Boris Fürtig

NMR Spectroscopy of RNA

NMR spectroscopy is a powerful tool for studying proteins and nucleic acids in solution. This is illustrated by the fact that nearly half of all current RNA structures were determined by using NMR techniques. Information about the structure, dynamics, and interactions with other RNA molecules, proteins, ions, and small ligands can be obtained for RNA molecules up to 100 nucleotides. This review provides insight into the resonance assignment methods that are the first and crucial step of all NMR studies, into the determination of base‐pair geometry, into the examination of local and global RNA conformation, and into the detection of interaction sites of RNA. Examples of NMR investigations of RNA are given by using several different RNA molecules to illustrate the information content obtainable by NMR spectroscopy and the applicability of NMR techniques to a wide range of biologically interesting RNA molecules.

Interplay of ‘induced fit’ and preorganization in the ligand induced folding of the aptamer domain of the guanine binding riboswitch

Riboswitches are highly structured elements in the 5′-untranslated regions (5′-UTRs) of messenger RNA that control gene expression by specifically binding to small metabolite molecules. They consist of an aptamer domain responsible for ligand binding and an expression platform. Ligand binding in the aptamer domain leads to conformational changes in the expression platform that result in transcription termination or abolish ribosome binding. The guanine riboswitch binds with high-specificity to guanine and hypoxanthine and is among the smallest riboswitches described so far. The X-ray-structure of its aptamer domain in complex with guanine/hypoxanthine reveals an intricate RNA-fold consisting of a three-helix junction stabilized by long-range base pairing interactions. We analyzed the conformational transitions of the aptamer domain induced by binding of hypoxanthine using high-resolution NMR-spectroscopy in solution. We found that the long-range base pairing interactions are already present in the free RNA and preorganize its global fold. The ligand binding core region is lacking hydrogen bonding interactions and therefore likely to be unstructured in the absence of ligand. Mg2+-ions are not essential for ligand binding and do not change the structure of the RNA-ligand complex but stabilize the structure at elevated temperatures. We identified a mutant RNA where the long-range base pairing interactions are disrupted in the free form of the RNA but form upon ligand binding in an Mg2+-dependent fashion. The tertiary interaction motif is stable outside the riboswitch context.

Three-state mechanism couples ligand and temperature sensing in riboswitches

Riboswitches are cis-acting gene-regulatory RNA elements that can function at the level of transcription, translation and RNA cleavage. The commonly accepted molecular mechanism for riboswitch function proposes a ligand-dependent conformational switch between two mutually exclusive states. According to this mechanism, ligand binding to an aptamer domain induces an allosteric conformational switch of an expression platform, leading to activation or repression of ligand-related gene expression. However, many riboswitch properties cannot be explained by a pure two-state mechanism. Here we show that the regulation mechanism of the adenine-sensing riboswitch, encoded by the add gene on chromosome II of the human Gram-negative pathogenic bacterium Vibrio vulnificus, is notably different from a two-state switch mechanism in that it involves three distinct stable conformations. We characterized the temperature and Mg2+ dependence of the population ratios of the three conformations and the kinetics of their interconversion at nucleotide resolution. The observed temperature dependence of a pre-equilibrium involving two structurally distinct ligand-free conformations of the add riboswitch conferred efficient regulation over a physiologically relevant temperature range. Such robust switching is a key requirement for gene regulation in bacteria that have to adapt to environments with varying temperatures. The translational adenine-sensing riboswitch represents the first example, to our knowledge, of a temperature-compensated regulatory RNA element.

Time-resolved NMR methods resolving ligand-induced RNA folding at atomic resolution

Structural transitions of RNA between alternate conformations with similar stabilities are associated with important aspects of cellular function. Few techniques presently exist that are capable of monitoring such transitions and thereby provide insight into RNA dynamics and function at atomic resolution. Riboswitches are found in the 5′-UTR of mRNA and control gene expression through structural transitions after ligand recognition. A time-resolved NMR strategy was established in conjunction with laser-triggered release of the ligand from a photocaged derivative in situ to monitor the hypoxanthine-induced folding of the guanine-sensing riboswitch aptamer domain of the Bacillus subtilis xpt-pbuX operon at atomic resolution. Combining selective isotope labeling of the RNA with NMR filter techniques resulted in significant spectral resolution and allowed kinetic analysis of the buildup rates for individual nucleotides in real time. Three distinct kinetic steps associated with the ligand-induced folding were delineated. After initial complex encounter the ligand-binding pocket is formed and results in subsequent stabilization of a remote long-range loop–loop interaction. Incorporation of NMR data into experimentally restrained molecular dynamics simulations provided insight into the RNA structural ensembles involved during the conformational transition.

High-resolution NMR structure of an RNA model system: the 14-mer cUUCGg tetraloop hairpin RNA

We present a high-resolution nuclear magnetic resonance (NMR) solution structure of a 14-mer RNA hairpin capped by cUUCGg tetraloop. This short and very stable RNA presents an important model system for the study of RNA structure and dynamics using NMR spectroscopy, molecular dynamics (MD) simulations and RNA force-field development. The extraordinary high precision of the structure (root mean square deviation of 0.3 Å) could be achieved by measuring and incorporating all currently accessible NMR parameters, including distances derived from nuclear Overhauser effect (NOE) intensities, torsion-angle dependent homonuclear and heteronuclear scalar coupling constants, projection-angle-dependent cross-correlated relaxation rates and residual dipolar couplings. The structure calculations were performed with the program CNS using the ARIA setup and protocols. The structure quality was further improved by a final refinement in explicit water using OPLS force field parameters for non-bonded interactions and charges. In addition, the 2′-hydroxyl groups have been assigned and their conformation has been analyzed based on NOE contacts. The structure currently defines a benchmark for the precision and accuracy amenable to RNA structure determination by NMR spectroscopy. Here, we discuss the impact of various NMR restraints on structure quality and discuss in detail the dynamics of this system as previously determined.

Adenosine-to-inosine RNA editing controls cathepsin S expression in atherosclerosis by enabling HuR-mediated post-transcriptional regulation

Adenosine-to-inosine (A-to-I) RNA editing, which is catalyzed by a family of adenosine deaminase acting on RNA (ADAR) enzymes, is important in the epitranscriptomic regulation of RNA metabolism. However, the role of A-to-I RNA editing in vascular disease is unknown. Here we show that cathepsin S mRNA (CTSS), which encodes a cysteine protease associated with angiogenesis and atherosclerosis, is highly edited in human endothelial cells. The 3′ untranslated region (3′ UTR) of the CTSS transcript contains two inverted repeats, the AluJo and AluSx+ regions, which form a long stem–loop structure that is recognized by ADAR1 as a substrate for editing. RNA editing enables the recruitment of the stabilizing RNA-binding protein human antigen R (HuR; encoded by ELAVL1) to the 3′ UTR of the CTSS transcript, thereby controlling CTSS mRNA stability and expression. In endothelial cells, ADAR1 overexpression or treatment of cells with hypoxia or with the inflammatory cytokines interferon-γ and tumor-necrosis-factor-α induces CTSS RNA editing and consequently increases cathepsin S expression. ADAR1 levels and the extent of CTSS RNA editing are associated with changes in cathepsin S levels in patients with atherosclerotic vascular diseases, including subclinical atherosclerosis, coronary artery disease, aortic aneurysms and advanced carotid atherosclerotic disease. These results reveal a previously unrecognized role of RNA editing in gene expression in human atherosclerotic vascular diseases.

Mapping the Landscape of RNA Dynamics with NMR Spectroscopy

Among the three major classes of biomacromolecules (DNA, RNA, and proteins) RNAs pronounced dynamics are the most explicitly linked to its wide variety of functions, which include catalysis and the regulation of transcription, translation, and splicing. These functions are mediated by a range of RNA biomachinery, including such varied examples as macromolecular noncoding RNAs, microRNAs, small interfering RNAs, riboswitch RNAs, and RNA thermometers. In each case, the functional dynamics of an interconversion is characterized by an associated rate constant. In this Account, we provide an introduction to NMR spectroscopic characterization of the landscape of RNA dynamics. We introduce strategies for measuring NMR parameters at various time scales as well as the underlying models for describing the corresponding rate constants. RNA exhibits significant dynamic motion, which can be modulated by (i) intermolecular interactions, including specific and nonspecific binding of ions (such as Mg(2+) and tertiary amines), (ii) metabolites in riboswitches or RNA aptamers, and (iii) macromolecular interactions within ribonucleic protein particles, including the ribosome and the spliceosome. Our understanding of the nature of these dynamic changes in RNA targets is now being incorporated into RNA-specific approaches in the design of RNA inhibitors. Interactions of RNA with proteins, other RNAs, or small molecules often occur through binding mechanisms that follow an induced fit mechanism or a conformational selection mechanism, in which one of several populated RNA conformations is selected through ligand binding. The extent of functional dynamics, including the kinetic formation of a specific RNA tertiary fold, is dependent on the messenger RNA (mRNA) chain length. Thus, during de novo synthesis of mRNA, both in prokaryotes and eukaryotes, nascent mRNA of various lengths will adopt different secondary and tertiary structures. The speed of transcription has a critical influence on the functional dynamics of the RNA being synthesized. In addition to modulating the local dynamics of a conformational RNA ensemble, a given RNA sequence may adopt more than one global, three-dimensional structure. RNA modification is one way to select among these alternative structures, which are often characterized by nearly equal stability, but with high energy barriers for conformational interconversion. The refolding of different secondary and tertiary structures has been found to be a major regulatory mechanism for transcription and translation. These conformational transitions can be characterized with NMR spectroscopy, for any given RNA sequence, in response to external stimuli.

The Nature of Hydrogen Bonds in Cytidine⋅⋅⋅H+⋅⋅⋅Cytidine DNA Base Pairs†

Formation of hydrogen bonds is an important stabilization mechanism during biomolecular folding. In DNA, specificity in hydrogen-bond formation is essential for the transfer of genetic information. The geometry of hydrogen bonds in Watson–Crick base pairs is well characterized. Besides canonical Watson–Crick base pairing, DNA can also adopt non-canonical structures for which special hydrogen bonds are found. A particular class of non-canonical structures involves protonated cytidine nucleotides: for example, C–G– C base triplets with one protonated cytidine interacting with the Hoogsteen side of the guanosine nucleotide are found in triplex DNA and hemiprotonated cytidine···H···cytidine base pairs (abbreviated as C·C) are formed in DNA imotifs at slightly acidic pH values. In these C–C base pairs, the nitrogen atoms in the 3-positions of the two cytidine residues on opposite strands share a single proton (Scheme 1b). The DNA i-motif structure is stabilized by the formation of C·C base pairs. The arrangement of this four-stranded structure is induced by the intercalation of the protonated base pairs. Cytidines in one base pair are arranged in parallel strands (Scheme 1c). In general, hydrogen bonding in an hemiprotonated N···H···N moiety can be described in either one of the following two ways: as 1) a symmetric hydrogen bond with a single-well potential or as 2) a double-well potential with a delocalized proton that oscillates between the two wells with associated transition rate constants ka and kb (Scheme 1a). C·C base pairs (Scheme 1b) were first observed in crystals of cytidine-5-acetic acid. Their incorporation into a d(TC5) tetramer leads to DNA i-motif formation at pH 6, which has been thoroughly investigated by Leroy et al. Based on line-width analysis, Leroy et al. postulated a hydrogen bonding with a double-well potential with a transition rate of 8 10 s 1 as upper limit for the proton transfer. From the NMR spectroscopy-based structure of an intramolecular i-motif (protein data bank (PDB) code: 1ELN) it can be deduced that the C·C base pairs are presumably planar with a maximal deviation of 188 and an N–N distance of 2.6– 2.8 . Herein, we performed a combined NMR spectroscopy and quantum chemical investigation of the N···H···N bond in the C·C hemiprotonated base pairs of the 21 nucleotide (nt) DNA with sequence d(CCCTAA)3CCC. By utilizing NMR spectroscopy of selectively isotope-labeled DNA strands, J(NH) coupling constants, H and N chemical shifts, as well as solvent exchange parameters have been measured and quantum chemical calculations have been performed to obtain information about hydrogen bonding geometry and strength in solution. From our combined NMR/QM approach, we infer that N···H···N bonds in the C·C hemiprotonated base pairs have to be described as hydrogen bonds with asymmetric double-well potentials ((2) in Scheme 1a). The DNA structure of the i-motif introduces local asymmetries resulting in two distinct wells for the proton transfer and different rates for back and forward transfer. The fast hopping rate leads to a single H chemical shift and to two distinct N chemical shifts (Table 1). The J(NH) coupling constants are averaged and from the measured couplings for the individual nitrogen atoms the populations of the conformations can be derived (Table 2). For our investigations, we utilized selectively labeled DNA sequences of the 21 mer DNA with sequence d(CCCTAA)3CCC in which only a single cytidine was 50% C and N enriched at a time, resulting in 12 different DNA Scheme 1. a) Symmetric (1) and asymmetric (2) N···H···N hydrogenbonding schemes in singly protonated C·C base pairs in DNA i-motif. b) N···H···N hydrogen bonds in hemiprotonated C·C base pairs. c) The intercalated, four-stranded DNA i-motif structure.

Multiple conformational states of riboswitches fine-tune gene regulation

Riboswitches are structured regions of mRNAs that modulate gene expression in response to specific binding of low molecular-weight ligands. They function by induced transitions between different functional conformations. The standard model assumed that the two functional states, the ligand-bound and ligand-free state, populated only two stable conformations. Recent discoveries of multiple conformations for the apo-state and holo-state of riboswitches challenge this model. Moreover, it becomes evident that detected conformational heterogeneity--mostly in the apo-state--provides sensitivity to multiple environmental inputs for riboswitch-based gene-regulation.

A caged uridine for the selective preparation of an RNA fold and determination of its refolding kinetics by real-time NMR.

The biological function of many RNAs is linked to a reversible conformational switching between active and inactive folds that differ in their secondary and/or tertiary structures. Regulatory RNAs, such as riboswitches, 3] and catalytic RNAs, such as ribozymes, undergo this conformational switching upon binding of metabolites or during catalysis. In general, the energetic differences between the alternative conformations are only small, and therefore the equilibrium distribution is strongly affected by ligand or substrate binding, or by small structural modifications, such as methylation of the nucleobases. Recently, we have introduced a method for studying the kinetics of RNA-conformational switching by time-resolved NMR spectroscopy. The method is based on RNA sequences modified with 1-(2-nitrophenyl)ethyl-substituted (NPE-substituted) nucleobases, which are designed to impair the formation of selected base pairs. When introduced at proper sequence position within the oligonucleotide fold, these photocleavable modifications lead to a specific destabilization of preselected folds. Rapid and traceless removal of the photolabile groups, carried out by coupling laser optics into the NMR tube, then allows NMR detection of the refolding process. The induced structural transition can be monitored from the time-dependent increase and decrease in intensity of imino proton resonances, which belong to specific base pairs present in specific folds. For medium-sized oligonucleotides, these resonances often show sufficient resolution in chemical shifts to be resolved in 1D experiments. In this context, we have already reported the refolding kinetics and a plausible refolding mechanism of the bistable 20-mer RNA sequence 5’-r[GACCGGAAGGUCCGCCUUCC]-3’. For this study, an (S)-NPE-substituted guanosine was introduced at position 6. The activation energy was Ea=26 kcalmol 1 and the frequency factor was A=10 s 1 for the forward process, or 31 kcalmol 1 and 10 s 1 for the backward process, respectively. These refolding rates are surprisingly low compared to the folding of similar systems from an unfolded state and are in agreement with a transient disruption of half of the base pairs. Together with an additional dynamical characterization by comparative water-exchange experiments, an associative mechanism was deduced, in which refolding is initiated by base-pair contacts between previously unpaired nucleotides. In this communication, we extend our real-time-NMR investigation of RNA secondary structural transitions to another bistable, 20-mer RNA sequence 5’-r(GAAGGGCAACCUUCGGGUUG)-3’ that was designed by Hçbartner et al. This RNA sequence adopts two almost equally stable hairpin folds A and B, each consisting of a five-base-pair stem, a stable tetraloop and six unpaired nucleotides (Figure 1C, D). In order to selectively destabilize fold B, we introduced an (R)-NPE-protected uridine at position U18 (Figure 1B). Structural and kinetic characterization were carried out by analyzing the imino proton NMR signals, each representing a specific base pair. The unmodified sequence shows 12 signals in the iminoproton region (Figure 1D), which were assigned to the base pairs of the two coexisting folds A and B by H,H-NOESY experiments. The signals assigned to fold A agreed well with the signals of the corresponding truncated 15-mer RNA hairpin (Figure 1A). After introduction of the (R)-NPE-protected U18 into the 20-mer RNA sequence, only the set of six imino proton signals characteristic of fold A were detected (Figure 1B). Importantly, this single NPE-modification completely suppressed the formation of fold B by preventing the formation of the associated U18–A9 base pair. Removal of the NPE group by photolysis and equilibration resulted in a NMR spectrum (Figure 1C) that was practically identical to the spectrum of the parent sequence (Figure 1D). To determine the kinetic parameters for refolding, the NPE group was released by a short continuous-wave laser pulse (duration 1.5 s) guided by an optical fibre into the NMR tube. Relaxation of the system towards equilibrium was monitored by the subsequent recording of 1D H spectra. 11] The refolding rate constants were obtained by fitting the time-dependent intensities of two selected pairs of imino proton signals according to the equation for a reversible first-order process (Figure 2A–D). The independently determined equilibrium constant KA!B was employed as an additional fit parameter. The thermodynamic parameters DHA!B=1.6 0.3 kcalmol 1 and DSA!B=6.0 1.0 calmol K 1 were determined from the temperature dependence of these K values (Figure S1 in the Supporting Information). The directly determined rate constants kA!B ranged between 0.005 0.001 s 1 at 283 K and 0.091 0.020 s 1 at 298 K; the rate constants kB!A were calculated from the kA!B and K values, and ranged between 0.005 0.002 s 1 at 283 K and 0.076 0.020 s 1 at 298 K. In each experiment, rates that were identical within the error range were determined for the two independently analyzed pairs of imino proton signals (all rates are given in Figure S2 in the Supporting Information). Arrhenius analysis revealed an activation energy of Ea,A!B=32.7 3.0 kcalmol 1 (Ea,B!A=31.0 [a] P. Wenter, A. Hainard, Prof. Dr. S. Pitsch Laboratory of Nucleic Acid Chemistry Ecole Polyt!chnique F!d!rale de Lausanne (EPFL) 1015 Lausanne (Switzerland) Fax: (+41)21-693-9355 E-mail : stefan.pitsch@epfl.ch [b] B. F6rtig, Prof. Dr. H. Schwalbe Johann Wolfgang Goethe University Institute for Organic Chemistry and Chemical Biology Center for Biomolecular Magnetic Resonance Marie-Curie Strasse 11, Frankfurt, 60439 Frankfurt/Main (Germany) Fax: (+49)69-7982-9515 E-mail : schwalbe@nmr.uni-frankfurt.de Supporting information for this article is available on the WWW under http://www.chembiochem.org or from the author.