Elucidation of the aberrant 3' splice site selection by cancer-associated mutations on the U2AF1

Activation of ERβ hijacks the splicing machinery to trigger R-loop formation in triple-negative breast cancer

Triple-negative breast cancer (TNBC) is a subtype of breast cancer with aggressive behavior and poor prognosis. Current therapeutic options available for TNBC patients are primarily chemotherapy. With our evolving understanding of this disease, novel targeted therapies, including poly ADP-ribose polymerase (PARP) inhibitors, antibody-drug conjugates, and immune-checkpoint inhibitors, have been developed for clinical use. Previous reports have demonstrated the essential role of estrogen receptor β (ERβ) in TNBC, but the detailed molecular mechanisms downstream ERβ activation in TNBC are still far from elucidated. In this study, we demonstrated that a specific ERβ agonist, LY500307, potently induces R-loop formation and DNA damage in TNBC cells. Subsequent interactome experiments indicated that the residues 151 to 165 of U2 small nuclear RNA auxiliary factor 1 (U2AF1) and the Trp439 and Lys443 of ERβ were critical for the binding between U2AF1 and ERβ. Combined RNA sequencing and ribosome sequencing analysis demonstrated that U2AF1-regulated downstream RNA splicing of 5-oxoprolinase (OPLAH) could affect its enzymatic activity and is essential for ERβ-induced R-loop formation and DNA damage. In clinical samples including 115 patients from The Cancer Genome Atlas (TCGA) and 32 patients from an in-house cohort, we found a close correlation in the expression of ESR2 and U2AF1 in TNBC patients. Collectively, our study has unraveled the molecular mechanisms that explain the therapeutic effects of ERβ activation in TNBC, which provides rationale for ERβ activation-based single or combined therapy for patients with TNBC.

Exonic splicing code and coordination of divalent metals in proteins

Exonic sequences contain both protein-coding and RNA splicing information but the interplay of the protein and splicing code is complex and poorly understood. Here, we have studied traditional and auxiliary splicing codes of human exons that encode residues coordinating two essential divalent metals at the opposite ends of the Irving-Williams series, a universal order of relative stabilities of metal-organic complexes. We show that exons encoding Zn2+-coordinating amino acids are supported much less by the auxiliary splicing motifs than exons coordinating Ca2+. The handicap of the former is compensated by stronger splice sites and uridine-richer polypyrimidine tracts, except for position -3 relative to 3' splice junctions. However, both Ca2+ and Zn2+ exons exhibit close-to-constitutive splicing in multiple tissues, consistent with their critical importance for metalloprotein function and a relatively small fraction of expendable, alternatively spliced exons. These results indicate that constraints imposed by metal coordination spheres on RNA splicing have been efficiently overcome by the plasticity of exon-intron architecture to ensure adequate metalloprotein expression.

Expanding PROTACtable genome universe of E3 ligases

Proteolysis-targeting chimera (PROTAC) and other targeted protein degradation (TPD) molecules that induce degradation by the ubiquitin-proteasome system (UPS) offer new opportunities to engage targets that remain challenging to be inhibited by conventional small molecules. One fundamental element in the degradation process is the E3 ligase. However, less than 2% amongst hundreds of E3 ligases in the human genome have been engaged in current studies in the TPD field, calling for the recruiting of additional ones to further enhance the therapeutic potential of TPD. To accelerate the development of PROTACs utilizing under-explored E3 ligases, we systematically characterize E3 ligases from seven different aspects, including chemical ligandability, expression patterns, protein-protein interactions (PPI), structure availability, functional essentiality, cellular location, and PPI interface by analyzing 30 large-scale data sets. Our analysis uncovers several E3 ligases as promising extant PROTACs. In total, combining confidence score, ligandability, expression pattern, and PPI, we identified 76 E3 ligases as PROTAC-interacting candidates. We develop a user-friendly and flexible web portal ( https://hanlaboratory.com/E3Atlas/ ) aimed at assisting researchers to rapidly identify E3 ligases with promising TPD activities against specifically desired targets, facilitating the development of these therapies in cancer and beyond.

Promoter R-Loops Recruit U2AF1 to Modulate Its Phase Separation and RNA Splicing

R-loops and guanine quadruplexes (G4s) are secondary structures of nucleic acids that are ubiquitously present in cells and are enriched in promoter regions of genes. By employing a bioinformatic approach based on overlap analysis of transcription factor chromatin immunoprecipitation sequencing (ChIP-seq) data sets, we found that many splicing factors, including U2AF1 whose recognition of the 3' splicing site is crucial for pre-mRNA splicing, exhibit pronounced enrichment at endogenous R-loop- and DNA G4-structure loci in promoter regions of human genes. We also revealed that U2AF1 binds directly to R-loops and DNA G4 structures at a low-nM binding affinity. Additionally, we showed the ability of U2AF1 to undergo phase separation, which could be stimulated by binding with R-loops, but not duplex DNA, RNA/DNA hybrid, DNA G4, or single-stranded RNA. We also demonstrated that U2AF1 binds to promoter R-loops in human cells, and this binding competes with U2AF1's interaction with 3' splicing site and leads to augmented distribution of RNA polymerase II (RNAPII) to promoters over gene bodies, thereby modulating cotranscriptional pre-mRNA splicing. Together, we uncovered a group of candidate proteins that can bind to both R-loops and DNA G4s, revealed the direct and strong interactions of U2AF1 with these nucleic acid structures, and established a biochemical rationale for U2AF1's occupancy in gene promoters. We also unveiled that interaction with R-loops promotes U2AF1's phase separation, and our work suggests that U2AF1 modulates pre-mRNA splicing by regulating RNAPII's partition in transcription initiation versus elongation.

FexSplice: A LightGBM-Based Model for Predicting the Splicing Effect of a Single Nucleotide Variant Affecting the First Nucleotide G of an Exon

Single nucleotide variants (SNVs) affecting the first nucleotide G of an exon (Fex-SNVs) identified in various diseases are mostly recognized as missense or nonsense variants. Their effect on pre-mRNA splicing has been seldom analyzed, and no curated database is available. We previously reported that Fex-SNVs affect splicing when the length of the polypyrimidine tract is short or degenerate. However, we cannot readily predict the splicing effects of Fex-SNVs. We here scrutinized the available literature and identified 106 splicing-affecting Fex-SNVs based on experimental evidence. We similarly identified 106 neutral Fex-SNVs in the dbSNP database with a global minor allele frequency (MAF) of more than 0.01 and less than 0.50. We extracted 115 features representing the strength of splicing cis-elements and developed machine-learning models with support vector machine, random forest, and gradient boosting to discriminate splicing-affecting and neutral Fex-SNVs. Gradient boosting-based LightGBM outperformed the other two models, and the length and nucleotide compositions of the polypyrimidine tract played critical roles in the discrimination. Recursive feature elimination showed that the LightGBM model using 15 features achieved the best performance with an accuracy of 0.80 ± 0.12 (mean and SD), a Matthews Correlation Coefficient (MCC) of 0.57 ± 0.15, an area under the curve of the receiver operating characteristics curve (AUROC) of 0.86 ± 0.08, and an area under the curve of the precision-recall curve (AUPRC) of 0.87 ± 0.09 using a 10-fold cross-validation. We developed a web service program, named FexSplice that accepts a genomic coordinate either on GRCh37/hg19 or GRCh38/hg38 and returns a predicted probability of aberrant splicing of A, C, and T variants.

Role of innate immunological/inflammatory pathways in myelodysplastic syndromes and AML: a narrative review

Dysregulation of the innate immune system and inflammatory-related pathways has been implicated in hematopoietic defects in the bone marrow microenvironment and associated with aging, clonal hematopoiesis, myelodysplastic syndromes (MDS), and acute myeloid leukemia (AML). As the innate immune system and its pathway regulators have been implicated in the pathogenesis of MDS/AML, novel approaches targeting these pathways have shown promising results. Variability in expression of Toll like receptors (TLRs), abnormal levels of MyD88 and subsequent activation of NF-κβ, dysregulated IL1-receptor associated kinases (IRAK), alterations in TGF-β and SMAD signaling, high levels of S100A8/A9 have all been implicated in pathogenesis of MDS/AML. In this review we not only discuss the interplay of various innate immune pathways in MDS pathogenesis but also focus on potential therapeutic targets from recent clinical trials including the use of monoclonal antibodies and small molecule inhibitors against these pathways.

