Parvalbumin, a cross-reactive fish allergen, contains IgE-binding epitopes sensitive to

Transcription

1 Parvalbumin, a cross-reactive fish allergen, contains IgE-binding epitopes sensitive to periodate treatment and Ca 2 depletion Agnes Bugajska-Schretter, MSc, a Lena Elfman, PhD, b Thomas Fuchs, MD, c Sonja Kapiotis, a Helmut Rumpold, MD, a Rudolf Valenta, MD, d and Susanne Spitzauer, MD a Vienna, Austria, Uppsala, Sweden, and Göttingen, Germany Background: Type I allergy to fish is a severe health problem in countries in which a large percentage of the population derive income from fishing. Objective: The aim of the study was to characterize crossreactive IgE-binding components in six different fish species (cod, tuna, salmon, perch, carp, and eel). The effect of reducing extraction conditions, periodate treatment, and depletion of Ca 2 on binding of IgE to the allergens was investigated. Methods: Extracts were prepared under nonreducing and reducing conditions. IgE-binding components were characterized by IgE immunoblotting, and cross-reactive epitopes were studied by IgE-immunoblot inhibition experiments. To reveal calcium-sensitive or carbohydrate-containing epitopes, nitrocellulose-blotted extracts were exposed to ethylene glycol bis( -aminoethyl ether)-n,n,n,n -tetraacetic acid (EGTA) and periodate. Results: Sera from all patients allergic to fish (n 30) displayed IgE reactivity to parvalbumin, a 12 kd protein present in fish extracts from six different species. Reducing extraction conditions had no effect on IgE binding to parvalbumins, whereas periodate treatment and depletion of proteinbound calcium led to a substantial reduction of IgE binding. Parvalbumins from six different species contained cross-reactive IgE epitopes. Conclusion: Parvalbumin represents a cross-reactive fish allergen. It contains IgE epitopes that are sensitive to periodate treatment and Ca 2 -depletion. (J Allergy Clin Immunol 1998;101:67-74.) Key words: Fish allergens, immunoblot, cross-reactivity, parvalbumin, Ca 2 -binding protein, IgE epitopes Fish is a valuable source of highly assimilated proteins and contains large amounts of polyunsaturated fatty acids and fat-soluble vitamins. Many physicians are therefore recommending a fish-enriched diet, thus leading to an increase of fish consumption. 1 However, fish From the a Institute of Medical and Chemical Laboratory Diagnostics, AKH, University of Vienna; b Pharmacia-Upjohn Diagnostics, Uppsala; c Department of Dermatology, Georg-August University Göttingen; and d the Institute of General and Experimental Pathology, AKH, University of Vienna. Supported by a grant of the Bürgmeisterfonds Vienna and by grant F0506 of the Austrian Science Foundation. Received for publication Feb. 21, 1997; revised Aug. 12, 1997; accepted for publication Aug. 19, Reprint requests: Rudolf Valenta, MD, Institute of General and Experimental Pathology, AKH, University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria. Copyright 1998 by Mosby, Inc /98 $ /1/85504 Abbreviations used EGTA: Ethylene glycol bis( -aminoethyl ether)- N,N,N,N -tetraacetic acid SDS-PAGE: Sodium dodecylsulfate polyacrylamide gel electrophoresis proteins, together with proteins from egg and milk, are a frequent cause of immediate food hypersensitivity. 2, 3 In individuals allergic to fish, ingestion of fish, inhalation of fish proteins, or skin contact cause IgE-mediated clinical symptoms, such as urticaria, dermatitis, angioedema, diarrhea, asthma, or anaphylactic reactions. 3-7 The incidence of fish allergy was found to be higher in children and young adults. 2, 3, 8, 9 In countries, in which a large percentage of the population lives on fishing and associated industries and in which fish consumption is high, the prevalence of fish allergy is 1 per 1000 of the general population. 9 Fish allergens have been characterized by immunoblotting, isoelectric focusing, and passive cutaneous transfer tests. 