Glycobiology Advance Access originally published online on August 3, 2005
Glycobiology 2005 15(11):1084-1093; doi:10.1093/glycob/cwj006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hydrophobic Man-1-P derivatives correct abnormal glycosylation in Type I congenital disorder of glycosylation fibroblasts
2 The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037; 3 Department of Applied Biological Science, Tokyo University of Agriculture and Technology, 3-5-8, Saiwai-cho, Fuchu, Tokyo 183-8509, Japan; and 4 Department of Chemistry, Swedish University of Agricultural Sciences, PO Box 7015, SE-75007, Uppsala, Sweden
1 To whom correspondence should be addressed; e-mail: hudson{at}burnham.org
Received on February 4, 2005; revised on June 15, 2005; accepted on June 30, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Patients with Type I congenital disorders of glycosylation (CDG-I) make incomplete lipid-linked oligosaccharides (LLO). These glycans are poorly transferred to proteins resulting in unoccupied glycosylation sequons. Mutations in phosphomannomutase (PMM2) cause CDG-Ia by reducing the activity of PMM, which converts mannose (Man)-6-P to Man-1-P before formation of GDP-Man. These patients have reduced Man-1-P and GDP-Man. To replenish intracellular Man-1-P pools in CDG-Ia cells, we synthesized two hydrophobic, membrane permeable acylated versions of Man-1-P and determined their ability to normalize LLO size and N-glycosylation in CDG-Ia fibroblasts. Both compounds, compound I (diacetoxymethyl 2,3,4,6-tetra-O-acetyl-
-D-mannopyranosyl phosphate) (C-I) and compound II (diacetoxymethyl 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate) (C-II), contain two acetoxymethyl (CH2OAc) groups O-linked to phosphorous. C-I contains acetyl esters and C-II contains ethylcarbonate (CO2Et) esters on the Man residue. Both C-I and C-II normalized truncated LLO, but C-II was about 2-fold more efficient than C-I. C-II replenished the GDP-Man pool in CDG-Ia cells and was more efficiently incorporated into glycoproteins than exogenous Man at low concentrations (25–75 mM). In a glycosylation assay of DNaseI in CDG-Ia cells, C-II restored glycosylation to control cell levels. C-II also corrected impaired LLO biosynthesis in cells from a Dolichol (Dol)-P-Man deficient patient (CDG-Ie) and partially corrected LLO in cells from an ALG12 mannosyltransferase-deficient patient (CDG-Ig), whereas cells from an ALG3-deficient patient (CDG-Id) and from an MPDU1-deficient patient (CDG-If) were not corrected. These results validate the general concept of using pro-Man-1-P substrates as potential therapeutics for CDG-I patients. Key words: congenital disorders of glycosylation / GDP-Man / Man-1-phosphate phosphomannomutase / N-glycosylation
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
The congenital disorders of glycosylation (CDG) are a diverse group of inherited diseases that impair protein glycosylation (Freeze, 2001
; Marquardt and Denecke, 2003) (Figure 1). Type I disorders result from mutations in the N-linked oligosaccharide pathway and lead to unoccupied glycosylation sites on some proteins. Twelve types of CDG (Ia-IL) in this group include defects in monosaccharide activation, glycosyltransferases, or in noncatalytic partners involved in the biosynthesis of the glucose (Glc)3-mannose(Man)9-N-acetyl glucosamine (GlcNAc)2-P-P-Dol lipid-linked oligosaccharide (LLO) (Jaeken 2003; Marquardt and Denecke, 2003).
|
By far, the most common disorder is CDG-Ia, which is caused by mutations in PMM2 encoding phosphomannomutase (PMM) (Van Schaftingen and Jaeken, 1995). This enzyme converts Man-6-P 
Man-1-P, the immediate precursor for GDP-Man. Patients with CDG-Ib are deficient in phosphomannose isomerase (PMI), which converts fructose-6-P Man-6-P, and they also synthesize insufficient GDP-Man. CDG-Ie patients have mutations in the synthesis of Dol-P-Man, the donor for the last four Man residues of the LLO molecule. Other patients have defects in GDP-Man or Dol-P-Man requiring LLO-specific mannosyltransferases (CDG-Id, -Ig -Ii, -Ik, and -IL) or in noncatalytic proteins (CDG-If) that reduce the supply of functional LLO.
Addition of 1 mM Man to fibroblasts from both CDG-Ia and -Ib patients corrects their LLO size and restores depleted GDP-Man pools (Panneerselvam and Freeze, 1996). CDG-Ib patients respond well to dietary supplements of Man since the sugar is transported into the cell and directly phosphorylated to Man-6-P, bypassing the defect (Niehues et al., 1998). In contrast, CDG-Ia patients do not respond to Man treatment (Mayatepek et al., 1997; Kjaergaard et al., 1998; Mayatepek and Kohlmüller, 1998).Unfortunately Man-1-P is neither transported nor membrane permeable, making it impossible to provide patients with the missing intermediate. Theoretically, a membrane-permeable derivative of Man-1-P might complement CDG-Ia cells by raising the GDP-Man pools. Since other types of CDG involve GDP-Man-dependent steps, increasing its intracellular concentration might benefit them as well.
