| Glycobiology | Pages |
The effect on IgG glycosylation of altering [beta]1,4-galactosyltransferase-1 activity in B cells
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
The effect on IgG glycosylation of altering [beta]1,4-galactosyltransferase-1 activity in B cells
Introduction
[beta]1,4-galactosyltransferase ([beta]4Gal-T) is located in the trans Golgi of most mammalian cells where it galactosylates GlcNAc residues on de novo synthesized glycoproteins. This process can generate oligosaccharide structures which are either fully galactosylated or only partially galactosylated. At Asn297 in IgG there is a conserved N-linked glycosylation site which possesses a complex-type oligosaccharide chain. It is known that B lymphocyte clones secrete a number of IgG glycoforms having variable levels of galactose on this oligosaccharide. Hypogalactosylation of IgG has been shown to affect some of the effector functions of the IgG molecule including binding to complement component C1q (Tsuchiya et al., 1989; Wright and Morrison, 1998), mannose-binding protein (Malhotra et al., 1995; Wright and Morrison, 1998), and, in some studies, Fc[gamma]R (Tsuchiya et al., 1989; Cant et al., 1994; Kumpel et al., 1994), although the particular effects observed depend upon the details of the analysis. It is currently unclear if the biological effects of IgG hypogalactosylation are a direct result of the absence of galactose or a secondary effect on the protein conformation within the Fc[gamma]2 region. Recently, it has been suggested that the orientation of the N-glycan chain on IgG affects its ability to activate complement (White et al., 1997).
Patients with rheumatoid arthritis (RA) have a higher frequency of IgG lacking galactose (referred to as IgG G0) when compared to age-matched controls (Parekh et al., 1988). This glycoform, which has terminal GlcNAc residues, was found to be more pathogenic than its galactosylated counterpart in an animal model of arthritis (Rademacher et al., 1994). The production of IgG G0 is a presecretory event (Bodman et al., 1992) and reduced [beta]4Gal-T activity has been reported in the B cells from patients with RA (Axford et al., 1987; Furukawa et al., 1990; Wilson et al., 1993). However, it has been questioned whether differences in [beta]4Gal-T levels affect the number of G0 structures on human IgG because three separate B cell lines with different endogenous [beta]4Gal-T activities were found to secrete IgG bearing similar amounts of galactose (Kumpel et al., 1994). Furthermore, it has very recently become apparent that [beta]4Gal-T activity is not mediated by a single enzyme, but that there are at least six functional [beta]4Gal-T genes in man (Almeida et al., 1997; Lo et al., 1998; Sato et al., 1998a). These enzymes are now designated [beta]4Gal-T1 for the 'classical" [beta]4Gal-T, and [beta]4Gal-T2 to [beta]4Gal-T6 for the recently cloned transferases. The fine substrate specificity of each of these enzymes has yet to be fully analyzed, but it is clear that they do not all function in exactly the same way. For example, while [beta]4Gal-T1 and [beta]4Gal-T2 exhibit lactose synthetase activity in the presence of [alpha]-lactalbumin, [beta]4Gal-T3 and [beta]4Gal-T4 do not (Almeida et al., 1997; Sato et al., 1998a,b).
In order to determine if the [beta]4Gal-T1 member of the [beta]4Gal-T family can utilize the oligosaccharides attached to the IgG polypeptide as an acceptor substrate, and to determine if it is possible to purposefully alter the glycosylation of IgG produced by a single B cell line, we have transfected B cells with vectors expressing sense or antisense human [beta]4Gal-T1 cDNA. The resulting transfectants, with different [beta]4Gal-T activities, were cultured and the secreted IgG assayed for terminal galactose.
Results
B cell [beta]4Gal-T1 transfectants express Golgi-associated but not cell surface enzyme
The anti-human [beta]4Gal-T mAb UCLgt1E7 (Keusch et al., unpublished observations) was used to identify [beta]4Gal-T in the transfected B cells. [beta]4Gal-T was only detected in B cell transfectants which had been fixed and permeabilized and not on the cell surface (Figure
Figure 1. Flow cytometric analysis of (a) nonpermeabilized and (b) fixed and permeabilized B cell transfectants, SpcDNA3-Gal-T1 (bold line) and ASpcDNA3-Gal-T1 (thin line) stained with anti-[beta]4Gal-T mAb UCLgt1E7 followed by rabbit F(ab[prime])2 anti-mouse IgG-FITC. The dotted line represents cells incubated with a negative isotype control MOPC-21 followed by rabbit F(ab[prime])2 anti-mouse IgG-FITC. Quantification of intracellular and secreted [beta]4Gal-T protein
The SpcDNA3-Gal-T1 B cell transfectants expressed approximately 5.5-fold more cell-associated [beta]4Gal-T protein (24.9 ± 2.25 ng/mg total protein) compared to the ASpcDNA3-Gal-T1 transfectants (4.0 ± 0.75 ng/mg total protein, p < 0.001, unpaired Student's t-test; Figure
Figure 2. [beta]4Gal-T protein levels were quantified in B cell transfectants SpcDNA3-Gal-T1(solid bars) and ASpcDNA3-Gal-T1 (shaded bars) using a sandwich ELISA. (a) Cellular [beta]4Gal-T (ng [beta]4Gal-T/mg total protein) and (b) secreted [beta]4Gal-T (ng [beta]4Gal-T/106 viable cells). The mean + 1 SEM are indicated for experiments measuring [beta]4Gal-T protein at eight different time points during continuous culture of the transfectants. [beta]4Gal-T enzyme activity is decreased in the antisense transfectants
The cell lysates from SpcDNA3-Gal-T1 B cell transfectants had a maximum 2.5-fold higher level of [beta]4Gal-T enzyme activity relative to those from the ASpcDNA3-Gal-T1 B cell transfectants (p < 0.001) using the exogenous acceptor substrate GlcNAc-pITC-BSA (Figure
Figure 3. [beta]4Gal-T activity in B cell transfectants assayed using the exogenous acceptor GlcNAc-pITC-BSA. Data in nmol/mg total cellular protein/h are presented for SpcDNA3-Gal-T1 (solid bar), ASpcDNA3-Gal-T1 (shaded bar), and control plasmid pcDNA3 without an insert (open bar). Error bars of 1 SEM are indicated. The results shown were obtained after 7 days of culture. The galactosylation of B cell substrates is altered
An increase in the number of endogenous cell-associated Gal[beta]1,4GlcNAc structures was observed in SpcDNA3-Gal-T1 relative to the ASpcDNA3-Gal-T1 B cell transfectants (Figure
Figure 4. Flow cytometric analysis of fixed and permeabilized B cell transfectants SpcDNA3-Gal-T1 (bold line) and ASpcDNA3-Gal-T1 (thin line) stained with biotinylated RCAI followed by streptavidin-FITC. The dotted line represents a negative control in which transfected cells were incubated with streptavidin-FITC only. No differences were seen between the two transfectants with the negative control.
Figure 5. IgG secreted from the B cell transfectants SpcDNA3-Gal-T1 (solid bar) and ASpcDNA3-Gal-T1 (shaded bar) assayed for terminal galactosylation using an enzyme-linked lectin assay. IgG G0 represents IgG lacking galactose. One SEM is indicated.
The particular glycoforms borne by proteins can profoundly influence their function (Wright and Morrison, 1997). It is therefore important to understand the mechanisms by which glycosylation is controlled, the regulation of these processes, and to develop strategies by which the glycosylation of a protein can be tailored in a very highly specific manner.
In order to examine the relationship between the level of [beta]4Gal-T1 and galactosylation, we modified [beta]4Gal-T1 expression in an IgG-secreting human B cell line. Alterations in the level of galactose on IgG are known to affect the biological properties of this molecule. Our approach was to specifically alter the level of the glycosyltransferase by stable transfection with vectors expressing [beta]4Gal-T1 in either the sense or the antisense orientations. A similar approach was used by Zeng and colleagues using antisense oligonucleotides for GM2- and GD3-synthase to modify glycolipid synthesis in HL60 cells (Zeng et al., 1995). Other approaches that have been used to alter protein glycosylation include treatment of the glycoprotein with individual exoglycosidases (Katoh et al., 1995), alteration of cell culture conditions (Kumpel et al., 1994), the use of transgenic (Prieto et al., 1995) and knockout (Asano et al., 1997; Lu et al., 1997) animals, and the use of Lec mutants of CHO cells in which individual glycosyltransferases are absent or overexpressed (Stanley et al., 1996).
In our experiments we found that a 2.5-fold difference in [beta]4Gal-T activity between the SpcDNA3-Gal-T1 and ASpcDNA3-Gal-T1 B cell transfectants resulted in different levels of galactosylation of cell-associated endogeneous B cell substrates. Furthermore, there was a substantial difference in Fc-associated IgG G0 structures between the transfectants. In other studies increasing the [beta]4Gal-T activity 3-fold in F9 cells by transfection with the [beta]4Gal-T1 gene did not result in any changes in lysosomal-associated membrane protein-1 (LAMP-1) galactosylation (Youakim and Shur, 1993). This difference between the two proteins could be due to different substrate preferences of [beta]4Gal-T1 in relation to other [beta]4Gal-T enzymes, and/or to the internal nature of the Fc-associated oligosaccharides on IgG. Culturing the F9 cells at 20°C, thereby increasing the transit time of glycoproteins through the Golgi, did enhance both the number and extent of polylactoaminoglycans attached to LAMP-1 and -2 (Wang et al., 1991).
Long and short forms of [beta]4Gal-T1 are produced by the use of alternative initiation codons (Russo et al., 1990). In the ASpcDNA3-Gal-T1 transfectants only the endogenous B cell [beta]4Gal-T is present (at a reduced level), whereas in the SpcDNA3-Gal-T1 transfectants there will additionally be overexpression of the full-length [beta]4Gal-T1 cDNA. The ratio of cellular to secreted [beta]4Gal-T protein remained relatively constant irrespective of the [beta]4Gal-T activity level in the B cell transfectants. [beta]4Gal-T along with a number of other Golgi proteins can occur as dimers (Bendiak et al., 1993; Fleischer et al., 1993). A dimer of ST6Gal I which accounted for a third of the total enzyme in the Golgi of HeLa cells was found to be catalytically inactive, but could still bind its acceptor substrate (Ma and Colley, 1996). If Golgi-associated [beta]4Gal-T consists of a mixture of monomers, dimers, and higher order oligomers, then the increased or decreased expression of [beta]4Gal-T1 may influence structural and functional aspects of the enzyme, and would explain the lack of a directly quantitative association between enzyme protein levels and enzyme activity. Modulation of [beta]4Gal-T activity might also occur through posttranslational modifications such as phosphorylation (Bunnell et al., 1990).
The binding of the anti-[beta]4Gal-T mAb to fixed and permeabilized B cells resulted in characteristic Golgi staining, the normal cellular location of [beta]4Gal-T. However, with the monoclonal antibody used (UCLgt1E7), cell surface [beta]4Gal-T was not detected, even after transfection with the [beta]4Gal-T1 cDNA in the sense orientation. Low but measurable cell surface [beta]4Gal-T has been reported on another EBV-transformed B cell line, JY (Mrkoci-Felner et al., 1997). This difference might be due either to inherent differences between the two cell lines, or related to different expression of [beta]4Gal-T epitopes recognized by different monoclonal antibodies. However, a greater than 12-fold expression of GalT1 activity in COS-7 cells transfected with SpcDNA3-GalT1, resulted in detection of the enzyme at the cell surface using the UCLgt1E7 mAb (data not shown).
Decreased levels of galactose on IgG are associated with certain rheumatological disorders. Lower [beta]4Gal-T enzyme levels have been described in these diseases although the actual [beta]4Gal-T enzyme(s) involved has yet to be established. While the Fc-associated oligosaccharides are not usually fully galactosylated, those in the Fab usually are (Rademacher et al., 1996).This suggests that the relative inaccessibility of the Fc-associated sugars which are enclosed within the C[gamma]2 domains plays a role in the regulation of galactosylation. Reduced [beta]4Gal-T enzyme activity could be due to an alteration in a [beta]4Gal-T (misfolding or posttranslational modification) which causes the enzyme to inefficiently galactosylate IgG or the enzyme may be fully functional but produced at a lower level. We show here that a reduction in the level of enzyme comparable to that seen in some patients with RA is sufficient to cause an increase in Fc-associated IgG G0.Although we have been unable to detect a direct relationship between B cell [beta]4Gal-T1 mRNA levels and IgG hypogalactosylation in patients with RA (Jeddi et al., 1996), it would be of interest to determine if [beta]4Gal-T gene transfection into the B cells from patients with RA is able to correct the glycosylation of their IgG. This would not be straightforward, however, due to the fact that EBV transformation itself has been reported to upregulate [beta]4Gal-T activity in lymphocytes isolated from patients with RA, and therefore they no longer reflect the natural in vivo situation with respect to increased IgG G0 (Wilson et al., 1993). The use of freshly isolated patients' B cells would also be problematic due to the necessity to maintain untransformed B cells in long term culture during the selection of transfectants. Although this can be achieved by culturing B cells in the presence of CD40L and appropriate cytokines (Rousset et al., 1995), the effect of such procedures on protein glycosylation are currently unknown. [beta]4Gal-T1 antisense/sense vector constructs
The full-length [beta]4Gal-T1 cDNA from clone CT7-J20 (Masri et al., 1988) was excised out of Bluescript KS and subcloned into the eukaryotic expression vector pcDNA3, under the control of the CMV promoter, in the sense (BamHI/XhoI digests, SpcDNA3-Gal-T1) and the antisense (HindIII/BamHI digests, ASpcDNA3-Gal-T1) directions. Ligation and correct orientation of the [beta]4Gal-T1 inserts into the vector was verified using differential restriction enzyme digests and DNA sequencing of the 5[prime] insert end using the Sequenase v2.0 DNA sequencing kit according to the manufacturer's instructions (United States Biochemical). B cell transfection with [beta]4Gal-T1 constructs
The JAC-10 EBV-transformed human B cell line (Kumpel et al., 1994) was cultured in RPMI-1640 with 25 mM HEPES and sodium bicarbonate and supplemented with 10% heat-inactivated fetal bovine serum (<0.1 µg/ml IgG, Life Technologies, Paisley, UK), 2 mM glutamine, 50 IU penicillin, and 50 µg/ml streptomycin ('complete medium"). Prior to transfection, B cells were washed free of antibiotics using Hanks' balanced salt solution, then resuspended at 4 × 106 cells/ml in complete medium (without antibiotics). Cells were distributed at 500 µl /well into 24 well tissue culture plates. Three micrograms of PEG-precipitated DNA at 1 mg/ml was added to 100 µl of RPMI and mixed gently with 100 µl of Lipofectin (Life Technologies, Paisley, UK) at 100 µg/ml in RPMI for 15 min at room temperature (Cumin et al., 1993). The DNA-Lipofectin complexes were added to the cells and incubated at 37°C in 5% CO2 for 8 h. The cells were then washed and resuspended in fresh complete medium at 0.5 × 106 cells/ml and recultured. Forty-eight hours later cells were washed and resuspended in complete medium containing 800 µg/ml G418(Santerre et al., 1984) (pretitrated against untransfected cells to determine the effective lethal dose). The cells were then continuously cultured in G418-containing complete medium, with resuspension at 0.5 × 106 cells/ml every 72 h in fresh medium. After 6 weeks of selection, aliquots of the transfectants were frozen down and stored in liquid nitrogen. All the data reported were obtained following recovery of the cells from storage and subsequent culture for 7 days (except where stated otherwise) of the transfectants in complete medium with G418. Cells were lysed and assayed for [beta]4Gal-T enzyme activity and protein, and culture supernatants (containing secreted IgG and secreted [beta]4Gal-T) were collected and stored in aliquots at -20°C prior to analysis. [beta]4Gal-T enzyme activity
After reculturing the transfectants for 7 days, [beta]4Gal-T activity was measured using GlcNAc-pITC-BSA as the acceptor substrate in an ELISA-based assay as described previously (Keusch et al., 1995). [beta]4Gal-T protein quantification
Cellular and secreted [beta]4Gal-T protein was measured in the B cell lysates and culture supernatants, respectively, using a sandwich ELISA which proved highly reproducible and sensitive down to 1 ng/ml (50 pg per well) of [beta]4Gal-T protein (Keusch et al., unpublished observations). In this assay Maxisorp ELISA plates were coated overnight at 4°C with 50 µl/well of streptavidin at 5 µg/ml in 0.15 M phosphate-buffered saline pH7.4 (PBS), and blocked with 100 µl of PBS-1% BSA. After washing, half of the plate was incubated with 50 µl/well of biotinylated isotype control MOPC-21 IgG1 at 5 µg/ml in PBS while the other half was incubated with a synergistic pair of biotinylated anti-human [beta]4Gal-T mAbs UCLgt1B6 and UCLgt1H11, each at 2.5 µg/ml in PBS, for 2 h at 37°C and then overnight at 4°C. Following washing, the plates were incubated with 50 µl/well of purified human milk [beta]4Gal-T standards at doubling dilutions from 60 ng/ml, or with 50 µl of test sample. All assay points were carried out in duplicates on both sides of the plates. Plates were incubated for 2 h at 37°C, then washed and incubated with 50 µl of a previously described affinity-purified rabbit anti-human [beta]4Gal-T (Watzele et al., 1991) at 5 µg/ml in PBS-Tween 20 0.05% (PBS-T)-1% BSA for 1 h at 37°C. After washing, plates were incubated with 50 µl/well of goat F(ab[prime])2 anti-rabbit IgG-horseradish peroxidase (Jackson ImmunoResearch Laboratories, West Grove, PA) at 1:2000 in PBS-T-1% BSA for 1 h at 37°C. Plates were washed and developed using 50 µl/well of 1 mg/ml o-phenylenediamine in 0.1 M citrate phosphate buffer, pH 5.0 with 0.03% H2O2. The reaction was stopped with 100 µl/well of 3N H2SO4 and the absorbances read at 490 nm using a Dynatech MR5000 ELISA plate reader. Immunofluorescence staining of B cell transfectants
Aliquots of 2.5 × 105 cells were washed in PBS and the cell pellets each resuspended in 100 µl of a fully characterized murine IgG1 anti-human [beta]4Gal-T mAb UCLgtIE7 (Keusch et al., unpublished observations) or IgG1 isotype control mAb MOPC-21 at 20 µg/ml in buffer (PBS-1% BSA with 15 mM NaN3), and incubated for 45 min at 4°C. Following two 1 ml washes, 50 µl of rabbit F(ab")2 anti-mouse IgG-FITC (Dako, Buckinghamshire, UK), diluted 1:10 in buffer was used to resuspend the cell pellet and incubated for 45 min at 4°C in the dark. Cells were washed twice in PBS and resuspended in 250 µl PBS. For intracellular staining, 1 × 106 cells were fixed and permeabilized by resuspension in 500 µl of working strength PermeaFix reagent (Ortho Diagnostic Systems, Buckinghamshire, UK) followed by a 40 min incubation at 20°C in the dark and two washes in 1 ml of buffer. The staining procedure was as described above except that an additional 30 min wash step was included following the incubation with the primary antibody. For the detection of exposed galactose, 100 µl of biotinylated Ricinus communis agglutinin I (RCAI; Vector Laboratories, Peterborough, UK) at 0.08 µg/ml in buffer was added to PermeaFixed cells and incubated as above. Cells were washed and 50 µl of 1:100 streptavidin-FITC (Serotec, Oxford, UK) added and incubated for 45 minutes at 4°C. Following washing, a minimum of 5000 stained cells were assessed using a Becton Dickinson FACScan. IgG galactosylation ELLA
This enzyme-linked lectin assay was performed as previously described (Sumar et al., 1990) on supernatants obtained following 3 days of reculture of the transfectants. The standards used contained known amounts of IgG G0, determined by the Department of Biochemistry, University of Oxford, using the hydrazinolysis method. A standard curve of the absorbance ratios at 410 nm of Bandeiraea simplicifolia II (BSII)/RCAI against known amounts of IgG G0 was plotted and the sample absorbances interpolated.
We thank Prof. Eric Berger (Institute of Physiology, University of Zurich) for the affinity-purified rabbit anti-human [beta]4Gal-T antibody and human milk [beta]4Gal-T, and Dr. Michiko Fukuda (La Jolla Cancer Research Foundation) for the [beta]4Gal-T1 cDNA clone CT7-J20. The JAC-10 EBV-transformed cell line was kindly donated by Belinda Kumpel (International Blood Group Reference Laboratory, Bristol, UK). We thank the Frances and Augustus Newman Foundation for supporting this research. J.K. was in receipt of an MRC UK Ph.D. studentship award.
ASpcDNA3-Gal-T1, antisense [beta]4Gal-T1 construct; BSII, Bandeiraea simplicifolia II; BSA, bovine serum albumin; ELISA, enzyme-linked immunosorbent assay; ELLA, enzyme-linked lectin assay; [beta]4Gal-T1, [beta]1,4 galactosyltransferase-1; GlcNAc, N-acetylglucosamine; GlcNAc-pITC-BSA, N-acetylglucosamine-phenylisothiocyanate-bovine serum albumin; IgG G0, IgG lacking galactose; mAb, monoclonal antibody; PBS, 0.15 M phosphate-buffered saline; pcDNA3, eukaryotic expression vector without an insert; RCAI, Ricinus communis agglutinin I; SpcDNA3-Gal-T1, sense [beta]4Gal-T1 construct; ST6Gal I, [alpha]2,6 sialyltransferase.
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 Dec 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
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 (11)
Request Permissions
Disclaimer
Google Scholar
Articles by Keusch, J.
Articles by Delves, P. J.
Search for Related Content
PubMed
PubMed Citation
Articles by Keusch, J.
Articles by Delves, P. J.
Social Bookmarking
What's this?