Dendritic Cell Immunoactivating Receptor, a Novel C-type Lectin Immunoreceptor, Acts as an Activating Receptor through Association with Fc Receptor γ Chain*

An increasing number of C-type lectin receptors are being discovered on dendritic cells, but their signaling abilities and underlying mechanisms require further definition. Among these, dendritic cell immunoreceptor (DCIR) induces negative signals through an inhibitory immunoreceptor tyrosine-based inhibitory motif (ITIM) in its cytoplasmic tail. Here we identify a novel C-type lectin receptor, dendritic cell immunoactivating receptor (DCAR), whose extracellular lectin domain is highly homologous to that of DCIR. DCAR is expressed similarly in tissues to DCIR, but its short cytoplasmic portion lacks signaling motifs like ITIM. However, a positively charged arginine residue is present in the transmembrane region of the DCAR, which may explain its association with Fc receptor γ chain and its stable expression on the cell surface. Furthermore, cross-linking of DCAR in the presence of γ chain activates calcium mobilization and tyrosine phosphorylation of cellular proteins. These signals are mediated by the immunoreceptor tyrosine-based activating motif (ITAM) of the γ chain. Thus, DCAR is closely related to DCIR, but it introduces activating signals into antigen-presenting cells through its physical and functional association with ITAM-bearing γ chain. The identification of this activating immunoreceptor provides an example of signaling via a dendritic cell-expressed C-type lectin receptor.

itory receptor and an immunoreceptor tyrosine-based activating motif (ITAM)-bearing activating receptor. Such pairs of receptors, when expressed on the same cell, allow for a balance between positive and negative cell signaling. In most cases, activating receptors are formed as complexes of a ligand-binding subunit with a short cytoplasmic tail and an ITAM-bearing signal-transducing subunit. This signal-transducing subunit can be shared by several receptors. These activating receptors were first identified on effector cells such as natural killer (NK), T, B, and mast cells and subsequently detected on myeloid cells including granulocytes, macrophages, and dendritic cells (DC). Now it is evident that such receptors are used rather ubiquitously, including in platelets (5) and even nonhematopoietic cells (6). Based on their domain structures, these receptors are divided into two subgroups, Ig superfamily and Ca 2ϩdependent (C-type) lectin family. Besides their structural similarity, the genes for these families are clustered (e.g. the Ig superfamily in the leukocyte receptor complex on human chromosome 19 or syntenic mouse chromosome 7, and the C-type lectin family in the NK gene complex (NKC) on human chromosome 12 or syntenic mouse chromosome 6, respectively) (7).
On the other hand, there have been many C-type lectin receptors detected on DC. Based on their molecular structures, they include two types of receptors: type I C-type lectin as a type I transmembrane protein with several carbohydrate recognition domains (CRDs) and type II C-type lectin as a type II transmembrane protein with a single CRD. Both types are considered to function mainly as pattern recognition receptors for antigen capture (8) and additionally in interactions of DC with other cells (9). For example, macrophage mannose receptor and DC-SIGN may act in DC trafficking (10,11), and DC-SIGN and DC-associated C-type lectin (Dectin)-1 can mediate an interaction of DC and T cells (12,13). Since some of the genes for type II lectin receptors on DC are mapped close to the NKC, one might suspect their additional functions in cellular signaling.
We have previously shown that one of these receptors, DC immunoreceptor (DCIR), originally identified as an ITIM-bearing type II lectin immunoreceptor expressed on antigen-presenting cells (14), actually acts as an ITIM-dependent inhibitory receptor in B cells (15). Here we describe the further identification of DC immunoactivating receptor (DCAR) as a novel immunoreceptor closely related with DCIR, and we successfully show its activating capacity through the ITAM of its associated Fc receptor (FcR) ␥ chain. Our findings illustrate that DC can express potentially paired signaling immunoreceptors with C-type lectin external domains. * This work was supported in part by Ministry of Education, Sciences, Sports, and Culture of Japan Grant 14770406 (to N. K.). 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.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AY230259 (DCAR␣) and AY230260 (DCAR␤).

