HIV-1 Packaging Visualised by In-Gel SHAPE.

HIV-1 packages two copies of its gRNA into virions via an interaction with the viral structural protein Gag. Both copies and their native RNA structure are essential for virion infectivity. The precise stepwise nature of the packaging process has not been resolved. This is largely due to a prior lack of structural techniques that follow RNA structural changes within an RNA-protein complex. Here, we apply the in-gel SHAPE (selective 2'OH acylation analysed by primer extension) technique to study the initiation of HIV-1 packaging, examining the interaction between the packaging signal RNA and the Gag polyprotein, and compare it with that of the NC domain of Gag alone. Our results imply interactions between Gag and monomeric packaging signal RNA in switching the RNA conformation into a dimerisation-competent structure, and show that the Gag-dimer complex then continues to stabilise. These data provide a novel insight into how HIV-1 regulates the translation and packaging of its genome.

Duplex formation between the template and the nascent strand in the transcription-regulating sequences is associated with the site of template switching in SARS - CoV-2.

Recently published transcriptomic data of the SARS-CoV-2 coronavirus show that there is a large variation in the frequency and steady state levels of subgenomic mRNA sequences. This variation is derived from discontinuous subgenomic RNA synthesis, where the polymerase switches template from a 3' proximal genome body sequence to a 5' untranslated leader sequence. This leads to a fusion between the common 5' leader sequence and a 3' proximal body sequence in the RNA product. This process revolves around a common core sequence (CS) that is present at both the template sites that make up the fusion junction. Base-pairing between the leader CS and the nascent complementary minus strand body CS, and flanking regions (together called the transcription regulating sequence, TRS) is vital for this template switching event. However, various factors can influence the site of template switching within the same TRS duplex. Here, we model the duplexes formed between the leader and complementary body TRS regions, hypothesizing the role of the stability of the TRS duplex in determining the major sites of template switching for the most abundant mRNAs. We indicate that the stability of secondary structures and the speed of transcription play key roles in determining the probability of template switching in the production of subgenomic RNAs. We speculate on the effect of reported variant nucleotide substitutions on our models.

YbeY is required for ribosome small subunit assembly and tRNA processing in human mitochondria.

Mitochondria contain their own translation apparatus which enables them to produce the polypeptides encoded in their genome. The mitochondrially-encoded RNA components of the mitochondrial ribosome require various post-transcriptional processing steps. Additional protein factors are required to facilitate the biogenesis of the functional mitoribosome. We have characterized a mitochondrially-localized protein, YbeY, which interacts with the assembling mitoribosome through the small subunit. Loss of YbeY leads to a severe reduction in mitochondrial translation and a loss of cell viability, associated with less accurate mitochondrial tRNASer(AGY) processing from the primary transcript and a defect in the maturation of the mitoribosomal small subunit. Our results suggest that YbeY performs a dual, likely independent, function in mitochondria being involved in precursor RNA processing and mitoribosome biogenesis. Issue Section: Nucleic Acid Enzymes.

METTL15 introduces N4-methylcytidine into human mitochondrial 12S rRNA and is required for mitoribosome biogenesis.

Post-transcriptional RNA modifications, the epitranscriptome, play important roles in modulating the functions of RNA species. Modifications of rRNA are key for ribosome production and function. Identification and characterization of enzymes involved in epitranscriptome shaping is instrumental for the elucidation of the functional roles of specific RNA modifications. Ten modified sites have been thus far identified in the mammalian mitochondrial rRNA. Enzymes responsible for two of these modifications have not been characterized. Here, we identify METTL15, show that it is the main N4-methylcytidine (m4C) methyltransferase in human cells and demonstrate that it is responsible for the methylation of position C839 in mitochondrial 12S rRNA. We show that the lack of METTL15 results in a reduction of the mitochondrial de novo protein synthesis and decreased steady-state levels of protein components of the oxidative phosphorylation system. Without functional METTL15, the assembly of the mitochondrial ribosome is decreased, with the late assembly components being unable to be incorporated efficiently into the small subunit. We speculate that m4C839 is involved in the stabilization of 12S rRNA folding, therefore facilitating the assembly of the mitochondrial small ribosomal subunits. Taken together our data show that METTL15 is a novel protein necessary for efficient translation in human mitochondria.

