UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Optimization of liposomal retention of vincristine Boman, Nancy L. 1994

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

Item Metadata


831-ubc_1994-953144.pdf [ 2.44MB ]
JSON: 831-1.0088027.json
JSON-LD: 831-1.0088027-ld.json
RDF/XML (Pretty): 831-1.0088027-rdf.xml
RDF/JSON: 831-1.0088027-rdf.json
Turtle: 831-1.0088027-turtle.txt
N-Triples: 831-1.0088027-rdf-ntriples.txt
Original Record: 831-1.0088027-source.json
Full Text

Full Text

OPTIMIZATION OF LIPOSOMAL RETENTION OF VINCRISTINEbyNANCY L. BOMANB.Sc. University of British Columbia, 1985M.D. University of British Columbia, 1989A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDepartment of BiochemistryWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAJune, 1994Nancy L. Boman, 1994In presenting this thesis in partial fulfillment of therequirements for an advanced degree at the University of BritishColumbia, I agree that the Library shall make it freely availablefor reference and study. I further agree that permission forextensive copying of this thesis for scholarly purposes may begranted by the head of my department or by his or herrepresentatives. It is understood that copying or publication ofthis thesis for financial gain shall not be allowed without mywritten permission.(Signature)Department of____________________The University of British ColumbiaVancouver, CanadaDate 94ABSTRACTVincristme is a widely used antineoplastic agent. It is a cell cycle specific drugarresting cell growth during metaphase. Since tumors contain dividing cells distributedthroughout the cell cycle, it is important to achieve as long an exposure time as possible.Encapsulation of drugs within liposomes can prolong circulation time in vivo as well asreduce toxic side effects. This thesis examines the effect of liposomal encapsulation onreducing toxicity as well as improving antineoplastic activity of vincristine by altering drugpharmacokinetic and biodistribution behaviour.The first focus of investigation demonstrates that encapsulation of vincristine withinliposomes greatly reduces soft-tissue toxicity of the drug. Subcutaneous injections ofliposomal vincristine are shown to demonstrate minimal toxic effects whereas similarinjections of free vincristine result in gross necrosis and ulceration. Free drug is rapidlycleared from the area of injection. Liposomal drug remains at the area of injection muchlonger, but remains trapped within the liposomes. Slow release rates presumably prevent itfrom exerting cytotoxic effects.The next topic concerns improving the retention of vincristine within liposomes. Theinfluence of lipid composition, internal pH and internal buffering capacity on the retentionproperties of vincristine loaded into LUVs in response to transmembrane pH gradients hasbeen assessed. It is shown that increasing the (saturated) acyl chain length of thephosphatidyicholine molecule, increasing the internal buffering capacity, and decreasing theinternal pH all result in increased drug retention. Further, a study of the pH dependence onIIthe rates of accumulation indicate that uptake proceeds via the neutral form of the vincristinemolecule. This uptake is associated with an activation energy of 37 kcallmol for DSPCICho1LUVs. It is shown that the major improvement in drug retention in vitro is achieved byemploying low initial internal pH values, where 90% retention is obtained over 24 h for aninitial internal pH of 2. Improved retention over the same system with an internal pH of 4in viva was also observed where a drug-to-lipid ratio approximately 4-fold greater at 24 hwas maintained.The third area of investigation concerns the incorporation of cationic lipids to furtherimprove vincristine retention within liposomes. The influence of both the incorporation of10 mol% cationic lipid (AL-l, stearylamine, or sphingosine) into DSPC/Chol (55:45;mol:mol) vesicles and lowering the internal pH to pH 2.0 on the circulation life-time andantitumor activity of liposomal vincristine systems (drug-to-lipid ratio of 0.1:1) has beenexamined. With an internal pH of 2.0, the incorporation of 10 mol% cationic lipid is shownto significantly increase drug retention within the liposomes without affecting lipid clearancetimes. The resulting increase in plasma drug concentration seen by the incorporation of 10mol% sphingosine results in a significant increase in therapeutic activity against the P388lymphocytic leukemia cell line in viva.The final area of investigation examines two different methods for increasing thecirculation longevity of vincristine encapsulated in liposomes. The first involvesincorporation of the ganglioside GM1, which acts to increase the circulation longevity ofliposomal carriers, while the second approach relies on modification of the vincristineencapsulation procedure to enhance drug retention. It is shown that these approaches aresynergistic and increase the circulation half-life of vincristine from approximately one hour111to greater than 12 hours. This results in a dramatic improvement in the therapeutic activityof liposomal vincristirie as measured using a murine P388 lymphocytic leukemia model. Atdoses above 2 mgfkg the optimized liposomal vincristine formulation cures greater than 50%of mice bearing the P388 tumor, whereas free vincristine results in no cures. The optimizedformulation is also shown to significantly increase solid tumor uptake of vincristine withinthe Lewis Lung tumor model and result in improved therapeutic activity.ivTABLE OF CONTENTSAbstract jjTable of Contents vList of Figures viiiList of Tables xiAbbreviations xiiAcknowledgements xivDedication xvChapter 1 Introduction 11.1 Liposomes as Drug Delivery Vehicles 11.2Liposomes 71.2.1 Multilamellar Vesicles (MLVs) 71.2.2 Small Unilamellar Vesicles (SUVs) 71.2.3 Large Unilamellar Vesicles (LUVs) 91.3 Lipid Structures 101.3.1 Phosphatidyicholine 101.3.2 Phase Transitions 141.3.3 Cholesterol 171.4 Transmembrane pH Gradients 201.5 Drug Trapping in Liposomes 211.5.1 Passive Entrapment 211.5.2 Active Entrapment 211.6 Lipid Asymmetry 231.7 Interaction of Liposomes In Vivo 251.7.1 Plasma Protein-Liposome Interactions 251.7.2 Interaction of Liposomes With Cells of the RES 271.8 Factors Affecting Liposome Circulation Lifetimes 271.8.1 Increasing Liposome Surface Hydrophilicity 301.9 Summary 30Chapter 2 Decreasing Vincristine Toxicity by Encapsulation WithinLiposomes 33V2.1 Introduction 332.2 Materials and Methods 342.2.1 Lipids and Chemicals 342.2.2 Liposomes 342.2.3 Drug Entrapment Procedure 352.2.4 Gross and Histological Studies 352.2.5 Skin Retention Studies 362.3 Results 362.3.1 Gross Studies 362.3.2 Histological Studies 392.3.3 Skin Retention Studies 392.4 Discussion 46Chapter 3 Optimizing Vincristine Retention In Liposomes 493.1 Introduction 493.2 Materials and Methods 503.2.1 Lipids and Chemicals 503.2.2 Liposomes 503.2.3 Drug Uptake and Release Experiments 513.2.4 In Vivo Pharmacokinetics Studies .513.2.5 Kinetic Analysis of Vincristine Uptake and Release 523.3 Results 533.3.1 Influence of Acyl Chain Length on Vincristine Retention. . . .533.3.2 Influence of Internal Buffering Capacity on VincristineRetention 553.3.3 Influence of Internal pH on Drug Retention 573.3.4 Influence of Temperature on Vincristine Uptake 613.3.5 Vincristine Retention in DSPC/Cholesterol LUVs in vivo. .. .613.4 DiscussionChapter 4 The Use of Cationic Lipids to Improve Liposomal VincristineRetention 694.1 Introduction 694.2 Materials and Methods 704.2.1 Lipids and Chemicals 704.2.2 Liposomes 704.2.3 Drug Entrapment Procedure 704.2.4 Pharmacokinetic Studies 714.2.5 Biodistribution Studies 714.2.6 Antitumor Studies 72v4.3 Results 724.3.1 Plasma Clearance and Biodistribution Studies 724.3.2 Antitumor Activity of Liposomal Vincristine 774.4 Discussion 80Chapter 5 The Use of Monosialoganglioside 6M1 to Increase LiposomeCirculation Longevity and Improve Liposomal Vincristine Retention 865.1 Introduction 865.2 Materials and Methods 875.2.1 Lipids and Chemicals 875.2.2 Liposomes 875.2.3 Drug Entrapment Procedure 875.2.4 Pharmacokinetic Studies .885.2.5 Biodistribution Studies 885.2.6 Antitumor Studies 885.2.7 Solid Tumor Loading and Efficacy Studies 885.3 Results 895.3.1 Plasma Clearance and In Vivo Drug Release Studies 895.3.2 Antitumor Activity of Liposomal Vincristine 935.3.3 Solid Tumor Loading and Efficacy Studies 965.4 Discussion 100Chapter 6 Summary 108References 113‘11LIST OF FIGURESFigure 1.1Structure of the Vincristine Molecule 3Figure 1.2Model for Vincristine Sensitivity of Tumors SFigure 1.3Multilamellar and Unilamellar Liposomes 8Figure 1.4Freeze-Fracture Electron Micrographs of LUVs Produced by Extrusion.. .11Figure 1.5Structures of Common Phospholipids 12Figure 1.6Phase Transition of Bilayer Lipids 15Figure 1.7Structure of Cholesterol 18Figure 1.8Calorimetric Tracings of DPPC Dispersions (Fully Hydrated)With and Without Cholesterol 19Figure 1.9Redistribution of Weak Bases in Response to Transmembrane pHGradients 22Figure 1.10Biodistribution of Liposomes Within the Organs of the RES 28Figure 1.11Structure of Monosialoganglioside GMI 31Figure 2.1Female BALB/c Mouse Given s.c. Free Vincristine 37Figure 2.2Female BALB/c Mouse Given s.c. Liposomal Vincristine 38vmFigure 2.3Histological Skin Sections following s.c. Free Vincristine 40Figure 2.4Histological Skin Sections following s.c. Liposomal Vincristine 42Figure 2.5Cutaneous Retention of Liposomal Vincristine 45Figure 3.1Vincristine Uptake with Varying Acyl Chain Length 54Figure 3.2Vincristine Release as a Function of Acyl Chain Length .56Figure 3.3Vincristine Release as a Function of Internal Buffering Capacity 58Figure 3.4The Effects of pH on Vincristine Uptake 59Figure 3.5Vincristine Release as a Function of Internal pH 62Figure 3.6Vincristine Uptake as a Function of Temperature 63Figure 3.7In Vivo Drug Retention 66Figure 4.1Plasma Clearance of Sphingosine-Containing Vesicles 74Figure 4.2Influence of Stearylamine and Internal pH on Lipid and Drug ClearanceIn Vivo 75Figure 4.3Influence of AL-i or Sphingosine on Vincristine Retention In Vivo 78Figure 4.4Liver and Spleen Accumulation of Liposomal Vincristine 79ixFigure 5.1Influence of GM1 and Internal pH on Lipid and Drug Clearance In Vivo.. .90Figure 5.2Liver Accumulation of Liposomal Vincristine 94Figure 5.3Spleen Accumulation of Liposomal Vincristine 95Figure 5.4Lewis Lung Tumor Accumulation of Liposomal Vincristine 98Figure 5.5Drug-to-Lipid Ratios Within the Lewis Lung Tumor Model 101Figure 5.6Lewis Lung Tumor Growth Following Vincristine Administration 102Figure 6.1Targeted Fusion of Liposomes 112xLIST OF TABLESTable 1.1Effect of exposure time on the in vitro cytotoxicity of vincristineagainst L1210 cells 4Table 1.2Names and structures of saturated fatty acids 13Table 1.3Effect of fatty acyl chain length and degree of saturation on phasetransition temperature 16Table 2.1Inflammatory response after free or liposomal vincristine injection 44Table 4.1P388 antitumor activity of free and liposomal vincristine inBDF1 mice (effect of sphingosine) 81Table 5.1P388 antitumor activity of free and liposomal vincristine inBDF1 mice (effect of GM1) 97xiABBREVIATIONS USEDAL-i rac 1 ,2-dioleoyl-3-N,N-dimethylaminopropaneApoA- 1 apolipoprotein A-iChol cholesterolDAPC diarachidoyl phosphatidyicholineDBPC dibehenoyl phosphatidyicholineDMPC dimyristoyl phosphatidyicholineDOPA dioleoyl phosphatidic acidDPPC dipalmitoyl phosphatidyicholineDSPC distearoyl phosphatidyicholineEDTA ethylenediaminetetraacetic acidegg-PC/EPC egg phosphatidyicholineGM1 monosialoganglioside GM1HBS hepes buffered salineHDL high density lipoproteinHEPES [4-(2-hydroxyethyl)]-piperazine ethanesulfonic acidi.m. intramusculari.p. intraperitoneali.v. intravascularLUV large unilamellar vesicleMLV multilamellar vesicleNBD-PC N-(4-nitrobenzo-2-oxa- 1 ,3-diazolyl)- phosphatidylcholinexl’PC phosphatidyicholinePE phosphatidylethanolaminePEG-PE phosphatidylethanolamine-polyethyleneglycolpH1 internal pHipH transmembrane pH gradientPS phosphatidylserineRES reticuloendothelial systemSA stearylaminexjnACKNOWLEDGEMENTSMy thesis would not have made it this far if it had not been for all the people who havemade the past 5 years so much fun! I’d like to thank Simon and Archie for pool games andbeer at the Grad Center in the early years, Mike for much of the original artwork in thisthesis and for just being there to kick around, Austin (a fellow chocaholic) who I’doccasionally bump into at Sarah McLachlan concerts, Dave for philosophical conversationsover coffee, Kim for ordering all my lipids when I needed them, Steven (I could neverunderstand a word he said), Barb (for being a hard working inspiration), Diane for promptlypreparing letters of reference, Conrad for being my “lab slave” during my maternity leave,Dana for always being so helpful, the Cancer Control crowd (Jeff, Shane, and Troy), Paulwho’s stories of life in Richmond made me thankful I didn’t live in the “core area”, Wendifor Figure 1.3, Sean, Sandy, and many others.I could not begin to thank Pieter enough for putting up with me for the past 5 years.Thank you also to Mick (second in command), Marcel (The Player) who can talk on thephone longer than anyone I know, Lawrence for teaching me how to make liposomes, Tom,Neil, Laval, and Lewis.Most importantly, I could not have survived the past few years without the superbgrandparenting skills of my Mom and Dad and the superb fathering skills of Bill. Not onlyhave they been a great support to Kylie but also a great emotional support to me. Finally,I would like to thank my beautiful daughter Kylie for making everything in life worthwhile.xivTO MY OLDEST FRIENDS, MY MOM AND DAD;MY BEST FRIEND, MY HUSBAND BILL;AND MY NEWEST FRIEND, MY DAUGHTER KYLIE.xvCHAPTER 1INTRODUCTION1.1 Liposomes as Drug Delivery VehiclesLiposomes have played an important role in the understanding of the structural andphysical properties of lipids in biological membranes. The aqueous interior of lipid vesiclesis able to contain ions and small, water-soluble molecules, sequestering them from theexternal environment, and thus provide simple “models’ of biological membranes. It hasalso been demonstrated that liposomes can encapsulate biologically active molecules suchas proteins, DNA, or conventional drugs (Gregoriadis, 1984; Kabanov et al., 1990; Frezardand Garnier-Suillerot, 1991; Madden et al., 1990). As lipids are biologically compatible andliposomes can potentially be targeted to disease sites, the possibility of utilizing liposomesas in vivo drug carrier systems holds much appeal.The appeal of liposomes as drug delivery systems is particularly strong for anticancerapplications. At the present time, anticancer chemotherapeutic drug therapy is often limitedby the devastating toxic side effects observed in organs or tissues not associated with the siteof disease. However, encapsulation of various antineoplastic agents within lipid vesicles hasbeen shown to decrease toxic side effects while maintaining or increasing therapeutic activity(Herman et al., 1983; Forssen and Tokes, 1981; Gregoriadis and Neerunjun, 1975; Woo etal., 1983; Mayhew and Rustum, 1985; Gabizon et al., 1982a; Kobayashi et al., 1977; Huntet al., 1979; Rahman et al., 1982; Gregoriadis, 1988; Fichtner et al., 1981; Gabizon et al.,1982b). Most of this work has been focused on doxorubicin, however similar observations1have been made for vincristine (Mayer et al., 1990c). Vincristine is an antineoplastic agentderived from the periwinkle plant (see Figure 1.1). It is an important anticancer drug in thatit displays effectiveness against a wide variety of neoplasms including both the Hodgkin’sand non-Hodgkin’s lymphomas, acute lymphoblastic leukemia, embryonalrhabdomyosarcoma, neuroblastoma, breast carcinoma, and Wilm’s tumor (Carter andLivingston, 1976; Sieber et aL, 1976). Vincristine is a cell-cycle specific drug which arrestscell growth exclusively during metaphase by attaching to the growing end of microtubulesand terminating their assembly (Owellen et al., 1976; 1972). For this reason, it ispresumably advantageous to expose neoplastic cells to the drug for prolonged periods oftime. This effect has been demonstrated in vitro by Jackson and Bender (1976), and hasbeen confirmed in our laboratory using the L1210 leukemic cell line (see Table 1.1, Mayeret al., 1993). Table 1.1 demonstrates that by increasing the exposure time for the drug from1 to 72 h, there is a 105-fold decrease in the concentration of vincristine necessary to yield50% cytotoxicity. Further, the potential importance of this relationship in the treatment ofhuman malignancies is supported by clinical trials where patients refractory to bolusvincristine therapy exhibited increased response rates when the drug was administered as a5-day infusion (Jackson et al., 1986a; 1986b). These data indicate that therapy withvincristine can be improved with the same dosage by prolonging tumor cell exposure time.This must be balanced against the threshold level required to maintain antimitotic ability,as illustrated in Figure 1.2.Previous work has shown that liposomal formulations of vincristine can exhibitreduced toxicity and enhanced efficacy compared to free drug (Mayer et al., 1990c). Thishas been related to the enhanced residence times of the drug in the circulation which is2Figure 1.1Structure of the vincristine molecule.c2a5K3Table 1.1Effect of exposure time on the in vitro cytotoxicity of vincristine againstL1210 cells.The drug concentration required to yield 50% cytotoxicity (IC50) decreases from12 4uM to 0.12 nM as the duration of drug exposure is increased from 1 to 72h. (Taken from Mayer et al., 1993).Exposure time (h)a (C50 (4)b1 12.0004 2.4006 2.40024 7.372 0.12a Duration of drug exposure starting at t = 0 over a total incubationperiod of 72 hb Vincristine concentration required to achieve 50% cytotoxicity4Figure 1.2Model for vincristine sensitivity of tumors.This figure indicates that for therapy, cellular levels of drug must be maintainedabove the concentration required to inhibit mitotic spindle formation (M phaseof the cell cycle). Further, since tumors contain dividing cells distributedthroughout the cell cycle it is important to achieve as long an exposure time aspossible.z0Izw0z00wzI-(I)0z>Extended CirculationAnti-mitoticconcentrationG1 S G2 M DCell Cycle/1 /1I, //TIME5achieved when using liposomal carriers, which in turn may be expected to lead to longerexposure of the drug to the tumor. An added advantage is that there tends to be increasedaccumulation within tumors for liposomes which exhibit longer circulation residence times(Vaage et al., 1993; Olson et al., 1979; Bally et al., 1990; Parr et al., 1993; Mayer et al.,1990b; Allen et aL, 1991a). However, a major problem associated with liposomal vincristinepreparations concerns drug leakage from the liposomal carrier. For example, the bestavailable retention achieved in previous work still results in approximately 85% release ofentrapped vincristine from DSPC/Chol liposomes in the blood within 24 h of i.v.administration (Mayer et al., 1990c). Despite this rapid release of vincristine from itsliposomal carrier, significant improvements in therapeutic activity are observed over freedrug. It could therefore be assumed that improvements in drug retention would result inincreased therapeutic activity.The main objective of this thesis is to examine the effect of various parameters onimproving vincristine retention within liposomes with the aim of reducing toxicity andimproving antitumor potency. Chapter 2 examines the ability of liposomal encapsulation ofvincristine to reduce drug toxicity after extravasation. Chapter 3 focuses on the use ofparameters such as lipid composition, internal buffering capacity, and internal pH to improveretention while Chapters 4 and 5 focus on the incorporation of specialized lipids to improvedrug retention.This introductory chapter will first describe liposome structure and the effects ofvarying lipid composition on the physical properties of these systems. The second partreviews the in vivo behaviour of liposomal drug delivery systems.61.2 LiposomesThe term “liposome” refers to one or more lipid bilayers forming an enclosed vesiclewhich is able to encapsulate a distinct aqueous volume. Liposomes can be multilamellar orunilamellar (see Figure 1.3) and can be large (>1 jim diameter) or small (50 nm diameter)(see Hope et aL, 1986; Szoka and Papahadjopoulos, 1980). The stability of liposomes (bothcirculation longevity and the ability to retain molecules) is strongly influenced by thelamellarity and size of the liposomes.1.2.1 Multilamellar Vesicles (MLVs)MLVs spontaneously form by mixing bilayer forming membrane lipids in water(Bangham et aL, 1965; Bangham, 1983). MLVs are typically heterogeneous in size, usuallyin the range of 1-10 jim diameter, and can exhibit unequal solute distributions across thebilayers where the concentration of trapped solute is lower than that in the external medium(Mayer et aL, 1985). Trapped volumes of MLVs can be increased by successive cycles offreezing and thawing (called FATMLVs, Mayer et al., 1985). MLVs are at a disadvantagefor experimental and practical usage due to their large and heterogeneous size as well as theirvariable lamellarity. This precludes quantitative studies on the permeability properties ofbilayers to various solutes, for example, and limits their use as drug delivery vehicles as largesystems are rapidly cleared from the circulation. Because of these disadvantages, procedureshave been developed to generate unilamellar liposomes of defined size.1.2.2 Small Unilamellar Vesicles (SUVs)SUVs range in size from 25-50 nm diameter and can be formed by the sonication of70 0 0D 3 m ‘-4. mcc0 0 1:.)0 43 30C z m I m C) r m Cl)2 3 m .. 