DEVELOPMENT OF AN ENCAPSULATION TECHNIQUE IN CONTROL OF PATHOGEN MOU HONGHUI NATIONAL UNIVERSITY OF SINGAPORE 2005 DEVELOPMENT OF AN ENCAPSULATION TECHNIQUE IN CONTROL OF PATHOGEN MOU HONGHUI (B.Eng (Hons). NUS) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF ENGINEERING DEPARTMENT OF CHEMICAL AND BIOMOLECULAR ENGINEERING NATIONAL UNIVERSITY OF SINGAPORE 2005 ACKOWNLEDGEMENTS Firstly, I would like to take this opportunity to thank my supervisors, Dr. J. Paul Chen and Prof. Jimmy Kwang, for their extensive support, guidance and encouragement throughout the course of this thesis. I would also like to thank Ms. Du Qingyun, Mr. Kit Wai and others in Prof Kwang’s laboratory for helpful advice during the work. I would like to thank Ms. Nitar for providing the partially purified HA and invaluable suggestions. I would like to acknowledge the help of several final year students, Mr. Benjamin Wong Choon Kiat, Mr. Tang Zhan Sheng, and Miss Toh Shi Ling in the Department of Chemical & Biomolecular Engineering for their help at different times and thank them for the same. I would like to extend my gratitude to Madam Susan Chia and Madam Li Xiang in the Department of Chemical & Biomolecular Engineering for help with equipment and various odds and ends. I also thank my colleagues in Dr Chen’s laboratory, Mr. Zou Shuaiwen and Mr. Yang Lei for their support and help. Last but not least, I would like to thank the Department of Chemical & Biomolecular Engineering for the financial and logistic support. i TABLE OF CONTENTS ACKNOWLEDGEMENTS i TABLE OF CONTENTS ii SUMMARY v NOMENCLATURE vii ABBREVIATIONS ix LIST OF FIGURES x LIST OF TABLES xii CHAPTER 1 INTRODUCTION 1 1.1 Motivation and objectives 1 1.2 Organization of the thesis 3 CHAPTER 2 BACKGROUND AND LITERATURE REVIEW 2.1 Influenza and H5N1 virus 4 4 2.1.1 Influenza and its virus 4 2.1.2 Influenza pandemic and its impact 7 2.1.3 Avian influenza (Bird flu) and H5N1 virus 9 2.1.4 Prevention and treatment 10 2.2 Vaccination through a delivery system 11 2.3 Common microencapsulation techniques 13 2.3.1 Solvent evaporation/extraction 13 2.3.2 Phase separation 15 2.3.3 Spray drying 16 2.3.4 Extrusion 16 ii 2.4 Electrostatic extrusion 20 2.4.1 Droplet formation under electrostatic potential 21 2.4.2 Modes of electrostatic dispersion 26 2.4.3 Governing factors on microcapsule formation 29 2.5 Alginate as a vaccine delivery system 30 2.5.1 Alginate 31 2.5.2 Ca-alginate matrix 33 2.5.3 Chitosan 34 2.5.4 Alginate-chitosan matrix 35 CHAPTER 3 MATERIALS AND METHODOLOGY 3.1 Materials 37 37 3.1.1 List of chemicals 37 3.1.2 Stock preparation 38 3.2 Experimental setup 39 3.2.1 Equipments and accessories 39 3.2.2 System configurations 39 3.3 Experimental procedures 41 3.3.1 Microcapsule formation 41 3.3.2 Antigen release 42 3.4 Determination of microcapsule properties 44 3.4.1 Particle size 44 3.4.2 Microcapsule morphology 45 3.4.3 BSA release 45 3.4.4 HA release 46 iii CHAPTER 4 RESULTS AND DISCUSSION 4.1 Optimum conditions for antigen encapsulation 47 47 4.1.1 Effect of applied voltage 49 4.1.2 Effect of flow rate 54 4.1.3 Effect of electrode spacing 58 4.1.4 Effect of needle size 61 4.2 Microcapsule morphology 63 4.2.1 Effect of pH on microcapsule surface 65 4.2.2 Effect of chitosan coating on microcapsule surface 66 4.3 Antigen release 68 4.3.1 BSA release from alginate matrices 69 4.3.2 HA release from alginate matrices 76 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 79 5.1 Conclusions 79 5.2 Recommendations for future work 80 REFERENCES 82 iv SUMMARY The transmission of avian H5N1 influenza viruses to 18 humans in Hong Kong in 1997 with six deaths established that avian influenza viruses can transmit to and cause lethal infection in humans. Concerns were raised about another global influenza pandemic because humans have absolutely no immunity against any H5 viruses. At present, vaccination remains the principal measure for preventing influenza and reducing the impact of epidemics. In this study, I investigated the use of microencapsulation in alginate matrix for antigen delivery. Bovine serum albumin (BSA) and partial purified hemagglutinin (HA) proteins were encapsulated in alginate matrix by electrostatic extrusion technique to form a delivery system which allowed for controlled release of protein. A positively charged needle with grounded collecting solution system was selected as the electrode configuration for the extrusion process. I found that applied voltage, flow rate, electrode spacing and needle size played their roles in electrostatic microcapsule formation and models were established between microcapsule size and two dominant factors, i.e., applied voltage and flow rate. As the voltage applied to the needle was increased, the formed droplets were reduced in size. However, any increase in applied voltage above its critical point did not result in much reduction in droplet size. Further reduction in droplet size would be achieved by reducing the flow rate of the polymer solution, which would give more time for counter ions to diffuse to droplet surface and reduce the surface tension. Meanwhile, either narrowing the electrode spacing or decreasing the needle size could also help reduce droplet size. v Alginate, a natural polysaccharide, was selected to form the polymeric matrix for antigen encapsulation, due to its excellent biocompatibility, biodegradability and nontoxicity. Alginate can be ionically crosslinked by the addition of divalent cations, such as Ca2+, Sr2+, and Ba2+, in aqueous solution. However, the resulting Ca-alginate microcapsules are very porous. My study showed a quick release of BSA proteins in Tris-HCl buffer solution. Therefore, polymeric coatings are often applied to ensure better isolation and retention of the encapsulated material. Poly-L-lysine and chitosan are the most commonly used polycations for microcapsule production. My study favoured chitosan over poly-L-lysine due to its low cost, easy access and naturalness. In vitro results showed protein release was affected by microcapsules size, pH, protein loading, and release medium. It would be significantly delayed if chitosan coating were applied on the Ca-alginate microcapsules. Further investigations were needed to explore the in vivo function of alginate matrix, strengthen alginate matrix by Ba2+, improve antigen loading efficiency and scale up the microcapsule production. vi Nomenclature Symbol Description d diameter of the droplet, m do diameter of the droplet under gravity only, m db diameter of the gel beads in the receiving solutions, m dc internal diameter of the needle, m de external diameter of the needle, m ds diameter of the pendant droplet neck, m D diffusion coefficient of the surfactant, m2/s Δx characteristic diffusional length, m εo permittivity of the vacuum (air), F/m F Faraday constant, C/mol Fe electric force under applied electrical potential, N g gravity constant, N/kg γ liquid surface tension, N/m γo liquid surface tension (under gravity only or pure water), N/m γe surface tension at its equilibrium value, N/m Γ adsorption amount of a surfactant at the droplet surface, mol/m2 h distance between the droplet and collecting solution, m js flux of the surfactant into the surface, mol/m2⋅s jv flow rate of the solution, m3/s k fitting parameter K electrical conductivity, (S/m) vii μ viscosity of the solution, Pa⋅s qo electrical charge accumulated, C ρd density of the dispersed phase, kg/m3 τad characteristic relaxation time of the adsorption, s τv characteristic relaxation time of the formation of the drop, s U applied electrical potential, V Ucr electrostatic potential at critical point, V viii Abbreviations BCA Bicinchoninic acid BSA Bovine serum albumin CDC Centers for Disease Control and Prevention DNA Deoxyribonucleic acid FITC-BSA Bovine serum albumin-fluorescein isothiocyanate H5N1 Hemagglutinin type 5 & Neuraminidase type 1 HA Hemagglutinin HPAI Highly pathogenic avian influenza M1 Matrix protein 1 M2 Matrix protein 2 NA Neuraminidase NP Nucleoprotein PLG Polylactide-co-glycolides RNA Ribonucleic acid vRNA viral RNA mRNA messenger RNA RNP Ribonucleoprotein SDS-PAGE Sodium dodecyl sulfate - polyacrylamide gel electrophoresis SEM Scanning electron microscope WHO World Health Organization W/O/W Water-in-oil-in-water ix List of Figures Figure 2.1 Microscopy image of influenza virus 4 Figure 2.2 A schematic drawing of influenza virus structure 6 Figure 2.3 Replication cycle of influenza virus 7 Figure 2.4 Double emulsion technique for microencapsulation 13 Figure 2.5 Extrusion devices 17 Figure 2.6 Droplet formation under simple gravity 21 Figure 2.7 Effect of electric field on the droplet formation 22 Figure 2.8 Effect of kinetic factors on droplet formation 24 Figure 2.9 Structure of alginate 31 Figure 2.10 “Egg-box” junction of Ca-alginate matrix 33 Figure 2.11 Structure of chitosan 34 Figure 3.1 Pictures of equipments and accessories 40 Figure 3.2 System configurations of electrostatic extrusion process 40 Figure 4.1 A typical result from the particle size analyzer 47 Figure 4.2 Satellite formation under an electrical potential 48 Figure 4.3 Effect of applied voltage on microcapsule formation 50 Figure 4.4 Transition from dripping mode to cone-jet mode 51 Figure 4.5 Stable and unstable jet emission from the needle 52 Figure 4.6 Linear models for microcapsule size estimation 53 Figure 4.7 Effect of flow rate on microcapsule formation 55 Figure 4.8 Scaling laws for 1% Na-alginate solution 57 x Figure 4.9 Modified scaling laws for 1% Na-alginate solution 58 Figure 4.10 Effect of electrode spacing on microcapsule formation 60 Figure 4.11 Effect of needle size on microcapsule formation 62 Figure 4.12 Ca-alginate microcapsule surface 64 Figure 4.13 pH effects on microcapsule surface structure 65 Figure 4.14 Ca-alginate microcapsule surface (No coating Vs. Coating) 66 Figure 4.15 SEM pictures of microcapsules with different coatings 67 Figure 4.16 SEM picture of freeze-dried Ca-alginate–chitosan microcapsule 68 Figure 4.17 The release of BSA from alginate matrix 70 Figure 4.18 Effect of microcapsule size on BSA release 70 Figure 4.19 Effect of pH on BSA release 71 Figure 4.20 Effect of BSA loading on BSA release 72 Figure 4.21 Effect of coating time of chitosan on BSA release 73 Figure 4.22 Effect of release medium on BSA release 74 Figure 4.23 BSA encapsulation efficiency and release conditions 75 Figure 4.24 HA release from alginate matrix in Tris-HCl buffer 77 Figure 4.25 Effect of chitosan coating on HA release 77 Figure 4.26 SDS-PAGE results of HA release 78 Figure 4.27 Western blotting results of HA release 78 xi List of Tables Table 2.1 Differences between influenza A, B, and C viruses 5 Table 2.2 Methods in droplet extrusion 17 Table 2.3 Characteristic features of the modes of electrostatic dispersion 27 Table 2.4 General properties of chitosan 35 Table 4.1 Control parameters for microencapsulation experiments 48 Table 4.2 Effect of applied voltage on microcapsule formation 49 Table 4.3 Effect of flow rate on microcapsule formation 54 Table 4.4 Physical properties of 1% Na-alginate solution 56 Table 4.5 Effect of electrode spacing on microcapsule formation 59 Table 4.6 Effect of needle size on microcapsule formation 61 Table 4.7 Factors in the control of microcapsule permeability 69 xii Chapter 1 CHAPTER 1 1.1 INTRODUCTION Motivation and objectives In 1997, an influenza virus (H5N1) known to infect only birds previously was found to infect human causing disease and death in Hong Kong and the outbreak involved 18 patients with six deaths (Tam, 2002). The government of Hong Kong decided on 28 December 1997, to cull all chickens in Hong Kong and 1.5 million of birds were killed from 29 to 31 December 1997. Since then, an epidemic of influenza in poultry, commonly referred to as “bird flu”, began sweeping through several countries in the Pacific Rim (Vietnam, Thailand, Japan, China, South Korea, Cambodia) in early 21st century. It caused serious financial damage, public health threat and mental distress to people in the region. Scientists also warned its potential to develop into another deadly global pandemic. World Health Organization (WHO) has urged both developed and developing countries to act swiftly and decisively to control and contain this outbreak. The most powerful thrust to the influenza control program focuses on vaccination (Straus, 1993). To curb spread of “bird flu”, Vietnam government plans to test vaccines in chickens and ducks from June to August 2005, leading to compulsory vaccination in high-risk areas in October. Ho Chi Minh City will conduct trial vaccination with 600,000 doses of Trovac avian influenza vaccine, produced by Merial France, while 2.7 million doses of a related Dutch vaccine will be tested elsewhere in the country, according to the official Vietnam News Agency. Most influenza vaccines are prepared from influenza virus strains grown in embryonated hens’ eggs. However, reactions are reported in some vaccinees given 1 Chapter 1 whole, inactive virus vaccine; thus many contemporary vaccines are vaccines containing only the surface hemagglutinin (HA) and neuraminidase (NA) glycoproteins, termed subunit vaccines (Potter, 1994). However, subunit antigens alone are usually only weak immunogens and require an adjuvant or carrier system to induce protective immunity. The quality of the resulting antibody-and/or cell-mediated immune response and its magnitude depends on appropriate antigen processing and the cytokine profile generated by the combined action of the antigen and the adjuvant (Felnerova et al., 2004). Many vaccine delivery systems have been proposed to provide sustained release and/or to increase the immune response. One approach to the development of delivery system for vaccines involves the use of polymer encapsulation of antigens (Singh and O’Hagan, 1998). The polymeric microcapsules that protect antigen from acidic and enzymatic degradation in the gastrointestinal tract also provide a stable vaccine vehicle with extended shelf life. Once encapsulated, antigen is released from the microspheres by diffusion through matrix pores and by matrix degradation (Morris et al., 1994). A number of different polymers have been evaluated for the development of oral vaccines, including naturally occurring polymers (e.g. starch, alginates and gelatin) and synthetic polymers (e.g. polylactide-co-glycolides (PLG), polyanhydrides and phthalates). In material selection, of primary concern are considerations of toxicity, irritancy and allergenicity, and the need for a biodegradable or a soluble coating. Usually, natural polymers are preferred. The advantages of using natural polymers include their low cost, biocompatibility and aqueous solubility (Singh and O’Hagan, 2 Chapter 1 1998). Among all the natural polymers, alginate is the most commonly employed polymer for cell encapsulation (Gerbsch and Buchholz, 1995). Many studies have also been carried out using alginate matrix as a vaccine delivery system (Kwok et al., 1989; Bowersock et al., 1994 and 1996; Wee et al., 1995). The main advantages of using alginate microcapsules for antigen delivery are the low cost of polymer, good biocompatibility and ease of preparation (Singh and O’Hagan, 1998). In my study, Bovine serum albumin (BSA) and partial purified hemagglutinin (HA) proteins were encapsulated in alginate matrix by electrostatic extrusion technique to form a delivery system which allowed for controlled release. The objectives of this study were: — Encapsulate antigens (BSA or HA) in alginate matrix by electrostatic extrusion — Study the governing factors of electrostatic extrusion process — Study the in vitro release of BSA or HA from Ca-alginate matrix — Study the effect of chitosan coating on the microcapsule structure and antigen release 1.2 Organization of the thesis This thesis is divided into five chapters. Chapter 1 is an introduction to the motivation and objectives of the project, followed by a description of the organization of the thesis. Chapter 2 is a literature review on the influenza A H5N1 virus, vaccination, microencapsulation by electrostatic extrusion and formation of alginate matrix. In Chapters 3, materials and methods used in this study are described. The experimental results as well as the discussion are presented in Chapter 4. Finally, a summary of the findings in this study and recommendations for future research are given in Chapter 5. 