Peracetylation of both mono- and disaccharides improves their membrane permeability (Schultz et al., 1993; Sarkar et al., 2000). Acetoxymethylation of phosphates neutralizes the negative charges and increases bioavailability since cytoplasmic esterases liberate the desired parent compound (Rudolf et al., 1998). Encouraged by preliminary studies by others (Rutschow et al., 2002; Muus et al., 2004), we synthesized two Man-1-P pro-compounds with acetoxymethylated phosphates and sugar alcohol groups protected by ethylcarbonate or acetyl groups and incubated patient fibroblasts with them. We found that these compounds normalize LLO synthesis in CDG-Ia cells and in several other types of CDG cells. Moreover, Man-1-P from the artificial compounds entered the GDP-Man pool of CDG-Ia cells more efficiently than free Man.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Synthesis of Man-1-P compounds
Since Man-1-P is not transported into cells and cannot diffuse through the plasma membrane, we synthesized two membrane-permeable derivatives. Compound I (C-I) is diacetoxymethyl-2,3,4,6-tetra-O-acetyl-
-D-mannopyranosyl phosphate (5 in Figure 2) and compound II (C-II) is diacetoxymethyl-2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate (10 in Figure 2). Starting from Man, per-O-acylation and subsequent treatment of the products (1 or 6) with hydrogen bromide in acetic acid gave the acylated pyranosyl bromides (2 or 7). Reaction with dibenzyl phosphate and silver carbonate generated the dibenzyl -mannopyranosyl phosphates (3 or 8). Subsequent hydrogenation on palladium yielded the corresponding -mannopyranosyl phosphates (4 or 9), which were reacted with acetoxymethyl bromide and diisopropylethylamine to yield C-I (5) and C-II (10) in 62 and 52% yield, respectively.
|
Stability of C-I and C-II in culture medium and human serum
To assess the stability and degradation patterns of the compounds in serum and culture medium two methods were used. Using high pressure liquid chromatography (HPLC) separation after back extraction of the compounds with ethyl acetate, the compound half lives under different conditions were determined (data not shown). C-I was considerably more unstable than C-II, with a half life in human serum of 
1.5 min and in cell culture medium of 6 min, whereas C-II has a half life of 2.5 min in serum and 15 min in medium. This potentially explains the lower efficiency of C-I to correct glycosylation, as described below.
To define the degradation products of the compounds in the presence of human serum or fetal bovine serum (FBS), C-I and C-II were added to the sera and the reactions were followed by mass spectrometry. After 30 min in human serum, neither of the compounds could be detected. C-I showed two main degradation products with the molecular weights 545.3 (acetoxymethyl-2,3,4,6-tetra-O-acetyl--D-mannopyranosyl phosphate sodium salt) and 371.3 (2,3,4,6-tetra-O-acetyl--D-mannopyranose sodium salt), respectively (data not shown). C-II showed the two corresponding degradation products, 665.4 (acetoxymethyl-2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate sodium salt) and 491.4 (2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranose sodium salt).
General characterization and toxicity
In the cell, C-I is hydrolyzed to Man-1-P + 6AcOH + 2CH2O and C-II generates Man-1-P + 2AcOH + 2CH2O + 4C2H5OH + 4CO2. We analyzed the effects of these potentially toxic products on cellular protein synthesis and growth. C-I does not affect cell growth (24 h), but 100 µM C-II inhibits growth by 50% and growth is completely inhibited at 250 µM. (Figure 3A). Up to 250 µM, C-I had no effects on protein synthesis (data not shown), but C-II decreased protein synthesis above 100 µM in control, CDG-Ia and CDG-Ib cells (Figure 3B). CDG-Ia fibroblasts incubated in 150 µM C-II for 25 h showed very few Trypan Blue-stained cells, whereas at 200 µM, more than 60% of the cells stained (data not shown). Since our primary goal is to determine the effects on LLO and glycoprotein synthesis, we concentrated on short-term incubations.
|
C-I and C-II normalize LLO synthesis in CDG-Ia cells
CDG-Ia cells lack sufficient GDP-Man and synthesize very little full size Glc3Man9GlcNAc2 LLO in the medium containing 2 mM Glc. Instead they make a series of truncated LLO species ranging in size from Man2 to Man5GlcNAc2, (Figure 4A, upper chromatogram). We incubated CDG-Ia fibroblasts with various amounts of either compound and analyzed LLO sugar chain size by HPLC (LLO profile with 200 µM C-II shown in Figure 4A, lower chromatogram). Both compounds completely normalized LLO size, but C-II was about 2-fold more effective than C-I (Figure 4B). These results suggested that the Man-1-P compounds supplemented the depleted GDP-Man pools in CDG-Ia cells. To confirm this, we directly measured the pools in control and CDG-Ia fibroblasts in the presence of 0, 100, or 250 µM C-II. The GDP-Man pool is about 3.5-fold smaller in CDG-Ia cells compared to control, but addition of C-II considerably increased it in the patient cells (Table I).
