Allergenicity attributes of different peanut market types

Four different market classes of peanut (Runner, Virginia Spanish, and Valencia) are commonly consumed in Western countries, but for some consumers peanuts are a main cause of food-induced anaphylaxis. Limited information is available on the comparative allergenicity of these distinct market classes. The aim of this study was to compare allergenicity attributes of different peanut cultivars.The protein content and protein profiles were highly comparable for all tested cultivars. All cultivar samples contained the major allergens Ara h 1, Ara h 2, Ara h 3 and Ara h 6, as assessed by SDS-PAGE and RP-HPLC, although some minor differences in major allergen content were found between samples. All samples were reactive in commercial ELISAs for detection and quantification of peanut protein. IgE-binding potency differed between samples with a maximum factor of 2, indicating a highly comparable allergenicity.Based on our observations, we conclude that peanuts from the main market types consumed in Western countries are highly comparable in their allergenicity attributes, indicating that safety considerations with regard to peanut allergy are not dependent on the peanut cultivar in question.     

The effect of the food matrix on in vivo immune responses to purified peanut allergens.

There is little knowledge about the factors that determine the allergenicity of food proteins. One aspect that remains to be elucidated is the effect of the food matrix on immune responses to food proteins. To study the intrinsic immunogenicity of allergens and the influence of the food matrix, purified peanut allergens (Ara h 1, Ara h 2, Ara h 3, or Ara h 6) and a whole peanut extract (PE) were tested in the popliteal lymph node assay (PLNA) and in an oral model of peanut hypersensitivity. In the PLNA, peanut proteins were injected into the hind footpad of BALB/c mice; in the oral exposure experiments C3H/HeOuJ mice were gavaged weekly with PE or allergens in the presence of cholera toxin (CT). Upon footpad injection, none of the allergens induced significant immune activation. In contrast, PE induced an increase in cell number, cytokine production, and activation of antigen-presenting cells. Furthermore, the presence of a food matrix enhanced the immune response to the individual allergens. Oral exposure to the purified allergens in the presence of CT induced specific IgE responses, irrespective of the presence of a food matrix. These results suggest that purified peanut allergens possess little intrinsic immune-stimulating capacity in contrast to a whole PE. Moreover, the data indicate that the food matrix can influence responses to individual proteins and, therefore, the food matrix must be taken into account when developing models for allergenic potential assessment.     

Partial characterization of red gram (Cajanus cajan L. Millsp) polypeptides recognized by patients exhibiting rhinitis and bronchial asthma

We sought to assess the allergenic potential of red gram by identifying and characterizing the responsible proteins. Immunoblotting was performed to detect IgE binding proteins. Identities of these proteins were confirmed by mass spectrometry. To evaluate allergenic potential, BALB/c mice were sensitized with red gram proteins and levels of specific immunoglobulins, histamine, Th2 cytokines were measured. Allergenic response was evident by significant increase in specific IgE, IgG1, histamine and Th2 cytokine levels. Prominent anaphylactic symptoms, discernible histopathological responses and down regulation of IFN-γ levels give strong support towards allergenicity of red gram proteins. IgE immunoblot detected five proteins; one of 66 kDa, three of 45 kDa (pI of ∼5.3, 5.9 and 6.6) and one of 30 kDa. All these proteins showed homology to known allergens of soybean (different subunits of β-conglycinin), lentil (Len c1 and Len c2), peanut (Ara h1) and pea (vicilin). In conclusion, five novel IgE binding proteins (namely Caj c1, Caj c2, Caj c3, Caj c4 and Caj c5) were identified as putative clinically relevant allergens.     

Clinical risk assessment of GM foods

The main concerns about adverse effects of genetically modified (GM) foods on health are the transfer of antibiotic resistance, toxicity and allergenicity. There are two issues from an allergic standpoint. First, the transfer of a known allergen may occur from a crop into a non-allergenic target crop. The second scenario is the creation of a neo-allergen where de novo sensitisation occurs in the population. The first scenario occurred in 1996 when the 2S albumen protein from Brazil nut was transferred into soy bean (N. Engl. J. Med. 334 (1996) 688). 2S albumen was found to be a major Brazil nut allergen and the newly expressed protein in transgenic soy retained its allergenicity. Patients allergic to Brazil nuts and not to soy bean now showed an IgE mediated response towards GM soy bean. We argue that it is possible to prevent such occurrences by doing IgE-binding studies and taking into account physico-chemical characteristics of proteins and referring to known allergen databases. The second possible scenario of de novo sensitisation does not easily lend itself to risk assessment. We compare GM technology to traditional plant breeding and food processing methods. There is no evidence that the technology used for the production of GM foods poses an allergic threat per se compared to other methodologies widely accepted in the food industry. We need to proceed cautiously in the future, assessing individual GM foods on the basis of their individual merits and risks prior to introducing them into the market.     