3 Chapter 2 CHAPTER 2 BACKGROUND AND LITERATURE REVIEW 2.1 Influenza and H5N1 virus 2.1.1 Influenza and its virus According to World Health Organization (WHO, 2003), influenza is caused by a virus that attacks mainly the upper respiratory tract – the nose, throat and bronchi and rarely also the lungs. The infection usually lasts for about a week. It is characterized by sudden onset of high fever, myalgia, headache and severe malaise, non-productive cough, sore throat, and rhinitis. Most people recover within one to two weeks without requiring any medical treatment. In the very young, the elderly and people suffering from medical conditions such as lung diseases, diabetes, cancer, kidney or heart problems, influenza poses a serious risk. In these people, the infection may lead to severe complications of underlying diseases, pneumonia and death. Influenza viruses (Fig. 2.1) belong to the family Orthomyxoviridae and are separated into types A, B and C according to antigenic differences in their respective nucleocapsid (WHO, 2002). Figure 2.1 Microscopy image of influenza virus (Stannard, 1995) 4 Chapter 2 There are significant differences in genetic organization, structure, host range, epidemiology, and clinical characteristics between the three influenza virus types, which are summarized in Table 2.1. Table 2.1 Differences between influenza A, B, and C viruses (Treanor, 1999) Influenza A Influenza B Influenza C Genetics 8 gene segments 8 gene segments 7 gene segments Structure 10 viral proteins 11 viral proteins 9 viral proteins M2 protein unique NB protein unique HEF protein unique Host range Humans, swine, equine, avian, marine mammals Humans only Humans and swine Epidemiology Antigenic shift and drift; Drift is generally linear Antigenic drift Antigenic drift only; More than only; Multiple one variant may co- variants circulate Clinical features May cause large pandemics with significant mortality in young disease Severe disease generally confined to elderly or those at high risk; pandemics not seen Mild disease without seasonality As shown in Fig. 2.2, influenza viruses are negative-strand RNA viruses with a segmented genome. The eight RNA segments of influenza viruses (seven for influenza C) are independently encapsulated by the viral nucleoprotein (NP) and each segment is associated with a polymerase complex. The subviral particle consisting of viral RNA (vRNA), NP and polymerase complex is called ribonucleoprotein (RNP) particle. The RNP particles are located inside a shell of M1 protein which lines with the viral lipid membrane. The surface of the particle contains three kinds of spike proteins: hemagglutinin (HA), neuraminidase (NA), and matrix protein (M2, only on type A). 5 Chapter 2 Hemagglutinin (HA) Neuraminidase (NA) Lipid membrane Matrix Protein (M1) M2 Protein (only on type A) Ribonucleoprotein (RNP): - vRNA - Polymerase - Nucleoprotein Figure 2.2 A schematic drawing of influenza virus structure Further classification of type A virus into subtype is dependent on differences in the HA and NA antigens, and 16 HA and 9 NA subtypes are known so far. Influenza A viruses are distinguished according to a formula, such as A/Hong Kong/1/68 (H3N2): this designation means that the virus is type A, was isolated in Hong Kong, is the first laboratory isolate made in 1968, and has the HA and NA molecule forms HA3 and NA2, respectively (Potter, 1994). The influenza virus replicates by entering a host cell and using this cell's resources to produce hundreds of copies of the viral RNA (Fig. 2.3). The virus attaches to the outside of the host cell and its RNA enters into the cell. Fusion and uncoating events, which are pH dependent, result in the release of the viral genomes into the cytoplasm. They are then imported into the nucleus for replication. Positive-sense viral messenger RNAs (mRNAs) are exported out of the nucleus into the cytoplasm for protein 6 Chapter 2 synthesis. Some of the proteins are imported into the nucleus to assist in viral RNA replication and packaging. In this way, the virus takes over the cell's productivity. Now, instead of producing only new cellular material, the cell produces hundreds of new virus particles. Those particles are eventually released from the cell and drift off. Figure 2.3 2.1.2 Replication cycle of influenza virus (Levine, 1992) Influenza pandemic and its impact Influenza viruses can change in two different ways. — Antigenic drift, the HA and NA proteins of the influenza virus can undergo minor changes or mutations, leading to new strains. It happens continually over time, and 7 Chapter 2 produces new virus strains that may not be recognized by antibodies to earlier influenza strains. — Antigenic shift, it is an abrupt, major change in the influenza A viruses, resulting in a new influenza virus that can infect humans and has a new HA or HA and NA protein combination that has not been seen in humans for many years. Antigenic shift results in a new influenza A subtype. If a new subtype of influenza A virus is introduced into the human population, if most people have little or no protection against the new virus, and if the virus can spread easily from person to person, a pandemic may occur. According to WHO (2003), three times in the last century, the influenza A viruses have undergone major genetic changes mainly in their H-component, resulting in global pandemics and large tolls in terms of both disease and deaths. The most infamous pandemic was “Spanish Flu” which affected large parts of the world population and is thought to have killed at least 40 million people in 1918-1919. More recently, two other influenza A pandemics occurred in 1957 (“Asian influenza”) and 1968 (“Hong Kong influenza”) and caused significant morbidity and mortality globally. Most recently, limited outbreaks of a new influenza subtype A (H5N1) directly transmitted from birds to humans have occurred in Hong Kong Special Administrative Region of China in 1997 and 2003. According to Centers for Disease Control and Prevention (CDC), many scientists believe it is only a matter of time until the next influenza pandemic occurs (U.S. CDC, 2005a). The severity of the next pandemic cannot be predicted, but modeling studies 8 Chapter 2 suggest that its effect in the United States could be severe. In the absence of any control measures (vaccination or drugs), it has been estimated that in the United States a “medium–level” pandemic could cause 89,000 to 207,000 deaths, between 314,000 and 734,000 hospitalizations, 18 to 42 million outpatient visits, and another 20 to 47 million people being sick. Between 15% and 35% of the U.S. population could be affected by an influenza pandemic, and the economic impact could range between $71.3 and $166.5 billion. 2.1.3 Avian influenza (Bird flu) and H5N1 virus Avian influenza (Bird flu) is an infection caused by avian influenza viruses. These flu viruses occur naturally among birds (U.S. CDC, 2005b). All birds are thought to be susceptible to infection with avian influenza, though some species are more resistant to infection than others. Infection causes a wide spectrum of symptoms in birds, ranging from mild illness to a highly contagious and rapidly fatal disease resulting in severe epidemics. The latter is known as “highly pathogenic avian influenza” (HPAI). This form is characterized by sudden onset, severe illness, and rapid death, with a mortality that can approach 100%. According to WHO (2004), fifteen subtypes of influenza virus are known to infect birds, thus providing an extensive reservoir of influenza viruses potentially circulating in bird populations. Of the 15 avian influenza virus subtypes, H5N1 is of particular concern for several reasons. H5N1 mutates rapidly and has a documented propensity to acquire genes from viruses infecting other animal species. Its ability to cause severe disease in humans has now been documented on several occasions. In addition, laboratory studies have demonstrated that isolates from this virus have a high 9 Chapter 2 pathogenicity and can cause severe disease in humans. Birds that survive infection excrete virus for at least 10 days, orally and in faeces, thus facilitating further spread at live poultry markets and by migratory birds. H5N1 virus demonstrated a capacity to directly infect humans in 1997, and succeeded again in Viet Nam in January 2004 (WHO, 2004). The spread of infection in birds increases the opportunities for direct infection of humans. If more humans become infected over time, the likelihood also increases that humans, if concurrently infected with human and avian influenza strains, could serve as the “mixing vessel” for the emergence of a novel subtype with sufficient human genes to be easily transmitted from person to person. Such an event would mark the start of an influenza pandemic. 2.1.4 Prevention and treatment Several measures can help minimize the global public health risks that could arise from large outbreaks of highly pathogenic H5N1 avian influenza in birds. An immediate priority is to halt further spread of epidemics in poultry populations. This strategy works to reduce opportunities for human exposure to the virus. Vaccination of persons at high risk of exposure to infected poultry, using existing vaccines effective against currently circulating human influenza strains, can reduce the likelihood of coinfection of humans with avian and influenza strains, and thus reduce the risk that genes will be exchanged. Workers involved in the culling of poultry flocks must be protected, by proper clothing and equipment, against infection. These workers should also receive antiviral drugs as a prophylactic measure (WHO, 2004). 10 Chapter 2 Vaccination is the principal measure for preventing influenza and reducing the impact of epidemics. Various types of influenza vaccines have been available and used for more than 60 years. They are safe and effective in preventing both mild and severe outcomes of influenza (WHO, 2002). The major treatment for influenza infections are the time-proven ones involving hydration, rest, and antipyretics, especially acetaminophen rather than aspirin (Straus, 1993). Antiviral drugs for influenza are an important adjunct to influenza vaccine for the treatment and prevention of influenza. When taken before infection or during early stage of the disease (within two days of illness onset), antivirals may help prevent infection, and if infection has already taken hold, their early administration may reduce the duration of symptoms by one to two days (WHO, 2003). 2.2 Vaccination through a delivery system Vaccines against influenza have been around for 60 years. The antigens of the influenza virus particle which stimulate immunity to subsequent infection have been identified. The results of challenge studies indicated that immunity is induced by the host responses to the virus haemagglutinin (HA) and, to a lesser extent, to the neuraminidase (NA). Therefore, antibody against HA is the most important component in the protection against influenza viruses (Potter, 2000). There are four types of influenza vaccines currently available: — Whole virus vaccines, these are the first influenza vaccines to be produced. The currently circulating strain of influenza is inoculated into embryonated eggs, 11 Chapter 2 harvested 2-3 days later and inactivated. However, systemic reactions, such as fever, headache and muscle pain have shown in some vaccinees. — Split virus vaccines, these are inactivated vaccine particles disrupted by detergents. Fewer side effects have shown while retaining effectiveness. — Subunit virus vaccines, these are vaccines containing only purified surface antigens (HA and NA). They are often used in aqueous suspension or may be absorbed to carriers such as alhydrogel. Vaccinees given subunit vaccines experience fewer reactions than those who taken the pervious two vaccines. — Live attenuated vaccines, these are produced by reassortment of wild-type strains of interest with carefully characterized attenuated strains. They will mimic pathogen infection to a large extent, but without causing disease. However, they can be dangerous to immuno-compromised individuals and may cause disease in otherwise healthy individuals. Other basic problems also remain particularly in the area of purification. The vaccine must also be shown to be attenuated and safe. Subunit virus vaccines were chosen in my study based on the above characteristics. Besides these existing vaccines, some researchers have also proposed a DNA-based vaccination technique (Tang et al., 1992), which is very much an approach for the future. The first clinical trials of such a vaccine for influenza have begun and there is a need for regulatory authorities worldwide to consider the requirements necessary for assuring the quality of these vaccines (Levandowski et al., 1996). Others prefer vaccination through a delivery system. For example, many researchers explored the adjuvanting properties of adsorbed lactide microparticulate vaccines (Wood et al., 1996; Coombes et al., 1998); some evaluated polymeric microspheres as an oral antigen delivery system for immunization with influenza virus (Moldoveanu et al, 12 Chapter 2 1993); others even suggest microencapsulated influenza antigen may have potential for a single dose vaccine delivery system with adjuvant properties (Hilbert et al., 1999). 2.3 Common microencapsulation techniques Microparticles prepared from certain polymers have the capacity to encapsulate and to surface adsorb influenza virus (Wood et al., 1996). Generally, microencapsulation involves two steps: drop formation (discrete droplet formation or emulsification), followed by droplet solidification (gelation, membrane formation, or other means) (Poncelet et al., 1999a). The most common techniques in microencapsulation are listed as follows. 2.3.1 Solvent evaporation/extraction Figure 2.4 Double emulsion technique for microencapsulation (Yang et al., 2000) 13 Chapter 2 Solvent evaporation is the most widely used method for the preparation of microparticles with entrapped antigens (Singh and O’Hagan, 1998). The method is usually based on the formation of a multiple emulsion (water-in-oil-in-water or w/o/w) from which the oil phase (organic solvent) is evaporated and/or extracted to yield free flowing microcapsules, as shown in Fig. 2.4. This technique is based on the evaporation of the internal phase of an emulsion by agitation (Benoit et al., 1996). Initially, the polymeric supporting material is dissolved in a volatile organic solvent. The active drug to be encapsulated is then dispersed or dissolved in the organic solution to form a suspension, an emulsion or a solution. Next, 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, and contains an appropriate tensioactive additive. Once the emulsion is stabilized, agitation is maintained and the solvent evaporates after diffusing through the continuous phase. Finally, the microcapsules are filtered, dried and characterized after the removal of the solvents. The organic solvent of the dispersed phase of the emulsion can also be eliminated in two steps: 1) Diffusion of the solvent in the dispersing phase or solvent extraction; 2) Elimination of the solvent at the dispersing phase-air interface or solvent evaporation. In theory, if one uses a continuous phase which will immediately extract the solvent of the dispersed phase, the evaporation stage is no longer necessary (Benoit et al., 1996). In practice, this can be achieved by using large volumes of dispersing phase with respect to the dispersed phase (Boisdron-Celle et al., 1995). 