Cloned T Cells Internalize Peptide from Bound Complexes of Peptide and Purified Class I1 Major Histocompatibility Complex Antigen*

Antigen presentation to helper T cells involves the formation of a trimolecular complex consisting of a class I1 major histocompatibility complex (MHC) anti- gen combined with an antigenic peptide on the surface of an antigen-presenting cell and a T cell receptor (TCR) on the T cell. The fate of the MHC class 11, peptide, or TCR moieties of the ternary complex fol- lowing antigen presentation is unknown. Using radiolabeled complexes of affinity-purified murine MHC class I1 molecules and peptides corresponding to T cell epitopes of myelin basic protein (MBP), this report presents evidence that the binding of preformed relevant MHC class 11-peptide complexes to cloned T cells in vitro results in internalization of the peptide moiety. Neither the restricting MHC class I1 molecule nor the TCR moiety of the trimolecular complex was internalized by T cells. The specificity of peptide internaliza- tion was demonstrated using complexes of syngeneic MHC class I1 with an irrelevant MBP peptide analog and by cloned T cells restricted for a different epitope of the same MBP antigen. Furthermore, the peptide translocation mediated by MHC class I1 and TCR was demonstrated by antibody-blocking experiments using anti-class I1 and anti-TCR monoclonal antibodies. The peptide internalization by T cells was markedly re- duced when binding was performed at 4 "C as compared with 37 *C. In addition, a significant inhibition of peptide translocation was observed in the presence of a metabolic inhibitor PMSF, followed by 5 bed volumes of containing 1% n-octyl (OG) buffer. the IA were eluted with 20 mM phosphate buffer, pH 11, containing 0.1 M NaCl, 1% OG, 0.02% sodium azide, and 1 mM PMSF. Each was neutralized with 1 M acetic acid to a final concentration of 12 mM, and the MHC class I1 molecules were concentrated using an Amicon Centriprep-10 concentrator. Affinity-purified IAk and IA" molecules were characterized by 12% SDS-polyacrylamide gel electrophoresis. MBP(1-14)A' the sequence Ac-ASQARPSQRHGSKY, Synthesis of Peptides-The rat myelin basic protein peptide ana-MBP(1-4)A3A'A6 the sequence Ac-ASAARASQRHGSKY, and MBP(89-101)YBg peptide representing the sequence YFKNIVTPRTPPP, were synthesized by the standard solid phase methodology using side chain-protected Fmoc (N-(g-flu- oreny1)methoxycarbonyl) amino acids and an Applied Biosystems automated peptide synthesizer. The deprotected, crude car- boxyl-terminal peptide amides were purified by reverse-phase high pressure liquid chromatography, and the homogeneity and identity of the purified peptides were confirmed by mass spectroscopic analysis. incubated The peptides separated lZ5I Specific activities of various MBP peptides cpm/pg.

Cloned T Cells Internalize Peptide from Bound Complexes of Peptide and Purified Class I1 Major Histocompatibility Complex Antigen* (Received for publication, December 11, 1992, and in revised form, March 15, 1993) Bishwajit Nag$, Shrikant V. Deshpande, Somesh D. Sharma, and Brian R. Clarke From Anergen Inc., Redwood City, California 94063 Antigen presentation to helper T cells involves the formation of a trimolecular complex consisting of a class I1 major histocompatibility complex (MHC) antigen combined with an antigenic peptide on the surface of an antigen-presenting cell and a T cell receptor (TCR) on the T cell. The fate of the MHC class 11, peptide, or TCR moieties of the ternary complex following antigen presentation is unknown. Using radiolabeled complexes of affinity-purified murine MHC class I1 molecules and peptides corresponding to T cell epitopes of myelin basic protein (MBP), this report presents evidence that the binding of preformed relevant MHC class 11-peptide complexes to cloned T cells in vitro results in internalization of the peptide moiety. Neither the restricting MHC class I1 molecule nor the TCR moiety of the trimolecular complex was internalized by T cells. The specificity of peptide internalization was demonstrated using complexes of syngeneic MHC class I1 with an irrelevant MBP peptide analog and by cloned T cells restricted for a different epitope of the same MBP antigen. Furthermore, the peptide translocation mediated by MHC class I1 and TCR was demonstrated by antibody-blocking experiments using anti-class I1 and anti-TCR monoclonal antibodies. The peptide internalization by T cells was markedly reduced when binding was performed at 4 "C as compared with 37 *C. In addition, a significant inhibition of peptide translocation was observed in the presence of a metabolic inhibitor (sodium azide) but not in the presence of cytochalasin B. These results together demonstrate that the in vitro interaction of soluble MHC 11-peptide complexes with cloned T cells is an active process associated with uptake of the antigenic peptide.
Antigen presentation to T helper cells is a complex event that requires the interaction of an antigen-specific TCR,' an appropriate molecule of the MHC, and a processed fragment of antigen (1,2). In this series of events, the exogenous antigen is first internalized by APC and proteolyzed intracellularly in the lysosomal compartment (3, 4). In the endosomal compartment, the resultant peptide fragments compete for an * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. MHC I1 peptide binding site, which because of sequence homology is inferred to be structurally and functionally similar to the peptide binding cleft of the MHC I molecule (5, 6).
The MHC 11-peptide complexes thus formed are then expressed on the surface of the antigen-presenting cell where they are recognized by TCRs on the surface of T cells (7)(8)(9).
Although the trimolecular model of antigen presentation (MHC class 11-antigenic peptide-TCR) is widely accepted, the fate of components of the ternary complex upon separation of apposed APC and T cell is still unknown. This report presents evidence that the in vitro binding of complexes, containing affinity-purified murine class I1 MHC and synthetic antigenic peptide, to cloned murine T cells was associated with internalization of only the peptide component of the ternary complex.

MATERIALS AND METHODS
Murine T cell clones (AJ1.2 and 4R3.9), restricted for IAk-MBP(l-14) complexes, were obtained from the laboratory of Patricia Jones, Stanford University (Stanford, CA). The HS-1 murine T cell clone, which recognizes IA"-MBP(89-101), was obtained from the laboratory of S. Sriram, University of Vermont (Burlington, VT). The hybridoma cell line producing 10-2.16 monoclonal antibody against murine IAk and IAs was obtained from American Type Culture Collection (Rockville, MD). Affinity-purified a/@-specific anti-TCR (H57-597) hamster monoclonal antibody and isotype match antibodies were purchased from PharMingen (San Diego, CA). Radioactive sulfur labeling reagent and lZ5I were purchased from Amersham Corp.
Affinity Purification of Murine IAk and ZA"--IAk and IA" were purified from Nonidet P-40 extracts of membrane, prepared from CH27 cells and SJL/J mouse spleen cells, respectively, using an affinity column prepared by coupling monoclonal antibody, 10-2.16 (specific for IAk and IA"), with Sepharose 4B beads by the standard cyanogen bromide-coupling method as described earlier (10). Briefly, a high speed (100,000 X g) membrane fraction was detergent-extracted in a buffer containing 10 mM Tris-HC1, pH 8.3,0.5% Nonidet P-40, 0.1 M NaCI, 5 mM EDTA, 0.02% sodium azide, and 1 mM PMSF. The lysate was recycled over the pre-equilibrated antibody column at 4 "C for 16 h. The column was washed with 10 bed volumes of deoxycholate buffer containing 10 mM Tris-HC1, pH 8.3, 0.5% deoxycholate, 0.1 M NaCl, 5 mM EDTA, 0.02% sodium azide, and 1 mM PMSF, followed by 5 bed volumes of PBS containing 1% n-octyl 8-D-glucopyranoside (OG) buffer. Finally, the IA molecules were eluted with 20 mM phosphate buffer, pH 11, containing 0.1 M NaCl, 1% OG, 0.02% sodium azide, and 1 mM PMSF. Each fraction was neutralized with 1 M acetic acid to a final concentration of 12 mM, and the MHC class I1 molecules were concentrated using an Amicon Centriprep-10 concentrator. Affinity-purified IAk and IA" molecules were characterized by 12% SDS-polyacrylamide gel electrophoresis.
Radiolabeling of Peptides and IA Molecules-Radiolabeling of peptides was achieved by the standard chloramine-T labeling procedure (11). Typically, 2.5 mg of peptide in a 500-ml volume was incubated with 2 mCi of Na[1261] in 0.1 M sodium phosphate buffer, pH 7.7. The labeled peptides were separated from free lZ5I by Sephadex G-10 gel filtration chromatography. Specific activities of various MBP peptides ranged from 0.5 to 2.7 X lo6 cpm/pg. Radiolabeling of class I1 molecules was performed either by lZ5I or by sulfur labeling reagent. Labeling of IAk with Iz5I was achieved using Pierce IODO-BEADS and the manufacturer's recommended procedure. Preloaded beads were prepared by washing with 50 mM phosphate buffer, pH 7.5 (1 ml of buffer for two beads), and dried on Whatman paper. Six beads were added to 50 pg of 0.1 mg/ml IAk in 100 mM phosphate buffer, pH 7.5, containing 1 mCi of Iz5I. The reaction mixture was incubated at room temperature for 10 min. Labeled IAk was separated from the beads, and residual free Iz5I was removed by extensive dialysis against PBS at 4 "C. The specific activity of 1251-labeled IAk was in a range of 0.4-0.5 X IO6 cpmlpg. 35S labeling of IAk with sulfur labeling reagent was performed according to the manufacturer's recommended procedure (Amersham). Briefly, 1 ml of [35S]sulfur labeling reagent (1 mCi) was evaporated to dryness under a gentle stream of argon. One mg of IAk at a concentration of 1 mg/ml in 0.1 M borate buffer, pH 8.5, containing 0.5% OG was added to the dry sulfur labeling reagent. The reaction mixture was mixed and incubated at room temperature for 30 min, and the labeling was stopped by adding 100 pl of 0.1 M borate buffer, pH 8.5, containing 0.2 M glycine. The excess reagent was removed by dialysis of the labeled IAk against PBS containing 0.5% OG. The specific activity of IAk with 35S label was 1.98 X lo6 cpm/pg. Peptide Binding Assay-Peptide binding to murine class I1 molecules was analyzed by silica gel thin layer chromatography (TLC) as described earlier (12, 13). IAk or IA" at a concentration of 200-500 pg/ml was incubated with a 50-fold molar excess of radiolabeled peptide at 37 "C for 48 h. The excess unbound peptide was removed by extensive dialysis against PBS containing 0.1% OG detergent at 4 'C for 36 h. One pl of complex was applied in triplicate onto a 5cm plastic-supported silica gel TLC plate (from EM sciences) and run in a solvent system of 5% ammonium acetate in 50% methanol. The plate was dried, and the distribution of radioactivity was estimated at R p 0-0.2 for calculating the percent of IAk or IA" occupied with labeled peptide. The percent of class I1 occupied with labeled peptide ranged from 25 to 35.
T Cell Binding Assay-Purified complexes of either radiolabeled peptide and nonradiolabeled IA molecules or 1251-MBP peptide and %-IAk in a total volume of 1 ml were incubated with 0.5-2.0 X lo6 resting T cells in 15-ml polypropylene tubes precoated with 1 mg/ml bovine serum albumin solution at 37 "C in a COz incubator. For antibody blocking experiments, cells were co-incubated with radiolabeled complexes in the presence of varying amounts of affinitypurified monoclonal antibodies. Purified isotype-matched standard antibodies were used as controls for these experiments. At the end of the incubation period, chilled PBS was added to each tube, the cells were resuspended gently and centrifuged at 2,000 X g, and the supernatant was carefully removed. The washing procedure was repeated three times, and the cell pellet was counted in either a y or p counter for or 36S, respectively.
Fluorescence-Actiuated Cell-scanning (FACS) Analysis-Flow cytometric analysis of 4R3.9 cloned T cells treated with affinity-purified anti-TCR(q3) monoclonal antibody was performed at various time intervals following complex incubation. 4R3.9 T cells (0.5 X lo6/ sample) were washed and incubated with 5 pg of preformed nonradiolabeled complex. For the control experiment, cells were incubated under identical conditions in the absence of complex. At varied time intervals, the cells were washed three times in isotonic saline buffer (FACS washing buffer), incubated for 2 h with 20 pg of anti-TCR(a@) antibody at 4 "C, washed, and kept on ice. At the end of 5 h, cells were stained with fluorescein isothiocyanate-conjugated goat antihamster secondary antibody and analyzed on a FACScan (Becton Dickinson). Unstained cells and cells treated with second antibody alone, anti-TCR alone, or isotype-matched antibody at 0 and 5 h were used as controls.

RESULTS
Specific association of MHC class I1 and antigenic peptide with MBP-specific cloned T cells was demonstrated using complexes of radiolabeled peptide and affinity-purified murine MHC class I1 molecules. Since T cells are very sensitive to detergent, all cell binding experiments used complexes that were extensively dialyzed against detergent-free PBS. The extent of aggregation in the final preparation was measured by high speed centrifugation and by Sephadex G-200 gel filtration chromatography of complexes of lZ5I-IAk and unlabeled MBP(1-14)A4 peptide in a separate experiment and was found to be approximately 60%. Complexes of IAk with lZ5I MBP( 1-14)A4 were prepared, and the effect of complex concentration on T cell binding was measured (Fig. 1). Incubation of 4R3.9 T cells with an increasing concentration of complexes showed a linear increase in the number of molecules associated per T cell. No significant decrease in T cell viability was observed up to a complex concentration of 5 pg/ml. However, incubation of T cells with a complex concentration greater than 5 pg/ml showed a decrease in T cell viability. At a complex concentration of 25 pg/ml the viability was reduced to 70%. Based on this observation, a complex concentration of 5 pg/ml or less was used in all subsequent experiments.
To demonstrate that the recognition of complexes by cloned T cells is peptide-specific, two acetylated N-terminal MBP The results presented in Fig. 3, 4 and B, demonstrate that the number of bound molecules/T cell was markedly inhibited in the presence of either anti-class I1 or anti-(aP)TCR monoclonal antibodies. In control experiments, equivalent amounts of isotype-matched antibodies co-incubated under identical conditions did not show any significant inhibition of radiolabeled peptide uptake by 4R3.9 T cells.
In all these experiments, the number of labeled peptide molecules associated with single T cells was substantially higher than the number of TCRs (20,000-50,000) reported on single T cells (15,16). To demonstrate that the increased number of labeled peptide molecules/T cell was not due to dissociation of the prebound peptide or release of free lZ5I from the tyrosine residue of the labeled peptide moiety at 37 "C, the stability of the purified complexes was studied. IAk-MBP( 1-14)A4 complexes containing radiolabeled peptide were incubated in the absence of T cells in RPMI-1640 medium containing 10% serum. At various time intervals, complexes were analyzed by silica gel TLC, and the dissociation of complexes was measured. As shown in Fig. 4B, complexes of IAk and radiolabeled MBP(1-14)A4 were fully stable for the 6-h incubation period without any significant dissociation. The rate constant measurements (kd) indicate that the complexes were 100% stable during the entire incubation period. In addition, complexes at time 0 and at the end of the 6-h incubation period were subjected to silica gel TLC to ensure there was no detectable release of bound peptide or free ' ' ' I in the medium (Fig. 4A).
Since the number of complex molecules associated with a single T cell (400,000-600,000) did not correspond to the reported number of TCRs (20,000-50,000) on the T cell surface, further experiments were designed to permit separate calculation of the average number of IAk heterodimer and peptide moieties associated with single T cells following the complex incubation. This was achieved using preformed complexes of known specific activity containing either '251-labeled IAk and unlabeled MBP(1-14)A4 peptide or complexes containing unlabeled IAk and ' ' ' I labeled MBP(1-14)A4 peptide.  The results of such an experiment are shown in Fig. 5A. Approximately 20,000-30,000 molecules of the IAk moiety of the IAk-peptide complex were associated with single AJ1.2 T cells compared with 400,000-500,000 molecules of peptide moiety/?' cell. This result was again confirmed in a separate experiment with 4R3.9 cloned T cells, specific for the same N-terminal MBP peptide fragment. In this study, T cells were incubated with a double labeled complex prepared from "Slabeled IAk and '251-labeled MBP(1-14)A4 peptide. The calculated number of cell-associated IAk or peptide moiety of the complex bound per T cell is shown in Fig. 5B. The results obtained from the experiment using double labeled complexes were similar to those obtained with single labeled complexes.
T o examine the fate of TCRs on the T cell surface following class 11-peptide complex treatment, cells were monitored for TCRs by FACS analysis using anti-TCR(cuB) monoclonal antibody followed by fluorescein isothiocyanate-conjugated secondary antibody. The flow cytometry results presented in Fig. 6 revealed no decrease in cell surface TCRs before and after treatment with relevant complex, suggesting no downregulation of T cell receptors during binding of complex and internalization of peptide molecules.
Finally, to address the question whether the peptide uptake by T cells from the trimolecular complex is an energy-dependent phenomenon or is mediated by microfilaments, the effect of sodium azide and cytochalasin B on the cell association of the IAk and peptide moieties of the doubly labeled complex was examined (Fig. 7). Sodium azide is known to inhibit the electron transport system and therefore can be used as an inhibitor of energy-dependent cellular uptake mechanisms. Cytochalasin B, however, is a fungal product widely used as an inhibitor of cell function via its binding to the actin moiety of actinomysin and inhibition of microfilament contraction (17). As shown in Fig. 7A, cloned 4R3.9 T cells incubated with double labeled complexes in medium containing 0.2% sodium azide showed a significant decrease in the uptake of peptide moiety, whereas 4R3.9 T cells incubated with up to 4 mM cytochalasin B did not show any significant decrease in peptide uptake (Fig. 7B). Similar results were obtained with cloned AJ1.2 T cells incubated with double labeled complex in the presence of each inhibitor. The viability of T cells was measured a t various time periods, and cells were found to be fully viable in all these experiments.

Antigen presentation to T cells involves the interaction of
MHC-peptide complexes on the surface of APC with TCRs on T cells (1, 2). In uiuo, such presentation leads to the activation of T cells (18). In vitro, however, the recognition of purified MHC class 11-peptide complexes by T cells in the absence of APCs and costimulatory signals results in induction of nonresponsiveness (19). In both cases the fate of the antigenic peptide, MHC class 11, and TCR molecules following presentation is not defined. In this report, using radiolabeled complexes of MHC class I1 and antigenic peptides we observed internalization of the peptide molecules from the trimolecular complexes. A direct association of labeled complexes (containing radiolabeled peptide, radiolabeled class 11, or both) with T cells was demonstrated using murine cloned T cells. AJ1.2 and 4R3. 9 T cells have been shown to recognize and respond to peptide containing the first nine amino acids of MBP in context to murine IAk (20). Both clones also recognize and respond to MBP (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) in association with IAk. Similarly, the HS-1 T cell clone restricted for IA" and MBP(90-101) has MHC Class II, Antigenic Peptide, and TCR Interaction been characterized (21). Complexes were prepared and purified, and the complete removal of unbound peptide was confirmed by analyzing complexes by silica gel TLC and autoradiography. Incubation of T cells with increasing concentra- -COMPLEX 0 hour tions of purified complexes resulted in an increase in cell association of these molecules as shown in Fig. 1. Incubation of cloned T cells with higher concentrations of complexes was associated with decreased cell viability. The possibility that mature T cells stimulated with very high levels of TCR occupancy in the absence of costimulatory signals first become functionally anergic and then die, whereas cells stimulated with lower levels of TCR occupancy in the absence of costimulation manifest only the first phenotype, has been speculated upon earlier (22). Recent results from our laboratory indicate that cells incubated with 50 pg/ml or higher concentration of complexes for 15 h lead to complete cell death.' Based on the specific activity of the peptide, the number of molecules associated with a single T cell was calculated and found to be approximately 600,000. The number of TCR molecules on a single T cell was reported to be 20,000-30,000 (15,16). Thus, the observed number of molecules associated per T cell in our experiments was significantly higher than the reported number of TCRs on T cells. Such an increased number of cell-associated molecules may be the result of one or more of the following possibilities: (i) internalization of the peptide moiety; (ii) rapid dissociation of labeled peptide from the MHC-peptide complexes; or (iii) release of free lZ5I from the tyrosine residue. The dissociation kinetics of IAk-1251-(MBP(1-14)A4 complex was performed at 37 "C and analyzed by TLC at times of 0-6 h. Since no detectable labeled peptide or free 1251 was observed for up to 6 h of incubation of the complex as shown in Fig. 4 A , the increased number of T cellassociated radiolabeled molecules is not due to either dissociation of peptide from the MHC-peptide complexes or release of free ' ' ' I in medium containing 10% serum.
The specificity of peptide uptake is demonstrated by three different approaches.  Fig. 2.4. Cells washed and exposed to pH 3.0 for 10 min at 0 "C (conditions known to release Fab' fragments bound to cell surface components with no decrease in cell viability (23)) resulted in no significant difference in the number of cellassociated molecules. (ii) That the binding and internalization of peptide moiety is TCR-restricted has been demonstrated by using a different T cell clone (HS-1) restricted for IA" and MBP(89-101). As shown in Fig. 2C, the incubation of IAk-MBP(1-14)A4 complexes with HS-1 T cell clone did not show significant uptake of peptide molecules. However, complexes of IA" with radiolabeled MBP(89-101) did show uptake of the MBP(89-101) peptide in a positive control experiment. (iii) Finally, the specificity of the binding was demonstrated by antibody blocking experiments using either specific anti-IAk (10-2.16) or &-specific anti-TCR (H57-597) monoclonal antibodies (24,25) as shown in Fig. 3. These data suggest that the internalization of peptide molecules is not a nonspecific phenomenon.
