Bioinformatic identification of candidate cis-regulatory elements involved in human mRNA polyadenylation

Fateful Decisions of Where to Cut the Line: Pathology Associated with Aberrant 3' End Processing and Transcription Termination

Aberrant gene expression lies at the heart of many pathologies. This review will point out how 3' end processing, the final mRNA-maturation step in the transcription cycle, is surprisingly prone to regulated as well as stochastic variations with a wide range of consequences. Whereas smaller variations contribute to the plasticity of gene expression, larger alternations to 3' end processing and coupled transcription termination can lead to pathological consequences. These can be caused by the local mutation of one gene or affect larger numbers of genes systematically, if aspects of the mechanisms of 3' end processing and transcription termination are altered.

The role of structure in regulatory RNA elements

Regulatory RNA elements fulfill functions such as translational regulation, control of transcript levels, and regulation of viral genome replication. Trans-acting factors (i.e., RNA-binding proteins) bind the so-called cis elements and confer functionality to the complex. The specificity during protein-RNA complex (RNP) formation often exploits the structural plasticity of RNA. Functional integrity of cis-trans pairs depends on the availability of properly folded RNA elements, and RNA conformational transitions can cause diseases. Knowledge of RNA structure and the conformational space is needed for understanding complex formation and deducing functional effects. However, structure determination of RNAs under in vivo conditions remains challenging. This review provides an overview of structured eukaryotic and viral RNA cis elements and discusses the effect of RNA structural equilibria on RNP formation. We showcase implications of RNA structural changes for diseases, outline strategies for RNA structure-based drug targeting, and summarize the methodological toolbox for deciphering RNA structures.

Evolutionary dynamics of polyadenylation signals and their recognition strategies in protists

The poly(A) signal, together with auxiliary elements, directs cleavage of a pre-mRNA and thus determines the 3' end of the mature transcript. In many species, including humans, the poly(A) signal is an AAUAAA hexamer, but we recently found that the deeply branching eukaryote Giardia lamblia uses a distinct hexamer (AGURAA) and lacks any known auxiliary elements. Our discovery prompted us to explore the evolutionary dynamics of poly(A) signals and auxiliary elements in the eukaryotic kingdom. We use direct RNA sequencing to determine poly(A) signals for four protists within the Metamonada clade (which also contains G. lamblia) and two outgroup protists. These experiments reveal that the AAUAAA hexamer serves as the poly(A) signal in at least four different eukaryotic clades, indicating that it is likely the ancestral signal, whereas the unusual Giardia version is derived. We find that the use and relative strengths of auxiliary elements are also plastic; in fact, within Metamonada, species like G. lamblia make use of a previously unrecognized auxiliary element where nucleotides flanking the poly(A) signal itself specify genuine cleavage sites. Thus, despite the fundamental nature of pre-mRNA cleavage for the expression of all protein-coding genes, the motifs controlling this process are dynamic on evolutionary timescales, providing motivation for future biochemical and structural studies as well as new therapeutic angles to target eukaryotic pathogens.

Control of mammalian gene expression by modulation of polyA signal cleavage at 5' UTR

The ability to control gene expression in mammalian cells is crucial for safe and efficacious gene therapies and for elucidating gene functions. Current gene regulation systems have limitations such as harmful immune responses or low efficiency. We describe the pA regulator, an RNA-based switch that controls mammalian gene expression through modulation of a synthetic polyA signal (PAS) cleavage introduced into the 5' UTR of a transgene. The cleavage is modulated by a 'dual-mechanism'-(1) aptamer clamping to inhibit PAS cleavage and (2) drug-induced alternative splicing that removes the PAS, both activated by drug binding. This RNA-based methodology circumvents the immune responses observed in other systems and achieves a 900-fold induction with an EC50 of 0.5 µg ml-1 tetracycline (Tc), which is well within the FDA-approved dose range. The pA regulator effectively controls the luciferase transgene in live mice and the endogenous CD133 gene in human cells, in a dose-dependent and reversible manner with long-term stability.

Dynamic regulation of alternative polyadenylation by PQBP1 during neurogenesis

Alternative polyadenylation (APA) is a critical post-transcriptional process that generates mRNA isoforms with distinct 3' untranslated regions (3' UTRs), thereby regulating mRNA localization, stability, and translational efficiency. Cell-type-specific APA extensively shapes the diversity of the cellular transcriptome, particularly during cell fate transition. Despite its recognized significance, the precise regulatory mechanisms governing cell-type-specific APA remain unclear. In this study, we uncover PQBP1 as an emerging APA regulator that actively maintains cell-specific APA profiles in neural progenitor cells (NPCs) and delicately manages the equilibrium between NPC proliferation and differentiation. Multi-omics analysis shows that PQBP1 directly interacts with the upstream UGUA elements, impeding the recruitment of the CFIm complex and influencing polyadenylation site selection within genes associated with the cell cycle. Our findings elucidate the molecular mechanism by which PQBP1 orchestrates dynamic APA changes during neurogenesis, providing valuable insights into the precise regulation of cell-type-specific APA and the underlying pathogenic mechanisms in neurodevelopmental disorders.

Light regulates widespread plant alternative polyadenylation through the chloroplast

Transcription of eukaryotic protein-coding genes generates immature mRNAs that are subjected to a series of processing events, including capping, splicing, cleavage, and polyadenylation (CPA), and chemical modifications of bases. Alternative polyadenylation (APA) greatly contributes to mRNA diversity in the cell. By determining the length of the 3' untranslated region, APA generates transcripts with different regulatory elements, such as miRNA and RBP binding sites, which can influence mRNA stability, turnover, and translation. In the model plant Arabidopsis thaliana, APA is involved in the control of seed dormancy and flowering. In view of the physiological importance of APA in plants, we decided to investigate the effects of light/dark conditions and compare the underlying mechanisms to those elucidated for alternative splicing (AS). We found that light controls APA in approximately 30% of Arabidopsis genes. Similar to AS, the effect of light on APA requires functional chloroplasts, is not affected in mutants of the phytochrome and cryptochrome photoreceptor pathways, and is observed in roots only when the communication with the photosynthetic tissues is not interrupted. Furthermore, mitochondrial and TOR kinase activities are necessary for the effect of light. However, unlike AS, coupling with transcriptional elongation does not seem to be involved since light-dependent APA regulation is neither abolished in mutants of the TFIIS transcript elongation factor nor universally affected by chromatin relaxation caused by histone deacetylase inhibition. Instead, regulation seems to correlate with changes in the abundance of constitutive CPA factors, also mediated by the chloroplast.

InPACT: a computational method for accurate characterization of intronic polyadenylation from RNA sequencing data

Alternative polyadenylation can occur in introns, termed intronic polyadenylation (IPA), has been implicated in diverse biological processes and diseases, as it can produce noncoding transcripts or transcripts with truncated coding regions. However, a reliable method is required to accurately characterize IPA. Here, we propose a computational method called InPACT, which allows for the precise characterization of IPA from conventional RNA-seq data. InPACT successfully identifies numerous previously unannotated IPA transcripts in human cells, many of which are translated, as evidenced by ribosome profiling data. We have demonstrated that InPACT outperforms other methods in terms of IPA identification and quantification. Moreover, InPACT applied to monocyte activation reveals temporally coordinated IPA events. Further application on single-cell RNA-seq data of human fetal bone marrow reveals the expression of several IPA isoforms in a context-specific manner. Therefore, InPACT represents a powerful tool for the accurate characterization of IPA from RNA-seq data.