Structure of the Caenorhabditis elegans m6A methyltransferase METT10 that regulates SAM homeostasis

In Caenorhabditis elegans, the N6-methyladenosine (m6A) modification by METT10, at the 3'-splice sites in S-adenosyl-l-methionine (SAM) synthetase (sams) precursor mRNA (pre-mRNA), inhibits sams pre-mRNA splicing, promotes alternative splicing coupled with nonsense-mediated decay of the pre-mRNAs, and thereby maintains the cellular SAM level. Here, we present structural and functional analyses of C. elegans METT10. The structure of the N-terminal methyltransferase domain of METT10 is homologous to that of human METTL16, which installs the m6A modification in the 3'-UTR hairpins of methionine adenosyltransferase (MAT2A) pre-mRNA and regulates the MAT2A pre-mRNA splicing/stability and SAM homeostasis. Our biochemical analysis suggested that C. elegans METT10 recognizes the specific structural features of RNA surrounding the 3'-splice sites of sams pre-mRNAs, and shares a similar substrate RNA recognition mechanism with human METTL16. C. elegans METT10 also possesses a previously unrecognized functional C-terminal RNA-binding domain, kinase associated 1 (KA-1), which corresponds to the vertebrate-conserved region (VCR) of human METTL16. As in human METTL16, the KA-1 domain of C. elegans METT10 facilitates the m6A modification of the 3'-splice sites of sams pre-mRNAs. These results suggest the well-conserved mechanisms for the m6A modification of substrate RNAs between Homo sapiens and C. elegans, despite their different regulation mechanisms for SAM homeostasis.

The Role of the m6A RNA Methyltransferase METTL16 in Gene Expression and SAM Homeostasis

The RNA methylation of adenosine at the N6-position (m⁶A) has attracted significant attention because of its abundance and dynamic nature. It accounts for more than 80% of all RNA modifications present in bacteria and eukaryotes and regulates crucial aspects of RNA biology and gene expression in numerous biological processes. The majority of m⁶A found in mammals is deposited by a multicomponent complex formed between methyltransferase-like (METTL) proteins METTL3 and METTL14. In the last few years, the list of m⁶A writers has grown, resulting in an expansion of our understanding of the importance of m⁶A and the methylation machinery. The characterization of the less familiar family member METTL16 has uncovered a new function of the m⁶A methylation apparatus, namely the fine-tuning of the cellular levels of the major methyl donor S-adenosylmethionine (SAM). METTL16 achieves this by adjusting the levels of the enzyme that synthesizes SAM in direct response to fluctuations in the SAM availability. This review summarizes recent progress made in understanding how METTL16 can sense and relay metabolic information and considers the wider implications. A brief survey highlights similarities and differences between METTL16 and the better-known METTL3/14 complex, followed by a discussion of the target specificity, modes of action and potential roles of METTL16.

Structural basis for the interaction between the first SURP domain of the SF3A1 subunit in U2 snRNP and the human splicing factor SF1

SURP domains are exclusively found in splicing-related proteins in all eukaryotes. SF3A1, a component of the U2 snRNP, has two tandem SURP domains, SURP1, and SURP2. SURP2 is permanently associated with a specific short region of SF3A3 within the SF3A protein complex whereas, SURP1 binds to the splicing factor SF1 for recruitment of U2 snRNP to the early spliceosomal complex, from which SF1 is dissociated during complex conversion. Here, we determined the solution structure of the complex of SURP1 and the human SF1 fragment using nuclear magnetic resonance (NMR) methods. SURP1 adopts the canonical topology of α1-α2-3₁₀ -α3, in which α1 and α2 are connected by a single glycine residue in a particular backbone conformation, allowing the two α-helices to be fixed at an acute angle. A hydrophobic patch, which is part of the characteristic surface formed by α1 and α2, specifically contacts a hydrophobic cluster on a 16-residue α-helix of the SF1 fragment. Furthermore, whereas only hydrophobic interactions occurred between SURP2 and the SF3A3 fragment, several salt bridges and hydrogen bonds were found between the residues of SURP1 and the SF1 fragment. This finding was confirmed through mutational studies using bio-layer interferometry. The study also revealed that the dissociation constant between SURP1 and the SF1 fragment peptide was approximately 20 μM, indicating a weak or transient interaction. Collectively, these results indicate that the interplay between U2 snRNP and SF1 involves a transient interaction of SURP1, and this transient interaction appears to be common to most SURP domains, except for SURP2.

Role of computational and structural biology in the development of small-molecule modulators of the spliceosome

INTRODUCTION

RNA splicing is a pivotal step of eukaryotic gene expression during which the introns are excised from the precursor (pre-)RNA and the exons are joined together to form mature RNA products (i.e a protein-coding mRNA or long non-coding (lnc)RNAs). The spliceosome, a complex ribonucleoprotein machine, performs pre-RNA splicing with extreme precision. Deregulated splicing is linked to cancer, genetic, and neurodegenerative diseases. Hence, the discovery of small-molecules targeting core spliceosome components represents an appealing therapeutic opportunity.

AREA COVERED

Several atomic-level structures of the spliceosome and distinct splicing-modulators bound to its protein/RNA components have been solved. Here, we review recent advances in the discovery of small-molecule splicing-modulators, discuss opportunities and challenges for their therapeutic applicability, and showcase how structural data and/or all-atom simulations can illuminate key facets of their mechanism, thus contributing to future drug-discovery campaigns.

EXPERT OPINION

This review highlights the potential of modulating pre-RNA splicing with small-molecules, and anticipates how the synergy of computer and wet-lab experiments will enrich our understanding of splicing regulation/deregulation mechanisms. This information will aid future structure-based drug-discovery efforts aimed to expand the currently limited portfolio of selective splicing-modulators.

The Biological and Clinical Consequences of RNA Splicing Factor U2AF1 Mutation in Myeloid Malignancies

Simple Summary

U2 small nuclear RNA auxiliary factor 1 (U2AF1) is one of the most important RNA splicing genes involved in regulating the alternative splicing of pre-mRNA. U2AF1 mutation is a genetic driver event in the initiation of myelodysplastic syndromes (MDSs) and frequently occurs in myeloid malignancies. U2AF1 mutation can severely impair hematopoiesis, drive tumor progression, adversely affect disease prognosis, and promote leukemic transformation. This review summarizes the biological and clinical implications of the oncogenic role of U2AF1 mutation in myeloid tumors. Our work provides important and comprehensive insights into the development of the U2AF1 mutation as a novel prognostic biomarker and therapeutic target for myeloid malignancies.

Abstract

Mutations of spliceosome genes have been frequently identified in myeloid malignancies with the large-scale application of advanced sequencing technology. U2 small nuclear RNA auxiliary factor 1 (U2AF1), an essential component of U2AF heterodimer, plays a pivotal role in the pre-mRNA splicing processes to generate functional mRNAs. Over the past few decades, the mutation landscape of U2AF1 (most frequently involved S34 and Q157 hotspots) has been drawn in multiple cancers, particularly in myeloid malignancies. As a recognized early driver of myelodysplastic syndromes (MDSs), U2AF1 mutates most frequently in MDS, followed by acute myeloid leukemia (AML) and myeloproliferative neoplasms (MPNs). Here, for the first time, we summarize the research progress of U2AF1 mutations in myeloid malignancies, including the correlations between U2AF1 mutations with clinical and genetic characteristics, prognosis, and the leukemic transformation of patients. We also summarize the adverse effects of U2AF1 mutations on hematopoietic function, and the alterations in downstream alternative gene splicing and biological pathways, thus providing comprehensive insights into the roles of U2AF1 mutations in the myeloid malignancy pathogenesis. U2AF1 mutations are expected to be potential novel molecular markers for myeloid malignancies, especially for risk stratification, prognosis assessment, and a therapeutic target of MDS patients.

