Microencapsulation (Drugs and the Pharmaceutical Sciences)

This Page Intentionally Left Blank Methods and Industrial Abblicalions edited by Simon Benita The Hebrew University ...

Author: Simon Benita

221 downloads 1927 Views 99MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

This Page Intentionally Left Blank

Methods and Industrial Abblicalions

edited by Simon Benita The Hebrew University of Jerusalem Jerusalem, Israel

Marcel Dekker, Inc. New York.

Basel Hong Kong

Library of Congress Cataloging-in-Publication Data Microencapsulation : methods and industrial applications/ edited by Simon Benita. p. cm. -(Drugs and the pharmaceutical sciences;v. 73) Includes index ISBN 0-8247-9703-5 (alk. paper) 1. Microencapsulation. 1. Benita,Simon. 11. Series. RS20 1.C3M27 1996 6 15'.1 9 d c 2 0

96- 1569 CIP

The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special SalesProfessional Marketing at the address below.

This book is printed on acid-free paper. Copyright 0 1996 by Marcel Dekker, Inc. All Rights Reserved. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writingfrom the publisher. Marcel Dekker,Inc. 270 Madison Avenue,New York, New York 10016 Current printing (last digit): l0987654321 PRINTED IN THE UNITED STATES OF AMERICA

Preface

Research, development, and sales of drug-delivery systems are increasing at a rapid pace throughout theworld. This worldwide trend will intensify in the next decade as cuts in public health expenses demand lowercosts and higher efficacy. To meet this demand, many efficient drugs currently in use will be reformulated within delivery systems that can be value-added for optimal molecular activity. In addition to the health sector, the cosmetic, agricultural, chemical, and foodindustries operate in an open marketplace where free and aggressive competition demands novel coating techniques with enhanced effectiveness at the lowest possible cost. Currently, microencapsulation techniques are most widely used in the development and production of improved drug- and food-delivery systems. These techniques frequently result in products containing numerous variably coated particles. The exact number of particles needed to form a single administered dose varies as a function of the final particle size and can lie in either the micro- or nanometer size range for micro- and nanoparticulate delivery systems, respectively. The microparticulate delivery systems include mainly pellets, microcapsules, microspheres, lipospheres, emulsions, and multiple emulsions. The nanoparticulate delivery systemsinclude mainly lipidor polymeric nanoparticles (nanocapsules and nanospheres), microemulsions, liposomes, and nonionic surfactant vesicles (niosomes). Generally, the microparticulate delivery systemsare intended for oral and topical use. Different types of coated particles can be obtained depending on the coating process used. The particles can be embedded within a polymeric or proteinic matrix network in either asolid aggregated state ora molecular dispersion, resulting in the formulation of microspheres. Alternatively, the particles can be coated by a solidified polymeric or proteinic envelope, leading to the formation of microcapsules. The profile and kinetic pattern governing the release rate of the entrapped active substance from the dosage form depend on the nature and morphology of the coated iii

iv

Preface

particles, which need to be established irrespective of the manufacturing method used. Microencapsulation techniques are normally used to enhance material stability, reduce adverse or toxic effects, or extend material release for different applications in various fields of manufacturing. Until now, the use of some interesting and promising therapeutic substances has been limited clinically because of their restrictive physicochemical properties, which have required frequent administration. It is possible that these substances may become more widely used in a clinical setting. if appropriate microencapsulation techniques can be designed to overcome theirintrinsic inconveniences. Investigators and pharmacologists have been trying to develop delivery systems that allow the fate of a drug to be controlled and the optimal drug dosage to arrive at the site of action in the body by means of novel microparticulate dosage forms. During the past two decades, researchers have succeededin part in controlling the drug-absorption process to sustain adequate andeffective plasma drug levels over a prolonged periodof time by designing delayed- or controlled-release microparticulate-delivery systems intended for either oral or parenteral administration. The ultimate objective of microparticulate-delivery systems is to control and extend the release of the active ingredient from the coated particle without attempting to modify the normalbiofate of the active molecules in the body after administration and absorption. The organ distribution and elimination of these molecules will not be modified and will depend only on their physicochemical properties. Ontheotherhand, nanoparticulatedelivery systems are usually intended for oral, parenteral,ocular, and topical use, with the ultimate objective being the alteration of the pharmacokinetic profile of the active molecule. In thepast decade, ongoing efforts have been made to develop systems or drug carriers capable of delivering the active moleculesspecifically to the intended target organ, while increasing the therapeutic efficacy. This approach involves modifyingthe pharmacokineticprofile of various therapeutic classes of drugs through their incorporation in colloidal nanoparticulate carriers in the submicron size range such as liposomes and nanoparticles. These site-specific delivery systems allow an effective drug concentration to be maintained for a longer interval in the target tissue and result in decreased side effects associated with lower plasma concentrations in the peripheral blood. Thus, theprinciple of drug targeting is to reduce the total amount of drug administered, while optimizingits activity. It should be mentioned that the scientific community was skeptical that such goals could be achieved, since huge investments of funds andpromising research studies have in many cases resulted in disappointing and nonlucrative results and havealso been

Preface

V

slow in yielding successfully marketed therapeutic nanoparticulate dosage forms. With the recent approval by health authorities of a few effective nanoparticulate products containing antifungal or cytotoxic drugs, interest in colloidal drug carriers has been renewed. A vast number of studies and reviews as well as several books have been devoted to thedevelopment, characterization, and potential applications of specific microparticulate- and nanoparticulate-deliverysystems. No encapsulation process developed to date has been able to produce the full range of capsules desired by potential capsule users. Few attempts have been made topresent anddiscuss in a single book the entire size range of particulate dosage forms covered in this book. The general theme and purpose here are toprovide the readerwith a current and general overview of the existing micro- and nanoparticulate-deliverysystems and to emphasize the various methods of preparation, characterization, evaluation, and potential applications in various areas such as medicine, pharmacy, cosmetology, and agriculture. The systematic approach used in presenting the various particulate systems should facilitate the comprehension of this increasingly complex field and clarify the main considerations involved in designing, manufacturing, characterizing, and evaluating a specific particulate-delivery system for a given application or purpose. Thus, the chapters, which have been contributed by leading authorities in the field, are arranged logically according to the methods of preparation, characterization, and applications of the various particulate-delivery systems. The first chapter is by C. Thies, a renowned scientist in the field of microencapsulation techniques. To provide an idea of which process is most appropriate for a specific application, the general principles of several microencapsulation processes are summarized and reviewed. This chapter focuses primarily on processes that have achieved significant commercial use. S . Magdassi and Y. Vinetsky present an interesting technique of oil-inwater emulsion microencapsulation by proteins following adsorption of the protein molecules onto the oil-water interface. J. P. Benoit and Drs. H. Marchais, H. Rolland, and V. Vande Velde have contributed a chapter on advances in the production technology of biodegradable microspheres. This chapter deals mainly with the preparation and use of microspheres. The potential o f the various technologies addressed is also discussed, with an emphasis on marketedproducts or those products currently underclinical evaluation. A. Markus demonstrates in his chapter the importance of applying microencapsulation techniques in the design of controlled-release pesticide formulations to meet the multifaceted demands of efficacy, suitability to mode of application, and minimal damage to the environment. The nanoparticulate-delivery systems are introduced by a chapter, authored by myself, B. Magenheim and P. WehrlC, that explains factorial

vi

Preface

design in the development of nanoparticulate systems. This chapter illustrates the application of the experim9ntal design technique not only for optimization but also for elucidation of the mechanistic aspects of nanoparticle formation by spontaneous emulsification. The second part of the book, which focuses on the evaluation and characterization of the various particulate-delivery systems, starts with an important chapter on microsphere morphology by J. P. Benoit and C. Thies. The chapter helps to clarify definitions and differences, which are very often confused. In addition, the chapter illustrates how morphology can becharacterized by usingdifferent techniques. C. Washington provides his valuable expertise in the presentation of the various kinetic models used to characterize drug-release profiles fromensembles or populations of microparticulate-delivery systems. It is worth noting that therelease mechanism of a drug from multiparticulate systems such as microcapsules or microspheres cannot be identified by a study of global release profiles, since it has been shown that overall or cumulative release profiles from ensembles of microcapsules are entirely different from those of single microcapsules. The discrepancy arises from the heterogeneous distribution of the parameters determiningrelease behavior in individual microcapsules, which isbeyond the scope of the present chapter. The following chapter, by P. Couvreur, G . Couarraze, JrP. Devissaguet, and F. Puisieux, presents a very detailed explanation of the preparation and characterization of nanoparticles. The authors first clearly define the morphology of nanocapsules and nanospheres, providing the background, information, and guidelines for choosing the appropriate methodfor a given drug to be encapsulated. K. Westesen and B. Siekmann havecontributed an important chapter on biodegradablecolloidal drug carrier systems based onsolid lipids. These new colloidal carriers differ from the otherwell-known and widely investigated lipidic colloidal carriers, including liposomes, lipoproteins, and lipid or submicron oil-in-water emulsions by exhibiting a solid physical state as opposed to theliquid or liquid crystallinestate of the above-mentionedand well-knownlipidiccolloidal carriers. The authors present the different methods of preparation and point out the advantagesof the novel dosage forms suchas biodegradability, biocompatibility, ease of manufacture, lack of drug leakage, and sustained drug release. Despite three decades of intensive research on liposomes as drug-delivery systems, the number of systems that have undergone clinical trials and become products on the market is quite modest. Even though there have been few successes with liposomes, the need for drug-delivery systems is as acute as ever, and the potential that liposomeshold, although somewhat tarnished, has not been substantially diminished according to R. Margalit and N. Yerushalmi. An interesting and original approach is presented in their chapter on thephar-

Preface

Vii

maceutical aspects of liposomes. Propositions are presented on how at least in some of the hurdles in research and development can be overcome and furthering the substantial strides that have been made in advancing liposomes from the lab to the clinic. An ingenious solution on how the drawis presented by D. Lasic in the backs of liposomes in vivo can be overcome chapter onstealth liposomes. He explains how the stability of liposomes in liposomicidal environments of biological systems presented a great challenge, which was only recently solved by coupling polyethylene glycol to the lipid molecules. An example of the potential of niosomes (a colloidal vesicular system prepared fromnonionic surfactants) for the topical application of estradiol is contributed by D. A. van Hal and J. A. Bouwstra, and H. E. Junginger. Niosomes have beenshown to increase the penetrationof a drug through human stratum corneum by a factor of 50 as compared with estradiol saturated in phosphate buffer solution, making this colloidal carrier promising for thetransdermal delivery of drugs. In thethird part of the book, thepotential applications of the various particulate-delivery systems are presented. The methods of preparation of microcapsules by interfacial polymerization and interfacial complexation and their applications are discussedby T. Whateley, a scientist who is extremelyknowledgeable in this field. The fast-growingfield of lipid microparticulate-delivery systems, particularly lipospheres, is explained and discussed by A. J. Domb, L. Bergelson, and S. Amselem. Lipospheres represent a new type of fat-based encapsulation technology developed for the parenteraldelivery of drugs and vaccines and the topical administration of bioactive compounds. In their comprehensive and exhaustive chapter, N. Garti andA. Aserin underline the potential of pharmaceutical applications of emulsions, multiple emulsions, and microemulsions, and emphasize the progress made in the last 15 years in understanding mechanismsof stabilization of these promising liquid dispersed-deliverysystems that open new therapeutic possibilities. J.-C.Leroux and E. Doelker and R. Gurny in their chapter on the use of drug-loaded nanoparticles in cancer chemotherapy cover the developments and progress made in the delivery of anticancer drugs coupled to nanoparticles, and the interactions of the latter with neoplastic cells and tissues. This is probably themost promisingand encouragingapplication of nanoparticles and by far themost advanced in the process of development into a viable commercial pharmaceutical product. G. Redziniak and P. Perrier have contributed a chapter on the cosmetic applications of liposomes that have beensuccessfully exploited over the last decade. To complete thewhole rangeof applications of capsular products, a final chapter, by M. Seiller, M.-C. Martini, and myself, discusses cosmetic uses of vesicular particulate-delivery systems. Cosmetics are definitely the largest mar-

viii

Preface

ket, asmanufacturers have demonstrated that marketedcosmetic products containing these vesicular carriers and tested by dermatologists improve cutaneous hydration and skin texture, increase skin glow, and decrease wrinkle depth. It is now taken forgranted that liposomes and othervesicular carriers represent a major step in cosmetics formulations. However, this field requires numerous research studies coupled with strict controls. It is my hope that the scientific information contained herein will modestly contribute to a better understanding of the various particulate systems of all sizes that are now available and to an improved comprehension of their current andpotential applications.

Simon Benita

Contents Preface Contributors

...

111

xiii

Part I: Methods of Encapsulation and Advances in Production Technology 1. A Survey of Microencapsulation Processes Curt Thies

2. Microencapsulation of Oil-in-Water Emulsions by Proteins Shlomo Magdassi and Yelena Vinetsky 3. Biodegradable Microspheres: Advances in

Production Technology Jean-Pierre Benoit, Hervt Marchais, Hervt Rolland, and Vincent Vande Velde

4.

Advances in the Technology of Controlled-Release Pesticide Formulations Arie Markus

5. The Use of Factorial Design in the Development of Nanoparticulate Dosage Forms Baruch Magenheim, Pascal WehrlE, and Simon Benita

21

35

73

93

Part I I: Evaluation and Characterizationof Micro- and Nanoparticulate Drug Delivery Systems 6. Microsphere Morphology Jean-Pierre Benoit and Curt Thies

133 iX

Contents Drug Release from Microparticulate Systems

155

Clive Washington Nanoparticles: Preparation and Characterization

183

Patrick Couvreur, Guy Couarraze, Jean-Philippe Devissaguet, and Francis Puisieux Biodegradable Colloidal Drug Carrier Systems Based on Solid Lipids

213

Kirsten Westesen and Britta Siekmann Pharmaceutical Aspectsof Liposomes: Perspectives in, and Integration of, Academic and IndustrialResearch & Development

259

Rimona Margalit and Noga Yerushalmi Stealth Liposomes

297

Danilo D. Lasic Nonionic Surfactant Vesicles (NSVs) Containing Estradiol for Topical Application Don A. van Hal, Joke A. Bouwstra, and Hans E. Junginger

329

Part IIk Applications of Particulate Delivery Systems 13.

Microcapsules: Preparation by Interfacial Polymerization and Interfacial Complexation and Their Applications Tony L. Whateley

14.

Lipospheres for ControlledDelivery of Substances Abraham J. Domb, Lev Bergelson, and Shimon Amselem

15.

Pharmaceutical Emulsions, Double Emulsions and Microemulsions

349

377

411

Nissim Garti and Abraham Aserin 16. The Use of Drug-Loaded Nanoparticles in Cancer Chemotherapy

Jean-Christophe Leroux, Eric Doelker, and Robert Gurny

535

Contents

Xi

17. CosmeticApplications of Liposomes G6rard Redziniak and Pierre Perrier

577

18. Cosmetic Applications of Vesicular Delivery Systems Monique Seiller, Mane-Claude Martini, and Simon Benita

587

Index

633


Contributors Shimon Amselem, Ph.D. Director of Pharmaceutics, Department of Pharmaceutics, Pharmos Ltd., Weizmann Industrial Park, Rehovot, Israel Abraham Aserin, Ph.D. Casali Institute of Applied Chemistry, School of Applied Science and Technology, The Hebrew University of Jerusalem, Jerusalem, Israel SimonBenita,Ph.D.* Professor, Department of Pharmacy,School of Pharmacy, The Hebrew University of Jerusalem, Jerusalem, Israel Jean-PierreBenoit,Ph.D. Professor, Laboratoire de Pharmacie GalCnique et Biophysique Pharmaceutique, UniversitC d’Angers, and Centre de Microencapsulation, Angers, France Lev Bergelson, Ph.D. Professor, Department of Pharmaceutical Chemistry, School of Pharmacy, Faculty of Medicine, The Hebrew University of Jerusalem, Jerusalem, Israel Joke A. Bouwstra, Ph.D. Associate Professor, Department of Pharmaceutical Technology, LeidedAmsterdam Center for Drug Research, Leiden University, Leiden, The Netherlands GuyCouarraze,Ph.D. Professor, Centred’EtudesPharmaceutiques, URA CNRS 1218, Universitt de Paris-Sud, Chbtenay-Malabry, France PatrickCouvreur,Ph.D. Professor, Centred’EtudesPharmaceutiques, URA CNRS 1218, UniversitCde Paris-Sud, Chitenay-Malabry, France

*Affiliated with the David Bloom Centerfor Pharmacy, Jerusalem, Israel.

xiii

xiv

Contributors

Jean-PhilippeDevissaguet,Ph.D. Professor, Centred'EtudesPharmaceutiques, UR4 CNRS 1218, Universite de Paris-Sud, Chtltenay-Malaby, France EricDoelker,Ph.D. Professor, School of Pharmacy,University of Geneva, Geneva, Switzerland Abraham J. Domb,Ph.D.* Professor, Department of Pharmaceutical Chemistry, School of Pharmacy, Faculty of Medicine, The HebrewUniversity of Jerusalem, Jerusalem, Israel Nissim Garti,Ph.D. Professor, Casali Institute of Applied Chemistry, School of Applied Science and Technology, The Hebrew University of Jerusalem, Jerusalem, Israel Robert Gurny,Ph.D. Head Professor, Department of Pharmaceutics and Biopharmaceutics, School of Pharmacy, University of Geneva, Geneva, Switzerland Hans E.Junginger, Ph.D. Professor, Pharmaceutical Deparhnent, Leided Amsterdam Center for Drug Research, Leiden University, Leiden, The Netherlands Danilo D. Lasic, Ph.D.** Department of Research, Liposome Technology, Inc. ,Menlo Park,California Jean-ChristopheLeroux,Ph.D. neva, Geneva, Switzerland

School of Pharmacy, University of Ge-

ShlomoMagdassi,Ph.D. SeniorLecturer, Casali Institute of Applied Chemistry, School of Applied Science and Technology, The Hebrew University of Jerusalem, Jerusalem, Israel BaruchMagenheim,M.Sc.+ Department of Pharmacy, School of Pharmacy, The Hebrew University of Jerusalem, Jerusalem, Israel Hem6 Marchais, Ph.D. Laboratoire de Pharmacie GalCnique et Biophysique Pharmaceutique,UniversitC #Angers, Angers, France *Affiliated with the David Bloom Center for Pharmacy, Jerusalem, Israel. **Current affiliation: Senior Scientist, MegaBios Corporation, Burlingame, California +Current affiliation: Agis Industries Ltd., Yeruham, Israel

Contributors

xv

Rimona Margalit, Ph.D. Professor, Department of Biochemistry, Faculty of Life Sciences, Tel Aviv University, Tel Aviv,Israel ArieMarkus,Ph.D. Institute for Chemistry and Chemical Technology, The Institutes for Applied Research, Ben-GurionUniversity of the Negev, Beer-Sheva, Israel Marie-Claude Martini, Ph.D. Professor, Institut Pharmaceutiques et Biologiques, Lyon, France ,

des

Sciences

Pierre Perrier, Ph.D. Parfums Christian Dior, Saint Jean de Braye, France FrancisPuisieux,Ph.D. Head Professor, Centre d’Etudes Pharmaceutiques, URA CNRS 1218, UniversitC de Paris-Sud, Chitenay-Malabry, France G6rard Redziniak, Ph.D. Applied Research Manager, Parfums Christian Dior, Saint Jean de Braye, France Hew6 Rolland,Ph.D. France

Director, Centre de Microencapsulation,Angers,

Monique Seiller, Ph.D. Professor, Unit6 d’Enseignement et de Recherche des Sciences Pharmaceutiques, UniversitC de Caen, Caen, France Britta Siekmann, Ph.D. Associate Director, Department of Pharmaceutics, Astra Arcus AB, Sodertalje,Sweden CurtThies,Ph.D. Professor, Department of Chemical Engineering, Washington University, St. Louis, Missouri Don A. vanHal, Ph.D. LeidedAmsterdamCenter for Drug Research, Leiden University, Leiden,The Netherlands Vincent Vande Velde,Ph.D. Technical Director, Centre deMicroencapsulation, Angers, France Yelena Vietsky, M.Sc. Casali Institute of Applied Chemistry, School of Applied Science and Technology, The Hebrew University of Jerusalem, Jerusalem, Israel Clive Washington, Ph.D. Lecturer in Pharmaceutics, Department of Pharmaceutical Sciences, The University of Nottingham, Nottingham, England

mi

Contributors

PascalWehrl6, Ph.D. Professor, Laboratorie de Pharmacotechnie, Centre de Recherches Pharmaceutiques,UniversitC Louis Pasteur, Strasbourg, France Kirsten Westesen, Ph.D. Professor, Department of Pharmaceutical Technology, Institute of Pharmaceutics, Friedrich-Schiller University Jena, Jena, Germany Tony L. Whateley,Ph.D. Reader, Department of Pharmaceutical Sciences, University of Strathclyde, Glasgow, Scotland NogaYerushalmi,Ph.D. Department of Biochemistry, Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel

1 A Survey o f Microencapsulation Processes Curt Thies Wizshington University, St. Louis, Missouri

I. Introduction 11. Processes for Preparing Microcapsules A.GeneralComments B. Type A Encapsulation Processes C. Type B Encapsulation Processes 111. Summary

1.

1 2 2 3 10 17

INTRODUCTION

Microcapsulesare small particles that contain an active agent or core material surrounded by a coating or shell. At present, there is no universally accepted size range that particles must have in order to be classified as microcapsules. However, many workers classify capsules smaller than 1 pm as nanocapsules andcapsules larger than 1000 pm as macrocapsules. Commercial microcapsules typicallyhave a diameter between3 and 800 pm and contain 10-90 wt. percent core. A wide range of core materials has been encapsulated, including adhesives, agrochemicals, live cells, active enzymes, flavors, fragrances, pharmaceuticals, and inks. Most capsule shell materials are organic polymers, but fats and waxes are also used. 1

Thies

2 Shell

Material

A

B

Fig. 1 Schematicdiagrams of tworepresentativetypes of microcapsules: (A) Continuous core/shell microcapsule. (B) Multinuclear microcapsule. (Courtesy of C. Thies.) Microcapsules can havea variety of structures. Some havea spherical geometry with a continuous coreregion surrounded by a continuous shell, as shown in Fig. 1A. Others have an irregular geometry and contain a number of small droplets or particles of core material, as shown in Fig.1B. Microcapsules are usedin a wide range of commercial products. They are thekey component of all carbonless copy papers and are present in several oraland injected drug formulations. Encapsulated adhesive resins coated on automotive fasteners are routinely used to assure that such fasteners are firmly set when installed. Microcapsules also are the basis for a number of long-actingcommercial pesticide and herbicide products. Improvement of these products and developmentof new ones is an ongoing process that involves a large number of development groups globally.

II.

PROCESSES FOR PREPARING MICROCAPSULES

A.GeneralComments

The concept of preparing small particles that carry a core material trapped within a shell material dates back at least to spray-drying work carried out in the 1930s. However, the first truly significant commercial productthat utilized such particles was carbonless copy paper. The microcapsules for this product were made by a process called complex coacervation. Since then many other methods forpreparing microcapsules have been developed and improved significantly. Some are based exclusivelyon physical phenomena. Some utilize polymerization reactions to produce a capsule shell. Others combine physical and chemical phenomena. Many investigators classify en-

Microencapsulation ofA Processes Survey

3

capsulation processes as either chemical or mechanical. I prefer toclassify them as type A or type B processes, since so-called mechanical processes may actually involve a chemical reaction and so-called chemical processes may rely exclusively on physical phenomena. Table 1 lists representative examples of both types of processes. Capsules produced by type A, or chemical processes, are formed entirely in a liquid-filled stirred tank or tubular reactor. Capsules produced by type B, or mechanical processes, utilize a gas phase at some stage of the encapsulation process. In such processes, capsules can be formed by either spraying droplets of coating material on acorematerial being encapsulated, solidifying liquid droplets sprayed into a gas phase, gelling droplets sprayed into a liquid bath, or carrying out apolymerization reaction at a solid-gas interface. No encapsulation process developed to dateis able to produce the full range of capsules desired by potential capsule users. Some readily produce small, liquid-filled capsules, whereas others produce relatively large capsules with a solid core material. In order to obtain a feeling for which process is most appropriate for a specific application, it is appropriate to summarize the general principles of several types A and B processes. This chapter focuses primarily on processes that have achieved significant commercial use. Several reviews giveadditional details[l-61. B. Type A Encapsulation Processes

l . Complex Coacervation Because of its original use in carbonless paper, the first type A process considered is complex coacervation. It is based on the ability of cat-

Table 1 List of Types A and B Encapsulation Processes Type A (chemical) processes

Type B (mechanical) processes

Complex coacervation Polymer-polymer incompatibility Interfacial polymerization in liquid media In situ polymerization In-liquid drying

Spray drying Spray chilling Fluidized bed Electrostatic deposition Centrifugal extrusion Spinning diskor rotational suspension separation Polymerization at liquid-gas or solid-gas interface Pressure extrusion or spraying into solvent extraction bath

Thermal andionic gelation in liquid media Desolvation in liquid media

Thies

4

ionic and anionic water-soluble polymers to interact in water to form a liquid, polymer-rich phase called a complex coacervate. Gelatinis normally the cationicpolymer used. A variety of natural and synthetic anionic watersoluble polymers interact with gelatin to form complexcoacervates suitable for encapsulation.When the complex coacervate forms,it is in equilibrium with a dilute solution called the supernatant. Inthis two-phase system, the supernatant acts as the continuous phase,whereas the complex coacervate acts as the dispersed phase. If a water-insoluble core material is dispersed in the system and the complex coacervate wets this core material, each droplet or particle of dispersed core materialis spontaneously coated with a thin film of coacervate. When this liquid film is solidified, capsules are formed. Fig. 2 outlines one version of a complex coacervation encapsulation process [7]. The first step is to disperse the core material in qn aqueous gelatin solution. This is normally done at40-60°C, a temperature range at which the gelatin solution is melted and liquid. After a polyanion or negatively charged polymer like gum arabic is added to thesystem, the pH and

tielalln

4

Hyoo (40-60C)

H

Form Emulsion of Core In Solulion Gelalln (40-60°C)

1 Gel Shell (515°C)

l

Core Malerial (Llquicl)

Form Liquid Coacervale by Adjusling pH I Adding Polyanion and Dilulion Waler (40-60'C)

7Crosslink with Glutaraldehyde

Fig. 2 Schematic flow diagram of encapsulation process based on complex coacervation. (Courtesy of C. Thies.)

A Survey of Microencapsulation Processes

5

concentration of polymer are adjusted so that a liquid complex coacervate forms. The pH atwhich this occurs is typically between 4.0 and 4.5. Once the liquid coacervate forms, the system is cooled to room temperature. The gelatin in the coacervate gels thereby forming capsules with a very rubbery shell. In order to increase the strength of the water-swollen shell and create a gel structure that is not thermally reversible, the capsules normally are further cooled to approximately 10°C and treatedwith glutaraldehyde. The glutaraldehyde crosslinks the gelatin by reacting with amino groupslocated on the gelatin chain. Glutaraldehyde-treated capsules can be dried to a free-flow powder or coated on a substrate and dried. Complex coacervation has been used to encapsulate many waterimmiscible liquids and is used in a variety of products, including inks for carbonless paper, perfumes for advertising inserts, and liquid crystals for display devices. This technology routinely produces single capsules of 20800 pm diameter that contain 80-90 wt. percent core material. Capsules outside this size range can be made, but considerable know-how may be required. Most coacervate capsules have a continuous core/shell structure (see Fig. lA),although the shell is not of uniform thickness. The mechanical and barrier propertiesof dry capsules formed by complex coacervation generally are sensitive to moisture. When stored at humidities above 70% or immersed in water, the shell of a complex coacervate capsule swells greatly, thereby facilitating transport of material into or outof the capsule. Postreatment of such capsules with urea and formaldehyde under acidic conditions has been used to reduce their moisture sensitivity. The value of complex coacervation encapsulation technology for food and pharmaceutical applications would be greatly enhanced if such capsules could be isolated commercially without the use of a chemical crosslinking agent.

2. Polymer-PolymerIncompatibility Polymer-polymer incompatibility is a second type A process that has been commercialized. This technology utilizes a polymer phase-separation phenomenon quitedifferent from complex coacervation. In complex coacervation, two oppositely charged polymers, gelatin and a polyanion, join together to form the complex coacervate and both polymers become part of the final capsule shell. In contrast,polymer-polymer incompatibility occurs because two chemically different polymers dissolved in a common solvent are incompatible and do not mix in solution. They essentially repel each other and form two distinct liquid phases. One phase is rich in polymer designed to act as the capsule shell. The otheris richin the second, incompatible polymer. The incompatible polymer is present in the system to cause formation of two phases. It is not designed to be part of the final capsule shell, although small amounts may remain entrapped in the final capsule as

Thies

6

% l

0000

@ V

Phase Separated Solution of Ethylcellulose+ Polyethylene in Hot (80'C) Cyclohexane

Core Material (Small Particles)

t

Dispersion of Core Material in Hot, Phase-Separated Polymer Solution

+ Cool to 20-c (Ethylcellulose Coating Solidifies)

Elhylmllulose Shell

Wash and Harvest Capsules

Fig. 3 Schematic flow diagram of encapsulation process based on polymerpolymer phase separation. (Courtesy of C. Thies.)

an impurity. This type of encapsulation process generally does not involve any chemicalreaction. Fig. 3 illustrates a specific polymer-polymerphase-separation encapsulation process used to produce commercial pharmaceutical grade microcapsules [8]. The first step is to disperse the core material in ahot (80°C) solution of ethyl cellulose in cyclohexane. Lowmolecular weight polyethylene, a polymer soluble in hot cyclohexane and incompatible with ethyl cellulose, is added to thesystem. This induces phase separationwith formation of an ethyl cellulose-rich phase and a polyethylene-rich phase. The


7

core material, a solid unaffected by 80°C cyclohexane, is dispersed in this two-phase system. Since the ethyl cellulose is more polar than polyethylene, it adsorbs preferentially on the surface of the core material and thereby causes a thin coating of shell material solution to engulf the particles of core material. When the system iscooled to room temperature, the ethyl cellulose precipitates, thereby solidifying the ethyl cellulose solution and formingsolid microcapsules that can beharvested. Aspirin and potassium chloride are examples of commercial encapsulated pharmaceutical products producedin this manner [g]. Polymer-polymer phase-separation processes normally are carried out in organic solvents and are used to encapsulate solids with a finite degree of water solubility. Many of the capsules produced commercially have an ethyl cellulose shell, are irregularly shaped 200- to 800-pm particles, and are loaded with solid drug particles. They are used to provide taste masking or prolonged oral drug delivery [g, 101. Smaller capsules (e.g., 20-120 pm in diameter) with a biodegradable poly(d,l-lactideglycolide) shell have been preparedby polymer-polymer phase separation and used asinjectable drug-delivery devices [ll].

3. Interfacial Polymerization Interfacial polymerization (IFP) is a third type A encapsulation process that has been commercialized. A unique featureof this technology is that thecapsule shell is formed at or on the surface of a droplet orparticle by polymerization of reactive monomers. This approach to encapsulation has evolved into a versatile technology able to encapsulate a wide range of core materials, including aqueous solutions, water-immiscible liquids, and solids. Many types of IFP reactions are used. Fig. 4 is a schematic diagram that illustrates how the capsule shell forms when a capsule loaded with a water-immiscible liquid is prepared by IFP. In suchcases, a multifunctional monomer is dissolved in the liquid core material. This monomer often is a multifunctional isocyanate but could be a multifunctional acid chloride or a combination of isocyanates and acid chlorides. In all cases, the resulting solution is dispersed to a desired drop size in an aqueous phase that contains a dispersing agent. A coreactant, usually a multifunctional amine, is then added to the aqueous phase. Thisproduces a rapid polymerization reaction at the interface which generates the capsule shell[12]. A polyurea capsule shellis obtained if the IFP reaction involves an isocyanate and an amine. A polyamide or nylon capsule shell is obtained if the IFP reaction involves an acid chloride and an amine. Reaction of an isocyanate with a hydroxylcontaining monomer dissolved in the aqueous phase produces a polyurethane capsule shell.

Thies

8 Polyurea Polyurea Shell

-

Core Material ["oil")

'. H20

+

EMULSION STABILIZER

Fig. 4 Schematic illustrationof an interfacial reaction frequently usedto prepare microcapsules by interfacial polymerization. (Courtesyof C. Thies.)

If the core material being encapsulated is an aqueous solution, the abovereactantadditionsequence is reversed.Thecoreactant,e.g.,a water-soluble amine,is dissolved in the aqueous phase.The resulting mixture is dispersed in a water-immiscible solvent.unti1the desired drop sizeis obtained. A solvent-soluble reactant like an isocyanate or acid chloride is added to the organic phase, thereby initiating rapid polymerization at the interface to produce a capsule shell. A variety of core materialsof biological interest (e.g., active enzymes) have been producedin this manner [13]. Solids can also be encapsulated by interfacial polymerization reactions, although the polymerization chemistry typically differs from that used to encapsulate liquids. Vinyl monomers that polymerize by free radical reactions generally are used to encapsulate solids. In one process, the solid to be encapsulatedis dispersed in aqueousmedia along with a dispersing agent. Vinyl monomer is added to thesystem and freeradical polymerization is initiated by a redox initiation system [14]. Success of this type of encapsulation process depends on forcing polymer formation to occur at the solid-liquid interface rather than allowing it to occur throughout the entire aqueous media. Another type of interfacial polymerization reaction involves the polymerization of p-xylylene at a gas-solid interface [15]. In this case,polymerization of vaporized p-xylylene radicals occurs spontaneously on contactwith a solid surface. Although interfacial polymerization can be used to produce large capsules, most commercial uses of this technology have focused on the preparation of small capsules that are coated or sprayed onto a substrate rather thanbeing isolated as afree-flow powder. For example, IFP is routinely used to produce 20- to 30-pm diameter

icroencapsulation ofA Processes Survey

9

capsules loaded with pesticides and herbicides. It also is used toform 3-to 6-pm diameter capsules loaded with carbonless paper ink. Capsules with polyurea or polyamide shells have liquid cores or cores that melt below 60-70°C. Because one of the reactantsused to create thecapsule shell is dissolved in the core material and is free to react with any reactive groups located on molecules of core material to create new molecular species, IFP encapsulation technology is viewed suspiciously by workers concerned with the encapsulation of fragrances, drugs, and flavors. Many agrochemicals like pesticides and herbicides are relatively water-insoluble low-melting solids or high-boiling point liquids free of groups that could react with the reagent(s) dissolved in the core material. Thus, many current commercial encapsulated agrochemical formulations are prepared by IFP [12]. Capsules formed by IFP often have a continuous core-shell structure and a spherical geometry. Significantly, the exterior surface of many IFP capsules is smooth and uniform. This is not the case for the interior surface. Capsule fracture studies show that the interior surface of many IFF' capsule shells is cratered and irregular. That is, the capsule shell is not uniformly deposited around the core by an IFP process except for a comparatively thin outer region of the shell. In such cases, the shell of an IFP capsule is not a membrane of uniform thickness. It is a thin skin membrane resting on a thick, porous support. In Situ Polymerization In situ polymerization is a type A encapsulation technology closely related to IFP. Like IFP, capsule shell formation occurs because of polymerization of monomers added to theencapsulation reactor. However, with in situ encapsulation processes, no reactive agents are added to thecore material. Polymerization occurs exclusively in the continuous phase and on the continuous-phase side of the interface formed by the dispersed core material and continuous phase. Polymerization of reagents located there producesa relatively low molecular weight prepolymer. As this prepolymer grows in size, it deposits onto thesurface of the dispersed core materialbeing encapsulated where polymerization with crosslinking continues to occur thereby generating a solid capsule shell. The first example of this technology was the encapsulation of various water-immiscibleliquids with shells formed by the reaction at acid pH of urea with formaldehyde in aqueous media [14]. Similar encapsulation reactions occur between melamine and formaldehyde in aqueous media. Addition of anionic polymers to theaqueous phasecan have a significant effect on thein situ encapsulation process [16]. In situ polymerization is used extensively to produce small(3- to 6-pm diameter) capsules loaded with carbonless paper inks or perfume for scented strips. Larger capsules loaded with mineral oil are used for cos-

4.

.

10

Thies

metic applications, whereas capsules filled with epoxy resin are used for fasteners. In all of these cases, the capsules have a continuous core-shell structure. Thetechnology can also be adapted to encapsulatesolids.

5. Centrifugal Force and Submerged Nozzle Processes Several interesting type A processes use centrifugal force or two-fluid submerged nozzles to form microcapsules. In oneprocess, a cup perforated with a series of fine holes is immersed in an oil bath. It is rotated while immersed in the oil, thereby extruding into the oil phase a stream of droplets of an oil-in-water emulsion [17]. The water phase of this emulsion is a concentrated solution of a water-soluble polymer that gels on cooling. Gelatin is a specific example. By controlling the temperature of the oil bath, the external phase of the extruded emulsion droplets is gelled to create oil-loaded gel beadlets that can be isolated and dried. When isolated, the capsules consist of a number of small oil droplets dispersed throughout a matrix of shell material. This process was developed in 1942 in order to produce capsules that improved the oxidative stability of vitamins and fish oils [17]. Capsules can also be produced by coextruding an aqueous gelatin solution and an oil to be encapsulated through a two-fluid nozzle into a moving fluidstream of an oil solution. Fig. 5 schematically illustrates such a nozzle. The gelatin solution surrounds the oil drop to be encapsulated at the moment of extrusion and is gelled thermally before the particles are harvested and dried. In the case shown in Fig. 5 , continuous shell-core capsules are obtained.

C. Type B EncapsulationProcesses Centrifugal force, extrusion,coextrusion, and formation of sprays are the principal means by which type B capsules are made. Type B methods of encapsulation predate type A encapsulation processes, since spray drying, a type B process, was developed in the 1930s. Because spray drying is the oldest type B encapsulation process, it is appropriate to discuss it first.

1. SprayDrying The first step in a spray-dry encapsulation processes is to emulsify or disperse the core material in a concentrated (40-60 wt. percent solids) solution of shell material.The core materialgenerally is a water-immiscible oil such as a fragrance, flavor, or vitamin. It is emulsified in a solution of shell material until 1-to 3-pmoil droplets are obtained.The shell material normally is a water-soluble polymer like gum arabic or a modified starch. These materialsdo not form high-viscosity solutions at thehigh-shell material concentrations favored for spray drying. Mixtures of these shell materi-


-

Core Material (e.g., Vitamin O i l )

Shell Material (e.g.. Warm Gelatin Solution)

+ 2-Fluid2-Fluid. Nozzle

Warm Carrier Phase (e.g.,Vegetable Oil)

@

l =l

Coil

@

z

@ /

Gel Beads that Core can be harvested Material and dried

Fig. 5 Schematicdiagram of a submergedtwo-fluidnozzleused microcapsules. (Courtesyof C. Thies.)

to prepare

als with hydrolyzed starches (maltodextrins) or hydrolyzed gelatins have been used. Water is the preferred solvent for most spray-drying encapsulations. Concerns about solvent flammability and toxicity severely restrict the use of organic solvents for conventional spray-drying encapsulation applications. However, several groups are exploring the use of spray drying from organic media to produce pharmaceutical capsules from biodegradable polymers. Once a suitable dispersion of core material in a solution of shell material has been prepared, the resulting emulsion is fed as droplets into the heated chamberof a spray drier. The droplets may be sprayed intothis chamber or spunoff a rotating disk. In either case, they are rapidly dehy-

72

Thies

drated in a heated chamber, thereby producing dry capsules. The capsules fall to the bottom of the spray-drying chamber where they are harvested. Capsules produced in this manner typically fall between 10 and 300 pm in diameter. They tend to have an irregular geometry and may be aggregates of a number of small particles. Each spray-dried capsule has a number of small droplets of core material dispersed throughout it. Spray-dry encapsulation has a number of advantages. It is a wellestablished technology, involves readily available equipment, andis able to produce large amounts of capsules. Many shell materials preferred for spray drying are approved for food use. Furthermore, these materials are water soluble and notchemically crosslinked. Thus, capsules prepared from them dissolve in water andrelease core material without leaving any capsule shell debris. In contrast, the shell of complex coacervate capsules crosslinked with glutaraldehyde swells greatly in water but does notdissolve. Such capsules can be ruptured, therebyreleasing their contents, butcapsule shell debris is left behind. For some applications, this is objectionable. Spray-dry encapsulation has several problems and limitations. For example, if water is the preferred solvent, spray-dry encapsulation is limited to shell materials soluble or dispersible in water. The list of candidate water-soluble shell materials is limited, because many candidate materials form aqueous solutions that are tooviscous to spray in spray-dry encapsulation protocols. Another limitation is the 20-30% core loading carried by most spray-dried capsules. Spray-drying protocols that allow core loading up to 50-60% have been reported [18], but current spray-dried capsules have lower loadings. A persistent problem with spray-dried capsules is free or surface core material. Because water evaporation from a capsule in the chamber of a spray drier occurs rapidly, it is not uncommon to harvest spray-dried capsules that have a finite amount of free or unencapsulated oil. The higher the core loading, the more pronounced this problem can become. This is undesirable, since free fragrance or flavor oil is susceptible to oxidation and the development of an off-odor or off-taste. Brenner [l81 noted that maximizing core loading and minimizing free core content involves a judicious choice of coating material, emulsifying agent, andspraydryeroperating conditions. Finally, it is important to stress that lowboiling point compounds with a finite degree of water solubility have posed a persistent problem to spray-dry encapsulation. Such compounds volatilize from the capsules in the spray-drying chamber. In summary, spray drying is a viable commercial method of forming microcapsules. It is an established, comparatively low-cost encapsulation technology that continues to develop. To date, spray drying has primarily been used to encapsulate fragrances and flavors.


13

2. Fluidized Bed Coaters Fluidized bed coaters are another type B encapsulation technology. They are limited to encapsulating solid particles or porous particles into which a liquid has been absorbed. Nevertheless, they are used extensively to encapsulate many different solids. Fluidized bed coaters function by suspending a bed or column of solid particles in a moving gas stream, usually air. A liquid coating formulation is sprayed onto the individual particles, and the freshly coated particles are cycled into a zone where the coating formulation is dried either by solvent evaporation or cooling. This coating and drying sequence is repeated until a desired coating thickness has been applied. Many variables affect the fluidized coating process, including dew point of the incoming air. Lehmann [l91 reviewed the operation of fluidized bed coaters. They represent a major capital investment, but are widely usedto produce encapsulated solids, especially for the pharmaceutical industry. A major advantage of fluidized bed coaters is their ability to handle an extremely wide range of coating formulations. They have been used to apply hot melts, aqueous latex dispersions, organic solvent solutions of shell material, and aqueous solutions of shell material. Enteric polymer solutions have been of particular interest, since they are used by the pharmaceutical industry to produce drug-dosage formulations that survive passage through the stomach. Enteric polymers are insoluble in gastric fluid (pH 2) and soluble in intestinal fluid (pH 7), so capsules or tablets coated with them can pass intact through the stomach and not disintegrate until they reach the intestine. Three types of fluidized beds are available: top-spray, tangentialspray, and bottom-spray. These units differ in location of the nozzle or nozzles used to apply the coating formulation. Fig. 6 offers schematic diagrams of a top-spray and a bottom-spray unit. In the top-spray unit (Fig. 6A), the coating formulation is sprayed into the fluidized bed. The droplets of spray leaving the nozzle move countercurrent to the gas stream until they impact the particles being coated. If the spray formulation contains a volatile solvent, evaporation of this solvent from the spray droplets occurs, thereby increasing the solids content, perhaps to such a degree that the spray droplets cannot spread on the particles being coated. This spraydrying effect has been used to explain why solvent-based coating formulations applied in top-spray fluidized bed coaters oftenyield coated particles with a degree of internal void volume and porous coatings. Enteric coatings applied as aqueous latex dispersions are an exception. Such coatings applied in a top-spray fluidized bed coater form a continuous coating analogous to that obtained with tangential-spray and bottom-spray units. Fats

Thies

14

A

Being Coated

Gas Flow

Expansion Chamber

Core Material Being Coated Formulation Gas Flow

Fig. 6 Schematic diagrams of two types of fludized bed coaters: (A) Top-spray unit. (B) Bottom-spray, or Wurster unit. (Courtesy of C. Thies.) are also applied by top-spray fluidized bed coaters. Top-spray coaters have limitations, but they are more simple to operate and have higher production capacity than the othertypes of units. In tangential-spray and bottom-spray units, the droplets of coating formulation move in the same direction as the particles being coated. The droplets of coating formulation travel a shorterdistance before theyimpact


15

the particles being coated. Solvent evaporation is minimized, and a more uniform film of coating material is deposited. The bottom-spray, or Wurster coater, shown in Fig.6B has becomea standard unit used to produce encapsulated solids, especially solidpharmaceuticals. The partition and gas-distribution plates of such units are important components, becausethey direct the particles being coated in a cyclic path past the nozzle that supplies the coating formulation. Particles move past this nozzle, receive a spray of coating material, and move up into the upper section of the unit. Here the coating is solidified either by solvent evaporation or cooling of a hot melt. The coated particles then fall back down to thebottom of the unit where a fresh coating is applied. This cyclic process is repeated until the desired weight of shell material has been applied. Application of the coating in a series of cyclic steps is both an advantage anda disadvantage. The disadvantage is that stepwise deposition takes time. A one-step coating process is much quicker, but with a multistep coating process, defects can beminimized or perhaps eveneliminated. This makes it possible to use the Wurster process to produce capsules with a shell that approaches the samebamer properties as a free-standing solvent-cast film. In a one-step coating process, capsule shell defects are not healed and, hence, can havea profound effect on capsule release behavior.

3. Centrifugal Extrusion Fig. 7 is a schematic diagram of a centrifugal extrusion process [20]. The core and shell material, two mutually immiscible liquids, are pumped through a spinning two-fluid nozzle. This produces a continuous two-fluid column or rod of liquid that spontaneously breaks up into a stream of spherical droplets immediately after it emerges fromthe nozzle. Each droplet contains a continuous core region surrounded by a liquid shell. How these droplets are convertedinto capsules is determined by the nature of the shell material. If the shell material is a relatively low-viscosity hot melt that crystallizes rapidly on cooling (e.g., a wax or wax toughened with a polymer), the droplets are converted into solid particles as they fall away from the nozzle. Suitable core materials typically are polar liquids like water or aqueous solutions, since they are immiscible with a range of hot melt shell materials like waxes. Alternatively, droplets emerging from the nozzle may have a shell that is an aqueous polymer solution able to be gelled rapidly. In this case, the droplets fall into a gelling bath wherethey are converted intogel beads. A specific example is the gelatin of an aqueous sodiumalginate shell by an aqueous CaCl, gelling bath. The gel beads produced can be dried to give capsules with a solid shell. Core materials suitable for this type of capsules shell are water-immiscible oils.

Thies

16 Core Material Shell

Shell

AIR ,Liquid Core

Shell gelled b y immersion in gelling bath

Shell solidifies

by cooling in air

Fig. 7 Schematicdiagram of centrifugaltwo-fluidnozzleused crocapsules. (Courtesyof C. Thies.)

to

produce rni-

Capsules prepared by centrifugal extrusion tend to be large with diameters typically ranging from over 250 p up to several millimeters. 4. Rotational Suspension Separation Fig. 8 is a schematic diagram of the rotational suspension separation process for preparing microcapsules. In this process, core material dispersed in a liquid shell formulation is fed onto a rotating disk. A flat disk is shown, but conical or bowl-shaped disks can be used. Individual core particles coated with a film of shell formulation are flung off the edge of the rotating disk along with droplets of pure coating material. When the shell

A Survey of Microencapsulation Processes "Spherical" Material

Coating Core Phase (e.g.. molten lot)

!

17

Rotate

Spinning Disc

0 h

t

Free Coating Material

Fig. 8 Schematicdiagram

t

Capsules

of rotational suspension separation encapsulation

process.

formulation is solidified, e.g., by cooling, discrete microcapsules are produced. The droplets of pure coating material also solidify, but they are said to collect in a discrete zone away from the capsules, as shown in Fig. 8. This technology is claimed to be a fast, low-cost, high-volume method of encapsulating a variety of materials that act like solids on the rotating disk, including particles below 150 pm in diameter [21]. In order toobtain optimal results, the core material must have a spherical geometry. Formation of this geometry may require a granulation step before encapsulation is attempted. A variety of hot-melt shell materials can be applied, butmelt viscosities below 5000 CP are favored. Capsule shell formulations that do not solidify rapidly are said to pose problems. Since solidification of the coating formmulation occurs rapidly and the coating formulations applied are crystalline materials susceptible to polymorphic transformations or changes in percent crystallinity on storage, it is reasonable to suggest that capsule shells produced by rotational suspension separation are notat thermodynamic equilibrium immediately after capsule formation is complete. 111.

SUMMARY

Most microcapsules of commercial significance are produced by the encapsulation processes described in this chapter. The commercialization of products that contain microcapsules is progressing as is the development of encapsulation processes. Today, a number of capsule-based products exist.

Thies

18

Several have significant sales. The biggest application, carbonless paper, consumes thousands of tons of capsules per year [5].Such capsules are produced by complex coacervation, IFP, or in situ polymerization. Large volumes of commercial capsules loaded with agrochemicals, especially herbicides, are madeby IFF. Although a numberof encapsulated food ingredients arebeing marketed [22,23],the volume of such capsules remains small relative to the huge potential that exists. The development of encapsulation technology and microcapsule-based products continues tobe pursued actively by a number of groups globally. Accordingly, it is reasonable to project that this field will experience steady growth for the foreseeable future. It will be interesting to see when the next high-volume capsulebased product appears and what it will be.

REFERENCES 1. P. B. Deasy, Microencapsulation and Related Drug Processes, Marcel Dekker, New York, 1984. 2. A. Kondo, Microcapsule Processing and Technology, Marcel Dekker, New York, 1979. 3. C. Thies, Kirk-Othmer Encyclopedia of Chemical Technology, Vol. 16, 4th ed., Wiley, 1995, pp. 628-651. 4. W. Slwika, Angew. Chem. Internat Edit., 14539-550 (1975). 5. C. A. Finch, Ullmann’s Encyclopedia of Industrial Chemistry, Vol. A16, 5th ed., VCH Publishers, New York, 1990, pp. 575-588. 6. I. C. Jacobs, and N. S. Mason, in Polymeric Delivery Systems, ACS Sympo-

sium Series520 (M. A. El-Nokaly, D. M. Piatt, and B. A. Charpentier,eds.), American Chemical Society, Washington, DC,1993, pp. 1-17. 7. B. K. Green, and L. Schleicher,U.S. Patent 2,800,457,1957. 8. J. L. Anderson, G. L. Gardner, and N. H. Yoshida, U.S. Patent 3,341,416, 9.

1967.

J. A. Bakan, in The Theory and Practice of Industrial Pharmacy, 2nd ed. (L. Lachman, H. A. Lieberman, and J. L. Kanig, eds.), Marcel Dekker, New York, 1986. 10. C. Thies, CRC Crit. Rev. Biomed. Eng.,1983,8:335-383 (1983). 1 1 . B. J. Floy, G . C. Visor, and L. M. Sanders in Polymeric Delivery Systems, ACSSymposiumSeries 520 (M. A. El-Nokaly, D. M. Piatt,and B. A. Charpentier, eds.), American Chemical Society, Washington, DC,1993, pp.

154-167. 12. H. B. Scher, in Pesticide Chemistry-Human Welfare and the Environment, Vol. 4 (J. Miyamoto and P. C. Kearny, eds.), Pergamon Press, Oxford, UK, 1983, pp. 295-300. 13. T. M. S. Chang, Artificial Kidney, Artificial Liver, and Artificial Cells, Plenum Press, New York,1978.


19

14. M. Cakhshaee, R. A. Pethrick, H. Rashid, and D. C. Shemngton, Polymer Commun., 26:185-192 (1985). E. 15. W. D.Jayne,inMicroencapsulation:ProcessesandApplications(J. Vandegaer, ed.), Plenum Press, New York, 1974, p. 103. 16. K. Dietrich, H. Herma, R. Nastke, E. Bonatz, and W. Tiege, Acta Polymerica, 40:243-251 (1989). 17. J. A. Reynolds, U.S.Patent 2,299,929, 1942. 8:40-44 (1983). 18. J. Brenner, Perfumer and Flavorist, 19. K.Lehmann, in Microcapsules and Nanoparticles in Medicine and Pharmacy (M. Donbrow, ed.), CRC Press, Boca Raton, FL, 1992, pp. 73-97. 20. J. T. Goodwin, and G . R. Somerville, Chemtech, 4:623 (1974). 21. R. E. Sparks, I. C. Jacobs, andN. S. Mason, in Polymeric Delivery Systems, ACSSymposiumSeries520(M. A. El-Nokaly,D.M. Piatt,andB.C. Charpentier, eds.), American Chemical Society, Washington, DC, 1993, pp. 145-153. 22. S. Hagenbart, Food Product Design, April:28-38 (1993). 23. R. Arshady, J. Microencapsulation, 10:413-436 (1993).

2 Microencapsulation of Oil-in-Water Emulsions by Proteins Shlomo Magdassi and Yelena Vinetsky Casali Institute ofApplied Chemisty, School ofApplied Science and Technology, The Hebrew University of Jerusalem, Jerusalem, Israel

Introduction I.

21

11. Methods of Encapsulation A. Phase Separation B. Spray Drying C. HeatDenaturation

23 23 25 25

1 1 1 . Surface Activity of Proteins

25

IV.

27

1.

PossibleFilm-Forming Proteins

INTRODUCTION

The formation of microcapsules is a widely used process that combines both basic chemical aspects and technological aspects. This process is usually very complex, since it involves various reactions and steps which take place atdifferent phases. For example, the initial step of a simple coacervation-based process isbased on adsorption of one of the wallforming components onto the core material-continuous phase interface. Therefore, the migration of the molecule from the bulk continuous phase to theinterface and the kinetics and reversibility of the adsorption process are of great importance.The second step is usuallybased on conversion of 21

22

and

Magdassi

Vinetsky

the adsorbed layer into a semisolid wall, with a defined and measurable thickness and permeation properties. This process involves the addition of other molecules, which lead to precipitation or coacervation onto the core material. Usually, this process is performed by adding an electrolyte (simple coacervation) or an oppositely charged polymer (complex coacervation). Since the first wall component may also be present in the bulk phase, and not only adsorbed at the core-continuous phase interface, it could also participate in the coacervation or precipitation process. It’is clear that such a situation is undesirable, as it results in waste of material, which could otherwise be used for the formation of the wall. Moreover, the free molecules in the continuous phase may also lead to aggregation of the individually dispersed particles or droplets of the core material, leading to formation of very large particles, which may sediment or float, and which are often unsuitable for most applications. The next step is usually based on hardening of the wall material by various crosslinking agents, which are added to the continuous phase. It is obvious that such a process should be carefully controlled, because, in addition to the hardening of the wall material of each microcapsule, it may also lead to crosslinking between several microcapsules, thus changing the physicochemical properties of the microcapsules. In most applications, it is desirable to obtain, a dry, microcapsules powder as a final product, which can be further redispersed in various liquid phases. Therefore, removal of the liquid continuous phaseis oftenrequiredand is achieved by a process such as spray drying or lyophylization. Such a process could often lead to agglomeration of the microcapsules, which should be prevented if a redispersion of the microcapsules is desired. These problems are often solved by including various additives in the liquid phase in a way that the removal of the liquid phase would cause the formation of an additional “barrier” between the individual microcapsules, which will disappear on redispersion. Obtaining dry, redispersible microcapsules is still not the end process. A caking phenomena that occurs as a result of agglomeration caused by the pressure gradient present in the storage containers may also be observed. This problem usually requires the use of anticaking agents, which can be added to the dry powder as well as to the final steps of removal of the liquid phase. The complexity of the process is more severeif the core material is also a liquid: In such cases, the microencapsulation process should also include an emulsificationstep in which a stable oil-in-water or water-in-oil emulsion could be obtained. The emulsification process is based on various parameters, such as interfacial tension, adsorption of the surfactant, or homogenization equipment. The use of a liquid core may also lead to manufacturing problems at the endof the process, during the removal of the continuous

Microencapsulation of Oil-inEmulsions Water

23

liquid phase, as in the case of encapsulation of volatile oils (such as orange oil), which are widely used as fragrance and flavor additives. This chapter deals with the formation of microcapsules in which the core material is a water immiscible oil and the continuous phase is an aqueous solution. The wall material described here is composed of various proteins and, in particular, gelatin. The principles for obtaining the microcapsules are similar, and therefore we will discuss in detail the various steps of the microencapsulation process with emphasis on the proteinbased microcapsules. II. METHODS OF ENCAPSULATION

Many different techniques have been proposed for the production of microcapsules, and it was suggested(1) that more than200 methods could be identified in the patent literature. Three methods which are significantly relevant to microcapsules that are based on proteins or polysaccharides, especially when the core material is an oil, are briefly discussed here. A.PhaseSeparation

As described in a publication by Finch [l], in the phase separation-based process, the core contents, in a solvent, are first suspended in a solution of the wall material. The wall polymer is induced to separate as a new, viscous, polymerrich phase by adding a nonsolvent, by lowering the temperature,by changing the pH,by adding a second polymer, or by changing other environmental conditions. It is essential that such changes would cause the polymer to come out of the solution and to aggregate around a core dropletto form a continuous encapsulating wall [2]. This process is shown schematically in Fig. 1, which includes the hardening of the coacervate layer, usually by a crosslinking agent. The coacervation of the polymer can be recognized bythe appearance of turbidity, droplets, or separation of the solution into two layers, which contain high and low concentrations of the polymer. Coacervation may be simple or complex: Simple coacervation is obtained by adding a water-miscible nonsolvent to theaqueous polymersolution (e.g., ethanol) or by adding an electrolyte, which causes polymer separation due to a “salting-out” mechanism. A typical example is the addition of sodium sulfate to an oil-in-water ( O M ) emulsion, whichis formed by using gelatin [3,4]. The system may be stabilized by altering the pH or the temperature by or addition of a crosslinking agent. Complex coacervation occurs with the mutual neutralization of two oppositely charged polymers. A widely used method is based on coacer-

Magdassi and Vinetsky

24

dissolved in water -

0

Emulsification

7I Film forming material, oppositely charged

Hardening of coacervate-capsule wall by chemical or physical means

-

V

Removal of water

Fig. 1 Schematic presentation of a formation of microcapsules of oil droplets in water by complex coacervation.

Microencapsulation of Oil-inEmulsions Water

25

vation of cationically charged gelatin (at pH below 8) by a negatively charged polymer such as gum arabic. This process was originallydeveloped for the formationof carbonless copying paper [5]. B. SprayDrying

Microencapsulationby spray drying is based on two steps: emulsification of the coreoil (or dispersion of particles) in the polymer solution followed by removal of the solvent by a hot stream of air. This method, in spite its limitations, is the most prevalent process used to encapsulate flavors [6]. The main problems relatedto this process are the nonuniformedsize of the microcapsules (due to coalescence of oil droplets at high temperature and aggregation of microcapsules) and incomplete encapsulation of the core component, which is often a volatileoil. The wall materials which can be used in sucha process can be various polysaccharides (starch, gum arabic) or proteins (gelatin, albumin, casein) [7,8]. It is obvious that while choosing the wall material, two important parameters should be taken into consideration: theability of the polymer to emulsify the oil phase and the behaviorof the polymer at high temperature. The latterhas a uniquemeaning regarding theuse of proteins, which can be denaturated specific at temperatures andhence form unsuitablewall material. C.HeatDenaturation Heat denaturation of proteins can also be used for encapsulation, as described recentlyin a patentby Janda et al. [g]. Briefly, the method is based on thefollowing steps:

1. Dispersionor emulsification of thecorematerialinasoluble protein slurry 2. Heating the slurry to form a protein melt followed by denaturation of the protein to render it insoluble in water in such a way that the insolublelayer coats the core droplets or particles The denaturation process can be achieved by heating only, bypH adjustment, orby using proteolyticenzymes which cause coagulation of the protein. It seems that this method could be suitable with various proteins, such as casein, albumin, corn and soy protein, wheat and rice gluten, and gelatin. 111.

SURFACE-ACTIVITY OF PROTEINS

All three methods described above have a common first step during the process of microencapsulation: adsorption of the proteinmolecule onto the

26

and

Magdassi

Vinetsky

oil-water interface. This adsorption may affect the emulsification of the oil phase (and hencethe droplet size distribution), the natureof the encapsulating wall (viapacking andinteractions between adsorbedmolecules), and the colloidal properties of the newly formed microcapsules (e.g., flocculation, zeta potential). Therefore, it is essential to understand the adsorption of proteins at oil-water interfaces, as well as the chemical parameters which may affect their surface activity. The proteins are macromolecules composed of a mixture of polar and nonpolar side chains and may have positive and negative electrical charges. These moleculesmay adsorb at theoil-water interface, through their hydrophobic side chains, and hencedecrease the interfacial tension of the emulsion system. The adsorption process is, in general, composed of three consecutive steps [lo]:(1) diffusion of the protein molecule fromthe bulk to the interface and attachment ontothe interface; (2) penetration of new (3) protein molecules, which have to overcome an energy barrier; and molecular rearrangement of the adsorbed molecules. It is interesting to note that the adsorption, or attachment at the interface, is achieved by small segments of protein having a unit surface area of 100-200 A2. The adsorption process obviously can lead to a drastic unfolding of the protein at theinterface, a phenomenon termedsurface denaturation. Under certain conditions, the protein present at the interface may coagulate; a result which may have implications on the formation of microcapsules by the protein-coated oil droplets. Since the first step in forming O/W microcapsules involvesan emulsification step, it is necessary to obtain an emulsion which has the required droplet size distribution and is stable, at least for the period of time required to complete the encapsulation process. Basically, it is preferred to use a protein which has some surface activity in terms of the initial adsorption and reduction of interfacial tension. Moreover, the proteins should be good emulsifiers, a property which can be obtained byusing nonrigid, charged protein that causes emulsion stabilization in both electrical and steric mechanisms. Such proteins are gelatin and casein, which are frequently used inprotein-based microencapsulation processes. If the emulsification step is achieved by using surfactants other than protein, the main property required of the coating protein is that it adsorbs onto the previously prepared emulsion droplets. Inthe case of simple or complex coacervation, it is required that the oil coacervate phase have suitable interfacial tensions, as shown by Arneodo etal. [ll],for gelatin coacervate phases and citrus oils. The surface activity of proteins is dependent on many factors [lo] such as themolecular parameters(MW, hydrophobicity, charge density and

Microencapsulation Emulsions Water of Oil-in-

27

isoelectricpoint,and rigidity) andthesolutionconditions(pH,ionic strength, temperature). Therefore,when forming protein-based microcapsules,oneshouldconsiderthe overall effectonboththeadsorption/ emulsification step and the formation of a wall around the oil droplets. IV. POSSIBLEFILM-FORMINGPROTEINS

As stated earlier, the of use certain proteins depends on the method used to prepare the microcapsules. From various publications, it appears thatthe number of proteins which can beused for microencapsulationis huge. In principle, any protein capable of forming a film is suitable for microencapsulationprovided the film can also be formed on the surface of the oil droplets. In a recent publication, Gennadioset al. [l21 review various edible coatings and films based on proteins. As shown, proteins such as zein, wheat gluten, soy protein, peanut protein, keratin, collagen, gelatin, milk proteins, andcasein can be used as film formers. Microcapsules were formed by emulsification and heat denaturation of egg albumin [13,14] around oil droplets; wheat gluten was used to encapsulate fat-soluble, edibleflavors by a spray-drying technique [15]; zein was used by dispersing the oil in alkaline protein solution,in which its pH was lowered to cause phase separation[16]; and casein was usedto encapsulate polyunsaturated fatsby spray drying [17-191. Qpical examples for theuse of various proteins to encapsulate oil droplets are given in Table 1. Of particular interestis the gelatin molecule, used either by itself or incombination with other componentsto form microcapsules [20]. Gelatin is a very abundant protein thatis produced from collagen of various origins. Gelatin is usually manufactured by two processes, “acid” and “basic,” which yield gelatin type A (positively charged below pH S) and type B (negatively charged above pH 5). The molecular weight of the proteins varies from 100,000 to 20,000, a property that is reflected in the bloom number, whichisindicative ofgel strength. Gelatin has unique surface activity and is a good emulsifier for various oil phases. Therefore, the various types of gelatin are often considered ideal candidates for the encapsulation process. For example, gelatinmay form capsules by the simple coacervation process, obtainedby addition of sodium sulfate [32], or by complex coacervation with gum arabic [7], sodium alginate [S], or even by using gelatin type Awith gelatin type B. We have recently reported [33] on the formation of gelatin-based microcapsules by formation of an insoluble complex of sodium dodecyl sulfate (SDS) with a positively charged gelatin. It was found that gelatin has the ability to bind anionic surfactants,such as SDS, and various soap molecules via electrostatic and hydrophobic interactions. At a certain mo-

28

v)

v1

0

E

Y

a

1"

i% C

.m

a 0

E

t-


E 0

.m

2 5

Y

m

Y

Q

Y

c 0 a

8

m

0

B

B

B

V

8

Y

.I

8

c)

.I

V

I;

0

E

W

N

I;

V

3

I;

3

c)

.I

Microencapsulation of Oil-in- Water Emulsions

E

c1

.I

0

M h

29


30

Soybean oil

Gelatin in water pH=6.1

T

V

Emulsification

non-adsorbed gelatin

Aqueous solution of SDS, pH=6.1

.

v

d

Mix

Crosslinking agents

c

V Microcapsules

V

Removal of water (lyophilization, evaporation)

Fig. 2 Schematic presentation of formation of microcapsules of oil droplets in water by interaction of anionic surfactant with a cationically chargedgelatin.

Microencapsulation Emulsions Water Oil-in- of

31

lar ratio of surfactant to protein, the protein becomes uncharged and, therefore, is insoluble in water. Further addition of SDS leads to redissolving of the gelatin-SDS inwater, probably due tohigh charge (negative) density. It was therefore assumed that if the precipitation of the insoluble gelatin-SDS complex could occur at the surface of the oil droplets, we could obtain microcapsule walls. Since both gelatin and SDS have surface activity, this process may combine the emulsification ability of the two components, with their binding properties. The formation of microcapsules by this method is schematically presented in Fig. 2. As shown,the process is based on threesteps: 1. Emulsification of the oil by usinggelatin type A as theemulsifier. The emulsions which are formed by simple homogenizer are stable for at least several days, having a mean droplet size of several micrometers. The pH adjustment is crucial, since the SDS binding is strongly dependent on thecharges of the protein molecule. After theemulsion is formed, the freegelatin should be removed simplybyrinsing the emulsion with water during filtration or centrifugation. 2. The second step is addition of SDS solution at a specific concentration, which is predetermined according to the optimal surfactant-protein ratio which causes precipitation. Fig. 3 gives an

! l

-

$I

-1

Fig. 3 Microcapsules of soybean oil prepared by interaction of SDS with gelatin type A.

32

Magdassi andVinetsky

example of such microcapsules which are obtained without further crosslinking. 3. The last step is the addition of a crosslinking agent suchas glutaraldehyde. The hardened wallof the microcapsules allows simple separation of the microcapsules from the continuous phase, andif the continuous aqueous phase is completely removed(by lyophilization or evaporation), a dry powder of microcapsules is obtained. From the description of this process, it is clear that formation of microcapsules of oil droplets can beachieved by various protein-surfactant combinations provided that either the protein or the surfactant has some emulsifying activity,and theinteraction between the two can lead to precipitation of a suitable film. REFERENCES 1. C. A. Finch, Spec. Pub1.-R. SOC. Chem. 138(Encapsulation and Controlled

2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

12. 13. 14. 15. 16. 17. 18.

Release):35 (1993). B. H. Kaye, Kona (Hirakata, Japan) 10:65 (1992). P. Spiegl, Austrian PatentDE 244, 1890 (1975). D. Fredj and F. Dietlin, French Patent 84-14653 840925, 1986. Y. Okumura, H. Takai, M. Shiozawa, and K. Kaneko, Japanese Patent 84143268 840712,1986. H. C. Greenblatt, M. Dombroski, W. Klishevich, et al., Spec. Pub1.-R. Soc. Chem. 138(Encapsulation and Controlled Release): 148 (1993). L. Y. Sheen, S. Y. Lin, S. J. Tsai, Zhangguo Nangye Huaxue Huizhi 30(3):307 (1992), C. Arneodo, J. P. Benoit, C. Thies,S. T. P.Pharma 2(15):303 (1986). J. Janda, D. Bernacchi, S. Frieders, U.S. PatentWO-91-US7278911004, 1992. S. Magdassi and N. Garti, in Interfacial Phenomena in Biological Systems (M. Bender, ed.), Marcel Dekker, New York, 1991, pp. 289-299. C. Arneodo, A. Baszkin,J.P. Benoit, and C. Theis, in Flavor Encapsulation (S. J. Risch and G . A. Reineccius, eds.), American Chemical Society, Washington, DC, 1988, pp. 132-147. A Gennadios, T. H. McHugh, C. L. Weller, et al., in Edible Coatings and Films to Improve Food Quality (J. M. Krochta,et al., ed.), Technomic Publishing, Basel, 1994, pp. 201-277. S. Soloway, U.S. Patent 3137631, 1964. J. Kosari and G. M. Atkins, U.S. Patent 3406119,1968. P. P. Noznick and C.W. Tatter, U.S. Patent 3351531,1967. C. Brynko and J. A. Bakan, U.S. Patent 3116206,1963. T. W. Scott, L. J. Cook, K. A. Ferguson et al., Austr. J. Sci 32:291 (1970). T. W. Scott, and G. D.L. Hills, U.S. Patent 4073960,1978.

Microencapsulation Emulsions Water Oil-in- of

33

19. G. J. Faichney, H. L.Davies,T.W.Scott et al., Austr. J. Biol. Sci 25:205 (1972). 20. L. S. Jackson and K. Zee, Lebensm. Wiss. Technol.24:289 (1992). 21. D. J. Mangold, W.W. Harlowe, andH. W. Schlamens, U.S. Patent 87-129503 871207, 1987. 22. Q.P. Corporation, Japanese Patent82-28757 820226,1983. 23. T. Kobayashi, French Patent 1568500 690523, 1969. 24. J. C. Soper, U.S. Patent 89401189,1991. 25. E. Bonatz, K.Golz, R. Nastke, German Patent90-342045 900625, 1991. 26. D. M. Whitaker, U.S. Patent 88-292495 881230,1991. 84-197776,1986. 27. S. Shimizu and H. Kato, Japanese Patent 28. R. M. Rawlings and D. Procter, Canadian Patent1086127 800923, 1980. 29. R. Shekerdzhiiski, S. Titeva, K. Mitikov et al., Formatsiya 35(4):23 (1985). 30. B. Sivik, J. Gruvmark, and M. Jakobsson, in Proc.-Scand. symp. Lipids,16th (G. Lambertsen, ed.), Lipidforum, Bergen,1991, pp. 147-152. 31. H. Jizomoto, E. Kanaoka, K. Sugita et al., Pharm. Res. 10(8):1115(1993). 32. J. Rosenblat, S. Magdassi, and N. Garti, J. Microencaps. 6515 (1989). 33. S. Magdassi andY. Vinetsky, J. Microencaps., accepted for publication.

Biodegradable Microspheres: Advancesin Production Techno Jean-Pierre Benoit Laboratoire de Phannacie Galbniqlue et BiophysiquePharmaceutique, Universitt? $Angers,and Centre dcE Microencapsulation, Angers, France

Hew6 Marchais Laboratoire de Pharmacie GaUnique et Biophysique Phannaceutique, Universitt? $Angers,Angers, France

Hew6 Rolland and Vincent Vande Velde Centre de Microencapsulation, Angers, France I. Introduction 11. Polymer Phase-SeparationMethods

111. Solvent Evaporation and Solvent Extraction Methods A. General Description of the Process B. Solvent Evaporation (Emulsification-Evaporation) C. Solvent Extraction (Emulsification-Extraction) IV. Spray-DryingMethods A. The NebulizingSystem B. Apparatus C. Applications D. AdvantagesandInconveniences V. Methods Using Fluids Under Supercritical Conditions A. Solvation of Compounds B. Precipitation of Pure Compounds C. Crystallization of Pure Compounds D. Formation of Microparticles VI. Milling Methods

36 36 43 43 45 49 50 52 53 53 55 57 58 59 60 60 62

35

Benoit et al.

36 1.

INTRODUCTION

Major advancesin the therapeutic sciences and substantial developments in the field of biotechnology have taken place over the past few years which have allowed the syntheses of peptides and proteins on industrial scales. The much valued pharmacological importance of these molecules has been somewhatoffset by their poor ability to penetrate thegastrointestinal barrier. Likewise, their high sensitivities and fragilities together with their very short biological half-lives seriously limit their therapeutic applications. The delivery of proteins and peptides for therapeutic purposes thus presents a major challenge to pharmaceutical scientists. Different strategies have been proposedto overcome these problems. Various modes of administration, such as the nasal or transdermal routes, have been investigated to facilitate the resorption of some proteins and peptides. However, the parenteral route remains the dominant means of delivery of peptides and proteins. From a formulation point of view, different drug-delivery systems involving, e.g., liposomes, mixed micelles, and nanoparticles, have shown much promise, as have controlled-release implants based on the use of biodegradable polymers. The success of the latter technique has contributed to the renaissance of the microencapsulation field. A large number of microencapsulation processes, and modifications of processes, have been reportedin the literature and in various patents [l].In this chapter, emphasis is placed on the description of selected processes leading to microparticle formation froma particular series of polymers used as coating materials; namely, poly(a-hydroxy-carboxylic acids) (PHCA). This category of polymers covers polymers and copolymers of lactic and glycolic acids (PLA, PLAGA),which are the sole excipients presently approved for use by international health authorities for designing biodegradable parenteralimplants. In referring to thevarious pharmaceutical formsdiscussed herein, the authors define microparticles as polymeric entities falling in the range of 11000 pm, covering two types of forms: Microcapsules, micrometric reservoir systems Microspheres, micrometricmatrix systems This chapter deals solelywith the preparation and use of microspheres. The potentials of the various technologies addressed hereafterare also discussed.

II. POLYMERPHASE-SEPARATIONMETHODS Polymer phase-separation, in nonaqueous media, by nonsolvents or polymer addition, also referred to as coacervation, is an excellent technique for

Biodegradable

37

the entrapment of water-soluble drugs such as peptides, proteins, or vaccines to beused as microencapsulatedpreparations. This technique has beenknown sincethe 1960s [2]. Since then numerous textbooks and articles have appeared describing the principles and preparation methodswhich, if adhered to, lead systematically to theformation of microcapsules [l-61. Under certain circumstances, it has been demonstratedthat this procedure leads to the production of microspheres. Thus, the coacervation of a polymer suchas poly(d,l-lactic acid-coglycolic acid) (PLAGA) dissolved in methylene chloride with a second polymer such as silicone oil allows the formation of matricial systems[7]. If crystals of active principles are placed in suspension at thebeginning of this process, they will be capturedin these matrices after the desolvation of PHCA. The coacervation process of a PHCA, as shown in Fig. 1, can be divided into four steps [7]:

1. In step 1 (Fig. lA), the amount of phase inducer added to the appearsto form a solution is low (l-5% v/v).Siliconeoil pseudoemulsion in the organic phase. 2. In step 2 (Fig. lB), for a greater amountof silicone oil introduced in the medium, the beginning of a phase separation is noticeable. The droplets of coacervate are unstable and merge together to give larger structures which expand andfinally rupture. 3. In step 3 (Fig. IC), the quantity of polymer added is sufficient to allow a stabledispersion of PLAGA coacervate droplets. 4. In step 4 (Fig. l D ) , extensive aggregation of coacervate droplets occurs which leads to their precipitation. To prepare wellindividualized microspheres, step 3 must be attained by carefully controlling the addition of the phase inducer to avoid massive aggregation. This sequenceof events can be conveniently schematized by a phase diagram. On such a diagram, each point corresponds to a defined weight percentage of methylene chloride, PLAGA, and silicone oil. Fig. 2 shows four ternary diagramsresulting from different PLAGA batches having very close mean molecular weights (Mgpc ranging from 35,000 to 55,000) but differing in the percentage of oligomers present: 2.0, 3.1, 4.5, and 40.0% for polymers 1, 2, 3, and 4, respectively. These ternary diagrams express the influence of the proportions of the three compounds on the sequence of events describing the PLAGA desolvation. In thefollowing discussion, we have designated step 3 as the “stability window.” Significant differences in the areas of stability windows are shown in Fig. 2. PLAGA 1 leads tothe largest zone, corresponding to step 3, whereas PLAGA 3 and 4 exhibit narrower windows. Increasing amounts of

38

Benoit et a/.

Fig. 1 Sequence of events occumng ina 5% PLAGA solution following the addition of 350 CSsilicone oil. silicone oil were required to be added to PLAGA solutions to induce the formation of stable coacervate droplets according to the sequence: polymers 4, 3,2 and 1, with polymer 1requiring the greatest quantity. These results concerning the influence of the physicochemical nature of PLAGA on the stability window indicate that a relationship exists between theamount of silicone oil added to induce the formation of stable coacervate droplets and the ratio of oligomers in polymer batches involved in the phase separation. The more monodisperse the copolymer, the more methylene chloride acts as a good solvent for the coating material. Consequently, increased quantities of silicone oil are needed to desolvate the copolymer. Thus, polymer 4, which contains the highest percentage of oligomers among all of the copolymers studied, does not readily dissolve in methylene chloride. In other words, it requires low amounts of silicone oil to be precipitated. Under such conditions, the formation of well-individualized microspheres is very difficult to control (i.e., small stability window).

Biodegradable Microspheres

39

SILICONE

SILICONE OIL

2c$Em

20 1

l .-c

METHYLENE CHLORIDE

MTEKLE~CHLORIDE

SILICONE

350 CS

1

/

-

METHYLENE CHLORIDE

Phase diagramsfor the coacervationof different PLAGA batches. step 2; step 3 (stability window); step 4.

step

From a formulation point of view, the width of the stability window can also be modified by changing the viscosity of the silicone oil. Fig. 3 illustrates the role of the viscosity of the silicone oil on the formation of stable coacervate droplets of polymer 1. For a low-viscosity-grade silicone oil (i.e., 20 CS), no stability window could be detected. By increasing the viscosity of the phase inducer up to a value of 12,500 CS,it was possible to enlarge the stability window. Above this limit, the silicone oil could not be handled for microencapsulation purposes. The evolution of the stability window versus the viscosity of the phase inducer is common for all the polymers studied. A phase inducer of 20 CS leads either to a medium of low viscosity that is unable to stabilize coacervate droplets or toits too rapid solvation by methylene chloride, which in turn induces uncontrolled precipitation of the wall material. These two phenomena may occur simultaneously, although in practice this is verydifficult to demonstrate. Silicone oils of higher viscosities produce the reverse effect and allow one to increase the area of the stability window.