Mitochondrial transcription and translation: overview.

Mitochondria are the major source of ATP in the cell. Five multi-subunit complexes in the inner membrane of the organelle are involved in the oxidative phosphorylation required for ATP production. Thirteen subunits of these complexes are encoded by the mitochondrial genome often referred to as mtDNA. For this reason, the expression of mtDNA is vital for the assembly and functioning of the oxidative phosphorylation complexes. Defects of the mechanisms regulating mtDNA gene expression have been associated with deficiencies in assembly of these complexes, resulting in mitochondrial diseases. Recently, numerous factors involved in these processes have been identified and characterized leading to a deeper understanding of the mechanisms that underlie mitochondrial diseases.

Defective mitochondrial rRNA methyltransferase MRM2 causes MELAS-like clinical syndrome.

Defects in nuclear-encoded proteins of the mitochondrial translation machinery cause early-onset and tissue-specific deficiency of one or more OXPHOS complexes. Here, we report a 7-year-old Italian boy with childhood-onset rapidly progressive encephalomyopathy and stroke-like episodes. Multiple OXPHOS defects and decreased mtDNA copy number (40%) were detected in muscle homogenate. Clinical features combined with low level of plasma citrulline were highly suggestive of mitochondrial encephalopathy, lactic acidosis and stroke-like episodes (MELAS) syndrome, however, the common m.3243 A > G mutation was excluded. Targeted exome sequencing of genes encoding the mitochondrial proteome identified a damaging mutation, c.567 G > A, affecting a highly conserved amino acid residue (p.Gly189Arg) of the MRM2 protein. MRM2 has never before been linked to a human disease and encodes an enzyme responsible for 2'-O-methyl modification at position U1369 in the human mitochondrial 16S rRNA. We generated a knockout yeast model for the orthologous gene that showed a defect in respiration and the reduction of the 2'-O-methyl modification at the equivalent position (U2791) in the yeast mitochondrial 21S rRNA. Complementation with the mrm2 allele carrying the equivalent yeast mutation failed to rescue the respiratory phenotype, which was instead completely rescued by expressing the wild-type allele. Our findings establish that defective MRM2 causes a MELAS-like phenotype, and suggests the genetic screening of the MRM2 gene in patients with a m.3243 A > G negative MELAS-like presentation.

Biallelic C1QBP Mutations Cause Severe Neonatal-, Childhood-, or Later-Onset Cardiomyopathy Associated with Combined Respiratory-Chain Deficiencies.

Complement component 1 Q subcomponent-binding protein (C1QBP; also known as p32) is a multi-compartmental protein whose precise function remains unknown. It is an evolutionary conserved multifunctional protein localized primarily in the mitochondrial matrix and has roles in inflammation and infection processes, mitochondrial ribosome biogenesis, and regulation of apoptosis and nuclear transcription. It has an N-terminal mitochondrial targeting peptide that is proteolytically processed after import into the mitochondrial matrix, where it forms a homotrimeric complex organized in a doughnut-shaped structure. Although C1QBP has been reported to exert pleiotropic effects on many cellular processes, we report here four individuals from unrelated families where biallelic mutations in C1QBP cause a defect in mitochondrial energy metabolism. Infants presented with cardiomyopathy accompanied by multisystemic involvement (liver, kidney, and brain), and children and adults presented with myopathy and progressive external ophthalmoplegia. Multiple mitochondrial respiratory-chain defects, associated with the accumulation of multiple deletions of mitochondrial DNA in the later-onset myopathic cases, were identified in all affected individuals. Steady-state C1QBP levels were decreased in all individuals' samples, leading to combined respiratory-chain enzyme deficiency of complexes I, III, and IV. C1qbp-/- mouse embryonic fibroblasts (MEFs) resembled the human disease phenotype by showing multiple defects in oxidative phosphorylation (OXPHOS). Complementation with wild-type, but not mutagenized, C1qbp restored OXPHOS protein levels and mitochondrial enzyme activities in C1qbp-/- MEFs. C1QBP deficiency represents an important mitochondrial disorder associated with a clinical spectrum ranging from infantile lactic acidosis to childhood (cardio)myopathy and late-onset progressive external ophthalmoplegia.

