137
Biochimica et Biophysica Acta, 5 4 2 ( 1 9 7 8 ) 1 3 7 - - 1 5 3 © E l s e v i e r / N o r t h - H o l l a n d Biomedical Press
BBA 28599
COMPARISON OF L A R G E U N I L A M E L L A R VESICLES P R E P A R E D BY A P E T R O L E U M E T H E R V A P O R I Z A T I O N METHOD WITH M U L T I L A M E L L A R VESICLES ESR, D I F F U S I O N AND E N T R A P M E N T ANALYSES H U G H S C H I E R E N *, S T E V E N R U D O L P H , M O R R I S F I N K E L S T E I N , P E T E R C O L E M A N and GERALD WEISSMANN **
Division of Rheumatology of the Department of Medicine, New York University School of Medicine, and the Departments of Biology and Physics, New York University, New York, N.Y. 10016 (U.S.A.) (Received November 28th, 1977)
Summary Large unilamellar vesicles, prepared by a petroleum ether vaporization method, were compared to multilamellar vesicles with respect to a number of physical and functional properties. Rotational correlation time approximations, derived from ESR spectra of both hydrophilic (3-doxyl cholestane) and hydrophobic (3-doxyl androstanol) steroid spin probes, indicated similar molecular packing of lipids in bilayers of multilamellar and large unilamellar liposomes. Light scattering measurements demonstrated a reduction in apparent absorbance of large unilamellar vesicles, suggesting loss of multilamellar structure which was confirmed by electron microscopy. Furthermore, large unilamellar vesicles exhibited enhanced passive diffusion rates of small solutes, releasing a greater percentage of their contents within 90 min than multilamellar vesicles, and reflecting the less restricted diffusion of a unilamellar system. The volume trapping capacity of large unilamellar vesicles far exceeded that of multilamellar liposomes, except in the presence of a trapped protein, soy bean trypsin inhibitor, which reduced the volume of the aqueous compartments of large unilamellar vesicles. Finally, measurement of vesicle diameters from electron micrographs of large unilamellar vesicles showed a vesicle size distribution predominantly in the range of 0.1--0.4 ~m with a mean diameter of 0.21 ~m.
* Submitted in partial fulfillment of the Ph.D. r e q u i r e m e n t s o f the New York U n i v e r s i t y . ** To w h o m reprint r e q u e s t s s h o u l d b e a d d r e s s e d at Division of Rheumatology, New York U n i v e r s i t y M e d i c a l C e n t e r , 550 First A v e n u e , N e w York, N.Y. 10016, U.S.A. Abbreviations: ASL, 17~-hydroxy-4',4'-dimethylspiro (5~-androstane-3,2'-oxazolidin)-3'-yloxyl (3-doxyl androstanol); 3-DC, 4',4'-dimethylspiro(5(~-cholestane-3,2'-oxazolidin)-3'-yloxyl (3-doxyl c h o l e s t a n e ) ; BAPNA, benzoyl-DL-arginine-p-nitroanilide - HCl.
138 Introduction Recently, Deamer and Bangham [ 1] introduced an ether vaporization method for the preparation of large volume liposomes (0.1--0.2 pm in diameter). The liposomes formed by this technique were essentially unilamellar, and were able to effect a substantial volume trapping (14 ± 6 1/mol lipid). Indeed, the large unilamellar vesicles formed with diameters considerably greater than those of small unilamellar vesicles produced by sonication [2], ethanol injection [3] or cholate solubilization [4] and they characteristically achieved a trapping efficiency which surpassed that of either these small unilamellar or multilamellar vesicles. In this paper we report an improved methodology for reproducably forming large unilamellar vesicles and describe some of the physiochemical properties of the resulting liposomes. We further explore the potential of these vesicles for entrapment of ionic (CrO~-) and non-ionic (D-glucose) solutes, as well as a protein, soy bean trypsin inhibitor. Materials and Methods Sources. Egg lecithin (chromatographically pure) and Dulbecco's phosphatebuffered saline (1X; Ca ~÷ and Mg2÷ free) were obtained from Grand Island Biological Co., Grand Island, N.Y. Dicetyl phosphate (>99%) was supplied by K and K Laboratories, Plainview, N.Y. Cholesterol (>99%), potassium chromate, petroleum ether (37.6--55.7°C boiling fraction) and ethyl ether (spectroanalyzed grade) were acquired from Fisher Scientific Co., Fair Lawn, N.J. Phosphatidylcholine ([Me-14C]choline) (>98% in toluene/ethanol (1 : 1, v/v)) and D-[1-3H (N)]glucose (>99%) were purchased from New England Nuclear, Boston, Mass. 4',4'-Dimethylspiro(5a-cholestane-3,2'-oxazolidin)-3'-yloxyl (3doxyl cholestane, 3-DC) (>95%) and 17~-hydroxy-4',4'-dimethylspiro(5aandrostane-3,2'-oxazolidin)-3'-yloxyl (3-doxyl androstanol, ASL) (>95%)were obtained from Sylva, Palo Alto, Calif. Soy bean trypsin inhibitor was procured from Worthington Biochemicals, Freehold, N.J. Benzoyl-DL-arginine-p-nitroanilide • HC1 (BAPNA) was provided by Sigma Chemical Co., St. Louis, Mo. Triton X-100 was acquired from Rohm and Haas, Philadelphia, Pa. The reagent purities stated are those specified by the supplier for each chemical lot. All other compounds were of reagent grade. Sephacryl S-200, Sepharose 2B, Sephadex G-200 and the chromatographic columns were purchased from Pharmacia, Uppsala, Sweden. Dialysis tubing ('A inch) was provided by Arthur H. Thomas Co., Philadelphia, Pa. The syringe drive, Model 355, for large unilamellar vesicles was obtained from Sage Instruments, Cambridge, Mass. The large unilamellar vesicle immersion circulating heater (Model 73) was acquired from Polyscience Corporation, Niles, Ill. The gas-tight syringes were purchased from Hamilton Co., Reno, Nev. Preparation o f large unilamellar vesicles. Large unilamellar vesicles were prepared according to the method of Deamer and Bangham [1] with these modifications. First, petroleum ether was utilized as the solvent rather than ethyl ether. This change was adopted because of the absence of an electron spin resonance (ESR) signal from spin probes within preparations of large unilamel-
139
I
I I
~O~~ 'SER
I
I
1 I
~--II,,,-~
l ~'-"r--~"~'"r~7~W~~AI i'\~R'R............
. . . . COD. . . . .
LOCKING NUT
LARGE UNILAMELLAR LIPOSOME APPARATUS
~
WASHER
:~::T::D~ . . . . . . ~ SHToOuPSCONCK
23 GAUGENEEDLE
STOPCOCK ASSEMBLY
I
~23 GAUGE NEEDLE
LOCKING NUT
~
STOPCOCKSLEEVE
'
19 GAUGE NEEDLE ALIGNMENT
II
PIN
Fig. 1. ( A ) L a r g e u n i l a m e l l a r vesicle a p p a r a t u s : basic c o m p o n e n t s , a n d (B) s p e c i f i c a t i o n s for c o n s t r u c t i o n of t h e t e f l o n s t o p c o c k a s s e m b l y .
