Dominant membrane uncoupling by mutant adenine nucleotide translocase in mitochondrial diseases

Abstract

Adenine nucleotide translocase (Ant) is the most abundant protein on the mitochondrial inner membrane (MIM) primarily involved in ADP/ATP exchange. Ant also possesses a discrete membrane uncoupling activity. Specific mis-sense mutations in the human Ant1 cause autosomal dominant Progressive External Ophthalmoplegia (adPEO), mitochondrial myopathy and cardiomyopathy, which are commonly manifested by fractional mitochondrial DNA (mtDNA) deletions. It is currently thought that the pathogenic mutations alter substrate preference (e.g. ATP versus ADP) thereby dominantly disturbing adenine nucleotide homeostasis in mitochondria. This may interfere with mtDNA replication, consequently affecting mtDNA stability and oxidative phosphorylation. Here, we showed that the adPEO-type A128P, A106D and M114P mutations in the yeast Aac2p share the following common dominant phenotypes: electron transport chain damage, intolerance to moderate over-expression, synthetic lethality with low Deltapsi(m) conditions, hypersensitivity to the uncoupler carbonyl cyanide m-chlorophenylhydrazone (CCCP) and mtDNA instability. More interestingly, the aac2(A137D) allele mimicking ant1(A123D) in mitochondrial myopathy and cardiomyopathy exhibits similar dominant phenotypes. Because Aac2(A137D) is known to completely lack transport activity, it is strongly argued that the dominant mitochondrial damages are not caused by aberrant nucleotide transport. The four pathogenic mutations occur in a structurally dynamic gating region on the cytosolic side. We provided direct evidence that the mutant alleles uncouple mitochondrial respiration. The pathogenic mutations likely enhance the intrinsic proton-conducting activity of Ant, which excessively uncouples the MIM thereby affecting energy transduction and mitochondrial biogenesis. mtDNA disintegration is a phenotype co-lateral to mitochondrial damages. These findings provide mechanistic insights into the pathogenesis of the Ant1-induced diseases.

Figure 1.

Dominant phenotypes associated with aac2 alleles equivalent to the adPEO-type pathogenic mutations in human Ant1. (A) Projected localization of Ala106, Met114, Ala128 and Ala137 (red) in the yeast Aac2p on the crystal structure of the bovine Ant1 in the cytosolic conformation bound by carboxyatractyloside (yellow) (2). M, matrix; IMS, inter-membrane space; H2, α-helix 2; H3, α-helix 3. (B) The aac2^A128P, aac2^A106D and aac2^M114P alleles dominant-negatively inhibit cell growth on YPD medium at 25°C and on YPD supplemented with EB. The strains 6A/UAU (+AAC2), 6A/UPU (+aac2^A128P), 6A/UDU (+aac2^A106D) and 6A/UP2U (+aac2^M114P) were grown in YPD at 30°C for overnight, serially diluted with water and spotted onto YPD, YPGly and YPD supplemented with EB. The plates were incubated at the temperatures as indicated for 5–7 days before being photographed. (C) Relative oxygen consumption rate of 6A/UAU (+AAC2), 6A/UPU (+aac2^A128P), 6A/UDU (+aac2^A106D) and 6A/UP2U (+aac2^M114P) cells incubated at 30°C, or at 25°C for approximately 5, 10 and 15 generations. The relative oxygen consumption rates are normalized against 6A/UAU. (D) Representative tetrads showing the growth phenotype of the meiotic segregants that express two extra copies of AAC2 and aac2^A128P (indicated by the arrows). The asci were dissected on YPD plates which were incubated at 30°C for 4 days.

Figure 2.

Petite formation and cell death in strains expressing one or two copies of chromosomally integrated aac2^A128P, aac2^A106D and aac2^M114P. Cells were grown at 30°C for 2–3 days, diluted in water and plated onto YPD (Glu) and YPGal+Raf (Gal+Raf) to an estimated cell density of 200 cells per plate. The plates were incubated at 30 or 25°C for 5–7 days.

Figure 3.

Dominant mitochondrial damages induced by the aac2^A137D allele. (A) Growth phenotype of cells of M2915-6A background co-expressing aac2^A137D and AAC2 on YPD, YPGly and YPD supplemented with EB at 30°C. (B) Petite formation and loss of cell viability in strains of W303 background expressing one or two copies of the chromosomally integrated aac2^A137D allele. Cells were grown at 30°C for 2–3 days, diluted in water and plated onto YPD (Glu) and YPGal+Raf (Gal+Raf) to an estimated cell density of 200 cells per plate. The plates were incubated at 30 or 25°C for 5–7 days. (C) A representative southern blot showing mtDNA content in M2915-6A (wild-type; lanes 1 and 4), CS1481 (lys2Δ::aac2^A137D-kan; lanes 2 and 5) and CS1487-1C (lys2Δ::aac2^A137D-kan, trp12Δ::aac2^A137D-URA3; lanes 3 and 6). Cells were grown in YPD at 25 and 30°C, respectively. (D) Western-blot analysis of isolated mitochondria showing the levels of the mtDNA-encoded Cox2p in the same cultures as in (C). Ilv5p is used as a marker for mitochondrial proteins. (E) A representative southern blot showing mtDNA content in M2915-6A (wild-type; lanes 1 and 3) and CS1487-1C (lys2Δ::aac2^A137D-kan, trp12Δ::aac2^A137D-URA3; lanes 2 and 4), grown in YPGal+Raf at 25 and 30°C, respectively. (F) Relative mtDNA copy number in CS1487-1C grown in YPGal+Raf at 25 and 30°C, respectively. Errors bars are standard deviations of three independent experiments.

Figure 4.

The mutant aac2 alleles are hypersensitive to low Δψ_m conditions in a dominant manner. (A) Hypersensitivity to yme1Δ. Diploid strains with the indicated phenotypes were sporulated and dissected on YPD. The plates were incubated at 30°C for 4–5 days. Circled are spores deduced to have the combination of yme1Δ::LEU2 with the chromosomally integrated copy of AAC2-URA3, aac2^A128P-URA3, aac2^A106D-URA3, aac2^M114P-URA3 or aac2^A137D-URA3. (B) Hypersensitivity to CCCP. The strains 6A/UAU (+AAC2), 6A/UPU (+aac2^A128P), 6A/UDU (+aac2^A106D), 6A/UP2U (+aac2^M114P) and CS1426/1 (+aac2^A137D) were grown on YPD at 30°C for 4 days in the presence or absence of CCCP (120 µm).

Figure 5.

The aac2^A128P, aac2^M114P, aac2^A106D and aac2^A137D alleles induce membrane uncoupling. Yeast cells were grown in YPGal at 30°C (A) or 25°C (B) and harvested at later exponential phase. Isolated mitochondria were immediately used for respiratory assays using ethanol as substrate. RCR was deduced by dividing the State 3 respiratory rate (induced by ADP) by the State 4 rate. A low RCR is an indication of uncoupled respiration and an RCR of 1 would indicate that the mitochondria are completely uncoupled. For all experiments, the values are averages of 4–6 measurements and are compared with the wild-type control. The standard errors and P-values (Unpaired Student's t-test) are indicated.