The Immunological Synapse: A Molecular Machine Controlling T Cell Activation
Tóm tắt
Từ khóa
Tài liệu tham khảo
E k -GPI and ICAM-1–GPI expressed in CHO and BHK cells were solubilized by probe sonication with buffered 1% Triton X-100 and were captured on 14-4-4 or YN1/1 agarose respectively. Proteins were labeled with Oregon green and Cy5 for E k -GPI and ICAM-1 (Molecular Probes and Amersham). Labeled E k -GPI and ICAM-1 were eluted at high pH and were individually reconstituted in phosphatidylcholine (egg) vesicles (33). Molecular density was determined by immunoradiometric assays. Iodinated 14-4-4 and YN1/1 were used for E k -GPI and ICAM-1–GPI respectively.
Web Movies can be seen at www.sciencemag.org/feature/data/1040037.shl.
We prepared planar bilayers by mixing the E k -GPI and ICAM-1–GPI liposomes 1:1 before incubating them on clean glass (33) in a parallel-plate flow cell (Bioptechs; Butler PA). The E k -GPI was loaded with peptides over 48 hours at 37°C (34). Loading of E k -GPI with MCC88-103 was directly assessed with the D4 monoclonal antibody (mAb) specific to this MHC-peptide complex (35). Lower densities of agonist-loaded complexes were generated by mixing in the appropriate null peptide while keeping the total peptide concentration at 100 μM. Cells were injected into the warmed (37°C) flow cells at time zero and then flow was stopped for the duration of the experiment. Thus any cell movement (migration) is active. The bilayers were imaged as described (23).
MHC-peptide complexes and ICAM-1 are laterally mobile on APCs (10 36). Regulation of lateral mobility of MHC-peptide complexes by cytoskeletal interactions augments some T cell responses but is not essential for T cell activation (36).
Splenocytes from 2B4 TCR transgenic mice were cultured with 1 μM MCC88-103 peptide for 3 days. The cells were then expanded in media with IL-2 (100 U/ml) and used on day 7. CHO cells expressing GPI-anchored 2B4 TCRs were cultured as described (37). Splenocytes from 3.L2 TCR transgenic mice with or without CD4 were depleted of CD8-positive cells by using magnetic beads (Dynal Great Neck NY) and stimulated with 1 μM Hb64-76 peptide. The 3.L2 CD4 −/− TCR transgenic cells are selected on the H-2 k haplotype in the thymus and normal numbers of mature T cells emerge in the periphery. The peripheral cells were either CD8 + which were depleted here or double-negative cells and both populations expressed the clonotypic TCR.
The efficiency of E k loading for different altered peptides was confirmed by competition with the MCC(88-103) peptide as detected by an E k /MCC-specific mAb D4 (35).
Proliferation of CD4-negative 3L.2 T cells on planar bilayer was less than one-tenth that obtained with CD4-positive 3L.2 T cells.
In FPR experiments the spot size of an argon ion laser (488 nm) was adjusted to match the fluorescent cluster (23 38). The cluster was exposed to a brief laser pulse to irreversibly bleach the fluorophores and the images were acquired with the cooled charge-coupled device camera. The recovery process is a function of both the kinetic rates for the receptor-ligand interaction and the ratio of accumulated fluorescence to the density of free ligand molecules in the bilayer. Thus these were matched between systems to be compared. The specific cluster fluorescence intensity at a given time ( I t = cluster intensity − bilayer intensity) divided by the fluorescence intensity of ligands in the bilayer ( I b ) was plotted against time (23 38).
Data from Fig. 1 were used to estimate the two-dimensional (2D) K d . FPR showed that 20% of the 50 000 TCRs on 2B4 T cells were laterally mobile. Our analysis considers only mobile TCRs (39). We assume that the surface area of the T cell is 500 μm 2 . Thus the 2D K d = {[80 free E k (MCC88-103)/μm 2 ] × [10 free TCRs/μm 2 ]} ÷ [100 engaged TCR–E k (MCC88-103)/μm 2 ] = 10 molecules per square micrometer. The entropy-corrected 3D K d is 7800 molecules per cubic micrometer (39). The 2D K d ÷ 3D K d is the confinement region 1.2 nm in this instance. The volume of the close contact in which TCR engagement occurs is ∼5 × 10 −17 liters.
The density and number of accumulated molecules in different regions of the contact area were calculated as the accumulated density (molecules/μm 2 ) = {[intensity in masked area (fluorescence units/μm 2 )] − [intensity in neighboring area (fluorescence units/μm 2 )]} ÷ specific activity (fluorescence units/molecule).
Planar bilayers with E k -GPI and ICAM-1–GPI at 200 molecules/μm 2 each were formed on 100-μm glass beads (Mo-Sci Rolla MO). Bilayers on large glass beads behave identically to planar bilayers. The E k on the coated glass beads (25 mg per well) was loaded with peptides as above. Beads were combined with 10 5 rested T cells in 200 μl of RPMI-1640 10% fetal bovine serum. [ 3 H]Thymidine (0.2 μCi) was added on day 3 the plates were incubated an additional 24 hours and then the cells were harvested on filters for scintillation counting.
We thank R. Houdei for technical assistance and N. Desai for advice on imaging and data analysis; D. Donermeyer for generation of the 3.L2 TCR transgenic mouse lacking CD4; E. Unanue for inspiring our efforts; E. Elson for help with FPR and the loan of the argon ion laser; J. Heuser and H. Hiyakawa for single-particle imaging; A. Chan G. Kersh M. Thomas E. Unanue and H. Virgin for critical reading of the manuscript; and J. Smith for final preparation of the manuscript. Supported by grants from the Whitaker Foundation (M.L.D.) the Arthritis Foundation (M.L.D.) the Howard Hughes Medical Institute (M.M.D.) and the NIH (M.L.D. M.M.D. P.M.A and A.S.S.).