lar vesicles extruded from ethyl ether. Conventional multilamellar vesicles generated appropriate signals. These observations suggested a decay of spin probe signal, mediated by peroxides formed upon the evaporation of ethyl ether to dryness [5] at the temperature used to form large unilamellar vesicles (60°C). Therefore, to preclude potential free radical membrane damage and to permit evaluation of ESR spectra, petroleum ether, utilized in the original solvent evaporation method [6], was adapted for this application with the use of a lower boiling fraction (37.6--55.7°C). In addition, petroleum ether may have fortuitously alleviated some solvent retention through its limited water solubility. Second, a stopcock assembly was added to the apparatus described by Deamer and Bangham (Fig. 1A). Developed to obviate several problems, this assembly permitted: (1) temperature equilibration of the aqueous solution while preventing mixing with the lipid phase prior to injection; (2) simulta-
140 neous running of multiple samples even with different lipid volumes; (3) work with hazardous or radioactive lipids and solvents due to the virtual elimination of spillage or evaporation by means of the closure possible at both ends of the system; (4) maneuverability in a minimum work area due to the flexibility of the teflon tubing serving as the syringe-stopcock connector. The stopcock was constructed as a one-piece housing (Fig. 1B). The stopcock plug sits firmly within a teflon sleeve which is inset into the main body. The adaptor section was machined to precisely fit through the cut-off b o t t o m of a Liebig condenser into its internal chamber, thereby preventing leakage. A 19 gauge needle (to minimize flow resistance) brings the lipid solution into the main body from the syringe, and a connecting 23 gauge needle (extending only slightly above the teflon adaptor section to minimize solvent pre-heating) introduces it into the aqueous solution. Anionic large unilamellar liposomes were prepared by solubilizing phosphatidylcholine, dicetyl phosphate and cholesterol in petroleum ether at appropriate molar percentages (70 : 20 : 10) in a final volume of 30 ml ( 2 ~ m o l lipid/ ml). ASL (in chloroform) or 3-DC (in petroleum ether) was included in this 30 ml mixture at a spin probe to lipid molar ratio of 1 : 150; phosphatidyl[14C]choline, when used, was also incorporated in this solution. These lipid phase constituents were aspirated in a gas-tight syringe and were injected with the aid of a mechanical drive at 0.25 ml/min through the stopcock into the aqueous phase, located within the internal chamber of the Liebig condenser. The internal chamber was maintained at 60°C by a circulating water bath in the outer jacket of the condenser. The deadspace of the injection apparatus was "also filled with the lipid solution to avoid volume errors. The aqueous phase consisted of 4.5 ml of phosphate-buffered saline (pH 7.3), used for ESR spectroscopy, or 4.5 ml of 100 mM K2CrO4 (pH 7.3) for entrapment and diffusion analyses. In other studies, the swelling solution consisted of 4.5 ml of phosphate-buffered saline (pH 7.3) containing 10.0 mM D-[~H]glucose (10 pCi/ml) or 10.0 mM D-[3H]glucose together with soy bean trypsin inhibitor (2.0 or 4.0 mg/ml). The 4.5 ml volume of aqueous solution initially added to the condenser was reduced through evaporation by approx. 12% during the 2 h period of lipid delivery at 60°C. Thus, the solute content in the aqueous solution, as well as in the vesicles, increased with time. Consequently, a 6% average evaporation correction, based on the premise of linear evaporation with time, was applied in the entrapment calculation. However, due to the very rapid equilibration of water expected upon the routine passage of large unilamellar vesicles through a column (below), the concentration within large unilamellar vesicles was considered to be isoosmotic and no further correction was made for changes in permeability or stability. At the conclusion of lipid injection, the resultant large unilamellar vesicle preparations were removed from the condenser, vortexed for 1 min, and allowed to come to room temperature. An aliquot of this 4 ml large unilamellar vesicle suspension (15 pmol lipid/ml) was layered on top of a Sephacryl S-200, or in a few cases a Sephadex G-200, column (1.6 X 20 cm), and eluted with phosphate-buffered saline (pH 7.3) at an initial flow rate of 40 or 22 ml/h, respectively. Alternatively, the suspension of large unilamellar vesicles was applied to a Sepharose 2B or Sephadex G-200 column (1.6 X 40 cm), and
141 eluted with phosphate-buffered saline (pH 7.3--7.4) at an initial flow rate of 15 ml/h. Flow rates slowed with time, but good resolution was obtained with these starting rates. The effluents, collected as 2-ml fractions, were assayed for lipid content by relating absorbancy at 520 or 750 nm (with chromate present) to that of standard dilutions of the original preparations, or by quantification of the phosphatidyl[14C]choline content by comparison with the radioactivity in the initial preparation. P r e p a r a t i o n o f m u l t i l a m e l l a r l i p o s o m e s . The preparation of multilamellar liposomes has been elaborated upon previously [7--10]. Briefly, anionic liposomes (phosphatidylcholine/dicetyl phosphate/cholesterol, 70 : 20 : 10 (mol/ mol)) were made by adding 90 pmol of the lipids, dissolved in 6 ml of either chloroform (99%), petroleum ether or ethyl ether to a 100 ml round b o t t o m flask. The hydrophilic (3-DC) or hydrophobic (ASL) spin probe was incorporated with the lipid components at a spin probe to lipid molar ratio of 1 : 150, as for large unilamellar vesicle samples. These mixtures were subjected to rotary evaporation for 30 min resulting in the formation of a uniformly thin film to serve as the lipid phase. The aqueous phase component of 6 ml of phosphate-buffered saline (pH 7.3) alone, used for ESR spectroscopy, or 6 ml of phosphate-buffered saline (pH 7.3--7.4) containing 10.0 mM 3H-labeled D-glucose (10 pCi/ml) or 6 ml of 100 mM K2CrO4 (pH 7.3) for diffusion and entrapment analyses, was added to the flask lined with the lipid film. The lipid layer was dispersed by vortexing, and the resulting multilamellar vesicle suspension (15 g mol lipid/ml) was chromatographed (as for large unilamellar vesicles) after standing for 2 h at room temperature. E l e c t r o n spin r e s o n a n c e . Multilamellar (3.0--4.5 ~mol lipid/ml) or large unilamellar {1.4--3.4 p mol lipid/ml) liposomes in phosphate-buffered saline (pH 7.3) were placed in an aqueous flat cell for analysis. Electron spin resonance (ESR) spectra were acquired of a Varian 4500 series X-band spectrometer with a rectangular TEl02 microwave cavity. Parameters of scans were: microwave power, 10 mW; microwave frequency, 9.5 GHz; field set, 3370 G; scan range, 100 G; scan time, 5 min; time constant, 1 s; modulation amplitude, I G; receiver gain, 3.2 • 103. The resulting o u t p u t signal from each of five successive scans was directed through a h o m e m a d e differential amplifier, a Vidar 240 voltage to frequency converter and a Hewlett-Packard 450 A wideband amplifier before it was stored and summed in a Hewlett-Packard series 5400 A multichannel analyzer. The data was punched out on a paper tape by a teletype terminal and input to a Hewlett-Packard 3000 computer. The data reduction program was designed to find the visual peaks of the spectrum lines by fitting the peaks to a quadratic expression (second-order polynomial) with a least square routine. The empirical rotational correlation time approximation (To), used as an indicator of probe mobility, was computed from these peak amplitudes and widths by the formula described previously [11,12]: To = g W o [ ( h o / h - 1 ) 1/' - - 1]
where Wo is the peak-to-peak mid-field line width in gauss, ho is the peak-topeak mid-field line height, h_l is the peak-to-peak high-field line height. The K value for the 3-DC probe was calculated as 6.5 • 10 -~° s [12], and obtained
142 from spin label crystal parameters [13]; the same K value was used for the ASL p ro b e which is also an oxazolidine nitroxide. This constant was u n c o m p e n s a t e d for variations in m e m b r a n e polarity, or for any deviation from isotropic rotation and Lorentzian lineshape for which the To equation is valid. Thus, r0 is an empirical a p p r o x i m a t i o n which yielded relative, although n o t absolute values of the rotational correlation time, while allowing useful comparisons of probe mobility in liposomes. Decreases in To represent enhanced mobility of the probe. Duplicate sets of five scans were averaged for each To value in an experiment. Diffusion of trapped solutes. The e x t e n t of c h r o m a t e anion or D-glucose release f r o m liposomes was measured by diluting 0.8- or 0.9-ml aliquots of large unilamellar (1--3 p mol lipid/ml) or multilamellar (2--6 g mol lipid/ml) vesicles to 1.0 ml with phosphate-buffered saline (pH 7.3--7.4). As a reference for total a m o u n t o f sequestered marker, replicate aliquots were diluted to 1.0 ml with T rito n X-100 so as to yield a 0.2% (v/v) final concentration. The samples (1.0 ml) were dialyzed in 0.25-inch tubing against 5 ml of phosphate-buffered saline (pH 7.3--7.4) at 37°C in a water bath. These sacs were progressively moved to fresh 5-ml solutions at 30-min intervals for an elapsed time of 90 rain; in some experiments an extra 24 h interval was included. Chromate anion release was measured by its absorbance at 370 nm directly from the dialysate at each interval. Analogously, for 3H-labeled D-glucose, samples (400 pl) of the dialysate were placed in vials with the subsequent addition of 4 ml of a scintillation fluid (consisting of 1 g p - b i s [ o - m e t h y l s t y r y l ] b e n z e n e and 7 g 2,5-dip h e n y l o x a z o l e dissolved in 340 ml Triton X-100 and 660 ml xylene), and were analyzed in a Beckman liquid scintillation counter. Percentage release was computed as the cumulative " l e a k " after 30, 60 or 90 min divided by the total Triton X-100-induced release at the end of 24 h (×100). In experiments witho u t 24 h interval readings, the 90-min totals were extrapolated to the percentage of the total release (Triton X-100) after 24 h by division with the ratio of the 90 m i n / 2 4 h leak from T r i t on X-100 samples. This ratio was found to be e x t r e m e l y constant, being 0.854 _+0.025 (n = 11) for glucose and 0.924 _+ 0.005 (n = 14) for ch rom a t e anion. Volume trapping. E n t r a p m e n t was determined by either the solute diffusion or the column chromatographic m e t h o d . In the solute diffusion m e t h o d , the a m o u n t of entrapped glucose or c h r o m a t e anion per pmol lipid was determined by incubating chromatographically resolved liposomes in the presence of 0.2% (v/v) T rito n X-100 and measuring the total marker released into the dialysate at 90 min. E n t r a p m e n t was then calculated from the equation: E = [(S -- b) a-lM] VD-IR-1L-I where the bracketed terms represent the measured a m o u n t of solute and the non-bracketed terms are corrections for dilution (V and D), solute behavior (R) and liposome sample size (L); E = the entrapped solute ( m o l / p m o l lipid), S = the total apparent absorbance (A) of c h r o m a t e anion or the cpm of D-[3H] glucose measured at 90 min f or a given volume of dialysate, b = the y-intercept of the c h r o m a t e absorbanee standard curve ( A / p m o l ) and is equal to zero when glucose is the marker, a = the slope of the c h r o m a t e standard curve ( A / p m o l ) or the D-[3H]glucose cpm (corrected for blank, spillover and evaporation) for
143 a given volume of the original, unfractionated liposome preparation, M = the product of the glucose concentration (mol/1) in the original swell solution and the volume (1) of this solution from which a was measured and is equal to one for the chromate marker, V = the volume of the buffer into which the dialysis sac is placed (5 ml in these experiments), D = the sampled volume (ml) of dialysate (as for S) analyzed for glucose cpm (0.4 ml in these experiments) or the dilution used to read the chromate absorbance, R = the ratio of 90 rain/24 h Triton X-100 marker release (0.924 for chromate and 0.854 for glucose), and L = the lipid concentration in the dialysis sac (pmol lipid/ml). In the chromatographic method, regardless of the sequestered species, the a m o u n t of aqueous c o m p a r t m e n t marker was related to the a m o u n t of lipid in the liposome peak. With liposomes containing radioactive glucose, a portion of each chromatographic fraction was evaluated for the presence of phosphatidyl[14C]choline and all-labeled D-glucose. Samples (diluted to 400 pl) were added to vials containing 4 ml scintillation fluid, as described earlier, and analyzed with a Beckman liquid scintillation counter set for two channel operation (partial 3H and '4C windows). All data was corrected for channel spillover. In liposomes with entrapped chromate, aliquots of each column fraction were assayed for chromate c o n t e n t at 370 nm, and for phospholipid levels at 750 nm. An appropriate adjustment was made for phospholipid absorption at 370 nm. Antiprotease assay. Soy bean trypsin inhibitor was assayed by inhibition of the tryptic hydrolysis of a synthetic chromogenic substrate [14]. Briefly, a 100 pl sample was incubated with an equal a m o u n t of 0.4% Triton X-100 (v/v) for 30 min, thus disrupting the liposomal integrity. Then 100 gl 0.03% trypsin (bovine pancreas) (w/v) in 0.0025 M HC1 was added along with 1200 pl 0.1 M Tris buffer containing 0.02 M CaC12 (pH 8.2). Following 10 min of preincubation at 25°C, 1500 t~l of the BAPNA substrate (435 mg/1) was added to initiate hydrolysis. Reactions were halted after 10 min through the addition of 30% (v/v) acetic acid. The apparent absorbances of the samples at 410 nm were determined with a Beckman Model 25 Spectrophotometer with a sipper microcell. The samples were diluted appropriately so as not to exceed 70% inhibition. Assays and blanks (minus trypsin) were performed in duplicate. Statistical analysis. Significance between data was assessed by the Student's t-test and the P values were obtained from the appropriate tables. Results and Discussion Electron spin resonance studies The cholesterol surrogates, 3-DC and ASL, employed as motional monitors of the hydrophilic and hydrophobic zones of the bilayers, respectively, were incorporated into the lipid phase of large unilamellar or multilamellar vesicles in order to compare the molecular arrangement of the phospholipids within these liposomes. Both 3-DC (Fig. 2) and ASL (Fig. 3) steroid spin probes exhibited spectra characteristic of anisotropic motion with rapid rotation around the y principal axis whether in large unilamellar vesicle or multilamellar vesicle membranes. This motional pattern is characteristic of bilayers whose structural geometry aligns the probes parallel to the phospholipid acyl chains and orients
144
5L~_OC,MLV vs LUV o
A
A B
20 GAUSS
H
I
D
I
F i g . 2. E l e c t r o n s p i n r e s o n a n c e s p e c t r a f r o m t h e h y d r o p h i l i c s p i n . p r o b e , 3 - D C , w i t h i n ( A ) v e s i c l e s (W 0 = 3 . 0 , h 0 / h _ l = 7 . 8 , T O = 3 . 5 • 1 0 - 9 s), a n d ( B ) l a r g e u n i l a m e l l a r v e s i c l e s (W 0 = 7 . 2 , T O = 3 . 4 • 1 0 - 9 s). E a c h s p e c t r u m is t h e c o m p o s i t e o f f i v e c o m p u t e r - a v e r a g e d s c a n s , a n d t o t h e s a m e m i d - f i e l d l i n e h e i g h t . D e t e r m i n a t i o n s o f l i n e w i d t h s o r h e i g h t s , a n d TO, as w e l l as p a r a m e t e r s u t i l i z e d , are as d e s c r i b e d i n M a t e r i a l s a n d M e t h o d s .
ASL, MLV
vs
multilamellar 3.1, h0/h_ 1 = is n o r m a l i z e d the ESR scan
LUV
o
A B
20
H im
F
GAUSS
I
F i g . 3. E l e c t r o n s p i n r e s o n a n c e s p e c t r a f r o m t h e h y d r o p b o b i c s p i n p r o b e , A S L , w i t h i n b i l a y e r s o f ( A ) m u l t i l a m e l l a r v e s i c l e s W 0 = 2 . 9 , h o / h _ l = 5 . 0 , T O = 2 . 3 • 1 0 - 9 s), a n d (B) l a r g e u n i l a m e l l a r v e s i c l e s (W 0 = 2 . 8 , h 0 / h _ 1 = 4 . B , r 0 = 2 . 2 . 1 0 - 9 s). S p e c t r a are t r e a t e d a s d e s c r i b e d i n F i g . 2.
145
TABLE
I
ROTATIONAL CORRELATION TIME APPROXIMATIONS (TO) -+ S.D. F O R PHILIC) A N D A S L ( H Y D R O P H O B I C ) SPIN PROBES IN L A R G E UNILAMELLAR ELLAR LIPOSOMES
THE 3-DC (HYDROAND MULTILAM-
L i p o s o m e composition in all experiments is phosphatidylcholinc/dicetyl phosphate/cholesterol (70 : 20 : 10). Spin probe
Liposome structure
n
TO X 1 0 9 S
3-DC 3-DC
Large unilarnellar Multilamellar
8 8
3.3 _+ 0 . 2 3 . 4 _+ 0 . 2
ASL ASL
Large unilamellar Multilamellar
4 4
2,1 + 0 . 1 2 . 2 _+ 0 . 1
them perpendicular to the membrane surface [15,16]. In addition, the rotational correlation time approximations, calculated as To + S.D. (X109 s), for 3-DC were 3.3 +_0.2 (large unilamellar vesicles) and 3.4 +_0.2 (multilamellar vesicles), and for ASL were 2.1 ± 0.1 (large unilamellar vesicles) and 2.2 ± 0.1 (multilamellar vesicles) (Table I). Thus, the observed increase in probe mobility of the hydrophobic (ASL) in relation to the hydrophilic (3-DC) spin probe indicates the existence of the predicted flexibility gradient [ 17,18] within both large unilamellar ~nd multilamellar liposomal bilayers. Furthermore, the nearly identical r0 values for both the 3-DC and ASL probes in large unilamellar and multilamellar liposomes demonstrate the similarity of the molecular packing within these t w o bilayer systems.
Light scattering studies The turbidity of large unilamellar vesicle and multilamellar vesicle systems was compared to confirm the absence of a multilamellar contaminant in the large unilamellar vesicle preparations. The slopes +_S.E. of absorbance as a function of liposome concentration (pmol lipid/ml) were 0.281 + 0.010 (n = 12) and 0.280 + 0.005 (n = 26) for large unilamellar vesicles at 750 and 520 nm, respectively, and 0.852 + 0.086 (n = 4) and 0.858 + 0.043 (n = 12) for multilamellar vesicles at 750 and 520 nm, respectively (Fig. 4). Thus, the light scattering of large unilamellar vesicles, in contrast to multilamellar vesicles, was markedly reduced at both 750 and 520 nm. Since the intralamellar order of the large unilamellar vesicle and multilamellar vesicle systems was shown to be similar by ESR, the diminished absorbance observed in the large unilamellar vesicle system is most likely a consequence of a decreased contribution by the interlamellar stacking (form) c o m p o n e n t of the birefringence. These data are concordant with the substantial elimination of multilamellar vesicle structures in the large unilamellar vesicle preparations and are further supported by electron photomicrographs which indicated the presence of primarily unilamellar vesicles (see below).
Diffusion studies of trapped solutes The permeability properties of large unilamellar and multilamellar liposomal membranes were analyzed with respect to their passive diffusion rates, calcu-
146 I
1
I
0.25
0.50
0.6
E
tO
W) I"--
0.4
0.2
0.6
E
0.4
O~
0.2
0.75
(/Jmol/ml)
Lipid Concentration
Fig. 4. A p p a r e n t a b s o r b a n c e s of m~tltilamellar a n d large u n i l a m e l l a r vesicle s u s p e n s i o n s as a f u n c t i o n of lipid c o n c e n t r a t i o n . A, at 7 5 0 n m , a n d B, at 5 2 0 n m .
25
I
I
I
""
@
I
60
..5
c7 20 b v
I I
50
t J
llli /
J
i1~1I
/ ///
40
15
30
S
20r
i
.,~
i i /! i
6~//'
o~ _ _ L _ _
Omi•
I
30min 60mio Time
I
I
90min
24hr
0rain
30rnin. 60rain Time
90min
24hr
Fig. 5. R a t e of solute release f r o m large u n i l a m e U a r vesicles a n d m u l t i l a m e l l a r vesicles w i t h (A) c h r o m a t e a n i o n , or (B) D-glucose. A l i q u o t s of l i p o s o m e s c o n t a i n i n g t h e s e s o l u t e s w e r e d i a l y z e d a g a i n s t p h o s p h a t e b u f f e r e d saline ( p H 7.3) f o r 9 0 rain, a n d in s o m e e x p e r i m e n t s f o r 2 4 h , a t 37°C. T h e s o l u t e release is e x p r e s s e d as t h e p e r c e n t ( m e a n ± S.E.) of t h e t o t a l T r i t o n X - 1 0 0 - i n d u e e d l e a k a g e ( 2 4 h), a n d is p l o t t e d as a f u n c t i o n of t i m e . T h e slopes o f release o v e r 90 m i n (% r e l e a s e / r a i n ) w e r e c a l c u l a t e d t h r o u g h linear r e g r e s s i o n analysis b y t h e m e t h o d o f least s q u a r e s .
147 lated b y linear regression analysis over a linear 90 min interval and expressed as the p e r c e n t ( m e a n + S.E.) o f t h e 24 h T r i t o n X - 1 0 0 release per min o f ionic ( c h r o m a t e anion) and n o n - i o n i c (glucose) solutes. Release o f c h r o m a t e anion f r o m large unilamellar vesicles ( 0 . 1 0 7 ± 0 . 0 0 5 , n = 8) was c o n s i d e r a b l y greater t h a n t h a t f r o m multilamellar vesicles ( 0 . 0 3 4 ± 0 . 0 0 7 , n = 5) P = < 0 . 0 0 1 (Fig. 5A). Similarly, the d i f f u s i o n o f n o n - i o n i c glucose m o l e c u l e s f r o m large unilamellar vesicles ( 0 . 2 0 0 _+ 0 . 0 1 4 , n = 6) e x c e e d e d t h a t f r o m multilamellar vesicles ( 0 . 1 2 4 + 0 . 0 0 7 , n = 7) P = < 0 . 0 0 1 (Fig. 5B). T h e s e large unilamellar vesicle flux d a t a are c o n s i s t e n t w i t h the p r e s e n c e o f a single-walled s t r u c t u r e w h o s e internalized a q u e o u s solutes e n c o u n t e r f e w e r diffusion barriers t h a n in m u l t i - l a y e r e d multilamellar vesicles. F u r t h e r m o r e , the ratio {mean + S.E.) o f the p e r c e n t m a r k e r release at 90 min to 24 h is larger f o r the large unilamellar vesicle system t h a n for t h e multilamellar vesicle s y s t e m with e i t h e r c h r o m a t e (large unilamellar vesicles = 0.78 _+ 0.01, n = 8; multilamellar vesicles ~ 0.37 _+ 0.07, n = 5; P = < 0 . 0 0 1 ) o r glucose (large unilamellar vesicles = 0.41 +_0.03, n = 5; multilamellar vesicles = 0.21 + 0.02, n = 3; P = ( 0 . 0 1 ) . Thus, in c o n t r a s t to m u l t i l a m e l l a r liposomes, a greater f r a c t i o n o f t h e sequestered c o n s t i t u e n t s released f r o m large unilamellar vesicles in 24 h pass t h r o u g h the single-bilayer barrier w i t h i n the first 9 0 min. This is in a c c o r d with the m o r e rapid equilibrat i o n e x p e c t e d f o r a unilamellar s t r u c t u r e . Also, the d a t a are c o n s i s t e n t with a m e m b r a n e selectivity in response to ionic and non-ionic solutes, b u t n o t with an intrinsic dissimilarity b e t w e e n m e m branes o f multilamellar and large unilamellar vesicles. Whereas significant differences were f o u n d b e t w e e n the p e r c e n t release o f c h r o m a t e anion f r o m large unilamellar vesicles at 24 h (22.6 + 0.9, n = 8) and multilamellar vesicles (14.0 _+ 2.6, n = 5) P = ( 0 . 0 0 5 , n o d i f f e r e n c e s w e r e f o u n d w h e n glucose was the e n t r a p p e d species (large unilamellar vesicles = 58.9 + 1.4, n = 5; multilamellar vesicles = 5 9 . 0 ± 0.4, n = 3). A l t h o u g h these d a t a c o u l d be i n t e r p r e t e d as a f u n c t i o n o f the stability o f t h e t w o l i p o s o m a l t y p e s , it is m o r e likely a reflect i o n o f the greater rate o f glucose flux in each system. In this case, the glucose d a t a w o u l d suggest equivalence o f multilamellar vesicle and large unilamellar vesicle m e m a b r a n e s which u l t i m a t e l y release c o m p a r a b l e levels o f solute. T h e anion d a t a w o u l d be in a g r e e m e n t w i t h this i n t e r p r e t a t i o n , assuming t h a t sufficient time had n o t y e t elapsed at 24 h f o r t h e slower diffusing c h r o m a t e (in multilamellar vesicles) t o r e a c h equilibrium. In initial e x p e r i m e n t s , w h e n t h e large unilamellar vesicles were p r e p a r e d with the lipids injected via an e t h y l e t h e r r a t h e r t h a n a p e t r o l e u m e t h e r solvent [ 1 ] , the large unilamellar vesicle m e m b r a n e s d i s p l a y e d d i f f e r e n c e s in t h e i r permeability properties. While the 90 min rate o f CrO~- " l e a k " f r o m e t h y l e t h e r multilamellar vesicles ( 0 . 0 2 3 + 0 . 0 0 4 , n = 2) vs. p e t r o l e u m e t h e r multilamellar vesicles ( 0 . 0 3 4 +_ 0 . 0 0 7 , n = 5) was n o t significantly d i f f e r e n t , t h e r e was a sign i f i c a n t decrease in the rate o f CrO~- release in e t h y l e t h e r large unilamellar vesicles ( 0 . 0 6 0 ± 0 . 0 0 7 , n = 2) vs. p e t r o l e u m e t h e r large unilamellar vesicles ( 0 . 1 0 7 ± 0.005, n = 8) P = < 0 . 0 0 5 . Since r e d u c t i o n in p e r m e a b i l i t y has b e e n a t t r i b u t e d to decrease in t h e radius o f c u r v a t u r e f o r small unilamellar vesicles [ 6 , 1 9 ] , these o b s e r v a t i o n s m a y be e x p l a i n e d analagously if t h e large unilamellar vesicle p o p u l a t i o n has a smaller m e a n d i a m e t e r ( f r o m smaller u n i f o r m vesicles or f e w e r large vesicles) w h e n s y n t h e s i z e d w i t h an e t h y l e t h e r , as o p p o s e d
148
to a petroleum ether solvent. This could be a function of differences in solubility, dispersal and vaporization characteristics between these two solvents. Moreover, release of chromate anion over the 90-min to 24-h interval was significantly greater when either multilamellar vesicles or large unilamellar vesicles were prepared using ethyl ether, rather than petroleum ether. Comparison of the ratios of the 24-h/90-min percent chromate release (mean + S.E.) from petroleum ether large unilamellar vesicles (1.29 ± 0.02, n = 8) and ethyl ether
70 60
100
,, t,~ 40
7~
i" 2
2~
t
k'
i
20
'*
1o
"'
0
O-Glucose
~b
6,000 °3
; ,
"4,000 ' 2-
5,000
100 75
,: ;
5 C'
,, '~
25
E
0
,7 !J
o
I~
EL
o
99
5C
o_
o~
~-~
~
3,000
#' a.'%~"~ ' ~,, %,.~ ~ ~,' , ~ , - - ~ , " -
Biochimica et Biophysica Acta, 5 4 2 ( 1 9 7 8 ) 1 3 7 - - 1 5 3 © E l s e v i e r / N o r t h - H o l l a n d Biomedical Press
BBA 28599
COMPARISON OF L A R G E U N I L A M E L L A R VESICLES P R E P A R E D BY A P E T R O L E U M E T H E R V A P O R I Z A T I O N METHOD WITH M U L T I L A M E L L A R VESICLES ESR, D I F F U S I O N AND E N T R A P M E N T ANALYSES H U G H S C H I E R E N *, S T E V E N R U D O L P H , M O R R I S F I N K E L S T E I N , P E T E R C O L E M A N and GERALD WEISSMANN **
Division of Rheumatology of the Department of Medicine, New York University School of Medicine, and the Departments of Biology and Physics, New York University, New York, N.Y. 10016 (U.S.A.) (Received November 28th, 1977)
Summary Large unilamellar vesicles, prepared by a petroleum ether vaporization method, were compared to multilamellar vesicles with respect to a number of physical and functional properties. Rotational correlation time approximations, derived from ESR spectra of both hydrophilic (3-doxyl cholestane) and hydrophobic (3-doxyl androstanol) steroid spin probes, indicated similar molecular packing of lipids in bilayers of multilamellar and large unilamellar liposomes. Light scattering measurements demonstrated a reduction in apparent absorbance of large unilamellar vesicles, suggesting loss of multilamellar structure which was confirmed by electron microscopy. Furthermore, large unilamellar vesicles exhibited enhanced passive diffusion rates of small solutes, releasing a greater percentage of their contents within 90 min than multilamellar vesicles, and reflecting the less restricted diffusion of a unilamellar system. The volume trapping capacity of large unilamellar vesicles far exceeded that of multilamellar liposomes, except in the presence of a trapped protein, soy bean trypsin inhibitor, which reduced the volume of the aqueous compartments of large unilamellar vesicles. Finally, measurement of vesicle diameters from electron micrographs of large unilamellar vesicles showed a vesicle size distribution predominantly in the range of 0.1--0.4 ~m with a mean diameter of 0.21 ~m.
* Submitted in partial fulfillment of the Ph.D. r e q u i r e m e n t s o f the New York U n i v e r s i t y . ** To w h o m reprint r e q u e s t s s h o u l d b e a d d r e s s e d at Division of Rheumatology, New York U n i v e r s i t y M e d i c a l C e n t e r , 550 First A v e n u e , N e w York, N.Y. 10016, U.S.A. Abbreviations: ASL, 17~-hydroxy-4',4'-dimethylspiro (5~-androstane-3,2'-oxazolidin)-3'-yloxyl (3-doxyl androstanol); 3-DC, 4',4'-dimethylspiro(5(~-cholestane-3,2'-oxazolidin)-3'-yloxyl (3-doxyl c h o l e s t a n e ) ; BAPNA, benzoyl-DL-arginine-p-nitroanilide - HCl.
138 Introduction Recently, Deamer and Bangham [ 1] introduced an ether vaporization method for the preparation of large volume liposomes (0.1--0.2 pm in diameter). The liposomes formed by this technique were essentially unilamellar, and were able to effect a substantial volume trapping (14 ± 6 1/mol lipid). Indeed, the large unilamellar vesicles formed with diameters considerably greater than those of small unilamellar vesicles produced by sonication [2], ethanol injection [3] or cholate solubilization [4] and they characteristically achieved a trapping efficiency which surpassed that of either these small unilamellar or multilamellar vesicles. In this paper we report an improved methodology for reproducably forming large unilamellar vesicles and describe some of the physiochemical properties of the resulting liposomes. We further explore the potential of these vesicles for entrapment of ionic (CrO~-) and non-ionic (D-glucose) solutes, as well as a protein, soy bean trypsin inhibitor. Materials and Methods Sources. Egg lecithin (chromatographically pure) and Dulbecco's phosphatebuffered saline (1X; Ca ~÷ and Mg2÷ free) were obtained from Grand Island Biological Co., Grand Island, N.Y. Dicetyl phosphate (>99%) was supplied by K and K Laboratories, Plainview, N.Y. Cholesterol (>99%), potassium chromate, petroleum ether (37.6--55.7°C boiling fraction) and ethyl ether (spectroanalyzed grade) were acquired from Fisher Scientific Co., Fair Lawn, N.J. Phosphatidylcholine ([Me-14C]choline) (>98% in toluene/ethanol (1 : 1, v/v)) and D-[1-3H (N)]glucose (>99%) were purchased from New England Nuclear, Boston, Mass. 4',4'-Dimethylspiro(5a-cholestane-3,2'-oxazolidin)-3'-yloxyl (3doxyl cholestane, 3-DC) (>95%) and 17~-hydroxy-4',4'-dimethylspiro(5aandrostane-3,2'-oxazolidin)-3'-yloxyl (3-doxyl androstanol, ASL) (>95%)were obtained from Sylva, Palo Alto, Calif. Soy bean trypsin inhibitor was procured from Worthington Biochemicals, Freehold, N.J. Benzoyl-DL-arginine-p-nitroanilide • HC1 (BAPNA) was provided by Sigma Chemical Co., St. Louis, Mo. Triton X-100 was acquired from Rohm and Haas, Philadelphia, Pa. The reagent purities stated are those specified by the supplier for each chemical lot. All other compounds were of reagent grade. Sephacryl S-200, Sepharose 2B, Sephadex G-200 and the chromatographic columns were purchased from Pharmacia, Uppsala, Sweden. Dialysis tubing ('A inch) was provided by Arthur H. Thomas Co., Philadelphia, Pa. The syringe drive, Model 355, for large unilamellar vesicles was obtained from Sage Instruments, Cambridge, Mass. The large unilamellar vesicle immersion circulating heater (Model 73) was acquired from Polyscience Corporation, Niles, Ill. The gas-tight syringes were purchased from Hamilton Co., Reno, Nev. Preparation o f large unilamellar vesicles. Large unilamellar vesicles were prepared according to the method of Deamer and Bangham [1] with these modifications. First, petroleum ether was utilized as the solvent rather than ethyl ether. This change was adopted because of the absence of an electron spin resonance (ESR) signal from spin probes within preparations of large unilamel-
139
I
I I
~O~~ 'SER
I
I
1 I
~--II,,,-~
l ~'-"r--~"~'"r~7~W~~AI i'\~R'R............
. . . . COD. . . . .
LOCKING NUT
LARGE UNILAMELLAR LIPOSOME APPARATUS
~
WASHER
:~::T::D~ . . . . . . ~ SHToOuPSCONCK
23 GAUGENEEDLE
STOPCOCK ASSEMBLY
I
~23 GAUGE NEEDLE
LOCKING NUT
~
STOPCOCKSLEEVE
'
19 GAUGE NEEDLE ALIGNMENT
II
PIN
Fig. 1. ( A ) L a r g e u n i l a m e l l a r vesicle a p p a r a t u s : basic c o m p o n e n t s , a n d (B) s p e c i f i c a t i o n s for c o n s t r u c t i o n of t h e t e f l o n s t o p c o c k a s s e m b l y .
lar vesicles extruded from ethyl ether. Conventional multilamellar vesicles generated appropriate signals. These observations suggested a decay of spin probe signal, mediated by peroxides formed upon the evaporation of ethyl ether to dryness [5] at the temperature used to form large unilamellar vesicles (60°C). Therefore, to preclude potential free radical membrane damage and to permit evaluation of ESR spectra, petroleum ether, utilized in the original solvent evaporation method [6], was adapted for this application with the use of a lower boiling fraction (37.6--55.7°C). In addition, petroleum ether may have fortuitously alleviated some solvent retention through its limited water solubility. Second, a stopcock assembly was added to the apparatus described by Deamer and Bangham (Fig. 1A). Developed to obviate several problems, this assembly permitted: (1) temperature equilibration of the aqueous solution while preventing mixing with the lipid phase prior to injection; (2) simulta-
140 neous running of multiple samples even with different lipid volumes; (3) work with hazardous or radioactive lipids and solvents due to the virtual elimination of spillage or evaporation by means of the closure possible at both ends of the system; (4) maneuverability in a minimum work area due to the flexibility of the teflon tubing serving as the syringe-stopcock connector. The stopcock was constructed as a one-piece housing (Fig. 1B). The stopcock plug sits firmly within a teflon sleeve which is inset into the main body. The adaptor section was machined to precisely fit through the cut-off b o t t o m of a Liebig condenser into its internal chamber, thereby preventing leakage. A 19 gauge needle (to minimize flow resistance) brings the lipid solution into the main body from the syringe, and a connecting 23 gauge needle (extending only slightly above the teflon adaptor section to minimize solvent pre-heating) introduces it into the aqueous solution. Anionic large unilamellar liposomes were prepared by solubilizing phosphatidylcholine, dicetyl phosphate and cholesterol in petroleum ether at appropriate molar percentages (70 : 20 : 10) in a final volume of 30 ml ( 2 ~ m o l lipid/ ml). ASL (in chloroform) or 3-DC (in petroleum ether) was included in this 30 ml mixture at a spin probe to lipid molar ratio of 1 : 150; phosphatidyl[14C]choline, when used, was also incorporated in this solution. These lipid phase constituents were aspirated in a gas-tight syringe and were injected with the aid of a mechanical drive at 0.25 ml/min through the stopcock into the aqueous phase, located within the internal chamber of the Liebig condenser. The internal chamber was maintained at 60°C by a circulating water bath in the outer jacket of the condenser. The deadspace of the injection apparatus was "also filled with the lipid solution to avoid volume errors. The aqueous phase consisted of 4.5 ml of phosphate-buffered saline (pH 7.3), used for ESR spectroscopy, or 4.5 ml of 100 mM K2CrO4 (pH 7.3) for entrapment and diffusion analyses. In other studies, the swelling solution consisted of 4.5 ml of phosphate-buffered saline (pH 7.3) containing 10.0 mM D-[~H]glucose (10 pCi/ml) or 10.0 mM D-[3H]glucose together with soy bean trypsin inhibitor (2.0 or 4.0 mg/ml). The 4.5 ml volume of aqueous solution initially added to the condenser was reduced through evaporation by approx. 12% during the 2 h period of lipid delivery at 60°C. Thus, the solute content in the aqueous solution, as well as in the vesicles, increased with time. Consequently, a 6% average evaporation correction, based on the premise of linear evaporation with time, was applied in the entrapment calculation. However, due to the very rapid equilibration of water expected upon the routine passage of large unilamellar vesicles through a column (below), the concentration within large unilamellar vesicles was considered to be isoosmotic and no further correction was made for changes in permeability or stability. At the conclusion of lipid injection, the resultant large unilamellar vesicle preparations were removed from the condenser, vortexed for 1 min, and allowed to come to room temperature. An aliquot of this 4 ml large unilamellar vesicle suspension (15 pmol lipid/ml) was layered on top of a Sephacryl S-200, or in a few cases a Sephadex G-200, column (1.6 X 20 cm), and eluted with phosphate-buffered saline (pH 7.3) at an initial flow rate of 40 or 22 ml/h, respectively. Alternatively, the suspension of large unilamellar vesicles was applied to a Sepharose 2B or Sephadex G-200 column (1.6 X 40 cm), and
141 eluted with phosphate-buffered saline (pH 7.3--7.4) at an initial flow rate of 15 ml/h. Flow rates slowed with time, but good resolution was obtained with these starting rates. The effluents, collected as 2-ml fractions, were assayed for lipid content by relating absorbancy at 520 or 750 nm (with chromate present) to that of standard dilutions of the original preparations, or by quantification of the phosphatidyl[14C]choline content by comparison with the radioactivity in the initial preparation. P r e p a r a t i o n o f m u l t i l a m e l l a r l i p o s o m e s . The preparation of multilamellar liposomes has been elaborated upon previously [7--10]. Briefly, anionic liposomes (phosphatidylcholine/dicetyl phosphate/cholesterol, 70 : 20 : 10 (mol/ mol)) were made by adding 90 pmol of the lipids, dissolved in 6 ml of either chloroform (99%), petroleum ether or ethyl ether to a 100 ml round b o t t o m flask. The hydrophilic (3-DC) or hydrophobic (ASL) spin probe was incorporated with the lipid components at a spin probe to lipid molar ratio of 1 : 150, as for large unilamellar vesicle samples. These mixtures were subjected to rotary evaporation for 30 min resulting in the formation of a uniformly thin film to serve as the lipid phase. The aqueous phase component of 6 ml of phosphate-buffered saline (pH 7.3) alone, used for ESR spectroscopy, or 6 ml of phosphate-buffered saline (pH 7.3--7.4) containing 10.0 mM 3H-labeled D-glucose (10 pCi/ml) or 6 ml of 100 mM K2CrO4 (pH 7.3) for diffusion and entrapment analyses, was added to the flask lined with the lipid film. The lipid layer was dispersed by vortexing, and the resulting multilamellar vesicle suspension (15 g mol lipid/ml) was chromatographed (as for large unilamellar vesicles) after standing for 2 h at room temperature. E l e c t r o n spin r e s o n a n c e . Multilamellar (3.0--4.5 ~mol lipid/ml) or large unilamellar {1.4--3.4 p mol lipid/ml) liposomes in phosphate-buffered saline (pH 7.3) were placed in an aqueous flat cell for analysis. Electron spin resonance (ESR) spectra were acquired of a Varian 4500 series X-band spectrometer with a rectangular TEl02 microwave cavity. Parameters of scans were: microwave power, 10 mW; microwave frequency, 9.5 GHz; field set, 3370 G; scan range, 100 G; scan time, 5 min; time constant, 1 s; modulation amplitude, I G; receiver gain, 3.2 • 103. The resulting o u t p u t signal from each of five successive scans was directed through a h o m e m a d e differential amplifier, a Vidar 240 voltage to frequency converter and a Hewlett-Packard 450 A wideband amplifier before it was stored and summed in a Hewlett-Packard series 5400 A multichannel analyzer. The data was punched out on a paper tape by a teletype terminal and input to a Hewlett-Packard 3000 computer. The data reduction program was designed to find the visual peaks of the spectrum lines by fitting the peaks to a quadratic expression (second-order polynomial) with a least square routine. The empirical rotational correlation time approximation (To), used as an indicator of probe mobility, was computed from these peak amplitudes and widths by the formula described previously [11,12]: To = g W o [ ( h o / h - 1 ) 1/' - - 1]
where Wo is the peak-to-peak mid-field line width in gauss, ho is the peak-topeak mid-field line height, h_l is the peak-to-peak high-field line height. The K value for the 3-DC probe was calculated as 6.5 • 10 -~° s [12], and obtained
142 from spin label crystal parameters [13]; the same K value was used for the ASL p ro b e which is also an oxazolidine nitroxide. This constant was u n c o m p e n s a t e d for variations in m e m b r a n e polarity, or for any deviation from isotropic rotation and Lorentzian lineshape for which the To equation is valid. Thus, r0 is an empirical a p p r o x i m a t i o n which yielded relative, although n o t absolute values of the rotational correlation time, while allowing useful comparisons of probe mobility in liposomes. Decreases in To represent enhanced mobility of the probe. Duplicate sets of five scans were averaged for each To value in an experiment. Diffusion of trapped solutes. The e x t e n t of c h r o m a t e anion or D-glucose release f r o m liposomes was measured by diluting 0.8- or 0.9-ml aliquots of large unilamellar (1--3 p mol lipid/ml) or multilamellar (2--6 g mol lipid/ml) vesicles to 1.0 ml with phosphate-buffered saline (pH 7.3--7.4). As a reference for total a m o u n t o f sequestered marker, replicate aliquots were diluted to 1.0 ml with T rito n X-100 so as to yield a 0.2% (v/v) final concentration. The samples (1.0 ml) were dialyzed in 0.25-inch tubing against 5 ml of phosphate-buffered saline (pH 7.3--7.4) at 37°C in a water bath. These sacs were progressively moved to fresh 5-ml solutions at 30-min intervals for an elapsed time of 90 rain; in some experiments an extra 24 h interval was included. Chromate anion release was measured by its absorbance at 370 nm directly from the dialysate at each interval. Analogously, for 3H-labeled D-glucose, samples (400 pl) of the dialysate were placed in vials with the subsequent addition of 4 ml of a scintillation fluid (consisting of 1 g p - b i s [ o - m e t h y l s t y r y l ] b e n z e n e and 7 g 2,5-dip h e n y l o x a z o l e dissolved in 340 ml Triton X-100 and 660 ml xylene), and were analyzed in a Beckman liquid scintillation counter. Percentage release was computed as the cumulative " l e a k " after 30, 60 or 90 min divided by the total Triton X-100-induced release at the end of 24 h (×100). In experiments witho u t 24 h interval readings, the 90-min totals were extrapolated to the percentage of the total release (Triton X-100) after 24 h by division with the ratio of the 90 m i n / 2 4 h leak from T r i t on X-100 samples. This ratio was found to be e x t r e m e l y constant, being 0.854 _+0.025 (n = 11) for glucose and 0.924 _+ 0.005 (n = 14) for ch rom a t e anion. Volume trapping. E n t r a p m e n t was determined by either the solute diffusion or the column chromatographic m e t h o d . In the solute diffusion m e t h o d , the a m o u n t of entrapped glucose or c h r o m a t e anion per pmol lipid was determined by incubating chromatographically resolved liposomes in the presence of 0.2% (v/v) T rito n X-100 and measuring the total marker released into the dialysate at 90 min. E n t r a p m e n t was then calculated from the equation: E = [(S -- b) a-lM] VD-IR-1L-I where the bracketed terms represent the measured a m o u n t of solute and the non-bracketed terms are corrections for dilution (V and D), solute behavior (R) and liposome sample size (L); E = the entrapped solute ( m o l / p m o l lipid), S = the total apparent absorbance (A) of c h r o m a t e anion or the cpm of D-[3H] glucose measured at 90 min f or a given volume of dialysate, b = the y-intercept of the c h r o m a t e absorbanee standard curve ( A / p m o l ) and is equal to zero when glucose is the marker, a = the slope of the c h r o m a t e standard curve ( A / p m o l ) or the D-[3H]glucose cpm (corrected for blank, spillover and evaporation) for
143 a given volume of the original, unfractionated liposome preparation, M = the product of the glucose concentration (mol/1) in the original swell solution and the volume (1) of this solution from which a was measured and is equal to one for the chromate marker, V = the volume of the buffer into which the dialysis sac is placed (5 ml in these experiments), D = the sampled volume (ml) of dialysate (as for S) analyzed for glucose cpm (0.4 ml in these experiments) or the dilution used to read the chromate absorbance, R = the ratio of 90 rain/24 h Triton X-100 marker release (0.924 for chromate and 0.854 for glucose), and L = the lipid concentration in the dialysis sac (pmol lipid/ml). In the chromatographic method, regardless of the sequestered species, the a m o u n t of aqueous c o m p a r t m e n t marker was related to the a m o u n t of lipid in the liposome peak. With liposomes containing radioactive glucose, a portion of each chromatographic fraction was evaluated for the presence of phosphatidyl[14C]choline and all-labeled D-glucose. Samples (diluted to 400 pl) were added to vials containing 4 ml scintillation fluid, as described earlier, and analyzed with a Beckman liquid scintillation counter set for two channel operation (partial 3H and '4C windows). All data was corrected for channel spillover. In liposomes with entrapped chromate, aliquots of each column fraction were assayed for chromate c o n t e n t at 370 nm, and for phospholipid levels at 750 nm. An appropriate adjustment was made for phospholipid absorption at 370 nm. Antiprotease assay. Soy bean trypsin inhibitor was assayed by inhibition of the tryptic hydrolysis of a synthetic chromogenic substrate [14]. Briefly, a 100 pl sample was incubated with an equal a m o u n t of 0.4% Triton X-100 (v/v) for 30 min, thus disrupting the liposomal integrity. Then 100 gl 0.03% trypsin (bovine pancreas) (w/v) in 0.0025 M HC1 was added along with 1200 pl 0.1 M Tris buffer containing 0.02 M CaC12 (pH 8.2). Following 10 min of preincubation at 25°C, 1500 t~l of the BAPNA substrate (435 mg/1) was added to initiate hydrolysis. Reactions were halted after 10 min through the addition of 30% (v/v) acetic acid. The apparent absorbances of the samples at 410 nm were determined with a Beckman Model 25 Spectrophotometer with a sipper microcell. The samples were diluted appropriately so as not to exceed 70% inhibition. Assays and blanks (minus trypsin) were performed in duplicate. Statistical analysis. Significance between data was assessed by the Student's t-test and the P values were obtained from the appropriate tables. Results and Discussion Electron spin resonance studies The cholesterol surrogates, 3-DC and ASL, employed as motional monitors of the hydrophilic and hydrophobic zones of the bilayers, respectively, were incorporated into the lipid phase of large unilamellar or multilamellar vesicles in order to compare the molecular arrangement of the phospholipids within these liposomes. Both 3-DC (Fig. 2) and ASL (Fig. 3) steroid spin probes exhibited spectra characteristic of anisotropic motion with rapid rotation around the y principal axis whether in large unilamellar vesicle or multilamellar vesicle membranes. This motional pattern is characteristic of bilayers whose structural geometry aligns the probes parallel to the phospholipid acyl chains and orients
144
5L~_OC,MLV vs LUV o
A
A B
20 GAUSS
H
I
D
I
F i g . 2. E l e c t r o n s p i n r e s o n a n c e s p e c t r a f r o m t h e h y d r o p h i l i c s p i n . p r o b e , 3 - D C , w i t h i n ( A ) v e s i c l e s (W 0 = 3 . 0 , h 0 / h _ l = 7 . 8 , T O = 3 . 5 • 1 0 - 9 s), a n d ( B ) l a r g e u n i l a m e l l a r v e s i c l e s (W 0 = 7 . 2 , T O = 3 . 4 • 1 0 - 9 s). E a c h s p e c t r u m is t h e c o m p o s i t e o f f i v e c o m p u t e r - a v e r a g e d s c a n s , a n d t o t h e s a m e m i d - f i e l d l i n e h e i g h t . D e t e r m i n a t i o n s o f l i n e w i d t h s o r h e i g h t s , a n d TO, as w e l l as p a r a m e t e r s u t i l i z e d , are as d e s c r i b e d i n M a t e r i a l s a n d M e t h o d s .
ASL, MLV
vs
multilamellar 3.1, h0/h_ 1 = is n o r m a l i z e d the ESR scan
LUV
o
A B
20
H im
F
GAUSS
I
F i g . 3. E l e c t r o n s p i n r e s o n a n c e s p e c t r a f r o m t h e h y d r o p b o b i c s p i n p r o b e , A S L , w i t h i n b i l a y e r s o f ( A ) m u l t i l a m e l l a r v e s i c l e s W 0 = 2 . 9 , h o / h _ l = 5 . 0 , T O = 2 . 3 • 1 0 - 9 s), a n d (B) l a r g e u n i l a m e l l a r v e s i c l e s (W 0 = 2 . 8 , h 0 / h _ 1 = 4 . B , r 0 = 2 . 2 . 1 0 - 9 s). S p e c t r a are t r e a t e d a s d e s c r i b e d i n F i g . 2.
145
TABLE
I
ROTATIONAL CORRELATION TIME APPROXIMATIONS (TO) -+ S.D. F O R PHILIC) A N D A S L ( H Y D R O P H O B I C ) SPIN PROBES IN L A R G E UNILAMELLAR ELLAR LIPOSOMES
THE 3-DC (HYDROAND MULTILAM-
L i p o s o m e composition in all experiments is phosphatidylcholinc/dicetyl phosphate/cholesterol (70 : 20 : 10). Spin probe
Liposome structure
n
TO X 1 0 9 S
3-DC 3-DC
Large unilarnellar Multilamellar
8 8
3.3 _+ 0 . 2 3 . 4 _+ 0 . 2
ASL ASL
Large unilamellar Multilamellar
4 4
2,1 + 0 . 1 2 . 2 _+ 0 . 1
them perpendicular to the membrane surface [15,16]. In addition, the rotational correlation time approximations, calculated as To + S.D. (X109 s), for 3-DC were 3.3 +_0.2 (large unilamellar vesicles) and 3.4 +_0.2 (multilamellar vesicles), and for ASL were 2.1 ± 0.1 (large unilamellar vesicles) and 2.2 ± 0.1 (multilamellar vesicles) (Table I). Thus, the observed increase in probe mobility of the hydrophobic (ASL) in relation to the hydrophilic (3-DC) spin probe indicates the existence of the predicted flexibility gradient [ 17,18] within both large unilamellar ~nd multilamellar liposomal bilayers. Furthermore, the nearly identical r0 values for both the 3-DC and ASL probes in large unilamellar and multilamellar liposomes demonstrate the similarity of the molecular packing within these t w o bilayer systems.
Light scattering studies The turbidity of large unilamellar vesicle and multilamellar vesicle systems was compared to confirm the absence of a multilamellar contaminant in the large unilamellar vesicle preparations. The slopes +_S.E. of absorbance as a function of liposome concentration (pmol lipid/ml) were 0.281 + 0.010 (n = 12) and 0.280 + 0.005 (n = 26) for large unilamellar vesicles at 750 and 520 nm, respectively, and 0.852 + 0.086 (n = 4) and 0.858 + 0.043 (n = 12) for multilamellar vesicles at 750 and 520 nm, respectively (Fig. 4). Thus, the light scattering of large unilamellar vesicles, in contrast to multilamellar vesicles, was markedly reduced at both 750 and 520 nm. Since the intralamellar order of the large unilamellar vesicle and multilamellar vesicle systems was shown to be similar by ESR, the diminished absorbance observed in the large unilamellar vesicle system is most likely a consequence of a decreased contribution by the interlamellar stacking (form) c o m p o n e n t of the birefringence. These data are concordant with the substantial elimination of multilamellar vesicle structures in the large unilamellar vesicle preparations and are further supported by electron photomicrographs which indicated the presence of primarily unilamellar vesicles (see below).
Diffusion studies of trapped solutes The permeability properties of large unilamellar and multilamellar liposomal membranes were analyzed with respect to their passive diffusion rates, calcu-
146 I
1
I
0.25
0.50
0.6
E
tO
W) I"--
0.4
0.2
0.6
E
0.4
O~
0.2
0.75
(/Jmol/ml)
Lipid Concentration
Fig. 4. A p p a r e n t a b s o r b a n c e s of m~tltilamellar a n d large u n i l a m e l l a r vesicle s u s p e n s i o n s as a f u n c t i o n of lipid c o n c e n t r a t i o n . A, at 7 5 0 n m , a n d B, at 5 2 0 n m .
25
I
I
I
""
@
I
60
..5
c7 20 b v
I I
50
t J
llli /
J
i1~1I
/ ///
40
15
30
S
20r
i
.,~
i i /! i
6~//'
o~ _ _ L _ _
Omi•
I
30min 60mio Time
I
I
90min
24hr
0rain
30rnin. 60rain Time
90min
24hr
Fig. 5. R a t e of solute release f r o m large u n i l a m e U a r vesicles a n d m u l t i l a m e l l a r vesicles w i t h (A) c h r o m a t e a n i o n , or (B) D-glucose. A l i q u o t s of l i p o s o m e s c o n t a i n i n g t h e s e s o l u t e s w e r e d i a l y z e d a g a i n s t p h o s p h a t e b u f f e r e d saline ( p H 7.3) f o r 9 0 rain, a n d in s o m e e x p e r i m e n t s f o r 2 4 h , a t 37°C. T h e s o l u t e release is e x p r e s s e d as t h e p e r c e n t ( m e a n ± S.E.) of t h e t o t a l T r i t o n X - 1 0 0 - i n d u e e d l e a k a g e ( 2 4 h), a n d is p l o t t e d as a f u n c t i o n of t i m e . T h e slopes o f release o v e r 90 m i n (% r e l e a s e / r a i n ) w e r e c a l c u l a t e d t h r o u g h linear r e g r e s s i o n analysis b y t h e m e t h o d o f least s q u a r e s .
147 lated b y linear regression analysis over a linear 90 min interval and expressed as the p e r c e n t ( m e a n + S.E.) o f t h e 24 h T r i t o n X - 1 0 0 release per min o f ionic ( c h r o m a t e anion) and n o n - i o n i c (glucose) solutes. Release o f c h r o m a t e anion f r o m large unilamellar vesicles ( 0 . 1 0 7 ± 0 . 0 0 5 , n = 8) was c o n s i d e r a b l y greater t h a n t h a t f r o m multilamellar vesicles ( 0 . 0 3 4 ± 0 . 0 0 7 , n = 5) P = < 0 . 0 0 1 (Fig. 5A). Similarly, the d i f f u s i o n o f n o n - i o n i c glucose m o l e c u l e s f r o m large unilamellar vesicles ( 0 . 2 0 0 _+ 0 . 0 1 4 , n = 6) e x c e e d e d t h a t f r o m multilamellar vesicles ( 0 . 1 2 4 + 0 . 0 0 7 , n = 7) P = < 0 . 0 0 1 (Fig. 5B). T h e s e large unilamellar vesicle flux d a t a are c o n s i s t e n t w i t h the p r e s e n c e o f a single-walled s t r u c t u r e w h o s e internalized a q u e o u s solutes e n c o u n t e r f e w e r diffusion barriers t h a n in m u l t i - l a y e r e d multilamellar vesicles. F u r t h e r m o r e , the ratio {mean + S.E.) o f the p e r c e n t m a r k e r release at 90 min to 24 h is larger f o r the large unilamellar vesicle system t h a n for t h e multilamellar vesicle s y s t e m with e i t h e r c h r o m a t e (large unilamellar vesicles = 0.78 _+ 0.01, n = 8; multilamellar vesicles ~ 0.37 _+ 0.07, n = 5; P = < 0 . 0 0 1 ) o r glucose (large unilamellar vesicles = 0.41 +_0.03, n = 5; multilamellar vesicles = 0.21 + 0.02, n = 3; P = ( 0 . 0 1 ) . Thus, in c o n t r a s t to m u l t i l a m e l l a r liposomes, a greater f r a c t i o n o f t h e sequestered c o n s t i t u e n t s released f r o m large unilamellar vesicles in 24 h pass t h r o u g h the single-bilayer barrier w i t h i n the first 9 0 min. This is in a c c o r d with the m o r e rapid equilibrat i o n e x p e c t e d f o r a unilamellar s t r u c t u r e . Also, the d a t a are c o n s i s t e n t with a m e m b r a n e selectivity in response to ionic and non-ionic solutes, b u t n o t with an intrinsic dissimilarity b e t w e e n m e m branes o f multilamellar and large unilamellar vesicles. Whereas significant differences were f o u n d b e t w e e n the p e r c e n t release o f c h r o m a t e anion f r o m large unilamellar vesicles at 24 h (22.6 + 0.9, n = 8) and multilamellar vesicles (14.0 _+ 2.6, n = 5) P = ( 0 . 0 0 5 , n o d i f f e r e n c e s w e r e f o u n d w h e n glucose was the e n t r a p p e d species (large unilamellar vesicles = 58.9 + 1.4, n = 5; multilamellar vesicles = 5 9 . 0 ± 0.4, n = 3). A l t h o u g h these d a t a c o u l d be i n t e r p r e t e d as a f u n c t i o n o f the stability o f t h e t w o l i p o s o m a l t y p e s , it is m o r e likely a reflect i o n o f the greater rate o f glucose flux in each system. In this case, the glucose d a t a w o u l d suggest equivalence o f multilamellar vesicle and large unilamellar vesicle m e m a b r a n e s which u l t i m a t e l y release c o m p a r a b l e levels o f solute. T h e anion d a t a w o u l d be in a g r e e m e n t w i t h this i n t e r p r e t a t i o n , assuming t h a t sufficient time had n o t y e t elapsed at 24 h f o r t h e slower diffusing c h r o m a t e (in multilamellar vesicles) t o r e a c h equilibrium. In initial e x p e r i m e n t s , w h e n t h e large unilamellar vesicles were p r e p a r e d with the lipids injected via an e t h y l e t h e r r a t h e r t h a n a p e t r o l e u m e t h e r solvent [ 1 ] , the large unilamellar vesicle m e m b r a n e s d i s p l a y e d d i f f e r e n c e s in t h e i r permeability properties. While the 90 min rate o f CrO~- " l e a k " f r o m e t h y l e t h e r multilamellar vesicles ( 0 . 0 2 3 + 0 . 0 0 4 , n = 2) vs. p e t r o l e u m e t h e r multilamellar vesicles ( 0 . 0 3 4 +_ 0 . 0 0 7 , n = 5) was n o t significantly d i f f e r e n t , t h e r e was a sign i f i c a n t decrease in the rate o f CrO~- release in e t h y l e t h e r large unilamellar vesicles ( 0 . 0 6 0 ± 0 . 0 0 7 , n = 2) vs. p e t r o l e u m e t h e r large unilamellar vesicles ( 0 . 1 0 7 ± 0.005, n = 8) P = < 0 . 0 0 5 . Since r e d u c t i o n in p e r m e a b i l i t y has b e e n a t t r i b u t e d to decrease in t h e radius o f c u r v a t u r e f o r small unilamellar vesicles [ 6 , 1 9 ] , these o b s e r v a t i o n s m a y be e x p l a i n e d analagously if t h e large unilamellar vesicle p o p u l a t i o n has a smaller m e a n d i a m e t e r ( f r o m smaller u n i f o r m vesicles or f e w e r large vesicles) w h e n s y n t h e s i z e d w i t h an e t h y l e t h e r , as o p p o s e d
148
to a petroleum ether solvent. This could be a function of differences in solubility, dispersal and vaporization characteristics between these two solvents. Moreover, release of chromate anion over the 90-min to 24-h interval was significantly greater when either multilamellar vesicles or large unilamellar vesicles were prepared using ethyl ether, rather than petroleum ether. Comparison of the ratios of the 24-h/90-min percent chromate release (mean + S.E.) from petroleum ether large unilamellar vesicles (1.29 ± 0.02, n = 8) and ethyl ether
70 60
100
,, t,~ 40
7~
i" 2
2~
t
k'
i
20
'*
1o
"'
0
O-Glucose
~b
6,000 °3
; ,
"4,000 ' 2-
5,000
100 75
,: ;
5 C'
,, '~
25
E
0
,7 !J
o
I~
EL
o
99
5C
o_
o~
~-~
~
3,000
#' a.'%~"~ ' ~,, %,.~ ~ ~,' , ~ , - - ~ , " -