40

Benoit et al.

-

METHYLENTCHL

METHYLENE CHLORIDE

M-NE

CHLORIDE

Fig. 3 Phase diagrams for the coacervation of pol mer 1. Influence of the phase inducer viscosity on the stability window. step 1; step 2; step 3 (stability window); step 4. It must be also noted that, as expected, the average molecularweight of the coating material also determines the existence, width, and displacement of the stability window within phase diagrams[8]. Fig. 4 summarizes the two main characteristics of a PLAGA batch which may influence the stability window and therefore theformulation of microspheres. Whenthe weight averagemolecular weight decreases, solvation in methylene chloride is good, and the amount of silicone oil necessary to coacervate the polymer is high. A displacementto theleft of the stability window is noted. Conversely, when the copolymer batchcontains low molecular weight compounds, i.e., oligomers, the overall hydroof more numerphobicity of the coating material decreases and the presence ous carboxyl and hydroxyl groups necessitates the use of smaller quantities of silicone oil. This induces a displacement to theright of the stability window often accompanied by a decrease in its area. The PHCA composition also exerts a strong effect on the stability window [g]. d,l-PLA appears to be less prone to desolvation induced by silicone oil addition than PLAGA copolymers of comparable molecular weights. The stability window which appears during the coacervation of d,l-PLA is therefore broadened compared with those of the copolymers.


41

METHYLEECHLORIDE eflicient polymer dissolution

WHEN

"-

bad polymer dissolution

" c -

kw DECREASESWHEN

HYDROPHOBICITY DECREASES

( /low

m : STABILITY

molecular weight compounds)

WINDOW

-

Fig. 4 Influence of polymer Mw and hydrophobicityon displacement of stability window. As a practical consequence,the isolation of well-individualizedmicrospheres is facilitated when d,l-PLA is used. The authors attribute this phenomenon to a nonrandom distribution of glycolic acid in the copolymer chains, since the solubility parameters of d,l-PLA and PLA25GA50 are quite similar: 16.4 and 16.8 (J m-3)112, respectively. Glycolic acidunits tend to form microcrystalline domains which are more sensitive to desolvation and precipitation. Numerous microparticular systems resulting from thecoacervation of a PHCA previously dissolved inhalogenated solvents have been described in the literature. A summary of the main operating conditions is given in Table 1. One must bear in mind that the investigations cited in Table 1 speak of "microcapsules," whereas in reality,matrix-type systems are more likely to be obtained. DCcapeptyl LP, 3.75 mg, a speciality polymer-drug preparation commercialized in Europe byIpsenA3iotech (France) and by Ferring (Germany), is prepared following this fabrication process. It consists ofmicrospheres of PLAGA 25 : 50 (25% l-lactic unit, 25% d-lactic unit, and 50% glycolic unit) containing an analogue of luteinizing hormone-releasing hormone (LHRH), triptoreline, used inthe treatmentof prostate cancer and endometriosis [19]. Formulation parameters have been adjusted to allow a constant release of triptoreline in vivoover 1month. This kinetic profile was extremely delicate to obtain. It is known, for instance, that a compromise

5

8

%

e 4 3

0

A Ae4

0

0

m

8I

r=! 0

N

R 8

Benoit et al.

Biodegradable

43

must be made between theintensity of the initial peptide loss and the time interval over which the sustained release occurs [20]. Again, the residual oligomer ratioin a PLAGA batch is one of the variables which can drastically change the release profile of the peptide. Lanreotide, which is a somatostatine analogue, has been approved and is in use in France for the treatment of acromegaly [21]. This drug is delivered from the same type of microsphere prepared according to the aforementioned technology. In conclusion, compared withspray-drying or solvent-evaporation methods, the polymer phase-separationtechnique may protect active principles from beingaltered by exposure to heat or from their partitioning out into dispersing phases. Residual solvent concentrations in resulting microspheres, however, can be very high, especially when heptane is chosen as the hardening agent [15,22]. 111.

SOLVENTEVAPORATIONANDSOLVENTEXTRACTION METHODS

The solvent evaporation technique was fully developed at the end of the 1970s [23]. Since then numerousstudies have been carried out on thebasis of this method. A. General Description of the Process

This technique is based on the evaporation of the internal phase of an emulsion by agitation. Initially, the polymeric supporting material is dissolved in a volatile organic solvent. The active medicinal principle to be encapsulated is then dispersed or dissolved inthe organic solution to form a suspension, an emulsion, or a solution. In the following step, the organic phase is emulsified under agitation in a dispersing phase consisting of a nonsolvent of the polymer, which is immiscible with the organic solvent, which contains an appropriate tensioactive additive. Once the emulsionis stabilized, agitation is maintained and the solvent evaporates after diffusing through the continuous phase. The result is the creation of solid microspheres. On the completion of the solvent evaporation process, the microspheres held in suspension in the continuous phaseare recovered by filtration or centrifugation and are washed and dried (Fig. 5) [24]. Although the concept of the solvent evaporation technique is relatively simple, the physicochemical phenomena governing this process are very complex. This system is characterized by the existence of several interfaces through which mass transfer occurs during particle formation (Fig. 6) [25]. The formation of solid microspheres is brought about by the

44

Benoit et al.

Y

Organic solution o f polymer with active principle dissolvcd o r dispcrsed with tensioactive agent

Dispersing medium

pq 0

0

/

Formation of emulsion under mechanical stirring

4)

Evaporation of solvent

Formation of solid microspheres

Fig. 5 The principle of the preparation of microspheres following the solvent evaporation technique.

evaporation of the volatile solvent L1 at interface L2/G. During the course of solvent evaporation, a partitioning is produced across the interface L1/ L2 from the dispersed phase to thecontinuous phase leading to theformation of solid microspheres. The partitioning across the interface L l L 2 is, however, not limited to the organic solvent; the active principle may also partition to some extent at this interface. The rapidity and importance of


45

Fig. 6 Schematic representationof the different interfaces in the solvent evaporation process: (1) Simple emulsion; (2) multiple emulsion.

Fig. 7 Differentmodes of incorporation of active principles in an organic solution of polymers. (a) Active principle dissolved. (b) Active principle dispersed. (c)

Active principle emulsified. these different transfer processes have a direct effect on microparticle formation. Thus, by studying the various governing parameters, the establishment of optimal conditions for theformulation of individual polymer/active principle couples can be achieved. Several systems may be envisaged based on the natureof the external phase (aqueous or nonaqueous), the incorporation mode of the active principle in the organic solution of the polymer (dissolved, dispersed, or emulsified) and the elimination procedure of the organic solvent (evaporation or extraction). (Fig. 7). The classification proposed by Aftabrouchad and Doelker [26] is in part adopted in the following discussion. '

B. SolventEvaporation(Emulsification-Evaporation)

1. Oil-in-WaterEmulsion Techniques calling on water as nonsolvent to thepolymer are in general preferred.In fact, theseprocesses are extremely economical and negate

46

Benoit et al.

the recycling of the external phase, which inturn results in the formationof well-individualized microparticles [27-291. In this system, the polymer is dissolved in an organic solvent such as methylene chloride or chloroform. The active principle is dissolvedor dispersed in the same medium, and then the entire mixture is emulsified inan aqueoussolution containing an appropriate surfactant. Beck et al. were thefirst to propose this procedure for the encapsulation of progesterone in PLA microparticles [23]. This technique, particularly suitable for theencapsulation of lipophilic active principles, has since been widely used for the encapsulation of different classes of therapeutic agents such as steroidal hormones [30-351, cytostatics [36-401, antiinflammatories [41-461, and neuroleptics [47,48]. Nevertheless, the microencapsulation of hydrophilic active principles by this process can pose problems. In actual fact, a partitioning phenomenon operates betweenthe dispersed and the dispersing phases which contributes to a substantial lowering of microencapsulation yields. The physicochemical characteristics of the active principle, such as its partition coefficient, its degree of ionization, or its surfactant character will play animportant role in its localization in one or otherof thetwo phases of the starting emulsion. Strategies permitting a reduction in the loss of water-soluble active principles have been proposed to increase microencapsulation yields. The solvation of medicinal substances in organic solutions of polymers can be achieved by the addition of hydrophilic cosolvents [49,50]. The advantage of solubilizing the active principle is related to the flexibility of producing particles of extremely small sizes(i.e., bordering onthe micrometer scale) whose internal structures are more homogeneousirrespective of the initial drug particle size. The water solubility of the active principle can be reduced by chemically modifying it prior its incorporation in the organic phase. This modification is simple to perform for water-soluble salts, as they are commonly lipophilic in their nonionized forms. Many medicaments have been encapsulated in this manner [51,52]. The chemical synthesis of lipophilic prodrugs fromwater-soluble species such as5-fluoro-2’-deoxyuridine[53] or cephradine [54] has likewise permitted higher microencapsulationyields. In this instance, however, structural modifications of the medicine may give rise to toxicological problems. One may also modifythe dispersing phase of the emulsion to reduce leakage of the active principle from the oily droplets. Several investigators have thereforesuggested saturating the continuous phase with the medicine [37,38,55-571, adjusting the pHof this same phase,or even addingelectrolytes [40]. It must be noted however, that thevalidity and effectiveness of these strategies vary from case to case. For example,chemical transformations only concern a limited number of active substances, and the saturation of external phases is onlyof value for substances with lowwater solubilities

Biodegradable

47

or inexpensive compounds. All the same, if the adjustment of the pH can prove advantageousfor ionizable medicines, this procedure generally accelerates thedegradation of polymers suchas poly(a-hydroxycarboxylic acid) by the hydrolysis of the ester bonds[55,58].

2. MultipleEmulsions:Water-in-Oil-in-Water Another approach has been proposed for the efficient encapsulation of water-soluble active principles by the solvent evaporation technique in aqueous continuousphases. In these systems, active principles to beencapsulated are incorporated in an aqueous solution, which is poured into a casting organic solution of the polymer to form an emulsion of the type water-in-oil (W/O). This primary emulsionisitselfemulsifiedin an external aqueous phase leading to a multiple emulsion of the type water-in-oilin-water (WlOnV). The organic phase acts as a barrier between the two aqueous compartmentspreventing the diffusion of the medicine towardthe external aqueous phase(Fig. 8 ) [29]. This technique was first proposed by Ogawa et al. [59-611 for the encapsulation of a peptide analogue of LHRH. This process proves much more effective when the watersolubility of the medicine is high(>900 mgl mL) [60] and partitioning between the organic phase is disfavorable. The same investigators have demonstrated thecrucial role of the viscosity of the primary emulsion (W/O) in the prevention of the diffusion of the active principle toward theexternal aqueous phase[59]. To this effect, they incorporated an agent (gelatine) capable of holding the active principle in the aqueous internalphase, whereas increasing this phase's viscosity. The effectiveness of such a formulation may nevertheless be compromisedby a clash of compatibility between theviscosity-enhancing agent (e.g., gelatine, pectine, agarose) and the active principle which limits its field of application. This process is particularly suitable for the encapsulation of active principles inweak doses, which are strongly water soluble (e.g., hormones, trophic factors) [62-651, and antigens [66-731. Using this technique, the Takeda Laboratories (Japan) havemarketedbiodegradable microspheres of PLA37.5GA.25 (37.5% 1-lactic unit, 37.5% d-lactic unit, and 25% glycolic unit) containing an analogue of LHRH, leuprolide acetate. In France, this speciality is marketed under the name of Enantone LPand is prescribed in thetreatment of prostate cancer along with DCcapeptylLP. In an identical manner, the release of the peptide is effected over a period of 1month [74]. 3. NonaqueousEmulsions This technique, also known under the name of oil-in-oil emulsion, is replaced by can berepresented by Fig. 5 if the continuous aqueous phase an oily phase. Naturally the dispersed phase (also of an oily nature) needs to betotally immiscible withthe continuousphase. The dispersing medium

'

48

et

Benoit

al.

Formation o f an crnuisic (water-in-oil type)

Aqueous solution

o f active principle .

..

L

Organic solution of polymer

Formation of cmulsion undcr mcchanical stirring (watcr-in-oil-in-water type)

3

Evaporation o f solvent

0

-

Formation of solid microspheres

Fig. 8 Preparation principle of microspheres according to the multiple emulsion technique (water/oiYwater). can by constituted by a mineral or vegetable oil or a nonvolatile organic solvent [57,75-771. Although this process essentially permits one to avoid the loss of water-soluble active principles, it has only been used for the encapsulation of a very limited number of active principles, including cytostatics [49,78-801, anti-inflammatories [81,82], antimalarials [83], and

Biodegradable

*

49

serum albumin [84,85]. A very interesting innovation was proposed by Iwata and McGinity [86]: multiphase microspheres of PHCA containing water-soluble drugs and proteins have been prepared using the multiple emulsion-solvent evaporation technique. A multiple emulsion of the W/O/ 0/0type was utilizedto first provide a protective barrier between the drug and the polymer-solvent phase and secondly to prevent the highly watersoluble drugs from partitioning out of the microspheres. To illustrate this, a drug was dissolved in an internal aqueous phase containing gelatin and Tween 80. This solution was then emulsifiedin soybean oil containing appropriate tensioactive agents. This primary emulsion was in turn dispersed in acetonitrile in which a PHCA was previously dissolved and a W/ 0/0emulsion was obtained. This emulsion was finallyadded to a hardening solution consisting of Span 80 and light mineral oil. It must be stressed that the protective oil phase and the external dispersing phase need to be immiscible withthe organic solution containing the polymer. In certain processes, solvent evaporation may be replaced by sublimation by the implementation of lyophilization after the emulsification step. Thus, DeLuca et al. dissolved an active principle as well as poly(glyco1ic acid) in hexafluoroacetone sesquihydrate and emulsified the mixture in carbon tetrachloride [87].The resulting medium was immersed in dry ice/ methanolandcooled to a temperature which froze the drug-polymersolvent phase and not the continuous phase. The suspension was then subjected to freeze drying, allowing the removal of the continuous phase solvent and solvent entrapped in embryonic microspheres. In addition to giving elevated microencapsulated yields for watersoluble components, this procedure canhelp prevent the eventual hydrolysis of the medicine or polymer. However, by comparison with aqueous emulsions, this technique exhibits a number of disadvantages related to the use of the nonaqueous solvents: these are often expensive and need to be recycled. Trace residues are oftendifficult to eliminate, which pose technological problems. C. SolventExtraction (Emulsification-Extraction)

In the emulsification-evaporation method, the organic solvent of the dispersed phaseof the emulsion iseliminated in two stages: 1. Diffusion of the solvent in the dispersingphase (solvent extraction) 2. Elimination of the solvent at the dispersing phase-air interface (solvent evaporation).

In theory, if one uses a continuous phase whichwill immediately extract the solvent(s) of the dispersed phase, the evaporation stage is

50

Benoit er al.

no longer necessary in the formation of microspheres. In practice, this can be achieved by using large volumes of dispersing phase with respect to the dispersed phase [88,89] or by choosing a dispersed phase consisting of cosolvents, ofwhich at least one has a great affinity forthe dispersing phase. One may even formulate a dispersing phase with two solvents in which one acts as a solvent extractor of the dispersed phase [68,90].

A hybrid technology may also be envisaged where emulsificationevaporation is initialized then interrupted before the volatile solvent is totally eliminated. At this point, the native microparticles are transferred into a large volume of the continuous phase(consisting mostoften of highpurity water) where remainingsolvent is eliminated by extraction. This method was advocated by Benita et al. [31] for the preventionof the development of active principle crystals at the microparticle surface during the solvent evaporation stage or in the continuous aqueous phase, leading to decreased efficiencies inthe microencapsulation of medicines. In fact, ithas been suggested that thetransfer of embryonic microparticles in a large volume of water leads to rapid extraction of organic solvent which is found at the periphery of dispersed globules which still contain some solvent. This leads to the formation of a polymeric barrier at the organic solvent-continuous phase interface, which slows downthe diffusion of the active principle from theorganic phase towardthe aqueous external phase and prevents crystal formation of the active principle [32,91]. An interesting modification of this technology was proposed by Leelarasamee et al. [92]. If the solvent extraction method is retained, the formation of dispersed globules is assured by the injection of a polymeric support solution containing the active principle in a solvent-extracting medium. The entire mixture is placed in a tubular system with reservoirs designed to permit the total extraction of the solvent from the polymer. This system replaces the mechanical stirring blade which assures emulsification in the preceding step. The concept behind this system is to enable the eventual production of microspheres ona continual basis.

IV. SPRAY-DRYINGMETHODS Nebulization, or spray drying, is widely used in the chemical, pharmaceutical, and foodindustries [93]. The principle of spray drying bynebulization rests on theatomization of a solution (containing the product to be dried) by compressed air or nitrogen through a desiccating chamber and drying across a current of warm air (Fig. 9).


L

I 8

I

Fig. 9 Laboratory spray dryer: (1) Drying air inlet + filtration; (2) heating; (3) desiccation chamber; (4) cyclone; ( 5 ) collector for drying powder-microspheres; (6) filtration + air outlet.(A) solution, suspension, emulsionto spray. (B) compressed or nitrogen air. (C) Spray nozzle (e.g., pneumatic, ultrasonic) Four separate phases may be distinguished: 1. Nebulization of the solution in theform of an aerosol 2. Contact of the nebulized solution with the warm air 3. Drying of the aerosol 4. Separation of thedriedproductandtheaircharged solvent

with the

This technique can be used to protect sensitive substances against oxidation (e.g., fish oils, essential oils, vitamins, colorants) [94]. When volatile substancesare encapsulated, the coating materialsmost often employed are maltodextrine and arabic gum.

Benoit et al.

52

A more recent application of this mode of drying has been adapted for the creation of matricial systems of the microspherical type starting from complex liquid mixtures comprising an active principle dissolved (or dispersed) with a polymer in an organic solvent. This type of process permits the isolation of microspheresor microcapsules depending on theinitial formulation, albeit in the form of a solution, a suspension, or anemulsion (Fig. 10).

A. TheNebulizingSystem W Oprincipal nebulizing systems exist[95]: 1. lbrbine method: nebulization by a rotating disk or rotary atomization 2. Atomizer method: nebulization by compressed air or pneumatic nozzle atomization

In the atomizer method,the internal diameter of the atomizer is often between 0.5 and 1.0 pm. Compressed air (or nitrogen) is then mixed directly in the central core along with the solution; nebulization thus being provoked simultaneously by the pressure applied to the interior of the liquid vein and by its ejection velocity from the nozzle. Compressed air subsequently arrives at the exit of the atomizer along with the liquid, inducing the shearing of the liquid vein. This system is convenient for the isolation of extremely fine particles ( 4 0 pm) from nonviscous solutions. In the turbine method, the pressure of the compressedair turns a disc with calibrated orifices at high speeds (25,000-35,000 rpm), provoking the

EMULSION SUSPENSION SOLUTION Liquid drug Polymer Solvent

Solid drug Polymer Solvent

Liquid drug Polymer Solvent

MICROSPHERE MICROCAPSULE

Fig. 10 '&pes of microparticlesobtained by spray drying accordingto the nature

of the active principle-polymer mixture.

Biodegradable

53

shearing of the solution from the moment of its passage through these orifices. The turbine methodis thus complementary tothe first technique, being adaptedfor very viscous solutions, and for the formation of particle sizes in the range of > l 0 pm. The turbine equipped with its nebulizers must havea desiccation chamber of >l m in diameter. B. Apparatus

A vast array of spray-drying equipment exists ranging from laboratory nebulizers (Minispray Dryer 190; Biichi,Switzerland) to pilot scale apparatus (model Minor Mobile; Niro Atomizer, Denmark) the first to productionscale equipment (Minor Productions; Niro Atomizer) and ultimately to drying towers of several meters in height. This process is thus industrializable, with users having at their disposal a range of accessories in function with the quantities of products to befabricated. Changes in particle sizes can thus beeffected in a progressive manner. Various factors, such as thermal exchanges and losses, variable yields, and the geometries of atomizers or turbines make it difficult to transform the laboratory-scale equipment to theproduction scale. Most reportsin the literature are based on results obtained from laboratory-scale equipment (Minispray Dryer190; Biichi). Being of reduced sizes, these models have limited evaporation capacities. Flow rates are typically of the orderof 15-20 mL min". The entrance temperature of the air (40-200°C) allows the user the choice of working with aqueous or organic solvents of low boilingpoints. Only a few industrial companies operate using equipment of greater dimensions [96,97], but notably none employs the atomization tower for the drying of powders. C. Applications

The formation of biodegradable microspheresby this technology, permitting the liberation of medicines over periods of months or longer from one unique injection, is a relatively recent development [96-1001. Polymers and copolymers of lactic and glycolic acids are used in the formation of some of these systems. To date, only bromocriptine (Parlodel LAR; Sandoz, Switzerland) is commercialized in the form of injectable microspheres, which originates from the spray drying method. This formulation is composed of PLAGA branched to D-glucose [96]. Other patents (busereline [loll, mifepristone [102]) make use of this process for the encapsulation of hydrophilic or lipophilic molecules, allowing their liberation over a period of 1 month after one injection only. The principal applicationsoriginating from this technology are detailed in Table 2.

54

I

0

m m

SE:

I

41

U

..

S

Y

0

3

..

E: c

8

Benoit et al.


55

D. AdvantagesandInconveniences

The interest in this technique lies, among other things, on the following factors: It can be likened to a continuous process, whereas, in the pharmaceutical industry, the preparationof batches is encountered. Certain models are able to discharge their material contentsin a dry, continual manner (during which thermal treatment processes for the drying of powders must be steadily maintained over the whole process and closely monitored). The residence times of products to be dried at elevated temperatures (>lOO"C) from aqueous solvents are very short. In the case of organic solvents, evaporation temperaturescan be even lower. For example, aworking temperature of 30-40°C is sufficient to evaporate methylene chloride. These parameters are important when considering the drying of thermally sensitive products. The majority of authors evaporate polymer-methylene chloride solutions at temperatures ranging between 50 and 70°C [90,98,105,108]. Particle size distributions areusually monodispersed with a Gaussian distribution more or less depending on the pulverizing head employed (e.g., pneumatic atomizer, disc) and the chosen nebulization parameters such as, for example, the pressure of compressed air, the internal diameter of the atomizer, theviscosity, and flow rate of the solution tobe nebulized. In the literature,particle sizes are commonly described as being less than 10 pm, since the geometries of atomizers and the viscosities of nebulized solutions are almost uniform. In published articles for certain polymers, for example, dl-PLA (209,000 MW) or PLAGA (22,000 M W ) [105], the obtention of microspheres is impossible when the Kquid vein cannot be sheared into fine droplets. In these cases, the final result is the formation of filaments. This phenomenonalso arises when the concentration of the polymer is increased in the organic phase, hence increasing the viscosity of the solution [98]. It also appears that theproduction of these filaments is more readily brought about employing polymers composed of linear chains [98]. The formation of these filaments at this point is also strongly influenced by the performance of the apparatus. The influence of the pressure of the nebulizing air, the geometry of the atomizer (the diameter of its orifice), and the flow rate of the nebulizing solution are important for filament formation. Nebulization of fine particles permits, under certain conditions, to keep active principles at the heart of polymeric matrices in amorphous forms.

56

Benoit et al.

On the other hand, Bodmeier and Chen[98]have observeda change in the crystalline form of progesterone nebulized on its own or in the presence of d,l-PLA. These state changes imposekinetic modifications of dissolutions, which may be exploited in pharmacokinetics. Spray drying thus permits the incorporation of lipophilic molecules in these types of copolymersin concentrations between 30 [l051 and 50% [98], withouttheappearance of crystals at particle surfaces. On theotherhand, for concentrations greaterthan 50%, or for hydrophilic molecules such as theophylline [98], the burst effect is more marked in the in vitro release kinetics. This can be explained for the former case by surface crystallizations and in the latter by weak interactions with hydrophobic polymers, which are not sufficient to prevent crystallizations. Residual levels of organic solvents are less than those from the emulsification and evaporation techniques. Levels of 0.1-0.2% can be found quoted in the literature [97]. Certain investigators have managed to arrive at concentrations of 0.05% and below [106]. For a pharmaceutical application, the concentration of organic solvent, by specified unit mass, must be inferior to thelevel set by the country of commercialization. For example, the standard set by the European Pharmacopaefor methylene chloride is less than 5OOppm (0.05%). Inferior residual solvent levels can be obtained by lyophilization of microspheres, and residual humidity levels are typically very low ( polymethylcyanoacrylate(PMCA) [16], Using a gel-permeation chromatographic method, it was shown that PACA nanospheres appearto be built by an entanglement of numerous small oligomeric subunits rather than by the rolling up of one or a few long polymerchains [17]. Chromatographic profiles suggested an uniform distribution of molecules with unexpected low molecular weights ranging between 500 and 1000. Furthermore, no monomeric residues were found. The low molecular weight of oligomeric subunits is consistent with later-discussed observations concerning excretion and metabolization of nanoparticles. The most significantadvantage of

Nanoparticles: Characterization Preparation and

187

alkylcyanoacrylate over otheracrylic derivatives previously used for preparing nanospheres is in the polymerization step. In contrast to other acrylic derivatives, requiring an energy input for this process that can affect the stability of the adsorbed drug, alkylcyanoacrylates can be polymerizedeasily without suchcontribution.

b. Drug incorporation and adsorption. A successful nanoparticle system may be one which has a high loading capacity to reduce the quantity of carrier required for administration. Two theoretical curves can be proposed to describe the adsorption of drugs onto nanoparticles: Langmuirian-type and constant partitioning-type-isotherms. In fact, it was found that nanoparticles can entrap a drug according to a Langmuirian adsorption mechanism, because of their large specific area [18]. The drug can either be incorporated into nanospheres during the polymerization process or beadsorbed ontothe surface of preformed particles. In the former case, the drug-polymer interaction may result in the covalent linkage of the drug with the polymer. This was observed with vidarabine, an antiviral agent, whose nucleophile N in positions 3 and 7 may playa role of initiator for the anionic polymerization mechanism of the cyanoacrylic monomer [19]. To a lesser extent, a similar interaction was described with peptidic compounds such as growth hormone-releasingfactor (GRF) [20]. In fact, with GRF, it was shownthat if the drugwas added very early after the beginning of the polymerization process (5 min), 50% of the peptide was found covalently linked to thepolymer. On thecontrary, when the peptidewas added later on(i.e., 60 min after the beginning of the polymerization), it was not chemically modified, but the loading capacity was poor. Thus, it was found that thereis a narrow window of time forthe addition of GRF to thepolymerization medium which results in both the preservation of the chemical structure of the peptide and a satisfactory drug-loading capacity [20]. In contrast, surface pressure and surface potential studies of polyisobutylcyanoacrylate monolayers spreadat theair-water interface in the presence of ampicillin revealed that theassociation which forms ampicillin with the polymer was weak and could be easily disrupted [21]. Based on the surface potential and surface pressure experiments, models of polyisobutylcyanoacrylate-ampicillin arrangements in the interfacial region have been proposed for the low- and high-polymer surface densities [22]. Concerning loading capacity of cyanoacrylic nanospheres, it was found that both the nature and quantities of the monomer used influenced the adsorption capacity of the carrier. Generally, the longer the alkyl chain length, the higher is the affinityof the drug. Taken in its entirety, the capacity of adsorption is related to the hydrophobicity of the polymer and

Couvreur et al.

188

to thespecific area of the carrier. Moreover, the percentage of drug adsorption generally decreases with the quantity of drug dissolved inthe polymerization medium, according to theLangmuirian isotherm[23]. Special attempts have been made for the association of synthetic fragments of DNA to PACA nanospheres [3]. In fact, the association of antisens oligonucleotides with nanoparticles was possible only inthe presence of a hydrophobic cation, such as triphenylphosphonium (Fig. 1) or quaternary ammoniumsalts. The poor yield of oligonucleotide association without cations could be explained by the hydrophilic character of nucleic acid chains that areknown to be soluble in water. In the proposed method, oligonucleotide adsorption on the nanosphereswas mediated by the formation of ion pairs between the negatively charged phosphate groupsof the nucleic acid chain and the hydrophobic cations [3]. The adsorption efficiency of oligonucleotide-cation complexes on nanospheres was found to be highly dependent on several parameters: oligonucleotide chain length, nature of the cyanoacrylic polymer, hydrophobicity of the cations used as ion-pairing agents, and ionic concentration of the medium [3]. When adsorbed onto nanospheres, oligonucleotides were found to be protected from the degradationby a 3'-exonuclease in vitro [3].

0

50

100

Hydrophobic ion

150

(PM)

Fig. 1 . Adsorption of oligothymidylate to polyalkylcyanoacrylatenanoparticles. Oligothymidylate (2 mM) complexed with increasing concentrations of positively charged tetraphenylphosphonium chloride (PP)was added to PIBCA (W) or PIHCA (+) nanoparticles (10 mg/mL). Adsorption of oligothymidylate to PIHCA nanoparticles in the presence of the negatively charged tetraphenylboron sodium (0)was measured under the same experimental conditions, as a control. (From ref. 3.)

Nanoparticles: Preparation and Characterization

c. Biodegradationanddrug

189

release. Another characteristic of cyanoacrylate polymers is that their rate of degradation is dependent on the length of their alkyl chain. The dominating mechanism of particle degradation was found to be a surface erosion process [24]. This process occurs through the hydrolysis by enzymes of the ester side chain of the polymer [25]. By this mechanism, the polymeric chain remains intact, butit gradually becomes more and more hydrophilic until it is water soluble. This degradation pathwayis consistent with the productionof alcohol during the bioerosion of PACA nanospheres in vitro in the presenceof esterases [25]. Indeed, the action of rat liver microsomes and tritosomes on the ester hydrolysis of polyisobutylcyanoacrylate nanospheres was clearly demonstrated [25]. A relationship between isobutanol production (i.e., nanosphere bioerosion) and esterase concentration was found. Therefore, we consider that the production of formaldehyde via chemical degradation (reverse Knoevenagel reaction) as proposed by several investigators [26] plays a marginal role in the overall degradation of PACA nanospheres under physiological conditions. Indeed, the amount of formaldehyde produced after nanospheredegradation was found to be5% of the theoretical quantity that would have been producedif the polymer had been entirely degrated by this pathway [25]. The suggestion of Vezin and Florence [26] that esterases do not affect the degradation rate is therefore contested. Since the biodegradability of polyalkylcyanoacrylate depends on the nature of the alkyl chain, it is possible to choose a monomer whose polymerized form has a biodegradability corresponding to theestablished program for drugrelease [14]. Indeed, by using a double radiolabeled technique (14Clabeled nanospheres loaded with [3H]actinomycin), it was found that drug was released from nanospheres as a direct ,consequence of the polymer’s bioerosion [25]. This was confirmed using GRF, another drug model[20]. With this peptidic compound, in the absence of esterases, no drugrelease was observed, whereas in the nanospheres suspension turbidity remained unchanged, indicating no polymerbioerosion. Drug release appeared to be dependent on the esterase content of the incubation medium. The liberation of GRF was quicker in rat plasma (content of esterases around 150 @mL) than in Ringer lactate medium containing esterases (100 pg/mL). At the same time, the turbidity (i.e., the expression of nanosphere bioerosion) decreased with the release of the peptide (Fig. 2). That the GRF release directly resulted from the polymer’s bioerosion was confirmed by the fact that themore slowly bioeroded polymerpolyisohexylcyanoacrylateyielded the slowest decrease of turbidity and the most progressive release of the peptide. Finally, the fact that the drugrelease was related to the turbidity decrease suggested that GRF was homogeneouslydispersed throughout the polymeric matrix rather than adsorbed atthe surface of the polymer.

Couvreur et a/.

790

GRF

f

f

f

"+

@

100

h

8

GiRF

GRF

400

80

300

v

W

v)

4

W

>.

-l

W

200

IT

2

s t5

60

40

m E 3

U 1oo

20

0

l-

0

0

2

6

4

8

10

Time (h) Fig. 2. Release of GRF from PIHCA nanoparticles is the consequence of polymer's bioerosion (decrease of the turbidity of the suspension). (From ref. 20.)

With rose bengal, a biphasic release from PACA nanospheres was observed [18]. The release profile of this compound was described by using a simple biexponential function: Pt = Ake-a

+ Be-b'

where Pt is the concentration of compound remaining in the nanosphere at time, t; the concentrations Aand B and the two rate constantsa and b can


191

be used to characterize the release of the compound. The initial release rate was faster when the marker was adsorbed only at the surface of the carrier, whereas it was reduced when the compound was incorporated into the polymeric networkof the nanospheres. By employing dextransulfate as the stabilizer in the preparation of PACA nanospheres, a slower release rate was obtained for the model drug rose bengal [27]. Thus, with this compound, the release was not only dependent on the kinetic of bioerosion of the polymer. B. Nanocapsules

After the development of nanospheres obtained by emulsion polymerization of alkylcyanoacrylates as described in the preceding section, it has been possible using the same monomer to manufacture particles of nanometric size containing an oil-filled cavity. This procedure, which was first developed by Al Khoury et al. [28], has beenapplied for insulin delivery byDamge et al. [29].The characterization of polyalkylcyanoacrylate nanocapsules has also been performed by different investigators [30-321. I . PreparationandCharacterization Nanocapsules of PACA are obtainedvia an interfacial polymerization process in emulsion. Nanocapsules are formedby mixing an organic phase with an aqueous phase. The organic phase is classically an ethanolic solution of the monomer mixed together with the oily core material and the lipophilic drug to be encapsulated and to which, in some cases, soya bean lecithin is added as an additional surfactant. Oils which have been used to prepare nanocapsules include Miglyol, benzylbenzoate, ethyl oleate, and Lipiodol. The encapsulation efficiency of a lipophilic drug generally depends on its partition coefficient between the oil and the aqueous phase,so the oil must be chosenaccordingly. The aqueous phase is a solution of a nonionic surfactant, usually Synperonic PE-F68 at 0.5% at a pH between 4 and 10. To form nanocapsules, the organic phase is allowedto runslowly-for example, through a wide-bore syringe needle or a micropipette tip-into the aqueousphase, which is magnetically stirred. The mixture immediately becomes opalescent. After stirring for 15-30 mins, the ethanolis removed by evaporation under reduced pressure. If required, the nanocapsules canbe furtherconcentrated by evaporation; this allows them to be rediluted in a physiological buffer for injection. Nanocapsules formedin this way have a mean diameter between200 and 300 nm witha narrow polydispersity. The speed of the magnetic stirring has noinfluence on theparticle size, which depends solely on the nature and

192

Couvreur et al.

the volumeof the oil and on thevolume of the diffusing organic phase. The proportion of oil to monomer must be correctly chosen to avoid simultaneous formation of either flakes of polymer or of a single oily emulsion. The presence of a surfactant in the aqueous phase is, in fact, not necessary for thesuccessful formation of nanocapsules, but it does prevent them from aggregating on storage to a cake which is difficult to disperse. Nanocapsules formedin this way are physically stable for several years at ambient temperature andmay be sterilized by autoclaving at 120°C for 20 min (the effect of such a procedure on any encapsulated drug must, of course, be considered.) However, as a result of their vesicular character, nanocapsules are not easily lyophilized, since they tend to collapse releasing the oily core. In fact, many factors could influence the type of colloid formed by anionic interfacial polymerization ofPACA: the nature of the aqueous phase, the pH, thecomposition of the organic phase (monomer, oil, ethanol), the ratio of monomer to the aqueous phase, and the emulsification conditions. The degree of polymerization and, therefore, the molecularweight, depend ona balance between initiation, propagation, and termination [33]. The number of growing chains depends on the concentration of initiators (OH-, CH,O-, CH,COO-, CN-). Forthe same quantity of monomer, when the numberof live chains is high, the degreeof polymerization (DP) is low. Similarly, DP is reduced when the concentration of terminating agents (H+) is high [34,35]. The concentration of the initiating and the terminating agents available for the monomer depends on thephysicochemical nature of the system in which the componentsare dispersed at themoment of the formation of the colloid. In thecase of a system composed of two immiscible phases, such as an emulsion, the presence and theconcentration of the solutes in the different phases is a function of the polarity of the solute molecules and of the dielectric constant of the medium. In contrast, at an interface between an aqueous and an organic medium, there can be large local variations in properties which can themselves cause changes in the interfacial properties in a dynamic system. For example, spontaneousemulsification depends on the interfacial tension. Thus, although the degree of polymerization depends on the propagation reaction, other characteristics of the colloid formed suchas particle size and morphology depend on interfacial phenomena inducted by the dynamic mixing of an organic phase with an aqueous phase. Thus, when the quantity of ethanol was increased, there was a tendency to form a suspension of polymeric flakes.However, with an increase in the volume of oil, an oily layer appeared onthe surface of the suspensions.


193

For systemshaving an oiVethano1ratio of some percentage, the macroscopic aspect is markedly more homogeneouswith the presence of a majority of nanocapsules. Since the preparation of nanocapsules of PACA is rather similar to the preparation of PACA nanospheres, it is necessary to verify that the colloidal suspension does not consist of a simple mixture of an oil-in-water emulsion containing polymeric nanospheres of similar size. Several different experiments confirmed that the preparation was indeed composed of oil-filled nanocapsules [32]. First, turbidity measurements showed a clear difference between nanocapsules and a mixture of nanospheres and the emulsion containing the same amount of polymer and oil. Second, the vesicular nature of nanocapsules prepared in this way was confirmed by electron microscopy. Scanning electron microscopy showed a smooth exterior to thenanocapsules,whereasthethincapsulewallandtheinternal cavity could be visualized by transmission electron microscopy [28]. Fig. 3 shows a polyisobutylcyanoacrylate nanocapsule after cryofracture andobservation by transmissionelectronicmicroscopy. The membrane thickness ranges from 5 to 10 nm. 2.

Mechanism of Formation Theoretically, nanocapsules result from the interfacial polymerization of the cyanoacrylic monomersat the surface of the oil droplets dispersed in the aqueous phase. A drugcan be encapsulated in the internalcavity of the polymerized membrane during the formation of the system provided that its oil-water partition coefficient is infavor of the oily phase. Nevertheless, a simple emulsification of an organic phase, containing alkylcyanoacryrate monomers and oil, in an aqueous phase does notallow nanocapsule formation. As a matter of fact, because of the affinity of alkylcyanoacrylates for the oily phase, a bulk polymerization of the monomers in the oil droplets occurs preferentially. Polymeric matrix systems, which can be compared with nanospheres in which the oil is dispersed throughout the polymeric network,result. To obtain nanocapsules, a dynamic process must be created which brings the monomer to the oil-water interface. This transfer is performed by the diffusion of a cosolvent from the organic phase to the aqueous phase. This cosolvent must be a solvent for the monomer and forthe oil on one hand and miscible with the aqueous phase on theother hand. Usually, ethanol can be used as diffusing solvent. So, nanocapsules are prepared by emulsificationof an organic phase containing the monomer in an aqueousphase. The main point is the compositionof the organic phase in which the oil and the diffusing solvent are mixed together. In the case of nanocapsules, the natureof the emulsification process is spontaneous.

194

Couvreur et a/.

Fig. 3. Polyisobutylcyanoacrylate nanocapsulesobserved by transmission electron microscopy after cryofracture. (From ref.32.)

In addition to their vesicular structure, the nanocapsules made from alkylcyanoacrylates differ from nanospheres by their size, the molecular mass of the polymer, and the characteristics of the drug release. Usually, nanocapsules are larger than the corresponding nanospheres and the degree of polymerization of the polymer is also significantly higher. These differences are theresult of the specific mechanismof nanocapsule formation. This mechanism is complex, becauseit results from boththe chemical characteristics of the polymerization and the physical characteristics of the polymerization medium (viscosity, interfacial tension, and cosolvent diffusion from the organic phase to theaqueous phase). These processes usually lead to theformation of mixed systems where nanocapsules coexist with other polymeric particles. Figure 4 shows photographs takenby transmission electronic microscopy of a nanocapsule preparation after separation by ultracentrifugation. Particles of lower density are true nanocapsules containing an oily cavity;


195

(b)

Fig. 4. Transmission electron micrographs of a nanocapsule preparation after separation by ultracentrifugation. (a) Floating layer.(b) Pellet. particles of higher density are in fact nanospheres with a size comparable to that of nanospheres obtained by emulsion polymerization. Nevertheless, the mechanism of formation of these nanospheres is not anemulsion polymerization, because the molecular mass of the polymer of these particles is high, similar to that associated with nanocapsules (about 100,000 versus about 1000 for nanospheres) [31]. Since the nanospheres obtained during the formation of nanocapsules have the same polymeric molecular mass as the nanocapsules, it is possible to conclude that thepolymerization mechanism is the same. Froma chemical point ofview, the anionic polymerization mechanism of alkylcyanoacrylate is not very much modifiedduring the formation of nanocapsules as

Couvreur et a/.

196

compared with nanospheres. It could be possible to consider that the nucleophilic initiation of alkylcyanoacrylate polymerization, which is normally performed by the hydroxyl ions of water, could be performed by ethoxy ions, although the dissociation constant of ethanol is very low. In fact, polymerization of alkylcyanoacrylates in the presence of ethanol is very restricted and leads to the formation of small oligomers with low molecular mass. Although the presence of ethanol seems not to modify the polymerization of the monomer from the chemical point ofview, the diffusion of ethanol is the major physical factor which imposes the conditions of polymerization and, thus, those of nanocapsules formation. Indeed, when the total volume of ethanol is kept constant, but is initially distributed differently between the organic phase and the aqueous phase, it appears thatthe molecular mass of thecreated polymer doesnotdependon the total amount of ethanol but only on that part of the ethanol diffusing from the organic phase to the aqueous phase. The very high molecular mass of the polymer making up the nanocapsules, is only obtained when the diffusing fraction of ethanol is more than half the total.The role of ethanol diffusion in the formationof nanocapsules is illustrated in Fin. - 5 . The rapid diffusion the aqueous phase leads to primary of ethanol from the organic phase to

Aqueous phase

2

Nanocapsules

/ 0

\

fragmentation of thepolymeric film

Marangoni effect:

69 @ @ Nanoparticles

interfacialpolymerization of monomer

Fig. 5. Mechanism of formation of nanocapsules

by interfacial polymerization.


197

convection currents, which are perpendicular to the interface, yielding an higher concentration of monomers on the peak of the surface undulations of the interface between the organic and the aqueous phases. Since alkylcyanoacrylate monomers have tensioactive properties, interfacial tension gradients appear on the interface, leading to secondary convection currents at a tangent to the interface (Marangoni effect). So primary and secondary convection currents promote spreading of the monomer at the interface where an interfacial polymerization will be able to propagate. This kind of polymerization leads to a higher molecular mass than ina single emulsion polymerization where the oligomers, being widelydispersed in the aqueousphase or in the micelles, have a greater probability of meeting cationic species (H,O+) which terminate the polymerization. In contrast, during interfacial polymerization, the oligomers are relatively protected from termination agents and the polymerization can develop owing to the monomer reservoir set up within the organic phase. The interfacial turbulences due to the secondary convection currents of the Marangoni effect initiate the fragmentation of the interfacial polymeric film and initiate the spontaneousemulsification of the system. During this emulsification, fragments of films fold back on themselves at the,interface around the oil droplets to form the nanocapsules, whereas the remainder of the film fragments lead to theformation of nanospheres of smaller size with a polymer molecular mass retaining the characteristics of interfacial polymerization (high molecular mass). The structure of the mixed systems (nanospheres and nanocapsules) formed during the preparation of nanocapsules is confirmed by the analysis of the drug-release kinetics. Fig. 6 shows the comparison of the in vitro release of tetracaine from polyisobutylcyanoacrylate nanospheres (true nanospheres) and from the two fractions (obtained after separation by ultracentrifugation) of the nanocapsule suspension (true nanocapsules and nanospheres). The similarity of the release kinetics of the two nanosphere systems confirms that they have an identical structure (matrix structure). Analysis of the release kinetics of nanocapsules, compared with those of nanospheres, shows that thetime constant of nanocapsules is superior to that of nanospheres which wouldhave the same diameter. This is coherent with the vesicular structure of the nanocapsules. 111.

NANOPARTICLES OBTAINED BY DISPERSIONOF PREFORMED MACROMOLECULES

Nanospheres A.

As presented above, nanoparticles can be obtained by inducing the polymerization of a monomer in certain conditions. However, many polymeric

Couvreur et al.

198 loo

I

80

nanwpsulcs obtained

0 nanosphcres obldncd

by inlerfacirl polymubtlon

* 0

2

h u e nannosphcrcs obtained

by emulsion polymerization

6

4

8

10

time (h)

Fig. 6. Invitroreleasekinetics nanoparticles.

of tetracain from polyisobutylcyanoacrlate

materials not capable of being prepared by emulsion polymerization, such as polyurethane, epoxy, polyester, and others,including semisynthetic polymers such as cellulose derivatives, remained unavailable for aqueousdispersion. This problem has led to the development of pseudolatexes (artificial latexes) obtained by dispersion of preformed polymers. For drug delivery purposes, following an intravenous injection, the polymeric material needs to meet physicochemical and biological requirements adapted and optimized for this specific application. Among these requirements, submicron size to avoid the risk of embolization following intravenous injection and biodegradability to nontoxic metabolites are of crucial importance, thus limiting the number of available polymers. If the method of preparation implies the use of nonsafe ingredients (solvents, stabilizers), steps of purification should be added in order to assure the required qualities of the final product. Most of the synthetic and artificial latexes preparedfor industrial applications did notmeetthese requirements,and various approaches were progressively developed


199

in order to obtain polymeric nanospheres fulfilling the pharmaceutical criteiia. Polymeric nanospheres prepared by emulsion polymerization may encounter somedrawbacks: With the exception of alkylcyanoacrylate, most of the monomers suitable for a micellar polymerization in an aqueous phaselead to slowly or nonbiodegradable polymers. The polymerization process is mainly limited to the vinyl addition reaction, and the molecular weight of the polymeric material cannot be fully controlled. Residues in the polymerization medium (monomer, oligomer, organic solvent, surfactant, initiator of polymerization) can be more or less toxic and may necessitate further purification of the colloidal material. During the polymerization process, activated monomer molecules may interact with the drug molecules, thus leading to theirinactivation (20). In order to avoid some of these limitations, the methods of obtaining nanospheres from various preformed macromolecular materials, whose physicochemical and biological properties can be well controlled, have been developed. Two approaches using the emulsification of a solution of the macromolecular material or using a controlled desolvation of this material have beensuccessively considered.

1. NanospheresPrepared by Emulsification The methods of preparing nanospheres byemulsification were adapted from theindustrial techniques available for obtaining the artificial latexes used in surface-coating applications (e.g., paints, adhesives, textilesizing, paper coating). Initially, artificial latexes were developed as waterbased dispersions in order to avoid economical and environmental problems encountered with the use of organic solvent-based coatings. They present the major advantage that any water-insoluble polymeric material can be used and all the industrial techniques for obtaining artificial latexes rely on oneof the threefollowing processes: 1. Solution-emulsification [36]: The polymer is dissolved ina volatile solvent immiscible with water. This organic phase is dispersed in an aqueous phase by conventional emulsification techniques and appropriate emulsifiers. The polymeric particles are formed from the oil-in-water ( O W ) droplets by removing the organic solvent following its vaporization.

200

b

Couvreur et al.

2. Phuse-inversion [37]:The polymer is first ground with a fatty acid. An alkaline aqueous solution is then added and mixed to form a water-in-polymer dispersion. Further addition of water leads to the inversion of phases and to a dispersion of the polymer in the aqueous solution. In this case, the emulsifier is the soap formed from the reaction between thefatty acid and the alkali. 3. Self-emulsification [38]:Some polar groups such as amine or quaternary ammonium, when attached to polymer molecules, may confer autoemulsioning properties and remove the need for any added emulsifier for the dispersion of the polymerin an aqueous phase. However, the presence of polar groups may also modify the water sensitivity, more specifically the pH sensitivity, of the colloid.

If pharmaceutical applications are considered, the emulsification techniques may present some limitations:

To obtain small ( ’. g . Otherapplications of microencapsulationby interfacial polymerization. A novel andinteresting application of semipermeable microcapsules was reviewed by 0’Neill[38l1 Magnetic ferric oxide.was.microencapsu1ated in a membrane consisting of a graft copolymer .of polyamide and poly(ethylenimine) (PEI). These were administered to animals together with

Microcapsules Polymerization by Complexation and

363

some known carcinogens, for example, nitrosoureas. The microcapsules were recoveredmagnetically and thespecies bound to thePE1 investigated in order tofollow metabolic changes andtypes of carcinogens occurring in the gut under various dietary situations. The microencapsulation of benzalkonium chloride based on the interfacial polycondensation of isocyanates has been reportedby Pense et al. [39-411. The use of thz HLB approach was shown to be appropriate for deciding the initial formulation [39,40]. An aqueous solution of benzalkonium chloride was poured into an oil phase (e.g., xylene or a blend of xylene and Miglyol) containing a polymeric surfactant (e.g., Hypermer A60). Polycondensationfollowed hydrolysisof the isocyanate and reaction between the formed amine and furtherisocyanate. The preparation and properties of polyurea microcapsules was reported by Yan et al. [42,43] usingcyanate and polyamine asthe monomers witha nonionic surfactant as the emulsifier. The same group [44] also studied polyamide microcapsules containing oily liquids. The preparation of oil-containing poly(terephtha1amide) microcapsules by interfacial polymerization has also been recently discussed by Alexandridou and Kiparissides [45]. 4. Studies of Microcapsules Prepared by Interfacial Polymerization Lee and Kondo [46] found at 20°C for KC1 a permeability of 10.2 X 10-7 cm S-' by following the diffusion though the membrane by an optical method. The same group [47] studied the effect of pH on thepermeability of poly(L-lysine-alt-terephthalic acid)microcapsules to electrolyte ions. A large increase in permeability in going from pH 4 to 6 (and above) was ascribed to an abruptincrease in the microcapsule size. Levy et al. [48,49] used Fourier transform infrared spectroscopy (FT" IR) to study the HSAmicrocapsules. By following the influence of reaction time, they were able to show after 5 min the progressive acylation of the hydroxy and carboxylate groups of HSA. Free amino groups were also determined by a back-titration method [48,50,51]. The interfacial polymerization of m-phenylenediamine andtrimesoyl chloride has been studied by Chai and Krantz [52] using the novel techniques of light reflection and pendant-drop tensiometry allowing rapid real-time assessment of the polyamide membrane formation.

B. Nonbiomedical Aspects of Microcapsules with Rigid, (Type 2 in Fig. 1) Nonpermeable Shells and Liquid Cores 1. The major application of microcapsules with rigid, nonpermeable and brittle membranes has been in the carbonless copy paper industry.

Whateley

364

2.

3.

4. 5.

111.

The underside of upper copies is coated with such inkcontaining microcapsules. These are ruptured by the pressure of writing, and the released ink interacts with the lower paper to leave the imprint. However, in general, such microcapsules are not prepared by interfacial polymerization. More specialized applications of this type of microcapsule allow preparation by the interfacial polymerization process, €or example, herbicide and pesticide microencapsulation for controlled release. As well as controlling the release, the active ingredient is also kept out of contact with the environment andhandlers. This topic is well covered in Wilkins [g]. Fragrances, for example, are microencapsulated for the “scratch ‘n sniff”-type application fixed to such articles as paper and “tee” shirts. Domestic gas samples are microencapsulatedto provide a simple and readily distributed means of indicating the smell of a gas leak. Pigments have been microencapsulated (e.g., see ref. 53). Duel1 and Pendergrass [54] have reported on the use of polyfunctional aziridines for interfacial polymerization for a variety of applications.

INTERFACIALADDITIONPOLYMERIZATION

Interfacial addition polymerization has features similar to interfacial polymerization, but the polymerisusuallycomposed of only one type of monomerand there is nootherproductformed in the reaction. This method, which does not generally require reactants in the aqueous droplets, should be particularly suitable for enzyme microencapsulation. However, in general, solid microspherestend to beproduced ratherthan aqueous-core-type microspheres. Alkylcyanoacrylates A.

There has been considerable interest in the use of the alkyl-cyanoacrylates to form microcapsules and nanoparticles for drug delivery and targeting by the groups of Couvreur andKreuter. Florence et al. [55]have also reported the preparation of biodegradable microcapsules from poly(alky1-2-cyanoacrylate) monomers. The anionic addition polymerization of analkyl cyanoacrylate is shown in Fig. 6. This process could be initiated, for example, by water in the aqueous phase of an water-in-oil emulsion with the cyanoacrylate dissolved in the oil phase.

Microcapsules by Polymerization and Complexation

365

CN I CH, = C -COOK

CN

I

1

CN CN I I HO-CH,- C - CH, - C - ClOOR I COOR

Fig. 6 Reaction for formation of poly(cyanoacry1ate).

Both the groups of Kreuter and Couvreur (e.g., seerefs. 56 and 57) have developedthese systems overanumber of years. Forexample, Kreuter et al. [58] studied the degradation of poly(butylcyanoacry1ate) and poly(hexylcyanoacry1ate) nanoparticles and their association with and toxicity toward isolated hepatocytes. Formaldehyde has been regarded as being the major problem in considering the toxicity of poly(cyanoacry1ate)(PCA) systems.However, the (low)toxicity toward hepatocytes (which did not take up nanoparticles to any significant extent) could not be attributed solely to formaldehyde during degradation. Leonard et al. [59] studied the degradation of the poly(alkylcyanoacry1ates) in detail. Troster and Kreuter [60] studied extensively the body distribution of radiolabeled poly(methyLmethacry1ate) nanoparticles with and without surface surfactants. Poloxamine 1508 was found to be the most effective in reducing liver uptake. The anticancer drug mitoxantrone was loaded into PCA nanoparticles and found to cause a significant volume reduction in the B16 tumor in rats [61]. Harmia et al.[62] and Harmia-Pulkkinen et al. [63] encapsulated pilocarpine in poly(butylcyanoacry1ate) nanoparticles for drug delivery to the eye. The effects of monomer and stabilizer concentrations on the properties of poly(buty1-2-cyanoacrylate) nanoparticles was investigated by -Nonso et al. [M].

Whateley

366

Poly(cyanoacry1ate) nanoparticles have been evaluated as drug carriers by other groups also; for example, Douglas et al. [65] and Puglisi et al. (661Detailed studies of the particle size distribution of poly(cyanoacry1ate) nanoparticles have been reported [67,68]. IV. INTERFACIAL COMPLEXATION OR INTERFACIAL POLYELECTROLYTE COMPLEXATION

A. AlginatelPolylysineMicrocapsules This process was introduced by Lim and Sun [69] and has been developed by their groups and Sefton et al. (e.g., see ref. 71). The topic has been reviewed recently by Sun et al. [70] and Chang [3]. The fact that calcium ions crosslink the alginate species (Fig. 7), initially in solution as the sodium salt, allows droplets of gel to be formed in the calcium chloride solution. The addition of an oppositely charged polyelectrolyte (in this case

poly(mannuronic) acid

poly(gu1uronic)

acid 0

poly(mannuronic) Coolt

- poly(gu1uronic) COOH

Fig. 7 Structures of alginate components.

acid

Microcapsules Polymerization by

and Complexation

367

the polylysine polycation) allows a layer of insoluble polyelectrolyte complex to form at theinterface. The calcium ions remaining in the interiorof the microcapsule are exchanged with an excess of sodium ions, rendering theinterior fluid again. Thus,an aqueous-cored, semipermeable microcapsule is formed under mild conditions (i.e., no extremes of pH, no chemical reactions). The polyelectrolyte complex membrane has proven to be very biocompatible, an aspect covered in more detaillater. l . Preparation of Microencapsulated Cells Cells are suspended in 0.8% sodium alginate (structure shownin Fig. 7) in saline: Droplets are extruded through a syringe into 1.5% calcium chloride solution. The crosslinking of the alginate macromolecules by the calcium ions causes the droplets to become gels (withthe cells entrapped in the gel sphere). The separatedgel microspheres are then resuspendedin a polylysine solution (0.02%, molecular weight, 35,000) for 5 min. An interfacial macromolecular complexation occurs at the surface of the gel microsphere, forming an insoluble (but semipermeable membrane). The gel microspheres are separated and resuspended in saline at pH 7.4; the calcium ions diffuse out of the gel interior of the microcapsule causing the contents to liquify. These microcapsules tend to be large (>l00 pm)perhaps macrocapsule is a better term? Mannuronic-rich alginate has been shown to link more strongly to the polylysineby Dupuy et al.[72], who used FT-IR to study polylysine/ alginate membranes containing various contents of mannuronic or guluronic acid residues (structures in Fig. 7). A novel method of preparing small alginate/polylysine microcapsules (in the range 5-15 pm) was reported by Kwoket al. [73]. Vaccine-containing microcapsules were preparedusing aTbrbotak air atomizer to spray sodium alginate solution into calcium chloride solution to form temporary microgel capsules which are subsequently crosslinked with polylysine. A different polyelectrolyte complexation reaction was used by Stange et al. [74] to encapsulate liver microsomes. The polyanion cellulose sulfate (sulfate substitution ratio 0.4-0.5)wasusedwith the polycation poly(dimethyldiallylammonium chloride) (molecular weight 40,000)The resulting membrane was semipermeable with an exclusion limit varying from 25,000 to 150,000 molecular weight depending on the preparation conditions. Increased enzymic activity and increased wall strength were claimed for this procedure. The microencapsulation and in vitro performance of various mammalian cells in polyacrylate membranes has been reviewed by the Sefton group at Toronto [71]. The cell suspension is coextruded with the polymer

368

Whateley

solution through a concentric needle assembly and the concept of this novel way for the controlled delivery of therapeutic agents is discussed. Drug encapsulation in alginate microspheres has been reported by by an emulsiWan et al. [75]. Calcium alginate microspheres were produced fication process and hardenedwith isopropyl alcohol or acetone. However, drug loading was found to below.

2. Applications of AlginatelPolylysine Microcapsules Because of the mild conditions in preparation and the biocompatible nature of the formed microcapsules, the vast majority of applications of microcapsules prepared by the interfacial complexation method are in the pharmaceutical, biomedical, and medical fields, and only these applications will be considered. These applications have been reviewed by Sun et al. [70] and Chang [3]. The transplantation of mammalian cells encapsulated in a protective, biocompatible, and semipermeable membrane has greatclinical potential for a range of diseases requiring enzyme- or cell-replacement therapy. The interfacial complexation process has been used to encapsulate islets of Langerhans, the cells which secrete insulin, for the treatment of diabetes. Limited success has been achieved [69,76,77]. Hepatocytes have also been microencapsulated by this process for the treatment of liver failure [78]. Another majorapplication of microencapsulated cells has beenin the production of monoclonal antibodies. The hybridoma cells, producting the monoclonal antibodies, are entrappedin the microcapsules: nutrients, oxygen, and so forth can enter through the semipermeable membrane. The monoclonal antibodies secreted by the encapsulated hybridoma cells remain trapped inside the microcapsules and accumulateto a greater concentration thanin the absenceof encapsulation. To recover the antibodies, the microcapsules are separated from the culture medium and ruptured to release the antibodies. In theabsence of encapsulation, the antibodies must be isolated from a large volume of culture medium.

3. Permeability of AlginatelPolylysine Microcapsules CoromiliandChang [79] haveused a heterogeneousmixture of dextrans with molecular weights 10,000-500,000 to study the membrane permeability of alginatelpolylysine microcapsules. Hybrid artificial organs must permit the diffusion of smaller molecules, includingpeptides and proteins, but must exclude leukocytes and immunoglobulins( M W >150,000) Fluorescein isothiocyanate-labeled dextrans were used by Vandenbossche et al. [80] to determine the molecular weight cut-off of similar microcapsules. The molecular weight cutoff of alginatelpolylysinemicrocapsules has


and Complexation

369

been shown to be dependent on preparative conditions [81]. Microcapsules prepared at 40°C showed a lower MW cutoff when the concentration of polylysine wasincreased from O.l%(w/v) The ultrastructural characterization of alginatellysine microcapsules was reported by Shimi et al. [82] Poly(1ysine) ofmolecular weight 22,000 was found to be optimum in forming robust microcapsules whichwere relatively impermeable to high molecular weight species such as immunoglobulins. 4. Biocompatibility of AlginatelPolylysine Microcapsules The topic of biocompatibility is a huge area in its own right: A few recent references are discussed here concerning the biocompatibility of alginate/polylysine microcapsules. The biocompatibility of alginate/polylysine microcapsules was enhanced by Sawhney et al. [83] by photopolymerizing a PEG-based hydrogel onto thesurface of alginate/polylysinemicrocapsules. A less inflammatory response with no fibrotic response was reported. The effect of composition on biocompatibility was examined by Clayton et al. [84], who found that in the peritoneum, high mannuronic acid alginate capsules provoked the weakest response. Islets of Langerhans inmicrocapsules of alginate/polylysine have been proposed as a artificial pancreas (e.g., see ref. 85), who studied the capacity of such microcapsules to activate macrophages in order tounderstand the foreign body reaction withfibrosiswhich has been observed around implanted microcapsules. The host reaction against alginate/polylysine microcapsules has been studied with both emptymicrocapsules [86] and with microcapsules containing living cells [87]. Histological evaluation showed a significantly higher number of cells stickingto microcapsules containing cells as comparedwith empty microcapsules. REFERENCES 1. R. Arshady, Biodegradable microcapsular drug delivery systems: Manufacturing methodology, release control and targeting prospects, J. Bioactive Compatible Polymers, 5:315-342 (1990). 2. R. Arshady,Preparation of microspheresandmicrocapsulesbyinterfacial polycondensation techniques, J. Microencaps., 6:13-28 (1989). 3. T. M. S. Chang, Recent advances in artificial cells based on microencapsulation, in Microcapsules and Nanoparticles in Medicine andPharmacy (M. Donbrow, ed.), CRC Press, Boca Raton, FL, 1992, pp. 323-339. 4. F. Lim, (ed.), Biomedical Applications of Microencapsulation, CRC Press, Boca Raton, FL, 1984.

370

Whafeley

5. M. Donbrow, (ed.), Microcapsules and Nanoparticles in Medicine And Phar-

macy, CRC Press, Boca Raton, FL, 1992. 6. T. L. Whateley (ed.), Microencapsulation of Drugs, Harwood Academic Publishers, Reading, UK, 1992. 7. T. M. S. Chang, Semipermeable microcapsules, Science, 146524-525 (1964). 8. T. M. S. Chang, F. C. Macintosh, andS. G . Mason, Semipermeable aqueous microcapsules, Can.J. Physiol. Pharmacol., 44:115-128 (1966). 9. R. M. Wilkins, (ed.), Controlled Delivery of Crop-Protection Agents, Taylor and Francis, London, 1990. 10. P.W. Morgan,CondensationPolymers:InterfacialandSolutionMethods, NY, PolymerReviewsSeries N0.10, IntersciencePublishers,JohnWiley, (1965). 11. F. Millich and C. E. Carraher, Jr., (eds.), Interfacial Synthesis, Volume 1: Fundamentals, Marcel Dekker,N Y , USA (1977). 12. Y. Wakamatsu, M. Koishi and T. Kondo, Microcapsules 17: Effect of chemical structureof acid chlorides and bisphenolson the formation of polyphenyl ester microcapsules, Chem. Pharm. Bull. 22,1319-1325 (1974). A. N.Hambly,Kineticsof interfacial 13. J. H.Bradbury, P.J.Crawfordand polycondensation reaction, Trans. Far. Soc. 64, 1337-1347 (1968). 14. M.Koishi, N. FukuharaandT.Kondo,Microcapsules2:Preparationof polyphthalamide microcapsules, Chem. Pharm. Bull. 17,804-809 (1969). 15. M. Koishi, N. Fukuhara and T. Kondo, Studies on microcapsules 4: Preparationandsomeproperties ofsulphonatedpolyphthalamidemicrocapsules, Can. J. Chem. 47,3447-3451 (1969). 16. Y. Shigeri, M. Koishi, T. Kondo, M. Shiba andS. Tomioka, Microcapsules 6: Effect of variationsin polymerization conditionson microcapsule size, KolloidZ Z.Polym., 249,1051-1055. Strength of interfacial polymerization films, 17. K. J. Mysels and W. Wrasidlo, Langmuir, 7:3052-3053 (1991). Etuk, andK.R.Murray,Formationofnylon6,lO 18. L.R.Weatherley,B. capsules by interfacial polymerization in a high-voltage electric-field, Powder Technol., 65227-233 (1992). T. Kondo,Preparation ofpolyamidemi19. N. Muramatsu,K.Shiga,and crocapsuleshavingnarrowsizedistributions,J.Microencaps.11:171-178 (1994). Levy, Polyhydroxamicmicrocapsules 20. D. Hettler, M.C.Andry,andM.C. Mipreparedfromproteins--Anoveltypeofchelatingmicrocapsules,J. croencaps., 11:213-224 (1994). 21. M.C. Levy, D. Hettler, M.C.Andry,and P. Rambourg, Polyhydroxamic serum albumin microcapsules: Preparation and chelating properties, Int. J. Pharm., 69:Rl-R4 (1991). New 22. T. M. S. Chang and R. Geyer (eds.), Blood Substitutes, Marcel Dekker, York, 1989. 23. T. M. S. Chang, Blood compatible coatingsof synthetic immunoadsorbents, Trans. Am. Soc. Intern. Artif. Organs, 26:546 (1980).


Complexation and

371

S. Chang, P. Barre,and S. Kuruvilla, Long termreducedtime haemoperfusion-haemodialysiscompared to standarddialysis,Trans.Am.

24. T.M.

Soc. Intern. Artif. Organs, 31572 (1985). 25. D. A. Wood, T. L.Whateley,andA. T. Florence, Microencapsulation of histidaseforenzymereplacementtherapy,J.Pharm.Pharmacol.,31:79P (1979). 26. T. Kondo, Enzyme inactivation in microencapsulation, in Microencapsulation (J. R. Nixon, ed.), Marcel Dekker, New York, 1976, pp. 67-75. T. L. Whateley,Astudyofenzymeand protein micro27. D.A.Woodand encapsulation-somefactorsaffecting the low apparentenzymicactivity yields, J. Pharm. Pharmacol.,3432-557 (1982). 28. M. Kidokoro, H. Ohshima, and T. Kondo, Interaction of poly(L-lysine-aftterephthalic acid) microcapsules with fibrinogen, J. Microencaps. , 8:63-70 (1991). on the thermostabilityof 29. K. Hoshino, N. Muramatsu, and T. Kondo, A study microencapsulated glucose oxidase, J. Microencaps. ,6:205-211 (1989). 30. M. Monshipouri and R.J. Neufeld, Activity and distribution of urease followingmicroencapsulationwithinpolyamidemembranes,EnzymeMicrobial Techn., 31:309-313 (1991). 31. R. B. Aisina, Effectof microencapsulated enzymes, in Microencapsulation of Drugs (T. L. Whateley, ed.), Harwood Academic Press, Reading, UK, 1992, pp. 215-231. 32. A. M. Wallace, Novel immunoassays for clinically important hormones based on microencapsulatedantibodies,inMicroencapsulationofDrugs(T.L. Whateley, ed.), Harwood Academic Press, Reading, UK, 1992, pp. 241-255. 33. L. A. Luzzi, M. A. Zoglio, and H. V. Moulding, Preparation and evaluation J. Pharm. Sci., 59:338of prolonged release properties of nylon microcapsules 341 (1970). 34. A. T. Florenceand A. W. Jenkins,Invitroassessmentofmicroencapsulated drug systems as sustained release parenteral dosage forms, in Microencapsulation,(J. R. Nixon,ed.),MarcelDekker,NewYork,1976,pp. 39-55. 35. G. W. Cuff and J. W. McGinity, Expanded versatility of microcapsules prepared by interfacial polymerisation,J. Microencaps. 1:343-347 (1984). 36. G. W. Cuff, A. B. Combs, and J. W. McGinity, Effect of formulation factors on the matrix pH of nylon microcapsules, J. Microencaps., 1:27-32 (1984). 37. U. Chiba, E. W. Neuse, J. C . Swarts, and G. J. Lamprecht, Water-soluble polyamides as potential-drug carriers. 7. Synthesis of polymers containing intrachain-type or extrachain-type amine ligands by interfacial polymerization, Angewandte Makromolekulare Chemie, 214:137-152 (1994). 38. I. O’Neill. Reactive microcapsules for detectionof carcinogen sources in the gut, J. Microencaps., 10:283-308 (1993). 39. A. M. Pense, C . Vauthier-Holtzscherer, and J.-P. Benoit, Microencapsulation of benzalkonoium chloride, in Microencapsulation of Drugs (T. L. Whateley, ed.), Harwood Academic Press, Reading, UK, 1-5.

372

Whateley

40. A. M. Pense, C. Vauthier, F. Puisieux, and J.-P. Benoit, Microencapsulation of benzalkonium chloride, Int. J. Pharm.,81:lll-117 (1992). 41. A. M. Pense, C. Vauthier, and J.-P. Benoit, Study of the interfacial polycondensationofisocyanateinthepreparation of benzalkoniumchloride loaded microcapsules, Colloid Polymer Sci., 272:211-219 (1994). Li, Size distribution and zeta-potential of micro42. N. Yan, M. Zhang and P. capsules. J. Membrane Science, 72, 163-169 (1992). of polyurea mi43. N. Yan, P. Ni, and M. Zhang, Preparation and properties crocapsules with non-ionic surfactant as emulsifier. J. Microencaps., 10:375383 (1993). on polyamide microcapsules containing 44. N. Yan, M. Zhang, and P. Ni, Study oily liquids, J. Microencaps., 11:365-372 (1994). 45. S. AlexandridouandC.Kiparissides,Productionofoilcontainingpoly(terephthalamide) microcapsulesby interfacial polymerisation: Effectof process variableson the microcapsule sue distribution,J. Microencaps., 11:603614 (1994). 46. K. B. Lee, and T. Kondo,Permeabilitycoefficientofmicrocapsulemembrane, J. Microencaps., 199-356 (1984). on pH of permeabil47. E. Miyauchi, Y.Togawa, K. Makino, et al., Dependence poly(L-lysine-alt-terephthalic acid)microitytowardselectrolyteionsof capsule membranes, J. Microencaps., 9:329-333,1992. 48. M. C. Levy, S. Lefebvre, M. Rahmouni, et al., Fourier-transform infrared spectroscopic studies of human serum-albumin microcapsules prepared by interfacialcross-linkingwith terephthaloylchloride-influence of polycondensation pHon spectra and relation with microcapsule morphology and size, J. Pharm. Sci., 80578-585 (1991). 49. M. C. Levy, S. Lefebvre, M. C.Andry, et al., Fourier-transform infrared spectroscopic studies of cross-linked human serum-albumin microcapsules. 2. Influence of reaction-time on spectra and correlation with microcapsule morphology and size, J. Pharm. Sci., 83:419-422 (1994). 50. F. Edwards-Levy, M. C. Andry, and M. C. Levy, Determination of free amino group content of serum-albumin microcapsules using trinitrobenzenesulfonic acid-Effect of variations in polycondensation pH, Int. J. Pharm., 96:85-90 (1993). 51. F. Edwards-Levy, M. C. Andry, and M.C. Levy, Determination of free amino group content ofserum-albuminmicrocapsules.2.Effectof variationsin reaction-timeandinterephthaloylchlorideconcentration,Int.J.Pharm., 103:253-257 (1994). 52. G . Y. Chai and W. B. Krantz, Formation and characterization of polyamide membranes via interfacial polymerization, J. Membr. Sci., 93:175-192 (1994). 53. H. S. Tan, T. H. Ng, and H.K. Mahabadi, Interfacial polymerization encapsulation of a viscous pigment mix-emulsification conditions and particle-size distribution, J. Microencaps., 8525-536 (1991). 54. B.L.Duel1 and D. B. Pendergrass, Microencapsulation by interfacial polymerization ofpolyfunctionalAziridines,AbstractPapersAm.Chem.Soc., 208(Pt. 2):336-P0ly (1994).

Microcapsules Polymerization by and 55.

56. 57. 58.

59. 60.

61.

62. 63.

64.

65. 66.

67. 68. 69. 70. 71.

COmpleXatiOn

373

A. T. Florence, T. L. Whateley, and D. A. Wood, Potentially biodegradable poly(alky1-2-cyanoacrylate) membranes, Pharm. J. microcapsules with Pharmacol., 31:422-424 (1979). P. Couvreur, Poly(cyanoacry1ates) as colloidal drug camers, CRC Crit. Rev. Ther. DrugCamer Sys., 51-20 (1988). P. Couvreur and C. Vauthier, Poly(alkylcyanoacry1ate) nanoparticles as drug carriers: Present state and perspectives, J. Controll. Rel., 17:187-198 (1991). J. Kreuter,C. G. Wilson,J. R. Fry, et al., Toxicityandassociationof poly(cyanoacry1ate) nanoparticles with hepatocytes, J. Microencaps. 1:253257 (1984). of F. Leonard, R.K. Kulkami, G . Brandes, et al., Synthesis and Degradation Poly(alky1 cyanoacrylates) J. Appl. PolymerSci., 10:259-272 (1966). S. D. Troster and J. Kreuter, Influenceof the surface properties of low contact angle surfactants on the body distributionof C-l4 poly(methy1 methacylate) nanoparticles, J. Microencaps. 9,19-28 (1992). I. Fichtner,Influenceofpoly(buty1P. Beck,J.Kreuter,R.Reszka,and on the efficacy and toxicity of the cyanoacrylate) nanoparticles and liposomes anticancerdrugmitoxantroneinmurinetumourmodels,J.Microencaps., 1O:lOl-114 (1993). T. Harmia, P. Speiser, and J. Kreuter, A solid colloidal drug delivery system for the eye: Encapsulation of pilocarpine in nanoparticles, J. Microencaps., 3~3-12(1986). T. Harmia-Pulkkinen, A. lbomi, and E. Kristoffersson, Manufactureof poly(alkylcyanoacrylate) nanoparticles with pilocarpine andbytimolol micelle polymerisation: Factors influencing particle formation, J. Microencaps., 6237-93 (1989). M. J. Alonso, A. Sanchez, D. Torres, et al., Joint effects of monomer and stabiliser concentrations on physico-chemical characteristics of poly(butyl-2cyanoacrylate) nanoparticles, J. Microencaps., 7517-526 (1990). S. J. Douglas, S. S. Davis, and L. Illum. Nanoparticles in drug delivery, CRC Crit. Rev. Ther. Drug Carrier Syst., 3,233-261 (1987). et al.,Evaluationofpoly(alky1G. Puglisi, G . Giammona,M.Fresta, cyanoacrylate) nanoparticles as a potential drug camer: Preparation, morphologicalcharacterisationandloadingcapacity,J.Microencaps.,10:353-366 (1993). K. Shankland and T. L. Whateley, Particle size distribution of polycyanaoacrylate nanoparticles, J. Pharm. Pharmacol., 41:133P (1989). K. Shanklandand T. L.Whateley,Determinationofrefractiveindicesof polycyanoacrylate particles, J. Coll. Interface Sci., 154:160-165 (1992). F. LimandA.M. Sun, Microencapsulated islets as bioartificial endocrine pancreas, Science, 210:908 (1980). A. M. Sun, I. Vacek, andI. Tai, Microencapsulation of living cells and tissues, in Microcapsules and Nanoparticles in Medicine and Pharmacy (M. Donbrow, ed.), 1992, pp. 315-322, CRC Press, Boca Raton, Florida. H.Uludag,L.Kharlip,and M. V. Sefton,Proteindelivery bymicroencapsulated cells, Adv. Drug Del. Rev., 10:115-130 (1993).

Whateley

374 72.

B.Dupuy,A.Arien,andA. P. MinnotFT-IRofmembranesmadewith alginate/polylysine complexes-variations with the mannuronic or guluronic 22,71content of the polysaccharides, Artif. Blood Subst. Immobil. Biotech. 82 (1994).

73.

K. K. Kwok, M.J. Groves, and D.J. Burgess, Production of5-15 micrometre diameter alginate-polylysine microcapsules by an air-atomization technique, Pharm. Res., 8:341-344 (1991). 74. J. Stange, E. Brugmann, D. Falfenhurst,et al., A new technique for encapsulation of liver microsomes, in Microencapsulation of Drugs(T. L. Whateley, ed.), Harwood Academic Press, Reading, UK,1992, pp. 257-262. 75. L. S. C. Wan, P. W. S. Heng, and L.W. Chan, Drug encapsulation in alginate microspheres by emulsification,J. Microencaps., 9:309-316 (1992). et al., An artificial endocrine 76. A.M. Sun, W. Parisius,H.G.Macmorine, pancreascontainingculturedislets of Langerhans,Artif.Organs, 4:275 (1980). 77.

M. Y. Fan, Z. P. Lum, X. W. Fu, et al., Reversal of diabetes in BB rats by transplantationof encapsulated pancreatic islets, Diabetes,39519 (1990). 78. Z.Cai, Z.Shi, M. Sherman, andA.M. Sun, Development and evaluation of asystemofmicroencapsulationofprimaryrathepatocytes,Hepatology, 10:855 (1989). 79. V. Coromili andT. M. S. Chang, Polydisperse dextran as a diffusing test solute

to study the membrane permeability of alginate polylysine microcapsules, Biomat. Artif. Cells Immobil. Biotech.21,427-444 (1993). 80. G . M. R. Vandenbossche, P. Vanoostveldt and J. P. Remon, A fluoroescence method for the determination of the molecular weight cut-off of alginate polylysine microcapsules, J. Pharmac. Pharmacal. 43,275-277 (1991). 81. G. M. R. Vandenbossche, P. Vanoostveldt, J. Demeester and J. P. Remon, The molecular weight cut-off of microcapsules is determined by the reaction 42, 381-386 betweenalginateandpolylysine,Biotechnol.andBioeng. (1993). 82. S. M. Shimi, E. L. Newman, D. Hopwood, and A. Cushieri, Semi-permeable microcapsulesforcellculture:Ultra-structuralcharacterisation, J. Microencaps., 8:307-316 (1991). 83. A. S. Sawhney, C. P. Pathak,and J. A.Hubbell,Interfacialphotopoly-

merization of poly(ethy1ene glycol)-based hydrogels upon alginate poly(L lysine) microcapsules for enhanced biocompatibility, Biomaterials, 14:10081016 (1993).

