Glycobiology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (16) Request Permissions Disclaimer Google Scholar Articles by Gao, N. Articles by Lehrman, M. A. Search for Related Content PubMed PubMed Citation Articles by Gao, N. Articles by Lehrman, M. A. Social Bookmarking          What's this?

Glycobiology, 2002, Vol. 12, No. 5 353-360
© 2002 Oxford University Press

Analyses of dolichol pyrophosphate–linked oligosaccharides in cell cultures and tissues by fluorophore-assisted carbohydrate electrophoresis

Ningguo Gao and Mark A. Lehrman¹

Department of Pharmacology, University of Texas–Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-9041, USA

Received on January 1, 2002; revised on February 7, 2002; accepted on February 24, 2002.

	Abstract

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
Lipid-linked oligosaccharides (LLOs) are the precursors of asparagine (N)-linked glycans, which are essential information carriers in many biological systems, and defects in LLO synthesis cause Type I congenital disorders of glycosylation. Due to the low abundance of LLOs and the limitations of the chemical and physical methods previously used to detect them, simple and sensitive nonradioactive methods for LLO analysis are lacking. Thus, almost all studies of LLO synthesis have relied on metabolic labeling of the oligosaccharides with radioactive sugar precursors. We report that LLOs in cell cultures and tissues can be easily detected and quantified with a sensitivity of 1–2 pmol by fluorophore-assisted carbohydrate electrophoresis (FACE). These analyses required efficient removal of contaminants, most likely trace quantities of glycogen breakdown products, that interfered with FACE. Studies with CHO-K1 cells showed that LLOs detected by FACE and by metabolic labeling had similar turnover rates. Glc₃Man₉GlcNAc₂-P-P-dolichol was the most prominent LLO detected by FACE in normal cultured cells and mouse tissues. However, the relative amounts of Glc_0-2Man_5–9GlcNAc₂-P-P-dolichol intermediates in tissues, such as liver and kidney, were unexpectedly greater than for cultured cells. IV injection of D-mannose, raising the circulatory concentration by three- to fourfold, did not affect LLO composition. Thus, the relative accumulation of LLO intermediates in mouse liver and kidney is not likely due to inadequate D-mannose in the circulation. In summary, FACE is a facile, accurate, and sensitive method for LLO analysis, permitting investigations not feasible by metabolic labeling.

Key words: dolichol/glycosylation/fluorophore-assisted carbohydrate electrophoresis/lipid-linked oligosaccharide

	Introduction

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
Asparagine (N)-linked glycoproteins are formed in the lumenal space of the endoplasmic reticulum (ER) (Kornfeld and Kornfeld, 1985; Varki et al., 1999). This occurs by transfer of preformed oligosaccharide units from lipid-linked oligosaccharide (LLO) donors to nascent polypeptides with asparaginyl residues in the context Asn-X-Ser/Thr. Completed LLOs have the structure Glc₃Man₉GlcNAc₂-P-P-dolichol, the preferred substrate for oligosaccharyl transferase compared with premature LLO intermediates (Karaoglu et al., 2001). LLO synthesis requires a series of highly conserved glycosyltransferase reactions. The first seven sugar transfer reactions take place on the cytoplasmic leaflet of the ER membrane, require two equivalents of UDP-GlcNAc and five of GDP-mannose, and generate the lipid-linked intermediate Man₅GlcNAc₂-P-P-dolichol. This intermediate then flips to the lumenal leaflet, where it is the acceptor substrate for seven additional glycosyltransferase reactions using four equivalents of mannose-P-dolichol and three of glucose-P-dolichol as donors.

It has recently been shown that the synthesis of Glc₃Man₉GlcNAc₂-P-P-dolichol is defective in the Type I congenital disorders of glycosylation (CDGs), a family of at least six distinct genetic diseases (Marquardt and Freeze, 2001; Schachter, 2001). After transfer of the oligosaccharide unit from Glc₃Man₉GlcNAc₂-P-P-dolichol to protein, the resultant N-linked oligosaccharide is processed in the ER by a series of glycosidases, in most cases ending in the structure Man₈GlcNAc₂. After exiting the ER, N-linked glycoproteins undergo additional processing reactions involving glycosidases and glycosyltransferases in the Golgi complex and trans-Golgi network. These N-linked oligosaccharides have specific roles in the folding, processing, and export of glycoproteins from the ER, as well as for functions that occur after exiting the secretory pathway (Lehrman, 2001).

A thorough knowledge of LLO assembly is therefore essential to understand both ER function and the pathophysiology of CDG Type I. Most procedures for LLO analysis require hydrolysis of the pyrophosphate bond, allowing the free water-soluble oligosaccharides to be fractionated and characterized, though recently an efficient fractionation of intact LLOs has been reported (Kelleher et al., 2001). The abundance of LLOs is low, typically on the order of 1 nmol/g tissue (Badet and Jeanloz, 1988; Gibbs and Coward, 1999), and detection of the oligosaccharides by physical and/or chemical means has been inefficient. For these reasons, it is common to study LLOs that have been made radioactive by incubations of cells, organelle preparations, or tissue slices with appropriate [³H]- or [¹⁴C]-labeled precursors. The incorporated isotopes then permit facile detection of the oligosaccharides.

Although isotopic approaches have been extremely useful for LLO analysis, they have a number of limitations.

1. The results obtained by metabolic labeling, which is typically done for a brief incubation period, may not reflect the true steady-state LLO compositions.

2. Due to isotope dilution, it is difficult to determine the actual molar quantity of each LLO species from the amount of radioactivity incorporated. For example, accurate measurements of the effects of mannose supplementation on LLO quantities in cultured CDG-Ia cells required analytical determination of the specific activity of GDP-[³H]mannose generated from exogenous [³H]mannose (Rush and Waechter, 1995; Rush et al., 2000). Furthermore, radioactivity per oligosaccharide must be normalized to the number of residues per oligosaccharide, and intermediates with few sugars may be difficult to detect. Due to variations in the fluxes of isotopic precursors, it cannot be assumed that the residues at each position of the oligosaccharide will be labeled with the same specific activity.

3. Isotopic labeling in culture medium is often inefficient unless the glucose concentration of the medium is lowered 5- to 10-fold below the physiological range, subjecting the cultures to potential glucose deprivation effects (Chapman and Calhoun, 1988).

4. Pool dilution and catabolism in living animals would require the use of very large quantities of radioactive compounds. Hence, few direct measurements of tissue LLO compositions have been reported.

LLO compositions from bovine pancreas (Badet and Jeanloz, 1988; Gibbs and Coward, 1999) have been studied by direct nonisotopic analysis of the oligosaccharides, though these approaches were laborious, required large amounts of tissue, and might not be practical with small experimental animals, such as the mouse or rat. An innovative and sensitive method for inferring LLO compositions from bovine and porcine pancreas has recently been reported (Kelleher et al., 2001), although this method involves the analysis of N-linked glycopeptides from oligosaccharyltransferase reactions rather than the free oligosaccharides themselves.

This study describes the use of fluorophore-assisted carbohydrate electrophoresis (FACE) (Jackson, 1996; Starr et al., 1996) to circumvent these problems. With the use of a commercial fluorescence scanner, multiple samples are easily processed. 1–2 pmol of oligosaccharide released from each LLO species is detected, and detection is independent of the number of residues per oligosaccharide. The FACE technique was used to characterize the LLO contents in single mouse tissues and permitted the dependence of LLO synthesis in liver and kidney on serum mannose concentrations to be assessed.

	Results

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
Detection of LLOs in parental and glycosylation-defective Chinese hamster ovary (CHO) K1 cells by FACE
FACE is a simple method for the detection, measurement, and structural characterization of monosaccharides and oligosaccharides. FACE was applied to dolichol pyrophosphate–linked oligosaccharides as described under Materials and methods. Cells and tissues were rapidly disrupted in methanol, and the resulting mixtures were dried under nitrogen. The dry material was extracted with chloroform:methanol 2:1 (CM) and water and then chloroform:methanol:water 10:10:3 (CMW) to recover LLOs. After cleavage of the pyrophosphate linkage with mild acid, the released oligosaccharides were reacted with 8-aminonaphthalene-1,3,6-trisulfonate (ANTS), resolved by gel electophoresis, and visualized with a commercial fluorescence scanner.

Preliminary analyses of such material from both cells and tissues revealed groups of contaminating oligosaccharides that gave ladder-like patterns differing by single sugar residues on FACE gels and strongly interfered with detection of authentic LLOs (data not shown). This material was routinely removed by adsorbing LLOs to diethylaminoethyl (DEAE)-cellulose (Materials and methods) and has been tentatively identified as small fragments of glycogen because they were digested by amyloglucosidase. FACE monosaccharide gels revealed the presence of glucose in the fragments, and pure mussel glycogen yielded similar fragments when subjected to the LLO extraction protocol (data not shown). Apparently, the contaminants originated from small amounts of cellular glycogen that entered the LLO-containing CMW phase and were then partially cleaved by the mild acid treatment used to liberate soluble oligosaccharides from dolichol-P-P. Though present in trace quantities, they were easily detected because of the high sensitivity of the FACE technique.

Previous metabolic labeling studies showed that the LLO pool in parental CHO-K1 cells consists mainly of Glc₃Man₉GlcNAc₂-P-P-dolichol, whereas in the mutant lines Lec15.2 and Lec35.1 the primary LLO is Man₅GlcNAc₂-P-P-dolichol (Camp et al., 1993; Anand et al., 2001). LLOs from these cells were thus used for an initial assessment of FACE (Figure 1). CHO-K1 samples had one abundant ANTS-labeled oligosaccharide, most likely Glc₃Man₉GlcNAc₂-ANTS (structural analyses presented later), which on average was 70% of the total LLO pool. However, lesser amounts of intermediates likely to be Man₅GlcNAc₂-ANTS were also detected. Glc₃Man₉GlcNAc₂-ANTS was the primary oligosaccharide detected by FACE in primary human dermal fibroblasts (Figure 1) and the human cell lines THP-1, HeLa, and C1R (data not shown), all of which are expected to synthesize LLO normally. In contrast, the major ANTS-labeled oligosaccharide in Lec15.2 and Lec35.1 (Figure 1), as well as a Lec1/Lec35 double mutant (not shown), was Man₅GlcNAc₂-ANTS. In the Lec15 sample but not the Lec35 sample, a second, slower-migrating oligosaccharide (asterisk, lane 6) was also detected. This was most likely Glc₃Man₅GlcNAc₂-ANTS, which metabolic labeling studies predict should be about 20% as abundant as Man₅GlcNAc₂-ANTS in Lec15.2 cells but negligible in Lec35.1 (Anand et al., 2001). As expected, failure to release oligosaccharides from dolichol-P-P by omitting treatment with weak acid eliminated their appearance on FACE gels. There were no effects with ceramidase digestions or ammonium hydroxide treatments designed to remove fatty acyl groups (not shown). The CM extraction step was necessary to remove interfering contaminants, but it also removed early LLO intermediates (up to Man₃GlcNAc₂-P-P-dolichol) with varying efficiency. Therefore, most of the work presented will be limited to LLOs with five or more mannose residues.

View larger version (26K): [in this window] [in a new window]   Fig. 1. FACE analyses of LLOs from cultured cells. LLOs were extracted and processed for FACE analysis as described under Materials and methods from approximately 5 x 107 cells of the following cultures: parental CHO-K1 cells (lane 1), CHO-K1 mutants Lec35.1 (lane 5) and Lec15.2 (lane 6), and two adult dermal fibroblast cultures from the American Tissue and Culture Collection (lanes 2 and 3). ANTS-labeled oligomers of glucose (maltooligosaccharide mixture, Pfahnstiehl item M-138), enriched in species with four to seven glucose units, were used as standards in this (lanes 4 and 7) and all other FACE analyses. In this and subsequent figures, the positions of Glc3Man9GlcNAc2-ANTS and Man5GlcNAc2-ANTS are indicated as G3M9 and M5, respectively. The asterisk denotes a species in lane 6 (Lec15) that may be Glc3Man5GlcNAc2-ANTS.

 
Characterization of ANTS-labeled oligosaccharides
Several lines of evidence showed that the major ANTS-labeled oligosaccharide identified in the CHO-K1 sample was indeed Glc₃Man₉GlcNAc₂-ANTS. As will be shown, mixtures of LLO intermediates from liver and other tissues provided convenient regularly spaced reference markers for determining the number of sugar residues removed by each treatment. If CHO-K1 cells were incubated with [³H]mannose prior to FACE, the major ANTS-labeled oligosaccharide was also the major radioactive oligosaccharide on FACE gels (data not shown). Endoglycosidase H (endo H) degraded all of the Glc₃Man₉GlcNAc₂-ANTS (Figure 2, lane 5), yielding ANTS-labeled material that migrated with the dye front (not shown) consistent with formation of GlcNAc-ANTS. Golgi endomannosidase (Spiro and Spiro, 2000) converted Glc₃Man₉GlcNAc₂-ANTS to a product that migrated as Man₈GlcNAc₂-ANTS (lane 3; see also the following). Glc₃Man₉GlcNAc₂-ANTS was digested with jackbean -mannosidase (lane 4), though the shift in mobility suggested a loss of only three mannose residues, consistent with the structure of Glc₃Man₆GlcNAc₂-ANTS in which two exposed -linked mannosyl residues remained. As indicated by earlier studies (Beeley, 1985; Cacan et al., 2001), Glc₃Man₄GlcNAc₂-ANTS (the theoretical limit digestion product) apparently did not form because cleavage by jackbean -mannosidase is inefficient when there is an exposed 1,3-linked mannose attached to an 1,6-linked mannose on glucosylated structures, such as Glc₃Man₆GlcNAc₂-ANTS.

View larger version (25K): [in this window] [in a new window]   Fig. 2. Enzymatic characterization of ANTS-labeled oligosaccharides from cell cultures. Substrates were either Glc3Man9GlcNAc2-ANTS prepared from CHO-K1 LLO and purified from a preparative FACE gel (lanes 2–5), similarly purified Man5GlcNAc2-ANTS from Lec35.1 LLO (lanes 6 and 7), or a mixture of ANTS-labeled oligosaccharides from mouse liver and Lec35.1 cells (lanes 8 and 9). Substrates were left untreated or were digested with either Golgi endomannosidase, jackbean -mannosidase, or endo H. Note that the product of endo H digestion of Glc3Man9GlcNAc2-ANTS, GlcNAc-ANTS, would have been expected to run off the bottom of the gel (lane 5), and the product of -mannosidase digestion of Man5GlcNAc2-ANTS, Man1GlcNAc2-ANTS, ran near the bottom of the gel (lane 7).

 
The major ANTS-labeled oligosaccharide from Lec35.1 cells behaved like Man₅GlcNAc₂-ANTS, exhibiting resistance to Golgi endomannosidase (not shown) while being converted to a faster migrating species after treatment with jackbean -mannosidase (Figure 2, lane 7). By comparison with a series of markers prepared by mixing partial mannosidase digestions (not shown), the limit -mannosidase product was identified as Man₁GlcNAc₂-ANTS. Man₅GlcNAc₂-ANTS was resistant to a 24-h treatment with endo H (lane 9), as anticipated (Lehrman and Zeng, 1989). To demonstrate that the endo H was active and exhibited the proper specificity, the Man₅GlcNAc₂-ANTS in this experiment was mixed with a sample from mouse liver containing the endo H substrates Glc₃Man_6–9GlcNAc₂-ANTS (lane 8).

LLOs detected in CHO-K1 cells with FACE are not derived from an inactive pool
LLOs detected by labeling with radioactive sugar precursors are rapidly diminished on addition of an excess of the appropriate unlabeled precursor. For example, pulse-labeled [³H]-LLOs from CHO-K1 cells incubated with [³H]mannose in the presence of 0.5 mM glucose were almost completely eliminated by a 20-min chase in medium with 5 mM glucose (not shown), in agreement with previous determinations (Hubbard and Robbins, 1980). It was necessary to demonstrate that the turnover rate of the LLO pool detected by FACE was similar to the rate for LLOs detected by metabolic labeling to rule out the possibility that the LLOs detected by FACE were derived from an accumulated pool of storage products such as found with lipofuscinosis (Hall et al., 1989; Faust et al., 1994). For this purpose, cells were incubated for increasing periods of time with tunicamycin (TN), which inhibits the GlcNAc-1-P transferase that initiates LLO synthesis. The cells were then labeled for 20 min with [³H]mannose, and the Glc₃Man₉GlcNAc₂-P-P-dolichol remaining was measured by incorporation of tritium and by FACE. As shown in Figure 3, LLOs detected by either method had similar turnover rates. In each case most of the Glc₃Man₉GlcNAc₂-P-P-dolichol was eliminated by 1 h of TN treatment, and none was detected after 2 h. These times were somewhat longer than measured by pulse chase as discussed, most likely due to the extra time required for TN to enter the cells and form complexes with the GlcNAc-1-P transferase.

View larger version (22K): [in this window] [in a new window]   Fig. 3. Metabolic turnover of LLOs detected by FACE in CHO-K1 cells. Dishes (100 mM) of CHO-K1 cells were incubated for the times indicated at 37°C in complete F-12 medium containing 5 mM glucose, supplemented with 20 µg/ml TN. Cells were then suspended in 0.1 ml F-12 medium with 0.5 mCi/ml [3H]mannose and 0.5 mM glucose and incubated for 20 min at 37°C in the continued presence of 20 µg/ml TN. CMW 10:10:3 extracts containing LLOs were prepared. One portion was used for Glc3Man9GlcNAc2-P-P-dolichol measurement by high-performance liquid chromatography ([3H]Glc3Man9GlcNAc2; closed circles), the other by FACE (Glc3Man9GlcNAc2-ANTS; open circles).

 
LLOs detected in rodent tissues
An important advantage of FACE is the potential to conveniently study LLOs extracted directly from small amounts of animal tissues. In Figure 4 (panels A–C) LLOs detected by FACE from eight single mouse tissues are displayed. FACE also detected LLOs in various rat tissues, including spleen and lymph node (not shown). As anticipated, Glc₃Man₉GlcNAc₂-P-P-dolichol was the major LLO in each tissue. However, it is also apparent that LLO intermediates account for a significant fraction of the total LLO pool in each mouse tissue (Figure 4, panels D and E), in some cases accounting for the majority of the pool. For most tissues, no appreciable variability was noticed from animal to animal. However, in the case of kidney, there was extensive fluctuation in the amounts of Glc_0-2Man₉GlcNAc₂-ANTS relative to Glc₃Man₉GlcNAc₂-ANTS. The two kidney samples shown in panel C are typical of this variation; the samples presented in panels D and E are from a kidney with relatively high Glc_0–2Man₉GlcNAc₂-ANTS. Although no factor that correlated with either pattern was identified, it was noted that analyses of individual kidneys from a single animal resulted in highly similar profiles, and animals received and processed together also had similar profiles (not shown). In the case of brain, prominent ANTS-positive materials that did not comigrate with any LLO-derived species were consistently noted (asterisks, panel B); their origins were not determined. The total amounts of LLO detected in different tissues varied about 10-fold, with kidney being most abundant (panel D). The values for tissues with the lowest amounts of LLO, 0.5 nmol/g, were similar to those reported previously (Badet and Jeanloz, 1988; Gibbs and Coward, 1999).

View larger version (51K): [in this window] [in a new window]   Fig. 4. Compositions and quantities of LLOs detected in mouse tissues by FACE. (A–C) LLOs from these amounts of various mouse tissues were analyzed by FACE. Unless indicated, the entire yield from a single tissue is shown. Brain, 0.350 g; heart, 0.201 g; kidney (one-fifth), 0.130 g; leg muscle, 0.302 g; liver (one-tenth), 0.220 g; lung, 0.271 g; spleen, 0.097 g; testis, 0.302 g. One profiling gel was used for panels A and B, and two were used for panel C. For brain, the asterisk denotes products that did not comigrate with other LLO-derived ANTS-oligosaccharides and whose origins were not determined. A kidney sample with relatively low Glc0–2Man9GlcNAc2-ANTS is shown in panel C, and kidney samples with high amounts of these species are shown in panels A and C. (D) The yields of ANTS-labeled LLO-derived oligosaccharides (nmol per g wet weight) are given for several mouse tissues. The amounts of total Glc0–3Man5–9GlcNAc2-ANTS (closed bars) and Glc3Man9GlcNAc2-ANTS (open bars) are shown. (E) Percentage compositions for Glc0–3Man5–9GlcNAc2-ANTS in liver (closed bars) and kidney (open bars) are shown; the kidney sample was one with a relatively high proportion of Glc0–2Man9GlcNAc2-ANTS. Results in panels D and E are the averages of two mice.

 
Structural characterization of the major ANTS-labeled oligosaccharide from mouse liver
The major ANTS-oligosaccharide from liver LLO comigrated with Glc₃Man₉GlcNAc₂-ANTS from CHO-K1 cells (not shown) and had similar susceptibilities to digestions with Golgi endomannosidase, jackbean -mannosidase, and endo H (Figure 5). In complete agreement with its known specificity (Spiro and Spiro, 2000), the endomannosidase digested all glucosylated molecules, with increased amounts of Man₈GlcNAc₂-ANTS appearing on the gel. Some liver samples also contained ANTS-labeled material just above Man₆GlcNAc₂-P-P-ANTS (lane 4, black dot), that did not correspond to any obvious LLO-derived ANTS-labeled products but was digested with endomannosidase. Endo H efficiently digested Glc_0–3Man_6–9GlcNAc₂-ANTS. As discussed, jackbean -mannosidase appeared to generate a mixture of partially digested glucosylated oligosaccharides, plus Man₁GlcNAc₂-ANTS from nonglucosylated oligosaccharides.

View larger version (25K): [in this window] [in a new window]   Fig. 5. Enzymatic characterization of ANTS-labeled products from liver LLO. As shown in lane 4, ANTS-labeled products from mouse liver LLO (see panel C, Figure 4) were combined with Man5GlcNAc2-ANTS from Lec35.1 cells (see lane 5, Figure 1) as an internal standard. Using Glc3Man9GlcNAc2-ANTS and Man5GlcNAc2-ANTS as reference points, each of the intermediates in the series Glc0–2Man6–9GlcNAc2-ANTS can be detected in lane 4 as a series of regularly spaced products (except the species denoted by the black dot; see text). This mixture was treated with the following enzymes. Lane 1, -mannosidase. -mannosidase digestion generated Man1GlcNAc2-ANTS (asterisk), the expected product of Man5GlcNAc2-ANTS. In addition, Glc3Man9GlcNAc2-ANTS was converted to ANTS-labeled species migrating as Glc3Man6–7GlcNAc2-ANTS. As described in the text, Glc3Man6GlcNAc2-ANTS is the expected product of Glc3Man9GlcNAc2-ANTS, and Glc3Man7GlcNAc2-ANTS would therefore result from incomplete digestion. Lane 2, endo H. All species with six or more mannose residues were digested. The apparent partial loss of Man5GlcNAc2-ANTS with this treatment (24 h) was repeatable. Reducing the digestion time to 1 h resulted in a more selective reaction in which no loss of Man5GlcNAc2-ANTS was detected, but all ANTS-labeled species with six or more mannose residues were still completely digested (not shown). Lane 3, Golgi endomannosidase. Glc1–3Man9GlcNAc2-ANTS were all converted to Man8GlcNAc2-ANTS, consistent with the known specificity of this enzyme. An unknown ANTS-labeled product (denoted by the black dot adjacent to lane 4, migrating just behind Man6GlcNAc2-ANTS) detected in some samples from liver was degraded by endomannosidase (lane 3), suggesting the presence of at least one glucose residue. Lane 4, no enzyme. Lane 5 contains maltooligosaccharide standards.

 
Liver LLOs detected with FACE are not highly degraded
Previous studies with tissue LLOs have raised the possibility of degradation during isolation (Kelleher et al., 2001). To determine whether this was a significant factor in the studies shown here, several controls were done for the analysis of mouse liver. Liver was chosen because it was an abundant source of LLOs and endocytic studies have shown it to be a tissue rich in lysosomal hydrolases, a potential cause of degradation. As shown in Figure 6, LLO samples from liver left at room temperature for 20 min before freezing in N₂(l) (lane 1) appeared only slightly degraded compared to tissue snap-frozen immediately after harvest (lane 2). This result demonstrated that the speed of initial tissue processing is not a crucial factor. Interestingly, a faint unknown ANTS-labeled product migrating between the positions expected for Man₄GlcNAc₂-ANTS and Man₅GlcNAc₂-ANTS (asterisk, lane 2) appeared to be highly degraded by the 20-min delay. Whether methanol homogenates of snap-frozen tissues were dried under N₂(g) immediately (lane 6), or left at room temperature for 24 h prior to drying (lane 5), similar results were obtained. Mixture of snap-frozen liver with Lec35 cell pellets (in which Man₅GlcNAc₂-P-P-dolichol is the major LLO) prior to disruption with a polytron (lane 4) gave the same pattern as a mixture of the corresponding CMW extracts prepared separately (lane 3). This demonstrates that the LLOs examined by this procedure are unlikely to be affected by -mannosidase activities. Taken together, there was no evidence of serious artifactual LLO degradation in the procedure reported here. Hence the LLO intermediates identified in liver and presumably other tissues are authentic.

View larger version (41K): [in this window] [in a new window]   Fig. 6. Mouse liver LLOs detected by FACE are not highly degraded. Lanes 1–2 and 3–6 were taken from two different gels. One mouse liver was cut into two equivalent portions and used for the two experiments in lanes 1–2. Another mouse liver was cut into four equivalent portions and used for the four experiments in lanes 3–6, except that 5 x 107 Lec35.1 cells were added for lanes 3 and 4. Samples were processed as described under Materials and methods, and half of each was loaded onto a profiling gel with the following exceptions. The liver was kept at room temperature for 20 min before submersing in liquid nitrogen (lane 1) or was snap-frozen in liquid nitrogen immediately after removal from the animal (lane 2). CMW 10:10:3 extracts of liver and Lec35.1 cells were prepared separately, then mixed prior to weak acid treatment (lane 3), or the liver portion and the Lec35 cells were combined during freezing in liquid nitrogen (lane 4). Methanol homogenates of snap-frozen liver were kept for 24 h at room temperature before drying (lane 5), or dried without delay (lane 6). Of these variables, only a 20-min delay in freezing in liquid nitrogen appeared to have any appreciable effect (lane 1 band intensities are less than for lane 2). However, comparisons of the relative intensities of the LLO-derived ANTS-oligosaccharides indicated that the distributions were similar (not shown). Therefore, the difference may be due to the inexactness of preparing two truly equal portions of the same liver. One species of unknown identity (asterisk, lane 2) was nearly completely eliminated by the 20-min delay.

 
Efficiency of LLO synthesis in mouse liver
Relative to the amounts of Glc₃Man₉GlcNAc₂-P-P-dolichol, cultured cells (Figure 7, panels A and B) appeared to have lower proportions of incompletely mannosylated LLO intermediates such as Man₆-₈GlcNAc₂-P-P-dolichol (triangles) than did mouse liver (panel C). As discussed earlier, kidney LLOs (panel D) also were atypical in that Glc_0–2Man₉GlcNAc₂-P-P-dolichol was variable and usually more enriched than in other tissues. Because defects that inhibit the production of mannosyl donors can cause undermannosylated LLO intermediates to accumulate, but supplementation of cell cultures with mannose can overcome this (Panneerselvam and Freeze, 1996), one possibility for the liver result was a limitation in the supply of mannose available in the circulation. As shown in Figure 7, panels G and H, mice were injected with vehicle (control), or vehicle containing D-mannose. After 20 min serum hexoses were measured and liver and kidney LLOs were analyzed by FACE. The 20-min time point was chosen because several studies, including those reported here, have indicated that 20 min is sufficient for essentially complete turnover of LLO pools. Though glucose concentrations were unchanged (average of 11 mM), circulatory mannose concentrations increased three- to fourfold by injection of D-mannose, reaching 600–1000 µM by 20 min. However, there were no apparent changes in the relative amounts of incompletely mannosylated LLO intermediates from liver (panel E). Thus, their relative accumulation does not appear to be due to lack of circulatory D-mannose. As expected, the glucosylation of kidney LLO (panel F) was unaffected by injection with D-mannose. As discussed, the two kidney samples (panels D and F) could be compared directly because the control and injected animals were received and processed together.

View larger version (45K): [in this window] [in a new window]   Fig. 7. The accumulation of LLO intermediates in tissues is not due to inadequate circulatory D-mannose. (A–F) Graphical Glc0–3Man5–9GlcNAc2-ANTS profiles are shown for cell cultures CHO-K1 (A) and CRL-1892 (B), liver (C), and kidney (D) from a control animal, and liver (E) and kidney (F) from a mannose-injected animal. To facilitate direct comparisons, the y-axes were adjusted so that the Glc3Man9GlcNAc2-ANTS peaks from the different samples were all the same height. The kidney samples in these experiments had relatively low amounts of Glc0–2Man9GlcNAc2-ANTS. The absence of an effect of mannose injection on liver and kidney LLOs was consistently observed (total of six mice). LLO intermediates with six to eight mannosyl residues are indicated by the open triangle. (G–H) Monosaccharide compositions were determined with sera from mice injected with vehicle (controls, samples 1 and 2) or with D-mannose (IV mannose, samples 3–6). Serum D-mannose concentrations (G, µM) are presented directly above an image of the monosaccharide profiling gel used (H). Hexose standards are shown; D-glucose was the most abundant serum sugar detected. Animals 1 (control) and 5 (IV mannose) were used for the LLO analyses in A–F, but the livers and kidneys in all mice were analyzed with consistent results.

 

	Discussion

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
LLOs in cell cultures and tissues can be analyzed simply and quantitatively with commercially available reagents by FACE, solving many prior limitations associated with the use of radiolabeled metabolic precursors. In our hands, 1–2 pmol of LLO was easily detected with a commercial fluorescence scanner. Even with a standard UV illumination box of the type used for detecting ethidium bromide–stained DNA in agarose gels, 10–20 pmol of LLO can be detected (data not shown). LLO-derived ANTS-labeled intermediates differing by a single residue, from Man₁GlcNAc₂ to Glc₃Man₉GlcNAc₂, were well resolved. In all cases, various genetic and enzymatic manipulations resulted in the expected ANTS-labeled products.

The elimination of radioisotopes for LLO analysis is important because it permits a true assessment of the steady-state pool of LLOs. FACE eliminates complications associated with factors such as the use of special "low sugar" labeling media and the lack of equilibration between exogenous radiolabeled and endogenous unlabeled precursors. Another major advantage is the simplicity by which LLOs can be studied from small amounts of snap-frozen tissues, such as from a single mouse (Figure 4). There was no serious artifactual degradation of LLOs detected by FACE, and attempts to deliberately allow such degradation to occur had little effect. Thus, FACE will be highly useful for analyses of animal models of human glycosylation diseases affecting LLOs, such as CDG Type I.

The FACE analyses of mouse tissues were somewhat surprising. LLO compositions of normal cultured cells, such as CHO-K1, determined by metabolic labeling have revealed a predominance of the mature LLO Glc₃Man₉GlcNAc₂-P-P-dolichol. To lesser extents, the intermediate Man₅GlcNAc₂-P-P-dolichol is also usually detectable. This was consistent with FACE results (Figure 7, panels A and B). However, FACE analyses of liver showed that Man_6–8GlcNAc₂-P-P-dolichol represented a greater proportion of the LLO pool (Figure 7, panel E) than for cultured cells. Elevation of the circulatory mannose concentration had no effect on the liver profile, indicating that metabolic precursors needed for mannosylation were probably not in limited supply. Kidney also gave unexpected results. The efficiency of glucosylation of Man₉GlcNAc₂-P-P-dolichol in kidney was variable, and in general kidney had a greater proportion of Glc_0–2Man₉GlcNAc₂-P-P-dolichol than other tissues. Clearly, the causes of these differences remain to be determined. Specifically, it will be interesting to examine whether LLOs in primary cell cultures are like LLOs in the original tissues or like LLOs in permanent cell lines.

	Materials and methods

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
Cell culture
Culture media and supplies were from Life Technologies/Gibco, except bovine sera (Atlanta Biologicals). Parental CHO-K1 cells, Lec15.2 mutant cells, Lec35.1 mutant cells (Camp et al., 1993), PIR-P3 cells containing both Lec1 and Lec35 defects (Zeng and Lehrman, 1991) (all cultured with Ham’s F-12 medium supplemented with 2% fetal bovine serum and 8% calf serum; Lehrman and Zeng, 1989), and human adult dermal fibroblasts (ATCC CRL-1892 and CRL-1987; cultured with RPMI 1640 medium supplemented with 10% fetal bovine serum) were grown in a humidified atmosphere with 5% carbon dioxide at 37°C degrees until 80–90% of confluence. In some experiments cells were incubated in medium with 0.5 mM glucose (i.e., 10% of glucose in complete medium) in the absence or presence of 50 µCi/ml [2-³H] mannose (23 Ci/mmol; Amersham) for 20 min at 37°C and either harvested immediately or subjected to a "chase" with complete medium containing 5 mM glucose. In all cases, cell monolayers were rinsed twice with ice-cold phosphate buffered saline (PBS), then scraped and sonicated with methanol to initiate the LLO extraction procedure.

Analysis of LLOs from cell cultures
The sonicated methanol suspensions were dried under N₂ gas and sequentially extracted with 2:1 CM, pure water, and 10:10:3 CMW (Lehrman and Zeng, 1989). For high-performance liquid chromatography analysis of [³H]mannose-labeled LLOs, the material in the CMW extract was treated with weak acid to generate soluble oligosaccharides and then reduced with NaBH₄ (Zeng and Lehrman, 1991). For FACE analysis of unlabeled LLOs, the CMW extract was loaded onto a column containing 1 ml DEAE-cellulose (Whatman) in the acetate form equilibrated with CMW. The column was washed with 10 ml CMW, then 10 ml 3 mM NH₄OAc in CMW to remove contaminating glycogen fragments (see Results).

LLOs were recovered by elution with 10 ml 300 mM NH₄OAc in CMW. To remove NH₄OAc, this fraction was supplemented with 4.4 ml chloroform and then 1.4 ml water to generate a two-phase system. After mixing and centrifugation, the lower phase containing LLOs was recovered and dried under nitrogen gas. The residue was mixed with 2 ml 0.1 N HCl in 50% isopropyl alcohol and reacted at 50°C for 60 min, releasing the pyrophosphate-linked oligosaccharides. Residual NH₄OAc was removed by repeated freeze-drying with water or by desalting with Dowex (Biorad) resins AG1-8X (hydrogen form) and AG50-8X (formate form). Using an oligosaccharide profiling kit (Glyko), oligosaccharides were labeled with 0.15 M ANTS and 1 M NaBCNH₃, and the resolved on an oligosaccharide profiling gel (Glyko) as described by the manufacturer. The gel was imaged with a Biorad Fluor-S MultiImager using a 530DF60 filter. Electronic gel images were generated, and individual ANTS-positive species were quantified, with Quantity One software supplied with the scanner. The sensitivity of LLO detection by FACE was determined with known quantities of various mono- and oligosaccharide standards subjected to the same ANTS labeling procedure described.

FACE analysis of LLOs from mouse tissues
Male adult ICR white outbred mice (typical weight, 20–25 g) from a local supplier were anesthetized with ether and then euthanized by IP injection of 0.2 ml 40 mg/ml pentobarbital (Sigma) according to institutionally approved animal handling procedures. The tissues were harvested immediately and frozen in liquid nitrogen. In some cases the frozen tissues were stored at –80°C up to 8 weeks. The frozen tissues were pulverized with a mortar and pestle chilled with liquid nitrogen, homogenized in ice-cold methanol with a polytron (Brinkmann Instrument model PT10/35), and dried under nitrogen gas. The remaining steps were the same as those described for cultured cells.

Determination of circulatory mannose and glucose concentrations after injection of mice with mannose
Mice were anesthetized with ether, and 0.2 ml of PBS with 0 or 100 mM D-mannose were injected into the tail vein (similar results were obtained when 0.15 M NaCl was used instead of PBS). Twenty minutes later, samples of blood were collected and the animals were euthanized. LLOs were harvested as described. The blood samples were allowed to clot at 4°C for 30 min, and serum was collected by centrifugation. Seven volumes of ice-cold 100% ethanol were added per three volumes of serum, and the mixtures were placed on ice for 15 min to precipitate protein. Soluble material was collected by centrifugation, dried under vacuum, dissolved in water, and deionized with Dowex AG50-8X (hydrogen) and AG1-8X (formate). Using a monosaccharide composition kit (Glyko), the samples were then labeled with 2-aminoacridone and resolved by gel electrophoresis as described by the manufacturer. 2-Aminoacridone-labeled monosaccharides were quantified as described for ANTS-labeled oligosaccharides.

Enzymatic digestion of ANTS-labeled oligosaccharides
Up to 1 nmol ANTS-labeled oligosaccharide was digested per reaction. In all cases controls were reactions without enzyme. Jackbean -mannosidase (Glyko): Oligosaccharides were dissolved in 2 µl 100 mM sodium acetate (pH 5.0), 0.1 U enzyme (1 µl) was added, and the reaction mixture was incubated at 37°C for 18–24 h. Endo H (New England Biolabs): Oligosaccharides were dissolved in 2 µl 50 mM sodium citrate (pH 5.0) containing 5% glycerol and 0.1% Triton X-100, 0.1 U enzyme (1 µl) was added, and the reaction mixture was incubated at 37°C for 1–24 h as indicated in the text. Golgi endomannosidase (recombinant; gift of Dr. Robert Spiro): Oligosaccharides were dissolved in 2 µl 100 mM Na–MES (pH 6.5) containing 0.2% Triton X-100, 1 µl (270 ng) endomannosidase was added, and the reaction mixture was incubated at 37°C for 18–24 h.

	Abbreviations

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
ANTS, 8-aminonaphthalene-1,3,6-trisulfonate; CDG, congenital disorder of glycosylation; CHO, Chinese hamster ovary; CM, chloroform:methanol; CMW chloroform:methanol:water; DEAE, diethylaminoethyl; ER, endoplasmic reticulum; FACE, fluorophore-assisted carbohydrate electrophoresis; endo H, endoglycosidase H; LLO, lipid-linked oligosaccharide; PBS, phosphate-buffered saline; TN, tunicamycin.

	Acknowledgments

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
Dr. Robert G. Spiro, Harvard Medical School, generously provided recombinant endomannosidase. This work was supported by NIH grant GM38545 and Welch Foundation grant I-1168.

	Footnotes

 
¹ To whom correspondence should be addressed

	References

Top Abstract Introduction Results Discussion Materials and methods Abbreviations Acknowledgments References

 
Anand, M., Rush, J.S., Ray, S., Doucey, M.A., Weik, J., Ware, F.E., Hofsteenge, J., Waechter, C.J., and Lehrman, M.A. (2001) Requirement of the Lec35 gene for all known classes of monosaccharide-P-dolichol-dependent glycosyltransferase reactions in mammals. Mol. Biol. Cell, 12, 487–501.[Abstract/Free Full Text]

Badet, J. and Jeanloz, R.W. (1988) Isolation and partial purification of lipid-linked oligosaccharides from calf pancreas. Carbohydr. Res., 178, 49–65.[CrossRef][ISI][Medline]

Beeley, J.G. (1985) Glycoprotein and proteoglycan techniques. New York: Elsevier.

Cacan, R., Duvet, S., Labiau, O., Verbert, A., and Krag, S.S. (2001) Monoglucosylated oligomannosides are released during the degradation process of newly synthesized glycoproteins. J. Biol. Chem., 276, 22307–22312.[Abstract/Free Full Text]

Camp, L.A., Chauhan, P., Farrar, J.D., and Lehrman, M.A. (1993) Defective mannosylation of glycosylphosphatidylinositol in Lec35 Chinese hamster ovary cells. J. Biol. Chem., 268, 6721–6728.[Abstract/Free Full Text]

Chapman, A.E. and Calhoun, J.C. (1988) Effects of glucose starvation and puromycin treatment on lipid-linked oligosaccharide precursors and biosynthetic enzymes in Chinese hamster ovary cells in vivo and in vitro. Arch. Biochem. Biophys., 260, 320–333.[CrossRef][Medline]

Faust, J.R., Rodman, J.S., Daniel, P.F., Dice, J.F., and Bronson, R.T. (1994) Two related proteolipids and dolichol-linked oligosaccharides accumulate in motor neuron degeneration mice (mnd/mnd), a model for neuronal ceroid lipofuscinosis. J. Biol. Chem., 269, 10150–10155.[Abstract/Free Full Text]

Gibbs, B.S. and Coward, J.K. (1999) Dolichylpyrophosphate oligosaccharides: large-scale isolation and evaluation as oligosaccharyltransferase substrates. Bioorg. Med. Chem., 7, 441–447.[CrossRef][Medline]

Hall, N.A., Jolly, R.D., Palmer, D.N., Lake, B.D., and Patrick, A.D. (1989) Analysis of dolichyl pyrophosphoryl oligosaccharides in purified storage cytosomes from ovine ceroid-lipofuscinosis. Biochim. Biophys. Acta, 993, 245–251.[Medline]

Hubbard, S.C. and Robbins, P.W. (1980) Synthesis of the N-linked oligosaccharides of glycoproteins. J. Biol. Chem., 255, 11782–11789.[Abstract/Free Full Text]

Jackson, P. (1996) The analysis of fluorophore-labeled carbohydrates by polyacrylamide gel electrophoresis. Mol. Biotechnol., 5, 101–123.[ISI][Medline]

Karaoglu, D., Kelleher, D.J., and Gilmore, R. (2001) Allosteric regulation provides a molecular mechanism for preferential utilization of the fully assembled dolichol-linked oligosaccharide by the yeast oligosaccharyltransferase. Biochemistry, 40, 12193–12206.[CrossRef][Medline]

Kelleher, D.J., Karaoglu, D., and Gilmore, R. (2001) Large-scale isolation of dolichol-linked oligosaccharides with homogeneous oligosaccharide structures: determination of steady-state dolichol-linked oligosaccharide compositions. Glycobiology, 11, 321–333.[Abstract/Free Full Text]

Kornfeld, R. and Kornfeld, S. (1985) Assembly of asparagine-linked oligosaccharides. Annu. Rev. Biochem., 54, 631–664.[CrossRef][ISI][Medline]

Lehrman, M.A. (2001) Oligosaccharide-based information in endoplasmic reticulum quality control and other biological systems. J. Biol. Chem., 276, 8623–8626.[Free Full Text]

Lehrman, M.A. and Zeng, Y. (1989) Pleiotropic resistance to glycoprotein processing inhibitors in Chinese hamster ovary cells. J. Biol. Chem., 264, 1584–1593.[Abstract/Free Full Text]

Marquardt, T. and Freeze, H. (2001) Congenital disorders of glycosylation: glycosylation defects in man and biological models for their study. Biol. Chem., 382, 161–177.[CrossRef][ISI][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.[ISI][Medline]

Rush, J.S. and Waechter, C.J. (1995) Method for the determination of cellular levels of guanosine-5'-diphosphate mannose based on a weak interaction with concanavalin A at low pH. Anal. Biochem., 224, 494–501.[CrossRef][ISI][Medline]

Rush, J.S., Panneerselvam, K., Waechter, C.J., and Freeze, H.H. (2000) Mannose supplementation corrects GDP-mannose deficiency in cultured fibroblasts from some patients with congenital disorders of glycosylation (CDG). Glycobiology, 10, 829–835.[Abstract/Free Full Text]

Schachter, H. (2001) The clinical relevance of glycobiology. J. Clin. Invest., 108, 1579–1582.[CrossRef][ISI][Medline]

Spiro, M.J. and Spiro, R.G. (2000) Use of recombinant endomannosidase for evaluation of the processing of N-linked oligosaccharides of glycoproteins and their oligosaccharide-lipid precursors. Glycobiology, 10, 521–529.[Abstract/Free Full Text]

Starr, C.M., Masada, R.I., Hague, C., Skop, E., and Klock, J.C. (1996) Fluorophore-assisted carbohydrate electrophoresis in the separation, analysis, and sequencing of carbohydrates. J. Chromatogr. A, 720, 295–321.

Varki, A., Cummings, R.D., Esko, J.D., Freeze, H.H., Hart, G.W., and Marth, J., eds. (1999) Essentials of glycobiology. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press.

Zeng, Y. and Lehrman, M.A. (1991) Isolation of Chinese hamster ovary cell lines producing Man₃GlcNAc₂ asparagine-linked glycans. Anal. Biochem., 193, 266–271.[CrossRef][ISI][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				N. Gao, J. Shang, and M. A Lehrman Unexpected Basis for Impaired Glc3Man9GlcNAc2-P-P-Dolichol Biosynthesis by Elevated Expression of GlcNAc-1-P Transferase Glycobiology, January 1, 2008; 18(1): 125 - 134. [Abstract] [Full Text] [PDF]

				M. A Lehrman Teaching dolichol-linked oligosaccharides more tricks with alternatives to metabolic radiolabeling Glycobiology, August 1, 2007; 17(8): 75R - 85R. [Abstract] [Full Text] [PDF]

				J. Shang, N. Gao, R. J. Kaufman, D. Ron, H. P. Harding, and M. A. Lehrman Translation attenuation by PERK balances ER glycoprotein synthesis with lipid-linked oligosaccharide flux J. Cell Biol., February 26, 2007; 176(5): 605 - 616. [Abstract] [Full Text] [PDF]

				S. K. Cho, N. Gao, D. A. Pearce, M. A. Lehrman, and S. L. Hofmann Characterization of lipid-linked oligosaccharide accumulation in mouse models of Batten disease Glycobiology, June 1, 2005; 15(6): 637 - 648. [Abstract] [Full Text] [PDF]

				N. Gao, J. Shang, and M. A. Lehrman Analysis of Glycosylation in CDG-Ia Fibroblasts by Fluorophore-assisted Carbohydrate Electrophoresis: IMPLICATIONS FOR EXTRACELLULAR GLUCOSE AND INTRACELLULAR MANNOSE 6-PHOSPHATE J. Biol. Chem., May 6, 2005; 280(18): 17901 - 17909. [Abstract] [Full Text] [PDF]

				J. Shang and M. A. Lehrman Metformin-stimulated Mannose Transport in Dermal Fibroblasts J. Biol. Chem., March 12, 2004; 279(11): 9703 - 9712. [Abstract] [Full Text] [PDF]

				C. E. Grubenmann, C. G. Frank, A. J. Hulsmeier, E. Schollen, G. Matthijs, E. Mayatepek, E. G. Berger, M. Aebi, and T. Hennet Deficiency of the first mannosylation step in the N-glycosylation pathway causes congenital disorder of glycosylation type Ik Hum. Mol. Genet., March 1, 2004; 13(5): 535 - 542. [Abstract] [Full Text] [PDF]

				N. Gao and M. A. Lehrman Alternative sources of reagents and supplies for fluorophore-assisted carbohydrate electrophoresis (FACE) Glycobiology, January 1, 2003; 13(1): 1G - 3G. [Full Text] [PDF]

				A. Gill, N. Gao, and M. A. Lehrman Rapid Activation of Glycogen Phosphorylase by the Endoplasmic Reticulum Unfolded Protein Response J. Biol. Chem., November 15, 2002; 277(47): 44747 - 44753. [Abstract] [Full Text] [PDF]

				N. Gao and M. A. Lehrman Coupling of the Dolichol-P-P-Oligosaccharide Pathway to Translation by Perturbation-sensitive Regulation of the Initiating Enzyme, GlcNAc-1-P Transferase J. Biol. Chem., October 11, 2002; 277(42): 39425 - 39435. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a