Effects of probucol on phase transition and fluidity of phosphatidylcholine membranes: A spin label study

Effects of Probucol on Phase Transition and Fluidity of Phosphatidylcholine Membranes: A Spin Label Study Witold K. Subczynski, Jerzy Wojas, Vida Pezeshk, and Abbas Pezeshk WKS, JW. Department of Biophysics, Institute of Molecular Biology, Jagiellonian University, Krakow, Poland.--JW. National Biomedical ESR Center, Medical College of Wisconsin, Milwaukee, Wisconsin.--JW, VP, AP. Department of Chemistry, Moorhead State University, Moorhead, Minnesota

ABSTRACT Spin labeling methods were applied to study the structure and dynamics of phosphatidylcholine membranes as a function of temperature and the mole fraction of probucol. Multilamellar liposomes made of dimyristoylphosphatidyclcholine, dipalmitoylphosphatidylcholine both saturated, and egg yolk phosphatidylcholine, an unsaturated membrane, were used. In fluid phase membranes probucol was found to increase the order and decrease the motional freedom of alkyl chains of lipids as shown with stearic acid spin labels. The effect of probucol on order and motional freedom is more pronounced in the membrane center (16-doxylstearic acid spin label position) than in the near polar headgroup region (5-doxylstearic acid spin label position). The presence of unsaturation in alkyl chains significantly decreased the ordering effect of probucol. The main phase transition temperature of saturated bilayers was lowered by 2°C in the presence of 3 mol% of probucol and significantly broadened at higher concentrations as measured with 2,2,6,6-tetramethylpiperidine-l-oxyl (TEMPO) partitioning. Also, pretransition was no longer observed in the presence of probucol. In gel phase membranes, the effect of probucol was complex. Close to the main phase transition the motion of alkyl chains was increased, showing a regulatory effect of probucol on membrane fluidity. It is proposed that probucol is located in the membrane center as opposed to vitamin E, which locates its phenolic -OH group at the membrane surface; therefore, it inhibits lipid peroxidation in this region which is less accessible to vitamin E.

INTRODUCTION Probucol, 4,4'-(isopropylidenedithio)bis(2,6-di-tert-butylphenol), is a hydrophobic drug with a unique bis-phenol structure. Two major actions of probucol are: (1)

Address reprint requests to: Dr. Abbas Pezeshk, Department of Chemistry, Moorhead State University, Moorhead, MN 56563.

JournaloflnorganicBiochemistry,55, 1-11 (1994) © 1994 Elsevier Science Inc., 655 Avenue of the Americas, NY, NY 1 0 0 1 0

1 0162-0134/94/$7.00

2

W K Subc.qn&i et al.

