Membrane surface potential measurements with amphiphilic spin labels

Ionizable Nitroxides for Studying Local Electrostatic Properties of Lipid Bilayers and Protein Systems by EPR

Electrostatic interactions are known to play a major role in the myriad of biochemical and biophysical processes. Here, we describe biophysical methods to probe local electrostatic potentials of proteins and lipid bilayer systems that are based on an observation of reversible protonation of nitroxides by electron paramagnetic resonance (EPR). Two types of probes are described: (1) methanethiosulfonate derivatives of protonatable nitroxides for highly specific covalent modification of the cysteine's sulfhydryl groups and (2) spin-labeled phospholipids with a protonatable nitroxide tethered to the polar head group. The probes of both types report on their ionization state through changes in magnetic parameters and degree of rotational averaging, thus, allowing the electrostatic contribution to the interfacial pKa of the nitroxide, and, therefore, the local electrostatic potential to be determined. Due to their small molecular volume, these probes cause a minimal perturbation to the protein or lipid system. Covalent attachment secures the position of the reporter nitroxides. Experimental procedures to characterize and calibrate these probes by EPR, and also the methods to analyze the EPR spectra by simulations are outlined. The ionizable nitroxide labels and the nitroxide-labeled phospholipids described so far cover an exceptionally wide range of ca. 2.5-7.0 pH units, making them suitable to study a broad range of biophysical phenomena, especially at the negatively charged lipid bilayer surfaces. The rationale for selecting proper electrostatically neutral interface for probe calibration, and examples of lipid bilayer surface potential studies, are also described.

Surface electrostatics of lipid bilayers by EPR of a pH-sensitive spin-labeled lipid

Many biophysical processes such as insertion of proteins into membranes and membrane fusion are governed by bilayer electrostatic potential. At the time of this writing, the arsenal of biophysical methods for such measurements is limited to a few techniques. Here we describe a, to our knowledge, new spin-probe electron paramagnetic resonance (EPR) approach for assessing the electrostatic surface potential of lipid bilayers that is based on a recently synthesized EPR probe (IMTSL-PTE) containing a reversibly ionizable nitroxide tag attached to the lipids' polar headgroup. EPR spectra of the probe directly report on its ionization state and, therefore, on electrostatic potential through changes in nitroxide magnetic parameters and the degree of rotational averaging. Further, the lipid nature of the probe provides its full integration into lipid bilayers. Tethering the nitroxide moiety directly to the lipid polar headgroup defines the location of the measured potential with respect to the lipid bilayer interface. Electrostatic surface potentials measured by EPR of IMTSL-PTE show a remarkable (within ±2%) agreement with the Gouy-Chapman theory for anionic DMPG bilayers in fluid (48°C) phase at low electrolyte concentration (50 mM) and in gel (17°C) phase at 150-mM electrolyte concentration. This agreement begins to diminish for DMPG vesicles in gel phase (17°C) upon varying electrolyte concentration and fluid phase bilayers formed from DMPG/DMPC and POPG/POPC mixtures. Possible reasons for such deviations, as well as the proper choice of an electrostatically neutral reference interface, have been discussed. Described EPR method is expected to be fully applicable to more-complex models of cellular membranes.

Nitroxide radical biostability in skin

Nitroxide radicals are important chemical tools in dermatologic research (e.g., for studying biophysical properties of skin lipids and epidermal membranes with the method of electron paramagnetic resonance, EPR, spectroscopy). However, nitroxides may loose their paramagnetic properties in biological tissues, which could limit their usefulness in biomedical applications. We analyzed the biostability of various chemical types of nitroxide radicals in keratinocytes, epidermis homogenate, and intact skin. EPR signal loss of imidazoline, pyrrolidine, piperidine, and oxazolidine nitroxides is attributed to their reduction to the corresponding hydroxylamine. The rate of nitroxide reduction in skin varies considerably with nitroxide ring structure and substitution. The order of nitroxide stability in isolated human keratinocytes, mouse epidermis homogenate, and intact mouse and human skin is imidazoline > pyrrolidine > di-t-butylnitroxide (DTBN) > piperidine > oxazolidine. Cationic nitroxides are reduced much faster than neutral or anionic probes, presumably due to transmembrane electron shuttle or internalization. The results indicate that imidazoline- and pyrrolidine-type nitroxides should be used when high biostability of nitroxides is needed. Piperidine-type nitroxides are versatile probes for studying one-electron transfer reactions in skin.

