- Email: [email protected]
Sensitive detection of GM1 lipid rafts and TCR partitioning in the T cell membrane
Journal of Immunological Methods 275 (2003) 161 – 168 www.elsevier.com/locate/jim
Sensitive detection of GM1 lipid rafts and TCR partitioning in the T cell membrane S. Thomas, R.S. Kumar, S. Casares, T.-D. Brumeanu* Department of Microbiology, Mount Sinai School of Medicine, 1 Gustave L. Levy Place, New York, NY 10029, USA Received 24 July 2002; received in revised form 9 January 2003; accepted 20 January 2003
Abstract The cholesterol-rich lipid rafts on T cell membrane play important role in the formation of T cell receptor (TCR) signalosome upon receptor ligation. Analytical studies on the kinetics of lipid rafts formation and recruitment of protein receptors to lipid rafts are still limited by the use of a large number of cells. Herein, we describe a strategy for detecting fine alterations in the amount and distribution of glycosphingolipid (GM1) lipid rafts, and in the formation of GM1 – TCR complexes in detergent-insoluble and -soluble compartments of the T cell membrane from a relative low number of cells. Using this strategy, we found that the GM1 moiety was physically associated with TCR in both detergent-insoluble and -soluble fractions. Shortly after ligation of CD3/TCR complex with a soluble CD3-q mAb, the TCR was found mainly in the detergentsoluble fraction of the T cell membrane. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Glycosphingolipid GM1; CD3-mediated TCR cross-linking; GM1 and TCR distribution; ELISA
1. Introduction The plasma membrane contains a lipid bilayer made of diffuse patches of cholesterol and sphingoliAbbreviations: BSA, bovine serum albumin; CTB – HRP, cholera toxin subunit B conjugated with horse radish peroxidase; CD, cluster of differentiation; CLSM, confocal laser scanning microscopy; ELISA, enzyme-linked immunosorbent assay; HA, hemagglutinin; IP, immunoprecipitation; IDB, immunodot blot; mAb, monoclonal antibody; MHC, major histocompatibility complex; MWCO, molecular weight cutoff; PBS, phosphate-buffered saline; PVDF, polyvinylidene fluoride; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; TcH, T cell hybridoma; TCR, T cell receptor; WB, Western blot. * Corresponding author. Tel.: +1-212-241-7551; fax: +1-212828-4151. E-mail address: [email protected] (T.-D. Brumeanu).
pid microdomains, which refers to the detergentinsoluble glycolipid-enriched complexes (DIGs), glycosphingolipid-enriched membranes (GEM), detergent-resistant membranes (DRMs), or lipid rafts (Simons and Ikonen, 1997). The lipid rafts play an important role in membrane trafficking and signal transduction, by means of reorganizing their composition in protein receptors with signaling ability (Harder and Simons, 1997). During the formation of immunological synapse following TCR engagement by MHC –peptide complexes on the surface of APCs, the lipid rafts in T cells recruit various protein receptors like CD2, CD3, CD4, CD28, and TCR (Ohtani et al., 2000) together with protein tyrosine kinases, i.e., p56lck, p59fyn, PI-3K (Hope and Pike, 1996; Simons and Ikonen, 1997; Zhang et al., 1998), and adaptor molecules like LAT and Grb2 (Montixi et al., 1998,
0022-1759/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0022-1759(03)00014-0
162
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
Zhang et al., 1998). Also, some protein phosphatases with role in TCR signaling such as SHP-1 are recruited into the lipid rafts, whereas others such as CD45 extracellular phosphatase are excluded (Janes et al., 1999). Based on these observations, it has been recently suggested that the lipid rafts represent foci of TCR signal transduction (Harder and Simons, 1999, Janes et al., 1999). Herein, we have analyzed the TCR distribution in the lipid rafts, shortly after cross-linking of CD3/TCR complex on the T cell membrane with a soluble CD3-q mAb-ligand. The cell membrane lipid rafts are basically isolated by gradient centrifugation methods using sucrose (Montixi et al., 1998) or Optiprep (Harder and Kuhn, 2000). At present, there is a limitation for measuring the amount of lipid rafts separated by sucrose gradient centrifugation fractions when low-frequency T cell phenotypes, i.e., CD4 and CD8 memory T cells, CD4 Th1 and Th2 cell subsets, CD4+CD25high or CD4+CTLA-4high regulatory T cell subsets, are required for the experiments. Using a modified ELISA protocol, we detected fine alterations in the amount and distribution of GM1 glycosphingolipid, and found transient changes in distribution of TCR in the T cell membrane. Shortly after cross-linking of CD3/TCR complex with a soluble CD3-q mAb, the TCR was found mainly in the detergent-soluble compartment of T cell membrane.
2. Materials and methods 2.1. Confocal laser scanning microscopy We used confocal laser scanning microscopy (CLSM) to visualize GM1 lipid rafts in the T cell membrane, before and after cross-linking of TCR with a soluble CD3 mAb (2C11, ATCC). The 14-3-1 T cell hybridoma (TcH) used in these experiments expresses the 14.3d T cell receptor (TCR) recognizing the HA110 – 120 CD4 immunodominant epitope of PR8 influenza virus in the context of I-Ed class II molecules (Weber et al., 1992). Nonstimulated and CD3stimulated TcH (2C11 mAb at 10 Ag/ml/106 cells) for 30 min, 2 h, and 4 h at 37 jC were washed, and treated with CTB – biotin conjugate (10 Ag/ml) for 30 min at 4 jC followed by washing in PBS containing 1% BSA and 0.05% sodium azide (2000 rpm, 4 jC).
