Regulated Folding of Tyrosinase in the Endoplasmic Reticulum Demonstrates That Misfolded Full-Length Proteins Are Efficient Substrates for C
http://www.100md.com
免疫学杂志 2005年第5期
Abstract
Short-lived protein translation products have been proposed to be the principal substrates that enter the class I MHC processing and presentation pathway. However, the biochemical nature of these substrates is poorly defined. Whether the major processing substrates are misfolded full-length proteins, or alternatively, aberrantly initiated or truncated polypeptides still remains to be addressed. To examine this, we used melanoma in which one-third of wild-type tyrosinase molecules were correctly folded and localized beyond the Golgi, while the remainder were present in the endoplasmic reticulum in an unfolded/misfolded state. Increasing the efficiency of tyrosinase folding using chemical chaperones led to a reduction in the level of substrate available to the proteasome and decreased the expression of a tyrosinase-derived epitope. Conversely, in transfectants expressing tyrosinase mutants that are completely misfolded, both proteasome substrate and epitope presentation were significantly enhanced. Proteasome substrate availability was a consequence of misfolding and not simply due to retention in the endoplasmic reticulum. Thus, the extent of folding/misfolding of a full-length protein is an important determinant of the level of epitope presentation.
Introduction
The nature of the protein substrates that enter the class I MHC Ag processing pathway has been the subject of intense investigation over several years. A significant percentage of cellular proteins with masses greater than 45 kDa is degraded by the proteasome within 60 min of synthesis (1). Newly synthesized proteins have also been shown to be an important source of substrates for the TAP (2) and to be the source of one virally derived epitope (3). It has been proposed that these newly synthesized and rapidly degraded proteins or peptides could correspond to defective protein translation products that were aberrantly initiated or prematurely terminated (4). In support of this, several epitopes have been identified as short polypeptides translated from alternative reading frames, some of which span exon-intron boundaries or lie entirely within introns, and others of which result from the use of nonstandard initiation codons (summarized in Ref.5). It has also been shown that the level of epitope presentation is enhanced using truncated forms of a protein (6). However, it is not clear what fraction of the total protein economy of the cell they represent.
An alternative source of rapidly degraded substrates for class I Ag processing is full-length proteins that are sampled: 1) after they become terminally misfolded; 2) while they are still in the process of folding; or 3) because they are short-lived based on N-end rule or other sequences that directly target them for degradation (7, 8, 9). Substitution of the N-terminal residue of proteins with N-end destabilizing residues, such as Arg, decreases the t1/2 of proteins by increasing the level of ubiquination, which in turn augments interaction with and degradation by the proteasome (10). In several studies, this maneuver led to an increased level of epitope presentation to T cells (11, 12, 13, 14, 15), while in others it had no effect (11, 16). However, because ubiquination is dependent on the N-terminal residue rather than protein-folding state, these studies did not address the importance of protein-folding state as a determinant of class I Ag presentation. The insertion of a sequence consisting largely of repeated Lys and Glu residues (KEKE) has also been shown to favor proteasome degradation and epitope presentation (15, 17). Although it has been proposed that the KEKE sequence directly augments interaction with the proteasome (18) or that it induces misfolding (15, 17), no definitive data distinguishing these possibilities have been published. Interestingly, however, this work also demonstrated that a protein with an intermediate rate of degradation generated class I MHC-restricted epitopes more efficiently than a more rapidly degraded form (15), and also demonstrated the apparent existence of two distinct pools of substrates, giving rise to epitopes based on kinetic measurements. However, the nature of these substrates was not directly examined in this or other work performed to date.
Taken together, these studies demonstrate that enhanced targeting of model protein Ags to the proteasome frequently augments the expression of class I-restricted epitopes, but may also make no difference or actually cause diminished expression. They also suggest that this could depend on the exact nature of the protein substrate for degradation, of which little is directly known. Indeed, while there is strong evidence that newly synthesized translation products are a significant source of class I MHC-restricted epitopes in virally infected cells (1, 2, 3, 15), there are little direct data that support this hypothesis for epitopes derived from stably expressed endogenous proteins (2). As has been pointed out elsewhere, it still remains to be established that full-length proteins that are either terminally misfolded or in the process of folding are significant substrates for the class I Ag processing pathway (4, 5). It is also not clear whether such proteins would be sampled after terminal misfolding or at early stages during the folding process.
One meaning to assess the impact of protein folding on the level of epitope presentation is to compare wild-type (WT)3 proteins with naturally occurring mutant forms that are known to misfold, or to alter other conditions that selectively affect the folding of individual proteins within cells. Tyrosinase, which represents a robust source of class I MHC-restricted peptides recognized by melanoma-specific T cells (19), offers a useful model for this evaluation. Tyrosinase is a membrane-associated N-linked glycoprotein that is synthesized and folds in the endoplasmic reticulum (ER), and is subsequently transported via the trans Golgi network to melanosomes (20, 21). Many natural variants of tyrosinase associated with type I oculocutaneous albinism fail to fold (22, 23, 24, 25), and unfolded or misfolded WT tyrosinase molecules frequently accumulate in melanoma cells as a consequence of abnormal cellular acidification (26, 27). These unfolded/misfolded mutant tyrosinase or WT tyrosinase molecules are completely retained in the ER through prolonged interaction with the ER chaperones calnexin and calreticulin rather than exiting to melanosomes, and are subsequently retrotranslocated to the cytosol and degraded by the proteasome (20, 24, 26, 27, 28, 29). In contrast, the tyrosinase substrates L-dopamine (L-Dopa) and L-tyrosine act as chemical chaperones to enhance the folding of WT tyrosinase molecules and their export from the ER to the Golgi (23). Work from our laboratory established that the Tyr369 epitope presented by HLA-A*0201 was produced after retrotranslocation of tyrosinase from the lumen of the ER to the cytosol, degradation by the proteasome, and reimport of peptides into the ER by TAP (28, 30). In this study, we used WT tyrosinase molecules and tyrosinase mutants that have been demonstrated to fail to fold (23, 24, 25, 26, 31), together with chemical chaperones, to evaluate the impact of folding of full-length molecules on the processing and presentation of the Tyr369 epitope.
Materials and Methods
Tyrosinase gene constructs and transfectants
The WT and R402Q tyrosinase genes (gift from M. Marks, University of Pennsylvania, Philadelphia, PA) were subcloned in the vector pEF6 (Invitrogen Life Technologies). The A206T mutation was introduced into pEF6 WT tyrosinase using the QuickChange XL site-directed mutagenesis kit (Stratagene). The melanoma cell line DM331 (tyrosinase–, HLA-A*0201+) (32) was transfected with plasmids encoding WT, R402Q, or A206T tyrosinase using Fugene (Roche Diagnostics). Bulk stable transfectant lines were generated by selection with blasticidin (10 µg/ml) (Invitrogen Life Technologies) and cloned. All experiments using stable transfectants were conducted on cells plated at 5000 cells/cm2 and grown for 3 days. Folding of tyrosinase was induced by incubation of cells for 3 days in medium supplemented with 20 µM L-Dopa (Sigma-Aldrich) or 400 µM L-tyrosine (Sigma-Aldrich). To ensure that the pH of the medium and the concentration of L-Dopa remain the same during the 3-day experiment, the medium and medium supplemented with L-Dopa or L-tyrosine were replaced every day. Retention of tyrosinase in the ER was induced in medium supplemented with 5 µg/ml brefeldin A (BFA) (Sigma-Aldrich) or by incubation at 20°C for 4 h. For experiments analyzing proteasome degradation, transfectants were incubated in presence of 50 µM N-acetyl-Leu-Leu-norleucinal (LLnL) (Calbiochem) for 3.5–4 h.
Cytofluorometry
Expression of tyrosinase was measured by intracellular staining after permeabilization for 30 min with BD Perm/wash (BD Pharmingen) and sequential addition of C19 (goat anti-tyrosinase; Santa Cruz Biotechnology), biotinylated donkey anti-goat IgG (Santa Cruz Biotechnology), and allophycocyanin-conjugated streptavidin (BD Pharmingen). Expression of HLA-A*0201 was measured by surface staining using the mAb CR11-351 and FITC-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratories). Data were acquired on a FACS and analyzed using CellQuest software (BD Pharmingen).
Immunoblotting and analysis of maturation of N-linked glycans
Cell pellets (106 cells) were solubilized in 100 µl of buffer containing 10 mM Tris-HCl, pH 7.5, 0.5% deoxycholate, 1% Igepal, 5 mM EDTA, 4 mM PMSF, 10 µg/ml aprotinin, 10 µM pepstatin A, 10 µg/ml leupeptin, and 100 µM iodoacetamide, and centrifuged at 21,000 x g for 30 min at 4°C. For analysis of maturation of N-glycans, proteins from 50 µl of supernatant were precipitated with 400 µl of cold acetone at –20°C overnight. The precipitates were resuspended in 0.5% SDS and 0.1 M 2-ME and boiled for 5 min. After cooling on ice for 2 min, the samples were digested with 2.5 mU endoglycosidase H (Endo H; Roche Diagnostics), according to the manufacturer’s recommendations. Alternatively, proteins from supernatant were directly boiled in sample buffer for 5 min. A total of 5 x 105 cell equivalents per lane was separated on 10% SDS-PAGE gels, transferred to NitroPure nitrocellulose (Osmonics), and blocked in 5% nonfat dried milk in PBS with 0.05% Tween 20. Blots were probed overnight with C19 goat anti-tyrosinase, washed, probed with HRP-conjugated donkey anti-goat IgG, and developed according to the Amersham ECL protocol (Amersham Biosciences). Films were scanned using Scan Maker III (Microtek), and images were processed using Adobe Photoshop software. Bands were quantified using NIH Image 1.62f software.
Immunofluorescence microscopy
Transfectants were grown on coverslips, washed with PBS, fixed for 10 min at room temperature with 4% paraformaldehyde in PBS, and permeabilized with BD Perm/wash (BD Pharmingen). Tyrosinase was detected either with C19 goat (Santa Cruz Biotechnology) or T3.11 mouse anti-tyrosinase Ab with similar results (25). The cells were incubated for 1 h with combinations of anti-tyrosinase Ab, rabbit anti-calnexin (StressGen Biotechnologies), or mouse anti-lysosome-associated membrane protein (LAMP)1 (BD Pharmingen). These primary Abs were detected with Alexa 488 anti-goat and Alexa 594 anti-rabbit or anti-mouse conjugates (Molecular Probes). Samples were mounted on glass slides with Vectashield (Vector Laboratories), visualized using an Olympus confocal microscope, and processed with Adobe Photoshop 6.0 software.
T cell assay
Tyr369 epitope-specific T cell lines were generated from C57BL/6 mice expressing a chimeric HLA-A*0201/H2Dd MHC class I molecule and maintained, as described (28). Resting T cells (50,000) were incubated with the indicated number of transfectants or Tyr369 peptide-pulsed DM331 cells for 16 h, and the level of IFN- secreted was measured by ELISA (eBiosciences).
