Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of liposomal bupivacaine using a transmembrane pH gradient Mowat, Jeffrey John 1994

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

Item Metadata

Download

Media
831-ubc_1995-0572.pdf [ 3.26MB ]
Metadata
JSON: 831-1.0098986.json
JSON-LD: 831-1.0098986-ld.json
RDF/XML (Pretty): 831-1.0098986-rdf.xml
RDF/JSON: 831-1.0098986-rdf.json
Turtle: 831-1.0098986-turtle.txt
N-Triples: 831-1.0098986-rdf-ntriples.txt
Original Record: 831-1.0098986-source.json
Full Text
831-1.0098986-fulltext.txt
Citation
831-1.0098986.ris

Full Text

Development of Liposomal Bupivacaine Using a Transmembrane pH Gradient by Jeffrey John Mowat B.Sc.(Honours), Mount Allison University, 1993 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Pharmacology and Therapeutics We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1995 © Jeffrey John Mowat, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract There is a clear clinical requirement for longer acting local anaesthetics, particularly for the management of post-operative and chronic pain. In this regard, liposomes have been suggested to represent a potentially useful vehicle for sustained drug release following local administration. In this thesis, a transmembrane pH gradient was employed to efficiently encapsulate bupivacaine within large unilamellar vesicles. The rate and extent of bupivacaine uptake into large unilamellar vesicles exhibiting a pH gradient (interior acidic) were determined and compared to drug association with control liposomes that did not exhibit a proton gradient. Subsequent studies examined the kinetics of bupivacaine release from the liposome systems in vitro. Using the Guinea Pig cutaneous wheal model, the rate of clearance of the liposome carrier was monitored following intradermal administration employing a radiolabeled lipid marker and the duration of nerve blockade produced by free and liposomal bupivacaine compared. Intraperitoneal injections of bupivacaine encapsulated in pH gradient vesicles, control (no pH gradient vesicles) and free drug were completed in mice to determine the relative toxicities. While bupivacaine is rapidly and efficiently accumulated within liposomes exhibiting a pH gradient (interior acidic), little uptake was seen for control vesicles. Using an in vitro model of drug clearance, liposomally encapsulated bupivacaine was found to be released more slowly and over a longer period of time compared to either the free drug or bupivacaine associated with control (no pH gradient liposomes). In the Guinea Pig cutaneous wheal model, over 85% of the liposomal carrier was found to remain at the site of administration over two days and the sustained drug release afforded by liposomes exhibiting a pH gradient resulted in a threefold increase in the duration of nerve blockade compared to either the free drug or bupivacaine in the presence of control (no pH gradient) liposomes. In the toxicity study, bupivacaine encapsulated in pH gradient vesicles showed a greater than fourfold increase in safety compared to the use of the free drug and control (no pH gradient) vesicles. The present results clearly establish that large unilamellar vesicles exhibiting a pH gradient can efficiently encapsulate bupivacaine and subsequently provide a sustained release system that greatly increases the duration of neural blockade and, in addition, reduces the toxicity. i i i Table of Contents Page Abstract ii Table of Contents iv List of Tables vi List of Figure : vii Abbreviations . Acknowledgment 1 Introduction 1.1 The Clinical Requirement for a Long Duration Local Anaesthetic with , Reduced Toxicity. ! 1.2 Liposomal Drug Delivery Systems' •  1.2.1 Classification of Liposomes. .3 1.2.2 Drug Uptake into LUVs Exhibiting a pH Gradient 5 1.3 Properties of Bupivacaine 1.3.1 Chemical Structure 8. 1.3.2 Mechanism of Action. 9 1.3.3 Bupivacaine Toxicity 1.3.3.1 Central Nervous System Toxicity 11 1.3.3.2 Cardiac Toxicity .....12 1.4 Previous Studies on Sustained Release Vehicles for Local Anaesthetics 1.4.1 Previous Studies on Non-Liposomal Delivery Systems for Local Anaesthetics 1.3. 1 4.2 Previous Studies on Liposomal Delivery Systems for Local Anaesthetics 17 2. Uptake of Bupivacaine into LUVs Exhibiting a Proton Gradient 2.1 Introduction 2.0 2.2 Methods 2.2.1 Preparation of Large Unilamellar Vesicles Exhibiting a Proton Gradient 23 2.2.2 Methods for Determining the Efficiency of Bupivacaine Loading into Liposomes 2.2.2.1 Gel Exclusion Chromatography. / 24 2.2.2.2 Density Gradient Chromatography. 25 2.2.2.3 Centrifuge Filtration... .25 2.2.3 Analytical Procedures 2.6 2.3 Results 2.3.1 Gel Exclusion Chromatography 26 2.3.2 Density Gradient Centrifugation...... 28 2.3.3 Filtration. 29 2.4 Discussion .3.5 iv 3. Release of Bupivacaine from LUVs 3.1 Introduction..; 39 3.2 Methods. 39 3.3 Results....; .40 3.4 Discussion 43 4. Duration of Nerve Blockade in Vivo , 4.1 Introduction. 46 4.2 Methods 4.2.1 Clearance Rate of LUVs Following Intradermal Injection into Guinea Pigs. 47 4.2.2 Guinea Pig Cutaneous Wheal Model for Measuring the Duration of Neural Blockade 48 4.3 Results 4.3.1 Intradermal Clearance of LUVs. 49 4.3.2 Duration of Nerve Blockade for POPC: Choi Liposomes 49 4.3.3 Duration of Nerve Blockade for DOPCChol Liposomes .50 4.4 Discussion .5.7 5. Toxicity 5.1 Introduction ......61. 5.2 Methods. .61. 5.3 Results ; ;.. 62 5.4 Discussion 72 6. Discussion 73 7. Bibliography 77 List of Tables Table 1 Solubilities of Some Common Local Anaesthetics at pH 7.37 vi List of Figures Page Figure 1.2.1 The Morphology of Liposomes ...4 Figure 1.2.2 The Mechanism of Drug Loading Using the pH Gradient Technique 6 Figure 1.3.1 The Chemical Structures of Some Common Local Anaesthetics. 8 Figure 2.2.1 Gel Exclusion Chromatographic Determination of Bupivacaine Uptake into DOPCChol Liposomes. 30 Figure 2.2.2 Gel Exclusion Chromatographic Determination of Bupivacaine Uptake into DSPCChol Liposomes .3.1. Figure 2.2.3 Density Gradient Centrifugual Determination of Bupivacaine Uptake into DOPCChol Liposomes 32; Figure 2.2.4 Determination of Bupivacaine Distribution by Filtration... 33 Figure 2.2.5 pH Dependant Solubility of Bupivacaine in Ftepes:NaCl Buffer 34 Figure 3.3.1 Bupivacaine Release Kinetics from POPCChol Liposomes 41 Figure 3.3.2 Bupivacaine Release Kinetics from DOPCChol Liposomes... 42 Figure 4.3.1 The Intradermal Clearance of Liposomes 52 Figure 4.3.2 Duration of Nerve Blockade for POPC: Choi Liposomes 53 Figure 4.3.3.1 Duration of Nerve Blockade for DOPCChol Liposomes (0.75%) 54 Figure 4.3.3.2 Duration of Nerve Blockade for DOPCChol Liposomes (2.0%) 55. Figure 4.3.3.3 The Effect of Buffering Capacity on the Duration of Nerve Blockade ...56 Figure 5.1 Percent Seizure Activity versus Dose of 0.75% Liposomal Bupivacaine. 64 Figure 5.2 Probit of Seizure Activity versus Log Dose of 0.75% Liposomal Bupivacaine: ...65 Figure 5.3 Percent Seizure Activity versus Dose of 2.0% Liposomal Bupivacaine. 66 vii Figure 5.4 Probit of Seizure Activity versus Log Dose of 2.0% Liposomal Bupivacaine. 67 Figure 5.5 Percent Death versus Dose of 0.75% Liposomal Bupivacaine. 68 Figure 5.6 Probit of Death versus Log Dose of 0.75% Liposomal Bupivacaine. 69 Figure 5.7 Percent Death versus Dose of 2.0% Liposomal Bupivacaine, 70 Figure 5.8 Probit of Death versus Log Dose of 2.0% Liposomal Bupivacaine. 7.1 viii Abbreviations DOPC Dioleoyl phosphatidylcholine DSPC Distearoyl phosphatidylcholine HPS Hepes (pH 7.4, 50 mM), NaCl (150 mM) LUV Large unilamellar vesicle M L V Multilamellar vesicle pH; pH inside of the liposome pHo pH outside of the liposome POPC Palmitoyl oleoyl phosphatidylcholine SUV Small unilamellar vesicle Acknowledgments There are several people who have been of great assistance throughout my experiments over the past two years, but none have devoted as much to my thesis as Dr. Tom Madden. I would sincerely like to thank him for his tutelage as my thesis supervisor. He was very encouraging and continuously offered many helpful suggestions. I would like to thank Dr. Bernard Macleod for his advice on models to use for ebmparing the duration of analgesia and for the implications of my research in clinical . anaesthesia. Dr. Richard Wall generously offered his HPLC equipment for bupivacaine analysis and his contribution is appreciated. . ' -All of the colleagues in our research group (Ed, Miranda, Mac, Cliff, Delara, Charmaine, Cathy, Julie, Cindy, Xue and John) were always available for assistance and were awesome buddies to work with. Both Ed and Miranda have been especially helpful with my experiments and should be singled out. Research cannot exist without financial support and I would like to thank the Medical Research Council of Canada for funding our project and for my studentship. Above all, the most important and deeply appreciated support comes from my family. Both of my parents and my sister Jacklyn are always very supportive in all of my endeavors (both athletic and academic) and I want to thank them for all of their encouragement. 1 Introduction 1.1 The Clinical Requirement for a Safe Long Duration Local Anaesthetic Local anaesthetics were first discovered by Carl Roller (1857-1944) in 1884.1 He showed that cocaine could be used for topical anaesthesia in the cornea.2 Ever since this discovery, scientists have been looking for new ways to improve anaesthesia. The longest acting local anaesthetic clinically available is bupivacaine. - Its duration of action lasts for 6-8 hours depending upon the location and the quantity of bupivacaine injected. There are several limitations with this local anaesthetic. The cardiotoxicity and the central nervous system (CNS) toxicity of bupivacaine are well documented and are presented in section 1.3.3. In addition, extended blockade with repeated injections is know to cause tachyphilaxis.3 Therefore, catheterization is necessary for long duration blockade, The development of a long acting local anaesthetic with reduced toxicity would be a welcomed addition to the anaesthetic's armament against pain. A direct quote by G. R. Strichartz on neural blockade describes the need for such an anaesthetic:4 "... Although local anaesthetics are usually reliable drugs, they can be improved in several respects. Fast reversibility, to produce block of rapidly controlled duration, is desirable. Drugs of very long, but reversible, action would be useful for nerve block, such as intercostal block for postoperative pain. Selective block of functionally defined fiber types also would be a desirable clinical feature, allowing a differential titration of sensory and motor activities. In addition, reduced toxicity, both from accidental intravascular injection and from systemic accumulation after repeated injections, is a highly sought objective. As our knowledge about the relation between the molecular properties of local anaesthetics and their pharmacologic actions increases, we come closer to the reality of designing drugs and procedures to fulfill these criteria." 1 Liposomes are a promising area of research in this regard. They possess the capacity to slowly release local anaesthetic over a prolonged duration, thus reducing peak plasma concentrations and therefore toxicity.5 In addition, the possibility of selective sensory blockade exists. Clinical trials using liposomal bupivacaine have already shown selective sensory analgesia without motor blockade.6 2 1.2 Liposomal Drug Delivery Systems 1.2.1 Classification of Liposomes Liposomes are microscopic spheres consisting of a phospholipid bilayer encapsulating an aqueous core. They are spontaneously formed upon hydration of most phospholipids with water. They have been used as systemic delivery systems for a number of antineoplastics, antifungal agents and antibiotics7 and have also been found to enhance the therapeutic properties of local anaesthetics. This later application takes advantage of the slow rate of clearance of liposomes following intradermal or subcutaneous injection8 9 to provide a depot from which the local anaesthetic can be slowly released thereby prolonging nerve blockade. There are three main types of liposomes and they are shown in Figure 1.2.1. The multilamellar vesicle (MLV) forms spontaneously upon hydration of phospholipid. MLVs are very large vesicles (> 1 um) and may have tens to hundreds of lamellae (bilayers). Large unilamellar vesicles (LUVs) consist of a single bilayer and are much smaller in size ranging from 50 to 200 nm. Typically, LUVs are formed by extrusion through a defined size membrane under high pressure or by dissolving the phospholipid with detergent and removing the detergent by dialysis Small unilamellar vesicles (SUVs) are less than 50 nm in size and are usually formed by sonication. MLVs, LUVs and SUVs each have different physical characteristics. The trapped volume is a term used to describe the volume of liquid encapsulated by the liposome and is measured in pi per umole phospholipid (PL) . It is desirable to have a large trapped 3 volume when delivering drugs that are dissolved in the aqueous core so that greater quantities of drug can be delivered per mole lipid: LUVs have the largest trapped volume ranging from 1 to 5 ul umole"1 PL . Both SUVs and MLVs have smaller trapped volumes due to the smaller diameter and to the hundreds of lamellae occupying the inner core, respectively. Small Unilamellar Vesicles Large Unilamellar Vesicles Multilamellar Vesicles (MLVs) (<oo nm) * ' (50-200 nm) (>1 pm) Figure 1.2.1 The Morphology of Liposomes Vesicle stability is another important characteristic of liposomes. SUVs are very smalliiposomes and due to their size and sharp curvature of the bilayer, the packing of the phospholipids is less stable than with larger liposomes. MLVs are also less stable due to their sensitivity to osmotic forces as a result of their larger diameter! LUVs are the.most stable type of liposome. 4 1.2.2 Drug Uptake into LUVs Exhibiting a pH Gradient The first reported case of using a pH gradient to load compounds into liposomes was by Nichols et al10 in 1976. They showed that catecholamines such as dopamine, norepinephrine and epinephrine are concentrated inside liposomes exhibiting a pH gradient. They used these observations to support a mechanism whereby pH gradients could contribute to the uptake of catecholamines by sub-cellular storage sites. The possibility of using this technique to deliver drugs encapsulated inside liposomes was never mentioned in their paper. Several drugs with titratable groups are capable of highly efficient loading into liposomes using the pH gradient technique.11 Another positive aspect of the pH gradient technique is that drugs can be remotely loaded. Traditionally, in the formation of drug loaded liposomes, the drug and the lipids are mixed together prior to hydration of the lipid mixture. This often yielded low recoveries (< 10%) as most of the drug is centrifuged and decanted off, a very inefficient and expensive procedure. Using remote loading, drugs can be added after the pH gradient liposomes are formed and the drugs will redistribute across the bilayer in response to the pH gradient. The criteria for drugs to be efficiently loaded into pH gradient liposomes is that they possess a membrane permeable neutral form and, once inside the liposome, can be protonated and become relatively impermeable to the bilayer. 5 pH Gradient Liposomes Non pH Gradient Liposomes [Bullae » [BH+]0Utside [BH+]inside = [BH+]ou1side Figure 1.2.2 - Mechanism of Drug Loading into Liposomes Possessing a pH Gradient •- • The mechanism for loading bupivacaine into liposomes exhibiting a pH gradient is diagrammed in Figure 1.2.2. Bupivacaine (pKa=8.1) is present outside of the liposome as both the protonated and non-protonated species, the relative ratio being dependant upon 1 the exterior pH. If the exterior pH is 7.4 then we could expect about 17% of bupivacaine outside of the liposome to be non-protonated and in the neutral membrane permeable form. The driving force for the accumulation of drug within the liposome is based upon two principles. Firstly, the drug's neutral form equilibrates across the bilayer and is present in equal concentrations on both sides of the biomembrarie. As the aqueous core is buffered to pH 4.0, most of the bupivacaine becomes protonated as it crosses the bilayer and enters the aqueous core. Secondly, since the ratio of protonated to non-protonated bupivacaine is dictated by the pKa of the drug (pKa=8.1) and the pH of the buffer (pH=4.0), then 99 96% of bupivacaine must be protonated inside of the liposome. In order to satisfy both principles, bupivacaine will redistribute into the aqueous core of the 6 liposome. The distribution of bupivacaine across the bilayer is in accordance with a derivation of the Henderson-Hasselbach equation were the concentration of bupivacaine inside/outside of the liposome is proportional to the pH inside/outside of the liposome. ; The protonated drug inside the liposome is practically impermeable to the bilayer and becomes concentrated inside the vesicles. The neutral form of the drug in the bilayer exist in equilibrium with the neutral form outside of the bilayer and therefore bupivacaine will not be released from the bilayer until the concentration of bupivacaine outside of the liposome is decreased. This is a perfect system for slow sustained release as drugs can be efficiently loaded into liposomes and slowly released as the drug concentration outside of the liposome dissipates. 7 1 1.3 Properties of Bupivacaine - . • i i 1.3.1 Chemical Structure Local anaesthetics have three characteristics regions in their structure and they are: 1) a lipophilic group; 2) an intermediate chain, and 3) an amine substituent. The intermediate chain is divided into two classes, the amide local anaesthetics and the ester local anaesthetics. The amide local anaesthetics, such as bupivacaine and lidocaine, are much longer acting since they have an amide bond that is resistant to hydrolysis by butyrylcholinesterases. The ester local anaesthetics are rapidly hydrolyzed by butyrylcholinesterases in the blood and are known to cause allergic reactions due to their metabolism to p-aminobenzoic acid derivatives in certain individuals.12 Figure 1.3.1 - The chemical structures of some common local anaesthetics. 8 1.3.2 Mechanism of Action Local anaesthetics exert their action by binding to the cytoplasmic face of N-type sodium channels and, therefore, blocking sodium conduction.13 The blockade of sodium conduction across the axon prevents its depolarization which inhibits the propagation of the neuronal action potential. • • \ The blockade of sodium channels by most local anaesthetics is both frequency and time dependant.14 Channels in the resting state (which predominate at more negative : membrane potential) have a much lower affinity for local anaesthetics than activated (open state) and inactivated channels (which predominate at more positive membrane potentials). Therefore, local anaesthetics are more effective in rapidly firing axons than resting axons. Modification of the calcium and potassium concentrations can.alter the membrane potential. Calcium increases the surface potential of the membrane and this favors the ; low-affinity resting state, while the elevation of extracellular potassium depolarizes the i membrane and this favors the higher affinity inactivated state. Thus, high levels of ' extracellular potassium enhances: the effects of local anaesthetics. Although local anaesthetics can be shown to block a variety of other channels, including chemically gated synaptic channels, there is no convincing evidence that such actions play an important role in the clinical effects of these drugs.15 " . ; Since local anaesthetics exert their action on the cytoplasmic side of the sodium: channel, the degree of protonation of the anaesthetic is very important. The pK a's of mbst amide local anaesthetics fall in the range of 7.8-8.2. The local anaesthetics must be able to cross the neuronal bilayer and this predominantly occurs as the uncharged species. This means that if the extracellular pH is too low, as occurs with local infections, then there will be a greater proportion of the drug in the protonated species. Under these conditions, the local anaesthetics will be unable to cross the bilayer and they will be less efficacious. It1 should also be pointed out that the pH inside of the axon is also very important, as the ;, local anaesthetics bind to the sodium channel as the charged species.16 Therefore, the • local anaesthetics must cross the neuronal bilayer as the uncharged species and, once inside, it must become protonated to bind to the inactivated sodium channel to block the propagation of the neuronal action potential i 10 1.3.3 Bupivacaine Toxicity 1.3.3.1 Central Nervous System Toxicity CNS toxicity is the most common side effect of local anaesthetics. Local anaesthetic induced cardiovascular depression occurs less frequently but tends to be more serious and more difficult to manage.17 The CNS symptoms reported from the lowest to the highest local anaesthetic plasma concentrations are: numbness of the tongue, lightheadedness, visual and auditory disturbances, muscular twitching, unconsciousness, convulsions, coma and respiratory depression. The mechanism by which local anaesthetics exert their CNS effects involves the selective blockade of inhibitory pathways in the cerebral cortex 1 8 The toxicity of the local anaesthetics is directly related to the potency of their action. Bupivacaine is one of the most toxic of the local anaesthic agents which is clinically used. The order of toxicity from highest to lowest is as follows: cocaine > dibucaine > tetracaine > bupivacaine > etidocaine > mepivacaine > lidocaine > prilocaine > procaine > 2-chloroprocaine. It should also be pointed out that local anaesthetic toxicity is potentiated by hypoxia and hypercalcemia, so acute management of toxicity must minimize these conditions.19 11 1.3.3.2 Cardiac Toxicity Bupivacaine is more cardiotoxic than other local anesthetics. Several case reports20'21 have noted that accidental intravenous injection of bupivacaine may lead not only to seizures but also to cardiac collapse, from which resusitation is extremely difficult or even unsuccessful22 The cardiotoxicity of bupivacaine has also been reported in several papers using animal models.23 2 4 This is due to the fact that bupivacaine's block of sodium channels in the heart is potentiated by the very long action potential duration of the cardiac fibers. Lidocaine dissociates rapidly from sodium channels at potentials negative to -85 mV, while bupivacaine dissociates much more slowly or incompletely from the resting channels. Whereas the recovery from lidocaine block is rapid (0.15 sec), bupivacaine is slower (1.5 sec) so that the block accumulates even at relatively low heart rates.25 The most common electrocardiographic finding with bupivacaine toxicity is slow * 12 idioventricular rhythm with broad QRS complexes and electromechanical dissociation T 12 1.4 Previous Studies on Sustained Release Vehicles for Local Anaesthetics 1.4.1 Previous Studies on Non-Liposomal Delivery Systems for Local Anaesthetics Several attempts have been made to extend the duration of local anaesthetics by altering their pharmacokinetics with the addition of a vehicle that slowly releases the drug. These included the addition of dextran or iophendylate (a lipid carrier) or the formation of microdroplets, microspheres, microcrystals, lipospheres or implants. Most of these delivery systems have shown significant prolongations of analgesia over the parent formulation and their design will be briefly outlined. The addition of dextran to local anaesthetics was thought to increase their duration of analgesia by increasing the viscosity of the injecting medium or by increasing the hydrophobic binding sites for the local anesthetics and thus limiting the rate of diffusion of the drug from the site of injection. However, the data available is inconclusive. Two of the early papers that used dextran to extend the duration of local anesthetics and reduce toxicity were successful in increasing the duration of analgesia.26'27 Yet, later papers ' - • • published in this area showed no significant effect upon the addition of dextran to the local anesthetic solution.28 2 9 It appears that dextran may extend the duration of shorter acting anesthetics such as 2-chloroprocaine or prilocaine, but has no significant effect on longer acting local anesthetics such as bupivacaine.30 Iophendylate is a radiopaque iodine containing compound that is sometimes used in radiography as a contrasting agent to show the spinal canal (myelogram). Langerman et al have used this lipid formulation to prolong the duration of epidural37 and spinal"2 13 anesthesia using procaine, lidocaine and tetracaine. Tetracaine had the longest duration of analgesia and this phenomenon was believed to be due to its hydrophobicity. The more hydrophobic the anaesthetic, the longer it will remain associated with the lipid carrier and the longer its duration of action. The use of iophendylate as a lipid carrier is somewhat • • 32 controversial as aseptic arachnoiditis was reported to occur after intrathecal injection, i Microdroplets of methoxyflurane, a general anaesthetic, have also been used in an effort to extend the duration of analgesia: The microdroplets were formed by suspending lecithin and methoxyflurane in buffer using a Polytron apparatus followed by sonication to produce microdroplets. Low speed centrifugation to remove smaller particles and ultrafiltration to remove particles with diameters larger than 4,500 nm were also completed before the samples were ready to use. The results from Haynes et aVs paper" showed a very long duration of blockade against electrical stimuli lasting 23 hours in the rat tail model with the pain threshold at 65 mV. In a paper later published with Haynes as one of the co-authors, lecithin coated tetracaine microcrystals were used to extend the duration of analgesia in the rat tail to 43 hours.34 The solution which blocked the tail for 43 hours contained 10% lecithin-coated tetracaine microcrystals. When a 10% solution of free tetracaine was injected, 3 of 5 animals died of severe toxicity and the other two survivors developed wet gangrene of the tail. To form the microcrystals, they placed tetracaine hydrochloride with 20 g of egg lecithin, 180 mg of methyl paraben, 20 mg of propyl paraben and 3 mg of gentamicin sulfate in a beaker. The compounds were then hydrated with mannitol buffer and sonicated for 45 minutes. The pH of the buffer was titrated to 7.2 - 7.4 and the vesicles \ 14 were stored in the refrigerator for 24 hours, allowing the tetracaine microcrystals to settle. The settled crystals were placed in sterile glass serum vials and sterilized with 1.5 megarads of gamma irradiation. Lipospheres were developed as a slow releasing drug delivery system by the Nova Pharmaceutical Corporation of Baltimore, Maryland. Lipospheres are water dispersible microparticles composed of a solid hydrophobic core of triglycerides containing the drug with phospholipids embedded on the surface of the core. Lipospheres containing bupivacaine are prepared by dissolving the drug in tristearin and subsequently adding egg lecithin and buffer. The mixture is homogenized with intermittent cooling using dry ice and acetone to obtain a uniform dispersion of microparticles. Bupivacaine (2.0%) lipospheres blocked the rat tail for 3 hours while 0.5% bupivacaine with epinephrine lasted for only 0.5 hours 3 5 Microspheres are synthetically prepared biodegradable polymers capable of slowly releasing drugs. Microspheres made of polylactic acid were prepared by a solvent evaporation process36 and averaged 50 um in diameter.37 The polylactic acid and the local anaesthetic were dissolved in an organic solvent with a low boiling point and added dropwise to a solution containing 1% gelatin or 1% sodium alginate. Heat or reduced pressure is applied to the organic solvent and the microspheres were collected by filtration or centrifugation. Using the guinea pig pin-prick test, the dibucaine microspheres showed an ultra long duration of blockade lasting over 450 hours compared to free dibucaine lasting a mere 90 minutes.38 Le Coree et at9 reported an 8 hour spinal block in the rabbit using bupivacaine in polylactic acid 9000 microspheres versus 2 hours for the free 15 bupivacaine solution. Polypropylene carbonate and polyethylene carbonate microspheres have also been made and show similar sustained release kinetics as the polylactic acid microspheres.40 Biodegradable implants have also been used to extend the duration of local anaesthetics which is similar to the Norplant® system, a contraceptive implant that slowly releases levenorgestrel over a period of 5 years The polymer used for local anaesthetic release relies upon the cleavage of polyanhydride bonds for its degradation. Masters et at" studied these biopolymers in the rat where the implants were surgically inserted next to the sciatic nerve. The implants showed a very long duration of neural blockade lasting from 2 to 6 days, depending upon the formulation studied. The formulations which lasted for 2 days showed less motor blockade than the formulations which lasted for 6 days. This would be expected as the shorter duration formulations had less bupivacaine present and it therefore made it much more difficult to block the myelinated A-oc fibers which are responsible for motor function and have a higher C m (minimum blocking concentration).42 There was no nerve damage reported on several biochemical assays completed with, the sciatic nerve. The obvious difficulty in using the implants is the cost and complications associated with surgical insertion Another potential problem associated with polymer implants is the formation of fibrous capsules which may prematurely slow the release of the drug below the anaesthetic's blocking dose and inhibit the drug from reaching the nerve. It is difficult to compare the various vehicles used for extended duration local anaesthesia due to a variety of factors: 1) the different animals used to compare the 16 duration of analgesia; 2) the different local anaesthetics used in the vehicles; 3) the different scales used to assess analgesia, and 4) the different locations of nerve block used in the experiments. Nonetheless, most of the vehicles used have increased the duration of analgesia and, since this is typically accomplished by a slow localized release of local anesthetic, we can expect the vehicles to also reduce toxicity. 1.4.2 Previous Studies on Liposomal Local Anaesthetics The concept of using liposomes as a method for increasing the duration of nerve blockade is not a novel idea. Several papers have been published on liposomal local anaesthetics; however, the techniques used to prepare the liposomes and the methods for drug loading have not made use of recent advances in liposome technology. Previous papers have typically relied upon hydrophobic association of the drug . with the liposomal bilayer as a mechanism for drug loading. In 1986, Djordjevich et al were the first researchers to extend the duration of action for bupivacaine using liposomes.43 They injected the free drug and the liposomal formulation in the base of the rat tail and noticed that the bupivacaine associated with MLVs blocked response to the tail immersion in water at 50°C for 34 hours, while the free drug lasted for only 5.7 hours. Legros et at4 have also relied upon hydrophobic association of bupivacaine with the bilayer as a mechanism of loading bupivacaine into liposomes. Typically, the procedure used among researchers to prepare liposomal local anaesthetics is quite similar. The free base form of the local anaesthetics are added to egg phosphatidylcholine : cholesterol (4:3, mole ratio) and dissolved in chloroform. The chloroform is removed 17 under an atmosphere of nitrogen to form a dry lipid/local anaesthetic powder. Multilamellar liposomes are formed by hydrating the powder with buffered saline. The 1 free bupivacaine not associated with the MLVs was removed by washing the MLVs with buffered saline and centrifuging, five times.44 The pellet was resuspended in isotonic buffer and used for injection. The drug release rates from these MLVs are slow since the drugs trapped within the inner core of the M L V must pass through several bilayers before escaping the vesicle. This technique is labor intensive and costly as most of the drug is lost during the washings. In spite of the procedure used, the technique yields good results. Legros et al showed that bupivacaine loaded in MLV liposomes has both a longer duration of action and a reduction in toxicity45 over the free drug. The reduction in toxicity was believed to be due to the lower peak plasma bupivacaine concentrations immediately after injection as a result of the drug's slower release from the liposome depot.5 Legros et al also used this formulation for epidural injection in humans in an open, non-randomized study.46 Twenty-six patients were chosen post-operatively. Half of the patients were to receive 0.5% bupivacaine with 1:200,000 adrenaline and the other half were receiving 0.5% liposomal bupivacaine. The liposomal bupivacaine formulation lasted for 6.25 hours compared to the free drug formulation of 3.2 hours. In addition, the liposomal formulation showed no signs of motor blockade in the 14 patients while for free drug 9 of 12 patients experienced motor blockade. Umbrain et al showed using scintillography that liposomes, when injected epidurally into rats, remained at the site of 18 injection over 24 hours.9 Clearly, the altered pharmacokinetics from the use of liposomes as a vehicle significantly altered the duration of nerve blockade. ! Other researchers have shown similar results with liposome formulations. Mashimo et a/*7used liddcaine with phosphatidylcholine:cholesterol MLVs to prolong canine epidural anesthesia from 61 min. with free lidocaine versus 170 min. with the liposomal formulation. Liposomes have also been used to prolong the action of alfentanil, an opioid analgesic. Bernards et at8 have used liposomal alfentanil to lower drug plasma levels and to prolong the duration of spinal analgesia in the rat. In all of the publications on liposomal local anaesthetics to date, the techniques used for drug loading and liposome preparation were quite different from the methods I employed. In the present work I have taken advantage of two important developments in liposome technology. First, I have employed liposomes of well defined size, large unilamellar vesicles, produced by an extrusion procedure. These systems, which exhibit relatively large trapped volumes, allow greater quantities of drug to be encapsulated.49. In addition, I have used a transmembrane pH gradient to efficiently encapsulate bupivacaine within the liposomal carrier. This technique has been termed "remote drug loading" and reflects the redistribution of drugs which are weak bases or weak acids in accordance with the Hendersen Hasselbalch equation.11 The remote loading procedure not only allows efficient drug accumulation within liposomes but should also provide a mechanism whereby release rates can be controlled. (Refer to Figure 1.2.2) 19 2. Uptake of Bupivacaine into LUVs Exhibiting a Proton Gradient 2.1 Introduction ! One of the challenges facing the use of LUVs to deliver local anaesthetics is loading the drug into the liposome. The pH gradient technique, as outlined in section 1.2, has circumvented the drug loading problem by concentrating the drugs in the aqueous core by their protonation and maintenance of the liposome's pH gradient. In tins section, the uptake of bupivacaine into LUVs will be examined. It is very important to achieve efficient loading of bupivacaine within the liposome with minimal free drug outside of the vesicle. Once the drug is loaded, it can then be slowly released to ' provide a longer duration of blockade and to reduce systemic toxicity. If most of the drug were not initially entrapped, then the potential benefits of using a liposomal system for -! drug delivery would not be realized. In order to determine the percentage of bupivacaine which resides inside of the liposome, we must separate free drug from encapsulated drug. I have used three techniques to examine this: 1) gel exclusion chromatography, 2) density gradient centrifugation, and 3) filtration. Each technique yields complimentary data that is useful to explain the differences in uptake between pH and no pH gradient liposomal bupivacaine. I In gel exclusion chromatography, we are limited to studying drug uptake at low lipid concentrations (5 mM) and using smaller diameter liposomes (100 nm). This is due to the nature of the column. Larger and more concentrated liposome solutions, which are used in the release kinetic and nerve block experiments, will not elute efficiently from the column. The gel exclusion method of studying drug uptake has been used in several other papers on drug encapsulation in liposomes.11 The data generated from the 100 nm 5 mM PL . vesicles may not reflect the actual uptake in the 600 nm, 100 mM P L liposomes due to four main factors. Firstly, the trapped volume of a 100 nm vesicle is about three fold smaller than a 600 nm vesicle. In vesicles with larger trapped volumes there is a greater aqueous compartment per mole lipid for the drug to be entrapped. Therefore at equivalent molar concentrations, there is a greater sink available for bupivacaine storage. In addition, larger trapped volume liposomes will have greater buffering capacities per mole lipid since the aqueous core is larger. Therefore, vesicles with larger trapped volumes should have more efficient drug loading than vesicles with smaller trapped volumes. Secondly, although the dfug-to-lipid ratios are similar, the concentration of drug and lipid used are much lower. Thirdly, as the liposomes pass down the column, free bupivacaine is lost. Therefore more bupivacaine will permeate out of the liposome to re-establish equilibrium, resulting in an underestimation of the trapping efficiency. Fourthly, bupivacaine is i dissolved in buffer prior to its addition to the liposomes, while in the release kinetic and nerve block experiments bupivacaine is added in solid form directly to the liposomes which may alter its uptake efficiency. Although this technique may have limitations in extrapolating data to our 600 nm, 100 mM liposome systems, it is a highly accurate and reliable technique for demonstrating the kinetics of drug loading into pH gradient liposomes. 