Development of a functional in vitro assay as a novel tool for the standardization of allergen extracts in the human system.

BACKGROUND: Biochemical and immunochemical methods used for batch control of allergen extracts rely on the binding of IgE molecules to allergens. They do not measure the ability of a protein to induce type I allergic reactions. Therefore, a biological assay was established that is based on the cellular mechanisms of allergies in order to assess the cross-linking capacity of allergens. METHODS: Rat basophilic leukaemia cells were transfected with cDNA coding for the human high affinity IgE receptor chains. The surface expression of the IgE-binding alpha-chain was detected by FACS analysis and the functional integration of the 'humanized' receptors into the signal transduction cascade was addressed by intracellular calcium mobilization. Mediator release was measured in response to human IgE and a variety of cross-linking allergen preparations. RESULTS: Several clones were obtained that were able to bind allergen-specific human IgE. The results of the biological assay were compared with those obtained by immunochemical methods. The biological assay was used to determine the potency of allergen extracts, including highly diluted products that cannot be analysed by conventional methods. CONCLUSION: A stable 'humanized' basophil cell line was established that will be a useful tool for the standardization and batch control of allergen extracts. Because of its high sensitivity, it can also be used to detect minute quantities of potentially allergenic proteins, e.g. in processed foods. In addition, the test may support the development of novel allergy vaccines, such as recombinant hypoallergenic molecules.

Molecular cloning and immunological characterisation of potential allergens from the mould Fusarium culmorum.

BACKGROUND: High quality and stability are essential requirements of commercial allergen preparations. Recently we have demonstrated the very low stability of protein allergens in an extract of the ubiquitous mould Fusarium culmorum. OBJECTIVE: The present study was performed to identify, isolate and characterise allergens of F. culmorum as a basis for a stable allergenic reference material. In addition, the significance of IgE binding to carbohydrate structures in the natural allergen source was investigated. METHODS: Sera of 52 subjects with suspected mould allergy were used to determine the IgE binding capacity of a commercial F. culmorum extract and an in-house extract by immunoblotting and enzyme allergo sorbent test (EAST). Binding of IgE-antibodies to putative carbohydrate structures located on glycoproteins was verified by periodate treatment of blot strips prior to immunodetection. A complementary (c)DNA expression library of F. culmorum was prepared and screened for IgE-binding clones using sera from F. culmorum-sensitive individuals. Positive clones were isolated, and the open reading frames were subcloned into expression vectors to produce recombinant proteins in E. coli. The recombinant proteins were tested for their IgE reactivity by immunoblotting and EAST. RESULTS: Using the in-house extract for EAST and immunoblot experiments 44% (23/52) of the sera were found to contain F. culmorum-specific IgE antibodies. Compared to the in-house extract, nearly all IgE-reactivties in the range of 15-30kD were lacking in the commercial preparation as examined by immunoblot analysis and only 10% (5/52) of the sera were found to contain F. culmorum-specific IgE by EAST. IgE binding to putative carbohydrate structures was observed in the high molecular weight range in approximately 50% (12/23) of the IgE-positive sera by both extracts. Three IgE binding clones were isolated from the cDNA-library. One clone (Fus c 1) is homologous to the highly conserved 60S acidic ribosomal protein P2 described as minor allergen in other moulds. The second (Fus c 2) shows high similarity (64%) to a respiratory allergen from the basidiomycete Coprinus comatus (Cop c 2). The third clone (Fus c 3) was not related to known proteins. With sera from 26 individuals sensitised to F. culmorum the IgE prevalence of recombinant proteins rFus c 1, rFus c 2 and rFus c 3 was found to be 35, 50, and 15%, respectively. CONCLUSIONS: F. culmorum may represent an underestimated source of aeroallergens. In contrast to highly labile and poorly standardised F. culmorum extracts, the new recombinant allergens may serve as stable allergenic reference material. A combination of rFus c 1 and rFus c 2 is suitable to diagnose 81% of F. culmorum-sensitised subjects. IgE reactivity to putative carbohydrate structures is relatively frequent, and can not be detected by these allergens.

IgE reactivity to profilin in pollen-sensitized subjects with adverse reactions to banana and pineapple.

BACKGROUND: The so-called 'latex-fruit syndrome' is a well-documented phenomenon in cross-reactive allergies. By contrast, there is a lack of information about allergy to exotic fruits in patients with a predominant pollen sensitization. Since the ubiquitous protein profilin has been identified as an allergen in natural rubber latex as well as in pollen-related foods, the aim of this study was to investigate the role of profilin in allergy to certain exotic fruits. METHODS: Recombinant profilins from banana and pineapple were cloned by a PCR technique after isolation of total RNA using degenerated profilin-specific primers. The unknown 5' ends of copy DNA (cDNA) were identified by rapid amplification of 5'cDNA ends (5'-RACE) and expression in Escherichia coli BL21(DE3) cells. The recombinant profilins were purified by affinity chromatography using poly-(L)-proline as the solid phase. IgE-binding capabilities were characterized by means of immunoblot and Enzyme Allergosorbent Test (EAST). The cross-reactivity to birch pollen profilin and latex profilin was studied by EAST as well as by immunoblot inhibition experiments. RESULTS: Both banana and pineapple profilin were found to consist of 131 amino acid residues with high amino acid sequence identity to known allergenic pollen and food profilins (71-84%). IgE binding to the recombinant profilins was observed in 7/16 sera from subjects with suspected banana allergy (44%) and in 8/19 sera from subjects with suspected pineapple allergy (42%). Inhibition experiments indicated similar IgE reactivity of natural and recombinant allergens. In addition, high cross-reactivity to birch pollen profilin Bet v 2 and latex profilin Hev b 8 was demonstrated by immunoblot inhibition as well as EAST inhibition experiments. CONCLUSIONS: Since a high IgE-binding prevalence of about 40% was obtained in both banana and pineapple allergy, we conclude that profilin is an important mediator of IgE cross-reactivity between pollen and exotic fruits.

Comparison of four variants of a major allergen in hazelnut (Corylus avellana) Cor a 1.04 with the major hazel pollen allergen Cor a 1.01.

The aim of this study was to produce the Bet v 1-related major hazelnut allergen Cor a 1.0401 and variants thereof as recombinant allergens, and to compare their immuno-reactivity with the major hazel pollen allergen using sera of patients whose hazelnut allergy recently was confirmed by double-blind placebo-controlled food challenges (DBPCFC) in a multicenter study. Total RNA was isolated from immature hazelnuts and transcribed into cDNA. Full length coding DNA obtained by PCR-strategy was subcloned into pTYB11 vector and expressed in E. coli ER2566 cells. Native non-fusion target proteins were purified by DTT-induced self-cleavage of the intein-tagged N-terminal fusion proteins. IgE reactivity of the recombinant allergens was tested by enzyme allergosorbent test (EAST), EAST-inhibition, immunoblot-inhibition and histamine release assays. Four recombinant allergens were produced showing deduced amino acid sequence identities among each other of 97-99%, and were considered as variants Cor a 1.0401 (GenBank Accession no.: AF136945), Cor a 1.0402 (AF323973), Cor a 1.0403 (AF323974) and Cor a 1.0404 (AF323975). Cor a 1.0402 and 03 only differed in a C4S exchange. Cor a 1.0404 had a unique proline residue in position 99. Surprisingly, only 63% identity was revealed with hazel pollen Cor a 1. EAST with 43 sera of patients with positive DBPCFC to hazelnut indicated IgE reactivity to Cor a 1.0401 in 95% of the sera, to Cor a 1.0402 in 93%, to Cor a 1.0403 in 91%, and in only 74% of the sera to the proline variant Cor a 1.0404. The allergenic activity of the four variants was confirmed by histamine release assays in 15 hazelnut-allergic patients stimulated with the four variants and controls. Eleven sera were positive with extract from native hazelnut, 13 with rCor a 1.0401, 12 with rCor a 1.0402, 11 with rCor a 1.0403, and only two with rCor a 1.0404 containing the proline exchange. The high IgE binding variant Cor a 1.0401 showed only partial IgE cross-reactivity with pollen Cor a 1. IgE-binding and histamine release capacity led to a concordant ranking of the allergenic activity of the recombinant variants: Cor a 1.0401>Cor a 1.0402 and 03>Cor a 1.0404 (the proline variant). Similar results for Cor a 1.0402 and 03 suggest a minor influence in IgE binding of cysteine in position 4, whereas proline in position 99 appears to be responsible for the decrease in IgE reactivity in Cor a 1.0404. It appears that the epitopes of hazelnut Cor a 1.04 are less related to pollen Cor a 1 than to Bet v 1 from birch pollen. Low IgE binding variants or mutants of Cor a 1.04 are candidate compounds for developing a novel and safe approach of specific immunotherapy of hazelnut allergy.

