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

J. Biol. Chem., Vol. 275, Issue 31, 23456-23461, August 4, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/31/23456    most recent M002693200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fukuta, K. Articles by Makino, T. Search for Related Content PubMed PubMed Citation Articles by Fukuta, K. Articles by Makino, T. Social Bookmarking                      What's this?

Control of Bisecting GlcNAc Addition to N-Linked Sugar Chains^*

Kazuhiro Fukuta, Reiko Abe, Tomoko Yokomatsu, Fumio Omae, Mineko Asanagi, and Tadashi Makino

From the Life Science Laboratory, Mitsui Chemicals, Inc., 1144 Togo, Mobara, Chiba 297-0017, Japan

Received for publication, March 30, 2000, and in revised form, May 9, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present study, experimental control of the formation of bisecting GlcNAc was investigated, and the competition between -1,4-GalT (UDP-galactose:N-acetylglucosamine -1,4-galactosyltransferase) and GnT-III (UDP-N-acetylglucosamine:-D-mannoside -1,4-N-acetylglucosaminyltransferase) was examined. We isolated a -1,4-GalT-I single knockout human B cell clone producing monoclonal IgM and several transfectant clones that overexpressed -1,4-GalT-I or GnT-III. In the -1,4-GalT-I-single knockout cells, the extent of bisecting GlcNAc addition to the sugar chains of IgM was increased, where -1,4-GalT activity was reduced to about half that in the parental cells, and GnT-III activity was unaltered. In the -1,4-GalT-I transfectants, the extent of bisecting GlcNAc addition was reduced although GnT-III activity was not altered significantly. In the GnT-III transfectants, the extent of bisecting GlcNAc addition increased along with the increase in levels of GnT-III activity. The extent of bisecting GlcNAc addition to the sugar chains of IgM was significantly correlated with the level of intracellular -1,4-GalT activity relative to that of GnT-III. These results were interpreted as indicating that -1,4-GalT competes with GnT-III for substrate in the cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A -1,4-linked N-acetylglucosamine residue attached to the mannose--1,4- in the trimannosyl core of N-linked sugar chains has been described in complex-type and hybrid-type sugar chains of various glycoproteins such as IgA, IgG, IgM, etc. (1-3). This GlcNAc residue has been termed a "bisecting" GlcNAc and is formed by -D-mannoside -1,4-N-acetylglucosaminyltransferase (GnT-III).¹ GnT-III is presumed to be involved in pathological conditions, because an increase in its expression is accompanied by malignant transformation or oncofetal changes (4, 5). Although the function of the bisecting GlcNAc is not well understood, this modification can inhibit the action of some enzymes (-mannosidase-II, GnT-II, GnT-V, core -1,6-fucosyltransferase) in the subsequent biosynthesis of N-linked sugar chains, suggesting a regulatory role in the formation of complex-type and hybrid-type sugar chains (6). It is also known that the occurrence of bisecting GlcNAc-containing sugar chains on IgG increases with age (7).

Here, we studied a method to control the addition of bisecting GlcNAc to N-linked sugar chains. We used human monoclonal IgM as a model glycoprotein for control of the attachment of the bisecting GlcNAc residue. Human IgM is a glycoprotein containing 7-12% carbohydrate distributed at five N-glycosylation sites in the constant region of the heavy chain at positions Asn-171, Asn-332, Asn-395, Asn-402, and Asn-563 (8-12). The sugar chains at Asn-402 and Asn-563 are high mannose-type chains, and those at Asn-171, Asn-332, and Asn-395 are complex-type chains. Sugar chains of the human IgM produced by hybridoma cells were analyzed at each of the five glycosylation sites on the µ-chain (13, 14). Although the function of the sugar chains of IgM is not well understood, it has been demonstrated that the sugar chain structure at Asn-402 on mouse IgM influences the ability of the IgM-antigen complex to bind complement (15). Bazin et al. (16) reported that murine hybridoma-produced IgM lacking the sugar chain at Asn-563 had increased avidity for antigen. The function of the sugar chains of antibodies has been well studied in the case of IgG. For example, hypogalactosylation of IgG has been shown to affect some of the effector functions of the IgG molecule including binding to complement C1q and mannose-binding protein (17). Patients with rheumatoid arthritis have a higher frequency of IgG lacking galactose (18).

Our original interest was to investigate the effect of hypogalactosylation on the function of IgM. For this purpose, we attempted to isolate -1,4-GalT-I-null B cells producing IgM. Although we failed to isolate -1,4-GalT-I-null cells, we isolated a single knockout cell clone in which the -1,4-GalT level was reduced to half due to disruption of one of the two -1,4-GalT-I alleles. In the case of the IgM produced by this -1,4-GalT-I single knockout clone, hypogalactosylation was not observed; however, we discovered that the extent of bisecting GlcNAc addition to the sugar chains of the IgM was increased. To explain this unexpected increase in the extent of bisecting GlcNAc addition, we speculated that -1,4-GalT might compete with GnT-III for substrate in the cells. It has been previously reported that, in an in vitro system using sugar chains in a free form or glycopeptides, GnT-III and -1,4-GalT react with an agalactosyl nonbisected biantennary sugar chain as a common substrate, suggesting that GnT-III and -1,4-GalT enzymatically compete for the substrate in vitro (19, 20) (Fig. 1). However, the occurrence of such competition in intact cells was viewed as questionable considering the subcellular localization of the two enzymes, as reported previously (21-24). In the present study, we attempted to demonstrate experimentally the competition between -1,4-GalT and GnT-III in intact cells. We demonstrate that the extent of bisecting GlcNAc addition to the sugar chains of IgM is significantly correlated with the level of intracellular -1,4-GalT activity relative to that of GnT-III. Our studies demonstrate that sugar chain structures can be systematically and quantitatively controlled by regulating the levels of expression of glycosyltransferases.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Reactions catalyzed by GnT-III and -1,4-GalT. The addition of galactose residues is known to inhibit the addition of bisecting GlcNAc in vitro.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Lines-- The experiments were performed using the following two IgM-producing cell lines. Clone No. 12 is an Epstein-Barr virus-transformed human lymphoblastoid cell line producing human IgM against Pseudomonas aeruginosa Homma serotype B, and clone 3-4 is a hybridoma established by fusion of clone No. 12 and an established human myeloma cell line P109 mutated so as not to produce IgM µ-chain (25). Hybridoma 3-4 produces the same µ-chain peptide of IgM as No. 12, and IgM from 3-4 cells reacts with the same serotype of Pseudomonas aeruginosa as IgM from No. 12 cells.

Cell Culture-- Cells producing human monoclonal IgM were routinely maintained by static culture at 37 °C (5% CO₂/air) in tissue culture flasks containing NYSF-404 medium (26) and 20% fetal bovine serum (FBS). G418 (800 µg/ml) was added to cultures of -1,4-GalT-I or GnT-III transfectant cells.

For IgM production, cells were cultured in flasks (175 cm²) containing 80 ml of the serum-free medium for 3 days. The supernatant samples were then collected, filtered, and stored at 20 °C.

Isolation of a Cell Line Having Low -1,4-GalT Activity-- A human -1,4-GalT-I gene was isolated from a human leukocyte genomic library (CLONTECH, catalog number HL1111j) by plaque hybridization using a probe corresponding to exon 2 of the human -1,4-GalT-I gene. A -1,4-GalT-I gene fragment (XhoI fragment) containing exon 2 and part of introns 1 and 2 was inserted into the XhoI site of the expression vector pBluescript II-KS(+) (Stratagene), and its NcoI/XhoI fragment was deleted. An XhoI/BamHI fragment of pMC1neo poly(A) (Stratagene) containing a neomycin resistance gene was made blunt-ended by treatment with mung bean nuclease, followed by addition of a NotI linker, and the resulting fragment was inserted into the NotI site of -1,4-GalT-I exon 2 of the above vector in order to disrupt the -1,4-GalT-I gene. A herpes simplex virus-thymidine kinase coding sequence (PvuII fragment of pHSV-106, Life Technologies, Inc.) was inserted to the HincII site of pUC19 (Life Technologies, Inc.), and a BamHI/HindIII fragment containing the herpes simplex virus-thymidine kinase sequence from the resulting plasmid was inserted downstream of the neomycin resistance gene in the targeting vector. The targeting vector was designated pGEXN-Neo-TK.

No. 12 cells (5 × 10⁶ cells) were suspended in NYSF-404 medium; then 200 µg of the targeting vector pGEXN-Neo-TK was added, and the cells were subjected to electroporation (220 V/0.4 cm, 960 microfarads). After standing at room temperature for 10 min, the cells were transferred to NYSF-404 medium containing 20% FBS and cultured on plates at an appropriate dilution for 2 days. Then they were further cultured in NYSF-404 medium containing 800 µg/ml G418 and 20% FBS. After approximately 2 weeks, G418-resistant cells were selected. The G418-resistant cells were further cultured in NYSF-404 medium containing 2.5 ng/ml ganciclovir (GANC) and 20% FBS, and GANC-resistant cells were selected.

Homologous recombination was initially detected by PCR using a Neo primer and a GalT-int primer and next by using an F01 primer and a GalT-int primer. PCR-positive clones were subjected to Southern blot analysis to confirm that homologous recombination had occurred. Southern hybridization was carried out according to standard methods (27) using a digoxigenin-labeled DNA probe. The probe (GalT-int probe) was obtained from genomic DNA by PCR using the primers 5'-GGAGAATCAGATTGATCTAAGAGG-3' and 5'-CGTGGAAGGGATACTGGGGTCCCCTT-3'.

Establishment of Cell Lines Overexpressing -1,4-GalT or GnT-III-- The -1,4-GalT-I expression vector pCXN2-GalT was constructed by inserting the entire human -1,4-GalT-I coding region (28) (EcoRI fragment of pCT7-J20, a kind gift from Dr. Michiko Fukuda, The Burnham Institute, La Jolla) into the EcoRI site of the expression vector pCXN2 (29) (a kind gift from Dr. Jun-ichi Miyazaki, Osaka University, Japan) containing a neomycin resistance gene, in which a foreign gene is driven by a -actin promoter. The GnT-III expression vector pCXN2-rGnT-III was constructed by inserting the entire rat GnT-III coding region (EcoRI fragment of vector Act-3) (30) (a kind gift from Dr. Naoyuki Taniguchi, Osaka University, Japan) into the EcoRI site of the expression vector pCXN2. 100 µg of plasmid DNA and 7.5 × 10⁶ cells were suspended in 0.5 ml of NYSF-404 medium, and the cells were subjected to electroporation (220 V/0.4 cm, 960 microfarads). After standing at room temperature for 10 min, the cells were transferred to NYSF-404 medium containing 20% FBS and seeded on plates at an appropriate dilution. Two days later, G418 was added at 800 µg/ml, and the cells were further cultured. After approximately 2 weeks, G418-resistant cells were picked up.

Assay of GnT-III and -1,4-GalT Activities-- Cells (5 × 10⁵) were suspended in 5 µl of buffer (10 mM HEPES buffer, 1% Triton X-100, pH 7.2) and disrupted by sonication. The lysates were used as the crude enzyme preparations for the assay of GnT-III and -1,4-GalT activities.

GnT-III and -1,4-GalT activities were measured using pyridylaminated agalactosyl biantennary sugar chain as a substrate as described in our preceding paper (36). The substrate was prepared as reported previously (31). The specific activities of GnT-III and -1,4-GalT were expressed as nanomoles of GlcNAc or galactose transferred per h/10⁶ cells.

Purification of IgM-- IgM was purified by immunoaffinity chromatography. Cell culture supernatant was loaded onto an immunoaffinity column (CHROMATOP immobilized anti-human IgM antibody column, Nihon-gaishi, Japan) equilibrated with phosphate-buffered saline. After washing with CHROMATOP Wash Buffer A (Nihon-gaishi, Japan), IgM was eluted with 0.2 M Gly-HCl, pH 2.5. The eluate was immediately neutralized with 1 M Tris-HCl, pH 8.0, and concentrated and desalted by ultrafiltration using an Ultrafree 15 centrifugal filter (Millipore).

Release of Sugar Chains from IgM and Pyridylamination-- Purified IgM was lyophilized and redissolved at 2 mg/ml in 200 mM Tris-HCl buffer, pH 8.2, containing 10 M urea, 150 mM NaCl, and 20 mM dithiothreitol and incubated at 37 °C for 15 h. Alkylation was performed by adding 60 mM iodoacetic acid, and the mixture was incubated at 37 °C for 90 min. Thereafter, the solution was dialyzed against 50 mM Tris-HCl buffer containing 2 mM CaCl₂, pH 8.2. Enzymatic digestion was then performed by adding thermolysin (Merck) in an amount corresponding to 1/50 (w/w) of the IgM in the dialyzed sample. After incubation at 37 °C for 15 h, the digest was loaded on a Sephadex G-25 (Amersham Pharmacia Biotech) gel filtration column, and glycopeptide fractions were collected. The glycopeptides were dissolved in 100 µl of 100 mM citrate/phosphate buffer, pH 5.0, and digested with 0.4 milliunits of glycopeptidase A at 37 °C for 15 h. The digestion products were applied to a Sep-Pak Plus C18 Cartridge (Waters) pretreated with 10 ml of methanol and 5 ml of water, and then sugar chains were eluted with 6 ml of 5% acetonitrile in 0.1% trifluoroacetic acid solution.

The sugar chains obtained were pyridylaminated by the method of Kuraya and Hase (32). Excess reagents were removed by gel filtration on a Sephadex G-15 (Amersham Pharmacia Biotech) column (1 × 40 cm) equilibrated with 10 mM NH₄HCO₃.

Structural Analysis of PA-Sugar Chains-- The PA-sugar chains were first analyzed by anion-exchange HPLC, and the content of sialic acid was determined. Next, the whole PA-sugar chains were desialylated with sialidase from Arthrobacter ureafaciens in 0.2 M ammonium acetate buffer, pH 5.0, for 20 h at 37 °C. The desialylated sugar chains were separated as a neutral fraction by anion-exchange HPLC using a Mono Q column (5 × 50 mm, Amersham Pharmacia Biotech) as described in our preceding paper (36).

The desialylated PA-sugar chains were analyzed by a two-dimensional sugar mapping technique (33). First, the desialylated PA-sugar chains were separated by reversed-phase HPLC on a Shim-pack CLC-ODS column (6 × 150 mm, Shimadzu, Japan). Each sugar chain fraction was collected separately and then applied to the second column, TSKgel Amide-80 (4.6 × 250 mm, Tosoh, Japan). Elution conditions for these columns were as described by Tomiya et al. (33). Each PA-sugar chain fraction was sequentially digested with exoglycosidases to verify the structural identification.

In all HPLC systems, PA-sugar chains were detected on the basis of fluorescence, and the excitation and emission wavelengths used were 320 and 400 nm, respectively.

Enzymatic digestion of PA-sugar chains was performed with the following enzymes at 37 °C for 18 h. The PA-sugar chains were incubated with 20 milliunits of -fucosidase (bovine kidney) in 20 µl of 0.2 M ammonium acetate buffer, pH 4.5; with 20 milliunits of -galactosidase (jack bean) in 20 µl of 0.1 M citrate/phosphate buffer, pH 4.0; or with 20 milliunits of -N-acetylhexosaminidase (jack bean) in 20 µl of 0.1 M citrate/phosphate buffer, pH 5.0. The reaction mixture was heated at 100 °C for 3 min to terminate digestion.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation of a Cell Line Expressing a Low Level of -1,4-GalT-I and Structural Analysis of the IgM Sugar Chains-- The targeting vector pGEXN-Neo-TK (containing a neomycin resistance gene and a herpes simplex virus-thymidine kinase gene, Fig. 2A) designed for disrupting the -1,4-GalT-I allele was introduced into IgM-producing B cell clone No. 12. At first, 1,100 clones resistant to G418 were selected. Next, GANC-resistant clones were selected from among the G418-resistant clones. Among 580 clones resistant to both G418 and GANC, two clones were shown to be positive for the fragment of the expected length by PCR using the Neo primer and the GalT-int primer. One of the two clones, designated Y6-17, was confirmed to be a true disruptant by PCR using the F01 primer and the GalT-int primer. By Southern blot analysis using an external probe, it was also confirmed that the expected homologous recombination had occurred in Y6-17 (Fig. 2B). In an attempt to isolate a -1,4-GalT-I double knockout clone, a second targeting vector was introduced into Y6-17 cells to disrupt the remaining -1,4-GalT-I allele in the cells. However, we failed to obtain such a double knockout clone. Therefore, the single knockout clone Y6-17 was used as a cell line expressing a low level of -1,4-GalT in the present study. The level of -1,4-GalT activity in the Y6-17 cells was found to be 63% of that in the parental cells (Table I). The levels of GnT-III activity and IgM production did not significantly differ comparing the Y6-17 cells and the parental cells.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Targeted disruption of the -1,4-GalT-I gene by homologous recombination. A, targeting strategy. Coding and non-coding exons of -1,4-GalT-I are indicated by open and hatched boxes, respectively. The neomycin resistance gene is indicated by a dotted box. Homologous recombination was detected by PCR first by using the Neo primer and the GalT-int primer and second by using the F01 primer and the GalT-int primer. The GalT-int probe shows the position of the external probe used for Southern blot analysis, and the expected HindIII/XhoI fragments are indicated by arrows. B, Southern blot analysis of targeted cells. Genomic DNA from parental No. 12 cells and targeted Y6-17 cells were digested with HindIII and XhoI and hybridized with the GalT-int probe. The expected DNA fragments in the case of the mutant allele (Mt) and wild-type allele (Wt) are indicated. kb, kilobase pairs.

View this table: [in this window] [in a new window]   Table I -1,4-GalT and GnT-III activities in parental cells and transfectants Activity is expressed as nmol of GlcNAc (or Gal) transferred per h per 106 cells. Values are the means ± S.D. of duplicate assays.

Sugar chains at five N-glycosylation sites on the IgM µ-chain were collectively excised and purified from the Y6-17 cells and the parental No. 12 cells. Sugar chain structures were analyzed by the two-dimensional mapping technique of Tomiya et al. (33) using pyridylaminated (PA-) sugar chains. Reversed-phase HPLC profiles of desialylated PA-sugar chains derived from each IgM are shown in Fig. 3. The proportion of each type of sugar chain is shown in Table II. Complex-type biantennary sugar chains comprised over 50% and high mannose-type sugar chains comprised about 30% of the total sugar chains of IgM in both the Y6-17 cells and the parental No. 12 cells. Other types of sugar chains comprised about 15% of the total, among which each of the individual types comprised merely 2% or less. Unexpectedly, biantennary sugar chains with an agalactosyl terminus were not detected in IgM from the Y6-17 cells that showed reduced levels of -1,4-GalT activity as not seen in IgM from the parental cells. On the other hand, a change in the extent of bisecting GlcNAc addition was observed. The rate was 62.1% in the IgM from No. 12 cells, whereas it was 73.8% in the IgM from Y6-17 cells. Thus, the extent of bisecting GlcNAc addition was increased in the case of the -1,4-GalT-I single knockout cells.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Reversed-phase HPLC of desialylated PA-sugar chains derived from IgM produced by parental No. 12 cells and transfectants. Structures of individual components are shown by symbols. Structures of high mannose-type sugar chains are indicated as M5-M9 according to the number of mannose residues.

Isolation of Clones Overexpressing -1,4-GalT-I and Structural Analysis of the IgM Sugar Chains-- The -1,4-GalT-I expression vector pCXN2-GalT (containing a neomycin resistance gene) was introduced into IgM-producing B cells (No. 12). By G418 selection, 170 clones were isolated. The transfectant B12/neo-1 served as a negative control (vector pCXN2 was introduced). Among the G418-resistant clones, 2 clones designated B12/G-1 and B12/G-2 were chosen as clones overexpressing -1,4-GalT-I. The levels of -1,4-GalT activity in the B12/G-1 and B12/G-2 cells were 5.7- and 2.7-fold higher, respectively, than that in the parental No. 12 cells (Table I). The levels of GnT-III activity and IgM production did not significantly differ comparing these -1,4-GalT-I transfectants and the parental No. 12 cells.

Sugar chains of IgM from the B12/G-1 and B12/G-2 cells were analyzed as described above (Fig. 3 and Table II). As in the case of IgM from the parental No. 12 cells, complex-type biantennary sugar chains comprised over 50% and high mannose-type sugar chains comprised about 15% of the total sugar chains of IgM in these two -1,4-GalT transfectants. The extent of bisecting GlcNAc addition to the biantennary sugar chains in IgM was 12.7 and 23.9% in the case of the B12/G-1 cells and the B12/G-2 cells, respectively. The extent in the case of the No. 12 cells was 62.1% as described above. Thus, the extent of bisecting GlcNAc addition was reduced in the case of these clones overexpressing -1,4-GalT-I.

Isolation of Clones Overexpressing GnT-III and Structural Analysis of the IgM Sugar Chains-- Cells overexpressing GnT-III were isolated and found to show an increased extent of bisecting GlcNAc addition, as compared with B cell clone No. 12 and hybridoma clone 3-4 which were used as the parental cell lines. GnT-III activity is suppressed in the 3-4 clone by the influence of its parent, clone p109, which displays no GnT-III activity, and the extent of bisecting GlcNAc addition is known to be very low in the 3-4 cells (37). The vector pCXN2-rGnT-III (containing a neomycin-resistance gene) was introduced into No. 12 cells and 3-4 cells. By G418 selection, 24 clones and 36 clones were isolated from No. 12 cells transfected with GnT-III and from 3-4 cells transfected with GnT-III, respectively. The transfectant clones B12/neo-2 and 3-4/neo served as negative controls (vector pCXN2 was introduced). One of the G418-resistant B cell transfectants, designated B12/III, showed 16-fold higher GnT-III activity than parental No. 12 cells (Table I). The levels of -1,4-GalT activity and IgM production did not significantly differ comparing the B12/III cells and the parental No. 12 cells. Three G418-resistant transfectants derived from clone 3-4 cells, designated 3-4/III-1, 3-4/III-2, and 3-4/III-3, were chosen as clones overexpressing GnT-III. The levels of GnT-III activity in 3-4/III-1, 3-4/III-2, and 3-4/III-3 cells were 1188-, 727-, and 15-fold higher than that in the parental 3-4 cells, respectively (Table I). The levels of -1,4-GalT activity and IgM production did not significantly differ comparing the three GnT-III transfectants derived from clone 3-4 and the parental clone 3-4.

The extent of bisecting GlcNAc addition to the biantennary sugar chains in IgM from B12/III cells was 90.1% (Fig. 3 and Table II), much higher than that in the case of the parental No. 12 cells (62.1% as described above). The extent was 4.0% in the case of the hybridoma clone 3-4. With clones 3-4/III-1, 3-4/III-2, and 3-4/III-3, the extents were 94.4, 90.0, and 9.5%, respectively, indicating that the extent of bisecting GlcNAc addition was increased in the case of these GnT-III transfectants as compared with the parental 3-4 cells (Fig. 4 and Table III). It should be noted that the extent observed in the case of clone 3-4/III-3 reflected a relatively small increase in GnT-III activity compared with the significant increases observed in the case of clones 3-4/III-1 and 3-4/III-2.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Reversed-phase HPLC of desialylated PA-sugar chains derived from IgM produced by hybridoma 3-4 cells and transfectants. Structures of individual components are shown as in Fig. 3.

Relationship between the Extent of Bisecting GlcNAc Addition to N-Linked Sugar Chains and the Level of Expression of -1,4-GalT Relative to That of GnT-III-- The relationship between the extent of bisecting GlcNAc addition to the biantennary sugar chains of IgM and the ratio of -1,4-GalT activity to GnT-III activity in the IgM-producing cells were examined (Fig. 5). A significant correlation was found between them.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Relationship between the extent of bisecting GlcNAc addition and the ratio of -1,4-GalT activity to GnT-III activity. The extent of bisecting GlcNAc addition was expressed as the percentage of bisected biantennary chain in the total of bisected and nonbisected biantennary chains.

Relationship between the Extent of Sialic Acid Addition and the Extent of Bisecting GlcNAc Addition-- PA-sugar chains derived from IgM produced by each cell line were analyzed by anion-exchange HPLC. The extent of sialic acid addition to the sugar chains was determined on the basis of their negative charges. PA-sugar chains were categorized on the basis of their elution positions into one neutral and two acidic fractions corresponding to asialo, monosialo, and disialo PA-sugar chains, respectively. The relative proportions of these are shown in Fig. 6. In the case of Y6-17, B12/III, 3-4/III-1, and 3-4/III-2 clones, where the extent of bisecting GlcNAc addition was increased, the proportion of asialo form was slightly increased, and the average number of sialic acid was slightly reduced compared with parental clones. On the other hand, in the case of B12/G-1 and B12/G-2 clones, where the extent of bisecting GlcNAc addition was reduced, the proportion of disialo form was slightly increased, and the average number of sialic acid was slightly increased compared with parental clone, No. 12. Thus, a rough tendency of the extent of addition of sialic acid residues to be somewhat diminished by the presence of bisecting GlcNAc residue was observed.

View larger version (40K): [in this window] [in a new window]   Fig. 6.   Analysis of sialylation. PA-sugar chains from each IgM were separated according to the negative charges by anion-exchange HPLC. The proportions of asialo, monosialo, and disialo sugar chains were determined on the basis of the peak area. Increase and decrease of values compared with parental clones are indicated by arrows.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We originally attempted to obtain -1,4-GalT-I-deficient B cells producing IgM to examine the effect of hypogalactosylation on the function of IgM. For this purpose, we isolated a single knockout clone in which the intracellular level of -1,4-GalT activity was reduced to half due to disruption of one of the two -1,4-GalT-I alleles. In the -1,4-GalT-I single knockout cells, under-galactosylated sugar chains were not observed in the IgM produced. Instead, we discovered an interesting phenomenon, the extent of bisecting GlcNAc addition to the sugar chains of IgM was increased in the cells, suggesting that -1,4-GalT is somehow involved in the formation of the bisecting GlcNAc.

It has been shown in vitro that GnT-III and -1,4-GalT can react with an agalactosyl nonbisected biantennary sugar chain as a common substrate (19, 20). When -1,4-GalT first catalyzes the conversion of this sugar chain to a galactosylated form, the latter product can no longer be a substrate for GnT-III, indicating that GnT-III and -1,4-GalT compete for an agalactosyl nonbisected biantennary sugar chain as a common substrate. Therefore, the increased extent of bisecting GlcNAc addition found in the case of -1,4-GalT-I single knockout cells could be explained by the competition between GnT-III and -1,4-GalT for substrate. Nishiura et al. (34) also suggested that the extent of bisecting GlcNAc formation and that of galactose addition in IgG are controlled by the balance between GnT-III activity and -1,4-GalT activity. For such competition to occur, however, GnT-III and -1,4-GalT should be located at the same subcellular site. It is well known that glycosyltransferases are located on the specific lumenal side of the Golgi apparatus in the order of the oligosaccharide processing pathway. -1,4-GalT is known to be localized on the trans Golgi stacks (21, 22), whereas GnT-III is considered to be present on the medial Golgi stacks (23, 24). That is to say, the two enzymes were considered to be located at different sites. Therefore, it has been thought that -1,4-GalT would not likely be able to compete with GnT-III in intact cells, even if the two enzymes compete in vitro when together they are allowed to react with an agalactosyl nonbisected biantennary sugar chain.

In the present study, we examined whether -1,4-GalT and GnT-III compete for a common substrate within the cells. As described above, the extent of bisecting GlcNAc addition to the sugar chains of IgM was increased in the case of -1,4-GalT-I single knockout clone Y6-17, whereas the extent was decreased in the case of the -1,4-GalT-I transfectants. In the GnT-III transfectants, the extent increased along with the increase in levels of GnT-III activity. The extent of bisecting GlcNAc addition to the sugar chains of IgM was significantly correlated with the level of intracellular -1,4-GalT activity relative to that of GnT-III. From these findings, the competition between the two enzymes in cells is apparent. Therefore, it was concluded that the addition of the bisecting GlcNAc residue was not only catalyzed by GnT-III but was also controlled by the level of expression of -1,4-GalT relative to that of GnT-III. Thus, we have demonstrated that the extent of bisecting GlcNAc addition to sugar chains can be controlled by changing the levels of expression of the competing enzymes, -1,4-GalT and GnT-III. The demonstrated competition between -1,4-GalT and GnT-III means that these two enzymes co-exist at the same site in the cells. However, since we still do not know where the overexpressed -1,4-GalT-I and GnT-III proteins are localized on the Golgi stacks in the transfectants, further study is needed to confirm completely the competition between these two enzymes in the cells.

It is noted that a rough tendency of addition of sialic acid residues to be somewhat suppressed by the presence of bisecting GlcNAc residue was observed. This result might indicate that the presence of the bisecting GlcNAc residue inhibits sialyltransferase reaction as reported in vitro (35).

A functional difference between IgM molecules with and without the bisecting GlcNAc has not been clearly demonstrated so far in our studies on the pharmacokinetics and binding to C1q (data not shown). The functional role of the bisecting GlcNAc residues added to the sugar chains of IgM should be elucidated through further investigation.

It is well known that sugar chains play important roles in defining the characteristics of glycoproteins such as their biological activity, immunogenicity, pharmacokinetics, solubility, and protease resistance. The functions of glycoproteins are expected to be improved by remodeling of the sugar chain structures. Techniques for controlling the sugar chain structures are necessary for this purpose. Here we have succeeded in controlling sugar chain structures systematically for the first time by regulating the levels of expression of glycosyltransferases, even though the control was performed only in terms of the attachment of bisecting GlcNAc. The present work should serve as a cornerstone for further studies aimed to control sugar chain structures.

	ACKNOWLEDGEMENTS

We thank Dr. Akira Kobata and Dr. Tamao Endo (Tokyo Metropolitan Institute of Gerontology, Japan) for their helpful comments. We are also grateful to Dr. Naoyuki Taniguchi (Osaka University, Japan) for providing rat GnT-III cDNA and to Dr. Michiko N. Fukuda (The Burnham Institute, La Jolla) for providing human -1,4-GalT-I cDNA. We also thank Dr. Jun-ichi Miyazaki (Osaka University, Japan) for providing vector pCXN2.

	FOOTNOTES

^* This work was supported by the New Energy and Industrial Technology Development Organization (NEDO) as a part of the Research and Development Projects of the Industrial Science and Technology Frontier Program in Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Life Science Laboratory, Mitsui Chemicals, Inc., 1144 Togo, Mobara, Chiba 297-0017, Japan. Tel.: 81-475-25-6727; Fax: 81-475-25-6553.

Published, JBC Papers in Press, May 17, 2000, DOI 10.1074/jbc.M002693200

	ABBREVIATIONS

The abbreviations used are: GnT-III, UDP-N-acetylglucosamine:-D-mannoside -1,4-N-acetylglucosaminyltransferase; -1, 4-GalT, UDP-galactose:N-acetylglucosamine -1,4-galactosyltransferase; PA, pyridylamino-; FBS, fetal bovine serum; PCR, polymerase chain reaction; HPLC, high performance liquid chromatography; GANC, ganciclovir.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. Klisch, E. Jeanrond, P.-C. Pang, A. Pich, G. Schuler, V. Dantzer, M. P Kowalewski, and A. Dell A Tetraantennary Glycan with Bisecting N-Acetylglucosamine and the Sda Antigen is the Predominant N-Glycan on Bovine Pregnancy-Associated Glycoproteins Glycobiology, January 1, 2008; 18(1): 42 - 52. [Abstract] [Full Text] [PDF]

				A. Kobata The history of glycobiology in Japan Glycobiology, August 1, 2001; 11(8): 99R - 105R. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/31/23456    most recent M002693200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fukuta, K. Articles by Makino, T. Search for Related Content PubMed PubMed Citation Articles by Fukuta, K. Articles by Makino, T. Social Bookmarking                      What's this?