Results
Maturation, folding, and intracellular localization of WT and mutant forms of tyrosinase in an amelanotic melanoma cell line
To examine the importance of protein folding and subcellular localization on the expression of tyrosinase-derived epitopes, we generated transfectants of the amelanotic melanoma cell DM331 with cDNA expression vectors encoding either WT human tyrosinase, or the tyrosinase mutants R402Q and A206T that have been previously shown to be retained in the ER in consequence of intrinsic misfolding (31). DM331 does not express detectable tyrosinase based on Western blot with anti-tyrosinase Abs (32) or PCR (our unpublished observations), but does express HLA-A*0201. Stable transfectant clones expressing similar levels of each tyrosinase construct were identified by intracellular staining using the Ab C19, which recognizes a C-terminal epitope of tyrosinase. Tyrosinase maturation in these transfectants was evaluated by digestion of postnuclear supernatants of cell lysates with Endo H, separation by SDS-PAGE, and immunoblotting with C19. In cells expressing WT tyrosinase, quantitation of total signal intensities of the bands by densitometry demonstrated that about one-third of the molecules were insensitive to Endo H, indicating that they had moved out of the ER and beyond the cis-/medial Golgi (Fig. 1A). Similar values have been obtained in five independent experiments involving four different cloned transfectants (data not shown). This observation is consistent with others documenting a relatively low level of mature tyrosinase in melanoma cells as opposed to melanocytes (26). However, in agreement with previous work (23), the amount of mature tyrosinase was increased markedly by growth of the transfectants in medium containing L-Dopa (Fig. 1A) or elevated levels of L-tyrosine (data not shown). In contrast, R402Q and A206T tyrosinase molecules were completely Endo H sensitive (Fig. 1A). L-Dopa (Fig. 1A) and L-tyrosine (data not shown) did not induce any significant maturation of A206T tyrosinase, and only a very minor increase in the maturation of R402Q. Using immunocytochemistry and confocal microscopy, WT tyrosinase was found to partially colocalize with calnexin, a marker of the ER, and with LAMP1, a marker of late endosomes, lysosomes, and melanosomes (Fig. 1B). In contrast, R402Q and A206T tyrosinase molecules colocalized strongly with calnexin and failed to colocalize with LAMP1. Taken together, these observations confirmed that a significant portion of WT tyrosinase molecules in the DM331 melanoma transfectants had matured beyond the cis-/medial Golgi, reflecting their capacity to fold, while R402Q and A206T tyrosinase molecules were completely immature, retained in the ER, indicating their inability to fold.
FIGURE 1. A significant fraction of WT tyrosinase is mature and localized in late endosomes, whereas R402Q and A206T tyrosinase are immature and retained in the ER. A, Cloned transfectants of DM331 melanoma cells stably expressing WT, R402Q, or A206T tyrosinase were grown for 3 days in the presence or absence of L-Dopa, lysed, digested with Endo H, analyzed by SDS-PAGE, and immunoblotted with the anti-tyrosinase Ab C19. Endo H-insensitive mature tyrosinase migrates at 70 kDa, while Endo H-sensitive immature tyrosinase migrates at 60 kDa. The membrane was then stripped and immunoblotted with anti-calnexin. Numbers representing the total signal intensities as quantified by scanning densitometry density and corresponding to mature and immature WT tyrosinase are indicated. Results represented in A and Figs. 2, 5, and 6, experiment 2, were obtained from experiments conducted in parallel and using duplicate aliquots of melanoma cells. B, Immunofluorescence and confocal microscopy of cloned transfectants of DM331 melanoma cells stably expressing WT, R402Q, or A206T tyrosinase using anti-tyrosinase and anti-calnexin or anti-LAMP1.
Proteasome-dependent degradation of R402Q and A206T tyrosinase is enhanced compared with WT tyrosinase
We next evaluated whether unfolded/misfolded R402Q and A206T full-length tyrosinase molecules were preferential substrates for the proteasome in comparison with WT tyrosinase. To do so, we quantified the amounts of degradation intermediates accumulating after incubation of the transfectants with the proteasome inhibitor LLnL. Treatment with 100 µM LLnL for 4 h, which we have previously shown to inhibit 90–97% of different proteasome activities (33), induced the accumulation of 60-kDa tyrosinase-derived bands in WT, R402Q, and A206T tyrosinase transfectants (Fig. 2). These bands have previously been shown to correspond to either unglycosylated and full-length (26, 29) or glycosylated and partially proteolyzed (29) tyrosinase molecules, many of which have been retrotranslocated to the cytosol (28, 29). Bands of the same mass are stabilized by LLnL (26, 28, 29), lactacystin (26, 29), MG132 (26), and epoxomycin (our unpublished results), and are apparently generated by a protease distinct from the proteasome. Quantitation by scanning densitometry established that in presence of LLnL, the relative amounts of the 60-kDa bands compared with those of full-length tyrosinase were 2.5- to 3-fold higher for R402Q and A206T tyrosinase than for WT tyrosinase. These results demonstrated that the amount of R402Q and A206T tyrosinase available to and degraded by the proteasome over 4 h is greater than for WT tyrosinase.
FIGURE 2. Proteasome degradation substrates from R402Q and A206T tyrosinase are present at relatively higher levels than those from WT tyrosinase. Cloned transfectants of DM331 melanoma cells stably expressing WT, R402Q, or A206T tyrosinase were incubated in presence or absence of LLnL for 4 h. Cell lysates were analyzed by SDS-PAGE and immunoblotted with the C19 anti-tyrosinase Ab. Films were exposed for 0.5–1 min to detect the broad bands centered at 70 kDa, which represent immature and mature intact forms of tyrosinase, and for 3 min to detect the bands at 60 kDa, which represent proteasome-sensitive deglycosylated or partially proteolyzed tyrosinase molecules. Numbers below each band represent the total signal intensities as quantified by scanning densitometry density, and the ratio of the intensities of the 60- to 70-kDa species is indicated. Results represented in this figure and Fig. 1A are obtained from experiments conducted in parallel and correspond to the same aliquots of melanoma cells.
Tyr369 epitope expression levels on melanoma transfectants expressing equivalent levels of full-length WT, R402Q, A206T tyrosinase
To evaluate whether intrinsically unfolded/misfolded full-length tyrosinase molecules were a preferential proteasome substrate for epitope presentation, we next evaluated the expression levels of the HLA-A*0201-restricted Tyr369 epitope on the different DM331 melanoma transfectants by measuring their ability to stimulate IFN- secretion from Tyr369-specific T cell lines. First, we established that the assay conditions would accurately reflect differences in epitope expression on transfectants expressing different levels of either WT tyrosinase or HLA-A*0201. Using melanoma transfectants that expressed similar levels of HLA-A*0201 molecules, we observed a correlation between the tyrosinase expression level in different transfectants and the secretion of IFN- (Fig. 3A, Expt. 1). Similarly, in transfectants expressing similar levels of tyrosinase, we observed a correlation between IFN- secretion and the level of HLA-A*0201 expression (Fig. 3A, Expt. 2). In parallel experiments, a 2-fold difference in the secretion of IFN- was correlated to a 2-log difference in the concentration of Tyr369 synthetic peptide used to pulse untransfected DM331 melanoma cells (Fig. 3B). Finally, we demonstrated that incubation of transfectants expressing WT tyrosinase with exogenous Tyr369 synthetic peptide led to a further increase in IFN- secretion (Fig. 3C). These experiments showed that this assay was sensitive to quantitative variations in Tyr369 epitope expression resulting from either tyrosinase, HLA-A*0201, or exogenous peptide expression differences.
FIGURE 3. The level of expression of WT tyrosinase or HLA-A*0201 determines the level of presentation of Tyr369. A, Top panels, Expression of WT tyrosinase in three independent clones of transfected DM331 was measured by intracellular staining with anti-tyrosinase and flow cytometry. Thin black line, WTcl1a or WTcl1b; thin gray line, WTcl2; thick line, WTcl8a or WTcl8b; gray fill, untransfected DM331. The arithmetic mean fluorescence intensity (MFI) of each distribution after subtraction of the DM331 value is shown in the second row of panels. Middle panels, Expression of HLA-A*0201 in the same clones was measured by surface staining with anti-HLA-A*0201 and flow cytometry. Bottom panels, The presentation of Tyr369 was evaluated as the level of IFN- secreted by Tyr369-specific T cell lines after incubation with the indicated DM331 transfectant clones. B, T cell recognition of DM331 transfectant clones expressing different levels of WT tyrosinase and the same number of untransfected DM331 cells pulsed overnight with different concentrations of exogenous Tyr369 peptide was evaluated by IFN- secretion. C, T cell recognition of untransfected DM331 cells and a DM331 transfectant expressing WT tyrosinase before or after overnight pulsing with 1 µM Tyr369 peptide.
We next compared the presentation levels of the Tyr369 epitope by transfectants expressing WT, R402Q, and A206T tyrosinase molecules. Using short-term bulk populations in which the mean levels of tyrosinase expression and the percentages of tyrosinase+ cells were similar, cells expressing R402Q tyrosinase stimulated the release of 1.8-fold more IFN- from Tyr369-specific T cells than did cells expressing the WT molecule (Fig. 4, Expt. 1). Similarly, when two different Tyr369-specific T cell lines were stimulated with long-term cloned transfectants that expressed uniform and comparable levels of WT and R402Q tyrosinase, the cells expressing R402Q stimulated the release of 1.8- to 2.1-fold more IFN- (Fig. 4, Expt. 2). This numerical range has been observed in five different experiments involving three distinct cloned transfectants expressing WT or R402Q tyrosinase molecules. Finally, a transfectant expressing A206T tyrosinase still stimulated a 1.6-fold higher level of IFN- secretion than did a transfectant expressing almost one-third more WT tyrosinase (Fig. 4, Expt. 3). In three independent experiments involving the same transfectant clones, the release of IFN- was 1.4- to 3.5-fold higher in the presence of the transfectant clone expressing A206T than in presence of the clone expressing WT molecules, with a mean of 2.2-fold. Collectively, the results of Figs. 2–4 indicate that the expression of unfolded or misfolded tyrosinase is associated with enhanced degradation by the proteasome and leads to a significantly augmented presentation of the processed Tyr369 epitope at the cell surface.
FIGURE 4. Presentation of Tyr369 from R402Q and A206T tyrosinase is enhanced compared with that of WT tyrosinase. Analyses were conducted using bulk populations (Expt. 1) or clones (Expts. 2 and 3) of DM331 transfectants expressing WT, R402Q, or A206T tyrosinase. Upper panels, Expression of tyrosinase was measured by intracellular staining and flow cytometry (thin line, WT tyrosinase transfectant or clone; thick line, R402Q in Expts. 1 and 2 and A206T in Expt. 3; gray fill, untransfected DM331). The background fluorescence observed for DM331 has been subtracted in the bar graphs of MFI. Middle panels, Expression of HLA-A*0201 was measured by surface staining with anti-HLA-A*0201 and flow cytometry. Bottom panels, The presentation of Tyr369 was evaluated as the level of IFN- secreted by Tyr369-specific T cell lines after incubation with transfectants. In experiment 1, DM331 cells were transfected with either WT or R402Q tyrosinase, and 100,000 or 30,000 short-term bulk transfectants in which the percentages of tyrosinase+ cells were similar were incubated with tyrosinase-specific T cell line 2. In experiments 2 and 3, 5,000 cloned transfectants were incubated with specific T cell lines 1 or 2, and the levels of secretion of IFN- shown are after subtraction of the background level of IFN- secreted in presence of untransfected DM331.
Folding, maturation, and export of WT tyrosinase from the ER diminish tyrosinase processing and presentation of Tyr369
As shown in Fig. 1 and elsewhere (23), growth of melanoma cells with the tyrosinase substrates L-Dopa or L-tyrosine enhanced folding and exit from the ER of WT tyrosinase, while having a minimal effect on the tyrosinase folding mutants. Consistent with this, growth of WT tyrosinase transfectants in L-Dopa for 3 days substantially reduced the relative amount of 60-kDa tyrosinase degradation intermediates that could be detected after a 4-h incubation of the cells with the proteasome inhibitor LLnL (Fig. 5, lanes 2 and 4). In contrast, the relative amount of 60-kDa tyrosinase fragments was actually somewhat higher in A206T transfectants treated with LLnL and L-Dopa compared with those treated with LLnL alone (Fig. 5, lanes 6 and 8). We therefore tested the hypothesis that addition of L-Dopa would lead to a reduction in the processing and presentation of Tyr369 from WT tyrosinase, but not from A206T tyrosinase. In three independent experiments, two of which are shown in Fig. 6A, we observed a significant decrease in the presentation of Tyr369 by transfectant expressing WT tyrosinase after growth in L-Dopa or L-tyrosine. However, no significant change in the presentation of Tyr369 by a transfectant expressing A206T tyrosinase was detected under the same conditions. The difference in the effect of L-Dopa on WT and A206T presentation of Tyr369 from WT and A206T tyrosinase was statistically significant (p < 0.005, Fig. 6B). Altogether, these results demonstrate that increased folding and export of tyrosinase from the ER to the Golgi decreased the fraction of tyrosinase molecules undergoing degradation and also decreased the magnitude of Tyr369 epitope presentation.
