Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Aug 1.
Published in final edited form as: Immunity. 2008 Jul 31;29(2):228–237. doi: 10.1016/j.immuni.2008.05.018

Novel Immune-Type Receptors Mediate Allogeneic Recognition

John P Cannon 1,2, Robert N Haire 1, Andrew T Magis 3, Donna D Eason 1,2, Kelley N Winfrey 4, Jose A Hernandez Prada 3, Kate M Bailey 3, Jean Jakoncic 5, Gary W Litman 1,2,4, David A Ostrov 3
PMCID: PMC2603606  NIHMSID: NIHMS66165  PMID: 18674935

SUMMARY

Novel immune-type receptors (NITRs) comprise an exceptionally large, diversified family of activating/inhibitory receptors that has been identified in bony fish. In this study, we characterize the structure of an activating NITR that is expressed by a cytotoxic NK-like cell line and specifically binds an allogeneic B cell target. A single amino acid residue within the NITR immunoglobulin variable (V)-type domain accounts for specificity of the interaction. Structures solved by x-ray crystallography reveal: (1) the V-type domains of NITRs form homodimers resembling heterodimers formed by rearranging antigen binding receptors and (2) both subunits of NITR dimers form ligand-binding surfaces in CDR1 that determine specificity for the nonself target. In the evolution of immune function, it appears that a specific NK-type of innate recognition may be mediated by a complex germline multigene family of V structures resembling those that are somatically diversified in adaptive immune responses.

INTRODUCTION

Studies in several different systems have established roles for activating/inhibitory immunoglobulin (Ig) domain-containing type I transmembrane proteins in a wide range of cell-cell interactions including various types of immunological recognition. The extracellular domains of these proteins can be of the C2-, I (intermediate)- or V (variable)-type. A particularly diverse family of these proteins, termed novel immune-type receptors (NITRs), has been identified in several major radiations of the bony fish (Strong et al., 1999; Yoder et al., 2004). Each NITR possesses either a single V domain or an N-terminal V domain and membrane-proximal I domain. The V domains are encoded in a diversified multigene family and can be classified in families corresponding to those seen in Ig and T cell antigen receptors (TCRs). NITRs vary in terms of the number of ectodomains, presence or absence of joining (J) motifs, presence or absence of charged transmembrane residues and composition of the cytoplasmic tail, including the presence or absence of an immunoreceptor tyrosine-based inhibitory motif (ITIM) or immunoreceptor tyrosine-based switch motif (ITSM). Despite the utility of the zebrafish (Danio rerio) and pufferfish (Spheroides nephelus) models, upon which most genomic information about the NITR gene complex is available, neither of these species is easily amenable to in vitro studies of immunological function.

The immune system of the channel catfish (Ictalurus punctatus) is the most extensively characterized of any bony fish or, for that matter, any ectothermic vertebrate. Cell lines have been established that exhibit the surface phenotypes and functions of T and B lymphocytes as well as cells that exhibit phenotypic characteristics resembling those seen in mammalian natural killer (NK) cells (Shen et al., 2004). At present, nine families of NITRs have been described in catfish. A number of different NITRs (Hawke et al., 2001; Evenhuis et al., 2007) are expressed by 3H9, a cell line shown previously to exhibit in vitro specificity for allogeneic B cell lines.

Based on both their domain organization and diverse V structures, we hypothesized that NITRs could be involved in allogeneic recognition (Yoder et al., 2001; Yoder et al., 2004; Hawke et al., 2001) and herein describe the binding of an activating catfish NITR to a non-self target B cell line. Alloreactive and nonreactive NITRs, as well as reciprocal gain and loss of function mutants, have been expressed, purified and crystallized to understand the structural basis for alternative binding specificities. Native NITR proteins and selenomethionyl (Se-Met) derivatives were crystallized and the structures were solved by x-ray crystallography using the single-wavelength anomalous dispersion technique (SAD) (Wang et al., 2004) and refined against x-ray diffraction data obtained from native NITR crystals. The results establish NITRs as structurally similar to antigen receptors in chain topology and dimerization mode. The crystal structures reconcile the functional binding specificity of NITRs with specific molecular features and reveal an additional mechanism of immune recognition by diversified V domain-containing proteins.

RESULTS

Interactions of Recombinant Catfish NITRs with the 1G8 B Cell Line

The 3H9 catfish cell line was cloned and selected from a long-term mixed lymphocyte reaction on the basis of stimulation by and killing of a γ—irradiated allogeneic B cell line. 3H9 possesses azurophilic cytoplasmic granules and is: surface Ig-, TCRα/β-, Sudan black B- and anti-NCC (5C6)-. 3H9 exhibits strong cytotoxicity for the B cell line 1G8 and comparatively weak cytotoxicity toward 3B11, an MHC I and MHC II haplotypic disparate B cell line (Shen et al., 2004). The ultrastructure, surface phenotype and specific cytotoxicity of 3H9 are characteristic of mammalian NK cells.

In order to examine potential interactions of NITRs with the 1G8 and 3B11 cell lines, a lentiviral vector was engineered to express FLAG epitope-tagged NITR ectodomains in a recombinant fusion protein containing the uncharged transmembrane region of MDIR2, which is an unrelated Ig superfamily (IgSF) receptor in the skate (Raja eglanteria) (Cannon et al., 2006), and the C-terminal cytoplasmic tail of mouse CD3ζ, which contains three separate immunoreceptor tyrosine-based activation motifs (ITAMs) (Fig. 1a). NITR-CD3ζ fusion constructs were transduced into the 43-1 T cell hybridoma line, which harbors a green fluorescent protein (GFP) reporter gene under the control of a nuclear factor of activated T cells (NFAT)-responsive promoter(Ohtsuka et al., 2004). Engagement of the receptor ectodomain on the surface of NITRCD3ζ-transduced 43-1 cells is expected to result in NFAT nuclear translocation and transactivation of GFP expression through CD3ζ-ITAM signaling. Potential weak or short-lived interactions at the cell surface can be detected. Similar approaches, using different constructs, have been used to assay lectin-type (Arase et al., 2002; Smith et al., 2002) and Ig-type (Khakoo et al., 2002) mammalian NK receptor-ligand interactions. The integrity and specificity of the CD3ζ signaling system was validated initially by measuring GFP expression following anti-epitope cross-linking of test constructs (data not shown).

Figure 1.

