Computational Exploration of the Effects of Mutations on GABA Aminotransferase in GABA Aminotransferase Deficiency

Abstract

Gamma-aminobutyric acid (GABA) transaminase—also called GABA aminotransferase (GABA-AT)—deficiency is a rare autosomal recessive disorder characterized by a severe neonatal-infantile epileptic encephalopathy with symptoms such as seizures, hypotonia, hyperreflexia, developmental delay, and growth acceleration. GABA transaminase deficiency is caused by mutations in GABA-AT, the enzyme responsible for the catabolism of GABA. Mutations in multiple locations on GABA-AT have been reported and their locations have been shown to influence the onset of the disease and the severity of symptoms. We examined how GABA-AT mutations influence the structural stability of the enzyme and GABA-binding affinity using computational methodologies such as molecular dynamics simulation and binding free energy calculation to understand the underlying mechanism through which GABA-AT mutations cause GABA-AT deficiency. GABA-AT 3D model depiction was carried out together with seven individual mutated models of GABA-AT. The structural stability of all the predicted models was analyzed using several tools and web servers. All models were evaluated based on their phytochemical values. Additionally, 100 ns MD simulation was carried out and the mutated models were evaluated using RMSD, RMSF, Rg, and SASA. gmxMMPBSA free energy calculation was carried out. Moreover, RMSD and free energy calculations were also compared with those obtained using online web servers. Our study demonstrates that P152S, Q296H, and R92Q play a more critical role in the structural instability of GABA-AT compared with the other mutated models: G465R, L211F, L478P, and R220K.

Discovery of GABA Aminotransferase Inhibitors via Molecular Docking, Molecular Dynamic Simulation, and Biological Evaluation

“…The crystal structure of human GABA was not available in Protein Data Bank (PDB) (https://www.rcsb.org/) (accessed on 1 March 2023). The primary structure GABA-AT was deduced from our previous study, in which the cDNA of the Sus scrofa brain was utilized for homology modeling of the human GABA-AT 3D structure [27]. The Sus scrofa GABA-AT enzyme sequence had 95.67% homology with the Homo sapiens enzyme, which authenticated studies with the Sus scrofa enzyme as being highly pertinent to humans.…”

“…Human GABA-AT is a homodimeric protein consisting of 461 amino acid residues. The three-dimensional structure of human GABA protein was obtained from our previously published data on the mutational analysis of GABA-AT, in which homology modeling was carried out against the reference model of Sus scrofa with 1.63 Å resolution (PDBID 4Y0D) and the sequence similarity of Sus scrofa was depicted to be 95.67% against human GABA-AT [27]. SWISS-MODEL was employed to generate the 3D model of human GABA-AT.…”

“…The function of a binding pocket within a protein is not only determined by its structure and location but also influenced by the specific group of amino acid residues that surround it [28]. The binding pocket residues of GABA-AT were chosen from already published data [27], in which we calculated the binding pocket residues of human GABA-AT using an online webserver, PrankWeb (https://prankweb.cz/) (accessed on 25 March 2023), and selected as Ile100, Ser102, Ala162, Cys163, Gly164, Ser165, Phe217, His218, Gly219, Arg220, Glu293, Asp326, Val328, Gln_329, Gln330, Ser356, Lys357, and Met360. Moreover, the binding pocket of GABA-AT was visualized using Discovery Studio and UCSF Chimera to confirm the residual binding position (Figure 2).…”

“…The interacting site in the protein's holo-structure most likely determines the binding pocket of the protein, where the active ligand binds [33]. The binding pocket residues were selected from already published data [27] as Ile100, Ser102, Ala162, Cys163, Gly164, Ser165, Phe217, His218, Gly219, Arg220, Glu293, Asp326, Val328, Gln_329, Gln330, Ser356, Lys357, and Met360. Furthermore, the binding pocket was visualized using UCSF Chimera and Discovery Studio for residual position comparison and validation.…”