14 Chapter 2 The specific organic solvent, temperature of solvent evaporation, rate of agitation, volume of organic phase per unit volume of aqueous phase, nature and amount of emulsifier dissolved in the aqueous phase, polymer structure and molecular weight, and antigen solubility all influence the final microcapsule obtained. Very small microcapsules can be produced with this technique, but the size dispersion is generally larger than with the extrusion techniques (Poncelet and Neufeld, 1996). Another potential problem is the removal of solvent contaminants from the final microcapsules becomes more difficult as microcapsule size decreases. 2.3.2 Phase separation Polymer phase separation, in nonaqueous media, by nonsolvent or polymer addition, also referred to as coacervation, is an excellent technique for the entrapment of watersoluble drugs such as peptides, proteins, or vaccines (Benoit et al., 1996). This technique utilizes solubilized polyesters and antigen which are then mixed with salts, non-solvents or, more commonly, with an incompatible second polymer resulting in a separation of the polyester phase that engulfs the antigen particles (Deasy, 1988). The liquid-polyester/antigen phase is desolvated and solidified by the addition of excess non-solvent at low temperatures. Phase separation technique may protect antigens from being altered by exposure to heat or from their partitioning out into dispersing phases. However, this approach is limited to antigens which remain insoluble in the organic solvent used to dissolve the polyester. And residual solvent concentrations in resulting microcapsules can be very high, especially when heptane is chosen as the hardening agent (Müller et al., 1993). 15 Chapter 2 2.3.3 Spray drying Spray drying is widely used in the chemical, pharmaceutical, and food industries (Nielsen, 1982). The principle of spray drying is the atomization of a solution by compressed air or nitrogen through a desiccating chamber and drying across a current of warm air. Knowledge of copolymer/antigen suspension rheology and nozzle design is used to control the microcapsule size and shape. Particle size distributions are usually monodispersed with a Gaussian distribution more or less depending on the pulverizing head employed (e.g., pneumatic atomizer, disc) and the other parameters such as, pressure of compressed air, internal diameter of the atomizer, viscosity, and flow rate of the solution (Benoit et al., 1996). One particular advantage for this technique is, residual levels of organic solvents are less than those from emulsification and evaporation techniques. However, there is only 20-30% core loading capacity for most spray-dried capsules. 2.3.4 Extrusion The above three methods involve organic solvents and/or extreme temperature or pH conditions which could inactivate or damage the materials being encapsulated. An alternative approach is the ionic cross-linking of polyelectrolytes with counterions whereby one of the polyions containing a biomaterial is extruded as droplets into a solution where the droplets gel and these droplets are subsequently coated by another polyion (Poncelet et al., 1992; Bugarski et al., 1993; Daly and Knorr, 1988). Various extrusion methods proposed in the literature are summarized in Table 2.2, and the devices used in the processes are shown in Fig. 2.5. 16 Chapter 2 Table 2.2 Methods in droplet extrusion (Dulieu et al., 1999) Extrusion methods Droplet formation Jet breakage • Under gravity • By vibrating device • Under coaxial flow • By cutting device • Under electrostatic potential Figure 2.5 Extrusion devices (Dulieu et al., 1999) 2.3.4.1 Extrusion under simple gravity The simplest method to form individual droplets is to let a liquid droplet fall from the tip of a needle under simple gravity, as shown in Fig. 2.5 A. In this scenario, a pendant droplet is formed by the flow of a liquid through a needle, and continuously grows 17 Chapter 2 until its mass reaches a critical value, when the droplet detaches from the tip of the tube and falls into a collecting solution. This method is often limited to a laboratory scale, due to a restriction on flow rate to prevent jet formation. The production capacity is limited by the speed of droplet formation at the tip of the needle, usually at an order of magnitude of ml/h. The other limitation is the large diameter of the beads, typically 2 to 3 mm, even for very small needle diameters. It compromises its application potential, as microcapsules with diameters larger than 1mm often induce significant mass transfer limitations and heterogeneous cell distribution within the beads (Groboillot et al., 1994). 2.3.4.2 Extrusion under coaxial air or liquid flow The application of a coaxial air jet around the needle has been proposed by Lane (1947) to increase the force acting on nascent drops (Fig. 2.5 B). Charwat (1977) replaced the air by a liquid jet, permitting a better control of the viscosity, surface tension, and density of the entraining phase, through selection of an appropriate liquid. Both methods produce beads or microcapsules ranging from a few micrometers to one millimetre. However, the flow rate remains very limited, to less than 30 ml/h, to avoid formation of a liquid jet. And the size dispersion increases drastically when the droplet diameter is decreased (Poncelet et al., 1993). For these reasons, the coaxial fluid jet systems have not been considered for scale-up. Even on the laboratory scale, this method is being replaced by the technologies described below. 18 Chapter 2 2.3.4.3 Extrusion under electrostatic potential Droplet formation is greatly improved by replacing the drag force with a high electrostatic potential between the needle and the collecting solution (Burgarski et al., 1993, 1994a; Poncelet et al., 1994). Electric potential may be applied between the needle and a stainless steel ring placed below the needle, as shown in Fig. 2.5 C. Under suitable conditions, the electrostatic pressure at the surface forces the liquid drop into a cone shape. Surplus charge is ejected by the emission of charged droplets from the tip of the liquid. The emission process depends on such factors as the needle diameter, distance from the collecting solution, and applied voltage (Nawab and Mason, 1958). Two primary advantages of electrostatic droplet generation, over, for example, an air jet extruder, are the production of much smaller beads with conventional needles and ease of bead size control by simply varying the applied potential (Goosen et al., 1997a). However, satellite peaks may be observed in the size distribution profile. Satellites are formed by breakage of the fine filament between the droplet and the needle tip just before separation, resulting in secondary peaks. Though the flow rate is still limited by the formation of the jet, electrostatic extrusion is a promising technique to obtain small microdroplets (down to 200 μm), at least at laboratory scale (Dulieu et al., 1999). Details of this technique will be discussed later. 2.3.4.4 Jet breakage by vibration If liquid exceeds a certain velocity, it exits from the needle as a jet. Rayleigh (1878) showed that if an external wave of the natural frequency is applied to the jet, the jet breaks into monodispersed droplets (Fig. 2.5 D). This method is one of the most efficient techniques to produce large capsules (1 to 3 mm) with a narrow size distribution. A multi-nozzle system would enable production in the order of hundreds 19 Chapter 2 of liters per hour. However, it appears more difficult to use this process for microcapsules less than 800 μm in diameter (Dulieu et al., 1999). 2.3.4.5 Jet breakage by rotating systems To overcome the limitations of vibrating system, Prusse et al. (1997) proposed a rotating device to cut the jet into small droplets, as shown in Fig. 2.5 E. The diameter of the cutting wires is the main parameter determining the effectiveness of this method. Rotating jet breakage appeared to be an easy, efficient, and scalable device for producing large quantities of relatively small microcapsules with narrow size distribution (Dulieu et al., 1999). 2.4 Electrostatic extrusion Electrostatic extrusion is a novel extrusion technique that uses electrostatic forces to disrupt a liquid surface at the capillary/needle tip forming a charged stream of small droplets (Nedovic et al., 2001). A polymer solution is extruded through a charged needle. The collecting solution directly beneath the needle is either grounded or has a charge opposite to that of the needle. As the polymer solution passes through the needle it accumulates charge and droplets formed at the end of the needle are pulled off by the electrostatic attractive force between the needle and the collecting solution. The result is a charged stream of fine droplets. 20 Chapter 2 2.4.1 Droplet formation under electrostatic potential 2.4.1.1 Size of droplets under simple gravity As shown in Fig. 2.6, under simple gravity, the droplet diameter, do, may be computed by equalizing the gravity force with the product of the interfacial tension, γο, and the perimeter of the tip, πds (Tate’s law): Figure 2.6 mg = πd s γ o Droplet formation under simple gravity (Poncelet et al., 1999b) or π 6 d o3 ρ d g = πd s γ o (2.1) Where, g is the gravity constant, ds is the diameter of the pendant droplet neck, which is approximately equal to the external diameter of the needle de, γο is the liquid surface tension under gravity only, assumed equal to pure water, ρd is the density of the dispersed phase, do is the diameter of the droplet under gravity only. 21 Chapter 2 And the size the droplet will be: d o = ( 6 d s γ o / ρ d g )1 / 3 ≈ ( 6 d eγ o / ρ d g )1 / 3 (2.2) However, the diameters db of the microcapsules in the collecting solutions may differ from the diameters of the falling droplets because of their contraction or swelling in the course of gelling or complex formation processes (Poncelet et al., 1999b) 2.4.1.2 Size of droplets in an electric field According to Poncelet et al. (1999b), under an electric filed (Fig. 2.7), the condition for the mechanical equilibrium of a pendant droplet may be written as: mg + Fe = πd s γ (U ) (2.3) Where, Fe is the electric force under applied electrical potential U, γ(U) is the surface tension of droplet under applied electrical potential U. Figure 2.7 Effect of electric field on the droplet formation (Poncelet et al., 1999b) For a single needle situated in the proximity of an infinite plane electrode configuration, the electrostatic force Fe can be obtained as: 22 Chapter 2 Fe ≅ (1 / 4πε o )(qo / 2h) 2 ≅ πε o (d / d c ) 2 (d / 2h) 2 U 2 (2.4) Where, εo is the permittivity of the vacuum (air), qo is the electrical charge accumulated, and qo = 2πεodcU, h is the distance between the droplet and the plane electrode, dc is the internal diameter of the needle, U is the applied electrical potential. And according to Lippman’s theory of electrocapillarity, the equilibrium surface tension γ of the charged surface of a liquid droplet decrease with increasing electrical potential U as: γ (U ) = γ o (1 − kε oU 2 / d c γ o ) = γ o [1 − (U 2 / U cr2 )] Where, U cr = d c γ o / kε o (2.5) (2.6) k ≤ 1, is the fitting parameter which accounts for the geometrical and kinetic approximations. By combining equations (2.3), (2.4), (2.5) and (2.6), I can have, d = d o {1 − [1 + α d / h ](U / U cr ) 2 }1 / 3 Where, α d / h = ( d / d c ) 2 ( d / 2h) 2 U 2 (2.7) (2.8) As αd/h 0 the surface tension will decrease tending to its equilibrium value γe at τv >> τad. In the case of dΓ/dt < 0, the surface tension will increase and tend to its maximal value γo (surface tension of pure water) at τv 1, then d ∝ jv1/2 (2.20) Where, ρd is the density of the dispersed phase, εo is the permittivity of the vacuum (air), γ is the liquid surface tension, K is the electrical conductivity, μ is the viscosity of the solution, d is the diameter of the droplet, jv is the flow rate of the solution. 2.4.2 Modes of electrostatic dispersion The droplet can exit the needle in various modes (Table 2.3). According to Sample and Bollini (1972), when the spraying configuration is highly symmetric, and under certain conditions of liquid pressure and applied voltage, the spraying process can be somewhat regular and periodic. There are mainly two spraying modes in my study: dripping mode and cone-jet mode. — Dripping mode, under this mode, the droplet will grow until its weight overcomes the net vertical component of the surface tension force. At this point the liquid nearest the needle forms a neck, which eventually ruptures, thereby allowing the 26 Chapter 2 Table 2.3 Characteristic features of the modes of electrostatic dispersion (Jaworek and Krupa, 1999) 27 Chapter 2 Table 2.3 (cont’d) Characteristic features of the modes of electrostatic dispersion (Jaworek and Krupa, 1999) 28 Chapter 2 main part of the drop to fall from the needle. When a dc voltage is applied to the needle, electric field between the needle and the plate causes a rise in the emission frequency and a reduction in drop size. — Cone-jet mode, for some liquids, the application of a sufficiently high electric field causes the convergence of the liquid jet into a conical shape, known as the Taylor cone, based on the pioneering work of Taylor (1964). In this case, the droplets detach from the tip of the cone, rather than from the needle, leading to the formation of a steady stream of fine droplets. 2.4.3 Governing factors on microcapsule formation Many researchers believed parameters such as applied voltage, needle size, electrode spacing, polymer concentration, flow rate and geometry played theirs roles in electrostatic microcapsule formation (Bugarski et al. 1994b; Klokk and Melvik, 2002). It has been proposed that the reduction in bead size with increasing potential is primarily a result of a fall in the surface tension of the droplets generated at the needle tip (Dulieu et al., 1999). However, any increase in applied potential above its critical point did not result in reduction in droplet size. Further reduction in microbead size could be achieved by decreasing the needle size or reduce the electrode spacing (Bugarski et al., 1994b; Goosen et al., 1997b; Klokk and Melvik, 2002; Nawab and Mason, 1958). Meanwhile, as the voltage was increased above Ucr, a high surface charge and an electric field on the surface of needle tip gave rise to a mechanical force causing needle vibration and resulting in an oscillating thread-like filament. The greatest decrease in microbead size was observed when natural needle oscillation occurred (Bugarski et al., 1994b). 29 Chapter 2 Klokk and Mevik (2002) also found that, an increase in concentration of the polymer solution caused an increase in bead diameter; furthermore, the bead size was more strongly correlated to solution viscosity than to the actual polymer concentration. The standard deviation also decreased at the lower polymer concentration due to a more uniform bead size distribution (Bugarski et al., 1994b; Goosen, 1996). It seemed impossible to compensate for the higher polymer solution concentration by increasing the electrostatic voltage further; however, reducing the flow rate of the polymer solution could amend the effect of higher solution viscosity (Klokk and Melvik, 2002). Because low flow rate will give more time for counter ions to diffuse to droplet surface and reduce the surface tension (Hallé et al., 1994). Bugarski et al. (1994a) also investigated with three types of electrode configurations: a parallel plate; a positively charged needle; and a grounded needle with alginate as the polymer. They concluded that the smallest droplet size was produced with the positively charged small diameter needle. 2.5 Alginate as a vaccine delivery system In the past, a whole range of different polymeric materials has been employed to encapsulate cells or tissues for implantation (Zimmermann et al., 1999). Among those, polymers such as alginate, CM-cellulose, κ-carrageenan, and chitosan, which come from natural sources, are of particular interest because they are of low cost and naturally redeemable. The overwhelming majority of the literature has employed sodium alginate for microcapsule formation, due to its excellent biocompatibility and biodegradability (Lim and Sun, 1980; Orive et al., 2004; Poncelet et al., 1992). 