ABSTRACT Title of dissertation: MICROENCAPSULATION OF PROBIOTIC BACTERIA IN XANTHAN-CHITOSAN POLYELECTROLYTE COMPLEX GELS Sanem Argin, Doctor of Philosophy, 2007 Dissertation directed by: Professor Y. Martin Lo Department of Nutrition and Food Science In recent years, increasing evidence indicating numerous health beneflts asso- ciated with the intake of probiotic bacteria has created a big market of probiotic foods worldwide. However, maintaining high numbers of viable cells in probiotic food products during the shelf life of the product and during gastrointestinal transit is a challenge. The goal of this research is to develop a novel microencapsulation system using xanthan gum and chitosan polyelectrolyte complex gels in order to protect the probiotic cells against adverse environmental conditions, and to increase their recovery rates. The extrusion method was used to form the xanthan-chitosan microcapsules. The efiects of initial polymer concentration and chitosan solution pH on the crosslink density of the capsule network were investigated by swelling studies and modulated difierential scanning calorimetry (MDSC) analysis. Once the capsule formulations resulting in a highly crosslinked network structure were determined, P. acidilactici cells were successfully encapsulated using these formulations. Efiorts were made to study the release kinetics of probiotic cells from these capsules in gastrointestinal conditions. Cell release was found to be negligible in simulated gastric juice. In simulated intestinal conditions, the release was relaxation controlled and complete cell release was achieved in at least 5 hours. After exposure to simulated gastric uid (pH=2.0), encapsulation with xanthan gum and chitosan provided up to six-log and four-log preservation of the freeze-dried probiotic cells over free suspending cells for 1 and 2 hours, respectively. These results suggest that xanthan-chitosan capsules have a good potential for delivery of probiotic cells to the intestines in high numbers where the cells can release and colonize to beneflt the consumer. MICROENCAPSULATION OF PROBIOTIC BACTERIA IN XANTHAN-CHITOSAN POLYELECTROLYTE COMPLEX GELS by Sanem Arg?n Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park in partial fulflllment of the requirements for the degree of Doctor of Philosophy 2007 Advisory Committee: Professor Y. Martin Lo, Chair/Advisor Professor Peter Koflnas, Co-Advisor Professor Srinivasa R. Raghavan Professor Mark Kantor Professor Mickey Parish xa9 Copyright by Sanem Arg?n 2007 DEDICATION To my daughter, Kayla Soysal, and my husband, Alkan Soysal. ii Acknowledgments First and foremost, I would like to thank my husband, Alkan Soysal, for his endless support, unconditional love, continuous encouragement, and tremendous patience. Without him, I would never have been able to come this far. He has been my coach, my best friend, and the shoulder I cried on throughout these years. Thank you for believing in me and for always being there for me. Thank you for flnding a solution to every problem I had and not complaining about it even once. Many, many thanks for the millions of small things you have done to make it possible for me to write this dissertation. I would like to thank my advisor, Dr. Y. Martin Lo, for giving me the oppor- tunity to pursue my Ph.D. in his laboratory and for his support during my studies. From my work with him I learned more about myself and what I am capable of achieving. I am extremely thankful to my co-advisor, Dr. Peter Koflnas, for ofiering his help at the time that I needed it the most and for welcoming me into his group. His guidance and support greatly impacted the progress of my research. Dr. Koflna sas eyxaristw poly gia ola. I would also like to thank Dr. Srinivasa Raghavan for not only being in my dissertation committee but also for everything I learned from him about polymers and for the valuable discussions he had with me about my research. I am also thankful to the other members of my former and flnal examination committee, Dr. Mickey Parish, Dr. Mark Kantor, Dr. Dallas Hoover, Dr. Inder Vijay, and Dr. iii Yang Tao, for their time and suggestions. I would like to thank my parents, F?usun and Ibrahim Arg?n, and my brother, Erman Arg?n, for their love and support throughout my life. I also owe thanks to all my friends in Dr. Lo?s and Dr. Koflnas? groups for their friendship, support, and for all the fun they brought into my life: Linden Bolisay, Brendan Casey, Arthur von Wald Cresce, Angela Fu, Ayan Ghosh, Daniel Janiak, Peter Machado, Daniel Reese, Karen Silagyi, Josh Silverstein, Pavan Kumar Soma, Patrick Williams, Ta-I Yang, and Afra Yeh. Finally, my special thanks go to my 10-week old daughter Kayla for being the biggest joy of my life and for giving me the motivation to flnish my work. iv TABLE OF CONTENTS List of Tables vii List of Figures viii 1 Introduction 1 1.1 Signiflcance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 Literature Review 3 2.1 Microencapsulation of probiotic bacteria . . . . . . . . . . . . . 3 2.2 Xanthan Gum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2.1 Source, structure and applications . . . . . . . . . . . . . 6 2.2.2 Rheological properties . . . . . . . . . . . . . . . . . . . . 8 2.3 Chitosan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.3.1 Source, structure and applications . . . . . . . . . . . . . 9 2.3.2 Rheological properties . . . . . . . . . . . . . . . . . . . . 11 2.4 Microencapsulation with xanthan gum and chitosan . . . . . . 12 2.4.1 Polyelectrolyte complex (PEC) gels by xanthan gum and chitosan . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2.4.2 pH-sensitiveswellingcharacteristicsofxanthan-chitosan gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 3 Efiects of Complexation Conditions on Xanthan-Chitosan Polyelec- trolyte Complex Gels 17 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2 Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . 18 3.2.1 Preparation of chitosan and xanthan solutions . . . . . 18 3.2.2 Capsule formation and Inverted microscopy . . . . . . . 19 3.2.3 Determination of the swelling degree . . . . . . . . . . . 19 3.2.4 Difierential Scanning Calorimetry (DSC) measurements 20 3.2.5 Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . 21 3.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3.1 Capsule formation . . . . . . . . . . . . . . . . . . . . . . . 21 v 3.3.2 Efiect of complexation conditions on the swelling de- gree of xanthan-chitosan capsules . . . . . . . . . . . . . 22 3.3.3 DSC analysis of xanthan-chitosan capsules . . . . . . . . 26 3.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4 The Release Kinetics and Swelling Behavior of Xanthan-Chitosan Capsules 33 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . 34 4.2.1 Calibration curve . . . . . . . . . . . . . . . . . . . . . . . 34 4.2.2 Microencapsulation of P. acidilactici . . . . . . . . . . . 35 4.2.3 Kinetics of cell release . . . . . . . . . . . . . . . . . . . . 35 4.2.4 Dynamic swelling behavior . . . . . . . . . . . . . . . . . 37 4.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 37 4.3.1 Calibration curve . . . . . . . . . . . . . . . . . . . . . . . 37 4.3.2 Kinetics of cell release . . . . . . . . . . . . . . . . . . . . 38 4.3.3 Dynamic swelling behavior . . . . . . . . . . . . . . . . . 43 4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 5 Protective efiects of Xanthan-Chitosan Encapsulation on P. acidi- lactici cells 47 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 5.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . 48 5.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 49 5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 6 Conclusions 56 Bibliography 58 vi LIST OF TABLES 2.1 Microorganisms considered as probiotics. . . . . . . . . . . . . . . . . 4 3.1 Swelling degrees of xanthan-chitosan capsules in DI water . . . . . . . 24 4.1 Analysis of cell release data from xanthan-chitosan capsules . . . . . 41 4.2 Analysis of dynamic swelling data for xanthan-chitosan capsules . . . 45 vii LIST OF FIGURES 2.1 Pentasaccharide repeating unit of xanthan gum. . . . . . . . . . . . . 7 2.2 Chemical structure of chitosan. . . . . . . . . . . . . . . . . . . . . . 10 2.3 pH-sensitive swelling of xanthan-chitosan complex . . . . . . . . . . . 16 3.1 Xanthan-chitosancapsulesformedwheninitialxanthanconcentration is (a) greater than 1.5% (w/v); (b) less than 1.5% (w/v); (c) Inverted microscopy image of a 2 mm-diameter xanthan-chitosan capsule pre- pared with uorescently labeled chitosan. . . . . . . . . . . . . . . . . 22 3.2 Swelling degree of xanthan-chitosan capsules formed by using 0.5% (w/v) and 0.7% (w/v) xanthan solutions. . . . . . . . . . . . . . . . . 23 3.3 MDSC curves of (a) chitosan; (b) xanthan gum. . . . . . . . . . . . . 28 3.4 MDSC curves of xanthan-chitosan capsules showing the efiect of ini- tial xanthan concentration on the resulting capsule network structure. 29 3.5 MDSC curves of xanthan-chitosan capsules showing the efiect of ini- tial chitosan solution pH on the resulting capsule network structure. . 31 4.1 Calibration curve of P. acidilactici. . . . . . . . . . . . . . . . . . . . 38 4.2 The release of P. acidilactici cells from xanthan- chitosan capsules. . 40 4.3 Fractional release of P. acidilactici cells from xanthan-chitosan cap- sules in SGF at pH=2.0 and in SIF at pH 6.8. . . . . . . . . . . . . . 42 4.4 (a)Dynamicswelling; (b)waterabsorptioncurveofxanthan-chitosan capsules in SGF at pH=2.0 and in SIF at pH 6.8. . . . . . . . . . . . 44 viii 5.1 Protective efiects of encapsulation on the viability of P.acidilactici cells against freeze-drying and SGF (pH=2.0) exposure (n=2) using (a) 0.7% and 1.0% chitosan. . . . . . . . . . . . . . . . . . . . . . . . 51 5.2 Relative size of capsules prepared by using a syringe and a nozzle. . . 52 5.3 Protective efiect of xanthan-chitosan capsules prepared by using a 0.7-mm nozzle on the viability of P. acidilactici cells against freeze- drying and SGF exposure for 1 hr. . . . . . . . . . . . . . . . . . . . 53 ix Chapter 1 Introduction 1.1 Signiflcance Life style and eating habits play an important role in overall health of in- dividuals. In recent years, increasing evidence indicating numerous health beneflts associated with the intake of probiotic bacteria has created a big market of probiotic foods worldwide [1]. The biggest challenge in the development of probiotic prod- ucts is to maintain the adequate number of viable cells during the shelf life of the product as well as during the gastrointestinal (GI)-tract transit after consumption, so that the claimed health beneflts can be delivered to the consumer [2{5]. Con- sequently, there has been a growing interest in developing techniques to enhance the survival of probiotic bacteria particularly during the GI-tract transit of the cells [6{14]. Microencapsulation of probiotic bacteria in hydrocolloid beads is one of the recent techniques studied to improve the viability and activity of the cells under unfavorable conditions by entrapping the bacteria within a bead matrix [15{18]. 1 Na-Alginate and ?-carrageenan, alone or combined with other polymers and cry- oprotectants, are the most commonly used hydrocolloids for the encapsulation of microorganisms [17,19{21]. However, each system has its own limitations such as susceptibility to ions, lack of mechanical strength and scale-up di?culties [22,23]. For this reason, there remains an important need for an efiective capsule design that ensures the proper delivery of probiotics to the target destination and enables the complete release of the cells at desired conditions to beneflt from the health promoting efiects of probiotic bacteria. 