1024
Biochemical Society Transactions (2006) Volume 34, part 6
Negative regulation of leucocyte functions by CD33-related siglecs T. Avril, H. Attrill, J. Zhang, A. Raper and P.R. Crocker1 Wellcome Trust Biocentre, University of Dundee, Dow Street, Dundee DD1 5EH, Scotland, U.K.
Abstract The siglecs (sialic acid-binding Ig-like lectins) are a family of transmembrane receptors expressed in the haemopoietic, immune and nervous systems. The CD33-related siglecs are a distinct subset mostly expressed in the innate immune system where they can function as inhibitory receptors by suppressing the signalling mediated by receptors coupled with ITAMs (immunoreceptor tyrosine-based activation motifs). CD33-related siglecs contain ITIMs (immunoreceptor tyrosine-based inhibitory motifs) that recruit and activate SHP-1 [SH2 (Src homology 2) domain-containing phosphatase-1] and SHP-2. In addition, the ITIMs of CD33-related siglecs can suppress siglec-dependent adhesion of sialylated ligands and mediate endocytosis. Siglec-H is a recently characterized murine CD33-related endocytic receptor that lacks intrinsic tyrosine-based signalling motifs and is expressed selectively on PDCs (plasmacytoid dendritic cells). Siglec-H depends on DAP12 (DNAX-activating protein of 12 kDa) for surface expression and cross-linking with anti-siglec-H antibodies can selectively inhibit interferon-α production by PDCs following TLR9 (Toll-like receptor 9) ligation. Thus CD33-related siglecs are able to mediate diverse inhibitory functions of leucocytes in the innate immune system via both ITIM-dependent and -independent pathways.
Introduction Cellular activation within the innate and adaptive immune systems often proceeds via cross-linking of receptors that are coupled with ITAMs (immunoreceptor tyrosine-based activation motifs), as exemplified by Fc receptors of myeloid cells, the T-cell receptor complex and the B-cell receptor [1]. This mode of signalling can be counterbalanced by membrane receptors containing cytoplasmic ITIMs (immunoreceptor tyrosine-based inhibitory motifs; consensus sequence I/L/S/VXYXXL/V, where X is any amino acid). Both ITAMs and ITIMs are phosphorylated by Src-like kinases following receptor clustering. ITAMs then recruit tandem SH2 (Src homology 2) tyrosine kinases, either Syk or ZAP-70 [ζ -chain (T-cell receptor)-associated protein kinase of 70 kDa] that trigger cellular activation, whereas ITIMs recruit and activate the protein tyrosine phosphatases SHP-1 (SH2 domain-containing phosphatase-1) and SHP-2 or the lipid phosphatase SHIP-1 (SH2 domain-containing 5 inositol phosphatase-1) and antagonize ITAM-dependent activation. A recent proteomic survey [2] revealed there are more than 100 membrane receptors containing one or more ITIMs, including many novel putative inhibitory receptors, thereby
Key words: endocytosis, immunoglobulin (Ig) superfamily, innate immune system, leucocyte, plasmacytoid dendritic cell, sialic acid. Abbreviations used: DAP12, DNAX-activating protein of 12 kDa; EAT2, Ewing’s sarcomaassociated transcript 2; IFN, interferon; ITAM, immunoreceptor tyrosine-based activation motif; ITIM, immunoreceptor tyrosine-based inhibitory motif; ITSM, immunoreceptor tyrosine-based switch motif; mAb, monoclonal antibody; NF-κB, nuclear factor κB; PD-1, programmed death 1; PDC, plasmacytoid dendritic cell; RBL, rat basophil leukaemia; SH2, Src homology 2; SHIP-1, SH2 domain-containing 5 inositol phosphatase-1; SHP, SH2 domain-containing phosphatase; siglec, sialic acid-binding Ig-like lectin; SLAM, signalling lymphocytic activation molecule; SAP, SLAM-associated protein; TLR, Toll-like receptor. To whom correspondence should be addressed (email
[email protected]).
1
C 2006
Biochemical Society
highlighting the extensive interplay between ITAMs and ITIMs in regulating cellular activation in the immune system. However, there is growing evidence that so-called ‘inhibitory’ receptors can trigger activation and conversely that ‘activatory’ receptors can trigger inhibition ([3], but see [3a], [4]). In the present review, we focus on the recently described CD33-related siglecs [sialic acid-binding Ig-like lectins], a major subset of inhibitory receptors of the innate immune system that appear to mediate their inhibitory functions using both ITIM-dependent and -independent pathways.
