Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M706237200 on October 19, 2007

J. Biol. Chem., Vol. 283, Issue 1, 244-254, January 4, 2008
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
283/1/244    most recent
M706237200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kario, E.
Right arrow Articles by Navon, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kario, E.
Right arrow Articles by Navon, A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

N-Linked Glycosylation Does Not Impair Proteasomal Degradation but Affects Class I Major Histocompatibility Complex Presentation*Formula

Edith Kario{ddagger}12, Boaz Tirosh§1, Hidde L. Ploegh, and Ami Navon, An incumbent of the Recanati career development chair of cancer research{ddagger}3

From the {ddagger}Department of Biological Regulation, The Weizmann Institute of Science, Rehovot 76100, Israel, the §Department of Pharmacology, School of Pharmacy, Faculty of Medicine, The Hebrew University, Jerusalem 91120, Israel, and the Whitehead Institute of Biomedical Research, Cambridge, Massachusetts 02142

Received for publication, July 30, 2007 , and in revised form, October 10, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The addition of N-linked glycans to nascent polypeptides occurs cotranslationally in the endoplasmic reticulum (ER). For many proteins the state of the glycans serves as an indicator, which allows the ER quality control system to monitor the conformation of polypeptides upon folding. Proteins that fail to fold in the ER are often dislocated to the cytoplasm, where they are subjected to proteasomal degradation. Although the addition of N-linked glycans occurs within the ER, non-lysosomal removal of the glycans occurs in the cytosol by the action of peptide N-glycanase (PNGase). In this study, we investigated the interplay between PNGase action and proteasomal degradation of ER misfolded proteins (i.e. whether PNGase acts prior to or following proteasomal degradation). Interestingly, we found that glycan removal from N-terminally extended peptides modulates the presentation of class I major histocompatibility complex-restricted epitopes. Our findings provide direct evidence that the proteasome is capable of degrading glycoproteins without prior removal of their glycans. This degradation is independent of either the identity of the glycosylated protein or the type and number of N-linked glycans it harbors. We also captured and characterized glycopeptides generated following proteasomal degradation of RNaseB. Although the carbohydrate moiety reduced the variability of the degradation products that include the glycosylated residue (local effect), the overall global digestion pattern of RNaseB was unaffected. Together with earlier findings by others, our data support a model in which PNGase may act both upstream and downstream to proteasomal degradation and demonstrates its important role in class I major histocompatibility complex antigen presentation.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Degradation of proteins in eukaryotes and bacteria is catalyzed primarily by large multimeric proteases, whose function is ATP-dependent (1–4). In eukaryotes these enzymes (most prominent of which is the proteasome) are essential for the rapid destruction of many regulatory proteins such as cyclins. The proteasome plays a key role also in the immune system by producing antigenic peptides suitable for presentation by class I major histocompatibility complex (MHC)4 molecules (5–9). In addition, many damaged polypeptides and proteins that fail to fold or assemble properly in the endoplasmic reticulum (ER) are also digested by the proteasome.

Many proteins that are targeted to the ER undergo cotranslational glycosylation as part of their maturation (10). In the ER an elaborate quality control system assists proteins in their folding and prevents misfolded proteins from reaching their final destination. Two mechanisms have evolved to prevent the accumulation of misfolded proteins. One, termed the "unfolded protein response," (UPR), involves the activation of an ER-to-nucleus signaling pathway that results in up-regulation of ER chaperones and augments its folding capacity (11–13). However, proteins cannot always be refolded, and therefore a second mechanism has evolved to destroy polypeptides that failed to pass the ER quality control system. Both mechanisms, the unfolded protein response and degradation of misfolded ER proteins, are coordinately regulated (14–17).

Degradation of ER misfolded proteins entails their dislocation from the ER to the cytoplasm, followed by substrate ubiquitination and proteasomal degradation (18). Because many of such misfolded proteins are glycosylated, the presence of the glycan may influence their proteasomal degradation. Although modification of proteins by N-linked glycans occurs only in the ER, non-lysosomal removal of the sugar moieties takes place in the cytosol. The removal of N-linked glycans is executed by the cytosolic enzyme peptide N-glycanase (PNGase). PNGase hydrolyzes the β-glycoamide bond between the asparagine residue and the GlcNAc group at the reducing end of the oligosaccharide, resulting in the release of a free glycan with the concomitant conversion of the asparagine to aspartic acid (19, 20).

Several studies addressed the exact step at which PNGase removes the glycans from glycosylated substrates in the course of their degradation. One model based on earlier studies suggests that PNGase is recruited to the 26 S proteasome and removes the glycan from the polypeptide prior to its degradation by the proteasome (21–24). In support of this model, both yeast and mammalian PNGase were shown to interact directly with the proteasome via the 19 S subunit S4, or indirectly through RAD23 (21, 22, 25–27). In addition, during the degradation of class I MHC heavy chains, which is mediated by the human cytomegalovirus-encoded glycoproteins US2 and US11, a deglycosylated intermediate accumulates in the cytosol when the 26 S proteasome is inhibited (24). In accord with this model yeast and mouse PNGase deglycosylate unfolded proteins in vitro; however, their properly folded counterparts do not undergo PNGase-mediated deglycosylation (28, 29). Presumably, the substrate of PNGase may be at least partially unfolded as it emerges from the ER. Alternatively, PNGase action may be coupled to the unfolding step that occurs during recognition of the ubiquitinated substrate by the ATPases at the base of the 19 S proteasome regulatory particle (21, 26). The observation that the N-linked glycan acts as a recognition module for the SCF (Fbx2) E3 ubiquitin ligase complex (30, 31) places PNGase downstream of ubiquitination and upstream of proteasomal degradation.

Both the yeast Saccharomyces cerevisiae and mammalian cells have a single gene encoding for cytoplasmic N-glycanase activity. Thus, if PNGase must act prior to proteasomal degradation of glycosylated substrates, blocking its activity should result in accumulation of glycoproteins in the cytoplasm, which must be deleterious to the cells. Surprisingly, yeast deleted for the Png1 gene are viable and present no growth defect or other noticeable phenotype (20). In fact, the only phenotype detected so far in Png1-deficient yeast includes a moderate decrease in the rate of CPY* degradation (a glycoprotein decorated with four N-linked glycans, commonly used as a model substrate for degradation of ER misfolded proteins) (20). Likewise, mammalian cells, in which PNGase expression was silenced using siRNA, showed only a minor delay in the degradation of class I MHC heavy chains, which carry a single N-linked glycan (23). The lack of an overt phenotype in yeast deficient in PNGase or mammalian cells in which PNGase expression was silenced by small interference RNA suggests an alternative mechanism in which PNGase activity is not a prerequisite for proteasomal degradation. Accordingly, PNGase may remove the glycan moiety from glycosylated digestion products following proteasomal degradation rather than act on the intact glycoprotein prior to its digestion. In agreement with this alternative, inhibition of PNGase by Z-VAD-fmk did not prevent proteasomal degradation of glycosylated substrates in living cells (19). Nonetheless, the requirement of PNGase for proteasomal degradation has never been assessed directly.

Here we monitored the ability of the proteasome (20 S and 26 S) to degrade N-linked glycoproteins without deglycosylation. We found that in vitro the proteasome digests proteins decorated with single or multiple glycans in a fashion that is independent of PNGase activity. Interestingly, although the carbohydrate moiety did not interfere with degradation per se, it altered the repertoire of the digestion products containing the glycosylation site. We further demonstrate that the presence of a glycan near a class I MHC epitope affects its presentation, which is suppressed when PNGase is inhibited.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Generation of EL4 Cells Expressing TCR{alpha} Mutants—Mutants of TCR{alpha} (Fig. 1 and supplemental Fig. S1) were cloned into pMIG, a retroviral vector, which harbors an IRES-GFP element. Retroviral infections were performed as previously described (32). GFP-positive EL4 cells were sorted to 100% purity and cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (HyClone Laboratories, Logan, UT), 2 mM glutamine, 50 units/ml penicillin, 50 µg/ml streptomycin, 50 µM β-mercaptoethanol, 25 mM HEPES, 1x nonessential amino acids, and 1 mM sodium pyruvate.

Pulse-Chase and Precipitation Experiments—After starvation of 45 min in methionine/cysteine-free Dulbecco's modified Eagle's medium, EL4 cells were metabolically labeled with 500 µCi/ml [35S]methionine/cysteine (PerkinElmer Life Sciences) at 37 °C for 30 min. Pulse-chase, cell lysis, and precipitation were performed as described previously (33). The precipitates were analyzed by SDS-PAGE followed by fluorography. EndoH was purchased from New England Biolabs and used according to the manufacture's recommendations.