1.2 Objectives The ultimate goal of this work was to develop a novel microencapsulation system using polyelectrolyte complex gels formed by xanthan gum and chitosan to efiectively sustain the viability of probiotics that are known to provide health beneflts to human and animals. The purpose was to protect the encapsulated cells against harsh environmental conditions, particularly against the highly acidic gastric conditions, and to establish a release mechanism so that the bacteria can colonize in the gut to improve the intestinal microbiota. Speciflcally, emphasis was placed on the efiects of complexation conditions on the capsule network structure, the release kinetics of the cells under GI-tract conditions, and the protective efiects of the xanthan-chitosan capsules on probiotics during GI-tract transit. 2 Chapter 2 Literature Review 2.1 Microencapsulation of probiotic bacteria Probiotics are deflned as live microbial food supplements that beneflt the host by improving its intestinal microbial balance [24]. Lactic acid bacteria are among the most important probiotic microorganisms (Table 2.1) associated with the human gastrointestinal (GI)-tract [25]. Known health beneflts of probiotic bacteria include suppressing the growth of undesirable microorganisms in the colon and small in- testine [26,27], controlling serum cholesterol levels [28], reducing the risk of colon cancer [29], stimulating the immune system [30], improving lactose utilization [31], and controlling the allergic in ammation associated with food allergy [32]. To act as probiotics, bacteria must arrive in intestines alive and in su?cient numbers which is suggested at 106-107 cfu/g product [26,33]. However, signifl- cant reduction of the number of viable and active cells in a food product occurs inevitably due to the environmental changes such as pH, temperature, and dis- 3 Table 2.1: Microorganisms considered as probiotics. Lactobacillus species Bifldobacterium species Other lactic acid bacteria Non-lactic acid bacteria L. acidophilus B. adolescentis Enterococcus faecalis? Bacillus cereus var. toyoi? L. amylovorus B. animalis/B. lactis Enterococcus faecium Escherichia coli strain nissle L. casei B. bifldum Lactococcus lactis Saccharomyces boulardii L. crispatus B. breve Leuconostoc mesenteroids Saccharomyces cerevisiae L. rhamnosus B. infantis Pediococcus acidilactici? Propionibacterium L. gallinarum? B. longum Sporolactobacillus inulinus? freudenreichii? L. gasseri Streptococcus thermophilus L. johnsonii L. paracasei L. plantarum L. reuteri L. delbrueckii subsp. bulgaricus ? Main application for animals Adopted from [25] solved oxygen [2{5]. Freeze drying of the probiotic cells during initial manufactur- ing [16,34]; the processing conditions of food carriers, especially the low pH and low temperature in fermented and frozen products; as well as the transportation and storage requirements [14,35,36] are some of the main reasons causing the loss of cell viability. Moreover, after their consumption, probiotic cells are subjected to various hydrolytic enzymes, acidic conditions of the stomach, and bile salts in the GI-tract [37,38]. Various approaches such as addition of difierent growth promot- ers [6,7,9,12,13], manipulation of fermentation and storage conditions of the food carriers [8,10,11,14], careful selection of the culture organisms according to their interrelationships and acid-bile resistance [14], and recently, using microencapsula- tion technology to reduce the cell loss [15{18] have been investigated to improve the survival rates of active cells arriving in the intestines. Microencapsulation of probiotics in hydrocolloid beads has been found to improve the viability and activ- ity of probiotics in food products and the intestinal tract by entrapping the cells within a bead matrix, thus, segregating them from adverse environmental condi- 4 tions, as well as protecting them against bacteriophages [21,33,39]. There are two common techniques applied to the microencapsulation of probiotic bacteria: The extrusion (droplet) method [40{43] and emulsion (two-phase) system [16,21,42,44]. Both systems achieve survival rates as high as 80-95%, but the extrusion method is simpler and cheaper [33]. Alginate is the most commonly used hydrocolloid for the entrapment of cells due to its simple and low cost gelling mechanism [19], excellent biocompatibility [20], and the reversibility of the immobilization [21]. However, al- ginate gels are unstable with phosphate and lactate ions as these ions can displace the calcium ions that stabilize the alginate gels [44]. Recent studies showed that Ca-alginate encapsulation does not signiflcantly improve the survival of the cells in gastric conditions [15,38,45]. On the other hand, ?-carrageenan capsules have been reported to provide better protection against refrigeration temperatures than Ca- alginate beads [17]. However, with ?-carrageenan, exible gels can only be formed in the presence of locust bean gum [44]. The formation of ?-carrageenan-locust bean gum capsules requires potassium ions, which can damage the probiotic cells [46]. Another approach was to use Ca-pectate gels as carriers since they are less suscep- tible to decalcifying and acidiflcation than Ca-alginate gels [47]. Combining other polymers (starch, pectin, etc) and/or whey proteins with these common hydrocol- loids has been shown to ofier better protection of cells under low temperature and pH than when these gums are used alone for the encapsulation [41,48,49]. More- over, difierent coating materials to coat the beads as well as some additives such as cryoprotectants were used to overcome the aforementioned disadvantages [39,50]. However, all of these approaches have had varying degrees of success and the costs 5 of operation remain a concern. The main problems of bacteria encapsulation with Ca-alginate and ?-carrageenan gels can be summarized as susceptibility to ions, lack of mechanical strength and scale-up di?culties [22,23]. Some other methods to encapsulate bacteria such as spray drying with starch reportedly did not ofier any protection against acidic conditions [15]. Sun and Gri?ths [51] reported that a mixture of gellan and xanthan gum, both of which are microbial exopolysaccha- rides, serves as a better encapsulating agent than Ca-alginate when the cells are exposed to the same environmental conditions. Produced by microorganisms and secreted to the environment to serve as a protection against desiccation, bacterio- phage attacks, variations in temperature, cell wall degrading enzymes, and osmotic stress [52{56], microbial exopolysaccharides may be suggested as promising agents for encapsulation of probiotic bacteria. 2.2 Xanthan Gum 2.2.1 Source, structure and applications Xanthan gum (Figure 2.1) is a microbial exopolysaccharide consisting of a cellulosic backbone with side chains of two mannose and one glucuronic acid on every second glucose residue [57,58]. In nature, the plant pathogenic bacterium Xanthomonas campestris produce highly viscous xanthan gum which attaches to cabbage and protects itself against dehydration and access of harmful substances, or as a way of binding and neutralizing bacteriophages [52,54]. Presently, commercial 6 Figure 2.1: Pentasaccharide repeating unit of xanthan gum. xanthan gum is produced from glucose by batch fermentation of X. campestris. The molecular weight of xanthan gum can reach up to 6 million Daltons, which makes it possible to create extremely viscous solutions at very low concentrations [59,60]. Pyruvate and acetate substitution may vary strongly depending on the par- ent bacterial strain and the growth conditions of the bacteria [61]. Side chains of xanthan gum represent a very high proportion of the molecule (60%). Due to the side chains, the polymer completely hydrates in water [62]. In addition, enzy- matic resistance of xanthan gum is thought to be due to the arrangement of the side chains which prevents the enzymes from attacking the fl-(1-4) linkages in the backbone, thereby preventing depolymerization by enzymes, acid and alkali [63]. In aqueous solution, the stable conformation of xanthan is double helical below a transition temperature [64]. Solution studies on the conformation of xanthan gum suggest a rod-like structure with some degree of exibility, depending on the molec- ular weight [65{67]. In addition to its enzymatic resistance, xanthan gum is stable 7 over a wide range of temperatures and pH, which flnds many applications in food, pharmaceutical, cosmetic, and oil-drilling industries [68{72]. Xanthan gum was ap- proved by the U.S. Food and Drug Administration (FDA) in 1965 and became the most widely used thickening agent in food products. 2.2.2 Rheological properties In aqueous solution, high molecular weight xanthan molecules form a semi- exible wormlike structure and develop a weak-gel network due to intermolecular interactions [66,67,73,74]. These intermolecular associations such as hydrogen bond- ing, electrostatic and hydrophobic interactions play a crucial role in the changes of rheological properties of the xanthan solutions [74,75]. Since the junctions between the ordered chain sequences in the gel network of xanthan gum are weaker than those in the true gel networks, they can be easily broken down under stress allowing the system to ow which gives xanthan the shear-thinning behavior [76]. Due to the presence of glucuronic acid and pyruvate in its side chains, xanthan is considered to be an anionic polyelectrolyte [77]. However, the response of xanthan solutions to increasing ionic strength difiers from other polyelectrolytes in that they maintain a high solution viscosity as ionic strength increases due to the increase in hydrody- namic radius [78,79]. At low salt concentrations (up to 0.05 M), the viscosity of the xanthan gum solutions slightly decreases and beyond this, no signiflcant efiect on the steady shear and dynamic properties is observed [79,80]. In addition, xanthan viscosity does not change signiflcantly with temperature in the presence of salt and 8 the ordered conformation exists at room or elevated temperatures [77,78,81]. 