0 43 -IC m r I m Cr) C) r m (1)E. — — •1 Qd. r.•.•••.••.rv...••el/ ••••••-.-----—MLVs (Huang, 1969). They can also be prepared using a “French press” process whichinvolves forcing the MLV suspension through a narrow orifice, employing very highpressures (Barenholz et al., 1979; Szoka and Papahadjopoulos, 1980). SUVs are relativelyeasy to prepare and are homogeneous in size. However, due to their high degree ofmembrane curvature they are often unstable and tend to fuse to form larger systems.Another disadvantage is that they have very small trapped volumes (<0.2 jiL/imolphospholipid). These disadvantages make SUVs inappropriate for many applications asmodel membrane systems as well as most forms of drug delivery. Therefore, severalprocedures have been developed to produce unilamellar vesicles of larger sizes.1.2.3 Large Unilamellar Vesicles (LUVs)There are several techniques which have been developed to produce LUVs. Thesimplest method involves the sequential extrusion of MLVs through polycarbonate filters ofvarious sizes (Olson et al., 1979; Hope et al., 1985). With this procedure, homogeneouslysized liposomes can rapidly be produced with vesicle diameters in the range from 50-200nm. The resulting vesicles are called LUVETs (Large Unilamellar Vesicles by ExtrusionTechniques). The major advantages of this technique include its speed, the avoidance ofsolvents or detergents, the high trapping efficiencies obtained, the homogeneous nature ofthe resulting vesicle populations, and the ability to select liposome size through filter poresize.Other methods for producing LUVs include ethanol injection (Kremer et al., 1977),ether infusion (Deamer and Bangham, 1976), reverse phase evaporation (Szoka andPapahadjopoulos, 1978), and detergent dialysis (Mimms et al., 1981). In these methods,9lipid is solubilized in organic solvent or detergent, followed by injection of the mixture intobuffer. The organic solvent or detergent can then be removed by one of several differenttechniques such as distillation, dilution, or dialysis. LUVs prepared by these methods haveaverage diameters of 50-200 nm and trapped volumes of between 1-3 L per mol of lipid(Szoka and Papahadjopoulos, 1980). All of these techniques are time consuming, result inliposome populations of heterogeneous size and may contain residual detergent or organicsolvents (Parente and Lentz, 1984). Due to these disadvantages, extrusion techniques (using100 nm pore size filters) were used to produce the vesicles used for all experiments presentedin this thesis. Electron micrographs of EPC LUVs can be seen in Figure Lipid Structures1.3.1 PhosphatidylcholineThis section will deal only with the primary lipids used for producing liposomesemployed in this thesis. These are phosphatidylcholine (PC) and cholesterol. Structures ofthe most common phospholipids can be seen in Figure 1.5. Phosphatidylcholine is azwitterion composed of a glycerol-phosphate ester with a choline headgroup and two acylchains esterified to the sn-i and sn-2 positions (Small, 1986). Phosphatidylcholine is themost common phospholipid in eukaryotic plasma membranes. Usually, the fatty acid at thesn-i position tends to be saturated, while the sn-2 acyl chain tends to be unsaturated (Small,1986). Tn this thesis, PCs with saturated fatty acids at both the sn-i and sn-2 positions werethe primary lipid species employed. The variety of saturated fatty acids attached to these twopositions is illustrated in Table 1.2. Variations in the acyl chain length and unsaturation canproduce marked effects on the physical properties of the lipid bilayers comprising these10Figure 1.4Freeze-Fracture Electron Micrographs of LUVsProduced by ExtrusionVesicles were prepared by extruding frozen and thawed egg-PC MLVs, at a concentration of 100mM lipid, 10 times through polycarbonate filters of various pore sizes. (A) 400 nm, (B) 200 nm,(C) 100 nm, and (D) 50 nm.11Figure 1.5Structures of common phospholipids.RJH20H2_O—OiH H OHH3C—(CH2)1phosphatidyl cholinephosphatidyl ethanolaminephosphatidyl serinephosphatidyl glycerolphosphatidyl InositolsphingomyelinH2CHN3NH3’H2(H12I—)000a0 O 0z S—flflflflflCl)—‘j—’,—‘,—‘..‘()(Th()fl9c9pocj—WI o,•ciphospholipids, as indicated below.1.3.2 Phase TransitionsLipid molecules in a bilayer can exist in a gel or frozen state which is ordered and rigidor the liquid-crystalline state which is disordered and fluid (see Figure 1.6). In the gel stateall the carbon-carbon bonds are in the extended, all-trans conformation, while in the liquid-crystalline state some are in the gauche conformation. Lateral diffusion occurs rapidly in theliquid-crystalline state but much more slowly in the gel state. For example, dipalmitoyl PCin the liquid-crystalline state exhibits a lateral diffusion rate of D 2 x 108 cm2/sec, whereasin the gel state, D < io cm2/sec (Cullis, 1976). Employing the relation x2 = 4Dt, where xis the distance diffused in time t, it is straight forward to show that in 1 sec a liquid-crystalline DPPC molecule will diffuse approximately 2.8 x 106 m, whereas in the gel statethe distance diffused is less than 0.3 x 106 m. In the more fluid liquid-crystalline state, thelipid acyl chains undergo fluctuations about individual carbon-carbon segments. There isa specific temperature at which a lipid transforms between these two states. Increasing thelength of the fatty acyl chains and increasing their degree of saturation raises this transitiontemperature (see Table 1.3).The transition between the ordered gel state and the fluid liquid-crystalline state canbe detected by calorimetric techniques, which measure the heat required to raise thetemperature of a sample containing the lipid dispersion. As illustrated in Figure 1.4, fordipalmitoyl PC, this transition occurs at approximately 41°C.14OCDCt,-C)CDCD .0p-t0—CD—-t‘<(D—-i1 CD0F—’oCDCD C--.oQ.CCD C) CD CD-I ACDI-—P. CDCD 0 0 P. CD -I 3 0 P. C’) P. 0 zr-c•)I,U)C,CD Cl,CDTable 1.3Effect of Fatty Acyl Chain Length and Degree of Saturationon Phase Transition TemperatureLipid Species Transition Temperature (°C)dimyristoyl PC (14:0, 14:0) 24dipalmitoyl PC (16:0, 16:0) 41distearoyl PC (18:0, 18:0) 55palmitoyl, oleoyl PC (16:0, 18:1)-1dioleoyl PC (18:1, 18:1)-19stearoyl, oleoyl PC (18:0, 18:1) 6stearoyl, linoleoyl PC (18:0, 18:2)-16stearoyl, linolenoyl PC (18:0, 18:3)-13stearoyl, arachidonyl PC (18:0, 20:4)-13161.3.3 CholesterolIn all liposomal preparations employed in this thesis, cholesterol is a major component(45 mol%). Cholesterol is the major neutral lipid component of eukaryotic plasmamembranes. Part of the molecule has a rigid steroid structure, with a polar 3- 3-hydroxylgroup on one end of the molecule (see Figure 1.7).In liposomal systems, the addition of cholesterol to saturated phosphatidylcholineprogressively decreases the enthalpy of the gel-liquid crystalline phase transition. At about30 mol% cholesterol or higher, the transition can no longer be detected (Chapman, 1975).An example of this progression for dipalmitoyl PC can be seen in Fig. 1.8. Cholesterolincreases the “order” in the acyl chains of PCs which are in the liquid-crystalline state anddecreases the order for PCs which are in the gel state (Demel and de Kruyff, 1976). Also,cholesterol can have a condensing effect with PC, where the volume of a mixture of liquid-crystalline PC and cholesterol is less than the volume of the two components separately(Hyslop et al., 1990).It has previously been shown that cholesterol stabilizes liposomal membranes asevidenced by an increase in liposomal solute retention (Papahadjopoulos et al., 1973; Inoue,1974). In vivo uptake of liposomes by RES cells in the liver and spleen is also reduced bythe addition of cholesterol (Patel et al., 1983). Further, there is decreased uptake andintracellular degradation of cholesterol-containing liposomes by cultured Kupffer cells(Roerdink et al., 1989).Several studies have established that liposome interactions with HDL result indestabilization of the liposomes and the transfer of liposomal phospholipid to HDL (Tall andGreen, 1981; Chobanian et al., 1979; Scherphofet al., 1978; Krupp et al., 1976). Cholesterol17HFigure 1.7Structure of cholesterol.CH3CH.Cholesterol18Figure 1.8Calorimetric tracings of DPPC dispersions (fully hydrated)with and without cholesterol.0IENDOTHERMICPure DPPC+5 mol % cholesterol+12.5 mol % cholesterol+20 mol % cholesterol+32 mol % cholesterol+ 50 mol % cholesterolI I290I I320 350Temperature (K)19appears to reduce the net transfer of phospholipid to high density lipoproteins in thepresence of serum (Kirby et al., 1980a; Kirby et al., 1980b; Gregoriadis and Davis, 1979).Kirby and coworkers monitored the fate of3H-phosphatidylcholine and‘4C-cholesteryl oleatein the liposomal membrane and of 6-carboxyfluorescein trapped in the aqueous interior ofliposomes. They found that by incorporation of increasing amounts of cholesterol intoliposomes, there was a greatly diminished release of 6-carboxyfluorescein in the presenceof serum and a decrease in the amount of phospholipid associated with HDL.1.4 Transmembrane pH GradientsMany biological compounds or pharmaceutical agents (such as vincristine) have protonaccepting (basic) groups. The dissociation constant (Ka) for the protonated base can bewritten as follows:Ka = [H][B] I [BHJ (1)where [Hj is the hydrogen ion concentration, [B] is the concentration of the neutral formof the weak base and [BHJ is the concentration of the protonated base. Assuming that Kais the same on both sides of the liposomal membrane, it follows that:Ka = [H][B] I [BH]1= [H]0B I [BH90 (2)where the subscripts 1and to the inside and outside of the liposome. Since it is knownthat the neutral form of ionizable molecules are usually orders of magnitude more membranepermeable than the charged species, (Addanki et al., 1968; Rottenberg, 1979), at equilibrium,the concentration of the neutral species will be the same on both sides of the membrane.Therefore,[BH]1I [BH]0= [H]1 I [H]0 (3)20This result indicates that the presence of a pH gradient across a liposome membrane (insideacidic) will drive the net uptake of weak bases. For example, for a three-unit pH gradient(ApH=3) (inside acidic), an equilibrium concentration of weak base inside the vesicle 1000times higher than outside should be achieved (see Figure 1.9).1.5 Drug Trapping in Liposomes1.5.1 Passive EntrapmentThis thesis focuses on the potential of liposomes as drug delivery vehicles forvincristine. Passive entrapment of drugs can be accomplished by simply preparingliposomes in a solution of the desired molecule for entrapment (Taylor et al., 1990). Thisprocedure results in extremely low drug-to-lipid ratios requiring a high dose of lipidadministration into patients. Previous work with doxorubicin, for example, has show as lowas 4% liposomal trapping efficiencies and drug-to-lipid ratios of 0.004:1 (wt:wt) (Shinozawaet al., 1981). If drug-to-lipid ratios of only 0.01:1 or less are achieveable, to inject atherapeutic dose of vincristme in mice of 2 mg/kg, a lipid dose of 200 mg/kg would need tobe administered. Drugs which have been passively entrapped within liposomes tend to leakout quite rapidly. Between 20-50% of passively loaded doxorubicin is released fromEPC/Chol (55:45; mol:mol) liposomes by 1 h at 37°C (Gabizon et al., 1982a; Mayer et al.,1985).1.5.2 Active EntrapmentDrugs which are weak bases can be actively entrapped within liposomes in responseto a pH gradient (inside acidic). This occurs essentially as outlined for weak bases in Section21Figure 1.9Redistribution of weak bases in response to transmembrane pH gradients;where D represents the drug/weak base of interest.[D]0[Hj0Ka= [DHjØDH1[D]1HjKa [DH]At equilibrium, if:Then:[D] = [D]10[DHj1[DH]f2— [HJ.—INSIDEpH 4.0OUTSIDEpH 7.5DH...—...—..————.—.—.———.———_-———— ———-——---—-—D+H1.4 (Mayer et al., 1990d; Mayer et al., 1986a; Mayer et al., 1986b) (see Figure 1.9). Asurvey of drug accumulation within liposomes exhibiting a pH-gradient has been performed(Madden et aL, 1990). These studies reveal that a large number of drugs which are weaklybasic in character could accumulate within unilamellar vesicles in the presence of a pH-gradient (interior acidic). A similar procedure had earlier been utilized to entrapcatecholamines in liposomes (Nichols and Deamer, 1976).Active entrapment is advantageous in that it allows for much higher drug-to-lipidratios and there is much slower drug leakage (Mayer et al., 1986b). For example, underappropriate conditions doxorubicin can be liposomally encapsulated in response to a pHgradient yielding transmembrane drug concentration gradients in excess of i03 and trappingefficiencies approaching 100% (Mayer et al., 1986b; Mayer et al., 1990d). In addition toinducing efficient encapsulation of doxorubicin, the transmembrane pH gradient alsoenhances drug retention within the liposomes. Less than 5% of doxorubicin is released fromEPCICho1 (55:45; mol:mol) liposomes by 24 h at 37°C (Mayer et al., 1985). Doxorubicinrelease from liposomes increases significantly upon dissipation of the pH gradient causedeither by decreasing the external pH or by addition of proton gradient uncouplers (Mayer etal., 1990d).1.6 Lipid AsymmetryIt is generally agreed that many biological membranes exhibit asymmetrictransmembrane distributions of lipids (Houslay and Stanley, 1982; Op den Kamp, 1979).An example of this can be seen in the erythrocyte membrane wherephosphatidylethanolamine (PE) and phosphatidylserine (PS) are localized to the inner23membrane (Houslay and Stanley, 1982; Op den Kamp, 1979). The asymmetric transbilayerdistributions of lipids commonly observed in biological membranes may be expected to playa role in membrane fusion in vivo. For example, erythrocytes which have lost lipidasymmetry fuse more readily than asymmetric erythrocytes with the fusogenic PE and PSmolecules localized to the inner monolayer (Tullius et al., 1989).The ability of transmembrane pH gradients to generate lipid asymmetry in modelmembrane systems has been described for both basic and acidic lipids (Hope and Cullis,1987; Eastman et al., 1989; Hope et al., 1989; Eastman et al., 1991). Specifically, thecomplete migration of both stearylamine and sphingosine to the inner monolayer has beenobserved for LUVs with acid interiors (Hope and Cullis, 1987). Since both stearylamine andsphingosine are amino-containing lipids which behave as weak bases, they permeate thevesicle membranes in their neutral form to reside on the inner monolayer in their protonatedstate.Chapter 4 examines the use of cationic lipids to enhance vincristine retention withinliposomes. These lipids are utilized to create a positive charge at the inner membranesurface thus decreasing the concentration of drug at that site. The charge repulsion betweenthe vincristine and the inner surface is expected to reduce the ability of vincristine topartition into the inner monolayer of the liposomes, decreasing the rate of release from theliposome. It is important to describe how these lipids behave under the experimentalconditions employed in these studies.241.7 Interaction of Liposomes In VivoIt has previously been demonstrated that MLVs are cleared from the circulatorysystem in a biphasic pattern, with an initial rapid rate followed by a slower rate ofelimination (Gregoriadis and Ryman, 1972). The same biphasic clearance pattern was laterseen for LIJVs and SUVs (for reviews see Gregoriadis, 1988; Hwang and Beaumier, 1988).Although the exact reasons for this clearance pattern are unclear, it is generally believed thatliposome clearance is a result of two major factors: (1) interactions with plasma proteinscausing liposomal membrane destabilization, and (2) the ingestion of liposomes by cells ofthe RES, primarily in the liver and spleen (Gregoriadis, 1988; Hwang and Beaumier, 1988).In order for liposomes to be effective drug delivery vehicles, they require extendedcirculation half-lives, in order to be able to accumulate within target areas. Because of this,the effect of varying physical characteristics on liposome clearance times will be addressedin this section.1.7.1 Plasma Protein-Liposome InteractionsIt has been shown that certain