H. A. Clayton, N. J. M. London, P. S. Colloby, et al., The effect of capsule J. Micomposition on the biocompatibility of alginate/poly(l-lysine) capsules, croencaps., 8:221-233 (1991). 85. M. E. Pueyo, S. Darquy, F. Capron and G. Reach. In-vitro activation of humanmacrophagesbyalginatepolylysinemicrocapsules, J. Biomat.Sci. Palp. Ed., 5,197-203 (1993). 86. G. M. R. Vandenbossche, M. E. Bracke, C. A. Cuvelier, H. E. Bortier, M. M. Mareel, andJ. P. Remon, Host reaction against empty alginate polylysine 84.


and Complexation

375

microcapsules-influence of preparation procedure, J. Pharmac. Pharmacol.,

45,115-120 (1993).

87. G. M. R. Vandenbossche, M. E. Bracke, C. A. Cuvelier, H. E. Bortier, M. M. Mareel and J.P. Remon, Host reaction against alginate polylysine microcapsules containing living cells, J. Pharmac. Pharmacol., 45, 121-125 (1993).


14 Lipospheres for Controlled Delivery of Substances Abraham J. Domb* and Lev Bergelson

TheHebrew Universityof Jerusalem, Jerusalem, Israel Shimon Amselem Phurmos Ltd., Weizmann IndustrialPark, Rehovot, Israel

Introduction I.

378

11. Preparation of Lipospheres

379

111. Physical Characterization of Lipospheres A. Morphology B. Structure C. Particle Size D. Viscosity and Drug Loading E. DrugDistribution F. InVitroDrugRelease

380 380 382 384 385 386 386

IV. Applications of Lipospheres A. Parenteral Delivery of Local Anesthetics B. Parenteral Delivery of Antibiotics C. Parenteral Delivery of Vaccines and Adjuvants D. Topical Delivery of Insect Repellent E. Nanolipospheres for Cell Targeting of Anticancer Drugs

388 389 394 399 403

V.

Summary

406 407

*Affiliated with the David Bloom Center for Pharmacy, Jerusalem, Israel.

377

378 1.

Domb et al.

INTRODUCTION

Many dispersion systems are currently in use as carriers of biologically active compounds. Dispersion systems used for pharmaceutical and cosmetic formulations can be categorized as either suspensions, emulsions, and dispersions. Suspensions are defined as solid particles ranging in size from a few nanometers up to hundreds of micrometers: dispersed in a liquid medium using suspending agents. Solid particles include microspheres, microcapsules, and nanospheres [l]. Emulsions can be defined as dispersions of one liquid in another, stabilized by an interfacial film of emulsifiers such as surfactants and phospholipids. Despite their long history, emulsions are used less often today than many other dosage forms owing to their inherent instability. Emulsion formulations include water-in-oil and oil-in-water dispersions, multiple water-in-oil-in-water emulsions, and microemulsions [2]. Lipid dispersions such as liposomes are defined as phospholipidvesicles, and they are obtained by dispersing phospholipids in aqueous medium. Liposomes may be unilamellar or multilamellar, depending on the number of lamellae or bilayers, separated each from the other by a water domain. Multilayer liposomes consist of a highly ordered assembly of concentric phospholipid membranes or bilayers entrapping internal aqueous compartments [3]. The rigidity and permeability of phospholipid bilayers can be adjusted by including other water-insoluble components such as sterols and amphiphilesin addition to thephospholipid matrix. Lipospheres represent a new type of fat-based encapsulation system developed forparenteral and topical drug delivery of bioactive compounds [4-lo]. Lipospheres consist of water-dispersible solid microparticles of particle size between 0.2 to 100 pm in diameter composed of a solid hydrophobic fatcore (triglycerides) stabilized by one monolayerof phospholipid molecules embedded in their surface. The internal core contains the bioactive compound dissolved or dispersed in the solid-fat matrix. The liposphere system has beenused for the controlled delivery of various types of drugs, including anti-inflammatory compounds, local anesthetics, antibiotics, and anticancer agents. They have alsobeen used successfullyas carriers of vaccines and adjuvants [7-lo]. Similar systems based on solid fats and phospholipids have beendescribed recently [U-131. Lipospheres have several advantages over other delivery systems, including emulsions, liposomes, and microspheres, forexample, better physical stability, low costof ingredients, ease of preparation and scale-up, high dispersability in an aqueous medium, high entrapment of hydrophobic

Lipospheres for Controlled Deliveryof Substances

379

drugs, controlled particle size, and extended release of entrapped drug after a single injection from a few hours to several days. This chapter describes the preparation and physicochemical properties of lipospheres and their use for the parenteral delivery of various drugs and vaccines and the topical administration of active agents. 11.

PREPARATION OF LIPOSPHERES

In contrast to certain oil emulsions, the liposphere approach utilizes naturally occumng biodegradable lipid constituents. The internal hydrophobic core of lipospheres is composed of fats, mainly solid triglycerides, whereas the surface activity of liposphere particles is provided by the surrounding lecithin layer composed of phospholipid molecules. The neutral fats used in the preparation of the hydrophobic core of the liposphere formulations described here include tricaprin, trilaurin, and tristearin, stearic acid, ethyl stearate, and hydrogenatedvegetable oil. The polymers used in the preparation of polymeric biodegradablelipospheres were low molecular weight poly(1actic acid) and poly(capro1actone). The phospholipids used to form the surroundinglayer of lipospheres were pure-egg phosphatidylcholine, soybean phosphatidylcholine, dimyristoyl phosphatidylglycerol, and phosphatidylethanolamine. Food-grade lecithin (96% acetone insoluble) was used inthe preparationof lipospheres for topioal and veterinary applications. Liposphere formulations are prepared by a solvent or melt process. In themelt method, theactive agent is dissolved or dispersed in the melted solid camer; that is, tristearin or polycaprolactone and a hot buffer solution is added at once along with the phospholipid powder. The hot mixture is homogenized for about 2-5 min using a homogenizer or ultrasound probe, after which a uniform emulsionis obtained. The milky formulation is then rapidly cooled down to about 20°C by immersing the formulation flask inan acetone-dry ice bath while homogenizationis continued to yield a uniform dispersion of lipospheres. Alternatively, lipospheres might be prepared by a solvent technique. In this case, the active agent, the solid camer, and phospholipid are dissolved in an organic solvent such as acetone, ethyl acetate, ethanol, or dichloromethane. The solvent is then evaporated and.tberesulting solid is mixed with warm buffer solution and mixing is continued until a homogeneous dispersion of lipospheres is obtained. In a typical preparation, dexamethasone(200mg), tristearin (400mg), and propylparaben (5 mg) are added to a 50-mL round-bottom glass flask. The flask isheated at75°C to melt the tristearin-dexamethasone mixture and

Domb et al.

380

hot 0.1 M phosphate buffer solution pH 7.4 (75"C, 9.3 g) isadded along with egg phosphatidylcholine (200 mg). The mixture is homogenized for 2 min until a uniform milky formulation is obtained. The hot formulation is rapidly cooled to below 20°Cby immersing the flask in adry ice-acetone bath with continued mixing to yield a white, thin dispersion.If needed, the pHof the formulation is adjusted to 7.4 with a 1N HC1 solution. The formulation may contain antioxidantssuch as tocopherol and preservatives such as parabens. Submicron-size lipospheres are prepared by passing (four times) the liposphere formulation by extrusion through a submicron series of filters at a temperature 5°C above the melting point of the liposphere core composition. Particlesize maybe reduced to about200 nm. Polymeric biodegradable lipospheres can also be prepared by a solvent or melt process. The difference between polymeric lipospheres and the standard liposphere formulations is the composition of the internal core of the particles. Standard lipospheresas those previously described consist of a solid hydrophobic fat core composed of neutral fats like tristearin, whereas in the polymeric lipospheres, biodegradable polymers such as polylactide or polycaprolactone substituted the triglycerides. Both typesof lipospheres are thought to be stabilized by one layer of phospholipid mole111). cules embedded in their surface (see Section Sterile liposphere formulations are prepared by sterile filtration of the dispersion in the hot stage during preparation through a filter at 0.2 pm a temperature5°C above themelting point of the liposphere corecomposition. Heat sterilization using a standard autoclave cycle decomposed the formulation. y-Sterilization of liposphere formulationsdid not affect their physical properties. Lipospheres formulations of 1 : 4 : 2 and 2 : 4 : 2 bupivacaine-tristearin-phospholipid(w/w % ratio) were irradiated with a samples were analyzed forparticlesize, dose of 2.33 Mrad,andthe bupivacaine content, in vitro release characteristics, and in vivo activity. The irradiated formulations had a similar particle size, bupivacaine content, release rate, and anesthetic effectiveness in the rat paw analgesia model to bupivacaine HCl solution (Marcaine). However, a more careful analysis of the formulation ingredients should be performed, since phospholipids may degrade during irradiation [14]. 111.

PHYSICAL CHARACTERIZATION OF LIPOSPHERES

A. Morphology Fig. 1 shows a light microscopic picture of a typical preparation of lipospheres composed of tristearin and lecithin consisting of particles having a uniform spherical shape. Microscopic examination of a typical liposphere

Lipospheres for Controlled Delivery

of Substances

38 1

Fig. 1 Light microscope pictureof a typical preparationof lipospheres composed of tristearin and lecithin with average particlesize of 10 pm.

Fig. 2 Transmission electron microscopy (TEM) of a single nanoliposphere (200 nm) .

formulation using transmission electron microscopy (TEM) showed spherical particles (Fig. 2). B. Structure

The phospholipid content on the surfaceof lipospheres was determined by 31Pnuclearmagneticresonance (NMR) before andaftermanogenase

Lipospheres Controlled forSubstances Delivery of

383

(Mn”) or proseodimium(Prt3) ion complexation [l51 and by trinitrobenzenesulfonic acid (TNBS) labeling using liposphere formulations containing phosphatidylethanolamine (PE) [16]. In both methods, the agents (Mn+2, Prt3, or TNBS) interact specifically with the exposed phosphateor amino groups of the phospholipid. Determination of the surface phospholipid using the TNBS methodshowed that 70-90% of the phospholipid polar heads are on the surface of the liposphere particles prepared from triglyceride-phospholipid at a 1 : 0.5 to 1 : 0.25 w/w ratio. Increasing the phospholipid content decreases the percentageof surface phospholipid polar heads, which indicates the formation of other phospholipid structures such as liposomesin the formulation. Similar results were obtained by 31P NMR analysis using Mn+’and Prf3ions for interaction with the phosphate polar heads [15]. This is illustrated in Fig. 3, which shows”P NMR spectra of drug-free liposphere vesicles obtained in the absence (A) and presence (B) of Mn+’. The decrease in the peak areaafter the addition of 5 mM Mn+’ represent therelative amount of phospholipid in the outer monolayer avail-

-200

-1 0 0

0

100

200

PPM

Fig. 3 31PNMR spectra of lipospheres composed of tricaprin-phospholipid in a weight ratio of 2 : 1 in the absence (A) and presence (B) of Mnt2. 0 ppm corresponds with the resonance position of diacylphosphatidylcholine undergoing isotropic motion.

384

Domb et al.

able for interaction with Mn+2. For lipospheres composed of tricaprinphospholipid at a weight ratio of 2 : 1 and 5 : 1, 75 and 90% of the phospholipid polar heads wereaccessible to theparamagnetic ions, respectively. For comparison, liposomes of the same composition and size but without tricaprin had only 40% of the phospholipid polar heads on the surface [151. These datasuggest that the proposed structure of a liposphere is that of spherical particle with a monolayer of phospholipid molecules surrounding theinternal solid fatcore,wherethehydrophobic chains of the phospholipids are embedded in the internal triglyceride core containing the active agent, asillustrated in Fig. 4.

C. Particle Size Analysis of the particle size distribution of lipospheres was performed using a LS 100 Coulter Counter Particle Size Analyzer (Coulter Corp., Hialeah, FL). This instrument can measure particles from 0.4 to 800 pm by particle size-dependent light diffraction patterns. The particle size of submicron lipospheres was estimated from the transmission electron microscope pictures and using a particle size analyzer for 0.01- to 3-pm particle size diameter range. Blank and drug-loaded lipospheres prepared by the melt methodwithout farther treatment had a unimodal shape with an average particle size

Fig. 4 Schematic illustrationof proposed structureof a liposphere particle.


385

between 5 and 15 pm with less than 2% of particles greater than 100 pm. This particle size formulation is useful for subcutaneous or intramuscular injection or fortopical applications. In an effort to reduce the particle size, lipospheres of 16.9 pm average diameter preparedby the melt method were passed through a laboratory Microfluidizer (Microfluidics, MA, USA). Thereby the average size was reduced to 10.8 pm after8 passes and to 9.4 pm after 40 cycles. Similar results were obtained when using a homogenizer. Lipospheres containing the malaria R32NS1 antigen and differingin their fat composition were prepared by the melt procedure [&lo]. Three groups of neutralfats were used: (1) solid fats such as tristearin and stearic acid with melting points in the rangeof 6570°C; (2) semisolid fats such as tricaprin and ethylstearate with melting points in the range of 30-35°C; and (3) liquid fats such as olive oil and corn oil. ?kro populations of liposphere particles usually coexisted, one in the size range of 1-10 pm in diameter (population A), and a second population with a diameter between10 and 100 pm (population B). No correlation was found between fat physical state (solid, semisolid, liquid) and particle size distribution of lipospheres formed. Lipospheresmade of tristearin were the most homogeneous formulations, with 100%of the particleshaving an average diameterof about 7 2 3 Pm 181. Biodegradable polymeric lipospheres made of polylactide and lecithin showed a very broad particle size distribution from 2 to 100 pm in diameter, with a mean average size of 30.6 & 25.9 pm and median (% of particles >50pm) equal to 21.6 [lo]. Polycaprolactone lipospheres showed a similar range of particle size distribution, but with a 1.5-fold mean average size (45.6 f 29.5 pm) and a doublemedian value (43.6) compared with polylactide lipospheres. Inclusion of lipid A in the composition of the polymeric lipospheres reduced their mean average particlesize by a factor of 0.25 regardless of the polymer type [lo]. All the liposphere formulations with a polymeric core prepared remained stable. during the3-month period of the study, and no phase separation or appearanceof aggregates were observed. D. Viscosity and Drug Loading

For drugs such as oxytetracycline [17], itraconazole, and dexamethasone, up to20% drug loading into lipospheres was obtained and the formulations were fluid enough to beinjected.Forotheragentslikebupivacaine, lidocaine, and chloramphenicol, drug loading above 10% produced aviscous lotion. Theviscosity of the liposphere formulation is dependent on the drug properties, the ionic strength andofpH the continuous aqueous solu-

386

Domb et a/.

tion, and the ratio and amount of phospholipids and triglycerides. An increase in the contentof the insoluble ingredients (drug and lipids) and in the salt concentration of the aqueousmedium increases the viscosity of the formulation. For topical applications, lotion and paste consistencies are desired, which are achieved by adding NaCl to thewater phase and increasing the fatand phospholipid content of the liposphere lipid phase [M].

E. DrugDistribution The drugdistribution in a liposphere formulation was determined by isolating the particles by centrifugation and determiningthe drug contentin the isolated cake after extraction of the free drug with, for example, acidic buffer solution for basic drugs. The amount of unencapsulated free drug can be estimated by microscopic analysis(the free drug appearsusually as crystals, whereas lipospheres appear as round shapes). A study was conducted to de". Ane bupivacaine distribution in the 2% bupivacaine formulation [19]. It was found that the nonencapsulated ,drug was observed as needle-shaped crystals dispersed among the liposphere particles. These lipospheres anddrug crystals were isolated by centrifugation, thebupivacaine crystals were extracted from the mixturewith acidic buffer solution, and the drug-loaded lipospheres were analyzed for bupivacaine contept after dissolution-extraction of the lipospheres in Triton solution. The drug loading was in the rangeof 70-85 weight% in the core and about20% of the drug was extracted by the acidic solution appearingasnonincorporated drug. About 4% of the drug was soluble in the aqueous solution, which is the solubility of bupivacaine in the buffer solution. In an attempt to determine the form of the unincorporated drug, bupivacaine and phospholipid weredispersed in aqueous mediumin the absence of the tristearin component. A uniform and stable submicrondispersion was obtained. Microscopic examination of this fat-free preparation showed that the dispersed drug microparticles are nonspherical but in the form of long needles. It should be noted thatbupivacaine free base is not dispersible in buffer solution without asurfactant such as phospholipids. Thus, it is apparent that the unincorporatedbupivacaine in the tristearin liposphere formulation is in a form of dispersible microparticles composed of the solid drug and phospholipids. Taxol, verapamil, and piridoxamine wereincorporated in submicronsize lipospheres up to 90% encapsulation yield. F. InVitro Drug Release

In vitro release studies were conductedusing a large-pore dialysis tubing of 300,000 molecular weight cut-off (MWCO) to minimize the effect of the

Lipospheres Controlled forDelivery

O+ 0

of Substances

387

I

I

1

I

12

24

36

48

Time (hours)

Fig. 5 Release of bupivacaine from 2 and 5% loaded liposphere formulation.

tubing on the release rate from the formulation (a regular pore size tubing of 12,000 MWCO did affectthe drug release rate). Thecontrol solutions of drugs such as bupivacaine (Marcaine-0.75%in solution for injection) were released through the large-pore tubing within a few hours. Drug releases from 2 and 5% bupivacaine-loaded liposphere formulations are shown in Fig. 5. Both formulations released the drug for 48 h following a first-order kinetics (R2=0.97).These release profiles are expected for formulations that contain 4% of the free drugsoluble in the aqueousvehicle which are released immediately throughthe dialysis tubing. The release of etoposide, a water-insoluble anticancer agent, from 2% loaded lipospheres placedin a dialysis tubing (300,000 MWCO) is shown in Fig. 6 . Over 90% of the drug was constantly released during a period of 80 h. Lipospheres containing I4C-diazepamwere preparedaccording to the melt technique. The lipospheres showed a very uniform particle size distribution with an average particle size of 8 pm. The lipospheres were evaluated in vitro by placing 0.5 mL of the formulation in a dialysis tubing (300,000 MWCO). The dialysis tubing was placed in a 0.1" phosphate

Domb et al.

388 6000

5000

4000

3000

2000

1000

Time (hours)

Fig. 6 Release of etoposidefromalipospheredispersion.Totalamount etoposide in the test specimen was 6.2 mg.

of

buffer, and the cumulative release of diazepam was determined by measuring the radioactivity in the release medium. The release medium was changed periodically to provide sink conditions. Sustained release of diazepam was obtained over a period of 3 days. The in vitro release was also determined by mixing a sample of the formulation in excessbuffer solution, specimens of the mixture were taken every few hours, and the drug released into the solution was determined after removalof the lipospheres by ultracentrifugation. The release rate by this method was faster than from dialysistubing. IV. APPLICATIONS OF LIPOSPHERES This section describes the use of lipospheres for parenteral administration of long-acting local anesthetics and oxytetracycline and fortopical application of N,N-diethyl-m-toluamide(DEET) insect repellent.

Lipospheres for Controlled Delivery of Substances A.

389

Parenteral Delivery of Local Anesthetics

Local anesthetics are preferred over general anesthetics because of the serious complications that can occur during general anesthesia. However, even local anesthetics, which are usually injected as an aqueous solution, are eventually absorbed from the site of application into the circulation. Frequent administration of local anesthetics may result in the development of systemic toxicity. A long-acting formulation that provides extended regional blockade may be useful for pain management following surgery or for chronic pain relief. Lipospheres wereused for the delivery of the common local anesthetics bupivacaine and lidocaine for the purpose of extending their effectiveness to a few daysafter a single injection. Liposphere formulations containing local anesthetics were prepared by the melt and solvent methods as described above. Sterile formulations wereprepared bydissolving bupivacaine (100 g) and egg phospholipid (100 g)in ethanol followed by sterile filtration through a 0.22-pm filter.Tristearin (200 g) was dissolved in hot ethanol andfiltered into the same flask and the ethanolwas evaporated to dryness. To the remaining semisolid, 5 L of hot (65°C) sterile 0.1" phosphate buffer solution containing 0.05% methyl paraben and 0.1% propyl paraben as preservatives was added and the mixture was homogenized for 5 min at high speed. The uniform milky preparation was rapidly cooled down to below 20°C by immersing the flask in a dry ice-acetone bath while homogenization continued. The formulation was filled into 10mL vials and stored underaseptic conditions at 4°C until used. A variety of models have been used for the evaluation of peripheral analgesic agents. However, only three have been extensively used to evaluate pharmaceutical compositions. There are the Randall-Selitto test, the abdominal construction writhing response to intraperitonealinjection of an irritant, and thepain response of mice after formalin injection [20,21]. The local effects as a function of time of liposphere formulations were tested using the Randall-Selitto experimental animal model. To induce hyperalgesia, animals were first anesthetized and then a yeast or carrageenan solution was injected through the foot pad nearest the first digit. The foot withdrawal score was then measured at the indicated times after the injection, on a scale from 0-25 (0 = could not stand any pressure; 25 = could stand extensive pressure) using the RandallSelitto instrument. Liposphere formulations to be tested were coadministered with the yeast solution. Control animals received an identical volume of water. For animals being tested for more than 24 h, the liposphere formulation was injected at time 0 and yeast was injected 24 h prior to

390

Domb et al.

liposphere injection, since yeast-induced hyperalgesia is often not measurable at times greater than 24 h postadministration. The hyperalgesic response was developed within the first hour after injection and maintained for 48 or 72 h. The effectiveness of several bupivacaine formulations was investigated. Liposphere formulations containing tristearin as the fat and phosphatidylcholine from eitheregg (PCE) or soybeans (PCS) and loaded with 1 or 2% bupivacaine produced long-lasting analgesia for at least 48 h and for almost 72 h [22]. The blank liposphere formulation did not produce any analgesia, and the Marcainecontrol solution (0.75% in saline) produced a strong analgesic effect which lasted for less than 3 h. A rat sciatic nerve preparation was developed to study prolonged local anesthetic blockade from bupivacaine lipospheres [23,24]. Blockade andor section of the rat sciatic nerve is a time-honored system to study regional anesthesia. The nerve is large and easily seen, and the effects of motor and sympathetic blockade in the foot can be detected. Sensory blockadecan also bemeasuredprovidedthere is no spillover in the saphenousnerve distribution. After anesthesia was induced, bilateral posterolateral incisions were made in the upper thighs, and the sciatic nerves were visualized. Sham vehicle wasinjected around the nerve onone side, and vehicle containing 5 and 10% bupivacaine (0.5-mL dose) was injected on the other side. The facia wasthen closed over the deepcompartment to restrict partially egress of drug formulation. Motor blockade was scored on a 4-point scale based on visual observation: 1, normal appearance; 2, impaired ability to splay toes when elevated by the tail;3, toes and foot remained plantar flexedwith no splaying ability; and 4, loss of dorsiflexion, flexion of toes, and impairmentof gait. Both 5 and 10% bupivacaine liposphere formulations showed significant levels of motor blockade throughday 3, and in some cases, day 4(Fig. 7). Motor function in all animals returned to normalby day 6. Sympathetic blockade was determined indirectly by foot pad temperature measurements. The foot receives sympathetic innervation largely from the sciatic nerve, although there may be a contribution from the saphenous nerve as well. Skin temperature measurementsof the blockadeside were also monitored as an indication of vascular tone. During the period when motor block was apparent (Fig. 7), the blockaded side was essentially always warmerthan the unblockaded side (Fig. 8). Temperature differences seemed to dissipate within 1 day after motor block resolved. Since sensory blockade measurementsin this experiment are complicated by several factors, it was important to find a method of detecting sensory blockade that is relatively independent of motor responses. Vocalizations in response to defined transcutaneous electrical stimulation of points on the feet were


of Substances

39 l

1

4.0

-

3.0

-

2.0

-

1.0

-

0

p 5%

1

2

3

4

5

(n=6)

6

TIME (days after injection) Fig. 7 Motor blockade of 5% (n = 6) and 10% (n = 3) liposphere formulations injected near the sciatic nerve of rabbits. The motor rating were:4, clubbed foot; 3, partial clubbing (does not splay); 2, obvious difference in splaying; and 1, normal.

used as a criterion for the measurementof sensory blockade. The threshold to hind paw electric shock and hind paw pad temperature measures of sympathetic block where both increased for 3-4 days. No impairments were observed on the contralateral control side. One week after liposphere administration, sciatic nerves were removed and histologically evaluated. No evidence of nerve damage andvery little inflammation of foreign body response wereobserved. . A study was conducted to evaluate the efficacy of 2% bupivacaine liposphere formulation to produce analgesia inthe ratformalin model [24]. The model was designed to assess the antinociceptive ability against chronic pain caused by a test compound. An additional foot flick thermal method was also performed utilizing the sameanimals. The formalin study

Domb et al.

392

Time (days)

Fig. 8 Vocalizationthreshold to hindpawelectricalstimulationafterbilateral sciatic nerve administration. compared the effects of administration of 2% loaded bupivacaine lipospheres with blank lipospheres, standard bupivacaine solution (Marcaine, 0.5% i i t h 1 : 200,000 epinephrine), and physiological saline on antinociception in the rat. Rats were pretreated at various times with test or control formulations by infiltration injection into the right popliteal fossa and theninjected with 5% formalin into thedorsal surface of the right hind paw. Nociception wasthen measuredin the form of paw flinches ina 5-min period during both the acute and tonic phases. Both 2% bupivacaine liposphere and Marcaine had an onset of action times within 10 minutes. However, the liposphere formulation was able to maintain significant antinociception for at least 9 h in both acute and tonic phases as compared with less than 3 h for Marcaine. Similar results were obtained in the foot flick thermal stimulus model. Sensory blockade was measured by the time


of Substances

393

required for each rat to withdraw its hind paw from a 56°C plate. Latency to withdraw eachhind paw from the hot plate was recorded by alternating the hot plate within 15 S, the trialwas paws. If no withdrawal occurred from terminated. Neither the Marcaine nor liposphere formulation caused a change in pawvolume, nordid they significantly influencethe development of formalin-induced edema. The long-acting effect of the bupivacaine liposphere formulation as compared with Marcaine solution was further confirmed bythe rattail flick model [25]. The study compared administration of the 2% bupivacaine liposphere, blank liposphere, standard Marcaine, andsaline. After administration of the formulations, tail flick latencies weredetermined.The liposphere formulation exceeded the anesthetic duration of Marcaine by 12-fold. Treatment with blank lipospheres was not significantly different from treatmentwith the saline control. A pilot pharmacokinetic study was performed in rabbits to compare the Cmax andTmax values obtainedafter intramuscular injection of Marcaine HC1 solution and a liposphere formulation (Fig. 9). A total amount equivalent to of 20 mg bupivacaine was injected to rabbits (n = 3) and blood was collected for 72 h. Bupivacaine bloodconcentrations were determined by high pressure liquid chromatography (HPLC) following a United States Pharmacopeia (USP) method. Lidocainewas used as internal standard with a linear calibration curve for bupivacaine between 50 and 1000 ng/mL.Bupivacaine blood levels for bupivacaine solution (Marcaine) and lipospheres after a single injection are given in Fig. 9. The maximal blood concentrations were 681 f 246 and 200 & 55 ng/mL for bupivacaine in solution and in lipospheres, respectively. The toxicity of bupivacaine lipospheres in rats was evaluated. Since the liposphere formulation consists of natural inert components, phospholipids, and triglycerides, they are expected to be biocompatible in vivo [26]. The incidence of microscopic observations after intramuscular injection of liposphere formulations was studied (Table 1). Blank lipospheres, 1% and 5% bupivacaine in lipospheres, 5% dextrose solution, and bupivacaine solution (0.1 mL) were injected in rats followed by histological examination of the site of injection at days 3, 7, and 14. The degree of inflammation, necrosis, and fibrosis was scaledfrom 0 to 4, where 0 means absent and 4 means marked. The degree of inflammation, necrosis, and fibrosiswas similar for all formulations. At day 3, some irritation and inflammation was observed, which was reduced after 14 days. In a second study, the local toxicity of a 2% bupivacaine formulation after daily injections for 2 weeks in dogs was estimated. Minimal local irritation was observed in these studies as determined by histological examination.

=

394

Domb et al.

"i ' 0 °

Y ~

I _

"

Solution

Liposphere

400

300

200

100

0

0

12

24

36

48

60

72

Time (hours)

Fig. 9 Bupivacainebloodlevelsafterintramuscularinjection of 2% loaded lipospheres as compared with bupivacaine HCl solution.Drug levels in blood were determined by HPLC. B. Parenteral Delivery of Antibiotics

Several antibiotics, including ofloxacin, norfloxacin, chloramphenicol palmitate, and oxytetracycline, and antigungal agents, such as nystatin and amphotericin B, have beenincorporated into lipospheres in highencapsulation yield. The use of lipospheres for antibiotic delivery was demonstrated by the development of liposphere oxytetracycline formulations for veterinary use [17]. Parenteral oxytetracycline (OTC) therapy in farm animals requires daily administration of the drug over several days, usually 3-5 days, in order to provide prolongedtherapeutic blood levels. Serum OTC concentrations of potential clinical and therapeutic values in the treatmentof OTC-sensitive organisms are estimated in the rangeof 0.15-1.5 mg/mL. The minimum inhibitory concentration (MIC) in milligramsper milliliter for certain patho-


395

Table 1. Incidence of Microscopic Observations After Livosvhere Administration

Formulation Blank lipospheres(0%) inflammation necrosis fibrosis Bupivacaine lipospheres(1%; inflammation necrosis fibrosis Bupivacaine lipospheres(5%) inflammation necrosis fibrosis Dextrose solution inflammation necrosis fibrosis Bupivacaine HC1 solution (0.75%) inflammation necrosis fibrosis

1Day

3 Days

14 Days

112 222 000

221 010 110

212 000 221

111 222 000

212 313 101

211 OOO 110

112 333 000

222 212 111

211 000 200

111 212 OOO

211 011 111

101 000 101

111 223 OOO

222 211 111

OOO

111

111

Rats (groups of three) were injected subcutaneously with either formulation and the injection site was histogically examined. The results for eachare ratgiven in the table.Liposphereformulationswerecomposed of bupivacaine-tristearin-phospholipid at a1:2:1 weight ratio. Dextrose used was 5% a solution, and bupivacaine HCI solution was commercial Marcaine solution(0.75% bupivacaine). The degree of inflammation, necrosis, and fibrosis was scored from 0 to 4, where 0 means absent and4 means marked. gens of farm animals are 0.15 for Pasteurella multocida, 0.3 for Staphylococcus and Pasteurella anatipestifer, 0.4 for Haemophyllis paragallinarum, 0.8 for Mycoplasma gallisepticum, and 1.5 for Escherichia coli. Blood levels of above 0.5 mg/mL are required for treatment of most bacterial infections. Several long-acting oxytetracycline formulations have been reported [27-291. These formulationshave been tested in various farm animals and showed adequate blood levels for 72 h after asingle injection at adose of 20 mg/kg* Oxytetracycline was encapsulated in good yields in solid triglyceride liposphere formulations. The microdispersion containing up to 15 wt% oxytetracycline was injectable through a 20-G needle and was stable for at least 1year when stored under refrigerated conditions.OTC was released

396

Domb et al.

in vitro from dialysis tubing for 5 days, whereas OTC was released from solution through the tubing in less than 6 h. Blood levels in turkey birds or rabbits were maintainedfor up to4-5 days for the8% loaded formulation. An increase in OTC concentration required a decrease in the content of triglyceride and phospholipid in the formulation resulting in a decrease in the durationof drug release. The formulation degraded in tissue, but remnants of the formulations remained for periods longer than 4 weeks. A study was conducted to determine theeffect of liposphere composition on its physical properties, drug release rate, and injectability through a 22-G needle. The parameters investigated were triglyceride composition, phospholipid type, aqueous phase composition, drug loading, and OTCtriglyceride : phopholipid ratio. Three triglycerides were used, tristearin (mp 75"C), trilaurin (mp 42"C), and tricaprin (mp 33°C). Three types of lecithin were utilized, highly pure (99%)egg and soybeanphospholipid and foot-grade lecithin (96% acetone-insoluble material). The continuous medium was either 0.1 M phosphate buffer or 5% dextrose solution. The formulations based on tricaprin, regardless of the type of phospholipid or aqueous medium composition, loaded with up to 15% OTC were stable and injectable and provided a prolonged drugrelease rate in vitro. Formulations containing less than 5% phospholipid or more than 15%OTC were less stable and often caused cloggingof the needle, and therefore they were discarded [171. Release studies were conducted in dialysis tubing with a molecular weight cut-off of 300,000 (Spectrum, CA)or without tubing. One milliliter of liposphere formulation was introduced into the prewashed dialysis tubing and placed in a jar containing 800 mL of 0.1 M phosphate buffer pH 4.5. Alternatively, 1mL dispersion was added directly to thebuffer solution and placed on anorbital shaker at 37°C. Samples were takenat discrete times, centrifuged, and analyzed by ultraviolet (UV) spectrophotometry at 365 nm to determine the drug release rate from the formulations. A commercial OTC solution (Dabecycline, 10% OTC concentration) used as control was placed in the dialysis tubing to determine if the tubing was limitingthe rate of release. The release was slower from thedialysis tubing; however, OTC was released from the formulation for about 3-5 days from both releasing systems. Two animal studies were conductedin order toevaluate the controlledrelease effect of the liposphere formulations by following the OTC blood levels and theelimination of the administered dose fromthe injection site. In the first study, four OTC-loaded liposphere formulations differing in their compositions were compared with an OTC solution (10% OTC in acidic solution) used as reference and a blank liposphere formulation as control (Table 2). The formulations were injected intramuscularly to groups of six

m

0


24

-Ei

28%

397

Y

Domb et a/.

398

3.0 ---h-

.."0." "m'a.,

0.0". 0.0

1 .o

.

; 2.0

4 8% in TS:PL 1:l B, 8% in n P L 1:l C, 12% in TLPL 1:l D, blank E, 8% TLPL 1:.5 Dabecvcline

.

; 3.0

.-v

..

. ....... ........,.-,

Y**"""

4.0

5.0

-t! 6.0

7.0

Time (days)

Fig. 10 Oxytetracyclineblood concentrations of several liposphere formulations.

OTC blood levelswere determined. The injection sites turkey birds and the were observed forresiduals 7,11, and 28 days postinjection. The average OTC blood levels for the six formulations are given in Fig. 10. All four liposphere formulationsshow OTC levels above MIC for at least 3 days. Formulation E, composed of trilaurin (80 mg/mL) and phospholipid in a ratio of 2 : 1, showed similar OTC blood levels to the formulations based on tristearin and the more concentrated (120 mg/mL) trilaurin-based formulation. The formulation composed of trilaurin-phospholipid in a 1 : 1 ratio was less effective. These results indicate that a lower phospholipid to triglyceride ratio improves the duration of drug release and that higher drug loading (formulation C) does not affect the formulation effectiveness. Although tristearin showed good results, it is not preferred, because itis less susceptible for elimination from the injection site, as discussed below.


of Substances

399

The residual amountsevaluatedat7 and 11 days were maximal (about 90% of the original dose) for tristearin-based formulation (A), about 50% for the trilaurin-based formulations (B, C, and E), and about 20%.for theblank formulation. The deposits contained less than 10% OTC of the original dose. After 28 days, only formulation A had significant amounts of deposits at the injection site, which was mostly tristearin. No OTC was detected in any of the deposits retrieved from theanimals after 28 days. These results indicate that formulations based on trilaurin, asemisolid fat at body temperature (mp = 44°C) are more rapidly eliminated from the injection site[17]. Based on these results, second a animal study was conducted in order to identify an optimal formulation which wouldprovide maximal OTC blood levels after a single injection. The composition of formulation E (Table 2) was tested in comparison with a mixture of formulation E and DabecyclineOTC solution in a 4 : 3 volume ratio. Dabecycline-OTC solution was usedas reference. Dabecycline was added to formulation E in order toincrease the initial blood levels, which were too low for the first 5 h after injection.The three formulations were administered intramuscularly to turkey birds (10 birds foreach formulation). The OTCsolution (Dabecycline 100) resulted in very high serum levels at 3and 6 h, 19.9 and 3.5 mg/mL, respectively, and drug levels of 2.1, 0.4, and olive oil > tristearin > tricaprin > corn oil stearic acid [8]. No correlation between liposphere particle sue or fat chemicalcharacteristics and immunogenicity was found. It is worth noting that the IgG antibody ELISA titers obtained on immunizing rabbits with liposphere R32NS1 were superior to those obtained following similar immunizations with the freeantigen absorbed to alum, which showed no antibodyactivity at the sameantigen concentrations. It was previously shown that this antigen was also poorly immunogenic in humans when injected alone as an aqueous solution or when adsorbed onalum [33]. Incorporation of a negatively charged phospholipid, dimyristoyl phosphatidylglycerol (DMPG) in the liposphere lipid phase, caused asignificant increase in the antibody response to the encapsulated R32NS1 antigen [9]. Enhancement of immunogenicity by inclusion of charged lipids have also beenobserved with certain antigens inliposomes.Negatively charged liposomes produced a better immune response to diphtheria toxoid than positively charged liposomes [34]. However, when liposomes were prepared with other antigens, positively charged liposomes workedequally as well as those bearing a negative charge [34]. Further studies are needed to determine whethernegative charges in lipospheres have general abilities to enhance immunogenicity or whether, as with liposomes, charge effects are dependent onindividual antigen composition. An interesting correlation was observed betweenthe liposphere fat to phospholipid (FPL) molar ratio, particle size, and immunogenicity. Low F/PL ratios (50.75) were found to induce the formation of lipospheres of small particle size (70% less than 10 pm in diameter), and this apparently resulted in increased antibody titers [9].Among the ratios tested, a maximal level of IgG antibody production was obtained at a F/PL ratioof 0.75, whereas at larger ratios, decreased antibody production was observed. Although the reason for this phenomenon is unknown, apossible explana-

402

Domb et a/.

tion may be theoccurrence of better antigen orientation and epitope exposures in the small lipospheres because of higher surface curvature. W Opopulations of particles usually coexist in vaccine-loaded liposphere formulations, one in the size range of 1-10 pm in diameter (population A) and a second population with a diameter between 10 and 80 pm (population B). As mentioned previously, the particle size distribution of lipospheres depends on the fat phospholipid to (FPL) molar ratio, and the immune response to liposphere-encapsulated R32NS1 was also dependent on the F P L ratio. The average size of the particles increases with increasing F/PL molarratio:Under conditions where theF P L ratio is high (22.5), the large particle population is predominant (approximately 80% of the particles had an averagesize of 73 pm) [9]. In order to examine the influence of different routes of administration of lipospheres on their immunogenicity, rabbits wereimmunized orally or parenterally (by subcutaneous, intraperitoneal, intramuscular, and intravenous routes) with lipospheres made of tristearin and lecithin (1 : 1 molar ratio) and containing the malaria antigen [S]. The immune response obtained was followed for a period of 12 weeks postimmunization. No antibody activity wasfound after oral immunizationin any of the individual rabbits immunized with the liposphere R32NS1 vaccine formulation. However, rabbit immunization by all parenteral routes tested resulted in enhanced immunogenicity with increased antibody IgG levels over the entire postimmunization period. The individual rabbit immune response shows that. immunization by subcutaneous injection was the most effective vaccination route among all parenteral routes of administration tested [S]. Incorporation of lipid A, the terminal portion of gram-negative bacterial lipopolysaccharide, in lipospheres increased significantly the immune response to R32NS1 malaria antigen resulting in double IgG levels compared with R32NS1 lipospheres lacking the lipid A. The adjuvanteffect of lipid A incorporated in lipospheres was observed evenafter 1600-fold dilution of the rabbit sera [g]. The adjuvant effect of different doses of lipid A in lipospheres was also examined by immunizing rabbits with lipospheres containing R32NS1 and prepared at different final concentrations of lipid A. A gradual increase .in IgG antibody titer with increasing lipid A dose was observed. The strongest antibody activitywas obtained with lipospheres containing 150 pg of lipid Nrabbit. Athigher lipid A dose(200 kg/ rabbit), a decrease in ELISA units was observed [9]. The preparation and use of polymeric biodegradablelipospheres as a studied. The potential vehicle for thecontrolled release of vaccines was also immunogenicity of polymericlipospheres composed of polylactide (PLD) or polycaprolactone (PCL) and containing the recombinant R32NS1 malaria


of Substances

403

antigen was tested in rabbits after intramuscular injection of the formulations [101. High levels of specificIgG antibodies were observedin the seraof the immunized rabbits up to 12 weeks after primary immunizationusing a solid phase ELISA assay. PCL lipospheres containing the malaria antigen were ableto induce sustained antibody activity after one single injection in the absence of immunomodulators.PCL lipospheres showedsuperior immunogenicity comparedwith PLD lipospheres, with the difference being attributed to thedifferent biodegradation rates of the polymers. D. Topical Delivery of Insect Repellent A common mean for repelling insects consists of applying the compound N,N-diethyl-m-toluamide(DEET) to theskin. The commercially available liquid DEET formulations contain between 15 and 100%DEET, and they are not recommended foruse on children. The toxicity of DEET has been extensively reported and related to its high skin absorption after topical administration [36-381. Previous studies have demonstratedthat as much as 50% of a topical dose of DEET is systematically absorbed [37,38]. In an effort to develop a new topical formulation for DEET thatpossesses reduced skin absorption as well as an increase in the durationof repellency, we have encapsulated DEET intolipospheres and studied its skin-absorption dynamics and durationof action [39,40].We hypothesized that theencapsulation of DEET will reduce its contact surface area with the skin and reduce its evaporation rate from theskin surface, resulting in reduced dermal uptake and extended repellent activity. The liposphere microdispersions containing DEET incorporated in solid triglyceride particles were prepared by the melt method using a m mon naturalingredients in one step without the use of solvents. The formulation was preserved by parabens, propylparaben in the oil phase, and methylparaben in the aqueous phase. The average particle size was in the range of 15 pm, and microscopic examination showed spherical particles. The formulations were stable for at least 1 year when stored at 4°C and 25°C in a closed glass container, and DEET content, particle size, and viscosity remained almost constant. The residual efficacyof liposphere formulations was evaluated on volunteers by applying the formulations to the skin and exposing the subjects to mosquitoes [40]. The time of 100% repellency (zero biting) was the index for determining the effectiveness of a formulation. The formulations were applied on four locations on the armof volunteers ata 2.5 mg/cm2on a total areaof 12-cm2skin surface. Mosquitoes were placed in a screen-bottomed (%mesh netting, 10-cm2exposure area) cylindrical cup containing 50 5- to 15-day old female mosquitoesdisplaying host-

Domb et al.

404

seeking behavior with access to the skin through the netting. The forearm was placed on the mosquito netting for 10 min every 30 min and the number of biting mosquitoes (evident by a blood meal) were recorded. Prior to any efficacy experiment, the mosquitoes weretested on untreated skin to confirm their host-seeking behavior. W Omosquito species were tested: Aedes aegypti and Anopheles stephemi, both aggressive biters. The results of this experiment is given in Table3.The formulations were repellent for a minimum of 2.5,3.5, and 6.3 h for the6.5, 10, and 20% DEETcontaining lipospheres, respectively. The DEET-freeformulation (control) and the untreated groupsdid not show any activity against Aedes aegypti, and the 10% DEET solution in alcohol was repellent for about 1.5 h. The blood concentrations after intravenous (IV) dosing and topical administration of 10% DEET in ethanol (group A) and 10% DEET in lipospheres (group B) were measured(Fig. 12). The AUC after dosing was 24,494 DPM/mL for group A (10% alcohol solution) and 8444 DPM/mL for groupB (liposphere formulation). Therefore, theabsolute bioavailabilTable 3. Repellency Effectivenessof DEET Liposphere Formulationsa

No.of biting Aedes aegypti and Anopheles stephemi % DEET in formulation

Time after application Untreatedb Controlb (min) 2

15 30

60

(4)

90 120 150 180 4 210 240 270 300 380

(5) 5 (4) 3(5) 3(5) 5 (5) 5 (3) 3 (5) 5 (3) 3 (4) 4 (6) 5 (6)

(4)

4 (0) 5 (0) 5 (3) 5 (3) 5 (5) 5 (5) 4 5 (3) 3 (4) 4 (3) 4 (5) 5 (5)

(2)

6.20 5

10

0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1(1) 1(1) 1(2) 2 3 (2)

0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (2) 0 (2) 1(2)

Referencec 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)

0 (0) 0 (0) 0 (0) 0 (0) 1(1) 0 (1) 1(2) 1(1) 2 (1)

a Cups with mosquitoes were placed for10 min every 30 min on the skin of volunteers treatedwith formulation(2.5 mg/cm2) and the number of bites were recorded. The results are averageof three independent tests (cup placements), the results in parenthesis areof Anopheles stephensi. Untreated group were without any application and control group was treated w DEET-free liposphere formulation. c Reference group received a10 wt% solution of deet in alcohol.

Lipospheres for Controlled Delivery of Substances m

405

-

c.

E

nRHANOL LIPOSPHEFIE

I 0

4

8

12

24

I

1

16

20

I

Time (hours)

Fig. 12 Bloodconcentrations(DPWmL)aftertopicaladministration of 10% DEET in ethanol or liposphere formulation administeredtopically (1 mL,20 pCi mL) to rabbits (surface area 64 cm’). Results representthe average of eight rabbits. ity of DEET from a 10% ethanol solution was 45%, whereas the bioavailability from DEET lipospheres was only 16%, a threefold reduction in the amount of DEET absorbed. About 74% of the IV administered dose was collected in the urine, and 39 and 19% of the topically administered doses were collected for the alcoholic and liposphere formulations, respectively. Assuming that the error in urine collection issimilar in all experiments, the difference in radioactivity contained in the urine after topical administration of the liposphere dosage form is about 50% of that of the alcoholic dose, which corresponds to theblood bioavailability calculations. The total amount of DEET recovered fromskin (washingof residual dose andextraction from skin) was similar for both formulations, indicating that both formulations were similarly exposed to the skin, and thus the results are comparable. The nonrecovered DEET after topical administration is probably due to evaporation of DEET from the nonocclusive patches to the environment. The lower DEET recovery of the liposphere formulation can be interpreted as more DEETis released to the environment resulting in more repellent activity [39].

406

Domb et al.

F. Nanolipospheres for Cell Targeting of Anticancer Drugs

The potential use of lipospheres loaded with paclitakel (Taxol) to overcome tumor cells’ acquired resistance to the drug was investigated [41]. It was assumed that, if absorbed by the cells, the encapsulation of Taxol willprevent the rapid expulsion of the drug outside the cell, and a sufficient cytotoxic level of drug concentration will be maintainedin cell plasma. %o liposphere formulations, one based on tricaprin and the other on polycaprolactone, were comparedwith liposomes of the same compositionbut without the core component(tricaprin or polycaprolactone). The formulations were preparedby the solvent method and extruded through a series of submicron filters to yield nanoparticles of 200-nmsize. The rateand amount of uptake of particles by cells wasdetermined using lipospheres or liposomes containing phosphatidylethanolamine fluorescein. The rate of uptake was followed by fluorescence microscopic visualizationof the amountof fluorescence accumulatedin cells or by measuring the amount of fluorescence in each cell usingthe FACS system. The formulations were incubatedwith the wild-type F98 glioma cellline or with the F98 cell lines resistant to 1 X 7.5 X and 8 X mMTaxol. The results indicate that (1) it takesabout 24 h of incubation of particles with cells to reach saturation of particle uptake, (2) cells accumulate higher concentrations of liposomes than lipospheres, and (3) cells which are more resistant to Taxol accumulate higher Taxol concentrations on incubation with Taxol liposomes and Taxol lipospheres than on incubation with the free drug. The cytotoxicity of liposphere- or liposome-encapsulated Taxol on the percentage of cell survivalof two celllines (wild type and F-98/1 X resistant cells) after 6 h of treatment with drug and further incubation of cells up to72 h was studied. F98 cells were incubated with a range of drug concentrations and different preparations for 6 h, then washed with fresh medium, and incubated for additional 66 h. The number of cells in each plate was counted daily. Thedata indicate that Taxol encapsulated in liposomes or lipospheres had a higher cytotoxic effect than freeTaxol. The results also demonstrate that although there is no significant difference betweenthe cytotoxic effect of free Taxol or Taxol encapsulated in liposomes on the wild-type cells,there is significantdifference in the effects on resistant cell lines. Taxolencapsulated in liposomes is about 30 and 50% more cytotoxic than free Taxol in cells resistant to and 10” m M Taxol, respectively. These preliminary results indicate that both lipospheres and liposomes were effective to overcome drug resistance, with the liposomes being moreeffective. The blood circulation time of nanolipospheres was compared with that of liposomes [42]. Lipospheres andliposomes of particle size between


407

100 and 200 nm containing radioactive cholesteryl hexadecyl ether were injected to mice and the tissue distribution was determined by following the radioactivity content in blood, liver, and spleen at 1,3,8,and 24 h. The cholesteryl hexadecyl ether in organs was extracted with chloroform and the radioactivity in the extract was determined. The liposphere formulation was excreted faster thanliposomes and was concentrated in the liver rather than in the spleen for liposomes. To increase the retention time of lipospheres in the bloodcirculation, lipospheres were coatedwith polyethylene glycol (PEG) of 5000 molecular weight, which has been shown to increase blood circulation [43]. Liposphere formulation made from phospholipid containing 10% w/w phosphatidylethanolaminewere reacted with aldehyde-terminatedmethoxy-PEG to form PEG-coated lipospheres. The imine-bound PEG was reduced to the amine linkage, which is more stable. Preliminary in vitro experiments indicate that PEG-coatedlipospheres are stable in serum [42].

V. SUMMARY Lipospheres are solid, water-insoluble microparticles composed of a solid hydrophobic core having a layer of a phospholipid embedded on the surface of the core. The hydrophobic core is made of solid triglycerides, fatty acid esters, or bioerodible polymers containing the active agent. Liposphere formulations were effective in delivering various drugs and biological agents, including local anesthetics, antibiotics, vaccines, and anticancer agents with a prolonged activity of up to 4-5 days. The results presented in this study demonstrate that enhanced immunogenic efficacy can be achievedbyusing liposphere-based antigen formulations indicating the potential usefulness of lipospheres in the formulation of humanand veterinary vaccines. The liposphere approach employs a fat-lipid environment to achieve several goals: to serve as a carrier to protect the antigen, to serve as “depot,”and to provide a surface interphase necessary for adjuvant activity. It is reasonable to presume that the immunogenic and adjuvant activity of lipospheres may be due toa combination of factors. These factors may include a focused and enhanced delivery of the antigen to an antigenpresenting cell (macrophage) andprotection of the antigen from metabolic destruction at other sites in the body that do noparticipate in the immune response. The liposphere-delivery system as a fat-based adjuvant formulation may provide both the surface interphase necessary for solubilization and proper orientation of the adjuvant-active material and aspotential carriers for vaccines, which may allowbetter position for processing and presentation of the incorporated antigens resulting in enhanced immunogenicity.

Domb et al.

408

The feasibility of polymeric biodegradable lipospheres as carriersfor the controlled release of a recombinant malaria antigen was also demonstrated. The microencapsulated peptide induced efficient and potent primary immune responses. Polymeric lipospheres containing R32NS1 malaria antigen were able to induce very high levels of antibody activity after one single injection in the absence of immunomodulators. REFERENCES 1. M. Donbrow (ed.), Microcapsules and Nanoparticles in Medicine and Phar-

macy, CRC Press, Boca Raton, FL, 1992. 2. P. Becher (ed.), Encyclopediaof Emulsion Technology, Marcel Dekker, New York, 1985. 3. G.Gregoriadis(ed.),LiposomeTechnology, 2nd ed., Vol. 1, CRC Press, Boca Raton, FL, 1993, pp. 527-616. 4. A. J. Domb, Lipospheres for Controlled Delivery of Substances. U.S. Patent 5,188,837, Feb. 1993. 5. A. J. Domb and M. Maniar. Lipospheres forthe Delivery of Local Anesthetics. U.S. Patent, Allowed 1993. 6. A. J. Domb, Lipospheres for the Delivery of Insect Repellent. U.S. Patent, Allowed 1993. 7. A. J. Domb, Lipospheres for the Delivery of Vaccines, U.S. Patent Application, 1990, allowed August 1992. 8. S. Amselem, C. Alving, and A. Domb, Lipospheres for the deliveryof vac(S. Cohen andH. Berncines, in Microparticulate Systems for Drug Delivery stein, eds.), Marcel Dekker,New York, 1996 (in press). as a vaccinecamer 9. S. Amselem, A. J. Domb, and C. R. Alving, Lipospheres system: Effects of size, charge, and phospholipid composition, Vaccine Res., 1~383-395 (1992). 10. S. Amselem, C. R. Alving, and A. J. Domb, Polymeric biodegradable lipospheres as vaccine delivery systems, Polym. Adv. Technol., 3:351-357 (1992). 11. S. Amselem, A. Yogev, E. Zawoznik, and D. Friedman, Emulsomes, a novel drug delivery technology, Proceed. Int. Symp. Control. Rel. Bioact. Mater., 21 ~668-669 (1994). 12. S. Amselem, A. Yogev, E. Zawoznik, and D. Friedman, Emulsomes, a New of LipidAssembly,inNonmedicalApplicationsofLiposomes (Y. BarenholzandD.D.Lasic,eds.),CRCPress,BocaRaton, FL, 1995(in press). W. Mehnert, Incorporation of 13. R. H. Muller, C. Schwarz, A. zur Muhlen, lipophilic drugs and drug release profiles of solid lipidnanoparticles, Proceed. Int. Symp. Control. Rel. Bioact. Mater., 21:146-147 (1994). 14. S. lanzini, L. Guidony, P. L. Indovina, et al., y-Irradiation effect on phosphatidyl choline multilayer liposomes: Calorimetric, NMR and spectrofluorimetric studies, Rad. Res., 98:154-166 (1984).

me


409

15. D. B. Fenske, Structural and motional propertiesof vesicles as revealed by nuclear magnetic resonance, Chem. Phys. Lipids, 64:143-162 (1993). 16. Y. Barenholz andS. Amselem, Quality control assays in the development and clinical use of liposome-based formulations, in Liposome Technology, 2nd ed., Vol. 1 (G. Gregoriadis, ed.), CRC Press, Boca Raton, FL, 1993, pp. 527-616. oxytetracycline-lipospheresformulations, 17. A. J. Domb, Long acting unjectable Int. J. Pharm., 124:271-278 (1995). 18. D. Hanibbal, M. Rock, and J. Knowles, DEET Bioavailability from lipospheres, Proceed. Intern. Symp. Control. Rel. Bioact. Mater., 19 (1992). pp. 446-447 19. M. Maniar, R. Burch, and A. Domb, In Vitro and In Vivo Evaluation of a Sustained Release Local Anesthetic Formulation,M P S Meeting, Washington, D.C., November 1991. 20. D. Dubuisson and S. G. Dannis, The formalin test: A quantitative study of the analgesic effectsof morphine, mepetidine, and brain stem stimulation in rats and cats, Pain, 4:161-174 (1977). 21. S. Haunskaar, 0.B. Fasmer, and K. Hole, Formalin test in mice, a useful techniqueforevaluatingmildanalgesics, J. Neurosci.Methods14:69-76 (1985). 22. A. J. Domb,Liposphereparenteraldeliverysystem,Proceed.Int.Symp. Control. Rel. Bioact. Mater.,20:121 (1993). 23. D. Masters and C. Berde, Drug delivery to peripheral nerves, in Polymer SiteSpecific Pharmacotherapy (A. J. Domb, ed.), Wiley, Chichester, UK, 1994, pp. 443-455. 24. D. Masters and A. J. Domb, Perinerve timed-release of local anesthetic and prolonged neural block from injectable liposphere formulations (submitted). 25. E. V. Hersh M. Maniar, M. Green, and S. A. Cooper, Anesthetic activity of the lipospheres bupivacaine delivery system in the rat, Anest. Prog., 39:197200 (1993). 26. M. J. Palham, Liposome phospholipid. Toxicological and environmental ad(0. Brown,H.C.Korting,and H. I. vantages,inLiposomeDermatics Maibach, eds.), Springer-Verlag, Berlin, 1992, pp. 57-68. 27. M.F.Landoniand J. 0. Errecalde, Tissue concentrations of a long-acting oxytetracycline formulation after intramuscular administration in cattle, Rev. Sci. Technol., 11:909-915 (1992). 28. M. Oukessou, V. Uccelli-Thomas, and P. L. Toutain, Pharmacokinetics and J. local toleranceof a long-acting oxytetracycline formulation in camels, Am. Vet. Res., 53:1658-1662 (1992). 29. D. A. Adawa, A. Z. Hassan, S. U. Abdullah, et al., Clinical trial of longacting oxytetracycline and peroxicam inthe treatment of canine ehrlichiosis, Vet. Q. 14:118-120 (1992). 30. C. R. Alving, Liposomes as carriers of vaccines, in Liposomes: From Biophysics to Therapeutics (M.J. Ostro, ed.), Marcel Dekker, New York, 1987, pp. 195-218.

410

Domb et al.

31. J. H. Eldridge, J.K. Staas, J. A. Meulbroek, et al., Biodegradable microspheres as a vaccine delivery system, Mol. Immunol., 28:287 (1991). 32. F. Stieneker, J. Kreuter,andJ.Lower,Highantibodytetersinmicewith polymethylmethacrylate nanoparticlesas adjuvant for HIV vaccines, AIDS, 5:431-435 (1991). 33. L. S. Rickman, D. M. Gordon, R. Wistar, Jr., et al., Use of adjuvant containing mycobacterial cell-wall skeleton, monophosphoryl lipidA, and squalene in malaria circumsporozoite protein vaccine, Lancet, 337:998-1001 (1991). 34. A. C. Allison and G. Gregoriadis, Liposomes as immunological adjuvants, Nature 252252 (1974). 35. T. D., Heath, D. C. Edwards, and B. E. Ryman, The adjuvant properties of liposomes, Biochem. Soc. Trans., 4:129 (1976). 36. J. R. Clem,D. F. Havemann,andM.A.Raebel.Insectrepellent(N,Ndiethyl-m-toluamide)cardiovasculartoxicityinanadult,Ann.Pharmacother., 27:289-293 (1993). 37. J. W. Lipscomb, J. E. Kramer, and J. B. Leikin. Seizure following brief exposure to the insect repellent N,N-diethyl-m-toluamide, Ann. Emerg. Med., 21~315-317 (1992). 38. H. L. Snodgrass, D. C. Nelson, and M. H. Weeks, Dermal penetration and potential for placental transfer of the insect repellent N,N-diethyl-m-toluamide, Am. Indust. Hyg. J., 43:747-753 (1982). 39. A. J. Domb,A.Marlinsky,M.Maniar,andL.Teomim,Insectrepellent formulationsofN,N-diethyl-m-touamide(DEET)inlipospheresystem,J. Am. Mosq. Control ASSOC., (1995). P. Patent, 40. A. J. Domb, Lipospheres for the delivery of insect repellent. U. 5,221,535 (1993). 41. A. Gur, Taxol Incorporated in Nanoliposphere Formulations Against Taxol ResistantCells,M.Sc.Thesis,TheHebrewUniversity,Jerusalem,Israel, 1994. 42. L. Lichtman-Teomim, Injectable systems for the delivery of insoluble anticanceragents,M.Sc.Thesis,TheHebrewUniversity,Jerusalem,Israel, 1994. 43. M.C. Woodle,M. S. Newman, andF. J.Martin, Liposome leakage and blood circulation: Comparison of absorbed block copolymers with covalent attachment of PEG, Intern.J. Pharm., 88:327-334 (1992).

15 Pharmaceutical Emulsions, Double Emulsions, and Microemulsions Nissim Garti and Abraham Aserin Casali Institute of Applied Chemisty, School ofApplied Science and Technology, The Hebrew University of Jerusalem, Jerusalem, Israel

I. Definitions andIntroduction

412

11.InterfacialPhenomena

418

111. EmulsionFormation

419

IV.EmulsificationTechniques

424

V. EmulsionStability VI. The

SurfactantandtheHydrophile-LipophileConcept

VII. PhaseInversion

VIII. StericStabilization IX. MechanicalStabilization X. EmulsionCharacterization XI. EmulsionApplications A. GeneralConsiderations B. ParenteralEmulsions C. LipidEmulsionsforDrugDelivery

427 432 438 440

444 445 448 448 449 453

41 l

412

Garti and Aserin

D. E. F. G.

Vaccine Emulsions Emulsions as Adjuvants Oral Emulsions Topical Emulsions H. Perlluoro Emulsions for Artificial Blood

XII. Double Emulsions A. Definitions and Formation B. Stability and Transport Mechanisms C. New Approaches to Improve Stability D. Naturally Occurring Macromolecular Amphiphiles

456 457 460 463 466 475 475 477 482 484

XIII. Double Emulsion Applications

494

XIV.

501

Microemulsions

XV. Microemulsions Applications A.Oral Administration B.Topical Drug Delivery C. PerfluoroMicroemulsions

XVI. Summary

511 514 515 517 518

1. DEFINITIONS AND INTRODUCTION

An emulsion is prepared by the dispersion of one immiscible liquid in another, and then it is stabilized by a third component, the emulsifier [l]. The art and science which describe and characterize these systems, in practice and in theory, are known as emulsion technology. The need to bring two immiscible liquids into a prolonged intimate contact with a high surface area has long been a task of emulsion technology. Emulsions.have been widely used in many areas of application-in industry [2], agriculture [3-51, food [6-121, pharmaceuticals [13-151, and cosmetics [16-181. Their industrial uses have an extremely wide range of possible applications such as explosives [19]; petroleum [20], concrete [21,22], rubber, leather,textile [23,24], metal [25], and paper[26] products; and mineral flotation [27]. Pharmaceutical and medical applications are, however, the most studied and are used mainly for drug administration. Emulsions in which the water is the internal dispersed phase are termed water-in-oil emulsions (W/O), whereas emulsions in whichthe oil is the dispersed phase and thewater is the continuous phase are known as O/ W emulsions [28-331. More complex systems, in which one emulsion is further dispersed into another continuous phase, are called double emul-

Emulsions and Microemulsions

413

multiple (oil) drop internal aqueous drops surfactant films

external aqueous phase

Fig. 1. Schematic illustration of a double-emulsion droplet. sions, multiple emulsions, or emulsified emulsions [34-421. Two main types of double emulsions wererecognized and studied-the W/O/W and the O/ W/O double emulsions (Fig. 1). The droplet-size distribution of emulsion droplets (sometimes also termed macroemulsion) is 0.1-50 pm. A more common averagedroplets size is0.5-5.0 pm. The inner droplet size distribution of the W/O emulsion in the double emulsionis usually smaller than 0.5 pm, whereas theouter, external double emulsionis quite large and can exceed 10 pm (Fig. 2). Amphiphilic compounds,used to stabilize emulsions (the emulsifier), possess two distinct groups in the samemolecule, which differ greatly in their solubility relationships. Such compounds were termed “amphipathics”(or amphiphiles) by Hartley [43,44] to denote the presence of a lipophilic group having an affinity for the solvent and a second group having an affinity to water. The amphiphile has, using more picturesque terms, an hydrophilic “head” anda hydrophobic (lipophilic) “tail.” Amphiphiles tendto concentrate as a monolayer at the water interface and reduce the surface tension (therefore also termed surfactants or surface-active agents). Any excess surfactant that is not accommodated at the surface migrates to thebulk and forms aggregates known as micelles (or reverse micelles, if the bulk is organic solvent or oil) [28,30,31] (Fig. 3). The micelles can accommodate organic molecules intheir hydrophobic core or may be associated only with the amphiphile within the interface layer. The increased solubility of a compound associated with the formation of micelles or inverted micells has been termed by McBain [45] “solubilization.” The solubilized compounds, when partially hydrophilic, may partition between the continuous phase,the interior of the micelle, and the interfacial region. Thus, the interfacial region

40

r

0

lnlernal droplets

10

20

-e

Mulliple droplets

30

40

50

60

Drop Diameter (pm) Fig. 2. Size distribution of the internal aqueous (W/O emulsion) and multiple drops (W/O/W emulsion) of a multiple emulsion at the preparation stage.

Adsorption at U L interface Adsorptionat

S/L

interface

Fig. 3. Modes of surfactant action on liquid-air interfaceon solid-liquid interface and in the liquid bulk. 474


415

may not consist entirely of the primary amphiphilic compound. Molecules that exhibit a substantial presence within the interfacial layers are sometimes called cosolvents or cosurfactants. These molecules are typically mediumchain alcohols. By changing the proportions of the surfactant to water and/or oil (constructing a complete phase diamgram), a micellar solution can progressively be converted into various other structures [46-501 (Fig. 4). The microstructures of the intermediate states are notfully understood. Depending on a large number of factors, including the structure of the amphiphile, its relative concentration, the temperature, and the presence orabsence of certain additives, several fascinating structures could be defined and characterized in the solution (Fig. 5). Some are isotropic solutions, amorphous in structure and mostly spherical in shape, called microemulsions, which are defined as S, (or L,) or S2 (or h)(depending on the continuous phase being water or organic solvent) [51-531. Some are isotropic solutions with a fluctuating structure (the continuous phase being oil and water alternatives) called bicontinuous structures or type I11 structures, and some are isotropic but more rigid in their structure and more viscous, appearing as gel-like, and these are called G and M, or M, phases or liquid crystalline phases. The intermediate liquid-crystalline phases may appear in a large variety of mesophases but usually in a given sequence. The most widely occurring liquid crystalline structure hasa lamellar structure [51-55] (Figs. 5 and 6 ) . The less-structured isotropic solutions, microemulsions, which are in essence “mixtures of aqueous solutions (sometimes with electrolytes), hy-

0

Fig. 4. Hypothetical phase regions of microemulsion system composed of oil (0),water (W), andsurfactant (S), showing a, inverted micelles; b, W/O microemulsions; c, cylinders; d, lamellae; e, O N microemulsion;f, normal micelles; and g, macroemulsion phases.

476

Garti and Aserin Long-chain

Alcohol ”-.

LAMELLAII LlOlJlD CRYSTAL

Ionic Surfactant 2_0

SPHERICAL MICELLES

Fig. 5. In a combination of water, a surfactant, and a long-chain alcohol, the areas for solutions of normal and inverse micelles are separatedby a micellar liquid crystal, drocarbons (‘oils’) and amphiphilic compounds,” have beenthe subject of extensive research (Fig. 7). The name microemulsions was introduced by Hoar and Schulman [56] to describe transparent or translucent systems obtained by titration of an ordinary emulsion havinga milky appearance to clarify it by the addition of medium-chain alcohol. There is muchdebate as to the term microemulsion beingused to describe such systems. Some prefer to call sucha system a swollenmicellar solution or solubilized micellar solution. However, microemulsions greatly differ from macroemulsions in the following characteristics: droplet sizes (the size ofmicroemulsions is less than 0.1 pm whereas the size of emulsions is 0.1 pm); appearance (transparent or translucent for microemulsions and milky for emulsions); thermodynamic stability (emulsions are unstable); and kinetic and time-dependent behavior (emulsion droplet size will grow with time and centrifugation); and modeof separation (microemulsionsare independent in the orderof mixing). Fig. 6. (a) M, mesophase. Rodlike micelles of indefinite length and arranged in hexagonal array. (b)G phase lamellar structures; sometimes called the Neat phase. of cubicpacking of (ModifiedfromRef.[56a])(c)Schematicrepresentation spheres. The spheres are generated by interaction of amphiphilic molecules and water.

Aserin 418

and

Garti

Fig. 7. Schematic structure of water-in-oiland oil-in-water microemulsions.

Although many investigators feel that one must distinguish between swollen micellar solutions and microemulsions in terms of the characteristics of the “reservoir” (the continuity of the oil) of the solubilized matter, there are no good instrumental methods to distinguish these differences. Thermodynamically speaking, microemulsions are in essence composed of bulk phases of water and oil separated by an interfacial region rich in surfactant, whereas for micellar systems, this is not clearly distinguished and most probably does not exist. However, with the lack of proper instrumentation, it is difficult to use different terms and, therefore, most scientists use the terms attheir convenience. II. INTERFACIALPHENOMENA

The most fundamental thermodynamic property of any interface is known as the interfacial tension [31]. The imbalanced perpendicular forces at the liquid-air and liquid-liquid interfaces were considered as the main reasons for two immiscible liquids to separate. The surface cannot be in mechanical equilibrium if there is a net force normal to the interface. The net forces at the interface were called interfacial tensions between liquids or liquid-air [30,57-591. However, many investigators concluded that it is not simple to interpret interfacial tension purely in terms of some net forces on molecules perpendicular to the surfaces, and therefore detailedstatistical mechanical models have been considered to try to describe the interfacial region [60,61]. Other approaches consider the density and the pressure profiles of the interfaces in order toexplain the complexity of the forces at the interface. Such extensive studies resulted in the discovery of various other interfacial forces, which play a significant role in explaining


419

the factors affecting the disruption of liquid interfaces and other possible recombinations. The basic macroscopic model of an interface and its properties still show two bulk phaseshaving uniformthermodynamic properties separated by an interfacial region of some thickness and over which the thermodynamic variables may possibly vary. Experimental techniques and theories (distribution functions [62,63], partition functions [59-631, summation of interaction forces in the whole interfacial region [58]) have been developed to quantify the interfacial forces (interfacial tensions). The existing methods fall roughly into two categories [64-661: those in which the properties of the meniscus is measured atequilibrium (e.g., pendant drop shape[58], sessile drop shape [67], and Wilhelmy plate methods [68]) and those measuring properties at nonequilibrium (e.g., the drop volume [69] and du Nouy ring methods [70], which are faster than any of the other methods). One has to bear in mind that often it is difficult to interpret results with regard to rupture (formation) recombination of liquids (coalescence) in terms of interfacial tensions alone and more complex rheological interfacial measurements may be required. Interfacial shear viscosity measurements (the resistance to shear stress of monolayered moleculesin the plane of the surface) and dilatational (compressional) measurements [58] (the interfacial potential [71,72] which results from spreading a changed monolayer) must be measured and considered as a major contributing factor to the formation and/or coalescence of immiscible liquids. The art and science of bringing two mutually insoluble (or only slightly soluble) liquid phases together in the form of droplets (Figs. 8 and 9) is known as emulsion technology[1,31-331. An emulsion is thermodynamically unstable and the tendencyof the two liquids to disrupt and separate is a natural and favorable process. The success of emulsion technology lies in “keeping” the system in a metastable state by opposing the mutual interfacial flow of liquids of one droplet to the otheronce thedroplets approach each other asa result of Brownian motionin thesystem. In practice, the emulsion technologists deal with. two separateprocesses: the formation of the emulsion and thedestruction (rupture) of the droplets and the consequent separation of the droplets into two immiscible liquids. Many technologists will call it the constant battle to keep the emulsion stable. 111.

EMULSION FORMATION

The formation of an emulsion is a “fast” process and takesplace in milliseconds. It is achieved by applying mechanical energy [l].First, the interface between the two phasesis deformed, forminga liquid film between the two liquids, which later deforms to such an extent that droplets form. These

Garti and Aserin

420

Oil

Water

Fig. 8. Schematicpresentation of normalemulsiondropletsstabilizedby the adsorption of an emulsifier at its surface in comparison with an oil-in-water microemulsion (emphasizing the size, curvature, and packing differences). droplets are mostly far too large and they are subsequently broken-up or disrupted into smaller ones. Hence, the disruption of droplets is a critical step in the emulsification (Fig. 10). The deformation of the droplets is opposed by the Laplace pressure. The pressure at the concave side of a curved interface with interfacial tension, y, is higher than thatat theconvex side by an amount of AP = (2y/r) (where r is the radius of a spherical droplet). Any deformation leads to an increase in AP. The interface be-

Emulsions and Microemulsions 42 1

h

t? E

m

422

Garti and Aserin

0; .

1

e

.

0;

6

0

e

Fig. 10. Processes takingplnceduring the later stage of emulsification. Surfactant is depicted by heavy lines and dots. Not to scale andhighly schematic. (After Ref. 1.)


423

tween the liquid deformation requires additionalenergy known as the viscous forces (velocity gradients) excreted by surrounding liquids, and a viscous stress has to be overcome (same magnitude as Laplace pressure). The pressure gradients(very smallportion of the totalenergy required, (35 kJ m-3) or the velocity gradients needed (approximately 3 mJ * mb3, which dissipates into heat) are mostly supplied by agitation and other external shearforces. However, certain surfactant molecules can be of significant help in reducing these gradient pressures. The molecules that are capable of adsorbing onto the droplets' surfaces and alter their properties and consequently reduce the interfacial tensions and hence stabilize the emulsion are the emulsifiers (Fig. 11). The emulsifier lowers y (e.g., from 40 to 5 mN m-') and thereby the Laplace pressure. The emulsifiers also help with the formation of film of a continuous phase between the droplets (which is a prerequisite for separation of the droplets). The adsorption and surface layer formation is mostly spread by the intense agitationof energy, and .it depends greatly on the natureand concentration of the surfactant. Adding a suitable surfactantmay reduce considerably the agitation energy (factor

-

(_Liquidphase 2

)

Adsorbed ions

I Surfactant stabilizers

t

Stablllzation by particulate solids

I Polymeric stabilizer

Fig. 11. Various possible adsorption mechanisms of surfactants on a'liquid dispersed droplet.

Aserin 424

and

Garti

of 10) needed to obtain a certain droplet, since the surfactant reduces the gradient viscosity pressure. One must realize that much of the surfactant is depleted from the interface during the emulsification process owing to the large increase in the interfacial surface area. The formation of emulsion immediately causes the droplets to seek ways to recombine and coalesce again. Luckily the coagulation process is comparatively “slow” and inpractical systems it takes minutes or months, with the net process resulting in the formation of metastable dispersed liquid droplets in the continuous phase termed emulsion droplets. An alternative way of making metastable emulsion with a large number of droplets is to start from a very tiny nucleus and then allow them to grow to the required size-the condensation method, which is relatively easy with solid particles dispersed in liquid, but is difficult for liquid-inliquid dispersions. The surfactant’s role in the emulsion formation is critical, since it largely determines which phase is going to be the continuous one: It was experimentally established by Bancroft’s rule that the continuous phase will be the one in which the surfactant is soluble. The explanation is presumably that droplets of the phase in which the surfactant is soluble are very unstable to coalesce or that a film devoid of surfactant cannot be made. IV. EMULSIFICATIONTECHNIQUES

Many devices have been designed to produce emulsions. The principles of some of those emulsifying machines are based on construction of vessels with strong flow (with baffles) on which the liquid can impinge (laminar and/or turbulent pipe flow); injection of one liquid into the other; rotorstator machines for vigorous stirring; colloid mills; ball and roller mills; high-pressure homogenizers; ultrasonic techniques; aerosol injection into liquids; and electrical and condensation devices [1,73]. It is beyond the scope of this chapter to elaborate on these various methods (Fig. 12). Emulsions in which water is the internal dispersed phase are termed water-in-oil emulsions (W/O), whereas emulsions in which oil is the dispersed phase and the water is the continuous phase are known as soil-inwater ( O N ) emulsions. More complex systems in which one emulsifier is further dispersed into anothercontinuous phase are called double or multiple emulsions (W/O/W or O/W/O emulsions). The way the substances are added is important. Usually the surfactant, which should be most soluble in the continuous phase, is indeed dissolved in that phase, butit is possible to dissolve it in the dispersed phase, which can result in the formation of smaller droplets. The in situ formation of an emulsifier is also possible (the so-called nascent-soap method). Slow addition of the dispersed phase into


Uttm T u r n

Colloid mill

425

HwnogenizePvolve

whistle’

Fig. 12. Principle of some emulsifying machines. The slit width in the homogenizer valve is very much exaggerated. Arrows indicate the directions of flow and rotation.

the continuous one during emulsification is often advantageous: It may enable addition of surfactant to the dispersed phase. If both phases are mixed in bulk, the one containing the surfactant usually becomes the continuous phase. Temperature plays a significant role in the emulsion formation. Often, it is required to first dissolve the phases or the emulsifier (if solid). Frequently, the solubility of the emulsifier and its surface tension reaction strongly depend (ethoxylated surfactants) on the temperature, and the hydrophilic water-soluble emulsifier becomes much more lipophilic and more water insoluble as the temperatureis raised. Therefore, it is essential to consider the emulsification temperature prior to any emulsification action. One must also remember that the emulsification process produces energy, the dissipation of the viscosity (velocity) gradient energy, which may affect many variables. The emulsification process can be evaluated in terms of emulsion capacity (the maximum amount of dispersed phases that can be emulsified under specified conditions). The emulsifier stability (the amount of phase separation that takes place with time under specified conditions) and the droplets size distribution (which seems to be the most practical way to study emulsification). Much has been written on the critical “events” that take place at the different stages of emulsion formation. Excellent discussions can be found in the Walstra’s chapter on “Formation of Emulsions’’ in the Encyclopedia of Emulsion Technology [l].The conclusions that should be stressed are (1) any emulsification process goes first through a film formation between the liquids which must exist for a little while; and (2) the film tends to drain rapidly under the influence of gravity. Interfacial forces tend to straighten the interface. The surfactant stabilizes the film. It allows an interfacial tension to exist, causing a considerable resistance to

Aserin 426

and

Garti

drainage of the film. As a result, it implies that slow stirring at the beginning of the emulsification (lower drainage) and high surfactant concentration are recommended. At the second stage, the plane (film) interface is disrupted into droplets. The plane disruption is accelerated by turbulence (only when y is verylow), capillary ripples (surface waves whichcan cause shredding of the droplets), and othereffects. The damping of low-amplitude capillary waves has been thoroughly studied [74]. It turns out that both the viscosity of the liquids and the presence of a surfactant enhance damping. Butif surfactant concentration is locallydifferent, as may occur in practice, the amplitudeof the wave may actually increase at the stepof lower y. It has been reasoned thatthis may be a cause of droplet formationduring emulsification. In some cases, an interface is unstable, that is, local perturbation tends to increase, and a Rayleigh-Taylor instability [75,76] occurs mostly when the interface is accelerated perpendicularly and directed from the lighter to the heavier phase (Fig. 13). Kelvin-Helmholtz instability [77] takes place inmany emulsions and arises when two phases move with different velocities parallel to the interfaces.

2

Fig. 13. Rayleigh-Taylor instability of an interface; the denser phase is below; the interface is accelerated in the downward direction. At the left side, pure Rayleigh-Taylor instabilityis depicted; at the right side, Kelvin-Helmholtz instability develops at the flanks of the waves. (From refs. 75 and 76.)

Emulsions and Microemulsions V.

