Trz1 chez levure est responsable du clivage endonucléolytique à l'extrémité 3 'au cours du processus de maturation des ARNt. Trz1 appartient à la famille des RNases de type b-lactamase, caractérisé par la présence d'un motif de séquence HxHxDH qui est impliqué dans la fixation des ions de zinc catalytique. La famille des RNaseZ est partagée en deux sous-familles de longueur de séquence différente: les formes courtes (300-400 acides aminés) et les formes longues (700-900 acides aminés). Les structures cristallines des enzymes RNaseZ de forme courte ont montré qu'ils sont actifs comme des homodimères. Une sous-unité englobe les ARNt substrat en utilisant un bras en saillie et l'autre fournit le site catalytique. Nous présentons ici la structure cristalline de Trz1, la première pour une RNase Z de forme longue. Trz1 est organisé en deux domaines reliés par un long peptide charnière. Chaque domaine est composé d'un repliement de type β-lactamase. Le domaine N-terminal a perdu ses résidus catalytiques au cours de l’évolution, mais il contient le bras long qui est important pour la liaison de l'ARNt; tandis que c’est l'inverse pour le domaine C-terminal. À partir des études protéomiques, on sait que Trz1 forme un complexe ternaire avec NUC1, une nucléase mitochondriale impliquée dans l'apoptose, et avec une mutarotase (codée par YMR099C). Nous avons purifié le complexe ternaire Trz1/NUC1/mutarotase caractérisé ses propriétés biochimiques. Trz1/NUC1/mutarotase forme in vitro un heterohexamère très stable en solution. A partir de nos données SAXS et MALLS nous proposons que l'homodimère NUC1 est au centre du complexe et que chaque sous-unité interagit avec une copie de Trz1 et une copie de mutarotase. / Yeast Trz1 is responsible for the endonucleolytic cleavage at the 3’-end during the maturation process of tRNAs. Trz1 belongs to the family of β-lactamase type RNases characterized by the presence of a HxHxDH sequence motif that is involved in the ligand formation of the catalytical required Zn-ions. The family consists of two subfamilies: the short forms with sequence lengths between and the long forms. A few crystal structures of short form RNase Z enzymes showed that they are active as homodimers. One subunit embraces the substrate tRNA using a protruding arm and the other provides the catalytic site. We here present the crystal structure of Trz1, the first of a long form RNase Z. Trz1 is organized in two domains connected by a large linker. Each domain is composed of a beta-lactamase type fold. The N-terminal domain has lost its catalytic residues, but contains the long arm that is important for tRNA binding; while it is the other way around of the C-terminal domain. From proteomics studies it is known that Trz1 forms a ternary complex with NUC1, a mitochondrial nuclease involved in apoptosis, and with a mutarotase (encoded by YMR099C). We purified the ternary Trz1/Nuc1/mutarotase complex and characterized its biochemical properties. Trz1/Nuc1/mutarotase forms in vitro a very stable heterohexamer in solution. From our SAXS and MALLS data we propose that the Nuc1 homodimer is at the centre of the complex and that each subunit interacts with one copy of Trz1 and mutarotase.

Transcription and translation of mammalian mitochondrial DNA (mtDNA) occurs within the mitochondrial matrix to produce oxidative phosphorylation subunits required for efficient energy production. These mtDNA-encoded subunits complex with mitochondrial-localized, nuclear-encoded subunits to form the respiratory chain, and aberrant production or function of these subunits can cause devastating human disease. In addition to 13 oxidative phosphorylation subunits, mtDNA encodes 2 rRNAs and 22 tRNAs. All proteins required for mitochondrial RNA transcription, processing, and translation are encoded in the nucleus and translocated into the mitochondria. Here, I characterize over 100 nuclear-encoded mitochondrial proteins with predicted RNA-binding domains. Using RNAi and an RNA profiling approach, MitoString, we further characterize previously identified RNA processing factors and identify the novel regulator FASTKD4, which influences the abundance of a subset of mitochondrial mRNAs. Next, we apply knowledge of the RNA degradation component SUPV3L1 gleaned from our RNAi studies and previous research to test whether a specific set of variants influence the function of this gene in patient fibroblasts. Using MitoString, we find no evidence of pathogenicity of these variants in our fibroblast model. Our approach highlights the value of a thorough understanding of mitochondrial proteins and the necessity of experimental techniques to validate the effect of variants found in exome-sequencing studies. Finally, we take an unbiased approach to characterizing the mitochondrial transcriptome of mouse liver by sequencing RNA from sequentially enriched mitochondrial fractions. Although we find an abundance of nuclear-encoded 5S rRNA, consistent with previous research, we fail to identify any imported nuclear-encoded tRNAs. Uniting genomics, biochemistry, and medicine, these findings advance our understanding of mitochondrial RNA biology.