Mitochondrial protein homeostasis: An investigation of substrate interplay between a processing peptidase APP3m and matrix AAA+ proteases

Submission note: A thesis submitted in total fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY to the Department of Biochemistry and Genetics, La Trobe Institute for Molecular Science, School of Health, Science and Engineering, La Trobe University, Bundoora.

Most mitochondrial matrix proteins are targeted to the organelle via an Nterminal presequence. Following import into the matrix, typically the presequence is cleaved by mitochondrial processing peptidase. For some matrix proteins, removal of a single N-terminal residue by intermediate cleaving peptidase 55 (scIcp55) is required to generate a mature protein. In yeast, the loss of scIcp55 generates immature matrix proteins that contain an exposed N-terminal hydrophobic residue (Y, F or L). This motif resembles a known degradation tag (N-degron) which facilitates protein turnover via a pathway known as the N-end rule. Lack of stability of these immature proteins in the mitochondrial matrix suggests the existence of a similar protein degradation pathway in mitochondria, which is potentially mediated by AAA+ proteases. However, to date neither is the protease(s) responsible for turnover of these proteins identified nor the existence of such a pathway in mammals known. In this study, human mitochondrial aminopeptidase P isoform 3 (hsAPP3m) was identified as a functional homolog of yeast Icp55. Depletion of hsAPP3m in human cells (coupled with a novel affinity chromatography method) permitted the identification of numerous putative N-end rule substrates in mitochondria. The ability of matrix AAA+ proteases to degrade model N-end rule substrates (both from yeast and humans) was examined. Interestingly, the presence of a single N-terminal hydrophobic residue on selected putative substrates enhanced the rate of turnover by hsLONM, suggesting that hsLONM is.. responsible for the turnover of immature proteins in mitochondria. Further dissection of the degron revealed that hsLONM-recognition was mediated by a bipartite signal composed of an N-terminal residue and a downstream hydrophobic motif. Interestingly, the spatial arrangement of this bipartite signal but not the identity of its N-terminal residue was critical for degradation by hsLONM. Nevertheless, hsLONM-mediated turnover of hsAPP3m substrates in vivo was not verified and requires further evaluation.