Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Factors influencing the stability of dehydrated liposomal systems Harrigan, Paul Richard 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1987_A6_7 H38.pdf [ 3.86MB ]
JSON: 831-1.0096955.json
JSON-LD: 831-1.0096955-ld.json
RDF/XML (Pretty): 831-1.0096955-rdf.xml
RDF/JSON: 831-1.0096955-rdf.json
Turtle: 831-1.0096955-turtle.txt
N-Triples: 831-1.0096955-rdf-ntriples.txt
Original Record: 831-1.0096955-source.json
Full Text

Full Text

FACTORS INFLUENCING THE' STABILITY OF DEHYDRATED LIPOSOMAL SYSTEMS by PAUL RICHARD HARRIGAN B.Sc. The U n i v e r s i t y of B r i t i s h Columbia, 1985 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n FACULTY OF GRADUATE STUDIES DEPARTMENT OF BIOCHEMISTRY We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA AUGUST 1987 (cT) PAUL RICHARD HARRIGAN 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date foXP&gfc V* , ABSTRACT Plant seeds, yeasts, bacterial spores, rotifers, and other organisms are capable of suspending their metabolism and entering a state of latency when dehydrated. These organisms may maintain this state for extremely long periods of time, yet upon rehydration resume normal metabolism, without evidence of severe membrane disruption. With many of these organisms, the ability to survive dehydration has been correlated to the production of large amounts of carbohydrates, including glycerol, glycogen and the disaccharide trehalose. Trehalose has been shown to protect isolated sarcoplasmic reticulum microsomes and phospholipid vesicles from dehydration damage , implying that the site of protective action of trehalose and other carbohydrates is the lipid portion of membranes . In this thesis, the effects of carbohydrate composition, vesicle size, and lipid composition on the protection of liposomes from dehydration was investigated, as was the structure of the solid lipid-trehalose complex. Electron microscopy of dried liposomes indicated that vesicles protected with trehalose remain essentially intact even when dry ,while vesicles not 31 13 protected by sugar are severely disrupted by drying . -P and -C NMR results suggested that the lipid of protected vesicles is in a similar phase as that of unprotected vesicles, and that this state is similar to powdered anhydrous phospholipid. Using carboxyfluorescein as a probe, it was demonstrated that trehalose, other sugars can prevent vesicle disruption upon dehydration. Different lipid compositions of the liposomes showed - i i -nearly identical behavior, with the exception of vesicles composed of dipalmitoylphosphatidylcholine and egg phosphatidylcholine, which showed greater and lower stability to dehydration respectively. Light scattering experiments indicated that a wide variety of carbohydrate and lipid vesicle combinations can withstand dehydration and maintain their original size when protected by sugars. The implications of these results in the development of liposomes as pharmaceuticals are discussed, and a hypothesis is advanced regarding the role of carbohydrates in the preservation of dry lipid membranes. -iii-TABLE OF CONTENTS ABSTRACT i i TABLE OF CONTENTS i v L I S T OF TABLES v i i L I S T OF FIGURES v i i i ABBREVIATIONS X ACKNOWLEDGEMENTS x i 1. INTRODUCTION 1.1 H i s t o r i c a l I n t r o d u c t i o n 1 1.2 H y p o t h e s e s o f C a r b o h y d r a t e P r o t e c t i o n o f Membranes... 4 1. 3 L i p o s o m e s a s M o d e l Membranes 5 31 13 1.4 U s e o f -P a n d -C NMR i n Membrane B i o c h e m i s t r y . . . . 6 1.5 G e l t o L i q u i d - C r y s t a l l i n e P h a s e T r a n s i t i o n 7 1.6 Membrane F l u i d i t y 9 1.7 O u t l i n e o f t h i s T h e s i s 10 - i v -2. MATERIALS AND METHODS 2.1 M a t e r i a l s 11 2.2 P u r i f i c a t i o n o f C a r b o x y f l u o r e s c e i n 11 2.3 V e s i c l e P r e p a r a t i o n and Work-up 12 2.4 Dehydration of Liposomes 12 2.5 F l u o r i m e t r y 13 2.6 Membrane P o t e n t i a l Determination 13 2.7 Glucose Release 14 2.8 Phosphorous Assay 14 31 13 2.9 -P and -C NMR P r e p a r a t i o n and Set-up 15 2.10 V e s i c l e S i z e Determination 16 2.11 S o l i d s t a t e F r e e z e - f r a c t u r e E l e c t r o n Microscopy 16 2.12 Determination of Doxorubicin Leakage 16 2.13 D e f i n i t i o n o f Dryness 17 2.14 S t r u c t u r e of Key Chemicals Used 17 3. RESULTS 3.1 S t r u c t u r e of the Dry T r e h a l o s e - V e s i c l e Complex 19 31 3.1.1 -P NMR Study of D r i e d V e s i c l e s 19 13 3.1.2 -C NMR Study of D r i e d V e s i c l e s 21 3.1.3 E l e c t r o n Microscopy of Dry V e s i c l e s 24 3.2 E f f e c t o f V a r i o u s Parameters on the Dehydration 26 -v-3.2.1 V a l i d i t y o f CF as a Marker of Contents 26 3.2.2 Dehydration of Liposomes w i t h and without T r e h a l o s e 32 3.2.3 P r o t e c t i v e E f f e c t o f Other Carbohydrates 35 3.2.4 Dehydration of liposomes of V a r y i n g F a t t y A c i d Composition 37 3.2.5 I n f l u e n c e of C h o l e s t e r o l on Dehydration 39 3.2.6 Dehydration of Liposomes of V a r y i n g P h o s p h o l i p i d Headgroups 39 3.2.7 E f f e c t o f V e s i c l e S i z e on Dehydration 41 3.3 A p p l i c a t i o n t o Liposomal Drug Therapy - Do x o r u b i c i n 43 3.3.1 Uptake of Doxorubicin - Response t o pH Gr a d i e n t 44 3.3.2 P r e p a r a t i o n of Dehydrated Liposomal D o x o r u b i c i n 46 4. CONCLUSIONS 4.1 An Hypothesis Regarding the P r e s e r v a t i o n of Dehydrated Membranes by Carbohydrates 48 4.2 General C o n c l u s i o n s 50 5. BIBLIOGRAPHY 51 - v i -LIST OF TABLES I. S i z e of Rehydrated V e s i c l e s 33 I I . P r o t e c t i v e e f f e c t of d i f f e r e n t carbohydrates on d e h y d r a t i o n of p h o s p h o l i p i d v e s i c l e s 36 I I I . Dehydration of liposomes of v a r y i n g f a t t y a c i d composition i n the presence of t r e h a l o s e 36 IV. I n f l u e n c e of c h o l e s t e r o l i n c o r p o r a t i o n on d e h y d r a t i o n 3 8 - v i i -LIST OF FIGURES 1 S t r u c t u r e of T r e h a l o s e and C a r b o x y f l u o r e s c e i n 18 31 2 S o l i d s t a t e -P NMR of egg PC v e s i c l e s i n the presence and absence of t r e h a l o s e 20 13 3 S o l i d s t a t e -C NMR of egg p h o s p h a t i d y l c h o l i n e v e s i c l e s i n the presence and absence of t r e h a l o s e 2 2 4 F r e e z e - f r a c t u r e e l e c t r o n microscopy of dehydrated v e s i c l e s i n the presence and absence of t r e h a l o s e 25 5 I n f l u e n c e of c a r b o x y f l u o r e s c e i n on g l u c o s e r e l e a s e from liposomes 27 6 Determination of membrane p o t e n t i a l due t o carboxyf l u o r e s c e i n 29 7 E f f e c t of ionophores on c a r b o x y f l u o r e s c e i n r e l e a s e from liposomes 30 - v i i i -8 Retention of Trapped Contents i n Rehydrated Vesicles 34 9 E f f e c t of d i f f e r e n t phospholipid headgroups on the dehydration of liposomes 4 0 10 Dehydration of liposomes of various si z e i n the presence of trehalose 42 11 Uptake of doxorubicin i n response to a proton gradient 45 12 Retention of doxorubicin upon dehydration i n the presence of varying amounts of glucose 47 - i x -ABBREVIATIONS USED C Carbon CF 5 ( 6 ) - c a r b o x y f l u o r e s c e i n CCCP Carbonyl cyanide m-chlorophenylhydrazone DOPC 1 , 2 , - d i o l e o y l - s n - g l y c e r o - 3 - p h o s p h o r y l c h o l i n e DPPC 1 , 2 , - d i p a l m i t o y l - s n - g l y c e r o - 3 - p h o s p h o r y l c h o l i n e E ( e <?9) L i p i d s d e r i v e d from hen egg y o l k POPC 1 - p a l m i t o y l , 2 - o l e o y l - s n - g l y c e r o - 3 - p h o s p h o r y l c h o l i n e Eu Europium HEPES [ 4 - ( 2 - H y d r o x y e t h y l ) ] p i p e r a z i n e e t h a n e s u l f o n i c a c i d I.R. I n f r a r e d + K Potassium MLV M u l t i l a m e l l a r v e s i c l e + Na Sodium NMR Nuclear magnetic resonance P Phosphorous PA P h o s p h a t i d i c a c i d PC P h o s p h a t i d y l c h o l i n e PG P h o s p h a t i d y l g l y c e r o l PI P h o s p h a t i d y l i n o s i t o l ppm p a r t s per m i l l i o n PS P h o s p h a t i d y l s e r i n e QELS Q u a s i - e l a s t i c l i g h t s c a t t e r i n g SCN Thiocyanate T Gel t o l i q u i d c r y s t a l l i n e phase t r a n s i t i o n temperature c -x-ACKNOWLEDGEMENTS I'd l i k e t o thank the Members of the C u l l i s Lab i n d i v i d u a l l y and c o l l e c t i v e l y f o r t h e i r unending a s s i s t a n c e and support. Thanks, eh. - x i --1-1.INTRODUCTION 1.1 HISTORICAL INTRODUCTION Tolerance to dehydration is a novel feature of many anhydrobiotic organisms, including tardigrades, rotifers and plant seeds (Crowe, 1971) . This adaptation has been recognized at least since the time of Leeuwenhoek (Leeuwenhoek, 1702) though , at least in some sense, the ability of seeds to survive drying has been recognized since the development of agriculture . Tolerance to drying has been associated with the production of large (up to 20% by weight of the dry organism) amounts of polyhydroxy compounds such as glycerol, glycogen, and trehalose (Clegg, 1965). These organisms are able to survive apparently complete desiccation for extremely long periods of time (as much as 1000 years) (Keilen, 1959) yet can return to normal metabolism upon rehydration. The role of these carbohydrates in the preservation of the structure and function of complete organisms, isolated lobster sarcoplasmic reticulum microsomes, and phospholipid vesicles has been investigated (Crowe, Crowe and Jackson, 1967; Crowe et al., 1983; Madden et al., 1985). These studies have implicated the lipid portion of the membranes as the site of action of the carbohydrates, although the precise mechanism is not clear. In the absence of any added carbohydrates, isolated lobster -2-sarcoplasmic reticulum microsomes show massive membrane disruption upon dehydration (Crowe et al., 1983) , undergoing fusion and loss of membrane +2 integrity. Using ATP-dependent Ca transport as a measure of microsomal integrity, it was found that exogenously added carbohydrates could prevent +2 both fusion and loss of ATP dependent Ca transport (Crowe et al., 1983, 1984a). The minimal weight ratio of carbohydrate required for retention of ATP-dependent transport varied with the specific carbohydrate tested, though, at high concentrations, a number of sugars share this protective property. Trehalose, a non-reducing disaccharide of glucose, appears to be the most effective at physiological concentrations, and functions at the lowest weight ratio of any of the sugars tested ( 0.5:1 g carbohydrate/g membrane for maximal retention of membrane integrity) (Crowe et al., 1984a). A series of experiments were undertaken by Crowe and associates to explain the protective properties of these carbohydrates. The number or position of axial or equatorial hydroxyl groups of the carbohydrates does not appear to be related to their minimum effective concentrations as determined using lobster sarcoplasmic reticulum microsomes (Crowe et al., 1984b). Using several techniques, Crowe has provided evidence for a direct interaction between the phospholipid headgroups and trehalose when the bilayer is dehydrated. Infrared spectroscopic studies of dry sarcoplasmic reticulum microsomes and of dry DPPC in the presence and absence of -1 trehalose show the band at 1246 cm , the phosphate headgroup stretching band , is dramatically depressed and broadened by the presence of trehalose (Crowe,Crowe and Chapman, 1984) . This, with other I.R. evidence, was taken as evidence of hydrogen bonding between the lipid phosphate groups and the -3-hydroxyls of the sugar. Additionally, an investigation of the binding of residual water to trehalose in the presence of liposomes has shown that trehalose is not available for binding water below trehalose:lipid ratios of 0.5:1 g/g . This was interpreted as a lipid-trehalose interaction preventing the trehalose from binding water (Crowe et al., 1986) , and as being consistent with hydrogen bonding between trehalose and lipid under conditions of low water activity. Recent computer generated molecular models of trehalose-phospholipid associations are also consistent with this H-bonding (Gaber et al., 1986). +3 Additionally, a study of the effect of Eu on the dehydration of +3 liposomes protected by sucrose indicated that Eu interfered with this +3 protection (Strauss and Hauser, 1986). Since it was shown that Eu ions 31 are in close proximity to the -P nucleus (Hauser et al., 1976), the results were interpreted as indicating that the metal ion competes with sucrose for a common binding site, most likely the phosphate group. The evidence for the existence of sugar-phospholipid hydrogen bonds is not totally consistent, however. Phospholipid monolayer expansion by carbohydrates, evidence for such a direct interaction (Crowe et al., 1884d; Johnston et al., 1984), has since been shown to arise from surfactant impurities in the sugars (Arnett et al., 1986). The relative strength of the carbohydrate-phospholipid interactions -4-closely correlated with the minimum amount of that carbohydrate required for membrane preservation for all compounds studied, with the exception of glycerol (Crowe et al., 1983,1984b, 1987). These results do suggest that a sugar-phospholipid interaction is somehow involved in the stabilization of dry membranes, but it is not clear how this interaction affords protection. A trehalose-dependent alteration of a wide range of lipid properties has been invoked to explain this behaviour, including alterations in gel-liquid crystalline phase transition temperatures (Crowe,Crowe and Chapman, 1984) , hexagonal phase formation (Crowe and Crowe, 1982) , membrane fluidity and lateral phase separation (Crowe,Spargo and Crowe, 1987) and the generation of novel phases (Lee et al., 1986). Two main hypotheses have emerged to explain the ability of carbohydrates to protect dehydrated membranes. 1.2 HYPOTHESES OF CARBOHYDRATE PROTECTION OF DRY MEMBRANES Lee has produced deuterium NMR evidence that, for dry DPPC above 46 o C (the gel-liquid crystalline phase transition temperature of a dry lipid-trehalose (1:1) mixture) trehalose can induce a novel phase they call "lambda-phase" (Lee et al., 1986). In this phase, the fatty acyl chains of the lipid are highly disordered, similar to the liquid crystalline state of the fully hydrated lipid. The phospholipid headgroups, however, are 31 immobilized, with a -P NMR spectrum similar to that of anhydrous DPPC in the absence of trehalose. Below the gel-liquid crystalline phase transition temperature, the acyl chains slowly revert to a gel phase with very little -5-chain mobility. It is suggested that the properties of this lambda-phase are important to the maintenance of membrane stability upon dehydration (Lee et al., 1986). An alternative hypothesis suggested by Crowe and Crowe (Crowe, Crowe, and Chapman, 1984; Crowe et al., 1987) is that trehalose acts as "replacement water". In particular, they propose that dry lipids in the presence of trehalose have physical properties similar to hydrated lipids. As mentioned above, they have produced evidence that the sugar hydrogen bonds to the phospholipid headgroup, replacing bound water. In the absence of trehalose, the gel to liquid crystalline phase transition temperature of anhydrous dipalmitoylphosphatidylcholine (DPPC) is 341 K, while a 1:1 weight ratio of trehalose to lipid lowers the T to 315 K, similar to the T of c c hydrated DPPC (Crowe, Crowe, and Chapman, 1984) . They suggest that trehalose protects anhydrous membranes by preventing lipids from entering the gel phase upon dehydration by increasing membrane fluidity (increasing acyl chain disorder). It is suggested that lipid phase transitions or lateral phase separations may be the cause of membrane damage during dehydration (Crowe et al., 1986, 1987). 1.3 LIPOSOMES AS MODEL MEMBRANES Intact cell membranes contain an astonishing variety of lipids, proteins, and other components. Such a system can be too complex for an -6-investigation of physical properties of individual components, and models must be constructed. Simple hydration of lipid yields multilamellar vesicles which can be used as a fairly coarse model of the lipid components of a cell membrane (Bangham et al., 1965) . The liposomes used in this thesis (Large Unilamellar Vesicles produced by Extrusion Techniques, or "LUVETS"), however, have many practical advantages over MLVs as a model membrane. As well as being of homogeneous size and unilamellar, they are quickly and easily prepared, contain no residual detergents, and their composition is easily modified with relatively well understood alterations in their biophysical properties (Hope et al., 1985) . As well, entrapment of an aqueous marker, such as 5(6) carboxyfluorescein can be achieved simply by preparing vesicles in its presence (Hope et al, 1985). Vesicle disruption can then be assessed by measuring the leakage of the marker and monitoring vesicle size changes. The very simplicity of these models means that results obtained with liposomal systems must be extrapolated to biological membranes only with great caution. 