The cell pellet was incubated with streptavidin –FITC conjugate (10 Ag/ml) for 30 min at 4 jC, washed and fixed in 1% paraformaldehyde, mounted in Vectashield medium (Vector Laboratories, Burlingame, CA), and sealed with Permount (Fisher Scientific, New Jersey). Sixteen cross sections (0.5 Am/section) of the stained cells were analyzed in an inverted Leica confocal laser scanning microscope equipped with a fluorescence filter set for excitation at 488 nm (Leica Lasertechnik, Heidelberg, Germany). 2.2. Sucrose gradient centrifugation The GM1 lipid rafts were isolated from the plasma membranes of nonstimulated and CD3-stimulated 143-1 TcH for 30 min at 37 jC, using a previously described sucrose gradient centrifugation protocol (Cottin et al., 2002) with some modifications. Briefly, TcH (50 106) in RPMI containing 10% BSA were first stimulated or not for 30 min at 37 jC with soluble 2C11 mAb (10 Ag/106 cells/ml), cells were washed, and re-suspended in 1 ml of buffer A (25 mM Tris, 150 mM NaCl, 5 mM EDTA supplemented with a cocktail of protease inhibitors (Roche Biochemicals, USA). Cells were sonicated on ice for five cycles of 10 s at 100 W per cycle, the lysates were cleared by 5min centrifugation at 300 g, and the cell membranes were solubilized for 1 h at 4 jC with 1% Brij 58 in buffer A. In contrast to Triton X-100, Brij 58 detergent preserves the noncovalent association between TCR and glycolipid moieties. The solubilized cell membranes (1 ml) were mixed with 1 ml of 80% sucrose in buffer A (w/v), and transferred to 13.5 ml ultra-clear centrifuge tubes followed by overlaying of 8 ml sucrose 35%, and 2 ml of 5% sucrose made in buffer A. Samples were centrifuged at 38,000 rpm (288,000 g) for 15 h at 4 jC. Nine fractions of 1 ml each were collected from the top of the tube, dialyzed in 0.0015 M saline using Spectra/Por 6 dialysis bags (MWCO 1000 kDa, Sigma, St. Louis, MO), and concentrated by speed vacuum centrifugation to a final volume of 100 Al per fraction. 2.3. ELISA To determine the amount of GM1 in T cell membrane fractions purified by sucrose gradient centrifugation from nonstimulated and CD3-stimulated T cell
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
for 30 min, we used a modified ELISA protocol. Several dilutions of the nine individual fractions (1:100, 1:5000, and 1:10,000 in PBS) were coated overnight at 4 jC onto plastic surface with high affinity for glycolipids (96-well Polysorp star well plates, 200 Al/well, Nunc, Denmark). The plates were blocked for 30 min at room temperature with 3% BSA/PBS, and then incubated for 1 h at 37 jC with 5 Ag/ml of CTB – HRP conjugate in 1% BSA/PBS. CTB is a natural ligand that binds strongly and specifically to GM1 moieties. The plates were washed with PBS/ 0.05% Tween 20, incubated for 20 min with 200 Al of HRP chromogen substrate (#SB01, Biosource International, Camarillo, CA), and the reaction was stopped by the addition of 25 Al of 2 M sulphuric acid. The optical density (k = 450 nm) was measured in a Powerwave X ELISA reader in triplicate wells, using the K4 analysis software (Bio-Tek Instruments, USA). To determine the amount of TCR in sucrose gradient fractions, we used 96-well Maxisorp ELISA plates with high avidity for proteins (Nunc). In parallel assays, we analyzed these fractions to find whether GM1 interacts with TCR, using Maxisorp ELISA plates pre-coated with CTB at 5 Ag/ml in carbonate buffer 0.1 M, pH 9.8. The plates were incubated overnight at 4 jC with a pool from fractions 1 – 4 (10 Al of each fraction re-suspended in 200 Al of 1% BSA/PBS, 200 Al/well), which showed the highest amount of GM1 (GM1high pool), or with a pool of fractions 5 –8, which showed the lowest amount of GM1 (GM1low pool) as found by Polysorp ELISA. Plates were blocked with 3% BSA in PBS for 2 h at 37 jC, and incubated overnight at 4 jC with 6.5.2 clonotypic mAb (5 Ag/ml in 1% BSA/PBS). The clonotypic 6.5.2 mAb is a rat IgG1/k monoclonal antibody obtained in the laboratory by affinity chromatography on a rat kappa chain mAb (MAR18.5, ATCC) coupled to Sepharose CL-4B (AmershamPharmacia, New Jersey). The 6.5.2 mAb recognizes the 14.3d TCR specific for HA110 –120 peptide, and it does not cross-react with other Vh8 chain-expressing TCR molecules (Weber et al, 1992). Plates were washed with PBS/0.05% Tween 20, and bound 6.5.2 mAb was revealed with Protein A –HRP conjugate (Sigma, 1:20,000 dilution in 1% BSA/PBS/0.05% Tween 20) for 2 h incubation at room temperature. Plates were washed, incubated with 200 Al of HRP chromogen substrate for 20 min, and the optical
163
density was measured in the ELISA reader, as described. 2.4. Immunoblotting SDS-PAGE in 4– 15% gradient PhastGels (Amersham-Pharmacia) was used to analyze the pattern of proteins in sucrose gradient fractions of nonstimulated and CD3-stimulated 14-3-1 TcH membranes (10 Ag/ ml/106 cells). Some 2 Al of each sucrose gradient fraction in sample buffer containing 0.1% SDS and 1% 2-ME were boiled for 5 min, separated in 4 –15% gradient PhastGels, and then stained with silver nitrate following the protocol recommended by the manufacturer. To detect the GM1 glycosphingolipid in sucrose gradient fractions, we used 2 Al of individual fractions in electrophoresis sample buffer. The proteins were separated in 4– 15% gradient PhastGels, the gels were electrotransferred onto PVDF membrane using the PhastSystem transfer unit, blocked overnight at 4 jC with 3% BSA/PBS/0.05% Tween 20, and then probed for 2 h at room temperature with 5 Ag/ml of CTB – HRP conjugate. Membranes were washed with PBS/ 0.05% Tween 20, and the enzyme activity was detected by chemiluminiscence (EDL kit, Roche Biochemical). For the immunodot blot analysis, we used 3 Al of each sucrose gradient fraction. Fractions were applied on PVDF membrane, the membrane was blocked with 3% BSA in PBS/0.05% Tween 20, and probed for 2 h at room temperature with 5 Ag/ml of CTB – HRP conjugate. The membrane was washed with PBS/ 0.05% Tween 20, and the HRP activity was detected by chemiluminiscence, as described. The presence of 14.3d TCR into the GM1high and GM1low pools from nonstimulated and CD3-stimulated TCH for 30 min were analyzed by Western blot. The GM1 high and GM1 low pools, prepared as described, were precipitated for 2 h at room temperature with 6.5.2 clonotypic mAb (2 Ag) bound to Sepharose-Protein A/G plus (20 Al of 50% slurry, Santa Cruz Biotech, Santa Cruz, CA). The immunoprecipitates were washed three times with PBS/0.05% Tween 20, boiled for 5 min in 25 Al of SDS/2-ME electrophoresis buffer, and separated by SDS-PAGE in 4– 15% PhastGels. Gels were electrotransferred on PVDF membrane, blocked overnight at 4 jC with 3%
164
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
BSA/PBS/0.05% Tween 20, and then probed overnight at 4 jC with 5 Ag/ml of F23.1 mAb (ATCC). The F23.1 mAb is a mouse IgG1/k mAb that recognizes the TCR Vh8 chain. Bound F23.1 mAb was revealed with Protein A – HRP conjugate for 2 h at room temperature, and the HRP activity was detected by chemiluminiscence.
sodium azide. Bound 6.5.2 mAb was revealed upon 30 min incubation on ice with a goat F(abV)2 anti-rat g1 Ab – FITC conjugate (BD Pharmingen, California). Cells were washed, fixed in 1% paraformaldehyde/ PBS, and analyzed for the fluorescence intensity using a Beckman Coulter Cytofluorimeter.
