RNA最新文献

tRNA-derived fragments have emerged as critical regulators in various biological processes, but reliable methods for their quantification remain a challenge due to their small size and extensive RNA modifications. In this study, we present the newly developed Complementary DNA Oligonucleotide Direct In-Gel Quantification (cDINGQ) method for tRF analysis and compare it with traditional radioactive [³²P] Northern blotting, non-radioactive approaches, and high-throughput Illumina sequencing under different experimental conditions. The cDINGQ method, utilizing Cy5-labeled hybridization probes, offers high specificity and sensitivity for detecting tRFs with significantly reduced processing time and costs. By applying these techniques to an Alzheimer's disease (AD) cell model, we demonstrate the reliability of these methods in detecting subtle variations in tRF abundance. Our findings highlight the sensitivity, specificity, and applicability of each method, addressing limitations such as RNA input requirements and probe hybridization conditions. The study further explores the utility of these methods for detecting tRFs in various biological contexts, emphasizing their potential for future research and biomarker discovery in disease-related studies.

Cells under stress shift their proteome by repressing cap-dependent translation initiation. RNA elements called internal ribosome entry sites (IRES) can allow key cellular transcripts to remain efficiently translated to support an effective stress response. We previously determined that the 5' untranslated region (5'UTR) of the insulin receptor mRNA possesses a capacity for IRES activity that is conserved from insects to mammals. Well-characterized IRESes depend on RNA structures that reduce the protein requirements for translation initiation, thus circumventing translation inhibition. While there are several examples of viral IRES structures solved in vitro, the RNA secondary structures of cellular IRESes remain elusive and little information exists about the secondary structures of these RNAs in vivo. Here we probe the secondary structure of the Insr 5'UTR IRES along with two well-studied viral IRESes from hepatitis C virus and encephalomyocarditis virus using dimethyl sulfate mutational profiling by sequencing (DMS-MaPseq) in vitro and in cells. We find that the structures of viral IRESes in a cellular environment are largely consistent with their known in vitro structures. Using DMS-MaPseq probing as a constraint, we generated a model of the RNA secondary structure of the mouse insulin receptor 5'UTR. With this model as a guide, we employed a mutation strategy which allowed us to identify a conserved segment of RNA, distal from the translation start codon, that is critical for Insr IRES function. This knowledge informed the design of a minimal IRES element with equivalent activity to the full-length Insr 5'UTR across translation contexts.

Argonaute proteins mediate gene silencing via small regulatory RNAs that are generated by distinctive biogenesis pathways. In animals, three main classes are recognized: ~21-24 nucleotide (nt) microRNAs (miRNAs), ~21-24 nt small-interfering RNAs (siRNAs) and ~24-32 nt Piwi-interacting RNAs (piRNAs). Mechanistic understanding of these pathways was gained from genetic, biochemical and genomic studies in a handful of model systems, where key ribonucleolytic events were identified that specify stereotyped positioning of small RNAs relative to their precursor transcripts. With burgeoning availability of assembled genomes and small RNA data, there are abundant opportunities to characterize the diversity of small RNAs across non-model organisms. While several tools are well-suited to analyze specific small RNA pathways, an integrated package that can help classify and interpret all three major classes of small RNAs is wanting. To address this need, we developed a simple and efficient R package (MiSiPi.Rna) that can generate a variety of plots and statistics for pre-selected loci, which enable the characterization of diverse biogenesis features of miRNAs, siRNAs and piRNAs. MiSiPi.Rna requires minimal computational expertise to run, and will facilitate efforts to annotate and analyze the major classes of Argonaute-based small regulatory RNAs in arbitrary species of choice.

Many non-protein-coding RNAs have been discovered and more are being discovered each year. At first we know them only by their sequences in a few organisms, but to understand their function and interactions, we need to understand what 3D structures they may form, in whole or in part. Many hairpin and internal loops are known to form structured 3D motifs, many of which are recurrent across different non-coding RNAs, for example, kink turn and sarcin-ricin internal loops, and GNRA, UNCG, and T-loop hairpin loops. As such, a new non-protein-coding RNA may well have one or more known structured 3D loop motifs. The goal of this paper is to introduce a tool which can identify loops in Rfam seed alignments that match well to known 3D loop motifs, and which makes those identifications easily accessible. JAR3D was developed to map sequences of hairpin and internal loops to known 3D motifs, and was extended for this work to 3-way and 4-way junction motifs. We applied JAR3D to 4,166 Rfam seed alignments from Rfam 15.0 and made the results accessible on the JAR3D web page, which makes it easy to inspect and evaluate the possible matches for each loop in each Rfam family. We provide several examples which validate JAR3D's ability to identify the correct loop motif, using 3D structures of RNAs outside of the training set. We created a new page to search for instances of a particular loop motif across all Rfam families, to facilitate studies of how widespread the occurrence of each motif is. We provide statistics on how many Rfam loops appear to match well to a known 3D motif. Match rates are much higher for internal loops than for hairpins or multi-helix junctions.

Porcine deltacoronavirus (PDCoV), an emerging enteropathogenic coronavirus, primarily causes diarrhea in piglets and has the potential for cross-species transmission to humans. The recent detection of PDCoV in Haitian children underscores the urgent need for developing antiviral therapeutic strategies. G-quadruplexes (G4s) are implicated in the modulation of viral infection; however, their identification and roles in the PDCoV life cycle remain unclear. Here, we identified a highly conserved G4 structure, designated PDCoV-G4, located within the coding region of PDCoV non-structural protein 8 (nsp8). PDS and TMPyP4, two well-known G4-binding ligands, were found to target PDCoV-G4 and exhibit anti-PDCoV activity. Interestingly, PDS stabilizes the structure of PDCoV-G4, while TMPyP4 disrupts it. The recombinant PDCoV with G4-disruptive mutations (rPDCoV-nsp8mut) displays resistance to both PDS and TMPyP4. Utilizing an embryonated chicken eggs (ECEs) infection model, we observed that TMPyP4 provides superior protective effects for rPDCoV-wt-infected ECEs compared to PDS. However, both PDS and TMPyP4 exhibited diminished protective effects on chicken embryos infected with rPDCoV-nsp8mut, relative to rPDCoV-wt, further confirming their in vivo antiviral activity through targeting PDCoV-G4. These findings demonstrate that the PDCoV-G4 plays a crucial regulatory role in the PDCoV life cycle and pathogenicity, representing a potential target for antiviral therapy.

Predicting the secondary structure of RNA is a core challenge in computational biology, essential for understanding molecular function and designing novel therapeutics. The field has evolved from foundational but accuracy-limited thermodynamic approaches to a new data-driven paradigm dominated by machine learning and deep learning. These models learn folding patterns directly from data, leading to significant performance gains. This review surveys the modern landscape of these methods, covering single-sequence, evolutionary-based, and hybrid models that blend machine learning with biophysics. A central theme is the field's "generalization crisis," where powerful models were found to fail on new RNA families, prompting a community-wide shift to stricter, homology-aware benchmarking. In response to the underlying challenge of data scarcity, RNA foundation models have emerged, learning from massive, unlabeled sequence corpora to improve generalization. Finally, we look ahead to the next set of major hurdles-including the accurate prediction of complex motifs like pseudoknots, scaling to kilobase-length transcripts, incorporating the chemical diversity of modified nucleotides, and shifting the prediction target from static structures to the dynamic ensembles that better capture biological function. We also highlight the need for a standardized, prospective benchmarking system to ensure unbiased validation and accelerate progress.

Membraneless organelles (MLOs) formed through phase separation play crucial roles in various cellular processes. Many MLOs remain spatially compartmentalized, avoiding fusion or engulfment. MLOs are formed by dynamic multivalent interactions, often mediated by proteins with intrinsically disordered regions (IDRs). However, the molecular principles behind how IDRs maintain MLO independence remain poorly understood. Here, we investigated the proline/glutamine (P/Q)-rich IDR of SFPQ, a protein identified as a key factor in segregating paraspeckles from nuclear speckles. Paraspeckle segregation analyses, using SFPQ mutants tethered to NEAT1_2 long noncoding RNA, revealed that P/Q residues within the SFPQ IDR, conserved from humans to zebrafish, are crucial for its segregation activity. Beyond amino acid composition, the blocky patterns of P/Q residues, in which proline- and glutamine-rich blocks are repetitively arranged, are required for the segregation from nuclear speckles. Among human IDRs exhibiting PQ-block patterns, BRD4 IDR shows strong sequence similarity to the SFPQ IDR, and exhibits comparable segregation activity. Molecular dynamics simulation suggests that the PQ-blocky patterns required for the paraspeckle segregation do not correlate with the IDR characteristics necessary for self-assembly. Thus, these data suggest that the PQ-blocky patterns in IDRs represent a previously uncharacterized property that contributes to MLO independence, possibly through a mechanism distinct from the conventional phase separation-promoting function of IDRs.