1, 4, 9, 10 The major allergen from cod, Gad c I, previousely called allergen M, with a molecular weight of 12 kd is a heat stable, acidic protein that is 10, 11 resistant to denaturation and proteolytic digestion. It corresponds to sarcoplasmatic parvalbumins present in the muscle of a variety of vertebrates 12 and shares amino acid sequence homologies in the range of 34% 13, 14 with parvalbumins from carp and hake. Studies on IgE-mediated cross-reactivity carried out by several groups indicated that sera from individuals allergic to fish contained IgE antibodies that cross-react 4-6, 12 with extracts from different fish species. In this study fish allergens were characterized by IgE immunoblotting. The effects of reducing extraction conditions, periodate treatment, and Ca 2 depletion on IgE binding to the allergens were studied. IgE-immunoblot inhibition experiments were performed to identify the cross-reactive components in extracts from different fish species. METHODS Patients and sera Sera from 30 patients who were allergic to fish and had a positive case history of type I allergy to fish were selected for 67

2 68 Bugajska-Schretter et al. J ALLERGY CLIN IMMUNOL JANUARY 1998 TABLE I. Serologic characterization of patients with fish allergy Fish-specific IgE binding by RAST Patient no. Total IgE concentration (ku/l) Cod (f3) Tuna (f40) Salmon (f41) Class kua/l Class kua/l Class kua/l ND ND ku/l, kilo units/liter; kua/l, kilo units antigen per liter; ND, not done. this study. Each of the patients experienced at least one of the typical clinical symptoms (dermatitis, urticaria, angioedema, diarrhea, asthma, or anaphylactic reaction) after ingestion, inhalation, or skin contact with fish proteins. The diagnosis of IgE-mediated fish allergy was verified by subsequent determination of fish-specific IgE (Table I) with the Pharmacia CAP- FEIA System (Pharmacia, Uppsala, Sweden). Nine patients were skin tested with at least one fish extract (cod) according to the recommendations of the European Academy of Allergology and Clinical Immunology. 15 The clinical and serologic characteristics of four representative patients used for the IgEinhibition experiment are displayed in Table II. Serum from a nonallergic individual with negative RAST results with fish allergens was used as a control. Natural allergen extracts The fish species studied belong to the class Osteichthyes of higher bony fish and represent the orders Gadiformes (cod: Gadus callarias), Perciformes (tuna: Thunnus thynnus; perch: Stizostedion lucioperca), Cypriniformes (carp: Cyprinus carpio) Salmoniformes (salmon: Salmon salar), and Anguilliformes (eel: Anguilla anguilla). One gram of raw fish muscle of each species was homogenized under liquid nitrogen, dissolved in Laemmli sample buffer 16 with or without 2-mercaptoethanol (1.25% vol/vol), and boiled for 10 minutes. In order to remove insoluble particles, extracts were centrifuged at 4500 rpm for 10 minutes at 4 C, and supernatants were stored in aliquots at 20 C. The protein content of all extracts was estimated by sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie brilliant blue (Coomassie Brilliant Blue R-250, Bio Rad Laboratories, Richmond, Calif.) staining. 17 SDS-PAGE Extracts were heated at 95 C for 3 minutes. Approximately 100 g of each fish extract and 15 g of bovine calmodulin (Sigma, St. Louis, Mo.), which was dissolved in aqua bidest. at a final concentration of 1 mg/ml, were loaded per centimeter of gel onto 12% gels, with 5% stacking gels, as described by Laemmli. 16 Molecular weights were estimated by using prestained protein standards (14.3 to 220 kd; Amersham, Buckinghamshire, UK). IgE immunoblots Proteins separated by SDS-PAGE were blotted onto nitrocellulose (Nitrocell, Pharmacia Biotech, San Francisco, Calif.) as described by Towbin. 18 Nitrocellulose sheets were cut into 0.5 cm-wide strips. Strips were blocked at room temperature in buffer A (phosphate-buffered saline, ph 7.5, and 0.5% vol/vol

3 J ALLERGY CLIN IMMUNOL VOLUME 101, NUMBER 1, PART 1 Bugajska-Schretter et al. 69 TABLE II. Clinical and serologic characterization of patients for IgE-immunoblot inhibition experiments (Fig. 5) Fish-specific IgE binding by RAST Patient no. Anamnesis* Clinical symptoms Skin prick test response Total IgE ku/l Cod (f3) Tuna (f40) Salmon (f41) Class kua/l Class kua/l Class kua/l 1 t, c, s, p, pl asthma, urticaria s:, pl: c, t, s, p, ca asthma, urticaria c:, t:, s:, p:, ca: c, s, ca, e, r, pl urticaria c, s:, ca:, e:, r:, pl: c, s, e, tr, h, pl urticaria, angioedema s:, e:, tr:, h: *Fish species that caused clinical symptoms. Fish extracts tested., positive reaction. c, cod; ca, carp; e, eel; h, herring; p, perch; pl, plaici; r, redfish; t, tuna, ku/l, kilo units per liter; kua/l, kilo units antigen per liter. Tween 20). Strips were then incubated overnight at 4 C with patients sera that had been diluted 1:10 in buffer A. Control strips were incubated with serum from a nonatopic individual and with buffer without serum. Strips were washed with buffer A three times for 30 minutes, and bound IgE antibodies were detected with I 125-labeled anti-human IgE antibodies (Pharmacia, Uppsala, Sweden). Strips were washed, dried, and exposed to autoradiography films (NEN, Boston, Mass.) with intensifying screens (NEN) at 70 C for 24 to 72 hours. Periodate treatment Nitrocellulose-blotted fish extracts were treated with periodate as described. 