2.3 Chitosan 2.3.1 Source, structure and applications Chitosan (Figure 2.2), poly-fl-(1 4)-D-glucosamine, is the only natural cationic polysaccharide and is produced by alkaline deacetylation of chitin [82,83]. Chitin is the second most abundant natural polysaccharide which is present in crustacea, insects and yeasts. The molecular weight of chitin and chitosan ranges between severalhundredthousandsto1millionDaltons[84]. Asoneofthemajorby-products of the crabbing and shrimp canning industry, chitin is also the major source of surface pollution in coastal areas [85]. Studies on chitin and chitosan have increased since the 1990?s to flnd value-added uses of these polysaccharides that show excellent biological properties such as biocompatibility, biodegradability, lack of toxicity, and adsorption, as well as relatively high percentage of nitrogen [82,86{88]. The degree of acetylation (DA) is a very important property of chitosan molecules since DA afiects biodegradation capability, solubility, gelling and reactivity of chitosan [89]. The unique properties of chitosan arise from its amino groups that carry pos- itive charges at pH values below 6.5, enabling it to bind to negatively charged materials such as enzymes, cells, polysaccharides, nucleic acids, hair and skin [90]. Being a linear polysaccharide, chitosan has both reactive amino groups and hydroxyl groups that can be used to alter its physical and solution properties [90]. Chitosan 9 Figure 2.2: Chemical structure of chitosan. is insoluble at alkaline and neutral pH but soluble in inorganic and organic acids such as hydrochloric acid, acetic acid, lactic acid and glutamic acid. Water solu- ble chitosan can be formed when hydrogen-bond formation is prevented by partial re-acetylation of chitosan molecules by several means [87,91]. Chitin and chitosan are being extensively used in the pharmaceutical industry (cosmetics, contact lenses, artiflcial skin, wound dressing), paper making, photogra- phy, solid state batteries, waste water treatment, chromatography, dietary supple- ments and animal feed [83,92]. Considerable amount of research has been conducted to use chitosan as a drug delivery vehicle, especially for the treatment of colon dis- eases, such as ulcerative colitis and Crohn?s disease and as a dietary supplement for lowering cholesterol and controlling overweight [84,87,89,93]. Chitosan for oral administration to humans is generally recognized as safe and approved for food use in Italy, Norway, and Japan [94,95]. In the U.S., chitosan is yet to be approved as a food additive by the U.S. Food and Drug Administration (FDA) but was approved in 1985 by Association of American Feed Control O?cials (AAFCO) for its use in 10 animal feed, as long as the level does not exceed 0.1% of the feed [96,97]. 2.3.2 Rheological properties Concentrated chitosan solutions exhibit shear-thinning behavior and the vis- cosity of dilute chitosan solutions are independent of shear rate [98]. Due to strong intermolecular hydrogen bonding even at low concentration, chitosan molecules have a tendency to entangle and form a network [99]. As a polyelectrolyte, the confor- mation of the chitosan molecule is afiected by the electric charge density [100]. For this reason, the rheological properties of chitosan depend on factors such as pH, ionic strength, solvent selection, concentration, molecular weight, DA, and the dis- tribution of the acetyl groups [101]. When the pH of the solution is increased, the intermolecular and intramolecular electrostatic repulsions between cationic charges are reduced and chitosan chains come closer together, lowering the hydrodynamic volume of the chitosan molecules. Such an efiect may enhance the interchain and intrachain hydrogen bonding. Similarly, as the ionic strength increases, the intrinsic viscosity decreases due to shielding efiect of counterions [85]. The apparent vis- cosities of chitosan solutions were found to be difierent even at the same pH when difierent types of acids were used to adjust the pH of the solution. The apparent viscosity of chitosan was higher when dissolved in organic acids than it was when hydrochloric acid was used. This can be attributed to the overall efiect of screening and steric efiects exerted by difierent anions on the conformation and chain stifiness of the chitosan molecules [98,102]. 11 2.4 Microencapsulation with xanthan gum and chitosan 2.4.1 Polyelectrolyte complex (PEC) gels by xanthan gum and chitosan Polyelectrolytes are macromolecules having many ionizable groups such as pro- teins and biopolymers. They fall apart into charged polyions and many oppositely charged counterions when dissolved in polar solvents like water [103]. Their size and shape depends on the charge and interaction with counterions. With increas- ing charge, the exible chain changes its shape from a contracted random coil to a fully extended one since the charges attached to the polymer repel each other [104]. When the polyelectrolyte chain stretches out, it occupies more space and the solu- tion becomes more viscous. On the other hand, addition of low-molecular weight salt causes a decrease in the range of the intramolecular coulomb force and polymer becomes a random coil again [103]. In general, the conflguration of the polyelec- trolytes depends on several factors such as pH, ionic strength, type of the solvent, concentration, molecular weight and distribution of the charged group. This con- flguration determines the dynamic response of the polyelectrolyte solution and the resultant rheology [105]. Polyelectrolyte complex (PEC) gels formed by the electrostatic attractions be- tween two oppositely charged polyelectrolytes mixed in an aqueous solution [106] are known to exhibit unique physical and chemical properties as the electrostatic in- 12 teractions are considerably stronger than most secondary binding interactions [107]. A variety of polyelectrolyte complexes can be obtained by changing the chemical structure of the components such as molecular weight, exibility, functional group structure, charge density, hydrophilicity, hydrophobicity, and stereoregularity, as well as changing reaction conditions such as pH, ionic strength, polymer concentra- tion, mixing ratio, and temperature [85]. In the last decade, there has been an increasing interest in the use of PEC gels formed by chitosan and polyanions, such as alginate, carrageenan, pectin, poly(acrylic acid), carboxymethylcellulose (CMC), and xanthan gum as carriers for drug delivery and in immobilized systems [108{113]. Xanthan gum and chitosan form a three dimensional network formed by reversible ionic linkages that can ab- sorb much more water than their own weight [113,114]. As polyelectrolyte hydrogels, xanthan-chitosan complexes were suggested as promising candidates for targeted de- livery and controlled release of encapsulated products for oral administration due to the fact that only nontoxic metabolites are produced during degradation and the complex has relatively high enzymatic resistance and pH-sensitive swelling charac- teristics [115{117]. Xanthan-chitosan PEC gels were studied as microcarriers mostly for encapsulation of enzymes [116{119] and the studies on the applicability of the system for bacterial cells are scarce [116]. 13 2.4.2 pH-sensitiveswellingcharacteristicsofxanthan-chitosan gels Chitosan-xanthan hydrogels are suggested as a potential candidate for targeted delivery since the pH-sensitive swelling allows the slow release of the encapsulant in the intestines while providing protection against acidic GI-tract conditions [117]. The swelling behavior of the xanthan-chitosan hydrogels is in uenced by polymer properties such as molecular weight and DA of chitosan as well as the complexa- tion conditions such as chitosan solution pH and complexation time, all of which determines the crosslink density of the complex [113,118]. In weak acid-weak base polyelectrolyte complexes as in the case of xanthan-chitosan, the number of elec- trostatic bonds varies with the ambient pH because of the change in the degree of dissociation [85]. The xanthan-chitosan complex swells in the range of pH = 10-12 in NaOH solution, with a maximum equilibrium swelling ratio (deflned as [Equilibrium diameter=Initial diameter]3) at pH = 10, and also at pH=0 in HCl solution. In the presence of NaCl, maximum swelling can be achieved at pH = 8 since the presence of ions weakens the ionic interactions and increases the osmotic pressure difierence between the ambient solution and inside of the gel [115]. The ion- izable groups of the xanthan gum and chitosan, which are a carboxyl group and an amino group respectively, are oppositely charged and are electrostatically attracted to each other during the gel formation. When the pH of the solution is increased, amino groups start to deionize while carboxyl groups hold the negative charge which will cause the electrostatic linkages to disappear (Figure 2.3). Meanwhile, the car- 14 boxyl groups attract Na+ ions and water difiuses into the complex. Both of these phenomena increase the osmotic pressure of the gel and consequently, the complex swells. However, equilibrium swelling ratio decreases in NaOH solution at pH values higher than 10 since too much increase in Na+ concentration decreases the difierence of osmotic pressure between the gel and ambient solution. Similarly, the swelling in acidic solutions can be explained by the neutralization of the negative charges of xanthan while the amino groups of chitosan hold their positive charge, resulting in the swelling of the gel complex [115]. Swelling capsules formed by xanthan-chitosan complexation can be described as hollow spheres that expand until equilibrium is reached in the release solution. Determination of the efiective difiusion coe?cient of hydrogels is very complex since their properties are afiected by many parameters. Due to the swelling and destruc- tion of the linkages between two polymers, the radius of the capsules and the amount of the impermeable segments (obstruction efiect) change with time which makes the calculations more complex. However, efiective difiusion coe?cient is expected to increase with the increased swelling and decreased polymer fraction [120]. 15 NaOH + + -- - - + + + + -- - - Chitosan Xanthan + :NH3+ - :COO- :NH2 Amino groups are deionized H2O H2O (a) + + + + -- - - Chitosan Xanthan NaOH NaCl Na+ Na+ Na+ Na+ -- - - + :NH3+ - :COO- :NH2 (b) Figure 2.3: pH-sensitive swelling of xanthan-chitosan complex: (a) In alkaline pH (10 to 12), amino groups deionize and electrostatic linkages disappear; whereas (b) attraction of Na+ ions increases osmotic pressure difierence and causes swelling of the gel where addition of salt helps achieve the swelling at a lower pH such as pH 8. 16 Chapter 3 Efiects of Complexation Conditions on Xanthan-Chitosan Polyelectrolyte Complex Gels 3.1 Introduction Polyelectrolyte hydrogels formed by xanthan gum and chitosan can be used for encapsulation and controlled release of food ingredients, cells, enzymes, and therapeutic agents. A variety of xanthan-chitosan PEC gels can be obtained by changing the molecular properties of the xanthan and chitosan polymers, such as molecular weight, degree of acetylation of chitosan, and pyruvic acid content of xanthan, as well as changing the complexation conditions, such as chitosan solution pH, polymer concentration, complexation time, and mixing ratio [113,118,119,121]. This study addresses the importance of polymer concentration and chitosan 17 solution pH in the complexation of xanthan gum and chitosan in the form of cap- sules. The swelling degree of the microcapsules formed by difierent combinations of xanthan and chitosan was studied as an indication of the crosslink density of the hydrogel membrane. Crosslink density is an important factor determining the sta- bility, pH-sensitive swelling behavior (thus the release properties), as well as the me- chanical strength of hydrogel networks. [113,122{124]. An increase in the crosslink density restricts the degree of swelling and reduces the pH-sensitivity by improv- ing the stability of the network. Hence, by controlling the amount of crosslinking, xanthan-chitosan microcapsules suitable for targeted release of probiotics can be prepared. 3.2 Materials and methods 3.2.1 Preparation of chitosan and xanthan solutions Chitosan from crab shells with a minimum deacetylation amount of 85% and a molecular weight of 370 000 (reported by the supplier) was purchased from Sigma- Aldrich Chemicals (St. Louis, MO). A known amount of chitosan was dissolved in 1 N HCl by heating and agitating. The desired solution pH was adjusted by 1 M NaOH and DI water was added to bring it to the flnal volume. Xanthan gum with a molecular weight of 1.02 million (TICAXAN r) was kindly supplied by TIC Gums (Belcamp, MD). A predetermined amount of xanthan gum was dissolved in DI water under heating and agitation. Both solutions were autoclaved before use. 18 3.2.2 Capsule formation and Inverted microscopy In this study, the extrusion (complex coacervation) method was used. Capsules were formed by dropwise addition of a solution of xanthan (50 mL) into a solution of chitosan (300 mL) using a manually operated syringe with a 0.7-mm cannula (Becton-Dickinson, Franklin Lakes, NJ). The chitosan solution was agitated con- tinuously for 40 min to allow crosslinking and avoid coalescence of capsules. The capsules were flltered through a 160 ?m Millipore nylon fllter, washed twice with DI water, and then freeze-dried for 24 hours. NHS-Fluorescein labeled chitosan, prepared according to Yi et al. [125], was kindly obtained from Dr. Payne?s lab in the Center of Biosystems Research at University of Maryland, College Park. Fluorescently-labeled capsules were examined under Axiovert 200 Inverted microscope (Carl Zeiss, Inc, Oberkochen, Germany). 3.2.3 Determination of the swelling degree The efiect of three complexation parameters, namely initial xanthan solution concentration (0.5, 0.7, 1.0, and 1.5% w/v), initial chitosan solution concentration (0.7 and 1.0% w/v) and chitosan solution pH (4.5, 5.5, and 6.2) on the degree of swelling of the resulting capsules were studied. Ten freeze-dried capsules were weighed and suspended in DI water overnight for each combination. The capsules were flltered, blotted to remove surface water, and weighed. The swelling degree (SD) values were calculated using the following equation: 19 SD (%) = Ws ?WdW d ?100 (3.1) where Ws and Wd are the weight of swollen capsules and that of dry capsules, respectively. The averages of four replicates for each combination were reported. 3.2.4 DifierentialScanningCalorimetry(DSC)measurements Difierential scanning calorimetry (DSC) measurements were carried out on a TA Instruments Q100 DSC device (New Castle, DE). The cell resistance and capacitance calibrations were performed in two steps. The flrst step was heating an empty cell and the second step was heating the cell with equal weight sapphire disks on the sample and reference platforms. The cell constant and temperature calibrations were performed with an indium standard. Standard DSC was used for the flrst heating and cooling runs. Approximately 5 mg of dry capsules were placed in sealed aluminum pans in small pieces. Each sample was heated up to 160x89 at a rate of 10x89/min and cooled back to 40x89 at a rate of 5x89/min to erase the thermal history of the polymers and eliminate the efiect of moisture. Modulated DSC curves were obtained from the second heating run at 2x89/min. Samples were heated from 100x89 to 175x89 with a modulation period of 60 seconds and modulation amplitude of 0.318x89. A nitrogen purge was applied for all experiments. The reversing signal was used to compare the glass transitions of the samples. In addition, chitosan in akes and xanthan in powder form were subjected to the same experimental procedure to determine their glass transition 20 temperatures. 3.2.5 Statistical analysis Statistical analyses were conducted using SAS 9.1.2 Software (Cary, NC). Fac- torial analysis of variance was used to analyze the efiect of xanthan concentration, chitosan concentration, and chitosan solution pH on the swelling degree of the cap- sules. Difierences in least square means were used for pairwise mean comparison. Analyses were performed using mixed procedure of SAS. 3.3 Results and discussion 3.3.1 Capsule formation By extruding the xanthan solution into moderate concentrations of chitosan (0.7 and 1.0% [w/v]), no stable capsules could be formed while xanthan concen- trations were below 0.5% (w/v). Xanthan concentrations exceeding 1.5% (w/v) resulted in formation of amorphous capsules (Figure 3.1-a, Figure 3.1-b). The pH of chitosan solution was controlled, ranging from 4.5, where the ionization degree of chitosan is unity, to 6.2, since chitosan precipitates above its pKa value of approxi- mately 6.3. Capsules formed by xanthan-chitosan complexation can be described as hol- low spheres. Figure 3.1-c shows a xanthan-chitosan capsule prepared by NHS- Fluorescein labeled chitosan under inverted microscope. 21 2m (a) 2m (b) (c) Figure 3.1: Xanthan-chitosan capsules formed when initial xanthan concentration is (a) greater than 1.5% (w/v); (b) less than 1.5% (w/v); (c) Inverted microscopy image of a 2 mm-diameter xanthan-chitosan capsule prepared with uorescently labeled chitosan. 3.3.2 Efiect of complexation conditions on the swelling de- gree of xanthan-chitosan capsules The swelling degree of microcapsules formed by difierent combinations of xan- than and chitosan was studied as an indication of the crosslink density of the hydro- gel membrane. The degree of swelling decreased by approximately 50% when using 0.7% (w/v) xanthan solution instead of 0.5% (w/v)(Figure 3.2). Therefore capsules formed from xanthan solutions at concentration of 0.5% (w/v) were not included in the statistical analysis and DSC studies. 22 Figure 3.2: Swelling degree of xanthan-chitosan capsules formed by using 0.5% (w/v) and 0.7% (w/v) xanthan solutions. The xanthan concentration was found to have a very signiflcant efiect on the swelling degree of xanthan-chitosan capsules at both chitosan concentrations (P < 0:0001) (Table 3.1). However, this efiect was found to be dependent on chi- tosan concentration and pH. When 0.7% (w/v) chitosan was used, the high the xanthan concentration yielded a lower swelling degree under all conditions studied except when xanthan concentration was increased from 1.0% to 1.5% (w/v) at pH 6.2. On the other hand, with 1.0% (w/v) chitosan, increasing xanthan concentration resulted in signiflcantly lower swelling degrees only when xanthan concentration was increased from 1.0% to 1.5% (w/v) at pH 4.5 and 5.5 (P < 0:05) and no signifl- cant difierence was observed at other conditions. At both chitosan concentrations, the decrease in the swelling degrees of the capsules with the increase in xanthan 23 concentration was more pronounced when chitosan solution pH was 5.5 and least signiflcant at pH 6.2. Table 3.1: Swelling degrees of xanthan-chitosan capsules in DI water Chitosan (w/v) 0.7% 1.0% Xanthan (w/v) pH 4.5 pH 5.5 pH 6.2 pH 4.5 pH 5.5 pH 6.2 0.7% ?29i 40 21m 24bgi 27cg 17dm 1.0% 21j 30 13an 22bhj 26ch 12dn 1.5% 13ek 16el 15aeo 16fk 11fl 15dfo Mean values with same letter are not signiflcantly difierent at P=0.05 level. * Swelling degree/100 The efiect of chitosan solution pH on the degree of swelling was more signiflcant when the chitosan solution concentration was 0.7% (w/v) (P < 0:0001) than when it was 1.0% (w/v) (P < 0:005) (Table 3.1). With 1.5% (w/v) xanthan concentra- tion, the increases in pH had no signiflcant efiect on the swelling degree (P > 0:05), whereas at other xanthan concentrations, the degree of swelling decreased signifl- cantly when chitosan pH was increased from 4.5 to 6.2. Such signiflcant decreases in swelling degree with increasing chitosan pH could be attributed to the changes in the chain exibility of chitosan polymer with the changes in the solution pH. The ionization degree of chitosan decreases from 1.0 to 0.5 as pH increases from 4.5 to 6.2 [126], which means that amino groups become less charged as pH increases. As a result, one may expect that fewer ionic linkages would occur between the two 24 polymers, resulting in higher swelling degrees as the pH approaches to its pKa value. However, since the charge density of the chitosan molecule is reduced by almost 50% as pH approaches 6.2 from a value of 4.5, the polymer chains become less extended with a smaller radius of gyration. This might result in a higher difiusion coe?cient for chitosan chains at pH 6.2, consequently enhancing difiusion of chitosan into the xanthan-chitosan network and forming more linkages during the specifled reaction time. This result suggests that since the formation of the xanthan-chitosan network is instantaneous upon contact, the difiusion of chitosan chains in the bulk solu- tion through this network plays an important role to achieve a highly crosslinked structure with a small swelling degree. Intriguingly, when 0.7 and 1.0% (w/v) xanthan solutions were extruded into 0.7% (w/v) chitosan solution, increases of chitosan pH from 4.5 to 5.5, flrst signifl- cantly increased the degree of swelling before it reached the lowest swelling degree at pH 6.2. This signiflcant increase in the swelling degree when chitosan pH was raised from 4.5 to 5.5 might be associated with the slight decrease in the charge density of chitosan (ionization degree ca. 0.9) that lessens the ionic linkages between the two polymers, rendering higher swelling degrees. Magnin et al. [113] have demonstrated similar results where the swelling degree of xanthan-chitosan matrix continued to increase with increasing chitosan solution pH from 3.5 to 5.8 in bulk form by us- ing 0.65 wt% chitosan and 0.65 wt% xanthan solutions. Moreover, our results also showed that when 1.0% (w/v) chitosan concentration was used, pH change from 4.5 to 5.5 chitosan solution had no signiflcant efiect on the swelling degree, suggesting that the swelling degree was less afiected by the decrease in the charge density of the 25 chitosan chains as pH approaches to 5.5 when chitosan concentration is increased. The efiect of chitosan solution concentration on the swelling degree of capsules was less pronounced than the efiect of xanthan solution concentration and chitosan solution pH. Increasing chitosan concentration from 0.7 to 1.0% (w/v) signiflcantly decreased the degree of swelling only when chitosan pH was 5.5 and at xanthan concentrations of 0.7 or 1.0% (w/v). No signiflcant difierence was observed at other conditions. These flndings indicate that the parameters studied cannot be viewed as independent parameters, as the efiect of one parameter on the degree of swelling depends on the other two parameters. While swelling studies were capable of iden- tifying the combined efiect of polymer concentration and chitosan solution pH on the crosslink densities of xanthan-chitosan hydrogel capsules, further investigation is needed to understand the difierences in the membrane structure as in uenced by the xanthan-chitosan hydrogel preparation conditions. Therefore, DSC analysis was performed to compare the changes in thermal transitions in order to elucidate the changes in the crosslinking density of the capsule network. 