30 Chapter 2 On the other hand, the use of synthetic polymers is limited because of the unphysiological conditions that are typically necessary for the encapsulation process and/or the potential risks associated with reactive monomers and/or impurities (Brodelius and Vandamme, 1987; Lodge and Monaco, 1994; Silver and Doillon, 1989). Until now, no synthetic polymer has existed that can encapsulate sensitive entities under mild conditions (Cohen et al., 1990). In view of regulatory concerns and safety issues for clinical applications, this higher level of safety is a major advantage of natural polymers such as alginates, as compared to synthetic polymers (Zimmermann et al., 1999). 2.5.1 Alginate Alginate, a naturally occurring biopolymer extracted from brown algae (kelp), has several unique properties that have enabled it to be used as a matrix for the entrapment and/or delivery of a variety of biological agents (Gombotz and Wee, 1998). Alginate polymers are a family of linear unbranched polysaccharides which contain varying amounts of 1,4’-linked β-D-mannuronic acid and α-L-guluronic acid residues (Fig. 2.9). After processing, alginates are available as water-soluble sodium alginates. Figure 2.9 Structure of alginate (MacGregor and Greenwood, 1980) 31 Chapter 2 Alginate can be ionically crosslinked by the addition of divalent cations in aqueous solution. The relatively mild gelation process has enabled not only proteins, but also cells and DNA to be incorporated into alginate matrices with retention of full biological activity. Furthermore, by selection of the type of alginate and coating agent, the pore size, degradation rate, and ultimately release kinetics can be controlled. The unique properties of alginate gels include: — A relatively inert aqueous environment within the matrix — A mild room temperature encapsulation process free of organic solvents — A high gel porosity which allows for high diffusion rates of macromolecules — The ability to control this porosity with simple coating procedures — Dissolution and biodegradation of the system under normal physiological conditions In the past, many researchers have worked on alginate as an ideal delivery system for protein antigens. Kwok et al. (1989) reported the potential feasibility of delivering live Bacillus Camette Guerin vaccine in alginate microbeads to the lung by either inhalation or intravenous injection. Bowersock et al. (1994; 1996) have evaluated the use of alginate hydrogels to deliver oral vaccines to different species of animals. Wee et al. (1995) showed that strong antibody responses were effectively produced when soluble antigens were encapsulated and released from alginate microbeads. And they believe that alginate could be successfully used as a mucosal drug delivery vehicle for the delivery of vaccines or drugs to the upper respiratory tract. 32 Chapter 2 2.4.3 Ca-alginate matrix Alginate beads can be prepared by extruding a solution of sodium alginate containing the desired protein, as droplets, into a divalent crosslinking solution such as Ca2+, Sr2+, or Ba2+. Monovalent cations and Mg2+ ions do not include gelation (Sutherland, 1991) while Ba2+ and Sr2+ ions produce stronger alginate gels than Ca2+ (Clark and RossMurphy, 1987). Other divalent cations such as Pb2+, Cu2+, Cd2+, Co2+, Ni2+, Zn2+ and Mn2+ will also crosslink alginate gels but their use is limited due to their toxicity (Gombotz and Wee, 1998). The gelation and crosslinking of the polymers are mainly achieved by the exchange of sodium ions from the guluronic acids with the divalent cations, and the stacking of these guluronic groups to form the characteristic egg-box structure (Fig. 2.10). Each alginate chain can dimerize to form junctions with many other chains and as a result gel networks are formed rather than insoluble precipitates (Rees and Welsh, 1977). Figure 2.10 “Egg-box” junction of Ca-alginate matrix (Rees, 1981) The monomeric composition, block structure, and molecular size of the alginate strongly affect the mechanical and swelling properties of the alginate beads (Martinsen et al., 1989). Reduction of mechanical properties are observed for alginate with low viscosity or molecular weight (Smidsrød, 1990; Johansen and Flink, 1986). The stability of alginate beads is quite low in the presence of chelating agents such as 33 Chapter 2 phosphate, lactate, and citrate due to shared affinity for calcium, destabilizing the gel. Some proposed using ions with better affinity for alginate such as Al3+, Sr2+, and Ba2+ to strengthen the gel, but these alternatives may be unacceptable for food applications because of potential toxicity. Other ions such as Pb2+ or Cu2+ are more efficient but are also more toxic (Dulieu et al., 1999; Gåserød et al., 1999; Groboillot et al., 1994; Poncelet et al., 2001). However, the resulting hydrogel beads are very porous. Polymer coats are often applied to ensure better isolation and retention of the encapsulated material. The external surface of the beads may also be modified to improve biocompatibility or bead strength (Dulieu et al., 1999). Poly-L-lysine and chitosan are the most commonly used polycations for capsule production (Gåserød, et al., 1998). My study favours chitosan over poly-L-lysine due to its low cost, easy access and naturalness. 2.5.3 Chitosan Figure 2.11 Structure of chitosan (Chourasia and Jain, 2003) Chitosan [poly (β-(1→4)-2-amino-2-deoxy-D-glucos)] is a linear polysaccharide derived from deacetylated chitin (Fig. 2.11), which is extremely abundant in nature, existing widely in mushrooms, yeasts, and hard outer shells of insects and crustaceans (Goosen, 1994). Chitosan is soluble in acidic solutions and the positive charge of 34 Chapter 2 chitosan interacts strongly with negative surfaces (Kas, 1997). Its properties are listed in Table 2.4. Table 2.4 General properties of chitosan (Kim et al., 1999) Chemical Properties Solution Properties Biological Properties • Linear polyamine (poly-D-glucosamine) • Soluble at pH6.5 • Biocompatible • • • Biodegradable Reactive amino groups Viscous solutions • Nontoxic Available reactive hydroxyl groups • • Shear thinning Therapeutic • • • Chelates many transitional metal ions Gel-forming with polycations • Soluble in alcoholwater mixtures • Strong affinity with polyanions Chitosan has been evaluated for a variety of therapeutic applications, including wound-healing, hemostatic, antimicrobial, osteoconductive, cholesterol-lowering, and drug delivery agents. Chitosan has demonstrated physiological compatibility with living tissues. It would be difficult to engineer another polysaccharide that could surpass chitosan in its variety of functions and applications for therapeutics and encapsulation (Kim et al., 1999). 2.5.4 Alginate-chitosan matrix Formation of a polyelectrolyte complex has been demonstrated to occur when a cationic and an anionic polymer are present simultaneously in aqueous solution. It was recognized that the formation of complexes between alginate and chitosan could be ascribed to this phenomenon (Murata et al., 1993). Polyanionic beads may be coated by polycationic membranes by simply suspending beads in the polycationic solution. 35 Chapter 2 This first coating may be followed by a second coating with a polyanionic polymer, and so on (Poncelet et al., 2001). The capsules which possessed a multiple membrane were much stronger and more flexible than their single-membrane counterparts. Consequently, there were fewer ruptured or broken capsules (Goosen, 1996). However, it is essential that the outer surface be biocompatible and negatively-charged (Hommel et al., 1990; Poncelet et al., 2001). Particularly effective coating materials for the outer layer are the highly purified alginates (Dorian and Cochrum, 1997). The binding of chitosan to an alginate gel can be described as almost irreversible. Once the capsules are formed, the exposure to solutions with higher concentrations of competing ions (H+ and Na+) have minor influence on the stability of the polyanionpolycation complex. This high stability is probably caused by cooperative ionic bonds between the oppositely charged polymers. The binding between alginate and chitosan is considerably stronger than the binding between poly-L-lysine and alginate (Gåserød, et al., 1999). McKnight et al. (1988) used chitosan to form a polyelectrolyte complex with calcium alginate beads, resulting in durable, strong, and flexible biocompatible polymeric membranes around the beads. Based on this knowledge, Liu et al. (1997) have prepared a new type of porous polyelectrolyte complex microspheres from alginate and chitosan, and used it to entrap the immune activating growth factor interleukin-2, and also bovine serum albuminfluorescein isothiocyanate (FITC-BSA). Polk et al. (1994) tried oral vaccination of fish through controlled release of antigens from chitosan-alginate microcapsules. They believed the encapsulation technique is a potentially simple, quick and inexpensive method of entrapping bioactive materials for vaccine delivery. 36 Chapter 3 CHAPTER 3 MATERIALS AND METHODOLOGY 3.1 Materials 3.1.1 List of chemicals Chemicals used in this study are listed below: − Acetic acid from Merck (Germany) − Alginic acid sodium salt from Sigma (USA) − Bovine serum albumin from Sigma (USA) − Calcium chloride dihydrate from Merck (Germany) − Chitosan (low molecular weight) from Sigma (USA) − Partial purified hemagglutinin (HA) protein cultured in Prof. Kwang’s Laboratory − Hydrochloric acid from Merck (Germany) − Sodium acetate from Merck (Germany) − Sodium chloride from Merck (Germany) − Sodium citrate from Fluka (Switzerland) − Trizma base from Sigma (USA) Acetic acid is used for preparation of chitosan solution and pH buffer. Sodium alginate is used for microcapsule formation. BSA is used for release study. Calcium chloride dehydrate is used for microcapsule formation. Chitosan is used as coating material. Partial purified HA is the antigen for vaccine delivery. Hydrochloric acid is used for pH buffer preparation. Sodium acetate is used for preparation of chitosan solution and pH buffer. Sodium chloride is used for saline preparation. Sodium citrate is used for dissolution of alginate matrix. Trizma base is used for Tris-HCl buffer preparation. 37 Chapter 3 3.1.2 Stock preparation 1. 1 % (w/v) sodium alginate solution 1) in a beaker introduce 500 ml deionised water 2) mix gently with a magnetic stirrer 3) slowly disperse 5 g alginate powder on the liquid surface 4) leave to stand for two hours to allow alginate grain swelling 5) mixing strongly for 2 hour 6) leave to stand overnight to deaerate Similar steps can be taken for sodium alginate solution at other concentrations. However, it was noted that beyond a 2.0% (w/v) concentration, the high viscosity of the solution often caused the equipment involved to fail and made uniform microcapsule formation difficult. For antigen loaded alginate solution, the target proteins (BSA or HA) and alginate were dissolved separately in 0.1 M Tris-HCl buffer, pH 7.4, at twice the desired final concentration. Equal volumes of the two solutions were mixed giving a final concentration. For example, mixing 50 ml 2% (w/v) alginate and 50 ml 1% (w/v) BSA resulted in 100 ml 1% (w/v) alginate and 0.5% (w/v) BSA solution. 2. 1 % (w/v) calcium chloride solution 1) in a beaker introduce 500 ml deionised water 2) mix gently with a magnetic stirrer 3) slowly disperse 6.62 g calcium chloride dihydrate powder on the liquid surface 4) mixing strongly for 30 minutes 5) leave to stand for 2 hours to deaerate 38 Chapter 3 3. 0.3 % (w/v) chitosan solution 1) introduce 500 ml 1 % acetic acid solution into a beaker 2) mix gently with a magnetic hot plate stirrer 3) slowly disperse 1.5 g chitosan powder on the liquid surface 4) mixing strongly until dissolution 5) raise the pH to 6.0 by sodium acetate 6) leave to stand overnight to deaerate 3.2 Experimental setup 3.2.1 Equipments and accessories Various equipments and accessories used in this study are listed in Fig. 3.1, (a) PVC tubing from Cole-Parmer (USA) (b) 10 ml and 20 ml plastic syringes from Terumo (Japan); 5 ml plastic syringes from Becton & Dickinson (USA) (c) 4” × 1” Petri dishes from Iwaki Glass (Japan) (d) Needles, 33G from Popper & Sons (USA), others from Becton & Dickinson (USA) (e) Single syringe infusion pump from KD Scientific (USA) (f) High voltage generator from Spellman HV Electronics (Germany) 3.2.2 System configurations The electrostatic extrusion system (Fig. 3.2) consisted of a syringe, storing antigen and sodium alginate solution, which was driven by a syringe pump; The solution was fed to a needle via PVC tubing; The needle serving as a spray nozzle, was charged by the high voltage generator; A Petri dish containing CaCl2 solution was placing on top of an adjustable stand, 1 - 4 cm below the needle, where microcapsules were formed. 39 Chapter 3 c a b d e f Figure 3.1 Figure 3.2 Pictures of equipments and accessories System configurations of electrostatic extrusion process 40 Chapter 3 3.3 Experimental procedures 3.3.1 Microcapsule formation The microcapsules containing active antigens were formed by the following steps: 1) The antigens were suspended uniformly in sodium alginate solution in Tris-HCl buffer or other buffers. 2) Introduced about 10 mL of the solution to a 10-mL plastic syringe, connected the syringe with the needle via PVC tubing, put back the plunger, and attached the syringe to the syringe pump. 3) Introduced 50 ml of 1% calcium chloride into a Petri dish. Positioned the supporting stand so that the needle tip was about 1-4 cm from the surface of the CaCl2 collection solution. To avoid surface aggregates, the surface tension of the CaCl2 solution was lowered with addition of 5% (v/v) of ethanol. 4) Mixed the collection solution with a magnetic stirrer at low speed. 5) Attached the positive electrode wire to the stainless steel needle and the ground wire to the collection solution. 6) Switched on the syringe pump and waited for the first few drops to come out of the end of the needle. After the first drop or two has been produced, switched on the voltage power supply. Desired voltage values were preset. 7) Needle oscillation might be observed as voltage rose to its critical value. This needle vibration would produce a bimodal microcapsule size distribution. 8) Left the microcapsules to stand for 30 min to ensure full gelation. 9) The microcapsules were centrifuged using an Allegra X-12R centrifuge (Beckman Coulter, UK) at 1000 rpm for 5 min and the supernatant decanted. 10) For chitosan coating, an outer layer was formed by dipping the microcapsules into 30 mL of a 0.3% (w/v) chitosan solution for 2-10 minutes. Then, those 41 Chapter 3 microcapsules were centrifuged at 1000 rpm for 5 min and the supernatant decanted. After that, microcapsules were washed with 30 mL each of 0.1 M Tris-HCl buffer and 1.0% CaCl2 solution. 11) Finally, the microcapsules were stored in Tris-HCl buffer or other mediums for release study. 3.3.2 Antigen release 3.3.2.1 Release medium preparation All mediums used in this study are 0.1 M Tris-HCl buffer pH 7.4 or otherwise stated. Proper pH buffers were prepared and used as release medium for antigen release study. 1. 0.1 M Tris-HCl buffer pH 9 1) Stock solution A: 0.2 M solution of tris(hydroxymethyl)aminomethane (24.2 g Trizma base in 1000 ml of distilled water) 2) Stock solution B: 0.2 M HCl (17.24 ml concentrated HCl dilute to 1000 ml with distilled water) 3) Mixing 50 ml of A and 5.0 ml of B and diluted to a total of 100 ml with distilled water 2. 0.1 M Tris-HCl buffer pH 7.4 1) Stock solution A: 0.2 M solution of tris(hydroxymethyl)aminomethane (24.2 g Trizma base in 1000 ml of distilled water) 2) Stock solution B: 0.2 M HCl (17.24 ml concentrated HCl dilute to 1000 ml with distilled water) 3) Mixing 50 ml of A and 41.4 ml of B and diluted to a total of 100 ml with distilled water 42 Chapter 3 3. Sodium acetate buffer pH 5.0 1) Stock solution A: 0.2 M acetic acid (11.5 ml glacial acetic acid dilute to 1000 ml with distilled water) 2) Stock solution B: 0.2 M solution of sodium acetate (27.2 g sodium acetate trihydrate or 16.4 g sodium acetate anhydrate in 1000 ml of distilled water) 3) Mixing 10.5 ml of A and 39.5 ml of B and diluted to a total of 100 ml with distilled water 4. Saline Dissolving 9 g sodium chloride in 1000 ml of distilled water 3.3.2.2 BSA release 1) Microcapsules, generated from 4 ml alginate-BSA (1% w/v alginate, 0.5% w/v BSA) solution, were added to 100 ml of release medium in 100 ml conical flask. 2) The flasks were then placed on a Forma Orbital Shaker (Thermo Electron Co., USA) at room temperature and stirred at 200 rpm. 3) At each sampling period, 3 ml of release medium was removed and replaced with fresh medium. 4) The removed sample was analyzed for BSA concentration via UV absorption at 280 nm; all tests were done in triplicate. 3.3.2.3 HA release 1) Microcapsules, generated from 2 ml alginate-HA (1% w/v alginate, 0.075% w/v partial purified HA) solution, were added with 1 ml of Tris-HCl buffer solution in 2 ml microtubes. 43 Chapter 3 2) The microtubes were then placed on a shaker at room temperature and stirred at 200 rpm. 3) At each sampling period, the microtubes were centrifuged using a Denville 2600 microcentrifuge (Denville Scientific, USA) at 10,000 rpm for 4 min; all supernatant was removed and replaced with fresh medium. 4) The removed samples were analyzed by a BCA protein assay kit (Pierce, USA) and SDS-PAGE/Western blotting method (see appendices); all tests were done in triplicate. 3.4 Determination of microcapsule properties 3.4.1 Particle size Microcapsule size was determined by laser diffraction using Coulter LS 230 Particle Size Analyzer (Beckman Coulter, UK). Microcapsule samples, generated from approximately 10ml of alginate solution, were dispersed in 1.0% (w/v) CaCl2 solution. This suspension solution was used instead of de-ionized water to minimize the swelling of the microcapsules during the size analysis process. The suspension was then added to the sample chamber of a Coulter LS 230 under moderate stirring. Particle size distribution was then determined as a function of the particle diffraction using the Coulter software, and plotted as a function of volume percentage. All tests were done in triplicate. To estimate the droplet size by scaling laws, the electrical conductivity of the sodium solution was measured by Oakton Conductivity meter (Eutech, Singapore). The density of the sodium solution was measured by balance (Mettler Toledo, B204-S, Switzerland). 44 Chapter 3 3.4.2 Microcapsule morphology Scanning electron microscopy (SEM) (JEOL, JSM-5600V, Japan) was used to determine microcapsule morphology. Microcapsules, after freeze-dried, were mounted on aluminium SEM stubs using double-sided NEM tape (Nisshin EM, Japan). The samples were sputter coated with platinum for 40 seconds under a vacuum environment with an auto fine coater (JEOL, JFC-1300, Japan). Coated samples were examined with SEM operating at an accelerating voltage of 15 kV. In addition to SEM, a fluorescence microscope (Leica DMLM, Germany) and an optical microscope (Olympus ULBD-2, Japan) were used to determine the morphology. In this way, the microcapsules need not go through freeze drying, and they were able to retain their original shape for morphology study. 3.4.3 BSA release For BSA release study, a 3 ml sample of each solution was removed periodically for analysis, and 3 ml medium was added after each removal to maintain a constant volume. The absorbance of each solution collected was determined spectrophotometrically at 280 nm using a UV-VIS spectrophotometer (UV Mini 1240, Shimadzu, Japan). Meanwhile, total concentration of BSA in the microcapsules was determined after the microcapsules were dissolved completely in sodium citrate (10% w/v) solution. And the amount of BSA lost to the external collecting solution during the extrusion process was also analyzed by UV spectrophotometer as described above. The encapsulation efficiency was calculated as the ratio of the actual and the theoretical BSA content. 45 Chapter 3 3.4.4 HA release HA concentrations in the samples were measured by BCA Protein Assay Kit (Pierce, USA) and SDS-PAGE/Western blotting methods (Gallagher and Smith, 2005a; Gallagher et al., 2005b; Sasse and Gallagher, 2005). Basically, BCA protein assay combines the well-known reduction of Cu+2 to Cu+1 by protein in an alkaline medium (the biuret reaction) with highly sensitive and selective colorimetric detection of the cuprous cation (Cu+1) using a unique reagent containing bicinchoninic acid (BCA). The purple-coloured reaction product of this assay is formed by the chelation of two molecules of BCA with one cuprous ion. This watersoluble complex exhibits a strong absorbance at 562 nm that is nearly linear with increasing protein concentration. Accordingly, protein concentrations could be determined and reported with reference to standards of a common protein such as BSA. Since proteins with different molecular weight would have different size, we could also use sodium dodecyl sulfate (SDS) to denature all proteins to the same linear shape, followed by polyacrylamide gel electrophoresis (PAGE). In this way, proteins were separated according to their size. Proteins separated by SDS-PAGE could be detected using Coomassie blue. After destaining, the blue protein bands appear against a clear background. The protein located around 35 kDa marks was believed to be HA protein. It could be further confirmed by Western blotting method, which allowed researchers to identify certain proteins and to measure relative amounts of the proteins present in different samples. 46 Chapter 4 CHAPTER 4 4.1 RESULTS AND DISCUSSION Optimum conditions for antigen encapsulation The size of the microcapsules formed is absolutely crucial to the successful implementation of this technology. As mentioned earlier, the microcapsules were analyzed using the particle size analyzer to determine its mean size. A typical result obtained from the particle size analyzer is shown in Fig. 4.1. Notably, satellite peaks are often observed in the size distribution profile. Volume (%) Differential Volume 35 30 25 20 15 10 5 0 1 10 100 1000 Particle Diameter (μm) Figure 4.1 A typical result from the particle size analyzer Satellites are formed by breakage of the fine filament between the droplet and the needle tip just before separation, resulting in secondary peaks (Fig. 4.2). 47 Chapter 4 Figure 4.2 Satellite formation under an electrical potential (Speranza et al., 2001) In the following sections, several factors governing the electrostatic extrusion process will be discussed, namely, applied voltage, pump flow rate, electrode spacing, and needle size. Experiments were designed to determine the optimum conditions for production of smaller microcapsules. A set of standard conditions is listed in Table 4.1. Table 4.1 Control parameters for microencapsulation experiments Control Parameters Set Values Units Applied Voltage 8.0 kV Syringe Pump Flow Rate 5.0 ml/h Electrode Spacing 2.0 cm Needle Size 30 G Sodium Alginate Concentration 1.0 % w/v Calcium Chloride Concentration 1.0 % w/v The set values would be replaced if it did not match the optimum condition for microcapsules production, for example, if smaller microcapsules formed at 7 kV, then the set value for applied voltage would become 7 kV for following experiments. 48 Chapter 4 4.1.1 Effect of applied voltage Table 4.2 and Figure 4.3 showed the experimental conditions and results of changes in applied voltage on the microcapsule sizes. Table 4.2 Effect of applied voltage on microcapsule formation Control Parameters y Syringe Pump Flow Rate — 5.0 ml/h y Electrode Spacing — 2.0 cm y Needle Size — 30 G y Na-Alginate Concentration — 1.0% w/v y Calcium Chloride Concentration — 1.0% w/v Variable Parameter & Results Applied Voltage (kV) 0 1 2 3 4 5 6 7 8 10 1383 1314 1105 1025 603 160 146 92.4 109 177 Coefficient of Variation (%) 10.2 33.7 52.7 38.2 42.3 46.4 62.7 Microcapsule size ( μm) Microcapsule sizes (μm) 12.4 14.9 20.3 1600 1400 1200 1000 800 600 400 200 0 0 2 4 6 8 10 12 Applied voltage (kV) (a) Applied voltage vs. microcapsule size 49 Chapter 4 70 60 C.V. % 50 40 30 20 10 0 0 2 4 6 8 10 12 Applied voltage (kV) (b) Applied voltage vs. size distribution 70 60 C.V.% 50 40 30 20 10 0 0 500 1000 1500 Microcapsule size (μm) (c) Figure 4.3 Microcapsule size vs. size distribution Effect of applied voltage on microcapsule formation As shown by Fig. 4.3 (a), it was clear that as electric potential increases, the size of the microcapsule decreased dramatically. However, as applied voltage was raised above a critical value, there was no significant further decrease in microcapsule diameter. As a result, the solution would exit the needle as a jet, and the size of the capsules was mainly determined by the jet instability rather than the applied voltage. Another interesting finding was, as electric potential increases, the size distribution of the 50 Chapter 4 microcapsules increased as well (Fig. 4.3 b). By plotting the size distribution as a function of microcapsule size; the size distribution increased when the microcapsule size decreased (Fig. 4.3 c). In my study, it was not possible to obtain small microcapsules with a narrow size distribution. As far as size distribution is concern, the rising trend was primarily caused by the change of dispersion mode, e.g., from dripping mode to cone-jet mode, under electrical field. Without electric field, the droplets were detached from the tip of the needle separately (Fig. 4.4 a). The droplet was gradually drawn into a cone by the electric field (Fig. 4.4 b), and as the droplet falls, a fluid filament connects the droplet to the next forming droplet. Beyond a critical applied voltage, droplets no longer formed; instead the solution poured from the needle in a steady stream (Fig. 4.4 c). (a) V = 0 kV Figure 4.4 (b) V = 4 kV (c) V = 6 kV Transition from dripping mode to cone-jet mode (by digital camera) As shown in Fig. 4.2, just prior to release from the needle, the droplets remain attached by a small liquid column. Turbulence or a vibration of even very small intensity was able to break this liquid cylinder. This secondary breakage leads to small droplet satellites. Sample and Bollini (1972) stated that, satellite formation usually 51 Chapter 4 accompanies with higher applied voltages, since under these conditions the elongation of the liquid neck prior to rupture is much more pronounced. Meanwhile, as the voltage was increased above its critical value, a high surface charge and an electric field on the surface of capillary tip gave rise to a mechanical force causing needle vibration and resulting in an oscillating thread-like filament. The greatest deduction in microcapsule size was observed when natural needle oscillation was observed, that means, oscillation in other frequencies will form larger microcapsules. Therefore, the smallest micropcapsules were observed at 7 kV of applied voltage in my experiment, which was also corresponding to a relatively narrow size distribution. As the applied voltage increased further to 10 kV, kink instability occurred, causing an unstable jet to be emitted from the needle, resulted in a much larger size distribution (Fig. 4.5). (a) V = 7 kV Figure 4.5 (b) V = 10 kV Stable and unstable jet emission from the needle Following the work done by Poncelet et al. (1999b), the microcapsules size may be estimated by equation (2.9), or ln d(U ) 1 ≅ ln [ 1 − ( U / U cr )2 ] . The numerical do 3 52 Chapter 4 estimation of the critical electrical potential by equation (2.6) or U cr = d c γ o / kε o gives the value Ucr = 1.6 kV, for γo = 0.072 N/m, dc = 0.3 mm, εo = 8.85 x 10-12 F/m and k = 1. This value is significantly different from the experimentally observed value of Ucr ranging from 4 kV to 5 kV. Therefore, an appropriate fitting parameter k must be assigned for valid estimation of Ucr value. Meanwhile, Poncelet et al. (1993), proposed another equation for estimation of Ucr: U cr = 9.419 h 0.2 (4.1) Where, h is the distance between the droplet and collecting solution. The estimation of the critical electrical potential by this equation gives the value Ucr = 4.3 kV, for h = 0.02 m. This value fits well within the experimental observed range of Ucr. 1 y = 0.4198x R2 = 0.9944 -ln(d/do) 0.8 0.6 y = 0.3333x R2 = 1 0.4 0.2 0 0 0.5 1 1.5 2 2.5 -ln(1-(U/Ucr )^2) Figure 4.6 My data Poncelet's model Linear (My data) Linear (Poncelet's model) Linear models for microcapsule size estimation Comparing Poncelet’s model, where d (U ) ≅ d o [1 − (U / U cr ) ] , with my 2 1/ 3 experimental data (Fig. 4.6), I found that my data also supported a good linear model for microcapsule size estimation, but with differences in exponential coefficient. It 53 Chapter 4 might due to the effect of droplet shrinkage during microcapsules formation. Therefore, the sizes measured after microcapsule formation would be different from the sizes of the droplets emitted from the needle. 4.1.2 Effect of flow rate The tests for flow rate were carried out at 7 kV, which was found to be the optimum electric potential for my experiment. Table 4.3 and Figure 4.7 showed the experimental conditions and results of changes in flow rate on the microcapsules sizes. Table 4.3 Effect of flow rate on microcapsule formation Control Parameters y Applied Voltage — 7.0 kV y Electrode Spacing — 2.0 cm y Needle Size — 30 G y Na-Alginate Concentration — 1.0% w/v y Calcium Chloride Concentration — 1.0% w/v Variable Parameter & Results Flow rate (ml/hr) 2 3 5 7 8 10 Microcapsule sizes (μm) 68.2 80.5 92.4 120.4 142.7 150.9 Coefficient of Variation (%) 40.3 41.2 42.3 44.5 46.5 62.1 54 Microcapsule size ( μm) Chapter 4 160 140 120 100 80 60 40 20 0 0 2 4 6 8 10 12 10 12 Flow rate (ml/h) (a) Flow rate vs. microcapsule size 70 60 C.V. % 50 40 30 20 10 0 0 2 4 6 8 Flow rate (ml/h) (b) Figure 4.7 Flow rate vs. size distribution Effect of flow rate on microcapsule formation When the pump flow rate was slower, a less volume of solution was assembled at the needle tip at the time of droplet detachment, and smaller droplets could be expected. Another explanation was, lower flow rate would give more time for counter ions, e.g., Na+, to diffuse to droplet surface under positive electric potential and reduce the surface tension, and thus smaller microcapsules could be generated. It was confirmed by my experimental data (Fig. 4.7 a). However, though a slower flow rate would 55 Chapter 4 generate smaller microcapsules, the production capacity also dropped significantly. On the other hand, as flow rate rose above 10 ml/h, the syringe pump could no longer work properly, i.e., the flow rate became inconsistent, and this limited the production capacity for my system. Microcapsule size distribution also increased steadily with increasing flow rate (Fig. 4.7 b). With increasing flow rate, more alginate solution was passing through the needle per time, resulted in larger microcapsules and surface tension. According to Poncelet et al. (1993), the greater the force required to break the droplet, the higher would be the resultant size dispersion, so were my results. After balancing the tradeoff among beads size, size distribution, and production capacity, the optimum condition for syringe pump flow rate was set to be 5 ml/h. Meanwhile, I also examined the scaling laws developed by Gañán-Calvo et al. (1997), which established a relationship between droplet size and volumetric flow rate, in my study. Table 4.4 Physical properties of 1% Na-alginate solution (* from Watanabe et al., 2003; others by my measurement) Density (kg/m3) Surface tension (N/m)* Viscosity (Pa s)* Electrical conductivity (S/m) Permittivity of the vacuum (F/m) 1002 0.0732 0.123 0.2024 8.85 x 10-12 Using the liquid properties listed in Table 4.4, the dimensionless viscous parameter δμ, calculated by equation (2.18), has a value of 0.002495 for the 1% (w/v) sodium 56 Chapter 4 alginate solution. Since δμ < 1, then I have d ∝ jv1/3 or d = k jv1/3. My data showed a Microcapsule size ( μm) poor correspondence to the scaling laws (Fig. 4.8). 