21 The density gradient centrifugation procedure was used to determine the extend of bupivacaine association to the 600 nm, 100 mM liposomes. During the nerve block studies in the guinea pigs backs, it was discovered that not all of the bupivacaine would dissolve in the control 2.0% no pH gradient solution. Therefore, we needed an experiment to look at the uptake of the drug in these systems. The control 2.0% no pH gradient solution was used to compare the effectiveness of pH gradient liposomes for both drug loading and duration of nerve blockade. A density gradient centrifugation procedure was developed whereby the liposomal bupivacaine sample would be placed on top of a glucose fraction of a higher specific gravity, By using this procedure, any bupivacaine which was not fully dissolved in the liposomal solution would, precipitate down into the glucose fraction. The top fraction would then be assayed for bupivacaine and DOPC. The drug-to-lipid ratios determined from this centrifugation experiment not only: reflect the amount of bupivacaine associated with the liposome, but also the amount of bupivacaine which is soluble in the HPS buffer. However, it should be noted that the solubility of bupivacaine in the HPS buffer is relatively low, in comparison to other local anaesthetics (Table 1), and the solubility is also pH dependant (Refer to Figure 2.2.5). Table 1 Solubility of Some Common Local Anaesthetics at pH 7.37 Agent Class pKa (25°C) Solubility (mg HC1 Salt /ml) bupivacaine amide 8.1 0:83 mepivacaine amide 7.8 15 lidocaine amide 7.9 . 24 tetracaine ester 8.46 1.4 22 2 .2 Methods 2.2.1 Preparation of Large Unilamellar Vesicles Exhibiting a Proton Gradient Mixtures of DOPC and cholesterol (55:45 molar ratio; including DPPC- 1 4C) were weighed and dissolved in benzene:methanol (95:5, v/v) at approximately 100 mg/kg. The solution was lyophilized by rapidly freezing the sample in liquid nitrogen then transferring it to a vacuum pump at. <50 millitorr for 5 hours with a liquid nitrogen trap attached. Multilamellar liposomes (110 mM) were prepared by the hydration of the lyophilized lipid mixture with pH 4.0 300 mM citrate or 150 mM NaCl, 50 mM HEPES pH 7.4 (HPS) buffer. The resulting MLVs were subjected to five freeze-thaw cycles by freezing the sample in liquid nitrogen for five minutes followed by incubation at 37°C for five minutes. Large unilamellar vesicles (LUVs) were made by passing the MLVs ten times through an Extruder (Lipex Biomembranes, Vancouver, British Columbia, Canada) at 37°C employing two stacked 600 nm Nuclepore polycarbonate filters. A transmembrane pH gradient was formed by dialyzing (Spectra Por 2 dialysis ; tubing, M.W.C.O.=I2-14,000) the LUVs against 200 volumes of HPS at 4 C for 48 hours with one change of external buffer after 24 hours. 23 2.2.2 Determination of the Trapping Efficiency for Bupivacaine in LUVs 2.2.2.1 Gel Exclusion Chromatography Since concentrated 600 nm liposomes at 100 mM phospholipid will not efficiently elute through a Sephadex G-50 gel exclusion column, the liposomes were sized to 100 nm instead of 600 nm and the lipid concentration was lowered from 79 mg/ml to 50 mg/ml. The technique to establish a pH gradient across the liposome was also changed from the method stated in section 2.2 1. A pH gradient was formed by passing the liposomes down a Sephadex G-50 gel exclusion column prepared with pH 7.4 HPS buffer. The pH gradient liposomes were collected from the column and the phospholipid concentration was determined by a standard phosphate assay.51 Bupivacaine uptake studies were performed using LUVs prepared with 100 nm pore size polycarbonate filters. Bupivacaine was added to LUVs (5 mM phospholipid) to give the final drug concentrations shown in Figure 2.2.1 and the samples then incubated at 37 C. At various times up to two hours, aliquots (100 pi) were removed and unencapsUlated drug separated from the liposomes by the use of 1 ml minicolumns.52 The columns were centrifuged at 2500 rpm for three minutes and the eluent then assayed for drug and phospholipid as described below under Analytical Procedures. Controls were also run consisting of liposomes that did not exhibit a pH gradient (pH 7.4 in and pH 7.4 out) to quantify non-specific bupivacaine binding. The uptake into liposomes composed of two different lipid mixtures (DOPCChol and DSPCChol) was studied. 24 2.2.2.2 Density Gradient Chromatography Liposomal bupivacaine (150ul) was carefully added on top of 100 ul of glucose (325 mM) in a polypropylene microcentrifuge tube. The microcentrifuge tube was held in a test tube and centrifuged for 5 minutes at 3000 rpm (Haerus Megafuge 1.0). The liposome fraction was assayed for bupivacaine and DOPC using HPLC and liquid scintillation counting, respectively. 2.2.2.3 Filtration In this procedure, the distribution of bupivacaine across the bilayer was determined by physically separating the liposomes from the external buffer through filtration. Microcentrifuge filtration tubes where purchased from Costar Scientific Corporation (Cambridge, MA) which had a M.W.C.O of 30,000. A 200 ul sample was placed on top of the filter and the sample was centrifuged for one hour at 15,000 rpm in a Sorvall MC 12V microcentrifuge by DuPont. A 15 ul filtered aliquot was collected and analyzed for bupivacaine by HPLC. Vesicles containing 3H-DPPC were used as a control to check the filter's impermeability to liposomes. Once the concentration of bupivacaine in the.aliquot was determined, the efficiency of encapsulation was determined based upon the trapped volume of the liposome and the concentration of the drug in the unfiltered sample. This procedure is the most accurate method for determining the efficiency of encapsulation and the amount of free drug in the 0.75% and 2.0% liposomal bupivacaine samples. The filtration procedure allows non-diluted vesicles of 600 nm in size to be used. 25 2.2.3 Analytical Procedures Bupivacaine was quantified by HPLC using mepivacaine as an internal standard. To the sample (300ul) was added mepivacaine (20 ul containing 83 ug ml"1 in H20) followed by acetonitrile (300ul). The mixture was then incubated at 60°C for five minutes to ensure rupture of the liposomes and release of the entrapped drug. The sample was then centrifuged at 2500 rpm for five minutes to pellet the lipid and the supernatant taken for HPLC analysis. Using a 10 cm ODS colurnn, bupivacaine and mepivacaine were eluted under isocratic conditions (35% sodium diphosphate, pH 5.0; 65% acetonitrile containing 20 mM triethylamine) at 1.0 ml min"1 and detected at 210 nm. All of the solvents used were of HPLC grade. In some experiments, phospholipid concentrations were determined by phosphate assay following perchloric acid digestion.51 Vesicle size distributions were determined by quasi-elastic light scattering using a Nicomp 370 submicron particle sizer as described by Madden et al.53 2.3 Results 2.3.1 Gel Exclusion Chromatography In Figure 2.2.1 the rate and extent of bupivacaine uptake by LUVs exhibiting a pH gradient for three different drug-to-lipid ratios are shown. Rapid drug accumulation is observed with uptake being essentially complete at 5 minutes for all samples. Higher 26 initial drug-to-lipid ratios achieved lower efficiencies of encapsulation. In the 3 mM bupivacaine pH gradient sample, the drug-to-lipid ratio that would represent 100% uptake is 600 umoles bupivacaine per mmole DOPC. The maximum drug-to-lipid ratio achieved for the 3 mM pH gradient sample was 260 umoles bupivacaine per mmole DOPC which represents 43% of the drug actually encapsulated. In the 1 mM bupivacaine sample the efficiency of drug uptake was 78%. The initial drug-to-lipid ratio also affected the stability of drug uptake in DOPC:Choi LUVs. In both the 2 mM and 3 mM bupivacaine pH gradient samples the drug-to-lipid ratio steadily decreased over the course of the experiment. In the 1 mM bupivacaine sample, the uptake was stable over the course of the experiment To confirm that the levels of bupivacaine uptake seen reflect drug redistribution in response to the imposed pH gradient, control LUVs that did not exhibit a pH gradient (pH 7.4 in and pH 7.4 out) were incubated with 3 mM bupivacaine under the same conditions. As shown in Figure 2.2.1, these control vesicles exhibited only low levels of drug binding. In Figure 2.2.2 drug uptake was compared in DSPCChol LUVs Rapid drug r , accumulation was also observed with uptake being essentially complete at 5 minutes for 0.89 mM and 1.33 mM bupivacaine and by 30 minutes for 1.78 mM drug. Unlike the i DOPCChol LUVs at high drug-to-lipid ratios, this uptake is stable under equilibrium conditions with no decrease in liposomal bupivacaine content throughout the experiment. As would be expected, uptake efficiency is dependant on the initial drug lipid ratio, increasing from about 70% at 1.78 mM drug:5 mM phospholipid to 80% at 0.89 mM drug.5 mM phospholipid. As in Figure 2.2.1, control liposomes were also used to confirm 27 that the levels of bupivacaine uptake seen reflect drug redistribution in response to the .; . • • j imposed pH gradient. The control LUVs were incubated with 1.78 mM bupivacaine | under the same conditions and exhibited only low levels of drug binding. 2.3.2 Density Gradient Centrifugation In Figure 2.2.3, the association of bupivacaine with DOPC: Choi LUVs was • - . i determined using density gradient centrifugation as shown. Within 15 minutes, maximum drug-to-lipid ratios (100% efficiency) were achieved for both the 0.75% (270/265 umoles bupivacaine mmole"1 DOPC) and 2 0% (720/716 umoles mmole"1 DOPC) pH gradient samples. In the control no pH gradient samples, drug-to-lipid ratios were 81% of maximum for both the 0.75% (215/266 umoles mmole*1 DOPC) and the 2^0% (570/706^ umoles mmole"1 DOPC) samples. In the 2.0% sample the drug-to-lipid ratio continued to rise over 48 hours, achieving a final drug-to-lipid ratio of 640 umoles mmole"1 which represents 90% of bupivacaine either in solution or associated with the liposomes. Large bupivacaine precipitate aggregates settled at the bottom of the glucose gradient during '• some of the initial time points in the control liposome samples (no pH gradient). It should also be noted that in the 2.0% liposomal bupivacaine samples, the liposomes exhibiting a pH gradient have a much faster rate of initial drug association versus the liposomes ; lacking a pH gradient. In Figure 2.2.5, the pH dependant solubility of bupivacaine in 50 mMHepes:150 mM NaCl buffer (HPS) is shown. Bupivacaine is readily soluble in HPS buffer (up to 26 mg/ml) at pH's lower than 6.0. However, at pH's above 7.0 the solubility is dramatically 28 reduced below 1.0 mg/ml. This is an unfortunate hurdle which has made the 2 0% control no pH gradient vesicles precipitate some bupivacaine. 2.3.3 Filtration : Two concentrations of bupivacaine, 0.75% and 2.0%, were analyzed for uptake | into pH gradient DOPCChol LUVs. In Figure 2.2.4, a diagram is shown to explain the ; concentration and percent distribution of bupivacaine both inside and outside of the liposomes. In the 0.75% sample, the concentration of bupivacaine outside of the liposome is 1.98 mg/ml. Based upon a trapped volume of 3 pi per umole lipid, this would, mean that 1.39 g of the 7.50 g in total of bupivacaine is free outside of the liposome in solution. Therefore 6.11 g of the bupivacaine is trapped inside of the liposome and the concentration of bupivacaine inside of the liposomes is 20.4 mg/ml. The, efficiency of I entrapment for the 0.75% bupivacaine sample is 82%, In the 2.0 % sample, the concentration of bupivacaine outside of the liposome is 10.26 mg/ml which means that ^ 718 g of the 20.0 g total bupivacaine is free outside of the liposome in solution. Therefore 12.82 g of the bupivacaine is trapped inside of the liposome and the concentration of bupivacaine inside of the liposomes is 42.7 mg/ml. The efficiency of entrapment for the 2.0% bupivacaine sample is 64%. 29 Figure 2.2; 1 - Gel Exclusion Chromatographic Determination of Bupivacaine Uptake into L U V DOPC:Choi Liposomes -1.0 mM (#), 2 0 mM (•), 3 0 mM (•) bupivacaine with 5 mM DSPC.Chol (55:45) liposomes at 37°C. The pH inside (pH;) th liposomes was 4.0 (300 mM citrate) and the pH outside (pHo) was 7.4 (50mM HEPES and 150 mM NaCl). Liposomes exhibiting no pH gradient (3.0 mM bupivacaine (V) v also passed through the minicolumns. 30 Figure 2.2.2 - Gel Exclusion Chromatographic Determination of Bupivacaine Uptake into LUV DSPCrChol Liposomes - 089 mM (#), 1.33 mM (A), 1.78 mM (•) bupivacaine with 5 mM DSPCChol (55:45) liposomes at 20°C. The pH inside (prL) the liposomes was 4.0 (300 mM citrate) and the pH outside (pFL) was 7.4 (50mM FfEPES and 150 mM NaCl). Liposomes exhibiting no pH gradient (1.78 mM bupivacaine (A) were also passed through the minicolumns. O QL CO Q Q) o £ £ o £ d .c o O <D "co o CO > CL CQ 300 250 h 200 h 150 100 H 20 40 60 80 Time (minutes) 100 120 31 Figure 2.2.3 - Density Gradient Centrifugual Determination of Bupivacaine Association with DOPC:Chol Liposomes - The association of bupivacaine with 600 nm DOPC: Choi liposomes (87 mM, PL.) was determined for 0.75% bupivacaine with (o) and without (A) a pH gradient and 2.0% bupivacaine with (•) and without (V) a pH gradient. 