Molecular dynamics simulations on structural differences of glucose-6-phosphate dehydrogenase deficiency variants among the Asian population

Abstract

Glucose-6-phosphate dehydrogenase (G6PD) is responsible for red blood cell protection against free radicals. There are over 186 G6PD deficient variants which adversely affected the enzyme activity. In its active state, G6PD exists in dynamic equilibrium as dimer and tetramer, influenced by its ligands. A human G6PD monomer has three ligands, the glucose-6-phosphate (G6P) substrate, a catalytic nicotinamide adenine dinucleotide phosphate (NADP) cofactor, and structural NADP. Ligands like G6P disrupt dimer formation, whereas NADP favour tetramer formation. G6PD enzyme activity is dependent on the structural integrity of the dimer interface. The mechanism of mutation-induced structural instability and the physiological significance of ligands on G6PD structure and function remains unclear till date. More than 400 G6PD variants exists, only 10 per cent of mutations were analysed in depth, none of which includes variants common to Asian population. In this study, ten common Asian variants (G410D, K275N, R387C, V291M, L128P, R459L, V431M, H32R, G163S, and G131V) were chosen for analysis using molecular dynamics simulation (MDS). Since G6PD dimerization is crucial for basic activity, a G6PD dimer with ligands was constructed using molecular docking and simulated using GROMACS for 100 ns. The simulated trajectories of the variants against the wild type (WT) were used to evaluate changes at the mutation site, and the dimer and tetramer interfaces. Alterations in protein-ligand affinities were evaluated by analysing the molecular binding profile coupled with free binding energy calculations. The wild type and variants with high enzyme activity such as G131V and G163S, showed high structural integrity at the dimer interface characterized by intermolecular hydrogen bonds between Asp 421-Asp 421 and Glu 419-Thr 423 at ßN, and salt bridges between Glu 206-Lys 407. The bonds spanned over both monomeric subunits, resulting in compact dimer indicated by low radius of gyration (Rg) values. The G6PD structures with low Rg exhibited increased distance between the ßI–ßJ loop, thus exposing the tetramer interface and tetramer salt bridge residues. The high solvent accessible surface area (SASA) characteristic indicates a high dimer-dimer affinity in tetrameric state. The ßE–ae loop responsible for positioning G6P and the catalytic NADP for G6PD catalysis was retained in variants with stable dimer structures. Ligand interplay between the G6P and the structural NADP was evident; G6P trajectory frames showing high affinity toward G6PD, led to a low or total loss affinity of NADP. High NADP binding pocket occupancy contributed to a low Rg of the structures. This was the first G6PD MDS study to relate in-silico findings with existing biochemical and kinetic data. In short, findings from this study would be beneficial for variant assessment, prognostic marker identification and drug development. This MDS study was successful in validating empirical observations from previous biochemical and structural studies such as the loss of an-ae interhelical interactions for R459L, impaired tetramerization for K275N and R459L, and protein-ligand affinities for the G410D, R387C, V291M, R459L, and G163S variants towards G6P and NADP.