ISSN 00063509, Biophysics, 2011, Vol. 56, No. 2, pp. 323–330. © Pleiades Publishing, Inc., 2011. Original Russian Text © V.V. Belov, E.L. Mal’tseva, N.P. Palmina, 2011, published in Biofizika, 2011, Vol. 56, No. 2, pp. 372–380.
DISCUSSIONS
Modification of the Structure of Plasmatic Membranes of the Liver by the Action of αTocopherol in Vitro V. V. Belov, E. L. Mal’tseva, and N. P. Palmina Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, Moscow, 119991 Russia Email:
[email protected] Received June 5, 2010; in final form, August 27, 2010
Abstract—The effect of the natural antioxidant αtocopherol in a broad concentration range (10–4– 10 ⎯25 M) on the viscosity characteristics and thermally induced structural transitions of a lipid bilayer of plasma membranes of murine hepatocytes in vitro has been studied. Changes in the rigidity of surface (~8 Å) and microviscosity of the deeplying regions (~20 Å) of the lipid bilayer were measured on a Bruker EMX EPR spectrometer (Germany) by the method of spin probes. Stable nitroxyl radicals of 5 and 16doxylstearic acid, localized at different depth in the membrane served as spin probes. It was shown that the concentration dependence of the effect of αtocopherol is linear and polymodal with three statistically significant increases in three ranges of its concentration: (1) in the range of traditional physiological concentrations 10–4–10–9 M, (2) in the range of superlow doses 10–9–10–17 M, and (3) in the range of «imaginary» concentrations 10–17– 10–25 M. The mechanisms of action of αtocopherol in each of the three ranges are discussed. When studying the temperature dependences of viscous characteristics, we discovered a new thermally induced structural transition in the range of «physiological» temperatures 309–313 K for those αtocopherol concentrations (including superlow ones) to which the maxima on the dose dependence curves at constant temperature of 293 K corresponded. Keywords: αtocopherol, superlow doses, spin probes, microviscosity of lipids, thermally induced structural transitions in lipids DOI: 10.1134/S0006350911020047
INTRODUCTION αTocopherol executes in biological membranes an important function on preservation of their integ rity, participating in regulation of lipid peroxidation (LPO) in the capacity of one of the most effective nat ural antioxidants [1, 2]. Owing to its lipophilicity, localizing in all biological membranes, αtocopherol exerts influence on their structural state, presenting as a central link in the cycle of LPO, on which depends the activity of many membranebound enzymes par ticipating in most important biochemical processes in the cell [3]. At the present time the modifying influence of α tocopherol on membrane structure is studied only in a rather limited range of physiological concentrations; meanwhile more and more topicality is gained by the problem of the action of biologically active substances (BAS) in supersmall doses (SSD) [4–6]. On this topic a vast experimental material has been accumulated (see review [4] and references therein), however the Editor’s Note: This text is the closest possible equivalent of the original publication with all its factual statements and terminol ogy, phrasing and style, meticulously prepared so that the reader may more clearly recognize the major problems with this line of scholarly activity. A.G.
mechanisms of this phenomenon have not totally been studied. It is supposed that in the role of one of critical targets, cellular and subcellular membranes may act. Thus earlier in a series of investigations [7, 8], includ ing those conducted in our laboratory [9–13], it has been discovered that upon introduction of BAS in supersmall doses into an animal organism, cell culture or suspension of biological membranes there occurs a change in the structural characteristics of the lipid bilayer. Plasma membranes (PM) present special interest as the place of localization of a series of vitally necessary regulatory systems, in particular systems of secondary messengers [14, 15] determining the gen eral metabolic status of the organism. The aim of the given work appeared as a study of the influence of α tocopherol in a broad range of concentrations (10–4– 10–25 M) on the structure of plasma membranes in vitro. EXPERIMENTAL Alcohol–water solutions of αtocopherol were obtained by the method of consecutive dilution in quartzware by next nearest order of its initial 10–1 M solution with alcohol of high rectification from Merck (Germany) to a concentration of 10–3 M, and then
323
324
BELOV et al. (a)
(b)
Probe С5 O
N–O⋅ O
OH
Probe С16 O
N–O⋅
OH O
2Amax 2Amin
I+
10 G
I0
I–
ΔH+
10 G ΔH0
Fig. 1. Structural formulae and EPR spectra of probes (a) C5 (5doxylstearic acid) and (b) C16 (16doxylstearic acid), incorpo rated into plasma membranes of mice liver cells. Probe concentration – 6 × 10–5 M. Protein concentration 2.5 mg/mL. Temper ature 293 K.