EXPERIMENTAL PROCEDURES
Cloning of Full-length Mouse DCAR cDNA-Total RNA, which was isolated from ICR mouse spleen using TRIzol reagent (Invitrogen), was reverse transcribed using oligo(dT) [12][13][14][15][16][17][18] primer and the SuperScript First-Strand Synthesis System (Invitrogen). Using this cDNA as a template, PCR was performed. TaKaRa Taq TM (Takara Shuzo Co. Ltd.) and the following primers were used: 5Ј-GGA GTT CTG GCC TGT TTG AAA G-3Ј (forward primer), 5Ј-AGT CCC TGA GTC ATA TCT TCA AG-3Ј (nested forward primer), and 5Ј-TGA TTC ATA AGT TTA TTT TCT TCA TCT G-3Ј (reverse primer). Conditions adapted for Gene-Amp® PCR System 9700 (Applied Biosystems) were as follows: 3-min denaturation at 95°C and 40 or 30 cycles of 94°C for 10 s, 55°C for 30 s, and 72°C for 60 s for the first or nested PCR, respectively. Rapid amplifications of cDNA ends (RACE) were performed using Marathon-Ready TM cDNA of BALB/c mouse spleen and Advantage TM 2 Polymerase Mix (BD Biosciences Clontech) as follows: 30-s denaturation at 94°C, 5 cycles of 94°C for 5 s and 72°C for 4 min, 5 cycles of 94°C for 5 s and 70°C for 4 min, and 30 or 20 cycles of 94°C for 5 s and 68°C for 4 min, for the first or nested PCR, respectively. The following primers were used: for 5Ј-RACE, 5Ј-CCC TTC ACT GAA GCA GGT CAA TGA ATG G-3Ј and 5Ј-CGC AAT GAA ACA GGT GCT GAG GAG TAA G-3Ј (nested primer); for 3Ј-RACE, 5Ј-AGT CCC TGA GTC ATA TCT TCA AG-3Ј and 5Ј-ATG GTT CAG GAA AGA CAG CTA CAA G-3Ј (nested primer). The amplified cDNA fragments were purified and subcloned into pGEM®-T Easy Vector (Promega), and their base sequences were determined. Handling of nucleotide and amino acid sequences and alignment of amino acid sequences were performed using GeneWorks® and MacVector™ software (Oxford Molecular Group Inc.), respectively.
Immunoprecipitation and Western Blotting-To detect association of DCAR and ␥ chain, transfected 293T cells were solubilized in 1% digi- The six cysteine residues that may be involved in formation of the C-type lectin fold are circled. B, genomic structure of the DCAR gene. TM, transmembrane domain. C, schematic view of the genes of DCAR and related C-type lectins localized on a short region of mouse chromosome 6. Each gene is shown as a box.

RESULTS
Identification of the cDNA of DCAR-Using a search of cDNA and amino acid sequences homologous to the mouse DCIR, one uncharacterized but highly homologous gene was found. This gene, 1810046I24Rik, was originally obtained from the fulllength cDNA library of mouse pancreas and appears in Gen-Bank TM under the accession number NM_027218 (30). To confirm the nucleotide sequence of its predicted coding region, FIG. 2. Comparison of DCAR amino acid sequences with those of related lectins. A, alignment of deduced amino acid sequences of DCAR and DCIR is shown. Identical and homologous residues are shaded dark and light, respectively. The potential transmembrane domain is underlined, and the ITIM of DCIR is double underlined. The cysteine residue that may be involved in dimer formation is indicated by a closed square, and the six conserved cysteine residues in the C-type lectin fold are shown by asterisks. The EPS motif, which may be involved in specificity of the recognizing sugar, is shown by a plus sign. B, amino acid sequence alignment of DCAR with other lectins having short cytoplasmic tails, Dectin-2 and hBDCA-2. Dark and light shadings represent identical and homologous amino acids, respectively. The potential transmembrane domain is underlined, and the positively charged amino acid residue is indicated by an asterisk.
