Wfold: A new method for predicting RNA secondary structure with deep learning

Precise estimations of RNA secondary structures have the potential to reveal the various roles that non-coding RNAs play in regulating cellular activity. However, the mainstay of traditional RNA secondary structure prediction methods relies on thermos-dynamic models via free energy minimization, a laborious process that requires a lot of prior knowledge. Here, RNA secondary structure prediction using Wfold, an end-to-end deep learning-based approach, is suggested. Wfold is trained directly on annotated data and base-pairing criteria. It makes use of an image-like representation of RNA sequences, which an enhanced U-net incorporated with a transformer encoder can process effectively. Wfold eventually increases the accuracy of RNA secondary structure prediction by combining the benefits of self-attention mechanism's mining of long-range information with U-net's ability to gather local information. We compare Wfold's performance using RNA datasets that are within and across families. When trained and evaluated on different RNA families, it achieves a similar performance as the traditional methods, but dramatically outperforms the state-of-the-art methods on within-family datasets. Moreover, Wfold can also reliably forecast pseudoknots. The findings imply that Wfold may be useful for improving sequence alignment, functional annotations, and RNA structure modeling.

Pioneering role of RNA in the early evolution of life

The catalytic, regulatory and structural properties of RNA, combined with their extraordinary ubiquity in cellular processes, are consistent with the proposal that this molecule played a much more conspicuous role in heredity and metabolism during the early stages of biological evolution. This review explores the pivotal role of RNA in the earliest life forms and its relevance in modern biological systems. It examines current models that study the early evolution of life, providing insights into the primordial RNA world and its legacy in contemporary biology.

SparseRNAfolD: optimized sparse RNA pseudoknot-free folding with dangle consideration

MOTIVATION

Computational RNA secondary structure prediction by free energy minimization is indispensable for analyzing structural RNAs and their interactions. These methods find the structure with the minimum free energy (MFE) among exponentially many possible structures and have a restrictive time and space complexity ( O ( n 3 ) time and O ( n 2 ) space for pseudoknot-free structures) for longer RNA sequences. Furthermore, accurate free energy calculations, including dangle contributions can be difficult and costly to implement, particularly when optimizing for time and space requirements.

RESULTS

Here we introduce a fast and efficient sparsified MFE pseudoknot-free structure prediction algorithm, SparseRNAFolD, that utilizes an accurate energy model that accounts for dangle contributions. While the sparsification technique was previously employed to improve the time and space complexity of a pseudoknot-free structure prediction method with a realistic energy model, SparseMFEFold, it was not extended to include dangle contributions due to the complexity of computation. This may come at the cost of prediction accuracy. In this work, we compare three different sparsified implementations for dangle contributions and provide pros and cons of each method. As well, we compare our algorithm to LinearFold, a linear time and space algorithm, where we find that in practice, SparseRNAFolD has lower memory consumption across all lengths of sequence and a faster time for lengths up to 1000 bases.

CONCLUSION

Our SparseRNAFolD algorithm is an MFE-based algorithm that guarantees optimality of result and employs the most general energy model, including dangle contributions. We provide a basis for applying dangles to sparsified recursion in a pseudoknot-free model that has the potential to be extended to pseudoknots.

The role of secondary structures in the functioning of 3' untranslated regions of mRNA: A review of functions of 3' UTRs' secondary structures and hypothetical involvement of secondary structures in cytoplasmic polyadenylation in Drosophila

3' untranslated regions (3' UTRs) of mRNAs have many functions, including mRNA processing and transport, translational regulation, and mRNA degradation and stability. These different functions require cis-elements in 3' UTRs that can be either sequence motifs or RNA structures. Here we review the role of secondary structures in the functioning of 3' UTRs and discuss some of the trans-acting factors that interact with these secondary structures in eukaryotic organisms. We propose potential participation of 3'-UTR secondary structures in cytoplasmic polyadenylation in the model organism Drosophila melanogaster. Because the secondary structures of 3' UTRs are essential for post-transcriptional regulation of gene expression, their disruption leads to a wide range of disorders, including cancer and cardiovascular diseases. Trans-acting factors, such as STAU1 and nucleolin, which interact with 3'-UTR secondary structures of target transcripts, influence the pathogenesis of neurodegenerative diseases and tumor metastasis, suggesting that they are possible therapeutic targets.

Comparative Study of Single-stranded Oligonucleotides Secondary Structure Prediction Tools

BACKGROUND

Single-stranded nucleic acids (ssNAs) have important biological roles and a high biotechnological potential linked to their ability to bind to numerous molecular targets. This depends on the different spatial conformations they can assume. The first level of ssNAs spatial organisation corresponds to their base pairs pattern, i.e. their secondary structure. Many computational tools have been developed to predict the ssNAs secondary structures, making the choice of the appropriate tool difficult, and an up-to-date guide on the limits and applicability of current secondary structure prediction tools is missing. Therefore, we performed a comparative study of the performances of 9 freely available tools (mfold, RNAfold, CentroidFold, CONTRAfold, MC-Fold, LinearFold, UFold, SPOT-RNA, and MXfold2) on a dataset of 538 ssNAs with known experimental secondary structure.

RESULTS

The minimum free energy-based tools, namely mfold and RNAfold, and some tools based on artificial intelligence, namely CONTRAfold and MXfold2, provided the best results, with [Formula: see text] of exact predictions, whilst MC-fold seemed to be the worst performing tool, with only [Formula: see text] of exact predictions. In addition, UFold and SPOT-RNA are the only options for pseudoknots prediction. Including in the analysis of mfold and RNAfold results 5-10 suboptimal solutions further improved the performances of these tools. Nevertheless, we could observe issues in predicting particular motifs, such as multiple-ways junctions and mini-dumbbells, or the ssNAs whose structure has been determined in complex with a protein. In addition, our benchmark shows that some effort has to be paid for ssDNA secondary structure predictions.

CONCLUSIONS

In general, Mfold, RNAfold, and MXfold2 seem to currently be the best choice for the ssNAs secondary structure prediction, although they still show some limits linked to specific structural motifs. Nevertheless, actual trends suggest that artificial intelligence has a high potential to overcome these remaining issues, for example the recently developed UFold and SPOT-RNA have a high success rate in predicting pseudoknots.

Structural basis for water modulating RNA duplex formation in the CUG repeats of myotonic dystrophy type 1

Secondary structures formed by expanded CUG RNA are involved in the pathobiology of myotonic dystrophy type 1. Understanding the molecular basis of toxic RNA structures can provide insights into the mechanism of disease pathogenesis and accelerate the drug discovery process. Here, we report the crystal structure of CUG repeat RNA containing three U-U mismatches between C-G and G-C base pairs. The CUG RNA crystallizes as an A-form duplex, with the first and third U-U mismatches adopting a water-mediated asymmetric mirror isoform geometry. We found for the first time that a symmetric, water-bridged U-H₂O-U mismatch is well tolerated within the CUG RNA duplex, which was previously suspected but not observed. The new water-bridged U-U mismatch resulted in high base-pair opening and single-sided cross-strand stacking interactions, which in turn dominate the CUG RNA structure. Furthermore, we performed molecular dynamics (MD) simulations that complemented the structural findings and proposed that the first and third U-U mismatches are interchangeable conformations, while the central water-bridged U-U mismatch represents an intermediate state that modulates the RNA duplex conformation. Collectively, the new structural features provided in this work are important for understanding the recognition of U-U mismatches in CUG repeats by external ligands such as proteins or small molecules.

Stress promotes RNA G-quadruplex folding in human cells

Guanine (G)-rich nucleic acids can fold into G-quadruplex (G4) structures under permissive conditions. Although many RNAs contain sequences that fold into RNA G4s (rG4s) in vitro, their folding and functions in vivo are not well understood. In this report, we showed that the folding of putative rG4s in human cells into rG4 structures is dynamically regulated under stress. By using high-throughput dimethylsulfate (DMS) probing, we identified hundreds of endogenous stress-induced rG4s, and validated them by using an rG4 pull-down approach. Our results demonstrate that stress-induced rG4s are enriched in mRNA 3'-untranslated regions and enhance mRNA stability. Furthermore, stress-induced rG4 folding is readily reversible upon stress removal. In summary, our study revealed the dynamic regulation of rG4 folding in human cells and suggested that widespread rG4 motifs may have a global regulatory impact on mRNA stability and cellular stress response.

An overview of structural approaches to study therapeutic RNAs

RNAs provide considerable opportunities as therapeutic agent to expand the plethora of classical therapeutic targets, from extracellular and surface proteins to intracellular nucleic acids and its regulators, in a wide range of diseases. RNA versatility can be exploited to recognize cell types, perform cell therapy, and develop new vaccine classes. Therapeutic RNAs (aptamers, antisense nucleotides, siRNA, miRNA, mRNA and CRISPR-Cas9) can modulate or induce protein expression, inhibit molecular interactions, achieve genome editing as well as exon-skipping. A common RNA thread, which makes it very promising for therapeutic applications, is its structure, flexibility, and binding specificity. Moreover, RNA displays peculiar structural plasticity compared to proteins as well as to DNA. Here we summarize the recent advances and applications of therapeutic RNAs, and the experimental and computational methods to analyze their structure, by biophysical techniques (liquid-state NMR, scattering, reactivity, and computational simulations), with a focus on dynamic and flexibility aspects and to binding analysis. This will provide insights on the currently available RNA therapeutic applications and on the best techniques to evaluate its dynamics and reactivity.

Conformational effects of a cancer-linked mutation in pri-miR-30c RNA

MicroRNAs (miRNAs) are small, noncoding RNAs that mediate post-transcriptional downregulation of specific target genes. These transcripts are the products of a two-step processing pathway; primary miRNAs (pri-miRNAs) are processed by Drosha into individual precursor miRNA (pre-miRNA) hairpins, which are subsequently processed by Dicer into mature miRNAs. Single nucleotide polymorphisms (SNPs) that occur in pri-miRNAs, pre-miRNAs and mature miRNAs have been shown to affect the processing of specific target genes by modulating Drosha and Dicer processing or interactions with RNA binding proteins (RBPs). Using NMR and single-molecule optical tweezer experiments, we have investigated the conformational effects of a cancer-linked G/A mutation in the terminal loop of pri-miR-30c RNA, and how this influences binding by the SRSF3 and hnRNP A1 RBPs, which are implicated in its processing. Our results reveal that the wildtype and G/A variant pri-miR-30c RNAs adopt very similar elongated stem-loop structures, both of which are bound by SRSF3. However, while both wildtype and G/A pri-miR-30c RNAs can form dimeric kissing hairpin structures, the G to A mutation results in partial destabilization of the dimer in the variant transcript. This promotes recognition and binding by hnRNP A1, an RBP that enhances pri-miR-30c processing. Our data provide structural insight into the conformational effects of a G/A mutation in pri-miR-30c RNA and how this could affect processing and promote cancer.

Using total RNA quality metrics for time since deposition estimates in degrading bloodstains

The physicochemical changes occurring in biomolecules in degrading bloodstains can be used to approximate the time since deposition (TSD) of bloodstains. This would provide forensic scientists with critical information regarding the timeline of the events involving bloodshed. Our study aims to quantify the timewise degradation trends and temperature dependence found in total RNA from bloodstains without the use of amplification, expanding the scope of the RNA TSD research which has traditionally targeted mRNA and miRNA. Bovine blood with ACD-A anticoagulant was deposited and stored in plastic microcentrifuge tubes at 21 or 4°C and tested over different timepoints spanning 1 week. Total RNA was extracted from each sample and analyzed using automated high sensitivity gel electrophoresis. Nine RNA metrics were visually assessed and quantified using linear and mixed models. The RNA Integrity Number equivalent (RINe) and DV200 were not influenced by the addition of anticoagulant and demonstrated strong negative trends over time. The RINe model fit was high (R²  = 0.60), and while including the biological replicate as a random effect increased the fit for all RNA metrics, no significant differences were found between biological replicates stored at the same temperature for the RINe and DV200. This suggests that these standardized metrics can be directly compared between scenarios and individuals, with DV200 having an inflection point at approximately 28 h. This study provides a novel approach for blood TSD research, revealing metrics that are not affected by inter-individual variation, and improving our understanding of the rapid RNA degradation occurring in bloodstains.

Towards SINEUP-based therapeutics: Design of an in vitro synthesized SINEUP RNA

SINEUPs are a novel class of natural and synthetic non-coding antisense RNA molecules able to increase the translation of a target mRNA. They present a modular organization comprising an unstructured antisense target-specific domain, which sets the specificity of each individual SINEUP, and a structured effector domain, which is responsible for the translation enhancement. In order to design a fully functional in vitro transcribed SINEUP for therapeutics applications, SINEUP RNAs were synthesized in vitro with a variety of chemical modifications and screened for their activity on endogenous target mRNA upon transfection. Three combinations of modified ribonucleotides-2'O methyl-ATP (Am), N6 methyl-ATP (m6A), and pseudo-UTP (ψ)-conferred SINEUP activity to naked RNA. The best combination tested in this study was fully modified with m6A and ψ. Aside from functionality, this combination conferred improved stability upon transfection and higher thermal stability. Common structural determinants of activity were identified by circular dichroisms, defining a core functional structure that is achieved with different combinations of modifications.

Photoswitches for controllable RNA binding: a future approach in the RNA-targeting therapy

RNA is an emerging drug target that opens new perspectives in the treatment of viral and bacterial infections, cancer and a range of so far incurable genetic diseases. Among the various strategies towards the design and development of selective and efficient ligands for targeting and detection of therapeutically relevant RNA, photoswitchable RNA binders represent a very promising approach due to the possibility to control the ligand-RNA and protein-RNA interactions by light with high spatiotemporal resolution. However, the field of photoswitchable RNA binders still remains underexplored due to challenging design of lead structures that should combine high RNA binding selectivity with efficient photochemical performance. The aim of this highlight article is to describe the development of photoswitchable noncovalent RNA binders and to outline the current situation and perspectives of this emerging interdisciplinary field.

The Role of RNA Secondary Structure in Regulation of Gene Expression in Bacteria

Due to the high exposition to changing environmental conditions, bacteria have developed many mechanisms enabling immediate adjustments of gene expression. In many cases, the required speed and plasticity of the response are provided by RNA-dependent regulatory mechanisms. This is possible due to the very high dynamics and flexibility of an RNA structure, which provide the necessary sensitivity and specificity for efficient sensing and transduction of environmental signals. In this review, we will discuss the current knowledge about known bacterial regulatory mechanisms which rely on RNA structure. To better understand the structure-driven modulation of gene expression, we describe the basic theory on RNA structure folding and dynamics. Next, we present examples of multiple mechanisms employed by RNA regulators in the control of bacterial transcription and translation.

RNAs as Regulators of Cellular Matchmaking

RNA molecules are increasingly being identified as facilitating or impeding the interaction of proteins and nucleic acids, serving as so-called scaffolds or decoys. Long non-coding RNAs have been commonly implicated in such roles, particularly in the regulation of nuclear processes including chromosome topology, regulation of chromatin state and gene transcription, and assembly of nuclear biomolecular condensates such as paraspeckles. Recently, an increased awareness of cytoplasmic RNA scaffolds and decoys has begun to emerge, including the identification of non-coding regions of mRNAs that can also function in a scaffold-like manner to regulate interactions of nascently translated proteins. Collectively, cytoplasmic RNA scaffolds and decoys are now implicated in processes such as mRNA translation, decay, protein localization, protein degradation and assembly of cytoplasmic biomolecular condensates such as P-bodies. Here, we review examples of RNA scaffolds and decoys in both the nucleus and cytoplasm, illustrating common themes, the suitability of RNA to such roles, and future challenges in identifying and better understanding RNA scaffolding and decoy functions.

Compaction of RNA Hairpins and Their Kissing Complexes in Native Electrospray Mass Spectrometry

When electrosprayed from typical native MS solution conditions, RNA hairpins and kissing complexes acquire charge states at which they get significantly more compact in the gas phase than their initial structure in solution. Here, we also show the limits of using force field molecular dynamics to interpret the structures of nucleic acid complexes in the gas phase, as the predicted CCS distributions do not fully match the experimental ones. We suggest that higher level calculation levels should be used in the future.

A New Approach to 3D Modeling of Inhomogeneous Populations of Viral Regulatory RNA

Tertiary structure (3D) is the physical context of RNA regulatory activity. Retroviruses are RNA viruses that replicate through the proviral DNA intermediate transcribed by hosts. Proviral transcripts form inhomogeneous populations due to variable structural ensembles of overlapping regulatory RNA motifs in the 5′-untranslated region (UTR), which drive RNAs to be spliced or translated, and/or dimerized and packaged into virions. Genetic studies and structural techniques have provided fundamental input constraints to begin predicting HIV 3D conformations in silico. Using SimRNA and sets of experimentally-determined input constraints of HIV^NL4-3 trans-activation responsive sequence (TAR) and pairings of unique-5′ (U5) with dimerization (DIS) or AUG motifs, we calculated a series of 3D models that differ in proximity of 5′-Cap and the junction of TAR and PolyA helices; configuration of primer binding site (PBS)-segment; and two host cofactors binding sites. Input constraints on U5-AUG pairings were most compatible with intramolecular folding of 5′-UTR motifs in energetic minima. Introducing theoretical constraints predicted metastable PolyA region drives orientation of 5′-Cap with TAR, U5 and PBS-segment helices. SimRNA and the workflow developed herein provides viable options to predict 3D conformations of inhomogeneous populations of large RNAs that have been intractable to conventional ensemble methods.

Intrinsic Regulatory Role of RNA Structural Arrangement in Alternative Splicing Control

Alternative splicing is a highly sophisticated process, playing a significant role in posttranscriptional gene expression and underlying the diversity and complexity of organisms. Its regulation is multilayered, including an intrinsic role of RNA structural arrangement which undergoes time- and tissue-specific alterations. In this review, we describe the principles of RNA structural arrangement and briefly decipher its cis- and trans-acting cellular modulators which serve as crucial determinants of biological functionality of the RNA structure. Subsequently, we engage in a discussion about the RNA structure-mediated mechanisms of alternative splicing regulation. On one hand, the impairment of formation of optimal RNA structures may have critical consequences for the splicing outcome and further contribute to understanding the pathomechanism of severe disorders. On the other hand, the structural aspects of RNA became significant features taken into consideration in the endeavor of finding potential therapeutic treatments. Both aspects have been addressed by us emphasizing the importance of ongoing studies in both fields.

Accurate inference of the full base-pairing structure of RNA by deep mutational scanning and covariation-induced deviation of activity

Despite the large number of noncoding RNAs in human genome and their roles in many diseases include cancer, we know very little about them due to lack of structural clues. The centerpiece of the structural clues is the full RNA base-pairing structure of secondary and tertiary contacts that can be precisely obtained only from costly and time-consuming 3D structure determination. Here, we performed deep mutational scanning of self-cleaving CPEB3 ribozyme by error-prone PCR and showed that a library of <5 × 104 single-to-triple mutants is sufficient to infer 25 of 26 base pairs including non-nested, nonhelical, and noncanonical base pairs with both sensitivity and precision at 96%. Such accurate inference was further confirmed by a twister ribozyme at 100% precision with only noncanonical base pairs as false negatives. The performance was resulted from analyzing covariation-induced deviation of activity by utilizing both functional and nonfunctional variants for unsupervised classification, followed by Monte Carlo (MC) simulated annealing with mutation-derived scores. Highly accurate inference can also be obtained by combining MC with evolution/direct coupling analysis, R-scape or epistasis analysis. The results highlight the usefulness of deep mutational scanning for high-accuracy structural inference of self-cleaving ribozymes with implications for other structured RNAs that permit high-throughput functional selections.

RNA secondary structure prediction using an ensemble of two-dimensional deep neural networks and transfer learning

The majority of our human genome transcribes into noncoding RNAs with unknown structures and functions. Obtaining functional clues for noncoding RNAs requires accurate base-pairing or secondary-structure prediction. However, the performance of such predictions by current folding-based algorithms has been stagnated for more than a decade. Here, we propose the use of deep contextual learning for base-pair prediction including those noncanonical and non-nested (pseudoknot) base pairs stabilized by tertiary interactions. Since only [Formula: see text]250 nonredundant, high-resolution RNA structures are available for model training, we utilize transfer learning from a model initially trained with a recent high-quality bpRNA dataset of [Formula: see text]10,000 nonredundant RNAs made available through comparative analysis. The resulting method achieves large, statistically significant improvement in predicting all base pairs, noncanonical and non-nested base pairs in particular. The proposed method (SPOT-RNA), with a freely available server and standalone software, should be useful for improving RNA structure modeling, sequence alignment, and functional annotations.

A Polymer Physics Framework for the Entropy of Arbitrary Pseudoknots

The accurate prediction of RNA secondary structure from primary sequence has had enormous impact on research from the past 40 years. Although many algorithms are available to make these predictions, the inclusion of non-nested loops, termed pseudoknots, still poses challenges arising from two main factors: 1) no physical model exists to estimate the loop entropies of complex intramolecular pseudoknots, and 2) their NP-complete enumeration has impeded their study. Here, we address both challenges. First, we develop a polymer physics model that can address arbitrarily complex pseudoknots using only two parameters corresponding to concrete physical quantities-over an order of magnitude fewer than the sparsest state-of-the-art phenomenological methods. Second, by coupling this model to exhaustive enumeration of the set of possible structures, we compute the entire free energy landscape of secondary structures resulting from a primary RNA sequence. We demonstrate that for RNA structures of ∼80 nucleotides, with minimal heuristics, the complete enumeration of possible secondary structures can be accomplished quickly despite the NP-complete nature of the problem. We further show that despite our loop entropy model's parametric sparsity, it performs better than or on par with previously published methods in predicting both pseudoknotted and non-pseudoknotted structures on a benchmark data set of RNA structures of ≤80 nucleotides. We suggest ways in which the accuracy of the model can be further improved.

Why Does RNA Collapse? The Importance of Water in a Simulation Study of Helix-Junction-Helix Systems

Using computer simulations, we consider the balance of thermodynamic forces that collapse RNA. A model helix-junction-helix (HJH) construct is used to investigate the transition from an extended to a collapsed conformation. Conventional Molecular Dynamics and Milestoning Simulations are used to study the free energy profile of the process for two ion concentrations. We illustrate that HJH folds to a collapsed state with two types of counterions (Mg²⁺ and K⁺). By dissecting the free energy landscape into energetic and entropic contributions, we illustrate that the electrostatic forces between the RNA and the mobile ions do not drive the RNA to a collapsed state. Instead, entropy gains from water expulsion near the neighborhood of the RNA provide the stabilization free energy that tilt HJH into more compact structures. Further simulations of a three-helix hammerhead ribozyme show a similar behavior and support the idea of collapse due to increased gain in water entropy.

Kissing loop interaction in adenine riboswitch: insights from umbrella sampling simulations

INTRODUCTION

Riboswitches are cis-acting regulatory RNA elements prevalently located in the leader sequences of bacterial mRNA. An adenine sensing riboswitch cis-regulates adeninosine deaminase gene (add) in Vibrio vulnificus. The structural mechanism regulating its conformational changes upon ligand binding mostly remains to be elucidated. In this open framework it has been suggested that the ligand stabilizes the interaction of the distal "kissing loop" complex. Using accurate full-atom molecular dynamics with explicit solvent in combination with enhanced sampling techniques and advanced analysis methods it could be possible to provide a more detailed perspective on the formation of these tertiary contacts.

METHODS

In this work, we used umbrella sampling simulations to study the thermodynamics of the kissing loop complex in the presence and in the absence of the cognate ligand. We enforced the breaking/formation of the loop-loop interaction restraining the distance between the two loops. We also assessed the convergence of the results by using two alternative initialization protocols. A structural analysis was performed using a novel approach to analyze base contacts.

RESULTS

Contacts between the two loops were progressively lost when larger inter-loop distances were enforced. Inter-loop Watson-Crick contacts survived at larger separation when compared with non-canonical pairing and stacking interactions. Intra-loop stacking contacts remained formed upon loop undocking. Our simulations qualitatively indicated that the ligand could stabilize the kissing loop complex. We also compared with previously published simulation studies.

DISCUSSION AND CONCLUSIONS

Kissing complex stabilization given by the ligand was compatible with available experimental data. However, the dependence of its value on the initialization protocol of the umbrella sampling simulations posed some questions on the quantitative interpretation of the results and called for better converged enhanced sampling simulations.