The siglec family Siglecs contain a homologous N-terminal Ig V-set domain that mediates sialic acid binding, and varying numbers of C2 set Ig domains that extend the sialic acid-binding site from the plasma membrane [5,6]. Sialic acids are commonly found at the exposed termini of oligosaccharide chains attached to proteins and lipids and there are probably hundreds or thousands of potential ligands for siglecs presented on cell surfaces. However, each siglec exhibits a preference for not only the type of sialic acid, but also its linkage to subterminal sugars. Binding specificity can also be strongly influenced by sugar residues that are positioned distally from the terminal sialic acid, probably via interactions with amino acids in the highly variable C-C β-strand loops of siglec V-set domains that function synergistically with the primary conserved sialic acid-binding sites positioned on the A, F and G β-strands [7]. Siglecs can be divided into two subgroups based on sequence similarity. Sialoadhesin (siglec-1), CD22 (siglec-2) and MAG (myelin-associated glycoprotein; siglec-4) are more distantly related (∼25–30% sequence identity) and have
Control of Immune Responses
Figure 1 ITIM-containing human CD33-related siglecs and their cellular expression profiles The ITIM motif is represented by a filled box in the cytoplasmic tails and ITIM-like motifs are open boxes. cDCs, conventional dendritic cells, pDCs, PDCs; NK cells, natural killer cells; (+), weakly expressed; +, expressed moderately or strongly.
clear-cut orthologues in all mammalian species examined. In comparison, the CD33-related siglecs share approx. 50– 90% identity and appear to be evolving rapidly via multiple processes involving gene duplication, exon shuffling, exon loss and gene conversion [8]. This has resulted in major differences in repertoires of CD33-related siglecs among mammalian species including the great apes. Several siglecs are expressed in a cell type-restricted pattern. For example, sialoadhesin is a macrophage-specific adhesion molecule and CD22 is a well-characterized B-cell-inhibitory receptor [6]. The CD33-related siglecs show a more complex pattern of expression on cells of the innate immune system (Figure 1) and, despite the differences in repertoires seen in humans and mice, they are represented collectively on the same cell populations [9].
Negative signalling mediated by CD33-related siglecs Most CD33-related siglecs contain a single conserved membrane-proximal ITIM and a membrane-distal ITIMlike motif. For most of the CD33-related siglecs, the distal motif corresponds to the ITSM (immunoreceptor tyrosinebased switch motif) (TXYXXI/V where X is any amino acid) defined originally in the SLAM (signalling lymphocytic activation molecule)/CD150 family of regulatory receptors [10]. In SLAM family molecules, this motif can bind to the small SH2 domain proteins, SAP (SLAM-associated protein) in its non-phosphorylated state or its homologue EAT2 (Ewing’s sarcoma-associated transcript 2) after tyrosine phosphorylation. SAP can recruit the haemopoietic-specific Fyn kinase, FynT, and trigger cellular activation [11]. However, binding of EAT2 seems to result in inhibitory signalling, notably in myeloid cells such as macrophages and dendritic cells where EAT2 is abundant [11]. As yet, there is no evidence that the ITSM-like motif of CD33-related siglecs can interact
with SAP or EAT2 as also noted for some other ITSMcontaining receptors such as the PD-1 (programmed death 1) receptor [12], and the potential of siglecs to trigger signalling via these pathways remains unclear. Several studies have used transfection-based approaches to demonstrate the inhibitory signalling functions of CD33related siglecs, including inhibition of the ITAM-containing receptor Fcγ RI (high-affinity IgE receptor)-dependent calcium flux in human U937 myelomonocytic cells [13,14] and inhibition of T-cell receptor signalling and proliferation in primary human T-cells [15] and the Jurkat T-cell line [16]. Another approach has been to modulate cellular functions using anti-siglec antibodies followed by cross-linking with secondary antibodies. For example, anti-CD33 and -siglec-7 mAbs (monoclonal antibodies) were found to inhibit proliferation of myeloid leukaemia cells [17,18] and anti-siglec8 and -siglec-9 mAbs could trigger apoptosis of human eosinophils and neutrophils respectively [19,20]. Studies from our laboratory have utilized the RBL-2H3 (where RBL is rat basophil leukaemia) model to investigate not only the signalling potential of CD33-related siglecs but also the relative importance of the ITIM and ITSM-like motifs for functions mediated by siglec-5, -7 and -9 [21,22]. RBL cells naturally express the γ -chain ITAM-coupled high-affinity IgRI which, when cross-linked, triggers the degranulation of RBL cells as measured through release of serotonin. The antibody-mediated co-cross-linking of transfected receptors with IgRI can be used to measure their inhibitory activity in a well-controlled manner. Results from these experiments showed that all three CD33-related siglecs examined were as potent in inhibition as a well-characterized inhibitory receptor, KIR2DL3 (CD158b), examined in parallel [21,22]. Similarly to several other related inhibitory receptors, the membrane-proximal ITIM of all CD33-related siglecs examined was essential for high-level, phosphotyrosinedependent recruitment of SHP-1 and SHP-2. The distal C 2006