Staining of EL4 Cells—EL4 cells were stained with 25D1.16 monoclonal antibody (kindly provided by Dr. Ron Germain, National Institutes of Health), which recognizes the H-2Kb/SIINFEKL complex (34). Alexa Fluor 647-conjugated anti-mouse (Molecular Probes) was used as a secondary antibody. Live cells were gated based on their light scattering characteristics.

OT-I T Cell Activation Assay—Naïve CD8 T cells were isolated from spleens of OT-I transgenic mice bred onto RAG-1-/- background. OT-I is a transgenic TCR (V2/V5) specific for SIINFEKL (an ovalbumin-derived peptide) restricted by H-2Kb. To obtain naïve OT-I CD8 T cells, cell suspensions from spleens were depleted for erythrocytes by LCK lysis. 106 OT-I cells were incubated for 16 h at 37 °C with 3 x 105 EL4 cells with or without 50 µM of the SIINFEKL peptide, in a final volume of 300 µl. The cells were then washed with ice-cold fluorescence-activated cell sorting staining buffer (phosphate-buffered saline, 0.5% bovine serum albumin, and 0.02% sodium azide), stained with phycoerythrin-anti-CD69 (BD Biosciences) and analyzed by flow cytometry. OT-I T cells were gated based on their smaller size than EL4 cells and the lack of GFP expression.

Preparation of Reduced and Alkylated Biotin-labeled Glycoproteins—10 mg of protein (RNaseB, ovalbumin, or transferrin receptor purchased from Sigma) dissolved in 20 mM Tris, pH 7.5, 5 mM EDTA, 20 mM DTT, 8 M urea were incubated for 30 min at 55 °C. The reaction mixture was then loaded onto a Superdex S-30 gel-filtration column (Amersham Biosciences) equilibrated with 20 mM sodium acetate, pH 6.0, 5 mM EDTA, 3 M guanidine, 8 M urea. For biotinylation, peak fractions were concentrated and incubated in the dark for 12 h at room temperature with 5 mg of biotin-polyethylene-oxide iodoacetamide (Pierce). The reaction mixture was then separated on a Superdex S-30 gel-filtration column equilibrated with 20 mM Tris, pH 7.5. Peak fractions containing the biotinylated protein were pulled and used for proteasomal degradation experiments. PNGaseF (New England Biolabs) was used for deglycosylation according to the manufacture's recommendations.

RNaseB Alkylation by Iodoacetamide—Twenty milligrams of RNaseB was dissolved in 50 mM Tris, pH 7.5. The protein was loaded on a mono S high-performance liquid chromatography column, and eluted with a linear gradient of 0–300 mM NaCl. Fractions containing RNaseB were dialyzed against 50 mM Tris, pH 7.5. The purified protein was then denatured and reduced as described for RNaseB-biotin, following by alkylation with 40 mg of iodoacetamide. The alkylated RNaseB (denoted RNaseB-IAA) was then purified as described for RNaseB-biotin and used for the degradation experiments, which were analyzed by mass spectrometry.

Expression and Purification of TCR{alpha} from HEK293 TREX Cells—The Flp-In expression system (Invitrogen) was used to stably express TCR{alpha}-6 His. TCR{alpha}-6 His from pCDNA3.1 expression vector (35) was subcloned into the pCDNA5 FRT/TO mammalian expression vector (Invitrogen). Subconfluent HEK293 TREX cells grown in Dulbecco's modified Eagle's medium were cotransfected with pCDNA5 FRT/TO TCR{alpha}-6 His and pOG44 (encoding the FLP recombinase) expression vectors. Forty-eight hours after transfection 100 µg/ml hygromycin B (Invitrogen) was added to the growth medium, and positive clones were selected. To induce TCR{alpha}-6 His expression, hygromycin-resistant clones were grown for 24 h in the presence of 1 µg/ml tetracycline. To purify the TCR{alpha}-6 His, cells were resuspended in lysis buffer (50 mM Hepes, pH 7.5, 10% glycerol, 150 mM NaCl, 1% Triton X-100) and preincubated on ice for 15 min. The lysate was supplemented with urea to a final concentration of 5 M. Following 15-min incubation on ice, the lysates were cleared by centrifugation for 15 min at 14,000 rpm, filtered, and loaded onto a nickel-nitrilotriacetic acid column (Amersham Biosciences). After thoroughly washing the column, TCR{alpha}-6 His was eluted with 0.5 M imidazole, dialyzed for 24 h against 20 mM Hepes, pH 7.5, and assayed by immunoblotting with anti-TCR{alpha} antibodies (Santa Cruz Biotechnology, Santa Cruz, CA).

Degradation of Glycosylated Proteins by the Proteasome: Preparation of Substrates—Prior to the 26 S degradation assays, OVA-biotin and TfR-biotin were preincubated with 6 M guanidine followed by brief desalting using a Superdex S-30 gel-filtration column. Fractions containing the biotinylated proteins were used in degradation assays immediately. The purified TCR{alpha} was boiled, chilled briefly on ice, and used without delay in the digestion reactions. For 20 S degradation reactions, purified TCR{alpha} and OVA-biotin were incubated for 30 min in 50 mM HEPES, pH 7.5, 6 M guanidine, 15 mM DTT at 55 °C, followed by desalting of the reaction mixture on G-50 (TCR{alpha}) or G-75 (OVA-biotin) spin columns (Amersham Biosciences). TfR-biotin was boiled and chilled briefly prior to usage.

Proteasomal Degradation Reaction Conditions and Analysis—RNaseB-biotin, TfR-biotin, OVA-biotin, and the purified TCR{alpha} were incubated at 37 °C with 50 mM HEPES pH 7.5, 1 mM DTT with or without 0.018% SDS, and 20 S proteasome purified from rabbit muscles. For degradation of substrates by the mammalian 26 S proteasome or the archaeal 20 S-PAN complex, the glycoprotein was incubated at 37 °C with 50 mM HEPES, pH 7.5, 5 mM MgCl2, 30 mM KCl, 1 mM DTT with or without 2 mM ATP and 26 S proteasome purified from rabbit muscles or 20 S-PAN complexes that were expressed and purified from bacteria. In all of the degradation reactions, the molar ratio of proteasomes to substrates was 1:1000. The concentration of the 20 S and the 26 S proteasome used was 1 nM, with 1 µM substrate. Samples from the reaction mixture were removed at different time intervals and subjected to immunoblot analysis with an anti-biotin horseradish peroxidase-conjugated antibody (Vector) or anti TCR{alpha} antibodies (Santa Cruz Biotechnology). The immunoblots were quantified using the Quantity One program (Bio-Rad). The data were fitted using linear or exponential decay (Y = Y0 + A1e-(x - x0)/t1) fitting functions (Origin laboratory).

Glycoprotein Staining with Concanavalin A—Staining of membranes with concanavalin A was used to detect glycosylated proteins transferred onto polyvinylidene difluoride membranes. The membrane was preincubated at room temperature for 30 min in wash solution (50 mM Tris, pH 7.5, 100 mM NaCl, 0.1% Tween 20) supplemented with 2.5% bovine serum albumin. Concanavalin A (Sigma, 20 µg/ml) was added for 30 min. The membrane was then washed three times with wash solution, followed by incubation with horseradish peroxidase (Sigma, 20 µg/ml) in wash solution supplemented with 2.5% bovine serum albumin for 30 min at room temperature. The membrane was washed three times, and the reactive bands were detected using ECL reagents (Amersham Biosciences).

Purification of Yeast PNGase—E. coli BL21(DE3) transformed with pET22-yPNGase 6 His expression vector were grown in LB medium (2 liters) for ~6 h at 37 °C with constant shaking until the A600 of the culture reached 0.6. Isopropyl 1-thio-β-D-galactopyranoside was added to a final concentration of 0.5 mM. The culture was grown for additional 3 h, and the bacteria were recovered by centrifugation. The pellet was resuspended in 50 mM HEPES, pH 7.5, 1 mM β-mercaptoethanol and sonicated for 3 min on ice. The sonicated lysate was then clarified by centrifugation. The clear lysate was filtered and loaded onto a nickel-nitrilotriacetic acid column (Amersham Biosciences) and eluted with 0.5 M imidazole. The eluted fractions were dialyzed against 50 mM HEPES, pH 7.5, 2 mM DTT for 24 h. The dialyzed eluant was concentrated and loaded onto a Superdex S-75 gel-filtration column equilibrated with 20 mM Tris, pH 7.5, 150 mM NaCl, 10% glycerol, 1 mM DTT, 5 mM EDTA. The peak fractions were pooled, concentrated, diluted with 50% glycerol, and stored at -20 °C.