liposome preparations become unstable in the presenceof plasma (Senior, 1987; Gregoriadis, 1973; Gregoriadis and Senior, 1980). Thus, plasmamust contain factors responsible for inducing this instability in liposomes. This section willdeal with the types of proteins which interact with liposomes. These include lipoproteins,complement factors, and other serum proteins.In vitro studies on the role of plasma lipoproteins in inducing liposome instabilityhave been extensively performed (Allen and Cleland, 1980; Mui et al., 1994). There is a nettransfer of liposomal lipids to lipoproteins (HDL) resulting in an increased leakage of25entrapped solutes (reviewed by Senior, 1987) (see Section 1.3.3). The major apolipoproteinof HDL, ApoA- 1 has been implicated in this destabilization effect (Klausner et al., 1985).Liposome instability can also be induced by complement factors as first suggestedby Finkeistein and Weissmann (1979). When the complement cascade is activated, theformation of membrane attack complexes occurs which leads to membrane lysis (reviewedby Muller-Eberhard, 1986). The mechanism whereby the membrane becomes damaged hasbeen extensively studied (reviewed by Alving and Richards, 1983; Muller-Eberhard, 1986).Studies indicate that release of solute from LUVs occurs through stable pores ofapproximately 10 nm diameter (Malinski and Nelsestuen, 1989). Complement damage canalso result in the release of phospholipids from liposomal membranes (Shin et aL, 1978;Kinoshita et al., 1977; Shin et al., 1977).Many other plasma proteins have been shown to bind to liposomes (reviewed byBonte and Juliano, 1986; Chonn et aL, 1992a). An example is the binding of vitamin Kdependent serum proteins via Ca2 ions to anionic phospholipids (reviewed by Jackson,1980). They enhance the conversion of the zymogen forms of the clotting proteins to activeproteases. This may enhance the removal of negatively charged liposomes from thecirculation. As another example, C-reactive protein binding to lipid membranes causesagglutination of liposomes and may affect their structural integrity (Richards et al., 1979).The most prominent protein that is associated with rapidly cleared liposomes is a 50kDa protein (Chonn et al., 1992a). Purified apolipoprotein H exhibits very similarelectrophoretic mobility to this 50 kDa protein (Chonn et al., unpublished). ApolipoproteinH has been shown to act as a cofactor for antiphospholipid antibody binding (McNeil et al.,1990; Galli et al., 1990).261.7.2 Interaction of Liposomes With Cells of the RESFollowing i.v. injection of liposomes, there is rapid uptake of liposomes by the cellsof the RES (Poste et al., 1982, 1983). The majority of these liposomes accumulate in thefixed RES cells of the liver and spleen as well as in the lung, lymph nodes, and bone marrow(Poste, 1983). Evidence of this can be seen from the biodistribution of DSPC/Cholliposomes (see Figures 5.2 and 5.3). The pattern of liposomal uptake can be seen in Figure1.10. This affinity of liposomes for RES cells can be utilized to enhance the therapeuticpotential of liposomally entrapped drugs. It has also been shown that liposome encapsulatedmolecules can accumulate at sites of inflammation (Williams et al., 1986), infection (Morganet al., 1981), and in solid tumors (Ogihara et al., 1986). For example, Morgan andcoworkers determined that approximately 13% of total i.v. injected anionic liposomesaccumulated in a site of thigh abscess within 30 mm following administration. This is incomparison to 4% accumulation in an uninfected thigh. Likewise, experiments in J6456lymphoma tumor-bearing mice demonstrated a 25-fold increase of the liposomeconcentration in the tumor when formulations with long and short blood residence timeswere compared (Gabizon and Papahadjopoulos, 1988). This passive targeting effect isclearly a potential therapeutic advantage as a higher concentration of drug accumulates at thesite of disease.1.8 Factors Affecting Liposome Circulation LifetimesIt is important for liposomes in therapeutic applications to have extended circulationtimes for several reasons. Firstly, they can provide a long-term release of encapsulated drug.Also, passive targeting of liposomes may be enhanced by extending liposome circulation27Figure 1.10Biodistribution of liposomes within the organs of the RES(spleen, liver, bone marrow, lungs, and lymph nodes).28time. Usually, 100 nm diameter vesicles composed of long chain saturated lipids such asDSPC combined with cholesterol are used as basic drug delivery vehicles (Bally Ct al., 1990;Gabizon and Papahadjopoulos, 1988). The circulation half-lives of these systems can beextended by incorporation of the ganglioside GM1 (Allen and Chonn, 1987) orpolyethyleneglycol derivatives (Blume and Cevc, 1990; Klibanov et al., 1990; Allen andHansen, 1991; Lasic et al., 1991; Allen et aL, 1989). These “extended lifetime” formulationsyield dose-independent clearance kinetics, with 5 to 30% of the injected dose remaining inthe (mouse) circulation at 24 h post injection. There are six major factors which play asignificant role in promoting liposome circulation lifetime. These include, as emphasizedpreviously, the presence of cholesterol in liposome membranes (Gregoriadis and Davis,1979) and using phospholipids with high phase transition temperatures (Blok et al., 1975;Senior et al., 1982). Liposomes are also rendered more long-lived by decreasing their size(Senior, 1987). For example, for DSPC/Chol systems, small unilamellar vesicles andvesicles passed through 400 nm diameter pores exhibited circulation half-lives of 7.5 and 0.2h, respectively. They can also be rendered more long-lived by injecting higher lipid dosesto saturate cells of the RES (Abra et al., 1980). A saturation of liver uptake with increasinglipid dose has been demonstrated for liposomes of various sizes, together with acorresponding increase in blood levels. An alternative theory suggests that saturation of theRES, per Se, does not occur due to inhibition of uptake by RE cells. Rather, the decreasedclearance rates observed for liposomes injected at higher doses may be due to a decrease inprotein binding to these liposomes (Chonn et al., 1992a; Oja et al., unpublished). Also, thepresence of encapsulated cytotoxic drug can enhance liposome circulation time (Bally et al.,1990; Parr et al., 1993). Lastly, as previously mentioned, recent studies have shown that29increasing the hydrophilic nature of the liposome surface can prolong liposome circulationtime, as summarized in the next section.1.8.1 Increasing Liposome Surface HydrophilicityLiposome surface hydrophilicity can be increased by the incorporation of moleculessuch as monosialoganglioside GM1 (see Figure 1.11) (Allen and Chonn, 1987; Gabizon andPapahadj opoulos, 1988) and phosphatidylethanolamine-polyethyleneglycol (PEG-PE)derivatives (Blume and Cevc, 1990; Klibanov et at, 1990; Allen and Hansen, 1991; Lasicet al., 1991; Allen et at, 1989). These molecules are postulated to provide a steric barrierthat inhibits the liposome association with plasma proteins (Chonn et al., 1992b; Allen et al.,1989; Klibanov et at, 1991; Senior et al., 1991a). Allen and coworkers have demonstratedthat addition of 10 mol% GM1 into DSPCICho1 (2:1; mol:mol) vesicles of 100 nm diameterincreases the recovery of liposomes in the circulation from 0% to 28% 24 h after i.v.injection. This is for an injected lipid dose of 25 mg/kg. Klibanov and coworkers havedemonstrated similar results with the incorporation of 10 mol% PEG-PE.1.9 SummaryEarlier work with liposomal vincristine demonstrated that encapsulation withinDSPC/Chol liposomes decreased the toxicity seen with free drug (Mayer et at, 1990). The50% lethal dose of 1.9 mg/kg in CD-i mice observed for free vincristine increased to 4.8mg/kg upon encapsulation of the drug within these liposomes. Liposomal encapsulation ofvincristine also enhanced the antitumor activity against murine P388 and L1210 lymphocyticleukemia models. This resulted from increased efficacy for liposomal vincristine at doses30Figure 1.11Structure of Monosialoganglioside GM1.Gal GalNac GalH(GlcCH2OCeramideIHOCNanOHR= Ih-OH1—OH31equal to free drug as well as the ability to administer increased doses of liposomalvincristine.These improvements were observed for systems which released approximately 85%of the drug within 24 h after injection. It may be expected that further therapeuticimprovements should be achievable for vincristine preparations with improved retentionproperties. Prior to this work, however, a liposomal system has not been developed whichis capable of retaining vincristine for longer periods of time. The purpose of this study isto determine the factors necessary for prolonging vincristine retention within liposomes andusing these factors to improve vincristine retention, and to correlate this with the antitumoreffects of such improved systems.The effect of liposomal encapsulation on decreasing drug toxicity is discussed inChapter 2. Physical factors such as lipid composition, internal buffering capacity, internalpH, and temperature and their role in improving drug retention are discussed in Chapter 3.In Chapters 4 and 5, the incorporation of novel lipids such as sphingosine andmonosialoganglioside GM1 and their effects on vincristine retention in liposomes is examined.32CHAPTER 2DECREASING VINCRISTINE TOXICITY BY ENCAPSULATION WITHINL1POSOMES2.1 IntroductionAs mentioned in Section 1.1, vincristine is a widely used antineoplastic agent thatdisplays effectiveness against a wide variety of neoplasms (Carter and Livingston, 1976;Sieber et aL, 1976). A dose limiting toxicity associated with vincristine use is its neurotoxiceffects which are manifest mainly as peripheral neuropathy. Vincristine is also known forits ability to produce soft tissue necrosis and ulceration if accidently extravasated during i.v.administration or injected i.m. (Bellone, 1981; Choy, 1979).Liposomal encapsulation of several antineoplastic drugs has been shown to decreasetoxic side effects while maintaining their therapeutic potential (Herman et al., 1983; Forssenand Tokes, 1981; Gregoriadis and Neerunjun, 1975; Woo et al., 1983; Mayhew and Rustum,1985; Gabizon et al., 1982a; Kobayashi et al., 1977; Hunt et al., 1979; Rahman et al., 1982;Gregoriadis, 1988; Fichtner et al., 1981; Gabizon et al., 1982b). This includes the ymcaalkaloid vincristine (Mayer et al., 1990c). It has previously been demonstrated thatliposomal encapsulation of doxorubicin decreases the vesicant properties of the drug(Balazsovits et al., 1989; Forssen and Tokes, 1983). However, minimal attention has beengiven to the ability of liposomal encapsulation to diminish the vesicant properties ofvincristine.Several studies have been devoted to analyzing the effectiveness of antidotes in the33treatment of accidental extravasation of vincristine (Loth and Eversman, 1986; Barr andSertic, 1981; Dorr and Alberts, 1985). However, in severe cases of extravasation, noantidote injections have been shown to significantly alter the outcome of the accident (Dorrand Fritz, 1982; Knoben and Anderson, 1983). In this chapter we assess the effect ofliposomal encapsulation on reducing the vesicant properties of vincristine, therebyalleviating the need for antidotes if inadvertant extravasation of the drug should occur.2.2 Materials and Methods2.2.1 Lipids and ChemicalsDistearoyl phosphatidyicholine (DSPC) was purchased from Avanti Polar Lipids andwas >99% pure. Cholesterol, HEPES, and citric acid were obtained from Sigma (St. Louis,MO, USA). Vincristine sulfate was purchased from Lynphomed (Markham, ON).[14C]cholesteryl hexadecyl ether was produced by special order from New England Nuclear(Ontario, Canada) and was >95% pure. It was chosen as a lipid, marker due to its stabilityin vivo (Derksen et al., 1987).[3H]vincristine was obtained from Amersham (Oakville,Ontario, Canada). Female BALB/c mice (retired breeders) were purchased from CharlesRiver Laboratories.2.2.2 LiposomesDSPC/Chol (55:45; mol:mol) liposomes were prepared by first dissolving the lipidmixture in 95% ethanol at 60°C for 30 mm. (100 mg lipid/mL). Multilamellar vesicles(MLVs) were formed by adding 300 mM citric acid pH 4.0 and vortex mixing vigorously(25 mg lipidlmL final mixture). The resulting MLVs were then incubated at 60°C for an34additional 30 mm to ensure equilibration of citrate buffer across the lipid bilayers.Following incubation, large unilamellar vesicles (LUVs) were produced by extruding theMLVs through an extruder containing 2 polycarbonate filters with 100 nm pore size. Theextrusion device was obtained from Lipex Biomembranes (Vancouver, British Columbia,Canada) and was equilibrated at 60°C. Following extrusion, the liposomes were dialyzedagainst two changes of 100 volumes of citric acid buffer (pH 4.0) over a 24-h period.Spectra/Por 2 dialysis tubing was used (cutoff 12-14 kDa).2.2.3 Drug Entrapment ProcedureVmcristine was loaded into the liposomes as follows. The vesicles were passed downa G-25 Sephadex column equilibrated with HBS pH 7.4 to exchange the external buffer.Vincristine sulfate was then added to the liposomes to achieve a drug-to-lipid ratio of 0.1:1.The resulting drug/lipid mixture was then incubated at 60°C for 10 mm. This procedureresults in greater than 95% trapping efficiency of the drug.2.2.4 Gross and Histological StudiesThe procedure used for skin toxicity experiments has been previously described (Dorret al., 1980). Briefly, an approximately 3-cm2 area of hair above the hindleg of the adultBALB/c mouse is removed by vigorous rubbing with Neet topical depilatory lotion(Whitehall Laboratories, New York, NY). This procedure causes no adverse skin effects initself (Doff et al., 1980). Twenty-four hours following hair removal, 10 .tg of either free orliposomal vincristine (diluted to 50 iL in normal saline) was injected subcutaneously usinga 25-gauge needle (bevel up). Mice were visually monitored twice daily for any evidence35of skin damage or gross ulceration. Mice appearing to be in any distress were immediatelysacrificed by cervical dislocation following anaesthetic.For histological examination, at varying time points, mice were anaesthetized withan i.p. mixture of ketamine 160 mg/kg and xylazine 20 mg/kg. Following anaesthetic, skinaround the site of injection was removed and immediately placed in 10% formalin as afixative. Histological sectioning and analysis were subsequently performed. Haematoxylinand eosin were used for staining.2.2.5 Skin Retention StudiesThe presence of liposomal vincristine in the skin was monitored at varying timepoints by radiolabelling. Skin was removed from the mice as in section 2.2.4, removing acircular section with a diameter as close as possible to 2 cm. The skin was thenhomogenized in saline using a Polytron homogenizer (Brinkmann Instruments, Rexdale,Ont.). Skin homogenates (total volume) were then digested with 500 jiL of “Solvable”(DuPont Canada, Inc., Mississauga, Ont.) for 3 h at 50°C. Subsequently, the samples werecooled to room temperature before decolorizing with 200 j.iL of 30% hydrogen peroxide.Samples were then analyzed using dual label liquid scintillation counting.2.3 Results2.3.1 Gross StudiesFigures 2.1 and 2.2 are photographs of mice given subcutaneous injections of freeand liposomal vincristine, respectively, on day 9. Neither treatment group displayed anyevidence of erythema or edema at the site of injection over the first 7 days following36Figure 2.1Female BALB/c mouse given s.c. free vincristine.The mouse was given a single s.c. injection (50 iL) of free vincristine (10 tg). The necrosis andulceration was prominent on day 9 following injection.I37Figure 2.2Female BALB/c mouse given s.c. liposomal vincristineThe mouse was given a single s.c. injection (50 jiL) of liposomal vincristine (10 jig drug, 100 jiglipid). There was no evidence of inflammation or ulceration at any time following the injection.This picture was taken on day 9 following injection.k38injection. On day 7, 5 of the 10 mice given free vincristine showed ulcerations withdiameters of 2-3 mm. By day 10, 9 mice showed evidence of ulceration, 4 of which hadulcerations of 5-6 mm diameter. By day 11, all mice had ulcerated, 6 mice havingulcerations of 5-8 mm diameter. The progressive severity of the lesions at this pointnecessitated sacrificing the animals. In contrast, mice injected with liposomal vincristineshowed no evidence of skin necrosis or ulceration throughout the time course of theexperiment. Control animals given saline or empty liposomes showed no evidence ofinflammatory response.2.3.2 Histological StudiesSkin sections were examined at 1, 3, 5, and 7 days following subcutaneous injectionof both free and liposomal vincristine. Injection of free drug resulted in a rapid infiltrationof inflammatory cells in the subcutaneous tissues (day 1) followed by a rapid resolution byday 3 (Figure 2.3). In contrast, when injected with liposomal vincristine, the initialinflammatory response in the subcutaneous area is less intense but is more prolonged, lastingthroughout the 7 days studied (Figure 2.4). The numbers of inflammatory cells in the dermalarea are relatively constant for both vincristine preparations and are 2-3 fold higher for theliposomal preparation. The estimated number of inflammatory cells is shown in Table Skin Retention StudiesSkin samples were analyzed at 1, 3, 5, and 7 days following subcutaneous injectionof free or liposomal vincristine. Free drug was cleared from the site of injection very rapidlywith only 0.4% of the drug remaining (0.04 .tg) on day 1 after injection (Figure 2.5). This39Figure 2.3Histological skin sections from BALB/c mice given free vincristine.*I,, -‘T •hfr•,?‘‘3•‘—•.4t.4jir ;•-_ .‘..11 “.*•V — -.—‘ .. —•d-—BEach mouse was given a single s.c. injection (50 pL) of free vincristine (10 fig). All sections areshown at 25X magnification and stained with haematoxylin and eosin. Photographs representsections at 1 (A), 3 (B), 5 (C), and 7 (D) days following injection.h w f‘ -4p_Ø4.I :.-. -- • • ,__l,’:-40CD41Figure 2.4Histological skin sections from BALB/c mice given liposomal vincristine.Each mrnise was given a single s.c. injection (50 iL) of liposomal vincristine (10 .tg). All sectionsare shown at 25X magnification and stained with haematoxylin and eosin. Photographs representsections at 1 (A), 3 (B), 5 (C), and 7 (D) days following injection.,1’ — -..‘.I — —— :: ..,..h4K •.a- I• —S ..- ,..‘ •b • •i.— % • .•..— a • -4-:‘ I• “I •‘-t r. • b • —p9—id I•I I-— ,.B.42•%.—-*-.-r- —0•‘—Jr.;.::hu1_,.it;p—.•.C,e_%...—••.•fr,.t p.*——.—‘..-4_——•I)_• •a-,—.9•—..I4I —*4’—4—I’•••_,,I•4——.—•...4•4•..—_—..s•—-.pa ,jp-# Is......—a-— •—,.—.•‘%;4••..44I.-•I44•p. -4,•_* _444I994•4PS.*AI.-•4-‘i•1•1•*-m—?