Citation
Synthesis and Characterization of Oil Core Silica Shell Nanocapsules for Biomedical Applications

Material Information

Title:
Synthesis and Characterization of Oil Core Silica Shell Nanocapsules for Biomedical Applications
Creator:
JOVANOVIC, ALEKS V ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Animals ( jstor )
Diameters ( jstor )
Eggshells ( jstor )
Hemolysis ( jstor )
Micelles ( jstor )
Molecules ( jstor )
Nanoparticles ( jstor )
Polymers ( jstor )
Surfactants ( jstor )
Toxicity ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Aleks V Jovanovic. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
11/30/2006
Resource Identifier:
496613764 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

SYNTHESIS AND CHARACTERIZATI ON OF OIL CORE SILICA SHELL NANOCAPSULES FOR BIOMEDICAL APPLICATIONS By ALEKSA V JOVANOVIC A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Aleksa V. Jovanovic .

PAGE 3

To Isabella and Ilka-thanks for everything

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my advisor, Dr . Randy Duran, for providing much needed support and guidance over the years that I was under his su pervision. I a ppreciate the patience, thoughtful critique and discussion that helped me as a scientist and as a person. My appreciation goes to my committee members Dr. Ken Wagener, Dr. John Reynolds, Dr. Lisa McElwee-White and Dr. D onn Dennis for taking the time and effort in reviewing the scientific merit of this work. I have been lucky to work with many gr eat people during the time at the University of Florida. My sincere appr eciation goes to Jason A. F lint, a person who opened many secret doors to the wonderful and exciting wo rld of biology. Furthermore, I thank Karen Kelly, a manager of Electron Microscopy (EM) at the Interdisciplinary Center for Biotechnology Research (ICBR), for unbelie vable patience during numerous hours I spent in this facility. I was fortunate to spe nd eight months total at Max Planck Institute for Polymer Research (MPIP) where I met my co-worker, now friend, Dr. Mark Pottek, who showed me how to be persistent and se lf-confident even thousands of miles from home. The Duran research group was and still is a very special group of individuals from many different countries and cultures that helped me during hard times in the lab, Stephen Carino, Royale Underhill, Andrew Skolnik, Jennifer Logan, Jorge Chavez, Thomas Joncheray, Henk Keizer, Jun Zhang a nd Raju Francis. My appreciation also goes

PAGE 5

v to many undergraduate students who worked w ith me at some point in time on various research endeavors. This work would not be a success without the enriching environment provided by the Butler Polymer Laboratory. I am most th ankful to the professors Dr. Randy Duran, Dr. Ken Wagener and Dr. John Reynolds who put great effort to provide us, the students, with state of the art apparatus and facilities. I would like to mention a couple of students from other Butler Laboratory groups like Avni Argun, Christoph Grenier, Ed Lehman, Tim Hopkins and John Sworen whose brillian ce and friendship helped me overcome the most challenging intellectual tasks during my studies. I will be always grateful to our Polymer Floor staff, most notably Ms. Lorraine Williams and Sara Klossner, for help and especially incredible patience in answering my (numerous) questions during the last five years. My special thanks go to Dr. George Butle r who put together the polymer program at the UF and made it recognizable in acad emic and industrial community. I am proud that I completed my studies at the Butler Polymer Laboratory. I also appreciate the invaluable help from Dr. Tim Morey, from the Shands Hospital at UF, Dr. Wolfgang Knoll at MPIP , Dr. Brij Moudgil, Dr. Hasan El-Shall and Dr. Dinesh Shah from the Particle Engineering Research Center (PERC). I am also grateful for the funding from th e PERC over the years, especially Anne Donnely for help and support. Naturally, I would like to express my appr eciation to the College of Liberal Arts and Sciences (CLAS) and especially to th e Department of Chemistry for providing me with this invaluable life experience. I will be always grateful to our Chemistry

PAGE 6

vi Department staff, most notably Ms. Lori Clark, Dr. Jim Deyrup and Dr. Ben Smith for help and guidance especially in the first months upon my arrival to UF. I am genuinely thankful to my parents Ma rija and Vito, my brother Vladislav and his son Andrija, my family and my friends in Serbia who believed in me, even when I was losing faith in myself. Finally I am wholeheartedly grateful to my wife Ilka and dau ghter Isabella for providing me with love and support re gardless on my success or failure. I am grateful to the One Almighty from whom all things come. My sincere apologies go to all those peopl e who believe their names should appear here and deserve my thanks but which I ina dvertently missed. I thank them nonetheless.

PAGE 7

vii TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...............................................................................................................x LIST OF FIGURES...........................................................................................................xi ABSTRACT...................................................................................................................xviii CHAPTER 1 LITERATURE BACKGROUND OF THE PROJECT................................................1 Introduction...................................................................................................................1 Template Synthesis.......................................................................................................1 Oil-In-Water Microemulsion as a Template.................................................................1 Microemulsion Technology...................................................................................2 Surfactants and Surface Thermodynamics............................................................5 Sol-Gel Chemistry......................................................................................................10 Host Guest Chemistry.................................................................................................10 Encapsulation of Guest Molecules......................................................................11 Nano vs Micro............................................................................................................11 Nano-sized Objects..............................................................................................12 Nanotechnology and Medicine...................................................................................13 Biosensors............................................................................................................14 Cell Adhesion...............................................................................................14 Drug Delivery......................................................................................................15 Drug Overdose and Clini cal Mistreatments...............................................................15 Drug Toxicity by Antidepressants-Mechanism of Action...................................16 Strategies Toward Drug Detoxification...............................................................18 2 SYNTHESIS OF CORE SHELL NANOC APSULES USING OIL IN WATER MICROEMULSION AS TEMPLATEs.....................................................................20 Introduction.................................................................................................................20 Mixed Surfactant Systems..........................................................................................20 General Remarks About Microemuls ion (ME) and Nanocapsules (Ncs) Formation................................................................................................................21 Experimental Methods................................................................................................22

PAGE 8

viii Quasi Elastic Light Scattering (QELS)...............................................................22 Transmission Electron Microscopy (TEM).........................................................25 Scanning Electron Microscopy (SEM)................................................................26 Results........................................................................................................................ .27 The Necessity of Mixed Surfactant Systems.......................................................27 Surfactant self-assembly and the solu bilization of oil. High S/O ratio........28 Surfactant self-assembly and the solubilization of oil. Medium Surfactant to Oil ratio.................................................................................................35 Oil templates size-Surfactants concentration influence...............................37 Oil template size-Oil co ncentration influence..............................................39 Silica Shell Formation.........................................................................................41 Control of Silica shell thickness-Role of TMOS and oil template size.......42 Control of Silica shell thickness-Role of OTMS.........................................46 Surfactant Self-Assembly and The Solubi lization of Oil -Low Surfactant to Oil Ratio...........................................................................................................48 Ostwald Ripening.........................................................................................49 Alternative Formulations.....................................................................................52 Formulations with n-Octyltrimethoxysilane................................................53 Formulations with 3-(trimethoxysilyl) methyl methacrylate (MTS)...........55 Silica shell formation for alternative formulations......................................57 Surface Modification of the nanocapsules...................................................57 Conclusions.................................................................................................................59 3 PHYSICO-CHEMICAL AND BIOMEDICAL CHARACTERIZATION OF CORE-SHELL NANOCAPSULES...........................................................................62 Introduction.................................................................................................................62 Core-Shell Systems.............................................................................................63 Experimental Methods................................................................................................63 Zeta Potential.......................................................................................................63 Uv-Vis Spectroscopy...........................................................................................64 Infrared Spectroscopy (Ir)...................................................................................66 High Performance Liquid Chromatography (Hplc)............................................66 Hemolysis............................................................................................................67 Thromboelastography (TEG)..............................................................................68 Qualitative interpretation of a TEG trace.....................................................70 Results........................................................................................................................ .71 Concentration of Nanocapsules...........................................................................71 Surface Properties of Nanocapsules....................................................................75 Aggregation of the ncs.................................................................................75 Surface charge of nanocapsules...................................................................77 Hemolytic properties of ncs.........................................................................79 Thromboelastography (TEG)..............................................................................84 Uptake Profile of Nanocapsules.................................................................................88 High Surfactant to Oil Ratio Samples-Uptake Ability........................................88 Medium Surfactant to Oil Ratio Samples-Uptake Ability..................................93

PAGE 9

ix Mechanism of Amitriptyline (AMI) Uptake.......................................................96 Conclusions...............................................................................................................103 4 DETOXIFICATION ABILIT Y OF NANOCAPSULES.........................................107 Introduction...............................................................................................................107 Isolated Heart.....................................................................................................107 Whole Animal Studies.......................................................................................107 Cardiac Myocytes Deposited on Microelctrode Arrays (MEAs)......................108 Experimental Methods..............................................................................................109 Isolated Heart Experiments..............................................................................109 Whole Animal Experiments..............................................................................110 Cardiac Myocytes on Microelect rode Arrays Experiments..............................111 Cardiac Myocyte Isolati on and Deposition on MEA........................................111 Results.......................................................................................................................113 Toxicity of Nanocapsules..................................................................................113 Preemptive Detoxification Experiments...................................................................117 Toxicity Reversal in Whole Animals................................................................117 Toxicity Reversal in Cardiac Myocytes............................................................119 Influence of AMI on cardiac myocytes......................................................119 Preemptive detoxification studies on MEAs..............................................120 Real Time Detoxi fication Studies.............................................................................125 Conclusions...............................................................................................................127 5 CONCLUSIONS AND FUTURE WORK...............................................................130 Microemulsions and Nanocapsules Synthesis..........................................................130 Microemulsions and Nanocapsules Synthesis-Future Work.............................132 Oil in Ethylene Glycol microemulsion.......................................................134 Double microemulsion technology............................................................135 Additional characterization experiments....................................................137 Biomedical Characterization and Drug Detoxification............................................138 Biomedical Characterization and Dr ug Detoxification-Future Work...............139 Silica nanoparticles with encapsulated 1-acid glycoprotein....................140 Additional Experiments..............................................................................144 APPENDIX A CHEMICAL STRUCTURES OF COMP OUNDS USED IN THIS WORK...........146 B INFRARED (IR) SPECTROSCOPY CHARCTERIZATION OF SURFACE MODIFIED NANOCAPSULES..............................................................................149 C LIST OF ABBREVIATIONS...................................................................................152 LIST OF REFERENCES.................................................................................................154 BIOGRAPHICAL SKETCH...........................................................................................165

PAGE 10

x LIST OF TABLES Table page 2-1 The formulations made with high Su rfactant to oil ratio and formulations without the exte rnal surfactant 1 ..............................................................................29 2-2 The formulations made with medium Surfactant to Oil ratio..................................35 2-3 The formulations made with low Surf actant to Oil ratio. (The oil phases in formulations 41-50 consisted of Ethyl Butyrate 7 and 1-dodecene 5 ).....................48 2-4 The alternative formulations made by using the polymerizable surfactants 16 and 24 .......................................................................................................................53

PAGE 11

xi LIST OF FIGURES Figure page 1-1 Example of possible phases in water, oil and surfactant ternary system, from right normal micelle, hexagonal phase, bi layer, inverted hexagonal phase and inverted micelle........................................................................................................3 1-2 The origin of surface tension in li quids: a) a molecule in the bulk, b) a molecule on the surface and c) su rfactant film on the surface.................................6 1-3 Structures of common surfactants, noni onic (left), anionic (middle) and anionic (right). Numbered compounds were used in this work............................................7 1-4 Various solution properties below and above cmc..................................................8 1-5 Packing parameter and its relation to shapes of aggregates (most of the surfactants used in this study had c pp<1/3, derived from Equation 1-2).................9 1-6 The picture of a human heart, SA (s inoatrial) node and AV (atrioventricular) node are labeled (right), and the repres entation of the QRS interval along with the types of heartbeat (left)....................................................................................17 2-1 Reactive Si-OH groups available af ter the polymerization of OTMS...................21 2-2 A typical shape of the corre lation function used in QELS....................................24 2-3 A picture of PDDLS photon co rrelator used in this study.....................................25 2-4 Hitachi Transmission Micros cope used in this study............................................26 2-5 Hitachi Scanning Electron Micr oscope used in this study.....................................27 2-6 The optical micrographs (upper row) and appearance of the formulation made without the exte rnal surfactant 1 (or 6 ). (Upper creamy oil phase and bottom clear water phase)..................................................................................................28 2-7 The QELS pattern of 1 alone ( ) and microemulsion of 1 and 2 ( )....................30 2-8 The QELS pattern of mixed micelles of 1 and 3 ( ) and ME formed with 1 , 2 and 3 ( ).................................................................................................................30 2-9 The polymerization of OTMS ( 3 )..........................................................................32

PAGE 12

xii 2-10 TEMs of: a) oil templates, b) nanoc apsules (obtain after the reaction with 4 ) and c) the same solution as in (b) after two weeks................................................34 2-11 TEM pictures of formulations ha ving large amount of lecithin ( 8 ), whole solution (top), creamy part of the sa me solution (middle) and microemulsion part of the solution.................................................................................................36 2-12 The effect of various components on the oil template size a) OTMS ( 3 ), b) Ethylbutyrate ( 7 ), c) Tween -80 ( 6 ), d) Lecithine ( 8 ).........................................38 2-13 TEM pictures of “small” (upper image) and “large” (bottom image) oil cores.....40 2-14 The kinetics of silica shell formation, the concentration of 4 was 0.07w%..........42 2-15 Nanocapsule diameter as a function of concentration of 4 for “small” oil core ( ) and “large” oil core ( ) formulations.............................................................43 2-16 TEM images of “small” core at a) 0.07 w% of 4 , b) 0.88 w% of 4 , “large” core at (c) 0.07 w% of 4 and (d) 0.28 w% of 4 ..............................................................45 2-17 The importance of OTMS: formulation ma de with all of the ingredients (left vial), solution with just compound 4 (middle) and formulation with all of the ingredients except 3 ...............................................................................................47 2-18 The influence of 3 on number of crosslinkable templates (left) and shell thickness of corresponding sample s (right) measured by QELS...........................47 2-19 The QELS analysis diameter integr ity for PS/O=0.12 (w/w) sample (left) and PS/O=0.23 (w/w) sample (right). ( ) no polymerization of 3 , ( ) after the polymerization of 3 ................................................................................................50 2-20 The QELS analysis of low S/O samples made with 7 alone ( ) S/O=4 and ( ) S/O=2 (left). Analysis of formulatio ns with overall S/O=4 and increasing concentration of non-polar oil 5 : ( ) 20w% of the total o il phase is non-polar oil, ( ) 40w%, (X) 80w% and ( ) 60w%. (Right).................................................50 2-21 The influence of surfactant size on stability of low S/O samples (S/O=4, 5 is 60w% of oil phase): ( ) Triton X-100 ( 23 , MW=648g/mol), ( ) Brij 97 ( 29 , MW=709g/mol) and (X) Tween-80 ( 6 , MW=1309g/mol)....................................51 2-22 SEMs of the formulation with S/O=1 ( upper row), oil templates (left) and after crosslinking with 4 (middle and right). TEMs of formulation with S/O=2.2 taken at different magnifications (bottom row).....................................................52 2-23 The stability of formulation ma de with Pluronic surfactant F-127 ( 17 ) and polymerizable surfactant 16 at S/O=3 ( ) and S/O=5 ( ).....................................54 2-24 TEMs of oil cores prepared using surfactants 16 , 6 and 27 (S/O=14)...................55

PAGE 13

xiii 2-25 TEMs of formulations made with polymerizable surfactant 24 and 6 (left) and 23 (right) as the ex ternal surfactant.......................................................................56 2-26 TEMs of formulations made with polymerizable surfactant 24 and 17 (left) and cationic surfactant 28 along with 6 (right) as the external surfactants..................56 2-27 TEMs of alternative formulation after the crosslinking reaction with 4 , at different magnifications.........................................................................................57 2-28 TEMs of nanocapsules with thin, medi um and thick silica shell (from left) prior to PEO modification (upper row) and after the PEO modification (bottom row)........................................................................................................................59 3-1 The standard curves for Tween-80 (top) and Brij-97 (bottom) obtained by measuring UV absorbance at max=233nm............................................................65 3-2 A Typical TEG Trace and important parameters...................................................68 3-3 The library of traces in TEG. Take n from Dennis, D.M. and T.E.Morey, University of Florida, Shands Hosp ital, Dept. of Anesthesiology, Private communication.......................................................................................................69 3-4 The relationship between the scattering intensity and particle concentration for polystyrene standards obtained from Duke Corporation, 100nm (top) and 20nm (bottom)..................................................................................................................71 3-5 The relationship between the scattering intensity and particle concentration for 100nm nanocapsules..............................................................................................72 3-6 The percent scatteri ng intensity of 20nm ( ) and 100nm Duke standards ( ) as a function of their weight ratio...............................................................................72 3-7 The diameter ratio parameter ( x from Equation 3-4) for 20nm and 100nm Duke standards as a function of their weight ratio..........................................................73 3-8 An example of QELS measurement fo r nanocapsules after the synthesis (top), and after the purification (bottom).........................................................................74 3-9 The change of nanocapsule samples diameter with time at pH=3 ( ), pH=7.4 ( ) and pH=11 ( ) for: a) silica surface, b) PE O surface, c) amino surface and d) acid surface........................................................................................................76 3-10 The aggregation of the amino (vial on the left) and acid (vial on the right) modified nanocapsules upon mixi ng (vial in the middle)......................................77 3-11 The ZETA potential curves for ncs with different surface ( ) silica surface, (o) PEO surface and ( ) PEO and succinic acid surface.............................................78

PAGE 14

xiv 3-12 Zeta Potential curves for amino modified () and succinic acid modified ( ) ncs..........................................................................................................................78 3-13 The hemolysis activity of Tween-80 ( ) and Tween-80/Ethyl Butyrate microemulsion ( ) at 0.1w%.................................................................................80 3-14 Hemolysis of RBCs in the presence of ncs having different surface at 0.1w% oil content...............................................................................................................82 3-15 The hemolysis of amino modified ncs ( ) and silica ncs ( ) for medium shell (top) and thick shell (bottom)................................................................................83 3-16 Maximum Amplitude (top left) and ktime (top right) for Tween-80 solutions....85 3-17 The MA for ncs with silica and PEO-modi fied surface a) thin shell, b) medium shell, c) thick shell and d) comparison of MA for ncs of different size at 0.1w%, silica surface and PEO-modified ( lines represents statistical trends).......86 3-18 The r-time for silica ( ) and PEO-modified ( )ncs: thin shell (top), medium shell (middle) and th ick shell (bottom)..................................................................88 3-19 Removal of quinoline from saline solution............................................................90 3-20 Removal of varying concentrations of bupivacaine as determined by HPLC.......92 3-21 The uptake efficiency of oil templa tes formed by changing the concentration of: a) compound 3 (plotted as w% of 3 vs. decrease in absorbance for clarity); b) compound 7 ; c) compound 6 (inset represents the decrease in absorbance for solutions containing just 6 ); d) compound 8 ..........................................................94 3-22 The uptake efficiency as a function of shell thickness according to Eq. 10 obtained by QELS (values for dtot and do are taken from Figure 2-15).................96 3-23 The Amitriptyline removal using Tween-80 micelles measured by HPLC...........97 3-24 The Amitriptyline removal comparison for 2w% ncs (white bars) and 1w% ncs and 1w% of 6 (grey bars).......................................................................................98 3-25 The removal of Amitriptyline measured for ncs of different shell thickness (same number of ncs) (top), the rem oval of Amitriptyline plotted versus relative surface area for the three populations of ncs (bottom).............................99 3-26 The uptake of Amitriptyline by silica (white bars) and PEO (grey bars) ncs......100 3-27 The difference in uptake ability for PEO modified (white bars) and mixed (PEO and COOH) modified ncs (grey bars)........................................................101

PAGE 15

xv 3-28 The spike rate of cardiac muscle cells in the presence of AMI alone ( ), AMI and PEO modified ncs ( ) and AMI and PEO/COOH modified ncs ( ).............102 3-29 The toxic drug uptake in human plasma measured by HPLC: control and three most efficient ncs samples (top) and control and Tween-80 at two different concentrations (bottom).......................................................................................103 4-1 The isolated heart set up......................................................................................109 4-2 The fluorescent micrograph of cardiac myocytes (blue-nuclei labeled by 4',6Diamidino-2-phenylindole (DAPI)) (left) and optical micrograph of cardiac myocytes deposited onto MEAs..........................................................................112 4-3 The principle experiment al idea, 1) cells are deposited on fibronectin coated MEAs, 2) the activity of cells is mon itored, 3) toxic drug is added and 4) nanocapsules addition with continuation of a measurement. Picture of the MEA device (lower right)....................................................................................113 4-4 The influence of high S/O ratio na nocaspules on QRS of a whole animal, normal QRS ( ) and QRS with ncs ( )...............................................................114 4-5 The influence of ncs on isolated heart parameters, QRS ( ), SA ( ) and SV ( )........................................................................................................................115 4-6 The influence of nanocapsules at 1w% on QRS in vivo , control ( ) and with ncs ( )..................................................................................................................116 4-7 The spike rate of cardiac myocytes in the presence of Tween-80 at 0.025 w% (left) and PEO and COOH modi fied nanocapsules (right)..................................117 4-8 The therapeutic action of nanocpasules, control ( ), AMI alone ( ) and AMI and ncs together ( ).............................................................................................118 4-9 The therapeutic action of dilu ted nanocapsules (0.5w%), control ( ), AMI alone ( ) and AMI and ncs together ( )..............................................................119 4-10 The Amitriptyline dose-response curve with respect to spike frequency of cardiac myocytes. (data acquired 15 min. after the drug addition)......................120 4-11 The preemptive detoxification of cardiac myocytes in the presence of nanocapsules, AMI was at 1 M: spike rate (uppe r) and spike amplitude (lower)..................................................................................................................121 4-12 The preemptive experiment with 2 M Amitriptyline: spike rate (left) and spike amplitude (ri ght)., control ( ), AMI alone ( ) and AMI and ncs together ( ). (* significance test, P<0.05).........................................................................123

PAGE 16

xvi 4-13 The preemptive experiment with 5 M Amitriptyline: spike rate (left) and spike amplitude (ri ght), control ( ), AMI alone ( ) and AMI and ncs together ( )........................................................................................................................124 4-14 The real time detoxifica tion of cardiac myocytes by nanocapsules, spike rate (top) and spike amplitude (bottom). The control conditions for spike amplitude are omitted for clarity...........................................................................................126 5-1 Left-The competition between the self condensation (outcome (b) red) and condensation on the droplet surface (desired process (a) blue). Right from top: sol-gel rods, unidentified high aspect ratio material and solid silica nanoparticles........................................................................................................133 5-2 The decrease in amount of water ne cessary to obtain clear microemulsion (ME) with increase of lipoph ilicity of the oil phase (t op left), the decrease in the amount of water to obtain clear ME with increase in HLB of surfactant mixture (bottom left) and TEMs of the core-shell nanocapsules using O/EG microemulsion as a template (left).......................................................................134 5-3 The scheme of the double microemulsi on (ME) process: (from left) 1) O/W ME is formed, 2) double O/W/O ME is formed, 3) sol-gel formation of the shell and 4) pyrolysis at elevated temperature to obtain hollow nanocapsules...135 5-4 The core-shell nanocapsules synthesi zed using double microemulsion method (top row) and hollow nanocapsules obtaine d after Thermogravimetric analysis (TGA). (a) particles at 100 nm, (b-d) particles at 70nm, (e) 70 nm particles after TGA, (f) 100 nm partic le after TGA (the inner part was 30 nm in each case).....................................................................................................................136 5-5 The scheme of silica nanoparticle s with AAG protein formation using W/O method..................................................................................................................140 5-6 The silica nanoparticles with em bedded AAG of different size..........................141 5-7 The morphologies of nanoparticles obtained with low (upper row) and high concentration of 3-aminopropyltr imethoxysilane (bottom row)..........................142 5-8 The IR spectra of (from top) neat AAG, bare Si nps and AAG@ nps................143 5-9 The comparison of uptake efficiency of nanoparticles with and without AAG (left), and two populations of nanoparticle s with different concentrations of AAG (right)..........................................................................................................144 A-1 The chemical structures of compounds used as surfactants and co-surfactants in this work..........................................................................................................146 A-2 The chemical structures of compounds used as oi l phases and model drugs for the uptake studies.................................................................................................147

PAGE 17

xvii A-3 The chemical structures of crosslinka ble surfactants, solgel agent and surface modifiers..............................................................................................................147 A-4 The scheme of nanocapsules synthe sis (medium Surfact ant-to-Oil ratio)...........148 B-1 The IR spectra of silica a nd PEO modified nanocapsules...................................149 B-2 The IR spectra of PEO and COOH modified (upper) and COOH alone silica nanocapsules........................................................................................................150 B-3 The IR spectrum of amino modified silica nanocapsules....................................151

PAGE 18

xviii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SYNTHESIS AND CHARACTERIZATI ON OF OIL CORE SILICA SHELL NANOCAPSULES FOR BIOMEDICAL APPLICATIONS By Aleksa V Jovanovic May 2006 Chair: Randolph S. Duran Major Department: Chemistry The possibility of synthesizing water-sol uble, biocompatible and shape persistent nanoparticles has been recently in the focus of both academia and industry. Aklylalkoxysilanes are able to form monolayer s in 2D at the air/water interface. This study aims to expand this chemistry to 3D using the Oil-in-Water (O/W) microemulsion as a template that contains polymerizab le surfactant headgroups available for polymerization at the oil/water interface. The goal is to synthesize the system that would be able to sequester the toxic drug in cas e of overdose or clinical mistreatment. The synthetic part of the project showed that it is possible to obtain stable crosslinkable microemulsion droplets using the aklyalkoxysilanes. The silica shell was built around these droplets to further improve th e stability of the system. The control over the size was demonstrated for both the oil co re templates and the silica shell thickness. The modification of the nanocapsules surface le d to particles of superior properties.

PAGE 19

xix Hemolysis and Thromboelastography (TEG) were the two simple tests used to assess the biomedical profile of the nanocapsules and relate these data to the particle surface structure. It was found that the polymer modified particles are more stable and biocompatible. Studies of the uptake poten tial showed the importance of surface characteristics as well. The detoxification potential was clearly dem onstrated in cardiac muscle cells plated on the Microelectrode arrays (MEAs). A g limpse of the therapeutic effect of ncs in vivo was shown through experiment s with living animals. Overall, the study showed the feasibility of the templated microemulsion approach toward the formation of the functional nano-ob jects. Considering the results from all of the characterization experiments, the nano capsule surface is the key feature that determines the overall performance of the particle.

PAGE 20

1 CHAPTER 1 LITERATURE BACKGROU ND OF THE PROJECT Introduction The motivation for this study was the investig ation of possibilities to synthesize oil core and silica shell nanocapsules (ncs). Th e physico-chemical and biomedical properties of these particulates have been thoroughly i nvestigated. The primary goal was to create a system for drug detoxification therapy in cases of accidental or intentional drug overdose. Template Synthesis A beginning of the synthesis of materials with properties predefined by presence of a specific adsorbent (i.e. “template”) was in the late 1940’s by Dickey,1 who used organic dyes as templates for formation of porous si lica gels. A template can be defined as a central structure around which a network is fo rmed, if the template is subsequently removed, a cavity that resembles mor phologically the template is formed.2 During the last couple of decades a variety of organic and inorganic material s were synthesized in this fashion. The most representative exampl es include carbon, silica and gold nanotubes,3-5 nanostructures and nanoparticles.6 Based on interactions involv ed in the procedure, the template synthesis can be divided in two gr oups: a) covalent and b) non-covalent (ionic, interaction, hydrogen bonding) synthesis.7 The former is further described in Chapter 2. Oil-In-Water Microemulsion as a Template Oil-in-water (O/W) microemulsions are tr ansparent, optically isotropic solutions, which are thermodynamically stable with oil droplet diameters less than 250nm,

PAGE 21

2 dispersed in a continuous domain of water. 8Self-assembled block copolymers, surfactants or microemulsion droplets are the most commonly used templates for the formation of mesostructured and nanostructured silica materials.9 The advantages of using microemulsion droplets as templates ar e a) easy preparation by mixing oil, water and surfactants and b) synthesis of mes oporous materials without further processing. Furthermore, the size of the template, i.e. microemulsion droplet, can be controlled by changing the oil/surfactant ra tio, leading to templates of desired size. A detailed description of this proc ess is in Chapter 2. Microemulsion Technology Microemulsions (ME) are thermodynamically stable homogeneous mixtures of two immiscible liquids (e.g., “water” and “oil”), stabilized by a surfact ant or a mixture of surfactants (surfactant and co-surfactants). The role of surfactant is to lower the interfacial tension between oil and water to ultra-low values, allo wing the dispersion of two liquids that normally do not mix. Microe mulsions (ME) can be oil-in-water ME (O/W) when water is a continuous phase or water-in-oil ME (W/O), when oil is a continuous phase. At low surfact ant concentrations ME phase can be in equilibrium with a pure continuous phase. Such phases have been characterized by the Windsor nomenclature: a) O/W ME with oil phase (Windsor I), W/O ME with water phase (Windsor II) and O/W and W/O ME separa ted by water/oil (3 phases) Windsor III.10 An isotropic solution that will be described in this work is often referred to as a Winsor IV system, i.e. only spherical droplets of the dispersed phase exist.11 Depending on the relative concentrations of su rfactants, oil and water, vari ous morphologies other than spherical droplets can also be obtained (Figure 1-1).12

PAGE 22

3 Figure 1-1. Example of possible phases in wate r, oil and surfactant ternary system, from right normal micelle, hexagonal phase, bi layer, inverted hexagonal phase and inverted micelle (image taken from http://www.nonequilibrium.com/complexfluids.htm ) Many applications of microemulsions stem from the fact that two immiscible liquids form a homogeneous mixture, with disc rete nanometer sized domains of dispersed liquid. Historically, one of the first large-scale uses of MEs was enchanced oil recovery,13 the idea was to “mobilize” the oil droplets tr apped in the rock pores by injection of solution that has ultra-low inte rfacial tension toward oil, t hus inducing the solubilization of oil. The oil droplets are collected from an oil bank and oil can be easily recovered. Another important industrial process is a liquidliquid extraction of meta ls from ores with low metal content. Briefly, the ore is solubi lized in extremely acidic or basic aqueous solutions, metal ions are then extracted with water-insoluble organic molecules by complexation and the organic material is th en solubilized in th e oil (i.e., lipophilic) domains of ME,14 while metal ions remain in the wa ter phase. There are also a number of other applications of ME th at are variants of these two major processes, such as washing,15 contaminated soil extraction16 and lubrication.17 Moreover, other applications relevant for this study include the usage of ME in cosmetics, medicine, separation science and catalysis. The interests of personal care industry in the application of ME (droplet size <250nm) technology are due to the following reasons:18

PAGE 23

4 1. small droplet size reduces gravity fo rces, no creaming (i.e. instability), sedimentation and flocculation will occur 2. penetration through skin is eased due to sm all size, leading to enhanced delivery of active compounds 3. transparency and fluidity provide a pleas ant aesthetic character and skin feel 4. potential usage as delivery vehicle for fr agrances and for alcohol free formulations MEs are also used for the delivery of drugs that are poorly soluble (i.e. “lipophilic”) in water. The ME droplets can solubilize the drug in their lipophilic interior, and due to their size transpor t it throughout the organism.19Good thermodynamic stability, ease of preparation and previously mentioned enhanced transdermal penetration make MEs among the most important vehicl es in the drug deli very industry today.20 Another important feature of ME droplets in th e drug delivery field is that they can serve as templates for synthesis of Solid-Lipid-Nanoparticles (“SLN’s”),21 Poly (alkylcyanoacrylate)22 nanoparticles and the formation of MEs bearing genetic material (i.e. “genetic vaccines”).23, 24 The genetic vaccines are considered safer for the patient since just a part of a virus is injected in the body, whereas classical methods include injection of whole-killed organisms. Recently, interesting applica tions of ME in analytical techniques such as High Performance Liquid Chromatography (HPLC) and Capillary Electrophoresis (CE) have been reported.25 The basis for the development of such techniques is that ME are nano dispersed chromatographic objects where part itioning occurs based on the solubility of the analyte in the oil or water. Ionic su rfactants are usually used in ME based Electrokinetic Capillary Ch romatography (MEEKC), a supe rior separation technique combining the aforementioned features of microe mulsions with the fact that the oil/water interface is charged. The separati on is two fold: a) neutral mo lecules are captured into oil

PAGE 24

5 reservoirs based on their hydrophobicity and b) charged molecules will be collected on the charged interface based on electrostatic interac tion with the interface.26 Finally, MEs are used to facilitate chemi cal reactions where certain components are immiscible with one another. The advant age of ME systems over classical ones, including e.g. mixture of two immiscible solvents or phase transfer catalysts, is the exceptionally large interfacial area, i.e. the contact area between the dispersed phase and the medium.27 For example, an elegant study by Erlington and Menger28 who showed that the substitution or oxidation of 2-chloroethyl et hyl sulfide (”half-mustard”) is accelerated two orders of magnitude when done in oil-in -water ME compared to two phase or phase transfer agent system. The water-insoluble reag ent was dispersed in th e ME droplets, thus becoming more available for the reaction with water-soluble reagent. Surfactants and Surface Thermodynamics Surfactants (or “amphiphiles”) are molecule s that have both, polar (“head”) and non-polar (“tail”) parts. Their primary role is to dramatical ly decrease the surface tension of the dispersed medium. Surface tension arises from the fact that resulting intermolecular forces are different for molecu les in the bulk and on the surface. Consider water molecules in Figure 1-2, each molecule in the bulk can form up to four hydrogen bonds, while surface molecules have fewer neighbors and hence a fewer number of hydrogen bonds. Consequently, water tries to minimize the surface to compensate for the loss of hydrogen bonding, which is a ma nifestation of surface tension ( ). A more general picture of surface tension is given in Figure 1-2. Surfactant can adsorb on the interface by placing the non-polar part in th e non-polar phase (e.g., oil or air) and polar part in polar

PAGE 25

6 phase (e.g. water). Thus it lowers the surf ace tension, because now all of the water molecules are energetically equal. The d ecrease in surface tension is given by: = 0 Equation (1-1) where is the surface pressure, 0 is the surface tension of pure water, and is the surface tension in the pres ence of the surfactant. Surfactant Monolayer Figure 1-2. The origin of surface tension in liquids: a) a molecule in the bulk, b) a molecule on the surface and c) su rfactant film on the surface. Thus a surfactant film is formed on the interface, one molecule thick, called a Langmuir monolayer . The most important consequence of thin film formation at the interface is that the su rface tension is greatly reduced, si nce all of the molecules in each phase are essentially the same, i.e., in th e bulk. Depending on the nature of the polar head, surfactants are divided into 3 groups : a) anionic, b) cationic and c) nonionic surfactants. Some common exam ples of all three groups are given in Figure 1-3. Once the film is formed, any excess of su rfactant molecules is organized into ordered structures called micelles , and the threshold concentration at which micelles are formed is called the critical micelle concentration (cmc).

PAGE 26

7 O O O * O O * * n n n O O n O OH10 O OH10 O OH20 O OH20 O S O ONa O N+BrN+Br-1 Brij-98 23 Triton X-100 6 Tween-80 19 Sodium Octanoate 27 Sodium Dodecylsulphate(SDS) 28 DodecyltrimethylammoniumBromide (DTAB) CetyltrimethylammoniumBromide (CTAB) ONa O Figure 1-3. Structures of co mmon surfactants, nonionic (lef t), anionic (middle) and anionic (right). Numbered compounds were used in this work. Formation of micelles is a phase equilib rium rather than chemical reaction equilibrium. The properties of surfactant solutions above and below the cmc are drastically different as can be measured by various techniques like surface tension measurements, solution conduc tivity, turbidity an d osmotic pressure measurements (Figure 1-4.). Micelles can be viewed as spheres in which polar headroups are at the surface and hydrophobic tails are buried deep in the interior of the micelle. Although the most common representation of surfactant self-assembly, micelles are not the only structure formed in surfactant solution. De pending on molecular structure, concentrations and other parameters such as temperature, su rfactants can be organi zed into spherical or cylindrical micelles, flexible bilayers, vesicles, planar bilayers, inverted micelles and a

PAGE 27

8 variety of other mesophases (some are show n on Figure 1-5) which are beyond the scope of this study. Figure 1-4. Various solution prop erties below and above cmc. The shape is a function of structure para meters that include headgroup area (ah), hydrophobic chain length (lc) and volume of the tail (vt), all of which are related by semiempirical equation = vt / (ah*lc) Equation (1-2) by Israelachvili,29 where is critical packing parameter (cpp) (Figure 1-5). This study focuses on aqueous surfactant so lutions, especially in regimes where spherical assemblies are expected, therefore this paragraph describes in detail the behavior of surfactant molecules in water. Organized surfactant structures are very dynamic in nature, the exchange of surfactant molecules from the micelle and the bulk is

PAGE 28

9 in the nanosecond range, whereas the life time of a micelle is in the order of 10-4 to 10-1s.30-32 Furthermore, the structure of or ganized surfactant molecules and the position of equilibrium are functions of soluti on temperature, ionic strength and pH. All of these variables are changing the structural parameters described in Equation 1-2. For example, increasing the salt concentration lead s to increased ion bindi ng at the surface of the micelle, which expels water molecules t hus making the headgroup area of the ionic surfactant essentially smaller. CPPCritical packing shapeStructure formed<1/3 1/3-1/2 1/2-1 ~1 >1 Spherical micelle Cylindrical micelle Flexible bilayers/vesicles Planar bilayers Inverted (reverse) micelles Surfactant structure Figure 1-5. Packing parameter and its relation to shapes of aggregates (most of the surfactants used in this study had cpp<1/3, derived from Equation 1-2). Furthermore, the addition of salt tends to amplify the difference between the polar and non-polar part of the surfact ant leading to a decrease in cmc. However, the influence of temperature on non-ionic surfactant behavior in water was extensively used in this study for the formulation of ME. The drivi ng force for micellizati on is a tendency of

PAGE 29

10 surfactant non-polar “tails” to minimize the cont act with water. As temperature increases, hydrogen bonding between surfactant polar “hea ds” and water becomes less energetically favorable (water-water hydrogen bonding more preferable), and consequently, decreases the water solubility of surfactant. At a certain temperature, which is called Cloud Point (c.p.), surfactant becomes esse ntially insoluble in water.33 In the case of a three component system, water, surfactant and oil, at the c.p. a phase inversion occurs, where an o/w ME, becomes a w/o ME in equilibrium with excess water. Furthermore, the solubility of oil in the surf actant phase increases profoundly34 at c.p. temperature, which will be widely used in this work for the formation of o/w ME. Sol-Gel Chemistry Initially, sol-gel or inorganic polymerizati on reactions were used for synthesis of ceramic materials and glasses at low temper atures. Recently, organics were employed in combination with sol-gel networks to produce hybrid materials with superior properties. There are numerous possible ap plications of such material s: a) electrical and NLO materials, b) reinforcements of plastics, c) catalysts and porous supports, d) chemical and biochemical sensors, and others.35 In this work, sol-gel chemistry will be employed to build the shell around the microemulsion template s, and therefore to impart the stability of the former ones. The size (“thickness”) of this sol-gel shell is mainly a function of concentration of sol-gel precursors, as described in Chapter 2. Host Guest Chemistry The controlled and specific interaction between a large molecular assembly (i.e. “host”) and a small molecule (i.e. “guest”) b ecame a powerful tool in modern chemistry, mainly because the resulting structures po ssess unique properties compared to starting materials. The interaction can be achieved through hydrogen bonding,36, 37 ionic,38, 39

PAGE 30

11 40 and hydrophobic interactions.41 A part of this study (Cha pter 3.) will focus on the interactions of nanocapsules (ncs) with small target molecules (i.e. toxic drugs), and ways to improve this interaction through modifica tion of size, architect ure and surface of the ncs. Encapsulation of Guest Molecules A broad definition of “encapsulation” is th e situation where the guest molecule is partially or totally surrounded by the host. Recently, molecular assemblies like molecular capsules42 and various supramolecular structures43 have shown great potential for the encapsulation of guest molecules. Nanometer-s ized capsules present one elegant strategy to confront the drug toxicity issue. The oil filled interior of th ese capsules provides an ideal environment for the encapsulated hydropho bic drug molecules. Interactions at the molecular level between ncs and guest drug mo lecules, along with factors influencing the efficiency of this interaction in media of different complexity (from normal saline to whole blood) will be described in detail in Chapter. 3. Nano vs Micro The first and the most obvious difference between micro and na no sized objects is the area exposed to the environment. The frac tion of the particulate material exposed to the environment increases with the decrease in particle diameter.44 The most important consequence of such relationship is that the contact area with the environment is larger for smaller (i.e. “nano”) than for larger (i.e. “micro”) particles. Morey et.al.45 showed that nanoparticles were much more efficient to ward drug detoxification than their micro counterparts, due to the increase in the cont act area between the drug and nanoparticles. Furthermore, particles of smaller size ha ve a different biodistribution within the body.46

PAGE 31

12 Our intention is to use nanocapsules as the dr ug scavengers in the bl oodstream, since they are much smaller than the diameter of the smallest (~5 m) capillaries in the body.47 Nano-sized Objects Due to their small size and high surfaceto volume ratio, materials with nano dimensions find application in many areas: electronics,48-50 medicine,51-53 cosmetics, and catalysis54 are among the most striking areas wher e miniaturization led to significant advances and breakthroughs. Metal nanoparticles have uni que electrical, optical and photophysical features due to their size. The quantum size effect happe ns when the dimension of nanoparticles is similar as de Broglie wavelength of valence electrons. As a conseque nce, there is a gap between the conduction and valence band in nan oparticles, similar to semiconductors and unlike bulk metals.55 Single electron transitions can be observed if the Ee = e2 /2C >kT Equation (1-3) where Ee is the electric field, e is the electron charge, C is the nanoparticle capacitance (very small due to the size of nanoparticle surface) and kT represents the thermal energy. This property of nanoparticles is of great interest in elec tronics technology.56 The application of nps in catalysis stem s from their high surface to volume ratio where they were used either as free57, 58 or supported particles.59, 60 Briefly, the keys for successful usage of nps in catalysis are the electronic and geometric effects. The former are responsible for enhanced availability of valence electrons for surface bond formation by drop in work function with increasing surface curvature61 and contraction of lattice constants which shifts d-el ectrons to higher energies.55 The latter is responsible for features like average coordination number of the surface atoms, abundance of defects and

PAGE 32

13 the concentration of active sites.62 One example of nps catalysis with high expectations in the future is the application in fuel cells.63 The nps most often used in fuel cells are comprised from precious metals. Therefore technology that offers low content (i.e. nanoparticles), yet efficient catalysts is im portant due to economic reasons. The aforementioned breakthroughs in na notechnology may have a great impact on wastewater and air purificat ion technology. One strategy for waste water treatment with nano-objects involves the conve rsion of toxic compounds pres ent in environment by the catalytic action of nanoparticles per se.64, 65 The other way is to encapsulate the reactive species that can transform the toxicants inside the nanoobjects.66 The third possible strategy, conceptually more sim ilar to this work (Chapter 3) is the encapsulation of the hydrophobic compounds in the hydro phobic interior of nanopart icles (i.e. “extraction dots”).67 Nanotechnology and Medicine Nanotechnology represents study, creation and manipulation of objects with dimensions smaller than 100nm.68 Besides the uniqueness of physical properties described earlier, such small objects are adva ntageous from a biological prospective as well. The scientific crossover between nanotechnology and medicine attracted great attention, due to advances former provi ded for improvement of health care.69 Miniaturization provides cost effective and more rapidl y functioning chemical and biological components. Moreover, some nanometer-sized obj ects also possess remarkable self-ordering and assembly patterns70 that make them the ideal systems for biosensors, drug delivery carriers, biorecognition entitie s, cell adhesion promoters and diagnostic tools. These unique behaviors are what make the application of nanotechnology in

PAGE 33

14 medicine possible, and better understanding of these processes, will increase the quality of human life. Biosensors Nanoparticles (nps) are emerging platform s for the development of biosensors since their surface can be modified with desired biosensing molecules.71-73 Commonly, a fluroscent dye (“marker”) is embedded inside the nps, and a chemical recognition agent is attached to the surface,74, 75,76 Consequently, sensing events are detected by tracking the changes in optical, magnetic and fluorescent pr operties of nps. Nanopa rticles prepared in this way can be used for the detection and analysis of DNA, enzyme catalyzed transformations and imaging of cancer cells. An interesting method for DNA detection is so-called molecular beacon assembly. A si ngle strand of DNA bears a nanoparticle on one end, a dye on other end and fluorescence is quenched by base pairing. Once the np and a dye are spatially separated by hybr idization with analyte DNA strand, the fluorescence is generated.77 The semiconductor core-shell nps (core CdSe, shell ZnS) were used to follow the dynamics of DNA rep lication and telomerizati on of cancer cells. Dye molecules were incorporated into newly formed DNA replica or telomers, to act as energy acceptors for fluorescence resonance energy transfer (FRET). The dynamics is monitored by UV-VIS spectra of nps that is a function of what process (replication or telomerization) occurred.78 Cell Adhesion A second important area of nps applica tion in biology is their usage as cell adhesion promoters. Cell adhesion is an impor tant event in migration, differentiation and communication of cells. The surface of nps can be func tionalized with various carbohydrates enabling the study of adhesion energetics.79 Furthermore, diagnostically

PAGE 34

15 relevant studies have been conducted using mo dified gold nps for adhesion and viability of cancer cells.80 A potential new strategy for breast cancer treatment was examined using thiamine-ligand modified gadolinium (Gd) nps. The idea is that cancer cells will specifically adhere to modifi ed nps by the over expression of the thiamine, and these in turn, will deliver the drug into cancer cells.81 Significant advances have been made in the field of artificial tissue engineering with the developments in nanotechnology.82 Polymeric materials, alone or in combinati on with nps, have excellent properties for cell adhesion and proliferation (i.e. “multiplying”) th at are important steps in the formation of an artificial tissue. 83-85 Drug Delivery Drug delivery systems that can target dr ugs to specific body sites or precisely control drug release have been a subject of intense interest in both, academia and industry. According to Langer et. al.,86, 87 three principle advantag es of advanced drug delivery systems compared to conventional pha rmaceuticals are: (a) a reduction in the toxicity of drugs (“drug targeting “), (b) an increase in the absorp tion of drugs, (c) the presence of a release profile. This study has exactly the opposite aim, that is, formation of devices to lower the concen tration of the drug already present in the body. However, these ncs can also be an inte resting starting point for drug de livery studies, provided that the drug is previously dissolved inside the oil template. Experiments and characterizations of nanocapsules synthesized in this work will be described in Chapter 3. Drug Overdose and Clinical Mistreatments A 2000-query showed, in the United States , more than 300,000 patients/year are admitted to emergency rooms for drug-related complications. Worldwide, there are more

PAGE 35

16 than 3 million insecticide poisonings/year resulting in 220,000 deaths.88 Tricyclic antidepressants are identified as the most frequently ingested substances in self poisonings together with paraceta mol, benzodiazepines and alcohol.89 Moreover, in the 1990’s, antidepressants were second after th e analgesics as the drugs most commonly taken in fatal drug overdose.90, 91 Finally, the rate of fatal drug poisonings is higher for tricyclic antidepressants than for any other antidepressants.92 The most common drugs used in suicides are Dothiepin and Amitriptyline (AMI), which are the most toxic tricyclics.93 Apart from intentional drug overdoses, desc ribed above, an important issue is the clinical mistreatment, especially in the ar ea of anesthesiology. Recent data from the American Society of Anesthesiology shows th at intravenous local anesthetic injections are major cause of cardiac bloc ks that had a fatal outcome.94 Drug Toxicity by Antidepressants-Mechanism of Action The toxicity of tricyclic antidepressants arises from four main pharmacological properties: 1) inhibi tion of norepinephrine reuptake at nerve terminals, 2) direct adrenergic block, 3) a membrane stabilizing effect on the myocardium and 4) anticholinergic action. The clinic al features of tricyclic ove rdose are categorized based on their effects on peripheral au tonomic system, the central nervous system and the cardiovascular system.95 The focus of this work is the investigation of the cardiovascular system in situations of drug overdose, so these effects will be explained in somewhat greater detail. The most commonly observed cardiovascular e ffect is a sinus tachycardia, which is attributable to the inhibition of norepinephr ine reuptake. However, the most important

PAGE 36

17 toxic effect of tricyclics is the slowing of depolarization of the cardiac action potential by the inhibition of the sodium current that cau ses a delay in propaga tion of depolarization through both myocardium and conducting tissue.96 Figure 1-6. The picture of a hu man heart, SA (sinoatrial) node and AV (atrioventricular) node are labeled (right), and the repres entation of the QRS interval along with the types of heartbeat (left). The im age was taken from the Heart Rhythm Society. (www.hrspatients.com) A simpler explanation is that the Atriove ntricular (AV) node receives electrical signals from the Sinoatrial (SA) node, the hear t’s major pacemaker, and transmits them to lower chambers (ventricles) of the heart. Th e electrical signals transmitted from SA to AV trigger the muscle contractions that pum p the blood out of the heart and into the lungs and body. In the presence of high concen trations of toxic drug, the time for the transmittance of the electrical signal is prolonged, consequently the QRS interval is prolonged, that causes the cardiac arrhythmias. In cases of overdose, the inhibition of sodium flux into myocardial cells can cause the depressed contractility with fatal outcome.97-99

PAGE 37

18 Strategies Toward Drug Detoxification In the clinical envir onment, the treatments in cases of anesthetic-i nduced toxicity are supportive measures such as rapid oxygenation, ventilation, seizure control and cardiovascular support.100 In cases of drug overdose, a common antidote is activated carbon, followed by gastric emptying.101 Furthermore, an intravenous injection of Sodium Bicarbonate is another way used in clinical world to fight drug overdose.102, 103 Several drugs have been proposed to serv e as antidotes to cardiac toxicity like Amiodarone, Epinephrine and others, however , with no clear and consistent effect.104 Apart from these traditional me thods, the infusion of insulin,105 propofol106 and lipid infusions107 are different routes to possible drug reversal in cases of toxicity. Despite some promising results showed by the aforementioned methods, an efficient “antidote” is still a subject of many studies. A more “chemical” approach toward this issue has been suggested as a platform of the GOAL IV: Nanoparticulate systems for drug detoxification of the NSF-supported Part icle Engineering Research Center at University of Florida. A multidisciplinary e ffort that includes Departments of Chemistry, Chemical Engineering, Materials Science and Anesthesiology (at Shands Hospital) has proposed several different routes for possible drug reversal in cases of overdose. The common theme of these approaches is th e nano-size of the possible drug scavenging entities. As explained on several occasions in this Chapter, nano objects have a large surface-to-volume ratio, which gives rise to increased probability of toxic drugdetoxifying agent interactions. A surface area effect was clearly demonstrated by Morey et. al.45, showing the superior drug removal in isol ated guinea pig heart compared to lipid macroemulsions. The most promising system to date with respect to drug removal efficiency and biological safety is described by Varshney et. al.108, a self assembled

PAGE 38

19 nanoemulsion made with polymeric (PLURONIC ) and medium alkyl chain anionic surfactants and Ethyl Butyrate as an li pophilic phase. High MW surfactant helped to overcome the common instability of microe mulsion upon dilution and negatively charged surfactant increased the efficacy of drug removal through electrostatic interaction. Furthermore, the authors showed that the interaction between the drug and nanoemulsion was a function of polymeric su rfactant structure (number of EO vs. PO units) as well. A second promising system consists of modified chitosan nanoparticles.109 This author’s idea is to use the interaction between electron rich benzene rings of antidepressants and electron deficient (dinitr o sulfonyl) aromatic rings of modified chitosan. Although a promising set of data was published,110 the system showed poor hydrolytic stability due to hydrolysis of sulfonyl es ter bonds within hours upon solubi lization in water. Another possible system for drug detoxification are poly (lactic-co-glycolide) (PLGA) nanoparticles that have excellent biocom patibile and biodegradable properties.111 Recently, nanotubes have also been suggested as a powerful tool for the separations of biomolecules and pharmaceuticals.112 The results of detoxification experiments using nanocapsules synthesized in this work will be the focus of Chapter 4.

PAGE 39

20 CHAPTER 2 SYNTHESIS OF CORE SHELL NANOC APSULES USING OIL IN WATER MICROEMULSION AS TEMPLATES Introduction This Chapter describes the results of nu merous trials conducted to produce oil-inwater (O/W) microemulsion droplets using vari ous combinations of surfactants and oil phases. The experimental methods used to asses the structure, architecture and concentration of templates/nanocapsules (n cs) will be also described. The surfactant structure, as well as the concentration of oil is used to understand the architecture and size of the templates. The role of external parameters, like temper ature and mixing time, on formation of microemulsions is also ex plained. A second step in the synthesis, building the silica shell around reactive microe mulsion templates is described, along with the factors influencing the shell thickne ss of these ncs. Finally, the siloxane polycondensation chemistry used for the m odification of ncs surface is explained. Mixed Surfactant Systems All of the formulations used for this st udy are systems composed of two or more surfactants. It is therefore n ecessary to introduce the basics of mixed surfactant behavior. Mixed micelles of non-ionic and ionic surfactan ts poses superior properties compared to single surfactant micelles with respect to colloi dal stability and ability to tune the size of aggregates.113 Most of the systems show non ideal mixing behavior114 due to the existence of interactions be tween surfactants, causing the different composition of mixed micelles.115 The extent of non-ideal mixing beha vior depends on the structure of

PAGE 40

21 surfactant, the length of hydrocar bon tail and charge of the po lar head. It is wortwhile to introduce at this point th e surfactant hydrophilic-lipophilic balance (HLB) concept.116 HLB is an arbitrary number, ranging between 1 and 20 that descri bes the solubilization tendency of surfactant towards water or oil. It is believe d that an optimum HLB of surfactant(s) is required for the formation of a stable microemulsion. Optimum in this case means most stable and the maximum solubilization power ME. Recent studies showed that by careful choice of non-ionic surfactant mixtures it is possible to obtain a range of HLB values that provides a plet hora of opportunities fo r ME formulations.117 General Remarks About Microemulsion (ME) and Nanocapsules (Ncs) Formation. The synthesis of ncs is divided in three di stinct steps, first be ing the formation of oil-in-water (O/W) microemulsion droplets, followed by a formation of silica shell around the droplets, and the last (optional) step is a covale nt modification of the silica shell. Figure 2-1. Reactive Si-OH groups availabl e after the polyme rization of OTMS. The uniqueness of this synthetic pathway is the presence of Octadecyltrimethoxysilane (OTMS 3 -see Appendix 1 for stru ctures of all bolded

PAGE 41

22 compounds), a polymerizable surfactant on the oil/water interface. As already shown in this group,118, 119 compound 3 can form a network (thin film ) on the air/water interface, and the condensation of methoxysiloxane gr oups can go up to 50% conversion, meaning that unreacted groups will be available for further crosslinking (Figure 2-1). There are two general ways for preparation of ME successfully used in this study: a) Phase Inversion Temperature (PIT used for the majority of formulations) and b) Phase Inversion method. PIT was introduced in late 1960’s by Shinoda,120 and the key notion is that the nonionic surfactant will become less soluble in water with increasing temperature, due to preferable water-water as oppose to water-E O hydrogen bonding (Hbonding) at higher temperatures. At the cloud point (c.p.), surfactant will be equally soluble in both water and oil, and further increase of temperature wi ll lead to a more efficient mixing of oil and surfactant. Upon cooling, small O/W droplet s (high surface-to-volume ratio) will be formed, due to favorable energetics of water-EO H-bonding. It s hould be noted that microemulsion can also be formed at r oom temperature (RT), however, the process would be dramatically more time consuming (in the order of weeks, as oppose to PIT where the solubilization time ra nge was 15 minutes to 48 hours). The Phase Inversion method uses the same ra tionale as PIT, with the exception that instead of temperature, the sol ubilization promoter is the change121 of ME type, going from W/O to O/W with the sl ow addition of water. Experimental Methods Quasi Elastic Light Scattering (QELS) The basis of this technique is the interact ion of light with matter, when the energy of an incident light beam is used to induce the oscillating dipoles from the electrons of

PAGE 42

23 the object. These dipoles can be regarded as new sources of light, which can be measured under fixed angle. Particles and macromolecu les are in a state of a constant motion (Brownian motion), which imparts for randomness in phase of scattered light, such that light added from two or more particles w ill interfere in constructive or destructive fashion. This means the intensity of scattered light will vary as a f unction of time. QELS measures the time-dependent fluc tuations of scattered light. The fluctuations are quantified by sec ond order correlation function measured by instrument: Equation (2-1) where I(t) is the scattered light at time t , is the delay, that is, the amount that a duplicate intensity trace is shifted from the original before the av eraging is performed. The scheme of the typical correla tion function is given on Figure 2-2. The correlation can be analyzed by: 2 exp2B g Equation (2-2) where B is the baseline of the correla tion function at infinite delay, is the correlation function amplitude at zero delay, and is the decay rate. can be connected to particle diffusion coefficient (D) by: 2q D Equation (2-3) here, q2 is the magnitude of the scat tering vector and is given by: 2 sin 40 0 n q Equation (2-4)

PAGE 43

24 Figure 2-2. A typical shape of the correlation function used in QELS. The rate of the decay of correlation function is related to pa rticle size where 0 is the wavelength of incident beam, n0 is the refractive index of the solvent and is the scattering angle. Finally, D can be related to particle radius by the Stokes-Einstein equation: D kT r6 Equation (2-5) where k is Boltzmann’s constant, T is the temperature, and is the solvent viscosity. A picture of PDDLS photon correlator used in this work is shown on Figure 2-3.

PAGE 44

25 Figure 2-3. A picture of PDDLS photon correlator used in this study. Transmission Electron Microscopy (TEM) This technique enables the imaging of s ub-micron particles at resolution of few nanometers. The apparatus consists of electron source, (usually tungsten wire) electromagnet condenser (lenses) and fluorescen t screen. The voltage used in this study was 50-100 kV, meaning that the theoretica l resolution was <1nm. The electron beam, accelerated by voltage, can be either refract ed from the specimen grid (dark field imaging) or it can pass through the specimen (bright field imaging). In case of bright field imaging, light will be transmitted depending on electron density of the features on the specimen grid. This implies that the grid has to be very thin <0.1mm to be feasible for TEM. The grids used in this study were made of Cu or Ni, covered by a thin film of carbon. Usually, the sample was diluted (<0.1w %) and carefully drop wise placed on the grid.

PAGE 45

26 Figure 2-4. Hitachi Transmission Micr oscope used in this study. Scanning Electron Microscopy (SEM) Similarly to TEM, SEM generates the electron beam in vacuum, which is collimated and focused by a set of electrom agnetic lenses. The difference between these two techniques is that in SE M, electron beams are scanned over the sample surface and excited secondary electrons are used for imaging rather than transmitted primary electrons (TEM). Secondary electrons are de tected by the scintillation materials that create flashes of light from the electrons. The light fl ashes are then amplified by a photomultiplier tube. Correlation of sample scan position with a resulting image will create an image that is very similar to what would have been seen through light microscope. Common prepara tions include placing a drop of sample (at higher concentration than TEM) on soni cated microscope glass slid e and drying in the vacuum. SEM was used in this study to c onfirm both, the size of microemulsion droplet/nanocapsule and the shape of these entities.

PAGE 46

27 Figure 2-5. Hitachi Scanning Electron Microscope used in this study. Results The Necessity of Mixed Surfactant Systems Optical microscopy revealed th at all attempted solutions containing any type of oil and only 3 as the surfactant were turbid due to formation of micron-si zed coarse emulsion droplets, or other phase s, instead of microemulsions (Figure 2-6). Furthermore all of the aforementioned samples were thermodynamically unstable and phase separation occurred within hours, im plying that an additional surfactant has to be added in the formulation. Although many ionic and polymer ic surfactants are able to dissolve oil and form O/W ME, only a non-ionic, me dium size surfactant of CnH2n-1(EO)p type, yielded ME phases with reasonable amount of OTMS as th e second (reactive, temp lating) surfactant. A second requirement for the su rfactant is the presence of a kink in the middle of the alkyl chain, like a cis double bond to inhibit the form ation of the lamellar phases introduced in Chapter 1. Oleic acid, C18H33 (9-Octadecenoic (Z) acid)) with its C18 is the fatty alkyl chain (CnH2n-1) most often used in this study, while EO, the ethyleneoxy group, formed the hydrophilic part of the surfactant.

PAGE 47

28 Figure 2-6. The optical microgra phs (upper row) and appearance of the formulation made without the exte rnal surfactant 1 (or 6 ). (Upper creamy oil phase and bottom clear water phase) The two classes of surfactants used had th e alkyl chain either covalently attached through ester (Tween type) or ether (Brij) bonds to the EO group. Surfactant self-assembly and the sol ubilization of oil. High S/O ratio Formulations can be divided into three groups based on the external surfactant-tooil ratio (S/O): a) high S/ O (50-20), b) medium S/O ( 12-5) and low S/O (4-0.8). Chronologically, high S/O samples were i nvestigated first (Table 1.), using primarily Brij-98 (compound 1 ) as a major surfactant and hexadecane (HD compound 2 see Appendix 1) as an oil phase. Surfactan t solution alone was checked with QELS

PAGE 48

29 yielding particles of 103nm, which corresponds to micelles as expected since the concentration of 10-15% is far above cmc of Brij-98 (0.0085w%). Table 2-1. The formulations made with high Surfactant to oil ratio and formulations without the exte rnal surfactant 1 . Formul. No.Surf (g) Co-surf (g) Oil (g) 2OTMS (g)TMOS (g)Saline (g)o. c. (nm)STDNcs (nm)STDoutcome 1 0,4g 10,2960,2258,8Phase. Sep. 2 0,4g 180,05g 190,02160,0388,8Phase. Sep. 30,70,18,8Phase. Sep. 4 1g 190,10,078,8Phase. Sep. 5 1,5g 10,0350,060,058,842129818Stable 6 1,5g 10,060,050,048,8378907Stable 7 1,5g 10,020,030,058,83510Unstable 8 0,7g 10,5g 190,050,048,8Phase. Sep. 9 0,7g 10,8g 180,060,05Phase. Sep. 10 1g 280,020,03Phase. Sep. The solubilization of oil re quires heating, since samples stirred at RT yielded unstable, coarse emulsions that phase separa ted within a couple of hours (Formulation 1 for example). Samples were hazy and turbid, until the temperature reached the cloud point (c.p.) at 65 oC, when the solutions become intensely white and phase separation occurred. The c.p. of pure surfactant solution is 70 oC, the addition of oil lowered the c.p. somewhat, since non-polar compounds tend to increase the aggreg ation number of surfactant, thus lowering the total amount of hydrogen bonded EO groups on the surface. Upon cooling, the solutions become gradually less milky, until totally transparent at RT. This heating/cooling cycle was repeated numerous times with the same outcome. Interestingly, the QELS analysis of these solu tions yielded features of the same size as surfactant alone, within the e xperimental error (Figure 2-7). This implies that oil is distributed evenly, without causing any structur al changes in the surfactant micelles.

PAGE 49

30 0 20 40 60 80 100 120 0102030405060 diameter (nm)Intensity (rel. units) Figure 2-7. The QELS pattern of 1 alone () and microemulsion of 1 and 2 (). Solubilization of OTMS (compound 3) requi red heating at c.p., the difference is that time for dissolution of OTMS was much longer than for the oil alone. This is surprising having in mind the reversibility of surfactant/oil system during heating and cooling. 0 20 40 60 80 100 120 020406080 diameter (nm)Intensity (rel. units) Figure 2-8. The QELS pattern of mixed micelles of 1 and 3 () and ME formed with 1 , 2 and 3 ().

PAGE 50

31 Upon cooling the solution had a faint blui sh color, indicati ng the formation of microemulsion.122 The QELS check revealed two populations of peaks, one at 103nm, and another one at 4212nm (Figure 2-8). Before any further assessments were ma de, a solution containing OTMS and Brij98 was tested to yield particles at 92nm. From this it was obvious that the smaller peaks in Figures 2-7 and 2-8 represent the unswo llen surfactant micelles and larger peaks in Figure 2-8 derive from ME droplets. A soluti on at this stage was stable for 2-3 days, before a phase separation occurred. In order to increase the stability of droplets and to provide reactive silanol groups at the surface, trimethoxysilane groups were subjected to a polymerization. The step-growth polymerization of trimet hoxysilane groups consis t of two distinct steps, hydrolysis and condensation. It is know n that the hydrolysis of alkoxysilanes is faster at lower pH (protonation of oxygen), whereas condensation is faster at higher pH (Sn2 attack of siloxide). Usi ng the results from previous work in this group on OTMS polymerization kinetics123, pH was lowered by the addition of sulphuric acid to pH=3 for 30 min. to complete the hydrol ysis. Moderatly rapid OTMS condensation was achieved at slightly basic buffered solution (pH=7.4) to prevent the formation of a gel. The formulations at this stage will be refereed to as the “oil templates” or “oil cores” being stable for 10 days. The condensation of th e siloxyl headgroups pr ovides a stabilizing "skin" around the oil droplets. However, when samples were prepared for Atomic Force Microscopy (AFM) and TEM anal ysis, the reacted microemuls ion droplets flattened on the supporting substrate, indicating that the "skin" layer is deformable.

PAGE 51

32 Figure 2-9. The polymerization of OTMS ( 3 ) To better retain the spherical shape, the silicon-based skin layer requires fortification. Fontaine and Goldmann124 showed that reaction of Tetramethoxysilane (TMOS, 4) beneath floating monolayers of 3 created a robust film. Molecule 4 provides the additional reactive groups to allow the silica network to extend away from the headgroups of 3 . Upon reaction, 4 formed a complex 3D cross-linked polysiloxane/silicate network which resulted in a film that is more stable than the skin layer formed by cross-linking 3 alone. Lindn et al.125 have shown that, at pH 7.0, the

PAGE 52

33 hydrolysis of alkoxysilanes is slow. Our work used these conditions to enhance reaction at the oil/water interface, wh ile minimizing the reaction of 4 with itself in the bulk aqueous phase. Figure 2-10 shows several TEM images of the nanocapsules synthesized. Figure 210a is the microemulsion before stabilization with 4 . This image reveals oil droplets 35 12 nm in diameter. The dark core is due to the oil, which was doped with 1-dodecene (compound 5 ) and subsequently stained with OsO4(g). TEM images of such droplets also showed a range of droplet diameters and some with noncircular shapes, consistent with the soft, fragile nature of the droplets hypothesized. The sa me solution, when examined 2 weeks later, was cloudy, indicating floccula tion, and TEM results gave no distinct particles. Upon st abilization with 4 , a different structure is observed as illustrated in Figure 2-10b. These spherical nanocapsules have a distinct core-shell structure with the darker core due to the osmium-stained doped o il and the lighter shell due to the presence of silica. The overall size measured from th e TEM images was 98 18 nm and the cores remain at 32 nm. Data from many other im ages (which were analyzed and are not shown) confirm that both the core diamet er and the shell diameters are fairly monodisperse in a given trial. Standard purif ication techniques, such as precipitation or column chromatography, were not possible due to the reactive hydroxyl groups on the nanocapsule surface. These purification techni ques led to nanocapsule aggregation. Once in contact, the nanocapsules react with each ot her or the column media, resulting in an insoluble mass. Consequently, the nanocapsule s were purified in a dispersed state, by dialysis. The molecular contaminants were removed by dialysis through a cellulose

PAGE 53

34 membrane with a 6000-8000 MW cutoff. After dialysis, the solutio ns retained their transparent blue color. Figure 2-10. TEMs of: a) oil templates, b) nanocapsules (obtain af ter the reaction with 4 ) and c) the same solution as in (b) after two weeks. The removal of molecular contaminants such as excess 4 increases the stability of the particles. Otherwise, 4 continue to react and eventually form a gel, causing the particles to become irretrievable.

PAGE 54

35 Surfactant self-assembly and the solubiliz ation of oil. Medium Surfactant to Oil ratio. Table 2-2. The formulations made wi th medium Surfactant to Oil ratio Formul. No. Surf. (g) 6 Co-surf. (g) 8 Oil (g) 7 OTMS (g)TMOS (g)Saline (g)o. c. (nm)STDNcs (nm)STDoutcome 110,30,10,10,0308,814737Stable 120,70,10,10,0308,811340Stable 131,10,10,10,0308,8467Stable 141,50,10,10,0308,84910Stable 1510,10,050,0308,84711Stable 1610,10,090,0308,85412Stable 1710,10,130,0308,89724Stable 1810,10,170,0308,816361Stable 1910,10,210,0308,8233100Stable 2010,020,10,0308,8317Stable 2110,050,10,0308,8516Stable 2210,10,10,0308,85514Stable 2310,150,10,0308,81590Unstable 2410,20,10,0308,8309107Phase. Sep. 2510,250,10,0308,8327127Phase. Sep. 2610,10,10,010,018,8541212318Stable 2710,10,10,020,018,8561610414Stable 2810,10,10,030,018,855146712Stable 2910,10,10,040,018,855156311Stable 3010,020,080,030,018,83177921Stable 3110,020,080,030,038,831713239Stable 3210,020,080,030,098,8317352114Stable 330,650,020,080,030,018,810637292103Stable 340,650,020,080,030,038,810637614124Stable 3510,020,1030,0510,039,67305527Stable 3610,020,1030,0510,0949,673059010Stable 3710,020,1030,0510,29,6730512515Stable High S/O ratio nanocapsules showed poor pr operties with regard to hemolysis and whole animal experiments (described in Ch apters 3 and 4). We believe that the high concentration of surfactant micelles as well as toxic (Hexadecane) oil phase were the two principal reasons for this behavior. Theref ore the system was reformulated as noted below. The relative concentration of surf actant was lowered, and MEs were formulated using more biocompatible oi l phase, Ethyl butyrate (compound 7 )126 and surfactants such as TweenTM-80 (compound 6 )127 and Lecithin (compound 8 ).128, 129

PAGE 55

36 The goal is to prove the generality of the templated microemulsion approach in the formation of nano-sized capsules using a ma ximum number of ingredients that are approved by the Food and Drug Administra tion (FDA) Center for Food Safety and Applied Nutrition (CFSAN).130 Controllability over particle size is demonstrated through the variation of the concentra tion of the formulation ingredie nts. The later is important since the fate of nano-objects in vivo is related to their size.131, 132 Furthermore, nanocapsules of different sizes may have different sequesteri ng efficiencies. In order to create oil templates of various sizes without the “shell” (c ompound 4), a series of samples systematically varying ingredients 3 and 6 8 were made (see Appendi x 1 for structures). Figure 2-11. TEM pictures of formulations having large amount of lecithin ( 8 ), whole solution (top), creamy part of the sa me solution (middle) and microemulsion part of the solution.

PAGE 56

37 A general synthetic route was similar to previous high S/O rati o formulations. In the approximate composition window of (3 -15):(0.5-2.5):(0.3-2.5):(0.1-0.4)(% w/w ratios of 6:7:8:3 ), most of the resulting solutions became clear upon heating, with the appearance of a bluish color. Formulations having relatively high concentrations of 8 (i.e. 2-2.5 w%) were turbid and phase separated to a microemulsion phase and unidentifie d milky dispersion133 after 24 hours (Figure 2-11). On subsequent cooling, the O/W microemulsion was stable at RT for at least 72 h after which coagulation occurred and the sample became cloudy. It should also be noted that while a large and rich literature exists of th e phase behavior of a diverse range of assemblies in lipid-based aqueous systems,134-137 this study centered on compositions where spherical microemulsion particles were reproduc ibly attained and any other phases were not extensively characterized. Oil templates size-Surfactants concentration influence The major component of the three surfactants used is Tween-80 ( 6 ). Compound 6 is soluble in water with a hydrophilic li pophilic balance (HLB) value of 15.4. As the amount of 6 was increased, the diameters of the resulting microemulsion droplets become smaller (Figure 2-12c). When no compound 6 is present, a coarse emulsion is formed. Although exact particle size was not determine d, they were large enough to be seen with minimum magnification under a light microsc ope, indicating that the droplets were micrometers in size. The decrea se in size as the amount of 6 was increased can be attributed to the surfactant adsorbing at the O/W interfac e and lowering the interfacial tension.138-140 The lower interfacial tension allows larger microemulsion droplets to break apart to form smaller droplets. This pro cess is limited by the amount of surfactant necessary to form a monolayer at the interf ace around the droplet. Each time a droplet

PAGE 57

38 breaks apart to form smaller ones, it creates more surface area. Once the surface area is large enough to utilize all th e surfactant at the O/W inte rface, the droplets will not subdivide. The resulting smaller droplets would be unstable. The second surfactant used in this system is lecithin ( 8 ), Figure 2-12d shows that as the amount of 8 is increased, the diameter of the resulting microemulsion droplets actually increases from 30nm to 160 nm. This is contrary to what was observed with compound 6 . With an HLB value of 9.2-9.5,141 8 is not soluble in saline, but is readily soluble in oil.142 High concentrations of 8 caused a phase transition from a clear, isotropic Figure 2-12. The effect of various compone nts on the oil template size a) OTMS ( 3 ), b) Ethylbutyrate ( 7 ), c) Tween-80 ( 6 ), d) Lecithine ( 8 ). liquid to a turbid, thermodynami cally unstable phase which was not further characterized. We believe that compound 8 partitions preferentially into th e interior of the oil droplets,

PAGE 58

39 leaving only a minimum amount at the interface. Furthermore, compound 8 may act to increase the interfacial te nsion when present in larg e enough quantity (Figure 2-12d). Both of these effects would result in the observed increase of the oil core diameter. A third surfactant used was OTMS ( 3 ), the diameter of the oil template remains essentially constant within the concentration range that allowed the formation of a microemulsion (Figure 2-12a). As with compound 8 , compound 3 is not soluble in water but is soluble in oil. It is likely that within the current concentration window, all molecules of 3 absorb at the O/W interface, with out partitioning into the oil phase 7 , thus keeping the diameter of oil template constant . From these results it is obvious that the surfactant which plays the most important role in reducing the size of the oil templates is TweenTM-80. Oil template size-Oil concentration influence The oil also plays a role in the size of th e microemulsion droplets. The weight of oil added is a direct measure of the overall volum e of oil present in the system. Increasing the total volume of oil can lead to two scenar ios. First, the number of droplets present can increase while their indivi dual size is constant. Second, the number of droplets may remain constant, and the increase in oil concentration will incr ease individual droplet volumes. Although the experimental data (Fi gure 2-12b) shows an increase in droplet size with added oil, it is not qua ntitatively consistent with either of these scenarios. In scenario one, if the diameter of the microemu lsion droplets is plotte d as a function of the total amount of oil in the system, the resulti ng line should be linear with a slope of zero. A similar plot for scenario two would result in a line, which could be fitted to the expression:

PAGE 59

40 11 0 bx a a y Equation 2-6 The exponent b=1/3 due to the fact that mass is proportional to volume and the diameter of a sphere is proportional to the cube root of the volume. Figure 2-13. TEM pictures of “small” (upper im age) and “large” (bottom image) oil cores Qualitatively, one can think about the cha nge in size in the following terms. The surfactant(s) will disperse a finite amount of oil, as limited by the amount of surfactant present. If the amount of oil is such that there is more surfacta nt present than is needed to

PAGE 60

41 saturate the O/W interface, th en the excess surfactant remain s molecularly dissolved in the oil phase or as unswollen micelles in th e aqueous phase. Once the amount of oil is increased to the point where a majority of surfactant is attached to the O/W interface, then any additional oil will add to the dropl ets already present rather than create new droplets thus maintaining the lowest po ssible surface area to volume ratio. The consequence of such action will also result in a decreased number of droplets due to unavailability of surfactant molecules. The comb ination of these effects is likely to cause the increase of oil template size (Figure 2-12b and 2-13). Silica Shell Formation The final component of the system, is tetramethoxysiloxane (compound 4 ), which forms the outer shell. Assuming that 4 is initially distributed evenly throughout the solution, then all microemulsion droplets ha ve an equal possibility of reacting with 4 . Thus if the total mass of 4 added to the solution is increased, the mass reacted at the surface of any given microemulsion droplet will also increase. The increase in 4 reacting at the surface of the droplet results in an increase in the thickness of the shell and an increase in the overall diameter of the re sulting nanocapsule. The kinetics of shell formation is shown in Figure 2-14, wh ere the initial concentration of 4 was 0.07 w%. The diameter steadily increases during the first eight hours of reaction, and then it remains constant even for prolonged reaction times, indicating the end of the reaction. The first part of the curve, the actual increase in nano capsule size, can be explained by Fick’s slow diffusion model. The diffusion is slow, m eaning that size will depend on how fast 4 can reach the surface. According to Fick’s Law: J=-DC/ Equation (2-7)

PAGE 61

42 where J is mass flux, D is diffusivity, C is the change in concentration and x is the diffusion length. The assumption is that the difference in shell thickness can be neglected compared to diffusion length of 4 , so x is constant. At this point Fick’s Law becomes: J=kC Equation (2-8) Figure 2-14. The kinetics of silica shell formation, the concentration of 4 was 0.07w%. which is a linear relations hip between concentration (C) and mass flux (J, the number of molecules of 4 that encounter the surface). Control of Silica shell thickness-Role of TMOS and oil template size The difference in thickness of the formed shell between oil templates of different size is a surface area issue. We prepared two samples with the same amount of oil phase, but with different oil template diameters (i .e. different S/O ratios) (Figure 2-15). The “large” core sample (oil core d=10337nm) has less overall surface to be covered, because there are fewer oil temp late particles, which leads to a more profound increase in shell thickness than the “small” core sample (oil core d=31.37nm). Indeed, if a trend between 0 and 0.07w% of 4 is observed for both cases, the data can be fit with a linear expression, y=ax+b,

PAGE 62

43 y=688.6x+31.3 and y=2657.1x+106 Equation (2-9) for “small” and “large” cores respectively. Theoretically, the ratio between two sl opes is calculated in following way: 2 1 12 4 d Aand 2 2 22 4 d A Equation (2-9) Figure 2-15. Nanocapsule diameter as a function of concentration of 4 for “small” oil core () and “large” oil core ( ) formulations. where A1 and A2, are surfaces of individual “small” and “large” droplet, respectively, d1 and d2 are corespondent diameters. The total surface is given by: 3 1 1 1 1 1tot) 2 d ( 4 m 3 ) (A N AA and 3 2 2 2 2 2tot) 2 d ( 4 m 3 ) (A N AA Equation (2-10) where A1tot and A2tot are respective total surfaces for “small” and “large” oil templates, and N1 and N2 are numbers of oil cores calculated as total mass of the sample divided by mass of an individual core. Equati on 2-10 can be simplified if the masses of

PAGE 63

44 samples analyzed and densities of both samp les are equal, which is experimentally confirmed. The ratio between two surfaces is then given by: 3 1 2 3 2 1 3 2 2 3 1 1 tot 2 tot 1) 2 d ( A ) 2 d ( A ) 2 d ( 4 m 3 A ) 2 d ( 4 m 3 A A A Equation (2-11) Combining the equations II-11 and II-9 and subsequent rearrangement gives: 3 . 3 d d ) 2 d ( ) 2 d ( 4 ) 2 d ( ) 2 d ( 4 A A1 2 3 1 2 2 3 2 2 1 tot 2 tot 1 Equation (2-12) This value is in a fair agreement with an experimental value given by the ratio of slopes (2657.1/688.6=3.9) of th e fitted lines (Figure 2-15). It was also of interest to see if the size of the oil core remains constant throughout the shell formation process. We prep ared solutions that contained compound 5 (20 %w/w of the oil phase), which was subsequently stained with OsO4 for better resolution of the oil core and shell by TEM. The results of th ese experiments are s hown in Figure 2-16(ad). The analysis of “small” core 0.07 w% of 4 image (Figure 2-16a) revealed oil core diameters 35 6nm and overall sizes of 77 10nm, which is in fair agreement with the results obtained with QELS (Figure 215). Diameters of “small” cores with 4 at 0.88 w% (Figure 2-16b) are 42 14 nm and 237 62 nm, for oil cores and overall nanocapsules respectively. The difference between TEM (237 nm) and QELS data (352 nm) for

PAGE 64

45 Figure 2-16. TEM images of “sma ll” core at a) 0.07 w% of 4, b) 0.88 w% of 4, “large” core at (c) 0.07 w% of 4 and (d) 0.28 w% of 4. “small” cores 0.88 w% is likely a consequen ce of greater sensitivity of QELS and any scattering technique to larger particles, in addition, the presence of any aggregate (inevitable at high concentration of 4) would overly bias QELS results to larger particles. Better agreement is also observed between TEM and QELS data for “large” core samples; for 0.07 w% of 4 the diameters are 116 23 nm for the core and 308 98 nm for the overall nanocapsule (Figur e 2-16c), and for 0.28 w% of 4, 127 36 nm core diameter and 646 152 nm for the overall nanocapsule (Figure 2-16d) by TEM. These

PAGE 65

46 data suggest that the diameter of an oil core remains essentially unchanged during the shell formation process. Control of Silica shell th ickness-Role of OTMS In order to illustrate the importance of compound 3 we have studied a number of ME formulations with and in the absence of OTMS. The results are shown on Figure 217., samples without 3 phase separate (sample with 4, 6, 7 and 8 is on the right, and sample containing just 4 is in the middle). Moreover, we have not observed any type of organized structures with samples made without 3. Finally, QELS data showed a good agreement with TEM, while samples without 3, are impossible to measure, even after dilution and prolonged sonicat ion. It was already men tioned that variation in concentration of 3 had almost no impact on the size of oil templates. However, QELS results (Figure 2-18) showed that the concentration of OTMS influences the number of crosslinkable templates. Therefore, the sh ell formed using formulations with high concentration of 3 will be smaller than those of lower amount of 3.

PAGE 66

47 Figure 2-17. The importance of OTMS: formulati on made with all of the ingredients (left vial), solution with just compound 4 (middle) and formulation with all of the ingredients except 3. Figure 2-18. The influence of 3 on number of crosslinkable templates (left) and shell thickness of corresponding sample s (right) measured by QELS.

PAGE 67

48 Surfactant Self-Assembly and The Solubilizat ion of Oil -Low Surfactant to Oil Ratio Although medium S/O systems were reproduc ibly synthesized, a majority of surfactant present in the form of the unswollen micelles had to be removed. This imposes the restrictions that the synthetic proce dure was time consuming and that ncs were synthesized in a very low yiel d. Furthermore, presence of un used (in a form of unswollen micelles) surfactant biased a whole set of m easurements (explained in detail in Chapter 3). Therefore, an improvement in syntheti c pathway had to be found to study the ncs properties in detail. Table 2-3. The formulations made with low Surfactant to Oil ratio . (The oil phases in formulations 41-50 consisted of Ethyl Butyrate 7 and 1-dodecene 5) Formul. No. Surf (g) 6 Co-surf (g) Oil (g) 7 OTMS (g)TMOS (g)Saline (g)o. c. (nm)STDNcs (nm)STDoutcome 38100,50,06010,7487Unstable 39100,50,115010,7466Stable 40100,250,06010,7355Unstable 41100,2+0,050,0610,7206Unstable 42100,15+0,10,060,210,7405955Stable 43100,1+0,150,060,210,7284876Stable 44100,05+0,20,0610,7305Stable 45100,8+0,20,11510,713015Stable 46 1g 2300,15+0,10,0610,7305Unstable 47 1g 2900,15+0,10,0610,7336Unstable It was found that clear and isotropic micr oemulsion can be formed using very low S/O ration (in extreme cases less than 1) if polymerizable surfactant (i.e 3, OTMS) is omitted. However, after the addition of 3 the solutions become irreversibly milky at low S/O ratios, even after prolonged heating. It was postulated that the thermodynamic stability of preformed microemulsion is a ma jor obstacle to efficient solubilization of a newly added surfactant. Therefore, preforme d mixed surfactant micelles of the main and polymerizable surfactant were prepared and th e oil was subsequently added. As expected, microemulsions formed in this way were optically clear at very low S/O ratios.

PAGE 68

49 Ostwald Ripening The dramatic increase in the amount of o il added to microemulsion caused some fundamental differences between low S/O comp ared to medium and especially high S/O systems. For example, the stability of the dr oplets was independent on the nature of the oil used (i.e. degree of hydrophobicity) fo r high and medium S/O samples, while profoundly dependent in the case of low S/O systems. Furthermore, the polymerizable surfactant crosslinking density (i.e. concentration of 3) was irrelevant with respect to the integrity of ME droplets for hi gher S/O ratios, while a major factor for low S/O systems. Both of these observations can be explaine d in terms of Ostwald Ripening. Ostwald Ripening is the process of larger particles growth at the expense of the smaller particles. The driving force of the process is the depende nce of the solubility of the particle on its size, as described by Kelvin equation: Equation (2-13) where c is the solubility, r is the radius of the particle, is the interfacial tension, Vm is the molar volume of the substance of the disp erse phase, R is the ga s constant and T is the absolute temperature. It is obvious th at the increase in Vm is going to favor the formation of larger particles. A mechanism of Ostwald Ripening consis ts of oil transfer from small droplets through dispersion phase (i.e . water) to larger droplets. It is important to note that this phenomenon occurs even in low and medium S/O systems, but due to the lower concentration of oil and high excess of surfactant the effects are much less profound. Obviously, the challenge is to attenua te the effects of Ostwald Ripening, thus increase the stability of oil templates and make them suitable for further polymerization.

PAGE 69

50 One strategy includes the increase in concen tration of polymerizable surfactant, thus forming a more crosslinked skin layer ar ound the oil droplet. Figure 2-19. shows the QELS results for two formulations made w ith different polymerizable surfactant/oil (PS/O) ratios. Figure 2-19. The QELS analysis diameter in tegrity for PS/O=0.12 (w/w) sample (left) and PS/O=0.23 (w/w) sample (right). ( ) no polymerization of 3, ( ) after the polymerization of 3. Figure 2-20. The QELS analysis of low S/O samples made with 7 alone ( ) S/O=4 and ( ) S/O=2 (left). Analysis of formulati ons with overall S/O=4 and increasing concentration of non-polar oil 5: ( ) 20w% of the total o il phase is non-polar oil, ( ) 40w%, (X) 80w% and ( ) 60w%. (Right) A lower PS/O formulation, after initia l stability (first 200 h) undergoes steady increase in diameter, regardless on whether the polymerization took place. However, the higher PS/O formulation experiences dramatic increase in diameter when polymerization

PAGE 70

51 was omitted (similarly to lower PS/O), while th e diameter of polymerized oil templates remains constant. The other possible strategy to increase the st ability of oil droplets stems from the mechanistic pathway of Ostwald Ripening, i. e. the transfer of oil through water phase. The data on Figure 2-20, shows the behavior of four formulations with identical S/O ratios, and the oil phase is a mixture of Et hyl Butyrate (polar oil) and 1-dodecene (5) (non-polar oil), where the rela tive w% of non-polar oil is in creased throughout the series. The oil water partition coefficient (pKo/w) of ethyl butyrate is 2, whereas pKo/w of 5 is close to 7, meaning that the later is mu ch less soluble in water. A general observation is that formulations having more insoluble (i .e. lipophilic) oil tend to be more stable as expected.143 Finally, a third way to minimize the Ostwald Ripening is the formation of dispersions using higher molecu lar weight surfactants. Figure 2-21. The influence of surfactant size on stability of low S/O samples (S/O=4, 5 is 60w% of oil phase): ( ) Triton X-100 (23, MW=648g/mol), ( ) Brij 97 (29, MW=709g/mol) and (X) Tween-80 (6, MW=1309g/mol). This is demonstrated on Figure 2-21, by monitoring the oil droplet diameter for formulations containing surfactants of differe nt size. It is obvious that bigger surfactant molecules are harder to depart from the interfaci al layer, thus it is less probable for the oil

PAGE 71

52 to leave the droplet in order to form a bi gger one. The shape and size of oil templates made with low S/O ratios are shown on Figure 2-22. Figure 2-22. SEMs of the formulation with S/O=1 (upper row), oil templates (left) and after crosslinking with 4 (middle and right). TEMs of formulation with S/O=2.2 taken at different magnifications (bottom row). Alternative Formulations A majority of investigation in this chapter describes the formation of ME using medium size, non-ionic surfactants as menti oned earlier. However, since the removal of excess Brij or Tween surfactant is a rather complicated and time consuming process, an effort was made late in the study to formul ate the oil templates using different types of surfactants. Surfactant 3 has a long (C18) alkyl chain, ther efore it is difficult to mix and self assembly in a ME, therefore th e polymerizable surf actants of Si(OCH3)3 type bearing a shorter alkyl chains were employed in formulations.

PAGE 72

53 Table 2-4. The alternative formulations ma de by using the polyme rizable surfactants 16 and 24. Formul.No.Surf (g) Co-surf (g)Oil (g) CxTMS (g)TMOS (g)Saline (g)o. c. (nm)STDNcs (nm)STDoutcome 48 2,4 180,043 270,27 70,15 2416256Stable 49 0,68 180,03 190,088 70,035 168245Stable 50 1g 170,08 190,1 70,035 1616275Stable 51 1,5 170,152 190,22 70,17 16242610Stable 52 1,5 170,06g 190,18g 70,13 2424407Stable 53 1 280,1 60,12 7 +0,04 20 0,1 240,1516100615015Stable 54 1 280,12 7 + 0,03 200,1 240,218467644Stable 55 0,31 270,11 60,03 200,11 240,1510514776Stable 56 0,4 270,25 60,1 200,21 166,5868Stable 57 0,82 230,07 190,16 210,16 246,58111Stable 58 0,6 280,14 220,072 247Unstable 59 2,0 230,4 210,64 160,0812385596Stable 60 1,6 230,42 211,1 1611426Stable 61 0,5 270,07 70,07 165Stable Formulations with n-Octyltrimethoxysilane In this work, a possibility of formi ng stable, crosslinkable ME using compound 16 (n-octyltrimethoxysilane) was studied. As expected, a shorter alkyl chain of 16, broadened the possibilities for formation of stable, crosslinkable ME s. Table 2-4. shows examples of formulations made by using 16 and different types of ionic and nonionic surfactants. A successful attempt to formulate ME without Brij or Tween surfactant was the usage of PLURONIC block copolymer surfactants of PEO-PPO-PEO type (F-127 17 and F-68 18) and the anionic surfactant Sodium Octanoate 19. High molecular weight (MW) block copolymer surfactants, being larg e, need more time to self-assemble and thus the formation of ME took weeks and sometimes (Formulation 51) months. Furthermore, since the h ydrophobic part of PLURONICs is a polymer chain of relatively polar PPO units, it wa s possible to formulate ME us ing just polar oils, like 7. All attempts to produce ME with more hydrophobic oil phases such as 1-dodecene (5), 1octene (20), Isopropyl Myristate (IPM, 21) and Hexamethyldisilane (HMDS, 22) were

PAGE 73

54 unsuccessful. However, due to their size and low mobility, once assembled, PLURONICs ME were by far the most stable overall (Figure 2-23). After numerous unsuccessful trials usi ng exclusively ionic surfactants, it was concluded that ionic surfactants alone are no t sufficient to produce optically clear ME. The scientific reasons for this behavior we re not studied in this work. In order to further decrease the amount of medium size non-ionics, anothe r set of formulations was made using majority of the ionic surf actants and some minimum amount of Tween or Triton (Triton X-100, 23) type (Figure 2-24). Figure 2-23. The stability of formulati on made with Pluron ic surfactant F-127 (17) and polymerizable surfactant 16 at S/O=3 ( ) and S/O=5 ( ). A significant difference in behavior of thes e types of MEs compared to ones made solely with Tween/Brij surfactan ts is the absence of cloud point. Therefore, the difference in solubilization rate between RT and HT was negligible, which can be of crucial importance for potential formul ations of MEs having biol ogically relevant hydrophobic compounds.

PAGE 74

55 Figure 2-24. TEMs of oil cores prepared using surfactants 16, 6 and 27 (S/O=14). Formulations with 3-(trimethoxysil yl) methyl methacrylate (MTS) One of the major problems in this st udy was a dynamic nature of MEs that accounted for sometimes short shelf life of oil templates and cause d difficulties during crosslinking reaction. Therefor e, an investigation toward fo rmulating a more rigid system was needed. Compound MTS (24) was chosen because it is relatively polar (i.e. short alkyl chain) and it has two crosslinkable functionalities: a trimethoxysilane group and a double bond. The idea was to polymerize compound 24 using radical initiators followed by the polycondensation of the surface TMS groups. It was po ssible to obtain clear and isotropic solutions with MTS, using almost all of the surfactants combinations. However, formulations made with Tween, Brij and Triton surfactants yielded poorly defined structures after the thermal radical polymerization of 24. Tween, Brij and Triton surfactants, as already menti oned, undergo phase transition (i.e. clouding) at elevated temperatures. As a consequence, the integrity of ME droplets is lost, leading to formation of ill-defined structures after the reaction (Fig 2-25).

PAGE 75

56 Figure 2-25. TEMs of form ulations made with polymerizable surfactant 24 and 6 (left) and 23 (right) as the external surfactant. Well-defined latex-like particles (i.e. oil templates) were obtained when ionic surfactants alone or in majority were used in formulations (Fig 2-26). Furthermore, better results in terms of shape, size, polydisper sity of the droplets where obtained when oil soluble initiators were used (AIBN 25) than water soluble ones (Potassium Persulfate 26), and when the oil phase had a poly merizable functionality, like 1-octene 20 (Figure 226, TEM on the right). Figure 2-26. TEMs of form ulations made with polymerizable surfactant 24 and 17 (left) and cationic surfactant 28 along with 6 (right) as the external surfactants.

PAGE 76

57 Silica shell formation for alternative formulations A second step in the synthesis (silica she ll formation) was attempted on all stable alternative formulations. Unfortunately, PLURONIC surfactant based ME droplets were unable to crosslink with 4, although they were the most stable templates overall synthesized in this work. One possible reason of this outcome is the steric hindrance at the interface where the reaction should take place. Compound 17, for example, is a block copolymer PEO101-PPO56-PEO101, and 18 PEO80-PPO27-PEO80, meaning that the hydrophilic PEO corona (or “outer shell”) of the micelle is ve ry dense, thus effectively shielding the reactive SiOH groups at the interface. The formulations prepared with mixtures of ionic and non-ionic surfactants due to time constraints, were not studied in detail. However, the initial experiments toward the assessment of silica shell formation suggest f easibility of this process with MEs made with ionic external surfactants and polymerizable surfactants 16 and 24 (Figure 2-27). Figure 2-27. TEMs of alternative formulati on after the crosslinking reaction with 4, at different magnifications. Surface Modification of the nanocapsules A third, and final step, in the synthesis of ncs, is the chemical modification of the silica surface. The scientific rationale fo r the surface modifications are outlined and

PAGE 77

58 explained in detail in Chapter 3 of this work. Here, it is worthwhile mentioning that the properties of ncs, especially if intended to be used in biol ogical or medical purposes are dominated by their surface chemistry. The presence of reactive silanol (Si-OH) groups on the ncs surface was used to covalently attach modifiers using the same silica condensation chemistry that was used for the skin layer polymerizati on and the growth of silica shell around the template. Compounds 9 (2[methoxy(polyethyleneoxy)propyl]trimet hoxysilane (MW 460-590), “PEO-TMS”), 10 (3-(Triethoxysilyl)propylsuccini c anhydride, “TMS-COOH”) and 11 (aminopropyl trimethoxysilane, “NH2-TMS”) were used to attach non-charged, negatively and positively charged moieties on the surface of the ncs. In case of compounds 9 and 10 the synthesis was rather straight forward, incl uding a step addition of modifier into the buffered stock solution of ncs at pH=8 and s ubsequent stirring after the complete addition for 48 hours. This method was not feasible for compound 11, since the amino group is protonated, thus positive at pH =8 and it is electrostatically attracted to negatively charged ncs silica surface. Therefore, it is not efficien t way to synthesize the amino modified ncs. In order to overcome this problem, the reaction with 11 was done in the acidic stock solution of ncs at pH=4, where the ncs surface is weakly negatively charged and electrostatic interaction is not profound. After the 48 hour stirring, the modified ncs were dialyzed against Milli Q water to remove the by-products and excess of modifiers. A convenient method for analysis of surface modi fied ncs was TEM, since all modifiers can be stained with OsO4.

PAGE 78

59 The difference in the appearance of the modified and non-modified ncs is shown on Figure 2-27, where non-modified ncs, have an easily detectable darker core and lighter shell, whereas modified ncs are uniforml y darker due to the staining with OsO4. Figure 2-28. TEMs of nanocapsules with thin, medium and thick silica shell (from left) prior to PEO modification (upper ro w) and after the PEO modification (bottom row). Conclusions. This Chapter analyzes the formations of mi croemulsions that can be crosslinked to form robust nanocapsules. Although compound 3, that served to introduce the crosslinkable groups in the ME is the surfact ant by definition (i.e. a presence of non-polar and polar end), it was not able to solubilize any type of oil and self-assemble into a microemulsion droplet by itself. Therefore, the presence of an external surfactant was necessary. A second feature that was common fo r great majority of MEs is the necessity

PAGE 79

60 of elevated temperatures during the solubi lization and assembling process. The most important parameter for the control of ME dr oplet size is the S/O ratio. A general trend noticed in this work is that the size of droplets is irreversib ly proportional to S/O ratio. It is found that the minimum droplet diameter was roughly 30nm in the range of infinitely high to medium S/O ratios. It was possible to obtain smaller micelles, however, but only in absence of 3. This result is probably the consequence of spacing required for the network formation of 3 and its possible cr ystallization during the polymerization. However, this is a speculation as this phenom enon was not in the focus of the study. A more precise control of the droplet size is achieved at medium S/O ratios by changing the relative concentrations of all formulation i ngredients. The increasing concentration of 3 increases the number of polymer izable droplets, while the droplet size remains constant. At low S/O ratios, ME droplet s are unstable and their growth is consistent with the Ostvald Ripening, that can be prevente d by increase in the concentration of polymerizable surfactant, using more hydr ophobic oil and higher molecular weight surfactants. A second synthetic step, the formation of silica shell around the ME droplet has been achieved using sol-gel agent 4. The most important c ondition for successful crosslinking is moderately basi c pH of the solution. In cases of either too low or too high pH, the resulting solution gelled within hours. The second requirement is the stepwise addition of 4, due to its reactivity and possibility of forming other types of structures like solid silica nanoparticles or disordered mesopor ous materials that resemble zeolites. The control of silica shell thickness was attemp ted and successfully achieved with medium S/O samples, and it is a func tion of the concentration of 4 and the increase is proportional

PAGE 80

61 to cube root of 4. A comparison between ME droplets ma de with the same amount of oil at different S/O ratios, the shell formed de pends on the surface area of droplets. The shell will be thinner for smaller droplets with highe r surface area compared to larger droplets with lower surface areas. The same rationale can be applied when droplets of the same size, prepared using different concentration of 3 when exposed to sol-gel reaction with 4. The thinner shell is observed with samples that contain a higher concentration of 3, due to the higher number (i.e. higher su rface area) of polymer izable templates. It was noted that uncontrolled crosslinking starts at concentrations hi gher than 2w% of 4, therefore the growth of a thick shell s hould be carried at low concentration of templates. The chemical structure at the surface offers a wide of range of opportunities for further modification of ncs. It was demonstrated that ncs can be cova lently modified with polymer, or compounds that have a variety of functional groups. A possibility of forming crosslinkable ME using polymerizable surfactants other than 3 is demonstrated. In essence, compounds more easily soluble and assembled than 3 were employed in the formulation. The af orementioned possibility of alternative polymerizable surfactants in turn broadened the range of external surfactants that are able to form ME. Although not studied in deta il, formulations made with compounds 16 and 24 are good candidates to overcome certain limitations and difficulties (i.e. higher temperatures) in preparations of conventional MEs. This will be important when a formulation contains hydrophobic enzymes or other temperature sensitive compounds.

PAGE 81

62 CHAPTER 3 PHYSICO-CHEMICAL AND BIOMEDICAL CHARACTERIZATION OF CORESHELL NANOCAPSULES Introduction This part of the study describes the char acterization of the nanocapsules (ncs), followed by the study of the interactions at the bio/nano interface and the assessment of ncs encapsulation efficiency toward guest molecules. The most important property of ncs in the solution is the stability with respect to aggregation. In Chapter 2, it was shown that shell prevented any changes in diameter of the oil core. However, the concern is whether the ncs will remain an independent entity in the solution and for how long. This question b ecomes more important with regard to the ncs primary application, that is, the inj ection of the ncs in the bloodstream. The bloodstream is an exceptionally harsh medium characterized by the high ionic strength and presence of many different biomolecules that can alter the properties of silica ncs. Therefore, a set of experimental methods was used to correlate the nanocapsules stability in water with the surface structure. The primary goal of this work was to design an injectable system able to lower the toxic drug concentration in the bloodstream . Therefore, an assessment of biological adversity of ncs was done using red blood cel ls (RBCs) and Thromboelstography (TEG). The uptake ability was studied as a func tion of the ncs con centration, size, architecture and surface structure. These st udies were done in order to understand the mechanisms of the guest molecule removal.

PAGE 82

63 Core-Shell Systems Objects having chemically distinct interior (“core”) and outer (“shell”) domains are called core-shell materials. There ar e many reasons why materials are coated: enhancement of thermal, mechanical and chem ical properties, durability and change of overall biochemical behavior of material. 144General procedures to obtain nano core-shell structures are: self-assembly of block copolymers used as templates145, 146, while shell can be formed by layer-by-layer (LBL ) deposition on particle surface147, 148 and sol-gel deposition on the template. The later method is used in this work to obtain core-shell nanocapsules. However, most of the sol-ge l deposition work rela tes to coating of quantum dots149 having excellent luminescent proper ties only in organic solvents. In order to prevent aggregation when transfe rred in polar (aqueous) solvents, a protecting silica shell is introduced. 150, 151 Protection shell preserve s quantum dot properties, enabling their application in catalysis and magnetization.152, 153 Organic latex particles are also used as seeds for silica coating, forming organicinorganic hybrids, with an option to rem ove the organic core by calcination, leaving hollow silica particles with very interesting applications.154, 155 Experimental Methods Zeta Potential ZETA Potential measurements were us ed to measure the charge of the nanocapsules in solutions of different pH. The values were obtained by measuring the electrophoretic mobility U of the particles, which is given by: U= /(V/L) Equation (3-1) where is the speed of particle (cm/s), V is the voltage and L is the distance between the electrodes. The ZETA Potential ( ) is then calculated by:

PAGE 83

64 = (4 U)/ Equation (3-2) where is the viscosity of the solution, and is the dielectric constant. The measurements were performed on self-calibrating Brookhaven ZetaPlus Particle Sizing Instrument. A stock solution of ncs was diluted to a final concentration of 0.05w% and pH was adjusted by adding appropr iate amount of 0.5M HCl or NaOH. Zeta potentials were estimated by a direct measur ement of electrophoretic mobility, employing the technique of electrophoretic light scattering. The data presented are the average with respective standard deviati ons (STD) of 3 independently made formulations. Uv-Vis Spectroscopy This analytical technique was used for two different purposes in this work. The first one is to monitor the concentration of surfactan t at various stages of preparation using the Lambert-Beer’s law linear dependence of absorbance on analyte concentration. Equation (3-3) where A is the absorbance, a is the molar absortivity (a n intrinsic constant for each compound), b is the length of the cuvette and c is the concentration of the analyte. A standard curve was constructed with known concentrations of surfactant and a linear equation describing the relati onship between an absorban ce and concentration was derived. All of the surfacta nts used in this work that had double bond ( a chromophore) were analyzed using UV ( max=233nm for Tween-80 and Brij -98). A second purpose was to monitor the uptake of guest molecules into the hydrophobic core of the ncs. Briefly, Quinoline (12), a hydrophobic drug mimic Po/w=115 (oi l/water partition coefficient) was monitored by its absorption peak at 299 nm. c b a A

PAGE 84

65 Figure 3-1. The standard curves for Tween -80 (top) and Brij-97 (bottom) obtained by measuring UV absorbance at max=233nm. The stock solution of oil templates/nanocapsules (200 L) was diluted with saline (buffered at pH 7.4) (1.8 mL), 10 L (0.085mmol) of 12 was added to the solution. Resulting solutions were stirred fo r 1 h, then filtered through 0.025 m pore filters (Millipore Corporation) to remove free 12 and the spectrum was obtained. A 10 L aliquot of the resulting soluti on was taken and diluted with 3.6 mL of normal saline in a cuvette. The decrease in absorption at 299nm between control and respective solutions was measured. The UV-visible spectra were obtained using a Vari an Cary UV-VisibleNIR instrument. Results are shown as rel.%=100%-(Ax/Atot)*100, where Atot. and Ax are absorbencies of the control sample and the sa mple of interest respectively. The detection limit for 12 was 2 mol and absorption in the concentration range used to asses the uptake ability of oil cores/nanocapsules obeye d the Beer-Lambert la w. All spectra were scanned in a double beam mode with normal salin e as the reference. Each point presented

PAGE 85

66 here is the average with re spective standard deviations of three independently made formulations. Infrared Spectroscopy (Ir) IR was used in this primarily to char acterize the difference in surface composition for PEO modified and non-modified ncs. The ncs were dried overnight in the oven, and the spectra were recorded as KBr pellets on a Nicolet model 510P spectrophotometer. High Performance Liquid Chromatography (Hplc) HPLC was used to quantify the uptake poten tial of ncs. The nanocapsule solution was diluted 10 and 100 times with PBS to pr oduce a 1% w/v and 0.1% w/v nanocapsule solution. A stock solution of 14 was prepared by dissolvi ng the hydrochloride salt in saline solution and added to diluted ncs solution to achieve the desired initial concentration of 14. The solutions of nanocapsules and compound 14 were stirred for 1 h at RT to establish equilibrium between absorbed 14 and free 14. Aliquots were then placed in Centrifree YM-30 cellulose centrifugal filter devices (Amicon Bioseparations) containing 30,000 MW cut-off filters and centrifuged at 3000 rpm for 30 min. The supernatant was then removed for HPLC analysis. A 4.6 250 mm reverse phase C-18 column (Waters) was used with a 30/70 acetonitrile/phosphate buffer (pH 3.5, 50 mM) eluent. A PDA detector (Millipore) was used at 210nm. The samples were analyzed in duplicate and the average was reported. Data presented here ar e average and standard deviation of 3 such measurements. All HPLC measurements were done under the superv ision of Dr. Manoj Varsheny, from Department of Chemical Engineering, University of Florida.

PAGE 86

67 Hemolysis Hemolysis assays were conducted in order to investigate the toxi city of ncs towards red blood cells. This simple biomedical test provided a plethora of information on adverse effects of ncs with respect to thei r size, concentration a nd surface properties. The experimental protocol was as follow s: 3ml of fresh human blood was dispersed in 11ml of phosphate buffer saline (PBS) and centrifuged at 2500 rpm for 15 minutes. The plasma was discarded and 10ml of PBS was added to the remaining red blood cells (RBCs). The solution was gently mixed and centrifuged as before. This procedure was repeated for 3 times. After the final washing step, 3ml of RBCs were added to 11ml of PBS. 0.1ml aliquots of samples were added in duplicate to Ependorf tubes and 0.9 ml of Millipore water (18.2 M ) and PBS were used as positive and negative probes respectively. 0.l ml of stock RBC mixture wa s added to each sample. The samples were mixed gently and incubated at RT for 20minut es. Aliquots of 0.2ml were taken from the tubes and centrifuged at 2500 rpm for 2 minutes to remove the RBCs or cellular debris. Samples analyzed at this point will be regarded as hours” samples. hours” samples were stored in the fridge w ith occasional gentle mixing a nd analysed after 24 hours, hours” were samples analyzed after 48 hours a nd so on. Supernatants were tested for the presence of free hemoglobin by UV-VI S spectroscopy. Briefly, a D2-lite deuterium/tungsten light source (World Pr ecision Instruments, Sarasota, Fl) was projected through fiber-optic ca bles to a CUV sample holder and its intensity was picked up by an S-2000 detector (WPI) attached to a computer. Data was acquired and analyzed using Spectraware 2.8 software (WPI). 100 l was added to 2ml of a 200:5 (v:v) Ethanol: HCl solution. Absorbance was read at =398nm. The absorbencies of PBS and De-

PAGE 87

68 ionized water were plotted against 0 and 100% respectivel y and a linear equation was calculated. This equation was used to transf orm hemoglobin absorbance in the samples to percent RBCs hemolyzed. Thromboelastography (TEG) The TEG test was used to asses the in fluence of ncs on the blood coagulation process. Therefore, this test was used to investigate the behavior of whole blood in the presence of ncs. Figure 3-2. A Typical TEG Trace and important parameters. Analysis of a TEG trace TEG generates a gr aphic representation of a clot quality, as the sample forms a clot. The analysis of this simple test is rather complicated, including several parameters that should be ta ken into account. The general scheme of a TEG trace is shown in Figure 3-2.

PAGE 88

69 Figure 3-3. The library of traces in TEG. Taken from Dennis, D.M. and T.E.Morey, University of Florida, Shands Hospit al, Dept. of Anesthesiology, Private communication. The analysis of a TEG trace includes: r time, which is a time necessary for th e beginning of the clot formation; k time is the period from the beginning of the clot formation till it reaches the maximum rate; angle is the slope of th e clot formation function; MA is a maximum amplitude which is a measur e of a clot overall strength( platelet interaction); A60 represents stability of the formed clot and it is measured 60 minutes after the maximum amplitude is achieved.

PAGE 89

70 Qualitative interpretation of a TEG trace After the iniatial analysis of a trace, th e results are reviewed and compared with a quality of library of different TEG traces suggesting the mechanism of an analyte interaction with blood clot. The librar y of traces is shown on Figure 3-3. Disposable cups and pins (Haemoscope Co rp.Niles, IN) were inserted into two Thromboelastograph units (Haemoscope Corp.) and the TEG machines were powered on and allowed to reach a temperature of 370C. 30 l of normal saline (control probe) and serially diluted samples (1, 0.1 and 0.01 w% of ncs) were placed in each of the four cups. 240 l of citrated whole blood was ad ded to the cups, followed by 30 l of 0.2M CaCl2. The pistons were then advanced and retracte d from the mixture five times to ensure proper mixing. The pistons were subsequently inserted fully in the caps. A drop of mineral oil was added in each cup to prevent the evaporation of the sample, immediately after the charting mechanism was engaged w ith simultaneous notations of the starting point on the paper reco rd. After 60 min. the experiment was terminated. Each essay was performed in triplicate using blood obtained from different subjects. The control was varied between and within coagulation an alyzers for different specimens to avoid spurious control data that might arise from using a singl e control lane. The following parameters were measured from TEG analysis: k time, r time, angle and maximal amplitude (MA). Briefly, k time is the durat ion of time measured from the point where the amplitude is 2mm to the point where th e amplitude is 20mm. If the TEG trace never reached 20mm mark, k time is arbitrarily taken as 60 min., i.e. the terminal duration of the experiment. The r-time is the duration of time from the beginning of an experiment to the point where the amplitude was 2mm. To determine the angle a longitudinal

PAGE 90

71 centerline was first drawn on the TEG trace. The second line was drawn from the centerline at the point of 2m m width to tangentially touch the amplitude at the 20mm width. The angle between these two lines was the angle (Figure 3-2). Results Concentration of Nanocapsules The most important step in the characteri zation of ncs after the synthesis is to investigate how many ncs are actually present in the stock solution, how stable they are toward aggregation and total composition wi th respect to ncs, unswollen micelles and microemulsion droplets. Figure 3-4. The relationship between the sca ttering intensity and particle concentration for polystyrene standards obtained from Duke Corporation, 100nm (top) and 20nm (bottom).

PAGE 91

72 Therefore we constructed standard curves using silica particle standards of 20nm and 100nm from Duke Corp by plotting the sca ttering intensity vs. weight percent of particles. Similarly to Lambert-Beer’s law, the goal is to find the concentration range where the aforementioned relationship is linear (Figure 3-4). Figure 3-5. The relationship between the sca ttering intensity and particle concentration for 100nm nanocapsules. Figure 3-6. The percent scat tering intensity of 20nm ( ) and 100nm Duke standards ( ) as a function of their weight ratio.

PAGE 92

73 The advantage of doing these experiments w ith standards of two different sizes is the possibility of establishing the relationship between the percent scattering intensity and the ratio of particle diameter. Therefore, it is possible to measure the relative ratio between surfactant micelles/soft microemulsi on droplets and ncs in any given sample. Based on experiments using the 20nm a nd 100nm standards mixed at different weight ratio (Figure 3-6), the scattering inte nsity as a function of diameter was calculated in a following way: xd d w w I I 20 100 100 20 20 100 Equation 3-4 where I100, w100 and d100 are the scattering intensity, weight percent and diameter of 100nm particles, and I20, w20 and d20 are the respective values for 20nm particles. Figure 3-7. The diameter ratio parameter (x from Equation 3-4) for 20nm and 100nm Duke standards as a functi on of their weight ratio. The data from Figure 3-7 are in fair agreem ent with theoretical value of x=3, since the scattering intensity is the function of vol ume, i.e. diameter raised to the power of

PAGE 93

74 three. An example of QELS analysis of na nocapsules subsequently after the synthesis (top) and after the purification ( bottom) is shown on Figure3-8. Figure 3-8. An example of QELS measuremen t for nanocapsules after the synthesis (top), and after the purification (bottom).

PAGE 94

75 The most striking difference is the re moval of micelles (15-20nm) after the purification. According to Equation 3-4, the weight ratio between the micelles and nanocapsules was 30 and 0.01 following the synthesis and after the purification respectively. A second experimental technique used to asses the concentration of remaining surfactant in the ncs samples and to follow the efficiency of purification steps was UVVIS (Figure 3-1). A standard curve was constructed with known amounts of surfactants used in the preparation. Therefore, we have related the data from QELS experiments with UV spectra to determine the precise concentr ation of the free and bound surfactant at the oil-silica shell interface. Unfortunately, the amount of surfactant that was bound to the interface, i.e. used for the formation of te mplates was just up to 5% of the initial concentration. Surface Properties of Nanocapsules It was already mentioned that the surf ace of nano-objects is crucial for their properties because of the high surface-tovolume ratio. This section of Chapter 3 describes the behavior of nanocapsules (aggregation, surface charge, biomedical properties) depending on ncs surface structure. Aggregation of the ncs The stability and aggregation of ncs was i nvestigated using QELS under different pH value and at ionic strength for ncs of differe nt surface chemistry. The ncs aggregation in solutions of different pH was monitored fo r a month and is shown on Figure 3-9. The aggregation profiles showed that the surface char ge is the key feature for stability of ncs. Negatively charged ncs, unmodi fied silica and acid modified (Figure 3-9a and d) samples

PAGE 95

76 tend to aggregate in acidic solutions due to the protonation of their respective conjugated bases Figure 3-9. The change of nanocapsule samples diameter with time at pH=3 ( ), pH=7.4 ( ) and pH=11 ( ) for: a) silica surface, b) PE O surface, c) amino surface and d) acid surface.

PAGE 96

77 (carboxylates) on the surface. This process, in turn, minimizes the surface charge, i.e. diminishes the electrostatic re pulsion between the particles that leads to aggregation. Similar rationale can be applied for behavior of amino modified ncs that are stable at acidic pH, while tending to aggregate in basi c environment. The aggregation profile of PEO modified ncs was independent of solution pH. The importance of surface chemistry is demonstrated on Figure 3-10, when clear solutions of acid and amino modified nano capsules immediately aggregated upon mixing. Figure 3-10. The aggregation of the amino (v ial on the left) and aci d (vial on the right) modified nanocapsules upon mixing (vial in the middle). Surface charge of nanocapsules The study of ncs aggregation clearly show ed the importance of surface charge on fundamental properties of these entities in solution. It was importa nt to quantify these conclusions by measuring the ZETA Potential of various ncs before any biomedical work was done. The results are shown of Figure 3-11.

PAGE 97

78 Figure 3-11. The ZETA potential curves for ncs with different surface ( ) silica surface, (o) PEO surface and () PEO and succinic acid surface. Figure 3-12. Zeta Potential cu rves for amino modified () and succinic acid modified () ncs.

PAGE 98

79 As expected, the hydrophilic polymer (PEO ) greatly reduced the charge on the surface and minimized the pH influence on ncs pr operties. It is interesting to note that even a small amount of succinic acid (PEO/C OOH=4:1 (w/w)) diminished the effect of PEO in the case of mixed ncs (Figure 3-11. ). Furthermore, the ZETA potential curves for amino and acid modified ncs are in good ag reement with results of the aggregation studies Figure 3-12. Hemolytic properties of ncs A first and the simplest biomedical te st was the hemolysis, i.e. studying the interaction of red blood cells (RBCs) with ncs. The cell integrity can be measured by UV monitoring of the free Hemoglobin released from broken RBCs. A standard way to perform hemolysis is to expose the cells to the analyte for 20-30 min., centrifuge the mixture and measure the absorbance at 398 nm. However, this method was inadequate for this study since almost all of the test ed samples showed no profound hemolysis after such a short exposure. On the other hand, su ch a behavior opened up a possibility of studying the interaction for prolonged exposure times, up to 72 hours. A first goal was to study the hemolytic prof ile of RBCs in the presence of particles at various stages of synthetic process. Therefore, RBCs were exposed to surfactant micelles and soft microemulsions. At 0.1w% both showed statistically similar toxicity toward RBCs (Figure 3-13.), which was expect ed since both are very dynamic species. The mechanism of RBC rupture in these cases is the intercalation of individual surfactant molecules in the cell membrane,156, 157 and its subsequent diss olution by detergency properties of surfactants. These experiments suggested that all ncs tested had to be thoroughly purified by dialysis, centrifugati on and filtration prior to hemolysis experiments.

PAGE 99

80 Figure 3-13. The hemolysi s activity of Tween-80 ( ) and Tween-80/Ethyl Butyrate microemulsion ( ) at 0.1w%. There is a range of biocompa tibility responses to silica ba sed materials. It is well known that silica causes a lung disease called Silicosis158 and quartz has been classified as a carcinogen by the International Agency for Research on Cancer (IARC) in 1997.159 However, there are reports on amorphous si lica based materials, such as Bioglass 160, 161showing good biocompatible proper ties in contact with other tissues. Most of the silica studies related to biocompatibility are cytot oxicity studies with respect to pulmonary diseases.162-164 The reports focused on the interac tion of silica in the blood, more specifically to red blood cells (RBCs) are rather scarce in th e literature. A study of Hunt et al.165 shows that chitosan-silica ae rogels have good cytotoxic pr ofile, but are toxic with respect to RBCs. One hypothesis is that silica is causing the hemolysis of RBCs,166 through the interaction of the negatively char ged surface of the silicaceous materials at

PAGE 100

81 physiological pH with the quaternary amines of the phospholipids at the cell membrane surface. This implies that the extent of hemo lysis would be a function of surface area, i.e. contact area between cells a nd particles. The hemolytic act ivity for ncs of distinct surfaces at three different thicknesses is repr esented on Figure 3-14. It is important to note that experiments were perf ormed on approximately the same number (i.e. same oil content) of ncs at three different thicknesses. All of the samples tested showed negligible hemolytic activity upon 20 min. exposure to RBCs with the exception of thick ncs with carbocyclic acid functionality on the surface (“ 0 hours” point in the Figure 3-14 a-c). Furthermore, hemolytic activity increased with prolonged exposure for every sample regardless of surface properties (i.e. from 0 to 24 hours, from 24 to 48 hours etc.). Analysis of samples w/o PEO showed that th ere is a statistically significant decrease in polymer modified ncs hemolysis of RBCs. This result is not surprising as it was shown that PEO greatly reduces cytotoxicity of Quan tum Dots (QDs) in the recent cell viability studies.167, 168 Moreover, if the extent of hemolysi s of ncs w/o PEO is compared for ncs with different thicknesses for 24 hours of expos ure, it is a function of the overall surface area for silica ncs, while the contact area is not a significant factor for PEO modified ncs (Figure 3-14d). It is obvious that hydrophilic, chemically inert polymer at the surface of ncs slows down the hemolytic pr ocess, although it still occurs. Th is result implies that the surface coverage is not complete, allowing low concentrations of unreacted Si-OH to eventually bind to the RBCs surface. It is impor tant to note that the hemolytic activity of ncs was at the same level even after ncs were modified under extreme conditions (high concentration of PEO-TMS at pH=11 for several days).

PAGE 101

82 Figure 3-14. Hemolysis of RBCs in the pres ence of ncs having different surface at 0.1w% oil content: a) thin shell (PEO and si lica ncs shown), b) medium shell (PEO, silica and PEO/Succinic acid shown) , c) thick shell (PEO, silica, PEO/Succinic acid and Succinic acid alone shown) ( a-c lines are guide to an eye) and d) hemolysis of RBCs afte r 24h for ncs with silica and PEO surface (d-lines represent statistical trends). ( ) silica surface, ( o) PEO surface, ( ) PEO and succinic acid surface and () succinic acid surface. (* P<0.05 for silica and PEO modified surface ncs). A deeper insight into the influence of surface on the RBCs hemolysis, is gained by testing ncs with succinic acid and polymer moiety together (PEO/Succinic anhydride at 80/20 w/w), and acid alone. The former showed activity statistically similar to that of non-modified ncs, while ncs with acid functionality alone were the most damaging of all tested ncs. According to these data the surface acidity gr eatly influences the RBC hemolysis, since succinic acid is

PAGE 102

83 a much stronger acid than silicic acid (p Ka=4.2 and 5.8 for succinic and pKa=9.8 and 13.8 for silicic acid). Figure 3-15. The hemolysis of amino modified ncs ( ) and silica ncs ( ) for medium shell (top) and thick shell (bottom). Although charged, amino modified were as toxic as non modified ncs as shown on (Figure 3-15). This result implies that the na ture of the charge is also important with

PAGE 103

84 respect to hemolysis, weakly positively ch arged (see Figure 3-12) are less damaging compared to negatively charged ncs. These experiments gave an indication of the relation between the surface properties of ncs and hemolysis. However, a more deta iled study involving th e preparation of the ncs with acids of different strength should be done in order to exact ly quantify the effects of acidity and hemolysis. Thromboelastography (TEG) A complementary test to hemolysis, TEG wa s used to study the toxic impact of ncs in whole blood. In the core of this analysis is monitoring the interaction of the analyte and blood, measuring the formation and the m echanical strength of the formed blood clot. Blood coagulation is one of the most complex processes that take pl ace in the biology of the bloodstream. It involves twelve cascade steps including synchronized action of a number of proteins and cells. Similarly to hemolysis, the goal was to test particles at each step of the synthesis. The results for surfactant micelles and soft microemulsion are shown on Figure 3-16. Several conclusions can be drawn from blood clothing profile in the presence of these dynamic species: r-time is the same as control, meaning that micelles and oil droplets are not influencing the proteins responsible fo r the clothing (not shown); k-time is dramatically reduced, which implies that the platelet function is reduced; MA is significantly shorter, revealing the poor mechanical properties of the clot, indicating reduced plat elet interaction.; According to the literature, this behavior is expected. Small non-ionic surfactants (as oppose to large block copolymer micelles) tend to block the platelet interaction, thus inducing the prolongation of k-time and MA.169

PAGE 104

85 Figure 3-16. Maximum Amplitude (top le ft) and k-time (top right) for Tween-80 solutions. The r-time for Tween-80 solutions (bottom left) and Tween-80 and Ethyl Butyrate soft microemulsions (bottom right). A next step was testing ncs of silica a nd PEO surface, at three different shell thicknesses (oil core was the same for all nc s). The two important TEG parameters are represented on Figure 3-17 for all three types of ncs: maximum amplitude (MA) and k time. Both parameters decreased in a concen tration dependence fashion (P<0.001). It is believed that ncs at 1w% are too toxic regard less of surface, while at 0.01w% too dilute to have any significant impact on coagulation process. The statistical difference for both, MA and k-time, is observed between ncs of different surface (w/o PEO) at the concentration of 0.1 w% for medium and thick ncs. At intermediate concentration, the k-

PAGE 105

86 Figure 3-17. The MA for ncs with silica and PEO-modified surface a) thin shell, b) medium shell, c) thick shell and d) comp arison of MA for ncs of different size at 0.1w%, silica surface and PEO-modified (lines represents st atistical trends). The k-time for ncs with silica and PEOmodified surface e) thin shell, f) medium shell, g) thick shell and h) comparison of k-times for silica and PEO modified ncs at 0.1 w% based on graphs 3 e-g (lines represents statistical trends). (* P<0.05 for silica and PEO-modified surfaces) (plots a-c and e-g (o) PEO-modified surface, ( ) silica surface, graphs d and h (o) PEO modified surface, silica surface)

PAGE 106

87 time for PEO modified ncs is essentially the same, regardless of surface area, while the k-time for unmodified ncs is prolonged dram atically with the increase of surface area (from thin to thick ncs). It is well known that PEO, in many aspects, is a “magical” polymer. The most striking feature is that PEO reduces protein adsorption,170 and the prolongation of circulation in th e bloodstream. It seems that th e former feature of PEO is not important with respect to blood coagula tion process, that is, r-time which is the reflection of protein cascade pr ocess (initiation of coagulation) is constant for all of the samples regardless of concentration and su rface chemistry (Figure 3-18). On the other hand, MA and k-time reflect the platelet activ ity, and are strongly affected by the nature of the surface of ncs. Non-m odified ncs dramatically reduce the platelet activity, which is contrary to what is commonly observed with artificial surfaces, su ch as titanium oxide171, which actually promoted plat elet adhesion and led to hype r coagulation. Non-ionic surfactants are, on the other hand, the co mmon platelet activity reducers. However, surfactant used in the synthesis of ncs wa s removed by multiple dialysis, filtration and sedimentation (for large ncs), to the point wh ere it was only at the in terface of an oil-inwater microemulsion (i.e. no surfactant was tr aceable by UV-VIS in the supernatant, and detection limit was at the critical micelle concentration (cmc).Moreover, the concentration at the interface was the same for modified and non-modified ncs regardless of size (i.e.the size of the oil templates was th e same for all three sizes of ncs). This fact leads to a conclusion that reduction of plat elet activity is likely caused by surface properties of ncs, rather than detrimental effect of the surfactant.

PAGE 107

88 As already mentioned, the TEG parameter th at describes the prot ein function in the beginning of the coagulation process, r-time wa s constant for all tested ncs regardless of concentration, surface chemistry and size. Figure 3-18. The r-time for silica ( ) and PEO-modified ( )ncs: thin shell (top), medium shell (middle) and thick shell (bottom). Uptake Profile of Nanocapsules One of the goals of this study was to s ynthesize nanocapsules that are able to sequester the lipophilic compounds from aque ous systems. This part of Chapter 3 describes the experiments done to quantif y and understand the mechanisms of the lipophilic compound rem oval by the ncs. High Surfactant to Oil Rati o Samples-Uptake Ability Nanocapsule (high surfactant-to-oil (S/O) ra tio) solutions were used to extract compound 12 from normal saline solutions. Compound 12 was chosen as a drug model because it is hydrophobic (lipophilic), a liqui d under ambient conditions, and the ring

PAGE 108

89 structure shows a UV absorbance that is no t obscured by the nanocapsule absorption band. A preliminary study of the nanocapsule sequestering of 12 from bulk solution was performed. When 12 was added (0.15M) to the nanocapsule solution ( 1.4% w/v oil content) in saline, it was easily visible as an insoluble droplet. U pon agitation, the droplet broke apart to form small droplets, which re mained insoluble in solution. With time, visible droplets of 12 disappeared as they were sequest ered into the hyd rophobic core of the nanocapsules. After 14 min 12 was completely sequestered. This is expected because the oil-water partition coefficient ( P (o/w) = 115) for 12 is large, indicating a higher affinity for oil phases. As a complimentary test, compound 5 was substituted for 12 since 9 is similarly insoluble in saline. Initially, molecule 5 appeared as a droplet on the bottom of the vial. With time, the droplet di sappeared, indicating the nanocapsules had sequestered it. Samples of this solution were stained with OsO4(g) and analyzed using TEM. Molecule 5 preferentially stained, a nd TEM images showed that 5 had partitioned into the core of the nanocapsules and was not just adsorbed onto the nanocapsule surface (Chapter 2). Both of these studies qual itatively examined the sequestering of a hydrophobic substance from bulk saline in to the oil-filled nanocapsules. The decrease in 12 with time was measured quantitatively using the UV absorption peak at 299.5 nm and is shown in Figure 319. Many lipophilic drugs, including tricyclic antidepressants (i.e., amitriptyline 14) and local anesthetics (i.e., bupivacaine 13), which the nanocapsule system could be used to se quester, are toxic at concentrations around 5 M (Chapter 4).172 Experiments were performed with 8 M compound 12.

PAGE 109

90 Figure 3-19. Removal of quinoline from salin e solution. The quinoline UV absorbance at 299.5 nm was monitored in a solution with 8 M quinoline in saline. Concentration of free quinoline with respect to time as affected by the concentration of nanoparticles: ( ) 1.4% w/v oil content, ( ) 0.8% w/v oil content, ( ) 0.3% w/v oil content, and ( ) 0.1% w/v oil content. The inset illustrates the removal of quinoline as a function of varying concentrations of nanocapsules. This concentration is 4 orders of magn itude lower than the above preliminary visual study. The free 12 concentration does not change with time in a pure saline solution without addition of the nanocapsules. Particle size analysis of high S/O samples shows that the nanocapsule solution, even after purification by extensive dialysis, contains mixed micelles and microemuls ion droplets. These can also bind 12, causing a drastic drop in free concentra tion. It was estimated that the micelles alone remove 70 5% of 12. When the nanocapsule solution is diluted, its ability to remove 12 decreased

PAGE 110

91 (Figure 3-19). TEM and light scattering conf irmed that dilution does not affect the nanocapsules themselves, likely due to the st ability provided by th e polysiloxane/silicate shell. In these diluted suspen sions, the bulk of compound 1 is trapped within the shell of the nanocapsules. The remainder is presumably within the micelles and microemulsion droplets. The cmc of 1 is 0.00154% w/v. This is only 100-fold lower than the total concentration used in the formulation. As a result, dilution destabilizes the micelles and microemulsion droplets. When diluted enough, they break apart. The results show that less 12 is bound at higher dilutions, due to the disappearance of the micellar and microemulsion moieties. Each dilu tion removed the maximum amount of 12 from the bulk solution in 15 min. The 1.4% w/v oil content nanocapsule solution removed >97% of 12, the 0.9% w/v oil cont ent removed 45 5% of 12, the 0.3% w/v oil content removed 35 5% of 12, and the 0.1% w/v oil co ntent removed 32 5% of 12 (Figure 319 inset). In all cases, a monoexponential decay equation with plateau gave the best fit. This is consistent with the ki netics observed by Ding and Liu,173 for the loading of Rhodamine B in water-soluble hollow nanosphe res. Specifically, a biexponential decay equation with plateau did not give a stat istically improved fit over a monoexponential decay equation with plateau for the 0.1% w/v oil content nanocapsule solution (P = 0.88), 0.3% w/v oil content nanocapsule solution (P = 0.36), 0.8% oil content nanocapsule solution (P = 0.93), and 1.4% w/v oil conten t nanocapsule solution (P = 0.10) data. The objective for creating th ese nanocapsules was the removal of drugs from blood so a preliminary uptake study was performed with bupivacaine (13), a local anesthetic. Molecule 13 does not have intense bands in th e UV-visible region of convenience to measure, so its removal from saline was m onitored using HPLC. Current emulsion-based

PAGE 111

92 treatments such as Intralipid, used to deliver the antifungal, propofol, are used at concentrations of 1% w/v total oil conten t in the blood. The uppe r concentration limit for the commercially available products is 1. 5% w/v oil content in blood. With this in mind, the stock nanocapsule solution was dilu ted to yield a soluti on that was 0.1% w/v oil content. To the 0.1% w/v oil content nano capsule solution, 13 was added to yield final drug concentrations of 7, 27, 49, 96, 181, 363, 536, 1030, 3400, and 10380 M. Figure 3-20. Removal of varying concentratio ns of bupivacaine as determined by HPLC. The data at low concentrations shows a linear fit with a correlation of 0.997; the dotted line shows the intersectio n indicating the capacity of the nanocapsules. These solutions were allowed to equilibrate at room temperature for >30 min, after which they were centrifuged in Centrifree filt er devices to remove the nanocapsules. The filtrate was analyzed by HPLC to determine the amount of free 13 present. As Figure 320 illustrates, for concentrations <100 M, the nanocapsules sequestered all of 13. The nanocapsules sequester >99% of 13 when the initial concentration is 200 M. The amount sequestered becomes constant at 1900 M while the initial concentration is

PAGE 112

93 increased. The leveling off of the amount of drug sequestered is reasonable since the nanospheres have a finite capacity and once th is is reached they cannot uptake more no matter how much is added. Figure 3-20 indicates the capacity of the nanocapsules at 0.1% w/v oil is 1900 M. Medium Surfactant to Oil Rati o Samples-Uptake Ability A next step in this part of investigatio n was to quantify the relationship between the oil core and ncs size and seque stering efficiency for medium S/O samples. Controlling the size of the nanocapsules also provides insi ght into how the architecture influences the efficiency of the system to sequester lipophilic compounds. Initially several uptake experiments we re performed on the microemulsions (oil templates) which were stabilized by crosslinking of 3, but without subsequent templating and growth of the shell. Each value wa s obtained one hour after the addition of compound 12, when the equilibrium was reached. Compound 12 is soluble in a micellar solution of 6, therefore there is a possibility th at the decrease in absorbance of 12 is due to unswollen micelle absorption. Several experiments were performed to understand the influence of this process on the obs erved decrease in absorbance of 12. Solutions containing just compound 6 at concentrations of 3, 7, 11 and 15 w% and compound 10 were monitored by UV-VIS spectroscopy to monitor the change in absorbance of 7. The results, presented in the inset Figure 321, show no effect. Therefore the observed decreases in absorbance can be accounted for by 12 being sequestered in side the oil cores or nanocapsules.

PAGE 113

94 Figure 3-21. The uptake efficiency of oil te mplates formed by changing the concentration of: a) compound 3 (plotted as w% of 3 vs. decrease in absorbance for clarity); b) compound 7; c) compound 6 (inset represents the decrease in absorbance for solutions containing just 6); d) compound 8. Compound 12 is readily soluble in the oil phase (7), and Figure 3-21 (a, c, d) shows uptake trends for oil cores of different diameter made by changing the amount of 3, 6 and 8 respectively while the concentration of oil phase 7 is constant at 1w%. A general

PAGE 114

95 observation based on these graphs is that oil templates of smaller diameter have better uptake capabilities than larger ones. The ra tio between total surf ace areas of two oil templates having different diameters is inversely proportional to their corresponding diameters. The smaller oil templates will have a larger interfacial area, which accounts for more efficient partitioning in the sense th at oil phase is more “available” to guest molecules. This interpretation is also in ag reement with the result s from the literature,174, 175 where solubilization or release from mice llar systems is proportional to the size of interface area ( i. e . number of droplets). Interestingly, the coulombic attraction between the negatively charged 8 (pI<6.5) and positively charged 12 at pH 7.4, seems not to have significant impact on uptake, although this phenomenon was not further quantified in the current study. It was surprising that the uptake remained fairly constant with the increase in the oil concentration (Figure 3-21b) . An advantage of increasing the oil concentration is the resulting decrease in the numb er of droplets (“oil cores”). This effect minimizes the surface to volume ratio as expected, since fewer surfactant molecules are available at higher concentrations of oil (Chapter 2). Therefore, the drug partitioning in the oil cores is proportional to the in terfacial area and is not a function of the oil phase concentration. Figure 3-22 shows that the drug partiti oning decreases as the shell thickness increases. Shell thickness is given by: 2 ) (0d d Sttot Equation (3-4) where dtot and do are respective diameters of nanocapsu le and of oil core. This result is not surprising as increasing the concentration of 4 results in a thicker shell (as illustrated in Chapter 2), thus diminishing the permeation of the drug.

PAGE 115

96 Figure 3-22. The uptake efficiency as a function of shell thickness according to Eq. 10 obtained by QELS (values for dtot and do are taken from Figure 2-15). The effect is at least partly due to the effect of larger shell thickness increasing the denominator of d /dx, where is the chemical potential, a driving force between the aqueous and oil phase for these particles. Mechanism of Amitriptyline (AMI) Uptake Although previous experiments with hydrophobic compound 12, were successful “proof of a principle” for the ncs detoxi fication application, an evaluation of in vitro partition of target drug (AMI, 14) in the ncs was necessary. Similarly to experiments with Bupivacaine (13), the AMI uptake was measured using HPLC method developed by Dr. M. Varshney, in the Department of Anesthes iology at UF. The target drug at certain concentration was allowed to equilibrate with ncs at RT for 1h, followed by centrifugation in Centrifree membrane tubes (molecular cut-off 30 kD) and analysis of the fraction that went through the membrane. Experiments were performed in the media

PAGE 116

97 of increased complexity, i.e. Milli Q water (no ions present), PB S buffered saline (1w% NaCl solution), human plasma and whole blood. Previously conducted experiments us ing hydrophobic probe revealed that unswollen surfactant micelles have high affi nity toward hydrophobic species present in aqueous systems. Moreover, the HPLC me thod was an opportunity to quantify the removal efficiency of the surfactant mice lles, since the weight of one micelle (diameter~7nm) was calculated to be 150,000 D. Figure 3-23. The Amitriptyline removal using Tween-80 micelles measured by HPLC The results (Fig 3-23) showed that initia lly uptake scales up linearly with the concentration of 6, reaching a plateau at ~0.6w% with ~70 M of the AMI bound. This uptake pattern resembles that of high S/O sa mples (Figure 3-20) implying that the uptake is dominated by the surfactant micelles. More importantly, these results pointed out the significance of the ncs purifica tion procedure, not only for biomedical reasons, but also for the accuracy of the ncs detoxification poten tial data (Chapter 4). On the other hand

PAGE 117

98 there was a possibility to study the comp etition between unswollen micelles and ncs toward the removal of the hydrophobic drug. The results are shown on Figure 3-24, unfortunately, as expected, surfactant micelle s are much more efficient towards removing of the target drug than ncs. 0 20 40 60 80 100 medium largeAMI bound (%) Figure 3-24. The Amitriptyline removal comp arison for 2w% ncs (white bars) and 1w% ncs and 1w% of 6 (grey bars). This result is expected having in mind th e difference in size of the two species (surface to volume ratio) and more impor tantly, the dynamic nature of micelles (formation and dissolution of micelles). Further experiments were done in order to understand the predominant interactions responsible for the removal of the target drug from aqueous systems. The ncs of 3 different sizes, where oil template is the same with different shell thickness (the same used in biomedical experiments) were test ed for toxic drug remova l. Contrary to our expectations, the uptake scaled up with th e diameter of the ncs (Fig. 3-25). The increasing diameter, or shell thickening, has two important consequences on ncs properties: a) a material transfer from the su rface to the core is suppressed with increased

PAGE 118

99 shell thickness (see Figure 322) and b) the surface area per ncs is increasing with increase in diameter. Figure 3-25. The removal of Amitriptyline meas ured for ncs of different shell thickness (same number of ncs) (top), the rem oval of Amitriptyline plotted versus relative surface area for the three populations of ncs (bottom). It is then obvious that the later explains the observed result, meaning that the AMI removal involves an adsorption process on the surface, whether or not is followed by the actual encapsulation in the ncs oil core. The a pparent disagreement with previous results using hydrophobic probe 12, is probably a function of different structure of 12 and AMI. Compound 12, has a positively charged group as the AMI, however AMI has a positive

PAGE 119

100 charge “protruding” from the end of the alkyl chain, causing the predominance of electrostatic effects on molecule behavior. This qualitative ex planation is consistent with the results obtained by Varshney176 using 13 and AMI as a model drug for in vitro binding. The AMI uptake using microemulsi on was highly dependent on pH (i.e. number of charged AMI molecules), while compound 13 showed no profound pH dependence. A next step in this part of the study wa s to investigate the influence of surface structure on uptake profiles of ncs. Figure 3-26. The uptake of Amitr iptyline by silica (white bars) and PEO (grey bars) ncs. The PEO modified and silica ncs have sim ilar uptake capabilities (Figure 3-26), as expected, since weakly negatively charged surf ace (Silica ncs) is exchanged for polar EO units of the polymer. In other words, the weak ly electrostatic ncs drug interaction (Silica ncs) is exchanged for hydrogen bonding of EO units and protonated amine of the drug for PEO modified ncs. However, mixed PEO and succinic anhydride (4:1 w/w) ncs showed

PAGE 120

101 substantial improvement of the uptake prof ile compared to PEO alone or silica ncs (Figure 3-27). Figure 3-27. The difference in uptake ability for PEO modified (white bars) and mixed (PEO and COOH) modified ncs (grey bars). The increase in the uptake efficiency can be explained by the presence of a stronger acid (i.e. stronger electrostatic inter action) at the surface of the ncs. Finally, the difference in uptake ability wa s demonstrated in the microelectrode array set up (explained in detail in Chapter 4) . Briefly, the toxic drug (AMI) was injected in cardiac muscle cell monolayer along w ith the PEO modified or PEO and COOH modified nanocapsules. The results are show n on Figure 3-28. The re lative spike rate of the cells is a function of the toxic drug c oncentration (Chapter 4), it is obvious that PEO/COOH ncs are more powerful detox ag ents than PEO modified particles.

PAGE 121

102 Figure 3-28. The spike rate of cardiac muscle cells in the presence of AMI alone ( ), AMI and PEO modified ncs ( ) and AMI and PEO/COOH modified ncs (). The envisioned site of action of the ncs will be the bloodstream, since most of the toxic drug will be dispersed in it. It was a logical step forward in the uptake characterization to perform the measurements in human plasma. Unfortunately, none of the tested ncs samples showed any appreciab le uptake under these conditions (Figure 329.). It should be noted that more than 95% of the drug is bound to bloodstream immunoproteins in the control probe, thus the concentration of the free drug is very low for the ncs to show any uptake. However, the surfactant micelles (Figure 3-29 bottom) were able to sequester the signi ficant portion of the free drug.

PAGE 122

103 Figure 3-29. The toxic drug upt ake in human plasma measured by HPLC: control and three most efficient ncs samples (t op) and control and Tween-80 at two different concentrations (bottom). Conclusions This part of the thesis explains the me thods to characterize the nanocapsules (ncs) after the synthesis. Some very important f eatures like concentration and aggregation are investigated by the combination of methods. Th e majority of propertie s are related to the ncs surface chemistry. Electrostatic repulsion is a phenomenon responsible for the long term water solubility and stability toward a ggregation of the ncs. The acid and silica ncs

PAGE 123

104 are the most stable in basic solutions when surface groups are fully deprotonated, thus negatively charged. Furthermore, amino modified ncs are more stable in acidic solutions due to protonation of the amino groups at the surface. A somewhat different rationale explains the stability of PEO coated ncs, the stability and lack of aggregation is a consequence of the hydrophilic polymer attached at the surface, i.e. polymer forms a hydrodynamic layer around the particle that effect ively protects particle from any type of interaction that could lead to aggregation. Moreover, the PEO modified ncs are the only particles that showed stability towa rd aggregation re gardless of pH. The second part of this Chapter was the study of nanocapsules and surfactant micelles (i.e. “soft” microemulsions) interacti ons with biologically relevant media. The hemolytic properties of ncs were studied us ing human red blood cells and monitoring the release of Hemoglobin as a consequence of cell damage. Microemulsions and simple surfactant micelles were extremely toxic with respect to hemolysis, probably due to their size and mobility. The extent of hemolysis fo r non-modified ncs is proportional to the contact area between the part icles and red blood cells, as de monstrated by separately monitoring the toxicity of the same number of ncs with different diameters (i.e. different surface areas). A different conclusion can be drawn for PEO modified particles that had similar hemolytic activity regardless of size. Moreover, the key f eature in reducing the hemolytic activity of silica co re-shell ncs is decreasing the surface charge, therefore preventing the electrostatic interaction betw een particles and positively charged lipids on the cell surface. Polymer modified ncs signifi cantly decreased the hemolysis of red blood cells compared to non-modified ones. Followi ng the aforementioned rationale, hemolysis is more profound when an acid stronger than silic ic acid is present at the particle surface.

PAGE 124

105 Succinic acid modified ncs were the most toxic since succinic acid is much stronger than silicic acid. The second test used in this study to asses the biomedical properties was Thromboelastography (TEG), the blood coa gulation process was monitored in the presence of the ncs and “soft” microemulsi on paticles. An interesting result was that neither ncs samples nor microemulsion solutions at any tested con centration influenced the protein cascade process duri ng the initiation of coagulation. Similarly to hemolysis, surfactant micelles were toxic, greatly influe ncing the mechanical properties of the clot. The extent of the clot damage is a function of ncs concentration, regardless of size and surface properties. However, the toxicity scaled with diameter of nc s for the silica ncs, while, like in hemolysis experiments, it was independent of size for the PEO ncs. Moreover, at equal concentrations, the PEO si gnificantly reduced toxicity with respect to blood coagulation compared to silica ncs. It is important to note the introduction of the standard clinical method (TEG) in the scie ntific investigation of nano-objects. After the biomedical investigation, a set of experiments was performed in order to measure and understand the mechanism of t oxic drug uptake. Experiments with high Surfactant-to-oil ratios revealed that mixed micelles together with certain portion of ncs are very powerful drug and model drug s cavengers. Unfortunately, subsequent experiments with separated mixed micelles and ncs (i.e. medium Surfactant-to-oil systems), showed that former are much superi or drug detoxification agents than the latter. Multiple model drug uptake experiments were conducted with formulations that yielded the oil templates (i.e. polymerized microemuls ions) of different size. The most important feature with respect to sequestering potential was the size of the inte rfacial area, i.e. the

PAGE 125

106 smaller droplets were more efficient towa rd drug uptake compared to bigger ones. Experiments with ncs of different sizes fu rthermore revealed that the drug uptake is proportional to surface area. An important conseque nce of this result is that the uptake of a target drug (i.e. Amitriptyline) is rath er the surface adsorption process than the encapsulation in the hydrophobic core of the ncs. A second important parameter that influenced the uptake potential was the nc s surface chemistry. The PEO modified samples had essentially equal drug detoxifica tion power as silica ncs, however, when a stronger acid (i.e. succinic acid) was atta ched on the surface together with PEO, it significantly improved the uptake profile. This result confirmed that the key interaction in the uptake process is the electrostatic attr action between the negatively charged surface and positive quaternary amine of the drug. Another confirmation of the uptake mechanism was the comparison of cardiac myoc yte monolayer spike rate in the presence of the toxic drug (lowered th e spike rate) and the ncs of different surface chemistry, where PEO and COOH mixed ncs showed superi or performance compared to particles modified with PEO alone. Although the ncs had an appreciable uptake profile in simple media like water or saline, when tested in more complex environment, human plasma or whole blood, ncs failed to lower the concentrat ion of the target dr ug. The only particles capable of removing the drug from the human plasma were the surfactant micelles. This result implies that further improvements are need ed in order for ncs to be applied in real life.

PAGE 126

107 CHAPTER 4 DETOXIFICATION ABILITY OF NANOCAPSULES Introduction This Chapter describes the experiments done in order to te st the ability of nanocapsules to remove selected toxic drugs in vivo and in vitro . Experiments toward the assessment of ncs toxicity in isolated heart, whole animal and isolated cardiac muscle cells are described. Finally, the envisioned a pplication of ncs, as a detoxification therapy is demonstrated in cardiac myocytes and in whole animal trials. Isolated Heart The isolated rodent heart set up is rela tively routinely done experiment in the Shands Hospital at University of Florida.177The electronic signals are used to measure the QRS interval of the electrogram that represen ts the summation of a ll sodium ion currents in the ventricles. Thus, the QRS serves as an indication of na nocapsule toxicity. Whole Animal Studies Experiments in vivo have been used to monitor the activity of organs such as the liver,178or malfunction of neruronal179 and cardiovascular180 tissues in the presence of tricyclic antidepressa nts. All experiments in vivo were carried out in the Shands Hospital at University of Florida, under the superv ision of Dr. Donn Denni s. Similarly to the isolated heart set up, the QRS signal was used as a parameter in the assessment of acute toxicity. The animals were then kept alive for a week to assess the possible long-term adverse effects of the drug or nanocapsules. The measurements were used to investigate the ncs toxicity profile and th eir detoxification potential.

PAGE 127

108 Cardiac Myocytes Deposited on Microelctrode Arrays (MEAs) Unfortunately, due to high costs of bot h, isolated heart and whole animal experiments, just a glimpse of full toxicity study could be achieved. Therefore, the focus was shifted toward the inves tigation of ncs toxic profile in cardiac myocyte monolayers plated onto Microelectrode A rrays (MEAs). Advantages of this technology also include ethical aspects of in-vitro studies (e.g. a tissue from just one animal can provide a plethora of information through multiple experiments on different MEAs). Before the data measurements cells are is olated from a sacrificed animal, followed by an enzymatic degradation of the connecti ng tissue. The cells we re then plated on a protein-coated MEA under the conditions that maintained physiologi cal activity. Because the microelectrodes are independently a ddressed, the experiments also allowed simultaneous multi-channel readings which were processed by powerful acquisition equipment. Furthermore, the application of microelectrode arrays (MEAs) for studying so-called “bioelectricity”, i.e. electrical phe nomena occurring within the tissues and cells, is one of the emerging fields of biotechnol ogy. This technology offers a long term noninvasive measurement of extracellular activity of excitable cells from neuron and cardiac tissues.181, 182 Understanding the connection between disordered electrical excitability (e.g. ion channel dysfunctions) to many diseases is important in areas such as medicine and pharmacology.183 The later is increasingly fo cusing on molecular level of bioelectricity (i.e. ion channe ls), since a number of diseases are the consequences of the ion channel dysfunction.184 The experiments were performed in colla boration with Dr. Mark Pottek from the Max Planck Institute for Polymer Research (MPIP). The measure of cardiac myocytes activity was the spike frequency and spike amplitude.

PAGE 128

109 Experimental Methods Isolated Heart Experiments Figure 4-1. The isolated heart set up. All experimental protocol s were reviewed and approved by the Animal Use Committee of the University of Florida H ealth Sciences Center. Guinea pig hearts isolated from animals weighing 450-500g we re perfused according to the Langendorff technique.177, 185 Briefly, the hearts were rapidly re moved and rinsed in ice-cold KrebsHenseleit (K-H) solution containing 1 17.9mM NaCl, 4.8mM KCl, 2.5mM CaCl2, 1.18mM MgSO4, 1.2mM KH2PO4, 0.5mM Na2EDTA, 0.14mM ascorbic acid, 5.5 mM glucose, 2mM pyruvic acid and 25mM NaHCO3. The ascending aorta was cannulated for perfusion with 8ml/min K-H solution gassed continuously with 95%O2 and 5%CO2. The

PAGE 129

110 oxygen tension, temperature and the pH of the K-H solution were maintained at 500-600 mm Hg, at 36oC and pH 7.4, respectively. Afte r the completion of dissection and instrumentation, the hearts were allowed to equilibrate for at least 20min. before the experiments were begun. The hearts were p aced (3-ms pulses at twice-treshold intensity) at an atrial cycle length of 250 ms (240 beat s/min) via a bipolar electrode placed on the high atrioseptal area. The isolation of the heart and measurements were done by Dr. Tim Morey and Mr. Jason Flint. Whole Animal Experiments Sprague Dawley rats (350-500g) were anes thetized using 5% Halothane in an induction chamber. Anesthesia was mainta ined using 1% Halothane via nose-cone apparatus. An intravenous catheter with th e nanocapsules soluti on was placed in the lateral tail vein. EKG electrodes were placed using surgiclips on the abdomen to measure the QRS interval along with the respiratory transducer. Data were acquired by Bio-Pac acknowledge software. Rats were monitored for QRS interval, heart rate (HR) and respiratory rate (RR). A baseline (i.e. normal QRS) was measured in a tranquilized animal for 1h. Amitriptyline (AMI) was given through the syringe needle at the concentration of 25mg/kg or 80mmol/kg into the intraperitoneal space (a bdomen). This concentration of AMI was chosen to induce the increase in the QRS interval of about 40% of the initial value. The measurement of AMI toxic effect wa s performed by the injection of drug at constant flow rate for 20 min., followed by an injection of a saline solution. The preemptive experiments included the simultaneou s injection of AMI and ncs solutions for 20 min., followed by the addition of saline. The QRS was measured for total of 1h in each case. The experiments were done by Mr. Jason Flint.

PAGE 130

111 Cardiac Myocytes on Microelec trode Arrays Experiments MEAs were purchased from Multi Channel Systems, Reutlingen, Germany. The MEAs were autoclaved and incubated with 250 l solution of fibronectin (50 m/ml) for 4 hours. The Fibronectin protein coating was n ecessary to achieve better adhesion of the cells on MEAs and to make the surface more biocompatible. Subsequently, the MEAs were washed three times with PBS to remove the excess protein and dried for 30 min., prior to cell seeding. Cardiac Myocyte Isolation and Deposition on MEA The rats were sacrificed and embryos were placed in PBS at the animal care station. The embryos were decapitated and hearts we re removed after thoracic insections. The hearts were collected and chopped in Hank’s balanced salt solution (HBSS) containing 136 mM NaCl, 5.2 mM KCl, 0.35 mM Na2HPO4, 0.44 mM KH2PO4, 0.81mM MgSO4, 1.26mM CaCl2, 5.5mM glucose and 0.028mM Phenol Red. The collected tissue was washed and gen tly redispersed three times with HBSS solution using a Pasteur pipette. The supernat ant was removed after the sedimentation of the tissue, followed by the addition of a fresh portion of HBSS. The F-10 medium solution is a HEPES buffered solution with fo llowing ingredients: in sulin, transferrin, selenium, penicillin, streptomycin and 10% fetal bovine serum. The F-10 will be also referred to as the “cell medium” in the following paragraphs describing MEA experiments. A stop solution is a F-10 medium that cont ains 30% fetal bovine serum and it is used to stop the enzymatic degradation of the tissue. The tissue was subjected to enzymatic degradation as follows. The tissue was first added to 2.5 ml of Trypsin soluti on, and incubated for 20 min. at 37oC. A stop solution and 20 l DNase I was added to the tissue suspension. The tissue at this stage was gently

PAGE 131

112 redispersed using Paster pi pette, followed by sedimentation. The supernatant was collected and centrifuged at 1300 rpm for 5 min. at 4oC. The supernatant was discarded and the pellet was re-suspende d with 10 ml of F10 medium . The remaining pellet was subjected to the same enzymatic procedure as described above and combined with the main suspension after. Figure 4-2. The fluorescent mi crograph of cardiac myocytes (blue-nuclei labeled by 4',6Diamidino-2-phenylindole (DAPI)) (left) and optical micrograph of cardiac myocytes deposited onto MEAs The final suspension was filtere d through a cellular filter (40 m pore diameter), transferred to a culture flask and incuba ted for two hours in the incubator. The supernatant was collected and centrif uged at 1300 rpm for 5 min. at 4oC and resuspended in F10 medium. The cell density was determined using a Neubauer cellcounting chamber (used 20 l of the suspension, count ce lls in four squares, mean number multiplied with 10,000=number of cells in 1ml of suspension). The cells were seeded on fibronectin coated MEAs and cover slips. The approximate cell number on MEA (17 mm di ameter) was 300,000. The cell seeded MEAs were incubated for 2 hours after the additional F10 was added. The F10 medium

PAGE 132

113 containing 3% of fetal bovine serum was re placed every two days, except when MEAs were used in experiments, when the me dium was added subsequently after the measurements. Figure 4-3. The principle expe rimental idea, 1) cells are de posited on fibronectin coated MEAs, 2) the activity of cells is mon itored, 3) toxic drug is added and 4) nanocapsules addition with continuation of a measurement. Picture of the MEA device (lower right). Results Toxicity of Nanocapsules The positive results of the basic biomedical tests described in the previous Chapter enable the study of biocompatibily of the ncs in more complex systems, such as isolated heart, whole animal and cardiac myocyte monolayers. The first tests were the influence of high surfactant-to-oil (S/O) formulations on the QRS time in a living guinea pig. Unfortunate ly, as already mentioned in the text, high

PAGE 133

114 concentration of the surfactant was extremely toxic, resulting in the death of animals within 5 min. after the ncs injection. (Figure 4-4). Figure 4-4. The influence of high S/O ratio nanocaspules on QRS of a whole animal, normal QRS ( ) and QRS with ncs ( ). After the successful synthesis of medium S/O formulation the be havior of ncs in the isolated heart was investigated. Obvious ly, this set up is simpler than the whole animal. Moreover it is convenient and rou tinely performed in the Department of Anesthesiology at UF. The results are shown on Fig. (4-5). The QRS interval of the isolated hear t gradually increased with increasing concentration of ncs up to 50% longer than the normal value at 1w% of ncs. Although this may seems to be a negative result, the ncs at 0.1w%, which is the anticipated therapeutical concentration, increased the QRS only <15%, which is considered benign. The other parameter monitored, SA, was essentially constant with increasing concentration of ncs. The SA is defined as th e time necessary for the electrical impulse to travel from the Sino-Atrial node across the atri a. The third parameter, SV time, is defined

PAGE 134

115 as the total time for an impulse to travel from the Sino-Atrial node to leave the AV node. It can be concluded from these experiments that the ncs most influenced the AV node of the isolated heart. Figure 4-5. The influence of ncs on isolated heart parameters, QRS ( ), SA ( ) and SV ( ). A next step in the toxicology characteri zation of ncs was th e monitoring of QRS time in vivo . The medium S/O nanocapsules were inj ected in a tranquilized rat and the QRS was monitored for 40 min. (time necessa ry for the observation of acute effects), followed by observation of overall animal health for a week, after which the animal was sacrificed. The ncs effect on the QRS interval of the live rats is showed on Figure 4-6,. It is somewhat surprising that the QRS remained constant at 1w% of ncs, compared to apparently more profound effect obser ved in the isolated heart set up.

PAGE 135

116 Figure 4-6. The influence of nanocapsules at 1w% on QRS in vivo , control ( ) and with ncs ( ). However, on further reflection there are sign ificant differences in the environments. There are many lipid reservoirs, immunoprot eins and other biomolecules that can attenuate the effects of the ncs on the heart f unctioning in the whole animal. The isolated heart on the other hand is rather unprotected tissue. The animals te sted survived for a week suggesting that ncs alone we re not damaging to the organism. As mentioned earlier, Microe lectrode Arrays (MEAs) we re more practical due to economic and ethical reasons for further inves tigations of the ncs t oxicity profile. The toxicology experiments using MEAs included the investigation of the major surfactant alone (Tween-80 for this particular formul ation) and purified ncs. As expected, the surfactant micelles are very toxic with respec t to cardiac myocyte ac tivity. The cells were no longer active just 3 min. afte r the exposure (Figure 4-7 left). The influence of ncs at four different concentrations on cardiac myocyt e activity is shown on Figure 4-7 (right).

PAGE 136

117 Figure 4-7. The spike rate of cardiac myocyt es in the presence of Tween-80 at 0.025 w% (left) and PEO and COOH m odified nanocapsules (right). Ncs at 0.025w% had identical spike rate patterns as the control, whereas 0.05w% and 0.1w% had no effect after 2 min. expos ure, followed by a continuous decrease to about 80% of the initial activity after 10 min. The highest conc entration used in this study was 0.2w%, which was characterized by a sharp and steady decrease to about 20% activity after 15 min. exposure. The therapeu tic concentration of ncs used in further experiments was 0.05w%. Preemptive Detoxification Experiments Toxicity Reversal in Whole Animals These experiments were done in order to ul timately test the abili ty of nanocapsules to lower the concentration of toxic drug (AMI in this case) in the living animal. An AMI dosage that would increase the QRS by 40% of the normal value was given to the animal (Figure 4-8).

PAGE 137

118 Figure 4-8. The therapeutic acti on of nanocpasules, control ( ), AMI alone ( ) and AMI and ncs together ( ). The second experiment included the simulta neous injection of the drug and ncs in the animal. The encouraging aspect of the re sults is that the nanocapsules lowered the concentration of AMI in the body, base d on decreased QRS time in these cases. However, 2 out of 3 animals died 24 and 48 hour s respectively after th e treatment. This is contrary to what was observed with ncs alone (Figure 4-6) where no apparent (short and long term) detrimental effects were observ ed. It was obvious that detoxification procedure with lower concentra tion of ncs had to be performed, and the results of the experiments with ncs at 0.5w% are shown on Fi gure 4-9. As expected, the effect of ncs with respect to decreasing th e AMI concentration was less profound in this case, however there were no damaging effects to the animal for the next 7 days, after which it was sacrificed.

PAGE 138

119 Figure 4-9. The therapeutic action of d iluted nanocapsules (0.5w%), control ( ), AMI alone ( ) and AMI and ncs together ( ). Toxicity Reversal in Cardiac Myocytes Influence of AMI on cardiac myocytes After the deposition of cardiac myocytes in physiologically active fashion on MEAs, our goal was to establish a dose response curve with respect to AMI concentration. AMI is long known to have a significant effect on the cardiovascular system above the therapeutic concentration, including the depression of myocardium and prolongation of conduction times.186, 187 Previous reports sugg ested that the primary target of AMI is blocking of a sodium ion channel in cardiac myocytes,188 at concentrations ~0.4 M.189 The influence of AMI on cardi ac myocytes with respect to spike frequency is shown on Figure 4-10. It is clear that the AMI at 0.4 M causes a 50% reduction in spike frequency, whereas only 30% of the initial activity is preserved at 1 M after 15 min. exposure.

PAGE 139

120 Figure 4-10. The Amitriptyline dose-response curve with resp ect to spike frequency of cardiac myocytes. (data acquired 15 min. after the drug addition) Preemptive detoxification studies on MEAs A step further in this study was the inve stigation of the ncs sequestering ability when the toxic drug is subseque ntly added to MEAs already containing cells and ncs. The comparison between cardiac myocytes activity in control, AMI alone and both, ncs and AMI conditions is shown on Figure 4-11. The cardiac myocyte activit y under the control conditions is constant, whereas an expected decrease is observed upon the addition of AMI at 1 M to about 30% of the initial frequency. However, if the ncs are present before the drug is added, the toxic e ffect of AMI is much less pr ofound. Although the activity is somewhat decreased under the preemptive conditions, the

PAGE 140

121 Figure 4-11. The preemptive detoxification of cardiac myocytes in the presence of nanocapsules, AMI was at 1 M: spike rate (uppe r) and spike amplitude (lower).

PAGE 141

122 magnitude of the decrease is much smaller (~70% of the initial activity) than in the case of AMI alone in the system. The detox potential of ncs is even more obvious if the spike amplitude is compared for the control, AMI and preemptive conditions . The control condition is characterized by constant spike amplitude over time. The addition of AMI results in a dramatic decrease of spike amplitude even after 2 min. exposure, wh ich is different than the spike rate effect. Moreover, once decreased, the spike amplitude remains at a constant level (~30%) of the initial value. Under the preemptive conditions, the spike amplitude of cardiac myocytes remains essentially the same as that of the control. The important thing to note is that cells remained unaffected within first 2 min. of exposure (time frame that AMI decreased the amplitude), suggesting that the uptake was instantaneous and quantitative. The tests with higher concentration of th e drug were also done to investigate the maximum detoxification ability of ncs. The sp ike rate and amplitude in the presence of 2 M AMI without the ncs are shown on Figure 4-12 ( in both graphs). The decrease in spike rate and the spike amplit ude is more profound than in 1 M AMI experiments as expected. Unfortunately, the effect of ncs in subsequent experiments was negligible for the first 10 min. of the exposure. However, after 10 min. exposure a clear indication of ncs action is manifested by 20% and 40% increase in spike rate , compared to control, for 10min. and 15min. points respectively. A similar observation was made for the spike amplitude, after the initial decrease, the spik e amplitude remains constant when ncs were present, while it keeps decreasing furthe r, when the drug alone was present. The difference in spike amplitude was 40% and >50% between the preemptive and drug alone experiments for 5min. and 10min. exposure respectively.

PAGE 142

123 Figure 4-12. The preemptive experiment with 2 M Amitriptyline: spike rate (left) and spike amplitude (ri ght)., control ( ), AMI alone ( ) and AMI and ncs together ( ). (* significance test, P<0.05) The highest attempted concentration was 5 M, an ultimately lethal concentration of AMI, manifested by a sharp and sudden decrease in both activity parameters even after 2 min. of exposure (Figure 4-13). Moreover, the cells that experienced such high

PAGE 143

124 concentration of the drug were not able to reco ver, even if they were flushed with freshly prepared cell medium and paced with artificial electric stimuli. Figure 4-13. The preemptive experiment with 5 M Amitriptyline: spike rate (left) and spike amplitude (ri ght), control ( ), AMI alone ( ) and AMI and ncs together ( ). In a preemptive set up, the spike rate sh arply decreased to ~30% of the initial value, but the important observa tion is that the spike rate was constant for an additional

PAGE 144

125 10min. of exposure. The spike amplitude quick ly decreased to about 20% of the initial value within 2min., followed by a slower but steady decrease to a bout 5% at 15 min. exposure. Similarly to spike rate, it is important to note that the cells were active even after 15 min. of 5 M AMI exposure. Although with a somewhat disappointing as pect, this experiment provided one additional proof of ncs therapeutic effect. The cells in a preemptive situation were washed with cell medium and the activity increased to the point of a full recovery after 24 hours. Real Time Detoxification Studies The ultimate goal in this study is to show the real-time detoxification potential of ncs. Since all drug poisoning cases or clinical mistreatments are situations when the toxic drug is already present in the body, it is cruc ial that the ncs are able to decrease the drug concentration in a timely fashion under these conditions. The results of real time detoxification with respect to spike rate a nd spike amplitude are shown on Figure 4-14. The first experiment was monitoring the spike rate when AMI alone was present, and F10 medium (10% of the total volume i.e. 0.1ml) was added after 10 min. The cells are then treated by multiple addition/removal of a fresh, drug free medium, to the point of full recovery. The second experiment included the addition of AMI, followed by an injection of concentrated ncs solution (10% volume, 0. 1ml, to the final 0.05w % of ncs) after 10 min. exposure. A sharp and instantaneous incr ease in cell activity was observed after the introduction of ncs, suggesti ng the decrease in co ncentration of AMI. Moreover, the difference in activity was ~30% between the AMI alone and detox medium after the first 2 min. of ncs introduction.

PAGE 145

126 Figure 4-14. The real time de toxification of cardiac myocytes by nanocapsules, spike rate (top) and spike amplitude (bottom) . The control conditions for spike amplitude are omitted for clarity. The spike amplitude under real time det ox conditions was also monitored (Figure 4-14 bottom). Although the therapeutic effect of the ncs is not as prof ound as in the case

PAGE 146

127 of spike rate (partially because of the large number of channels and amplitudes of different magnitude that make STD very large), the diffe rence between AMI alone and detox medium was observed. The maximum di fference is achieved 4 min. after the ncs introduction (~40%). Conclusions The work in this Chapter is related to th e application side of the whole project and describes the attempts to show the therapeutic potential of ncs. Furthermore, the toxicity profile of formulations is established in vivo (whole animal model) and in ex-vivo (isolated heart and cardiac myocytes). Simila rly to biomedical studies in simple media (i.e. blood and RBCs) described in Chapter 3, the most toxic species are surfactant micelles and soft oil filled microemulsion dr oplets. Although not studied in great detail, the probable origins of observed toxicity of these entities are high mobility, great solubilization power and the presence of a l ong alkyl chain of surfactant that can disrupt the cell membrane. This is the main reason why simple and easy to prepare low molecular weight surfactant solutions and mi croemulsions may not be suitable for these applications, despite the excellent scavenging profiles (Chapter 3). The purified formulations showed no apparent toxicity at certain concentrations. In more simple systems, like the isolated hear t and cardiac myocytes, the ncs were harmless at concentrations up to 0.5w%. The highe r concentrations caused the prolongation of QRS time (isolated heart) and decrease of spike frequency (cardiac myocytes), probably by blocking the AV node as suggested by isolat ed heart experiments. As expected, the harmless concentration of the ncs in the whole animal was significantly higher (1w%) than in tissue alone. However, an interesti ng (and unfortunate) re sult is the increased

PAGE 147

128 toxicity of the ncs and toxic drug mixture in the animal. It seems that AMI amplified the toxicity of the ncs in a mechanism not yet understood. The detoxification potential of the ncs was studied in whole animal and cardiac myocyte set ups. The former included only so called “preemptive” studies, meaning that ncs were already present in the moment of t oxic drug injection. The encouraging result is that the ncs decreased the QRS time compar ed with drug alone situation, up to 50%. Unfortunately, at this concentration of ncs, the animals experienced death in couple of days time. The two times lower concentra tion of ncs had a less profound therapeutic effect, suggesting that that th e decrease in QRS time was th e function of ncs present in the animal. Furthermore, diluted ncs were apparently harmless with respect to animal’s health. It is important to note that only PEO modified ncs were tested in vivo , while more efficient PEO and COOH modified ncs were omitted due to shortage in the project’s funding. It is reasonable to assume that more potent scavenging particles will require lesser therapeutic concentration, so that th e intrinsic toxicity might well be reduced. A more detailed study on ncs detox prof ile was done using cardiac myocytes deposited onto MEAs. The ncs presence in the system significantly diminished the toxic effect of AMI with respect to both, spik e rate and amplitude. The most encouraging feature of the preemptive experiments was ra pid and efficient action of nanocapsules on the MEAs. In case of high drug concentrations (i.e. 2 and 5 M), the ncs effect was less profound and slower than in the 1 M experiment. A very significant observation is that cells subjected to lethal drug concentrations were able to recover only if the ncs were present during the exposure. Moreover, we were able to proceed to a real time detoxification experiments based on thes e results. We designed the real time

PAGE 148

129 detoxification experiments in way to fully mimic the potential drug overdose,190 that is the drug was present at toxic level for 10 min. after the detoxification therapy was administered. The ncs improved the activity of cells already contaminated with toxic drug, especially with respect to spike ra te. The obvious difference between the two treatments happened 2 min. after the inject ion of ncs, showing promising sequestering ability with possible clin ical significance. .

PAGE 149

130 CHAPTER 5 CONCLUSIONS AND FUTURE WORK Microemulsions and Nano capsules Synthesis The formation of the microemulsions pr oved to be a feasible process in the presence of polymerizable alkylalkoxysilanes. It started as a scientific idea to extend the methodology from previously studied alkylalkox ysilane network (thin film) formation at the air/water interface (2D), to the oil/wate r interface of microemulsion droplets in the bulk (3D). Most of the work is related to incorporation of the Oc tadecyltrimethoxysilane (OTMS, 3) partially because the 2D study was done with this compound. The most important consequence of the li pophilicity of OTMS wa s the necessity of an external surfactant to succe ssfully formulate stable micr oemulsions. Moreover, all of the samples with OTMS had to be heated for period of time to self assemble in microemulsion droplets. The heating time scal ed up with the concen tration of OTMS. The external surfactants (most often 1 and 6) were chosen in accordance with US Food and Drug Administration (FDA) Guidelines to increase the overall biocompatibility profile of nanocapsules. A variety of oil phases were used in th is work, from very non-polar Hexadecane (2) to more polar and FDA approved Ethyl Butyrate (7). A general trend observed is that polar oils are easier to mix (i.e. shorter hea ting times), compared to non-polar oil phases. On the other hand, microemulsions formulated using the non-polar oil phases were more stable toward coalescence and creaming (i.e. more stable).

PAGE 150

131 Chronologically, formulations with a large excess of the external surfactant (i.e. high Surfactant-to-oil ratio) were initially synthesized.191 These formulations provided the synthetic proof of princi ple, and helped study the formation of microemulsion droplets starting from the unswollen surfactan t micelles by Quasi Elastic Light Scattering (QELS). However, their properties were ove rshadowed by the surfactant and were not studied further. After the initial toxicity and biomedical tests with the aforementioned formulations, the necessity of surfactant removal or decreas e in the concentration became the focus of this study. Therefore, formulations with lower surfactant-to-oil ratio (S/O) were synthesized (i.e. medium S/O ratio).192 These samples were studied in detail, more specifically the formulation aspects (i.e. the relative concentration of formulation ingredients) on properties like, droplet size, architecture and silica shell thickness. As expected, the increase in relativ e concentration of external surfactant lowered the droplet size, while the increase in o il concentration had the opposit e effect. A very important conclusion is that the all formulations w ithin a certain concentration window of the ingredients were stable at least for the time n ecessary to build the silica shell. This means that the droplets of different size can be templated to obtain the nanocapsules. The study of the silica shell formation showed that th e thickness is a function of sol-gel reagent (Tetramethoxysilane TMOS) concentration, th e number and size of the polymerizable templates. The latter implies that the sili ca shell formed around the smaller droplets will be thinner compared to bigger droplets (equal oil content!) at given concentration of the sol-gel reagent. These results led to the successful and reproducible nanocapsule syntheses in the range of 50 to 600nm in diameter.

PAGE 151

132 Although successfully synthesized, medium S/O samples still had a significant excess surfactant that had to be removed. Th erefore, further decrea se in the surfactant concentration was investigated (low S/O formulations). The apparent differences (stability most importantly) in behavior fr om previous two groups of microemulsions were the consequences of high oil concentra tion. The mechanism of droplet growth was consistent with the Ostwald ripening model. Se veral successful strategies to prevent the ripening were employed, like the increase in the OTMS concentr ation, addition of extremely non-polar oil and formulating the system with higher molecular weight surfactants. The OTMS limited the choice of the external surfactants to Brij or Tween type that were not ideally biocompatible as revealed by biomedical part of the study. Therefore, more soluble alkylalkoxysilanes were us ed in the formulations (Alternative formulations). The shorter alkyl tails of Octyltrimethoxysilane (OTS 16) and 3(trimethoxysilyl) methyl methacrylate (MTS 24) expanded the formulation options to various ionic and polymeric surfactants. The surface modification of nanocapsules wa s achieved by using the modifier (i.e. compounds 9-11) compounds bearing the trialkoxysila ne groups on one chain end and desired functionality (i.e. polymer, acid, amine) on the other. It is important to note that modification chemistry described in this work is just a glimpse of all possibilities to synthesize functiona l nanoparticles. Microemulsions and Nanocaps ules Synthesis-Future Work There are two main concerns with respect to synthetic work described in this study: a) the excess of surfactant, not only due to high toxicity, but also surfactant can direct the sol-gel synthesis towards fo rmation of other, undesired types of materials, like the solid silica nanoparticles for example.193-195

PAGE 152

133 b) it is very difficult and probably impossi ble to fully direct the condensation of TMOS on the microemulsion droplet surface, due to competing self condensation reactions The scheme of the latter pro cess is shown on Figure 5-1. Figure 5-1. Left-The competition between the self condensation (outcome (b) red) and condensation on the droplet surface (desired process (a) blue). Right from top: sol-gel rods, unidentified high aspect ratio material and solid silica nanoparticles. The formation of various undesired material s is illustrated on Figure 5-1 (right TEM images). Therefore, a strategy toward diminishing the aforementioned processes has to be devised if nanocapsules are to be used in a larger scale.

PAGE 153

134 Oil in Ethylene Glycol microemulsion One method to overcome these difficulties is to use Ethylene Glyc ol as a dispersion medium instead of water, i.e. to formulate Oil-in-Ethylene Glycol (O/EG) microemulsions. The benefits of this system include the direction of TMOS condensation solely to the surface of nanocapsules, EG is not nearly as good as a catalyst for the hydrolysis of TMOS as is water, therefore the reaction will be si gnificantly slower, and decrease in the population of su rfactant micelles, since the cr itical micelle concentration of surfactant is dramatically high er in EG compared to water. Figure 5-2. The decrease in amount of wate r necessary to obtain clear microemulsion (ME) with increase of lipoph ilicity of the oil phase (top left), the decrease in the amount of water to obtain clear ME with increase in HLB of surfactant mixture (bottom left) and TEMs of the core-shell nanocapsules using O/EG microemulsion as a template (left).

PAGE 154

135 The initial experimental efforts are shown on Figure 5-2. Unfortunately, the isotropic microemulsion can not be synthe isized using pure EG, however, upon titration with small amounts of water (a s low as 4-5w%) a clear and stable microemulsion can be formed (Figure 5-2). The amount of wate r necessary to produ ce optically clear microemulsion is the function of oil phase lipophilicty and the HLB number of surfactant mixture (Figure 5-2). This synthetic pathwa y should be studied in greater detail, especially since the EG can be easily replaced after the synthesis by water through dialysis. Double microemulsion technology. A second possible solution to the aforemen tioned problems is the formation of the nanocapsules using double microemulsi on technology. It is well known that monodisperse silica nanoparticles can be obt ained through the solgel condensation of Tetraethoxysilane (TEOS) in a water-in-oil (W/O) microemulsion.196 Figure 5-3. The scheme of the double microemu lsion (ME) process: (from left) 1) O/W ME is formed, 2) double O/W/O ME is formed, 3) sol-gel formation of the shell and 4) pyrolysis at elevated temperature to obtain hollow nanocapsules. The size of the nanoparticle is the functi on of water droplet size where the TMOS condensation takes place. The idea is to form oil-in-water (O/W) microemulsion droplets, as described in this work, and to conduct a polymerization of the “skin” layer to increase the stability of these structures. Thus, prep ared oil templates will be added to an oil

PAGE 155

136 solution, to form the water-in-oil microemulsi on. Therefore, a double, i.e. oil-in-water-inoil (O/W/O) microemulsion is obtained. Figure 5-4. The core-shell nanocapsules synt hesized using double microemulsion method (top row) and hollow nanocapsules obtaine d after Thermogravimetric analysis (TGA). (a) particles at 100 nm, (b-d) particles at 70nm, (e) 70 nm particles after TGA, (f) 100 nm part icle after TGA (the inner part was 30 nm in each case)

PAGE 156

137 The main advantage of this system is that the TMOS condensation will be restricted to nanometer sized water droplets (“wa ter pools”) containi ng the previously synthesized oil templates. Moreover, excess surfactant after the formation of the O/W microemulsion will be easily removed, i.e. solubilized in the oil phase when double microemulsion is formed. The purification is simple, the ethanol is added and the nanocapsules are centrifuged and redispersed in water. The dried particles revealed less electron-dense holes in the middle of the pa rticles. The size of the holes was 10-15nm, while the total size was 100+/-7nm (Figure 5-4 a) and 71+/-5nm (F igure 5-4 b-d). The TEMs of particles after the TGA (Figure 5-4 e, f) revealed the le ss electron-dense holes of 25+/-4nm, that is in a fair agreement with the measured size of oil droplets by QELS (~30nm)! Additional characteri zation experiments The most important future experi ments toward bette r understanding of microemulsion and nanocapsule behavior is th e investigation of the individual component dynamics. In other words, the diffusion pattern of the surfactant and oil molecules, i.e. whether they stay inside the core and how l ong. Furthermore, the porosity of the shell and its dependence on shell thickness are also essen tial for practical and fundamental reasons. It is very difficult to obtain molecular we ight and true architec ture of poly (OTMS) molecules presumably formed at the ME oil/w ater interface. Therefore experiments with dimethoxy polymerizable surfactants should be done in order to c onfirm the postulated mechanism of polymerization and the presen ce of unreacted silanols that serve as “anchoring points” for the silica shell formation.

PAGE 157

138 Biomedical Characterization and Drug Detoxification Probably the most detailed part of the study is the relation of the surface chemistry and the biomedical behavior of nanocapsules.197 The benefits of surface modification with poly (ethylene oxide) are increased resist ance toward aggregation and stability in water solutions. Moreover, in creased biocompatibility of the PEO modified ncs is demonstrated by hemolysis and Thromboels tography (TEG) assays. The particle size effect, that is, the increased t oxicity of the ncs with an incr ease in the size observed with non-modified ncs was diminished by the presence of PEO. The absence of any surfactant micelles or soft microemulsion droplets was crucial for the application of the ncs in detoxificati on tests. The reasons for such high toxicity were not in the focus of this study. Intuit ively, small size, dynam ic nature and great solubilization power of the surfactant micelles are some of the factors contributing to the observed behavior. The uptake profile was also greatly infl uenced by the presence of surfactant micelles. Micelles and soft microemulsion droplets were by far the best scavenging entities tested in this work for the same reasons as mentioned earlier. Despite these difficulties, the uptake potential of oil cores as a function of the ingredient concentrations was investigated. The uptake capability of the oil cores is determined by the size, i.e. the amount interfacial area. These experiments s uggested that the guest molecules partition more likely at the oil/water interf ace, than in the oil reservoirs. The experiments with nanocapsules reveal ed somewhat different mechanism of uptake. The electrostatic inter actions are the driving force for the removal of the toxic drug. Furthermore, presence of a modifier on the surface that enhances this interaction (i.e. carboxylic acid) led to prof ound increase in uptake ability.

PAGE 158

139 The toxicity of nanocapsules was studied in isolated heart, cardiac muscle cells plated on the Microelectrode arrays (MEAs) and in vivo . In most cases, at medium concentrations (e.g. 2-5mg/ml) no a pparent toxicity was observed. It is somewhat surprising that ncs were able to decrease the concentration of the toxic drug in vivo , having in mind the disappointing upt ake profile in human plasma and whole blood. Unfortunately, due to high costs of these experiments, the quantification of the therapeutical effects was not concluded. Th e proof of principle i nvestigation of ncs as the potential detox agents was nonetheless achieved on cardiac myocytes plated on MEAs. The detoxification was demonstrated by the preservation of cell activity in preemptive conditions, when both ncs and th e drug were added simultaneously. More importantly, the nanocapsules showed therapeutic potential by increasing the cells activity parameters, when injected in cell suspensions 10 min. after the drug. Biomedical Characterization and Drug Detoxification-Future Work It was already mentioned that the dr ug uptake experiments in blood plasma (Chapter. 3) failed to give a clear indicat ion of antidote action of nanocapsules. The reason for this outcome is a high affinity of AMI to bind to immunoproteins in the blood stream, more specifically 1-acid glycoprotein (AAG) that is responsible for the removal of 91-95% of AMI from the bloodstream.198, 199 Henry et. al.200 used the aforementioned property of AAG to demonstrate the AMI drug reversal in isolated cardiac myocytes by patch clamp methods. Another strategy toward drug det oxification using unmodified biomolecules is described in the work of Brunn et. al.201 where the drug specific antibody fragment (F (ab’)2) was used to lower the concentration of desipr amine (toxic drug) in rats.

PAGE 159

140 Silica nanoparticles with encapsulated 1-acid glycoprotein Figure 5-5. The scheme of silica nanopart icles with AAG protein formation using W/O method As already mentioned, the formation of monodispersed silica nanoparticles is

PAGE 160

141 conveniently done through the waterin-oil (W/O) microemulsion method. Figure 5-6. The silica nanoparticles w ith embedded AAG of different size. The idea for future research is to use the 1-acid glycoprotei n (AAG) aqueous solution instead of water in the method, t hus producing the nanoparticles with protein

PAGE 161

142 embedded or encapsulated inside (Figure 5-5) . It is reasonable to assume that these particles would be efficient drug scav engers even in the bloodstream. The advantages of this technology include the ease particle size control by changing the Surfactant-to-water ratio (shown on Figure 5-6) , convenient purification and possibility of in situ surface modifications. Since AAG is negatively char ged protein (pI~2), an am ino based alkylalkoxysilane (3-aminopropyltrimethoxysilane) was added before TMOS/TEOS in attempt to obtain the core-shell nanoparticles. Figure 5-7. The morphologies of nanoparticles obtained w ith low (upper row) and high concentration of 3-aminopropyl trimethoxysilane (bottom row). The presence of protein in particles, as well as modification of the surface, can be investigated by the IR spectroscopy.

PAGE 162

143 Figure 5-8. The IR spectra of (from t op) neat AAG, bare Si nps and AAG@ nps. The preliminary measurements in saline suggest that the presence of AAG significantly improved the uptake prop erties of Si nanoparticles.

PAGE 163

144 Figure 5-9. The comparison of uptake effici ency of nanoparticles with and without AAG (left), and two populations of nanoparticle s with different concentrations of AAG (right). Additional Experiments Unfortunately, the complex and expensive experiments of in vivo detoxification assessment and biodistribution could not be done in this work. It would be interesting to see the difference (if any!) in the biological fate of nanocapsules with different surface chemistries. The outcome of these experiment s would likely decide the future research direction, i.e. whether they are suitable fo r biosensors, imaging, cell adhesion and drug delivery. More detailed study should be done to understand the precise nature of the interaction between the nanocapsules and the guest molecules. The dynamics of the guest molecule, once captured by the ncs would be the most interesting scientific topic in this regard. Analytical techniques as NMR Diffusiometry, Cyclic Voltametry and Fluoroscence Spectroscopy should be used in or der to study the stability of the assembly and the kinetics of guest molecule encapsulati on. These findings would lead to changes in the synthesis procedure to obtain mo re efficient scavenging systems.

PAGE 164

145 The application of PLGA copolymers in the synthetic proces s is one of the alternatives for the forma tion of nanocapsules from biodegradable and biocompatible materials. Furthermore, the nanocapsules synt hesis should be very reproducible (5-10% standard deviation) in order to meet st rict US FDA requirements for the clinical application. This study summarizes the attempts to understand the possibility of using crosslinkable microemulsions for the template synthesis of functional nano-objects. The future challenges are outlined and possible scientific solutions are described. I hope this work will serve as a basis for further research directed toward practical and large scale usage in a real life, since this was my goa l from the point I st arted working on this project.

PAGE 165

146 APPENDIX A CHEMICAL STRUCTURES OF COMP OUNDS USED IN THIS WORK Figure A-1. The chemical struct ures of compounds used as su rfactants and co-surfactants in this work.

PAGE 166

147 Figure A-2. The chemical stru ctures of compounds used as oil phases and model drugs for the uptake studies. Figure A-3. The chemical structures of cr osslinkable surfactants, sol-gel agent and surface modifiers.

PAGE 167

148 Figure A-4. The scheme of nanocapsules s ynthesis (medium Surfactant-to-Oil ratio), Chapter 2.

PAGE 168

149 APPENDIX B INFRARED (IR) SPECTROSCOPY CHARCTERIZATION OF SURFACE MODIFIED NANOCAPSULES Figure B-1. The IR spectra of sili ca and PEO modified nanocapsules.

PAGE 169

150 Figure B-2. The IR spectra of PEO and CO OH modified (upper) and COOH alone silica nanocapsules.

PAGE 170

151 Figure B-3. The IR spectrum of amin o modified silica nanocapsules.

PAGE 171

152 APPENDIX C LIST OF ABBREVIATIONS 1. AMI-Amitriptyline 2. EB-Ethyl Butyrate 3. Tw-80-Tween-80 4. ME-microemulsion 5. S/O-Surfactant-to-oil ratio 6. PS/O-Polymerizable Surfactant-to-oil ratio 7. QELS-Quasi Elastic Light Scattering 8. TEM-Transmission Electron Microscopy 9. SEM-Scanning Electron Microscopy 10. O.C.-Oil cores 11. Ncs-nanocapsules 12. Nps-Nanoparticles 13. PEO-Poly (ethylene oxide) 14. TMS-Trimethoxysilane 15. OTMS-Octadecyltrimethoxysilane 16. TEG-Thromboelastography 17. RBC-Red blood cells 18. C.P.-Cloud point 19. HLB-Hydrophilic-L ypophilic balance 20. SA-Sinoatrial node

PAGE 172

153 21. AV-Atrioventricular 22. Cmc-Critical micelle concentration 23. Cpp-Critical Packing Parameter 24. HPLC-High Performance Liquid Chromatography 25. O/W-Oil in water microemulsion 26. W/O-Water in oil microemulsion 27. AAG1-acid glycoprotein 28. O/EG-Oil in Ethylene Glycol microemulsion

PAGE 173

154 LIST OF REFERENCES 1. Dickey, F. H., Proc. Natl. Acad. Sci. 1949, 35 , 227 2. Beck, J. S. V., J.C.; Kennedy, G.J.; Kresge, C.T.; Roth, W.J.; Schramm, S.E. , Chem. Mater. 1994, 6 , 1816. 3. Kohli, P. H., C.C.; Cao, Z.; Gasparac, R.; Tan, W.; Martin, C.R, Science 2004, 305 , 5686. 4. Zheng, F. L., L.; Gao, Y.; Sukamto, J.H.; Aardahl, C.L. , Nano Lett. 2002, 2 , 729 5. Adachi, M. H., T.; Harada, M. , Langmuir 2000, 16 , 2376. 6. Caruso, F. C., R.A.; Mohnwald, H. , Science 1998, 282 , 1111. 7. Raman, N. K. A., M.T.; Brinker, C.J . , Chem. Mater. 1996, 8 , 1682. 8. Safran, S., Micelles, Membranes, Microemulsions and Monolayers . SpringerVerlag: New York, 1994. 9. Schmidt-Winkel, P. G., C.J.; Stucky, G.D . , Langmuir 1999, 16 , 356. 10. Winsor, P. A., Trans. Faradey. Soc 1948, 44, 376. 11. Saito, H. S., K. , J. Coll. And Interface Sci. 1968, 26, 70. 12. Van Nieuwkoop, J. S., J . , J. Colloid Interface Sci. 1985, 103, 417. 13. Kahlweit, M. S., R. , Angew. Chem. Intl. Ed. Engl. 1985, 24, 654. 14. Bauer., D., Anal. Chem. 1983, 55, 662. 15. Schambil, F. S., M. , J. Colloid. Polym. Sci. 1987, 265, 1009. 16. Bockelen, A. N., R., Anal. Bioanal. Chem. 1993, 346, 435. 17. Butt, H.-J. G., K.; Kappl, M. , Physics and Chemistry of Interfaces . 2nd ed.; Wiley: New York, 2003. 18. Tadros, T. I., P.; Esquena, J.; Solans, C. , Adv. Coll. Interface Sci. 108-109, 303. 19. Kostarelos, K., Adv. Coll. Interface Sci. 2003, 106, 147.

PAGE 174

155 20. Kreilgaard, M., Adv. Drug Delivery Rev. 2002, 54, S77. 21. Heydenreich, A. V. W., R.; Pedersen, N.; Poulsen, H.S.; Kristensen, H.G. , Int. J. Pharm. 2003, 1, 83. 22. Watnasirichaikul, S. R., T.; Tucker, T.G.; Davies, N.M. , Int. J. Pharm. 2002, 1-2, 237. 23. Glenn, G. M. R., M.; Matyas, G.R.; Alving, C.R. , Nature 1998, 391, 581. 24. Cui, Z. M., R.J. , J. Controlled Release 2002, 81, 173. 25. Marsh, A. C., B.; Broderick, M.; Power, J.; Donegan, S.; Altria, K. , Electrophoresis 2004, 25, 3970. 26. Klotz, W. L. S., M.R.; Foley, J.P. , Journal of Chromatography 2001, 930, 145. 27. Holmberg, K., Curr. Opinion Coll. Interface Sci. 2003, 8, 187. 28. Erlington, A. R.; Menger, F. M., J. Am. Chem. Soc. 1991, 113, 9621. 29. Israelachvili, J. N., Intermolecular and Surface Forces . 2nd ed.; Academic Press: New York, 1991. 30. Patist, S. G.; Oh, R.; Leung, R.; Shah, D. O., Coll. And Surf. A:Physicochem. Eng. Aspects 2001, 176, 3. 31. Raju, B. B.; Costa, S. M. B., Phys. Chem. Chem. Phys. 1999, 1, 5029. 32. Messaoud, T.; Duplatre, G.; Michels, B. ; Waton, G. J. P. C. B., 108, 13137. , J. Phys. Chem. B 2004, 108, 13137. 33. Saito, H.; Shinoda, K., J.Coll. Interface Sci. 1968, 26, 70. 34. Saito, H.; Shinoda, K.; , J.Coll. Interface Sci. 1968, 26, 10. 35. Wen, J.; Wilkes, G. L., Chem.Mater. 1996, 8, 1667. 36. Sontjens, S. H. M.; Sijbesma, R. P.; van Genderen, M. H. P.; Meijer, E. W., J. Am. Chem. Soc. 2000, 122, 7487. 37. Boas, U.; Karlsson, A. J.; de Waal, B. F. M.; Meijer, E. W. J. O. C., 66, 2136., J. Org. Chem. 2001, 66, 2136. 38. Corbellini, F.; Mulder, A.; Sartori, A.; Ludden, M. J. W.; Casnati, A.; Ungaro, R.; Huskens, J.; Crego-Calama, M.; Reinhoudt, D. N., J. Am. Chem. Soc. 2004, 126, 17050. 39. Palmer, L. C.; Rebek, J., Org. Biomol. Chem. 2004, 2, 3051.

PAGE 175

156 40. Kishikawa, K.; Tsubokura, S.; Kohmoto, S.; Yamamoto, M., J. Org. Chem. 1999, 64, 7568. 41. Piatnitski, E. L.; Flowers, R. A.; Deshayes, K., Chem. Eur. J. 2000, 6, 999. 42. Palmer, L. C.; Ribek, J. J., Org. Biomol. Chem. 2004, 2, (3051). 43. Sijbesma, R. P.; Meijer, E. W., Curr. Op. Coll. Surf. Sci. 1999, 4, 24. 44. Hiemenz. P.C.; Rajagopalan, R., Principles of Surface and Colloidal Chemistry . 3rd ed.; Marcel Dekker, Inc.: New York, 1997; p 6. 45. Morey, T. E.; Varshney, M.; Flint, J. A.; Rajasekaran, S.; Shah, D. O.; Dennis, D. M. N. L., 4, 757., Nano. Lett. 2004, 4, 757. 46. Jung, T.; Kamm, W.; Breitenbach, A.; Kaiserling, E.; Xiao, J. X.; Kissel, T., Eur. Jour. Pharm. Biopharm. 2000, 50, 147 47. Brenner, B. M.; Rector, F. C., The Kidney . Philadelphia, 1991; Vol. 1. 48. He, J. L.; Chen, B.; Flatt, A. K.; Stephenson, J. J.; Doyle, C. D.; Tour, J. M., Nat. Mater. 2006, 1, 63. 49. Sanchez, C.; Julian, B.; Belleville, P.; Popall, M., J. Mater. Chem. 2005, 15, 3559. 50. Schenning, A. P. H. J.; Meijer, E. W.; , Chem. Comm. 2005, 26, 3245. 51. Gordon, E. M.; Hall, F. L., Oncology Reports 2005, 13, 1003. 52. Ai, H.; Jones, S. A.; Lvov, Y. M., Cell Biochem. Biophys. 2003, 39, 23. 53. Kobayashi, H.; Kawamoto, S.; Sakai, Y.; Choyke, P. L.; Star, R. A.; Brechbiel, M. W.; Sato, N.; Tagaya, Y.; Mo rris, J. C.; Waldmann, T. A., Journal of the National Cancer Institute 2004, 96, 703. 54. Simon, P. F. W.; Ulrich, R.; Spiess, H. W.; Wiesner, U., Chem. Mater. 2001, 13, 3464. 55. Hammer, B.; Norskov, J. K.; Adv. Catal. 2000, 71., Adv. Catal. 2000, 45, 71. 56. Andres, R. P.; Bein, T.; Dorogi, M.; Fe ng, S.; Jenderson, J. I.; Kubiak, C. P.; Mahoney, W.; Osifchin, R. G.; Reifenverger, R., Science 1996, 272, 1323 1996, 272, 1323. 57. Bonnermann, H.; Braun, G.; Brijoux, G. B.; Brinkman, R.; Tilling, A. S.; Schulze, S. K.; Siepen, K., J. Organomet. Chem. 1996, 520, 143. 58. Mayer, A. B. R., Polym. Adv. Technol. 2001, 12, 96.

PAGE 176

157 59. Kralik, M.; Biffis, A., J. Mol. Catal. A : Chem. 2001, 177, 113. 60. Thomas, J. M.; Johnson, B. F. G.; Raja, R. S., G.; ; Midgley, P. A., Acc. Chem. Res. 2003, 36, 20. 61. Poppa, H., Vacuum, 1984, 34, 1081. 62. Raimondi, F.; Scherer, G. G.; Kotz, R.; Wokaun, A., Angew. Chem. Int. Ed 2005, 44, 2190 63. Landsman, D. A.; Luczak, F. J., Handbook of Fuel Cells . Wiley: New York, 2003; p 821. 64. Kanel, S. R.; Manning, B.; Charlet, L.; Choi, H., Environ. Sci. Technol. 2005, 39, 1291. 65. Li, P.; Miser, D. E.; Rabiei, S.; Yadav, R. T.; Hajaligol, M. R., App. Catal. B : Environ. 2003, 43, 151. 66. Hudson, M. J.; Jackson, D. B.; Ward, J. L.; Chinn, M. J., Chem. Comm. 2003, 2970. 67. Kim, J.-Y.; Cohen, C.; Shuler, M. L.; Lion, L. W., Environ. Sci. Technol. 2000, 34, 4133. 68. Jianrong, C.; Yuding, M.; Nongyue, H.; Xiaohua, W.; Sijiao, L., Biotech. Adv. 2004, 22, 505. 69. Tolles, W. M.; Rath, B. B., Curr. Science 2003, 85, 1746. 70. Emerich, D. F.; Thanos, C. G., Curr. Exp. Biol. Ther. 2003, 3, 655. 71. Chan, W. C.; Maxwell, D. J.; Gao, X.; Bailey, R. E.; Han, M.; Nie, S., Curr. Opinion Biotech. 2002, 13, 40. 72. Weizmann, Y.; Patolsky, F.; Lioubashevski, O.; Willner, I., J. Am. Chem. Soc. 2004, 126, 1073. 73. El-Sayed, I. H.; Huang, X. H.; El-Sayed, M. A., Nano. Lett. 2005, 5, 829. 74. Zhao, X.; Bagwe, R. P.; Tan, W., Adv. Mater. 2004, 16, 173. 75. Zhao, X.; Tapec-Dytioco, R.; Tan. W. J. Am. Chem. Soc. 2003, 11474; , J. Am. Chem. Soc. 2003, 125, 11474. 76. Xiao, Y.; Pavlov, V.; Levine, S.; Niazov, T.; Markovitch, G.; Willner, I., Angew. Chem. Int. Ed 2004, 43, 4519 77. Dubertet, B.; Calame, M.; Libehaber, A. J., Nat. Biotechnol. 2001, 19, 365.

PAGE 177

158 78. Patolsky, F.; Gill, R.; Weizmann, Y.; Mokari, T.; Banin, U.; Willner, I., J. Am. Chem. Soc. 2003, 125, 13918. 79. De la Fuente, J. M.; Eaton, P.; Barrientos, A. B.; Menendez, M.; Penades, S., J. Am. Chem. Soc. 2005, 127, 6194. 80. Du, D.; Liu, S.; Chen, J.; Ju, H.; Lian, H.; Li, J., Biomaterials 2005, 26, 6487. 81. Oyewumi, M. I.; Liu, S. Q.; Moscow, J. A.; Mumper, R. J., Bioconj. Chem. 2003, 14, 404. 82. Krol, S.; Diaspro, A., IEEE Trans. Nanobiosci. 2004, 3, 32. 83. Hersel, U.; Dahmen, C.; Kessler, H., Biomaterials 2003, 24, 4385. 84. Trentin, D.; Hall, H.; Wechsler, S.; Hubbell, J. A., Proc. Nat. Acad. Sci. 2006, 103, 2506. 85. Ito, A.; Ino, K.; Hayashida, M.; Kobayashi, T.; Matsunuma, H.; Kagami, H.; Ueda, M.; Honda, H., Tissue Engineering 2005, 11, 1553. 86. Langer, R.; Peppas, N.; Frechet, J., C&EN 2002, 39. 87. Langer, R.; Peppas, N., AICHE Jour. 2003, 49, 2990. 88. Dr. Donn M. Dennis, U. o. F., Departme nt of Anesthesiology: , Query of the ASA Closed Claims Project Data base located at the Univer sity of Washington in 2001. 89. Buckley, N. A.; Whyte, I. M.; Dawson, A. H., Med. J. Aust. 1995, 164, 190. 90. Obafunwa, J. O.; Busutill, A., Hum. Exp. Toxicol. 1994, 13, 401. 91. Coleridge, J.; Cameron, P. A.; Drummer, O. H., Med. J. Aust. 1992, 157, 459. 92. Henry, J. A.; Alexander, C. A.; Sener, E. K., Brit. Med. Jour. 1995, 310, 221. 93. Buckley, N. A.; Dawson, A. H.; Whyte, I. M., Lancet 1994, 343, 159. 94. Lee, L. A.; Posner, K. L.; Domino, K. B.; Caplan, R. A.; Cheney, F. W., Anesthesiology 2004, 101, 143. 95. Kerr, G. W.; McGuffie, A. C.; Wilkie, S., Emerg. Med. J. 2001, 18, 236. 96. Jarvis, M. R., Psychopharmacol. Bull. 1991, 27, 541. 97. Marshall, J. B.; Forker, A. D., Am. Heart J. 1982, 103, 401. 98. Taylor, D. J. E.; Braithwaite, R. A., Br. Heart J. 1978, 40, 1005.

PAGE 178

159 99. Thorstrand, C., Acta. Med. Scand. 1976, 199, 337. 100. Renehan, E. R.; Enneking, F. K.; Varshney, M.; Partch, R.; Dennis, D. M.; Morey, T. E., Reg. Anesth. Pain. Med. 2005, 30, 380. 101. Bosse, G. M.; Barefoot, J. A.; Pfeifer, M. P.; Rodgers, G. C., J. Emerg. Med. 1995, 13, 203. 102. Hoffman, J. R.; Votey, S. R.; Bayer, M.; Silver, L., Am. J. Emerg. Med. 1993, 11, 336. 103. Koppel, C.; Wiegreffe, A.; Tenczer, J., Hum. Exp. Toxicol. 1992, 11, 458. 104. Weinberg, G. L., Reg. Anesth. Pain. Med. 2002, 27, 568. 105. Cho, H. S.; Lee, J. J.; Chung, I. S.; Shin, B. S.; Kim, J. A.; Lee, K. H., Anesth. Analg. 2000, 91, 1096. 106. Ohmutra, S.; Ohta, T.; Yamamoto, K.; Kobayashi, T. A., Anesth. Analg. 1999, 88, 155. 107. Weinberg, G. L.; VadeBoncouer, T.; Ramaraju, G. A.; Garcia-Amaro, M. F.; Cwik, M. J., Anesthesiology 1998, 88, 1071. 108. Varshney, M.; Morey, T. M.; Shah, D. O.; Flint, J. A.; Moudgil, B. M.; Seubert, C. N.; Dennis, D. M., J. Am. Chem. Soc. 2004, 126, 5108. 109. Lee, D. W.; Baney, R. H., Biomacromolecules 2004, 5, 1310. 110. Lee, D. W.; Flint, J. A.; Morey, T. E. ; Dennis, D. M.; Partch, R.; Baney, R. J., J. Pharm. Sci. 2005, 94, 373. 111. Kim, D.; El-Shall, H.; Dennis, D.; Morey, T., Coll. Surf. B:Biointerfaces 2005, 40, 83. 112. Kohli, P. M., C.R. Nature Rev. 2003, 2, 29, Nature Rev. 2003, 2, 29. 113. Hierrezualo, J. M.; Aguiar, J.; C., C. R., Langmuir 2004, 20, 10419. 114. Penfold, J.; Staples, E.; Thompson, L.; Tu cker, I.; Hines, J.; Thomas, R. K.; Lu, J. R.; Warren, N., J. Phys. Chem. B 1999, 103, 5204. 115. Wydro, P.; Paluch, M., Coll. Surf. A 2004, 245, 75. 116. Griffin, W. C., J. Soc. Cosmet. Soc 1949, 1, 311. 117. Izquiredo, P.; Feng, J.; Esquena, J.; Tadros , T. F.; Dederen, J. C.; Garcia, M. J.; Azemar, N.; Solans, C., J. Coll. Surf. Sci. 2005, 285, 388.

PAGE 179

160 118. Carino, S. R.; Tostmann, H.; Underhill, R. S.; Logan, J.; Weerasekera, G.; Culp, J.; Davidson, M.; Duran, R. S., J. Am. Chem. Soc. 2001, 123, 767. 119. Carino, S. R.; Underhill, R. S.; Tost mann, H.; Skolnik, A. M.; Logan, J. L.; Davidson, M. R.; Culp, J. T.; Duran, R. S., Langmuir 2003, 19, 10514. 120. Shinoda, K.; Arai, J., J. Phys. Chem. B 1964, 68, 3485. 121. Porras, M.; Solans, C.; Gonzales, C.; Ma rtinez, A.; Guinart, J. M. G., J.M.;, Coll. And Surf. A:Physicochem. Eng. Aspects 2004, (249), 115. 122. Liu, S.; Weaver, J. V. M.; Tang, Y.; Billingham, N. C.; Armes, S. P., Macromolecules 2002, 35, 6121. 123. Carino, S. R.; Duran, R. S., Macromol. Chem. Phys. 2005, 206, 83. 124. Fontaine, P.; Goldmann, M.; Rondelez, F., Langmuir 1999, 15, 1348. 125. Lindn, M.; Slotte, J. P.; Rosenholm, J. B., Langmuir 1996, 12, 4449. 126. Warisnoicharoen, W.; Lansley, A. B.; Lawrence, M. J., Int. Jour. Pharm. 2000, 198, 7. 127. Kawakami, K.; Yoshikawa, T.; Moroto, Y.; Kanaoka, E.; Takahashi, K.; Nishihara, Y.; Masuda, K. J., J. Controlled Release 2002, 81, 65. 128. El-Shabouri, M. H., Stp. Phar. Sci. 2002, 12, 97. 129. Ichikawa, S.; Sugira, S.; Naka jima, M.; Sano, M.; Furusaki, S., Biochem. Eng. Jour. 2000, 6, 193. 130. In Food and Drug Administration (FDA) Center for Food Safety and Applied Nutrition (CFSAN). 131. Hillyer, J. F.; Albrecht, R. M., J. Pharm. Sci. 2001, 90, 1927. 132. Dauty, E.; Remy, J.-S.; Blessing, T.; Behr, J.-P., J. Am. Chem. Soc. 2001, 123, 9227. 133. Mele, S.; Murgia, F.; Caboi, F.; Monduzzi, M., Langmuir 2004, 20, 5241. 134. Seddon, J. M., Biochim. Biophys. Acta 1990, 1031, 1. 135. Safinya, C. R., Coll. And Surf. A:Physicochem. Eng. Aspects 1997, 128, 183. 136. Gruner, S. M., Annu. Rev. Biophys. Chem. 1985, 14, 211. 137. Schwuger, M. J.; Stickdorn, K.; Schoemaker, R., Chem. Rev. 1995, 95, 849.

PAGE 180

161 138. Tatsumi, Y.; Masanori, T.; Hiroyuki, K. J., J. Chem. Eng. Jap. 2003, 36, 963. 139. Izquierdo, P.; Esquena, J.; Tadros, T. F. ; Dederen, C.; Garcia, M. J.; Azemar, N.; Solans, C., Langmuir 2002, 18, 26. 140. Chanana, D. C. S., B.B. , J. Phar. Sci. 1995, 49, 71. 141. Kunieda, H.; Ohyama, K., J.Coll. Int. Sci. 1990, 136, 432. 142. Shinoda, K.; Araki, M.; Sadaghiani, A.; Khan, A.; Lindman, B., J. Phys. Chem. B 1991, 95, 989. 143. Tadros, T.; Izquierdo, P.; Esquena, J.; Solans, C., Adv. Coll. Int. Sci. 2004 2004, 108-109, 303. 144. Caruso, R. A.; Antonietti, M., Chem. Mater. 2001, 13, 3272. 145. Magbitang, T.; Lee, V. Y.; Cha, J. A.; Wang, H. L.; Chung, R.; MIller, R. D.; Dubois, G.; Volksen, W.; Kim, H. C.; Hedrick, J. L., Angew. Chem. Int. Ed 2005, 44, 7574. 146. Chang, J. H.; Kim, J. A., J. Sol-Gel Sci. Tech. 2005, 33, 15. 147. Rhodes, K. H.; Davis, S. A.; Caruso, F.; Zhang, B.; Mann, S., Chem. Mater. 2000, 12, 2832. 148. Shenoy, D. B.; Antipov, A. A.; Sukhorukov, G. B.; Mohvald, H., Biomacromolecules 2003, 4, 265. 149. Nann, T.; Mulvaney, P., Angew. Chem. Int. Ed 2004, 43, 5393. 150. Ung, T.; Liz-Marzan, L. M.; Mulvaney, P., Langmuir 1998, 14, 3740. 151. May, S. R.; Davis, S. A.; Mann, S., Langmuir 2000, 16, 1454. 152. Ung, T.; Liz-Marzan, L. M.; Mulvaney, P., J.Phys. Chem.B 1999, 103, 6770. 153. Vestal, C. R.; Zhang, Z. J., Nano Lett. 2003, 3, 1739. 154. Tissot, I.; Novat, C.; Lefebvre, F.; Bourgeat-Lami, E., Macromolecules 2001, 34, 5737. 155. Tissot, I.; Reymond, J. P.; Lefebvre, F.; Bourgeat-Lami, E., Chem. Mater. 2002, 14, 1325. 156. Park, M. J. C., Y.C.; Chun, C.B., Coll. And Surf. B:Biointerfaces 2003, 32, 11. 157. Imanidis, G.; Hartner, K. C.; Mazer, N. A., Int. J. Pharm. 1995, 120, 41.

PAGE 181

162 158. Mossman, B. T.; Churg, A., Am. J. Respir. Crit. Care 1998, 157, 1666. 159. Silica, Some Silicates Coal Dust and Para-aramid Fibrils. Monograph on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. In IARC Press: 1997; Vol. 68. 160. Hench, L. L., J. Am. Ceram. Soc. 1991, 81, 1497. 161. Hench L.L.; Paschall, H. H., J. Biomed. Mater. Res. 1974, 5, 49. 162. Fenoglio, I.; Croce, A.; Di Renzo, F.; Tiozzo, R.; Fubini, B., Chem. Res. Toxicol. 2000, 13, 489. 163. Johnston, C. J.; Driscoll, K. E.; Finkelstein, J. N.; Baggs, R.; O’Reilly, M. A.; Carter, J.; Gelein, R.; Oberdorster., Tox. Sci. 2000, 56, 405. 164. Schins, R. P. F.; Duffin, R.; Hohr, D.; Knaapen, A. M.; Shi, T.; Weisaupt, C.; Stone, V.; Donaldson, K.; Borm, P. J. A., Chem. Res. Toxicol. 2002, 15, 1166. 165. Ayers, M. R.; Hunt, A. J., J. Non-Crys. Sol. 2001, 285, 123. 166. Depasse, J.; Warlus, J., J. Coll. Int. Sci. 1976, 56, 618. 167. Selvan, S. T.; Tan, T. T.; Ying, J. Y., Adv. Mater. 2005, 17, 1620. 168. Kirchner, C.; Liedl, T.; Kudera, S.; Pellegrino, T.; Ja vier, A. M.; Gaub, H. E.; Stolze, S.; Fertig, N.; Paraj, W. J., Nano. Lett. 2005, 5, 331. 169. Morey, T. E.; Varshney, M.; Flint, J. A.; Seubert, C. N.; Smith, W. B.; Bioraker, D. G.; Shah, D. O.; Dennis, D. O., J. Nano. Res. 2004, 6, 159. 170. Mosqueira, V. C. F.; Legrand, P.; Gulik, A.; Bourdon, O.; Gref, R.; Labarre, D.; Barratt, G., Biomaterials 2001, 22, 2967. 171. Kenausis, G. L.; Voros, J.; Elbert, D. L.; Huang, N.; Hofer, R.; Ruiz-Taylor, L.; Textor, M.; Hubbell, J. A.; Spencer, N. D., J. Phys. Chem. B 2000, 104, 3298. 172. Goodman & Gilman's the pharmacol ogical basis of therapeutics . McGraw-Hill: New York, 2001; p 1714. 173. Ding, J. L., G. , J. Phys. Chem. B 1998, 102, 6107. 174. Plucinski, P.; Nitsch, W., J. Phys. Chem. B 1993, 97, 8983. 175. Ktistis, G. J., Disper. Sci. Technol. 1997, 18, 49. 176. Varshney, M.; Flint, J. A.; Morey, T. E.; Dennis, D. M., In Shands Hospital: 2003.

PAGE 182

163 177. Morey, T.; Martynyuk, A. E.; Napolitano, C. A.; Raatikainen, M. J.; Guyton, T. S.; Dennis, D. M., Anesthesiology 1997, 87, 1172. 178. Haduch, A.; Wojcikowski, J.; Daniel, W. A., Pharm. Rep. 2005, 57, 774. 179. Estebe, J. P.; Gentli, M. E.; Le Corre, P.; Leduc, C.; Moulinoux, J. P.; Eccoffey, C., Brit. Jour. Anesth. 2004, 93, 705. 180. Estorch, M.; Carrio, I.; Mena, E.; Fl otats, A.; Camacho, V.; Fuertes, J.; Kulisewsky, J.; Narula, J., Eur. Jour. Nucl. Med. Mol. Imaging 2004, 31, 1575. 181. Pancrazio, J. J.; Bey, P. P.; Loloee, A.; Manne, S.; Chao, H. C.; Howard, L. L.; Gosney, W. M.; Borkholder, D. A.; Kovacs, G. T. A.; Manos, P.; Cuttino, D. S.; Stenger, D. A., Biosensors & Bioelectronics 1998, 13, 971. 182. Kovacs, G. T. A., Proc. IEEE 2003, 91, 915. 183. Stett, A.; Egert, U.; Guenther, E.; Hofmann, F.; Meyer, T.; Nisch, W.; Haemmerle, H., Anal. Bioanal. Chem. 2003, 377, 486. 184. Lehmann-Horn, F.; Jurkat-Rott, K., Physiol. Rev. 1999, 79, 1317. 185. Alphin, R. S.; Martens, J. R.; Dennis, D. M., Anesthesiology 1995, 83, 382. 186. Nattel, S.; Keable, H.; Sasyniuk, B. I. J., Cardiovasc. Pharmacol. 1984, 6, 83. 187. Nattel, S., Circulation 1985, 72, 898. 188. Hille, B., Mechanism of block, in Ionic Channels of Excitable Membranes . Sinauer Associate Inc.,: Sunderla nd, Massachusetts 1997; p pp 390. 189. Barber, M. J.; Starmer, C. F.; Grant, A. O., Circ. Res. 1991, 69, 677. 190. Dr.D.M. Dennis; Dr. T.E. Morey. Departme nt of Anesthesiology, U. o. F., Shands Hospital Private communication The reduc tion in drug concentration has to be achieved in first 15 min. upon the drug is fully dispersed in the bloodstream in 2002. 191. Underhill, R. S.; Jovanovic, A. V.; Carino, S. R.; Varshney, M.; Shah, D. O.; Dennis, D. O.; Morey, T. E.; Duran, R. S., Chem. Mater. 2002, 14, 4919. 192. Jovanovic, A. V.; Underhill, R. S.; Bucholz, T. L.; Duran, R. S., Chem. Mater. 2005, 17, 3375. 193. Lu, Q. F.; Chen, D. R.; Jiao, X. L., Chem. Mater. 2005, 17, 4168. 194. Christiansen, S. C.; Zhao, D. Y.; Janick e, M. T.; Landry, C. C.; Stucky, G. D.; Chmelka, B. F., J. Am. Chem. Soc. 2001, 123, 4519.

PAGE 183

164 195. Janicke, M. T.; Landry, C. C.; Christia nsen, S. C.; Birtalan, S.; Stucky, G. D.; Chmelka, B. F., Chem. Mater. 1999, 11, 1342. 196. Bagwe, R. P.; Yang, C.; Hilliard, L. R.; Tan, W., Langmuir 2004, 20, 8336. 197. Jovanovic, A. V.; Flint, J. A.; Varshney , M.; Morey, T.; Dennis, D. M.; Duran, R. S., Biomacromolecules 2006, 7, 945. 198. Borga, O.; Azarnoff, D. L.; Forshell, G. P.; Sjoqvist, F., Biochem. Pharmacol. 1969, 18, 2135. 199. Borga, O.; Hamberger, B.; Malmfors, T.; Sjoqvist, F., Clin. Pharmacol. Ther. 1970, 11, 581. 200. Ma, Y. L. H., J.A. , Toxicol. 2001, 169, 133. 201. Brunn, G. J.; Keyler, D. E.; Ross, C. A.; Pond, S. M.; Pentel, P. R., Int. J. Immunopharm. 1991, 13, 841.

PAGE 184

165 BIOGRAPHICAL SKETCH Aleksa V Jovanovic was born in Serbia , on July 11, 1973. He obtained his bachelor’s degree in chemistry in 1999 at the University of Belgrade, under the supervision of Dr. Petar Pfendt in the area of geochemistry. He left Serbia in 2000, to pursue a PhD degree at the University of Flor ida where he joined the research group of Dr. Randy Duran. His current interests are in the area of nanotec hnology and colloidal chemistry. Aleksa and Ilka have a daughter, Is abella, born on June 21, 2005.