News and Publications

Catalytic Mechanisms of RNA Enzymes

M.J. Fedor, C.P. Donahue, H.A. Erlacher, S. Nesbitt

The discovery that some biochemical reactions are catalyzed by RNAs, and not proteins, gave rise to a new field of RNA enzymology that seeks to explain the mechanisms used by RNAs to accomplish catalysis. The defining feature of a catalyst is the ability to accelerate a chemical transformation by lowering the energy barrier between the ground state and the transition state. Ribozymes adopt highly ordered structures that bind substrates, metal cations, and small molecules with high affinity and specificity. Consequently ribozymes, like protein enzymes, can position substrates in the active site to create favorable interactions with the transition state and unfavorable interactions with the ground state.

The dilemma for ribozymes is the absence of functional groups that are as adept as amino acid side chains at catalytic chemical reactions. The mechanism of phosphoester cleavage by small ribozymes involves in-line nucleophilic attack of the 2´ oxyanion on phosphorus to form the transition state with 5 oxygens transiently bound to phosphorus (Fig. 1). Breaking the 5´ oxygen-phosphorus bond generates products with 5´ hydroxyl and 2´,3´-cyclic phosphate termini. This reaction is the same one that is catalyzed by RNase A. However, nucleosides lack functional groups, such as the histidine imidazoles of RNase A, that could remove a proton from an attacking nucleophile. Furthermore, no ribonucleoside functional groups are positively charged, such as the /epsilonbdy/ amino of lysine, to stabilize negative charge in the transition state.

Until recently, this dilemma was thought to be solved in all catalytic RNAs by recruitment of (1) metal cations for charge stabilization and (2) hydrated metal cations for general base catalysis. An approach used to establish a catalytic role for metal cations exploits the difference between hard metal cations, such as magnesium ions, and soft metal cations, such as cadmium and manganese ions, in the ability to bind to oxygen or sulfur ligands. When the proRp nonbridging oxygen of the reactive phosphate is replaced with sulfur, hammerhead ribozyme cleavage rates decrease in buffers that contain magnesium ions, which bind to sulfur ligands poorly, but return to normal in buffers that contain manganese ions, which bind to sulfur ligands well. This change in metal-cation specificity implicates direct coordination of a metal cation to the proRp oxygen for charge stabilization of the transition state.

Surprisingly, we have found no evidence of direct coordination of metal cations in the hairpin ribozyme catalytic mechanism. The hairpin cleavage rate constant is only 4-fold lower for substrates with Rp phosphorothioate substitutions than for substrates with normal phosphates in buffers with magnesium ions, and the magnitude of the "thio effect" remains the same in buffers with manganese ions. Because pro-Rp and pro-Sp oxygens are close in space, a small difference in the geometry of the active site might change the stereospecificity of a large thio effect. However, cleavage rate constants for substrates with Sp phosphorothioates are even higher than for substrates with normal phosphates in buffers containing magnesium ions. Thus, for the hairpin ribozyme, neither diastereomer has a thio effect comparable to the greater than 100-fold inhibition of hammerhead cleavage of Rp phosphorothioate substrates. This absence of a metal ion--dependent thio effect points to a unique catalytic strategy for the hairpin ribozyme.

We found similar thio effects in buffers containing Co(NH₃)₆Cl₃ rather than MgCl₂ or MnCl₂. Unlike chloride ligands of magnesium or manganese ions, amine ligands of cobalt ions do not exchange with phosphate or water. Thus, the ability of Co(NH₃)₆Cl₃ to support activity excludes direct metal-cation coordination to nonbridging phosphate oxygens, or to any ligands, in the hairpin mechanism.

We are now focusing on understanding this novel catalytic strategy. Hairpin catalysis might be limited by a slow conformational change. A metal cation--independent chemical step, such as breaking the phosphorus-5´ oxygen bond, might be rate determining. Even more exciting is the possibility that functional groups within the hairpin ribozyme participate directly in catalytic chemical reactions.

PUBLICATIONS

Donahue, C.P., Fedor, M.J. Kinetics of hairpin ribozyme cleavage in yeast. RNA 3:961, 1997.

Fedor, M.J. Capturing a speeding locomotive. Cell 88:589, 1997.

Fedor, M.J. Ribozymes. Curr. Biol., in press.

Nesbitt, S., Hegg, L.A., Fedor, M.J. An unusual pH-independent and metal-ion-independent mechanism for hairpin ribozyme catalysis. Chem. Biol. 4:619, 1997.