Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Synthesis of a model lipid and observation of behaviour of model liposomes by electrophoresis Song, Xu Chun 1994

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

Item Metadata


831-ubc_1994-0300.pdf [ 1.41MB ]
JSON: 831-1.0059587.json
JSON-LD: 831-1.0059587-ld.json
RDF/XML (Pretty): 831-1.0059587-rdf.xml
RDF/JSON: 831-1.0059587-rdf.json
Turtle: 831-1.0059587-turtle.txt
N-Triples: 831-1.0059587-rdf-ntriples.txt
Original Record: 831-1.0059587-source.json
Full Text

Full Text

SYNTHESIS OF A MODEL LIPID AND OBSERVATION OF BEHAVIOUR OF MODEL LIPOSOMES BY ELECTROPHORES IS By XU CHUN SONG  B.Sc.., University of Science and Technology of China Hefei, Anhui, P.R.China, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMISTRY  We accept this thesis as conforming  THE UNIVERSITY OF BRITISH COLUMBIA April © Xu Chun  1994 Song, 1994  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.  (Signature)  Department of  CftkLSTP1  The University of British Columbia Vancouver, Canada Date  DE-6 (2188)  Apr.  j 4  ABSTRACT  A model lipid consisting of a cholesterol base, tetraethoxy— spacer and glucuronic acid head group was synthesized.  First, the head group was  prepared by acetylation and esterification of glucuronolactone to produce methyl (1, 2,3,4-tetra-O— acetyl——D—glucopyran)uronate (j.) which was then brominated to produce methyl (2,3,4—tri-O—acetyl—--D—glucopy ranosyl bromide)uronate  (Z). The combination of the cholesterol part and tetra—  ethoxy chain was made by reacting cholesteryl—p—toluenesulfonate and tetraethylene glycol, to produce 3—O—(i1-hydroxy—3, 6, 9—trio xaundecyl)cholest —5-ene (tetra-EC)  (3).  reported previously.  The above steps were carried out with methods The coupling of the head group and tetra-EC employed  a different method, which had been used by others in the coupli ng reaction of the same head group and cholesterol, by using silver oxide as the catalyst instead of silver carbonate.  Methyl t3—O--(3,6,9—trioxaundecyl)  cholest—5—en—3-yl—2,3, 4—tri—O—acetyl——D—glucopyranosid3uro nate  () was  produced from the coupling reaction with a yield estimated to be  —  50’!.,  higher than that of the reaction with silver carbonate as the catalys t. The final step was to remove the methyl group and the acetyl protect ing groups on the head group by using excess MaCH in a specific solven t system and acidifying with HC1, to obtain crude 3—O—(3,6,9—trioxun decyl) cholest—5—en—3—yl——D—glucopyranosiduronic acid  ().  The crude acid  product was primarily purified by adjusting the pH of the suspension of the acid in warn ethanol and water and the salt form was obtained. product,  The salt  (i), was precipitated pure from chloroform solution by addition of  mixed ethyl acetate arid hexane.  ii  The product (tetra—ECG), which has a negative charge on the head group, was used with other lipids to prepare liposomes.  The liposomes,  which vary in poly(ethylene glycol) (PEG) chain density and charge location were made for the purpose of mimicking the glycocalyx region in actual biomembranes.  Particle electrophoresis was used to measure the mobilities  of the liposomes in the solutions with variation of pH (1.8—9.9), ionic strength (O.OO1M—O.1H) and PEG chain density (O-60’A by molar ratio).  The  classical theory for particle electrophoresis was applied to calculate the mobility and the apparent charge density of the liposoznes.  The pKa of  tetra—ECG was determined with a plot of the mobility of the tetra—ECGcontaining liposome against pH on the surface of the particle.  A numerical  model, which has been developed as a computational program, was used to interpret the results of electrophoresis as a function of ionic strength, in terms of several parameters which describe the surface properties of the liposomes.  iii  TABLE OF CONTENTS Page ABSTRACT  jj  TABLE OF CONTENTS  iv  LIST OF FIGURES  vi  LIST OF TABLES  vii  SYMBOLS AND ABBREVIATIONS  viii  ACKNOWLEDGEMENTS Chapter 1  ix  Synthesis of Lipids  1  1.0 Objectives  1  1.1 Introduction  1  1.1.1 Lipids in Biomembranes  2  1.1.2 Chemical Structure of Lipids in Biomembranes  2  1.1.3 The Model Lipid  4  1.2 Methods  5  1.2.1 Synthesis of Tetraethoxycholesterol(tetra—EC)  7  1.2.2 Synthesis of Protected and Brominated Glucuronate  9  1.2.3 Coupling Reaction  10  1.2.4 Deprotection  13  1.3 Experimental  17  Preparation of Methyl (1,2,3, 4—tetra—O—acetyl——D— glucopyran)uronate Preparation of Methyl (2,3,4-tri—O—acetyl——D— glucopyranosyl bromide)uronate Preparation of 3—O—(11—hydroxy—3, 6, 9—trioxaundecyl) cholest—5—ene Preparation of Methyl [3—O—(3,6,9—trioxaundecyl)cholest— 5—en—3—yl—2, 3, 4-tri—O-acetyl—-D—glucopyranosid]uronate Preparation of (3—O—(3,6,9—trioxaundecyl)cholest—5—en—3— -yl--D-glucopyranosid]uronic acid and Sodium [3—O-(3,6,9-trioxaundecyl)cholest— 5-en-3-yl--D-glucopyranosid]uronate  iv  18 18 19 ..  20  21  1.4 Discussion  .  Chapter 2 Electrophoresis of Liposomes  24  26  2.1 Introduction  26  2.2 Theories of Particle Electrophoresis  28  2.2.1 The Theory for Charged Particle with A Smooth Surface  28  2.2.2 pH at Surface of the Liposomes  32  2.2.3 The Theory for NHairy  33  Model Liposomes  2.3 Methods  35  2.3.1 Preparation of Liposomes  35  2.3.2 Electrophoresis Equipment  37  2.4 Experimental  39  2.4.1 Liposome Preparation  39  2.4.2 Electrophoresis  41 pH Dependence Studies  41 Ionic Strength Dependence Studies  41 Different Compositions of Liposomes  43  2.5 Result and Discussion  43  2.5.1 pH Dependence  45  2.5.2 Composition Dependence  47  2.5.3 Ionic Strength Dependence  49  2.6 Conclusions  52  REFERENCES  53  LIST OF APPENDICES  56  V  LIST OF FIGURES  Page Figure 1.1  The Modified Fluid Mosaic Model of Biomembrane  3  Figure 1.2  The Model Lipid  4  Figure 1.3  Procedure of Lipid Synthesis  6  Figure 1.4  Two Anomers of Protected Sugar  9  Figure 1.5  Koenig—Knorr Coupling Reaction  11  Figure 1.6  Scheme of Schneider’s Coupling Reaction  ii  Figure 1.7  Procedure of Deprotection of tetra—ECPG  14  Figure 1.8  TLC Result for Side Product in Deprotection  16  Figure 2.1  Structure of Liposome  27  Figure 2.2  Concepts of Electric Double Layer and Electrokinetic Potential (C)  30  Figure 2.3  The Liposome Model of Glycocalyx  34  Figure 2.4  Preparation of Liposomes  36  Figure 2.5  Equipment for Particle Electrophoresis  38  Figure 2.6  Deducted Mobility of Liposome EL* 2 vs pH at Surface of Liposome  48  Mobility of Liposomes EL* 7—1, 7—2, 7—3, and EL* 8-1, 8-2, 8-3 vs PEG Chain Concentration  48  Mobility of Liposomes EL* 3 vs Ionic Strength  50  Figure 2.7  Figure 2.8  vi  —  6  LIST OF TABLES  Table 1.1  Page  Brief Review of Syntheses of PEG—Cholesterol Derivatives  8  Table 1.2  Yields of Products  Table 2.1  Composition of Liposomes for Electrophoresis  Table 2.2  Mobility and Apparent Surface Charge Density of EL* 1 and EL* 2 with Variation of pH  44  Mobility and Apparent Surface Charge Density of EL* 7 EL* 8 with Variation of Composition  44  Table 2.3  25 ....  —  Table 2.4  Mobility and Apparent Surface Charge Density of EL* 3 EL* 6 with Variation of Ionic Strength —  Table 2.5  Table 2.6  ..  40  45  Deducted Mobility, Net Surface Charge Density and pH at Particle Surface of EL* 2 with Variation of pH  46  Comparison of Mobility Model Parameters for Four Liposome Preparation  51  vii  ABBREVIATIONS  CI  chemical ionization  Chol.  cholesteryl or cholesterol  d  doublet  DCI  desorption chemical ionization  DPPG  dipalmitoyl phosphatidyl glycerol  egg PC  egg phosphatidyicholine  El  electron ionization  HPLC  high pressure liquid chromatography  LSIMS  liquid secondary ion mass spectrometry  m  multiple  NNR  nuclear magnetics resonance  PEG  poly(ethylene glycol)  s  singlet  tetra—EC  tetraethoxycholesterol  tetra—ECG  sodium [3—O—(3,69—trioxauridecyl)cholest— 5-en-3-yl---D-glucopyranosid)uronate  tetra—ECPG  Methyl [3—O—(3,6,9—trioxaundecyl)cholest— 5—en-3—yl—2, 3, 4—tri—O—acetyl—-D— glucopyranosidluronate  TLC  thin layer chromatography  tr  triplet  tri—EC  triethoxycholesterol  qt  quartet  viii  (In N)ffi results)  (In NMR results)  (In NMR results)  (in NNR results)  (in N)4R results)  AcKNOWLEDGE24ENTS  I would like to express my special thaiilcs to my supervisor, Don Brooks, for his patience, assistance and encouragement throughout this work. I am especially grateful to Johan Janzen, who has helped me so much during almost all y project.  Thanks to everyone in this group for  their valuable suggestions and assistance.  ix  Chapter 1  1.0 Objectives The objectives of this project are: 1)  to modify the synthesis of a model lipid which has a cholesteryl  base, poly(ethylene glycol) (PEG) chain and glucuronic acid head group; 2)  to prepare liposomes, which have different (negative) charge  location and different densities of PEG chains, as models of the glycocalyx region in actual biomembranes; 3)  to use particle electrophoresis to measure the mobilities of the  charged liposomes and apply both the classical theory and a numerical model of liposomes to interpret the results.  1.1 Introduction Biological membranes are directly involved in many biological processes of living organisms.  They are composed primarily of lipids and  proteins, with some associated oligosaccharides and small amounts of mono— and divalent ions binding to ionic groups of lipids and proteins, and water. For more than 50 years, biomembranes have been extensively studied in order to understand the relationship between their structure and the function (1, 2).  In 1925 Gorter and Grendell (3) first described the structure of a  biomembrane as a bilayer lipid matrix.  Since then,  various models have  been proposed to describe membrane organization, resulting in our present understanding of membranes described by the Singer and Nicholson fluid mosaic model (4), which has been proved to be useful for presentation of 1  the gross organization and structure of lipids and proteins in biomembranes.  Figure 1.1 is a diagram which is based on the model given by  Singer et al. and modified to include a description of carbohydrates on the surface of the membranes (2).  1.1.1 Lipids in Biomembranes Biological membranes contain an astonishing variety of lipids, both in amount and in kind.  Phospholipids are usually found in biological systems,  as well as sphingolipids, glycolipids and steroids.  Not only do lipids  function as a matrix for association of membrane proteins and provide a permeability barrier between the exterior and interior of a cell, they also participate in a variety of specialized biological processes (5).  1.1.2 Chemical Structure of Lipids in Biomembranes Generally the structure of lipids in biomembranes consists of two parts: a polar or hydrophilic region, and a nonpolar or hydrophobic region. The chemical nature of these two sections can vary substantially.  In water  the polar regions tend to orient toward the aqueous phase while the nonpolar regions are withdrawn from water.  For phospholipids bearing two  alkyl chains the lowest free energy is achieved through formation of a two dimensional bilayer which is the basis of biological membranes. In some cases, models of lipids may be described to have a third part, the spacer region, between the terminal of the hydrophilic head group and the hydrophobic tail in a bilayer or micelle (6).  In recent work on the  synthesis of artificial lipids, poly(ethylene glycol)  (PEG) has been applied  as a spacer (7).  2  Figure 1.1  A modified fluid mosaic model of biomembrane(2)  It offers a three—dimensional and cross—sectional description of protein, lipid and carbohydrate in the model of a biomembrane. Lipid and protein form a structural bilayer matrix and carbohydrate moieties extend from the surface of membranes into the aqueous solution.  3  Another characteristics of most lipids is that they have charged head groups, such as phosphate, sulfate, and carboxylate (anionic), as well as ammonium groups (cationic) (8).  These provide biomembranes which contain  such lipids with net surface charges.  1.1.3 The Model Lipid Our synthesis of lipids is aimed at making liposomes to mimic biomembrane structures to allow studies of the properties of a model biomembrane surface.  The detailed purpose is described in Chapter 2.  The model lipid consists of three sections:  Figure 1.2  The Model Lipid  R  =  H  (3—O—(3, 6, 9—trioxaundecyl)cholest—5-en—3—yl— -D-glucopyranosid]uronic acid  R  =  Na  sodium [3—O—(3,6,9—trioxaundecyl)cholest—5—en-3—yl--D—glucopyranosid)uronate  4  The cholesteryl group is placed as a hydrophobic anchor in this lipid, because it is present in a wide variety of biological membranes and participates actively in many biological processes (9). The hydrophilic section is glucuronic acid, which provides the saccharide character.  It has been widely observed that carbohydrates on  the surfaces of biomembranes are common groups in living organisms.  In  general they can act as recognition reactants, structural materials and energy stores and are involved in a multitude of interactions with other organisms and biological reagents (10, 11).  They are also considered to  stabilize the membranes against disruption in some biological systems (12). The spacer section of the model lipid is PEG, because it is soluble in water and most organic solvents and apparently has a high compatibility with biological systems (13). was chosen.  A compound with four ethylene glycol units  This particular lipid had been synthesized by Paula J. Sather,  a previous student in our laboratory, in 1990 (7).  However, she obtained a  low yield and was unable to isolate enough product to be used in model membrane studies.  1.2 Methods The synthesis of the model lipid employed modifications of the procedure described in P.J.Sather’s thesis (7). major steps:  5  It consisted of three  0  U  HO\  OH  o  HOrjY/—OH  tretraathylena qlycol H OH qlucuroo1acton.  A0Ac ‘oAc  I  cholesteryl-p-toluerie sulfonate  H  methyl t.tra-O-&cety].— -D qlueopyranuronata (1)  Me000  1  ACH ‘OAc Compound  ta>  I  +  Br  Compound (1)  MeOOC  OAo H Compound CL)  0—1----Compound  Figure 1.3  () & (>  Procedure of Lipid Synthesis 6  The first step is connection of tetra—ethylene glycol to the cholesteryl group by displacing the tosylate from cholesteryl—p—toluene sulfonate.  The second step is coupling between compound  and the protected and brominated glucuronic acid  -  () (tetra—EC)  compound  (a),  and the  final step is removal of the protecting acetyl groups from the glucuronic moiety.  1.2.1 Synthesis of Tetraethoxycholesterol (tetra—EC) The starting materials, tetraethylene glycol and cholesteryl—p—toluene sulfonate, are commercially available.  Cholesteryl—p—toluene sulfonate has  very high reactivity with the compounds which contain hydroxyl groups, especially with water.  For this reason, the reaction between tetraethylene  glycol and cholesteryl—p—toluene sulfonate is carried out under anhydrous conditions and in the absence of oxygen. Brockerhoff and Ramsammy (14).  This method was reported by  Patel et al.  (15) used it to synthesize  triethoxycholesterol (tri—EC) and the yield was >90X.  Sather (7) applied  the same procedures to make a series of oligo—ethoxy—cholesterols and reported the yields given in Table 1.1. From Sather’s work, a key to a high coupling yield was to use anhydrous 2,4-dioxane (which had been dried with sodium metal and then distilled prior to the reaction) as the solvent. Liquid chromatography of mixed organic solvent systems on silica gel columns is the usual way of purification of these compounds. this procedure was 58.9’/.  7  The yield of  Table 1.1  Brief Review of Recent Syntheses of PEG—Cholesterol Derivatives  Hf[O  REF4CE  Reaction Pathway  Molar Ratio of Reagents: cTS*/pEG  Reaction Time (hr)  Yield  (•/.)  (14)  3  a  (15)  3  b  1/25.3  (16)  1  b  3  b  (20)  3  b  (13)  3  b  1 / 25.4  24  —100  4  b  1/4.5  24  56.5  6  b  1/5  60  26  —-  ——  ——  (18)  a  Ethoxy Unit (n)  3  --  -—  ——  2  92  1 / 25—30  2—4  81  1/25-30  2—4  92  2—3  ——  -—  b  reported by Fong et al in Lipids,  Vol.12, iQ 857—62(1977) excess PEG and cholesteryl—p—toluenesulfonate were stirred b in ref lux of dioxane(dried by ref lux with Na) for several hours under N 2 [Patel et al, 1984, Ref. (15)] * Cl’S: cholesteryl—p—toluene sulfonate  8  1.2.2 Synthesis of Protected and Brominated Glucuronate (2) Two steps are taken to obtain the compound (2) (See Figure 1.3). were reported initially by Bollenback et al.  They  (17) in 1955 and utilized  widely since then as the traditional method to make this compound (7). The first step includes esterification of glucuronolactone (to produce methyl glucuronate) arid acetylation of the four hydroxyl groups with acetic anhydride.  The solvents are anhydrous methanol and pyridine, the latter  acting as a catalyst of acetylation. obtained by crystallization at  —  In this part, the crude product is  4C from the solution of the reaction  mixture and the pure crystals are obtained by recrystallization in absolute ethanol.  Sometimes the color of the reaction mixture is so dark that the  solution of crude product needs to be decolorized by carbon (ref lux under reduced pressure) to remove the colored impurities before recrystallization. Compound  (1)  has two anomers:  has been reported to be the  and (See Figure 1.4).  anomer based on the NNR evidence (7).  work the yield was 33.5%.  type  type  Figure 1.4  The product  Two Anomers of Compound (1)  9  In our  The second step of the synthesis of compound anomeric carbon of compound().  (Z) is bromination at the  Hydrobromic acid in glacial acetic acid  (3O% by weight) reacts with the acetyl group, eliminating one molecule of AcOH and forming the C-Br bond.  The crude product is crystallized with  absolute ethanol. The product was  anomer of compound  () (because of the anomeric  effect of the suger ring) and the yield was 857. in Bollenback et al.’s work and 737. in our synthesis.  The  NMR spectrum of the product in Figure 1.5  indicates, from the coupling constant between the protons connected directly to the sugar ring, that the bromine is axial in product (18) that is the  anomer of compound  (a).  —  (See Appendix II)  The procedure of bromination had also been used by Sather (7) with very satisfactory results.  An important point is that the brominated  compound (2) should be used in the next coupling reaction as soon as possible because the Br is sensitive to both oxygen and water, especially in light and heat, which might produce the impurities that could complicate subsequent reactions.  1.2.3 Coupling Reaction Typically, catalysts are used in the coupling reaction between the brominated sugar and tetra—EC to activate the bromine first; then the sugar is attacked by the hydroxyl group on tetra—EC to form the ether linkage. Koenig Knorr’s method was the first one dealing with this type of coupling, with silver carbonate as the catalyst (See Figure 1.5).  Again, Sather  followed this method in her work, but the yields reported were very low, around 30% or less, after purification by column chromatography (7). 10  MeOOC  O\  AcO  Compound ()  H  +  Br  Compound (2)  CO 1 Ag 3 Benzene  AcO 3 Compound  Figure 1.5  (i.)  Koenig-Knorr Coupling Reaction  Mo00  + Compound  (2.)  Cholesterol  AgO  AcO AcO.  Figure 1.6 Scheme of Schneider et al.’s Coupling Reaction (20) 11  This method was considered and tried in our project.  The coupled  product did form, but the yields varied from a few percent to less than 30°A, similar to Sather’s results.  Even though silver carbonate was freshly made  and dried in the dark (which did affect its catalytic efficiency), there was no significant improvement in the yield. A mixed catalytic system of mercury bromide and mercury oxide, based on Goodrich et al.’s work (19), was tried in our system, but the results were also not satisfactory. Another approach was an older method reported in 1969 by Schneider and Bhacca (20), which had been used in Goodrich et al.’s previous work (21) Figure 1.6).  (See  In this method, silver oxide (freshly made) had been used  instead of silver carbonate.  The reaction was between exactly the same  sugar used in our system, and simple cholesterol.  The yield reported by  Schneider et al. was very high (and so was the yield of their deprotection reactions). This method was applied in our system and brought the yield of coupling reaction of the sugar and tetra-EC up to 50%.  (The yield was estimated  from 1 H—NMR spectra of several samples of impure products, because of the difficulty of purification). The crude product was a syrup which was easily dissolved in an organic solvent.  Silica gel provided an appropriate stationary phase to separate  the compounds in liquid chromatography. and proved inadequate)  (Aluminum oxide column was tried  The problem in the purification is that both  tetra-EC and the product (3), tetra—ECPG, have very strong affinity for silica gel.  Even though they can be separated very well on TLC silica  plates in small scale, with the mixed solvent system of ethyl acetate, chloroform and hexane, the separation on a column was quite different because the product, tetra—ECPG, always came off contaminated with a small 12  amount of tetra—EC (as detected by TLC).  Controlling the amount and manner  of loading material did not improve the separation.  The pure product had  to be collected from only a few of fractions in each run of chromatography, since most of the fractions remained as mixtures of the product and tetra—EC (by TLC results).  The practical reaction yield, >50%, was  estimated roughly from the sum of the pure product and the product which remained in the mixture with tetra—EC.  For determination of the molar  ratio of tetra—ECPG and tetra—EC in the mixtures, the method is described in Appendix II.  TLC analysis of the mixtures indicated that most of the  components were those two compounds. The yield from the purification was so low that several chromatography runs had to be made, starting from a few grams of the crude product, to provide sufficient material for the next step.  1.2.4 Deprotection In organic syntheses, it is an old problem to make specific fimctionalized compounds by retaining some of the functional groups while changing others in reactions.  Many reagents have been applied in  protection, deprotection or catalysis for both of these two reactions under different circumstances. In our project, deacetylation and deesterification were required. The latter is easy to be approached by hydrolysis (13, 19, 20). the deacetylation step took much more effort.  However,  Although there have been  many methods of deacetylation used, with sodium methoxide in methanol as the most commonly used reagent system (22), most of them were applied only for small molecule systems.  13  compound  (4)  (i) addition of NaOH—arthydrous MeOH, methylene chloride(5 hrs) (ii) addition of THF, aqueous NaOH, MeOH and water(1.5 hrs) (iii)pH -* 3, concentration of the mixture to 10—30 ml to give the precipitate of the acid product compound (crude)  ()  (i) suspension in warm ethanol—water, pH - 7 (ii) addition of ethanol to precipitate the salt product compound (crude)  ()  Figure 1.7  compound (pure)  recrystallization from chloroform solution ethyl acetate/hexane  ()  A diagrammatic description of the procedure of the deprotection of the compound(4)(tetra—ECPG)  A procedure, which had been reported by Schneider et al.  (20) and used  in Sather’s deacetylation step (7), was first applied in deacetylation of compound (4).  It is a complex system of NaOH in anhydrous methanol (sodium  methoxide), methylene chloride, THF and then aqueous NaOH solution. procedure is described in Figure 1.7.  The  Following adjustment of the pH to 3,  the crude acid product is separated from the mixture by centrifugation. The primary purification is transformation of the crude acid salt  () by adjusting pH to 7 in aqueous ethanol.  This step removes the  impurities which are not soluble in water and ethanol.  14  () to the  The procedures, including the quantities of reagents given, were exactly followed, but the reactions did not go to completion (See the TLC results in Figure 1.8).  A series of other compounds appeared, making purification  very difficult and greatly lowering the deprotection yield. reaction times produced no improvement in the results.  Longer  It seems that some  uncompletely deprotected compounds form producing poor yields.  These  compounds were identified because the amounts were very small. After several failures, the sodium hydroxide concentration in the first part of the reaction was increased. improved.  The yield of crude material was  After the primary purification, further purification again  became a significant problem because of the polyhydroxylated nature of the compounds  () and (i). (See Section 1.3 TLC results).  TLC showed that the deprotected product interacted strongly with silica gel for both acidic and salt forms.  In the usual (not highly polar)  solvent systems, the product (acid or salt) almost always remained at the origin of the plate.  The product dot moved from the origin when a polar  solvent, e.g., methanol, was added, but the solubilities of the products were limited in methanol.  Hence, chromatography on a silica gel column was  not considered for the purification. Instead, recrystallization of the salt product, compound  (i), from  chloroform (which is the best solvent for the products) was attempted, with solvents which were expected to lower the solubility of the product and crystallize them from solution.  A precipitate, which was the pure product,  resulted upon addition of a mixed solvent of ethyl acetate and hexane. gave an indication of its purity.  The integration information from proton  NMR and LSIMS results in different matrices were also useful.  15  TLC  Column  1  2  3  4  5  6  7  8  9  10  9  ....L.iApr.F6 .  A1  Figure 1.8  TLC Results for the Compounds in the Deprotection Step (solvent system: ethyl acetate/chioroform/hexane  3:3:1)  Columns 1 7 are the results of TLC monitoring of the first 5—hour reaction (See Figure 1.7), with sampling times indicated below. Column 8 is the result for the sample taken from the reaction system after addition of aqueous NaOH solution. Column 9 is for the mixture extracted from the finished reaction solution with chloroform, in which there are a series of compounds. Column 10 is the reference of the compound before deprotect ion. -‘  16  The yield from this procedure was 32.6%, but not all of the product was recovered.  A portion of the product remained in solution with the  impurities. The practical yield of the reaction was not estimated, therefore, because it was difficult to determine the type and concentration of the impurities. Appendix II.  The NMR analysis of pure deprotected product (6) is in The product  (i), called tetra—ECG, was used to prepare  liposome in the next part of this project.  1.3 Experimental General Details Proton NNR spectra were recorded on Bruker 141-1—400 MHz and Bruker AC—200E (200 MHz) spectrometers in CDC]. 3 or CD OD as the solvent. 3  Mass  spectra were recorded on a Kratos MS—40 (El), a Delsi I4ermag Rb—b  C Mass  Spectrometer (CI & DCI) with ammonia reagent grade gas and a Kratos Concept II HQ Mass Spectrometer(LSIMS) with matrices of glycerol and thioglycerol.  Solvents and reagents were used as purchased with the exception of those listed below: Chloroform (solvent for making tetra—EC), benzene, pyridine, dichioromethane and tetrahydrofuran (THF) were distilled prior to use. Dioxane was distilled from a ref lux with sodium for 1 hour prior to use. Methanol was distilled from a ref lux with magnesium and iodine for 0.5 hours, stored over  4A molecular sieves, and distilled again from calcium  hydride or sodium hydride, prior to use.  Cholesteryl—p—toluene sulfonate,  tetra—ethylene glycol, and glucuronolactone were dried under vacuum. 17  0 2 Ag  was made from hot NaOH and AgNO 3 solution in the dark and dried prior to use.  Ag CO  23  was made from AgNO  3  and Na CO  23  solutions in the dark.  SYNTHESES Preparation of Methyl (1,2,3, 4-tetra-O-acetyl-j3-D-glucopyran)uronate (1) This reaction was based on the procedure described by Bollenback et al.  (17), and also by Sather (7).  The reagents and amounts were:  Glucuronolactone  10 g, 56.78 mmol  Sodium hydroxide  0.08 g(in MeOH)  Anhydrous methanol  75 ml  Acetic anhydride  60 ml  Pyridine  25 ml in 5 ml acetic anhydride  The product was decolorized with carbon if necessary and recrystallized from absolute ethanol.  The yield was 7.15 g, 19.00 inmol.  NMR (400 MHz) CDC1  2.03(2s,9H), 2.11(s,3H), 3.73(s,3H), 4.16(d,1H),  5.12(tr,1H), 5.26(m,2H), 5.76(d,1H) Mass Spectrum  DCI: m/e 394(M+NH ), m/e 317(M—59), m/e 257(317—59—H) 4  m/e 228(257—O—CH), m/e 114(317—3x59-2CH) or (257—0-CH) 44  Preparation of Methyl (2,3,4-tri-O-acetyl-a-D-glucopyranosyl bromide) uronate (2) This reaction procedure was also from Bollenback et al.  (17). The  reagents and amounts were: Compound (1)  1.8114 g, 4.813 inmol  Hydrobromic acid (30% in acetic acid)  7.2 ml(in 3.6ml)  Chloroform  30 ml  Sodium bicarbonate (saturated)  30 ml  18  After reaction and removal of acetic acid, the crude product dissolved in chloroform was washed with NaHIDO remove the residual acid. ethanol.  3  saturated solution and then water to  The product was recrystallized from 100%  The yield was 1.389 g, 3.497 mmol.  NNR (400 MHz) CDC1 3  2.04(2s,6H), 2.09(s,3H), 3.75(s,3H), 4.57 (d,1H),  4.84(qt1H), 5.22(tr1H), 5.59(tr,1H), 6.62(d,1H)  Preparation of 3-O-(11-hydroxy--3, 6, 9-trioxaundecyl)cholest-5-ene (3) Tetraethylene glycol (11.0162 g, 56.7 mmol), cholesteryl—p--toluene sulfonate (0.9928 g, 1.836 mniol) and dioxane (20 ml) were added to a 50 ml round bottom flask.  A condenser was added and the mixture was stirred at  ref lux under a nitrogen atmosphere for 6 hours (after 3 hours the reaction was finished according to the results of TLC).  The dioxane was removed by  rotary evaporation and the residue was dissolved in 30 ml water and extracted with diethyl ether (5 x 80 ml) in a 250 ml separatory furmel. The organic extracts were combined, washed with 10°!. NaC0 solution (1 x 40 ml), 3 and then water (5 x 60 ml).  The organic phase was dried over anhydrous  sodium sulfate, filtered and the filtrate was evaporated under reduced pressure.  The residue was loaded onto a silica gel column (20 g), packed  with 1:1 (volume ratio) ethyl acetate/chloroform.  The column was eluted  with the same solvent system and the product was collected in 1 ml fractions (each fraction was analyzed by  mc).  The fractions containing  product were combined and evaporated under reduced pressure to give 0.609 g, 1.082 mmol of product.  19  TLC  R= 0.24  NMR CDC1  ,  silica 60, ethyl acetate/chloroform 1:1(by volume ratio)  (400 MHz)  0.64(s,3H), 0.84(qt,6H), 0.88(d,3H), 0.97(s,3H),  0.98”2.40 (m, -29H), 3.16(m,1H), 3.60-’3.75(m, “-16H), 5.32(d,1H) Mass Spectrum  nile 560(M—2) , nile 368(M—194), mle 3 4 353(368_CH ) ,  El  nile 255(368ll3side  chain),  DCI (NH ) 3  mle 247(368—C H ), 1 3 nile 195(M—368+H) nile 581 & 3 580(M+NH ) , mle  563(M)’,  nile 369(M-193), nile 353(369-CH—1), nile 195(M-368+H)’  Preparation of Methyl [3-O-(3, 6, 9-trioxaundecyl)cholest-5-en-313--yl-2, 3,4-tri-O-acetyl-13-D-glucopyranosid]uronate Compound  ()  (2) (0.4642 g, 0.8247 mmol), () (1.2415 g, 3.126 mmol),  freshly made silver oxide (0.3919 g, 1.691 mmol) and benzene (10 ml), were stirred together at room temperature in a 25 ml round bottom flask in the dark.  After 24 hours, adding a few grams of Celite, the mixture was  filtered and the filtrate evaporated under reduced solvent.  pressure to remove  The residue was separated on a silica gel column (100 g) packed  with ethyl acetate/chloroformlhexane 3:1:2 (by volume ratio).  The column  was eluted with the same solvent system, and 1 ml fractions were collected and monitored by TLC.  The crude product, which by TLC contained mostly  unreacted (3) and the product (4), was 0.6523 g.  From 1 H NMR, the molar  ratio of (3) and (4) was calculated to be 1.15—1.4. The practical yield of the product (4) was 0.397-0.433 mmol, 0.349-0.381 g. The crude product was purified with about 10 runs of the same chromatography as above, and the pure product was obtained.  20  111  NMR CDC1 3 (200 MHz)  0.65(s, 3H), 0.83(d, 3H), 0.85(d, 3H), 0.89(d,  311), 0.98(s, 3H), 1.0 2.4( 29H), 1.98,2.00 & 2.02(3s. 9H), 3.62( 16H), 3.74(s, 3H), 4.03(d, 1H), 4.65(d, 1H), 4.99(tr, 1H), 5.22(2tr, 2H), 5.32(d, 1H) TLC (pure product) Rf  =  0.45, silica 60,  ethyl acetatelchloroformlhexane 3:1:2 (by volume ratio) Mass Spectrum (pure product) DCI (Nil ) 3 nile ) 4 + 3 929(M+2NH NH ”, nile ) 4 + 3 913(M+NH NH , nile 4 897(M+NH ) ”, nile 877(M—1)’, nile 837(M—CHCO+1)”, nile 528(M+NH—368), nile 396(528—3C 0 4 H 2 )’, nile 368(Chol.—16)”, nile 2 352(396—C 0 4 H )”,mle 336(352—0)’, nile 334(352-NH)’, nile 317(334—0—1)”, nile C 3 308(352—C O—1)”, H nile 292(308—0)’, nile 276(292—20)”, nile 4 257(276—NH — 1)”, nile C 3 234(276—C O+1)”, H nile 218(234-0)”, nile 4 216(234-NH ) ”  Preparations of [3-0-(3,6, 9-trioxaundecyl)cholest-5-en-313-yl13-D-glucopyranosid]uronic acid (5) and Sodium [3-O-(3, 6, 9-trioxaundecyl)cholest-5-en-3(3-yl3-D-glucopyranosid Juronate (4)  (0  (0.460 g, 0.524 mmol) was dissolved in anhydrous ether (6 ml) and  dichloromethane (6 ml).  0.5 N sodium hydroxide in anhydrous methanol (0.48  ml, 0.24mmol) was added and the reaction was carried out at room temperature, monitored by TLC.  After 4 hours, it was found by TLC with  solvent system of ethyl acetatelchlorofornilhexane 3:1:2 that a series of compounds formed added.  .  0.7m1 sodium hydroxide—methanol (0.66 N, 0.462mmol) was  After 1 hour further, TLC showed only one product.  THF(40 ml) was  added with initiation of stirring, followed by addition of aqueous 1 M sodium hydroxide (12 ml, l2mmol) in one portion and methanol (35 ml). 21  Finally water (75 ml) was added slowly.  Stirring was continued for another  1.5 hours at room temperature.  The pH of the solution was adjusted to 3  with the addition of 10°!. HC1.  The solution was concentrated with  difficulty by rotary evaporation under reduced pressure to about 10—30 ml. The precipitate was centrifuged and the supernatant was concentrated again to provide more material.  The precipitate, the crude acid (5), was  suspended in warm aqueous ethanol and sufficient aqueous sodium hydroxide was added to adjust the pH to around 7 at which point all the acid had dissolved.  Ethanol was added and the salt precipitate formed and separated  with centrifuging. product  The yield of crude salt  The crude  () was dissolved in chloroform and then precipitated by ethyl  acetate and hexane (3:1, volume ratio). proton NMR and LSIMS to be pure product  TLC  () was 295.9 mg.  [Salt, 1 R  (6)]  =  The precipitate was proven by TLC,  (h), 130 ing, 0.171 mmol.  0 for:  ethyl acetate/chloroform/hexane 3:1:2, chloroform/ethyl acetate 3:1, ethyl acetate/chloroform/benzene 10: 10:3, chloroform, ethyl acetate/CHC1/ethanol 2:2:1, R[Salt,  (6)]  =  0.56 for methanol, 0.55 for ethanol, 0.17—0.6 for chloroform/methanol 7:1, 0.57—0.86 for chloroform/methanol 1:1, 0.39—0.48 for ethyl acetate/methanol 6:1;  Rf[Acid.  (i)]  =  0 for chloroform/methanol 7:1, 0.07 for ethyl acetate/methanol 6:1  H NMR COC1 1  0.65—Z.40(cholesteryl section, —46H by integration),  3.0—4.8(broad, —24H), 5.34(d, 1H) LSIMS (low resolution) matrix: thioglycerol 22  nile 4 761(M+1) ’ , ni/e 693(761— COONa), ni/e 585(M—l76sugar part+2C), ni/e 475(chol.+CH CII OCH CH O+2H), ni/e 443(chol.i-CH CII OCH )‘, ni/e 22 22 22  2  413(chol.+CHCH)’, ni/e 369(chol.—16) 4 LSIMS (high resolution) matrix: thioglycerol Mass range: 761—761 Mass: 761.48241  No. peaks:  1  Base  mt.:  168501  Carbon 41 Hydrogen 70 Oxygen 11 Sodium 1  (The theoretical formula for the product is H Na) 4 C 1 O 6 1  23  1.4 Discussion In this part of the work, compounds  (4), () and () were produced.  Table 1.2 is the list of all the products synthesized. some problems which remained unsolved. of these types of lipids.  However, there are  The first problem was purification  Problems occurred in almost all the procedures,  from synthesis of tetra—EC through production of the deprotected lipid. Since the purifications of some of the products are on a hundred— milligram scale, HPLC seems appropriate but optimizing the solvent systems will be very important for improvement of the purification. Better methods of purification on a milligram scale need to be developed for the deprotected product, tetra—ECG, and an effort should be made to increase the yield of the reaction.  Although it has been improved  in this work, several methods suggested themselves, based on more recent publications: (i) OH—resin as a reagent for deacetylation (22). In this reference,  IRA—400(OH) resin was applied in sugar and nucleoside  systems to remove acetyl groups with yields of 70°!. up to 91%. (ii) Enzymes (23). It was reported that seven enzymes could catalyze the partial deacetylation of sucrose derivatives in phosphate buffer or phosphate buffer—organic co—solvent with variable yields.  Basically, these two methods were again effective for small molecules and the conditions of the reactions were selected only for the particular circumstances in their syntheses.  However, these could be good starting  points to improve the deprotection of tetra—ECPG.  24  Table 1.2 Yield of Products (See Figure 1.3, the procedure of the synthesis) Product  Catalyst/Solvent  Purification  Yield(%)  (1)  0 2 pyridine/(AcO)  crystallization (from reaction solution)  33.5  ()  ——/acetic acid  recrystallization (from 100% EtOH)  72.7  (3)  dioxane, N 2  (4)  0/benzene 2 Ag  tetra—EC  tetra—ECPG (6) tetra—ECG  * **  ——/methanol(etc.)  chromatography (ethyl acetate/ chloroform 1:1, silica gel)  58.9  48.1_52.5* chromatography (ethyl acetate/chloroform /hexane 3: 1:2, silica gel) precipitation (from chloroform solution by ethyl acetate/hexane)  32.6**  the yield was estimated from 1 H NMR results the yield was for the pure product obtained in the experiment  25  Chapter 2  2.1 Introduction Liposomes are artificial lipid vesicles which are composed of one or more bilayers and have an aqueous interior or interlameller phase (See Figure 2.1).  They can be classified into small unilamellar vesicles  (SUV), large unilamellar vesicles (LUV) and multilamellar vesicles (MV) by their size and the number of concentric bilayers present (24). Liposomes are structurally similar to biomembranes and have been utilized as models for studying many aspects of biological membrane properties, e.g., membrane lipid chemistry, lipid—protein interactions, transport phenomena and ligand binding to membranes.  They also have been  developed as an approach to controlled drug—delivery systems (25). The liposomes in this work were made from the model lipid synthesized in Chapter 1, tetra-ECG, which has a cholesteryl base, a PEG spacer and an acidic monosaccharide head group.  While cholesterol does not form bilayers  spontaneously, addition of PEG chains allows these molecules to form bilayers and liposomes (15).  The liposomes not only have a bilayer matrix  of cholesterol derivatives, but also have carbohydrates as well as PEG chains linked on the surface.  They therefore can be employed as a model to  mimic the behavior of actual membranes using pure molecules with well defined properties.  The liposomes are then examined by a technique which  is thought to be sensitive to the properties of the head group region, particle electrophoresis.  The result is interpreted in terms of a theory  for the electrophoretic mobility which contains parameters incorporating the chain concentration, thickness and charge location within the surface region (26). 26  . .  •  VVATER •  I  I  .  •  -  a I •  •  .  •  •  :  •  •  0  •  .  •  •  •  I I  •  •  I •.  • f(_o’•  ••-. •  • :  •  •.cz  I. I.  I  .•  . •  •ê 1  • • • •  •.•I•I• I  I  I  •  •  I  •  •  I  I  I  •  •  •  •  I  I•  •  •  •  . .  •  • •  • •  •  I  •  •  2.1  Liposome structure (24) 27  •  if  -‘ 0  • I  Figure  • I  Liposome I  I  •  •  I  0  •  I  I  p  Using the method of particle electrophoresis, the electrophoretic mobility of charged particles in an electric field can be measured.  The  mobility, which is the particle velocity per unit electric field strength, is related to the physico—chemical properties of the particle surface by the theory described below.  Since biological surfaces are normally  negatively charged, particle electrophoresis has been used in research on many biological organisms and mammalian cells, especially red blood cells (27).  There has been a lot of work which focused on the relationship  between the surface properties and behavior of single red cells, using particle electrophoresis.  Usually the mobilities are interpreted using  theories which apply to smooth charged surfaces.  Cell membranes, however,  contain many glycolipids and glycoproteins whose head groups extend into solution some distance and carry charged residues which are distributed throughout the depth of the head group region, or glycocalyx.  Hence, it is  not appropriate to interpret the electrophoretic mobilities of cells with theories which assume smooth charged surfaces. In this project, particle electrophoresis was applied to liposomes of known compositions with head groups whose properties modeled those found on biological cells.  A theory expressly designed for surfaces bearing such  structures was used to interpret the results.  2.2  Theories of Particle Electrophoresis  2.2.1 The Theory for Charged Particles with A Smooth Surface The classical theory for dealing with the behaviour of smooth charged particles, assumed to be locally flat, in electrophoresis is as follows: 28  An electrical double layer forms near the surface of charged particles because the counterions in solution are attracted to the surface. (See Figure 2.2)  The electrostatic potential at the surface of charged is related to the surface charge density, a’, by (29):  particles, (o),  =  (I)  (5)2}  _1{  [11  c’12  which reduces to (27):  where a’ ic  ZF4’o) RT  if  ice  <<  1  [21  surface charge density(esu/cm ) 2 Debye-Hückel parameter: r  ‘8irNave  K =  [  2  ,1/2  Ii  (cm  j  ckT 3 1O  -i  ),  is called the electric double layer thickness(cm) e  dielectric constant of water (78.5)  k  erg/e) 16 Boltzmann’s constant (1.3805 x 1O’  ,  Nay  Avogadro s number (6.025 x 10  T  temperature (°K)  e  electron charge(4.8 x 10 esu)  I  ionic strength  23  )  (298°K in our experiments)  I  =  4- ions CZ  (M)  1 C  molar concentration of i—th ionic species  1 Z  valence of i—th ionic species  F  Faraday constant  F  =  Nay  e  The electrostatic potential decays with distance from the surface of particles; if  ZFo RT W(x)  1, (o) is given by (27): .  = ‘I’(O)  e  -Kx  29  [3]  ab I  I  SOLUTION -l--’V-——-’ ± 4 )——  --  T”  —  H’ ‘I  I  I  a  b 1/K  Figure 2.2  I  distance from solid  Concepts of Electric Double Layer and Electrokinetic Potentail() The electric potential, ‘, decays with the distance. a—-the Stern layer of “fixed” charges; b-—shearing boundary of the solution when solid moves; 1/K——electric double layer; C——electrokinetic potential or zeta potntial (28) 30  The motion of the charged particle is determined by the direct electrical force, the fluid drag (from the viscosity of the solution) and the electroosmotic retardation (due to motion of the electric double layer in the electric field in the direction opposite to velocity of the particles).  The famous Helmholtz—Smoluchowski formula describes this  behaviour, for the case in which the radius of curvature of the particle surface is large compared with 1/ic, the electric double layer thickness:  E where U  [4]  4in  mobility of particles dielectric constant  r  medium viscosity (0.009 poise)  E  electric field  ?  (zeta potential) is the electrostatic potential at shear plane(or the nonslip surface)  The electric field, E, can be calculated from either the voltage or current.  Although the total voltage in the experiments was controlled to be  constant (40V)  ,  because of differences in cross—sectional area, it is  best to calculate E from the conductivity of the solution in which the liposomes are suspended. From [4], the electric field was calculated from the measured current: E— where V  ,  1  _iR 1  —  —  1  aAc  i and R are voltage, current and resistance respectively.  [5] 1 is  the electric length related to the distance between two electrodes (27). a is the cross—sectional area of the electrophoresis chamber in the viewing region (the radius of the viewing section of the chamber is 1.362 mm, given by the manufactor) and A is the equivalent conductivity of the solution (30).  c is the concentration of the electrolyte solution. 31  Generally, for smooth charged particles, an assumption is made that zeta potential is equal to the potential at particle surface,  P(O).  From  [2] and [3], an expression for the surface charge density is obtained (for low surface potentials):  [6] 2.2.2 pH at Surface of the Liposomes In aqueous systems, pH is one of the most important parameters that affect the electrostatic behaviour of the liposomes because H or OH  ions  participate in the processes of ion binding to surface molecules on the liposomes. According to the model for smooth charged particles, the concentration of any ion at the surface of liposome, Cs, is different with that of the ion in bulk solution, Cb.  The relationship between these two concentrations  is (27): Cs=Cbexp{—  ZF’I’ (0) RT  ]  Considering the H distribution, applying [7] and solving for pH  =  -log H gives: pHs  =  pHb  +  Fo) RT  =  pHb  +  e 2.3O3kT  [8]  where pHb is the value of pH in the bulk solution (measured).  2.2.3 The Theory for “Hairy” Model Liposomes The above relations have been widely used to describe the electrophoresis of biological cells, even though the surfaces of cells are actually more complicated than that of a smooth particle (31).  The region  of the glycocalyx in the real cells, as mentioned in Section 2.1, contains the polyelectrolyte or polymer chains which are penetrated by ions or small  32  electrolytes, such as carbohydrates.  It causes a distribution of fixed  charges throughout the glycocalyx (rather than a uniform charge distribution over the smooth surface particles) and hydrodynamic resistance produced by polymer segments. There has been some work published which focuses on the description of the behaviour of real cells in electrophoresis (32).  Levine et al.  (31)  offered a mathematical treatment of the cell surface, which considered the charge distribution (within the surface layer) and the hydrodynamic flow. This treatment was improved by Sharp et al.  (32) and a model to represent  the behaviour of liposomes bearing charged glycolipids in electrophoresis was developed by McDaniel et al (26). The approach used to interpret the electrophoretic mobilities of liposomes containing tetra—EC or tetra—ECG is an adaptation of that described by Sharp and Brooks (32).  Briefly, for particles large enough  that their radius is large compared to the double layer thickness, the electrophoretic mobility is calculated from the electroosmotic velocity resulting from motion of the electrical double layer adjacent to the charged particle surface in the electric field.  The resulting fluid  velocity a long way from the particle surface is set equal and opposite to the particle mobility (31).  For smooth particles the velocity distribution  near the surface is determined only by the ion concentration profile in the double layer, increasing in magnitude with surface charge and with decreasing ionic strength.  When the surface carries polymer chains which extend away  from the surface into solution, however, they exert an additional drag which reduces the electroosmotic velocity and hence the mobility.  At lower ionic  strengths where the double layer is expanded, if the double layer, and hence the region in which electroosmosis takes place, extends beyond the surface polymer layer the effective drag associated with the chains will be reduced 33  C.)  ITj  p.  U)  z  0  I-..  I  0  (I)  0  CD Ii I....  rt  0 0  w  CD  G)  ‘rJ  0  r  tl C)  .Q  p  CD  ‘-I)  0  ‘:s  ii  o  ‘-3  ::i (API —0  ,.Jp’  Ph  m ‘t-  ft  0  Lii C)  1  ItIi  CD  ‘.3  op ij  (DO  CD  Ii  0  (DO  C)  I-..  wp, (DI,  •p.  0  ‘I  *0  0  and the mobility will tend to behave more like that of a smooth particle. Hence, the ionic strength dependence is a function of the depth of the surface layer, which allows its estimation. If the fixed surface charge is located within the region occupied by the chains the effect on the velocity profile is to reduce the overall drag relative to the case in which the charge is distributed over the smooth surface to which the polymer chains are anchored. This increases the mobility. In the calculations utilized here, the numerical integration program described by Sharp and Brooks (32) was used.  The hydrodynamic drag exerted  by the polymer chains is considered to be equal to the Stokes drag exerted by segments of hydrodynamic radius a present at a uniform density throughout a thickness (3.  The larger the segment concentration, a or  drag and the lower the predicted mobility. program are:  (3, the greater the  The parameters used in this  i) thickness of the glycocalyx, (3; ii) polymer chain density  (number per unit area); iii) the polymer segment radius, a  (A); iv) the  fixed charge density, o 2 (esu/cm ) ; v) the location of the fixed charge; vi) the location of the shear plane.  2.3 Methods 2.3.1 Preparation of Liposomes Liposome behaviour and stability depend on particle size, number of bilayers, chemical composition and the composition of the aqueous phase in which liposomes are formed.  It is essential to control all these factors  if one is to understand the relationship between each of these properties and liposome function (33). There have been many techniques for preparation of liposomes. procedure (33) can be divided into three stages (See Figure 2.4): 35  The  Preparation of Aqueous Phase  Hydration of Lipids: Formation of Liposomes (primary)  Preparation of Mixture of Lipids: Removal of Solvents  Secondary Processing: Formation of Specified Liposomes  Figure 2.4 Procedure of Liposome Preparation  first, preparation of the aqueous phase and lipid mixture; second, lipid hydration and third, optional steps to make specific sorts of liposomes. In preparation of the aqueous phase, several factors must be considered: osniolarity, ionic strength, pH, choice of materials and their concentration for both interior and exterior phases and contaminants. For making a particular molecular mixture of lipids, each of them is first dissolved in a single or mixed solvent.  Generally, all the lipids  should be soluble at the desired concentrations in the chosen solvent system. A dry lipid mixture is made by removing the solvent uniformly since the form of dry lipids can seriously affect the hydration and formation of liposomes.  Usually, a continuous film on the container wall is considered  optimal. The usual hydration method is to disperse the dried lipid mixture into the aqueous phase by shaking.  Following this step, particular techniques 36  are applied for making specific liposomes, for instance of a particular size (34).  In this work, however, only shaken, multilayered liposomes were  used.  2.3.2 Electrophoresis Equipment There have been many kinds of equipment used in particle electrophoresis research for various purposes. was described by Seaman et al.  The equipment in our project  (27) in their work on electrophoresis of  red cells (See Figure 2.5a & 2.5b). In Figure 2.5a, a constant temperature is maintained in a water bath containing a stirring device and a thermostated heater(e, thermostat).  The  calibrated vertical traverse, b, and the dial test indicator, k, provide the readings of the vertical and horizontal position of the chamber.  The  ocular with a fitted graticule, h, serves for the measurement of the distance that a particle moves in a measured interval. The most important part in the electrophoresis apparatus is the chamber, which is mounted horizontally (between a and g); its vertical and horizontal positions are adjusted until the axis of the microscope is located at right angles and passes through the center line of the chamber. There are two kinds of chambers, cylindrical and rectangular, which can provide accurate and reliable results.  In our experiment, a cylindrical  chamber similar to that described in Figure 2.5b was used, except that the stopcocks were replaced by plugs in ground glass joints. The chamber can be divided functionally into two parts: the ends containing two compartments for the electrodes (in KC1 solution) and the central observing section, separated by two pieces of sintered glass discs.  37  1  Figure a, d, h, j,  I  2.5a  Cylindrical microelectrophoresis apparatus: Tube holder; b calibrated vertical traverse; c, crossbar; locking screw; e, thermostat; f, microscope tube; g,objective; ocular with fitted graticule/reticule; i, light source; microscope fine adjustment; k, dial test indicator. (27)  Gloss lmpregnoled RT.F.E. (Fluon) Stopcock Key  Figure  2.Sb All glass small volume cylindrical electrophoresis chamber incorporating a Ag/AgC1/KC1 electrode system with fused—in sintered glass discs. (27) 38  Current was supplied to the electrodes by a power supply operated in constant voltage mode; the voltage and current were read from two digital multimeters. Ag/AgC1 (/KC1) electrodes have been used previously and proven to be the most satisfactory system.  They were treated as described in Section 2.4.  The refurbishing of Ag/AgC1 electrode is performed with nitric acid and ammonia respectively (to recover Ag), followed by replating in a KC1 solution (the detailed procedures are described in Section  2.4 Experimental General Details Centrifugation was carried out with a Micro Centaur Centrifuge (Johns Scientific Inc.).  Solution pH was measured with an Acumet pH Meter 915  (Fisher Scientific) to ±0.02 units.  Current was supplied with a Hewlett  Packard 6212A Power Supply; the voltage and current were read from two Hewlett Packard 3438A Digital Multimeters.  The water used in preparation  of aqueous solutions was from a Millipore Milli—Q Plus ultra—pure water system. All organic solvents and reagents were used as purchased. and tetra-ECG were synthesized in Chapter 1 (See Section 1.4).  Tetra—EC Particular  pH solutions were made by adjusting pH prior to usage with HC1, NaOH or 3 solutions of the same ionic strength as the solution being adjusted. NaHCO  2.4.1 Liposome Preparation The same procedure for preparing liposomes was used for all lipid compositions(See Table 2.1)  39  Table 2.1 Composition of Liposomes in Particle Electrophoresis Composition of Liposomes (molar ratio)  Liposome Number  egg PC 60 60 50 50 60 55 30 30 30 40 40 40  EL*1 EL*2  EL*3 EL*4 EL*5 EL*6 EL* 7-1 EL* 7-2 EL* 7-3 EL* 8-1 EL* 8-2 EL* 8-3  DPPG  Cholesterol  —-  ——  —-  ——  10 10  40 10 10 10  —-  5 10 10 10  30 50  ——  ——  ——  30 50  ——  —-  tetra—EC 40 30 ——  30 20 25 60 30 10 50 20 ——  tetra—ECG ——  10 ——  ——  10 5 ——  ——  —-  10 10 10  Sodium chloride 0.6867 g(11.75 mmol) and sodium azide (NaN ) 0.049 g 3 (0.75 mmol) were dissolved in a 250 ml volumetric flask as the aqueous phase for liposomes.  The lipids in desired concentration were dissolved in  chloroform (DPPG was warmed up to around 43°C), and transferred into a 250 ml round bottom flask (which was cleaned with chromic acid, rinsed thoroughly in water and dried prior to usage). adjusted to 5  —  6 ml with chloroform.  The volume of solution was  The mixture then was dried slowly  with rotary evaporation under reduced pressure at room temperature. Addition of about 5 ml NaC1-NaN 3 solution was followed by incubation of the mixture in a 42°C water bath.  The suspension was centrifuged for 10 to 30  minutes (13000 IPM; radius 65 mm) and the supernatant was removed.  The  liposomes were washed twice in 5 ml of the medium in which they were to be examined by electrophoresis.  The procedure was to pipette the supernatant  40  carefully, resuspending the liposomes with addition of the solution of the desired pH and ionic strength, followed by centrifugation.  The same  resuspension and centrifugation were repeated to ensure the complete replacement by medium of the desired pH and ionic strength.  2.4.2 Electrophoresis General Preparation(See Figure 2.5 for details) All solutions were de—gassed before they were used.  The cylindrical  chamber was cleaned with chromic acid (Cr03/112S04) and thoroughly rinsed before use.  Ag/AgC1 electrodes were washed with nitric acid and ammonia  sequentially, and replated in KC1 solutions at a current density of 0.43 mA/cm . 2 The water bath tank was filled and temperature was set up to be controlled at 25.00+0.02°C.  The optical microscope was focused at the  stationary level, equal to 0.293 x (radius ,i.362 mm), from the inner wall (35).  At this location the fluid velocity caused by electroosmosis  along the walls of the chamber is zero. The electrode cells at both ends of the chamber were filled with KC1 solution and stoppered by electrodes. solution as well when not in use.)  (The center cell was filled with KC1  The voltage was adjusted and controlled  at nominally 40 V. pH Dependence Studies The experimental salt solution (NaCl 50 mM) with adjusted pH, which was also used to wash the liposomes (EL* 1 & EL# 2), was degassed and warmed in the water bath.  The center chamber was filled, after being rinsed  several times, with the warmed solution (See Section 2.3.2 & Figure 2.3b). One of the outlets of the center chamber was stoppered. 41  The liposome  sample was taken in suspension in the same solution in a 1.0 ml syringe and approximately 0.1 ml injected into the viewing region (“Optical Flat”) by a tubing coimected to the tip of the syringe. dilute it and provide a roughly region of the chamber.  The sample was stirred to  uniform distribution near the viewing  The other outlet was stoppered to keep the system  stable during the measurement.  The electric field was switched on and the  motion of particles in the field was observed through the microscope at a magnification of x 320.  The time taken by a particular particle to transit  a fixed distance on the eyepiece graticule was recorded.  The direction of  the current was reversed, producing migration of the same particle in the opposite direction; the mobility of the particle was calculated from the average of the velocities which were first calculated from the pair of readings. Usually, ten particles were observed for each pH solution and the values of voltage and current were read before and after each measurement, as well as the pH values of the NaCl solution.  In the pH dependence  experiments, the range from 1.8-9.9 was examined. When an experiment was finished, the sample could be recovered with the syringe and the center chamber was washed with the next salt solution three times.  For these experiments, the concentration of KC1 solution in the  electrode cells was kept constant (0.050 M), equal to that of NaCl in various pH solutions. Ionic Strength Dependence Studies The procedure for the ionic strength experiments was almost the same as the previous one, except that the concentration of KC1 solution in the electrode cells was changed to be the same as that of the NaCl solution. 42  The ionic strength of the salt solution varied from 0.001 M to 0.100 M, with a constant pH  The liposome samples were EL* 3  7.  —  —  EL* 6. Different Compositions of Liposoes The procedure was as described above. NaCl with pH  7.  The salt solution was 0.050 M  The liposomes were EL* 7—1, EL* 7—2, EL* 7-3, EL* 8—1,  EL* 8-2, and EL* 8—3.  From EL* 7—1 to 7-3, the molar ratio of PEG chains  decreased, while the charged lipid, DPPG, was held constant at 10%.  From  EL# 8-1 to EL* 8—3, the molar ratio of PEG chain concentration varied while the charged lipid, tetra-ECG, was held constant at 10%.  2.5 Result and Discussion The mobility and apparent charge density of liposomes for all the experiments were calculated from the velocity data in Appendix III, with the classical theory of electrophoresis of smooth particles (See Section 2.2), giving Table 2.2, Table 2.3 and Table 2.4. Section 2.2.1.  The constants are given in  In calculations of the apparent charge densities, the  equivalent conductivity A is given below for NaC1 solution at 298°K (30):  concentration (M) A (cm ) 1 equiv 21  0.15 103.89  0.1 106.74  43  0.05 111.06  0.02  0.005  0.001  115.76  120.64  123.74  Table 2.2  Mobility and Charge Density with Variation of pH at Constant Ionic Strength I = 0.050 M for Liposomes EL* 1 & ELI 2 C’—” and 11+11 are the signs of charges) Electrophoretic Mobility (jim-cm/sec-volt: ±S.D.)  pH (±)  Apparent Charge Density (esu/cm2: ±S.D.)xO.001  Liposome *1  Liposome *2  Liposome *1  Liposome *2  9.88±0. 14  —0.77±0.10  —2.23±0.09  —2.64±0.34  -7.67±0.32  9.43±0.09  -1.53±0.19  —3.36±0.19  —5.27±0.66  —11.56±0.65  7. 97±0. 17  —0.48±0.11  —2.40±0.24  —1.66±0.36  -8.24±0.81  7. 30±0. 11  0.34±0.12  —2.34±0.26  1. 18±0.39  -8.05±0.90  7.03±0.05  0.07±0.07  —1.92±0.15  0.23±0.26  -6.60±0.51  6.87±0.06  0.16±0.05  —2.08±0.19  0.55±0.19  —7.15±0.67  5.99±0.02  0.24±0. 15  —2.03±0.26  0.83±0.50  —6.99±0.89  4.95±0.02  0.24±0.07  —1.48±0.24  0.82±0.23  —5.07±0.82  4.91±0.02  0.23±0.03  —1.47±0.21  0.80±0.11  -5.07±0.73  3.98±0.02  0.35±0.06  —0.64±0.17  1.20±0.20  —2.18±0.60  2.97±0.02  0.60±0.07  0.17±0.09  2.07±0.24  0.58±0.29  1.90±0.02  0.80±0.05  0.74±0.06  2.75±0.16  2.55±0.21  1.89±0.03  0.88±0.04  0.82±0.09  3.01±0.15  2.83±0.32  Table 2.3 Mobility and Apparent Charge Density with Variation of Compositions at Constant pH — 7 and Constant Ionic Strength 0.05 M *  Liposome  EL*7-1 ELI 7_2A* ELI 7_2B* EL*7—3 EL*8—1 EL* 8—2 EL*8—3  A and B are two measurements for one composition Composition (molar ratio): egg—PC/cholesterol/tetra—EC /DPPG/tetra-ECG 30/ 0/60/10/ 0 30 / 30 / 30 / 10 / 0 30 / 30 / 30 / 10 / 0 30/50/10/10/0 40/ 0/50/ 0/10 40 / 30 / 20 / 0 / 10 40/50/ 0/ 0/10  44  Electrophoretic Mobility (pm-cm/sec-volt: ±S.D.) —3.47±0.45 —3.46 ± 0.39 —2.93 ± 0.83 —3.13±0.31 —1.53±0.16 —1.85 ± 0.02 —2.29±0.11  Table 2.4 Mobility and Apparent Charge Density with Variation of Ionic Strength at Constant pH around 7 for Liposomes *3,4,5 & 6 *  repeated measurement  Ionic Strength (M) 0.001 0.005 0.005* 0.020 0.050 0.100 Ionic Strength (M) 0.001 0.005 0.005* 0.020 0. 050 0.100  Electrophoretic Mobility (pm—cm/sec-volt: ±S.D.)  Apparent Charge Density (esu/cm’2: S.D. )xO.001  Liposome *3  Liposome *4  Liposome *3  Liposome *4  -4.39±0.40 -2.85±0.32 —3.58±0.33 —3.13±0.26 —2.71±0.20 -1.55±0.27  —3.78±0.40 —2.14±0.31 —3.14±0.38 —1.78±0.19 —1.20±0.19 —0.91±0.14  -1.23±0.11 -1.79±0.20 -2.24±0.21 -3.92±0.33 -5.37±0.41 -4.35±0.77  —1.06±0.11 —1.34±0.19 -1.97±0.24 -2.23±0.23 -2.39±0.38 -2.56±0.39  Electrophoretic Mobility (pm—cm/sec—volt: ±S.D.)  Apparent Charge Density (esu/cm2: S.D. )xO.001  Liposome *5  Liposome *6  Liposome 415  Liposome *6  —3.79±0. 16 —3.50±0. 13 —3.27±0. 18 —3.27±0.16 —2.65±0.19 —1.90±0.18  —4. 11±0. 18 —3.54±0.35 —3.67±0.25 —2.71±0.14 -2.11±0.15 -1.60±0.45  —1.06±0.04 —2.20±0.08 —2.05±0.11 —4.10±0.20 —5.25±0.38 —5.33±0.52  -1.15±0.05 —2.22±0.22 —2.30±0. 16 —3.40±0.17 —4.18±0.31 —4.49±1.26  2.5.1 pH Dependence Since the liposome *2 (EL* 2) has the composition of tetra—ECG instead of tetra—EC in the liposome *1 (EL* 1), the mobility should differ due to the presence of PEG chains and the location of the charges, even though the net charge density is expected to be the same. From Table 2.2, a phenomenon was observed that EL*1 with a neutral surface had low mobilities in all the pH range, because the surface of liposomes can adsorb ions which are concentrated in solution by van der Waals interactions.  The net charge density was calculated with deduction  of the results of EL* 1 from those of EL* 2.(See Table 2.5)  45  Table 2.5  Calculation of Net Electrophoretic Mobility, Net Charge Density of Liposome *2 and pH at Surface of Liposomes with Variation of pH at Constant I = 0.050 M for Charged Liposome EL* 2 (o’=o’—c)  pH (measured)  Electrophoretic Mobility U’ = U2 Ui (aim—cm/sec—volt: ±S.D.)  Net Surface Charge Density (esu/cm’2)x 0.001  pH at surface of liposome  9.88±0. 14  —1.46 ± 0.14  —5.03 ± 0.47  9.56±0. 14  9.43±0.09  -1.83 ± 0.27  —6.28 ± 0.93  9.03±0.10  7.97±0.17  -1.92 ± 0.26  —6.59 ± 0.89  7.55±0.18  7.30±0. 11  -2.69 ± 0.29  —9.23 ± 0.98  6.71±0. 13  7.03±0.05  -1.99 ± 0.17  —6.83 ± 0.57  6.59±0.06  6.87±0.06  -2.24 ± 0.20  —7.70 ± 0.70  6.38±0.07  5.99±0.02  -2.27 ± 0.30  —7.82 ± 1.02  5.49±0.07  4.95±0.02  -1.72 ± 0.25  —5.90 ± 0.85  4.57±0.06  4.91±0.02  -1.71 ± 0.21  —5.87 ± 0.73  4.54±0.05  3.98±0.02  -0.99 ± 0.18  —3.38 ± 0.63  3.76±0.04  2.97±0.02  —0.44 ± 0.11  —1.50 ± 0.38  2.87±0.03  1.90±0.02  —0.06 ± 0.08  —0.29 ± 0.27  1.89±0.03  1.89±0.03  —0.05 ± 0.10  —0.18 ± 0.35  1.88±0.04  -  Theoretically, the surface charge density of liposome *2 is constant and can be calculated from its lipid composition.  The surface charge  density is expressed as the net charge on the surface divided by the surface area.  The composition of EL* 2 is known as 60Y. egg PC, 30°?.  tetra-EC and 10°!. tetra—ECG.  The surface area occupied per molecule of  egg PC was estimated by De Young et al. (36) as 50 A (in presence of 40% 2 cholesterol).  With the assumption that the surface area taken per lipid  for tetra—EC or tetra—ECG is the same as that of cholesterol, which was reported to be 37 A 2 (36), the surface charge density was calculated to be 1.07 x 10  4  esu/cm  2 .  It is obvious from Table 2.2 that the results  46  calculated from electrophoresis were lower than this estimate. The pH at the particle surface was calculated with the classical theory and the mobilities of the liposome particles were plotted as a function of surface pH (see Table 2.5 & Figure 2.6). The pKa of the charged liposome, which is also the pKa of the charged lipid (tetra-ECG), calculated from Figure 2.6, was equal to 3.9.  2.5.2 Composition Dependence Two groups of liposomes, with variation of PEG chain density at the surface of the particles, were observed.  The plot of the mobility vs  the molar percentage of PEG chains in each liposome is given in Figure 2.7. In the liposomes *7—1, 7-2 and 7-3, the negative charge was contributed from lOX (molar) DPPG and located on the surface of the bilayer.  The  mobilities can be considered unchanged within the experimental error when the PEG chain density was varied. In the results of the liposomes *8—1, 8—2 and 8—3, the mobility slightly decreased while the PEG chain concentration increased. (The charge, contributed from tetra—ECG in these liposomes was located at the outer plane of the glycocalyx.) density increases.  This is the usual case when the polymer chain  The resistance of the motion of liposomes increases,  causing mobility to decrease; while the charge density is kept constant. For the liposomes *7—1, 7-2 and 7—3 (with constant lOX DPPG), however, the mobility was almost unchanged while the PEG chain density increased.  47  1•  ---------  U)  I  ‘  0  3  —4  I  4  1  Figure 2.6  I  2  4  3  1  4  4  4  4  j  5  vs  I  I  6  I  pH(surface)  7  4  I  4  8  4  9  I  I  10  4  11  Net Liposome Mobility pH at Surface of Liposomes for EL*2 at constant ionic strength 0.050 H  0  -0.5  x  EL#7-1,7-2 & 7-3 EIA8-1,8-2 & 8-3  -1 o  E E  a.  zo  -I.e  48_i  8-2  -2  .8-3  -2.5 7—2B  -3  -4  cf7  72A  4.5 0  4  10  4  20  4  30  40  I  50  PEG Chain Concentration (molar%)  vs  70  Figure 2.7 Electrophoretic Mobility PEG Chain Concentration for Different Compositions at ionic strength 0.050 H and pH 48  ‘-  7  2.5.3 Ionic Strength Dependence The mobilities of the liposomes EL* 3, EL* 4, EL* 5 & EL* 6 are plotted as a function of ionic strength in Figure 2.8. EL*3 and ELM have the same charge locations (DPPG, inner plane) and charge density, but different PEG chain concentrations.  The EL*3 particle,  which has lower PEG chain density, moves more quickly in the electric field than EL*4 at all ionic strengths. EL*4, EL*5 and EL*6 have the same PEG chain density and charge density, but different charge locations.  The mobilities of EL*6 are  between those of EL*4 and ELlIS at high ionic strength, as expected since the charges in EL*4, ELliS and EL*6 are distributed at the inner plane, outer plane and both positions, respectively.  At low ionic strength, the  mobilities are almost the same within experimental error. The numerical model was used to fit the data.  The results are listed  in Table 2.6 and the figures with a more detailed description of the fitting procedure in Appendix IV.  49  L#3 A L#4  4  +  [2.4  -4  -o  0.00  0.02  Figure 2.8a  0.04  I  I  0.06 0.08 Ionic Strength (M)  I  0.10  0.12  Electrophoretic Mobility vs Ionic Strength for EL*3 and *4 at constant pH — 7  0•  A L#4  -0.5  + 0.00  Figure 2.8b  0.02  0.04  •I  0.06  Ionic Strength (M)  0.08  Electrophoretic Mobility vs Ionic Strength for EL*4, *5 and *6 at constant pH — 7 50  0.12  Table 2.6  Comparison of Mobility Model Parameters  for Four Liposome Preparation (EL*3,4,5 & 6)  Preparation  Ionic Strength (mM)  Charge Density (esu/cm  2  Molecular Density  )  (molec./cm  Extension 2  )  (A)  EL*3: CHEMI CAL DPPG 10% egg PC 50% MOBILITY  >10 <10  10700  2.2E+13 1.1E+14  8 8  8500 8500  1.3E+14 1.3E+14  8 >8  10700  2.2E+-13 1.1E+14 6.6E+13  8 8 15  9500  1.3E+14 6.6E+13  EL*4: CHEMI CAL DPPG 10% egg PC 50°!. tetra-EC 30% MOBILITY  1—150  8 15—73  EL*5: CHEMICAL egg PC 60% tetra-EC 20% tetra—ECG 10% MOBILITY  1.3E+14 4.4E+13 2.2E+13  8 15 21  2.2E+13 4.4E+13 2.2E+13 4.4E+13  21 15 21 15  5350  1.1E+13 1.2E+14 5.5E+13 1.1E+13  8 8 15 21  4250 2875  6.2E+13 2.2E+14  15 21  10700 >10  5750  <10  <5000  EL*6: CHEMI CAL DPPG 5°!. egg PC 55°!. tetra—EC 25% tetra-ECG 5% NOBILITY  5350  >10  51  2.6 Conclusion Tetraethoxycholesterol and tetra-EC terminated with glucuronic acid, tetra—ECG, were successfully synthesized and purified in sufficient yield to allow their investigation in model membranes.  A series of liposomes  composed of egg PC, DPPG, cholesterol, tetra—EC and tetra-ECG were made and observed with particle electrophoresis.  The pKa of tetra—ECG was estimated  to be 3.9 from the pH dependence of tetra—ECG—containing liposomes, taking into account the difference in pH between the surface region and the bulk phase when the surface is negatively charged. The electrophoretic mobilities of the liposomes were measured as a function of ionic strength of the suspending medium.  The classical theory  for smooth particles was found not to describe the data, particularly when PEG chains were anchored in the surface.  The model of Sharp and Brooks was  found to be more successful in describing the general effects of tetra—EC and tetra-ECG, allowing the experimental data to be fit with physically reasonable parameter values for chain extension and charge density at ionic strengths above 10 mM.  The data taken at low ionic strengths did not fit  either theory in the presence or absence of surface polymer chains, however, suggesting that the surface charge density was not constant under these conditions, possibly due to the adsorption phenomena.  52  REFERENCES  1.  Jam, M.K., Nonrandom Lateral Organization in Bilayers and Biomembranes, in Membrane Fluidity in the Biolov Vol.1, ed. Aloia, R.C., Academic Press, New York, 1983  2.  Cullis, P.R., Hope, M.J., Physical Properties and Functional Roles of Lipids in Membranes, in Biochemistry of Lipids and Membranes , ed. Vance, D.E., Vance, J.E., 1992  3.  Gorter, E., Grendel, F., Journal of Experimental Medicine, 41, 439£143pp, 1925  4.  Singer, S.J., Nicolson, G.L., Science, 175, 72O—73Opp, 1972  5.  Brockerhoff, H., Molecular Designs of Membrane Lipids, in Bioorganic Chemistry, lpp, 1977  6.  Ponpipom, M.M., Shen, T.Y., Baldeschwieler, J.D., Wu, P.-S., Liposome Technology, Vol. III, 95—ll5pp, 19  7.  Sather, P.J., “Synthesis of Cholesterol Based Model Glycolipids”, Master Thesis, The University of British Columbia, Canada, 1990  8.  (a)Kunitake, T., Okahata, Y., Bulletin of Chemical Society of Japan, Vol.51(6), 18T1—1879pp, 1978 (b)Farhood, H., Bottega, R., Epand, R. M., Huang, L., Biochimica et Biophysica Acta, 1111, 239—246pp, 1992  9.  Demel, R.A., De Kruyff, B., Biochimica et Biophysica Acta, 457, 132pp, 1976  109-  10. Johnston, D. S., Coppard, E., Parera, G. V., Chapman, D., Biochemistry, 23, 69l2-69l9pp, 1984 11. Crowe, J.H., Crowe, L.M., Carpenter, J.F., Rudolph, A.S., Wistrom, C.A., Spargo, B.J., Anchordoguy, T.J., Biochimica et Biophysica Acta, 974(2), 367—384pp, 1988 12. Chandrasekhar, I., Gaber, B.P., Journal of Biomolecular Structure & Dynamics, Vol.5, ll63—ll7lpp, 1988 13. Harris, J.M., Introduction to Biotechnical and Biomedical Applications of Poly(ethylene glycol), in Poly(Ethylene Glycol) Chemistry: Biotechnical and Biomedical Applications, ed. Harris, J.M., Pleum, New York, 1992 14. Brockerhoff, H., Ramsammy, L.S., Biochimica et Biophysica Acta, 691, 22’T—232pp, 1982 15. Patel, K.R., Li, M.P., Schuh, J.R., Baldeschwieler, J.D., Biochimica et Biophysica Acta, 797, 2O-26pp, 1984 53  16. Patel, K.R., Li, M.P., Schuh, J.R., Baldeschwieler, J.D., Biochimica et Biophysica Acta, 814, 256-264pp, 1985 17. Bollenback, G.N., Long, J.W., Benjamin, D.G., Lindquist, J.A., Journal of the American Chemical Society, 77, 33lOpp, 1955 18. Paudler, W.W., Nuclear Magnetic Resonance, Chapter 3, Allyn and Bacon, Boston, 1971 19. Goodrich, R.P., Crowe, J.H., Crowe, L.M., Baldeschwieler, J.D., Biochemistry, 30, 5313-S3l8pp, 1991 20. Schneider, J. J., Bhacca, N. S., Journal of Organic Chemistry, 34(6), 199O—1993pp, 1969 21. Goodrich, R.P., Handel, T.M., Baldeschwieler, J.D., Biochimica et Biophysica Acta, 938, l4Z3—lS4pp, 1988 22. Pathak, V.P., Synthetic Communications, 23(1), 83-SSpp, 1993 23.  (a)Wu, S.-H., Ong, G.-T., Hsiao, K.-F., Wang, K.-T., Journal of the Chinese Chemical Society, 39, 6’TS—682pp, 1992 (b)Chang, K. -Y., Wu, S. -H., Wang, K. -T., Carbohydrate Research, 222, l2lpp, 1991  24. Bangham, A.D., Liposomes: An Historical Perspective, in Liposomes, ed. Ostro, M.J., 1—26pp, Marcel Deckker Inc., New York, 1983 25. Juliano, R.L., Interaction of Proteins and Drugs with Liposomes, Liposomes, ed. Ostro, N.J., Marcel Dekker Inc., New York, 1983  in  26. McDaniel, R.V., Sharp, K., Brooks, D.E., McLaughlin, A.C., Winiski, A.P., Cafiso, D., MaLaughlin, S., Biophysical J., 49, 74l-752pp, 1986 27. Seaman, G.V., Electrokinetics Behavior of Red Cells, in The Red Blood Cell, Vol.2, ed. Surgenor, D.M., Academic Press, New York, 1975 28. Popiel, W.J., Introduction to Colloid Science, Chapter 9, lS2pp, Exposition Press, New York, 1978 29. Davies, J.T., Rideal, E.K., Interfacial Phenomena, Chapter 2, 75pp, Academic Press, London, 1963 30. Robinson, R.A., Stokes, R.H., Electrolyte Solutions, 4l—45pp & 46S—466pp, Butterworth, London, 1959 31. Levine, S., Levine, M., Sharp, K.A., Brooks, D.E., Biophysical Journal, 42, l27-l35pp, 1983 32. Sharp, K. A., Brooks, D. E., Biophysical Journal,  4i  S63-566pp, 1985  33. Woodle, M.C., Papahadjopoulos, D., Liposome Preparation and Size Characterization, in Methods in Enzvmology, Vol.171, ed. Colowick, S.P., Kaplan, N.0., 1989  54  34. Jousma, H., Taisma, H., Spies, F., Joosten, J.G.H., Junginger, H.E.,  Crommelin, D. J. A., International Journal of Pharmaceutics, 35,  263—274pp, 1987  35. From personal communication with Robert J. Knox, Western Biomedical  Research Institute, P.O.Box 22510, Eugene, Oregon 97402-0419  36. De Young, L.R., Dill, K.A., Biochemistry,  55  ,  5281—5289pp, 1988  LIST OF APPENDICES  Page Appendix  I  Appendix  II  Intel III: View of the Model Lipid  Al  (a) Proton NMR analysis of Brominated Sugar  A2  (b) Estimation of the Coupling Reaction Yield of Tetra-ECPG with Proton NNR  A3  (c) Proton NMR Result of Tetra—ECG  A4  Appendix III  Velocity Data Table of Electrophoresis  AS  Appendix  Data Fit to Equations for Mobilities of  IV  Hairy Particles  56  A8  Appendix I  Molecular Model of Tetra-ECG  (Builded and studied with INSIGHT II Modeling System)  Al  Appendix II (a) 1 H NMR Analysis of Brominated and Protected  Sugar()  Part of the whole spectrum of proton NMR 400 MHz is shown with the chemical shift varying from 4.3 to 7.3 ppm. The single peak around 7.2 is from the solvent deuterium chloroform. The signals at 4.57, 4.84 , 5.22, 5.59 and 6.62 ppm refer tollS, H2, H4, 113 and Hi respectively. The coupling constant of Hi and 112, .312, is measured and calculated to be 4 Hz, smaller than that of the axial—axial coupling, 9—14 Hz (17); while .323, .334 and J45 are the same 11Hz.  114  CH  1 OC  0  640  i  7.0  .  I  4.6  I  I  I  6.5  I  I  6.0  A2  I  I  I  I  5.5  I  I  I  I  5.0  ‘  45  Appendix II (b)  ‘H NNR Analysis of the Compounds in Coupling Reaction The estimation of the molar ratio of tetra—EC and tetra—ECPG is based on the assumption that the mixture only composes of these two compounds. At the left below a table shows whether or not and which proton(s) of tetra—EC or tetra—ECPG contribute to the NNR integral of a particular signal.  tetra—EC  Shift(ppm) 0.4-2.6  44Hc  3,2 3.44.5 4.5—5.6 a cd  s  a  p  iHo l6Hp lHcd  Assume the molar ratio of tetra—EC and tetra-ECPG is Xi : X2. Assign the relations as:  tetra-ECPG 44Hc+9Ha  = ITGL 1 =ITGL xi 2 16i + 2 OX2 = ITGL 3 Xi 5X2=ITGL4  44x1  l6Hp+3Ha+lHs 4Hs+lHcd  acety]. group c cholesterol double bond in cholesterol methyl group on sugar part hydroxyl group PEG chain s sugar ring  + 53X2  The ratio x’ : X2 can be given by any two of these equations.  I  -J  w I-.  8.0  7.0  6.0  2 T ’ o jT i’o i  PPM  A3  0.0  A4  Appendix II (c) 200 MHz I1 NMR Analysis of Deprotected Compound ()  Chemical Shift (ppm)  Integral/Ratio 412.7/45.9 217.4/24.2 9.00/1  0.4—2.6 3.0—4.8 5.4 c cd o  tetra-ECG  44Hc  l6Hp+3Ho+5Hs lHcd  K  sugar ring • cholesterol double bond in cholesterol p PEG chain hydroxy group  11  /•  ‘IL  A trJj  -< CD ‘Ii I  z 14  •  I  10.0  I  I  I  I  9.0  6.0  6.0  5.u  PPM  0  1.0  0.0  Appendix III  Particle Velosity Data Tables  Velosity of liposome particle is:  v  =  dlt  =  where  d is the distance the particle moves in particular time; D Is the length per division, 0.024mm; n is the number of division.  The average velosity is obtained from the average of the inversions of all the time readings from motion of different particles, multiplied by d. Liposoine 1 1 (EL* 1) (constant Ionic strength at 0.05014) pH (±)  9.88 ± 0.14 9.43 ± 0.09 7.97±0.17 7.30 ± 0.11 7.03 ± 0.05 6.87 ± 0.06 5.99 ± 0.02 4.95 ± 0.02 4.91 ± 0.02 3.98 ± 0.02 2.97 ± 0.02 1.90 ± 0.02 1.89 ± 0.03  Charge Sign (+1—) —  —  —  + + + + + + + + + +  Average Velosity (cmlsec:±S.D.) x 10000  —2.63 ± 0.34 —5.26 ± 0.66 —1.67±0.37 1.12 ± 0.47 0.23 ± 0.26 0.55 ± 0.19 0.80 ± 0.52 0.82 ± 0.23 0.81 ± 0.11 1.21 ± 0.20 2.19 ± 0.25 4.63 ± 0.28 5.12 ± 0.25  Current  Voltage  (mA)  CV)  1.11 1.11 1.12 1.10 1.12 1.11 1.11 1.11 1.12 1.12 1.18 1.87 1.89  40.01 39.91 40.00 39.91 40.01 40.02 40.01 40.01 40.02 40.01 39.98 39.91 40.00  Liposome * 2 (EL* 2) (constant ionic strength at 0.05CM) pH (±) 9.88 9.43 7.97 7.30 7.03 6.87 5.99 4.95 4.91 3.98 2.97 1.90 1.89  ± ± ± ± ± ± ± ± ± ± ± ± ±  0.14 0.09 0.17 0.11 0.05 0.06 0.02 0.02 0.02 0.02 0.02 0.02 0.03  Charge Sign (+1—) —  —  —  —  —  — —  —  —  —  + + +  Average Velosity (cm/sec:±S.D.) x 10000 —7.66 —11.53 —8.23 —7.94 —6.58 —7.14 —6.98 —5.06 —5.08 —2.20 0.61 4.31 4.80  AS  ± ± ± ± ± ± ±  ± ± ± ± ±  0.32 0.65 0.81 0.88 0.51 0.67 0.87 0.82 0.73 0.63 0.31 0.36 0.53  Current  Voltage  (mA)  CV)  1.11 1.11 1.11 1.10 1.12 1.11 1.11 1.11 1.12 1.12 1.18 1.88 1.89  40.00 39.91 40.00 39.95 40.02 40.02 40.02 40.02 40.01 40.01 39.99 39.91 40.00  Appendix III  Particle Velosity Data Tables  Liposome *3 (EL* 3) (constant pH) Ionic pH Strength (M) (±0.02) 0.005 0.020 0.050 0.100  6.94 6.91 6.88 6.95  Average Velosity (cm/sec: S.D.) x 10000 -15.00 —11.89 9.27 5.36 -  -  ± ± ± ±  1.37 0.99 0.70 0.94  Current  Voltage  (mA)  (my)  0.127 0.461 1.11 2.10  39.82 39.89 39.90 39.94  Current  Voltage  Liposome *4 (EL* 4) (constant pH) Ionic pH Strength CM) (±0.02) 0.005 0.020 0.050 0.100  6.94 6.91 6.88 6.95  Average Velosity (cm/sec: S.D.) x 10000 -13.16 6.77 4.12 3.81  -  -  -  ± ± ± ±  1.60 0.70 0.65 0.49  (mA)  (mV)  0.127 0.462 1.11 2.11  39.84 39.91 39.92 39.95  Current  Voltage  Liposome *5 (EL* 5) (constant pH) Ionic pH Strength (H) (±0.02) 0.005 0.020 0.050 0.100  6.97 6.91 6.88 6.95  Average Velosity (cm/sec: S.D.) x 10000 -15.03 -12.51 9.08 6.57  —  —  ± ± ± ±  0.56 0.60 0.65 0.63  (mA)  (mV)  0.130 0.464 1.11 2.10  39.99 39.99 39.93 39.93  Current  Voltage  Liposome *6 (EL* 6) (constant pH) Ionic pH Strength CM) (±0.02) 0.005 0.020 0.050 0.100  6.94 6.91 6.88 6.95  Average Velosity (cm/sec: S.D.) x 10000 -15.40 —10.34 7.24 5.59  — —  ± ± ± ±  A6  1.06 0.52 0.53 1.56  (mA)  (mV)  0.127 0.463 1.11 2.12  39.84 39.97 39.94 39.99  Appendix III  Particle Velosity Data Tables Velosity of Electrophoresis for Liposome Composition Dependence (constant pH arid ionic strength)  Liposome with different composition EL* EL* ET_ EL* EL* EL* EL  7—1 7_2A* 7—2B 7—3 8—1 8-2 8-3  Average Velosity (cm/sec: S.D.) x 10000 —11.79 —11.76 9.97 —10.65 5.19 6.31 7.79  —  —  —  -  ± ± ± ± ± ± ±  1.54 1.31 2.82 2.19 0.54 0.69 0.36  Current  Voltage  (mA)  (mV)  1.10 1.10 1.10 1.10 1.10 1.10 1.10  40.00 39.91 39.92 39.89 39.93 39.93 39.92  PH 6.96 ± 0.02, Ionic Strength 0.050 H A & B are two measurements for composition EL* 7—2  A7  Appendix IV  Data Fit to Equations for )4obilities of Hairy Particles Contributed by X. Song, J. Janzen and D. E. Brooks  Model Parameters The mobility model assumes a region, the glycocalyx, exterior to the bilayer which contains surface associated molecules and suspending medium. Some of these associated molecules bear charged groups.  The mobility is  calculated as a function of the ionic strength of the suspending medium.  The composition data for the liposomes and structural data for the molecular species were used to calculate parameter values which were expected to fit the mobility data. improve the fit.  The parameter values were then varied to  Assessment of the degree of fit was qualitative.  The mole percent data for the liposome components were first converted into molecular surface densities, i.e., molecules/cm . 2 used the reported areas per molecule for egg PC (50 (37  ) 2 A  in 40 mole h cholesterol bilayers (36).  ) 2 A  This calculation and cholesterol  Surface charge densities  were calculated from molecular densities assuming one ionized group per carboxylic acid and phosphodiester residue.  Ten mole percent of a species  with a single charge per molecule corresponds to 2.2 x 1013 molecules/cm 2 and 10700 esujcm . 2  In the mobility calculation the electrostatic charge density contributes a positive term to the magnitude of the particle mobility while the interaction of anchored molecular chains with the suspending medium is resistive.  In the current program, electrostatic charge may be specified  independently at the inner and outer interfaces of the glycocalyx and as a A8  diffuse charge within the glycocalyx.  In the latter case the volume  density of the charge is calculated from the surface charge density and either a gaussian distribution function or a uniform distribution function. Diffuse charge distributions were not used in these calculations.  Stokes resistance is characterized by a radius which is estimated here from structural data. characteristic.  In reality this parameter is a hydrodynamic  This parameter is referred to here as the segment radius  and was set equal to one—half the unit cell length of the head group chain  of tetraethoxycholesterol (tetra—EC), 1.85  A.  The resistive term varies with distance from the bilayer due to the velocity gradient across the glycocalyx and changes in the volume density of resistive segments.  In the particle frame of reference the hydrodynamic  shear plane determines the zero velocity location. was set at the bilayer to glycocalyx interface (0  In the calculations this  A).  The volume density of  segments is calculated from the surface density of segments and the length over which these are distributed.  Two resistive elements, independent  except for the common segment radius parameter, may be used.  One (polymer  I), associated with the glycocalyx, assumes a uniform distribution over the glycocalyx limits.  The other (polymer II) assumes a uniform distribution  over lower and upper limits freely set between 0 X and 120 V. of the glycocalyx’s depth.  The latter was originally included to model adsorbed  polymer but is used here to include the effect of different dimensions for tetra-EC and the glucuronic acid derivative of tetra—EC (tetra-ECG) when they are present in the same liposome preparation.  The surface density of segments is calculated from the surface A9  density of molecular species and the number of segments per molecule.  The  latter was calculated from molecular lengths and the assumed segment radius.  Fitting Results  The mobilities of three liposome preparations EL*3, EL*4 and EL*5 were fitted as functions of ionic strength.  A fourth, EL*6, was intermediate in  composition and behaviour to EL*4 and EL*5.  The EL*3 preparation was 50 7. egg PC (PC), 107. DPPG (PG) and 40 ‘4 cholesterol.  The model for the surface region has the charge at the  bilayer—glycocalyx interface, i.e., inner charge at 0 surface density was 1.32 x 1014 molecules/cm . 2  A.  The combined  Phospholipid and cholesterol  head group resistance was not considered initially but its effect was subsequently examined.  First, the charge density was varied while the head group resistance was held at zero (Fig. A4-1). Under this condition mobilities above 10 mM NaC1 were best fit by a surface charge of 4500 esu/cm . 2  This is much lower  than the value calculated from the chemical composition of the lipid mixture from which the liposomes were formed, 10,700 esu/cm . 2  Below 10 mM  NaC1 successively lower charge densities were required.  The head group resistance was then included (Figure A4—2) by allowing the PC and PG molecules to extend from the bilayer 8 the molecular structure.  Two cases were considered.  A, as estimated from First, the number of  segments per unit volume was varied in proportion to the depth. AlO  Thus the  segment volume density was held constant, implying that changing the length of the molecule changes the molecular mass. This does not coincide with experimental reality, however.  Secondly, the segment volume density was  held constant at the value used for the BA depth.  In this case the segment  volume density varies inversely with the extension, but the molecular mass is constant.  This models a molecule which could collapse towards the  surface, for instance.  Comparison of the curves for the two case showed  almost no difference over the length variation so in subsequent fitting only the constant molecular mass case was calculated.  The mobility was significantly reduced at all ionic strength by adding the head group resistance.  The charge density had to be increased to  8,500 esulcm 2 to match the high ionic strength data, a value near to that calculated from the assumed chemical composition.  The EL*4 preparation was 50 % PC, 10’!. PG. 10°!. cholesterol and 30% tetra-EC.  The model for the surface region has the charge at 0  and PG parameters are as in preparation EL*3.  A.  The PC  The PEG chain on tetra-EC was  estimated from molecular models to extend about 15  A and the surface  density was 6.6 x 10 molecules/cm . 2  The data was fit by assuming (Figure A4—3).  =  15  A and varying the charge density  A value of 9,000 esu/cin , very close to that used to fit the 2  bare PC/PG surface, fit the mobility value at 0.10 M. density constant strengths.  With the charge  had to be varied to fit the points at the lower ionic  The thicknesses had to be continuously increased to fit the  data as lower ionic strength data was considered, the range being 15 to 73  A. All  A set of calculations was also carried out with both an 8  A layer for  the PC/PG head groups and a greater extension for the tetra—EC chains as described above.  It was found that the resistance increase was small.  A similar mobility fitting to that achieved above required the charge density to be increased to 9,500 esulcm . 2  However, the same range of  thicknesses fit the data at lower ionic strengths when the charge density was held at 9,500 esuJcm . 2  The EL*5 preparation was 60 X PC, 10% cholesterol, 20% tetra—EC and 10% tetra-ECG.  The model for the surface region has the charge at the  outside of the glycocalyx adjacent to the free medium. density is 1.32 x 1014 molecules/cm . 2 4.4 x  1013  The PC surface  The tetra—EC surface density was  2 and its extension taken to be 15 molecules/cm  head group was estimated to extend about 21 2.2 x 1013 molecules/cm . 2  A.  The tetra—ECG  A and the surface density was  The tetra—ECG surface charge density was varied  (Figure A4—4) and found to fit the mobility in 0.10 M NaC1 at 5,750 esulcm 2 located at the extreme end of the molecule.  The next two ionic strength  points were likewise fit with this parameter set, but the two lowest values fell well above the line.  It is evident from the curves for other  values that the reduction in  mobility with increase in molecular extension is much less in this instance than that obtained when the surface is modeled with all the charge at the base of the chains.  The mobility is less sensitive to the chain extension  when the charge is moved away from the the solid surface, presumably because the chains near the fixed charge plane offer less drag than the solid surface, at which the relative velocity must be zero. A12  The EL*6 preparation contained 25% tetra—EC and 5% tetra—ECG with 5% PG, 55% PC and 10% cholesterol.  Hence, the overall chain density was as above  but the charge was distributed half on the ends of the tetra-ECG chains and half at their base.  In this case (Figure A4—5) an acceptable fit to the  three highest ionic strength points was obtained if the charge values determined from best fits for EL*4 and ELN5 were each halved and assigned to the inner and outer charge layers in EL*6 and the polymer extensions taken as 15 A and 21 A for the tetra—EC and tetra—ECG chains respectively, demonstrating the consistency of the modeling at moderate ionic strengths.  Discussion  It is clear from the above results that the model of Sharp and Brooks can explain the general features of the effect on the electrophoretic mobility of adding neutral polymer chains to charged liposomes providing the data is not taken at ionic strengths below 10 mM.  The values for the  charge density agree reasonably well with those expected from the lipid composition provided the charge is located at the plane of the solid surface.  In order to fit data obtained when the charge was located at the  end of the chains on tetra—ECG the value was somewhat lower, however.  The inability of the theory to fit the data in any system at low ionic strength, in the presence or absence of surface—associated polymer chains, suggests that the charge density in fact varied as a function of ionic strength.  This could reflect the binding of cations or impurities to  the phosphate and acidic sugar moieties, a possibility that was not examined quantitatively.  The fact that B had to be increased as the ionic A13  strength was decreased when the charge was located at the base of the PEG chains could be interpreted to mean the chain extension increased in lower ionic strength media.  This seems an unlikely explanation, however, as the  chain is electrically neutral and and its configuration would not be expected to be very sensitive to ion concentration in the range examined here.  It seems more likely the effect is associated with the relatively  low values of mobility observed in low ionic strengths, the reason for which is not known.  A14  A15  Figure A4-1  LIPOSOME MOBILITY in pH 6.9 NaCI  C)  a)  0?  0  E 9 E D  > -Q  0  0.0100  Ionic Strength (I)  A16  Figure A4-2  (3  ci)  -1  LIPOSOME MOBILITY in pH 69 NaCI  EL3 glycocalyx thickness (A) 20.7 polymer II between (A) 0.0, 8.0 polymer II #segments/cm2 2.9E+14 inner charge (esu/cm2): variable  0  E  ci)  E  >  4-,  0  0  + + Ionic Strength (I)  A17  Figure A4-3  C-) G) C?  -1  LIPOSOME MOBILITY in pH 6.9 NaCI  EL4 glycocalyx thickness (A) 15.0 polymer I #segments/cm2 2.7E+14 inner charge (esu/cm2): variable  +  +  0  2 9 2D  8000 9000 10700  0 -4.  0.  0.0010  0.0100  Ionic Strength (I)  0. 1000  A18  Figure A4-4  LIPOSOME MOBILITY in pH 69 NaCI  0-  EL5 glycocalyx thickness (A) 20.7 polymer I #segments/cm2 1.2E+14 polymer II between (A) 0.0, 15.0 polymer II #segments/cm2 1.8E+14 outer charge (esu/cm2): variable  () a) 4-,  0  E  -  -2-  5000 5500 6000 6500 7000  (I)  2 D  >  -3-  / 10700  +  0 -4-  1  0.0001  I  I  I  liii  0.0010  Ionic Strength (I)  I  I  I III  0.1000  I  I  I  III  1.0000  A19  LIPOSOME MOBILITY in pH 69 NaCI  Figure A4-5 0-  -  1-  0  EL6 glycocalyx thickness (A) 20.7 polymer I #segments/cm2 0.6E+14 polymer II between (A) 0.0, 15.0 polymer II #segments/cm2 2.2E+14 inner,outer charge (esu/cm2): 4250, 2875  /  E 9 2 :3  %-.-.  >  -3-  -Q  + ——  0.0001  I  111111  0.0010  I  I  111111  0.0100  I  Ionic Strength (I)  I  111111  0.1000  1.0000  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items