Metal Interactions with a GAAA RNA Tetraloop Characterized by ³¹P NMR and Phosphorothioate Substitutions

Abstract

A metal site in a 5'-GAAA-3' tetraloop, a stabilizing and phylogenetically conserved RNA motif, is explored using (31)P NMR spectroscopy and phosphorothioate modifications. Similar to previous reports [Legault, P., and Pardi, A. (1994) J. Magn. Reson., Ser. B 103, 82-86], the (31)P NMR spectrum of a 12-nucleotide stem-loop sequence 5'-GGCCGAAAGGCC-3' exhibits resolved features from each of the phosphodiester linkages. Titration with Mg(2+) results in distinct shifts of a subset of these (31)P features, which are assigned to phosphodiesters 5' to A6, A7, and G5. Titration with Co(NH(3))(6)(3+) causes only a slight upfield shift in the A6 feature, suggesting that changes caused by Mg(2+) are due to inner-sphere metal-phosphate coordination. R(p)-Phosphorothioate substitutions introduced enzymatically 5' to each of the three A residues of the tetraloop provide well-resolved (31)P NMR features that are observed to shift in the presence of Cd(2+) but not Mg(2+), again consistent with a metal-phosphate site. Analysis of (31)P NMR spectra using the sequence 5'-GGGCGAAAGUCC-3' with single phosphorothioate substitutions in the loop region, separated into R(p) and S(p) diastereomers, provides evidence for an inner-sphere interaction with the phosphate 5' to A7 but outer-sphere or structural effects that cause perturbations 5' to A6. Introduction of an R(p)-phosphorothioate 5' to A7 results in a distinct (31)P NMR spectrum, consistent with thermodynamic studies reported in the accompanying paper that indicate a unique structure caused by this substitution. On the basis of these results and existing structural information, a metal site in the 5'-GAAA-3' tetraloop is modeled using restrained molecular dynamics simulations.

Solution probing of metal ion binding by helix 27 fromEscherichia coli16S rRNA

“…This observation is consistent with previous studies that have shown that GNRA tetraloops bind Mg 2+ tightly (Hermann and Westhof 1998;Maderia et al 2000;Rudisser and Tinoco 2000;Mundoma and Greenbaum 2002). Other resonances that display significant downfield shifts were those of U904 NH3 (0.16 ppm) and G887 NH1 (0.12 ppm), with relative Mg 2+ half-titration points of Mg 1/2 = 2.0 6 0.6 mM and Mg 1/2 = 6.1 6 1.1 mM, respectively ( Fig.…”

“…Bacterial 30S ribosomal subunit crystal structures depicting magnesium ion binding sites consistently place two such Mg 2+ binding sites at opposite ends of H27 Carter et al 2000;Ogle et al 2001Ogle et al , 2002: one (Mg73) adjacent to the G886-U911 pair and another one (Mg126) within the GNRA tetraloop (Fig. 1B), a common metal ion binding site (Hermann and Westhof 1998;Maderia et al 2000;Rudisser and Tinoco 2000;Mundoma and Greenbaum 2002). The metal-RNA distances are all >2.8 Å , consistent with the notion that these metal ions are outersphere (diffusely) bound.…”

Specific metal-ion binding sites in a model of the P4-P6 triple-helical domain of a group I intron

“…Large ribozymes require divalent metal ions for tertiary folding and structure+ Positively charged ions screen the negative charge of the RNA backbone and reduce repulsions between electronegative groups within the RNA structure+ Most of the divalent metal ions bind the RNA in a nonspecific manner+ However, a subset interacts with functionally important metal binding sites, where the metal ion can play a structural and/or catalytical role, interacting with electron pair donors of the bases and sugar-phosphate backbone (Feig & Uhlenbeck, 1999)+ The metal ion may bind to pockets already existing in the secondary structure, or to sites that appear during tertiary folding (Tinoco & Bustamante, 1999)+ Specific interactions between magnesium and nonbridging phosphate oxygens can be revealed by phosphorothioate interference experiments+ In these, one utilizes the fact that a hard Lewis acid (magnesium) shows a much stronger affinity for a hard Lewis base (oxygen) as compared to a soft Lewis base (sulfur)+ If a phosphate oxygen is part of a specific metal binding site, where a magnesium ion crucial for ribozyme structure and/or catalysis is situated, a sulfur substitution at that position results in a reduced or abolished activity of the ribozyme+ Manganese, being a less hard Lewis acid than magnesium, has a greater propensity to interact with sulfur and can thereby often replace magnesium and restore activity+ This kind of manganese rescue implies that the site of interest is a specific magnesium-binding site of importance for ribozyme structure and/or activity+ Generally, the exchange of ligand atom (oxygen r sulfur) and metal-ion identity (magnesium r manganese) of a phosphorothioate interference experiment does not have any major influence on the original structure of the RNA molecule (Feig & Uhlenbeck, 1999;Feig, 2000;Maderia et al+, 2000b)+ However, recent studies have in some cases described substantial effects on RNA structure as a consequence of phosphorothioate substitutions (Horton et al+, 2000;Maderia et al+, 2000a;Smith & Nikonowicz, 2000)+ The effects seem to be localized to regions of irregular secondary structure, where a phosphorothioate substitution might modify the local structure considerably, thus changing the availability of functional groups+ Other studies have shown that a metal-ion switch can lead to RNA-metalion interactions not present in the original structure (Basu & Strobel, 1999;Shan & Herschlag, 2000), this due to the different tendencies of the metal ions in binding various ligands (Cowan, 1998;Bock et al+, 1999)+ The potential structural disruption and/or alternate RNAmetal-ion interactions must be taken into account when interpreting results obtained from phosphorothioate interference experiments+…”

Structural Basis for a Lethal Mutation in U6 RNA,

“…We find that the U80G ISL displays stereospecific chemical shift changes at G80 upon cadmium addition that are qualitatively similar to those of the wild-type ISL ( Figure 6A). Cadmium binding to phosphorothioates typically produces a diagnostic upfield change in the 31 P chemical shift, as observed for the S p resonance 5′ to nucleotide 80 for both the wild-type and mutant ISL RNAs ( Figure 6A) (12,33,34). As previously observed for the wild-type U6 ISL, these data cannot be fit to a single-binding site model, because cadmium ions also bind nonspecifically to the electronegative major groove of the ISL, and sample aggregation and line broadening occur with higher molar equivalents of cadmium (12).…”