19 In brief, nitrocellulose strips were incubated with 50 mmol/l sodium acetate (Merck, Darmstadt, Germany) buffer, ph 4.5, with or without 20 mmol/l sodium m-periodate (Sigma) at room temperature for 1 hour. Strips were then washed and blocked with buffer A three times for 30 minutes. IgE immunoblotting was performed as described above. Ca 2 -depletion experiments The effects of Ca 2 depletion on IgE reactivity to nitrocelluloseblotted cod proteins were investigated as described, 20 with the following modifications: Patients sera were diluted 1:10 either in buffer A containing 0.5 mmol/l CaCl 2 or 5 mmol/l EGTA, ph 7.5. Bound IgE antibodies were detected by using I 125-labeled anti-human IgE antibodies. Reduction of IgE binding was quantified by gamma counting (Wizzard, Automatic Gamma Counter, Wallac, Uppsala, Sweden) of equally sized strips. IgE inhibition studies Four sera from representative patients with fish allergy (see Table II: patients 1, 3, 29, and 30) were diluted 1:60 and preadsorbed with nitrocellulose-blotted proteins from cod or, for control purposes, with empty nitrocellulose membranes, overnight at 4 C. Preadsorbed sera were then probed with nitrocellulose-blotted extracts from tuna, salmon, perch, carp, and eel. Bound IgE was detected with I 125-labeled anti-human IgE antibodies. Inhibition of IgE binding was visualized by autoradiography and quantified by gamma counting. RESULTS Distribution of IgE-binding components in nitrocellulose-blotted cod extract Sera from 28 patients with fish allergy and, as controls, serum from a nonallergic individual and buffer without serum were tested for IgE reactivity with nitrocelluloseblotted cod extract (Fig. 1). All sera contained IgE antibodies directed against a 12 kd component, presumably representing parvalbumin, the major cod allergen Gad c I. Sera that had bound to the 12 kd cod allergen also reacted with purified parvalbumin from carp (data not shown). Twenty-four sera showed intense IgE binding to cod parvalbumin and four sera (from patients 2, 6, 9, and 10) reacted weakly (see Fig. 1). Sera from 7 patients (patients 1, 2, 6, 9, 10, 26, and 27) additionally recognized proteins of 200, 100, 80, 66, 35, 30, 25, 20, and 15 kd, which therefore represent minor allergens. Parvalbumin contains periodate-sensitive IgE epitopes To investigate whether IgE binding to parvalbumin is directed against carbohydrate-containing epitopes, nitrocellulose-blotted extracts from cod and carp were treated with periodate (Fig. 2; lanes) before testing with sera from three representative patients with fish allergy (patients 1, 2, and 20). Periodate treatment led to a substantial reduction of IgE binding to parvalbumin. The slight increase in IgE reactivity of sera 1 and 2 to the 66 kd component and of serum 2 to the 100 kd moiety observed after periodate treatment may result from the exposure of hidden epitopes. IgE binding to the other minor allergens was not affected by periodate treatment. Parvalbumin contains Ca 2 -sensitive IgE epitopes Parvalbumins belong to a family of Ca 2 -binding proteins. To investigate whether IgE-binding epitopes of parvalbumin require protein-bound calcium, nitrocellulose strips containing blotted cod extract were exposed to EGTA during IgE immunoblotting. Sera from 14 patients with fish allergy were incubated with nitrocellulose-blotted fish proteins in the presence (Fig. 3; lanes) or absence (see Fig. 3; lanes) of calcium. A significant reduction of IgE binding to parvalbumin of more than 50% (estimated by gamma counting) was observed in 9 out of 14 sera (sera 1, 4, 5, 6, 7, 8, 9, 13, and 15). No significant reduction of IgE binding to minor allergens was observed to be caused by Ca 2 depletion.

