Genetic differences in omega-gliadins involved in two different immediate food hypersensitivities to wheat

BACKGROUND

Anti-gliadin IgE are expressed in patients with food allergy associated to skin immediate hypersensitivity to hydrolyzed wheat proteins (IHHWP). It is not known if they react with omega5-gliadins, the major allergens in wheat dependant exercise-induced food anaphylaxis (WDEIA), encoded on wheat chromosomes 1B.

METHODS

Unmodified gliadins from 14 wheat varieties expressing most of the 1B omega-gliadin alleles, were immunoprobed after SDS-PAGE and blotting, with four sera from patients with IHHWP, and two with WDEIA. Gliadins reacting with IgE were visualized using chemiluminescence and identified according to their mobility and typical SDS-PAGE pattern. The resulting signal was also measured to compare their IgE reactivity.

RESULTS

IHHWP and WDEIA sera exhibited distinct patterns of reactivity. IgE of patients with IHHWP reacted mainly with all omega-gliadins alleles and one gamma-gliadin encoded respectively on chromosomes 1D and 1B, but not with any omega5-gliadins alleles as for WDEIA. A few other reactive alleles of omega-gliadins were encoded on chromosomes 1A. Unassigned additional bands of the whole gliadin pattern were also reactive. The four patients with IHHWP exhibited almost the same pattern of reactivity. Main differences concerned band reactivity which modulated the overall reactivity of each wheat variety.

CONCLUSIONS

The IgE epitopes involved in IHHWP and WDEIA are different. This suggests that the protein state and the route of exposure to very similar gluten structures, probably orientate the pattern of epitope reactivity and the wheat food allergy manifestations.

Accurate detection of both glycoproteins and total proteins on blots: control of side reactions occurring after periodate oxidation of proteins

The detection of glycoproteins based on the periodate oxidation of their carbohydrate moiety, and their conjugation to digoxigenin hydrazide directly on blots, leads to high background staining, especially with polyvinylidene difluoride (PVDF) membranes. The addition of Tween-20 to all incubation solutions, except at the oxidation step, strongly reduced the background staining and allowed us to detect lower amounts of glycans fixed on the blots through the protein moiety. The presence of polysaccharides during the oxidation step was shown to produce side reactions that led to the staining of nonglycosylated proteins; it demonstrated risks that may occur with crude extracts when periodate oxidation is performed in solution. The phenomenon was used to design a total protein-staining assay that can be included along with positive and negative controls in a general strategy based on blotting, which was delineated for the identification of glycoproteins in complex tissue extracts.

The combination of Indian ink staining with immunochemiluminescence detection allows precise identification of antigens on blots: application to the study of glycosylated barley storage proteins

The localization after blotting of specific spots in two-dimensional electrophoretic protein pattern was achieved using, in that order, Indian ink protein staining and immunodetection with chemiluminescence on the same membrane. Indian ink did not inhibit significantly the antibody reactions even after overnight staining. It produces permanent staining that did not quench the chemiluminescent signal, recorded on a film. This allowed perfect matching between the specific and the total protein patterns. The procedure was applied to the identification of glycoproteins present in barley storage protein preparations.

Identification of glycosylated forms of wheat storage proteins using two-dimensional electrophoresis and blotting

Two-dimensional electrophoresis with acid-polyacrylamide gel electrophoresis (PAGE), followed by sodium dodecyl sulfate (SDS)-PAGE and SDS-PAGE of unreduced polypeptides followed by SDS-PAGE under reducing conditions, were used to separate and identify the different subgroups of gliadins and glutenins and to distinguish between covalent and noncovalent polymers of glutenins. Gels were blotted under semidry conditions according to Laurière (Anal. Biochem. 1993, 212, 206-211) to allow large polymers of glutenins to be transferred efficiently. Glycosylated polypeptides were detected on blots using either the method of Haselbeck and Hösel (Glycoconjugate J. 1990, 7, 63-74), or using anti-(xylose-containing N-glycan) antibodies (Laurière et al., Plant Physiol 1989, 90, 1182-1188). High and low molecular weight glutenin subunits were shown to aggregate through both disulfide bridges and noncovalent protein-to-protein interactions. Aggregated gamma-gliadins were also demonstrated. Glycans were detected on both gliadin and glutenin polypeptides. Covalently aggregated low molecular weight glutenins were shown to contain N-glycans with xylose, which demonstrated their sorting in the Golgi apparatus.

A semidry electroblotting system efficiently transfers both high- and low-molecular-weight proteins separated by SDS-PAGE

A general procedure is described for the simultaneous recovery by electroblotting on nitrocellulose or polyvinylidene difluoride (PVDF) membranes of both high- and low-molecular-weight proteins already separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. It is based on the use of an original set of buffers which creates a stable pH boundary between the two faces of the blotted membrane. The gel to be transferred is placed on the basic side (pH 8.4) of this boundary, which also contains sodium dodecyl sulfate. Under these conditions even the high-molecular-weight proteins are efficiently eluted out of the acrylamide. Their migration is then stopped on the membrane, due to the change of pH with the acid side (pH 3.8) of the boundary, and the presence of methanol in it. This increases the adsorption on the membrane of the low-molecular-weight polypeptides which have a low affinity for nitrocellulose or PVDF. The procedure permits a quantitative transfer of complex samples which contain poorly soluble proteins or with molecular masses outside the range of 20-70 kDa, like grain storage proteins of cereals.

Characterization of a xylose-specific antiserum that reacts with the complex asparagine-linked glycans of extracellular and vacuolar glycoproteins

Antibodies were raised against carrot (Daucus carota) cell wall beta-fructosidase that was either in a native configuration (this serum is called anti-betaF(1)) or chemically deglycosylated (anti-betaF(2)). The two antisera had completely different specificities when tested by immunoblotting. The anti-betaF(1) serum reacted with beta-fructosidase and many other carrot cell wall proteins as well as with many proteins in extracts of bean (Phaseolus vulgaris) cotyledons and tobacco (Nicotiana tabacum) seeds. It did not react with chemically deglycosylated beta-fructosidase. The anti-betaF(1) serum also reacted with the bean vacuolar protein, phytohemagglutinin, but not with deglycosylated phytohemagglutinin. The anti-betaF(2) serum reacted with both normal and deglycosylated beta-fructosidase but not with other proteins. These results indicate that the betaF(2) antibodies recognize the beta-fructosidase polypeptide, while the betaF(1) antibodies recognize glycan sidechains common to many glycoproteins. We used immunoadsorption on glycoprotein-Sepharose columns and hapten inhibition of immunoblot reactions to characterize the nature of the antigenic site. Antibody binding activity was found to be associated with Man(3)(Xyl)(GIcNAc)(2)Fuc, Man(3)(Xyl)(GIcNAc)(2), and Man(Xyl) (GIcNAc)(2) glycans, but not with Man(3)(GIcNAc)(2). Treatment of phytohemagglutinin, a glycoprotein with a Man(3)(Xyl)(GIcNAc)(2)Fuc glycan, with Charonia lampas beta-xylosidase (after treatment with jack-bean alpha-mannosidase) greatly diminished the binding between the antibodies and phytohemagglutinin. We conclude, therefore, that the antibodies bind primarily to the xylosebeta, 1--> 2mannose structure commonly found in the complex glycans of plant glycoproteins.

Characterization of beta-fructosidase, an extracellular glycoprotein of carrot cells

Seedlings and suspension-cultured cells of carrot (Daucus carota) contain a cell wall associated as well as a soluble form of beta-fructosidase (beta F). These two forms have different pH optima: 4.6 for cell wall beta F and 5.6 for soluble beta F. Soluble beta F is relatively more abundant in the seedlings and cell wall beta F is relatively much more abundant in the cultured cells. Protoplasts of cultured cells have only the soluble form (pH optimum 5.6) indicating that the cell wall associated form is indeed extracellular in situ. Cell wall beta F was purified to homogeneity and has an Mr = 63,000. Antibodies raised against the deglycosylated enzyme cross-reacted with two soluble enzyme forms: in cultured cells, the soluble enzyme has an Mr = 58,000 and, in seedlings, there are two forms of Mr = 58,000 and 52,000. Treatment of purified cell wall beta F with endoglycosidase H and trifluoromethanesulfonic acid (complete deglycosylation) indicated that the enzyme probably has one high mannose and two complex glycans. This was confirmed by HPLC analysis of [3H]GlcNAc- and [3H]fucose-labeled glycopeptides obtained after trypsin digestion of radioactively-labeled beta F. The amino acid composition shows that cell wall beta F has 18.6% glycine.

[Nature and fractionation of barley proteins extracted by ethanol, isopropanol and n-propanol in different proportions]

Barley alcoholsoluble protein extractabilities by aqueous ethanol, isopropanol and n-propanol were measured at room temperature. The quantities extracted by each alsohol strongly depend on the concentration of the alcohol. The most efficient concentration for the three alcohols were by increasing order: 45 per cent ethanol, 40 per cent isopropanol, 35 per cent (w/w) n-propanol. Hrodein preparations extracted by these three alcoholic solutions and by 75 per cent (w/w) ethanol were compared by means of the flour nitrogen percentage they contain and by electrophoresis, amino-acid analysis and Sephadex G 100 gel filtration. The preparations studied do not differ markedly in their amino-acid composition or electrophoretic pattern which shows at least 17 different bands. On the other hand, Sephadex G 100 gel filtration separates two main groups of proteins, The first one is present at the same level in all preparations studied and consists of electrophoretically typical hordeins (already described hordeins). The other group represents a fraction of the preparation, the more abundant as the solvent is more effective. This second group is excluded on Sephadex G 100 chromatography and does not give well defined bands by starch gel electrophoresis. Consequently it is related to some glutelins. Nevertheless its amino-acid composition is very close to the mean hordein composition. Electrophoretic comparison with glutelins extracted by acetic acid and with hordeins, all reduced and alkylated, discloses a great similitude between this fraction, the glutelins and some hordein fast components alpha, beta and gamma.