Glycobiology Advance Access originally published online on December 21, 2005
Glycobiology 2006 16(4):349-357; doi:10.1093/glycob/cwj071
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Immunoglobulin G specifically binding plant N-glycans with high affinity could be generated in rabbits but not in mice
2 Department of Chemistry and 3 Department of Biotechnology, University of Natural Resources and Applied Life Sciences (BOKU), 1190 Vienna, Austria; 4 Institute of Pathophysiology and 5 Clinical Institute of Pathology, Medical University Vienna, 1090 Vienna, Austria
1 To whom correspondence should be addressed; e-mail: friedrich.altmann{at}boku.ac.at
Received on November 8, 2005; revised on December 13, 2005; accepted on December 16, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
Xylosylated and core
1,3-fucosylated N-glycans from plants are immunogenic, and they play a still obscure role in allergy and in the field of plant-made protein pharmaceuticals. We immunized mice to generate monoclonal antibodies (mAbs) binding plant N-glycans specifically via the epitope containing either the xylose or the core 1,3-fucose residue. Splenocytes expressing N-glycan-specific antibodies derived from C57BL/6 mice previously immunized with plant glycoproteins were preselected by cell sorting to generate hybridoma lines producing specific antibodies. However, we obtained only mAbs unable to distinguish fucosylated from xylosylated N-glycans and reactive even with the pentasaccharide core Man3GlcNAc2. In contrast, immunization of rabbits yielded polyclonal sera selectively reactive with either fucosylated or xylosylated N-glycans. Purification of these sera using glyco-modified neoglycoproteins coupled to a chromatography matrix provided polyclonal sera suitable for affinity determination. Surface plasmon resonance measurements using sensor chips with immobilized glyco-modified transferrins revealed dissociation constants of around 10–9 M. This unexpectedly high affinity of IgG antibodies toward carbohydrate epitopes has repercussions on our conception of the binding strength and significance of antiglycan IgE antibodies in allergy. Key words: binding affinity / carbohydrate epitope / glycoprotein / immunogenic N-glycan
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
In complex-type N-glycans from plants, an
1,3-fucose is attached to the proximal glucosamine residue and a ß1,2-xylose residue to the ß-mannose. Being absent in mammals, these two residues are responsible for the fact that the glycan moiety of plant glycoproteins by itself constitutes an immunogenic determinant in mammals (Faye and Chrispeels, 1988
; Lauriere et al., 1988; Prenner et al., 1992; Wilson et al., 1998; van Ree et al., 2000). Sera from around 20–30% of allergic patients contain IgE antibodies against plant N-glycans which cause an enormous in vitro cross-reactivity of such sera (Aalberse et al., 1981; Tretter et al., 1993; Wilson et al., 1998; Bencurova et al., 2004). The diagnostic and clinical consequences of the cross-reactivity of antiglycan IgE for allergic patients have been exposed recently (Foetisch and Vieths, 2001; van Ree, 2004; Kochuyt et al., 2005; Malandain, 2005) and will be debated to some degree in the discussion. Among nonallergic blood donors, an even higher percentage was found to have IgG and IgM directed against xylosylated and 1,3-fucosylated N-glycans (Bardor et al., 2003).
Plants as production systems for recombinant biopharmaceuticals, such as antibodies, have received considerable attention during the last years (Raskin et al., 2002; Fischer et al., 2003; Yano et al., 2004). However, glycoproteins will be furnished with complex-type N-glycans containing the said carbohydrate epitopes, unless special measures are undertaken to prevent a plant-like glycosylation (Koprivova et al., 2004; Strasser et al., 2004; Faye et al., 2005). The presence of this immunogenic determinant on a glycoprotein therapeutic appears problematic if not intolerable because it will result in a reduced circulatory half-life and hence reduced efficacy of the drug. Antiglycan antibodies which were either preexisting or elicited by repeated administration of the drug may even lead to more severe adverse effects (Faye et al., 2005). At the extreme, allergic individuals with IgE antibodies against plant-type N-glycans may experience anaphylactic reactions. It is at the moment difficult to quantitate these hazards arising from a plant-type glycosylation, as for example the biological significance of IgE to plant N-glycans is questionable (van der Veen et al., 1997). Nevertheless, it is obvious that this issue deserves attention and considerable additional research before plant-made glycoprotein may become acceptable as topic pharmaceuticals for humans.
Following the discovery of the immunogenicity of plant N-glycans in rabbits and goats, it was soon discovered that the core 1,3-fucose on one hand and the ß1,2-linked xylose on the other hand together with the lack of antennae (sugar chains on the nonreducing end) constituted the key features which made plant N-glycans immunogenic (Kaladas et al., 1983; Lauriere et al., 1988; McManus et al., 1988; Kurosaka et al., 1991; Prenner et al., 1992). The importance of the fucose residue became evident from studies on bee venom-allergic patients (Weber et al., 1987; Tretter et al., 1993; Hemmer et al., 2001). The separation of rabbit anti-plant N-glycan sera into a "fucose-" and a "xylose-specific" fraction using immobilized honeybee phospholipase A2 indicated that these oligosaccharides contained two independent immune determinants (Faye et al., 1993), and this result was later corroborated by a study on commercial anti-horseradish peroxidase serum from rabbit with defined synthetic glyco-antigens containing either xylose or fucose (Bencurova et al., 2004). As expected, rabbit anti-bee venom serum exclusively bound to fucosylated glycans (Bencurova et al., 2004). It is important to notice that the terms fucose-specific or xylose-specific are meant in the sense that these residues constitute an essential part of the epitope but together with other groups which, however, by themselves are not able to bind antibody (Kurosaka et al., 1991).