2.5. Cytofluorimetric analysis
3. Results and discussion
The expression of 14.3d TCR on nonstimulated vs. CD3-stimulated TcH for 30 min (2C11 mAb 10 Ag/ ml/106 cells) was determined by FACS. In parallel assays, we measured the extent of TCR internalization using cells previously incubated for 4 h with Monensin (50 AM), and then stained for 30 min on ice with 6.5.2 mAb (2 Ag/106 cells) in 1% BSA/PBS/0.5%
T cells stimulated for 4 h with a soluble CD3-q mAb and analyzed by CLSM showed enlarged GM1 patches on cell surface, which clearly indicated an increase in the amount of GM1 lipid rafts on T cell membrane upon CD3 ligation (Fig. 1c and d) as compared with nonstimulated cells (Fig. 1a and b). There was no significant increase in GM1 patches in
Fig. 1. Visualization of GM1 lipid rafts in T cell membrane upon CD3/TCR cross-linking with a soluble CD3-q mAb-ligand. Panels a and b show two CLSM cross sections (0.5 Am) from nonstimulated T cells. Panels c and d show two CLSM cross sections (0.5 Am) from CD3stimulated T cells for 4 h at 37 jC. Increased green fluorescence of cell membrane in a middle range cross section (panel c), and enlarged green fluorescent patches closed to the apical area of the cell (panel d) indicate the formation of GM1 lipid rafts, as identified with CTB ligand – FITC conjugate.
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
cells stimulated with the soluble 2C11 CD3-q mAb for only 30 min. However, large GM1 patches were visible 2 h after CD3 stimulation, suggesting that organization of GM1 lipid rafts upon CD3/TCR cross-linking with a soluble CD3 ligand is a relatively slow, ongoing process.
165
We compared the GM1 content in individual sucrose gradient fractions from the membrane of nonstimulated and CD3-stimulated T cells (for 30 min) using a modified ELISA protocol. The highest amount of GM1 detected by ELISA, as well as by Western blotting and immunodot blotting, was in the fractions
Fig. 2. Partitioning of GM1 lipid rafts in the membrane of nonstimulated and CD3-stimulated T cells. Panel A: Direct coating of sucrose gradient fractions on Polysorp ELISA plates for the analysis of GM1 distribution in detergent-insoluble and -soluble compartments of nonstimulated and CD3-stimulated T cell membrane (for 30 min) with soluble 2C11 mAb. The background of CTB – HRP conjugate against BSA blocking reagent was subtracted from each sample. Values represent the mean of triplicate wells F S.D. Panels B and C: Identification of GM1 distribution in the same fractions by Western blot and immunodot blot analyses, respectively. GM1 was revealed by chemiluminiscence using CTB – HRP conjugate. Panel D: The protein pattern in the same fractions as analyzed by SDS-PAGE under denaturing and reducing conditions in 4 – 15% gradient PhastGels, and developed by silver staining. The M lane indicates the molecular weight markers.
166
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
1 – 4 from both nonstimulated and CD3-stimulated cells (Fig. 2A –C). However, ELISA showed a 1000fold greater sensitivity than Western blot and immunodot blot analyses. Up to 5 104 cells could be used in these experiments, and the GM1 moieties would still be detected by this ELISA protocol. This would make possible the analysis of GM1 moieties in populations of T cells with a low frequency, such as memory or regulatory cells. The amount of GM1 was almost two times higher in fractions 1 – 4 from CD3-stimulated T cells than in nonstimulated ones. A 350-fold increase in the amount of GM1 in human CD4 T cell clones stimulated for 30 min with a CD3 mAb has been found using an ELISA protocol, in which GM1 was captured by CTB-coated plates and revealed with anti-CTB – HRP conjugate (Tuosto et al., 2001). The authors suggested that the increase was the result of intensive de novo synthesis of GM1 moieties. The difference between these results and ours could be explained by different T cell lines, ELISA protocol, and the type of CD3 mAb-ligand used in their experiments. Interesting enough, we found that GM1 was present in the detergent-soluble fractions (#5 – 8) from CD3stimulated T cells, although at lesser extent than in the detergent-insoluble fractions (#1 – 4). The increased amount of GM1 –protein complexes upon CD3 ligation may explain why GM1 moiety was found in higher sucrose density gradients. SDS-PAGE analysis in individual fractions from CD3-stimulated T cells showed a redistribution of several protein bands from the detergent-insoluble (#1– 3) to detergent-soluble fractions (#4 – 8) (Fig. 2D). Particularly, some low-molecularsize protein bands were decreased in the detergentinsoluble fractions, whereas some high-molecular-size protein bands were enriched in the detergent-soluble
fractions from CD3-stimulated cells (Fig. 2D). Montixi et al. demonstrated that the stimulation of T cells by specific ligands led to distribution of TCR molecules mostly in the detergent-insoluble fraction (Montixi et al., 1998, Xavier et al., 1998), which suggested that membrane proteins are differentially recruited to GM1 lipid rafts. It has been demonstrated that high-molecular species like CD45 extracellular phosphatase are excluded from lipid rafts during the formation of immunological synapse, whereas small molecular species are preferentially recruited to lipid rafts (Janes et al., 1999). Since we found that GM1 moieties are also present in the detergent-soluble fractions of CD3stimulated T cells, we questioned whether (1) GM1 could form complexes with 14.3d TCR, (2) CD3/ TCR cross-linking with a soluble 2C11 CD3-q mAb may favor the formation of such complexes, and (3) whether GM1 – TCR complexes could form preferentially in detergent-insoluble (GM1high) or in detergent-soluble (GM1low) compartments. ELISA showed that 14.3d TCR was almost equally distributed in the detergent-insoluble and -soluble fractions of nonstimulated TcH, whereas its amount was three times higher in the detergent-soluble fractions of CD3-stimulated cells (Fig. 3A). This was also confirmed by Western blot analysis (Fig. 3B). These data indicated that CD3/TCR ligation by a soluble ligand can lead to the enrichment of detergentsoluble compartment of T cell membrane in TCR molecules. It is unlikely that this was the result in TCR upmodulation in this compartment upon CD3 ligation, since it has been reported that TCR is usually dowmmodulated upon CD3 ligation. Our cytofluorimetric analysis showed that, apparently,
Fig. 3. Partitioning of TCR in the membrane lipid rafts from nonstimulated and CD3-stimulated T cells. Panel A: ELISA assay for measuring the amount of 14.3d TCR in the GM1high and GM1low pooled fractions in nonstimulated and CD3-stimulated T cell membrane (for 30 min) with soluble 2C11 mAb. The background of 6.5.2 mAb and Protein A – HRP conjugate against BSA blocking reagent was subtracted from each sample. Values represent the mean of triplicate wells F S.D. Panel B: Western blot analysis under denaturing and reducing conditions of GM1high and GM1low pooled fractions. The major band of 55 kDa represents the heavy chain of clonotypic 6.5.2 precipitating mAb revealed by Protein A – HRP conjugate used to develop the F23.1 mAb binding to Vh8 chain of 14.3d TCR. Panel C: Cytofluorimetric analysis of 14.3d TCR expression on nonstimulated and CD3-stimulated T cells for (30 min) with soluble 2C11 mAb, which were either non-incubated (left panel) or previously incubated with Monensin (right panel). The TCR expression identified with 6.5.2 clonotypic mAb was revealed with a secondary antibody (goat F(abV)2 anti-rat g1 Ab – FITC conjugate). The TCR expression was calculated in relation to the log fluorescence intensity of bound 6.5.2-secondary Ab – FITC complexes to the cells. The signal-to-noise background from the secondary (2.5 – 3%) was subtracted and eliminated from the histograms. The median of peak fluorescence in each sample was used to calculate the difference in fluorescence intensity. Panel D: ELISA capture assay for determining the amount of GM1 – TCR complexes in GM1high and GM1low pooled fractions from nonstimulated and CD3-stimulated T cell membrane (for 30 min) with soluble 2C11 mAb. The background of 6.5.2 mAb and protein A – HRP conjugate against CTB coated plates blocked with BSA was subtracted from each sample. Values represent the mean of triplicate wells F S.D.