427

EMULSION STABILITY

Emulsions consisting of microscopic droplets of 0.1-50 pm and dispersed in a continuousphase are thermodynamically unstable and tendto coalesce so that thereis a net reduction in the interfacial area. Hence, coalescence is a thermodynamic spontaneous process. Nevertheless, there are numerous examples of spontaneous, naturally occurring emulsions, which are stable. Milk is a good example. Stable emulsions exist in some areas of application; for example, in food,; (mayonnaise), pharmaceutics (intravenous fat emulsions), cosmetics (some skin creams), and agriculture (some pesticide emulsions). Much has been done in the last decades to understand the factors affectingsuch unusual stabilities. In order to better understand the concepts of stabilityhtability, it is essential to study the stepsand mechanisms that lead to phase separation of the liquids. Three to four main “events” can take place either in parallel orin sequence in each emulsion: creaming, flocculation, coalescence, and in some cases also Ostwald ripening (Fig. 14). All the processes lead eventually to an “emulsion breaking” and phase separation. It is essential to distinguish between the processes and to learn to control them to some extent. Creaming is the rise of dispersed droplets under the action of gravity, with the droplets remaining separated when they touch. Creaming takes place in any dispersion if the phases are not exactly equal in density and if the dispersion medium is truly fluid. It is well known that forsmall spherical undistorted dropletscreaming will depend strongly on the radius of the droplets, the difference in the density of the two phases, and the viscosity of the continuous phase. The rate of creaming is governed by Stoke’s equation:

is the creaming rate, pc-pdis the density difference between the two phases, 7 is the viscosity of the continuous phase,and r is the radiusof the droplets. Stoke’s law assumes that thefollowing conditions are valid [77]: (1) a spherical shape, (2) no droplet deformation, (3) a single particle (dilute conditions), (4) Newtonian fluid of the continuous phase with continuous properties, and (5) no diffusion. Emulsion droplets under rest are spherical, but they can be deformed in the shear field caused by the creaming. With large droplets or droplets stabilized by solid interfacial particles, distortion of the dropletscan occur owing to changes in pressure onthe different sides of the droplets (topvs bottom) which willlead to an increase in the surface area and resultin variation from Stoke’s equation. The deformation effect

U

Garti and Aserin

428

Coalescence

Primary emu1si on

Breaking

-

Flocculation

Creaming

Fig. 14. Schematic representation of the instability processes occurring in emulsions. takes place (as can be easily calculated) only for very large droplets (>l00 pm) andtherefore it is not a severe drawback. However, Stoke's law has an additional limitation, since it is good only for noninteracting droplets, which means it is good only for diluted systems. For more concentratedsystems, the term mass rate of creaming has been adopted (i.e., the motion of the center o f the mass of the dispersed phase) (Greenwald) [ 7 8 ] . Additional complexation is related to the fact that the emulsion droplets vary in size (polydispersity), which also affects the creaming rate, and finally parallel phenomena such as flocculation and coalescence should not be neglected when creaming rates are measured or evaluated. Creaming can be hastened by applying (for charged emulsions) an external electrostatic field across the end of the vessel, centrifugation, or dilution. Creaming rates can be slowed down by decreasing the emulsion droplet sizes, equalizing the densities of the two phases; and increasing the


429

viscosity of the continuous phase (by adding viscosity builders with high specific gravitysuch as polysaccharides). Flocculation is the act of stickingtogether of droplets and the formation of three-dimensional clusters. In otherwards, flocculation is a process in which an aggregation of droplets takes place as a result of collisions in combination with adhesive interdroplets forces. It is clear that since in both creaming and flocculation, there is no flow of liquid from one dispersed droplet to another, these two phenomena are reversible. Flocculation is a complex phenomenon caused by Brownian movement, gravity, and shear motions of the droplets and isverydifficult for quantitative evaluation. The free energy change of flocculation, for weakly interacting droplets AG,,, can be positive or negative depending on the droplets’ concentration, and thus one can assume a critical flocculation concentration belowwhich the emulsionis thermodynamically stable with respect to flocculation but above which reversible flocculation to an equilibrium extent occurs. For strongly interacting droplets, this critical droplet concentration is so low as to be practically insignificant, and the flocculation may be classified as being irreversible. The flocculation process is then sometimes referred to as coagulation. In thermodynamic flocculated unstable systems, flocculation may be effectively prevented if a large enough free energybarrier exists between the flocculated droplets. The free energy barrier is derived from the interfacial forces such as long-range Van der Waals forces and electrostatic forces. The net force of the two sets of interactions will determine the behavior of the emulsion droplets to flocculate and further coalesce or to stay stable in the kinetic sense (Fig. 15). A quantitative treatment of the three different fundamental mechanisms which mayinduce flocculation (Brownian, shear-induced, and, gravity induced) was recently done by Bergenstahl [77]. The Brownian flocculation was first introduced by Smoluchowski [79]and described as the reaction (interaction) between twoparticles colliding and producinga new aggregate. The flocculation rates depend onthe number of particles present in the dispersion (concentration) and on the interaction time. Bergenstahl [77] mentioned twoadditional factors that affect the Brownian flocculation: the surface forces (surface potential or surface charges) and the hydrodynamic interactions. Fig. 16illustrates the two effects on theemulsion stability to flocculate. The two factorsare verysignificant whenBrownian flocculation takes place and must be strongly addressed if such flocculation has to beminimized. Bergenstahl [77] also found that the shear-induced flocculation (also known as the orthorhombic flocculation) as described by Smoluchowski [81] has similar limitations to those of the Brownian flocculation; that is, surface forces were not included, hydrodynamic interactions were notcon-

Garti and Aserin

430

h/n m

Fig. 15. Calculated examples of repulsive(GR), attractive(GA), and total interacas a function of surface separation distance, h, of two identical tion free energy(GT) (DLVO); (b) steric spheres: (a) electrostatic repulsion and van der Waals attraction repulsion and van der Waals attractions.

10

2

Fig. 16. The stability to flocculation factor including viscous interactions versus the surface potential and the salinity (surface charge) of the emulsion [77,80].


43 1

sidered, and deflocculation was neglected. Gravity-induced flocculation, which is due to settling of particles, was first estimated by Saffman and -mer [82] assuming that every gravity-induced collision leads to attachment of the particles. Bergenstahl added to the basic Stokes’ law for settling of particles, the hydrodynamic interactions, and the surface interactions (as described by Melik and Fogler [83,84] and was able to predict stability areas forgravity-induced sedimentation as, a function of the droplet size, surface potential, and density difference between the particles. The mathematical expressions are very complicated but allow better evaluation of the possible mechanisms leading to flocculation, better predictions of the factors affecting the emulsions, and good evaluation of the stability to fluctuation that any emulsion can have. Coalescence is joining of small droplets in an emulsion to form large droplets or the process inwhichtwo droplets form one larger droplet leading ultimately to the formation of two separate liquid layers (Fig. 17).

Flocculation

l

0 Fig. 17. The primary destabilization of an emulsion is the flocculation that leaves aggregates of droplets(A). Subsequent coalescence leaves an emulsion with a wide distribution of droplet size (B).

Garti and Aserin

432

Table 1. Factors InfluencingTime for Coalescence of Emulsionsa Parameter

Effects

Droplet size

Exponential increase increasing with radius Exponential decreaseincreaswith 23 ing radius Interfacial tension 25increasing withIncrease interfacial tension Decreaseincreasing with interfacial tension.

References 23,24

26

‘Measured as time for separation rate of droplets’ size increase or as droplets’ lifetime

The coalescence process, examined from the point of view ofthe individual droplets in an emulsion, contains four basic steps: (1)The approach of two droplets toward each other achieve to an adhesive contact. (2) the drainage of the film between droplets in contact. The rate of the drainage process determines how rapid the critical thickness of the film is reached. (3) The rupture of the film. This is a stochastic process, the probability of which is determined by the thickness of the film. (4) The merging of the droplets. If the viscosity is low the process is rapid. However, emulsions with high viscosity will merge very slowly and may even be disrupted during the process. Table 1 summarizes the factors influencing time (rate) of coalescence. It can be seen that the droplets size and the interfacial tensions are key factors in the coalescence times. The coalescence process is very difficult to explain in simple mechanical events and the literature is still quite confusing. The main treatments arebased on thekinetics of the individual droplets, as was first suggested byVan den Tempe1 [S51 and later was elaborated by others.The main steps include creaming, consolidation (flocculation) (Fig. 18), film drainage, and film rupture. Bergenstahl [77], in his recent close examination and critical review of the coalescence process, examined a set of principal effects that can influence the coalescence process (Table 2). The large number of parameters affecting the coalescence emphasize the complexity of the process that takes place in the course of the collision of two droplets. VI. THE SURFACTANT AND THE HYDROPHILE-LIPOPHILE CONCEPT

Surfactants play a significant role in any of the two destabilization phenomena (flocculation and coalescence) by adsorbing onto the oil-water interface and contributing to the repulsive forces and the London dispersion


I

433

Garti and Aserin

434

Table 2. Principal Effects of Different Variableson the Various Process Steps Involved inthe Phase Separation Consolidation Drainage Rupture Creaming Variable

viscosity Droplet viscosityphase Continuous size Droplet Volume fractionof disphase persed difference Density tension Surface viscosity Surface elasticity Surface Depth of attractive minima interaction Repulsive thicknesslayerAdsorbed Solubilityof emulsifier in phasedispersed the HLB (hydrophilicflipophilic of balance emulsifier) the

0

+

0 0

0

+

-

+

+ +

+ +

+

-

-

0 0 0 0 0 0

0 0 0

-

-

+

+ + + 0 + +

0

0

-

0

0

-

+

+

+

+ 0

+ 0

0

0 0

(+) denotes that the variable has a positive effect; (-) denotes a negative effect; (0) the variable hasno effect on the phenomenon (process).

forces. Mixed surfactants are particularly more efficient than single surfactants with respect to the coalescence rate. The enhanced stability is attributed to the formation of intermolecular complexes at the oil-water interface, forming a densely packed layer and reducing the interfacial tension to very low values of about 0.1 mN * m-'. In addition, the interfacial complexes are capable of increasing the interfacial viscosity (although each component giveslowviscosity). Some investigators [86] claim that the synergism is derived from the enhanced adsorption kinetics of the surfactant when such two surfactant mixtures are used. The coherent, strong interfacial film should act as a barrier preventing coalescence by virtue of its rheological properties; for example, its high dilutional viscoelectricity. As a result, most technologists prefer using mixtures of two surfactants in most emulsion applications. The selection of surfactants is still quite difficult and requires much experience and experimental work. Griffin [87,88] has introduced the empirical concept of hydrophile-lipophile balance (HLB) for selecting a surfactant or a blend of surfactants. Surfactants with low HLB will form W/O emulsions and those with high HLB will form O/W emulsions (Table 3). Several methods have been proposed to determine the HLB of some nonionic surfactants (the HLB values are meaningful only for nonionic

W/O


435

Table 3. Summary of HLB Ranges and Their Application

HLB range

Application

3-6 7-9

8-18 13-15 15-18

agent Wetting OIW emulsifier Detergent Solubilizer

surfactants and practically meaningless for any charged surfactants) and tables with the HLBvalues are available in any textbook. Moore andBell [89] suggestedthe H L numbers (hydrophile-lipophile ratio) as an improvedcriteria for selection of surfactants. Greenwald [90] suggested a water titration procedure for surfactant in organic solution, but this method is not widely used. Davies [91,92] has developed a method by which he calculated the HLB values of surfactants directly from their chemical formula, using empirically determined group numbers: HLB = 7

+ 2 hydrophilic group number - 2 lipophilic group number

The values of the group numbersare given ina simple HLB table(Table 4). The match betweenthe Davies method for evaluating HLB and the empirical work is quite good. The experimental determination of HLB of a surfactant is also not an easy task. Greenwald [90] developed a titration procedure andRacz and Orban [93] used a calorimetric method (based on the heat of hydration of ethoxylated surfactants) (Fig. 19). Becher and Birkmeier [94] suggested an HLB determination by gas-liquid-chromatography (GLC). In any event, there was a need to find the relationship between the surfactant’s HLB and the coalescence (stability) of emulsion. As a result, different oils were assigned to different required HLBs in order to obtain maximum stability (Table 5 ) [88,95]. The formulator is asked to match the required HLB of a given “oil”or mixture of organic solvents to the HLB of the surfactant (or mixtureof surfactants) by varying the natureof the surfactants and their relative proportions in the blend. Davies [91,92] has also demonstrated the effect of the partition of the surfactant between the oil and water phases in the emulsion coalescence and related it in an empirical equation to the HLB surfactant. He showed that the HLB group numberis proportional to theenergy barrier to coalescence set up by the water which is firmly bound to the hydroxyl or ester group on thesurface-active agent molecule. He also explained the correlation between HLB and thecloud point of nonionic surfactants.

p

Garti and Aserin

436

Table 4. HLB GroupNumbers Groups Hydrophilic -S04Na+

-COO-H+ -COO-Na+ N (tertiary amine) Ester (sorbitan ring) Ester (free) -COOH -0-

CH (sorbitan ring) Lipophilic -CH"CH, -CH-,-CHDerived -CHZ-CH2-0-CHZ-CHz-CHZ-O-

38.7 21.2 19.1 9.4 6.8 2.4 2.1 1.3 0.5 0.475

0.33 -0.15

Source: From ref. 91.

One must realize that the HLB concept is based on a simplified picture of the coalescence and the formationprocess and does not take into account the important role of liquid-film thinning between droplets and the importance of shear and dilational properties of the adsorbed surfactant layer around the droplets. Therefore, it is not surprising that in practice there aremany examples of emulsifications that do not obey the HLBrules and concepts. Many investigators have cast doubt on the validity of the concept [73,75,76,96,97]. In spite of the criticism, the HLB index can be very usefuland practical consideration in the process of making emulsions. In practice, in formulating a given emulsion, one may take any pairof emulsifying agents, which fallat opposite ends of the HLBscale, for example, ' b e e n 80 (sorbitan monooleate, with 20 ethylene oxide units, HLB=5), and use themin various proportions to cover a wide range of HLBs. Emulsions are made in a given way for all combinations with fewpercentages of the emulsifying blend. The stability of the emulsions is then evaluated at each HLB number, and from the rate of droplet coalescence, the most effective HLB value, that is, the HLBvalue providing optimum stability, can be found. Various other surfactant pairs are compared at these HLBnumbers until the most effective pair is found [2].

18 -

/ /

16 1L

Wlth ethylene oxlde e+

, 1

/.

,

-

1210 m

-

86-

oxide

1-

Water number

Fig. 19. Correlation of HLB in the water number. (From ref. 90.) Table 5. HLB Values for Various Oil Phases Application Cream, all-purpose Cream, antiperspirant Cream, cold Cream, stearic acid Creams and lotions Lotions Oil, perfume Oil, mineral Oil, vegetable Oil, vitamin Ointment bases absorption washable Ointment, emollient Polishes

type

Emulsion

HLB range

OIW OIW OIW OIW

6-8 14-17 7-15 6-15 4-6 6-18 9-16 9-12 7-12 5-10

W10 OIW OIW OIW OIW OIW

W10 OIW OIW OIW

2-4 10-12 8-14 8-12

437

438

Gatti and Aserin

VII. PHASEINVERSION

In many cases, when an emulsion is made, a sudden event takes place which implies the inversion of the internal phase of the emulsion into the continuous external phase; for example, O/W emulsion inverts into W/O or vice versa. This phenomenonis related to the HLBvalue of the surfactant used, or theinternal phase volume. It has been shown that ata given emulsifierconcentration, the viscosity of the emulsion increases as theinternal phase volume(G) increases. At there is a sharp decrease in the viscosity of the emula certain critical GC, sion, which corresponds to the inversion of the emulsion. The @c depends on the emulsifier concentration, and inversion is expected to take place at higher GCas theemulsifier concentration increases [75,98]. In theearly days of emulsion technology, the inversion was attributed toa packing factorof the droplets. Thus, it was assumed that at a phase volume of db0.74, of the highemulsion droplets cannotaccommodatetogetherbecause density packing, and any attempt to increase the phase volume will cause an inversion or breaking of the emulsion. New theories and recent examples show that an inversion can also take place at low volume phases [99,100] and doesnot occur atvery highlypacked systems (99% by volume) [99]. It seems thatthe inversion isrelated more to the natureof the emulsifier than to theinternal phase volume, andthe emulsion tends to invert if the emulsifier used does not have the proper hydrophilic-lypophilic balances and tendsto migrate from the interface to thecontinuous phaseat a given concentration or temperature. Shinoda [101-1041 has shown that W/O emulsions prepared with nonionic emulsifiers tend to invert if the temperature is raised (since nonionic surfactants become less water soluble in water asthe temperatureincreases and leave the interface). In a series of papers Shinoda and coworkers [96,97,100-1101 have introduced a new concept for emulsifierperformance-the HLB-PIT relationship-and have shown that the phaseinversion temperature (PIT) of nonionic emulsifiers is influenced the by surfactant HLB number (Figs. 20 and 21) [96,97].The following conclusionswere drawn:(1) the size of emulsion droplets depends on the temperature and the HLB of emulsifiers; (2) the droplets are less stable toward coalescence close to the PIT; (3) relatively stable O/W emulsions are obtained when the PITof the system is some 2065°C higher than thestorage temperature; (4) a stable emulsionis obtained by rapid cooling after formation at thePIT; and( 5 ) the optimum stability of an emulsion is relativelyinsensitive to changes of HLB value or PITof the emulsifier, but instability is very sensitive to the PITof the system. Later, Shinoda et al. [l031 studied the effect of the molar mass and the molar mass distribution of the hydrophilic chain lengths of alkyl or


16

439

-

14

-

L

QI

n

E,

z

m

2

120' 0'6

IO

3

d4 20

40 60 80 100 Phase Inversion Temperature ('C)

120

Fig. 20. Correlation between the HLB numbers of nonionic surfactants and the phase inversion temperatures (PIT,HLB temperature) incyclohexane-wateremulsions stabilized with the surfactants(3 wt% system). 1. Tween 40 9. i-R8C6H40(CH2CH20)8,5H 2. i - ~ C 6 H 4 0 ( C H 2 C H , ~ , 7 , 7 H 10. i-R,2C6H40(CHzCH~)g~7H 3. Tween60 11. i-R,,0(CH,CH20)6,,H 4. i-RgC6H40(CH,CH20)14H 12. i-R,,C,H,O(CH,CH~),,H 5. i-R,,C6H40(CH2CH, O)& 13. i-R9C6H40(CH2CH20)7,4H 6. RL?.0(CH2CH20)10.8H 14. R,20(CH2CH,0)4,2H 7. i-R8C6H40(CH~CH2O),,,H 15. i-R&H4O(CH2CH,0)6H 8. i-R9C6H40(CH,CH20),,,H 16. i-R9C6H4C(CH,CH20)6.2H

alkyl aryl polyoxyethylene ethers and emulsifiers (having the same PIT) on the stability of O/W and W/O emulsions. They found that stability against coalescence increases markedly as the molar mass of the lipophilic and hydrophilic groups increases. Moreover, the emulsions showed maximum stability when the distribution of the hydrophilic groups was fairly broad. It was also found that in those cases where the distribution of the hydrophilic chains is broad, the cloud point is lower and the PIT is higher than in the corresponding case for narrow-size distributions. Thus, the PIT and HLB numbers are directly related parameters provided the distribution of the hydrophilic chains of the emulsifiers is similar. The HLB number of an

Garti and Aserin

440 16

14

m

1

12

IO

20

100 Phase Inversion Temperature (‘C)

60

140

Fig. 21. Correlation between HLB numbers andPITS in various oil-water (1:l) emulsions stabilizedwith nonionic surfactants(1.5% system). 1. i-%C6H40(CH2CH,03,,,H 4 i-R9C6H,O(CHZCH,O),,,H 2. i-&C6H,0(CH,CH,0),,H 5. i-%C6H,0(CH,CH,0)6,2H 3. i-R9C6H4O(CH,CH,O),,H 6 . i-R9C6H,0(CH,CH,0),,3H

emulsifier of broad distribution is lower than that of the pure emulsifier having the same PIT. A plot of HLB value and PIT forcyclohexane-water emulsions (1 : 1by volume), stabilized by a number of nonionic surfactants [96],is shown in Fig. 22. This figure is useful for the estimation of the change of the HLB value and the optimum hydrophilic chain length from the change of the PIT atvarious temperatures. VIM. STERICSTABILIZATION

Emulsions can be stabilized also by naturally occurring macromolecular substances such as gums and proteins. Such stabilization takes place quite often in food and pharmaceutical emulsions. In the past, stabilization by amphiphilic macromolecules was attributed to the formation of “films” atthe interface. The adsorption of

44 7


Fig. 22. Relationshipbetween HLB and PIT forcyclohexane-wateremulsions stabilized with nonionic surfactant(5%).

macromolecules at theliquid-liquid interface is characteristic, for example, protein like casein, bovine serum albumin (BSA), and soy proteins, and strongly depends onthe molar mass distribution of the adsorbedmolecule. The molecules form an adsorbedlayer comprising loops and tails that are anchored in the continuous phase. Thus, when two oil droplets approach each other, through aBrownian encounter, a repulsive force is generated owing to the presence of the adsorbed layers of the amphiphilic polymer. Such stabilization is known as steric stabilization. The steric interaction is effective when droplets approach a distance of b0.5; that is, just beyond conditions for thestabilizing polymer.

In this way, Napper [l111 was able to demonstrate aclose correlation between the critical flocculation temperature (CFT) of a dispersion and the corresponding e temperature of the stabilizing polymer, and also between the critical flocculation volume fraction (CFV) of a dispersion and the volume fractionof added nonsolvent required to give a 8 solvent mixture for the stabilizing polymer. With regard to the CFT,some dispersions flocculate on cooling and others onheating. Generally speaking (but not always), the former occurs when a nonaqueoussolvent is the external phase,whereas the latteroccurs when water is the external phase. This behavior mirrors the temperature phase diagrams for thepolymer plus solvent system in question. There has not been a great deal of work carried out to date on the application of these ideas of emulsion systems. Recently, we [l141 have demonstrated that excellent steric stabilization can be obtained in double emulsions in whichthe outeroil droplets are very large (20 pm) and with specially designed block copolymers (with siliconic backbone) that strongly anchor to the oil phase. Similar results have been obtained with BSA, a naturaloccurring amphiphilic protein.

IX. MECHANICALSTABILIZATION Emulsions can be stabilized also with solid particles which are not amphiphilic in their nature (inorganic compounds) provided that the solids


445

Fig. 26. Schematic illustration of solid particles adsorbed at the oil-water interface to provide mechanical stabilization.

adsorb onto theinterface (Fig. 26). Such stabilization is knownas mechanical stabilization,’’ since the solid particles provide a mechanical barrier between the droplets to prevent them from colliding and coalescing. The mechanical stabilization has only limited applications and has not been widely studied. The most impressing attempt was done by Oza and Frank [112,113] to stabilized double emulsions in the presence of a mechanical barrier obtained fromstrong adsorption of colloidal microcrystalline cellulose (CMCC). The double emulsion isa reservoir for a drug thatis strongly entrapped in the inner water phase and does not leach “on the shelf” (storage conditions). The emulsion is stable, and the entrapped drug is released slowly through the mechanical barrier. X.

EMULSIONCHARACTERIZATION

Emulsion are mostly characterized by the size distribution of the droplets and other physical properties such as dielectric properties, optical properties, thermal behavior, rheological properties, and other microscopic and macroscopic observations. The size distribution of the droplets is a very important parameter when characterizing any emulsion. Creaming, flocculation, and coalescence can be evaluated and influenced by the droplet size distribution. Therefore, it is essential to seek adequate methods to determine quantitatively the droplets’ distribution of an emulsion as a function of temperature, time, pH, and other factors that are responsible for the changes in droplet distribution. Practically speaking, only a limited number of drop-

lets are ever examined, but the resulting measurements are treated as if they constitute a continuous distribution of sues. Usually, when dealing with multivalued variables, statistical methodswill be utilized to limit the number of variables. Emulsion droplet datacan be obtained in terms of the number of droplets of a specific diameter, D, by means of optical or electron microscopy [115-1161; as diameter versus volume, V, by electrical resistance counting [117], and by diameter, D, versus mass, M, by x-ray coupled with sedimentation [118]. As a result, terms such as number of droplets, diameter (length), area, volume, and mass can be estimated or evaluated. Table 6 demonstrates the parameters that can be calculated. The datacan be presented in of droplets in range, diameter table form (droplets diameter range, number interval, and percentage on a numberof boxes of the range, the less then maximum diameter, and so forth), in histograms and in cumulative plots (of number, surface area,and volume (Fig. 27).

Table 6. Mean Diameter Definitions for Emulsion Droplets ~

Symbol ession8 literature)in

mean

Garti and Aserin

446

~

_

_

_

Mathematical

Descriptive name found (alternate

Number-length mean (arithmetic Number-surface (surface mean, diameterof average Number-volume mean (volume mean, diameter of average volume) mean Length-surface (linear mean, length diameter mean) mean Length-volume (volume-diameter mean) Surface-volume mean (surface mean, Sauter) Volume-(or weight-) moment mean (volume-(or weight-) mean, weight average particle size, De Brouckere) Could be expressed in terms numbers (AN).

. D,

(W) 122

(?%F) D,

U3

D, D, D*, D,

or D,

of number percentages (AP) as well as in actual


12

14

18

18

447

20

22

24 28 26

Fig. 27. Histogram of dropletwithinrange droplet diameter range (pm).

30

32

34

36

38

40

42

44

(% by number) as a function of

Out of all the techniques mentioned above, the most useful information is the droplets’ size distribution behavior with time (indicating the rates of coalescence and flocculation); with temperature, electrolytes, and the presence or absence of various emulsifiers (as single blends and/or in combination with macromolecules). Much can be learned on the emulsion stability from its rheological properties. Emulsions exhibit a non-Newtonian behavior which is emphasized by its variation in the viscosities as a function of its flow properties (shear rate orshear stress). A creep compliance as a function of small shear stress over a period of time gives important information relevant to emulsion stability. For more detailed information on the rheological properties of emulsion, the reader is referred tothe Encyclopedia of EmulsionTechnology, Vol. 1 [l]. A sharp decrease in viscositymeasurements (at high or medium shear rates) of highly concentrated emulsions is clearly associated with the increase in mean drop size. Several such examples exist both with W/O and O N emulsions [119-1211.

46

448

Garti and Aserin

XI. EMULSIONAPPLICATIONS A.GeneralConsiderations

Emulsions are widely used ina range of industrial and foodapplications as well as agricultural, medical, pharmaceutical, and cosmetics. This chapterwill obviouslyconcentrate only on themedical andpharmaceutical applications. The use of emulsions in pharmaceuticals was widely spread in the early 1960s and 1970s, but its importance has been diminishing in the past years, mostly because more attractive ways of intake and dosage forms have become available. The main reason for this is the increasing need for formulations that canprovide sustained, prolonged, slow, and controlled release of active substances. Furthermore, there has been a rising interest in many solid dosages of active matter such as, tablets, capsules, microspheres, microcapsules, nanoparticulated matter, and powders entrapped onpolymeric matrices. Emulsions, microemulsions, and double emulsionsalso have to compete with various types of vehicles suchas liposomes. They also suffer from a lack of thermodynamic stability and, therefore, their shelf lives are restricted and requirespecial storage conditions. Microemulsions, although thermodynamicallystable, suffer from low-entrapment capacity and require large amounts of emulsifiers, which is not necessarily permitted for drug application. Any change in the emulsion’s or microemulsion’s stability (temperature, storage, decomposition of components) will affect the drugrelease, which willimmediately affect the stability. However, since emulsions are very inexpensive and easy to prepare, they will always be attractive to formulators for certain applications. The progress which has occurred in recent years in the area of double emulsions (better stability, smaller droplet size, high capacity, and better controlled release of active matter) has been attracting renewed attention, and new formulations have beensuggested and tried. The dropletsize distribution of the emulsion is yet another significant problem. Most emulsions cannot be used for parenteral applications because of blockage (emboli) problems. Microemulsionsare small enough in size but are made of emulsifiers that do not seem to have health permits. Their applications will be considered according to their delivery methods: by injection (parenteral), by mouth (oral), and by the skin and eyes (topical). It is beyond the scope of this chapter to discuss all the possible applications mentioned in the literature orin patents, and itwould be difficult to list all the constraints related to the use of emulsifiers and oils (health and physical stability, as well as chemical or biological stability). An attempt is


449

made to bring some interesting examples of novel systems for controlled release, and discussions are brought only if the example helps in understanding or solving basic and/or critical problems related to the emulsion technology. B. Parenteral Emulsions

Parenteral emulsions are used as vehicles for carrying lipid-soluble materials, to control therelease of drugs, and, if possible, also to targetit. Most of the emulsions mentioned for parenteral use are of O/W or W/O/W type. Intravenous emulsions have very strict requirements regarding droplet size. Large droplets (over 5 pm) cause blockage (emboli) in the body. Toxicity effects have been reported in fat emulsions when meansize rises above 1.5 pm (Table 7) [122].Physicochemical parameters such as the viscosity of the external phase, phase volume ratios, and the partition coefficient of the drug are key factors influencing the release of a drug [123,124]. Variables such as droplet size distribution, surface characteristics, and surface charge are relevant factors in the relationship between physicochemical properties and physiological responses. It has been demonstrated [125,126] that fine particles “clear” from the blood slower than coarser particle-size emulsions; charged droplets clear quicker than neutral droplets and emulsionsstabilized by low molecular emulsifiers are cleared than those stabilized by high molecular weight emulsifiers [127-1291. Emulsions stabilized using phosphatides (phosphatidyl choline) were cleared quickly, whereas those stabilized with thenonionicpoloxamer (Pluronic F108) were cleared slowly. A mixed emulsifier system of phosphatide and Pluronic had intermediate characteristics (Fig. 28) [129]. The addition of drugs to the oil phase had a modifying effect on the clearance. The results support the concept that the emulsifying agent is of major importance for the removal of the emulsion from the bloodstream. Geyer Table 7. Effect of Particle Size on the Toxicity of Intravenous Lipid Emulsions Administered Intravenouslyto Rats Mean

Maximum

(pm) size (pm) System size 0.8 Fine

Medium Coarse

0.35 0.79 1.65

- 112 (103-122) 5.0

% Greater than 1pm 0 22 84

LDS0 (95% confidence limits) (mJJb)

(104-124)114

820% fat emulsionof soybean oil stabilizedby egg lecithin.

60 (53.5-66.5)

Garti and Aserin

450

2 t 0

b 100

200 lime (min)

300

Fig. 28. Effect of emulsification on the elimination of lipid emulsion. A, Pluronic F108:0, egg lecithin; 0 ,mixture of Pluronic F108 and egg lecithin.

[l271 has studied the effect of the emulsifier molecular weight on its clearance and found that clearance of fat emulsionsstabilized by the poloxamers with high molecular weight surfactants were cleared more slowly than from the blood than those emulsified with lowmolecular weight poloxamers. In spite of all those limitations, the administration of essential nutrients via the parenteral route is common practice. Glucose, amino acids, electrolytes, and vitamins are between the water-soluble matters intravenously injected. Fats areimportant sources of energy andmany efforts have been invested in introducing fat-emulsion for parenteralintake. Some commercially available lipid emulsions are listed in Table8. The oils were almost exclusively of vegetable origin, although synthetic glycerides, including simulated human fats and acetoglycerides, have been studied. Soybean oil and safflower oil [l301 appear to give rise to a low incidence of toxic reactions when purifiedand are resistant to oxidative changes (rancidity). Natural and synthetic compounds have both been considered as possible emulsifying agents. The natural agents were various types of lecithin (phospholipids) and the synthetics were nonionic polyoxyethylene-polyoxypropylenederivatives. A wide range of nonionic materials have been investigated as potential emulsifying agents for intravenous fats: polyethylene glycol stearate,


45 1

Table 8. Some Commercially Available Lipid Emulsions Trade phase name Oil

(%)

Emulsifier (%)

Intralipid Soybean 10 orEgg lecithin 20 1.2 (Kabi-Vitrum) Lipofundin S Soybean 10 or 20 Soybean lecithin (Braun) 0.75 or 1.2 Lipofundin Cottonseed 10 Soybean lecithin 0.75 (Braun) lecithin Egg Liposyn Safflower 10 1.2 (Abbott) 20) (and Travemulsion Soybean 10 or 20 Egg lecithin 1.2 (Travenol)

Other components (%) Glycerol 2.5 Xylitol 5.0 Sorbitol 5.0 Glycerol 2.5 Glycerol 2.5

diacetyltartarateester of monoglyceride (DATEM), partially esterified polyglycerol, polyoxyethylene monostearate (Myrj), polyethylene glycol (Carbowax), nonylphenyl ethers (Tergitol), and polyoxyethylene sorbitan monoesters (Tweens). However, all of these materials gave rise to toxic reactions of one form or another. Only one rangeof nonionics was found to be free from toxic effects: the poloxamers (polyoxyethylene-polyoxypropylene derivatives). These have been widelyused for fat emulsions either ontheir own or as coemulsifiers withlecithin [131,132]. Although Black and Popovich [l331have attemptedto give guidelines and recommendations. Burnham and others [l341 have reported that a stable fat emulsion (Intralipid) intravenous-based feeding mixture have been preparedto simplify nutritional support for patients with gastrointestinal disease. Mixing of Intralipid with other nutrients before administration allows a constant infusion through peripheral veins. The physical properties of fat particles in Intralipid formulations were studied before and aftermixing withthree different combinations of amino acids, dextrose, and electrolyte. The effect of storage on lipid particle size was also examined. The mean particle size of the dropletsincreased slightly over a 48-h storage period at 4"C, but no large droplets were observed. Measurement of the electrophoretic mobility of the droplets indicated that the mixture of a fat emulsionwith an aminoacid, dextrose, and electrolyte did not produce any permanent change in the stabilizingfilm of phospholipid. Total concentrations of divalent cations are greater than 2.5 mmol/L, such as calcium and magnesium, couldlead to aggregation of fat droplets and the separation of the oil as a cream layer. These systems had poor short-term stability, and consequently Burnham et al. [l341 recorn-

452

and

Garti

Aserin

mended that feeding mixtures containing amounts of Ca2' and M e above this limit should beused with care. In similar fashion, Hardy et al. [l351 have shown that mixtures of amino acid, electrolytes, trace elements, glucose, and 10% soybean oil lipid emulsions werestable when prepared aseptically in3-L plastic containers and stored at refrigerated temperature for 72 h. The pH-buffering property of the amino acid solution was believed to be an important contributor to the stability of the mixture [136]. The lipid emulsion (mixed with amino acids and glucose for parenteral administration of dogs) was eliminated andmetabolized to triglyceride, free fatty acids, andphospholipids in a fashion similar to that observed during separateadministration of the same nutrients. Clinical data supportthese findings [137]. The clinical use of fat emulsions in parenteral nutrition has been described in detail in various textbooks and reviews [130,138-1411, and a variety of products is available, including those containing amino acids together with fat emulsions [130,142,143]. The possibilities of administering complete intravenous nutrition with fat as one of the sources of energy has been demonstrated in many investigations. Emulsified vegetable oils were used as test systems to explore the characteristics of the reticuloendothelial system. The role of a,-macroglobulin in the phagocytic process (in rats) using a lipid emulsion has been investigated by Molnar et al. [144]. They found thata,-macroglobulin was an important component, but thatthis material was not related to the a,macroglobulin in humans. They pointed out that plasma contains at least 60 proteins, and this does not include enzymes, clotting factors, complementing components, and hormone carriers. Ashworth and others [l451 examined the uptake of small lipid droplets by hepatic and Kupffercells. The uptakeof small particles was dependent on thesize and potencyof the hepatic sinusoids. In addition, the size, composition, structure, and charge on the particles was considered to be relevant. A critical particle size of 100 nm was proposed. Tonaki et al. [l461 used a gelatin test emulsion to investigate plasma components that promote phagocytosis (opsonins). Particles coated with human serum albumin werealso employed. Different uptakes in liver and spleen regions as well as small uptake in the lung were reported. They indicated thatthere was some specific interaction between reticuloendothelial cells and particles which was dependent on the nature of the particle surface. Intravenous O/W emulsion systems are also employed as diagnostic agents in the form of injectable radiopaques. Kunz et al. [l471 described in detail the preparation,sterilization, and stability of 50% oil-in-water emul-

roemulsions Emulsions and

453

sions of iodized oil, iophendylate injection, and ethiodized oil (Ethiodol). Three emulsions of each oil injected intraperitoneally into rats gave an excellent radiopaque outline of the peritoneal cavity. Emulsion technology isquite a mature science withregard to pharmaceutical preparations. Most studies were carried out onmicroencapsulation as a release vehicle. Only a limited number of studies are available on improvements or innovations related to emulsions. Some of the key issueswere fat emulsions for parenteral applications:

1. Studies related to the stability assesment and its relation to the nature of the phospholipids [148-1501. Highly negatively charged droplets were prepared by using purified phospholipids enriched with negatively charged phosphatides; use of amino acids, urea, or carboxylic or sulfonic acid to form stable emulsions even after mixing with blood plasma [151]. 2. Attempts to make bettercharacterized parenteral emulsions(by Coulter counter [152,153]) and to improve their productiontechniques (microfluidizer [1541). 3. Stability studies related to thermal treatments[155,156]. 4. Effect of various nutient ratios on the emulsion stability of total nutrient admixture [157]. 5. Elimination and metabolism of a fat emulsioncontaining mediumchain triglycerides [1581. 6. Transdermal new formulations for special applications have been also studied. For example, new nicotinic devices and their usein the treatment of withdrawal symptoms associated with smoking cessation [1591. C. Lipid Emulsions for Drug Delivery

Low water soluble drugs are difficult to administer and, therefore, are good candidates for formulation into O N emulsion. Fat emulsions are known and used in parenteral nutrition (artificialcylomicra have been developed in the form of fat emulsions). The two-phase emulsion system should havean advantage overa solubilized system(use of cosolvents such as polyols) in that the drugcontained therein will not be able to precipitate as solid (harmful) particles when diluted. Furthermore, the presenceof the bulk of the drug in nonaqueous environment may lead to an increased stability of the drug (e.g., reduced hydrolysis) as well as to a possible controlled-release system. The use of soybean oil emulsionsas carriers for lipid-soluble drugs has

Garti and Aserin

454

30.8 LO.0 Dose ( m g I kg I

52.0

Fig. 29. Administration of thiopentone in aqueous and emulsion forms.0 ,emulsion; 0,solution. (After ref. 164.)

been pioneeredby Jeppsson and Ljunbergin Sweden [160-1631. The drugs, dissolved in the oil phase, have included barbituric acids, cyclandelate, diazepam, andlocal anesthetics and the routes used have been intravenous, intra-arterial, subcutaneous, intramuscular, and intraperitoneal. The greatest interest hascentered around the use of fat emulsions asvehicles for the intravenous administration of drugs [1601. For thecase of the barbituric acids [MO], a prolongation of anesthesia was observed when the drug was administered in the oil phase of a soybean emulsion comparedwith a solution of the corresponding sodiumsalt (Fig. 29). The results were explained either by a slow release of the drug from the oil particles or thepossibility of a more specific delivery of the drugsto the central nervous system (CNS) when the drug is confined in the oil droplets, the latter being due to the type of close association of droplets and thevascular lining of the CNS. As evidence for the possible close association of lipid particles with vascular linings, Jeppsson and Ljunberg[l641 cited the work of Schoefl and French [165], who demonstrated that in the mammary glands of lactating mice, chylomicra and artificial fat particles (soybean oil emulsion) were concentrated against the luminal surface of the endotheliumof small blood vessels and appeared to adhere closely to it. In some cases, the fatparticles appeared to have sunk deeply into thecytoplasm, or it appeared that thin flaps had encircled them. In contrast, in the lung, the fat particles were freely suspended in the blood plasma. The role of an enzyme, believed to


455

be lipoprotein lipase, in this adhesion process and the transport of fat through the vessel wall was discussed. Many more reports exist in the literature on drug delivery by the intravenous administration of fat emulsions containing drugs. Examples are the administration of lignocaine [164,166] (which protects the heart from electrically induced arrhythmias), diazepam [167-1701, hexobarbital [l701 and phenyl butazone[170]. In spite theextensive work invested in stabilizingthe drug-containing fat emulsions with various peimitted emulsifiers, it is not yet clear whether the emulsifiers and the fats decompose in the system and whether theyplay any role in the release and/or targeting of the drug. Some investigators claim that certain effects (such as sustained release) could well be related to metabolic changes induced by the presence of the emulsion rather than delayed drugrelease or drug targeting [170,171]. Narcotic antagonists are importantcandidates for slow-release action and, therefore, have been tested in various formulations to induce physical dependence in rats following subcutaneousdepot injection [172,173]. Valinomycin [l751 and other anticancer drugs have been introduced into intravenous emulsions [173-1751. Tests in animals showed that the drug contained in an emulsion had effects similar to those of an aqueous suspension, but that the emulsion formulation required a 20-fold lower dose to produce these effects [176]. Artificial intravenous emulsions havealso been tested as “on-target” delivery systems for drugs that arevery oil soluble. The drugis dissolvedin the oil and the emulsion is injected. The fat particles (chylomicrons) can deposit and close contact between the fatparticles, and the endothelium can allow passage of the drug from the oil phase tothe tissue without it necessarily passing into theplasma. The excess chylomicrons willbe cleared by the liver, adipose tissues, heart muscle, and lactating mammary glands. In this connection, it is of interest to review the work of Meyerson [177], who found a prolonged-release effect after dissolving progesterone in soybean oil and thenemulsifymg it to give an O/W emulsion forintravenous administration. Similarly, Mitushima et al. [l781 have employed a lipid emulsion as a novel carrier for corticosteroids. Their studies in rats have suggested that corticosteroids incorporated inlipid emulsions are taken up by the reticuloendothelial system and certain inflammatory cells to a greater extent than free corticosteroids, thus resulting in stronger antiinflammatory activity [178]. Thus, it was believed that corticosteroids in lipid emulsion could be of value in the treatmentof certain diseases such as rheumatoid arthritis, where phagocytic inflammatory cells play an important role in the pathogenesis.

Aserin 456

and

Garti

D. Vaccine Emulsions

Adjuvant methods of immunization havethe advantage of achieving both the primary and secondary immune responses with only one injection of antigen and of maintaining the antibody level over a long period of time. Adjuvants have been used recently in tumor immunotherapy [179]. An adjuvant effect is obtained by adsorbing the antigen onto a mineral gel (aluminum hydroxideor phosphate). This is widely usedas an immunizing agent for human use [179]. Water-in-oil emulsions, consisting of anaqueous antigen solution dispersed in a mineral oil phase with the aid of a lipophilic emulsifier (mannide-monooleate), have been prepared to act as efficient immunological adjuvants. These emulsified systems presumable enhance the antibody response (slowly releasing secondary stimulation of antibodies) and the stability of the vaccine on storage [180,181]. Various mechanisms have been suggested to explain the high antibody response which these adjuvants produce, and it has been assumed that the emulsion acts as an inert depot, whichslowly releases antigen over a long period of time [179,182-1841. Steinberg [l831 has suggested that in addition to the depoteffect, the oil attracts mononuclear cells about the antigen which can take part in antibody production andgive a high antibody level. Freund and McDermott [l851 were some of the first to demonstrate adjuvant activity of mineral oil emulsions. These simple water-in-oil emulsions, subsequently termed incomplete adjuvants, were later modified to give a greater antigenic response by incorporating dried, heat-killed Mycobacterium tuberculosis in the mineral oil phase. Such emulsions are termed Freund's complete adjuvants [186]. The complete adjuvantinitiates a cellular response andis too reactive for immunization in humans, although it is used in experimental work [186-1891. Problems have been encountered in the use of mineral oil emulsion adjuvants, since mineral oil isnot metabolized andwill persist at theinjection site [179,187,188]. This has limited the use of mineral oil emulsions in humans. Studies have, therefore, beencarried out in the search for alternative oil phases for adjuvant preparations (hexadecane, squalane, sesame oil, and peanut oil. Stewart-Tu11and others have found that low molecular weight hydrocarbons of chain length C,&,, were goodalternatives to mineral oil). Early studies with vegetable oils .were unsuccessful, as the oils were unstable and gave lessantibody response thanmineral oils [189]. However, further work produced a suitable adjuvant of peanut oil [187]. This adjuvant consists of a water-in-oil emulsion of aqueous antigen together

Emulsions

457

with an equal volume of an oil phase. Arlacel A is the emulsifier and aluminum monostearateis the stabilizer. A more recent report by Reynolds [l901 describes the adjuvantactivity of a refined peanut oil base emulsified inaqueous vaccines with glycerol and lecithin. This formulation was found to be metabolizable and caused no granulomasor abscess formation. E. Emulsions as Adjuvants

Oil-in-water emulsions have beenused as adjuvant-type systems in cancer chemotherapy [191].Successful tumor regression has been achieved by intratumoral injections of viable Bacille Calmette-GuCrin (BCG)vaccine, and a potential antitumor preparation has been constructed by combining trehalose dimycolate purified from BCG with aqueous soluble extracts of Re mutants of Salmonella species (Re glycolipid). An essential requirement of activity has been the formulation of an O/W emulsion. A drug injected into the interstitial spaces in muscle tissue is transported away from the injection site by the circulating blood, but someof it also reaches the regional lymph nodes. The amount of drug so absorbed depends on the injection formulation and the site of the injection. It is known that lipids, fatty acids, and high molecular weight substances are absorbed mainly to the lymph system and lower molecular weight compounds are absorbed into the blood. Many important drugs are water soluble with relatively low molecular weights; therefore, a specific drugdelivery system in the form of an emulsion or other colloidal system is necessary to deliver such agents to thelymphatic system. Sezaki, Hashida and other workers [192-1981 have investigated the use of a water-in-oil emulsion to provide drug-delivery systems using a series of anticancer drugs in a series of formulations; water-in-oil emulsion, water-in-oil emulsion containing gelatin, and also simple oily and aqueous solutions and aqueous intravenous injections for comparison. Injections were given by intramuscular, intraperitoneal, and intragastic routes. Concentrations of both the drug and the oil were measured in the blood, regional lymph, thoracic lymph, and lymph in other parts of the body and also at the injection site following injection at various sites. It was suggested that absorption of a drug from its injection site could be absorbed into the bloodstream or into the regional lymph nodes. It passes to the thoracic duct and thento thebloodstream. It can travel as theisolated drug or in the emulsion formulation. Using mitocycin C and bleomycin, it was noted that the water-in-oil emulsion produceda much greater specific delivery of drug into the lymph system than did an oil-in-water emulsion [195-196]. Sezaki et al. [195],

L

458

Garti and Aserin

therefore, suggested that encapsulating a water-insoluble drug in oil was highly advantageous forthe transport of drug fromthe interstitial spaces to the lymphaticcapillaries. They also noted that theconcentration of oil was much higher in the lymph nodes thanin the thoracic lymph and that it rose to a maximum quickly and then decreased in the former, whereas it was relatively constant in the latter. This suggested that theoil was transported to the lymph nodes, where it accumulated and was gradually released downstream. An oil-in-water emulsion containing gelatin as an emulsifying agent was found to be intermediate in its lymphatic targeting ability between the water-in-oil emulsionand the oil-in-water emulsionprepared by using polysorbate 80 [195-1961. The gelatin-containing emulsion was investigated, andit was found that themitomycin C was partially bound to theoil drops [195]. In subsequent experiments, a water-in-oil emulsion, using a relatively high concentration of gelatin as part of the emulsifying system, was compared with the simple water-in-oil emulsion. The concentration of the gelatin used was20% of the aqueousphase, and the gelatin was thought to form solidified microspheresin the oil. It was noted thata high drug concentration surroundingthe injection site was prolonged with the emulsion formulations, especially the gelatin emulsion. A sustained release of drug into both lymph and blood systems was significantly improved with the two emulsions comparedwith the aqueoussolutions [194-1981. The time during which the drug concentration in the regional lymph node was higher than that in other lymph nodes (in the orderof several hours with both emulsion formulations as compared with halfan hourfor the aqueoussolution). Also, the time delay to reach a maximum or peak concentration of drug in both blood and lymph was greater with the emulsion formulations. Microscopical observations made of the muscle fibers and the regional lymph nodes showed that multiple water-in-oil-in-interstitial fluid emulsion was formed in the muscle tissue on injection. With the aqueous gelatin-in-oil emulsion, microspheres (1-2 pm in diameter) were observed to be dispersed in the oil droplets. The investigators [193,194] suggested that this system wasmore stable than the simple water-in-oil emulsion and that itprolonged the release of the drug forthis reason. This suggests that the release of drug is due primarily to the breakdownof the emulsion. Measurements of the total amountsof drug absorbed intothe lymph system revealed that the microsphere-in-oil emulsion was superior to the water-in-oil formulation in promoting specific lymphatic absorption. It was concluded [l941 that water-in-oil and microsphere-in-oil emulsions would be advantageous for use in cancer chemotherapy. In the light of these experiments, bleomycin in the form of a microsphere-in-oil emulsion was injected into the appendix of rabbits having a carcinoma on the organ after surfkal removal of the carcinoma [198]. A


459

mixture of medium-chain triglycerides (MCTs) wasused as theoil phase of the vehicle, since this had a lower viscosity than the sesameoil previously used and, therefore, was easier to inject. The MCT-based emulsion improved the original delivery characteristics of the microsphere-in-oil emulsion with sesame oil as the oil base. It produced a definite improvement in the extentof metastasis. The metastatic cells were seento bedestroyed after 1month (observedmicroscopically) compared with an aqueousformulation which slightlydamaged diseased cells, but these recovered within a month. It also delivered little drug IO the lung or liver, where it is especially toxic. Takahaski et al. [l991 investigated the use of water-in-oil emulsions as a vehicle for the injection of anticancer agents (bleomycin and mitomycin C) directly into tumorsfor antitumor activity and also for the preventionof lymphatic metastasis, thereby making use of the sustained-release effect of such a formulation. They found in experimental animals a prolonged level of drug in the tumor using a water-in-oil emulsion, and also a decrease in tumor size of 50% in 16 days and 75% survival rate compared with a 5% decrease in tumor size and a 40% survival rate with an aqueous injection (Fig. 30). Clinical trials on carcinomas produced encouraging results. In some cases, a multiple emulsion was employed as the delivery system.

-1 00

2

4

6

8

1014 12

16

Time after injection (davrt

Fig. 30. Changes in mean tumor diameterof rat carcinoma aftera single 0.5-mg dose of bleomycin in various forms. 0 , untreated control; A, aqueous solution intravenous; V, aqueous solution intratumoral; 0, emulsion intratumoral. (After ref. 199.)

460

Garti and Aserin

The chemical composition of a digestible oil may also be critical, as this will determine the physiological mechanisms stimulated during digesof tion. Forexample,thehydrocarbon chain length andthedegree unsaturation of the fatty acids liberated by triglyceride hydrolysis determine the extent of changes in gastrointestinal motility, with the longerchain unsaturated fatty acids havingthe greaterinhibitory effect [200,201]. Incorporation of long-chain fatty acids into the bile salt micelle increases the solubilizing capacity of the micelle for lipophilic materials within the intestinal lumen [202,203], and incorporation of unsaturated long-chain fatty acids may result in an increase in the permeability of the mucosal membrane [204,205]. Short- and medium-chain fatty acids are absorbed via the portal route, whereas long-chain fatty acids (>C,,) are absorbed into the lymph, with the unsaturated acids stimulating chylomicron synthesis. Lipophilic compounds selectively absorbed into the lymph are carried in the chylomicron [206], the natural fat particles, and therefore administration of long-chain unsaturated fatty acids may stimulate lymphatic absorption. It is apparent that the chemical composition of the oil phase of the emulsion is a major factor in determining the extent of drug absorption following oral administration. F. Oral Emulsions

Oral administration of drugs in an emulsion formhas been known for over 100 years and is easier to use than intravenous injection. Its main advantage is the morefacile absorption of certain drugs from the gastrointestinal tract. The ONV emulsions arerelatively easy to prepare, since there are no strict requirements on droplet size, nature of surfactant and its purity, nature of its fat, norphysical, chemical and biological properties. The range of emulsifying agents is greater than that available for intravenous emulsions. Materials suchas acacia, tragacanth gums, and methylcellulose as well as various nonionic surface-active agents have been employed in pharmaceutical formulations for oraluse [207,208]. The ionic surfactants are notnormally usedfor internal preparations but arereserved for topical and cosmetic applications. The G U S (generally regarded as safe) listings of the U.S. Food and Drug Administration are a useful source of information [209]. Vegetable oils suchas cornoil, peanut oil, and soybeanoil are used as emulsion vehicles for drugdelivery. Mineral oils (liquid paraffin) are emulsified to provide a laxative effect but are not used for drug administration unless the drugis intended forthe same pharmacologicaleffect. Many attempts have been madeto entrapvarious drugs in O N emulsions for oral adsorption.


461

Higher levels of drug in the plasma during the first 4 h after oral administration of acetyl sulfisoxazole O/W emulsion were observed in Svenson's early studies [210]. The emulsion was also shown to be three to five times more active against streptococcal and pneumococcal infections in mice. Similar results were reported with sulfonamides [211]. In most formulations, greater advantages were found in drugs with limited water solubility (poor dissolution), such as indoxole [212], griseofulrin [213,214] (Fig. 31), phenytoin [215], and theophylline [216]. Formulation of nitroglycerin in a sesame oil-in-water emulsion yielded lower and later peak plasma levels than the equivalent aqueous solution, butbioavailability was unaffected [217]. The delay in drug absorption from the emulsion was attributed to the time required forthe drug to move from the oil phase into the aqueousphase prior to absorption andto effects of the oil on gastric emptying. Thus, the emulsion provided some sustained-release characteristics to nitroglycerin absorption, whereas not affecting the extent of absorption.

3 r

'1

Time (hrl

Fig. 31. Administration of griseofulvin in different dosage forms (30 mglkg of micronized griseofulvin inrats). 0, aqueous suspension;A,corn oil suspension; 0 , corn oil-in-water emulsion containing suspended griseofulvin.(After ref.213.)

462

Garti and Aserin

Absorption of the fat-soluble vitamin, vitamin A, occurs primarily via the lymphatic pathway [215,218]. Emulsification of vitamin A palmitate and vitamin A alcohol in corn oil has been found to increase the total absorption of vitamin A in normal patients and in those with cysticfibrosis [219,220]. These observations have important implications on both the dietary management of patients with cystic fibrosisand, moregenerally, on the use of the lymphatic system for drugdelivery. Drug absorption into the lymph has twomain advantages overthe portal route. First, the possibility of targeting high concentrations of anticancer agents into the lymph to prevent metastasis along the lymphatic pathway. Second, reduction of firstpass metabolism by delivery of the drug directly into the general blood circulation. The lymph is the primary route of absorption for lipophilic dietary compounds suchas cholesterol, long-chain fatty acids, and the fatsoluble vitamins. As seenwith vitamin A, adsorption may be improved by administration with lipids. Similarly, it has been suggested that the lymphatic absorption of lipophilic drugs may be stimulated by coadministration of lipids [221]. The oral absorption of heparin was shown to be insignificant in rats and gerbils following the intraduodenal administration in aqueous solution in micellar solutions of monoolein or sodium taurocholate or in sodium taurocholate-stabilized mineral oilemulsion[222]. However, significant absorption was achieved by administration in trioctanoin, corn oil,or peanut oil emulsions stabilized with sodiumtaurocholate. Although no mechanism for the enhanced absorption of heparin was proposed, the apparent dependence of this effect on thedigestibility of the oil used in the emulsion was noted. Thiswas seen again in absorption studies with the glycoprotein, urogastrone [223]. Absorption was unaltered following administration in aqueous solution, 0.2% (w/v) Ween 80 solution, liquid paraffin emulsion, and diethylphthalate emulsion but was significantlyincreased by administration in trioctanoin and in olive oil emulsions. Studies byBloedow and Hayton [224], using lipids rather than emulsions, suggested that generally polar digestible lipids increase the bioavailability of lipophilic, poor watersoluble drugs, whereas nonpolarnondigestible lipids have no effect on the bioavailability but may reduce the rateof absorption [225,226]. The mechanism of intestinal absorption of drugs from oil-in-water emulsion systems has been explored by Kakemi et al. [227-2291 using an in situ large intestinal rat gut loop model. Their results suggested that the partition coefficient of the drug and the absolute volume of the aqueous phase is of importance, with absorption occurring mainly from the aqueous phase. For a drug with a partition coefficient of less than 1, drug transfer into the aqueous phaseis rate limiting rather thantransfer from the water


463

to theabsorptive membrane. However,Noguchi et al. [230]proposed that for alipid-soluble, water-insoluble drug, the absorption process consists of adsorption of vehicleoils to the absorptive membrane, partitioning of drugs intothe membrane lipids with simultaneous partition into hydrolyzed oils, release from the membrane lipids or hydrolyzed oils with movement into the inner compartments, and then transport into the portal or lymphatic systems. The absorption of ephedrinefrom orally administered emulsions given to dogs was investigated byLin et al.[231], whoattempted to rationalize their results using the HLB concept. Correlations of ephedrine present in urine in vivo with the total amount released in vitro using a dialysis method were significant at five different HLB values in the range 10-14. Although emulsionsmay enhance oral drug absorption and biological availability, the underlying mechanisms causing this effect remain unclear. Both physiochemical and physiological functions appear to be involved; for example, dissolution rate, oil-water partition coefficient, gastrointestinal mobility, bile flow, lymphatic absorption, and membranepermeability. The exact way in which these factors interact to affect drug absorption will depend largely on the natureof the drug and theoil phase. Oral administration of drugs in emulsions, therefore, seems to have great potential as a dosage form provided the problems of stability and drug release can becontrolled. G. Topical Emulsions

Emulsion systems can be administered to theexternal surfaces of the body and to body cavities as topical formulations in the form of O/W or W/O creams containing drugs. General reviews on the structureof the skin, the absorption of drugs through the skin, and pharmaceutical and cosmetic products for topical administration can be found in the relevant textbooks [232-2341. A good topical formulation should be one in which the dosage form has both physical and chemical stability, has cosmetic acceptability, and also provides the optimum environment for the active ingredient to reach the skin surface [235]. For the corticosteroids, which may be regarded as representative molecules, the topical therapeutic activity may be considered as the result of three interactions: release, penetration, and antiinflammatory activity. These, in turn, result from the interactions between corticosteroid, vehicle, and skin. In the design of topical formulations, all three factors must be considered to optimize drug delivery [233].

464

Garti and Aserin

In general, a suitable topical preparation may be produced by:

1. Optimization of the drug concentration in order that all the drug be in the solution 2. Minimizing the amount of solvent used so that thechemical potential of the drug is maximized inthe solution 3. Ensuring that the vehicle components affect the permeability of the stratum corneumin a favorable manner In emulsion systems, the distribution of the drug between the various phases and the total drug concentration will define the overall concentration gradient that exists across the skin. Also, the relative proportions of the lipid and aqueous phases in the emulsionwill affect the degree of hydration of the stratum corneum and hence drug penetration rates [236]. In general, emulsions will be less occlusivethan greasy materials. The pharmaceutical andcosmetic of a topical product is related toits viscosity, which will affect its consistency, spreadability, and extrudability and will determine the rate which at the active drug candiffuse to the outer layers of the stratum corneum [234,237]. If a thick layer is applied to the skin, the drug in the upper regions of a viscous preparation will not reach the skin during the normal application time. Surfactants present in emulsion systems mayalso influence the rateof release from the formulation as well as the rate of absorption [238]. Many publications have indicated that the presence of soaps can enhance the penetration rates of poorly absorbed materials [239-2421. The surfactants . may also promote the diffusion from the base itself and hence influence therapeutic performance[236,242]. However, simple relationships between the HLB and surfactant action could not be established, and it was concluded that the surfactants had a direct action on the protein rather than the lipid components of the stratum corneum [243]. The following types of emulsifying agent are used pharmaceutically:

1. Zonic: examples of anionic surfactants commonlyused are the alkali salts of the higher fatty acids and sulfate esters of the higher fatty alcohols; for example, sodium cetyktearyl sulfate. Cationic surfactants employed include the long-chain quaternary ammonium compounds; for example, cetrimide. If ionic surfactants are included in a formulation, care must be taken that they do not interact with the active ingredient [243]. 2. Nonionic: a large number of nonionic emulsifiers existfor stabilizing topical emulsions. The long-chain alcohols (e.g., stearyl) are very hydrophobic and are used to stabilize W/O emulsions. The


465

corresponding polyoxyethylene glycol ethers have higher HLB values and therefore better emulsifying capacity and are used in the preparation of O/W emulsions. The overall properties of nonionic surface-active agents depend onthe relative proportions of the constituent hydrophilic and lipophilic moieties of the molecule (HLB value). 3. Mixture of emulsifiers: to obtain a stable emulsion, it is often useful to use mixed-surfactant systems. A combination provides a more rigid film at the lipid-aqueous interface and an increased reduction in the interfacial tension. An example of a complex emulsifiersystemiscetyl alcohol, sodium lauryl sulfate, and glyceryl monostearate. The rheological properties of semisolid emulsions made frommixed emulsifiers have been studied extensively by Barry and Eccleston [2442471. The system comprising oil-alcohol (usually a mixture of cetyl and stearyl alcohols)-surfactant (cetrimide, sodium lauryl sulfate, or Cetomacrogol) was found to producea viscoelastic structurethatgavea bodying effect to theformulation through the formation of a gel network of liquid crystals. The rheological properties were highly dependent on the quantity and natureof the surfactant and the alcohol. Interestingly, the use of pure alcohols of different chain lengths gave rise to the formation of semisolid products with poor stability. Oil-in-water creams are water miscible and hence washable andmay be used in a range of active ingredients. After application, the film produced is less occlusivethan a W/O emulsion system.They spreadeasily and their advantage is that as the water evaporates, they exert a cooling effect at the skin surface. They also produce little irritation, yet their major disadvantage is that despite their fat content,they dry out. Water-in-oil emulsions have some advantages in certain disease conditions, where an emollient effect on the stratum comeurn is required. They also provide some protection barrier for the skin and generally provide occlusion. The occlusion effect increases the hydration of the stratum of some drugs. corneum and hence enhances the permeation Examples of the effect of formulation on thepercutaneous absorption of methyl nicotinate are shown in Table 9. The data show that aqueous cream BP releases methyl nicotinate more readily than oily cream BP. The table also showsthe effect of decreasing the drugconcentration on thetime of onset of erythema. Simple formulation changes can markedly affect drug release, as shown by comparing the time of onset of erythema after the addition of glycerol to aqueous creamBP.

and

466

Gatfi

Aserin

Table 9. Percutaneous Penetration of Methyl Nicotinate from Four Emulsion Vehicles Concentration of methyl nicotinate 1 4.1 0.5 0.1 0.05 0.01

Time of onset of erythema (min) aqueous cream BP 3.3 6.3 4.0 6.3 4.6 16.7 5.9 7.4

oily cream BP

aqueous cream

aqueous cream glycerol

+ 40% glycerol + 60%

4.5 5.2 6.8 10.5

4.6 5.6 6.7 11.7 17.6

-

-

Many publications have discussedthe effect of formulation (including emulsion systems) on the activity of topical corticosteroids [248,249]. The factors influencing activity concern the physicochemical characteristics of the steroid and the choice of the vehicle. Simple constants such as the partition coefficient and the solubility of the steroid are important. It appears that the greatest release is obtained from preparations in which the drug is dissolved at its maximal solubility concentration. Other factors, such as the viscosity of the vehicle and the effect of the vehicle on the degree of hydration of the stratum corneum, arealso significant. The influence of vehicle on the penetration of individual potent corticosteroids is also very significant. Marked differences in the activity of the same corticosteroid in preparations of different companies could be attributedto vehicle-dependent effects showing the significance of formulations. Many formulations forcreams and semisolid products areavailable in the literature and patents. The scientific studies, however, concentrate on the various aspects of the release of drugs or active matter from theemulsion to the skin or to various “skins” such as membranes and cells [2502521. Most mechanisms treat the release in terms of Fick‘s second law of diffusion controlled process [250,253,254]. The effect of various nonionic emulsifiers [255-2581 on the release of various antimicrobial agents, various oils and various adjuvants on the rate of release was studied. The release seems to be diffusion controlled [250].

H. Perfluoro Emulsions for Artificial Blood

A formulation for artificial blood (total blood replacement) has been a goal for many investigations [259-2791. Various perfluoro emulsions were tried, including perlluorotributylamine, perfluorobutyltetrahydrofuran,


467

perfluoro decalin, perfluoromethyl decalin, and perfluoromethylcyclohexane. The emulsified fluorocarbon provides the necessary oxygenexchange, and the hydroxyethyl starch acts as a plasma expander. The emulsifier isthe polyoxyethylene-polyoxypropylene copolymer (Pluronic F68) [261]. Some encouragingresults were reported by Geyer [261] on complete replacement of blood in rats. The formation of a suitable perfluorochemical emulsion has been difficult, since those oils that provide stable emulsions are not cleared from the body in a satisfactory way and the choice of surfactants is extremely limited [262,263]. Long-term room temperatureshelf stability of ready-for-use concentrated fluorocarbon emulsions is necessary in order fully to exploit their rherapoic potential. The degradation mechanisms of the fluorocarbons, the factors that have impact on their stability and the means of proplonging their shelf life by using fluorinated surfactants or coemulsifiers, have seen carefully studied [259]. Generally, the fluorinated amphiphiles addmuch to the stability. Davis et al. [262,263] found that mixtures of lecithin and poloxamer were the best systems for emulsion stability. Of interest was the fact that the preferred oil, perfhorodecalin, although originally giving a very fine emulsion onemulsification, had poor stability. The fluorocarbon-emulsifying and emulsion-stabilizing properties of F-alkylated surfactants when used alone or in combination with Pluronic F68 or egg yolk phospholipids (EYP), the two surfactants used in the presently developed fluorocarbon emulsions, were investigated. The stability of these emulsions was evaluated by monitoring the increase in average particle sizes and changes in particle size distribution over time [264,265]. Generally speaking, F-alkylated amphiphiles aremuch more efficient as emulsifiers or coemulsifiers or coemulsifiers for fluorocarbons than their hydrocarbon analogues. Emulsions significantly more stable, than those obtained with egg yolkphospholipids (and a fortiori, Pluronic F68) can be obtained with several of the monodisperse F-alkylated amphiphiles even when used alone, which by itself is remarkable; mixtures of amphiphiles (which includes Pluronic and EYP) are indeed known to be considerably more efficient emulsifiers than any of their constituents taken separately. However, the exact prevision of the behavior of individual surfactants is still uncertain. Some structurally closely related surfactants were found to have widely different emulsion-stabilizing characteristics [266,267]. Examples of various distinct types of behavior aregiven below. Stable, medium concentrated (50% w/v; i.e., 27% v/v) to highly concentrated (up to 100% w/v; i.e., 52% v/v) fluorocarbon emulsions were

Garti and Aserin

468 0.8 0.7

4%

EYP

-

/

L

m

aJ

0.4

M

m

L

aJ

e

0.3

0.2 0.1

I 0

I

I

100

200

Days Fig. 32. Comparison of average particle sue vs time variation for sterilized Foctyl bromideemulsions (90%w/v) prepared(a) with 1% of sodium D-glucose 6-[2(F-octy1)ethyl phosphate]and, (b) with 4%of egg yolk phospholipides(EYP). (After ref. 269.)

obtained with several F-alkylated amphiphiles, including F-alkylated glucosyl-phosphates [268,269], trehalose ester[270], carnitine esters[271], some amino acids of type, amine oxides, phosphocholines, and phosphatidylcholines [272]. Fig. 32 compares, as an example, the average particle size increase over time for a 90% w/v concentrated perfluoro octyl bromide (PFOB) emulsion prepared with the F-alkylated glucosyl-phosphates and with EYF! It is noteworthy that even very low amounts of the fluorinated surfactant result in more stable emulsions than when four times larger amounts of EYP are used [268]. F-Alkylated fluorinated surfactant polymers were shown to be more effective emulsifiers than theirnonfluorinated parent compounds [273]. Long-term shelf lives of perfluordecalin (FDC) emulsions (20% w/v) were, however, not achieved. Somewhat better results were obtained with the F-alkylated Tris (hydroxymethyl) acrylaminomethane (THAM) telomers [274,275]. On the other hand, the anionic poly(ethyleneglyco1) phos-


469

phate and theneutral ones were foundto be less efficient emulsifiersthan Pluronic F68 [276]. Further improvement in fluorocarbon emulsion stability is expected when mixtures of surfactants are used, especially if they present complementary attributes. Therefore, F-alkylated amphiphiles havealso been investigated as cosurfactants with Pluronic F68, EYP, or other F-alkylated surfactants. Pluronic F68-type poloxamers are not very effective in reducing the interfacial tension between a fluorocarbon and water, but they exercise considerable steric stabilization. The addition of a fluorinated cosurfactant was expected to compensate forthis lack of effectiveness. Polyhydroxylated surfactants were chosenfor this purpose, because they could hydrogen bond Fto Pluronic. Considerable stabilization was indeedachievedwhen alkylated maltoside- or xylitol-derived surfactants were usedin conjunction with Pluronic F68 [277]. This synergistic stabilization effect is illustrated in Fig. 33 for a 50% w/v concentrated FDC emulsion preparedwith 5% w/v ofa mixture of 2-(F-octyl)ethyl maltoside and Pluronic F68. The stability of the emulsion, which is rather poorwhen Pluronic F68 is used alone, is significantly improved when 20430% of the' poloxamer is replaced by the Falkylated maltoside in the surfactant mixture. It should be noted that this Falkylated surfactant alone (right side of Fig. 33) is totally unable to stabilize the emulsion. The glucosyl-phosphates were found to act synergistically with EYP [268]. Therefore 90% w/v PFOB emulsions were prepared in which the EYP/F-(octy1)ethyl phosphate ratio was decreased stepwise, with the total amount of surfactant being held constant (Fig. 34). The replacement of only one eighth of the EYP by F-(octy1)ethyl phosphate was sufficient to obtain both a definite stabilizing effect and, interestingly, a decrease in particle size which, moreover, is better preserved during sterilization and ageing than in the emulsions preparedwith EYP alone. Synergistic stabilization was also observed when mixtures of two Falkylated phosphocholines were used in the formation of 100% w/v FDC emulsions [278]. These concentrated emulsions were significantly more stable than those formulated with EYP. Mixtures of F-alkylated THAM telomers of different length or of these telomers with the F-alkylated xylitol derivative also led to some improvementsin FDC or PFOB emulsion stability [275].

1. Biological Evaluation of F-Alkylated Amphiphiles (in Vivo Tests) The acute toxicity of F-alkylated surfactants has been evaluated by intravenous injection in mice of about 0.5 mL (i.e., 25 mWkgbody weight, caudal vein) of their solutions and/or dispersions. The reported maximum

Garti and Aserin

470

'

0.0 4

0

20

40

60

80

100 %

Maltoside / Pluronic F-68(w/w)

Fig. 33. Average particle sizes initially (0),after 1 month (O), after 1 year (m) at 25"C, in 50% w/v F-decalin emulsion asa function of a ratio of 2-(F-octyl)-ethyl maltoside to Pluronic 68 in thesurfactantmixture, with thetotalamount of surfactant being held constant(5% w/v). tolerated dose (MTD) compatible with the .survival of all injected mice (n=10) and/or lethal dose causing the deathof 50% of the mice (LDS0)are collected in tables [279]. The highest tolerance, by far, is observedfor the F-alkylated double-chain glycerophosphocholines. They display considerably higher intravenous tolerance than the correspondingsingle-chain analogues, with MTD values for compounds superior to 8000 mg/kgbodyweightcompared with 25-125 mgkg for compounds. Obviously, the nature of the hydrophilic head is also important. In the single-chain compound series, the following sequence of increasing tolerance was observed: phosphocholine phosphocholine mono-N-ethyl < amino acid < trehalose sucrose

-

-


471

n.xn T I 40°C

x % wlv of 81b

After 6.7 nrollths

T

IL!

L

After 3 m o n t h s

+

I

0.00 0

1

2

3

4

X%

Fig. 34. Averageparticlesizesin 90% perlluorooctyl bromide emulsions, prepared by microfluidation, asa function of the amount, x%, of sodium D-glucose6[2-(F-octyl)ethyl phosphate], with the remaining (4-x)% consisting of egg yolk phospholipides.

< maltose < xylitol < poly(ethy1ene-glycol) phosphate < THAM telomer. The high in vivo tolerance of the poly(ethyleneglyco1) phosphate derivatives is seen further to increase with increasing the size of the polar head and its hydrophilic character. These trends indicate that for similar hydrophobic tails, intravenous tolerance improves along the sequence: zwitterionic single-chain < neutral single-chain < zwitterionic double-chain compounds. It also appears that increased hydrophilic character (from neutral to zwitterionic or toanionic compounds) needs to bebalanced by an increase in hydrophobic character or by an increase of the size of its hydrophilic head. No obvious relation is, however, found between hydrophobic chain length an in vivo toxicitywhatever the hydrophilic head. , All in all, F-alkylated glycerophosphocholines, poly(ethy1ene-glycol) phosphates and THAM telomers, l-O-[3-(F-octyl)-Zpropenyl]xylitol,2(F-octy1)ethyl maltoside, and 6-0-[3'-(F-octyl)propanoyl]trehalose stand out as themost promising candidates for biomedical preparations.

Aserin 472

and

Garti

One of the primary goals for which many of the F-alkylated amphiphiles were synthesized was the obtaining of stable fluorocarbon emulsions to be used as injectable oxygen carriers and in particular as blood substitutes. Consequently, some of the new surfactants were incorporated in emulsions which were tested in isovolemic exchange-perfusion experiments Therein rats. This was the case for 1-0-[3’-(F-octyl)-2’-propenyl]xylitol. fore, adispersion of 0.5& of the octyl-xylitol in grams per literof Pluronic F68 was used to prepare a 10% (W/.) emulsion of lY2-bis(F-butyl)ethene (F-ME). This emulsion was tested on a series of 10 conscious rats whose red blood cell count was reduced to about one third of normal (hematocrit 15%) [280].The behavior of the animals during and after the exchange was comparable to thatof control animals, and the survival ratio was 10/10 after the 6-month observation period. This test was also applied to the mixed fluorocarbon-hydrocarbon “dowel” molecules. One of these compounds, C,Fl,CH=CHCl,H21, was administered intravenously inemulsifiedform at doses 300-600 times larger than those required to stabilize a clinically relevant dose of fluorocarbon emulsion. None of the 33 treated animals died. ”F-NMR (nuclear magnetic resonance) analysis of the organtissues indicated no metabolism; the halfof the compound in the liver (where fluorocarbon droplets are transiently stored) is about 25 5 days. In conclusion, compared with their hydrogenated analogues, fluorinated surfactants display significantly enhanced surface activity both in terms of effectiveness and efficiency and, in particular, strongly reduced water-fluorocarbon interfacial tensions. Fluorinated amphiphiles also show an increased tendency to form organized supramolecular assemblies, including liposomes, and impart unique properties to these assemblies. There is no doubt thattheir potentialas membrane components and modifiers should be further explored. Their fluorocarbon emulsion-stabilizing capacity, when they are used as surfactants andor as cosurfactants, has been demonstrated butremains unpredictable. The assessment of thebiological properties of fluorinated surfactants has, however, hardly begun. Preliminary evaluation indicates somewhat reduced acute toxicity and definitely lower hemolytic activity compared with their hydrogenated analogues. It is now necessary to establish their fate in the organism and to explore theirbiological effects both when alone and when incorporated in emulsions, liposomes, or other supramolecular assemblies. The influence of fluorinated surfactants on the recognition of such particulate matterin vivo deserves special attention. The stage now. is set forsuch investigations. Only few detailed studieshave been published so far concerning the biological evaluation of F-alkylated amphiphiles. Some attention has, how-

*


473

ever, been given to F-alkanoic acids because of their many industrial applications related to their thermal and chemical stability and to their antiwetting and surface activity. Some simple biological tests (in vitro action on cell cultures and on red blood cells) and in vivo tests (acute toxicity in mice after intravenous injection) havebeenreported on a range of other F-alkylated amphiphiles. In a few cases, isovolemic blood exchange-perfusion experiments in rats with emulsions containing such surfactants have been performed. Concerning the F-alkanoic acids, these were found to be toxic with, for example, intraperitoneal LD, (rat) values of 189 and 41 mgkg for Foctanoic and F-decanoicacid, respectively. The major symptoms observed include anorexia, marked bodyweight loss, andhepatomegaly with a concomitant increase in peroxisomesand swelling. But the fluorinated metabolites of these F-acids have not been identified yet. It was reported that F-alkanoic acids covalenty bind to proteins both in vivo and in vitro, but the relevance of this binding to toxicity remains unknown. An increased incidence of Leydig cell adenomas was observed in rats fed with ammonium F-octanoate (300 ppm daily) for 2 years; an endocrine-related mechanism was proposed. The use of diverse F-alkanoic acids was nevertheless deemed acceptable for stabilizing micronized inhalation drug suspensions in 1,1,1,2-tetrafluoroethane7one of the CFC substitutes presently being developed. Studies on F-glucopyranosides also indicate the existence of toxicity due torapid hydrolysis and cleavage. These studies indicate that it is preferable to avoid the direct connection of a F-alkyl chain on a metabolic site which would then give either F-alkanoic acids or unstable F-alcohols. It should be preferable to insert a hydrocarbon spacer between the F-alkyl chain and the possible metabolic sites of the polar head.

2. Effect on the Growth and Viability of Cell Cultures (in Vitro Tests) The growth and viability of cells in the presence of the fluorinated surfactants tested are compared with those of control cultures grown under the same conditions without the surfactant. Cell culture tests were used both to assess direct cell toxicity and as a means of monitoring the final purification steps. Different cell strains were used, including lymphocytes, Hela, Molt-4, hybridoma, K562, or Namalva lymphoblastoid cells. No significant toxic effect of the Namalva cell's growth and viability anionic phosphate esters, the was found at a 1 g L concentration for the three 2-(F-octyl)ethyl D-maltoside, the 6-0-[5'-(F-hexyl)pentanoyl]trehalose, and for the 6-O-[ll'-(F-hexyl)undecanoyl]trehalose in spite of their high surface activity.

