International Immunology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (22) Request Permissions Google Scholar Articles by Kamada, N. Articles by Taniguchi, M. Search for Related Content PubMed PubMed Citation Articles by Kamada, N. Articles by Taniguchi, M. Social Bookmarking          What's this?

International Immunology, Vol. 13, No. 7, 853-861, July 2001
© 2001 Japanese Society for Immunology

Crucial amino acid residues of mouse CD1d for glycolipid ligand presentation to V14 NKT cells

Noriaki Kamada^1,2, Hiroshi Iijima³, Kaname Kimura³, Michishige Harada¹, Eiko Shimizu¹, Shin-ichiro Motohashi¹, Tetsu Kawano¹, Hiroshi Shinkai², Toshinori Nakayama¹, Teruyuki Sakai³, Laurent Brossay⁴, Mitchell Kronenberg⁴ and Masaru Taniguchi^{1 1} CREST (Core Research for Evolutional Science and Technology) and Department of Molecular Immunology, Graduate School of Medicine, and
² Department of Dermatology, School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan
³ Pharmaceutical Research Laboratory, Kirin Brewery Co., Ltd, 3 Miyahara, Gunma 370-12, Japan
⁴ La Jolla Institute for Allergy and Immunology, 10355 Science Center Drive, San Diego, CA 92121, USA

Correspondence to: M. Taniguchi

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
A novel lymphocyte, NKT cells bearing an invariant V14 antigen receptor, specifically recognizes -galactosylceramide (-GalCer) exclusively presented by mouse CD1d (mCD1d). However, the precise molecular interaction remains unclear. For the basis of functional analyses, a docking model of -GalCer with the crystal structure of mCD1d was constructed. Possible residues involved in the -GalCer–mCD1d interaction were found to be Arg79, Glu83 and Asp80 for carbohydrate recognition, and Asp153 for interaction with the amide group on the fatty acyl chain. The -GalCer-presenting ability of various transfectants expressing mutant mCD1d was completely abrogated if a single amino acid mutation was induced at positions 79, 80, 83 or 153, suggesting that the polar amino acids above the F' pocket are crucial for -GalCer presentation to activate V14 NKT cells. The possibility that Glu83 is a contact site for the NKT cell receptor is also discussed.

Keywords: -galactosylceramide, NKT cell receptor, docking modeling

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
V14 NKT cells are known to belong to a lymphoid cell lineage distinct from conventional ß T lymphocytes (1–3), and are characterized by their expression of the invariant antigen receptor encoded by V14 and J281 gene segments associated mainly with V_ß8.2 in mouse (4–6), and by V24 and JQ segments with V_ß11 in human (7,8). Both mouse and human NKT cell antigen receptors interact with -galactosylceramide (-GalCer) selectively presented by a non-classical MHC-like molecule, CD1d, that is monomorphic in nature among species (9–16).

Crystallographic analysis has indicated that the mouse CD1d (mCD1d) has a deep, narrow and very hydrophobic ligand-binding groove with two pockets denoted A' and F' (17). In fact, the interaction of glycolipids with mCD1d has been demonstrated by biochemical analysis (18,19). It has also been suggested that the two long alkyl chains of the ligand are able to be accommodated in the two hydrophobic pockets (A' and F'), while the hydrophilic head group of the antigen is exposed for recognition by the NKT cell antigen receptor (20).

The analysis of the structure–activity relationship of glycosylceramides strongly suggest that the antigenicity is highly susceptible to the molecular structure of the glycosylceramides (11,21–24). Especially important are the 1''- linkage between the sugar and ceramide moieties, the existence of the 2''-OH in the sugar pyranose ring, and the existence of the 3-OH on the sphingosine. Possible binding conformations of -GalCer analogues with mCD1d are calculated in which active analogues could share important pharmacophoric groups (25) and the seven possible conformations are suggested by the calculation. However, the active conformation could not be determined from the information obtained from the structure–activity relationship study alone.

In the present study, the molecular basis of the binding and recognition of glycolipids with mCD1d is addressed by computer modeling and by functional assay. The docking modeling and functional study using a series of mutant mCD1d-transfected cells suggest that the polar amino acid residues located over the F' pocket, including Arg79, Asp80 and Glu83 in the ₁-helix, and Asp153 in the ₂-helix of mCD1d, and a backbone carbonyl group (Val149) on the ₂-helix are essential for the efficient activation of V14 NKT cells.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Mice
V14 NKT (RAG-KO/V14tg/V_ß8.2tg) mice were established and back-crossed 4 times with C57BL/6 mice (11). All mice used were maintained under specific pathogen-free conditions in our animal facility. Animal care was in accordance with the guidelines of Chiba University.

Glycolipids
The -GalCer (KRN7000) was prepared in the Pharmaceutical Research Laboratories, Kirin Brewery (Gunma, Japan). Antigen stock solution (100 µg/ml) was stored in DMSO at 4°C.

Computer modeling
Software.
The modeling study was aided by the SYBYL software package (Tripos Associates, St Louis, MO). SYBYL Programming Language was utilized for programming specific procedures developed in this study.

Step 1: ligand binding sites of mCD1d.
In our previous study we have shown that the long alkyl chains of -GalCer (AGL517) and -glucosylcceramide (-GlcCer) (AGL563) (Fig. 1A) would bind to the inside the A' and F' pockets via hydrophobic interaction (11). Although most of the amino acid residues near the binding groove are hydrophobic, His68, Ser76, Arg79, Asp80, Glu83, Lys148, Asp153, Thr156, Thr159, Asp166 and Thr167 are polar residues that might interact by hydrogen bonding with the amide or hydroxyl groups of -GalCer. Most of these residues are located near the center of the binding helices except His68, Lys148 and Asp166. Thus we chose the side chains of Ser76, Arg79, Asp80, Glu83, Asp153, Thr156, Thr159 and Thr167 as candidate hydrogen bond donor/acceptor sites, several of which interact with hydrogen acceptor/donor atoms of ligand molecules (Fig. 1B). Distances between pairs of heteroatoms in the side chains of the above-mentioned amino acid residues were calculated. The accuracy factor (or grid size) of the distance was set to 0.8 Å, i.e. if a distance was in the range of 0.0–0.8 Å, it was considered to be 0.4 Å, and a value 1.2 Å was assigned for a distance between 0.8 and 1.6 Å.