Figure 1

Open in a new tab

(a) An in vitro signaling assay for characterizing NITR-ligand interactions. The mouse T cell hybridoma line 43-1 contains a GFP transgene under the control of an NFAT-responsive promoter. A lentiviral vector containing an NITR ectodomain, the uncharged transmembrane region of skate MDIR2 and the cytoplasmic region of mouse CD3ζ, has been transduced into 43-1. (b-e) GFP fluorescence of 43-1 cells transduced with NITR11-CD3ζ, after incubation with either 3B11 (b, d) or 1G8 cells (c, e). b, c: Hoechst 33258 staining; d, e: GFP fluorescence. (f, g) Fluorescent staining of 1G8 B cells with hFcγ chimeras of NITR10 (f) or NITR11 (g); light micrographs of cells in each field are inset. GFP, green fluorescent protein; ITAM, immunoreceptor tyrosine-based activation motif; MDIR2, modular immune-type receptor 2; NFAT, nuclear factor of activated T cells; NITR, novel immune-type receptor; TM, transmembrane region.

cDNA fragments encoding the extracellular domains of nine separate catfish NITRs (NITR1 and NITR4-11), representing three different NITR subfamilies, were cloned into CD3ζ lentiviral reporter constructs, each of which was used to transduce the 43-1 hybridoma. Transduced hybridoma cells were co-incubated with either 1G8 or 3B11 and assayed for GFP expression after 20 hours. GFP induction was not observed when 43-1 cells expressing eight of the nine catfish NITR-CD3ζ fusions were exposed to either 1G8 or 3B11 cells; however, 43-1 cells expressing the NITR11-CD3ζ fusion demonstrated significant GFP expression after co-incubation with 1G8 but not 3B11 (Fig. 1b-e).

Binding of Soluble NITR11 to the 1G8 B Cell Line

The ectodomains of NITR10 and NITR11, derived specifically from the 3H9 NK-like cell line (Hawke et al., 2001), were cloned into a vector that allows expression of soluble human IgG1 Fc (hFcγ) chimeras in mammalian cells. NITR10 was chosen as the primary negative reference on the basis of both its failure to produce a signal in the chimeric protein signaling assays and its relatively close relationship to NITR11 (93% amino acid identity in the Ig domain and 81% identity across the entire extracellular region). The predicted N-terminus of mature NITR10 differs from that of NITR11 by the presence of four additional amino acids. Therefore, two variants of the NITR10 ectodomain were tested in hFcγ chimeras: NITR10 (wildtype [wt]), which bears an N-terminus identical to that of the NITR11 (wt) ectodomain, and NITR10 (wt-VTTS), which contains the four additional residues at the N-terminus (see Supplementary Fig. 2). When hFcγ chimeras of NITR10 (wt) and NITR11 (wt) were incubated with the catfish 1G8 and 3B11 cell lines, only NITR11 (wt) bound to the 1G8 target (Fig. 1f-g); neither construct demonstrated detectable binding to 3B11 (data not shown). hFcγ chimeras of NITR10 (wt-VTTS) behaved identically to NITR10 (wt), showing no detectable binding to either cell line (data not shown).

As a first step in addressing the nature of the interaction between NITR11 and its cell surface ligand(s), 1G8 cells were stained with NITR11-hFcγ chimeras after treatment with pH 2.5 glycine, which has been shown to influence antibody recognition of conformed MHC I in mammals by disruption of the interaction of MHC I α-chains with β2-microglobulin (β2m) (Gonen-Gross et al., 2005; Poláková et al., 2003). HeLa cells were included in the experiment as a positive control for disruption of conformed MHC I. Whereas a 50% decrease in mean fluorescence intensity was observed in anti-human MHC I antibody-stained HeLa cells following acid treatment, neither the mean fluorescence intensity nor the relative percentage of positive cells was altered in acid-treated 1G8 cells stained with NITR11-hFcγ (Supplementary Fig. 1). Acid treatment did not affect staining with an anti-heat shock protein 70 (anti-hsp70) antibody in either cell line. These results confirm earlier observations regarding human MHC I and do not show an effect of acid treatment on the interaction of NITR11 with 1G8.

Interactions of NITR10 and NITR11 with the 1G8 B Cell Line Depend upon a Single Amino Acid

The differential binding activities of NITR10 and NITR11 suggested that mutational analysis could provide useful information regarding the structural basis for the binding of NITR11 to 1G8. Reciprocal exchange constructs between NITR10 and NITR11 were cloned and their binding to the 1G8 cell line was assayed using soluble hFcγ chimeras. Based on the results of an initial subdomain exchange experiment (Supplementary Fig. 2), the critical region for interaction with the 1G8 cell line was localized to a pentapeptide stretch corresponding to positions 47-51 (GenPept AAL35551 and AAL35552) of NITR10/11, within the predicted loop between the B and C strands of the V domain (corresponding to the CDR1 region of antigen receptors). Because both initial exchange constructs contain the additional VTTS tetrapeptide from the predicted N-terminus of NITR10, this motif was ruled out as a basis for the differential binding to 1G8. All other sequence differences between the extracellular domains of NITR10 and NITR11, including variation in length of the membrane-proximal “stalk” regions, did not influence the interaction with 1G8 in the direct binding assays.

Further investigation of the critical five amino acid region was performed using site-directed mutagenesis (Supplementary Fig. 3). Ultimately, it was possible to demonstrate that a point mutation of NITR11 (N50D, substituting asparagine for aspartic acid at position 50) abolishes the interaction of NITR11 with 1G8. Conversely, the reciprocal point mutation of NITR10 (D50N) results in strong reactivity with the 1G8 cell line (Fig. 2). Neither mutation alters the pattern of potential glycosylation sites (Ohtsubo and Marth, 2006) as compared to wildtype; consistent with this prediction, both mutant NITR-hFcγ forms migrate identically to their wildtype counterparts in denaturing polyacrylamide gel electrophoresis (data not shown). Retrospective examination revealed that the identity of the amino acid at position 50 (Asp or Asn) is predictive of 1G8 binding activity in all twelve of the NITR10/11 variants that were tested (NITR10, NITR11, two subdomain exchange constructs and eight site-directed mutants): all NITR10/11 forms that contain Asn50 bind strongly to 1G8, whereas all of those containing Asp50 fail to exhibit robust binding (Fig. 2 and Supplementary Figs. 2 and 3). Together, these data demonstrate that the identity of the amino acid at position 50 of both NITR10 and NITR11 has a critical effect on the interaction of these receptors with 1G8.

Figure 2.

Figure 2

Open in a new tab

Pairwise flow cytometric comparisons of the interactions of NITR10, NITR11 and two NITR point mutants with the 1G8 catfish B cell line. (a) NITR10 (wildtype [wt]) vs. NITR11 (wt). (b) NITR11 (N50D) vs. NITR11 (wt). (c) NITR10 (wt) vs. NITR10 (D50N). (d) NITR10 (D50N) vs. NITR11 (wt). A change of Asn50 to Asp results in the loss of binding of NITR11 to 1G8. In NITR10, a corresponding gain of function mutation results from a change of Asp50 to Asn and confers binding activity approaching that of wildtype NITR11. Blue and green lines indicate NITR-hFcγ staining profiles; the identities of the specific NITR domains used as binding probes are indicated above each profile. A red line in all graphs indicates the control staining profile (secondary antibody only). All profiles indicate R-phycoerythrin fluorescence. Flow cytometric data were analyzed using FlowJo software version 8.5 (Tree Star Inc., Ashland, OR).

Crystal Structures Reveal that NITRs are Structurally Similar to Antigen Receptor V Set Ig Domains and Dimerize in an Analogous Manner

Four NITR proteins: NITR10, NITR11, as well as NITR10 (D50N) and NITR11 (N50D), which reciprocally convert alloreactive binding specificities, were expressed, refolded and purified. The four native proteins and their corresponding Se-Met derivatives were crystallized and their structures were solved by molecular replacement or SAD and refined against x-ray data from native crystals (Supplementary Tables 1 and 2). Solved structures were compared to each other as well as to structures in the protein data bank (PDB). NITR10, NITR11 and their respective single point mutant forms exhibit V-set chain topology (front sheet, strands A’GFCC’C”, back sheet, strands ABED), as predicted by the presence of canonical V-type framework residues in the NITRs sequences (Strong et al., 1999). TCR Vα, PDB code 2ICW (Wang et al., 2007), is the most similar solved structure to NITR11, with root mean square deviation (r.m.s.d.) of 1.9 Å for Cα atoms (Fig. 3a). An amino acid sequence alignment is provided (Supplementary Figure 4) to allow comparison of strand topology between NITR10/11 and TCR Vα.

Figure 3.

Figure 3

Open in a new tab

NITR11 adopts the same dimerization mode as antigen receptors. (a) TCR Vα•Vβ, PDB code 2ICW, is shown in which the secondary structure of Vα is depicted as ribbon for β-strands in cyan, Vβ in blue. The molecular surface of Vβ is shown in partial transparency and colored light blue for carbon, dark blue for nitrogen, red for oxygen. (b) The NITR11 homodimer is shown with secondary structure depicted in gold. One subunit of NITR11 is shown with a partial transparency and colored gold for carbon, blue for nitrogen, red for oxygen. Phe98 in Vα, Phe103 in Vβ and Phe103 in NITR11 are shown as spheres in which carbon is depicted as green, blue for nitrogen, red for oxygen. Panels (c) and (d) show that NITRs and antigen receptors dimerize similarly due to conservation of key residues in front sheet edge strands involved in TCR Vα•Vβ and Ig VH•VL packing. Side chains of key residues in the edge strands that fold over the central strands of their β sheets and form the cores of dimer interfaces are shown: for Vα (c), Pro44, Tyr96 and Phe98; for NITR11 (d), Pro44, Tyr101 and Phe103 (cyan for carbon, blue for nitrogen, red for oxygen).

The highly curled NITR front sheets interlock to form dimers in both solution and crystal states, (Fig. 3b, Supplementary Table 2). The dimeric arrangement of NITR molecules was not imposed by crystallographic symmetry, as there were single and multiple copies of dimers within the asymmetric units of the solved NITR crystal structures (Supplementary Table 2). The homodimeric arrangement of the NITR11 V domains (Fig. 3b), similar to the unique front-sheet-to-front-sheet dimerization mode observed in the corresponding position of antigen receptors (Fig. 3a), positions the CDR3-analogous loops from both subunits in close proximity at the subunit interface. As in antigen binding receptors, the CDR-analogous loops of NITRs present a relatively flat surface with dimensions in the ranges described for Fab and TCR contact sites with antigen or peptide/MHC complexes (33-266 Å2 buried at interfaces) (MacCallum et al., 1996; Garcia et al., 1999). Interactions between elements in the G strands on the front sheets form portions of the V domain interfaces in both antigen binding receptors (Fig. 3c) and NITRs (Fig. 3d). Sequences in this region of NITRs (Yoder et al., 2004) are homologous to elements encoded by J gene segments in rearranging antigen binding receptors. Phe103 residues from both NITR11 subunits interact by van der Waals contact and form a central portion of the dimer interface (Fig. 3a and b, Supplementary Figure 5), corresponding to the FGXG sequence motif present in the J gene segments of antigen receptors. The FGXG motif is exhibited in the G strand of the front sheets of antigen receptors as a β-bulge, thus breaking the extended β-strand conformation in the central portion of antigen receptor G strands (Chothia et al., 1985) (Fig. 4). The relative position of Phe103 at the dimer interface of NITR11 is nearly identical to the corresponding J-encoded Phe in the central portion of Vα/Vβ and VH/VL interfaces (Chothia et al., 1985) (Fig. 3d), thus conserving this feature of the antigen combining site through an alternative structural form in at least these two NITR molecules.

Figure 4.

Figure 4

Open in a new tab

Structural comparison of the G strands in antigen receptors, shown in cyan, with NITR11, shown in gold. In NITR11, the G strand is in an extended β-strand geometry, whereas antigen receptors (Vα, Vβ, VH, VL) exhibit β bulges at the position corresponding to FGX/G.

To determine if the dimerization mode of NITR extracellular fragments is impacted by the intrachain disulfide bond linking each subunit of NITR homodimers, a crystal structure was solved to a resolution limit of 2.8 Å. for an NITR11 covalent homodimer. This crystal structure includes the extracellular stalk region (C terminal to the V domain and N terminal to the transmembrane region) and is stabilized through an interdomain disulfide bond. The dimerization mode is indistinguishable from that of the non-disulfide linked NITR11 crystal structure and bears overall structural similarity to the paired V domains of Ig and TCR (Fig. 5) (Chothia et al., 1985).

Figure 5.

Figure 5

Open in a new tab

The crystal structure of the disulfide-linked form of NITR11 exhibits the front-sheet-to-front-sheet dimerization mode observed in antigen receptors. One subunit of NITR11 is shown in magenta and the other in cyan. Atoms in cysteine residues that participate in intra- and interchain disulfide bonds are shown as spheres (carbon, magenta or cyan; sulfur, yellow; nitrogen, blue; oxygen, red).

NITR10 and NITR11 exhibit the most significant structural differences from Ig and TCRs in the conformations of the front sheet G strands (Figs. 3c-d and 4). The G strands in NITRs are in the extended β-strand conformation, whereas those of antigen receptors exhibit a conserved β-bulge in the central portion of the G strands. It is not known if this structural difference has functional significance.

A Single-Residue Change has Drastic Effects on NITR Functional Specificity

Analysis of the NITR crystal structures reveals the context of amino acid position 50, which is critical for allogeneic recognition (Fig. 6). The conservative substitution in NITR11 (N50D), which ablates allogeneic recognition, results in a significant shift in the electrostatic potential of the solvent accessible portion of the CDR1 analogous loop, creating an additional area of negative charge (Fig. 6b), resembling that observed in wildtype NITR10 (Fig. 6c). Conversely, the substitution in NITR10 (D50N), which converts the inactive wildtype domain into a functioning and specific receptor (a single carboxy→amide substitution, Oδ2 in D50, Nδ2 in N50), alters the electrostatic potential in this area into a form (Fig. 6d) that resembles wildtype NITR11 more closely (Fig. 6a). These data indicate that the residue at position 50 of NITR10 and NITR11 directly affects the negative electrostatic potential of the CDR1 analogous loop in the V domain and that this altered charge topography may have a direct effect on allogeneic recognition by the NITR.

Figure 6.

Figure 6

Open in a new tab

Electrostatic potential of the CDR-analogous regions of (a) NITR11, (b) NITR11 (N50D), (c) NITR10 and (d) NITR10 (D50N) reveals a structural basis for ligand binding. Molecular surfaces are displayed over one subunit from each NITR dimer and colored by electrostatic potential (red: acidic, white: neutral, blue: basic). For clarity, the molecular surface from the corresponding NITR subunits is shown without the molecular surface. Secondary structure is depicted by gold for strands, red for α-helices, gray for loop regions.

