Mitochondrial F(0) F(1) -ATP synthase is a molecular target of 3-iodothyronamine, an endogenous metabolite of thyroid hormone.

BACKGROUND AND PURPOSE: 3-iodothyronamine (T1AM) is a metabolite of thyroid hormone acting as a signalling molecule via non-genomic effectors and can reach intracellular targets. Because of the importance of mitochondrial F(0) F(1) -ATP synthase as a drug target, here we evaluated interactions of T1AM with this enzyme. EXPERIMENTAL APPROACH: Kinetic analyses were performed on F(0) F(1) -ATP synthase in sub-mitochondrial particles and soluble F(1) -ATPase. Activity assays and immunodetection of the inhibitor protein IF(1) were used and combined with molecular docking analyses. Effects of T1AM on H9c2 cardiomyocytes were measured by in situ respirometric analysis. KEY RESULTS: T1AM was a non-competitive inhibitor of F(0) F(1) -ATP synthase whose binding was mutually exclusive with that of the inhibitors IF(1) and aurovertin B. Both kinetic and docking analyses were consistent with two different binding sites for T1AM. At low nanomolar concentrations, T1AM bound to a high-affinity region most likely located within the IF(1) binding site, causing IF(1) release. At higher concentrations, T1AM bound to a low affinity-region probably located within the aurovertin binding cavity and inhibited enzyme activity. Low nanomolar concentrations of T1AM increased ADP-stimulated mitochondrial respiration in cardiomyocytes, indicating activation of F(0) F(1) -ATP synthase consistent with displacement of endogenous IF(1,) , reinforcing the in vitro results. CONCLUSIONS AND IMPLICATIONS: Effects of T1AM on F(0) F(1) -ATP synthase were twofold: IF(1) displacement and enzyme inhibition. By targeting F(0) F(1) -ATP synthase within mitochondria, T1AM might affect cell bioenergetics with a positive effect on mitochondrial energy production at low, endogenous, concentrations. T1AM putative binding locations overlapping with IF(1) and aurovertin binding sites are described.

The nucleotide-independent Fe(III)-binding site is located on beta subunit of the mitochondrial F(1)-ATPase.

Upon separation of a crude preparation of beta subunit ("beta fraction") from mitochondrial F(1)-ATPase containing one equivalent of Fe(III) in the nucleotide-independent site (1Fe(III)-loaded MF(1)), Fe(III) is almost completely recovered. CD spectra show that "beta fraction" maintains the structural changes induced by Fe(III) in the whole enzyme. In accordance, EPR reveals that the Fe(III) site geometry is conserved in "beta fraction." Moreover, the EPR spectra of 1Fe(III)-loaded MF(1) and its "beta fraction" undergo similar changes of the line-shape upon Pi binding at the catalytic site, indicating that the Pi and Fe(III) are proximal on beta. Highly purified beta in nucleotide-free form binds 1mol of Fe(III)/mol of protein. MF(1) "freezed" by inhibitors with two beta in closed conformation and one beta in open or half-closed conformation binds 1mol of Fe(III)/mol of enzyme. Therefore, the Fe(III) site location in the unique beta subunit not adopting the closed conformation is proposed.

Effects of Fe(III) binding to the nucleotide-independent site of F1-ATPase: enzyme thermostability and response to activating anions.

Mitochondrial F1-ATPase was induced in different conformations by binding of specific ligands, such as nucleotides. Then, Fourier transform infrared spectroscopy (FT-IR) and kinetic analyses were run to evaluate the structural and functional effects of Fe(III) binding to the nucleotide-independent site. Binding of one equivalent of Fe(III) induced a localised stabilising effect on the F1-ATPase structure destabilised by a high concentration of NaCl, through rearrangements of the ionic network essential for the maintenance of enzyme tertiary and/or quaternary structure. Concomitantly, a lower response of ATPase activity to activating anions was observed. Both FT-IR and kinetic data were in accordance with the hypothesis of the Fe(III) site location near one of the catalytic sites, i.e. at the alpha/beta subunit interface.

