Role of the major histocompatibility complex class II transmembrane region in antigen presentation and intracellular trafficking

Abstract

While a sorting signal in the cytoplasmic tail of the major histocompatibility complex (MHC) class II molecules is known to influence their endocytic transport, potential effects of the transmembrane (TM) domain of the MHC class II molecules on endocytic transport remain unclear. We have examined the role of the TM domain by comparing antigen-presenting functions of the wildtype (WT) I-A^b and mutant (MT) I-A^b molecule substituted in the β-chain TM with α chain TM. A20 cells transfected with WT I-A^b were able to present antigen (hen egg lysozyme) better to some hybridomas, while those transfected with MT I-A^b consistently outperformed WT for other hybridomas recognizing different epitopes. This difference in antigen processing and presentation is not caused by the differences in H-2M (DM) requirement or association with Ii. The time required for processing of specific epitopes appears to be different, suggesting sequential involvement of various endocytic compartments in the antigen processing. Although both WT and MT molecules were found in the early endocytic (transferrin receptor-rich) compartments, MT molecules accumulated in these compartments in higher quantities for longer time periods. Similarly, the MT molecule is retained for a longer time period than WT in late endocytic (LAMP-1 associated) compartments. Together, our data indicate an important role of the TM domain of the MHC class II molecules in the intracellular trafficking and, consequently, antigen processing and presentation.

Introduction

Presentation of specific peptides by antigen-presenting cells (APC) to T lymphocytes in the context of major histocompatibility complex (MHC) class II molecules is a central step in the initiation of an immune response. Interaction of antigenic peptide/MHC class II complex on the APC with the antigen-specific T-cell receptor (TCR) on the surface of the T cell result in complex events, such as T-cell activation, proliferation, and cytokine production.^1–5 Antigenic peptides following endocytosis of an antigen from the cell surface are loaded onto the MHC class II molecules in various compartments of the endocytic pathway and transported back to the surface for presentation to T cells. Many investigations conducted during the last decade have elucidated many steps of antigen processing and presentation by class II molecules, including but not limited to, locations of the antigen loading compartments, changes in intracellular signalling patterns upon MHC class II engagement, and the role of MHC class II associated molecules in antigen processing. Structural and functional analyses of various mutants of MHC class II have been extensively investigated to assign the role of different segments of the molecule in signal transduction as well as antigen processing/presentation.² In particular, the cytoplasmic (CY) domain of the MHC class II molecule attracted most attention in functional studies. Numerous researchers demonstrated that deletion, truncation or substitutions of the CY domain of the MHC class II molecules results in severe defects in intracellular signalling and ultimately antigen processing and presentation.^6–8 Truncation of the CY domain abrogates the antigen-presenting ability for certain hybridomas.⁷ The modification of antigen presentation function may reflect the deletion of di-leucine sorting signal located in the CY domain of the MHC class II β chain, related to the similar signal in the invariant chain CY domain.⁹ In contrast, very few studies have focused on the correlation between the transmembrane (ΤΜ) domain and the function of class II molecules. It has been previously shown that MHC class II molecules with TM domains of α and β chains substituted with those of the interleukin-2 (IL-2) receptor were incapable of presenting certain (‘conformational’) epitopes in vitro¹⁰ suggesting a role of TM in antigen presentation. In addition, both TM and CY regions are important for B-cell activation, and yet the TM-mediated signal remains unknown, unlike the CY region which is associated with cAMP induction.^6,11 Involvement of the TM region in trafficking class II molecules to endocytic compartments has also been implicated based on the indirect evidence that the class II molecules are transported even in the absence of associated invariant (Ii) chain carrying sorting sequences.¹² However, these studies do not provide information on the antigen-presenting functions of individual class II polypeptides. Thus, the role of the TM domain of MHC class II molecules in antigen processing and presentation remains unclear.

We have previously observed that membrane-bound immunoglobulin M (IgM), carrying the I-A_α TM substitution were preferentially able to deliver antigen complexes to certain preferred endocytic compartments.¹³ This result suggests that the TM domain of I-A molecules is capable of directing the molecules to selected endocytic compartments. We have here analysed and compared antigen processing and presentation in vitro by WT and MT I-A^b molecules carrying homodimeric A_αTM; the TM domain of its β chain substituted with α chain. While transfectants expressing both wildtype and mutant molecules are competent to present hen egg lysozyme (HEL) to I-A^b restricted hybridomas, substantial differences in their functional capacity are observed. WT I-A^b molecules are able to present antigen significantly better than MT I-A^b to the hybridomas recognizing epitopes generated in the late endocytic compartments, and vice versa to the hybridomas recognizing epitopes generated in the early compartment. Further investigation of the transport of I-A^b WT and MT molecules by confocal microscopy revealed that MT molecules appear to be trapped in early endocytic as well as late lysosomal compartments as compared to WT counterparts. These results strongly suggest that TM of MHC class II molecules may play an important role in the transport of the molecules in the endocytic compartments, ultimately affecting the outcome of antigen processing/presentation to T cells.

Materials and methods

Cell lines

HEL-specific, I-A^b restricted hybridomas, A 2.5 (HEL_83–93), A69.5 (HEL_76–85) and PCH4.1 (HEL_51–60), were previously generated in our laboratory and used extensively.^14,15 BOH4 hybridoma specific for HEL_74–90¹⁶ was a gift from Dr Eli Sercarz (La Jolla Rearch Institute of Immunology, San Diego, CA). A20 murine B-cell lymphoma (American Type Culture Collection, Manassas, VA) was used to express I-A^b WT and MT molecules. 3A5.¹⁷ H-2M (DM)-deficient A20 cell line was a kind gift from Dr John Monaco (University of Cincinnati, Cincinnati, OH). Schematic representations of the WT and MT I-A^b molecules are presented in Fig. 1. The generation of cDNA clones encoding the wildtype α, β and mutant TMβ chains was previously described.⁶ A20 cells were transfected by electroporation with 10 µg of α chain coding plasmid and 10 µg of either WT or MT β chain. pcDNA3 plasmid (1 µg) containing the selectable marker gene for neo^R was cotransfected with the above-mentioned combination of plasmids. After the initial selection in G418, transfectants were further subcloned by limiting dilution and maintained in Dulbecco's modified Eagle's minimal essential medium containing 800 µg/ml of G418. Surface expression of I-A^b was evaluated by fluorescence-activated cell sorting (FACS) analysis with fluoroscein isothiocyanate (FITC)-conjugated anti-I-A_β^b antibody.

Figure 1.

Schematic representation of the structure of the wild type (WT) and mutant (MT) I-A^b molecules and the level of surface expression in the transfectants. A20 cells were transfected by electroporation with the wild type α and β chains of I-A^b molecule or with the wild type α and mutant β chains. After selection by limiting dilution, surface expression level of the molecules on the transfectants was analysed by FACS using FITC-conjugated anti-mouse I-A^b antibody.

Antibodies

Antibodies utilized in this study for FACS analysis, surface cross-linking and confocal microscopy were as follows: FITC conjugated goat anti-mouse IgG (H + l) (Southern Biothechnology Associates, Birmingham, AL), mouse anti-mouse I-A_β^b, rat anti-mouse anti-B7.1 and B7.2 (Pharmingen, San Diego, CA), and phycoerythrin (PE)-conjugated rat anti-mouse transferrin receptor (Caltag Laboratories, Burlingame, CA), rat anti-mouse LAMP1 (ID4B, DSHB, University of Iowa), AlexaFluor™ 568 conjugated goat anti-rat IgG (H + l) (Molecular Probes, Eugene, OR), and FITC-conjugated goat anti-rat immunoglobulin (BioSource International, Camarillo, CA).

Antigen presentation to T-cell hybridomas

Presentation of HEL to HEL-specific, I-A^b restricted T cell hybridomas was evaluated by the level of IL-2 production in the culture supernatants. Briefly, hybridoma cells (5 × 10⁴ per well) were cocultured in 96-well culture plates with 1 × 10⁵ of the parental cell line (A20) and transfectants expressing WT or MT I-A^b molecules in the presence of various concentrations of HEL or phosphate-buffered saline (PBS) for 24 hr. For experiments determining the kinetics of antigen processing and presentation, transfectants (1 × 10⁵ cells/ml) were incubated with HEL (70 µm) for varying time period and then fixed with 0·1% paraformaldehyde (Sigma Chemical, St. Louis, MO) according to previously described methods.¹⁸ The fixation was terminated by addition of an excess amount of cold 0·5% glycyl glycine in PBS. The cells were then washed three times with PBS containing 10% fetal calf serum (FCS), and further incubated for 1 hr at 37° in RPMI-1640 culture medium. The fixed cells (2 × 10⁵ cells/well) were then used to stimulate T hybridomas. The production of IL-2 in the culture supernatant was determined by its ability to support the proliferation of an IL-2-dependent cell line CTLL-2. Culture supernatants (100 µl per well) were removed and added to 7·5 × 10³ CTLL-2 cells in 100 µl of culture medium. After 24 hr, cultures were pulsed with 1 µCi of ³H-TdR per well for 14–18 hr. Cells were then harvested for measurement of incorporated ³H-TdR labels. Results are expressed as Δc.p.m. ± the standard error of the mean for triplicate cultures.

Western blot

A20 cells (1 × 10⁶) expressing either wild-type or mutant I-A^b molecules were harvested and washed three times with PBS. Cold lysis buffer (1 ml), containing 1 mm phenylmethylsulphonyl fluoride, 10 µg/ml aprotinin and 10 µg/ml leupeptin, was added to the cell pellet. Cell debris was removed by centrifugation and supernatant was preabsorbed with 20 µl of 20% protein G slurry (Amersham Pharmacia Biotech). Either 1 µg mouse antibody against I-A^b (AF6-120·1, Pharmingen) or rat antibody against invariant chain (Pharmingen), together with protein G slurry, was added and the mixture was incubated at 4° for 1 hr. After three washes, immunoprecipitated samples were applied to 12% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE) and transferred to nitrocellulose membrane. The membrane was blotted with either anti-I-A^b or anti-invariant chain antibodies and the specific bands were visualized with either secondary anti-mouse or anti-rat IgG–horseradish peroxidase (HRP) in the presence of ECL™ (Amersham Pharmacia Biotech).

Confocal microscopy

Microscopic analysis of intracellular trafficking patterns of WT and MT I-A^b molecules was performed with Olympus IX-70 inverted microscope and analysed with SoftWorx software. For analysis of MHC molecules trafficking into early transferrin receptor-rich endocytic compartments, transfectants were labelled with PE-conjugated anti-transferrin and FITC-conjugated anti-I-A^b antibodies for 30 min at 4°, washed 2× with cold PBS, and then shifted to 37° for 15, 30 or 60 min After incubation, cells were fixed with 3·75% formaldehyde in PBS, washed, transferred into 70° ethanol, dropped on poly L-lysine coated slides, and mounted in antifade media (VectaShield, Vector). For every time point of each transfectant, 10 random fields of vision were analyzed. Trafficking of WT and MT molecules to late endocytic compartments (LAMP1⁺) was similarly analysed. After fixation, cells were permeabilized with 0·05% saponin in PBS and then incubated with FITC-conjugated anti-I-A^b antibody and rat anti-mouse LAMP1 plus Alexa Fluor™-conjugated goat anti-rat antibody.

Results

Peptide presentation efficacies of WT and MT I-A^b transfectants are similar

To study the potential role of the TM region of an I-A molecule, we have generated A20 B-cell tumour lines expressing the wildtype (WT) and mutant (MT) I-A^b molecules. The MT molecules consist of A_β^b, whose TM is substituted with that of A_α^b together with the wildtype A_α^b chain (Fig. 1a). The transfectants expressing an equivalent level of WT and MT I-A^b molecules on the cell surface as well as in the cytoplasm (Fig. 1b) were used to investigate the efficiency of processing/presentation of a protein antigen, HEL. In addition, these transfectants displayed similar levels of endogenous I-A^d molecules on the surface (data not shown). In order to assess the ability of these transfectants to present known epitope peptides without processing, two representative T-cell hybridomas were tested with respective epitope peptides (Fig. 2). When specific peptides (e.g. HEL_81–93 for A2.5 and HEL_46–60 for PCH4.1) rather than whole protein antigen were used, no substantial differences in their ability to stimulate these hybridomas were detected between the transfectants. These results indicate that there is no deficiency in the epitope presentation function of the MT I-A^b molecules. These results also suggest that the potential interisotypic pairing of I-A molecules¹⁹ is not significantly increased because of the alteration in the TM domain. If a significant population of I-A molecules are interisotypic hybrids, a corresponding difference in the I-A^b peptide presentation between WT and MT is expected.

Figure 2.

Peptide presentation by WT and MT transfectants. To compare the ability to present peptides to T cells by WT and MT molecules, the parent A20 (PT), WT and MT transfectants (1 × 10⁵) were cultured for 24 hr with two representative T-cell hybridomas, A2.5 and PCH4.1 in the presence of various epitope peptides (HEL_81–93 for A2.5 and HEL_46–60 for PCH4.1, respectively). Culture supernatants were assessed for the presence of IL-2 by the level of ³H-TdR uptake using 1 × 10⁴ CTLL-2 cells.

Efficacy of WT and MT I-A^b transfectants to process/present native antigen differs depending on the epitopes

As can be seen in the Fig. 3, both WT and MT I-A^b-expressing cells were capable of presenting antigen to I-A^b restricted, HEL-specific T-cell hybridomas. However, the efficiencies of antigen presentation by the transfectants were different based on antigen concentrations required for T-cell stimulation. While WT I-A^b transfectant consistently presented the antigen better to A2.5 and A69.5 hybridomas, MT I-A^b transfectant presented better to PCH 4.1 and BOH4 hybridomas (Fig. 3). Our results clearly indicate that the β-chain TM of class II molecules provides a unique function in processing and presentation of a native antigen, and this cannot be replaced by the α-chain TM.

Figure 3.

Antigen presentation by WT and MT transfectants. Antigen presentation to A2.5, A69.5, PCH4.1 and BOH4 hybridomas. T hybridoma cells (5 × 10⁴) were cultured for 24 hr with HEL as an antigen in the presence of 1 × 10⁵ irradiated untransfected A20 cells, WT or MT I-A^b transfectants. Culture supernatants were assessed for the presence of IL-2 by the level of ³H-TdR uptake using CTLL-2 cells. A representative of four separate experimental results are shown here.

Because the level of costimulatory molecules may influence T-cell activation, the base level of several costimulatory molecules on the surface of parental cell line as well as WT and MT I-A^b transfectants were examined. All the unactivated cell lines expressed surface B7.2 (CD86) and none displayed significant levels of B7.1 (CD80) or CD40 (data not shown). These data suggest that altered antigen presentation capacity of WT and MT I-A^b is not due to the different levels of costimulatory molecules on the transfectants.

Different antigen processing time is required for presentation by WT and MT I-A^b

In order to assess the potential differences between WT and MT I-A^b transfectants in the kinetics of antigen processing/presentation, splenic antigen presenting cells were exposed to HEL for various time periods before fixation with paraformaldehyde. Fixed cells were then used to stimulate hybridoma cells for 18 hr (Fig. 4). The epitope generation (HEL_83–93) for A2.5 T-cell hybridoma was much more efficient in APC expressing WT I-A^b than MT I-A^b molecules. The generation of the epitope in APC expressing MT I-A^b peaked at 2 hr and gradually reduced thereafter, whereas APC expressing WT continuously increased during the 6–18-hr period. Interestingly, this epitope presentation by APC expressing WT I-A^b becomes far stronger than MT I-A^b at 18 hr (Fig. 4), similar to that shown at 72 hr in Fig. 2. The epitope presentation patterns to A69.5 and BOH4 by A20 cells expressing WT and MT I-A^b were similar to those to the above A2.5 and PCH4.1 hybridomas, respectively (data not shown). These results strongly suggest that the kinetics of I-A^b presentation of epitopes is significantly different between these transfectants, suggesting differential stability of the epitope-I-A^b complexes and/or transport of WT and MT I-A^b molecules to the antigen processing compartments.

Figure 4.

Assessment of processing/presentation time required for HEL_81–93 and HEL_47–60 epitopes in WT and MT I-A^b transfectants. Transfectants (1 × 10⁵ cells/ml) were incubated with HEL for varying time periods, fixed with 0·1% paraformaldehyde, and then further incubated for 24 hr with T-cell hybridomas in the absence of additional antigen. Supernatants of the cultures were assessed for the production of IL-2 by ³H-TdR uptake using CTLL-2 as described above.

In contrast, the generation of a cryptic epitope for I-A^b, HEL_46–60 recognized by PCH4.1²⁰ was severely limited in APC expressing WT I-A^b compared to very efficient processing in APC expressing MT I-A^b molecules (Fig. 4). These differences in the epitope generation strongly suggest the potential differences in the trafficking of the I-A^b molecules for efficient peptide loading, as these APC are different only in the expressed I-A^b molecules. It is also noteworthy that the generation of normally cryptic HEL_46–60 epitope in C57BL/6 low responder mice, which exhibited 10-fold lower avidity to I-A^b molecules compared to HEL_83–93 region (data not shown), can be drastically altered by a modification in the TM of class II molecules.

Processing/presentation difference is not caused by differences in association with Ii or DM

We next examined the possibility that the differences in the processing and presentation between WT and MT class II molecules may represent a deficiency in association with Ii or DM required for efficient I-A protection/transport or peptide exchange, respectively.²¹ Extensive previous studies indicate that cytoplasmic tails of Ii as well as DM molecules associated with class II molecules exhibit sorting signals involved in transport of the complexes to endocytic compartments.^22–24 Therefore, deficiencies in the molecular associations of I-A, Ii and/or DM may influence the ability to transport class II molecules into processing and/or loading compartments. In order to compare the levels of association between transfected I-A^b and Ii, I-A^b- and Ii-associated molecules were isolated from WT I-A^b and MT I-A^b transfectants by immunoprecipitation using specific antibodies to I-A^b and Ii, respectively. The immunoprecipitates were then analysed by Western blotting (Fig. 5). The results clearly indicate that there is no detectable deficiency in the association between MT I-A^b molecules and Ii as compared to WT I-A^b and Ii. In addition, both WT and MT I-A^b molecules in the transfectants appear to form an SDS-stable A_α/A_β complex since the major reactivity of anti-I-A^b antibody is found in the high molecular weight band corresponding to the A_α/A_β complex. These results suggest that the differences in the antigen presentation by WT and MT I-A^b molecules are not caused by the differences in the ability to associate with Ii. This is not surprising as the region of class II molecules involved in association with Ii is located in the α1 domain of A_β^25,26 although alteration in the TM of A_β may affect the conformation of the α1 region involved in association with Ii chains.

Figure 5.

Western blot analyses of I-A^b molecules associated with Ii in WT and MT I-A^b transfectants. (a) Ii chains were isolated from WT or MT I-A^b transfectants by immunoprecipitation using anti-Ii antibody and then analysed by Western using I-A^b antibody. (b) I-A^b molecules were isolated by immunoprecipiation using anti-I-A^b antibody and then similarly analysed by Western blotting using anti-Ii antibody.

To further determine whether the differential antigen presentation by WT and MT I-A^b molecules reflects differences in requirement of DM, A20 cells deficient in DM expression¹⁷ were transfected with WT and MT I-A^b and the ability to stimulate T-cell hybridomas were subsequently examined (Fig. 6). The level of I-A^b transfection was equivalent based on their cell surface expression (not shown) and the ability to present an epitope peptide (HEL_81–93) to a T-cell hybridoma (Fig. 6). However, in the absence of DM molecules, these epitopes from HEL for both types of T-cell hybriodmas are apparently not generated, because none of the T-cell hybridomas are activated in the presence of HEL. These results strongly suggest that the generation of epitopes for these hybridomas requires the presence of DM molecules. Thus, the requirement of DM is not altered for MT molecules.

Figure 6.

Processing and presentation of HEL in the absence of DM molecules. A20 cells deficient in DM molecules were transfected with either WT or MT I-A^b molecules and their ability to process and present HEL to T cell hybridomas (A2.5 and PCH4.1) were assessed. In the absence of DM, HEL cannot be processed and presented to either T-cell hybridoma.

WT and MT I-A^b molecules differ in their endocytic trafficking patterns

In attempts to elucidate the mechanisms underlying differential antigen-presenting capacity of WT and MT I-A^b molecules, the endocytic trafficking patterns of these molecules were monitored by using fluorochrome tagged antibodies and confocal microscopy (Fig. 7). Within 15 min, both WT and MT I-A^b molecules could be detected within the early endocytic compartments (transferrin receptor-rich compartments). In 30 min, the level of MT I-A^b molecules increased much higher than WT I-A^b in these compartments. Both WT and MT I-A^b molecules appear to recirculate back to the cell surface and differences between the transfectants disappear after 60 min (data not shown). A similar result was observed for trafficking of the WT and MT I-A^b molecules to the late endocytic compartments (LAMP1-positive). No difference was observed between WT and MT I-A^b transfectants after 15 min (Fig. 7 and Table 1); the same percentage of WT and MT I-A^b molecules was localized in late endocytic compartments. However, approximately 1·5 times more MT I-A^b molecules than WT were localized at 30 min within late endocytic compartments. After 60 min, almost all WT I-A^b molecules have left this compartment and yet nearly half of MT molecules were still present. Thus, the MT I-A^b molecule carrying a homodimeric α-chain TM region (lacking β-chain TM region) provides alterations in the endocytic trafficking of the molecules, ultimately influencing the efficiencies of antigen processing and presentation of certain epitopes. It appears that MT I-A^b molecules tend to accumulate at a higher level in both early and late endocytic compartments and remain in these compartments for longer time intervals than the WT counterparts. Perhaps MT I-A^b molecules are trapped longer in the late endocytic compartments because of the lack of a β-chain TM domain, which may be involved in facilitating transport to the cell surface.

Figure 7.

Confocal microscopy of the intracellular localization of WT and MT I-A^b molecules in transfectants. Cells (2 × 10⁶) were incubated on ice with FITC-conjugated anti-mouse I-A^b and PE-conjugated anti-mouse CD71 (transferrin receptor) antibodies, and either fixed immediately with 3·75% formaldehyde in PBS (0 min time point) or fixed after incubation for 15–60 min at 37°. Fixed cells were washed, mounted and then analysed under the confocal light microscope. To determine late endocytic compartments, rat anti-mouse LAMP-1 followed by AlexaFluor™-conjugated goat anti-rat IgG antibodies were similarly applied instead of anti-CD71 antibody. Ten random fields of view were analysed for each transfectant at all time points and a representative cell at each time point is shown.

Table 1.

Transport patterns of I-A^b WT and MT molecules to LAMP-1 associated endocytic compartments

No of cells	0 min	15 min	30 min	60 min*
I-Ab WT	0/10	3/10	3/10	0/10
	(0%)	(30%)	(30%)	(0%)
I-Ab MT	0/10	3/10	6/12	5/11
	(0%)	(30%)	(50%)	(45%)

The percentage of cells containing colocalized MHC class II and LAMP-1 were assessed under confocal microscopy as shown in Fig. 7.

^*The difference between WT and MT at 60 min is significant (P = 0·03) based on the two-sided Fisher exact test.

Discussion

MHC class II molecules are essential for activation of CD4⁺ T cells via the TCR in an antigen-specific manner. To present an epitope peptide to the TCR for antigen-specific T-cell activation, class II molecules should exhibit the peptide bound to the class II molecules. The binding of class II molecules and peptides occurs during passage through various endocytic compartments, in conjunction with a series of proteolytic processing of antigens. Studies investigating the structural and functional relationship of class II molecules indicate that the cytoplasmic domain of the MHC class II molecule significantly affects the efficiency of antigen processing and presentation.^6–8 In this paper, we have focused on the potential role of the TM regions of the class II molecule in antigen processing and presentation. In the absence of an A_β TM region by substitution with an A_α TM region, the mutant class II molecules displayed an altered transport of the class II molecules to the endocytic compartments, resulting in a different pattern of antigen processing and presentation. Therefore, these results suggest that the A_β TM region is involved in the trafficking of class II molecules, affecting consequent processing and presentation of select epitopes of an antigen molecule.

A number of previous studies have demonstrated that select epitopes of the same antigen can be generated in different compartments of the endocytic pathway.^13,27 Therefore, different trafficking patterns of WT and MT I-A^b molecules may provide differential peptide-loading to I-A^b molecules for altered antigen presentation. For example, if certain epitopes are preferentially generated from a native antigen in early endocytic compartments and MT I-A^b molecules are accumulated in this compartment for longer time periods than WT molecules, the MT molecules will have a definite advantage over WT molecules in binding and, subsequently, presenting these epitopes to the T cells. It is also interesting to note that HEL_46–60 is a cryptic epitope in C57BL/6 mice^18,20 and this epitope is preferentially presented by MT I-A^b molecules (Fig. 3). This enhanced epitope presentation by MT I-A^b molecules appear to result from a combination of preferential generation of the epitope peptide in early endocytic compartments and altered trafficking and retention of MT I-A^b favouring the early compartments. It is not expected that the rate of epitope generation for a given epitope at the same endocytic compartment of the WT and MT transfectants is any different as both transfectants are from the A20 B-cell line. Thus, it is likely that the differences in the trafficking pattern of these class II molecules are responsible for such differential presentation.

In contrast, HEL_83–93 and potentially HEL_76–85 epitopes are preferentially generated in late endocytic compartments following prolonged processing, and are much more efficiently presented by WT I-A^b molecules. Thus, dominance of T-cell responses to protein epitopes may be influenced by differences in trafficking of the class II molecules and preferential generation of the epitopes in particular endocytic compartments. Consequently, alteration in the TM region affecting transport of class II molecules may potentially change the immunodominance of epitope-specific T-cell responses. Therefore, our finding for an association between the TM region of the class II molecules and presentation of native antigen may have an influence on immunity, which is associated with polymorphism of MHC class II haplotypes differing in the TM sequences.²⁸

The molecular mechanism leading to alterations in antigen processing/presentation in transfectants expressing MT I-A^b molecule is not yet clear. MHC class II molecules with TM domains of class II α and β chains substituted with those of the IL-2 receptor failed to present certain epitopes in vitro.¹⁰ We have here ruled out the possibility that alterations in antigen presentation by substitution of the TM region significantly affect the ability of the mutant class II molecules to associate with Ii or reflect the altered function of associated DM molecules (Figs 5 and 6). Therefore, these studies, including our current study, strongly suggest that the TM regions of class II molecules may play an important role in antigen presentation. Previous studies indicate that the CY domains of MHC class II, Ii as well as DM molecules influence transport of class II molecular complexes to the endocytic compartments and consequent antigen presentation.^7,22–24 Therefore, it is conceivable that the TM substitution may modify the sorting signal located in the CY domain of the MHC class II β chain⁷ and this in turn may affect the transport of I-A molecules.

Alternatively, the modification in the TM region may alter the efficiencies of Ii and/or DM molecules, which are known to influence the transport of class II molecules to various endocytic compartments.^9,22–24 Our immunoprecipitation studies suggest that there is no gross differences in the levels of class II/Ii complexes in the transfectants (Fig. 5). This is not surprising, considering the contact residues are located away from the TM region.^25,26 In addition, the presence of DM appears to be essential for processing/presentation of both epitopes, suggesting that DM may not directly be involved in the differential epitope presentation. However, we cannot exclude the possibility that modification in the TM sequences may influence stability and consequent transport of class II/Ii/DM complexes, which is sufficient to affect the antigen processing and presenting function for certain epitopes. In addition, A_β TM region appears to be important for B-cell activation^6,11 and modification in the activation signal may influence the trafficking of the class II molecules. Nevertheless, further studies will be required to elucidate the molecular mechanisms of the TM regions of class II molecules affecting the antigen processing/presentation of select epitopes.

Abbreviations

TM

transmembrane

CY

cytoplasmic

WT

wildtype

MT

mutant

HEL

hen egg lysozyme