If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To whom correspondence should be addressed: Glycobiology Research Group, Tokyo Metropolitan Institute of Gerontology, Foundation for Research on Aging and Promotion of Human Welfare, 35-2 Sakaecho, Itabashi-ku, Tokyo 173-0015, Japan. Tel.: 81-3-3964-3241 (ext. 3080); Fax: 81-3-3579-4776
* This study was supported by a research grant for nervous and mental disorders (17A-10) and research on psychiatric and neurological diseases and mental health from the Ministry of Health, Labor, and Welfare of Japan, a grant-in-aid for scientific research on priority area (14082209) and on scientific research (17657039) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and the Japan Society for the Promotion of Science Core-to-Core Program. 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. The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1–S5.
O-Mannosyl glycans are important in muscle and brain development. Protein O-mannosyltransferase (POMT) catalyzes the initial step of O-mannosyl glycan biosynthesis. To understand which serine (Ser) and threonine (Thr) residues POMT recognizes for mannosylation, we prepared a series of synthetic peptides based on a mucin-like domain in α-dystroglycan (α-DG), one of the best known O-mannosylated proteins in mammals. In α-DG, the mucin-like domain spans amino acid residues 316 to 489. Two similar peptide sequences, corresponding to residues 401–420 and 336–355, respectively, were strongly mannosylated by POMT, whereas other peptides from α-DG and peptides of various mucin tandem repeat regions were poorly mannosylated. Peptides 401–420 and 336–355 contained four and six Ser and Thr residues, respectively. Substitution of Ala residues for the Ser or Thr residues showed that Thr-414 of peptide 401–420 and Thr-351 of peptide 336–355 were prominently modified by O-mannosylation. Matrix-assisted laser desorption ionization time-of-flight mass spectrometry and Edman degradation analysis of the mannosylated peptide 401–420 indicated that Thr-414 was the Thr residue that was most prominently modified by O-mannosylation and that O-mannosylation occurred sequentially rather than at random. Based on these results, we propose a preferred amino acid sequence for mammalian O-mannose modification.
O-Mannosyl glycans are important in muscle and brain development (
The abbreviations used are: α-DG, α-dystroglycan; MALDI-TOF MS, matrix-assisted laser desorption ionization time-of-flight mass spectrometry; PMT and POMT, protein O-mannosyltransferase; WWS, Walker-Warburg syndrome; Con A, concanavalin A; HPLC, high pressure liquid chromatography.
2The abbreviations used are: α-DG, α-dystroglycan; MALDI-TOF MS, matrix-assisted laser desorption ionization time-of-flight mass spectrometry; PMT and POMT, protein O-mannosyltransferase; WWS, Walker-Warburg syndrome; Con A, concanavalin A; HPLC, high pressure liquid chromatography.
predominantly include O-mannosyl glycan Siaα2–3Galβ1–4GlcNAcβ1–2Man (
). Mutations in POMT1 and POMT2 genes are considered to be the cause of Walker-Warburg syndrome (WWS: OMIM 236670), an autosomal recessive developmental disorder associated with congenital muscular dystrophy, neuronal migration defects, and ocular abnormalities (
). POMT1 and POMT2 are homologous to members of the family of protein O-mannosyltransferases (PMTs) in yeast. PMTs were shown to catalyze the transfer of a mannosyl residue from dolichyl phosphate mannose to Ser/Thr residues of certain proteins (
), suggesting the presence of some sequence for recognition by PMTs, but the sequence was not identified. On the other hand, in mammals, O-mannosylated proteins are rare and O-mannosylation may require a specific sequence because we detected POMT activity when a glutathione S-transferase-fused mucin-like domain of α-DG (amino acid residues 316–489) was used as an acceptor (
). Previous studies suggested that the mucin-like domain is highly glycosylated and that certain glycans of α-DG play an important role in its binding to ligand proteins such as laminin, neurexin, and agrin (
). To address the biological function and regulation of O-mannosylation, it is important to determine whether there is a preferential amino acid sequence. In this study we synthesized a series of peptides that fully covered the mucin-like domain of α-DG (Fig. 1). Then we examined whether these peptides worked as acceptors for protein O-mannosylation. Further, the enzymatic products were analyzed by matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) and Edman degradation to determine the number of mannose residues transferred and the sites to which they were transferred. Based on the results of these studies, we proposed a preferred sequence for mammalian O-mannosylation.