474

Garti and Aserin

3. HemolyticActivity Unexpectedly, avery drastic difference in hemolytic potency was found betweensurfactants with hydrocarbon chains and with fluorocarbon chains of similar length. The hemolytic activity is consistently and considerably lower and often totally suppressed for the F-alkylated surfactants as compared with their hydrocarbon analogues. Moreover, the hemolytic activity is seen to decrease with increasing length of the F-alkyl chain, hence with increasing surface activity. The protective effect of the F-alkyl segment against hemolysis takes place despite the definitely higher surface activity if the F-alkylated compounds. The length of the hydrocarbon spacer, or moreexactly its relative length compared with that of the fluorocarbon segment, has also a decisive impact on hemolysis. Thus, the 11-(F-hexyl)-10-undecenyl maltoside, with a long CH==CH(CH,), hydrocarbon spacer, is hemolytic at a lowconcentration (1 g L ) , whereas the 5-(F-hexyl)-4-pentenyl maltoside, with a shorter CH=CH(CH,), spacer, is not hemolytic at 50 g/L. Identical results are found in the carnitine and phosphocholine series, where the compounds witha (CH,),, spacer are significantly more hemolytic than those with shorter spacers. The natureof the polar head also influences the hemolytic potencyof fluorinated surfactants. Thus, for the same hydrophobic(F-hexy1)allyl tail, the maltose derivative issignificantlyless hemolytic than the galactose analogue, which is less hemolytic than that of xylitol. This appears to indicate that hemolytic potency is lower for those surfactants with the larger, more hydrophilic heads: It is also noteworthy in this respect, and somewhat unexpected, that the anionic glucose-derived phosphate esters are not more hemolytic than their nonionic F-alkylated sugar or glycoside counterparts. Others have described the partial and total replacement of blood or the use of perfluorochemicals in emulsion form fororgan perfusion [283-2871.

In conclusion, the most recent work in the area of emulsions for pharmaceutical applications is related to practical improvements of the size of droplets with regard to their toxicity and coalescence and polydispersity [288]. Emulsions are used andcurrently evaluated for parenteraldelivery of nutrients and drug substances [289-2911. Some of the commercial total parenteral nutrition (TPN)emulsionsare Emulsan, Intralipid, Lyposyn, and Lipovenos. All of these emulsions are stabilized with egg lecithin and are based on soybean oil except Liposyn in which amixture of soybean oil and safflower oil is used. Other TPN emulsions contain cottonsed oil and soybean phosphatides[292,293]. The droplet size and compositionof the emul-


475

sion droplets are reminiscent of the endogenously occurring chylomicrons. Chylomicrons are fat globules, 0.5-1.0 pm in size, composed of triglycerides, proteins, cholesterol, and phospholipids. The chylomicrons are generated by the liver and the intestines. The relatively low toxicity'of TPN emulsions has made them an attractive alternative as carriers forlipophilic drugs [294,295]. In addition, much work has been done onusing emulsion forms in the delivery of vaccines [13]. The main efforts are relatedto artificial blood substitute [296]. The advantages of using the emulsion form in the delivery of drugs are (1) no precipitation of the drug substance will occur upon dilution [297], (2) drug stability may be improved, that is, hydrolysis will be reduced, and (3) a slow release of drug may be obtained. There is a risk that the emulsion will become unstable on injection into the blood stream through, if the emulsifier is too water solubleor if it sensitive to increased electrolyte concentration. There is much scientific literature describing emulsions as a drugdelivery system [297-3011. Examples of commercially occurring products in which emulsions are used for parenteraldelivery of active substances are Diprivan (propofolum), Stesolid Novum (diazepam), and Vitalipid Adult and Vitalipid Infant (vitamins). XII. A.

DoubleEmulsions DefinitionsandFormation

Multiple emulsions are complex systems, termed emulsions of emulsions, with the dropletsof the dispersed phase themselves containing even smaller dispersed droplets. Each dispersed globule in the double emulsion forms a vesicular structure with singleor multiple aqueous compartments separated from the aqueous phase by a layer of oil phase compartments [302-3081. Multiple emulsions were first described by Seifriz [306] in 1925,but it was only in the past 20 years that they have been studied in more detail. The two major types of multiple emulsions are the water-in-oil (WIOW) and oil-water-oil (O/W/O) double emulsions. A schematic representation of a W/O/W double-emulsion droplet is shown in Fig. 1. Multiple emulsions have been prepared in two main modes: one-step emulsification and two-step emulsification.There have been several reports on the one-stepemulsification for the preparationof W/O/W double emulsions which included strong mechanical agitation of the water phase containing an hydrophilic emulsifier and an oil phase containing large amountsof hydrophobic surfactant. A W/O emulsion is formed, but it tends to invert and form a W/O/W double emulsion [306] (Fig. 35). In addition, double

476

Garti and Aserin

Fig. 35. Possible sequence of events leading to the final formation of an OiW emulsion via a transient (WlOiW)emulsion, when hydrophilic surfactant is initially located in the oil phase.

emulsions can be prepared by forming W/O emulsion with a large excess of relatively hydrophobic emulsifier and small amount of hydrophilic emulsifier followed by heat treatingthe emulsion until, at least in part, it willinvert. At a proper temperature, and with the right HLB of the emulsifiers, W/O/W emulsion can be found in the system. However, there is usuallylittle chance of reproducing these “accidental” preparations. In most of the recent studies, doubleemulsions are preparedin a twostep emulsificationprocess by twosets of emulsifiers: a hydrophobic emulsifier I (for thewater-in-oil emulsion) and ahydrophilic emulsifier I1 (for the oil-in-water emulsion) (Fig. 36). The primary W/O emulsion is prepared under high shear conditions (ultrasonification, homogenization), whereas the secondary emulsification step is carried out without any severe mixing (an excess of mixing can rupture the dropsresulting in a simple oil-in-water emulsion). The composition of the multiple emulsion is of significant importance, since the different surfactants along with the nature and concentration of the oil phase willaffect the stability of thedouble emulsion [303,308-3101. Much work was done on the natureof the oils and their influence on the manufacturing conditions, as well as on the stability of the double emulsion [309]. Ionic and nonionic surfactants have been used for different applications in accordance with health restrictions. It was, however, well established that combinations of emulsifiers at the outerphase have a beneficial effect on stability, and that the inner hydrophobic emulsifier I must be used


hydrophilic surfactant

477

w/o/w multiple emulsion

Fig.36. Schematicillustration of a two-stepprocessinformation emulsion.

of double

in great excess (10-30 wt% of the inneremulsion), whereas the hydrophilic emulsifier I1 must be used in low concentration (0.5-5.0 wt%). The inner emulsifier was found to migrate in part to the outerinterface and influence the outer emulsifiers (Fig. 37) [310-3121. The HLB of the outer'emulsion was found to be a weighted HLB of the contribution of the two types of emulsifiers [313,314]. In addition, parameters such as the oil phase volume and the nature of the entrapped materials in the inner phase have been discussed and optimized [310-3131. B. Stability and Transport Mechanisms

Many review articles have been written on the potential practical applications of the multiple emulsions and on the main problems associated with this technology-their inherent thermodynamic instability [304,308,309]. It was concluded that the classic double emulsions prepared with two sets of monomeric emulsifiers; one hydrophobic in nature (to stabilize the inner W/O interface) and the other hydrophilic in nature (to stabilize the

Garti and Aserin

478 100

E

@

Tween 80 and POE-oleyl ether POE-lauryl ether

, , , 8 8

+

Tween 20

x

POE-nonylphenyl ether

POE l

1

I

I

I

Illll

10

I

and ester

I

- 1

I

polyoxyethylene I 1 1 1

100

Weight ratio of Span 80 to hydrophilicemulsifier

\

Fig. 37. Weight ratio of Span 80 in the oilphase to hydrophilic emulsifiers inthe aqueoussuspendingfluidaffectingtheformation of water-liquid paraffin-water emulsions due to thetwo separated stepsof emulsification.

outer O N interface) cannot provide long-term stability to thedouble emulsion. A s a result, relatively large W / O N droplets with short-term stability are obtained,which cannot be used in practice [315]. The complex nature of the double emulsions has caused significant difficulties in the assessment of the stability and in detecting rupture and coalescence phenomena. The main technique is based on measurement of number and size of the double-emulsion drops over a period of time. Such measurements produce ,only limited information on double-emulsion stability, since no information on the coalescence of the inner droplets can be deduced. Similar information is obtained from photomicrography (see Fig. 2). It was and still is very difficult to determine if the internal droplets tend to aggregate or to coalesce. Several more sophisticated techniques have been applied, including freeze-etching, viscosity measurements, and quantitative estimation of addenda transportedfrom the inner phase to the outer phase and vise versa. In addition, engulfment or shrinkage of the double emulsion in the presence of water migration in or out of the droplets has been studied and interpreted in terms of stability (Fig. 38). Several possible mechanisms by which materials may be transported across the oil layer in the W/O/W systems have been proposed and dis-


479

Fig. 38. Possible pathways for the transportof water and entrapped matter from the inner interface tothe outer aqueousphase.

cussed [303,307,310]. The most common is the diffusion-controlled mechanism for ionized lipid-soluble materials. It will be dependent onthe nature of the material (including its dissociation constant) andthe oil, as well as on the pH of the aqueous phase. This system could be used, for example, in the treatment of overdosage of acidic drugs such as aspirin and barbiturates [316]. At low pH values, the barbiturate would exist almost exclusively as the un-ionized form and so would be readily soluble in the oil phase. The drug could therefore easily pass across the oil layer to theinternal aqueous phase containing a basic buffer which wouldionize the addenda thatis now insoluble in the oil phase and would become trapped within the internal aqueous phase. This would then be carried out with the emulsion as it is voided from the gastrointestinal tract (Fig. 39). The drug transport was found to follow the first-order kinetics according to Fick’s law [316]. Ionized compounds arenotthe only materials to be transported across the oil membrane. It has been demonstrated that both watermolecules and nonelectrolytes can easily migrate through the oil membrane without affecting the double-emulsion stability [317-3191. A mechanism based on “micellar transport” from one phase to the other has been described and determined. It was also demonstrated that one can alter the diffusion rates through the oil membrane by changing the natureof the oil. This suggests that the diffusion of the addenda through the oil is the ratedetermining step, and that the inner water phase does not have any effect on the determination of release rates. Kita, et al. [20] have suggested two

480

Garti and Aserin

Oil (liquid membrane) phase ion i zed dr uq

basic

buffer

Un-ionized drug

Fig. 39. W/O/W “liquid membrane” system for removalof acidic drugs form an aqueous system.

possible mechanisms for the permeation of water and water-soluble materials through the oil phase; the first being via the reverse micellar transport (Fig. 40) and the second by diffusion across a very thin lamella of surfactant formed where the oil layer is very thin (Fig. 41). Our detailed studies on double emulsions stabilized by monomeric nonionic emulsifiers (Span and R e e n ) [317-3191 on the releaseof electrolytes from the inner aqueous phase to the outeraqueous phase have indicated that theosmotic pressure gradient between the two aqueousphases is a strongdriving force for mutual migration of addenda and water fromone phase through the other by both mechanisms. However, it was clearly demonstrated that even if the osmotic pressure of the two phases was equilibrated and no visual coalescence took place (neither of the inner phase droplets nor of the outer phase droplets), electrolytes tend to be transported outmostly through a reverse micellar mechanism controlled by the viscosity and the natureof the oil membrane. The mechanism is similar to the one described by Higuchi for release from polymeric matrices“slab into the sink” (Fig. 42) [320-3221.


48 1

Fig. 40. Schematic illustration of a model for micellar transport of water from the outer aqueous phase to the inner aqueous phase through the oil layer inWIOIW emulsion.

A modified diffusion-controlled equation was adapted to explain the release results [317-3191. It has been demonstrated that the release factor B,

3

3 2[1 - (1 - F)3n]- F

B =- De-t r,”C0 (where F is the release fraction of the marker soluble in the inner phase; De, effective diffusion coefficient;ro,radius of the droplets;t, time of release; C,, initial concentrationof marker) can be plotted against l/Co and t, with excellent correlationcoefficients suggesting“diffusion controlled releasemechanism of reverse micellar transport” [318,319]. The “effective” as well as the “real” diffusion coefficient couldtherefore be calculated. Although the release mechanism was clarified to some extent, it is still very difficult to control the rate of release (to slow it down) mainly B=

’

wafer

Fig. 41. Schematicillustration of amodelforwatertransportthroughthis lamellae of surfactant due to fluctuation in the thicknessof the oil layer.

Garti and Aserin

482

multiple drop

Fig. 42. -0-dimensional model for diffusion-controlled transport of un-ionized materials from the external aqueous phaseto the internal aqueous droplets in W/ a O W system. owing to thefact that the monomericemulsifiers that have been used serve both as stabilizing moieties and as transport species. In addition, the fast exchange that those emulsifiers will undergo between the two interfaces and the fact that shearshould be avoided in the second emulsification stage lead to theformation of relatively large double-emulsion droplets with very limited thermodynamic stability.

C. New Approaches to Improve Stability Fig. 43 demonstrates in part the different processes that can lead to the destabilization of the double emulsion and the processes involved in the release of various molecules (water or entrapped matter) from the inner phase to the outer aqueous phase. One can think of several approaches to overcome instability and release problems in double emulsions. Some of those ideas can be summerized as follows: 1. Stabilizing the inner W/O emulsion by reducing its droplet size

and by forming h-microemulsions ormicrospheres or by increasing the viscosity of the innerwater.


483

t

External drop coalescence

growth

/ -

Empty primary contlnuous phase drop

Fig. 43. Various possible mechanisms of emulsion stabilization.

2. Modifying the nature of the oil phase by increasing its viscosity or by adding carriers or complexants to theoil. 3. Stabilizing the inner and/or the outer emulsion by using polymeric emulsifiers, macromolecular amphiphiles (protiens, polysaccharides), or colloidal solid particles to form strong and more rigid film at the interface. One can consider both naturally occurring macromolecules (gums, proteins) or synthetic grafted block copolymers with surface activities. It should be noted, although it is beyond the scope of this text, that polymerizable nonionic surfactants can form an “in situ” crosslinked membranes (after adsorption and carrying out a polymerization process). This concept was well documented and tested [323-3251. The polymeric complex that was formed was able to withstand extensive thinning (caused by osmotic driven influx of water) and the resulting swelling of the internal water droplets [326].

Aserin 484

and

Garti

D. Naturally Occurring Macromolecular Amphiphiles

The stability of double emulsions can be improved (as explained above) by forming a polymeric film or macromolecular complexacross the oil-water interfaces. Omotosho and Florence have suggested using macromolecules and nonionic surfactants to form such stabilizing complexes. The film is formed throughan interfacial interaction between macromolecules such as albumin andnonionic surfactants [327,328]. Release rates of methotrexate (MTX) encapsulated in the internal phase of W/O/W emulsions stabilized by a film, formed as a result of an interfacial interaction between albumin and sorbitan monooleate (Span80), were measured as functions of two formulation variables-the oil phase and the secondary emulsifier composition. The release rate was significantly affected by the natureof the oil phase and decreasedin the orderof: isopropyl myristate > octadecane > hexadecane > dodecane > octane which was a reflection of the increasing internal droplet size of the emulsion (Fig. 44). The release rate data conformswith first-order kinetics. Comparison of the effective permeability coefficients-calculated from the experimental

80

70

IPM

60

octadecane

S-” 50

hexadecane

h

v

cd

2

dodecane

40

W

octane

30 20 10 0 0

1

2

3

4

5

6

7

Time (hours) Fig. 44. Release of MTX from multiple W/O/W emulsions. The emulsions contained 2.5% Span 80 and 0.2% BSA as primary emulsifiers, with MTX (1 mg/mL) in the internal phase andthe following oil phases: *, octane: A, dodecane; +, hexadecane; 0 ,octadecane; and a, isopropylmyristate.


485

apparent first-order rate constants-with the effective permeability coefficient of water in planar oil layers containing nonionic surfactants (determined by a microgravimetric method) supportedthe hypothesis of diffusion of MTX via loaded inverse micelles. Surfactants with highHLB values, used as the secondary hydrophilic emulsifier, increased the release rates, primarily byincreasing the rateof diffusion of MTX through the nonaqueous liquid membrane (Fig. 45). Omotosho and Florence have also reported use of other macromolecule complexes [327,328]. The formation of multiple W/O/W emulsions with improved stability due to the formation of interfacial complex film between acacia, gelatin, polyvinyl pyrrolidone (PVP), and sorbitan monooleate have been described [327,328]. The long-term stability as assessed by microscopy, showed no significant time changes in droplet size distribution of W/O/W emulsions prepared with acacia, gelatin, or PVP in the internal phase (Fig. 46). Multiple emulsions containing chloroquine phosphate in the internal phase that had been stored for2 weeks surprisingly showed a reduced rate of release of chloroquine phosphate as compared with freshly prepared emulsions, suggesting the release of chloroquine phosphate from these systems occurs by the process of diffusion as opposed to thephysical breakdown of emulsions [328] (Fig. 46). The intramuscular administration of

L

, 1

1

*l+ 3

S 7 Time (hrs)

24

Fig. 45. Effect of the secondaryemulsifier(emulsifier 11) on the release of MTX from multiple WIOTW emulsions prepared with 2.5% Span 80 and 0.2% BSA as primary emulsifiers with isoprophymyristateas the oil phase.

Aserin 486

and

L,

1 0

2

Garti

..W" A BI

I

I 4

I

6

Time (hours) Fig. 46. Profile of chloroquine phosphate release from WIONV multiple emulsions. A,, freshly prepared PVP emulsions;A,, PVP emulsions stored for2 weeks; B,, freshly prepared gelatin emulsions;B,, gelatin emulsions stored for 2 weeks; C,, emulsion prepared with gum acacia;G, acacia emulsion storedfor 2 weeks. chloroquine in the form of W/O/W emulsions has a significant advantage, since it could reduce frequency of administration, improve patient compliance, andincrease the therapeuticefficacy of chloroquine. The drug can be formulated as a combination of two types of doses: starting dose which is incorporated into the external phase and a maintenance dose which is encapsulated in the internal phase of the doubleemulsion. Wasan et al. [329,330] have published a novel method for forming stable hemoglobin-oil-in-water (Hb/O/W) multiple emulsion for use as an artificial red cell substitute. A concentrated Hb solution was emulsified in oil to form microdroplets (with Pluronic F101 and Span 80) followed by dispersion of the primary emulsion into an outer aqueous phase containing hydrophilic surfactants (Pluronic F68 or Tween SO). The addition of human serum albumin into both the inner and the outerphases along with added dextran into the outer phase seems to be stabilizing the emulsion. The average diameter of the prepared multiple emulsions after homogenization and filtration was 2-3 pm with good hydrodynamic stability (sensitivity to shear). The formulation showed a very small release of Hb from the primary emulsion to the outer aqueous and good stability of the multiple emulsion during short-term storage (Fig. 47). Frank et al. [331,332] have suggested the use of colloidal microcrystalline cellulose (CMCC) and various monomeric surfactants to stabi-

Emulsions and Microemulsions B

100 -I U W

5

D

W

-!

80 70 50

0 [I:

n

U.

0

8

0

+

B 0

+

60-

U)

a

D

m

90

U

2

487

-

40-

20 -

B

30

10

-

0

0

B 0

Without Albumln

.L

o

~

Q

2

Wilh Albumln

~

4

l

6

~

8

l

10

12

'

i

14

16

~

18

l

~

l

20

DROPLET DIAMETER (microns)

Fig. 47. Effect of albumin on size distribution of multiple emulsion.

lize double emulsions via a mechanical stabilization mechanism. It was shown that the double emulsions were stable over a period of 1 month (monitored by microscopy). Slow release of certain drugs was achieved by gelling the oil phase (Table 10). Recently [302], the authors of this chapter have usedBSA as a polymeric emulsifier added both to the inner and to the outerinterfaces in the absence and in the presence of conventional monomeric emulsifiers (such as Span 80 and Tween 80). We have tried to evaluate in more detail the mechanism of the release of an electrolyte, NaCl, from the inner phase to the outerphase. Since it was believed that therelease mechanism is associated with a micellar transport which is diffusion controlled, attempts were made to obtain stable double emulsionswith relatively small droplets anda minimum oil micellization capacity. Double emulsions were prepared with 10 wt% Span 80, 0-0.5 wt% BSA, and 2 wt% NaCl inthe inner aqueous phase. The outerinterface was stabilized with 5 wt% of Span 80-Tween 80. It should be noted thatafter 25 h of aging, the emulsion oil droplets size did not change significantly. The percentage release plot of NaCl versus wt% of the inner BSA reveals that BSA retards the release of NaCl (Fig. 48). Its maximum effect was obtained at 0.2 wt% BSA. It seems that BSA adsorbs at theinterface together with Span 80 and contributes both to the stability of the double emulsion and to therelease rates.

~

l

Garti and Aserin

488

Table 10. Comparison of Release of Lidocaine Base From Different Formulations % Lidocaine Effective incorpobase diffusion rated within coefficient Time for Formulation formulation Dx106 (cm2/s) release (h) after

CMC"

2% 25.4 1 dispersion 1

W/O/Wb

- 6.9

0.8

Innermost

0.03 4.9

30%

Release 5 h (%)

186

aqueous (4:6/1:1), phase CMCC 2% *Release of lidocaine from a 2 wt% CMCC dispersion containing 1 wt% lidocaine base. bRelease of lidocaine from W/O/W a (4:6/1:1) emulsion containing1wt% lidocaine base (25+O.l0C).

0

0.1

02

03

0.4

05

BSA (wt %) in inner phase

Fig. 48. PercentagereleasewithtimeofNaClfromdoubleemulsionprepared with 10 wt% Span 80 and various BSA concentrations in the inner phase and 5 wt% Span + Tween (1:9) in the outer aqueous phase.

Emulsions

489

In accordance with the previous studies, it was suggested that the polymeric surfactant forms a complex with the monomeric lipophilic surfactant. The complex is probably a thick, strong gelled film that imparts elasticity and resistance to ruptureof the innerdroplets. The film improves the mechanical and steric stability of the double emulsions and slows the coalescence rates. In addition,it appears that it depresses the formation of reverse micelles in the oil phase and slows the transport of the electrolyte via the reverse micelle or mechanism. However, the monomeric emulsifiers covering the outer interface do not prevent the coalescence of the outer droplets in the double emulsions. In fact, after 1 week of aging, a significant increase in droplet size distribution, as well as strongflocculation, was detected. Therefore, BSA was also added to the outer aqueous phase (during the second step of the emulsification) in addition to themonomeric Span 80-%een 80. The Coulter counter measurements (Fig. 49) clearly indicate that BSA when present in the outer phase contributes.effectively

30

r 0

CUIWE 1 CURVE2

0

10 2 03 0

U

CURVE3

o

CURVE4

4 0 5 0 6 0 70 8 0

90 100

DIAMETER Cm)

Fig. 49. Droplet size distribution of four emulsions prepared with(1) Span 80 + Tween 80 (9:l) in the inner phase and 0.2 wt% BSA in the outer aqueous phase;(2) with Span 80 in the inner phase and withoutBSA in the outer phase;(3) Span 80 in the inner phase and withoutBSA in the outer phase;(4) Span 80 in the inner phase and without BSA in the outer phase.

Aserin 490

and

Garti

to the stability of the double emulsions. Double emulsions prepared with BSA (in the outerinterface) consisted of significantly smaller droplets than any other emulsions preparedwithout the BSA. The improved dropletsize distribution was found for any BSA emulsion prepared with any level of monomeric emulsifiers. The photomicrographs and theCoulter counter measurementsof the droplets size distribution after 6 weeks of aging showpractically no change in droplets size distribution. The release rates as a function of the BSA concentration in the outer phase (emulsifier 11) show a minimum at 0.2 wt% BSA and a slight increase in the release rates at higher BSA concentration (Fig. 50). The Higuchi model for therelease of drug from solid polymeric matrix [321] as well as other models such Garti's as modification [318,319] for multiple emulsions (based on Fick's diffusion) have been used for testing the double emulsions prepared with BSA in the innerphase. The B parameter (as described previously) was plotted against time (t) in order to determine the effective diffusion coefficient, De. Figures 51 and 52 are typical plots of the parameter B versus l/Coof entrapped NaCl and B versus the time of aging of the emulsion. The De values for each BSA concentration calculated from plot B against l/Co and against t"(n=l) arelinear and quite similar, indicating that

h

E

5.

20

v

I(0)

W

5.

t ( l week)

3

-J

B 8

U

t(6 weeks)

10

0 0

25

50

75

100

125

DIAMETER (pm) Fig. 50. Percentage release with time of NaCl from double emulsion prepared with 10 W% Span 80 in the inner phase and 0.1 wt% BSA in the outer phase.


491

Time (hrs)

Fig. 51. Factor B (see text) plotted against timeof release for aouble emulsion stabilized with 10wt% Span 80 + 0.2 wt% BSA in the inner phase and5 wt% of Span 80-Tiveen 80(9:l) in the outer phase.

0

50

100

150

200

/co(wt %l” of NaCl Fig. 52. Same as Fig. 19 except plotting factor B against l/C,of the NaCl concentration in the inner aqueous phase.

Gatti and Aserin

492

the Higuchi model dominates the release mechanism. However, better correlation coefficients (R,=0.998-1.0) will be obtained if the B parameter is plotted against t”, where n (termed arbitrarily as the “diffusion order”) varies from 0.5 to 3.0 as a function of the BSA concentration (Table 11). Plots of parameter B versus to for double emulsions stabilized with BSA in the outerphase showed similar trends of linearity. Best correlation coefficients for linearity of the curves were obtained fort‘ in which n was in the range of 0.5-1.0 for 0-0.5 wt% BSA in W, (see Table 11). The differences between the functionality and performance of the BSA present in the inner(W,) or the outer(W,) interface can be seen from the plot of the time exponent n (the “diffusion order”) versus the BSA concentration (Fig. 53). The effective diffusion coefficients (De) were calculated from the corrected Higuchi equation and plotted versus the BSA concentrations in the inner and outer phases (Fig. 54). Significant differences in the performance of BSA have been found. The outer BSA has only a limited retarding effect on the electrolyte transport, limited to concentration at 0-0.1 wt%, whereas the inner BSA (at 0.02 wt% levels) has a strong slowing effect on the releaseof NaC1. It has

Table 11. n Values (“Diffusion Order”) Obtained from Plotsof B Factor Corresponding to the Percentageof Release of NaCl versus t” from Double Emulsions Prepared with BSA in the Inner Phase(W,) and in the Outer Phase (W,) (see text) BSA concentrationsin W, or W, (wt%)

%la

~Wzb

0 0.05 0.1 0.2 0.3 0.5

0.5 1.25 2.0 2.0 2.0 3.0

0.5 0.5 0.5 0.5 1.0 1.0 ~

anwlis calculated from the best fit for linearity of the plotof B against t” for double emulsions stabilized with BSA added to the inner water phase (W,) bnwZ is calculated from the best fit for linearity of the plotof factor B against t” of double emulsions stabilizedwith BSA added to the outer water phase (W,).


493

n(BSA in Wl) n(BSA in W2)

0.1

0.0

0.2

0.3

0.4

0.5

0.6

CONCENTRATION of BSA (wt%) Fig. 53. Plot of exponent n of time (t") againstBSA concentration in both inner

(0) and outer (

) interfaces.

I

-16.0 'p

6)

R

B

U

M

log(De)(BSA

in W 4

Lag(De)(BSA ih WZ)

0

%4

0.0

0.1

0.2

0.3

0.4

0.5

0.6

CONCENTIWTION ofBSA (@h] Fig. 54. Plot of log De (effective diffusion coefficient) of NaCl vs BSA concentration in the inner and the outer interfaces.

494

Garti and Aserin

Fig. 55. Schematic illustration of the two interfaces of the double emulsion stabilized by a combinationof monomeric and polymeric (BSA) emulsifiers. been assumed that the parameter nis a reflection of the nature of the film that is formed on the interface. When n s l , the film is rather thin and no significant viscoelastic gelfilm (complex between the Span and BSA) is formed.When n z l , the film is viscous owingto theformation of a strong Span-BSAcomplex. From Fig. 53 it can be concluded that the internal W/O film (W, interface) is more pronounced and stronger than the O N (W, interface) film. The W, film develops as the BSA concentration increases andis well defined at 0.1-0.2 wt% BSA. On the other hand, the BSA in the outer phase (W,) does not contribute to the formation of a viscoelastic network with the hydrophilic combination of Span-Tween and has only a limited effect on thediffusion coefficient. of the Fig. 55 is a schematic illustration of theorganization monomeric and polymeric emulsifiers onto the inner (W,) and the outer (W,) interfaces. Note thatthe BSAis coadsorbedtogether with the monomeric emulsifier at the inner interfaceand serves only as protective colloid at the outerinterface. From a careful evaluation of the release and stability results, it is possible to formulate an optimal double emulsion consisting of Span 80BSA in the outer interface and Span 80-BSA in the inner interface. XIII. DOUBLE EMULSIONAPPLICATIONS Double emulsions are excellent and exciting potential systems for slow or controlled release of active entrapped compounds. The fact that the inner W/O emulsion serves as large confined reservoir of water is very attractive

croemulsions Emulsions and

495

property fordissolving init significant amounts of water-soluble drugs. The oil membrane seems to serve as good transport barrier for the confined ionized and/or nonionized water-soluble drugs. The two amphiphilic interfaces are yet an additional barrier. The possibility to manipulate transport and release characteristics of the formulations seems to be feasible. However in spite of20 years of research with exiting in vitro release results [334-3661, no pharmaceuticalpreparation using the double-emulsiontechnology (neither for intravenous nor intramuscular administration, nor for oral or topical applications) exist inthe marketplace. It seems that themain reasons are thedroplets instability (shelf life)and theuncontrolled release. We therefore present a short review summary of the studies mentioning the use of various drugs for different applications and the potential advantages that such formulation might have. Multiple emulsions haveshown significantpromise in many technologies, particularly in pharmacology and separation science. Their potential biopharmaceutical applications, as a consequence of the dispersal of one phase inside droplets of another, include uses such as adjuvant vaccines [334,335], prolonged drug-delivery systems [336-3411, sorbent reservoirs of drug overdosage treatments[342,315], taste masking [313,314], and immobilization of enzymes [340,341]. In some disciplines, certain multiple emulsions have been termed “liquid membranes,” as the liquid film which separates one liquid phase from the other liquid phases acts as a thin semipermeable film through which solute must diffise as moves it from one phase to another. Theuse of multiple emulsions in the separation field has included, for example, separation of hydrocarbons andthe removal of toxic materials from waste water [342,345]. The following systemshave beenrecently studied and considered for pharmaceutical andmedical uses:

1. Hemoglobin multiple emulsion as a red blood cell substitute [330,346]. 2. The use of thickening agents and emulsifying agents to obtain slow release (retention to pharmaceutical dosage) of mebeverine from double emulsions[347]. 3. Prolonged release of pentazocine from O/W/Odouble emulsions [348]. 4. Sustained release of lidocaine from double emulsions stabilized [113,3491. An interesting compariwith microcrystalline cellulose son among O/W, W/O, and W / O N systems for the targeting of labeled 5-fluorouracil to lymph nodes after intratesticular administration is shown in Fig. 56. The profound response achieved

496

Garti and Aserin

0.5

1

2

I 6

I 12

I 24

Time alter administration (hr)

Fig. 56. Uptake of 5-fluorouracil into regional lymph nodes after intratesticular injection into rats. W, aqueous solution; O N , oil-in-water emulsion; W/O, waterin-oil emulsion; WlOnV, multipleemulsion; IV, intravenousinjection(control). (After ref. 349.)

5.

6.

7. 8.

with the multiple emulsion is striking. Unfortunately, the reason for such an effect has not been elucidated as yet [350]. Sustained release of fluorouracil for intramuscular administration [350,351]. The use of albumin and monomeric nonionic emulsifier to control the transport of methotrexate from the internal phase of multiple emulsion to the external aqueousphase [328]. Effect of encapsulated insulin in double emulsion on the blood glucose efficiencyof experimental diabeticmice by oral administration [352]. Use of double emulsions for controlled release of topical drugs [353].


497

9. Bioavailability and adsorption of pentazocine entrapped in double emulsion (oral intake). Its slow release effect on kidney, liver, and lung metabolism [354,355]. 10. Release of terbutaline sulfate from double emulsions prepared with paraffin oils and peanut oils, effect of pH, temperature, agitation, typeof oil on the release rates[355]. 11. Effect of acacia, gelatin, and polyvinylpyrrolidone on the chloroquine phosphase transport from W/O/W double emulsion [356]. Formation of interfacial complex films of the polymers and the surfactants and its effect on the transport rates (see Fig. 46). Reduced rate of release of chloroquine was observed. 12. Adsorption and lymphatic uptake of 5-fluorouracil in rats following oral administration of W/O/W double emulsions [350]. 13. The release of ephedrine hydrochloride from W/O/W emulsions. Efficiency of formation and rates of release as a function composition, mixing time, emulsifier concentration, and the HLB of oil and the emulsifiers. It was found that there is a linear relationship between the amount of ephedrine hydrochloride released and the square rootof the dialysis time of the double emulsions [357]. 14. In vivo releaseof 2.5% Xylocaine (lidocaine) hydrochloride from simple and multiple emulsion systems was compared with that from aqueous and micellar solution and anesthetic effects such as duration of action and tolerability were compared. The double emulsions showed a longer duration of action, less eye irritation, and improved efficacy compared with aqueous solutions [358]. 15. The use of double emulsions as useful vehicles for the administration of vaccines that gave rise to an improvement in antibody titer [359-3641. 16. In vitro releaseand pharmacokinetic and tissue distribution studied of doxorubicin hydrochloride (Adriamycin HCl) encapsulated in lipidolized W/O emulsions and W/O/W multiple emulsions. The results indicate that the release was sustained for both emulsions when HCO-60 (Hydrogenated CastorOil ethoxylated with 60 EO units) was used as emulsifier.The clearance of some W/O/W emulsions also decreased with the increaseof the emulsifier concentration (Figs. 57 and 58) [365]. 17. Study of the release of cytarabine and fluorouracil (5-FU) from double emulsion. It was demonstrated that the release of cytarabine (Fig. 59) was very prolonged and affected by the increase of theinternal phase volume and theECO-10(Castor Oil

Garti and Aserin

498 Conc.of

1( r g m ~ )

Conc.of

1 [rotmlj

2.0-

2.0Formulation A

a

Formulation 0

o FormulatlonC

o

Formulotlon D

m

10-

l.0.

+

Formulallon E

m Tormulation F

Formulation G

Formulation U

I

0.1-

025

(a)

0.5

0.75

* 1.0 2

Tame [h]

10

0.25

x) 24

(b)

05

0.75

10 2 10 Time [h]

20

Fig. 57. Serum concentrationof doxorubicin asa function of time after administration of into caroid arteryof SD rats, mean k SD, (n=6). (a) W/O/w emulsion; (b) in W/O emulsion. ethoxylated with 10EO units) (the emulsifier) concentration. The release of the 5-FU was faster and strongly depended on thepH (best at pH 10) [366]. 18. Prolonged release of bleomycin from parenteralgelatin spheresin-oil-in-water multiple emulsions. 19. Phenylephrine hydrochloride was formulated in different emulsion systems with and without viscosity builders (methylcellulose). Results indicate that the diffusion coefficient decreased with increasing viscolizer concentration. The mydriatic and intraocular pressure (IOP) were followed. Areaunderthecurve (AUC), maximum response (CMR), the time of maximum response (TMR), and the duration of drug action (DA) were found to improve in comparisonwith aqueous solution (Figs. 60 and 61). 20. Multiple W/OW emulsions containing pentazocine were prepared and tested in vitro and in vivo. The in vitro results indicated a well-controlled and higher drug release from the W/OW emulsions than the W/O emulsion. The invivo data showed prolonged tissue levels of pentazocine after oral administration of W/OW emulsions to mice in comparison with aqueous drug solution and W/O emulsion (Fig. 62).

24

Emulsions and Microemulsions COnC.Ol

1 [PQlQ]

0 Formulotlon A

FormulotlonC FormulatlonD

Formulotlon B

CON. 01 1 [pg~rni]

8-1 m Formulotlon E 61

(b)

499

Formulotlon F

Lung

Liver

l

a

T

FormuiotiooG

Q FormulotionH

Heart

Spleen Kidney

Fig. 58. Tissuedistributionofdoxorubicinat 24 h afteradministration of via carotid artery of SD rats. (a) W/O emulsion; (b) WIOiW emulsion.

U

TlN, h

500

Garti and Aserin

50 7

Emulsions and Microemulsions a

0

GO

120

180

MC

240

300

360

Time (minutes)

Fig. 61. Changeinpupildiameter of rabbit’s eye postinstallation of 2.5% phenylephrine HCl ophthalmic emulsions containing different concentrations o f (a) methylcellulose and (b) tylose.

XIV.

MICROEMULSIONS

A. DefinitionsandCharacterization The existence of a macroscopically stable homogeneous fluid, optically transparent, and isotropic has attracted much attention in the last 20 years [46,48,50-581. The original microemulsion was first defined Hoar by and Schulmanin 1959 [56] and consisted of water, benzene, hexanol, and potassium oleate. Most of Schulman’s work dealt with four-component systems: hydrocarbons, ionic surfactant, cosurfactant, four- to eight-carbon chain aliphatic alcohol, and an aqueousphase. The microemulsion wasformed only when the surfactant-cosurfactant blend formeda mixed filmof the oil-water interface, resulting in interfacial pressure exceedingthe initial positive interfacial

502

Garti and Aserin

tension (so-called negative interfacial tension). The emulsion was, therefore, produced spontaneously. Rosano etal. [367,368] measured the change in the water-oil interfacial tension while alcohol was injected into oneof the phases. It was found that theinterfacial tension may be temporarily lowered to zero while the alcohol diffused through the interface and redistributed itself between the water and oil phases. It would, therefore, be possible for a dispersion to occur spontaneously (while yi=O).Rosano et al. [367,368] have stressed that the spontaneousformation of these dispersed swollen micelles is not dependent on simple thermodynamic stability but rather, at least in part, on theoccurrence of kinetic conditions favorable to thedispersion of the dispersed phase into the O/W system. Microemulsions are not necessarily four-component systems. It is well documented that ternary water-oil nonionic surfactant can form microemulsions [369-3721.


503

The classic structural model for microemulsion is that of a monodispersed population of dynamic microglobules in a continuous medium. The optical transparency microemulsions indicates that the diameter microglobules cannot be larger than 0.1 pm or so. Since hydrocarbon or water is the continuous phase, the microglobules will be called “direct” or “inverse” microglobules, respectively. Such organizations do not necessarily exist in any microemulsion. Inmany compositions, one finds comparable amounts of water andoil, which can bedescribed as local organizations of each of the phasesin a dynamic bicontinuous structure (as suggested by Scriven [373,374] and others[375-3811). The phase behaviorof three-component systems can, atfixed pressure and temperature,best be represented using a ternary diagram [51,382-3841. These diagramsprovide a simple perspective of phase behavior thatis difficult to capture in any other way. Phase boundaries of systems containing more thanthree components canalso be represented in a phase diagram. A ternary diagram depicting a two-phase region and a single phase region is shown in Fig. 63 [385,386]. Any system whose overall composition lies within the two-phase region will exist as two phases whose composition is represented by the endsof the tie lines. In accordance with the phase rule, the surfactant concentration in the phaselabeled M can bevaried independently over a restricted range. Also showninFig. 64 is a plait point, or critical point, which is designated as P. The ties existing in the two-phase

AMPHlPHlLE

I

WATER

-

I

MISCIBILITY GAP

011

Fig. 63. Ternary diagram representation of two-phase region. The sloping lines are tie lines connecting conjugate phases.

,

Garti and Aserin

504 AMPHIPHILE

Fig. 64. Ternary diagram showing a three-phase invariant triangular region.

region surrounding this point are quite short, indicating that thetwo continuous phases have nearly the same composition. If three phases coexist, the system isinvariant. Then thereis a region of the ternary diagram where systems whose overall composition fall within it and divide into three phases, having compositions which are invariant and represented by the corners of the tie triangle (Fig. 65). Any M point will represent three phases with compositions represented by corresponding to thevertices A, B, and C of the triangle. Different points within the triangle will all have compositions of A, B, and C but will differ in the proportions of the quantities of each of the phases. Fig. 66 represents a real commercial surfactant system, hexane and water. The shaded triangle in the tie-triangle with three phases, each of them with a composition represented by corresponding vertices. Thus, the composition within the tie-triangle is composed of a micellar solution in equilibrium with excess oil and water phases. The micellar phase is often called the surfactant phase or themiddle phase (since it appears, in the test tube, sandwiched between the upper excess oil and the lower excess water). The tie-triangle is bounded by three two-phase regions. The twophase regions are bounded by a single-phase isotropic micellar solution designated L. The micellar solution represented by the surfactant-rich vertex of the tie-triangle is intermediate between S, and S,.


505

C SURFACTANT

,A

Single phase

I

,

I

I

A WATER

I

l

l

B OIL

Three phases

Fig. 65. Three-phase region for ternarysystemhaving three binarymiscibility gaps.

9.2

Fig. 66. A phase diagramof cyclohexane, nonylphenol ethoxylated NP9, and water at 62°C.

506

Garti and Aserin

AMPHIPHILE ......

OIL ..... .....

MICELLAR SOLUTION

PLAIT .AIT POINT WATER

OIL

Fig. 67. A typical S1 system, also known as Winsor I type system.

At high amphiphilic concentration, two additional one-phase regions can bedepicted. The phases are also knownas theliquid crystalline phases. The D phase is lamellar and the Y phase, also known as the M, phase, is hexagonal. A typical S, system is shown in Fig. 67. The P point is a region rich with oil, and therefore the two-phase region will have a two-phase one, which is practically just oil, and a micellar solution, in which the water phase will contain practically all the surfactant. Such a system is called Winsor type I. Fig. 68 describes a Winsor type I1 system, in which the water is separated from it and an oil phase, rich with surfactant, is orga'nized ona micellar structure [383,384].The transformation from S, compositions to anS, and vice versa can take place progressively by changing the temperature. Transformation from type I to type I1 systems can occur through an intermediate sequenceof three-phase systems. These can bedesignated as type I11 (Fig. 69). The microstructure of a type I system may be visualized as swollen micelles surrounded by water. The hydrocarbon present in such type I system is dissolved inthe interior of the micelle, forcing the micelle, owing to packing constraint, to become spherical in shape. S, or type I systems may, therefore, be imagined to be oil droplets surrounded by a sheath of amphiphiles, which separates the oil core from the continuous aqueous phase. The oil droplets are small enough so that the system is transparent


507

AMPHIPHILE MICELLAR SOLUTION WATER

WATER

OIL

Fig. 68. A typical Winsor I1 type system.

and considered to be single phased even though substantial regions of oil do exist. Determination of the structures of the different phases was, and in many cases still is, a difficult task. Methods such as light scattering [3873891, small-angle x-ray diffractions (SAXS), small angle neutron scattering (SANS) [390,394], quasi-electric light scattering (QELS) [395-3991, sedimentation [400,401], and dielectric measurements have been exercised. In spite of the recent great progress in developing the proper models (QELS and SANS) for determining the size and shapeof droplets within S, and S,, D or M areas, many investigators are not yet clear as towhat exactly is the structure of borderline composition in the phase diagrams, mainly the ones that are rich both in water and oil. When substantial oil or water quantities are added, the droplets are no longer spherical and the structure seems to be bicontinuous; that is, both the aqueous region and the oil regions are continuous and the interface separating them has essentially a constant mean curvature [373-3801. The first mesophase to appear from a swollen rodlike micelle is the middle phase, known also as the hexagonal I phase or theM, phase. If the amphiphile is dispersed in an organic phase, the middle phase will be termed reverse hexagonal I1 and will be defined as the M, phase. This phaseconsists of infinitely long rods in hexagonal array, as depicted in Fig. 6. X-ray diffracidentifition in the low-angle region is one of the most important methods for cation of lyotropic mesophases (Luzatti [402]). The hexagonal system is periodic in two dimensions and consists of cylindrical aggregates of which

Garfi and Aserin

508

A

A

A

THREE PHASES TIE TRIANGLE

CRITICAL END POINT

-

W

\ / i fhLC ESOLUTION M l

A

0

THREE-PHASE SYSTEM

\ W

0

A

A

TIE

W

Fig. 69. A sequence of ternary systems which are continuously transformed from type I to type 111 to type 11. This is denoted as a 1-111-11 transformation. The existence of two critical tie lines is most notable. there are two types. In hexagonal I (also called the middle phase), the lipophilic chains of the amphiphile occupy the linear coreof the cylinders, and the polar groups are arranged on the outer side in contact with the continuous waterphase. In hexagonal I1phase, the inner coreof the cylinders consists of water surroundedby a shell of polar groups of the amphiphile and the hydrocarbon chains of the amphiphile make up the continuous outer phase between the cylinders. At polarized light under the microscope, the appearance of the hexagonal I and the lamellar phases is easily identified. Distinction between hexagonal I and I1 is not possible by this method. The dilution with water will reveal the differences. Type I can be diluted, since spherical micelles are formed. Dilution with water is not possible for typeI1 because of the lipophilic nature of the amphiphile.


509

At higher amphiphile concentrations, the Neat phase with lamellar structure may appear [404]. This structure is designated as the G phase, and its structureis depicted in Fig. 6. Between the M, and G phases other mesophases sometimes occur with unidentified structures. At very high amphiphile concentrations and within the M, and G phases, a more viscous phase sometimes forms. This phase is termed theV-form or theviscous-phase. Recently, the designation cubic phase was adopted. The structureof the cubic phase has beenmuch discussed and is still debatable [402,404-4061 (Fig. 6). Liquid-crystalline structures are mesophases between crystals and isotropic solutions. Some investigators treat such systems from this perspective (Table 12). The micellar or surfactant phases are of great importance, since they have practicalapplications. It is the goal of the food technologists to find the Table 12. Designation of Some of the Common Lyotropic Phases

Basic

Phase

1. Micellar solutions More

l

1

1

or spherical, less swollen, S1 (optically isotropic) micelles containing solubilized, organic liquids. 2. Middle phase, normal Indefinitely long, mutually parallel M, (anisotropic) rods in hexagonal array. The rods consist of more or less radically arranged amphiphiles. The hydrophiles are in contact with the surrounding continuous aqueous phase. 3. Neat phase (anisotropic) Coherent double layers of G amphiphilic molecules with the with interfaces the in hydrophiles intervening layersof water. Indefinitely long mutually parallel M, 4. Middle phase, reversed rods in hexagonal array. The (anisotropic) lipophiles are arrangedso that the surrounding continuous organic phase is in contact with the lipophile. 5 . Inverted micellarsolutionsMore or lesssphericalinverted S23 L2 (optically isotropic) micelles containing solubilized water.

Source: From Ref. 403.

Aserin 570

and

Garti

proper oils and theright amphiphilesthat will give one-phase regions in the phase diagram,with maximumarea. Largeisotropic areas meanin practice a thermodynamically stablesurfactant phase thatis rich inboth oil and water. If, for example, one is interested in solubilizingsubstantial amounts of water (about 30 wt%), it wouldbe important tofind the surfactant or a combination of surfactants that will exhibit large L,areas (Fig. 70). Larger areas in the phase diagram mean moresolubilized water (Fig. 71). The three-phase tie-triangle is also very important, since it was found both experimentally and by theoretical calculations that any macroemulsion prepared with the composition represented by the corners of the tie-triangle will have better thermodynamic stability than any other emulsion prepared with a composition of any two-phase regions in the diagram. Qpical phase diagrams of soybean oil-water and various amphiphilesare shown in Fig. 72, demonstrating the differences in the areas of the isotropic phases as a function of the type of the amphiphile. For further reading, the book by Bourrel and Schechter [386]on microemulsions andrelated systems is highlyrecommended. One must also note that themicroemulsion regions are inequilibrium with the lamellar liquid crystallineregions. This is one of the reasons that in

HEXADECANE

WATER

AOT-TS/ARlACEL 20

Flg. 70. Pseudoternary phase diagram at 24°C for the water/hexadecane/Arlacel 2O/AOT-75 system. The AOT-75: Arlacel20 weight ratio is 40 : 60 (A A), 30 : 70 (0 O),20 : 80 (m.), 10 : 90 (-), and 100% (----).


51 1

HEXADECANE

20

Fig. 71. Pseudoternaryphasediagramat25°Cforthe waterkexadecanelAOT7YArlacel20 system. The ratio of AOT-75 to Arlacel20 are as follows:(- - -) for 57/ 43 wt ratio, whereas the solid line (-) is for the 54/46 AOT-75/Arlacel 20 ratio. Note the two-phase region within the single-phase region for the 54/46 AOT-75/ Arlacel20 system. practice we add cosurfactant, with medium-chain length (C, alcohol) to the lamellar liquid crystalline structure in order to destabilize the mesophase and transform it into microemulsion. The C,-OH will “upset” the lamellar packing and yill facilitate formation of very small spherical panicles with high curvature. The change in curvature (from flat or almost flat to highly curved) leads to a pronounced change in the freeenergy, (e.g., the bending component of the surface free energy) and leads to thermodynamic stability (Fig. 73). XV. MICROEMULSIONSAPPLICATIONS As previously explained, microemulsions are dynamic systems in whichthe

amphiphiles, water and oil, exchange very fast within each other. Drugs incorporated in microemulsions will, therefore, separatebetween the aqueous andoil phases depending on their lipophilicity. The mass transfer constants of several drugs through the surfactant phase (hydrophilic membrane) 4 were studied [407]. It was clearly demonstrated that a linear relationship

Garti and Aserin

512 Soybean 011

25°C

Emulsifier

Fig. 72. Completephasediagram of soybeanoil-water-emulsifier: (--.----.) polyglycerollinoleate, (-W--) polyglycerololeate, (-) monoglyceride stearate, and (--;-) R e e n 80.

Water A

0 Oil

0

&?h

Water

Fig. 73. The curvature in an emulsion droplets (A) is extremely small and the bending componentof the surface energyis not significant.A change in curvature (B), on does not lead toa change in the free energy. In the microemulsion droplet the other hand, a change in free energy in curvature to leads a pronounced change in free energy; e.g., the bending componentof the surface energy is pronounced.


513

between the release rate of the drug and itsoil-water partition coefficient exists. Several workers have reported studies in whichthe lipophilicity of the drug has beenincreased to enhance its solubility in the dispersed oil droplets. In this way, a reservoir of the drug is produced anda sustained release effect is achieved as the drug continuously transfers from the oil droplets to the continuous phaseto replace drug release from the microemulsion. An excellent example of such studies is theattemptmade by Gallarate [408] to add timolol into a microemulsion consisting of egg lecithin, butanol, isopropyl myristate, and octanoic acid solution buffered to pH 7.4. The drug was ion paired with octanoic acid, so that its lipophilicity wassignificantly increased. In addition, it wasshown that the in vitro permeability constants of timolol through a lipophilic membrane atpH 7.4 were about seven times higher when the timolol was present as anion pair, suggesting that an improved corneal absorption of this drug might be achieved by topical administration in this microemulsion vehicle. Similarly, the lipophilicity of propranolol has also been enhancedby the formation of lipophilic ion pairs with octanoic acid in O/W microemulsions containing isopropyl myristate, polysorbate 60, butanol, and buffer pH 6.5 [409]. The partitioning of the propranolol into the dispersed oil phase was increased by the addition of octanoic acid. As a consequence, its release rate through a hydrophilic membrane was decreased owing to its decrease in concentration in the continuous phase, producing a prolongation of drug release. Gallarate has also tried [410] to increase the solubility of the azelaic acid (dicarboxylic acid) in the dispersed oil phase of liveen 20 (ethoxylated sorbitan monolaurate)/butanol/decanolin water by lowering the pHof the aqueous phase and by adding propyleneglycol to further reduce its dissociation. The partition into thelipophilic phase was improved asthe propylene glycol concentration was increased. The azelaic acid dissolved better and could be incorporated into a cream as well as be released from the microemulsion corein amore controlled or prolonged manner(within hours). A 10-fold increase in the amount of drug released was demonstrated. A similar retention effect could have been achieved if an interaction between the drug and the amphiphile was obtained. The release rates of doxorubicin [411,412] from O/W microemulsions prepared with Sodium diis0 octyl sulfosuccinate (Aerosol-OT or AOT) or polysorbate 80 and W/O microemulsions prepared with lecithin were greatly reduced owing to the formation of lipophilic complexes between this water-soluble drug and each of the surfactants. A similar modification of drug release was later noted when l-demethoxy daunorubicinwas formulated in the same W/O lecithin microemulsion [413]. In view of the appreciable influence that thecosurfactant exerts on the

514

Garti and Aserin

properties of the microemulsion system, it is not unexpected thatit should also significantlyaffect drug release from the microemulsion. The influence of the cosurfactant concentration on the rate of release of six steroids with a range of lipophilicities from a series of O N microemulsions obtained by adding increasing volumes of butanol to fixed amounts of isopropyl myristate, water, and AOT was reported by Trotta etal. [414]. For all the drugs, the apparentpartition coefficient of drug intothe dispersed phase increased with butanol addition, producing a decrease in the release rate of the drug.A similar correlation between this partition coefficient and drug release was noted whenthese steroids were incorporated into microemulsionsof fixed composition preparedwith a range of alcohols as cosurfactants. The above conceptions have been exercised with slight modifications by several investigators for a variety of microemulsions indifferent areas of application. Some of the examples are grouped togetheraccording to their mode of delivery. A. Oral Administration A cyclosporine preparation was administered as a coarse dispersion, formed by mixing a vehicleof olive oil, alcohol, and polyoxyethylated oleic glycerowing ides with water. The preparation proved to be erratic and incomplete to the pair dispersibility of the drug in the microemulsion vehicle [415]. A cyclosporine [416], preparation using a W/O microemulsion containing a sorbitan ester-polyoxyethylene glycol mono ether mixture of surfactant, a low molecular weight alcohol, fatty ester, and water as the vehicle for the drugwas administered. A gel-like consistency was obtained throughthe addition of pyrogenic silica (Cab-o-Sil) and the microemulsions were administered in hard gel capsules. The better bioavailability of the drug [417,418] in comparison with anyother preparation (including an intravenous one) was attributed to the microemulsion droplets (Table 13). Ritschel [419] investigated the gastrointestinal absorption of three peptides using a series of O N microemulsion formulations. The peptides used were dissolved in the aqueous phaseat a suitable pH if water-soluble (insulin, vassopressin) or otherwise added to the microemulsion (cyclosporine) and dispersed by sonication. Three forms of microemulsion were used: a fluid microemulsion for in situ studies on the isolated segment rat model, a microemulsion gel formed by the addition of silicium dioxide (for rectal studies), and an encapsulated microemulsion gel for peroralabsorption studies). Improvements in the bioavailability of these peptides when administered orally were again not solely dependent on droplet size. The author Ritschel [419] concluded that the systemic peptide uptake from micro-


515

Table 13. Absolute (F) and Relative (EBA) Bioavailability in Rats of Cyclosporin A (AverageS D )After PerOs Solution and Two Peroral Microemulsion Formations p.0.

sion p.0. bcroemulsion Solution Parameters Absolute bioavailability F (%) Relative bioavailability EBA(%) Rate of bioavailability C,, (pg/mL) L a x (h)

p.0. BC 118

&

41.4 2.8

100 (standard)

1.95 4.35

*

4.36 0.03 0.59

f

447.1

9.00

15.0 18.1

-+

287.68 147.2

f

1.6Y

f 4.253.47

3.72

f

3.3

&

26.5

f f

1.35 0.50

8Significantlydifferentfromp.0.solutions. P .05 bMicroemulsion A: W/O microemulsion containing a long chain fatty acid as lipid lowmolecularweightalcoholas phase,Arlacel and Brijassurfactants,a cosurfactant, and distilled water. CMicroemulsion B: as A except that branched alkyl fatty esters were used as lipid phase. Source: From ref. 388. emulsion in the gastrointestinal tract was dependent additionally on the following formulation factors: type of lipid phase of the microemulsion, digestibility of the lipid used, and types of surfactant in the microemulsion. A detailed hypothesized mechanism of peptide absorption from microemulsions given perorally was proposed. B. Topical Drug Delivery

The effect of microemulsions on a drug vehicle was tested in various percutaneous absorptions through topical administration. The microemulsion consisted of AOT-octanol-water [420]. It was demonstrated that the tranderual flux increased sixfold as the water content of the microemulsion increased from 15 to 68%. The high water concentration served as transport vehicle to the addenda, leading to higher flux. Octanol and AOT had a synergistic effect as a penetration enhancer. Transport of glucose across human cadaver skin was demonstrated [421] using microemulsionscontaining up to68% water. A 30-fold enhancement of the glucose transport was achieved. FBvrier et al. [422] have reported in vitro experiments designed to simulate the percutaneous penetration of tyrosine when administered using an O W microemulsion composed of a betaine derivative (as surfactant),

516

Gatfi and Aserin

benzyl alcohol, hexadecane, and water. The release of radiolabeled tyrosine from this vehicle was compared with that from a liquid-crystal system andan emulsion using a diffusion cellequipped with rat skin. Both the microemulsion and the liquid-crystal formulation enhanced the penetration of tyrosine through the epidermis when compared with the emulsion. However, cutaneousirritation studies showed a strongly irritant effect from the liquid-crystal formulation but none from themicroemulsion. In a similar study by Ziegenmeyer and Fuhrer [423], tetracycline hydrochloride was reported to show enhanced percutaneous absorption from a microemulsion compared with conventional systems. Complete diffusion from the microemulsion occurred within 5-6 h comparedwith cornplete diffusion after 12 h from a gel and after more than24 h froma cream. Martini et al. [424] showed in an in vivo study that the percutaneous absorption of D-c~-(5-methyl-3H)tocopherol in rats was more rapid from a emulsion or white petroleumjelly an O/W microemulsion than fromW/O (Vaseline). Hence, the microemulsion not only enhances the rate of penetration butalso appears toinfluence the distribution and elimination of the drug. of the difficulty Gasco et al. [425] have recently addressed the problem in applying these formulations to the skin in the development of a microemulsion for the topical administration of azelaic acid, which has reported therapeutic effects on some pigmentarydisorders and on acnevulgaris. In an earlier study [380] discussedabove, these workers demonstrated the formulation conditions required to reduce the dissociation of this acidic drug. Its consequent enhanced partition into thedispersed phase of an O W microemulsion produced a reservoir of dissolved drug and a high rate of release. The viscosity of the O/W microemulsions used in this study was increased to make themsuitable for topical administration by the inclusion of Carbopol 934. A comparison of the in vitro release profiles through hairless skin membrane from this “viscosized” microemulsion with those from a previously reported gel containing similar components (propylene glycol, Carbopol934) is shown in Fig. 74. Clearly,a pronounced enhancement of penetration is achieved; after 8 h about 35% of the azelaic acid present in the microemulsion had been transported compared with only 1.8% from the gel. The release could be furtherincreased by the addition of a penetration enhancer, dimethyl sulfoxide, to themicroemulsion. Trottaet al. [426] havereported a similar enhancement of skin permeation of diazepam from “viscosized” O/W microemulsion prepared using egg lecithin, polysorbate 20,benzyl alcohol, isopropyl myristate, andwaterpropylene glycol mixtures. The potential application of microemulsions for theocular administration of timolol was investigated [427] usingO/W lecithin microemulsionsin


2

4

6

51 7

a

time (h)

Fig. 74. Permeation profiles of azelaic acid from a "viscosized" microemulsion (A)and from a gel ( A ) . (From ref. 425.) which this drug was present as an ion pair with octanoate. The microemulsion, a solution of the ion pair, and a solution of timolol alone were instilled in the conjunctival sac of rabbits and the bioavailability of timolol from eachwas compared (Fig. 75). The areasunder the curve fortimolol in aqueous humorafter administration of the microemulsion andthe ion pair solution were 3.5 and 4.2 times higher, respectively, than that observed after administration of timolol alone. A prolonged absorption was achieved using the microemulsion with detectable amounts of the drug still present in the aqueous humor120 minafter administration. The enhanced cornealabsorption of timolol as an ion pair might suggest the possibility of lowering the doses instilled in topical therapy, leading to a possible reduction of the side effects of this drug.

C. PerfluoroMicroemulsions Fluorinated surfactants have also been used to prepare fluorocarbon microemulsions (i.e., spontaneously formed, thermodynamicallydisperstable sions) in view of solving the long-term shelf-stability issue. F-Alkylated amine oxides, including XMO-10 and alcohols, were proposedas surfactants and cosurfactants for this purpose [428], but notoxicity data were provided. Likewise, spontaneously formeddispersions of fluorcarbons were achieved with mixtures of two F-alkylated polyoxyethylenederivatives [429]. Unfortunately, these preparations turned outto betoxic. Delpuech et al. obtained microemulsions with one single, appropriately chosen monodisperse F$

Garti and Aserin

518

0

A

A I

A

v

0

80

40

t

A

L

120

time (mln)

Flg. 75. Aqueous humor concentration-time profiles following multiple installation in rabbits’ eyes.+ ,timolol alone;0,timolol as an ion pairin solution; andA, timolol as an ion pair in microemulsion. (From ref.427.) alkylated polyoxyethylenesurfactant, but biocompatibility was not achieved there either [430]. One difficulty in the microemulsion approach lies in the large proportion of surfactant required, which makes the biocompatibility issue all the morearduous.

XVI. SUMMARY It isa very difficulttask to summarize in one chapter, 100 years of extensive work done in the area of dispersed liquids in pharmaceutical and drug designs. Micelles (direct and reverse) have always been attractive vehicles for drug confinement (entrapment) andrelease, but owing to its dynamic nature and the fast exchange betweenthe continuous and the internal reservoirs, it seems to be animpractical method for encapsulation of liquid or dissolved drugs. The liposomes havesignificant better capacities and seem to be bettervehicles for similar formulations. Emulsions, in spite of their thermodynamic instability and endless number of restrictions (on size of droplets, nature of the oil, and the surfactants), are excellent liquid preparations for many medical and pharmaceutical needs that requirecontrolled and slow release. Intravenous emulsionssuffer from many limitations, but topical and oral preparation have gained a great deal of interest and areused in many such applications.


519

Double emulsions are excellent potential systems for controlled and sustained releaseof drugs, but they stillsuffer from a lack of stability and a lack of the means to control the transport of matter across the two interfaces. It is our belief that futureliquid formulations will be based mainly on such formulations. Intravenous intakeseems to be very difficultbecause of size limitations, but topical and oral preparations show much promise, mainly in view of the use of macromolecular amphiphiles such as modified proteins, polysaccharides, glycolipids, and other naturally occurring molecules that form viscoelastic films at the interfaces and control (slow) the release of the drugs. The progress made in the last 15 years in understanding the mechanisms of stabilization (steric) and transport of matter across the oil membrane (reverse micelles) and across the interfaces opennew possibilities in the field of liquid dispersed formulations for drugs and pharmaceuticals. REFERENCES

I

of Emulsion Technology, 1. P. Walstra, Formationof emulsions, in Encyclopedia Vol. 1(P. Becher, ed.), Marcel Dekker, New York,1985, p. 57. 2. M. W. Lvnch and W. C. Griffin, Food emulsions, in Emulsions and Emulsion Technoldgy (K. J. Lissant, ed.), Marcel Dekker, New York, 1974, pp. 249290. 3. P. L. Lindner, Agricultural emulsions, in Emulsions and Emulsion Technology, Part I (K. J. Lissant, ed.), Marcel Dekker, New York,1974, pp. 179-236. of Emulsion Tech4. D. Z.Becher, Applications in agriculture, in Encyclopedia nology, Vol. 2 (P. Becher, ed.), Marcel Dekker, New York,1985, p. 239. 5. P. J. Mulqueen, Surfactants for agrochemical formulation, in Industrial Applications of SurfactantsI1 (D. R. Karsa, ed.), Royal Society of Chemistry, 1990, p. 279. 6. K.Larsson and S. E. Friberg (eds.), Food Emulsions, Marcel Dekker, New York, 1990. 7. E. DickinsonandG.Stainsby(eds.),AdvancesinFoodEmulsionsand Foams, Elsevier Applied Science, London,1988. 8. E.Dickinson andG . Stainsby (eds.), Colloids in Food, Applied Science Publishers, London, 1982. 9. E. Dickinson (ed.), Food Emulsions and Foams, Royal Society of Chemistry, London, 1987. 10. N. J. b o g , T. H. Riisom, and K. Larsson, Applications inthe food industry: I, in Encyclopedia of Emulsion Technology, Vol. 2 (P. Becher, ed.), Marcel Dekker, New York, 1985, p. 321. 11, in Encyclopediaof Emul11. E.N. Jaynes, Applications in the food industry: sion Technology, Vol. 2 (P. Becher, e d . ) , Marcel Dekker, New York, 1985,

520

Garti and Aserin

12. H.D.Graham,FoodColloids,AviPublishingCompany,Inc.,Westport, Connecticut,USA 1977. 13. S. S. Davis, J. Hadgraft, andJ. K. Palin, Medical and pharmaceutical applications of emulsions, in Encyclopedia of Emulsion Technology, vol. 2 (P. Becher, ed.), Marcel Dekker, New York, 1983, p. 159. 14. B. A. Mulley, Medical emulsions, in Emulsions and Emulsion Technology, Part I (K. J. Lissant, ed.), Marcel Dekker, New York, 1974, p. 291. 15. J. Kreuter, Nanoparticles, in Colloidal Drug Delivery Systems (J. Kreuter, ed.), Marcel Dekker,New York, 1994, p. 219. 16. M. M. Breuer, Cosmetic emulsions, in Encyclopedia of Emulsion Technology, vol. 2 (P. Becher, ed.), Marcel Dekker, New York, 1983, p. 159. 17. M.M.Rieger (ed.), Surfactants in Cosmetics, Vol. 16, Surfactant Science Series, Marcel Dekker,New York, 1984. 18. C. Fox, in Cosmetic Science, Vol. 2 (M. M. Breuer, ed.), Academic Press, London, 1980. of 19. H.A.BampfieldandJ.Cooper,EmulsionsexplosivesinEncyclopedia Emulsion Technology, vol. 3 (P. Becher, ed.), Marcel Dekker, New York, 1988, p. 281. 20. B. W. Davis, Applications in petroleum industry, in Encyclopedia of Emulsion Technology, vol. 3 (P. Becher,ed.), Marcel, Dekker,New York, 1988, p. 307. 21. B. El-Jazairi, Surfactants widely used in the concrete industry, in Industrial Applications of Surfactants (D. R. Karsa, ed.), Royal Society of Chemistry, 1987, p. 33. 22. A. D. James and D. Stewart, Cationic surfactants in road construction and repair, in Industrial Applicationsof Surfactants I1 (D. R. Karsa, ed.), Royal Society of Chemistry, 1990, p. 338. 23. K. R. F. Cockett, Surfactants in textile processing, in Industrial Applications of Surfactants (D. R. Karsa, ed.), Royal Society of Chemistry, 1987, p. 195. 24. D. Klamann, Lubricants and Related Products, Verlag Chemie, Weinheim, Germany, 1984. 25. Datyner, A. (ed.), Surfactantsin Textile Processing, Vol. 14, Surfactant Science Series, Marcel Dekker, New York, 1983. 26. R. A. Morland and N. Morgan, Surfactants in the paper and board industry, in Industrial Applicationsof Surfactants I1(D. R. Karsa, ed.), Royal Society of Chemistry, 1990, p. 356. 27. F. J. Kenny, Use of surfactantsin mineral flotation, Surfactants in the paper and board industry, in Industrial Applications of Surfactants I1 (D. R. Karsa, ed.), Royal Societyof Chemistry, 1990, p. 366. 28. D. Myers, Surfactant Science and Technology, VCH Publishers, New York, 1988, p. 209. NewYork, 29. D. Myers, Surfaces, Interfaces and Colloids, VCH Publishers, 1991. 30. M. J. Rosen, Surfactants and Interfacial Phenomena, Wiley, New York, 1989, p. 304. 31. Th. F. Tadros and B. Vincent, Emulsion stability, in Encyclopedia of Emul-


32. 33. 34. 35. 36. 37. 38. 39. 40.

41. 42. 43.

44. 45. 46. 47. 48.

49. 50.

51. 52. 53. 54.

52 l

sion Technology, Vol. 1 (P. Becher, ed.), Marcel Dekker, New York, 1983, p. 129. P. Becher and M. N. Yudenfreund (eds.), Emulsions, Lattices, and Dispersions, Marcel Dekker, New York, 1978. S. E. Friberg, Emulsion stability, in Emulsions-A Fundamental and PracticalApproach (J. Sjoblom,ed.),NATOAS1Series,KluwerAcademic, Dordrecht, The Netherlands, 1989, p.1. A. T. Florence and D. Whitehill, J. Colloid InterfaceSci., 79:243 (1981). S. Matsumoto and WW . . Kang, J. Dispersion Sci. Technol., 10:455 (1989). S. Matsumoto, Y. Kita, and D. Yonezawa, J. Colloid Interface Sci., 5 7 3 3 (1976). A. T. Florence and D. Whitehill, Int.J. Pharm., 11:277 (1982). Garti, N., Frenkel, M., and Shwartz, R., J. Dispersion Sci. Technol., 4:237 (1983). S. Matsumoto, J. Texture Studies, 17:141 (1986). S. S. Davis, Liquid membranes and multiple emulsions, Chem. Ind., 19583 (1981). S. Matsumoto, WIOIW-type multiple emulsions, in Nonionic Surfactants, Vol. 23 (Schick, M.J., ed.), Marcel Dekker, New York, 1987, p. 549. N. Garti, Double emulsions for food applications, Food Microstructure Dec., 14, (1995) (in press). G. S. Hartley, Paraffin-Chain Salts, Hermann& Cie, Pans, 1936. G. S. Hartley, Rep. Prog. Appl. Chem., 45x33 (1948). J. W. McBain, Colloidal Science, D. C. Heath, Lexington, MA, 1950. of liquid crystalline phases P. Ekwall, Composition, properties and structures in systems of amphiphilic compounds, in Advancesin Liquid Crystals, Vol.1 (Brown, G . H., ed.), Academic Press, New York, 1975, p.1. G. H. Brown and J. J. Wolken, Liquid Crystals and Biological Structures, Academic Press, London, 1979. K. Larsson, Emulsions of reversed micellar phases and aqueous dispersions of cubic phases of lipids: Some food aspects, in Microemulsions and Emulsions in Foods (M. El-Nokaly and D. Cornell, eds.), American Chemistry Society, Washington, D.C., 1991, p. 44. K. Larsson and P. Dejmek, Crystal and liquid crystal structures of lipids, in Food Emulsions (K.Larsson and S. E. Friberg, eds.), Marcel Dekker, New York, 1990, p. 97. R. G. Laughlin, Phase equilibria and mesophases in surfactant systems, in Surfactants (Th.F. Tadros, ed.), Academic Press, London, 1984, p. 53. M. Bourrel and R. S. Schechter, Microemulsions and Related Systems, vol. . 30, Marcel Dekker, New York, 1988, p. 102. I. D. Robb (ed.), Microemulsions, Plenum Press, New York, 1982. S. E. Friberg and B. Lindman (eds.), Organized Solutions, Surfactant Science Series, Vol. 44, Marcel Dekker, New York, 1992. S. E. Friberg and R. L. Venable, Microemulsions, in Encyclopedia of Emulsion Technology, Vol. I. (P. Becher, ed.), Marcel Dekker, New York, 1983, p. 287.

522 55. R.Zana(ed.),SurfactantSolutions:

Garti and Aserin

NewMethodsofInvestigation,Surfactant Science Series, Vol. 22, Marcel Dekker, New York, 1987. 56.T.P. Hoar, and J. H. Schulman, Nature, 152:102 (1943). 56a. F. B. Rosevear, Liquid Crystals: The mesomorphic phases of surfactant cornpositions. J. Soc. Cosmet. Chem. 19581 (1968). 57. J. T. Davies and E. K. Rideal, Interfacial Phenomena, Academic Press, New York, 1961, p. 16. 58. Th. F. TadrosandB.Vincent,Liquidniquidinterfaces,inEncyclopediaof EmulsionTechnology,Vol. 1 (P.Becher, ed.), Marcel Dekker, New York, 1983, p. 1. 59.N.K.Adam,ThePhysicsandChemistryofSurfaces,3rd ed., University Press, Oxford, UK, 1941, p.5 . 60. J. S. Rowlinson, Chem. Soc. Rev., 7:329 (1978). 61. B. Widom and J. S. Rowlinson, J. Chem. Phys., 52:1670 (1970). 62. J. E. Kirkwood and F. P. Buff, J. Chem. Phys., 17:338 (1949). 63. F. C. Goodrich, in Surface and Colloid Science, Vol. 3, (E. Matijevic, ed.), Wiley-Interscience, New York, p.1. 64. S. Middleman, The Flow of High Polymers, Interscience, New York, 1968. 65. R. Zana and L. G. Leal, in Proceedings of the International Colloquium on Drops and Bubbles (D. J. Collins, S. M.Plesset, andM. M. Saffren, eds.), Jet Propulsion Laboratory, Passadena, CA, 1974, p. 428. 66. R.W. Flumerfelt, Ind. Eng. Chem. Fund., 11:312 (1972). 67. R. Shinnar and J. M. Church, Ind. Eng. Chem., 52:253 (1960). 68. Y. Mlynek and W. Resnick, Am. Inst. Chem. Eng. J., 18:122 (1972). 69.J.W.Van Heuven, Ph.D. Thesis, Technical University of Delft, 1969. S. Noro, S. Ando, andM.Koishi,Chem.Pharm.Bull., 70.A.Takamura, 252617 (1977). 71. E. Lange, Z. Elektrochem., 5576 (1951). 72. R. Parsons, Modem Aspects of Electrochemistry (J. 0. M. Bockris and B. E. Conway, eds.), Butterworths, London,1954. 73. P. Becher, Emulsions: Theory and Practice, Krieger, New York, 1977, p. 267. 74. J. Lucassen-ReyndersandLucassen,Adv.ColloidInterfaceSci.,2:347 (1969). 75. E. S. R.Gopal,inEmulsionScience(P.Sherman,ed.),AcademicPress, London and New York, 1968, pp. 1-75. ed.),Pergamon, 76. E. S. Rajagopal, inRheologyofEmulsions(P.Sherman, London, 1963, pp. 15-25. 77. B. Bergenstahl, Topics in Food Emulsions, Ph.D. Thesis, University of Lund, Sweden, 1994. 78. H. L. Greenwald, J. Soc. Cosmetic Chem., 6:164 (1955). 79.M.Smoluchovski,Phys. Z.,17593 (1916). 80. L. A. Spielman, J. Colloid Interface Sci., 33562 (1970). 81. M. Smoluchowski, Z. Phys. Chem., 92:129 (1917). 82.P. F. Saffman and J. S. 'Tbmer, J. Fluid Mech., 1:16 (1956). 83. D. H. Melik and H. S. Fogler, J. Colloid Interface Sci., 101:72 (1984).

Emulsions and Microemulsions 84.

85.

86. 87. 88.

89. 90. 91.

92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111.

112. 113. 114. 115. 116. 117. 118.

523

D. H. Melik and H. S. Fogler, in Encyclopedia of Emulsion Technology, Vol. 3 (P. Becher, ed.), Marcel Dekker, New York, 1988, p. 3. M. Van den Temple, Rec. Trav. Chim., 72:433 (1953). Th. F. Tadros, Colloids and Surfaces 1:3 (1980). W. C. Griffin, J. Soc. Cosmetic Chemist 1:311(1949). W.C. Griffin, J. Soc. Cosmetic Chemist5249 (1954). C. D. Moore andM. Bell, Soap, Perfum. Cosmet. 29:893 (1956). H. L. Greenwald, G. L. Brown, and M. N. Fineman, Anal. Chem. 28:1693 (1956). J. T. Davies andE. K. Rideal, Interfacial Phenomena, Academic Press, New York, 1963, pp. 371-383. J. T. Davies,Proc. Int. Congr.SurfaceActivity, 2nd ed., Vol. 1, Butterworths, London, 1959, p. 426. I. R a n and E. Orban,J. Colloid Sci., 20:99,1965. P. Becher andR. L. Birkmeier, J. Am. Oil Chem. Soc. 41:169 (1964). P. Becher, Am. Perfumer., 76:33 (1961). K. Shinoda and H. Kunieda, Phase properties of emulsions: PIT and HLB, in Encyclopedia of EmulsionTechnology,Vol. 1 (P.Becher, ed.), Marcel Dekker, New York, 1983, p. 337. S. Friberg,EmulsionsandSolubilization,Wiley-InterK.Shinodaand science, New York, 1986, p.55. S. Ross, J. Soc. Cosmetic Chemists 6:184 (1955). S. U.Pickering, J. Chem. Soc. (Lond.) 91:2002 (1907). K. Shinoda and H. Saito,J. Colloid InterfaceSci., 30:2,258 (1969). K. Shinoda, J. Colloid Interface Sci., 25:396 (1969). K. Shinoda, Proceedingsof the International Congress Surface Activity, 5th ed., Vol. 2, Butterworths, London,1968, p. 275. K. Shinoda, H.Saito, and H. Arai, J. Colloid Interface Sci., 35:624 (1971). K. Shinoda and H. Takeda,J. Colloid Interface Sci., 32642 (1970). K. Shinoda, J. Colloid Interface Sci., 24:4 (1967). K. Shinoda and H. Sagitani,J. Colloid Interface Sci., 64:68 (1978). K. Shinoda, H. Kunieda,N. Obi, andS. E. Friberg,J. Colloid InterfaceSci., 80:304 (1981). H. Kunieda and K. Shinoda,J. Colloid Interface Sci.,75601 (1980). K. Shinoda andT. Ogawa, J. Colloid Interface Sci.,2456 (1967). K. Shinoda and H. Sagitani,J. Phys. Chem., 87:2018 (1983). D. H. Napper, J. Colloid Interface Sci., 58:390 (1977). K. P. Oza and S. G. Frank, J. Dispersion Sci. Technol., 7543 (1986). K. P. Oza and S. G. Frank,J. Dispersion Sci. Technol., 10:187 (1989). Y. Sela, S. Magdassi andN. Garti, Colloids and Surfaces, A., 83:143 (1994). J. A. Davidson and E. A. Collins, J. Colloid Interface Sei., 47:459 (1974). C. Holt, A. M. Kimber, B. Brooker and J. H. Prentice, J. Colloid Interface Sci., 65:555 (1978). P. Walstra and H.Oortwijn, J. Colloid InterfaceSci., 29:424 (1969). W.P. Hendrix and C. Orr,in Particle Sue Analysis (M. J. Groves and J. L.

524

Garti and Aserin

Wyatt-Sargent, eds.), Society for Analytical Chemistry, London, 1972, pp. 133-144.

P. Sherman, J. Colloid Interface Sci., 24:97 (1967). P. Sherman, J. Phys. Chem., 67:2531(1963). P. Sherman, J. Colloid Interface Sci., 27:282 (1968). K.Yokoyama, 19th Conference on IndustrialPharmaceuticalResearch, Mitsugane-San, Aichi, Japan,1982. 123. J. J. Windheuser, M. L. Best, and J. H. Penin, Bull. Parenter. Drug Assoc.,

119. 120. 121. 122.

