Mechanisms of iron and copper-frataxin interactions.

Frataxin is a mitochondrial protein whose deficiency is the cause of Friedreich's ataxia, a hereditary neurodegenerative disease. This protein plays a role in iron-sulfur cluster biosynthesis, protection against oxidative stress and iron metabolism. In an attempt to provide a better understanding of the role played by metals in its metabolic functions, the mechanisms of mitochondrial metal binding to frataxin in vitro have been investigated. A purified recombinant yeast frataxin homolog Yfh1 binds two Cu(ii) ions with a Kd1(CuII) of 1.3 × 10-7 M and a Kd2(CuII) of 3.1 × 10-4 M and a single Cu(i) ion with a higher affinity than for Cu(ii) (Kd(CuI) = 3.2 × 10-8 M). Mn(ii) forms two complexes with Yfh1 (Kd1(MnII) = 4.0 × 10-8 M; Kd2(MnII) = 4.0 × 10-7 M). Cu and Mn bind Yfh1 with higher affinities than Fe(ii). It is established for the first time that the mechanisms of the interaction of iron and copper with frataxin are comparable and involve three kinetic steps. The first step occurs in the 50-500 ms range and corresponds to a first metal uptake. This is followed by two other kinetic processes that are related to a second metal uptake and/or to a change in the conformation leading to thermodynamic equilibrium. Frataxin deficient Δyfh1 yeast cells exhibited a marked growth defect in the presence of exogenous Cu or Mn. Mitochondria from Δyfh1 strains also accumulated higher amounts of copper, suggesting a functional role of frataxin in vivo in copper homeostasis.

Maghemite nanoparticles coated with human serum albumin: combining targeting by the iron-acquisition pathway and potential in photothermal therapies.

Human serum albumin (HSA), the most abundant plasma protein in human blood, is a natural transport vehicle with multiple ligand binding sites. It, therefore, constitutes an attractive candidate for drug delivery. Targeting may occur via the most known interaction of the protein with the neonatal Fc receptor (FcRn). Here, we investigate another HSA delivery path, involving the transferrin receptor, and we elaborate a maghemite-HSA nanohybrid, opening up new opportunities for medical applications. Fluorescence spectrophotometric titration and size-exclusion chromatography were used to substantiate, in cell-free assays, an interaction between HSA and the transferrin receptor R1. This occurs with a dissociation constant, KD of 6.7 nM. This interaction was confirmed in HeLa cell culture where, by confocal microscopy, rhodamine-labeled HSA is shown to be internalized. HSA was then covalently conjugated onto maghemite nanoparticles (NPs) to give a NP-HSA nanohybrid. The therapeutic potential of this hybrid was demonstrated through its heating capacity in magnetic hyperthermia (MH) and near-infrared (NIR) photothermia (PT). In particular, the Specific Absorption Rate (SAR) in the PT Therapy (PTT) mode, using a 808 nm NIR-LASER (1 W cm-2) and at iron concentration as low as 2.5 mM, was found to be very high, equal to 1870 W g-1 with a temperature increment of 9.2 °C. The nanohybrids incubated with HeLa cells were mainly localized at the cell surface. When the PTT mode was applied under the same conditions as in vitro, mortality was higher in HeLa cells than in fibroblasts (non-malignant cells). Cytotoxicity was checked in both cell lines without the PTT mode; the nanohybrids do not seem to affect cell viability. These results make the nanohybrids very promising agents for NIR-PT and for targeting in cancer therapy, since non-malignant cells were not damaged.

Design and synthesis of 3-isoxazolidone derivatives as new Chlamydia trachomatis inhibitors.

Chlamydia trachomatis (Ct) is a bacterial human pathogen responsible for the development of trachoma, the worldwide infection leading to blindness, and is also a major cause of sexually transmitted diseases. As iron is an essential metabolite for this bacterium, iron depletion presents a promising strategy to limit Ct proliferation. The aim of this study is to synthesize 3-isoxazolidone derivatives bearing known chelating moieties in an attempt to develop new bactericidal anti-Chlamydiaceae molecules. We have investigated the paths by which these new compounds affect Ct serovar L2 development in HeLa cells, in the presence or absence of exogenously added iron. The iron-chelating properties of these molecules were also determined. Our data reveal important bactericidal effects which are distinguishable from those due to iron chelation.