Chemicals—Synthetic peptides were purchased by made-to-order system on the web site of Sigma-Aldrich (www.genosys.jp, Tokyo, Japan), and the quality of the synthesis was ascertained by HPLC analysis and mass spectrometry. A series of peptides, whose average length is ∼20 amino acids, fully covered the mucin-like domain of α-DG (Table 1). Triton X-100 was purchased from Nacalai Tesque (Kyoto, Japan). n-Octyl-β-d-thioglucoside was from Dojindo Laboratories (Kumamoto, Japan). Tritium-labeled or unlabeled mannosylphosphoryldolichol were purchased from American Radiolabeled Chemical, Inc. (St. Louis, MO).
TABLE 1Kinetics analysis of POMT activity against various peptides
Mannosyl-threonine was synthesized as follows. To a solution of phenyl 2,3,4,6-tetra-O-benzyl-1-thio-α-d-mannopyranoside (1.0 g, 1.6 mmol) and N-benzyloxycarbonyl-l-threonine benzyl ester (813 mg, 2.4 mmol) in toluene (30 ml) and CH2Cl2 (10 ml), molecular sieves 4A (2.0 g) was added, and the mixture was stirred for 1 h at room temperature and then cooled to 0 °C. The cooled mixture were added to N-iodosuccinimide (720 mg, 3.2 mmol) and trifluoromethanesulfonic acid (30 μl, 0.32 mmol), and the mixture was further stirred for 2 h at 0 °C. The precipitates were filtered off through Celite and washed with CH2Cl2. Combined filtrate and washings were washed with saturated Na2CO3 and saturated Na2S2O3, dried, and concentrated. Column chromatography of the residue on silica gel (AcOEt:Hexane = 1:4) gave O-(2,3,4,6-tetra-O-benzyl-α-d-mannopyranosyl)-N-benzyloxycarbonyl-l-threonine benzyl ester (1.05 g, 77%). To a solution of O-(2,3,4,6-tetra-O-benzyl-α-d-mannopyranosyl)-N-benzyloxycarbonyl-l-threonine benzyl ester (500 mg, 0.58 mmol) in MeOH (50 ml) 10% Pd on activated carbon (500 mg) and acetic acid (50 μl) were added. H2 gas was bubbled into the mixture, which was stirred for 2 h at room temperature. The precipitates were filtered off and washed with water. The filtrate and washings were combined and concentrated to solid, which was chromatographed on Sephadex G-50 (eluent: water) to give O-(α-d-mannopyranosyl)-l-threonine (170 mg, quantitative). The structure was identified by 1H-NMR and MALDI-TOF MS.
Mannosyl peptide 401–420(T414Man) (IRPTMTIPGYVEPT(Man)AVATPP) was synthesized essentially as described in our previous report (
). The structure was identified by 1H-NMR and MALDI-TOF MS.
POMT Enzyme Source—The microsomal membrane fraction of human embryonic kidney 293T cells co-transfected by POMT1 and POMT2 was used for POMT enzyme source. A cDNA containing the most common splicing variant of human POMT1 (which lacks bases 700–765, corresponding to amino acids 234–255) and human POMT2 cDNA were cloned into pcDNA 3.1 (Invitrogen) as described previously (
). Human embryonic kidney 293T cells were maintained in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% fetal bovine serum, 2 mm l-glutamine, 100 units/ml penicillin, and 50 μg/ml streptomycin at 37 °C with 5% CO2. Expression plasmids were transfected into human embryonic kidney 293T cells using Lipofectamine plus regent (Invitrogen) according to the manufacturer's instructions. The cells were incubated for 3 days at 37 °C to produce POMT1 and POMT2 proteins. The cells were homogenized in 10 mm Tris-HCl, pH 7.4, 1 mm EDTA, 250 mm sucrose, 1 mm dithiothreitol with a protease inhibitor mixture (3 μg/ml pepstatin A, 1 μg/ml leupeptin, 1 mm benzamidine-HCl, 1 mm phenylmethylsulfonyl fluoride). After centrifugation at 900 × g for 10 min, the supernatant was subjected to ultracentrifugation at 100,000 × g for 1 h. The precipitate was used as the microsomal membrane fraction. Protein concentration was determined by BCA assay.
Assay for POMT Activity—POMT activity was based on the amount of [3H]mannose transferred from mannosylphosphoryldolichol to synthetic peptides. The reaction mixture containing 20 mm Tris-HCl, pH 8.0, 100 nm mannosylphosphoryldolichol (125,000 dpm/pmol), 2 mm dithiothreitol, 10 mm EDTA, 0.5% n-octyl-β-d-thioglucoside, 0.25–4 mm synthetic peptide; 40 μg of microsomal membrane fraction in 20 μl of total volume was incubated for 60 min at 25 °C. After boiling for 3 min, the mixture was analyzed by reversed-phase HPLC with a Wakopak 5C18–200 column (4.6 × 250 mm; Wako Pure Chemical Industries, Ltd., Osaka, Japan) or a Mightysil RP-18GP Aqua column (4.6 × 250 mm; Kanto Chemical Co., Inc., Tokyo, Japan). Solvent A was 0.1% trifluoroacetic acid in distilled water, and solvent B was 0.1% trifluoroacetic acid in acetonitrile. The peptide was eluted at a flow rate of 1 ml/min using a linear gradient of 1–50% solvent B. The peptide separation was monitored continuously at 215 nm, and the radioactivity of each fraction was measured using a liquid scintillation counter.
Peptide Sequencing and Mass Spectrometry—A mannosylated peptide 401–420 was prepared using concanavalin A (Con A)-Sepharose 4B beads (GE Healthcare) as follows. The reaction mixture containing 20 mm Tris-HCl, pH 8.0, 300 μm unlabeled mannosylphosphoryldolichol, 2 mm dithiothreitol, 10 mm EDTA, 0.5% n-octyl-β-d-thioglucoside, 1 mm peptide 401–420; 120 μg of microsomal membrane fraction in 30 μl of total volume was incubated for 15–120 min at 25 °C. The reaction mixture was mixed with 400 μl of 20 mm Tris-HCl, pH 7.4, 0.5 m NaCl, 1 mm MnCl2, 1 mm CaCl2, and 1 mm MgCl2 and centrifuged at 10,000 × g for 10 min. The supernatant was loaded onto the Con A-Sepharose column (100 μl of bed volume) and washed with 2 ml of the same buffer. The bound fraction was obtained by eluting with 250 μl of 200 mm α-methylmannoside in 20 mm Tris-HCL, pH 7.4, 0.5 m NaCl, 1 mm MnCl2, 1 mm CaCl2, and 1 mm MgCl2 and separated by reversed-phase HPLC as described above.
Con A-purified mannosyl peptide preparation was fractionated by reversed-phase HPLC on an Inertsil ODS-3 column (1 × 100 mm; GL Sciences Inc., Tokyo) with a model 1100 series liquid chromatography system (Agilent Technologies, Waldbronn, Germany). Mannosyl peptides were eluted at a flow rate of 20 μl/min using a linear gradient of 0–50% solvent B, where solvents A and B were 0.085% (v/v) aqueous trifluoroacetic acid and 0.075% (v/v) trifluoroacetic acid in 80% (v/v) acetonitrile, respectively. Three peptides thus obtained were subjected to Edman degradation using a Procise HT protein sequencing system (Applied Biosystems, Foster City, CA). “Cart pulsed liquid” and “Flask normal” were used for default programs of reaction and conversion cycle, respectively. At proline residues (3rd, 8th and 13th), the “Cart PL Proline” program was used for reaction cycles. Except for 8-mm peptide supports (Beckman Coulter Inc., Fullerton, CA), all sequence programs and reagents were supplied by Applied Biosystems. To reduce chemicals and background peaks, the chromatogram of the previous cycle was subtracted from one of the current cycle with the 610A data analysis system (Applied Biosystems) using the “Subtraction Mode”. To identify mannosyl-threonyl residue, 100 pmol of synthetic mannosyl-threonine was subject to protein sequencer. The peaks of 3-phenyl-2-thiohydantoin-mannosyl-threonine and related molecule were eluted on the first cycle of the sequencer. A MALDI-TOF mass spectrum was obtained on a Ultraflex mass spectrometer (Bruker Daltonics, Bremen, Germany) in a reflector mode using 2,5-dihydroxybenzoic acid as a matrix. For MS/MS spectra, parent ion was selected ± 15 Da from the observed MH+ value using time gate and re-accelerated (LIFT mode).
Detection of POMT Activity against Various Peptides—We previously detected POMT activity when POMT1 and POMT2 were co-expressed in human embryonic kidney 293T cells using glutathione S-transferase-α-DG as an acceptor (
). In the present study, we examined whether or not synthetic peptides worked as an acceptor. We prepared nine ∼20-amino acid peptides that covered the mucin-like domain of α-DG (Fig. 1). When we measured the POMT activity using these peptides, these peptides were mannosylated to different degrees. The peptide sequence 401–420 worked as a prominent acceptor, followed by the peptide sequence 336–355 (Fig. 2). The POMT activity was 4.03 pmol/h/mg total protein against peptide 401–420 and 2.32 pmol/h/mg total protein against peptide 336–355. The POMT activities against the other peptides were less than peptide 401–420 and 336–355 (<0.63 pmol/h/mg total protein), suggesting the occurrence of α-DG peptide-specific O-mannosylation.
Mucin type O-linked glycosylation is initiated by the action of a family of UDP-GalNAc, polypeptides N-acetylgalactosaminyltransferase (pp-GalNAc-T), which catalyze the transfer of GalNAc from the nucleotide sugar UDP-GalNAc to the hydroxyl group of either Ser or Thr. Peptides derived from the tandem repeat region of mucins, which have a high density of Ser/Thr, were utilized efficiently by all pp-GalNAc-Ts (
). Therefore, we examined whether or not several peptides of mucin tandem repeat regions worked as an acceptor of O-mannosylation (Fig. 1B). Among these peptides, MUC2 and MUC3 showed slight POMT activities (0.7 and 0.49 pmol/h/mg total protein, respectively). These results indicated that mucin peptides worked as poor acceptors for O-mannosylation, suggesting that there are preferred protein sequences for O-mannosylation.
Kinetics Analysis of POMT Activity against Various Peptides—Among seven peptides (peptides 336–355, 364–383, 385–403, 401–420, 451–470, MUC2, and MUC3) examined, peptides 336–355 and 401–420 had the lowest Km values (0.63 and 0.73 mm, respectively) (Table 1 and supplemental Fig. S1). The Km/Vmax values for peptides 336–355 and 401–420 were 0.16 and 0.10, respectively, and the Km/Vmax values for other peptides were 8 times higher than that of peptide 336–355. These results suggest that the affinities of POMT for peptides 336–355 and 401–420 are comparable and that peptides 336–355 and 401–420 are the most suitable acceptors for POMT reaction under these experimental conditions.
Substrate Specificities and Effect of Change of Amino Acids of Peptides—Peptide 336–355 and peptide 401–420 have six and four potential O-mannosylated Ser and Thr residues, respectively (Fig. 1B). To elucidate which Ser or Thr was preferentially modified by O-mannosylation, each Ser or Thr of the peptides was changed to Ala one by one and then the O-mannosylation efficiencies of the mutated peptides were examined. Six mutated peptides were prepared from peptide 336–355 and four peptides were prepared from peptide 401–420 (Fig. 3A). For peptide 336–355, changing Ser-336 or Ser-344 to Ala did not affect the acceptor efficiency of the peptides, whereas changing Thr-341 or Thr-351 to Ala greatly reduced the acceptor efficiency (by 71.7 and 85.1%, respectively) and changing Thr-343 or Thr-353 to Ala moderately reduced the acceptor efficiency (by 48.5 and 39.8%, respectively) (Fig. 3B). For peptide 401–420, changing Thr-418 to Ala slightly affected the acceptor efficiency, whereas changing Thr-404, Thr-406, or Thr-414 to Ala greatly reduced the acceptor efficiency (by 82.3, 69.7, and 92.7%, respectively) (Fig. 3C). The results indicate that Thr-351 of peptide 336–355 and Thr-414 of peptide 401–420 were prominently modified by O-mannosylation. Next, to elucidate the preference of O-mannosylation of Ser and Thr residues, four Thr residues of peptide 401–420 were changed to Ser (Fig. 3A, peptide T-S) and then the O-mannosylation efficiency of peptide T-S was examined. Changing Thr to Ser reduced the acceptor efficiency by 62.2% (Fig. 3C). This result indicates that the POMT O-mannosylates both Thr and Ser but prefers Thr to Ser.
Interestingly, peptides 336–355 and 401–420 have similar sequences. In an alignment analysis of these peptides and adjacent amino acid residues (Fig. 3D), the two sequences have nine identical amino acids (indicated by asterisks) and seven similar amino acids (indicated by dots). Taken together, these results indicate that the similar O-mannosylation efficiencies of the two peptides are due to the similarity of their sequences.
Next, we tried to identify a consensus sequence for O-mannosylation. First, to examine the necessity of C-terminal Pro-Pro and N-terminal Ser-Arg, we prepared peptide 338–357 (Fig. 4A). The POMT activity against peptide 338–357 was almost the same as the activity against peptide 336–355 (by 97.8%) (Fig. 4B). This result indicates that the C-terminal Pro-Pro and N-terminal Ser-Arg have a negligible effect on O-mannosylation efficiency. Then we examined the contribution of nonidentical amino acids between peptides 336–355 and 401–420 on O-mannosylation. To address this question, we prepared peptide X-A in which Ala was substituted for all nonidentical amino acids (Fig. 4A). The mutated peptide X-A reduced the acceptor efficiency by 46.2% (Fig. 4B), suggesting that nonidentical amino acids also are important for efficient O-mannosylation. Then we prepared the peptide 338–355 XA in which Ala-342 was reverted to Pro and Ala-353 was reverted to Thr (indicated by asterisks). The reverted peptide 338–355 XA restored the acceptor efficiency to 89.3% of peptide 336–355. Taken together, these results suggest that the identical amino acid sequence between peptide 401–420 and peptide 336–355 is necessary for efficient O-mannosylation and that (IXPT(P/X)TXPXXXXPTX(T/X)XX) is a consensus sequence for O-mannosylation.
MALDI-TOF MS Analysis and Edman Degradation Analysis—The enzymatic products of peptide 401–420 were analyzed by MALDI-TOF MS and Edman degradation to determine the number of attached mannose residues and their sites of attachment. The mannosylated peptides from the POMT reaction using peptide 401–420 as an acceptor were purified by Con A-Sepharose column followed by reversed-phase HPLC. The POMT reaction mixture showed several peaks around 22∼23 min (supplemental Fig. S2A). A large peak at 23 min derived from the original peptide 401–420 was obtained from the unbound fraction of the Con A-Sepharose column (supplemental Fig. S2B, indicated by closed triangle). Three peaks around 22–23 min recovered from the bound fraction of the Con A-Sepharose column appear to be mannosyl peptides (supplemental Fig. S2C, indicated by open triangle).
The purified mannosylated peptides were subjected to MS analysis (Fig. 5). The monoisotoic mass value calculated from the peptide 401–420 is 2111.1 for (M+H)+. On the other hand, the spectrum of the products showed three m/z values for (M+H)+ of 2597.3, 2435.2, and 2273.2, suggesting that three, two, and one mannosyl residues have been transferred to the peptide, respectively (Fig. 5A). Then each peak was subjected to MS/MS analysis to determine the sites of attachment by mannose. MS/MS with selection of the ion at m/z 2597 resulted in detection of a demannosylated ion m/z 2438 (peptide 401–420 + 2Man) and b-series ion m/z 1685 (b12, peptide 401–412 + 2Man). MS/MS with selection of the ion at m/z 2435 resulted in detection of ions m/z 2274 (peptide 401–420 + Man) and m/z 1522 (b12, peptide 401–412 + Man). MS/MS with selection of the ion at m/z 2273 resulted in detection of ions m/z 2113 (peptide 401–420) and m/z 1358 (b12, peptide 401–412) (Fig. 5B). These results indicated that the POMT enzymatic products of peptide 401–420 have three, two, and one Man residues.
Analysis of the amino acid sequences of these mannosylated peptides to determine the mannosylated Thr residues (supplemental Fig. S3) showed that the mono-mannosylated peptide was mannosylated mainly at Thr-414 and slightly at Thr-404 and Thr-406, the di-mannosylated peptide was mannosylated at Thr-406 and Thr-414, and the tri-mannosylated peptide was mannosylated at Thr-404, Thr-406, and Thr-414. These results indicate that 1) Thr-414 was the residue that was most frequently modified by O-mannosylation, 2) Thr-404 and Thr-406 were sequentially mannosylated after Thr-414, and 3) Thr 418 was not mannosylated. To confirm the conclusions, we performed a time course study of O-mannosylation. As shown in supplemental Fig. S4, mono-mannosylated peptide appeared at the early stage of the reaction (∼15 min), and then the amount of di- and tri-mannosylated peptides increased at later stages. Taken together, these results indicate that O-mannosylation did not occur at random but occurred sequentially.
Effect of Mannosylation of Thr-414 on Substrate Efficiency—To examine whether the presence of mannose on the peptide affects O-mannosylation efficiency, we synthesized a monomannosyl peptide (IRPTMTIPGYVEPT(Man)AVATPP, named as peptide 401–420(T414Man)) and compared kinetic parameters with peptide 401–420 (supplemental Fig. S5). The Km value for peptide 401–420(T414Man) (0.01 mm) was remarkably lower than that of peptide 401–420 (0.73 mm), and the Vmax value for peptide 401–420(T414Man) (9.63 pmol/h/mg) was higher than that of peptide 401–420 (7.11 pmol/h/mg). These results indicate that the mannosylation of Thr-414 in peptide 401–420 facilitates subsequent mannosylation.
In this study, we identified a preferred amino acid sequence for mammalian O-mannosylation using a series of synthetic peptides whose sequences were derived from the mucin-like domain of human α-DG. Our data show that the peptide sequences 336–355 and 401–420 from α-DG are suitable acceptors for O-mannosylation, whereas other sequences including various mucin tandem repeat regions were poor acceptors. Interestingly, peptides 336–355 and 401–420 had very similar amino acid sequences and had comparable Km values. These findings suggest that these peptides have the same affinity for POMT and have what appears to be a consensus sequence (IXPT(P/X)TXPXXXXPTX(T/X)XX) for mammalian O-mannosylation. The positions of Pro and Thr residues in the two peptides are especially similar. A BLAST search for proteins with this sequence turned up only α-DG, suggesting that the primal O-mannosylated protein is α-DG.