Inter-laboratory comparisons of assessment of the allergenic potential of proteins in mice

Assessment of the potential allergenicity of novel proteins, including those expressed in genetically modified plants, is an important issue. In previous studies, we have shown that the IgE measurement induced by systemic exposure of BALB/c mice to a range of proteins correlates broadly with what is known of their allergenic potential in humans. The approach used a homologous passive cutaneous anaphylaxis (PCA) assay that reflects IgE-dependent biological activity and is of sufficient sensitivity to detect IgE production in the absence of adjuvant. In previous studies, the immunization phase was conducted independently in two separate facilities, and the subsequent analytical work (PCA) conducted in a single facility. The purpose here was to further evaluate the transferability of this approach. To this end, BALB/c mice were exposed to a range of doses of peanut agglutinin or ovalbumin, allergenic proteins of peanut and hen's egg, respectively, in two independent laboratories. Serial doubling dilutions of serum pooled for each treatment group were analyzed for specific IgE. At higher doses of allergen very similar, or identical, IgE titers were achieved in both laboratories, although at lower doses, responses were somewhat more variable. These data demonstrate that, although technically demanding, the measurement of protein allergen-induced IgE antibody production in mice using PCA is relatively robust and is transferable and reproducible between laboratories. This approach may provide a useful tool for the safety assessment of novel proteins and suggests that continued evaluation of the approach with a wider range of protein allergens and non-sensitising proteins is justified.     

The value of short amino acid sequence matches for prediction of protein allergenicity.

Typically, genetically engineered crops contain traits encoded by one or a few newly expressed proteins. The allergenicity assessment of newly expressed proteins is an important component in the safety evaluation of genetically engineered plants. One aspect of this assessment involves sequence searches that compare the amino acid sequence of the protein to all known allergens. Analyses are performed to determine the potential for immunologically based cross-reactivity where IgE directed against a known allergen could bind to the protein and elicit a clinical reaction in sensitized individuals. Bioinformatic searches are designed to detect global sequence similarity and short contiguous amino acid sequence identity. It has been suggested that potential allergen cross-reactivity may be predicted by identifying matches as short as six to eight contiguous amino acids between the protein of interest and a known allergen. A series of analyses were performed, and match probabilities were calculated for different size peptides to determine if there was a scientifically justified search window size that identified allergen sequence characteristics. Four probability modeling methods were tested: (1) a mock protein and a mock allergen database, (2) a mock protein and genuine allergen database, (3) a genuine allergen and genuine protein database, and (4) a genuine allergen and genuine protein database combined with a correction for repeating peptides. These analyses indicated that searches for short amino acid sequence matches of eight amino acids or fewer to identify proteins as potential cross-reactive allergens is a product of chance and adds little value to allergy assessments for newly expressed proteins.     

The utility of an international sera bank for use in evaluating the potential human allergenicity of novel proteins.

In the safety assessment of novel foods produced through biotechnology, careful consideration is given to determining the allergenic potential of newly introduced proteins. IgE serum screening is one tool for evaluating whether the protein in question has sequence identity to a known allergen or if the source of the gene encoding the protein is a known allergenic food. A "specific" serum screen involves testing a gene product with sera from patients with documented clinical allergy to a specific allergen to confirm that the gene product of interest is not the same protein to which the patient produces IgE antibodies. A "targeted" serum screen involves testing the gene product of interest with sera from patients sensitive to food or aeroallergens from the same broad group. The concept of a global sera bank with accessible, well-characterized sera for use in such assays is an appealing option. This paper summarizes the consensus elements from a workshop to evaluate the potential utility of an international sera bank for evaluating the allergenicity of novel proteins. Areas of agreement following the workshop included the following: (1) specific sera screens are appropriate for exploring potentially cross-reactive proteins that have been identified through bioinformatics analyses; however, additional validation is needed, particularly for targeted sera screens, (2) practical and ethical considerations may preclude the formation of a global sera bank, and therefore, (3) a regional network of clinicians who could serve as sources of patient sera or be approached to conduct sera studies would be the most practical alternative.     

