Dissection of the structural and functional role of a conserved hydration site in RNase T1.

The reoccurrence of water molecules in crystal structures of RNase T1 was investigated. Five waters were found to be invariant in RNase T1 as well as in six other related fungal RNases. The structural, dynamical, and functional characteristics of one of these conserved hydration sites (WAT1) were analyzed by protein engineering, X-ray crystallography, and (17)O and 2H nuclear magnetic relaxation dispersion (NMRD). The position of WAT1 and its surrounding hydrogen bond network are unaffected by deletions of two neighboring side chains. In the mutant Thr93Gln, the Gln93N epsilon2 nitrogen replaces WAT1 and participates in a similar hydrogen bond network involving Cys6, Asn9, Asp76, and Thr91. The ability of WAT1 to form four hydrogen bonds may explain why evolution has preserved a water molecule, rather than a side-chain atom, at the center of this intricate hydrogen bond network. Comparison of the (17)O NMRD profiles from wild-type and Thr93Gln RNase T1 yield a mean residence time of 7 ns at 27 degrees C and an orientational order parameter of 0.45. The effects of mutations around WAT1 on the kinetic parameters of RNase T1 are small but significant and probably relate to the dynamics of the active site.

Dissecting histidine interactions of ribonuclease T1 with asparagine and glutamine replacements: analysis of double mutant cycles at one position.

His92 of Ribonuclease T1 combines functional and structural features involving both imidazole nitrogens. To evaluate the use of Asn and Gln substitutions in dissecting the properties of histidines, we analysed the consequences of the His92Gln and His92Asn substitutions on the enzyme's structure, function, and conformational stability by protein engineering and X-ray crystallographic methods. In the X-ray structures of wild-type and His92Gln RNase T1 in complex with 2'-GMP the His92-N epsilon 2 and Gln92-N epsilon 2 atoms are isosterically equivalent. Similarly, the His92N delta 1H...OAsn99 hydrogen bond observed in wild-type is replaced by an equivalent Asn92N delta 2H...OAsn99 in the His92Asn mutant structure. Double mutant cycles at a single position were used to analyse the intermolecular and intramolecular interactions of the exchangeable proton and the individual histidine nitrogens. Urea denaturation measurements as a function of pH revealed that the exchangeable proton of His92, rather than its imidazole ring is contributing about 1 kcal/mol to the conformational stability of RNase T1. The stabilizing and the destabilizing effects of the (His-->Gln) and the (His-->Asn) mutations on urea denaturation of RNase T1 at pH 9.0 suggest that the unprotonated N delta 1 and N epsilon 2 atoms contribute in a compensating way to the conformational stability of RNase T1. A comparative study of the kinetics of all mutants suggests that the protonated His92 imidazole is a strictly co-operative catalytic device.

Additivity of protein-guanine interactions in ribonuclease T1.

It has been established that Tyr-42, Tyr-45, and Glu-46 take part in a structural motif that renders guanine specificity to ribonuclease T1. We report on the impact of Tyr-42, Tyr-45, and Glu-46 substitutions on the guanine specificity of RNase T1. The Y42A and E46A mutations profoundly affect substrate binding. No such effect is observed for Y45A RNase T1. From the kinetics of the Y42A/Y45A and Y42A/E46A double mutants, we conclude that these pairs of residues contribute to guanine specificity in a mutually independent way. From our results, it appears that the energetic contribution of aromatic face-to-face stacking interactions may be significant if polycyclic molecules, such as guanine, are involved.

A catalytic function for the structurally conserved residue Phe 100 of ribonuclease T1.

The function of the conserved Phe 100 residue of RNase T1 (EC 3.1.27.3) has been investigated by site-directed mutagenesis and X-ray crystallography. Replacement of Phe 100 by alanine results in a mutant enzyme with kcat reduced 75-fold and a small increase in Km for the dinucleoside phosphate substrate GpC. The Phe 100 Ala substitution has similar effects on the turnover rates of GpC and its minimal analogue GpOMe, in which the leaving cytidine is replaced by methanol. The contribution to catalysis is independent of the nature of the leaving group, indicating that Phe 100 belongs to the primary site. The contribution of Phe 100 to catalysis may result from a direct van der Waals contact between its aromatic ring and the phosphate moiety of the substrate. Phe 100 may also contribute to the positioning of the pentacovalent phosphorus of the transition state, relative to other catalytic residues. If compared to the corresponding wild-type data, the structural implications of the mutation in the present crystal structure of Phe 100 Ala RNase T1 complexed with the specific inhibitor 2'-GMP are restricted to the active site. Repositioning of 2'-GMP, caused by the Phe 100 Ala mutation, generates new or improved contacts of the phosphate moiety with Arg 77 and His 92. In contrast, interactions with the Glu 58 carboxylate appear to be weakened. The effects of the His 92 Gln and Phe 100 Ala mutations on GpC turnover are additive in the corresponding double mutant, indicating that the contribution of Phe 100 to catalysis is independent of the catalytic acid His 92. The present results lead to the conclusion that apolar residues may contribute considerably to catalyze conversions of charged molecules to charged products, involving even more polar transition states.