‘b_-..ATable 2.1Inflammatory Response Following Subcutaneous Injection ofFree or Liposomal VincristineSample Injected Dermal Subcutaneousinflammatory cells inflammatory cells(cells/mm2) (cells/mm2)Controls 8.2 x iO 9.8 x i0Free vincristine (1Oig)day 1 1.9 x i0 1.3 x 106day 3 1.8 x i05 6.8 x i0day5 1.5x 10 9.Ox i0day7 1.lx 105 9.Ox 10Liposomal vinc (10.tg)dayl 4.3x 10 6.9x 10day3 2.Ox 10 5.4x 105day5 4.6x 10 6.6x 10day7 7.2x 10 9.Ox 1044Figure 2.5Cutaneous Retention of Liposomal VincristineCutaneous retention of vincristine (A) and lipid (B) following s.c. injection of10 tg of free (.) or liposomal (•) vincristine. The drug-to-lipid ratio ofliposomal vincristine was 0.1:1 (wt:wt). Error bars represent standard deviationsof four mice.1.0 A0.9 -0.8-z 0.70.6-Liiz 0.5 -1—0.4-C-)z 0.3 ->0.10.0 - -01009080706050403020100I I I1 2 3 4 5 6 7B0 1 2 3 4 5 6 7TIME (DAYS)45residual amount remained constant over the 7 day time course. By comparison, there wasapproximately 6% of the drug remaining at the site of injection for the liposomal preparationon day 1 (0.6 rig). This remaining drug level gradually declined over the time course of theexperiment until approximately 0.3% remained on day 7. Approximately 45% of theliposomes remained at the subcutaneous site on day 1 and decreased to approximately 30%by day 7. This corresponds to a drug-to-lipid ratio of approximately 0.01 on day 1 followinginjection or 90% leakage of the drug. This value at 24 h corresponds to the same drug-to-lipid ratio seen in the circulation for this liposomal preparation 24 h after i.v. administration(Mayer et al., 1990c).2.4 DiscussionThis chapter investigates the effect of liposomal encapsulation on the vesicantproperties of vincristine. It has previously been shown that encapsulation of the drug reducesvincnstine toxicity as evidenced by a decrease in weight loss over time in mice (Mayer etal., 1990c, Boman et al., 1994). The soft tissue toxicity of liposomal vincristine, however,has received little attention until now. In this chapter we focus on the vesicant properties ofthe drug since extravasation can be devastating and seriously affect the patienttsquality oflife, particularly since vincristine is widely used in paediatric patients for the treatment ofacute lymphoblastic leukemia. Extravasation injuries have been shown to be highest in thepaediatric and geriatric age groups (Upton et al., 1979). Extravasation is also a relativelycommon occurrence, occuring in as many as 1-2% of chemotherapy infusions (Spiegel,1981).In patients, following antineoplastic drug extravasation, blistering and skin loss46become apparent in a few days followed by a progressive tissue necrosis which can continuefor as long as 3 months (Reilly et al., 1977). Full thickness skin necrosis can eventuallyensue, exposing underlying tendons and neurovascular structures (Spiegel, 1981). Due tothe seriousness and frequency of antineoplastic extravasation, it is felt that minimizing theseeffects by liposomal encapsulation is an important factor to be studied.The mechanism by which soft-tissue necrosis occurs is widely assumed to be due toa directly cytotoxic effect of the drug. Histologic analysis has been reported on two patientsfollowing inadvertent extravasation of cytotoxic drugs (Rudolph et al., 1976; hait andDinner, 1975). These studies revealed a nonspecific chronic inflammation with a patentmicrovasculature. Regardless of the mechanism and severity of vincristine-induced necrosis,various antidote therapies have questionable efficacy (Loth and Eversman, 1986; Dorr andAlberts, 1985).Liposomal encapsulation of doxorubicin has been shown to dramatically reduce thevesicant properties of the drug (Balazsovits et al., 1989; Forssen and Tokes, 1983). Usingthe same approach to decrease the vesicant properties of vincristine, it was found thatliposomal encapsulation also dramatically reduced soft-tissue damage by the drug. Therewas virtually no evidence of inflammatory response seen grossly when liposomal vincristinewas administered subcutaneously. In contrast, free drug produced gross ulceration by day9 following injection. On histological analysis, liposomal vincristine produced a chronic andmild inflammatory response whereas free drug produced an acute and more intenseinflammatory response. This is likely due to the fact that the drug was shown to remain inthe area of injection for a much longer time period when liposomally encapsulated. Thedrug leaks slowly from the liposome interior, resulting in a long-term, low dose of free drug47being exposed to the tissue. In comparison, when free drug is administered, there is a briefexposure of the full drug dose to the tissue before being absorbed by the circulatory system.In summary, it is clear by this study that the liposomal encapsulation of vincristinegreatly reduces its potential for producing tissue necrosis upon accidental extravasation. Aspreviously mentioned, the drug’s therapeutic potential remains the same. This coulddramatically improve the quality of life of the cancer patient and allow the drug to beadministered more safely.48CHAPTER 3OPTIMIZING VINCRISTINE RETENTION IN LIPOSOMES3.1 IntroductionSince the ymca alkaloids are cell-cycle-specific cytotoxic drugs, it is likely that anability to maintain high plasma levels of vincristine for extended periods would beadvantageous. A liposome system which provides extended drug retention in the circulationis therefore desireable. Previous work with doxorubicin has shown that increased drugretention within liposomes leads to an increase in drug circulation time in vivo (B ally et al.,1990) and increased anti-tumor activity (Mayer et al., 1990b). However, currently availableretention properties for vincristine are not optimal, as the best available formulation ofliposomal vincristine releases drug in vivo with a half-life of less than 8 h, leading to morethan 90% release by 24 h.In this chapter we explore three different parameters for improving the liposomalretention of vincristine. These include the use of phospholipids with increased acyl chainlength, reduction of the interior pH, and increased interior buffering capacity. The latter twoparameters are important factors in the loading of lipophilic amines, such as vincristine, intovesicles exhibiting a transmembrane pH gradient (ApH; inside acidic) (Mayer et al., 1990a;1990c; 1990d). It is shown that the major factor resulting in improved retention is theinterior pH, where initial interior pH values of 2.0 result in nearly 50% retention at 24 h invivo.493.2 Materials and Methods3.2.1 Lipids and ChemicalsDimyristoylphosphatidylcholine (DMPC), dipalmitoyl PC (DPPC), distearoyl PC(DSPC), diarachidoyl PC (DAPC), and dibehenoyl PC (DBPC) were purchased from AvantiPolar Lipids, while cholesterol and all salts were obtained from Sigma (St. Louis, MO,USA). Vincristine sulfate was obtained from the British Columbia Cancer Agency(Vancouver, British Columbia, Canada).[‘4C]cholesteryl hexadecylether was chosen as a lipid marker due to its stability invivo (Derksen et aL, 1987). It was purchased from New England Nuclear (Ontario, Canada)and was >95% pure. Tritiated vincristine was obtained from Amersham (Oakville, Ontario,Canada).Normal mouse serum was purchased from Cedar Lane Laboratories and femaleBDF1 mice (6-8 weeks old) were purchased from Charles River Laboratories.3.2.2 LiposomesPC/cholesterol (55:45; mol:mol) liposomes were prepared as outlined in Section2.2.2. Each mixture was incubated at or above the transition temperature for each particularPC derivative (65°C for DMPC, DPPC and DSPC; 85°C for DAPC; and 100°C for DBPC).These MLVs were then extruded ten times through two stacked 0.1 jim polycarbonate filters.The extrusion device was heated to the appropriate temperature for each sample.503.2.3 Drug Uptake and Release ExperimentsFor the vincristine uptake experiments, drug was added to the various liposomepreparations to achieve a drug-to-lipid ratio of 0.1:1 (wt:wt). The external pH of the vesicleswas then raised to pH 7.0-7.2 with 0.5 M Na2HPO4and incubated at 37°C over a 4 h period.Aliquots were removed at various time points for determination of vincristine uptake.External untrapped vincristine was removed by running the samples over G-50 Sephadexcolumns prior to dual label scintillation counting of the liposomal fractions contained in thevoid volume.For the drug release studies, vincristine was initially loaded into the liposomes usingthe same procedure as for the drug uptake experiments, except that the samples wereimmediately heated to their lipid transition temperatures for 10 mm. This ensured >95%trapping efficiencies at a drug-to-lipid ratio of 0.1:1 (wt:wt) for all lipid compositions studied(Mayer et al., 1990c). The liposomal vincristine was then diluted 10-fold in either HBS (pH7.4) or normal mouse serum. These samples were dialyzed using SpectralPor 2 dialysistubing against 200 vols. of HBS (pH 7.4) at 37°C. Aliquots were removed at various timepoints, run down G-50 Sephadex columns, and retained vincristine analyzed by dual labelscintillation counting.3.2.4 In Vivo Pharmacokinetics StudiesLiposomal vincristine was injected into BDF1 mice via a lateral tail vein (2 mg/kgvincristine, 20 mg/kg lipid). At varying time points, mice were anaesthetized with i.p.ketamine (160 mg/kg) and xylazine (20 mg/kg). Blood was removed via cardiac puncture51and placed into EDTA-coated microtainer tubes (Becton Dickenson). Samples were thencentrifuged and plasma was analyzed for lipid and vincristine content by dual label liquidscintillation counting.3.2.5 Kinetic Analysis of Vincristine Uptake and ReleaseAn initial-rates treatment of the uptake of lipophilic amino containing drugs intovesicles with an acidic interior has been previously developed for doxorubicin (Harrigan etal., 1993). Assuming that only the neutral form of the drug can cross the lipid bilayer, weconsider first weak bases with a single ionizable group with a dissociation constant (Kd)where Kd << [Hj where [Hi0 is the external proton concentration. Assuming that V0>>Vm, where V0 is the aqueous volume and Vm is the volume of the membrane, it can be shownthat[D(t)]1= [D(eq)]1 (1ekt) (1)where [D(t)] is the interior concentration of the drug at time t, [D(eq)] is the equilibriuminterior concentration at t = infinity and k is the rate constant associated with the uptakeprocess. The rate constant k can be written as (Harrigan et al., 1993)k=PAmK Kd (2)V0[H]where P is the membrane permeability coefficient for the neutral form of the drug, Am is the52area of the membrane, and K is the lipid-water partition coefficient of the drug. A similaranalysis for a drug such as vincristine, which contains two basic functions can be performedunder the assumption that [H’10 >> K1, Kd2 (where Kdl, Kd2 are the dissociation constantsassociated with the two basic groups). It can be shown thatk=PAmK KdlKd2 (3)V0[H]23.3 Results3.3.1 Influence of Acyl Chain Length on Vincristine RetentionIt has previously been shown that both doxorubicin (Mayer et al., 1990a; 1990d) andvincristine (Mayer et al., 1990c) display enhanced retention within liposomes composed ofDSPC-cholesterol as compared to EPC-cholesterol after loading in response to a pH-gradient(inside acidic). One could extend this observation to predict that the presence of longersaturated acyl chain PCs will further improve the retention properties of vincristine. Theuptake and release properties of LUVs composed of DAPC and DBPC, in combination withcholesterol, were therefore investigated.LUVs for drug loading were prepared from mixtures of cholesterol with DMPC (C14),DPPC (C16), DSPC (C18), DAPC (C20), and DBPC (C22). The uptake of vincristine into theseLUVs at 37°C is shown in Fig. 3.1. As expected, the rate of uptake was fastest for theDMPC-cholesterol system, and decreased progressively for DPPC-cholesterol and DSPCcholesterol systems. Surprisingly, this progression was reversed for DAPC- and DBPC53Figure 3.1Vincristine Uptake With Varying Acyl Chain Length.Vincristine uptake across a pH gradient at 37°C for DMPC/Chol (0), DPPC/Chol (O),DSPC/Chol (s), DAPC/Chol (A), and DBPC/Chol (n). Vincristine was added to vesiclepreparations at a potential drug-to-lipid ratio of 0.1:1. Internal pH was 4.0 and external pH was7.5. All vesicles were sized through 100 nm filters. Error bars representing standard deviationsof three trials are too small to be visible.1007O60]./D 503’40)(I,30/51000 60 120 180 240 300 360 420 480TIME (MIN)54cholesterol systems, which exhibited rates of vincristine uptake which increased as the acylchain length increased.Vmcristine release from the same liposomal formulations was also investigated (Fig.3.2). Liposomal vincnstine was incubated at 37°C for 24 h in the presence of buffer andmouse serum. The DMPC-cholesterol system exhibits the most rapid leakage, as expected,whereas the DAPC-cholesterol and DBPC-cholesterol exhibit the best drug retention, withapproximately 40% of the drug remaining at 24 h in the presence of mouse serum.3.3.2 Influence of Internal Buffering Capacity on Vincristine RetentionIt has been previously documented that the accumulation of weakly basic drugs inresponse to a pH-gradient is extremely dependent on the internal buffering capacity(Harrigan et al., 1993). This can easily be explained by the fact that the drugs permeateacross the liposome bilayer in the neutral form and are protonated on reaching the interior,thus consuming a proton and raising the interior pH. This, consequently, will limit theequilibrium uptake of drug. In the case of a drug which contains two basic functions, suchas vincristine, it can be shown that, in the absence of membrane partitioning effects, andassuming that [H10, [H]>> Kdl, Kd2 that[Drug]1 [H] (4)[Drug]0 [H]02and thus the amount of drug entrapped will decrease as the square of the internal protonconcentration as the internal pH rises.55Figure 3.2Vincristine Release as a Function of Acyl Chain Length.Vincristine release over time at 37°C from liposomes incubated in HBS (A) and mouse serum (B)for DMPC/Chol (•), DPPC/Chol (A), DSPC/Chol (s), DAPC/Chol (V), and DBPC/Chol (+).Initial drug-to-lipid ratio was 0.1:1 (wt:wt). Error bars representing standard deviations of threetrials are too small to be visible.10080z60wz400z>20010080z60wzI400z>200A0 4 8 12 16 20 24B0 4 8 12 16 20 24TIME (HOURS)56The drug retention properties of DBPC-cholesterol LUVs with varying internalcitrate concentrations was investigated (Fig. 3.3). Samples were incubated in both buffer andserum at 37°C. As expected, higher initial internal citrate levels resulted in improvedretention, however, these improvements were not significant for internal citrateconcentrations in excess of 400 mM. At internal citrate concentrations of both 400 and 500mM, >50% drug retention was achieved at 24 h when incubated in mouse serum.3.3.3 Influence of Internal pH on Drug RetentionFrom the model outlined in Section 3.2.5, the rate constant for transbilayer movementof vincristine should be proportional to the inverse square of the proton concentration. Wewould therefore predict that the efflux of entrapped vincristine would be significantly slowerif the internal pH is lowered. The dependence of the rate constant k on the pH was initiallyexamined. As shown in Section 3.2.5, when [H]0>> Kdl, Kd2 it is expected that k ct [H902and, thus, log k c 2pH0. Thus, a plot of log k vs. pH0 should result in a straight line with aslope of 2. Uptake of vincristine into DSPC-cholesterol vesicles over the external pH range4-5 was observed. The uptake rates were seen to vary significantly (Fig. 3.4A). The rateconstants k can be calculated from the slopes of the semilogarithmic plots shown in Fig.3.4B, leading to a plot of log k vs. pH0 (Fig. 3.4C) which exhibits a slope of 1.6. Since thefirst pK of vincristine is so low (pK1=5.0), the condition [H]0>> K1 is not well observed.Therefore, there is deviation from the predicted slope of 2.These results predict that vincristine retention within liposomes is extremely sensitiveto the internal pH. Lower internal pH values should dramatically improve drug retention.This prediction was therefore tested by preparing DSPC-cholesterol (55:45) LUVs with57Figure 3.3Vincristine ReLease as a Function of Internal Buffer Salt Concentration.Effect of varying the internal buffer salt concentration on vincristine release from DBPC/Cholvesicles incubated at 37°C in HBS (A) and mouse serum (B). Internal citrate concentrations were500 mM (•), 400 mM (a), 300 mM (A), and 200 mM (V). Internal pH was 4.0 for allpreparations. Initial drug-to-lipid ratios were 0.1:1 (wt:wt). Error bars representing standarddeviations of three trials are too small to be visible.10080z60LiizI-.400z>20010080z60UzI400z>200A0 4 8 12 16 20 24B0 4 8 12 16 20 24AE (HOURS)Figure 3.4The Effects of pH on Vincristine Uptake.(A) Time-course for vincristine uptake into 100 nm DSPC/Chol vesicles fordifferent external pH values. The internal pH for all systems was 3.0. Theexternal pH values were 4.00 (0), 4.25 (Li), 4.50 (Q), 4.75 (v), and 5.00 (c).All samples were loaded at 60°C with a potential drug-to-lipid ratio of 0.1:1.(B) Plot of 1n{([A(eq)J - [A(t)j / [A(eq)],} vs. time, where [A(t)J1 is the internalconcentration of the accumulated drug at time t and [A(eq)]1 is the internalconcentration at equilibrium. The slopes of these lines give the rate constant (k)for the transport of vincristine across the liposome membrane. (C) Plot of logk vs. external pH. The slope of this line is 1.60.59-0.25rc—Q5Q‘z.tL.—0.751.25—1.50—1.75A80 100 120CTIME (Mlii) 80 20 40 60TIME (Miii)—1.00—2.0-2.54.00 4.25604.50 4.75 5.00(TNs’J.. pHinitial interior pH values of 2.0, 3.0,4.0, and 5.0, using 300 mM citrate buffer. As evidencedin Fig. 3.5, the internal pH has a profound effect on vincristine retention. With (initial) pHvalues of 3 or less, there is essentially complete retention of contents for 24-h incubationsin the presence of both buffer and serum. Lowering the internal pH appears to exhibit itsgreatest effect on improving drug retention initially. The slopes of the curves in Fig. 3.5 arenot as markedly different after the 1 h time point.3.3.4 Influence of Temperature on Vincristine UptakeA final variable which would be expected to influence vincristine uptake (and, byextension, release) is temperature. It has been shown elsewhere that weak bases such asdoxorubicin (unpublished data), as well as amino acid and peptide derivatives (Chakrabartiet al., 1992) can exhibit high activation energies for uptake rates in the range of 30 kcallmol.An activation energy of 30 kcallmol corresponds to an uptake rate which increases byapproximately a factor of 5 for every 10°C increase in temperature. Vincristine uptake intoDSPC-cholesterol LUVs was therefore monitored over the temperature range 30-60°C whichresulted in remarkable differences in uptake rates as shown in Fig. 3.6A. An Arrhenius plot(Fig. 3.6C) of the rate constants derived from these data resulted in an activation energy of37 kcallmol.3.3.5 Vincristine Retention in DSPC/Cholesterol LUVs in vivoA basic aim of these studies was to identify parameters which would lead to aformulation of liposomal vincristine which is able to better retain the drug in vivo to allowfor extended circulation lifetime and payout characteristics. It is clear from the studies61Figure 3.5Vincristine Release as a Function of Internal pH.Vincristine release from 100 nm DSPC/Chol vesicles incubated in buffer (A) and mouse serum(B) at 37°C for internal pH of 2.0 (0), 3.0 (•), 4.0 (a), and 5.0 (A). Internal buffer saltconcentration was 300 mM citrate for all systems. Initial drug-to-lipid ratios were 0.1:1. Errorbars representing standard deviations of three trials are too small to be visible.A10090C,z 8070wiLlzI—Cl’ 4030> 20100100____________________________90C,z80z70iLlzI.cn4O>201000 4 8 12 16 20 24H4I’AA0 4 8 12 16 20 24)4E (HOURS)Figure 3.6Vincristine Uptake as a Function of Temperature.(A) Time-course of vincristine uptake into 100 nm DSPC/Chol vesiclesexhibiting a tpH (pH1 = 3.0; pH0 = 5.0). Uptake was conducted at 30 (0), 40(•), 45 (L), and 60°C (A). (B) Plot of ln{([A(eq)] - [A(t)1,) I [A(eq)]1}vs. t,where [A(t)]1 and [A(eq)11 are the same as for Fig. 3.4. (C) Arrhenius plot ofthe rate constants (k) for vincristine uptake. The activation energy calculatedfrom the slope of this plot is 37 kcal/mol.630.070.06t•j —1.0i It-1.5—I.nt-0.5-1.5-2.5.z-3.5£-4-5-5.5-6.5A0 20 40 60E MR4)80 100 120C—7.53.0 3.1 3.2 3.3ut cx 3)64presented above that the internal pH is the most important variable for retention, and thatDSPC/cholesterol LUVs prepared with an (initial) pH1 of 3 or less exhibit retention of 90%or more over 24 h in the presence of buffer or serum. However, it is also known thatliposome leakage in vivo is usually more extensive than in vitro (Mayer et al., 1989). Therelease properties in vivo, of DSPC/cholesterol LUVs with (initial) interior pH values of 2and 4 and loaded with vincristine were assessed by monitoring the drug-to-lipid ratio inplasma (Fig. 3.7). It may be observed that whereas the 90% retention over 24 h obtained invitro was not achieved for the pH1 2 formulation, a value of 40% was achievable. This isapproximately a factor of 5 higher than obtained with the pH1 = 4.0 formulation. It is onceagain apparent, as in vitro, that the most marked difference in vincristine release rates is seenbefore 1 h post-injection. It is of interest that DBPC/cholesterol (55:45) preparations with0.5 M internal citrate at pH1 = 2.0 did not result in improved retention in vivo over theDSPC/cholesterol systems (results not shown). It should be noted that >99% of injected freevincristine (no liposomal carrier) is cleared from the circulation within 5 mm post injection(Mayer et al., unpublished data).3.4 DiscussionThis chapter presents a detailed study of factors leading to improved retention ofvincristine in liposomal systems. These results may also be expected to extend to othermembers of the large class of drugs which are lipophilic weak bases. Here we discuss theinfluence of the experimental parameters investigated on vincristine uptake and release andthe implications for design of liposomal formulations of lipophilic amino containing drugs.Increases in acyl chain length are shown to exhibit the type of retention65C0a0D0Figure 3.7In Vivo Drug Retention.Drug-to-lipid ratios for DSPC/Chol vesicles in vivo with internal pH of 2.0(•) and 4.0 (0). Both systems were loaded with an initial drug-to-lipid ratio of0.1:1. Each point represents the average value obtained from four BDF1 mice. 4 8 12 16 20ThAE (HOURS)2466improvements expected. Thus, the half-times for vincristine release at 37°C in bufferincrease from approximately 1 h for DMPC/Chol LUVs to approximately 12 h for DAPCand DBPC-containing systems. The uptake of vincristine into these systems exhibitsanomalous behavior in that the rates of uptake first decrease, as expected, as acyl chainlength is increased to 18 carbons (DSPC), and then increase markedly for the DAPC andDBPC systems. It is possible that this reflects an increased lipid-water partition coefficientfor vincristine for the outer monolayer of the DAPC and DBPC systems. An increasedhydrophobicity of this interface would be consistent with packing effects expected for longerchain lipids in small vesicular systems, and this is reflected by the increased tendency of thelonger chain DAPC and DBPC LUV systems to aggregate after extrusion. Conversely, forthe inner monolayer, it would be expected that these effects would result in tighter packingin the headgroup region, and correspondingly reduced leakage, as shown experimentally.With regard to interior buffer capacity, the results presented here show that increasingthe interior citrate concentration above 400 mM does not result in significant improvementsin vincristine retention. This is consistent with the osmotic properties of extruded LUVsystems. As shown for 100 nm egg PC/cholesterol systems (Harngan et aL, 1993), extrudedLUVs exhibit tubular ‘sausage’ shapes and respond to osmotic gradients (high interiorosmolar.ity) first by ‘rounding up’ to increase the interior volume and subsequently undergoosmotically induced lysis. For EPC/Chol LUVs the ‘effective’ osmotic difference that canbe sustained is approximately 650 mosmol/kg. The osmolality of 400 mM citrate, pH 4.0,is 700 mosmol/kg, resulting in an effective osmotic imbalance of 400 mosmollkg when theliposomes are in normal saline solution (osmolality 300 mosmol/kg). This initial effectiveosmotic imbalance will increase on drug loading. Vincristine, after crossing the liposomal67bilayer in its neutral form, becomes protonated in the vesicle interior due to the low internalpH, consequently raising the internal osmolality.Lower internal pH values are clearly critical for improving vincristine retention inliposomal systems. This is due to the dependence of the rate constant for vincristinemovement on the inverse square of the proton concentration (Eqn. 3). This predicts a 100-fold reduction in leakage rates for every unit the interior pH is lowered.In conclusion, the internal pH is the most important parameter in enhancingliposomal vincristine retention both in vitro and in vivo. These effects can likely be extendedto enhance retention of other basic drugs in liposomes in vitro and in vivo.68CHAPTER 4THE USE OF CATIONIC LiPIDS TO IMPROVE LIPOSOMAL VINCRISTINERETENTION4.1 IntroductionSeveral mechanisms for improved vincristine retention within liposomes wereexamined in Chapter 3. Of these, lowering the internal pH to 2.0 had substantial effects onimproving drug retention both in vitro and in vivo. This improved retention may be partiallydue to the fact that at pH 2.0, the phospholipid phosphate groups are nearing their P”a value.The phosphatidyicholine molecules wifi thus exhibit some positive charge, creating a chargerepulsion with the positively charged vincristine molecules on the vesicle interior.In order to further develop the hypothesis that lowering the internal pH increasesdrug retention by a charge repulsion effect, it was decided to incorporate cationic lipids intothe liposomal membrane to determine their effect on vincristine retention. A potentialproblem with this approach is that liposomes containing positively charged lipids arerecognized more readily by the RES and are cleared from the circulation more rapidly thanneutral lip osomes (Senior et al., 1991). It was necessary, therefore, to utilize positivelycharged lipids that could be induced to reside only in the inner monolayer of the liposomes,thereby preventing their recognition by the RES. It has previously been shown that weaklybasic simple lipids such as stearylamine and sphingosine can rapidly and completely migrateto the inner monolayer in the presence of a transbilayer pH-gradient (inside acidic) (Hopeand Cuffis, 1987; Hope et al., 1989; Eastman et aL, 1989; Eastman et al., 1991). Since theseare the same conditions necessary for loading the weakly basic drug into the vesicles (Mayeret al., 1990c), one would expect to see the migration of these lipids to the inner monolayer69as the drug is loaded into the vesicles.In this Chapter, it is demonstrated that the incorporation of positively charged lipidscan indeed prolong vincristine retention within liposomes and that this incorporation ofweakly basic lipids has no effect on liposome circulation time in vivo.4.2 Materials and Methods4.2.1 Lipids and ChemicalsDSPC was obtained from Avanti Polar Lipids. Cholesterol, stearylamine, andsphingosine were purchased from Sigma Chemical Company (St. Louis, MO). rac 1,2-dioleoyl-3-N,N-dimethylaminopropane (AL-i) was synthesized in our laboratory usingstandard procedures (Bailey and Cullis, 1994). Vincristine sulfate was obtained from AdriaLaboratories of Canada. [‘4C]-cholesteryl hexadecylether was specially synthesized for usby Amersham (Oakvffle, Ontario). It was chosen as a lipid marker since it is not exchangedor metabolized in vivo (Derksen et al., 1987).[3H]-vincristine was also purchased fromAmersham. Female BDF1 mice (18-22 g) were obtained from Charles River Laboratories.4.2.2 LiposomesDSPC/Chol (55:45; mol:mol), DSPC/ChoIISA (45:45: 10; mol:mol:mol),DSPC/Cho1IAL- 1 (45:45:10; mol:mol:mol), or DSPC/Chollsphingosine (45:45:10;mol:mol:mol) solutions were prepared by the same procedure used in Section Drug Entrapment ProcedureVincristine was entrapped in the liposomes using the ApH loading procedure70described elsewhere (Mayer et al., 1990c). Liposome preparations were run over G-25Sephadex columns equilibrated with HBS (pH 7.4) to achieve a pH-gradient across thevesicle membranes. The vesicles were then added to vincristine (vincristine sulfate solution,1 mg vincristine/mL) to achieve a drug-to-lipid ratio of 0.1:1. The resulting exterior pH ofthe liposome/vincristine mixture was raised to pH 7.4 with 0.5 M Na2HPO4and immediatelyheated to 60°C for 10 mm. This procedure ensured >95% trapping efficiencies in all cases.4.2.4 Pharmacokinetic StudiesPharmacokinetic studies were performed by injecting liposomal vincristine intoBDF1 mice via a lateral tail vein (2 mgfkg vincristine, 20 mg/kg lipid). It has previouslybeen shown that this dose of liposomal vincristine exhibits measurable levels of antitumoractivity in the P388 ascites tumor model (Mayer et a!., 1990c). At varying time points, micewere anaesthetized with i.p. ketamine (160 mg/kg) and xylazine (20 mg/kg). Blood wasremoved via cardiac puncture and collected in EDTA-coated microtainer tubes (BectonDickenson). The samples were then centrifuged (500 x g for 10 mm.) to pellet the bloodcells and obtain plasma samples. Liposomal lipid and/or vincristine were then assayed usingdual label scintillation counting.4.2.5 Biodistribution StudiesBiodistribution studies were performed on the same mice used for pharmacokineticstudies. Following heart puncture, mice were terminated by cervical dislocation, and theliver and spleen were removed from each animal and weighed. 10% (w/v) homogenates insaline were achieved using a Polytron homogenizer (Brinkmann Instruments, Rexdale,71Ontario). 500 1iL of each homogenate was digested with 500 pL of “Solvab1e (DuPontCanada, Inc., Mississauga, Ontario) for 1 h at 50°C. After cooling to room temperature, thesamples were decolorized with 200 juL of 30% hydrogen peroxide and maintained at 4°Covernight to prevent excessive foaming. Samples were then counted using Ultima Gold(Packard) scintillation cocktail. The statistical significance of both the plasma clearance andbiodistribution results were determined using the student’s t-test.4.2.6 Antitumor StudiesThe antitumor effects of liposomal vincristine were monitored using the P388lymphocytic leukemia model. BDF1 mice (5 per group) were injected i.p. with 1 x i05 P388cells. The indicated doses of saline or liposomal vincristine were administered (i.v.) 24 hafter tumor inoculation. Animal weights and mortality were monitored daily. Mediansurvival times as well as the statistical significance of the results were determined using theMann-Whitney-Wilcoxon procedure.4.3 Results4.3.1 Plasma Clearance and Biodistribution StudiesPrevious work has shown that pH-gradients across liposomal membranes (insideacidic) induce the complete migration of cationic lipids such as sphingosine and stearylamineto the inner monolayer (Hope and Cullis, 1987). It has also been shown that the presenceof cationic lipids on the liposome surface decreases blood circulation times (Senior et al.,1991). For these reasons, the plasma clearance of empty DSPC/Chollsphingosine (45:45:10;mol:mol:mol) vesicles in the presence or absence of a pH-gradient was investigated (Figure724.1). Vesicles with an interior and exterior pH of 7.5 had more rapid clearance times thanvesicles with an interior pH of 4.0 and exterior pH of 7.5 (ApH = 3.5). This was particularlyevident at the 1 hr time point indicating that the presence of cationic lipid on the liposomesurface increases initial clearance rates. The presence of the pH-gradient drives sphingosinemigration to the inner monolayer, creating an asymmetric liposomal membrane whichexhibits essentially identical liposome clearance times to vesicles composed solely ofDSPCICho1 (55:45; mol:mol).The previous chapter demonstrated that lowering the internal pH to pH, 2.0 ofDSPCICho1 (55:45; mol:mol) liposomes significantly improves vincristine retention both invitro and in vivo. This is, in part, due to the fact that a larger proportion of vincristinemolecules remain protonated (and hence, positively charged) on the interior of theliposomes. It was thought that if a small percentage of cationic lipids were incorporated intothe liposomal membrane, this would repel the positively charged vincristine molecules,preventing their partitioning into the inner monolayer and, therefore, leakage from thevesicles.The data presented here strongly indicate that the incorporation of cationic lipidsincrease vincristine retention within DSPCICho1 (55:45; mol:mol) liposomes in vivo.Liposomes containing 10% stearylamine result in a substantial increase in vincristinecirculation times with an internal pH of 2.0 (Fig 4.2A) without changing liposome clearancetimes (Fig 4.2B). This is.due to a significant increase in liposomal retention of drug asevidenced by an increase in drug-to-lipid ratios over a 24 h period (Fig 4.2C). At 24 h, thereis approximately a 6-fold higher drug-to-lipid ratio for the stearylamine-containingliposomes at pH 2.0 than for the DSPC/Chol (55:45; mol:mol) liposomes at pH 4.0. There73criaE00U0-Jc,)Figure 4.1Plasma Clearance of Sphingosine-Containing VesiclesPlasma clearance of DSPC/Chol/sphingosine empty liposomes with internal pHof 7.5 (•) and internal pH of 4.0 (‘ and DSPC/ChoI pH1 7.5 (A). Error barsrepresent standard deviations of four mice.4540353025201510500 4 8 12 16 20 24TIME (HOURS)74Figure 4.2Influence of Stearylamine and Internal pH on Lipid and DrugClearance In Vivo.Vincristine clearance (A), liposome clearance (B), and drug-to-lipid ratios (C)were determined following i.v. administration in BDF1 mice of DSPC/Chol pH14.0 (0), DSPC/Chol pH1 2.0 (•), DSPC/ChoI/SA pH1 4.0 (L), andDSPC/Chol/SA pH1 2.0 (A). Vincristine was encapsulated at a drug-to-lipidratio of 0.1:1 (wt:wt). Error bars represent standard deviations of four mice.75x 0 C (I,N 0DRUG—TO—UPIDRA11O000000000000000000000-’0-NCACO.JCD(D0UPID(jsg/100p1PLASMA)-‘-‘NMCAC1CnppnpcnenOLno0000000000oN 0VINCRIS11NE(pL/100jLPLASMA)00.NNCA0U0Ui0Ui00-a N -a 0)-a N0)N 00)NwNNis also a 2-fold increased drug retention over the DSPCICho1 pH 2.0 vesicles at 24 h.Essentially identical results are seen with the incorporation of 10 mol% AL-i orsphingosine (Fig 4.3). The presence of cationic lipid substantially improves drug retentionat pH 2.0 but not at pH 4.0. Liposome clearance times are not affected by the addition ofthese lipids (results not shown).Since sphingosine is a naturally occurring lipid found in biological membranes(Sabbadini et at, 1993; Tao et at, 1973; Wang and Schick, 1981), it was decided to performfurther in vivo experiments solely with this lipid. Biodistribution studies were performed towitness any differences in liposomal drug accumulation in the liver and spleen, two majororgans of the RES.Liposome accumulation in the liver was found to be independent of internal pH andindependent of liposome composition (results not shown). However, due to the increaseddrug retention seen with the sphingosine-containing pH1 2.0 preparation, there aresubstantially higher drug levels in the liver at the 24 h time point (Fig 4.4A). Interestingly,despite seeing the highest drug accumulation in the spleen at 24 h for the sphingosinecontaining pH 2.0 preparation (Fig. 4.4B), we see similar liposome uptake for all vesiclecompositions. Any differences at the 1 and 4 h time points are insignificant.4.3.2 Antitumor Activity of Liposomal VincristineIt was decided to compare the antitumor activity of the DSPC/ChoLfsphingosine pH2.0 formulation to that of the DSPCICho1 pH 4.0 preparation. Consistent with previousresults (Boman et al., 1994; Mayer et al., i990c), both liposomal formulations of vincristinewere significantly more efficacious in the P388 tumor model when compared to free drug77Figure 4.3Influence of AL-i or Sphingosine on Vincristine Retention In Vivo.Drug-to-lipid ratios were determined with the incorporation of 10 mol% AL-i(A) or 10 mol% sphingosine (B). Vincristine retention was determinedfollowing i.v. administration in BDFI mice for both systems at pH1 4.0 (0) andat pH1 2.0 (•). Vincristine was encapsulated at a drug-to-lipid ratio of 0.1:1(wt:wt).0.100.09o 0.08I—°° 4 8 12 16 20 24B0 4 8 12 16 20 24TIME (HOURS)78Figure 4.4Liver and Spleen Accumulation of Liposomal Vincristine.Drug accumulation in the liver (A) and spleen (B) were determined 24 hoursafter i.v. injection in BDF1 mice. Error bars represent standard deviations offour mice.ALUDC’)CoI—I—UD0C)0I IB0(f)0.U00.C’JC-)C’,U0C)C 0C.0C-)00Cl)0C)COc’iC.cj,Z-cC.)Cl)U79(Table 4.1). Most interestingly, the combined incorporation of sphingosine into liposomesand the lowering of the internal pH to 2.0 significantly increased the antitumor activity ofvincristine against P388 leukemia when compared to DSPC/Chol liposomal vincristine (pH14.0) at drug dosages of 2, 3, and 4 mg/kg (p<O.Ol at all drug dosages). There was also adecrease in drug toxicity, as judged by drug-induced weight loss, for theDSPC/Chollsphingosine (pH1 2.0) vesicles compared to DSPC/Chol vesicles (pH 4.0). TheDSPC/ChoLfsphingosine (pH1 2.0) formulations at drug dosages of 3 and 4 mg/kg producedlong-term survivors with median survival times of >60 d. When these surviving mice wereagain challenged with i.p. inoculation of 1 x i05 P388 cells on day 62, they exhibited mediansurvival times similar to the control group (data not shown).4.4 DiscussionThe results presented in this Chapter demonstrate that in the presence of a pH-gradient (inside acidic), catiomc lipids such as sphingosine are completely transferred to theinner monolayer of DSPC/Chol vesicles, creating a liposomal clearance pattern essentiallysimilar to vesicles containing no cationic lipid. It has also been demonstrated that vincristineretention within liposomes is greatly increased by the simultaneous incorporation ofsphingosine in DSPC/Chol vesicles as well as lowering the internal pH to 2.0. Further, thisincreased drug retention results in a substantial increase in vincristine circulation time.Finally, the large increase in drug circulation time seen with the sphingosine-containing pH2.0 preparation results in a significant increase in antitumor activity over liposomalvincristine without sphingosine (p<O.Ol at all drug dosages). Each of these observationsshall be discussed below.80TABLE 4.1P388 Antitumor Activity of Free and Liposomal Vincristine in BDF1 MiceDrug Lipid % wt 60-day Median %ILS LJFbSample Dose Dose change Survival Survival(mg/kg) (mg/kg) on day 7 (days)Saline + 14.4 0/15 10.0controlFree 2.0 + 6.0 0/10 14.0 40vine 3.0- 3.6 0/10 12.0 204.0- 29.8 0/5 8.5 -15DSPC/Chol 2.0 20- 2.1 0/10 27.0 170 1.93pH 4.0 3.0 30- 12.0 0/10 31.0 210 2.58Lipovinc 4.0 40- 24.9 0/10 32.0 220 3.76DSPC/Chol/sphingo 2.0 20 + 1.8 1/5 36.0 260 2.57pH 2.0 3.0 30- 7.0 3/5 >60.0 ND NDLipovinc 4.0 40- 18.5 3/5 >60.0 ND ND3Percentage of ILS values were determined from median survival time comparing treated andsaline control groups. If greater than 50% of the animals survived for greater than 70 days,median survival times and %IL.S were not calculated.“LJF (liposomal/free) values were calculated by dividing the median survival time for theliposomal vincristine group by the median survival time for the equivalent dosage of free drug.‘ND. - not determined.81As shown in Chapter 3, when the internal pH of liposomes is lowered to pH, 2.0,there is a dramatic increase in vincristine retention both in vitro and in vivo. At this pH, theheadgroups of the phosphatidyicholine molecule is beginning to become positively charged,perhaps creating a charge separation between the inner monolayer of the liposome and thepositive charge on the vincristine molecule at low pH. In order to examine this hypothesis,the incorporation of cationic lipids into liposomal membranes was investigated. A potentialproblem with this approach is that cationic lipids have been shown to increase liposomeclearance from the circulation (Senior et al., 1991). Therefore, it was necessary to utilizecationic lipids which could be induced to migrate to the liposome inner monolayer,producing an asymmetric vesicle membrane.Previous work has demonstrated that positively charged lipids such as sphingosineand stearylamine can be induced to migrate to the inner monolayer in the presence of a pH-gradient (inside acidic) (Hope and Cullis, 1987; Hope et al., 1989; Eastman et al., 1989;Eastman et al., 1991). We examined this effect on the in vivo clearance times ofsphingosine-containing vesicles in the presence and absence of a pH-gradient across theliposome bilayer. It is clearly evident that in the presence of a pH-gradient, the vesiclesdisplay an essentially identical clearance pattern to DSPC/Chol vesicles containing nocationic lipid, indicating that the sphingosine is completely migrating to the inner monolayerof the vesicle.When the internal pH of cationic lipid-containing vesicles is decreased from pH1 2.0,there is a substantial increase in drug retention in vivo. This is likely due to the fact that thestrong positive charge on the sphingosine molecules (and stearylamine or AL-i) at the inner82monolayer at this pH (PKa = 5.30 for sphingosine) repels the positive charge on thevincristine molecules on the liposome interior. This charge separation prevents the drugfrom partitioning into the inner monolayer of the liposome, the first step necessary fordiffusion across the lipid bilayer to the exterior of the vesicle.It is of interest to note that with an internal pH of 4.0 there is no increase in drugretention with the incorporation of stearylamine, AL- 1, or sphingosine. With thesimultaneous movement of cationic lipids to the inner monolayer and the loading ofvincristine across a pH-gradient, there is a greater drop in the internal proton concentrationthan for the loading of drug alone. This increase in the internal pH allows for not only ahigher percentage of the vincristine molecules to be in their neutral form, but also for ahigher percentage of the sphingosine (or other cationic lipid) molecules to be in the neutralform. This results in a lower charge density for the liposome inner monolayer, being lessable to repel the positive charges on the vincristine molecules. This allows the vincristineto more easily partition into the inner monolayer and diffuse across the membrane into theexternal compartment.All liposome compositions studied displayed essentially identical lipid clearancetimes in vivo. Since the liver and spleen are major organs of the RES, one would expect allliposomal systems studied to have similar patterns of uptake in these organs. This is indeedthe case. Neither the lowering of the internal pH to 2.0 nor the incorporation of sphingosineinto DSPCICho1 liposomes appears to increase liposome recognition by the cells of the RES.The pattern of drug uptake by the liver and spleen is quite different. The vincristineuptake by these organs is determined by two parameters: the liposome uptake by theseorgans and the drug retention within the liposomes. Since its shown that liposome uptake83is essentially identical for all four preparations studied, any differences should be attributedto vincristine retention within the liposomes. Therefore, we would expect to see the highestdrug accumulation in both organs for the sphingosine-containing pH 2.0 preparation. Thisis indeed the case as we see a 2 to 3-fold increase in drug accumulation in both the liver andspleen at 24 h for the sphingosine-containing pH1 2.0 preparation.By analogy with previous studies with liposomal anticancer agents (Mayer et al.,1990b; 1990c; Bally et al., 1990; Mayer et al., 1993; Vaage et al., 1993), increased drugconcentration in the plasma would be expected to result in an increase in antitumor activity.Such an effect has been seen previously for P388 and L1210 leukemia when the vincristinecirculation time was increased by encapsulation within DSPCICho1 vesicles (Mayer et al.,1990c). The encapsulated drug displayed increased antitumor activity over free drug as wellas drug encapsulated in leaky liposomes (eggPCIChol). The increased plasma drugconcentrations observed here by incorporating sphingosine into vesicle membranes anddecreasing the internal pH results in a substantial increase in antitumor activity in the P388leukemia model. This is particularly evident at drug dosages above 3 mgfkg. At thesedosages, we see long term survivors with median survival times in excess of 60 days. Thus,the benefits from extended circulation times are clearly evident.The DSPC/Chollsphingosine pH 2.0 system also displays a considerable decrease intoxicity as evidenced by decreased weight loss. This is possibly due to the fact that the drugis more tightly held within the liposome, preventing toxic side effects.In summary, this Chapter has demonstrated that, in the presence of a pH-gradient(inside acidic), sphingosine, stearylamine, or AL-i can be induced to migrate to the innermonolayer of liposomes, where, at low enough pH, the increase in the charge density of the84vesicle inner monolayer repels the positively charged vincristine molecules, reducing drugleakage. This results in substantially increased drug circulation times. The systemcontaining sphingosine at pH1 2.0 displays dramatic improvement in therapeutic activityagainst the P388 leukemia cell in in vivo resulting in long term survival.85CHAPTER 5THE USE OF MONOSIALOGANGLIOSIDE GM1 TO INCREASE LIPOSOMECIRCULATION LONGEVITY AND IMPROVE LIPOSOMAL VINCRISTINERETENTION5.1 IntroductionAs previously mentioned, liposomal formulations of vincristine can exhibit reducedtoxicity and enhanced efficacy compared to free drug (Mayer et al., 1990c). This has beenrelated to the enhanced residence times of the drug in the circulation which is achieved whenusing liposomal carriers. The incorporation of monosialogangliosides (GM1) andpolyethyleneglycol (PEG) derivatives can further enhance liposome circulation time(Gabizon et al., 1988; Klibanov et al., 1991; Allen et al., 1989; Senior et al., 1991a; Liu etal., 1990; Klibanov et al., 1990; Allen et al., 1987; Allen et al., 1991b). Incorporation ofthese lipids reduces protein/liposome association (Chonn et aL, 1992), and results in adecreased uptake by the reticuloendothelial system (RES) (Allen et al., 1987; Allen et al.,1991b). They also have been shown to decrease macrophage uptake when coated oncolloidal particles (ifium et aL, 1986). An added advantage is that there tends to be increasedaccumulation within tumors for liposomes which exhibit longer circulation residence times(Gabizon et al., 1988; Gabizon et al., 1983; Fichtner et al., 1981; Richardson et al., 1978;Proffitt et al., 1983; Gabizon et al., 1990). However, until recently, such extendedcirculation lifetimes have been of questionable value for liposomal vincristine preparationssince approximately 85% of the entrapped vincristine is released from DSPC/Chol pH 4.0liposomes in the blood within 24 h of i.v. administration.In Chapter 3, we identified conditions which may circumvent the drug leakageproblem cited above (Boman et al., 1993). Specifically, decreasing the pH of the entrapped86citrate buffer to 2.0 dramatically reduces vincristine leakage in the presence of serum invitro. In view of this development, we have investigated the pharmacokinetic,biodistribution, and efficacy properties of vincristine entrapped in DSPC/Chol liposomescontaining pH 4.0 and pH 2.0 citrate buffers. Further, we have examined the effects ofincluding the ganglioside GM1 in the liposomes on the in vivo properties of these systems.These studies reveal that combining the effects of decreased entrapped pH and including GM1synergistically enhance the phannacological and therapeutic activity of liposomal vincristine.5.2 Materials and Methods5.2.1 Lipids and Chemicals??Oncovinfl (vincristine sulfate) was obtained from the B.C. Cancer Agency(Vancouver, British Columbia, Canada). DSPC was purchased from Avanti Polar Lipids,and was greater than 99% pure. Monosialoganglioside GMI, cholesterol, and all salts wereobtained from Sigma Chemical Company (St. Louis, MO). Cholesteryl hexadecylether (‘4C)was specially synthesized for us by Amersham (Oakville, Ontario). Female BDF1 mice (6-8weeks old) were purchased from Charles River Laboratories.5.2.2 LiposomesDSPC/Chol (55:45; mol:mol) or DSPC/Ch0IIGM1(45:45:10; mol:mol:mol) wereprepared by the same procedure outlines in Section Drug Entrapment ProcedureVincristine was entrapped in the liposomes using the ApH loading procedure87described in Section Pharmacokinetic StudiesPlasma clearance studies were performed as outlined in Section Biodistribution StudiesBiodistribution studies were performed as outlined in Section 4.2.5 except that thelung and kidney were also removed from the animals.5.2.6 Antitumor StudiesThe antitumor effects of liposomal vincristine were monitored using the P388lymphocytic leukemia model. BDF1 mice (5 per group) were injected i.p. with 1 x i05 P388cells. The indicated doses of saline or liposomal vincristine were administered (i.v.) 24 hafter tumor inoculation. Animal weights and mortality were monitored daily. Mean andmedian survival times as well as the statistical significance of the results were determinedusing the Mann-Whitney-Wilcoxon procedure.5.2.7 Solid Tumor Loading and Efficacy StudiesTumor loading of liposomal vincristine was determined in BDF1 mice using theLewis Lung tumor model. 3 x i05 tumor cells were injected subcutaneously above the hindleg. Free vincristine, DSPC/Chol pH 4.0 liposomal vincristine, and DSPC/CholJGMl pH 2.0liposomal vincristine were injected i.v. 7 to 9 days after inoculation of Lewis Lung tumorcells. At varying time points, tumors were excised, digested, and lipid and drug contents88were determined by liquid scintillation counting. Efficacy studies were performed by dailymonitoring of tumor size following drug injection.5.3 Results5.3.1 Plasma Clearance and In Vivo Drug Release StudiesIt was shown in Chapter 3 that the retention of vincristine entrapped insideDSPC/Chol liposomes in response to a transmembrane pH gradient (inside acidic) could beimproved by decreasing the pH of the intravesicular citrate buffer below that previouslyemployed for such systems (300 mM citrate, pH 4.0) (Boman et al., 1993). Theseobservations were expanded on here to determine whether such in vitro results translated toincreased circulation lifetimes for vincristine in vivo. Also,G1-containing liposomes wereemployed to determine whether increasing the circulation lifetime of the liposomal carriersystem could provide additional improvements in the longevity of vincristine in the bloodcompartment after i.v. administrationFigure 5. 1A demonstrates that decreasing the interior pH of DSPC/Chol 100 nmliposomes from 4.0 to 2.0 has a modest effect on circulating vincristine levels over 24 h posti.v. injection (2 mg/kg drug, 20 mg/kg lipid) to BDF1 mice. This effect appears to be relatedprimarily to the enhanced ability of the pH1 2.0 liposomes to retain entrapped vincristine.This is evidenced by the fact that circulating liposomal lipid levels for these two systems arecomparable over the time course studied (Figure 5. 1B) whereas the circulating drug-to-lipidratio is increased for the pH 2.0 system 1.2, 1.3, and 9.1-fold over the pH1 4.0 system at 1h, 4 h, and 24 h respectively (Figure 5. 1C).Inclusion of 10 mol% GM1 in the DSPC/Chol liposomes with an entrapped pH1 4.089Figure 5.1Influence of GMI and Internal pH on Lipid and Drug ClearanceIn Vivo.Vincristine clearance (A), liposome clearance (B), and drug-to-lipid ratios (C)were determined following i.v. administration in BDF1 mice of DSPC/Chol pH4.0 (•), DSPC/Chol pH1 2.0 (v), DSPC/Chol/GM1 pH1 4.0 (•), andDSPC/Chol/GM1pH1 2.0 (A). Vincristine was encapsulated at a drug-to-lipidratio of 0.1:1 (wt:wt). Error bars represent standard deviations of four mice.900) I’) 0DRUG-TO-LIPIDRATIOp0000000000oobobàoooö-’o-rC.).0)0)-I0)(DOLIPID(pg/i00pLPLASMA)-C.)00000U’ 0F’) 0VINCRISTINE(pg/lOGpLPLASMA)PP-‘F’)))00)00)0(SC0(TI0(Il000m 0 C ;i:i CI)0)F’)1\) 00)0wF’)I\)citrate buffer results in circulating vincristine levels that are increased compared to thecanier system in the absence of GM1 and equivalent to those observed employing DSPCICho1liposomes with the pH 2.0 entrapped buffer (Figure 5.1A). In this case the increased plasmavincristine concentrations arise from the extended circulation lifetime of the DSPC/Ch0IJGM1liposomes. Plasma liposomal lipid levels are increased approximately 2.5-fold 24 h after i.v.administration when GM1 is incorporated into DSPCICho1 liposomes prepared at pH1 4.0(Figures 5. 1B). Corresponding circulating drug-to-lipid ratios for these two systems are verysimilar over the time course studied (Figure 5. 1C).Incorporation of GM1 into the vesicle membrane in combination with the use of thepH 2.0 entrapped citrate buffer resulted in unexpected phannacokinetic effects for both thelipid and drug components. Plasma vincristine levels for this system were approximately1.5-fold higher than the DSPC/Chol pH1 2.0 and DSPC/Ch0LIGM1pH1 4.0 preparations at 1and 4 h after i.v. injection and approximately 6-fold higher at 24 h (Figure 5. 1A). As maybe expected, the GM1 pH 2.0 liposomes display enhanced circulation longevity compared tosystems devoid of GM1 (Figure 5. 1B), however, significantly lower liposomal lipidconcentrations than the GM1 pH1 4.0 preparation are seen over 24 h (p<O.Ol at all timepoints). Consequently, the increased plasma vincristine levels arise from the ability of theGM1 pH 2.0 liposomes to retain vincristine in the circulation. This is shown in Figure 5.1Cwhere, in contrast to the pH 4.0 systems, pH1 2.0G1-containing liposomes displaycirculating drug-to-lipid ratios that are 1.3-, 1.4-, and 2.2-fold higher than observed for theDSPC/Chol pH1 2.0 preparation at 1 h, 4 h, and 24 h after injection, respectively. This ratiois increased to 7.2-fold at 24 h when compared to the pH 4.0 liposomal systems.The increase in circulation longevity of the liposomes is reflected by a decreased92uptake in the liver and spleen, two major organs of the RES. This is illustrated in Figures5.2B and 5.3B where it is shown thatG1-containing liposomes are accumulated in theseorgans much less than the DSPCICho1 liposomes. The data indicate significant differences(p.<O.O5) at 24 h in the liver and for all time points in the spleen (p<O.Ol). Lung and kidneyuptake were identical for all liposomal compositions (data not shown). Lowering the internalpH of the DSPC/Ch0IJGM1vesicles to pH1 2.0 increases both liver and splenic uptake(p<O.005) as would be expected by the decreased circulation time of these liposomes.Lowering the internal pH of the DSPCICho1 vesicles, however, does not appear to affecttheir biodistribution.For liposomal drug preparations at pH 4.0, there was increased vincristine uptake byboth the liver and spleen in the absence of GM1 (p.