Mass Spectrometry Analysis—Following degradation by the 26 S proteasome the peptides were resolved by capillary reversed-phase chromatography and eluted with linear gradients of 5 to 95% of acetonitrile with 0.1% formic acid in water. Mass spectrometry was performed by an ion-trap mass spectrometer (Orbitrap, Finnigan, San Jose, CA) in a positive mode using repetitively full mass spectrometry scan followed by collision induces dissociation of the three most dominant ions selected from the first mass spectrometry scan. The mass spectrometry data were analyzed using the Sequest software (J. Eng and J. Yates, University of Washington and Finnigan, San Jose) searching the RNaseB sequence. The peptides were identified with Xcorr of at least 3 {Delta}CN above 0.1 and were manually evaluated as well. In addition, the peptides were analyzed using MALDI TOFTOF mass spectrometer (4700, Applied Biosystems), enabling a better coverage of the differences between the samples.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
The Presence of a Sugar Moiety Near a Class I MHC Epitope Reduces the Efficiency of Its Presentation—A small fraction of the proteasomal degradation products is transported into the ER and loaded onto class I MHC molecules. The peptides suitable for presentation must be of certain length (8–10 amino acids) and composed such that they can be presented by a specific type of class I MHC molecule. These strict requirements markedly limit the number of peptides generated from each antigenic protein that are suitable for presentation. Although most of the peptides generated are further degraded by cytosolic peptidases, only ~35% of the peptides emerge from the proteasome as mature epitopes, correct in size for presentation, or as N-extended versions (36–38). The N-extended epitopes can be further trimmed to the correct size for presentation by amino-peptidases, either in the cytosol or within the ER. On the other hand, the identity of the C terminus is determined by proteasomal cleavage. Because the average half-life of a peptide in the cytosol is only ~5s, >99% of the proteasomal degradation products suitable for presentation are destroyed before they are translocated into the ER (38, 39). For example, whereas ovalbumin is composed of 386 residues, only two ovalbumin-derived epitopes are presented: the major epitope SIINFEKL and a minor epitope KVVRFDKL (40). During degradation of ovalbumin, products containing the SIINFEKL epitope, suitable for presentation, are destroyed >90% of the time (36). To examine the effects of PNGase activity on the generation and presentation of class I MHC-restricted epitopes, we investigated whether the presence of a glycan moiety in proximity to a class I MHC-restricted epitope would alter its presentation. To assay effective presentation, we inserted the ovalbumin-derived SIINFEKL epitope in-frame into the backbone of TCR{alpha}. The SIINFEKL epitope was demonstrated previously to be suitable for the study of N-extended epitopes (36, 41). Because the target glycosylation consensus sequence is Asn-X-Ser/Thr (where X stands for any amino acid but proline), it is possible to insert the SIINFEKL sequence two residues downstream of the glycosylation site such that its serine residue serves as part of the consensus sequence.

When TCR{alpha} is expressed in the absence of a cognate β chain, it is routed to proteasome-mediated degradation (35, 42). TCR{alpha} contains a 20-amino acids signal sequence at its N terminus that is cleaved upon maturation, a large extracellular domain decorated with four N-linked glycans (Asn-44, -168, -202, and -213), a short transmembrane domain, and a C-terminal intracellular domain (Fig. S1). SIINFEKL was inserted after the first glycosylation site Asn-44 (mutant denoted S1). As a control we generated an S1 mutant without a signal peptide (S1{Delta}SP), which is not inserted into the ER and thus would not carry any glycans. In addition, we generated a mutant in which the Asn residue of TCR{alpha} adjacent to the SIINFEKL sequence was converted to Asp (S1-N44D). This mutant lacks the glycosylation site adjacent to the SIINFEKL sequence. The different TCR{alpha} constructs (supplemental Fig. S1) were cloned into pMIG, a retroviral expression vector that contains an IRES-GFP element. We transduced EL4 cells (mouse thymoma cells) with pMIG vectors that encode the different TCR{alpha} mutants, and GFP-positive cells were isolated by cell sorting (Fig. 1, A and B). We verified that the different TCR{alpha}-SIINFEKL mutants are expressed in EL4 cells and processed in the expected manner (Fig. 1C). The cells were pulse-labeled with [35S]methionine/cysteine, lysed, and TCR{alpha} was immunoprecipitated. To verify that all the mutants are glycosylated as expected, a portion of the immunoprecipitates was treated with endoglycosidase H (EndoH), to cleave the high mannose N-linked glycans. Samples of control and EndoH-treated immunoprecipitates were separated by SDS-PAGE, followed by fluorography. The untreated samples resolved into two forms of TCR{alpha}, which differ by their molecular weight (Fig. 1C). A higher molecular weight form was observed in cells expressing either the WT TCR{alpha} or the S1 mutant, and a lower molecular weight form was detected in cells in which TCR{alpha} contains a point mutation of the asparagine residue (mutant S1-N44D). In the latter case, the TCR{alpha} appeared smaller due to the loss of one glycosylation site. However, when the different immunoprecipitates were exhaustively digested with EndoH, the molecular weights of the mutant proteins were reduced to the same size (Fig. 1C), indicating that the difference in molecular weight indeed resulted from a reduced number of glycans. These results suggest that the different TCR{alpha} mutants are expressed in EL4 cells at comparable levels and processed to contain the expected number of glycans.


Figure 1
View larger version (36K):
[in this window]
[in a new window]

  		FIGURE 1.
Expression of TCR{alpha}-SIINFEKL mutants in EL4 cells. A and B, EL4 cells were transduced with the pMIG retroviral vector that encodes TCR{alpha} (or mutants) (see "Experimental Procedures"), and GFP-positive cells were sorted by fluorescence-activated cell sorting. Depicted in the graphs are GFP levels of control non transduced cells (A) and EL4 cells transduced with pMIG-TCR{alpha} (B). C, EL4 cells transduced with different TCR{alpha} mutants were labeled with [35S]methionine for 30 min, and then lysed in 1% SDS. The lysate was diluted to 0.07% SDS with Nonidet P-40 lysis mix followed by precipitation with an anti-TCR{alpha} antibody. Precipitates were treated with EndoH, where indicated, and analyzed by SDS-PAGE followed by fluorography.

 
If PNGase indeed cleaves all glycans of TCR{alpha} prior to proteasomal degradation, the S1 and the S1-N44D should yield the same polypeptide sequence in the cytoplasm and therefore, the release of the SIINFEKL epitope by proteasomal proteolysis should occur in the exact same manner. Hence, equal levels of SIINFEKL presentation on H-2Kb are expected. On the other hand, if the levels of H-2Kb/SIINFEKL are different between the S1- and the S1-N44D-expressing cells, we can infer that at some point in the course of degradation the proteasome encountered the glycosylated TCR{alpha}. This type of analysis requires a quantitative measurement of the levels of H-2Kb/SIINFEKL at the cell surface. To examine the presentation of the SIINFEKL peptide, derived from the different mutants, we initially used a T-cell activation assay. SIINFEKL presentation by H-2Kb molecules leads to the stimulation of CD69 expression on OTI CD8 T cells, which specifically recognize the H-2Kb/SIINFEKL complex (43). Up-regulation of surface CD69 on OTI cells was evident after incubation with EL4 cells expressing the different SIINFEKL-containing TCR{alpha} derivatives (Fig. 2A). In contrast, EL4 cells that were transduced with WT TCR{alpha} (control), which lacks the SIINFEKL sequence, did not activate the OTI T cells (Fig. 2A). We conclude that the SIINFEKL epitope is processed and presented from the TCR{alpha} backbone.


Figure 2
View larger version (34K):
[in this window]
[in a new window]

  		FIGURE 2.
SIINFEKL generated from TCR{alpha}-SIINFEKL mutants is presented by H2-Kb molecules on the cell surface of EL4 cells. A, 106 OTI cells were incubated with either 50 µM SIINFEKL peptide, or with 3 x 105 EL4 cells expressing the different TCR{alpha}-SIINFEKL mutants, or a combination of both, for 16 h. The cells were then washed and stained with phycoerythrin-anti-CD69, followed by fluorescence-activated cell sorting analysis. x axis, forward scatter; y axis, CD69 levels. The percentage indicates the portion of activated T cells in the assay. B, EL4 cells expressing the different TCR{alpha} mutants (red, WT-TCR{alpha}; green, S1; pink, S1 N44D; and blue, S1{Delta}SP) were treated with or without Z-VAD-fmk (50 µM) for 24 h. The cells were then stained with 25D1.16 monoclonal antibody, recognizing the H2-Kb/SIINFEKL complex, followed by staining with Alexa Fluor 647-conjugated anti-mouse used as a secondary antibody. Live cells were gated based on their light scatter characteristics. To ensure similar levels of expression, a narrow GFP gate (see supplemental Fig. S2) was selected when the levels of SIINFEKL expression in the different cell types were analyzed.