|
|
CDG-Ia cells use C-II-derived Man-1-P more efficiently than Man-1-P derived from Man
Man (1 mM) can correct GDP-Man pools and normalize LLO profiles in CDG-Ia and CDG-Ib cells (Panneerselvam and Freeze, 1996). To determine whether Man-1-P derived from C-II is used more efficiently than Man, we measured the ability of each to reduce incorporation of tracer [2-3H]Man into glycoproteins. The more efficiently an unlabeled substrate contributes to glycoprotein synthesis, the more efficiently it will decrease [2-3H]Man incorporation into glycoproteins. We labeled cells with [2-3H]Man and [35S]-cysteine/methionine (Tran[35S]LabelTM), together with increasing concentrations of the two substrates. C-II is more efficient than Man in CDG-Ia cells (n = 4) at low (nontoxic) concentrations, suggesting that C-II efficiently enters the glycosylation pathway. C-II and Man are about equally efficient in control cells (n = 2), but in PMI-deficient CDG-Ib cells, Man contributes much more efficiently than C-II (n = 2) (Figure 5A–C).
|
C-II corrects glycosylation of DNaseI in CDG-Ia fibroblasts
To investigate the effects of Man versus C-II on the rate of N-glycosylation we employed a glycosylation assay described by Nishikawa and Mizuno (2001). In this assay, a construct of DNaseI, having one potential glycosylation site (glycosylated to 70–80% in control cells), is introduced through adenoviral transduction into fibroblasts. The effect of added compounds on its glycosylation can then be monitored by 35S-amino acid metabolic labeling in the presence of compound, immunoprecipitation, and separation of glycosylated and nonglycosylated species by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). Phosphorimaging is used to quantify the amount of label in each species. In the CDG-Ia fibroblasts used here, we found 60% of the label in the glycosylated version of DNaseI (Figure 6). Addition of 200 µM Man did not change this proportion, whereas addition of C-II in increasing concentrations increased the percentage of glycosylated molecules. At 200 µM C-II, a clear toxicity was noted with a decrease in total label, but at 150 µM C-II, only a slight decrease in label incorporation was noted, where most (75%) of the label was in the glycosylated version. This shows that C-II not only alters the steady-state distribution of LLOs but has a direct effect on the extent of protein N-glycosylation.
|
C-II corrects truncated LLO synthesis in other Type I congenital disorders of glycosylation cells
Deficiencies in enzymes or proteins that require GDP-Man or Dol-P-Man (Marquardt and Denecke, 2003) cause other types of Type I congenital disorders of glycosylation (CDG-I) and produce truncated LLO glycans. Since these defects are usually partial, increasing GDP-Man might normalize truncated LLO structures, especially where the mutations increase the Km for GDP-Man.
We chose several types of CDG-I cells, incubated them with C-I or C-II, and analyzed the LLO structures. CDG-Ie results from mutations in DPM1 (Imbach et al., 2000; Kim et al., 2000), the catalytic subunit for Dol-P-Man synthase (Dol-P + GDP-Man Æ Dol-P-Man). A homozygous point mutation in DPM1 of patient C-II increases the Km of GDP-Man about 6-fold yielding a truncated Man5GlcNAc2-P-P-Dol precursor (Figure 7, upper chromatogram), but providing either C-I (300 µM) or C-II (300 µM) corrected synthesis toward full-size LLO. Again, C-II is more efficient than C-I (Figure 7, middle and lower chromatograms). Another patient (MA) with an unknown defect (CDG-Ix) synthesizes primarily Man5GlcNAc2-P-P-Dol precursor (Figure 8, upper chromatogram). 300 µM C-I partially corrects the abnormal pattern (middle chromatogram) whereas 300 µM C-II fully corrects it (lower chromatogram), suggesting that the defect may involve an enzyme that utilizes Dol-P-Man. Because C-II has greater efficiency, further studies only tested this compound. Fibroblasts from a CDG-Ig patient (MH), who has a deficiency in ALG12, synthesizes only Man7GlcNAc2-P-P-Dol precursor. C-II (150 µM) slightly improved the LLO profile (5–10% Glc3Man9GlcNAc2-P-P-Dol), but even 500 µM did not completely correct it (20% full size LLO, Table II). We also tested cells from a CDG-If patient (AL) who has a deficiency in MPDU1. This gene encodes a noncatalytic protein needed for Dol-P-Man and Dol-P-Glc addition to LLO. C-II (500 µM) only slightly improved the abnormal LLO pattern (Table II). Finally, one CDG-Id patient with a deficiency in ALG3, the mannosyltransferase required to convert Man5GlcNAc2-P-P-Dol to Man6-GlcNAc2-P-P-Dol showed no improvement in LLO pattern with 300 µM C-II. The results as summarized in Table II clearly show that synthetic Man-1-P precursor improves truncated LLO in several types of CDG. However, the extent of improvement depends on both the type of CDG and the specific mutations in those genes.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Cultured fibroblasts from CDG-Ia patients have small GDP-Man pools, which can be corrected with 1 mM Man (Panneerselvam and Freeze, 1996