View larger version (26K): [in this window] [in a new window]   Fig. 1. Structure of glycolipids and mCD1d. (A) Structures of -glycosylceramides used in the present study. Rax'' and Req'', axial and equatorial configuration of hydroxyl (OH) groups on the carbohydrates; R, OH on the sphingosine; m, number of methylene groups on the sphingosine; n, number of methylene groups on the fatty acyl chain. (B) Hydrophilic residues on the -helices of mCD1d. The protein (mCD1d) structure is indicated as a ribbon (-helix, orange; ß-sheet, green). The side chains of the polar amino acid residues are represented in space-filling mode (oxygen atoms, red; nitrogen atoms, violet; carbon atoms, white). The two hydrophobic pockets of mCD1d are indicated as A' and F'.

 
Step 2: hydrogen-bonding site mapping of ligand molecules.
The conformation of the ligand (AGL517 or AGL563) was fixed to one of the putative active conformations obtained in the previous structure–activity relationship study (25). Dummy atoms were added to all the heteroatoms in the ligand molecules that might interact with mCD1d. The dummy atoms were placed at a fixed distance (2.8 Å from heteroatoms) and in the direction of the hydrogen-bonding partner atom (Fig. 2A). These dummy atoms correspond to the hydrogen-bonding pair sites of the receptor. At this step, short alkyl chain (m = 0, n = 0; Fig. 1 A) models were used. The torsion angles about the bonds specified, as indicated in Fig. 2(B), were varied in a systematic manner with an increment of 30° and the resulting conformers were examined for unfavorable van der Waal's atomic contacts. Conformers without significant van der Waal's contacts were selected and the distances between the dummy atoms were recorded (distance map). Here an accuracy factor of 0.8 Å was again employed.

View larger version (29K): [in this window] [in a new window]   Fig. 2. Docking simulation. (A) Geometry of dummy atoms. The dummy atoms correspond to possible hydrogen-bond pair atoms on mCD1d. Top, amide bond; middle, a hydroxyl group; bottom, an ether oxygen. (B) Generation of distance matrices for a ligand molecule. By fixing the torsion angles to one of the seven conformations obtained in the previous structure–activity relationship study, only torsion angles that change the positions of the dummy atoms (numbered in the figure) were scanned to map positions of the possible hydrogen-bond pair. The dummy atoms each have an atomic radius of 1 Å because they represent heteroatoms of mCD1d. However, the dummy–dummy contacts were ignored because one heteroatom on the receptor could interact with multiple heteroatoms of the ligand. Distances between dummy atoms were recorded. This figure displays only distances between four dummy atoms for simplicity (dashed lines). (C) Placing a ligand model in the binding pocket. Note that dummy atoms originating from 2-OH and the carbonyl oxygen (1') overlap and correspond to the -oxygen of Asp80.

 
Step 3: docking study.
Docking ligands to mCD1d was guided by the potential of the hydrogen-bonding interaction. Inter-atomic distances between the specified dummy atoms were extracted from the distance map calculated in Step 2. The six distances determine the relative orientation of the dummy atoms (Fig. 2B). The distance that describes the quadruplet was then subjected to a search for the distances calculated for the receptor molecule in Step 1. As the same grid resolution (0.8 Å) was employed in Step 1 and Step 2, the comparison in this step was straightforward. For every distance combination matched for the ligand and the receptor, the ligand molecule was placed in the receptor groove according to the orientation (Fig. 2C). Figure 2(C) summarizes the docking and alkyl chain conformation search procedure. The models were sorted by interaction energy and the lowest 20 energy models were selected for each putative active conformer.

Step 4: superimposition of AGL517 and AGL563 in the frame of the mCD1d structure.
In the previous step, 20 docking models were calculated. Using mCD1d as the frame of reference, the docked models of AGL517 and AGL563 were compared for pharmacophore overlap. The goodness of overlap was measured by r.m.s. fit value for the 2''-, 4''- and 3-OH groups. Extending the alkyl chains of AGL517 produced the docking model of -GalCer (KRN7000) (m = 24, n=13). Finally, structure optimization was applied exclusively to the extended carbon chains.

Mutagenesis of mCD1d and their transfectants
The mCD1d cDNA was established by L. Brossay and M. Kronenberg (26), and was in a pHßAprmCD1.1neo plasmid. For mutagenesis, the insert mCD1d cDNA was ligated with a cloning vector pGEM-11zf using unique SalI and BamHI sites (P2371; Promega Madison, WI). Mutant mCD1d cDNAs were obtained with a Gene Editor in vitro site-directed mutagenesis system (Q9280; Promega) using mutagenic oligonucleotides as follows: R74A: 5'-GTTTCAAGTCTATGCAGTCAGCTTTACCAGGG-3'; V75A, 5'-CAAGTCTATCGAGCCAGCTTTACCAGGGAC-3';S76A, 5'-GTCTATCGAGTCGCCTTTACCAGGGACATAC-3'; F77A, 5'-CTATCGAGTCAGCGCTACCAGGGACATACAGG-3'; T78A, 5'-CGAGTCAGCTTTGCCAGGGACATACAGGAATT-3'; R79A, 5'-CGAGTCAGCTTTACCGCGGACATACAGGAATTAGTC-3'; D80A, 5'-CTTTACCAGGGCCATACAGGAATTAGTCAA-3'; I81A, 5'-GCTTTACCAGGGACGCACAGGAATTAGTCAAAATG-3'; Q82A, 5'-CCAGGGACATAGCGGAATTAGTCAAAATGATG-3'; E83A, 5'-GGACATACAGGCATTAGTCAAAATGATGTCAC-3'; L84A, 5'-GACATACAGGAAGCAGTCAAAATGATGTCACC-3'; V85A, 5'-CATACAGGAATTAGCCAAAATGATGTCACCTAA-3'; K86A, 5'-CAGGAATTAGTCGCAATGATGTCACCTAAAGAAG-3';K148A,5'-GTTAGACTTGCCCATCGCAGTGCTCAACGCTGATC-3';N151A, 5'-CATCAAAGTGCTCGCCGCTGATCAAGGGACA-AG-3'; D153A,5'-GCTCAACGCTGCTCAAGGGACAAGTGCAAC-3'; Q154A, 5'-GCTCAACGCTGATGCAGGGACAAGTGCAACCG-3'; G155A, 5'-CAACGCTGATCAAGCGACAAGTGCAACCGTGC-3'; D166A, 5'-GCTCCTGAATGCCACCTGCCCCCTAT-3'.

