Many biological processes are RNA-mediated, but higher-order structures for most RNAs are unknown, making it difficult to understand how RNA structure governs function. Here we describe SHAPE mutational profiling (SHAPE-MaP) that makes possible de novo and large-scale identification of RNA functional motifs. Sites of 2’-hydroxyl acylation by SHAPE are encoded as non-complementary nucleotides during cDNA synthesis, as measured by massively parallel sequencing. SHAPE-MaP-guided modeling identified greater than 90% of accepted base pairs in complex RNAs of known structure and was used to define a second-generation model for the HIV-1 RNA genome. The HIV-1 model contains all known structured motifs and previously unknown elements, including experimentally validated pseudoknots. SHAPE-MaP yields accurate and high-resolution secondary structure models, enables analysis of low abundance RNAs, disentangles sequence polymorphisms in single experiments, and will ultimately democratize RNA structure analysis.

SHAPE chemistries exploit small electrophilic reagents that react with the 2′-hydroxyl group to interrogate RNA structure at single-nucleotide resolution. Mutational profiling (MaP) identifies modified residues based on the ability of reverse transcriptase to misread a SHAPE-modified nucleotide and then counting the resulting mutations by massively parallel sequencing. The SHAPE-MaP approach measures the structure of large and transcriptome-wide systems as accurately as for simple model RNAs. This protocol describes the experimental steps, implemented over three days, required to perform SHAPE probing and construct multiplexed SHAPE-MaP libraries suitable for deep sequencing. These steps include RNA folding and SHAPE structure probing, mutational profiling by reverse transcription, library construction, and sequencing. Automated processing of MaP sequencing data is accomplished using two software packages. ShapeMapper converts raw sequencing files into mutational profiles, creates SHAPE reactivity plots, and provides useful troubleshooting information, often within an hour. SuperFold uses these data to model RNA secondary structures, identify regions with well-defined structures, and visualize probable and alternative helices, often in under a day. We illustrate these algorithms with the E. coli thiamine pyrophosphate riboswitch, E. coli 16S rRNA, and HIV-1 genomic RNAs. SHAPE-MaP can be used to make nucleotide-resolution biophysical measurements of individual RNA motifs, rare components of complex RNA ensembles, and entire transcriptomes. The straightforward MaP strategy greatly expands the number, length, and complexity of analyzable RNA structures.

The hepatitis delta virus (HDV) ribozyme uses the nucleobase C75 and a hydrated Mg(2+) ion as the general acid-base catalysts in phosphodiester bond cleavage at physiological salt. A mechanistic framework has been advanced that involves one Mg(2+)-independent and two Mg(2+)-dependent channels. The rate-pH profile for wild-type (WT) ribozyme in the Mg(2+)-free channel is inverted relative to the fully Mg(2+)-dependent channel, with each having a near-neutral pKa. Inversion of the rate-pH profile was used as the crux of a mechanistic argument that C75 serves as general acid both in the presence and absence of Mg(2+). However, subsequent studies on a double mutant (DM) ribozyme suggested that the pKa observed for WT in the absence of Mg(2+) arises from ionization of C41, a structural nucleobase. To investigate this further, we acquired rate-pH/pD profiles and proton inventories for WT and DM in the absence of Mg(2+). Corrections were made for effects of ionic strength on hydrogen ion activity and pH meter readings. Results are accommodated by a model wherein the Mg(2+)-free pKa observed for WT arises from ionization of C75, and DM reactivity is compromised by protonation of C41. The Brønsted base appears to be water or hydroxide ion depending on pH. The observed pKa's are related to salt-dependent pH titrations of a model oligonucleotide, as well as electrostatic calculations, which support the local environment for C75 in the absence of Mg(2+) being similar to that in the presence of Mg(2+) and impervious to bulk ions. Accordingly, the catalytic role of C75 as the general acid does not appear to depend on divalent ions or the identity of the Brønsted base.

