Stable nuclear expression of ATP8 and ATP6 genes rescues a mtDNA Complex V null mutant

Abstract

We explore the possibility of re-engineering mitochondrial genes and expressing them from the nucleus as an approach to rescue defects arising from mitochondrial DNA mutations. We have used a patient cybrid cell line with a single point mutation in the overlap region of the ATP8 and ATP6 genes of the human mitochondrial genome. These cells are null for the ATP8 protein, have significantly lowered ATP6 protein levels and no Complex V function. Nuclear expression of only the ATP8 gene with the ATP5G1 mitochondrial targeting sequence appended restored viability on Krebs cycle substrates and ATP synthesis capabilities but, failed to restore ATP hydrolysis and was insensitive to various inhibitors of oxidative phosphorylation. Co-expressing both ATP8 and ATP6 genes under similar conditions resulted in stable protein expression leading to successful integration into Complex V of the oxidative phosphorylation machinery. Tests for ATP hydrolysis / synthesis, oxygen consumption, glycolytic metabolism and viability all indicate a significant functional rescue of the mutant phenotype (including re-assembly of Complex V) following stable co-expression of ATP8 and ATP6 Thus, we report the stable allotopic expression, import and function of two mitochondria encoded genes, ATP8 and ATP6, resulting in simultaneous rescue of the loss of both mitochondrial proteins.

Figure 1.

Description and detection of the m.8529G→A (A8/A6^mut) mutation. (A) Sequence of WT 143B and mutant m.8529G→A mtDNA showing the point mutation and impact on the coding of ATP8 and ATP6. (B) Detection of wild-type (WT) and mutant sequences using ARMS qRT-PCR. (C) Confirmation of mutant homoplasmy by qRT-PCR. ΔCt = C_T mutant – C_T wt, normalized using MT-CYB as the housekeeping gene. * No discernible cell survival under these conditions.

Figure 2.

A8/A6^mut cells stably express ATP8 alone or ATP8 + ATP6 proteins and target to mitochondria. Purified mitochondrial fractions from stable cell lines expressing ATP8 (A8/A6^mut + A8F) and/or ATP6 (A8/A6^mut + A8F + A6F, A8/A6mut + A6F). (A) Immuno-detection for anti-ATP8 (left panel, the ⁺ at ∼26 kD indicates a cross-reacting band), anti-FLAG (center panel; reprobe of left panel) and anti-ATP6 (right panel) antibodies. Resolved on a 4–12% denaturing PAGE. TIM23 protein was used as loading control (bottom panel). (B) Transient (lanes 2 and 3) and stable (lanes 5 and 6) expression of ATP8 protein in WT and A8/A6^mut. Resolved on a 4–12% denaturing PAGE. (C) Stable expression of G1MTS-ATP8-FLAG and G1MTS-ATP6-FLAG proteins in A8/A6^mut cells. Resolved on a 12% denaturing PAGE. TIM23 protein was used as loading control (bottom panel). (D) Monitoring tryptic peptide LITTQQWLIK (at m/z 622.37 with z = 2) derived from protein ATP6 by mass spectrometry. Data-independent SWATH acquisitions were analyzed using the Skyline algorithm, extracted ion chromatograms (XICs) were processed and normalized peak areas for LITTQQWLIK were used to quantify relative changes in the abundance of ATP6 comparing different human cell lines WT, A8/6^mut and A8/6^mut + A8F cells. Each cell line was acquired in biological triplicates, the A8/6^mut + A8F strain was analyzed from four biological replicates, and normalized peak areas from all biological replicates and strains are displayed. A statistically significant ratio of 0.13 was measured comparing relative ATP6 protein levels between A8/6^mut versus WT strains with a p-value of 0.009(**). The reduction in relative ATP6 protein levels in the mutant strain A8/6^mut were significantly lower compared to the ATP6 levels in the rescue mutant A8/6^mut + A8F strain. The relative ATP6 protein ratio comparing A8/6^mut + A8F versus WT strains was measured at 0.36 with a P-value of 0.018 (*). (E) Co-localization of FLAG (Green) with Mitotracker (Red) in A8/A6^mut cells stably expressing G1MTS-ATP8-FLAG and G1MTS-ATP6-FLAG.

Figure 3.

OxPhos complexes reassemble upon exogenous expression of ATP8 and ATP6 proteins in the A8/A6^mut cell line. Mitochondria-enriched fractions (∼25 μg protein/lane) from WT (lane 1), A8/A6^mut + A8F + A6F (lane 2), A8/A6^mut + A8F (lane 3) and A8/A6^mut (lane 4) were electrophoresed on 4–12% Bis-Tris BN-PAGE gels. Proteins were transferred onto PVDF membranes and immunodetected with the following antibodies: (A) panel i: anti-ATP8, ii: anti-FLAG, iii: anti-ATP5h, iv: anti-ATP5O (OSCP), and (B) panel i: anti-NDUFS4 (Complex I), ii: anti-SDHB (Complex II), iii: anti-Core 2 protein (Complex III), iv: anti-MT-CO2 (Complex IV). Protein standards are indicated on the left. * Immunodetected protein band, ** CV dimers, ⁺degraded products of Complex III.

Figure 4.

Restoration of Complex V ATPase activity. Mitochondrial proteins (∼25 μg) from WT (lane 1), A8/A6^mut (lane 2), A8/A6^mut + A8F (lane 3) and A8/A6^mut + A8F + A6F (lane 4) were electrophoresed in 3–12% Bis-Tris, NativePAGE protein gels and incubated with assay buffer for 24 h. The in-gel activity was documented and subsequently the same gel was stained with 0.2% Coomassie blue R250. (A) Coomassie stain and (B) in-gel ATPase activity. Protein standards are indicated on the left.

Figure 5.

Metabolic preference of mutant cells switches with expression of ATP8/6. Cells were pre-incubated with an excess of glucose and sodium pyruvate. After basal rates were measured, CV function was inhibited by addition of oligomycin. Electron transport was then uncoupled by addition of FCCP, and finally all electron transport was inhibited by addition of rotenone and antimycin A. X-axis denotes measurement points. Cell lines were tested in quadruplicate in each of 3 independent experiments and results were normalized to cell number as estimated by DNA content. (A) Proton Production Rate (PPR) as a crude measure of glycolysis. (B) Oxygen consumption rate (OCR).

Figure 6.

Influence of carbon source and OxPhos inhibitors on total cellular ATP content. (A) Total cellular ATP levels (normalized to total DNA content) were determined after 16 h of incubation in respective carbon sources, glucose (5 mM), galactose (5 mM) or pyruvate (5 mM) by luciferase assay as described in Materials and Methods. Values are expressed as percent ATP content relative to glucose as the carbon source for WT. *P = 1.1 × 10⁻³ and **P < 1 × 10⁻⁵. (B) Cell lines were grown in complete medium for 24 h and then shifted to complete medium plus the respective inhibitor: antimycin A 5 μM, oligomycin 5 μg/ml and valinomycin 5 μM for 16 h, washed once with PBS and the total ATP content measured by luciferase assay as in A. Values are expressed percent relative to untreated WT sample. *P < 0.001 and ** P = 1.4 × 10⁻³. Error is ± SEM, N = 6.

Figure 7.

Viability of mutant and ATP8/6 expressing cells under restrictive growth conditions. Cells were cultured in media containing no glucose but supplemented with 25 mM galactose. Equal numbers of cells were plated in glucose containing media in replicates of 5, 24 h later cells were switched to galactose media. After another 24 h, one set of cells was collected and counted (day 1) and another on each successive day. Cells were counted by hemocytometer and viability determined by trypan blue exclusion. Results are an average of 5 independent experiments and error bars are SEM.