Transferrins: iron release from lactoferrin.

Iron loss in vitro by the iron scavenger bovine lactoferrin was investigated in acidic media in the presence of three different monoanions (NO(3)(-), Cl(-) and Br(-)) and one dianion (SO(4)(2-)). Holo and monoferric C-site lactoferrins lose iron in acidic media (pH< or =3.5) by a four-step mechanism. The first two steps describe modifications in the conformation affecting the whole protein, which occur also with apolactoferrin. These two processes are independent of iron load and are followed by a third step consisting of the gain of two protons. This third step is kinetically controlled by the interaction with two Cl(-), Br(-) and NO(3)(-) or one SO(4)(2-). In the fourth step, iron loss is under the kinetic control of a slow gain of two protons; third-order rate-constants k(2), 4.3(+/-0.2)x10(3), 3.4(+/-0.5)x10(3), 3.3(+/-0.5)x10(3) and 1.5(+/-0.5)x10(3) M(-2) s(-1) when the protein is in interaction with SO(4)(2-), NO(3)(-), Cl(-) or Br(-), respectively. This step is accompanied by the loss of the interaction with the anions; equilibrium constant K(2), 20+/-5 mM, 1.0(+/-0.2)x10(-1), 1.5(+/-0.5)x10(-1) and 1.0(+/-0.3)x10(-1) M(2), for SO(4)(-), NO(3)(-), Cl(-) and Br(-), respectively. This mechanism is very different from that determined in mildly acidic media at low ionic strength (micro<0.5) for the iron transport proteins, serum transferrin and ovotransferrin, with which no prior change in conformation or interaction with anions is required. These differences may result from the fact that in the transport proteins, the interdomain hydrogen bonds that consolidate the closed conformation of the iron-binding cleft occur between amino acid side-chain residues that can protonate in mildly acidic media. With bovine lactoferrin, most of the interdomain hydrogen bonds involved in the C-site and one of those involved in the N-site occur between amino acid side-chain residues that cannot protonate. The breaking of the interdomain H-bond upon protonation can trigger the opening of the iron cleft, facilitating iron loss in serum transferrin and ovotransferrin. This situation is, however, different in lactoferrin, where iron loss requires a prior change in conformation. This can explain why lactoferrin does not lose its iron load in acidic media and why it is not involved in iron transport in acidic endosomes.

Transferrin, is a mixed chelate-protein ternary complex involved in the mechanism of iron uptake by serum-transferrin in vitro?

Iron uptake by transferrin from triacetohydroxamatoFe(III) (Fe(AHA)3) in the presence of bicarbonate has been investigated between pH 7 and 8.2. The protein transits from the opened apo- to the closed holoform by several steps with the accumulation of at least three kinetic intermediates. All these steps are accompanied by proton losses, probably occurring from the protein ligands and the side-chains involved in the interdomain H-bonding nets. The minor bihydroxamatoFe(III) species Fe(AHA)2 exchanges its iron with the C-site of apotransferrin in interaction with bicarbonate without detectable formation of any intermediate protein-iron-ligand mixed complex; direct second-order rate constant k1=4.15(+/-0.05)x10(7) M(-1) s(-1). The kinetic product loses a single proton and undergoes a modification in its conformation followed by the loss of two or three protons; first-order rate constant k2=3.25(+/-0.15) s(-1). This induces a new modification in the conformation; first-order rate constant k3=5.90(+/-0.30)x10(-2) s(-1). This new modification in conformation rate controls iron uptake by the N-site of the protein and is followed by a single proton loss; K3a=6.80 nM. Finally, the holoprotein or the monoferric transferrin in its thermodynamic equilibrated state is produced by a last modification in the conformation occurring in about 4000 seconds. But for the Fe(AHA)3 dissociation and the involvement of Fe(AHA)2 in the first step of iron uptake, this mechanism is identical to that reported for iron uptake from FeNAc3. This implies that the exchange of iron between a chelate and serum-transferrin occurs by a single general mechanism. The nature of the iron-providing chelate is only important for the first kinetic step of the exchange, which can be slowed to such an extent that it rate limits the exchange of iron.

