- Email: [email protected]
Isolation and biochemical characterization of a thaumatin-like kiwi allergen
Isolation and biochemical characterization of a thaumatin-like kiwi allergen Marija Gavrovic´-Jankulovic´, PhD,a Tanja C´irkovic´, MS,a Olga Vucˇ kovic´, MD,b Marina Atanaskovic´-Markovic´, PhD,c Arnd Petersen, PhD,d Gordana Gojgic´, PhD,a Lidija Burazer, MS,b and Ratko M. Jankov, PhDa Belgrade, Yugoslavia, and Borstel, Germany
From athe Department of Biochemistry, Faculty of Chemistry, University of Belgrade, Belgrade; bthe Department of Allergology, Institute for Immunology and Virology Torlak, Belgrade; cthe Department of Allergology, University Children’s Hospital Belgrade; and dResearch Center Borstel, Biochemical and Molecular Allergology, Borstel. Supported by grant No. 1802 from the Ministry of Science, Technologies, and Development, Republic of Serbia. Received for publication April 26, 2002; revised July 29, 2002; accepted for publication August 13, 2002. Reprint requests: Marija Gavrovic´-Jankulovic´, PhD, Faculty of Chemistry, Department of Biochemistry, Studentski trg 16, 11000 Belgrade, Yugoslavia. © 2002 Mosby, Inc. All rights reserved. 0091-6749/2002 $35.00 + 0 1/81/128947 doi:10.1067/mai.2002.128947
9.5 and molecular weight of 24 kd), which belongs to the family of pathogenesis-related proteins. The isolated protein expressed antifungal activity toward S carlsbergensis and C albicans. (J Allergy Clin Immunol 2002;110:805-10.) Key words: Allergen purification, antifungal activity, food allergen, kiwi fruit, pathogenesis-related protein, thaumatin-like protein
Kiwi fruit allergy, together with its association with hypersensitivity to other foods and to tree and grass pollen, has been widely reported in the last few years. The clinical symptoms range from localized symptoms confined to the oral mucosa to severe anaphylactic reactions. Common allergenic structures have been identified in kiwi and hazel nuts, birch, timothy, rye, mugwort, and fescue meadow pollen.1-6 It is important to identify allergenic molecules to improve the quality of allergen extracts.7 Identification of food allergens is a priority in the management of food allergy8 because well-characterized relevant allergens might replace allergen extracts in a component-based allergy diagnosis. The rapid progress made in the field of molecular allergen characterization appears likely to considerably improve the use of recombinant allergens in diagnosis and specific immunotherapy.9 Although loss of allergenicity was reported for recombinant thaumatin-like cherry protein, probably because of incorrect folding in the polypeptide chain,10 Scheurer et al11 indicated that it is possible to replace cherry extract with 3 recombinant allergens without affecting the results of in vitro diagnosis. However, the isolation and characterization of natural allergens is indispensable in the creation of their structural, as well as functional, recombinant counterparts. Namely, researchers need to compare biochemical and immunologic characteristics of recombinant molecules with those of native molecules to verify that molecular expression in heterologous systems does not significantly modify the IgE response pattern, which would make recombinant proteins useless for diagnostic purposes.8 Although several studies reported that specific IgE immunodetection revealed approximately 12 allergens in kiwi extract, only one 30-kd allergen has been isolated; it was identified as actinidin (Act c 1).12 Recently, we reported that a 24-kd kiwi protein was a potential major allergen in a group of 8 patients with oral allergy syndrome (OAS).6 805
Food and drug reactions and anaphylaxis
Background: Kiwi fruit allergy, as well as its association with hypersensitivity to other foods and to pollen, has been extensively reported in the last few years. Several IgE-binding components have been detected in kiwi extract, but only one 30-kd allergen has been isolated; it was identified as actinidin (Act c 1). Recently, we have reported a 24-kd kiwi protein to be a potential major allergen in a group of patients with oral allergy syndrome (OAS). Objective: The aim of this study was to purify and characterize the 24-kd kiwi allergen biochemically. Methods: Seven polysensitized patients with OAS to kiwi were used in this study. The kiwi allergen was isolated by using a combination of gel permeation, ion exchange, and immobilized metal ion affinity chromatography. Its biochemical characterization included determination of its isoelectric point, molecular weight, N-terminal sequencing, concanavalin A–binding ability, digestibility in simulated gastric fluid, and antifungal activity. Western blotting, 2-dimensional PAGE immunoblotting, and skin prick tests were performed to characterize the isolated protein immunochemically. Results: All 7 patients recognized the isolated 24-kd kiwi protein as an allergen. The isolated protein consisted of 2 isoforms with isoelectric points of 9.4 and 9.5 migrated as one protein band of 20 kd after SDS-PAGE under nonreducing conditions or at 24 kd under reducing conditions. The partial N-terminal sequence revealed that it is a thaumatin-like protein (TLP) with concanavalin A–binding ability. The protein showed antifungal activity toward Saccharomyces carlsbergensis, and Candida albicans. The protein was degraded by the simulated gastric fluid within 1 minute. Both isoforms bound IgE from a pool of sera in a 2-dimensional PAGE immunoblot. The TLP elicited positive skin prick test responses in 4 (80%) of 5 patients with OAS. Conclusion: This study reported isolation and full characterization of a new kiwi allergen, TLP (isoelectric points of 9.4 and
806 Gavrovic´-Jankulovic´ et al
J ALLERGY CLIN IMMUNOL NOVEMBER 2002
Abbreviations used OAS: Oral allergy syndrome pI: Isoelectric point PR: Pathogenesis related SGF: Simulated gastric fluid SPT: Skin prick test TLP: Thaumatin-like protein
TABLE I. Clinical data, total and specific IgE levels, and skin test results of the patients allergic to kiwi Patient no.
Kiwi symptoms
Total IgE (kU/L)
Specific IgE (kUA/L)
1 2 3 4 5 6 7
OS/RC OS/BA OS/RC AE U/A OS/RC OS/RC
797 2709 382 732 ND 910 151
0 2.89 0 0.90 ND 0.71 0.70
Kiwi TLP SPT* SPT*
16 16 7 16 36 ND ND
7 7 — 7 36 ND ND
OS, Oral syndrome; RC, rhinoconjunctivitis; BA, bronchial asthma; AE, angioedema; U, urticaria; A, systemic anaphylaxis; ND, not determined. *SPT results are given in square millimeters of wheal area.
The aim of this study was to isolate and characterize the 24-kd kiwi allergen.
METHODS Patient sera We have studied seven patients who had symptoms suggestive of an IgE-mediated reaction after the ingestion of kiwi, which are summarized in Table I. All the patients tolerated other kinds of exotic fruits, and some of them had airborne allergen sensitization to grass pollens. Open challenge with kiwi was performed according to instructions given in an article by the European Academy of Allergology and Clinical Immunology.13 All sera had positive immunoblotting results with self-prepared kiwi extract. In addition, a pool of sera from 3 subjects (nonatopic) was used as the negative control.
Skin tests
Food and drug reactions and anaphylaxis
Skin prick tests (SPTs) with the kiwi extract (500 µg/mL) and the purified allergen (30 µg/mL) were performed according to the standard procedure14 in the volar side of the forearm. Histamine phosphate (10 mg/mL) and PBS were used as positive and negative controls, respectively. A mean wheal area of 7 mm2 (diameter ≥3 mm) or greater compared with that produced by the negative control 20 minutes after puncture was considered a positive response.
Total and specific serum IgE determination The patients’ total and specific IgE levels were determined by using the CAP System IgE FEIA (Pharmacia Diagnostics), according to the manufacturer’s instructions. Levels of specific IgE of greater than 0.35 kU/L (≥class 1) were considered positive.
