Challenges in microRNAs' targetome prediction and validation

Mitigating off-target effects of small RNAs: conventional approaches, network theory and artificial intelligence

Three types of highly promising small RNA therapeutics, namely, small interfering RNAs (siRNAs), microRNAs (miRNAs) and the RNA subtype of antisense oligonucleotides (ASOs), offer advantages over small-molecule drugs. These small RNAs can target any gene product, opening up new avenues of effective and safe therapeutic approaches for a wide range of diseases. In preclinical research, synthetic small RNAs play an essential role in the investigation of physiological and pathological pathways as silencers of specific genes, facilitating discovery and validation of drug targets in different conditions. Off-target effects of small RNAs, however, could make it difficult to interpret experimental results in the preclinical phase and may contribute to adverse events of small RNA therapeutics. Out of the two major types of off-target effects we focused on the hybridization-dependent, especially on the miRNA-like off-target effects. Our main aim was to discuss several approaches, including sequence design, chemical modifications and target prediction, to reduce hybridization-dependent off-target effects that should be considered even at the early development phase of small RNA therapy. Because there is no standard way of predicting hybridization-dependent off-target effects, this review provides an overview of all major state-of-the-art computational methods and proposes new approaches, such as the possible inclusion of network theory and artificial intelligence (AI) in the prediction workflows. Case studies and a concise survey of experimental methods for validating in silico predictions are also presented. These methods could contribute to interpret experimental results, to minimize off-target effects and hopefully to avoid off-target-related adverse events of small RNA therapeutics. LINKED ARTICLES: This article is part of a themed issue Non-coding RNA Therapeutics. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v182.2/issuetoc.

Identification and characterization of host miRNAs that target the mouse mammary tumour virus (MMTV) genome

The intricate interplay between viruses and hosts involves microRNAs (miRNAs) to regulate gene expression by targeting cellular/viral messenger RNAs (mRNAs). Mouse mammary tumour virus (MMTV), the aetiological agent of breast cancer and leukaemia/lymphomas in mice, provides an ideal model to explore how viral and host miRNAs interact to modulate virus replication and tumorigenesis. We previously reported dysregulation of host miRNAs in MMTV-infected mammary glands and MMTV-induced tumours, suggesting a direct interaction between MMTV and miRNAs. To explore this further, we systematically examined all potential interactions between host miRNAs and the MMTV genome using advanced prediction tools. Leveraging miRNA sequencing data from MMTV-expressing cells, we identified dysregulated miRNAs capable of targeting MMTV. Docking analysis validated the interaction of three dysregulated miRNAs with the MMTV genome, followed by confirmation with RNA immunoprecipitation assays. We further identified host targets of these miRNAs using mRNA sequencing data from MMTV-expressing cells. These findings should enhance our understanding of how MMTV replicates and interacts with the host to induce cancer in mice, a model important for cancer research. Given MMTV's potential zoonosis and association with human breast cancer/lymphomas, if confirmed, our work could further lead to novel miRNA-based antivirals/therapeutics to prevent possible MMTV transmission and associated cancers in humans.

Gra-CRC-miRTar: The pre-trained nucleotide-to-graph neural networks to identify potential miRNA targets in colorectal cancer

Colorectal cancer (CRC) is the third most diagnosed cancer and the second deadliest cancer worldwide representing a major public health problem. In recent years, increasing evidence has shown that microRNA (miRNA) can control the expression of targeted human messenger RNA (mRNA) by reducing their abundance or translation, acting as oncogenes or tumor suppressors in various cancers, including CRC. Due to the significant up-regulation of oncogenic miRNAs in CRC, elucidating the underlying mechanism and identifying dysregulated miRNA targets may provide a basis for improving current therapeutic interventions. In this paper, we proposed Gra-CRC-miRTar, a pre-trained nucleotide-to-graph neural network framework, for identifying potential miRNA targets in CRC. Different from previous studies, we constructed two pre-trained models to encode RNA sequences and transformed them into de Bruijn graphs. We employed different graph neural networks to learn the latent representations. The embeddings generated from de Bruijn graphs were then fed into a Multilayer Perceptron (MLP) to perform the prediction tasks. Our extensive experiments show that Gra-CRC-miRTar achieves better performance than other deep learning algorithms and existing predictors. In addition, our analyses also successfully revealed 172 out of 201 functional interactions through experimentally validated miRNA-mRNA pairs in CRC. Collectively, our effort provides an accurate and efficient framework to identify potential miRNA targets in CRC, which can also be used to reveal miRNA target interactions in other malignancies, facilitating the development of novel therapeutics. The Gra-CRC-miRTar web server can be found at: http://gra-crc-mirtar.com/.