 
The similar magnitude of OTI activation by the different EL4 cells may be due to saturation. Presumably, maximal activation is triggered even by low levels of H2-Kb/SIINFEKL complexes. To detect the differences in the levels of H-2Kb/SIINFEKL presented at the surface of the EL4 cells we measured the complex levels directly. We gated the EL4 cells by a narrow window of GFP expression to ensure similar levels of transgene expression (supplemental Fig. S2). The selected cells were analyzed by flow cytometry for the levels of H-2Kb/SIINFEKL complexes by staining with 25-D1.16, a monoclonal antibody specific for the complex (34). Cells expressing WT-TCR{alpha} represented the background levels (Fig. 2B). In cells expressing the S1 mutant (in which the SIINFEKL epitope was inserted adjacent to the glycosylated Asn-44) the SIINFEKL peptide was presented by H-2Kb to a lower extent compared with that of either S1{Delta}SP or S1-N44D (the mutants lacking glycosylation at position 44) (Fig. 2B). These results indicate that the presence of a glycan in proximity to the nominal epitope markedly reduced its presentation. This conclusion was further supported by the additional effect of treatment with Z-VAD-fmk, an inhibitor of PNGase, which resulted in a further reduction in SIINFEKL presentation by the S1-expressing cells, but did not affect the presentation of SIINFEKL derived either from S1{Delta}SP or S1-N44D. Note that the scale of Fig. 2B is logarithmic; thus, the effect on antigen presentation is significant.

To verify that differences in the efficiency of SIINFEKL presentation originating from each of the mutants are not due to differences in kinetics of degradation, we monitored the rate of decay of the S1 mutant by pulse-chase analysis. As seen in Fig. S3 the S1 mutant was degraded at similar rate to WT-TCR{alpha}. Thus, the alteration in the level of presentation must be correlated with the presence of the glycan moiety. These observations indicate that the proteasome acts directly on the glycosylated TCR{alpha}, probably because of non-efficient removal of all glycans by PNGase prior to proteasomal degradation, and some of the carbohydrates may be removed following proteolysis.

The role of PNGase in class I MHC presentation was previously demonstrated by its ability to generate epitopes through the conversion of the asparagine residue to aspartic acid in the course of glycan removal. This accounts for T-cell recognition of the tyrosinase-derived epitope YMNGTMSQV that is presented on class I MHC as YMDGTMSQV (44), and for an MHC-restricted peptide originating from the lymphocytic choriomeningitis virus GP1 glycoprotein (45). Our data implicate PNGase as a modulator of presentation also in cases where the glycan is located close to the nominal epitope, and not only by the mere conversion of the Asn residue to Asp. One possibility is that the N-terminal extended version of the epitope containing the glycan moiety interferes with further processing by amino peptidases (cytosolic and/or ER resident) to the appropriate length suitable for loading onto class I MHC molecules. Alternatively, it is possible that the glycan moiety at the N terminus prohibits the translocation via the TAP complex into the ER.

The Mammalian Proteasome Directly Degrades N-Linked Glycosylated Substrates in Vitro—If PNGase acts exclusively prior to proteasomal degradation, its inhibition by Z-VAD-fmk should reduce SIINFEKL presentation for both S1 and S1-N44D, because the remaining three glycans of TCR{alpha} should a priori inhibit the entry of the glycosylated form into the proteasome. Our findings, however, indicate that, upon inhibition of PNGase, SIINFEKL presentation was reduced only for the S1 and not for the S1-N44D variant (Fig. 2B). This supports the hypothesis that deglycosylation does not occur entirely prior to proteasomal degradation and implies that in some cases PNGase may act following proteasomal degradation of glycoproteins. Our findings agree with the observations of Altrich-VanLith et al. (46), which demonstrated a decrease in the presentation of the tyrosinase-derived epitope (which contains a glycosylation site in its sequence) upon PNGase blockade. Our results may also explain the lack of phenotype upon deletion of PNGase in yeast or inhibition of its activity in mammalian cells (20, 23).

We monitored the global effect of png1 deletion on the transcription profile of the yeast S. cerevisiae using DNA microarray analysis. We detected only minor alterations in the transcriptional profile when the png1-deficient strain was compared with its WT parental strain (data not shown). This lack of a global alteration in gene expression supports our notion that loss of PNGase activity does not prevent proteasomal degradation. It is noteworthy that yeast PNGase is expressed at low levels and its activity is barely detectable in yeast cytosol in contrast to the rapid de-N-glycosylation observed in mammalian cytosol (47). Thus, it is likely that inhibition of PNGase activity would affect differently glycoprotein turnover in yeast and mammalian cells.

To examine whether the mammalian 26 S proteasome can degrade N-linked glycoproteins without the involvement of N-glycanase, we analyzed the degradation of bovine pancreatic ribonuclease B (RNaseB), which is frequently used as a model substrate for PNGase. RNase is a small soluble protein (124 residues) that contains four disulfide bonds. RNase occurs in two major forms; RNaseA and RNaseB. The latter differs from RNaseA by the presence of a single, high mannose-type N-linked glycan attached to Asn-34 (48). Because RNaseA (and -B) are substrates for proteasomal degradation only after unfolding and reduction (29), we reduced and blocked the eight cysteine residues of RNaseB by alkylation with iodoacetamidobiotin. Biotin was used to enhance the detection of the protein by immunoblotting and to allow the retrieval of biotinylated fragments. First, we subjected the biotinylated RNaseB (denoted RNaseB-biotin) to degradation by the 20 S proteasome core particle. RNaseB-biotin was incubated for various time intervals with 20 S proteasomes purified from rabbit muscle, in the presence or absence of SDS. Low concentrations of SDS induce the gate opening of the 20 S proteasome {alpha} ring, a prerequisite for efficient substrate entry into the proteasome catalytic core (49, 50). RNaseB was degraded by the 20 S proteasome (Fig. 3A), and SDS treatment enhanced its degradation. To verify the presence of the N-linked glycan on the biotinylated substrate, we subjected RNaseB-biotin to deglycosylation with a recombinant bacterial PNGaseF, which yielded the expected reduction in molecular weight (Fig. 3B). Although normally deglycosylation by PNGaseF proceeds more readily after unfolding of a glycosylated substrate, no additional treatment was needed to enhance glycan removal from RNaseB-biotin, consistent with its lack of stable tertiary structure in the reduced state (51).

The degradation of RNaseB-biotin by the 20 S proteasome proves that in principle, the proteasome can digest glycan-containing substrates, and that there is no a priori size restriction that excludes glycosylated substrates from entering the 20 S particle. Yet, the presence of a glycan may inhibit entry into, and translocation through the 19 S regulatory complex. We therefore analyzed the ability of the eukaryotic 26 S proteasome to digest glycosylated and deglycosylated RNaseB-biotin. The two substrates were incubated with purified rabbit muscle 26 S proteasomes in the presence or absence of ATP. Both the glycosylated and the deglycosylated RNaseB-biotin were degraded by the 26 S proteasome in an ATP-dependent manner (Fig. 3C). Similar results were obtained when an anti-RNase polyclonal antibody was used for detection (data not shown). The presence of the glycan on glycosylated RNaseB-biotin was demonstrated by blotting the membrane with the lectin concanavalin A (Fig. 3C, ConA, lower panel). Together, these results indicate that the high mannose glycan of RNaseB-biotin does not interfere with its degradation by the mammalian 20 S and 26 S proteasomes in vitro. Nevertheless, the presence of the glycan resulted in reduction of RNase degradation rate. Although the half-life of the deglycosylated RNase was ~30 min, the presence of the glycan prolonged the half-life to ~90 min (Fig. 3C). Likewise, Misgahi et al. (19) noted that, upon inhibition of PNGase in mammalian cells, the degradation rate of glycosylated substrates was reduced. RNaseB-biotin contains 8 cysteine residues modified by biotin, which are spaced along the whole molecule (aa 26, 40, 58, 65, 72, 84, 95, and 110). Thus partial degradation or protection of a large fragment by the glycan should have resulted in the appearance of biotinylated intermediates. Such an intermediate was not detected in our immunoblot analysis. The complete degradation of RNaseB is also demonstrated by mass spectrometry analysis of the degradation products (see Fig. 6 below).