1.4 USE OF NUCLEAR MAGNETIC RESONANCE IN MEMBRANE BIOCHEMISTRY NMR, a non-perturbing technique, entails exciting nuclei with radiofrequency radiation and monitoring the energy absorbed. This gives rise to a resonance characterized by the "chemical shift" parameter, which reflects the frequency separation between the resonance of interest and a given standard divided by the radiation frequency. The chemical shift -7-reflects the environment of the nucleus. In proton-decoupled -C NMR of molecules whose nuclei are capable of relatively free motion, for example, a 13 carbohydrate in solution, the -C NMR spectrum consists of a series of sharp peaks at various chemical shift values depending on the environment of the nucleus. However, when motion is restricted, such as for a crystalline carbohydrate, additional interaction with neighboring nuclei are not averaged out , and nuclei will resonate at various frequencies according to 13 their orientation . Thus -C NMR of a crystalline carbohydrate gives rise 31 to a very broad resonance peak. Similarly for -P NMR, the restriction of motional averaging produced by the lipid being in, say, a bilayer of a large (greater than 200 nm) vesicle results in a lineshape characteristic of the motion available. For lipids in the phase ( a fluid bilayer) the primary mode of available motion is rapid rotation about the long axes of the molecules . This results in a characteristic "bilayer" lineshape with a low-field shoulder and high-field peak separated by an effective chemical shift anisotropy of about 40 ppm (Cullis and de Kruijff, 1979) . Correlation 31 between X-ray diffraction data and -P NMR spectra confirms the general 31 utility of -P NMR in determining the phase properties of lipids (Tilcock et al., 1984). 1.5 GEL-LIQUID CRYSTALLINE PHASE TRANSITION The thermotropic behavior of a large number of pure lipids has been -8-investigated using calorimetric techniques. Pure phospholipids in aqueous suspension undergo a phase transition from a rigid gel state to the more fluid liquid crystalline state at a temperature characteristic of the specific lipid. The temperature of this transition can depend on, among other factors, the length and degree of saturation of the fatty acid and on the headgroup composition of the lipid (Chapman, 1975) . In addition, the temperature of the transition also depends on the degree of hydration of the lipid (Janiak et al., 1979) . It is important to note that the alterations in the phase behavior discussed in this thesis are lvotropic (depending on the proportion of water in the system) rather than thermotropic (depending on changes induced by temperature). For example, at a constant temperature o of 50 C, a fully hydrated sample of DPPC is in a liquid crystalline phase (L^ ). As the proportion of water is lowered to about 7 water molecules per phospholipid, the DPPC enters the gel phase (L ,^) (Janiak et al., 1979). From a thermotropic point of view, hydrated samples and dry samples of DPPC o have transition temperatures of 41 and 68 C respectively. To further complicate matters, the presence of trehalose can alter the lyotropic phase o behavior of phospholipids. Thus, at 50 C dry DPPC is in the gel phase, but a dry trehalose-DPPC mixture (1:1 w/w ratio) is not (Crowe,Crowe and Chapman, 1984). The liquid crystalline phase of a dry trehalose-DPPC mixture above the transition temperature is not a true phase commonly referred to as "liquid-crystalline" (Lee et al., 1986) - this term refers to the fact that -9-the hydrocarbon chains have motional freedom reminiscent of a liquid, but a degree of alignment that results in anisotropy of the bulk properties of the system (Silver, 1985). It is not a phase designation. For this reason the trehalose-DPPC mixture was referred to as being in an L -like or "lambda" phase. Lateral phase separation (the co-existence of a gel phase and a liquid crystalline phase in the same membrane) has been shown to occur in some model systems composed of two lipids of widely separated T where one c lipid is above its Tc and another is below its T c (Van Dijck et al, 1977). 1.6 MEMBRANE FLUIDITY The "fluidity" of membranes is an ill-defined property related to the degree of disorder of the phospholipid acyl chains. Strictly, the fluidity parameter is the reciprocal of the viscosity, which in turn is inversely related to the rotational and lateral diffusion rates of the lipid (Cullis et al., 1985) . For example, the incorporation of cholesterol, commonly thought to decrease the fluidity of membranes due to a "condensing" effect on liquid crystalline lipids, has little or no effect on diffusion rates, but does increase the electron spin resonance and NMR " order parameters" of the hydrocarbon matrix . It is the degree of order of the hydrocarbon matrix that is often referred to as "fluidity", though there are dangers in equating the two. One text urges the reader to consider a bowl of sleeping earthworms to demonstrate that a low degree of order may not correlate to high fluidity (Silver, 1985). -10-1.7 OUTLINE OF THIS THESIS The results of this thesis are divided into three main parts. The first is a study of the physical properties of the dry lipid vesicle-trehalose complex using nuclear magnetic resonance spectroscopy and electron microscopy. The second part examines the influence of various parameters on the stability (defined here as retention of vesicle size and contents) of phospholipid vesicles in response to a cycle of dehydration and rehydration. This stability was monitored using quasi-elastic light scattering to examine changes in vesicle size, and relief of fluorescence self-quenching of carboxyf luorescein to quantify release of aqueous contents. The final section investigates the exploitation of this natural adaptation for practical purposes. Liposomes have significant potential as a pharmaceutical preparation, however aqueous suspensions of drug loaded liposomes could be expected to exhibit significant problems with drug leakage and drug or lipid degradation over the months or years between their manufacture and administration to patients. The ability to produce a dehydrated liposomal doxorubicin preparation could be of great utility in avoiding these significant storage problems. - 1 1 -2.MATERIALS AND METHODS 2.1 MATERIALS 14 Phospholipids were obtained from Avanti Polar Lipids , and - C 14 glucose and - C K S C N obtained from New England Nuclear. Carboxyfluorescein (Molecular Probes) was purified as below. A l l other chemicals and buffers were obtained from Sigma Chemicals and used without further purification. 2.2 PURIFICATION OF CARBOXYFLUORESCEIN 5(6)-carboxyfluorescein as commercially obtained can contain significant levels of hydrophobic impurities which can alter membrane structure (Ralston et al., 1981). It was purified by treatment with activated charcoal in boiling ethanol, followed by filtration through Whatman 50 paper and precipitation at -20 C from ethanol/water (1:2, v/v). 10 mL of a 2 M solution of C F (pH 7.4) was passed over a 2.5 X 40 cm LH-20 column eluted at room temperature with distilled water (Lelkes, 1985) . The resulting solution is relatively free of hydrophobic impurities, and is an equal mixture of the 5 and 6 isomers of CF. These isomers are believed to act essentially identically in vitro (Weinstein et al. 1985). -12-2.3 VESICLE PREPARATION AND WORK UP Large unilamellar vesicles were prepared using the L U V E T (large unilamellar vesicles by extrusion techniques) technique employing an "Extruder" (LipEx Biomembranes, Vancouver B.C.) (Hope et al., 1985) . Unless otherwise stated, 40 /zmol lipid was hydrated with 1 mL of 100 mM C F , 150 mM NaCl, 20 mM HEPES (pH 7.4) containing 250 mM trehalose. The mixture was dispersed by vortexing, subjected to 5 freeze-thaw cycles in liquid nitrogen, then passed ten times through two stacked polycarbonate filters of 100 nm pore size (Nucleopore,Pleasanton,CA) under moderate pressure (less than 800 psi). This procedure produce vesicles with a narrow size distribution, with a mean near the filter pore size (Hope et al., 1985) Multilamellar vesicles (MLVs) were prepared as above, omitting both the freezing and extrusion steps. Unentrapped C F was removed by passing the vesicles through a column (1.4 x 15 cm) of Sephadex G-50 (fine) equilibrated with the appropriate buffer. 