Molecular Threat of Splicing Factor Mutations to Myeloid Malignancies and Potential Therapeutic Modulations

Splicing factors are frequently mutated in myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). These mutations are presumed to contribute to oncogenic transformation, but the underlying mechanisms remain incompletely understood. While no specific treatment option is available for MDS/AML patients with spliceosome mutations, novel targeting strategies are actively explored, leading to clinical trials of small molecule inhibitors that target the spliceosome, DNA damage response pathway, and immune response pathway. Here, we review recent progress in mechanistic understanding of splicing factor mutations promoting disease progression and summarize potential therapeutic strategies, which, if successful, would provide clinical benefit to patients carrying splicing factor mutations.

Multiomics analysis of the NAD+-PARP1 axis reveals a role for site-specific ADP-ribosylation in splicing in embryonic stem cells

The differentiation of embryonic stem cells (ESCs) into a lineage-committed state is a dynamic process involving changes in cellular metabolism, epigenetic modifications, post-translational modifications, gene expression, and RNA processing. Here we integrated data from metabolomic, proteomic, and transcriptomic assays to characterize how alterations in NAD⁺ metabolism during the differentiation of mouse ESCs lead to alteration of the PARP1-mediated ADP-ribosylated (ADPRylated) proteome and mRNA isoform specialization. Our metabolomic analyses indicate that mESCs use distinct NAD⁺ biosynthetic pathways in different cell states: the de novo pathway in the pluripotent state, and the salvage and Preiss-Handler pathways as differentiation progresses. We observed a dramatic induction of PARP1 catalytic activity driven by enhanced nuclear NAD⁺ biosynthesis during the early stages of mESC differentiation (e.g., within 12 h of LIF removal). PARP1-modified proteins in mESCs are enriched for biological processes related to stem cell maintenance, transcriptional regulation, and RNA processing. The PARP1 substrates include core spliceosome components, such as U2AF35 and U2AF65, whose splicing functions are modulated by PARP1-mediated site-specific ADP-ribosylation. Finally, we observed that splicing is dysregulated genome-wide in Parp1 knockout mESCs. Together, these results demonstrate a role for the NAD⁺-PARP1 axis in the maintenance of mESC state, specifically in the splicing program during differentiation.

Predicting RNA splicing from DNA sequence using Pangolin

Recent progress in deep learning has greatly improved the prediction of RNA splicing from DNA sequence. Here, we present Pangolin, a deep learning model to predict splice site strength in multiple tissues. Pangolin outperforms state-of-the-art methods for predicting RNA splicing on a variety of prediction tasks. Pangolin improves prediction of the impact of genetic variants on RNA splicing, including common, rare, and lineage-specific genetic variation. In addition, Pangolin identifies loss-of-function mutations with high accuracy and recall, particularly for mutations that are not missense or nonsense, demonstrating remarkable potential for identifying pathogenic variants.

Precision analysis of mutant U2AF1 activity reveals deployment of stress granules in myeloid malignancies

Splicing factor mutations are common among cancers, recently emerging as drivers of myeloid malignancies. U2AF1 carries hotspot mutations in its RNA-binding motifs; however, how they affect splicing and promote cancer remain unclear. The U2AF1/U2AF2 heterodimer is critical for 3' splice site (3'SS) definition. To specifically unmask changes in U2AF1 function in vivo, we developed a crosslinking and immunoprecipitation procedure that detects contacts between U2AF1 and the 3'SS AG at single-nucleotide resolution. Our data reveal that the U2AF1 S34F and Q157R mutants establish new 3'SS contacts at -3 and +1 nucleotides, respectively. These effects compromise U2AF2-RNA interactions, resulting predominantly in intron retention and exon exclusion. Integrating RNA binding, splicing, and turnover data, we predicted that U2AF1 mutations directly affect stress granule components, which was corroborated by single-cell RNA-seq. Remarkably, U2AF1-mutant cell lines and patient-derived MDS/AML blasts displayed a heightened stress granule response, pointing to a novel role for biomolecular condensates in adaptive oncogenic strategies.

The XRN1-regulated RNA helicase activity of YTHDC2 ensures mouse fertility independently of m6A recognition

The functional consequence of N⁶-methyladenosine (m⁶A) RNA modification is mediated by "reader" proteins of the YTH family. YTH domain-containing 2 (YTHDC2) is essential for mammalian fertility, but its molecular function is poorly understood. Here, we identify U-rich motifs as binding sites of YTHDC2 on 3' UTRs of mouse testicular RNA targets. Although its YTH domain is an m⁶A-binder in vitro, the YTH point mutant mice are fertile. Significantly, the loss of its 3'→5' RNA helicase activity causes mouse infertility, with the catalytic-dead mutation being dominant negative. Biochemical studies reveal that the weak helicase activity of YTHDC2 is enhanced by its interaction with the 5'→3' exoribonuclease XRN1. Single-cell transcriptomics indicate that Ythdc2 mutant mitotic germ cells transition into meiosis but accumulate a transcriptome with mixed mitotic/meiotic identity that fail to progress further into meiosis. Finally, our demonstration that ythdc2 mutant zebrafish are infertile highlights its conserved role in animal germ cell development.

Recurrent Spliceosome Mutations in Cancer: Mechanisms and Consequences of Aberrant Splice Site Selection

Splicing alterations have been widely documented in tumors where the proliferation and dissemination of cancer cells is supported by the expression of aberrant isoform variants. Splicing is catalyzed by the spliceosome, a ribonucleoprotein complex that orchestrates the complex process of intron removal and exon ligation. In recent years, recurrent hotspot mutations in the spliceosome components U1 snRNA, SF3B1, and U2AF1 have been identified across different tumor types. Such mutations in principle are highly detrimental for cells as all three spliceosome components are crucial for accurate splice site selection: the U1 snRNA is essential for 5′ splice site recognition, and SF3B1 and U2AF1 are important for 3′ splice site selection. Nonetheless, they appear to be selected to promote specific types of cancers. Here, we review the current molecular understanding of these mutations in cancer, focusing on how they influence splice site selection and impact on cancer development.

Transcription Factors, R-Loops and Deubiquitinating Enzymes: Emerging Targets in Myelodysplastic Syndromes and Acute Myeloid Leukemia

Simple Summary

The advent of DNA massive sequencing technologies has allowed for the first time an extensive look into the heterogeneous spectrum of genes and mutations underpinning myelodysplastic syndromes (MDSs) and acute myeloid leukemia (AML). In this review, we wish to explore the most recent advances and the rationale for the potential therapeutic interest of three main actors in myelo-leukemic transformation: transcription factors that govern myeloid differentiation; RNA splicing factors, which ensure proper mRNA maturation and whose mutations increase R-loops formation; and deubiquitinating enzymes, which contribute to genome stability in hematopoietic stem cells (HSCs).

Abstract

Myeloid neoplasms encompass a very heterogeneous family of diseases characterized by the failure of the molecular mechanisms that ensure a balanced equilibrium between hematopoietic stem cells (HSCs) self-renewal and the proper production of differentiated cells. The origin of the driver mutations leading to preleukemia can be traced back to HSC/progenitor cells. Many properties typical to normal HSCs are exploited by leukemic stem cells (LSCs) to their advantage, leading to the emergence of a clonal population that can eventually progress to leukemia with variable latency and evolution. In fact, different subclones might in turn develop from the original malignant clone through accumulation of additional mutations, increasing their competitive fitness. This process ultimately leads to a complex cancer architecture where a mosaic of cellular clones—each carrying a unique set of mutations—coexists. The repertoire of genes whose mutations contribute to the progression toward leukemogenesis is broad. It encompasses genes involved in different cellular processes, including transcriptional regulation, epigenetics (DNA and histones modifications), DNA damage signaling and repair, chromosome segregation and replication (cohesin complex), RNA splicing, and signal transduction. Among these many players, transcription factors, RNA splicing proteins, and deubiquitinating enzymes are emerging as potential targets for therapeutic intervention.