Genome-wide kinetic profiling of pre-mRNA 3' end cleavage

Cleavage and polyadenylation is necessary for the formation of mature mRNA molecules. The rate at which this process occurs can determine the temporal availability of mRNA for subsequent function throughout the cell and is likely tightly regulated. Despite advances in high-throughput approaches for global kinetic profiling of RNA maturation, genome-wide 3' end cleavage rates have never been measured. Here, we describe a novel approach to estimate the rates of cleavage, using metabolic labeling of nascent RNA, high-throughput sequencing, and mathematical modeling. Using in silico simulations of nascent RNA-seq data, we show that our approach can accurately and precisely estimate cleavage half-lives for both constitutive and alternative sites. We find that 3' end cleavage is fast on average, with half-lives under a minute, but highly variable across individual sites. Rapid cleavage is promoted by the presence of canonical sequence elements and an increased density of polyadenylation signals near a cleavage site. Finally, we find that cleavage rates are associated with the localization of RNA polymerase II at the end of a gene, and faster cleavage leads to quicker degradation of downstream readthrough RNA. Our findings shed light on the features important for efficient 3' end cleavage and the regulation of transcription termination.

Alternative cleavage and polyadenylation of the Ccnb1 mRNA defines accumulation of cyclin protein during the meiotic cell cycle

Progression through the mitotic and meiotic cell cycle is driven by fluctuations in the levels of cyclins, the regulatory subunits controlling the localization and activity of CDK1 kinases. Cyclin levels are regulated through a precise balance of synthesis and degradation. Here we demonstrate that the synthesis of Cyclin B1 during the oocyte meiotic cell cycle is defined by the selective translation of mRNA variants generated through alternative cleavage and polyadenylation (APA). Using gene editing in mice, we introduced mutations into the proximal and distal polyadenylation elements of the 3' untranslated region (UTR) of the Ccnb1 mRNA. Through in vivo loss-of-function experiments, we demonstrate that the translation of mRNA with a short 3' UTR specifies Cyclin B1 protein levels that set the timing of meiotic re-entry. In contrast, translation directed by a long 3' UTR is necessary to direct Cyclin B1 protein accumulation during the MI/MII transition. These findings establish that the progression through the cell cycle is dependent on the selective translation of multiple mRNA variants generated by APA.

Elevated pre-mRNA 3' end processing activity in cancer cells renders vulnerability to inhibition of cleavage and polyadenylation

Cleavage and polyadenylation (CPA) is responsible for 3' end processing of eukaryotic poly(A)+ RNAs and preludes transcriptional termination. JTE-607, which targets CPSF-73, is the first known CPA inhibitor (CPAi) in mammalian cells. Here we show that JTE-607 perturbs gene expression through both transcriptional readthrough and alternative polyadenylation (APA). Sensitive genes are associated with features similar to those previously identified for PCF11 knockdown, underscoring a unified transcriptomic signature of CPAi. The degree of inhibition of an APA site by JTE-607 correlates with its usage level and, consistently, cells with elevated CPA activities, such as those with induced overexpression of FIP1, display greater transcriptomic disturbances when treated with JTE-607. Moreover, JTE-607 causes S phase crisis and is hence synergistic with inhibitors of DNA damage repair pathways. Together, our data reveal CPA activity and proliferation rate as determinants of CPAi-mediated cell death, raising the possibility of using CPAi as an adjunct therapy to suppress certain cancers.

The emerging roles of CFIm25 (NUDT21/CPSF5) in human biology and disease

The mammalian cleavage factor I subunit CFIm25 (NUDT21) binds to the UGUA sequences of precursor RNAs. Traditionally, CFIm25 is known to facilitate 3' end formation of pre-mRNAs resulting in the formation of polyadenylated transcripts. Recent studies suggest that CFIm25 may be involved in the cyclization and hence generation of circular RNAs (circRNAs) that contain UGUA motifs. These circRNAs act as competing endogenous RNAs (ceRNAs) that disrupt the ceRNA-miRNA-mRNA axis. Other emerging roles of CFIm25 include regulating both alternative splicing and alternative polyadenylation (APA). APA generates different sized transcripts that may code for different proteins, or more commonly transcripts that code for the same protein but differ in the length and sequence content of their 3' UTRs (3' UTR-APA). CFIm25 mediated global changes in 3' UTR-APA affect human physiology including spermatogenesis and the determination of cell fate. Deregulation of CFIm25 and changes in 3' UTR-APA have been implicated in several human diseases including cancer. In many cancers, CFIm25 acts as a tumor suppressor. However, there are some cancers where CFIm25 has the opposite effect. Alterations in CFIm25-driven 3' UTR-APA may also play a role in neural dysfunction and fibrosis. CFIm25 mediated 3' UTR-APA changes can be used to generate specific signatures that can be used as potential biomarkers in development and disease. Due to the emerging role of CFIm25 as a regulator of the aforementioned RNA processing events, modulation of CFIm25 levels may be a novel viable therapeutic approach. This article is categorized under: RNA Processing > 3' End Processing RNA in Disease and Development > RNA in Disease.

Slight Variations in the Sequence Downstream of the Polyadenylation Signal Significantly Increase Transgene Expression in HEK293T and CHO Cells

Compared to transcription initiation, much less is known about transcription termination. In particular, large-scale mutagenesis studies have, so far, primarily concentrated on promoter and enhancer, but not terminator sequences. Here, we used a massively parallel reporter assay (MPRA) to systematically analyze the influence of short (8 bp) sequence variants (mutations) located downstream of the polyadenylation signal (PAS) on the steady-state mRNA level of the upstream gene, employing an eGFP reporter and human HEK293T cells as a model system. In total, we evaluated 227,755 mutations located at different overlapping positions within +17..+56 bp downstream of the PAS for their ability to regulate the reporter gene expression. We found that the positions +17..+44 bp downstream of the PAS are more essential for gene upregulation than those located more distal to the PAS, and that the mutation sequences ensuring high levels of eGFP mRNA expression are extremely T-rich. Next, we validated the positive effect of a couple of mutations identified in the MPRA screening on the eGFP and luciferase protein expression. The most promising mutation increased the expression of the reporter proteins 13-fold and sevenfold on average in HEK293T and CHO cells, respectively. Overall, these findings might be useful for further improving the efficiency of production of therapeutic products, e.g., recombinant antibodies.

Alternative polyadenylation regulation: insights from sequential polyadenylation

The processing of the proximal and distal poly(A) sites in alternative polyadenylation (APA) has long been thought to independently occur on pre-mRNAs during transcription. However, a recent study by our groups demonstrated that the proximal sites for many genes could be activated sequentially following the distal ones, suggesting a multi-cleavage-same-transcript mode beyond the canonical one-cleavage-per-transcript view. Here, we review the established mechanisms for APA regulation and then discuss the additional insights into APA regulation from the perspective of sequential polyadenylation, resulting in a unified leverage model for understanding the mechanisms of regulated APA.

Intronic Polyadenylation in Acquired Cancer Drug Resistance Circumvented by Utilizing CRISPR/Cas9 with Homology-Directed Repair: The Tale of Human DNA Topoisomerase IIα

Simple Summary

DNA topoisomerase IIα (170 kDa, TOP2α/170) resolves nucleic acid topological entanglements by generating transient double-strand DNA breaks. TOP2α inhibitors/poisons stabilize TOP2α-DNA covalent complexes resulting in persistent DNA damage and are frequently utilized to treat a variety of cancers. Acquired resistance to these chemotherapeutic agents is often associated with decreased TOP2α/170 expression levels. Studies have demonstrated that a reduction in TOP2α/170 results from a type of alternative polyadenylation designated intronic polyadenylation (IPA). As a consequence of IPA, variant TOP2α mRNA transcripts have been characterized that have resulted in the translation of C-terminal truncated TOP2α isoforms with altered biological activities. In this paper, an example is discussed where circumvention of acquired TOP2α-mediated drug resistance was achieved by utilizing CRISPR/Cas9 specific gene editing of an exon/intron boundary through homology directed repair (HDR) to reduce TOP2α IPA. These results illustrate the therapeutic potential of CRISPR/Cas9/HDR to impact drug resistance associated with aberrant IPA.