Secondary structural motifs play essential roles in the folding and function of RNA and DNA molecules. Previous work from our lab compared the folding of small DNA and RNA hairpin loops containing a sheared GA pair [Moody, E. M., Feerar, J. C., and Bevilacqua, P. C. (2004) Biochemistry 43, 7992-7998]. We found that the small DNA hairpins fold in a highly cooperative manner with indirect coupling, while their RNA counterparts fold in a much less cooperative fashion and display direct coupling. Herein, we extend this study to the double-stranded helix. We carried out double mutant cycles on base pairs having identical nearest-neighbor contexts but located in either external or internal helical registers. In the external register, both RNA and DNA exhibit extensive folding cooperativity between the penultimate and terminal base pair, which is independent of mismatch identity. In contrast, DNA exhibits virtually no folding cooperativity in the center of the helix, while RNA maintains substantial coupling, which is dependent on mismatch identity. Two models account for these non-nearest-neighbor effects: one involves the unfavorable entropy of helix initiation common to DNA and RNA, and the other involves steric and electrostatic strain peculiar to RNA. These data show that RNA can display cooperativity less than, greater than, or equal to that of DNA depending on context and position.

Secondary structure plays critical roles in nucleic acid function. Mismatches in DNA can lead to mutation and disease, and some mismatches involve a protonated base. Among protonated mismatches, A(+).C wobble pairs form near physiological pH and have relatively minor effects on helix geometry, making them especially important in biology. Herein, we investigate effects of helix position, temperature, and ionic strength on pK(a) shifting in A(+).C wobble pairs in DNA. We observe that pK(a) shifting is favored by internal A(+).C wobbles, which have low cooperativities of folding and make large contributions to stability, and disfavored by external A(+).C wobbles, which have high folding cooperativities but make small contributions to stability. An inverse relationship between pK(a) shifting and temperature is also found, which supports a model in which protonation is enthalpically favored overall and entropically correlated with cooperativity of folding. We also observe greater pK(a) shifts as the ionic strength decreases, consistent with anticooperativity between proton binding and counterion-condensed monovalent cation. Under the most favorable temperature and ionic strength conditions tested, a pK(a) of 8.0 is observed for the A(+).C wobble pair, which represents an especially large shift ( approximately 4.5 pK(a) units) from the unperturbed pK(a) value of adenosine. This study suggests that protonated A(+).C wobble pairs exist in DNA under biologically relevant conditions, where they can drive conformational changes and affect replication and transcription fidelity.

Structured RNA elements within messenger RNA often direct or modulate the cellular production of active proteins. As reviewed here, RNA structures have been discovered that govern nearly every step in protein production: mRNA production and stability; translation initiation, elongation, and termination; protein folding; and cellular localization. Regulatory RNA elements are common within RNAs from every domain of life. This growing body of RNA-mediated mechanisms continues to reveal new ways in which mRNA structure regulates translation. We integrate examples from several different classes of RNA structure-mediated regulation to present a global perspective that suggests that the secondary and tertiary structure of RNA ultimately constitutes an additional level of the genetic code that both guides and regulates protein biosynthesis.

RNA plays essential roles in much of biology. These functions are dictated by structures mediated by hydrogen bonding, stacking, electrostatics, and steric interactions. Roles of unsatisfied hydrogen bond functionalities in these structures are less well understood. Herein, we evaluated the energetic contributions of unsatisfied hydrogen bonding groups by placing chemically modified substituents in select internal positions in RNA helices and conducting thermodynamic studies. We find that unsatisfied carbonyl groups make exceptional contributions to structure formation (approximately 3 kcal/mol in free energy), most likely due to a combination of strain and dehydration effects. Thus, unsatisfied hydrogen bonding groups are likely key determinants in the folding energetics and specificity of many RNA and DNA molecules and may be especially important in tertiary structure interactions.

RNA SHAPE chemistry yields quantitative, single nucleotide resolution structural information based on the reaction of the 2′-hydroxyl group of conformationally flexible nucleotides with electrophilic SHAPE reagents. However, SHAPE technology has been limited by the requirement that sites of RNA modification be detected by primer extension. Primer extension results in loss of information at both the 5′ and 3′ ends of an RNA and requires multiple experimental steps. Here we describe RNase-detected SHAPE (Selective 2′-Hydroxyl Acylation analyzed by Protection from Exoribonuclease) that uses a processive, 3′→5′ exoribonuclease, RNase R, to detect covalent adducts in 5′-end labeled RNA in a one-tube experiment. RNase R degrades RNA but stops quantitatively three and four nucleotides 3′ of a nucleotide containing a covalent adduct at the ribose 2′-hydroxyl or the pairing face of a nucleobase, respectively. We illustrate this technology by characterizing ligand-induced folding for the E. coli thiamine pyrophosphate riboswitch RNA. RNase-detected SHAPE is a facile, two-day approach that can be used to analyze diverse covalent adducts in any RNA molecule, including short RNAs not amenable to analysis by primer extension and RNAs with functionally important structures at their 5′ or 3′ ends.