A mouse model for food allergy using intraperitoneal sensitization

Abstract

Food allergy is an important health issue. With the increasing interest in novel foods derived from transgenic crop plants, there is a growing need for the development of approaches for the characterization of the allergenic potential of proteins. Although most foreign proteins are immunogenic (able to induce IgG antibody responses), relatively few are important food allergens with the capacity to provoke IgE antibody production. There is currently no validated animal model for the determination of allergenic potential of food proteins. One approach that appears to show some promise is outlined in the current chapter. BALB/c strain mice are immunized by intraperitoneal injection and the potential to cause allergenicity assessed as a function of the induction of specific IgE antibody, measured by homologous passive cutaneous anaphylaxis. Progress to date with this method is summarized, and comparisons are made with other experimental models, including considerations of route of exposure, use of adjuvants and selection of appropriate end points.

Introduction

The prevalence of food allergy has been estimated recently at 3.5–4% of adults in the USA and approximately 6–8% of young children and infants [1], [2]. With increased interest in the development of novel foods, including foods and food products derived from genetically modified crops, there is a need for the provision of appropriate safety assessment strategies. One important issue is whether the products of novel genes introduced into crop plants have the potential to induce allergic sensitization [3], [4], [5]. This concern is not simply academic. Given that the introduction into the diet of conventionally produced novel foods, such as that of kiwi fruit in the UK, has resulted in the appearance, and a steady increase in the number, of cases of food allergy to this product [6]. However, there is good reason to suppose that not all proteins are equally allergenic. Only a small proportion of the food proteins consumed regularly is associated with allergic disease. Indeed, most cases of food allergy in the USA and Western Europe are associated with a relatively limited range of produce; most commonly peanuts, tree nuts, hens’ egg, cows’ milk, wheat, soybeans, fish and shellfish [2], [7], [8], [9]. These observations suggest that there is indeed a spectrum of allergenicity among food proteins.Food allergy is a complex disease, with genetic predisposition, environmental factors and exposure conditions all contributing to inter-individual differences in susceptibility [10]. It is therefore very unlikely that a single method using experimental animals will be developed that is capable of accurately predicting all aspects of the likely prevalence, persistence and severity of food allergy among human populations exposed to a novel allergen in the diet. Despite this, however, the first step of any safety assessment process is to identify accurately intrinsic hazard, or lack of it, and it is for this purpose that current approaches to the development of animal models of food allergy are directed. Once intrinsic hazard has been identified, the next steps in the risk assessment process will be to determine the characteristics of that hazard and to define the likely conditions and extent of exposure, and on those bases assess likely risks to human health.

In 1996, a collaboration between the International Food Biotechnology Council (IFBC) and the International Life Sciences Institute (ILSI), Allergy and Immunology Institute outlined the first systematic approach for such allergenicity testing [11]. A hierarchical strategy was suggested that incorporated consideration of the serological identity of the novel protein with proteins known to be allergenic in humans, examination of amino acid homology with, and/or structural similarity to, allergenic proteins and measurement of resistance to proteolytic digestion in a simulated gastric fluid [11], [12]. However, it must be recognized that these approaches (individually or collectively) may not provide a definitive evaluation of inherent sensitizing potential. The IFBC/ILSI decision tree provides a useful approach to the identification of proteins that are likely to have sensitizing activity based upon their antigenic, molecular or structural similarities with known allergens. However, the absence of serological or structural homology with already characterized protein allergens does not automatically preclude inherent sensitizing potential. Furthermore, although there exists a correlation between the relative resistance of a protein to digestion by pepsin in a simulated gastric fluid and allergenic activity, this relationship is not absolute, with reports of both pepsin resistant non-allergens [13], and pepsin sensitive allergens [14].

In the light of these considerations there is a need for methods that will provide a more holistic and more definitive assessment of allergenic potential, and that will have the ability to identify novel proteins with inherent sensitizing potential that lack structural homology or serological cross-reactivity with known allergens. As a consequence there has been a growing interest in the design and development of appropriate animal models and their potential integration into safety assessment paradigms [10], [15]. Progress in this area was acknowledged in a re-evaluation of the 1996 ILSI/IFBC decision tree reported by the joint Food and Agricultural Organization of the United Nations and the World Health Organization (FAO/WHO) Expert Consultative Committee on the Allergenicity of Foods Derived from Biotechnology [16]. One of the conclusions reached in this report was that sufficient evidence has now accumulated to suggest that some animal models may provide valuable information regarding the potential allergenicity of foods derived from biotechnology [16]. Investigators have explored the use of various species for the assessment of allergenic potential, including rats, dogs and swine [17], [18], [19]. An alternative approach is the development of mouse models of sensitization to food proteins [20], [21], [22], [23], [24]. The mouse has a number of advantages compared with other animal models, particularly with respect to the availability of inbred high IgE responder strains and of various immunological and molecular reagents.

It is important to emphasize that currently none of these approaches has been validated (or even evaluated thoroughly) for the purposes of hazard identification in the context of a safety assessment. However, the available evidence suggests that the judicious use of an accurate and robust animal model, in tandem with the other approaches summarized above, would be of considerable utility in safety assessment. In this chapter, we describe one method that is being developed currently; in which inherent allergenic potential is evaluated as a function of the ability of proteins to induce the production of IgE; antibody of this isotype being the major effector of immediate-type allergic reactions, including food allergic reactions [25], [26]. In this regimen, test protein is administered systemically (by intraperitoneal injection) to BALB/c strain mice in the absence of adjuvant, allowing the intrinsic allergenic potential of each protein to be assessed [20], [21], [22], [27]. The BALB/c strain mouse was selected as such mice are high IgE responders, equivalent to an atopic phenotype. Specific IgG antibody production is measured by enzyme-linked immunosorbent assay (ELISA) and specific IgE antibody responses assessed by homologous passive cutaneous anaphylaxis (PCA) assay. The strategy is to identify as potential allergens those proteins that have the ability to induce IgE antibody responses. These can be distinguished from non-allergenic proteins that despite being antigenic, and therefore able to provoke IgG antibody production, either fail to elicit IgE, or stimulate only low titre IgE antibody. Although only a relatively limited number of proteins has been examined to date, marked and very significant differences in the ability of allergens and presumed non-allergens to stimulate IgE responses have been demonstrated over a wide range of doses, and under conditions where the same proteins are of equivalent overall immunogenicity in terms of IgG antibody production [20], [21], [22], [27]. Experience to date suggests that the measurement of antibody (IgE) responses in BALB/c mice serves to identify food allergens accurately, and to distinguish them from those materials that apparently lack significant allergenic potential.

This method using BALB/c strain mice is described, progress to date is summarized, and comparisons are made with other experimental models, including considerations of route of exposure, use of adjuvants and selection of appropriate end points.