Transferrins--a mechanism for iron uptake by lactoferrin.

Iron uptake by bovine lactoferrin from nitrilotriacetatoFe(III) [FeN(Ac)3] in the presence of bicarbonate has been investigated at pH 7.1-8.7. Deprotonated apolactoferrin interacting with bicarbonate or carbonate extracts iron from nitrilotriacetatoFe(III); the direct second-order rate constant k1 = (4.90 +/- 0.20)x10(4) M(-1) s(-1), a reverse second-order rate constant k(-1) = (1.80+/-0.05)x10(5) M(-1) s(-1), and the iron-exchange equilibrium constant K1 = 0.25+/-0.05. The newly formed iron-protein complex loses a single proton with proton dissociation constant K3a = (17+/-0.5) nM, then undergoes a modification in its conformation followed by the loss of two or three protons; the first-order rate constant k2 = (1.0+/-0.10) s(-1). This induces a new modification in the conformation; the first-order rate constant k3 = (8.75+/-0.40)x10(-3) s(-1). This second modification in conformation controls the rate of iron uptake by the N site of the protein and is followed by a single proton loss; K5a = 8.0 nM. Finally, the holoprotein or the monoferric lactoferrin in their final equilibrated states are produced by a third modification in the conformation occurring in about 9000 s. The mechanism of iron uptake by lactoferrin is very similar to that of serum transferrin with a cooperativity between the C and N sites upon iron uptake but with lower rates, higher affinities and at least one more proton loss involved. These differences may be the result of slight discrepancies in the intimate structures of binding sites for serum transferrin and lactoferrin. In order to analyse the cooperativity between these iron-binding sites, the three-dimensional position of the chain of amino acid residues separating the N and C lobes of human apo-, holo- and dicopper-lactoferrin have been compared by the recognition of the three-dimensional shape dissimilarity program. The interlobe peptides of human hololactoferrin and apolactoferrin showed only 75.5 % tridimensional similarity, indicating that iron uptake affects the three-dimensional structure of the interlobe chain.

Transferrins. Hen ovo-transferrin, interaction with bicarbonate and iron uptake.

Fe(III) uptake by the iron-delivery and iron-scavenging protein, hen ovotransferrin has been investigated in vitro between pH 6.5 and 9. In the absence of any ferric chelate, apo-ovotransferrin loses two protons with K1a = 50 +/- 1 nM and K2a = 4.0 +/- 0.1 nM. These acid-base equilibria are independent of the interaction of the protein with bicarbonate. The interaction with bicarbonate occurs with two different affinity constants, KC = 9.95 +/- 0.15 mM and KN = 110 +/- 10 mM. FeNAc3 exchanges its Fe(III) with the C-site of the protein in interaction with bicarbonate, direct rate constants k1 = 650 +/- 25 M-1 s-1, reverse rate constant k-1 = (6.0 +/- 0.1) x 10(3) M-1 s-1 and equilibrium constant K1 = 0.11 +/- 0.01. This iron-protein intermediate loses then a single proton, K3a = 3.50 +/- 0.35 nM, and undergoes a first change in conformation followed by a two or three proton loss, first order rate constant k2 = 0.30 +/- 0.01 s-1. This induces a new modification in conformation followed by the loss of one or two protons, first order rate constant k3 = (1.50 +/- 0.05) x 10(-2) s-1. These modifications in the monoferric protein conformation are essential for iron uptake by the N-site of the protein. In the last step, the monoferric and diferric proteins attain their final state of equilibrium in about 15,000 s. The overall mechanism of iron uptake by ovotransferrin is similar but not identical to those of serum transferrin and lactoferrin. The rates involved are, however, closer to lactoferrin than serum transferrin, whereas the affinities for Fe(III) are lower than those of serum transferrin and lactoferrin. Does this imply that the metabolic function transferrins is more related to kinetics than to thermodynamics?