), but Man therapy failed in patients (Marquardt and Denecke, 2003). As an alternative therapy for them, we synthesized two hydrophobic, membrane permeable versions of Man-1-P, using a different strategy than previously reported (Rutschow et al., 2002; Muus et al., 2004). Acetoxymethyl groups neutralized the negative charges on the phosphate and the hydrophobicity of the hydroxyl groups was increased by their conversion to either acetyl (C-I) or ethyl carbonate (C-II) esters. Once inside the cells, cytosolic esterases generate the parent compounds. C-I appeared to be nontoxic even at high concentrations, whereas C-II began to inhibit protein synthesis above 100 µM. This suggested that either C-I was less permeable than C-II or that the products released from C-II on hydrolysis were more toxic than the acetate released from C-I. Since both C-I and C-II contain two acetoxymethyl esters, they are unlikely to cause the toxicity. Although it is expected that the same amount of formaldehyde would be generated in the controls as in the experimental samples, it cannot be ruled out that the formaldehyde has some toxic effect. The half life of C-I is much shorter than that of C-II, both in culture medium and human serum. This may, at least partially account both for the higher efficiency and higher toxicity of C-II. More studies are needed to assess the basis of toxicity and to design other less toxic derivatives. The main goal of these studies was to prove that these derivatives could rescue defective glycosylation in CDG cells. Both C-I and C-II almost completely corrected the synthesis of truncated LLO in two different CDG-Ia cell lines. C-II gave 50% correction at about 65 µM while C-I was about 2-fold less efficient, requiring about 130 µM for 50% correction. C-II increased the GDP-Man pool size in CDG-Ia cells, as expected, but there was also a small increase in the GDP-Man pool in normal cells. Since 1 mM Man fully corrected LLO size and GDP-Man pool in CDG-Ia and CDG-Ib cells, it was important to compare the efficiency of Man-1-P and Man. In control fibroblasts, C-II and Man entered the biosynthetic pathway with equal efficiency. This outcome was not predictable since there are many competing and counteracting factors (Figure 1). For instance, Man efficiently enters the cell through facilitated diffusion transporter, but C-I and C-II do not. Once inside the cell, the great majority of Man-6-P (95–98%) derived from Man is rapidly converted to fructose-6-P by PMI, leaving only a small amount for glycosylation. The relative activities and substrate affinities of the enzymes competing for Man-6-P and Man-1-P determine whether Man or the compounds are used more efficiently. In PMM-deficient CDG-Ia cells, C-II is more efficient than Man, because the normal PMI activity catabolizes most of the Man-6-P and prevents its accumulation, whereas reduced PMM activity encourages entry of Man-1-P into glycosylation. In contrast, PMI-deficient CDG-Ib cells use Man more efficiently than C-II. This finding underscores the role of PMI as both a provider of Man-6-P for glycosylation and a regulator against accumulation of Man-6-P within the cell. C-II was also able to correct glycosylation of an over-expressed glycoprotein (DNaseI) in CDG-Ia cells, whereas Man had no effect. This shows that the effects of C-II seen on LLO correction not only reflect a change in the steady-state of the LLO intermediates, but actually provides a dose-dependent increase in glycosylation.
C-II slightly increased the GDP-Man pool in control fibroblasts, suggesting that fibroblasts from other CDG-I patients with defects in LLO synthesis might benefit from C-II. Indeed, CDG-Ie and CDG-Ix both produced normal-sized LLO when given C-II. Cells from CDG-Id, CDG-If, and CDG-Ig were not corrected or only slightly corrected. It is likely that both the gene and its specific mutations determine whether increasing GDP-Man improves LLO. Km have not been determined for most of the LLO biosynthetic reactions, and it is unknown which mutations affect catalytic activity or protein stability. An exception is Dol-P-Man synthase (CDG-Ie) where the Km for GDP-Man is 6 µM, but a homozygous point mutation in patient CH increases the Km to about 40 µM (Kim et al., 2000). Increasing GDP-Man is predicted to have a positive effect and increase the amount of Dol-P-Man synthesized. By analogy, the corrected CDG-Ix patient probably has mutations that increase the Km for a GDP-Man-dependent reaction, perhaps involving the Dol-P-Man donor.
These initial experiments provide evidence that hydrophobic Man-1-P compounds may be useful as a potential therapeutic. Synthesis of other compounds with lower toxicity and greater stability must be addressed in future studies.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Materials
Minimal essential medium-alpha (
-MEM) and Dulbecco’s modified Eagle’s medium (DMEM) w/o Glc, L-glutamine and antibiotics were all obtained from Invitrogen (Carlsbad, CA). FBS was from HyClone (Logan, UT), [2–3H]Man (20 Ci/mmol) from Perkin Elmer (Boston, MA), and the Microsorb-MV NH2 HPLC column from Varian Instruments (Walnut Creek, CA). Tran[35S]LabelTM 1175 Ci/mmol; 70% L-methionine was from MP Biomedicals (Irvine, CA). GDP-Man and routine reagents were purchased from Sigma-Aldrich Chemical (St Louis, MO). Human serum was prepared from blood drawn just before the experiment.