3.3.3 DSC analysis of xanthan-chitosan capsules Conventional DSC was used for the flrst heating and cooling runs. The flrst heating run of each sample gave single endothermic peak at about 100x89, which was attributed to the absorbed water. Samples were cooled back to 40x89 before the second heating. The modulated difierential scanning calorimetry (MDSC) technique was used for the second heating. MDSC applies two simultaneous heating rates to 26 the sample. The linear heating rate provides total heat ow as conventional DSC and sinusoidal (modulated) heating rate provides the heat capacity-related (reversing) component of the total heat ow. The reversing signal was used to quantify the glass transition since it separates the glass transition completely from other non-reversing processes [127,128]. The transition enthalpies were calculated by integration of the peaks on the reversing heat capacity (Rev Cp) curves as is usually done for flrst-order phase transitions [129]. The small glass transitions observed for both chitosan and xanthan can be explained by the fact that both polymers are partially crystalline (Figure 3.3). The in ection points of the peaks on the Rev Cp curves correspond to the glass transition temperature (Tg) of the samples. The Tg of chitosan was determined to be approx- imately 148x89 and the enthalpy of this transition (H) was calculated as 0.016 J/g. The Tg for xanthan gum was found to be at approximately 143x89 with a transition enthalpy of H=0.01 J/g. Representative reversing heat ow curves from the second heating runs of the freeze-dried hydrogel capsules are shown in Figures 3.4 and 3.5. The transition enthalpiesandtheglasstransitiontemperaturesweredeterminedasdescribedabove. It is evident from these data that the polymer complex shows weak transitions as expected in physically crosslinked networks. For this reason, the transition enthalpy of xanthan gum, 0.01 J/g, was selected as the threshold enthalpy to difierentiate the noise from the actual transitions appearing on the MDSC curves of the hydrogel capsules. Figure 3.4 shows the efiect of initial xanthan concentration on the capsule 27 (a) (b) Figure 3.3: MDSC curves of (a) chitosan; (b) xanthan gum. 28 (a) (b) Figure 3.4: MDSC curves of xanthan-chitosan capsules showing the efiect of initial xanthan con- centration on the resulting capsule network structure. (a) Chitosan: 0.7% (w/v) pH=6.2; (b) Chitosan 1.0% (w/v) pH=4.5. 29 network structure. In Figure 3.4-a, the MDSC curves of capsules made from 0.7% (w/v) chitosan solution at pH 6.2 are shown. When the xanthan solution was 0.7% (w/v), two glass transition temperatures at approximately 144x89 and 149x89 were observed. This indicates the presence of uncrosslinked xanthan gum and chitosan in the network. However, these transitions disappeared in the samples formed using 1.0% (w/v) xanthan, which may indicate complete crosslinking of both polymers. This increase in crosslink density explains the signiflcant decrease in the swelling degree of capsules when the initial xanthan concentration was increased from 0.7% (SD = 2049) to 1.0% (SD = 1305). On the other hand, increasing xanthan con- centration to 1.5% (w/v) did not result in a signiflcant difierence in the degree of swelling (SD = 1520) since the network was already completely crosslinked when 1.0% (w/v) xanthan was used. Figure 3.4-b shows the MDSC curves of capsules prepared from 1% (w/v) chitosan solution at pH 4.5. The transitions appeared approximately at 147x89 and 148.5x89 in the 0.7% and 1.0% xanthan curves, respectively, can be attributed to the glass transition of chitosan and the transition appeared in 1.0% (w/v) xanthan curve at approximately 143x89 corresponds to the glass transition of xanthan gum. This suggests the presence of uncrosslinked polymer chains in both networks. When xanthan concentration was increased to 1.5% (w/v), no transitions were observed suggesting a completely crosslinked network. These results are in good agreement with the swelling study, since there was no signiflcant difierence in the swelling degree when xanthan concentration was increased from 0.7% (w/v) (SD = 2375) to 1.0% (w/v) (SD = 2164) and with further increase in xanthan concentration to 30 1.5% (w/v), the degree of swelling decreased signiflcantly (SD = 1573). The efiect of pH on the network structure of the capsules made from 1.0% (w/v) chitosan and 1.0% (w/v) xanthan solutions is shown in Figure 3.5. Both transitions from xanthan gum and chitosan observed at pH 4.5 were eliminated when the pH was increased to 6.2. This result might suggest an incomplete crosslinking between the two polymers at pH 4.5 and a complete crosslinking at pH 6.2. The swelling data also shows that there is a signiflcant decrease in the degree of swelling when pH is increased from 4.5 (SD = 2164) to 6.2 (SD = 1187), which can be explained by the increase in the crosslink density. Figure 3.5: MDSC curves of xanthan-chitosan capsules showing the efiect of initial chitosan solution pH on the resulting capsule network structure. Xanthan: 1% (w/v), Chitosan: 1% (w/v). 31 3.4 Conclusions Characterization of factors contributing to the crosslink density of xanthan- chitosan network is important in developing hydrogels with particular mechanical and controlled release properties. Results from the swelling degree and DSC experi- ments showed that the crosslink density of xanthan-chitosan network was dependent on the complexation conditions employed in the present study. Xanthan concen- tration was found to be the most critical parameter in xanthan-chitosan network formation. The hydrogel capsules were completely crosslinked at all conditions stud- ied when initial xanthan solution concentration was at 1.5% (w/v). The increase in xanthan concentration signiflcantly afiected the degree of swelling of the hydrogel at both chitosan concentrations. On the other hand, the efiect of chitosan solution pH on the degree of swelling was more pronounced at 0.7% (w/v) than at 1.0% (w/v) chitosan concentration. The swelling degree was less dependent on chitosan con- centration than xanthan concentration and chitosan solution pH. Conformational changes of chitosan polymer chains, which is dependent on the solution pH, were critical in determining the crosslinked network structure that afiects the swelling degrees of resulting gels. Results from this study showed that pH and concentration efiects on the xanthan-chitosan network properties are dependent on each other. It can be concluded that, the xanthan-chitosan network properties can be easily modulated by changing operationally controllable parameters, especially xanthan concentration and chitosan solution pH. 32 Chapter 4 The Release Kinetics and Swelling Behavior of Xanthan-Chitosan Capsules 4.1 Introduction Xanthan-chitosan hydrogels swell in response to changes in pH, temperature, and ionic strength of the aqueous environment. In swelling controlled systems, the molecular structure as well as the nature of the polymer system control the mechanism by which a solute may be released from a polymer network [130] . At high or very low pH values, the electrostatic linkages between xanthan gum and chitosan start to disappear, allowing the network to expand and imbibe water [115]. The release of probiotic cells by such a pH-sensitive swelling-controlled mechanism is related to the macromolecular chain relaxation and to the difiusion of the cells 33 through the polymer membrane under countercurrent difiusion of water or biological uids into the capsules. It is important that the encapsulation system developed in this study keeps the cellreleasetoaminimumuntilthecapsulesreachthesmallintestineswheretherapid release of the cells is desirable for their colonization. For this reason, the kinetics of cell release under difierent conditions need to be studied in order to understand the response of the system to environmental changes, particularly variations in ambient pH. The main goal of this study was to characterize the swelling and release behavior under simulated GI-tract conditions. 4.2 Materials and Methods 4.2.1 Calibration curve Pediococcus acidilactici (MA18/5M, National Collection of Microorganism Culture, Pasteur-France) was kindly provided by Imagilin Technology LLC (Po- tomac, MD). One gram of bacterial powder was hydrated in 9 mL of DI water for 30 min by shaking at 260 rpm. One mL hydrated bacteria was inoculated to 99 mL MRS broth and incubated for 24 hours at 35x89. To construct the calibration curve, a 25 mL sample was taken from MRS broth, centrifuged at 3000 rpm for 30 minutes and washed with sterile distilled water under the same conditions. After necessary dilutions, OD values at 600 nm were recorded by a Hefl IOS Spectropho- tometer (ThermoSpectronic, Rochester, NY) and the corresponding samples were 34 oven dried overnight for mass determination. 4.2.2 Microencapsulation of P. acidilactici One milliliter of hydrated P. acidilactici cells was inoculated into 99 mL MRS broth and incubated at 35x89for 24 hours. Actively growing cells were recovered from MRS broth by centrifuging at 10000 rpm for 10 minutes and then were washed twice with sterile phosphate bufiered saline (PBS) solution under the same centrifugation conditions. DI water was added to the cell pellet and vortexed. Xanthan and chitosan solutions were prepared as described in section 3.2.1. Encapsulation was achieved by dropwise addition of xanthan and P. acidilactici mixture (9:1 v/v) into chitosan solution using a manually operated syringe with 0.7-mm cannula. The chitosan solution was agitated continuously for 40 min to allow crosslinking and to avoid coalescence of capsules. The capsules were flltered through a 160?m Millipore nylon fllter, washed twice with DI water, and then freeze-dried for 24 hours. 4.2.3 Kinetics of cell release A known amount of freeze-dried capsules were suspended in release solutions, namely simulated gastric uid (Fisher Chemicals, Suwanee, GA), simulated intesti- nal uid (Fisher Chemicals, Suwanee, GA), and DI-water. The cell concentration in the solutions was monitored over time. A 1 mL sample was taken periodically in order to monitor the changes in OD values at 600 nm by using a HeflIOS Spec- trophotometer (ThermoSpectronic, Rochester, NY). Consequently, 1 mL of release 35 solution was added after each sample is taken in order to avoid any errors in calcula- tions. Dilutions were performed when necessary. The calibration curve of the strain was used to relate the OD values to cell concentration. Averages of two replicates for each sample were reported. Cell release kinetics were studied in deionized water for 72 hours at room temperature (under 150 rpm); in simulated intestinal uid (composed of pancreatin, KH2PO4 and NaOH) at pH=6.8 for 24 hours at room temperature (under 150 rpm); in simulated gastric uid (composed of pepsin, NaCl and HCl) at pH=1.0 and pH=2.0 (37x89 under 150 rpm); and flnally in simulated intestinal uid after 2 hour exposure to simulated gastric uid at pH=2.0 (37x89 under 150 rpm). The cell release data were analyzed according to the Higuchi equation [131], eqn (4.1); Korsmeyer-Peppas equation [132], eqn (4.2); and Peppas-Sahlin equation [133], eqn (4.3) using MATLAB version 6 R13 (MathWorks Inc, Natick, MA). Non- linear least squares fltting method was used to determine the parameters in each equation. Mt Mf = k 0t1=2 (4.1) Mt Mf = kt n (4.2) Mt Mf = kdt m +krt2m (4.3) where Mt is the concentration of the cells released at time t, Mf is the concentra- tion of the cells released at equilibrium, k and k0 are constants incorporating the structural and geometric characteristics of the hydrogel, and n is the release expo- nent describing the mode of the transport mechanism, and m is the purely Fickian 36 difiusion exponent for a system of any geometrical shape. 