24:286 (1970). 124. A. H. Goldberg, W. I. Higuchi, N. F. H. Ho, and G. Zografi, J. Pharm.Sci., 56:1432 (1967). 125. K. Yakayama,K.Yamanouchi,M.Watanabe, et al.,Fed.Proc., 34:1478 (1975). 126. A. Fricker, W. Griem, and K. Lang, Klin. Wschr., 47:735 (1967). 127. R. P. Geyer,Uptakeandretention of fattyacidsbytissueculturecells, Wistar. Inst. Symp. Monogr. 6:33 (1967). 128. C. J. van Oss, C. F. Gillman, and A. W. Neumann, Phagocytic Engulfment 129. 130. 131. 132. 133. 134.

and Cell Adhesiveness as Cellular Surface Phenomena, Marcel Dekker, New York, 1975. R. Jeppsson and S. Rossner, Acta Pharmacol. Toxicol., 37:134 (1975). H.JoyeuxandB.Astruc,Traitedenutritionartificialle del'adulte, I, SSTNA, Montpellier, France,1980. 0.Schuberth and A. Wretlind, Acta Chir. Scand. Suppl., 278 (1961). A. R. Vela, 0. L. Hartwig, M. Atik, et al., Am. J. Clin. Nutr., 16:80 (1965). C. D. Black and N. G. Popovich, Drug Intell. Clin. Pharm., E 1 8 4 (1981). W. R.Burnham, P.K. Hansrani,C.Knott,J.etal.,Int. J. Pharm., 13:9

(1983). 135. G. Hardy, R. Cotter, and R. Dawe, in Advances in Clinical Nutrition (I. D. A. Johnson, ed.), MTP, Lancaster, UK,1983, p. 241. 136. C. Kawilarang, K. Georghiou, and M. J. Groves, J. Clin. Hosp. Pharm., 551 (1980). 137. R. I. Lawrence, W. K. Flukes, V. J. Rust, and P. A. Braithwaite, Aust. J. Hosp. Pharm., 11540 (1981). 138. C. Solassol and H. Joyeux, Ann. Anaesthsiol. Fr. Spec., 2:75 (1974). 139. C. Solassol, H. Joyeux, R. Serrou, et al., J. Chir., 105:15 (1973). 140. I. D. A. Johnson(ed.), Advances in Clinical Nutrition,MTP, Lancaster, UK, 1983. 141. H. C. .Meng and D. W. Wilmore (eds.), Fat Emulsions in Parenteral Nutrition, American Medical Association, Chicago,1976. 142. S. S. Davis, in Advances in Clinical Nutrition (I. D. A. Johnson,ed.), M T P , Lancaster, UK, 1983, p. 213. 143. A. Wretlind, in Fat Emulsions in Parenteral Nutrition (H. C. Meng and D. W. Wilmore, eds.), American Medical Association, Chicago, 1976, p. 109. 144. J.Molnar, S. McLain,C.Allen, et al.,Biochim.Biophys.Acta, 493:37 (1977).


525

145. C.T.Ashworth,N.R.DiLuzio,and S. J.Riggi,Exp.Mol.Pharm., l(Suppl.):83 (1963). 146. H. Tonaki, T. M. Sata, L. W. Mayron, and E. Kaplan, Exp. Mol. Pharm., l(Suppl.):83 (1963). 147. A. L. Kunz, R. E. Lewis, and G. J. Sperandio, Bull. Parenter. Drug ASSOC., 19:13 (1965). 148. S. Muchtar, M. Y. Levy, S. Sarig, and S. Benita, S.T.P. Pharm. Sci., 1:130 (1991). 149. C. Washington, Int. J. Pharm., 66:l (1990). 150. C.Washington,A.Athersuch,andD.J.Kynoch,Int.J.Pharm.,64:217 (1990). 151. A. Wretlind and B. Ajaxon, Eur. Patent ApplicationEP 220,152,1987. 152. J. D. Sherwood, Anal. Proc., 24:277 (1987). 153. J. D. Sherwood, Anal. Proc., 24:279 (1987). 154. C. Washington andS. S. Davis, Int. J. Pharm., 44:169 (1988). 155. S. Muehlebach, R. B. Graf,andK.Sommermeyer,Pharm.ActaHelv., 62:130 (1987). 156. J.P.Reynier, J. Danho, P. Faurenfies, and A. Bovis, Pharm. Acta Helv., 62:93 (1987). 157. V.A. Parry, K. R. Harrie, and N. L. McIntosh-Lowe, Am. J. Hosp. Pharm., 43:3017 (1986). 158. U. Julius andW. Leonhardt, J. Parenter. Enteral Nutr.,12:116 (1988). 159. Y. B. Bannon, J. Corish, 0. I. Corrigan, et al., Eur. Patent Application EP 289,342 (1988). 160. R. Jeppsson, Acta Pharm. Suec., 9:81 (1972). 161. R. Jeppsson and S. Ljunberg, Acta Pharm. Suec., 10:129 (1973). 162. R. Jeppsson, Acta Pharm. Suec. 13(Suppl.):43 (1976). 163. R. Jeppsson and S. Ljunberg, Acta Pharm. Toxicol., 36:312 (1975). 164. R. Jeppsson, Acta Pharm. Toxicol., 36:299 (1975). 165. G. I. Schoefl and J. E. French, Proc. R. Soc., B169:153 (1968). 166. G. J. Jesmok, G. M. Walsh, W. Ditzler, and E. F. Woods, Proc. Soc. Exp. Biol., 162:458 (1979). 167. R. Jeppsson, J. Clin. Pharm., 1:181 (1976). 168. 0. von Dardel, C. Mebius, and T. Mossberg, Opusc. Med., 20:255 (1975). M. Ajaxon, U.S. Patent 169. K. A. Wretlind, S. Ljunberg, I. Hakansson, and B. 4,073,943 (1978). 170. R. Jeppsson, Acta Pharm. Toxicol.,3756 (1975). 171. F. J. Laska and M. R. Fennessy, Clin. Exp. Pharm. Ther.,3587 (1976). 172. H. 0. J. Collier, D. L. Francis, and C . Schneider, Nature 237:220 (1972). 173. C. L. Fortner, W. R. Grove, D. Bowie, and M. D. Walker, Am. J. Hosp. Pharm., 32582 (1975). 174. C. L.Litterst, E. G.Mimnaugh,A.C.Cowles,J.Pharm.Sci.,63:1718 (1974). 175. A.J. Repta, in Topics in Pharmaceutical Sciences (D. D. Breimer and P. Speiser, eds.), Elsevier, Amsterdam, 1981, p. 181.

526

Garti and Aserin

176. A. A. A. El-Sayed and A. J. Repta, Int. J. Pharm., 13:303 (1983). 177. B. Meyerson, Horm. Behav., 3:l (1972). 178. Y. Mizushima, T. Hamano, and K. Yokoyama,J. Pharm. Pharmacol., 34:49 (1982). 179. D. M. Weir, Immunology, 4thed., Churchill, London, 1977. 180. P. Jolles and A. Paraf, Chemical and Biological Bases of Adjuvants, Chapman & Hall, London, 1973. on Adjuvantsof 181. P. MuggletonandM.Hilton,InternationalSymposium Immunity, Symposium Serieson Immunobiology, Standard6,Utrecht, 1966, Karger, Basel, 1966,p. 29. 182. W. J. Herbert, Immunology, 14:301 (1968). 51:238 (1973). 183. P. Steinberg and P. S. Norman, J. Allergy Clin. Immunol., 184. W. H. 0.Scientific Group, Who Tech. Rep. Ser. 595,World Health Organization, Geneva, 1976. 185. J. Freund and K. McDermott, Proc. Soc. Exp. Biol. Med., 49548 (1942). 186. W.J. Herbert, in Handbook of Experimental Immunology, Vol. 3 (D. M. Weir, ed.), Blackwell Scientific, London,1978,p. A31. 187. M. R. Hilleman, Prog. Med. Virol., 8:131 (1966). 188. E. G. White, inModeh Trends in Immunology, Vol.2 (R. Cruickshank and D. M. Weir, eds.), Butterworth, London,1967,p. 28. 189. J. Lazarus and L. Lachmann, Bull. Parenter. Drug Assoc.,21:184 (1967). 190. J. A. Reynolds, Infect. Immun., 28(B):937 (1980). 191. C.A.McLaughlin, E. E. Ribi,M.B. Goren,andR.Tbubiana,Cancer Immunol. Immunother., 4:109 (1978). 192. K. Kakemi, H. Sezaki, K. Okumura, and S. Ashida, Chem. Pharm. Bull., 17:1332 (1969). 193. M.Hashida, Y. Takahashi, S. Muranishi,and H. H. Sezaki, J. Pharmacokinet. Bipharm., 5:225 (1977). 194. M.Hashida, S. Muranishi,and H. Sezaki,Chem.Pharm.Bull., 25:2410 (1977). 195. Y.Nakamoto, M. Fujiwara, T. Noguchi, et al., Chem. Pharm. Bull., 23:2232 (1975). 196. Y. Nakamoto,M.Hashida, S. Muranishi,andH.Sezaki,Chem.Pharm. Bull., 23:3125 (1975). 197. M. Hashida, Y. Takahashi, S. Muranishi,andH. H. Sezaki, J. Pharmacokinet. Bipharm., 5241 (1977). 198. M. Hashida, S. Muranishi, and H. Sezaki, et al., IntJ. Pharm., 2:245 (1979). 199. T. Takahashi, S. Ueda, K. Kono, and S. Majimi, Cancer, 38:1507 (1976). 200. J. N. Hunt and M.T. Knox, J. Physiol. (Lond.), 194:327 (1968). 201. A. Cooperman and S. A. Cook, Surg. Clin. North Am.,54:1277 (1976). 202. A. F. Hofman and D. M. Small, Annu. Rev. Med., 18:333 (1967). 203. C. R. Treadwell and G. V.Vahouny, in Handbook of Physiology, Sect. 6, Vol. 3 (C. F. Code, ed.), American Physiological Society, Washington, D.C., 1968,p. 1407-1438. 204. S. Muranishi, N. Muranishi, and H. Sezaki, Int. J. Pharm., 2:107 (1979).


527

205. N.Muranishi,M.Kinugawa, Y. Nakajima, et al.,Int. J.Pharm.,4:271 (1980). 206. S. M. Sieber, V. H. Cohn, andW. T. Wyn, Xenobiotica, 4:265 (1974). 207. J. E. F. Reynolds (ed.), Martindale, The Extra Pharmacopoeia, 28th ed., Pharmaceutical Press, London, 1982. 208. J. F. Hoover (ed.), Remington’s Pharmaceutical Sciences, 15th ed., Mack, Easton, PA, 1975. 209. T. Furia (ed.), Handbook of Food Additives, Chemical Rubber Company, Cleveland, 1968, p. 626. 210. S. E. Svenson, W.F. Delorenzo, R. Engelberg, et al.,Antibiot.Med., 11:148 (1956). 211. C. W. Daeschner, W. R. Bell, P. C. Stivrins,et al., Am. Med. Assoc. J. Dis. Child., 93:370 (1957). 212. P.J. Carrigan, Ph.D. Thesis, University of Connecticut, Storrs, 1974. 213. P. J. Carrigan and T. R. Bates,J. Pharm. Sci., 62:1476 (1973). Sci., 64:793 (1975). 214. T. R. Bates and J. Sequira, J. Pharm. 215. G. Gagnon and A. M. Dawson, Proc. Soc. Exp. Biol. Med., 127:99 (1968). 216. L. Diamond, Arch. Int. Pharmacodyn., 185246 (1970). 217. H. Ogata and H. Leung-Fung, Int. J. Rharm., 5:335 (1980). 218. D. Hollander, Am.J. Physiol., 239:G210 (1980). 219. J. M. Lewis, S. Q. Colhar, and A. Messina, Pediatrics,5425 (1950). 220. W. J. Warwick, L. G. Hansen, and H. Sharp, Clin. Pediatr., E 8 0 7 (1976). 221. T. De Marco and R. Levine, J. Pharmacol. Exp. Ther., 169:142 (1969). Soc. Exp.Biol.Med.,129:772 222. R. H. Engel and M. J. Fahrenbach, Proc. (1968). 223. R. Hori, K. Okumura, K. Inui,et al., Chem. Pharm. Bull., =:l974 (1977). 224. D. C. Bloedow and W. L. Hayton, J. Pharm. Sci., 65:328 (1976). 225. F. H. Mattson, R. J. Jandacek, and M. R. Webb, J. Nutr., 102:1177 (1976). 226. R. A. Volpenhein, D. R. Webb, and R. J. Jandacek, J. Toxicol. Environ. Health, 6:679 (1980). 227. K. Kakemi, H. Sezaki,S. Muranishi et al., Pharm. Bull., 20:708 (1972). 228. K.Kakemi,H.Sezaki, S. Muranishi et al.,Chem.Pharm.Bull.,20:715 (1972). 229. K. Kakemi,H.Sezaki, H. Ogata,andC.Nagai,Chem.Pharm.Bull., 20:1053 (1972). 230. T. Noguchi, Y. Tokunaga, M. Ichikawa, et al., Chem. Pharm. Bull., 25:223 (1977). Sci., 63566 231. G. M.Lin, F. Sadik, W. F. Gilmore, and J. H. Fincher, J. Pharm. (1974). 232. H.Schaefer,A.Zesch,andG.Stuttgen,inHandbuchderHautund Geschlechtskrankheiten Erganzungswerk, Vol. 1/4B, Springer-Verlag, Berlin, 1981. 233. M. Katz and B. J. Poulsen, in Handbook of Experimental Pharmacology, Vol. 28/1 (B. B. Brodie and J. Gillette, eds.), Springer-Verlag, Berlin, 1971, pp. 103-174.

528

Garti and Aserin

234. N. J.VanAbbe, R. I. C. Spearman, and A. Jarrett, Pharmaceutical and Cosmetic Products for Topycal Administration, Heinemann, London, 1969. 235. M. Katz and B. J. Poulsen, J. Soc. Cosmetic Chemists, 23565 (1972). 236. M. Parr, J. Pharm. Sci., 51:395 (1962). for Topical Admin237. N. J. VanAbbe, in Pharmaceutical and Cosmetic Products istration, Heinemann, London, 1969, p. 98. 238. W. G. Waggoner and J. H. Fincher, J. Pharm. Sci., 60:1830 (1971). 239. F. R. Bettley, Br. J. Dermatol., 73:448 (1961). 240. I. H. Blank, E. Gould,andA.Theobald,J.Invest.Dermatol.,37:485 (1961). 241. J. Scala, D. E. McOsker,andH. H. Reller,J.InvestDermatol.,50:371 (1968). 242. L. H. McDonald and K. Himelick, J. Pharm. Assoc. Sci. Ed., 37:368 (1948). 243. P. H. Dugand and R. J. Scheuplein, J. Invest. Dermatol., 60:263 (1973). 244. B.W. Barry and G. M. Saunders, J. Colloid Interface Sci., 41:331 (1972). 245. B. W. Barry and G. M. Eccleston, J. Pharm. Pharmacol., 25244 (1973). 246. B. W. Barry, Adv. Colloid Interface Sci., 5:37 (1975). 247. G . M. Eccleston, J. Colloid Interface Sci., 57:66 (1976). 248. I. Sarkany and J. W. Hadgraft, Br. J. Dermatol., 81:98 (1969). 249. B. W. Barry and R. Woodford, Br. J. Dermatol., 91:323 (1974). 250. Z.T. Chowhan and R. Pritchard, J. Pharm. Sci., 64:754 (1975). 251. N.F. Billups and N. K. Patel, Am. J. Pharm. Ed., 34:190 (1970). 252. T. Koizumi and W. I. Higuchi, J. Pharm. Sci., 57237 (1968). 253. D. V. Barker, H. G. DeKay, and J. E. Christian, J. Am. Pharm. ASSOC., 45527 (1956). 254. L. D. Lockie andT. B. Sprowis, J. Am. Pharm. ASSOC., 38:222 (1949). 255. D. V. Barker, J. E. Christian, and H. G. DeKay, J. Am. Pharm. ASSOC., 45501 (1956). 256. J. A. Wood and L. W. Rising, J. Am. Pharm. ASSOC., 42:481(1952). 257. A. H. Fenton andM. Warren, Phann. J., 1885 (1962). 258. W. S. Singleton and M. L. Brown, J. Am. Oil Chem. Soc., 42:312 (1965). 259. J. Greiner, J. G. Riess, and P. Vierling, in Organofluorine Compounds in Medicinal Chemistry and Biomedical Applications(R. Filler, ed.), Elsevier, The Netherlands, 1993, p. 339. 260. J. G. Riessand M. Postel,Proceedingsof theInternationalSymposium BloodSubstitute,Montreal,1991,Biomat. Art.Cells,Immob.Biotech. 20:819 (1992). 261. R. P. Geyer, in Drug Design, Vol. 7 (A. J. Ariens, ed.), Academic Press, London, 1976, p. 1. 262. S. S. Davis, H.P. Round, T. S. Buscall, et al., in Recent Advances in Colloid and Interface Science, Vol. 2 M. Kerker, ed.), Academic Press,New York, 1976, p. 265. 263. S. S. Davis, H. P. Round, andT. S. Purewal,J. Colloid InterfaceSci., 82:307 (1981). 264. T. M. S. Changand R. P. Geyer, eds., in Blood Substitutes, p, 411,M. Dekker, Inc., New York (1989).


529

265. J. G. Reiss, J. L. Dalfors, G. K. Hanna, et al., Biomat., Art. Cells, Immob. Biotech., 20:839 (1992). 266. J. G. Reiss, J. Greiner, S. Abouhilale, and A. Milius, Prog. Colloid Polym. Sci., 88:123 (1993). 267. A. Milius, J. Greiner, andJ.G. Reiss, New J, Chem., 16:771 (1992). 268. A. Milius, J. Greiner, andJ. G. Reiss, Colloids Surfaces, 63:281 (1992). 269. J. Greiner, A. Milius, and J.G. Reiss, Proceedings of the 3rd International Surfactants Congress, London, Sect. C, 1992, p. 31. 270. S. Abouhilale,J.Greiner, andJ.G . Reiss,J. Am. Oil Chem.Soc., 68:1(1991). 271. J. B. Nivet, M. LeBlanc, andJ. G. Reiss, Eur.J. Med. Chem., 26:953 (1991). 272. Santaella, P. Vierling, and J.G. Riess, New J. Chem., 16:399 (1992). E. A.O'Rear, J. ColloidInterface Sci., 273. M. Gangoda,B.M.Fung,and 116:230 (1987). J. G.Riess,andL.Zarif,Makromol.Chem., 274. A.A.Pavia,B.Pucci, 193:2505 (1992). J. FluorineChem., 58:194 275. J. Greiner,E.Myrtil,E.Dere,etal.,Pavia, (1992). 276. S. Gaentzler andP. Vierling, New J. Chem., 17:337 (1993). 277. L. Zarif, A. Manfredi, C. Varescon, et al., J. Am. Oil Chem. Soc., 66:1515 (1989). J.. Rolland, P. Vierling, 'andJ. G. Riess, New J. Chem., 278. P.M.P. Krafft, P 14:869 (1990). 279. A. M. Mahe, J. Manoux,A.Valla, et al.,Biomat.,Art.Cells,Immob. Biotech., 20:1025 (1992). 280. L. Zarif, J. Greiner, S. Pace, andJ. G. Riess, J. Med. Chem., 33:1262 (1990). 281. T. Fujita, T. Sumaya, andK. Yokoyama, Eur. Surg. Res., 3:436 (1971). 282. L. Illum and S. S. Davis, J.Parenter. Sci. Technol., 36:242 (1982). 283. H. A. Sloviter, M. Petkovic, S. Ogoshi, and H. Yamada, J. Appl. Physiol., 27566 (1969). 284. K.Yokoyama, K. Yamanouchi, M. Watanabe, et al., Fed. Proc., Fed. Am. Soc. Exp. Biol., 34:1478 (1975). 285. H.Ohyanagi, K. Toshina, M. Sekita, et al., Clin. Ther., 2:306 (1979). 286. K. Honda, S. Hoshino, M. Shafi, et al., N. Engl. J. Med., 303:391 (1980). 287. K. K. Tremper, R. Lapin, E. Levine, et al., Crit. Care Med., 8:738 (1980). 288. T. Fujita, T. Sumaya, andK. Yokoyama, Eur. Surg. Res., 3:436 (1971). 289. M. Bury and C. Boymond,S. T. P. Pharm., 6474 (1990). 290. S. S. Davis, C. Washington, P. West, et al., Lipid Emulsions as Drug Delivery Systems, Ann. NY Acad. Sci., 507:75-88 (1987). 291. S. S. Davis, in Phospholipids (I. Hanin and G. Pepeu, eds.), Plenum Press, New York, pp. 69-82 (1990). 292. J. B. Boyett and C. W. Davis, in Pharmaceutical dosage forms: Disperse Systems, Vol. 2 (H. A. Lieberman, M. M. Rieger, and G . S. Banker, eds.), Marcel Dekker, New York, 1989, pp. 379-416. 293. B. Idson, in Pharmaceutical dosage forms: Disperse Systems,Vol. 1 (H. A. Lieberman, M. M. Rieger, and G. S. Banker, eds.), Marcel Dekker, New York, 1989, pp. 199-243. ,

530

Garti and Aserin

294. S. S. Davis, Optimization of Drug Delivery (H. Bundgard, A. Bagger, Hansen and H. Kofod, eds.), Alfred Benzon Symposium 17, Copenhagen, 1982, pp. 333-347. 295.M.Y. Levy and S. Benita, Int. J. Pharm., 54:103 (1989). 296. S. Zheng, Y. Zheng, R. L. Beissinger, D. T. Wasan and D. L. McCormick, Biochim. Biophys. Acta, 115865 (1993). 297.L. C.Collins-Gold,R. T. Lyons,andL.C.Bartholow,Adv.DrugDeliv. Rev., 5:189 (1990). 298. 0.von Dardel, C. Mebius, and T. Mossberg, Acta Anaesth. Scand., 20:221 (1976). 299. S. Benita, D. Friedman,andM.Weinstock, Int. J. Pharm.,30:47(1986). 300. R. J. Prankerd and V. J. Stella, J. Parent. Sci. Technol., 44:139 (1990). 301. M. Singh and L. J. Ravin, J. Parent. Sci. Technol., 4 0 9 (1986). 302.P.Becher,EmulsionTheoryandPractice, 2nd ed., Reinhold, NewYork, (1965), p. 2. 303. S. S. Davis, J. Hadgraft, and K. J. Palin, in Encyclopediaof Emulsion Technology, Vol. 2, (P. Becher, ed.), Marcel Dekker, New York, 1985, p. 159. 304. A.T. Florence and D. Whitehill, Int.J. Pharm., 11,277 (1982). 305. P. Docic and P. Sherman, Coll. Polym. Sci., 258:1159 (1980). 306. W. Seifriz, J. Phys. Chem. ,29, 738 (1980). 307. T. J. Lin, H. Kurihara, and H. Ohta, J. Soc. Cosmet. Chem., 26121 (1975). 308. S. Matsumoto, WlOiW-type multiple emulsions, in Nonionic SurfactantsPhysical Chemistry, Surfactants Science Series, Vol. 23, (M.J. Schick, ed.), Marcel Dekker, New York, 1985, p. 549. 309. A. T. Florence and D. Whitehill, J. Colloid InterfaceSci., 79:243 (1981). 310. S. Matsumoto and WW . . Kang, J. Dispersion Sci. Technol., 10455 (1989). 311. S. Matsumoto, Y. Kida, and D. Yonezawa, J. Colloid Interface Sci., 57:353 (1976). 312. A. T. Florence and D. Whitehill, J. Colloid Interface Sci., 79:243 (1981). 313.M. Frenkel,R.Shwartz,and N. Garti, J. ColloidInterface Sci., 94:174 (1983). 314. N. Garti, M.Frenkel, and R. Schwartz, J. Dispersion Sci. Technol.,4:237 (1983), 315. A. F. Brodin and S. G. Frank, Acta Pharm. Suec., 15111 (1978). 316.C.Chiang,G.C.Fuller, J. W. Frankenfeld, and C. T. Rhodes, J. Pharm. Sci., 67:63 (1978). 317. S. Magdassi,M. Frenkel,and N. Garti, J. DispersionSci.Technol.,5:49 (1984). 318. S. Magdassi,M. Frenkel, and N. Garti, Drug Ind. Pharm., 11:791 (1985). 319. S. Magdassi and N. Garti, J. Controll. Rel., 3:273 (1986). 320. Y. Kita, S. Matsumoto, and D. Yonezawa, J. Colloid Interface S c i . , 62237 (1977). 321. T. Higuchi, J. Pharm. Sci., 52:1145(1963). 322. T. K. Law, A. T. Florence, and T. L. Whateley,J. Pharm. Pharmacol., 3650 (1984).


53 1

323. T. K. Law, T. L. Whateley, andA. T. Florence, Int. J. Pharm., 21:277 (1984). 324. T. K. Law, T. L.Whateley,and A. T. Florence, J. Controll.Rel., 3:279 (1986). 325. A. T. Florence, T. K. Law,andT.L.Whateley,J.Colloid Interface Sci., 107584 (1985). 326. J. A. Omotosho, T. K. Law, T. L. Whateley, and A. T. Florence, Colloid and Surfaces, 20:133 (1986). 327. J. A. Omotosho, T. K. Law, T. L. Whateley, and A. T. Florence, J. Pharm. Pharmacol., 38:865 (1986). 328. J.A.Omotosho,T. K. Law, T. L. Whateley,andA. T. Florence, J. Microencaps., 6:183 (1989). 329. R. L. Beissinger, D. T. Wasan, L. R. Sehgal, and A. L. Rosen, UK Patent Application GB2,221,912 (1990). 330. S. Zheng, R. L. Beissinger, andD. T. Wasan, J. Colloid Interface Sci., 144:72 (1991). 331. K. P. Oza and S. G. Frank, J. Dispersion Sci. Technol., 7543 (1986). 332. K. P. Oza and S. G. Frank, J. Dispersion Sci. Technol., 10:187 (1989). 333. N. Garti, A. Aserin, andY. Cohen, J. Controll. Rel., 29:41 (1994). 334. W.J. Herbert, Lancet, 2:771(1965). 335. R. H. Engel, S. J. Riggi, and M. J. Fahrenbach, Nature, 219:856 (1968). 336. J. A. Omotosho, Int. J. Pharm., 62:81(1990). 337. Y. Morimoto, K. Sugibayashi, Y. Yamaguchi, and Y. Kato, Chem. Pharm. Bull., 27:3188 (1979). 338. P. J. Taylor, C. L. Miller, T. M. Pollock,et al., Hygiene, 67:485 (1969). V. Mitchley, A. J.Collings,andR.Schneider,Eur. J. 339. L.A.Elson,B.C. Clin. Biochem. Res., 1587 (1970). 340. A. J. Collings, U.K. Patent 1,235,667,1971. 341. C. H.Benoy, L. A. Elson, and R. Schneider, Rev. Eur. Etudes Clin. Biol., 45:135 (1972). 342. A. F. Brodin and S. G. Frank, Acta Pharm. Suec., 151(1978). 343. N.N. Li, U.S. Patent 3,410,794,1968. 344. N.N. Li, Amer. Inst. Chem. Eng. J. (A.I.Ch.E.J.), 17:459 (1971). L. Shrier, Recent Developments in Separation Science, 1, 345. N. N. Li and A. Chem. Rubber Co., Cleveland,1972, p. 163. 346. S. Zheng, Diss. Abstr. Int. B, 51:2485(1980). 347. M. S. Mohamed, F. S. Ghazy, M. A. Mahdy, and M. A. Gad, J. Pharm. Sci., 5 5 6 (1989). 348. K. P. Oza and S. G. Frank, J. Dispersion Sci. Technol., 10:163 (1989). 349. T. Takahashi, M. Mizuno, Y. Fujita, et al., Gann, 64:345 (1973). 350. J. A. Omotosho, A. T. Florence, and T. L. Whateley, Int. J. Pharm., 6151 (1990). 351. J.A.Omotosho, T. L.Whateley,andA.T. Florence,Biopharm.Drug Dispos., 10:257 (1989). 352. Q. Wu, Q. Ping, and G. Liu, Zongguo Yoyao Gongye Zashi,21:445 (1990). 353. S. C. Kundu, Diss. Abstr. Int. B 51:1763 (1990).

532

354. 355. 356. 357. 358. 359. 360. 361. 362. 363. 364. 365. 366. 367. 368. 369. 370. 371. 372. 373. 374. 375. 376. 377. 378. 379. 380. 381. 382. 383.

Garti and Aserin

B. Mishra and J. K. Pandit, J. Controll. Rel.,1453 (1990). M. Mathew andC. S. Thampi, East Pharm., 33:385 (1990). J. A. Omotosho, Int. J. Pharm., 62:81 (1990). I. Eros, J. Balazs, I. Peter, and M. Tacsi, Pharmazie, 45419 (1990). S. Ismail, A. A. Mohamed, and S. A. Hasan, Bull. Pharm. Sci. Assuit Univ., 12:68 (1989). W.J. Herbert, Lancet,2:771(1965). W. J. Herbert,Int. Symp. on Adjuvants of Immunity,Symp.Seri.Immunobiol. Standard 6, Utrecht, 1966, Karger, Basel, 1966, p. 89. W.J. Herbert, U.K. Patent 1,080,994 (1967). L. E. Elson, B. C. V. Mitchley, A. J. Collings, and R. Schneider, Eur. J. Clin. Biochem. Res., E 8 7 (1970). C.J.Benoy, L. A. Elson, andR.Schneider,Br.J.Pharmacol.,45:135P (1972). P. A. Gresham, M. Barrett, S. V. Smith, and R. Schneider, Nature, 234:149 (1971). S. Y. Lin, W. H. Wu, and W.Y. Lui, Pharmazie 46:439 (1992). S. Fukushima, K. Juni,and M.Nakanu,Chem.Pharm.Bull.,31:4048 (1983). H.L. Rosano, R. C. Peiser, andE. A. Eydt,Rev. Franc. Corps Gras, 16:249 (1969). H. L. Rosano, J. Soc. Cosmet. Chem., 25:609 (1974). S. E. Friberg, in Microemulsions: Theory and Practice (L. M. Prince, ed.), Academic Press, New York, 1977, pp. 133-146. D. 0.Shah, V. K. Bansal,K.Chan,and W. C.Hsieh,inImprovedOil RecoverybySurfactantandPolymerFlooding (D. 0. ShahandR. S. Schechter, eds.), Academic Press, New York, 1977, pp. 293-337. L. MittalandB.Lindman, Th. F. Tadros,inSurfactantsinSolution(K. eds.), Vol. 111, Plenum Press, New York, 1984, pp. 1501-1532. A. M. Cazabat and D. Langevin, J. Chem. Phys., 74:314 (1981). L. E. Scriven, Nature, 263:123, (1976). L. E. Scriven, in Micellization, Solubilization and Microemulsions, Vol. 2 (K. L.Mittal, ed.), Plenum Press,New York, 1977, p. 877. Y. Talmon and S. Prager, Nature, 267:333 (1977). Y. Talmon and S. Prager, J. Chem. Phys., 69:2984 (1978). Y. Talmon and S. Prager, J. Chem. Phys., 76:1535 (1982). E. W. Kaler and S. Prager, J. Colloid InterfaceSci., 86:359 (1982). J. Jouffroy, P. Levinson, and P. G. De Gennes, J. Physiol. (Paris), 43:1241 (1982). E. W. Kaler, K. E. Bennet, H.T. Davis, and L. E. Scriven, J. Chem. Phys., 795673 (1983). P. A. Winsor, Trans. Faraday Soc., 44:376 (1984). L. S. Polatnik and A. I. Landau, Phase Equilibria in MulticomponentSystems, Holt, Rinehart and Winston,New York, 1964, p. 247. P. A. Winsor, Trans. Faraday Soc., 44,376,1984.

icroemulsions Emulsions and

533

384. P. A. Winsor, Solvent Propertiesof Amphiphilic Compounds, Butterworth. London, 1954. Clause, (eds.),MicroemulsionSystems,Vol.24, 385. H. L.RosanoandM. Surfactant Science Series, Marcel Dekker, New York, 1987. 386. M. Bourrel and R.S. Schechter, Microemulsions and Related Systems, Formulation, Solvency and Physical Properties, Vol. 30, Surfactant Science Series, Marcel Dekker, New York, 1988, p. 127. 387. A. Graciaa, J. Lachaise, A. Martinez, et al., C. R. Acad. Sci., B282547 (1976). 388. A. Graciaa, J.Lachaise,A.Martinez, et al., C. R. Acad. Sci., B285295 (1977). 389. A. A. Galje,W. G. M. Agterof, and A. Vrij, in Micellization, Solubilization and Microemulsions, Vol. 1 (K. L. Mittal. ed.), Plenum Press, NewYork, 1977. 390. C. Cabos andP. Delord, J. Appl. Crystallogr., 12502 (1979). 391. C. Cabos andP. Delord, J. Phys. Lett., 41:L455 (1980). 392. C. Taupin, J. P. Cotton, and R. Ober, J. Appl. Crystallogr., 11:613 (1978). 393. M. Dvolaitzky, M. Guyot, M. Lagues, J. P. Le Pesant, R. Ober, C. Sautery and C. Taupin, J. Phys. Chem., 69:3279 (1978). S. Huang, J. Phys.Chem.,86:3273 394. M.Kotiarchyk, S. H. Chen,andJ. (1982). 395. S. Qutubuddin, C. A. Miller, G. C.* Berry, T. Fort, Jr. and A. Hussam, in Surfactants in Solution, Vol. (K. 3 L. Mittal and B. Lindman, eds.), Plenum Press, New York, 1984. pp. 1693-1708. 396. A. Vrij, E. A. Nieuwenhuis, N. M. Fijnant, andW. G. H. Agterof, Faraday Discuss. Chem. Soc., 65:101(1978). 397. A. M. Cazabat, D. Langevin, and A. Pouchelon, J. Colloid Interface Sci., 73:1(1980). 398. C. Hermansky and R. A. Mckay, J. Colloid Interface Sci., 73:324 (1980). 399. E. Gulari, B. Bedwell, and S. Alkhafaji, J. Colloid Interface Sci., 77:202 (1980). 400. M. B. Mathews and E. Hirschhorn, J. Colloid Interface Sci., 8:86 (1953). 401. R. N. Hwan, Ph.D. Thesis, A Mechanism for Ultralow Interfacial Tension in Systems Containing Microemulsions: Theoretical Considerations and Experiments with Ultracentrifuge, Carnegie-Mellon University, Pittsburgh, 1978. 402. V. Luzzatti andF. Husson, J. Cell. Biol., 12:207, 1962. 403. Ekwall,P.,Composition,propertiesandstructuresofliquidcrystalline of amphiphilic compounds, in Advances in Liquid Crystals, phases in systems Vol. 1 (G. H. Brown, ed.), Academic Press, New York, 1975, p.1. 404. K.Larsson and P. Dejmek, Crystal and liquid crystal structures of lipids, in Food Emulsions, (K. Larsson andS. E. Friberg, eds.), Marcel Dekker,New York, 1990, p. 97. 405. K. Larsson, J. Phys. Chem., 73:7304,1989. 406. K. Larsson, J. Colloid Interface Sci., 113:299, 1986. 407. M. Trotta, M. R. Gasco, and S. Morel, J. Controll. Rel., 10:237 (1989).

Aserin 534

and

Garti

408. M. R. Gallarate, M. R. Gasco, and M. Trotta, Acta Pharm. Technol., 34:102 (1988). 409. M.R.Gasco,M. E. Carlotti, and M. Trotta, Int. J.Cosmet. Sci., 10:263 (1988). 410. M. R. Gallarate, M. R. Gasco, and G. Rua, Acta Pharm. Jugoslav. 40533 (1990). 411. M. R. Gasco, F. Pattarino, and I. Voltani, I1 Farmaco Ed. Pr., 43:3 (1988). 412. M. R. Gasco, S. Morel, and R. Mazoni, I1 Farmaco Ed. Pr., 43:373 (1988). 413. F. Pattarino, M. R. Gasco, and M. Trotta, Il Farmaco Ed. Pr.,44:339 (1989). 414. M. Trotta, M. R. Gasco, andF. Pattarino, Acta Pharm. Tech.36:226 (1990). 415. B. D. Kahan M. Ried, and J. Newburger, Transplant Proc., 15446 (1983). 416. B. Tam and S. H. Yalkowsky, Pharm. Res., 6:40 (1989). 417. W. A. Ritschel, S. Adolph, G . B. Ritschel and T. Schroeder, Meth. Find. Exp. Clin. Pharmacol., 12:127 (1990). 418. W. A. Ritschel, G. B. Ritschel, A. Sabouni, et al., Meth. Find. Exp. Clin. Pharmacol., 11:281(1989). 419. W. A. Ritschel, Meth. Find. Exp. Clin. Pharmacol., 13:205 (1991). 420. D. W. Osborne, A. J. I. Ward, and K. J. O’Neill, Drug Dev. Ind. Pharm., f 14:1203 (1988). 421. D. W. Osborne, A. J. I. Ward, and K. J. O’Neill, J. Pharm. Pharmacol., 43:451(1991). 422. F. FCvrier, M. F. Bobin, C. Lafforgue, and M. C. Martini, S.T.P. Pharm. Sci., 1:60 (1991). 423. J. Zeigenmeyer and C. Fuhrer, Acta Pharm. Technol., 26:273 (1980). 424. M. C. Martini, M.F. Bobin, H. Flandin, et al., Pharm. Belg.,39:348 (1984). 425. M. R. Gasco, M. Gallarate, and F. Pattarino, Int. J. Pharm., 69:193 (1991). 426. M. Trotta, M. R. Gasco, andF. Pattarino, Proceedings of 10th Pharmaceutical Technology Conf., Bologna, Italy,1991, p. 388. 427. M. R. Gasco, M. Gallarate, M. Trotta, et al., J. Pharm. Biomed. Anal., 7:433 (1989). 428. H. L. Rosano and W. E. Gerbacia,U.S. Patent, 3,778,381,1973. 429. P. Chabert, L. Foulletier, and A. Lantz, Ger. Offen., 2,452,513 (1975). 430. J.J.Delpuech,G.Mathis,J.C.Ravey,etal.,Bull.Soc.Chim.Fr., 578 (1985).

16 The Use of Drug-Loaded Nanoparticles in Cancer Chemotherapy Jean-Christophe Leroux,Eric Doelker, and Robert Gurny School ofPharmacy, University of Geneva, Geneva,Switzerland

I.

Introduction

11. In Vitro Uptake of Nanoparticles by Tumoral Cells A. Interaction of Uncoated Nanoparticles with Cultured Cells B. Interaction of Antibody-Coated Nanoparticles with Cultured Cells 111. Distribution and Pharmacokinetics of Anticancer Drugs Coupled to Nanoparticles A. Distribution of Anticancer Drugs Coupled to Nonmagnetic Nanoparticles B. Distribution of Anticancer Drugs Coupled to Magnetic Nanoparticles IV. V.

536 536 536 545 546 547 555

In Vivo Activity and Toxicity of Anticancer Drugs Coupled to Nanoparticles

557

Concluding Remarks

565

535

Leroux et al.

536 1.

INTRODUCTION

Currently, the limiting factor in cancer chemotherapy is the lack of selectivity of anticancer drugs toward neoplastic cells. Generally, rapidly proliferating cells such asthose of the bone marrow or the gastrointestinal tract are affected by the cytotoxic action of these drugs. This results in a narrow therapeutic index of most anticancer compounds. Furthermore, the emergence of resistant cell sublines during the chemotherapeutic treatmentmay require the use of higher doses of anticancer drugs or the elaboration of dosing protocols combining different anticancer drugs. This, in return, may to the toxicity and enhance thetoxicity of the treatment. In order decrease to enhance the selectivity of existing drugs, many drug-delivery systems have been developed ,in recent years [1,2]. These systems include soluble drug-polymer conjugates [3], nanoparticles [4], liposomes [ 5 ] , and microparticles [6]. Nanoparticles have received a growing interest for drug targeting, because they can beeasily prepared with well-definedbiodegradable polymers [7]. The reason of targeting tumors with nanoparticles is because certain neoplastic cells have been found to exhibit an enhanced endocytotic activity [ 8 ] . In addition, since some particular tumors are blood supplied by capillaries having an increased vascular permeability, one can anticipate that nanoparticles will gain access to extravascular tumoral cells [9,101.

This chapter covers the developments andprogress made in the deliv-

ery of anticancer drugs coupled to nanoparticles and theinteractions of the latter with neoplastic cells and tissues. Throughout the chapter, the term nunoparticles will generally refer to submicronic carriers ( 5 1 pm) consisting of either nanospheres (matrix systems) or nanocapsules (vesicular systems) [H]. Because many submicronic carriers havebeen in the past named microparticles, it is possible that some studies dealing with the present topic but using this term have been unintentionally omitted. II. IN VITRO UPTAKE OF NANOPARTICLES BY TUMORAL CELLS

The interaction of nanoparticles with cultured cells may be either passive or mediated via determined moieties such as monoclonalantibodies which can bindspecifically to targeted cells. A. Interaction of Uncoated Nanoparticles with Cultured Cells

In vitro studies involving nanoparticles were initially performed to determine if malignant cells were able to internalize nanoparticles to a signifi-

Drug-Loaded Nanoparticles

in Cancer Chemotherapy

537

cant extent (Table 1). However, targeting drug carriers to homogeneous cell cultures is of limited value, because in this case, the carriers are directly addedtothetarget cells. Intheseconditions, no anatomical barriers prevent the nanoparticles from acceding to target cells, and no extensive clearance of the drug is encountered [10,12]. Furthermore, it still remains unclear if an increase of drug efficiency in vivoresulting from the binding of the drug to nanoparticles is related to a greater uptake of the drug by tumoral cells in vitro [13-151. So far, in vitro experiments have led to inconsistent results regarding any increase of efficacy of nanoparticle-bound anticancer drugs. In fact, the free drug remained often more or as effective as its bound counterpart [16-191. These results may be explained by the limited access of the bound drug to the internal cell compartments. Indeed, Astier et al.[20] have reported that doxorubicin-loaded poly(alky1 methacrylate) (DXR PAMA) nanospheres were more effective in inhibiting human U-937 leukemia cell growth than free DXR. However, U-937cellsbelong toa monocytelike cell line with marked phagocytic properties. On the other hand, in vitro studies are useful to help to understand the mechanisms of action of bound anticancer drugs. By coupling covalently DXR to polyglutaraldehyde (PGL) nanospheres, TokCs et al. [l61 and Rogers et al. [l81 have demonstrated that DNA intercalationof DXR was not essential for its pharmacological action, and that the cell membrane was a major siteof action of these DXR nanospheres. Moreover,in vitro experimentshave revealed the roleof the nanoparticlepolymer in the cytotoxic action of drug-loaded nanoparticles. In another study where it was observed thatthe binding of DXRto poly(alky1 cyanoacrylate) (PACA) nanospheres could increase its efficiency [21], it was also found that the polymers used contributed to thecytotoxicity. Although, as stated by the investigators, the polymer itself could act as a second chemotherapeutic agent, it has to be pointed out thatcolloidal carriers have a propensity to concentrate in the reticuloendothelialsystem (RES) [7,10] and thus could possibly damage the latter, especially if they are loaded with highly cytotoxic agents [lo]. Selection of appropriate polymers remains a key issue, because it was shown that even well-established biodegradable polymers such as poly(1actic acid) (PLA) cannot be considered as inert vehicles [22,23]. Regarding PACA nanospheres, it was demonstrated that themost cytotoxic polymers for tumoral cells were those with shorter side chains [15], confirmingprevious results obtained with normal cells [24]. However, the evaluation of the polymer toxicity from in vitro tests must take into account the fact that the amount of nanospheres incubated with tumoral cells isoften excessively high'and sometimes corresponds to what isadministered to onemouse [15].

538

cl

2

.-C

U

a

B

C

.e

B E!

S

B m

m

2

Leroux et al.

cl

2

C

V

.d

B

g 3 n m W

S

v,

D

a

e

2

d

4

5:

3 5:

2

z

3

v)

Drug-Loaded Nanoparticlesin Cancer Chemotherapy

2 2

el

2 z

4 v,

z

v)

m W

5:

E 2

3 % 5 :s

539

540

0

a C L.

D

8

4a c 0

m m

2 d m

2

2

Leroux et al. mm

m m

a

2: N

z

V

m m m m

22


54 1

Leroux et al.

542

One of the most interesting properties of anticancer drug-loaded nanoparticles is their capacity for overcoming pleiotropic drug resistance (see Table 1). In vitro studies evaluating the efficacy of nanoparticles on multidrug-resistant cells have been so farperformed onlywith DXR [16,18,19,21,29,34,36,37]. Generally, the sensitivity of the resistant cell sublines was completely restored when DXR-loaded nanospheres were used. The mechanisms involved inpleiotropic drugresistance are complex. Among them, the overexpression of a membrane glycoprotein, namely glycoprotein P, which pumps out the anticancerdrug from theinside of the tumor cells, has been extensively investigated [2,42-441. By coupling DXR to nanospheres, DXR efflux from tumoral U-937 cells was considerably reduced (Fig. 1) [20,34]. Accordingly, increased drug retention, possibly DXR because of the inability of glycoprotein P to reject nanosphere-bound outside the cell, may partly explain how drug resistance can be counteracted. Recently, Colin de Verdibre et al. [38] demonstratedthat poly(isobutyl cyanoacrylate) (PIBCA) nanospheres were not endocytosed by P388 leukemic murine cells, and that the accumulation of doxorubicin in the resistant cells could not be explained, in this case, by a reduced efflux.

0

10

20

30

40

50

Incubation time after wash (h)

Fig. 1 Efflux of DXR from U-937 cells. Exponentiallygrowing cells were exposed for 4 h at 37°C to 0.5 pg/mL of free DXR (0)or nanosphere (NS) bound DXR (0)(4 x lo4 NSlcell). After they had been washed in phosphate-buffered saline (PBS), the cells were resuspended in drug-free medium for 48 h and harvested at various times. Each point represents the mean of two separate experiments performed in quadruplicate (bars, SD). (From ref. 20.)

Orug-Loaded Nanoparticles in

Cancer Chemotherapy

543

Indeed, the increase uptakemay be attributed to the formationof an ion pair between doxorubicin and soluble oligomers of PIBCA [45]. This complex would be able to betterdiffuse through the cell membrane. Another mechanism also seems to be involved in the increase efficiency of DXR nanospheres against DRX-resistant cells. DXR bound to nanospheres can efficiently interact with the membrane of resistant cells, inducing at thefinal stage a perforationof the cytoplasmic membrane [181. This mechanism is dependent on the nanospheredensity per cell, because for thesame drug concentration, theactivity of DXR increaseswith higher nanosphere densities[20]. This issue ispotentially important for the elaboration of improved in vivo drug-administration protocols. Indeed, the drug activity may be dependent not only on its dose but also on the amountof nanoparticles administered. More recently, immunomodulators that activatemacrophages to render them cytotoxic for tumor cells have attracted increasing interest for theirincorporationinto colloidal carriers.Immunomodulators such as muramyl dipeptide (MDP)have a very short half-life whichprevents accumulation of the drug intomacrophages [22]. Therefore, alipid derivative of MDP, namely muramyl dipeptide-L-alanyl-cholesterol(MTP-Chol), was incorporated intonanocapsules. These nanocapsules are made from an oily core surrounded by a polymeric wall and can be loaded with lipophilic drugs [41,46-481. MTP-Chol-loaded nanocapsules were able to activate rat alveolarmacrophages for a cytostatic effect on tumoralcells [33].However, the nanocapsule formulation was not superior to the nanoemulsion, micellar solution, and liposome formulations [33]. The possible mechanisms of action of MTP-Chol nanocapsules are depicted in Fig. 2. At low immunomodulator concentrations, MTF"Cho1 nanocapsules are generally more effective than free MDP, probably because of an improved intracellular delivery by phagocytosis [40,41]. Nanocapsules containing MTPChol were found to be slightly toxic for rat alveolar macrophages [22,33, 351. This toxicity couldbe explained by an increased sensitivity of activated macrophages to nitric oxide. Nitric oxide is produced by the immunomodulator during theactivation of the macrophages. It is a highly reactive molecule playing an important role in cytotoxic rat macrophages, but itis able to reduce the mitochondrial electron transport activity of macrophages. Since MTP-Chol nanocapsules are internalized by phagocytosis, which isan energy-dependentprocess, it is possible that the internalization of the immunomodulator by this mechanism could sensitize the macrophages to nitric oxide [22]. In recent years, there has been a growing interest for the biological therapies of cancer [49]. For instance, the insertion of new or foreigngenes into a tumoralcell may represent an alternative to conventional drug ther-

544

Leroux et a/.

MTPChol-loaded NC

0

MTPChol j

0+ Lysosome

I

MDP

"NO

MDP

YI

-b

Intracellular receptor

I Macrophage

NO-Synthase

]

Fig. 2 Different pathways by which encapsulated MTP-Chol or free hydrophilic MDP could enter rat alvelolar macrophages and induce an activation of these cells mediated by a release of NO: (1) Phagocytosis of intact nanocapsules (NC); (2) passive diffusion of released MTP-Chol; (3) uptake of MDP by receptor mediated endocytosis or pinocytosis. (From ref. 35.)

apy. Liposomes and nanoparticles have been proposed as alternative carriers toviral capsids for gene transfer [50,51]. Bertling et al. [51] have shown that itwas possible to incorporateefficiently DNA into PIBCA nanospheres and accordingly to protectDNA from degradation. However, the DNAwas not any longer transcriptionally active. Other studies are needed toassess the potentialityof different nanoparticletypes to increase the bioavailability of DNA to the cell nucleus, but at present time, liposomes appear to be better candidates. Promising results were recently obtained with antisense oligonucleotides adsorbed onpoly(isohexy1 cyanoacrylate) (PIHCA) nanowhen spheres [52]. It was demonstrated that the adsorbed oligonucleotides, administered locally, were able to inhibit the neoplasticgrowth of tumoral cells in nude mice. However, these preliminary results need to be further confirmed.

Drug-Loaded Nanoparticles in Cancer Chemotherapy

545

B. Interaction of Antibody-Coated Nanoparticles with Cultured Cells

To achieve tumor-specific targeting, nanoparticle formulations should be able to recognize specific cell determinants belonging only to target cells. The development of monoclonal antibody technologyby Kohlerand Milstein [53] has allowed the production of specific antibodies able to interact preferentially with tumor cell surface antigens [54,55]. From a purely pharmaceutical standpoint, the production of a drug-antibody complex is limited by the number of functional groups available per antibody molecule which can besuccessfully usedwithout significant lossof antibody activity [56]. Accordingly, a colloidal system in the form of nanoparticles with a large carrier capacity, coated with monoclonal antibodies, has been proposed as an alternative [57-591. Monoclonal antibodies (MAbs) can be fixed on nanoparticles either by nonspecific adsorption [57,58,60,61], specific adsorption [56,61-631, or covalent linkage [56,59,64-661. Nonspecific adsorption of MAbs on nanoparticles is realized by incubating the MAbs with the nanoparticles. Using this technique, it has been estimated that, theoretically, a maximum of 2000 MAbs can be adsorbed on 170 nm of poly(hexy1 cyanoacrylate) (PHCA) fixed nanospheres [57]. In vitro, the noncovalent link seems stable and the MAbs can recognize specifically tumor cells with a ratio of nonspecific binding to specific binding reaching 3:l-4:l [57,58]. These findings suggest that some MAbs are fixed to nanospheres by their Fc part with their antigen-binding site Fab protruding into the medium and thus being accessible to the antigens (Fig. 3). This type of binding can be achieved if the surface of the nanosphere is relatively hydrophobic [56].Although MAbs fixed by adsorption appear tightly bound to the nanoparticles, competitive displacement by plasma proteins can occur andmay represent a drawback following intravenous administration of MAb-coated nanoparticles [57,67]. Kubiak et al. [60] showed that radiolabeling of MAbs with iodine modified the physicochemical interactions between the MAbs and the nanospheres, resulting in adecreased affinity ofthe antibody for thecarrier. Therefore, it is of prime iinportance to assess the stability of the binding in biological fluids before conducting any in vivo experiments with labeled MAbs coupled to nanospheres. Apart from their possible use in drug targeting and radioimaging, Manil et al. [61] have suggested that nanospheres with adsorbed MAbs may be used as a solid phase for immunoradiometric assay for the analytical determination of tumor markers such as Q-fetoprotein. The specific adsorption of MAbs is achieved byfirst coupling to nanoparticles aligandwhich can specifically bind antibodies, Then,

Leroux et al.

546

Tumor (antigen)

\

Fa b

Antibody

Nanoparticle Fig. 3 Schematic diagram of the attachment of monoclonal antibody to a nanoparticle. (From ref. 57.)

the MAbs are incubated with the precoated nanoparticles and bind to the spacer molecule via noncovalent links. Some studies report the use of staphylococcal protein A as a spacer molecule which can bind the Fc portion of most subclasses of immunoglobulin G (IgG) [56,58,62,63,68]. AlthoughMAbs linked to protein A-precoated nanoparticles retain their specificity, it has been found in one study [61]that the presence of the protein A did not enhance the immunoreactivity of the adsorbed MAbs. Covalently bound MAbs cannot be displaced from the nanoparticle surface by plasma proteins and thus appear to be better candidates for efficient drug targeting. However, few studies have been performed in vitro with MAbs covalently linked to nanoparticle [56,64,66], and noneof them used tumoral cells. Suchexperiments have been carried out in vivo and will be discussed inSection II1.A along with the potential implications of the in vivo use of MAb-coated nanoparticles. 111. DISTRIBUTIONANDPHARMACOKINETICS OF ANTICANCER DRUGS COUPLED TO NANOPARTICLES

The localization of nanoparticles in specific tissues maybe dependent only on theintrinsic characteristics of the carrier (e.g., size, surface) or bepartly governed by an external mechanism (e.g., magnetic field). Therefore, this section makes a distinction between the natural distribution of a nonmagnetized carrier and thatof magnetically directed carrier.


547

A. Distribution of Anticancer Drugs Coupled to Nonmagnetic Nanoparticles The tissue distribution and pharmacokinetics of an anticancer drug can be altered by its incorporation into nanoparticles (Table 2). Generally, following intravenous administration, nanoparticles are rapidly and extensively taken up by the RES [7]. Accordingly, as soon as a few minutes after the injection of nanoparticles, the anticancer drug mainly accumulates in the liver and spleen [69-741. Thereafter,the RES drug level gradually decreases over several days depending on the biodegradability of the polymer and on the drug-release kinetics. Following intravenous injection of 14C-DXR PIHCA nanospheres, Verdun et al. [74] found that the liver concentration of radioactivity remained high during the first day and decreased rapidly during days 2 and 3 as a consequence of fecal excretion. After 8 days, only about 30% of the labeled drug could be detected in the liver. Indeed, PACA nanoparticles are biodegradable and accordingly are rapidly eliminated from the organism [4,75]. This is not the case of other nanoparticle polymers such as PAMA. PAMA nanoparticlesare well tolerated invivo but are veryslowly biodegradable [7,76]. Considering that some tumoral diseases may alter the RES, the long-term consequences of the accumulation of PAMA nanoparticles following multiple dosings of anticancer drugs should be seriously considered [ n ] . Rolland [76]suggested that the clinical use of DXR-loaded PAMA nanospheres would still be compatible with a monthly injection protocol. However, in our opinion, and because of the importance of the RES in the host defense [78,79], it seems questionable to justify the use of such polymers when other well-characterized biodegradable polymers are available. of liver DXR concentration Couvreur etal. [80] reported no increase following intravenousadministration of 3H-DXR-loaded poly(methy1 cyanoacrylate) (PMCA) nanospheres. In fact, an increase of DXR concentration in the gut and lungswas noticed. Although this unusual distribution pattern may be due to the higher hydrophilicityof PACA made of short side chains [81], these resultshave not been further exploited,probably because of the higher toxicity of PMCA nanospheres[24]. More recently, Bapat etal. [82] reported that DXR concentrations in the liver and spleen following intravenous injection of DXR-loaded poly(buty1 cyanoacrylate) (PBCA) nanospheres was lower than the in control and did not increase over time. They attributed these low concentrations to the slow breakdown of the polymer in the body. Nevertheless, they did not specify if their extraction procedure was able to separate the free drug from its nanosphere-bound counterpart. According to Kattanet al. [83], most ofconventional methods

548

2

M

n 2

U

C

S

U

9)

C

S

c

Leroux et al.

c

e

a . 3 0 9 )

2 2 2

8

2

.-?

.-4

S

.-? d

E

.-e ? .-

W

8 N

?

N

2

.d

W

8 d

s

2

W

C

2

W

W

2

m

2

W

%

€

.M

e


8

'5 l-

i3

d

S a

U

V

E

.0, 0

m

z v ) 2

W

m m

2

m

2

m

549

550

P l

i

I

!

2

W W

2

W W

.-e N

Leroux et al.

.-e

i ? g

Q

z .-?

?

.m

4

m Q)

?

.a

: E 0

2

B

V

.m

C

4 U

m

m


.-?

4

U

. I

2

B

8

S

B

m m

2

S? m

55 1

552

Leroux et al.

of drug determination do not discriminate between nanoparticle-associated drug and freedrug. Because nanoparticles are mainly taken upby the RES, nonphagocytic cells such as heart and musclecells could not theoretically accumulate nanoparticles. The incorporation of anticancer drugs into nanoparticles would thus appearsuitable for avoiding the deposition of highly toxic compounds intonontarget organs. DXR is a potent antitumoragent that is active against a wide spectrum of malignancies, but it is associated with acute and delayed cardiotoxicity [42]. When administered in the nanoparticulate form, the concentration of DXR in the mouse heart was generally lower thanin the control [74,82,89]. Surprisingly, when usingrabbit as the experimental animal model, Rolland [74] did not find any significant difference in tissue concentration between free and nanosphere-bound DXR in the spleen, lungs, and heart. He suggested that the bound drug could still produce a decrease of cardiac toxicity byits binding to othersubcellular compartments not involved in toxicity and by a slow release of thedrugfrom the nanospheres. In the same study, no differences were found between the pharmacokinetic parametersof the bound andfree DXR[73]. These findings are in disagreement with other studies, where the administration of DXR-loaded nanospheresresulted in a reduction of DXR blood clearance during the first fewminutes after intravenous injection [74,87]. The first clinical trial using anticancer drug-loaded nanospheres was carried out by Rolland in 1989 [76].He compared the pharmacokinetics of 20-30 mg PAMA nanosphere-bound DXR intravenously administered to four patients with hepatoma against 20 mg of free DXR administered to one patientwith hepatoma. This resulted in a 15-48% increase in the dosenormalized area under the drug concentration-time curve for the patients receiving the nanosphere-bound DXR [1,76]. More recently, Kattan et al. [S31 conducted a clinical trial using DXR-loaded PIHCA nanospheres.The pharmacokinetic profiles of DXR were examined on three patients with refractory solid tumors receiving 60-75 mg/mz DXR-loaded nanospheres. These profiles were comparedwith those obtained with the administration of free DXR at the same doseto thesame patients. Contrary to theprevious study, three controls with free DXR wereevaluated in this trial, and it was found thatthe variability of kinetic parameters of free and boundDXR was such that no convincing conclusions could be drawn from the results. Any other comparison of these two studies is difficult, because the plasma concentrations were monitored between0.5 and 24 h in the first study and .between 5 and 90 min inthe otherone. $Althoughit has been shown that some phagocytic cells may have a higher endocytotic activity than normal cells [8,30], it still remains unclear if nanoparticles can concentrate in tumors. According to Grislain et al.

Drug-Loaded Nanoparticles Cancer Chemotherapy in

553

[M], 4 h after intravenous administration of PBCA nanospheres to mice bearing a subcutaneous Lewis lung carcinoma, accumulation of the carrier in the tumoral tissue was higher than in the underlying tissue and the concentration of the nanospheres in the lungs was as much as five times higher than that of the liver. The investigators concluded that the deposition of the carrier in the lungs was possibly the consequence of nanosphere uptake by lung metastases of the primary tumor. However, as underlined by Douglas et al. [81], no histological examinations were performed in this study, and since the Lewis lung carcinoma is an hemorragic tumor that possesses poorly formed sinusoidal channels in which an endothelial lining is lacking [95],it is possible that the conditions encountered in this case were optimal for the uptake of nanospheres by tumoral cells. Other results on thedistribution of nanosphere-bound anticancer drugsto tumors are not as conclusive [70,86,88]. Even when high concentrations of bound drug were found in the tumor in comparison with the underlying tissue, the amount of nanospheres detected in the targeted area was still relatively low [88]. Chiannilkuchai et al. [l31 have administered DXR-loaded PIHCA nanospheres to mice bearing reticulosarcoma M5076 metastases. They found thatthe drug concentrated dramatically in the healthy hepatic tissue, which in turn could act as a reservoir for the anticancer drug (Fig. 4). This

0

10

20

30

40

Time (h)

Fig. 4 DXR concentration vs time curves in healthy hepatic(circles) and tumoral (squares) tissues after intravenous administration of DXR (10 mgkg corresponding to 133 mg/kg PIHCA) in its free (solid symbols) or nanosphere-bound form (open symbols). (From ref. 13.)

554

Leroux et al.

probably enabled a higher exposure of liver metastases to thedrug. These pharmacokinetic findings were consistent with histological examinations where a considerable number of nanospheres weredetected in the Kupffer cells, whereas nanospherescould not be foundin neoplastic cells [13]. Distribution of MAb-coated nanospheres havenot led to the results anticipated with previous in vitro experiments [59,67]. Apart from nanosphere clearance by the RES, poor localization of the nanospheres in the tumors could be explained by several factors, including competitive displacement of adsorbed MAbs, a secondary coating process where the MAb-coated nanospheres receive an opsonic layer, and most probablyan insufficient access of the nanospheres to tumor cells [10,67]. Maybe this latter issue has been so far underestimated. Initial enthusiasm for therapies using MAbs, drug-MAb complex, and MAb-macromolecular systems has been dampenedby the fact that macromolecules cannot reach all regions of a tumor in adequate quantity [3,55,96]. One method of increasing the transport of antibodies to tumoralcells could be, for instance, to use lower molecular weight agents; for example, antibody fragmentsF(ab), and Fab [96]. Nanoparticles are confronted to a greater problem in that they are much bigger than antibody molecules. Accordingly, it can be anticipated that the use of MAb-coupled nanospheresmay still have potential applications, but in a limited number of cases, such as nonsolid tumors, intratumoral therapy and for the treatment of solid tumors which are blood supplied by discontinuous capillaries or sinusoidal blood channels[lo]. Finally, the extensive clearance of nanoparticles and MAb-coated nanoparticles has to be lowered to achieve a better deposition of nanoparticles in tumors. Recent studies have demonstrated the efficacy of coating polystyrene nanospheres withblock copolymers of poloxamer and poloxamine and polyethylene glycol (PEG) for reducing nanosphere uptake by macrophages [97-1011 and allowing a longer in vivo circulation time of the nanospheres or their accumulation in thebone marrow [102,103]. Studies performed with Stealth@liposomes eitherstabilized with PEG organglioside GM1 haveshown a greater accumulationof the stabilized carrier over the control liposomesin tumors [104,105]. Such investigations need to be performed to a greater extent with biodegradable nanoparticles [106,107]. A study carried out by our group demonstrated that coating PLA nanospheres with PEG 20,000 could produce a substantial increase of the blood circulation time of the photoactivable compound ZnPcF,,, (hexadecafluorinated zinc phthalocyanine) ascompared with plain spheres (see Table 2) [94]. After 24 and 168 h, the cumulated uptake of the compound in the liver and spleen represented 61 and 4 4% for plain nanospheres versus 50 and 29% for PEG-coated nanospheres,respectively. The reduction of the uptakeof the nanospheresby the RES and theresult-



555

ing longer blood circulation time wereassociated with a three-fold increase of the compound concentration in the tumor after 24 h. B. Distribution of Anticancer Drugs Coupled to Magnetic Nanoparticles

In order to minimize colloidal camer uptake by the RES and to enhance their extravasation from the capillaries, magnetic nanospheres weredeveloped in the late 1970s and have been reviewed in detail by Gupta and Hung [108]. Only considerations related to their use in cancer chemotherapy will be discussed in this section. These nanospheres (200-1000 nm) contain a magnetically active component (e.g., Fe,O,) and can be guided by an externally placed electromagnet. An electromagnetic field ranging from 1000 to 8000 G is generally applied at the target site for 5-60 min after dosing [109-1171. Very highanticancer drug concentrations in the target area (generally a section of the tail) can be obtained following the administration of magnetic 60 min nanospheres. Forinstance, Senyei et al. [l101 showed that as late as postinjection, 0.05 mg/kg DXR administered intra-arterially via magnetic nanospheres resulted in almost twicethe local tissue concentration than was achieved by a 100-fold higher intravenous dose solution. The use of magnetic nanospheres canincrease the extravasation and partly overcome the limited access of conventional nanoparticles to extravascular tumors. Widder et al. [l111 demonstrated that30 min after infusion of albumin nanospheresto rats bearing subcutaneous Yoshida sarcoma, 1000-nm nanospheres were found in the extravascular compartment adjacent to tumorcells and occasionally in tumor cells. By 24and 72 h, nanospheres werestill detectable within tumor cells. Furthermore, it was shown elsewhere [l121 that DXR-loaded nanospheres could traverse the vascular endothelium asearly as 2 h afterdosing and that thepharmacodynamic characteristics of the drug were not altered by its entrapment intonanospheres. In order to evaluate the efficiencyof drug delivery via magnetic nanospheres, Galloet al. [l131 introduced two indices; namely, the relative exposure (re) and the targeting efficiency (t,):

where AUC, is the dose-normalized area under the drug concentrationtime curve (AUC) for the ith tissue and the superscripts NS and S refer to

Leroux et al.

556

the nanosphereand the solution dosage forms, respectively; (AUC),,,,, and (AUC)nontargct are the AUC for the target site and nontarget site, respectively. These two indices provide more valuable information on the drugdelivery than simple drug concentrations at thetarget site at differenttime points and should also be used for nonmagnetic nanospheres. Usingthese parameters, it wasfound thatDXR delivery to a selected target area can be greatly improved by its incorporation into magnetic nanospheres [113,114]. Magnetic nanospheres can also decrease the nontarget tissue exposure to as DXR. Forinstance, the administration of 2 mgkg DXR to ratsa solution or via magnetic albumin nanospheres provided revalues inferior to 0.6 for the heart and kidney aand reof 1.63 at thetarget site [1131.The efficiency of drug targeting is highly dependent of the strength of the magnetic field [114]. Table 3 suggests that the 1000-G magnetic field was only partly able to control the distribution of DXR, because the liver uptake of the drug remained high. considering a possible human application of magnetic nanospheres, this observation emphasizes the importance of defining the minimal magnetic field strength to apply to ensurean acceptable drug targeting. Moreover, it wasshown thatthepharmacokinetics of magnetic nanospheres could be altered by the dose administered [116,117]. Indeed, a reduction of the carrier dose has proved to increase the targeting efficiency. The reduction of the dose decreased moreparticularly the amount of DXR delivered to the liver and heart. As underlined by Gupta et al. [116], such a dose-dependence of kinetic parameters is important since the carrier dose may often need to be altered, depending on factorslike the weight of the patient and the volume of the target tissue.

Table 3. Drug Targeting Efficiency (tJ of Magnetic Albumin Nanoparticles in the Delivery of 0.4 mgkgDoxorubicin Administered Intra-arterially (Tail) to Wistar Rats inthe Presence or Absence of a 1000- or 8000-G Magnetic Field Applied at Target Site for 30 Min

. L.

Tissue ~~

1000

G

Control

8000 G

~~

Tail upstream from target site target site downstream from targetsite Liver Source: Adapted from ref. 114.

0.62 1.001.00

1.38 1.00 0.88

0.51

0.99

0.43

15.54 6.12 6.73



557

Finally, .it has to be pointed out that, in most cases, the target area was easily accessibleand thatsimple intratumoral injection of drug-loaded nanospheres could provide similar results. The magnetic carriers were also often administered via the arterial route. Althoughthis mode of administration allows the carrier to avoid the hepatic first pass effect, it suffers from a lack of clinical convenience. Furthermore, in the case of disseminated tumoral diseases or nonsuperficial tumors, it can be anticipated that such highly specifictissue distribution of anticancer drugs obtainedwith animals bearing subcutaneous tumors may not be as easily achievable. In the future, target areas other than themouse or rat tail should be more investigated. Ibrahim et al. [l181 showed that the kidneylevels of DXR could be enhanced by its incorporation into 220 nm-PIBCA magnetic nanospheres injected intravenously. However, in this study, the data werecollected only 10 min after dosing, providing no indication on thepossible retention of the carrier at thetarget site. IV. IN VIVO ACTIVITY AND TOXICITY OF ANTICANCER DRUGS COUPLEDTO NANOPARTICLES

Preliminary studies regarding the in vivo activity of PACA nanospheres against sarcoma implanted subcutaneously demonstratedthe efficiency of carried anticancer drugs in reducing the tumor size [24,86,119] (Table 4). The increased efficiency of the nanoparticle dosage form was first attributed to an improved delivery of the bound drug to malignant cells. [86,119]. Although this hypothesis may be relevant for some specific tumors [lo], the modification of the drug’s pharmacokinetic parameters, as discussed earlier, may be more important. Generally, an increase of life span was noted for the animals treated with nanoparticle-bound anticancer drugs. This prolonged survival was observed for animals bearing solid tumors as Interestingly, Cuvier et al. well as leukemia [24,34,84,120,122,127-1291. [34] demonstrated that it was possible using DXR-loaded nanospheres to prolong significantly the survival time of mice which were inoculated with DXR-resistant leukemiacells (Table5 ) . This bypass of tumor cell resistance confirmed the previous in vitro observations. Recently, Allkmann et al. [l361 showed that the photodynamic activity of ZnPcF,, against EMT-6 mouse mammary tumor could be greatly improved afterincorporation of the compound in PEG-coated nanospheres. One week after treatment, 63% of the mice had no macroscopic sign of tumor regrowth andat 3 weeks complete healing was observed for these mice. In contrast, with an O N emulsion formulation only 14% tumor regression was observed. In vitro investigations have revealed the importance of the binding of the drug to thecarrier on the activity of the nanoparticle suspension. A loss

,,

Q\

m

2

W W

B

W

B

W

2

0

5

5

vl

2

m m

d

2

1

.

W

2

v

l

2 4

m

W

Leroux et al.

562

Table 5. Effect of Free DXR or DXR-Loaded PIHCA Nanosphereson the Survival of DBA2 Mice Bearing P388-DXR-Resistant Ascitis(Intraperitoneal Injection of lo6 P388-DXR-Resistant Cells at Day0) 50% Survival time (days)

Groups of 10 mice Nontreated (C) DXR (TI DXR-loaded NS (T) 14 15 11 14

1 2

3 4

Mean T/C P (T vs C)

11 14 11 11 11.7

13.5

115%

Not significant

21 21 17 18 20.2

164% 0.003

NS, nanospheres. See text for definitions of other abbreviations. DXR or DXR-loadedNS was given byintraperitoneal route at days 1-4 at a dose of 2 mg/kg body weighttday; the activityof unloaded NS was not significant.P is the probability that Tis statistically different from C, using Student’s t-test. Source: From ref. 34.

of efficiency may occur if a mixture of unloaded nanoparticles and free drug is administered instead of the drug-loaded nanoparticle formulation [127,129]. This indicates that in vivo,the nanoparticle polymer is lesslikely to exerta cytostatic effect, as was noted in vitro (see Table 1). On the other hand, anticancer drugs bound to nanoparticles have often been associated with a higher acute toxicity;namely, premature deaths [86,107,119]. The increase of toxicity could be attributed to the accumulation of the carrier in the organs of the RES [86,119] or to the injection of agglomerated nanospheres that may have provoked embolisms [107]. This confirms the importance of developing drug carriers that can avoid the RES (when it is not the target site) and of making sure that the nanoparticle suspension is always fully dispersed before the injection. An increase of the bone marrow [l341 and renal toxicity [l351 has also been reported for doxorubicin-loaded nanospheres. Conversely, a reduction of the general toxicity of anticancer drugs boundto nanoparticles has beendemonstrated in a number of cases [17,24,123]. From these findings, one can infer that even if the incorporation of a cytostatic agent into nanoparticles is not always followed by an increase of drug efficiency, a reduction of the toxicity may produce an increase of the therapeuticindex [17]. One of the most promising applications of anticancer drug-loaded nanoparticles may be their usein the treatment of hepatic metastases

Drug-Loaded Nanopatticlesin Cancer Chemotherapy

563

200

T

150

T

T

50

No treatment

7.5

10.0

Dose (mrng)

Fig. 5 Number of hepatic metastases forC57/BL6 mice after treatmentat days 11 and 13 with intravenous free DXR (shaded columns) or nanosphere-bound DXR (closed columns) comparedwith control (open column). The mice were innoculated intravenously at day0 with 7 X Id tumoral cells (reticulum cell sarcoma M 5076). The study was performed with groups of 8-10 mice (mean 2 SD). **, significantly different from control (P C.001,Student’s t-test). (Adapted from ref. 129).

[128,129]. Administration of DXR-loaded PIHCA nanospheres to mice intravenously inoculated with tumoral cells resulted in a marked reduction of the number of hepatic metastases (Fig. 5 ) and in an increase of survival time (Fig. 6). These interesting results may be partly explained by the fact that in this study, the tumoral cells inoculated were macrophagic in origin [129]. Thus, these cells should have a propensity to take upcolloidal drug carriers. The activity of the immunomodulator MTP-Chol coupled to nanocapsules against metastases has also been evaluated. Because nanoparticles can incorporate peptides efficiently [l13 and have a propensity to accumulate in macrophages, MW-Chol-loaded nanocapsules could provide a stable and active pharmaceutical formulation. Indeed, MTP-Chol coupled to nanocapsules was partially able to inhibit the metastatic proliferation that normally occurs following the intravenous injection of tumoral cells. However, this formulation was efficient only when given as a prophylactic treatment [131,132,137]. According to Yu et al. [132], this would correspond in the clinic to patients undergoing surgery for a primary tumor and who could develop metastases. Furthermore, the efficacy of MTP-Chol-loaded nanocapsules is limited, and this formulation will certainly need to be combined to other chemotherapeutic drugsto ensure optimal results. It is well established that opsonization of nanoparticles

Leroux et al.

564

5

10

15

20

25 30 Time (d)

35

40

45

Fig. 6 Survivaltime of mice(bearinghepaticmetastases)untreated

(A) or

treated with 10 mglkg intravenously of free DXR (0) or PIHCA nanosphere-bound DXR (0).Single administrationon day 7 . (From ref. 129.)

by some specific plasma proteins enhances the uptake of the carrier by macrophages [101,138].Accordingly, one possible approach to increase the efficiency of encapsulated immunomodulators could be, for instance, to use nanoparticles precoated with opsonic proteins. Encouraging results have already beenobtained in vitro and invivowith protein-coated poly(L-lactic-co-glycolic acid) microspheres leading toan increase of macrophage activity toward tumoral cells [139]. Peroral administration of M”-Chol-loaded nanocapsules has led to inconsistent results, and further studies‘are needed to evaluate the potentialities of this route of administration [131]. A phase I study recently performed with DRX-loaded PIHCA nanospheres has shown that the nanosphere formulation seems to berelatively well tolerated [83]. Allergic reactions occurring at the beginning of the study were attenuated by increasing the infusion time from 10 to 60 minutes. Moreover, dextran 70, an adjuvant contained in the drug mixture, was suspected of being the causative agent of the remaining allergic episodes. Although unexpected side effects such as fever or bone pain were recorded, therewas no cardiac toxicity among the 18 patients evaluated. It was not possible to affirm that DXR-loaded nanospheres can completely avoid cardiotoxicity, because the number of patients was small and the maximal cumulative dose waslow(180mg/mz).Normally, the range of cumulative dose required to produce cardiotoxicity is 50-700 mg/m’ [42].

Drug-Loaded Nanoparficles


565

As for free DXR, the dose-limiting factor of the DXR-loaded nanospheres was neutropenia. Interesting results have been obtained with magnetic nanospheres [121,124,125]. The administration of anticancerdrug-loaded magnetic nanospheres to rats bearing subcutaneous Yoshida sarcoma (tail)was associated with a high rate of tumor remission. No deaths orwidespread metastases occurred in the nanosphere-treated groups. As discussed in Section III.B, such results could be explained partly by the facilitated accessibility of the tumor to the magnetic field and by the intra-arterial injectionof the magnetic nanospheres. Amongthestudies involving nanoparticles coupled toanticancer drugs, some important issues have sometimes been neglected. In many cases,thetreatment was initiated atthesamemomentor few hours following the tumor cells inoculation [84,86,107,120,122]. According to Poste [140], the implantation of tumor cells into host tissue disrupts the local microvasculature and enhances the vascular permeability at the injection site. Vascular repair is generally achieved within 2-3 days. Premature injection of nanoparticles may result in unusually high amounts of cytostaticagent in the tumor. Accordingly, positive resultsobtained with nanoparticles injected much later after the inoculation of tumoral cells providemorereliable information (see Table 4). Furthermore,transplanted tumors growing subcutaneously fail to provide a sufficiently demanding model for evaiuating the effectiveness of agents in,treating metastases [1401. In some studies (see Table 4), failures in demonstrating any difference between the activity of a nanoparticle drug formulation over the solution dosage form may be caused by the presence of a too high amount of unbound anticancerdrug in the nanoparticlesuspension. In vivo, administration of a nanoparticle suspension containing some unbound drugmay be of little consequence when the entrapmentefficiency of the preparation procedure is high. But, when the proportion of the unbound drug in the preparation reaches, for instance,85-92% of the total administered dose [107], it seems unlikely to anticipate important variationsof the activity of the nanoparticlesuspension from the solution formulation. V.

CONCLUDING REMARKS

Over the last 20 years, considerable progress has been made in the preparation of well-characterized nanoparticle formulations loadedwith a variety of anticancer agents [7]. Despite this considerable amount of work, still little is known about the type of neoplasia which may respond beneficially to a nanoparticulate dosage form. More in vitro and in vivo experiments

566

Leroux et al.

are needed to determine the best conditions requiring the administration of anticancer drug-loaded nanoparticles. It seems possible that nanoparticles will have interesting applications only in limited tumors such as those of the mononuclear phagocyte system (e.g., monocytic leukemia, hairy cell leukemia) or for activating macrophages’ tumoricidal properties [141]. Increased efficiency of nanoparticle formulations against some drug-resistant leukemia and strong activity against hepatic metastasesare examples which illustrate thepossible applications of nanoparticles in the future. Nanoparticles may also be used for sustained-release delivery of cytostatic agents in some specific circumstances. Magnetic targeting isan efficient but expensive and specialized technical approach. The possible accumulation of the Fe,O, in the body will probably restrict its use to the treatmentof severe superficial malignancies that are refractory to other treatments [142]. The extensive uptake of the nanoparticles by the RES and their limited access to extravascular tumors are major obstacles to an efficient drug targeting. Surface modifications of drug-loaded biodegradable nanoparticles mayincrease the half-life of the carrier and/or allow its accumulation in a wider variety of target areas. This latter approachmay open new attractive perspectives in the chemotherapy of cancer. REFERENCES 1. P. K. Gupta, Drug targeting in cancer chemotherapy: A clinical perspective, J.