Isolation and characterization of the allergen Kiwi fruit (Actinidia chinensis) purchased in a local store was used as the starting material for allergen purification. Peeled kiwi (50 g) was cut into pieces, and then 100 mmol/L sodium bicarbon-
ate buffer containing 2% (wt/vol) polyvinylpolypyrrolidone, pH 9.3 (1:2 wt/vol), was poured over the kiwi, followed by homogenization in a blender for 1 minute. After extraction (3 hours at 4°C), the slurry was passed through a cheesecloth and centrifuged (15 minutes at 10,000g). The resulting protein content was 0.75 mg/mL. Ammonium sulfate was dissolved in the extract (100 mL) to achieve 60% saturation. After overnight standing (4°C), the solution was centrifuged (15 minutes at 10,000g). The obtained pellet was dissolved in a minimal volume of the starting buffer and dialyzed extensively against the same buffer. The dialysate was spun and subjected to gel permeation chromatography on a Bio-Gel P-60 (Bio Rad) column (39 × 1.2 cm). Enriched protein fractions were pooled, concentrated twice, and subjected to ion-exchange chromatography on a QAE-Sephadex A-50 column (8 × 1 cm) pre-equilibrated with 100 mmol/L sodium bicarbonate buffer, pH 9.3. The unbound protein fractions were pooled, concentrated twice by means of ultrafiltration, and dialyzed against 150 mmol/L NaCl in 50 mmol/L sodium acetate buffer, pH 6.0, overnight. The protein sample was applied on a copper iminodiacetate Sepharose CL 4B (Cu2+-IDA-Sepharose) column (3.5 × 0.7 cm). Finally, the unbound protein sample was rechromatographed on a Superdex 75 prep grade (Pharmacia) column (30 × 1.2 cm). The distribution of the protein through the various chromatographic steps was monitored by means of SDSPAGE. The amount of purified protein quantified according to the method of Lowry and colleagues15 was 6 mg from 75 mg of starting material. Analytic SDS-PAGE was performed on a polyacrylamide gel (4% stacking gel and 10% resolving gel), according to the method of Laemmli.16 After electrophoresis, proteins were either stained with CBB or semidry transferred (1 mA/cm2) to a nitrocellulose membrane (Serva) for further examination. In 2-dimensional PAGE isoelectric focusing was performed in the model 2117 Multiphor cell (LKB Pharmacia), according to the manufacturer’s instructions. The protein sample (8 µg) was separated under conditions, as previously described.17 The resolved thaumatin-like kiwi protein in 2-dimensional PAGE was blotted to a polyvinylidene difluoride membrane and visualized by staining the blot with 0.1% CBB G-250, followed by destaining with several changes of 50% methanol. Spots with thaumatin-like kiwi protein were excised, and automated Edman degradation was performed. A concanavalin A–horseradish peroxidase complex was used to detect carbohydrate moieties.17
Digestion with simulated gastric fluid The digestibility of the self-prepared kiwi extract and isolated thaumatin-like protein (TLP) in the simulated gastric fluid (SGF) was examined according to the method of Yagami et al.18 Briefly, a protein sample (700 µg of crude proteins or 40 µg of TLP) was dissolved in 200 µL of prewarmed SGF (US Pharmacopoeia) containing 0.32% wt/vol percentage of pepsin A (Sigma Chemical Co). Digestion proceeded at 37°C with continuous shaking, and an aliquot (20 µL) of the digest was periodically withdrawn (at 0.5, 1, 2, 4, 8, 16, and 60 minutes). The digestion was stopped with 0.2 mol/L Na2CO3 (6.0 µL), and samples were mixed with a sample buffer for SDS-PAGE analysis.
Immunodetection The protein samples were separated by means of SDS-PAGE (4% stacking gel and 10% resolving gel) with reducing conditions and then electrotransferred onto a nitrocellulose membrane, as described by Towbin et al.19 Immunodetection was performed according to the method of Harlow and Lane.20 In 2-dimensional PAGE immunoblotting, a pool of 7 sera was used for detection of IgE-binding components.
Gavrovic´-Jankulovic´ et al 807
J ALLERGY CLIN IMMUNOL VOLUME 110, NUMBER 5
In vitro antifungal activity assay The antifungal activity assay was performed in vitro according to the method of Liu et al,21 with a modification. A culture of S carlsbergensis (0.2 mg/mL) was washed and suspended in 50 mmol/L phosphate buffer, pH 7.6. Purified thaumatin-like kiwi allergen, buffer, or BSA was added to the fungal culture and incubated at 37°C for 1 hour. After centrifugation (3 minutes at 3000g), supernatants were collected and analyzed by means of SDS-PAGE. Fungistatic testing on viable yeast cells was performed according to the method of Amsterdam.22 Cultures of Saccharomyces carlsbergensis and Candida albicans (ATCC 10259, 1 × 104 colony-forming units/mL) were maintained on Sabouraud dextrose liquid medium (Institute for Immunology and Virology,Yugoslavia) and subcultured to the same fresh medium before use in the assay. The fungistatic effect was monitored in 2 mL of liquid cultures containing 50 µg/mL isolated TLP. Cultures were started at an OD660nm of 0.01 to 0.05 from diluted overnight cultures incubated at 28°C with shaking for 24 hours, and the OD660nm reached was determined with appropriate dilutions.
Seven patients were recruited for in vivo and in vitro studies. Their clinical characteristics, total and specific IgE levels, and SPT results are given in Table I. All patient sera were tested individually by means of specific IgE immunodetection after SDS-PAGE of the crude self-prepared extract. Several IgE-binding bands ranging from 20 to 94 kd were detected (Fig 1). Patients 2 to 5 demonstrated similar IgE-binding patterns, differing only in their intensities. A component with a molecular weight of approximately 24 kd was recognized as the most prominent band in the kiwi extract by all 7 sera. Additionally, actinidin of approximately 30 kd, as well as bands of approximately 27 kd and 25 kd, were also recognized by all sera, although with different intensities. Homogeneity of the purified kiwi protein was checked by means of reductive SDS-PAGE and 2-dimensional PAGE, during which the allergen showed a single band of approximately 24 kd and 2 spots with the same molecular weight and isoelectric point (pI) of approximately 9.4 and 9.5 (Fig 2, lane c). These 2 isoforms migrated as one protein band of 20 kd after SDS-PAGE under nonreducing conditions or 24 kd under reducing conditions (Fig 2). The allergenicity of both isoforms was shown by the pool of patients’ sera in the 2-dimensional PAGE immunoblot (Fig 2, lane d). To further characterize the isolated allergen, its N-terminal amino acid sequence was determined, and the sequence was found to be A-T-F-N-I-I-N-N-?-P-F-?-V. A search of the Swiss Protein Bank (accession no. P81370) showed that this sequence corresponded to amino acids 1 to 13 of the TLP fragment. Moreover, thaumatin-like kiwi protein revealed a concanavalin A–binding ability (not shown). The in vitro assay for antifungal activity of the isolated thaumatin-like protein revealed protein leakage from the culture of S carlsbergensis cells. Biomolecules up to 70 kd were detected in the supernatant after 1 hour of incubation with thaumatin-like kiwi protein (Fig 3). The isolated TLP elicited positive SPT responses in 4 (80%) of 5 patients with kiwi allergy. In the experiment of digestion stability, isolated TLP was digested by the
FIG 1. Binding patterns of patients’ IgE to self-prepared kiwi fruit extract after Western blotting. Lane m, Molecular weight marker; lane a, protein stained by means of CBB stain; lanes 1-7, individual patient sera; lane b, buffer control; lane n, normal human sera.
FIG 2. SDS-PAGE of isolated TLP. Lane m, Molecular weight markers; lane a, separation in reducing conditions; lane b, separation in nonreducing conditions and 2-dimensional PAGE immunoblot of isolated TLP; lane c, nitrocellulose stained with Ponceau S; lane d, IgE reactivity of pool of patients’ sera.
SGF within 1 minute, whereas kiwi extract was decomposed after 8 minutes (not shown).