Synthesis of Man-1-P derivatives
General
All solvents and reagents used were reagent grade or better and were used as received. The analytical thin-layer chromatography plates was performed on Merck silica gel 60 F254 plates (glass backed, 0.25 mm). Compounds were visualized by dipping the plates in a cerium sulfate-ammonium molybdate solution followed by heating. Liquid column chromatography was performed on silica gel (standard grade, 60 Å, 32–63 µm) provided by Fluka. 1H and 13C-nuclear magnetic resonance (NMR) spectra were recorded at 300 MHz and 75 MHz, respectively (Varian Inova 300). The proton and carbon chemical shifts were referenced to the solvent residual peak for the samples in CDCl3 (
H = 7.27 and C = 77.0 ppm). The electrospray ionization mass spectrometry spectra were recorded on a Waters Micromass ZQ (Waters, Milford, CT) with a cone voltage of 60V, source temperature of 90°C, and a desolvation temperature of 300°C.
Preparation of acetoxymethyl bromide
Anhydrous zinc chloride (freshly fused, 100 mg, 0.84 mmol) was added at 25°C to a solution of acetyl bromide (11.70 g, 9.50 mmol) in methylene chloride (CH2Cl2, 10 mL), and the mixture was cooled to 0°C. Paraformaldehyde (3 g, 100 mmol) was added, and the reaction mixture was stirred overnight at 25°C. Distillation of the mixture at 750 mm Hg using a Vigreux headpiece (15 cm) gave first CH2Cl2 and unreacted acetyl bromide (74–77°C), followed by the crude product boiling at 120–140°C/750 mm Hg. This compound was found to be of good purity and was used as it is in the subsequent experiments. 1H NMR (CDCl3), : 2.13 (s, 3H), 5.78 (s, 2H); 13C NMR (CDCl3), : 20.8 (CH3), 57.0 (CH2), 168.8 (C=O).
Diacetoxymethyl 2,3,4,6-tetra-O-acetyl--D-mannopyranosyl phosphate (C-I, 5)
D-Mannose (3.0 g) in pyridine (50 mL) and acetic anhydride (25 mL) was stirred overnight at room temperature. Coevaporation with toluene yielded 1,2,3,4,6-penta-O-acetyl-/ß-D-mannopyranose as a syrup (1, 7.1 g), which was dissolved in CH2Cl2 (10 mL) and HBr in acetic acid (30%, 30 mL) was added. The mixture was stirred at room temperature and its progress was monitored by thin-layer chromatography (TLC) (silica gel, toluene-EtOAc 6:4). After 2 h, the reaction was poured into ice water/CH2Cl2 mixture, and the organic layer was washed with ice-cold saturated aqueous sodium bicarbonate, water, dried (magnesium sulfate), and evaporated. The residual crude 2,3,4,6-tetra-O-acetyl--D-mannopyranosyl bromide (2, 6.5 g) was stored at 4°C. This material (1.6 g) was dissolved in CH2Cl2 (12 mL) containing dibenzyl phosphate (1.6 g) and freshly activated powdered molecular sieves (4Å, 4 g) and was stirred at room temperature for 30 min, after which silver carbonate (4.0 g) was added. Progress of the reaction was monitored by TLC (toluene-EtOAc 6:4). A slower, UV-absorbing spot appeared, with the disappearance of the bromo-sugar. After 12 h, the mixture was filtered through Celite, washed with aqueous sodium thiosulfate, water, dried (magnesium sulfate), and evaporated. The residue was purified by chromatography on silica gel (toluene-EtOAc, 6:4), and fractions containing the main compound were combined and evaporated. The residue (1.40 g) showed a strong peak at m/z 631.1 (M + Na) in the positive ion ESI–MS spectrum, as expected for dibenzyl 2,3,4,6-tetra-O-acetyl--D-mannopyranosyl phosphate (3). This material (600 mg) was taken up in EtOAc (18 mL) and added to a slurry of Pd/C (200 mg) in EtOAc (2 mL). The mixture was stirred under an atmosphere of hydrogen at room temperature until the TLC (EtOAc-acetic acid-methanol-water, 12:3:3:2) showed complete debenzylation of the phosphate moiety. The mixture was filtered through Celite, and the filtrate was evaporated and coevaporated with dry acetonitrile. The residue, containing crude 2,3,4,6-tetra-O-acetyl -D-mannopyranosyl phosphate (4) was taken up in dry acetonitrile (10 mL) and stirred at room temperature while adding acetoxymethyl bromide (1.6 mL) and diisopropylethylamine (1.6 mL, 9.6 mmol). Stirring was continued overnight while TLC analysis showed a major new, fast-migrating spot (Rf 0.95 in EtOAc-acetic acid-methanol-water, 12:3:3:2; Rf 0.3 in toluene-EtOAc 6:4). The mixture was partitioned between EtOAc and aqueous sulfuric acid (0.5 M), washed with aqueous saturated sodium bicarbonate, and evaporated. The resulting oil was purified by column chromatography (toluene-EtOAc 4:6). Appropriate fractions were collected and concentrated to give diacetoxymethyl 2,3,4,6-tetra-O-acetyl--D-mannopyranosyl phosphate as a syrup (5, 350 mg). The material showed the expected strong peak at m/z 595.15 (M + Na) in the positive ion ESI-MS spectrum. 1H NMR, : 5.70–5.63 (m, 5H), 5.34–5.31 (m, 3H), 4.27 (dd, 1H), 4.19 (m, 1H, 4.13 (dd, 1H), 2.17 (s, 6H), 2.16 (s, 3H), 2.10 (s, 3H), 2.05 (s, 3H), 1.99 (s, 3H); 13C, : 170.5–169.1 (6 C), 95.7 (d, 2JC-1,P = 5.25, C-1), 82.6 (d), 82.5 (d), 70.6, 68.4 (d, J = 11.9), 68.1, 65.0, 61.8, 20.7–20.5 (6 C) ppm.
