Sequence- and Structure-Based Immunoreactive Epitope Discovery for Burkholderia pseudomallei Flagellin

Abstract

Burkholderia pseudomallei is a Gram-negative bacterium responsible for melioidosis, a serious and often fatal infectious disease that is poorly controlled by existing treatments. Due to its inherent resistance to the major antibiotic classes and its facultative intracellular pathogenicity, an effective vaccine would be extremely desirable, along with appropriate prevention and therapeutic management. One of the main subunit vaccine candidates is flagellin of Burkholderia pseudomallei (FliCBp). Here, we present the high resolution crystal structure of FliCBp and report the synthesis and characterization of three peptides predicted to be both B and T cell FliCBp epitopes, by both structure-based in silico methods, and sequence-based epitope prediction tools. All three epitopes were shown to be immunoreactive against human IgG antibodies and to elicit cytokine production from human peripheral blood mononuclear cells. Furthermore, two of the peptides (F51-69 and F270-288) were found to be dominant immunoreactive epitopes, and their antibodies enhanced the bactericidal activities of purified human neutrophils. The epitopes derived from this study may represent potential melioidosis vaccine components.

Fig 1. FliC_Bp B-cell linear sequence-based epitope predictions using three predictor methods: BepiPred (threshold = 0.35), BCPred (classifier specificity = 75%), and AAP (classifier specificity = 75%).

Epitope predictions were carried out on the full-length amino acid sequence of FliC_Bp, using three web-accessible prediction servers: BepiPred, BCPred and AAP [21,22]. Predicted epitope residues by BepiPred, BCPred and AAP are shown in red, blue and green font, respectively on the amino acid sequence (residue numbers are indicated). Grey shaded boxes indicate consensus positive residues identified by at least two epitope predictors.

Fig 2. Recognition of predicted epitope peptides by rabbit anti-FliC_Bp sera as detected by Indirect ELISA.

Controls (A) composed of human serum albumin (HSA), crude B. pseudomallei, recombinant FliC_Bp, without primary antibody (No 1⁰ Ab) and without secondary antibody (No 2⁰ Ab) and F51-69, F96-111, F270-288 and F213-231 peptides (B) were coated at various concentration (1, 3 and 10 μg/ml) onto a 96-well polystyrene plate and probed with diluted rabbit antibodies to FliC_Bp and quantified by indirect ELISA. Results are represented by the Absorbance index (O.D._test−O.D._uncoated / O.D._uncoated). Experiments were performed in duplicate and results represent the mean of the Absorbance index ± SE.

Fig 3. Distribution of human antibody against B. pseudomallei related proteins and peptides among seronegative (S-; n = 24), seropositive (S+; n = 24) and melioidosis recovered individuals (R; n = 24), detected by Indirect ELISA.

Crude B. pseudomallei antigens (A) recombinant FliC_Bp (B), FliC_Bp peptide F213-231 (C), F51-69 (D), F96-111 (E), F270-288 (F) were coated onto a 96-well polystyrene plate and probed with diluted plasma samples from healthy and recovered individuals and quantified by indirect ELISA. Results are represented by the Absorbance index (O.D._test−O.D._uncoated / O.D._uncoated). Experiments were performed in duplicate and results represent the mean of the Absorbance index ± SE. * P < 0.05, ** P < 0.01, *** P < 0.001, ns = not significant compared between plasma sample groups using one-tailed Mann-Whitney U test.

Fig 4. FliC_Bp peptides induce human IFN- and IL-10 production from PBMCs.

PBMCs (5 x 10⁵ cells/well) from 20 seropositive healthy donors were stimulated with culture medium alone (Med), fixed Bps (PBMCs:organism = 1:30), 3 μg/ml of phytohaemagglutinin (PHA), 10 μg/ml of FliC_Bp protein, and 50 g/ml of FliC_Bp peptides (F51-69, F96-111, F270-288 and F213-231) for 48 h, and the IFN- and IL-10 levels were quantified by ELISA. Vertical lines represent the median and dashes indicate the detection limit.

Fig 5. Rabbit anti-FliC_Bp predicted peptide antibodies enhance phagocytosis (A), oxidative burst (B) and bacterial killing (C) against B. pseudomallei in purified human PMNs (N = 5).

Rabbit anti-FliC_Bp, pre-bleed anti-sera at dilution 1:50 or anti-predicted peptide antibodies at 40 g/ml were used for opsonization. * P < 0.05, ** P < 0.01, *** P < 0.001, ns, not significant compared between with or without antibody groups using one-tailed paired t test.

Fig 6. Domain organization and 3D crystal structure of FliC_Bp.

Schematic secondary structure ribbon representation of the crystal structure of FliC_Bp. The N- and C-termini are labeled as are the diverse secondary structural features with α-helices in cyan and β-strands in magenta. Peptides 96–111, 213–233 and 270–288 are highlighted in red, green and black, respectively. This figure was produced using MacPymol.