Not all pseudouridine synthases are potently inhibited by RNA containing 5-fluorouridine

Abstract

RNA containing 5-fluorouridine has been assumed to inhibit strongly or irreversibly the pseudouridine synthases that act on the RNA. RNA transcripts containing 5-fluorouridine in place of uridine have, therefore, been added to reconstituted systems in order to investigate the importance of particular pseudouridine residues in a given RNA by inactivating the pseudouridine synthase responsible for their generation. In sharp contradiction to the assumption of universal inhibition of pseudouridine synthases by RNA containing 5-fluorouridine, the Escherichia coli pseudouridine synthase TruB, which has physiologically critical eukaryotic homologs, is not inhibited by such RNA. Instead, the RNA containing 5-fluorouridine was handled as a substrate by TruB. The E. coli pseudouridine synthase RluA, on the other hand, forms a covalent complex and is inhibited stoichiometrically by RNA containing 5-fluorouridine. We offer a hypothesis for this disparate behavior and urge caution in interpreting results from reconstitution experiments in which RNA containing 5-fluorouridine is assumed to inhibit a pseudouridine synthase, as normal function may result from a failure to inactivate the targeted enzyme rather than from the absence of nonessential pseudouridine residues.

FIGURE 1.

Complex formation between TruB or RluA and RNA stem–loops containing f⁵U by gel shift in SDS-PAGE. (A) Incubation of TruB (37.3 kD; marked with arrow) with [f⁵U]TSL does not result in a shifted band over a broad time range; TruB was present at 12.5 μM in all incubations; (MW) molecular weight standards. (B) Incubation of RluA (26.2 kD; marked with arrow) with [f⁵U]ASL produces a complex that survives denaturation by formamide and SDS (shifted band), with the amount of complex reflecting the ratio of enzyme to RNA. If the samples are heated before electrophoresis, the complex decomposes, and some decomposition is apparent even without heating.

FIGURE 2.

Inhibition of TruB and RluA by RNA stem–loops containing f⁵U. TruB is only very modestly inhibited by preincubation with [f⁵U]TSL (open bars); the same degree of inhibition (18%) is seen with two concentrations of [f⁵U]TSL that differ by 10-fold, which is inconsistent with the product stem–loop serving simply as a competitive, noncompetitive, or uncompetitive inhibitor. RluA is inhibited by preincubation with [f⁵U]ASL in essentially stoichiometric fashion (solid bars), which is consistent with irreversible inhibition by the formation of a covalent adduct. The concentrations given are those in the preincubation mixtures, which were diluted 10-fold for the activity assays.

FIGURE 3.

Conversion of [f⁵U]TSL into products upon incubation with TruB. (A) A portion of the C₁₈ HPLC traces of intact stem–loop before (broken line) and after (solid line) incubation with TruB showing quantitative conversion of [f⁵U]TSL into products. (B) A portion of the C₁₈ HPLC traces of the nucleosides resulting from digestion of [f⁵U]TSL before (broken line) and after (solid line) incubation with TruB; the void peak is labeled with an asterisk. (C) MALDI-TOF mass spectrum of intact [f⁵U]TSL before (broken line) and after (solid line) incubation with TruB; the shift of 18 m/z is that expected for hydration. (D) The structures of the likely products from f⁵U; the stereochemistry at C5 is assigned on the basis of the appearance of the depicted cis isomer (within the RNA stem–loop) in the cocrystal structure of TruB (Hoang and Ferre-D’Amare 1991).

FIGURE 4.

Reaction routes that could explain the disparate behavior of Ψ synthases toward [f⁵U]RNA and the rearranged and hydrated product observed in the TruB•[f⁵U]TSL cocrystal structure (Hoang and Ferre-D’Amare 1991). In all three depicted schemes, the reaction can proceed by either proposed mechanism to generate the rearranged product. (A) The essential aspartic acid serves as a general base in the direct hydration of the rearranged product. (B) The essential aspartic acid residue performs a Michael addition at C6, followed by ester hydrolysis, which could be fast or slow; it is also possible that the ester forms as part of the mechanism involving Michel addition at C6, so that the depicted rearranged product never exists. (C) The rearranged product is released by the enzyme into solution and becomes hydrated, likely in a stereorandom manner.