Mature miRNAs form secondary structure, which suggests their function beyond RISC

The generally accepted model of the miRNA-guided RNA down-regulation suggests that mature miRNA targets mRNA in a nucleotide sequence-specific manner. However, we have shown that the nucleotide sequence of miRNA is not the only determinant of miRNA specificity. Using specific nucleases, T1, V1 and S1 as well as NMR, UV/Vis and CD spectroscopies, we found that miR-21, miR-93 and miR-296 can adopt hairpin and/or homoduplex structures. The secondary structure of those miRNAs in solution is a function of RNA concentration and ionic conditions. Additionally, we have shown that a formation of miRNA hairpin is facilitated by cellular environment.Looking for functional consequences of this observation, we have perceived that structure of these miRNAs resemble RNA aptamers, short oligonucleotides forming a stable 3D structures with a high affinity and specificity for their targets. We compared structures of anti-tenascin C (anti-Tn-C) aptamers, which inhibit brain tumor glioblastoma multiforme (GBM, WHO IV) and selected miRNA. A strong overexpression of miR-21, miR-93 as well Tn-C in GBM may imply some connections between them. The structural similarity of these miRNA hairpins and anti-Tn-C aptamers indicates that miRNAs may function also beyond RISC and are even more sophisticated regulators, that it was previously expected. We think that the knowledge of the miRNA structure may give a new insight into miRNA-dependent gene regulation mechanism and be a step forward in the understanding their function and involvement in cancerogenesis. This may improve design process of anti-miRNA therapeutics.

Detection of a novel intergenogroup recombinant Norovirus from Kolkata, India

Mutation and recombination are recognized as important driving forces of evolution among RNA viruses. An intergenogroup recombinant norovirus strain [Hu/Kol/NLV/L8775/AB290150/2006/India] was detected in the faecal specimen of a 17 year old male, who had suffered from acute watery diarrhea and severe dehydration. Sequence analysis confirmed that this novel recombinant strain had a polymerase gene fragment that closely resembled a Norovirus (NoV) genogroup-I genotype-3 virus (HuCV/NLV/GI.3/VA98115/AY038598/1998/USA) and a capsid gene resembling NoV genogroup-II genotype-4 virus (NoV/Hu/GII.4/Terneuzen70/EF126964/2006/NL). The crossing over and recombination was observed at nucleotide (nt) 790 of NoV GI VA98115 strain and nt808 of NoV GII Terneuzen70 strain. In both parent strains conserved nucleotide sequence and hairpin structure (DNA secondary structure) were reported at the junction point of ORF1 and ORF2, exhibiting the mechanism of recombination in these viruses. Thus this novel recombinant NoV is another step in evolution among NoVs, indicating that constant surveillance is important to successfully monitor emergence of these strains.

Modifications of aminoglycoside antibiotics targeting RNA

The increased awareness of the central role of RNA has led to realization that RNA, as structural and functional information accumulation, is also drug target to small molecular therapy. Aminoglycosides are a group of well-known antibiotics, which function through binding to specific sites in prokaryotic ribosomal RNA (rRNA) and affecting the fidelity of protein synthesis. Unfortunately, their clinical practice has been curtailed by toxicity and rapid increasing number of resistant strains. Therefore, it is highly desirable to design new modified aminoglycosides that will overcome the undesirable properties of natural occurring aminoglycosides. On the other hand, aminoglycosides as potential antiviral (HIV) agents were also reported. Herein, we survey the current efforts to develop new aminoglycoside derivatives with modification and reconstruction on each sugar ring and review the latest advances in structure-activity relationships (SAR).

Mutational analysis of the influenza virus cRNA promoter and identification of nucleotides critical for replication

Replication of the influenza A virus virion RNA (vRNA) requires the synthesis of full-length cRNA, which in turn is used as a template for the synthesis of more vRNA. A "corkscrew" secondary-structure model of the cRNA promoter has been proposed recently. However the data in support of that model were indirect, since they were derived from measurement, by use of a chloramphenicol acetyltransferase (CAT) reporter in 293T cells, of mRNA levels from a modified cRNA promoter rather than the authentic cRNA promoter found in influenza A viruses. Here we measured steady-state cRNA and vRNA levels from a CAT reporter in 293T cells, directly measuring the replication of the authentic influenza A virus wild-type cRNA promoter. We found that (i) base pairing between the 5' and 3' ends and (ii) base pairing in the stems of both the 5' and 3' hairpin loops of the cRNA promoter were required for in vivo replication. Moreover, nucleotides in the tetraloop at positions 4, 5, and 7 and nucleotides forming the 2-9 base pair of the 3' hairpin loop were crucial for promoter activity in vivo. However, the 3' hairpin loop was not required for polymerase binding in vitro. Overall, our results suggest that the corkscrew secondary-structure model is required for authentic cRNA promoter activity in vivo, although the precise role of the 3' hairpin loop remains unknown.

