Activating NK cell receptors transduce signals through ITAM-containing adaptors, including FcRγ and DAP12. Although the caspase recruitment domain (CARD)9-Bcl10 complex is essential for FcRγ/DAP12-mediated NF-κB activation in myeloid cells, its involvement in NK cell receptor signaling is unknown. Herein we show that the deficiency of CARMA1 or Bcl10, but not CARD9, resulted in severe impairment of cytokine/chemokine production mediated by activating NK cell receptors due to a selective defect in NF-κB activation, whereas cytotoxicity mediated by the same receptors did not require CARMA1-Bcl10-mediated signaling. IκB kinase (IKK) activation by direct protein kinase C (PKC) stimulation with PMA plus ionomycin (P/I) was abrogated in CARMA1-deficient NK cells, similar to T and B lymphocytes, whereas CARD9-deficient dendritic cells (DCs) exhibited normal P/I-induced IKK activation. Surprisingly, CARMA1 deficiency also abrogated P/I-induced IKK activation in DCs, indicating that CARMA1 is essential for PKC-mediated NF-κB activation in all cell types, although the PKC-CARMA1 axis is not used downstream of myeloid ITAM receptors. Consistently, PKC inhibition abrogated ITAM receptor-mediated activation only in NK cells but not in DCs, suggesting PKC-CARMA1-independent, CARD9-dependent ITAM receptor signaling in myeloid cells. Conversely, the overexpression of CARD9 in CARMA1-deficient cells failed to restore the PKC-mediated NF-κB activation. Thus, NF-κB activation signaling through ITAM receptors is regulated by a cell type-specific mechanism depending on the usage of adaptors CARMA1 and CARD9, which determines the PKC dependence of the signaling.

Natural killer cells play the role of a sentinel in protect immunity by exerting a direct cytotoxic effect on target tumor cells and intracellular pathogen-infected cells as well as secreting an array of inflammatory cytokines and chemokines that recruit and activate other cells of the innate and adaptive immune systems. NK cell activation is controlled by multiple germline-encoded activating and inhibitory receptors (activating and inhibitory NK cell receptors) that recognize specific ligands on target cells (1, 2). Whereas signaling through inhibitory NK cell receptors depends on their cytoplasmic ITIMs, activating NK cell receptors do not contain cytoplasmic domains capable of transducing signals, but instead associate with transmembrane adaptor molecules that contain ITAMs, such as DAP12, FcRγ, and CD3ζ (3).

ITAM was initially found as a common and essential sequence in the cytoplasmic domain of signaling chains associated with Ag receptors of lymphocytes, TCRs and BCRs, and certain Fc receptors (4). Upon engagement of Ag receptors, an activation signal cascade is initiated with phosphorylation of specific tyrosines in ITAMs by Src family kinases and subsequent recruitment of tyrosine kinase Syk or ZAP-70 to the phosphorylated ITAMs (5). This results in the activation of a number of downstream effector molecules, including phospholipase C (PLC)γ3; calcium mobilization; protein kinase C (PKC); Vav1; Ras family members; PI3K; MAPKs, such as JNK, ERK, and p38; inhibitor of NF-κB (IκB) kinase (IKK); and, finally, the activation of various transcription factors, including AP-1, NFAT, and NF-κB (6, 7). Although a signaling machinery analogous to that of the Ag receptor has been implicated (8), the precise molecular mechanisms coupling the triggering of activating NK cell receptors to defined NK cell functions, such as cytotoxicity and induction of cytokine/chemokine gene expression, are only partially understood.

In addition to TCRs, BCRs, and activating NK cell receptors, there is accumulating evidence that myeloid lineage cells, such as macrophages and dendritic cells (DCs), also express multiple receptors that transduce activation signals through ITAMs (9, 10). These myeloid ITAM receptors and activating NK cell receptors are similar in terms of structure and signaling mechanisms; that is, most of them belong to either Ig superfamily or C-type lectin family receptors; they associate with the same ITAM-containing adaptor, DAP12 or FcRγ; and signaling is dependent on Syk/Zap-70 activity.

Caspase recruitment domain (CARD) is a protein-binding module that mediates the assembly of CARD-containing proteins. One of the CARD proteins, Bcl10 (11), functions together with MALT1 adaptor protein (12, 13) as a central regulator of TCR- and BCR-mediated NF-κB activation. CARMA1 (also called CARD11 and Bimp3), a CARD-containing protein of the MAGUK (membrane-associated guanylate kinase) family, was found to associate with Bcl10 via CARD-CARD interaction and to be essential for TCR- and BCR-mediated NF-κB activation through the regulation of PKC-dependent lipid raft recruitment of Bcl10 and IKK proteins; thus, it is crucial for acquired immunity (14, 15). These three adaptors, CARMA1, Bcl10, and MALT1, were shown to interact with each other to form the so-called CARMA1-Bcl10-MALT1 (CBM) complex. The involvement of the CBM complex in TLR4 signaling in B cells has also been implicated (12, 14, 16).

We have recently demonstrated that the CARD-containing adaptor molecule CARD9 is essential for NF-κB activation and cytokine production in macrophages and DCs mediated by DAP12- and FcRγ-associated myeloid ITAM receptors, but not for TCR- and BCR-mediated lymphocyte activation (17). CARD9 is structurally related to CARMA1 and is expressed predominantly in myeloid lineage cells among hematopoietic cells (17, 18). Like CARMA1, CARD9 associates with Bcl10 through its CARD to synergistically induce NF-κB activation (18). Analysis of Bcl10−/− and Card11−/− (CARMA1-deficient) mice revealed that the CARD9-mediated signaling through myeloid ITAM receptors involves Bcl10 but not CARMA1 (17). These results suggest that TCRs/BCRs and myeloid ITAM receptors use the CARMA1-Bcl10 complex and CARD9-Bcl10 complex for NF-κB activation, respectively. However, it is not clear why CARMA1 and CARD9 play a nonredundant role in different types of cells, albeit having similar molecular structure and function, despite the fact that CARMA1 is substantially expressed also in myeloid cells.

Bcl10 may play a role in ITAM-mediated NF-κB activation also in NK cells. One recent report has shown that Bcl10 deficiency as well as pharmacological inhibition of PKC or NF-κB impaired activating NK cell receptor-mediated cytokine/chemokine production (19). However, the precise molecular reasons for the defect in cytokine/chemokine production resulting from Bcl10 deficiency, and the molecular connection among Bcl10, PKC, and NF-κB activation downstream of activating NK cell receptors, have not been elucidated.

In this study, by analyzing Card9−/−, Card11−/−, and Bcl10−/− mice, we found that in contrast to myeloid cells, CARD9 deficiency did not affect DAP12- or FcRγ-associated receptor-mediated cytokine/chemokine production in NK cells. Instead, loss of CARMA1 as well as Bcl10 abrogated cytokine/chemokine production in response to stimulation through various activating NK cell receptors due to impaired NF-κB activation, whereas cytotoxicity mediated by the same NK cell receptors did not require CARMA1-Bcl10-mediated signaling. Moreover, we provide new insights into ITAM receptor signaling for NF-κB activation by showing evidence that the selective involvement of CARMA1 but not CARD9 in lymphoid cells, as well as the opposite tendency in myeloid cells, is attributed to cell type-specific regulation through CARMA1 and CARD9 based on PKC dependence, and not simply to the specific expression of these adaptors in certain immune cells.

Card11−/−, Bcl10−/−, and Card9−/− mice were generated previously (14, 16, 17). These mice were backcrossed for at least six generations to C57BL/6J mice. C57BL/6J mice were purchased from Clea Japan. All mice were maintained at animal facilities of the RIKEN Research Center for Allergy and Immunology or Saga University according to institutional guidelines.

Abs specific for Erk (9102), phospho-Erk (9101), p38 (9212), phospho-p38 (9102), Jnk (9258), phospho-Jnk (9251), IκBα (9242) and phospho-IκBα (2859) were from Cell Signaling Technology; Abs specific for Ly49A (JR9–318), Ly49D (4E5), Ly49G2 (4D11), Ly49C/F/E (14B11), NKG2A/C/E (20d5), CD49b (DX5), CD122 (TM-β1), CD11b/Mac1 (M1/70), CD69 (H1.2F3), CD244.2 (2B4), CD43 (S7), CD3ε (145-2C11), NK1.1 (PK136), CD107a/lysosomal-associated membrane protein (LAMP)1 (1D4B), and CD107b/LAMP2 (ABL-93) were from BD Pharmingen; anti-actin (sc-8432) was from Santa Cruz Biotechnology; anti-phospho-tyrosine (4G10) and anti-Vav1 (05-219) were from Upstate Biotechnology; anti-phospho-Vav1 (44–482) was from BioSource International; anti-biotin (Z021) was from Zymed Laboratories; anti-NKG2D (CX5) was from eBioscience; anti-TREM1 (174031) was from R&D Systems; anti-OSCAR was provided by Dr. T. Takai (Tohoku University); F(ab′)2 fragment donkey anti-mouse IgG (H+L) (715-006-150) and F(ab′)2 fragment donkey anti-rat IgG (H+L) (712-006-150) were from Jackson ImmunoResearch Laboratories; and the F(ab′)2 fragment of anti-NK1.1 was prepared from purified anti-NK1.1 (PK136) by using ImmunoPure F(ab′)2 preparation kit (Pierce). Rae1β-human Ig Fc-fusion protein (Rae1β-Fc) was from R&D Systems, and m157-Ig Fc-fusion protein (m157-Fc) was prepared as described (20). NaClO-oxidized zymosan was provided by Dr. N. Ohno (Tokyo University of Pharmacy and Life Science). Recombinant mouse IL-12, IL-15, and IL-18 were from PeproTech. PMA and ionomycin (calcium ionophore A231877) were from Sigma-Aldrich. Rottlerin was from Calbiochem. pcDNA3.1-Carma1 and pcDNA3-Card9 were created using a PCR method. pBXIV-Luc was provided form Dr. W. C. Yeh (Amgen). pRL-TK was from Promega.