The mutant mCD1d cDNAs were inserted into the expression vector pHßAprneo, after the alanine substitution was confirmed by DNA sequencing. Transfection was performed using RMA-S cells lacking TAP-1 as described previously (26). The expression of wild-type and mutant mCD1d was detected by staining with anti-mCD1d antibodies (1B1 and 1H1; PharMingen, San Diego, CA).

V14 NKT cell lines and clones
V14 NKT cell lines were generated as described previously (27) with a slight modification. In brief, spleen cells (1x10⁵/well) from V14 NKT mice were cultured in anti-CD3 mAb (145-2C11, 100 µg/ml)-coated 96-well round-bottomed plates (#3077; Falcon, Franklin Lakes, NJ) in the presence of 100 U/ml of recombinant mouse IL-4 for 3 days and then expanded in six-well plates (#3516; Costar, Corning, NY) with 100 U/ml of recombinant mouse IL-4 for 2 days. The cells were then transferred and cultured with 30 U/ml of recombinant mouse IL-4 for another 2 days in new plates. A CD1d-restricted phosphatidylinositol-reactive V14 NKT cell clone, 24.8A1 (28), was kindly provided by Drs M. Brenner and S. Behar (Harvard Medical School, Boston).

Measurement of cytokine production
V14 NKT cell lines or the 24.8A1 V14 NKT cell clone were co-cultured with wild-type or mutant mCD1d expressing RMA-S cells or non-transfected parental RMA-S cells, which had been preincubated with -GalCer (100 ng/ml) for 12 h, in RPMI 1640 supplemented with 10% FCS, kanamycin (100 mg/ml), 2 mM L-glutamine and 50 mM 2-mercaptoethanol in 96-well round-bottomed culture plates under humidified 5% CO₂ at 37°C. The amounts of IFN- or IL-2 in the culture supernatants were assessed using a standard sandwich ELISA according to the manufacturer's protocol (PharMingen, San Diego, CA).

	Results

Top Abstract Introduction Methods Results Discussion References

 
Computer-aided molecular modeling analysis of the interaction between mCD1d and -GalCer
We have previously shown that the hydrophobic interaction between the long carbon chains of the ceramide portion of -GalCer and the two hydrophobic A'/F' pockets of CD1d are essential for stimulation of V14 NKT cells (11). Moreover, the hydrogen-bonding interactions between -GalCer and CD1d are also suggested to be important for stimulation of V14 NKT cells. In fact, the 1''--anomeric sugar linkage, the configuration of the 2''-OH group of the sugar moiety and the 3-OH residue on the sphingosine base of the ceramide are critical for the activation of V14 NKT cells (11,21). It has also been suggested that -galactosylsphingosine lacking all carbon chains of the fatty acyl chains but retaining the amide structure and intact sphingosine base on the ceramide has a significant stimulatory effect on the allo-mixed lymphocyte reaction (29) or V14 NKT cell hybridoma (21,30), indicating the importance of the sphingosine and amide structure on the alkyl chain of the ceramide for the activation of V14 NKT cells.

To identify a possible hydrogen-bonding interaction between mCD1d and -GalCer, computer-aided molecular modeling was carried out. The modeling study utilized the results from the structure–activity relationship study carried out previously (25). In the present study, we calculated the putative active conformations for -GalCer (AGL517) and -GlcCer (AGL563) (Fig. 1 A) by assuming that these active analogues bind to a receptor molecule (mCD1d) with a common geometric arrangement of the 2''-OH on the sugar, the 3-OH on the sphingosine and the 4''-OH on the sugar, which represents the orientation of the sugar ring. It was found that there are seven possible ways for the active analogues to arrange their hydroxyl groups in space (25).

The docking models were then searched using the following assumptions:

1. Several hydrogen bonds are formed between -monoglycosylceramides and mCD1d.
   
2. Considering that the crystal structure of mCD1d contains an unidentified ligand molecule (17), the conformation of each side chain in the crystal structure available from the Protein Data Bank (1CD1) will accept the active -monoglycosylceramides, -GalCer (AGL517) and -GlcCer (AGL563), for docking.
   
3. The active -monoglycosylceramides bind to mCD1d in one of the seven putative active conformations.
   
4. In the docking models of mCD1d/AGL517 and mCD1d/AGL563, the important hydroxyl groups of AGL517 and AGL563 must be superimposed using mCD1d as the frame of reference.
   
5. -Monoglycosylceramides bind to mCD1d with their alkyl chains inside the A' and F' pockets.
   

Under the above assumptions, we first selected several hydrophilic amino acid residues that locate near the binding groove of mCD1d (Fig. 1B). Some of the heteroatoms in the side chains of these hydrophilic residues would interact with -monoglycosylceramides by forming hydrogen bonds. These atoms were considered to be possible hydrogen-bonding sites of the receptor (Assumptions 1 and 2). On the other hand, dummy atoms were added to the heteroatoms of -monoglycosylceramides. These dummy atoms represented hydrogen-bonding partner sites on the mCD1d (Fig. 2A and B; Assumption 1). By fixing the conformation of the -monoglycosylceramides in one of the seven possibilities (Assumption 3), the bonds indicated in Fig. 2(B) were systematically rotated to map the possible hydrogen-bonding sites. Mapping of the possible hydrogen-bonding sites was expressed as the distances between the dummy atoms.

Similarly a distance map that describes the hydrogen-bonding sites on the receptor was prepared. Then the maps from the ligand were screened with the map from the receptor. In this study, we screened the ligand maps that found more than six matches of distances in the map from the receptor (Assumptions 1–3). According to the match in the distances, a model of -monoglycosylceramide could be positioned in the mCD1d (Fig. 2C).

A large number of docking models of mCD1d/AGL517 and mCD1d/AGL563 were obtained. The models were screened by the following criteria: (i) the absence of unfavorable van der Waal's atomic contacts and (ii) the energy of interaction. For each of the seven conformations, 20 energetically favored models were selected. Then, Assumption 4 was applied to score the goodness of the docking models. Using mCD1d as the frame of reference, the docked models of AGL517 and AGL563 were compared for goodness of overlap of the 2''-OH and 4''-OH groups on the sugar moiety, and the 3-OH group on the sphingosine base expressed as r.m.s. fit values.