2.4 DEHYDRATION OF LIPOSOMES Samples (0.5 mL) were dehydrated to a mobile powder at room temperature under high vacuum (60 mTorr) (Virtis Freeze Drier) for 24 hrs. Samples were rehydrated to their original volume with distilled water at 30 o C, and the vesicles dispersed by gentle vortexing. -13-2.5 FLUORIMETRY In the present work, 5(6)-carboxyfluorescein is used as an aqueous marker. Following entrapment at self-quenching concentrations, its release can be monitored as a relief of fluorescence quenching. This procedure is preferable to the use of column chromatography which can exhibit varying lipid recovery if massive fusion occurs during dehydration (Crowe et al. 1984b). Fluorescence measurements were performed on an SLM Amico SPF 300C ratio spectrofluorimeter using an excitation wavelength of 492 nm (bandpass 0.25 nm) and an emission wavelength of 520 nm (bandpass 10 nm). Measurements were made at a lipid concentration of between 15 and -1 30 nmol mL in the appropriate external buffer with or without 25 mM octyl -D glucopyranoside. These concentrations of lipid and detergent have little effect on the fluorescence intensity of CF. 2.6 MEMBRANE POTENTIAL DETERMINATION Membrane potentials were determined from the distribution of the + probe KSCN, as described previously for MTPP (Bally et al., 1985). Briefly, 14 1 juCi of - C KSCN was added to 2 mL of a suspension of vesicles, and incubated at 20 C. -14-At various times, a 100 J J L aliquot was withdrawn and the unentrapped radiolabel removed by passage down a 1 mL Sephadex G-50 mini-column. The trapped label was determined by liquid scintillation counting, and lipid determinations made by phosphorous determination (see below). Knowledge of the trapped volumes of these systems allows the concentration of the probe inside and outside the vesicles to be calculated, and the membrane potential may be calculated from the Nernst equation: A * (mV)= 59 log [SCN-] i n s. d e / [SCN-] o u t s. d e 2.7 GLUCOSE RELEASE Glucose release was determined following preparation of vesicles in 14 -1 150 mM NaCl, 10 mM glucose (containing 4 fxCi -C glucose mL , 20 mM HEPES pH 7.4 with or without 100 mM carboxyfluorescein). Unentrapped glucose was removed by Sephadex G-50 column chromatography, and subsequent leakage determined with 1 mL Sephadex G-50 mini-columns, similar to determination of membrane potential above. 2.8 PHOSPHOROUS ASSAY Phospholipid was determined according by phosphorous assay (Fiske and Subbarrow, 1925) Reagent I was 70% perchloric acid ,Reagent II contained 0.22% ammonium molybdate made up in 2% (v/v) ^SO^, and Reagent III was prepared by dissolving 30 g of sodium bisulfite, 1 g sodium sulfite, -15-and 0.5 g of l-amino-2-napthol-4-sulfonic acid in distilled water at 40 C. Between 0.1 and 0.3 umol of phospholipid was hydrolysed by addition of 0.5 mL of reagent I, and digested for at least 90 min at 190 C. The tubes were cooled, 7 mL reagent II and 0.5 mL of reagent III were added, followed by heating in a boiling water bath for 20 minutes. The absorbance at 815 nm was measured, and correlated to phosphate concentrations through the construction of a standard curve (1 mM potassium phosphate). 2.9 31 -P AND 13-C NMR PREPARATION AND SETUP 2 g samples of dried lipid were prepared in 0.5 mL lots as above, 31 then placed into NMR tubes and sealed. -P NMR spectra were obtained using a Bruker WP-200 Fourier transform NMR spectrophotometer operating at 81.0 MHz. Free induction decays were accumulated for up to 1000 transients employing a radiofrequency pulse width of 11 us, a sweepwidth of 50 kHz, 0.8s interpulse time and gated high power broad band proton decoupling. An exponential filter corresponding to 50 Hz line-broadening was applied prior to Fourier transform. 13 Natural abundance proton decoupled -C spectra were obtained in a similar manner, at 50.3 MHz. -16-2.10 VESICLE SIZE DETERMINATION Vesicle size was determined by quasi-elastic light scattering using a Nicomp 200 Laser Particle Sizer (Nicomp Instruments,Goleta CA) operating at 632.8 nm and 5 mW. As well, some samples were freeze-fractured and replicated (see Madden et al., 1985) 2.11 FREEZE-FRACTURE ELECTRON MICROSCOPY Samples (containing 25% by volume glycerol) were placed directly on cups and frozen in liquid Freon. The frozen samples were dehydrated for 24 hours, then quickly placed into the fracture apparatus and fractured. Replicas were produced and visualized as in Madden et al., 1983. Replicas visualized with the assistance of Dr. M. Hope. 2.12 DETERMINATION OF DOXORUBICIN LEAKAGE Doxorubicin release was quantified by its absorbance at high pH (Mayer et al. 1987) . To a suspension of vesicles, NaOH was added to raise the external pH to 10.5. At this pH a solution of doxorubicin becomes bright blue, while at the pH of the interior of the liposomes, doxorubicin is a bright red color . The absorbance at 600 nm was determined before and after the addition of 0.5% Triton X-100. The ratio of absorbances reflects the amount of doxorubicin free and entrapped in the vesicles. -17-2.13 DEFINITION OF DRYNESS The vesicles in this study were dehydrated to a dry mobile powder, but these preparations were not necessarily anhydrous. It is not certain how many, if any, water molecules remain tightly bound to the lipid or trehalose-lipid mixture. Thus, if only one molecule of water remains bound per lipid molecule, these "dry" vesicles are in fact composed of equal parts water and phospholipid. For comparative purposes, vesicles dried in the presence of tritiated water retained 0.2% of their original tritium counts, while trehalose and trehalose-phospholipid mixtures retained 2.2%. The majority of these tritium counts reflect tritium exchange with the most exchangeable protons of the phospholipid and carbohydrate (see Crowe et al., 1986). 2.14 STRUCTURE OF KEY CHEMICALS USED For ease of reference, the structures of trehalose and carboxyfluorescein are shown in Fig. 1. -18-Figure 1. Structure of Trehalose and Carboxyfluoresceln. structure of trehalose (A) and carboxyfluoresceln (B). -19-3. RESULTS 3.1 STRUCTURE OF THE DRY TREHALOSE-VESICLE COMPLEX To investigate the protective effect of trehalose in preventing the disruption of dehydrated phospholipid vesicles, the physical structure of 31 13 the dried vesicles was examined by solid state -P NMR, -C NMR, and by freeze-fracture electron microscopy, in the presence and in the absence of trehalose. 3J. 3.1.1 -P NMR STUDY OF DRIED VESICLES 31 Solid state -P NMR spectra of egg PC vesicles are similar, whether the vesicles were dried in the presence or absence of sufficient trehalose to protect membrane integrity after rehydration (Fig.2). These spectra are similar to the powder-pattern obtained from anhydrous egg PC , dried from solution in an organic solvent rather than a suspension of vesicles in water. Such a powder-pattern reflects the superposition of all possible orientations of the phospholipid headgroup, oriented essentially at random with respect to the applied field in a "rigid lattice" . This -20-A B I 1 T 1 1 1 - 1 0 0 0 1 0 0 P P m Figure 2. S o l i d - s t a t e 3 l - P NMEl'of egg p h o s p h a t i d y l c h o l i n e v e s i c l e s i n the presence and absence of t r e h a l o s e . 81.0 MHz proton decoupled s p e c t r a of egg P.C. v e s i c l e s d r i e d i n the presence (A) and absence (B) of 250 mM t r e h a l o s e . For d e t a i l s see Section 2.9. -21-pattern indicates that motional averaging from vesicle tumbling, rapid diffusion around vesicles, and rotation around the long axis of the lipid of phosphate portion of the molecule is essentially absent in both the dry lipid and lipid-trehalose mixtures. The breadth of the lineshape, and the near identity of the trehalose and non-trehalose samples suggest that trehalose does not maintain the dried phospholipid in a state similar to the hydrated lipid, at least at 31 the phosphate part of the headgroup. The -P NMR spectrum of hydrated phospholipids in the L q phase typically consists of a low field shoulder and high field peak separated by about 40 ppm (Cullis and de Kruijff (1979). This contrasts with to the suggestion of Crowe (Crowe, Crowe and Chapman, 1984) that trehalose acts as a "water replacement" , maintaining dehydrated membranes in a similar state as hydrated membranes. Analogous results using powdered DPPC and DPPC-trehalose have since been reported [Lee et al., 1986) 3.1.2 13-C NMR STUDY OF DRIED VESICLES 31 13 As was the case with -P NMR, the proton decoupled -C NMR signals of vesicles dried in the presence and absence of trehalose are 13 virtually identical (Figure 3), and these spectra are similar to the -C spectrum of powdered EPC directly subjected to NMR (data not shown). -22-B T 1 1 1 r 5 0 0 ppm F i g u r e 3. S o l i d - s t a t e 1 3-C NMR of egg phosphatidylcholine v e s i c l e s In the presence and absence of t r e h a l o s e . Natural abundance proton decoupled 1 3-C NMR s p e c t r a of egg P.C. v e s i c l e s d r i e d i n the presence (A) and absence (B) of 250 mM t r e h a l o s e . For d e t a i l s see s e c t i o n 2.9. -23-Spectra both in the presence and absence of trehalose show well defined signals from the fatty acid acyl chain carbons centered at about 25 ppm, and weaker signal from the N-methyl groups of the choline moeity of the phosphatidylcholine , centered about 25 ppm further downfield . These values for chemical shifts are very similar to those found for these groups in vesicles in suspension ( K. Wong personal communication ). This indicates that these groups are capable of some motional freedom on the NMR time scale, since the signals from immobile carbon atoms are so broadened