heminested RT-PCR was performed using spleen mRNA as a template. The obtained products were separated in agarose gel electrophoresis, and one major band and minor bands with various sizes were observed. DNA was extracted from the major band to determine its base sequence, and two isoforms were identified as the result. The major isoform was 99 base pairs longer than the other minor one, identical to NM_027218, and these were named DCAR ␣ and ␤, respectively. To isolate the 5Ј-and 3Ј-untranslated regions of DCAR, RACE experiments were performed, and the base sequences of the obtained cDNA fragments were determined. A contiguous contig representing the full-length cDNA of DCAR was composed by alignment of these three sequences including results of 5Ј-RACE, RT-PCR, and 3Ј-RACE, each of which represented at least six different clones. This contig is 737 base pairs in length, excluding the poly(A) tail, and contains a putative open reading frame of 627 base pairs (Fig. 1A). Given a potential start codon is present in the consensus Kozak sequence (31) and a stop codon exists at an upstream position (TAG: nucleotides 89 -91), the encoded protein should contain 209 amino acids. The presence of a putative hydrophobic signal anchor, consisting of 27 amino acids (residues 15-41; underlined in Fig. 1A), indicates that this polypeptide is a type II transmembrane protein. Since its extracellular domain contains a single CRD, DCAR represents a type II C-type lectin.
Genomic Structure and Localization of the DCAR Gene-The 1810046I24Rik gene, identical to DCAR␤, is localized on mouse chromosome 6 supercontig Mm6_WIFeb01_114 (GenBank TM accession number NW_000264). Nucleotide sequence alignment of DCAR and this supercontig enabled us to demonstrate the genomic structure of the DCAR gene with appropriate exon-intron boundaries except for exon III, which was deleted in the ␤ isoform. The DNA fragments containing introns II and III were obtained by PCR, and their exon-intron boundaries were confirmed. The whole structure of the DCAR gene is shown in Fig. 1B. It is composed of six exons representing functional domains. Exon I encodes the 5Ј-untranslated region and 8 amino acids in the cytoplasmic domain, exon II encodes 30 amino acids mainly containing the transmembrane domain (TM in Fig. 1), exon III encodes 33 amino acids representing the neck domain, and the remaining CRD is encoded by 3 exons (IV, V, and VI). DCAR maps close to DCIR, and their genes form a cluster with other related C-type lectins including Dectin-2, macrophage-restricted C-type lectin (MCL) (32), and macrophage-inducible C-type lectin (Mincle) (33) in a short region of the NW_000264 supercontig (Fig. 1C). The human genomic locus syntenic to this cluster is reportedly located on the telomeric end of the NKC (34).
DCAR Is Related Closely to DCIR-Amino acid sequence alignment of DCAR and DCIR revealed that their CRDs were highly homologous (91% identity), whereas their cytoplasmic, transmembrane, and neck domains were quite different ( Fig.   2A). The cytoplasmic domain of DCAR was much shorter than that of DCIR and lacked tyrosine residue and any signaling motif such as ITIM. These results suggest that DCAR and DCIR form putative paired immunoreceptors.
DCAR Contains a Conserved Charged Amino Acid in Its Transmembrane Domain-Dectin-2, identical to NKCL (35), and human BDCA-2 (hBDCA-2) (36), identical to human DC lectin (37), are also C-type lectins, having short cytoplasmic domains without any signal transduction motif. Amino acid sequence alignment of DCAR and these lectins demonstrated that they contain quite similar numbers of amino acids and considerable sequence homology (Fig. 2B), especially the cytoplasmic to neck portions (82% identity for DCAR and Dectin-2). Notably, the charged arginine (DCAR and Dectin-2) and lysine (hBDCA-2) residues were present at the conserved position in their transmembrane domains, and these have been reported to be responsible for an association with adaptor molecules in many immunoreceptors.
Expression Profile of DCAR Compared with That of DCIR-Specific expression of DCAR was only examined by RT-PCR, because it was considered to be quite difficult to produce specific Ab recognizing the extracellular domain of DCAR, due to the high sequence homology between the CRDs of DCAR and DCIR as well as the neck domains of DCAR and Dectin-2. The useful forward primer specific for DCAR was prepared in its 5Ј-untranslated region. Among normal tissues, DCAR transcripts were detected strongly in lung and spleen and weakly in skin and lymph node, whereas DCIR transcripts were more ubiquitously observed (Fig. 3). During differentiation of BM-DC, DCAR expression reached a maximum at day 8 versus day 10 in the case of DCIR. BM-NK cells expressed neither DCAR nor DCIR. These data indicate that the expression pattern of DCAR is similar to but not identical to DCIR.
