The ADP/ATP translocator is not essential for the mitochondrial permeability transition pore

Abstract

A sudden increase in permeability of the inner mitochondrial membrane, the so-called mitochondrial permeability transition, is a common feature of apoptosis and is mediated by the mitochondrial permeability transition pore (mtPTP). It is thought that the mtPTP is a protein complex formed by the voltage-dependent anion channel, members of the pro- and anti-apoptotic BAX-BCL2 protein family, cyclophilin D, and the adenine nucleotide (ADP/ATP) translocators (ANTs). The latter exchange mitochondrial ATP for cytosolic ADP and have been implicated in cell death. To investigate the role of the ANTs in the mtPTP, we genetically inactivated the two isoforms of ANT in mouse liver and analysed mtPTP activation in isolated mitochondria and the induction of cell death in hepatocytes. Mitochondria lacking ANT could still be induced to undergo permeability transition, resulting in release of cytochrome c. However, more Ca2+ than usual was required to activate the mtPTP, and the pore could no longer be regulated by ANT ligands. Moreover, hepatocytes without ANT remained competent to respond to various initiators of cell death. Therefore, ANTs are non-essential structural components of the mtPTP, although they do contribute to its regulation.

Figure 1

Preparation of a CRE-conditional Ant2-null mutant allele in mouse embryonic stem cells. a, Top to bottom, maps of the wild-type Ant2 locus, the region of homology within the targeting construct, and the floxed (Ant2^fl) and mutant (Ant2^–) alleles. Exons (black boxes), loxP sites flanking exons 3 and 4 (triangles), PGKneo selection cassette within 3′UTR (hatched box). a, b, Arrows indicate sizes of Xba I (X) and Dra III (D) restriction fragments used to identify correctly targeted clones by Southern analysis of embryonic-stem-cell genomic DNA using probes A and B. b, Southern blot of targeted and untargeted embryonic-stem-cell clones.

Figure 2

Inactivation of ANT in liver mitochondria. a, Western blots of mitochondria using ANT1 and ANT2-specific antibodies. Heart mitochondria are a positive ANT1 control. ANT1 and ANT2 are both 30 kDa (arrows). The Ponceau-S-stained blot (below) is the loading control. b, ADP-stimulated mitochondrial respiration. Arrowheads denote addition of 600 μg of mitochondrial protein, 125 nmol of ADP and 75 nM FCCP. c, OXPHOS enzyme activities of ANT-control (white bars) or ANT-deficient (black bars) liver mitochondria (mean of three mice each). OXPHOS complexes I–IV and fumarase (Fum.) are shown. Error bars show the standard error. d, Western blots of mitochondria from three independent mice for COI (56 kDa), UCP2 (33 kDa), cytochrome (cyt) c (11 kDa). The Ponceau S blot (below) is the loading control.

Figure 3

Effect of ANT-deficiency on mtPTP activation in liver mitochondria. a, ANT-control and b, ANT-deficient mtPTP activation by ΔP depolarization (FCCP) monitored by mitochondrial swelling. Black trace, no CsA; red trace, with CsA. c, Amount of Ca²⁺ required per mg of mitochondrial protein to activate the mtPTP and collapse ΔP. d, Western blots of cytochrome c released into the supernatant fraction versus COI retained in the mitochondrial pellet following Ca²⁺-activation of the mtPTP. e, ANT-control and f, ANT-deficient Ca²⁺-activation of the mtPTP in the presence of positive and negative effectors.

Figure 4

Induction of cell death in ANT-deficient mouse hepatocytes. a, Transmission electron microscope images of ANT-deficient hepatocytes. Control, untreated; Br-A23187, treated for 1 h with 10 μM Br-A23187. b, Calcium ionophore (Br-A23187) induced cell death (% of LDH released) of ANT-deficient treated hepatocytes relative to untreated hepatocytes, without or with pretreatment by either CsA or z-VAD. c, Receptor-mediated apoptosis (Hoechst 33258 analysed nuclear fragmentation) induced by TNF-α or soluble Fas ligand treatment, with or without actinomycin D (ActD). KO, knockout.