More specifically to demonstrate the internalization of only the peptide moiety and not the class I1 moiety from the ternary complex, double labeled complexes were prepared using 35S-IAk and "'I-MBP( 1-14)A4 peptide where both components can be monitored simultaneously in the same experiment. The number of IAk molecules and peptide molecules associated with a single 4R3.9 T cell was calculated and found to be approximately 23,000 and 400,000, respectively, which correlate well with the data obtained with complexes where IAk and MBP( 1-14)A4 were separately labeled with 12' 1 (Fig.  5, A and B ) . FACS analysis of T cells with anti-TCR monoclonal antibody at various time intervals following complex treatment did not show any down-regulation of TCRs (Fig.  6).
The effect of sodium azide and cytochalasin B was examined to determine if the uptake of peptide by T cells is due to an active process and not due to a microfilament-mediated phenomenon. The internalization of peptide molecules from the ternary complexes is an active process as the uptake was significantly reduced at 4 "C. Similarly, the number of cellassociated radiolabeled peptides was markedly decreased when complexes were incubated with T cells in the presence of varied concentrations of sodium azide ranging from 0.01 to 0.2%. Sodium azide is known to inhibit the electron transport system and therefore can be used as an inhibitor of energydependent cellular uptake mechanisms. In contrast, incubation of T cells with relevant complex in the presence of cytochalasin B up to 4 mM did not show any significant inhibition of cell-associated peptide molecules/T cell. Cytochalasin B is a fungal product widely used as an inhibitor of cell function via its binding to the actin moiety of actinomycin and inhibition of microfilament contraction (17). Cytochalasin B at concentrations between 0.02 and 0.2 pg/ml has been shown to be slightly stimulatory to lymph node cells, whereas at higher concentrations of 5-10 pg/ml or above, it shows inhibition of lymphocyte responses (26). Incubation of 4R3.9 T cells with up to 4 mM cytochalasin B did not show any significant decrease in peptide uptake by these T cells.
Estimates of k d for dissociation of MHC-peptide complex from TCR have been reported both for recombinant soluble MHC 1-peptide complexes (27) and recombinant soluble MHC 11-peptide complexes (28). In the latter study, the k d value ranged from 4 x to 6 X lo-' for soluble, monomeric lEk-peptide complex bound to TCR on T cells or T hybridomas. These data would suggest that very little MHC 11-peptide complex will bind to T cells at the concentrations of complexes used in our study. However, the present study uses purified native MHC I1 with the transmembrane domain intact. Therefore, one would not expect an appreciable fraction of the complex to exist in a soluble, monomeric form in the absence of detergent. In fact, we have shown that the native MHC I1 used in our studies is 60-90% aggregated. Similar binding of purified MHC 11-peptide complexes to several T cell clones was observed in other studies from our laboratory (10,29). We propose that the aggregated material is multivalent and binds with a higher apparent affinity to multiple TCR on the T cell surface, and therefore lower concentrations of relevant complex effec,tively bind to the cognate T cell. Another possibility is that the MHC 11-peptide complex binds to the TCR with high affinity, but a rapid translocation of the peptide moiety into the cell results in a rapid release of the empty MHC I1 molecule from the ternary complex, resulting in a lower apparent kd.
The results presented in this study suggest that the in vitro interaction of purified MHC 11-peptide complex with T cell receptor on cloned T helper cells leads to internalization of only the peptide moiety, most likely by an energy-dependent translocation process. Although the function served by the proposed internalization of the peptide moiety of the cognate complex is not clear at present, one may speculate that it could facilitate disengagement of apposed antigen-presenting cell and T cell after interaction and/or that it may be involved in immunoregulation of T cell function. A model for T cell activation has been proposed in which the last step is a decline in the enhanced adhesiveness between APC and T cell prior to detachment (27). The proposed disengagement would be hindered by multiple attachments remaining between the apposed cells as a result of intercellular MHC-peptide-TCR ternary molecular bridges stabilized by CD4 or CD8 molecules that are known to cross-link MHC to the CD3-TCR complex.
Internalization of the peptide moiety may serve to destabilize these bridges and facilitate separation.