Maturation of selected human mitochondrial tRNAs requires deadenylation.

Human mitochondria contain a genome (mtDNA) that encodes essential subunits of the oxidative phosphorylation system. Expression of mtDNA entails multi-step maturation of precursor RNA. In other systems, the RNA life cycle involves surveillance mechanisms, however, the details of RNA quality control have not been extensively characterised in human mitochondria. Using a mitochondrial ribosome profiling and mitochondrial poly(A)-tail RNA sequencing (MPAT-Seq) assay, we identify the poly(A)-specific exoribonuclease PDE12 as a major factor for the quality control of mitochondrial non-coding RNAs. The lack of PDE12 results in a spurious polyadenylation of the 3' ends of the mitochondrial (mt-) rRNA and mt-tRNA. While the aberrant adenylation of 16S mt-rRNA did not affect the integrity of the mitoribosome, spurious poly(A) additions to mt-tRNA led to reduced levels of aminoacylated pool of certain mt-tRNAs and mitoribosome stalling at the corresponding codons. Therefore, our data uncover a new, deadenylation-dependent mtRNA maturation pathway in human mitochondria.

Regulation of Mammalian Mitochondrial Gene Expression: Recent Advances.

Perturbation of mitochondrial DNA (mtDNA) gene expression can lead to human pathologies. Therefore, a greater appreciation of the basic mechanisms of mitochondrial gene expression is desirable to understand the pathophysiology of associated disorders. Although the purpose of the mitochondrial gene expression machinery is to provide only 13 proteins of the oxidative phosphorylation (OxPhos) system, recent studies have revealed its remarkable and unexpected complexity. We review here the latest breakthroughs in our understanding of the post-transcriptional processes of mitochondrial gene expression, focusing on advances in analyzing the mitochondrial epitranscriptome, the role of mitochondrial RNA granules (MRGs), the benefits of recently obtained structures of the mitochondrial ribosome, and the coordination of mitochondrial and cytosolic translation to orchestrate the biogenesis of OxPhos complexes.

Human mitochondrial ribosomes can switch their structural RNA composition.

The recent developments in cryo-EM have revolutionized our access to previously refractory structures. In particular, such studies of mammalian mitoribosomes have confirmed the absence of any 5S rRNA species and revealed the unexpected presence of a mitochondrially encoded tRNA (mt-tRNA) that usurps this position. Although the cryo-EM structures resolved the conundrum of whether mammalian mitoribosomes contain a 5S rRNA, they introduced a new dilemma: Why do human and porcine mitoribosomes integrate contrasting mt-tRNAs? Human mitoribosomes have been shown to integrate mt-tRNAVal compared with the porcine use of mt-tRNAPhe We have explored this observation further. Our studies examine whether a range of mt-tRNAs are used by different mammals, or whether the mt-tRNA selection is strictly limited to only these two species of the 22 tRNAs encoded by the mitochondrial genome (mtDNA); whether there is tissue-specific variation within a single organism; and what happens to the human mitoribosome when levels of the mt-tRNAVal are depleted. Our data demonstrate that only mt-tRNAVal or mt-tRNAPhe are found in the mitoribosomes of five different mammals, each mammal favors the same mt-tRNA in all tissue types, and strikingly, when steady-state levels of mt-tRNAVal are reduced, human mitoribosome biogenesis displays an adaptive response by switching to the incorporation of mt-tRNAPhe to generate translationally competent machinery.

