Yeast as a system for modeling mitochondrial disease mechanisms and discovering therapies

Abstract

Mitochondrial diseases are severe and largely untreatable. Owing to the many essential processes carried out by mitochondria and the complex cellular systems that support these processes, these diseases are diverse, pleiotropic, and challenging to study. Much of our current understanding of mitochondrial function and dysfunction comes from studies in the baker's yeast Saccharomyces cerevisiae. Because of its good fermenting capacity, S. cerevisiae can survive mutations that inactivate oxidative phosphorylation, has the ability to tolerate the complete loss of mitochondrial DNA (a property referred to as 'petite-positivity'), and is amenable to mitochondrial and nuclear genome manipulation. These attributes make it an excellent model system for studying and resolving the molecular basis of numerous mitochondrial diseases. Here, we review the invaluable insights this model organism has yielded about diseases caused by mitochondrial dysfunction, which ranges from primary defects in oxidative phosphorylation to metabolic disorders, as well as dysfunctions in maintaining the genome or in the dynamics of mitochondria. Owing to the high level of functional conservation between yeast and human mitochondrial genes, several yeast species have been instrumental in revealing the molecular mechanisms of pathogenic human mitochondrial gene mutations. Importantly, such insights have pointed to potential therapeutic targets, as have genetic and chemical screens using yeast.

Keywords: Drug screening; Genetic suppressors; Mitochondrial disease; OXPHOS; Yeast.

Fig. 1.

Overview of mitochondrial processes and components. This is a schematic cross-section of a human mitochondrion, showing a number of components involved in mitochondrial function. (1) Import and sorting of proteins of nuclear origin: the translocase of the outer membrane (TOM) complex mediates translocation of proteins across or into the outer membrane (OM); sorting and assembly machinery (SAM; also known as TOB) facilitates protein insertion from the intermembrane space (IMS) into the OM; TIM23 takes in proteins with a cleavable mitochondrial targeting sequence (MTS), directing them either into the matrix (M) (when associated to PAM) or the inner membrane (IM) (when associated to Tim21); Twin-cys (the so-called mitochondrial disulfide relay system) mediates, in a redox-dependent manner, the delivery into the IMS of proteins containing specific cysteine motifs; TIM22, together with small soluble proteins in the IMS (called Tim), delivers into the IM the proteins of the so-called mitochondrial carrier family (MCF) that lack a cleavable MTS; OXA1 helps the insertion of proteins from the matrix into the IM. (2) Mitochondrial quality control: misfolded and damaged mitochondrial proteins and organelles are eliminated by proteases and chaperones present in the IM (i-AAA, m-AAA) or the matrix (Lon, ClpXP) by the cytosolic ubiquitin proteasome system (UPS), and by the PINK1 and Parkin proteins at the surface of mitochondria. Fusion (which is mediated by MNF1, MNF2, OPA1L and OPA1S) and fission (mediated by DRP1) of mitochondria contribute also to mitochondrial quality surveillance. (3) mtDNA maintenance and expression: mtDNA is packaged into structures called nucleoids that contain proteins involved in mtDNA maintenance (PolG, Twinkle, mt-SSB), RNA synthesis (TFAM), and the processing of RNAs into messenger (mRNA), transfer (tRNA) and ribosomal (rRNA) RNAs, which are then used to translate the mtDNA-encoded proteins on mitochondrial ribosomes. (4) OXPHOS assembly: the nDNA-encoded subunits of the OXPHOS system (Nuc OXPHOS subunits; CI-V) assemble with their partner subunits of mitochondrial origin (Mt OXPHOS subunits; all except CII, which is entirely encoded by nDNA) together with their redox prosthetic groups (heme and FeS, which are in part synthetized in the mitochondria, and Cu²⁺). CI-IV together with ubiquinone (Q) and cytochrome c (c) transfer electrons to oxygen from reduced cofactors (NADH, FADH) produced by the Krebs cycle, which is coupled to the pumping of protons out of the matrix. The protons are transported back into the matrix by CV, which is coupled to ATP synthesis from ADP and inorganic phosphate (Pi) (see Fig. 2 for details). (5) Transport of metabolites: systems in the OM (VDAC; also known as porin) and IM [MCF (mitochondrial carrier family)] enable the transport of small solutes and ions into and outside the organelle. Parts of the IM protrude into the matrix, forming the cristae, at the basis of which narrow tubular structures termed ‘cristae junctions’ are maintained by proteins of the mitochondrial inner membrane organizing system (MINOS) complex.

Fig. 2.

Mammalian versus yeast OXPHOS system. The figure shows the main enzymatic systems involved in mitochondrial oxidative phosphorylation (OXPHOS) in yeast and mammals. In mammals (top), complexes I-IV together with ubiquinone (Q) and cytochrome c (cyt c) transfer electrons to oxygen from the NADH and succinate produced by the Krebs cycle. These transfers are, at the level of complexes I, III and IV, coupled to proton translocation from the matrix into the intermembrane space (IMS). The resulting proton gradient across the inner mitochondrial membrane (IM) is used by complex V (F1Fo ATP synthase) to produce ATP from ADP and inorganic phosphate (Pi). Part of the ATP produced in the matrix is exchanged against cytosolic ADP by the ADP/ATP translocase (ANT) to provide the whole cell with energy and to maintain the ADP phosphorylation capacity of mitochondria. The OXPHOS system of S. cerevisiae (bottom) is highly similar to the mammalian system except that complex I is replaced by a non-proton-translocating NADH dehydrogenase (Ndi1p) at the inner side of the IM. There are also in S. cerevisiae two NADH dehydrogenases on the external side of the IM (Nde1p, Nde2p) that deliver electrons at the level of ubiquinone. The protein structures are from the Protein Data Bank (PDB) and are at the same scale (indicated by the scale bar).

Fig. 3.

Construction of yeast models of a human mtDNA pathogenic mutation. Schematic of the steps used to create a yeast model of a mutation of the human mitochondrial ATP6 gene, which causes neuropathy ataxia retinitis pigmentosa (NARP) syndrome. (A) In this approach, a plasmid containing a mutant version of the yeast ATP6 gene that carries the NARP-associated mutation is created (ATP6-NARP). This is introduced into the mitochondria of a ρ⁰ arg8Δ kar1 strain, which is devoid of mtDNA (ρ⁰), has a null allele of the nuclear ARG8 gene (arg8Δ) and a mutation (kar1) that strongly delays nuclear fusion, which allows the transfer of mtDNA from one nuclear haploid background to another (Conde and Fink, 1976). (B) The resulting ρ⁻ synthetic strain, which fails to grow in the absence of external arginine (ARG⁻), is crossed with (C) an arginine prototrophic (ARG⁺) strain that contains wild-type (ρ⁺) mtDNA but is deleted for ATP6 (atp6Δ). ARG8m is a mitochondrial version of a nuclear gene (ARG8) and encodes a yeast mitochondrial protein involved in arginine biosynthesis (Steele et al., 1996). (D) Because the ARG8m clone used to delete ATP6 is flanked on each side by ∼100 bp of the ATP6 locus, homologous recombination (E) can mediate the replacement of ARG8m with the ATP6-NARP gene. (F) Mitotic segregation then produces ρ⁺ cells with the NARP-associated ATP6 mutation in a pure (homoplasmic) form that can be identified by virtue of their inability to grow in the absence of arginine.

Fig. 4.

A yeast-based assay to identify drugs that are active against mitochondrial disorders. (A) A respiratory-deficient yeast model of a mitochondrial disease is grown in glucose. (B) Subsequently, yeast cells are spread onto a solid medium containing a non-fermentable substrate (glycerol), on which they grow very poorly. (C) Small sterile filters are placed on the agar surface and (D) spotted with compounds from a chemical library; the plate is then incubated for several days. (E) After incubation, active drugs that improve mitochondrial function in the yeast disease model result in the appearance of a halo of enhanced growth around the corresponding filters.