Diacetoxymethyl 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate (C-II, 10)
D-Mannose (3 g, 16.67 mmol) in pyridine (50 mL) was cooled in an ice bath and ethyl chloroformate (15.8 mL) was added dropwise, and the reaction was stirred overnight at room temperature. TLC (toluene-EtOAc 8:2) showed the formation of a major spot (Rf 0.5). Water (2 mL) was added to the mixture, and after 15 min the mixture was partitioned between EtOAc and water. The organic phase was washed with water, aqueous sulfuric acid (1 M), aqueous saturated sodium bicarbonate, dried (magnesium sulfate), and evaporated. The resulting orange-colored syrupy residue (8.9 g) was an anomeric mixture of 1,2,3,4,6-penta-O-ethyloxycarbonyl-D-mannopyranoses (6). This material (2.0 g) was dissolved in CH2Cl2 (2 mL) and HBr in acetic acid (30 %, 8 mL) was added. The mixture was stirred at room temperature until TLC (toluene-EtOAc 8:2) showed accumulation of a product with a slightly higher Rf value ( 4h). The mixture was partitioned between ice water and CH2Cl2, the organic layer was washed with ice-cold aqueous saturated sodium bicarbonate, water, dried (magnesium sulfate), and evaporated. This material, 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl bromide (7, 1.16 g) was dissolved in dry CH2Cl2 (12 mL), and dibenzyl phosphate (1.0 g) and activated powdered molecular sieves (4Å, 2.0 g) were added and the mixture stirred for 30 min, after which silver carbonate (2 g) was added. The mixture was stirred at room temperature overnight, and the reaction was monitored by TLC (toluene-EtOAc 6:4). A major slower, UV-absorbing spot developed, as the bromo-sugar disappeared. The mixture was filtered through celite, washed with aqueous sodium thiosulfate, water, dried (magnesium sulfate), and evaporated. The residue was purified by chromatography on silica gel (toluene-EtOAc, 7:3), and fractions containing the main compound were combined and evaporated. The residue (0.67 g) showed a strong peak at m/z 751.2 (M + Na) in the positive ion ESI-MS spectrum, as expected for dibenzyl 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate (8). This material (400 mg) was taken up in EtOAc (8 mL) and added to a slurry of Pd/C (130 mg) in EtOAc (2 mL). The mixture was stirred under an atmosphere of hydrogen at room temperature until the TLC (EtOAc-acetic acid-methanol-water, 12:3:3:2) showed complete deprotection of the phosphate. The mixture was filtered through Celite and the filtrate was evaporated and coevaporated with dry acetonitrile. The residue, containing crude 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate (9), was taken up in dry acetonitrile (6 mL), and acetoxymethyl bromide (0.8 mL, 8 mmol) and diisopropylethylamine (0.8 mL, 4.8 mmol) were added. The mixture was stirred overnight at room temperature. TLC showed essentially a single new spot (Rf 0.4 in toluene-EtOAc 6:4). The mixture was partitioned between EtOAc and aqueous sulfuric acid (0.5 M), washed with aqueous saturated sodium bicarbonate, brine, and evaporated. The residue was purified by chromatography on silica gel (toluene-EtOAc 6:4) to afford diacetoxymethyl 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate (10, 200 mg) as a colorless oil. The material showed the expected strong peak at m/z 715.35 (M + Na) in the positive ion ESI-MS spectrum. NMR data, 1H, : 5.77 (dd, 3JH-1,H-2 = 1.9 Hz, 3JH-1,P = 6.0 Hz), 5.58–5.72 (m, 4H), 5.28 (t, 1H), 5.16–4.98 (m, 2H), 4.30–4.12 (m, 11H), 2.15 (s, 3H), 2.14 (s, 3H), 1.35–1.22 (m, 12H) ppm; 13C NMR, : 169.2, 169.1, 154.7, 154.0, 153.9, 153.7, 95.3 (d, 2JC-1,P = 5.30, C-1), 82.7, 82.6, 71.7, 71.4 (d, J = 12.4 Hz), 70.4, 69.0, 65.2, 65.1, 64.90, 64.8, 64.4, 20.6, 20.5, 14.2–14.0 (4 C) ppm.
Half-life determination by HPLC
About 50 mM solutions of the compounds in ethyl acetate were added to either 100% human serum or to cell culture medium (DMEM) of a final concentration of 300 µM and incubated for 0–60 min. After incubation, the compounds were back extracted in an equal volume of ethyl acetate, and the upper phase was removed and evaporated under a stream of nitrogen. The efficiency of the back extraction was close to 100% as determined by mass spectrometry. The dried upper phase was reconstituted in ethyl acetate and separated using a linear gradient of acetonitrile (35–65% over 20 min) on a 4.5 x 250-mm Microsorb-MV (NH2) column. The compounds were monitored by UV absorption at 210 nm. The half-lives were determined by measuring the remaining amount of the compound investigated at four different time points and plotting the degradation slope. To assure that the UV absorbing material only consisted of the compound assayed, all peaks were characterized by mass spectrometry.
Cells
Control human dermal fibroblasts were obtained from ATCC (Rockville, MD). CDG patient cells were obtained by informed consent from physicians treating the patients. Cells were grown in -MEM with 5 mM Glc, 2 mM L-glutamine, and 10% FBS at 37°C in a 5% CO2 atmosphere.
Growth inhibition and Trypan Blue exclusion
To assess the cytotoxicity of C-I and C-II, control fibroblasts were plated in 6-well plates (23,500–50,000 cells/well) and incubated with DMEM, 10% FBS, for 16 and 25 h at different concentrations of C-I and C-II. The cells were thereafter trypsinized and counted. An aliquot was mixed (1:1) with 0.2% Trypan Blue solution and the percentage of cells with cytoplasmatic staining was assessed by standard light microscopy.
LLO correction
[2-3H]Man labeling for LLO analysis used cells grown in 60-mm dishes to 75–80% confluency. Medium was removed and replaced with DMEM containing 2 mM Glc. Compounds were dissolved in ethanol at 70 mM and diluted to the appropriate concentration in prewarmed medium and added to cells for 15 min at 37°C before addition of 100 µCi/mL of [2-3H]Man. Incubation was continued for 30 min, and the LLOs were isolated and analyzed as described previously (Eklund et al., 2005).
Competition of [2-3H]Man incorporation into glycoproteins
Cells were grown in 12-well plates to approximately 80% confluency. C-II and Man were dissolved in ethanol at 70 mM and diluted in preheated DMEM with 10% dialyzed FBS to appropriate concentrations. The cells were washed with PBS, added the substances in DMEM (1 mL/well), and after 15 min incubation 40 µCi [2-3H]Man and 5 µCi Tran[35S]-labelTM per well were added. The incubations were terminated after another hour by removing the medium, washing 3 times with PBS, and scraping the cells in 500 µL/well 1% Triton X-100. Aliquots were TCA precipitated and the protein content determined using the bicinchoninic acid ZQ method. Other aliquots were TCA precipitated in the presence of 200 µg BSA, the resulting pellets dissolved in 50 mM NaOH and the radioactivity determined using liquid scintillation counting. The relative decrease in 3H-incorporation compared to untreated control was determined as the ratio of 3H/35S in each sample normalized to the 3H/35S ratio in untreated and expressed as percentage of control. The protein synthesis was determined as 35S/mg protein and expressed as percentage of control.
Glycosylation of DNaseI
The effect on glycosylation of DNaseI in CDG-Ia fibroblasts by Man or C-II was investigated essentially as described earlier (Nishikawa and Mizuno, 2001). However, due to the short half-life of C-II, the labeling time was changed to 15 min in the presence of Man or C-II at different concentrations, followed by a 2 h chase in cold medium. The amount of 35S-labeled Cys/Met was 0.5 mCi/mL during the labeling.
Measurement of GDP-Man pools
Control and CDG-Ia patient fibroblasts were grown in 150 mm plates to 80% confluency. They were added -MEM with 10% dialyzed FBS containing 0, 100, or 250 µM C-II and incubated for 30 min. The cells were then scraped in serum-free media, recovered by centrifugation and extracted three times with chloroform/methanol/water (10:20:8). Chloroform and water (final ratio 10:10:9 (chloroform/methanol/water) were added to the pooled extracts, which were mixed thoroughly and centrifuged at 420 x g for 5 min. The aqueous phase was collected and subjected to HPLC separation on Dionex Carbopac. Aliquots of these samples were run twice, either in water or 10 mM NaOH, with an increasing gradient of sodium acetate (0.0–1.0 M from 10 to 60 min). Standard nucleotide sugars were used to calibrate the Dionex column. The results were normalized to the protein content of the extracted pellets and expressed as pmole/mg.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Dr. G. Srikrishna is acknowledged for advice and technical assistance. Supported by R01 DK065091 to H.H.F, a postdoctoral stipend from STINT/VR, Sweden, (K2004-99PK-14887-02B) to E.A.E. and the Collaborative Project Grant R24 GM61894 (Principal Investigator Dr. A. Varki, UCSD, San Diego, CA). The Glycotechnology Core Facility at the Glycobiology Research and Training Center at UCSD is acknowledged for the nucleotide sugar pool measurements.
|
Abbreviations |
|---|
CDG, congenital disorders of glycosylation; CDG-I, Type I congenital disorders of glycosylation; C-I, compound I (diacetoxymethyl 2,3,4,6-tetra-O-acetyl-
-D-mannopyranosyl phosphate); C-II, compound II (diacetoxymethyl 2,3,4,6-tetra-O-ethyloxycarbonyl--D-mannopyranosyl phosphate); DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; Glc, glucose; GlcNAc, N-acetyl glucosamine; HPAC, high pressure liquid chromatography; LLO, lipid-linked oligosaccharide; Man, mannose; PMI, phosphomannose isomerase; PMM, phosphomannomutase, -MEM, minimal essential medium alpha |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Eklund, E.A., Newell, J.W., Sun, L., Seo, N.S., Alper, G., Willert, J., and Freeze, H.H. (2005) Molecular and clinical description of the first US patients with congenital disorder of glycosylation Ig. Mol. Genet. Metab., 84, 25–31.[CrossRef][Web of Science][Medline]
Freeze, H.H. (2001) Update and perspectives on congenital disorders of glycosylation. Glycobiology, 11, 129R–143R.