mRNA-based therapeutics are commonly produced through T7 RNA Polymerase-mediated in vitro transcription. Introducing these exogenous RNAs into human cells activates an RNA sensor Protein Kinase R (PKR), which suppresses translation initiation and reduces their therapeutic effectiveness. Incorporating uridine analogs into these transcripts prevents PKR activation and translation shutdown, but the underlying mechanism remains unclear. Here, we demonstrate that treating T7 RNA Polymerase-produced transcripts with RNase III, which selectively degrades double-stranded RNA (dsRNA), blocks PKR activation and downstream translation-inhibition events, including eIF2α phosphorylation and stress granule formation in human cells. Interestingly, dsRNAs generated with uridine analogs robustly induce eIF2α phosphorylation and stress granules to the same extent as dsRNA containing uridine. These findings indicate that uridine analogs do not prevent PKR from detecting dsRNA. Instead, we show that uridine analogs decrease the production of T7 RNA Polymerase byproducts, including antisense RNA and dsRNA, which activate PKR and downstream stress responses. Finally, we demonstrate that higher amounts of exogenous RNA, lacking T7 RNA Polymerase byproducts, can induce stress granules independently of PKR and phospho-eIF2α, but dependent on stress granule scaffold proteins G3BP1 and G3BP2. Together, our findings show that uridine analogs mitigate PKR signaling not by blocking mRNA-PKR interactions, but by minimizing dsRNA byproducts from T7 Polymerase transcription. Furthermore, stress granule formation in response to high levels of exogenous RNA can occur through a mechanism that does not depend on PKR but relies on G3BP1 and G3BP2. These insights clarify the role of uridine analogs in PKR activation and may inform future therapeutic RNA design.

Nicotinamide adenine dinucleotide (NAD) is a ubiquitous enzyme cofactor that serves as a carrier of hydride ions for metabolic oxidation-reduction reactions. NAD is also sometimes used as a source of activated adenosine monophosphate (AMP) for adenylation reactions or as a precursor of ADP-ribose upon removal of nicotinamide. Many bacterial riboswitch classes are known to sense nucleotide-derived enzyme cofactors, but NAD is one of several ancient cofactors that have few or no known riboswitch representatives. Two rare riboswitch classes, named NAD⁺-I and NAD⁺-II, have been reported that regulate genes relevant to NAD biosynthesis and transport. However, these RNAs exhibit unusual functional and structural properties. Here we report that miniature NAD⁺-II riboswitches, named mini-NAD⁺-II, are more abundant and widespread than the longer RNAs that were used to define the original consensus model for this class. The newfound examples are commonly found within lactic acid bacteria, which are notable for varied metabolic fermentation strategies used to maintain sufficient NAD⁺ Furthermore, the simple H-type pseudoknot core of mini-NAD⁺-II aptamers is similar to that of class I preQ₁ riboswitch (preQ₁-I) aptamers. Thus, H-type pseudoknots might serve as a versatile architecture for the natural or synthetic construction of ligand-binding aptamers.

Although protein-RNA interactions are crucial for many biological processes, predicting their binding free energies (ΔG) is a challenging task due to limited available experimental data and the complexity of these interactions. To address this issue, we developed a machine learning-based model designed to predict energy-based scores for protein-RNA complexes, called PANTHER Score. By applying a local-to-global approach, we proposed a methodology further subdivided into five steps: (1) We derived 87,117 pairwise local interaction energies from 331,744 MD-derived interactions across 46 curated protein-RNA complexes; (2) we trained ML models on pairwise interaction features to predict local interaction energies without performing MD simulations; (3) we integrated predicted local interaction energies using a local-to-global methodology, to compute model-specific PANTHER Score; (4) we evaluate model-specific PANTHER Score on an independent test set of seven complexes; and (5) we validated and selected the optimal model using an external stress set of 110 complexes with experimental ΔG values for implementation in the PANTHER Scoring pipeline. Among the regression models developed, Random Forest Regression exhibited the highest predictive performance as a model-specific PANTHER Score, achieveing a Pearson correlation (r) of 0.80 and MAE of 1.79 kcal/mol on the test set. It maintained strong predictive capabilities on the stress set (r = 0.64, MAE = 1.63 kcal/mol). Benchmarking against existing tools on the stress test set, the PANTHER Score demonstrated superior accuracy and reliability. This study highlights the effectiveness of MD and machine learning in addressing data limitations through innovative strategies, positioning the PANTHER Score as a robust tool for predicting protein-RNA binding affinities in biomolecular research, drug discovery and mainly in RNA-therapeutics.