lowering the lipid level, especially the cholesterol level, in blood plasma, which was shown in animals and in patients with hypercholesterolemia [l, 21 and (2) inhibition of formation of atherosclerotic lesions as demonstrated in animals [3, 41. The mechanisms through which these actions are achieved are not fully understood on the molecular level. It is thought that different processes contribute to reduction in serum lipids and cholesterol and to reduction in lesion formation [4-S]. The last effect of probucol as an effective antioxidant limits oxidative modification of low density lipoproteins (LDL) which results in the development of macrophage into foam cells [9, 101. The macrophage-derived form cells which contain most of the stored lipids are considered to be the initial form of the fatty streak-lesion [ll, 121. After oral administration probucol is primarily transported by blood. Prolonged treatment of animals and humans with probucol shows its accumulation in blood, where 90% of the drug remains in blood plasma and 10% in erythrocytes [13]. In plasma almost all probucol is located in lipoprotein particles, 44% in LDL, 38% in very low density lipoprotein (VLDL), and 13% in high density lipoprotein (HDL). After prolonged treatment the mean plasma level is about 7 mg/L. Probucol slowly accumulates in adipose tissue, reaching concentrations of about 100 times higher than in plasma, in adrenal glands about 25 times, in liver and in heart about 2 times higher. Also, the elimination half-time for probucol is very long and was measured as 23-47 days 121. Recently, McLean et al. [14] measured the effect of probucol on cholesteryl ester physical states in dry mixtures, phospholipid-containing dispersions, and cells using differential scanning calorimetry and polarized light microscopy. They found that probucol decreases the transition temperature of each system by 2°C at concentrations varying from l-10 mol% in these systems. They also reported that probucol changes the phase state of cholesteryl ester droplets in cells to a more fluid phase. Probucol is a lipophilic drug and is located in organisms in lipid environments such as lipoproteins of blood plasma, fats of adipose tissue, and membranes of erythrocytes and other cells, where it acts, giving wide therapeutic effects. Not much is known about its localization, motion, and effect on the structure of these systems. The main attention has been paid to the lipoprotein particles [15, 161 and not much to model and biological membranes. To our knowledge, only one investigation on the effect of probucol on phase transition of dimyristoylphosphatidylcholine (DMPC) has been reported ([17], for review see Ref. 8). There are two major approaches to study the interaction of drugs with membrane systems using the electron spin resonance (ESR) spin-labeling technique. In the first approach, a spin label is covalently attached to the drug and its distribution and membrane permeability can be determined [18-201. In the second approach, the effect of the drug on the structure and dynamics of membrane can be detected using lipid soluble spin labels [21-231. Since probuco1 has a large degree of hydrophobicity for having four tert-butyl groups, the second approach has been used to study its effect on fluidity and phase transition of different phosphatidylcholine membranes. The most probable location of probucol within the lipid bilayer is proposed and compared with the localization of vitamin E. Finally, the significance of localization of probucol in membranes for its therapeutic action is discussed. Figure 1 illustrates the chemical structures of probucol, vitamin E, and spin labels and also their approximate localization in DMPC bilayers.

EFFEmS

OF PROBUCOL

ON PHASE TRANSITION

AND FLUIDITY

3

VITAMIN E

PROBUCOL

5-stvx isSASL

AQUEOUS PHASE

HEAD GROUP REGION

HYDROCAfWON PHASE

AQUEOUS PHASE

FIGURE 1. Cross-sectional drawing of DMPC bilayer including probucol, vitamin E, and spin labels. This figure provides chemical structures of molecules and their approximate locations within the bilayer.

MATERIALS AND METHODS Materials Spin labels were purchased from Molecular Probe, Inc. (Eugene, OR), and probucol from Aldrich (Milwaukee, WI). Phospholipids from Sigma (St. Louis, MO) were DMPC and dipalmitoylphosphatidylcholine (DPPC) with saturated alkyl chains containing 14 and 16 carbon atoms/chain, respectively, and egg yolk phosphastidylcholine (EYPC) containing a mixture of phospholipids with both saturated and unsaturated alkyl chains. Sample Preparations The membranes used in this work were multilamellar dispersions of lipids containing various amounts of probucol (from 0 to 20 mol%) and when indicated 1 mol% of stearic acid spin label. Briefly, membranes were prepared according to the following method [24]: chloroform solutions of the lipids, probucol, and when indicated, spin labels, were mixed (containing lo-’ moles of total amount of molecules) and the chloroform was evaporated with a stream of nitrogen gas and then under a reduced pressure (0.1 mm Hg) for at least 12 hr. Buffer (0.1 mL) was added to dried lipid at about 20°C above the phase transition temperature of PC membranes and vortexed vigorously. The buffer used for the study was 0.1 M borate at pH 9.5. To ensure that all stearic acid spin label (SASL) probe carboxyl groups are ionized in phosphatidylcholine (PC) membranes a rather high pH was chosen [25, 261. The structures of PC membranes are not altered at this pH [27, 281. For phase transition experiments the buffer contained 5 ‘X 10v4 M of 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO). For other experiments the lipid dispersion was centrifuged briefly at 4”C, and a portion of the fluffy pallet (containing about 20% lipid wt/wt> was

4

W K Subczynski et al.