Biochemical Society
1025
1026
Biochemical Society Transactions (2006) Volume 34, part 6
ITSM-like motif was important for SHP-1 recruitment but its mutation did not affect SHP-2 recruitment. Surprisingly, tyrosine to alanine mutations of these motifs either alone or in combination had no effect on siglec-5 inhibitory signalling, but a low-level phosphotyrosine-independent recruitment of SHP-1 was observed [21]. In vitro phosphatase assays showed that both non-phosphorylated cytoplasmic tail constructs and tyrosine to alanine double mutations of the ITIM and ITSM-like motifs could weakly activate SHP-1. These findings support a model whereby CD33-related siglecs can mediate weak interactions with SHP-1 even in the absence of tyrosine phosphorylation, leading to tonic suppression of cellular activation. This could be important in both cis and trans interactions of siglecs with sialylated ligands expressed on cell surfaces. In addition to suppression of cellular activation, the ITIMs of CD33-related siglecs are important for additional functions, including inhibition of siglec-dependent adhesion to sialylated ligands and endocytosis [21–25]. Mutation of the membrane-proximal ITIM tyrosine to either alanine or phenylalanine resulted in enhanced sialic acid-dependent binding of CD33 and siglec-5, -7 and -9 to human red blood cells [21–23]. The mechanisms are unknown, but since the mutations did not lead to changes in surface expression, the enhanced adhesion is likely to involve clustering effects at the cell surface that result in increased binding avidity to the sialylated red blood cell ligands. Enhanced adhesion of siglec-5, -7 and -9 mutants was found to correlate with reduced recruitment of SHP-2 and introduction of a dominantnegative form of SHP-2 was able to reverse the adhesionpromoting functions of siglec-5 [21], arguing for a role of this particular tyrosine phosphatase in downstream signalling, leading to reduced adhesive properties of these siglecs. An important question is how sialic acid recognition by siglecs translates into their inhibitory signalling functions, and recent work on CD22/siglec-2 has provided some insights. CD22 contains up to six tyrosine-based motifs, of which three function as ITIMs. These become tyrosinephosphorylated by Lyn (a Src-like kinase) following B-cell receptor ligation, leading to recruitment and activation of the SHP-1 tyrosine phosphatase and inhibition of cellular activation mediated via various pathways [26]. In CD22deficient mice, B-cells are hyperactivated, whereas in mice that lack CD22 ligands via inactivation of the ST6GalI sialyltransferase, B-cells are refractory to activation [27]. B-cells from CD22−/− ST6GalI−/− double knockout mice behave similarly to those from CD22-deficient mice, showing that the immunodeficiency of ST6GalI-deficient mice depends on the presence of CD22 [28,29]. Since CD22 appears to exist as homo-multimeric complexes on the surface of B-cells that limit the cis interactions of CD22 with other sialylated counter-receptors [30], its lectin function may serve to regulate non-sialic acid-dependent interactions with the Bcell receptor, preventing excessive inhibitory signalling and endocytosis [31]. This role in fine-tuning B-cell responses probably reflects a long-standing co-evolution of CD22 with the B-cell lineage and the highly orchestrated expression C 2006
Biochemical Society
of the ST6Gal1 in B-cells to regulate CD22 homotypic interactions. It seems rather unlikely that CD33-related siglecs mediate inhibitory functions via similar mechanisms since most members of this subgroup differ significantly from CD22 in several important respects. These include a broader range of binding specificities for sialylated glycans that involve multiple sialyltransferases, expression on a range of different cell types and marked species-dependent differences in repertoire. It is possible that CD33-related siglecs, which are known to be masked at the cell surface, mediate weak and transient sialic acid-dependent interactions in cis with multiple neighbours on the cell membrane, the net effect of which could be suppression of cellular activation potential depending on the cell type and its glycosylation status. Trans interactions with ‘high-affinity’ ligands on other cells may also occur and contribute to the dampening of autoreactivity by cells of the innate immune system.