DISCUSSION There have been an increasing number of reports of allergic reactions to kiwi fruit, as well as reports of its association with hypersensitivity to tree and grass pollen.2-4 Thus far, actinidin, designated as Act c 1, has been characterized as a major kiwi allergen, and proteins with molecular weights of 28, 24, and 12 kd were designated as important allergens.12 Recently we6 reported that a protein with a molecular weight of approximately 24 kd was recognized by 8 patients with kiwi allergy, as well as by anti-grass group 4 mAb (2D8).
Food and drug reactions and anaphylaxis
RESULTS
808 Gavrovic´-Jankulovic´ et al
J ALLERGY CLIN IMMUNOL NOVEMBER 2002
FIG 3. Antifungal activity assay of TLP to brewer’s yeast cells. Lane m, Molecular weight markers; lane a, TLP (50 µg/mL); lane b, cells incubated with TLP (50 µg/mL); lane c, control; lane d, BSA (50 µg/mL); lane e, cells incubated with BSA (50 µg/mL).
TABLE II. N-terminal amino acid sequence of the 24-kd kiwi allergen aligned with the TLP fragment and several PR-5 proteins TLP
24 kd Kiwi Cherry Thaumatin II Osmotin Zeamatin
N-terminal amino acid sequence
A-T-F-N-I-I-N-N-?-P-F-?-V A-T-F-N-I-I-N-N-C-P-F-T-V A-T-I-S-F-K-N-N-C-P-Y-M-V A-T-F-E-I-V-N-R-C-S-T-T-V A-T-I-E-V-R-N-N-C-P-Y-T-V A-V-F-T-V-V-N-Q-C-P-F-T-V
Accession no.
P81370 P50694 P02884 P14170 P33679
Food and drug reactions and anaphylaxis
The isolation of proteins and glycoproteins in a native form is the prerequisite for extraction from food stuffs.23 In addition to the method of Bjorksten et al,24 a suitable but slightly demanding procedure for vegetable food extraction is the low-temperature acetone powder method developed by Vieths et al.25 Quality problems with allergen extracts from vegetables can be overcome with extraction procedures adapted to the specific source material.26 In our experiment, because of the high content of ascorbic acid in kiwi fruit, a bicarbonate buffer was used to mitigate pH lowering during protein extraction and to avoid proteolysis. Under these conditions, we obtained an extract for 24-kd protein isolation, which contained proteins with a broad range of molecular weights (94-17 kd). The new kiwi fruit allergen, isolated by means of classical biochemical methods, consists of 2 components with pI values of 9.4 and 9.5 and the same molecular weight (approximately 24 kd), according to SDS-PAGE and 2-dimensional PAGE in reducing conditions. Densitometric analysis showed that both isoforms are present in
equal amounts and bound IgE from a pool of patients’ sera. The N-terminal amino acid sequence of the new kiwi fruit allergen found in this study corresponds to the N-terminus of the TLP from kiwi.27 Table II shows alignment of this sequence with several other pathogenesisrelated (PR-5) proteins. Anomalous migration of PR proteins in SDS-PAGE has been previously reported28,29 and should be ascribed to the unusually high number of S-S bridges, a peculiarity common for TLPs. The conserved positions of 16 cysteine residues of the TLP sequences are critical for a common structure-function relationship.10 Higher-order plants accumulate PR proteins in response to infections by pathogens, such as fungi, bacteria, or viruses; wounding; or the application or chemicals that mimic the effect of such an infection or induce similar stresses.30 Their tissue-specific expression during development and consistent localization in the apoplast, as well as in the vacuolar compartment, and their differential induction by endogenous and exogenous signaling compounds suggest that PR proteins have important functions extending beyond a role in adaptation to biotic stress conditions.31 The family of PR-5 proteins comprises unique proteins with diverse functions, including antifungal activity. Accumulation of TLPs in ripening or ripe fruits has been reported.32,33 Because of the sequence homologies between PR-5 proteins and thaumatin, members of this family are referred to as TLPs,30 and they belong to the class of β proteins.34 PR-5 proteins have both acidic and basic isoforms: the acidic forms are secreted into the intercellular space, and the basic forms are sequestered in the vacuole.35 It seems that PR-5 protein from tobacco (osmotin) activates the mitogen-activated protein kinase signal transduction pathway to defeat the cell wall–based resistance of yeast, suggesting the presence of G protein–coupled osmotin receptors on the cell surface.36 When we found that the isolated allergen belongs to the family of PR-5 proteins, we were interested in revealing possible biologic functions (ie, examining whether it possesses antifungal activity), considering that cherry TLP showed no antifungal activity in preliminary assays.29 Antifungal activity of the isolated TLP from kiwi, expressed as protein leakage, was shown in culture of S carlsbergensis cells. Furthermore, the protein revealed a fungistatic effect (fungal growth was inhibited for approximately 20% compared with the control value) on viable cells of C albicans and S carlsbergensis (not shown). An explanation for the lack of antifungal activity observed for cherry TLP compared with the kiwi homologue could be the difference in the protein 3dimensional structure (ie, surface topology) because the first represents an acidic form (pI 4.2) and the second represents a basic form (pI 9.4-9.5) of TLPs. Because the mechanism of antifungal activity of TLPs at the molecular level is not fully clarified, it is too early to say whether their biologic function might contribute to allergic inflammation in the sense of plasma membrane permeabilization of effector cells (ie, mast cells or basophils).
Allergenic TLPs are known from apple (Malus domestica) as Mal d 2,33 cherry (Prunus avium) as Pru av 2,10 and bell pepper (Capsicum annuum) as Cap a 1.37 Mal d 2 and Pru av 2, designated as fruit group 2 allergens, represent, in addition to Bet v 1 homologues, LPTs, and class I chitinases, a new group of potential fruit panallergens.38 TLP from kiwi is the first fruit thaumatin-like allergen with proven antifungal activity. Kiwi fruit, as well as avocado, banana, and chestnut, are the foods most frequently associated with latex allergy and most prone to induce anaphylaxis. Class I chitinases are described as major panallergens in fruit associated with this syndrome, and its antifungal activity has been demonstrated.39 Considering that the new kiwi allergen also belongs to PR proteins with proven antifungal activity, further investigation should reveal the possible contribution of TLP to the latex-fruit syndrome. OAS after consumption of fruits is frequently associated with certain pollen allergies.3,40,41 Patients with OAS usually have IgE antibodies that cross-react with allergens or epitopes present in pollen, as well as plantderived food. Kazemi-Shirazi et al42 suggested that contact with pollen allergens through the respiratory system could induce, maintain, and boost the production of cross-reactive IgE antibodies that are responsible for plant food allergy in patients with OAS. Food allergens, which are able to react with IgE, induce allergic reactions, and sensitizations are considered to be complete allergens.43 Positive SPT responses with the TLP in 4 of 5 patients showed that the isolated allergen is capable of bridging cell-bound IgE on effector cells and triggering a histamine release. This feature makes this allergen a good candidate for component-resolved diagnosis of patients with kiwi allergy. Consensus among the scientific community seems to be that a protein that is resistant to proteolytic digestion in the digestive tract would retain sufficient structural integrity to have an increased probability of stimulating immune reactions.44 In vitro digestion assays provide an estimate of the relative integrity of a protein and thus the probability of eliciting allergic reactions. According to our findings, the isolated TLP is capable of binding IgE and eliciting an allergic reaction, but because of digestibility in SGF, such as Mal d 1,43 it might be considered a nonsensitizing elicitor or an incomplete food allergen. REFERENCES 1. Vocks E, Borga A, Szliska C, et al. Common allergenic structures in hazelnut, rye grain, sesame seeds, kiwi, and poppy seeds. Allergy 1993;48:168-72. 2. Gall H, Kalveram K-J, Forck G, Sterry W. Kiwi fruit allergy: a new birch pollen-associated food allergy. J Allergy Clin Immunol 1994;94:70-6. 3. Pastorelo EA, Pravettoni V, Ispano M, et al. Identification of the allergenic components of kiwi fruit and evaluation with timothy and birch pollens. J Allergy Clin Immunol 1996;98:601-10. 4. Voitenko V, Poulsen LK, Nielsen L, Norgaard A, Bindslev-Jensen C, Stahl Skov P. Allergenic properties of kiwi-fruit extract: cross-reactivity between kiwi-fruit and birch-pollen allergens. Allergy 1997;52:136-43. 5. Rudeschko O, Fahlbusch B, Steurich F, Schlenvoigt G, Jager L. Kiwi allergens and their cross-reactivity with birch, rye, timothy, and mugwort pollen. Invest Allergol Clin Immunol 1998;8:78-84.
