Effects of 3'-terminal phosphates in RNA produced by ribozyme cleavage

During the course of a structural investigation of a selenocysteine insertion element from rat Type 1 iodothyronine 59-deiodinase (D1 SECIS), we wished to prepare milligram quantities of an RNA that comprised the upper stem/loop of this RNA structure (41 nt; Fig+ 1A)+ This stem/loop is a stimulator required for the insertion of selenocysteins at a specific internal UGA stop codon (Berry et al+, 1991)+ Construction of the upper stem/ loop RNA required that the base of the stem end with four noncanonical base pairs (Walczak et al+, 1996), with no additional base pairing below this terminal quartet, as determined from in vitro assays (Martin et al+, 1998)+ This RNA has been produced synthetically for crystallization trials, so we sought a thorough comparison between this molecule and the RNA produced by run-off transcription+ Because the RNA to be studied begins with an adenosine and ends in a uridine, a ribozyme was used at each end to produce homogeneous RNA+ The system used to accomplish this has been described previously (Ferre-D™Amare & Doudna, 1996) and the resulting T7-produced RNA would differ from the synthetic RNA only by the 39- cyclic phosphate+ In making preliminary imino proton NMR assignments for both of these RNAs, we concluded that the base pairing for both RNAs is nearly identical, as evident from the one-dimensional NMR spectra collected in 90% H2O (Fig+ 2)+ Close inspection shows that there are subtle differences near the RNA helix end