Fig 1.
It shows all the steps that were applied in this work for the rational vaccine design to get highly immunogenic epitopes from EsxG and EsxH mycobacterial proteins. These are efficiently recognized by MHC-II and induce antibodies production using bioinformatics tools and Molecular Dynamics Simulations.
Fig 2.
(A). Linear B cell epitope prediction. Prediction in BepiPred 2.0. Residues that present a score higher than 0.46 are indicated with an E at the top of the sequence. Residues with values greater than 0.46 are colored in dark orange. (B). Discontinuous B cell epitope prediction. Prediction in DiscoTope 2.0. Epitopes predicted are shown as spheres. (C). Antigenic determinant prediction. Kolaskar and Tongaonkar´s prediction method was employed; x-axis—residue number; y-axis—antigenic propensity. Residues with antigenic propensity greater than 1.000 were predicted as potential antigenic sites.
Table 1.
Amino acid sequence from epitopes selected of protein EsxG and EsxH.
Fig 3.
Global structure behavior from EsxG EsxH dimer at 310 K.
(A). 0 ns of simulation. Arrows indicate regions that underwent folding. (B). 149 ns of simulation. (C). 500 ns of simulation. Blue depicts EsxG monomer; orange depicts EsxH monomer. (D). First principal component covariance matrix. The matrix is color-coded, from red (correlated displacements) to blue (non-correlated displacements). The diagonal line stands for the correlation between the residues paired with themselves; the blue stands for the correlation between each residue pair during the 500 ns simulation. (E). Contact map. The matrix is color-coded, blue (farthest) to red (closest). The diagonal line represents the zero distance between the residues paired with themselves, while spots represent the distances (nm) for each residue pair during the 500 ns simulation.
Fig 4.
Global structure behavior of EsxG at 310 K.
(A). at 0 ns of simulation. (B). at 57 ns of simulation. (C). at 250 ns of simulation. (D). First principal component covariance matrix. The matrix is color-coded, from red (correlated displacements) to blue (non-correlated displacements). The diagonal line stands for the correlation between the residues paired with themselves, while the color stands for the correlation between each residue pair during the 250 ns simulation. Global structure of the first cluster from EsxH at 310 K. (E). 0 ns of simulation. (F). 66 ns of simulation. (G). 250 ns of simulation. (H). First principal component covariance matrix.
Fig 5.
Average structures from EsxG EsxH dimer thermal unfolding simulations.
(A). 0 ns of simulation. (B). 5 ns of simulation at 350 K. (C). 50 ns of simulation at 350 K. (D). 5 ns of simulation at 400 K. (E). 5 ns of simulation at 450 K. (F). 5 ns of simulation at 500 K. (G). 5 ns of simulation at 550 K.
Fig 6.
Average structures from epitopes in solution and HLA-epitopes complexes.
(A). G1 epitope. (B). G2 epitope. (C). H1 epitope. (D). H2 epitope.
Table 2.
Hydrogen bonds occupancy in HLA-epitopes complexes.
Fig 7.
Solvent accessible surface area (SASA).
(A). G1 epitope. (B). G2 epitope. (C). H1 epitope. (D). H2 epitope. The blue bars depict the SASA from the epitope in solution. The green bars depict the SASA from the epitope in the MHC. The yellow bars depict the ΔSASA, subtracting the SASA of the epitopes in solution from the SASA of the epitopes forming the complex.
Table 3.
Characteristics of the selected epitopes.
Comparation of characteristics from epitopes in solution and when they are presented into HLA complex.
Table 4.
Solvent accessible surface area in tryptophan residues.
Fig 8.
Cleft formation in EsxG EsxH dimer at 310 K.
(A). 0 ns. (B). 149 ns. (C). 500 ns. Blue represents the EsxG monomer. Orange represents the EsxH monomer. Residues involved in cleft formation are presented in light blue (EsxG) and yellow (EsxH).