FIGURE 5. Induction of folding of WT tyrosinase with chemical chaperones decreases the amount of proteasome degradation substrates. Cloned transfectants of DM331 melanoma cells stably expressing WT or A206T tyrosinase were grown for 3 days in the presence or absence of L-Dopa, and then further incubated for 4 h in the presence or absence of LLnL. Cell lysates were analyzed by SDS-PAGE and immunoblotted with anti-tyrosinase. Films were exposed for 0.5–1 min to detect the broad bands centered at 70 kDa, which represent immature and mature intact forms of tyrosinase, and for 3 min to detect the bands at 60 kDa, which represent proteasome-sensitive deglycosylated or partially proteolyzed tyrosinase molecules. Numbers below each band represent the total signal intensities as quantified by scanning densitometry density, and the ratio of the intensities of the 60- to 70-kDa species is indicated. Results shown in this figure and Fig. 1A are obtained from experiments conducted in parallel and correspond to the same aliquots of melanoma cells.
FIGURE 6. Induction of folding of WT tyrosinase decreases presentation of Tyr369. A, Cloned transfectants of DM331 melanoma cells stably expressing WT or A206T tyrosinase were grown for 3 days in the absence () or presence of L-Dopa () or L-tyrosine (). Upper panels, Expression of tyrosinase was measured by intracellular staining and flow cytometry. The background fluorescence observed for DM331 has been subtracted in the bar graphs of MFI. Middle panels, Expression of HLA-A*0201 was measured by surface staining and flow cytometry. ND, not done. Bottom panels, Presentation of Tyr369 was evaluated as the level of IFN- secreted by a Tyr369-specific T cell line after incubation with cloned transfectants. The levels of secretion of IFN- shown are after subtraction of the background level of IFN- secreted in presence of untransfected DM331. Results shown in this figure, experiment 2, and Figs. 5 and 1A are obtained from experiments conducted in parallel and correspond to the same aliquots of melanoma cells. B, Percentage of inhibition of IFN- released by Tyr369-specific T cells as a consequence of melanoma cell culture in the presence of L-Dopa, calculated from the data of A, experiments 1 and 2. The p value was calculated using two-sample Student‘s t test and Minitab 13.0 software.
Proteasome-dependent degradation of tyrosinase is increased by misfolding regardless of ER retention and by ER retention separately of intrinsic misfolding
The results above suggested that the failure of tyrosinase to fold in the ER led directly to an increase in epitope expression based on enhanced interaction with the ER quality control system. However, they did not exclude the possibility that simple enhanced representation of the protein in the ER, regardless of its folding state, might also lead to enhanced processing and presentation. To assess the impact of misfolding on tyrosinase processing independent of differences in ER retention, we used BFA (34) or incubation of cells at 20°C (35, 36) to block anterograde transport of either WT or R402Q tyrosinase from the ER to the Golgi, and compared the accumulation of tyrosinase degradation intermediates after inhibition of proteasome function with LLnL. In BFA-treated cells, we found that the relative amount of 60-kDa tyrosinase degradation intermediates for R402Q tyrosinase was still 2.5-fold higher than that for WT tyrosinase (Fig. 7). A similar result was obtained after incubation of cells at 20°C. This demonstrates that the extent of tyrosinase misfolding determines retrotranslocation from the ER independently of its impact on retention in the ER. In addition, we observed that the relative amount of the 60-kDa degradation intermediates from WT tyrosinase was 3-fold higher in presence of BFA and LLnL than in presence of LLnL alone, while, as expected, the addition of BFA in the presence of LLnL only slightly increased the amount of 60-kDa tyrosinase degradation intermediates for R402Q tyrosinase (Fig. 7). This suggested that either properly folded WT tyrosinase was a substrate for retrotranslocation because of its retention, or residence in the ER led to an unfolding of properly folded molecules. Therefore, while we demonstrated that the extent of tyrosinase misfolding influences extraction from the ER independently of localization in the ER, we also provide evidence that retention in the ER independently of intrinsic misfolding favors exit from the ER and degradation by the proteasome.
FIGURE 7. Proteasome substrate levels generated from WT and R402Q tyrosinase retained in the ER. Clones of transfectants expressing WT or R402Q tyrosinase were incubated at 37°C in presence or absence of BFA for 30 min, or at 20°C, and then further incubated in the presence or absence of LLnL for 3.5 h. Cell lysates were analyzed by SDS-PAGE and immunoblotted with anti-tyrosinase. Films were exposed for 0.5–1 min to detect the broad bands centered at 70 kDa, which represent immature and mature intact forms of tyrosinase, and for 3 min to detect the bands at 60 kDa, which represent proteasome-sensitive deglycosylated or partially proteolyzed tyrosinase molecules. Numbers below each band represent the total signal intensities as quantified by scanning densitometry density, and the ratio of the intensities of the 60- to 70-kDa species is indicated.
Discussion
Short-lived proteins and defective protein translation products, which include aberrantly initiated, prematurely terminated, and unfolded or misfolded full-length proteins, have been proposed to be the principal substrates that enter the class I MHC processing and presentation pathway (1, 2, 3, 15). However, it is not clear to what extent proteasome substrates are represented by aberrant translation products that are inherently unable to fold, or alternatively, full-length proteins that are targeted for degradation either because of a conformational abnormality, because they take a longer time to fold properly, or because they contain signals that confer a short t1/2 (4). Many studies have shown that Ag presentation can be altered by the introduction of sequences that enhance proteasome targeting without regard for protein folding state, or by the creation of truncated proteins (6, 11, 12, 13, 14, 15). However, no study to date has directly evaluated the impact of protein folding on the efficiency of Ag processing and presentation, nor evaluated whether full-length proteins, as opposed to aberrant translation products, are substrates for processing (4, 5).
Using tyrosinase as a model, we have now provided direct insight into these issues. This is due in part to the fact that the C19 Ab detects an epitope located in the C-terminal tail of tyrosinase, which is located on the opposite side of the membrane from the lumenal domain whose folding determines tyrosinase maturation and degradation. By using this Ab, we thus evaluate full-length tyrosinase molecules, as opposed to prematurely terminated or aberrantly initiated products. In addition, the folding behavior of WT tyrosinase and the tyrosinase mutants R402Q and A206T has been well established in previous work. Unfolded or misfolded forms of WT tyrosinase are commonly found in melanoma cells (26) as a consequence of abnormal cellular acidification (27) and the absence of other melanocyte differentiation proteins (37, 38). Only about one-third of WT tyrosinase expressed in melanoma cells folds properly, while the remainder fails to fold, but can be rescued in presence of the chemical chaperones L-Dopa or L-tyrosine (23). R402Q tyrosinase molecules are completely retained in the ER at 37°C, but can be partially rescued and exported to the Golgi at low temperature or by addition of chemical chaperones (27, 31). A206T tyrosinase is a mutation leading to oculocutaneous albinism. It is completely retained in the ER through prolonged interaction with the chaperone calnexin (24), and cannot be rescued by low temperature incubation or chemical chaperones. Neither of these molecules exhibits enzymatic activity in the unfolded state, strongly suggesting that misfolding is responsible for their complete retention in the ER. Finally, N-glycans are attached to tyrosinase during synthesis, undergo structural modification as a consequence of folding and export from the ER, and are removed during retrotranslocation as a prelude to degradation by the proteasome. Based on our ability to distinguish among these forms, we have shown a direct relationship between the levels of the unfolded/misfolded and reverse translocated forms of full-length tyrosinase and the amount of epitope presented.
In the study presented in this work, we have directly assessed the influence of misfolding/unfolding in the case of a full-length protein, regardless of its impact on ubiquination, on the efficiency of Ag processing and presentation. We showed that increasing the folding efficiency of WT tyrosinase using chemical chaperones led to a consequent decrease in the amount of immature WT tyrosinase in the ER and proteasome-sensitive tyrosinase degradation intermediates correlated with a decrease in Tyr369 presentation. Conversely, in transfectants expressing R402Q and A206T, which are unable to fold and are completely retained in the ER, the proteasome-dependent degradation of tyrosinase molecules and the presentation level of Tyr369 were significantly enhanced compared with transfectants expressing WT tyrosinase. To exclude the possibility that the availability of tyrosinase molecules for degradation and epitope presentation was strictly a function of representation in the ER independent of folding state, we evaluated the impact of misfolding on degradation after retention of R402Q and WT tyrosinase in the ER using BFA or low temperature. Proteasome substrate availability was a consequence of misfolding state, and not simply due to retention in the ER. Our results demonstrate that the extent of folding or misfolding of full-length protein is an important determinant of the final level of epitope presentation.
We cannot fully exclude the possibility that prematurely terminated forms of tyrosinase, which we would not detect using C19, represent an additional source of Ag for class I presentation. However, to the extent that these protein forms contain a full-length lumenal domain whose folding is augmented by L-Dopa or L-tyrosine, and disrupted by R402Q and A206T mutations, they are functionally equivalent to full-length molecules in their contribution to Ag processing and presentation. Conversely, we would expect that aberrantly initiated tyrosinase molecules would remain largely unfolded/misfolded, even in the presence of L-Dopa or L-tyrosine, so that the differences in epitope presentation shown in this study would not have been observed. In addition, aberrantly initiated WT tyrosinase molecules would fail to enter the ER because they lack a signal sequence, and would be synthesized in the cytosol. We have previously shown that cells expressing a truncated form of tyrosinase in the cytosol present two forms of the Tyr369 epitope, one containing a genetically encoded Asn371 and the other containing a deamidated Asp at the same position (28). In contrast, cells expressing full-length tyrosinase express only the latter epitope (28, 39). Collectively, these results indicate that aberrantly initiated tyrosinase molecules are an insignificant source of class I MHC-restricted epitopes.
Somewhat surprisingly, we also found that enforced retention of WT tyrosinase in the ER using BFA enhanced proteasome-dependent degradation independently of intrinsic misfolding, although the level of proteasome substrate observed was still lower than that arising from R402Q tyrosinase under normal conditions. BFA collapses the cis-/medial Golgi into the ER to create a single compartment (34). This structure would then contain not only immature unfolded or misfolded tyrosinase, but also some molecules that are sufficiently well folded to have moved to the Golgi. The processing of immature unfolded/misfolded tyrosinase molecules should not be increased by enforced ER retention because these forms are normally retained in the ER. Properly folded tyrosinase molecules are also improbable sources for this enhanced retrotranslocation and degradation. Previous studies have shown that correct folded glycoproteins are not reglucosylated by the uridine diphosphate-glucose glucosyl transferase (40), and consequently, they do not interact with the calnexin/ER degradation-enhancing -mannosidase-like protein molecular complex that has recently been shown to play an important role in retrotranslocation from the ER (41, 42). Instead, we favor the hypothesis that the increase in retrotranslocation and degradation observed when ER retention is enforced results from the processing of a form of tyrosinase that is partially folded, and has a lower, but still significant probability of interaction with uridine diphosphate-glucose glucosyl transferase and calnexin. We hypothesize that this species is normally able to be exported from the ER to the Golgi, but is retrotranslocated and degraded by the proteasome when it is forced to be retained in the ER. Tyrosinase normally complexes with and is stabilized by other melanosomal proteins (37). It is possible that tyrosinase molecules expressed in the absence of these proteins, as is the case in our transfectants, represent such partially folded species.
Our work provides insights into the Ag processing of proteins synthesized in the ER and establishes a link between quality control in the ER and epitope processing. The mechanisms of Ag processing of membrane-associated proteins have been poorly investigated in comparison with cytosolic proteins, although a significant proportion of all cellular proteins is translocated into the lumen of the ER and all enveloped viruses produce excess glycoproteins in the ER of the host cells (43).
Unfolded/Misfolded proteins that accumulate as a consequence of cellular stress, viral infection, or cellular transformation may be important sources for enhanced presentation of peptides. Hypoxia, which commonly occurs in cancer cells, has been shown to promote accumulation of unfolded/misfolded proteins (44). Physiological responses associated with the accumulation of unfolded proteins have been shown to occur as a consequence of dysregulated glucose homeostasis in diabetes (45), and as a consequence of excess protein production in cells infected with hepatitis C or B viruses (46, 47). Other cellular insults, such as perturbation in redox status or in calcium homeostasis, deprivation of glucose, have been associated with the accumulation of unfolded/misfolded proteins in the ER (44). Finally, misfolding and aberrant ER retention of tyrosinase due to dysregulation of pH homeostasis is frequently observed in metastatic melanomas, which are often amelanotic even when they express WT tyrosinase (26, 27). In the present work, we have shown that this leads to enhanced Ag presentation. It remains to be demonstrated that these other physiological stressors have a similar effect.
Disclosures
The authors have no financial conflict of interest.
Footnotes
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.
1 This work was supported by U.S. Public Health Service Grants AI20963 and AI33134 to V.H.E.
2 Address correspondence and reprint requests to Dr. Victor H. Engelhard, Beirne Carter Center for Immunology Research, Box 801386, University of Virginia School of Medicine, Charlottesville, VA 22908. E- mail address: vhe@virginia.edu
3 Abbreviations used in this paper: WT, wild type; BFA, brefeldin A; Endo H, endoglycosidase H; ER, endoplasmic reticulum; L-Dopa, L-dopamine; LAMP, lysosome-associated membrane protein; LLnL, N-acetyl-Leu-Leu-norleucinal.
Received for publication August 18, 2004. Accepted for publication December 3, 2004.
References
Schubert, U., L. C. Anton, J. Gibbs, C. C. Norbury, J. W. Yewdell, J. R. Bennink. 2000. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature 404:770.
Reits, E. A., J. C. Vos, M. Gromme, J. Neefjes. 2000. The major substrates for TAP in vivo are derived from newly synthesized proteins. Nature 404:774.
Khan, S., R. de Giuli, G. Schmidtke, M. Bruns, M. Buchmeier, B. M. van den, M. Groettrup. 2001. Cutting edge: neosynthesis is required for the presentation of a T cell epitope from a long-lived viral protein. J. Immunol. 167:4801.
Yewdell, J.. 2002. To DRiP or not to DRiP: generating peptide ligands for MHC class I molecules from biosynthesized proteins. Mol. Immunol. 39:139.
Shastri, N., S. Schwab, T. Serwold. 2002. Producing nature’s gene-chips: the generation of peptides for display by MHC class I molecules. Annu. Rev. Immunol. 20:463.
Tevethia, S. S., M. J. Tevethia, A. J. Lewis, V. B. Reddy, S. M. Weissman. 1983. Biology of simian virus-40 (Sv40) transplantation antigen (Trag). IX. Analysis of Trag in mouse cells synthesizing truncated Sv40 large T-antigen. Virology 128:319.
Glotzer, M., A. W. Murray, M. W. Kirschner. 1991. Cyclin is degraded by the ubiquitin pathway. Nature 349:132.
Rogers, S., R. Wells, M. Rechsteiner. 1986. Amino-acid sequences common to rapidly degraded proteins: the pest hypothesis. Science 234:364.
Murakami, Y., S. Matsufuji, T. Kameji, S. Hayashi, K. Igarashi, T. Tamura, K. Tanaka, A. Ichihara. 1992. Ornithine decarboxylase is degraded by the 26S proteasome without ubiquitination. Nature 360:597.
Varshavsky, A.. 1992. The N-end rule. Cell 69:725
Townsend, A. R., J. Bastin, K. Gould, G. Brownlee, M. Andrew, B. Coupar, D. Boyle, S. Chan, G. Smith. 1988. Defective presentation to class I-restricted cytotoxic T lymphocytes in vaccinia-infected cells is overcome by enhanced degradation of antigen. J. Exp. Med. 168:1211.
Grant, E. P., M. T. Michalek, A. L. Goldberg, K. L. Rock. 1995. Rate of antigen degradation by the ubiquitin-proteasome pathway influences MHC class I presentation. J. Immunol. 155:3750.
Sijts, A. J., I. Pilip, E. G. Pamer. 1997. The Listeria monocytogenes-secreted p60 protein is an N-end rule substrate in the cytosol of infected cells: implications for major histocompatibility complex class I antigen processing of bacterial proteins. J. Biol. Chem. 272:19261.
Tobery, T., R. F. Siliciano. 1999. Cutting edge: induction of enhanced CTL-dependent protective immunity in vivo by N-end rule targeting of a model tumor antigen. J. Immunol. 162:639.
Princiotta, M. F., D. Finzi, S. B. Qian, J. Gibbs, S. Schuchmann, F. Buttgereit, J. R. Bennink, J. W. Yewdell. 2003. Quantitating protein synthesis, degradation, and endogenous antigen processing. Immunity 18:343.
Goth, S., V. Nguyen, N. Shastri. 1996. Generation of naturally processed peptide/MHC class I complexes is independent of the stability of endogenously synthesized precursors. J. Immunol. 157:1894.
Anton, L. C., U. Schubert, I. Bacik, M. F. Princiotta, P. A. Wearsch, J. Gibbs, P. M. Day, C. Realini, M. C. Rechsteiner, J. R. Bennink, J. W. Yewdell. 1999. Intracellular localization of proteasomal degradation of a viral antigen. J. Cell Biol. 146:113.
Realini, C., S. W. Rogers, M. Rechsteiner. 1994. KEKE motifs: proposed roles in protein-protein association and presentation of peptides by MHC class I receptors. FEBS Lett. 348:109.
Slingluff, C. L., Jr. 1996. Tumor antigens and tumor vaccines: peptides as immunogens. Semin. Surg. Oncol. 12:446.
Petrescu, S. M., N. Branza-Nichita, G. Negroiu, A. J. Petrescu, R. A. Dwek. 2000. Tyrosinase and glycoprotein folding: roles of chaperones that recognize glycans. Biochemistry 39:5229
Jimbow, K., J. S. Park, F. Kato, K. Hirosaki, K. Toyofuku, C. Hua, T. Yamashita. 2000. Assembly, target-signaling and intracellular transport of tyrosinase gene family proteins in the initial stage of melanosome biogenesis. Pigm. Cell. Res. 13:222
Oetting, W. S.. 2000. The tyrosinase gene and oculocutaneous albinism type 1 (OCA1): a model for understanding the molecular biology of melanin formation. Pigm. Cell. Res. 13:320.
Halaban, R., E. Cheng, S. Svedine, R. Aron, D. N. Hebert. 2001. Proper folding and endoplasmic reticulum to Golgi transport of tyrosinase are induced by its substrates, DOPA and tyrosine. J. Biol. Chem. 276:11933.
Toyofuku, K., I. Wada, R. A. Spritz, V. J. Hearing. 2001. The molecular basis of oculocutaneous albinism type 1 (OCA1); sorting failure and degradation of mutant tyrosinases results in a lack of pigmentation. Biochem. J. 355:259.
Luckey, C. J., G. M. King, J. A. Marto, S. Venketeswaran, B. F. Maier, V. L. Crotzer, T. A. Colella, J. Shabanowitz, D. F. Hunt, V. H. Engelhard. 1998. Proteasomes can either generate or destroy MHC class I epitopes: evidence for nonproteasomal epitope generation in the cytosol. J. Immunol. 161:112.
Berson, J. F., D. W. Frank, P. A. Calvo, B. M. Bieler, M. S. Marks. 2000. A common temperature-sensitive allelic form of human tyrosinase is retained in the endoplasmic reticulum at the nonpermissive temperature. J. Biol. Chem. 275:12281.
Slingluff, C. L., T. A. Colella, L. Thompson, D. D. Graham, J. C. A. Skipper, J. A. Caldwell, L. Brinkerhoff, D. J. Kittlesen, D. H. Deacon, C. Oei, et al 2000. Melanomas with concordant loss of multiple melanocytic differentiation proteins: immune escape that may be overcome by targeting unique or undefined antigens. Cancer Immunol. Immunother. 48:661.
Luckey, C. J., J. A. Marto, M. Partridge, E. Hall, F. M. White, J. D. Lippolis, J. Shabanowitz, D. F. Hunt, V. H. Engelhard. 2001. Differences in the expression of human class I MHC alleles and their associated peptides in the presence of proteasome inhibitors. J. Immunol. 167:1212.
Lippincott-Schwartz, J., L. C. Yuan, J. S. Bonifacino, R. D. Klausner. 1989. Rapid redistribution of Golgi proteins in the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56:801.
Saraste, J., G. E. Palade, M. G. Farquhar. 1986. Temperature-sensitive steps in the transport of secretory proteins through the Golgi complex in exocrine pancreatic cells. Proc. Natl. Acad. Sci. USA 83:6425.
Lippincott-Schwartz, J., J. S. Bonifacino, L. C. Yuan, R. D. Klausner. 1988. Degradation from the endoplasmic reticulum: disposing of newly synthesized proteins. Cell 54:209.
Skipper, J. C. A., R. C. Hendrickson, P. H. Gulden, V. Brichard, A. Van Pel, Y. Chen, J. Shabanowitz, T. Wolfel, C. L. Slingluff, T. Boon, et al 1996. An HLA-A2 restricted tyrosinase antigen on melanoma cells results from post-translational modification and suggests a novel processing pathway for membrane proteins. J. Exp. Med. 183:527.
Sousa, M., A. J. Parodi. 1995. The molecular basis for the recognition of misfolded glycoproteins by the UDP-Glc:glycoprotein glucosyltransferase. EMBO J. 14:4196.
Molinari, M., V. Calanca, C. Galli, P. Lucca, P. Paganetti. 2003. Role of EDEM in the release of misfolded glycoproteins from the calnexin cycle. Science 299:1397.
Oda, Y., N. Hosokawa, I. Wada, K. Nagata. 2003. EDEM as an acceptor of terminally misfolded glycoproteins released from calnexin. Science 299:1394.
Kaufman, R. J.. 1999. Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene transcriptional and translational controls. Genes Dev. 13:1211
Kaufman, R. J.. 2002. Orchestrating the unfolded protein response in health and disease. J. Clin. Invest. 110:1389.
Harding, H. P., D. Ron. 2002. Endoplasmic reticulum stress and the development of diabetes: a review. Diabetes 51:(Suppl. 3):S455.
Meyer, M., W. H. Caselmann, V. Schluter, R. Schreck, P. H. Hofschneider, P. A. Baeuerle. 1992. Hepatitis B virus transactivator MHBst: activation of NF-B, selective inhibition by antioxidants and integral membrane localization. EMBO J. 11:2991.
