Binding and antigen presentation of ceramide-containing glycolipids by soluble mouse and human CD1d molecules - PubMed (original) (raw)

Binding and antigen presentation of ceramide-containing glycolipids by soluble mouse and human CD1d molecules

O V Naidenko et al. J Exp Med. 1999.

Abstract

We have purified soluble mouse and human CD1d molecules to assess the structural requirements for lipid antigen presentation by CD1. Plate-bound CD1d molecules from either species can present the glycolipid alpha-galactosyl ceramide (alpha-GalCer) to mouse natural killer T cells, formally demonstrating both the in vitro formation of antigenic complexes, and the presentation of alpha-GalCer by these two CD1d molecules. Using surface plasmon resonance, we show that at neutral pH, mouse CD1 and human CD1d bind to immobilized alpha-GalCer, unlike human CD1b, which requires acidic pH for lipid antigen binding. The CD1d molecules can also bind both to the nonantigenic beta-GalCer and to phosphatidylethanolamine, indicating that diverse lipids can bind to CD1d. These studies provide the first quantitative analysis of monomeric lipid antigen-CD1 interactions, and they demonstrate that the orientation of the galactose, or even the nature of the polar head group, are likely to be more important for T cell receptor contact than CD1d binding.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Purification of soluble CD1 proteins. (A) Coomassie blue–stained 12% SDS polyacrylamide gel showing FPLC-purified mCD1 and hCD1d proteins. Molecular weight (MW) standards are in the first lane with sizes (in kD) indicated. (B) ELISA confirming the identity of purified CD1d proteins. 200 ng of protein was coated per well; the blank wells were coated with BSA only. HRP-conjugated goat anti–mouse Ig was used as a secondary reagent to detect NOR3.2, BBM.1, and 7F11, and HRP-conjugated streptavidin was used to detect biotin-1B1. The values represent the average of duplicates of the absorbance values after addition of the HRP substrate.

Figure 2

Figure 2

Immobilized purified CD1d molecules are able to present α-GalCer to NK T cells. (A) 1 μg/well mCD1 or hCD1d proteins preincubated with 100 ng of α-GalCer or β-GalCer and immobilized on a 96-well plate was used to stimulate the T cell hybridoma, 3C3. Some wells had antigen alone or CD1d protein alone. 15 μg/ml anti-mCD1 mAb 1B1 or isotype control antibody (Isotype) was added to some wells immediately before the addition of T cells. This experiment is representative of five independent experiments. (B) Stimulation of the 1.2 and 2C12 NK T cell hybridomas by immobilized mCD1 and hCD1d proteins (1 μg/well), either alone or with 50 ng/well of α-GalCer. This experiment is representative of five experiments.

Figure 3

Figure 3

Biotin-modified α-GalCer is antigenic for NK T cells. (A) Antigen structures. α-GalCer is shown on the top, with the α-linkage of the sugar indicated by an arrow and the acyl chain of the ceramide moiety enclosed within a box. The position 1–4 carbons of the sphingosine are indicated. The biotin-modified acyl chain of biotin–α-GalCer and biotin–β-GalCer is shown below. (B) Biotin–α-GalCer can be recognized by NK T cells. mCD1-transfected or control, mock-transfected A20 cells were pulsed with 100 ng/ml of lipid antigen, or with 0.1% DMSO vehicle control, washed, and added to cultures of the 3C3 mouse NK T cell hybridoma for 16 h. IL-2 release was measured by ELISA. Data shown are from one experiment that was repeated five times. (C) Dose-dependent stimulation of the 3C3 hybridoma in response to increasing amounts of either α-GalCer or biotin–α-GalCer presented by 1 μg/well mCD1 immobilized on a microtiter plate. This experiment is representative of three independent experiments.

Figure 4

Figure 4

mCD1 and hCD1d can bind to immobilized biotin–α-GalCer and biotin–β-GalCer. (A) Binding of mCD1 (0.7 μM), hCD1d (0.5 μM), and H2-M3 (0.9 μM) to biotin–α-GalCer in PBS at 10 μl/min flow rate. I, injection; D, dissociation (washing of the chip in the running buffer). To control for refractive index changes, the RU values for the control flow cell (Fc1) have been subtracted from the RU values for the lipid-coupled flow cell (Fc2) to generate the sensograms shown. The noticeable dip in the binding curve during the last 20–25 s of the injection is due to the increased nonspecific sticking of the protein to flow cell 1, which generates a decreased binding signal in the subtracted (Fc2 − Fc1) sensogram. (B) Binding of 0.04, 0.06, 0.13, 0.22, 0.33, and 0.72 μM mCD1 to immobilized biotin–α-GalCer in PBS at 50 μl/min flow rate. (A and B) Representative data from one of three independent experiments. (C) Fitting the dissociation phase of mCD1 on biotin–α-GalCer surface to a double exponential model produces k off fast 0.1 s−1 and k off slow 0.004 s−1. The contribution of the fast component is 22%. The plot includes observed data and fitted graph. Goodness of the fit is determined by the difference between the fitted values and the observed values (open symbols, plotted as Residual, RU). A residual RU with a value of less than the noise of the instrument (±3) indicates a good fit. (D) Association rate of mCD1 binding to biotin–α-GalCer surface. The slope of the plot of k obs versus mCD1 concentration gives a k on value of 6.7 × 104 M−1s−1. (E) Binding of 0.13, 0.16, 0.22, 0.33, 0.5, and 1.3 μM mCD1 to immobilized biotin–β-GalCer. All conditions were the same as in B.