Figure 3(A) represents the schematic topology of the docking model. In this model, the fatty acyl chain stretched toward the A' pocket and the sphingosine base toward the F' pocket. The model (r.m.s. fit score 1.3A) was ranked fifth best. The top four models (r.m.s. fit scores 1.0, 1.1, 1.2 and 1.2A) were omitted from consideration, even though the important hydroxyl groups overlapped well for AGL517 and AGL563 because one or both of the alkyl chains of the ligand was placed outside of the binding groove (Assumption 5).

View larger version (50K): [in this window] [in a new window]   Fig. 3. Docking model of -GalCer with the crystal structure of mCD1d. (A) Views from the side (upper panel) and top (lower panel) of the binding cleft of mCD1d are shown. The protein (mCD1d) structure is indicated as a ribbon and the ligand (-GalCer) as a space-filling model. The ribbon diagram of mCD1d is shown with -helices in orange and ß-sheets in green. Also the -GalCer is shown with hydrogen atoms in light blue, oxygen atoms in red, nitrogen atoms in violet and carbon atoms in white. (B) Hydrogen-bond network near the sugar–ceramide linkage. The side chains of Arg79, Asp80, Glu83, Asp153 and -GalCer are shown with oxygen atoms in red and nitrogen atoms in violet. The -helices and carbon atoms of mCD1d and carbon atoms of -GalCer are colored in yellow and white respectively. A possible hydrogen-bond network is indicated (dotted lines). (C) Molecular surface of the mCD1d–-GalCer complex. The 1- and 2-helices are in magenta. The side chains of Arg79, Glu83 and -GalCer are colored (carbon and hydrogen atoms of the amino acid residues of mCD1d in green, hydrogen atoms of -GalCer in light blue, nitrogen atoms of mCD1d and -GalCer in violet, and oxygen atoms of mCD1d and -GalCer in red).

 
The docking model shown in Fig. 3(B) suggests that the side chains of Arg79 and Asp80 of mCD1d interact with the 2''-OH on the sugar moiety through a complex hydrogen bond network. Asp80 is common in both mCD1d and human CD1d (hCD1d), but is replaced by Gly (hCD1a) or Glu (hCD1b and hCD1c) in other CD1 molecules that have no -GalCer presenting activity to V14 NKT cells (15), indicating Asp80 is critical for -GalCer binding to CD1d. Exchanging Asp with Glu is generally considered to be a conservative mutation, however, it is obvious that an elongation of one methylene unit in the side chain will have a significant effect on the topology of the hydrogen-bond network used for molecular recognition. It is also suggested that the 3''-OH group on the sugar moiety interacts with Arg79 and Glu83 on the ₁-helix of mCD1d. The sugar moiety generates a hump-shaped molecular surface with Arg79 and Glu83 that might be important for the T cell activation (Fig. 3C).

The model also suggests that the amide group on the fatty acyl chain of -GalCer interacts with Asp153. Asp153 is conserved in both hCD1d and mCD1d, but is replaced by Asn (hCD1a) or Tyr (hCD1b and hCD1c) in other species, again indicating the importance of Asp153 for the -GalCer interaction. The 3-OH group on the sphingosine base, which is essential for V14 NKT cell activation (21), resides between the groove helices, suggesting that this hydroxyl group interacts with a backbone carbonyl group (Val149) on the ₂-helix of mCD1d (Fig. 3B).

The final model of -GalCer (KRN7000)/mCD1d was subjected to a structure optimization calculation using Merck molecular force field (31–35) and atomic charge parameters employing a dielectric constant of 4. For the important hydrogen-bonding pair atoms, the distance constraint energy term, E_{dist_c} = k(d – d₀)², was added where d is the distance between the two atoms, d₀ = 2.8 Å, and k, the force constant, was set to 4 (kcal/mol/Ų). Structural optimization was carried out until a minimum energy change of 0.05 kcal/mol was attained. No significant change in the overall structure was observed after optimization, supporting the docking model as being energetically acceptable.

Functional activity of mutant mCD1d.
According to the suggestions obtained from the molecular modeling, we established a panel of transfectants expressing mutant mCD1d molecules with alanine substitutions. RMA-S cells were used as a host cell line, since they express marginal levels of mCD1d (Fig. 4A) and undetectable levels of MHC class I (data not shown). The mutations covered amino acid residues Arg79, Asp80, Glu83 (₁-helix) and Asp153 (₂-helix) of mCD1d, all of which were suggested by molecular modeling to be important for the interaction with -GalCer. The amino acid residues around the four residues (₁-helix: 75, 76, 77, 78, 81, 82, 84 and 85, and ₂-helix: 151, 154 and 155), and nearby charged amino acid residues (₁-helix: 74 and 86, and ₂-helix: 148 and 166) were replaced by alanine. The mutation sites are indicated in Fig. 4(A).

View larger version (74K): [in this window] [in a new window]   Fig. 4. Failure to stimulate V14 NKT cells with -GalCer presented by mutant mCD1d transfectants. (A) Expression patterns of mCD1d in RMA-S cell transfectants. Transfectants expressing wild-type or mutated mCD1d and non-transfected RMA-S cells were stained with anti-mCD1d antibody (1B1) and their fluorescence profiles are illustrated. Capital letters and numbers indicate amino acid residues with alanine substitutions (A) and their position in mCD1d. (B) IFN- production from V14 NKT cell lines. We used RMA-S cells expressing wild-type or mutated mCD1d as APC and they were pre-pulsed with 100 ng/ml -GalCer for 12 h. V14 NKT cell lines (2x105/well) were stimulated with APC (5x104/well) for 72 h in 100 µl culture. IFN- produced in the culture supernatants were measured by ELISA. The data were expressed as mean in triplicates ± SD. Three other experiments gave similar results.

 
The expression levels of wild-type and mutated mCD1d on the RMA-S transfectants were determined by flow cytometry analysis. As shown in Fig. 4(A), transfectants with notably high and similar levels of fluorescence intensity were selected by anti-mCD1d 1B1 mAb. The expression patterns of CD1d were also confirmed by another anti-CD1d 1H1 mAb (data not shown). The results suggested that these mutants do not have significant conformational changes for antigen presentation. V14 NKT cell lines were stimulated with transfectants preincubated with -GalCer and IFN- production was assessed. As shown in Fig. 4(B), most of the mutant mCD1d transfectants stimulated V14 NKT cells to the same extent as the wild-type mCD1d transfectant. However, no stimulatory activity was detected if a mutation was induced at Arg79, Asp80 or Glu83 in the ₁-helix, or at Asp153 in the ₂-helix of mCD1d. Control groups, such as vehicle-pulsed transfectants or no antigen-presenting cells (APC) with 100 ng/ml of -GalCer treatment, showed no activity (data not shown). Furthermore, the increased numbers of transfectant APC or those of the V14 NKT cell line did not influence the results (data not shown). These results clearly suggest that the four residues on the -helices of mCD1d are crucial for the activation of V14 NKT cells.

