Conformational Epitope Prediction of Birch Betv 1 and Hazel Cor A1 Towards B-Cells

Background: White birch and hazel allergens, namely Bet v1 and Cor a1 are known allergens, but their allergen specificity is not yet characterized. Objective: To map the antigenic determinants responsible for IgE binding utilizing in silico modelling and docking of the peptides against IgE antibody. Methods: The antigen sequences were cut into peptides are docked against the IgE antibody and those with the highest docking scores are further studied for the bond interactions. The overlapping sequences of the high score peptides are observed in the whole antigen model to predict their position. The residues at bond interactions also been reported for these overlapping peptide sequences. Results: The validation is done by antigen-antibody docking studies to confirm the predicted epitope. 25% of the world population suffers from allergic rhinitis and 15% of them develop asthma. Conclusion: Negative binding energies of the studied pollen allergens with IgE confirm their allergenicity. Based on the results of overlapping peptides PF 3,4 and PF 16,17 to play a key role in the allergenic response of white birch and Common hazel.


INTRODUCTION
In the adaptive immune system, B cells play an essential role in protecting the human body against various pathogenic molecules. Specifically, B cells belong to humoral immunity that is mediated through antibodies. In response to exposure to pathogens, B cells develop antibodies that bind to and neutralize the target. However, pathogens are not identified by B-cells as a whole, but through molecular components known as antigens. The component of an antigen that is detected by the immune system, primarily by B cells or T cells, is the epitope, also known as the antigenic determinant. The majority of the B cell epitopes are conformational (discontinuous), while the remaining (merely 10 percent) are linear B-cell epitopes (continuous). For immunologists, accurate detection of conformational B-cell epitopes is still a major hurdle [1]. The major plant allergens of white birch (Betula verrucosa) and common hazel (Corylus avellana) are the antigens chosen for this analysis. Allergic hypersensitivity or allergy is a reaction that happens in an individual when the same allergen is introduced into that person who has developed IgE antibodies in response to that antigen or allergen priorly [2]. The most common causative agents of allergic reactions are pollen grains from plants, foods, bee stings, dust mites, molds, fungal spores, animal epithelia, fur and feathers, animal dander, latex. Studies have indicated that allergic disorders such as anaphylaxis, hay fever, atopic dermatitis, eczema, asthma, and many other respiratory and pulmonary diseases affected about 25% of the world population which are primarily caused by aero-allergens. The major hazel allergen Cor a1 and major birch pollen allergen Bet v1 are homologous to each other [3]. According to researchers, about 53% of people who are allergic to birch pollen also have crossreactivity to Cor a1 as well. People who are allergic to birch pollen (70%) also exhibit hypersensitive responses towards different seeds, fruits, nuts, and roots than those without allergy to birch pollen (19%). Hazel allergy is very widespread across European countries and has also been shown to be the most prevalent source of food allergy mediated by IgE [4]. The present work focused on the in silico molecular characterization of white birch and common hazel Bet v1 and Cor a1 allergen-derived peptides, respectively.

Retrieval of Allergen Sequences and Antibody Structure
Bet v1 (P15494) and cor a1 (Q08407) allergens were retrieved for the FASTA sequences of the major allergens from uniprot belonging to a white birch and hazelnut. In general, immunoglobulin E (IgE) is an instinctive response shown by the human immune system to any antigen that has breached into the body. The IgE antibody structure with zero mutations and having a Fab macromolecule also was retrieved from the RSCB PDB database [5][6].

Overlapping and Cutting of Peptides
To determine the epitope present in the allergen, the individual peptides are docked against the antibody. For this, the entire allergen sequences that were retrieved from UniProt are cut into short overlapping peptide fragments using Sigma-Aldrich tools [Peptide Library design and calculator tool]. Sigma Aldrich's "Overlapping Peptide Fragment Library" tool is used to chop the peptides with a convenient amino acid gap into the appropriate lengths and also hydropathy index. This tool of Sigma Aldrich is used to cut the amino acid sequence into short peptides of length 10 and with five overlapping amino acids [7][8].

Protein-Peptide Docking
HPEPDOCK is used to conduct a docking check for the binding site attributable to a protein receptor structure and a peptide sequence, allowing the peptide to be completely flexible and predicting the protein-peptide complex structure, beginning from random peptide conformations and locations. Computational docking techniques are used to scan for rotational space between a protein receptor and its peptide-binding partner in all possible binding modes. Antibody is docked against each overlapping peptide fragment of the pollen sequence. All potential models for each peptide fragment along with their docking scores are estimated by the HPEPDOCK docking. Peptide model with the highest docking score is identified and picked for every overlapping peptide fragment and docked against the antibody, among all potential models that have been predicted by the server [9].

Docking Analysis
The antibody and peptide interactions are studied through Schrodinger's software, for the presence of non-covalent bonds and pi-pi interactions between antibody and antigen peptides and are analyzed through the Maestro Viewer (data not provided). The 3D structures of both the antibody and the antigen peptide model with the highest docking score according to HPEPDOCK are viewed in the Maestro workspace to identify the bond interactions between them. The type of bond that is formed between the atoms, the name & number of the atom, residues, and chains at where the bonds are formed for both antibody and peptide are noted [10].

Homology Modelling
Template selection is done through the BLAST tool for the prediction of the secondary structure of the antigen sequence that is retrieved from UniProt. Areas of similarities amongst various biological sequences are detected through BLAST, which helps in the comparison of protein or nucleotide sequences against database sequences and measures their statistical significance. Due to the very high similarity between both the antigen sequences, the protein data structure (ID:4a86, Birch major allergen) with 72.33% of similarity was selected as a template for building secondary structure for Hazel major allergen (Cor a1). The model is built or generated using SWISS-MODEL [11].

Antigen-Antibody Docking Studies
The complete antigen 3D models of both Birch (Bet v1) and Hazel (Cor a1) are docked against the IgE antibody for validation. The structure of the Birch allergen is retrieved from Protein Data Bank (PDB) (ID: 4A86). Whereas the structure of Hazel allergen is modelled through SWISS-MODEL. Both these structures are docked against IgE antibody whose structure is retrieved from PDB (ID: 2vxq). Docking is carried out through the software ClusPro which is a webbased server that is useful for direct proteinprotein docking. From all the models that are predicted by the ClusPro server, the models with high scores for both birch and hazel are considered for validation [12].

Overlapping Peptide Fragments
The birch pollen (Bet v1) and hazel pollen (Cor a1) sequences are cut into a length of 10-mers with a gap of five amino acids. Both the allergen sequences of chosen plant species are cut into 31 peptides each with the help of the Sigma Aldrich tool (Table 1). Peptide fragments with highest docking score acquired through HPEPDOCK ( Table 2). The bond interactions is seen between overlapping peptide fragments of Bet v1, Cor a1 and IgE antibody (Tables 3 and 4). Overlapping peptide fragments of Bet v1 and IgE antibody is shown in Fig. 1 and overlapping peptide fragments of Cor a1 and IgE antibody is shown in Fig. 2 respectively.  Pi Salt Salt

Homology Modelling
Template selection: The target sequence was searched with BLAST against the primary amino acid sequence contained in the SMTL. A total of 108 templates were found. Among them, the template with the highest sequence identity, MAJOR POLLEN ALLERGEN BET V1-A (PDB ID:4a86) is selected as a template with 72.33% similarity as shown in Table 5. Threedimensional structures of Birch pollen (Bet v1) were available in the PDB (4A86). Hazel pollen (Cor a1) structure was determined using the homology modelling application of the SWISS-MODEL as shown in Fig. 3.
Top hit from the Blastp analysis of Hazel pollen (Cor a1) with PDB ID 4A86 as a template and energy-based model was developed. The structural alignment of the model as evaluated by Ramachandran plot indicated that most of the (93.4%) amino acids fit into the most favored regions, 5.8% of the modelled Cor a1 residues fall into the additional allowed regions and the remaining were found in generously allowed regions. The ERRAT overall quality factor was 93.617, specifying that the model predicted was good. To identify allergen-IgE interacting sites, an IgE-allergen (protein-protein) docking study was undertaken. IgE antibody (PDB ID: 2VXQ) was retrieved and prepared by using Schrödinger's protein preparation wizard. Concurrently, all the simulated trajectory frames of the modelled allergen of Cor a1 were clustered based on the energy and deviations. The cluster center frame showing minimal energy, deviations, and fluctuations was chosen for docking studies. Tail-end sequences of the allergen were found intact with the paratope region of the antibody by the end of docking studies. To validate the importance of other amino acids in the allergen, the sequence was divided into overlapping peptides. The allergen sequences of Bet v1, as well as Cor a1, were processed using overlapping peptide fragment library software, and 31 different 10-mer peptides, were designed (Bet v1-31 and Cor a1-31) with an overlap of five amino acids.

DISCUSSION
Bet v1 is responsible for 60% of allergies with birch (Betula verrucosa) pollen released into the air affecting millions of people in spring [13]. The birch pollen allergen has different isoforms, all of which exhibit identical conformations, but different allergenic potentials [14]. IgE and IgG antibodies of patients with allergy to birch pollen serve as tools to define the allergen [15]. Up to 90% of the Bet v1-exposed individuals do exhibit IgE-mediated allergic cross-reactions (oral allergy syndrome) to Bet v1-homologous food allergens, such as hazel nut [16]. The threedimensional structure of Bet v1 and related pollen and food allergens including Cor a1 from hazelnut belong to the family of class 10 pathogenesis-related proteins (PR-10) within the Bet v1 superfamily. PR-10 proteins comprise about 160 amino acid residues with a molecular weight of 17.5 kDa. These proteins exhibit a canonical fold consisting of a seven-stranded antiparallel -sheet (1-7) and three  helices (1-3). The two short, consecutive helices 1 and 2 interrupt the -sheet between strands 1 and 2 while the long C-terminal helix 3 is located above the -sheet, creating a large and fairly hydrophobic cavity in the protein interior [17]. Cor a1 shares 67% sequence identity with Bet v1 and shared similar tertiary structures based on the homology modelling. As with Cor a11, structural flexibility in Bet v1 is distributed across the entire PR-10 scaffold, including secondary structure elements and loops [18]. Whether an allergen induces strong immediatetype hypersensitivity reactions in sensitized allergic patients is largely determined by its ability to induce IgE-mediated degranulation of mast cells and basophils [19]. The process of degranulation is dependent on cross-linking of cell-bound IgE antibodies and hence requires the presence of at least two IgE epitopes on the allergen [20]. The IgE antibodies appear to recognize primary conformational epitopes on allergens [21]. Conformational epitope mapping using conventional strategies such as testing for IgE reactivity to recombinant or synthetic allergen fragments is not easy because fragmentation of proteins often leads to the loss of the threedimensional structure of the protein and hence to loss of IgE reactivity [22]. The onset of birch pollen-related food allergy is believed to be induced by primary sensitization to pollen allergens and subsequent development of secondary food allergy caused by IgE crossreactivity between homologous pollen and food allergens [23]. Bet v1-specific IgE antibodies were shown to cross-react at the T-cell level with Cor a1. [24]. In order to characterize IgE binding as a measure for allergenicity, we characterized the antibody-binding behavior of the Bet v1 [25]. IgE recognition of Bet v1 is not influenced by the bound ligands such as flavonoids [26]. We also sought to map the IgE epitopes on the threedimensional structure of Cor a1 [27]. Due to the lack of crystal structure of Cor a1, a homology 3D-model was employed for characterizing the epitopes on the surface of Cor a1 [28]. For each of the allergens, namely, Bet v1 and Cor a1, and their interaction profiles with Ig E antibodies, the antigen sequence was fragmented into a series of overlapping peptides and their binding modes against IgE were studied. RMSD and RMSF from the simulation results were found to be in the acceptable range of 1-3 A°. The ERRAT score indicates the overall stability of the modelled Cor a1 protein. Sequential IgE epitope analysis was performed to study IgE epitopes that recognize birch pollen and hazelnut allergens at the level of peptides [29]. Our results confirmed a few sequential IgE epitopes, which were found in similar locations and the homology of the amino acid composition of the epitopes of the two allergens was relatively high [30]. The identified sequential epitopes mapped to the Bet v1 threedimensional structures indicate that these residues are exposed on the protein surface and are spread over the 1-1 regions, 6, 5 and 4 regions in case of Bet v1 [31].

CONCLUSION
The generated model could be supportive to understand the functional characteristics of Cor a1 and Bet v1 against IgE. The in silico molecular modeling and validation studies is helpful to understand the structure, function and mechanism of proteins action. We here display the usefulness of allergen-specific IgE antibody as a tool in studies of the crucial molecular interaction taking place at the initiation of an allergic response. Such studies may aid us in development of better diagnostic tools and guide us in the development of new therapeutic compounds.

CONSENT
It is not applicable.

ETHICAL APPROVAL
It is not applicable.