4.2.4 Dynamic swelling behavior Ten freeze-dried capsules were weighed and placed in simulated gastric uid followed by simulated intestinal uid. The capsules were carefully removed from the solutions, blotted for the removal of surface water and weighed at specifled time intervals. The swelling ratio, q, was calculated as follows: q = WsW d (4.4) where Ws is the weight of swollen capsules and Wd is the weight of the dried capsules. The averages of three replicates for each combination were reported. The water absorption data was analyzed according to the Higuchi equation, eqn (4.1); Korsmeyer-Peppas equation, eqn (4.2); and Peppas-Sahlin equation, eqn (4.3) using MATLAB version 6 R13. Non-linear least squares fltting method was used to determine the parameters in each equation. 4.3 Results and Discussion 4.3.1 Calibration curve The optical density of the cells in suspension was found to be related to cell concentration by a factor of 4.0773 (Figure 4.1). 37 y = 4.0773xR2 = 0.9782 00.2 0.40.6 0.81 0 0.05 0.1 0.15 0.2 0.25Conc (g/L) OD60 0 Figure 4.1: Calibration curve of P. acidilactici. 4.3.2 Kinetics of cell release In swelling-controlled-release systems, solute transport through the polymer network is controlled by several physico-chemical phenomena such as polymer water uptake, gel layer formation, and polymeric chain relaxation [134,135]. Delivery of the entrapped cells from hydrogel networks is a mass-transfer controlled process and depend on environmental factors such as pH, temperature, and ionic strength. The xanthan-chitosan capsules with difierent combinations were exposed to simulated gastric and intestinal uid to study the cell release kinetics of the capsules under GI-tract conditions. Figure 4.2 shows the cell release from xanthan-chitosan capsules (prepared from 0.7% [w/v] chitosan solution at pH 6.2 and 1% [w/v] xanthan solution) in DI- water, in SIF and in SGF at pH=1.0 and 2.0. The negligible release in DI-water for 72 hours (only flrst 26 hours is shown) indicates that capsules are stable in DI-water. 38 In SIF, the release from the capsules was very slow in the beginning and the release rate increased after 8 hours. Maximum release of the cells from the capsules was achieved after 24 hours in SIF. When capsules were suspended in SGF at pH=2.0, the cell release was in negligible amounts for 2 hours. However, cells were released rapidly in SGF at pH=1.0 and the OD values started to decrease after 2.5 hours, indicating cell death due to low pH. These results suggest that the pH of the SGF solution is more critical in determining the release properties of xanthan-chitosan capsules than the presence of enzyme (pepsin). The enzymatic resistance of the complex can be explained by the conformational changes that occur in chitosan after complexation with enzymatically resistant xanthan gum. These changes might make the sensitive sites of chitosan less accessible to hydrolytic action of the media [117]. Figure 4.3 shows the cell release proflles from xanthan-chitosan capsules for four difierent matrix formulations. Capsules were kept in SGF at pH 2.0 for 2 hours prior to be suspended in SIF at pH 6.8. For all combinations, the cell release in SGF was negligible. A time lag was observed in the early phase of the cell release from the capsules in SIF, indicating a highly-crosslinked network structure [136]. The time lag was longer for the capsules prepared from chitosan solution at pH 6.2 (60 min) than at pH 4.5 (30 min). Complete release of the encapsulated cells in SIF was achieved after 5.5 hours and 5 hours of exposure when initial chitosan solution pH was controlled at 6.2 and 4.5, respectively. Longer time lag and longer release time observed in the case of capsules prepared from chitosan solution at pH 6.2 suggests a more crosslinked hydrogel network compared to the membrane network of capsules prepared from chitosan solution at pH 4.5. These results are in agreement with the 39 (a) (b) Figure 4.2: The release of P. acidilactici cells from xanthan- chitosan capsules (Chitosan: 0.7% (w/v), pH=6.2; Xanthan: 1.0% (w/v)) in (a) DI water and SIF; (b) in SGF at pH=1.0 and pH=2.0. 40 swelling degree experiments reported in Chapter 3. Table 4.1: Analysis of cell release data from xanthan-chitosan capsules using Higuchi eqn, Korsmeyer-Peppas eqn, and Peppas-Sahlin eqn. Capsule Higuchi eqn? Korsmeyer-Peppas eqn? Peppas-Sahlin eqn? combinations k0 r2 n k r2 kd kr r2 with 0.7% Ch (min?0:5) (min?n) (min?0:43) (min?0:86) pH=6.2; X=0.7% 0.2374 0.7888 2.043 0.03598 0.9924 -0.5222?? 0.473 0.9936 pH=6.2; X=1.0% 0.0432 0.8704 1.233 0.00098 0.9467 -0.044?? 0.0116 0.94881 pH=4.5; X=0.7% 0.3304 0.9459 1.082 0.2422 0.9954 -0.1275?? 0.3724 0.9946 pH=4.5; X=1.0% 0.1954 0.8878 1.284 0.1001 0.9959 -0.1646?? 0.2624 0.9951 ?k0 (min?0:5), Higuchi kinetic constant; k (min?n), Korsmeyer-Peppas kinetic constant; kd (min?0:43), difiusional constant; kr (min?0:86), relaxational constant; r2, correlation coe?cient. ??The negative values obtained for kd should be interpreted in terms of a difiusion process insigniflcant compared to the relaxation mechanism [137]. The portion of the cell release curve with a fractional release (Mt/Mf) between 0.1 and 0.6 was analyzed according to the Higuchi equation in (4.1); Korsmeyer- Peppas equation in (4.2); and Peppas-Sahlin equation in (4.3). The Higuchi model is applicable if the release is largely governed by difiusion through water-fllled pores in the matrix. A good flt to Kormeyer-Peppas equation indicates the combined efiect of difiusion and relaxation mechanisms for the release. For spherical systems, n = 0:43 indicates Fickian (difiusion controlled) transport, 0:43 < n < 0:85 indicates non- Fickian (anomalous) transport and n = 0:85 implies Case II, zero-order (relaxation- controlled) transport [133,138]. Values of n higher than 0.85 for release from spheres are considered to be Super Case II kinetics [132]. The relative contribution of the difiusion and relaxation processes to the release mechanism can be analyzed by Peppas-Sahlin equation. The purely Fickian difiusion exponent, m, for xanthan- chitosanmicrocapsules corresponds toavalueof 0.43 since spherical matrices present 41 (a) (b) Figure 4.3: Fractional release of P. acidilactici cells from xanthan-chitosan capsules in SGF at pH=2.0 and in SIF at pH 6.8. (a) Chitosan: 0.7 % (w/v), pH=6.2; Xanthan: 0.7% (w/v) and 1.0% (w/v); (b) Chitosan: 0.7% (w/v), pH=4.5; Xanthan: 0.7% (w/v) and 1.0% (w/v). 42 an aspect ratio of 1 [133]. The main parameter values from difierent equations are summarized in Table 4.1 to describe the mode of transport mechanism. A good flt to Korsmeyer-Peppas (n > 0:85) and Peppas-Sahlin (kr kd) equations suggests that the cell release from xanthan-chitosan capsules in SIF (af- ter 2 hrs exposure to SGF) is controlled by polymer relaxation (Table 4.1). Since Higuchi model is only applicable to difiusion-controlled release mechanism, the re- lease data showed a poor flt to this equation. The n values higher than 0.85 reveals a Super Case II transport mechanism for the release of cells from xanthan-chitosan capsules regardless of the combination used. Super Case II transport mechanism might result from an increased plasticization at the relaxing boundary (gel layer) due to a reduction of the attractive forces among polymeric chains [139,140]. 4.3.3 Dynamic swelling behavior Xanthan-chitosanhydrogelsexhibitpH-sensitiveswellingcharacteristicsdueto the deionization of the functional groups in the hydrogel which signiflcantly afiects the penetrant transport mechanism of the polymer networks [115, 130]. Highly swollen hydrogels contain large amounts of unbound water which allows greater solute release [135]. The objective was to study the dynamic swelling behavior of xanthan-chitosan capsules under simulated GI-tract conditions. Figure 4.4 shows the dynamic swelling of xanthan-chitosan capsules, prepared from 0.7% (w/v) chitosan at pH=6.2 and 0.7% (w/v) or 1.0% (w/v) xanthan in SGF at pH=2.0 and in SIF. The swelling ratio, q, was constant in SGF and increased 43 (a) (b) Figure 4.4: (a) Dynamic swelling ; (b) water absorption curve of xanthan-chitosan capsules (pre- pared from 0.7% [w/v] chitosan at pH=6.2 in combination with 0.7% and 1.0% [w/v] xanthan) in SGF at pH=2.0 and in SIF at pH 6.8. 44 rapidly for the flrst two hours in SIF for both combinations. This rapid swelling observed in the early stage of swelling also supports the flndings in Section 4.3.2 suggesting that the rate-limiting factor for the cell release is chain relaxation. The capsules swelled to at least 80 times of their dry weights at equilibrium. While capsules with 1% (w/v) xanthan reached the equilibrium swelling ratio after two hours, at least 6 hours was needed for the capsules to reach equilibrium for capsules with 0.7% (w/v) xanthan gum. The water absorption data was analyzed according to the Higuchi equation, eqn (4.1); Korsmeyer-Peppas equation, eqn (4.2); and Peppas-Sahlin equation, eqn (4.3). The main parameter values from difierent equations are summarized in Ta- ble 4.2. A good flt to Higuchi model, the low n values determined by Korsmeyer- Peppas equation and signiflcantly higher values obtained for kd compared to kr by Peppas-Sahlin equation suggests that the swelling of xanthan-chitosan capsules in SIF (after 2 hrs exposure to SGF) is difiusion-controlled for both combinations. Table 4.2: Analysis of dynamic swelling data for xanthan-chitosan capsules using Higuchi eqn, Korsmeyer-Peppas eqn, and Peppas-Sahlin eqn. Capsule Higuchi eqn? Korsmeyer-Peppas eqn? Peppas-Sahlin eqn? combinations k0 r2 n k r2 kd kr r2 with 0.7% Ch (min?0:5) (min?n) (min?0:43) (min?0:86) pH=6.2; X=0.7% 0.0536 0.9792 0.3966 0.0094 0.9926 0.0863 -0.0007?? 0.9971 pH=6.2; X=1.0% 0.0911 0.9537 0.326 0.1889 0.9826 0.158 -0.0057?? 0.9859 ?k0 (min?0:5), Higuchi kinetic constant; k (min?n), Korsmeyer-Peppas kinetic constant; kd (min?0:43), difiusional constant; kr (min?0:86), relaxational conctant; r2, correlation coe?cient. ??The negative values obtained for kr should be interpreted in terms of a relaxation mechanism insigniflcant com- pared to the difiusion process [137]. 45 4.4 Conclusions In swelling-controlled systems, the coupling of difiusion and macromolecular relaxation control the release mechanism. The xanthan-chitosan capsules were found to be stable in DI water. The pH of SGF was critical in determining the release propertiesofthecapsules. ThecellreleaseinSGFatpH2.0wasnegligiblesuggesting that xanthan-chitosan capsules have a good potential for delivery of probiotic cells to intestines. The cell release from xanthan-chitosan capsules in SIF after 2 hr exposure to SGF at pH 2.0, exhibited a Super Case II transport mechanism (n > 0:85) regardless of the formulation used. The complete release of the cells from the capsules was achieved in at least 5 hours. Xanthan-chitosan capsules were found to swell by a difiusion-controlled mechanism. The rapid swelling observed also supported the flndings that the cell release is relaxation-controlled (kr kd). 