Procaine incorporation into human erythrocyte membrane--a spin label study

The positioning of procaine among the components of human erythrocyte membranes was investigated by electron paramagnetic resonance (EPR) using several spin labelled molecules: fatty acids 5-, 12- and 16-doxyl-stearic acids (5DSA, 12DSA, 16DSA), procaine (SLP) and a quaternary ammonium salt (CAT4). It was proved that while procaine significantly modified EPR parameters of both CAT4 and 5DSA, it had no effect at the level of the twelfth or sixteenth carbon atom of the membrane lipid chains. Simultaneous incorporation of 5DSA and SLP and SLP in erythrocyte membranes revealed a shorter than 10 A distance between paramagnetic centers of 5DSA and SLP. It is inferred that procaine interacts with the membrane not only at lipid water interface, but also (at least) up to the fifth carbon atom of the hydrophobic tail of phopolipids. Procaine inclusion does not change the transition point (40 degrees C) at which a break appears in the temperature dependence of the erythrocyte volume, hence no strong interaction between the procaine and the cytoskeleton or the integral proteins responsible for this phase transition occurs. In the temperature domain of 25-42 degrees C, procaine seems to be freely diffusible within the membrane (Debye model can be applied).

Surface potential changes of mitoplasts in the presence of pyridoxal phosphate modified cytochromes c

1. The addition of native cytochrome c to mitoplasts leads to a decrease of surface potential of the mitoplast membrane. However the surface potential is slightly decreased (approximately 3 mV) when PLP(Lys 86)-cytochrome c and PLP(Lys 79)-cytochrome c were added. 2. The native and PLP-modified cytochromes c do not influence the order parameters S and isotropic constant a when both spin probe I and probe II were used. It is shown that cytochrome c binding to the membrane does not affect the hydrophobic intermembrane area as well as the lipid arrangements of the mitoplast membrane. 3. At low ionic strength there was observed a significant difference in the membrane potential when PLP-cytochromes c were added to the mitoplasts. 4. At high ionic strength the addition of native or PLP-modified cytochromes c does not change the membrane potential.

Nitroxide-stimulated H2O2 decomposition by peroxidases and pseudoperoxidases

Nitroxide free radicals interact with Hb/metHb, Mb/metMb and with peroxidases/phenols to induce a catalase-like conversion of H2O2 to O2 (catalatic activity), without being substantially consumed in the process. The mechanism of this reaction is postulated to involve a one-electron oxidation of the nitroxide to the immonium oxene, which then reacts further to release oxygen and the nitroxide. An involvement of the immonium oxene in the reaction mechanism is consistent with ferryl heme reduction by nitroxides and a detection of the reduced nitroxide when the reaction mixture is supplemented with the two-electron reductant sodium borohydride. The nitroxide-induced catalatic activity is completely inhibited when the reaction mixture is supplemented with glutathione. Nitroxides suppress free radical formation by hydroperoxide-activated heme proteins, as inferred from their inhibition of the spin-trapping of glutathionyl radicals. H2O2 decomposition and a suppression of reactive free radical formation by heme proteins appears to be an antioxidant activity of nitroxides, which is distinct from their previously reported superoxide dismutating activity and which may be a factor in their protective action in models of cardiac reperfusion injury.

Free radical reduction mechanisms in mouse epidermis skin homogenates

Scavenging mechanisms for persistent free radicals were investigated using nitroxide-type radicals as model compounds. The free radical reducing activity of a) isolated thioredoxin reductase, a flavin containing oxidoreductase, b) skin homogenates, and c) the epidermis of hairless mice was studied by electron spin resonance spectroscopy. In all three systems, reduction rates of different classes of nitroxide free radicals exhibited the following order: oxazolidinoxy greater than piperidinoxy greater than dihydropyrroloxy. The main reductant for piperidinoxy radicals in mouse skin homogenate is ascorbic acid. Other reducing activities were stimulated by NAD(P)H and could be inhibited by N-ethyl maleimide, suggesting involvement of thiol-dependent processes. Mammalian thioredoxin, a competitive inhibitor of nitroxide reduction by thioredoxin reductase, significantly stimulates nitroxide scavenging in skin homogenate. Thioredoxin reductase did not significantly participate in nitroxide reduction in skin homogenates. At the surface of mouse epidermis a cationic dihydropyrroloxy nitroxide, which was stable in the presence of mammalian thioredoxin reductase was readily reduced. The epidermal reduction was inhibited by zinc, N-ethyl maleimide, and by heat (70 degrees C, 5 min). At least for mouse epidermis, reduction of a variety of nitroxides is a complex phenomenon involving enzymatic and nonenzymatic mechanisms and cannot be used as a specific assay for an enzyme, e.g., thioredoxin reductase. The study indicates the epidermis contains an effective antioxidant system that scavenges ascorbate-sensitive piperidinoxy nitroxides as well as more reducing radicals exemplified by dihydropyrroloxy nitroxides.