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
167
168
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
the TCR expression was increased by 30% after 30 min cross-linking of CD3/TCR complex with 2C11 mAb, as indicated by the fluorescence of 6.5.2 mAb – FITC conjugate bound to TCR. However, pretreatment of cells with Monensin, which inhibits protein translocation on the cell surface but does not affect protein internalization, showed a decrease by 10– 12% in TCR expression, which clearly indicated a TCR internalization. The apparent increase in TCR expression in CD3-stimulated cells in the absence of Monensin pretreatment was most probably the result of enhanced accessibility of 14.3d TCR to the 6.5.2 binding mAb as a result of allosteric changes mediated by the CD3/TCR cross-linking. The fact that both GM1 and TCR were found increased by ELISA in the detergent-soluble fractions of CD3-stimulated T cells raised the question of whether GM1 may co-exist with TCR in the same complex. Using a CTB capture ELISA, we found that TCR was physically associated with GM1 in both detergent-insoluble and -soluble fractions (Fig. 3D). However, the amount of GM1– TCR complexes was lower in the detergent-soluble fraction than in insoluble one, suggesting that the GM1 –TCR interactions are more stable in the detergent-insoluble fraction. As indicated by Western blot analysis carried out under denaturing and reducing conditions, these interactions were noncovalent, since GM1 was identified as a single, low-molecular band in each sucrose gradient fraction, both in nonstimulated and CD3-stimulated cells. A more detailed analysis will be required to find out whether GM1 may preferentially interact with other protein receptors on the T cell membrane. In conclusion, our results indicated that the shortterm ligation of CD3/TCR complexes on T cells with a soluble CD3 mAb-ligand led to accumulation of TCR in the detergent-soluble compartment of T cell membrane. The GM1 lipid rafts interacted physically with TCR in both detergent-insoluble and -soluble compartments, although more stable in the detergent-insoluble compartment. Subsequent to these findings, several important questions need to be addressed: (1) what is the structural basis of GM1 –TCR complex formation, (2) what is the rate of TCR internalization in the membrane detergent-insoluble compartment vs. the soluble one, (3) whether GM1 –TCR complex trafficking between the plasma membrane microdomains is possible, and under what conditions, and (4) what is the
nature of TCR signaling in the detergent-soluble compartment of T cell membrane. Acknowledgements This work received supported through grants from the National Institutes of Health (1RO1 DK61927 and 1R21 DK61326) to T-D.B. References Cottin, V., Doan, J.E.S., Riches, D.W.H., 2002. Restricted localization of the TNF receptor CD120a to lipid rafts: a novel role of the death domain. J. Immunol. 168, 4095. Harder, T., Kuhn, M., 2000. Selective accumulation of lipid raft associated membrane protein LAT in T-cell receptor signaling assemblies. J. Cell Biol. 151, 199. Harder, T., Simons, K., 1997. Caveolae, DIGs and the dynamics of sphingolipid – cholesterol microdomains. Curr. Opin. Cell Biol. 9, 534. Harder, T., Simons, K., 1999. Clusters of glycolipid and glycosylphospatidylinositol anchored proteins in lymphoid cells: accumulation of actin regulated by local tyrosine phosphorylation. Eur. J. Immunol. 29, 556. Hope, H.R., Pike, L.I., 1996. Phosphoiniositides and phosphoinositide utilizing enzymes in detergent insoluble lipid domains. Mol. Biol. Cell 7, 843. Janes, P.W., Ley, S.C., Magee, A.I., 1999. Aggregation of lipid rafts accompanies signaling via the T-cell antigen receptor. J. Cell Biol. 147, 447. Montixi, C., Langlet, C., Bernard, A.-M., Thimonier, J., Dubois, C., Wurbel, M.-A., Chauvin, J.P., Pierres, M., He, H.-T., 1998. Engagement of T-cell receptor triggers its recruitment to low density detergent insoluble membrane domains. EMBO J. 17, 5334. Ohtani, Y.Y., Zhou, X.-Y., Toyooka, K., Tai, X.-G., Park, C.S., Hamaoka, T., Abe, R., Miyake, K., Fujiwara, H., 2000. NonCD28 costimulatory molecules present in T-cell rafts induce Tcell co-stimulation by enhancing the association of TCR with rafts. J. Immunol. 164, 1251. Simons, K., Ikonen, E., 1997. Functional rafts in cell membranes. Nature 387, 569. Tuosto, L., Parolini, I., Schroder, S., Sargiacomo, M., Lanzavecchia, A., Viola, A., 2001. Organization of plasma membrane functional rafts upon T cell activation. Eur. J. Immunol. 31, 345. Weber, S., Trannecker, A., Olivery, F., Gerhard, W., Karjalainen, K., 1992. Specific low affinity recognition of major histocompatibility complex plus peptide by a soluble T-cell receptor. Nature 356, 793. Xavier, R., Brennan, T., Li, Q., McCormack, C., Seed, B., 1998. Membrane compartmentation is required for efficient T-cell activation. Immunity 8, 723. Zhang, W., Trible, R.P., Samelson, L.E., 1998. LAT palmitoylation: its essential role in membrane microdomain targeting and tyrosine phosphorylation during T-cell activation. Immunity 9, 239.