Phenylcoumaran benzylic ether and isoflavonoid reductases are a new class of cross-reactive allergens in birch pollen, fruits and vegetables.

We investigated the biochemical function of the birch pollen allergen Bet v 6 and its role in the IgE-cross-reactivity between birch pollen and plant foods, and characterized Pyr c 5, a Bet v 6-related food allergen, from pear; the proteins were expressed as His-Tag fusion proteins in Eschershia coli and purified by Ni-chelate affinity chromatography under native conditions. Nonfusion proteins were obtained by factor Xa protease treatment. The highest degree of amino-acid sequence identity of Pyr c 5 and Bet v 6 was found with a plant protein related to a defense mechanism, which we have named phenylcoumaran benzylic ether reductase (PCBER) based on its ability to catalyze the NADPH-dependent reduction of 8-5' linked lignans such as dehydrodiconiferyl alcohol to give isodihydrodehydrodiconiferyl alcohol. Enzymatic assays with recombinant Pyr c 5 and Bet v 6 showed PCBER catalytic activity for both recombinant allergens. Both Pyr c 5 and Bet v 6 allergens had similar IgE binding characteristics in immunoblotting and enzyme allergosorbent tests (EAST), and bound IgE from 10 sera of birch-pollen-allergic patients including six pear-allergic subjects. EAST inhibition experiments with Pyr c 5 as the solid phase antigen suggested that homologous allergens may be present in many vegetable foods such as apple, peach, orange, lychee fruit, strawberry, persimmon, zucchini (courgette), and carrot. In extracts of pear, apple, orange, and persimmon, the presence of proteins of approximately 30-35 kDa containing Bet v 6 cross-reactive epitopes was demonstrated with two Bet v 6-specific monoclonal antibodies. Recombinant Pyr c 5 triggered a strong, dose-dependent mediator release from basophils of a pear-allergic subject, suggesting that Pyr c 5 has the potential to elicit type I allergic reactions.

Recombinant food allergens.

Allergenic (glyco)proteins are the elicitors of food allergies and can cause acute severe hypersensitivity reactions. Recombinant food allergens are available in standardised quantity and constant quality. Therefore, they offer new perspectives to overcome current difficulties in the diagnosis, treatment and investigation of food allergies. This review summarises the expression strategies and characteristics of more than 40 recombinant food allergens that have been produced until today. Their IgE-binding properties are compared to those of their natural counterparts, in addition their application as diagnostic tools, the generation of hypoallergenic recombinant isoforms and mutants for therapeutic purposes, the determination of epitopes and cross-reactive structures are described.

Pyr c 1, the major allergen from pear (Pyrus communis), is a new member of the Bet v 1 allergen family.

Pear is known as an allergenic food involved in the 'oral allergy syndrome' which affects a high percentage of patients allergic to birch pollen. The aim of this study was to clone the major allergen of this fruit, to express it as bacterial recombinant protein and to study its allergenic properties in relation to homologous proteins and natural allergen extracts. The coding region of the cDNA was obtained by a PCR strategy, cloned, and the allergen was expressed as His-Tag fusion protein. The fusion peptide was removed by treatment with cyanogen bromide. Purified non-fusion protein was subjected to allergenicity testing by the enzyme allergosorbent test (EAST), Western blotting, competitive inhibition assays, and basophil histamine release. The deduced protein sequence shared a high degree of identity with other major allergens from fruits, nuts, vegetables, and pollen, and with a family of PR-10 pathogenesis related proteins. The recombinant (r) protein was recognised by specific IgE from sera of all pear-allergic patients (n = 16) investigated in this study. Hence, the allergen was classified as a major allergen and named Pyr c 1. The IgE binding characteristics of rPyr c 1 appeared to be similar to the natural pear protein, as was demonstrated by EAST-inhibition and Western blot-inhibition experiments. Moreover, the biological activity of rPyr c 1 was equal to that of pear extract, as indicated by basophil histamine release in two patients allergic to pears. The related major allergens Bet v 1 from birch pollen and Mal d 1 from apple inhibited to a high degree the binding of IgE to Pyr c 1, whereas Api g 1 from celery, also belonging to this family, had little inhibitory effects, indicating epitope differences between Bet v 1-related food allergens. Unlimited amounts of pure rPyr c 1 are now available for studies on the structure and epitopes of pollen-related food allergens. Moreover, the allergen may serve as stable and standardised diagnostic material.

Cross-reactivity within the profilin panallergen family investigated by comparison of recombinant profilins from pear (Pyr c 4), cherry (Pru av 4) and celery (Api g 4) with birch pollen profilin Bet v 2.

Profilin is a panallergen which is recognised by IgE from about 20% of birch pollen- and plant food-allergic patients. Little is known about epitope diversity among these homologous proteins, and about the correlation between IgE-cross-reactivity and allergenic reactivity. Plant food profilins from pear (Pyr c 4) and cherry (Pru av 4) were cloned by polymerase chain reaction and produced in Escherichia coli BL21. The profilins were purified as non-fusion proteins by affinity chromatography on poly-(L-proline)-Sepharose and characterized by immunoblotting, IgE-inhibition experiments and histamine release assays. The coding regions of the cDNA of pear and cherry profilin were identified as a 393 bp open reading frame. The deduced amino acid sequences showed high identities with birch pollen profilin Bet v 2 (76-83%) and other allergenic plant profilins. Pyr c 4 and Pru av 4 were investigated for their immunological properties in comparison with profilins from celery (Api g 4) and birch pollen (Bet v 2). Fourty-three of 49 patients (88%), preselected for an IgE-reactivity with Bet v 2 showed specific IgE-antibodies to the recombinant pear protein, 92% of the sera were positive with the recombinant cherry allergen and 80% of the sera were reactive with the celery protein. Inhibition experiments showed a strong cross-reactivity of IgE with profilins from plant food and birch pollen. However, IgE binding profiles also indicated the presence of epitope differences among related profilins. All investigated profilins, Pyr c 4, Pru av 4, Api g 4 and Bet v 2, presented almost identical allergenic properties in cellular mediator release tests. Therefore, cross-reactivities between related profilins may explain pollen-related allergy to food in a minority of patients. The nucleotide sequences reported have been submitted to the Genbank database under accession numbers AF129424 (Pyr c 4) and AF129425 (Pru av 4).

Recombinant allergens Pru av 1 and Pru av 4 and a newly identified lipid transfer protein in the in vitro diagnosis of cherry allergy.

BACKGROUND: In central and northern Europe food allergy to fruits of the Rosaceae family is strongly associated with birch pollinosis because of the existence of IgE cross-reactive homologous allergens in birch pollen and food. By contrast, in the Mediterranean population allergic reactions to these fruits frequently are not related to birch pollen allergy and are predominantly elicited by lipid transfer proteins (LTPs). OBJECTIVE: We sought to determine the prevalence of IgE sensitization to the recombinant cherry allergens Pru av 1 and Pru av 4 in comparison with cherry extract within a representative group of patients who were allergic to cherries recruited in Germany and to compare the relevance of IgE to cherry LTPs in Italian patients. METHODS: Recombinant Pru av 1 and rPru av 4 were available from earlier studies. The cDNA of the cherry LTPs was obtained by using a PCR-cloning strategy. The protein was expressed in Escherichia coli and purified by means of metal chelate affinity chromatography. Sera from 101 German patients with birch pollinosis and oral allergy syndrome to cherry and sera from 7 Italian patients with cherry allergy were investigated by using enzyme allergosorbent tests for IgE reactivity with cherry extract, rPru av 1, rPru av 4, and the recombinant cherry LTP. Inhibition experiments were performed to compare the IgE reactivity of natural and recombinant cherry LTPs and to investigate potential cross-reactivity with birch pollen allergens. RESULTS: The LTP from cherry comprises 91 amino acids and a 26 amino acid signal peptide. The mature cherry LTP shows high amino acid sequence identity with allergenic LTPs from peach (Pru p 3, 88%), apricot (Pru ar 3, 86%), and maize (Zea m 14, 59%) and displays no IgE cross-reactivity with birch pollen. The IgE prevalences in the German patients were as follows: LTP, 3 of 101 (3%); rPru av 1, 97 of 101 (96.0%); rPru av 4, 16 of 101 (16.2%); and cherry extract, 98 of 101 (97%). All 7 Italian patients had IgE against the cherry LTP. CONCLUSIONS: Recombinant allergens are useful tools for a more accurate in vitro IgE-based diagnosis of cherry allergy. Taken together, they mimic the allergenic activity of cherry extract, having slightly higher biologic activity. Sensitization to the cherry LTP is relevant for a minority of patients recruited in Germany, but our data indicate that it may be a major allergen in Italy.

Cloning of the minor allergen Api g 4 profilin from celery (Apium graveolens) and its cross-reactivity with birch pollen profilin Bet v 2.

BACKGROUND: Profilin is a panallergen that is recognized by IgE from about 20% of birch pollen- and plant food-allergic patients. A subgroup of celery-allergic patients shows IgE-reactivity with this minor allergen. To investigate the IgE-binding potential and cross-reactivity of celery profilin at the molecular level, this study was aimed at the cloning and immunological characterization of this allergen. OBJECTIVES: Cloning, expression and purification of profilin from celery tuber to characterize its immunological properties and its cross-reactivity with birch pollen profilin. METHODS: Cloning of celery profilin was performed by polymerase chain reaction using degenerated primers and a 5'RACE method for the identification of the unknown 5'-end of the cDNA. Expression was carried out in Escherichia coli BL21 (DE3) using a modified vector pET-30a. The recombinant profilin was purified by affinity chromatography on poly L-proline coupled to sepharose. Immunological characterization was performed by immunoblotting, EAST and IgE-inhibition experiments. RESULTS: The coding region of the cDNA of celery profilin was identified as a 399-bp open reading frame, coding for a protein of 133 amino acids with a calculated molecular weight of 14.3 kDa. The deduced amino acid sequence of the corresponding protein showed high identity with other plant profilins (71-82%) recently described as allergens. Celery profilin was isolated as highly pure nonfusion protein. The IgE-reactivity of celery profilin was similar to that of natural protein. Seven of 17 celery-allergic patients tested presented specific IgE-antibodies to the recombinant protein tested by immunoblotting. Inhibition experiments showed high cross-reactivity of IgE with both profilins from celery and birch pollen. Moreover, the biological activity of recombinant celery profilin was demonstrated by a histamine release assay. CONCLUSIONS: Celery profilin is an important allergenic compound in celery and shows high homology to birch pollen profilin, Bet v 2. According to the revised IUIS allergen nomenclature, we suggest naming the celery profilin Api g 4. In addition to the cross-reacting major allergens Api g 1 and Bet v 1, birch pollinosis and associated allergies to celery can therefore additionally be explained by the cross-reactivity between homologous profilins. Moreover, recombinant Api g 4 may be used for target-specific diagnosis and structural analyses.

Molecular cloning and characterization of a birch pollen minor allergen, Bet v 5, belonging to a family of isoflavone reductase-related proteins.

BACKGROUND: Birch pollen is a major cause of pollinosis and is responsible for cross-reactive oral allergies to fruits, nuts, and vegetables. The major allergen, Bet v 1, has been extensively characterized, and 3 minor allergens, Bet v 2, Bet v 3, and Bet v 4, have been cloned and sequenced. Recently, another birch pollen protein with an apparent mass of 35 kd was described as a new IgE-binding protein in birch pollen with cross-reacting homologues in plant foods. OBJECTIVE: The aim of this study was to determine the primary structure of the 35-kd birch pollen allergen and to investigate its immunologic properties. METHODS: On the basis of a known complementary DNA fragment, a PCR strategy was applied to obtain the full-length nucleotide sequence of the coding region. The protein was expressed as His-Tag fusion protein in Escherichia coli and purified by Ni-chelate affinity chromatography. Nonfusion protein was obtained by cyanogen bromide treatment of the fusion protein. IgE-binding characteristics and potential allergenicity were investigated by immunoblot, immunoblot inhibition analysis, rat basophil leukemia-cell mediator release assay, and basophil histamine release and compared with those of natural (n) Bet v 5, recombinant (r)Bet v 1, and rBet v 2. RESULTS: Recombinant Bet v 5 has a mass of 33 kd, an isoelectric point of 9.0, and sequence identity of 60% to 80% to isoflavone reductase homologue proteins from various plants. On immunoblots the recombinant Bet v 5 bound IgE from 9 (32%) of 28 sera from patients allergic to birch pollen with a CAP class of at least 3; Bet v 1 was detected by 89% of these patients. IgE immunoblot and inhibition experiments showed that nBet v 5 and rBet v 5 shared identical epitopes. A rabbit antiserum raised against pea isoflavone reductase and patients' IgE reacted with Bet v 5 and proteins of similar size in several vegetable foods, including exotic fruits. A similar reaction pattern was obtained with 2 Bet v 5-specific mAbs. Furthermore, Bet v 5 triggered a dose-dependent mediator release from rat basophil leukemia 2H3 cells passively sensitized with murine anti-birch pollen IgE and from basophils of a Bet v 5-reactive subject with birch pollen allergy. In contrast, no mediator release could be induced from basophils of a subject who was monosensitized to Bet v 1. CONCLUSIONS: This 33-kd protein, designated as Bet v 5, is a new minor allergen in birch pollen and may be responsible for pollen-related oral allergy to specific foods in a minority of patients with birch pollen allergy. Amino acid sequence comparison and immunoreactivity to anti-isoflavone reductase serum indicate that Bet v 5 is related to isoflavone reductase, a protein family that is involved in plant defense reactions.

Involvement of carbohydrate epitopes in the IgE response of celery-allergic patients.

BACKGROUND: This study was performed to get further insights into antibody responses to cross-reactive carbohydrate determinants (CCD), including initial experiments to prove the biological activity of anti-CCD IgE. Earlier studies have shown that IgE specific for CCD occurs in about 25% of celery-allergic patients. The clinical significance of these antibody specificities is doubtful. METHODS: Patient sera were selected on the basis of a positive case history of celery allergy and multiple binding to high molecular weight celery allergens on immunoblots. Specific IgE to native and heated celery tuber was determined by the enzyme allergosorbent test (EAST). N-glycans were purified after extensive digestion of specific glycoproteins, such as pineapple stem bromelain, bovine fibrin, and human IgG, and used as antigens in an IgE ELISA as well as in EAST and immunoblotting inhibition experiments. Dose-related histamine release was performed with BSA neoglycoproteins containing 3-4 units of the purified glycopeptides. RESULTS: Seven celery-allergic patients were identified who clearly presented IgE against the N-glycan purified from bromelain which is a common structure within the plant kingdom. Chemical defucosylation showed that alpha1, 3-fucose is a key structure for IgE binding. In patients with anti-CCD IgE, the maximal inhibition of celery EAST by the bromelain glycan ranged from 22 to 100%. Inhibition of celery immunoblots by preincubation of patient serum with this glycan led to a quenching of multiple bands at masses >40 kD. After linking the bromelain glycopeptide to BSA, a strong dose-related histamine release was obtained in a celery-allergic patient, occurring at lower concentrations than with the recombinant major protein allergen from celery, Api g 1. CONCLUSIONS: Our results demonstrate that IgE specific for CCD is common in celery-allergic patients, and can represent the major proportion of IgE against this food. alpha1, 3-fucose was confirmed to be an essential part of the IgE epitope. Immunoblotting inhibition indicated the presence of this carbohydrate determinant on multiple glycoproteins in celery extract. Although histamine release was only performed in 1 patient, our data show that proteins carrying multiple glycan units can be biologically active in patients sensitized to CCD.

Pollen-related food allergy: cloning and immunological analysis of isoforms and mutants of Mal d 1, the major apple allergen, and Bet v 1, the major birch pollen allergen.

BACKGROUND: Mal d 1, the major apple allergen, cross-reacts with IgE specific for the major birch pollen allergen, Bet v 1, and is responsible for birch pollen related food allergy to apple. Isoforms of Bet v 1 showing minor sequence variations display different binding capacity for specific IgE antibodies from allergic patients. Moreover, strain-dependent variation of allergenicity has been reported for apples. OBJECTIVE: To investigate the occurrence of strain-dependent isoforms of Mal d 1 which may differ in their allergenic potential, to obtain data on structures essential for binding of Mal d 1 to the antibody, and to gain insights into the structures responsible for its IgE cross-reactivity to Bet v 1. METHODS: The cDNA of Mal d 1 from various apple strains was amplified by a PCR strategy based on conserved regions of known Mal d 1-sequences, and sequenced. Two major isoforms of Mal d 1 were expressed as recombinant proteins and purified, as were different variants of the major birch pollen allergen, Bet v 1. Together with already existing recombinant birch pollen and apple allergens, these were subjected to allergenicity testing by IgE-immunoblotting, enzyme allergo sorbent test and dose related mediator release. "Hot-spots" for IgE-reactivity were identified by site-directed mutagenesis. RESULTS: Twelve Mal d 1-clones were sequenced from 7 apple varieties and compared to 3 known Mal d 1 sequences. The clones were clustered into two groups, each showing a high degree of sequence identity to one of the known sequences and specific differences to the third sequence. No strain-specific sequences were identified. In contrast, apple strains with reported differences in allergenicity showed different expression levels of the major allergen. Immunologic testing of recombinant allergens revealed high IgE binding capacity of 2 major isoforms, named GD26 and GS29, with a slightly higher IgE binding capacity of GD26. Moreover, the allergenicity was similar to another r Mal d 1 reported in the literature, representing the isoform divergent from our clones. Mutational analysis of our Mal d 1 allergens identified serine in position 111 as essential for IgE binding. Allergenicity was almost depleted by changing this residue into a proline. Moreover, the corresponding serine residue, present in position 112 of Bet v 1, was in a similar manner crucial for the allergenicity of the birch pollen allergen. CONCLUSION: We conclude that divergent allergenicity of apple strains mainly depends on different expression levels of the major allergen. Introduction of a proline residue in position 111 of Mal d 1 and in position 112 of Bet v 1 led to a drastic reduction of allergenicity of both the pollen and the food allergen, obviously also removing the cross-reactive epitope. Mutants with reduced IgE-reactivity but maintained T-cell reactivity may represent new candidates for a safer specific immunotherapy with reduced side-effects.

Cross-reactivity and epitope analysis of Pru a 1, the major cherry allergen.

A high percentage of birch pollen allergic patients experiences food hypersensivity after ingestion of fresh fruits and vegetables. The cross-reactivity of the major allergens of sweet cherry (Pru a 1), apple (Mal d 1), pear (Pyr c 1), celery tuber (Api g 1) and carrot (Dau c 1) is due to structural similarities which are reflected by high amino acid sequence identities with Bet v 1a, the major birch pollen allergen. Apart from a strong cross-reactivity to Bet v 1a, IgE inhibition experiments with Mal d 1, Pru a 1 and Api g 1 demonstrated the presence of common and different epitopes among the tested food allergens. Secondary structure prediction of all investigated allergens indicated the presence of almost identical structural elements. In particular, the 'P-loop' region is a common domain of the pollen related food allergens and of pathogenesis related proteins. To identify the IgE binding epitopes, five overlapping recombinant Pru a 1 fragments representing the entire amino acid sequence with lengths of approximately 60-120 residues were investigated. Weak IgE binding capacity was measured exclusively with Pru a IF4 (1-120) by immunoblotting, whereas none of the fragments showed allergenicity in the rat basophil leukaemia cell mediator release assay. Site-directed mutagenesis experiments with Pru a 1 revealed that amino acid S112 is critical for IgE binding of almost all patients sera tested. This reduced IgE binding was also observed with a single point mutant of Bet v 1a (S112P) and thus indicated serine 112 as an essential residue for preserving the structure of a cross-reactive IgE epitope. Moreover, two Pru a 1 mutants with an altered 'P-loop' region, showed a lowered IgE binding capacity for IgE from a subgroup of allergic patients. The investigation of essential features for preserving cross-reactive IgE-epitopes provides the structural basis for understanding the clinically observed cross-allergenicity between pollen and fruits. Moreover, non-anaphylactic allergen fragments or variants derived from the IgE-inducing pollen allergens may serve as useful tools for a new strategy of specific immunotherapy.

Determination of the allergenic activity of birch pollen and apple prick test solutions by measurement of beta-hexosaminidase release from RBL-2H3 cells. Comparison with classical methods in allergen standardization.

BACKGROUND: A murine in vitro model of the allergic type I reaction was set up to determine the biologic activity of extracts without involvement of human beings. It is based on beta-hexosaminidase release from passively sensitized RBL cells after allergen challenge. The intended application of this RBL cell assay in the field of quality control of allergenic extracts requires its comparison with established methods. METHODS: The activity of five standardized birch-pollen prick test solutions was determined in parallel by RBL assay, direct IgE binding, IgE-binding inhibition, major allergen content, histamine-release assay, and skin testing. RESULTS: The RBL cell-release assay corresponded well to other methods if a reagin raised against natural birch-pollen extract was used for passive sensitization. However, in the case of a reagin against recombinant Bet v 1, only a decreased activity was observed, presumably because a reduced number of epitopes were recognized by the monospecific reagin. In contrast to standardized birch-pollen extracts, nonstandardized apple extracts showed poor activity in all assays. CONCLUSIONS: This murine model might be a useful tool in the quality control of allergenic extracts. It combines properties of assays based on standardized antisera and of assays that consider IgE cross-linking properties.

IgE antibodies specific for carbohydrates in a patient allergic to gum arabic (Acacia senegal).

The present study deals with the detailed investigation of the IgE antibody response of a gum arabic-allergic patient. The patient showed multiple serologic and skin test sensitizations to a range of pollen, other inhalants and foods, and bee venom, and to the recombinant allergens Bet v 1 and Bet v 2. Moreover, the patient's serum reacted strongly to gum-arabic extract. The NaIO4-treated and thus deglycosylated extract showed no binding to IgE. In contrast, removal of the protein backbone by basic hydrolysis did not deplete the IgE reactivity. Therefore, it is concluded that the gum arabic-specific IgE antibodies of this patient were mainly directed against the carbohydrate fraction of this material. In IgE-inhibition assays, cross-reactions occurred in the range of 60% between gum arabic and known immunogenic N-glycans containing alpha1-3-linked fucose. Since the inhibition graphs were not parallel and the inhibition was not complete with heterologue antigens, the cross-reacting epitopes of gum arabic appeared to be different from the latter well-known cross-reactive carbohydrate determinants (CCD). Inhibition may have been caused by a partial immunologic identity of the investigated carbohydrate moieties. A strong IgE response to the fucose-containing glycan from bromelain was measured in a glycan ELISA that utilizes purified glycopeptides at the solid phase. This response, which may explain the multiple sensitizations without clinical significance diagnosed in the patient, could originate from inhalation of pollen, which is known to contain similar glycans, or from occupational sensitization during work as a baker and confectioner. Since the gum-arabic protein showed only very weak participation in the IgE reactivity, the clinical symptoms of the patient caused by gum arabic may be attributed to carbohydrate epitopes. Due to the repetitive polysaccharide sequence of gum arabic, several epitopes for the cross-linking of IgE should exist.

Factors influencing the quality of food extracts for in vitro and in vivo diagnosis.

Food extracts for diagnostic purposes often lack sufficient activity and consistency. Biologically standardized food extracts are not available on the market. Using extracts from plant-derived foods as examples, we investigated factors which may be important for the quality of such extracts. Divergent allergenic activities were found between strains of apples, but not within varieties of celery tuber (celeriac), hazelnut, and peanut, respectively. Heating of the food remarkably reduced the activity of apple, hazelnut, and celeriac, but had little effect on peanut. By contrast, heating of semipurified protein extracts from celery tuber and apple for 30 min at 100 degrees C did not deplete the immunoreactivity of the major allergens, indicating that this is an inappropriate test for identifying labile food allergens. Due to their high endogenous enzyme activities, apples and other fruits require special extraction procedures applying either low temperature or enzyme inhibitors. Variation of extraction conditions had little effect on the composition and activity of extracts from hazelnut. The storage stability of skin test solutions from plant foods can be improved by avoiding phenol as an additive and by including 50% of glycerol. For model studies considering neoallergens, IgE was raised in mice against native and heated celery tuber, respectively. When extracts from nonthermally and thermally processed celeriac were subjected to an RBL-cell mediator release assay with these sera, an inverse ranking was obtained with anti-heated celeriac IgE and anti-native celeriac IgE, respectively. These data indicated that new epitopes had been formed by the heating process. Since all parameters were tested in model experiments with either human or murine IgE, their relevance has to be proven in further clinical investigations.