Gra-CRC-miRTar: The pre-trained nucleotide-to-graph neural networks to identify potential miRNA targets in colorectal cancer

Colorectal cancer (CRC) is the third most diagnosed cancer and the second deadliest cancer worldwide representing a major public health problem. In recent years, increasing evidence has shown that microRNA (miRNA) can control the expression of targeted human messenger RNA (mRNA) by reducing their abundance or translation, acting as oncogenes or tumor suppressors in various cancers, including CRC. Due to the significant up-regulation of oncogenic miRNAs in CRC, elucidating the underlying mechanism and identifying dysregulated miRNA targets may provide a basis for improving current therapeutic interventions. In this paper, we proposed Gra-CRC-miRTar, a pre-trained nucleotide-to-graph neural network framework, for identifying potential miRNA targets in CRC. Different from previous studies, we constructed two pre-trained models to encode RNA sequences and transformed them into de Bruijn graphs. We employed different graph neural networks to learn the latent representations. The embeddings generated from de Bruijn graphs were then fed into a Multilayer Perceptron (MLP) to perform the prediction tasks. Our extensive experiments show that Gra-CRC-miRTar achieves better performance than other deep learning algorithms and existing predictors. In addition, our analyses also successfully revealed 172 out of 201 functional interactions through experimentally validated miRNA-mRNA pairs in CRC. Collectively, our effort provides an accurate and efficient framework to identify potential miRNA targets in CRC, which can also be used to reveal miRNA target interactions in other malignancies, facilitating the development of novel therapeutics.

MicroRNAs and long non-coding RNAs during transcriptional regulation and latency of HIV and HTLV

Human immunodeficiency virus (HIV) and human T cell leukemia virus (HTLV) have replicative and latent stages of infection. The status of the viruses is dependent on the cells that harbour them and on different events that change the transcriptional and post-transcriptional events. Non-coding (nc)RNAs are key factors in the regulation of retrovirus replication cycles. Notably, micro (mi)RNAs and long non-coding (lnc)RNAs are important regulators that can induce switches between active transcription-replication and latency of retroviruses and have important impacts on their pathogenesis. Here, we review the functions of miRNAs and lncRNAs in the context of HIV and HTLV. We describe how specific miRNAs and lncRNAs are involved in the regulation of the viruses' transcription, post-transcriptional regulation and latency. We further discuss treatment strategies using ncRNAs for HIV and HTLV long remission, reactivation or possible cure.

Impact of MICA 3'UTR allelic variability on miRNA binding prediction, a bioinformatic approach