800 Q. m 200 100 0 10 15 20 25 30 35 40 45 50 55 Time (hours) 32 Figure 2.2.4 - Determination of Bupivacaine Distribution by Filtration The rectangular box represents a 1.0 ml sample of liposomal bupivacaine. The left compartment respresents the 0.3 ml volume inside the liposome and the right comparment represents the 0.7 ml volume outside of the liposome. Distribution of 0.75% Liposomal Bupivacaine Liposome Free Concentration of Concentration of Bupivacaine=20.4 mg/ml Bupivacaine=1.98 mg/ml % of total Bupivacaine % of total Bupivacaine =6.11mg/7.5mg*100% =1.38mg/7.5mg*100% = 8 2 % 0.300 ml = 1 8 % 0.700 ml Distribution of 2.0% Liposomal Bupivacaine Liposome Free Concentration of Concentration of Bupivacaine=42.7 mg/ml Bupivacaine=10.3 mg/ml % of total Bupivacaine % of total Bupivacaine =12.8mg/20.0mg*100% =7.18mg/20mg*100% = 6 4 % 0.300 ml = 3 6 % 0.700 ml Figure 2.2.5 NaCl - pH Dependant Solubility of Bupivacaine in 50 mM Hepes, 150 34 2.4 Discussion The uptake studies using gel exclusion chromatography showed that bupivacaine is rapidly loaded into both DOPCChol and DSPCChol LUVs exhibiting a pH gradient. In both systems, higher drug-to-lipid ratios exhibited lower efficiencies of drug uptake. The reason the uptake into LUVs is limited at high drug-to-lipid ratios is mostly due to the rise in the pH inside of the liposome as more drug is accumulated. When bupivacaine enters the aqueous core of the liposome it becomes protonated and this raises the internal pH of the liposome. The pH gradient across the bilayer is the driving force for drug accumulation inside of the liposome. As the pH gradient is reduced, from the accumulation of increasing amounts of bupivacaine, bupivacaine is no longer driven into the liposome. Therefore, lower quantities of bupivacaine are more efficiently encapsulated than larger quantities of bupivacaine for the same concentration of pH gradient liposomes. The stability of drug uptake is dependant upon the lipid composition of the liposome. In DOPC:Choi LUVs bupivacaine uptake was unstable at higher drug-to-lipid ratios, while in the DSPCChol liposomes uptake was stable at all drug-to-lipid ratios \ tested. At higher drug-to-lipid.ratios higher bupivacaine concentrations will partition into the membrane bilayer. This may destabilize the bilayer and render it more permeable to counter ions such as Na + or K +. An influx of Na + or K 7 will reduce the pH gradient and: allow the efflux of bupivacaine. The reason DOPC shows more unstable uptake at higher drug-to-lipid ratios as compared to DSPC is likely due to differences in bilayer fluidity. Both phospholipids contain two eighteen carbon chains; however, DOPC contains two. unsaturated bonds in the acyl chain while the acyl chains of DSPC are completely 35 saturated. The unsaturated double bondss in DOPC produce a kink in the hydrophobic chains which force neighboring phospholipids to be further apart. This increases the , fluidity of the DOPC bilayer. Bupivacaine, at high concentrations in the DOPC:Choi bilayer, may increase the permeability of the membrane to Na+ or K + ions resulting in a decrease in the magnitude of the pH gradient and subsequent release of bupivacaine. The DSPC bilayer is much more rigid and Na+ or K + ions may not be permeable even at larger drug-to-lipid ratios. High drug-to-lipid ratios required longer incubation times for the DSPC vesicles. This would be expected as there would be more bupivacaine to be loaded into equal quantities of vesicles. A comparison of bupivacaine's loading capacity in liposomes exhibiting a pH gradient versus those lacking a pH gradient shows dramatically different levels of drug uptake. The pH gradient sample had 80% of the bupivacaine associated ! with the DSPCChol LUVs while the no pH gradient sample had merely 3%. This indicates that the pH gradient technique is a highly efficient method to load bupivacaine into liposomes. Similar efficiencies of encapsulation were also achieved for the DOPCChol pH gradient LUVs at 1 mM bupivacaine: 5 mM DOPC where 78%. o f the drug was encapsulated while in the control vesicles lacking a pH gradient there was little drug loading. In the density gradient centrifugation procedure, the fraction containing pH gradient liposomes exhibited maximum drug concentrations upon incubation with 0.75% and 2.0% bupivacaine after 15 minutes. This would be expected as most of the bupivacaine would be present in the aqueous core of the liposome. In the case of vesicles 36 lacking a pH gradient, the liposomal fraction following incubation with either 0.75% or 2.0% bupivacaine exhibited only 81% of the maximum potential drug concentration after 1 hour. The lower percent association of bupivacaine with non pH gradient liposomes is in good agreement with observations from gel exclusion chromatography. Bupivacaine is not actively loaded inside of the control no pH gradient vesicles and, therefore, less drug is associated with this fraction. If only the bupivacaine present in the aqueous core and in the membrane bilayer were included in the liposomal fractions, we would expect based on the gel exclusion chromatography data, significantly lower drug concentrations for the no pH gradient liposomes than represented in the data from the density centrifugation experiment. The higher levels seen are likely due to factors. First a larger proportion of the bupivacaine partitioning into the hydrophobic milieu of the bilayer. Second, the centrifugation procedure includes in the liposomal fraction any free bupivacaine that is dissolved in the, buffer. Further, in the minicolumn procedure, passage of the LUVs down the gel column will results in the loss of any bupivacaine associated with the bilayer.' In order to accurately determine the extent of bupivacaine association with pH gradient liposomes under similar conditions to those used in the animal experiments (i.e. at high lipid concentrations with 600 nm vesicles) we employed a filtration procedure. The data obtained from this experiment are in good agreement with the gel exclusion data In the 0.75% samrjle, 82% of bupivacaine is encapsulated as determined through filtration.' Using the gel exclusion method, 78% of the drug was entrapped at a drug-to-lipid ratio of 200 nmoles bupivacaine per umole DOPC. While the drug-to-lipid ratio for 0.75% 37 liposomal bupivacaine is slightly higher (231 bupivacaine per umole phospholipid), both methods show good levels of drug uptake. In the 2.0% sample, 64% of the drug is encapsulated as determined through filtration. The gel exclusion method showed much lower levels of drug uptake at similar drug-to-lipid ratios (43% at 600 mmoles bupivacaine per mmole DOPC). The discrepancy between the data obtained from the two methods is likely due to the release of bupivacaine as LUVs pass down the gel exclusion column. The liposomes were designed to release bupivacaine as more bupivacaine is lost outside of the vesicles. As the liposomes pass down the gel column, bupivacaine is released as more of the free drug is removed. This will cause an underestimation of the true efficiency of encapsulation using the gel exclusion method. None the less, the data obtained from the gel exclusion chromatography is -excellent evidence to demonstrate the very rapid and highly efficient uptake that is achieved by using a pH gradient to load bupivacaine inside liposomes. Rapid and efficient drug uptake is key to the success of establishing a long lasting local anaesthetic with reduced toxicity. Another important attribute of the pH gradient liposomes is the ease of sample preparation. Upon demand, pH gradient liposomes prepared outside of the hospital clinic could be mixed with the free drug at room temperature for 1 hour and be ready for use. This would help extend the shelf life of the liposomal formulation and speed its acceptance into hospitals as pharmacists would not have to be trained in liposome preparation. 38 3 Release of Bupivacaine from Large Unilamellar Vesicles 3.1 Introduction Understanding the release kinetics of bupivacaine from the liposome carrier in vitro aids the interpretation of data from animal models. Once bupivacaine was loaded into LUVs, the liposomal bupivacaine was placed inside a dialysis bag and dialyzed against a large excess of HPS buffer. Samples were then removed from within the dialysis bag over time to determine the rate of release of bupivacaine. This system is more straightforward than in vivo release kinetic analysis due to the lack of confounding variables such as serum proteins, macrophages, etc. which makes understanding the liposomal system significantly more complex. 3.2 Methods LUVs were prepared as stated in section 2.2.1 and bupivacaine (266 umole/mmole DOPC) was incubated with the liposomes (100 mM DOPC) at 37°C for one hour. The drug loaded vesicles (2 0 ml) were placed inside Spectra Por 2 dialysis tubing (molecular weight cutoff=12-14,000) and dialyzed against 500 volumes of pH 7.4 HPS buffer, changing the buffer after four and eight hours. At various times up to 24 hours, samples were removed from inside the dialysis bags and assayed for bupivacaine and phospholipid as previously described in section 2.2.3. 39 3.3 Results Bupivacaine release rates from both the POPC and DOPC lipid mixtures exhibiting a pH gradient were similar. As shown in Figure 3.3.1, bupivacaine release from POPC:Cholesterol LUVs is initially very rapid over the first hour and thereafter exhibits slower more sustained drug payout. In DOPC Choi LUVs, the rate of bupivacaine release was compared between liposomes exhibiting a pH gradient (pHi=4.0) versus those lacking a pH gradient (pHi=7.4). As shown in Figure 3.3.2, DOPC control vesicles (lacking a pH gradient) had much faster rates of drug loss than the DOPC LUVs exhibiting a pH gradient. By forming a pH gradient, the rate of bupivacaine loss from the LUVs was significantly reduced. 40 Figure 3.3.1 Bupivacaine Release Kinetics from POPC:Choi Liposomes - Ail initial drug-to-lipid ratio of 266 umole bupivacaine mmole"1 POPC was added into dialysis ' tubing and dialyzed against FTPS buffer. 5 10 15 20 25 Time (hours) •41 Figure 3.3.2. Bupivacaine Release Kinetics from DOPCrChol Liposomes with and without a pH Gradient - An initial drug-to-lipid concentration of 266 umole bupivacaine mmole"1 DOPC was added into dialysis tubing and dialyzed against FTPS buffer. Liposomes with a pH; of 4.0 (#) and pH ; of 7.4 (•) are shown. 300 0 5 10 15 20 Time (hours) 42 3.4 Discussion The similarity of bupivacaine release kinetics from both POPC and DOPC LUVs would suggest, to a large extent, that the rate of bupivacaine release is not dictated by the lipid composition. The release kinetics are, however, quite different for DOPC LUVs with or without a pH gradient. This would suggest that the release of bupivacaine is fettered to the slow dissipation of the pH gradient. The release of bupivacaine from LUVs exhibiting a pH gradient has two characteristic components; a rapid initial release followed by a slower sustained payout. The rapid release of bupivacaine from the dialysis bag is due to two main factors. Firstly, there is a small fraction of the free drug outside of the liposome that can easily diffuse out of the dialysis bag. This is useful as a rapid initial payout is desirable to achieve a loading dose for rapid onset of analgesia. Secondly, the higher internal pH of the liposome when the drug is maximally loaded causes a much larger concentration of bupivacaine to be present in the unprotonated form. Since bupivacaine is vastly more membrane permeable in the unprotonated species than in the protonated form, initially, there is much more bupivacaine present inside of the liposome that can readily diffuse outward. Therefore,! bupivacaine is rapidly released at the start and progressively slows as more of the unprotonated species is removed. The ratio of bupivacaine concentrations inside to outside of the liposome must always be in equilibrium with the ratio of proton concentrations inside to outside of the liposome. A derivation of the Henderson-Hasselbalch equation can be used to explain this phenomenon were: [Bupivacaine• H ~ } w e _ [HiO*]»«•«/« [Bupivacaine • H~]ou<sije [HsO ]<mu,Jc 43 As the bupivacaine outside of the liposome diffuses away, bupivacaine inside of the liposome must cross the lipid bilayer to replace it. As more neutral bupivacaine leaves the liposome, more of the protonated species must be deprotonated to the neutral form By this action, a proton is released inside of the liposome causing the internal pH to drop (i.e. the interior of the liposome becomes more acidic ) The rate of payout begins to slow because the ratio of Bup/BupH+" inside of the liposome decreases as the pH inside of the liposome decreases. The slower release is due to the progressive loss of the permeable unprotonated form inside of the liposome as more bupivacaine is released. By comparing the release kinetics of bupivacaine from liposomes with and without a pH gradient, the liposomes lacking a pH gradient release bupivacaine much more rapidly than liposomes exhibiting a pH gradient. In the control no pH gradient vesicles, the drug is not loaded and therefore the drug is readily released. Altering the internal pH of the liposome will change bupivacaine release rates. If faster release kinetics are desirable, then a higher internal pH can be used. Likewise, if slower release kinetics are desirable then a lower internal pH can be used. The extent of the pH gradient is, however, known to affect the uptake efficiency into the liposome. The smaller the pH gradient the less efficient the uptake. If the pH gradient is too large, then there will be too much bupivacaine protonated and too little of the membrane permeable neutral form to be released and blocking neural conduction becomes more difficult. In . addition, highly acidic buffers inside of liposomes are known to catalyze the hydrolysis of ester bonds linking the glycerol backbone with the hydrophobic acyl chain. Free fatty ! acids are known to destabilize the bilayer, increase the membrane permeability and 44 decrease the shelf life. It is desirable, therefore, to choose a pH gradient that allows both efficient uptake, to load bupivacaine inside of the liposome, and reasonable release kinetics, to provide a slow sustained release of bupivacaine over a prolonged period. 