Mitochondrial Sco proteins are involved in oxidative stress defense

Abstract

Members of the evolutionary conserved Sco protein family have been intensively studied regarding their role in the assembly of the mitochondrial cytochrome c oxidase. However, experimental and structural data, specifically the presence of a thioredoxin-like fold, suggest that Sco proteins may also play a role in redox homeostasis. In our study, we addressed this putative function of Sco proteins using Saccharomyces cerevisiae as a model system. Like many eukaryotes, this yeast possesses two SCO homologs (SCO1 and SCO2). Mutants bearing a deletion of either of the two genes are not affected in their growth under oxidative stress. However, the concomitant deletion of the SOD1 gene encoding the superoxide dismutase 1 resulted in a distinct phenotype: double deletion strains lacking SCO1 or SCO2 and SOD1 are highly sensitive to oxidative stress and show dramatically increased ROS levels. The respiratory competent double deletion strain Δsco2Δsod1 paved the way to investigate the putative antioxidant function of SCO homologs apart from their role in respiration by complementation analysis. Sco homologs from Drosophila, Arabidopsis, human and two other yeast species were integrated into the genome of the double deletion mutant and the transformants were analyzed for their growth under oxidative stress. Interestingly, all homologs except for Kluyveromyces lactis K07152 and Arabidopsis thaliana HCC1 were able to complement the phenotype, indicating their role in oxidative stress defense. We further applied this complementation-based system to investigate whether pathogenic point mutations affect the putative antioxidant role of hSco2. Surprisingly, all of the mutant alleles failed to restore the ROS-sensitivity of the Δsco2Δsod1 strain. In conclusion, our data not only provide clear evidence for the function of Sco proteins in oxidative stress defense but also offer a valuable tool to investigate this role for other homologous proteins.

Keywords: Mitochondria; Oxidative stress response; ROS; Saccharomyces cerevisiae; Sco proteins; Sod1.

Graphical abstract

Fig. 1

Growth analysis under oxidative stress. Cells of wild type (WT), the rho⁰-strain KL14-4a and the indicated single and double deletion strains were dropped in a dilution series (10⁴–10¹ cells) onto YPD plates with the indicated concentrations of menadione (MD), paraquat (PQ) and L- ascorbic acid (L-AA). Growth was documented after incubation at 30 °C for three days.

Fig. 2

Measurement of ROS levels in yeastSCOandSODdeletion mutants. Wild type (WT), the rho⁰-strain KL14-4a and the indicated single and double deletion strains were grown in YPD with or without the addition of 1 mM PQ for 24 h, stained with DCF-DA and analyzed by flow cytometry. Mean fluorescence intensities (MFI) of the cell populations were measured and given values indicate the ratio of treated (+PQ) to untreated (-PQ) sample (± standard deviation). Data derive from four independent experiments. Mean values were compared with WT using the unpaired two-tailed t-test; ** p-value≤ 0.01, *** p-value≤ 0.001.

Fig. 3

SOD (A) and COX (B) activity inSCOandSODsingle and double deletion mutants. A. Mitochondria and cytoplasmic fractions were prepared and SOD activity was measured and calculated as described in the material and methods section. Mean values derive from three independent measurements (± standard deviation). B. COX activity was determined in purified mitochondria by measuring the conversion of reduced cytochrome c to its oxidized form. Mean values derive from triplicates of three independent experiments (± standard deviation). Values were compared with the respective WT sample using the unpaired two-tailed t-test; * p-value ≤ 0.05, ** p-value ≤ 0.01, *** p-value ≤ 0.001.

Fig. 4

Complementation analysis withSCOhomologs from different organisms. A. Overview of the analyzed Sco homologs: their structural features and sequence similarity to ySco2 are depicted. The positions of the mitochondrial target sequence (orange), transmembrane domain (red), thioredoxin-like domain (yellow), putative copper-binding motif CxxxC (green) and conserved histidine residue (black) are highlighted. B. Cells of the WT, the Δsco2Δsod1 mutant strain and its transformed derivatives expressing the indicated Sco homologs were dropped in a dilution series (10⁴ to 10¹ cells) onto YPD plates with or without 100 µM PQ. Growth was documented after incubation at 30 °C for three days.

Fig. 5

Levels of extracellular H₂O₂(A) and lipid peroxidation levels (B) in deletion and recombinant yeast strains. The indicated yeast strains were grown in YPD in the presence or absence of 100 µM PQ for 24 h. A. The cells were incubated with Amplex Red and the amount of hydrogen peroxide (µM H₂O₂/10⁹ cells) was calculated as described in the material and methods section. B. The malondialdehyde (MDA) concentration as an indicator for cellular lipid peroxidation was measured and normalized to the total protein amount (µM MDA/mg protein). Given data indicate the ratio of treated (+PQ) to untreated (-PQ) sample (± standard deviation) and are mean values from at least three independent experiments. Values were compared with WT using the unpaired two-tailed t-test; * p-value ≤ 0.05, ** p-value ≤ 0.01, *** p-value ≤ 0.001.

Fig. 6

Complementation assay with hSco2 variants harboring pathogenic point mutations. A. Scheme of the location of selected pathogenic mutations in the structure of hSco2. The protein backbone of the soluble hSco2 is given in grey. Mutations are shown at their exact positions on the protein structure and are highlighted in different colors. The figure was prepared with PyMOL by rearranging the available structure data retrieved from PDB (PDB: 2RLI). B. Cells of the wild type (WT), the Δsco2Δsod1 mutant strain and its derivatives expressing different hSco2 variants were dropped in a dilution series (10⁴–10¹ cells) onto YPD plates with or without 100 µM PQ. Growth was documented after incubation at 30 °C for three days.