Association of DCAR with FcR ␥ Chain-When transfected in 293T cells, FLAG-tagged DCAR by itself was expressed very weakly on the cell surface, in contrast to strong expression of DCIR and almost no expression of Dectin-2 on the transfectants (Fig. 4A, top column). To determine whether an adaptor molecule can associate with DCAR, 293T cells were transfected with a mixture of FLAG-tagged DCAR and each one of the known adaptor molecules, including DAP12, DAP10, CD3 chain, and FcR ␥ chain, and analyzed with FCM. Only in the case of cotransfection with ␥ chain was surface expression of DCAR significantly enhanced (Fig. 4A, middle two columns). Transfection of ␥ chain alone could not induce any protein expression on the cell surface detected by anti-FLAG M2 (data not shown). The association of DCAR and ␥ chain was further confirmed by immunoprecipitation. 293T cells transfected with FLAG-tagged DCAR and ␥ chain were lysed, and the lysates were immunoprecipitated with anti-FLAG M2. The immunoprecipitates were separated by SDS-PAGE, and the associated ␥ chain was visualized with immunoblotting (Fig. 4B). These RT-PCR was performed using specific primer pairs of DCAR, DCIR, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) on various mouse tissues and cell populations. Amplification of 30 cycles was performed for DCAR and DCIR, whereas amplification of 25 cycles was performed for glyceraldehyde-3-phosphate dehydrogenase. Obtained fragments were separated with 2% agarose gel electrophoresis. data indicate that DCAR and ␥ chain interact in the cotransfected 293T cells and that this interaction is involved in transport of DCAR to the cell surface. Furthermore, to examine the contribution of membranous arginine residue of DCAR to this association, two kinds of mutant DCAR, whose arginine was mutated to noncharged isoleucine (R/I) and negatively charged aspartic acid (R/D), were prepared and used for cotransfection experiments. Unexpectedly, R/I and R/D mutations only partially but similarly reduced the cell surface expression of DCAR in the presence of ␥ chain (bottom column of Fig. 4A), indicating that the conserved charged arginine residue in the transmembrane domain of DCAR is at most partly responsible for the association with ␥ chain. Mutation of ITAM-tyrosines in the ␥ chain (amino acids 65 and 76 mutated to phenylalanines; Y/F) did not influence the expression level of DCAR cotransfected with intact ␥ chain (bottom column of Fig. 4A). This result indicates that the ITAM of ␥ chain has no effect in this association.
Calcium Mobilization after Ligation of DCAR-FcR through the ITAM of Associated ␥ Chain-An Fc ␥ receptor-negative derivative of mouse A20 B cells, IIA1.6, which we previously used for detection of inhibitory function of DCIR (15), did not express ␥ chain (28). We used these cells to detect the activating capacity of DCAR as well as the role of ␥ chain in DCARmediated activation. For this purpose, a chimeric receptor containing the cytoplasmic to transmembrane portion of DCAR and extracellular domain of mouse Fc ␥ receptor IIB (DCAR-FcR) was constructed and transfected into A20 IIA1.6 cells. After analyzing the expression level of DCAR-FcR on the surface of the transformants with FCM, one clone clearly expressing DCAR-FcR was selected. After sequential transfection of ␥ chain or Y/F mutant ␥ chain, the clones showing comparably enhanced surface expression levels of DCAR-FcR were selected (Fig. 5A). The comparable expression levels of ␥ chain and mutant ␥ chain in the obtained transformants were confirmed by immunoprecipitation followed by Western blotting using anti-␥ (Fig. 5B).
We first examined the effect of DCAR-FcR ligation on Ca 2ϩ mobilization using Fura Red as an indicator. IC were composed of anti-Fc ␥ receptor (CD16/CD32) 2.4G2 and mouse anti-rat IgG F(abЈ) 2 and used for ligation of DCAR-FcR (27). F(abЈ) 2 anti-mouse IgG was used as the positive control stimulator, which can aggregate B cell receptor and induce cellular activation (15). After the addition of F(abЈ) 2 anti-mouse IgG, all of the transformants showed an immediate decrease and gradual recovery of cellular Fura Red, indicating a transient rapid increase of intracellular Ca 2ϩ concentration (Fig. 6A, lower column). A similar increase in cellular Ca 2ϩ concentration was observed when IC were added to the transformant expressing both DCAR-FcR and ␥ chain, but not when IC were added to the transformant expressing DCAR-FcR alone or the one expressing DCAR-FcR and Y/F mutant ␥ chain (Fig. 6A, upper  column). These data indicate that ligation of cytoplasmic DCAR connected with ␥ chain induces Ca 2ϩ mobilization and that the ITAM of ␥ chain is responsible for this effect.