Overall, the tertiary and quaternary structures of NITR10, NITR10 (D50N), NITR11 and NITR11 (N50D), which all dimerize in the front-sheet-to-front-sheet mode utilized by antigen receptors (Chothia et al., 1985), are extremely similar, with r.m.s.d. values of less than 1.2 Å for Cα atoms. The functionally relevant alterations in the NITR mutant proteins have a localized structural effect on the solvent accessible electrostatic potential rather than distal effects that impinge on dimerization or other quaternary effects.

DISCUSSION

We previously described the NITRs as a large multigene family of diversified V domain-containing receptors in several species of bony fish (Strong et al., 1999; Yoder et al., 2001; Yoder et al., 2004; Hawke et al., 2001). NITRs have been characterized best in zebrafish, in which representatives of 12 different NITR V families map to chromosome 7 and members of two additional families are located on chromosome 14 (Yoder et al., 2008). NITRs of the pufferfish S. nephelus exhibit a similar level of organizational complexity (Strong et al., 1999), as do the NITRs of another species of pufferfish and also the Japanese medaka (Oryzias latipes). Based on these and other studies, NITRs appear to be confined to several lineages of bony fish, exhibit a particularly high level of diversity and possess signaling features shared with several other activating/inhibitory types of IgSF transmembrane receptors found in different leukocyte lineages, including NK cells. Despite these general similarities to molecules encoded by the mammalian Leukocyte Receptor Complex (LRC), the amino acid sequences of the Ig domains of NITRs more closely resemble those seen in the rearranging antigen binding receptors, although no evidence has been provided to suggest that NITR genes undergo lineage-restricted somatic variation. The basic question underlying these investigations is: what structural and functional relationships are shared between NITRs and the V region-containing antigen binding receptors?

NITR10/11, two activating-type NITRs that encode a single positive charged residue in the transmembrane and lack inhibitory motifs in their cytoplasmic tails (Yoder et al., 2004; Wei et al., 2007), were the primary focus of these investigations. Bony fish activating receptors have been shown to redirect cytolysis when expressed in transfected human NK cells and also to partner with zebrafish orthologs of Dap12. When cross-linked, the activating receptor-Dap12 complex activates the phosphytidylinositol 3-kinase→AKT→extracellular signal-regulated kinase pathway, suggesting that a DAP12-based activating pathway is common between bony fish and mammals (Wei et al., 2007). The transmembrane region of NITR10/11 associates with DAP12 in a cross-linking dependent manner, also consistent with physiological responses, (Wei, Yoder and Cannon, unpublished).

Definitive understanding of primary leukocyte-specific functions of activating/inhibitory receptors can be daunting, especially outside of mammalian systems where reagent development and genetic characterization range from being far less comprehensive to essentially nonexistent. Cell lines with defined phenotypes are critical to investigations of allorecognition. Four assumptions or observations are relevant to the design of these investigations: 1) NK functions most likely emerged prior to the radiations of mammals, 2) 3H9, a bony fish NK-like cell line that distinguishes allogeneic targets, was shown to express multiple NITRs, 3) the predicted V domains in NITRs suggest that like the conventional antigen binding receptors, NITRs may directly bind a specific determinant (s) and 4) NITRs generally resemble NK receptors of the Ig-type, which are associated with considerable haplotypic and allelic polymorphism (reviewed in (Parham, 2005)).

In addition to these considerations, the genomes of two species of bony fish, medaka and zebrafish, have been sufficiently resolved and adequately annotated to make possible the consideration of other gene families in the context of what we understand to be the general features of multigene families encoding the polymorphic NK receptors seen in higher vertebrates. Specifically, no compelling evidence exists for sequence orthologs of either KIRs or class V C-type lectin-related receptors (Zelensky and Gready, 2004; Panagos et al., 2008), members of which constitute MHC I receptors on human and mouse NK cells, respectively. The relationship of a specific gene family in catfish (leukocyte immune-type receptors, LITRs) (Stafford et al., 2006a) to LRC sequences is remote. The modular domain immune-type receptors (MDIRs), encoded by an additional large multigene family in bony fish, possess complex ectodomain structures that differ from both NITRs and LRC molecules (Cannon et al., 2006). In the broader context of activating/inhibitory transmembrane receptors, a number of Fc receptors have been identified in bony fish and in vitro functions have been established, making it unlikely that NITRs effect these types of interactions (Stafford et al., 2006b; Shen et al., 2003). Nonsaturating pilot surveys of NITR10/11-related sequences in genomic DNA from eight individual fish have identified numerous variants of these genes (including a sequence very similar to the NITR10D50N mutant described in this report), consistent with interaction of NITR10/11 with a polymorphic set of ligands. It seems reasonable to propose that NITRs could constitute at least one type of NK cell receptor, although mammalian-equivalent NK functional criteria are currently absent in bony fish.

Recognition by NK and NK-like cells is mediated through different modes of binding such as by: 1) C-type lectin domains (e.g., NKG2, mouse Ly49), 2) C2-type Ig domains (e.g., KIR, NKp46), 3) somatically rearranged, semi-invariant TCR (e.g. Vα14-Jα18 and Vα24-Jα18 in mouse and human NKT cells, respectively) (Bendelac et al., 2007) and 4) non-or minimally polymorphic V-type Ig domains (e.g., NKp30, NKp44). Binding specificity for the allogeneic determinant in NITR11 resides in the V region and is dependent on a single, solvent-exposed residue in the V region CDR1-analogous loop. Identification of the ligand(s) recognized by NITR11 thus far has proven refractory to a number of conventional approaches that have identified ligands for mammalian NK-type receptors, including the use of Fc pulldown assays on 1G8 cells lysed by a number of different detergents (including Triton X-100, Nonidet P-40, CHAPS, digitonin, Brij 35, Brij 97, n-octyl-β-D-glucopyranoside and MEM-PER) to solubilize membrane complexes. However, mild acid treatment of the target cells (Poláková et al., 2003), which has been shown to disrupt MHC I-β2m interactions in mammalian cell lines, did not influence the quantitative binding of NITR11 to the cell surface, distinguishing the NITR interaction from this aspect of MHC I receptor binding.

The crystal structures of the NITR Ig domains solved in this study classify as V-type and demonstrate a mode of dimerization that is structurally equivalent to antigen binding receptors. However, all of the NITR structures bear an important distinction as compared to those of antigen receptor V domains. Antigen receptors are derived through somatic rearrangement and contain joining (J) gene segments that are highly polymorphic yet show strict conservation of a sequence motif (translated as FGXGTXLXV) in the 3′ portions. This conserved sequence motif is exhibited in the G strands of antigen receptors and similar sequences are found in NITR genes. This site (the G strand) plays a key role in the ability of antigen receptors to form functional dimers by forming binding sites comprised of both subunits (e.g. TCR Vα•Vβ). In antigen receptors, the conserved Phe98 in G strands (Fig. 4) of opposing subunits participate in conserved interactions that form part of a “three layer” packing interface in which side chains from residues in edges of the front sheet (A’FGCC’C”) fold over the central strands of the front sheet to form contacts (Fig. 3c). The conformation of the G strand, which plays a critical direct role in the central portion of antigen receptors dimer interfaces, is distinct in NITRs (Fig. 4). However, despite these structural differences, NITRs bear an unexpectedly close structural resemblance to antigen receptors in their dimerization modes. The unique nature of the V domain interfaces, which thus far has been seen exclusively in molecules involved in immune recognition, suggests that NITR proteins, TCR Vα•Vβ and Ig VH•VL form a common structural family of which all form dimers that function in binding non-self ligands through diversified loop elements in both of their constituent subunits. Two members of this family exhibit the effects of extensive somatic modification and one (NITRs) does not.

On a broad scale, the studies described in this report establish alternative utilization of the V-set Ig domain that has factored prominently in immune recognition throughout the extended chordate lineages (Hernandez Prada et al., 2006; Pende et al., 1999; Cantoni et al., 2003). If one is to accept the hypothesis that NITRs mediate NK-type recognition, the allogeneic binding demonstrated here by NITRs, which combine a level of V region germline diversity equivalent to that seen in the rearranging antigen binding receptors with the signaling mechanisms of NK (and other immune) receptors, suggest that at one point in the development of vertebrate immune competence, NK-type recognition may well have been even more complex than what we presently understand it to be in human and mouse, perhaps exhibiting a range of recognition potential that could extend to include diverse self and nonself determinants.

EXPERIMENTAL PROCEDURES

Cell Culture

43-1 T cell hybridoma and 293T human embryonic kidney cell lines were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mM glutamine, 1 mM sodium pyruvate, 100 IU/ml penicillin and 100 μg/ml streptomycin. 1G8 and 3B11 catfish B cell lines were maintained in catfish cell culture medium as described previously (Miller et al., 1994).

Construction of Lentiviral Vectors for Expression of Catfish NITR-CD3ζ Chimeric Signaling Proteins

The pLF plasmid (Yoder et al., 2004) encodes a Kozak consensus methionine start codon (Kozak, 1996) upstream of the secretion signal peptide (SP) of zebrafish NITR3, as well as a FLAG epitope tag immediately downstream from the predicted NITR3 SP cleavage site. cDNA fragments encoding the uncharged transmembrane region of Raja eglanteria MDIR2 (GenBank DQ278451) (Cannon et al., 2006) and the cytoplasmic signaling domain of mouse CD3ζ (GenBank NM_031162) were inserted downstream of the FLAG epitope sequence in pLF. An oligonucleotide linker containing asymmetric Sfi I sites was inserted between the FLAG epitope sequence and the MDIR2 transmembrane region (Cannon et al., 2006), allowing insertion of compatible Sfi I fragments of catfish NITR cDNAs. A Spe I/Xba I fragment containing the entire SP-FLAG-Sfi I-MDIR2-CD3ζ cassette was inserted into a variant of pEF.GFP (Zufferey et al., 1998) that had been modified by replacement of the GFP cDNA with a short polylinker sequence and mutation of an Sfi I site elsewhere in the pEF.GFP plasmid to form the pEF.CD3ζ lentiviral vector plasmid.

Individual catfish NITR cDNAs were amplified from previously isolated plasmid cDNA clones (Hawke et al., 2001) using Sfi I-flanked oligonucleotides (Supplementary Table 3). The translation product of each cDNA amplicon extends from predicted N-terminus of the mature polypeptide (Emanuelsson et al., 2007) to the end of the extracellular portion of the NITR (predicted at http://www.cbs.dtu.dk/services/TMHMM/). PCR products were cleaved with Sfi I and subcloned into Sfi I-cleaved pEF.CD3ζ to produce each lentiviral vector. Lentiviral plasmid clones were verified by sequencing before packaging and transduction (see below).

GFP Assay for NITR Interactions with Catfish B Cell Lines

pEF.CD3ζ clones containing NITR cDNA fragments were packaged in 293T cells by cotransfection with the pMD.G and pCMVΔR8.91 plasmids as described previously (Zufferey et al., 1998; Zufferey et al., 1997). Supernatants from transfected cells were harvested, centrifuged at 1500 rpm for 5 minutes and used at full strength to transduce 43-1 cells. After 48 hours, FLAG-positive 43-1 cells were isolated by incubation with the anti-FLAG M2 mouse monoclonal antibody (Sigma) followed by anti-mouse IgG magnetic cell sorting (MACS, Miltenyi Biotec). FLAG-positive populations were expanded to produce sufficient numbers for GFP assays.

To perform NITR interaction assays, MACS-sorted, FLAG-positive 43-1 cells expressing various NITR ectodomain-CD3ζ fusions were mixed with either 1G8 or 3B11 catfish B cells at a ratio of 1:1 and incubated for 20 hours at 37°C in 43-1 culture medium. After incubation, cells were fixed in 1% paraformaldehyde in Dulbecco’s phosphate buffered saline (PBS) containing 2 μg/ml Hoechst 33258. GFP expression by 43-1 cells was assessed by fluorescence microscopy.

Construction and Expression of NITR-hFcγ Chimeric Proteins

The Hind III/Xba I fragment from pCP03 (Arase et al., 2002), which encodes the Fc region of human IgG1 (hFcγ) downstream of the start codon and SP of mouse CD150, was subcloned into pcDNA3 (Invitrogen). The original Xho I cloning site in the pCP03 fragment, between the CD150 SP and hFcγ cDNA sequences, was subsequently modified by the insertion of asymmetric Sfi I sites, as well as sequences encoding two epitope tags (FLAG and Avitag) downstream of the Sfi I sites and upstream of the hFcγ fragment, to generate pcDNA3-hsIgG1Fc. Sfi I-flanked fragments of the catfish NITR10 and NITR11 cDNAs, which had been subcloned in the pEF.CD3ζ lentiviral vector described above, were shuttled directly into pcDNA3-hsIgG1Fc at its Sfi I sites. The resulting modifications allow expression of secreted, homodimeric fusion proteins containing paired N-terminal NITR ectodomains and a C-terminal hFcγ fragment, with FLAG and Avitag epitope tags present at the junction between these two moieties for use in other investigations. All NITR10 and NITR11 mutant constructs used for 1G8 cell staining (see Supplementary Methods) also were subcloned in this vector.

Plasmids encoding recombinant NITR10 (wt)-, NITR11 (wt)-, NITR10 (D50N)- and NITR11 (N50D)-hFcγ chimeric proteins were isolated by miniprep (Qiagen) and expressed in 293T cells after DNA transfection using Lipofectamine 2000™ (Invitrogen) in 293T cell culture medium without added antibiotics. Cell culture supernatants were harvested every 3 days (for a total of up to 9 days) and subjected to affinity isolation using protein A sepharose beads (Amersham). After elution of the protein from the gel matrix, the concentration of each hFcγ chimera was estimated by ELISA using a standard curve of human IgG1 (Sigma), after correction for mass differences relative to human IgG1 molecules.

Cell Surface Staining using NITR-hFcγ Chimeric Proteins

Protein A purified NITR-hFcγ chimeric proteins were assayed for binding at a concentration of ∼5 μg/ml in 0.9X PBS + 1% (w/v) IgG-free bovine serum albumin (PBS/BSA). Cell surface staining was performed by incubation of 5 × 106 1G8 or 3B11 cells in 200 μl of NITR-hFcγ solution for 30 min on ice. Cells were washed once in 1 ml of PBS/BSA and stained with 100 μl of a 1:100 dilution of R-phycoerythrin-conjugated goat anti-human-IgGFc (Jackson ImmunoResearch) in PBS/BSA for 30 min on ice. Cells were then washed twice with 1 ml PBS/BSA and fixed in 1% paraformaldehyde in PBS prior to fluorescence microscopy or analysis by an LSR II Flow Cytometer (BD Biosciences).

Generation of Soluble NITR Proteins for Crystallization

cDNA fragments encoding the Ig ectodomains of NITR10, NITR11 and their respective mutant forms were amplified from plasmids and subcloned into the pETBlue-1 vector for IPTG-inducible expression in E. coli. Fragments encoding the V regions of all four NITRs were amplified using PCR primers NITR1011xc-S and NITR1011xc-A; the fragment encoding the V region + stalk region of NITR11 (wt) was amplified using NITR1011xc-S and NITR1011stalkXC-A (Supplementary Table 3).

Coding sequences of the NITR V region proteins begin at the first amino acid of the predicted N-terminus of the mature NITR protein and extend through residues corresponding to the predicted G strand/J-like sequence (residues 22-131 of GenPept entries AAL35551 [NITR10] and AAL35552 [NITR11]). The coding sequence of the V domain and membrane proximal stalk region of NITR11 begins at the same position as the V-only constructs and extends through the final predicted extracellular residue (residue 157 of GenPept AAL35552) rather than ending at the J-like sequence. All constructs initiate with an artificial methionine start codon and terminate with tandem stop codons. After sequence verification, each construct was used to transform E. coli TUNER cells for induction and expression as inclusion bodies, which were purified from induced cultures and refolded as described (Hernandez Prada et al., 2006).

Crystallization

NITR proteins were crystallized using the hanging drop method of vapor diffusion(McPherson, 1999) at 18° C. Both NITR10 native and Se-Met crystals were grown in 1.5M sodium citrate at pH 8.0. NITR10 (D50N) native crystals were grown in 24% PEG 4000, 0.16M magnesium chloride, 20% glycerol and 0.08M Tris HCl at pH 8.5. NITR10 (D50N) Se-Met crystals were grown in 30% PEG 8000, 0.2M sodium acetate trihydrate and 0.1M sodium cacodylate trihydrate at pH 6.5. NITR11 native crystals were grown in 30% PEG 4000, 0.2M sodium acetate and 0.1M Tris HCl at pH 8.5. NITR11 Se-Met crystals were grown in 25.5% PEG 4000, 0.17M ammonium acetate, 15% glycerol and 0.085M sodium citrate at pH 5.6. NITR11 (N50D) native crystals were grown in 18% PEG 8000, 0.2M sodium acetate and 0.1M sodium cacodylate at pH 6.5. NITR11 (N50D) Se-Met crystals were grown in 22% PEG 2000 and 0.1M Tris HCl at pH 8.6. The NITR11 stalk disulfide linked dimer crystals were grown in 22% PEG 4000, 0.2M lithium sulfate monohydrate and 0.1M Tris HCl at pH 8.5. Crystals of NITR11 and NITR11 (N50D) formed after 3 days at 18° C, crystals of NITR10 and NITR10 (D50N) required 3 weeks of growth at 18° C and crystals of NITR11 stalk formed after 7 days.

Data Collection

Data collection for all crystals was performed at the Brookhaven National Laboratory NSLS beamline X6A. Beamline X6A uses an ADSC Quantum 210 CCD detector with an Oxford Danfysik toroidal focusing mirror. The absorption edge for all Se-Met crystals was determined using x-ray fluorescence scanning, from which the peak, remote and inflection wavelengths could be calculated. An NITR10 native crystal diffracted to 1.56 Å at a wavelength of 0.9791 Å. A single NITR10 Se-Met crystal diffracted to 1.65 Å and produced datasets collected at a peak wavelength of 0.9791 Å and a remote wavelength of 0.9393 Å. A single data set was collected to 2.44 Å on a native crystal for NITR10 (D50N) at a wavelength of 0.9791 Å. Two NITR10 (D50N) Se-Met crystals diffracted to 1.9 Å and produced datasets collected at a peak wavelength of 0.9792 Å and a remote wavelength of 9537 Å. An NITR11 native crystal diffracted to 1.98 Å at a wavelength of 0.9791 Å. Datasets were collected on an NITR11 Se-Met crystal at a peak wavelength of 0.9789 Å and a remote wavelength of 0.9322 Å. X-ray data were collected on a single NITR11 (N50D) native crystal that diffracted to 1.90 Å at a wavelength of 0.9322 Å. An NITR11 N50D Se-Met crystal diffracted to 1.65 Å at a wavelength of 0.9322 Å. A crystal of an NITR11 protein including the stalk region and interchain disulfide bond diffracted to 2.8 Å and native data was collected at 0.9537 Å.

Structure Determination

All datasets were indexed and scaled using the software package HKL2000 (Otwinowski and Minor, 1997). The XPREP (Sheldrick, 1994) program was used to verify crystal space group identification and produce reflection files containing merged and unmerged anomalous signals for the Se-Met datasets. The SHELX (Uson and Sheldrick, 1999) package was used to phase the resulting reflection files using SAD. Finally, ARP/wARP (Lamzin and Wilson, 1993; Morris et al., 2003; Perrakis et al., 1999) was used to trace the resulting phases and produce an initial structure model. All Se-Met models were refined using the CNS (Brunger et al., 1998) package until the Rfree was below 30.0%. Refinement of the structures for NITR10, NITR11, NITR10 (D50N), NITR11 (N50D) and the NITR11 stalk dimer was performed using x-ray diffraction data obtained from native protein crystals with CNS. The PDB codes are: NITR10: 2QHL; NITR10 (D50N): 2QJD, 3B5T; NITR11: 2QQQ; NITR11 (N50D): 2QTE; NITR11 stalk dimer: 3BDB.

Structural Analysis

Cα alignments, r.m.s.d. calculations and structural similarity searches were performed with EMBL Secondary Structure Matching (Krissinel and Henrick, 2004). Dimer interfaces were analyzed using both EMBL Protein Interfaces, Surfaces and Assemblies (Krissinel and Henrick, 2007) and the Protein-Protein Interaction Server (Jones and Thornton, 1996). Protein surfaces and secondary structures were visualized and ray-traced using PyMOL (DeLano, 2006). Superpositions were generated and electron density maps were visualized using Coot (Emsley and Cowtan, 2004).

Supplementary Material

01

ACKNOWLEDGEMENTS

We thank L. Lanier for helpful discussion regarding the GFP assay, providing the pCP03 plasmid and suggesting the acid dissociation analyses. We thank T. Saito for provision of the 43-1 T cell hybridoma line. We thank B. Pryor for editorial assistance and M. O’Driscoll for technical assistance. Flow cytometry was performed by M. Morrow at the Immunodiagnostics Laboratory at the University of South Florida/All Children’s Hospital Children’s Research Institute. This work was supported by the National Institutes of Health (R01 AI23338 and R01 AI57559 to GWL; R01 DE013883 and R21 HL080222 to DAO); DAO is additionally supported by the Cure Autism Now Foundation 2908051-12.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Reference List

  1. The CCP4 suite: programs for protein crystallography. Acta Crystallogr. D. Biol. Crystallogr. 1994;50:760–763. doi: 10.1107/S0907444994003112. [DOI] [PubMed] [Google Scholar]
  2. Arase H, Mocarski ES, Campbell AE, Hill AB, Lanier LL. Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science. 2002;296:1323–1326. doi: 10.1126/science.1070884. [DOI] [PubMed] [Google Scholar]
  3. Bendelac A, Savage PB, Teyton L. The biology of NKT cells. Ann. Rev. Immunol. 2007;25:297–336. doi: 10.1146/annurev.immunol.25.022106.141711. [DOI] [PubMed] [Google Scholar]
  4. Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, Read RJ, Rice LM, Simonson T, Warren GL. Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D. Biol. Crystallogr. 1998;54:905–921. doi: 10.1107/s0907444998003254. [DOI] [PubMed] [Google Scholar]
  5. Cannon JP, Haire RN, Mueller MG, Litman RT, Eason DD, Tinnemore D, Amemiya CT, Ota T, Litman GW. Ancient divergence of a complex family of immune-type receptor genes. Immunogenet. 2006;58:362–373. doi: 10.1007/s00251-006-0112-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cantoni C, Ponassi M, Biassoni R, Conte R, Spallarossa A, Moretta A, Moretta L, Bolognesi M, Bordo D. The three-dimensional structure of the human NK cell receptor NKp44, a triggering partner in natural cytotoxicity. Structure (Camb.) 2003;11:725–734. doi: 10.1016/s0969-2126(03)00095-9. [DOI] [PubMed] [Google Scholar]
  7. Chothia C, Novotny J, Bruccoleri R, Karplus M. Domain association in immunoglobulin molecules. The packing of variable domains. J. Mol. Biol. 1985;186:651–663. doi: 10.1016/0022-2836(85)90137-8. [DOI] [PubMed] [Google Scholar]
  8. DeLano WL. The PyMOL molecular graphics system. DeLano Scientific LLC; San Carlos, CA: 2006. [Google Scholar]
  9. Emanuelsson O, Brunak S, von Heijne G, Nielsen H. Locating proteins in the cell using TargetP, SignalP and related tools. Nat. Protoc. 2007;2:953–971. doi: 10.1038/nprot.2007.131. [DOI] [PubMed] [Google Scholar]
  10. Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D. Biol. Crystallogr. 2004;60:2126–2132. doi: 10.1107/S0907444904019158. [DOI] [PubMed] [Google Scholar]
  11. Evenhuis J, Bengten E, Snell C, Quiniou SM, Miller NW, Wilson M. Characterization of additional novel immune type receptors in channel catfish, Ictalurus punctatus. Immunogenet. 2007;59:661–671. doi: 10.1007/s00251-007-0230-x. [DOI] [PubMed] [Google Scholar]
  12. Garcia KC, Teyton L, Wilson IA. Structural basis of T cell recognition. Ann. Rev. Immunol. 1999;17:369–397. doi: 10.1146/annurev.immunol.17.1.369. [DOI] [PubMed] [Google Scholar]
  13. Gonen-Gross T, Achdout H, Arnon TI, Gazit R, Stern N, Horejsí V, Goldman-Wohl D, Yagel S, Mandelboim O. The CD85J/leukocyte inhibitory receptor-1 distinguishes between conformed and β2-microglobin-free HLA-G molecules. J. Immunol. 2005;174:4866–4874. doi: 10.4049/jimmunol.175.8.4866. [DOI] [PubMed] [Google Scholar]
  14. Hawke NA, Yoder JA, Haire RN, Mueller MG, Litman RT, Miracle AL, Stuge T, Shen L, Miller N, Litman GW. Extraordinary variation in a diversified family of immune-type receptor genes. Proc. Natl. Acad. Sci. USA. 2001;98:13832–13837. doi: 10.1073/pnas.231418598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Prada J.A. Hernandez, Haire RN, Allaire M, Jakoncic J, Stojanovic N, Cannon JP, Litman GW, Ostrov DA. Ancient evolutionary origin of diversified variable regions revealed by crystal structures of an amphioxus VCBP. Nat. Immunol. 2006;7:875–882. doi: 10.1038/ni1359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jones S, Thornton JM. Principles of protein-protein interactions derived from structural studies. Proc. Natl. Acad. Sci. USA. 1996;93:13–20. doi: 10.1073/pnas.93.1.13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Khakoo SI, Geller R, Shin S, Jenkins JA, Parham P. The D0 domain of KIR3D acts as a major histocompatibility complex class I binding enhancer. J. Exp. Med. 2002;196:911–921. doi: 10.1084/jem.20020304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kozak M. Interpreting cDNA sequences: some insights from studies on translation. Mamm. Genome. 1996;7:563–574. doi: 10.1007/s003359900171. [DOI] [PubMed] [Google Scholar]
  19. Krissinel E, Henrick K. Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions. Acta Crystallogr. D. Biol. Crystallogr. 2004;60:2256–2268. doi: 10.1107/S0907444904026460. [DOI] [PubMed] [Google Scholar]
  20. Krissinel E, Henrick K. Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 2007;372:774–797. doi: 10.1016/j.jmb.2007.05.022. [DOI] [PubMed] [Google Scholar]
  21. Lamzin VS, Wilson KS. Automated refinement of protein models. Acta Crystallogr. D. Biol. Crystallogr. 1993;49:129–147. doi: 10.1107/S0907444992008886. [DOI] [PubMed] [Google Scholar]
  22. MacCallum RM, Martin AC, Thornton JM. Antibody-antigen interactions: contact analysis and binding site topography. J. Mol. Biol. 1996;262:732–745. doi: 10.1006/jmbi.1996.0548. [DOI] [PubMed] [Google Scholar]
  23. McPherson A. Crystallization of Biological Macromolecules. Cold Spring Harbor: 1999. [Google Scholar]
  24. Miller NW, Rycyzyn MA, Wilson MR, Warr GW, Naftel JP, Clem LW. Development and characterization of channel catfish long term B cell lines. J. Immunol. 1994;152:2180–2189. [PubMed] [Google Scholar]
  25. Morris RJ, Perrakis A, Lamzin VS. ARP/wARP and automatic interpretation of protein electron density maps. Methods Enzymol. 2003;374:229–244. doi: 10.1016/S0076-6879(03)74011-7. [DOI] [PubMed] [Google Scholar]
  26. Ohtsubo K, Marth JD. Glycosylation in cellular mechanisms of health and disease. Cell. 2006;126:855–867. doi: 10.1016/j.cell.2006.08.019. [DOI] [PubMed] [Google Scholar]
  27. Ohtsuka M, Arase H, Takeuchi A, Yamasaki S, Shiina R, Suenaga T, Sakurai D, Yokosuka T, Arase N, Iwashima M, Kitamura T, Moriya H, Saito T. NFAM1, an immunoreceptor tyrosine-based activation motif-bearing molecule that regulates B cell development and signaling. Proc. Natl. Acad. Sci. USA. 2004;101:8126–8131. doi: 10.1073/pnas.0401119101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997;276:307–326. doi: 10.1016/S0076-6879(97)76066-X. [DOI] [PubMed] [Google Scholar]
  29. Panagos PG, Dobrinski KP, Chen X, Grant AW, Traver D, Djeu JY, Wei S, Yoder JA. Immune-related, lectin-like receptors are differentially expressed in the myeloid and lymphoid lineages of zebrafish. Immunogenet. 2008;58:31–40. doi: 10.1007/s00251-005-0064-3. [DOI] [PubMed] [Google Scholar]
  30. Parham P. MHC class I molecules and KIRs in human history, health and survival. Nat. Rev. Immunol. 2005;5:201–214. doi: 10.1038/nri1570. [DOI] [PubMed] [Google Scholar]
  31. Pende D, Parolini S, Pessino A, Sivori S, Augugliaro R, Morelli L, Marcenaro E, Accame L, Malaspina A, Biassoni R, Bottino C, Moretta L, Moretta A. Identification and molecular characterization of NKp30, a novel triggering receptor involved in natural cytotoxicity mediated by human natural killer cells. J. Exp. Med. 1999;190:1505–1516. doi: 10.1084/jem.190.10.1505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Perrakis A, Morris R, Lamzin VS. Automated protein model building combined with iterative structure refinement. Nat. Struct. Biol. 1999;6:458–463. doi: 10.1038/8263. [DOI] [PubMed] [Google Scholar]
  33. Poláková K, Bennink JR, Yewdell JW, Bystrická M, Bandzuchová E, Russ G. Mild acid treatment induces cross-reactivity at 4H84 monoclonal antibody specific to nonclassical HLA-G antigen with classical HLA class I molecules. Hum. Immunol. 2003;64:256–264. doi: 10.1016/s0198-8859(02)00777-2. [DOI] [PubMed] [Google Scholar]
  34. Sheldrick GM. Siemens Analytical X-ray Instruments. Madison, WI: 1994. XPREP SHELXTL. [Google Scholar]
  35. Shen L, Stuge TB, Bengten E, Wilson M, Chinchar VG, Naftel JP, Bernanke JM, Clem LW, Miller NW. Identification and characterization of clonal NK-like cells from channel catfish (Ictalurus punctatus) Dev. Comp. Immunol. 2004;28:139–152. doi: 10.1016/s0145-305x(03)00119-8. [DOI] [PubMed] [Google Scholar]
  36. Shen L, Stuge TB, Evenhuis JP, Bengten E, Wilson M, Chinchar VG, Clem LW, Miller NW. Channel catfish NK-like cells are armed with IgM via a putative FcmicroR. Dev. Comp. Immunol. 2003;27:699–714. doi: 10.1016/s0145-305x(03)00042-9. [DOI] [PubMed] [Google Scholar]
  37. Smith HRC, Heusel JW, Mehta IK, Kim S, Dorner BG, Naidenko OV, Iizuka K, Furukawa H, Beckman DL, Pingel JT, Scalzo AA, Fremont DH, Yokoyama WM. Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc. Natl. Acad. Sci. USA. 2002;99:8826–8831. doi: 10.1073/pnas.092258599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Stafford JL, Bengten E, Du Pasquier L, McIntosh RD, Quiniou SM, Clem LW, Miller NW, Wilson M. A novel family of diversified immunoregulatory receptors in teleosts is homologous to both mammalian Fc receptors and molecules encoded within the leukocyte receptor complex. Immunogenet. 2006a;58:758–773. doi: 10.1007/s00251-006-0134-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Stafford JL, Wilson M, Nayak D, Quiniou SM, Clem LW, Miller NW, Bengten E. Identification and characterization of a FcR homolog in an ectothermic vertebrate, the channel catfish (Ictalurus punctatus) J. Immunol. 2006b;177:2505–2517. doi: 10.4049/jimmunol.177.4.2505. [DOI] [PubMed] [Google Scholar]
  40. Strong SJ, Mueller MG, Litman RT, Hawke NA, Haire RN, Miracle AL, Rast JP, Amemiya CT, Litman GW. A novel multigene family encodes diversified variable regions. Proc. Natl. Acad. Sci. USA. 1999;96:15080–15085. doi: 10.1073/pnas.96.26.15080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Uson I, Sheldrick GM. Advances in direct methods for protein crystallography. Curr. Opin. Struct. Biol. 1999;9:643–648. doi: 10.1016/s0959-440x(99)00020-2. [DOI] [PubMed] [Google Scholar]
  42. Wang JW, Chen JR, Gu YX, Zheng CD, Jiang F, Fan HF, Terwilliger TC, Hao Q. SAD phasing by combination of direct methods with the SOLVE/RESOLVE procedure. Acta Crystallogr. D. Biol. Crystallogr. 2004;60:1244–1253. doi: 10.1107/S0907444904010674. [DOI] [PubMed] [Google Scholar]
  43. Wang L, Zhao Y, Guo Y, Jones LL, Kranz DM, Mourad W, Li H. Crystal structure of a complete ternary complex of TCR, superantigen and peptide-MHC. Nat. Struct. Mol. Biol. 2007;14:169–171. doi: 10.1038/nsmb1193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wei S, Zhou J, Chen X, Shah RN, Liu J, Orcutt TM, Traver D, Djeu JY, Litman GW, Yoder JA. The zebrafish activating immune receptor Nitr9 signals via Dap12. Immunogenet. 2007;59:813–821. doi: 10.1007/s00251-007-0250-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Yoder JA, Cannon JP, Litman RT, Murphy C, Freeman JL, Litman GW.Evidence for a transposition event in a second NITR gene cluster in zebrafish Immunogenet in press.2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Yoder JA, Litman RT, Mueller MG, Desai S, Dobrinski KP, Montgomery JS, Buzzeo MP, Amemiya CT, Trede NS, Wei S, Djeu JY, Humphray S, Jekosch K, Prada J.A. Hernandez, Ostrov DA, Litman GW. Resolution of the NITR gene cluster in zebrafish. Proc. Natl. Acad. Sci. USA. 2004;101:15706–15711. doi: 10.1073/pnas.0405242101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yoder JA, Mueller MG, Wei S, Corliss BC, Prather DM, Willis T, Litman RT, Djeu JY, Litman GW. Immune-type receptor genes in zebrafish share genetic and functional properties with genes encoded by the mammalian lymphocyte receptor cluster. Proc. Natl. Acad. Sci. USA. 2001;98:6771–6776. doi: 10.1073/pnas.121101598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Zelensky AN, Gready JE. C-type lectin-like domains in Fugu rubripes. Genomics. 2004;5:51. doi: 10.1186/1471-2164-5-51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zufferey R, Dull T, Mandel RJ, Bukovsky a., Quiroz D, Naldini L, Trono D. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J. Virol. 1998;72:9873–9880. doi: 10.1128/jvi.72.12.9873-9880.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Zufferey R, Nagy D, Mandel RJ, Naldini L, Trono D. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat. Biotech. 1997;15:871–875. doi: 10.1038/nbt0997-871. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS66165-supplement-01.pdf (482.1KB, pdf)

RESOURCES