Transferrin--interactions of lactoferrin with hydrogen carbonate.

The interaction of apolactoferrin with hydrogen carbonate (bicarbonate) has been investigated in the pH range 6.5-9.2. In the absence of bicarbonate apolactoferrin loses a single proton with pK1a of 8.10. This proton loss is independent of the interaction with the synergistic anion. The C-site of apolactoferrin interacts with bicarbonate with a very low affinity (K(-1)C = 3.2 M(-1)). This process is accompanied by a proton loss, which is probably provided by the bicarbonate in interaction with the protein. This proton loss can possibly be the result of a shift in the proton dissociation constant, pKa, of the bicarbonate/carbonate acid/base equilibrium, which would decrease from pKa 10.35 to pK2a 6.90 in the bicarbonate-lactoferrin adduct. The N-site of the protein interacts with bicarbonate with an extremely low affinity, which excludes the presence of the N-site-synergistic anion adduct in neutral physiological media. Contrary to serum transferrin, the concentration of the apolactoferrin in interaction with bicarbonate is pH dependent. Between pH 7.4 and pH 9 with [HCO3-] about 20 mM, the concentration of the serum transferrin-bicarbonate adduct is always about 30%, whereas that of the apolactoferrin-synergistic anion adduct varies from 25% at pH 7.5 to 90% at pH 9. This implies that, despite an affinity for bicarbonate two orders of magnitude lower than that of serum transferrin, lactoferrin interacts better with the synergistic anion. This can be explained by the possible interaction of lactoferrin with carbonate in neutral media, whereas transferrin only interacts with bicarbonate.

A mechanism for iron uptake by transferrin.

Iron uptake by transferrin from iron nitrilotriacetate (FeNAc3) in the presence of bicarbonate has been investigated in the pH range 6.5-8. Apotransferrin, in interaction with bicarbonate, extracts iron from FeNAc3, without the formation of an intermediate protein-iron-ligand mixed complex (iron-exchange-equilibrium constant, K1=1 +/- 0.05; direct second-order-rate constant, k1=8.0x10(4) +/- 0.5x10(4)M(-1)s(-1)., reverse second-order-rate constant, k-1=7.5x10(4) +/- 0.5x10(4)M(-1)s(-1). The newly formed iron-protein complex loses a single proton (proton-dissociation constant, Ka=16 +/- 1.5nM) and then undergoes a modification of its conformation followed by loss of two or three protons (first-order-rate constant, K2=2.80 +/- 0.10s-1). This includes a new modification in the conformation (first-order-rate constant, K2=6.2x10(2) +/- 0.3x10(-2)s(-1). This second modification in conformation controls the rate of iron uptake by the N-site of the protein and is followed by loss of one proton (K3a=6.80 nM). Finally, the holoprotein or the monoferric transferrin in its final equilibrated state is produced by a third modification in the conformation that occurs after approximately 3000 s. Iron uptake by the N-site does not occur when the apotransferrin interacts with bicarbonate. Nevertheless, it occurs with the monoferric transferrin, in which iron is bound to the C-site, in its final state of equilibrium by a mechanism similar to that of iron uptake by the C-site of apotransferrin. These modifications in the conformation of the protein occur after iron uptake by the C-site and may be important for the recognition of the protein by its receptor prior to iron delivery by endocytosis.

Transferrin, a mechanism for iron release.