Gavrovic´-Jankulovic´ et al 809
6. Gavrovic-Jankulovic M, Cirkovic T, Burazer L, Vuckovic O, Jankov RM. IgE cross-reactivity between meadow fescue pollen and kiwi fruit in patients’ sera with sensitivity to both extracts. J Invest Allergol Clin Immunol 2002. In press. 7. Becker W-M, Reese G. Immunological identification and characterization of individual food allergens. J Chromatogr B Biomed Sci Appl 2001;756:131-40. 8. Pastorello EA, Trambaioli C. Isolation of food allergens. J Chromatogr B Biomed Sci Appl 2001;756:71-84. 9. Valenta R, Vrtala S, Laffer S, Spitzauer S, Kraft D. Recombinant allergens. Allergy 1998;53:552-61. 10. Inschlag C, Hoffmann-Sommergruber K, O’Riordain G, et al. Biochemical characterization of Pru a 2, a 23 kD thaumatin like protein representing a potential major allergen in cherry (Prunus avlum). Int Arch Allergy Immunol 1998;116:22-8. 11. Scheurer S, Pastorello E, Wangorsch A, Kastner M, Haustein D, Vieths. Recombinant allergens Pru av 1 and Pru av 4 and a newly identified lipid transfer protein in the in vitro diagnosis of cherry allergy. J Allergy Clin Immunol 2001;107:724-31. 12. Pastorello EA, Conti A, Pravettoni V, et al. Identification of actinidin as the major allergen of kiwi fruit. J Allergy Clin Immunol 1998;101:531-7. 13. Bruijnzeel-Koomen C, Ortolani C, Aas K, et al. Adverse reaction to food. Allergy 1995;50:623-35. 14. Malling HJ. Methods of skin testing. In: Dreborg S, Frew A, editors. Allergen standardization and skin tests. Allergy 1993;48(suppl):55-6. 15. Lowry OH, Rosebrough NJ, Farr AL, Randal RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-75. 16. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-5. 17. Gavrovic-Jankulovic M, Cirkovic T, Bukilica M, Fahlbusch B, Petrovic S, Jankov RM. Isolation and partial characterization of Fes p 4 allergen. Invest Allergol Clin Immunol 2000;10:361-7. 18. Yagami T, Haishima Y, Nakamura A, Osuna H, Ikezawa Z. Digestibility of allergens extracted from natural rubber latex and vegetable foods. J Allergy Clin Immunol 2000;106:752-62. 19. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc Natl Acad Sci USA 1979;76:4350-4. 20. Harlow E, Lane D. Antibodies: a laboratory manual. 1st ed. New York: Cold Spring Harbor Laboratory; 1988. p. 473-510. 21. Liu D, Ragotham KG, Hasegawa PM, Bressan RA. Osmotin overexpression in potato delays development of disease symptoms. Proc Natl Acad Sci U S A 1994;91:1888-92. 22. Amsterdam D. Susceptibility testing of antimicrobials in liquid media. In: Lorian V, editor. Antibiotics in laboratory medicine. 4th ed. Baltimore: Williams & Wilkins; 1996. p. 52. 23. Steinhart H. Wigotzki M, Zunker K. Introducing allergists to food chemistry. Allergy 2001;56(suppl 67):9-11. 24. Bjorksten F, Halmepuro L, Hannuksela M, Lahti A. Extraction and properties of apple allergens. Allergy 1980;35:671-7. 25. Vieths S, Schoning A, Petersen A. Characterization of the 18-kDa apple allergen by two-dimensional immunoblotting and microsequencing. Int Arch Allergy Immunol 1994;104:399-404. 26. Vieths S, Scheurer S, Reindl J, et al. Optimized allergen extracts and recombinant allergens in diagnostic applications. Allergy 2001;56(suppl 67):78-82. 27. Wurms K, Greenwood D, Sharrock K, Long P. Thaumatin-like protein in kiwifruit. J Sci Food Agric 1999;79:1448-52. 28. Cusack M, Pierpoint WS. Similarities between sweet protein thaumatin and a pathogenesis-related protein from tobacco. Phytochemistry 1988;27:3817-21. 29. Fils-Lycaon BR, Wiersma PA, Eastwell KC, Sautiere P. A cherry protein and its gene, abundantly expressed in ripening fruit, have been identified as thaumatin-like. Plant Physiol 1996;111:269-73. 30. Breiteneder H, Ebner C. Molecular and biochemical classification of plant-derived food allergens. J Allergy Clin Immunol 2000;106:27-36. 31. van Loon LC, van Strien EA. The families of pathogenesis-related proteins, their activities, and comparative analysis of PR-1 type proteins. Physiol Mol Plant Pathol 1999;55:85-97. 32. Pressey R. Two isoforms of NP24: a thaumatin-like protein in tomato fruit. Phytochemistry 1997;44:1241-5. 33. Hsich LS, Moos M, Lin Y. Characterization of apple 18 and 31 kd aller-
Food and drug reactions and anaphylaxis
J ALLERGY CLIN IMMUNOL VOLUME 110, NUMBER 5
810 Gavrovic´-Jankulovic´ et al
34. 35. 36.
37.
38. 39.
gens by microsequencing and evaluation of their content during storage and ripening. J Allergy Clin Immunol 1995;96:960-70. Hubbard TJP, Ailey B, Brenner SE, Murzin A, Chothia C. SCOP: a Structural Classification of Proteins database. Nucleic Acids Res 1999;27:254-6. Bol JF, Linthorst HJM, Cornelissen NJC. Plant pathogenesis-related proteins induced by virus infection. Annu Rev Phytopathol 1990;28:113-38. Yun D-J, Zhao Y, Pardo JM, et al. Stress proteins on the yeast cell surface determine resistance to osmotin, a plant antifungal protein. Proc Natl Acad Sci U S A 1997;94:7082-7. Jensen-Jarolim E, Santner B, Leitner A, et al. Bell peppers (Capsicum annuum) express allergens (profilin, pathogenesis-related protein P23 and Bet v 1) depending on the horticultural strain. Int Arch Allergy Immunol 1998;116:103-9. Salcedo G, Diaz-Perales A, Sanchez-Monge R. Fruit allergy: plant defence proteins as novel potential panallergens. Clin Exp Allergy 1999;29:1158-60. Sanchez-Monge R, Blanco C, Diaz-Perales A, et al. Class I chitinases, the panallergens responsible for the latex-fruit syndrome, are induced by eth-
J ALLERGY CLIN IMMUNOL NOVEMBER 2002
40.
41.
42.
43. 44.
ylene treatment and inactivated by heating. J Allergy Clin Immunol 2000;106:190-5. van Ree R, Fernandez-Rivas M, Cuevas M, van Wijngaarden M, Aalberse RC. Pollen-related allergy to peach and apple: An important role of profilin. J Allergy Clin Immunol 1995;95:726-34. Fritsch R, Bohle B, Vollmann U, et al. Bet v 1, the major birch pollen allergen, and Mal d 1, the major apple allergen, cross-react at the level of allergen-specific T helper cells. J Allergy Clin Immunol 1998; 102:679-86. Kazemi-Shirazi L, Pauli G, Purohit A, et al. Quantitative IgE inhibition experiments with purified recombinant allergens indicate pollen-derived allergens as the sensitizing agents responsible for many forms of plant food allergy. J Allergy Clin Immunol 1999;105:116-25. Aalberse RC. Structural biology of allergens. J Allergy Clin Immunol 2000;106:228-38. Fu T-J. Digestion stability as a criterion for protein allergenicity assessment. Ann N Y Acad Sci 2002;964:99-110.
Food and drug reactions and anaphylaxis
From athe Department of Biochemistry, Faculty of Chemistry, University of Belgrade, Belgrade; bthe Department of Allergology, Institute for Immunology and Virology Torlak, Belgrade; cthe Department of Allergology, University Children’s Hospital Belgrade; and dResearch Center Borstel, Biochemical and Molecular Allergology, Borstel. Supported by grant No. 1802 from the Ministry of Science, Technologies, and Development, Republic of Serbia. Received for publication April 26, 2002; revised July 29, 2002; accepted for publication August 13, 2002. Reprint requests: Marija Gavrovic´-Jankulovic´, PhD, Faculty of Chemistry, Department of Biochemistry, Studentski trg 16, 11000 Belgrade, Yugoslavia. © 2002 Mosby, Inc. All rights reserved. 0091-6749/2002 $35.00 + 0 1/81/128947 doi:10.1067/mai.2002.128947
9.5 and molecular weight of 24 kd), which belongs to the family of pathogenesis-related proteins. The isolated protein expressed antifungal activity toward S carlsbergensis and C albicans. (J Allergy Clin Immunol 2002;110:805-10.) Key words: Allergen purification, antifungal activity, food allergen, kiwi fruit, pathogenesis-related protein, thaumatin-like protein
Kiwi fruit allergy, together with its association with hypersensitivity to other foods and to tree and grass pollen, has been widely reported in the last few years. The clinical symptoms range from localized symptoms confined to the oral mucosa to severe anaphylactic reactions. Common allergenic structures have been identified in kiwi and hazel nuts, birch, timothy, rye, mugwort, and fescue meadow pollen.1-6 It is important to identify allergenic molecules to improve the quality of allergen extracts.7 Identification of food allergens is a priority in the management of food allergy8 because well-characterized relevant allergens might replace allergen extracts in a component-based allergy diagnosis. The rapid progress made in the field of molecular allergen characterization appears likely to considerably improve the use of recombinant allergens in diagnosis and specific immunotherapy.9 Although loss of allergenicity was reported for recombinant thaumatin-like cherry protein, probably because of incorrect folding in the polypeptide chain,10 Scheurer et al11 indicated that it is possible to replace cherry extract with 3 recombinant allergens without affecting the results of in vitro diagnosis. However, the isolation and characterization of natural allergens is indispensable in the creation of their structural, as well as functional, recombinant counterparts. Namely, researchers need to compare biochemical and immunologic characteristics of recombinant molecules with those of native molecules to verify that molecular expression in heterologous systems does not significantly modify the IgE response pattern, which would make recombinant proteins useless for diagnostic purposes.8 Although several studies reported that specific IgE immunodetection revealed approximately 12 allergens in kiwi extract, only one 30-kd allergen has been isolated; it was identified as actinidin (Act c 1).12 Recently, we reported that a 24-kd kiwi protein was a potential major allergen in a group of 8 patients with oral allergy syndrome (OAS).6 805
Food and drug reactions and anaphylaxis
Background: Kiwi fruit allergy, as well as its association with hypersensitivity to other foods and to pollen, has been extensively reported in the last few years. Several IgE-binding components have been detected in kiwi extract, but only one 30-kd allergen has been isolated; it was identified as actinidin (Act c 1). Recently, we have reported a 24-kd kiwi protein to be a potential major allergen in a group of patients with oral allergy syndrome (OAS). Objective: The aim of this study was to purify and characterize the 24-kd kiwi allergen biochemically. Methods: Seven polysensitized patients with OAS to kiwi were used in this study. The kiwi allergen was isolated by using a combination of gel permeation, ion exchange, and immobilized metal ion affinity chromatography. Its biochemical characterization included determination of its isoelectric point, molecular weight, N-terminal sequencing, concanavalin A–binding ability, digestibility in simulated gastric fluid, and antifungal activity. Western blotting, 2-dimensional PAGE immunoblotting, and skin prick tests were performed to characterize the isolated protein immunochemically. Results: All 7 patients recognized the isolated 24-kd kiwi protein as an allergen. The isolated protein consisted of 2 isoforms with isoelectric points of 9.4 and 9.5 migrated as one protein band of 20 kd after SDS-PAGE under nonreducing conditions or at 24 kd under reducing conditions. The partial N-terminal sequence revealed that it is a thaumatin-like protein (TLP) with concanavalin A–binding ability. The protein showed antifungal activity toward Saccharomyces carlsbergensis, and Candida albicans. The protein was degraded by the simulated gastric fluid within 1 minute. Both isoforms bound IgE from a pool of sera in a 2-dimensional PAGE immunoblot. The TLP elicited positive skin prick test responses in 4 (80%) of 5 patients with OAS. Conclusion: This study reported isolation and full characterization of a new kiwi allergen, TLP (isoelectric points of 9.4 and
806 Gavrovic´-Jankulovic´ et al
J ALLERGY CLIN IMMUNOL NOVEMBER 2002
Abbreviations used OAS: Oral allergy syndrome pI: Isoelectric point PR: Pathogenesis related SGF: Simulated gastric fluid SPT: Skin prick test TLP: Thaumatin-like protein
TABLE I. Clinical data, total and specific IgE levels, and skin test results of the patients allergic to kiwi Patient no.
Kiwi symptoms
Total IgE (kU/L)
Specific IgE (kUA/L)
1 2 3 4 5 6 7
OS/RC OS/BA OS/RC AE U/A OS/RC OS/RC
797 2709 382 732 ND 910 151
0 2.89 0 0.90 ND 0.71 0.70
Kiwi TLP SPT* SPT*
16 16 7 16 36 ND ND
7 7 — 7 36 ND ND
OS, Oral syndrome; RC, rhinoconjunctivitis; BA, bronchial asthma; AE, angioedema; U, urticaria; A, systemic anaphylaxis; ND, not determined. *SPT results are given in square millimeters of wheal area.
The aim of this study was to isolate and characterize the 24-kd kiwi allergen.
METHODS Patient sera We have studied seven patients who had symptoms suggestive of an IgE-mediated reaction after the ingestion of kiwi, which are summarized in Table I. All the patients tolerated other kinds of exotic fruits, and some of them had airborne allergen sensitization to grass pollens. Open challenge with kiwi was performed according to instructions given in an article by the European Academy of Allergology and Clinical Immunology.13 All sera had positive immunoblotting results with self-prepared kiwi extract. In addition, a pool of sera from 3 subjects (nonatopic) was used as the negative control.
Skin tests
Food and drug reactions and anaphylaxis
Skin prick tests (SPTs) with the kiwi extract (500 µg/mL) and the purified allergen (30 µg/mL) were performed according to the standard procedure14 in the volar side of the forearm. Histamine phosphate (10 mg/mL) and PBS were used as positive and negative controls, respectively. A mean wheal area of 7 mm2 (diameter ≥3 mm) or greater compared with that produced by the negative control 20 minutes after puncture was considered a positive response.
Total and specific serum IgE determination The patients’ total and specific IgE levels were determined by using the CAP System IgE FEIA (Pharmacia Diagnostics), according to the manufacturer’s instructions. Levels of specific IgE of greater than 0.35 kU/L (≥class 1) were considered positive.
Isolation and characterization of the allergen Kiwi fruit (Actinidia chinensis) purchased in a local store was used as the starting material for allergen purification. Peeled kiwi (50 g) was cut into pieces, and then 100 mmol/L sodium bicarbon-
ate buffer containing 2% (wt/vol) polyvinylpolypyrrolidone, pH 9.3 (1:2 wt/vol), was poured over the kiwi, followed by homogenization in a blender for 1 minute. After extraction (3 hours at 4°C), the slurry was passed through a cheesecloth and centrifuged (15 minutes at 10,000g). The resulting protein content was 0.75 mg/mL. Ammonium sulfate was dissolved in the extract (100 mL) to achieve 60% saturation. After overnight standing (4°C), the solution was centrifuged (15 minutes at 10,000g). The obtained pellet was dissolved in a minimal volume of the starting buffer and dialyzed extensively against the same buffer. The dialysate was spun and subjected to gel permeation chromatography on a Bio-Gel P-60 (Bio Rad) column (39 × 1.2 cm). Enriched protein fractions were pooled, concentrated twice, and subjected to ion-exchange chromatography on a QAE-Sephadex A-50 column (8 × 1 cm) pre-equilibrated with 100 mmol/L sodium bicarbonate buffer, pH 9.3. The unbound protein fractions were pooled, concentrated twice by means of ultrafiltration, and dialyzed against 150 mmol/L NaCl in 50 mmol/L sodium acetate buffer, pH 6.0, overnight. The protein sample was applied on a copper iminodiacetate Sepharose CL 4B (Cu2+-IDA-Sepharose) column (3.5 × 0.7 cm). Finally, the unbound protein sample was rechromatographed on a Superdex 75 prep grade (Pharmacia) column (30 × 1.2 cm). The distribution of the protein through the various chromatographic steps was monitored by means of SDSPAGE. The amount of purified protein quantified according to the method of Lowry and colleagues15 was 6 mg from 75 mg of starting material. Analytic SDS-PAGE was performed on a polyacrylamide gel (4% stacking gel and 10% resolving gel), according to the method of Laemmli.16 After electrophoresis, proteins were either stained with CBB or semidry transferred (1 mA/cm2) to a nitrocellulose membrane (Serva) for further examination. In 2-dimensional PAGE isoelectric focusing was performed in the model 2117 Multiphor cell (LKB Pharmacia), according to the manufacturer’s instructions. The protein sample (8 µg) was separated under conditions, as previously described.17 The resolved thaumatin-like kiwi protein in 2-dimensional PAGE was blotted to a polyvinylidene difluoride membrane and visualized by staining the blot with 0.1% CBB G-250, followed by destaining with several changes of 50% methanol. Spots with thaumatin-like kiwi protein were excised, and automated Edman degradation was performed. A concanavalin A–horseradish peroxidase complex was used to detect carbohydrate moieties.17
Digestion with simulated gastric fluid The digestibility of the self-prepared kiwi extract and isolated thaumatin-like protein (TLP) in the simulated gastric fluid (SGF) was examined according to the method of Yagami et al.18 Briefly, a protein sample (700 µg of crude proteins or 40 µg of TLP) was dissolved in 200 µL of prewarmed SGF (US Pharmacopoeia) containing 0.32% wt/vol percentage of pepsin A (Sigma Chemical Co). Digestion proceeded at 37°C with continuous shaking, and an aliquot (20 µL) of the digest was periodically withdrawn (at 0.5, 1, 2, 4, 8, 16, and 60 minutes). The digestion was stopped with 0.2 mol/L Na2CO3 (6.0 µL), and samples were mixed with a sample buffer for SDS-PAGE analysis.
Immunodetection The protein samples were separated by means of SDS-PAGE (4% stacking gel and 10% resolving gel) with reducing conditions and then electrotransferred onto a nitrocellulose membrane, as described by Towbin et al.19 Immunodetection was performed according to the method of Harlow and Lane.20 In 2-dimensional PAGE immunoblotting, a pool of 7 sera was used for detection of IgE-binding components.
Gavrovic´-Jankulovic´ et al 807
J ALLERGY CLIN IMMUNOL VOLUME 110, NUMBER 5
In vitro antifungal activity assay The antifungal activity assay was performed in vitro according to the method of Liu et al,21 with a modification. A culture of S carlsbergensis (0.2 mg/mL) was washed and suspended in 50 mmol/L phosphate buffer, pH 7.6. Purified thaumatin-like kiwi allergen, buffer, or BSA was added to the fungal culture and incubated at 37°C for 1 hour. After centrifugation (3 minutes at 3000g), supernatants were collected and analyzed by means of SDS-PAGE. Fungistatic testing on viable yeast cells was performed according to the method of Amsterdam.22 Cultures of Saccharomyces carlsbergensis and Candida albicans (ATCC 10259, 1 × 104 colony-forming units/mL) were maintained on Sabouraud dextrose liquid medium (Institute for Immunology and Virology,Yugoslavia) and subcultured to the same fresh medium before use in the assay. The fungistatic effect was monitored in 2 mL of liquid cultures containing 50 µg/mL isolated TLP. Cultures were started at an OD660nm of 0.01 to 0.05 from diluted overnight cultures incubated at 28°C with shaking for 24 hours, and the OD660nm reached was determined with appropriate dilutions.
Seven patients were recruited for in vivo and in vitro studies. Their clinical characteristics, total and specific IgE levels, and SPT results are given in Table I. All patient sera were tested individually by means of specific IgE immunodetection after SDS-PAGE of the crude self-prepared extract. Several IgE-binding bands ranging from 20 to 94 kd were detected (Fig 1). Patients 2 to 5 demonstrated similar IgE-binding patterns, differing only in their intensities. A component with a molecular weight of approximately 24 kd was recognized as the most prominent band in the kiwi extract by all 7 sera. Additionally, actinidin of approximately 30 kd, as well as bands of approximately 27 kd and 25 kd, were also recognized by all sera, although with different intensities. Homogeneity of the purified kiwi protein was checked by means of reductive SDS-PAGE and 2-dimensional PAGE, during which the allergen showed a single band of approximately 24 kd and 2 spots with the same molecular weight and isoelectric point (pI) of approximately 9.4 and 9.5 (Fig 2, lane c). These 2 isoforms migrated as one protein band of 20 kd after SDS-PAGE under nonreducing conditions or 24 kd under reducing conditions (Fig 2). The allergenicity of both isoforms was shown by the pool of patients’ sera in the 2-dimensional PAGE immunoblot (Fig 2, lane d). To further characterize the isolated allergen, its N-terminal amino acid sequence was determined, and the sequence was found to be A-T-F-N-I-I-N-N-?-P-F-?-V. A search of the Swiss Protein Bank (accession no. P81370) showed that this sequence corresponded to amino acids 1 to 13 of the TLP fragment. Moreover, thaumatin-like kiwi protein revealed a concanavalin A–binding ability (not shown). The in vitro assay for antifungal activity of the isolated thaumatin-like protein revealed protein leakage from the culture of S carlsbergensis cells. Biomolecules up to 70 kd were detected in the supernatant after 1 hour of incubation with thaumatin-like kiwi protein (Fig 3). The isolated TLP elicited positive SPT responses in 4 (80%) of 5 patients with kiwi allergy. In the experiment of digestion stability, isolated TLP was digested by the
FIG 1. Binding patterns of patients’ IgE to self-prepared kiwi fruit extract after Western blotting. Lane m, Molecular weight marker; lane a, protein stained by means of CBB stain; lanes 1-7, individual patient sera; lane b, buffer control; lane n, normal human sera.
FIG 2. SDS-PAGE of isolated TLP. Lane m, Molecular weight markers; lane a, separation in reducing conditions; lane b, separation in nonreducing conditions and 2-dimensional PAGE immunoblot of isolated TLP; lane c, nitrocellulose stained with Ponceau S; lane d, IgE reactivity of pool of patients’ sera.
SGF within 1 minute, whereas kiwi extract was decomposed after 8 minutes (not shown).
DISCUSSION There have been an increasing number of reports of allergic reactions to kiwi fruit, as well as reports of its association with hypersensitivity to tree and grass pollen.2-4 Thus far, actinidin, designated as Act c 1, has been characterized as a major kiwi allergen, and proteins with molecular weights of 28, 24, and 12 kd were designated as important allergens.12 Recently we6 reported that a protein with a molecular weight of approximately 24 kd was recognized by 8 patients with kiwi allergy, as well as by anti-grass group 4 mAb (2D8).