Pavio, N., P. R. Romano, T. M. Graczyk, S. M. Feinstone, D. R. Taylor. 2003. Protein synthesis and endoplasmic reticulum stress can be modulated by the hepatitis C virus envelope protein E2 through the eukaryotic initiation factor 2 kinase PERK. J. Virol. 77:3578.(Marina Ostankovitch, Vale)
Short-lived protein translation products have been proposed to be the principal substrates that enter the class I MHC processing and presentation pathway. However, the biochemical nature of these substrates is poorly defined. Whether the major processing substrates are misfolded full-length proteins, or alternatively, aberrantly initiated or truncated polypeptides still remains to be addressed. To examine this, we used melanoma in which one-third of wild-type tyrosinase molecules were correctly folded and localized beyond the Golgi, while the remainder were present in the endoplasmic reticulum in an unfolded/misfolded state. Increasing the efficiency of tyrosinase folding using chemical chaperones led to a reduction in the level of substrate available to the proteasome and decreased the expression of a tyrosinase-derived epitope. Conversely, in transfectants expressing tyrosinase mutants that are completely misfolded, both proteasome substrate and epitope presentation were significantly enhanced. Proteasome substrate availability was a consequence of misfolding and not simply due to retention in the endoplasmic reticulum. Thus, the extent of folding/misfolding of a full-length protein is an important determinant of the level of epitope presentation.
Introduction
The nature of the protein substrates that enter the class I MHC Ag processing pathway has been the subject of intense investigation over several years. A significant percentage of cellular proteins with masses greater than 45 kDa is degraded by the proteasome within 60 min of synthesis (1). Newly synthesized proteins have also been shown to be an important source of substrates for the TAP (2) and to be the source of one virally derived epitope (3). It has been proposed that these newly synthesized and rapidly degraded proteins or peptides could correspond to defective protein translation products that were aberrantly initiated or prematurely terminated (4). In support of this, several epitopes have been identified as short polypeptides translated from alternative reading frames, some of which span exon-intron boundaries or lie entirely within introns, and others of which result from the use of nonstandard initiation codons (summarized in Ref.5). It has also been shown that the level of epitope presentation is enhanced using truncated forms of a protein (6). However, it is not clear what fraction of the total protein economy of the cell they represent.
An alternative source of rapidly degraded substrates for class I Ag processing is full-length proteins that are sampled: 1) after they become terminally misfolded; 2) while they are still in the process of folding; or 3) because they are short-lived based on N-end rule or other sequences that directly target them for degradation (7, 8, 9). Substitution of the N-terminal residue of proteins with N-end destabilizing residues, such as Arg, decreases the t1/2 of proteins by increasing the level of ubiquination, which in turn augments interaction with and degradation by the proteasome (10). In several studies, this maneuver led to an increased level of epitope presentation to T cells (11, 12, 13, 14, 15), while in others it had no effect (11, 16). However, because ubiquination is dependent on the N-terminal residue rather than protein-folding state, these studies did not address the importance of protein-folding state as a determinant of class I Ag presentation. The insertion of a sequence consisting largely of repeated Lys and Glu residues (KEKE) has also been shown to favor proteasome degradation and epitope presentation (15, 17). Although it has been proposed that the KEKE sequence directly augments interaction with the proteasome (18) or that it induces misfolding (15, 17), no definitive data distinguishing these possibilities have been published. Interestingly, however, this work also demonstrated that a protein with an intermediate rate of degradation generated class I MHC-restricted epitopes more efficiently than a more rapidly degraded form (15), and also demonstrated the apparent existence of two distinct pools of substrates, giving rise to epitopes based on kinetic measurements. However, the nature of these substrates was not directly examined in this or other work performed to date.
Taken together, these studies demonstrate that enhanced targeting of model protein Ags to the proteasome frequently augments the expression of class I-restricted epitopes, but may also make no difference or actually cause diminished expression. They also suggest that this could depend on the exact nature of the protein substrate for degradation, of which little is directly known. Indeed, while there is strong evidence that newly synthesized translation products are a significant source of class I MHC-restricted epitopes in virally infected cells (1, 2, 3, 15), there are little direct data that support this hypothesis for epitopes derived from stably expressed endogenous proteins (2). As has been pointed out elsewhere, it still remains to be established that full-length proteins that are either terminally misfolded or in the process of folding are significant substrates for the class I Ag processing pathway (4, 5). It is also not clear whether such proteins would be sampled after terminal misfolding or at early stages during the folding process.
One meaning to assess the impact of protein folding on the level of epitope presentation is to compare wild-type (WT)3 proteins with naturally occurring mutant forms that are known to misfold, or to alter other conditions that selectively affect the folding of individual proteins within cells. Tyrosinase, which represents a robust source of class I MHC-restricted peptides recognized by melanoma-specific T cells (19), offers a useful model for this evaluation. Tyrosinase is a membrane-associated N-linked glycoprotein that is synthesized and folds in the endoplasmic reticulum (ER), and is subsequently transported via the trans Golgi network to melanosomes (20, 21). Many natural variants of tyrosinase associated with type I oculocutaneous albinism fail to fold (22, 23, 24, 25), and unfolded or misfolded WT tyrosinase molecules frequently accumulate in melanoma cells as a consequence of abnormal cellular acidification (26, 27). These unfolded/misfolded mutant tyrosinase or WT tyrosinase molecules are completely retained in the ER through prolonged interaction with the ER chaperones calnexin and calreticulin rather than exiting to melanosomes, and are subsequently retrotranslocated to the cytosol and degraded by the proteasome (20, 24, 26, 27, 28, 29). In contrast, the tyrosinase substrates L-dopamine (L-Dopa) and L-tyrosine act as chemical chaperones to enhance the folding of WT tyrosinase molecules and their export from the ER to the Golgi (23). Work from our laboratory established that the Tyr369 epitope presented by HLA-A*0201 was produced after retrotranslocation of tyrosinase from the lumen of the ER to the cytosol, degradation by the proteasome, and reimport of peptides into the ER by TAP (28, 30). In this study, we used WT tyrosinase molecules and tyrosinase mutants that have been demonstrated to fail to fold (23, 24, 25, 26, 31), together with chemical chaperones, to evaluate the impact of folding of full-length molecules on the processing and presentation of the Tyr369 epitope.
Materials and Methods
Tyrosinase gene constructs and transfectants
The WT and R402Q tyrosinase genes (gift from M. Marks, University of Pennsylvania, Philadelphia, PA) were subcloned in the vector pEF6 (Invitrogen Life Technologies). The A206T mutation was introduced into pEF6 WT tyrosinase using the QuickChange XL site-directed mutagenesis kit (Stratagene). The melanoma cell line DM331 (tyrosinase–, HLA-A*0201+) (32) was transfected with plasmids encoding WT, R402Q, or A206T tyrosinase using Fugene (Roche Diagnostics). Bulk stable transfectant lines were generated by selection with blasticidin (10 µg/ml) (Invitrogen Life Technologies) and cloned. All experiments using stable transfectants were conducted on cells plated at 5000 cells/cm2 and grown for 3 days. Folding of tyrosinase was induced by incubation of cells for 3 days in medium supplemented with 20 µM L-Dopa (Sigma-Aldrich) or 400 µM L-tyrosine (Sigma-Aldrich). To ensure that the pH of the medium and the concentration of L-Dopa remain the same during the 3-day experiment, the medium and medium supplemented with L-Dopa or L-tyrosine were replaced every day. Retention of tyrosinase in the ER was induced in medium supplemented with 5 µg/ml brefeldin A (BFA) (Sigma-Aldrich) or by incubation at 20°C for 4 h. For experiments analyzing proteasome degradation, transfectants were incubated in presence of 50 µM N-acetyl-Leu-Leu-norleucinal (LLnL) (Calbiochem) for 3.5–4 h.
Cytofluorometry
Expression of tyrosinase was measured by intracellular staining after permeabilization for 30 min with BD Perm/wash (BD Pharmingen) and sequential addition of C19 (goat anti-tyrosinase; Santa Cruz Biotechnology), biotinylated donkey anti-goat IgG (Santa Cruz Biotechnology), and allophycocyanin-conjugated streptavidin (BD Pharmingen). Expression of HLA-A*0201 was measured by surface staining using the mAb CR11-351 and FITC-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratories). Data were acquired on a FACS and analyzed using CellQuest software (BD Pharmingen).
Immunoblotting and analysis of maturation of N-linked glycans
Cell pellets (106 cells) were solubilized in 100 µl of buffer containing 10 mM Tris-HCl, pH 7.5, 0.5% deoxycholate, 1% Igepal, 5 mM EDTA, 4 mM PMSF, 10 µg/ml aprotinin, 10 µM pepstatin A, 10 µg/ml leupeptin, and 100 µM iodoacetamide, and centrifuged at 21,000 x g for 30 min at 4°C. For analysis of maturation of N-glycans, proteins from 50 µl of supernatant were precipitated with 400 µl of cold acetone at –20°C overnight. The precipitates were resuspended in 0.5% SDS and 0.1 M 2-ME and boiled for 5 min. After cooling on ice for 2 min, the samples were digested with 2.5 mU endoglycosidase H (Endo H; Roche Diagnostics), according to the manufacturer’s recommendations. Alternatively, proteins from supernatant were directly boiled in sample buffer for 5 min. A total of 5 x 105 cell equivalents per lane was separated on 10% SDS-PAGE gels, transferred to NitroPure nitrocellulose (Osmonics), and blocked in 5% nonfat dried milk in PBS with 0.05% Tween 20. Blots were probed overnight with C19 goat anti-tyrosinase, washed, probed with HRP-conjugated donkey anti-goat IgG, and developed according to the Amersham ECL protocol (Amersham Biosciences). Films were scanned using Scan Maker III (Microtek), and images were processed using Adobe Photoshop software. Bands were quantified using NIH Image 1.62f software.
Immunofluorescence microscopy
Transfectants were grown on coverslips, washed with PBS, fixed for 10 min at room temperature with 4% paraformaldehyde in PBS, and permeabilized with BD Perm/wash (BD Pharmingen). Tyrosinase was detected either with C19 goat (Santa Cruz Biotechnology) or T3.11 mouse anti-tyrosinase Ab with similar results (25). The cells were incubated for 1 h with combinations of anti-tyrosinase Ab, rabbit anti-calnexin (StressGen Biotechnologies), or mouse anti-lysosome-associated membrane protein (LAMP)1 (BD Pharmingen). These primary Abs were detected with Alexa 488 anti-goat and Alexa 594 anti-rabbit or anti-mouse conjugates (Molecular Probes). Samples were mounted on glass slides with Vectashield (Vector Laboratories), visualized using an Olympus confocal microscope, and processed with Adobe Photoshop 6.0 software.
T cell assay
Tyr369 epitope-specific T cell lines were generated from C57BL/6 mice expressing a chimeric HLA-A*0201/H2Dd MHC class I molecule and maintained, as described (28). Resting T cells (50,000) were incubated with the indicated number of transfectants or Tyr369 peptide-pulsed DM331 cells for 16 h, and the level of IFN- secreted was measured by ELISA (eBiosciences).