Analysis of the cloverleaf element in a human rhinovirus type 14/poliovirus chimera: correlation of subdomain D structure, ternary protein complex formation and virus replication

RNA genomes of enteroviruses and rhinoviruses contain a 5'-terminal structure, the cloverleaf (CL), which serves as signal in RNA synthesis. Substitution of the poliovirus [PV1(M)] CL with that of human rhinovirus type 2 (HRV2) was shown previously to produce a viable chimeric PV, whereas substitution with the HRV14 CL produced a null phenotype. Fittingly, the HRV14 CL failed to form a complex with PV-specific proteins 3CD(pro)-3AB or 3CD(pro)-PCBP2, considered essential for RNA synthesis. It was reported previously (Rohll et al., J Virol 68, 4384-4391, 1994) that the major determinant for the null phenotype of a PV/HRV14 chimera resides in subdomain Id of the HRV14 CL. Using a chimeric PV/HRV14 CL in the context of the PV genome, stem-loop Id of HRV14 CL was genetically dissected. It contains the sequence C(57)UAU(60)-G, the underlined nucleotides forming the loop that is shorter by 1 nt when compared to the corresponding PV structure (UUGC(60)GG). Insertion of a G nucleotide to form a tetra loop (C(57)UAU(60)GG(61)) did not rescue replication of the chimera. However, an additional mutation at position 60 (C(57)UAC(60)GG(61)) yielded a replicating genome. Only the mutant PV/HRV14 CL with the UAC(60)G tetra loop formed ternary complexes efficiently with either PV proteins 3CD(pro)-3AB or 3CD(pro)-PCBP2. Thus, in the context of PV RNA synthesis, the presence of a tetra loop in subdomain D of the CL per se is not sufficient for function. The sequence and, consequently, the structure of the tetra loop plays an essential role. Biochemical assays demonstrated that the function of the CL element and the function of the cis-acting replication element in the 3D(pol)-3CD(pro)-dependent uridylylation of VPg are not linked.

Fitness of a turnip crinkle virus satellite RNA correlates with a sequence-nonspecific hairpin and flanking sequences that enhance replication and repress the accumulation of virions

satC, a satellite RNA associated with Turnip crinkle virus (TCV), enhances the ability of the virus to colonize plants by interfering with stable virion accumulation (F. Zhang and A. E. Simon, unpublished data). Previous results suggested that the motif1-hairpin (M1H), a replication enhancer on minus strands, forms a plus-strand hairpin flanked by CA-rich sequence that may be involved in enhancing systemic infection (G. Zhang and A. E. Simon, J. Mol. Biol. 326:35-48, 2003). In this study, sequence and structural requirements of the M1H were further assayed by replacing the 28-base M1H with 10 random bases and then subjecting the pool of satellite RNA to functional selection in plants. Unlike previous results with 28-base replacement sequences (G. Zhang and A. E. Simon, J. Mol. Biol. 326:35-48, 2003), only a few of the 10-base SELEX (systematic evolution of ligands by exponential enrichment) assay winners contained short motifs in their minus-sense orientation that were similar to TCV replication elements. However, all second- and third-round winning replacement sequences folded into hairpins flanked by CA-rich sequence predicted to be more stable on plus strands than minus strands. Plus strands of several of the most fit satellite RNAs contained insertions of CA-rich sequence at the base of their hairpins whose presence correlated with enhanced replication and reduced detection of virions. Deletion of the M1H resulted in no detectable virions despite very low satellite accumulation. These results support the hypothesis that a sequence-nonspecific plus-strand hairpin brings together flanking CA-rich sequences in the M1H region that confers fitness to satC by reducing the accumulation of stable virions.

Stem-loop III in the 5' untranslated region is a cis-acting element in bovine coronavirus defective interfering RNA replication

Higher-order structures in the 5' untranslated region (UTR) of plus-strand RNA viruses are known in many cases to function as cis-acting elements in RNA translation, replication, or transcription. Here we describe evidence supporting the structure and a cis-acting function in defective interfering (DI) RNA replication of stem-loop III, the third of four predicted higher-order structures mapping within the 210-nucleotide (nt) 5' UTR of the 32-kb bovine coronavirus (BCoV) genome. Stem-loop III maps at nt 97 through 116, has a calculated free energy of -9.1 kcal/mol in the positive strand and -3.0 kcal/mol in the negative strand, and has associated with it beginning at nt 100 an open reading frame (ORF) potentially encoding an 8-amino-acid peptide. Stem-loop III is presumed to function in the positive strand, but its strand of action has not been established. Stem-loop III (i) shows phylogenetic conservation among group 2 coronaviruses and appears to have a homolog in coronavirus groups 1 and 3, (ii) has in all coronaviruses for which sequence is known a closely associated short, AUG-initiated intra-5' UTR ORF, (iii) is supported by enzyme structure-probing evidence in BCoV RNA, (iv) must maintain stem integrity for DI RNA replication in BCoV DI RNA, and (v) shows a positive correlation between maintenance of the short ORF and maximal DI RNA accumulation in BCoV DI RNA. These results indicate that stem-loop III in the BCoV 5' UTR is a cis-acting element for DI RNA replication and that its associated intra-5' UTR ORF may function to enhance replication. It is postulated that these two elements function similarly in the virus genome.