In order to characterize the effect of the mutations in the direct interaction with the NKT antigen receptor, the mCD1d mutants were tested for their ability to activate a V14 NKT cell clone 24.8.A1 which is activated by mCD1d in the absence of -GalCer (28). Endogenous phosphatidylinositol is suggested to be an inherent antigen that the 24.8.A1 NKT cell recognizes (28), while it is not formally proved. Since the 24.8A1 NKT cell clone is an autoreactive clone, it is thus likely to assume that the reactivity of the clone reflects on its NKT antigen receptor binding site on the CD1d molecule if endogenous and exogenous ligands behave in a similar manner. Among the four critical residues, Asp80 and Asp153 locate inside the binding groove of mCD1d, indicating the difficulty in the direct interaction of the NKT cell receptor with these residues (Fig. 3B). In contrast, the side chains of Arg79 and Glu83 are exposed outside the -helix of mCD1d, suggesting possible contact sites with the NKT cell receptor. We, thus, chose R79A and E83A mutants for analysis. As shown in Fig. 5, the R79A mutant successfully activated the 24.8.A1 NKT cell clone, while the E83A did not, indicating that at least Arg79 does not interact with the NKT cell receptor directly, but possibly with the 2''-OH of the carbohydrate moiety as suggested by the data in Fig. 3(B). The control mutation of Gln154 that also locates outside the helix, however, does not influence the ability to stimulate the clone.

View larger version (19K): [in this window] [in a new window]   Fig. 5. Recognition of mutant mCD1d by autoreactive V14 NKT cell clone. The autoreactive V14 NKT cell clone, 24.8.A1 (1x105/well), was cultured with RMA-S cells (1x105/well) transfected with wild-type or mutated mCD1d for 16 h in 150 µl culture. IL-2 in the culture supernatants was measured by ELISA. The data were expressed as mean in triplicates ± SD. Two other experiments gave similar results.

 

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Based on the functional analysis using mutant mCD1d transfectants and the docking model of -GalCer with the crystal structure of mCD1d, four amino acid residues in mCD1d, Arg79, Asp80 and Glu83 (₁-helix), and Asp153 (₂-helix), were suggested to be crucial for the activation of V14 NKT cells. Arg79 and Asp80 may interact with the 2''-OH, while Arg79 and Glu83 may interact with the 3''-OH on the carbohydrate moiety of -GalCer as suggested by the docking model. However, functional analyses using CD1d mutants raised the possibility of the direct interaction of some residues with the NKT cell receptor. In fact, Asp80 and Asp153 locate inside the binding groove, while Arg79 and Glu83 locate outside. This suggests that Arg79 and Glu83 are able to interact with the NKT cell receptor, whereas Asp80 and Asp153 are not.

The docking model suggests that Asp80 interacts with the 2''-OH on the carbohydrate moiety, whereas the amide moiety of the acyl chain interacts with Asp153 and also possibly with Asp80. Thus, the Asp80 and Asp153 residues would be covered by the ligand and not be accessible from the molecular surface of the mCD1d–-GalCer complex, suggesting that these amino acids are responsible for -GalCer interaction rather than for direct contact with the NKT cell receptor. This notion is also supported by the fact that Asp80 is common to both mCD1d and hCD1d, but is replaced by Gly (hCD1a) or Glu (hCD1b and hCD1c) in other CD1 molecules having no -GalCer-presenting activity to V14 NKT cells (15).

Based on the topological point of view, Arg79 and Glu83 are able to interact with the NKT cell receptor. However, Arg79 is essential for the -GalCer interaction. This is because the R79A mutant mCD1d transfectant successfully stimulated the 24.8.A1 NKT cell clone (Fig. 5) which is stimulated by the recognition of endogenous phosphatidylinositol in the absence of -GalCer (28). Distinct from Arg79, the E83A mutation, however, failed to stimulate the 24.8.A1 NKT cell clone (Fig. 5). A simple interpretation is that Glu83 is the contact site for the NKT cell receptor rather than for -GalCer binding. However, it is possible that Glu83 is important for the binding of endogenous phosphatidylinositol to mCD1d. Preferably, Glu83 is likely to interact with the 3''-OH of the carbohydrate moiety of -GalCer as suggested by the docking model. In fact, an -GalCer analogue that carries an additional glycosylation at the 3'' position of -GalCer, Galß1–3Gal1–1'Cer, has been shown to be a potent antigen for the stimulation of V14 NKT cells (11). Given that the Galß1–3Gal1–1'Cer is presented in a similar manner, the existence of the bulky substitution at the 3''-OH might disturb the interaction of Glu83 with the NKT cell receptor. Taken collectively, the role of Glu83 remains unclear from the present study. We have to wait for the results of the crystallographic analysis to obtain the final conclusion.

In our current model, the 3-OH group on the sphingosine base of -GalCer, which is critical for the activation of V14 NKT cells, may not interact with any of the four residues. Instead, it is likely that it interacts with a backbone carbonyl group (Val149) on the ₂-helix of mCD1d. Under the assumptions made for the docking modeling and the results of the mutation experiments, we could exclude the possibility that the 3-OH of the sphingosine interacts with any of the four charged amino acid residues by the following calculations. First, the 3-OH is hypothesized to form a hydrogen bond with either Asp80 or Asp153. Search calculations are carried out to find suitable docking models that satisfy (i) the interaction of Arg79 and Glu83 with the ligand, and (ii) the occupancy of the hydrophobic pockets of mCD1 with the alkyl chains of the ligand. As a result, no model was found (data not shown), thus eliminating the possibility that the 3-OH interacts with Asp80 or Asp153. If we assume that the 3-OH interacts with either Arg79 or Glu83, which are located outside the binding groove, it is obviously impossible to satisfy Assumption 5. Based on the above estimations, the 3-OH on the sphingosine base is likely to interact with the backbone carbonyl group of Val149. However, the interaction between the 3-OH group of the sphingosine base and the backbone atoms of mCD1d cannot be demonstrated experimentally by mutation studies.

