Satoru Ueno

A comparative study of the ability of ferric nitriloacetate and other iron chelators to assist membrane lipid peroxidation by superoxide radicals

Abstract This study examined some of the variables determining the efficiency of lipid peroxidation in egg yolk phosphatidylcholine liposomes and in microsomes exposed to enzymatically-generated superoxide radicals. The initiation of peroxidation required the presence of preformed lipid peroxides and a chelated metal catalyst. Comparison of the relative effectiveness of four iron chelating agents showed that the chelate must bind to the membrane by coulombic attraction between the charged membrane and a chelate carrying an opposite net charge. Of the chelates tested, only the carcinogenic ferric nitriloacetate (Fe3+-NTA) was an effective catalyst of oxidation of all membranes, whether carrying a net charge, or not. We postulate that the unique catalytic capacity of the ferric nitriloacetate (Fe3+-NTA) can be explained by its existence in two forms at neutral pH, each binding to oppositely charged membranes and initiating their peroxidation. This gives the complex the unique ability to bind to any membrane, which may be a factor in its carcinogenicity.

Photocurrent of purple membrane adsorbed onto a thin polymer film: action characteristics of the local anesthetics

Abstract The proton pumping activity of bacteriorhodopsin (bR) in the purple membrane adsorbed onto a thin polymer film as a solid support for electrical measurements has been examined in the presence of local anesthetics and 1-alcohols as an anesthetic model. This membrane adsorbed system provided high reproducibility of the photocurrents in bR due to the mechanical and the chemical stability and the electric properties of the thin polymer film. As the concentrations of the local anesthetics increased, the photocurrents generated by the proton pump of bR were cooperatively suppressed and the changes in the photocurrents were reversible. From the dose–response curves for the anesthetics, the concentration (EC 50 ) required for a 50% suppression showed a marked specificity in the order of lidocaine>bupivacaine>tetracaine>dibucaine. The suppression of the photocurrent in bR was more effective for the uncharged form of the local anesthetics than for the charged one. The absorption and fluorescence spectra suggested that the charged form of the anesthetics was bound to the purple membrane surface, while their uncharged form interacted with the hydrophobic portions of the purple membrane interior rather than with the membrane surface. From the dose–response curves for the 1-alcohols, an increase in hydrophobicity in their molecules effectively suppressed the photocurrent of bR. We found that the binding of hydrophobic organic cations such as tetracaine hydrochloride and bupivacaine hydrochloride to the blue membrane with loss of the proton pump, which was prepared by removal of the cations from the purple membrane, could regenerate the proton pumping activity. The photocurrent in bR in the purple membrane adsorbed onto a thin solid film sensitively responded to different local anesthetics.

Ability of ferric nitrilotriacetate complex with three pH-dependent conformations to induce lipid peroxidation.

This study examined the generation of reactive oxygen species (ROS) and the induction of lipid peroxidation by carcinogenic iron(III)-NTA complex (1:1), which has three conformations with two pKa values (pKa1≈4, pKa2≈8). These conformations are type (a) in acidic conditions of pH 1-6, type (n) in neutral conditions of pH 3-9, and type (b) in basic conditions of pH 7-10. The iron(III)-NTA complex was reduced to iron(II) complex under cool-white fluorescent light without the presence of any reducer. The reduction rates of three species of iron(III)-NTA were in the order type (a)≫type (n) ≫ type (b). Iron(III)-NTA-dependent lipid peroxidation was induced in the presence and absence of preformed lipid peroxides (L-OOH) through processes associated with and without photoreduction of iron(III). The order of the abilities of the three species of iron(III)-NTA to initiate the three mechanisms of lipid peroxidation was: (1) type (a) ≫ type (n) ≫ type (b) in lipid peroxidation that is induced L-OOH- and H2O2-dependently and mediated by the photoreduction of iron(III); (2) type (b) ≫ type (n) ≫ type (a) in lipid peroxidation that is induced L-OOH- and H2O2-dependently but not mediated by the photoreduction of iron(III); (3) type (n) ≫ type (b) ≫ type (a) in lipid peroxidation that is induced peroxide-independently and mediated by the photoactivation but not by the photoreduction of iron(III). The rate of lipid peroxidation induced L-OOH-dependently is faster than that induced H2O2-dependently in the mechanism (1), but the rate of lipid peroxidation induced H2O2-dependently is faster than that induced L-OOH-dependently in the mechanism (2). In the lag process of mechanism (3), L-OOH and/or some free radical species, not 1O2, were generated by photoactivation of iron(III)-NTA. These multiple pro-oxidant properties that depend on the species of iron(III)-NTA were postulated to be a principal cause of its carcinogenicity.