that they cannot be detected at the frequency range used in this experiment. There is no detectable signal from the trehalose in Fig 3(a), thus it appears to form an immobile , perhaps crystalline matrix. Other signals which are missing from the spectra of the dried samples, but present in vesicles in suspension include those of the glycerol backbone and the 31 methylene groups of the choline. This is consistent with the -P NMR results which suggest that the phosphate groups are immobile. The somewhat surprising picture of immobile headgroups with more mobile acyl chains is similar to that of the "lambda-phase" proposed by Lee et al. for a dry lipid-trehalose mixture, but this appears to be the case even in the absence of trehalose in our systems. The most important and most easily interpretable characteristic of these four NMR spectra is that the presence of trehalose does not 31 13 drastically alter either the -P or -C signals of the dried lipid, and -24-thus may not be altering the phase behavior of dry lipids.. This is in contrast to the results predicted by the hypothesis of Crowe and Crowe, and of that of Lee et al.. Both hypotheses propose trehalose-dependent phase differences between dry lipid and dry lipid-trehalose mixtures. There may, of course, be subtle phase differences in these vesicle preparations which are not detectable using NMR, and biological membranes may act in a different manner than liposomes, but the results are suggestive. 3.1.3 FREEZE-FRACTURE ELECTRON MICROSCOPY OF DRIED VESICLES Freeze-fracture electron micrographs of vesicles dried in the presence of a 250 mM trehalose (a 4:1 weight ratio of trehalose to lipid) show what appear to be essentially intact vesicles distributed in a matrix of trehalose (Fig. 4a). In contrast, vesicles dried in the absence of any sugar do not form recognizable structures (Fig.4b). Previous freeze-fracture studies have produced similar results (Crowe et al., 1979), however the previous electron microscopy of vesicles dried in the presence of trehalose showed the vesicles to be much more highly distorted "collapsed cups". The difference between the earlier work and this study may be due to the fact that the previous study suspended the dried liposomes in paraffin oil before fracturing. This oil could be responsible for the distortion of the vesicles. The disorganized lipid structures of Fig. 4b are similar to those previously published (Crowe et al., 1979). -25-B Figure 4. Freeze-Fracture E l e c t r o n Microscopy of Dehydrated V e s i c l e s V e s i c l e s prepared as i n S e c t i o n 2.11 prepared i n the presence (A) and absence (B) of 250 mM t r e h a l o s e . -26-3.2 EFFECT OF DIFFERENT PARAMETERS ON DEHYDRATION OF LIPOSOMES In order to better understand the phenomenon of carbohydrate mediated protection of dehydrated membranes, the effect of altering various parameters on the dehydration of liposomes was monitored. In particular, the effects of altering the carbohydrate or lipid composition or the size of the vesicle systems was investigated using the fluorescent dye 5(6) carboxyfluorescein (CF) as a marker of aqueous contents, and quasi-elastic light scattering as a measure of vesicle fusion following a dehydration and rehydration cycle. Following entrapment of CF at self - quenching concentrations its release can be monitored as a relief of fluorescence quenching. This procedure is preferable to the use of column chromatography, which can show varying lipid recovery if massive fusion occurs during dehydration (Crowe et al. 1984c) and has become the method of choice for assessing vesicle integrity (Weinstein et al., 1985) 3.2.1 VALIDITY OF CF AS A MARKER OF AQUEOUS CONTENTS Before using CF as an aqueous marker to monitor vesicle leakage, it -27-8 12 Time (hr) 16 Figure 5. Influence of Carboxyfluorescein on Membrane Permeability. The permeation of 14-C glucose from large unilamellar v e s i c l e s i n the presence , • , or absence, O , of 100 mM carboxyfluorescein was followed as described i n Section 2.7. -28-must be established that the self-quenching concentration employed does not in itself alter membrane permeability. As well, the probe should be entrapped within the vesicle as a result of the lipid permeability barrier, not merely retained as a result of electrostatic forces. To validate the 14 first criterion, -C glucose was entrapped in vesicles in the presence and absence of 100 mM CF. As shown in Fig. 5 glucose efflux from the vesicles is not altered by the presence of the fluorophore, indicating no major perturbation of the membrane by CF. Bramhall has reported that small unilamellar vesicles composed of DPPC are permeable to CF (Bramhall, 1984) . He suggests that the release of CF from vesicles generates a membrane potential which limits the further + release of the fluorophore to the rate of influx of counterions such as Na . To investigate the possibility that this is also the case in the larger systems used in this thesis, a number of experiments were performed. The existence of a membrane potential in CF loaded vesicles was probed directly 14 by monitoring the distribution of the lipophilic anion -C thiocyanate across the vesicle membrane (Fig.6). This molecule would respond to a membrane potential (inside positive) by accumulating in the vesicles in accordance with the Nernst equation (Pick, 1983). Control vesicles in which a positive membrane potential was generated using a potassium thiocyanate 14 gradient (150 mM) rapidly accumulate -C thiocyanate, indicating a potential of greater than 100 mV after 90 min (Fig.6). Similar vesicles loaded with CF, however, show no measurable potential, nor do control vesicles containing 150 mM thiocyanate both inside and outside the membrane. -29-F i g u r e 6. Measurement of membrane p o t e n t i a l due t o ^ F . Membrane p o t e n t i a l as determined by the d i s t r i b u t i o n o f -C KSCN as i n S e c t i o n 2.6. P o t e n t i a l o f v e s i c l e s w i t h a transmembrane g r a d i e n t o f 100 mM c a r b o x y f l u o r e s c e i n ( • ), a transmembrane g r a d i e n t o f 250 mM KSCN ( Q ) , and w i t h 250 mM KSCN on both s i d e s o f the membrane. -30-Figure 7. E f f e c t of ionophores on CF release^ from liposomes. The influence of gramicidin (5 x 10 M) , valinomycin (5_x 10~ M i n the presence of 10 mM KCl), or CCCP (2 x 10 M) on the fluorescence i n t e n s i t y of a suspension of carboxyfluorescein loaded v e s i c l e s i s shown. F u l l r e l i e f of quenching i s observed on v e s i c l e l y s i s using 25 mM o c t y l - j6 -D-glucopyranoside (OGP) . The l i p i d concentration was 20 pM. -31-If the efflux of CF were limited by the influx of counterions, then increasing membrane permeability to such ions should cause rapid release of the fluorophore. In Fig. 7 , the effect of various ionophores on CF leakage is shown. Neither gramicidin, valinomycin, nor CCCP, which allow the + + equilibration of Na , K , and protons respectively, alter the fluorescence of CF loaded vesicles. If, however, the vesicles are lysed with the detergent octyl glucoside release of the fluorophore with consequent relief of self-quenching is observed (Fig.7). These results indicate that electrostatic forces are not preventing CF efflux in this system. The suitability of CF as an aqueous marker is further demonstrated by the fact that similar levels of retention upon dehydration and rehydration are found using either radiolabeled methotrexate or phenylalanine (results not shown). Following removal of external CF by passage of lipid vesicles over a gel filtration column 94% fluorescence self-quenching is observed. This background fluorescence is not reduced by further column chromatography and is stable for several days. It may arise from the fluorophore partitioning into the bilayer. Given that the butanol/water partition coefficient of CF -4 at pH 7.4 is 8.0x10 (Weinstein et al., 1985), a rough estimate of the probe -5 concentration in the membrane is of the order of 10 M. At this concentration in aqueous solution,self-quenching is nearly fully relieved. Consistent with CF in the membrane itself being responsible for background -32-fluorescence, this level is decreased by 50% for vesicles composed of DPPC. At room temperature, this phospholipid is in the gel state and reduced partitioning of the probe into the bilayer would be expected. 