Mouse Ba/F3 pro-B cells transduced with m157 (Ba/F3-m157) were provided by Dr. H. Arase. Ba/F3-Rae1β was made by retroviral transfection as previously described (21) using pMX-IRES-GFP-Rae1β, which was generated by cloning an Rae1β-encoding fragment into the retroviral vector pMX-IRES-GFP (kindly provided by Dr. T. Kitamura). For retroviral infection, concentrated virus supernatant was added to the culture of Ba/F3 cells, followed by centrifugation at 800 × g for 1 h at 32°C. Two days after infection, Rae1β-expressing GFP+ cells were sorted with FACSAria (BD Biosciences). For cell surface biotinylation, Ba/F3 cells were suspended and incubated for 1 h on ice in labeling buffer (0.01 M HEPES (pH 8.0), 0.15 M NaCl) containing 100 μg/ml of sulfo-NHS-Biotin (Pierce), followed by washing three times with PBS. JPM50.6, a mutant Jurkat cell line lacking the expression of CARMA1 (22), was kindly provided by Dr. X. Lin (M. D. Anderson Cancer Center, Houston, TX).

NK cells were purified from spleens using NK cell isolation kit (Miltenyi Biotec). Purity was >90% CD3DX5+ by flow cytometry. For expansion, isolated splenic NK cells were cultured for 5–7 days in the presence of 50 ng/ml of IL-15 (PeproTech). Liver mononuclear cells were prepared as previously described (23). Bone marrow (BM)-derived dendritic cells (BMDCs) were prepared by culturing BM cells with 20 ng/ml GM-CSF (PeproTech) for 7 days.

For flow cytometry, single-cell suspensions of BM, spleen, and liver mononuclear cells were stained with FITC-, PE-, APC-, or biotin-conjugated Abs. Biotinylated Abs were visualized with streptavidin-PerCP (BD Pharmingen). Abs used were against Ly49D, Ly49G, Ly49A, Ly49C/F/I/E, NKG2D, NKG2A/C/E, DX5, CD122, Mac1, CD69 2B4, CD43, CD16/32, NK1.1, CD3ε, CD107a (LAMP1), and CD107b (LAMP2). The stained cells were analyzed with a FACSCalibur using CellQuest software (both from BD Biosciences).

Total RNA was extracted from cells using TRIzol reagent (Invitrogen). For cDNA synthesis, total RNA was reverse transcribed with SuperScript II (Invitrogen) using random hexamer primers (Invitrogen) according to the manufacturer’s instructions. Semiquantitative PCR was performed by using TaqDNA polymerase (Roche), and samples were amplified at an annealing temperature of 57°C. The primer sequences used in this study were as follows: mouse β-actin, forward primer (5′-TGGAATCCTGTGGCATCCATGAAAC) and reverse primer (5′-TAAAACGCAGCTCAGTAACAGTCCG); gata-3, forward primer (5′-GAAGGCATCCAGACCCGAAAC) and reverse primer (5′-ACCCATGGCGGTGACCATGC); t-bet, forward primer (5′-CATCACTAAGCAAGGACGGCGA) and reverse primer (5′-AACAGATGCGTACATGGACTCAA); and irf-2, forward primer (5′-GATGGGACGTGGAAAAGGATG) and reverse primer (5′-TGGTCATCATCTCTCAGTGGT). β-actin expression was used to control quantity of cDNA preparation.

Freshly isolated or in vitro IL-15-expanded NK cells were used as effector cells and cytotoxicity was measured in a 4-h 51Cr-release assay as previously described (24). Target cells (Yac1 mouse thymoma, P815 mouse mastocytoma, Ba/F3, Ba/F3-Rae1β, Ba/F3-m157, or surface-biotinylated Ba/F3 cells) were loaded with 200 μCi of 51Cr (ICI Pharmaceuticals). For Ab-dependent cell cytotoxicity (ADCC), 51Cr-labeled surface-biotinylated Ba/F3 cells were preincubated with 10 μg/ml of anti-biotin or control Abs and then mixed with NK cells. For redirected ADCC, NK cells were incubated with anti-NK1.1 mAb or control Ab, and 51Cr-labeled FcR+ P815 cells were added and incubated. After incubation, cells were centrifuged and the culture supernatant was collected. Released 51Cr was measured with a Packard γ-counter (PerkinElmer). Specific cytotoxicity was calculated as previously described (24).

Freshly isolated or IL-15-expanded NK cells were stimulated with immobilized mouse IgG for FcγR crosslinking with or without 10 μg/ml of soluble anti-CD16 (2.4G2), or stimulated with immobilized anti-CD16, F(ab′)2 fragment of anti-NK1.1 mAb, anti-Ly49D mAb, Rae1β-Fc, m157-Fc, isotype-control rat IgG (for anti-Ly49D) or human IgG (for Rae1β-Fc and m157-Fc), or recombinant mouse IL-12 and/or IL-18. For blocking signals through FcγR, stimulation with anti-Ly49D mAb, Rae1β-Fc, m157-Fc, and isotype-control Abs was performed in the presence of 10 μg/ml of soluble 2.4G2. After culture for 16–20 h, cytokines (IFN-γ, TNF-α, and GM-CSF) and C-C chemokines (MIP-1α, MIP-1β, and RANTES) in the culture supernatants were analyzed with ELISA kits (for cytokines, BD Biosciences; for chemokines, R&D Systems).

As a marker for degranulated cells, cell surface exposure of LAMP1 (CD107a) and LAMP2 (CD107b) was analyzed as previously described (25). Equal numbers of IL-15-expanded NK cells and Yac1 cells were mixed and cultured. After incubation for 1 h, the cells were placed on ice and stained with PE-conjugated anti-NK1.1 and FITC-conjugated anti-CD107a and anti-CD107b. Then, NK1.1-positive cells were analyzed for LAMP1/2 expression by flow cytometry.

For calcium mobilization, IL-15-expanded NK cells were loaded with 3 μM INDO-1-AM (Molecular Probes) for 30 min at 37°C. After loading, the cells were incubated with anti-NK1.1 or anti-NKG2D for 20 min at 4°C, washed, and crosslinked at 37°C with F(ab′)2 fragment donkey anti-mouse IgG (H+L) or F(ab′)2 fragment donkey anti-rat IgG (H+L), respectively. Cytosolic Ca2+ flux was analyzed by LSR-FACS (BD Biosciences).

For immunoblot analysis, IL-15-expanded NK cells were stained on ice with anti-CD16, followed by crosslinking with F(ab′)2 fragment donkey anti-rat IgG (H+L) (Jackson ImmunoResearch Laboratories) or stimulation with PMA (20 ng/ml) plus ionomycin (1 μM). After incubation at 37°C for various times, cells were lysed in ice-cold lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1.0% Triton X-100, 20 mM EDTA, 1 mM Na3VO4, 1 mM NaF, and protease inhibitors). Cell lysates were analyzed on SDS-PAGE and proteins were transferred onto a polyvinylidene difluoride membrane. The membrane was incubated with Abs for phospho-tyrosine, Vav1, phospho-Vav1, Erk, phospho-Erk, p38, phospho-38, Jnk, phospho-Jnk, IκBα, and actin, and then with HRP-conjugated secondary Abs, and developed with the ECL detection system (Amersham Biosciences). For image quantification, band intensities were quantified with an LAS3000 imaging system (Fujifilm).

For analysis of NF-κB DNA binding activity, BMDCs were stimulated for 8 h with immobilized anti-CD16 (10 μg/ml) or IL-18 (10 ng/ml). Nuclear extract was prepared from the stimulated cells, and the binding activity of NF-κB subunit p65 in the extracts was measured with a Mercury TransFactor kit (BD Clontech).

JPM50.6 cells (5 × 105) in 6-well plates were transfected with 1.5 μg of the NF-κB-dependent luciferase (firefly) reporter plasmid pBXIV-Luc and 0.2 μg of the HSV-TK promoter-regulated Renilla luciferase reporter plasmid pRL-TK plus indicated amount of expression plasmids (pcDNA3.1-CARD9 or pcDNA3.1-CARMA1) or a mock vector (pcDNA3.1) using Lipofectamine LTX (Invitrogen) according to the manufacture’s instructions. The transfected cells were cultured for 24 h and then stimulated with PMA (20 ng/ml) plus ionomycin (1 μM) for 8 h. The cells were lysed, and luciferase activity in the lysates was measured by a dual-luciferase kit (Promega).

We have previously reported that Bcl10 is essential for cytokine response to DAP12- and FcRγ-associated receptor stimulation in myeloid cells such as macrophages and DCs (17). Since activating NK cell receptors also deliver activation signals by associating with DAP12 or FcRγ, similar to the case of myeloid cells, we examined whether Bcl10 is involved in NK cell receptor signaling. When we stimulated Bcl10−/− NK cells through FcRγ-associated receptors NK1.1 (26) and FcγRIII (CD16) with anti-NK1.1 and anti-CD16, respectively, and through DAP12-associated receptors NKG2D and Ly49D with human Ig Fc-fusion protein of the NKG2D ligand Rae1β (Rae1β-Fc) and anti-Ly49D, respectively, inflammatory cytokine production (Fig. 1,A) and C-C chemokine production (Fig. 1 B) were severely impaired compared with those in wild-type (WT) mice, consistent with a previous report (19). Thus, Bcl10 is critically involved in DAP12- and FcRγ-associated receptor signaling in NK cells as well as myeloid cells.

FIGURE 1.

CARD9 is dispensable for NK cell activation. ELISA for the indicated cytokine (A) and C-C chemokine (B) from IL-15-expanded splenic WT and Bcl10−/− NK cells stimulated for 18 h with immobilized anti-CD16 mAb (10 μg/ml), F(ab′)2 fragment of anti-NK1.1 mAb (10 μg/ml), Rae1β-Fc (15 μg/ml), or anti-Ly49D mAb (15 μg/ml). Note that cytokine/chemokine production by WT and Bcl10−/− NK cells was comparable or undetectable when cells were unstimulated or stimulated with control Abs. Data are indicated as percentage of cytokine or chemokine concentration produced by WT NK cells. C, ELISA for the production of TNF-α and IL-6 by WT and Card9−/− BMDCs (upper) or NK cells (lower) stimulated for 16 h with immobilized anti-CD16 mAb. D–F, ELISA for TNF-α and IFN-γ production by IL-15-expanded splenic WT and Card9−/− NK cells stimulated for 16 h with immobilized F(ab′)2 fragment of anti-NK1.1 mAb (D) or Rae1β-Fc (E), or with IL-18 (5 ng/ml) in the presence or absence of IL-12 (1 ng/ml) (F). Soluble anti-CD16 mAb (10 μg/ml) was added for Fc blocking during stimulation with Rae1β-Fc or anti-Ly49D mAb. Human IgG was used as stimulation control Ab in E. Data are means ± SD of triplicates and are representative of two independent experiments.