Synthesis, crystal structure and electronic properties of a 2Fe–2S complex with a bulky thiolato ligand, [NEt4]2[Fe2S2(SC6H2Me3-2,4,6)4]

The complex [NEt4]2[Fe2S2(tmbt)4](tmbt = 2,4,6-trimethylbenzenethiolate) was synthesised and its molecular structure determined. The crystal system is monoclinic with a= 15.241(6), b= 16.174(5), c= 16.768(6)A, β= 135.84(2) and Z= 2, in space group P21/c. The Fe–S*(S*= inorganic sulphide) bond lengths are 2.195(6) and 2.204(6)A and the Fe–S(C) are 2.299(8) and 2.318(6)A. Steric congestion at the thiolate ligands causes rotation of two of the for Fe–S(C) torison angles from the stable staggered position to the restrictedly eclipsed position. The eclipsed Fe–S(C) has a wide Fe–S–C angle [112.35(76)°] with π-bond character and another staggered Fe–S(C) possesses a narrow Fe–S–C angle [104.95(64)°] with less π-bond character.

Effect of water activity on unfolding of adsorbed protein at the interface

Abstract The rates of change in film pressure, II, of proteins during the adsorption of human serum albumin, γ-globulin, lysozyme, myoglobin, concanavalin A, and β-casein at the air-water interface were monitored using the Wilhelmy plate method. Assuming that unfolding of an adsorbed protein is initiated at the surface of an aqueous solution through the reaction of surface-activated water with the polar internal bond of the protein, the number of water molecules, n , causing a loss in local rigidity of the protein interior was estimated from the kinetic data of protein adsorption. The magnitude of the n values for these proteins with different secondary structures were closely correlated with the kinetic data of the hydrogen-deuterium exchange reaction in protein solution. The present results suggested that the driving force inducing a loss in the internal structure of proteins would be the access of water into the interior of the protein.

Monovalent anions stabilize the structure of bacteriorhodopsin at the air-water interface

Abstract The structural stability of bacteriorhodopsin (bR) in purple membrane (PM) and the effects of monovalent anions on the secondary structure of bR at the air-water interface were examined using a monolayer technique and Fourier transform IR spectrometry. The π- A isotherms of PM monolayer films were condensed in the presence of 50 mM sodium halides. The order of increasing effect was I − >Br − >Cl − >F − . The α helical bR molecule in the PM film at the interface is partially transformed into a β sheet with incubation time after spreading of the PM film. Compared with the α helix content of 62% for bR prepared from the subphase water (no ion), the contents in the presence of these ions were 74.4% for fluoride, 78.5% for chloride, 80.4% for bromide and 83.8% for iodide. The addition of anions to the subphase significantly protected the conformational change of bR at the interface. This is due to the direct binding of the anions to the bR molecule in the PM.

Dispersion—polar surface free energies of polypeptide films

Abstract The dispersion energy contribution γ d s and the polar contribution γ p s to the total surface free energy γ s of ten kinds of synthetic polypeptides were determined by contact angle measurements. A linear relationship for α-helical polypeptides existed between the ratio of γ d s to γ s and γ s . The γ d s contribution to the γ s is essentially governed by the non-polar moieties of the radially distributed side-chains around an α-helix backbone. The deviation from the linear relationship for β-sheeted poly( l -valine)(PLVal) suggests that the isopropyl side-chains of PLVal act with difficulty as an effective shield for the polypeptide backbone. The γ d s contribution to the γ s for polypeptide films is smaller for the β sheet than for the α helix.

