ИСТИНА

Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is a homotetrameric glycolytic enzyme catalysing oxidative phosphorylation of glyceraldehyde 3-phosphate to glycerate 1,3-bisphosphate coupled with reduction of NAD+ to NADH. Humans possess two homologous isoenzymes of GAPDH with 68% identical sequences: somatic (GAPDH-1) and testis-specific (GAPDH-2). GAPDH-1 is a well-studied enzyme which is present in all cells. It was discovered to exhibit negative cooperativity: the affinity for NAD+ lowers with the amount of NAD+ already bound [1]. GAPDH-2 is found only in sperm tails. Recent studies by our group established that GAPDH-2 is also present in certain malignant tumors, viz. melanomas (data to be published). In this work, we apply both experimental and bioinformatical approaches to investigate NAD+ binding characteristics of human GAPDH isoenzymes. This does not only provide new data on little-studied GAPDH-2, but also give some insight into the interactions resulting in negative cooperativity. The NAD+ binding constants of human GAPDH isoenzymes were estimated by means of fluorescence quenching titrations. GAPDH-2 was found to exhibit higher affinity for NAD+ and weaker negative cooperativity than GAPDH-1. This peculiarity seems for us to be an adaptation to the spermatozoa metabolism. In contrast to the majority of human cells, spermatozoa are known to generate energy mainly in the course of glycolysis [2]. It leads to the necessity in glycolytic enzymes capable of working in low NAD+/NADH conditions such as GAPDH-2. It is significant that malignant tumors expressing GAPDH-2 are also known to utilize glycolysis as the main energy source. The limitations of experimental methods forced us to switch to bioinformatics for more detailed study of interactions between GAPDH subunits. Ligand-protein affinities were estimated from molecular dynamics simulations by means of linear interaction energy (LIE) method [3]. The NAD+ binding free energies exhibited by each of 4 binding sites in GAPDH tetramers were calculated to be -3 kJ/mol, -7 kJ/mol, -33 kJ/mol and -60 kJ/mol for GAPDH-1, as well as -18 kJ/mol, -37 kJ/mol, -47 kJ/mol and -54 kJ/mol for GAPDH-2. These values meet the experimental ones confirming the model quality. Further analysis of human GAPDH-1 crystal structure (PDB ID 1znq) revealed that each of 4 subunits (usually designated as O, P, Q and R) makes contacts with 2 other subunits, e.g. subunit O makes contacts with subunits P and R. Therefore, negative cooperativity should be mediated by either OP-type or OR-type interactions. To distinguish between these two possibilities, the affinities for NAD+ were estimated for both OP and OR dimers. OR dimer was discovered to retain negative cooperativity (-24 kJ/mol and -56 kJ/mol), while OP dimer was not (-43 kJ/mol and -54 kJ/mol). Free energy of interactions between O and R subunits (-348 kJ/mol) was calculated to be 3 times greater than between O and P subunits (-105 kJ/mol). These findings mean that GAPDH tetramer may be considered as a dimer of dimers (OR + PQ) and negative cooperativity of NAD+ binding is mediated by interplay within OR and PQ dimers. The work was supported by Russian Foundation for Basic Researches (grants 09-04-01122-a and 09-04-92740-NNIOM_a). Molecular dynamics simulations were performed at SKIF-MGU “Chebyshev” supercomputer. 1. J.E.Bell, K.Dalziel (1974) Studies of coenzyme binding to rabbit muscle glyceraldehyde-3-phosphate dehydrogenase, Biochim Biophys Acta., 391:249–258. 2. C.Mukai, M.Okuno (2004) Glycolysis plays a major role for adenosine triphosphate supplementation in mouse sperm flagellar movement, Biol Reprod., 71:540–547. 3. J.Aqvist et al. (2002) Ligand binding affinities from MD simulations, Acc Chem Res., 35:358-365.