Abstract

Intronic polyadenylation (IPA) plays a critical role in malignant transformation, development, progression, and cancer chemoresistance by contributing to transcriptome/proteome alterations. DNA topoisomerase IIα (170 kDa, TOP2α/170) is an established clinical target for anticancer agents whose efficacy is compromised by drug resistance often associated with a reduction of nuclear TOP2α/170 levels. In leukemia cell lines with acquired resistance to TOP2α-targeted drugs and reduced TOP2α/170 expression, variant TOP2α mRNA transcripts have been reported due to IPA that resulted in the translation of C-terminal truncated isoforms with altered nuclear-cytoplasmic distribution or heterodimerization with wild-type TOP2α/170. This review provides an overview of the various mechanisms regulating pre-mRNA processing and alternative polyadenylation, as well as the utilization of CRISPR/Cas9 specific gene editing through homology directed repair (HDR) to decrease IPA when splice sites are intrinsically weak or potentially mutated. The specific case of TOP2α exon 19/intron 19 splice site editing is discussed in etoposide-resistant human leukemia K562 cells as a tractable strategy to circumvent acquired TOP2α-mediated drug resistance. This example supports the importance of aberrant IPA in acquired drug resistance to TOP2α-targeted drugs. In addition, these results demonstrate the therapeutic potential of CRISPR/Cas9/HDR to impact drug resistance associated with aberrant splicing/polyadenylation.

DeeReCT-APA: Prediction of Alternative Polyadenylation Site Usage Through Deep Learning

Alternative polyadenylation (APA) is a crucial step in post-transcriptional regulation. Previous bioinformatic studies have mainly focused on the recognition of polyadenylation sites (PASs) in a given genomic sequence, which is a binary classification problem. Recently, computational methods for predicting the usage level of alternative PASs in the same gene have been proposed. However, all of them cast the problem as a non-quantitative pairwise comparison task and do not take the competition among multiple PASs into account. To address this, here we propose a deep learning architecture, Deep Regulatory Code and Tools for Alternative Polyadenylation (DeeReCT-APA), to quantitatively predict the usage of all alternative PASs of a given gene. To accommodate different genes with potentially different numbers of PASs, DeeReCT-APA treats the problem as a regression task with a variable-length target. Based on a convolutional neural network-long short-term memory (CNN-LSTM) architecture, DeeReCT-APA extracts sequence features with CNN layers, uses bidirectional LSTM to explicitly model the interactions among competing PASs, and outputs percentage scores representing the usage levels of all PASs of a gene. In addition to the fact that only our method can quantitatively predict the usage of all the PASs within a gene, we show that our method consistently outperforms other existing methods on three different tasks for which they are trained: pairwise comparison task, highest usage prediction task, and ranking task. Finally, we demonstrate that our method can be used to predict the effect of genetic variations on APA patterns and sheds light on future mechanistic understanding in APA regulation. Our code and data are available at https://github.com/lzx325/DeeReCT-APA-repo.

Precise gene models using long-read sequencing reveal a unique poly(A) signal in Giardia lamblia

During pre-mRNA processing, the poly(A) signal is recognized by a protein complex that ensures precise cleavage and polyadenylation of the nascent transcript. The location of this cleavage event establishes the length and sequence of the 3' UTR of an mRNA, thus determining much of its post-transcriptional fate. Using long-read sequencing, we characterize the polyadenylation signal and related sequences surrounding Giardia lamblia cleavage sites for over 2600 genes. We find that G. lamblia uses an AGURAA poly(A) signal, which differs from the mammalian AAUAAA. We also describe how G. lamblia lacks common auxiliary elements found in other eukaryotes, along with the proteins that recognize them. Further, we identify 133 genes with evidence of alternative polyadenylation. These results suggest that despite pared-down cleavage and polyadenylation machinery, 3' end formation still appears to be an important regulatory step for gene expression in G. lamblia.

Regulation of the Alternative Neural Transcriptome by ELAV/Hu RNA Binding Proteins

The process of alternative polyadenylation (APA) generates multiple 3' UTR isoforms for a given locus, which can alter regulatory capacity and on occasion change coding potential. APA was initially characterized for a few genes, but in the past decade, has been found to be the rule for metazoan genes. While numerous differences in APA profiles have been catalogued across genetic conditions, perturbations, and diseases, our knowledge of APA mechanisms and biology is far from complete. In this review, we highlight recent findings regarding the role of the conserved ELAV/Hu family of RNA binding proteins (RBPs) in generating the broad landscape of lengthened 3' UTRs that is characteristic of neurons. We relate this to their established roles in alternative splicing, and summarize ongoing directions that will further elucidate the molecular strategies for neural APA, the in vivo functions of ELAV/Hu RBPs, and the phenotypic consequences of these regulatory paradigms in neurons.

Alternative polyadenylation by sequential activation of distal and proximal PolyA sites

Analogous to alternative splicing, alternative polyadenylation (APA) has long been thought to occur independently at proximal and distal polyA sites. Using fractionation-seq, we unexpectedly identified several hundred APA genes in human cells whose distal polyA isoforms are retained in chromatin/nuclear matrix and whose proximal polyA isoforms are released into the cytoplasm. Global metabolic PAS-seq and Nanopore long-read RNA-sequencing provide further evidence that the strong distal polyA sites are processed first and the resulting transcripts are subsequently anchored in chromatin/nuclear matrix to serve as precursors for further processing at proximal polyA sites. Inserting an autocleavable ribozyme between the proximal and distal polyA sites, coupled with a Cleave-seq approach that we describe here, confirms that the distal polyA isoform is indeed the precursor to the proximal polyA isoform. Therefore, unlike alternative splicing, APA sites are recognized independently, and in many cases, in a sequential manner. This provides a versatile strategy to regulate gene expression in mammalian cells.

Benchmarking sequencing methods and tools that facilitate the study of alternative polyadenylation

BACKGROUND

Alternative cleavage and polyadenylation (APA), an RNA processing event, occurs in over 70% of human protein-coding genes. APA results in mRNA transcripts with distinct 3' ends. Most APA occurs within 3' UTRs, which harbor regulatory elements that can impact mRNA stability, translation, and localization.

RESULTS

APA can be profiled using a number of established computational tools that infer polyadenylation sites from standard, short-read RNA-seq datasets. Here, we benchmarked a number of such tools-TAPAS, QAPA, DaPars2, GETUTR, and APATrap- against 3'-Seq, a specialized RNA-seq protocol that enriches for reads at the 3' ends of genes, and Iso-Seq, a Pacific Biosciences (PacBio) single-molecule full-length RNA-seq method in their ability to identify polyadenylation sites and quantify polyadenylation site usage. We demonstrate that 3'-Seq and Iso-Seq are able to identify and quantify the usage of polyadenylation sites more reliably than computational tools that take short-read RNA-seq as input. However, we find that running one such tool, QAPA, with a set of polyadenylation site annotations derived from small quantities of 3'-Seq or Iso-Seq can reliably quantify variation in APA across conditions, such asacross genotypes, as demonstrated by the successful mapping of alternative polyadenylation quantitative trait loci (apaQTL).

CONCLUSIONS

We envisage that our analyses will shed light on the advantages of studying APA with more specialized sequencing protocols, such as 3'-Seq or Iso-Seq, and the limitations of studying APA with short-read RNA-seq. We provide a computational pipeline to aid in the identification of polyadenylation sites and quantification of polyadenylation site usages using Iso-Seq data as input.

On the Cutting Edge: Regulation and Therapeutic Potential of the mRNA 3' End Nuclease

Cleavage of nascent transcripts is a fundamental process for eukaryotic mRNA maturation and for the production of different mRNA isoforms. In eukaryotes, cleavage of mRNA precursors by the highly conserved endonuclease CPSF73 is critical for mRNA stability, export from the nucleus, and translation. As an essential enzyme in the cell, CPSF73 surprisingly shows promise as a drug target for specific cancers and for protozoan parasites. In this review, we cover our current understanding of CPSF73 in cleavage and polyadenylation, histone pre-mRNA processing, and transcription termination. We discuss the potential of CPSF73 as a target for novel therapeutics and highlight further research into the regulation of CPSF73 that will be critical to understanding its role in cancer and other diseases.

CRISPRpas: programmable regulation of alternative polyadenylation by dCas9

Most human protein-coding genes produce alternative polyadenylation (APA) isoforms that differ in 3' UTR size or, when coupled with splicing, have variable coding sequences. APA is an important layer of gene expression program critical for defining cell identity. Here, by using a catalytically dead Cas9 and coupling its target site with polyadenylation site (PAS), we develop a method, named CRISPRpas, to alter APA isoform abundance. CRISPRpas functions by enhancing proximal PAS usage, whose efficiency is influenced by several factors, including targeting strand of DNA, distance between PAS and target sequence and strength of the PAS. For intronic polyadenylation (IPA), splicing features, such as strengths of 5' splice site and 3' splice site, also affect CRISPRpas efficiency. We show modulation of APA of multiple endogenous genes, including IPA of PCF11, a master regulator of APA and gene expression. In sum, CRISPRpas offers a programmable tool for APA regulation that impacts gene expression.

Known sequence features can explain half of all human gene ends

Cleavage and polyadenylation (CPA) sites define eukaryotic gene ends. CPA sites are associated with five key sequence recognition elements: the upstream UGUA, the polyadenylation signal (PAS), and U-rich sequences; the CA/UA dinucleotide where cleavage occurs; and GU-rich downstream elements (DSEs). Currently, it is not clear whether these sequences are sufficient to delineate CPA sites. Additionally, numerous other sequences and factors have been described, often in the context of promoting alternative CPA sites and preventing cryptic CPA site usage. Here, we dissect the contributions of individual sequence features to CPA using standard discriminative models. We show that models comprised only of the five primary CPA sequence features give highest probability scores to constitutive CPA sites at the ends of coding genes, relative to the entire pre-mRNA sequence, for 41% of all human genes. U1-hybridizing sequences provide a small boost in performance. The addition of all known RBP RNA binding motifs to the model, however, increases this figure to 49%, and suggests an involvement of both known and suspected CPA regulators as well as potential new factors in delineating constitutive CPA sites. To our knowledge, this high effectiveness of established features to predict human gene ends has not previously been documented.

Alternative Polyadenylation in Stem Cell Self-Renewal and Differentiation

Cellular function is shaped by transcriptional and post-transcriptional mechanisms, including alternative polyadenylation (APA). By directly controlling 3'- untranslated region (UTR) length and the selection of the last exon, APA regulates up to 70% of all cellular transcripts influencing RNA stability, output, and protein isoform expression. Cell-state-dependent 3'-UTR shortening has been identified as a hallmark of cellular proliferation. Hence, quiescent/dormant stem cells are characterized by long 3'-UTRs, whereas proliferative stem/progenitor cells exhibit 3'-UTR shortening. Here, the latest studies analyzing the role of APA in regulating stem cell state, self-renewal, differentiation, and metabolism are reviewed. The new role of APA in controlling stem cell fate opens novel potential therapeutic avenues in the field of regenerative medicine.

On the function and relevance of alternative 3'-UTRs in gene expression regulation

Messanger RNA (mRNA) isoforms with alternative 3'-untranslated regions (3'-UTRs) are produced by alternative polyadenylation (APA), which occurs during transcription in most eukaryotic genes. APA fine-tunes gene expression in a cell-type- and cellular state-dependent manner. Selection of an APA site entails the binding of core cleavage and polyadenylation factors to a particular polyadenylation site localized in the pre-mRNA and is controlled by multiple regulatory determinants, including transcription, pre-mRNA cis-regulatory sequences, and protein factors. Alternative 3'-UTRs serve as platforms for specific RNA binding proteins and microRNAs, which regulate gene expression in a coordinated manner by controlling mRNA fate and function in the cell. Genome-wide studies illustrated the full extent of APA prevalence and revealed that specific 3'-UTR profiles are associated with particular cellular states and diseases. Generally, short 3'-UTRs are associated with proliferative and cancer cells, and long 3'-UTRs are mostly found in polarized and differentiated cells. Fundamental new insights on the physiological consequences of this widespread event and the molecular mechanisms involved have been revealed through single-cell studies. Publicly available comprehensive databases that cover all APA mRNA isoforms identified in many cellular states and diseases reveal specific APA signatures. Therapies tackling APA mRNA isoforms or APA regulators may be regarded as innovative and attractive tools for diagnostics or treatment of several pathologies. We highlight the function of APA and alternative 3'-UTRs in gene expression regulation, the control of these mechanisms, their physiological consequences, and their potential use as new biomarkers and therapeutic tools. This article is categorized under: RNA Processing > 3' End Processing RNA Interactions with Proteins and Other Molecules > Protein-RNA Interactions: Functional Implications RNA in Disease and Development > RNA in Disease.

Aptardi predicts polyadenylation sites in sample-specific transcriptomes using high-throughput RNA sequencing and DNA sequence

Annotation of polyadenylation sites from short-read RNA sequencing alone is a challenging computational task. Other algorithms rooted in DNA sequence predict potential polyadenylation sites; however, in vivo expression of a particular site varies based on a myriad of conditions. Here, we introduce aptardi (alternative polyadenylation transcriptome analysis from RNA-Seq data and DNA sequence information), which leverages both DNA sequence and RNA sequencing in a machine learning paradigm to predict expressed polyadenylation sites. Specifically, as input aptardi takes DNA nucleotide sequence, genome-aligned RNA-Seq data, and an initial transcriptome. The program evaluates these initial transcripts to identify expressed polyadenylation sites in the biological sample and refines transcript 3'-ends accordingly. The average precision of the aptardi model is twice that of a standard transcriptome assembler. In particular, the recall of the aptardi model (the proportion of true polyadenylation sites detected by the algorithm) is improved by over three-fold. Also, the model-trained using the Human Brain Reference RNA commercial standard-performs well when applied to RNA-sequencing samples from different tissues and different mammalian species. Finally, aptardi's input is simple to compile and its output is easily amenable to downstream analyses such as quantitation and differential expression.

Nuclear mRNA maturation and mRNA export control: from trypanosomes to opisthokonts

The passage of mRNAs through the nuclear pores into the cytoplasm is essential in all eukaryotes. For regulation, mRNA export is tightly connected to the full machinery of nuclear mRNA processing, starting at transcription. Export competence of pre-mRNAs gradually increases by both transient and permanent interactions with multiple RNA processing and export factors. mRNA export is best understood in opisthokonts, with limited knowledge in plants and protozoa. Here, I review and compare nuclear mRNA processing and export between opisthokonts and Trypanosoma brucei. The parasite has many unusual features in nuclear mRNA processing, such as polycistronic transcription and trans-splicing. It lacks several nuclear complexes and nuclear-pore-associated proteins that in opisthokonts play major roles in mRNA export. As a consequence, trypanosome mRNA export control is not tight and export can even start co-transcriptionally. Whether trypanosomes regulate mRNA export at all, or whether leakage of immature mRNA to the cytoplasm is kept to a low level by a fast kinetics of mRNA processing remains to be investigated. mRNA export had to be present in the last common ancestor of eukaryotes. Trypanosomes are evolutionary very distant from opisthokonts and a comparison helps understanding the evolution of mRNA export.

Advances in the Bioinformatics Knowledge of mRNA Polyadenylation in Baculovirus Genes

Baculoviruses are a group of insect viruses with large circular dsDNA genomes exploited in numerous biotechnological applications, such as the biological control of agricultural pests, the expression of recombinant proteins or the gene delivery of therapeutic sequences in mammals, among others. Their genomes encode between 80 and 200 proteins, of which 38 are shared by all reported species. Thanks to multi-omic studies, there is remarkable information about the baculoviral proteome and the temporality in the virus gene expression. This allows some functional elements of the genome to be very well described, such as promoters and open reading frames. However, less information is available about the transcription termination signals and, consequently, there are still imprecisions about what are the limits of the transcriptional units present in the baculovirus genomes and how is the processing of the 3′ end of viral mRNA. Regarding to this, in this review we provide an update about the characteristics of DNA signals involved in this process and we contribute to their correct prediction through an exhaustive analysis that involves bibliography information, data mining, RNA structure and a comprehensive study of the core gene 3′ ends from 180 baculovirus genomes.

Molecular mechanism for the interaction between human CPSF30 and hFip1

In this study, Hamilton and Tong set out to elucidate the molecular basis for the CPSF30–hFip1 interaction. Using structural, biophysical, and biochemical experiments, the authors define how the mammalian polyadenylation specificity factor (mPSF) is organized and add to our understanding of the pre-mRNA 3′-end processing machinery.

Most eukaryotic pre-mRNAs must undergo 3′-end cleavage and polyadenylation prior to their export from the nucleus. A large number of proteins in several complexes participate in this 3′-end processing, including cleavage and polyadenylation specificity factor (CPSF) in mammals. The CPSF30 subunit contains five CCCH zinc fingers (ZFs), with ZF2–ZF3 being required for the recognition of the AAUAAA poly(A) signal. ZF4–ZF5 recruits the hFip1 subunit of CPSF, although the details of this interaction have not been characterized. Here we report the crystal structure of human CPSF30 ZF4–ZF5 in complex with residues 161–200 of hFip1 at 1.9 Å resolution, illuminating the molecular basis for their interaction. Unexpectedly, the structure reveals one hFip1 molecule binding to each ZF4 and ZF5, with a conserved mode of interaction. Our mutagenesis studies confirm that the CPSF30–hFip1 complex has 1:2 stoichiometry in vitro. Mutation of each binding site in CPSF30 still allows one copy of hFip1 to bind, while mutation of both sites abrogates binding. Our fluorescence polarization binding assays show that ZF4 has higher affinity for hFip1, with a K_d of 1.8 nM. We also demonstrate that two copies of the catalytic module of poly(A) polymerase (PAP) are recruited by the CPSF30–hFip1 complex in vitro, and both hFip1 binding sites in CPSF30 can support polyadenylation.

SANPolyA: a deep learning method for identifying Poly(A) signals

MOTIVATION

Polyadenylation plays a regulatory role in transcription. The recognition of polyadenylation signal (PAS) motif sequence is an important step in polyadenylation. In the past few years, some statistical machine learning-based and deep learning-based methods have been proposed for PAS identification. Although these methods predict PAS with success, there is room for their improvement on PAS identification.

RESULTS

In this study, we proposed a deep neural network-based computational method, called SANPolyA, for identifying PAS in human and mouse genomes. SANPolyA requires no manually crafted sequence features. We compared our method SANPolyA with several previous PAS identification methods on several PAS benchmark datasets. Our results showed that SANPolyA outperforms the state-of-art methods. SANPolyA also showed good performance on leave-one-motif-out evaluation.

AVAILABILITY AND IMPLEMENTATION

https://github.com/yuht4/SANPolyA.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

DeepPASTA: deep neural network based polyadenylation site analysis

MOTIVATION

Alternative polyadenylation (polyA) sites near the 3' end of a pre-mRNA create multiple mRNA transcripts with different 3' untranslated regions (3' UTRs). The sequence elements of a 3' UTR are essential for many biological activities such as mRNA stability, sub-cellular localization, protein translation, protein binding and translation efficiency. Moreover, numerous studies in the literature have reported the correlation between diseases and the shortening (or lengthening) of 3' UTRs. As alternative polyA sites are common in mammalian genes, several machine learning tools have been published for predicting polyA sites from sequence data. These tools either consider limited sequence features or use relatively old algorithms for polyA site prediction. Moreover, none of the previous tools consider RNA secondary structures as a feature to predict polyA sites.

RESULTS

In this paper, we propose a new deep learning model, called DeepPASTA, for predicting polyA sites from both sequence and RNA secondary structure data. The model is then extended to predict tissue-specific polyA sites. Moreover, the tool can predict the most dominant (i.e. frequently used) polyA site of a gene in a specific tissue and relative dominance when two polyA sites of the same gene are given. Our extensive experiments demonstrate that DeepPASTA signisficantly outperforms the existing tools for polyA site prediction and tissue-specific relative and absolute dominant polyA site prediction.

AVAILABILITY AND IMPLEMENTATION

https://github.com/arefeen/DeepPASTA.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

DeeReCT-PolyA: a robust and generic deep learning method for PAS identification

Motivation

Polyadenylation is a critical step for gene expression regulation during the maturation of mRNA. An accurate and robust method for poly(A) signals (PASs) identification is not only desired for the purpose of better transcripts’ end annotation, but can also help us gain a deeper insight of the underlying regulatory mechanism. Although many methods have been proposed for PAS recognition, most of them are PAS motif- and human-specific, which leads to high risks of overfitting, low generalization power, and inability to reveal the connections between the underlying mechanisms of different mammals.

Results

In this work, we propose a robust, PAS motif agnostic, and highly interpretable and transferrable deep learning model for accurate PAS recognition, which requires no prior knowledge or human-designed features. We show that our single model trained over all human PAS motifs not only outperforms the state-of-the-art methods trained on specific motifs, but can also be generalized well to two mouse datasets. Moreover, we further increase the prediction accuracy by transferring the deep learning model trained on the data of one species to the data of a different species. Several novel underlying poly(A) patterns are revealed through the visualization of important oligomers and positions in our trained models. Finally, we interpret the deep learning models by converting the convolutional filters into sequence logos and quantitatively compare the sequence logos between human and mouse datasets.

Availability and implementation

https://github.com/likesum/DeeReCT-PolyA

Supplementary information

Supplementary data are available at Bioinformatics online.

Recent molecular insights into canonical pre-mRNA 3'-end processing

The majority of eukaryotic messenger RNA precursors (pre-mRNAs) undergo cleavage and polyadenylation at their 3' end. This canonical 3'-end processing depends on sequence elements in the pre-mRNA as well as a mega-dalton protein machinery. The cleavage site in mammalian pre-mRNAs is located between an upstream poly(A) signal, most frequently an AAUAAA hexamer, and a GU-rich downstream sequence element. This review will summarize recent advances from the studies on this canonical 3'-end processing machinery. They have revealed the molecular mechanism for the recognition of the poly(A) signal and provided the first glimpse into the overall architecture of the machinery. The studies also show that the machinery is highly dynamic conformationally, and extensive re-arrangements are necessary for its activation. Inhibitors targeting the active site of the CPSF73 nuclease of this machinery have anti-cancer, anti-inflammatory and anti-protozoal effects, indicating that CPSF73 and pre-mRNA 3'-end processing in general are attractive targets for drug discovery.