MicroRNAs (miRNAs) are small non-coding RNAs that participate as powerful genetic regulators. MiRNAs can interfere with cellular processes by interacting with a broad spectrum of target genes under physiological and pathological states, including cancer development and progression. Major histocompatibility complex major histocompatibility complex class I-related chain A (MICA) belongs to a family of proteins that bind the natural-killer group 2, member D (NKG2D) receptor on Natural Killer cells and other cytotoxic lymphocytes. MICA plays a crucial role in the host's innate immune response to several disease settings, including cancer. MICA harbors various single nucleotide polymorphisms (SNPs) located in its 3'-untranslated region (3'UTR), a characteristic that increases the complexity of MICA regulation, favoring its post-transcriptional modulation by miRNAs under physiological and pathological conditions. Here, we conducted an in-depth analysis of MICA 3'UTR sequences according to each MICA allele described to date using NCBI database. We also systematically evaluated interactions between miRNAs and their putative targets on MICA 3'UTR containing SNPs using in silico analysis. Our in silico results showed that MICA SNPs rs9266829, rs 1880, and rs9266825, located in the target sequence of miRNAs hsa-miR-106a-5p, hsa-miR-17-5p, hsa-miR-20a-5p, hsa-miR-20b-5p, hsa-miR-93, hsa-miR-1207.5p, and hsa-miR-711 could modify the binding free energy between -8.62 and -18.14 kcal/mol, which may affect the regulation of MICA expression. We believe that our results may provide a starting point for further exploration of miRNA regulatory effects depending on MICA allelic variability; they may also be a guide to conduct miRNA in silico analysis for other highly polymorphic genes.

MicroRNA-Target Identification: A Combinatorial In Silico Approach

Contemporary computational target prediction tools with their distinctive properties and stringency have been playing a vital role in pursuing putative targets for a solitary miRNA or a subcategory of miRNAs. These tools utilize a defined set of probabilistic algorithms, machine learning techniques, and information of experimentally validated miRNA targets to provide the best selection. However, there are numerous false-positive predictions, and a method to choose an archetypal approach and put the data provided by the prediction tools into context is still lacking. Moreover, sensitivity, specificity, and overall efficiency of a single tool have not yet been achieved. Therefore, a systematic combination of selective online tools combining elementary and advanced factors of miRNA target identification might reinforce the current target prediction regime. The focus of this study was to build a comprehensive workflow by combining six available online tools to facilitate the current understanding of miRNA-target prediction.

Comprehensive transcriptomic analysis of two RIL parents with contrasting salt responsiveness identifies polyadenylated and non-polyadenylated flower lncRNAs in chickpea

Salinity severely affects the yield of chickpea. Understanding the role of lncRNAs can shed light on chickpea salt tolerance mechanisms. However, because lncRNAs are encoded by multiple sites within the genome, their classification to reveal functional versatility at the transcriptional and the post-transcriptional levels is challenging. To address this, we deep sequenced 24 salt-challenged flower transcriptomes from two parental genotypes of a RIL population that significantly differ in salt tolerance ability. The transcriptomes for the first time included 12 polyadenylated and 12 non-polyadenylated RNA libraries to a sequencing depth of ~50 million reads. The ab initio transcriptome assembly comprised ~34 082 transcripts from three biological replicates of salt-tolerant (JG11) and salt-sensitive (ICCV2) flowers. A total of 9419 lncRNAs responding to salt stress were identified, 2345 of which were novel lncRNAs specific to chickpea. The expression of poly(A+) lncRNAs and naturally antisense transcribed RNAs suggest their role in post-transcriptional modification and gene silencing. Notably, 178 differentially expressed lncRNAs were induced in the tolerant genotype but repressed in the sensitive genotype. Co-expression network analysis revealed that the induced lncRNAs interacted with the FLOWERING LOCUS (FLC), chromatin remodelling and DNA methylation genes, thus inducing flowering during salt stress. Furthermore, 26 lncRNAs showed homology with reported lncRNAs such as COOLAIR, IPS1 and AT4, thus confirming the role of chickpea lncRNAs in controlling flowering time as a crucial salt tolerance mechanism in tolerant chickpea genotype. These robust set of differentially expressed lncRNAs provide a deeper insight into the regulatory mechanisms controlled by lncRNAs under salt stress.

Decoding microRNA drivers in Atherosclerosis

An estimated 97% of the human genome consists of non-protein-coding sequences. As our understanding of genome regulation improves, this has led to the characterization of a diverse array of non-coding RNAs (ncRNA). Among these, micro-RNAs (miRNAs) belong to the short ncRNA class (22–25 nucleotides in length), with approximately 2500 miRNA genes encoded within the human genome. From a therapeutic perspective, there is interest in exploiting miRNA as biomarkers of disease progression and response to treatments, as well as miRNA mimics/repressors as novel medicines. miRNA have emerged as an important class of RNA master regulators with important roles identified in the pathogenesis of atherosclerotic cardiovascular disease. Atherosclerosis is characterized by a chronic inflammatory build-up, driven largely by low-density lipoprotein cholesterol accumulation within the artery wall and vascular injury, including endothelial dysfunction, leukocyte recruitment and vascular remodelling. Conventional therapy focuses on lifestyle interventions, blood pressure-lowering medications, high-intensity statin therapy and antiplatelet agents. However, a significant proportion of patients remain at increased risk of cardiovascular disease. This continued cardiovascular risk is referred to as residual risk. Hence, a new drug class targeting atherosclerosis could synergise with existing therapies to optimise outcomes. Here, we review our current understanding of the role of ncRNA, with a focus on miRNA, in the development and progression of atherosclerosis, highlighting novel biological mechanisms and therapeutic avenues.

Regulation of human UDP-glycosyltransferase (UGT) genes by miRNAs

The human UGT gene superfamily is divided into four subfamilies (UGT1, UGT2, UGT3 and UGT8) that encodes 22 functional enzymes. UGTs are critical for the metabolism and clearance of numerous endogenous and exogenous compounds, including steroid hormones, bile acids, bilirubin, fatty acids, carcinogens, and therapeutic drugs. Therefore, the expression and activities of UGTs are tightly regulated by multiple processes at the transcriptional, post-transcriptional and post-translational levels. During recent years, nearly twenty studies have investigated the post-transcriptional regulation of UGT genes by miRNAs using human cancer cell lines (predominantly liver cancer). Overall, 14 of the 22 UGT mRNAs (1A1, 1A3, 1A4, 1A6, 1A8, 1A9, 1A10, 2A1, 2B4, 2B7, 2B10, 2B15, 2B17, UGT8) have been shown to be regulated by various miRNAs through binding to their respective 3' untranslated regions (3'UTRs). Three 3'UTRs (UGT1A, UGT2B7 and UGT2B15) contain the largest number of functional miRNA target sites; in particular, the UGT1A 3'UTR contains binding sites for 12 miRNAs (548d-5p, 183-5p, 214-5p, 486-3p, 200a-3p, 491-3p, 141-3p, 298, 103b, 376b-3p, 21-3p, 1286). Although all nine UGT1A family members have the same 3'UTR, these miRNA target sites appear to be functional in an isoform-specific and cellular context-dependent manner. Collectively, these observations demonstrate that miRNAs represent important post-transcriptional regulators of the UGT gene superfamily. In this article, we present a comprehensive review of reported UGT/miRNA regulation studies, describe polymorphisms within functional miRNA target sites that may affect their functionalities, and discuss potential cooperative and competitive regulation of UGT mRNAs by miRNAs through adjacently located miRNA target sites.

Role of microRNAs in inflammatory upper airway diseases

MicroRNAs (miRNAs) are a conserved family of small endogenous noncoding RNA molecules that modulate post-transcriptional gene expression in physiological and pathological processes. miRNAs can silence target mRNAs through degradation or inhibition of translation, showing their pivotal role in the pathogenesis of many human diseases. miRNAs play a role in regulating immune functions and inflammation and are implicated in controlling the development and activation of T and B cells. Inflammatory chronic upper airway diseases, such as rhinitis and rhinosinusitis, are spread all over the world and characterized by an exaggerated inflammation involving a complex interaction between immune and resident cells. Until now and despite allergy, little is known about their etiology and the processes implicated in the immune response and tuning inflammation of these diseases. This review highlights the knowledge of the current literature about miRNAs in inflammatory chronic upper airways diseases and how this may be exploited in the development of new clinical and therapeutic strategies.

A combinatorial in silico approach for microRNA-target identification: Order out of chaos

Contemporary computational microRNA(miRNA)-target prediction tools have been playing a vital role in pursuing putative targets for a solitary miRNA or a group of miRNAs. These tools utilise a set of probabilistic algorithms, machine learning techniques and analyse experimentally validated miRNA targets to identify the potential miRNA-target pairs. Unfortunately, current tools generate a huge number of false-positive predictions. A standardized approach with a single tool or a combination of tools is still lacking. Moreover, sensitivity, specificity and overall efficiency of any single tool are yet to be satisfactory. Therefore, a systematic combination of selective online tools combining the factors regarding miRNA-target identification would be valuable as an miRNA-target prediction scheme. The focus of this study was to develop a theoretical framework by combining six available online tools to facilitate the current understanding of miRNA-target identification.

Integrative approaches for analysis of mRNA and microRNA high-throughput data

•

Review on tools and databases linking miRNA and its mRNA targetome.

•

Databases show little overlap in miRNA targetome predictions suggesting strong contextual effects.

•

Deconvolution and deep learning approaches are promising new approaches to improve miRNA targetome predictions.

Advanced sequencing technologies such as RNASeq provide the means for production of massive amounts of data, including transcriptome-wide expression levels of coding RNAs (mRNAs) and non-coding RNAs such as miRNAs, lncRNAs, piRNAs and many other RNA species. In silico analysis of datasets, representing only one RNA species is well established and a variety of tools and pipelines are available. However, attaining a more systematic view of how different players come together to regulate the expression of a gene or a group of genes requires a more intricate approach to data analysis. To fully understand complex transcriptional networks, datasets representing different RNA species need to be integrated. In this review, we will focus on miRNAs as key post-transcriptional regulators summarizing current computational approaches for miRNA:target gene prediction as well as new data-driven methods to tackle the problem of comprehensively and accurately dissecting miRNome-targetome interactions.

MiRNA Regulatory Functions in Photoreceptors

MicroRNAs (miRNAs) are important regulators of gene expression. These small, non-coding RNAs post-transcriptionally silence messenger RNAs (mRNAs) in a sequence-specific manner. In this way, miRNAs control important regulatory functions, also in the retina. If dysregulated, these molecules are involved in several retinal pathologies. For example, several miRNAs have been linked to essential photoreceptor functions, including light sensitivity, synaptic transmission, and modulation of inflammatory responses. Mechanistic miRNA knockout and knockdown studies further linked their functions to degenerative retinal diseases. Of note, the type and timing of genetic manipulation before, during, or after retinal development, is important when studying specific miRNA knockout effects. Within this review, we focus on miR-124 and the miR-183/96/182 cluster, which have assigned functions in photoreceptors in health and disease. As a single miRNA can regulate hundreds of mRNAs, we will also discuss the experimental validation and manipulation approaches to study complex miRNA/mRNA regulatory networks. Revealing these networks is essential to understand retinal pathologies and to harness miRNAs as precise therapeutic and diagnostic tools to stabilize the photoreceptors’ transcriptomes and, thereby, function.

Transcriptomic analysis reveals essential microRNAs after peripheral nerve injury