then transferred to a capillary tube for ESR measurements. For experiments with TEMPO, pH 7.0 was occasionally used where no change was observed in the results. ESR Measurements ESR spectra were obtained with either a Varian E-109 or a Varian E-231 X-band spectrometer with Varian temperature control accessories. The sample was taken into a gas-permeable capillary (0.9 mm i.d.) made of methylpentene polymer called TPX [29]. This plastic is permeable to oxygen and nitrogen and is substantially impermeable to water. The TPX capillary was placed inside the ESR dewar insert and equilibrated with nitrogen gas that was used for temperature control. The sample was thoroughly deoxygenated yielding correct ESR line shape and preventing possible oxidation of the sample. Temperature of the sample was measured with a copper-constantan thermocouple located in the sample tube just above the active volume of the ESR cavity. RESULTS AND DISCUSSION Effect on Phase Transitions The TEMPO partition coefficient parameter, f, defined as H/(H + P> (Fig. 2) is roughly proportional to the fraction of the lipid which is in the fluid state [30]. Plots off vs temperature have been used to determine the main phase transition and pretransition temperature for model membranes [311. This method has been employed for the measurements of the influence of probucol on the main phase transition and pretransition of DMPC and DPPC membranes. The results are presented in Figure 2. Addition of 3 mol% probucol lowers the temperature of the main phase transition of DMPC and DPPC membranes by about 2°C and causes its broadening; also, the pretransition is not observed. Increasing the probucol concentration to 10 mol% in DMPC membrane causes the disappearance of the abrupt change in the magnitude off as observed for pure lipid at the

a60

0

10

20 30 TEMPERATURE

40

50

(‘C)

FIGURE 2. TEMPO partition coefficient parameter versus temperature (cooling experiment) for DMPC liposomes (left) and DPPC liposomes (right). Arrows indicate main phase transitions of pure lipids. Membranes contain: no probucol (x); 3 mol% (a); 10 mol% (A 1,or 20 mol% of probucol(0).

EFFECTS

OF PROBUCOL

ON PHASE TRANSITION

AND FLUIDITY

5

main phase transition. The parameter f decreases slowly with decreasing temperature in a wide temperature region, with the deepest slope at 20-21°C. Further, increasing the probucol concentration does not influence these changes. In DPPC membranes, increasing the probucol concentration to 10 mol% causes a further decrease of phase transition temperature by about 2.5”C and its broadening. Further increasing of probucol concentration to 20 mol% causes only broadening of transition without additional shifting. Effect of probucol on main phase transition of DPPC is weaker than on DMPC. Similar effects on phase transition were reported by McLean and Hagaman with 5 mol% probucol using fluorescence probe measurements in DMPC membranes [17]. Measurements with 5- and 16-doxylstearic acid spin labels (5- and 16-SASL) also show shifting and broadening of the main phase transition of DMPC and DPPC in the presence of probucol. Effect on Motion of Alkyl Chains Spin labels 5- and 16-SASL were used to examine the influence of probucol on the order and mobility of alkyl chains in saturated and unsaturated PC membranes. Because of structural similarities, it seems apparent, as noted previously, that a stearic acid spin label is a good probe of the alkyl chain mobility of phospholipids [24, 321. The maximum splitting values of ESR spectra have been used as a parameter to monitor motional freedom of the nitroxide group of 5-SASL. Maximum splitting parameters decrease as motional freedom of 5-SASL increases. Figure 3 shows the temperature profiles (cooling experiments) of maximum splitting of 5-SASL in DMPC membranes containing: 0, 3, 10, and 20 mol% probucol. There are three remarkable features in this figure. (1) In fluid phase membranes the presence of probucol decreases the mobility of 5-SASL. (2) The phase transition, indicated by an abrupt changing of the maximum splitting, shifts to lower temperatures by about 2°C in the presence of 3 mol% of probucol and disappears at 10 and 20 mol%. (3) In gel phase membranes, probucol increases alkyl chain motion of 5-SASL for temperatures close to the main phase transition, but decreases alkyl chain motion at lower temperatures.

TEMPERATURE

(‘Cl

FIGURE 3. Maximum splitting values of 5SASL in DMPC bilayers plotted as a function of temperature (cooling experiment) without (x) and in the presence of 3 mol% (O), 10 mol% (A 1, or 20 mol% of probucol ( 0).