3.2.2 DEHYDRATION OF LIPOSOMES WITH AND WITHOUT TREHALOSE Large unilamellar vesicles (mean diameter 100 nm) prepared in the presence of CF are stable for at least three days, showing no detectable release of CF over this time period. If dehydrated and rehydrated in the absence of any carbohydrate the vesicles undergo massive fusion, increasing in size to considerably greater than 1 /zM in diameter, and release nearly all of their originally entrapped aqueous contents (Table I and Fig.8). Similar vesicles prepared in the presence of trehalose and subjected to this cycle are able to maintain their original size and greater than 70% of their contents, in agreement with previous results using isolated sarcoplasmic reticulum microsomes (Crowe et al., 1984). The presence of trehalose appears to stabilize the dry liposomes, allowing nearly total dehydration without loss of vesicle integrity. Thus, the protection of liposomes appear to be a reasonable model for the protection of biological membranes. Also as shown previously, the trehalose must be present on both sides of the vesicle membrane for optimal protection. If the sugar is present only on the outside of the vesicles, only 30% of the entrapped CF is retained following dehydration and rehydration. Interestingly, the vesicles retain their original size distribution. The presence of trehalose only inside the vesicles, however, has no protective effect against either fusion or leakage (Table I, Figure 8). -33-TABLE I SIZE OF REHYDRATED VESICLES CONDITIONS MEAN DIAMETER MEAN DIAMETER OF BEFORE DEHYDRATION AFTER REHYDRATION DEHYDRATION (+ S.D.) (+ S.D.) (nm) (nm) 2 No sugar present 101+24 > 1000 250 mM trehalose on both sides of 107 + 19 105 + 30 vesicle membrane 250 mM trehalose added to outside of vesicles 104 + 36 112 + 56 only 250 mM trehalose 108 + 21 > 1000 trapped only inside vesicles 1 2 - Standard deviations from NICOMP Gaussian Analysis - Vesicles too large to be sized by NICOMP Sizer ( >1000 nm) -34-Figure 8. Retention of entrapped contents i n rehydrated v e s i c l e s . Retention of CF by large u n i l a m e l l a r v e s i c l e s dehydrated and rehydrated as i n S e c t i o n 2.4. V e s i c l e s were dehydrated l n the absence of t r e h a l o s e ( a ) ; i n the presence of 250 mM t r e h a l o s e on both s i d e s of. the membrane ( b ) ; with 250 t r e h a l o s e only on the outside of the v e s i c l e s ( c ) ; or with 250 mM t r e h a l o s e only on the i n s i d e o.f the v e s i c l e s ( d ) . -35-3.2.3 PROTECTIVE EFFECT OF OTHER CARBOHYDRATES The ability to protect phospholipid vesicles is not unique to trehalose, though it appears to be one of the few sugars that can protect at physiological concentrations (Crowe et al., 1984) . While physiological concentrations of trehalose in anhydrobiotic organisms are very high (up to 20% by weight of the organism) (Clegg, 1965), trehalose also appears to be more effective than other sugars at low concentrations. Therefore it was of interest to examine the effect of different sugars at the same concentration, to determine whether there is any difference in the degree of protection afforded by a given carbohydrate. From Table II , there is little difference in the level of retention of CF among the sugars tested, and all samples retained their original size distribution. Thus, a range of carbohydrates share the remarkable stabilizing effects of trehalose. Trehalose was used to protect the liposomes in most of the rest of this thesis since its interactions with dry membranes have been the most thoroughly investigated. The range of carbohydrates effective in protecting these vesicles suggests that the prevention of vesicle disruption upon dehydration may be a general property of sugars, and that any proposed mechanism for this protection should take account of this relative non-specificity. As well, it indicates that alternative sugars can replace the expensive, and non-injectable, sugar trehalose as an agent to preserve liposomes from damage from dehydration. This could be of importance if liposomally encapsulated drugs become a common dosage form. (See Section 3.3). -36-TABLE II PROTECTIVE EFFECT OF DIFFERENT SUGARS  ON DEHYDRATION OF LIPOSOMES SUGAR RETENTION OF CF UPON REHYDRATION (%) Trehalose 73 Glucose 69 Lactose 71 Galactose 70 TABLE II: 40 /zmol egg PC vesicles (100 nm mean diameter) was dehydrated in the presence of 250 mM of the sugar listed, and the retention of carboxyfluorescein upon rehydration was monitored as in Methods. Results shown are the average of two experiments. TABLE III DEHYDRATION OF LIPOSOMES OF VARYING FATTY ACID COMPOSITION  IN THE PRESENCE OF TREHALOSE FATTY ACID RETENTION OF CF UPON REHYDRATION (%) DPPC 88 + 2 POPC 69 + 6 EPC 73 + 4 DOPC 72 + 2 TABLE III: Large unilamellar vesicles (mean diameter 100 nm) were prepared as in Methods using 40 umol of the lipid shown, then dehydrated in the presence of 250 mM trehalose. Results and standard deviations are the average of four experiments. -37-3.2.4 DEHYDRATION OF LIPOSOMES OF VARYING FATTY ACID COMPOSITION The influence of the phospholipid acyl chain composition on the ability of trehalose to protect against dehydration is examined in Table III. These samples, and all others in this thesis unless otherwise stated, maintained their original size distribution after dehydration in the presence of trehalose, but showed dramatic increases in size and loss of trapped CF when dehydrated without trehalose. As the degree of unsaturation of the acyl chains increases (ie. DPPC<POPC=EPC<DOPC) there is little change in the retention of CF by trehalose protected vesicles, despite a large o (approximately 60 C) difference in the gel to liquid crystalline phase transition temperatures of the different phospholipids (Table III). The results of the dehydration of DPPC vesicles are particularly illuminating. The hypothesis of Crowe is that the role of trehalose is to prevent lateral phase separation of different lipid classes by maintaining the phospholipids in a liquid-crystalline phase and increasing membrane "fluidity" (Crowe and Crowe, 1987). However, vesicles composed of DPPC, in the gel state at room temperature both when fully hydrated and when dehydrated in the presence of trehalose, show the highest retention. Further, since these vesicles show the typical fusion and release of CF when dehydrated without trehalose, lateral phase separation is unlikely to be the cause of vesicle disruption, because there is only one lipid class is present. Thus, for liposomes at least, phase separations and membrane fluidity may not be related to dehydration damage. (It must be noted that Crowe's hypothesis is directed mainly at biological systems.) Finally, these results indicate that the "lambda" phase (Lee et al. , 1986) may not play a critical role in protecting dehydrated membranes since the sample was o maintained below the temperature required (40 C) for its formation. -38-TABLE IV INFLUENCE OF CHOLESTEROL INCORPORATION ON DEHYDRATION Sample Carboxyfluorescein Retention (%) EPC 73 ± 4 EPC : CHOL (9:1) 76 + 3 EPC : CHOL (8:2) 74 + 5 EPC : CHOL (7:3) 75 ± 2 EPC : CHOL (6:4) 75 + 4 TABLE IV: 40 /zmol of total lipid was used to prepare liposomes of mean diameter 100 nm as in Methods. These were dehydrated in the presence of 250 mM trehalose, and the retention of trapped fluorophore determined. Results and standard deviations determined from at least four experiments. -39-3.2.5 INFLUENCE OF CHOLESTEROL ON DEHYDRATION OF LIPOSOMES The incorporation of cholesterol into the bilayer is another mechanism whereby the "fluidity" of phospholipid bilayers can be manipulated. As well, it effectively abolishes cooperative gel to liquid crystalline phase transitions (Ladbroke et al., 1968). From Table IV no significant changes in the retention of CF are seen with changes in cholesterol content, up to 45 mol% cholesterol. (Above this amount of cholesterol formation of vesicles becomes difficult). Thus membrane "fluidity" may not be a critical factor in carbohydrate protection during dehydration. That cholesterol does not interfere with such protection is fortunate, since it is likely to be a major component of most liposomal preparations (See Section 3.3). 3.2.6 DEHYDRATION OF PHOSPHOLIPIDS WITH DIFFERENT HEADGROUPS In Fig. 9 the retention of CF by trehalose protected vesicles containing varying ratios of phosphatidylcholine and phosphatidylglycerol is shown. Clearly the presence of the anionic phospholipid destabilizes the dried vesicles. Control samples of the different lipid mixtures not subjected to dehydration showed no leakage over several days. Only the dehydrated sample of 100% P.G. showed an increased mean vesicle size following rehydration . This leakage of CF upon dehydration is not, however, a general feature of anionic phospholipids, since vesicles composed -40-100 75 50 2 5 -0 PC PA PI PS PC(3) (1) (1) PG(1) (1) (3) Figure 9. Effect of different phospholipid headgroups on the protection of dehydrated vesicles. Large unilamellar vesicles of the above lipid composition were dehydrated in the presence of 250 mM trehalose, and retention of entrapped carboxyfluorescein monitored. Numbers in brackets refer to the molar ratio of PG and PC used for those preparations. -41-solely of PI, PA, and PS dehydrated in the presence of trehalose showed similar levels of retention as those of 100% PC. The reason for this effect of PG is not clear. 3.2.7 EFFECT OF VESICLE SIZE ON ABILITY TO WITHSTAND DEHYDRATION A final parameter investigated was the effect of the original vesicle size on the ability of the vesicles to survive dehydration in the presence of trehalose. Vesicles of various sizes were prepared in the presence of trehalose by the extrusion technique with filters ranging in pore size from 30 to 400 nm (Hope et al., 1985). As the pore size decreased below 100 nm, the resulting vesicles had decreased mean diameters, but these did not correspond to the filter pore size, perhaps due to elastic deformation during vesicle formation (Fig.10). It is evident that vesicles extruded through the smaller pore size filters are much more resistant to dehydration damage than the larger systems. It should be noted that vesicles extruded through pore sizes of 200 nm and larger contain some systems containing more than one phospholipid bilayer. Multilamellar vesicles, the largest systems tested, show the lowest levels of retained fluorophore. This result is also difficult to rationalize with a dehydration destabilization mechanism based on the phase properties of the lipid (unless the phase properties change with vesicle size) but suggests that the mechanism of protection may depend on a purely physical property of the trehalose and phospholipid mixture. Storage of MLVs may present a problem. Both dehydration and freezing of these veseicles in the presence of trehalose and/or glycerol result in considerable leakage of CF (unpublished observations). -42-« 1 00 T g 30 50 ,100 200 400 MLVs (76) (84) (105) (184) (312) ( )2000) FILTER PORE SIZE, nm. (MEAN VESICLE DIAMETER, nm.) Figure 10. Dehydration of v e s i c l e s of various s i z e l n the presence of t r e h a l o s e . V e s i c l e s were prepared by e x t r u s i o n through f i l t e r s of va r y i n g pore s i z e as described i n Sectio n 2.3. The mean v e s i c l e diameter as determined by QELS i s i n d i c a t e d i n ' b r a c k e t s below the f i l t e r pore s i z e . Dehydration l n the presence of "250 mM trehalose was as described l n Se c t i o n 2.4. -43-3.3 APPLICATION TO LIPOSOMAL DRUG THERAPY - DOXORUBICIN As mentioned in the Introduction, liposomes containing entrapped drugs have therapeutic potential, with the the immediate prospect of using liposomes to buffer the toxicity of certain drugs, and the eventual possibility of their use as targeted drug delivery systems (Poste, 1986) Drugs which are candidates for liposomal entrapment include the anthracycline anticancer agent doxorubicin (Olson et al., 1982), the antifungal agent amphotericin B, (Lopez-Berestein et al., 1985) and certain antibacterials, antivirals, and antiparasites (Graybill et al., 1982; Koff and Fidler, 1985; Alving and Swartz 1985). In addition, immunomodulating agents entrapped in liposomes show therapeutic promise for potentiating macrophage-mediated destruction of cancer metastases (Fidler et al., 1982). Doxorubicin was chosen as a representative drug. It is active against a wide range of carcinomas , but its toxicity is dose limiting . When administered in a liposomally encapsulated form, both its acute and chronic toxicities are reduced, while its efficacy is either undiminished or, in the case of liver carcinomas, actually increased (Olson et al., 1982). The therapeutic index of the drug is thus enhanced. It is not practical to store aqueous suspensions of this and other drugs entrapped in liposomes, as the liposomes would be expected to exhibit considerable drug leakage and lipid or drug degradation over the time scale required between their manufacture and their administration to patients. This storage problem could hinder their large scale utilization. For this -44-reason, the preparation of dehydrated liposomes is an attractive possibility if conditions can be employed which maintain liposomal integrity and prevent loss of entrapped material. It was therefore of interest to see if a pharmaceutical^  practical dehydrated liposomal preparation of doxorubicin could be obtained using the information gained from the earlier portions of this thesis. The intravenous route of administration of doxorubicin places certain restrictions on the formulation of both the liposomes and the carbohydrates used in the preparation. For example, the sugar trehalose may not be injected intravenously, and the vesicle preparation used must contain cholesterol to prevent large amounts of drug leakage upon exposure to plasma. 3.3.1 UPTAKE OF DOXORUBICIN IN RESPONSE TO pH GRADIENTS Previous work from this laboratory has demonstrated that preparing liposomes with transmembrane ion gradients can result in the uptake of a variety of lipophilic drugs. This "remote-loading" results in much higher trapping efficiencies and drugrlipid ratios , as well as lower rates of drug efflux than passively loaded liposomes (Bally et al., 1987). The advantages of this method suggest that it is a likely procedure for the production of liposomally encapsulated drugs. Since the need for highly toxic ionophores is eliminated if the ion gradient is one of protons, a pH gradient (interior acidic) was used to trap the drug. As seen from Fig. 11, doxorubicin accumulates into the vesicles in response to the pH gradient. -45-21 O CO ZD or o x o Q o LJJ < r -1 0 0 -120 TIME (min) Figure 11. Uptake of doxorubicin ln response to a proton gradient. Large unilamellar vesicles (100 nm diameter) (40 mM EPC:cholesterol (2.2:1 g/g)) Incubated in 13 mM doxorubicin. Conditions were: (a) pll 4 inside and outside vesicle; (b) pit 7.5 inside and outside, the vesicles; or (c) pH 4 inside, pli 7.5 outside.. -46-According to Henderson-Hassalbach relationships, the accumulation continues until [AH +]. n/[AH +] o u t equals [H +]jn/[H +]o u t at equilibrium, where AH + is the protonated from of doxorubicin. Note that this equation may not adequately describe the distribution of doxorubicin in the case of the systems used here, as equilibrium may not have been achieved. Without an applied pH gradient, very little of the drug is associated with the membrane at either at pH 4 or pH 7.5 (Figure 11). 3.3.2 PREPARATION OF DEHYDRATED LIPOSOMAL DOXORUBICIN Using the information from earlier portions of this thesis that other carbohydrates can be substituted for trehalose, that the inclusion of cholesterol does not hinder dehydration and that smaller vesicles withstand dehydration better than larger systems, liposomally entrapped doxorubicin as prepared in 3.3.1 was dehydrated in the presence of varying amounts of glucose. Glucose was chosen because it is acceptable for injection. From Fig. 12, a dehydrated liposomal preparation prepared without glucose present showed roughly 50% retention of doxorubicin. These vesicles underwent fusion to form MLVs of greater than 1 uM diameter, so, for remote loaded doxorubicin, this level of apparent drug retention does not imply that the vesicles withstood dehydration. More importantly, the other samples of Figure 12 did not show changes in vesicle size, and preparations of doxorubicin showing 100% retention of entrapped drug upon rehydration can be prepared using a 50:1 ratio of glucose to lipid. This implies that 12 g of glucose would be needed for an average dose of 50 mg of doxorubicin (at a 3:1 lipid:drug ratio). This rather large amount of glucose could be easily administered as a standard 5% glucose solution in 250 mL, a reasonable amount, since doxorubicin is generally administered via a slow infusion. - 47 -GLUCOSE:LIPID. RATIO ( g / g ) Figure 12. Retention of d o x o r u b i c i n upon dehydration i n the presence of glucose. V e s i c l e s loaded with d o x o r u b i c i n as l n S e c t i o n 3.3.1 dehydrated In the presence of va r y i n g amounts of glucose. Glucose c o n c e n t r a t i o n was constant a t 250 mM, but v a r y i n g volumes corresponding to the weight r a t i o s shown were added e x t e r n a l l y a f t e r remote l o a d i n g of the v e s i c l e s w i t h d o x o r u b i c i n as l n S e c t i o n 3.3.1. -48-4.CONCLUSTONS 4.1 A HYPOTHESIS REGARDING THE MECHANISM OF PROTECTION OF  DEHYDRATED MEMBRANES BY CARBOHYDRATES The results of Section 3.1 of this thesis show that phase differences between dry lipid and dry lipid-trehalose mixtures are not 31 13 detectable using -P or -C NMR. As well, from Section 3.2, phospholipid vesicles of a variety of lipid compositions (including cholesterol) undergo fusion and release of entrapped contents when dehydrated in the absence of any carbohydrates, but are protected from damage if dehydrated with trehalose present. Further, this protection appears to be a general property of carbohydrates, though they may vary in their effectiveness. The results of the NMR experiments and the relative non-specificity of the protection suggests that trehalose-dependent lipid phase differences may have little to do with the protective effect of carbohydrates in preserving dry membranes. Further suggestive evidence that changes in phase properties may not be involved comes from the fact that membranes are protected from lyophilization (freeze-drying) by trehalose in a similar fashion as they are protected from drying without prior freezing. If the prevention of phase transitions were critical in the protection of membranes -49-by carbohydrates, one might expect that the simultaneous reduction of the temperature to that of liquid nitrogen, reduction of the pressure to the milliTorr range, and the near complete removal of water would have a drastic effect on lipid phase properties. Nevertheless, microsomes and phospholipid vesicles emerge from this process essentially intact. These results suggest that a purely physical mechanism may be responsible for carbohydrate dependent membrane protection. Since it is generally accepted that hydration repulsion (the repulsion arising from the the work that must be done to remove water that is tightly bound to the phospholipid headgroups) is primarily responsible for preventing molecular contact and fusion between closely apposed bilayers, it may be that carbohydrates play a similar role. One possibility is that the carbohydrate, hydrogen bonded to the phospholipid headgroups as previously demonstrated, acts as a simple "spacer". Such a role is consistent with the observation that disaccharides protect vesicles at half the molar concentration as monosaccharides, and that glycerol not effective in protecting membranes from dehydration damage, since it may be too small a molecule to function as a spacer . If trehalose is present only on the outside of the vesicles, considerable leakage is observed but no fusion, consistent with a "spacer" role. This leakage indicates that membrane integrity can be compromised even in the absence of fusion. When trehalose is trapped only inside liposomes, both fusion and complete release of contents are observed. Here no effective spacer molecules exist between the vesicles to prevent apposition and subsequent fusion. -50-4.2 GENERAL CONCLUSIONS 31 13 -P and -C NMR of liposomes dried with trehalose are similar to spectra of liposomes dried without trehalose, despite a significant difference in morphology. Electron micrographs of the dried vesicle:trehalose mixture show the vesicles to be remarkably intact, despite being dehydrated, while vesicles dried without trehalose appear severely disrupted. Upon rehydration, trehalose protected vesicles retain their original size and the majority of their contents, while unprotected vesicles undergo fusion with release of contents. A variety of carbohydrates appear to be share the protective ability of trehalose, suggesting that it may be a general property of sugars. A range of phospholipid headgroup and fatty acid compositions, as well as the presence of equimolar cholesterol are not detrimental to protection by trehalose, with the exception of phosphatidylglycerol. PG liposomes are not stable to dehydration even in the presence of trehalose, and the addition of PG destabilizes the dehydration of PC liposomes. Vesicle size is seen to be a critical factor in the protection by trehalose, with smaller systems being most resistant to dehydration damage. Large MLVs show release of trapped contents, suggesting that these systems may not be candidates for preservation by dehydration. Finally it is shown that liposomes with "remote-loaded" drugs, as well as those with passively trapped contents are able to be dehydrated. Retention approaching 100% of the originally entrapped contents is possible under pharmaceutically realistic conditions. -51-5. BIBLIOGRAPHY Abra, R.M., Hunt, A. and Lau, D.J. (1984) J. Pharm. Sci. 73. 203-206. Alving, CR. and Swartz, G.M. (1985) in Liposome Technology, Vol. II, G. Gregoriadis (Ed.), CRC Press Boca Raton, Florida, pp. 55-68. Arnett, E.M., Harvey, N., and Johnson, E.A., (1986) Biochemistry, 25, 5239-5242. Bangham, A.D., Standish, M.M., and Watkins, J.D. (1965) J.Mol. Biol.Ji, 238-244 Bally, M.B.B., Hope, M.J., van Echteld, C.J.A. and Cullis, P.R., (1985) Biochim. Biophys. Acta 8J2 65-70 Bally, M.B., Hope, M.J., Mayer, L.D., Madden, T.D. and Cullis, P.R. (1987) in Liposomes as Drug Carriers: Recent Trends and Progress, G. Gregoriadis (Ed.) CRC Press, Boca Raton, FL. Bramhall, J. (1984) Biochim. Biophys. Acta 778, 393-399. Chapman, D., (1975) Quart Rev. Biophys._8 185-235 Clegg, J. (1965) Comp. Biochem. Physiol. 14 135-143 Crowe, J.H. (1971) American Naturalist 105, 563-574 Crowe, J.H. Crowe, L.M. and Jackson, S.A. (1983) Arch. Biochem. Biophys._218, 477-484 Crowe, L.M. and Crowe, J.H. (1982) Arch. Biochem. Biophys._ 217. 582-587. Crowe, L.M., Mouradian, R., Crowe, J.H., Jackson, S.A. and Womersley, C. (1984) Biochim. Biophys. Acta 769, 141-150. Crowe, J.H., Crowe, L.M. and Chapman, D. (1984) Science 223, 701-703. Crowe, J.M., Crowe, L.M., Rudolph, A.S., Womersley, CW. and Appel, L. (1985) Arch. Biochem. Biophys. 242. 240-247. Crowe, J.H., Spargo, B.J. and Crowe, L.M. (1987) Proc. Natl. Acad. Sci. USA 84, 1537-1540. Cullis, P.R. (1976) FEBS Lett.J70 223-225, 1976 Cullis, P.R., and de Kruijff, B., (1979) Biochim. Biophys Acta 559, 399-420 -52-Fidler, I.J., Barnes, Z., Fagler, W.E, Kirsh, R.M., Bugelski, P. and Poste, G. (1982) Cancer Res._42, 496-501. Finklestein, M.C. and Weissmann, G. (1979) Biochim. Biophys. Acta 587, 202-216. Gabizon, A., Meshorer, A. and Barenholz, Y. (1986) JNCI 77, 459-467. Gaber, B.P., Chandrasekhar, I., and Pattabiraman, N. (1986) in Membranes, Metabolism, and Dry Organisms, Cornell University Press, Ithica N.Y. pp 231-242 Graybill, J.R., Craven, P.C., Taylor, R.L., Williams, D.M. and Magee, W.E. (1982) J. Infect. Dis. 145. 748-752. Hauser, H., Phillips, M.C, Levine, B.A., and Williams, R.J.P. 1976. Nature 26J., 390-394 Janiak, M.J., Small, D.M., and Shipley, G.G.,(1979) J.Biol. Chem. 254 6068-6084 Hope., M.J., Bally, M.B., Webb, G., and Cullis, P.R. (1985) Biochim. Biophys. Acta 812. 55-65. Juliano, R.L. and Layton, D. (1980) in Drug Delivery System: Characteristics and Biomedical Applications, R.L. Juliano (Ed.) Oxford University Press, London pp. 189-236. Keilen, D. (1959), The Problem of anabiosis or latent life: history and current concept. Proc. Royal Soc. London B150: 141-191. In Anhydrobiosis. Dowden , Hutchison and Ross, Strousburg, PA. 1973 Kirby, L., Clarke, J. and Gregoriadis, G (1980) Biochem. J._ 186. 591-598. Koff, W.C. and Fidler, I.J. (1985) Antiviral Res. 5, 179-190. Ladbroke, B.D., Williams, R.M. and Chapman, D. (1968 Biochim. Biophys. Acta 156 333-340 Lee, C.W.B., Waugh, J.S. and Griffin, R.G. (1986) Biochemistry 25, 3737-3741. Leeuwenhoek, A. van, (1702) On certain animicules found in the sediment in gutters of the roofs of houses.Jn The Select Works of Antony van Leeuwenhoek. Letter 144. Translated by Samuel Hoole, 1798. Vol. 2, pp 207-213. Cited by Keilen, 1959 -53-Lelkes, P.I. (1985) in Liposome Technology, Vol.3, 225-246. CRC Press, Boca Raton, FL., pp. 183-195. Lopez-Berestein, G., Fainstein, V., Hopfer, R., Mehta, K., Sullivan, M.P., Keating, M., Rosenblum, M.G., Mehta, R., Luna, M., Hersh, E.M., Reuben, J., Juliano, R.L. and Bedey, G.P. (1985) J. Infect. Dis. 151. 704-710. Madden, T.D., Bally, M.B., Hope, M.J., Cullis, P.R., Schieren, M.P. and Janoff, A.S. (1985) Biochim. Biophys. Acta 817, 67-74. Madin, K.A.C. and Crowe, J.H. (1985) J. Exp. Zool. 193. 335-342. Mayer, L.D., Hope, M.J. and Cullis, P.R. (1986) Biochim. Biophys. Acta 858, 161-168. Olsen, F., Mayew,E., Maslow,D., Rustmus,Y., and Szoka.F., (1982) Eur. J. Cancer Clin. Oncol._J8, 167-176. Pick, U. (1981) Arch. Biochem. Biophys._212, 186-194. Poste, G. (1986) Cancer Treat. Rep._70, 183-199. Ralston, E., Hjelmeland,L.M., Klausner,R.D., Weinstein,J.N., and Blumenthal,R. (1981) Biochim. Biophys. Acta 649, 133-137. Rudolph, A.S. and Crowe, J.H. (1985) Crybiology 22, 367-375. Senior, J., Crawley, J.C.W-. and Gregoriadis, G. (1985) Biochim. Biophys. Acta 839, 1-8. Silver, B.L. (1985) in The Physical Chemistry of Membranes The Solomon Press, NY Strauss, G., and Hauser, H. (1986) in Membranes, Metabolism, and Dry Organisms. Cornell University Press, Ithica, NY Tilcock, C.P.S., Bally, M.B.B., Farren, S.B., and Gruner, S.M. (1984) Biochemistry 1\ 2696-2703 Weinstein, J.N., Ralston, E., Leserman, L.D., Klausner, R.D., Dragsten, P., Henkart, P. and Blumethal, R. (1985) in Liposome Technology (Vol III) G. Gregoriadis (Ed.) CRC Press, Boca Raton, FL., pp. 183-195. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items