Food and drug reactions and anaphylaxis
RESULTS
808 Gavrovic´-Jankulovic´ et al
J ALLERGY CLIN IMMUNOL NOVEMBER 2002
FIG 3. Antifungal activity assay of TLP to brewer’s yeast cells. Lane m, Molecular weight markers; lane a, TLP (50 µg/mL); lane b, cells incubated with TLP (50 µg/mL); lane c, control; lane d, BSA (50 µg/mL); lane e, cells incubated with BSA (50 µg/mL).
TABLE II. N-terminal amino acid sequence of the 24-kd kiwi allergen aligned with the TLP fragment and several PR-5 proteins TLP
24 kd Kiwi Cherry Thaumatin II Osmotin Zeamatin
N-terminal amino acid sequence
A-T-F-N-I-I-N-N-?-P-F-?-V A-T-F-N-I-I-N-N-C-P-F-T-V A-T-I-S-F-K-N-N-C-P-Y-M-V A-T-F-E-I-V-N-R-C-S-T-T-V A-T-I-E-V-R-N-N-C-P-Y-T-V A-V-F-T-V-V-N-Q-C-P-F-T-V
Accession no.
P81370 P50694 P02884 P14170 P33679
Food and drug reactions and anaphylaxis
The isolation of proteins and glycoproteins in a native form is the prerequisite for extraction from food stuffs.23 In addition to the method of Bjorksten et al,24 a suitable but slightly demanding procedure for vegetable food extraction is the low-temperature acetone powder method developed by Vieths et al.25 Quality problems with allergen extracts from vegetables can be overcome with extraction procedures adapted to the specific source material.26 In our experiment, because of the high content of ascorbic acid in kiwi fruit, a bicarbonate buffer was used to mitigate pH lowering during protein extraction and to avoid proteolysis. Under these conditions, we obtained an extract for 24-kd protein isolation, which contained proteins with a broad range of molecular weights (94-17 kd). The new kiwi fruit allergen, isolated by means of classical biochemical methods, consists of 2 components with pI values of 9.4 and 9.5 and the same molecular weight (approximately 24 kd), according to SDS-PAGE and 2-dimensional PAGE in reducing conditions. Densitometric analysis showed that both isoforms are present in
equal amounts and bound IgE from a pool of patients’ sera. The N-terminal amino acid sequence of the new kiwi fruit allergen found in this study corresponds to the N-terminus of the TLP from kiwi.27 Table II shows alignment of this sequence with several other pathogenesisrelated (PR-5) proteins. Anomalous migration of PR proteins in SDS-PAGE has been previously reported28,29 and should be ascribed to the unusually high number of S-S bridges, a peculiarity common for TLPs. The conserved positions of 16 cysteine residues of the TLP sequences are critical for a common structure-function relationship.10 Higher-order plants accumulate PR proteins in response to infections by pathogens, such as fungi, bacteria, or viruses; wounding; or the application or chemicals that mimic the effect of such an infection or induce similar stresses.30 Their tissue-specific expression during development and consistent localization in the apoplast, as well as in the vacuolar compartment, and their differential induction by endogenous and exogenous signaling compounds suggest that PR proteins have important functions extending beyond a role in adaptation to biotic stress conditions.31 The family of PR-5 proteins comprises unique proteins with diverse functions, including antifungal activity. Accumulation of TLPs in ripening or ripe fruits has been reported.32,33 Because of the sequence homologies between PR-5 proteins and thaumatin, members of this family are referred to as TLPs,30 and they belong to the class of β proteins.34 PR-5 proteins have both acidic and basic isoforms: the acidic forms are secreted into the intercellular space, and the basic forms are sequestered in the vacuole.35 It seems that PR-5 protein from tobacco (osmotin) activates the mitogen-activated protein kinase signal transduction pathway to defeat the cell wall–based resistance of yeast, suggesting the presence of G protein–coupled osmotin receptors on the cell surface.36 When we found that the isolated allergen belongs to the family of PR-5 proteins, we were interested in revealing possible biologic functions (ie, examining whether it possesses antifungal activity), considering that cherry TLP showed no antifungal activity in preliminary assays.29 Antifungal activity of the isolated TLP from kiwi, expressed as protein leakage, was shown in culture of S carlsbergensis cells. Furthermore, the protein revealed a fungistatic effect (fungal growth was inhibited for approximately 20% compared with the control value) on viable cells of C albicans and S carlsbergensis (not shown). An explanation for the lack of antifungal activity observed for cherry TLP compared with the kiwi homologue could be the difference in the protein 3dimensional structure (ie, surface topology) because the first represents an acidic form (pI 4.2) and the second represents a basic form (pI 9.4-9.5) of TLPs. Because the mechanism of antifungal activity of TLPs at the molecular level is not fully clarified, it is too early to say whether their biologic function might contribute to allergic inflammation in the sense of plasma membrane permeabilization of effector cells (ie, mast cells or basophils).
Allergenic TLPs are known from apple (Malus domestica) as Mal d 2,33 cherry (Prunus avium) as Pru av 2,10 and bell pepper (Capsicum annuum) as Cap a 1.37 Mal d 2 and Pru av 2, designated as fruit group 2 allergens, represent, in addition to Bet v 1 homologues, LPTs, and class I chitinases, a new group of potential fruit panallergens.38 TLP from kiwi is the first fruit thaumatin-like allergen with proven antifungal activity. Kiwi fruit, as well as avocado, banana, and chestnut, are the foods most frequently associated with latex allergy and most prone to induce anaphylaxis. Class I chitinases are described as major panallergens in fruit associated with this syndrome, and its antifungal activity has been demonstrated.39 Considering that the new kiwi allergen also belongs to PR proteins with proven antifungal activity, further investigation should reveal the possible contribution of TLP to the latex-fruit syndrome. OAS after consumption of fruits is frequently associated with certain pollen allergies.3,40,41 Patients with OAS usually have IgE antibodies that cross-react with allergens or epitopes present in pollen, as well as plantderived food. Kazemi-Shirazi et al42 suggested that contact with pollen allergens through the respiratory system could induce, maintain, and boost the production of cross-reactive IgE antibodies that are responsible for plant food allergy in patients with OAS. Food allergens, which are able to react with IgE, induce allergic reactions, and sensitizations are considered to be complete allergens.43 Positive SPT responses with the TLP in 4 of 5 patients showed that the isolated allergen is capable of bridging cell-bound IgE on effector cells and triggering a histamine release. This feature makes this allergen a good candidate for component-resolved diagnosis of patients with kiwi allergy. Consensus among the scientific community seems to be that a protein that is resistant to proteolytic digestion in the digestive tract would retain sufficient structural integrity to have an increased probability of stimulating immune reactions.44 In vitro digestion assays provide an estimate of the relative integrity of a protein and thus the probability of eliciting allergic reactions. According to our findings, the isolated TLP is capable of binding IgE and eliciting an allergic reaction, but because of digestibility in SGF, such as Mal d 1,43 it might be considered a nonsensitizing elicitor or an incomplete food allergen. REFERENCES 1. Vocks E, Borga A, Szliska C, et al. Common allergenic structures in hazelnut, rye grain, sesame seeds, kiwi, and poppy seeds. Allergy 1993;48:168-72. 2. Gall H, Kalveram K-J, Forck G, Sterry W. Kiwi fruit allergy: a new birch pollen-associated food allergy. J Allergy Clin Immunol 1994;94:70-6. 3. Pastorelo EA, Pravettoni V, Ispano M, et al. Identification of the allergenic components of kiwi fruit and evaluation with timothy and birch pollens. J Allergy Clin Immunol 1996;98:601-10. 4. Voitenko V, Poulsen LK, Nielsen L, Norgaard A, Bindslev-Jensen C, Stahl Skov P. Allergenic properties of kiwi-fruit extract: cross-reactivity between kiwi-fruit and birch-pollen allergens. Allergy 1997;52:136-43. 5. Rudeschko O, Fahlbusch B, Steurich F, Schlenvoigt G, Jager L. Kiwi allergens and their cross-reactivity with birch, rye, timothy, and mugwort pollen. Invest Allergol Clin Immunol 1998;8:78-84.
