Abstract

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a well-known glycolytic enzyme, exhibits moonlighting functions, including catalysis of glyceryl trinitrate (GTN) to release vasoactive nitrite (NO2–) or nitric oxide (NO). GTN catalysis generates a thionitrate (E-Cys-NO2) intermediate at the catalytic thiol of GAPDH. Thus, we investigated decomposition of CH3SNO2, a model compound for E-Cys-NO2, using state-of-the-art quantum mechanics techniques. We showed that the well-studied homolysis pathway, which releases NO, is energetically unfavorable. Instead, hydrolysis and thiolysis pathways that release NO2– are energetically favorable. The release of NO2– upon the attack of anionic nucleophiles (e.g. OH– or CH3S–) along the S–N bond is barrierless which could explain the instability of thionitrates in aqueous solutions. We also looked at the effect of protonation on the stability of the S–N bond of CH3SNO2 to mimic the effect of proton donation from the protein. We found that O- and S-protonation stabilize and destabilize the S–N bond, respectively. We evaluated the electronic structures of the protonated isoforms of CH3SNO2 to elucidate the observed stability/instability of the S–N bond. Further investigation of nitrite release by hydrolysis of E-Cys-NO2 in the active site of GAPDH using quantum mechanics / molecular mechanics (QM/MM) techniques revealed similar activation barriers as shown by the QM methods.
Each subunit of the GAPDH homotetramer binds the cofactor nicotinamide adenine dinucleotide (NAD+). The tetramer binds NAD+ with negative cooperativity. We carried out normal mode analysis and molecular dynamics (MD) simulations of GAPDH-NAD+ to probe the mechanism of negative cooperativity and define the subunit interactions that contribute to this phenomenon. We compared the dynamics of GAPDH-NAD+, GAPDH-NADH, and apo-GAPDH and observed concerted motions between subunits of GAPDH-NAD+ and GAPDH-NADH. These motions are lost in apo-GAPDH, indicating that NAD+ binding induces these dominant and functionally relevant motions of GAPDH. We also indicated the changes in the NAD binding-site residues and active-site residues that are observed in the crystal structure of apo- vs. holo-GAPDH.
Many moonlighting functions of GAPDH depend on its oligomeric state. MD simulations are thus carried out for the three oligomeric states, tetramer, dimer or monomer. Our study shows that the dimer and monomer are less stable than the tetramer and functionally important motions are centered on the NBDs in tetramer and dimer, but they are centered on the S-loop in the monomer. The importance of the dynamics of the S-loop are highlighted. This disordered loop in the dimer and particularly monomer could promote binding of the GAPDH dimer and monomer with multiple protein partners to affect different functions.
We modeled GAPDH and the seven in absentia homolog 1 (Siah1) protein-protein interactions. By investigating complex formation between the different oligomeric states of each protein, we show that the GAPDH monomer binds more tightly to Siah1 than the GAPDH tetramer. Thus, GAPDH monomer could stabilize Siah1 and the complex could be translocated to the nucleus to initiate apoptosis.
Overall, our study provides molecular-level insights into novel functions of GAPDH. We have highlighted the importance of the dynamics of its various structural domains, which contribute to the different functions of GAPDH.