TRMT5 Mutations Cause a Defect in Post-transcriptional Modification of Mitochondrial tRNA Associated with Multiple Respiratory-Chain Deficiencies.

Deficiencies in respiratory-chain complexes lead to a variety of clinical phenotypes resulting from inadequate energy production by the mitochondrial oxidative phosphorylation system. Defective expression of mtDNA-encoded genes, caused by mutations in either the mitochondrial or nuclear genome, represents a rapidly growing group of human disorders. By whole-exome sequencing, we identified two unrelated individuals carrying compound heterozygous variants in TRMT5 (tRNA methyltransferase 5). TRMT5 encodes a mitochondrial protein with strong homology to members of the class I-like methyltransferase superfamily. Both affected individuals presented with lactic acidosis and evidence of multiple mitochondrial respiratory-chain-complex deficiencies in skeletal muscle, although the clinical presentation of the two affected subjects was remarkably different; one presented in childhood with failure to thrive and hypertrophic cardiomyopathy, and the other was an adult with a life-long history of exercise intolerance. Mutations in TRMT5 were associated with the hypomodification of a guanosine residue at position 37 (G37) of mitochondrial tRNA; this hypomodification was particularly prominent in skeletal muscle. Deficiency of the G37 modification was also detected in human cells subjected to TRMT5 RNAi. The pathogenicity of the detected variants was further confirmed in a heterologous yeast model and by the rescue of the molecular phenotype after re-expression of wild-type TRMT5 cDNA in cells derived from the affected individuals. Our study highlights the importance of post-transcriptional modification of mitochondrial tRNAs for faithful mitochondrial function.

Mitochondrial transcript maturation and its disorders.

Mitochondrial respiratory chain deficiencies exhibit a wide spectrum of clinical presentations owing to defective mitochondrial energy production through oxidative phosphorylation. These defects can be caused by either mutations in the mitochondrial DNA (mtDNA) or mutations in nuclear genes coding for mitochondrially-targeted proteins. The underlying pathomechanisms can affect numerous pathways involved in mitochondrial biology including expression of mtDNA-encoded genes. Expression of the mitochondrial genes is extensively regulated at the post-transcriptional stage and entails nucleolytic cleavage of precursor RNAs, RNA nucleotide modifications, RNA polyadenylation, RNA quality and stability control. These processes ensure proper mitochondrial RNA (mtRNA) function, and are regulated by dedicated, nuclear-encoded enzymes. Recent growing evidence suggests that mutations in these nuclear genes, leading to incorrect maturation of RNAs, are a cause of human mitochondrial disease. Additionally, mutations in mtDNA-encoded genes may also affect RNA maturation and are frequently associated with human disease. We review the current knowledge on a subset of nuclear-encoded genes coding for proteins involved in mitochondrial RNA maturation, for which genetic variants impacting upon mitochondrial pathophysiology have been reported. Also, primary pathological mtDNA mutations with recognised effects upon RNA processing are described.

Bayesian robot system identification with input and output noise.

For complex robots such as humanoids, model-based control is highly beneficial for accurate tracking while keeping negative feedback gains low for compliance. However, in such multi degree-of-freedom lightweight systems, conventional identification of rigid body dynamics models using CAD data and actuator models is inaccurate due to unknown nonlinear robot dynamic effects. An alternative method is data-driven parameter estimation, but significant noise in measured and inferred variables affects it adversely. Moreover, standard estimation procedures may give physically inconsistent results due to unmodeled nonlinearities or insufficiently rich data. This paper addresses these problems, proposing a Bayesian system identification technique for linear or piecewise linear systems. Inspired by Factor Analysis regression, we develop a computationally efficient variational Bayesian regression algorithm that is robust to ill-conditioned data, automatically detects relevant features, and identifies input and output noise. We evaluate our approach on rigid body parameter estimation for various robotic systems, achieving an error of up to three times lower than other state-of-the-art machine learning methods.

Efficient learning and feature selection in high-dimensional regression.

We present a novel algorithm for efficient learning and feature selection in high-dimensional regression problems. We arrive at this model through a modification of the standard regression model, enabling us to derive a probabilistic version of the well-known statistical regression technique of backfitting. Using the expectation-maximization algorithm, along with variational approximation methods to overcome intractability, we extend our algorithm to include automatic relevance detection of the input features. This variational Bayesian least squares (VBLS) approach retains its simplicity as a linear model, but offers a novel statistically robust black-box approach to generalized linear regression with high-dimensional inputs. It can be easily extended to nonlinear regression and classification problems. In particular, we derive the framework of sparse Bayesian learning, the relevance vector machine, with VBLS at its core, offering significant computational and robustness advantages for this class of methods. The iterative nature of VBLS makes it most suitable for real-time incremental learning, which is crucial especially in the application domain of robotics, brain-machine interfaces, and neural prosthetics, where real-time learning of models for control is needed. We evaluate our algorithm on synthetic and neurophysiological data sets, as well as on standard regression and classification benchmark data sets, comparing it with other competitive statistical approaches and demonstrating its suitability as a drop-in replacement for other generalized linear regression techniques.

Variational Bayesian least squares: an application to brain-machine interface data.

An increasing number of projects in neuroscience require statistical analysis of high-dimensional data, as, for instance, in the prediction of behavior from neural firing or in the operation of artificial devices from brain recordings in brain-machine interfaces. Although prevalent, classical linear analysis techniques are often numerically fragile in high dimensions due to irrelevant, redundant, and noisy information. We developed a robust Bayesian linear regression algorithm that automatically detects relevant features and excludes irrelevant ones, all in a computationally efficient manner. In comparison with standard linear methods, the new Bayesian method regularizes against overfitting, is computationally efficient (unlike previously proposed variational linear regression methods, is suitable for data sets with large numbers of samples and a very high number of input dimensions) and is easy to use, thus demonstrating its potential as a drop-in replacement for other linear regression techniques. We evaluate our technique on synthetic data sets and on several neurophysiological data sets. For these neurophysiological data sets we address the question of whether EMG data collected from arm movements of monkeys can be faithfully reconstructed from neural activity in motor cortices. Results demonstrate the success of our newly developed method, in comparison with other approaches in the literature, and, from the neurophysiological point of view, confirms recent findings on the organization of the motor cortex. Finally, an incremental, real-time version of our algorithm demonstrates the suitability of our approach for real-time interfaces between brains and machines.

Incremental online learning in high dimensions.

Locally weighted projection regression (LWPR) is a new algorithm for incremental nonlinear function approximation in high-dimensional spaces with redundant and irrelevant input dimensions. At its core, it employs nonparametric regression with locally linear models. In order to stay computationally efficient and numerically robust, each local model performs the regression analysis with a small number of univariate regressions in selected directions in input space in the spirit of partial least squares regression. We discuss when and how local learning techniques can successfully work in high-dimensional spaces and review the various techniques for local dimensionality reduction before finally deriving the LWPR algorithm. The properties of LWPR are that it (1) learns rapidly with second-order learning methods based on incremental training, (2) uses statistically sound stochastic leave-one-out cross validation for learning without the need to memorize training data, (3) adjusts its weighting kernels based on only local information in order to minimize the danger of negative interference of incremental learning, (4) has a computational complexity that is linear in the number of inputs, and (5) can deal with a large number of-possibly redundant-inputs, as shown in various empirical evaluations with up to 90 dimensional data sets. For a probabilistic interpretation, predictive variance and confidence intervals are derived. To our knowledge, LWPR is the first truly incremental spatially localized learning method that can successfully and efficiently operate in very high-dimensional spaces.