Fe(III) binding to Bacillus PS3 F(1)ATPase, alphabeta subcomplexes and isolated alpha- and beta-subunits.

Isolated alpha- and beta-subunits of Thermophilic Bacillus PS3 F(1)ATPase (TF(1)) bind about 1 Fe(III) equivalent. Upon reassembling in the symmetric alpha(3)beta(3) hexamer, Fe(III) binding capacity decreases, as this complex binds about three Fe(III) equivalents. In accordance, when the hexamer is dissociated in the alpha(1)beta(1) heterodimer, each heterodimer binds about one Fe(III) equivalent. On the contrary, native TF(1) exhibits a single Fe(III) site. CD spectra in far UV indicate that upon Fe(III) binding both the whole complex and the isolated beta-subunit undergo structural modifications accompanied by decrease of alpha-helix content, while alpha-subunit doesn't. As in alpha(3)beta(3) and in the whole enzyme the number of bound Fe(III) equivalents is consistent with the number of beta-subunits in the "empty" conformation, it is inferred that the single Fe(III) site in TF(1) is probably located in beta(E).

EPR detection of protein-derived radicals in the reaction of H(2)O(2) with Fe bound in mitochondrial F(1)ATPase.

A severe inactivation is obtained upon the addition of H(2)O(2) to bovine heart F(1)ATPase samples containing Fe(III) in the nucleotide-independent site, and Fe(II) in the ATP-dependent site. EPR spectra at 4.9 K of these samples indicate that H(2)O(2) produces the complete oxidation of Fe(II) to Fe(III) and the concomitant appearance of two protein-derived radical species. The two signals (g = 2.036 and g = 2.007) display a different temperature dependence and saturation behavior. The relaxation properties of the radical at g = 2.036 suggest magnetic interaction with one of the two iron centers. Such events are not observed when H(2)O(2) is added either to native F(1)ATPase containing a high amount of Fe(II) and low amount of Fe(III) or to F(1)ATPase deprived of endogenous Fe and subsequently loaded with only Fe(III) in both sites. It is hypothesized that in F(1)ATPase samples containing both Fe(III) and Fe(II), intramolecular long-range electron transfer may occur from Fe(II) to a high oxidation state species of Fe formed in the nucleotide-independent site upon oxidation of Fe(III) by H(2)O(2).

Melatonin and oxidative damage in mice liver induced by the prooxidant antitumor drug, adriamycin.

Antioxidant properties have been attributed to melatonin; it seemed therefore worthwhile to determine its effects in relation to the prooxidant action of adriamycin, which contributes to its toxic and therapeutic effects. Melatonin effectively acts as a direct free radical scavenger in the concentration range of 20-100 microM as determined in vitro, using Fenton reaction as a source of free radicals that were determined by EPR using spin trapping method. Following the administration of a single i.v. dose of 28 mg/Kg or of 3 repeated i. p. doses of 5 mg/Kg adriamycin to CBA mice, glutathione levels in the liver cells were significantly reduced. When the treatment with adriamycin was preceded by the s.c. administration of 2 mg/Kg melatonin, the decrease in total and reduced glutathione concentrations was significantly prevented. A significant increase in lipid peroxidation was observed in liver cells after a single administration of adriamycin which was not attenuated by pretreatment with melatonin. These results indicate that further examination of the possible protective action of melatonin on the toxic effects of prooxidant antitumor drugs on normal and neoplastic tissues would be of interest also in relation to their chronotoxicological properties.

Decrease of cytochrome c oxidase protein in heart mitochondria of copper-deficient rats.

Copper deficiency has been reported to be associated with decreased cytochrome c oxidase activity, which in turn may be responsible for the observed mitochondrial impairment and cardiac failure. We isolated mitochondria from hearts of copper-deficient rats: cytochrome c oxidase activity was found to be lower than in copper-adequate mitochondria. The residual activity paralleled copper content of mitochondria and also corresponded with the heme amount associated with cytochrome aa3. In fact, lower absorption in the alpha-band region of cytochrome aa3 was found for copper-deficient rat heart mitochondria. Gel electrophoresis of protein extracted from mitochondrial membranes allowed measurements of protein content of the complexes of oxidative phosphorylation, revealing a lower content of complex IV protein in copper-deficient rat heart mitochondria. The alterations caused by copper deficiency appear to be specific for cytochrome c oxidase. Changes were not observed for F0F1ATP synthase activity, for heme contents of cytochrome c and b, and for protein contents of complexes I, III and V. The present study demonstrates that the alteration of cytochrome c oxidase activity observed in copper deficiency is due to a diminished content of assembled protein and that shortness of copper impairs heme insertion into cytochrome c oxidase.

Melatonin decreases bone marrow and lymphatic toxicity of adriamycin in mice bearing TLX5 lymphoma.

When CBA male mice bearing TLX5 lymphoma were treated in the evening with a single i.v. dose of adriamycin (20-40 mg/Kg), the administration of a single pharmacological dose of melatonin (10 mg/kg s.c.) 1 hr earlier reduced the acute mortality from 10/24 to 2/24. The increase in survival time caused by adriamycin over drug untreated controls was not reduced by melatonin. The administration of melatonin alone did not cause any antitumor or evident toxic effect. Melatonin also attenuated the reduction caused by adriamycin in the number of bone marrow GM-CFU, and of CD3+, CD4+ and CD8+ splenic T-lymphocyte subsets. Reduced and total glutathione levels were decreased in the bone marrow and in the liver cells of the animals treated with adriamycin, and were significantly restored by melatonin. Moreover, lipid peroxidation by adriamycin was reduced by melatonin, as indicated by malondialdehyde measurement in the liver of the treated animals. These data indicate that the protective effects of melatonin against the host toxicity of the prooxidant antitumor drug, adriamycin, might be attributed at least partially to its antioxidant properties. These findings appear of interest in relation to the physiological rhythmic levels of endogenous melatonin and to the chronotoxicology of anthracyclines.

Severe energy impairment consequent to inactivation of mitochondrial ATP synthase as an early event in cell death: a mechanism for the selective sensitivity to H2O2 of differentiating erythroleukemia cells.

Irreversible damage to Friend's erythroleukemia cells was caused by induction of endogenous heme biosynthesis with the differentiating agent N,N'-hexamethylene bisacetamide followed by a 30-min exposure to 0.25 mM H2O2. Early irreversible ATP depletion was observed concomitant with oxidative inactivation of the mitochondrial ATP synthase. Cell proliferative capacity was also impaired within 2 h of the treatment, and progressive delayed cell lethality, starting 2 h after the insults, was also found. Based on the prevention provided by specific antioxidants and on the absence of malodialdehyde production, all the effects were ascribed to the oxidant action of .OH radicals, or closely related species, generated through iron-catalyzed reactions of H2O2, which apparently caused site-directed oxidative modifications of iron-binding proteins, in particular mitochondrial ATP synthase, rather than peroxidation of membrane lipids. Similar effects were mimicked even in the parental cell line when oligomycin was used to inhibit selectively mitochondrial ATP synthase activity, thereby lowering the enzyme activity to a level similar to that found in H2O2-damaged differentiating cells. Hence, induction of erythroid differentiation makes the mitochondrial ATP synthase a major target of H2O2 by enhancing the availability of redox-active iron in the local environment of the enzyme. Subsequent oxidative inactivation of the mitochondrial ATP synthase, resulting in severe energy impairment, leads to loss of cell growth capacity. Erythroleukemia cells may serve as a model system for the combination of two selective properties: (1) the capacity for carrying out efficient heme synthesis and/or for undergoing iron overload-like state; and (2) subsequent enhanced sensitivity to reactive oxygen species generators. Early severe mitochondrial dysfunction and energy impairment may be a major part of the mechanism of the sensitivity.