Siglec-H is a PDC (plasmacytoid dendritic cell)-associated molecule lacking ITIMs and implicated in negative regulation of IFN-α (interferon-α) production During a PCR-based screen for novel murine siglecs, we identified a murine CD33-related siglec, siglec-H, that lacks an ITIM or ITIM-like motif. By raising antibodies to a soluble recombinant form of this protein, we found that siglec-H has a remarkable restriction to PDCs [32]. PDCs are key cells in mammalian first-line defence to viruses since they are able to produce extremely high levels of type I IFN following activation of TLR (Toll-like receptor)-dependent and -independent signalling pathways. Besides inhibiting viral replication, type I IFN has a number of important effects on the immune system and can modulate the functions of T-cells, B-cells, NK (natural killer) cells and conventional dendritic cell populations. Thus PDCs are considered as a key link between the innate and adaptive immune systems. Siglec-H was identified independently by Colonna’s group as the antigen recognized by the PDC-specific mAb 440c, which had been shown to inhibit CpG-dependent production of IFN-α via TLR9 ligation [33]. Interestingly, siglec-H expression depends on the presence of DAP12 (DNAXactivating protein of 12 kDa), an ITAM-containing adaptor normally associated with cellular activation events. Siglec-H contains a lysine residue within the transmembrane region typical of DAP12-coupled receptors, suggesting that these proteins are physically associated in the plasma membrane [34]. In our laboratory, we failed to see any effects of siglec-H ligation on production of cytokines such as TNF-α and IL6 (interleukin-6) that depend on NF-κB (nuclear factor κB) activation [32], but recent studies have shown that TLR9 signalling occurs in two distinct endosomal environments, with IRF7 (IFN regulatory factor 7)-dependent IFN-α production occurring in early endosomes and NF-κB-dependent cytokine production in late endosomes [35]. Thus targeting of siglec-H–DAP12 complexes to specific endocytic compartments could lead to selective inhibition of IFN-α
Control of Immune Responses
Figure 2 Siglec-H is a DAP12-coupled receptor expressed on PDCs When cross-linked with anti-siglec-H antibodies, this leads to selective inhibition of IFN-α production following TLR9 ligation with CpG-containing oligonucleotides.
3 3a 4 5 6 7
8 9 10
11 12 13 14 15 16 17
18 19
production by mechanisms that remain to be established (Figure 2). Another function that could be mediated by siglec-H is cross-presentation of antigens on MHC class I antigens to trigger CD8 T-cell responses to viruses, a crucial component of adaptive immunity to viral infections. This was demonstrated by coupling siglec-H antibodies with ovalbumin followed by injection into mice in the presence of CpG [32]. Although the ligands recognized by siglec-H are unknown, one interesting possibility is that siglec-H has evolved to function as a pattern-recognition molecule that binds viral or other pathogen ligands. By delivering these to the endosomal compartment, it could favour interactions with TLR7 and TLR9 as well as the antigen-processing machinery and provide a negative feedback mechanism to limit production of IFN-α. Work in P.R.C.’s laboratory is supported by a Wellcome Trust Senior Fellowship.
20 21 22 23 24 25 26 27 28 29 30 31
32 33 34
References 1 Humphrey, M.B., Lanier, L.L. and Nakamura, M.C. (2005) Immunol. Rev. 208, 50–65 2 Staub, E., Rosenthal, A. and Hinzmann, B. (2004) Cell. Signalling 16, 435–456
35
Barrow, A.D. and Trowsdale, J. (2006) Eur. J. Immunol. 36, 1646–1653 Erratum (2006) Eur. J. Immunol. 36, 2276 Hamerman, J.A. and Lanier, L.L. (2006) Science STKE 2006, re1 Varki, A. and Angata, T. (2006) Glycobiology 16, 1R–27R Crocker, P.R. (2005) Curr. Opin. Pharmacol. 5, 431–437 Attrill, H., Takazawa, H., Witt, S., Kelm, S., Isecke, R., Brossmer, R., Ando, T., Ishida, H., Kiso, M., Crocker, P.R. and van Aalten, D.M. (2006) Biochem. J. 397, 271–278 Angata, T., Margulies, E.H., Green, E.D. and Varki, A. (2004) Proc. Natl. Acad. Sci. U.S.A. 101, 13251–13256 Zhang, J.Q., Biedermann, B., Nitschke, L. and Crocker, P.R. (2004) Eur. J. Immunol. 34, 1175–1184 Shlapatska, L.M., Mikhalap, S.V., Berdova, A.G., Zelensky, O.M., Yun, T.J., Nichols, K.E., Clark, E.A. and Sidorenko, S.P. (2001) J. Immunol. 166, 5480–5487 Veillette, A. (2006) Nat. Rev. Immunol. 6, 56–66 Chemnitz, J.M., Parry, R.V., Nichols, K.E., June, C.H. and Riley, J.L. (2004) J. Immunol. 173, 945–954 Paul, S.P., Taylor, L.S., Stansbury, E.K. and McVicar, D.W. (2000) Blood 96, 483–490 Ulyanova, T., Shah, D.D. and Thomas, M.L. (2001) J. Biol. Chem. 276, 14451–14458 Nguyen, D.H., Hurtado-Ziola, N., Gagneux, P. and Varki, A. (2006) Proc. Natl. Acad. Sci. U.S.A. 103, 7765–7770 Ikehara, Y., Ikehara, S.K. and Paulson, J.C. (2004) J. Biol. Chem. 279, 43117–43125 Vitale, C., Romagnani, C., Falco, M., Ponte, M., Vitale, M., Moretta, A., Bacigalupo, A., Moretta, L. and Mingari, M.C. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 15091–15096 Balaian, L. and Ball, E.D. (2004) Leuk. Res. 28, 821–829 Nutku, E., Aizawa, H., Hudson, S.A. and Bochner, B.S. (2003) Blood 101, 5014–5020 von Gunten, S., Yousefi, S., Seitz, M., Jakob, S.M., Schaffner, T., Seger, R., Takala, J., Villiger, P.M. and Simon, H.U. (2005) Blood 106, 1423–1431 Avril, T., Freeman, S.D., Attrill, H., Clarke, R.G. and Crocker, P.R. (2005) J. Biol. Chem. 280, 19843–19851 Avril, T., Floyd, H., Lopez, F., Vivier, E. and Crocker, P.R. (2004) J. Immunol. 173, 6841–6849 Taylor, V.C., Buckley, C.D., Douglas, M., Cody, A.J., Simmons, D.L. and Freeman, S.D. (1999) J. Biol. Chem. 274, 11505–11512 Walter, R.B., Raden, B.W., Kamikura, D.M., Cooper, J.A. and Bernstein, I.D. (2005) Blood 105, 1295–1302 Biedermann, B., Gil, D., Bowen, D.T. and Crocker, P.R. (2006) Leukocyte Res., doi:10.1016/j.leukres.2006.05.026 Nitschke, L. (2005) Curr. Opin. Immunol. 17, 290–297 Hennet, T., Chui, D., Paulson, J.C. and Marth, J.D. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 4504–4509 Ghosh, S., Bandulet, C. and Nitschke, L. (2006) Int. Immunol. 18, 603–611 Collins, B.E., Smith, B.A., Bengtson, P. and Paulson, J.C. (2006) Nat. Immunol. 7, 199–206 Han, S., Collins, B.E., Bengtson, P. and Paulson, J.C. (2005) Nat. Chem. Biol. 1, 93–97 Grewal, P.K., Boton, M., Ramirez, K., Collins, B.E., Saito, A., Green, R.S., Ohtsubo, K., Chui, D. and Marth, J.D. (2006) Mol. Cell. Biol. 26, 4970–4981 Zhang, J., Raper, A., Sugita, N., Hingorani, R., Salio, M., Palmowski, M.J., Cerundolo, V. and Crocker, P.R. (2006) Blood 107, 3600–3608 Blasius, A., Vermi, W., Krug, A., Facchetti, F., Cella, M. and Colonna, M. (2004) Blood 103, 4201–4206 Blasius, A.L., Cella, M., Maldonado, J., Takai, T. and Colonna, M. (2006) Blood 107, 2474–2476 Guiducci, C., Ott, G., Chan, J.H., Damon, E., Calacsan, C., Matray, T., Lee, K.D., Coffman, R.L. and Barrat, F.J. (2006) J. Exp. Med. 203, 1999–2008
Received 31 July 2006
C 2006
Biochemical Society
1027