Gavrovic´-Jankulovic´ et al 809
6. Gavrovic-Jankulovic M, Cirkovic T, Burazer L, Vuckovic O, Jankov RM. IgE cross-reactivity between meadow fescue pollen and kiwi fruit in patients’ sera with sensitivity to both extracts. J Invest Allergol Clin Immunol 2002. In press. 7. Becker W-M, Reese G. Immunological identification and characterization of individual food allergens. J Chromatogr B Biomed Sci Appl 2001;756:131-40. 8. Pastorello EA, Trambaioli C. Isolation of food allergens. J Chromatogr B Biomed Sci Appl 2001;756:71-84. 9. Valenta R, Vrtala S, Laffer S, Spitzauer S, Kraft D. Recombinant allergens. Allergy 1998;53:552-61. 10. Inschlag C, Hoffmann-Sommergruber K, O’Riordain G, et al. Biochemical characterization of Pru a 2, a 23 kD thaumatin like protein representing a potential major allergen in cherry (Prunus avlum). Int Arch Allergy Immunol 1998;116:22-8. 11. Scheurer S, Pastorello E, Wangorsch A, Kastner M, Haustein D, Vieths. Recombinant allergens Pru av 1 and Pru av 4 and a newly identified lipid transfer protein in the in vitro diagnosis of cherry allergy. J Allergy Clin Immunol 2001;107:724-31. 12. Pastorello EA, Conti A, Pravettoni V, et al. Identification of actinidin as the major allergen of kiwi fruit. J Allergy Clin Immunol 1998;101:531-7. 13. Bruijnzeel-Koomen C, Ortolani C, Aas K, et al. Adverse reaction to food. Allergy 1995;50:623-35. 14. Malling HJ. Methods of skin testing. In: Dreborg S, Frew A, editors. Allergen standardization and skin tests. Allergy 1993;48(suppl):55-6. 15. Lowry OH, Rosebrough NJ, Farr AL, Randal RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-75. 16. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-5. 17. Gavrovic-Jankulovic M, Cirkovic T, Bukilica M, Fahlbusch B, Petrovic S, Jankov RM. Isolation and partial characterization of Fes p 4 allergen. Invest Allergol Clin Immunol 2000;10:361-7. 18. Yagami T, Haishima Y, Nakamura A, Osuna H, Ikezawa Z. Digestibility of allergens extracted from natural rubber latex and vegetable foods. J Allergy Clin Immunol 2000;106:752-62. 19. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc Natl Acad Sci USA 1979;76:4350-4. 20. Harlow E, Lane D. Antibodies: a laboratory manual. 1st ed. New York: Cold Spring Harbor Laboratory; 1988. p. 473-510. 21. Liu D, Ragotham KG, Hasegawa PM, Bressan RA. Osmotin overexpression in potato delays development of disease symptoms. Proc Natl Acad Sci U S A 1994;91:1888-92. 22. Amsterdam D. Susceptibility testing of antimicrobials in liquid media. In: Lorian V, editor. Antibiotics in laboratory medicine. 4th ed. Baltimore: Williams & Wilkins; 1996. p. 52. 23. Steinhart H. Wigotzki M, Zunker K. Introducing allergists to food chemistry. Allergy 2001;56(suppl 67):9-11. 24. Bjorksten F, Halmepuro L, Hannuksela M, Lahti A. Extraction and properties of apple allergens. Allergy 1980;35:671-7. 25. Vieths S, Schoning A, Petersen A. Characterization of the 18-kDa apple allergen by two-dimensional immunoblotting and microsequencing. Int Arch Allergy Immunol 1994;104:399-404. 26. Vieths S, Scheurer S, Reindl J, et al. Optimized allergen extracts and recombinant allergens in diagnostic applications. Allergy 2001;56(suppl 67):78-82. 27. Wurms K, Greenwood D, Sharrock K, Long P. Thaumatin-like protein in kiwifruit. J Sci Food Agric 1999;79:1448-52. 28. Cusack M, Pierpoint WS. Similarities between sweet protein thaumatin and a pathogenesis-related protein from tobacco. Phytochemistry 1988;27:3817-21. 29. Fils-Lycaon BR, Wiersma PA, Eastwell KC, Sautiere P. A cherry protein and its gene, abundantly expressed in ripening fruit, have been identified as thaumatin-like. Plant Physiol 1996;111:269-73. 30. Breiteneder H, Ebner C. Molecular and biochemical classification of plant-derived food allergens. J Allergy Clin Immunol 2000;106:27-36. 31. van Loon LC, van Strien EA. The families of pathogenesis-related proteins, their activities, and comparative analysis of PR-1 type proteins. Physiol Mol Plant Pathol 1999;55:85-97. 32. Pressey R. Two isoforms of NP24: a thaumatin-like protein in tomato fruit. Phytochemistry 1997;44:1241-5. 33. Hsich LS, Moos M, Lin Y. Characterization of apple 18 and 31 kd aller-
Food and drug reactions and anaphylaxis
J ALLERGY CLIN IMMUNOL VOLUME 110, NUMBER 5
810 Gavrovic´-Jankulovic´ et al
34. 35. 36.
37.
38. 39.
gens by microsequencing and evaluation of their content during storage and ripening. J Allergy Clin Immunol 1995;96:960-70. Hubbard TJP, Ailey B, Brenner SE, Murzin A, Chothia C. SCOP: a Structural Classification of Proteins database. Nucleic Acids Res 1999;27:254-6. Bol JF, Linthorst HJM, Cornelissen NJC. Plant pathogenesis-related proteins induced by virus infection. Annu Rev Phytopathol 1990;28:113-38. Yun D-J, Zhao Y, Pardo JM, et al. Stress proteins on the yeast cell surface determine resistance to osmotin, a plant antifungal protein. Proc Natl Acad Sci U S A 1997;94:7082-7. Jensen-Jarolim E, Santner B, Leitner A, et al. Bell peppers (Capsicum annuum) express allergens (profilin, pathogenesis-related protein P23 and Bet v 1) depending on the horticultural strain. Int Arch Allergy Immunol 1998;116:103-9. Salcedo G, Diaz-Perales A, Sanchez-Monge R. Fruit allergy: plant defence proteins as novel potential panallergens. Clin Exp Allergy 1999;29:1158-60. Sanchez-Monge R, Blanco C, Diaz-Perales A, et al. Class I chitinases, the panallergens responsible for the latex-fruit syndrome, are induced by eth-
J ALLERGY CLIN IMMUNOL NOVEMBER 2002
40.
41.
42.
43. 44.
ylene treatment and inactivated by heating. J Allergy Clin Immunol 2000;106:190-5. van Ree R, Fernandez-Rivas M, Cuevas M, van Wijngaarden M, Aalberse RC. Pollen-related allergy to peach and apple: An important role of profilin. J Allergy Clin Immunol 1995;95:726-34. Fritsch R, Bohle B, Vollmann U, et al. Bet v 1, the major birch pollen allergen, and Mal d 1, the major apple allergen, cross-react at the level of allergen-specific T helper cells. J Allergy Clin Immunol 1998; 102:679-86. Kazemi-Shirazi L, Pauli G, Purohit A, et al. Quantitative IgE inhibition experiments with purified recombinant allergens indicate pollen-derived allergens as the sensitizing agents responsible for many forms of plant food allergy. J Allergy Clin Immunol 1999;105:116-25. Aalberse RC. Structural biology of allergens. J Allergy Clin Immunol 2000;106:228-38. Fu T-J. Digestion stability as a criterion for protein allergenicity assessment. Ann N Y Acad Sci 2002;964:99-110.
Food and drug reactions and anaphylaxis