A new set of mutations in the second transmembrane helix of the Cox2p-W56R substantially improves its allotopic expression in Saccharomyces cerevisiae

Abstract

The dual genetic control of mitochondrial respiratory function, combined with the high mutation rate of the mitochondrial genome (mtDNA), makes mitochondrial diseases among the most frequent genetic diseases in humans (1 in 5,000 in adults). With no effective treatments available, gene therapy approaches have been proposed. Notably, several studies have demonstrated the potential for nuclear expression of a healthy copy of a dysfunctional mitochondrial gene, referred to as allotopic expression, to help recover respiratory function. However, allotopic expression conditions require significant optimization. We harnessed engineering biology tools to improve the allotopic expression of the COX2-W56R gene in the budding yeast Saccharomyces cerevisiae. Through conducting random mutagenesis and screening of the impact of vector copy number, promoter, and mitochondrial targeting sequence, we substantially increased the mitochondrial incorporation of the allotopic protein and significantly increased recovery of mitochondrial respiration. Moreover, CN-PAGE analyses revealed that our optimized allotopic protein does not impact cytochrome c oxidase assembly, or the biogenesis of respiratory chain supercomplexes. Importantly, the most beneficial amino acid substitutions found in the second transmembrane helix (L93S and I102K) are conserved residues in the corresponding positions of human MT-CO2 (L73 and L75), and we propose that mirroring these changes could potentially help improve allotopic Cox2p expression in human cells. To conclude, this study demonstrates the effectiveness of using engineering biology approaches to optimise allotopic expression of mitochondrial genes in the baker's yeast.

Keywords: allotopic expression; engineering biology; mitochondria; random mutagenesis; yeast.

Fig. 1.

a) Schematic representation of the design used to replicate expression vectors from Supekova et al. 2010 using the YeastFab assembly. b) Spot test presenting the growth benefit conferred by the different COX2-W56R allotopic constructs in the Δcox2::ARG8^m strain, when hosted on a low- or a high-copy plasmid after 3 days at 30°C. c) Oxygen consumption rate (OCR, nmol.min⁻¹.OD_600nm unit⁻¹, mean ± SD, n = 3) measured in whole cells at the basal level and following addition of Ethanol (EtOH) and CCCP. d) Western blot-based quantification of the Cox2p abundance relative to the Pgk1p control [mean (expressed as a percentage of the WT level) ± SD, n = 3]. A, B, C and D: Denotes a significant difference between the conditions (P < 0.05). U, unprocessed; M, mature.

Fig. 2.

a) Normalized GFP signal (GFP signal/OD_600nm) measured in the BY4741 strain expressing the different GFP constructs in glucose-containing rich media (2% glucose for 24 hours, upper chart) or ethanol-containing rich media (2% EtOH for 48 hours, lower chart) (n = 3). b) Spot test performed to assess the growth benefit conferred by the COX2-W56R allotopic construct expressed under the selected promoters after 7 days at 30°C (from ACT1, ADH1, ICL1, or JEN1). c) Oxygen consumption rate (OCR, nmol.min⁻¹.OD_600nm unit ⁻¹) measured in whole cells at the basal level and following addition of Ethanol (EtOH) and CCCP. d) Western blot-based quantification of the Cox2p abundance relative to the control Pgk1p [mean (expressed as a percentage of the WT level) ± SD, n = 3]. AFU, arbitrary fluorescence unit. A, B, C and D: denotes a significant difference between the conditions (P < 0.05). U, unprocessed; M, mature.

Fig. 3.

a) Spot test of the 9 isolated best growing epPCR clones, grown for 3 days at 30°C. b) Spot assay testing the additivity of the mutations identified in the mutational hot spot in TMH2, grown for 4 days at 30°C. c) Oxygen consumption rate (OCR, nmol.min⁻¹.OD_600nm unit ⁻¹) measured in whole cells at the basal level and following addition of ethanol (EtOH) and CCCP. d) Western blot-based quantification of Cox2p abundance relative to the control Pgk1p [mean (expressed as a percentage of the WT level) ± SD, n = 3]. e) Annotation of the functional mutations identified in the protein sequence in the isolated epPCR clones E6, E8, C7, and F9 and their corresponding switch in hydrophobicity on the Cox2p protein structure generated using the SWISS-MODEL tool on the ExPASy Server and coloured based on amino acids hydrophobicity (red meaning hydrophobic and white hydrophilic) using PyMol. A, B, C, D, and E: denotes a significant difference between the conditions (P < 0.05). U, unprocessed; M, mature.

Fig. 4.

a) Schematic representation of constructs for the MTS screening using the sub-optimal allotopic construct from epPCR clone E6. b) Fluorescence imaging of the BY4741 rho⁺ strain expressing the different MTS-GFP constructs. Scale bar = 5 μm. c) Spot test presenting the respiratory growth benefit conferred by the screened MTSs as compared to OXA1 MTS after 4 days at 30°C. d) Oxygen consumption rate (OCR, nmol.min⁻¹.OD_600nm unit ⁻¹) measured in whole cells at the basal level and following addition of ethanol (EtOH) and CCCP. e) Western blot-based quantification of the Cox2p abundance relatively to the control Pgk1p [mean (expressed as a percentage of the WT level) ± SD, n = 3]. A, B, C, D, and E: denotes a significant difference between the conditions (P < 0.05). U, unprocessed; M, mature.

Fig. 5.

a) Spot test comparing the growth benefit conferred by the initial construct from supekova and collaborators and the most optimized expression system generated in this study. Shown is growth after 4 days at 30°C. b) Western blot–based quantification of the Cox1p and Cox2p abundances relatively to the control Porin [mean (expressed as a percentage of the WT level) ± SD, n = 3]. c) Oxygen consumption rate (OCR, nmol.min⁻¹.OD_600nm unit ⁻¹) and ATP synthesis rate (nmol Pi.min⁻¹.mg prot⁻¹) measured in isolated mitochondria. d) CN-PAGE analysis performed on digitonin-solubilized mitochondria. Lane 1: WT; Lane 2: original construct on high-copy vector; Lane 3: original construct on low-copy vector; Lane 4: optimized version on low-copy vector. NADH DH: NADH dehydrogenase. A, B, C and D: denotes a significant difference between the conditions (P < 0.05). *Denotes a significant difference as compared to the WT (*P < 0.005; ***P < 0.005). OC-HC: previously described construct⁵ hosted on a high-copy vector. OC-LC, previously described construct⁵ hosted on a low-copy vector; OV, the optimized version from this study hosted on a low-copy vector.