Sensitive detection of GM1 lipid rafts and TCR partitioning in the T cell membrane S. Thomas, R.S. Kumar, S. Casares, T.-D. Brumeanu* Department of Microbiology, Mount Sinai School of Medicine, 1 Gustave L. Levy Place, New York, NY 10029, USA Received 24 July 2002; received in revised form 9 January 2003; accepted 20 January 2003
Abstract The cholesterol-rich lipid rafts on T cell membrane play important role in the formation of T cell receptor (TCR) signalosome upon receptor ligation. Analytical studies on the kinetics of lipid rafts formation and recruitment of protein receptors to lipid rafts are still limited by the use of a large number of cells. Herein, we describe a strategy for detecting fine alterations in the amount and distribution of glycosphingolipid (GM1) lipid rafts, and in the formation of GM1 – TCR complexes in detergent-insoluble and -soluble compartments of the T cell membrane from a relative low number of cells. Using this strategy, we found that the GM1 moiety was physically associated with TCR in both detergent-insoluble and -soluble fractions. Shortly after ligation of CD3/TCR complex with a soluble CD3-q mAb, the TCR was found mainly in the detergentsoluble fraction of the T cell membrane. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Glycosphingolipid GM1; CD3-mediated TCR cross-linking; GM1 and TCR distribution; ELISA
1. Introduction The plasma membrane contains a lipid bilayer made of diffuse patches of cholesterol and sphingoliAbbreviations: BSA, bovine serum albumin; CTB – HRP, cholera toxin subunit B conjugated with horse radish peroxidase; CD, cluster of differentiation; CLSM, confocal laser scanning microscopy; ELISA, enzyme-linked immunosorbent assay; HA, hemagglutinin; IP, immunoprecipitation; IDB, immunodot blot; mAb, monoclonal antibody; MHC, major histocompatibility complex; MWCO, molecular weight cutoff; PBS, phosphate-buffered saline; PVDF, polyvinylidene fluoride; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; TcH, T cell hybridoma; TCR, T cell receptor; WB, Western blot. * Corresponding author. Tel.: +1-212-241-7551; fax: +1-212828-4151. E-mail address: [email protected] (T.-D. Brumeanu).
pid microdomains, which refers to the detergentinsoluble glycolipid-enriched complexes (DIGs), glycosphingolipid-enriched membranes (GEM), detergent-resistant membranes (DRMs), or lipid rafts (Simons and Ikonen, 1997). The lipid rafts play an important role in membrane trafficking and signal transduction, by means of reorganizing their composition in protein receptors with signaling ability (Harder and Simons, 1997). During the formation of immunological synapse following TCR engagement by MHC –peptide complexes on the surface of APCs, the lipid rafts in T cells recruit various protein receptors like CD2, CD3, CD4, CD28, and TCR (Ohtani et al., 2000) together with protein tyrosine kinases, i.e., p56lck, p59fyn, PI-3K (Hope and Pike, 1996; Simons and Ikonen, 1997; Zhang et al., 1998), and adaptor molecules like LAT and Grb2 (Montixi et al., 1998,
0022-1759/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0022-1759(03)00014-0
162
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
Zhang et al., 1998). Also, some protein phosphatases with role in TCR signaling such as SHP-1 are recruited into the lipid rafts, whereas others such as CD45 extracellular phosphatase are excluded (Janes et al., 1999). Based on these observations, it has been recently suggested that the lipid rafts represent foci of TCR signal transduction (Harder and Simons, 1999, Janes et al., 1999). Herein, we have analyzed the TCR distribution in the lipid rafts, shortly after cross-linking of CD3/TCR complex on the T cell membrane with a soluble CD3-q mAb-ligand. The cell membrane lipid rafts are basically isolated by gradient centrifugation methods using sucrose (Montixi et al., 1998) or Optiprep (Harder and Kuhn, 2000). At present, there is a limitation for measuring the amount of lipid rafts separated by sucrose gradient centrifugation fractions when low-frequency T cell phenotypes, i.e., CD4 and CD8 memory T cells, CD4 Th1 and Th2 cell subsets, CD4+CD25high or CD4+CTLA-4high regulatory T cell subsets, are required for the experiments. Using a modified ELISA protocol, we detected fine alterations in the amount and distribution of GM1 glycosphingolipid, and found transient changes in distribution of TCR in the T cell membrane. Shortly after cross-linking of CD3/TCR complex with a soluble CD3-q mAb, the TCR was found mainly in the detergent-soluble compartment of T cell membrane.
2. Materials and methods 2.1. Confocal laser scanning microscopy We used confocal laser scanning microscopy (CLSM) to visualize GM1 lipid rafts in the T cell membrane, before and after cross-linking of TCR with a soluble CD3 mAb (2C11, ATCC). The 14-3-1 T cell hybridoma (TcH) used in these experiments expresses the 14.3d T cell receptor (TCR) recognizing the HA110 – 120 CD4 immunodominant epitope of PR8 influenza virus in the context of I-Ed class II molecules (Weber et al., 1992). Nonstimulated and CD3stimulated TcH (2C11 mAb at 10 Ag/ml/106 cells) for 30 min, 2 h, and 4 h at 37 jC were washed, and treated with CTB – biotin conjugate (10 Ag/ml) for 30 min at 4 jC followed by washing in PBS containing 1% BSA and 0.05% sodium azide (2000 rpm, 4 jC).
The cell pellet was incubated with streptavidin –FITC conjugate (10 Ag/ml) for 30 min at 4 jC, washed and fixed in 1% paraformaldehyde, mounted in Vectashield medium (Vector Laboratories, Burlingame, CA), and sealed with Permount (Fisher Scientific, New Jersey). Sixteen cross sections (0.5 Am/section) of the stained cells were analyzed in an inverted Leica confocal laser scanning microscope equipped with a fluorescence filter set for excitation at 488 nm (Leica Lasertechnik, Heidelberg, Germany). 2.2. Sucrose gradient centrifugation The GM1 lipid rafts were isolated from the plasma membranes of nonstimulated and CD3-stimulated 143-1 TcH for 30 min at 37 jC, using a previously described sucrose gradient centrifugation protocol (Cottin et al., 2002) with some modifications. Briefly, TcH (50 106) in RPMI containing 10% BSA were first stimulated or not for 30 min at 37 jC with soluble 2C11 mAb (10 Ag/106 cells/ml), cells were washed, and re-suspended in 1 ml of buffer A (25 mM Tris, 150 mM NaCl, 5 mM EDTA supplemented with a cocktail of protease inhibitors (Roche Biochemicals, USA). Cells were sonicated on ice for five cycles of 10 s at 100 W per cycle, the lysates were cleared by 5min centrifugation at 300 g, and the cell membranes were solubilized for 1 h at 4 jC with 1% Brij 58 in buffer A. In contrast to Triton X-100, Brij 58 detergent preserves the noncovalent association between TCR and glycolipid moieties. The solubilized cell membranes (1 ml) were mixed with 1 ml of 80% sucrose in buffer A (w/v), and transferred to 13.5 ml ultra-clear centrifuge tubes followed by overlaying of 8 ml sucrose 35%, and 2 ml of 5% sucrose made in buffer A. Samples were centrifuged at 38,000 rpm (288,000 g) for 15 h at 4 jC. Nine fractions of 1 ml each were collected from the top of the tube, dialyzed in 0.0015 M saline using Spectra/Por 6 dialysis bags (MWCO 1000 kDa, Sigma, St. Louis, MO), and concentrated by speed vacuum centrifugation to a final volume of 100 Al per fraction. 2.3. ELISA To determine the amount of GM1 in T cell membrane fractions purified by sucrose gradient centrifugation from nonstimulated and CD3-stimulated T cell
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
for 30 min, we used a modified ELISA protocol. Several dilutions of the nine individual fractions (1:100, 1:5000, and 1:10,000 in PBS) were coated overnight at 4 jC onto plastic surface with high affinity for glycolipids (96-well Polysorp star well plates, 200 Al/well, Nunc, Denmark). The plates were blocked for 30 min at room temperature with 3% BSA/PBS, and then incubated for 1 h at 37 jC with 5 Ag/ml of CTB – HRP conjugate in 1% BSA/PBS. CTB is a natural ligand that binds strongly and specifically to GM1 moieties. The plates were washed with PBS/ 0.05% Tween 20, incubated for 20 min with 200 Al of HRP chromogen substrate (#SB01, Biosource International, Camarillo, CA), and the reaction was stopped by the addition of 25 Al of 2 M sulphuric acid. The optical density (k = 450 nm) was measured in a Powerwave X ELISA reader in triplicate wells, using the K4 analysis software (Bio-Tek Instruments, USA). To determine the amount of TCR in sucrose gradient fractions, we used 96-well Maxisorp ELISA plates with high avidity for proteins (Nunc). In parallel assays, we analyzed these fractions to find whether GM1 interacts with TCR, using Maxisorp ELISA plates pre-coated with CTB at 5 Ag/ml in carbonate buffer 0.1 M, pH 9.8. The plates were incubated overnight at 4 jC with a pool from fractions 1 – 4 (10 Al of each fraction re-suspended in 200 Al of 1% BSA/PBS, 200 Al/well), which showed the highest amount of GM1 (GM1high pool), or with a pool of fractions 5 –8, which showed the lowest amount of GM1 (GM1low pool) as found by Polysorp ELISA. Plates were blocked with 3% BSA in PBS for 2 h at 37 jC, and incubated overnight at 4 jC with 6.5.2 clonotypic mAb (5 Ag/ml in 1% BSA/PBS). The clonotypic 6.5.2 mAb is a rat IgG1/k monoclonal antibody obtained in the laboratory by affinity chromatography on a rat kappa chain mAb (MAR18.5, ATCC) coupled to Sepharose CL-4B (AmershamPharmacia, New Jersey). The 6.5.2 mAb recognizes the 14.3d TCR specific for HA110 –120 peptide, and it does not cross-react with other Vh8 chain-expressing TCR molecules (Weber et al, 1992). Plates were washed with PBS/0.05% Tween 20, and bound 6.5.2 mAb was revealed with Protein A –HRP conjugate (Sigma, 1:20,000 dilution in 1% BSA/PBS/0.05% Tween 20) for 2 h incubation at room temperature. Plates were washed, incubated with 200 Al of HRP chromogen substrate for 20 min, and the optical
163
density was measured in the ELISA reader, as described. 2.4. Immunoblotting SDS-PAGE in 4– 15% gradient PhastGels (Amersham-Pharmacia) was used to analyze the pattern of proteins in sucrose gradient fractions of nonstimulated and CD3-stimulated 14-3-1 TcH membranes (10 Ag/ ml/106 cells). Some 2 Al of each sucrose gradient fraction in sample buffer containing 0.1% SDS and 1% 2-ME were boiled for 5 min, separated in 4 –15% gradient PhastGels, and then stained with silver nitrate following the protocol recommended by the manufacturer. To detect the GM1 glycosphingolipid in sucrose gradient fractions, we used 2 Al of individual fractions in electrophoresis sample buffer. The proteins were separated in 4– 15% gradient PhastGels, the gels were electrotransferred onto PVDF membrane using the PhastSystem transfer unit, blocked overnight at 4 jC with 3% BSA/PBS/0.05% Tween 20, and then probed for 2 h at room temperature with 5 Ag/ml of CTB – HRP conjugate. Membranes were washed with PBS/ 0.05% Tween 20, and the enzyme activity was detected by chemiluminiscence (EDL kit, Roche Biochemical). For the immunodot blot analysis, we used 3 Al of each sucrose gradient fraction. Fractions were applied on PVDF membrane, the membrane was blocked with 3% BSA in PBS/0.05% Tween 20, and probed for 2 h at room temperature with 5 Ag/ml of CTB – HRP conjugate. The membrane was washed with PBS/ 0.05% Tween 20, and the HRP activity was detected by chemiluminiscence, as described. The presence of 14.3d TCR into the GM1high and GM1low pools from nonstimulated and CD3-stimulated TCH for 30 min were analyzed by Western blot. The GM1 high and GM1 low pools, prepared as described, were precipitated for 2 h at room temperature with 6.5.2 clonotypic mAb (2 Ag) bound to Sepharose-Protein A/G plus (20 Al of 50% slurry, Santa Cruz Biotech, Santa Cruz, CA). The immunoprecipitates were washed three times with PBS/0.05% Tween 20, boiled for 5 min in 25 Al of SDS/2-ME electrophoresis buffer, and separated by SDS-PAGE in 4– 15% PhastGels. Gels were electrotransferred on PVDF membrane, blocked overnight at 4 jC with 3%
164
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
BSA/PBS/0.05% Tween 20, and then probed overnight at 4 jC with 5 Ag/ml of F23.1 mAb (ATCC). The F23.1 mAb is a mouse IgG1/k mAb that recognizes the TCR Vh8 chain. Bound F23.1 mAb was revealed with Protein A – HRP conjugate for 2 h at room temperature, and the HRP activity was detected by chemiluminiscence.
sodium azide. Bound 6.5.2 mAb was revealed upon 30 min incubation on ice with a goat F(abV)2 anti-rat g1 Ab – FITC conjugate (BD Pharmingen, California). Cells were washed, fixed in 1% paraformaldehyde/ PBS, and analyzed for the fluorescence intensity using a Beckman Coulter Cytofluorimeter.
2.5. Cytofluorimetric analysis
3. Results and discussion
The expression of 14.3d TCR on nonstimulated vs. CD3-stimulated TcH for 30 min (2C11 mAb 10 Ag/ ml/106 cells) was determined by FACS. In parallel assays, we measured the extent of TCR internalization using cells previously incubated for 4 h with Monensin (50 AM), and then stained for 30 min on ice with 6.5.2 mAb (2 Ag/106 cells) in 1% BSA/PBS/0.5%
T cells stimulated for 4 h with a soluble CD3-q mAb and analyzed by CLSM showed enlarged GM1 patches on cell surface, which clearly indicated an increase in the amount of GM1 lipid rafts on T cell membrane upon CD3 ligation (Fig. 1c and d) as compared with nonstimulated cells (Fig. 1a and b). There was no significant increase in GM1 patches in
Fig. 1. Visualization of GM1 lipid rafts in T cell membrane upon CD3/TCR cross-linking with a soluble CD3-q mAb-ligand. Panels a and b show two CLSM cross sections (0.5 Am) from nonstimulated T cells. Panels c and d show two CLSM cross sections (0.5 Am) from CD3stimulated T cells for 4 h at 37 jC. Increased green fluorescence of cell membrane in a middle range cross section (panel c), and enlarged green fluorescent patches closed to the apical area of the cell (panel d) indicate the formation of GM1 lipid rafts, as identified with CTB ligand – FITC conjugate.
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
cells stimulated with the soluble 2C11 CD3-q mAb for only 30 min. However, large GM1 patches were visible 2 h after CD3 stimulation, suggesting that organization of GM1 lipid rafts upon CD3/TCR cross-linking with a soluble CD3 ligand is a relatively slow, ongoing process.
165
We compared the GM1 content in individual sucrose gradient fractions from the membrane of nonstimulated and CD3-stimulated T cells (for 30 min) using a modified ELISA protocol. The highest amount of GM1 detected by ELISA, as well as by Western blotting and immunodot blotting, was in the fractions
Fig. 2. Partitioning of GM1 lipid rafts in the membrane of nonstimulated and CD3-stimulated T cells. Panel A: Direct coating of sucrose gradient fractions on Polysorp ELISA plates for the analysis of GM1 distribution in detergent-insoluble and -soluble compartments of nonstimulated and CD3-stimulated T cell membrane (for 30 min) with soluble 2C11 mAb. The background of CTB – HRP conjugate against BSA blocking reagent was subtracted from each sample. Values represent the mean of triplicate wells F S.D. Panels B and C: Identification of GM1 distribution in the same fractions by Western blot and immunodot blot analyses, respectively. GM1 was revealed by chemiluminiscence using CTB – HRP conjugate. Panel D: The protein pattern in the same fractions as analyzed by SDS-PAGE under denaturing and reducing conditions in 4 – 15% gradient PhastGels, and developed by silver staining. The M lane indicates the molecular weight markers.
166
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
1 – 4 from both nonstimulated and CD3-stimulated cells (Fig. 2A –C). However, ELISA showed a 1000fold greater sensitivity than Western blot and immunodot blot analyses. Up to 5 104 cells could be used in these experiments, and the GM1 moieties would still be detected by this ELISA protocol. This would make possible the analysis of GM1 moieties in populations of T cells with a low frequency, such as memory or regulatory cells. The amount of GM1 was almost two times higher in fractions 1 – 4 from CD3-stimulated T cells than in nonstimulated ones. A 350-fold increase in the amount of GM1 in human CD4 T cell clones stimulated for 30 min with a CD3 mAb has been found using an ELISA protocol, in which GM1 was captured by CTB-coated plates and revealed with anti-CTB – HRP conjugate (Tuosto et al., 2001). The authors suggested that the increase was the result of intensive de novo synthesis of GM1 moieties. The difference between these results and ours could be explained by different T cell lines, ELISA protocol, and the type of CD3 mAb-ligand used in their experiments. Interesting enough, we found that GM1 was present in the detergent-soluble fractions (#5 – 8) from CD3stimulated T cells, although at lesser extent than in the detergent-insoluble fractions (#1 – 4). The increased amount of GM1 –protein complexes upon CD3 ligation may explain why GM1 moiety was found in higher sucrose density gradients. SDS-PAGE analysis in individual fractions from CD3-stimulated T cells showed a redistribution of several protein bands from the detergent-insoluble (#1– 3) to detergent-soluble fractions (#4 – 8) (Fig. 2D). Particularly, some low-molecularsize protein bands were decreased in the detergentinsoluble fractions, whereas some high-molecular-size protein bands were enriched in the detergent-soluble
fractions from CD3-stimulated cells (Fig. 2D). Montixi et al. demonstrated that the stimulation of T cells by specific ligands led to distribution of TCR molecules mostly in the detergent-insoluble fraction (Montixi et al., 1998, Xavier et al., 1998), which suggested that membrane proteins are differentially recruited to GM1 lipid rafts. It has been demonstrated that high-molecular species like CD45 extracellular phosphatase are excluded from lipid rafts during the formation of immunological synapse, whereas small molecular species are preferentially recruited to lipid rafts (Janes et al., 1999). Since we found that GM1 moieties are also present in the detergent-soluble fractions of CD3stimulated T cells, we questioned whether (1) GM1 could form complexes with 14.3d TCR, (2) CD3/ TCR cross-linking with a soluble 2C11 CD3-q mAb may favor the formation of such complexes, and (3) whether GM1 – TCR complexes could form preferentially in detergent-insoluble (GM1high) or in detergent-soluble (GM1low) compartments. ELISA showed that 14.3d TCR was almost equally distributed in the detergent-insoluble and -soluble fractions of nonstimulated TcH, whereas its amount was three times higher in the detergent-soluble fractions of CD3-stimulated cells (Fig. 3A). This was also confirmed by Western blot analysis (Fig. 3B). These data indicated that CD3/TCR ligation by a soluble ligand can lead to the enrichment of detergentsoluble compartment of T cell membrane in TCR molecules. It is unlikely that this was the result in TCR upmodulation in this compartment upon CD3 ligation, since it has been reported that TCR is usually dowmmodulated upon CD3 ligation. Our cytofluorimetric analysis showed that, apparently,
Fig. 3. Partitioning of TCR in the membrane lipid rafts from nonstimulated and CD3-stimulated T cells. Panel A: ELISA assay for measuring the amount of 14.3d TCR in the GM1high and GM1low pooled fractions in nonstimulated and CD3-stimulated T cell membrane (for 30 min) with soluble 2C11 mAb. The background of 6.5.2 mAb and Protein A – HRP conjugate against BSA blocking reagent was subtracted from each sample. Values represent the mean of triplicate wells F S.D. Panel B: Western blot analysis under denaturing and reducing conditions of GM1high and GM1low pooled fractions. The major band of 55 kDa represents the heavy chain of clonotypic 6.5.2 precipitating mAb revealed by Protein A – HRP conjugate used to develop the F23.1 mAb binding to Vh8 chain of 14.3d TCR. Panel C: Cytofluorimetric analysis of 14.3d TCR expression on nonstimulated and CD3-stimulated T cells for (30 min) with soluble 2C11 mAb, which were either non-incubated (left panel) or previously incubated with Monensin (right panel). The TCR expression identified with 6.5.2 clonotypic mAb was revealed with a secondary antibody (goat F(abV)2 anti-rat g1 Ab – FITC conjugate). The TCR expression was calculated in relation to the log fluorescence intensity of bound 6.5.2-secondary Ab – FITC complexes to the cells. The signal-to-noise background from the secondary (2.5 – 3%) was subtracted and eliminated from the histograms. The median of peak fluorescence in each sample was used to calculate the difference in fluorescence intensity. Panel D: ELISA capture assay for determining the amount of GM1 – TCR complexes in GM1high and GM1low pooled fractions from nonstimulated and CD3-stimulated T cell membrane (for 30 min) with soluble 2C11 mAb. The background of 6.5.2 mAb and protein A – HRP conjugate against CTB coated plates blocked with BSA was subtracted from each sample. Values represent the mean of triplicate wells F S.D.