Studies using surface plasmon resonance have demonstrated that the affinities of both the mCD1d and hCD1d for -GalCer are in the 0.1–1 and 0.01–0.1 µM ranges respectively (18,19). Likewise, despite their inability to stimulate V14 NKT cells, mCD1d and hCD1d can bind to non-stimulatory ß-GalCer, N-dipalmitoyl-L--phosphatidylethanolamine or gangliosides (18). Thus, CD1d appears to have an ability to bind a variety of lipid-containing antigens regardless of their stimulatory activities. Residues whose side-chain atoms might interact with the fatty acyl chain of -GalCer include Trp63, Met69 and Phe70. Side-chain atoms of Leu84, Val149 and Leu150 interact with the sphingosine base. These residues are well conserved among both hCD1d and mCD1 groups (36). In contrast, the 4 amino acid residues important for -GalCer binding are conserved only in CD1d, but not in other CD1 families of both human and mouse. The findings are in good agreement with the functional data that -GalCer is presented by both human and mCD1d but not by other hCD1 family members (15,37,38).

The model is compared with a crystal structure of the TCR–MHC–peptide complex (39). The position of the alkyl chain (fatty acid moiety) of -GalCer seems to correspond well with the backbone of the peptide antigen. The sugar moiety of -GalCer resides well within the space between mCD1d and the NKT cell antigen receptor. The position of the sugar moiety is similar to the position of the peptide (Tyr8^p) bound to the B7/HLA-A2 complex, suggesting that the sugar moiety is recognized by CDR3ß or CDR1ß of the V_ß8.2 chain. Recently, it has been reported that a consensus motif of CDR3ß is preferentially used by human NKT cells after selection by -GalCer (12). This seems to support the suggestion that the sugar moiety is recognized in part by the CDR3ß.

The 4''-OH and 6''-OH groups on the sugar moiety make no direct interaction with mCD1d, but might interact with the NKT cell antigen receptor. However it should be noted that our model remains hypothetical regarding the interaction with the NKT cell receptor, considering the diversity found in several MHC–TCR complex structures determined by X-ray crystallography (40,41). Recently, the structural basis of mCD1d for antigen presentation has been reported (42), and it was concluded that some of the acidic and polar amino acids in the CDR3 region of the TCR ß chain could be involved in the formation of hydrogen bonds with the galactose moiety of the antigen. Taken together with the surface plasmon resonance results, the abrogation of the response by these mutations may be due to a decreased stability of the mCD1d–-GalCer complex or to inappropriate binding orientation for V14 NKT receptor recognition, rather than a simple failure in the binding of -GalCer to mCD1d.

	Acknowledgments

 
We thank Professor Garland R. Marshall for reading the manuscript, and Drs M. Brenner and S. Behar (Harvard Medical School, Boston) for the generous gift of V14 NKT cell clone, 24.8A1. We also thank Ms Kazuko Higashino for her excellent technical support and Ms Hiroko Tanabe for preparation of this manuscript. This work was supported by HFSP (RG0168/2000-M).

	Abbreviations

 