Tyrosine Phosphorylation of Cellular Proteins after Ligation of DCAR-FcR through the ITAM of Associated ␥ Chain-We further examined the effect of DCAR-FcR ligation on tyrosine phosphorylation of various cellular proteins with immunoprecipitation followed by Western blotting using anti-Tyr(P) PY20. Although bands of tyrosine-phosphorylated proteins were faintly observed before stimulation, the number and intensity of the bands increased after IC were added to the transformant expressing DCAR and ␥ chain. The degree of this increase was as high as that seen after the addition of F(abЈ) 2 anti-mouse IgG ligating B cell receptor (Fig. 6B, left panel). In contrast, when IC were added to the transformant expressing DCAR-FcR alone and the one expressing DCAR-FcR and Y/F mutant ␥ chain, the number and intensity of the bands did not significantly change compared with those seen when IC were added to the parental cells (Fig. 6B, right panel). These data indicate that ligation of cytoplasmic DCAR connected with ␥ chain is also able to induce tyrosine phosphorylation of cellular proteins and that the ITAM of ␥ chain is responsible for this effect. DISCUSSION We report the identification and characterization of a novel C-type lectin immunoreceptor, DCAR. Similar to other paired immunoreceptors, DCAR and DCIR share highly homologous extracellular domains, whereas their intracellular domains are quite different; the DCAR domain is very short and lacks signaling motifs, whereas that of DCIR is longer and contains an ITIM. The presence of a charged arginine residue in the transmembrane domain of DCAR encouraged us to investigate the possibility that DCAR acts as an activating receptor in association with an adaptor molecule. The most widely used ITAM- bearing adaptor molecule would be DAP12 (38), and its critical role in DC antigen-priming capacity as well as in NK cell activation was demonstrated by the analysis of DAP12-deficient mice (39,40). In fact, myeloid DAP12-associating lectin-1 (41), triggering receptor expressed on myeloid cells-2 (42), and signal-regulatory protein ␤ (43,44) have been identified as DAP12-associating immunoreceptors expressed in myeloid cells. In contrast, Ig-like transcript-1 (45) and paired Ig-like receptor-A (28) are associated with FcR ␥ chain. Our results clearly show that DCAR is an additional receptor physically and functionally associating with ␥ chain. Tomasello et al. (44) compared the positions of charged amino acid residues in the transmembrane domains of DAP12-associating receptors with those of ␥ chain-associating receptors and showed their distinctive structural features: the former in the center of the transmembrane domain and the latter close to the extracellular domain. It should be noted that a position near the ectodomain in a type I membrane protein corresponds to a position close to the cytoplasm in the case of type II proteins, such as NKR-P1A. Consistent with this role, the position of the membranous arginine residue of DCAR, which associates with ␥ chain, is very close to its intracellular domain. Since a previous study showed a critical role for the negatively charged aspartic acid of DAP12 in its association with myeloid DAP12-associating lectin-1 (41), we examined the importance of the positively charged arginine of DCAR. Unexpectedly, mutation analyses could not demonstrate an essential role of this charged amino acid for an interaction with ␥ chain, only a partial contribution.
Dectin-2 and hBDCA-2 also have a charged arginine and lysine residue, respectively, in the conserved position of their transmembrane domains, suggesting that these two receptors also work as activating receptors in the presence of appropriate adaptor molecules. However, cotransfection with any of the four adaptor molecules (DAP12, DAP10, CD3, and FcR␥) could not enhance the surface expression of Dectin-2 at all (data not shown). As Fernandes et al. (35) proposed, Dectin-2 (NKCL) might associate with an as yet unidentified signaling molecule. Another possibility might be that Dectin-2 with no signaling capacity can form a heterodimer with DCAR and/or DCIR, similar to the CD94 NK cell receptors, which form heterodimers with activating NKG2C and inhibitory NKG2A molecules (46). In contrast, triggering hBDCA-2 can induce Ca 2ϩ mobilization and protein tyrosine phosphorylation, suggesting that these activation signals are mediated by some associated adaptor molecule (36). The FcR ␥ chain would be the most possible candidate, considering its sequence similarity with DCAR. Although it has been proposed that Dectin-2 represents the mouse homologue of hBDCA-2 (36), we have newly identified the predicted human counterpart of Dectin-2, which is clearly distinct from hBDCA-2, 2 and it is possible that Dectin-2 and hBDCA-2 have distinctive roles.