Pharm. Sci., 79:949 (1990). 2. R. K. Y. &e-Cheng and C. C. Cheng, Delivery of anticancer drugs, Methods Find. Exp. Clin. Pharmacol., 11:439 (1989). 3. R. Duncan, Drug-polymer conjugates: potential for improved chemotherapy, Anti-Cancer Drugs, 3:175 (1992). 4. P. Couvreur,Polyalkylcyanoacrylatesascolloidaldrugcarriers,Crit.Rev. Ther. Drug Carrier Syst.,5:l (1988). 5. P. Couvreur,Liposomesencliniquehumaine:lepoint sur lesrksultats thkrapeutiques, in Les Liposomes (J.Delattre, P. Couvreur, F. Puisieux, et al., eds.), Inserm, Paris, 1993, p. 214. 6. P.M.J.Flandroy,C.Grandfils,and R. J. JkrBme,Clinicalapplications of microspheres in embolization and chemoembolization: A comprehensive review and perspectives, in Pharmaceutical Particulate Carriers. Therapeutic Applications (A. Rolland, ed.), Marcel Dekker,New York, 1993, p. 321. 7. E. Allkmann,R.Gurny,and E. Doelker,DrugloadednanoparticlesPreparation methods and drug targeting issues, Eur. J. Pharm. Biopharm., 39:173 (1993). 8. J. Kreuter, Evaluation of nanoparticles as drug-delivery systems 111: Materials, stability, toxicity, possibilities of targeting, and use, Pharm. Acta Helv., 58:242 (1983).



567

9. R. K. Jain, Transport of molecules across tumor vasculature, Cancer Metastasis Rev., 6559 (1987). of 10. G. Poste, Drug targeting in cancer therapy, Receptor-mediated Targeting Drugs (G. Gregoriadis, G. Poste, J. Senior, and A. Trouet, eds.), Plenum Press, New York, 1984, p. 427. 11. P. Couvreur and F. Puisieux, Nano- and microparticles for the delivery of polypeptides and proteins, Adv. Drug Deliv. Syst., 10:141 (1993). 12. W.T. Bellamy, Prediction of response to drug therapy. A reviewof in vitro assays, Drugs, 44:690 (1992). 13. N. Chiannilkulchai, N. Ammoury, B. Caillou, et al., Hepatic tissue distribution ofdoxorubicin-loadednanoparticlesafteri.v.administrationinreticulomaM5076metastasis-bearingmice,CancerChemother.Pharmacol., 26: 122 (1990). 14. S. Maassen, E. Fattal, R. H. Muller, and P. Couvreur, Cell cultures for the assessmentoftoxicityanduptakeofpolymericparticulatedrugcarriers, S.T.P. Pharma Sci., 3:11 (1993). 15. E. M. Gipps, P. Groscurth, J. Kreuter,and P.P. Speiser,Theeffects of polyalkylcyanoacrylate nanoparticles on human normal and malignant mesenchymal cells in vitro, Int.J. Pharm., 40:23 (1987). 16. Z. A. Tokts, K. E. Rogers,andA.Rembaum,Synthesisofadriamycincoupled polyglutaraldehyde microspheres and evaluation of their cytostatic activity, Proc. Natl. Acad. Sci.USA, 79:2026 (1982). 17. R. C. Oppenheim,E. M. Gipps, J. F. Forbes, and R. H. Whitehead, Development and testing of proteinaceous nanoparticles containing cytotoxics, in Microspheres and Drug Therapy (S. S. Davis, L. Illum, J. G. McVie, and E. Tomlinson, eds.), Elsevier, Amsterdam, 1984, p. 117. 18. K. E. Rogers,B. I. Carr, andZ. A. TokBs, Cell surface-mediated cytotoxicity ofpolymer-boundadriamycinagainstdrug-resistanthepatocytes,Cancer Res., 43:2741 (1983). 19. L. Treupel, M. F. Poupon, P. Couvreur, and F. Puisieux, Vectorisation de la doxorubicine dans des nanosphtres et rkversion de la rksistance plCiotropique des cellules tumorales, C.R. Acad.Sci. Paris, 313:171(1991). 20, A. Astier, B. Doat, M.J. Ferrer, et al., Enhancement of adriamycin antitumor activity by its binding with an intracellular sustained-release form, polymethacrylate nanospheres, in U-937 cells, Cancer Res., 48:1835 (1988). 21, C. Kubiak, P. Couvreur, L. Manil, and B. Clausse, Increased cytotoxicity of nanoparticle-camed adriamycin in vitro and potentiation by verapamil and amiodarone, Biomaterials, 10553 (1989). 22. C. Morin, G. Barratt, H. Fessi, et al., Nitric oxide synthesis in rat alveolar macrophages is stimulated by an encapsulated immunomodulator, Proc. 6th Int. Conf. Pharm. Technol., 550 (1992). 23. A. Smith andI. M. Hunneyball, Evaluation of poly(1actic acid) as a biodegradable drug delivery system for parenteral administration,J. Pharm., Int. 30:215 (1986). 24.P. Couvreur, L. Grislain, V. Lenaerts, et al., Biodegradable polymeric nano-

568

25. 26. 27. 28. 29.

30.

31. 32. 33. 34. 35. 36.

37.

Leroux et al.

particles as drugcamer for antitumor agents, in Polymeric Nanoparticles and Microspheres (P. Guiot andP. Couvreur, eds), CRC Press, Boca Raton, 1986, p. 27. P. A. Kramer and T. Burnstein, Phagocytosis of microspheres containing an anticancer agent by tumor cells in vitro, Life Sci., 19515 (1976). R. C. Oppenheim and N. F. Stewart, The manufacture and tumour cell uptake of nanoparticleslabelledwithfluoresceinisothiocyanate,Drug.Dev.Ind. Pharm., 5563 (1979). Z . A. Tokb, K. E. Rogers,A.M.Daniels,and J. R. Daniels, Increased cytotoxic effects by polymer-bound adriamycin are mediated through the cell surface, Proc.Am. Assoc. Cancer Res., 24:255 (1983). P. Guiotand P. Couvreur,Quantitativestudy of theinteractionbetween polybutylcyanoacrylate nanoparticles and mouse peritoneal macrophages in culture, J. Pharm. Belg., 38:130 (1983). Z.A. Tokts, K. L. Ross, andK. E. Rogers, Use of microspheres to direct the cytotoxic actionof adriamycin to the cell surface, in Microspheres and Drug Therapy (S. S. Davis,L.Illum, J. G.McVie,and E. Tomlinson,eds.), Elsevier, Amsterdam, 1984, p. 139. J. M. BtguC, A. Rolland, G. Merdrignac, et al., Etude de la ptnttration de nanoparticules de polymtthacrylate dans des cultures primaires d’hbpatocytes de rat et d’homme et des lignCes cellulaires d‘htpatomes: application A la vectorisation d’agents cytostatiques utilisCs en chimiothtrapie anticanckreuse, Arch. Int. Physiol. Biochim. 96:A413 (1988). A. Astier, G. Benoit, B. Doat, et al., Activitt antitumorale de l’adriamycine J. lite B des nanoparticules sur 2 ligntes cellulaires canctreuses humaines, Pharm. Clin., 7:43 (1988). A.Rolland, J. M. Btgut, R. Le Verge,andA.Guillouzo,Increaseof doxorubicin penetration in cultured rat hepatocytes by its binding to polymethacrylic nanoparticles, Int.J. Pharm., 53:67 (1989). G. M. Barrat, W. P.Yu, H. Fessi, et al.,DeliveryofMDP-L-alanylcholesterol to macrophages: Comparison of liposomes and nanocapsales, Cancer J., 2:439 (1989). C. Cuvier, L. Roblot-Treupel,J. M. Millot, et al., Doxorubicin-loaded nanospheres bypass tumor cell multidrug resistance, Biochem. Pharmacol., 44509 (1992). C. Morin,G . Barratt, H. Fessi, et al., Biodegradable nanocapsules containing a lipophilic immunomodulator: drug retention and tolerance towards macrophages in vitro,J. Drug Targeting, 1:157 (1993). F. Nemati, A. C. De Verditre, C. Dubernet, et al., Cytotoxicity and uptake of free doxorubicin and doxorubicin-loaded nanoparticles in sensitive and multidrug-resistantleukaemicmurinecells,Proc.6thInt.Conf.Pharm. Technol., 3:93 (1992). F. NCmati, C. Dubernet, A. C. De Verditre, et al., Some parameters influencing cytotoxicity of free doxorubicin and doxorubicin-loaded nanoparticles in sensitive and multidrug resistant leucemic murine cells: incubation time, num ber of nanoparticles per cell, Int.J. Pharm., 10255 (1994).


569

38. A. C. De VerdiBre, C. Dubernet, F. Nemati, et al., Uptake of doxorubicin from loaded nanoparticles in multidrug-resistant leukemic murine cells, Cancer Chemother. Pharmacol.,33504 (1994). 39. S. Bennis, C. Chapey, P. Couvreur, and J. Robert, Enhanced cytotoxicity of doxorubicin encapsulated in polyisohexylcyanoacrylate nanospheres. against multidrug-resistant tumour cells in culture, Eur. J. Cancer, 30A:89 (1994). 40. C. Morin, G. Barratt, H. Fessi, et al., Improved intracellular delivery ofa J. Immunopharmuramyl dipeptide analog by means of nanocapsules, Int. macol., 16:451 (1994). 41. I. Seyler, C. Morin, G. Barratt, et al., Induction of macrophage NO-synthase by an immunomodulator entrapped within polymeric nanocapsules, Eur. J. Pharm. Biopharm., 41:49 (1995). 42.P. A. J. Speth, Q. G.C.M.VanHoesel,andC.Haanen,Clinicalpharmacokinetics of doxorubicin, Clin. Pharmacokinet., 1515 (1988). 43.V.Ling,MultidrugresistanceandP-glycoproteinexpression,Ann.N.Y. Acad. Sci., 507:7 (1987). 44. T. Ban, Pleiotropic, multidrug-resistant phenotype and P-glycoprotein:A review, Chemotherapy, 38:191 (1992). F. Nemati,C. De VerdiBre, et al.,Mechanism of actionof 45.C.Dubernet, polyalkylcyanoacrylate doxorubicin-loaded nanoparticles in the reversion of resistance of P388/ADR, Proc. 1st World Meeting APGI/APV, 1:473 (1995). 46. N.A1 Khouri Fallouh, L. Roblot-Treupel, H. Fessi, et al., Development of a new process for the manufacture ofpolyisobutylcyanoacrylatenanocapsules, Int. J. Pharm., 28:125 (1986). 47. S. Bonduelle, C. Foucher, J. C. Leroux, et al., Association of cyclosporin to isohexylcyanoacrylate nanospheres and subsequent release in human plasma in vitro, J. Microencaps., 9:173 (1992). 48. A. Labib, V. Lenaerts, F. Chouinard, et al., Biodegradable nanospheres containing phtalocyanines and naphthalocyanines for targeted photodynamic tumor therapy, Pharm. Res., 8:1027 (1991). 49. B.A.Chabner,Biologicalbasisforcancertreatment,Ann.Intern.Med., 118:633 (1993). 50. D. Lasic, Liposomes, Am. Sci., 80:20 (1992). 51.W.M. Bertling,M.Gareis, V. Paspaleeva, et al.,Useofliposomes,viral capsids,andnanoparticlesasDNAcarriers,Biotechnol.Appl.Biochem., 13:390 (1991). 52. G. Schwab,C.Chavany, I. Duroux, et al.,Antisenseoligonucleotidesadsorbed to polyalkylcyanoacrylatenanoparticlesspecificallyinhibitmutated Ha-ras-mediated cell proliferation and tumorigenicity in nude mice, Proc. Natl. Acad. Sci. USA, 91:10460 (1994). 53. G. Kohler and C. Milstein, Continuous cultures of fused cells secreting antibody of predefined specificity, Nature, 256:495 (1975). vie, Immunoconjugates:applicationsintargeteddrug 54.B.P.RamandP. delivery for cancer therapy, Pharm. Res., 4:181 (1987). 55. C. Kosmas, H. Linardou, and A. Epenetos, Review: advances in monoclonal antibody tumour targeting,J. Drug Targeting,1:81 (1993).

56. L. Illum and P. D. E. Jones, Attachment of monoclonal antibodies to microspheres, Methods Enzymol., 112:67 (1985). 57. L. Illum, P. D. E. Jones, J. Kreuter, et al., Adsorption of monoclonal antibodies to polyhexylcyanoacrylatenanoparticlesandsubsequentimmunospecific binding to tumour cells in vitro, Int. J. Pharm., 17:65 (1983). 58. P. Couvreur and J. Aubry, Monoclonal antibodies for the targeting of drugs: application to nanoparticles,inTopics in PharmaceuticalSciences(D.D. Breimer and P. Speiser, eds.), Elsevier, Amsterdam, 1983, p. 305. 59. Y. Akasaka, H. Ueda, K. Takayama, et al., Preparation and evaluation of bovine serum albumin nanospheres coated with monoclonal antibodies, Drug Design Deliv., 3:85 (1988). 60. C. Kubiak, L. Manil, and P. Couvreur, Sorptive propertiesof antibodies onto cyanoacrylic nanoparticles, Int. J. Pharm., 41:181 (1988). 61. L. Manil, L. Roblot-Treupel, and P. Couvreur, Isobutyl cyanoacrylate nanoparticlesasasolidphaseforanefficientimmunoradiometricassay,Biomaterials, 7:212 (1986). 62. K. J. Widder, A. E. Senyei, H. Ovadia, and P. Y. Paterson, Magnetic proteinAmicrospheres:arapidmethodforcellseparation,Clin.Immunol. Immunopathol., 14:395 (1979). 63. K.J.Widder, A. E. Senyei, H. Ovadia,and P. Y. Paterson,Specificcell binding using staphylococcal protein A magnetic microspheres, J. Pharm. Sci., 70:387 (1981). 6 4 . M. R. Kaplan, E. Calef, T. Bercovici, and C. Gitler, The selective detection of cell surface determinants by means of antibodies and acetylated avidin attached to highlyfluorescentpolymermicrospheres,Biochim.Biophys. Acta, 728:112 (1983). Le Verge, andB. Genetet, Anew immunoreagent 65. D. Bourel, A. Rolland, R. for cell labeling-CD3 monoclonal antibody covalently coupled to fluorescent polymethacrylic nanoparticles, J. Immunol. Methods, 106:161 (1988). Le Verge, Monoclonal antibodies 66. A. Rolland, D. Bourel, B. Genetet, and R. covalently coupled to polymethacrylic nanoparticles: in vitro specific targeting to humanT lymphocytes, Int. J. Pharm., 39:173 (1987). 67. L. Illum, P. D. E. Jones, R. W. Baldwin, and S. S. Davis, Tissue distribution of poly(hexy1 2-cyanoacrylate) nanoparticles coated with monoclonal antibodies in mice bearing human tumor xenografts, J. Pharmacol. Exp. Ther., 230:733 (1984). X. 68. T. Laakso, J. Anderson, P. Artursson, et al., Acrylic microspheres in vivo. Elimination of circulating cells by active targeting using specific monoclonal antibodies boundto microparticles, Life Sci., 38:183 (1986). Y .Kato, Drug-camer property of 69. K. Sugibayashi,Y. Morimoto, T. Nadai, and albumin microspheres in chemotherapy. I. Tissue distribution of microsphereentrapped 5-fluorouracil in mice, Chem. Pharm. Bull., 25:3433 (1977). of albumin 70. K. Sugibayashi,Y. Morimoto, T. Nadai, et al., Drug-carrier property microspheres in chemotherapy. 11. Preparation and tissue distribution in mice of microsDhere-entraDDed 5-fluorouracil.Chem. Pharm. Bull., 27:204 (1979).



571

71. T. Yoshioka, M. Hashida, S. Muranishi, and H. Sezaki, Specific delivery of

mitomycin Cto the liver, spleen and lung: nano- and microspherical ofcarriers gelatin, Int. J. Pharm., 81:131 (1981). 72. B. Kante, P. Couvreur, V. Lenaerts, et al., Tissue distribution of [3H]actinomycin D adsorbed on polybutylcyanoacrylate nanoparticles, Int. J. Pharm., 7:45 (1980). 73. A. Rolland, Pharmacokinetics and tissue distributionof doxorubicin-loaded polymethacrylic nanoparticles in rabbits, Int.J. Pharm., 42:145 (1988). of doxorubicin 74. C. Verdun, F. Brasseur, H. Vranckx, et al., Tissue distribution 75. 76. 77. 78. 79.

80.

associated with polyisohexylcyanoacrylate nanoparticles, Cancer Chemother. Pharmacol., 26:13 (1990). P. Couvreur and C. Vauthier, Polyalkylcyanoacrylate nanoparticles as drug J. Control. Rel., 17:187 (1991). carrier: present state and perspectives, A. Rolland, Clinical pharmacokinetics of doxorubicin in hepatoma patients after a singleintravenousinjection of free or nanoparticle-bound anthracycline, Int. J. Pharm., 54;113 (1989). J. E. O'Mullane, P. Artursson, and E. Tomlinson, Biopharmaceutics of microparticulate drug carriers, Ann. N.Y. Acad. Sci., 507:120 (1987). K. J. Widder, A. E. Senyei, and D. F. Ranney, Magnetically responsive microspheresandothercarriersforthebiophysicaltargeting of antitumor agents, Adv. Pharmacol. Chemother., 16:213 (1979). I. J. Fidler, Macrophages and metastases. A biological approach to cancer therapy, Cancer Res.,45:4714 (1985). P. Couvreur, B. Kante, V. Lenaerts, et al., ,Tissue distributionof antitumor J. Pharm. Sci., drugs associated with polyalkylcyanoacrylate nanoparticles,

69:199 (1980). S. J. Douglas, S. S. Davis, and L. Illum, Nanoparticles in drug delivery, Crit. Rev. Ther. Drug Carrier Syst., 3:233 (1987). of colloidal carrierson the disposition and 82. N. Bapat and M. Boroujerdi, Effect

81.

83. 84.

85. 86.

87.

tissue uptakeof doxorubicin: 11. Conjugation with isobutylcyanoacrylate nano' particles, Drug. Dev. Ind. Pharm., 19:2667 (1993). J. Kattan, J. P. Droz, P. Couvreur, et al., Phase I clinical trial and pharmacokinetic evaluation of doxorubicin carried bypolyisohexylcyanoacrylate nanoparticles, Invest. New Drugs, 10:191 (1992). Y. Morimoto, M. Akimoto, K. Sugibayashi, et al., Drug-carrier property of albumin microspheres in chemotherapy. IV. Antitumor effect of single-shot or multiple-shot administration of microsphere-entrapped 5-fluorouracilon Ehrlich ascites or solid tumor in mice, Chem. Pharm. Bull., 28:3087 (1980). L.Grislain, P. Couvreur, V. Lenaerts, et al., Pharmacokinetics and distribution of a biodegradable drug-carrier, Int.J. Pharm., 15:335 (1983). J. Kreuter and H. R. Hartmann, Comparative study on the cytostatic effects andthetissuedistributionof5-fluorouracilinafreeformandbound to polybutylcyanoacrylate nanoparticles in sarcoma 180-bearing mice, Oncology, 40:363 (1983).

L. Grislain,. P. Couvreur,andM.Roland,Modification

de lapharma-

572

Leroux et a/.

cocinttique de moltcules associbes aux nanoparticles, S.T.P. Pharma, 1:1038 (1985). J. Kreuter, et al.,Distribution ofpolyhexyl 88. E. M.Gipps,R.Arshady, cyanoacrylate nanoparticles in nude mice bearing human osteosarcoma, J. Pharm. Sci., 75256 (1986). 89. C. Verdun, P. Couvreur, H. Vranckx, et al., Development of a nanoparticle controlled-release formulation for human use, J. Control. 3:205 Rel., (1986). 90. M. Simeonova, T. Ivanova, E. Raikova, et al., Tissue distribution of polybutylcyanoacrylate nanoparticles carrying spin-labelled nitrosourea, Int. J. Pharm., 43:267 (1988). 91. G. Mukherji, R.S. R. Murthy, and B. D. Miglani, Preparation and evaluation ofpolyglutaraldehydenanoparticlescontaining5-fluorouracil, Int. J. Pharm., 50:15 (1989). 92. G. Mukherji, R. S. R. Murthy, and B. D. Miglani, Preparation and evaluation of cellulose nanospheres containing 5-fluorouracil, Int. J. Pharm., 65:l (1990). 93. V. Lenaerts, A. Labib, F. Chouinard, et al., Nanocapsules with a reduced liveruptake:targeting ofphthalocyaninestoEMT-6mousemammary tumour in vivo, Eur. J. Pharm. Biopharm.,41:38 (1995). 94. E. AllBmann, N.Brasseur, 0. Benrezzak, et al., PEG-coated poly(1actic acid) nanoparticles for the delivery of hexadecafluoro zinc phthalocyanine to EMT6 mouse mammary tumours,J. Pharm. Pharmacol., 47:382-387 (1995). 95. S. E. James and A. J. Salsbury, Effect of (~)-1,2-bis(3,5-dioxopiperazin-ly1)propane on tumor blood vessels and its relationshipto the antimetastatic 342339 (1974). effect in the Lewis lung carcinoma, Cancer Res., 96. R. K. Jain, Physiological barriers to delivery of monoclonal antibodies and other macromolecules in tumors, Cancer Res., 50:814s (1990). 97. S. E. Dum, A. Brindley, M. C. Davies, et al., Studies on in-vitro uptake by Kupffer cells and the in-vivo biodistribution of a range of novel polymeric colloids, J. Pharm. Pharmacol., 44(Suppl.):1082(1992). 98. S. Rudt and R. H. Muller, In vitro phagocytosis assay of nano- and microparticles by chemiluminescence. 11. Effect of surface modificationcoatby ing of particles with poloxameron the phagocytic uptake, J. Control. Rel., 2551 (1993). 99. S. Rudt, H. Wesemeyer, and R. H. Muller, In vitro phagocytosis assay of IV.Effect of surfacemodifinano- and microparticles by chemiluminescence. cation by the coating particles with poloxamine and Antarox CO on the phagocytic uptake, J. Control. Rel., 25:123 (1993). 100. J. C. Leroux, F. De Jaeghere, B. M. Anner, et al., An investigation on the role of plasma and serum opsonins on the internalization of biodegradable poly(D,L-lactic acid) nanoparticles by human monocytes, Life Sci., 57;695 (1995). 101. J. C. Leroux, P. Gravel, L. Balant, et al., Internalizationof poly(D,L-lactic acid) nanoparticles by isolated human leukocytes and analysis of plasma 28:471 (1994). proteins adsorbed onto the particles, J Biomed. Mater. Res.,

7:85



573

102. S. D. Troster and J. Kreuter,Influenceofthesurfacepropertiesoflow contact angle surfactants on the body distribution of 14C-poly(methyl methacrylate) nanoparticles,J. Microencaps., 9:19 (1992). 103. L. Illum andS. S.Davis, Targetingof colloidal particlesto the bone marrow, Life Sci., 40:1553 (1987). 104. D. Papahadjopoulos, T.M. Allen, A. Gabizon, et al., Sterically stabilized liposomes: improvement in pharmacokinetics and antitumor therapeutic efficacy, Proc. Natl. Acad. Sci. USA, 88:11460 (1991). 105. S. Unezaki, K. Maruyama, 0. Ishida, et al., Enhanced tumor targeting of J. Drug doxorubicin by ganglioside GM1-bearing long-circulating liposomes, Targeting, 1:287 (1993). 106. S. J. Douglas, S. S. Davis, and L. Illum, Biodistribution of poly(buty1 2cyanoacrylate) nanoparticles in rabbits, Int.J. Pharm., 34:145 (1986). 107. P. Beck, J. Kreuter,R.Reszka,and I. Fichtner,Influenceofpolybutylcyanoacrylate nanoparticles and liposomes on the efficacy and toxicity ofthe J. Microencaps., anticancer drug mitoxantrone in murine tumour models, 1O:lOl (1993). 108. P. K. Gupta and C. T. Hung, Magnetically controlled targeted micro-camer systems, Life Sci., 44:175 (1989). 109. K. J. Widder, A. E. Senyei, and D. G. Scarpelli, Magnetic microspheres: a model system for site specific drug delivery in vivo, Proceed Soc. Exp. Biol. Med., 58:141 (1978). 110. A. E. Senyei, S. D. Reich, C. Gonczy, and K. J. Widder, In vivo kinetics of magnetically targeted low-dose doxorubicin, J. Pharm. Sci., 70:389 (1981). 111. K. J. Widder, P. A. Marino, R. M. Morris,et al., Selective targeting of magnetic albumin microspheresto the Yoshida sarcoma: Ultrastructural evaluation of microsphere disposition, Eur. J. Cancer Clin. Oncol., 19:141 (1983). 112. P. K. Gupta,C. T. Hungand N. S. Rao,Ultrastructuraldisposition of adriamycin-associated magnetic albumin microspheres in rats, J. Pharm. Sci. 78:290 (1989). 113. J. M. Gallo, P. K. Gupta, C. T. Hung, and D. G. Perrier, Evaluation of drug delivery following the administration of magnetic albumin microspheres containing adriamycin to therat, J. Pharm. Sci., 78:190 (1989). 114. P. K. Gupta and C. T. Hung, Albumin microspheres. V. Evaluation of parameters controlling the efficacy of magnetic microspheres in the targeted delivery of adriamycin in rats, Int. J. Pharm., 5957 (1990). 115. J. M. Gallo,C. T. Hung, P. K. GuptaandD. G. Perrier,Physiological pharmacokinetic model of adriamycin delivered via magnetic albumin microspheres in therat, J. Pharmacokinet. Biopharm., 17:305 (1989). 116. P. K. Gupta and C. T. Hung, Effect of carrier dose on the multiple tissue disposition of doxorubicin hydrochloride administered via magnetic albumin microspheres in rats,J. Pharm. Sci., 78:745 (1989). 117. P. K. Gupta and C. T. Hung, Targeted delivery of low dose doxorubicin hydrochloride administered via magnetic albumin microspheres in rats, J. Microencaps., f

574

Leroux et al.

118. A. Ibrahim, P. Couvreur, M. Roland, and P. Speiser, New magnetic drug camer, J. Pharm. Pharmacol., 3559 (1983). 119. F. Brasseur, P. Couvreur, B. Kante, et al.,ActinomycinDadsorbed on polymethylcyanoacrylate nanoparticles: Increased efficiency against an experimental tumor, Eur. J. Cancer, 16:1441 (1980). 120. K. Sugibayashi, M. Akimoto,Y. Morimoto, et al., Drug-carrier property of albumin microspheres in ckemotherapy. 111. Effect of microsphere-entrapped 5-fluorouracilon Ehrlich ascites carcinoma in mice, J. Pharm. Dyn., 2:350 (1979). 121. K.J. Widder, R. M. Moms, G. Poore, et al., 'Ihmor remission in Yoshida sarcoma-bearing rats by selective targeting of magnetic albumin microspheres containing doxorubicin, Proc. Natl. Acad. Sci. USA, 78579 (1981). 122. Y. Morimoto, K. Sugibayashi, and Y. Kato, Drug-camer property of albuV. Antitumoreffectofmicrosphereminmicrospheresinchemotherapy. entrapped adriamycin on liver metastasis of AH 7974 cells in rats, Chem. Pharm. Bull., 29:1433 (1981). 123. P. Couvreur, B. Kante,L. Grislain, et al., Toxicity of polyalkylcyanoacrylate nanoparticles 11: doxorubicin-loaded nanoparticles, J. Pharm.Sci.,71:790 (1982). 124. K. J. Widder, R. M. Moms,G.A.Poore, et al.,Selectivetargeting of magnetic albumin microspheres containing low-dose doxorubicin: Total remission in Yoshida sarcoma-bearingrats, Eur.J. Cancer Clin. Oncol., 19:135 (1983). 125. R. M. Morris, G. A. Poore, D. P. Howard, and J.A. Sefranka, Selective targeting of magnetic albumin microspheres containing vindesine sulfate: total remission in Yoshida sarcoma-bearing rats, in Microspheres and Drug Therapy (S.S. Davis, L. Illum, J. G . McVie,and E. Tomlinson,eds.), Elsevier, Amsterdam, 1984, p. 439. 126. J. Kreuter, Factors influencing the body distribution of polyacrylic nanoparticles, in Drug Targeting, (P. Buri and A. Gumma, eds.), Elsevier, Amsterdam, 1985, p. 51. 127. F: Brasseur, C. Verdun, P. Couvreur, et al., Evaluation experimentale de I'efficacitC thkrapeutique de la doxorubicine associee aux nanoparticules de polyalkylcyanoacrylate, Proc. 4th Int. Conf. Pharm. Technol., 5:177 (1986). 128. N. Chiannilkulchai, Z. Driouich, J. P. Benoit, et al.,Nanoparticules de doxorubicine: vecteurs colloidaux dans le traitement des mttastases hepatiques chez I'animal, Bull. Cancer, 76:845 (1989). 129. N. Chiannilkulchai, Z.Driouich, J. P. Benoit, et al.,Doxorubicin-loaded nanoparticles: increased efficiency in murine hepatic metastases, Sel. Cancer Ther., 5:l (1989). 130. P. K. Gupta, C.T. Hung, andF. C. Lam, Applicationof regression analysisin the evaluationof tumor response following the administration of adriamycin either as a solutionor via magnetic microspheres to the rat, J. Pharm. Sci., 79:634 (1990). 131. W. P. Yu, G . M. Barrat, J.Ph. Devissaguet, andF. Puisieux, Anti-metastatic


132. 133. 134. 135. 136. 137. 138.

139. 140. 141. 142.


575

activityinvivoof MDP-L-alanyl-cholesterol (MTP-CHOL)entrappedin nanocapsules, Int.J. Immunopharmacol., 13:167 (1991). W.P.Yu, C. Foucher, G. Barratt,et al., Antimetastatic activityof muramyl dipeptide-L-alanyl-cholesterol incorporated into various typesof nanocapsules, Proc. 6th Int. Conf. Pharm. Techno]., 3:83 (1992). P. Blagoeva, R. M. Balansky, T. J. Mircheva, and M. I. Simeonova, Diminished genotoxicity of mitomycin C and farmorubicin included in polybutylcyanoacrylate nanoparticles, Mutat. Res., 268:77 (1992). S. Gibaud, J. P. Andreux, C. Weingarten, et al., Increased bone marrow toxicityofdoxorubicinboundtonanoparticles,Eur. J. Cancer,30A:820 (1994). L. Manil, P. Couvreur, and P. Mahieu, Acute renal toxicity of doxorubicin (adriamycin)-loaded cyanoacrylate nanoparticles, Pharm. Res., 12:85 (1995). E. AIlCmann, S. V. Kudrevich, K. Lewis, et al., In vivo photodynamic activity of hexadecafluoro zinc phthalocyanin loaded in PEG-coated nanoparticles, Proc. 1st World Meeting APGIIAPV, 1:487 (1995). G . Barratt, F. Puisieux, W. P. Yu, et al., Antimetastatic activity of MDP-Lalanyl-cholesterol incorporated into various types of nanocapsules, Int. J. Immunopharmacol., 16:457 (1994). T. Blunk, D. F. Hochstrasser, J. C. Sanchez, et al., Colloidal camers for on surface intravenous drug targeting: plasma protein adsorption patterns modifiedlatexparticlesevaluatedbytwo-dimensionalpolyacrylamidegel electrophoresis, Electrophoresis, 14:1382 (1993). Y. Tabata andY. Ikada, Protein precoatingof polylactide microspheres containingalipophilicimmunopotentiatorforenhancementofmacrophage phagocytosis and activation, Pharm. Res., 6:296 (1989). G. Poste and R. Kirsh, Site-specific (targeted) drug delivery in cancer chemotherapy, Biotechnology, 1:869 (1983). R. Kirsh, P. J. Bugelski, and G. Poste, Drug delivery to macrophages forthe therapy of cancer and infectious diseases, Ann. N.Y. Acad. Sci., 507:141 (1987). D. F. Ranney and H. H. Huffaker, Magnetic microspheres for the targeted controlled release of drugsanddiagnosticagents,Ann.N.Y.Acad.Sci., 507:104 (1987).


17 Cosmetic Applications of Liposomes Gdrard Redziniak and Pierre Perrier Parfums Christian Dior, Saint Jean de Braye, France

577

I. Introduction 11. Definition: A Phospholipid Vesicle the Gram to the Ton

579 579 580

Liposomes and Cutaneous Cell Interactions

580

In Vivo Applications

582

111. Manufacturing-From A. Methods B. Composition IV. V.

1.

578

INTRODUCTION

Liposomes were first described by Alec Bangham [l] 30 years ago. At the end of the 1960s, liposomes were regarded as efficient and specific drug carriers capable of carrying the drugs they encapsulated directly to a targeted cellular site. However, in spite of a considerable amount of enthusiasm in the early literature, therewas still no product containing liposomes on the marketby the end of 1985. At that time, the majority of investigators were rather pessimistic: “the initialexcitement has given way to intense scientific activity in a num577

578 Perrier

and

Redziniak

ber of laboratories all over the world, but a marketable realistic product appears to beelusive” [2]. Bangham [3] was more optimistic when he wrote“gazing into a crista1 ball I think I can see a bright future [for liposomes].” In 1986, the first commercial productincorporating liposomes identical to those described by Bangham appeared in the market (Capture@).At the same time, a synthetic one made by nonionic surfactants [4] also was launched (Niosomes@),but these first products were not drugs; they were cosmetics. II. DEFINITION: A PHOSPHOLIPID VESICLE

Liposomes are artificial spherical microcapsules made up of phospholipids that are able to encapsulate active ingredients in their structure (Fig. 1). The same phospholipids are the main constituent of cell membrane(s) where they act as a selective barrier and as a holder of the membrane proteins [6]. Phospholipids are amphiphilic molecules of the glycerophospholipids class made of an hydrophobic “tail” (fatty acids) and a polar hydrophilic “head” (e.g., choline, serine, inositol). When phospholipids are dispersed in water, hydrophobic “tails” aggregate together as far as possible from water molecules, whereas the hydrophilic “heads” come in contact with

Hydrophilic molecule

holipids

Amphiphilic molecule Cholesterol

Fig. l Schematic representation of a multilamellar liposome.

Cosmetic Applications of Liposomes

579

water. Thus, a layer is created in which the fattyacids tails are directed to the inside of the membrane and the polar head is the outside. When phospholipids swell and hydrate in water, they form vesicles made of one orseveral concentric bilayers which are either surrounded or divided by one orseveral aqueous compartments. These vesicles were first used by biophysicians as biological membrane models in order to study their physiological properties, especially their permeability. In 1965, Bangham called these vesicles liposomes and demonstrated that they are closed system capable of developing almost spontaneously from natural or synthetic phospholipids when they are in presence of an aqueous medium. 111.

MANUFACTURING-FROM THE GRAM TO THE TON

A. Methods The first liposome preparation method was naturally proposed by Bangham using a rotary evaporator. This method is still today considered as the reference technique. Other techniques have been described: the socalled ethanol injection technique[7];the reverse phase evaporation technique [8]; the dialysis process [9]; the freezing-thawing method [lo]; the spray-drying technique [ll]. In this latter process, natural or synthetic phospholipids and other liposome membrane constituents (e.g., sterols, lipophilic ccactive7’ ingredient, antioxidants) are dissolvedin an organic solvent (chloroform, dichloromethane, methanol). This solution is then pulverized in a fluidized gas stream to yield a “preliposomes”fine powder (Fig. 2). Then this powder is rehydrated in an aqueous solution containing hydrophilic “active”ingredient(s)(e.g.,peptides, vitamins, coenzymes, enzyme activators or inhibitors; anti-free radicals) and a dispersion of large liposomes is obtained by simple stirring. This dispersion is then refined by extrusion under high pressure [12]. Before homogenization, liposomes are multilamellar, and their size depends on the phospholipidic mixture, but varies between 0.5 and a few micronmeters. After homogenization, the average size of a liposome is about 0.1 pm. These liposomes are mainly small unilamellar vesicles (SMV) small unilamellar vesicles (SUV) (Fig. 3). Encapsulation ratesvaries from 1 to 30% for hydrophilic molecules. It can reach 100% for hydrophobic molecules entrapped in the phospholipidic bilayer. Used at industrial scale, this method allows the productionof large quantities of liposomes ( > l ton by batch).

580 Perrier

and

Redziniak

Fig. 2 Preliposomes powder observed by scanning electronic microscopy (x5000). A mixture of hydrogenatedphosphatidylcholin(LucasMeyer,Germany), p. Sitosterol (Kaukas, Finland), Asiaticoside (Indena, Italy), antioxidants are dissolved in an proportion of 80/10/9/1 (w/w)insolvents (dichloromethan-methanol, Ul). Then this solution is spray-dried.

B. Composition The chemical composition, and moreparticularly the choice of the phospholipids used, is an essential step in the development of liposomes. Natural (yolk or soya) or synthetic and hemisynthetic phosphatidylcholins are mostly used. Apart from this phospholipid, it can be useful to add a sterol (cholesterol, beta-sitosterol) to modulate the“transition temperature” and the lipidic bilayer microviscosity. Electrocharged componentssuchas dicetyl-phosphate or stearylamine can also be added to provide the liposomes with a negative or positive charge. These charges generate repulsions among sheets and among vesicles and improveencapsulation capacity and stability. IV. LIPOSOMES AND CUTANEOUS CELL INTERACTIONS

Because of their biochemical nature andtheir structural identity with cellular membranes, liposomesare able to interact with cutaneous cells. Differ-


58 l

Fig. 3 Liposome suspension observed by transmission electronic microscopy after freeze-fracture (~45000).A “preliposomes” powderis dispersed in an aqueous solution of peptides and stirred during1 hour. Then the dispersion is homogenized in the high-pressure extruder.

ent interaction types can be foreseen depending on theliposome composition and the type of cells encountered [13]: endocytosis, transfer or exchange of phospholipids, fusion. Some publicationshave allowed us to support or to contradict these various possibilities; studies performed in our laboratories enable us to bring new information. It is well knownthat in various cellular types (e.g., neurons,lymphocytes), the membranes of the old cells are more “rigid” than those of the [14].Invesyoung cells, as they contain more cholesterol sphingomyelin and tigators such as Shinitzky [l51 state that the decrease, with age, of membrane fluidity isconsistent with cellular metabolism slowing down. In vitro techniqueshave shown the influence of different effectorson the fluidity of the membranes of keratinocytes [16].In these techniques, the membrane fluidity is monitored by fluorescence polarization analysis of (DPH). the hydrophobic probe l-6-diphenyl-l-3-5-hexatriene

Redziniak and Perrier

This test has demonstrated that the addition of cholesterol in the ester culture medium reduces the membranefluidity and blocks the endocytosis. To the contrary, a specific composition of phospholipids in the form of a calibrated liposomal dispersion (0.1 pm) reverses the effect of the cholesterol ester by increasing the membrane fluidity and reinstoring the endocytosis of the keratinocytes. As an essential element for the rigidity or fluidity of the membraneis the quantitative relationship between cholesterol and phospholipids, one can easily conceive that the unsaturated phospholipids of the liposomes, fluid as they are above their transition temperature,fuse or exchange with the cell membrane. Other in vitro investigations on cell culture have shown an increase of the efficacy of vitamin A esters and particularly the vitamin A propionate when it is encapsulated in liposomes compared with the same substance,atthesameconcentration, and in afree form [17]. These results allow us to assume that, at least in vitro, liposomes interact with the cell membrane of skin cells and transfer their contents into the cell. This interaction can be increased by a specific targeting using a “lectinsugar binding strategy’’ [18-191. of the By the discovery of particular endogenous lectins at the surface basal layer keratinocytes and human melanocytes [20], we have built new targeted liposomes for cosmetics application that also can be useful in therapeutical skin treatment.

V. IN VIVO APPLICATIONS Even if about 40 pharmaceutical products containing liposomes are at different development levels, one can wonder why the cosmetic industry first marketed products containing liposomes. The final aimof the liposomes isthe same: they are carriershaving to convey an active principle to the target cells. For this purpose, the liposomes have to go through different obstacles in maintaining their structure, without releasing their encapsulated molecule(s). These liposomes should be able to recognize the target cells and to adhere to them. The main difficulty encountered in the preparation of liposomes for therapeutical application is their purification; that is, the completeelimination of nonencapsulated active principles, especially if they are toxic. This purification requires sophisticated processes such as gel filtration chromatography, ultracentrifugation, dialysis, and ultrafiltration [21]. In the case of cosmetic products and considering the lack of toxicity of the active ingredients, it is possible to avoid this process by over-

of Applications Cosmetic

583

concentrating the hydrophilic active principle, allowing its presence both inside and outside of the liposome. Another problem can be faced in therapeutics: stability and targeting or specificity after oral or parenteraladministration. It seems that we do not have to face such problems with cutaneous application: There is no need toavoid bile salts or lytic enzymes or to evade the reticuloendothelial system. The target tissue, the skin, is very visible. In cosmetics, at least for thefirst generation of active principles used, the main concern is to reach cutaneous cells while limiting the passage in blood circulation as much aspossible. In this field, the reference works are those of Mezei [22], who has studied the cutaneous penetration of labeled triamcinolone incorporated into various vehicles (ointment, lotion,and gel withor without liposomes). Compared with other vehicles, utilization of the liposomal form allows an increase five times the molecular concentration in the epidermis and three times in the dermis. On the other hand, there is an extremely low quantity of the molecule in the blood circulation. This first study on the liposome penetration in the skin was followed by numerous works confirmingthe Mezei data [23-281. We can note that the liposomes increase the active ingredient’s concentration around the targeted cells and then, at the same time, they increase the actionof these active ingredients on thecells. It seems logical to expect an increaseof the activity of products containing liposomes. Incorporated in the lipidicphase of the vesicle, vitamin A acid has been shown to be 10 times more efficient than when presented in free form[29]. Besides, we have demonstrated that marketed products containing liposomes and tested by dermatologists revealed a whole series of positive effects;forexample, improvement of cutaneoushydration, visibleimprovement of skin texture, increase in skin glow, decrease in the depth of wrinkles, decrease in eye puffiness, and decrease in the number aging spots. When looking at the results, we understand the extreme interest of this tool in cosmetics in different claims [30]; however, the difficulties and pitfalls must not be neglected, which makes theutilization of the liposome very ticklish. In every new formula containing liposomes, it is essential to check the following: Safety of the preparation,as it can potentiate theeffect of ingredients applied on the skin Physical and chemical stability, since a marketed product must be stored for along time without modification of its properties 4

584 Perrier

and

Redziniak

To be efficient, such controls need sophisticated equipment, for instance electronic microscopy, and checking of size of the liposomes by laser light scattering. It is now taken for granted that liposomes representa major step in cosmetics formulation. However, this requires research studies and severe controls. The choice of the type of liposome andof its formulation must be according to theactive ingredient to be encapsulated and also according to the required target. The final product must be formulated and tested regarding its stability as well as its safety and efficacy. REFERENCES 1. A. D. Bangham, M. M. Standish, and J. C. Watkins, J. Mol. Biol., 13:238

(1965). 2. M. J. Groves, S.T.P. Pharma Sci., 3:664 (1987). 3. A. D. Bangham, in Lipids and Membranes: Past, Present and Future (J.A.F. Op den Kamp, B. Roelofen, andK. Witz, eds.), Elsevier, Amsterdam, 1986, p. 106. 4. R. M. Handjani-Vila, A. Ribier, B. Rondot, and G. Vanlerberghe, Int. J. Cosmet. Sci., 1:303 (1979). 5. J. F. Danielli andH. Dawson, J. Cell Comp. Physiol., 5:495 (1935). 6. S. J. Singer and G. L. Nicolson, Sciences, 175:720 (1972). 7. J. Batzri and E. D. Korn, J. Cell. Biol., 66:621 (1975). 8. F. Szoka and D. Papahadjopoulos, P.N.A.S., 75:4194 (1978). 9. 0.Zumbuehl and H. G. Weder, Biochem. Biophys. Acta, 640:252 (1981). 10. G. Strauss in liposome Technology, Vol. 1 (G. Gregoriadis, ed.), CRC Press, Boca Raton, FL, 1984, p. 197. 11. G.RedziniakandA.Meybeck, U.S.Patent 4,508,703,ParfumsChristian Dior Assignee, 1985. 12. G. RedziniakandA.Meybeck, U.S. Patent 4,621,023,ParfumsChristian Dior Assignee, 1986. 13. G. Redziniak, Seminaire INSERM, 214:129 (1991). 14. G. Rouser and A. Yamamoto, Lipids, 3:284 (1968). 15. M. Shinitzky, in Physiologyof Membrane Fluidity, Vol.1(M. Shinitzky, ed.), CRC Press, Boca Raton,FL, 1984, p. 1. 16. T. M.Callaghan, P. Metezeau,H.Gachelin, et al., J. Invest.Dermatol., 92:410 (1989). 17. J. Franchi, M.C. Coutadeur, J. C. Archambault, et al., Nouv. Dermatol., 12:443 (1993). 18. D.Cerdan, C. Grillon, M. Monsigny, et al., Biol. Cell., 73:35 (1991). 19. D. Cerdan, G. Redziniak, C. A. Bourgeois, et al., Exp. Cell Res., 203:164 (1992). 20. A. Denis, C. Kieda, P.Monsigny,andG.Redziniak, Eur. Patent 464077, Parfums Christian Dior Assignee, 1989.


21. 22. 23. 24. 25. 26.

585

F. J. Martin, Drug Pharm. Sci., 41:267 (1990).

M. Mezei andV. Gulasekharam, LifeSci., 26:1473 (1980). J. Lash and W. Wohlrab, Biomed. Biochem. Acta, 45:1295 (1986). W. Wohlrab and J. Lash, Dermatologica, 174:18 (1987). A. Kato, I. Yasuo, and M. J. Yasuo, Pharm. Pharmacol., 39:399 (1987). R. Natmki, S. Tomomichi, R. Matsuo, et al., Pharmacobiol. Dyn., 9:S-12

(1986). 27. N. Weiner, N. Williams, G. Birch, et al., Microb. Agents Chemother., 34:107 (1990). 28. K.Egbaria, C. Ramachandran,D. Kittayanond, and N. Weiner, Antimicrob. Agents Chemother., 34:107 (1990). 29. A. Meybeck, R. Michelon, C. Montastier, and G. Redziniak, French Patent 2591105, Parfums Christian Dior Assignee, 1985. Dermatics,GriesbachConference (0. Braun30. A. MeybeckinLiposome H. I. Maibach,eds.),Springer-Verlag,Berlin, Falco,M.C.Korting,and 1992, p. 341.


18 Cosmetic Applicationsof Vesicular Delivery Systems Simon Benita The Hebrew University ofJerusalem, Jerusalem, Israel

Marie-Claude Martini Institut des Sciences Phunnaceutiques etBwlogiques, Lyon, France

Monique Seiller Universitk de Caen, Caen, France

I. Introduction

11. %es of Vesicular Delivery Systems A. Liposomes B. Nanoparticulate Systems C. Microemulsions D. MultipleEmulsions E. LiquidCrystals

588 589 589 592 593 594 595

111. Composition A. Liposomes B. NanoparticulateSystems C. Microemulsions D. MultipleEmulsions E. LiquidCrystals

597 597 598 599

W. Production

602 602

A. B. C. D.

Liposomes NanoparticulateSystems Microemulsions MultipleEmulsions E. LiquidCrystals

600 601

605 608 608

611 587

588

Benita et al.

V. Characterization A. Microscopic Analyses B. Rheological Analysis C. Encapsulation or Incorporation Efficiency of the Active Substance

VI. Stability A. B. C. D.

Liposomes Nanoparticles Microemulsions Multiple Emulsions E. Liquid Crystals

VII. Cosmetic Uses A. B. C. D.

Liposomes Nanoparticles Microemulsions Multiple Emulsions E. Liquid Crystals

1.

611 611 614 615 615 615 616 616 617 617 617 617 621 621 622 624

INTRODUCTION

Cosmetic technology is constantly evolving in term of raw materials, excipients, and formulations of active ingredients. The new surfactant molecules, the search for original active substances and efficient combinations, and the design of novel vehicles or carriers led to the implementation of new cosmetic systems incontrast to theclassic forms such ascreams or gels. The achievements of the extensive research conducted over the last 15 years have resulted in the development of well-controlled innovative delivery systems. Some of these systems have been extensively investigated for their therapeutic potential at the same time that they were being examined quitesuccessfully for theirpossible cosmetic uses. The main objective of this chapter is to concentrate on and fully describe the preparation, characterization,and fate of the various delivery systems following topical application. The recent sophisticated cosmetic preparations based on theinnovative carriers are more attractive than the regular cosmetic preparations, because the sensitive and active cosmetic substances are more fully protected when incorporated in the innovative carrier systems. Furthermore, the carrierscan promote and enhance the cutaneous permeation of specific substances that normally exhibit low skin permeability. This can occur because the stratum corneum, the outermost skin layer, consists of dense,

Cosmetic Applications

of Vesicular Delivery Systems

589

overlapping laminates of dead cells with each cell packed with keratin filaments in an amorphousmatrix of proteins with lipids and water-soluble substances, The novel carrier thereforefavors the penetrationof the active substance through the stratum corneum, which is recognized as the ratelimiting barrier to the ingress of materials. This is an important feature, since the novel camer can retain oreven sustain the releaseof these active substances in the epidermis leading to skin targeting. Among the various delivery systems or camers, the liposomes have been the most widely studied and successfully marketed by the cosmetic industry. Furthermore,other vesicularsystems(such as nanocapsules, nanospheres, and multiple emulsions) also have been developed and are currently under investigation. If their claimed efficiency and potential is proven, there is no doubt that these vesicular systemswill be rapidly developed and will be present in the cosmetic market in the next decade. It is believed that effectiveand pioneering cosmetic applications of these vesicular systems stillremain to be discovered. When discovered and developed, such systemswill probably open up a new era of effective and sophisticated cosmetic preparations. II. TYPES OF VESICULAR DELIVERY SYSTEMS

A. Liposomes

Liposomes were first studied around 1965 as models of biological membranes [l-61. By 1970, their structureand physical-chemicalcharacteristics had led researchers in a number of fields to investigate the potential of liposomes as camers of therapeutical active ingredients. Research on the use of liposomes in the cosmetics sector started more recently. This research has included attempts to achieve an optimal modulation in the release of active substances introduced within this type of vesicle by means of phospholipids. Although research on liposomesin the cosmetics field started fairly late, thewell-established ability of liposomes to protectencapsulated active substances and above all the similarity between most of their components and cutaneouslipids turned liposomes quite quickly into the prime camers of dermatological ingredients. Furthermore, they are compatible withmany active, biodegradable substances of limited toxicity. Thus, for aperiod of someyears now, the use of liposomes in cosmetology has continued togrow. Liposomes are spherical vesicles with anaqueous cavity at their center (Fig. 1). The encasing envelope is made up of a varying number of bimolecular sheets (lamellae)composed of phospholipids. They are either unilamellar, and thus have a single bilayer, or oligolamellar, with multiple

Benita et a/.

590

Fig. 1 Schematic descriptions of unilamellar liposomes. bilayers, or multilamellar, with a large number of bilayers. Liposomes are either homogeneous with a narrow distribution, or heterogeneous with a broad distribution. Their size varies from approximately 15 to 3500 nm. The type of liposomes-small unilamellar vesicles (SUV),large multilamellar vesicles (MLV), and so forth-depends mainly on: Nature of the amphiphilic lipids selected Composition of the iso-osmotic dispersing solution Method of preparation (ultradispersion) Since their appearance in 1971 as carriers of active substances, a whole series of names other than liposomes can be found in the literature, and have been given to commercial products depending on the nature of the components which form their envelope. Thus, if the envelope is formed of sphingolipids, the perfectly correct nameof sphingosome is given; if some stabilizing agent or active substance is introduced into thevesicles, a more or less imaginative trade name is given; for example, Dermosome, Glycosome, and Brookosome. Thetrade name Ufasome, forexample, was coined to describe vesicles made of a long chain of unsaturated fattyacids. Vesicles whose envelopes are made up of nonionic surfactants are called Niosomes. Their discovery resulted from the physicochemical problems exhibited by liposomes. In fact, the large extent of leakage andthe mediocre stability of liposomes led a number of researchers to design new systems containing amphiphilic lipids different from those used for liposomes; thus, synthetic nonionic lipids were created, allowing the formation of

Cosmetic Applications of Vesicular Delivery Systems

59 1

vesicles whichpresent the same propertiesexhibited by liposomes without the disadvantages previously mentioned. These systems have also found their place in all sorts of cosmetic preparations. These syntheticnonionic lipid systems varyin size, shape, and structure butremain in the sameranges as liposomes. They appeared as commercial products in the market shortly after liposomes. The first publication concerning original amphiphilic lipids capable of forming such vesicles, and the first patents dealing with these lipids, came out in1972 and 1975, respectively. llvo other systems may be likened to liposomes, particularly as they are composed primarily of phospholipids. 1. Supramolecular biocarriers (Fig. 2) are very small vesicles (20nm) formed by a gelified polysaccharide hydrophilic core capable of capturing the active substances in the links of a network. This central core is surrounded by a crown of fatty acids attached to the coreby covalent bonds. The whole iscovered by an external sheetof phospholipids attached to the lipid crown by hydrophobic interactions, with their polar heads facing the periphery. The hydrophilic active substances are attached with more orless stability to theheart of the core, and the active lipophilic substances penetrate through the double-lipid membrane. These supramolecular carriers can be either of nautral or syntheticorigin. (Fig. 3). 2. A second liposome-like system is totally lipid in nature and actsas a carrier forlipophilic substances (Lipomicrons). The phospholipid molecules, fatty acids, and cholesterol are arranged so as to form spherical

I

I I

Fig. 2 Schematic description of supramolecularbiocamers (biovectors).

Benita et al.

592

Fig. 3 Comparativedescriptionbetween pramolecular biocarriers.

(A) naturaland

(B) syntheticsu-

globules of 400-500 nm. Their periphery is hydrophilic, but their interior, completely lipophilic, may be loaded with liposoluble vitamins or sunscreen agents.

B. NanoparticulateSystems The entities known as particulate systems are becoming increasingly popular, just as the delivery systems discussed above, both in the fields of cosmetology and pharmacy. Although in principle particulate systems are generally well known, their proliferation and sometimes their definitions cause some confusion as to their capacity as carriers, as well as their biofate in the stratum corneum. The size of particulate systems, which ranges from a few nanometers to a millimeter, allows the formation of two groups which offer completely different carrier possibilities, as theboundary is around 1pm. For particulate sizes smaller than 1pm (nanoparticles), the insertion of vesicles between corneal cells is possible. Thus, theseare true carriers of active agents, the release kinetics of which are a function of the stability and structureof the system. For sizes larger than 1 pm, the fate of the particulate forms remains superficial, with the possible release of agents or surface activity likely due to the components of the matrix itself. These latter systems are not addressed in this chapter because of their lack of uniqueness as compared with that of nanoparticles, and also because of the tremendous lack of

Cosmetic Application

Vesicular ofDelivery Systems

A

593

B

Fig. 4 Schematic description of a nanosphere (A) and a nanocapsule (B).

precision in their names anddefinitions. Microspheres, microcapsules, millispheres, thallaspheres, microbeads, or pearls are included in this category. All these different systems share a common spherical form, which is either a matrix or reservoir type. They vary in size ranging from 5 pm (microsphere) to some millimeters (microbeads), and in the materials from which they are composed (e.g. , polyamides, collagens, polysaccharides). Among nanoparticulate systems [7], nanospheres are generally distinguished from nanocapsules. Nanospheres are matrix systems of polymers composed of a solid core with a porous structure and a discontinuous envelope (Fig. 4A). Substances are adsorbed uponthe polymeric materials and generally dissolve inthe polymerization environmental medium,which is most often oriented toward hydrophilic ingredients. Nanocapsulesare reservoir systems made upof a continuous polymerized envelope surrounding a liquid or gelified core (Fig. 4B). The substances are most often of a lipophilic nature, andthey may be composedof a dispersion or oily mixture. Certain nanoparticles are hybrid nanocapsules-nanospheres insofar as theactive substances are intimately integrated into thegelified core of a nanocapsule.

C. Microemulsions The study of microemulsions [8] is often coupled with the study of micellar solutions because of their structural similarity. The distinctions between the two systems remainto be defined (Fig. 5 ) . Microemulsionsare stable dispersions of a liquid inthe form of spherical droplets whose diameter is lessthan 100 nm. Theyare composed of oil, water, and one ormore surface-active agents. The dispersion is micellar in nature, formed by the aggregation of amphiphilic molecules around either an aqueous core (normal micelles) or around a lipid core (inverse micelles),

594

Benita et al.

A$ 7nm

5nm

Solution

Micelle

20nm

Solubilization Microemulsion

Fig. 5 Formation and schematic description of a microemulsion. resulting in oil-in-water ( O N ) or water-in-oil (W/O) microemulsions, respectively. They form spontaneously without the use of energy subject to the appropriate choice of surface-active agents that must sufficiently lower the interfacial tension to make it either negligible or a negative value. The main characteristics of microemulsions are low viscosity associated with a Newtonian-type flow, a transparent or translucid appearance associated with the isotropic character of the system, and thermodynamic stability within a specific temperature setting. Certain microemulsions may thus be obtainedwithout heating simply by mixingthe components as long as they are in a liquid state. D. MultipleEmulsions

Multiple emulsions are emulsions of emulsions [9,10]. They can comprise, in each of theirthree constituent phases active hydrosoluble and/or liposoluble substances. Although interesting as cosmetological forms, multiple emulsions are not widely usedin cosmetic preparations. Nevertheless, it is anticipated that these limitations will be removed in the very near future, and that multiple emulsions will be very much in demand in the coming years, particularly if their extended-release abilities are confirmed. The use of multiple emulsions could be almost as broad as thatof the simple emulsions from which they are derived and at least equal to other vesicular systems to which they are similar. One of the considerations in their favor is that they are capable of exhibiting the same properties as those of simple emulsions, and, like these, are applicable directly to the skin, since they can be constituted in the form of white, oily creams with a consistency well suited to proper spreading. Multiple emulsions are emulsions in which the dispersion phase contains another dispersion phase. Thus, a water-in-oil-in water (WlOnV) emulsion is a system in which the globules of water are dispersed in globules of oil, and the oil globules are



595

themselves dispersed in an aqueous environment. A parallel arrangement exists in oil-in-water-in oil (O/W/O) type of multiple emulsions in which an internal oily phase is dispersed in aqueous globules, which are themselves dispersed within an external oily phase (Fig. 6). These emulsions are composed of at least two nonmiscible liquids. Emulsifiers, most often selected from amongsyntheticsurface-active agents, are thus required in order to formulate them. In orienting themselves along the two interfaces, these synthetic surface-active agentscalled, respectively, primary surface-active agent (SAA 1) and secondary surface-active agent ( S A A 2)-each forms a film which, during a longeror shorter period of time, gives the system a certain degreeof stability. In the case of W/O/W-type emulsions, the only ones which will be described in this chapter, the S A A 1 molecules, which tend to be lipophilic, are arranged on the internal W/O interface, and the S A A 2 molecules, which have a hydrophilic tendency, are arrangedalong the externalO/W surface. Thus, if the natureof the surface-active agents is welladapted to that of the oily phase, two monomolecular films will be formed: the apolar part of each emulsifier is localized in the oil, whereas the polar part is situated in the internal orexternal aqueous phase. It follows that the emulsifiers are organized into abimolecular layer along the interfaces,forming with the oil the envelopeof the vesicle itself.The idea that thesesystems could serve as useful vehicles for the transport of active ingredients has crystallized only during the 1990s despite thefact that multiple emulsions have been known for a numberof decades. Certain productsnow on the market arecalled triphase emulsions or triple-phase emulsions. Some of these are not truemultiple emulsions but rather complex preparations composed, for example,of three main components, a simple gelified emulsion, a simple emulsion containing microparticles in suspension, or a simple emulsion in which three active substances are incorporated and present three different activities. One must remember that true multiple emulsions are also sometimes called triple emulsions. Although in theory quadruple and even quintuple emulsions exist, a more precise name would be double emulsion, since in these systems two emulsions co-exist: one emulsion with a continuous oily phase and one emulsion with a continuous aqueous phase.

E. Liquid Crystals Discovered a centuryago, liquid crystals have been used in many different fields for a number of decades, but their value in cosmetology has been confirmed only in the last few years [11,12]. Liquid crystals are intermediate anisotropic fluids which are between the conventional solid and liquid

596

et

Benita Water

w/o/w

Oil

emulsion

Water

@/W/O emulsion

Oil

0-

hydrophobic ADS

+. hydrophilic ADS

Fig. 6 Schematic representation of O/W/O and WIONV multiple emulsions.

al.


597

5 nm

Fig. 7 Schematic representationof lamellar phases.

isotropic phases. This structure is attributable to a very specific morphology of molecules which must be elongated and present an irregular distribution of electrical charges. These molecules must be arranged in a specific order (Fig. 7). 111.

COMPOSITION

A. Liposomes

1. Phospholipids Phospholipids and specific additives are considered the most important primarymaterials for vesicular delivery systems[13-181. Some phospholipids, like phosphatidylcholine, are of natural origin, and some, like dimyristoyl or phosphatidylcholine dipalmitoyle, are of semisynthetic or synthetic origin. The phospholipids that are most commonly used are the glycerophospholipids. In these substances, glycerol, whichmay be considered the skeleton of the molecule, is esterified in positions 1and 2 by long-chain fatty acids and in position 3 by phosphoric acid. A number of different hydrophilic molecules can be attachedto this acid, for example, choline and ethanolamine. Although the length of the chain of fatty acids and/or the degree of saturation may vary from C14 to C18 and from1to 2 double bonds,the lipophilic nature of the alkyl chain is relatively constant compared with the hydrophilic part of the chain. In fact, this hydrophilic part of the chain presents different properties and has considerable influence on thecharacteristics of liposomes. A few other phospholipids shouldbementioned in addition to glycerophospholipids. These are less commonly used, but because of their

598

Benita et a/.

great similarity to the lipidsin the skin, they are good candidates for forming liposomes for dermatological use. Sphingolipids are a good example of these: They have a sphingosine skeleton upon which a number of derivatives are grafted. All these phospholipids have the fundamental property of forming flat lamellar sheetsin the presence of water within whichaqueous compartments alternate with lipid sheets. Under certain temperature and agitation conditions, these flat bilamellar structures may fragment, fold in on themselves, and ultimately become fastened at their ends. They thus become the external wall of water-trapping vesicles, which are themselves in suspension in the water. By choosing certain specific phospholipid fractions, even dry mixtures can be achieved. In thepresence of water, these allow the immediate formation of unilamellar liposomes of the LUV type (proliposomes). 2. Additives I In addition to thephospholipids which constitute the main envelope, two types of additives may be used (see Table 1).The first isa sterol (including phytosterol, dihydrocholesterol, and cholesterol). By localizing themselves in the phospholipid bilayer, these sterols permit phospholipids to modulate the physical and chemical characteristicsof the envelope,which becomes morerigid. As a result of the modified compactness, the permeability will change according to the proportion and the location of these substances. The second type of additive, in addition to, for example, buffers, electrolytes, pH modifiers, and preservatives, comprises ionic substances. Theseareanionic derivatives (phosphatidic acid, dicetylphosphate) or cationic derivatives (stearylamine). The function of these additives is to confer a negative or positive charge on vesicles, thus giving them a greater stability with regard to their aggregation and fusion. They may also cause an increase in the interlamellar space resulting in a greater capacity for the encapsulation of certain active substances. Thus, it is said that, except for liposomes, the essential components of the nonionic surfactant agent vesicle envelopes are, in this case, not phospholipids but rather nonionic synthetic surfactants. They areessentially surface-active agents containing ester or etherbonds. Their hydrophilic part consists of condensation products of polyoxyethylene, polyoses, and most of all polyglycerols (the most effective). The lipophilic part is usually composed of one ortwo hydrocarbon chains, between C12 and C18, either saturated or unsaturated. The additives used are the sameas for liposomes and have the same function. B. NanoparticulateSystems

The primary constituents of nanospheres and nanocapsules are identical for both of these types of systems [19]. In essence, they areacrylic deriva-



599

tives, most frequently of the polyalkylcyanoacrylatetype (PACA) or derivatives of copolymers of styrene, lactic and glycolic acids, cross-linked polysiloxames, or biological macromolecules such as albumin, gelatine, or dextran. In order to ensureand to control theparticle sue of nanoparticles of the SAA type, dodecylsulfate or sodium oleate and polysorbates are added. Sometimes, cross-linking agents (glutaraldehyde) orstabilizing substances (polyvinyl alcohol, cellulose derivatives), salt buffers, and protective colloids are incorporated in order to facilitate the formation of the individual nanoparticles in either an aqueous oroily environment. The active substances thus incorporated may be hydrophilic in the case of nanospheres or lipophilic in the case of nanocapsules (Table 2). It should be added that both types of active ingredients may be incorporated invariably into these two types of nanoparticles but that depends on the methods of fixation or incorporationused as depicted in Fig. 8. C. Microemulsions

Microemulsions are generally systems with four components: water, oil, surfactant, and co-surfactant [20-261. The aqueous phase may contain hydrophilic active ingredients and preservatives, and the fattyphase may be composed of mineral oil,silicone oil, vegetable oil, or esters of fatty acids, all of which are classic ingredients

Table 1. TypicalFormula of a Liposomal Suspension 200 mg Soya lecithin 25 mg Cholesterol Phosphatidic acid 30 mg Hyaluronicic acid 10 mg 5-10 mg Preservative Water to l o g Table 2. Typical Formula of a Nanocapsule Suspension acid Polylactic Mink oil Poloxamer 188 (Puronic F 68) Phosphatidylcholine Purified water to

125 mg 0.5 m1

75 mg 75 mg 20 m1

Benita et al.

600

A

C

0 D

Fig. 8 Description of the different incorporation patternsof active ingredients in nanospheres and nanocapsules. Active substance association: (A) dissolved in the nanosphere matrix, (B) dissolved in the liquid phase nanocapsule, (C) adsorbed at the nanosphere surface, (D) adsorbed at the nanocapsule surface.

for cosmetic products. The lipid phase may also contain lipophilic active ingredients. The surfactants chosen are generally among the nonionic group because of their good cntaneous tolerance. Certain derivatives of sugar are currently being studied (Table 3). To a lesser degree, and only for specific cases, amphoters are being investigated. The cosurfactants were originally short-chain fatty alcohols (pentanol,hexanol, benzyl alcohol).Theseare most often polyols, esters of polyol, derivatives of glycerol, and organic acids. Theirpurpose is to make the interfacial filmfluidbywedging themselves between the surfactant molecules. Thus, they create a bicontinuous structure allowing a continuous phase inversion; the undetectable transition from a W/O microemulsion into a O W microemulsion when the water to oil proportions are altered.

D. MultipleEmulsions The nature of the componentshas no significant effect either on themultiple character or on the stability of multiple emulsions [27-301 provided that a specific ratio is maintained between the SAA 1 and SAA 2 rates, the HLB of their mixture, and an optimal concentration of additives. The primary materials capable of producing this sort of system are quite numerous and practically identical to those generally used for simple



60 7

emulsions. The oily phases most frequently utilized in order to formmultiple W/O/W emulsions are hydrocarbons, esters, and triglycerides. The emulsifiers are usually nonionic surfactant agents. Someexamples are: for SAA 1 emulsions, the long hydrocarbon chain ester sorbitan, perfluorocarbons derivatives,and most importantly polymeric surfactants; for SAA 2 emulsions, esters of polyoxyethylene sorbitan, copolymers of ethylene propylene oxide, strongly ethoxylated fatty alcohols, and condensation products of polyglycerol (Table4). As in vesicular systems, additives are often introduced in order to control and increase the stability of the system. Along with electrolytes, sugars or glycols are used, most frequently hydrophilic polymers (xantham gum, cellulose derivatives, and carboxyvinylic compounds) introduced in one of the aqueousphases, most commonly in the external phase. Lipophilic substances (waxes, acids or fatty alcohols, silicone derivatives) are introduced into theoily phase. D. Liquid Crystals

There are two types of liquid crystals: thermotropics and lyotropics [31]. Thermotropic liquid crystals are made up of identical molecules or of a Table 3. Example of a Formulation of O N Microemulsion Sucroester Glycerol Jojoba oil Elastin Preservative Purified water q.s.p.

30 5 2 2 0.1

1 0 0

Table 4. Example of a Formulation of OIWIO Multiple Emulsion

Almond oil Urea Sodium Sorbitan oleate (Span 80) sulfate Magnesium Polyoxyethylated sorbitan stearate (Tween 60) Preservative water Distilled q.s.p.

25 2 3 8 0.7 1 0.1

1 0 0

Benita etal.

602

Table 5. Example of a Gel Formulation Containing Lyotropic Liquid Crystals

Oxyethylaled (2 OE) oleyl alcohol Oxyethylated (10 OE) oleyl alcohol cellulose Hydroxyethyl Preservative s.p. water Distilled

2 0.1 100

combination of molecules of the same geometrical form. They are “cholesteric” liquid crystals-so named because of their helical structure even though they are not always derived from cholesterol. In fact, liquid crystals dispersed in cosmetic preparations are generally basedon derivatives of methyl butyl Rhenol (Licritherm) or equivalent molecules. There are also combinations of cholesteric and synthetic chiral nematic esterson themarket. Liquid crystals constitute more or less thick mesomorphic structures in which the molecules are arranged in a certain order, leading to characteristics somewhere between those of a liquid state and a crystalline state. The kind of liquid crystals known as lyotropics are systems with two or three constituents: water, oil, and surfactant (Table 5). The soluble molecules are amphiphilic. One example is compounds of glycerol stearate and polyethylene glycol, which provide a lamellar structure beyonda given concentration, generally higher than 40%. They may be formed from concentrated micellar solutions or during the production of microemulsions wherein theproportions of the various components lead to a gelified phase. These lamellar structures may also form at the interface of an emulsion. IV. PRODUCTION A. Liposomes

There area number of processes available for the productionof liposomes [32-401, and appropriate selection would depend on the type of liposome desired: multilamellar or unilamellar and large or small. Only twoof these are described briefly below. The reference method (schematically presented in Fig. 9) was initially developed by Bangham, allowing the obtention of multilamellar vesicles. The process is carried out in four main stages:

Cosmetic Application of Vesicular Delivery Systems

603

Dissolution of phospholipids and other constituents of the wall coating in a volatile solvent

H/

Solution of wall constituents

U

Evaporation in a rotorevaporator

pJ

Addition of an aqueous solution

.... .*.

Suspension of liposomes

Fig. 9 Description of a typicalprocess for thepreparation of multilamellar liposomes.

b

1. Dissolution of the components of the envelope in a volatile solvent 2. Evaporation of the solvent under reduced pressure in a rotating evaporator in order to form a lamellar layer of phospholipids along the wall of the evaporator 3. Addition of a generally buffered aqueous solution at a temperature higher than the phase-transition temperature of the phospholipid 4. Stirring the suspension until it has cooled completely

604

Benita e?al.

The active substance tobe encapsulated is added to the organic solvent,if it is lipophilic, or in the aqueous solution,if it is hydrophilic. This method generally produces multilamellar liposomes and is carried out under nitrogen atmosphere. This is one of the oldest methods in use; for purposes of comparison, one of the most recent methods is mentioned below. The stages areas follows: Preparation of a first liquid phase made up primarily of a solution of phospholipids in a single solvent or acombination of solvents Preparation of a second liquid phase, miscible in the solvent or solvents described above, and insoluble for phospholipids, and made up primarily of water Addition, with moderate stirring,of the first phase to the second followed by partial or total elimination of the pure solvent in a way which willpermit the production a colloidal suspension of liposomes adequately loaded Just as with the preceding procedure, the active ingredients are added to the solvent most suited to themaccording to their solubility properties. This procedure yields multilamellar and especially oligolamellar liposomes of a very uniform size. Small unilamellar liposomes form when multilamellar liposomes obtained through some of the methods described above, for example, are subjected to high-frequency ultrasound waves during a fairly long period of time or to a high-pressure homogenization process with a two-stage homogenizing valve assembly. It is also possible to obtain small unilamellar liposomes by means of other methodswhich circumvent the need for ultrasound; like the method which involves injection of an ethanolic phospholipid solution into the aqueous phase or the method known as “detergent removal,” which consists of preparing mixed micellesof phospholipid detergent and theneliminating the latter. Just as unilamellar liposomes may be obtained from multilamellar liposomes, it is also possible to obtain large unilamellar liposomes from small unilamellar liposomes. Small vesiclesfuse and provoke the formation of large lamellar structures by the addition of calcium ions which function as complexing agents promoting the formationof large vesicles. The above method, which requires thepresence of phosphatidic acid, is less popular than two methods which make use of the injection of ether and evaporation in the inverse phase. The first of these two methods consists of dissolving the phospholipids in ether and then injecting this solution slowly into the aqueoussolution of the active hydrophilic substance to be encapsulated at a specified temperature. The evaporation of the ether causes the spontaneous formationof vesicles inthe aqueous phase.


605

The second method makes use of the production of an emulsion with a continuous oily phase after dissolving phospholipids in a volatile lipophilic solvent and the active hydrophilic ingredient to be encapsulated in the aqueous phase. Large liposomes form spontaneously after evaporation of the externalphase of the emulsion. All of the methods cited involve both advantages and disadvantages which will not be detailed in this chapter, since they have been described thoroughly in the literature. Each process is characterized by its possible induced degradation, the extent of its encapsulation efficiency whichcan be attained, its simplicity, its reproducibility, and its control of the particle size distribution of the liposomes obtained. As for this last point, when the liposome populations are tooheterogeneous, but include the type and size of liposomes desired, common separation methods such as gel filtration, ultrafiltration, ultracentrifugation, dialysis, or membrane diffusion under pressure are employed. None of the methodsused for liposomes is satisfactory for theproduction of large quantities of monionic surfactant vesicles. Indeed, large quantities of solvents are necessary, and.moreover, ultimately the encapsulation rate is low, since the dispersions are not rich enough in vesicles. A number of patents have been issued on specific methods that do not present any of the disadvantages mentioned above. The most common (and simplest) of these makes use of the formation of a lamellar phase. In summary, it involves the following stages: fusion of the amphiphilic lipid phase at a very high temperature; formation of a lamellar phase by mixing of the amphiphilic lipid phase with the aqueous phase containing the hydrophilic substances; gradual introduction of an iso-osmotic aqueous solution to the aqueous phase previously described; and homogenization, with the help of an efficient homogenizer, during a predetermined period of time, during which the preparation is allowed to reach the ambient temperature and the vesicles are formed. B. NanoparticulateSystems

As for liposomes, a large number of processes exist for obtaining nanoparticulate systems [41,42], and theprocess selected depends on thetype of nanoparticles desired (either nanospheres or nanocapsules), their particle size, and the materials used to form the nanoparticles. Nanospheres are most commonly obtained using polymerization reactions, purified natural macromolecules, or preformed polymers. Nanocapsules are most commonlyobtained through the use of interfacial polymerization reactions preformed polymers. A typical method for the preparation of nanospheres is described in Fig. 10.

606

Benita etal.

4

Anionic Polymerization

Fig. 10 Schematic representationof the nanosphere preparationtechnique using the anionic polymerization approach. Most of the methodsused for obtaining nanospheres are based on the polymerization of monomers introducedin the dispersed phase of an emulsion or W/O microemulsion or dissolved in a nonsolvent polymer environment. 73vo separate phases are distinguished, a nucleation phase and a growing phase.

1. Methods of Producing Nanospheres There are fourmain methods for obtaining nanospheres. a. Preparationinemulsion. Thecontinuous phase is composed of a mixture of monomer S A A water. The monomer + S A A combination produces micelles from 1to 10 nm. Polymerization occurs in the interiorof the micelle, and the particles expand owing to the continuous insertionof monomer molecules into the interior of the micelle until they reach asize of 200 nm. The monomer is soluble in the continuous phase. It is in this phase of polymerthat freeradicals (FR) are formed,which promote the initiation ization with the formation of an insoluble polymer. This process involves three stages: (1)the monomeris in the micelle; (2) a gradual incorporation

+

+

Cosmetic Applicationsof Vesicular Delivery Systems

607

of monomer molecules into the micelle is produced, giving rise to a polymerization initiated by FRS, and subsequently a reorganization of oligomers surroundedby SAA, which causes further growth of the micelle; and (3) The fusion of two micelles leads to a certain increasein size.

b. Dispersion-polymerization. The monomer is dissolved in the continuous phase, and nucleation is directly induced in the monomer solution without any diffusion. The oligomers which have formed then turn into aggregates stabilized by molecules of S A A . The polymer is obtained by growth of the aggregates. c. Polymerization in inverse microemulsion. Dilatednonaqueous micelles are stabilized by a layer of SAA and dispersed in the oil.The process of polymerization is continuous. The number of polymer particles which have formed increases over time, but theirsize remains constant. The formation of polymer particles results from the fusion of many micelles by collision. Thus, each polymer particle contains relatively few chains, which is in contrast to particles formed through other processes. d. Evaporation of the solvent. An organic polymer solution is emulsified in anaqueous phase and followed by evaporation of the solvent.If the solvent and nonsolvent are not chosen carefully, the organic solvent in which the polymer is dissolved can diffuse rapidly into the aqueous solution resulting in the precipitation of the polymer. Therefore, the choice of solvent and nonsolvent is a very delicate matter. They must be conducive to the formation of nanoparticles. Once the nanospheres are formed, the polymer solvent is evaporated under reduced pressure. 2. Methods of Producing Nanocapsules There aretwo principal methods of obtaining nanocapsules. Emuls@cation. Fora lipophilic substanceto be encapsulated, two phases, A and B, are mixed: monomer + alcohol + lipophilic substance represents the dispersed phase, whereas water + S A A represents the continuous phase. The monomer, which isinsoluble in water, is polymerized at the interfaceof the O M emulsion. For a hydrophilic substance, two phases, A and B, are both mixed. The hydrophilic substance is dissolved in water which is emulsified in the lipophilic phase containing the monomerusing anappropriate S A A resulting in a W/O emulsion. The monomer can be methyl, ethyl, or butyl cyanoacrylate. The monomer, which isinsoluble in water, is polymerized at the interfaceof the W/O emulsion. In case of an inverse microemulsion (W/O), the individual size is small. Rinsing is necessary for the solvent to be eliminated. a.

Benita et al.

608

b. Evaporation of the solvent. W Odifferent environments areused: S1 = polymer in solution in an organic solvent, and S2 = polymer nonsolventsolution. Oil to becapsulated must be miscible withS1and nonmiscible inS1+ S2. Oil must be minutelydispersed in S1 S2. The polymer capsulates the oil. inwhich lipophilic substances are dissolved. The encapsulation of lipophilic substances can thus be performed easily, whereas the encapsulation of hydrophilic substances is difficult. Whatever method is used, the critical point is the thickness of the membrane, which ranges between 3 and 10 nm, for a total size reaching between 150 and 800 nm.

+

C. Microemulsions

Owing to the stable thermodynamic character of microemulsions, it is straightforward and easy to obtain them [43-481. A stable W/Oemulsion obtained with a lipophilic SAA may be used as a base for preparing an O/W microemulsion. To this emulsion, a hydrophilic aqueous SAA solution is added followed by stirring. A gelified phase appears becauseof the cubic structure of the product. If hydrophilic SAA is again added, an O/W microemulsion is obtained. In order to preparea W/O microemulsion, anO/W emulsion should be used stabilized either by an ionic or a nonionic SAA. By means of titration, a cosurfactant (COS) is added. As in the preceding example, a gelified phase appearswhich becomes fluidand results in the formation of a W/O microemulsion. However,these methods are empirical and relatively crude. Likewise, a microemulsion is almost always created by the establishment of a pseudoternary diagram for which a ratio of SAA/CoS is fixed, representing a sole constituent. The establishment of a ternary diagram is generally accomplished for the purpose of locating the microemulsion or the microemulsion zonesby titration. Using a specific ratio of SAA/CoS, various combinations of oil and SAA/CoSare produced. The water is added drop by drop. After the addition of each drop, the mixtureis stirred and examined through a crossed polarized filter. The appearance (transparence, opalescence, isotropy) is recorded, along with the number of phases. In this way, an approximate delineation of the boundaries can be obtained in whichit is possible to refine through the production of compositions point by point beginning with the four basic components.

D. MultipleEmulsions The operational technique plays an even more important role in the production of multiple emulsions than in the production of simple emulsions [49-541.


Aqueous Phase

Oil + lipophilic surfactant

W10 Emulsion

Hydrophilic surfactant in water

WIOW Emulsion

609

(4

WKb Emulsion (B)

Fig. 11 Preparation of a multiple emulsion by a two-step procedure. (A) Step 1: formulation of W/O emulsion. (B) Step 2: formulation of W/Oi?V emulsion. Four main types of protocols are recommended:

1. The two-stage procedure as depicted inFig. 11 (the most frequently utilized method). The name is not precisely indicative, since other methodsinvolve two stages. 2. Dispersion of a lamellar phase. This method is still rarely used. It is similar to the proceduredescribed above for obtaining nonionic surfactant vesicles. 3. Dispersion of an isotropic oil solution in water. This procedure is recommended more for the formation of W/O/W-type microemulsions. 4. Phase inversion. This procedure isbecoming more commonly used. Here again, the name is inaccurate, since it does not actually involvea phase inversion. Each of the fourprotocols is executed in almost identical fashion: First, either a continuous oily phase simple emulsion,alamellar phase, or an isotropic oil solution is produced at 70-80°C ? 1. Whichever system is used, the water, the oil, and the emulsifiers (the proportionsof the three components vary depending on which

610

Benita et a/. type of system is desired) are combined with the help of a classic turbine mixer during a period of approximately 30 min. IA e r , for the protocol designated double-stage, the continuous oil phase emulsion is poured slowly into the aqueous phase. For the other three procedures, it is the water which is gradually introduced either into the lamellar phase or theisotropic oil emulsion or the continuous oily phase emulsion. The second dispersion is likewise produced by means of a turbine stirrer, most commonly at room temperature, during a period of approximately 30 min.

Of course, each procedure offers bothadvantages and disadvantages. The two-stage procedure has the major advantage of having very well-regulated steps; in theory, it is possible to fix the amount of internal water. Its disadvantage is that it is not very reproducible, although this is likely because of the second emulsification, which is a critical stage in the procedure. In fact, theprimary W/O emulsion is quite viscous and is difficult to disperse. This is the stage when a shearingshould be executed, and certain globules of oil which have just formed are atrisk of breaking down. If this happens, someof the internal water may become mixed withsome of the externalwater. The process which involvesdispersion in the waterof a lamellar phase has the advantage of necessitating only a single emulsification stage. The initial phase is an anisotropic phase; it is thermodynamically stable, and easy to achieve. One of the limitations of this procedure stems from thefact that notall of the surfacantsform a lamellar phase. Where such a phase does exist,the HLB (hydrophile lipophile balance) is often elevated,which is undesirable for thestability of a multiple emulsion. Furthermore, only a small quantity of oil (rarely exceeding 10%) is incorporated in the lamellar phase. This procedure is best adapted to obtainingnonionic surfactant vesicles, since, in this kind of vesicle, the amount of apolar substancesis smaller. The procedure thatinvolves the dispersion in water of an isotropic oil solution presents more or less the same advantage as the procedure detailed in the preceding paragraph, in that only one emulsification process is involved. The initial phase is a pure phase, stable, andeasy to obtain. The main disadvantage is the incorporationof a low concentration of water-soluble compounds in inverse micelles (seldomly exceeding 10%). The disadvantages which are common to these two procedureshave to do with the necessarily elevatedquantities of surface-active agents needed. Furthermore,because of the dispersion in water of the lamellaror isotropic oil phases, in which the initial water is not truly emulsified in the form of globules, it is difficult to know whether the water transported by


61 1

this dispersion still remains soluble, and if so, then to determine the total water quantityin the internal aqueousphase. The advantages of thephase inversion procedure arethat itis easy to perform and that it provides a precisely known rate of internal aqueous phase similar to thedouble-stage procedure.Moreover, unlike the doublestage procedure,it is reproducible. However, it entails the disadvantageof having two emulsification stages, andof being a very delicate procedure to perform. Indeed, even a very slight excess of water might be sufficient to transform the multiple emulsion into an aqueous simple emulsion type. E. LiquidCrystals

Special laboratory skills are necessary in order toobtain thermotropicliquid crystals [55,56]. These crystals are, like any chemical molecule, added at the same time as the other constituents of the vehicle in whichthey must, following a change in temperature, present a certain appearance. Obtaining lyotropic liquid crystals is quite easy, and any formulator can produce them. The process consists of mixing, usually by means of a simple turbine mixer, the solvent and the ingredients (e.g., fatty ethoxyl alcohol, phospholipid stearate of glycerol or PEG) which will produce the desired structureabove a certain concentrationlevel. Thus, a lamellar structure could easily be obtained. Most often this structure is not preparedby itself but rather is obtained duringthe preparation of the cosmetic formulations. Thus, acream or gel may be formed on the spot with ingredients which will produce liquid crystals equally well.

V. CHARACTERIZATION Since all the particulate systems described above are basically dispersions, the approach used for their characterizationis almost identical [57-731. The following features pertain to all of them: particle size distribution, morphology analysis, electrical charge nature, creaming or sedimentation rate, and so forth; rheological behavior; and rate.and extent of the encapsulated active substances. W Oapproaches are used for the characterization of these systems. The first is simple and of a systematic nature. Its objectiveis to determine rapidly and without difficulty the type of system obtained. A.MicroscopicAnalyses

Microscopic examination is the first test performedin order toidentify the type of particulate systems obtained [57-651. Moreover, it constitutes an,

612

Benita et al.

Fig. 12 TEM photomicrograph of multilamellar liposomes following negative staining with phosphotungstic acid.

excellent means of following up on the physical stability of these systems as a functionof prolonged storage times. The most studied characteristic of liposomes is their dimensions. If the vesicles are medium sized (on the orderof 1pm), they can be examined under any ordinary optical microscope, a polarized light microscope, or a fluorescent microscope using a markerlike carboxyfhoresceine. However, examination under an electron microscope is necessary for nanoparticles and liposomes smaller than 1 pm, as depicted in Fig. 12. In any event, in order to measure the size, and in order to ascertain the particle distribution of these two types of systems, other methods which are faster and more accurate than microscopic techniques must be used. As for any other dispersed system, the technique used involves the electronic counting of the particles for vesicles that have (for the most part) a diameter greater than 600 nm. Photon-correlation spectroscopy is used for size determination of vesicles of smaller size. For multiple emulsions, optical microscopy isalso used as a common method of analysis, as shown in Fig. 13, to keep track of all these systems and



613

Fig. 13 Photograph of a WIOIW multiple emulsion obtained with a normal light microscope (at a magnificationof 1OOO).

to determine the particle size distribution. This method allows the direct measurement of the size of multiple globules with a diameter greater than 0.5 pm, as well as a rough estimate of the percentage of multiple globules compared with simple globules. Furthermore, thesize of the internal aqueous globules (usually on the orderof a micrometer) can be determinedby this method. Prior to observation, and in order tomeasure the actual size of the globules, the multiple emulsions must be well diluted with a solution whose osmotic pressure should be similar to that of the internal aqueous phase. Indeed, dilution with a solution of lower molarconcentration would, after the ingress of external water intothe internal water, cause swelling and then rupture of the internal aqueous globules. Inversely, dilution with a more highly concentrated solution would precipitate the escape of the internal water and hence ashrinking of the internal aqueous globules. Furthermore, examination under polarized light could sometimesdetect a texture corresponding to a lamellar phase indicating the directional orientation which the two emulsifiers take in the oily phase. As for the vesicular systems described above, microscopy following cryofracture of the multiple emulsion is used in order to enablea precise analysis of the internal aqueous globules. The objective here is to generate accurate information regarding their exact morphology.

Benita et al.

614

Fig. 14 Photograph of lyotropic liquid crystalsforms with polarizing microscope. Some researchers instead of obtaining a direct estimate following observation prefer to use photographs for measurement, and in this way, they follow up on thesize and stability of the emulsions. Finally, examination under a polarizing microscopeis also a valid tool for detecting andobserving liquid crystals, as shown in Fig. 14.

B. RheologicalAnalysis In order to gain a deeper knowledge of particulate systems, we often turn to rheology [66-701. Because of the diversity offered by this technique, itis possible to characterize the structure of such systems and to follow their evolution over time. In termsof rheological analyses, it means: l . ViscoelasticOscillator Analyses. The oscillatory experiment consists of applying a sinusoidal stress and recording theconsecutive strain defined as follows: T(f) = T,COSWt E(t) = €,COS(Wt

+ S)

where T, and eo are maximal amplitudes of stress and strain, W = 27rN, with N the frequency of strain and S the phase angle of stress/strain. The basic viscoelastic parameters which describe the rheological behavior are:


615

G* = TJE,, is the rigidity modulus, and S the phase angle of the stress with respect to the strain (for a viscoelastic material, it is included between 0 and 90"; for apurely elastic material, it equals0" and for a is the critical stress at perfect newtonian liquid, it equals 90"); and (T")~ which the sample becomes more viscous than elastic.The comparison of (T")~and G*/27r values gives information about the form and the homogeneity of the droplets.

2. Analyses of the Permanent Flow System. During the process of sweeping, the system is sheared in a cycle of increasing constraint, then constant constraint, and then diminishing constraint. Thiskind of test provides information about the functionof shearing in the destructuration of the dispersion. It also provides information about themore or less reversible character of the dispersion. More simply stated, in this kind of analysis, the Newtonian character of these systems, taking microemulsions for example,can be exposed.

C. Encapsulation or Incorporation Efficiencyof the Active Substances The assay of the active substance incorporated into one of the dispersion phases provides useful information on the encapsulation ability of the systems [71-731. The location of the active substance in the dispersion phase depends on its affinity for the various constituents of the formulation. Hydrophilic substances are dissolved in the aqueous phases, whereas lipophilic substances are dissolved in the oily phases. The rateof encapsulation depends on a number of factors, like the concentration of the active substance, the nature of the constituents, the type and size of the vesicles. Prior to the determination of the amountof active substance encapsulated, it is nevertheless preferable, once the dispersed systems have been formed, to eliminate the nonencapsulated active substance. Separation is accomplished by means of various procedures, for example filtration on gel, ultracentrifugation, ordialysis. VI.

STABILITY

A.

Liposomes

To date, the production of stable liposomes isstill delicate and chancy [74-791. It should nevertheless be kept in mind that, as for all dispersions, these vesicles have a tendency toward flocculation and fusion and later sedimentation.

Benita et al.

616

Likewise, instability is also associated with the increase in permeability of the envelope and the resulting leakage of the encapsulated active substance, as, for example, it is likelyto occur when the rateof cholesterol in the envelope is insufficient. Degradation of phospholipids is largelyrelated to autoxidation, which can be markedly decreasedif antioxidants are added. Furthermore,oxidation of liposomes can be avoided by using hydrogenated phospholipids. Another cause of degradation is hydrolysis of the constituents of the membrane which results in the release of fatty acids, an increase in the an in membrane permeability. All of fluidity of the membrane, and increase these phenomena can lead to fusion of the vesicles and eitherpartial or total leakage of the encapsulated active ingredient. Protection against hydrolysis and autoxidation as well can be ensured by coating the liposomal vesicles with a membrane comprising biologicalmacromolecules,atelocollagen, and glycosaminoglycanes. The second coating ensures a diminution in the liposomal membrane fluidity as a result of interaction between the phospholipids and the macromolecules. Thisleads to a reduction in the permeability of the membranes. Various physical procedures, such as irradiation with gamma rays or cryodessication, promise to provide answers to increasing the stability of these systemsin the future. Generally speaking, the stability of liposome in gelified aqueous or hydroalcoholic environments ranges between 2 and 3 years at temperature between 4" and 25°C. However, liposomes remain stable for only a few months if they are dispersed (and this is quite frequently the case in cosmetic products) in a lipid-rich environment or in a solution containing surfactants in which the phospholipid envelop dissolves or gradually becomes soluble. B.

Nanoparticles

As long as organic solvents, residual monomers, and polymerization inducers are eliminated, then problems in the stability of nanoparticles are practically nonexistent. Nevertheless, in order toavoid anyeventual degradation of the polymeric materials in aqueous suspension or hydrolysis of the active substance after release, cryodessication can be performed without causing any alteration in the size.

C. Microemulsions Since microemulsions are stable thermodynamic systemswithin a defined temperature range, they present no problems for storage under normal conditions.


617

D. MultipleEmulsions

Multiple emulsions are unstable thermodynamic systems; over time they progress inevitably toward their rupture [77-791. In addition to the traditional factorsof instability inherent in simple emulsions (separation, creaming, coalescence, phase inversion), specificunstable conditionsoccur within multiple emulsions such as the diffusion of internal water into the external aqueous continuous phase. These instabilities are all irreversible. Theymay appear separately orconcurrently. Today, through the use of very effective polymeric surface-active agents and thickeners that increase the viscosity of the various phases (particularly cross linking of the intermediateoily phase through theuse of chemical or physical processes like gamma radiation), we are able significantly to slow down the destabilization process of these systems. In fact, stablemultiple emulsions can be prepared for a periodof 2 years without any alteration of their characteristics or releaseof the encapsulated substance. E. LiquidCrystals Liquid crystals do not exhibit any problem of stability provided they are stored within specific temperature ranges, since above such temperatures they lose theirproperties (including even lyotropic liquid crystals). It should be noted that the presence of lamellar structures at the emulsion interface greatly increases theirstability. VII. COSMETICUSES

A. Liposomes Liposomes are theprincipal vehicles for the transportof active ingredients, which, depending on their molecular size and solubility properties, are localized at different sitesof the liposomes [80-991. If the active ingredients are small hydrophilic molecules such as sugars, aminoacids, peptides, ornormal moisturizing factors (NMF), they are localized at the centerof the vesicle or between the bilayers. If they are lipophilic molecules such as liposoluble vitamins and their esters, they are located in the lipid bilayer. The transformation of these lipid vesicular carriers when they come is still the subjectof a strong controinto contactwith the stratum corneum versy. The presence of globular structures in the first layers of corneal cells was firstassumed and later confirmed through various transmission electron microscopic (TEM) techniques. Furthermore, the presence of multilamellar structures in the deeplayers of the stratum corneum has also been observed through cryofracture.Given the fact that intercellular distances are eitherof

618

Benita et al.

6.4 or 13.4 nm, itis surprising to find entirely intactvesicular structures with a diameterof between 50 nm and 0.5 pm in the interiorof bilayers with such anarrow thickness. It isnow known thatthe molecules constituting liposomes become dispersed during thecourse of their intercellularmigration, and their reactions with the cutaneouslipids are more or less a function of their chemical structure. An accumulation of phospholipids develops at certain lipid sites, andthis leads to the formation of new vesicular structures at these sites. It has been demonstrated that phosphatidylcholine, having a relatively modest hydrophilic moiety, is more likely to react than phosphatidyl inositol choline, since the latter has a heavier hydrophilic moiety group. It is thus a dynamic structure whose initial globular form is indispensable in penetrating the cutaneous barrier. Extensive and thorough knowledge regarding interactions between liposomes and living cells is currently available. These interactions studied with routes of administration other than topical may call into play four different phenomena:

1. Absorption, facilitated by the charge on the vesicles due to the ionization of the primary or secondary component. 2. Lipidic transfer, through a mediation of surface proteins (here an analogy between lipids, liposomes, and membrane lipidscan be seen). It is important to emphasize the role played by molecules such as lipoproteins in the constitutionof the membrane. 3. Endocytosis, also influenced by the charge on the particles. This permits thedigestion of liposomes by lysosomes inthe interiorof the cell, and this promotes the releaseof the active ingredient at the same location. 4. Fusion, which results from the insertion of constituents of the liposome membrane into thecell membrane. This simple and attractive hypothesis has, however, not been proven to date. Because of their biomimetism with the cellular membranes, all of these lipid systems are designed as vehicles for carrying active ingredients, particularly hydrophilic ones, sometimes lipophilic, andensuringprolonged release of such ingredients. Although the natureof interactions between liposomes and the skin remains unexplained to date, these systems have been used continuously since the initial creationof a "liposomed" cosmetic product manufactured by Dior more than a decadeago and called Capture. The first cosmetic applications of liposomes were in the field of hydration of the skin. It was demonstrated that products containingliposomes, into which propylene glycol was added, possessed excellent moisturizing properties. This hydrating capacity was prolonged and extended with a


619

product containing liposomes constituted from glucosylceramides, an ester of glucose of fatty acid, and an ester of sucrose or trehalose of fatty acid. The following products have been tested: Creams or lotions made up of lysophospholipids, an aqueous phase containing monovalent and polyvalent alcohols, an oily phase able to containhydrocarbons, esters, triglycerides, fatty acids, and fatty alcohol as well as silicone. Dispersions constituted from a mixture of an emulsion containing elastin, collagen, casein, and fibroin, and a liposomal suspension containing vitamins E, F, and A. Liposomes made of cutaneous lipids dispersed in gel.

1

Currently, a large number of active substances used in cosmetology have been encapsulated, presented, and dispersed in hydroalcoholic liquids, aqueous gels, or creams with a continuous aqueous phase. Most cosmetic compounds, whether intended foruse on the face, the hands, the body, or thehair, are appreciatedby those who use them. Because of their bilamellar structure, nonionic surfactant vesicles have the same propertiesas liposomes. However, since the nature of lipids found in the envelope is different, their properties are not entirely identical. Of course,thesame controversies existwith regardtononionic surfactant vesicles as with regard to liposomes insofar as their passage through the skin is concerned. Even if we accept that only vesicles on the penetrate the corneal layer intact, it is still order of 20 nm in size can undeniable that vesicles with nonionic surface-active agents have some effect after topical application. The first results were reported by Handjani-Vila and colleagues [88,91,96,97], whoshowed that sodium carboxylate pyrolidone has the greatest hydrating power when it is encapsulated in nonionic surfactant vesicles and not when it is encapsulated in emulsions or even in liposomes. Recently, nonionic surfactant vesicles without any active ingredients, composed of fatty ethoxyl alcohols and cholesterol, were tested on human skin. Microscopic examination following cryofracture of a cross section of skin reveals that the vesicles would be located in intercellular lipid sites in a depth of a few micrometers. This confirms the stratum corneum to results whichshow,with the aid of images obtained by Electron Paramagnetic Resonance (EPR) and plates obtained through microscopy following cryofracture of biopsy a of human skin, that intercellular restructuring of the epidermis takes place after contact withsuch systems. Like liposomes, nonionic surfactant vesicles will then restore to a delipidified corneal layer the lamellar structures normally contained therein. It was shown that incorporation of estradiol in nonionic surfactant vesicles (par-

620

Benita et al.

ticularly multilamellar structures) substantially enhances penetration of this ingredient into the stratum corneum of a piece of skin. Although themechanism by which these carriers operate remains to be explained, theynevertheless constitute very useful systems for cosmetic applications. Without any active substance, like liposomes, the nonionic surfactant vesicles enhance the supply of lipids and water of the stratum corneum. Even when they do not penetrate the stratum corneum intact, theyfacilitate theaccumulation of hydro- or lipophilic active substances in theupper layers of the epidermis. Finally, like liposomes, nonionic surfactant vesicles exhibit a high cutaneous tolerance. Preparations containing nonionic surfactant vesicles were first placed on the market at approximately the same time as the liposomes described in the preceding sections. They too achieve satisfactory results for people who use them knowledgeably. For these systems as well, the first cosmetic applications were intended as hydrating agents, andthey were followedby self-tanning ingredients of the skin. Prevention of dry skin has been studied with empty vesicles. With their component ingredients, thesesystems work against water loss by forming an occlusive film on the surfaceof the epidermis. As mentioned earlier, their hydrating capacity has been studied by encapsulating sodium carboxylate pyrolidone (PCNa), a componentof the skin’s natural hydrating system (NMF). The hygroscopic substances which make up the NMF are often used as hydrating agents in cosmetic compositions. However, these composites, which are soluble in water, develop a weak affinity for the corneal layer once they are incorporated into aqueous solutions or in oily emulsions within water. They penetrate the stratum corneum with onlymoderate success, and are, aside from that,easily eliminated by washing with water. If such substances are administeredin a lipid environment, their diffusion and the resistance to washing are increased. The application of nonionic surfactant vesicles containing 10% PCNa will, even afterrinsing, markedly improve the hydration of the skin by 40-90% as compared with the initial moisturizing level before treatment. One of the first tests in the study of coloration of the skin was performed with vesiclescontaining 0.6% of tartaric aldehyde and,as controls, aqueous solutionsof various concentrations of the same component. Four hours after application, both before and after washing, the intensityof coloration producedby these vesicles was as high as the intensity produced by the aqueous solutioncontaining 10 times the quantityof tartaric aldehyde. A second, parallel test was conducted with vesicles containing 1.5% of tartaric aldehyde and 3% of dihydroxyacetone and an O/W emulsion. An intense tan was obtained with these vesicles, which was resistant to

Cosmetic Applications Vesicular of Delivery Systems

62 1

washing with soap and water, whereas with the O/W emulsion, the tan, which was already weak before washing, disappeared almost completely after washing. These nonionic surfactant vesicles have been in continuous use since their first commercial exploitation as Niosomes by L'Oreal. They are more and more commonly used, and a number of active substances have been encapsulated. Evenif, as for liposomes, the precise mechanism by which they operate on application to the skin remains yet to be determined, their biomimetic approach, like that of liposomes, has inspired a growing number of researchers, as their vesicular carriers constitute an undeniably valuable asset to cosmetology. B. Nanoparticles

Although cosmetic applications of nanoparticles proliferate (numerous patents have been granted),publications, studies, or reports on their biofate in the skin following topical application have been rare. The incorporationof active cosmetic hydrophilic substances (e.g., amino acids, vegetable extracts, organicand mineral elements) in the nanospheres attempts to modulate the releaseof the substances in the skin. Where nanocapsules are concerned, theactive substances (AS) are usually of lipophilic nature, and they can be composed of an oily compound or a dispersion. Here again the objective is to control the release of the AS, since the molecule is protected, during a shorter longer or period, from biodegradation in the organism. The release profile of the AS (by, e.g., erosion, diffusion, elution) depends on the natureof the constituents.Recently, Lancome launched a cosmetic product containing nanocapsules of vitamin E (Primordiale). As for the vesicular systems described in the preceding sections, except for very small sizes which can be detected in the pilosebaceous apparatus, it is unlikely that they penetrate thestratum corneum intact, since their size varies from 100 to 800 nm. C. Microemulsions

Over the last 10years, many studies have dealt with the percutaneous absorption of various active ingredients carried by microemulsions both from a pharmacological point of view and a cosmetological perspective [100-107]. Overall, hydrosoluble active ingredients have been the most sought after and (rathercuriously) the ingredients most often added to the external phase of O/W microemulsions. Various investigators have shown that microemulsions acted as absorptionenhancers of both liposoluble and hydrosoluble active substances. This action was not just attributed to the high proportion of surface-active agents. Still, when the hydrosoluble active element is in the internal phaseof a W/O microemulsion, its physicochemical characteristics are altered. Be-

622

Benita et al.

cause of this, and owing to the components of the fatty phase, it can accumulate in the cutaneous lipophilic structures or in the cutaneous fats (orotic acid).Liposoluble active ingredients have been introduced both in the internaland in the externalmicroemulsion phases compared with W/O or O/W emulsions or with petroleum jelly (Vaseline). Nevertheless, they are more frequently incorporated in the internal phase, where they are solubilized. This is the case with alpha-tocopherol, azelaic acid, and octyl dimethyl para-aminobutyric acid, for which the absorption rate is significantly increased, and which no longer behave like lipophilic compounds but like hydrophilic ones. When estradiol or prednisone is introduced in the internalO/W microemulsion phase, it is stored in the intercellularlipids of the stratum corneum. Estradiol placed in the external phase is better retained. Thus, it can be observed that depending on the formulation and the type of the microemulsion, the absorption of an active ingredient can be modulated on request. There are numerous cosmetic products in the form of microemulsions; these products range from body care to facial and hair treatments. They include bath oils, body thinning products, fixatives for hair, hardeners fornails, hydrating products, antiwrinkle products, productsto prevent seborrhea, and antiaging serums marketed principally in France, Italy, Belgium, and the United States.

D. Multiple Emulsions The behavior of the multiple emulsions on the skin following topical application has not yet been addressed [108-1101. Do small-sized globule vesicles in the range of 20 nm penetrate intact? In what manner does the rupture of larger-sized globule vesicles take place? Are they reformed on contact with the intercellular lipids? For multiple emulsions, these questions do not present a problem. Although the releaseof the encapsulated active substance is complicated, since a number of different mechanisms exist, themultiple emulsion’s behavior after application to theskin appears to be relatively simple, since it is quite similar to the behavior observed with regard to simple emulsions. If evaporation of the water after application to the skin leading to other structuresis not taken into consideration, two principal hypotheses may be proposed. In the first hypothesis, the encapsulated active substance has a near zero diffusion rate through the oily membrane. Theactive substance is thus released in the internal phase only by virtue of the rupture of the multiple oily globules. This rupture takes place either by shearing of the preparation or afterswelling of the internal aqueousglobules. Shearing may be induced by massaging or rubbing performed when the creamis applied to theskin.


623

Swelling of the internal aqueous globules may be causedby dilution of the multiple emulsion in water. It is possiblethat theinstructions for use could require ,mixing immediately prior to application of a given quantity of the multiple emulsion with a specific volume of water. The strong osmotic gradient thus created causes the external water to enter the internal phase, causing swelling of the internal water globules. When theseglobules reach their maximum size, they burst. In both of these cases, rupture by shearing or rupture afterswelling, the active substance, having reached the external aqueous phase, becomes immediately available. The multiple emulsion behaveslike a simple emulsion with a continuous aqueous phase. In this case, multiple emulsions serve the purpose of protecting (by encapsulation) an active substance or of permitting the incorporation into the same preparation of incompatible active substances which will not come intocontact until the cream is used. In the second hypothesis, the encapsulated active substance diffuses through the oily membrane, and the multiple emulsion keeps its multiple globules intact when applied to theskin. The release profile of the active substance from a multiple emulsion would, in this case, not be similar to that froma simple emulsion. It would be slower and gradual, and it would depend on the following factors: the 1ocalization.andphase partition distribution of the active substance and its permeability or diffusion rate through the oily membrane; the rigidity, the viscosity, and the thickness of the interfacial film as well as thepresence or absence of liquid crystals at this level; the particle size and distribution of the dispersion, and otherfactors. Since they do notconcentrate or accumulate theactive substance at the vesicular systems described above do, level of the stratum corneum as the they at least extend the release rate of the active substance. Their fate after application to theskin has been thesubject of few publications to date;only the works of Kundu [l101 and Ferreira [l081 on thesubject are well known. Multiple emulsions used as cosmetic preparations generally come in the form of lotions or creams of varying density. As soon as they are constituted, they can be used directly; unlike the vesicular systems described above, it is not necessary to disperse them in a gel environment or in a creamin order toobtain an acceptable cosmetic product. Cosmetic applications of multiple emulsions have been protected in the patents issued for theircomposition. One example of an application is perfume encapsulated in the internal phase: very small amounts of it are released over a long period of time. However, it could be released instantaneouslyby intentionally breaking the primary emulsion; for example, by rubbing during application to the skin. The patents show that multiple emulsions are highly recommended for all kinds of cosmetic applications:

'

al. 624

et

Benita

for example, sunscreens, make-up removers, cleansers, nutritive, hydrating, and refreshing and cooling products. The mixture of a continuous oily phaseemulsion with an aqueous solution containing, in addition to a hydrophilic surface-active agent, saccharides (such as xylose, lactose, or sorbitol), polysaccharides (such as xantham gum, pectin, carraghenate, or dextran), or synthetic polymers(such as acrylic derivatives) allows the obtention of multiple emulsions presenting cosmetic qualities of good spreading and unctuousness, which can be modified on request. In these systems, it is possible to incorporate a number of active substances both in the internal andexternal aqueous phases and even in the intermediateoily phase without significantly altering their stability. We cite as an examplea formula which produces a fairly viscous cream that remains stable for a period of 36 months at25°C. All other things being equal, W/O/W multiple emulsions are more effective than simple emulsionswith a continuous aqueous phase and are more pleasant to use than simple emulsions with a continuous oily phase because of a less greasy feel. Multiple emulsions will certainly enjoy a future in cosmetics at least as bright as that of the vesicular systems described above. Simple emulsions are the building blocks of cosmetology. Multiple emulsions, like their simpler form, give skin both water and oil; they can comprise a large number of ingredients, both hydrophilic and lipophilic; they are easy to administer, since they can be applied directly to the skin; and theypresent good cosmeticqualities. Like any vesicular system, multiple emulsions constitute a new form which could prove extremely fruitful even if only for the possibilities it presents for the protection of encapsulated substances and perhaps for prolonged release of substances. The firstcommercial use of a W/O/W-type multiple emulsion is Unique Moisturizing by Lancaster, which was introduced to the marketin 1991.

E. Liquid Crystals For reasons of cosmetic appeal dueto thecolored appearance theygive to preparations into which they are introduced, and for reasons of active substance solubilization, or simply because they increase the stability of dispersed systems, liquid crystals are enjoying a growing popularity [ l l l ] . Until now, liquid crystalshave beenselected only rarely as absorption enhancers, because they contain such a high proportion of surface-active agents. In fact, poor cutaneoustolerance was observed whenliquid crystals were usedas the principal vehicle. After dispersion in a gel, their tolerance is no longer a problem.


625

Thermotropic liquid crystals were introduced in a cosmetic product for the first time by the Estee Lauder Company, with the launching of Eyxone, a translucid gel.Today, liquid crystals incorporated in microcapsules made of gelatin which rupture ontopical application are available (Merck). Lyotropicliquid crystals are incorporatedin special dermatological formulations thatexhibit hydrating properties (Liphaderm). Most of all, liquid crystals are used as excipients to protect sensitive, active substances (vitamins, antioxidants, oils). Liquid crystals may enhance the stability of emulsions while creating arheological barrier resulting inan increasein the viscosity and a decreasein coalescence by modification of Van der Waals forces. It is also claimedthat they enhance cutaneous penetration. REFERENCES 1. G. Gregoriadis,P. D. Leathwood, and B. E. Ryman, Enzyme entrapment in liposomes, FEBS Lett.,14:95-99 (1971). 2. C. G . Knight, Liposomes, in Physical Structure to Therapeutic Applications, Elsevier, Amsterdam,1981. 3. M. J. Ostro, Liposomes, Marcel Dekker, New York,1983. 4. F. Puisiew and J. Delattre,LesLiposomes:applicationstherapeutiques, Lavoisier Tecet Doc, Paris,1985. 5. G. Gregoriadis, J. Wiley, and S. Chichester, Liposomes as Drug Carriers, Wiley, Chichester, UK,1988. 6. J. Delattre, P. Couvreur, F. Puisiew, et a1 al., Lesliposomes:Aspects 7. 8.

9. 10. 11. 12. 13. 14.

technologiques, biologiques, et pharmacotechniques, Lavoisier Tec et Doc, Paris, 1993. B. Magenheim andS. Benita, Nanoparticles characterisation. A Comprehensive physicochemical approach,STP Pharma Sci., 4:221-241 (1991). F. Fevrier, Etude de microemulsionsnon ioniques comme agentde penetration cutanee, Doctoral thesis, de 1’Universitie de Technologie de Compiegne, France, 1991. W. J. Herbert, Multiple emulsion, a new form of mineral oil antigen adjuvant, Lancet, 2:771-773 (1965). M. DeLuca,C.Vaution,A.Rabaron,andM.Seiller,Classification et obtention des emulsions multiples,STP Pharma Sci., 4:679-687 (1988). J. P. Caquet andT. Bemoud, Les cristaux liquides dans les cosmetiques, Parf. Cosmet. Aromes, 91:77-84 (1990). G. Cioca and L. Calvo, Liquid crystals and cosmetic applications, Cosmet. Toiletries, 10557-62 (1990). A. D. Bangham, M. W. Hill, and N. G. A. Miller, Preparation and use of liposomes as a model of biological membranes, in Methods in Membrane Biology (E. D. Kom, ed.), Plenum Press,New York, 1974, pp. 1-68. J. M. Gebicki and M. Hicks, Preparation and propertiesof vesicles enclosed by. fatty acid membranes, Chem. Phys. Lipids,16:142-160 (1976).

626

Benita et a/.

15. C. Kirby, J. Clarke, and G. Gregoriadis, Effect of cholesterol content of small unilamellarliposomes on their stability in vivo and in vitro, Biochem.J., 186~591-598 (1980). 16. A. W. T. Konings, Lipid peroxidation in liposomes, in Liposome Technology (G. Gregoriadis, ed.), CRC Press, Boca Raton,FL, 1984,pp. 139-161. 17. S. Stainmesse, H. Fessi,J. Devissaguet, andP. Puisieux, Procede de preparation de systemes colloidaux dispersibles de lipides amphiphilessous forme de liposomes submicroniques, Brevet No.88.08.874,France, June,1988. 18. D. Marsh, Handbook of Lipid Bilayers, CRC Press, Boca Raton, FL, 1990. 19. D. J. A. Crommelin, Influence of lipid composition and ionic strengthon the physical stability of liposomes, J. Pharm. Sci.,73:1559-1263 (1990). 20. L. M. Prince, Microemulsions: “Theory and Practice,” Academic Press, New York, 1977,p. 177. 21. L.M.Prince,Micellization,solubilizationandmicroemulsions,inMicellization,SolubilizationandMicroemulsions, (K. L. Mittal,ed.),Plenum Press, New York, 1977,pp. 45-54. 22. S. E. Friberg and L. G a m o , Microemulsions with esters, J. Soc. Cosmet. Chem., 34:73-81 (1983). 23. I. Rico, Les microemulsions: definition et applications pratiques, Vol. VII, Les entretiens du Carla,(P. Fabre, ed.), Castres,1986,pp. 22-26. 24. A. Ceglie, K. P. Das,andB.Lindman,Microemulsionstructureinfourcomponent system for different surfactants, Colloid and Surfaces, 28:29-40 (1987). on the microscopic 25. A. Ceglie, K. P. Das,andB.Lindman,Effectofoil structure in four component cosurfactant microemulsions, J. Colloid and Interface Sci. B, 115:115-120 (1987). 26. C.H.ChewandL. M. Gan,Monohexyletherofethyleneglycolanddiethylene glycol as microemulsion cosurfactants, J. Dispersion sci. Technol., 11~49-68 (1990). 27. A. T. Florence and D. Whitehill, The formulation and stability of multiple emulsions, Int.J. Pharm, 11:277-308 (1982). 28. S. Magdassi, M.Frenkel, andN. Garti, Correlation between nature of emulsifiersandmultipleemulsionstability,DrugDev.Ind.Pharm., 11:791-798 (1985). 29. C. Fox, An introduction to multiple emulsions, Cosmet. Toiletries, 62:lOl112 (1986). 30. C. Prybilski, M. de Luca, J. L. Grossiord, et al., w/o/w multiple emulsions 106:97manufacturing and formulation considerations, Cosmet. Toiletries, 100 (1990). 31. T. Suzuki, M. Nakamura, H. Sumida, and I. Shigeta, Liquid crystal make-up remover:conditionsofformationanditscleansingmechanisms,J.Soc. Cosmet. Chem., 43:21-36 (1992). 32. L. Saunders, J. Pemn, and D. B. Gammack, Aqueous dispersion of phospholipids by ultrasonic radiations, J. Pharm. Pharmacol., 14567-572 (1962). 33. A. D. Bangham, M. M. Standish, and J. C. Watkins, Diffusion of univalent


34. 35. 36. 37. 38. 39. 40. 41. 42. 43.

44. 45.

46.

47.

48. 49. 50.


627

ions across the lamellae of swollen phospholipids, J. Mol. Bioi., 13:238-252 (1965). C. H. Huang, Studies on phosphatidylcholine vesicles. Formation and physical characteristics, Biochemistry, 8:344-352 (1969). S. Batzri and E. D. Kom, Single bilayer liposomes prepared without sonication, Biochem. Biophys. Acta, 298:1015-1019 (1973). D. W. Deamer and A. D. Bangham, Large volume liposomes by an ether evaporation method, Biochem. Biophys. Acta, 443:629-634 (1976). J. M. H. Kremer,M.W, J. Esker, N.C.Pathmamanohara,and P. H. Wiersema, Vesicles of variable diameter prepared by a modified injection method, Biochemistry, 16:3932-3935 (1977). F. C. Szoka and D. Papahadjopoulos, Procedure for preparation of liposomes with large internal aqueous surfacehigh andcaptureby reverse phase evaporation, Proc. Natl. Acad. Sci. USA, 75:4194-4198 (1978). Y.Barenholz, S. Amselem, and D. Lichtenberg, A new method for preparation of phospholipid vesicles (liposomes), FEBS Lett., 99:210-215 (1979). T.Oshawa, H. Miura, and K. Harada, A novel method for preparing liposomes with a high capacity to encapsulate protein and drugs, Chem. Pharm. Bull., 32:2442-2445 (1984). C. Vauthier-Holzcherer, S. Benabbou, G. Spenlehauer, et al., Methodology for the preparation of ultradispersedpolymersystems, STP PharmaSci., 1:109-116 (1991). M. Bornschein, P. Melegari, C. Bismarck, andS. Keipert, Mikro- und Nanopartikeln alsarzneistofftragersysteme unter besonderer Berucksichtigung der herstellungsmethoden, Pharmazie, 44585-593 (1989). N. Kamenka, G. Haouche, B. Brun, and B. Lindman, Microemulsions in zwitterionic surfactant systems, J. Colloid Interface Sci., 86:359-369 (1982). J. Franz (S. A. Sandoz), Compositions pharmaceutiques topiques sous forme d’une microemulsion, Fr. Patent Application 2-502-951 (30/03/1982). B. W. Muller and P. Kleinebudde, Untersuchengen an sogenannten mikroemulsions systemen1Teil: Unter suchungen an arzneistoffhaltigen systemen, Pharm. Ind., 50:370-378 (1988). B.W. Muller and P. Kleinebudde, Untersuchengen an sogenannten Mikroemulsions systemen 2 Teil: Unter suchungen an arzneistoffhaltign systemen, Pharm. Ind., 50:1301-1306 (1988). F. Fevrier, Formulation de microemulsions cosmetiques, Nouv. Dermatol., 10:84-87 (1991). M. F. Bobin, N. Dejour, A. Revillon, and M.C. Martini, Formulation of ultrafine emulsions with surcroesters for Pharmaceutic and Cosmetic applications, Congres Mondialde l’emulsion, Paris, France, 1993. S. Matsumoto, Y. Kita, and D. Yonesawa, An attemptat preparing water in oil in water multiple phase emulsions, J. Colloid Interface Sci., 57:353-361 (1976). D. R. Kavaliunas and S. G. Franck, Liquid crystal stabilization of multiple emulsions, J. Colloid InterfaceSci., 66586-588 (1978).

628

et

Benita

al.

S. Matsumoto andP. Sherman, A preliminary study of wlolw emulsions with a view to possible food applications,J. Texture Studies,12:243-257 (1981). 52. S. Matsumoto, Developmentof wlolw type dispersion during phase inversion of concentratedwlo emulsions,J. Colloid and Interface Sci.,94:362-368 (1983). 53. S. Magdassi,M. Frenkel, N. Garti, On thefactorsaffectingtheyield of 51.

preparation and stability of multiple emulsions, J. Dispersion Sci. Technol.,

5~49-59(1984). 54. M. de Luca,Les emulsions multiplesWLJH. Obtention, validation,et libera191, 1991. tion, These de 1’Universite Paris XI, No. 55. C.Colinet,ContributionaI’etudedesystemesmesomorphesapartir de surfactifs non ioniques, These de pharmacie, Lyon I, 1991. 56. E. Coupier, Contribution a l’etude de la stabilite des systemes mesomorphes. 57.

58. 59.

60. 61. 62.

63.

Influence de la temperature et des divers adjuvants, These de pharmacie, Lyon I, 1991. P. Schurtenberger and H. Hauser, Characterization of the size distributionof unilamellar vesicles by gel filtration quasi-elastic light scattering and electron microscopy, Biochim. Biophys. Acta,778:470-480 (1984). H. Ruf, Y. Georgalis, and E. Grell, Dynamic laser light scatteringto determine size distributions of vesicles, Methods Enzymol., 172:364-372(1989). Z.Kojro, S. Q. Lin, E. Grell, and H. Ruf, Determination of internal volume and volume distribution of lipid vesicles from dynamic light scatteringdata, Biochim. Biophys. Acta,9851-8 (1989). R. R. C.New,Characterization of liposomes,inLiposomes,APractical Approach (R. R.C. New, ed.), IRL Press, McLean, Virginia,1990. D. Atwood andG . Ktistis, A light scattering study on olw microemulsion,Int. J. Pharm, 52:165-171 (1989). R. C. Baker,A. T. Florence, R. H. Ottewill, andT. H. Tadros, Investigations into the formation and characterizationof microemulsions. 11. Light scattering conductivity and viscosity studies of microemulsions,J.Colloid Interface Sci., 100:332-349 (1984). S. Keipert, I. Siebenbrodt, F. Luders, and M. Bornschein, Mikroemulsionen 44:433-444 undihrepotentiellepharmazentischenutzung,DiePharmazie, (1989).

64. J. C. Selser, Y.Yeh,and R. J. Baskin, A light scattering measurement

of membrane vesicle permeability, Biophysics, 16:135-139 (1976). 65. S. S. Davis, A.S. Burbage, Electron micrographyof wlolw emulsions, J. Coll. Interface, 62(2):361-362 (1977). 66. Y. Kita, S. Matsumoto, and D. Yonezawa, Viscosimetric method for estimating the stabilityof wlolw type multiphase emulsions, J. Colloid Interface Sci., 67.

62:87-94 (1977). S. Matsumoto, T. Inoue, M.Koda,and

T. Ota, Anattempttoestimate J. Colstability of the oil layer in wlolw emulsions by means of viscosimetry, loid Interface Sci.,77564-564 (1980). 68. J. I. Zatz and G . H. Cueman, Assessment of stability in water in oil in water multiple emulsions,J. Soc. Cosmet. Chem., 39:211-222 (1988).



629

69. A. A. Elbary, S. A. Nour, and I. Ibrahim, Physical stability and rheological propertiesofw/o/wemulsionsasafunctionofelectrolytes,Pharm. Ind., 52~357-363 (1990). 70. J. L. Grossiord, M. Seiller, andF. Puisieux, Apport des analyses rheologiques dans l'etude des emulsions multiples I", Rheologica Acta, 32:168-180 (1993). 71. E. H. Ralston,L. M. Jelmeland, R. D. Klausner, et al., Carboxyfluoresceine as a probefor liposome cell interactions. Effect of impurities and purification of the dye, Biochim. Biophys. Acta, 649:133-137 (1981). 72. S. S. Davis and I. Walker, Measurement of the yield of multiple emulsion droplets by a fluorescent technique, Int. J. Pharm., 17:203-213 (1983). 73. J. A. Omotosho, T. L. Whateley,T. K. Law, and A. T. Florence, The nature of the oil phase and release of solutes from multiple wlolw emulsions, J. Pharm. Pharmacol., 58:865-870 (1986). 74. P. A. Poly, Liposomes d'un produitde contraste iode. Influence de la formulation sur l'encapsulation Etude de la conservation par lyophillisation. Doctoral thesis No. 65,3"'' cycle, UniversitC Paris-Sud, 1983. 75. S. Henry Micheland, P. A. Poly, F. Puisieux, et al., Etude du comportement desliposomeslorsd'experiencesde congelatioddecongelation et de lyophillisation. Influence des cryoprotecteurs, 3rd International Conference on Pharmaceutical Technology APGI, Paris, 1983, pp. 223-133. et al., Prevention of fusion and 76. L. M. Crowe, C. Womersley, J. H. Crowe, leakage infreeze-driedliposomes,Biochim.Biophys.Acta,861:131-140 (1986). 77. T. K. Law, T. L. Whateley, A. T. Florence, Stabilization of wlolw multiple emulsions by interfacial complexation of macromolecules and non-ionic surfactants, J. Controlled Rel., 3:279-280 (1986). 78. A. 02, P. Kamles, P. Franck, and G. Sylvan, Multiple emulsions stabilized by colloidal microcrystalline cellulose, J. Dispersion Sci. Technique, 10:163-185 (1989). 79. M. de Luca, J. L. Grossiord, J. M. Medard, et al., A stable w/o/w multiple emulsion, Cosmet. Toiletries, 10565-69 (1990). 80. M. Mezei andV. Gulasekharam, Liposomes. A selective drug delivery system for the topical route of administration, Life Sci., 26:1473-1477 (1980). 81. A. Rougier, D. Dupuis, C. Lotte, et al., In vivo correlation between stratum corneumreservoirfunctionandpercutaneousabsorption.J.Invest.Dermatol., 81:275-278 (1983). 82. J. Lasch, S. Zellmer, W. Pfeil, and R. Schubert, Interaction of liposomes with the human skin lipid bamer: hSCLLD as model system-DSc of intact stratum corneum and in situ CLSM of human skin, J. Liposome Res., 5:99-108 (1995). P. Pemer, New vehicleto enhancethe biologi83. G. Redziniak, A. Meybeck, and cal performances of active cosmetic ingredient, Vol.11, XIVth International Federation Societies of Cosmetic Chemists Congress, Barcelona, 1986, pp. 299-307. ;&l 8 4 . Skincosmeticscontainingoilinwaterdispersionscontainingnaturalsur-

630

Benira er al.

factantsandliposomewaterdispersions,PolaChemicalIndustries,Inc., French Patent 261 4787 (10111988). 85. Skin treatment products containing lysophospholipids, Eur. Patent 25 59 37 (17 02 1988). 86. Topical compositions containing liposomes and acrylic gels, Estee Lauder 06 1989). Inc., Eur. Patent 3 196 38 (14 87. A. Meybeck, F. Bonte, G. Redziniak, Improvement of comedolytic activity andtoleranceofvitaminAacidintopicalliposomeformulation,Vth Congres Int. Tech. Pharm., Pans, 1989. 88. R. M. Hanjani-Vila andJ. Guesnet, Les liposomes. Un avenir prometteur en dermatologie, Ann. Dermatol. Venerol., 116:423-430 (1989). 89. W. Abraham and D. T. Downing, Interaction between corneocytes and strain vitro, Biochim. Biophys. Acta, 1021:119-125 tum corneum lipid liposomes (1990). 90. H. Lautenschlager, Liposomes in dermatological preparations. Part I, Cosmet. Toiletries, 105:89-96 (1990). 91. E. Clar and A. Ribier, Lipid vesicles in cosmetology, Cosmet. Toiletries Mf., 2~138-147 (1991). 92. M. Ghyczy, J. Roding, and E. Hoff, Control of skin humidity by liposomes, Cosmet. Toiletries Mfg., 2:148-152 (1991). 93. H. E. J. Hofland, J. A. Bouwstra, F. Spies, et al., Interactions betweennonionic surfactant vesicles and human stratum corneum in vitro, J. Liposome Res., 5241-263 (1995). 94. F. Bonte, J. M. Chevalier, and A. Meybeck, Determination of retinoic acid liposomalassociationlevelinatopicalformulation,DrugDev.Indus. Pharm., 20:25-34 (1994). 95. R. M. Handjani-Vila and G. Vanlerberghe, Les Niosomes, in Formes Pharmaceutiques nouvelles. Aspects technologique, biopharmaceutique, et medical, (P. Bun, F. Puisieux, and J. B. Benoit eds.), Lavoisier Tec et Doc, Pans, 1985. Les Niosomes des structures 96. G.VanlerbercheandR.M.Handjani-Vila, restructurantes pour les soins de laNouv. peau,Dermatol., 5259-262 (1986). 97. G . Vanlerberghe, R. M. Handjani-Vila, and A. Ribier, Les niosomes. Une nouvelle famille de vesicules a base d’amphiphiles non ioniques Colloques Nationaux du CNRS No. 938, Masson,Pans, 1989, pp. 303-311. 98. V. Gabnjedeie, M. Sentjure, and J. Kristl, Evaluation of liposomes as drug J. Pharm., carriers into the skin by onedimensionalEPRimaging,Int. 62~75-79 (1990). 99. H. Schreier, J. Bouwstra, Liposomes and niosomes as topical drug carriers: dermal and transdermal drug delivery, J. Controlled Rel., 3O:l-15 (1994). 100. M. C. Martini, M. F. Bobin, H. Flandin, J. Cotte, Absorption and fate of orotic acid after topical application in different vehicles, J. Appl. Cosmetol., 2~19-26 (1984). 101. M. C. Martini, M. F. Bobin, H. Flandin, and F. Caillaud, Role des microemulsions dans l’absomtion Dercutaneede l’alpha-tocopherol,J. Pharm. Belg., 39:348-354 (1984).



631

102. M. Gallarate, M. R. Gasco, and G. Rua,In vitro release of azelaic acid from oil in water microemulsions, Acta Pharm. Jugosl.,40533438 (1990). 103. E. E.Linn, R. C. Pohlan, and T. K. Byrd, Microemulsion for intradermal deliveryofcetylalcoholandoctyldimethylPABA,DrugDev. Indus. Pharm., 16:899-920 (1990). 104. F. Fevrier, M. F. Bobin,C.Lafforgue,andM.C.Martini,Advancesin STP Pharma microemulsions and transepidermal penetration of tyrosine, Sci., 1:60-63 (1991). 105. V. Burigana, Mise au point de dosagede la N-acetyl hydroxyproline.Etude I, de l'absorption cutanee de cette molecule, These de pharmacie, Lyon 1993. In vitro cutaneous penetrationof 106. F. Fevrier, M. F. Bobin, and M. C. Martini, proline with microemulsion formulation compared to emulsion, Symposium International Federation Societiesof Cosmetic Chemists, Barcelona, 1993. 107. F. Fevrier, Microemulsions in cosmetology, LCR Intetnational Federation Societies of Cosmetic Chemists Congress, pp. 105-118, Barcelona, 16-19/9/ 1993. 108. L. A. M. Ferreira, Emulsions multiples WWH et simples LEI et WL. Etude comparative pur l'approche de la voie topique, These Universite de ParisSud, No. 370,1994. 109. M.Seiller,A.M.Orecchioni,andC.Vautions,VesicularSystemsin cosmetology, in Cosmetic Dermatology (R. Baran andH. I. Maibach, eds.), London, Martin Dunitz, 1994, pp. 27-35. 110. S. C. Kundu, Preparation and evaluation of multiple emulsions as controlled release topical drug delivery systems, Avail Univ. Microfilms, Dep. Abstra. Int. Order no. DA 9224664 from Dirs Abstr. Int. B 1990 51 (1990) No. 4 S1763-4. 112. G. Cioca and L. Calvo, Liquid crystals and cosmetic application. Cosmet. Toiletries, 10557-62 (1990).

a


Index

Adsorbents, 359 Aedes aegvpti, 404 Aggregation, 307 Albumin, 166, 172,599 Alginate, 366 Alkylcyanoacrylates,364 Alum, 401 Amphotericin B, 394 Anesthetics local, 389 Anomalous diffision, 174 Anopheles srephensi, 404 Anthracyclines, 320 Antibiotics, 394 Antibody, 351,362 IgG, 401 monoclonal, 545-546,554,368 Antifly case study, 22,23,24 Arabic gum, 51 Arginase, 360 Atomization, 50,61 tower, 53,57,61 Atomizer, 55,57 method, 52 AUC, 404 Bangham, 602 Benzalkonium chloride, 363 Biocarrier, 591 Biocompatibility, 369

Biodegradable polymers, 36,385 Biovector, 591 Blood cells, 359 Box and Wilson test, 104 Brij96,33 1 Bromocriptine, 53 Bupivacaine, 385,389 Busereline, 53

c

Caking, 22 Carbon dioxide, 59,61 Carrageenan, 389 Cells, 35 1 artificial, 360 microencapsulated, 367 Centrifugation, 160, 161 filtration, 161 Centrisart@,161 Characterization, 611 Chemotherapy for cancer, 535-537 Chloroform, 356 Chlorpromazine, 163 Cholesteryl hexadecyl ether, 407 Chymotrypsin, 360 Coacervation, 22,23,27,36-37,41 Coated particles, 168 Colloidal drug carrier, 214-216 lipid-based, 215-218

633

Index

634 [Colloidal drug carrier] polymeric, 214-215 Composition, 597 liposomes, 597 liquid crystals, 601 microemulsions, 599 multiple emulsions, 600 nanoparticulate systems, 598 Copolymer of ethylenepropylene oxide, 601 of styrene, 599 Cosmetics, 299 Cosmetic uses liposomes, 617-621 liquid crystals, 624 microemulsions, 621 multiple emulsions, 622-624 nanoparticles, 621 Co-surfactant, 599 Counting of particles electronic, 612 Critical point, 57 Crystallinity of colloidal glyceride crystals, 239 of stored tripalmitate nanoparticles, 242

Crystallization in the dispersed state, 232 Cyclohexane, 356 Cytotoxicity, 406 Deconvolution, 162 DEET, 403 Denaturation heat, 25 surface, 26 Dermal uptake,403 Dexamethasone, 379 Dextran, 599 Dialysis, 159, 164 reverse, 160 tubing, 386 Diazepam, 160,387 Differential scanning calorimetry, 236 of melt-emulsified glyceride dispersions, 236-243 Differential thermal analysis (DTA), 135

Diffision, 156-181 Diffisional exponent, 174

l-B-Diphenyl-l-3-5-hexatriene, 581 DLVO, 312 Doxorubicin, 321,537,542-543,547,552553,555-557,562-565

DPH (see 1-6-Diphenyl-l-3-5-hexatriene)

Drug delivery of, 299,362 leakage of, 216-252 release of, 156-181 resistance, 542-543,557 EGF, 265-266,288 Embryonic microparticles, 50 Emulsifiers, 413,424,432 in baked goods, 448 fany acids, 464 fatty alcohols, 464 HLB of, 435 lecithin, 450,457 lysolecithins, 474 in milk, 427,448 monodiglycerides, 45 1,465,5 12 monomeric, 464 polymeric, 419,440,466 spans, 465,480,484,487 Sodium stearoyl lacrylate (SSL), 464 stabilization by ,419,440,466,484 Tweens, 465,480,486,487 Emulsion, 24,26,3 1, 159, 163 double vs. multiple, 475,494 adjuvants, 494 anticancer, 497 diffusion controlled, 479 droplet size distribution, 473,375 flux, 483 formation, 475 intravenous, 495 oil-in-water-in-oil (OIWIO), 475, 494

oral intake, 496 osmotic pressure, 480 release of drugs, 494 release of foods, 494 release mechanism, 477,482 by reverse micelles, 480 stability, 477,482 transdennal, 495 transport through, 477,482 water-in-oil-in-water (W/OIW), 475, 494

multiple, 47,49, 170, 172, 608,616,622 non aqueous,47 oil-in-oil, 47 oil-in-water, 45

oil-in-water-in-oil, 594 stabilization, 440,444 aggregation, 427,4267 bridging flocculation, 427 coalescence, 419

594,

599,

Index

635

[Emulsion] colloidal, 445 creaming, 427 depletion, 428, DLVO theory, 430 electrostatic, 429 flocculation, 427 HLB, 435 by hydrocolloids, 440,460 mechanical, 444 by monomeric emulsifiers, 419 by polymeric emulsifiers, 440,466, 484

by proteins, 440,484 required HLB, 435 sedimentation steric, 440 water-in-oil, 351 water-in-oil-in-water, 47,594 Encapsulation, 22,23,453,518 or incorporation efficiency of the active substances, 615 Endocytosis, 581 Enzymes, 61,351,359 kinetics, 361 Epitaxical effect of emulsifier in lipid-in-water dispersions, 246 of phospholipids in melt-emulsified glyceride dispersions, 239 Ester sorbitan, 601 Estradiol, 336 Ethoxylated fatty alcohols, 601 Etoposide, 388 Experimental design, 97 principle application, 97 of the Z3type, 104 of the 33type, 105 methodology, 100 Exponential function, 165-168, 175 Fibrinogen, 360,361 Film, 27 Filtration, 161 Flow system permanent, analyses of, 615

Foods,421,440,448 colloids, 448 emulsifiers, 424,427,448 emulsions, 427,448 margarine, 421,427 Formaldehyde, 365 Freeze-drying, 164 Fusion, 581

y-irradiation, 380 Gelati, 172 Gelatin, 25,27,31,599 Gel formation of lipid suspensions, 227-229,230 prevention by coemulsifying agents, 227

of phospholipid stabilized solid lipid nanoparticles, 232-233 in phospholipid-stabilizedtripalmitate suspensions, 242-249 model of, 247-248 prevention of, 243-245 studied by transmission electron .. microscopy, 245-249 Genes as therapy, 543,544 Glycolic acid, 36,53,599 Grinding technique, 62 Growth factors, 265-266,276,288 Hadamard matrix, 100 Hard fat nanoparticles isothermal recrystallization of, 241 recrystallization of, 236 supercooling of, 240-241 synchroton radiation x-ray diffraction Of, 234-236 Hepatocytes, 365 Hexamethylenediamine, 352,353 1,6-Hexanediamine, 352,,353 High pressure homogenization, 225,229 Higuchi law, 165, 169 Hisitidase, 360 Histidinemia, 360 HLB, 357 Hyaluronic acid, 172, 173 Hydrocolloids, 460,486 emulsifiers, 484 galactomannans, 460 gelatin, 486 gum arabic, 486 stabilizers, 484 xanthan gum,460 Hydrolysis, 356 Hydrophile-lipophile-balance(HLB), 357 Immune system, 299 Immunoadsorbents,359 Immunoliposomes, 272,275-277 . Immunomodulator muramyl dipeptide, 543 muramyl dipeptide-L-alanyl-cholesterol, 543,563,564

Indomethacin, 169

636 Insect repellent, 403 Interfacial addition polymerization, 349 complexation, 349,366 condensation polymerization, 349 polyelectrolyte complexation, 350,366 polymerization, 349,352 reactions, 605 transport, 163,166-168 Intralipid, 156 Ion exchange, 166 Islets of Langerhans, 368 Isocarbacyclin, 163,164 Kelvin equation, 240 Lack of fit test, 107 Lactic acid, 36 ,53,599 Lamellar structure, 602,611 Lanreotide, 43 Lattice, imperfections, 236,239 Lecithin, 164 Leuprolide, 47 LHRH analogue of, 41,47,62 Linked polysiloxames, 599 Lipid A, 402 Lipid emulsion, 216 Lipid nanopellets, 218-219 Lipid suspensions (see also Solid lipid nanoparticles) as alternative delivery system for poorly water-soluble drugs, 216-218 incorporation of drugs into, 249-251 of melt-emulsified glycerides, 225-231 crystallinity of, 236-243,250 crystallinity index of, 239,242 degree of crystallinity, 236,243 gel formation of, 243-249 recrystallization process of, 236-243 time course of polymorphic transitions, 236 prepared by melt-emulsification, 225231 physical state of, 233-236 physicochemical characterization of, 232-249 synchrotron radiation x-ray diffraction of, 234-236,240 prepared by precipitation in emulsified organic solvents, 223-225 polarized microscopy of, 246-247 Lipomicron, 591 Lipoproteins, 215-216 low-density (LDL),215-216

Index Liposomes, 164, 168, 173,215,259-295, 297,329,333,401,578,589,597, 602,615,617 (see also Vesicles, lipid) bilayer mechanics, 305 bioadhesive, 276 cutaneous cell interactions, 580 drug biological activities. 287-290 drug encapsulation, 261,262,267,278279,281-283 drug release from, 261,267,279,281, 283-286 as pharmaceutical products, 259-295 stability, 260-261,267,276-281,300 biological, 309 chemical, 306 colloidal, 307 physical, 301 sterically stabilized, 300,311 sterility, 277-278 surface modification, 274-277 thermodynamics, 303 types and sizes, 260-264 Lipospheres, 377-410 antibiotics, 394 blank, 393 characterization, 380 composition, 396 drug. dlstribution, 386 loading, 385 release from, 386,396 elimination of, 399 insect repellent, 403 for intramuscularor subcutaneous injection, 222 local anesthetics, 389 morphology, 380 nabolipospheres, 406 particle size, 384 preparation, 379 prepared from microemulsions, 219-221 stearic acid, 220 sterilization, 382,384 for sustained drug release, 221-223 timolol-containing, 221 toxicity of, 393 vaccines, 399 Liquid crystals, 595,601,611,617,624 cubic, 417,509 hexagonal I, 417,506 lamellar, 415,506 reverse hexagonal, 417,506 stabilization, 415 L-lysine, 353,354

Index L W (see Vesicles, large unilamellar) Lyotropics, 601 Magnetic, 362 Maltodextrin, 51 Marangoni effect, 197,205 Marcaine, 387 Matricial systems, 52 Melt-homogenization, 225 Melting point depression of melt-emulsified lipid suspensions, 240 of dispersed glycerides, 239 Membrane fluidity, 582 rigidity, 582 Methotrexate, 172 Micelles, 329,334 Micellizations aggregation number, 427 critical micellar concentration, 413 formation, 413 rod-like micelles, 417,507 size, 4 17 solubilization, 413 spherical, 417,506,508 surfactants exchange, 427 swollen micelles, 416,418,502 Microcapsules, 36-37,52,61,349,350 definition, 1 shell materials of alginate, 15 enteric polymers, 13 ethyl cellulose, 6 fats, 13 gelatin, 3, 10 gum arabic, 3, 10 hot-melt, 15, 17 maltodextrins, 11 melamine-fomaldehyde, 9 modified starch, 10 nylon, 7 polyamide, 7 polyurea, 7 polyurethane, 7 urea-formaldehyde, 9 Microemulsions, 593,599,608,616,621 Microencapsulationprocesses, 36,46,50 centrifugal extrusion in, 15 classification of, 2,3 complex coacervation in, 3 fluidized bed coating in, 13 in situ polymerization in, 9 interfacial polymerization in, 7 polymer/polymer incompatibility in, 5

637 [Microencapsulation processes] rotating submerged orifice in, 10 rotational suspension separation in, 16 spray drying in, 10,25,27,50,56 static submerged nozzle in, 10 Microparticle, 156-18 1 Microscope analyses, 611 fluorescent, 612 ordinary optical, 612 polarized light, 612 Microsomes, 367 Microspheres, 36-38,41-44,47,50-53,55, 61,350,448,458,482

aging of, 138 cyanoacrylate, 163 lipid, 218 magnetic, 161, 173 morphology cisplatin, 151 hydrocortisone, 144 lomustine, 145 poly(d,l lactide), 134-135 progesterone, 135 wax, 222 Mifepristone, 53 Milling methods, 62 Mitomycin C, 159 MLV (see Vesicles, multilamellar large) Mosquitoes, 403 Motor blockade, 391 Nanocapsules, 191-197,204-206,207,593 definition, 94 Nanolipospheres, 406 Nanoparticles, 183-208,616,621 cholesteryl acetate, 223-225 morphology of, 224 transmission electron microscopy of, 223-224

definition, 94 fatty acid, 219-221 formation mechanism, 109, 124 hard fat, 234 magnetic cancer chemotherapy, 565-566 distribution, 555-557 polymeric, 2 14-2 15 systems, 592,599,605 tripalmitate, 224,234 Nanospheres, 184-191, 197-204,207,593 albumin, 204 definition, 94 gelatin, 203 polyacrylamide, 202

638 [Nanospheres] polyalkylcyanoacrylate,202 poly-c-caprolactone, polylactide-coglycolic, 202 polymethacrylate, 202 polystyrene, 202 Nebulization, 50, 52,55 laboratory, 53 Niosomes, 590 N,N-diethyl-m-tolumide, 403 Nonionic surfactant vesicles, 329,330 drug entrapment, 329,333 interaction with skin, 336 penetration enhancement of, 377 transport, 336 physical state, 329,333 preparation, 329,331 ether injection method, 332 handshaking method, 332 reversed phase evaporation, 332 sonication method, 331 Nylon 6-10,353 Oil silicone, 37-40 Opsonization, 564 Oxytetracycline, 394 Particle, 156-181 size, 384 distribution, 156, 163 Partition coefficient, 159,355 PEG, 407 Perfluorocarbon derivative, 601 Permeability, 363,368 Pesticides, 1 formulation q.c., 15, 16, 17 introduction to formulations of, 1-5 pH-activity curve, 361 Pharmacokinetics, 393,546-557 Phase inversion, 609 separation, 10, 13, 14,23 PHCA, 36-37,40-41,49,62-63 Phenylketonuria, 359 pH-Activity curve, 361 pH measurement, 163 Phospholipids, 379,589,597 Piperazine, 352,353 PLA, 36,40-41,46,55-57,61-62,95,98,109 PLAGA, 36-38,40-41,53,55,57,61 Plasmodiumfalciparum, 400 Polarizing microscopy of lipid-in-water dispersions, 232,233, 246-247

Index Polarography, 163 Poloxamer, 228 as steric stabilizer of lipid suspensions, 228 Polyacrylate, 367 Poly(a-hydroxy-carboxylicacids), 35,47 Polyalkylcyanoacrylate,599 Poly(alkyl-2-cyanoacrylates), 364 Polyamide, 351 Polyamine, 353 Polyanhydrides, 171 Polycaprolactone, 402 Poly(d,l-lactic acid-coglycolic acid), 37 Poly(d-l lactide), 169, 171 Polyester, 95, 353 Polyethylene glycol, 300,407 Polyglycerol, 601 Poly(iminocarbonates), 173 Poly(1actide-co-glycollide), 169, 171, 173 Polylysine, 366 Polymer albumin, 555-556 brush, 312 desolvatation, 202 membrane, 35 1 phase-separation 36,43 poly(alky1 cyanoacrylate), 537,547,557 poly(alky1 methacrylate), 537,547,552 poly(buty1 cyanoacrylate), 547,553 polyglutaraldehyde,537 poly(hexy1cyanoacrylate), 545 poly(isobuty1cyanoacrylate), 542-543, S57 ". poly(isohexy1cyanoacrylate), 544,547, 552-553.563-564 poly(1actic acid), 537,554 Polymeric barrier, 50 core, 385 surfactant, 601 Polymerization, 183-208 anionic, 195 dispersion, 198 emulsion, 185, 191, 195, 198, 199 interfacial 6,9, 191, 192, 193, 197, 198, 204,349 interfacial condensation, 349 Polymorphic transitions of glycerides, 233 in melt-emulsifiedglyceride dispersions, 236-243 time course of, 241 Poly(ortho-esters), 173 Polyoxyethylene alkyl ether surfactants, 331

lndex Polyoxyethylene sorbitan, 601 Polyurea, 353,363 Polyurethane, 353 Preferential adsorption of surfactants to glyceride nanocrystals, 248-249 Production liposomes, 602 liquid crystals, 61 1 microemulsions, 608 multiple emulsions, 608 nanoparticulate systems, 605 Products thermally sensitive, 55 Progesterone, 46 Proliposomes, 598 Proteins, 61,351,440,484 absorption on latex, 440 absorption on liquid interfaces, 5 18 Bovine serum albumin (BSA),440,443, 487,490 denaturation, 203-204 as emulsifiers, 440 as food colloids, 440,484 lysosyme, 440,484 modified chemically, 519 modified enzymatically, 519 soy proteins, 440,484 Pulsed release, 173 Randall-Selitto instrument, 389 Recrystallization from dispersed lipid melt, 232,240-241 of melt-emulsified glyceride nanoparticles, 236-243 of melt-emulsified trilaurate, 229 Release measurement, 158-163 model, 157,165-176 rate, 156-181 Response surface methodology, 99 Rheological analysis, 614 Rose bengal, 162 Sebacoyl chloride, 352,353,354 Sensory block, 391 Skin absorption, 403 SMV (see Vesicles, small multilamellar) Sodium dodecyl sulfate, 27,31 Sodium glycocholate, 228 Solidification of lipid suspensions, 230 Solid tipid nanoparticles (see also Lipid suspensions) drug release from, 253

639 [Solid lipidnanoparticles] incorporation of drugs into, 249-251 in vitro cytotoxicity of, 252 of melt-emulsified glycerides, 225-23 1 crystallinity of, 236-243,250 crystallinity index of, 239,242 degree of crystallinity, 236,243 recrystallization process of, 236243 time course of polymorphic transitions, 236 prepared by melt-emulsification,22523 1 gel formation of, 227-229,230 lyophilization of, 230-231 mean particle size of, 227 spray-drying of, 230 storage stability of, 225,227-228 prepared by sonication, 227 Solvent evaporation, 43,45 extraction residual, 43,56,61-62 Somatostatin analogue of, 62 Spectroscopy photon correlation, 612 Spheronization, 62-63 Spray congealing, 62 Stability, 615 liposomes, 615 liquid crystals, 617 microemulsions, 616 multiple emulsions, 617 nanoparticles, 616 window, 37-40 Steric stabilization, 228 Sterol, 598 Storage stability of solid lipid nanoparticles, 225,227228,252 Sucrose ester Wasag-7,331 Supercooling degree of, 232 of hard fat and tripalmitate dispersions, 240-241 of melt-emulsified trilaurate, 229 Supercritical conditions, 57-60 fluid, 58,60-61 phases, 58,61 Supramolecularbiocarriers, 591 Surface activity, 25,26 Surface modification of colloidal drug carriers, 231

lndex

640 [Surface modification] by modifying agents, 554-557(see also Antibody, monoclonal) to reduce uptake by reticuloendothelial system, 231 of tripalmitate nanoparticles, 231 Surfactants, 357,413,424,432,599 anionic, 464,467 cationic, 464 HLB, 435,465,477,497 hydrophilicity, 438 micellization, 413,487 non-ionic, 438,450,460,464,600, non-ionic vesicles, 590,610 surface tension, 472,502 wetting, 473 zwitterionic, 471 SUV (see Vesicles, small unilamellar) Targeting, 259-277 Taxol, 406 Technologies solvent-free, 62 Terephthaloyl chloride, 352,353,354 Ternary diagrams, 37 Therapy photodynamic, 554455,557 Thermotropics, 601 Tissue distribution, 407 Toxicity, 537,543,562 cardiotoxicity, 552,564 embolism, 562 neutropenia, 565 Transferosomes, 329,334 Transmission electron microscopy of cholesteryl acetate nanoparticles, 223-224 of melt-emulsified tripalmitate suspensions, 245-249 Triglycerides polymorphic behavior of, 233 Tripalmitate nanoparticles crystallinity of stored, 242 crystal shape of, 245 DSC heating scans of, 239 gel formation of, 243-249 gel structure of, 246 prepared by precipitation in emulsified solvent, 224 stabilized by phospholipids only, 239 DSC of, 239 supercooling of, 240-241 surface-modificationof, 23 1 synchrotron radiation x-ray diffraction Of, 234-236

pripalmitate nanoparticles] transmission electron microscopy of, 245-246 ubidecarenone-loaded, 250 zeta potential of, 23 1 Triptoreline, 43 Tumor EMT-6 mouse mammary tumor,557 hepatoma, 552 leukemia, 537,557,566 Lewis lung carcinoma, 553 metastases, 553454,562463,565466 models, 322 sarcoma, 553,555,557,565 Turbine method, 52 Two-stage procedure, 609 Tyloxapol, 228 as steric stabilizer of lipid suspensions, 228 Ultracentrifugation, 164 Ultrafiltration, 161 Ultraviolet spectroscopy, 161-163 Uptake cellular, 536-546 Urea, 359 Urease, 359 Vaccines, 399 Vesicles large unilamellar, 330 lipid, 330 multilamellar large, 330,590 phospholipid, 578 skin interaction with, 329,335 small multilamellar, 579 small unilamellar, 330,579,590 in colloidal tripalmitate suspensions, 246,247-248 transdermal application of, 329,334 Viscoelastic oscillator analyses, 614 Vitamin A acid, 583 Witepsol W35,233,249 X-ray diffraction, 234 synchrotron radiation, 234 of melt-homogenized glyceride dispersions, 234-236 time-resolved during temperature scans, 236 Zeta potential of surface-modifiedtripalmitate nanoparticles, 23 1

Microencapsulation (Drugs and the Pharmaceutical Sciences)

Containment in the Pharmaceutical Industry (Drugs and the Pharmaceutical Sciences)

Microencapsulation: Methods and Industrial Applications, 2nd edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Project Management. Drugs and the Pharmaceutical Sciences

Handbook of Pharmaceutical Analysis (Drugs and the Pharmaceutical Sciences)

Advanced Pharmaceutical Solids (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Project Management, 2nd edition (Drugs and the Pharmaceutical Sciences)

Transport Processes in Pharmaceutical Systems (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Engineering: Second Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Powder Compaction Technology (Drugs and the Pharmaceutical Sciences)

Excipient Toxicity and Safety (Drugs and the Pharmaceutical Sciences)

Handbook of Drug Screening (Drugs and the Pharmaceutical Sciences)

Transdermal Drug Delivery, Second Edition, (Drugs and the Pharmaceutical Sciences)

Handbook of Drug Screening (Drugs and the Pharmaceutical Sciences)

Biomarkers in Clinical Drug Development (Drugs and the Pharmaceutical Sciences)

Handbook of Bioequivalence Testing (Drugs and the Pharmaceutical Sciences)

Handbook of Bioequivalence Testing (Drugs and the Pharmaceutical Sciences)

Modified-Release Drug Delivery Technology (Drugs and the Pharmaceutical Sciences)

Nanoparticle Technology for Drug Delivery (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fourth Edition, Revised and Expanded (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fourth Edition, Revised and Expanded (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fifth Edition (Drugs and the Pharmaceutical Sciences)

Transport Processes in Pharmaceutical Systems (Drugs and the Pharmaceutical Sciences: a Series of Textbooks and Monographs)

Handbook of Pharmaceutical Analysis (Drugs and the Pharmaceutical Sciences: a Series of Textbooks and Monographs)

Pharmaceutical Statistics: Practical and Clinical Applications, Fifth Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fifth Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Engineering (Drugs and the Pharmaceutical Sciences: a Series of Textbooks and Monographs)

Percutaneous Absorption: Drugs, Cosmetics, Mechanisms, Methods, Fourth Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Validation: An International Third Edition (Drugs and the Pharmaceutical Sciences)

Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, 3rd Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Engineering: Second Edition, Volume 195 (Drugs and the Pharmaceutical Sciences)

Microencapsulation (Drugs and the Pharmaceutical Sciences)

Containment in the Pharmaceutical Industry (Drugs and the Pharmaceutical Sciences)

Microencapsulation: Methods and Industrial Applications, 2nd edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Project Management. Drugs and the Pharmaceutical Sciences

Handbook of Pharmaceutical Analysis (Drugs and the Pharmaceutical Sciences)

Advanced Pharmaceutical Solids (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Project Management, 2nd edition (Drugs and the Pharmaceutical Sciences)

Transport Processes in Pharmaceutical Systems (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Engineering: Second Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Powder Compaction Technology (Drugs and the Pharmaceutical Sciences)

Excipient Toxicity and Safety (Drugs and the Pharmaceutical Sciences)

Handbook of Drug Screening (Drugs and the Pharmaceutical Sciences)

Transdermal Drug Delivery, Second Edition, (Drugs and the Pharmaceutical Sciences)

Handbook of Drug Screening (Drugs and the Pharmaceutical Sciences)

Biomarkers in Clinical Drug Development (Drugs and the Pharmaceutical Sciences)

Handbook of Bioequivalence Testing (Drugs and the Pharmaceutical Sciences)

Handbook of Bioequivalence Testing (Drugs and the Pharmaceutical Sciences)

Modified-Release Drug Delivery Technology (Drugs and the Pharmaceutical Sciences)

Nanoparticle Technology for Drug Delivery (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fourth Edition, Revised and Expanded (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fourth Edition, Revised and Expanded (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fifth Edition (Drugs and the Pharmaceutical Sciences)

Transport Processes in Pharmaceutical Systems (Drugs and the Pharmaceutical Sciences: a Series of Textbooks and Monographs)

Handbook of Pharmaceutical Analysis (Drugs and the Pharmaceutical Sciences: a Series of Textbooks and Monographs)

Pharmaceutical Statistics: Practical and Clinical Applications, Fifth Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Statistics: Practical and Clinical Applications, Fifth Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Engineering (Drugs and the Pharmaceutical Sciences: a Series of Textbooks and Monographs)

Percutaneous Absorption: Drugs, Cosmetics, Mechanisms, Methods, Fourth Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Validation: An International Third Edition (Drugs and the Pharmaceutical Sciences)

Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, 3rd Edition (Drugs and the Pharmaceutical Sciences)

Pharmaceutical Process Engineering: Second Edition, Volume 195 (Drugs and the Pharmaceutical Sciences)

Recommend Documents