FIGURE 1.

CARD9 is dispensable for NK cell activation. ELISA for the indicated cytokine (A) and C-C chemokine (B) from IL-15-expanded splenic WT and Bcl10−/− NK cells stimulated for 18 h with immobilized anti-CD16 mAb (10 μg/ml), F(ab′)2 fragment of anti-NK1.1 mAb (10 μg/ml), Rae1β-Fc (15 μg/ml), or anti-Ly49D mAb (15 μg/ml). Note that cytokine/chemokine production by WT and Bcl10−/− NK cells was comparable or undetectable when cells were unstimulated or stimulated with control Abs. Data are indicated as percentage of cytokine or chemokine concentration produced by WT NK cells. C, ELISA for the production of TNF-α and IL-6 by WT and Card9−/− BMDCs (upper) or NK cells (lower) stimulated for 16 h with immobilized anti-CD16 mAb. D–F, ELISA for TNF-α and IFN-γ production by IL-15-expanded splenic WT and Card9−/− NK cells stimulated for 16 h with immobilized F(ab′)2 fragment of anti-NK1.1 mAb (D) or Rae1β-Fc (E), or with IL-18 (5 ng/ml) in the presence or absence of IL-12 (1 ng/ml) (F). Soluble anti-CD16 mAb (10 μg/ml) was added for Fc blocking during stimulation with Rae1β-Fc or anti-Ly49D mAb. Human IgG was used as stimulation control Ab in E. Data are means ± SD of triplicates and are representative of two independent experiments.

Close modal

CARD9 forms a complex with Bcl10 to synergistically induce NF-κB activation. We have recently reported that CARD9 is essential for the activation of myeloid cells through DAP12- or FcRγ-associated receptors (17). Whereas Bcl10 is critical for cytokine/chemokine gene expression through DAP12- or FcRγ-associated receptors on NK cells, the role of CARD9 in NK cell activation remains unknown. Thus, we examined the influence of CARD9 deficiency on NK cell receptor-mediated cytokine production. As shown in Fig. 1,A and consistent with our previous study (17), the lack of CARD9 resulted in marked impairment in the production of proinflammatory cytokines, such as TNF-α and IL-6, by BMDCs upon CD16 crosslinking (Fig. 1,C, upper). However, TNF-α and IFN-γ production upon crosslinking of the same receptor on NK cells was not compromised by CARD9 deficiency (Fig. 1,C, lower). Similarly, TNF-α, IFN-γ, and GM-CSF (not shown) production in response to stimulation through NK1.1 (Fig. 1,D) and NKG2D (Fig. 1 E) was also comparable between WT and Card9−/− NK cells. Thus, these results suggest that CARD9 is dispensable for FcRγ- and DAP12-associated receptor signaling in NK cells, in contrast to myeloid cells.

The IL-18 receptor delivers signals via Myd88 and stimulates NK cells to produce abundant IFN-γ by itself as well as in synergy with IL-12 (27). We have previously reported that CARD9 is involved in Myd88-mediated signaling in BMDCs (17). Thus, we analyzed the involvement of CARD9 in IL-18 receptor signaling. However, we found that IFN-γ production in response to IL-18 or to IL-18 plus IL-12 was normal in Card9−/− NK cells (Fig. 1 F).

Taken together, these results indicate that, despite its essential role in myeloid cells, CARD9 is not required for the activation of NK cells through ITAM-associated and Myd88-associated receptors.

Since CARMA1 is also capable of regulating Bcl10-mediated signaling and is abundantly expressed in lymphoid lineage cells, we examined whether CARMA1 is involved in NK cell activation through activating NK cell receptors. In addition to the receptors tested in Fig. 1, we also analyzed Ly49H, a DAP12-associated receptor for the MCMV-encoded protein m157 (20). Stimulation of NK cells through FcγRIII with immobilized mouse IgG (Fig. 2,A), NK1.1 with anti-NK1.1 (Fig. 2,B), NKG2D with Rae1β-Fc (Fig. 2,C), Ly49D with anti-Ly49D (Fig. 2,D), or Ly49H with a human Ig Fc-fusion protein of m157 (m157-Fc, Fig. 2,E) resulted in IFN-γ, TNF-α, and GM-CSF production by WT NK cells. However, none of these stimuli induced the production of these cytokines in Card11−/− (CARMA1-deficient) NK cells. These defects could not be overcome even with high doses of the activating stimuli. These results suggest that CARMA1 is essential for the activation of NK cell receptor-mediated cytokine production in NK cells. Consistent with the results of cytokine production, cytokine mRNA expression after stimulation with anti-NK1.1 was markedly decreased in Card11−/− NK cells (Fig. 2,F), suggesting that the impairment of cytokine production is regulated on the transcriptional level. In contrast to activating NK cell receptor stimulation, IL-12- and/or IL-18 receptor stimulation led to comparable cytokine production by WT and Card11−/− NK cells (Fig. 2 G), indicating that Card11−/− NK cells were capable of cytokine production and, therefore, the defect was specific to signaling mediated by ITAM-associated NK cell receptors. Similar to cytokine production, MIP-1α, MIP-1β, and RANTES production was severely impaired after stimulation through activating NK cell receptors but not through IL-12 and IL-18 receptors in Card11−/− NK cells (data not shown).

FIGURE 2.

Impaired production of inflammatory cytokines upon stimulation of Card11−/− NK cells through activating NK cell receptors. ELISA (A–E and G) and real-time PCR analysis (F) for indicated cytokines from IL-15-expanded splenic WT and Card11−/− NK cells stimulated for 16 h with immobilized mouse IgG for FcγR crosslinking in the presence or absence of Fc blocker (soluble anti-CD16) (A), or stimulated with immobilized F(ab′)2 fragment of α-NK1.1 mAb (B and F), Rae1β-Fc (C), anti-Ly49D mAb (20 μg/ml) (D), or m157-Fc (20 μg/ml) (E), or with IL-12 (1 ng/ml) (G, top), IL-18 (5 ng/ml) (G, middle), or IL-12 plus IL-18 (G, bottom). Soluble anti-CD16 mAb (10 μg/ml) was added for Fc blocking during stimulation (C–E). Human (C and E) or rat (D) IgG was used as stimulation control Ab. Data are means ± SD of triplicates and are representative of three independent experiments.

FIGURE 2.

Impaired production of inflammatory cytokines upon stimulation of Card11−/− NK cells through activating NK cell receptors. ELISA (A–E and G) and real-time PCR analysis (F) for indicated cytokines from IL-15-expanded splenic WT and Card11−/− NK cells stimulated for 16 h with immobilized mouse IgG for FcγR crosslinking in the presence or absence of Fc blocker (soluble anti-CD16) (A), or stimulated with immobilized F(ab′)2 fragment of α-NK1.1 mAb (B and F), Rae1β-Fc (C), anti-Ly49D mAb (20 μg/ml) (D), or m157-Fc (20 μg/ml) (E), or with IL-12 (1 ng/ml) (G, top), IL-18 (5 ng/ml) (G, middle), or IL-12 plus IL-18 (G, bottom). Soluble anti-CD16 mAb (10 μg/ml) was added for Fc blocking during stimulation (C–E). Human (C and E) or rat (D) IgG was used as stimulation control Ab. Data are means ± SD of triplicates and are representative of three independent experiments.

Close modal

Taken together, these results clearly suggest that the CARMA1-Bcl10 complex, but not the CARD9-Bcl10 complex, relays ITAM-mediated signals for cytokine/chemokine gene expression in NK cells.

Next, we investigated the involvement of CARMA1 and Bcl10 in NK cell cytotoxicity by triggering activating NK cell receptors. We first examined natural killing activity against NK-sensitive Yac1 cells, which is recognized predominantly through NKG2D (28), and NK-resistant P815 cells. We found that the lack of CARMA1 (Fig. 3,A) or Bcl10 (Fig. 3,B) did not affect NK cell cytotoxicity against these target cells. Similar results were obtained when freshly isolated NK cells were used (Fig. 3,C). Consistent with this, Card11−/− NK cells killed Ba/F3 mouse pro-B cells transduced with the NKG2D ligand Rae1β, in the same manner as WT NK cells (Fig. 3,D). Similarly, Card11−/− NK cells also killed m157-expressing Ba/F3 cells at an efficiency comparable to WT NK cells (Fig. 3 E). Thus, NK cell cytotoxicity through DAP10- or/and DAP12-associated receptors does not require CARMA1-Bcl10 signaling.

FIGURE 3.

CARMA1 is dispensable for NK cell cytotoxicity. IL-15-expanded (A, B, D–G, and H) or freshly isolated (C) splenic WT, Card11−/− (A, C–F, and H), and Bcl10−/− (B and G) NK cells were subjected to a 4-h 51Cr-release assay for cytotoxicity against indicated target cells at the indicated E:T ratios. A–C, Natural cytotoxicity against Yac1 and P815 cells. D and E, Cytotoxicity against (D) Rae1β- or (E) m157-expressing Ba/F3 cells. F and G, ADCC against surface-biotinylated Ba/F3 cells in the presence of α-biotin or control Ab. H, Redirected ADCC against FcR+ P815 cells coated with anti-NK1.1 mAb or control IgG. Data are means ± SD of triplicates and are representative of three independent experiments.

FIGURE 3.

CARMA1 is dispensable for NK cell cytotoxicity. IL-15-expanded (A, B, D–G, and H) or freshly isolated (C) splenic WT, Card11−/− (A, C–F, and H), and Bcl10−/− (B and G) NK cells were subjected to a 4-h 51Cr-release assay for cytotoxicity against indicated target cells at the indicated E:T ratios. A–C, Natural cytotoxicity against Yac1 and P815 cells. D and E, Cytotoxicity against (D) Rae1β- or (E) m157-expressing Ba/F3 cells. F and G, ADCC against surface-biotinylated Ba/F3 cells in the presence of α-biotin or control Ab. H, Redirected ADCC against FcR+ P815 cells coated with anti-NK1.1 mAb or control IgG. Data are means ± SD of triplicates and are representative of three independent experiments.

