1 in antigen
receptor signaling. Semin. Immunol. 16, 429– -435 (2004).
Bonizzi, G. & Karin, M. The two NF-kB activation pathways and their role in
innate and adaptive immunity. Trends Immunol. 25, 280– -288 (2004).
Thome, M. CARMA1, BCL-10 and MALT1 in lymphocyte development and
activation. Nature Rev. Immunol. 4, 348– -359 (2004).
Bertin, J. et al. CARD9 is a novel caspase recruitment domain-containing
protein that interacts with BCL10/CLAP and activates NF-kB. J. Biol. Chem.
275, 41082– -41086 (2000).
Matsumoto, R. et al. Phosphorylation of CARMA1 plays a critical role in T Cell
receptor-mediated NF-kB activation. Immunity 23, 575– -585 (2005).
Sommer, K. et al. Phosphorylation of the CARMA1 linker controls NF-kB
activation. Immunity 23, 561– -574 (2005).
10. Thome, M. & Tschopp, J. TCR-induced NF-kB activation: a crucial role for
Carma1, Bcl10 and MALT1. Trends Immunol. 24, 419– -424 (2003).
11.Di Carlo, F. J. & Fiore, J. V. On the composition of zymosan. Science 127,
756– -757 (1958).
12. Brown, G. D. & Gordon, S. Immune recognition. A new receptor for b-glucans.
Nature 413, 36– -37 (2001).
13. Underhill, D. M. et al. The Toll-like receptor 2 is recruited to macrophage
phagosomes and discriminates between pathogens. Nature 401, 811– -815 (1999).
14. Brown, G. D. & Gordon, S. Fungal b-glucans and mammalian immunity.
Immunity 19, 311– -315 (2003).
15. Villamon, E. et al. Toll-like receptor-2 is essential in murine defenses against
Candida albicans infections. Microbes Infect. 6, 1– -7 (2004).
16. Gantner, B. N., Simmons, R. M., Canavera, S. J., Akira, S. & Underhill, D. M.
Collaborative induction of inflammatory responses by dectin-1 and Toll-like
receptor 2. J. Exp. Med. 197, 1107– -1117 (2003).
Figure 5 | Card9 signals via Bcl10 and Malt1. a, Card9 and Bcl10 synergize
Data are means ^ s.d. of triplicates. b, Bcl10 is required for Card9-induced
NF-kB activation. Card9, Carma3 or Nod1 expression plasmids were
introduced into Bcl10þ/2or Bcl102/2murine embryonic fibroblasts and
NF-kB luciferase assays performed. Data are means ^ s.d. of triplicates.
c, Dectin-1 ligation with zymosan augments Card9–Bcl10-induced NF-kB
induction. 293T cells were transfected with Dectin-1 (50ng per well in 96
wells) and/or a combination of Card9 and Bcl10 expression vectors (5ng
each per well in 96 wells). NF-kB luciferase assays were performed with or
without zymosan stimulation. One representative result from four
from wild type, Bcl102/2or Malt12/2mice that were left unstimulated
(Medium) or stimulated with TLR ligands PGN (20mgml21), LPS
(100ngml21) or CpG (1mM). ND, not detected. Data are means ^ s.d. of
triplicates. e, Bcl10 and Malt1 are required for zymosan-induced cytokine
production. BMDCs from wild type, Bcl102/2or Malt12/2mice were
stimulated with zymosan and TNF-a, IL-2 and IL-10 production was
assayed. Results are means ^ s.d. of triplicates and representative of three
independent experiments. f, Model for the differential control of
Bcl10–Malt1 signalling in innate and adaptive immunity. Y, ITAM. (See text
© 2006 Nature Publishing Group Download full-text
17. Rogers, N. C. et al. Syk-dependent cytokine induction by Dectin-1 reveals a
novel pattern recognition pathway for C type lectins. Immunity 22, 507– -517
18. Brown, G. D. et al. Dectin-1 mediates the biological effects of b-glucans. J. Exp.
Med. 197, 1119– -1124 (2003).
19. Brown, G. D. Dectin-1: a signalling non-TLR pattern-recognition receptor.
Nature Rev. Immunol. 6, 33– -43 (2005).
20. Takeuchi, O. et al. Differential roles of TLR2 and TLR4 in recognition of gram-
negative and gram-positive bacterial cell wall components. Immunity 11,
443– -451 (1999).
21. Aliprantis, A. O. et al. Cell activation and apoptosis by bacterial lipoproteins
through toll-like receptor-2. Science 285, 736– -739 (1999).
22. Takeuchi, O., Hoshino, K. & Akira, S. Cutting edge: TLR2-deficient and
MyD88-deficient mice are highly susceptible to Staphylococcus aureus infection.
J. Immunol. 165, 5392– -5396 (2000).
23. Kawai, T., Adachi, O., Ogawa, T., Takeda, K. & Akira, S. Unresponsiveness of
MyD88-deficient mice to endotoxin. Immunity 11, 115– -122 (1999).
24. Ruland, J. et al. Bcl10 is a positive regulator of antigen receptor-induced
activation of NF-kB and neural tube closure. Cell 104, 33– -42 (2001).
25. McAllister-Lucas, L. M. et al. Bimp1, a MAGUK family member linking protein
kinase C activation to Bcl10-mediated NF-kB induction. J. Biol. Chem. 276,
30589– -30597 (2001).
26. Eliopoulos, A. G., Das, S. & Tsichlis, P. N. The tyrosine kinase Syk regulates
TPL2 activation signals. J. Biol. Chem. 281, 1371– -1380 (2006).
27. Ruland, J., Duncan, G. S., Wakeham, A. & Mak, T. W. Differential requirement for
Malt1 in T and B cell antigen receptor signaling. Immunity 19, 749– -758 (2003).
28. Klemm, S. et al. The Bcl10-Malt1 complex segregates Fc1RI-mediated nuclear
factor kB activation and cytokine production from mast cell degranulation.
J. Exp. Med. 203, 337– -347 (2006).
29. Hammer, M. et al. Control of dual-specificity phosphatase-1 expression in
activated macrophages by IL-10. Eur. J. Immunol. 35, 2991– -3001 (2005).
30. Sparwasser, T. et al. Bacterial DNA and immunostimulatory CpG
oligonucleotides trigger maturation and activation of murine dendritic cells. Eur.
J. Immunol. 28, 2045– -2054 (1998).
Supplementary Information is linked to the online version of the paper at
Acknowledgements We thank H. Wagner, R. Lang, R. Rupec and M. Thome for
discussions; S. Bauer and K. Pechloff for critically reading the manuscript;
B. Holzmann for providing Myd882/2bone marrow; A. Walch for access to a
confocal microscope; M. Neuenhahn for help with intravenous injections; and
S. Weiss, S. Leeder and K. Meiners for technical assistance. This work was
supported by SFB grants from Deutsche Forschungsgemeinschaft to I.F. and J.R.
and by a Max-Eder-Program grant from Deutsche Krebshilfe to J.R.
Author Information Reprints and permissions information is available at
npg.nature.com/reprintsandpermissions. The authors declare no competing
financial interests. Correspondence and requests for materials should be
addressed to J.R. (firstname.lastname@example.org).