A new metal-binding site for yeast phosphoglycerate kinase as determined by the use of a metal-ATP analog.

Suicide substrate beta, gamma-bidentate Rh(III)ATP (RhATP) was used to map the metal ion-binding site in yeast phosphoglycerate kinase (PGK). Cleavage of the RhATP-inactivated enzyme with pepsin and subsequent separation of peptides by reverse-phase high-performance liquid chromatography gave two Rh-nucleotide bound peptides. One of the peptides corresponded to the C-terminal residues of PGK, and the other to a part of helix V. Of the four glutamates present in the C-terminal peptide, Glu 398 may be a likely metal coordination site. Therefore, importance of the C-terminal residues in PGK catalysis may be attributed, in part to the coordination of metal ion of the metal-ATP substrate. Metal coordination may then align the C-terminal peptide to extend toward the N-terminal domain and form the "closed" active site. Results presented in this paper suggest that one or more side chains of the enzyme may be coordinated to the metal ion in the PGK.3-phospho-D-glycerate-RhATP complex, and that exchange-inert metal-ATP analogs could be used to determine metal coordination sites on kinases and other metal-ATP-utilizing enzymes.

Proton NMR studies of a large protein. pH, substrate titrations, and NOESY experiments with perdeuterated yeast phosphoglycerate kinase containing [1H]histidine residues.

Fully deuterated yeast phosphoglycerate kinase ([2H]PGK) was prepared biosynthetically with only histidine side chains of normal (1H) isotopic composition. The 1H NMR spectrum of this enzyme ([1H]His[2H]PGK) showed that the histidine side chains are clearly visible as sharp signals. Thus detailed structural studies by 1H NMR became feasible with isotope-hybrid phosphoglycerate kinase which is otherwise too large (M(r) approximately 46,000) for conventional 1H NMR studies. Proton signals of bound substrates were visible in the 1H NMR spectrum even with a substrate-to-enzyme ratio of less than 1/2 (mol/mol). The 2D NOESY spectrum of enzyme-MgdATP-glycerol 3-phosphate complex showed that, although protein concentration was very high (1.5 mM), no intraprotein cross peaks were observed other than those of intraresidue histidine NOE cross peaks. In addition, intrasubstrate NOEs and intermolecular NOEs between histidine and substrate protons were visible at a 1.5/1 substrate/enzyme (mol/mol) ratio. Paramagnetic effects of a substrate analog, Cr(III)ATP, on some of the histidine side chains indicated that the formation of the ternary enzyme-substrate complex causes large conformational changes in the enzyme.

Substrate activity of Rh(III)ATP with phosphoglycerate kinase and the role of the metal ion in catalysis.

An exchange-inert Rh(H2O)4ATP complex showed a one-turnover substrate activity with yeast phosphoglycerate kinase. Transfer of the phosphoryl group between ATP and 3-phosphoglycerate occurs with both substrates in the coordination sphere of the metal ion. Because of the slow ligand exchange rates of Rh3+, the reaction product 1,3-diphosphoglycerate (1,3-dPGA) remained coordinated to the metal ion. During the course of the reaction, the enzyme was inactivated, suggesting that the metal ion is coordinated to a protein side chain. Thus the product Rh(H2O)nADP.1,3-dPGA remained bound to the enzyme even after removal of excess substrate. These results suggested that the metal ion may not only act as an electron sink to activate the electrophile, but it may also help to optimally align both substrates for phosphoryl transfer by coordination to both substrates. It is therefore likely that entry of 3-phospho-D-glycerate into the coordination sphere of metal of a metal-ATP complex may start the proposed hinge-bending motion of yeast phosphoglycerate kinase to form a "closed" active site between the two substrate binding domains of the enzyme.