- Top of page
- Materials and methods
Background: The development of animal models developing specific immunoglobulin (Ig)E presenting the same specificity as human IgE and similar clinical symptoms as those observed in allergic patients are of great interest for the understanding of mechanisms involved in the induction and regulation of food allergy.
Methods: Balb/c female mice were sensitized with whole peanut protein extract (WPPE) by means of intraperitoneal (i.p.) injections with alum or gavages with cholera toxin (CT). The WPPE specific IgE, IgG1 and IgG2a were monitored. Th2 cells activation was analysed assaying interleukin (IL)-4 and IL-5 vs IFNγ on reactivated splenocytes. Local anaphylactic reaction was evaluated by assaying histamine in faecal samples. The oral sensitization protocol was further extended to cow's milk proteins (CMP).
Results: Balb/c mice developed high peanut-specific IgE and IgG1 responses either after i.p. or oral sensitizations. In both cases, antibodies were specific to polymer of glycinin fragments, containing polypeptides from Ara h3/4, and to a lesser extent to Ara h1 and Ara h2. Interleukin-4 and IL-5 production were evidenced. Balb/c mice could also be sensitized to CMP, as demonstrated by CMP-specific IL-4 and IL-5 secretions and induction of IgE specific for whole caseins, β-lactoglobulin, serum bovine albumin and lactoferrin. Of interest was the occurrence of a local anaphylactic reaction in the peanut and CM models.
Conclusions: In contrast with previous authors, Balb/c mice were sensitized and evidenced an allergic reaction after oral administrations of peanut or CMP plus CT, providing an interesting model for further studies on immunopathogenic mechanisms.
Cow's milk allergy affects 1.9–2.8% of young children in Europe (1), and can develop in adulthood (2). Cow's milk allergic patients may be sensitized to, i.e. presented immunoglobulin (Ig)E specific to, various proteins, mainly β-lactoglobulin (BLG) and casein (3, 4). Although most CM allergic young patients outgrow their allergy by 3 years of age, IgE mediated-CM allergy predisposes them to the development of others food allergies and asthma (5). Unlike cow's milk allergy, peanut allergy is rarely outgrown and is often implicated in fatal and near-fatal food-induced anaphylactic reactions (6). A recent study in North America demonstrated that peanut allergy prevalence exceeds 1% of schoolchildren aged 5–9 years (7), and two sequential cohort studies on the Isle of Wight (UK) clearly demonstrated that sensitization to peanut is increasing (8). Major allergens of peanut were demonstrated to be Ara h1 (vicilin) and Ara h2 (conglutinin-like protein), which were recognized by more than 65% of peanut-allergic patients (9–14). Ara h3/4 (glycinin) is recognized by 44% (13) to 53% (15) of peanut-allergic patients.
In order to obtain more information on the immunopathogenic mechanisms of the IgE-mediated food allergic reaction, mouse models of high allergen-specific IgE induction are required, which need the use of adjuvants. These models should reproduce the IgE fine specificity and the symptoms as observed in allergic humans upon challenge. Immune response is influenced by various factors (16), and of interest are CM and peanut allergy models proposed by Li et al., who used C3H/HeJ mice, the intragastric (i.g.) route for sensitization and challenge, cholera toxin (CT) as efficient mucosal adjuvant, and optimized antigen doses. In those models, i.g. challenge of sensitized mice induced general anaphylactic reactions and symptoms involving multiple target organs such as gastrointestinal tract, respiratory system and skin. Moreover, allergen-specific Th2 cells were induced, as demonstrated by cytokine assays of interleukin (IL)-4 vs IFNγ (17, 18). However, in a later study, the same authors demonstrated Balb/c mice failed to produce CM-specific IgE and did not reproduce symptoms after CM or peanut challenge, even if a slight peanut-specific IgE response was observed. At the cellular level, Th1 but not Th2 cells were evidenced. The authors then concluded that Balb/c is a responder strain for parenteral sensitization, but a nonresponder strain for oral sensitization, in contrast with C3H/HeJ strain (19). However, these results are conflicting with others demonstrating sensitization of Balb/c mice via the oral route with purified antigen alone (20), or purified antigen mixed with CT (4, 21, 22). Successful challenges of sensitized Balb/c mice was also observed (22). Moreover, Balb/c mice have been shown to be of great interest for the study of the sensitizing potency of (novel) proteins (23), as well as for study of the elicitation of the allergic reaction (24). Then, we performed optimized oral administrations of peanut or CM proteins mixed with CT. We demonstrated that, in those conditions, Balb/c mice produced peanut- and CM-specific IgE and Th2 cytokines. The IgE specificity was shown to be close to the response characterized in allergic human patients. Moreover, we could demonstrate by a noninvasive method the occurrence of a local anaphylactic reaction, further validating this model.
- Top of page
- Materials and methods
In this study, we demonstrated that high peanut-specific IgE and IgG1 responses can be obtained in Balb/c mice after i.p. or oral sensitization. These responses resulted from the activation of allergen-specific Th2 cells, as demonstrated by IL-4 and IL-5 secretions by splenocytes. Balb/c mice could also be orally sensitized to CM. In the peanut and CM models, we demonstrate the occurrence of a local anaphylactic reaction.
These results are in accordance with studies demonstrating the value of Balb/c mice in oral sensitization protocols. In fact, the oral sensitization of Balb/c mice after administration of ovalbumin (Ova) alone (20) or mixed with CT (22) was evidenced. In the latter study, sensitized Balb/c mice were further i.p. challenged with Ova, and an increase in serum histamine levels, comparable with that induced in C3H/He mice, was demonstrated. Oral sensitization of Balb/c mice was also successfully performed after gavage with BLG mixed with 10 μg of CT, allowing optimal specific IgE induction (21).
However, our results are in conflict with those comparing the sensitivity of C3H/HeJ and Balb/c strains after oral administration of antigens and CT (19). In this work, 3- or 5-week old C3H/HeJ mice developed CM and peanut allergic reactions, respectively, whereas age-matched Balb/c mice exhibited no hypersensitivity reactions when subjected to the same CM and peanut sensitization and challenge regimens. Balb/c only developed slight peanut-specific IgE responses, whereas no CM-specific IgE were detectable. At the cellular level, Balb/c mice developed a clear Th1 phenotype whereas C3H/HeJ mice developed a clear Th2 one. In other studies using C3H/HeJ mice, increases in the CT and the allergen doses and the number of gavages raised specific IgE levels, anaphylaxis scores and plasma histamine concentrations (33, 34), showing that optimal sensitization requires optimization of protocols of allergen administration. Accordingly, another study clearly demonstrates in Balb/c mice that CT has a clearly dose-dependent effect on the specific antibody response, whereas the kinetics of the response were more affected by the antigen dose (4). In our CM model, we sensitized 3-week old mice with threefold less antigen and twofold more CT than in Morafo's model. However, these discrepancies only partially explain the conflicting results, and others main contributing factors can be source of allergens, volume and form of administration (20 μl/g of emulsified CM or WPPE in our study), pretreatment or not before sensitizations (anti-acidic treatment in Li's studies vs no pretreatment in ours), mice origin (Jackson Lab. vs CERJ), composition of food (milk-free in our experiments), conditions of maintenance of the mice… All those factors are often variable from one lab to another one, and the implication of all those factors further demonstrates the difficulty developing a suitable and reproducible animal model in the field of allergy.
In the CM allergy model proposed by Li et al. (17), systemic anaphylaxis was proposed to be the consequence of intestinal anaphylaxis events, which allow CM antigen to pass rapidly into the systemic circulation and induce mediator release from mast cells and other cells of extra-intestinal sites. Intestinal anaphylaxis was demonstrated in C3H mice after oral sensitization with hen's egg lysozyme and CT, using Ussing chambers (22) and was also shown to be associated with mast cell degranulation in the intestinal wall in orally sensitized NMRI mice (35). All these results together suggest that intense degranulation in intestinal wall may lead to increased permeability of the mucosal barrier, and then possibly to histamine passage from mucosa to lumen. This correlate with our results then demonstrating that oral sensitization with CT in Balb/c mice also acts as allergen challenge and leads to the elicitation of an allergic reaction.
After i.p. or oral sensitization of Balb/c mice, we demonstrated that IgE and IgG1 shared common specificity, with major recognition of polymer of glycinin fragments, containing proteolytically processed Ara h3/4 polypeptides. Identification of Ara h3 as a peanut allergen recognized by 44–53% of peanut allergic patients depended on recombinant Ara h3, expressed as a 36 or 60 kDa recombinant protein (13, 15). However, ELISA tests performed in our lab clearly identified the natural, peanut-derived, glycinin polymer as a major peanut allergen (H. Bernard et al., unpublished data). Concomitant with these results, peanut-derived Ara h3 has recently been described as a heteromultimeric protein with an apparent molecular weight (Mw) of approximately 400 kDa, comprising polypeptide subunits of Mw ranging from 14 to 45 kDa. Ara h3 adopts the structural organization of soy glycinin, with an 11S behaviour, and was shown to be proteolytically processed in peanut (25). Further studies should be done in human allergic patients to assess the actual clinical relevance and importance of this natural allergen vs the recombinant one which is not proteolytically processed and associated as its natural counterpart.
The IgE induced in orally sensitized Balb/c mice also recognized, although to a lesser extent, Ara h1 and slightly Ara h2, which have been shown to be major peanut allergens in numerous studies. These results were comparable to the immunoblot demonstration that peanut-specific IgE induced in C3H/HeJ mice recognized Ara h1 and Ara h2.
In the CM-sensitization model, induced IgE were specific to whole caseins, BLG, serum bovine albumin and lactoferrin. Specific IgE have been analysed in 80 sera collected in the late 1990s from CM-allergic patients and the frequency of sensitization to BLG, α-lactalbumin and lactoferrin was 51, 19 and 21, respectively, lower values than those described 10 years ago. In contrast, sensitization to caseins and BSA largely increased, reaching 54 and 72% of the studied population (3). Major human allergens, i.e. casein, BSA and BLG, were then identified in our mouse model. Surprisingly, anti-α-lactalbumin IgE were undetectable whereas high anti-lactoferrin IgE titres were noted. The highest reaginic response induced in rat i.p. sensitized with CM and carrageenan was specific to caseins and lactoferrin (36), in accordance with our results.
In conclusion, our results further validate the use of CT and oral sensitization in Balb/c mice to study food allergy. Additional studies will be performed to analyse the fine specificity of specific IgE in Balb/c mice, i.e. against different isoforms and linear epitopes.