Background

Human mitochondria possess their own genome and house all the components necessary for transcription, RNA maturation, and translation of the encoded genetic information (Dennerlein et al., 2017). The mitoribosome, one of the most important machineries within mitochondria, is responsible for the translation of essential mitochondrial mRNAs encoding components of oxidative phosphorylation (OXPHOS) complexes (Ott et al., 2016). Since all proteins translated by the human mitoribosome are membrane proteins, knowledge of the mechanism of co-translational membrane protein insertion is central to understanding mitoribosome translation (Hildenbeutel et al., 2012; Stiller et al., 2016). In human mitochondria, the oxidase assembly 1-like (Oxa1L) translocase plays a central role in the membrane insertion of mitochondrially encoded products (Haque et al., 2010; Stiller et al., 2016); however, due to the unique features of mitochondria, little is known regarding the molecular and quality control mechanisms underlying mitochondrial translation.

Owing to the structures obtained by single-particle electron cryo-microscopy (cryo-EM), opportunities now arise to comprehensively study the molecular mechanisms underlying the translational action and quality control of the human mitoribosome (Rathore, 2020). Although the existence of quality control in mitochondria has been predicted based on analogy with bacterial and eukaryotic cytosols (Ayyub et al., 2020), biochemical or structural evidence of ribosome-associated quality control in mitochondria remain elusive. We reasoned that any attempt to induce translational stalling may trap mitoribosomes in various stages of the translation cycle during the act of nascent chain insertion into the inner mitochondrial membrane, and generate intermediates suitable for structural analysis by cryo-EM. To this end, we purified mitoribosomes from a genetically engineered human cell line lacking the 2’-5’ phosphodiesterase exonuclease 12 (PDE12). PDE12 facilitates the maturation of mt-tRNA; therefore, a PDE12 knockout can lead to aberrant polyadenylation of the 3’ ends of mt-tRNAs and, consequently, mitoribosome stalling (Rorbach et al., 2011; Pearce et al., 2017). As expected, we observed a substantial proportion of stalled mitoribosomes and discovered a mitoribosome-associated quality control pathway (mtRQC) to rescue the elongational stalling ribosomes (Desai et al., 2020). Our report of the first mitoribosome-associated quality control pathway provides novel insights into the regulation of mitochondrial translation. Since the cell needs to detect and respond rapidly even to subtle changes in translation rates, quality control is intimately coupled to elongation. During purification, we included GMPPCP, a non-hydrolysable GTP analog, to prevent dissociation of GTPases from mitoribosomes, and successfully captured five additional structures of virtually every elongating ribosomal state. Purifying the mitoribosome in complex with its translocase Oxa1L was challenging since membrane proteins are unstable and tend to dissociate during complicated isolation steps. Accordingly, we added n-dodecyl β-D-maltoside (β-DDM) along with cardiolipin to solubilize mitoribosomes anchored to the mitochondrial inner membrane via the Oxa1L translocase. We also trapped Oxa1L on all our active and stalled ribosomes, giving us a first glimpse at the co-translational insertion of mitochondrial inner membrane proteins.

This protocol was modified from prior methods developed in our lab (Amunts et al., 2015; Brown et al., 2017) and by others (Greber et al., 2014 and 2015; Aibara et al., 2018), as well as in our current work (Desai et al., 2020). The method described here can be used to obtain mitoribosomes in complex with Oxa1L, which has advantages such as the applicability for mitochondria isolation from large volumes of suspension cell culture, the use of common media and buffers, ease of handling, and absence of additional affinity purification steps. In our previous publication and experiments, we observed a good reproducibility between HEK293 cell batches using this methodology.