46 Chapter 5 Protective efiects of Xanthan-Chitosan Encapsulation on P. acidilactici cells 5.1 Introduction For probiotic bacteria to be beneflcial to the host, they should be able to survive gastric transit and reach the small intestine in su?cient numbers [26,33]. TheharshenvironmentoftheGI-tract(thelowpHconditionsofthestomachandthe presence of bile in the intestines) adversely afiect the viability of probiotic cultures [37,38]. In this study, Pediococcus acidilactici cells were encapsulated in xanthan- chitosan PEC gels to increase the survival rates during GI-tract transit. Among the available techniques for encapsulating bacterial cells, entrapment in xanthan-chitosan capsules has not been used for the encapsulation of probiotic 47 bacteria. The only application of xanthan-chitosan PEC gels as microcarriers for bacterial cells was reported by Chu et al. [116] for the encapsulation of Corynebac- terium glutamicum, a Gram (+) soil bacterium. The immobilization was carried out by mixing xanthan-chitosan PEC solution with the cell suspension and adding this mixture to distilled water through a syringe. In this work, a simple and cost-efiective extrusion method was used to en- capsulate the probiotic cells. The goal was to determine the protective efiects of encapsulation with xanthan gum and chitosan on P. acidilactici cells against freeze- drying and simulated gastric uid at pH 2.0. 5.2 Materials and Methods P. acidilactici cells were encapsulated as described in Section 4.2.2. Fifty milliliters of cell-xanthan mixture were dropped into 200 mL of chitosan solution by using a manually operated syringe to form the capsules. Freeze-dried capsules were subjected to simulated gastric uid (SGF) at pH 2.0 for 1 hour and 2 hours at 37x89 under 150 rpm. Two difierent combinations (chitosan: 0.7% [w/v], pH 6.2; xanthan: 1.0% [w/v] and chitosan: 1.0% [w/v], pH 6.2; xanthan: 0.7% [w/v]) were used to encapsulate the bacteria. Capsules were flltered from SGF solution and suspended in SIF solution for 5 hours at 37x89 under 150 rpm to release the cells from the capsules. One milliliter aliquots from the release solutions were serially diluted and 3 replicates for each dilution were plated on MRS agar and incubated for 48 hours at 35x89. The experiments were repeated twice for each combination. 48 Smaller capsules were prepared by spraying the xanthan gum solution at 0.7% (w/v) through a 0.7 mm nozzle (B?uchi B-290 Spray Dryer, Flawil, Switzerland) into chitosan solutions at 3 difierent concentrations (0.4, 0.7, and 1.0 % [w/v]) and 2 difierent pH values (4.5 and 6.2). Moreover, the efiect of increasing xanthan concentration from 0.7 to 1.5% (w/v) on the protective properties of the capsules was studied. Capsules were subjected to SGF at pH 2.0 for 1 hour and cells were released from capsules in SIF solution. One milliliter aliquots from the release solutions were serially diluted and 3 replicates for each dilution were plated on MRS agar and incubated for 48 hours at 35x89. The experiments were repeated three times for each combination. Free cells were mixed with 10% (w/v) skim milk (9:1 v/v), transferred to freeze-drying ask and frozen in dry ice before connected to the freeze dryer (Ther- moSavant, Holbrook, NY). Freeze-dried cells were suspended in SGF for 1 hour and 2 hours under 150 rpm. One milliliter samples from free-cell solution were serially diluted and 3 replicates for each dilution were plated on MRS agar and incubated for 48 hours at 35x89. 5.3 Results and Discussion P. acidilactici are used in dietary supplements and in animal feed for their probiotic efiects. They can produce large amounts of lactic acid that helps keep a proper balance of micro ora in the digestive system [141]. Nevertheless, to act as a probiotic, the bacteria must survive in acidic stomach conditions and arrive in 49 the intestines in high numbers. Gastric juice has a very low pH of 0.9 in healthy adults. However, the presence of food may raise the pH value to 3.0 and it takes 2 to 4 hours for stomach to empty after ingestion of food [142]. In this work, free and encapsulated P. acidilactici cells were subjected to SGF at pH=2.0 for 1 hr or 2 hours at 37x89 with moderate shaking to evaluate the efiect of encapsulation on the survival of P. acidilactici cells in stomach conditions. Two difierent combinations based on the swelling degree and DSC studies (Chapter 3) were chosen to encapsulate the bacteria in the capsules prepared by us- ing a syringe. The capsules formed with these combinations were highly crosslinked with relatively small swelling degrees (Table 3.1). Xanthan-chitosan capsules pre- pared from 0.7% (w/v) chitosan at pH 6.2 and 1.0% (w/v) xanthan showed six-log preservation of P. acidilactici cells over free suspending cells after exposed to SGF (pH=2.0) for 1 hr and over four-log retention when exposure time was 2 hours (Fig- ure 5.1-a). The protection efiect of the capsules prepared from 1.0 (w/v) chitosan at pH 6.2 and 0.7% (w/v) xanthan was more than flve logs and three logs after 1hr and 2 hours in SGF, respectively (Figure 5.1-b). The resistance of lactic acid bacteria to freeze-drying and low pH is strain speciflc. It has been reported that survival rates of P. acidilactici in simulated gastric juice (SGF) is similar to the survival rates of Lactococcus lactis subsp. lactis and slightly lower than Lactobacillus casei subsp. casei [143]. Xanthan-chitosan microcapsules were found to provide better protection to P. acidilactici cells against freeze-drying and SGF than the protection provided by alginate capsules to L. casei cells [144,145]. 50 (a) (b) Figure 5.1: Protective efiects of encapsulation on the viability of P.acidilactici cells against freeze- drying and SGF (pH=2.0) exposure (n=2) using (a) 0.7% and 1.0% chitosan. 51 The higher protection observed with the capsules prepared from 0.7% (w/v) chitosan at pH 6.2 and 1.0% (w/v) xanthan might be attributed to higher con- centration of xanthan gum used compared to the other combination. Xanthan gum is known to be enzymatically resistant and is stable at low pH in contrast to other gums possessing carboxylic acid function, which are greatly in uenced by pH [70,71,117,146]. Sun and Gri?ths [51] also reported that encapsulation of bifldobacteria in xanthan-gellan capsules increased their tolerance of high acid en- vironments compared to alginate encapsulation. Smaller size capsules were prepared by using a 0.7-mm nozzle to see the efiect of capsule size on the protective properties of xanthan-chitosan capsules (Figure 5.2). Figure 5.2: Relative size of capsules prepared by using a syringe and a nozzle. Although encapsulation in smaller capsules provided some protection against freeze-drying and low pH, protective efiects observed in the case of smaller capsules (Figure 5.3-a) were signiflcantly lower than that of provided by larger capsules (Fig- ure 5.1). For instance, when capsules were prepared from 1.0% (w/v) chitosan at pH 6.2 and 0.7% (w/v) xanthan gum, encapsulation in larger capsules improved the cell 52 (a) (b) Figure 5.3: Protective efiect of xanthan-chitosan capsules prepared by using a 0.7-mm nozzle on the viability of P. acidilactici cells against freeze-drying and SGF exposure for 1 hr. (a) Efiect of initial chitosan solution concentration and pH (Xanthan= 0.7 % (w/v), n=3), (b)Efiect of initial xanthan gum concentration (Chitosan=0.7 % (w/v) & pH 4.5, n=3). 53 viability at least three logs compared to the encapsulation in small capsules. Simi- lar results have been reported by other researchers. Lee and Heo [45] reported that survival of bacterial cells under in vitro gastric conditions decreased with decreasing capsule size (diameter 1-2.6 mm). Sheu and Marshall [21] indicated that larger bead diameters provided more protection for Lactobacillus bulgaricus in frozen desserts. Moreover, it has been reported that small beads of size less than 100 ?m did not signiflcantly protect the probiotic bacteria in simulated gastric juice compared to free cells [147]. Finally, the efiect of increasing initial xanthan concentration on the protective properties of xanthan-chitosan capsules was studied for the capsules prepared by nozzle. It was found that increasing initial xanthan gum concentra- tion from 0.7 to 1.5% (w/v) only slightly improved the cell viability when chitosan solution concentration and pH were 0.7% (w/v) and 4.5, respectively (Figure 5.3-b). These results suggest that xanthan-chitosan hydrogels have a promising pro- tective mechanism for probiotic preservation. 5.4 Conclusions Encapsulation of probiotic bacteria in capsules composed of xanthan gum and chitosan increased the survival rate of the cells during freeze-drying and in simulated gastric uid compared to non-encapsulated cells. The viability of freeze-dried P. acidilactici cells in simulated gastric conditions was found to increase with increased capsule size. This approach might be useful in delivery of probiotic cultures to the GI-tract in high numbers to beneflt from the health promoting efiects of probiotic 54 cells. 55 Chapter 6 Conclusions In this dissertation, a microencapsulation system using polyelectrolyte com- plex gels formed by xanthan gum and chitosan was developed to sustain the viability of probiotic bacteria in simulated gastrointestinal tract conditions. First, the efiects of initial polymer concentration and chitosan solution pH on the crosslinking density of the capsule network were investigated by swelling studies and modulated difieren- tial scanning calorimetry analysis. Xanthan concentration was found to be the most critical parameter in xanthan-chitosan network formation. Conformational changes of chitosan polymer chains as pH approaches to 6.2 were critical in determining the crosslinked network structure. Swelling degree was less dependent on chitosan concentration than xanthan concentration and chitosan solution pH. Results from this study showed that when initial xanthan concentration was 1.5% (w/v) complete crosslinking can be achieved at all conditions studied. However, it was also possi- ble to obtain a completely crosslinked network at 1% (w/v) xanthan when chitosan solution pH was 6.2. 56 In the second part of this dissertation, kinetics of cell release from highly crosslinked xanthan-chitosan capsules was studied under difierent conditions. It was found that xanthan-chitosan capsules were stable in DI water and cell release was negligible in simulated gastric uid (SGF) at pH 2.0 for 2 hrs. These flndings suggest that xanthan-chitosan capsules have a good potential for the delivery of probiotic cells to intestines with minimum damage to the cells in stomach conditions. The cell release from xanthan-chitosan capsules in simulated intestinal uid (after 2 hrs exposure to SGF at pH 2.0) was found to be relaxation-controlled and exhibited Super Case II transport mechanism. The maximum cell release was achieved at least in 5 hrs. This rapid release in intestinal conditions was desirable for the colonization of the cells in the intestines. Finally, the protection efiects of capsules on probiotic cells against freeze- drying and simulated gastric uid at pH 2.0 were investigated. Capsules with high crosslinking densities were chosen to encapsulate P. acidilactici cells. Xanthan- chitosan capsules showed up to six-log and four-log preservation of probiotic cells in SGF after 1 hr and 2 hrs, respectively. The viability of the cells was found to increase with the increase in capsule size. In conclusion, encapsulation of probiotic bacteria in xanthan-chitosan capsules might be useful in delivery of probiotic cultures to the intestines in high numbers to beneflt from their health promoting efiects. Based on the results reported in this study, capsules prepared from 0.7% (w/v) chitosan at pH 6.2 and 1.0% (w/v) xanthan gum might be suggested as the optimum combinations to encapsulate the probiotic cells. 