ABBREVIATIONS

APA: alternative polyadenylation; β-CASP: metallo-β-lactamase-associated CPSF Artemis SNM1/PSO2; CTD: C-terminal domain; CF: cleavage factor; CPF: cleavage and polyadenylation factor; CPSF: cleavage and polyadenylation specificity factor; CstF: cleavage stimulation factor; DSE: downstream element; HAT: half a TPR; HCC: histone pre-mRNA cleavage complex; mCF: mammalian cleavage factor; mPSF: mammalian polyadenylation specificity factor; mRNA: messenger RNA; nt: nucleotide; NTD: N-terminal domain; PAP: polyadenylate polymerase; PAS: polyadenylation signal; PIM: mPSF interaction motif; Poly(A): polyadenylation, polyadenylate; Pol II: RNA polymerase II; pre-mRNA: messenger RNA precursor; RRM: RNA recognition module, RNA recognition motif; snRNP: small nuclear ribonucleoprotein; TPR: tetratricopeptide repeat; UTR: untranslated region; ZF: zinc finger.

Pathogenic diversity of RNA variants and RNA variation-associated factors in cancer development

Recently, with the development of RNA sequencing technologies such as next-generation sequencing (NGS) for RNA, numerous variations of alternatively processed RNAs made by alternative splicing, RNA editing, alternative maturation of microRNA (miRNA), RNA methylation, and alternative polyadenylation have been uncovered. Furthermore, abnormally processed RNAs can cause a variety of diseases, including obesity, diabetes, Alzheimer’s disease, and cancer. Especially in cancer development, aberrant RNAs caused by deregulated RNA modifiers or regulators are related to progression. Accumulating evidence has reported that aberrant RNAs promote carcinogenesis in many cancers, including liver cancer, leukemia, melanoma, lung cancer, breast cancer, and other cancers, in which abnormal RNA processing occurs in normal cells. Therefore, it is necessary to understand the precise roles and mechanisms of disease-related RNA processing in various cancers for the development of therapeutic interventions. In this review, the underlying mechanisms of variations in the RNA life cycle and the biological impacts of RNA variations on carcinogenesis will be discussed, and therapeutic strategies for the treatment of tumor malignancies will be provided. We also discuss emerging roles of RNA regulators in hepatocellular carcinogenesis.

A single gene can generate a variety of RNA products, and changes in this final RNA output can directly contribute to cancer onset and progression. The initial transcription of each DNA sequence yields a raw RNA strand that subsequently undergoes a variety of modification processes that shape its function. Hee Doo Yang and Suk Woo Nam, The Catholic University of Korea, Seoul, have reviewed the potential impact of these mechanisms on malignancy. Some cancers, for example, express RNA sequences that have been inappropriately edited by an enzymatic process called splicing, yielding abnormal RNAs that drive metastasis. Other tumors contain RNAs with atypical chemical modifications, or in which individual nucleotides have been enzymatically converted into other nucleotides. A deeper understanding of these RNA alterations and their impacts could lead to more effective and targeted cancer treatments.

Differential Alternative Polyadenylation Landscapes Mediate Hematopoietic Stem Cell Activation and Regulate Glutamine Metabolism

Alternative polyadenylation (APA) is emerging as an important regulatory mechanism of RNA and protein isoform expression by controlling 3' untranslated region (3'-UTR) composition. The relevance of APA in stem cell hierarchies remains elusive. Here, we first demonstrate the requirement of the APA regulator Pabpn1 for hematopoietic stem cell (HSC) function. We then determine the genome-wide APA landscape (APAome) of HSCs and progenitors by performing low-input 3' sequencing paired with bioinformatic pipelines. This reveals transcriptome-wide dynamic APA patterns and an overall shortening of 3'-UTRs during differentiation and upon homeostatic or stress-induced transition from quiescence to proliferation. Specifically, we show that APA regulates activation-induced Glutaminase (Gls) isoform switching by Nudt21. This adaptation of the glutamine metabolism by increasing the GAC:KGA isoform ratio fuels versatile metabolic pathways necessary for HSC self-renewal and proper stress response. Our study establishes APA as a critical regulatory layer orchestrating HSC self-renewal, behavior, and commitment.

Stiff matrix instigates type I collagen biogenesis by mammalian cleavage factor I complex-mediated alternative polyadenylation

Alternative polyadenylation (APA) is a widespread and important mechanism in regulation of gene expression. Dysregulation of the 3' UTR cleavage and polyadenylation represents a common characteristic among many disease states, including lung fibrosis. In this study, we investigated the role of mammalian cleavage factor I-mediated (CFIm-mediated) APA in regulating extracellular matrix production in response to mechanical stimuli from stiffened matrix simulating the fibrotic lungs. We found that stiff matrix downregulated expression of CFIm68, CFIm59 and CFIm25 subunits and promoted APA in favor of the proximal poly(A) site usage in the 3' UTRs of type I collagen (COL1A1) and fibronectin (FN1) in primary human lung fibroblasts. Knockdown and overexpression of each individual CFIm subunit demonstrated that CFIm68 and CFIm25 are indispensable attributes of stiff matrix-induced APA and overproduction of COL1A1, whereas CFIm did not appear to mediate stiffness-regulated FN1 APA. Furthermore, expression of the CFIm subunits was associated with matrix stiffness in vivo in a bleomycin-induced mouse model of pulmonary fibrosis. These data suggest that stiff matrix instigates type I collagen biogenesis by selectively targeting mRNA transcripts for 3' UTR shortening. The current study uncovered a potential mechanism for regulation of the CFIm complex by mechanical cues under fibrotic conditions.

Alternative polyadenylation of mRNA and its role in cancer

Alternative polyadenylation (APA) is a molecular process that generates diversity at the 3′ end of RNA polymerase II transcripts from over 60% of human genes. APA is derived from the existence of multiple polyadenylation signals (PAS) within the same transcript, and results in the differential inclusion of sequence information at the 3′ end. While APA can occur between two PASs allowing for generation of transcripts with distinct coding potential from a single gene, most APA occurs within the untranslated region (3′UTR) and changes the length and content of these non-coding sequences. APA within the 3′UTR can have tremendous impact on its regulatory potential of the mRNA through a variety of mechanisms, and indeed this layer of gene expression regulation has profound impact on processes vital to cell growth and development. Recent studies have particularly highlighted the importance of APA dysregulation in cancer onset and progression. Here, we review the current knowledge of APA and its impacts on mRNA stability, translation, localization and protein localization. We also discuss the implications of APA dysregulation in cancer research and therapy.

Inference of the human polyadenylation code

Motivation

Processing of transcripts at the 3′-end involves cleavage at a polyadenylation site followed by the addition of a poly(A)-tail. By selecting which site is cleaved, the process of alternative polyadenylation enables genes to produce transcript isoforms with different 3′-ends. To facilitate the identification and treatment of disease-causing mutations that affect polyadenylation and to understand the sequence determinants underlying this regulatory process, a computational model that can accurately predict polyadenylation patterns from genomic features is desirable.

Results

Previous works have focused on identifying candidate polyadenylation sites and classifying tissue-specific sites. By training on how multiple sites in genes are competitively selected for polyadenylation from 3′-end sequencing data, we developed a deep learning model that can predict the tissue-specific strength of a polyadenylation site in the 3′ untranslated region of the human genome given only its genomic sequence. We demonstrate the model’s broad utility on multiple tasks, without any application-specific training. The model can be used to predict which polyadenylation site is more likely to be selected in genes with multiple sites. It can be used to scan the 3′ untranslated region to find candidate polyadenylation sites. It can be used to classify the pathogenicity of variants near annotated polyadenylation sites in ClinVar. It can also be used to anticipate the effect of antisense oligonucleotide experiments to redirect polyadenylation. We provide analysis on how different features affect the model’s predictive performance and a method to identify sensitive regions of the genome at the single-based resolution that can affect polyadenylation regulation.

Supplementary information

Supplementary data are available at Bioinformatics online.