Monoclonal antibodies (mAbs) of similar specificity have not been described so far. A mouse mAb raised against pollen extract from Cupressus arizonica was demonstrated to be directed against the plant N-glycan Man3XylFucGlcNAc2 (MMXF3) (for structures see Figure 1) (Iacovacci et al., 2001); however, no detailed analysis about its sugar specificity had been performed.
|
A rat mAb (YZ1/2.23) raised against elderberry abscission tissues was also found to be specific for plant MMXF3 structure (McManus et al., 1988), but further investigation showed that the paratope of YZ1/2.23 covered both the fucosyl and the xylosyl residues and that it bound to some degree even with the pentasaccharide core alone. As a consequence, YZ1/2.23 appears as an unsuitable tool for the detection of plant-specific glycosylation, and it certainly cannot be used to discriminate between fucosylated and xylosylated glycoproteins (Bencurova et al., 2004). A likewise very broad reactivity was observed for an mAb obtained from Schistosome-infected mice (van Remoortere et al., 2003). Recently, C57BL/6 mice were reported to give a much stronger immune reaction toward plant glycoproteins than the more widely used BALB/c mice (Bardor et al., 2003). This inability of BALB/c mice may also explain that the first and best-characterized monoclonal antiglycan antibody (YZ1/2.23) was from rat.
In none of the abovementioned studies, the affinity of binding between antibodies and glycoproteins has been determined. This is regrettable, as biological consequences of antibody reaction may depend on binding affinity at least as much as on antibody titer. Interactions between carbohydrates and proteins, for example lectins or antibodies, are usually considered to be much weaker (in the micromolar rang) than protein–protein interactions (usually in the nanomolar range) (Wilson and Stanfield, 1995; Weis, 1997; Thomas et al., 2002).
Binding affinities can be measured with various methods, the most attractive probably being surface plasmon resonance (SPR) (Haseley et al., 1999). SPR and other techniques require the use of pure, well-defined carbohydrate structures. However, plant glycoproteins with uniform and defined glycosylation are hardly available, especially, if one aims at evaluating the relative contributions of fucose and xylose or other structural features to antibody binding. The previous success in glyco-modifying human transferrin with the help of recombinant xylosyl and fucosyltransferase opened the opportunity to study the binding kinetics of antiglycan antibodies to structurally defined glycoforms (Bencurova et al., 2004).
In this article, we report on the attempt to produce mAbs against plant N-glycans in C57BL/6 mice and on the more successful attempt to produce structural specific polyclonal sera in rabbits. Last, but not least, the binding affinities of purified rabbit antibodies against MMX, MMF, and also MMXF glycans (see Figure 1 for structures) to their ligands were investigated by SPR technology, taking advantage of the availability of defined "plantified" glycoforms of human transferrin.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
Mouse mAbs
After repeated failure to generate hybridoma lines which produce mAbs against plant N-glycans (Tretter V and Altmann F, unpublished data), a recent report on the much higher immunogenicity of plant glycoproteins in C57BL/6 as compared with BALB/c mice (Bardor et al., 2003
) made us plan a new trial to obtain structure-specific antibodies against plant complex-type N-glycans. We wanted to specifically boost the glycan-directed immune response by challenging the mice at consecutive immunizations with different proteins bearing similar N-glycan structures. Even then, however, the immune response of the mice was far from overwhelming (data not shown). Nevertheless, spleen cells of a few mice were fused, and by applying a sophisticated sorting strategy (Böhm et al., 2005), we got 164 clones, 36 of which secreted glycan-directed antibody as measured using MMX and MMF coupled to human transferrin (MMX- and MMF-Tf). However, a closer characterization of these mAbs revealed that the individual mAbs could bind both to MMX- and to MMF-Tf (Figure 2). Hence, they did not discriminate between fucosylated and xylosylated N-glycans. Moreover, the mAbs even exhibited high binding to MM-Tf and unmodified transferrin (Figure 2). Thus, despite great efforts and the use of C57BL/6 mice, we were not able to elicit a strong immune response against the plant N-glycans and to generate hybridoma lines producing structure-specific mAbs.
|
Rabbit polyclonal antisera
To overcome the problems encountered with mice, rabbits were chosen for the production of antiglycan sera. Specificity should be introduced by the immunization protocol using only MMF-Tf as antigen to obtain "fucose-specific" sera or MMX-Tf for "xylose-specific" sera. All rabbits showed immune response to the antigens, even though only one rabbit from each group of three developed a satisfyingly high titer, which may be attributed to the use of incomplete adjuvant. Notably, the same result was obtained for horseradish peroxidase (HRP), a usually very potent glyco-antigen as we judge from our long experience with commercial rabbit and goat anti-HRP serum (Tretter et al., 1993; Wilson et al., 1998) as well as for both transferrin glycoforms. Before immunization, no detectable binding activity to HRP, MMX- and MMF-Tf was found in control rabbits (not shown). Sera from rabbits which showed highest binding activity to xylose or fucose residues were used for a two-step affinity purification. To detect residual anti-transferrin antibodies in the purified Xyl- and Fuc-specific sera, they were tested by western blot with native and with glyco-modified transferrins (Figure 3). Obviously, the antibodies directed against the protein backbone of transferrin had been efficiently removed by affinity chromatography. After separation from (most of) the albumin, the three antiglycan antibodies were characterized with regard to their glycan binding using various transferrin glycoforms (Figure 1). In contrast to the mouse mAbs, the rabbit sera, even though polyclonal, exhibited high specificity and were able to discriminate between fucosylated and xylosylated glycoproteins. Remarkably, the anti-MMX serum bound xylosylated glycoforms irrespective of the presence or absence of nonreducing terminal GlcNAc residues. Anti-HRP also had this ability but to a lesser extent. In contrast, anti-MMF antibody only bound truncated fucosylated glycoforms (MMF, MUXF, and MMXF). The same results were obtained when the specificities of purified antiglycan antibodies were investigated by ELISA using glyco-modified transferrins as well as bovine serum albumin (BSA) conjugates (Figure 4) or glyco-modified human 1-antitrypsins (data not shown). The binding to glyco-modified transferrins and to BSA conjugates with the same glycans was comparable, which argues against any essential influence of the protein backbone. Finally, the reaction of the purified antibodies with natural glycoproteins HRP (MMXF), ascorbate oxidase (MMXF and MMX; Altmann, 1998), hemocyanin from Helix pomatia (MMX and many undefined structures; van Kuik et al., 1985), honeybee venom phospholipase A2 (PLA, an insect glycoprotein containing a core (1,3)-fucose; Kubelka et al., 1993), and lectin from Erythrina corallodendron (MMXF and other structures; Ashford et al., 1991) was determined (Figure 5). Consistent with the structures of their carbohydrate chains, plant glycoproteins bound both the anti-MMF and the anti-MMX serum, whereas the insect and the snail glycoprotein selectively reacted with either anti-MMF or anti-MMX, respectively.
|
|
|
Goat anti-HRP serum
Commercial goat anti-HRP serum was subjected to ELISA with glyco-modified transferrins. This serum exhibited a similar specificity to that of rabbit anti-HRP, that is it bound strongly to all fucosylated and/or xylosylated transferrins but almost not at all when these residues were absent (data not shown). The same results were obtained with BSA conjugates and glyco-modified human 1-antitrypsins (data not shown).
Binding affinities of rabbit antibodies
The problems associated with the determination of binding affinities of polyclonal antibodies have recently been discussed, and a strategy to obtain representative average values has been devised (Bakker et al., 1995; Sem et al., 1999; Cachia et al., 2004; Hantusch et al., 2005). The situation in the present study is somewhat simpler as in the case of anti-MMF and anti-MMX as we are dealing with sera directed against a single epitope, whereas in natural allergens an unknown number of different epitopes is present.
The on- and off-rates of polyclonal xylose- and fucose-specific IgG and anti-HRP binding to MMX-, MMF-, and MMXF-Tf were determined in SPR experiments. In all instances, the reference cell of chip was coated with unmodified transferrin to subtract any trace unspecific binding to the transferrin backbone. Decreasing amounts of plant N-glycan-specific IgG and anti-HRP were injected in triplicates onto the sensor chip (Figure 6). Results of the association rate constant, ka (1/Ms), dissociation rate constant, kd (1/s), and the calculated equilibrium dissociation constant, Kd (M) are summarized in Table I. The values of ka were determined from the association phase of the binding curve, whereas kd were determined from the dissociation phase. The larger the association rate, the faster is the binding and in combination with a low dissociation rate, the higher is the affinity of an antibody. Notably, the binding of the xylose-specific antibody was faster to the MMXF surface than to the specific surface with MMX glycans (Table I). As the dissociation rates from the MMX and the MMXF surface were similar, this difference of the association rate caused the combination anti-MMX with MMXF-Tf to exhibit the highest binding affinities of 2.5 x 10–10 M (Table I). Possibly, the concomitant presence of xylose and fucose on the glycans favors a conformation that facilitates antibody binding to the epitope containing the rather flexible xylose residue (Lommerse et al., 1995). The affinity of anti-HRP may be seen as sort of average value between those of anti-MMX and anti-MMF as anti-HRP actually consists of two antibody populations, one binding to xylosylated and one to fucosylated glycans.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
From this study as well as from previous experiences in our laboratory and from other groups (van Remoortere et al., 2003
), it appears that the immune system of mice and also of rats (Wilson et al., 1998; Bencurova et al., 2004) cannot respond to plant N-glycans with the production of structurally specific antibodies as is seen with rabbits or goats. It also seems from this study and a previous one (Bardor et al., 2003) that depending upon strain and maybe immunization regime, glycoproteins, which in other animals lead to a strong anti-carbohydrate immune response (Kaladas et al., 1983; Faye and Chrispeels, 1988; Lauriere et al., 1988; Prenner et al., 1992; Faye et al., 1993), do not necessarily lead to a nearly comparable antibody production in mice. In this light, the recent finding that a tobacco-produced mAb does not elicit an immune response in mice (Chargelegue et al., 2000) cannot be taken as a last affirmation that plant-made glycoprotein pharmaceuticals may be regarded as immunologically safe for human therapy. In other words, mice must be rated an inappropriate animal model for the evaluation of the carbohydrate-related immunogenicity of plant-derived glycoproteins.
The immunological distinctiveness of mice unfortunately deprives us of the possibility of obtaining highly specific mAbs. In this study, we have therefore chosen affinity-purified polyclonal rabbit antibodies. Other options may be the use of phage display libraries, which, however, disrupts the pairing of heavy and light chain unless camelid antibodies are considered (Nguyen et al., 2003). However, rabbits, when immunized with defined glyco-antigens such as our glyco-modified transferrins, respond in a very specific manner. The quality of the sera can be enhanced by removal of the inadvertently generated anti-protein antibodies by immobilized transferrin and/or by using immobilized glyco-antigen for affinity purification.
Another small but notable point can be made from the ELISA comparison of antibodies raised against HRP or modified transferrins and tested against a panel of glycoproteins. The protein backbone of the glyco-allergen on the ELISA does not affect the result. This argues very much against the sometimes heard notion that the sugars only modulate the binding of antibodies to nearby peptide epitopes.
The monovalent, albeit polyclonal, anti-MMF and anti-MMX antibodies were finally used to assess their binding affinities to glycoproteins. The motive for this effort lays in the enigmatic role of carbohydrate determinants in allergy. While on the one hand plant N-glycans have been shown to elicit effector release (Bublin et al., 2003; Foetisch et al., 2003; Westphal et al., 2003; Wicklein et al., 2004), they are considered as innocuous regarding their potency to evoke clinical symptoms (van der Veen et al., 1997). Does the carbohydrate nature of this determinant explain this observation? For immunologists, all epitopes are equal. But for allergologists, some epitopes—the peptide epitopes—appear to be more equal. A possible explanation may lay in a low binding affinity of antibodies to these epitopes as it was found for lectin–carbohydrate interactions, where the Kd often lies in the micromolar range (Haseley et al., 1999). However, even for lectins, higher affinities have been reported for RCA120 and DSA to terminal Gal residue in asialofetuin (Kd, 3.8 x 10–9 and 3.7 x 10–9 M, respectively) (Shinohara et al., 1994). The affinities of antibodies against carbohydrate epitopes were often found to be quite high, for example in the case of a 3-deoxy-D-manno-octulosonic acid trisaccharide on Chlamydia outer membrane LPS, high-affinity mAbs could be obtained (Kd, 
10–9 M) (Muller-Loennies et al., 2000). Examples of lower affinity can be found as well, e.g. a Kd of 2.3 x 10–7 M for an mAb against O-polysaccharide from Salmonella serogroup B because of its rapid dissociation rates (MacKenzie et al., 1996).
The antibodies investigated in this study exhibited Kd values of around 10–9 M and thereby almost reach the theoretical limit of antibody affinity (10–10 M) (Foote and Eisen, 1995). Therefore, cross-reactive carbohydrate determinants (CCDs) can no longer be regarded as "less equal" in the sense of allowing only low-affinity binding. Even though the affinities of human IgE and IgG remain to be determined, we dare to speculate that low binding affinity of anti-CCD antibody drops out as an explanation for the usually observed inoffensiveness of CCDs as allergens. However, this should not remain a speculation for long as the current work forms the basis for the determination of binding affinities of anti-CCD antibodies in human sera.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
Preparation of neoglycoproteins
BSA conjugates were prepared as previously described with some modifications (Wilson et al., 1998
). GnGn fibrin glycopeptides were first cross-linked to BSA to prepare GnGn-BSA. Then GnGn-BSA was modified by purified recombinant Arabidopsis thaliana core 1,3-fucosyltransferase and rice ß1,2-xylosyltransferase expressed in Pichia pastoris (Bencurova et al. 2003; Leonard et al., 2004). The resultant GnGnX-, GnGnF-, and GnGnXF-BSA were treated with N-acetylglucosaminidase from Streptococcus pneumoniae (Sigma-Aldrich, St Louis, MO, USA) to obtain MMX-, MMF-, and MMXF-BSA. MUXF-BSA and MUX-BSA were generated by cross-linking bromelain glycopeptides or defucosylated glycopeptides to BSA (Wilson et al., 1998). The BSA conjugates were subjected to amino sugar analysis (Altmann, 1992) and found to contain 2–3 mol glycopeptide per mol of BSA.
Glyco-modified human transferrins (MM-, MMX-, MUX-, MMF-, and MMXF-Tf) were prepared as previously described except using purified recombinant ß1,2-xylosyltransferase from rice instead from A. thaliana (Bencurova et al. 2003; Leonard et al., 2004). The purity and integrity of neoglycoproteins were verified by SDS-PAGE, and the structures of their glycan moieties were checked by matrix-assisted laser-desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) as described previously (Kolarich and Altmann, 2000). It should be noted that the glycosyltransferases and glycosidases used were not of plant origin, and thus glyco-contamination from this source were excluded.
Generation and characterization of mouse mAbs
Nine eight-week-old C57BL/6 mice in three groups (Charles River Wiga, Sulzfeld, Germany) were randomly divided into three groups immunized with different combinations of natural or modified plant glycoproteins, mixed with incomplete Freund’s adjuvant. The dosage per injection of the glycoproteins was as follows: 100 µg of HRP (Sigma-Aldrich); 30 µg of honeybee venom phospholipase (PLA, prepared as described; Tretter et al., 1993); 50 µg of zucchini ascorbate oxidase (AO, Sigma-Aldrich); 100 µg of Sophora japonica lectin (SJA, Sigma-Aldrich); and 50 µg of bromelain (BRO, prepared as described; Tretter et al., 1993). The first three injections were given intraperitoneally on days 0, 21, 42, 63, and 84. The two final injections were given intravenously without adjuvant and were followed 4 days later by cell fusion.
The sequences of antigens were HRP-PLA-AO-HRP-SJL, HRP-BRO-AO-BRO-PLA, and HRP-AO-BRO-SJL-AO for the first, second, and third group of mice, respectively. The spleen cells from the three mice, which showed highest titers to glyco-modified transferrin (MMX- and MMF-Tf), were fused with an NS-0 myeloma cell line according to ClonaCell-HY for hybridoma production’s protocol (Stemcell Technologies, Vancouver, British Columbia, Canada).
After 10 days of hypoxanthine-aminopterin-thymidine (HAT) selection, cells were stained with the affinity matrix secretion assay (Böhm et al., 2005) using a sheep-anti mouse IgG–biotin conjugate (Sigma-Aldrich) as capture antibody. Staining of specific hybridoma cells was achieved with the respective glyco-modified transferrin and an FITC- conjugated anti-transferrin (Capricorn Products, Portland, ME, USA). Cells secreting transferrin-specific antibodies were then sorted into 96-well plates at two to five cells per well. After approximately 2 weeks, samples of the culture supernatants of wells containing monoclonal colonies were taken and were screened by qualitative ELISA using MM-, MMX-, MMF-, MMXF-Tf, or NaNa-Tf (= unmodified transferrin) similarly as recently described (Bencurova et al., 2004). However, 1:1000 diluted alkaline phosphatase-conjugated anti-mouse IgG 
-chain (Sigma-Aldrich) was used as the detection antibody. Bound antibody was detected by reaction with p-nitrophenylphosphate in 0.1 M diethanolamine at pH 9.8. The reaction was stopped after 20 min by addition of 5 M sodium hydroxide, and the absorbance at 405 nm was read.
Immunization of rabbits
Chinchilla Bastard rabbits were also divided into groups of three, but the animals in each group repeatedly received the same antigen, either HRP or MMX-Tf or MMF-Tf. All antigens were dissolved in phosphate-buffered saline (PBS) at a concentration of 1 mg/mL. Intradermal injections consisting of 200 µL antigen solution and the same volume of Freund’s incomplete adjuvant were administered in 3-week intervals. After the fourth immunization, the antisera were collected and kept at –20°C.
For ELISAs with rabbit sera, the plates were coated with plant glycoproteins or neoglycoprotein at a concentration of 5 µg/mL in 50 mM sodium carbonate, pH 9.6, for 60 min at 37°C. Bound antibody was detected by a 1:2000 dilution of alkaline phosphatase-conjugated anti-rabbit IgG (Sigma-Aldrich) as described (Wilson et al., 1998).
Affinity purification of xylose or fucose-specific IgG
From each group of three rabbits, the one serum which showed the highest antiglycan antibody titer was used for affinity purification. Affinity columns were prepared by coupling unmodified human transferrin, MMX-BSA or MMF-BSA conjugates to Affi-Gel 15 (BioRad, Richmond, CA) according to the manufacturer’s instructions. The sera were diluted with 0.1 M Tris-HCl, pH 7.2, containing 0.1 M NaCl, 2.5 mM MgCl2, and 0.05% Tween 20 (TTBS) and applied at first to the transferrin column to eliminate anti-transferrin antibodies. The breakthrough was applied to either MMX-BSA- or MMF-BSA-coupled column depending on the antiserum. After extensive washing with TTBS, bound antibodies were eluted with 0.2 M glycine–HCl, pH 2.5, and neutralized immediately with 1 M Tris–HCl, pH 7.5. To remove rabbit serum albumin, effluents were loaded on a 1 mL bed of SP Sepharose Fast Flow (Amersham Biosciences AB, Uppsala, Sweden). Elution of IgG was performed with a gradient from 0.02 to 1.0 M NaCl in 50 mM MES, pH 5.0. Buffer exchange as well as concentration was performed immediately by filter devices with 10 kDa cut-off (Millipore, MA, USA).
The specificities of xylose or fucose-specific IgG were determined by ELISA and western blot. The amount of IgG was quantified by comparison of Coomassie-stained SDS-PAGE bands with a dilution series of pure human IgG. The xylose-specific IgG had an estimated concentration of 0.1 mg/mL, the fucose-specific pool of 0.2 mg/mL.
For western blot analysis, transferrin glycoforms and BSA conjugates (0.5–1 µg) were separated by SDS-PAGE, electroblotted to a nitrocellulose membrane and stained with the help of rabbit anti-HRP (1:2000) or purified rabbit xylose-specific or fucose-specific antibody (1:500) as described (Bencurova et al., 2004).
SPR analysis
The binding kinetics of the purified antiglycan antibodies were determined by SPR using a BIAcore X biosensor system (Biacore, Uppsala, Sweden). The running buffer used for all experiments was HBS-P buffer (Biacore), pH 7.4, containing 10 mM HEPES, 150 mM NaCl, 3 mM EDTA, and 0.005% surfactant P20 (Biacore). The immobilization to research grade CM5 sensor chips was carried out using a standard amine coupling procedure (Johnsson et al., 1991). Transferrin glycoforms were applied in 10 mM sodium acetate buffer, pH 4.0. The two channels of each chip were set up as follows: MMX-, MMF-, and MMXF-Tf were immobilized on channel 1, whereas unmodified transferrin was bound to channel 2 of each chip as a control surface. Amounts of 3000 ± 1000 RU of each glycoform were immobilized onto the chip surfaces. One RU corresponds to an immobilized protein concentration of about 1 pg/mm2 (Stenberg et al., 1991).
All measurements were carried out in HBS-P buffer. Analyses were performed at 25°C and at a flow rate of 5 µL/min. Between cycles, the chip surface was regenerated with two injections of 10 µL of 10 mM glycine–HCl, pH 2.2, or 0.5% SDS at a flow rate of 5 µL/min followed by washing the surface with HBS-P buffer until the signal reached baseline.
For the evaluation of association and dissociation rate constants, samples were diluted in HBS-P buffer and analyzed at various concentrations. Purified antibody at the specified concentration was pumped over the sensor chip surface for 6 min at a flow rate of 5 µL/min to observe the association phase, and dissociation phase was monitored after switching to protein-free buffer for another 6 min.
Sensorgrams were fitted to a single site-binding model (1:1 Langmuir binding), using the numerical integration functions of the BIAevaluation 3.1 software package. Association and dissociation rate constants (ka and kd) and the dissociation constant (Kd) were derived by both separate and simultaneous fitting of the association and dissociation phase of individual sensorgrams. Sensorgrams were also fitted globally using data collected at a series of concentrations. The steady-state binding values, Req, were then plotted as a function of the applied antibody concentrations and fitted to first-order kinetics, assuming a monovalent antibody–modified transferrin interaction (Karlsson et al., 1991).
|
Supplementary data |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
Supplementary data are available at Glycobiology online (http://glycob.oxfordjournals.org).
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
This work was supported by project S8803 and, in the case of B.H., by project F1808-B04 from the Austrian Science Fund.
|
Footnotes |
|---|
* These authors contributed equally to this work.
|
Abbreviations |
|---|
BSA, bovine serum albumin; CCD, cross-reactive carbohydrate determinant; HAT, hypoxanthine-aminopterin- thymidine; HRP, horseradish peroxidase; SPR, surface plasmon resonance
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and MethodsSupplementary dataAcknowledgmentsReferences |
|---|
Aalberse, R.C., Koshte, V., and Clemens, J.G. (1981) Immunoglobulin E antibodies that crossreact with vegetable foods, pollen, and Hymenoptera venom. J. Allergy Clin. Immunol., 68, 356–364.[CrossRef][Web of Science][Medline]
Altmann, F. (1992) Determination of amino sugars and amino acids in glycoconjugates using precolumn derivatization with o-phthalaldehyde. Anal. Biochem., 204, 215–219.[CrossRef][Web of Science][Medline]
Altmann, F. (1998) Structures of the N-linked carbohydrate of ascorbic acid oxidase from zucchini. Glycoconj. J., 15, 79–82.[CrossRef][Web of Science][Medline]
Ashford, D.A., Dwek, R.A., Rademacher, T.W., Lis, H., and Sharon, N. (1991) The glycosylation of glycoprotein lectins. Intra- and inter-genus variation in N-linked oligosaccharide expression. Carbohydr. Res., 213, 215–227.[CrossRef][Web of Science][Medline]
Bakker, R., Lasonder, E., and Bos, N.A. (1995) Measurement of affinity in serum samples of antigen-free, germ-free and conventional mice after hyperimmunization with, 2,4-dinitrophenyl keyhole limpet hemocyanin, using surface plasmon resonance. Eur. J. Immunol., 25, 1680–1686.[Web of Science][Medline]
Bardor, M., Faveeuw, C., Fitchette, A.C., Gilbert, D., Galas, L., Trottein, F., Faye, L., and Lerouge, P. (2003) Immunoreactivity in mammals of two typical plant glyco-epitopes, core alpha(1,3)-fucose and core xylose. Glycobiology, 13, 427–434.
Bencurova, M., Hemmer, W., Focke-Tejkl, M., Wilson, I.B., and Altmann, F. (2004) Specificity of IgG and IgE antibodies against plant and insect glycoprotein glycans determined with artificial glycoforms of human transferrin. Glycobiology, 14, 457–466.
Bencurova, M., Rendic, D., Fabini, G., Kopecky, E.M., Altmann, F., and Wilson, I.B. (2003) Expression of eukaryotic glycosyltransferases in the yeast Pichia pastoris. Biochimie, 85, 413–422.[Medline]
Böhm, E., Grillari, J., Voglauer, R., Gross, S., Ernst, W., Ferko, B., Kunert, R., Katinger, H., and Borth, N. (2005) Establishment of a strategy for the rapid generation of a monoclonal antibody against the human protein SNEV (hNMP200) by flow cytometric cell sorting. J. Immunol. Methods, 307, 13–23.[CrossRef][Web of Science][Medline]
Bublin, M., Radauer, C., Wilson, I.B., Kraft, D., Scheiner, O., Breiteneder, H., and Hoffmann-Sommergruber, K. (2003) Cross-reactive N-glycans of Api g 5, a high molecular weight glycoprotein allergen from celery, are required for immunoglobulin E binding and activation of effector cells from allergic patients. FASEB J., 17, 1697–1699.
Cachia, P.J., Kao, D.J., and Hodges, R.S. (2004) Synthetic peptide vaccine development: measurement of polyclonal antibody affinity and cross-reactivity using a new peptide capture and release system for surface plasmon resonance spectroscopy. J. Mol. Recognit., 17, 540–557.[CrossRef][Web of Science][Medline]
Chargelegue, D., Vine, N.D., van Dolleweerd, C.J., Drake, P.M., and Ma, J.K. (2000) A murine monoclonal antibody produced in transgenic plants with plant-specific glycans is not immunogenic in mice. Transgenic Res., 9, 187–194.[CrossRef][Web of Science][Medline]
Faye, L., Boulaflous, A., Benchabane, M., Gomord, V., and Michaud, D. (2005) Protein modifications in the plant secretory pathway: current status and practical implications in molecular pharming. Vaccine, 23, 1770–1778.[CrossRef][Web of Science][Medline]
Faye, L. and Chrispeels, M.J. (1988) Common antigenic determinants in the glycoproteins of plants, molluscs and insects. Glycoconj. J., 5, 245–256.
Faye, L., Gomord, V., Fitchette-Laine, A.C., and Chrispeels, M.J. (1993) Affinity purification of antibodies specific for Asn-linked glycans containing alpha, 1
3 fucose or beta 12 xylose. Anal. Biochem., 209, 104–108.[CrossRef][Web of Science][Medline]
Fischer, R., Twyman, R.M., and Schillberg, S. (2003) Production of antibodies in plants and their use for global health. Vaccine, 21, 820–825.[CrossRef][Web of Science][Medline]
Foetisch, K. and Vieths, S. (2001) N- and O-linked oligosaccharides of allergenic glycoproteins. Glycoconj. J., 18, 373–390.[CrossRef][Web of Science][Medline]
Foetisch, K., Westphal, S., Lauer, I., Retzek, M., Altmann, F., Kolarich, D., Scheurer, S., and Vieths, S. (2003) Biological activity of IgE specific for cross-reactive carbohydrate determinants. J. Allergy Clin. Immunol., 111, 889–896.[CrossRef][Web of Science][Medline]
Foote, J. and Eisen, H.N. (1995) Kinetic and affinity limits on antibodies produced during immune responses. Proc. Natl. Acad. Sci. U. S. A., 92, 1254–1256.
Hantusch, B., Scholl, I., Harwanegg, C., Krieger, S., Becker, W.M., Spitzauer, S., Boltz-Nitulescu, G., and Jensen-Jarolim, E. (2005) Affinity determinations of purified IgE and IgG antibodies against the major pollen allergens Phl, p 5a and Bet v 1a: discrepancy between IgE and IgG binding strength. Immunol. Lett., 97, 81–89.[CrossRef][Web of Science][Medline]
Haseley, S.R., Talaga, P., Kamerling, J.P., and Vliegenthart, J.F. (1999) Characterization of the carbohydrate binding specificity and kinetic parameters of lectins by using surface plasmon resonance. Anal. Biochem., 274, 203–210.[CrossRef][Web of Science][Medline]
Hemmer, W., Focke, M., Kolarich, D., Wilson, I.B., Altmann, F., Wohrl, S., Gotz, M., and Jarisch, R. (2001) Antibody binding to venom carbohydrates is a frequent cause for double positivity to honeybee and yellow jacket venom in patients with stinging-insect allergy. J. Allergy Clin. Immunol., 108, 1045–1052.[CrossRef][Web of Science][Medline]
Iacovacci, P., Pini, C., Afferni, C., Barletta, B., Tinghino, R., Schinina, E., Federico, R., Mari, A., and Di Felice, G. (2001) A monoclonal antibody specific for a carbohydrate epitope recognizes an IgE-binding determinant shared by taxonomically unrelated allergenic pollens. Clin. Exp. Allergy, 31, 458–465.[CrossRef][Web of Science][Medline]
Johnsson, B., Lofas, S., and Lindquist, G. (1991) Immobilization of proteins to a carboxymethyldextran-modified gold surface for biospecific interaction analysis in surface plasmon resonance sensors. Anal. Biochem., 198, 268–277.[CrossRef][Web of Science][Medline]
Kaladas, P.M., Goldberg, R., and Poretz, R.D. (1983) Rabbit anti-carbohydrate antibody elicited by the lymphocyte mitogenic glycoprotein from Wistaria floribunda seeds. Mol. Immunol., 20, 727–735.[CrossRef][Web of Science][Medline]
Karlsson, R., Michaelsson, A., and Mattsson, L. (1991) Kinetic analysis of monoclonal antibody–antigen interactions with a new biosensor based analytical system. J. Immunol. Methods, 145, 229–240.[CrossRef][Web of Science][Medline]
Kochuyt, A.M., Van Hoeyveld, E.M., and Stevens, E.A. (2005) Prevalence and clinical relevance of specific immunoglobulin E to pollen caused by sting-induced specific immunoglobulin E to cross-reacting carbohydrate determinants in Hymenoptera venoms. Clin. Exp. Allergy, 35, 441–447.[CrossRef][Web of Science][Medline]
Kolarich, D. and Altmann, F. (2000) N-Glycan analysis by matrix-assisted laser desorption/ionization mass spectrometry of electrophoretically separated nonmammalian proteins: application to peanut allergen Ara h 1 and olive pollen allergen Ole e 1. Anal. Biochem., 285, 64–75.[CrossRef][Web of Science][Medline]
Koprivova, A., Stemmer, C., Altmann, F., Hoffmann, A., Kopriva, S., Gorr, G., Reski, R., and Decker, E. (2004) Targeted knockouts of Physcomitrella lacking plant specific immunogenic N-glycans. Plant Biotechnol., 2, 517–523.[CrossRef]
Kubelka, V., Altmann, F., Staudacher, E., Tretter, V., Marz, L., Hard, K., Kamerling, J.P., and Vliegenthart, J.F. (1993) Primary structures of the N-linked carbohydrate chains from honeybee venom phospholipase A2. Eur. J. Biochem., 213, 1193–1204.[Web of Science][Medline]
Kurosaka, A., Yano, A., Itoh, N., Kuroda, Y., Nakagawa, T., and Kawasaki, T. (1991) The structure of a neural specific carbohydrate epitope of horseradish peroxidase recognized by anti-horseradish peroxidase antiserum. J. Biol. Chem., 266, 4168–4172.
Lauriere, C., Lauriere, M., Sturm, A., Faye, L., and Chrispeels, M.J. (1988) Characterization of beta-fructosidase, an extracellular glycoprotein of carrot cells. Biochimie, 70, 1483–1491.[Medline]
Leonard, R., Kolarich, D., Paschinger, K., Altmann, F., and Wilson, I.B. (2004) A genetic and structural analysis of the N-glycosylation capabilities. Plant Mol. Biol., 55, 631–644.[CrossRef][Web of Science][Medline]
Lommerse, J.P., Kroon-Batenburg, L.M., Kamerling, J.P., and Vliegenthart, J.F. (1995) Conformational analysis of the xylose-containing N-glycan of pineapple stem bromelain as part of the intact glycoprotein. Biochemistry, 34, 8196–8206.[CrossRef][Medline]
MacKenzie, C.R., Hirama, T., Deng, S.J., Bundle, D.R., Narang, S.A., and Young, N.M. (1996) Analysis by surface plasmon resonance of the influence of valence on the ligand binding affinity and kinetics of an anti-carbohydrate antibody. J. Biol. Chem., 271, 1527–1533.
Malandain, H. (2005) IgE-reactive carbohydrate epitopes – classification, cross-reactivity, and clinical impact. Allerg. Immunol., 37, 122–128.
McManus, M.T., McKeating, J., Secher, D.S., Osborne, D.J., Ashford, D., Dwek, R.A., and Rademacher, T.W. (1988) Identification of a monoclonal antibody to abscission tissue that recoginses xylose/fucose-containing N-linked oligosaccharides from higher plants. Planta, 175, 506–512.[CrossRef]
Muller-Loennies, S., MacKenzie, C.R., Patenaude, S.I., Evans, S.V., Kosma, P., Brade, H., Brade, L., and Narang, S. (2000) Characterization of high affinity monoclonal antibodies specific for chlamydial lipopolysaccharide. Glycobiology, 10, 121–130.
Nguyen, V.K., Zou, X., Lauwereys, M., Brys, L., Bruggemann, M., and Muyldermans, S. (2003) Heavy-chain only antibodies derived from dromedary are secreted and displayed by mouse B cells. Immunology, 109, 93–101.[CrossRef][Web of Science][Medline]
Prenner, C., Mach, L., Glossl, J., and Marz, L. (1992) The antigenicity of the carbohydrate moiety of an insect glycoprotein, honey-bee (Apis mellifera) venom phospholipase A2. The role of alpha, 1,3-fucosylation of the asparagine-bound N-acetylglucosamine. Biochem. J., 284(2), 377–380.
Raskin, I., Ribnicky, D.M., Komarnytsky, S., Ilic, N., Poulev, A., Borisjuk, N., Brinker, A., Moreno, D.A., Ripoll, C., Yakoby, N., et al. (2002) Plants and human health in the twenty-first century. Trends Biotechnol., 20, 522–531.[CrossRef][Web of Science][Medline]
Sem, D.S., McNeeley, P.A., and Linnik, M.D. (1999) Antibody affinities and relative titers in polyclonal populations: surface plasmon resonance analysis of anti-DNA antibodies. Arch. Biochem. Biophys., 372, 62–68.[CrossRef][Web of Science][Medline]
Shinohara, Y., Kim, F., Shimizu, M., Goto, M., Tosu, M., and Hasegawa, Y. (1994) Kinetic measurement of the interaction between an oligosaccharide and lectins by a biosensor based on surface plasmon resonance. Eur. J. Biochem., 223, 189–194.[Web of Science][Medline]
Stenberg, E., Persson, B., Roos, H., and Urbaniczky, C. (1991) Quantitative determination of surface concentration of protein with surface plasmon resonance by using radiolabeled proteins. J. Colloid Interface Sci., 143, 513–526.[CrossRef]
Strasser, R., Altmann, F., Mach, L., Glossl, J., and Steinkellner, H. (2004) Generation of Arabidopsis thaliana plants with complex N-glycans lacking beta1,2-linked xylose and core alpha1,3-linked fucose. FEBS Lett., 561, 132–136.[CrossRef][Web of Science][Medline]
Thomas, R., Patenaude, S.I., MacKenzie, C.R., To, R., Hirama, T., Young, N.M., and Evans, S.V. (2002) Structure of an anti-blood group A Fv and improvement of its binding affinity without loss of specificity. J. Biol. Chem., 277, 2059–2064.
Tretter, V., Altmann, F., Kubelka, V., Marz, L., and Becker, W.M. (1993) Fucose alpha, 1,3-linked to the core region of glycoprotein N-glycans creates an important epitope for IgE from honeybee venom allergic individuals. Int. Arch. Allergy Immunol., 102, 259–266.[Web of Science][Medline]
van der Veen, M.J., van Ree, R., Aalberse, R.C., Akkerdaas, J., Koppelman, S.J., Jansen, H.M., and van der Zee, J.S. (1997) Poor biologic activity of cross-reactive IgE directed to carbohydrate determinants of glycoproteins. J. Allergy Clin. Immunol., 100, 327–334.[CrossRef][Web of Science][Medline]
van Kuik, J.A., van Halbeek, H., Kamerling, J.P., and Vliegenthart, J.F. (1985) Primary structure of the low-molecular-weight carbohydrate chains of Helix pomatia alpha-hemocyanin. Xylose as a constituent of N-linked oligosaccharides in an animal glycoprotein. J. Biol. Chem., 260, 13984–13988.
van Ree, R. (2004) Clinical importance of cross-reactivity in food allergy. Curr. Opin. Allergy Clin. Immunol., 4, 235–240.[CrossRef][Medline]
van Ree, R., Cabanes-Macheteau, M., Akkerdaas, J., Milazzo, J.P., Loutelier-Bourhis, C., Rayon, C., Villalba, M., Koppelman, S., Aalberse, R., Rodriguez, R., et al. (2000) Beta(1,2)-xylose and alpha(1,3)–fucose residues have a strong contribution in IgE binding to plant glycoallergens. J. Biol. Chem., 275, 11451–11458.
van Remoortere, A., Bank, C.M., Nyame, A.K., Cummings, R.D., Deelder, A.M., and van Die, I. (2003) Schistosoma mansoni-infected mice produce antibodies that cross-react with plant, insect, and mammalian glycoproteins and recognize the truncated biantennaryN-glycan Man3GlcNAc2-R. Glycobiology, 13, 217–225.
Weber, A., Schroder, H., Thalberg, K., and Marz, L. (1987) Specific interaction of IgE antibodies with a carbohydrate epitope of honey bee venom phospholipase A2. Allergy, 42, 464–470.[Web of Science][Medline]
Weis, W.I. (1997) Cell-surface carbohydrate recognition by animal and viral lectins. Curr. Opin. Struct. Biol., 7, 624–630.[CrossRef][Web of Science][Medline]
Westphal, S., Kolarich, D., Foetisch, K., Lauer, I., Altmann, F., Conti, A., Crespo, J.F., Rodriguez, J., Enrique, E., Vieths, S. et al. (2003) Molecular characterization and allergenic activity of Lyc e 2 (beta-fructofuranosidase), a glycosylated allergen of tomato. Eur. J. Biochem., 270, 1327–1337.[Web of Science][Medline]
Wicklein, D., Lindner, B., Moll, H., Kolarich, D., Altmann, F., Becker, W.M., and Petersen, A. (2004) Carbohydrate moieties can induce mediator release: a detailed characterization of two major timothy grass pollen allergens. Biol. Chem., 385, 397–407.[CrossRef][Web of Science][Medline]
Wilson, I.B., Harthill, J.E., Mullin, N.P., Ashford, D.A., and Altmann, F. (1998) Core alpha1,3-fucose is a key part of the epitope recognized by antibodies reacting against plant N-linked oligosaccharides and is present in a wide variety of plant extracts. Glycobiology, 8, 651–661.
Wilson, I.A., and Stanfield, R.L. (1995) A Trojan horse with a sweet tooth. Nat. Struct. Biol., 2, 433–436.[CrossRef][Web of Science][Medline]
Yano, A., Maeda, F., and Takekoshi, M. (2004) Transgenic tobacco cells producing the human monoclonal antibody to hepatitis B virus surface antigen. J. Med. Virol., 73, 208–215.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Jin, F. Altmann, R. Strasser, L. Mach, M. Schahs, R. Kunert, T. Rademacher, J. Glossl, and H. Steinkellner A plant-derived human monoclonal antibody induces an anti-carbohydrate immune response in rabbits Glycobiology, March 1, 2008; 18(3): 235 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Poltl, O. Ahrazem, K. Paschinger, M. D. Ibanez, G. Salcedo, and I. B.H. Wilson Molecular and immunological characterization of the glycosylated orange allergen Cit s 1 Glycobiology, February 1, 2007; 17(2): 220 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bakker, G. J. A. Rouwendal, A. S. Karnoup, D. E. A. Florack, G. M. Stoopen, J. P. F. G. Helsper, R. van Ree, I. van Die, and D. Bosch An antibody produced in tobacco expressing a hybrid beta-1,4-galactosyltransferase is essentially devoid of plant carbohydrate epitopes PNAS, May 16, 2006; 103(20): 7577 - 7582. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||