with distilled water. Plasma membranes were isolated of liver cells of mice (40–50 animals) of line F1(C57 × DBA2) by consecutive centrifugation by the method described in work [16]. Protein concentration was determined by the Lowry method [17]. The obtained membranes were resuspended in the isolation medium; protein concentration was brought to 2.5 mg/mL sus pension. Then the suspension was poured out into eppendorfs of volume 1 mL each and stored in a freez ing chamber at a temperature of –50°C in the course of 2–3 months. At the day of experiment a necessary amount of membranes was thawed, suspension shaken and used for work in 150 min with thermostating at 293 K (the physicochemical properties of the given membranes in the course of this time and in the pro cess of storage remained immutable). The structural dynamic state of membrane was studied by the method of spin probes [18, 19] with the aid of a Bruker EMX EPR spectrometer (FRG). We determined the viscous characteristics at constant temperature 293 K, such as rigidity of the surface regions of lipids (~8 Å) and microviscosity of the deep lying ones (~20 Å), and also the thermoinduced struc tural transitions in the lipid bilayer. The soughtfor parameters were calculated from EPR spectra obtained as a result of three acquisitions, in a semiau tomatic mode with the help of program Origin 6.1. In the capacity of a spin probe for study of the rigidity of the surface regions of the membrane we used a stable nitroxyl radical – 5doxylstearic acid (probe C5) (Fig. 1а). Membrane rigidity in the region of localiza
tion of this probe (~8 Å) was described with an order ing parameter S [19], depending on the amplitude of deviation of the large axis of the ellipsoid of rotation of the probe from the mean direction of orientation of the surrounding lipid molecules: S = 1.66(Amax – Amin/Amax + 2Amin), where Amax and Amin are values corresponding to half the distance between outer and inner extrema of the EPR spectrum (Fig. 1a), the inaccuracy of measuring which in a wellresolved spectrum did not exceed 0.2 G. For the study of microviscosity of the deeplying regions of the lipid bilayer in the capacity of a spin probe we applied a stable nitroxyl radical – 16doxyl stearic acid (probe C16) (Fig. 1b). Membrane micro viscosity in the region of localization of this probe (~20 Å) was characterized by the times of its rotational correlation τc1 and τc2, which were calculated by for mulae for rapid anisotropic rotation of a radical [18]: τc1 = 6.65ΔH+{(I+/I–)1/2 – 1} × 10–10 s, τc2 = 6.65ΔH0{(I0/I–)1/2 – 1} × 10–10 s, where ΔH+ and ΔH0 are widths of lowfield and central components of the spectral line respectively (inaccu racies not exceeding 0.6%), while I+, I0 and I– – intensities of lowfield, central and highfield compo nents of the EPR spectrum respectively (inaccuracies not exceeding 1.5%) (Fig. 2b). In the Results and Dis cussion section the data are analyzed on the basis of BIOPHYSICS
Vol. 56
No. 2
2011
PLASMA MEMBRANE MODIFICATION BY IMAGINARY TOCOPHEROL
where ϕ is parameter τc or S calculated by experimen tal data; σ – standard deviation. As a result for τc the relative random error in determining the mean value did not exceed 1%, while for S – 0.2%. In the experiment into a membrane suspension of volume 200 μL with incorporated spin probe (final concentration 6 × 10–5 M) we added 3.3 μL of a corre sponding solvent and placed into a long, narrow planeparallelfaced quartz cuvette of high Q, owing to which throughout the volume we achieved the same temperature. Thermostating was actualized with the aid of a Bruker ER 4131 VT thermoattachment. The precision of thermostating was controlled by a ther mocouple placed into the resonator and constituted 0.05°C. Spectra were registered every 3–5 min in the course of 35–40 min. After this the membranes were transferred into a tube, supplemented with αtoco pherol in a definite concentration, the membrane sus pension was thoroughly stirred on a shaker with a pre set speed in the course of 1 min, again placed into the cuvette and then into the EPR resonator. In this way, control and test measurements were conducted on one and the same membranes. In separate control experi ments it was established that repeated addition of 3.3 μL of solvent, additional pouring over and shaking of the membrane suspension does not influence the determined parameters. Inasmuch as we used homo geneous material, then the relative standard errors of the mean values obtained with account of all controls for membranes of one isolation did not differ from the aboveindicated values. It should be noted that the deviations of mean values of τc and S for various mem branes found in the literature sources of different years (even in those cases when there was no strict thermo stating) were close to the values obtained by us [20–24]. For obtaining temperature dependences of the dynamic characteristics of the membrane the measure ments were conducted in a temperature range from 285 to 320 K with an interval of 2 K. Structural transitions in the lipid bilayer were imaged by inflection points between linearized stretches of temperature depen dences presented in Arrhenius coordinates – log (dynamic parameter) vs. 1/T. As inflection points we took those points the addition of which to the lin earized stretch of the graph brought the correlation coefficient beyond the limits of the norms determined by the number of degrees of freedom and statistical reliability 95%. For every concentration of αtoco BIOPHYSICS
Vol. 56
No. 2
2011
(a)
3.0 2.5 Effect, %
t ( 95%, f )σ ( s ϕ ) rand = , n
3.5
2.0 1.5 1.0 0.5 0
–0.5 4
6
3.5
8 10 12 14 16 18 20 22 24 26 –log [αtocopherol] (b)
3.0 2.5 Effect, %
parameter τc1, inasmuch as the results on τc2 are in a significant degree identical. The main contribution to the error in determining the mean values of quantities S and τc was provided by a random component, determined first of all by the statistical character of the studied quantities:
325
2.0 1.5 1.0 0.5 0
–0.5 4
6
8 10 12 14 16 18 20 22 24 26 –log [αtocopherol]
Fig. 2. Action of αtocopherol on the rigidity of the surface (a) and microviscosity of the deeplying (b) regions of plasma membranes at a temperature of 293 K. Results are presented in percentages relative to control. Protein con centration in the membrane suspension 2.5 mg/mL. In the control, S = 0.644 ± 0.001, τc = (2.10 ± 0.04) × 10–9 s.
pherol on one preparation of isolated membranes we conducted from three to five measurements. In total we have conducted two series of experiments on mem branes isolated in different time of the year. The graphs present the mean values of effects of αtocopherol, expressed in percentages relative to control. The mag nitude of effect was estimated by formula: ϕ test – ϕ control Effect = 100%, ϕ control where ϕ is parameter calculated from experimental data (in our experiments τc or S). Relative standard errors for these effects were obtained after statistical treatment of all results by methods of parametric and nonparametric statistics with the use of computer pro gram packages Statistica and Origin 6.1 at a statistical reliability of 95%.
326
BELOV et al. –logτc 8.9
(a)
8.8
8.7
8.6
8.5 3.15 3.20 3.25 3.30 3.35 3.40 3.45 3.50 1/T × 103, K–1 –logτc (b) 8.9
8.8
8.7
8.6
8.5 3.15 3.20 3.25 3.30 3.35 3.40 3.45 3.50 1/T × 103, K–1 Fig. 3. Temperature dependences of τc in Arrhenius coor dinates in control and upon action of αtocopherol in vitro in concentrations: (a) asterisk – control; dark triangle 10 ⎯4; light circle – 10–7; dark square – 10–10 M; (b) asterisk – control; light square – 10–14; dark circle – 10 ⎯18; light triangle – 10–22 M. The maximal relative error of the obtained values –log[τc] in a series of three mea surements for each value of temperature did not exceed 2%. Experiments were repeated thrice for membranes of each isolation.
RESULTS AND DISCUSSION The viscous characteristics of a membrane present as important parameters describing its structural dynamic state at a set temperature, therefore it was of interest to study the influence of αtocopherol on the rigidity of the surface and microviscosity of the deep lying regions of lipids at a temperature of 293 K. These data are presented in Fig. 2a,b and expressed in per centages relative to control, at that each point repre sents a result of averaging from 6 to 10 independent measurements (for each concentration). The pre sented dose dependences have a polymodal character, typical of the action of BAS in a broad range of con centrations, including the range of supersmall doses, where maxima corresponding to an increase in rigidity
of the surface and microviscosity of the deeplying regions of lipids are observed in definite intervals of doses, separated between each other by socalled «dead zones» in which the effect does not manifest itself. In the range of traditional «physiological» con centrations (10–4–10–9 M) of αtocopherol and in the range of supersmall doses (10–9–10–17 M) the effects are observed both in surface (~8 Å) and in deeplying (~20 Å) regions of membrane. However in the range of socalled «imaginary» concentrations (