Differences in allergenic potential of food extracts following oral exposure in mice reflect differences in digestibility: potential approaches to safety assessment.

An animal model for food allergy is needed to assess genetically modified food crops for potential allergenicity. The ideal model must produce allergic antibody (IgE) to proteins differentially according to known allergenicity before being used to accurately identify potential allergens among novel proteins. The oral route is the most relevant for exposure to food antigens, and a protein's stability to digestion is a current risk assessment tool based on this natural route. However, normal laboratory animals do not mount allergic responses to proteins administered orally due to oral tolerance, an immunologic mechanism which specifically suppresses IgE. To circumvent oral tolerance and evoke differential IgE responses to a panel of allergenic and nonallergenic food extracts, female C3H/HeJ mice were exposed subcutaneously or orally with cholera toxin as an adjuvant. All foods elicited IgE by the subcutaneous route. Oral exposure, however, resulted in IgE to allergens (peanut, Brazil nut, and egg white) but not to nonallergens (spinach and turkey), provided that the dose and exposures were limited. Additionally, in vitro digestibility assays demonstrated the presence of digestion-stable proteins in the allergenic food extracts but not in the nonallergenic foods. Our results suggest that the subcutaneous route is inadequate to distinguish allergens from nonallergens, but oral exposure under the appropriate experimental conditions will result in differential allergic responses in accordance with known allergenicity. Moreover, those foods containing digestion-resistant proteins provoke allergic responses in this model, supporting the current use of pepsin resistance in the decision tree for potential allergenicity assessment.     

A multi-laboratory evaluation of a common in vitro pepsin digestion assay protocol used in assessing the safety of novel proteins

Rationale. Evaluation of the potential allergenicity of proteins derived from genetically modified foods has involved a weight of evidence approach that incorporates an evaluation of protein digestibility in pepsin. Currently, there is no standardized protocol to assess the digestibility of proteins using simulated gastric fluid. Potential variations in assay parameters include: pH, pepsin purity, pepsin to target protein ratio, target protein purity, and method of detection. The objective was to assess the digestibility of a common set of proteins in nine independent laboratories to determine the reproducibility of the assay when performed using a common protocol. Methods. A single lot of each test protein and pepsin was obtained and distributed to each laboratory. The test proteins consisted of Ara h 2 (a peanut conglutin-like protein), beta-lactoglobulin, bovine serum albumin, concanavalin A, horseradish peroxidase, ovalbumin, ovomucoid, phosphinothricin acetyltransferase, ribulose diphosphate carboxylase, and soybean trypsin inhibitor. A ratio of 10U of pepsin activity/microg test protein was selected for all tests (3:1 pepsin to protein, w:w). Digestions were performed at pH 1.2 and 2.0, with sampling at 0.5, 2, 5, 10, 20, 30, and 60min. Protein digestibility was assessed from stained gels following SDS-PAGE of digestion samples and controls. Results. Results were relatively consistent across laboratories for the full-length proteins. The identification of proteolytic fragments was less consistent, being affected by different fixation and staining methods. Overall, assay pH did not influence the time to disappearance of the full-length protein or protein fragments, however, results across laboratories were more consistent at pH 1.2 (91% agreement) than pH 2.0 (77%). Conclusions. These data demonstrate that this common protocol for evaluating the in vitro digestibility of proteins is reproducible and yields consistent results when performed using the same proteins at different laboratories.     

Induction of IgE antibody responses by protein allergens: inter-laboratory comparisons.

There is a growing interest in the development of methods for the evaluation of the allergenic potential of novel proteins. One approach is the measurement of specific IgE antibody production stimulated by systemic (intraperitoneal; i.p.) exposure of BALB/c strain mice. In the current investigations, inter-laboratory comparisons have been performed of IgE antibody production induced in mice by food proteins of differing sensitizing potential. Female BALB/c strain mice (n=5) were exposed to 0.1% peanut agglutinin, an allergenic constituent of peanuts, to 2% ovalbumin (OVA), a major allergenic constituent of hens' egg, or to a protein considered to lack significant allergenicity, potato agglutinin (5%). Specific IgE antibody was measured by homologous passive cutaneous anaphylaxis assay and IgG and IgG1 antibody production was analysed by enzyme-linked immunosorbent assay (ELISA). Two independent experiments were conducted in each laboratory, but with all serological analyses conducted in one of the laboratories. Each of the proteins induced vigorous IgG and IgG1 antibody responses, with no statistically significant differences in titres recorded between laboratories. Furthermore, OVA and potato agglutinin induced responses of equivalent immunogenicity with respect to both IgG and IgG1 antibody titres. Administration of peanut agglutinin and OVA each stimulated marked IgE antibody responses in every experiment. In the two laboratories, titres ranged from 1:32 and 1:64 for peanut agglutinin, and from 1:8 and 1:32 for OVA. In contrast, exposure to potato agglutinin failed to induce vigorous IgE production, with no detectable IgE (negative with neat serum), or titres of 1 (positive with neat serum only) recorded. These data demonstrate that the induction of IgE antibody by food proteins of differing allergenic potential is a relatively robust phenomenon and transferable between laboratories. Furthermore, these results provide additional evidence that the measurement of antibody (IgE) responses in BALB/c mice may allow discrimination between allergens and those materials that apparently lack allergenicity.     

Cytotoxic and allergenic potential of bioactive proteins and peptides

This review article deals with the assessment of cytotoxic and allergenic potential of bioactive proteins and peptides. It is evident that 'novel' foods or nutraceuticals containing bioactive proteins and peptides must fulfill their proposed "health claim". Furthermore, there is a need to assess their potential to exert adverse effects before they can be made widely available to consumers. A brief overview of compounds (i.e. proteins and peptides of animal and plant origin) and mechanisms involved in cytotoxic and allergenic (adverse) reactions is given along with some recent results obtained from ongoing studies. There are numerous proteins and peptides of plant and animal origin that are known to exhibit cytotoxic effects. There is evidence that many cytotoxic compounds described in the literature exclusively affect malignant cells leading to the assumption that a cancer protective effect could exist for such bioactive proteins and peptides. All the constituents that are responsible for the allergenicity of foods (as well as of pollens) are proteinaceous in nature. Some protein breakdown products, i.e. peptide fragments, may conserve part of the allergenicity of the native protein and thus can also be considered as allergens. The molecular basis of IgE recognition underlying cow's milk protein allergy is described. Some results from studies on volunteers fed caseinophosphopeptides or potentially hypotensive milk protein hydrolysates illustrate the major difference between allergenicity and immunogenicity. The data presented on the relationship between the structure of food proteins and peptides and their allergenicity shows the difficulty in trying to assess the "non-allergenicity" of products derived from an allergenic source, even if the process used involved extensive hydrolysis of the native protein(s). A 'weight of evidence approach' for assessing the potential allergenicity of a novel protein with no history of prior allergenicity is also presented with regard to the current EU Regulations.     

The influence of an endogenous beta3 subunit on recombinant GABA(A) receptor assembly and pharmacology in WSS-1 cells and transiently transfected HEK293 cells

Cell lines are commonly used for studying recombinant heterooligomeric ion channels with defined subunit composition. Such studies often ignore the contribution of endogenous proteins in the assembly of mature channels. We examined whether an endogenous subunit was required for the functional expression of gamma-aminobutyric acid type A (GABA(A)) receptors in WSS-1 cells, HEK293 cells stably expressing recombinant alpha1 and gamma2 subunits. Our pharmacological and RT-PCR analyses of GABA(A) receptors and their mRNAs in WSS-1 cells confirm the presence of alpha1 and gamma2 subunits and suggest the existence of an endogenous beta3 subunit. Whole-cell GABA-evoked currents recorded from untransfected WSS-1 cells were blocked by bicuculline methiodide and enhanced by anesthetics and anticonvulsants including the subunit-selective compounds diazepam and loreclezole. These data suggest that, in addition to the gamma2 subunit, WSS-1 cell receptors also contain beta2/3 subunits. RT-PCR revealed that WSS-1 cells and parental HEK293 cells contain beta3 mRNA. We examined the contribution of the beta3 subunit in the function of receptors formed by expression of alpha1 and gamma2S subunits. Untransfected HEK293 cells were unresponsive to GABA. Cells transfected with alpha1 and gamma2S cDNAs displayed small diazepam and loreclezole responsive GABA-activated currents. By contrast, the expression of alpha1 and gamma2S cDNAs in the neuroblastoma NB41A3 cell line, that lacks beta subunit mRNAs, failed to produce functional receptors. These data reaffirm that alpha1 and gamma2S subunits alone do not form functional GABA(A) receptors and that receptors of WSS-1 cells contain alpha1, beta3 and gamma2S subunits.     

Information management for the study of allergies

Microarrays and other large-scale screening technologies produce quantities of increasingly complex allergy data. These data link molecular and clinical measurements and observations and provide fertile ground for improving our understanding of the processes involved in allergic reactions. Information technology is employed in gathering, storage, retrieval and analysis of these data. The increasing proportion of allergy data are generated from genomics and proteomics approaches. The major activity focuses on characterization of allergens including IgE reactivity, structural properties, and mapping of IgE and T-cell epitopes. Because of the complexity of allergy data, their utilization requires bioinformatics approaches. Allergen data are stored in the general and specialist databases. At least a dozen of important allergen databases and data repositories have been developed to date. These data are analysed using general and specialist bioinformatics tools. The major applications of bioinformatics include support for allergen characterization, assessment of allergenicity, and identification of allergic cross-reactivity. These applications in turn support the development of vaccines and therapies for allergic disease. In this article we review allergen databases and tools for the analysis of allergens, and discuss the new directions in the field supported by large scale screening involving genomics, proteomics, and bioinformatics support.     

Allergenicity of refined vegetable oils.

Several commercially important refined vegetable oils are derived from plants which are recognized as potent food allergens (e.g. peanut, soy). Full refining of oils results in the almost complete removal from oils of protein, which is responsible for allergic reactions. However, it is uncertain whether the minute amounts remaining could provoke allergic reactions in highly susceptible individuals. This has led to a vigorous debate about the safety of refined oils and specifically whether to label each oil individually because of the potential risk of allergenicity. Peanut oil has been the most thoroughly studied. It has been shown, in well-designed studies, that refined peanut oil can be safely consumed by the vast majority of peanut-allergic individuals, whereas unrefined oil can provoke reactions in some of the same individuals. However, some other studies report cases of allergic individuals reacting to oils, which are presumed to be refined. While it is likely that the discrepancy between these observations is due to differences in the processing of the oils, and possibly the protein content, this has not been formally demonstrated. Few data exist on the potential allergenicity of other edible vegetable oils; what data there are suggest that the major oils (soy, maize, sunflower, palm) do not provoke allergic reactions in susceptible individuals. Determining the content and immunoreactivity of the residual protein of refined oils is crucial to assessing the allergenic risk they present. Current methodology is inadequate and has not been validated for use with oils and aqueous extracts from oils. Little is known about the importance of different processing steps on allergenicity, although this information is crucial to risk assessment, particularly when considering process modifications. Available data suggest that the protein content of crude oils is of the order of 100-300 microg and that refining results in levels up to about 100-fold lower. The review concludes that peanut oil, and by extrapolation other edible vegetable oils, presents no risk of provoking allergic reactions in the overwhelming majority of susceptible people. However, there is a need to standardize and validate methodology for measuring the protein content and immunoreactivity of such so that they can be used to maintain process specifications. Thresholds of reactivity to allergens in man also need to be established in order to assess fully the risk from very small amounts.     

Structural, functional and hybridization studies of the glutathione S-transferases of rat liver

We have purified five forms of glutathione S-transferase from rat liver. One form was the glutathione S-transferase B (ligandin), which is composed of two non-identical subunits with molecular weights of 22,000 (Ya) and 25,000 (Yc). Two of the other transferases were Ya and Yc homodimers. The other two transferases were also homodimers, but their subunit, Yb, had a molecular weight of 24,000. The three proteins containing either Ya or Yc subunits had similar substrate specificities, and all three contained peroxidase activity. The greatest peroxidase activity was present in proteins containing the Yc subunit. Enzymes composed of Yb subunits had minimal peroxidase activity in addition to different substrate specificities. The Ya and Yc containing enzymes bound the ligands bilirubin, and indocyanine green with high affinity (KD less than 5 microM), although the KD values of the YcYc protein were consistently 4- to 12-fold greater than those of the other two transferases. Studies were performed to define the origins of the various isozymes. There was no evidence for conversion of Yc to either Ya or Yb during storage or under conditions favorable to proteolysis. Hybridization studies were performed under denaturing conditions (6 M guanidine-HCl), and a YaYc hybrid was formed from the YaYa and YcYc proteins. In addition, both YaYa and YcYc hybrids were formed from transferase B. The hybrids were functionally similar to the proteins isolated originally from the liver. Attempts to form a YaYb hybrid from the YbYb and YaYa transferases were unsuccessful. This result is consistent with the lack of this enzyme form in the liver. Glutathione S-transferase B and the Ya and Yc homodimers appeared to be hybrids of common subunits. These three transferases had very similar functional and structural characteristics and differed from the transferases that are composed of Yb subunits.     

Comparative assessment of multiple criteria for the in silico prediction of cross-reactivity of proteins to known allergens

Genetically modified crops are becoming important components of a sustainable food supply and must be brought to market efficiently while also safeguarding the public from cross-reactivity of novel proteins to known allergens. Bioinformatic assessments can help to identify proteins warranting further experimental checks for cross-reactivity. This study is a large-scale in silico evaluation of assessment criteria, including searches for: alignments between a query and an allergen having ?35% identity over a length ?80; any sequence (of some minimum length) found in both a query and an allergen; any alignment between a query and an allergen with an E-value below some threshold. The criteria and an allergen database (AllergenOnline) are used to assess 27,243 Viridiplantae proteins for potential allergenicity. (A protein is classed as a “real allergen” if it exceeds a test-specific level of identity to an AllergenOnline entry; assessment of real allergens in the query set is against a reduced database from which the identifying allergen has been removed.) Each criterion’s ability to minimize false positives without increasing false negative levels of current methods is determined. At best, the data show a reduction in false positives to ∼6% (from ∼10% under current methods) without any increase in false negatives.     

Modulation of ethanol consumption by genetic and pharmacological manipulation of nicotinic acetylcholine receptors in mice

Rationale

Alcohol and nicotine are commonly co-abused. Genetic correlations between responses to these drugs have been reported, providing evidence that common genes underlie the response to alcohol and nicotine. Nicotinic acetylcholine receptors (nAChRs) in the mesolimbic dopamine system are important in mediating nicotine response, and several studies suggest that alcohol may also interact with these nAChRs.

Objective

The aim of this study was to examine the role of nAChRs containing α7 or β2 subunits in ethanol consumption.

Methods

A two-bottle choice paradigm was used to determine ethanol consumption in wild-type and nAChR subunit knockout mice. Challenge studies were performed using the α4β2 nAChR partial agonist varenicline.

Results

Mice lacking the β2 subunit consumed a similar amount of ethanol compared to their wild-type siblings in an ethanol-drinking paradigm. In contrast, mice lacking the α7 nAChR receptor subunit consumed significantly less ethanol than wild-type mice but consumed comparable amounts of water, saccharin, and quinine. In C57BL/6J mice, varenicline dose-dependently decreased ethanol consumption with a significant effect of 2 mg/kg, without affecting water or saccharin consumption. This effect of varenicline was not reversed in mice lacking either the α7 or β2 subunit, providing evidence that nAChRs containing one of these subunits are not required for this effect of varenicline.

Conclusions

This study provides evidence that α7 nAChRs are involved in ethanol consumption and supports the idea that pharmacological manipulation of nAChRs reduces ethanol intake. Additional nAChRs may also be involved in ethanol intake, and there may be functional redundancy in the nicotinic control of alcohol drinking.     

Characterization of the allergenic potential of proteins: an assessment of the kiwifruit allergen actinidin

Assessment of the potential allergenicity (IgE‐inducing properties) of novel proteins is an important challenge in the overall safety assessment of foods. Resistance to digestion with pepsin is commonly measured to characterize allergenicity, although the association is not absolute. We have previously shown that specific IgE antibody production induced by systemic [intraperitoneal (i.p.)] exposure of BALB/c strain mice to a range of proteins correlates with allergenic potential for known allergens. The purpose of the present study was to explore further the utility of these approaches using the food allergen, actinidin. Recently, kiwifruit has become an important allergenic foodstuff, coincident with its increased consumption, particularly as a weaning food. The ability of the kiwifruit allergen actinidin to stimulate antibody responses has been compared with the reference allergen ovalbumin, and with the non‐allergen bovine haemoglobin. Haemoglobin was rapidly digested by pepsin whereas actinidin was resistant unless subjected to prior chemical reduction (reflecting intracellular digestion conditions). Haemoglobin stimulated detectable IgG antibody production at relatively high doses (10%), but failed to provoke detectable IgE. In contrast, actinidin was both immunogenic and allergenic at relatively low doses (0.25% to 1%). Vigorous IgG and IgG1 antibody and high titre IgE antibody responses were recorded, similar to those provoked by ovalbumin. Thus, actinidin displays a marked ability to provoke IgE, consistent with allergenic potential. These data provide further encouragement that in tandem with analysis of pepsin stability, the induction of IgE after systemic exposure of BALB/c strain mice provides a useful approach for the prospective identification of protein allergens. Copyright © 2013 John Wiley & Sons, Ltd.