Slowed enzymatic turnover allows characterization of intermediates by solid-state NMR.

EPSP (5-enolpyruvylshikimate-3-phosphate) synthase catalyzes condensation of shikimate 3-phosphate (S3P) and phosphoenolpyruvate (PEP) to form EPSP, a precursor to the aromatic amino acids. S3P and [2-13C]POP were bound to mutant or wild type E. coli forms of the enzyme prior to lyophilization. CPMAS-echo and rotational-echo double-resonance (REDOR) NMR experiments, employing a slow catalytic EPSP synthase mutant and a long prelyophilization incubation interval, allowed our observation of the gradual formation of a strong 31P-13C coupling consistent with the well characterized tetrahedral intermediate. However, after shorter low temperature incubation intervals of substrates with mutant or wild-type enzymes, carbon CPMAS-echo NMR spectra showed the 13C label at 155 ppm, consistent with sp2 geometry of this carbon. REDOR revealed that the phosphorus of PEP was cleaved. However, phosphorus at a distance of 7.5 A was observed, due to the phosphate of a nearby bound S3P. Heating the sample allowed the reaction to progress, as shown by the diminution of the 155 ppm peak and growth of a peak at 108 ppm. The sp3 geometry implied by the 108 ppm peak strongly suggested formation of a S3P-PEP condensation product. REDOR indicated that phosphorus was still distant, but now only 6.1 (wild type) or 5.9 A (mutant) distant. We think that the early intermediates with peaks at 155 and 108 ppm are covalently bound to the enzyme. We also think that the tetrahedral intermediate that we observed was formed after product was generated.

Analysis of site-directed mutants locates a non-redox-active metal near the active site of cytochrome c oxidase of Rhodobacter sphaeroides.

Substoichiometric amounts of Mn are bound by the aa3-type cytochrome c oxidase of Rhodobacter sphaeroides and appear in the EPR spectrum of the purified enzyme as signals that overlay those of CuA in the g = 2.0 region. The Mn is tightly bound and not removed by a high degree of purification or by washing with 50 mM EDTA. The amount of bound Mn varies with the ratio of Mg to Mn in the growth medium. Oxidase containing no EPR-detectable Mn can be prepared from cells grown in low Mn/Mg, while high Mn/Mg in the growth medium gives rise to near stoichiometric levels (0.7 mol/mol of aa3). Incubation of purified Mn-deficient oxidase with 1 mM Mn does not allow incorporation into the tight binding site, indicating that this site is not accessible in the assembled protein. When bound Mn is depleted by growth in high Mg, there is no change in electron transfer activity, suggesting that Mg may substituted for Mn and maintain protein structure. Analysis of site-directed mutants in an extramembrane loop close to the active site of cytochrome oxidase identifies His-411 and Asp-412 of subunit I as probable ligands of the Mn. Mutation of either residue leads to lower activity and loss of Mn binding, even in cells grown in elevated concentrations of Mn. Since Mn binding correlates with the [Mn] to [Mg] ratio in the culture medium, we propose that Mn competes for the site that normally binds a stoichiometric Mg ion in aa3-type cytochrome c oxidases.

A continuous wave and pulsed EPR characterization of the Mn2+ binding site in Rhodobacter sphaeroides cytochrome c oxidase.

The ligation environment of the tightly bound Mn2+ in cytochrome c oxidase from Rhodobacter sphaeroides has been characterized by electron paramagnetic resonance (EPR) and electron spin echo envelope modulation (ESEEM). The EPR data show that the Mn2+ is six-coordinate and located in a highly symmetric binding site. Analyses of X- and Q-band EPR spectra show that the zero field splitting parameter D is 115 +/- 25 G (0.0107 +/- 0.0023 cm-1) in the fully oxidized enzyme and 125 +/- 15 G (0.0117 +/- 0.0014 cm-1) in the fully reduced enzyme. For both redox forms of the enzyme the value of E is < or = 25 G (0.0023 cm-1). By comparison with crystal structures of Mn2+ binding proteins, the structural changes at the Mn2+ binding site upon redox state change of the enzyme are estimated to be < or = 0.2 A in ligand bond lengths and < or = 10 degrees in bond angle. This analysis indicates that little modification occurs at the Mn2+ site upon redox change at the other metal centers. Considering the proximity of the Mn2+ site to heme a and heme a3-CuB [Hosler, J. P., Espe, M. P., Zhen, Y., Babcock, G. T., & Ferguson-Miller, S. (1995) Biochemistry 34, 7586-7592], we interpret these results to imply also that there is no large protein conformational change near the heme a and heme a3-CuB sites upon a change in their redox states. Multifrequency 3-pulse ESEEM results provide direct evidence for a nitrogen ligand to the Mn2+, which is assigned to a histidine by comparison with ESEEM studies of Mn(2+)-bound lectins [McCracken, J., Peisach, J., Bhattacharyya, L., & Brewer, F. (1991) Biochemistry 30, 4486-4491] and specifically to His-411 in subunit 1 on the basis of mutagenesis studies (Hosler et al., 1995). From these results a partial model of the Mn2+ binding site has been constructed.