Sequence determinants of polyadenylation-mediated regulation

The cleavage and polyadenylation reaction is a crucial step in transcription termination and pre-mRNA maturation in human cells. Despite extensive research, the encoding of polyadenylation-mediated regulation of gene expression within the DNA sequence is not well understood. Here, we utilized a massively parallel reporter assay to inspect the effect of over 12,000 rationally designed polyadenylation sequences (PASs) on reporter gene expression and cleavage efficiency. We find that the PAS sequence can modulate gene expression by over five orders of magnitude. By using a uniquely designed scanning mutagenesis data set, we gain mechanistic insight into various modes of action by which the cleavage efficiency affects the sensitivity or robustness of the PAS to mutation. Furthermore, we employ motif discovery to identify both known and novel sequence motifs associated with PAS-mediated regulation. By leveraging the large scale of our data, we train a deep learning model for the highly accurate prediction of RNA levels from DNA sequence alone (R = 0.83). Moreover, we devise unique approaches for predicting exact cleavage sites for our reporter constructs and for endogenous transcripts. Taken together, our results expand our understanding of PAS-mediated regulation, and provide an unprecedented resource for analyzing and predicting PAS for regulatory genomics applications.

Mechanistic insights into mRNA 3'-end processing

•

Integrated structural biology approaches have provided new insights into the mechanism of eukaryotic mRNA 3′-end processing.

•

The polymerase modules of yeast and human cleavage and polyadenylation factors share a conserved architecture.

•

CryoEM structures of human CPSF have revealed the mechanism of AAUAAA polyadenylation signal recognition.

•

Cleavage and polyadenylation of mRNA 3′-ends likely involves a dynamic assembly of CPF/CPSF and accessory factors.

The polyadenosine (poly(A)) tail found on the 3′-end of almost all eukaryotic mRNAs is important for mRNA stability and regulation of translation. mRNA 3′-end processing occurs co-transcriptionally and involves more than 20 proteins to specifically recognize the polyadenylation site, cleave the pre-mRNA, add a poly(A) tail, and trigger transcription termination. The polyadenylation site (PAS) defines the end of the 3′-untranslated region (3′-UTR) and, therefore, selection of the cleavage site is a critical event in regulating gene expression. Integrated structural biology approaches including biochemical reconstitution of multi-subunit complexes, cross-linking mass spectrometry, and structural analyses by X- ray crystallography and single-particle electron cryo-microscopy (cryoEM) have enabled recent progress in understanding the molecular mechanisms of the mRNA 3′-end processing machinery. Here, we describe new molecular insights into pre-mRNA recognition, cleavage and polyadenylation.

Modeling of Genome-Wide Polyadenylation Signals in Xenopus tropicalis

Alternative polyadenylation (APA) is an important post-transcriptional modification event to process messenger RNA (mRNA) for transcriptional termination, transport, and translation. In the present study, we characterized poly(A) signals in Xenopus tropicalis using 70,918 highly confident poly(A) sites derived from 16,511 protein-coding genes to understand their roles in the regulation of embryo development and gender difference. We examined potential factors, including the gene length, the number of introns in a gene, and the intron length, that may affect the prevalence of APA. We observed 12 prominent poly(A) signal patterns, which accounted for approximately 92% of total APA sites in Xenopus tropicalis. Among them, three patterns are specific to X. tropicalis, so they are absent in other animals such as humans or mice. We catalogued APA sites based on their genomic regions and developed a bioinformatics pipeline to identify over-represented signal patterns for each class. Then the schema of cis elements for APA sites in each genomic region was proposed. More importantly, APA usage is dramatically dynamic in embryos along five developmental stages and well-coordinated with the maternal-to-zygotic transition event. We used an entropy-based method to identify developmental stage-specific APA sites and identified significant signal patterns around specific sites and constitutive sites. We found that the APA frequency in different genomic regions varies with developmental stages and that those sites located in intron or coding sequence regions contribute most to the dynamics of gene expression during developmental stages. This study deciphers the characteristics and poly(A) signal patterns for both canonical APA sites and non-canonical APA sites across different developmental stages and gender dimorphisms in X. tropicalis, providing new insights into the dynamic regulation of distal and proximal APA.

A Deep Neural Network for Predicting and Engineering Alternative Polyadenylation

Alternative polyadenylation (APA) is a major driver of transcriptome diversity in human cells. Here, we use deep learning to predict APA from DNA sequence alone. We trained our model (APARENT, APA REgression NeT) on isoform expression data from over 3 million APA reporters. APARENT's predictions are highly accurate when tasked with inferring APA in synthetic and human 3'UTRs. Visualizing features learned across all network layers reveals that APARENT recognizes sequence motifs known to recruit APA regulators, discovers previously unknown sequence determinants of 3' end processing, and integrates these features into a comprehensive, interpretable, cis-regulatory code. We apply APARENT to forward engineer functional polyadenylation signals with precisely defined cleavage position and isoform usage and validate predictions experimentally. Finally, we use APARENT to quantify the impact of genetic variants on APA. Our approach detects pathogenic variants in a wide range of disease contexts, expanding our understanding of the genetic origins of disease.

Regulation of Intronic Polyadenylation by PCF11 Impacts mRNA Expression of Long Genes

Regulation of cleavage and polyadenylation (CPA) affects gene expression and polyadenylation site (PAS) choice. Here, we report that the CPA and termination factor PCF11 modulates gene expression on the basis of gene size. Although downregulation of PCF11 leads to inhibition of short gene expression, long genes are upregulated because of suppressed intronic polyadenylation (IPA) enriched in large introns. We show that this regulatory scheme, named PCF11-mediated expression regulation through IPA (PEIPA), takes place in cell differentiation, during which downregulation of PCF11 is coupled with upregulation of long genes with functions in cell morphology, adhesion, and migration. PEIPA targets distinct gene sets in different cell contexts with similar rules. Furthermore, PCF11 is autoregulated through a conserved IPA site, the removal of which leads to global activation of PASs close to gene promotors. Therefore, PCF11 uses distinct mechanisms to regulate genes of different sizes, and its autoregulation maintains homeostasis of PAS usage in the cell.

Long-read sequencing uncovers a complex transcriptome topology in varicella zoster virus

BACKGROUND

Varicella zoster virus (VZV) is a human pathogenic alphaherpesvirus harboring a relatively large DNA molecule. The VZV transcriptome has already been analyzed by microarray and short-read sequencing analyses. However, both approaches have substantial limitations when used for structural characterization of transcript isoforms, even if supplemented with primer extension or other techniques. Among others, they are inefficient in distinguishing between embedded RNA molecules, transcript isoforms, including splice and length variants, as well as between alternative polycistronic transcripts. It has been demonstrated in several studies that long-read sequencing is able to circumvent these problems.

RESULTS

In this work, we report the analysis of the VZV lytic transcriptome using the Oxford Nanopore Technologies sequencing platform. These investigations have led to the identification of 114 novel transcripts, including mRNAs, non-coding RNAs, polycistronic RNAs and complex transcripts, as well as 10 novel spliced transcripts and 25 novel transcription start site isoforms and transcription end site isoforms. A novel class of transcripts, the nroRNAs are described in this study. These transcripts are encoded by the genomic region located in close vicinity to the viral replication origin. We also show that the ORF63 exhibits a complex structural variation encompassing the splice sites of VZV latency transcripts. Additionally, we have detected RNA editing in a novel non-coding RNA molecule.

CONCLUSIONS

Our investigations disclosed a composite transcriptomic architecture of VZV, including the discovery of novel RNA molecules and transcript isoforms, as well as a complex meshwork of transcriptional read-throughs and overlaps. The results represent a substantial advance in the annotation of the VZV transcriptome and in understanding the molecular biology of the herpesviruses in general.

