Data on the epitope mapping of soybean A2 and A3 glycinin

The data information provided in this article relate to our research article “Using patient serum to epitope map soybean glycinins reveals common epitopes shared with many legumes and tree nuts” (Saeed et al., 2016) [1]. Here we provide western blot detection of glycinin subunits by soy-sensitive human sera, ELISA screens with overlapping synthetic peptides (epitope mapping), and various database/server epitope searches.

Epitope sequence similarity searches were done using the SDAP website: (http://fermi.utmb.edu/) B-cell epitope predictions were done using the following servers: ABCpred (http://www.imtech.res.in/raghava/abcpred/) BepiPred  It may also contribute to the generation of hypoallergenic soybean cultivars. Provide risk assessment tools for the evaluation and characterization of the allergenicity of novel foods.

Data
The data presented here show the western blot detection of A2 or A3 subunits by soy-sensitive human sera (Fig. 1) and ELISA screens (Figs. 2 and 3) of these patient sera with overlapping synthetic peptides (Pepsets). Serum specificity is also confirmed by cross-screening the A2 Pepset with a serum that does not bind to the A2 cluster on western blot (Fig. 4). Also contained in this article is SDAP (Structural Database of Allergenic Proteins) sequence similarity search results (Tables 1 and 2) of the epitopes reported by Saeed et al. (2016) [1] and theoretical B-cell epitope prediction data on the full length sequences of A2 and A3 subunits (Table 3).

Patient serum
Soy-sensitive human sera used in the western blots and epitope mapping are previously described [1].

Immunoblot analysis
Western blotting of human sera was conducted as previously described [2]. Membranes were hybridized with serum dilutions ranging from 1/50 to 1/500.

Epitope mapping
Two peptide sets representing the mature amino acid sequences of glycinin A2 (P04405, 90 peptides) and A3 (BAB15802, 104 peptides) were synthesized and biotinylated by Mimotopes (http:// www.mimotopes.com) via parallel array platform. Quality Control Assurance was provided for both peptide synthesis and biotinylation by reverse phase HPLC (RP-HPLC), and by mass spectrometry (MS) respectively. The biotinylated 12-mer peptides, frame-shifted by three residues were used as per manufacturer's instructions (Application/Method PT3013). DMSO was used to resuspend the dry peptides and streptavidin-coated high capacity plates (Pierce #15500) pre-blocked with SuperBlock™ buffer were used to capture the biotinylated peptides. Serum was diluted at 1/50 in TBS-BSA 2% except for Patients 4 (1/100) and 5 (1/50 or 1/100). The secondary mouse anti-human IgE-HRP (Southern Biotech, Birmingham, Alabama, #9160-05) was diluted at 1/4000 in TBS-BSA 2%. SureBlue Reserve TM TMB microwell peroxidase substrate (KPL, Gaithersburg, Maryland, #53-00-01) was added to the plate, the reaction was stopped by acidification and colorimetric detection was performed on a Tecan Sunrise microplate reader with Magellan™ data analysis software (Tecan group AG, Männedorf, Switzerland) at 450 nm. Each experiment was performed in duplicate. Negative controls were performed using the same protocol, but the addition of human sera was omitted. The data was normalized by calculating the ratio of experimental to negative control and graphed.

B-cell epitope prediction servers
Three popular B-cell epitope prediction servers were tested with the A2 and A3 sequences. ABCpred server predicts B cell epitopes using a recurrent neural network (machine based technique) using fixed length patterns [3]. Lengths of epitopes varying from 10-16 amino acids were tested. BepiPred 1.0 server uses a combination of a hidden Markov model and a propensity scale method [4]. SVMTriP uses support vector machine integrating tri-peptide similarity and propensity scores [5], where epitope lengths varying from 10-20 amino acids were tested. In all cases, a higher score reflects a higher probability that a sequence is an epitope.    Amino acid number corresponds to position of full length sequence. A dash (-) indicates that no epitope was found. Scores are listed for the 3 different methods tested. Only the highest score is listed if the epitope was found in multiple lengths tested (10-20 mer).