Salvatore Nesci, Alessandra Pagliarani
Department of Veterinary Medical Sciences, University of Bologna, Ozzano Emilia Via Tolara di Sopra 50, 40064 Bologna, Italy.
Corresponding author:salvatore.nesci@unibo.it
Running title: H+-translocation driven by ATP hydrolysis by the Ca2+-activated F1FO-ATPase
Abstract
The mitochondrial F1FO-ATPase in the presence of the natural cofactor Mg2+ acts as the enzyme of life by synthesizing ATP, but it can also hydrolyze ATP to pump H+. Interestingly, Mg2+ can be replaced by Ca2+, but only to sustain ATP hydrolysis and not ATP synthesis. When Ca2+ inserts in F1, the torque generation built by the chemomechanical coupling between F1 and the rotating central stalk was reported as unable to drive the transmembrane H+ flux within FO. However, the failed H+translocation is not consistent with the oligomycin-sensitivity of the Ca2+-dependent F1FO-ATP(hydrol)ase. New enzyme roles in mitochondrial energy transduction are suggested by recent advances. Accordingly, the structural F1FO-ATPase distortion driven by ATP hydrolysis sustained by Ca2+is consistent with the permeability transition pore signal propagation pathway. The Ca2+-activated F1FO-ATPase, by forming the pore, may contribute to dissipate the transmembrane H+ gradient created by the same enzyme complex.
Keywords: Ca2+ cofactor; F1FO-ATPase; mitochondria; H+ pump; oligomycin; permeability transition pore; bioenergetics.
Introduction
The mitochondrial F1FO-ATPase is a multisubunit complex arranged in dimers or oligomers and placed at the edge of the cristae of the inner mitochondrial membrane (IMM)1. The monomer is formed by two domains, named F1 and FO functionally and structurally linked to a stator (lateral stalk) and a rotor (central stalk). The F1 portion, namely the hydrophilic domain that protrudes in the mitochondrial matrix, has a conspicuous lollipop shape formed byα 3, β 3, γ ,δ , and ε subunits. An alternated arrangement of αand β subunits forms a globular hexamer around the γsubunit (Fig. 1A). The structure functions as a reversible rotary molecular motor which can build or hydrolyze ATP depending on the rotation direction, which in turn is driven by the transmembrane proton-motive force (Δp ).In vitro the γ subunit of F1-ATPase was shown to rotate within the surroundingα 3β 3 subunits, synthesizing or hydrolysing ATP in three separate catalytic sites on theα /β subunit interface. The catalytic sites are alternated with the non-catalytic sites, which can only bind adenine nucleotides2. During the kinetic reactions, the three non-equivalent conformation β E (empty),β DP (which hosts ADP) andβ TP (contains ATP or ADP) of the catalytic sites, with increasing affinity for ATP, change their conformation and binding properties every 120° with the rotation of the rotor3. In addition, the F1FO-ATPase catalytic and non-catalytic sites in their different conformations can also bind metal divalent cations 1. In mammals, the membrane-embedded domain is composed by the a subunit, the transmembrane α-helices ofb subunit, the c n subunits (n= eight in mammals) which arranged as a cylindric palisade form the c -ring, A6L subunit, and the supernumerary subunits e , f ,g , DAPIT (Diabetes-Associated Protein in Insulin-sensitive Tissue), 6.8 KDa proteolipid (PL) (Fig. 1A) 4. The H+ translocation sector arises from a/c -ring interactions by forming two asymmetric half-channels with unexpected horizontal membrane-intrinsic α-helices in the a subunit. These two half-channels are mutually offset, while the H+binding sites are located on the C-terminal α-helix of each csubunit 5. In the mammalian F1FO-ATPase the a and A6L membrane subunits are encoded by the mitochondrial DNA. The central stalk within the F1 domain contains the γsubunit, which joined to the δ and ε subunits, forms a sort of foot which interacts with the loop region of c -ring. The lateral or peripheral stalk joins the two FO and F1 enzyme domains (Fig. 1A). The b subunit spans the complete length of the lateral stalk and interacts with OSCP, F6 andd subunits which belong to the soluble enzyme section. All these subunits connect the soluble stator subunits withα TP subunit of F1 domain. In addition, the top of α TP,α DP, α E and theβ DP and β E are only linked with OSCP. Some subunits of the lateral stalk, namely the membrane embedded portion of b , f and A6L subunits6 and the supernumerary subunits, are transmembrane subunits 4. The lateral stalk shows a spectacular flexibility that plays the role of resisting the torque generation of the rotor by coupling F1 catalysis to H+ translocation 7,8.