To explore how the proteasome handles substrates decorated with either complex-type glycans or multiple glycans, and to verify that the ability of the 26 S proteasome to degrade RNaseB is not a rare exception, we monitored the proteasomal degradation of additional types of glycoproteins. We used transferrin receptor (TfR), which contains a single complex-type N-linked glycan, TCR{alpha}, which carries four high mannose N-linked glycans and ovalbumin, a glycoprotein with a single high mannose glycan. TfR was reduced and alkylated with iodoacetamido-biotin (termed TfR-biotin). TCR{alpha} with a C-terminal His6 tag was purified from mammalian cells (see "Experimental Procedures"). The purified TCR{alpha} was unfolded either by boiling and immediate chilling, or by incubation with guanidinium-HCl, followed by immediate desalting to remove the denaturant. These treatments confer on the substrate a molten globule-type structure prior to its incubation with the proteasome. Ovalbumin was also reduced and biotinylated (termed OVA-biotin), as were RNaseB and TfR. We verified by PNGaseF treatment, as well as by concanavalin A binding, that the glycans survived the biotinylation and were present on all the input substrates (data not shown). We then subjected these substrates to degradation by the 20 S and 26 S proteasomes (Fig. 4). As indicated in Fig. 4 (A, C, and E), these substrates were degraded by the SDS-activated 20 S proteasome. In addition, the three glycoproteins were also destroyed following incubation with 26 S proteasomes in the presence of ATP (Fig. 4, B, D, and F). Thus, the mammalian proteasome can degrade glycoproteins containing high mannose glycans, such as RNaseB (Fig. 3) or ovalbumin (Fig. 4, E and F), as well as complex-type glycan-containing proteins such as TfR (Fig. 4, A and B) and even substrates with multiple glycans, such as TCR{alpha} (Fig. 4, C and D).


Figure 3
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 3.
RNaseB is degraded by the eukaryotic proteasome. A, RNaseB-biotin was incubated at 37 °C with eukaryotic 20 S proteasomes, with or without 0.018% SDS for the time intervals indicated. Equal amounts of sample were removed at each time point and separated by SDS-PAGE, followed by blotting with an anti-biotin antibody. A quantification of the immunoblot is presented in the graph (right panel). B, RNaseB-biotin was incubated for 1 h at 37 °C with or without PNGaseF. The reaction products were analyzed by immunoblot analysis with an anti-biotin antibody. C, RNaseB-biotin was incubated at 37 °C with eukaryotic 26 S proteasomes in the presence or absence of 2 mM ATP for the time intervals indicated. Equal amounts of sample were removed at the time points indicated and separated by SDS-PAGE, followed by blotting with the indicated antibodies. A quantification of the immunoblot is presented in the graph below.

 
As TCR{alpha} is decorated with 4 N-linked glycans (Asn-44, -168, -202, and -213), partial degradation from either or both termini of TCR{alpha} must result in the accumulation of a degradation intermediate of 169 residues due to protection of the two terminal glycans (Asn-44 and Asn-213). This product would contain the four glycans and thus should have also strong reactivity with concanavalin A, as well as anti-TCR{alpha} polyclonal antibodies. As seen in Fig. S4 (an overview of the complete molecular weight range of a degradation reaction of TCR{alpha}), we did not detect such partial degradation products (which could not have escaped detection due to its expected size and stoichiometric amounts with the starting material). Thus, our results do not support proteasomal degradation of glycoproteins through multiple rounds of partial degradation.

The ability to degrade glycoproteins is an inherent and conserved property of proteasomes from archaea to mammals, as the archaeal proteasome also degraded those glycosylated substrates (Fig. S5). Clearly, the narrow gated opening in the {alpha} ring of the 20 S proteasome (13–15 Å) (52–55) and the dimensions of the ATPase ring in the base of the 19S particle (56) do not prevent entry into the proteasome. Given the size of the central compartment of the proteasome, accommodation of a glycosylated peptide is certainly within the range of possibilities. The ability of the gated {alpha} ring of the proteasome to extend beyond its crystallographic dimensions was already proposed by Jentsch and co-workers (57, 58), who suggested that loop-formation occurs during proteasomal degradation of certain membrane proteins. This suggestion was corroborated by Liu et al. (59), who showed that the proteasome can accommodate hairpin-loop type of substrates and cleave a polypeptide even when both termini are blocked by large non-translocatable moieties such as folded GFP. Together with our current results, it is clear that the proteasome can digest complex structures that have a diameter larger than that of a single fully unfolded polypeptide chain, and the gated opening in the 20 S {alpha} ring must be more permissive than may be inferred from its crystal structure (52–55). The glycan moiety, which is attached to a glycoprotein through the side chain of an asparagine residue, is relatively flexible, and thus may assume a conformation that allows transport of the glycosylated substrate into the core proteasome. In this sense glycosylation appears no different than other modifications, including biotinylation, phosphorylation, or addition of large fluorescent probes (36, 51), all of which allow proteasomal degradation. The experiments shown here address the ability of the proteasome to degrade glycosylated substrates in the absence of ubiquitination. However, because polyubiquitin chains are responsible for the targeting of substrates to the proteasome upstream to the translocation of the polypeptide chain into the catalytic core, we do not think that addition of polyubiquitin would interfere with degradation. On the contrary, if at all, the presence of polyubiquitin chains may facilitate the degradation of the glycosylated substrates (60, 61). The glycan moiety serves in vivo as an anchor for E3 ubiquitin ligases (30, 31). In addition, glycan moieties also play a role in the extraction phase of ER-associated degradation substrates from the ER (62, 63). It should be emphasized that our in vitro analysis did not take the effect of substrate ubiquitination into account but, rather, investigated the direct interaction of the glycoprotein with the proteasome. Therefore, the presence of ubiquitin chains may affect the kinetics of degradation either by influencing the interaction with the proteasome or affecting de-ubiquitination, which must occur prior to the proteasomal degradation itself. Proper investigation of these aspects would require generation of homogenous pool of ubiquitinated glycoprotein.


Figure 4
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 4.
The glycan moiety of transferrin receptor, TCR{alpha}, and ovalbumin does not prevent proteasomal degradation. A, C, and E, 20 S proteasomal degradation; TfR-biotin (A), TCR{alpha} (C), or ovalbumin-biotin (E) were incubated at 37 °C with eukaryotic 20 S proteasomes, with or without 0.018% SDS, for the time intervals indicated. Equal amounts of sample were removed at each time point and separated by SDS-PAGE, followed by blotting with anti-biotin (A and E) or anti-TCR{alpha} (C) antibodies. B, D, and F, 26 S proteasomal degradation; TfR-biotin (B), TCR{alpha} (D), or ovalbumin-biotin (F) were incubated at 37 °C with eukaryotic 26 S proteasomes in the presence of 2 mM ATP for the times indicated. Equal amounts of sample representing each time point were separated by SDS-PAGE followed by blotting with anti-biotin (B and F) or anti-TCR{alpha} (D) antibodies. Quantifications of the immunoblots analysis are presented in Fig. S6.

 
Proteasome-associated PNGase Is Not Essential for Glycoprotein Degradation—Both yeast and mammalian PNGase associate either directly or indirectly with the proteasome (21, 22, 25, 26). This raises the possibility that our proteasome preparations might contain traces of bound PNGase that can act on the substrates prior to our in vitro degradation assays. To exclude this possibility, we used Z-VAD-fmk, which irreversibly blocks PNGase (19). To verify that Z-VAD-fmk inhibits PNGase in vitro (in our experimental conditions), we used recombinant yeast PNGase expressed and purified from bacterial cells. The purified yeast PNGase (yPNGase) was active, as evident by the reduction in the molecular weight of RNaseB-biotin upon incubation with yPNGase. Deglycosylation by yPNGase was inhibited when Z-VAD-fmk was present, but not by the proteasome inhibitor MG132 (Fig. 5A). We then examined the proteasomal degradation of RNaseB-biotin in the presence of either Z-VAD-fmk or MG132. As expected, degradation of RNaseB-biotin was inhibited in the presence of MG132 (Fig. 5B), whereas Z-VAD-fmk treatment did not affect its degradation. We did not detect a complete removal of the glycan, which is associated with PNGase activity, or any partial trimming of the carbohydrates mediated by other enzymes, such as endoglycosidases or mannosidases. Likewise, these enzymes were not detected in mass spectrometry analysis of our proteasomes preparations (20 S and 26 S, data not shown). We conclude that, even if PNGase is physically coupled to the proteasome, as was previously suggested (21, 22, 25, 26), its activity is not mandatory for the degradation of a glycosylated substrate by the mammalian 26 S proteasome.

