SOLID-STATE NUCLEAR-MAGNETIC-RESONANCE INVESTIGATION OF SOLVENT DEPENDENCE OF TYROSYL RING MOTION IN AN ENZYME

Tyrosyl ring motions in alpha-lytic protease were investigated by soli d-state deuterium nuclear magnetic resonance(NMR) spectroscopy in lyop hilized enzyme powder, in powder suspended in organic solvents, and in aqueous crystals. Ring flipping rates were determined by examining de uterium quadrupole echo line shapes. Of the four Tyr residues in the e nzyme, one was flipping at the slow (less-than-or-equal-to 10(3) s-1) and one at the fast (greater-than-or-equal-to 10(7) s-1) exchange limi t of the line shape experiment in all the environments tested. Flippin g rates of the remaining two Tyr residues depended markedly on the sol vent, with the lowest flipping rates (less-than-or-equal-to 10(3) s-1 for both residues) observed in the enzyme powder, whether dry or suspe nded in hydrophobic tert-butyl methyl ether. In hydrophilic dioxane an d acetonitrile, the mobility of these residues increased to 10(4) and 10(5) s-1. The latter rate rose further to 10(6) s-1 in the hydrated h ydrophilic solvents and to greater-than-or-equal-to 10(7) s-1 in aqueo us crystals. The deuterium spectrum of native alpha-lytic protease was compared with that of the enzyme whose active center was covalently m odified with an inhibitor, which binds next to Tyr-123, constraining i ts ring. This experiment revealed that water addition to acetonitrile specifically increased the flipping rate of this active center residue . Librational motions (''wobbling''), estimated by their effect on spi n-lattice relaxation times, were slowest in the anhydrous solvents, in termediate in the hydrated solvents, and fastest in the aqueous crysta ls. Thus, alpha-lytic protease is more rigid in organic solvents than in water, as judged by mobility of its tyrosyl residues. Water strippi ng by hydrophilic solvents did not increase enzyme rigidity, nor were there clear correlations between mobility and either enzymatic activit y or solvent dielectric constant.