Results
Maturation, folding, and intracellular localization of WT and mutant forms of tyrosinase in an amelanotic melanoma cell line
To examine the importance of protein folding and subcellular localization on the expression of tyrosinase-derived epitopes, we generated transfectants of the amelanotic melanoma cell DM331 with cDNA expression vectors encoding either WT human tyrosinase, or the tyrosinase mutants R402Q and A206T that have been previously shown to be retained in the ER in consequence of intrinsic misfolding (31). DM331 does not express detectable tyrosinase based on Western blot with anti-tyrosinase Abs (32) or PCR (our unpublished observations), but does express HLA-A*0201. Stable transfectant clones expressing similar levels of each tyrosinase construct were identified by intracellular staining using the Ab C19, which recognizes a C-terminal epitope of tyrosinase. Tyrosinase maturation in these transfectants was evaluated by digestion of postnuclear supernatants of cell lysates with Endo H, separation by SDS-PAGE, and immunoblotting with C19. In cells expressing WT tyrosinase, quantitation of total signal intensities of the bands by densitometry demonstrated that about one-third of the molecules were insensitive to Endo H, indicating that they had moved out of the ER and beyond the cis-/medial Golgi (Fig. 1A). Similar values have been obtained in five independent experiments involving four different cloned transfectants (data not shown). This observation is consistent with others documenting a relatively low level of mature tyrosinase in melanoma cells as opposed to melanocytes (26). However, in agreement with previous work (23), the amount of mature tyrosinase was increased markedly by growth of the transfectants in medium containing L-Dopa (Fig. 1A) or elevated levels of L-tyrosine (data not shown). In contrast, R402Q and A206T tyrosinase molecules were completely Endo H sensitive (Fig. 1A). L-Dopa (Fig. 1A) and L-tyrosine (data not shown) did not induce any significant maturation of A206T tyrosinase, and only a very minor increase in the maturation of R402Q. Using immunocytochemistry and confocal microscopy, WT tyrosinase was found to partially colocalize with calnexin, a marker of the ER, and with LAMP1, a marker of late endosomes, lysosomes, and melanosomes (Fig. 1B). In contrast, R402Q and A206T tyrosinase molecules colocalized strongly with calnexin and failed to colocalize with LAMP1. Taken together, these observations confirmed that a significant portion of WT tyrosinase molecules in the DM331 melanoma transfectants had matured beyond the cis-/medial Golgi, reflecting their capacity to fold, while R402Q and A206T tyrosinase molecules were completely immature, retained in the ER, indicating their inability to fold.
FIGURE 1. A significant fraction of WT tyrosinase is mature and localized in late endosomes, whereas R402Q and A206T tyrosinase are immature and retained in the ER. A, Cloned transfectants of DM331 melanoma cells stably expressing WT, R402Q, or A206T tyrosinase were grown for 3 days in the presence or absence of L-Dopa, lysed, digested with Endo H, analyzed by SDS-PAGE, and immunoblotted with the anti-tyrosinase Ab C19. Endo H-insensitive mature tyrosinase migrates at 70 kDa, while Endo H-sensitive immature tyrosinase migrates at 60 kDa. The membrane was then stripped and immunoblotted with anti-calnexin. Numbers representing the total signal intensities as quantified by scanning densitometry density and corresponding to mature and immature WT tyrosinase are indicated. Results represented in A and Figs. 2, 5, and 6, experiment 2, were obtained from experiments conducted in parallel and using duplicate aliquots of melanoma cells. B, Immunofluorescence and confocal microscopy of cloned transfectants of DM331 melanoma cells stably expressing WT, R402Q, or A206T tyrosinase using anti-tyrosinase and anti-calnexin or anti-LAMP1.
Proteasome-dependent degradation of R402Q and A206T tyrosinase is enhanced compared with WT tyrosinase
We next evaluated whether unfolded/misfolded R402Q and A206T full-length tyrosinase molecules were preferential substrates for the proteasome in comparison with WT tyrosinase. To do so, we quantified the amounts of degradation intermediates accumulating after incubation of the transfectants with the proteasome inhibitor LLnL. Treatment with 100 µM LLnL for 4 h, which we have previously shown to inhibit 90–97% of different proteasome activities (33), induced the accumulation of 60-kDa tyrosinase-derived bands in WT, R402Q, and A206T tyrosinase transfectants (Fig. 2). These bands have previously been shown to correspond to either unglycosylated and full-length (26, 29) or glycosylated and partially proteolyzed (29) tyrosinase molecules, many of which have been retrotranslocated to the cytosol (28, 29). Bands of the same mass are stabilized by LLnL (26, 28, 29), lactacystin (26, 29), MG132 (26), and epoxomycin (our unpublished results), and are apparently generated by a protease distinct from the proteasome. Quantitation by scanning densitometry established that in presence of LLnL, the relative amounts of the 60-kDa bands compared with those of full-length tyrosinase were 2.5- to 3-fold higher for R402Q and A206T tyrosinase than for WT tyrosinase. These results demonstrated that the amount of R402Q and A206T tyrosinase available to and degraded by the proteasome over 4 h is greater than for WT tyrosinase.
FIGURE 2. Proteasome degradation substrates from R402Q and A206T tyrosinase are present at relatively higher levels than those from WT tyrosinase. Cloned transfectants of DM331 melanoma cells stably expressing WT, R402Q, or A206T tyrosinase were incubated in presence or absence of LLnL for 4 h. Cell lysates were analyzed by SDS-PAGE and immunoblotted with the C19 anti-tyrosinase Ab. Films were exposed for 0.5–1 min to detect the broad bands centered at 70 kDa, which represent immature and mature intact forms of tyrosinase, and for 3 min to detect the bands at 60 kDa, which represent proteasome-sensitive deglycosylated or partially proteolyzed tyrosinase molecules. Numbers below each band represent the total signal intensities as quantified by scanning densitometry density, and the ratio of the intensities of the 60- to 70-kDa species is indicated. Results represented in this figure and Fig. 1A are obtained from experiments conducted in parallel and correspond to the same aliquots of melanoma cells.
Tyr369 epitope expression levels on melanoma transfectants expressing equivalent levels of full-length WT, R402Q, A206T tyrosinase
To evaluate whether intrinsically unfolded/misfolded full-length tyrosinase molecules were a preferential proteasome substrate for epitope presentation, we next evaluated the expression levels of the HLA-A*0201-restricted Tyr369 epitope on the different DM331 melanoma transfectants by measuring their ability to stimulate IFN- secretion from Tyr369-specific T cell lines. First, we established that the assay conditions would accurately reflect differences in epitope expression on transfectants expressing different levels of either WT tyrosinase or HLA-A*0201. Using melanoma transfectants that expressed similar levels of HLA-A*0201 molecules, we observed a correlation between the tyrosinase expression level in different transfectants and the secretion of IFN- (Fig. 3A, Expt. 1). Similarly, in transfectants expressing similar levels of tyrosinase, we observed a correlation between IFN- secretion and the level of HLA-A*0201 expression (Fig. 3A, Expt. 2). In parallel experiments, a 2-fold difference in the secretion of IFN- was correlated to a 2-log difference in the concentration of Tyr369 synthetic peptide used to pulse untransfected DM331 melanoma cells (Fig. 3B). Finally, we demonstrated that incubation of transfectants expressing WT tyrosinase with exogenous Tyr369 synthetic peptide led to a further increase in IFN- secretion (Fig. 3C). These experiments showed that this assay was sensitive to quantitative variations in Tyr369 epitope expression resulting from either tyrosinase, HLA-A*0201, or exogenous peptide expression differences.
FIGURE 3. The level of expression of WT tyrosinase or HLA-A*0201 determines the level of presentation of Tyr369. A, Top panels, Expression of WT tyrosinase in three independent clones of transfected DM331 was measured by intracellular staining with anti-tyrosinase and flow cytometry. Thin black line, WTcl1a or WTcl1b; thin gray line, WTcl2; thick line, WTcl8a or WTcl8b; gray fill, untransfected DM331. The arithmetic mean fluorescence intensity (MFI) of each distribution after subtraction of the DM331 value is shown in the second row of panels. Middle panels, Expression of HLA-A*0201 in the same clones was measured by surface staining with anti-HLA-A*0201 and flow cytometry. Bottom panels, The presentation of Tyr369 was evaluated as the level of IFN- secreted by Tyr369-specific T cell lines after incubation with the indicated DM331 transfectant clones. B, T cell recognition of DM331 transfectant clones expressing different levels of WT tyrosinase and the same number of untransfected DM331 cells pulsed overnight with different concentrations of exogenous Tyr369 peptide was evaluated by IFN- secretion. C, T cell recognition of untransfected DM331 cells and a DM331 transfectant expressing WT tyrosinase before or after overnight pulsing with 1 µM Tyr369 peptide.
We next compared the presentation levels of the Tyr369 epitope by transfectants expressing WT, R402Q, and A206T tyrosinase molecules. Using short-term bulk populations in which the mean levels of tyrosinase expression and the percentages of tyrosinase+ cells were similar, cells expressing R402Q tyrosinase stimulated the release of 1.8-fold more IFN- from Tyr369-specific T cells than did cells expressing the WT molecule (Fig. 4, Expt. 1). Similarly, when two different Tyr369-specific T cell lines were stimulated with long-term cloned transfectants that expressed uniform and comparable levels of WT and R402Q tyrosinase, the cells expressing R402Q stimulated the release of 1.8- to 2.1-fold more IFN- (Fig. 4, Expt. 2). This numerical range has been observed in five different experiments involving three distinct cloned transfectants expressing WT or R402Q tyrosinase molecules. Finally, a transfectant expressing A206T tyrosinase still stimulated a 1.6-fold higher level of IFN- secretion than did a transfectant expressing almost one-third more WT tyrosinase (Fig. 4, Expt. 3). In three independent experiments involving the same transfectant clones, the release of IFN- was 1.4- to 3.5-fold higher in the presence of the transfectant clone expressing A206T than in presence of the clone expressing WT molecules, with a mean of 2.2-fold. Collectively, the results of Figs. 2–4 indicate that the expression of unfolded or misfolded tyrosinase is associated with enhanced degradation by the proteasome and leads to a significantly augmented presentation of the processed Tyr369 epitope at the cell surface.
FIGURE 4. Presentation of Tyr369 from R402Q and A206T tyrosinase is enhanced compared with that of WT tyrosinase. Analyses were conducted using bulk populations (Expt. 1) or clones (Expts. 2 and 3) of DM331 transfectants expressing WT, R402Q, or A206T tyrosinase. Upper panels, Expression of tyrosinase was measured by intracellular staining and flow cytometry (thin line, WT tyrosinase transfectant or clone; thick line, R402Q in Expts. 1 and 2 and A206T in Expt. 3; gray fill, untransfected DM331). The background fluorescence observed for DM331 has been subtracted in the bar graphs of MFI. Middle panels, Expression of HLA-A*0201 was measured by surface staining with anti-HLA-A*0201 and flow cytometry. Bottom panels, The presentation of Tyr369 was evaluated as the level of IFN- secreted by Tyr369-specific T cell lines after incubation with transfectants. In experiment 1, DM331 cells were transfected with either WT or R402Q tyrosinase, and 100,000 or 30,000 short-term bulk transfectants in which the percentages of tyrosinase+ cells were similar were incubated with tyrosinase-specific T cell line 2. In experiments 2 and 3, 5,000 cloned transfectants were incubated with specific T cell lines 1 or 2, and the levels of secretion of IFN- shown are after subtraction of the background level of IFN- secreted in presence of untransfected DM331.
Folding, maturation, and export of WT tyrosinase from the ER diminish tyrosinase processing and presentation of Tyr369
As shown in Fig. 1 and elsewhere (23), growth of melanoma cells with the tyrosinase substrates L-Dopa or L-tyrosine enhanced folding and exit from the ER of WT tyrosinase, while having a minimal effect on the tyrosinase folding mutants. Consistent with this, growth of WT tyrosinase transfectants in L-Dopa for 3 days substantially reduced the relative amount of 60-kDa tyrosinase degradation intermediates that could be detected after a 4-h incubation of the cells with the proteasome inhibitor LLnL (Fig. 5, lanes 2 and 4). In contrast, the relative amount of 60-kDa tyrosinase fragments was actually somewhat higher in A206T transfectants treated with LLnL and L-Dopa compared with those treated with LLnL alone (Fig. 5, lanes 6 and 8). We therefore tested the hypothesis that addition of L-Dopa would lead to a reduction in the processing and presentation of Tyr369 from WT tyrosinase, but not from A206T tyrosinase. In three independent experiments, two of which are shown in Fig. 6A, we observed a significant decrease in the presentation of Tyr369 by transfectant expressing WT tyrosinase after growth in L-Dopa or L-tyrosine. However, no significant change in the presentation of Tyr369 by a transfectant expressing A206T tyrosinase was detected under the same conditions. The difference in the effect of L-Dopa on WT and A206T presentation of Tyr369 from WT and A206T tyrosinase was statistically significant (p < 0.005, Fig. 6B). Altogether, these results demonstrate that increased folding and export of tyrosinase from the ER to the Golgi decreased the fraction of tyrosinase molecules undergoing degradation and also decreased the magnitude of Tyr369 epitope presentation.
FIGURE 5. Induction of folding of WT tyrosinase with chemical chaperones decreases the amount of proteasome degradation substrates. Cloned transfectants of DM331 melanoma cells stably expressing WT or A206T tyrosinase were grown for 3 days in the presence or absence of L-Dopa, and then further incubated for 4 h in the presence or absence of LLnL. Cell lysates were analyzed by SDS-PAGE and immunoblotted with anti-tyrosinase. Films were exposed for 0.5–1 min to detect the broad bands centered at 70 kDa, which represent immature and mature intact forms of tyrosinase, and for 3 min to detect the bands at 60 kDa, which represent proteasome-sensitive deglycosylated or partially proteolyzed tyrosinase molecules. Numbers below each band represent the total signal intensities as quantified by scanning densitometry density, and the ratio of the intensities of the 60- to 70-kDa species is indicated. Results shown in this figure and Fig. 1A are obtained from experiments conducted in parallel and correspond to the same aliquots of melanoma cells.
FIGURE 6. Induction of folding of WT tyrosinase decreases presentation of Tyr369. A, Cloned transfectants of DM331 melanoma cells stably expressing WT or A206T tyrosinase were grown for 3 days in the absence () or presence of L-Dopa () or L-tyrosine (). Upper panels, Expression of tyrosinase was measured by intracellular staining and flow cytometry. The background fluorescence observed for DM331 has been subtracted in the bar graphs of MFI. Middle panels, Expression of HLA-A*0201 was measured by surface staining and flow cytometry. ND, not done. Bottom panels, Presentation of Tyr369 was evaluated as the level of IFN- secreted by a Tyr369-specific T cell line after incubation with cloned transfectants. The levels of secretion of IFN- shown are after subtraction of the background level of IFN- secreted in presence of untransfected DM331. Results shown in this figure, experiment 2, and Figs. 5 and 1A are obtained from experiments conducted in parallel and correspond to the same aliquots of melanoma cells. B, Percentage of inhibition of IFN- released by Tyr369-specific T cells as a consequence of melanoma cell culture in the presence of L-Dopa, calculated from the data of A, experiments 1 and 2. The p value was calculated using two-sample Student‘s t test and Minitab 13.0 software.
Proteasome-dependent degradation of tyrosinase is increased by misfolding regardless of ER retention and by ER retention separately of intrinsic misfolding
The results above suggested that the failure of tyrosinase to fold in the ER led directly to an increase in epitope expression based on enhanced interaction with the ER quality control system. However, they did not exclude the possibility that simple enhanced representation of the protein in the ER, regardless of its folding state, might also lead to enhanced processing and presentation. To assess the impact of misfolding on tyrosinase processing independent of differences in ER retention, we used BFA (34) or incubation of cells at 20°C (35, 36) to block anterograde transport of either WT or R402Q tyrosinase from the ER to the Golgi, and compared the accumulation of tyrosinase degradation intermediates after inhibition of proteasome function with LLnL. In BFA-treated cells, we found that the relative amount of 60-kDa tyrosinase degradation intermediates for R402Q tyrosinase was still 2.5-fold higher than that for WT tyrosinase (Fig. 7). A similar result was obtained after incubation of cells at 20°C. This demonstrates that the extent of tyrosinase misfolding determines retrotranslocation from the ER independently of its impact on retention in the ER. In addition, we observed that the relative amount of the 60-kDa degradation intermediates from WT tyrosinase was 3-fold higher in presence of BFA and LLnL than in presence of LLnL alone, while, as expected, the addition of BFA in the presence of LLnL only slightly increased the amount of 60-kDa tyrosinase degradation intermediates for R402Q tyrosinase (Fig. 7). This suggested that either properly folded WT tyrosinase was a substrate for retrotranslocation because of its retention, or residence in the ER led to an unfolding of properly folded molecules. Therefore, while we demonstrated that the extent of tyrosinase misfolding influences extraction from the ER independently of localization in the ER, we also provide evidence that retention in the ER independently of intrinsic misfolding favors exit from the ER and degradation by the proteasome.
FIGURE 7. Proteasome substrate levels generated from WT and R402Q tyrosinase retained in the ER. Clones of transfectants expressing WT or R402Q tyrosinase were incubated at 37°C in presence or absence of BFA for 30 min, or at 20°C, and then further incubated in the presence or absence of LLnL for 3.5 h. Cell lysates were analyzed by SDS-PAGE and immunoblotted with anti-tyrosinase. Films were exposed for 0.5–1 min to detect the broad bands centered at 70 kDa, which represent immature and mature intact forms of tyrosinase, and for 3 min to detect the bands at 60 kDa, which represent proteasome-sensitive deglycosylated or partially proteolyzed tyrosinase molecules. Numbers below each band represent the total signal intensities as quantified by scanning densitometry density, and the ratio of the intensities of the 60- to 70-kDa species is indicated.
Discussion
Short-lived proteins and defective protein translation products, which include aberrantly initiated, prematurely terminated, and unfolded or misfolded full-length proteins, have been proposed to be the principal substrates that enter the class I MHC processing and presentation pathway (1, 2, 3, 15). However, it is not clear to what extent proteasome substrates are represented by aberrant translation products that are inherently unable to fold, or alternatively, full-length proteins that are targeted for degradation either because of a conformational abnormality, because they take a longer time to fold properly, or because they contain signals that confer a short t1/2 (4). Many studies have shown that Ag presentation can be altered by the introduction of sequences that enhance proteasome targeting without regard for protein folding state, or by the creation of truncated proteins (6, 11, 12, 13, 14, 15). However, no study to date has directly evaluated the impact of protein folding on the efficiency of Ag processing and presentation, nor evaluated whether full-length proteins, as opposed to aberrant translation products, are substrates for processing (4, 5).
Using tyrosinase as a model, we have now provided direct insight into these issues. This is due in part to the fact that the C19 Ab detects an epitope located in the C-terminal tail of tyrosinase, which is located on the opposite side of the membrane from the lumenal domain whose folding determines tyrosinase maturation and degradation. By using this Ab, we thus evaluate full-length tyrosinase molecules, as opposed to prematurely terminated or aberrantly initiated products. In addition, the folding behavior of WT tyrosinase and the tyrosinase mutants R402Q and A206T has been well established in previous work. Unfolded or misfolded forms of WT tyrosinase are commonly found in melanoma cells (26) as a consequence of abnormal cellular acidification (27) and the absence of other melanocyte differentiation proteins (37, 38). Only about one-third of WT tyrosinase expressed in melanoma cells folds properly, while the remainder fails to fold, but can be rescued in presence of the chemical chaperones L-Dopa or L-tyrosine (23). R402Q tyrosinase molecules are completely retained in the ER at 37°C, but can be partially rescued and exported to the Golgi at low temperature or by addition of chemical chaperones (27, 31). A206T tyrosinase is a mutation leading to oculocutaneous albinism. It is completely retained in the ER through prolonged interaction with the chaperone calnexin (24), and cannot be rescued by low temperature incubation or chemical chaperones. Neither of these molecules exhibits enzymatic activity in the unfolded state, strongly suggesting that misfolding is responsible for their complete retention in the ER. Finally, N-glycans are attached to tyrosinase during synthesis, undergo structural modification as a consequence of folding and export from the ER, and are removed during retrotranslocation as a prelude to degradation by the proteasome. Based on our ability to distinguish among these forms, we have shown a direct relationship between the levels of the unfolded/misfolded and reverse translocated forms of full-length tyrosinase and the amount of epitope presented.
In the study presented in this work, we have directly assessed the influence of misfolding/unfolding in the case of a full-length protein, regardless of its impact on ubiquination, on the efficiency of Ag processing and presentation. We showed that increasing the folding efficiency of WT tyrosinase using chemical chaperones led to a consequent decrease in the amount of immature WT tyrosinase in the ER and proteasome-sensitive tyrosinase degradation intermediates correlated with a decrease in Tyr369 presentation. Conversely, in transfectants expressing R402Q and A206T, which are unable to fold and are completely retained in the ER, the proteasome-dependent degradation of tyrosinase molecules and the presentation level of Tyr369 were significantly enhanced compared with transfectants expressing WT tyrosinase. To exclude the possibility that the availability of tyrosinase molecules for degradation and epitope presentation was strictly a function of representation in the ER independent of folding state, we evaluated the impact of misfolding on degradation after retention of R402Q and WT tyrosinase in the ER using BFA or low temperature. Proteasome substrate availability was a consequence of misfolding state, and not simply due to retention in the ER. Our results demonstrate that the extent of folding or misfolding of full-length protein is an important determinant of the final level of epitope presentation.
We cannot fully exclude the possibility that prematurely terminated forms of tyrosinase, which we would not detect using C19, represent an additional source of Ag for class I presentation. However, to the extent that these protein forms contain a full-length lumenal domain whose folding is augmented by L-Dopa or L-tyrosine, and disrupted by R402Q and A206T mutations, they are functionally equivalent to full-length molecules in their contribution to Ag processing and presentation. Conversely, we would expect that aberrantly initiated tyrosinase molecules would remain largely unfolded/misfolded, even in the presence of L-Dopa or L-tyrosine, so that the differences in epitope presentation shown in this study would not have been observed. In addition, aberrantly initiated WT tyrosinase molecules would fail to enter the ER because they lack a signal sequence, and would be synthesized in the cytosol. We have previously shown that cells expressing a truncated form of tyrosinase in the cytosol present two forms of the Tyr369 epitope, one containing a genetically encoded Asn371 and the other containing a deamidated Asp at the same position (28). In contrast, cells expressing full-length tyrosinase express only the latter epitope (28, 39). Collectively, these results indicate that aberrantly initiated tyrosinase molecules are an insignificant source of class I MHC-restricted epitopes.
Somewhat surprisingly, we also found that enforced retention of WT tyrosinase in the ER using BFA enhanced proteasome-dependent degradation independently of intrinsic misfolding, although the level of proteasome substrate observed was still lower than that arising from R402Q tyrosinase under normal conditions. BFA collapses the cis-/medial Golgi into the ER to create a single compartment (34). This structure would then contain not only immature unfolded or misfolded tyrosinase, but also some molecules that are sufficiently well folded to have moved to the Golgi. The processing of immature unfolded/misfolded tyrosinase molecules should not be increased by enforced ER retention because these forms are normally retained in the ER. Properly folded tyrosinase molecules are also improbable sources for this enhanced retrotranslocation and degradation. Previous studies have shown that correct folded glycoproteins are not reglucosylated by the uridine diphosphate-glucose glucosyl transferase (40), and consequently, they do not interact with the calnexin/ER degradation-enhancing -mannosidase-like protein molecular complex that has recently been shown to play an important role in retrotranslocation from the ER (41, 42). Instead, we favor the hypothesis that the increase in retrotranslocation and degradation observed when ER retention is enforced results from the processing of a form of tyrosinase that is partially folded, and has a lower, but still significant probability of interaction with uridine diphosphate-glucose glucosyl transferase and calnexin. We hypothesize that this species is normally able to be exported from the ER to the Golgi, but is retrotranslocated and degraded by the proteasome when it is forced to be retained in the ER. Tyrosinase normally complexes with and is stabilized by other melanosomal proteins (37). It is possible that tyrosinase molecules expressed in the absence of these proteins, as is the case in our transfectants, represent such partially folded species.