4 70 Bugajska-Schretter et al. J ALLERGY CLIN IMMUNOL JANUARY 1998 FIG. 1. IgE-binding components in cod extract. Sera from 28 patients with fish allergy (lanes 1-28) were tested for IgE reactivity with nitrocellulose-blotted cod proteins. Serum from a nonatopic individual (lane N) and buffer without serum (lane B) were used as controls. FIG. 2. Periodate sensitivity of IgE binding to parvalbumin. Nitrocellulose-blotted extracts from cod and carp were exposed to periodate ( lanes) or sodium acetate buffer ( lanes) prior to testing with sera from three patients with fish allergy (lanes 1, 2, 20). Serum from a nonatopic individual (lane N) and buffer without serum (lane B) were used as controls. Sera from patients 1, 12, and 20, who were allergic to fish, were tested for IgE binding to bovine calmodulin. No IgE binding to bovine calmodulin was observed (data not shown). Comparison of IgE-binding components in extracts from six different fish species Three representative sera (sera 1, 2, and 3) that contained IgE antibodies against most of the cod allergens were probed with nitrocellulose-blotted fish extracts from six fish species (cod, tuna, salmon, perch, carp, and eel), prepared under nonreducing (Fig.4, A) or reducing (Fig. 4, B) conditions. Parvalbumin, the 12 kd major allergen, was recognized by IgE antibodies of patients 1 and 3 in extracts from all fish species tested. Serum from patient 2 reacted weakly with parvalbumin from cod and salmon and did not react with parvalbumin from tuna, perch, carp, and eel. A 66 kd minor allergen, detected by serum 1 was also present in all fish extracts. Extract preparation under reducing conditions did not alter the intensity of IgE binding to these two allergens nor did it change their electrophoretic mobility. Minor allergens with molecular weights of 100, 46, 35, 25, 20, and 15 kd showed a variable distribution in different fish extracts. Reduction of extracts with 2-mercaptoethanol led, in part, to changes of their migration properties.

