A Walk in the Memory, from the First Functional Approach up to Its Regulatory Role of Mitochondrial Bioenergetic Flow in Health and Disease: Focus on the Adenine Nucleotide Translocator
ANT research has been highly informed by two unique classes of ANT inhibitors. The first is represented by ATR, a toxic molecule from the Atractylis gummifera thistle plant [39]. It, together with the chemically related CAT [37], locks the carrier in a cytoplasmic state (c-state) in which the substrate-binding site is accessible to the intermembrane space, which is confluent with the cytosol [40]. As such, ATR is only able to inhibit ANT function when applied to the exterior of mitochondria [22,41]. Notably, ATR inhibition can be overcome via competition with ADP, which drives ANT proteins back into the m-state [41], while CAT locks ANT-proteins in the c-state irreversibly.
The second class is represented by BKA, a respiratory toxin produced by the bacteria Burkholderia gladioli pathovar cocovenenans [42]. It is another potent inhibitor of ATP/ADP exchange, but, unlike ATR, BKA binds exclusively to the m-state conformation from inside the mitochondrial lumen [36,37,41], i.e., with the substrate-binding site accessible to the matrix side. BKA can be used to lock ANT proteins in the m-state [36,43].
The differential targeting of these two inhibitors has proved invaluable in determining the kinetic and structural features of ANT proteins.
Another important structural consideration of the ANT protein is that it tightly binds molecules of cardiolipin (CL), which is required for protein function and stability [44,45,46].
The importance of lipids in maintaining and establishing membrane barriers is accepted as dogma; however, at its most elemental, the lipid bilayer is much more than simply a random matrix providing structure for membrane proteins. In fact, within this pool of lipids, specific protein–lipid interactions critical for the structure, incorporation, and/or assembly of proteins occur [47].
As a signature phospholipid of mitochondria, CL acts as a determinant of membrane protein structure and function. It is almost exclusively found in association with the IMM, where it is synthesized, thus indicating that CL is critically important for this compartment, where it assumes a role of the utmost importance for the structural organization of the respiratory complexes in higher-order structures of functional importance [47,48]. CL also interacts with all of the major players in OXPHOS, including respiratory complexes I, III, IV, and V and the two members of the mitochondrial carrier family required for this process, ANT and the phosphate carrier [47]. In particular ANT excels in unusually strong interaction with CL [47] in a ratio of 3:1, as evidenced by 31P-NMR and ESR spectroscopic measurements [49], and by CL presence in crystal structures despite stringent washes during protein purification [50,51].
Although this aspect concerning the interaction between CL and ANT is beyond the scope of this review, because the presence of CL is critical for its functioning, deficits in ANT function should be critically evaluated in all the assorted pathologies that have been associated with disturbances in CL metabolism [52].
Consequently, it is intuitive to deduce that CL oxidation, reduced CL levels, and/or altered molecular composition can cause mitochondrial dysfunction associated with aging, ischemia and reperfusion, heart failure, and Barth syndrome [47]; deficits in CL synthesis and increased CL catabolism have been connected to diabetic cardiomyopathy [53]. Consistent with the concept that ANT may represent a downstream target subsequent to alterations to CL, CL peroxidation has been shown to inactivate mammalian ANT, resulting in apoptosis [54].
A key question that needed to be resolved is whether the mitochondrial ADP/ATP carrier requires dimerization to achieve its function. It is not surprising that these studies are often marred with controversy, in our opinion. In fact, all the experiments dealing with membrane proteins are technically challenging and many complicating factors often lead to incorrect interpretations. An aspect to take into account in these studies is that mitochondrial carriers are highly dynamic membrane proteins with very few polar interactions stabilizing their structures, causing major issues with almost every aspect of this analysis.
In one of their most recent works, Kunji and Ruprecht [55] dealt with this topic in a very accurate and detailed study.
As the authors remarked in their review article, all of the sequence, structural, biophysical, and functional data show that the mitochondrial ADP/ATP carrier exists and functions as a monomer. Studies on the structure clearly demonstrated that its structural fold is monomeric, lacking a conserved dimerization interface. Therefore, this is the only reference to which we refer the interested reader to issues concerning the ability of the protein to function as a monomer or as dimer.
Humans possess four distinct ANT isoforms encoded by four genes (gene names in brackets): ANT1 (SLC25A4), ANT2 (SLC25A5), ANT3 (SLC25A6), and ANT4 (SLC25A31) [55], whose transcription depends on the cell type, developmental stage, cell proliferation, and hormone status. They have ~90% homology [56], except for ANT4, which has ~70% homology to the others [57]. The expression of the four isoforms is tissue specific and may depend on the proliferative capacity of the tissue and its energy requirements in terms of glycolysis or OXPHOS [58].
ANT1 is predominantly expressed in post-differentiated tissues such as heart and skeletal muscle [59]. ANT2 is specifically expressed either in undifferentiated cells, such as lymphocytes, or in tissues that are able to proliferate and regenerate, such as those of the kidney and liver [60]. The expression of ANT2, which has been shown to be growth-dependent, is considered a marker of cell proliferation [61]. ANT2 gene expression is downregulated in differentiated cell lines and remains unexpressed, or only slightly expressed, in most tissues [60]. ANT3 is ubiquitously expressed [60,62] and ANT4 is specifically expressed in lung and germ tissues [63,64].
In 1999, a systematic screen for proapoptotic genes led to the identification of ANT1 as a pro-apoptotic factor [65]. Apoptosis induced by ANT1 overexpression was accompanied by enhanced cytochrome c release and caspases-9 and -3 activation, indicating that its expression facilitates mitochondria-dependent apoptosis [65,66]. In support of its pro-apoptotic role in vivo, ANT1 transfection in breast adenocarcinoma cells in a nude mouse model induced apoptosis, increased Bax expression, and finally stimulated tumor regression, validating the hypothesis that ANT1 could be a potent therapeutic target for the treatment of cancer [56]. The direct implication of ANT in apoptosis offers one more reason to focus on ANT in cancer development. Furthermore, ANT works with Bax, cyclophilin D (CyP-D) of the matrix, and VDAC of the OMM to form a permeability transition pore (PTP) (see below, Section 3.6.2), a lethal pore during apoptosis [60].
Since the expression of ANT isoforms varies with mitochondrial energy metabolism, overexpression of ANT isoforms reveals variations in apoptotic behavior. Therefore, the overexpression of ANT1 and ANT3 can increase ATP export from the matrix and lead to mitochondrial de-energization that induces apoptosis. Conversely, in tumor cell lines, the import of ATP and the so-called matrix conformation of ANT are favored; therefore, the ANT2 isoform can be a direct inhibitor of the PTP function of ANT [67] and then considered an anti-apoptotic oncoprotein [68,69] that functions as an inhibitor of mitochondrial membrane permeability.
In differentiated adult cells, ANT1 is highly expressed and ANT2 expression is minimal [69], whereas when it turns cancerous—switching from aerobic respiration to aerobic glycolysis—it downregulates ANT1 and upregulates ANT2 expression, which is linked to the rate of glycolytic metabolism and is an important indicator of carcinogenesis. Under OXPHOS impairment conditions, ANT2 imports glycolytically produced ATP into the mitochondria. In the mitochondrial matrix, the F1Fo-ATPase complex hydrolyzes the ATP, pumping out a proton into the intermembrane space. The reverse operations of ANT2 and F1Fo-ATPase under glycolytic conditions contribute to maintaining the mitochondrial membrane potential, ensuring cell survival and proliferation. Indeed, in the majority of cancer cell lines, ANT2 is very high [69], unlike that of ANT1. ANT3 expression is low across the board.
If ANT1 and ANT3 are associated with aerobic respiration, thus exporting the ATP produced by OXPHOS from the mitochondria into the cytosol while importing ADP [60], by contrast ANT2 is associated with aerobic glycolysis [60]. Its expression is a marker for rapid proliferation and/or cancer. Hence, specific inhibition of ANT2 is a prospective anti-cancer strategy.
That cancer cells use mitochondrial ATP synthase to hydrolyze glycolytic ATP in order to maintain ΔΨ is certain; however, whether in cancer cells, the entry of glycolytic ATP into mitochondria occurs by an ANT-dependent pathway or something other than ANT [70] is unknown to date. Thinking about it, it could be through the electroneutral ATP-Mg/Pi carrier, which exchanges ATP (Mg-ATP−) for Pi (HPO42−) [57], but this is not the only hypothesis. In fact, since the human genome encodes 48 different mitochondrial carriers [71] and a big proportion of them have not yet had their function assigned, it could be that one or more of these “orphan” transports facilitates ATP entry into mitochondria.
In a recent paper, Forrest [57] identified molecular targets and operations unique to cancer cells; in particular, the author, suggesting that ANT2 does import ATP into mitochondria of cancer cells, predicted that ANT2 is inversely orientated in the IMM compared to normal cells, making it resistant to inhibition by CAT and BKA, which are both inhibitors of ANT. This provides the explanation for the inability of these drugs to block ATP entry into the mitochondria of cancer cells [70]. It follows that ANT2 has a different drug interaction profile, which can be leveraged for a selective cancer therapy. A further distinction is that this opposite orientation of ANT2 may make it less predisposed to joining in to make the permeability transition pore (PTP) in apoptotic scenarios [57].
In addition to functioning as exchanger, a large body of experimental evidence demonstrates that ANTs are multifunctional proteins that are poised to participate in several aspects of mitochondrial biology.
We believe that this aspect deserves at least to be mentioned, albeit briefly. Therefore, here we describe how ANTs participate in the formation of mRC supercomplexes, as well as how they constitute a pore-forming component of the mitochondrial permeability transition pore (mPTP), a structure that forms in the IMM that is thought to underlie regulated cell death. ANT proteins are also critical for promoting mitophagy.
There is solid evidence that mitochondria coordinate the activity of diverse enzymatic processes involved in mitochondrial oxidative metabolism by grouping proteins into supercomplexes to locally increase efficiency towards the goal of producing ATP.
“ATP synthasome” is one mitochondrial supercomplex. It contains the F1Fo-ATP synthase as well as ANT and phosphate carrier proteins in a 1:1:1 stoichiometric ratio so that the two transporters that have to do with OXPHOS are proximal to the site of ATP production [37,72]. It is thought that this structure ensures a robust supply of inorganic phosphate and ADP to the F1Fo-ATP synthase as well as rapid export of produced ATP.
The respirasome, the main “respiratory chain supercomplex” (mRCS), involving electron chain complexes I, III, and IV [37,73], also associates with ANT.
Although ANT association with RCS is not required for the individual function of either electron chain complex or ANT proteins themselves, it has been noted that when RCS assembly is inhibited, overall ATP/ADP exchange rates are decreased [73], suggesting that RCS integrity is important to ANT function. Interestingly, many if not all of these interactions depend on CL (see above, Section 3.3.1). In particular, two years ago, Parodi-Rullán et al. [74] sought to elucidate whether ANT knockdown affects RCS formation in H9c2 cardiomyoblasts. Their study confirmed that ANT is involved in RCS assembly, although RCS may not be solely dependent on ANT. Moreover, ANT can physically interact with RC complexes I, III, and IV [73] and thus be involved in the respirasome structure or play a regulatory role in the formation/maintenance of mRCS assembly. However, the same authors declared that further studies are required to elucidate the role of ANT in the structural integrity and regulation of mRCS and other mitochondrial supercomplexes (e.g., ATP synthasome) in cardiac cells.
The reversible Ca2+-dependent increase of IMM permeability to solutes with molecular masses up to about 1500 Da is defined as a mitochondrial permeability transition (mPT) and is mediated by the mitochondrial permeability transition pore (mPTP) [75]. The proposal that the permeability transition is linked to the reversible opening of a pore rather than to unspecific membrane damage is based on the inhibitory effect of cyclosporin A (CsA) used in nanomolar concentrations [76]. mPTP opening causes both (i) mitochondrial swelling and the rupture of the outer membrane, a process leading to the release of soluble pro-apoptotic proteins from the intermembrane space and (ii) the dissipation of the mitochondrial H+ gradient, thus collapsing the Δψ and in turn abolishing the mitochondrial capacity to produce ATP [37], a real energetic catastrophe that invariably leads to cell death.
It is commonly believed that mPTP is a multi-protein system. Originally, only regulatory components were identified. The first unambiguously established component was CypD (in the matrix), which remains the only protein whose involvement in mPTP pore formation and activity regulation is undisputed [77]. The identity of other proteins comprising the PTP is to date debated; however, mPTP is usually described as a polyprotein complex located at the mitochondrial contact site formed from the voltage-gated anion channel (VDAC, in the OMM), members of the Bax/Bcl2 pro and antiapoptotic protein family [78], and additional proteins such as HK (in the cytosol), creatine kinase, peripheral benzodiazepine receptor (PBR, in the OMM) [79], and the phosphate carrier [80]. To these must also be added ANT, widely accepted as a candidate for PTP formation and first proposed by Halestrap and Davidson in 1990 [81].
Therefore, ANT appears to act as a bi-functional protein, contributing, on the one hand, to a crucial step in aerobic energy metabolism, the ADP/ATP translocation, and on the other hand, converting into a pro-apoptotic pore under the control of onco- and anti-oncoproteins from the Bax/Bcl-2 family. In particular, in response to multiple pro-apoptotic stimuli, Bax moves from the cytoplasm to mitochondrial membranes. Upon insertion into OMM, Bax can form a pore, presumably via oligomerization with itself or with its homologs (Bak, Bad, t-Bid) and/or interactions with mPTP components, such as VDAC and/or ANT [82]. As reported by Belzacq et al. [82], in a hypothetical cooperation model Bcl-2 would stimulate the translocator function of ANT, whereas Bax, after a conformational change induced by ATR, would stimulate the pore function of ANT. On the contrary, Bax would diminish the translocator function and ATP, ADP, and Bcl-2 would inhibit the pore function of ANT. Shortly after, the same authors, utilizing a variety of different systems (proteoliposomes containing purified ANT, Bcl-2, and/or Bax; isolated mitochondria and mitoplasts; intact cells), reported that Bcl-2 and Bax, in addition to modulating pore formation by ANT, importantly also influence the enzymatic function of ANT as an ADP/ATP antiporter, thus shedding new light on the apoptosis-regulatory functions of Bcl-2/Bax-like proteins [67]. These observations are corroborated by the fact that several ANT ligands and inhibitors modulate mPTP opening activity [37].
However, although a number of studies have been carried out with the aim of demonstrating unequivocally the involvement of the ADP/ATP carrier in mPTP formation, the conclusions are controversial.
In this regard, we refer to the latest review on this topic by Professor Bernardi [78,83,84], an example of excellence in the field, who has taken care over the years to follow the discoveries regarding the mPTP since the idea appeared in 1990 [81].