Metabolism in rat liver microsomes of the nitroxide spin probe tempol

Paramagnetic nitroxide spin labels have been extensively used to probe various biophysical and biochemical properties of the cellular environment. Recently nitroxides have been proposed as contrast enhancing agents in proton magnetic resonance imaging and contrast enhancement has been demonstrated in animal studies. Nitroxides, possessing a stable unpaired electron, increases the relaxation rates of protons, providing an enhancement of contrast. Nitroxides are metabolized intracellularly principally via reversible reduction to hydroxylamines. Rates of reduction depend on the physical characteristics of the nitroxides, in general 5-membered pyrrolidine ring are reduced more slowly than those with a 6-membered piperidine ring. Oxidation back to the nitroxide is relevant for lipid soluble hydroxylamines, while is low for water soluble ones. It is known that nitroxides are metabolized by subcellular fractions (cytosol, mitochondria, microsomes), though the enzymatic and non-enzymatic systems involved are poorly characterized. In the present study, the first of the necessary steps toward a systematic study of the metabolism of nitroxides by subcellular organelles, we have chosen to study the metabolism of 4-hydroxy 2,2,6,6-tetramethylpiperidine-N-oxyl in isolated rat liver microsomes. Microsomes were able to reduce Tempol slowly without any substrate addition; when NADPH was added, the reduction rate substantially increased. In phenobarbitone induced rats the reduction rate was significantly higher than in not-induced microsomes. NADPH-dependent reduction rate was inhibited by thallium chloride (an inhibitor of the flavin-centered cytochrome P-450 reductase), superoxide dismutase, and by N-ethylmaleimide; menadione increased it. The Tempol reduction rate was not significantly affected by various cytochrome P-450 inhibitors with the sole exception of metyrapone. A solution containing purified cytochrome P-450 reductase and NADPH readily reduced Tempol. Microsomes fortified with NADPH were able to reduce Tempol at an appreciable rate. In order to distinguish between reduction of nitroxides to hydroxylamine or destruction of nitroxides following nitroxide reduction, microsomal suspensions were treated with a mild oxidant (ferricyanide 0.5-10 mM). The recovery varied from 40 to 60%, indicating a process of probe destruction leading to as yet unknown metabolites. The present study clearly indicates that, in this model system, cytochrome c (P-450) reductase and not cytochrome P-450 is responsible for the observed Tempol metabolism; along with hydroxylamine formation, other Tempol derived metabolites are formed during the process.

Amine and carboxylate spin probe permeability in red cells

Permeabilities for a homologous series of amine and carboxylate nitroxide spin probes were measured in human red blood cells by an electron paramagnetic resonance (EPR) method. Permeabilities determined in this study are much lower than would be predicted for a sheet of bulk hydrocarbon and the polarity of the rate-limiting region is shown to be greater than bulk hydrocarbon. This suggests that the rate-limiting region for permeation of these nonelectrolytes is somewhere in the membrane periphery rather than in the center of the membrane. The red cell membrane does not discriminate between these probes on the basis of molecular volume, as might be predicted by a simple free-volume theory of membrane permeation.

Enhancement by tetraphenylboron of the interaction of the 1-methyl-4-phenylpyridinium ion (MPP+) with mitochondria

Inhibition of mitochondrial energy production by MPP+ may be the key step in chemically-induced Parkinson's disease. Tetraphenylboron (TPB-) markedly enhances the effect of MPP+. Inhibition of respiration and uptake of MPP+ are accelerated, the former by up to two orders of magnitude. TPB increases the final concentration of MPP+ in the matrix by 2-3 fold, insufficient to explain the rapid inhibition of respiration. TPB- lowers the membrane surface potential by only about 20%, but increases the partitioning of MPP+ into organic solvent by one order of magnitude. TPB- also enhances the effect of MPP+ on inverted membranes, reducing the I50 by an order of magnitude. We suggest that TPB- acts by ion pairing with MPP+ to facilitate penetration into mitochondria as well as access to a hydrophobic inhibition site on NADH dehydrogenase.