<O.0O5 for all time points) (Figures 5.2Aand 5.3A). The opposite effect, however, was seen for the pH1 2.0 preparations. For theDSPC/Ch0IJGM1systems, lowering the internal pH to 2.0 results in increased liver andsplenic drug accumulation. This is likely due to the high drug-to-lipid ratios seen for thisformulation. Again, no significant differences were seen in uptake by the lung or kidney(data not shown).5.3.2 Antitumor Activity of Liposomal VincristineTumor efficacy studies were conducted to determine whether the pharmacokineticeffects described above influenced the antitumor activity of liposomal vincristine. Theantitumor activity of the DSPC/Ch0IJGM1pH 2.0 formulation was compared with that of theDSPC/Chol pH 4.0 preparation. Consistent with previous results (Mayer et al., 1990c), bothliposomal formulations of vincristine were significantly more efficacious in the P388 tumor93110 -100 -90 -80-Cl)Cl)F-Ui 60-50-40--J) 3020100-Figure 5.2Liver Accumulation of Liposomal Vincristine.Drug accumulation (A) and lipid accumulation (B) were determined followingi.v. administration in BDF1 mice of DSPC/ChoI pH1 4.0 (0), DSPC/Chol pH1 2.0(D), DSPC/Chol/GML pH1 4.0 (•), and DSPC/Chol/GMI pH1 2.0 (a). Vincristinewas encapsulated at a drug-to-lipid ratio of 0.1:1 (wt:wt). Error bars representstandard deviations of four mice.87LUD6U)Cl)I—Ui0D3100 4 8 12 16 20 24B0 4I I8 12 16 20 2494TIME (HOURS)Figure 5.3Spleen Accumulation of Liposomal Vincristine.Drug accumulation (A) and lipid accumulation (B) were determined followingi.v. administration in BDF1 mice of DSPC/Chol pH1 4.0 (0), DSPC/Chol pH1 2.0(c), DSPC/Chol/GMI pH1 4.0 (•), and DSPC/ChoI/GMI pH1 2.0 (•). Vincristinewas encapsulated at a drug-to-lipid ratio of 0.1:1 (wt:wt). Error bars representstandard deviations of four mice.15 -u-IDa,a,I— 10 -LUC)C,D5-C)A4812I---16 20 240-0600-500-LUD400 -I—I-u.i300-200 -100 -0-BI I I I0 4 8 12 16 20 24TIME (HOURS)95model when compared to free drug (Table 5.1) (p<O.Ol for all drug dosages). Mostinterestingly, the combined incorporation of GM1 into liposomes and the lowering of theinternal pH to 2.0 significantly increased the antitumor activity of vincristine against P388leukemia when compared to DSPC/Chol liposomal vincristine (pH1 4.0) at drug dosages of2, 3, and 4 mg/kg (p<O.Ol). There was no improvement in therapeutic activity, however, atthe 1 mg/kg drug dose. There was also a decrease in drug toxicity, as judged by drug-induced weight loss, for the DSPC/Chol/GM1(pH1 2.0) vesicles compared to DSPC/Cholvesicles (pI-l 4.0). The DSPC/Ch0IIGMI (pH1 2.0) formulations at drug dosages of 2, 3, and4 mg/kg all produced long-term survivors with median survival times of >70 d. When thesesurviving mice were again challenged with i.p. inoculation of 1 x i05 P388 cells on day 72,they exhibited median survival times similar to the control group (data not shown).5.3.3 Solid Tumor Loading and Efficacy StudiesVincristme loading in the Lewis Lung tumor as well as tumor efficacy studies wereperformed to determine any correlations which may be present. Lewis lung tumor loadingof free vincristine, DSPC/Chol pH 4.0 liposomal vincristine, and DSPC/Ch0IJGM1pH1 2.0liposomal vincristine was determined. Both preparations of liposomal drug displayedsignificantly higher levels of tumor drug uptake than when free drug was administered(Figure 5.4A). Free vincristine initially accumulated in the solid tumor before decreasingto a level near zero 24 h following i.v. administration. In contrast, when vincristine wasencapsulated within liposomes composed of DSPC/Chol pH1 4.0, much higher tumor levelsof drug were achieved initially as well as a much slower release of drug from the site oftumor growth resulting in drug levels nearing zero at approximately 72 h following i.v.96TABLE 5.1P388 Antitumor Activity of Free and Liposomal Vincristine in BDF1 MiceSample Drug Lipid % wt 60-day Median %ILS L/F’Dose Dose change survival survival(mg/kg) (mg/kg) on day 7 (days)Saline + 14.4 0/15 10.0controlFree 1.0 + 1.9 0/5 14.0 40Vinc 2.0 + 6.0 0/10 14.0 403.0- 3.6 0/10 12.0 204.0 -29.8 0/5 8.5 -15DSPC/ 1.0 10 + 2.8 0/5 22,0 120 1.57chol 2.0 20 - 2.1 0/10 27.0 170 1.93pH 4.0 3.0 30 -12.0 0/10 31.0 210 2.58Lipovinc 4.0 40 -24.9 0/10 32.0 220 3.76DSPC/ 2.0 20 0.0 2/5 31.0 210 2.21chol 3.0 30 -10.2 2/5 36.0 260 3.00pH 2.0 4.0 40 -19.0 5/5 >60.0 NDC NDLipovincDSPC/ 2.0 20 - 4.1 0/5 21.0 110 1.50chol/GMI 3.0 30 - 8.4 0/5 24.0 140 2.00pH 4.0 4.0 40 -22.8 1/5 24.0 140 2.82LipovincDSPC/ 1.0 10 + 3.3 1/5 20.0 100 1.43chol/GMI 2.0 20 + 0.2 8.10 >70.0 ND NDp1-I 2.0 3.0 30 -10.9 10/10 >70.0 ND NDLipovinc 4.0 40 -14.4 10/10 >70.0 ND NDPercentage ILS values were determined from median survival times of treated and saline control groups. If morethan 50% of the animals survived for more than 70 days, median survival times and % ILS are indicated asgreater than 70 days.L/F (liposomal/free) values were calculated by dividing the median survival time for the liposomal vincristinegroup by the median survival time for the group receiving the equivalent dosage of free drug.CND- not determined.97Figure 5.4Lewis Lung Tumor Accumulation of Liposomal VincristineDrug accumulation (A) and lipid accumulation (B) were determined followingi.v. administration in BDF1 mice of free vincristine (•), DSPC/Chol pH1 4.0liposomal vincristine (a), and DSPC/Chol/GMI pH1 2.0 (A). Vincristine wasencapsulated at a drug-to-lipid ratio of 0.1:1 (wt:wt). Error bars representstandard deviations of eight tumors.40DI.0)zI.C’,0z>0807060504030201000D0)0)0-J0 12 24 36 48TIME (HOURS)9831A0 12 24 36 48 60 72B60 72injection. Vesicles composed of DSPC/Ch0IIGM1pH1 2.0 resulted in even better drugretention within the site of tumor growth. For this liposomal drug preparation, despitesimilar levels of tumor drug accumulation to DSPCICho1 pH1 4.0 liposomal vincristine at 24h, the release of drug from the tumor site is much slower, resulting in drug levels ofapproximately 2 ig/g tumor at 72 h following injection.Initial tumor drug loading for the two liposomal preparations show quite differentpatterns of uptake. For the DSPCICho1 pH 4.0 preparation, initial tumor drug uptake is veryrapid reaching close to its maximum value within the first hour following i.v. administration.The DSPC/ChoL’GMl pH 2.0 formulation, however, displays a much slower drugaccumulation within the site of tumor growth. At 1 h following i.v. injection, even lowerlevels of drug accumulation are seen than for the administration of free drug (Figure 5.4A).These vesicles result in peak tumor drug concentrations at approximately 24 h following i.v.injection. The reasons for these differences may be due to the ability of the differentliposomes to gain access to the tumor. This is reflected in Figure 5.4B. At 1 h followingadministration, the concentration of DSPC/Chol pH1 4.0 liposomes is approximately 5-foldhigher than the concentration of DSPC/Ch0IJGM1pH1 2.0 liposomes within the tumor. Theconcentration of DSPC/Chol pH1 4.0 liposomes increases to reach a maximum at 24 h beforesteadily declining. In contrast, the accumulation of of DSPC/Ch0IJGM1pH1 2.0 liposomeswithin the tumor progressively increases over the 72 h experiment.The accumulation of liposomal vincristine within sites of tumor growth is determinednot only by the ability of the liposomes to accumulate within the tumor, but also on theability of the liposome to retain the drug within its aqueous interior. In order to betterunderstand the factors affecting tumor uptake of drug, it is of interest to examine the drug-to99lipid ratios of both liposomal vincristine preparations within the site of tumor growth (Figure5.5). DSPC/CholIGMl pH 2.0 liposomes retain vincristine much better than DSPC/Chol pH,4.0 liposomes within the Lewis Lung tumor. It appears that this difference is mainlydependent on the initial drug retention since vincristine release rates for both systemsfollowing 1 h post injection are essentially similar.Lewis Lung tumor efficacy studies are shown to reflect the differences seen in tumordrug accumulation (Figure 5.6). Control mice demonstrate a continual increase in tumorweight until all mice are dead by day 9 following i.v. injection of saline. When 2 mg/kg offree vincristine is administered, a minimal decrease in tumor growth is observed but all miceare dead by day 9 following drug administration. The administration of DSPC/Chol pH1 4.0liposomal vincristine results in a substantial decrease in tumor growth as compared tocontrols and an increase in survival time of approximately 2-fold. The best results, however,are seen with the DSPC/Chol/GM1pH1 2.0 preparation where tumor growth is essentiallyhalted for the first 10 days following i.v. administration of the drug. Tumor growth thenbegins to progress resulting in mouse death by day 21 following drug administration.5.4 DiscussionA large amount of attention in liposomal drug delivery has been focussed recentlyon the use of “stealth” lipids such as GM1 and PEG-derivatized phospholipids to increase thecirculation lifetime and therapeutic activity of certain liposomal drugs. This approach hasbeen utilized successfully for liposomal formulations of doxorubicin and cytosinearabinoside (Gabizon and Papahadjopoulos, 1988; Allen et al., 1992). However, an implicitfeature of the “stealth” strategy is that the entrapped drug must be efficiently retained inside1000a-.0I-;0D0Figure 5.5Drug-to-Lipid Ratios Within the Lewis Lung Tumor ModelDrug-to-lipid ratios within the Lewis Lung tumor over time for DSPC/Chol pH14.0 liposomal vincristine (R) and DSPC/Chol/GM1pH1 2.0 liposomal vincristine(A). Vincristine was encapsulated at an initial drug-to-lipid ratio of 0.1:1(wt:wt). 12 24 36 48 60 72TIME (HOURS)101D)I—CDw0DF-Figure 5.6Lewis Lung Tumor Growth Following Vincristine AdministrationTumor weight (as calculated from size) was measured daily following i.v.administration of saline (0), free vincristine (•), DSPC/Chol pH1 4.0 liposomalvincristine (•), and DSPC/Chol/GM1 pH1 2.0 liposomal vincristine (A).Vincristine was administered at a dose of 2 mg/kg. Error bars representstandard errors of four mice.2100 1 2 3 4 5 6 7DAYS FOLLOWING DRUG ADMINISTRATION102the liposome in order to take advantage of the long circulation lifetime of such systems.Previously, there has been minimal interest in utilizing GM1 in liposomal vincristinepreparations due to the relatively rapid release of the drug from liposomes in the circulation(Mayer et al., 1993). Recent developments in the use of pH-gradient liposome systems toenhance vincristine entrapment, however, has now made the use of stealth lipids for in vivovincristine delivery of interest.Decreasing the pH of the entrapped citrate buffer for DSPC/Chol liposomes leads toincreased vincristine retention in the circulation as evidenced by the 3.1-fold increase in thecirculating drug-to-lipid ratio compared to DSPC/Chol pH1 4.0 preparations 24 h after i.v.administration. These results, together with the fact that including GM1 into DSPC/Cholliposomes at pI-I 4.0 results in a 2.4-fold increase in circulating liposomal lipid levels at 24h, suggest that plasma vincristine concentrations for GM1 pH1 2.0 liposomes should beapproximately 7.5-fold greater than observed for the DSPC/Chol pH 4.0 systems studiedextensively in previous investigations (Mayer et al., 1990c; 1993). The data presented hereindicate, however, that lowering the pH of the entrapped buffer to 2.0 and including GM1 inthe membrane synergistically combine to dramatically increase the vincris tine concentrationin the plasma. Specifically, 24 h after i.v. administration, circulating drug levels areincreased 18.3-fold whenG1-containing pH1 2.0 DSPC/Chol liposomes are utilizedcompared to DSPC/Chol pH 4.0 systems.It is interesting to note that the enhanced pharmacokinetic properties of GM1 pH1 2.0liposomal vincristine systems are obtained even though liposomes prepared at pH, 2.0 displaysomewhat decreased circulation lifetimes (Figure 5. 1B) and corresponding increased RESuptake (Figure 5.2). The reasons for this difference are not yet understood, however it is103clear that being able to eliminate this effect could potentially result in further increases incirculating vincristine levels.The studies described in this chapter demonstrate that the enhanced pharmacokineticcharacteristics provided by the GM1 pH1 2.0 liposomal vincristine preparations translate tosubstantially improved antitumor activity. We have previously shown that increasing thecirculation lifetime of vincristine entrapped in PCICho1 systems improved antitumor efficacy(Mayer et al., 1990c; 1993; Boman et al., 1994). The results obtained with the long-lived,non-leaky GM1 pH 2.0 liposomes extend from this initial observation, thereby transforminga drug with minimal activity against the munne P388 tumor model into one where high curerates are achieved.The mechanisms whereby the use of GM1 and pI-l 2.0 entrapped buffer synergisticallystabilize liposomal vincristine preparations is not well understood. One possibility is thatG1-containing vesicles are able to withstand an osmotic gradient approximately 10% greaterthan that withstood by DSPC/Chol liposomes when incubated in mouse serum (results notshown). On exposure to large osmotic gradients, liposomal membranes form transient poresallowing entrapped solute to escape (Mui et aL, 1993). By theG1-containing vesicles beingable to withstand a greater osmotic gradient, they are less likely to release entrapped drug.It is of interest to note that with an internal pH of 4.0 there is no increase in drugretention with the incorporation of GM1. However, at pH1 2.0, there is a substantial increasein drug retention when GM1 is added. This is likely due to the fact that lowering the internalpH and adding GM1 act via different mechanisms to increase drug retention. It has beenshown previously that vincristine is released from liposomes as the pH gradient across theliposome membrane decreases (Boman et al., 1993). By increasing the initial pH gradient104across the membrane, the drug can be retained by the liposome more efficiently. In contrast,the incorporation of GMI decreases the amount of plasma protein binding to the liposome(Chonn et al., 1992). This enables the membrane to remain more stable thereby preventingdrug leakage. At pH1 4.0, drug leakage is so rapid that the stabilizing effects of GM1 areunable to further enhance the efficiency of drug retention in vivo.By analogy with previous studies with liposomal anticancer agents (Mayer et al.,1990b; Mayer et al., 1989), increased drug concentration in the plasma would be expectedto result in an increase in antitumor activity. Such an effect has been seen previously forP388 and L1210 leukemia when the vincristine circulation time was increased byencapsulation within DSPCICho1 vesicles (Mayer et al., 1990c). The encapsulated drugdisplayed increased antitumor activity over free drug as well as drug encapsulated in leakyliposomes (eggPCIChol). The increased plasma drug concentrations observed here byincorporating GMI into vesicle membranes and decreasing the internal pH results in asubstantial increase in antitumor activity in the P388 leukemia model. This is particularlyevident at drug dosages above 2 mg/kg. At these dosages, we see long term survivors withmedian survival times in excess of 70 days. Thus, the benefits from extended circulationtimes are clearly evident.The extended drug circulation times seen with the incorporation of GM1 with aninternal pH of 2.0 also prove to be beneficial in the treatment of a murine solid tumor model.Not only are increased tumor drug levels observed but also a substantial decrease in tumorgrowth is noted in vivo. This improvement in tumor drug accumulation appears to be duenot only to the ability of the liposomes to retain the drug, but also to the ability of theliposomes to remain within the site of tumor growth for extended periods of time. This is105likely due to the extended plasma circulation times seen for this preparation.It is entirely possible that similar liposomal systems can be utilized to increase drugretention of other remote loading drugs. By increasing their retention within liposomes,there would likely be an increase in therapeutic activity.The DSPC/Ch0IJGM1pH1 2.0 system also displays a considerable decrease in toxicityas evidenced by decreased weight loss. This could be due to the fact that the drug is moretightly held within the liposome, thus preventing toxic side effects.It is important to emphasize that the improved therapeutic activity observed here isobtained following a single i.v. dose of encapsulated vincristine. It could be argued thatsimilar results would be achieved with a systemic infusion of free drug. Results from thislaboratory (Mayer et al., 1990b) and others (Gabizon and Papahadjopoulos, 1988), however,suggest that increased therapeutic activity achieved with liposomal anticancer drugs is dueprimarily to accumulation of drug loaded liposomes in the region of tumor cell growth.Further, recent pharmacological studies (Mayer, L.D., unpublished observations) indicatethat systemic exposure to free vincristine is lower for drug administered in liposomal formcompared to unentrapped vincristine. Therefore, improved drug retention characteristicsshould result in improved specificity of drug delivery to the tumor and hence, increasedtherapeutic activity. Comparable drug delivery to a defined region of disease growth wouldnot be expected using a continuous i.v. infusion of free vincristine.In summary, this chapter demonstrates that lowering the internal pH andincorporation of GM1 into liposomal membranes results in a synergistic increase in drugretention in vivo and increased drug circulation times. This system shows substantialimprovement in therapeutic activity against the P388 leukemia cell line in vivo resulting in106long term survival. It also displays a considerable decrease in solid tumor growth rate inmice inoculated with Lewis Lung tumor cells. Although the synergistic effects of GM1-incorporation and lowering the internal pH to 2.0 are not well understood, many of theseareas warrant further investigation.107CHAPTER 6SUMMARYUntil recently (Mayer et aL, 1990c), liposomal encapsulation of vincristine has beenof marginal benefit due to the inability to retain the drug within the vesicles for extendedperiods of time. This thesis has first shown that liposomal encapsulation employing thetechnique of Mayer et al. (1990) ameliorates the toxic side effects usually observed onextravasation, and has then examined factors which further improve the ability of liposomesto retain the drug in vivo.Firstly, it was shown in Chapter 2 that the use of liposomal encapsulation, employingthe procedures of Mayer et aL, 1990, dramatically decreases the soft-tissue necrosis andulceration seen for the free drug when extravasated. This observation is important due to thedevastating necrotic effects that can occur on accidental extravasation of vincristine (Bellone,1981; Choy, 1979). Although the liposomally encapsulated drug remains in the tissue forextended periods of time after extravasation, the trapped drug is unable to exert cytotoxiceffects. The tissue is exposed to a low, long-term dose of free drug as the vincristine slowlyleaks from the liposomes.Next, Chapter 3 investigated physical parameters such as lipid composition, internalbuffering capacity, internal pH, and temperature on vincristine uptake and release fromvesicles. Through these studies, it was determined that reduction of the internal pH had themost profound effect on improving vincristine retention within liposomes in vivo.Chapters 4 and 5 examined the incorporation of factors into liposomes to improvevincristine circulation time. In Chapter 4 the hypothesis that cationic lipids would improve108drug retention was studied, the thought being that the presence of positively charged lipidson the inner monolayer of vesicles would repel the positive charge on the vincristinemolecule at low pH, thus inhibiting release. Effects consistent with this interpretation wereobserved. Vincristine retention was dramatically improved by the incorporation of 10 mol%cationic lipid at low pH. The resultant increase in vincristine circulation time translated toa substantial improvement in therapeutic activity.Chapter 5 examined a different approach for improving vincristine circulation timewhich involved the use of monosialoganglioside GM1. Surprisingly, it was found that theincorporation of GM1 combined with the lowering of the internal pH to 2.0 showed asynergistic effect in improving vincristine retention within the liposomes. This effectcombined with the effect of GM1 on increasing liposome circulation time, greatly improvedthe circulation time