Rearrangement of substrate secondary structure facilitates binding to the Neurospora VS ribozyme

The Neurospora VS ribozyme differs from other small, naturally occurring ribozymes in that it recognizes for trans cleavage or ligation a substrate that consists largely of a stem-loop structure. We have previously found that cleavage or ligation by the VS ribozyme requires substantial rearrangement of the secondary structure of stem-loop I, which contains the cleavage/ligation site. This rearrangement includes breaking the top base-pair of stem-loop I, allowing formation of a kissing interaction with loop V, and changing the partners of at least three other base-pairs within stem-loop I to adopt a conformation termed shifted. In the work presented, we have designed a binding assay and used mutational analysis to investigate the contribution of each of these structural changes to binding and ligation. We find that the loop I-V kissing interaction is necessary but not sufficient for binding and ligation. Constitutive opening of the top base-pair of stem-loop I has little, if any, effect on either activity. In contrast, the ability to adopt the shifted conformation of stem-loop I is a major determinant of binding: mutants that cannot adopt this conformation bind much more weakly than wild-type and mutants with a constitutively shifted stem-loop I bind much more strongly. These results implicate the adoption of the shifted structure of stem-loop I as an important process at the binding step in the VS ribozyme reaction pathway. Further investigation of features near the cleavage/ligation site revealed that sulphur substitution of the non-bridging phosphate oxygen atoms immediately downstream of the cleavage/ligation site, implicated in a putative metal ion binding site, significantly altered the cleavage/ligation equilibrium but did not perturb substrate binding significantly. This indicates that the substituted oxygen atoms, or an associated metal ion, affect a step that occurs after binding and that they influence the rates of cleavage and ligation differently.

Intramolecular secondary structure rearrangement by the kissing interaction of the Neurospora VS ribozyme

Kissing interactions in RNA are formed when bases between two hairpin loops pair. Intra- and intermolecular kissing interactions are important in forming the tertiary or quaternary structure of many RNAs. Self-cleavage of the wild-type Varkud satellite (VS) ribozyme requires a kissing interaction between the hairpin loops of stem-loops I and V. In addition, self-cleavage requires a rearrangement of several base pairs at the base of stem I. We show that the kissing interaction is necessary for the secondary structure rearrangement of wild-type stem-loop I. Surprisingly, isolated stem-loop V in the absence of the rest of the ribozyme is sufficient to rearrange the secondary structure of isolated stem-loop I. In contrast to kissing interactions in other RNAs that are either confined to the loops or culminate in an extended intermolecular duplex, the VS kissing interaction causes changes in intramolecular base pairs within the target stem-loop.

Yeast three-hybrid screening of rous sarcoma virus mutants with randomly mutagenized minimal packaging signals reveals regions important for gag interactions

We previously showed that the yeast three-hybrid system provides a genetic assay of both RNA and protein components for avian retroviral RNA encapsidation. In the current study, we used this assay to precisely define cis-acting determinants involved in avian leukosis sarcoma virus packaging RNA binding to Gag protein. In vivo screening of Rous sarcoma virus mutants was performed with randomly mutated minimal packaging sequences (MPsi) made using PCR amplification after cotransformation with GagDeltaPR protein into yeast cells. Colonies with low beta-galactosidase activity were analyzed to locate mutations in MPsi sequences affecting binding to Gag proteins. This genetic assay delineated secondary structural elements that are important for efficient RNA binding, including a single-stranded small bulge containing the initiation codon for uORF3, as well as adjacent stem structures. This implies a possible tertiary structure favoring the high-affinity binding sites for Gag. In most cases, results from the three-hybrid assay were well correlated with those from the viral RNA packaging assays. The results from random mutagenesis using the rapid three-hybrid binding assay are consistent with those from site-directed mutagenesis using in vivo packaging assays.

Secondary structure analysis of a minimal avian leukosis-sarcoma virus packaging signal

We previously identified a 160-nucleotide packaging signal, MPsi, from the 5' end of the Rous sarcoma virus genome. In this study, we determine the secondary structure of MPsi by using phylogenetic analysis with computer modeling and heterologous packaging assays of point mutants. The results of the in vivo studies are in good agreement with the computer model. Additionally, the packaging studies indicate several structures which are important for efficient packaging, including a single-stranded bulge containing the initiation codon for the short open reading frame, uORF3, as well as adjacent stem structures. Finally, we show that the L3 stem-loop at the 3' end of MPsi is dispensable for packaging, thus identifying an 82-nucleotide minimal packaging signal, microPsi, composed of the O3 stem-loop.

A hairpin loop at the 5' end of influenza A virus virion RNA is required for synthesis of poly(A)+ mRNA in vitro

We present evidence, based on extensive mutagenesis, that a hairpin loop at the 5' end of influenza A virus virion RNA (vRNA) is required for the synthesis of polyadenylated mRNA from model vRNA templates in vitro. The hairpin loop, which we term the vRNA 5' hook, contains a stem of 2 bp formed by the second and third residues pairing with the ninth and eighth residues, respectively, and a 4-nucleotide loop composed of the intervening residues 4 to 7. Disruption of the base pairs of the vRNA 5' hook by introducing point mutations prevented polyadenylation, except in two mutants where a G-U base pair reformed. The polyadenylation activity of point mutants could be rescued by constructing double mutants with reformed base pairs in the stem of the vRNA 5' hook. These results suggest that base pairing rather than a particular nucleotide sequence was critical. We also show that mutation of the analogous region in the 3' arm of vRNA did not interfere with the synthesis of polyadenylated mRNA, suggesting that a hook structure in the 3' arm is not required for transcription of polyadenylated mRNA in vitro.