45 4. Duration of Nerve Block in Vivo 4.1 Introduction The prolongation of neural blockade through the use of liposomal bupivacaine was the central goal in my project. A model was needed where the duration of blockade for several formulations could be tested in a single animal. This reduces inter-animal variation and makes comparisons between different formulations simpler to analyze. The model for testing the duration of neural blockade was the pin-prick test in the guinea pig back.54 This model can compare up to five formulations including a saline control per guinea pig. Injections are made intradermally and the duration of analgesia is easily measured by scoring the response of the wheal over time with a standardized probe stimulus. One of the disadvantages of using this particular model is the uncertainty as to which class of nerve fibres are actually blocked. Different types of nerve fibers are blocked at different concentrations of local anaesthetic.. Typically, unmyelinated narrow • • • • . 1 diameter nerve fibers such as type C fibres responsible for pain and temperature are • blocked long before myelinated large diameter type A alpha nerve fibres responsible for motor function. In the back of the guinea pig, we are uncertain which of the fibres (C (pain/temperature), A delta (pain/temperature) or A beta (touch pressure)) respond to the stimuli. The muscle twitch response can be elicited by merely touching the backs of the guinea pigs which would suggest A beta fibre stimulation. As the A beta fibres are large diameter myelinated fibres, they have one of the highest C m (minimum blocking 46 concentration) values. Since the pain fibres are blocked at much lower concentrations, this may indicate that higher bupivacaine concentrations surrounding the nerve are reached than may be necessary to relieve pain. 4.2 Methods 4.2.1 Clearance Rate of LUVs Following Intradermal Injection into Guinea Pigs Clearance of liposomes from the site of injection was followed using 14C-cholesteryl hexadecyl ether, a non-metabolizable, non-exchangeable lipid marker. At various time points, intradermal injections (100 u.1) of DOPCChol LUVs (100 mM phospholipid) were made into the backs of the Guinea Pigs. A zero time point injection was also made immediately prior to sacrificing the animal. The skin was then removed and the injection sites dissected. To monitor the possibility of leakage of liposomes into surrounding wheals, muscle tissue below the wheal and the dermal layers surrounding the wheal were also dissected. Tissue samples were weighed and digested in 4 ml of Solvable® overnight at 65°C Samples were then cooled to 4°C and two 250 pi aliquots of H 2 0 2 (35%) were added to decolorize, with the second peroxide aliquot added only after foaming had subsided. The sample was then divided into four 1 ml aliquots and 15 ml of UltimaGold liquid scintillation cocktail added to each vial. Samples were vortexed and counted the following day on a Beckman LS 3801 liquid scintillation counter after overnight storage in the dark. 47 4.2.2 Guinea Pig Cutaneous Wheal Model for Measuring the Duration of Neural Blockade The duration of nerve blockade produced by free and liposomal bupivacaine was compared using the cutaneous wheal model54 in Dunkin Hartley Guinea Pigs from the University of British Columbia Animal Care Facility. To reduce stress, the animals' backs were shaved the day before the experiment was conducted. Wheals were formed by injecting 100 ul of the sample into the intradermal layer of the Guinea Pig's back. As a control, a blank saline or liposome sample was also injected in each animal. Wheals were marked and pricked five times at intervals using a standardized probe. A response was scored if the back muscles twitched in response to the probe stimulus. Inter-animal variation was reduced by testing all four samples on the same animal in each experiment. Each animal was used three times with a minimum 48 hour interval between experiments. Wheals were never generated in the same location twice. A random number generator was used to assign the formulations to various regions on the back and to determine the order of testing each wheal. In addition, the person testing the response to probe stimuli was blind to the contents of the sample injected in each wheal. 48 4.3 Results 4.3.1 Intradermal Clearance of LUVs The clearance of DOPC: Choi LUVs from the intradermal layers in the backs of guinea pigs was studied to determine the duration over which the liposomes would remain at the site of injection. This offers valuable data in explaining the results from the differences in duration of nerve blockade in the Guinea Pig experiments- As shown in Figure 4.3.1, over 85% of the liposomes remain at the site of injection over 48 hours. 4.3.2 Duration of Nerve Blockade for POPC:Choi Liposomes The duration of nerve blockade was compared using 600 nm POPC:Choi liposomes exhibiting a pH gradient loaded with 0.75% and 2.0% bupivacaineHCl versus the free drug. Both the 0.75% and the 2.0% liposomal bupivacaine solutions showed dramatic improvements over the free drug. The 0.75% free bupivacaine showed a half maximal response (R2.5) after 1,5 hours while the 0.75% liposomal bupivacaine sample showed a R2.5 after 4.5 hours. It should be pointed out, however, that the 0.75% POPC:Choi liposomal bupivacaine sample never achieved a complete block. The lowest number of twitches was 1 and this occurred after 1.5 hours. In the 2.0% samples, the free drug had an R2.5 after 3 hours while the liposomal POPC:Chol sample had an R2.5 of 7.5 hours. In addition, the 2.0% liposomal bupivacaine samples had a lower number of ; twitches over the first 5 hours than the 0.75% solution. The blank saline control also averaged 5 out of 5 twitches throughout the experiment. 49 4.3.3 Duration of Nerve Blockade for DOPC:ChoI Liposomes The duration of neural blockade was compared in the guinea pig wheal model at two different bupivacaine concentrations, 0.75% and 2.0%, between three different formulations; free bupivacaine, bupivacaine in DOPCChol (55:45) liposomes with and without a pH gradient. In Figure 4.3.3.1, the three different bupivacaine formulations were compared at a concentration of 0.75%. Free bupivacaine showed a half maximal response (R2.5) after 2 hours while the liposomes lacking a pH gradient showed a Ris.after 3.5 hours. In the liposome formulation with a pH gradient, the R2.5 occurred after 6.5 hours. It should also be pointed out that this particular formulation had a delayed onset of neural blockade. In Figure 4.3.3.2, the three different bupivacaine formulations were compared at a concentration of 2.0%. All of the curves at the higher concentrations were shifted to the right, denoting an increase in the duration of neuronal blockade Both the free bupivacaine sample and the liposome formulation lacking a pH gradient showed similar R2.5 of 4.5 hours. In the liposomal formulation with a pH gradient, the R2.5 was increased to 12 hours. In addition, the delayed onset of analgesia was eliminated as j • compared to the 0.75% liposomal bupivacaine exhibiting a pH gradient. The blank saline control injection also scored 5 out of a possible 5 twitches throughout this-experiment. In Figure 4.3.3:3, the effects of internal buffering capacity on the duration of nerve blockade were examined. The concentration of pH 4.0 citrate buffer used to buffer the aqueous core of the liposomes was changed from 300 mM to 100 mM. The duration of nerve blockade for the 100 mM citrate buffered liposomes was significantly reduced to 5 50 hrs from 12 hrs when the liposomes are internally buffered with 300 mM citrate. The duration of blockade for free bupivacaine was 4 hrs and this is consistant with previous experiments. '51 Figure 4.3.1 - Clearance of Liposomes - Liposomal bupivacaine (2.0%) was injected into the backs of Dunkin Hartley guinea pigs. The clearance of liposomes (#) was monitored with 3H-cholesteroyl hexadecyl ether. The error bars indicate standard error andN=3. 100 c 'c 'co E CD a: _co o •a co CD CD CL 20 30 Time (hours) 52 I 1 Figure 4.3.2 - Duration of Nerve Blockade with POPCrChol Liposomes - Injections of 100 ul were placed into the intradermal layers in the backs of Dunkin Hartley guinea pigs. Three samples were administered: Free bupivacaine in 150 mM NaCl (#), 0.75%(A) and 2.0% (•) bupivacaine in L U V POPCChol. The error bars indicate standard error and N=9. hours 53 Figure 4.3.3.1 - Duration of Nerve Blockade with DOPC:Chol Liposomes at 0.75% Bupivacaine - Injections of 100 ul were placed into the intradermal layers in the backs of Dunkin Hartley guinea pigs. Three samples were administered: 0.75% bupivacaine in 150. mM NaCl (A) and 0.75% bupivacaine in LUV DOPCChol (100 mM PL.) with pHi=7,4 (O) and pH;=4.0 (•). The error bars indicate standard error and N=12. Time (hours) 54 Figure 4.3.3.2 - Duration of Nerve Blockade with DOPC:Chol Liposomes at 2.0% Bupivacaine - Injections of 100 ul were placed into the intradermal layers in the backs of Dunkin Hartley guinea pigs. Three samples were administered: 2.0% bupivacaine in 150 mM NaCl (A) and 2.0% bupivacaine in LUV DOPCChol (100 mM P.L) with pH;=7.4 (O) and pHj=4.0 (•). The error bars indicate standard error and N=12. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Time (hours) 55: Figure 4.3.3.3 - Effect of Buffering Capacity on Duration of Nerve Blockade for DOPC:Chol Liposomes - Injections of 100 ul were placed into the intradermal layers in the backs of Dunkin Hartley guinea pigs. Three samples were administered: 2.0% bupivacaine in 150 rhM NaCl (#), 2.0% bupivacaine in LUV DOPCChol liposomes with pH 4.0 buffered 100 mM citrate (•) and a 150 mM NaCl control (not shown). The last line is 2.0% liposomal bupivacaine (•) with pH 4.0 buffered 300 mM citrate and was : extracted from Figure 4.3.3.2 for comparison. 0 2 4 6 8 10 12 14 Time (hours) 56 J 4.4 Discussion Liposomes were shown largely to remain at the site of injection with over 85% present after 48 hours. This suggests that it is not the removal of the liposomes from the site of injection that causes the loss of neuronal blockade. Most of the liposomes that 1 were cleared did so during the initial eight hours. This could be due to some heterogeneity in vesicle size with the smaller diameter liposomes being cleared at a faster rate than larger systems. Smaller liposomes have been previously shown to clear much more readily than larger liposomes following subcutaneous injection.8 The POPCChol liposomes showed a dramatic increase in the duration of neural: blockade for both 0.75% and 2.0% bupivacaine concentrations compared to the free drug. This phenomenon is most likely due to slower bupivacaine payout from the liposome depot. In the free bupivacaine sample, high concentrations of the drug will readily diffuse away from the site of injection. In the liposomal sample, the liposomes remain localized at the site of injection over a period of days slowly releasing the local anaesthetic. Another point of interest is the slightly increased duration of blockade in the 2.0% versus the 0.75% POPCChol liposomes. An explanation is rather complex as it pertains both to the drug-to-lipid ratio and release kinetics. Recall from Section 3 that it is only the neutral form of bupivacaine which crosses the bilayer in any significant quantities and as more bupivacaine enters.the liposome the pH inside of the liposome will rise. As the pH rises inside of the liposome, more of the membrane permeable neutral form will be present. In the 2.0% sample the final pH after all the bupivacaine is loaded will be higher than in the 0.75% sample. Therefore, 2.0% liposomal bupivacaine will release more bupivacaine 57 because it contains more of the drug as the neutral membrane permeable species. In addition, the duration of nerve blockade should be longer in the 2.0% sample as there is. more free drug available. Both of these factors resulted in an increased duration of blockade for the 2.0% liposomal bupivacaine over the 0.75% liposomal bupivacaine. For DOPC: Choi LUVs, duration of nerve block studies were completed on the free drug and on liposomes with and without a pH gradient. In comparing the three 0.75% bupivacaine samples, each showed substantially different R 2 5 values. Free . bupivacaine had the shortest duration of neural blockade of 2 hours (R2.5) because most of the drug could rapidly diffuse away from the nerve The R2.5 value increased to 3.5 hours for the no pH gradient liposomes most likely due to some hydrophobic association of bupivacine with the vesicles which resulted in more drug remaining at the site of injection over a longer period. The pH gradient sample had the longest duration of neuronal blockade of 7 hours (R2.5). Based on in vitro dialysis experiments, the extended duration likely reflects sustained release of bupivacaine at the site of injection. The pH gradient sample showed a three fold increase in the duration of nerve blockade compared to free drug. The onset of neural blockade was the slowest for the pH gradient formulation since it had the lowest initial concentration of free drug. In comparison of the same three formulations at 2.0% local anaesthetic, both free drug and bupivacaine with liposomes lacking a pH gradient showed similar durations of neural blockade of 4.5 hours (R2.5) The pH gradient formulation with a R 2 5 of 11.5 hours showed a nearly three fold increase in the duration of neural blockade over these two samples. There was also a decrease in the time required for analgesia onset in the pH 58 gradient sample, as compared to the 0.75% pH gradient sample. The decrease in onset time is due to more free drug being present outside of the liposome at higher drug-to-lipid ratios and more membrane permeable drug present inside of the 2.0% sample due to the higher internal pH after drug loading The effect of a lower buffering capacity on the duration of nerve blockade is shown in Figure 4.3.3.3. For LUVs buffered with 100 mM citrate the R2.s was 4 5 hours, while in the liposomes buffered with 300 mM citrate the R 2 5 was 11.5 hours. Buffering: the liposomes with 300 mM citrate showed a nearly threefold increase in the duration of nerve blockade over liposomes buffered with 100 mM citrate. In addition, some CNS side effects were observed during the experiments using 100 mM citrate, suggesting that higher peak systemic concentrations of bupivacaine were reached. These observations support: the pH gradient hypothesis of drug loading. While both the 100 mM and 300 mM citrate buffers were prepared at pH 4 0, the 100 mM buffer had a higher internal pH after drug loading than the 300 mM buffer due the lower buffering capacity. The higher internal pH causes more of the membrane permeable protonated form to be present inside of the liposome, which can readily diffuse outward once the sample is injected. The liposomes made with 300 mM citrate buffer has a lower final pH after drug loading. Therefore, there is less of the membrane permeable form present and bupivacaine is released at a slower rate. In addition to the rapid release of bupivacaine, the drug may also not be efficiently loaded at 100 mM citrate. Bupivacaine uptake studies were not completed with liposomes buffered with 100 mM citrate at pH 4.0; however, based on previous studies we would anticipate that less drug loading would have occured. Both the higher concentration of, 59 membrane permeable bupivacaine and the reduced drug loading would contribute to shorter R 2 5 values in liposomes containing 100 mM citrate versus 300 mM citrate. 60 5. Toxicity 5.1 Introduction Bupivacaine is one of the most toxic of the clinically used local anaesthetics. Accidental intravenous injections has been reported to cause severe CNS toxicity, arrhythmia or even death from cardiac failure. ; Along with the promise of longer duration of action, liposomal local anaesthetics may also provide an extended margin of safety. The design of the liposome delivery system is to alter the normal biodistribution of the free drug by slowly releasing the local anaesthetic over prolonged duration. If bupivacaine is released at a sufficiently slow rate, the drug is transformed to safer metabolites before the parent compound can accumulate to toxic levels. A comparison of the toxicities between bupivacaine encapsulated in pH gradient liposomes, bupivacaine with control liposomes (no pH gradient) and free drug was performed. In this study, mice were given intraperitoneal injections and both the convulsive and lethal doses were determined for each of the three samples. 5.2 Methods In this study, female CD 1 mice purchased from Charles River in Quebec were given intraperitoneal injections to determine the dose were 50% of the mice convulsed (CD5o)^ Liposomally encapsulated bupivacaine (pH gradient) was compared with control no pH gradient vesicles and free drug at two different bupivacaine concentrations, 0.75% 61 and 2.0%. Subsequent to convulsions, the mice frequently died and, therefore, the lethal dose (LD 5 0) was also determined. Injections were completed between the hours of 10 am to 1 pm to avoid circadian rhythm effects 5 5 The mice were observed for a minimum of 30 minutes post injection to score seizure activity or death. If the mice were not fully recovered by this time, they remained under observation until they appeared normal! Mice were also housed for one week post injection to observe any long term effects. The experiment was double blind and randomized to minimize experimental bias. Each mouse was given a number on its tail and randomly selected for injection using a random number generator. Both the order of injection and the type of sample given were also randomized. In addition, the person scoring the mice for seizure activity was blind to the sample given to the mice Both the CD 5 0 and the L D 5 0 was determined through a regression analysis of the probit versus log dose plot. 5.3 Results . For 0.75% bupivacaine samples in both Figures 5.1 and 5.2, the pH gradient sample was significantly less toxic than either the free or control no pH gradient sample. The CD 5 0 (dose where 50% of the mice convulsed) for free bupivacaine was 49.3 mg/kg. This is similar to the value reported by DeJong and Bonin of 57.7 mg/kg56 The CD 5 0 for the no pH gradient sample was slightly higher than the free drug at 62.0 mg/kg. The CD50 for the pH gradient liposomes was much greater than 200 mg/kg as only 2 of 40 mice in this dose range experienced seizures. 62 In the LD 5 0 data accumulated from the 0.75% samples in both Figures 5.5 and 5.6, the pH gradient sample was again clearly superior in terms of its margin of safety over both the free drug and control liposomes (no pH gradient). The LD 5 0 for the free drug was 68.1 mg/kg which is similar to DeJong,and Bonin's value of 58.7 mg/kg56 and Bruguerolle et aVs value of 52 mg/kg57. The control no pH gradient sample had an L D 5 0 value of 81.1 mg/kg. Only 1 of the 40 mice injected died over the dose range studied up to 200 mg/kg. Therefore, the LD 5 0 value for the pH gradient sample was greater than 200 mg/kg which represents a greater than fourfold increase in the margin of safety. Both the CD50 and LD50 values are taken from the regression of the probit analysis. For the 2.0% bupivacaine samples the pH gradient liposomes again showed a dramatic improvement in the margin of safety over both the free drug and the no pH gradient control. The CD 5 0for the control no pH gradient liposomes was 42.6 mg/kg while the CD 5 0 value for pH gradient liposomes was 64.9 mg/kg, as seen from Figures 5.3 and 5.4. This represents a 30% increase in the safety of using pH gradient liposomes over the free drug, which had a CD50 of 49.3 mg/kg. In the LD50 studies at the 2.0% concentration shown in Figures 5.7 and 5.8, the control no pH gradient liposomes had an LD 5 0 value of 69.1 mg/kg which is very similar to the free drug at 68.1 mg/kg. In the pH gradient liposomes the LD50 value was 102.5 mg/kg and the margin of safety was improved by 51% over the free drug. During the one week post injection observation period, all of the surviving mice remained healthy. 63 Figure 5.1 - Percent Seizure Activity versus Dose of 0.75% Bupivacaine. (•) Free Bupivacaine, (•) 0.75% No pH Gradient LUVs, (•) 0.75% pH Gradient LUVs Figure 5.2 - Probit of Seizure Activity versus Log Dose of 0.75% Bupivacaine. (•) Free Bupivacaine, (••) 0.75% No pH Gradient LUVs, (A) 0.75% pH Gradient LUVs Figure 5.3 - Percent Seizure Activity versus Dose of 2.0% Bupivacaine. (•) Free Bupivacaine, (•) 2.0% No pH Gradient LUVs, (A) 2.0% pH Gradient LUVs Figure 5.4 - Probit of Seizure Activity versus Log Dose of 2.0% Bupivacaine. (•) Free Bupivacaine, (•) 2.0% N o pH Gradient LUVs, (A) 2.0% pH Gradient LUVs U 1 _ J 1 I | J | I I 1 30 40 50 60 70 80 100 125 150 175200 Dosage (mg/kg) 67 Figure 5.5 - Percent Death versus Dose of 0.75% Bupivacaine. (•) Free Bupivacaine, (•) 0.75% No pH Gradient LUVs, (A) 0.75% pH Gradient LUVs 68 Figure 5.6 - Probit of Death versus Log Dose of 0.75% Bupivacaine. (•) Free Bupivacaine, (•) 0,75% No pH Gradient LUVs, (A) 0.75% pH Gradient LUVs Figure 5.7 - Percent Death versus Dose of 2 . 0 % Bupivacaine. (•) Free Bupivacaine, (•) 2.0% No pH Gradient LUVs, (A) 2.0% pH Gradient LUVs Figure 5.8 - Probit of Death versus Log Dose of 2.0% Bupivacaine. (•) Free Bupivacaine, (•) 2.0% No pH Gradient LUVs, (A) 2.0% pH Gradient LUVs • • 30 40 50 60 70 80 100 125 150 175200 Dosage (mg/kg) 71 5.4 Discussion The toxicity associated with the use of bupivacaine for neural blockade is drastically reduced by loading bupivacaine into liposomes. A greater than fourfold increase in safety was observed for the 0.75% sample and a 51% increase in safety for the 2.0% sample. Clearly, an increase in safety of this magnitude is a remarkable advance in controlling clinical pain. The 0.75% pH gradient liposomes were less toxic than the 2.0% pH gradient liposomes. As shown in Figure 2.2.4, 36% of bupivacaine in the 2.0% pH gradient sample is free while only 18% of bupivacaine in the 0.75% pH gradient sample is free. Obviously, the larger quantity of free drug in the 2.0% sample would readily diffuse from the injection site to achieve toxic levels more rapidly. In addition, the pH inside of the liposomes in the 2.0% sample is much higher than the 0.75% sample and, therefore, more bupivacaine will be present in the non-protonated permeable species to readily diffuse out of the liposome. 72 6. Discussion The use of liposomes exhibiting a transmembrane pH gradient offers considerable advantages for long term neural bloackade. Such systems have been shown to rapidly accumulate bupivacaine by simply mixing drug and liposomes together. The extent of drug uptake can be contrasted to liposomes lacking a pH gradient which show very poor loading. Uptake of bupivacaine into pH gradient vesicles is also very efficient with over 80% of the drug loaded in the 0.75% sample. Another major benefit of the pH gradient loading system is that the rate of drug release from the vesicle can be controlled. Delivering local anaesthetics is no different than delivering other drugs as a loading dose is required to saturate local distribution of the drug. In the pH gradient system, a rapid loading dose is provided with 20% to 30% of the total drug concentration being free. Additionally, bupivacaine present inside of the liposome is initially released rapidly as there is more of the membrane permeable species present due to the high intravesicular pH following drug loading. Drug release rates could ,4 be controlled by either altering the pH gradient or changing the lipid composition of the bilayer. The major objective behind this research on liposomal bupivacaine was the development of a long term local anaesthetic which could be used in the management of chronic, severe pain. It has been clearly shown in the guinea pig model that liposomal bupivacaine is a significantly longer duration local anaesthetic than the free drug alone. Recall that the pH gradient liposomes had a nearly threefold increase in duration over both the free and control (no pH gradient) liposomes, 11.5 hrs versus 4 firs respectively. This 73 longer blockade may allow for novel procedures to be used by anaesthetists in the clinical management of pain. Not only has the duration of analgesia increased with the use of pH gradient LUVs, but the margin of safety is also increased compared to the free drug. At both bupivacaine concentrations tested, toxicity was dramatically reduced for the liposomal formulation. An increase in the safety margin of 51% was observed for the 2.0% sample compared to the free drug (LD 5 0 of 103 mg/kg for 2.0% liposomal bupivacaine versus 68 mg/kg for the free drug). Using 0.75% bupivacaine, no LD50 could be found up to a dose of 200 mg/kg. This represents at least a fourfold increase in the