Although our results demonstrate that, in principle, deglycosylation is not required prior to proteasomal degradation, our findings do not exclude the possibility that some glycosylated substrates are deglycosylated by PNGase prior to proteasomal degradation, whereas certain other glycoproteins are digested by the proteasome independent of glycan removal. Most likely, the order of events of deglycosylation and degradation is dictated by the nature of the protein and other components, which are involved in substrate processing prior to its proteasomal degradation that require the presence of the carbohydrate moiety (30, 31, 62, 63). Thus, it is possible that different glycosylated proteins are processed differently. Although in certain cases the glycan may serve as a cue for dislocation and ubiquitination, in other instances the processing of glycosylated substrates prior to proteasomal degradation might be completely independent of the glycan itself. This is also exemplified by the different degradation patterns of TCR{alpha}, which is degraded without prior deglycosylation, and that of class I MHC heavy chains, which undergo deglycosylation prior to their degradation. Either way PNGase inhibition did not block their proteasomal degradation (19, 23). It is noteworthy that TCR{alpha} belongs to a class of glycoproteins in which the presence of the glycan is essential for recognition and ubiquitination by their E3 ubiquitin ligases (30, 31). Similarly, during the degradation of the yeast model substrate CPY*, which is also dependent on one of its glycans for dislocation, no deglycosylated intermediates are detected upon proteasome inhibition (20).


Figure 5
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 5.
PNGase is not present in the 26 S preparation, nor is it required for RNaseB degradation. A, yPNGase was preincubated for 15 min at 30 °C in the presence or absence of MG132 (10 µM) or Z-VAD-fmk (100 µM). Subsequently, RNaseB-biotin was added, and the reaction was incubated for additional 30 min at 30 °C, followed by SDS-PAGE and blotting with an anti-biotin antibody. B, mammalian 26 S proteasome was preincubated for 15 min at 37 °C in the presence or absence of MG132 (10 µM) or Z-VAD-fmk (100 µM). Following the incubation, RNaseB-biotin was added, and the reaction was incubated for an additional hour at 37 °C, followed by SDS-PAGE and blotting with an anti-biotin antibody.

 
The Presence of a Glycan Limits the Digestion Sites Close to the Glycan Moiety—Our findings demonstrate that glycoproteins can be degraded by the proteasome (Figs. 3 and 4). To monitor how a glycan affects the digestion pattern of glycosylated proteins, we analyzed the repertoire of peptides generated following 26 S proteasomal degradation by mass spectrometry. Glycosylated and deglycosylated RNaseB reduced and alkylated with iodoacetamide (RNaseB-IAA) were used as model substrates. The overall proteasomal digestion products profile of RNaseB and deglycosylated RNaseB were similar (Fig. 6). Comparable results were recorded also for glycosylated and deglycosylated OVA-biotin and TfR-biotin (data not shown). Thus, the presence of glycans does not globally alter the manner in which the glycosylated substrate is handled by the proteasome, and any effect, if exists, is local (i.e. changes in cleavage sites close to the glycosylation site).

The proteasomal degradation of an intact glycosylated protein (described in Figs. 3 and 4) must result in the generation of glycosylated peptides. Such a glycosylated peptide should be absent from the pool of products generated by digestion of the deglycosylated protein. Following in vitro proteasomal degradation of RNaseB-IAA (performed as in Fig. 3C), the reaction mixture was analyzed for the presence of a glycosylated peptide. In principle, it is possible to detect glycopeptides generated by proteasomal degradation by two methods. One involves the direct detection of glycosylated peptides by monitoring an increase in the peptide's molecular weight, contributed by the glycan. Alternatively, it is possible to compare degradation products of glycosylated peptides before and following treatment with a recombinant PNGase. As a result of PNGase activity, a distinct mass corresponding to the glycopeptide (in which the glycosylated Asn was converted to Asp) will appear. This mass must be missing in the peptide profile of the untreated sample. Due to the heterogeneity of the glycan moiety (64), which translates to multiple masses related to a single glycopeptide sequence, we chose to use the latter method. As a reference, deglycosylated RNaseB-IAA was digested by the proteasome, and the sequence of the different proteasomal degradation products was determined by MALDI TOFTOF mass spectrometry (Table 1 and Fig. S7). Within the pool of deglycosylated RNaseB degradation products, six different peptides containing the original glycosylation site were detected. In parallel, the glycosylated RNaseB was digested by the 26 S proteasome, and the degradation products were treated by PNGaseF, and analyzed similarly. Surprisingly, only three of the peptides detected among the degradation products of the deglycosylated RNaseB were also generated during glycosylated RNaseB degradation (Table 1). In a control experiment, mass spectrometry of the degradation products of glycosylated RNaseB was carried without a following PNGase treatment (Fig. S7). As expected, the peptides containing the glycosylation site could not be detected due to an increase in their molecular weight attributed to the glycan moiety.


View this table:
[in this window]
[in a new window]

  		TABLE 1
RNaseB degradation products detected by MALDI TOFTOF RNaseB-IAA proteasomal degradation products containing the glycosylation site detected by MALDI TOFTOF mass spectrometry. The peptides contained an Asp residue instead of the original glycosylated Asn (underlined). –, indicates a peptide that was not detected in the sample; +, indicates a peptide detected in the sample.

 


Figure 6
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 6.
Mass spectrometry analysis of RNAseB-IAA degradation products. Glycosylated and deglycosylated RNaseB-IAA were subjected to degradation by the 26 S proteasome (as described in Fig. 3). The digestion products were subjected to liquid chromatography-tandem mass spectrometry on a DECA-XP mass spectrometer. Peptides containing the glycosylation site that were detected by either MALDI TOFTOF or Orbitrap (see Tables 1 and 2) were also added. Underlined, peptides that appeared in both samples and quantified (see Table 2); *, peptides that were detected only among deglycosylated RNaseB-IAA degradation products; **, peptides detected only among glycosylated RNaseB-IAA degradation products. The glycosylated Asn residue of RNaseB is highlighted in light gray.

 
To identify additional peptides, not detected by MALDI TOFTOF, and to quantify the relative amounts of degradation products identified in both reactions we subjected the same samples to analysis by an ion-trap mass spectrometer (Table 2). We identified two peptides, which appeared in the pool of degradation products of both glycosylated and deglycosylated RNaseB (peptides 4* and 8, Table 2). This approach helped us to detect a unique peptide among the glycosylated RNaseB degradation products (peptide 7, Table 2), and two unique products from the deglycosylated RNaseB pool of digestion products (peptides 2* and 5*, Table 2). Interestingly, the relative amount of peptide 8 (Table 2) was higher by 2.5-fold in the deglycosylated RNaseB products relative to its glycosylated counter-part. Even more pronounced was the difference in the generation of peptide 4* (Table 2), which was 6-fold more abundant in the deglycosylated RNaseB products relative to the products of the glycosylated protein (see Table 2).


View this table:
[in this window]
[in a new window]

  		TABLE 2
RNaseB degradation products detected by ORBITRAP RNaseB-IAA proteasomal degradation products containing the glycosylation site detected by the Orbitrap mass spectrometry. RNaseB-IAA digestion products treated with PNGaseF, and deglycosylated RNaseB-IAA 26 S proteasomal degradation products (see Table I and supplemental Fig. S7) were analyzed by the Orbitrap (see "Experimental Procedures"). Listed in the table are peptides containing the glycosylation site. As expected those peptides contained an Asp residue instead of the original glycosylated Asn (underlined). -, indicates a peptide that was not detected in the sample; +, indicates a peptide that was detected in the sample. Semiquantitative comparative measurements were done according to the overall intensity of the relevant peaks. The relative amounts of peptide 8 and 4* were calculated based on the facts that the intensities of peptides other than the presented in the table were similar, and the conditions of the MS analyses were identical.

 
Our results demonstrate that the presence of a glycan protects the microenvironment surrounding the glycan. Accordingly, longer peptides are produced from the digestion of the glycosylated region, in which the digestion sites are more distant from the glycosylated residue.