Close modal

We also checked NK cell cytotoxicity through FcRγ-associated receptors. Card11−/− (Fig. 3,F) and Bcl10−/− NK cells (Fig. 3,G) exhibited normal ADCC through FcγRs against surface-biotinylated Ba/F3 cells in the presence of α-biotin Ab. Similarly, redirected ADCC against FcR+ P815 cells coated with anti-NK1.1 mAb was also comparable between WT and Card11−/− NK cells (Fig. 3 H). Thus, cytotoxicity through FcRγ-associated receptors also does not require CARMA1-Bcl10 signaling.

Taken together, these results indicate that CARMA1-Bcl10 signaling is dispensable for NK cell cytotoxicity through activating NK cell receptors, although it is essential for cytokine/chemokine production by the same receptor stimulation.

NK cell functions, including cytokine/chemokine production and cytolysis, are associated with the state of NK cell maturation (20). We therefore investigated whether the functional defects in Card11−/− NK cells are due to defective development and/or maturation of NK cells. The percentages of CD3NK1.1+DX5+ NK cells in BM, liver, and spleen (Fig. 4,A, left) and the absolute numbers of CD3NK1.1+ NK cells in spleen (Fig. 4,B) of Card11−/− mice were not significantly different from those of WT mice. Moreover, the expression of other NK cell markers, including CD122, 2B4, CD16, and CD69, was also comparable between WT and Card11−/− NK cells (Table I).

FIGURE 4.

Normal NK cell development in Card11−/− mice. A, Flow cytometric analysis of NK cells in BM, spleen, and liver of WT and Card11−/− mice. Numbers in quadrants indicate percentage of positive cells in the region. B, Absolute numbers of NK cells in spleens of WT and Card11−/− mice. C, Semiquantitative PCR analysis of T-bet, GATA-3, and IRF-2 expression in purified splenic NK cells from WT and Card11−/− mice. Four-fold serial dilutions of cDNA were tested. β-actin was used as control for normalization.

FIGURE 4.

Normal NK cell development in Card11−/− mice. A, Flow cytometric analysis of NK cells in BM, spleen, and liver of WT and Card11−/− mice. Numbers in quadrants indicate percentage of positive cells in the region. B, Absolute numbers of NK cells in spleens of WT and Card11−/− mice. C, Semiquantitative PCR analysis of T-bet, GATA-3, and IRF-2 expression in purified splenic NK cells from WT and Card11−/− mice. Four-fold serial dilutions of cDNA were tested. β-actin was used as control for normalization.

Close modal
Table I.

Expression of NK cell markersa

MarkerWTCard11−/−
 % Positive  
Ly49D 53.4 ± 6.3 49.9 ± 6.3 
Ly49G2 51.8 ± 5.0 48.9 ± 2.4 
Ly49A 16.0 ± 2.6 13.2 ± 2.0 
Ly49C/F/I/E 83.4 ± 4.8 78.7 ± 6.6 
NKG2A/C/E 41.5 ± 4.6 42.5 ± 2.6 
NKG2D 90.2 ± 2.6 86.7 ± 4.8 
DX5 89.2 ± 4.4 88.7 ± 3.8 
CD122 90.7 ± 3.3 86.7 ± 4.8 
Mac1 83.2 ± 6.2 90.3 ± 3.3 
CD69 4.1 ± 1.3 2.9 ± 1.4 
2B4 95.1 ± 2.0 96.9 ± 1.7 
CD43 91.6 ± 6.2 92.6 ± 4.8 
CD16/32 72.2 ± 4.3 69.2 ± 8.8 
MarkerWTCard11−/−
 % Positive  
Ly49D 53.4 ± 6.3 49.9 ± 6.3 
Ly49G2 51.8 ± 5.0 48.9 ± 2.4 
Ly49A 16.0 ± 2.6 13.2 ± 2.0 
Ly49C/F/I/E 83.4 ± 4.8 78.7 ± 6.6 
NKG2A/C/E 41.5 ± 4.6 42.5 ± 2.6 
NKG2D 90.2 ± 2.6 86.7 ± 4.8 
DX5 89.2 ± 4.4 88.7 ± 3.8 
CD122 90.7 ± 3.3 86.7 ± 4.8 
Mac1 83.2 ± 6.2 90.3 ± 3.3 
CD69 4.1 ± 1.3 2.9 ± 1.4 
2B4 95.1 ± 2.0 96.9 ± 1.7 
CD43 91.6 ± 6.2 92.6 ± 4.8 
CD16/32 72.2 ± 4.3 69.2 ± 8.8 
a

NK1.1+CD3 splenic NK cells from WT and Card11−/− mice of 12–17 wk of age were analyzed for the expression of surface markers. Values indicate the means ± SD percentages of NK cells positive for the indicated marker by flow cytometry from five mice per group.

Integrin Mac-1 expression is regulated during NK cell maturation, being high in splenic NK cells and low in BM NK cells (29). This is correlated with the acquisition of functional competence, including cytotoxicity and IFN-γ productivity. The percentages of fully mature Mac-1high NK cells in BM, liver, and spleen were comparable between WT and Card11−/− mice (Fig. 4,A). Additionally, the expression of another maturation marker, CD43 (29), in splenic NK cells was also comparable between WT and Card11−/− mice (Table I).

NK cells express a “repertoire” of NK cell receptors, including CD94, that forms complexes with NKG2A/C/E, as well as various activating and inhibitory receptors of Ly49 C-type lectin gene family (30). It has been suggested that the NK cell receptor repertoire is acquired sequentially during development in BM; CD94/NKG2A/C/E appear first, followed by Ly49 (29, 31). Additionally, NK cells acquire functional competence through the “licensing” effect by expressing inhibitory Ly49 receptors specific for self-MHC class I molecules (32). In line with this, it has been reported that CD45−/− mice having similar functional impairment of NK cells to Card11−/− mice show normal development and maturation of NK cells but an alteration in the repertoire of Ly49 receptors (23). Therefore, the imbalanced expression of Ly49 and/or the CD94/NKG2 receptor repertoire might influence the overall functional competence of the peripheral NK cell pool. In this regard, we analyzed the repertoire of NK cell receptors in Card11−/− NK cells. The percentage expression of NKG2D, NKG2A/C/E, the activating receptor Ly49D, and inhibitory receptors Ly49G, Ly49A, and Ly49D/F/I/E in splenic CD3NK1.1+ NK cells was not significantly modified in Card11−/− mice in comparison to WT mice (Table I).

Transcription factors GATA-3, T-bet, and IRF-2 have been suggested to control the terminal maturation of NK cells (33, 34, 35). NK cells deficient for these transcription factors display similar phenotypes to those in Card11−/− mice: NK cells are poor producers of IFN-γ; nevertheless, they retain normal or only moderately reduced cytotoxic potential. Thus, the functional defects in Card11−/− NK cells might be attributed to loss or low expression of these transcription factors. However, the mRNA levels of GATA-3, T-bet, and IRF-2 in splenic Card11−/− NK cells were not significantly changed compared with those in WT NK cells (Fig. 4 C).

Taken together, these results indicate that CARMA1 is dispensable for the development, phenotypic maturation, and repertoire formation of NK cells, and they suggest that the defective cytokine/chemokine production observed in Card11−/− NK cells is not attributed to abnormal NK cell differentiation.

To identify the molecular mechanism(s) that account for the defect of specific NK function in Card11−/− and Bcl10−/− NK cells, we analyzed signaling pathways downstream of activating NK cell receptors in Card11−/− NK cells. Upon stimulation through ITAM-coupling receptors CD16 (Fig. 5,A), NK1.1 (Fig. 5,B), or NKG2D (Fig. 5, B and C), total tyrosine phosphorylation levels (Fig. 5,A) and signaling events associated with cytotoxicity, such as calcium mobilization (Fig. 5,B), Vav1 phosphorylation (Fig. 5,A), and cell surface exposure of lysosomal protein LAMP1/2, which is concomitantly induced with the degranulation of lytic granules of NK cells (Fig. 5,C), were induced normally in Card11−/− NK cells. These results were consistent with the competent NK cell receptor-mediated cytotoxicity in Card11−/− NK cells (Fig. 3).

FIGURE 5.

CARMA1 regulates NF-κB but not MAPK activation upon stimulation through activating NK cell receptors. A, Immunoblot analysis of tyrosine phosphorylation of total protein and Vav1 levels in IL-15-expanded splenic NK cells from WT and Card11−/− mice upon stimulation with FcγRIII crosslinking. B, Calcium influx of Indo-1-AM-loaded WT and Card11−/− NK cells upon stimulation with anti-NK1.1 or anti-NKG2D mAb. FL5/FL4 fluorescence ratio was monitored for 200 s. C, Flow cytometric analysis of cell surface expression of LAMP1/2 on WT and Card11−/− NK cells after incubation with Yac1 cells for 1 h. D, Immunoblot analysis of activation of MAPKs (ERK, JNK, and p38) in WT and Card11−/− NK cells upon stimulation with anti-CD16 mAb. E–G, Immunoblot analysis of I-κBα degradation in WT and Card11−/− NK cells upon stimulation with anti-CD16 (E), TNF-α (10 ng/ml) (F), or IL-18 (10 ng/ml) (G). Band intensities of I-κBα were quantified and the values were normalized to the intensity of actin at the same time points. Normalized results are indicated in histograms appearing at the bottom of the panels, as values relative to those at time 0 (set at 100). Data are representative of two independent experiments. H and I, Assay for DNA binding activity of NF-κB p65 in nuclear extract of WT or Card11−/− NK cells unstimulated or stimulated with immobilized anti-CD16 mAb (10 μg/ml) (H) for indicated times or with IL-18 (10 ng/ml) for 8 h (I). Relative DNA binding was colorimetrically measured.

FIGURE 5.