Binding of lipophilic cations to the liposomal membrane: thermodynamic analysis

Lipophilic ions are widely used as probes for measuring membrane potentials. Since binding of the probes to the membrane interferes with the accurate estimation of the membrane potential, it is necessary to clarify the characteristics of probe binding to membranes. The present paper deals with the binding of lipophilic cations to liposomes. The results can be summarized as follows: (1) The binding of triphenylmethylphosphonium, its homologues and tetraphenylphosphonium to liposomes of dipalmitoylphosphatidylcholine followed the Langmuir adsorption isotherm. (2) Spin-labeled lipophilic cations were synthesized and the binding to liposomes of egg phosphatidylcholine was examined. The binding also followed the Langmuir adsorption isotherm. The dissociation constant (the concentration giving half-maximal binding), K, was independent of the temperature, indicating that the binding is entropy-driven. (3) The binding was influenced by the fluidity of the membrane. Except in the case of triphenylmethylphosphonium (TPMP+), K and A (maximum amounts of binding) increased above the transition temperature. In other words, above the phase transition temperature the binding affinity is decreased, while maximum amounts of binding are increased for all phosphoniums used except TPMP+.

Free radical reduction by thioredoxin reductase at the surface of normal and vitiliginous human keratinocytes

Cell cultures of human keratinocytes contain membrane-associated thioredoxin reductase that is extremely active in reducing radicals on the outer plasma membrane. This enzyme activity was confirmed by its purification from cultures of stratified human keratinocytes by affinity column chromatography. The enzyme was assayed both in vivo and in vitro using a spin-labeled quaternary ammonium compound as the substrate, under saturating conditions in free radical substrate. Specific activities were determined by monitoring the sequential decrease in the amplitude of the electron spin resonance signal per unit of cell protein. The following properties were found: Cultures of adult stratified cells have approximately twice the thioredoxin reductase activity of neonatal cells. The enzyme is inhibited by thioprotein inhibitors (i.e., parachloromecuribenzoate and dinitrochlorobenzene). The activity is regulated by calcium concentrations of the cell culture medium. Stratified keratinocytes are half as active in medium containing 2 mM Ca++ compared with 0.1 mM Ca++ concentration. Product inhibition of the enzyme occurs with oxidized coenzyme NADP+ (i.e., 87% inhibition of enzyme activity over 30 min). The enzyme is heat stable at temperatures of 70 degrees C for 10 min. It is inactivated at 75 degrees C. A comparative study of thioredoxin reductase activity on stratified differentiated and undifferentiated rapidly growing cells was performed. Also, enzyme activity was quantitated for cultured keratinocytes isolated from vitiliginous and normal skin of the same donor. The results of this study, and the connection between this enzyme activity and UV-generated free radicals are reconciled in terms of the mechanism of action and metabolic activity of thioredoxin reductase.

Internalization of the spin-labelled surface potential probe CAT12 by energized mitochondria

Rat liver mitochondria briefly incubated in the presence of 50 microM 4-(dodecyl dimethyl ammonium)-1-oxyl-2,2,6,6-tetramethyl piperidine bromide (CAT12) bind 1.9 nmol of this spin-labelled membrane probe per mg mitochondrial protein in a form which is not reducible by ascorbate. Upon energization with ATP the amount of the non-reducible CAT12 increases by about 37%. It is concluded that CAT12 which is not reducible by externally added ascorbate is bound to the inner surface of the inner mitochondrial membrane and/or accumulated in the matrix compartment. Therefore, CAT12 is not suitable to monitor the surface potential of mitochondria and other organelles which develop a transmembrane potential, negative inside.

Hydrophobic ion interactions with membranes. Thermodynamic analysis of tetraphenylphosphonium binding to vesicles

The thermodynamic properties for the interaction of the hydrophobic ion tetraphenylphosphonium (TPP+) with egg phosphatidylcholine vesicles were studied in detail by equilibrium dialysis and spin label techniques. A partition coefficient of beta = 4.2 + 0.4 x 10(-6) cm (K congruent to 100) was determined. Electrostatic saturation sets in at approximately 600 microM (about one absorbed TPP+ molecule per 100 lipids), and is not screened by salt. The temperature dependence of binding was determined, which reveals that the binding is entropy-driven with a positive (repulsive) enthalpy of binding, a result to be compared with hydrophobic anions in which the binding enthalpy is negative. The membrane dipole potential may be responsible for this binding difference. Activity coefficients are determined and shown to be significantly different from those of most common salts, an important result that should be considered in all hydrophobic ion studies. Comparison of the TPP+ results with those of its anionic structural analogue, tetraphenylboron (TPB-), permits a general analysis of hydrophobic ion interactions with membranes. A theoretical model consistent with the entire set of data is developed in an accompanying article.