6

W K Subqnski et al.

The maximum splitting value is directly related to the order parameter of .5-SASL only in fluid phase membranes where 5-SASL undergoes a rapid anisotropic motion about the long axis of spin label [33]. The maximum splitting increases with an increase in alkyl chain order. The influence of lipid alkyl chain length and unsaturation on the effect of probucol on maximum splitting of 5-SASL in fluid phase PC membranes is shown in Figure 4. When compared at the same temperature (25‘0, it can be seen that probucol has a strong ordering effect on 5-SASL in DMPC bilayers while its ordering effect in EYPC bilayers is very weak. At 50°C probucol shows a visible ordering effect on 5-SASL in DMPC bilayer but no ordering effect in DPPC bilayer and a weak disordering effect in EYPC bilayer. The data show that an increase in temperature would decrease the ordering effect of probucol on 5-SASL in EYPC membranes. It is, however, difficult to evaluate the influence of temperature on the effect of probucol in DMPC bilayers. The changing of the maximum splitting caused by 20 mol% probucol would decrease when temperature would increase. The changing caused by probucol could be expressed in terms of a change in temperature, which has the same effect as probucol. At 3O”C, 20 mol% probucol causes the same effect as lowering the temperature by 7”C, while at 50°C the effect of probucol is comparable to a temperature decrease of 6°C. On the basis of this comparison, in DMPC bilayers an increase in temperature does not weaken the effect of probucol. In DMPC bilayers and at 25°C 20 mol% probucol causes the same effect as about 10 mol% of cholesterol [24] or 10 mol% of polar carotenoids [32]. The effect of probucol and polar carotenoids on the order of 5-SASL depends on alkyl chain length, while the effect of cholesterol showed no dependence on alkyd chain length. The effect of all these modifiers is much weaker in unsaturated membranes. In fluid phase membranes, 16-SASL has an isotropic motion, thus a different analysis of ESR spectra should be used. The effective correlation time, assuming isotropic rotational diffusion of 16-SASL, can be calculated from the linear term of the line width parameter: 72B= 6.51 x 10-10AH,[(h,/h_)“2

- (h,/h+)“2]

s;

and with the quadratic term: 7

2c = 6.51 x 10-‘“AH,[(h,/h_)1’2

+ (h,,/h+)“2

- 21 s.

FIGURE 4. Maximum splitting values of 5-SASL in fhiid phase DMPC membranes (*I. DPPC membranes (A) and EYPC membranes (0) are plotted as functions of mole fractions of probucol at different temperatures.

EFFECTS

OF PROBUCOL

ON PHASE TRANSITION

AND FLUIDITY

7

AH, is the peak-to-peak width of the central (M, = 0) line in gauss and h,, h,, and h_ are heights of the low (M, = + 11, central (M, = 01, and high (M, = - 1) field peaks, respectively. When 728 and rzc are similar it is argued that the motional model is fairly good and motion is isotropic [34]. Addition of probucol decreases motional freedom of 16-SASL free radical moiety which is monitored by a large increase in correlation time (Fig. 5). At high temperatures, 45-55°C r2a and 72~ are very similar, which suggests that probucol decreases the rate of motion but does not influence its isotropy. At lower temperatures, an increase in probucol concentration would increase the difference between r2a and 72c, indicating the onset of anisotropic rotational diffusion. 20 mol% of probucol

I

I 03

10 PROBUCOL

I

03

I

I

I

10

20

PROBUCOL FIGURE

5.

20 MOL%

I

MOL%

Effective rotational correlation time of 16-SASL in DMPC (A), DPPC (B), and EYPC (C) membranes plotted as functions of mole fractions of probucol at different temperatures (0) Tag; (0) Tag.

8

W K Subczynskiet al.

causes the same effect as about 10 mol% of polar carotenoids at 15°C in EYPC or at 25°C in DMPC membranes. At 50°C the effect is the same as 5-7 mol% of polar carotenoids in EYPC and DMPC membranes [321. In DMPC membranes, the effect of 20 mol% of probucol can be compared with the decrease of temperature by about 10°C for all temperature regions (Fig. 5A). In EYPC membranes, 20 mol% of probucol causes the same effect as decreasing the temperature by 8°C (at 25”C), (6°C (at 35”(Z), and 5°C (at 50°C) (Fig. 5C); 3 mol% of probucol causes a similar effect in DPPC and DMPC (Fig. 5A and B), while at higher probucol concentrations, saturation of the probucol effect in DPPC bilayer was observed. From an Arrhenius display of the data (log T versus l/T), the activation energy of rotational motion of the nitroxide moiety of 16-SASL was calculated for DMPC and EYPC containing 0, 10, and 20 mol% of probucol, and for DPPC containing 0 and 3 mol% of probucol. The results for DMPC are collected in Table 1 showing a decrease of activation energy in the presence of probucol. The plots of log7 versus l/T for EYPC are not linear and it is difficult to characterize rotational diffusion of the 16-SASL free radical moiety in EYPC membranes based on a single set of activation-energy. However, the data show that at any given temperature, probucol decreases the activation barrier for rotational diffusion of 16-SASL. In similar experiments we have previously shown that polar carotenoids cause a decrease in activation energy and the rate of motion of 164ASL in DMPC membranes [321. This suggests that probucol has a similar effect on the motion of 16-SASL in PC membranes as polar carotenoids, zeaxanthin, and violaxanthin. Polar carotenoids cross the membrane with a rigid rod-shaped molecule perpendicular to the membrane surface 1351.Their ordering effect on lipid alkyl chains is especially pronounced in the membrane center [32]. GENERAL DISCUSSION The present data indicate that in lipid bilayers probucol increases alkyl chains order and decreases their reorientational motion. The effect of probucol is similar to that of polar carotenoids, zeaxanthin, and violaxanthin. Ordering effect of probucol at the C-5 position decreases very quickly when the membrane thickness increases. Also, the ordering effect of probucol decreases with increasing the temperature, giving a disordering effect at 50°C in EYPC membranes. At the membrane center, the effect of probucol on alkyl chain order and TABLE 1. Activation Energy for Rotational Motion of 16-SASL in DMPC Bilayers in Fluid Phase Membranes Calculated from 72B and 72c Way of calculation From 72B From 72c

Probucol

AE, (Kcal mole - ’ )

0 mole % 10 mole % 20 mole % 0 mole % 10 mole % 20 mole %

7.68 6.40 5.79 8.61 8.46 7.66

Temperature Range WI 25-70 25-50 25-50 25-70 25-50 25-50

EFFECTS

OF PROBUCOL

ON PHASE TRANSITION

AND FLUIDITY

9

motion is somewhat weaker in EYPC and DPPC membranes than in DMPC bilayers, but does not depend on alkyl chain length as strongly as at the C-5 position. Probucol also decreases the activation energy of reorientational motion of 164ASL, as polar carotenoids do. On the basis of the above data, we conclude that probucol is located in the membrane center with its long axis preferentially parallel to the alkyl chains. Probucol does not possess any groups which could “anchor” somewhere in the membrane, because its polar phenolic -OH groups are strongly shielded by teti-butyl groups (see Figs. 1 and 6). It therefore should diffuse freely within the lipid bilayer with major localization in the membrane center as discussed above. It should be transported easily within the cell using the extended membrane system as routes [36], but escape from the membrane with difficulty. Probucol transport across cytosol systems does not seem to be effective. This also explains its long elimination half-time from organisms [2]. Localization and the effect of probucol on membrane structure is illustrated in Figure 6 for short-chain (left) and long-chain membranes (right). Different localizations of probucol and vitamin E are also shown. In animal and human membranes which contain large amounts of cholesterol, free space in the central part of the membrane is created.because the cross section of steroid ring is larger than its hydrocarbon tail [37]. It is, therefore, feasible that probucol is located in the central part of the membrane. In Figures 1 and 6 it is shown that the localization of phenolic -OH groups of vitamin E and probucol, which are responsible for termination reactions with lipid peroxyl radicals, differs significantly. Lipid peroxyl radicals are generated in the central region of lipid bilayers and appear at the membrane surface only due to vertical fluctuations (bending) of alkyl chains of phospholipids. Since vitamin E locates its phenolic -OH group within the polar headgroup region [38], it reacts with peroxyl radicals during collisions at the membrane surface. Phenolic rings of probucol, on the other hand, are located in the membrane center between the alkyl chains and react with peroxyl radicals in the lipid core where they are generated. Based on this model membrane study, it is possible to propose that probucol is located in the inner part of the lipid core of lipoproteins. This proposal is supported by a recent study by Gotoh et al. [16], who compared the antioxident activity of probucol with vitamin E against lipid peroxidation. They suggested that vitamin E is predominantly located at or near the surface of LDL and that probucol is within the core of LDL as well as outer

FIGURE 6. Schematic drawing showing the interaction of probuco1 with PC membranes of short allcyl chains as DMPC (left) and long all@ chains as DPPC or

o==s=-JR%-PC

PROBUCOL

VITAMIN E CHOLESTEROL

EYPC including cholesterol (right). The membranes include vitamin E to indicate different localization of phenolic -OH groups (-OH is represented as a black circle) within the bilayer.

10 W. K Subczynski et al.

monolayers. This explains pharmaceutical effects of probucol to prevent lipid peroxidation. Probucol provides an additional pathway to vitamin E for termination of chain reactions during lipid peroxidation. It should be noted that there is a hypothesis by Barclay and Ingold [39] that lipid peroxyl radicals, because of their polarity, immediately after their formation within the lipid alkyl chains diffuse to the membrane surface where the phenolic -OH groups of vitamin E are located. However, there is no proof for this assumption and on the basis of today’s knowledge, we prefer the explanation that involves the dynamic vertical fluctuation of peroxyl radicals. This research was supported by Grant No. RI5 GM-42066 &lP) and by the National Biomedical ESR Center through NIH Grants Nos. RR 01008 and GM-22923. Partial support by the Moorhead State University Foundation is also appreciated

REFERENCES 1. C. Cortese, C. B. Marenah, N. E. Miller, and B. Lewis, Atherosclerosis 44,319 (1982). 2. R. Fellin, A. Gasparotto, G. Valerio, M. R. Baiocchi, R. Padrini, S. Lamon, E. Vitale, G. Baggio, and G. Grapaldi, Atherosclerosis 59,47 (1986). 3. T. Kita, Y. Nagano, M. Yokoda, K. Ishii, N. Kume, A. Ooshima, H. Yoskida, and C. Kawai, Proc. NatL Acad. Sci. USA 84,5928 (1987). 4. T. E. Carew, D. C. Schwenke, and D. Steinberg, Proc. Natl. Acad. Sci. USA 84,772s (1987). 5. S. Parthasarathy, S. G. Young, 3. L. Witztum, R. L. Pittman, and D. Steinberg, J. Clin. Invest. 77, 641 (1986). 6. A. Yamamoto, S. Takaichi, H. Hara, 0. Nishikawa, and S. Yokoyama, Atherosclerosis 62, 209 (1986). 7. M. M.-T. Buckley, K. L. Goa, A. H. Price, and R. N. Brogden, Drugs 37, 761 (1989). 8. M. Kuzuya and F. Kuzuya, Free Radical Biol. and Med. 14,67 (1993). 9. M. S. Brown and J. L. Goldstein, Annu. Rev. Biochem 52, 223 (1983). 10. B. Kalayanaraman, V. M. Darley-Usmar, J. Wood, J. Joseph, and S. Parthasarathy, J. Biol. Chem. 267, 6789 (1992). 11. D. Steinberg, American Journal of Cardiology 57, 16H (1986). 12. D. Steinberg, S. Parthasarathy, T. E. Carew, J. C. Khoo, and J. L. Witztum, N. En@. J. Med. 320, 915 (1989). 13. S. Urien, P. Riant, E. Albengres, R. Brioude, and J. P. Tillement, Mol. Pharmacol. 26, 322 (1984). 14. L. R. McLean, G. E. Thomas, B. Weintraub, and K. A. Hogaman, J. Biol. Chem. 267, 12291 (1992). 15. L. R. McLean and K. A. Hagaman, Biochemistty 28, 321 (1989). 16. N. Gotoh, K. Shimizu, E. Komuro, J. Tsuchiga, and N. Noguchi, Biochim. Biophys. Acta 1128, 147 (1992). 17. L. R. McLean and K. A. Hagaman, Biochem. Biophys. Acta 959, 201 (1988). 18. G. Sosnovslq Pure and Applied Chem. 62,289 (1990). 19. A. Pezeshk, V. Pezeshk, J. Wojas, and W. K. Subczynski, J. Inotg. B&hem. 46, 67 (1992). 20. A. Pezeshk, V. Pezeshk, A. Firles, J. Wojas, and W. K. Subczynski, Life Sciences 52, 1071 (1993). 21. S. Schreier, W. A. Frezzatti Jr., P. S. Aranjo, H. Chaimovich, and I. M. Cuccovia, Biochim. Biophys. Acta 769, 231 (1984).

EFFECTS OF PROBUCOL ON PHASE TRANSITION AND FLUIDITY

11

22. E. Lissi, M. L. Bianconi, A. T. de Amaral, E. de Paula, L. E. B. Blanch, and S. Schreier, Biochim Biophys. Acru 1021,46 (1990). 23. S. R. Wassall, C. R. Yang, L. Wang, J. M. Pheleps, W. Ehringer, and W. Stillwel, Bul. Mag. Reson. 12, 60 (1990). 24. A. Kusumi, W. K. Subczynski, M. Pasenkiewier-Gierula, J. S. Hyde, and H. Merkle, Biochem. Biophys. Actu 854, 307 (1986). 25. M. Egret-Charlie, A. Sanson, M. Ptak, and 0. Bouloussa, FEBS Lett. 87, 313 (1978). 26. A. Kusumi, W. K. Subczynski, and J. S. Hyde, Fed. Proc. 41, 1394 (1982). 27. H. Trauble and H. Eibl, Proc. N&l. Acad. Sci. USA 71, 214 (1974). 28. A. Kusumi, W. K. Subczynski, and J. S. Hyde, Proc. Natl. Acad. Sci. USA 79, 1854 (1982). 29. J. S. Hyde and W. K. Subczynski in Biological Magnetic Resonance, L. J. Berliner and J. Reuben, Eels., Plenum, New York, 1989, Vol. 8, pp. 399-425. 30. H. M. McConnell, in Spin Labeling. Theory and Applications, L. J. Berliner, Ed., Academic Press, New York, 1976, pp. 535-560. 31. E. J. Shimshick and H. M. McConnell, Biochemisty 12, 2351 (1973). 32. W. K. Subczynski, E. Markowska, W. L. Gruszecki, and J. Sielewiesuk, B&hem. Biophys. Actu. 1105, 97 (1992). 33. B. J. Gaffney, in Spin Labeling, i%eov andApplications, L. J. Berliner, Ed., Academic Press, New York, 1976, pp. 567-571. 34. L. J. Berliner, Methods Enzymol. 49, 466 (1978). 35. A. Milon, G. Wolff, G. Ourisson, and Y. Nakatani, Helu. Chim. Acta 69, 12 (1986). 36. V. P. Skulachev, J. Membr. Biol. 114, 97 (1990). 37. W. K. Subczynski, W. E. Antholine, J. S. Hyde, and A. Kusumi, Biochemistry 29,7936 (1990). 38. S. Srivastava, R. S. Phadke, G. Govid, and C. N. R. Rao, Biochem. Biophys. Acta 734, 353 (1983). 39. L. R. C. Barclay and K. U. Ingold, J. Am. Chem. Sot. 103, 6478 (1981). Received July 19, 1993; accepted September 16, 1993