Studies have shown that microRNAs (miRNAs) mediate posttranscriptional regulation of target genes and participate in various physiological and pathological processes, including peripheral nerve injury. However, it is hard to select key miRNAs with essential biological functions among a large number of differentially expressed miRNAs. Previously, we collected injured sciatic nerve stumps at multiple time points after nerve crush injury, examined gene changes at different stages (acute, sub-acute, and post-acute), and obtained mRNA expression profiles. Here, we jointly analyzed mRNAs and miRNAs, and investigated upstream miRNAs of differentially expressed mRNAs using Ingenuity Pathway Analysis bioinformatic software. A total of 31, 42, 30, and 23 upstream miRNAs were identified at 1, 4, 7, and 14 days after rat sciatic nerve injury, respectively. Temporal expression patterns and biological involvement of commonly involved upstream miRNAs (miR-21, let-7, miR-223, miR-10b, miR-132, miR-15b, miR-127, miR-29a, miR-29b, and miR-9) were then determined at multiple time points. Expression levels of miR-21, miR-132, miR-29a, and miR-29b were robustly increased after sciatic nerve injury. Biological processes involving these miRNAs include multicellular organismal response to stress, positive regulation of the epidermal growth factor receptor signaling pathway, negative regulation of epithelial cell differentiation, and regulation of myocardial tissue growth. Moreover, we constructed mechanistic networks of let-7, miR-21, and miR-223, the most significantly involved upstream miRNAs. Our findings reveal that multiple upstream miRNAs (i.e., let-7, miR-21, and miR-223) were associated with gene expression changes in rat sciatic nerve stumps after nerve injury, and these miRNAs play an important role in peripheral nerve regeneration. This study was approved by the Experimental Animal Ethics Committee of Jiangsu Province of China (approval No. 20190303-18) on March 3, 2019.

Integrated miRNA/mRNA Counter-Expression Analysis Highlights Oxidative Stress-Related Genes CCR7 and FOXO1 as Blood Markers of Coronary Arterial Disease

Our interest in the mechanisms of atherosclerosis progression (ATHp) has led to the recent identification of 13 miRNAs and 1285 mRNAs whose expression was altered during ATHp. Here, we deepen the functional relationship among these 13 miRNAs and genes associated to oxidative stress, a crucial step in the onset and progression of vascular disease. We first compiled a list of genes associated to the response to oxidative stress (Oxstress genes) by performing a reverse Gene Ontology analysis (rGO, from the GO terms to the genes) with the GO terms GO0006979, GO1902882, GO1902883 and GO1902884, which included a total of 417 unique Oxstress genes. Next, we identified 108 putative targets of the 13 miRNAs among these unique Oxstress genes, which were validated by an integrated miRNA/mRNA counter-expression analysis with the 1285 mRNAs that yielded 14 genes, Map2k1, Mapk1, Mapk9, Dapk1, Atp2a2, Gata4, Fos, Egfr, Foxo1, Ccr7, Vkorc1l1, Rnf7, Kcnh3, and Mgat3. GO enrichment analysis and a protein–protein-interaction network analysis (PPI) identified most of the validated Oxstress transcripts as components of signaling pathways, highlighting a role for MAP signaling in ATHp. Lastly, expression of these Oxstress transcripts was measured in PBMCs from patients suffering severe coronary artery disease, a serious consequence of ATHp. This allowed the identification of FOXO1 and CCR7 as blood markers downregulated in CAD. These results are discussed in the context of the interaction of the Oxstress transcripts with the ATHp-associated miRNAs.

Precision Medicine in Neurodegenerative Diseases: Some Promising Tips Coming from the microRNAs' World

: Novel insights in the development of a precision medicine approach for treating the neurodegenerative diseases (NDDs) are provided by emerging advances in the field of pharmacoepigenomics. In this context, microRNAs (miRNAs) have been extensively studied because of their implication in several disorders related to the central nervous system, as well as for their potential role as biomarkers of diagnosis, prognosis, and response to treatment. Recent studies in the field of neurodegeneration reported evidence that drug response and efficacy can be modulated by miRNA-mediated mechanisms. In fact, miRNAs seem to regulate the expression of pharmacology target genes, while approved (conventional and non-conventional) therapies can restore altered miRNAs observed in NDDs. The knowledge of miRNA pharmacoepigenomics may offers new clues to develop more effective treatments by providing novel insights into interindividual variability in drug disposition and response. Recently, the therapeutic potential of miRNAs is gaining increasing attention, and miRNA-based drugs (for cancer) have been under observation in clinical trials. However, the effective use of miRNAs as therapeutic target still needs to be investigated. Here, we report a brief review of representative studies in which miRNAs related to therapeutic effects have been investigated in NDDs, providing exciting potential prospects of miRNAs in pharmacoepigenomics and translational medicine.