Mechanistic Insights of Aberrant Splicing with Splicing Factor Mutations Found in Myelodysplastic Syndromes

Pre-mRNA splicing is an essential process for gene expression in higher eukaryotes, which requires a high order of accuracy. Mutations in splicing factors or regulatory elements in pre-mRNAs often result in many human diseases. Myelodysplastic syndrome (MDS) is a heterogeneous group of chronic myeloid neoplasms characterized by many symptoms and a high risk of progression to acute myeloid leukemia. Recent findings indicate that mutations in splicing factors represent a novel class of driver mutations in human cancers and affect about 50% of Myelodysplastic syndrome (MDS) patients. Somatic mutations in MDS patients are frequently found in genes SF3B1, SRSF2, U2AF1, and ZRSR2. Interestingly, they are involved in the recognition of 3' splice sites and exons. It has been reported that mutations in these splicing regulators result in aberrant splicing of many genes. In this review article, we first describe molecular mechanism of pre-mRNA splicing as an introduction and mainly focus on those four splicing factors to describe their mutations and their associated aberrant splicing patterns.

IntSplice2: Prediction of the Splicing Effects of Intronic Single-Nucleotide Variants Using LightGBM Modeling

Prediction of the effect of a single-nucleotide variant (SNV) in an intronic region on aberrant pre-mRNA splicing is challenging except for an SNV affecting the canonical GU/AG splice sites (ss). To predict pathogenicity of SNVs at intronic positions −50 (Int-50) to −3 (Int-3) close to the 3’ ss, we developed light gradient boosting machine (LightGBM)-based IntSplice2 models using pathogenic SNVs in the human gene mutation database (HGMD) and ClinVar and common SNVs in dbSNP with 0.01 ≤ minor allelic frequency (MAF) < 0.50. The LightGBM models were generated using features representing splicing cis-elements. The average recall/sensitivity and specificity of IntSplice2 by fivefold cross-validation (CV) of the training dataset were 0.764 and 0.884, respectively. The recall/sensitivity of IntSplice2 was lower than the average recall/sensitivity of 0.800 of IntSplice that we previously made with support vector machine (SVM) modeling for the same intronic positions. In contrast, the specificity of IntSplice2 was higher than the average specificity of 0.849 of IntSplice. For benchmarking (BM) of IntSplice2 with IntSplice, we made a test dataset that was not used to train IntSplice. After excluding the test dataset from the training dataset, we generated IntSplice2-BM and compared it with IntSplice using the test dataset. IntSplice2-BM was superior to IntSplice in all of the seven statistical measures of accuracy, precision, recall/sensitivity, specificity, F1 score, negative predictive value (NPV), and matthews correlation coefficient (MCC). We made the IntSplice2 web service at https://www.med.nagoya-u.ac.jp/neurogenetics/IntSplice2.

m6 A-mediated alternative splicing coupled with nonsense-mediated mRNA decay regulates SAM synthetase homeostasis

Alternative splicing of pre-mRNAs can regulate gene expression levels by coupling with nonsense-mediated mRNA decay (NMD). In order to elucidate a repertoire of mRNAs regulated by alternative splicing coupled with NMD (AS-NMD) in an organism, we performed long-read RNA sequencing of poly(A)⁺ RNAs from an NMD-deficient mutant strain of Caenorhabditis elegans, and obtained full-length sequences for mRNA isoforms from 259 high-confidence AS-NMD genes. Among them are the S-adenosyl-L-methionine (SAM) synthetase (sams) genes sams-3 and sams-4. SAM synthetase activity autoregulates sams gene expression through AS-NMD in a negative feedback loop. We furthermore find that METT-10, the orthologue of human U6 snRNA methyltransferase METTL16, is required for the splicing regulation in␣vivo, and specifically methylates the invariant AG dinucleotide at the distal 3' splice site (3'SS) in␣vitro. Direct RNA sequencing coupled with machine learning confirms m⁶ A modification of endogenous sams mRNAs. Overall, these results indicate that homeostasis of SAM synthetase in C. elegans is maintained by alternative splicing regulation through m⁶ A modification at the 3'SS of the sams genes.