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
167
168
S. Thomas et al. / Journal of Immunological Methods 275 (2003) 161–168
the TCR expression was increased by 30% after 30 min cross-linking of CD3/TCR complex with 2C11 mAb, as indicated by the fluorescence of 6.5.2 mAb – FITC conjugate bound to TCR. However, pretreatment of cells with Monensin, which inhibits protein translocation on the cell surface but does not affect protein internalization, showed a decrease by 10– 12% in TCR expression, which clearly indicated a TCR internalization. The apparent increase in TCR expression in CD3-stimulated cells in the absence of Monensin pretreatment was most probably the result of enhanced accessibility of 14.3d TCR to the 6.5.2 binding mAb as a result of allosteric changes mediated by the CD3/TCR cross-linking. The fact that both GM1 and TCR were found increased by ELISA in the detergent-soluble fractions of CD3-stimulated T cells raised the question of whether GM1 may co-exist with TCR in the same complex. Using a CTB capture ELISA, we found that TCR was physically associated with GM1 in both detergent-insoluble and -soluble fractions (Fig. 3D). However, the amount of GM1– TCR complexes was lower in the detergent-soluble fraction than in insoluble one, suggesting that the GM1 –TCR interactions are more stable in the detergent-insoluble fraction. As indicated by Western blot analysis carried out under denaturing and reducing conditions, these interactions were noncovalent, since GM1 was identified as a single, low-molecular band in each sucrose gradient fraction, both in nonstimulated and CD3-stimulated cells. A more detailed analysis will be required to find out whether GM1 may preferentially interact with other protein receptors on the T cell membrane. In conclusion, our results indicated that the shortterm ligation of CD3/TCR complexes on T cells with a soluble CD3 mAb-ligand led to accumulation of TCR in the detergent-soluble compartment of T cell membrane. The GM1 lipid rafts interacted physically with TCR in both detergent-insoluble and -soluble compartments, although more stable in the detergent-insoluble compartment. Subsequent to these findings, several important questions need to be addressed: (1) what is the structural basis of GM1 –TCR complex formation, (2) what is the rate of TCR internalization in the membrane detergent-insoluble compartment vs. the soluble one, (3) whether GM1 –TCR complex trafficking between the plasma membrane microdomains is possible, and under what conditions, and (4) what is the
nature of TCR signaling in the detergent-soluble compartment of T cell membrane. Acknowledgements This work received supported through grants from the National Institutes of Health (1RO1 DK61927 and 1R21 DK61326) to T-D.B. References Cottin, V., Doan, J.E.S., Riches, D.W.H., 2002. Restricted localization of the TNF receptor CD120a to lipid rafts: a novel role of the death domain. J. Immunol. 168, 4095. Harder, T., Kuhn, M., 2000. Selective accumulation of lipid raft associated membrane protein LAT in T-cell receptor signaling assemblies. J. Cell Biol. 151, 199. Harder, T., Simons, K., 1997. Caveolae, DIGs and the dynamics of sphingolipid – cholesterol microdomains. Curr. Opin. Cell Biol. 9, 534. Harder, T., Simons, K., 1999. Clusters of glycolipid and glycosylphospatidylinositol anchored proteins in lymphoid cells: accumulation of actin regulated by local tyrosine phosphorylation. Eur. J. Immunol. 29, 556. Hope, H.R., Pike, L.I., 1996. Phosphoiniositides and phosphoinositide utilizing enzymes in detergent insoluble lipid domains. Mol. Biol. Cell 7, 843. Janes, P.W., Ley, S.C., Magee, A.I., 1999. Aggregation of lipid rafts accompanies signaling via the T-cell antigen receptor. J. Cell Biol. 147, 447. Montixi, C., Langlet, C., Bernard, A.-M., Thimonier, J., Dubois, C., Wurbel, M.-A., Chauvin, J.P., Pierres, M., He, H.-T., 1998. Engagement of T-cell receptor triggers its recruitment to low density detergent insoluble membrane domains. EMBO J. 17, 5334. Ohtani, Y.Y., Zhou, X.-Y., Toyooka, K., Tai, X.-G., Park, C.S., Hamaoka, T., Abe, R., Miyake, K., Fujiwara, H., 2000. NonCD28 costimulatory molecules present in T-cell rafts induce Tcell co-stimulation by enhancing the association of TCR with rafts. J. Immunol. 164, 1251. Simons, K., Ikonen, E., 1997. Functional rafts in cell membranes. Nature 387, 569. Tuosto, L., Parolini, I., Schroder, S., Sargiacomo, M., Lanzavecchia, A., Viola, A., 2001. Organization of plasma membrane functional rafts upon T cell activation. Eur. J. Immunol. 31, 345. Weber, S., Trannecker, A., Olivery, F., Gerhard, W., Karjalainen, K., 1992. Specific low affinity recognition of major histocompatibility complex plus peptide by a soluble T-cell receptor. Nature 356, 793. Xavier, R., Brennan, T., Li, Q., McCormack, C., Seed, B., 1998. Membrane compartmentation is required for efficient T-cell activation. Immunity 8, 723. Zhang, W., Trible, R.P., Samelson, L.E., 1998. LAT palmitoylation: its essential role in membrane microdomain targeting and tyrosine phosphorylation during T-cell activation. Immunity 9, 239.