We conclude that, although the proteasome is capable of degrading both the glycosylated and deglycosylated RNaseB, the glycan moiety affects the local proteasome digestion characteristics (i.e. close to the glycosylation site). Although we detected seven degradation products (peptides 1–6 and 8, Tables 1 and 2) containing the glycosylation site among the deglycosylated RNaseB degradation products, we could detect only four of those products (peptides 1, 3, 4, and 8, Tables 1 and 2), and an additional unique longer product (peptide 7, Table 2), following the degradation of glycosylated RNaseB. In the presence of the carbohydrate the ability of the proteasome to generate short peptides was markedly reduced (compare peptides 2, 5, and 6, Tables 1 and 2). The results presented in Tables 1 and 2 are also in agreement with our in vivo observations (Fig. 2), in which the reduced presentation of SIINFEKL was directly correlated with the presence of the glycan moiety. Accordingly, the glycan moiety of RNaseB led to generation of an N-terminally extended glycopeptide (peptide 7, Table 2). Presumably, TCR{alpha}-S1 degradation also involved generation of N-extended SIINFEKL glycosylated epitopes. This was reflected by the further reduction of SIINFEKL presentation upon PNGase inhibition.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Here we show that N-linked glycosylation may affect class I MHC presentation of epitopes located near a glycosylation site. Our findings suggest that, in the absence of PNGase activity, a glycosylated N-extended epitope is not presented following proteasomal degradation. Likewise, Engelhard and colleagues (46) demonstrated a similar dependence on PNGase activity for efficient presentation of an epitope with an integrated glycosylation site. Taken together, PNGase is an important modulator of MHC class I antigen presentation of epitopes that contain or reside adjacent to a glycosylation site. This is further supported by the analysis of PNGase distribution across different tissues, which indicates that its levels are elevated mainly in cells of the immune system (University of California, Santa Cruz microarray data base).

Following proteasomal degradation, products suitable in size for presentation (8 or 9 residues) undergo two major competing processes. A small portion of the peptides is transported into the ER where they are loaded on class I MHC en route to be presented at the cell surface. Alternatively, these peptides can be further processed to smaller peptides not suitable for presentation. Because the half-life of a peptide generated in the course of proteasomal degradation is estimated to be seconds (39), even a small attenuation in transport of potential epitope to the ER (if it contains a glycan) may facilitate its final destruction by cytoplasmic peptidases and, accordingly, reduce its presentation. We propose that the kinetics of glycan removal may be a crucial step in the balance between presentation and destruction of class I MHC epitopes. Thus, dense glycosylation, as found on certain viral proteins, may help them evade class I MHC presentation. Although we found that glycosylation per se does not impair proteasomal degradation independently of the type and number of glycans, we suggest that PNGase can act upstream or downstream to proteasomal degradation. Most likely, in some cases where the glycosylation site is exposed, PNGase may act prior to proteasomal degradation, whereas in other cases, where the glycan is less accessible, proteasomal degradation proceeds without prior deglycosylation.


   FOOTNOTES
 
* This work was supported in part by the Israeli Foundation of Science (Grant ISF 802/04), The German Minerva foundation (Grant 8527), the German Israeli Foundation for scientific research and development (Grant GIF I-845-210.9/2004), and a special donation from Rolando Uziel. 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. Back

Formula The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1–S7. Back

1 Both authors contributed equally to this work. Back

2 A recipient of an Eshkol PhD fellowship (3-3407). Back

3 To whom correspondence should be addressed: Tel.: 972-8-934-3719; Fax: 972-8-934-4116; E-mail: ami.navon{at}weizmann.ac.il.

4 The abbreviations used are: MHC, major histocompatibility complex; ER, endoplasmic reticulum; PNGase, peptide N-glycanase; PNGaseF, peptide N-glycanase F; yPNGase, yeast peptide N-glycanase; EndoH, endoglycosidase H; E3, ubiquitin-protein isopeptide ligase; Z, benzyloxycarbonyl; fmk, fluoromethyl ketone; GFP, green fluorescent protein; DTT, dithiothreitol; IAA, iodoacetamide; TfR, transferrin receptor; MALDI, matrix-assisted laser desorption ionization; TOF, time of flight; TCR, T cell receptor; RNaseA, -B, ribonucleases A and B; aa, amino acid(s). Back


