Fig 1.
Structural representation of wild-type hGOT1 based on crystallographic coordinates PDB ID 3II0.
Fig 2.
(A) Closed and (B) open states of hGOT1 produced by 1 μs MD simulation. The cofactor is represented by sticks and colored by atom, and the P15-R32 “door” is shown in yellow. The color scheme of the protein corresponds to Fig 1. (C) RMSF calculations for the substrate bound and unbound simulations of hGOT1. Both monomers A and B were averaged in RMSF analysis. The blue plot represents the simulations of WT without the substrate (unbound form) whereas the red plot describes simulation of the enzyme with substrate bound. The large domain is in white, the small domain is in gray, and the loop residues are highlighted in yellow.
Fig 3.
PLP dynamics in the binding pocket of hGOT1.
(A) PLP RMSD values based on four 1 μs simulations of WT, E266K, R267H, and P300L. The blue plot represents WT, whereas E266K is shown in yellow, R267H is shown in purple, and P300L is shown in green. (B) RMSD distribution of PLP in each variant and WT. WT shows a much narrower distribution than the 3 variants. The color scheme is the same as in panel A. The representative PLP snapshots in the pocket are shown by (C) WT, (D) E226K, (E) R26H, and (F) P300L. The opposite monomer is colored in purple, and the amino acid substitutions are shown in green.
Fig 4.
PLP coupled hydrogen bond network for WT, E266K, R267H, and P300L complexed with PLP.
(A) Key residues involved in PLP stabilization in the hGOT1 binding site based on the PDB ID 3II0. (B) Refinement of the PLP hydrogen bond network in WT by MD simulation from the crystal structure. Blue bars represent the average occupancies from MD data and pink bars represent crystal data. (C) Differences in hydrogen bond occupancies between WT (blue bars), E266K (yellow bars), R267H (purple bars), and P300L (green bars) produced by MD simulation.
Fig 5.
Misalignment of the hGOT1 dimer interface determined by the first principal component (PC1).
A) Visualization of PC1 positive displacement (shown by yellow arrows) from PCA. Negative displacements correspond to the opposite direction of the arrows. The protein representation is the average coordinates of the inner Cα atoms (see Methods). B) The PC1 displacement of each variant as a function of simulation time. This displacement represents the magnitude of dimer interface misalignment. C) The monomer energy (EM) as a function of PC1 displacement. Ellipses show the standard deviation on both axes, and the mean values are located at the centers. D) and E) show the dimerization (EE) and PLP binding (EL) free energy versus PC1 displacement, respectively. The variants and WT follow the same color scheme as Fig 1. All values are relative to WT.
Table 1.
Absolute energies calculated for WT, E226K, R267H, and P300L systems with their standard deviations.
The monomer energy EM denotes the combined energy of the two monomers in the dimer, whereas EE represents the dimerization energy. PLP binding free energy is defined as EL. Both binding sites were included in the EL calculation and the sum of two are reported. The total energy (ET) in the last column is the sum of all energy components.
Fig 6.
1 μs MD simulation of the hGOT1 monomer with PLP.
A) Starting structure of the hGOT1 monomer showing PLP bound (colored in cyan) in the pocket (the monomer structure is based on PDB 3II0). The displacement of R267 from the PLP binding pocket observed after 100 ns of MD monomer simulation, shown as the transition from yellow to orange. B) The flipped out R267 removed essential binding interactions with PLP. The subsequent destabilization of PLP lead to its release from the binding pocket after 1 μs of MD simulation.
Fig 7.
Residue pairwise decomposition of interaction energies (kcal/mol) of the dimer interface and PLP binding.
To clarify the difference between WT and variants, the values for the E266K, R267H, and P300L were plotted relative to WT. Segments that have no interaction have been removed to increase the resolution of significant residues. The data represent the mean interaction energy over 800 ns, with vertical lines representing the standard deviation. A) Plots showing the individual contributions of each residue to the PLP binding energy (EL). B) Plots showing the individual contributions of each residue to the dimerization energy (EE).