APC antigen-presenting cell

GalCer galactosylceramide

GlcCer glucosylceramide

hCD1d human CD1d

mCD1d mouse CD1d

	Notes

 
The first two authors contributed equally to this work

Transmitting editor: M. Miyasaka

Received 14 February 2001, accepted 22 March 2001.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Makino, Y., Kanno, R., Koseki, H. and Taniguchi, M. 1996. Development of V14⁺ NK T cells in the early stages of embryogenesis. Proc. Natl Acad. Sci. USA 93:6516.[Abstract/Free Full Text]
2. Cui, J., Shin, T., Kawano, T., Sato, H., Kondo, E., Toura, I., Kaneko, Y., Koseki, H., Kanno, M. and Taniguchi, M. 1997. Requirement for V14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623.[Abstract/Free Full Text]
3. Sato, H., Nakayama, T., Tanaka, Y., Yamashita, M., Shibata, Y., Kondo, E., Saito, Y. and Taniguchi, M. 1999. Induction of differentiation of pre-NKT cells to mature V14 NKT cells by granulocyte/macrophage colony-stimulating factor. Proc. Natl Acad. Sci. USA 96:7439.[Abstract/Free Full Text]
4. Imai, K., Kanno, M., Kimoto, H., Shigemoto, K., Yamamoto, S. and Taniguchi, M. 1986. Sequence and expression of transcripts of the T-cell antigen receptor -chain gene in a functional, antigen-specific suppressor-T-cell hybridoma. Proc. Natl Acad. Sci. USA 83:8708.[Abstract/Free Full Text]
5. Lantz, O. and Bendelac, A. 1994. An invariant T cell receptor chain is used by a unique subset of major histocompatibility complex class I-specific CD4⁺ and CD4^–8^– T cells in mice and humans. J. Exp. Med. 180:1097.[Abstract/Free Full Text]
6. Makino, Y., Kanno, R., Ito, T., Higashino, K. and Taniguchi, M. 1995. Predominant expression of invariant V14⁺ TCR chain in NK1.1⁺ T cell populations. Int. Immunol. 7:1157.[Abstract/Free Full Text]
7. Dellabona, P., Padovan, E., Casorati, G., Brockhaus, M. and Lanzavecchia, A. 1994. An invariant V24–JQ/V_ß11 T cell receptor is expressed in all individuals by clonally expanded CD4^–8^– T cells. J. Exp. Med. 180:1171.[Abstract/Free Full Text]
8. Porcelli, S., Gerdes, D., Fertig, A. M. and Balk, S. P. 1996. Human T cells expressing an invariant V24–JQ TCR are CD4^– and heterogeneous with respect to TCR ß expression. Hum. Immunol. 48:63.[Web of Science][Medline]
9. Brossay, L., Chioda, M., Burdin, N., Koezuka, Y., Casorati, G., Dellabona, P. and Kronenberg, M. 1998. CD1d-mediated recognition of an -galactosylceramide by natural killer T cells is highly conserved through mammalian evolution. J. Exp. Med. 188:1521.[Abstract/Free Full Text]
10. Burdin, N., Brossay, L., Koezuka, Y., Smiley, S. T., Grusby, M. J., Gui, M., Taniguchi, M., Hayakawa, K. and Kronenberg, M. 1998. Selective ability of mouse CD1 to present glycolipids: -galactosylceramide specifically stimulates V14⁺ NK T lymphocytes. J. Immunol. 161:3271.[Abstract/Free Full Text]
11. Kawano, T., Cui, J., Koezuka, Y., Toura, I., Kaneko, Y., Motoki, K., Ueno, H., Nakagawa, R., Sato, H., Kondo, E., Koseki, H. and Taniguchi, M. 1997. CD1d-restricted and TCR-mediated activation of V14 NKT cells by glycosylceramides. Science 278:1626.[Abstract/Free Full Text]
12. Kawano, T., Tanaka, Y., Shimizu, E., Kaneko, Y., Kamada, N., Sato, H., Osada, H., Sekiya, S., Nakayama, T. and Taniguchi, M. 1999. A novel recognition motif of human NKT antigen receptor for a glycolipid ligand. Int. Immunol. 11:881.[Abstract/Free Full Text]
13. Nicol, A., Nieda, M., Koezuka, Y., Porcelli, S., Suzuki, K., Tadokoro, K., Durrant, S. and Juji, T. 2000. Human invariant V24⁺ natural killer T cells activated by -galactosylceramide (KRN7000) have cytotoxic anti-tumour activity through mechanisms distinct from T cells and natural killer cells. Immunology 99:229.[Web of Science][Medline]
14. Nieda, M., Nicol, A., Koezuka, Y., Kikuchi, A., Takahashi, T., Nakamura, H., Furukawa, H., Yabe, T., Ishikawa, Y., Tadokoro, K. and Juji, T. 1999. Activation of human V24NKT cells by -glycosylceramide in a CD1d-restricted and V24TCR-mediated manner. Hum. Immunol. 60:10.[Web of Science][Medline]
15. Spada, F. M., Koezuka, Y. and Porcelli, S. A. 1998. CD1d-restricted recognition of synthetic glycolipid antigens by human natural killer T cells. J. Exp. Med. 188:1529.[Abstract/Free Full Text]
16. Takahashi, T., Nieda, M., Koezuka, Y., Nicol, A., Porcelli, S. A., Ishikawa, Y., Tadokoro, K., Hirai, H. and Juji, T. 2000. Analysis of human V24⁺ CD4⁺ NKT cells activated by -glycosylceramide-pulsed monocyte-derived dendritic cells. J. Immunol. 164:4458.[Abstract/Free Full Text]
17. Zeng, Z., Castano, A. R., Segelke, B. W., Stura, E. A., Peterson, P. A. and Wilson, I. A. 1997. Crystal structure of mouse CD1: an MHC-like fold with a large hydrophobic binding groove. Science 277:339.[Abstract/Free Full Text]
18. Naidenko, O. V., Maher, J. K., Ernst, W. A., Sakai, T., Modlin, R. L. and Kronenberg, M. 1999. Binding and antigen presentation of ceramide-containing glycolipids by soluble mouse and human CD1d molecules. J. Exp. Med. 190:1069.[Abstract/Free Full Text]
19. Sakai, T., Naidenko, O. V., Iijima, H., Kronenberg, M. and Koezuka, Y. 1999. Syntheses of biotinylated -galactosylceramides and their effects on the immune system and CD1 molecules. J. Med. Chem. 42:1836.[Web of Science][Medline]
20. Grant, E. P., Degano, M., Rosat, J. P., Stenger, S., Modlin, R. L., Wilson, I. A., Porcelli, S. A. and Brenner, M. B. 1999. Molecular recognition of lipid antigens by T cell receptors. J. Exp. Med. 189:195.[Abstract/Free Full Text]
21. Brossay, L., Naidenko, O., Burdin, N., Matsuda, J., Sakai, T. and Kronenberg, M. 1998. Structural requirements for galactosylceramide recognition by CD1-restricted NK T cells. J. Immunol. 161:5124.[Abstract/Free Full Text]
22. Morita, M., Motoki, K., Akimoto, K., Natori, T., Sakai, T., Sawa, E., Yamaji, K., Koezuka, Y., Kobayashi, E. and Fukushima, H. 1995. Structure–activity relationship of -galactosylceramides against B16-bearing mice. J. Med. Chem. 38:2176.[Web of Science][Medline]
23. Uchimura, A., Shimizu, T., Morita, M., Ueno, H., Motoki, K., Fukushima, H., Natori, T. and Koezuka, Y. 1997. Immunostimulatory activities of monoglycosylated -D-pyranosylceramides. Bioorg. Med. Chem. 5:2245.[Medline]
24. Motoki, K., Morita, M., Kobayashi, E., Uchida, T., Akimoto, K., Fukushima, H. and Koezuka, Y. 1995. Immunostimulatory and antitumor activities of monoglycosylceramides having various sugar moieties. Biol. Pharm. Bull. 18:1487.[Web of Science][Medline]
25. Iijima, H., Kimura, K., Sakai, T., Uchimura, A., Shimizu, T., Ueno, H., Natori, T. and Koezuka, Y. 1998. Structure–activity relationship and conformational analysis of monoglycosylceramides on the syngeneic mixed leukocyte reaction. Bioorg. Med. Chem.6:1905.[Medline]
26. Teitell, M., Holcombe, H. R., Brossay, L., Hagenbaugh, A., Jackson, M. J., Pond, L., Balk, S. P., Terhorst, C., Peterson, P. A. and Kronenberg, M. 1997. Nonclassical behavior of the mouse CD1 class I-like molecule. J. Immunol. 158:2143.[Abstract]
27. Kaneko, B. Y., Harada, M., Kawano, T., Yamashita, M., Shibata, Y., Gejyo, F., Nakayama, T. and Taniguchi, M. 2000. Augmentation of V14 NKT cell-mediated cytotoxicity by interleukin 4 in an autocrine mechanism resulting in the development of concanavalin A-induced hepatitis. J. Exp. Med. 191:105.[Abstract/Free Full Text]
28. Gumperz, J. E., Roy, C., Makowska, A., Lum, D., Sugita, M., Podrebarac, T., Koezuka, Y., Porcelli, S. A., Cardell, S., Brenner, M. B. and Behar, S. M. 2000. Murine CD1d-restricted T cell recognition of cellular lipids. Immunity 12:211.[Web of Science][Medline]
29. Sakai, T., Ueno, H., Natori, T., Uchimura, A., Motoki, K. and Koezuka, Y. 1998. Effects of - and ß-galactosylated C2-ceramides on the immune system. J. Med. Chem. 41:650.[Web of Science][Medline]
30. Makowska, A., Kawano, T., Taniguchi, M. and Cardell, S. 2000. Differences in the ligand specificity between CD1d-restricted T cells with limited and diverse T-cell receptor repertoire. Scand. J. Immunol. 52:71.[Web of Science][Medline]
31. Halgren, T. A. 1996. Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94. J. Comput. Chem. 17:490.[Web of Science]
32. Halgren, T. A. 1996. Merck molecular force field. II. MMFF94 van der Waals and electrostatic parameters for intermolecular interactions. J. Comput. Chem. 17:520.[Web of Science]
33. Halgren, T. A. 1996. Merck molecular force field. III. Molecular geometries and vibrational frequencies for MMFF94. J. Comput. Chem. 17:553.[Web of Science]
34. Halgren, T. A. and Nachbar, R. B. 1996. Merck molecular force field. IV. Conformational energies and geometries for MMFF94. J. Comput. Chem. 17:587.[Web of Science]
35. Halgren, T. A. 1996. Merck molecular force field. V. Extension of MMFF94 using experimental data, additional computational data, and empirical rules. J. Comput. Chem. 17:616.[Web of Science]
36. Porcelli, S. A. 1995. The CD1 family: a third lineage of antigen-presenting molecules. Adv. Immunol. 59:1.[Web of Science][Medline]
37. Kawano, T., Nakayama, T., Kamada, N., Kaneko, Y., Harada, M., Ogura, N., Akutsu, Y., Motohashi, S., Iizasa, T., Endo, H., Fujisawa, T., Shinkai, H. and Taniguchi, M. 1999. Antitumor cytotoxicity mediated by ligand-activated human V24 NKT cells. Cancer Res. 59:5102.[Abstract/Free Full Text]
38. Exley, M., Garcia, J., Balk, S. P. and Porcelli, S. 1997. Requirements for CD1d recognition by human invariant V24⁺ CD4^–CD8^– T cells. J. Exp. Med. 186:109.[Abstract/Free Full Text]
39. Ding, Y. H., Smith, K. J., Garboczi, D. N., Utz, U., Biddison, W. E. and Wiley, D. C. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA- A2/Tax peptide complex using different TCR amino acids. Immunity 8:403.[Web of Science][Medline]
40. Garcia, K. C., Degano, M., Pease, L. R., Huang, M., Peterson, P. A., Teyton, L. and Wilson, I. A. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptide–MHC antigen. Science 279:1166.[Abstract/Free Full Text]
41. Reinherz, E. L., Tan, K., Tang, L., Kern, P., Liu, J., Xiong, Y., Hussey, R. E., Smolyar, A., Hare, B., Zhang, R., Joachimiak, A., Chang, H. C., Wagner, G. and Wang, J. 1999. The crystal structure of a T cell receptor in complex with peptide and MHC class II. Science 286:1913.[Abstract/Free Full Text]
42. Burdin, N., Brossay, L., Degano, M., Iijima, H., Gui, M., Wilson, I. A. and Kronenberg, M. 2000. Structural requirements for antigen presentation by mouse CD1. Proc. Natl Acad. Sci. USA 97:10156.[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				M. Taniguchi, T. Tashiro, N. Dashtsoodol, N. Hongo, and H. Watarai The specialized iNKT cell system recognizes glycolipid antigens and bridges the innate and acquired immune systems with potential applications for cancer therapy Int. Immunol., October 27, 2009; (2009) dxp104v2. [Abstract] [Full Text] [PDF]

