HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Branza-Nichita, N. Articles by Petrescu, S. M. Search for Related Content PubMed PubMed Citation Articles by Branza-Nichita, N. Articles by Petrescu, S. M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 11, 8169-8175, March 17, 2000

Mutations at Critical N-Glycosylation Sites Reduce Tyrosinase Activity by Altering Folding and Quality Control^*

Norica Branza-Nichita^§, Gabriela Negroiu, Andrei J. Petrescu^§, Elspeth F. Garman^¶, Fran M. Platt^§, Mark R. Wormald^§, Raymond A. Dwek^§, and Stefana M. Petrescu^§**

From the  Institute of Biochemistry of the Romanian Academy, Splaiul Independentei 296, 77700 Bucharest 17, Romania, the ^§ Oxford Glycobiology Institute, Department of Biochemistry, University of Oxford, South Parks Road, Oxford OX1 3QU, United Kingdom, and the ^¶ Laboratory of Biophysics, Department of Biochemistry, University of Oxford, South Parks Road, Oxford OX1 3QU, United Kingdom

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Tyrosinase is a copper-containing enzyme that regulates melanin biosynthesis in mammals. Mutations at a single N-glycosylation sequon of tyrosinase have been reported to be responsible for oculocutaneous albinism type IA in humans, characterized by inactive tyrosinase and the total absence of pigmentation. To probe the role that each N-glycosylation site plays in the synthesis of biologically active tyrosinase, we analyzed the calnexin mediated folding of tyrosinase N-glycosylation mutants. We have determined that four of the six potential N-glycosylation sites, including that associated with albinism, are occupied. Analysis of the folding pathway and activity of 15 tyrosinase mutants lacking one or more of the occupied N-glycosylation sites shows that glycans at any two N-glycosylation sites are sufficient to interact with calnexin and give partial activity, but a specific pair of sites (Asn⁸⁶ and Asn³⁷¹) is required for full activity. The mutants with less than two N-glycosylation sites do not interact with calnexin and show a complete absence of enzyme activity. Copper analysis of selected mutants suggests that the observed partial activity is due to two populations with differential copper content. By correlating the degree of folding with the activity of tyrosinase, we propose a local folding mechanism for tyrosinase that can explain the mechanism of inactivation of tyrosinase N-glycosylation mutants found in certain pigmentation disorders.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Tyrosinase (monophenol, dihydroxyphenylalanine:oxygen oxidoreductase, EC 1.14.18.1) is a copper-containing enzyme responsible for catalyzing the first two steps of the melanin synthesis pathway, hydroxylation of tyrosine to dihydroxyphenylalanine (DOPA)¹ and its subsequent oxidation to DOPA quinone (1). The presence of two copper atoms in the active site is essential for activity. The enzyme also has six N-glycosylation sequons (Fig. 1). All six potential N-glycosylation sites are conserved in human and mouse tyrosinase (2), suggesting that the glycosylation may be functionally relevant. The absence of tyrosinase activity is associated with oculocutaneous albinism in many animal species, including humans (3). This inborn disorder is characterized by the absence of melanin in the skin, hair, and eyes and a series of related abnormalities of the ocular system. Several mutations within the tyrosinase gene of individuals with type I oculocutaneous albinism have been reported, including a substitution in codon 371 causing a putative amino acid change from threonine (ACA) to lysine (AAA) causing abolition of the sixth potential N-glycosylation site (4, 5). The expressed tyrosinase gene produces a protein with no activity, leading to an absence of melanin synthesis (5). We have found that following treatment of mouse melanoma cells with N-butyldeoxynojirimicin (NB-DNJ), an inhibitor of -glucosidases I and II, the resulting expressed tyrosinase, which carries immature Glc₃Man_7-9GlcNAc₂ N-glycans, is catalytically inactive and fails to initiate melanin biosynthesis (6).

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Schematic representation of the sequence of mouse tyrosinase (numbered according to SwissProt AC P11344) showing the positions of the six potential N-glycosylation sites (numbered 1-6). SS, signal sequence; TM, trans-membrane region; CYS, regions rich in conserved cysteine residues; Cu2+, proposed binding sites for the two copper atoms based on sequence comparisons with other copper binding proteins.

We and others have shown that tyrosinase activity is dependent on the interaction between tyrosinase and calnexin during the initial stages of folding (7, 8). Calnexin and calreticulin are two lectin-like chaperones that interact transiently with partially trimmed N-linked oligosaccharides of the newly synthesized glycoprotein, their substrates being incompletely folded, monoglucosylated proteins (9). Calnexin is part of the folding/quality control mechanism within the endoplasmic reticulum, along with other endoplasmic reticulum chaperones and the enzymes -glucosidase II and UDP-glucose:glycoprotein glucosyl transferase (GT) (10, 11). Monoglucosylated N-glycans are generated by the actions of -glucosidases I and II (12) and GT, the latter only recognizing denatured or incompletely folded proteins (13, 14). Thus, the substrates of these chaperones are driven through a cycle of deglucosylation/reglucosylation and returned to the cycle by GT as long as the native conformation is not achieved (15). Interference in the early stages of glycan processing prevents the interaction of tyrosinase with calnexin and leads to inactive enzyme being produced (6, 8). Although it is now accepted that calnexin promotes folding of tyrosinase by interacting with monoglucosylated N-glycans (7, 8), it remains unclear as to whether glycans at all six of the potential N-glycosylation sites of the polypeptide chain are involved in these interactions and hence in folding.

The object of this paper is to probe the role that each of the six individual N-glycosylation sites plays in the synthesis of biologically active tyrosinase. We have identified the N-glycosylation sites that are normally occupied in tyrosinase (four of the six). Fifteen tyrosinase mutants have been constructed lacking one or more of these four sites using site-directed mutagenesis and have been transiently expressed in CHO cells. We measured the enzymatic activity of the wild type and recombinant tyrosinases and investigated their kinetics of folding and calnexin interaction. These studies show that the mutants with less than two glycans do not interact with calnexin and have no enzymatic activity. Any pair of the four normal glycans is sufficient to give an interaction with calnexin resulting in partial activity, but only one specific pair reproduces the folding kinetics of the wild type protein and gives full activity. The analysis of copper content for two of the mutants suggests that the observed partial activity is a result of incomplete copper loading. These results show that individual N-glycans play distinct roles in the tyrosinase folding kinetics, suggesting a local folding mechanism for this glycoprotein.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reagents-- NB-DNJ was a gift from Searle/Monsanto (St. Louis, MO). The rabbit anti-tyrosinase antiserum (PEP7) was a gift from Dr. V. J. Hearing (NCI, National Institute of Health, Bethesda, MD). The expression plasmid, pHDmcTyr 1, was a kind gift from Dr. Günther Schüts (Institute of Cell and Tumor Biology, Heidelberg, Germany). [³⁵S]Methionine/cysteine was from I.C.N. Flow, (Thame, Oxfordshire, UK), CHAPS was purchased from Pierce. All other chemicals were from Sigma.

Cell Culture-- CHO cells (European Collection of Animal Cell Cultures, Porton Down, UK), were cultured in RPMI 1640 medium (Life Technologies, Inc.) containing 10% fetal calf serum (Sigma), 50 units/ml penicillin, and 50 mg/ml streptomycin (Life Technologies, Inc.) and maintained at 37 °C with 5% CO₂.

Tyrosinase N-Glycosylation Mutants-- All mutant constructs were based on the mouse tyrosinase cDNA expression plasmid, pHDmcTyr1 (2). Tyrosinase mutants lacking single N-glycosylation sites were created by changing the codon for Asn (TAT) in each of the six potential N-glycosylation sequons to the codon for Gln (CAA), using synthetic oligonucleotides (22-30-mers) and a site-directed mutagenesis kit (Stratagene, La Jolla, CA). The sequences of the mutagenic oligonucleotides are given in Table I. The resultant expressed proteins were named (1), (2), (3), (4), (5), and (6) according to the N-glycosylation site removed (Table II).

View this table: [in this window] [in a new window]   Table II Results of the calnexin interaction and enzyme activity assays for wild type tyrosinase and N-glycosylation site deletion mutants expressed in CHO cells The notation (x,y) refers to the protein obtained following expression of mutant tyrosinase cDNA lacking N-glycosylation sequons x and y (see Fig. 1).

All the other tyrosinase mutants lacking more than one N-glycosylation site were obtained using the cDNAs coding for (1), (2), (3), (4), (5), and (6), generated as described above, and recombining the DNA fragments resulting from hydrolysis with restriction enzymes, as follows: (1,4), (1,5), and (1,6): the cDNAs coding for (1), (4), (5), and (6) were digested with ApaI and ClaI, and the 5' fragment of (1) cDNA was ligated to the 3' fragment of the cDNAs coding for (4), (5), or (6), respectively, using standard techniques; (5,6) was obtained through partial SphI hydrolysis of the cDNAs coding for (5) and (6), followed by the ligation of the DNA fragments containing the mutation at the N-glycosylation sites 5 and 6, respectively; (4,5), (4,6), and (4,5,6): the cDNAs coding for (4), (5), (6), and (5,6) were digested with BsaAI and ClaI and the 5' fragment of (4) cDNA was ligated to the 3' fragment of the cDNAs coding for (5), (6), or (5,6), respectively; and (1,4,5), (1,4,6), (1,5,6), and (1,4,5,6) were obtained using the same ApaI and ClaI hydrolysis approach described above, except that the substrates were the cDNAs coding for (1), (4,5), (4,6), (5,6), and (4,5,6). Nucleotide sequences of all constructs were confirmed by the dye terminator double-stranded DNA sequencing method.

Transfection of CHO Cells and Metabolic Labeling-- For transfection, 75-cm² culture flasks of CHO cells in logarithmic phase growth were used to transiently express different variants of tyrosinase N-glycosylation mutants, using TransIT^TMLT transfection kit (PanVera Corporation), according to the manufacturer's instructions. In all experiments, controls were included in which the cells were transfected with the expression plasmid lacking the fragment coding for tyrosinase. For metabolic labeling, 48 h post-transfection, CHO cells were starved in the cysteine/methionine-free medium for 1 h, pulse labeled with 100-150 µCi of [ ³⁵S]cysteine/methionine (Tran³⁵S-label, 1100 Ci/mmol, ICN Flow, Thame Oxfordshire, UK), for 30 min and chased for the indicating time. When the effect of -glucosidases I and II inhibition was followed, NB-DNJ was added to the cells either 1 h before or immediately after the pulse labeling and was maintained through out the chase period. Labeled cells were harvested, washed twice with cold phosphate-buffered saline, and incubated in 20 mM N-ethylmaleimide for 1 h to block the free sulfhydryl groups, by alkylation. Cells were then lysed with lysis buffer (2% CHAPS in 50 mM Hepes, 200 mM NaCl, pH 7.5, containing a mixture of protease inhibitors (Sigma)), for 1 h, on ice.

Immunoprecipitation and SDS-PAGE-- After lysis as described above, cell extracts were first precleared by incubation with 20 µl of protein A-Sepharose for 1 h, at 4 °C. Proteins were then immunoprecipitated in supernatants with either  PEP 7 anti-tyrosinase antibodies (20, 21) or anti-calnexin antibodies followed by anti-tyrosinase antibodies. The bound proteins were eluted with either 1% SDS for 30 min, at room temperature (nonreducing conditions, under which activity is retained) or by boiling in reducing buffer with 5% 2-mercaptoethanol, for 5 min (reducing conditions). The eluted proteins were then subjected to SDS-PAGE and analyzed by autoradiography.

Tyrosinase Assay and Western Blotting-- The DOPA-oxidase assay measuring the second catalytic activity of tyrosinase, the conversion of DOPA to DOPAchrome was performed as described before (22). Briefly, 48 h after transfection with different mutant tyrosinase cDNAs, CHO cells grown in 75-cm² culture flasks were harvested and lysed in 100 µl of lysis buffer. 50 µl of cell extracts were mixed with 950 µl of 1 mM DOPA in 0.1 M sodium phosphate buffer, pH 7.2, for 30 min at 37 °C. The appearance of DOPAchrome was measured spectrophotometrically at 475 nm. The specific activity of different tyrosinase N-glycosylation mutants was defined as the enzyme activity/mg of total protein. Protein concentration in the cell lysates was determined using the BCA Protein Assay kit (Pierce).

The remaining 50 µl of each cell extract was boiled for 5 min in reducing sample buffer and loaded on 7.5% acrylamide gels. Electrophoretically separated proteins were transferred to Immobilon membrane (Amersham Pharmacia Biotech). Blots were incubated with 1:100 dilutions of rabbit anti-tyrosinase antibodies (PEP 7), for 1 h, at 37 °C. Immunoreactivity of proteins was detected by enhanced chemiluminescent Western blotting (ECL, Amersham Pharmacia Biotech), according to the manufacturer's instructions.

Copper Analysis-- Samples for copper analysis were prepared by 7.5% SDS-PAGE under nonreducing conditions. The bands of interest were identified by DOPA oxidase activity, excised from the gel, and resolubilized in phosphate-buffered saline. The presence of tyrosinase was confirmed by Western blotting. The proteins were dialyzed against phosphate-buffered saline followed by exhaustive dialysis against water to remove chloride ions (which interfere with the analysis of sulfur). The samples were concentrated by ultrafiltration using a 30-kDa cut-off membrane to a final concentration of 1 µg of protein in 100 µl. Three or four 0.5-µl drops of each sample were spotted on to 2-µm-thick mylar film sample holders and allowed to dry by evaporation overnight. Direct inspection showed that the majority of the protein was concentrated on the circumference of the drop but not uniformly distributed.

The elemental composition of each sample was measured by proton induced x-ray emission spectroscopy using a scanning proton microprobe (23). Briefly, this involves irradiating the sample with a proton beam of 1-µm diameter and monitoring the emitted x-rays. Each element that is irradiated by the beam gives a distinctive x-ray peak that can be identified and integrated with an accuracy of ± 5%. The beam can be scanned across the sample to produce a spatial map of elemental distribution or focused at a single point to obtain accurate spectra. The procedure used here was initially to generate a map of sulfur distribution. The only source of sulfur present in the sample comes from methionine and cysteine side chains and residual SDS, and so this gives a map of the protein distribution. Three or four points on each sample that gave a high sulfur peak were then selected to obtain point spectra. The copper content at each point was measured and standardized to the amount of protein present at that point by taking the ratio of copper to sulfur (23).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Folding of Wild Type Tyrosinase in CHO Cells and Its Association with Endogenous Calnexin-- We have previously investigated the folding and maturation of tyrosinase in the melanotic B16 melanoma cell line (8). CHO cells do not normally synthesize tyrosinase and do not contain melanosomes. Transfection of CHO cells with the expression vector pHDmcTyr1 containing the full-length mouse tyrosinase cDNA (2) resulted in the production of fully active enzyme.

Tyrosinase biosynthesis was monitored in vivo by pulse-chase analysis, followed by immunoprecipitation of labeled cell lysates with anti-tyrosinase antibodies (PEP7) and analysis by nonreducing SDS-PAGE (Fig. 2A). A slower migrating (upper) band is observed immediately after 30 min of pulse, which disappears within about 3 h. A faster migrating (lower) band appears approximately 2 h post pulse labeling and is still present up to 7 h of chase. On reducing gels, tyrosinase runs as a single 72-kDa band for all pulse-chase times examined (Fig. 2B), confirming that the two species resolved in nonreducing gels are two folding intermediates.

View larger version (45K): [in this window] [in a new window]   Fig. 2.   Kinetics of folding and calnexin interaction of wild type tyrosinase expressed in CHO cells. CHO cells transfected with wild type mouse tyrosinase cDNA were incubated in starvation buffer for 1 h before a 30 min pulse with 35S. Cells were chased for the indicated periods of time, and cell lysates were subjected to immunoprecipitation with either anti-tyrosinase antibodies (A and B) or anti-calnexin antibodies followed by anti-tyrosinase antibodies (C and D). The immunoprecipitated proteins were analyzed by nonreducing (A and C) or reducing (B and D) SDS-PAGE followed by autoradiography. X-ray films were exposed to the dried gels for 10 days.

Because tyrosinase synthesized in melanoma cells associates with the chaperone calnexin all along its folding pathway (8), we analyzed whether this association occurs in CHO cells transfected with tyrosinase cDNA. Interactions between tyrosinase and calnexin were probed using double immunoprecipitation with anti-calnexin followed by anti-tyrosinase antibodies and analysis of the bound proteins by nonreducing SDS-PAGE (Fig. 2C). The resulting gel is almost identical to that in Fig. 2A except that the lower band disappears 3 h post-pulse. Under reducing conditions, calnexin bound tyrosinase runs as a single band at 72 kDa (Fig. 2D). Tyrosinase could not be precipitated by anti-calnexin antibodies in the presence of NB-DNJ (data not shown), demonstrating the requirement of calnexin for monoglucosylated glycans.

All these results are similar to those obtained previously (8) in melanoma cells (B16 cell line). Thus, CHO cells are capable of producing active tyrosinase with maturation kinetics and calnexin binding properties similar to those of melanocytes.

Occupancy of Tyrosinase N-Glycosylation Sites by Site-directed Mutagenesis-- Tyrosinase recombinants lacking single N-glycosylation sites were used to determine the occupancy of the six putative N-glycosylation sites. CHO cell lysates expressing the above mutants were subjected to electrophoresis under reducing conditions followed by Western blotting and tyrosinase immunodetection. Four of the six mutants, (1), (4), (5), and (6), showed faster migration than the wild type (Fig. 3A; see Table II for explanation of notation). (2) and (3) showed similar migration to the wild type (Fig. 3A). Thus sites 1, 4, 5, and 6 at Asn residues 86, 230, 337, and 371 are fully occupied, whereas sites 2 and 3 at Asn residues 111 and 161 are unoccupied.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Analysis of N-glycan site occupancy of tyrosinase expressed in CHO cells by selective removal of the potential N-glycan sites. A, Western blot of CHO cell lysates expressing wild type or tyrosinase mutants lacking one potential N-glycosylation site. The samples were reduced prior to electrophoresis. Lane 1, wild type; lane 2, (1); lane 3, (2); lane 4, (3); lane 5, (4); lane 6, (5); lane 7, (6). (x) corresponds to the tyrosinase mutant missing N-glycosylation site x. B, Western blot of CHO cell lysates expressing tyrosinase mutants lacking one or more normally occupied N-glycosylation sites. The samples were reduced prior to electrophoresis. Lane 1, (1); lane 2, (1,6); lane 3, (1,5,6); lane 4, (1,4,5,6); lane 5, wild type after PNGase digestion.

Folding and in Vivo Calnexin Interaction of Tyrosinase N-Glycosylation Mutants-- Tyrosinase mutants lacking one, two, three, or four of the normally occupied N-glycosylation sites were constructed, and their expression in CHO cells were analyzed by Western blotting. Migration patterns of these mutants were consistent with the molecular weight of the mutants lacking one to all of the four glycans. The mutant (1,4,5,6) has a similar migration to that of the wild type protein digested with PNGase F, confirming the absence of any N-linked oligosaccharides (Fig. 3B). Folding of these mutants was analyzed by single and double immunoprecipitation with anti-tyrosinase and anti-calnexin antibodies (summarized in Table II).

All 15 mutants gave two bands on nonreducing SDS-PAGE after immunoprecipitation with anti-tyrosinase antibodies (typical data given in Fig. 4, A, C, and E). (4), (5), and (4,5) show behavior similar to that of the wild type protein (Fig. 4C). All the mutants lacking site 1 had 5-10% of the upper band remaining up to 7 h post-pulse (for example, Fig. 4, A and E). The remaining mutants all show intermediate behavior with the upper band disappearing and the lower band appearing between 0 min and 1 h (Table II).

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Kinetics of folding and calnexin interaction of tyrosinase mutants lacking one or more normally occupied N-glycosylation sites expressed in CHO cells. CHO cells transfected with tyrosinase cDNAs coding for (1) (A and B), (4,5) (C and D), and (1,6) (E and F) were incubated in starvation buffer for 1 h before a 30-min pulse with 35S. Cells were chased for the indicated periods of time, and cell lysates were subjected to immunoprecipitation with either anti-tyrosinase antibodies (A, C, and E) or anti-calnexin antibodies followed by anti-tyrosinase antibodies (B, D, and F). The immunoprecipitated proteins were analyzed by nonreducing SDS-PAGE followed by autoradiography. X-ray films were exposed to the dried gels for 10 days.

The mutants lacking three or four normally occupied N-glycosylation sites cannot be co-immunoprecipitated by anti-calnexin antibodies. All the remaining mutants can be precipitated by anti-calnexin antibodies (typical data given in Fig. 4, B, D, and F). Again, (4), (5), and (4,5) show behavior similar to that of the wild type protein (Fig. 4D). The other mutants also show two bands, but the lower band disappears between 30 min and 2 h post-pulse (Fig. 4, B and F, and Table II). The mutants lacking three or four normally occupied N-glycosylation sites ((1,4,5,6) shown in Fig. 5A) show behavior similar to that of the wild type protein in the presence of NB-DNJ (Fig. 5B).

View larger version (47K): [in this window] [in a new window]   Fig. 5.   Kinetics of folding of aglycosyl tyrosinase and of wild type tyrosinase expressed in CHO cells, in the presence of NB-DNJ. A, same as Fig. 2A except that CHO cells were transfected with cDNA coding for (1,4,5,6). B, same as Fig. 2A except that 1 mM NB-DNJ was added to the cell suspension 30 min before the pulse.

We have previously shown in B16 cells that wild type tyrosinase binds irreversibly to calnexin in the absence of -glucosidase II activity (by post-pulse inhibition using NB-DNJ), leading to premature degradation (8). We have repeated this experiment on the (1,6) and (4,5) mutants, both of which show identical behavior to the wild type protein (data not shown), demonstrating that both mutants bind irreversibly to calnexin in the absence of -glucosidase II activity.

DOPA Oxidase Activity of Tyrosinase N-Glycosylation Mutants-- The enzymatic activity of wild type or mutant tyrosinase synthesized in CHO cells was determined by measuring their DOPA oxidase activity (Table II). (4), (5), and (4,5) show activity similar to the wild type. Significantly lower activities were measured for the single mutants (1) and (6), i.e. 70 and 64%, respectively, from the wild type activity. In contrast to the high activity of the double mutant (4,5), the mutant (1,6) retains only 30% of the wild type activity. All the mutants lacking three or four normally occupied N-glycosylation sites show no activity. The remaining mutants show intermediate activity, ranging from 30 to 70%.

Copper Analysis of Wild Type, (4,5), and (1,6)-- Samples for copper analysis of wild type tyrosinase and the mutants (4,5) and (1,6) were prepared by SDS-PAGE. The bands with DOPA oxidase activity were excised and dialyzed. All three samples contained approximately the same concentration of protein as determined by DOPA oxidase activity, allowing for the lower activity of (1,6). This method produces samples contaminated with other proteins but was the only method of obtaining sufficient quantities for elemental analysis. However, the contaminants are normally synthesized CHO cell proteins and thus lack DOPA oxidase activity and are the same in every sample. The copper content was measured by proton induced x-ray emission spectroscopy using a scanning proton microprobe. Each measurement is accurate to ±5%, but inhomogeneities in samples dried by evaporation lead to a greater variation in measurements taken at different points on the sample. Thus three or four measurements were taken for each sample and the copper content is reported as the ratio of copper to sulfur (see "Materials and Methods").

The four measurements for the wild type sample gave copper:sulfur values of 1:79, 1:58, and 1:48 and one reading below the detection limit. For the last point, the sulfur content was also low, indicating a low level of total protein at that point. The average copper:sulfur ratio from the first three values was 1:62. The four measurements for the (4,5) mutant gave copper:sulfur values of 1:74, 1:42, 1:53, and 1:50, giving an average of 1:55. The three measurements for the (1,6) mutant were below the detection limit for copper. Each of these measurements gave a relatively high sulfur value and the detection limit was calculated as 1:185, 1:232, and 1:264. Thus, the (1,6) mutant contains at least three times less copper relative to total protein than the wild type and (4,5) mutant. This is consistent with the ~100% activity for the wild type and (4,5) mutant compared with the ~30% activity for the (1,6) mutant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Tyrosinase is a type I membrane glycoprotein localized in the membrane of melanosomes, comprising 533 amino acids, six potential glycosylation sites, 17 cysteine residues grouped in two cysteine-rich domains, and two copper binding domains (2). To address the role of individual N-glycans in tyrosinase folding, knowledge of the occupancy of the sites is required. Expression in CHO cells of tyrosinase mutants lacking single N-glycosylation sites shows that sites 1, 4, 5, and 6 (at Asn residues 86, 230, 337, and 371) are fully occupied, whereas sites 2 and 3 (at Asn residues 111 and 161) are unoccupied. Although the exact locations have not been identified, four asparagine-linked oligosaccharides chains/molecule have been found on hamster tyrosinase (16), and both hamster (16) and mouse (17) tyrosinase show a similar range of oligomannosidic and sialilated complex antennary structures. These data, together with the observation that mouse and human tyrosinase have 85% sequence identity and identical potential N-glycosylation sites, indicate an interesting and possibly functionally relevant conservation of N-glycosylation between tyrosinases from different mammalian species. It remains unclear as to why the two unoccupied potential N-glycosylation sites are also conserved.

Wild type tyrosinase expressed in CHO cells shows very similar kinetics of folding and interaction with calnexin to the enzyme synthesized in B16 cells (8). Calreticulin has also been shown to interact with tyrosinase but only in the first 30 min after synthesis.² In both B16 and CHO cells, two inactive calnexin-bound folding intermediates of tyrosinase can be observed, resolved on nonreducing SDS-PAGE but not on reducing gels. Treatment with NB-DNJ results in inactive enzyme in both cell lines. These data suggest that the folding machinery used by tyrosinase is common to both melanoma and CHO cells and that CHO cells are a suitable model system to use to study the effects of glycosylation mutants on the folding of tyrosinase.

Tyrosinase mutants with less than two occupied N-glycosylation sites expressed in CHO cells cannot be co-immunoprecipitated with anti-calnexin, whereas all other mutants can. This implies that the interaction between tyrosinase and calnexin is multivalent in vivo. Similar results have been obtained in vitro for ribonuclease (10) and influenza hemagglutinin (18). We cannot rule out the possibility that complexes involving monoglycosylated proteins are formed, the interaction being too weak to be observed by co-immunoprecipitation. However, tyrosinase mutants with only one occupied site have no enzymatic activity and show the same folding kinetics as wild type tyrosinase expressed in the presence of NB-DNJ. Thus any interactions between monoglycosylated tyrosinase and calnexin have no effect on protein folding. The tyrosinase mutant with sites 1 and 6 occupied, (4,5), shows the same kinetics of interaction with calnexin and final activity as the wild type enzyme. This implies that calnexin is either divalent or is acting as a dimer during its interaction with tyrosinase, glycans at sites 1 and 6 being sufficient to reproduce the wild type behavior.

Tyrosinase mutants with three occupied glycosylation sites behave equivalently to the corresponding double mutant carrying the pair that gives the longest interaction time with calnexin. For example, the mutant with sites 1, 4, and 5 occupied ((6)) shows the same kinetics of folding and enzymatic activity as the double mutant with sites 1 and 5 occupied ((4,6)). This confirms the concept that the interaction between tyrosinase and calnexin is divalent.

The location of N-glycans dramatically affects both folding kinetics and final enzymatic activity of the mutants. For instance the mutant (4,5) has wild type activity, whereas (1,6) is only 30% active. Copper analysis of the wild type protein, the (4,5) mutant and the (1,6) mutant shows that the copper content is significantly lower in the (1,6) mutant. Because the presence of copper is critical to enzymatic activity, the (1,6) mutant must contain a high proportion of completely inactive protein, explaining its overall 30% level of activity.

A wide range of activities from 0 to 100% is seen for the whole range of N-glycosylation mutants. The simplest explanation of these results is that the expressed protein from each mutant consists of a mixture of active and inactive protein, the latter lacking copper. The observed percentage of activity represents the percentage of fully active protein present and thus the efficiency of folding of that mutant in vivo. The presence of any two occupied glycosylation sites is sufficient to produce some fully active protein, but the efficiency of folding depends on which sites are occupied, and wild type folding efficiency is only obtained with sites 1 and 6 occupied.

This difference in folding efficiency cannot be due to the difference in affinity of the glycosylation mutants for calnexin, because both the (1,6) and (4,5) mutants bind irreversibly to calnexin in the absence of -glucosidase II activity. It has previously been reported that some of the N-linked oligosaccharides of Schizosaccharomyces pombe double mutant Gls2/alg6 are not recognized by GT under denaturing conditions (19). It is possible that although the glycans at all the occupied N-glycosylation sites of tyrosinase can interact with calnexin, some of the glycans are not accessible to GT. Alternatively, the regions of the protein local to some of the glycan sites may fold rapidly and correctly with limited chaperone assistance and so may not be recognized by GT. The presence of glycans in regions of the protein structure that spontaneously fold less efficiently may be necessary for the correct function of the GT quality control system and hence may explain the conservation of glycosylation sites in tyrosinase.

It is interesting to note that there is a linear relationship that holds for the wild type and all the N-glycosylation mutants between the interaction time with calnexin and the enzymatic activity of the mature protein. Accelerated folding of tyrosinase mutants is accompanied by a lower enzymatic activity of the mature protein (Fig. 6). The in vivo acquisition of activity of tyrosinase appears to be zero-order with respect to tyrosinase, suggesting that one or more additional components are rate-limiting, for example other chaperones assisting tyrosinase during calnexin-mediated folding or the acquisition of copper at a later stage. We have found recently that Erp57 interacts with tyrosinase (data not shown). Other chaperones are currently being investigated to understand the whole picture of tyrosinase folding in the endoplasmic reticulum.³ Removal of glycans has been shown in vitro to result in the accelerated folding of influenza hemagglutinin monomers but not to impair final function (18). The data here indicate that for tyrosinase only a subpopulation (indistinguishable on the basis of its electrophoretic mobility) is able to reach the active conformation, and this increases with the time spent by the glycoprotein in the calnexin cycle.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Plot of interaction time with calnexin versus DOPA-oxidase activity of wild type tyrosinase and N-glycosylation mutants (data in Table II).

In conclusion, the efficiency of in vivo folding of tyrosinase is dependent on the occupancy of its N-glycan sites. In the absence of interaction with calnexin, no activity is obtained. Recruitment into the calnexin chaperone system occurs for any form of the enzyme that has two occupied N-glycan sites, leading to the production of some fully active molecules. Specific glycans are required for efficient operation of this pathway, leading to maximum activity. We propose that the quality control system operates locally with individual N-glycans having distinctive roles. Mutations at glycosylation sites critical to the operation of the quality control mechanism lead to a reduction in the activity of the expressed protein. It is interesting to note that the removal of site six has the most significant effect on the activity of the expressed enzyme, and this is the site at which mutations have been implicated in oculocutaneous albinism.

	ACKNOWLEDGEMENTS

We thank G. W. Grime (Oxford University, Oxford, UK) for help with the scanning proton microprobe analysis, G. Schüts for providing the plasmid pHDmcTyr1, V. Hearing (NIH, Bethesda, MD) for anti-tyrosinase antibodies, J. J. M. Bergeron (McGill University, Montreal, Canada) for anti-calnexin antibodies, and Searle/Monsanto for NB-DNJ.

	FOOTNOTES

^* This work was supported by Wellcome Trust Collaborative Research Initiative Grant 053441 and by grants from the Romanian Academy and Romanian Ministry of Research and Technology.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Lister Institute Research Fellow.

^** To whom correspondence should be addressed. Tel.: 401-223-90-69; Fax: 401-223-90-68; E-mail: Stefana.Petrescu@biochim.ro.

² S. M. Petrescu, N. Branza-Nichita, G. Negroiu, A. J. Petrescu, E. F. Garman, F. M. Platt, M. R. Wormald, and R. A. Dwek, personal communication.

³ A. Hillebrand, personal communication.

	ABBREVIATIONS

The abbreviations used are: DOPA, dihydroxyphenylalanine; NB-DNJ, N-butyldeoxynojirimicin; GT, UDP-glucose:glycoprotein glucosyl transferase; CHAPS, 3-(3-chloramidopropyl)-dimethylammonino-1-propanesulfate; CHO, Chinese hamster ovary; PAGE, polyacrylamide gel electrophoresis.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				D. N. Hebert and M. Molinari In and Out of the ER: Protein Folding, Quality Control, Degradation, and Related Human Diseases Physiol Rev, October 1, 2007; 87(4): 1377 - 1408. [Abstract] [Full Text] [PDF]

				G.-E. Costin and V. J. Hearing Human skin pigmentation: melanocytes modulate skin color in response to stress FASEB J, April 1, 2007; 21(4): 976 - 994. [Abstract] [Full Text] [PDF]

				C. I. Popescu, A. Mares, L. Zdrentu, N. Zitzmann, R. A. Dwek, and S. M. Petrescu Productive Folding of Tyrosinase Ectodomain Is Controlled by the Transmembrane Anchor J. Biol. Chem., August 4, 2006; 281(31): 21682 - 21689. [Abstract] [Full Text] [PDF]

				A. Singh, S. Severance, N. Kaur, W. Wiltsie, and D. J. Kosman Assembly, Activation, and Trafficking of the Fet3p{middle dot}Ftr1p High Affinity Iron Permease Complex in Saccharomyces cerevisiae J. Biol. Chem., May 12, 2006; 281(19): 13355 - 13364. [Abstract] [Full Text] [PDF]

				C. I. Popescu, C. Paduraru, R. A. Dwek, and S. M. Petrescu Soluble Tyrosinase is an Endoplasmic Reticulum (ER)-associated Degradation Substrate Retained in the ER by Calreticulin and BiP/GRP78 and Not Calnexin J. Biol. Chem., April 8, 2005; 280(14): 13833 - 13840. [Abstract] [Full Text] [PDF]

				A. Lavado, C. Olivares, J. C. Garcia-Borron, and L. Montoliu Molecular Basis of the extreme dilution mottled Mouse Mutation: A COMBINATION OF CODING AND NONCODING GENOMIC ALTERATIONS* J. Biol. Chem., February 11, 2005; 280(6): 4817 - 4824. [Abstract] [Full Text] [PDF]

				G. Tian, D. Wilcockson, V. H. Perry, P. M. Rudd, R. A. Dwek, F. M. Platt, and N. Platt Inhibition of {alpha}-Glucosidases I and II Increases the Cell Surface Expression of Functional Class A Macrophage Scavenger Receptor (SR-A) by Extending Its Half-life J. Biol. Chem., September 17, 2004; 279(38): 39303 - 39309. [Abstract] [Full Text] [PDF]

				Z. Frenkel, M. Shenkman, M. Kondratyev, and G. Z. Lederkremer Separate Roles and Different Routing of Calnexin and ERp57 in Endoplasmic Reticulum Quality Control Revealed by Interactions with Asialoglycoprotein Receptor Chains Mol. Biol. Cell, May 1, 2004; 15(5): 2133 - 2142. [Abstract] [Full Text] [PDF]

				A.-J. Petrescu, A.-L. Milac, S. M. Petrescu, R. A. Dwek, and M. R. Wormald Statistical analysis of the protein environment of N-glycosylation sites: implications for occupancy, structure, and folding Glycobiology, February 1, 2004; 14(2): 103 - 114. [Abstract] [Full Text] [PDF]

				G. Negroiu, R. A. Dwek, and S. M. Petrescu The Inhibition of Early N-Glycan Processing Targets TRP-2 to Degradation in B16 Melanoma Cells J. Biol. Chem., July 11, 2003; 278(29): 27035 - 27042. [Abstract] [Full Text] [PDF]

				C. Olivares, F. Solano, and J. C. Garcia-Borron Conformation-dependent Post-translational Glycosylation of Tyrosinase. REQUIREMENT OF A SPECIFIC INTERACTION INVOLVING THE CuB METAL BINDING SITE J. Biol. Chem., April 25, 2003; 278(18): 15735 - 15743. [Abstract] [Full Text] [PDF]