We found that Thr residues modified by O-mannosylation in the peptides 336–355 and 401–420 were mannosylated sequentially rather than at random. The substitution analysis showed that Thr-341, Thr-343, Thr-351, and Thr-353 of peptide 336–355 and Thr-404, Thr-406, and Thr-414 of peptide 401–420 are mannosylated (Fig. 3). Interestingly, the effectiveness of substituted Thr depended on the position; the third Thr residues from the N terminus of these peptides (Thr-351 and Thr-414) were most effective, followed by the first Thr residues (Thr-341 and Thr-404) and then the second Thr residues (Thr-343 and Thr-406). These results indicate that three Thr residues in the consensus sequence are necessary for O-mannosylation. Because the third and first Thr residues in the peptides 336–355 and 401–420 are located next to a Pro residue, POMT may prefer a PT motif as an acceptor. The result that the acceptor efficiency of peptide 338–355 XA is higher than that of peptide X-A (Fig. 4) also suggests the presence of a PT motif is necessary for preferential O-mannosylation. However, the PT motif by itself does not seem to be enough for effective O-mannosylation. Although several of the peptides shown in Fig. 1B have a PT motif, some of them did not work as an acceptor at all and others worked poorly as acceptors (Figs. 1B and 2, and Table 1). These results suggest that the preferential common sequence for effective O-mannosylation is present in peptides 336–355 and 401–420. Indeed, the presence of a Thr residue that is preferentially mannosylated was confirmed by MALDI-TOF MS analysis and amino acid sequence analysis of mannosylated peptides. We obtained mono-, di-, and tri-mannosyl peptides from peptide 401–420. Analysis of these products showed that a mono-mannosylated peptide was mannosylated mainly at Thr-414 and a di-mannosylated peptide was mannosylated at Thr-406 and Thr-414, and Thr-404 is the last of the Thr residues to be O-mannosylated. These results indicate that O-mannosylation of the Thr residues is highly ordered. Furthermore, peptide 401–420(T414Man) was a more effective acceptor than peptide 401–420, indicating that the mannosylation of Thr-414 leads to effective subsequent O-mannosylation. Therefore, α-DG appears to be a prominent acceptor of O-mannosylation because it has both the PT motif and the preferential sequences.
O-GalNAc glycosylation is the most common protein modification and is initiated by the action of a family of pp-GalNAcTs. Rules are thought to exist that specify which Ser and Thr residues become decorated with O-GalNAc. However, despite intense investigation, no consensus sequence has emerged that is both necessary and sufficient for O-GalNAc glycosylation to occur. Many nuclear and cytosolic proteins are O-GlcNAc glycosylated at a variety of attachment sites, but a consensus sequence for O-GlcNAc transferase has not been reported (
). On the other hand, O-Fuc glycosylation exists in direct O-linkage to Ser or Thr residues in two different types of Cys knot motifs, epidermal growth factor-like repeats and thrombospondin type 1 repeats. The enzyme responsible for adding O-Fuc to epidermal growth factor repeats was identified as protein O-fucosyltransferase 1 (POFUT1), and the enzyme for adding to thrombospondin type 1 repeats was identified as POFUT2 (
). A consensus sequence for O-Fuc glycosylation in epidermal growth factor is proposed, CysX4–5(Ser/Thr)Cys between the second and third Cys residues, and a consensus sequence for O-Fuc in thrombospondin type 1 repeats is TrpX5CysX2/3Ser/ThrCysX2G between the first and second Cys residues, respectively (
). Both POFUT1 and POFUT2 require a specific sequence for O-Fuc glycosylation. Our results indicate that α-DG has preferred sequences for O-mannosylation. However, it may be that the efficiency of O-mannosylation depends not only on the sequence but also the secondary and tertiary structures.
Protein O-mannosylation is an essential post-translational modification (
). Protein O-mannosylation has also been suggested to be involved in the endoplasmic reticulum quality control system. In yeast, protein O-mannosylation is necessary for intracellular protein trafficking (
). For example, it was found that nonnative proteins are O-mannosylated in the endoplasmic reticulum, which causes them to be excreted from the cell without aggregation and accumulation of aberrant proteins in the endoplasmic reticulum (
), little is known about the molecular pathomechanism of WWS. Our results show that two peptides derived from the mucin domain of α-DG are highly O-mannosylated by POMT but that little or no O-mannosylation occurs in mucin tandem repeat peptides, suggesting that α-DG has multiple O-mannosylated glycans but other proteins have little, if any. The binding of ligands to glycans is known to depend on how the glycans are clustered. Binding of extracellular matrixes such as laminin, neurexin, and agrin may require a cluster of O-mannosyl glycan on α-DG. Therefore, a defect of O-mannosyl glycans in patients with WWS may severely affect α-DG functions but not other protein functions. Future studies may reveal how α-DG glycosylation contributes to muscular dystrophy and neuronal migration disorder and how normal glycosylation can restore functions of DG. Such studies may lead to the development of therapies for incomplete glycosylation-induced dystroglycanopathies.
We thank Fumihiko Ito for assistance in synthesizing mannosyl-threonine.