CARMA1 regulates NF-κB but not MAPK activation upon stimulation through activating NK cell receptors. A, Immunoblot analysis of tyrosine phosphorylation of total protein and Vav1 levels in IL-15-expanded splenic NK cells from WT and Card11−/− mice upon stimulation with FcγRIII crosslinking. B, Calcium influx of Indo-1-AM-loaded WT and Card11−/− NK cells upon stimulation with anti-NK1.1 or anti-NKG2D mAb. FL5/FL4 fluorescence ratio was monitored for 200 s. C, Flow cytometric analysis of cell surface expression of LAMP1/2 on WT and Card11−/− NK cells after incubation with Yac1 cells for 1 h. D, Immunoblot analysis of activation of MAPKs (ERK, JNK, and p38) in WT and Card11−/− NK cells upon stimulation with anti-CD16 mAb. E–G, Immunoblot analysis of I-κBα degradation in WT and Card11−/− NK cells upon stimulation with anti-CD16 (E), TNF-α (10 ng/ml) (F), or IL-18 (10 ng/ml) (G). Band intensities of I-κBα were quantified and the values were normalized to the intensity of actin at the same time points. Normalized results are indicated in histograms appearing at the bottom of the panels, as values relative to those at time 0 (set at 100). Data are representative of two independent experiments. H and I, Assay for DNA binding activity of NF-κB p65 in nuclear extract of WT or Card11−/− NK cells unstimulated or stimulated with immobilized anti-CD16 mAb (10 μg/ml) (H) for indicated times or with IL-18 (10 ng/ml) for 8 h (I). Relative DNA binding was colorimetrically measured.

Close modal

Three MAPKs, ERK, JNK, and p38, are involved in the regulation of cytokine/chemokine gene expression in immune cells, including NK cells (36). Since our previous report showed that TCR- and BCR-mediated JNK activation is impaired in Card11−/− T and B lymphocytes (14), we next analyzed the activation of MAPKs. However, stimulation of WT and Card11−/− NK cells with anti-CD16 revealed no apparent differences in the phosphorylation of ERK, JNK, and p38 (Fig. 5 D), indicating that the defective cytokine/chemokine production observed in Card11−/− NK cells is not attributed to impaired MAPK activation.

As it has been suggested that Bcl10-mediated signaling is essential for NF-κB activation through TCR, BCR, FcεRI, and myeloid ITAM receptors (11, 17, 37), we investigated NK cell receptor-mediated NF-κB activation in Card11−/− NK cells. The degradation of I-κBα upon CD16 crosslinking was markedly impaired in Card11−/− NK cells (Fig. 5,E). Additionally, DNA binding activity of the p65-containing NK-κB complex was markedly decreased in the nuclei of Card11−/− NK cells after anti-CD16 stimulation (Fig. 5,H). In contrast, the impairment of I-κBα degradation (Fig. 5, F and G) and of the DNA binding activity of NF-κB (Fig. 5,I) were not observed when Card11−/− NK cells were stimulated through other NF-κB-activating receptors, such as TNFR1 (Fig. 5,F) and IL-18R (Fig. 5 G). A similar deficiency of NF-κB activation was observed in Bcl10−/− NK cells (data not shown). Collectively, these results suggest that CARMA1-Bcl10 selectively mediates the NF-κB activation pathway but not other signaling pathways through ITAM-associated receptors but not other receptors, which may be responsible for the involvement of CARMA1-Bcl10 in specific effector functions of NK cells mediated by specific receptors on the cells.

Our next question was why CARMA1 and CARD9 function in specific types of cells. It might be possible that the phenotypes of Card9−/− and Card11−/− mice simply reflected the specific expression of CARD9 and CARMA1 in myeloid and lymphoid cells, respectively. Indeed, CARD9 is highly expressed in myeloid cells but barely detectable in lymphoid cells (17). In contrast, CARMA1 is substantially expressed both in myeloid and lymphoid cells (38) (Fig. 6 A). Thus, we hypothesized that there is a cell type-specific regulation of ITAM receptor-mediated signaling through CARMA1 and CARD9, probably due to their structural differences. Previous reports have suggested that the phosphorylation of CARMA1 by PKC is crucial for its function in Ag receptor signaling (39, 40). As a result, Card11−/− or Bcl10−/− T and B cells are hyporesponsive to direct PKC stimulation with P/I (11, 14). However, the fact that CARD9 lacks PKC phosphorylation sites suggests the possibility that CARD9-mediated signaling cannot be regulated by PKC.

FIGURE 6.

Differential PKC-based regulation in CARD9- and CARMA1-mediated NF-κB activation signaling. A, Expressions of mouse Carma1 mRNA in immune cells were analyzed by means of a reference database of quantitative mRNA, RefDiC (http://refdic.rcai.riken.jp/welcome.cgi), and the values are shown by a histogram. Data are representative of the analyses using two different probes. BMMC, bone marrow-derived mast cells. B–E, ELISA for indicated cytokine production by WT, Card11−/− (B), and Bcl10−/− (C) NK cells, and by WT, Card9−/− (D), and Bcl10−/− (E) BMDCs stimulated for 18 h with PMA (20 ng/ml) plus ionomycin (1 μM) (P/I). Data are means ± SD of triplicates. F and G, Immunoblot analysis of phosphorylated (P-) I-κBα and I-κBα degradation in NK cells (F) and BMDCs (G) from WT, Card9−/−, and Card11−/− mice upon stimulation with P/I. H, ELISA for the production of IFN-γ by NK cells (left panels) and of TNF-α by BMDCs (right panels) from C57BL/6 mice unstimulated or stimulated for 16 h with 10 μg/ml of indicated immobilized Abs or 20 μg/ml of NaClO-oxidized zymosan (OX-Zym) in the presence or absence of 5, 20, or 50 μM of the PKC inhibitor rottlerin. Soluble anti-CD16 (10 μg/ml) was added for Fc blocking during stimulation with anti-Ly49D, anti-NKG2D, anti-OSCAR, and anti-TREM1. Data are means ± SD of triplicates. I, Jurkat T (lane 8) or JPM50.6 (lanes 1–7) cells were transfected with plasmids encoding NF-κB-dependent luciferase reporter gene plus 0.1 (lanes 2 and 5), 0.3 (lanes 3 and 6), or 1 (lanes 4 and 7) μg of expression plasmid encoding CARMA1 (lanes 2–4) or CARD9 (lanes 5–7), or 1 μg of a mock plasmid (lanes 1 and 8). Twenty-four hours after transfection, the transfected cells were stimulated for 8 h with P/I. The cells were lysed and luciferase activity was determined. All of the samples were also cotransfected with HSV-TK promoter Renilla luciferase plasmid. Constitutively expressed Renilla luciferase activity was used to normalize transfection efficiency. All data are representative of three independent experiments.

FIGURE 6.

Differential PKC-based regulation in CARD9- and CARMA1-mediated NF-κB activation signaling. A, Expressions of mouse Carma1 mRNA in immune cells were analyzed by means of a reference database of quantitative mRNA, RefDiC (http://refdic.rcai.riken.jp/welcome.cgi), and the values are shown by a histogram. Data are representative of the analyses using two different probes. BMMC, bone marrow-derived mast cells. B–E, ELISA for indicated cytokine production by WT, Card11−/− (B), and Bcl10−/− (C) NK cells, and by WT, Card9−/− (D), and Bcl10−/− (E) BMDCs stimulated for 18 h with PMA (20 ng/ml) plus ionomycin (1 μM) (P/I). Data are means ± SD of triplicates. F and G, Immunoblot analysis of phosphorylated (P-) I-κBα and I-κBα degradation in NK cells (F) and BMDCs (G) from WT, Card9−/−, and Card11−/− mice upon stimulation with P/I. H, ELISA for the production of IFN-γ by NK cells (left panels) and of TNF-α by BMDCs (right panels) from C57BL/6 mice unstimulated or stimulated for 16 h with 10 μg/ml of indicated immobilized Abs or 20 μg/ml of NaClO-oxidized zymosan (OX-Zym) in the presence or absence of 5, 20, or 50 μM of the PKC inhibitor rottlerin. Soluble anti-CD16 (10 μg/ml) was added for Fc blocking during stimulation with anti-Ly49D, anti-NKG2D, anti-OSCAR, and anti-TREM1. Data are means ± SD of triplicates. I, Jurkat T (lane 8) or JPM50.6 (lanes 1–7) cells were transfected with plasmids encoding NF-κB-dependent luciferase reporter gene plus 0.1 (lanes 2 and 5), 0.3 (lanes 3 and 6), or 1 (lanes 4 and 7) μg of expression plasmid encoding CARMA1 (lanes 2–4) or CARD9 (lanes 5–7), or 1 μg of a mock plasmid (lanes 1 and 8). Twenty-four hours after transfection, the transfected cells were stimulated for 8 h with P/I. The cells were lysed and luciferase activity was determined. All of the samples were also cotransfected with HSV-TK promoter Renilla luciferase plasmid. Constitutively expressed Renilla luciferase activity was used to normalize transfection efficiency. All data are representative of three independent experiments.

Close modal

To investigate whether CARMA1- or CARD9-mediated induction of cytokine/chemokine production in NK cells or BMDCs, respectively, is regulated by PKCs, we stimulated Card9−/−, Card11−/−, or Bcl10−/− NK cells or BMDCs with P/I. IFN-γ and TNF-α production was severely impaired in Card11−/− (Fig. 6,B) and Bcl10−/− (Fig. 6,C) NK cells, indicating that CARMA1-Bcl10 signaling is essential for PKC-mediated cytokine production in NK cells. However, the lack of CARD9 (Fig. 6,D) or Bcl10 (Fig. 6,E) in BMDCs did not affect IL-6 and TNF-α production in response to stimulation with P/I, indicating that CARD9-Bcl10 signaling is not required for PKC-mediated cytokine production. Consistent with the impaired cytokine production, P/I-induced phosphorylation and degradation of IκBα were completely abolished in Card11−/−, but not Card9−/−, NK cells, whereas ERK activation was not affected (Fig. 6,F). In contrast, Card9−/− BMDCs exhibited normal IκBα phosphorylation and degradation following the same stimulation (Fig. 6,G). Intriguingly, loss of CARMA1 also abrogated P/I-induced phosphorylation and degradation of IκBα in BMDCs, similar to NK cells (Fig. 6 G). These results indicate that CARMA1 acts downstream of PKC and is essential for coupling PKC stimulation to NF-κB activation not only in NK cells but also in BMDCs, and they suggest that the PKC-CARMA1 signaling axis is not used for NF-κB activation and cytokine production through myeloid ITAM receptors, despite the existence of such a pathway and its potential to activate NF-κB.

Next, we examined the effect of the PKC inhibitor rottlerin in ITAM receptor-mediated activation of NK cells and DCs. Rottlerin has previously been shown to inhibit TCR- and BCR-mediated NF-κB activation (41, 42). The PKC inhibitior effectively blocked cytokine production including IFN-γ mediated by all the ITAM-associated receptors tested in NK cells (Fig. 6,H, left), whereas it did not affect ITAM receptor-mediated cytokine production in BMDCs (Fig. 6 H, right), indicating that PKC activation is essential for ITAM receptor-mediated cytokine production in NK cells, but is dispensable in DCs.