Redox properties of iron in the binding site(s) of F1ATPase from mammalian mitochondria and thermophilic bacterium PS3: a comparative study.

Iron ions in the two iron centers of beef heart mitochondrial F1ATPase, which we have been recently characterized (FEBS Letters 1996, 379, 231-235), exhibit different redox properties. In fact, the ATP-dependent site is able to maintain iron in the redox state of Fe(II) even in the absence of reducing agents, whereas in the nucleotide-independent site iron is oxidized to Fe(III) upon removal of the reductant. Fe(III) ions in the two sites display different reactivity towards H2O2, because only Fe(III) bound in the nucleotide-independent site rapidly reacts with H2O2 thus mediating a 30% enzyme inactivation. Thermophilic bacterium PS3 bears one Fe(III) binding site, which takes up Fe(III) either in the absence or presence of nucleotides and is unable to maintain iron in the redox state of Fe(II) in the absence of ascorbate. Fe(III) bound in thermophilic F1ATPase in a molar ratio 1:1 rapidly reacts with H2O2 mediating a 30% enzyme inactivation. These results support the presence in mitochondrial and thermophilic F1ATPase of a conserved site involved in iron binding and in oxidative inactivation, in which iron exhibits similar redox properties. On the other hand, at variance with thermophilic F1ATPase, the mitochondrial enzyme has the possibility of maintaining one equivalent of Fe(II) in its peculiar ATP-dependent site, besides one equivalent of Fe(III) in the conserved nucleotide-independent site. In this case mitochondrial F1ATPase undergoes a higher inactivation (75%) upon exposure to H2O2. Under all conditions the inactivation is significantly prevented by PBN and DMSO but not by Cu, Zn superoxide dismutase, thus suggesting the formation of OH radicals as mediators of the oxidative damage. No dityrosines, carbonyls or oxidized thiols are formed. In addition, in any cases no protein fragmentation or aggregation is observed upon the treatment with H2O2.

Characterization of the binding of Fe(III) to F1ATPase from bovine heart mitochondria.

The binding Fe(III) to F1ATPase purified from beef heart mitochondria has been characterized by chemical analyses and EPR spectroscopy. F1ATPase binds 2 mol of Fe(III)/mol of protein selectively in the presence of saturating concentrations of ATP. In the absence of nucleotides or in the presence of either saturating ADP or limiting ATP concentrations, the enzyme binds 1 equivalent of Fe(III). F1ATPase pretreated with 5'-p- fluorosulfonylbenzoyladenosine, that selectively modifies the non-catalytic sites, binds only 1 mol of Fe(III)/mol of protein in the presence of either saturating ATP or ADP, Fe(III)-loaded F1ATPase containing either 1 or 2 equivalents of Fe(III) show identical EPR signals at g=4.3. The signals are not perturbed by the binding of nucleotides to the enzyme while they are altered by phosphate addition. These results indicate that F1ATPase contains two distinct Fe(III)-binding sites, which differ from nucleotide-binding sites, and that one of these sites is opened up for Fe(III) uptake by conformational changes induced by binding of ATP to the loose non-catalytic site.

Enhancement of daunomycin toxicity by the differentiation inducer hexamethylene bisacetamide in erythroleukemia cells.

