Displacement of Mn 2+ from RNA by K + , Mg 2+ , Neomycin B, and an Arginine-Rich Peptide: Indirect Detection of Nucleic Acid/Ligand Interactions Using Phosphorus Relaxation Enhancement Jack S. Summers,* ,† John Shimko, † Fredric L. Freedman, † Christopher T. Badger, †,‡ and Michael Sturgess † Contribution from Message Pharmaceuticals, Inc., 30 Spring Mill Rd, MalVern, PennsylVania 19355, and CooperatiVe Education Associate, UniVersity of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland 21250 Received July 23, 2002 Abstract: We have developed a novel method to study the interactions of nucleic acids with cationic species. The method, called phosphorus relaxation enhancement (PhoRE), uses 1 H-detected 31 P NMR of exogenous probe ions to monitor changes in the equilibrium between free Mn 2+ and Mn 2+ bound to the RNA. To demonstrate the technique, we describe the interactions of four RNA molecules with metal ions (K + and Mg 2+ ), a small molecule drug (neomycin b), and a cationic peptide (RSG1.2). In each case, cationic ligand binding caused Mn 2+ to be displaced from the RNA. Free Mn 2+ was determined from its effect on the T2 NMR relaxation rate of either phosphite (HPO3 2- ) or methyl phosphite (MeOPH, CH3OP(H)O2 - ). Using this method, the effects of [RNA] as low as 1 μM could be measured in 20 min of accumulation using a low field (200 MHz) instrument without pulsed field gradients. Cation association behavior was sequence and [RNA] dependent. At low [K + ], Mn 2+ association with each of the RNAs decreased with increasing [K + ] until ∼40 mM, where saturation was reached. While saturating K + displaced all the bound Mn 2+ from a 31-nucleotide poly-uridine (U31), Mn 2+ remained bound to each of three hairpin-forming sequences (A-site, RRE1, and RRE2), even at 150 mM K + . Bound Mn 2+ was displaced from each of the hairpins by Mg 2+ , allowing determination of Mg 2+ dissociation constants (Kd,Mg) ranging from 50 to 500 μM, depending on the RNA sequence and [K + ]. Both neomycin b and RSG1.2 displaced Mn 2+ upon binding the hairpins. At [RNA] ∼ 3 μM, RRE1 bound a single equivalent of RSG1.2, whereas neither RRE2 nor A-site bound the peptide. These behaviors were confirmed by fluorescence polarization using TAMRA-labeled peptide. At 2.7 μM RNA, the A-site hairpin bound a single neomycin b molecule. The selectivity of RSG1.2 binding was greatly diminished at higher [RNA]. Similarly, each hairpin bound multiple equivalents of neomycin at the higher [RNA]. These results demonstrate the utility of the PhoRE method for characterizing metal binding behaviors of nucleic acids and for studying RNA/ligand interactions. Introduction RNA molecules have only recently become the targets of small molecule drug discovery efforts, and optimal strategies are not yet established. 1 We believe an effective strategy for developing such molecules will include a systematic effort to identify and characterize likely drug binding sites on structured regions of RNA and that these sites will commonly be sites that also bind divalent metal ions. Several lines of evidence indicate that the cationic aminoglycoside antibiotics (such as neomycin b) target RNAs at divalent metal binding sites: (1) Footprinting studies indicate that aminoglycosides protect RNAs from metal-induced cleavage reactions. (2) The Mg 2+ depen- dence of the neomycin inhibition of some ribozymes indicates a competition for the drug binding site. 2-4 (3) A recent crystallographic study showed that neomycin b bound to a transfer RNA at sites occupied by divalent metal ions in the absence of the drug. 4 (4) Molecular modeling studies suggest that aminoglycosides bind RNAs at sites of high negative electronic charge and that, in general, these drugs bind RNA with displacement of divalent metal ions. 5 In addition to ribozymes, attractive RNA-containing drug targets include RNA/RNA binding protein (RBP) complexes. * Corresponding author. Current address: 600 Lucia Avenue, Baltimore, MD 21229. Tel: (410) 644-6263. E-mail: jack@hhmi.umbc.edu. † Message Pharmaceuticals. ‡ University of Maryland Baltimore County. (1) (a) Cheng, A. C.; Calabro, V.; Frankel, A. D. Curr Opin. Struct. Biol. 2001, 11, 478-484. (b) Hermann, T. Angew. Chem. Int. Ed. 2000, 39, 1890- 1905. (c) Afshar, M.; Prescott, C. D.; Varani, G. Curr. Opin. Biotechnol. 1999, 10, 59-63. (d) Hermann, T.; Westhof, E. Combin. Chem., High Throughput Screen. 2000, 3, 219-234. (2) Hoch, I.; Berens, C.; Westhof, E.; Schroeder, R. J. Mol. Biol. 1998, 282, 557-569. (3) Rogers, J.; Chang, A. H.; von Ahsen, U.; Schroeder, R.; Davies, J. J. Mol. Biol. 1996, 259, 916-925. (4) Mikkelsen, N. E.; Johansson, K.; Virtyanen, A.; Kirsebom, L. A. Nature Struct. Biol. 2001, 8, 510-514. (5) Hermann, T.; Westhof, E. J. Mol. Biol. 1998, 276, 903-912. Published on Web 11/21/2002 14934 9 J. AM. CHEM. SOC. 2002, 124, 14934-14939 10.1021/ja027829t CCC: $22.00 © 2002 American Chemical Society