Figure 4

Figure 4

mCD1 and hCD1d can bind to immobilized biotin–α-GalCer and biotin–β-GalCer. (A) Binding of mCD1 (0.7 μM), hCD1d (0.5 μM), and H2-M3 (0.9 μM) to biotin–α-GalCer in PBS at 10 μl/min flow rate. I, injection; D, dissociation (washing of the chip in the running buffer). To control for refractive index changes, the RU values for the control flow cell (Fc1) have been subtracted from the RU values for the lipid-coupled flow cell (Fc2) to generate the sensograms shown. The noticeable dip in the binding curve during the last 20–25 s of the injection is due to the increased nonspecific sticking of the protein to flow cell 1, which generates a decreased binding signal in the subtracted (Fc2 − Fc1) sensogram. (B) Binding of 0.04, 0.06, 0.13, 0.22, 0.33, and 0.72 μM mCD1 to immobilized biotin–α-GalCer in PBS at 50 μl/min flow rate. (A and B) Representative data from one of three independent experiments. (C) Fitting the dissociation phase of mCD1 on biotin–α-GalCer surface to a double exponential model produces k off fast 0.1 s−1 and k off slow 0.004 s−1. The contribution of the fast component is 22%. The plot includes observed data and fitted graph. Goodness of the fit is determined by the difference between the fitted values and the observed values (open symbols, plotted as Residual, RU). A residual RU with a value of less than the noise of the instrument (±3) indicates a good fit. (D) Association rate of mCD1 binding to biotin–α-GalCer surface. The slope of the plot of k obs versus mCD1 concentration gives a k on value of 6.7 × 104 M−1s−1. (E) Binding of 0.13, 0.16, 0.22, 0.33, 0.5, and 1.3 μM mCD1 to immobilized biotin–β-GalCer. All conditions were the same as in B.

Figure 5

Figure 5

hCD1d binds to immobilized biotin-DPPE with slow association and slow dissociation. (A) Structure of _N_-(biotinoyl)dipalmitoyl-

l

-α-phosphatidylethanolamine (biotin-DPPE). (B) Binding of 1.4, 4, 5.6, and 14 μM hCD1d to immobilized biotin-DPPE in PBS (0.005% Tween) at 2 μl/min flow rate. (C) Fitting the association phase of hCD1d on biotin-DPPE surface gives a k on value of 1.4 × 103 M−1s−1. The plot includes observed data and fitted graph. Goodness of the fit is determined by the difference between the fitted values and the observed values (plotted as Residual, RU). (D) Fitting the dissociation phase of hCD1d on biotin-DPPE surface to a single exponential model produces k off 2.7 × 10−4 s−1.

Figure 6

Figure 6

mCD1 can bind to both peptide and glycolipid ligands. (A) GM1 inhibits binding of mCD1 (1.8 μM) and hCD1d (1 μM) to the biotin–α-GalCer chip. Data are representative of three independent experiments. (B) Gangliosides GM1 and GD1a inhibit α-GalCer presentation by plate-bound mCD1 to the 3C3 hybridoma, whereas the control hexasaccharide has no effect. 1 μg/well (0.2 μM) mCD1 was preincubated with 25 ng/well α-GalCer (0.3 μM), with or without indicated concentrations of competitors. (C) Binding of 2 μM mCD1 to immobilized biotin-P99 peptide, and competition by 0.2 and 23 μM soluble P99 peptide. All conditions were the same as in the legend to Fig. 4 B. (D) mCD1 (1.8 μM) binding to the biotin-P99 chip (dashed lines) and to the biotin–α-GalCer chip (solid line). Biotin-P99 binding is inhibited by increasing concentrations of free P99 but not by GM1. P99 does not affect mCD1 binding to biotin–α-GalCer. All data shown are representative of at least three experiments in each case.

Similar articles

Cited by

References

    1. Porcelli S.A. The CD1 familya third lineage of antigen-presenting molecules. Adv. Immunol. 1995;59:1–98. - PubMed
    1. Bradbury A., Belt K.T., Neri T.M., Milstein C., Calabi F. Mouse CD1 is distinct from and co-exists with TL in the same thymus. EMBO (Eur. Mol. Biol. Organ.) J. 1988;7:3081–3086. - PMC - PubMed
    1. Calabi F., Bradbury A. The CD1 system. Tissue Antigens. 1991;37:1–9. - PubMed
    1. Porcelli S., Morita C.T., Brenner M.B. CD1b restricts the response of human CD4−8− T lymphocytes to a microbial antigen. Nature. 1992;360:593–597. - PubMed
    1. Beckman E.M., Melian A., Behar S.M., Sieling P.A., Chatterjee D., Furlong S.T., Matsumoto R., Rosat J.P., Modlin R.L., Porcelli S.A. CD1c restricts responses of mycobacteria-specific T cells. Evidence for antigen presentation by a second member of the human CD1 family. J. Immunol. 1996;157:2795–2803. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources