A total of 12 subjects were included in the study, comprising two patient groups: (A) 6 allergic rhinitis patients (aged 15–34; 3 males and 3 females; recruited between January and June 2015) with positive skin prick test (allergens were supplied by ALK-Abelló, Copenhagen, Denmark) and positive serum IgE test to the P. acerifolia pollen extract by using ImmunoCAP assay (Phadia AB, Uppsala, Sweden); (B) 6 healthy controls (aged 20–45; 3 males and 3 females; recruited in June 2015). The study protocol was approved by the ethical committee of the First Affiliated Hospital of Nanjing Medical University (Nanjing, China). Written informed consent for the use of blood samples were obtained from all participants before study entry according to the Declaration of Helsinki.

The complete nucleotide acid and amino acid sequences of Pla a 3 allergen were retrieved from the GenBank database. The open reading frames (ORF) of Pla a 3 allergen had 354 bases pairs, encoding 118 amino acids. Because this ORF contained a 26 amino acid signal peptide, the mature Pla a 3 allergen had 276 bases pairs, encoding 92 amino acids.

The nucleotide acid of mature Pla a 3 allergen was synthesized by GenScript (Piscataway, NJ, USA) and sub-cloned into pSUMO-Mut vector using Stu I and Xho I sites and verified by DNA sequencing. The recombinant pSUMO-Mut-Pla a 3 plasmid was transformed into Arctic Express™ (DE3) RP E. coli host strain.

A positive colony was selected to inoculate in a 3 ml lysogeny broth (LB)-kanamycin broth (Shanghai Sanger Biotech Technology, Co., Ltd., Shanghai, China), and incubated at 37°C overnight. A total of 1% of the overnight culture of the transformant was transferred into fresh LB-kanamycin broth and cultured at 37°C with shaking at 200 × g up to the absorbance 0.6–0.8 at 600 nm. The culture was induced with 1 mM isopropyl-b-D-thiogalactopyranoside (IPTG) and harvested following an additional 4 h incubation. The culture was disrupted by sonication at 60 kHz, 4 sec pulse-on, 8 sec pulse-off and centrifuged at 10,000 × g for 30 min at 4°C. The results indicated that recombinant Pla a 3 was mainly in supernatant fraction (Fig. 1) and purified by Nickel affinity chromatography (GenScript). The washing buffer contains 100 mM NaH₂PO₄, 20 mM Tris-HCl, 10 mM imidazole, pH 8.0, and the eluting buffer contains 100 mM NaH₂PO₄, 20 mM Tris-HCl, 250 mM imidazole, pH 8.0.

An IgE binding assay was conducted as previously described. Briefly, proteins were analyzed via 12% sodium dodecylsulfatc-polyacrylamide gel electrophoresis (SDS-PAGE) (21). The gel was transferred to polyvinylidene difluoride membranes (22). The blots were blocked in 5% skim milk for 2 h, then incubated with serum of P. acerifolia pollen allergic patients (diluted 1:40 in PBS) as the first antibody overnight at 4°C. Then, IgE that bound to the allergen was detected with a horseradish peroxidase-conjugated goat anti-human IgE mAb (1:3,000, catalog no. A9667, Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) at 4°C for 1 h and then detected using a ImageQuant LAS 4000 mini detection system (GE Healthcare Life Sciences, Chalfont, UK). A total of 6 healthy individuals were used as a negative serum control.

The Pla a 3 allergen protein sequence was searched for homology in the PDB. As well, the homologous templates suitable for Pla a 3 allergen were selected from SWISS-MODEL server (http://swissmodel.expasy.org/) (23). SWISS-MODEL Repository is a database of 3D protein structure models generated by The Pla a 3 allergen protein sequence Homology modeling. The Pla a 3 allergen protein sequence was searched for homology in the Protein Data Bank. In addition, the homologous templates suitable for the Pla a 3 allergen were selected from the SWISS-MODEL server (http://swissmodel.expasy.org/) (23). The SWISS-MODEL Repository is a database of 3D protein structure models generated by the SWISS-MODEL homology-modeling pipeline (24). The best template was retrieved from the results of previous methods and used for homology modeling. Pla a 3 allergen modeled protein structure was built through alignment mode in SWISS-MODEL using the complete amino acid sequence.

A total of three immunoinformatics tools including the DNAStar protean system, the Bioinformatics Predicted Antigenic Peptides (BPAP) system (http://imed.med.ucm.es/Tools/antigenic.pl) and the BepiPred 1.0 server (http://www.cbs.dtu.dk/services/BepiPred/) were used to predict the B cell epitopes of the Pla a 3 allergen. The ultimate consensus epitope results were obtained by combining the results of the three tools together (25). In the DNAStar protean system, four properties (hydrophilicity, flexibility, accessibility and antigenicity) of the amino acid sequence were chosen as parameters for epitopes prediction (26). The peptide regions with good hydrophilicity, high flexibility, surface accessibility and high antigenic index were chosen as candidate epitopes for further investigation.

T cell epitopes are principally predicted indirectly by identifying the binding of peptide fragments to the major histocompatibility complexes (MHCs). The binding significance of each peptide to the given MHC molecule was based on the estimated strength of binding exhibited by a predicted nested core peptide at a set threshold level. SYFPEITHI and NetMHCII-2.2 (http://www.cbs.dtu.dk/services/NetMHC II-2.2) (27) were used to predict T cell epitopes of the Pla a 3 allergen. In the current study, HLA-DR10101 and HLADR50101 were used to predict HLA-DR-based T cell epitope prediction. As a result, the ultimate results of T epitopes were obtained by combining the results of the SYFPEITHI and NetMHCII-2.2. B cell and T cell epitopes identified by immunoinformatics tools were mapped onto a linear sequence and presented on the three dimensional model of the Pla a 3 allergen to determine their position.

The Pla a 3 allergen was subcloned into a pSUMO-Mut vector and transformed into the E. coli host strain. The culture was induced with IPTG and expressed at 37°C for 4 h and the resulting protein presented a band with a apparent molecular weight of 28 kDa (Fig. 1A; lane 2 and lane 3), and no such band was seen in the non-induced cells (Fig. 1A; lane 1). Protein was produced primarily in the supernatant fraction following sonication (Fig. 1A; lane 5) and purified by Ni column. Finally, the purity of the purified Pla a 3 allergen was identified by SDS-PAGE. As presented in Fig. 1B, the recombinant Pla a 3 allergen was successfully purified by elution buffer.

The ability of recombinant Pla a 3 allergen to bind IgE from the P. acerifolia pollen allergy patients' serum was determined by western blotting. Fig. 2 demonstrated that mixed P. acerifolia pollen allergy patients' serum revealed positive IgE reactivity to the Pla a 3 allergen, but healthy controls failed to.

Searching for the proteins with known tertiary structure in the Protein Data Bank, 1t12.1.A (PDB accession number) reported the highest sequence identity (62.64%) to the Pla a 3 allergen. Therefore, the best template 1t12.1.A was used for homology modeling. Fig. 3A demonstrated the overall 3D structure of the Pla a 3 allergen, and Fig. 3B demonstrated the predicted B cell and T cell epitopes superimposition on the surface of the Pla a 3 allergen structure.

Surface accessibility and fragment flexibility are important features for predicting antigenic epitopes. In addition, the existence of regions with high hydrophobicity provides strong evidence for epitope identification. Antigenic index directly indicated the epitope forming capacity of the Pla a 3 allergen sequence. Based on these sequence properties, the final predicting regions of the Pla a 3 allergen by DNAstar were obtained as: 35–45 and 81–86 (Table I). The predicted results of the BPAP system were 4–17, 19–34, 45–54 and 63–88. The predicted results of the BepiPred 1.0 server were 22–26, 35–47, 57–59 and 81–87. Furthermore, the final potential B cell epitopes of the Pla a 3 allergen were selected on the basis of the results of these three tools. The ultimate results of the three immunoinformatics tools finally predicted two peptides (35–45 and 81–86) and these peptides are presented in Fig. 3.

SYFPEITHI and NetMHCII-2.2 were used to identify the T cell epitope of the Pla a 3 allergen. The predicted results of SYFPEITHI were 4–18, 29–37, 57–63, 67–73 and 77–85 (Table I). For HLA-DR-based T cell epitope prediction, the final predicting regions of HLA-DR10101 and HLA-DR 50101 are presented in Table I. As a result, the Pla a 3 allergen was predicted to have four T cell epitope sequences including 2–15, 45–50, 55–61 and 67–73, as presented in Table II.