Here we present a time-resolved crystallographic analysis of the hydrolysis of exo (Sp) guanosine 2',3'-cyclophosphorothioate by RNase T1. The use of a slow substrate and fast crystallization methods made it possible to perform the study with conventional data-collection techniques. The results support the idea that the hydrolysis reaction proceeds through a mechanism that is the inverse of the transesterification reaction. In addition, the structures provide an explanation for the differential behavior of RNase T1 towards exo- and endo-cyclic thiophosphates.

We previously presented evidence that ribonuclease T1 (RNase T1; EC 3.1.27.3) contains a subsite that, by interacting with the leaving nucleoside N of GpN dinucleoside phosphate substrates, contributes to catalysis. The kcat values for transphosphorylation follow the order GpC greater than GpA greater than GpU whereas the equilibrium dissociation constants for these substrates are very similar [Steyaert, J., Wyns, L., & Stanssens, P. (1991) Biochemistry (preceding paper in this issue)]. Consistent with this notion, we find that the rate of transesterification of the synthetic substrate GpMe, in which the leaving nucleoside is replaced by a methanol group, is at least 3 orders of magnitude lower than that of GpN substrates. The enzyme's affinity for GpMe is very similar to that for the various GpN substrates, indicating that the apparent contribution of the leaving nucleoside to ground-state binding is minimal. To identify the side chains that belong to the RNase T1 subsite, we searched for amino acid substitutions that differentially affect the transesterification kinetics of GpNs versus GpMe. The Asn36Ala, Tyr38Phe, His92Gln, and Asn98Ala mutants have been analyzed. Of these, the Asn36Ala and Asn98Ala substitutions reduce the transphosphorylation rate of the different GpNs considerably whereas they have virtually no effect on the rate of GpMe transphosphorylation. This observation shows that the Asn36 and Asn98 amide functions are part of the RNase T1 subsite. The sum of the contributions of the two residues accounts quite precisely for the differences in turnover rates among GpC, GpA, and GpU.

No abstract

Reaction of ethyl 4-thioxo-3,4-dihydropyrimidine-5-carboxylate derivatives 1a,b and ethyl 4-oxo-3,4-dihydropyrimidine-5-carboxylate 1c with 2,3,4,6-tetra-O-acetyl-alpha-D-glucopyranosyl bromide in KOH or TEA afforded ethyl 2-aryl-4-(2',3',4',6'-tetra-O-acetyl-beta-D-glucopyranosylthio or/ oxy)-6-methylpyrimidine-5-carboxylate 6a-c. The glucosides 6a and 6b were obtained by the reaction of 1a and 1b with peracetylated glucose3 under MW irradiation. Mercuration of 1a followed by reaction with acetobromoglucose gave the same product 6a. The reaction of 1a-c with peracetylated ribose 4 under MW irradiation gave ethyl 2-aryl-4-(2',3',5'-tri-O-acetyl-beta-D-ribofuranosylthio)-6-methylpyrimidine-5-carboxylate 8a-c. The deprotection of 6a-c and 8a-c in the presence of methanol and TEA/H(2)O afforded the deprotected products 7a-c and 9a-c. The structure were confirmed by using (1)H and (13)CNMR spectra. Selected members of these compounds were screened for antimicrobial activity.