The H+-translocating F1FO-ATPase sustains either ATP synthesis or hydrolysis 9. In the “forward” mode the Mitchell’s proton motive force Δp created by mitochondrial respiration drives ATP formation from ADP and Pi. In the so-called “reverse ” mode, the phosphorylation potential generated by ATP breakdown is exploited by the enzyme complex to pump H+ and energize the IMM when the Δp drops10. Both ATP synthase and hydrolase activities are opposite F1FO-ATPase functions that depend on the bioenergetic state of mitochondria. The bi-functional ATP synthesis/hydrolysis mode coupled to H+ translocation of F1FO-ATPase is a mechanism unique in biology sustained by the natural cofactor Mg2+. The F1FO-ATPase can replace Mg2+ by Ca2+ losing the ATP synthesis function, but preserving the F1FO-ATP(hydrol)ase activity11. High Ca2+ concentrations in mitochondria activate of F1FO-ATPase by direct Ca2+ binding to theα 3β 3 globular hexamer that dissociates ATP hydrolysis from H+ pumping. In this case F1 was reported to become uncoupled from FO domain 12. However, this assumption becomes questionable on considering the recent cryo-EM enzyme structure and structure activity relationship data on the effect of small molecules 13 and specific F1 and FO domain inhibitors 11,14. So, in search for a different interpretation of the findings up to now obtained, experimental and literature data were combined to draw a pattern of the mechanism involved.
  1. Materials and Methods
  2. Preparation of the mitochondrial fractions and F1FO‐ATPase activity assays
Swine hearts (Sus scrofa domesticus ) were collected at a local abattoir and transported to the lab. From heart tissue homogenized and then subjected to differential centrifugation 11 the mitochondrial preparations, obtained in a divalent cation-free medium, were characterized as described in 14. To evaluate the mitochondrial F1FO-ATPase activities, the mitochondrial suspensions were added to the reaction system that contained 3 mM ATP and 2 mM Ca2+ or Mg2+ in 75 mM ethanolammine–HCl buffer, pH 8.8 and spectrophotometrically detected as reported in 15.
Protein model
The structural details of the protein arrangement in the F1FO-ATPase subunits were obtained by the Chem3D program of ChemOffice Professional 19.1.1 software16 using the deposited structures in PDB.
Statistical analysis
In each set of experiments, the data represent the mean ± SD of the number of analyses carried out on at least three distinct mitochondrial preparations. The differences between the enzyme activity data in differently treated mitochondria were evaluated by one way ANOVA followed by Student-Newman-Keuls’ test when F values indicated significance (P ≤0.05). Percentage data werearcsin -transformed before statistical analyses to ensure normality.
