Structural and Functional Characterization of Allergens

During the past year, this project has succeeded in determining the structures of two important allergens. The structure of the cockroach allergen Bla g 1 revealed a previously uncharacterized protein fold that binds to lipid ligands, which is a common property of many allergens. The structure of the allergen Cat r 1 is the first example of a cyclophilin allergen derived from plants. Additionally, this project is beginning to move beyond the determination of single allergen structures and to investigate the interactions of allergens with the molecules in the immune system. This approach is currently proceeding with two molecular systems related to peanut allergy. First, we are looking at the chemical modification of peanut allergens due to roasting. Modifications such as those that result from roasting are commonly recognized by the receptor for advanced glycation end products (RAGE). These studies are the first to demonstrate that peanut allergens with modifications bind specifically to RAGE. Second, we are beginning an attempt to determine co-structures of antibodies bound to the major peanut allergen Ara h 2. Further accomplishments of each project are summarized below.Sensitization to cockroach allergens is a major risk factor for asthma. The cockroach allergen Bla g 1 and other homologous insect proteins are characterized by a tandem repeat of two sequences with about 25% identity, but the fold of the protein and the exact biological function are unknown. The structure of Bla g 1 was determined by X-ray crystallography. The Bla g 1 tandem repeat forms a novel fold with 6 helices that encapsulate a large and nearly spherical hydrophobic cavity of 3,750 A3. The X-ray data could identify some electron density in the cavity that might be long acyl-chain lipids. Using NMR and mass spectrometry, different lipid ligands were found associated with Bla g 1 depending on the source of the protein. Recombinant Bla g 1 from E. coli or P. pastoris bound phospholipids PE, PG, PS, and PC with various affinities, and cockroach frass was found to not contain phospholipids, but instead simple stearate and oleic acids. Further studies are needed to test whether or not these lipid ligands may contribute to the allergenicity of Bla g 1. Various phospholipids are known to bias the immune response towards a Th0/Th2 via presentation by dendritic cells on CD1 molecules, contributing to allergic disease.Cyclophilin allergens are considered pan-allergens due to their high crossreactivity; i.e. patients sensitized to just one source are usually highly allergic to the allergens from all other sources. This crossreactivity can include autoreactivity where the immune system mistakenly reacts against self-antigens. Indeed, some patients with chronic allergic disease, either asthma or atopic dermatitis, have demonstrated both humoral and cell-mediated autoreactivity. The reason for the crossreactivity is the high sequence identity between members of this protein family. In this period we determined the structure of the allergen Cat r 1, derived from the rosy periwinkle using NMR techniques. This is the first structure of a cyclophilin protein from plants. <P>Using the structure, we have been able to better understand the important residues that likely account for the cross reactivity between plant and mould allergens, and potential residues involved in autoreactivity with human cyclophilins. This knowledge will help in the rationale design of immunotherapeutics in that researchers may now design hypo-allergens that also avoid encouraging autoreactivity. The protein Ara h 2 is the most potent peanut allergen recognized by >90% of peanut allergic patients. The natural allergen and the recombinant construct used to determine the structure showed different patterns of recognition by patient sera. Based on these comparisons a major site of interaction (an epitope) for about 50% of patients was identified. This success has encouraged us to further map the patient epitopes using a panel of antibodies with various specificities for Ara h 2 and the homologous Ara h 6 allergen. Currently we have selected and produced the Fab fragment of several antibodies and are currently performing crystallization trials with complexes of Ara h 2.<P> It is our goal to further identify conformational epitopes on peanut allergens in order to better understand the patient response to peanut and to determine whether specific eptiope recognition correlates with any aspect of peanut allergic disease, e.g. risk of anaphylaxis, emergency room visits, or response to oral therapy. To do this we are collaborating Wesley Burks (UNC) in order obtain sera with well-documented patient histories. It is anticipated that further epitope mapping will provide information for the rational design of Ara h 2 hypoallergens for more effective and safer immunotherapies.It has been suggested that one of the reasons for the hypersensitivity to peanut allergens is due to food processing. Indeed, patient IgE more commonly recognizes roasted peanut proteins than raw. Roasting commonly leads to modifications of lysine side chains and formation of advanced glycation end products (AGEs). Recent immunonlogical data suggests that dendrictic cells preferentially uptake AGE modified protein and consequently upregulate expression of the receptor for AGE (RAGE).<P> In our studies, AGE modifications were found on Ara h 1 and Ara h 3 that included carboxymethyllysine, carboxyethyllysine, malondialdehyde, and dihyropyridine, in both raw and roasted peanut extract. Very few modifications were found on Ara h 2. Based on mass spectrometric analysis and Western blotting with allergen specific antibodies, RAGE was demonstrated to selectively pull down Ara h 1, Ara h 3, and AHY-3 from peanut extract. No Ara h 2 binding to RAGE was detected by Western analysis. Recombinant Ara h 1 with no AGE modifications did not bind RAGE, however after AGE modification Ara h 1 bound to RAGE. If the suggestion that sensitization to peanut allergens occurs in dendritic cells recognizing AGE modified peanut proteins is correct, these cells are likely interacting with modified Ara h 1, and Ara h 3, and not Ara h 2. With a better understanding of the immune pathways that lead to sensitization, we will hopefully be able to identify novel future therapies.