Finally, to investigate whether CARD9 potentially functions to couple PKC stimulation to NF-κB activation in lymphoid cells, we examined whether CARD9 expression is able to compensate for CARMA1 deficiency by using a CARMA1-deficient Jurkat cell line, JPM50.6 (22), which we found to be defective for P/I-induced NF-κB activation (Fig. 6,I, lower, lane 1). Transfection of JPM50.6 with elevated doses of an expression plasmid for CARD9 (Fig. 6,I, upper, lanes 5–7) induced a dose-dependent elevation of basal NF-κB activity to levels comparable to or higher than those in cells transfected with the same amounts of an expression plasmid for CARMA1 (Fig. 6,I, upper, lanes 2–4), indicating that CARD9 is able to function in lymphoid cells for NF-κB activation presumably through Bcl10. However, upon P/I stimulation of the transfected cells, whereas a minimal dose of CARMA1 expression was sufficient to completely restore P/I-induced NF-κB activation in JPM50.6 (Fig. 6,I, lower, lanes 2–4), the overexpression of CARD9 failed to do so at all (Fig. 6 I, lower, lanes 5–7). Thus, these results indicate that CARD9 is not capable of coupling PKC stimulation to NF-κB activation even when it exists abundantly in lymphoid cells.

Taken together, these results indicate that the selective involvement of CARMA1 but not CARD9 in T, B, and NK cells and the involvement of CARD9 but not CARMA1 in DCs and macrophages in ITAM receptor signaling are attributed to cell type-specific regulation mechanisms based on the ability of these adaptors to be a target for PKC stimulation, but not to the lineage-specific expression of these adaptors. Thus, CARD9 and CARMA1 are functionally not interchangeable.

We have shown herein that CARMA1-Bcl10 signaling is essential for cytokine/chemokine production but not for cytotoxicity mediated by ITAM-associated NK cell receptors. These results indicate that the signaling pathways required by these two effector functions triggered by NK cell receptors are regulated by distinct signaling machineries. Several previous studies using gene-knockout mice support this idea.

Using mice deficient for ITAM-bearing adaptor molecule DAP12, CD3ζ, or FcRγ, and the critical scaffold protein of ITAM-dependent signal, LAT (linker for activation of T cells) or NTAL (non-T cell activation linker), Chiesa et al. showed that NK cell cytotoxicity and IFN-γ secretion are initiated by ITAM-dependent and -independent as well as LAT/NTAL-dependent and -independent pathways (43).

CD45−/− NK cells displayed defective cytokine/chemokine production upon NK cell receptor stimulation but normal cytotoxicity (23, 44), similar to Card11−/− and Bcl10−/− NK cells. However, the molecular mechanisms underlying the defect seem to be different. MAPK activation after receptor engagement was mainly perturbed in CD45−/− NK cells (23, 44), whereas NF-κB activation was markedly inhibited in Card11−/− NK cells (Fig. 5). Analysis of NF-κB activation in CD45−/− NK cells may clarify the connection of CD45-regulated signaling to the CARMA1-Bcl10-mediated NF-κB activation.

We showed that CARMA1-Bcl10 signaling is also essential for cytokine/chemokine production mediated by NKG2D (Fig. 2), which associates with DAP10 in naive NK cells or with both DAP12 and DAP10 in activated NK cells (45, 46). The cytoplasmic regions of DAP12 and DAP10 contain an ITAM that binds to Syk or ZAP-70 and the YINM sequence that binds to PI3K, respectively. The latter is also capable of binding to Grb-2, which mediates recruitment and activation of downstream effector molecules, such as PLCγ2, and Vav1, as well as other Rho family proteins (47, 48). It has been shown that NK cells in Zap-70−/−Syk−/− mice as well as DAP12−/− mice retained the ability to lyse tumor targets via NKG2D, but exhibited abrogated cytokine production in response to NKG2D stimulation (28). In contrast, the PI3K inhibitor wortmannin completely abrogates NKG2D-mediated cytotoxicity (49). Thus, DAP12 and DAP10 appear to activate independent signaling pathways leading to distinct cellular responses triggered by NKG2D; that is, cytokine production and cytotoxicity, respectively. These reports and our results collectively indicate that the CARMA1-Bcl10 complex functions downstream of DAP12-ITAM-Syk/Zap-70-mediated signaling through NKG2D. However, NKG2D-induced cytokine production in freshly isolated naive NK cells that express mainly the long isoform of NKG2D (NKG2D-L), which associates exclusively with DAP10 (46), was also markedly reduced in Card11−/− mice (data not shown), suggesting that the CARMA1-Bcl10 pathway is also capable of mediating DAP10-PI3K-dependent cellular activation. This is also consistent with a previous report that CARMA1-deficient Jurkat cells abrogated Akt-mediated NF-κB activation (50).

In contrast to CARMA1 and Bcl10 deficiency, Vav1 deficiency results in normal IFN-γ production but reduced tumor lysis and exocytosis of cytotoxic granules triggered by various activating NK cell receptors (24). In line with this, it was reported that natural cytotoxicity and ADCC are both decreased in human NK cell lines in the presence of a dominant-negative form of Rac1 (51). These results suggest that the Vav1-Rac1 pathway is required for ITAM-mediated cytotoxicity but not for the CARMA1-Bcl10-mediated pathways for cytokine production in NK cells. These are consistent with our observation that Vav1 phosphorylation following CD16 crosslinking was not affected in Card11−/− NK cells (Fig. 5).

It has been reported that mouse NK cells require PLCγ2-mediated signals for both cytotoxicity and cytokine production (52). Stimulation with P/I mimics PLCγ-mediated generation of 1,2-diacylglycerol and intracellular calcium flux, thus directly stimulating PKCs, Ras-MAPKs, and calcium-dependent signaling. We found that Card11−/− NK cells showed impaired cytokine/chemokine production upon stimulation with P/I due to abrogated NF-κB activation but normal MAPK activation, indicating that CARMA1 acts downstream of PKCs and selectively regulates NF-κB but not MAPK activation. This is consistent with studies showing the critical phosphorylation of CARMA1 linker region for TCR- or BCR-mediated NF-κB activation (39, 40). Therefore, signaling pathways required for NK cell cytotoxicity and cytokine/chemokine production branch at least downstream of PLCγ2 and PKCs; the former pathway probably involves calcium mobilization and Vav1-Rac1 activation, and the latter involves CARMA1-Bcl10, leading to NF-κB activation.

We have shown herein that CARMA1 is dispensable for the development and terminal maturation of NK cells. ITAM-mediated signals are known to be crucial for the development of T and B lymphocytes (53). NK cells and lymphocytes are thought to originate in common precursor cells (54), and, indeed, usage of similar/common intracellular signaling components and machinery by TCRs/BCRs and activating NK cell receptors has been suggested (8). However, in contrast to lymphocyte development, the deficiency of genes involved in activating NK cell receptor signaling, such as CD3ζ/FcRγ/DAP12, LAT/NTAL (43), and Syk/ZAP-70 (55), has no or minimal effect on NK cell development. Similarly, whereas T and NK T cell development was impaired in Vav1−/− and CD45−/− mice, NK cell development and maturation in those mice were not compromised except for the absolute numbers of NK cells in spleen and BM, and they were increased (23, 24). These results indicate that the development and maturation of NK cells do not depend on ITAM-mediated activation signals. In contrast, the observation that CD45−/− and Syk−/−Zap-70−/− mice have an alteration in the Ly49 repertoire in peripheral NK cells (23, 55) indicates that some ITAM-mediated signals might influence repertoire formation of NK cell receptors. However, we showed that the lack of CARMA1 did not influence the repertoire formation (Table I). Regarding lymphocyte development, although CARMA1, Bcl10, or MALT1 deficiency also does not affect conventional T and B cell development (11, 12, 13, 14), it has been reported that PKCθ-Bcl10-IKKβ signaling is required for the normal development of NK T and regulatory T cells (56). Therefore, the requirement of CARMA1-Bcl10 signaling in the development of lymphoid cells may reflect the requirement of NF-κB activation in the development of lymphoid cell populations.

We have previously reported significant reductions in the percentages and absolute numbers of splenic NK cells in Card11−/− mice with 129J and C57BL/6 mixed genetic background (14). However, in the present analyses of mice that were backcrossed to C57BL/6 mice more than six times, we observed no such developmental abnormality in NK cells of Card11−/− mice between 3 and 17 wk of age (Fig. 4 and data not shown). Thus, although the reasons are still unclear, under the condition of CARMA1 deficiency, some non-MHC background genes might influence the development and/or peripheral maintenance of NK cells.

Bcl10-MALT1 signaling is essential for NF-κB activation mediated by TCRs and BCRs in lymphocytes (11, 12, 13), FcεRI in mast cells (37), and FcγR, OSCAR, TREM1, MAIRII, and dectin-1 in macrophages and DCs (17). All of those receptors deliver signals by associating with ITAM-containing signaling adaptors or by their own cytoplasmic ITAMs. Our data extend that knowledge to NK cells by demonstrating that Bcl10 is also essential for NF-κB activation through various ITAM-associated activating NK cell receptors (Fig. 7). However, studies have suggested that adaptors coupling receptors to the Bcl10-MALT1 unit vary by receptor and cell type. In lymphocytes, CARMA1 is essential for Bcl10-mediated NF-κB activation through TCRs and BCRs (14), whereas CARD9 is dispensable for the signaling pathway (17) (Fig. 7). In contrast, CARMA1 is dispensable while CARD9 is essential for the ITAM-associated receptor-mediated NF-κB activation in myeloid cells (17) (Fig. 7). Recent reports have suggested that G protein-coupled receptor-induced NF-κB activation also depends on the Bcl10-MALT1 unit and is mediated by another CARD-containing MAGUK family protein, CARMA3 (37, 57). These reports have also suggested that CARMA1- or CARMA3-coupled receptors, TCR, BCR, and G protein-coupled receptor, are regulated by PKC activation. In the present study, despite the fact that ITAM receptors in myeloid and NK cells deliver signals through common adaptors DAP12 and FcRγ, we found that CARD9 was dispensable for the Bcl10-mediated ITAM receptor signaling in NK cells, whereas CARMA1 was the essential molecule mediating this pathway as well as PKC activation, similar to TCR and BCR signaling (Fig. 7). These results indicate that the signal diversity depending on the PKC dependence of the Bcl10-MALT1-mediated NF-κB activation is determined by usage of upstream Bcl10-binding adaptors, CARMA1 and CARD9, but not receptor-coupled ITAM-containing adaptors. Although the cell type specificity of these two adaptors might be attributed to their specific expression (i.e., CARMA1 in lymphoid lineage cells and CARD9 in myeloid lineage cells), our data suggested that CARD9 and CARMA1 are functionally not interchangeable (Fig. 6). The disability of CARD9 in lymphoid ITAM-mediated signaling should be explained by the lack of a site in CARD9 to be phosphorylated by PKC. However, it remains a mystery why CARMA1 is not able to function in myeloid ITAM-mediated signaling despite its abundant expression and importance in PKC-mediated NF-κB activation in myeloid cells (Fig. 6). Nevertheless, it clearly indicates that there is an unknown myeloid cell-specific regulation mechanism coupling ITAM receptor stimulation to the CARD9-Bcl10 complex independent of PKC (Fig. 7). Yang et al. have recently shown that the VHL tumor suppressor gene, pVHL, bound to casein kinase 2 (CK2) promotes the inhibitory phosphorylation of CARD9 by CK2, thereby reducing renal carcinogenesis by downregulating NF-κB activity (58). Thus, it might be one candidate for the myeloid cell-specific signaling mechanism linking ITAM receptors to CARD9.

FIGURE 7.

Cell type-specific regulation mechanism of ITAM receptor-mediated NF-κB activation through L-CBM and M-CBM. Activation of NF-κB in immune cells through ITAM receptors is mediated by two different complexes containing Bcl10-MALT1: the lymphoid type (L)-CARMA1-Bcl10-MALT1 (L-CBM) complex and the myeloid type (M)-CARD9-Bcl10-MALT1 (M-CBM) complex. L-CBM is regulated by PKC and acts in TCR, BCR, and activating NK cell receptor signaling, while M-CBM acts in ITAM receptor signaling independent of the PKC-CARMA1 system in macrophages and DCs.

FIGURE 7.

Cell type-specific regulation mechanism of ITAM receptor-mediated NF-κB activation through L-CBM and M-CBM. Activation of NF-κB in immune cells through ITAM receptors is mediated by two different complexes containing Bcl10-MALT1: the lymphoid type (L)-CARMA1-Bcl10-MALT1 (L-CBM) complex and the myeloid type (M)-CARD9-Bcl10-MALT1 (M-CBM) complex. L-CBM is regulated by PKC and acts in TCR, BCR, and activating NK cell receptor signaling, while M-CBM acts in ITAM receptor signaling independent of the PKC-CARMA1 system in macrophages and DCs.

Close modal

Collectively, in innate and adaptive immunity, the activation of lymphoid (T, B, and NK) and myeloid (macrophage and dendritic) cells through ITAM receptors leading to NF-κB activation is mediated by two different complexes containing Bcl10-MALT1: the lymphoid-type (L)-CARMA1-Bcl10-MALT1 (L-CBM) complex and the myeloid-type (M)-CARD9-Bcl10-MALT1 (M-CBM) complex, respectively (Fig. 7). Additionally, signaling mediated by these complexes is differently regulated by cell-type-specific mechanisms based on PKC dependence. Our findings demonstrate that therapeutic approaches targeting L-CBM, M-CBM, or their specific regulator(s), such as PKC, would enable the specific modulation of lymphoid cells and myeloid cells, effectively activating or inhibiting their functional responses.

During submission of this paper, a similar study that ITAM-containing receptors on NK cells mediate activation signals through the CBM complex has recently been reported (Gross et al., Blood online January 2008, DOI 10.1182/blood-2007-11-123513).

We thank N. Suzuki, N. Shinobu, S. Yamasaki, T. Yokosuka, A. Tane, and T. Imanishi for discussion, H. Arase and X. Lin for reagents, and H. Yamaguchi for secretarial assistance.

The authors have no financial conflicts of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by a Grant-in-Aid for Research in Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan; Takeda Science Foundation; National Cancer Institute Grant R01 CA87064 (to S.W.M.); Cancer Center CORE Grant CA21765; and the American Lebanese Syrian Associated Charities, St. Jude Children’s Research Hospital.

3

Abbreviations used in this paper: PLC, phospholipase C; ADCC, Ab-dependent cell cytotoxicity; BM, bone marrow; BMDC, bone marrow-derived dendritic cell; CARD, caspase recruitment domain; CBM, CARMA1-Bcl10-MALT1; DC, dendritic cell; IKK, IκB kinase; LAMP, lysosomal-associated membrane protein; L-CBM, lymphoid-type CARMA1-Bcl10-MALT1; M-CBM, myeloid-type CARD9-Bcl10-MALT1; P/I, PMA plus ionomycin; PKC, protein kinase C; WT, wild type.