   ACKNOWLEDGMENTS
 
We are grateful to Dr. Tamar Ziv and Dr. Arie Admon from the Technion (Haifa, Israel) for mass spectrometry analysis of proteasomal degradation products.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Confalonieri, F., and Duguet, M. (1995) BioEssays 17, 639-650[CrossRef][Medline] [Order article via Infotrieve]
  2. Gottesman, S., and Maurizi, M. R. (1992) Microbiol. Rev. 56, 592-621[Abstract/Free Full Text]
  3. Gottesman, S., Maurizi, M. R., and Wickner, S. (1997) Cell 91, 435-438[CrossRef][Medline] [Order article via Infotrieve]
  4. Voges, D., Zwickl, P., and Baumeister, W. (1999) Annu. Rev. Biochem. 68, 1015-1068[CrossRef][Medline] [Order article via Infotrieve]
  5. Bonifacino, J. S., and Lippincott-Schwartz, J. (2003) Nat. Rev. Mol. Cell. Biol. 4, 409-414[CrossRef][Medline] [Order article via Infotrieve]
  6. Goldberg, A. L. (1995) Science 268, 522-523[Free Full Text]
  7. Hershko, A., and Ciechanover, A. (1998) Annu. Rev. Biochem. 67, 425-479[CrossRef][Medline] [Order article via Infotrieve]
  8. Hochstrasser, M. (1996) Annu. Rev. Genet. 30, 405-439[CrossRef][Medline] [Order article via Infotrieve]
  9. Wiertz, E. J., Tortorella, D., Bogyo, M., Yu, J., Mothes, W., Jones, T. R., Rapoport, T. A., and Ploegh, H. L. (1996) Nature 384, 432-438[CrossRef][Medline] [Order article via Infotrieve]
  10. Helenius, A., and Aebi, M. (2001) Science 291, 2364-2369[Abstract/Free Full Text]
  11. Cox, J. S., and Walter, P. (1996) Cell 87, 391-404[CrossRef][Medline] [Order article via Infotrieve]
  12. Mori, K. (2000) Cell 101, 451-454[CrossRef][Medline] [Order article via Infotrieve]
  13. Sidrauski, C., Cox, J. S., and Walter, P. (1996) Cell 87, 405-413[CrossRef][Medline] [Order article via Infotrieve]
  14. Casagrande, R., Stern, P., Diehn, M., Shamu, C., Osario, M., Zuniga, M., Brown, P. O., and Ploegh, H. (2000) Mol. Cell 5, 729-735[CrossRef][Medline] [Order article via Infotrieve]
  15. Friedlander, R., Jarosch, E., Urban, J., Volkwein, C., and Sommer, T. (2000) Nat. Cell Biol. 2, 379-384[CrossRef][Medline] [Order article via Infotrieve]
  16. Travers, K. J., Patil, C. K., Wodicka, L., Lockhart, D. J., Weissman, J. S., and Walter, P. (2000) Cell 101, 249-258[CrossRef][Medline] [Order article via Infotrieve]
  17. Urano, F., Calfon, M., Yoneda, T., Yun, C., Kiraly, M., Clark, S. G., and Ron, D. (2002) J. Cell Biol. 158, 639-646[Abstract/Free Full Text]
  18. Meusser, B., Hirsch, C., Jarosch, E., and Sommer, T. (2005) Nat. Cell Biol. 7, 766-772[CrossRef][Medline] [Order article via Infotrieve]
  19. Misaghi, S., Pacold, M. E., Blom, D., Ploegh, H. L., and Korbel, G. A. (2004) Chem. Biol. 11, 1677-1687[CrossRef][Medline] [Order article via Infotrieve]
  20. Suzuki, T., Park, H., Hollingsworth, N. M., Sternglanz, R., and Lennarz, W. J. (2000) J. Cell Biol. 149, 1039-1052[Abstract/Free Full Text]
  21. Park, H., Suzuki, T., and Lennarz, W. J. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 11163-11168[Abstract/Free Full Text]
  22. Suzuki, T., Park, H., Kwofie, M. A., and Lennarz, W. J. (2001) J. Biol. Chem. 276, 21601-21607[Abstract/Free Full Text]
  23. Blom, D., Hirsch, C., Stern, P., Tortorella, D., and Ploegh, H. L. (2004) EMBO J. 23, 650-658[CrossRef][Medline] [Order article via Infotrieve]
  24. Wiertz, E. J., Jones, T. R., Sun, L., Bogyo, M., Geuze, H. J., and Ploegh, H. L. (1996) Cell 84, 769-779[CrossRef][Medline] [Order article via Infotrieve]
  25. Biswas, S., Katiyar, S., Li, G., Zhou, X., Lennarz, W. J., and Schindelin, H. (2004) Biochem. Biophys. Res. Commun. 323, 149-155[CrossRef][Medline] [Order article via Infotrieve]
  26. Katiyar, S., Li, G., and Lennarz, W. J. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 13774-13779[Abstract/Free Full Text]
  27. Lee, J. H., Choi, J. M., Lee, C., Yi, K. J., and Cho, Y. (2005) Proc. Natl. Acad. Sci. U. S. A. 102, 9144-9149[Abstract/Free Full Text]
  28. Hirsch, C., Blom, D., and Ploegh, H. L. (2003) EMBO J. 22, 1036-1046[CrossRef][Medline] [Order article via Infotrieve]
  29. Hirsch, C., Misaghi, S., Blom, D., Pacold, M. E., and Ploegh, H. L. (2004) EMBO Rep. 5, 201-206[CrossRef][Medline] [Order article via Infotrieve]
  30. Yoshida, Y., Chiba, T., Tokunaga, F., Kawasaki, H., Iwai, K., Suzuki, T., Ito, Y., Matsuoka, K., Yoshida, M., Tanaka, K., and Tai, T. (2002) Nature 418, 438-442[CrossRef][Medline] [Order article via Infotrieve]
  31. Yoshida, Y., Tokunaga, F., Chiba, T., Iwai, K., Tanaka, K., and Tai, T. (2003) J. Biol. Chem. 278, 43877-43884[Abstract/Free Full Text]
  32. Tirosh, B., Iwakoshi, N. N., Lilley, B. N., Lee, A. H., Glimcher, L. H., and Ploegh, H. L. (2005) J. Virol. 79, 2768-2779[Abstract/Free Full Text]
  33. Rehm, A., Stern, P., Ploegh, H. L., and Tortorella, D. (2001) EMBO J. 20, 1573-1582[CrossRef][Medline] [Order article via Infotrieve]
  34. Porgador, A., Yewdell, J. W., Deng, Y., Bennink, J. R., and Germain, R. N. (1997) Immunity 6, 715-726[CrossRef][Medline] [Order article via Infotrieve]
  35. Huppa, J. B., and Ploegh, H. L. (1997) Immunity 7, 113-122[CrossRef][Medline] [Order article via Infotrieve]
  36. Cascio, P., Hilton, C., Kisselev, A. F., Rock, K. L., and Goldberg, A. L. (2001) EMBO J. 20, 2357-2366[CrossRef][Medline] [Order article via Infotrieve]
  37. Rock, K. L., York, I. A., and Goldberg, A. L. (2004) Nat. Immunol. 5, 670-677[CrossRef][Medline] [Order article via Infotrieve]
  38. Yewdell, J. W., Reits, E., and Neefjes, J. (2003) Nat. Rev. Immunol. 3, 952-961[CrossRef][Medline] [Order article via Infotrieve]
  39. Reits, E., Griekspoor, A., Neijssen, J., Groothuis, T., Jalink, K., van Veelen, P., Janssen, H., Calafat, J., Drijfhout, J. W., and Neefjes, J. (2003) Immunity 18, 97-108[CrossRef][Medline] [Order article via Infotrieve]
  40. Chen, W., Khilko, S., Fecondo, J., Margulies, D. H., and McCluskey, J. (1994) J. Exp. Med. 180, 1471-1483[Abstract/Free Full Text]
  41. Ben-Shahar, S., Komlosh, A., Nadav, E., Shaked, I., Ziv, T., Admon, A., DeMartino, G. N., and Reiss, Y. (1999) J. Biol. Chem. 274, 21963-21972[Abstract/Free Full Text]
  42. Yu, H., Kaung, G., Kobayashi, S., and Kopito, R. R. (1997) J. Biol. Chem. 272, 20800-20804[Abstract/Free Full Text]
  43. Schott, E., Bertho, N., Ge, Q., Maurice, M. M., and Ploegh, H. L. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 13735-13740[Abstract/Free Full Text]
  44. Skipper, J. C., Hendrickson, R. C., Gulden, P. H., Brichard, V., Van Pel, A., Chen, Y., Shabanowitz, J., Wolfel, T., Slingluff, C. L., Jr., Boon, T., Hunt, D. F., and Engelhard, V. H. (1996) J. Exp. Med. 183, 527-534[Abstract/Free Full Text]
  45. Hudrisier, D., Riond, J., Mazarguil, H., Oldstone, M. B., and Gairin, J. E. (1999) J. Biol. Chem. 274, 36274-36280[Abstract/Free Full Text]
  46. Altrich-VanLith, M. L., Ostankovitch, M., Polefrone, J. M., Mosse, C. A., Shabanowitz, J., Hunt, D. F., and Engelhard, V. H. (2006) J. Immunol. 177, 5440-5450[Abstract/Free Full Text]
  47. Romisch, K., and Ali, B. R. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 6730-6734[Abstract/Free Full Text]
  48. Joao, H. C., and Dwek, R. A. (1993) Eur. J. Biochem. 218, 239-244[Medline] [Order article via Infotrieve]
  49. Kisselev, A. F., Kaganovich, D., and Goldberg, A. L. (2002) J. Biol. Chem. 277, 22260-22270[Abstract/Free Full Text]
  50. Kohler, A., Cascio, P., Leggett, D. S., Woo, K. M., Goldberg, A. L., and Finley, D. (2001) Mol. Cell 7, 1143-1152[CrossRef][Medline] [Order article via Infotrieve]
  51. Navon, A., and Goldberg, A. L. (2001) Mol. Cell 8, 1339-1349[CrossRef][Medline] [Order article via Infotrieve]
  52. Groll, M., Bajorek, M., Kohler, A., Moroder, L., Rubin, D. M., Huber, R., Glickman, M. H., and Finley, D. (2000) Nat. Struct. Biol. 7, 1062-1067[CrossRef][Medline] [Order article via Infotrieve]
  53. Groll, M., Ditzel, L., Lowe, J., Stock, D., Bochtler, M., Bartunik, H. D., and Huber, R. (1997) Nature 386, 463-471[CrossRef][Medline] [Order article via Infotrieve]
  54. Lowe, J., Stock, D., Jap, B., Zwickl, P., Baumeister, W., and Huber, R. (1995) Science 268, 533-539[Abstract/Free Full Text]
  55. Whitby, F. G., Masters, E. I., Kramer, L., Knowlton, J. R., Yao, Y., Wang, C. C., and Hill, C. P. (2000) Nature 408, 115-120[CrossRef][Medline] [Order article via Infotrieve]
  56. Walz, J., Erdmann, A., Kania, M., Typke, D., Koster, A. J., and Baumeister, W. (1998) J. Struct. Biol. 121, 19-29[CrossRef][Medline] [Order article via Infotrieve]
  57. Hoppe, T., Matuschewski, K., Rape, M., Schlenker, S., Ulrich, H. D., and Jentsch, S. (2000) Cell 102, 577-586[CrossRef][Medline] [Order article via Infotrieve]
  58. Piwko, W., and Jentsch, S. (2006) Nat. Struct. Mol. Biol. 13, 691-697[CrossRef][Medline] [Order article via Infotrieve]
  59. Liu, C. W., Corboy, M. J., DeMartino, G. N., and Thomas, P. J. (2003) Science 299, 408-411[Abstract/Free Full Text]
  60. Lam, Y. A., Lawson, T. G., Velayutham, M., Zweier, J. L., and Pickart, C. M. (2002) Nature 416, 763-767[CrossRef][Medline] [Order article via Infotrieve]
  61. Hanna, J., Leggett, D. S., and Finley, D. (2003) Mol. Cell. Biol. 23, 9251-9261[Abstract/Free Full Text]
  62. Kostova, Z., and Wolf, D. H. (2005) J. Cell Sci. 118, 1485-1492[Abstract/Free Full Text]
  63. Spear, E. D., and Ng, D. T. (2005) J. Cell Biol. 169, 73-82[Abstract/Free Full Text]
  64. Lee, B. S., Krishnanchettiar, S., Lateef, S. S., and Gupta, S. (2005) Rapid Commun. Mass Spectrom. 19, 1545-1550[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Immunol.Home page
M. Ostankovitch, M. Altrich-VanLith, V. Robila, and V. H. Engelhard
N-Glycosylation Enhances Presentation of a MHC Class I-Restricted Epitope from Tyrosinase
J. Immunol., April 15, 2009; 182(8): 4830 - 4835.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
283/1/244    most recent
M706237200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kario, E.
Right arrow Articles by Navon, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kario, E.
Right arrow Articles by Navon, A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2008 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement