- Top of page
- Materials and methods
Background: Mosquito salivary proteins cause allergic reactions in humans. The allergenicity, clinical relevance, and species cross-reactivity of a new 37-kDa recombinant mosquito (Aedes aegypti) salivary allergen, rAed a 2, were evaluated.
Methods: rAed a 2 was expressed using a baculovirus/insect cell system and purified. Its allergenicity was examined using an enzyme-linked immunosorbent assay (ELISA), ELISA inhibition tests, immunoblots, and skin tests. Epicutaneous tests with the allergen, mosquito whole body extracts, and mosquito bite tests were performed on 48 volunteers. Serum rAed a 2-specific immunoglobulin E (IgE) was measured in individuals with positive mosquito saliva-specific IgE and negative controls.
Results: Both immunoblots and ELISAs demonstrated that rAed a 2 bound to the IgE of mosquito-allergic individuals. The binding could be inhibited by the addition of a natural mosquito preparation. Furthermore, rAed a 2 induced immediate and delayed skin reactions. Ten per cent of 31 participants with a positive mosquito bite test had positive skin reactions to rAed a 2, compared with 32% for mosquito whole body extract. None of the participants with a negative bite test showed positive reactions to either of the two extracts. Forty-three per cent of individuals with positive saliva-IgE had positive rAed a 2-IgE. rAed a 2 was a species-shared allergen, being present in the saliva of the 11 species studied.
Conclusions: rAed a 2 has identical antigenicity and biologic activity to its native form. It can be used in the diagnosis of mosquito allergy, and is an important species-shared antigen.
Due to the lack of availability of mosquito salivary preparations, allergic reactions to mosquito bites are under-diagnosed and under-treated (1). Mosquito bites commonly trigger an immediate wheal and flare and/or delayed induration in the skin, but in individuals who are allergic to mosquito saliva, large local reactions and even systemic reactions such as generalized urticaria, angioedema, asthma, and anaphylaxis may occur (2–9). The mechanisms underlying allergic reactions to mosquito bites involve immunoglobulin E (IgE) antibodies, lymphocyte proliferation, and probably IgG antibodies as well (1, 2, 7, 10). Immunotherapy with injections of mosquito whole body extracts may prevent allergic reactions to subsequent bites (3–6, 11, 12), however, mosquito whole body extracts, currently available commercially, contain few salivary proteins and many nonsalivary proteins (13) that may cause new sensitizations and side effects. Recent advances in clinical immunology involving recombinant DNA technology have produced many recombinant allergens and their innovative derivatives (14–19). Utilization of molecular biologic techniques to produce pure mosquito salivary allergens will make the diagnosis and treatment of mosquito allergy more accurate and safe, and will facilitate the development of innovative allergen immunotherapy possible.
Mosquito salivary antigens have been identified by immunoblot analyses. A number of mosquito antigens with molecular masses ranging from 14 to 126 kDa have been reported in various mosquito species (20–22). To date, only four recombinant mosquito salivary proteins, designated rAed a 1 (23), rAed a 2, rAed a 3 (24), and rAed a 4 (25) have been identified as allergens. Aed a 2, a 37-kDa protein belonging to the D7 family of mosquito salivary proteins with unknown function, is found in the adult female salivary glands of the mosquito Aedes aegypti (26). In this study, we expressed rAed a 2 using a baculovirus/insect cell system, purified it, and examined its allergenicity, clinical relevance, and cross-reactivity with 11 mosquito species.
- Top of page
- Materials and methods
In the saliva of Ae. aegypti, the most common mosquito pest worldwide, there are at least eight allergens (22), and new ones are being discovered (32). The Aed a 2 antigen is the original protein (26) that defines a large family of D7-related proteins. A large number of cDNAs corresponding to these proteins have been isolated and characterized from salivary gland-specific libraries of a variety of hematophagous insects (31). Two types of D7 proteins have been identified, the so-called long (molecular weights in the 30+ kDa range) and short (molecular weights in the 16–20 kDa range) forms. It is not surprising that Aed a 2 antibody cross-reacts with saliva or salivary gland extracts from other mosquitoes (Fig. 5). Most of the cross-reactivity with short molecular weight forms is in Cx. quinquefasciatus, An. sininesis and Oc. triseriatus, although there also seems to be some reactivity with the higher molecular weight forms in Cx. quinquefasciatus. Our data confirm the general antigenicity of this family of salivary gland proteins.
Baculovirus/insect cell systems have been used extensively for the production of large amounts of recombinant proteins with biologic activity (23, 33). The ability of insect cells to perform many of the post-translational protein modifications found in eukaryotic cells is a major advantage over prokaryotic expression systems. Moreover, the growth of insect cells in serum-free medium makes the purification of the recombinant protein relatively easy using simple laboratory techniques. Proteins expressed using this system can be used in both in vitro and in vivo tests (14). In contrast, proteins expressed by using prokaryotic expression systems, e.g. Escherichia coli, have no carbohydrates and are generally only used in in vitro assays (15, 17). The efficiency of expression of baculovirus system varies up to 1000-fold from gene to gene. In our study, about 10 mg of recombinant protein was obtained from 1 litre of culture medium using small-scale batch fermentation in spinner cultures.
rAed a 2, expressed using a baculovirus/insect cell system, has shown to be immunogenic in both in vitro and in vivo tests. It is extremely difficult to isolate native Aed a 2 from mosquito saliva for a direct comparison with rAed a 2. However, indirect evidence indicates that rAed a 2 has identical immunogenicity and biologic activity to native Aed a 2. First, in Western blotting, rAed a 2 bound to the serum IgE antibodies, which were induced by natural Aed a 2 in mosquito saliva, i.e. the IgE in mosquito allergic sera (Fig. 1, left), and natural Aed a 2 in the saliva was able to bind to the rabbit antibody immunized with rAed a 2 (Fig. 1, right). Secondly, the binding of rAed a 2 to human IgE in ELISA could be inhibited by addition of natural Aed a 2 in mosquito head and thorax extract (Fig. 2). In addition to the above in vitro experiments, rAed a 2 itself induced skin immediate and delayed reactions, as the mosquito whole body extract and mosquito bite tests did, indicating that rAed a 2 has biologic activity. Significant correlation of serum rAed a 2- IgE levels with mosquito saliva-IgE levels provides additional evidence supporting its identical immunogenicity to native form in the saliva.
The in vivo allergenicity of rAed a 2 has also been investigated in mice (34). Injections of rAed a 2 without adjuvant induced a significant increase in IgE and IgG1 (Th2-driving antibodies), but not IgG2a (Th1-driving antibody) in BALB/c and C57BL/6 mice. In all sensitized mice, a positive immediate skin reaction was observed. In immunoblot analysis, the mouse IgE antibodies elicited by rAed a 2 bound not only to rAed a 2 but also its natural form in mosquito saliva.
The positivity rate of epicutaneous tests with rAed a 2 (10%) and the mosquito whole body extract (32%) seems to be low in mosquito bite-test positive individuals. In mosquito bite tests, fresh salivary proteins are injected directly into the skin. In our previous experiments, each salivation of a female adult Ae. aegypti produces approximately 0.2 μg of saliva, while in epicutaneoud tests, each prick introduced about 3 × 10−6 ml of the test solution into the epidermis (30), that is, the equivalent of 0.0030 μg of rAed a 2 or 0.0095 μg of the whole body extract was introduced in each epicutaneous test. The small amount of antigen extract introduced may account for the low positivity rate of both rAed a 2 and the whole body commercial extract. As Aed a 2 is one of eight allergens in the saliva of Ae. aegypti, it is to be expected that the positivity rate induced by a single allergen in the saliva would be lower than by the mosquito whole body extract. However, in comparison with the 68-kDa rAed a 1, which induces positive skin reactions in 29% (23) of the participants with a positive mosquito bite test, rAed a 2 is a less common allergen.
Our long-term goal is to clone and express all of the eight allergens identified in the saliva of Ae. aegypti (22). Using a cocktail of all the recombinant allergens, we will eventually develop highly sensitive immunoassays for the diagnosis of mosquito allergy. If positive, then using each of the allergens separately will further identify the particular allergen(s) to which the individual is sensitized. Forty-three per cent of individuals with a positive serum IgE to saliva had a positive IgE to rAed a 2, and as a species-shared allergen present in all Aedes mosquito studied and other two species, rAed a 2 may have broad applicability in the diagnosis of mosquito allergy.
In conclusion, we report the second recombinant mosquito salivary allergen rAed a 2. In addition, characterization of the third and fourth recombinant salivary allergens of Ae. aegypti, rAed a 3 (24), and rAed a 4 (25), are in progress. These recombinant salivary allergens will greatly facilitate the diagnosis and immunotherapy of mosquito allergy.