Our work provides insights into the Ag processing of proteins synthesized in the ER and establishes a link between quality control in the ER and epitope processing. The mechanisms of Ag processing of membrane-associated proteins have been poorly investigated in comparison with cytosolic proteins, although a significant proportion of all cellular proteins is translocated into the lumen of the ER and all enveloped viruses produce excess glycoproteins in the ER of the host cells (43).
Unfolded/Misfolded proteins that accumulate as a consequence of cellular stress, viral infection, or cellular transformation may be important sources for enhanced presentation of peptides. Hypoxia, which commonly occurs in cancer cells, has been shown to promote accumulation of unfolded/misfolded proteins (44). Physiological responses associated with the accumulation of unfolded proteins have been shown to occur as a consequence of dysregulated glucose homeostasis in diabetes (45), and as a consequence of excess protein production in cells infected with hepatitis C or B viruses (46, 47). Other cellular insults, such as perturbation in redox status or in calcium homeostasis, deprivation of glucose, have been associated with the accumulation of unfolded/misfolded proteins in the ER (44). Finally, misfolding and aberrant ER retention of tyrosinase due to dysregulation of pH homeostasis is frequently observed in metastatic melanomas, which are often amelanotic even when they express WT tyrosinase (26, 27). In the present work, we have shown that this leads to enhanced Ag presentation. It remains to be demonstrated that these other physiological stressors have a similar effect.
Disclosures
The authors have no financial conflict of interest.
Footnotes
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.
1 This work was supported by U.S. Public Health Service Grants AI20963 and AI33134 to V.H.E.
2 Address correspondence and reprint requests to Dr. Victor H. Engelhard, Beirne Carter Center for Immunology Research, Box 801386, University of Virginia School of Medicine, Charlottesville, VA 22908. E- mail address: vhe@virginia.edu
3 Abbreviations used in this paper: WT, wild type; BFA, brefeldin A; Endo H, endoglycosidase H; ER, endoplasmic reticulum; L-Dopa, L-dopamine; LAMP, lysosome-associated membrane protein; LLnL, N-acetyl-Leu-Leu-norleucinal.
Received for publication August 18, 2004. Accepted for publication December 3, 2004.
References
Schubert, U., L. C. Anton, J. Gibbs, C. C. Norbury, J. W. Yewdell, J. R. Bennink. 2000. Rapid degradation of a large fraction of newly synthesized proteins by proteasomes. Nature 404:770.
Reits, E. A., J. C. Vos, M. Gromme, J. Neefjes. 2000. The major substrates for TAP in vivo are derived from newly synthesized proteins. Nature 404:774.
Khan, S., R. de Giuli, G. Schmidtke, M. Bruns, M. Buchmeier, B. M. van den, M. Groettrup. 2001. Cutting edge: neosynthesis is required for the presentation of a T cell epitope from a long-lived viral protein. J. Immunol. 167:4801.
Yewdell, J.. 2002. To DRiP or not to DRiP: generating peptide ligands for MHC class I molecules from biosynthesized proteins. Mol. Immunol. 39:139.
Shastri, N., S. Schwab, T. Serwold. 2002. Producing nature’s gene-chips: the generation of peptides for display by MHC class I molecules. Annu. Rev. Immunol. 20:463.
Tevethia, S. S., M. J. Tevethia, A. J. Lewis, V. B. Reddy, S. M. Weissman. 1983. Biology of simian virus-40 (Sv40) transplantation antigen (Trag). IX. Analysis of Trag in mouse cells synthesizing truncated Sv40 large T-antigen. Virology 128:319.
Glotzer, M., A. W. Murray, M. W. Kirschner. 1991. Cyclin is degraded by the ubiquitin pathway. Nature 349:132.
Rogers, S., R. Wells, M. Rechsteiner. 1986. Amino-acid sequences common to rapidly degraded proteins: the pest hypothesis. Science 234:364.
Murakami, Y., S. Matsufuji, T. Kameji, S. Hayashi, K. Igarashi, T. Tamura, K. Tanaka, A. Ichihara. 1992. Ornithine decarboxylase is degraded by the 26S proteasome without ubiquitination. Nature 360:597.
Varshavsky, A.. 1992. The N-end rule. Cell 69:725
Townsend, A. R., J. Bastin, K. Gould, G. Brownlee, M. Andrew, B. Coupar, D. Boyle, S. Chan, G. Smith. 1988. Defective presentation to class I-restricted cytotoxic T lymphocytes in vaccinia-infected cells is overcome by enhanced degradation of antigen. J. Exp. Med. 168:1211.
Grant, E. P., M. T. Michalek, A. L. Goldberg, K. L. Rock. 1995. Rate of antigen degradation by the ubiquitin-proteasome pathway influences MHC class I presentation. J. Immunol. 155:3750.
Sijts, A. J., I. Pilip, E. G. Pamer. 1997. The Listeria monocytogenes-secreted p60 protein is an N-end rule substrate in the cytosol of infected cells: implications for major histocompatibility complex class I antigen processing of bacterial proteins. J. Biol. Chem. 272:19261.
Tobery, T., R. F. Siliciano. 1999. Cutting edge: induction of enhanced CTL-dependent protective immunity in vivo by N-end rule targeting of a model tumor antigen. J. Immunol. 162:639.
Princiotta, M. F., D. Finzi, S. B. Qian, J. Gibbs, S. Schuchmann, F. Buttgereit, J. R. Bennink, J. W. Yewdell. 2003. Quantitating protein synthesis, degradation, and endogenous antigen processing. Immunity 18:343.
Goth, S., V. Nguyen, N. Shastri. 1996. Generation of naturally processed peptide/MHC class I complexes is independent of the stability of endogenously synthesized precursors. J. Immunol. 157:1894.
Anton, L. C., U. Schubert, I. Bacik, M. F. Princiotta, P. A. Wearsch, J. Gibbs, P. M. Day, C. Realini, M. C. Rechsteiner, J. R. Bennink, J. W. Yewdell. 1999. Intracellular localization of proteasomal degradation of a viral antigen. J. Cell Biol. 146:113.
Realini, C., S. W. Rogers, M. Rechsteiner. 1994. KEKE motifs: proposed roles in protein-protein association and presentation of peptides by MHC class I receptors. FEBS Lett. 348:109.
Slingluff, C. L., Jr. 1996. Tumor antigens and tumor vaccines: peptides as immunogens. Semin. Surg. Oncol. 12:446.
Petrescu, S. M., N. Branza-Nichita, G. Negroiu, A. J. Petrescu, R. A. Dwek. 2000. Tyrosinase and glycoprotein folding: roles of chaperones that recognize glycans. Biochemistry 39:5229
Jimbow, K., J. S. Park, F. Kato, K. Hirosaki, K. Toyofuku, C. Hua, T. Yamashita. 2000. Assembly, target-signaling and intracellular transport of tyrosinase gene family proteins in the initial stage of melanosome biogenesis. Pigm. Cell. Res. 13:222
Oetting, W. S.. 2000. The tyrosinase gene and oculocutaneous albinism type 1 (OCA1): a model for understanding the molecular biology of melanin formation. Pigm. Cell. Res. 13:320.
Halaban, R., E. Cheng, S. Svedine, R. Aron, D. N. Hebert. 2001. Proper folding and endoplasmic reticulum to Golgi transport of tyrosinase are induced by its substrates, DOPA and tyrosine. J. Biol. Chem. 276:11933.
Toyofuku, K., I. Wada, R. A. Spritz, V. J. Hearing. 2001. The molecular basis of oculocutaneous albinism type 1 (OCA1); sorting failure and degradation of mutant tyrosinases results in a lack of pigmentation. Biochem. J. 355:259.
Luckey, C. J., G. M. King, J. A. Marto, S. Venketeswaran, B. F. Maier, V. L. Crotzer, T. A. Colella, J. Shabanowitz, D. F. Hunt, V. H. Engelhard. 1998. Proteasomes can either generate or destroy MHC class I epitopes: evidence for nonproteasomal epitope generation in the cytosol. J. Immunol. 161:112.
Berson, J. F., D. W. Frank, P. A. Calvo, B. M. Bieler, M. S. Marks. 2000. A common temperature-sensitive allelic form of human tyrosinase is retained in the endoplasmic reticulum at the nonpermissive temperature. J. Biol. Chem. 275:12281.
Slingluff, C. L., T. A. Colella, L. Thompson, D. D. Graham, J. C. A. Skipper, J. A. Caldwell, L. Brinkerhoff, D. J. Kittlesen, D. H. Deacon, C. Oei, et al 2000. Melanomas with concordant loss of multiple melanocytic differentiation proteins: immune escape that may be overcome by targeting unique or undefined antigens. Cancer Immunol. Immunother. 48:661.
Luckey, C. J., J. A. Marto, M. Partridge, E. Hall, F. M. White, J. D. Lippolis, J. Shabanowitz, D. F. Hunt, V. H. Engelhard. 2001. Differences in the expression of human class I MHC alleles and their associated peptides in the presence of proteasome inhibitors. J. Immunol. 167:1212.
Lippincott-Schwartz, J., L. C. Yuan, J. S. Bonifacino, R. D. Klausner. 1989. Rapid redistribution of Golgi proteins in the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56:801.
Saraste, J., G. E. Palade, M. G. Farquhar. 1986. Temperature-sensitive steps in the transport of secretory proteins through the Golgi complex in exocrine pancreatic cells. Proc. Natl. Acad. Sci. USA 83:6425.
Lippincott-Schwartz, J., J. S. Bonifacino, L. C. Yuan, R. D. Klausner. 1988. Degradation from the endoplasmic reticulum: disposing of newly synthesized proteins. Cell 54:209.
Skipper, J. C. A., R. C. Hendrickson, P. H. Gulden, V. Brichard, A. Van Pel, Y. Chen, J. Shabanowitz, T. Wolfel, C. L. Slingluff, T. Boon, et al 1996. An HLA-A2 restricted tyrosinase antigen on melanoma cells results from post-translational modification and suggests a novel processing pathway for membrane proteins. J. Exp. Med. 183:527.
Sousa, M., A. J. Parodi. 1995. The molecular basis for the recognition of misfolded glycoproteins by the UDP-Glc:glycoprotein glucosyltransferase. EMBO J. 14:4196.
Molinari, M., V. Calanca, C. Galli, P. Lucca, P. Paganetti. 2003. Role of EDEM in the release of misfolded glycoproteins from the calnexin cycle. Science 299:1397.
Oda, Y., N. Hosokawa, I. Wada, K. Nagata. 2003. EDEM as an acceptor of terminally misfolded glycoproteins released from calnexin. Science 299:1394.
Kaufman, R. J.. 1999. Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene transcriptional and translational controls. Genes Dev. 13:1211
Kaufman, R. J.. 2002. Orchestrating the unfolded protein response in health and disease. J. Clin. Invest. 110:1389.
Harding, H. P., D. Ron. 2002. Endoplasmic reticulum stress and the development of diabetes: a review. Diabetes 51:(Suppl. 3):S455.
Meyer, M., W. H. Caselmann, V. Schluter, R. Schreck, P. H. Hofschneider, P. A. Baeuerle. 1992. Hepatitis B virus transactivator MHBst: activation of NF-B, selective inhibition by antioxidants and integral membrane localization. EMBO J. 11:2991.
Pavio, N., P. R. Romano, T. M. Graczyk, S. M. Feinstone, D. R. Taylor. 2003. Protein synthesis and endoplasmic reticulum stress can be modulated by the hepatitis C virus envelope protein E2 through the eukaryotic initiation factor 2 kinase PERK. J. Virol. 77:3578.(Marina Ostankovitch, Vale)