safety for 0.75% liposomal bupivacaine compared to the free drug. The liposomal system which has been developed seems to have met all of the requirements that were initially sought. Nonetheless, improvements in the system can be envisaged and there are several experiments and modifications that should be completed. The first and most important experiment would be to determine the release kinetics in vivo. For this study, radiolabeled bupivacaine would be required. The experiment would be similar to that described herein to determine liposome clearance except that radiolabeled bupivacaine would be included in addition to the radiolabeled lipid. In this study, we could correlate the rate of bupivacaine release in vivo with the duration of analgesia and adjust the bupivacaine release rate accordingly. Using vesicles of smaller size (200nm) instead of 600 nm may also be beneficial. LUVs of smaller size are less osmotically sensitive than larger vesicles and would therefore be more durable. Although the viscosity is not a problem with the 600 nm 74 sample, the viscosity in the 200 nm liposomes should be reduced. Vesicles of 200 nm would also be more convenient for pharmaceutical preparation as the liposomal bupivacaine could be simply filtered through 0.22um pore filters for sterilization. Clearance studies must also be completed to determine if vesicle size is correlated to its rate of clearance from the injection site. Since the trapped volume will be changed by using smaller diameter vesicles, trapped volume measurements should be completed to determine if drug uptake is altered. Studies using other lipid mixtures should also be completed to determine if the duration of action can be extended further. If it is found that there is still large quantities of bupivacaine trapped inside of the liposome as analgesia is lost, then a bilayer could be fabricated that increases in permeability subsequent to injection. The gradual increase in permeability would dissipate the pH gradient and serve to release the final amounts of bupivacaine trapped inside. Remarkable developments have been made in this area in our group. John Holland and Tom Madden have succeeded in developing a liposomal membrane that fuses with erythrocytes. The theory behind this bilayer is based on the progressive loss of stabilizing lipids subsequent to their combination with acceptor vesicles like erythrocytes. The pH gradient is the driving force for uptake into our delivery system and therefore the buffering capacity of the liposome is very important. Citrate buffer is the buffer used in our LUVs and it has three pKa's (3.1, 4.8 and 6.4). If another buffer could be used that contained several more titratible groups (more pKa's) then we would increase the buffering capacity without altering the internal osmolarity and have osmotic stability of 75 our vesicles. A larger buffering capacity would also allow us to increase the initial pH prior to drug loading to 4.5 instead of 4.0 to increase the bupivacaine release rate in the latter stages of nerve blockade. This would hopefully extend the duration of analgesia. The final suggestion, which may serve to both increase the duration of analgesia and reduce the toxicity of bupivacaine, is to add epinephrine. Bupivacaine ampules are presently available with epinephrine to reduce toxicity. However, epinephrine negligibly increases the duration of analgesia for bupivacaine. It is added to the shorter duration anaesthetics such as lidocaine to act as a vasoconstrictor and reduce the clearance of drug from the injection site, thus prolonging neural blockade. Epinephrine is not as usefulto prolong neural blockade in longer duration local anaesthetics as the anaesthetic acts longer than the vasoconstrictory effect. If, however, epinephrine were loaded into the liposome with the local anaesthetic then both drugs would be released at a slower rate and hence prolong the duration of analgesia. 76 7. Bibliography I Fink BR: Neural Blockade in Clinical Anesthesia and Management of Pain (2nd ed). Cousins MJ, Bridenbaugh PO ed. J.B. Lippincott Company, Philadelphia, 1988, p.3. 2Koller C: On the use of cocaine for producing anaesthesia on the eye Lancet 2:990, 1884 3 Simcock MJ: Bupivacaine for regional analgesia in labour. A double-blind comparison with lignocaine. Med J Australia. 1:889-91, 1970 4 Strichartz GR: Neural Blockade in Clinical Anesthesia and Management of Pain (2nd ed). Cousins MJ, Bridenbaugh PO ed. J.B. Lippincott Company, Philadelphia, 1988, p,43 5 Boogaerts JG, Lafont ND, Luo H, Legros FG; Plasma concentrations of bupivacaine after brachial plexus administration of liposome-associated and plain solutions to rabbits. Can J Anaesth 40:1201-04,1993 6 Boogaerts JG, Lafont ND, Declerq AG, Luo HC, Gravet ET, Bianchi JA, Legros FJ: Epidural administration of liposome associated bupivacaine for the management of postsurgical pain: a first study. J Clin Anesth 6:315-20, 1994 7Ostro MJ, Cullis PR: Use of liposomes as injectable-drug delivery systems. Amer J Hosp Pharm. 46:1576-87, 1989 8 Allen TM, Hansen CB, Guo LS: Subcutaneous administration of liposomes: a comparison with the intravenous and intraperitoneal routes of injection. Biochim Biophys Acta 1150:9-16, 1993 i. 9Umbrain V, D'Haese J, Alafandy M, De Roover E., Legros FJ, Camy F: Fate of epidural liposomes in rats.(Abstract) Anesthesiology 81:A979, 1994 1 0 Nichols JW, Deamer DW: Catecholamine uptake and concentration by liposomes maintaining pH gradients. Biochim Biophys Acta 455:269-271, 1976 I I Madden TD, Harrigan R, Tai LCL, Bally MB: The accumulation of drugs within large unilamellar vesicles exhibiting a proton gradient: a survey. Chem. Phys. Lipids 53: 37-46, 1990 12Hondeghem LM, Miller RD: Basic and Clinical Pharmacology, 5th ed. Katzung BG ed., Appleton & Lange, Norwalk, Connecticut, 1992, p. 369 1 3 Ragsdale DS, McPhee JC, Scheuer T, Catterall WA: Molecular determinants of state-dependent block of Na+ channels by local anesthetics. Science 265:1724-8, 1994 1 4 Courtney KR: Mechanism of frequency dependant inhibition of sodium currents in frog myelinated nerve by the lidocaine derivateive GEA. J Pharmacol Exp Ther 195: 225,1975' 1 5 Hondeghem LM, Miller RD: Basic and Clinical Pharmacology, 5th ed. Katzung BG ed., Appleton & Lange, Norwalk, Connecticut, 1992, p. 366 1 6 De Jong RH: Neural blockade by local anesthetics. JAMA 238:1383-5, 1977 1 7Covino BG: Neural Blockade in Clinical Anesthesia and Management of Pain (2nd ed). Cousins MJ, Bridenbaugh PO ed. J.B. Lippincott Company, Philadelphia, 1988, p.121 77 1 S Tanaka K, Yamasaki M: Blocking of cortical inhibitory synapses by intravenous lidocaine. Nature: 209: 207, 1966 1 9McCaugherW: Adverse effects of local anaesthetics. Drug Safety 7: 178-89, 1992 Zornow MH, Scheller MS: Cardiac arrest during spinal anaesthesia II. Anesthesiology 68: 970-1, 1988 2 1 Brown DL, Carpenter RL, Moore DC, Bridenbaugh LD, Rupp SM, Ramsey DH, Thompson GE, Mulroy MF: Cardiac arrest during spinal anaesthesia III. Anesthesiology 68: 971-2, 1988 2 2 Albright GA: Cardiac arrest following regional anesthesia with etidiocaine or bupivacaine. Anesthesiology 51: 285, 1979 2 3 Moller RA, Covino BG: Cardiac electrophysiologic effects of lidocaine and bupivacaine. .Anesth Analg 67: 107-14, 1988 2 4 Zavisca FG, Badgwell JM: A rodent model for studying four well defined toxic endpoints during bupivacaine infusion. Reg Anesth 16: 223-7, 1991 2 5 Clarkson CW, Hondegham L: Mechanism for bupivacaine depression of cardiac conduction: fast block of sodium channels during the action potential with slow recovery from block during diastole. Anesthesiology 62: 396-405, 1985 2 6Loder RE: A local anaesthetic solution with longer action. Lancet 2: 346-7, 1960. 2 7 Strauss RG, Dase D, Doering PL: Prolonging paracervical block anesthesia: addition of dextran to 2-chloroprocaine. Am J Obstet Gynecol 133: 891-3, 1979 2 8 Buckley FP, Fink BR: Duration of nerve blocks produced by mixtures of local anesthetics and low molecular weight dextran. studies in rat infraorbital nerve blocks. AnesthAnalg 60: 142-5, 1981 29Scurlock JE, Curtis BM: Dextran local anesthetic interactions. Anesth Analg 59: 335-6, 1980 , 0 Hassan HG, Renck H, Lindberg B, Akerman B, Hellquist R: Effects of adjuvants to local anaesthetics on their duration. Acta. Anaesthesiol Scand 29: 375-9, 1985 Langerman L, Grant GJ, Zakowski M, Golomb E, Ramanathan S, Turndorf H: Prolongation of epidural anesthesia using a lipid drug carrier with procaine, lidocaine and tetracaine. Anesth Analg 75: 900-5, 1992 , 2 Langerman L, Grant GJ, Zakowski M, Ramanathan S, turndorf H: Prolongation of spinal anesthesia. Anesthesiology 77: 475-481, 1992 " Haynes DH, Kirkpatrick AF: Ultra-long duration local anaesthesia produced by injection of lecithin coated methoxyflurane microdroplets. Anesthesiology ,63: 490-499, 1985 ",4Boedeker BH, Lojeski EW, Kline MD, Haynes, DH: Ultra-long duration local anesthesia produced by injection of lecithin coated tetracaine microcrystals. J Clin Parmacol 34: 699-702, 1994 J 3 Hersh EV, Maniar M, Green M, Cooper SA: Anesthetic activity of the lipospheres bupivacaine delivery system in the rat. Anesth Prog 39: 197-200, 1992 ~'6 Beck LR, Cowsar DR, Cosgrove RJ, Riddle CT, Lowry SL, Epperly T: A new long acting injectable microcapsule system for the administration of progesterone. Fertil Steril 31: 545, 1979 78 3 7 Wakiyama N, Juni K, Nakano M: Preparation and evaluation in vitro of polylactic acid microspheres containing local anaesthetics Chem Pharm Bull 29: 3363-8, 1981 j 8 Wakiyama N, Juni K, Nakano M: Preparation and evaluation in vitro and in vivo of polylactic acid microspheres containing dibucaine. Chem Pharm Bull 30: 3719-27, 1982 j 9LeCorre P, Estebe JP, Chevanne F, Malledant Y, Le Verge R: Spinal controlled delivery of bupivacaine from dl-lactic acid oligomer microspheres. J Pharm Sci 84: 75-8, 1995 4 0Kojima T, Nakano M, Juni K, Inoue S, Yoshida Y: Preparation and evaluation in vitro of polycarbonate microspheres containing local anaesthetics. Chem. Pharm Bull. 32:2795-2802,1984 4 1 Masters DB, Berde CB, Dutta SK, Griggs CT, Delphine H, Kupsky W, Langer R: Prolongued regional nerve blockade by controlled release of local anesthetic from a biodegradable polymer matrix. Anesthesiology 79: 340-6, 1993 4 2 Cousins MJ, Bridenbaugh PO: Neural Blockade in Clinical Anesthesia and Management of Pain (2nd ed). J.B. Lippincott Company, Philadelphia, 1988, p.36. 4" Djordjevich L,Tankovich A, Chigurupati R, Woronowicz A, McCarthy R: Efficacy of liposome encapsulated bupivacaine. (Abstract) Anesthesiology 65:A185, 1986 4 4 Legros F, Ruysschaert JM: Pharmaceutical composition containing a local anesthetic and/or centrally acting analgesic encapsulated in liposomes. U. S. Patent 5,244,678 Sept. 14, 1993 4 5 Boogaerts JG, Declercq A Lafont N, Benameur H, Akodad EM, Dupdnt JC, Legros FJ? Toxicity of bupivacaine encapsulated into liposomes and injected intravenously: comparison with plain solutions. Anesth Analg 76:553-5, 1993 4 6 Boogaerts JG, Lafont ND, Declerq AG, Luo HC, Gravet ET, Bianchi J A Legros FJ: Epidural administration of liposome associated bupivacaine for the management of postsurgical pain: a first study. J Clin Anesth 6:315-20, 1994 4 7 Mashimo T, Ichiro U, Pak M, Shibata A, Nishimura S, Inagaki Y, Yoshiya I.: Prolongation of canine epidural anesthesia by liposome encapsulation of bupivacaine. Anesth Analg 74:827-34, 1992 4 8 Bernards CM, Luger TJ, Malmberg AB, Hill HF, Yaksh TL: Liposome encapsulation prolongs alfentanil spinal analgesia and alters systemic redistribution in the rat. Anesthesiology 77:529-35, 1992 4 9 Mayer LD, Hope MJ, Cullis PR: Vesicles of variable sizes produced by a rapid extrusion procedure. Biochim Biophys Acta 858:161-8,1986 5 0 Tucker GT, Laurence ME: Neural Blockade in Clinical Anesthesia and Management of Pain (2nd ed). Cousins MJ, Bridenbaugh PO ed. J.B. Lippincott Company, Philadelphia, 1988, p. 50-51 5 1 Fiske CM, Subbarow Y: The colorimetric determination of.phosphorous. J. Biol. Chem. 66:375-400, 1925 ' . 5 2 Pick U: Liposomes with a large trapping capacity prepared by freezing and thawing of sonicated phospholipid mixtures. Arch Biochem Biophys 212:186-194, 1981 79 "Madden TD, Tilcock CP, Wong K, Cullis, PR: Spontaneous vesiculation of large unilamellar vesicles composed of saturated phosphatidylcholine and phosphatidylglycerol mixtures. Biochem. 27:8724-30, 1988 4Bulbring E, Wajda, I: Biological comparison of local anaesthetics. J. Pharmacol. 78:84, 1945. 5 Moore PA: Orcadian variations in the convulsant response and pharmacokinetics of lidocaine. Thesis, University of Pittsburgh, 1977 Taken from De Jong and Bonin (1980), Anesth Analg 59:6 p:401-5 6DeJong RH, Bonin JD: Deaths from local anesthic induced convulsions in mice. Anesth Analg 59: 401-5, 1980 , 7Bruguerolle B, Attolini L, Gantenbein M: Acute toxicity of bupivacaine metabolites in mice. Clin Exp Pharm Phys 21:997-9, 1994 80 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098986/manifest

Comment

Related Items