PON-P2: Prediction Method for Fast and Reliable Identification of Harmful Variants

More reliable and faster prediction methods are needed to interpret enormous amounts of data generated by sequencing and genome projects. We have developed a new computational tool, PON-P2, for classification of amino acid substitutions in human proteins. The method is a machine learning-based classifier and groups the variants into pathogenic, neutral and unknown classes, on the basis of random forest probability score. PON-P2 is trained using pathogenic and neutral variants obtained from VariBench, a database for benchmark variation datasets. PON-P2 utilizes information about evolutionary conservation of sequences, physical and biochemical properties of amino acids, GO annotations and if available, functional annotations of variation sites. Extensive feature selection was performed to identify 8 informative features among altogether 622 features. PON-P2 consistently showed superior performance in comparison to existing state-of-the-art tools. In 10-fold cross-validation test, its accuracy and MCC are 0.90 and 0.80, respectively, and in the independent test, they are 0.86 and 0.71, respectively. The coverage of PON-P2 is 61.7% in the 10-fold cross-validation and 62.1% in the test dataset. PON-P2 is a powerful tool for screening harmful variants and for ranking and prioritizing experimental characterization. It is very fast making it capable of analyzing large variant datasets. PON-P2 is freely available at http://structure.bmc.lu.se/PON-P2/.     

Pathology-related mutation A7526G (A9G) helps in the understanding of the 3D structural core of human mitochondrial tRNAAsp

More than 130 mutations in human mitochondrial tRNA (mt-tRNA) genes have been correlated with a variety of neurodegenerative and neuromuscular disorders. Their molecular impacts are of mosaic type, affecting various stages of tRNA biogenesis, structure, and/or functions in mt-translation. Knowledge of mammalian mt-tRNA structures per se remains scarce however. Primary and secondary structures deviate from classical tRNAs, while rules for three-dimensional (3D) folding are almost unknown. Here, we take advantage of a myopathy-related mutation A7526G (A9G) in mt-tRNA^Asp to investigate both the primary molecular impact underlying the pathology and the role of nucleotide 9 in the network of 3D tertiary interactions. Experimental evidence is presented for existence of a 9-12-23 triple in human mt-tRNA^Asp with a strongly conserved interaction scheme in mammalian mt-tRNAs. Mutation A7526G disrupts the triple interaction and in turn reduces aspartylation efficiency.     

Nucleotide correlations that suggest tertiary interactions in the TV-replacement loop-containing mitochondrial tRNAs of the nematodes, Caenorhabditis elegans and Ascaris suum.

In the predicted secondary structures of 20 of the 22 tRNAs encoded in mitochondrial DNA (mtDNA) molecules of the nematodes, Caenorhabditis elegans and Ascaris suum, the T psi C arm and variable loop are replaced with a loop of 6 to 12 nucleotides: the TV-replacement loop. From considerations of patterns of nucleotide correlations in the central regions of these tRNAs, it seems highly likely that tertiary interactions occur within five sets of binary and ternary combinations of nucleotides that correspond in location to nucleotides known to be involved in tertiary interactions in yeast tRNA(Phe) and other standard tRNAs. These observations are consistent with the nematode TV-replacement loop-containing mt-tRNAs being folded into a similar L-shaped functional form to that demonstrated for standard tRNAs, and for the bovine DHU (dihydrouridine) arm replacement-loop-containing mt-tRNA(Ser(AGY)). However, the apparent occurrence in nematode mt-tRNAs of tertiary bonds common to standard tRNAs contrasts with the situation in bovine mt-tRNA(Ser(AGY)) where the functional form is dependent on an almost unique set of tertiary interactions. Because three of the proposed conserved tertiary interactions in the nematode mt-tRNAs involve nucleotides that occur in the variable loop in standard tRNAs, it seems more likely that in nematode mt-tRNAs it is the T psi C arm rather than the variable loop that has undergone the greatest proportional decrease in nucleotide number.     

Tissue-specific expression atlas of murine mitochondrial tRNAs

Mammalian mitochondrial tRNA (mt-tRNA) plays a central role in the synthesis of the 13 subunits of the oxidative phosphorylation complex system (OXPHOS). However, many aspects of the context-dependent expression of mt-tRNAs in mammals remain unknown. To investigate the tissue-specific effects of mt-tRNAs, we performed a comprehensive analysis of mitochondrial tRNA expression across five mice tissues (brain, heart, liver, skeletal muscle, and kidney) using Northern blot analysis. Striking differences in the tissue-specific expression of 22 mt-tRNAs were observed, in some cases differing by as much as tenfold from lowest to highest expression levels among these five tissues. Overall, the heart exhibited the highest levels of mt-tRNAs, while the liver displayed markedly lower levels. Variations in the levels of mt-tRNAs showed significant correlations with total mitochondrial DNA (mtDNA) contents in these tissues. However, there were no significant differences observed in the 2-thiouridylation levels of tRNA^Lys, tRNA^Glu, and tRNA^Gln among these tissues. A wide range of aminoacylation levels for 15 mt-tRNAs occurred among these five tissues, with skeletal muscle and kidneys most notably displaying the highest and lowest tRNA aminoacylation levels, respectively. Among these tissues, there was a negative correlation between variations in mt-tRNA aminoacylation levels and corresponding variations in mitochondrial tRNA synthetases (mt-aaRS) expression levels. Furthermore, the variable levels of OXPHOS subunits, as encoded by mtDNA or nuclear genes, may reflect differences in relative functional emphasis for mitochondria in each tissue. Our findings provide new insight into the mechanism of mt-tRNA tissue-specific effects on oxidative phosphorylation.     

Proteomic Changes during B Cell Maturation: 2D-DIGE Approach

B cells play a pivotal role in adaptive immune system, since they maintain a delicate balance between recognition and clearance of foreign pathogens and tolerance to self. During maturation, B cells progress through a series of developmental stages defined by specific phenotypic surface markers and the rearrangement and expression of immunoglobulin (Ig) genes. To get insight into B cell proteome during the maturation pathway, we studied differential protein expression in eight human cell lines, which cover four distinctive developmental stages; early pre-B, pre-B, plasma cell and immature B cell upon anti-IgM stimulation. Our two-dimensional differential gel electrophoresis (2D-DIGE) and mass spectrometry based proteomic study indicates the involvement of large number of proteins with various functions. Notably, proteins related to cytoskeleton were relatively highly expressed in early pre-B and pre-B cells, whereas plasma cell proteome contained endoplasmic reticulum and Golgi system proteins. Our long time series analysis in anti-IgM stimulated Ramos B cells revealed the dynamic regulation of cytoskeleton organization, gene expression and metabolic pathways, among others. The findings are related to cellular processes in B cells and are discussed in relation to experimental information for the proteins and pathways they are involved in. Representative 2D-DIGE maps of different B cell maturation stages are available online at http://structure.bmc.lu.se/BcellProteome/.     

Assigning pathogenicity to mitochondrial tRNA mutations: when "definitely maybe" is not good enough

Some mutations in mitochondrial tRNA (mt-tRNA) genes cause devastating disease, whereas others have no clinical consequences. We understand little of the factors determining the pathogenicity of specific mt-tRNA mutations, making prediction of clinical outcome extremely difficult. Using extensive sequence databases, we compared the characteristics of neutral variations with those of pathogenic mutations. We recommend that the location of the proposed mutation within the secondary structure of the mt-tRNA molecule and the disruption it causes to Watson-Crick base pairing should be considered when assessing the pathological significance of a novel mt-tRNA mutation.     

Mitochondrial tRNA 5′-Editing in Dictyostelium discoideum and Polysphondylium pallidum

Mitochondrial tRNA (mt-tRNA) 5′-editing was first described more than 20 years ago; however, the first candidates for 5′-editing enzymes were only recently identified in a eukaryotic microbe (protist), the slime mold Dictyostelium discoideum. In this organism, eight of 18 mt-tRNAs are predicted to be edited based on the presence of genomically encoded mismatched nucleotides in their aminoacyl-acceptor stem sequences. Here, we demonstrate that mt-tRNA 5′-editing occurs at all predicted sites in D. discoideum as evidenced by changes in the sequences of isolated mt-tRNAs compared with the expected sequences encoded by the mitochondrial genome. We also identify two previously unpredicted editing events in which G-U base pairs are edited in the absence of any other genomically encoded mismatches. A comparison of 5′-editing in D. discoideum with 5′-editing in another slime mold, Polysphondylium pallidum, suggests organism-specific idiosyncrasies in the treatment of U-G/G-U pairs. In vitro activities of putative D. discoideum editing enzymes are consistent with the observed editing reactions and suggest an overall lack of tRNA substrate specificity exhibited by the repair component of the editing enzyme. Although the presence of terminal mismatches in mt-tRNA sequences is highly predictive of the occurrence of mt-tRNA 5′-editing, the variability in treatment of U-G/G-U base pairs observed here indicates that direct experimental evidence of 5′-editing must be obtained to understand the complete spectrum of mt-tRNA editing events in any species.     

PathogenFinder - Distinguishing Friend from Foe Using Bacterial Whole Genome Sequence Data

Although the majority of bacteria are harmless or even beneficial to their host, others are highly virulent and can cause serious diseases, and even death. Due to the constantly decreasing cost of high-throughput sequencing there are now many completely sequenced genomes available from both human pathogenic and innocuous strains. The data can be used to identify gene families that correlate with pathogenicity and to develop tools to predict the pathogenicity of newly sequenced strains, investigations that previously were mainly done by means of more expensive and time consuming experimental approaches. We describe PathogenFinder (http://cge.cbs.dtu.dk/services/PathogenFinder/), a web-server for the prediction of bacterial pathogenicity by analysing the input proteome, genome, or raw reads provided by the user. The method relies on groups of proteins, created without regard to their annotated function or known involvement in pathogenicity. The method has been built to work with all taxonomic groups of bacteria and using the entire training-set, achieved an accuracy of 88.6% on an independent test-set, by correctly classifying 398 out of 449 completely sequenced bacteria. The approach here proposed is not biased on sets of genes known to be associated with pathogenicity, thus the approach could aid the discovery of novel pathogenicity factors. Furthermore the pathogenicity prediction web-server could be used to isolate the potential pathogenic features of both known and unknown strains.     

Methyltransferase METTL8 is required for 3-methylcytosine modification in human mitochondrial tRNAs

A subset of eukaryotic tRNAs is methylated in the anticodon loop, forming 3-methylcytosine (m³C) modifications. In mammals, the number of tRNAs containing m³C modifications has been expanded to include mitochondrial (mt) tRNA-Ser-UGA and mt-tRNA-Thr-UGU. However, whereas the enzymes catalyzing m³C formation in nuclear-encoded tRNAs have been identified, the proteins responsible for m³C modification in mt-tRNAs are unknown. Here, we show that m³C formation in human mt-tRNAs is dependent upon the methyltransferase-Like 8 (METTL8) enzyme. We find that METTL8 is a mitochondria-associated protein that interacts with mitochondrial seryl-tRNA synthetase, as well as with mt-tRNAs containing m³C. We demonstrate that human cells deficient in METTL8 exhibit loss of m³C modification in mt-tRNAs, but not nuclear-encoded tRNAs. Consistent with the mitochondrial import of METTL8, the formation of m³C in METTL8-deficient cells could be rescued by re-expression of WT METTL8, but not by a METTL8 variant lacking the N-terminal mitochondrial localization signal. Notably, we found METTL8-deficiency in human cells causes alterations in the native migration pattern of mt-tRNA-Ser-UGA, suggesting a role for m³C in tRNA folding. Altogether, these findings demonstrate that METTL8 is required for m³C formation in mt-tRNAs and uncover a potential function for m³C modification in mitochondrial tRNA structure.     

Thio modification of yeast cytosolic tRNA is an iron-sulfur protein-dependent pathway

Defects in the yeast cysteine desulfurase Nfs1 cause a severe impairment in the 2-thio modification of uridine of mitochondrial tRNAs (mt-tRNAs) and cytosolic tRNAs (cy-tRNAs). Nfs1 can also provide the sulfur atoms of the iron-sulfur (Fe/S) clusters generated by the mitochondrial and cytosolic Fe/S cluster assembly machineries, termed ISC and CIA, respectively. Therefore, a key question remains as to whether the biosynthesis of Fe/S clusters is a prerequisite for the 2-thio modification of the tRNAs in both of the subcellular compartments of yeast cells. To elucidate this question, we asked whether mitochondrial ISC and/or cytosolic CIA components besides Nfs1 were involved in the 2-thio modification of these tRNAs. We demonstrate here that the three CIA components, Cfd1, Nbp35, and Cia1, are required for the 2-thio modification of cy-tRNAs but not of mt-tRNAs. Interestingly, the mitochondrial scaffold proteins Isu1 and Isu2 are required for the 2-thio modification of the cy-tRNAs but not of the mt-tRNAs, while mitochondrial Nfs1 is required for both 2-thio modifications. These results clearly indicate that the 2-thio modification of cy-tRNAs is Fe/S protein dependent and thus requires both CIA and ISC machineries but that of mt-tRNAs is Fe/S cluster independent and does not require key mitochondrial ISC components except for Nfs1.     

A proposed consensus panel of organisms for determining evolutionary conservation of mt-tRNA point mutations

Assigning pathogenicity to mt-tRNA variants requires multiple strands of evidence. Evolutionary conservation is often considered mandatory, but lack of a standard panel of organisms to assess conservation complicates comparison between reports and undermines the value of conservation-based evidence. We demonstrate that intra-species MTT sequence variation is sufficiently low for sequence data from a single organism to adequately represent a species. On this basis, we propose a standardised panel of organisms for conservation assessment and describe integration of this conservation panel into a pathogenicity scoring system designed to assess mt-tRNA variation associated with mitochondrial disease.     

Mtu1-Mediated Thiouridine Formation of Mitochondrial tRNAs Is Required for Mitochondrial Translation and Is Involved in Reversible Infantile Liver Injury

Reversible infantile liver failure (RILF) is a unique heritable liver disease characterized by acute liver failure followed by spontaneous recovery at an early stage of life. Genetic mutations in MTU1 have been identified in RILF patients. MTU1 is a mitochondrial enzyme that catalyzes the 2-thiolation of 5-taurinomethyl-2-thiouridine (τm⁵s²U) found in the anticodon of a subset of mitochondrial tRNAs (mt-tRNAs). Although the genetic basis of RILF is clear, the molecular mechanism that drives the pathogenesis remains elusive. We here generated liver-specific knockout of Mtu1 (Mtu1^LKO) mice, which exhibited symptoms of liver injury characterized by hepatic inflammation and elevated levels of plasma lactate and AST. Mechanistically, Mtu1 deficiency resulted in a loss of 2-thiolation in mt-tRNAs, which led to a marked impairment of mitochondrial translation. Consequently, Mtu1^LKO mice exhibited severe disruption of mitochondrial membrane integrity and a broad decrease in respiratory complex activities in the hepatocytes. Interestingly, mitochondrial dysfunction induced signaling pathways related to mitochondrial proliferation and the suppression of oxidative stress. The present study demonstrates that Mtu1-dependent 2-thiolation of mt-tRNA is indispensable for mitochondrial translation and that Mtu1 deficiency is a primary cause of RILF. In addition, Mtu1 deficiency is associated with multiple cytoprotective pathways that might prevent catastrophic liver failure and assist in the recovery from liver injury.     

Predicting the Functional Effect of Amino Acid Substitutions and Indels

As next-generation sequencing projects generate massive genome-wide sequence variation data, bioinformatics tools are being developed to provide computational predictions on the functional effects of sequence variations and narrow down the search of casual variants for disease phenotypes. Different classes of sequence variations at the nucleotide level are involved in human diseases, including substitutions, insertions, deletions, frameshifts, and non-sense mutations. Frameshifts and non-sense mutations are likely to cause a negative effect on protein function. Existing prediction tools primarily focus on studying the deleterious effects of single amino acid substitutions through examining amino acid conservation at the position of interest among related sequences, an approach that is not directly applicable to insertions or deletions. Here, we introduce a versatile alignment-based score as a new metric to predict the damaging effects of variations not limited to single amino acid substitutions but also in-frame insertions, deletions, and multiple amino acid substitutions. This alignment-based score measures the change in sequence similarity of a query sequence to a protein sequence homolog before and after the introduction of an amino acid variation to the query sequence. Our results showed that the scoring scheme performs well in separating disease-associated variants (n = 21,662) from common polymorphisms (n = 37,022) for UniProt human protein variations, and also in separating deleterious variants (n = 15,179) from neutral variants (n = 17,891) for UniProt non-human protein variations. In our approach, the area under the receiver operating characteristic curve (AUC) for the human and non-human protein variation datasets is ∼0.85. We also observed that the alignment-based score correlates with the deleteriousness of a sequence variation. In summary, we have developed a new algorithm, PROVEAN (Protein Variation Effect Analyzer), which provides a generalized approach to predict the functional effects of protein sequence variations including single or multiple amino acid substitutions, and in-frame insertions and deletions. The PROVEAN tool is available online at http://provean.jcvi.org.     

Pathogenic mutations in antisense mitochondrial tRNAs

Pathogenic mutations in mitochondrial tRNAs are 6.5 times more frequent than in other mitochondrial genes. This suggests that tRNA mutations perturb more than one function. A potential additional tRNA gene function is that of templating for antisense tRNAs. Pathogenic mutations weaken cloverleaf secondary structures of sense tRNAs. Analyses here show similar effects for most antisense tRNAs, especially after adjusting for associations between sense and antisense cloverleaf stabilities. These results imply translational activity by antisense tRNAs. For sense tRNAs Ala and Ser UCN, pathogenicity associates as much with sense as with antisense cloverleaf formation. For tRNA Pro, pathogenicity seems associated only with antisense, not sense tRNA cloverleaf formation. Translational activity by antisense tRNAs is expected for the 11 antisense tRNAs processed by regular sense RNA maturation, those recognized by their cognate amino acid’s tRNA synthetase, and those forming relatively stable cloverleaves as compared to their sense counterpart. Most antisense tRNAs probably function routinely in translation and extend the tRNA pool (extension hypothesis); others do not (avoidance hypothesis). The greater the expected translational activity of an antisense tRNA, the more pathogenic mutations weaken its cloverleaf secondary structure. Some evidence for RNA interference, a more classical role for antisense tRNAs, exists only for tRNA Ser UCN. Mutation pathogenicity probably frequently results from a mixture of effects due to sense and antisense tRNA translational activity for many mitochondrial tRNAs. Genomic studies should routinely explore for translational activity by antisense tRNAs.     

MP3: A Software Tool for the Prediction of Pathogenic Proteins in Genomic and Metagenomic Data

The identification of virulent proteins in any de-novo sequenced genome is useful in estimating its pathogenic ability and understanding the mechanism of pathogenesis. Similarly, the identification of such proteins could be valuable in comparing the metagenome of healthy and diseased individuals and estimating the proportion of pathogenic species. However, the common challenge in both the above tasks is the identification of virulent proteins since a significant proportion of genomic and metagenomic proteins are novel and yet unannotated. The currently available tools which carry out the identification of virulent proteins provide limited accuracy and cannot be used on large datasets. Therefore, we have developed an MP3 standalone tool and web server for the prediction of pathogenic proteins in both genomic and metagenomic datasets. MP3 is developed using an integrated Support Vector Machine (SVM) and Hidden Markov Model (HMM) approach to carry out highly fast, sensitive and accurate prediction of pathogenic proteins. It displayed Sensitivity, Specificity, MCC and accuracy values of 92%, 100%, 0.92 and 96%, respectively, on blind dataset constructed using complete proteins. On the two metagenomic blind datasets (Blind A: 51–100 amino acids and Blind B: 30–50 amino acids), it displayed Sensitivity, Specificity, MCC and accuracy values of 82.39%, 97.86%, 0.80 and 89.32% for Blind A and 71.60%, 94.48%, 0.67 and 81.86% for Blind B, respectively. In addition, the performance of MP3 was validated on selected bacterial genomic and real metagenomic datasets. To our knowledge, MP3 is the only program that specializes in fast and accurate identification of partial pathogenic proteins predicted from short (100–150 bp) metagenomic reads and also performs exceptionally well on complete protein sequences. MP3 is publicly available at http://metagenomics.iiserb.ac.in/mp3/index.php.