N1-methylation of adenosine (m1A) in ND5 mRNA leads to complex I dysfunction in Alzheimer's disease.

One mechanism of particular interest to regulate mRNA fate post-transcriptionally is mRNA modification. Especially the extent of m1A mRNA methylation is highly discussed due to methodological differences. However, one single m1A site in mitochondrial ND5 mRNA was unanimously reported by different groups. ND5 is a subunit of complex I of the respiratory chain. It is considered essential for the coupling of oxidation and proton transport. Here we demonstrate that this m1A site might be involved in the pathophysiology of Alzheimer's disease (AD). One of the pathological hallmarks of this neurodegenerative disease is mitochondrial dysfunction, mainly induced by Amyloid ß (Aß). Aß mainly disturbs functions of complex I and IV of the respiratory chain. However, the molecular mechanism of complex I dysfunction is still not fully understood. We found enhanced m1A methylation of ND5 mRNA in an AD cell model as well as in AD patients. Formation of this m1A methylation is catalyzed by increased TRMT10C protein levels, leading to translation repression of ND5. As a consequence, here demonstrated for the first time, TRMT10C induced m1A methylation of ND5 mRNA leads to mitochondrial dysfunction. Our findings suggest that this newly identified mechanism might be involved in Aß-induced mitochondrial dysfunction.

Machine learning algorithm for precise prediction of 2'-O-methylation (Nm) sites from experimental RiboMethSeq datasets.

Analysis of epitranscriptomic RNA modifications by deep sequencing-based approaches brings an essential contribution to the general knowledge on their precise locations and relative stoichiometry in cellular RNAs. To reveal RNA modifications, several analytical approaches have been proposed, including antibody-driven enrichment, analysis of RT-signatures and specific chemical treatments. However, analysis and interpretation of these massive datasets, especially for low abundant cellular RNAs (e.g. mRNA and lncRNA) is not easy nor straightforward, since the insufficient specificity and selectivity are leading to massive false-positive and false-negative identifications. The main issue in the application of these methods relies on a subjective classification of potentially modified positions, mostly based on arbitrarily defined threshold values for different scores. Such approach using pre-defined scores' values was revealed to be appropriate for limited complexity datasets (for tRNA and/or rRNA analysis), but application to longer reference sequences requires much better classification algorithms. In this work we applied a machine learning algorithm (Random Forest, RF) to create a predictive model for analysis of 2'-O-methylated sites in RNA using RiboMethSeq datasets. Model's training was performed on a large collection of human rRNA datasets with well-known modification profiles and the performance of the prediction was assessed using experimentally defined profiles for other eukaryotic rRNAs (S.cerevisiae and A.thaliana). Application of this Random Forest prediction model for detection of other RNA modifications and to more complex datasets is discussed.

NOseq: amplicon sequencing evaluation method for RNA m6A sites after chemical deamination.

Methods for the detection of m6A by RNA-Seq technologies are increasingly sought after. We here present NOseq, a method to detect m6A residues in defined amplicons by virtue of their resistance to chemical deamination, effected by nitrous acid. Partial deamination in NOseq affects all exocyclic amino groups present in nucleobases and thus also changes sequence information. The method uses a mapping algorithm specifically adapted to the sequence degeneration caused by deamination events. Thus, m6A sites with partial modification levels of ∼50% were detected in defined amplicons, and this threshold can be lowered to ∼10% by combination with m6A immunoprecipitation. NOseq faithfully detected known m6A sites in human rRNA, and the long non-coding RNA MALAT1, and positively validated several m6A candidate sites, drawn from miCLIP data with an m6A antibody, in the transcriptome of Drosophila melanogaster. Conceptually related to bisulfite sequencing, NOseq presents a novel amplicon-based sequencing approach for the validation of m6A sites in defined sequences.

HydraPsiSeq: a method for systematic and quantitative mapping of pseudouridines in RNA.

Developing methods for accurate detection of RNA modifications remains a major challenge in epitranscriptomics. Next-generation sequencing-based mapping approaches have recently emerged but, often, they are not quantitative and lack specificity. Pseudouridine (ψ), produced by uridine isomerization, is one of the most abundant RNA modification. ψ mapping classically involves derivatization with soluble carbodiimide (CMCT), which is prone to variation making this approach only semi-quantitative. Here, we developed 'HydraPsiSeq', a novel quantitative ψ mapping technique relying on specific protection from hydrazine/aniline cleavage. HydraPsiSeq is quantitative because the obtained signal directly reflects pseudouridine level. Furthermore, normalization to natural unmodified RNA and/or to synthetic in vitro transcripts allows absolute measurements of modification levels. HydraPsiSeq requires minute amounts of RNA (as low as 10-50 ng), making it compatible with high-throughput profiling of diverse biological and clinical samples. Exploring the potential of HydraPsiSeq, we profiled human rRNAs, revealing strong variations in pseudouridylation levels at ∼20-25 positions out of total 104 sites. We also observed the dynamics of rRNA pseudouridylation throughout chondrogenic differentiation of human bone marrow stem cells. In conclusion, HydraPsiSeq is a robust approach for the systematic mapping and accurate quantification of pseudouridines in RNAs with applications in disease, aging, development, differentiation and/or stress response.

2'-O-methylation within prokaryotic and eukaryotic tRNA inhibits innate immune activation by endosomal Toll-like receptors but does not affect recognition of whole organisms.

Bacterial RNA has emerged as an important activator of innate immune responses by stimulating Toll-like receptors TLR7 and TLR8 in humans. Guanosine 2'-O-methylation at position 18 (Gm18) in bacterial tRNA was shown to antagonize tRNA-induced TLR7/8 activation, suggesting a potential role of Gm18 as an immune escape mechanism. This modification also occurs in eukaryotic tRNA, yet a physiological immune function remained to be tested. We therefore set out to investigate the immune modulatory role of Gm18 in both prokaryotic and eukaryotic microorganisms, Escherichia coli and Saccharomyces cerevisiae, and in human cells. Using RiboMethSeq analysis we show that mutation of trmH in E. coli, trm3 in S. cereviase, and CRISPR/Cas9-induced knockout of TARBP1 in H. sapiens results in loss of Gm18 within tRNA. Lack of Gm18 across the kingdoms resulted in increased immunostimulation of peripheral blood mononuclear cells when activated by tRNA preparations. In E. coli, lack of 2'-O-methyltransferase trmH also enhanced immune stimulatory properties by whole cellular RNA. In contrast, lack of Gm18 in yeasts and human cells did not affect immunostimulation by whole RNA preparations. When using live E. coli bacteria, lack of trmH did not affect overall immune stimulation although we detected a defined TLR8/RNA-dependent gene expression signature upon E. coli infection. Together, these results demonstrate that Gm18 is a global immune inhibitory tRNA modification across the kingdoms and contributes to tRNA recognition by innate immune cells, but as an individual modification has insufficient potency to modulate recognition of the investigated microorganisms.

Data Analysis Pipeline for Detection and Quantification of Pseudouridine (ψ) in RNA by HydraPsiSeq.

Pseudouridine, a modified RNA residue formed by the isomerization of its parental U nucleotide, is prevalent in a majority of cellular RNAs; its presence was reported in tRNA, rRNA, and sn/snoRNA as well as in mRNA/lncRNA. Multiple analytical deep sequencing-based approaches have been proposed for pseudouridine detection and quantification, among which the most popular relies on the use of soluble carbodiimide (termed CMCT). Recently, we developed an alternative protocol for pseudouridine mapping and quantification. The principle is based on protection of pseudouridine against random RNA cleavage by hydrazine/aniline treatment (HydraPsiSeq protocol). This "negative" detection mode requires higher sequencing depth and provides a precise quantification of the pseudouridine content. All "wet-lab" technical details of the HydraPsiSeq protocol have been described in recent publications. Here, we describe all bioinformatics analysis steps required for data processing from raw reads to the pseudouridylation profile of known or unknown RNA.

Quantification of substoichiometric modification reveals global tsRNA hypomodification, preferences for angiogenin-mediated tRNA cleavage, and idiosyncratic epitranscriptomes of human neuronal cell-lines.

Modification of tRNA is an integral part of the epitranscriptome with a particularly pronounced potential to generate diversity in RNA expression. Eukaryotic tRNA contains modifications in up to 20% of their nucleotides, but not all sites are always fully modified. Combinations and permutations of partially modified sites in tRNAs can generate a plethora of tRNA isoforms, termed modivariants. Here, we investigate the stoichiometry of incompletely modified sites in tRNAs from human cell lines for their information content. Using a panel of RNA modification mapping methods, we assess the stoichiometry of sites that contain the modifications 5-methylcytidine (m5C), 2'-O-ribose methylation (Nm), 3-methylcytidine (m3C), 7-methylguanosine (m7G), and Dihydrouridine (D). We discovered that up to 75% of sites can be incompletely modified and that the differential modification status of a cellular tRNA population holds information that allows to discriminate e.g. different cell lines. As a further aspect, we investigated potential causal connectivity between tRNA modification and its processing into tRNA fragments (tiRNAs and tRFs). Upon exposure of cultured living cells to cell-penetrating angiogenin, the modification patterns of the corresponding RNA populations was changed. Importantly, we also found that tsRNAs were significantly less modified than their parent tRNAs at numerous sites, suggesting that tsRNAs might derive chiefly from hypomodified tRNAs.

Non-Redundant tRNA Reference Sequences for Deep Sequencing Analysis of tRNA Abundance and Epitranscriptomic RNA Modifications.

Analysis of RNA by deep-sequencing approaches has found widespread application in modern biology. In addition to measurements of RNA abundance under various physiological conditions, such techniques are now widely used for mapping and quantification of RNA modifications. Transfer RNA (tRNA) molecules are among the frequent targets of such investigation, since they contain multiple modified residues. However, the major challenge in tRNA examination is related to a large number of duplicated and point-mutated genes encoding those RNA molecules. Moreover, the existence of multiple isoacceptors/isodecoders complicates both the analysis and read mapping. Existing databases for tRNA sequencing provide near exhaustive listings of tRNA genes, but the use of such highly redundant reference sequences in RNA-seq analyses leads to a large number of ambiguously mapped sequencing reads. Here we describe a relatively simple computational strategy for semi-automatic collapsing of highly redundant tRNA datasets into a non-redundant collection of reference tRNA sequences. The relevance of the approach was validated by analysis of experimentally obtained tRNA-sequencing datasets for different prokaryotic and eukaryotic model organisms. The data demonstrate that non-redundant tRNA reference sequences allow improving unambiguous mapping of deep sequencing data.

Holistic Optimization of Bioinformatic Analysis Pipeline for Detection and Quantification of 2'-O-Methylations in RNA by RiboMethSeq.

A major trend in the epitranscriptomics field over the last 5 years has been the high-throughput analysis of RNA modifications by a combination of specific chemical treatment(s), followed by library preparation and deep sequencing. Multiple protocols have been described for several important RNA modifications, such as 5-methylcytosine (m5C), pseudouridine (ψ), 1-methyladenosine (m1A), and 2'-O-methylation (Nm). One commonly used method is the alkaline cleavage-based RiboMethSeq protocol, where positions of reads' 5'-ends are used to distinguish nucleotides protected by ribose methylation. This method was successfully applied to detect and quantify Nm residues in various RNA species such as rRNA, tRNA, and snRNA. Such applications require adaptation of the initially published protocol(s), both at the wet bench and in the bioinformatics analysis. In this manuscript, we describe the optimization of RiboMethSeq bioinformatics at the level of initial read treatment, alignment to the reference sequence, counting the 5'- and 3'- ends, and calculation of the RiboMethSeq scores, allowing precise detection and quantification of the Nm-related signal. These improvements introduced in the original pipeline permit a more accurate detection of Nm candidates and a more precise quantification of Nm level variations. Applications of the improved RiboMethSeq treatment pipeline for different cellular RNA types are discussed.

RNA ribose methylation (2'-O-methylation): Occurrence, biosynthesis and biological functions.

Methylation of riboses at 2'-OH group is one of the most common RNA modifications found in number of cellular RNAs from almost any species which belong to all three life domains. This modification was extensively studied for decades in rRNAs and tRNAs, but recent data revealed the presence of 2'-O-methyl groups also in low abundant RNAs, like mRNAs. Ribose methylation is formed in RNA by two alternative enzymatic mechanisms: either by stand-alone protein enzymes or by complex assembly of proteins associated with snoRNA guides (sno(s)RNPs). In that case one catalytic subunit acts at various RNA sites, the specificity is provided by base pairing of the sno(s)RNA guide with the target RNA. In this review we compile available information on 2'-OH ribose methylation in different RNAs, enzymatic machineries involved in their biosynthesis and dynamics, as well as on the physiological functions of these modified residues.

Next-Generation Sequencing-Based RiboMethSeq Protocol for Analysis of tRNA 2'-O-Methylation.

Analysis of RNA modifications by traditional physico-chemical approaches is labor  intensive,  requires  substantial  amounts  of  input  material  and  only  allows  site-by-site  measurements.  The  recent  development  of  qualitative  and  quantitative  approaches  based  on   next-generation sequencing (NGS) opens new perspectives for the analysis of various cellular RNA  species.  The  Illumina  sequencing-based  RiboMethSeq  protocol  was  initially  developed  and  successfully applied for mapping of ribosomal RNA (rRNA) 2'-O-methylations. This method also  gives excellent results in the quantitative analysis of rRNA modifications in different species and  under varying growth conditions. However, until now, RiboMethSeq was only employed for rRNA,  and the whole sequencing and analysis pipeline was only adapted to this long and rather conserved  RNA species. A deep understanding of RNA modification functions requires large and global  analysis datasets for other important RNA species, namely for transfer RNAs (tRNAs), which are  well known to contain a great variety of functionally-important modified residues. Here, we  evaluated the application of the RiboMethSeq protocol for the analysis of tRNA 2'-O-methylation in  Escherichia coli and in Saccharomyces cerevisiae. After a careful optimization of the bioinformatic  pipeline, RiboMethSeq proved to be suitable for relative quantification of methylation rates for  known modified positions in different tRNA species.