5 J ALLERGY CLIN IMMUNOL VOLUME 101, NUMBER 1, PART 1 Bugajska-Schretter et al. 71 FIG. 3. Ca21 sensitivity of IgE binding to parvalbumin. Sera from 14 patients with fish allergy (lanes 1, 2, 4-15) diluted in buffer A containing either CaCl2 (1 lanes) or EGTA ( lanes) were probed with nitrocellulose-blotted cod extract. FIG. 4. Comparison of IgE binding to different fish extracts. Sera from three patients with fish allergy (lanes: 1-3) were probed with nitrocellulose-blotted extracts from cod, tuna, salmon, perch, carp, and eel, prepared under (A) nonreducing or (B) reducing conditions. Serum from a nonatopic individual (lane N) and buffer without serum (lane B) were used as controls. Parvalbumin represents the major cross-reactive fish allergen The presence of common IgE-binding epitopes in various fish species was investigated by IgE-immunoblot inhibition experiments (Fig. 5). By testing serial serum dilutions of representative sera, a dilution of 1:60 was determined to contain allergens in excess of antibodies. Sera from four representative patients with fish allergy (see Table II; patients 1, 3, 29, and 30) who had displayed clinical symptoms after ingestion of more than one fish species and who had positive skin prick test results with several fish extracts were used for the cross-inhibition experiments. The sera were preadsorbed either with nitrocellulose-blotted proteins from cod (Fig. 5; 1 lanes) or, as control, with empty nitrocellulose membrane (see Fig. 5; lanes). Preadsorption of sera

6 72 Bugajska-Schretter et al. J ALLERGY CLIN IMMUNOL JANUARY 1998 FIG. 5. Inhibition of IgE binding to parvalbumin by preadsorption with cod proteins. Sera from four representative patients with fish allergy (patients 1, 3, 29, and 30; see Table II for characteristics) that had been preadsorbed either with cod proteins ( lanes) or, as a control, with empty nitrocellulose membrane ( lanes) were probed with nitrocellulose-blotted fish (tuna, salmon, perch, carp, and eel) extracts. TABLE III. Inhibition of serum reactivity to fish extracts after preadsorption with cod proteins Fish species Percentage inhibition of IgE binding serum: Tuna * 84 Salmon Perch Carp Eel The percentage inhibition of IgE binding to the various extracts after preadsorption of sera with cod proteins was quantified by gamma counting. *Serum did not bind to extract. with cod proteins led to a significant inhibition of IgE binding to parvalbumin (12 kd) present in tuna, salmon, perch, carp, and eel. IgE binding of serum 1 to a 66 kd allergen that was detected in all tested fish extracts was not inhibited. Serum from patient 29 who had no significant clinical symptoms after ingestion of tuna also failed to mount detectable IgE reactivity to Western blots of tuna extract. The degree of inhibition of IgE binding to tuna, salmon, perch, carp and eel obtained after preadsorption of the four sera with cod extract was determined by gamma counting (Table III). Inhibition of IgE binding ranged between 43% and 97%, with an average inhibition of 78.2%. DISCUSSION Fish proteins cause IgE-mediated allergic reactions, particularly in atopic children and young adults. 2-9 Contact with fish can lead to dermatitis and urticaria, inhalation of fish particles may cause asthma, and ingestion can provoke diarrhea, angioedema, and severe anaphylactic reactions. 1, 4, 7, 12 In the present study we used serum IgE from 30 patients with fish allergy who had experienced clinical symptoms after ingestion of more than one fish species to study IgE-binding components in different fish extracts by immunoblotting. All sera displayed IgE binding to parvalbumin, a 12 kd component in cod extract, which represents the major cod allergen Gad c I. Nine out of the 30 sera also contained IgE antibodies that reacted with minor allergens of higher molecular weight. To further characterize the effects of different treatments on IgE binding to parvalbumin, a series of exper-