of the drug, resulting in high cure rates when employed against P388lymphocytic leukemia in mice.There are several directions for future research. The first is to characterize howmonosialoganglioside GM1 contributes to improved drug retention. It would be of interestto examine the use of other lipids which improve liposome circulation time such aspolyethyleneglycols and phosphatidylinositol (Gabizon and Papahadjopoulos, 1988). It haspreviously been demonstrated that GM1 increases liposome circulation time by reducing thetotal amount of blood protein bound to the liposome surface (Chonn et aL, 1992). It wouldbe interesting to determine if any correlation exists between plasma protein binding andvincristine leakage from liposomes. A wide variety of liposome compositions could bestudied with varying abilities to bind plasma proteins.To this point, we have demonstrated that the slower the leakage of vincristine from109liposomes and the more drug associated with the vesicles, the lower the toxicity and thehigher the therapeutic potential. Previous work by Layton and Trouet (1980) demonstratedno advantage with liposomal encapsulation of vincristine either in terms of reduced toxicityor increased therapeutic index. This was likely due to the fact that preparations were utilizedwith low drug trapping efficiencies and poor drug retention abilities. Since vincristine mustgain access to tumor cells in order to exert its cytotoxic effects, there is likely an optimallevel of drug retention within liposomes, above which therapeutic activity will againdecrease. It would be of considerable interest to develop means of further improvingvincristine retention within liposomes to determine the optimum level of drug retention fortherapeutic activity.The next area which warrants further investigation is the ability of various liposomesto access sites of disease, such as tumors. Although a liposome may be better able to retaina drug, if that drug is less able to access the target cells, it may be of little or no benefit.Since the majority of this thesis has focused on the antineoplastic effects against P388lymphocytic leukemia injected intraperitoneally, access to the peritoneal cavity for thevarious liposomal compositions studied should be determined. Factors such as liposomesize, liposome composition, and the presence or absence of internal drug should beinvestigated as to their effects on peritoneal accumulation. The accumulation should alsobe investigated in both the presence and absence of tumor growing in the mouse. Anotherinteresting avenue for further investigation would include the use of drugs which promotevascular permeability, thereby allowing the liposomes to pass through the vasculature andachieve better access to the tumor.Along the same lines, various factors should be investigated for their abilities to110promote or inhibit liposome accumulation within solid tumors. In Chapter 5 it was seen thatalthough GM1 resulted in higher vincristine accumulation within the Lewis Lung tumor, thelipid accumulation was drastically decreased.Very high drug retention levels could be important for targeting these liposomalsystems directly to tumor cells. At the present time liposomes have successfully beentargeted to some tissues (Gabizon et al., 1990; Ahmad and Allen, 1992; Lundberg et al.,1993; Ahmad et al., 1993). if targeting to tumor cells could be accomplished with very highdrug retention levels, the drug could then be released to expose the target area to a very highconcentration of drug, leaving non-target tissues with lower potential drug toxicity. Meansby which the drug could be released at its target site by dissipating the pH-gradient acrossthe liposomal membrane. This could be accomplished with the use of proton gradientuncouplers (Mayer et aL, 1990d). More effectively, the targeted liposomes could be inducedto fuse with the membranes of target cells, thereby releasing the drug directly into the cytosolof tumor cells. Liposome fusion has already been accomplished with the use of fusogeniclipids (Allen et al., 1990; Wilschut et al., 1992; Eastman et al., 1992) (see Figure 6.1).In conclusion, it has been shown that improved retention of vincristine in liposomesleads to decreased drug toxicity and improved antineoplastic activity. These studies may beused as a model for the liposomal encapsulation of other drugs of weakly basic nature whichcan be loaded into liposomes across a pH gradient. The potential for reduced toxicity andimproved therapeutic activity is of obvious benefit.111Figure 6.1Targeted fusion of liposomes.The targeted carrier binds to the target cell and fuses with it. The entrapped material is then releaseddirectly into the cell interior.monoclonalantibodyfree drugtarget cellosi(<1 hour)target cellfusionprotein112REFERENCESAbra, R.M., Bosworth, M.E., and Hunt, C.A. (1980). Res. Comm. Chem. Pathol.Pharmacol. 29, 349-360.Addanki, S., Cahill, F.D., and Sotos, J.F. (1968). J. Biol. Chem. 243, 2337-2343.Ahmad, I. and Allen, T.M. (1992). Cancer Res. 52(7), 48 17-4820.Ahmad, I., Longenecker, M., Samuel, J., and Allen, T.M. (1993). Cancer Res. 53(7), 1484-1488.Allen, T.M., Austin, G.A., Chonn, A., Lin, L., and Lee, K.C. (1991a). Biochim. Biophys.Acta 1061, 56-64.Allen, T.M. and Chonn, A. (1987). FEBS Lett. 223, 42-46.Allen, T.M. and Cleland, L.G. (1980). Biochim. Biophys. Acta 597, 418-426.Allen, T.M. and Hansen, C. (1991). Biochim. Biophys. Acta 1068, 133-141.Allen, T.M., Hansen, C., and Rutledge, J. (1989). Biochim. Biophys. Acta 981, 27-35.Allen, T.M., Hansen, C., Martin, F., Redemann, C., and Yau-Young, A. (1991b). Biochim.Biophys. Acta 1066, 29-36.Allen, T.M., Hong, K., and Papahadjopoulos, D. (1990). Biochemistry 29(12), 2976-2985.Allen, T.M., Mehna, T., Hansen, C., and Chin, Y.C. (1992). Cancer Res. 52, 243 1-2439.Alving, C.R. and Richards, R.L. (1983) in Liposomes. Ostro, M., ed. Marcel Dekker, NewYork, 209-287.Bailey, A.L. and Cullis, P.R. (1994). Biochemistry, in press.Balazsovits, J.A.E., Mayer, L.D., Bally, M.B., Cullis, P.R., McDonall, M., Ginsberg, R.S.,and Fallc, R.E. (1989). Cancer Chemother. Pharmacol. 23, 8 1-86.Bally, M.B., Nayar, R., Masin, D., Hope, M.J., Cullis, P.R., and Mayer, L.D. (1990).Biochim. Biophys. Acta 1023, 133-139.Bangham, A. (1983) in A.D. Bangham (ed.) Liposome Letters, Academic Press, London113261-268.Bangham, A.D., Standish, M.M., and Watkins, J.D. (1965). J. Mol. BioL 13, 238-244.Barenholz, Y., Amselem, S., and Lichtenberg, D. (1979). FEBS Lett. 99, 210-214.Barr, R.D. and Sertic, J. (1981). Br. J. Cancer 44, 267-269.Bellone, J.D. (1981). JAMA 245, 343.Blok, M.C., van der Neut-Kok, E.C.M., van Deenan, L.L.M., and de Grier, J. (1975).Biochim. Biophys. Acta 406, 187-192.Blume, G. and Cevc, G. (1990). Biochim. Biophys. Acta 1029, 9 1-97.Boman, N.L., Mayer, L.D., and Cullis, P.R. (1993). Biochim. Biophys. Acta 1153, 253-258.Boman, N.L., Masin, D., Mayer, L.D., Cullis, P.R., and Bally, M.B. (1994). Cancer Res.,in press.Bonte, F. and Juliano, R.L. (1986). Chem. Phys. Lip. 40, 359-372.Carter, S.K. and Livingston, R.B. (1976). Cancer Treat. Rep. 60, 1141-1156.Chait, L.A. and Dinner, M.I. (1975). S. Afr. Med. J. 49, 1935-1936.Chakrabarti, A.C., Clark-Lewis, I., Harrigan, P.R., and Cullis, P.R. (1992). Biophys. 1 61,228-234.Chapman, D. (1975). Q. Rev. Biophys. 8, 185-235.Chobanian, J.V., Tall, A.R., and Brecher, P.1. (1979). Biochemistry. 18, 180-187.Chonn, A., Cullis, P.R., and Devine, D.V. (1991). J. Immunol. 146, 4234-4241.Chonn, A., Semple, S.C., and Cullis, P.R. (1992a). J. Biol. Chem. 267, 18759-18765.Chonn, A. and Cullis, P.R. (1992b). J. Liposome Res. 2(3), 397-410.Choy, D.S. (1979). JAMA 241, 695.Cullis, P.R. (1976). FEBS Lett. 70, 223-228.Deamer, D.W. and Bangham, A.D. (1976). Biochem. Biophys. Acta 443, 629-634.114Demel, R.A. and de Kruyff, B. (1976). Biochem. Biophys. Acta 457, 109-132.Derksen, J.T.P., Morselt, H.W.M., and Scherphof, G.L. (1987). Biochim. Biophys. Acta931, 33-40.Dorr, R.T. and Alberts, D.S. (1985). JNCI74, 113-120.Dorr, R.T., Alberts, D.S., and Chen, H.S. (1980). J. Pharmacol. Methods 4, 237-250.Dorr, R. and Fritz, W. (1982) in Cancer chemotherapy handbook. Elsevier Science, NewYork, p. 388-401.Eastman, S.J., Hope, M.J., and Cullis, P.R. (1991). Biochemistry 30, 1740-1745.Eastman, S.J., Wilschut, J., Cullis, P.R., and Hope, M.J. (1989). Biochim. Biophys. Acta981, 178-184.Eastman, S.J., Hope, M.J., Wong, K.F., and Cullis, P.R. (1992). Biochemistry 31(17), 4262-4268.Fichtner, I., Reszka, R., Elbe, B., and Arndt, D. (1981). Neoplasma 28, 141-149.Finkeistein, M.C. and Weissmann, G. (1979). Biochim. Biophys. Acta 587, 202-209.Forssen, E.A. and Tokes, A.A. (1981). Proc. Nati. Acad. Sci. USA 78, 1873-1877.Forssen, E.A. and Tokes, Z.A. (1983). Cancer Treat. Rep. 67, 48 1-484.Frezard, F. and Garnier-Suillerot, A. (1991). Biochemistry 30(20), 5038-5043.Gabizon, A. and Papahadjopoulos, D. (1988). Proc. Nati. Acad. Sci. USA 85, 6949-6953.Gabizon, A., Dagan, A., Goren, D., Barenholz, Y., and Fuks, Z. (1982a). Cancer Res. 42,4734-4739.Gabizon, A., Goren, D., Fuks, Z., Barenholz, Y., Dagan, A., and Meshorer, A. (1982b).Cancer Res. 43, 4730-4735.Gabizon, A., Price, D.C., Huberty, J., Bresalier, R.S., and Papahadjopoulos, D. (1990).Cancer Res. 50, 637 1-6378.Gaffi, M., Comfurius, P., Maassen, C., Hemker, H.C., DeBaets, M.H., VanBreda-Vriesman,P.J.C., Barbui, T., Zwaal, R.F.A., and Bevers, E.M. (1990). Lancet 335, 1544-1547.115Gregoriadis, G. (1984). Nature. 310(5974), 186-187.Gregoriadis, G. (ed.) (1988). Liposomes as Drug Carriers, Wiley, New York.Gregoriadis, G. (1973). FEBS Lett. 36, 292-296.Gregoriadis, G. and Davis, C. (1979). Biochem. Biophys. Res. Commun. 89, 1287-1293.Gregoriadis, G. and Neerunjun, E.D. (1975). Res. Commun. Chem. Pathol. Pharmacol. 10,351-362.Gregoriadis, G. and Ryman, B.E. (1972). Eur. J. Biochem. 24, 485-491.Gregoriadis, G. and Senior, J. (1980). FEBS Lett. 119, 43-46.Harrigan, P.R., Wong, K.F., Redelmeier, T.E., Wheeler, J.J., and Cullis, P.R. (1993).Biochim. Biophys. Acta 1149, 329-338.Herman, E.H., Rahman, A., Ferrans, V.J., Vicks, J.A., and Schein, P.S. (1983). Cancer Res.43, 5427-5432.Hope, M.J., Bally, M.B., Mayer, L.D., Janoff, A.S., and Cullis, P.R. (1986). Chem. Phys.Lipids 40, 89-108.Hope, M.J., Bally, M.B., Webb, G., and Cullis, P.R. (1985). Biochim. Biophys. Acta. 812,55-65.Hope, M.J. and Cullis, P.R. (1987). J. Biol. Chem. 262(9), 4360-4366.Hope, M.J., Redelmeier, T.E., Wong, K.F., Rodrigueza, W., and Cullis, P.R. (1989).Biochemistry 28, 418 1-4187.Houslay, M.D. and Stanley, K.K. (1982). Dynamics of Biological Membranes, John Wileyand Sons, Toronto.Huang, C.H. (1969). Biochemistry 8, 344-352.Hunt, C.A., Rustum, Y.M., Mayhew, E., and Papahadjopoulos D. (1979). Drug Metab.Dispos. 7, 124-128.Hwang, K.J. and Beaumier, P.L. (1988) in Liposomes as Drug Carriers: Recent Trends andProgress. Gregoriadis, G., ed. John Wiley and Sons Ltd., 19-36.Hyslop, P.A., Morel, B., and Sauerheber, R.D. (1990). Biochemistry 29, 1025-1038.116Ilium, L., Hunneyball, I.M., and Davis, S.S. (1986). mt. J. Pharmaceutics 29, 53.Inoue, K. (1974). Biochim. Biophys. Acta. 339, 390-402.Jackson, C.M. (1980). Ann. Rev. Biochem. 49, 765-811.Jackson, D.V., Jr. and Bender, R.A. (1979). Cancer Res. 39, 4346-4349.Jackson, D.V., Jr., Jobson, V.W., Homesley, H.D., Welander, C., Hire, E.A., Pavy, M.D.,Votaw, M.L., Richards, F. 2d, and Muss, H.B. (1986a). Gynecol. Onc. 25(2), 212-216.Jackson, D.V., White, D.R., Spurr, C.L., Hire, E.A., Pavy, M.D., Robertson, M., Legos,H.C., and McMahan, R.A. (1986b). Am. J. of Clin. Onc. 9(5), 376-378.Kabanov, A.V., Levashov, A.V., Alakhov, V.Y., Martinek, K., and Severin, E.S. (1990).Biomedical Science. 1(1), 33-36.Kinoshita, T., Inoue, K., Okada, M., and Akiyama, Y. (1977). J. Immunol. 119, 73-76.Kirby, C., Clarke, J., and Gregoriadis, G. (1980a). Biochem. J. 186, 591-598.Kirby, C., Clarke, J., and Gregoriadis, G. (1980b). FEBS Lett. 111, 324-328.Klausner, R.D., Bluthmenthal, R., Innerarity, T., and Weinstein, J.N. (1985). J. Biol. Chem.260, 13719-13727.Klibanov, A.L., Maruyama, K., Beckerleg, A.M., Torchilin, V.P., and Huang, L. (1991).Biochim. Biophys. Acta 1062, 142-148.Klibanov, A.L., Maruyama, K., Torchilin, V.P., and Huang, L. (1990). FEBS Lett. 268,235-237.Knoben, J. and Anderson, P. (1983) in Handbook of clinical drug data. Drug IntelligencePublications, Hamilton, p. 347.Kobayashi, T., Kataoka, T., Tsukagoshi, S., and Sakurai, Y. (1977). mt. J. Cancer 20, 581-587.Kremer, J.M.H., Esker, M.W.J., Pathmamanohovan, C., and Wiersema, P.H. (1977).Biochemistry 16, 3932-3935Krupp, L., Chobanian, A.V., and Brecher, I.P. (1976). Biochim. Biophys. Res. Commun. 72,125 1-1258.117Lasic, D.D., Martin, F.J., Gabizon, A., Huang, S.K., and Papahadjopoulos, D. (1991).Biochim. Biophys. Acta 1070, 187-192.Layton, D and Trouet A. (1980). Europ. J. Cancer 16, 945-950.Liu, D. and Huang, L. (1990). Biochim. Biophys. Acta 1022, 348-354.Loth, T.S. and Eversman, W.W., Jr. (1986). J. Hand Surg. hA, 388-396.Lundberg, B., Hong, K., and Papahadjopoulos, D. (1993). Biochim. Biophys. Acta 1149(2),305-312.McNeil, H.P., Simpson, R.J., Chesterman, C.N., and Krilis, S.A. (1990). Proc. Nati. Acad.Sci. USA. 87, 4 120-4124.Madden, T.D., Harrigan, P.R., Tai, L.C.L., Bally, M.B., Mayer, L.D., Redelmeier, T.E.,Loughrey, H.C., Tilcock, C.P.S., Reinish, L.W., and Cullis, P.R. (1990). Chem.Phys. Lipids 53, 37-46.Malinski, J.A. and Nelsestuen, G.L. (1989). Biochemistry 28, 6 1-70.Mayer, L.D., Bally, M.B., and Cullis, P.R. (1986a). Chem. Phys. Lip. 40, 333-345.Mayer, L.D., Bally, M.B., and Cullis, P.R. (1986b). Biochim. Biophys. Acta 857, 123-126.Mayer, L.D., Bally, M.B., and Cullis, P.R. (1990a). J. Liposome Res. 1, 463-480.Mayer, L.D., Bally, M.B., Cullis, P.R., Wilson, S.L., and Emerman, J.T. (1990b). CancerLett. 53, 183-190.Mayer, L.D., Baily, M.B., Loughrey, H., Masin, D., and Cullis, P.R. (1990c). Cancer Res.50, 575-579.Mayer, L.D., Hope, M.J., Cullis, P.R., and Janoff, A.S. (1985). Biochim. Biophys. Acta 817,193- 196.Mayer, L.D., Nayar, R., Thies, R.L., Boman, N.L., Cullis, P.R., and Bally, M.B. (1993).Cancer Chemother. Pharmacol. 33, 17-24.Mayer, L.D., Tai, L.C.L., Bally, M.B., Mitilenes, G.N., Ginsberg, R.S., and Cullis, P.R.(1990d). Biochim. Biophys. Acta 1025, 143-151.Mayer, L.D., Tai, L.C.L., Ko, D.S.C., Masin, D., Ginsberg, R.S., Cullis, P.R., and Bally,M.B. (1989). Cancer Res. 49, 5922-5930.118Mayhew, E. and Rustum, Y.M. (1985). Prog. Clin. Biol. Res. 172B, 301-3 10.Mimms, L.T., Guido, Z., Nozaki, Y., Tanford, C., and Reynolds, J.A. (1981). Biochemistry20, 833-840.Morgan, J.R., Williams, K.E., Davies etal., (1981). J. Med. Microbiol. 14, 213-217.Mui, B.L.S., Cuffis, P.R., Evans, E.A., and Madden, T.D. (1993). Biophys. J. 64, 443-453.Mui, B.L.S., Cullis, P.R., Pritchard, P.H., and Madden, T.D. (1994). J. Biol. Chem. 269,7364-7370.Muller-Eberhard, H.J. (1986). Annu. Rev. Immunol. 4, 503-528.Nichols, J.W. and Deamer, D.W. (1976). Biochim. Biophys. Acta 455, 269-271.Ogihara, I., Kojima, S., and Jay, M. (1986). Eur. J. Nuci. Med. 11, 405-411.Olson, F., Hunt, C.A., Szoka, F.C., Vail, W.J., and Papahadjopoulos, D. (1979). Biochim.Biophys. Acta 557, 9-23.Op den Kamp, J.A.F. (1979). Annu. Rev. Biochem. 48, 47-71.Ostro, M. and Cullis, P.R. (1989). Am. J. Hosp. Pharm. 46, 1576-1587.Owellen, R.J., Hartke, C.A., Dickerson, R.M., and Hal, F.O. (1976). Cancer Res. 36,1499-1502.Owellen, R.J., Owens, A.H., Jr., and Donigian, D.W. (1972). Biochem. Biophys. Res.Commun. 47, 685-691.Papahadjopoulos, D., Jacobson, K., Nir, S., and Isac, T. (1973). Biochim. Biophys. Acta.311, 330-348.Parente, R.A. and Lentz, B.R. (1984). Biochemistry 23, 2353-2362.Parr, M.J., Bally, M.B., and Cullis, P.R. (1993). Biochim. Biophys. Acta 1168, 249-252.Patel, H.M., Tuzel, N.S., and Ryman, B.E. (1983). Biochim. Biophys. Acta 761, 142-151.Poste, G. (1983). Biol. Cell 47, 19-38.Poste, G., Bucana, C., Raz, A., Bugeiski, P., Kirsh, R., and Fidler, I.J. (1982). Cancer Res.42, 1412-1422.119Proffitt, R.T., Williams, L.E., Presant, C.A., Tin, G.W., Uliana, J.A., Gamble, R.C., andBaldeschwieler, J.D. (1983). J. Nuci. Med. 24, 45-51.Rahman, A., More, N., and Schein, P.S. (1982). Cancer Res. 42, 1817-1825.Reilly, J.J., Neifield, J.P., and Rosenberg, S.A. (1977) Cancer 40, 2053-2056.Richards, R.L., Gewurz, H., Osmand, A.P., and Alving, C.R. (1977). Proc. Nati. Acad. Sci.USA 74, 567 2-5676.Richards, R.L., Gewurz, H., Siegel, J., and Alving, C.R. (1979). J. immunol. 122, 1185-1189.Richardson, V.J., Jeyasingh, K., Jewkes, R.F., Ryman, B.E., and Tattersall, M.H.N. (1978).J. Nuci. Med 19, 1049-1054.Roerdhk, F.H., Regis, J., Handel, T., Sullivan, S.M., Baldeschwieler, J.D., and Scherphof,G.L. (1989). Biochim. Biophys. Acta 980, 234-240.Rottenberg, H. (1979). Methods Enzymol. 55, 547-569.Rudolph, R., Stein, R.S., and Patillo, R.A. (1976). Cancer 38, 1087-1094.Sabbadini, R., McNutt, Wm., Jenkins, G., Betto, R., and Salviati, G. (1993). Biochem.Biophys. Res. Commun. 193, 752-758.Scherphof, G., Roerdink, F., Waite, M., and Parks, J. (1978). Biochim. Biophys. Acta. 542,296-307.Senior, J.H. (1987) in “CRC Critical Reviews in Therapeutic Drug Carrier Systems”, Vol.3, 123-193.Senior, J. and Gregoriadis, G. (1982). Life Sci. 30, 2123-2136.Senior, J., Delgado, C., Fisher, D., Tilcock, C., and Gregoriadis, G. (1991a). Biochim.Biophys. Acta 1062, 77-82.Senior, J.H., Trimble, K.R., and Maskiewicz, R. (1991b). Biochim. Biophys. Acta 1070:173-179.Shin, M.L., Paznekas, W.A., and Mayer, M.M. (1978). J. Immunol. 120, 1996-2002.Shin, M.L., Paznekas, W.A., Abramovitz, A.S., and Mayer, M.M. (1977). J. Immunol. 119,1358-1364.120Shinozawa, S., Araki, Y., and Oda, T. (1981). Acta Medica Okayama 35(6), 395-405.Sieber, S.M., Mead, J.A.R., and Adamson, R.H. (1976). Cancer Treat. Rep. 60, 1127-1139.Small, D.M. (1986) in The Physical Chemistry of Lipids from Alkanes to Phospholipids,Plenum Press, New York.Spiegel, R.J. (1981) Cancer Treat. Rev. 8, 197-207.Szoka, F. and Papahadjopoulos, D. (1978). Proc. Natl. Acad. Sci. 79, 4194-4198.Szoka, F. and Papahadjopoulos, D. (1980). Ann. Rev. Biophys. Bioenerg. 9, 467-508.Tall, A.R. and Green, P.H.R. (1981). J. Biol. Chem. 256, 2035-2044.Tao, R.V.P., Sweeley, C.C., and Jamieson, G.A. (1973). J. Lipid Res. 14, 16-25.Taylor, K.M.G., Taylor, G., Kellaway, I.W., and Stevens, J. (1990). mt. Journal ofPharmaceutics 58, 49-55.Tullius, E.K., Williamson, P., and Schlegel, R.A. (1989). Biosci. Rep. 9(5), 623-633.Upton, J., Mulliken, J.B., and Murray, J.E. (1979). Am. J. Surg. 137, 497-506.Vaage, J., Donovan, D., Mayhew, E., Uster, P., and Woodle, M. (1993). mt. J. Cancer 54,959-964.Wang, C.T. and Schick, P.K. (1981). .1. Biol. Chem. 256, 752-756.Williams, B.D., O’Sullivan, M.M., Saggu, G.S. et al., (1986). Br. J. Rheum. 25, 98-103.Wilschut, J., Scholman, J., Eastman, S.J., Hope, M.J., and Cullis, P.R. (1992). Biochemistry31(10), 2629-2636.Woo, S.Y., Dilhiplane, P., Rahman, A., and Sinks, L.F. (1983). Cancer Drug Deliv. 1, 59-62.121


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