Imbach, T., Schenk, B., Schollen, E., Burda, P., Stutz, A., Grunewald, S., Bailie, N.M., King, M.D., Jaeken, J., Matthijs, G., and others. (2000) Deficiency of dolichol-phosphate-mannose synthase-1 causes congenital disorder of glycosylation type Ie. J. Clin. Invest., 105, 233–239.[Web of Science][Medline]
Jaeken, J. (2003) Komrower lecture. Congenital disorders of glycosylation (CDG): it’s all in it! J. Inherit. Metab. Dis., 26, 99–118.[CrossRef][Web of Science][Medline]
Kim, S., Westphal, V., Srikrishna, G., Mehta, D.P., Peterson, S., Filiano, J., Karnes, P.S., Patterson, M.C., and Freeze, H.H. (2000) Dolichol phosphate mannose synthase (DPM1) mutations define congenital disorder of glycosylation Ie (CDG-Ie). J. Clin. Invest., 105, 191–198.[Web of Science][Medline]
Kjaergaard, S., Kristiansson, B., Stibler, H., Freeze, H.H., Schwartz, M., Martinsson, T., and Skovby, F. (1998) Failure of short-term mannose therapy of patients with carbohydrate-deficient glycoprotein syndrome type 1A. Acta Paediatr., 87, 884–888.[CrossRef][Web of Science][Medline]
Marquardt, T. and Denecke, J. (2003) Congenital disorders of glycosylation: review of their molecular bases, clinical presentations and specific therapies. Eur. J. Pediatr., 162, 359–379.[Web of Science][Medline]
Mayatepek, E. and Kohlmüller, D. (1998) Mannose supplementation in carbohydrate-deficient glycoprotein syndrome type I and phosphomannomutase deficiency. Eur. J. Pediatr., 157, 605–606.[CrossRef][Medline]
Mayatepek, E., Schroder, M., Kohlmuller, D., Bieger, W.P., and Nutzenadel, W. (1997) Continuous mannose infusion in carbohydrate-deficient glycoprotein syndrome type I. Acta Paediatr., 86, 1138–1140.[Web of Science][Medline]
Muus, U., Kranz, C., Marquardt, T., and Meier, C. (2004) Cyclosaligenyl-mannose-1-monophosphates as a new strategy in CDG-Ia therapy: hydrolysis, mechanistic insights and biological activity. Eur. J. Org. Chem., 2004, 1228–1235.[CrossRef]
Niehues, R., Hasilik, M., Alton, G., Korner, C., Schiebe-Sukumar, M., Koch, H.G., Zimmer, K.P., Wu, R., Harms, E., Reiter, K., and others. (1998) Carbohydrate-deficient glycoprotein syndrome type Ib. Phosphomannose isomerase deficiency and mannose therapy. J. Clin. Invest., 101, 1414–1420.[Web of Science][Medline]
Nishikawa, A. and Mizuno, S. (2001) The efficiency of N-linked glycosylation of bovine DNase I depends on the Asn-Xaa-Ser/Thr sequence and the tissue of origin. Biochem. J., 355, 245–248.[CrossRef][Web of Science][Medline]
Panneerselvam, K. and Freeze, H.H. (1996) Mannose corrects altered N-glycosylation in carbohydrate-deficient glycoprotein syndrome fibroblasts. J. Clin. Invest., 97, 1478–1487.[Web of Science][Medline]
Rudolf, M.T., Traynor-Kaplan, A.E., and Schultz, C. (1998) A membrane-permeant, bioactivatable derivative of Ins(1,3,4)P3 and its effect on Cl(-)-secretion from T84 cells. Bioorg. Med. Chem. Lett., 8, 1857–1860.[CrossRef][Medline]
Rutschow, S., Thiem, J., Kranz, C., and Marquardt, T. (2002) Membrane-permeant derivatives of mannose-1-phosphate. Bioorg. Med. Chem., 10, 4043–4049.[Medline]
Sarkar, A.K., Brown, J.R., and Esko, J.D. (2000) Synthesis and glycan priming activity of acetylated disaccharides. Carbohydr. Res., 329, 287–300.[CrossRef][Web of Science][Medline]
Schultz, C., Vajanaphanich, M., Harootunian, A.T., Sammak, P.J., Barrett, K.E., and Tsien, R.Y. (1993) Acetoxymethyl esters of phosphates, enhancement of the permeability and potency of cAMP. J. Biol. Chem., 268, 6316–6322.
Van Schaftingen, E. and Jaeken, J. (1995) Phosphomannomutase deficiency is a cause of carbohydrate-deficient glycoprotein syndrome type I. FEBS Lett., 377, 318–320.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||