Determination of osmotic volumes and pH gradients of plant membrane and lipid vesicles using ESR spectroscopy

Volumes and pH gradients were determined with spin probes in liposomes and zucchini membrane vesicles by quantitating the internal concentrations of probes in the presence of an impermeable line-broadening agent, manganese + EDTA. Volume shrinkage in response to increasing external concentrations of MnEDTA was consistent with perfect osmotic behavior of both vesicle populations. Buffer additions were used to impose pH gradients on the vesicles; liposome gradients measured with a spin-labeled weak acid were slightly smaller than the maximum theoretical imposed gradients, whereas above a threshold magnitude, measured gradients for the plant membranes were significantly smaller than imposed gradients. However, the residual pH gradient in the zucchini vesicles decreased at about the same rate as the liposome gradient. Moreover, this residual gradient was not completely collapsed in the presence of the proton ionophore, FCCP, indicating that the vesicles were impermeable to ions; indeed, ion permeabilities of both vesicle preparations appeared to be similar during the slow phase of the pH gradient collapse. Thus, zucchini membrane vesicles are tightly sealed and appear to have a mechanism for dissipating pH gradients rapidly when these gradients exceed some threshold value.

Determination of the electric potential at the external and internal bilayer-aqueous interfaces of the human erythrocyte membrane using spin probes

Partitioning of oppositely charged amphipathic spin probes indicates that the electric potential at the external bilayer-aqueous interface of the human erythrocyte is insignificant, and that protruding sialic acids do not contribute to this potential. This potential at the surface is distinguished from the electrokinetic potential due to all charged groups within the hydrodynamic surface of shear. By contrast, using inside-out erythrocyte membrane vesicles, a substantial potential is observed at the cytoplasmic membrane surface. This can be attributed to the asymmetric distribution of acidic phospholipids on the two sides of the erythrocyte membrane bilayer.

Membrane-bound lactoferrin alters the surface properties of polymorphonuclear leukocytes

Polymorphonuclear leukocytes (PMN) aggregate and avidly attach to endothelium in response to chemotactic agents. This response may be related in part to the release of the specific granule constituent lactoferrin (LF). We found by using immunohistology and biochemical and biophysical techniques that LF binds to the membrane and alters the surface properties of the PMN. Upon exposure of PMN treated with 5 micrograms/ml cytochalasin B to 2 x 10(-7) M formyl-methionine-leucine-phenylalanine for 5 min, the PMN mobilized LF to their surface as observed by immunoperoxidase staining for LF. At added LF levels ranging from 4 to 15 micrograms/10(7) PMN there was a dose-dependent reduction in PMN surface charge reaching 4 mV, when the partitioning into the membrane of a charged amphipathic nitroxide spin label was measured by electron spin resonance spectroscopy, whereas transferrin was without effect. When 125I-FeLF was added to human PMN in increasing amounts and the results corrected for the residual amount of free LF contaminating the cells, the PMN were saturated with LF at concentrations between 100 and 200 nM in the medium. Human PMN bound 1.35 x 10(6) molecules per cell and the calculated value for the association constant for these receptors was 5.2 x 10(6) M-1. Additionally, 6 micrograms/ml LF served as an opsonin for rabbit MN to promote PMN uptake by rabbit macrophages, when assessed by electron microscopy, but lysozyme did not. These studies indicate that LF can bind to the surface of the PMN and reduce its surface charge. This correlates with enhanced "stickiness" leading to a variety of cell-cell interactions.

Surface charge changes in purple membranes and the photoreaction cycle of bacteriorhodopsin

The surface potential of purple membrane fragments, determined from the distribution of the aqueous free and the membrane-bound positively charged, paramagnetic, amphiphilic probe 4-(dodecyldimethylammonium)-1-oxyl-2,2,6,6-tetramethylpiperidine bromide varied almost 60 mV as a function of ionic strength and 50 mV as a function of pH of the medium. Light-induced changes in surface potential followed the changes observed in the M412 intermediate of the photocycle of bacteriorhodopsin as a function of pH, temperature, and response to antibiotics beauvericin and valinomycin. The number of induced charges per M412 appearing at the surface of purple membranes decreased from about 0.75 to 0.45 as the surface potential became more negative. The stoichiometry would be twice as large if the charge changes were localized exclusively on one side of the purple membrane. Laser flash-induced kinetics of the rise and decay of surface charge changes were slightly slower than the kinetics of the rise and decay of M412 which is associated with the reversible deprotonation of the retinal Schiff base nitrogen in the chromophore. It is suggested that the light-induced charge changes monitor a dissociable amino acid residue which may be a step in the movement of protons across the purple membrane.