1
Yokoyama, W. M., B. F. Plougastel.
2003
. Immune functions encoded by the natural killer gene complex.
Nat. Rev. Immunol.
3
:
304
-316.
2
Lanier, L. L..
2005
. NK cell recognition.
Annu. Rev. Immunol.
23
:
225
-274.
3
Tassi, I., J. Klesney-Tait, M. Colonna.
2006
. Dissecting natural killer cell activation pathways through analysis of genetic mutations in human and mouse.
Immunol. Rev.
214
:
92
-105.
4
Isakov, N..
1998
. Role of immunoreceptor tyrosine-based activation motif in signal transduction from antigen and Fc receptors.
Adv. Immunol.
69
:
183
-247.
5
Samelson, L. E..
2002
. Signal transduction mediated by the T cell antigen receptor: the role of adapter proteins.
Annu. Rev. Immunol.
20
:
371
-394.
6
Jain, J., C. Loh, A. Rao.
1995
. Transcriptional regulation of the IL-2 gene.
Curr. Opin. Immunol.
7
:
333
-342.
7
Schulze-Luehrmann, J., S. Ghosh.
2006
. Antigen-receptor signaling to nuclear factor κB.
Immunity
25
:
701
-715.
8
Vivier, E., J. A. Nunes, F. Vely.
2004
. Natural killer cell signaling pathways.
Science
306
:
1517
-1519.
9
Humphrey, M. B., L. L. Lanier, M. C. Nakamura.
2005
. Role of ITAM-containing adapter proteins and their receptors in the immune system and bone.
Immunol. Rev.
208
:
50
-65.
10
Robinson, M. J., D. Sancho, E. C. Slack, S. LeibundGut-Landmann, C. Reis e Sousa.
2006
. Myeloid C-type lectins in innate immunity.
Nat. Immunol.
7
:
1258
-1265.
11
Ruland, J., G. S. Duncan, A. Elia, I. del Barco Barrantes, L. Nguyen, S. Plyte, D. G. Millar, D. Bouchard, A. Wakeham, P. S. Ohashi, T. W. Mak.
2001
. Bcl10 is a positive regulator of antigen receptor-induced activation of NF-κB and neural tube closure.
Cell
104
:
33
-42.
12
Ruefli-Brasse, A. A., D. M. French, V. M. Dixit.
2003
. Regulation of NF-κB-dependent lymphocyte activation and development by paracaspase.
Science
302
:
1581
-1584.
13
Ruland, J., G. S. Duncan, A. Wakeham, T. W. Mak.
2003
. Differential requirement for Malt1 in T and B cell antigen receptor signaling.
Immunity
19
:
749
-758.
14
Hara, H., T. Wada, C. Bakal, I. Kozieradzki, S. Suzuki, N. Suzuki, M. Nghiem, E. K. Griffiths, C. Krawczyk, B. Bauer, et al
2003
. The MAGUK family protein CARD11 is essential for lymphocyte activation.
Immunity
18
:
763
-775.
15
Hara, H., C. Bakal, T. Wada, D. Bouchard, R. Rottapel, T. Saito, J. M. Penninger.
2004
. The molecular adapter Carma1 controls entry of IκB kinase into the central immune synapse.
J. Exp. Med.
200
:
1167
-1177.
16
Xue, L., S. W. Morris, C. Orihuela, E. Tuomanen, X. Cui, R. Wen, D. Wang.
2003
. Defective development and function of Bcl10-deficient follicular, marginal zone and B1 B cells.
Nat. Immunol.
4
:
857
-865.
17
Hara, H., C. Ishihara, A. Takeuchi, T. Imanishi, L. Xue, S. W. Morris, M. Inui, T. Takai, A. Shibuya, S. Saijo, et al
2007
. The adaptor protein CARD9 is essential for the activation of myeloid cells through ITAM-associated and Toll-like receptors.
Nat. Immunol.
8
:
619
-629.
18
Bertin, J., Y. Guo, L. Wang, S. M. Srinivasula, M. D. Jacobson, J. L. Poyet, S. Merriam, M. Q. Du, M. J. Dyer, K. E. Robison, et al
2000
. CARD9 is a novel caspase recruitment domain-containing protein that interacts with BCL10/CLAP and activates NF-κB.
J. Biol. Chem.
275
:
41082
-41086.
19
Malarkannan, S., J. Regunathan, H. Chu, S. Kutlesa, Y. Chen, H. Zeng, R. Wen, D. Wang.
2007
. Bcl10 plays a divergent role in NK cell-mediated cytotoxicity and cytokine generation.
J. Immunol.
179
:
3752
-3762.
20
Arase, H., E. S. Mocarski, A. E. Campbell, A. B. Hill, L. L. Lanier.
2002
. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors.
Science
296
:
1323
-1326.
21
Yamasaki, S., E. Ishikawa, M. Sakuma, K. Ogata, K. Sakata-Sogawa, M. Hiroshima, D. L. Wiest, M. Tokunaga, T. Saito.
2006
. Mechanistic basis of pre-T cell receptor-mediated autonomous signaling critical for thymocyte development.
Nat. Immunol.
7
:
67
-75.
22
Wang, D., Y. You, S. M. Case, L. M. McAllister-Lucas, L. Wang, P. S. DiStefano, G. Nunez, J. Bertin, X. Lin.
2002
. A requirement for CARMA1 in TCR-induced NF-κB activation.
Nat. Immunol.
3
:
830
-835.
23
Huntington, N. D., Y. Xu, S. L. Nutt, D. M. Tarlinton.
2005
. A requirement for CD45 distinguishes Ly49D-mediated cytokine and chemokine production from killing in primary natural killer cells.
J. Exp. Med.
201
:
1421
-1433.
24
Colucci, F., E. Rosmaraki, S. Bregenholt, S. I. Samson, V. Di Bartolo, M. Turner, L. Vanes, V. Tybulewicz, J. P. Di Santo.
2001
. Functional dichotomy in natural killer cell signaling: Vav1-dependent and -independent mechanisms.
J. Exp. Med.
193
:
1413
-1424.
25
Penack, O., C. Gentilini, L. Fischer, A. M. Asemissen, C. Scheibenbogen, E. Thiel, L. Uharek.
2005
. CD56dimCD16neg cells are responsible for natural cytotoxicity against tumor targets.
Leukemia
19
:
835
-840.
26
Arase, H., N. Arase, T. Saito.
1996
. Interferon γ production by natural killer (NK) cells and NK1.1+ T cells upon NKR-P1 cross-linking.
J. Exp. Med.
183
:
2391
-2396.
27
Munder, M., M. Mallo, K. Eichmann, M. Modolell.
1998
. Murine macrophages secrete interferon γ upon combined stimulation with interleukin (IL)-12 and IL-18: a novel pathway of autocrine macrophage activation.
J. Exp. Med.
187
:
2103
-2108.
28
Zompi, S., J. A. Hamerman, K. Ogasawara, E. Schweighoffer, V. L. Tybulewicz, J. P. Di Santo, L. L. Lanier, F. Colucci.
2003
. NKG2D triggers cytotoxicity in mouse NK cells lacking DAP12 or Syk family kinases.
Nat. Immunol.
4
:
565
-572.
29
Kim, S., K. Iizuka, H. S. Kang, A. Dokun, A. R. French, S. Greco, W. M. Yokoyama.
2002
. In vivo developmental stages in murine natural killer cell maturation.
Nat. Immunol.
3
:
523
-528.
30
Lanier, L. L..
1998
. NK cell receptors.
Annu. Rev. Immunol.
16
:
359
-393.
31
Raulet, D. H., R. E. Vance, C. W. McMahon.
2001
. Regulation of the natural killer cell receptor repertoire.
Annu. Rev. Immunol.
19
:
291
-330.
32
Kim, S., J. Poursine-Laurent, S. M. Truscott, L. Lybarger, Y. J. Song, L. Yang, A. R. French, J. B. Sunwoo, S. Lemieux, T. H. Hansen, W. M. Yokoyama.
2005
. Licensing of natural killer cells by host major histocompatibility complex class I molecules.
Nature
436
:
709
-713.
33
Samson, S. I., O. Richard, M. Tavian, T. Ranson, C. A. Vosshenrich, F. Colucci, J. Buer, F. Grosveld, I. Godin, J. P. Di Santo.
2003
. GATA-3 promotes maturation, IFN-γ production, and liver-specific homing of NK cells.
Immunity
19
:
701
-711.
34
Townsend, M. J., A. S. Weinmann, J. L. Matsuda, R. Salomon, P. J. Farnham, C. A. Biron, L. Gapin, L. H. Glimcher.
2004
. T-bet regulates the terminal maturation and homeostasis of NK and Vα14i NKT cells.
Immunity
20
:
477
-494.
35
Taki, S., S. Nakajima, E. Ichikawa, T. Saito, S. Hida.
2005
. IFN regulatory factor-2 deficiency revealed a novel checkpoint critical for the generation of peripheral NK cells.
J. Immunol.
174
:
6005
-6012.
36
Dong, C., R. J. Davis, R. A. Flavell.
2002
. MAP kinases in the immune response.
Annu. Rev. Immunol.
20
:
55
-72.
37
Klemm, S., J. Gutermuth, L. Hultner, T. Sparwasser, H. Behrendt, C. Peschel, T. W. Mak, T. Jakob, J. Ruland.
2006
. The Bcl10–Malt1 complex segregates FcεRI-mediated nuclear factor κB activation and cytokine production from mast cell degranulation.
J. Exp. Med.
203
:
337
-347.
38
Hijikata, A., H. Kitamura, Y. Kimura, R. Yokoyama, Y. Aiba, Y. Bao, S. Fujita, K. Hase, S. Hori, Y. Ishii, et al
2007
. Construction of an open-access database that integrates cross-reference information from the transcriptome and proteome of immune cells.
Bioinformatics
23
:
2934
-2941.
39
Matsumoto, R., D. Wang, M. Blonska, H. Li, M. Kobayashi, B. Pappu, Y. Chen, X. Lin.
2005
. Phosphorylation of CARMA1 plays a critical role in T cell receptor-mediated NF-κB activation.
Immunity
23
:
575
-585.
40
Sommer, K., B. Guo, J. L. Pomerantz, A. D. Bandaranayake, M. E. Moreno-Garcia, Y. L. Ovechkina, D. J. Rawlings.
2005
. Phosphorylation of the CARMA1 linker controls NF-κB activation.
Immunity
23
:
561
-574.
41
Khoshnan, A., D. Bae, C. A. Tindell, A. E. Nel.
2000
. The physical association of protein kinase Cθ with a lipid raft-associated inhibitor of κB factor kinase (IKK) complex plays a role in the activation of the NF-κB cascade by TCR and CD28.
J. Immunol.
165
:
6933
-6940.
42
Krappmann, D., A. Patke, V. Heissmeyer, C. Scheidereit.
2001
. B-cell receptor- and phorbol ester-induced NF-κB and c-Jun N-terminal kinase activation in B cells requires novel protein kinase C’s.
Mol. Cell. Biol.
21
:
6640
-6650.
43
Chiesa, S., M. Mingueneau, N. Fuseri, B. Malissen, D. H. Raulet, M. Malissen, E. Vivier, E. Tomasello.
2006
. Multiplicity and plasticity of natural killer cell signaling pathways.
Blood
107
:
2364
-2372.
44
Hesslein, D. G., R. Takaki, M. L. Hermiston, A. Weiss, L. L. Lanier.
2006
. Dysregulation of signaling pathways in CD45-deficient NK cells leads to differentially regulated cytotoxicity and cytokine production.
Proc. Natl. Acad. Sci. USA
103
:
7012
-7017.
45
Gilfillan, S., E. L. Ho, M. Cella, W. M. Yokoyama, M. Colonna.
2002
. NKG2D recruits two distinct adapters to trigger NK cell activation and costimulation.
Nat. Immunol.
3
:
1150
-1155.
46
Raulet, D. H..
2003
. Roles of the NKG2D immunoreceptor and its ligands.
Nat. Rev. Immunol.
3
:
781
-790.
47
Billadeau, D. D., J. L. Upshaw, R. A. Schoon, C. J. Dick, P. J. Leibson.
2003
. NKG2D-DAP10 triggers human NK cell-mediated killing via a Syk-independent regulatory pathway.
Nat. Immunol.
4
:
557
-564.
48
Upshaw, J. L., L. N. Arneson, R. A. Schoon, C. J. Dick, D. D. Billadeau, P. J. Leibson.
2006
. NKG2D-mediated signaling requires a DAP10-bound Grb2-Vav1 intermediate and phosphatidylinositol-3-kinase in human natural killer cells.
Nat. Immunol.
7
:
524
-532.
49
Bonnema, J. D., L. M. Karnitz, R. A. Schoon, R. T. Abraham, P. J. Leibson.
1994
. Fc receptor stimulation of phosphatidylinositol 3-kinase in natural killer cells is associated with protein kinase C-independent granule release and cell-mediated cytotoxicity.
J. Exp. Med.
180
:
1427
-1435.
50
Narayan, P., B. Holt, R. Tosti, L. P. Kane.
2006
. CARMA1 is required for Akt-mediated NF-κB activation in T cells.
Mol. Cell. Biol.
26
:
2327
-2336.
51
Billadeau, D. D., K. M. Brumbaugh, C. J. Dick, R. A. Schoon, X. R. Bustelo, P. J. Leibson.
1998
. The Vav-Rac1 pathway in cytotoxic lymphocytes regulates the generation of cell-mediated killing.
J. Exp. Med.
188
:
549
-559.
52
Caraux, A., N. Kim, S. E. Bell, S. Zompi, T. Ranson, S. Lesjean-Pottier, M. E. Garcia-Ojeda, M. Turner, F. Colucci.
2006
. Phospholipase C-γ2 is essential for NK cell cytotoxicity and innate immunity to malignant and virally infected cells.
Blood
107
:
994
-1002.
53
DeFranco, A. L..
1995
. Transmembrane signaling by antigen receptors of B and T lymphocytes.
Curr. Opin. Cell Biol.
7
:
163
-175.
54
Ikawa, T., H. Kawamoto, S. Fujimoto, Y. Katsura.
1999
. Commitment of common T/natural killer (NK) progenitors to unipotent T and NK progenitors in the murine fetal thymus revealed by a single progenitor assay.
J. Exp. Med.
190
:
1617
-1626.
55
Colucci, F., E. Schweighoffer, E. Tomasello, M. Turner, J. R. Ortaldo, E. Vivier, V. L. Tybulewicz, J. P. Di Santo.
2002
. Natural cytotoxicity uncoupled from the Syk and ZAP-70 intracellular kinases.
Nat. Immunol.
3
:
288
-294.
56
Schmidt-Supprian, M., J. Tian, E. P. Grant, M. Pasparakis, R. Maehr, H. Ovaa, H. L. Ploegh, A. J. Coyle, K. Rajewsky.
2004
. Differential dependence of CD4+CD25+ regulatory and natural killer-like T cells on signals leading to NF-κB activation.
Proc. Natl. Acad. Sci. USA
101
:
4566
-4571.
57
Grabiner, B. C., M. Blonska, P. C. Lin, Y. You, D. Wang, J. Sun, B. G. Darnay, C. Dong, X. Lin.
2007
. CARMA3 deficiency abrogates G protein-coupled receptor-induced NF-κB activation.
Genes Dev.
21
:
984
-996.
58
Yang, H., Y. A. Minamishima, Q. Yan, S. Schlisio, B. L. Ebert, X. Zhang, L. Zhang, W. Y. Kim, A. F. Olumi, W. G. Kaelin, Jr.
2007
. pVHL Acts as an adaptor to promote the inhibitory phosphorylation of the NF-κB agonist Card9 by CK2.
Mol. Cell
28
:
15
-27.