Redox potentials of the oriented film of the wild-type, the E194Q-, E204Q- and D96N-mutated bacteriorhodopsins

The redox potentials of the oriented films of the wild-type, the E194Q-, E204Q- and D96N-mutated bacteriorhodopsins (bR), prepared by adsorbing purple membrane (PM) sheets or its mutant on a Pt electrode, have been examined. The redox potentials (V) of the wild-type bR were -470 mV for the 13-cis configuration of the retinal Shiff base in bR and -757 mV for the all-trans configuration in H(2)O, and -433 mV for the 13-cis configuration and -742 mV for the all-trans configuration in D(2)O. The solvent isotope effect (DeltaV=V(D(2)O)-V(H(2)O)), which shifts the redox potential to a higher value, originates from the cooperative rearrangements of the extensively hydrogen-bonded water molecules around the protonated C=N part in the retinal Schiff base. The redox potential of bR was much higher for the 13-cis configuration than that for the all-trans configuration. The redox potentials for the E194Q mutant in the extracellular region were -507 mV for the 13-cis configuration and -788 mV for the all-trans configuration; and for the E204Q mutant they were -491 mV for the 13-cis configuration and -769 mV for the all-trans configuration. Replacement of the Glu(194) or Glu(204) residues by Gln weakened the electron withdrawing interaction to the protonated C=N bond in the retinal Schiff base. The E204 residue is less linked with the hydrogen-bonded network of the proton release pathway compared with E194. The redox potentials of the D96N mutant in the cytoplasmic region were -471 mV for the 13-cis configuration and -760 mV for the all-trans configuration which were virtually the same as those of the wild-type bR, indicating that the D to N point mutation of the 96 residue had no influence on the interaction between the D96 residue and the C=N part in the Schiff base under the light-adapted condition. The results suggest that the redox potential of bR is closely correlated to the hydrogen-bonded network spanning from the retinal Schiff base to the extracellular surface of bR in the proton transfer pathway.

Alkane derivative-bacteriorhodopsin interaction: proton transport and protein structure.

The effects of alkane derivatives, 1-alcohols (ROH), aliphatic amine hydrochlorides (RNH(2).HCl) and sodium aliphatic carboxylates (ROONa), on the proton pumping activity of bacteriorhodopsin (bR) in a purple membrane have been examined. Photocurrents in bR in the purple membrane adsorbed onto polyester thin film were recorded before and after exposure to these test substances. The peak photocurrent in bR was reversibly suppressed by each substance. From the dose-response curve, the concentrations required to reduce the peak capacitive current by 50% were determined for each homolog and then the standard free energies per CH(2) for the adsorption of the alkane derivatives to the site of action were estimated: -3.13 kJ mol(-1) for ROH, -3.05 kJ mol(-1) for RNH(3)(+), and -2.95 kJ mol(-1) for ROO(-). The proton pumping activity of bR was mainly suppressed by the hydrophobic interaction with the additive. The relative potencies of the functional groups of the alkane derivatives were almost comparable between 1-octanol (C(8)OH) and octylamine hydrochloride (C(8)NH(3)(+)) and about 10 times less effective for sodium octanoate (C(8)OO(-)) than for others. The addition of C(8)OH or C(8)OO(-) changed the absorption spectra of bR with a maximum at 560 nm to the spectra of the intermediate state with a maximum at 480 nm, while the C(8)NH(3)(+) decreased the intensity of the 560 nm band only with no blue-shift by the 480 nm band. We conclude that the action of the alkane derivatives is nonspecific and directed to all organized purple membrane structures and that the binding sites of the ROH and ROO(-) are different from that of RNH(3)(+).

Penetration of 1-alkanols into monolayers of α-helical polypeptides with hydrophobic side chains

Abstract Penetration of 1-alkanols into monolayers of α-helical polypeptides with hydrophobic side chains at two different compressional states is studied by surface potential and surface pressure measurements. It was found that at an initial surface pressure of 3.0 mN m −1 there existed a time difference in achieving the final equilibrium between the surface potential and the surface pressure for poly (γ-dodecyl-L-glutamate) (PDOLG) and poly (γ-methyl-L-glutamate) (PMLG) monolayers after adsorption of 1-alkanol. This disagreement was accounted for by the time-dependent reorientation of polypeptide side chains at the interface after adsorption. The free energies of adsorption per CH 2 group of 1-alkanol to the monolayers spread at the initial surface pressures of 3.0 and 13.5 mN m −1 were −2.90 and −2.46 kJ mol −1 for PDOLG, and −2.70 and −2.57 kJ mol −1 for PMLG, respectively. The affinity of 1-alkanol to the PDOLG monolayer was stronger than that of the PMLG monolayer.