				M. I. Zimmer, H. P. Nguyen, B. Wang, H. Xu, A. Colmone, K. Felio, H.-J. Choi, P. Zhou, M.-L. Alegre, and C.-R. Wang Polymorphisms in CD1d affect antigen presentation and the activation of CD1d-restricted T cells PNAS, February 10, 2009; 106(6): 1909 - 1914. [Abstract] [Full Text] [PDF]

				K. S. Wun, N. A. Borg, L. Kjer-Nielsen, T. Beddoe, R. Koh, S. K. Richardson, M. Thakur, A. R. Howell, J. P. Scott-Browne, L. Gapin, et al. A minimal binding footprint on CD1d-glycolipid is a basis for selection of the unique human NKT TCR J. Exp. Med., April 14, 2008; 205(4): 939 - 949. [Abstract] [Full Text] [PDF]

				L. Kjer-Nielsen, N. A. Borg, D. G. Pellicci, T. Beddoe, L. Kostenko, C. S. Clements, N. A. Williamson, M. J. Smyth, G. S. Besra, H. H. Reid, et al. A structural basis for selection and cross-species reactivity of the semi-invariant NKT cell receptor in CD1d/glycolipid recognition J. Exp. Med., March 20, 2006; 203(3): 661 - 673. [Abstract] [Full Text] [PDF]

				B. Giabbai, S. Sidobre, M. D. M. Crispin, Y. Sanchez-Ruiz, A. Bachi, M. Kronenberg, I. A. Wilson, and M. Degano Crystal Structure of Mouse CD1d Bound to the Self Ligand Phosphatidylcholine: A Molecular Basis for NKT Cell Activation J. Immunol., July 15, 2005; 175(2): 977 - 984. [Abstract] [Full Text] [PDF]

				J. S. Im, K. O. A. Yu, P. A. Illarionov, K. P. LeClair, J. R. Storey, M. W. Kennedy, G. S. Besra, and S. A. Porcelli Direct Measurement of Antigen Binding Properties of CD1 Proteins Using Fluorescent Lipid Probes J. Biol. Chem., January 2, 2004; 279(1): 299 - 310. [Abstract] [Full Text] [PDF]

				S. Sidobre, O. V. Naidenko, B.-C. Sim, N. R. J. Gascoigne, K. C. Garcia, and M. Kronenberg The V{alpha}14 NKT Cell TCR Exhibits High-Affinity Binding to a Glycolipid/CD1d Complex J. Immunol., August 1, 2002; 169(3): 1340 - 1348. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (22) Request Permissions Google Scholar Articles by Kamada, N. Articles by Taniguchi, M. Search for Related Content PubMed PubMed Citation Articles by Kamada, N. Articles by Taniguchi, M. Social Bookmarking          What's this?

Online ISSN 1460-2377 - Print ISSN 0953-8178

Copyright © 2009 Japanese Society for Immunology

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics