The isolation and reconstitution of the ADP/ATP carrier from wild-type Saccharomyces cerevisiae. Identification of primarily one type (AAC-2)

Methods for isolation of the ADP/ATP carrier (AAC) from yeast (Saccharomyces cerevisiae) are described which allow separation of the carrier from the initially copurified porin which poses a specific problem in yeast. The procedure varies according to whether one wishes to obtain a stable CAT-AAC complex, the free and active AAC for reconstitution, or the SDS-denatured pure AAC peptide. CNBr cleavage of AAC enabled us to differentiate clearly between isogenes AAC-1 and AAC-2 recently found in yeast, due to the exclusive occurrence of a methionine (M-115) residue at the end of the first domain in AAC-2. Thus the AAC isolated from wild-type yeast is primarily or exclusively AAC-2. The isolated AAC is active in ADP/ATP exchange in reconstituted liposomes with a Vmax of 1100 mumol/min per g protein and Km = 15 microM for ADP, and a Vmax of 900 mumol/min per g protein and Km = 9 microM for ATP.     

The Reconstituted ADP / ATP carrier from mitochondria is both inhibited and activated by anions

Anions were found to have a number of different effects on the reconstituted ADP / ATP carrier from mitochondria. (1) Binding of adenine nucleotides to the active site of the translocator is competitively inhibited by various anions. These anions can be arranged in a sequence of increasing competitive effect due to their order in a lyotropic series, and also due to increasing charge. (2) Apart from this competition effect, the presence of a sufficiently high concentration of anions turned out to be absolutely essential for functional ADP / ATP exchange in the reconstituted system. The activating anions too can be arranged in sequence, similar to that of the competition effect. The adenine nucleotide transport shows sigmoidal dependence on the stimulating anions with a Hill coefficient of n = 2. Addition of anions does not change the basic amount of functionally active translocator molecules. (3) The different effects of anions, i.e., inhibition and activation, were shown to take place at different sites and to be due to different mechanisms. Anions compete with substrates both at the outer (cytosolic) and at the inner (matrix) active site, whereas anion activation is observed solely by interaction with the cytosolic side of the translocator protein. (4) Activation of the reconstituted ADP / ATP exchange by anions could be discriminated from an activating influence of anionic phospholipids in the surroundings of the carrier protein.     

The reconstituted ADP/ATP carrier from mitochondria is both inhibited and activated by anions

Anions were found to have a number of different effects on the reconstituted ADP/ATP carrier from mitochondria. (1) Binding of adenine nucleotides to the active site of the translocator is competitively inhibited by various anions. These anions can be arranged in a sequence of increasing competitive effect due to their order in a lyotropic series, and also due to increasing charge. (2) Apart from this competition effect, the presence of a sufficiently high concentration of anions turned out to be absolutely essential for functional ADP/ATP exchange in the reconstituted system. The activating anions too can be arranged in sequence, similar to that of the competition effect. The adenine nucleotide transport shows sigmoidal dependence on the stimulating anions with a Hill coefficient of n = 2. Addition of anions does not change the basic amount of functionally active translocator molecules. (3) The different effects of anions, i.e., inhibition and activation, were shown to take place at different sites and to be due to different mechanisms. Anions compete with substrates both at the outer (cytosolic) and at the inner (matrix) active site, whereas anion activation is observed solely by interaction with the cytosolic side of the translocator protein. (4) Activation of the reconstituted ADP/ATP exchange by anions could be discriminated from an activating influence of anionic phospholipids in the surroundings of the carrier protein.     

The dynamic dimerization of the yeast ADP/ATP carrier in the inner mitochondrial membrane is affected by conserved cysteine residues

The ADP/ATP carrier (AAC) that facilitates the translocation of ATP made in mitochondria is inserted at the inner mitochondrial membrane by the TIM10-TIM22 protein import system. Here we addressed the state of the AAC precursor during insertion (stage IV of import) and identified residues of the carrier important for dimerization. By a combination of (i) import of a mix of His-tagged and untagged versions of AAC either 35S-labeled or unlabeled, (ii) import of a tandem covalent dimer AAC into wild-type mitochondria, and (iii) import of monomeric AAC into mitochondria expressing only the tandem covalent dimer AAC, we found that the stage IV intermediate is a monomer, and this stage is probably the rate-limiting step of insertion in the membrane. Subsequent dimerization occurs extremely rapidly (within less than a minute). The incoming monomer dimerizes with monomeric endogenous AAC suggesting that the AAC dimer is very dynamic. Conserved Cys residues were found not to affect insertion significantly, but they are crucial for the dimerization process to obtain a functional carrier.     

Activation of the ADP/ATP carrier from mitochondria by cationic effectors

The ADP/ATP carrier from the mitochondrial inner membrane was found to be influenced by cationic substances from the hydrophilic surroundings. Under low-ionic-strength conditions, addition of these cationic effectors fully activated the reconstituted adenine nucleotide translocator. The list of activators included divalent cations, polyamines, peptides and cationic proteins. The minimum requirement for an activator to be effective was the presence of at least two positive net charges, regardless of the size of the molecule. Cationic molecules were not activating when an intramolecular charge compensation was possible or when the two charges were too far apart from one another. The affinity of these activators varied from several hundred microM (diaminoalkanes, divalent cations) to 1 microM (cytochrome c, spermine) and even down to a few nM (polylysine). The activation by cations was fully reversible and was not due to fusion processes. It was not mediated by an interaction with the anionic substrates ADP and ATP, nor by interaction with the liposomes. The stimulation could directly and functionally be correlated to the reconstituted carrier protein. Activation was not observed in intact mitochondria, but could be demonstrated when the outer mitochondrial membrane had been removed by treatment with digitonin. These mitoplasts were stimulated by polycations similar to the ADP/ATP carrier in the reconstituted system.     

The structure of the second cytosolic loop of the yeast mitochondrial ADP/ATP carrier AAC2 is dependent on the conformational state

To detect structural changes in the second cytosolic loop of the mitochondrial ADP/ATP carrier of Saccharomyces cerevisiae AAC2, we prepared 20 single cysteine mutants by replacing each amino acid in the S213 to L232 region. All single cysteine mutants were fully functional, because they could restore growth on glycerol of a yeast strain lacking functional ADP/ATP carriers. First, these single-Cys mutants were treated with carboxyatractyloside to lock the carrier in the cytosolic state or with bongkrekic acid to generate the matrix state, and then with the membrane-impermeable SH reagent eosin-5-maleimide (EMA) to probe accessibility. The amino acid residues S213C, L214C, F231C and L232C were not labeled, indicating that these 4 residues must have been buried in the membrane, whereas the region between residues K215 and S230 is accessible to labeling and must, therefore, have protruded into the aqueous phase. Residue L218C showed strong resistance against EMA labeling regardless of the state of the carrier, but the reason for such behavior is unclear. On the contrary, the labeling of the residues between F227C and S230C was strongly dependent on the state of the carrier. Thus, the C-terminal region of the second cytosolic loop in AAC2 changes its environment when the carrier cycles between the matrix and cytosolic state.     

Subunits of the yeast mitochondrial ADP/ATP carrier: cooperation within the dimer

The mitochondrial ADP/ATP carrier, or Ancp, is a member of the mitochondrial carrier family (MCF). It exchanges ADP and ATP between matrix and intermembrane space. It is postulated from numerous experiments that the inactive Ancp bound to one of its inhibitors (CATR or BA) is a dimer, and it is inferred that the active unit is a dimer, too. However, the structure of beef Ancp bound to CATR obtained at high resolution is that of a monomer. To ascertain the dimeric organization of Ancp, we have constructed covalent tandem dimers of which one "subunit" (protomer) is the wild type and the other is inactive for ADP/ATP exchange. We have chosen either the op1 mutant or another member of the MCF, the phosphate carrier (Picp). Activities of the chimeras were first evaluated in vivo. The Ancp/op1 constructs exchange the adenine nucleotides. The Anc/Pic chimeras are considered as bifunctional forms since they exchange ADP and ATP and transport P(i) within the same cells. We have then controlled the fact that the chimeras are stable in vivo and in vitro. Proteinase K digestion showed that both protomers of Ancp/op1 have similar organization in the membrane. Analyses of kinetic properties indicated that protomers of Ancp/op1 chimeras crosstalk during the nucleotide exchange unlike those of Anc/Pic. However, full inhibition of phosphate uptake by CATR, a very specific inhibitor of Ancp, strongly suggests that the native functional unit of Ancp, and thus of Picp, is a dimer.     

ADP/ATP carrier is required for mitochondrial outer membrane permeabilization and cytochrome c release in yeast apoptosis

Adenine nucleotide translocator (ANT) is a mitochondrial inner membrane protein involved in the ADP/ATP exchange and is a component of the mitochondrial permeability transition pore (PTP). In mammalian apoptosis, the PTP can mediate mitochondrial outer membrane permeabilization (MOMP), which is suspected to be responsible for the release of apoptogenic factors, including cytochrome c. Although release of cytochrome c in yeast apoptosis has previously been reported, it is not known how it occurs. Herein we used yeast genetics to investigate whether depletion of proteins putatively involved in MOMP and cytochrome c release affects these processes in yeast. While deletion of POR1 (yeast voltage-dependent anion channel) enhances apoptosis triggered by acetic acid, H(2)O(2) and diamide, CPR3 (mitochondrial cyclophilin) deletion had no effect. Absence of ADP/ATP carrier (AAC) proteins, yeast orthologues of ANT, protects cells exposed to acetic acid and diamide but not to H(2)O(2). Expression of a mutated form of Aac2p (op1) exhibiting very low ADP/ATP translocase activity indicates that AAC's pro-death role does not require translocase activity. Absence of AAC proteins impairs MOMP and release of cytochrome c, which, together with other mitochondrial inner membrane proteins, is degraded. Our findings point to a crucial role of AAC in yeast apoptosis.     

Functional reconstitution of yeast mitochondrial ADP/ATP carrier by removing detergent with Amberlite treatment

The ADP/ATP carrier (AAC) from yeast mitochondria has been reconstituted in phospholipid vesicles essentially according to the procedure described for the reconstitution of AAC from bovine heart mitochondria (Kr?mer and Heberger (1986) Biochim. Biophys. Acta, 863, 289-296). Liposomes were prepared from the mixed micelles of dodecyl octaoxyethylene ether (C12E8)-solubilized protein and egg yolk phosphatidylcholine by removing the detergent with Amberlite treatment. The micelles were treated with Amberlite either by repeatedly passing through small columns filled with Amberlite XAD-2 beads or by stepwise addition of Amberlite beads to the micelles. All the important variables of the reconstitution components were kept at optimal level and the liposomes obtained by both the methods of Amberlite treatment were analysed for (3H)CAT binding, orientation of AAC and nucleotide exchange activity. Reconstituted AAC showed 80% right side out orientation in the liposomes prepared by either procedure. Lipsomes prepared by the Amberlite column procedure exhibited higher CAT binding but lower ADP exchange activity. Liposome preparation by the stepwise addition of Amberlite is suggested to be the method of choice for functional reconstitution of yeast AAC in view of the higher nucleotide transport activity associated with the liposomes prepared by this method.     

Characterization of loops of the yeast mitochondrial ADP/ATP carrier facing the cytosol by site-directed mutagenesis

To characterize structural features of the regions of the yeast type 2 ADP/ATP carrier (yAAC2) facing the cytosol, we prepared its Cys-less mutant, in which all four cysteine residues were replaced by alanine residues. The Cys-less mutant functioned like native yAAC2, showing that the cysteine residues are not essential. We then prepared cysteine mutants by substituting Ser(21) in the putative N-terminal region, Ala(124) and Ser(222) in the first and second loops facing cytosol, respectively, and Leu(312) in the C-terminal region of the Cys-less mutant for cysteine and examined the labeling of the substituted cysteine residues of the mutants with the membrane-impermeable SH reagent eosin-5-maleimide (EMA) from the cytosol. EMA labeled all the mutants, showing that all regions containing mutated residues faced the cytosolic side. The effects of transport inhibitors on EMA labeling were also examined. From the results, the location and conformation of the region around mutated residues were discussed.     

Functional reconstitution of the import of the yeast ADP/ATP carrier mediated by the TIM10 complex

Import of the ADP/ATP carrier (AAC) into mitochondria requires the soluble TIM10 complex to cross the intermembrane space. We report here that Tim9 and Tim10 purified from Escherichia coli can form a complex of the same size as the endogenous complex from yeast mitochondria. This shows that no other mitochondrial protein is required for the formation of the TIM10 complex. Co-expression of both proteins rendered Tim9 more soluble and allowed purification of the reconstituted complex in a single step. Urea/EDTA treatment of recombinant Tim10 allowed its import into tim10-ts mitochondria that lack endogenous Tim10 and cannot import AAC. In this way, we were able to (i) reconstitute the TIM10 complex in the intermembrane space and (ii) restore import of AAC to almost wild-type levels. The reconstituted TIM10 complex not only facilitated passage of AAC across the outer membrane but also ensured its accurate membrane insertion. We conclude that the TIM10 complex can be formed exclusively from Tim9 and Tim10 and that the reconstituted complex efficiently restores AAC import in a strain lacking the TIM10 complex.     

Identification and characterization of the cAMP binding proteins of yeast by photoaffinity labeling.

Modulation of a membrane glycoprotein, approximate molecular weight 200,000, in concert with active ionic flux has been shown in a human neuroblastoma cell line. The modulating agent was 2% dimethyl sulfoxide. Other neuronal properties, acetylcholinesterase and choline acetyltransferase, were also modulated but to a lesser extent. The appearance of this glycoprotein on the surface of both human and mouse neuroblastoma cells only under conditions of differentiation leads to the suggestion that it is directly involved with the active Na⁺ channels.     

Kinetics of electrogenic transport by the ADP/ATP carrier.

The electrogenic transport of ATP and ADP by the mitochondrial ADP/ATP carrier (AAC) was investigated by recording transient currents with two different techniques for performing concentration jump experiments: 1) the fast fluid injection method: AAC-containing proteoliposomes were adsorbed to a solid supported membrane (SSM), and the carrier was activated via ATP or ADP concentration jumps. 2) BLM (black lipid membrane) technique: proteoliposomes were adsorbed to a planar lipid bilayer, while the carrier was activated via the photolysis of caged ATP or caged ADP with a UV laser pulse. Two transport modes of the AAC were investigated, ATP(ex)-0(in) and ADP(ex)-0(in). Liposomes not loaded with nucleotides allowed half-cycles of the ADP/ATP exchange to be studied. Under these conditions the AAC transports ADP and ATP electrogenically. Mg(2+) inhibits the nucleotide transport, and the specific inhibitors carboxyatractylate (CAT) and bongkrekate (BKA) prevent the binding of the substrate. The evaluation of the transient currents yielded rate constants of 160 s(-1) for ATP and >/=400 s(-1) for ADP translocation. The function of the carrier is approximately symmetrical, i.e., the kinetic properties are similar in the inside-out and right-side-out orientations. The assumption from previous investigations, that the deprotonated nucleotides are exclusively transported by the AAC, is supported by further experimental evidence. In addition, caged ATP and caged ADP bind to the carrier with similar affinities as the free nucleotides. An inhibitory effect of anions (200-300 mM) was observed, which can be explained as a competitive effect at the binding site. The results are summarized in a transport model.     

Fatty acid-promoted mitochondrial permeability transition by membrane depolarization and binding to the ADP/ATP carrier

The mechanism by which non-esterified long-chain fatty acids (FFA) promote mitochondrial permeability transition (MPT) is not clear. We examined with energized rat liver mitochondria the role of two possible actions of FFA in MPT, (i) the reduction of the transmembrane potential (Δψ) and (ii) the increase of the negative surface charge of the inner mitochondrial membrane [Broekemeier, K.M. and Pfeiffer, D.G., Biochemistry 43, (1995) 16440–16449]. It was found that the ability of FFA to stimulate large amplitude swelling is clearly related to their uncoupling activity. Moreover, compared with classical protonophores (FCCP) FFA increase the sensitivity of the pore opening process to Δψ changes. In addition, FFA interact like their thioester derivatives in a structure-dependent manner with the ADP/ATP carrier (measured as inhibition of [³H]atractyloside binding to the AAC protein). It is suggested that not only the protonophoric action of FFA, but also a presumable stabilization of the ‘cytosolic' conformation of AAC contribute to the FFA-promoted MPT.     

Chimers of two fused ADP/ATP carrier monomers indicate a single channel for ADP/ATP transport

The mitochondrial ADP/ATP carrier (AAC) is generally believed to function as a homodimer (Wt. Wt). It remains unknown whether the two monomers possess two independent but fully anticooperative channels or they form a single central channel for nucleotide transport. Here we generated fusion proteins consisting of two tandem covalent-linked AAC monomers and studied the kinetics of ADP/ATP transport in reconstituted proteoliposomes. Functional 64-kDa fusion proteins Wt-Wt and Wt-R294A (wild-type AAC linked to a mutant having low ATP transport activity) were expressed in mitochondria of yeast transformants. Compared to homodimer Wt. Wt, the fusion protein Wt-Wt retained the transport activity and selectivity of ADP versus ATP. The strongly divergent selectivities of Wt and R294A were partially propagated in the Wt-R294A fusion protein, suggesting a limited cooperativity during solute translocation. The rates of ADP or ATP transport were significantly higher than those predicted by the two-channel model. Fusion proteins for Wt-R204L (Wt linked to an inactive mutant) and R204L-Wt were not expressed in aerobically grown yeast cells, which contained plasmid rearrangements that regenerated the fully active 32-kDa homodimer Wt. Wt, suggesting that these fusion proteins are inactive in ADP/ATP transport. These results favor a single binding center gated pore model [Klingenberg, M. (1991) in A Study of Enzymes, Vol. 2: pp. 367-388] in which two AAC subunits cooperate for a coordinated ADP/ATP exchange through a single channel.     

Cardiolipin dynamics and binding to conserved residues in the mitochondrial ADP/ATP carrier

Cardiolipin in eukaryotes is found in the mitochondrial inner membrane, where it interacts with membrane proteins and, although not essential, is necessary for the optimal activity of a number of proteins. One of them is the mitochondrial ADP/ATP carrier, which imports ADP into the mitochondrion and exports ATP. In the crystal structures, cardiolipin is bound to three equivalent sites of the ADP/ATP carrier, but its role is unresolved. Conservation of residues at these cardiolipin binding sites across other members of the mitochondrial carrier superfamily indicates cardiolipin binding is likely to be important for the function of all mitochondrial carriers. Multiscale simulations were performed in a cardiolipin-containing membrane to investigate the dynamics of cardiolipin around the yeast and bovine ADP/ATP carriers in a lipid bilayer and the properties of the cardiolipin-binding sites. In coarse-grain simulations, cardiolipin molecules bound to the carriers for longer periods of time than phosphatidylcholine and phosphatidylethanolamine lipids—with timescales in the tens of microseconds. Three long-lived cardiolipin binding sites overlapped with those in the crystal structures of the carriers. Other shorter-lived cardiolipin interaction sites were identified in both membrane leaflets. However, the timescales of the interactions were of the same order as phosphatidylcholine and phosphatidylethanolamine, suggesting that these sites are not specific for cardiolipin binding. The calculation of lipid binding times and the overlap of the cardiolipin binding sites between the structures and simulations demonstrate the potential of multiscale simulations to investigate the dynamics and behavior of lipids interacting with membrane proteins.     

Function-related asymmetry of the specific cardiolipin binding sites on the mitochondrial ADP/ATP carrier

The ADP/ATP carrier (AAC) transports matrix ATP and cytosolic ADP across the inner mitochondrial membrane (IMM). It is well known that cardiolipin (CL) plays an important role in regulating the function of AAC, yet the underlying mechanism still remains elusive. AAC is composed of three homologous domains, and three specific CL binding sites are located at the domain-domain interfaces near the matrix side. Here we report an in-depth investigation on the dynamic properties of the bound CL within the three specific sites through all-atom molecular dynamics simulations of up to 13 μs in total. Our results highlight the importance of the basic and polar residues in CL binding. The basic residues from the linker helix and/or the [Y/W/F][K/R]G motif enable the bound CL to form an intra-domain binding mode, and the canonical inter-domain binding mode only forms when these basic residues are occupied by an additional phospholipid. Of special significance, differences in the basic and polar residues lead to remarkable asymmetry among the three specific CL binding sites. We found that the bound CL at the interface of domains 2 and 3 predominantly adopts inter-domain binding mode, while CLs at the other two sites have much more intra-domain populations. This is consistent with the asymmetric crystal structure of the matrix state (m-state) AAC which implies an asymmetric transport mechanism. The dynamic equilibrium between the inter-domain and intra-domain binding modes observed in our simulations could be highly important for the bound CLs to adapt to the movements during state transitions.     

Dialectics in carrier research: The ADP/ATP carrier and the uncoupling protein

A concise review is given of the research in our laboratory on the ADP/ATP carrier (AAC) and the uncoupling protein (UCP). Although homologous proteins, their widely different functions and contrasts are stressed. The pioneer role of research on the AAC, not only for the mitochondrial but also for other carriers, and the present state of their structure-function relationship is reviewed. The function of UCP as a highly regulated H⁺ carrier is described in contrast to the largely unregulated ADP/ATP exchange in AAC. General principles of carrier catalysis as derived from studies on the AAC and UCP are elucidated.