Iron release from transferrin has been investigated in mildly acidic and acidic media in the presence of formate, acetate and citrate. It occurs first from the N-terminal iron-binding site (N-site) of the holoprotein. It is independent of the nature and the concentration of competing ligands and is controlled by a slow proton transfer; second-order rate constant k1 = (7.4 +/- 0.5) x 10(4) M-1 s-1 which can be attributed to a rate-limiting slow proton gain by a protein ligand subsequent to a fast decarbonation of the N-site. Iron loss from the C-terminal iron-binding site (C-site) is slower than that from the N-site and occurs by two pathways. The first is favoured below pH 4 and does not involve the formation of an intermediate ternary complex. It can be controlled by a rate-limiting slow proton-triggered decarbonation of the binding site; second-order rate constant k3 = (2.25 +/- 0.05) x 10(4) M-1 s-1. The second pathway is favoured above pH 4 and involves a mixed protein-ligand iron complex. It takes place through the slow protonation of the mixed ternary complex and depends on the nature of the competing ligand. It is faster in the presence of citrate than in that of acetate; second-order rate constant k4 = (1.75 +/- 0.10) x 10(3) M-1 s-1 for citrate and (85 +/- 5) M-1 s-1 for acetate. All these phenomena can possibly describe proton-triggered changes of conformation of the binding sites.

The mechanism of iron release from transferrin. Slow-proton-transfer-induced loss of nitrilotriacetatoiron(III) complex in acidic media.

The role of protonation of amino acid ligands involved in iron release from human serum transferrin, previously saturated with nitrilotriacetatoiron(III) complex, has been elucidated in acidic media. Iron loss occurs first from the N-terminal site at pH < 6 and is followed at pH < 4 by iron release from the C-terminal iron-binding site. Nitrilotriacetatoiron(III) release from the N-terminal site is controlled by the slow protonation of the mixed protein/nitrilotriacetatoiron(III) complex; the second-order rate constant was k3a = 9.95 +/- 0.35 x 10(4) M-1.s-1. Protonation of an amino acid ligand in the C-terminal site leads to a new protein-site-C-loaded mixed complex with dissociation constant K4 = 0.300 +/- 0.025 mM. Nitrilotriacetatoiron(III) release is the result of mixed complex dissociation and the slow rate-limiting protonation of the iron-free protein with a proton dissociation constant K5a = 0.100 +/- 0.010 mM and a second-order rate constant k5a = 4.20 +/- 0.40 x 10(3) M-1.s-1. The mechanism of iron uptake and release seems to imply that slow proton transfers can induce complex formation between iron and the amino acid ligands of each of the protein iron-binding sites. These slow proton transfers may be controlled by the change of conformation of the binding sites upon iron loss.

Phospholipid transmembrane domains and lateral diffusion in fibroblasts.

The lateral diffusion of fluorescent phospholipids in cultured Chinese hamster lung fibroblasts was examined by modulated fringe pattern photobleaching. When cells were labeled and maintained at 7 degrees C, the fluorescence remained localized at the plasma membrane. N-[6-(7-Nitrobenz-2-oxa-1,3-diazol-4-yl-amino)caproyl] sphingosylphosphocholine (C6-NBD-SphPCho) and 1-acyl-2-[6-(7-nitrobenz-2-oxa-1,3-diazol-4-yl-amino)caproyl] phosphatidylcholine (C6-NBD-PtdCho) both diffused with the same apparent lateral diffusion coefficient (D1 approximately 0.3 x 10(-9) cm2/s). By contrast, the phosphatidylserine derivative (1-acyl-2-[6-(7-nitrobenz-2-oxa-1,3-diazol-4-yl-amino)caproyl] phosphatidylserine (C6-NBD-Ptd-Ser)) gave rise to two diffusional components: a slow component, D1, analogous to that measured with the choline-containing lipids, and a fast component (D2 approximately 2 x 10(-9) cm2/s). The fast component only exists in ATP-containing cells. It was shown to be associated with C6-NBD-PtdSer translocated to the inner leaflet. This indicates that the two leaflets form very different membranous domains. At higher temperature, the same difference in mobility was observed between the choline-containing lipids and the aminolipid. However, with C6-NBD-SphPCho, a fraction of very slowly diffusing or quasi-immobilized probes gradually appeared with time. This could be attributed to sphingomyelin located in small organelles after internalization. From the amplitude of this component registered at different intervals, we calculated that approximately 50% of the plasma membrane sphingomyelin is recycled in less than 30 min in Chinese hamster fibroblasts by an ATP- and microtubule-dependent process.