160 y = 63.563x R2 = 0.8317 140 120 100 80 60 40 20 0 0 0.5 1 1.5 2 2.5 Jv^(1/3) Figure 4.8 Scaling laws for 1% Na-alginate solution Therefore, the scaling laws need to be modified to a more general form to provide better droplet size estimation in my case. According to Gañán-Calvo et al (1997), if charge is advected with a velocity of the order of the bulk velocity jv/d2 (highly viscous and conducting liquids) one has d ~ jv1/3; on the other hand if charge is advected with a larger velocity, almost independent of jv (liquid with sufficiently low viscosity and conductivity) then d ~ jv1/2. Therefore, I proposed the following modified scaling laws: d ∝ jvn (4.2) d = k' jvn (4.3) ln d = n ln jv + ln k' (4.4) 57 Chapter 4 6 y = 0.5083x + 3.8265 R2 = 0.9555 5 ln(d) 4 3 2 1 0 0 0.5 1 1.5 2 2.5 ln(Jv) Figure 4.9 Modified scaling laws for 1% Na-alginate solution As shown in Fig. 4.9, a much better corresponding relationship between flow rate and microcapsule size was presented by the modified scaling laws. In the experiment, the results show n ≈ ½, which means the viscosity or the surface tension of the solution has been reduced significantly under the electrostatic force. 4.1.3 Effect of electrode spacing The electrode spacing was defined as the distance between the tip of the needle and the surface of the collecting solution. Table 4.5 and Figure 4.10 showed the experimental conditions and results of changes in electrode spacing on the microcapsules sizes. 58 Chapter 4 Table 4.5 Effect of electrode spacing on microcapsule formation Control Parameters y Applied Voltage — 7 kV y Syringe Pump Flow Rate — 5.0 ml/h y Needle Size — 30 G y Na-Alginate Concentration — 1.0% w/v y Calcium Chloride Concentration — 1.0% w/v Variable Parameter & Results 1 1.5 2 2.5 3 4 Microcapsule sizes (μm) 95.4 88.0 92.4 119.3 130.0 147.8 Coefficient of Variation (%) 45.0 39.3 42.3 50.7 53.8 63.6 Microcapsule size ( μm) Electrode Spacing (cm) 160 140 120 100 80 60 40 20 0 0 1 2 3 4 5 Electrode spacing (cm) (a) Electrode spacing vs. microcapsule size 59 Chapter 4 70 60 C.V. % 50 40 30 20 10 0 0 1 2 3 4 5 Electrode spacing (cm) (b) Figure 4.10 Electrode spacing vs. size distribution Effect of electrode spacing on microcapsule formation In the study of electrostatic extrusion, I understand that electrostatic force Fe depends on the geometric parameters of the experimental dropping set-ups. For my system, the electrostatic force Fe can be expressed by equation (2.4) or Fe ∝ (1/h)2 (4.5) Where, h is the distance between the droplet and collecting solution. Therefore, a reduction in the electrode spacing would increase the electrical force and intensity of the electric field between the needle and the solution. According to equation (2.3), any increase in Fe would help counteract the surface tension, and produce smaller droplets. Hence, microcapsule sizes would decrease as the electrode spacing decreases (Fig. 4.10 a). However, the size distribution of microcapsules increased with the electrode spacing (Fig. 4.10 b). One possible explanation for this phenomenon was, when the various droplets penetrate the collecting solution, they might break again, giving rise to even smaller droplets and/ or satellites. The larger distances between the needle tip and the 60 Chapter 4 collecting solution, the more likely the droplet would break into satellites when it reached solution surface, hence increased the size distribution. The optimum distance was found to be 1.5 cm, below this value, sparks might be produced across the electrical fields due to ionization of the air. It induced system instability, resulted in relatively larger microcapsule size and wider size distribution at 1 cm electrode spacing, as shown in Fig. 4.10. 4.1.4 Effect of needle size Needles range from 22 Gauge to 33 Gauge were used in this study, however, 33 G needle was highly prone to clogging, and the flow was extremely unstable. Therefore, results from 33 G needle were excluded in this discussion. Table 4.6 and Figure 4.11 showed the experimental conditions and results of changes in needle size on the microcapsule sizes. Table 4.6 Effect of needle size on microcapsule formation Control Parameters y Applied Voltage — 7 kV y Syringe Pump Flow Rate — 5.0 ml/h y Electrode Spacing — 1.5 cm y Na-Alginate Concentration — 1.0% w/v y Calcium Chloride Concentration — 1.0% w/v Variable Parameter & Results 23 25 26 27 30 0.64 0.51 0.46 0.41 0.30 Microcapsule sizes (μm) 411.9 203.3 186.0 120.2 88.0 Coefficient of Variation (%) 23.5 27.7 30.2 34.5 39.3 Needle Size Gauge O.D. (mm) 61 Microcapsule size ( μm) Chapter 4 500 400 300 200 100 0 0.2 0.3 0.4 0.5 0.6 0.7 Needle size (O.D., mm) (a) Needle size vs. microcapsule size 50 C.V. % 40 30 20 10 0 0.2 0.3 0.4 0.5 0.6 0.7 Needle Size (O.D., mm) (b) Figure 4.11 Needle size vs. size distribution Effect of needle size on microcapsule formation According to Tate’s law, surface tension of the droplet is proportional to the diameter of the pendant droplet neck, which is approximately equal to outer diameter of the needle. Therefore, from equation (2.1) and (2.2), I can see a larger do with a larger needle; and from equation (2.6), I can see a larger Ucr value for a larger needle. By combining these two factors in equation (2.9), I can expect a rise in microcapsule size by increasing the needle size, as shown in Fig. 4.11 (a). In practice, with a smaller 62 Chapter 4 orifice, the base area of the Taylor cone was reduced, forcing a sharper apex to be formed at the same flow rate since the volume of liquid expelled remains unchanged. This gave rise to a lower jet diameter and consequently, smaller droplets upon jet collapse. According to Poncelet et al. (1993), any perturbations in electric field or vibration during droplet generation will lead to variation in droplet diameter and formation of satellite droplets. These perturbations act more on small drops or jets, and the small jets are themselves more easily broken than the larger jets. Their opinion was clear reflected in my results, as show in Fig. 4.11 (b). Overall, the optimum conditions for microcapsules formation in my experimental set up were: 7 kV for applied voltage, 5 ml/h for syringe pump flow rate, 1.5 cm electrode spacing and 30 G needle. I also understood that, narrowing the size distribution in the electrostatic extrusion process required a careful design of the system configurations. The needle must be precisely centered to avoid any source of vibration. The distance between the needle tip and the collecting solution must be carefully adjusted. The flow rate driven by the syringe pump must be reliable and consistent. 4.2 Microcapsule morphology Ca-alginate matrix is very porous, as shown in Fig. 4.12, results in a quick release of entrapped antigen from the matrix. Therefore, it is not desired in this study. 63 Chapter 4 (a) SEM picture of Ca-alginate microcapsule surface (b) Optical microscope picture of Ca-alginate microcapsule surface (c) Fluorescence microscope picture of Ca-alginate microcapsule surface Figure 4.12 Ca-alginate microcapsule surface 64 Chapter 4 Gombotz and Wee (1998) suggested that, the porosity of an alginate gel can be significantly reduced by partially drying the microcapsules. Complete dehydration of alginate beads, however, can result in surface cracking which can facilitate the surface erosion of the beads upon rehydration. Batich and Vaghefi (2000) even warned that, any drying distorted the shape and size of the capsules and resulted in a shrunken mass of solid material. Even lyophilization and critical point drying caused distortion. Therefore this approach is not considered in this study. 4.2.1 Effect of pH on microcapsule surface A reduction in pore size of an alginate matrix may also be achieved by exposing the microcapsules to lower pH. Ca-alginate microcapsules formed at different pH will have different surface structure, as shown in Fig. 4.13 (under optical microscope). (a) pH = 5 Figure 4.13 (b) pH = 7.4 (c) pH = 9 pH effects on microcapsule surface structure It was shown that at pH 5, a layered structure matrix was observed. And those formed at pH 9 were shown to be more porous than at pH 7.4, which would result in a faster release of the beads at higher pH values. This different surface structure would affect the BSA release from the alginate matrix latterly. A possible explanation for it may be the lower the pH, the higher the [H+]. As Ca-alginate microcapsule is negatively charged, the H+ will help neutralizing the negative charge and coating a hydrous layer on the microcapsule surface, hence make it smoother. 65 Chapter 4 4.2.2 Effect of chitosan coating on microcapsules surface The surface porosity of microcapsule will be reduced significantly by polymer coating, which ensures better isolation and retention of the encapsulated material. In my case, the ca-alginate microcapsules were coated with 0.3% chitosan solution. The results were shown in Fig. 4.14. The additional layer of chitosan coating on the calcium alginate microcapsules can be clearly seen in the optical microscope after coating in chitosan for 2 minutes (Fig. 4.14 b). No coating (a) Fluorescence microscope pictures of Ca-alginate microcapsule surface No coating (b) With chitosan coating With chitosan coating Optical microscope picture of Ca-alginate microcapsule surface layer Figure 4.14 Ca-alginate microcapsule surface (No coating Vs. Coating) 66 Chapter 4 Prolonging the coating time resulted in the formation of thicker and more compact membranes due to the greater number of ionic bonds established between chitosan and the alginate core. This could be seen in the SEM pictures in Figure 4.15. The surface was observed to be much more compact after coating with chitosan for 10 minutes. (a) No chitosan coating (b) 2 minutes chitosan coating (c) 4 minutes chitosan coating (d) 10 minutes chitosan coating Figure 4.15 SEM pictures of microcapsules with different coatings Similar results were also presented by Taqieddin and Amiji (2004), the calcium alginate chitosan beads in the Fig. 4.16 can be seen as having a highly porous core, resulting from freeze-drying of liquid alginate, and a uniform chitosan layer, that hindered the diffusion of BSA out of the calcium alginate chitosan beads. 67 Chapter 4 Figure 4.16 4.3 SEM picture of freeze-dried Ca-alginate–chitosan microcapsule (Taqieddin and Amiji, 2004) Antigen release Antigens entrapped in alginate matrices are released by two mechanisms: i) diffusion of the proteins through the pores of the polymer network, and ii) degradation of the polymer network. Analysis of Ca-alginate microcapsules by microscopy has shown that the pore size ranges from 5 nm to 200 nm in diameter (Fig. 4.12). There is a more constricted polymer network on the microcapsule surface than in the microcapsule core (Fig.4.16). Table 4.7 lists several factors that may be exploited to control membrane permeability in microencapsulation systems. In my study, effects of differences in size, pH, loadings, chitosan coating time and release mediums were studied for BSA release from the alginate matrix, while only effect of coating was examined in HA release. 68 Chapter 4 Table 4.7 Factors in the control of microcapsule permeability (Goosen, 1994) Capsule Wall Process Factors Capsule Solute 4.3.1 • Polymer molecular weight • Physical integrity • Thickness • Chemical structure • Multiple membranes • Membrane polymer concentration • Reaction time • pH • Additives • Purity • Size • Swelling • Shape • Size • Shape • Electrical charge BSA release from alginate matrices A typical release profile for BSA is shown in Fig. 4.17. Initially, a burst of BSA was observed due to release of BSA located near the microsphere surface. This was followed by a period of slow release which was attributed to degradation of the microsphere and diffusion of the drug out of the microsphere. The third phase was described as a secondary burst of BSA, and was attributed to the increased solubilization and erosion of the polymeric matrix. 69 Chapter 4 BSA released (%) 100 80 60 40 20 0 0 5 10 15 20 25 Time (days) Figure 4.17 The release of BSA from alginate matrix (Parameters: Size 0.15 mm; 0.5% BSA loading; No coating & Tris-HCl buffer with pH 7.4) 4.3.1.1 Effect of microcapsule size The effect of microcapsule size on the release profiles of BSA is shown in Fig. 4.18. While small microcapsules had a fast release profile, larger microcapsules had slower release rates. BSA released (%) 100 80 60 40 20 0 0 5 10 15 20 25 Time (days) 0.15 mm Figure 4.18 0.7 mm 1.6 mm Effect of microcapsule size on BSA release 70 Chapter 4 The microcapsules with mean particle size of 150 μm exhibited an initial release of about 40% of BSA loaded within the first day, comparing to about 25% for 700 μm microcapsules and about 25% for 1600 μm microcapsules. This was attributed to the small microcapsule size and resultant increase in surface area. And mass transfer is directly proportional to the surface area of the spherical microcapsules. 4.3.1.2 Effect of pH As the alginate gel network is negatively charged, the pH will also influence the diffusion of charged substrates and products. The release profiles of the BSA from Caalginate microcapsules, sized around 150 μm, at pH 5, pH 7.4 and pH 9 were studied. BSA released (%) 100 80 60 40 20 0 0 5 10 15 Time (days) pH 5 Figure 4.19 pH 7.4 pH 9 Effect of pH on BSA release As shown in Fig. 4.19, the rate of release of BSA from alginate microcapsules increases with increasing pH as a result of the increased negative charges on the protein and increased pore size (Fig. 4.13). After 24 h, only 25% of the entrapped BSA was released from the capsules at pH 5.0, compared with 40% release at pH 7.4 and 55% release at pH 9.0. 71 Chapter 4 A protein with overall net positive charge can potentially interact with the negatively charged alginate polymer, thus inhibiting diffusion from the gel. And according to Gombotz and Wee (1998), acidic proteins are less likely to interact with the anionic alginate polymer; while basic proteins interact with the alginate polymer network and hence diffusion through the pores is greatly hindered. BSA has an isoelectric point of 4.90, and belongs to acidic proteins. It will be negatively charged at pH 7.4 and 9, and almost neutral at pH 5. However, an attempt to entrap BSA at pH 4 failed, as precipitation occurred while preparing BSA-alginate solutions. 4.3.1.3 Effect of BSA loading Different loadings of BSA would result in different BSA release patterns from the BSA released (%) microcapsules (Fig 4.20). 100 90 80 70 60 50 40 30 20 10 0 0 5 10 15 20 25 Time (days) 0.75% BSA Figure 4.20 0.5% BSA 0.25% BSA Effect of BSA loading on BSA release Generally, a higher initial loading of BSA would result in a faster release from the microcapsules, due to higher driving force for diffusion from the microcapsules. For 72 Chapter 4 instance, most BSA was released out from the microcapsules during the first 2 weeks at a 0.75% w/v BSA loading. 4.3.1.4 Effect of chitosan coating Figure 4.21 shows the effect of coating time of chitosan on the release profiles of BSA. BSA released (%) 100 80 60 40 20 0 0 5 10 15 20 25 Time (days) No coating Figure 4.21 2 min coating 10 min coating Effect of coating time of chitosan on BSA release As shown in Fig. 4.21, the release of BSA decreased when coating time increased. With coating period longer than 10 min, the mass transfer was seriously hindered. This was attributed to the additional chitosan layer formed around the Ca-alginate microcapsules. With chitosan coating, surface of the microcapsules became much less porous and thus reduced the rate of diffusion significantly. When Ca-alginate microcapsules were suspending in the chitosan solution, the formation of polyelectrolyte complexes between alginate and chitosan occurred. With short coating time, it might not give the sufficient interaction time for the chitosan- 73 Chapter 4 alginate matrix. Therefore, the Ca-alginate microcapsules coated for 2 minutes showed faster release behavior that those coated for 10 minutes. 4.3.1.5 Effect of release environment BSA released (%) The release profiles of BSA in different mediums are shown in Fig. 4.22. 100 80 60 40 20 0 0 5 10 15 20 25 Time (days) Tris-HCl Buffer (pH 7.4) Saline Tris-HCl Buffer (No shaking) Saline (Total replacement) Figure 4.22 Effect of release medium on BSA release A significant drop in release rate was observed if I do not shake the release medium. Without shaking, there would be an accumulation of boundary layer around the microcapsules, which would delay the release of BSA from the microcapsules. The release of BSA in saline solution was slower than in Tris-HCl buffer, because saline solution has a pH around 5.5, resulting from ionic strength effect and dissolution of CO2 from the air. Hence, it followed the pattern of pH effect on BSA release. However, if I replaced the entire saline medium during sampling, a much faster release of BSA from the microcapsules would be observed, because the excess 74 Chapter 4 Na+ would accelerate the erosion of Ca-alginate microcapsules. If fact, all the microcapsules were dissolved within one week. Encapsulation efficiency 4.3.1.6 Encapsulation efficiency 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% Microcapsule size pH BSA loading Coating Time Release medium Release conditions 0.15 mm / pH 5 / 0.25% BSA / No coating / Tris-HCl buffer (pH 7.4) 0.7 mm / pH 7.4 / 0.5% BSA / 2 min coating / Saline 1.6 mm / pH 9 / 0.75% BSA / 10 min coating / Figure 4.23 BSA encapsulation efficiency and release conditions Figure 4.23 showed the relationship between BSA encapsulation efficiency and the different release conditions. — Effect of size: the encapsulation efficiency increased with increase of microcapsule size, it was assumed that the BSA lost to CaCl2 solution during the microcapsule formation was directly proportional to the specific surface area of the microcapsules, i.e., large microcapsules lost less BSA to the collecting solution. — Effect of pH: pH has limited effect on encapsulation efficiency; however, the efficiency has a relatively higher value at pH 5 than others. BSA has an isoelectric point of 4.90, so it’s almost neutral at pH 5, hence it’s less repulsive with the negatively charged alginate polymer comparing to higher pH values, thus inhibiting diffusion from the microcapsules. Therefore, less BSA will be lost 75 Chapter 4 during encapsulation at pH 5. Meanwhile, pH 5 buffer (sodium acetate) consists of Na+, which would accelerate the exchange between Na+ and Ca2+, hence stimulated the crosslinking and formation of microcapsule wall, which eventually reduced the loss of BSA from the microcapsules. — Effect of BSA loading: With a higher BSA loading, more BSA would be present in the CaCl2 solution during ionic crosslinking, which could prohibit diffusion of BSA from microcapsules, and reduce the loss. Therefore, microcapsules generate with 0.75% BSA loading have the highest encapsulation efficiency at about 27%. — Effect of chitosan coating: as coating occurred after the formation of Ca-alginate microcapsules, some BSA would be certainly lost during the coating process, therefore the encapsulation efficiency would decrease as coating time increased. — Effect of release medium: same as sodium acetate buffer, saline consists of Na+, which would accelerate the formation of microcapsule wall, and hence reduced the loss of BSA from the microcapsules. Therefore, encapsulation efficiency was higher in saline medium compared to Tris-HCl buffer. 4.3.2 HA release from alginate matrices Slightly truncated HA proteins were firstly generated by the baculovirus expression system, in which insect cells (Spodoptera frugiperda clone 9 or Sf-9 cells) were infected with recombinant baculovirus encoded HA genes. These Sf-9 cells were later suspended in lysis buffer, lysed using ultrasonication. The partially purified HA proteins were finally obtained after centrifugation. In vitro release pattern of HA protein in Tris-HCl buffer (0.1M, pH 7.4), which was generated in terms of the cumulative release (%) versus time, was shown in Fig. 4.24. 76 Chapter 4 The total amount of partial purified HA proteins encapsulated was calculated to be 0.22 mg. An initial burst over the first three days was observed. About 20% proteins were released within the first 24 h. And only 70% of proteins were released from Caalginate microcapsules after 4 weeks. Cumulative release (%) 100 80 60 40 20 0 0 5 10 15 20 25 30 Time (days) Figure 4.24 HA release from alginate matrix in Tris-HCl buffer After chitosan coating, the release rate was much slower, and similar patterns were Cumulative release (%) shown by BSA samples (Fig. 4.25). 100 80 60 40 20 0 0 5 10 15 20 25 30 Time (days) HA Figure 4.25 HA+coating BSA BSA+coating Effect of chitosan coating on HA release 77 Chapter 4 SDS-PAGE was also performed to investigate the release pattern of HA (Fig. 4.26). The bands fainted as time went on, indicating the descending of HA release rate. And the identity of HA was further confirmed by Western blotting, as shown in Fig. 4.27. kDa 1 2 3 4 5 6 7 8 9 175 83 62 47.5 32.5 25 16.5 Figure 4.26 kDa SDS-PAGE results of HA release (lane 1, protein marker; lane 2, HA control; lane 3-9, HA samples after 1, 3, 5, 7, 14, 21, 28 days in vitro release) 1 2 3 4 5 6 7 8 9 175 83 62 47.5 32.5 25 16.5 6.5 Figure 4.27 Western blotting results of HA release (lane 1, protein marker; lane 2, HA control; lane 3-9, HA samples after 1, 3, 5, 7, 14, 21, 28 days in vitro release) 78 Chapter 5 CHAPTER 5 5.1 CONCLUSIONS AND RECOMMENDATIONS Conclusions Electrostatic extrusion is a novel extrusion technique that uses electrostatic forces to form a charged stream of small droplets. Alginate, a naturally polysaccharide, can be ionically crosslinked by the addition of divalent cations in aqueous solution. The combination of these two aspects has enabled me to develop a technique for the entrapment and/or delivery of antigen in control of pathogen. I found that applied voltage, flow rate, electrode spacing and needle size played theirs roles in microcapsules formation. As applied voltage increased until a critical value, the size of the microcapsules decreased dramatically, but the size distribution of the microcapsules increased. As a result, it was not possible to obtain small microcapsules with a narrow size distribution in present conditions. A modified linear relationship was observed between microcapsules size and applied voltage based on Poncelet’s model. Above the critical value, there was no significant decrease in microcapsule diameter by higher voltage. Further reduction in microcapsule size was achieved by reducing flow rate, which would give more time for counter ions to diffuse to droplet surface and minimize the surface tension. I also modified the scaling laws, developed by Gañán-Calvo and fellows, to show a better corresponding relationship between flow rate and microcapsule size in my study. As expected, microcapsule sizes decreased as the electrode spacing decreased, due to increase in electrical force and intensity of the electric field between the needle and the solution. And smaller needles produced smaller microcapsules. The overall 79 Chapter 5 optimum conditions for microcapsule production with 1% w/v Na-alginate and 1% w/v CaCl2 were: 7 kV for applied voltage, 5 ml/h for syringe pump flow rate, 1.5 cm electrode spacing and 30 G needle. Pictures taken by SEM and other microscopes showed that, Ca-alginate matrix has very porous surface structure. A reduction in pore size of an alginate matrix was observed by exposing the microcapsules to lower pH. The surface porosity was clearly reduced by applying chitosan coating. The additional layer of chitosan coating on the calcium alginate microcapsules was confirmed by observation under microscopes. In vitro results showed protein release was affected by microcapsule size, pH, protein loading, and release medium. A faster protein release was observed for smaller microcapsules, at higher pH, with higher loading, and excess Na+ presence. However, there would be significantly delay in protein release if chitosan coating were applied on the Ca-alginate microcapsules. The release of HA was also confirmed by SDSPAGE and Western blotting. The results suggested the potential applications for a controlled release of HA antigen for vaccination against H5N1 influenza virus. 5.2 Recommendations for future work While my understanding of electrostatic extrusion and alginate matrix formation is rapidly increasing, there is still much work to be done for its implementations. Firstly, the in vivo function of alginate matrix must be studied for antigen delivery through alginate matrix in animals. Its effectiveness in vaccination should be compared with solo antigen injection and existed antigen delivery vehicles. 80 Chapter 5 The stability of alginate matrix should also be improved. The approaches involved alginate microcapsules made from crosslinked barium ions would be interesting. Barium alginate microcapsules were believed to be chemically and physically more stable than calcium alginate microcapsules under both in vitro and in vivo conditions. My results also reflected a very low antigen encapsulation efficiency, therefore it is very important to improve its efficiency before any implementation would be possible. Usually, higher polymer concentrations would help improve the encapsulation efficiency and minimizing the loss of antigen during microcapsules formation. Therefore, study on different polymer concentration would also be crucial. Finally, scaling-up of this technique is necessary for the manufacturing process. It would be quite challenging to improve production capacity while maintaining a controlled environment, operational discipline and rigorous quality controls. For multiple-needle operation, the breakage force must be applied carefully and uniformly to each needle, increasing the operational difficulties as the number of ports increases. 81 REFERENCES 1. Batich, C. and F. Vaghefi. Process for Microencapsulating Cells. US Patent 6,033,888. 2000. 2. Benoit, J.P., H. Marchais, H. Rolland and V. Vande Velde. Biodegradable Microspheres: Advances in Production Technology. In Microencapsulation: Methods and Industrial Applications, ed by S. Benita, pp. 35-72. New York: Marcel Dekker. 1996. 3. Boisdron-Celle, M., P. Menei and J.P. Benoit. Preparation and Characterization of 5-Fluorouracial-Loaded Microparticles as Biodegradable Anticancer Drug Carriers, J. Pharm. Pharmacol., 47, pp. 108-114. 1995. 4. Bowersock, T.L., H. HogenEsch, M. Suckow, E. Davis-Synder, D. Borie, H. Park and K. Park. Oral Administration of Mice with Ovalbumin Encapsulated in Alginate Microspheres, Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 35(2), pp. 405-406. 1994. 5. Bowersock, T.L., H. HogenEsch, M. Suckow, R.E. Porter, R. Jackson and K. Park. Oral Vaccination with Alginate Microsphere Systems, J. Controlled Release, 39, pp. 209-220. 1996. 6. Brodelius, P. and E.J. Vandamme. Immobilized Cell Systems. In Biotechnolgy: A Comprehensive Treatise in 8 Volumes, Vol. 7a, ed by H.J. Rehm and G. Reed, pp. 405-464. Weinheim: Verlag Chemie. 1987. 7. Bugarski, B., Smith, J., Wu, J., and Goosen, M.F.A. Methods for Animal Cell Immobilization Using Electrostatic Droplet Generation, Biotechnol. Tech., 7(9), pp. 677-682. 1993. 82 8. Bugarski, B., B. Amsden, R.J. Neufeld, D. Poncelet and Goosen, M.F.A. Effect of Electrode Geometry and Charge on the Production of Polymer Microbeads by Electrostatics, Can. J. Chem. Eng., 72, pp. 517-521. 1994a. 9. Bugarski, B., Q. Li, M.F.A. Goosen, D. Poncelet, R.J. Neufeld and G. Vunjak. Electrostatic Droplet Generation: Mechanism of Polymer Droplet Formation, AIChE J., 40(6), pp. 1026-1031. 1994b. 10. Centers for Disease Control and Prevention. Information about Influenza Pandemics. Fact sheet, Department of Health and Human Services, USA. March 8, 2005. 11. Centers for Disease Control and Prevention. Information about Avian Influenza (Bird Flu) and Avian Influenza A (H5N1) Virus. Key Facts, Department of Health and Human Services, USA. March 18, 2005. 12. Charwat, A.F. Generator of Droplet Tracers for Holographic Flow Visualization in Water Tunnels. Rev. Sci. Instrum., 48(8), pp. 1034-1036. 1977. 13. Chourasia, M. K. and S. K. Jain. Pharmaceutical Approaches to Colon Targeted Drug Delivery Systems, J. Pharm. Pharmaceut. Sci., 6(1), pp. 33-66. 2003. 14. Clack, A.H. and S.B. Ross-Murphy. Structure and Mechanical Properties of Biopolymer Gels, Adv. Polym. Sci., 83, pp. 57-192. 1987. 15. Cohen, S., M.C. Bañó, K.B. Visscher, M. Chow, H.R. Allcock and R. Langer. Ionically Cross-linkable Polyphosphazene: A Novel Polymer for Microencapsulation, J. Am. Chem. Soc., 112, pp. 7832-7833. 1990. 16. Coombes, A.G.A., D. Major, J.M. Wood, D.J. Hockley, P.D. Minor and S.S. Davis. Resorbable Lamellar Particles of Polylactide as Adjuvants for Influenza Virus Vaccines, Biomaterials, 19, pp. 1073-1081. 1998. 83 17. Daly, M.M. and D. Knorr. Chitosan-Alginate Complex Coarcevate Capsules: Effects of Calcium Chloride, Plasticizers, and Polyelectrolytes on Mechanical Stability, Biotech. Progr., 4(2), pp.76-80. 1988. 18. Deasy, P.B. Microencapsulation of Drugs by Pan and Air Suspension Techniques, Crit. Rev. Ther. Drug. Carrier. Syst., S(2), pp. 99-139. 1988. 19. Dorian, R.E. and K.C. Cochrum. Electrostatic Process for Manufacturing Coated Transplants and Products. US Patent 5,639,467. 1997. 20. Dulieu, C., D. Poncelet and R.J. Neufeld. Encapsulation and Immobilization Techniques. In Cell Encapsulation Technology and Therapeutics, ed by W.M. Kühtreiber, R.P. Lanza and W.L. Chick, pp. 3-17. Boston: Birkhäuser. 1999. 21. Felnerova, D., J.F. Viret, R. Glück and C. Moser. Liposomes and Virosomes as Delivery Systems for Antigens, Nucleic Acids and Drugs, Curr. Opin. Biotechnol., 15, pp. 518-529. 2004. 22. Gallagher, S.R. and J.A. Smith. Electrophoretic Separation of Proteins. In Current Protocols in Immunology, Vol. 2, ed by J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach and W. Strober, pp. 8.4.1 – 8.4.21. USA: John Wiley & Sons. 2005a. 23. Gallagher, S., S.E. Winston, S.A. Fuller and J.G.R. Hurrell. Immunoblotting and Immunodetection. In Current Protocols in Immunology, Vol. 2, ed by J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach and W. Strober, pp. 8.10.1 – 8.10.21. USA: John Wiley & Sons. 2005b. 24. Gañán-Calvo, A.M., J. Dávila and A. Barrero. Current and Droplet Size in the Electrospraying of Liquids. Scaling Laws, J. Aerosol Sci., 28(2), pp. 249-275, 1997. 84 25. Gåserød, O., O. Smidsrød and G. Skjåk-Bræk. Microcapsules of AlginateChitosan. I. A Quantitative Study of the Interaction between Alginate and Chitosan, Biomaterials, 19, pp. 1815-1825. 1998. 26. Gåserød, O., A. Sannes and G. Skjåk-Bræk. Microcapsules of AlginateChitosan. II. A Study of Capsule Stability and Permeability, Biomaterials, 20, pp. 773-783. 1999. 27. Gerbsch, N. and R. Buchholz. New Processes and Actual Trends in Biotechnology, FEMS Microbiol. Rev., 16, pp. 259-269. 1995. 28. Gombotz, W.R. and S.F. Wee. Protein Release from Alginate Matrices, Adv. Drug. Deliv. Rev., 31, pp. 267-285. 1998. 29. Goosen, M.F.A. Fundamentals of Microencapsulation. In Pancreatic Islet Transplantation, Vol. III: Immunoisolation of Pancreatic Islets, ed by R.P. Lanza and W.L. Chick, pp. 21-44. Austin: RG Landes. 1994. 30. Goosen, M.F.A. Microencapsulation of Living Cells. In Immobilised Living Cell Systems: Modelling and Experimental Methods, ed by R.G. Willaert, G.V. Baron and L. De Backer, pp. 295-322. New York: John Wiley & Sons. 1996. 31. Goosen, M.F.A., E.S.E. Mahmud, A.S. Al-Ghafri, H.A. Al-Hajri, Y.S. AlSinani and B. Bugarski. Immobilization of Cells Using Electrostatic Droplet Generation. In Methods in Biotechnology, Vol. 1: Immobilization of Enzymes and Cells, ed by G.F. Bickerstaff, pp. 167-174. Totowa: Humana Press. 1997a. 32. Goosen, M.F.A., A.S. Al-Ghafri, O. El Mardi, M.I.J. Al-Belushi, H.A. AlHajri, E.S.E. Mahmoud and E.C. Consolacion. Electrostatic Droplet Generation for Encapsulation of Somatic Tissue: Assessment of High-Voltage Power Supply, Biotechnol. Prog., 13, 497-502. 1997b. 85 33. Groboillot, A., D.K. Boadi, D. Poncelet and R.J. Neufeld. Immobilization of Cells for Application in the Food Industry, Crit. Rev. Biotechnol., 14(2), pp. 75-107. 1994. 34. Hallé, J.P., F.A. Leblond, J.F. Pariseau, P. Jutras, M.J. Brabant and Y. Lapage. Studies on Small (< 300 microns) Microcapsules: II - Parameters Governing the Production of Alginate Beads by High Voltage Electrostatic Pulses, Cell Transplant., 3(5), pp. 365-372. 1994. 35. Hilbert, A.K., U. Fritzsche and T. Kissel. Biodegradable Microspheres Containing Influenza A Vaccine: Immune Response in Mice, Vaccine, 17, pp. 1065-1073. 1999. 36. Hommel, M., A.M. Sun and M.F.A. Goosen. Droplet Generation. US Patent 4,956,128. 1990. 37. Jaworek A. and A. Krupa. Classification of the Modes of EHD Spraying, J. Aerosol Sci., 30(7), pp. 873-893. 1999. 38. Johansen, A. and J.M. Flink. A New Principle for Immobilized Yeast Reactors Based on Internal Gelation of Alginate, Biotechnol. Lett., 8(2), pp. 121-126. 1986. 39. Kas, H.S. Chitosan: Properties, Preparations and Application to Microparticulate Systems, J. Microencapsulation, 14(6), pp. 689-711. 1997. 40. Kim, S.K., J.H. Choi, E.A. Balmaceda and C.K. Rha. Chitosan. In Cell Encapsulation Technology and Therapeutics, ed by W.M. Kühtreiber, R.P. Lanza and W.L. Chick, pp. 3-17. Boston: Birkhäuser. 1999. 41. Klokk T.I. and J.E. Melvik. Controlling the Size of Alginate Gel Beads by Use of a High Electrostatic Potential. J. Microencapsulation, 19(4), pp. 415-424. 2002. 86 42. Kwok, W.W., M.J. Groves and D.J. Burgess. Sterile Microencapsulation of BCG in Alginate-Poly-L-Lysine by an Air Spraying Technique, Proc. Int. Symp. Control. Release Bioact. Mater., 16, pp. 170-171. 1989. 43. Lane, W.R. A microburette for Producing Small Liquid Drops of Known Size. Rev. Sci. Instrum., 24(4), pp. 98-101. 1947. 44. Levandowski, R., J.S. Robertson and J.M. Wood. Regulatory Issues for Influenza Vaccines. In Options for the Control of Influenza III, ed by L.E. Brown, A.W. Hampson and R.G. Webster, pp. 844-846. Amsterdam: Elsevier. 1996. 45. Levine, A.J. Viruses. pp. 155-176, New York: Scientific American Library. 1992. 46. Lim F. and A.M. Sun. Microencapsulated Islets as Bioartificial Endocrine Pancreas, Science, 210, pp. 908-910. 1980. 47. Liu, L.S., S.Q. Liu, S.Y. Ng, M. Froix, T. Ohno and J. Heller. Controlled Release of Interleukin-2 for Tumour Immunotherapy Using Alginate/Chitosan Porous Microspheres, J. Controlled Release, 43, pp. 65-74. 1997. 48. Lodge J.P. and A.P. Monaco. Artificial Organs. In Implantation Biology, the Host Response and Biomedical Devices, ed by R.S. Greco, pp. 347-361. Boca Raton: CRC Press. 1994. 49. MacGregor, E.A. and C.T. Greenwood. Polymers in Nature. pp. 240-329. New York: Wiley. 1980. 50. Martinsen, A., G. Skjåk-Bræk and O. Smidsrød. Alginate as Immobilization Material: I. Correlation between Chemical and Physical Properties of Alginate Gel Beads, Biotechnol. Bioeng., 33, pp. 79-89. 1989. 87 51. McKnight, C.A., M.F.A. Goosen, D. Sun and C. Penney. Synthesis of Chitosan-Alginate Microcapsule Membranes, J. Bioact. Comp. Polym., 3, pp. 334-354. 1988. 52. Moldoveanu, Z., M. Novak, W. Huang, R.M. Gilley, J.K. Staas, D. Schafer, R.W. Compans and J. Mestecky. Oral Immunization with Influenza Virus in Biodegradable Microspheres, J. Infect. Dis., 167(1), pp. 84-90. 1993. 53. Morris, W., C.M. Steinhoff and P.K. Russell. Potential of Polymer Microencapsulation Technology for Vaccine Innovation, Vaccine, 12(1), pp. 511. 1994. 54. Müller, B.W., J. Bleich and B. Wagenaar. Microparticle Production without Organic Solvent. In Proc. 9th International Symposium on Microencapsulation, September 1993, Ankara, Turkey, pp. 29-40. 55. Murata, Y., T. Maeda, E. Miyamoto and S. Kawashima. Preparation of Chitosan-Reinforced Alginate Gel Beads—Effects of Chitosan on Gel Matrix Erosion, Int. J. Pharm., 96, pp. 139-145. 1993. 56. Nawab, M.A. and S.G. Mason, S.G. The Preparation of Uniform Emulsion by Electrical Dispersion, J. Colloid Sci., 13, pp. 179-187. 1958. 57. Nedovic, V.A., B. Obradovic, I. Leskosek-Cukalovic, O. Trifunovic, R. Pesic and B. Bugarski. Electrostatic Generation of Alginate Microbeads Loaded with Brewing Yeast, Process Biochem., 37, pp. 17-22. 2001. 58. Nielsen, F. Spray Drying Pharmaceuticals, Manuf. Chem., 53(7), pp. 38-39. 1982. 59. Orive, G., R.M. Hernandez, A.R. Gascon, R. Calafiore, T.M.S. Chang, P. de Vos, G. Hortelano, D. Hunkeler, I. Lacik and J.L. Pedraz. History, Challenges 88 and Perspectives of Cell Microencapsulation, Trends Biotechnol., 22(2), pp. 87-92. 2004. 60. Polk, A.E., B. Amsden, D.J. Scarratt, A. Gonzal, A. O. Okhamafe and M.F.A. Goosen. Oral Delivery in Aquaculture: Controlled Release of Proteins from Chitosan-Alginate Microcapsules, Aquacult. Eng., 13, pp. 311-323. 1994. 61. Poncelet, D., R. Lencki, J.P. Beaulieu, R.J. Neufeld and A. Fournier. Production of Alginate Beads by Emulsification/Internal Gelation. I. Methodology, Appl. Microbiol. Biotechnol., 38, pp. 39-45. 1992. 62. Poncelet, D., B. Poncelet DeSmet, C. Beaulieu and R.J. Neufeld. Scale-up of Gel Bed and Microcapsule Production in Cell Immobilization. In Fundamentals of Animal Cell Encapsulation and Immobilization, ed by M.F.A. Goosen, pp. 113-142. Boca Ration: CRC Press. 1993. 63. Poncelet D., B. Burgarski, B. Amsdem, J. Zhu, R.J. Neufeld and M.F.A. Goosen. A Parallel Plate Electrostatic Droplet Generator: Parameters Affecting Microbead Size, Appl. Microbiol. Biotechnol., 42, pp. 251-255. 1994. 64. Poncelet, D. and R.J. Neufeld. Fundamentals of Dispersion in Encapsulation Technology. In Immobilized Cells: Basics and Applications, ed by R.H. Wijffels, R.M. Buitelaar, C. Bucke and J. Tramper, pp. 47-54. Amsterdam: Elsevier. 1996. 65. Poncelet, D., R.J. Neufeld, M.F.A. Goosen, B. Burgarski and V. Babak. Formation of Microgel Beads by Electric Dispersion of Polymer Solutions, AIChE J., 45(9), pp. 2018-2023. 1999a. 66. Poncelet, D., V.G. Babak, R.J. Neufeld, M.F.A. Goosen and B. Bugarski. Theory of Electrostatic Dispersion of Polymer Solutions in the Production of 89 Microgel Beads Containing Biocatalyst, Adv. Colloid Interface Sci., 79, pp. 213-228. 1999b. 67. Poncelet, D., C. Dulieu and M. Jacquot. Description of the Immobilisation Procedures. In Immobilized Cells, ed by R.H. Wijffels, pp. 15-30. New York: Springer-Verlag. 2001. 68. Potter, C.W. Influenza Virus. In Handbook of Neurovirology, ed by R.R McKendall and W.G. Stroop, pp. 575-581. New York: Marcel Dekker. 1994. 69. Potter, C.W. Influenza. In Principles and Practice of Clinical Virology, ed by A.J. Zuckerman, J.E. Banatvala and J.R. Pattison, pp. 253-278. New York: John Wiley & Sons. 2000. 70. Prusse, U., F. Bruske, J. Breford and K.D. Vorlop. Encapsulation by the Jet Cutting Method: New Developments. In Proc. International Workshop on Bioencapsulation VI, August 29 - September 1, 1997, Barcelona, Spain, P6, S. 1-4. 71. Rayleight, J.W.S. On the Instability of Jet, Proc. London Math. Soc., 10, pp. 413. 1878. 72. Rees, D.A. and E.J. Welsh. Secondary and Tertiary Structure of Polysaccharides in Solution and Gels, Angew. Chem. Int. Ed. Engl., 16, pp. 214-224. 1977. 73. Rees, D.A. Polysaccharide Shapes and Their Interactions—Some Recent Advances, Pure Appl. Chem., 53, pp. 1-14. 1981. 74. Sample, S.B. and R.J. Bollini. Production of Liquid Aerosols by Harmonic Electrical Spraying, J. Coll. and Interface Sci., 41(2), pp. 185-193. 1972. 75. Sasse, J. and S.R. Gallagher. Detection of Proteins. In Current Protocols in Immunology, Vol. 2, ed by J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, 90 E.M. Shevach and W. Strober, pp. 8.9.1 – 8.9.25. USA: John Wiley & Sons. 2005. 76. Silver, F. and C. Doillon. Biocompatibility: Interactions of Biological and Implantable Materials, Vol. 1. Polymer. pp. 261-297, New York: VCH Publishers. 1989. 77. Singh, M. and D. O’Hagan. The Preparation and Characterization of Polymeric Antigen Delivery Systems for Oral Administration, Adv. Drug. Deliv. Rev., 34, pp. 285-304. 1998. 78. Smidsrød, O. and G. Skjåk-Bræk. Alginate as Immobilization Matrix for Cells, Trends Biotechnol., 8, pp. 71-78. 1990. 79. Speranza, A., M. Ghadiri, M. Newman, L.S. Osseo and G. Ferrari. ElectroSpraying of a Highly Conductive and Viscous Liquid, J. Electrostatics, 51-52, pp. 494-501. 2001. 80. Stannard, L.M. Department of Medical Microbioloy. University of Cape Town. 1995. http://web.uct.ac.za/depts/mmi/stannard/linda.html. 81. Straus, S.E. Influenza and its Virus. In Mechanisms of Microbial Disease, ed by M. Schaechter, G. Medoff and B.I. Eisenstein, pp. 445-458. Baltimore: Williams & Wilkins. 1993. 82. Sutherland, I.W. Alginates. In Biomaterials: Novel Materials from Biological Sources, ed by D. Byron, pp. 309-331. Basingstoke: Macmillan. 1991. 83. Tam, J.S. Influenza A (H5N1) in Hong Kong: An Overview, Vaccine, 20, pp. S77-S81. 2002. 84. Tang, D., M. DeVit and S. Johnston. Genetic Immunization is a Simple Method for Eliciting an Immune Response, Nature, 356, pp. 152-154. 1992. 91 85. Taqieddin, E. and M. Amiji. Enzyme Immobilization in Novel Alginate– Chitosan Core-Shell Microcapsules, Biomaterials, 25, pp. 1937–1945. 2004. 86. Taylor. G. Disintegration of Water Drops in an Electric Field, Proc. R. Soc. London, Ser. A, 280(1381), pp. 383-397. 1964. 87. Treanor, J.J. Influenza A and B Viruses. In Viral Infections of the Respiratory Tract, ed by R. Dolin and P.F. Wright, pp. 105-160. New York: Marcel Dekker. 1999. 88. Watanabe, H., T. Matsuyama and H. Yamamoto. Experimental Study on Electrostatic Atomization of Highly Viscous Liquids, J. Electrostatics, 57, pp. 183-197. 2003. 89. Wee. S.F., W.R. Gombotz and W. Fanslow. Evaluation of Alginate Microbeads for Intranasal Delivery of Ovalbumin, Proc. Int. Symp. Control. Release Bioact. Mater., 22, pp. 566-567. 1995. 90. Wood, J.M., A.G.A. Coombes, D. Major, P.D. Minor and S.S. Davis. Polylactide Microparticles as Adjuvants for Parenteral Delivery of Influenza Virus Vaccines. In Options for the Control of Influenza III, ed by L.E. Brown, A.W. Hampson and R.G. Webster, pp. 810-814. Amsterdam: Elsevier. 1996. 91. World Health Organization. Influenza Vaccines, Wkly. Epidemiol. Rec., 77(28), pp. 229-240. 2002. 92. World Health Organization. Influenza. Fact sheet No. 211, World Health Organization. March, 2003. 93. World Health Organization. Avian Influenza: Avian Influenza ("Bird Flu") and the Significance of its Transmission to Humans. Fact sheet, World Health Organization. January 15, 2004. 92 94. Yang, Y.Y., T.S. Chun, X.L. Bai and W.K. Chan. Effect of preparation conditions on morphology and release profiles of biodegradable polymeric microspheres containing protein fabricated by double-emulsion method, Chem. Eng. Sci., 55, pp. 2223-2236. 2000. 95. Zimmermann, U., C. Hasse, M. Rothmund and W. Kühtreiber. Biocompatible Encapsulation Materials: Fundamentals and Application. In Cell Encapsulation Technology and Therapeutics, ed by W.M. Kühtreiber, R.P. Lanza and W.L. Chick, pp. 40-52. Boston: Birkhäuser. 1999. 93 [...]... cycle of influenza virus (Levine, 1992) Influenza pandemic and its impact Influenza viruses can change in two different ways — Antigenic drift, the HA and NA proteins of the influenza virus can undergo minor changes or mutations, leading to new strains It happens continually over time, and 7 Chapter 2 produces new virus strains that may not be recognized by antibodies to earlier influenza strains — Antigenic... shift, it is an abrupt, major change in the influenza A viruses, resulting in a new influenza virus that can infect humans and has a new HA or HA and NA protein combination that has not been seen in humans for many years Antigenic shift results in a new influenza A subtype If a new subtype of influenza A virus is introduced into the human population, if most people have little or no protection against the... existing vaccines effective against currently circulating human influenza strains, can reduce the likelihood of coinfection of humans with avian and influenza strains, and thus reduce the risk that genes will be exchanged Workers involved in the culling of poultry flocks must be protected, by proper clothing and equipment, against infection These workers should also receive antiviral drugs as a prophylactic... Differences between influenza A, B, and C viruses (Treanor, 1999) Influenza A Influenza B Influenza C Genetics 8 gene segments 8 gene segments 7 gene segments Structure 10 viral proteins 11 viral proteins 9 viral proteins M2 protein unique NB protein unique HEF protein unique Host range Humans, swine, equine, avian, marine mammals Humans only Humans and swine Epidemiology Antigenic shift and drift; Drift... Chapter 2 Vaccination is the principal measure for preventing influenza and reducing the impact of epidemics Various types of influenza vaccines have been available and used for more than 60 years They are safe and effective in preventing both mild and severe outcomes of influenza (WHO, 2002) The major treatment for influenza infections are the time-proven ones involving hydration, rest, and antipyretics,... direct infection of humans If more humans become infected over time, the likelihood also increases that humans, if concurrently infected with human and avian influenza strains, could serve as the “mixing vessel” for the emergence of a novel subtype with sufficient human genes to be easily transmitted from person to person Such an event would mark the start of an influenza pandemic 2.1.4 Prevention and... acetaminophen rather than aspirin (Straus, 1993) Antiviral drugs for influenza are an important adjunct to influenza vaccine for the treatment and prevention of influenza When taken before infection or during early stage of the disease (within two days of illness onset), antivirals may help prevent infection, and if infection has already taken hold, their early administration may reduce the duration of. .. combined action of the antigen and the adjuvant (Felnerova et al., 2004) Many vaccine delivery systems have been proposed to provide sustained release and/or to increase the immune response One approach to the development of delivery system for vaccines involves the use of polymer encapsulation of antigens (Singh and O’Hagan, 1998) The polymeric microcapsules that protect antigen from acidic and enzymatic...Abbreviations BCA Bicinchoninic acid BSA Bovine serum albumin CDC Centers for Disease Control and Prevention DNA Deoxyribonucleic acid FITC-BSA Bovine serum albumin-fluorescein isothiocyanate H5N1 Hemagglutinin type 5 & Neuraminidase type 1 HA Hemagglutinin HPAI Highly pathogenic avian influenza M1 Matrix protein 1 M2 Matrix protein 2 NA Neuraminidase NP Nucleoprotein PLG Polylactide-co-glycolides... surface of the particle contains three kinds of spike proteins: hemagglutinin (HA), neuraminidase (NA), and matrix protein (M2, only on type A) 5 Chapter 2 Hemagglutinin (HA) Neuraminidase (NA) Lipid membrane Matrix Protein (M1) M2 Protein (only on type A) Ribonucleoprotein (RNP): - vRNA - Polymerase - Nucleoprotein Figure 2.2 A schematic drawing of influenza virus structure Further classification of type ... poultry, using existing vaccines effective against currently circulating human influenza strains, can reduce the likelihood of coinfection of humans with avian and influenza strains, and thus reduce... cycle of influenza virus (Levine, 1992) Influenza pandemic and its impact Influenza viruses can change in two different ways — Antigenic drift, the HA and NA proteins of the influenza virus can... proteins M2 protein unique NB protein unique HEF protein unique Host range Humans, swine, equine, avian, marine mammals Humans only Humans and swine Epidemiology Antigenic shift and drift; Drift