Reconstitution of mammalian cleavage factor II involved in 3' processing of mRNA precursors

Cleavage factor II (CF II) is a poorly characterized component of the multiprotein complex catalyzing 3' cleavage and polyadenylation of mammalian mRNA precursors. We have reconstituted CF II as a heterodimer of hPcf11 and hClp1. The heterodimer is active in partially reconstituted cleavage reactions, whereas hClp1 by itself is not. Pcf11 moderately stimulates the RNA 5' kinase activity of hClp1; the kinase activity is dispensable for RNA cleavage. CF II binds RNA with nanomolar affinity. Binding is mediated mostly by the two zinc fingers in the C-terminal region of hPcf11. RNA is bound without pronounced sequence-specificity, but extended G-rich sequences appear to be preferred. We discuss the possibility that CF II contributes to the recognition of cleavage/polyadenylation substrates through interaction with G-rich far-downstream sequence elements.

A compendium of conserved cleavage and polyadenylation events in mammalian genes

Cleavage and polyadenylation is essential for 3' end processing of almost all eukaryotic mRNAs. Recent studies have shown widespread alternative cleavage and polyadenylation (APA) events leading to mRNA isoforms with different 3' UTRs and/or coding sequences. Here, we present a compendium of conserved cleavage and polyadenylation sites (PASs) in mammalian genes, based on approximately 1.2 billion 3' end sequencing reads from more than 360 human, mouse, and rat samples. We show that ∼80% of mammalian mRNA genes contain at least one conserved PAS, and ∼50% have conserved APA events. PAS conservation generally reduces promiscuous 3' end processing, stabilizing gene expression levels across species. Conservation of APA correlates with gene age, gene expression features, and gene functions. Genes with certain functions, such as cell morphology, cell proliferation, and mRNA metabolism, are particularly enriched with conserved APA events. Whereas tissue-specific genes typically have a low APA rate, brain-specific genes tend to evolve APA. In addition, we show enrichment of mRNA destabilizing motifs in alternative 3' UTR sequences, leading to substantial differences in mRNA stability between 3' UTR isoforms. Using conserved PASs, we reveal sequence motifs surrounding APA sites and a preference of adenosine at the cleavage site. Furthermore, we show that mutations of U-rich motifs around the PAS often accompany APA profile differences between species. Analysis of lncRNA PASs indicates a mechanism of PAS fixation through evolution of A-rich motifs. Taken together, our results present a comprehensive view of PAS evolution in mammals, and a phylogenic perspective on APA functions.

A compendium of conserved cleavage and polyadenylation events in mammalian genes

Cleavage and polyadenylation is essential for 3' end processing of almost all eukaryotic mRNAs. Recent studies have shown widespread alternative cleavage and polyadenylation (APA) events leading to mRNA isoforms with different 3' UTRs and/or coding sequences. Here, we present a compendium of conserved cleavage and polyadenylation sites (PASs) in mammalian genes, based on approximately 1.2 billion 3' end sequencing reads from more than 360 human, mouse, and rat samples. We show that ∼80% of mammalian mRNA genes contain at least one conserved PAS, and ∼50% have conserved APA events. PAS conservation generally reduces promiscuous 3' end processing, stabilizing gene expression levels across species. Conservation of APA correlates with gene age, gene expression features, and gene functions. Genes with certain functions, such as cell morphology, cell proliferation, and mRNA metabolism, are particularly enriched with conserved APA events. Whereas tissue-specific genes typically have a low APA rate, brain-specific genes tend to evolve APA. In addition, we show enrichment of mRNA destabilizing motifs in alternative 3' UTR sequences, leading to substantial differences in mRNA stability between 3' UTR isoforms. Using conserved PASs, we reveal sequence motifs surrounding APA sites and a preference of adenosine at the cleavage site. Furthermore, we show that mutations of U-rich motifs around the PAS often accompany APA profile differences between species. Analysis of lncRNA PASs indicates a mechanism of PAS fixation through evolution of A-rich motifs. Taken together, our results present a comprehensive view of PAS evolution in mammals, and a phylogenic perspective on APA functions.

Targeting the Polyadenylation Signal of Pre-mRNA: A New Gene Silencing Approach for Facioscapulohumeral Dystrophy

Facioscapulohumeral dystrophy (FSHD) is characterized by the contraction of the D4Z4 array located in the sub-telomeric region of the chromosome 4, leading to the aberrant expression of the DUX4 transcription factor and the mis-regulation of hundreds of genes. Several therapeutic strategies have been proposed among which the possibility to target the polyadenylation signal to silence the causative gene of the disease. Indeed, defects in mRNA polyadenylation leads to an alteration of the transcription termination, a disruption of mRNA transport from the nucleus to the cytoplasm decreasing the mRNA stability and translation efficiency. This review discusses the polyadenylation mechanisms, why alternative polyadenylation impacts gene expression, and how targeting polyadenylation signal may be a potential therapeutic approach for FSHD.

Alu exaptation enriches the human transcriptome by introducing new gene ends

In mammals, transposable elements are largely silenced, but under fortuitous circumstances may be co-opted to play a functional role. Here, we show that when Alu elements are inserted within or nearby genes in sense orientation, they may contribute to the transcriptome diversity by forming new cleavage and polyadenylation sites. We mapped these new gene ends in human onto the Alu sequence and identified three hotspots of cleavage and polyadenylation site formation. Interestingly, the native Alu sequence does not contain any canonical polyadenylation signal. We therefore studied what evolutionary processes might explain the formation of these specific hotspots of novel gene ends. We show that two of the three hotspots might have emerged from mutational processes that turned sequences that resemble polyadenylation signals into full-blown canonical signals, whereas one hotspot is tightly linked to the process of Alu insertion into the genome. Overall, Alu elements may lie behind the formation of 302 new gene end variants, affecting a total of 243 genes. Intergenic Alu elements may elongate genes by creating a downstream cleavage site, intronic Alu elements may lead to gene variants which code for truncated proteins, and 3'UTR Alu elements may result in gene variants with alternative 3'UTR.

Structural basis of AAUAAA polyadenylation signal recognition by the human CPSF complex

Mammalian mRNA biogenesis requires specific recognition of a hexanucleotide AAUAAA motif in the polyadenylation signals (PAS) of precursor mRNA (pre-mRNA) transcripts by the cleavage and polyadenylation specificity factor (CPSF) complex. Here we present a 3.1-Å-resolution cryo-EM structure of a core CPSF module bound to the PAS hexamer motif. The structure reveals the molecular interactions responsible for base-specific recognition, providing a rationale for mechanistic differences between mammalian and yeast 3' polyadenylation.

Useful Bicistronic Reporter System for Studying Poly(A) Site-Defining cis Elements and Regulation of Alternative Polyadenylation

The link between polyadenylation (pA) and various biological, behavioral, and pathological events of eukaryotes underlines the need to develop in vivo polyadenylation assay methods for characterization of the cis-acting elements, trans-acting factors and environmental stimuli that affect polyadenylation efficiency and/or relative usage of two alternative polyadenylation (APA) sites. The current protein-based CAT or luciferase reporter systems can measure the polyadenylation efficiency of a single pA site or candidate cis element but not the choice of two APA sites. To address this issue, we developed a set of four new bicistronic reporter vectors that harbor either two luciferase or fluorescence protein open reading frames connected with one Internal Ribosome Entry Site (IRES). Transfection of single or dual insertion constructs of these vectors into mammalian cells demonstrated that they could be utilized not only to quantify the strength of a single candidate pA site or cis element, but also to accurately measure the relative usage of two APA sites at both the mRNA (qRT-PCR) and protein levels. This represents the first reporter system that can study polyadenylation efficiency of a single pA site or element and regulation of two APA sites at both the mRNA and protein levels.