Results and Discussion
The catalytic and non-catalytic subunits of the F1FO-ATPase show specific amino acid residues and secondary structure motifs required for the molecular interaction with adenine nucleotides and divalent cations. An eight amino acids sequence, GXXXXGKT , conserved in all ATPases17 is the basic structural feature of the P-loop, known as a phosphate binding loop, on α and β subunits (Fig. 1B). The motif interacts with Mg2+ and phosphate (Pi) groups of ATP by coordinating β-Pi and γ-Pi to exchange the terminal γ-Pi when the ATP is synthesized or hydrolysed. The positions and specific amino acid composition in the pig sequence are158GGAGVGKT 165 and169GDRGTGKT 176 in the βand α subunits, respectively. T 165 is the only residue that plays a key role in coordinating Mg2+ in the β subunits of enzyme during ATP hydrolysis, while T 176 of α subunits could bind the cofactor. Similarly to Mg2+, Ca2+ can also bind to all the catalytic sites and probably also to the non-catalytic sites 4. The relative affinities for divalent cations and ATP in the reverse reaction of ATP hydrolysis are modulated by mutagenesis of these specific residues of β subunits 18. However, Ca2+, which has higher steric hindrance than Mg2+, can change the coordination geometry of the Mg2+-binding site from the octahedral bipyramide which binds six ligands up to allow eight ligands when Ca2+is inserted in replacement of Mg2+19. Therefore, the rigid octahedral complex changes to a less rigid geometry with irregular bond distances and angles and variable coordination number. This flexible arrangement may explain the non-competitive Ca2+ inhibition of the Mg2+-activated F1FO-ATPase 14. The Ca2+-dependent F1FO-ATP(hydrol)ase is capable of sustaining torque generation of the rotor. The rotational motion was found to be similar to that induced by Mg2+ in the F1-ATPase 20. Ca2+binding could have the functional consequence to prevent the building of the transmembrane H+ gradient, as shown by ACMA fluorescence quenching 21. However, these results cannot exclude that the rotation driven by ATP hydrolysis stimulated by Ca2+ 20 is coupled to H+ translocation. The Mg2+-activated F1FO-ATPase can display H+ flow across FO in the absence of adenine nucleotides bound to F1. This uncoupled proton leakage, known as “proton slip”, is associated with a free-wheeling of the central stalk under non-physiological conditions10. In addition, the proton slip is abolished by FO inhibitors (e .g . oligomycin), but it is insensitive to F1 inhibitors. Conversely, the Ca2+-dependent F1FO-ATP(hydrol)ase activity was shown to inhibited by various F1 inhibitors14. Since the F1FO-ATPase in the presence of Ca2+shows a four orders of magnitude lower enzyme activity than the Mg2+-activated F1FO-ATPase (Fig. 2A), ATP hydrolysis sustained by Ca2+ may be unable to support a significant H+ pumping to energize the membrane. Moreover, the Ca2+-activated F1FO-ATPase is now generally recognized to play an important role in the permeability transition pore (PTP) formation and opening 22–26, which can dissipate the Δp 11. The loss of F1FO-ATPase structural-functional integrity emerges as the most likely event involved in the decreased oligomycin sensitivity when the F1 catalysis is not coupled to H+ transport by FO27. However, the Ca2+-dependent F1FO-ATP(hydrol)ase is inhibited by oligomycin 28. A similar behaviour was described in pea stem mitochondria, where, since the Ca-ATPase activity was fully sensitive to oligomycin, ATP hydrolysis could be coupled to H+ translocation 29. In swine heart mitochondria oligomycin displays a higher inhibition efficiency on the Ca2+-activated F1FO-ATPase than on the Mg2+-activated F1FO-ATPase, as shown by the lower IC50 value (Fig. 2B). The coupling index (the ratio between the total F1FO-ATPase activity and the oligomycin-sensitive F1FO-ATPase activity) is 94.7±1.8% and 91.6±3.7% in presence of Mg2+ and Ca2+, respectively. Therefore oligomycin blocks H+ translocation coupled to ATP hydrolysis irrespective of the divalent cation (Fig. 2C). Consistently, these data suggest that mechanochemical coupling of Ca2+-dependent F1-ATP(hydrol)ase works as a rotary chemical motor to drive H+ translocation in the FO domain 11,15. The fact that the H+-pumping activity driven by Ca2+ may not energize IMM is not surprising, being supported by the new “bent-pull” model of the c -ring gated channel 30 and by the cryo-EM maps of the enzyme exposed to Ca2+ 4. The lack of apparent H+ translocation with Ca2+-dependent F1FO-ATP(hydrol)ase may be rather due to H+ backflow through the open PTP 31. Indeed, different Ca2+F1FO-ATPase states during ATP hydrolysis were not identified in the Mg2+-activated F1FO-ATPase. Moreover recent data show that the PTP opens when the Ca2+-enzyme in disassembled conformation has the peripheral stalk twisted and the F1 detached from FO 4(Fig. 3). In all likelihood, oligomycin inhibits ATP hydrolysis sustained by Ca2+ in the first conformational stages of the Ca2+-dependent F1FO-ATP(hydrol)ase when F1 is still coupled to FO. Indeed, oligomycin, venturicidin, and DCCD, which block H+translocation by binding to the c -ring, can reduce the calcein quenching rate 32, while small-molecules obtained from the oligomycin structure target the c subunits and inhibit the PTP 33.
To sum up, the H+-translocating Ca2+-dependent F1FO-ATP(hydrol)ase is a (mono)functional mode of the mitochondrial F-type ATPase complex. The F1 domain which hydrolyzes ATP in the presence of Ca2+ drives the mechanical-power transmission which results in FO conductance to H+. Consistently, the poor H+-pumping activity of the Ca2+-dependent F1FO-ATP(hydrol)ase fails to energize the IMM, mainly because the same enzyme activity is a key PTP constituent, and the PTP opening prevents and masks Δp formation31,34.
Conflicts of interest
None.
Acknowledgments
This work was supported by the CARISBO Foundation Grants n° 2019.0534, Bologna, Italy to SN.
References
1 Hahn A, Parey K, Bublitz M, Mills DJ, Zickermann V, Vonck J et al. Structure of a Complete ATP Synthase Dimer Reveals the Molecular Basis of Inner Mitochondrial Membrane Morphology. Mol Cell 2016;63 : 445–456.
2 Boyer PD. The ATP synthase–a splendid molecular machine. Annu Rev Biochem 1997; 66 : 717–749.
3 Yoshida M, Muneyuki E, Hisabori T. ATP synthase–a marvellous rotary engine of the cell. Nat Rev Mol Cell Biol 2001; 2 : 669–677.
4 Pinke G, Zhou L, Sazanov LA. Cryo-EM structure of the entire mammalian F-type ATP synthase. Nat Struct Mol Biol 2020. doi:10.1038/s41594-020-0503-8.
5 Nesci S, Pagliarani A, Algieri C, Trombetti F. Mitochondrial F-type ATP synthase: multiple enzyme functions revealed by the membrane-embedded FO structure. Crit Rev Biochem Mol Biol 2020; : 1–13.
6 Gu J, Zhang L, Zong S, Guo R, Liu T, Yi J et al. Cryo-EM structure of the mammalian ATP synthase tetramer bound with inhibitory protein IF1. Science 2019; 364 : 1068–1075.
7 Hahn A, Vonck J, Mills DJ, Meier T, Kühlbrandt W. Structure, mechanism, and regulation of the chloroplast ATP synthase.Science 2018; 360 : eaat4318.
8 Junge W, Sielaff H, Engelbrecht S. Torque generation and elastic power transmission in the rotary F(O)F(1)-ATPase. Nature 2009;459 : 364–370.
9 Junge W, Lill H, Engelbrecht S. ATP synthase: An electrochemical transducer with rotatory mechanics. Trends in Biochemical Sciences 1997; 22 : 420–423.
10 Feniouk BA, Mulkidjanian AY, Junge W. Proton slip in the ATP synthase of Rhodobacter capsulatus: induction, proton conduction, and nucleotide dependence. Biochim Biophys Acta 2005; 1706 : 184–194.
11 Nesci S, Trombetti F, Ventrella V, Pirini M, Pagliarani A. Kinetic properties of the mitochondrial F1FO-ATPase activity elicited by Ca(2+) in replacement of Mg(2+). Biochimie 2017; 140 : 73–81.
12 Hubbard MJ, McHugh NJ. Mitochondrial ATP synthase F1-beta-subunit is a calcium-binding protein. FEBS Lett 1996; 391 : 323–329.
13 Vincenzo Algieri, Cristina Algieri, Loredana Maiuolo, Antonio De Nino, Alessandra Pagliarani, Matteo A. Tallarida et al.1,5-Disubstituted-1,2,3-triazoles as inhibitors of the mitochondrial Ca2+-activated F1FO-ATP(hydrol)ase and of the permeability transition pore. Ann N Y Acad Sci 2020. doi:10.1111/nyas.14474.
14 Algieri C, Trombetti F, Pagliarani A, Ventrella V, Bernardini C, Fabbri M et al. Mitochondrial Ca2+ -activated F1 FO -ATPase hydrolyzes ATP and promotes the permeability transition pore. Ann N Y Acad Sci 2019; 1457 : 142–157.
15 Algieri C, Trombetti F, Pagliarani A, Ventrella V, Nesci S. Phenylglyoxal inhibition of the mitochondrial F1FO-ATPase activated by Mg2+ or by Ca2+ provides clues on the mitochondrial permeability transition pore. Arch Biochem Biophys 2020; 681 : 108258.
16 Trombetti F, Pagliarani A, Ventrella V, Algieri C, Nesci S. Crucial aminoacids in the FO sector of the F1FO-ATP synthase address H+ across the inner mitochondrial membrane: molecular implications in mitochondrial dysfunctions. Amino Acids 2019; 51 : 579–587.
17 Ramakrishnan C, Dani VS, Ramasarma T. A conformational analysis of Walker motif A [GXXXXGKT (S)] in nucleotide-binding and other proteins. Protein Eng 2002; 15 : 783–798.
18 Du Z, Tucker WC, Richter ML, Gromet-Elhanan Z. Assembled F1-(alpha beta ) and Hybrid F1-alpha 3beta 3gamma -ATPases from Rhodospirillum rubrum alpha, wild type or mutant beta, and chloroplast gamma subunits. Demonstration of Mg2+versus Ca2+-induced differences in catalytic site structure and function. J Biol Chem 2001; 276 : 11517–11523.
19 Casadio R, Melandri BA. CaATP inhibition of the MgATP-dependent proton pump (H+-ATPase) in bacterial photosynthetic membranes with a mechanism of alternative substrate inhibition. JBIC 1996;1 : 284–291.
20 Tucker WC, Schwarz A, Levine T, Du Z, Gromet-Elhanan Z, Richter MLet al. Observation of calcium-dependent unidirectional rotational motion in recombinant photosynthetic F1-ATPase molecules. J Biol Chem 2004; 279 : 47415–47418.
21 Papageorgiou S, Melandri AB, Solaini G. Relevance of divalent cations to ATP-driven proton pumping in beef heart mitochondrial F0F1-ATPase.J Bioenerg Biomembr 1998; 30 : 533–541.
22 Alavian KN, Beutner G, Lazrove E, Sacchetti S, Park H-A, Licznerski Pet al. An uncoupling channel within the c-subunit ring of the F1FO ATP synthase is the mitochondrial permeability transition pore.Proc Natl Acad Sci USA 2014; 111 : 10580–10585.
23 Bonora M, Bononi A, De Marchi E, Giorgi C, Lebiedzinska M, Marchi Set al. Role of the c subunit of the FO ATP synthase in mitochondrial permeability transition. Cell Cycle 2013;12 : 674–683.
24 Giorgio V, von Stockum S, Antoniel M, Fabbro A, Fogolari F, Forte Met al. Dimers of mitochondrial ATP synthase form the permeability transition pore. Proc Natl Acad Sci USA 2013; 110 : 5887–5892.
25 Mnatsakanyan N, Llaguno MC, Yang Y, Yan Y, Weber J, Sigworth FJet al. A mitochondrial megachannel resides in monomeric F1FO ATP synthase. Nat Commun 2019; 10 : 5823.
26 Urbani A, Giorgio V, Carrer A, Franchin C, Arrigoni G, Jiko Cet al. Purified F-ATP synthase forms a Ca2+-dependent high-conductance channel matching the mitochondrial permeability transition pore. Nat Commun 2019; 10 : 4341.
27 Devenish RJ, Prescott M, Boyle GM, Nagley P. The Oligomycin Axis of Mitochondrial ATP Synthase: OSCP and the Proton Channel. J Bioenerg Biomembr 2000; 32 : 507–515.
28 Nesci S, Ventrella V, Trombetti F, Pirini M, Pagliarani A. Preferential nitrite inhibition of the mitochondrial F1FO-ATPase activities when activated by Ca(2+) in replacement of the natural cofactor Mg(2+). Biochim Biophys Acta 2016; 1860 : 345–353.
29 De Col V, Petrussa E, Casolo V, Braidot E, Lippe G, Filippi Aet al. Properties of the Permeability Transition of Pea Stem Mitochondria. Front Physiol 2018; 9 : 1626.
30 Mnatsakanyan N, Jonas EA. ATP synthase c-subunit ring as the channel of mitochondrial permeability transition: Regulator of metabolism in development and degeneration. J Mol Cell Cardiol 2020;144 : 109–118.
31 Bernardi P, Rasola A, Forte M, Lippe G. The Mitochondrial Permeability Transition Pore: Channel Formation by F-ATP Synthase, Integration in Signal Transduction, and Role in Pathophysiology.Physiol Rev 2015; 95 : 1111–1155.
32 Bonora M, Morganti C, Morciano G, Pedriali G, Lebiedzinska-Arciszewska M, Aquila G et al. Mitochondrial permeability transition involves dissociation of F1FO ATP synthase dimers and C-ring conformation. EMBO Rep 2017; 18 : 1077–1089.
33 Morciano G, Preti D, Pedriali G, Aquila G, Missiroli S, Fantinati Aet al. Discovery of Novel 1,3,8-Triazaspiro[4.5]decane Derivatives That Target the c Subunit of F1/FO-Adenosine Triphosphate (ATP) Synthase for the Treatment of Reperfusion Damage in Myocardial Infarction. J Med Chem 2018; 61 : 7131–7143.
34 Nesci S. The mitochondrial permeability transition pore in cell death: A promising drug binding bioarchitecture. Medicinal Research Reviews 2020; 40 : 811–817.
Figure 1. Representative structure of F1FO-ATPase monomers in mammalian mitochondria (A). The enzyme subunits are drawn as ribbon representations obtained from modified PDB ID codes: 6TT7. Δp , Mitchell’s proton motive force, IMM, inner mitochondrial membrane. The letter colors are the same as those of the subunit to which belong. B) Catalytic binding site of F1FO-ATPase. The ATP substrate and Mg2+ cofactor (in ball and stick representation) are located in the β and α subunits, drawn as ribbon model (modified PDB ID code: 6J5J) inβ TP and α TP conformation, respectively, which show the position of key amino acid residues that bind Mg2+. The P-loop is in light blue in both subunits. The binding sites are viewed from the γ subunit (upper panel) and between the observer and the γ subunit (lower panel).
Figure 2. Effect of divalent cations on ATP hydrolysis by the mitochondrial F1FO-ATPase. A) F-ATPase activities in the presence of Ca2+ or Mg2+ are shown as bar chart. B) Dose-response curve of oligomycin on the F1FO-ATPase activated by Ca2+ or Mg2+ expressed as percentage of the enzyme activity in the absence of oligomycin. C) The oligomycin-sensitive ATPase activity (█) and the oligomycin-insensitive ATPase activity in presence of 3 μg/ml of oligomycin (█) are expressed as percentages of the total mitochondrial ATPase activity sustained by Ca2+ or Mg2+, respectively. Data expressed as column chart represent the mean ± SD (vertical bars) from three experiments carried out on different mitochondrial preparations. * indicates significantly different values (P ≤0.05).
Figure 3. F1FO-ATPase activity raised by Mg2+ or Ca2+ as cofactors. ATP hydrolysis sustained by Mg2+ (i ) or Ca2+ (ii ) is coupled to H+translocation. The different size of the two cofactors changes the F1FO-ATPase conformation. Indeed, the transition of the Ca2+-dependent F1FO-ATP(hydrol)ase from the assembled (ii ) to the disassembled state (iii ) could induce the loss of H+-translocation. Consequently, the PTP opens when a retracted e subunit pulls the lyso-phophatidylserine plug out of the c -ring at the inner mitochondrial membrane side, while the F1FO destabilization pulls out phospatidylserine at the matrix side.