We have successfully shown the in vitro activatory effects of cross-linking chimeric DCAR-FcR and its associated FcR ␥ chain, as in Ca 2ϩ mobilization and protein tyrosine phosphorylation. The association with ITAM-bearing ␥ chain is critical for the signal transduction we observed. Compared with the inhibitory effects of DCIR shown by the similar experiments (15), the function of DCAR shows a clear contrast with DCIR, further suggesting that they form a functional pair. Unfortunately, their natural ligands are unknown, and their in vivo functions remain to be resolved. Similar to many paired immunoreceptors, their extracellular CRDs are highly homologous but are not completely the same. This fact raises two possibil-ities; the same ligand can bind both receptors with different affinity, or similar but distinct molecules bind to the specific receptors. Indeed, the EPS sequence is conserved in both receptors as well as the human counterpart of DCIR, but specific carbohydrates that bind to the CRD containing this motif have not yet been identified. Human BDCA-2, which has activating properties in vitro, further inhibits interferon-␣/␤ production by plasmacytoid DC (36). Engering et al. (47) have proposed an immune escape theory in which virus binding to hBDCA-2 can down-regulate plasmacytoid DC function by inhibiting activation signals mediated by pathogen recognition receptors such as Toll-like receptors. Both DCAR and DCIR can also possibly modulate intracellular signaling induced when pathogens or self-antigens act through their specific receptors.
Our RT-PCR analyses showed that the specific transcripts of these receptors had a similar but clearly distinct pattern. The observed ubiquitous expression of inhibitory however, DCIR in a variety of tissues may possibly contribute to the maintenance of the immunological homeostasis. It should be noted that DCAR mRNA expression by itself does not represent its surface protein expression, because DCAR can sufficiently be expressed on the cell surface only in the presence of FcR ␥ chain. Considering that the expression level of ␥ chain does not remarkably change during maturation of BM-DC (data not shown), however, DCAR mRNA expression may control its surface protein expression in these cells. On day 8 and day 10, mRNA expression of DCAR and DCIR reached maximum, respectively, possibly contributing to the observation that T cell activating capacity reached a maximum on day 8, whereas the number of apoptotic cells increased on day 10 (16). Although expression of DCAR and DCIR was not specifically observed in DC, their expression in each DC subset or regulation of their expression should be examined to further determine their roles in the DC system.
In our previous report, rabbit polyclonal Abs raised against the two polypeptides in the CRD of DCIR were used for detecting DCIR protein in vivo (15). Amino acid sequence alignment revealed that one polypeptide (GHRQWQWVDQTPYEES) was fully identical to the corresponding region of DCAR and that the other (QSQEEQDFITGILDTH) was not the same but was quite similar to the corresponding DCAR polypeptide (14 of 16 residues identical). Thus, the published data of the FCM analysis using the Ab for the former polypeptide (15) should be reinterpreted to indicate that either DCIR or DCAR is expressed on the surface of major histocompatibility complex class II-positive antigen-presenting cells. Although FCM analysis with both Abs revealed almost the same expression pattern (15), suggesting the expression of DCIR and DCAR on the same cell surface, precise expression profiles for each molecule need to be determined at the single cell level. For this reason, we cannot definitely conclude that DCIR and DCAR are paired immunoreceptors. Preparation of monoclonal Ab recognizing each molecule would resolve this issue.
In conclusion, we have reported the molecular cloning and functional characterization of DCAR, a novel C-type lectin immunoreceptor expressed on DC. DCAR mediates activating signals through associated FcR ␥ chain and represents a potent immunoreceptor forming a pair with inhibitory DCIR. The presence of such putative paired immunoreceptors among DCexpressing C-type lectins provides a possible model of their signaling capacity.