Cytotoxic effects of daunomycin were investigated upon differentiation of Friend erythroleukemia cells induced with hexamethylene bisacetamide, a process during which a 20-fold increase in the hemoglobin content occurred. Daunomycin proved to be more toxic to differentiated Friend cells than to their undifferentiated counterparts. No changes in the daunomycin uptake rates of the two cell types were detectable. Externally added catalase and desferrioxamine mesylate protected against the additional cytotoxicity of daunomycin in differentiated cells, pointing to hydrogen peroxide and iron ions as mediators of the toxic effect. Daunomycin-dependent, cyanide-insensitive oxygen consumption of control and induced cells did not differ significantly, and the rate of formation of the daunomycin semiquinone radical electron paramagnetic resonance signal was similar in both cell types, indicating that the difference in toxicity was not due to increased drug activation by plasma membrane enzymes. Differentiated cells had a lowered catalase content; the cellular iron content was shown to increase by 2.8-fold upon cell differentiation, with hemoglobin-bound iron being about 50% of the total. Altogether, the results suggest increased intracellular hydrogen peroxide generation mediated by hemoglobin, combined with a decrease in catalase activity and an increase in accessible iron, as responsible for the higher sensitivity to daunomycin shown by differentiated Friend cells. This represents the first experimental system where the increase in anthracycline cytotoxicity upon cell differentiation can be attributed to redox activation and the formation of reactive oxygen species.

Differentiation potentiates oxidant injury to mitochondria by hydrogen peroxide in Friend's erythroleukemia cells.

Oxidative damage to mitochondrial functions was investigated upon non-lethal treatment with H2O2 of Friend's erythroleukemia cells induced to differentiate, in comparison with the parental cell line. Both respiration and maximal ATP synthase capacity were more severely diminished by H2O2 in induced cells. The effects were mediated by intracellular redox-active iron and OH. radicals. Specifically, the mechanisms of the selective oxidant injury to F0F1 ATP synthase observed in differentiating cells likely involved impairment of F0-F1 coupling sensitive to oligomycin. We suggest a Fenton-like reaction of H2O2 with iron ions, more available in the differentiating cells, as occurring at the surface and/or in the lipid bulk phase of the inner mitochondrial membrane, thus injuring subunits responsible for the coupling of F0F1 ATP synthase through generation in situ of the actual damaging species. Besides, we propose heme iron as the most likely candidate for such reaction in induced cells actively synthesizing heme. In accordance, pretreatment of uninduced cells with hemin made H2O2-damage qualitatively identical.

H2O2-induced damage to beef heart mitochondria F0F1 ATP synthase complex: differential sensitivity of the F1 and F0 moieties.

Exposure of purified mitochondrial F0F1 ATP synthase to H2O2 resulted in a marked inhibition of the ATPase activity, irrespective of the purification procedure used and of the incorporation of the enzyme complex into phospholipid vesicles. The inactivation appeared consequent to oxidative modifications of the F1 moiety, whereas damage to the F0 sector, leading to low enzyme activity through impaired binding with F1, seemed not to occur. In fact, when H2O2-inactivated complex was deprived of F1, no loss of the capacity of the F0 sector thus obtained to properly reassemble with untreated purified F1 was apparent, because the resulting enzyme complex showed full activity and oligomycin sensitivity. On the contrary, the exposure of the isolated components F1 or F0 to H2O2, followed by reassembly with untreated F0 and F1 respectively, resulted in both cases in lower catalytic activity of the reconstituted complexes, whereas low oligomycin sensitivity was exhibited only in the case of F0 treatment, suggesting the inactivation in this case as due to oxidative modifications leading to impaired binding with F1.

Plasma membrane as a site of redox activation of daunomycin in intact human erythrocytes. Quantitative evaluation of the hydrogen peroxide produced by the membrane with respect to the cytosol.

The relative importance in human red blood cells of the plasma membrane as a site of redox activation of anthracyclines as compared to hemoglobin was evaluated by assaying the H2O2 produced upon exposure to daunomycin. The method of H2O2-dependent irreversible inhibition of catalase (EC 1.11.1.6) activity by 3-amino-1,2,4-triazole was applied to intact erythrocytes, as well as to isolated membranes with added purified catalase. The results obtained indicate a secondary role in daunomycin activation for the plasma membrane from a quantitative point of view, although membrane pathways can be more harmful than cytosolic pathways, especially towards extracellular targets, when the high efficiency of the cytosolic antioxidative defences and the external location of the membrane activation site are considered.

The inactivation of mitochondrial F1 ATPase by H2O2 is mediated by iron ions not tightly bound in the protein.

Exposure to purified mitochondrial F1 ATPase to continuous flux of H2O2 resulted in significant loss (up to 60%) of the ATP hydrolytic activity. The presence of chelating agents including desferrioxamine or previous selective removal of the iron ions not tightly bound in the protein completely prevented the inactivation, whereas re-loading of the enzyme with F3+ restored the sensitivity to H2O2. A marked protective effect was provided as well by mannitol or by Cu,Zn superoxide dismutase. The results indicated the decomposition of H2O2 by redox-active iron-protein adducts as responsible for the enzyme inactivation, probably through site-directed generation of more highly reactive oxygen species. A possible role for iron associated to F1 component in the oxidation, aging and turnover of ATP synthase complex in vivo may be suggested on the basis on these results.

Enhancement of hyperthermic damage on M14 melanoma cells by liposome pretreatment.

Exponentially growing human melanoma cells (M14 cell line) were pretreated with various amounts of dipalmitoylphosphatidylcholine-containing multilamellar liposomes and then exposed to heat treatment (42.5 degrees C). Cell damage produced by the treatments, given separately or in combination, was evaluated in terms of cell survival. Our results demonstrate that the cell survival at 37 degrees C was not affected by liposome concentrations up to 1000 nmol of phospholipid/2.5 x 10(6) cells, while liposome treatment of cells before heat exposure determined a marked damaging effect even at 100 nmol of phospholipid/2.5 x 10(6) cells. The mechanisms of liposome-cell interaction have been investigated by electron microscopy or by electron spin resonance measurements of spin-labeled membranes of intact cells. Evidence has been obtained that liposomal lipids are either taken up by M14 cells or become incorporated in the cell membrane. The present data suggest the possibility that liposome treatments per se could be of potential value as a therapeutic approach, by increasing the effect of heat therapy.

Cytotoxicity of a low molecular weight Cu2Zn2 superoxide dismutase active center analog in human erythroleukemia cells.

The cytotoxicity of SOD-mimics was studied in human K562 erythroleukemia cells. CuPUPY, a low molecular weight copper complex with properties typical of a Cu2Zn2 SOD active center analog was shown to display pronounced toxicity upon incubation with human K562 erythroleukemia cells, while the ligand, CuSO4 or CuEDTA did not affect vitality. Externally added catalase decreased the cytotoxic effects of CuPUPY by 50% indicating an involvement of hydrogen peroxide in toxicity. An increased oxygen uptake and glutathione oxidation by K562 cells in the presence of CuPUPY suggested that toxicity might be due to a copper-mediated redox-cycle. In fact addition of glutathione to a solution of CuPUPY resulted in glutathione oxidation, O2-consumption and H2O2-generation. CuPUPY proved to be less toxic to human lymphocytes than to K562 cells. This selectivity may be related to the low content of antioxidative enzymes in K562 cells.

Generation of daunomycin radicals on the outer side of the erythrocyte membrane.

The generation of the daunomycin semiquinone was studied in intact red blood cells under CO atmosphere by ESR spectroscopy. The undialyzed hemolysates and the spin broadening agent chromium oxalate quenched the ESR signal, suggesting external location of the ESR-detectable radicals and their slow diffusion inside. A constant outward flow of O2- was detected by monitoring the approach to the steady state of the ESR signal of Cu,Zn superoxide dismutase externally added to red blood cells plus daunomycin in air. This suggests a reductase on the outer side of the erythrocyte membrane as the source of daunomycin radicals.

Liposome-mediated increase of the superoxide dismutase content in human erythrocytes: characterization by electron spin resonance.

Room temperature electron spin resonance (ESR) spectroscopy, together with enzyme activity measurements, have been used to study the liposome-mediated enrichment of human red blood cells with superoxide dismutase. The ESR technique was found to be useful not only for qualitative and quantitative determinations, but also for demonstration of enzyme activity inside intact cells. The molecular properties of superoxide dismutase in relation to pharmacological and biotechnological applications are reviewed.