7 J ALLERGY CLIN IMMUNOL VOLUME 101, NUMBER 1, PART 1 Bugajska-Schretter et al. 73 iments was performed. Previously it has been shown that IgE epitopes of parvalbumin are resistant to boiling and denaturation. 10, 11 Although Western blot analysis of cod extract, performed under reducing conditions, led to changes in the migration properties of certain minor allergens, no changes in the electrophoretic mobility or IgE-binding capacity of parvalbumin was noted. Parvalbumin from cod is an acidic glycoprotein with a single glucose residue linked to Cys To investigate whether the protein-bound sugar moiety contributes to the IgE-binding capacity of parvalbumin, nitrocellulose-blotted carp and cod extracts were subjected to periodate treatment before exposure to serum IgE. Deglycosylation did not affect IgE binding to minor allergens, and in the case of certain minor allergens (66 kd and 110 kd components), even slightly increased the IgE reactivity, probably because of exposure of hidden epitopes. In contrast, a significant reduction of IgE binding to parvalbumin was observed after periodate treatment. It is equally possible that IgE antibodies are directed against carbohydrate epitopes or that sugar residues are necessary to maintain the protein conformation required for IgE binding. Alternatively, it is possible that oxidation of amino acids (e.g., cysteine residues) by periodate reduces the IgE-binding capacity of the proteins. Gad c I and related parvalbumins from other fish species belong to a family of Ca 2 -binding proteins. 12, 13, Recently it has been demonstrated that IgE binding to certain Ca 2 -binding plant allergens requires protein-bound calcium. 20, We were therefore interested in determining whether IgE binding to parvalbumin is affected by Ca 2 depletion. Indeed, the majority of sera showed a substantial reduction of IgE binding to parvalbumin after Ca 2 depletion. Structural studies performed with different Ca 2 -binding proteins demonstrated dramatic conformational changes not only in the Ca 2 -binding regions but also in distant parts of these proteins Two mutually nonexclusive models for reduction of IgE binding to parvalbumin after Ca 2 depletion may therefore be considered. The first model would suggest that IgE antibodies bind directly to the Ca 2 -binding loops of parvalbumin. These loops consist of stretches of mostly acidic amino acids EF-hand, which constitute the most conserved common structural motif in different Ca 2 -binding proteins. 22, 24 If indeed the Ca 2 -binding loops represent the major IgE epitopes, extensive cross-reactivity of parvalbumin-specific IgE antibodies with other Ca 2 - binding proteins would be expected. Sera that showed Ca 2 -dependent IgE reactivity to parvalbumin failed to bind to bovine calmodulin, another EF-hand protein, indicating that a different mechanism of binding must be present. Thus, parvalbumin-specific IgE antibodies may be directed against an epitope located at a site, distant from the Ca 2 -binding loop, that can be affected by conformational changes of the protein structure induced by Ca 2 depletion. Our finding that parvalbumin may contain conformational IgE epitopes that are sensitive to Ca 2 depletion may also have clinical implications. Blood sampling in vials containing Ca 2 -chelating agents may affect serologic determinations of parvalbumin-specific IgE antibodies. Recombinant parvalbumin derivatives or fragments containing modified or disrupted calcium-binding sites would be expected to have reduced IgE-binding capacity and may hence represent candidate molecules, with reduced anaphylactic activity, for specific immunotherapy. In the past, several studies on IgE cross-reactivities that were based on skin-testing, oral food challenge tests, immunoblotting, and iselectric focusing analysis of fish allergens provided evidence for cross-reactive components in different fish species. 1, 4, 5, 7 The presence of serum IgE antibodies that cross-react with different fish species has been demonstrated by ELISA and RASTinhibition studies. 1, 4, 5 In order to identify cross-reactive allergens in different fish extracts, IgE-immunoblot inhibition studies were performed. Sera from patients who had experienced clinical symptoms after ingestion of more than one fish species and who had positive skin prick test reactivity to different fish extracts also displayed IgE cross-reactivity with parvalbumins from cod, tuna, salmon, perch, carp, and eel. The amino acid sequences of parvalbumin from cod, carp, and hake have been determined, and these show significant sequence homology, supporting our notion that parvalbumin represents a cross-reactive fish allergen. 13, 21, 23 The fact that almost 70% of the patients tested reacted exclusively with parvalbumin indicates that parvalbumin represents the major cross-reactive fish allergen. That only partial inhibition of IgE binding was observed after preadsorption with parvalbumins from certain fish species indicates the presence of species-specific IgE epitopes. It is therefore not unlikely that patients were sensitized initially against parvalbumin from only one or a few fish species. Those IgE antibodies that are synthesized against cross-reactive structures may form the molecular basis for allergic reactions after contact with other fish species. REFERENCES 1. Bernhisel-Broadbent J, Scanlon SM, Sampson HA. Fish hypersensitivity. I. In vitro and oral challenge results in fish-allergic patients. J Allergy Clin Immunol 1992;89: Crespo JF, Blanco C, Contreras J, Pascual C, Martin Esteban M. Food allergy: a clinical and epidemiological study [abstract]. J Allergy Clin Immunol 1992;89: Kajosaari M. Food allergy in Finnish children aged 1 to 6 years. Acta Paediatr Scand 1982;71: Pascual C, Martin Esteban M, Crespo JF. Fish allergy: evaluation of the importance of cross-reactivity. J Pediatr 1992;121: De Martino M, Novembre E, Galli L, De Marco A, Botarelli P, Marano E, Vierucci A. Allergy to different fish species in codallergic children: in vivo and in vitro studies. J Allergy Clin Immunol 1990;86: De Besche A. On asthma bronchiale in man provoked by cat, dog and different other animals. Acta Med Scand 1937;42: Tuft L, Blumenstein GI. Studies in food allergy. V. Antigenic relationship among members of fish family. J Allergy 1946;17: Dannaeus A, Inganäs M. A follow-up study of children with food

8 74 Bugajska-Schretter et al. J ALLERGY CLIN IMMUNOL JANUARY 1998 allergy: clinical course in relation to serum IgE-and IgG-antibody to milk, egg and fish. Clin Allergy 1981;11: Aas K. Fish allergy and the cod allergen model. In: Brostoff J, Challacombe ST, editorss. Food allergy and intolerance. London: Balliere Tindall, 1987: Elsayed S, Aas K. Characterization of a major allergen (cod). Observations on effect of denaturation on the allergenic activity. J Allergy 1971;47: Aas K, Elsayed S. Characterization of a major allergen (cod). Effect of enzymic hydrolysis on the allergenic activity. J Allergy 1969;44: O Neil C, Helbling AA, Lehrer SB. Allergic reactions to fish. Clin Rev Allergy 1993;11: Elsayed S, Bennich H. The primary structure of allergen M from cod. Scand J Immunol 1975;4: Lindstrom, CD-V, Van Do T, Hordvik I, Endresen C, Elsayed S. Cloning of two distinct cdnas encoding parvalbumin, the major allergen of atlantic salmon (Salmo salar). Scand J Immunol 1996;44: Dreborg S. EAACI: skin tests for diagnosis of IgE-mediated allergy. Allergy 1989;44(suppl 10): Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227: Bradford MM. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 1976;72: 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: Woodward MP, Young Jr WW, Bloodgood RA. Detection of monoclonal antibodies specific for carbohydrate epitopes using periodate oxidation. J Immunol Methods 1985;78: Seiberler S, Scheiner O, Kraft D, Lonsdale D, Valenta R. Characterization of a birch pollen allergen, Bet v III, representing a novel class of Ca 2 binding proteins; specific expression in mature pollen and dependence of patients IgE binding on protein-bound Ca 2. EMBO J 1994;13: Coffee CJ, Bradshaw RA. Carp muscle calcium-binding protein. I. Characterization of the tryptic peptides and the complete amino acid sequence of component B. J Biol Chem 1973;248: Kretsinger RH, Nockolds CE. Carp muscle calcium-binding protein. II. Structure determination and general description. J Biol Chem 1973;248: Pechere JF, Capony JP, Ryden L. The primary structure of the major parvalbumin from hake muscle. Eur J Biochem 1971;23: Heizmann CW, Hunziker W. Intracellular calcium-binding proteins: more sites than insights. Trends Biochem Sci 1991;16: Seiberler S, Scheiner O, Kraft D, Valenta R. Homology of two cdnas coding for birch pollen allergens with calmodulin: proteinbound Ca 2 affects the IgE binding capacity. Int Arch Allergy Immunol 1992;99: Twardosz A, Hayek B, Seiberler S, Vangelista L, Elfman L, Grönlund H, et al. Molecular characterization, expression in Escherichia coli and epitope analysis of a two EF-hand calciumbinding birch pollen allergen, Bet v 4. Biochem Biophys Res Comm 1997; in press. 27. Suphioglu C, Ferreira F, Knox RB. Molecular cloning and immunological characterization of Cyn d 7, a novel calcium-binding allergen from Bermuda grass pollen. FEBS Lett 1997;402: Declercq JP, Tinant B, Parello J, Rambaud J. Ionic interactions with parvalbumins. Crystal structure determination of pike 4.10 parvalbumin in four different ionic environments. J Mol Biol 1991;220: Strynadka N, James M. Towards an understanding of the effects of calcium on protein structure and function. Curr Opin Struct Biol 1991;1: Trave G, Lacombe PJ, Pfuhl M, Saraste M, Pastore A. Molecular mechanism of the calcium-induced conformational change in the spectrin EF-hands. EMBO J 1995;14: