High-resolution imaging reveals compartmentalization of mitochondrial protein synthesis in cultured human cells

Abstract

Human mitochondria contain their own genome, mitochondrial DNA, that is expressed in the mitochondrial matrix. This genome encodes 13 vital polypeptides that are components of the multisubunit complexes that couple oxidative phosphorylation (OXPHOS). The inner mitochondrial membrane that houses these complexes comprises the inner boundary membrane that runs parallel to the outer membrane, infoldings that form the cristae membranes, and the cristae junctions that separate the two. It is in these cristae membranes that the OXPHOS complexes have been shown to reside in various species. The majority of the OXPHOS subunits are nuclear-encoded and must therefore be imported from the cytosol through the outer membrane at contact sites with the inner boundary membrane. As the mitochondrially encoded components are also integral members of these complexes, where does protein synthesis occur? As transcription, mRNA processing, maturation, and at least part of the mitoribosome assembly process occur at the nucleoid and the spatially juxtaposed mitochondrial RNA granules, is protein synthesis also performed at the RNA granules close to these entities, or does it occur distal to these sites? We have adapted a click chemistry-based method coupled with stimulated emission depletion nanoscopy to address these questions. We report that, in human cells in culture, within the limits of our methodology, the majority of mitochondrial protein synthesis is detected at the cristae membranes and is spatially separated from the sites of RNA processing and maturation.

Mitochondrial translation occurs preferentially in the peri-nuclear mitochondrial network of cultured human cells

“…The focus of this mini-series is mitochondrial heterogeneity. Our data presented here and that of another recent publication [19] show clearly that for at least certain human cell lines there is evidence of marked heterogeneity in the location of translation within the mitochondrial network. Normalising mitochondrial signal using markers to either the outer mitochondrial membrane (TOM20) or the cristae membranes (FoF1 ATP synthase) confirm new protein synthesis is more markedly concentrated within the perinuclear mitochondrial network rather than towards the cell periphery.…”

“…By exploiting the technique of click chemistry in a process termed mitochondrial FUNCAT (Fluorescent Noncanocial Aminoacid Tagging), our laboratory was able to show that newly synthesised proteins were mostly located at the cristal membranes [19]. We now report that further use of mitochondrial FUNCAT reveals a heterogeneous distribution of translation within the mitochondrial network in cultured cells, with a far greater level of synthesis occurring in the perinuclear region of the network as compared to the periphery after normalisation to mitochondrial mass.…”

“…For FUNCAT experiments, cells were first seeded onto glass coverslips and cultured for 1-2 days. In vivo labelling of mitochondrial translation products was performed essentially as described [19]. Cells were pulsed (30 min, unless otherwise stated) with methionine-free DMEM containing the methionine analogue HPG (Jena Bioscience) with concomitant inhibition of cytosolic translation by cycloheximide (50 µg/ml).…”

“…Initial data from mitochondrial FUNCAT visualising mitochondrial protein synthesis in fibroblasts, U2OS and HeLa cells had suggested an enrichment of translation in the perinuclear region of the cell compared to the periphery when normalized to a marker of mitochondrial mass [19]. To extend this analysis we subjected human cultured U2OS cells to mitochondrial FUNCAT as previously described [19,20]. All cell images were manually masked such that signals were assigned either to the perinuclear or peripheral regions of each cell.…”

RNA Granules in the Mitochondria and Their Organization under Mitochondrial Stresses

“…Until now, the precise identity of the transcripts present in the MRGs has not been revealed, and it would be interesting to know whether polycistronic, immature transcripts occupy the same or a different location as the mature transcripts ready for translation. A recent study by Zorkau et al [17] has suggested that the mitoribosomes could be loaded with mature transcripts that are close to MRGs but not within them. The authors measured the spatiotemporal kinetics of mitochondrial protein synthesis by using fluorescent noncanonical amino acid tagging.…”