57 BIBLIOGRAPHY [1] R. Agheyisi. Ga-121 probiotics: Ingredients, supplements, foods. Technical report, Business Communication Company, Inc, Norwalk, CT, 2005. [2] S. K. Hood and E. A. Zottola. Efiect of low pH on the ability of Lactobacillus- acidophilus to survive and adhere to human intestinal-cells. Journal of Food Science, 53(5):1514{1516, 1988. [3] F. A. M. Klaver, F. Kingma, and A. H. Weerkamp. Growth and survival of bifldobacteria in milk. Netherlands Milk Dairy Journal, 47:151{164, 1993. [4] N. P. Shah and W. E. V. Lankaputhra. Improving viability of Lactobacillus acidophilus and Bifldobacterium spp. in yogurt. International Dairy Journal, 7:349{356, 1997. [5] D. B. Hughes and D. G. Hoover. Bifldobacteria: Their potential for use in american dairy products. Food Technology, 45(4):74{83, 1991. [6] B. Ahmed and B. K. Mital. Efiect of magnesium and manganese ions on the growth of Lactobacillus acidophilus. Journal of Food Technology, 27:228{229, 1990. [7] V. Babu, B. K. Mital, and S. K. Garg. Efiect of tomato juice addition on the growth and activity of Lactobacillus acidophilus. International Journal of Food Microbiology, 17:67{70, 1992. [8] V. M. Marshall. Inoculated ecosystems in a milk environment. Journal of Applied Bacteriology, 73:S127{S135 Suppl. S, 1992. [9] T. Mitsuoka. Role of intestinal ora in health with special reference to dietary control of intestinal ora. In B. .H Hga and Y. K. Lee, editors, Microbiology applications in food biotechnology, pages 135{148. Elsevier Science Publishers Ltd, London, 1992. [10] T. Tanaka and K. Hatanaka. Application of hydrostatic pressure to yogurt to prevent its after acidiflcation. Journal of the Japanese Society for Food Science and Technology, 39:173{177, 1992. 58 [11] D. Supriadi, K. Kailasapathy, and J. A. Hourigan. Efiect of partial replace- ment of skim milk powder with whey protein concentrate on bufiering capacity of yogurt. In B. Dixion and Muller L., editors, The XXIV International Dairy Congress, Melbourne, 1994. The Australian National Committee of the Inter- national Dairy Federation. [12] R. I. Dave and N. P. Shah. Ingredient supplementation efiects on viability of probiotic bacteria in yogurt. Journal of Dairy Science, 81:2804{2816, 1998. [13] A. M. P. Gomes, F. X. Malcata, and F. A. M. Klaver. Growth enhance- ment of Bifldobacterium lactis Bo and Lactobacillus acidophilus Ki by milk hydrolyzates. Journal of Dairy Science, 81:2817{2825, 1998. [14] N. P. Shah. Probiotic bacteria: Selective enumeration and survival in dairy foods. Journal of Dairy Science, 83:894{907, 2000. [15] K. O?Riordan, D. Andrews, K. Buckle, and P. Conway. Evaluation of mi- croencapsulation of a bifldobacterium strain with starch as an approach to prolonging viability during storage. Journal of Applied Microbiology, 91:1059{ 1066, 2001. [16] C. Desmond, R. P. Ross, E. O?Callaghan, G. Fitzgerald, and C. Stanton. Improved survival of Lactobacillus paracasei NFBC 338 in spray-dried powders containing gum acacia. Journal of Applied Microbiology, 93:1003{1011, 2002. [17] K. Adhikari, A. Mustapha, and I. U. Grun. Survival and metabolic activity of microencapsulated Bifldobacterium longum in stirred yogurt. Journal of Food Science, 68(1):275{280, 2003. [18] V. Chandramouli, K. Kailasapathy, P. Peiris, and M. Jones. An improved method of microencapsulation and its evaluation to protect Lactobacillus spp. in simulated gastric conditions. Journal of Microbiological Methods, 56:27{35, 2004. [19] P. Brodelius and K. Nilsson. Entrapment of plant-cells in difierent matrices- a comparative study. FEBS Letters, 122:312{316, 1980. [20] H. Prevost and C. Divies. Cream fermentation by a mixed culture of lacto- cocci entrapped in two-layer calcium alginate gel beads. Biotechnology Letters, 14:583{588, 1992. [21] T. Y. Sheu and R. T. Marshall. Microentrapment of Lactobacilli in calcium alginate gels. Journal of Food Science, 54(3):557{561, 1993. [22] D. Roy, J. Goulet, and A. LeDuy. Continuous production of lactic acid from whey permeate by free and calcium alginate entrapped Lactobacillus helveti- cus. Journal of Dairy Science, 70:506{513, 1987. 59 [23] A. F. Groboillot, C. P. Champagne, G. D. Darling, D. Poncelet, and R. J. Neufeld. Membrane formation by interfacial cross-linking of chitosan for microencapsulation of Lactococcus lactis. Biotechnology and Bioengineering, 42:1157{1163, 1993. [24] R. Fuller. Probiotics in man and animals. Journal of Applied Bacteriology, 66:365{378, 1989. [25] W. H. Holzapfel, P. Haberer, R. Geisen, J. Bjorkroth, and U. Schilinger. Tax- onomy and important features of probiotic microorganisms in food and nutri- tion. American Journal of Clinical Nutrition, 73 (suppl):365S{373S, 2001. [26] K. Kailasapathy and J. Chin. Survival and therapeutic potential of probiotic organisms with reference to Lactobacillus acidophilus and Bifldobacterium spp. Immunology and Cell Biology, 78:80{88, 2000. [27] K. Orrhage and C.E. Nord. Bifldobacteria and Lactobacilli in human health. Drugs Under Expperimental and Clinical Resesarch, 26(3):95{111, 2000. [28] H. Nakajima, Y. Suzuki, H. Kaizu, and T. Hirota. Cholesterol lowering activity of ropy fermented milk. Journal of Food Science, 57(6):1327{1329, 1992. [29] J. Singh, A. Rivenson, M. Tomita, S. Shimamura, N. Ishibashi, and B. S. Reddy. Bifldobacterium longum, a lactic acid-producing intestinal bacterium inhibits colon cancer and modulates the intermediate biomarkers of colon car- cinogenesis. Carcinogenesis, 18(4):833{841, 1997. [30] A. C. Ouwehand and S. J. Salminen. The health efiects of cultured milk products with viable and non-viable bacteria. International Dairy Journal, 8:749{758, 1998. [31] H. S. Kim and S. Giliand. Lactobacillus acidophilus as a dietary adjunct for milk to aid lactose digestion in humans. Journal of Dairy Science, 3(4):253{ 257, 1983. [32] P.V. Kirjavainen, E. Apostolou, S.J. Salminen, and E. Isolauri. New aspects of probiotics-a novel approach in the management of food allergy. Allergy, 54:909{951, 1999. [33] W. Krasaekoopt, B. Bhandari, and H.F Deeth. Evaluation of encapsulation techniques of probiotics for yoghurt. International Dairy Journal, 13:3{13, 2003. [34] C. P. Champagne, N. Morin, R. Couture, C. Gagnon, P. Jelen, and C. Lacroix. The potential of immobilized cell technology to produce freeze-dried, phage- protected cultures of Lactococcus lactis. Food Research International, 25:417{ 427, 1992. 60 [35] S. E. Gilliand and M. L. Speck. Instability of Lactobacillus acidophilus in yogurt. Journal of Dairy Science, 60:1394{1398, 1977. [36] H. Iwana, H. Masuda, T. Fujisawa, H. Suzuki, and T. Mitsuoka. Isolation and identiflcation of bifldobacterium ssp in commercial yogurt sold in Europe. Bifldobacteria micro ora, 12:39{45, 1993. [37] W.E.V. Lankaputhra and N.P. Shah. Survival of Lactobacillus acidophilus and Bifldobacterium spp in the preence of acid and bile salts. Cultured Dairy Products Journal, 30(3):2{7, 1995. [38] K. Sultana, G. Godward, N. Reynolds, R. Arumugaswamy, P. Peiris, and K. Kailasapathy. Encapsulation of probiotic bacteria with alginate-starch and evaluation of survival in simulated gastrointestinal conditions and in yoghurt. International Journal of Food Microbiology, 62:47{55, 2000. [39] L. Kearney, M. Upton, and A. McLoughlin. Enhancing the viability of Lac- tobacillus plantarum inoculum by immobilizing the cells in calcium-alginate beads incorporating cryoprotectants. Applied and Environmental Microbiol- ogy, 56(10):3112{3116, 1990. [40] C. P. Champagne, C. Gaudy, D. Poncelet, and R. J. Neufeld. Lactococcus lactis release from calcium alginate beads. Applied and Environmental Microbiology, 58(5):1429{1434, 1992. [41] T. Jankowski, M. Zielinska, and A. Wysakowska. Encapsulation of lactic acid bacteria with alginate/starch capsules. Biotechnology Techniques, 11(1):31{34, 1997. [42] B. F. McNamee, E. D. O?Riordan, and M. O?Sullivan. Emulsiflcation and microencapsulation properties of gum arabic. Journal of Agricultural and Food Chemistry, 46:4551{4555, 1998. [43] E. Selmer-Olsen, S. E. Birkeland, and T. Sorhaug. Efiect of protective solutes on leakage from and survival of immobilized Lactobacillus subjected to drying, storage and rehydration. Journal of Applied Microbiology, 87:429{437, 1999. [44] P. Audet, C. Paquin, and C. Lacroix. Sugar utilization and acid production by free and entrapped cells of Streptococcus salivarius subsp. thermophilus, Lactobacillus delbrueckii subsp. bulgaricus, and Lactococcus lactis subsp. lac- tis in a whey permeate medium. Applied and Environmental Microbiology, 55(1):185{189, 1989. [45] K. Y Lee and T. R. Heo. Survival of Bifldobacterium longum immobilized in calcium alginate beads in simulated gastric juices and bile salt solution. Applied and Environmental Microbiology, 66(2):869{873, 2000. 61 [46] C. Paquin, M. Leroy, and C. Lacroix. Bifldobacterium longum ATCC 15707 production using free and immobilized fermentation in a whet permeaste based medium. In Proceedings of the 23rd International Dairy Congress, page 32, Brussels, belgium, 1990. International Dairy Federation. [47] P. Gemeiner, J. Nahalka, A. Vikartovska, J. Nahalkova, M. Tomaska, E. Stur- dik, O. Markovic, A. Malovikova, I. Zatkova, and M. Ilavsky. Calcium pectate gel could be a better alternative to calcium alginate gel in multiple applications of immobilized cells. In RH Wijfiels, RM Buitelaar, C Bucke, and J Tram- per, editors, Immobilized cells: Basics and Applications. Elsevier, Netherlands, 1996. [48] G. Godward and K. Kailasapathy. Viability and survival of free, encapsu- lated and co-encapsulated probiotic bacteria in yoghurt. Milchwissenschaft, 58(7/8):396{399, 2003. [49] D. Guerin, J-C. Vuillemard, and M. Subirade. Protection of Bifldobacteria encapsulated in polysaccharide-protein gel beads against gastric juice and bile. Journal of Food Protection, 66(11):2076{2084, 2003. [50] W. Krasaekoopt, B. Bhandari, and H. Deeth. The in uence of coating materi- als on some properties of alginate beads and survivability of microencapsulated probiotic bacteria. International Dairy Journal, 14:737{743, 2004. [51] W. Sun and M. W. Gri?ths. Survival of Bifldobacteria in yogurt and simulated gastric juice following immobilization in gellan-xanthan beads. International Journal of Food Micobiology, 61:17{25, 2000. [52] R. Weiner, S. Langille, and E. Quintero. Structure, function and immuno- chemistry of bacterial exopolysaccharides. Journal of Industrial Microbiology, 15:339{346, 1995. [53] P. L. Looijesteijn, L. Trapet, E. deVries, T. Abee, and J. Hugenholtz. Physi- ological function of exopolysaccharides produced from Lactococcus lactis. In- ternational Journal of Food Microbiology, 64:71{80, 2001. [54] I. W. Sutherland. Microbial exopolysaccharides from gram-negative bacteria. International Dairy Journal, 11:663{674, 2001. [55] M. Ak?celik and P. S?anl?baba. Characterisation of an exopolysaccharide pre- venting phage adsorption in Lactococcus lactis subsp. cremoris MA39. Turkish Journal of Veterinary and Animal Sciences, 26:1151{1156, 2002. [56] P. Ruas-Madiedo, J. Hugenholtz, and P. Zoon. An overview of the functional- ity of exopolysaccharides produced by lactic acid bacteria. International Dairy Journal, 12:163{171, 2002. 62 [57] P. Jannson, L. Kenne, and B. Lindberg. Structure of extracellular polysac- charide from Xanthomonas campestris. Carbohydrate Research, 45:275{282, 1975. [58] L. D. Melton, L. Mindt, D. A. Rees, and G. R. Sanderson. Covalent structure of the polysaccharide from Xanthomonas campestris: evidence from partial hydrolysis. Carbohydrate Research, 46(2):245{257, 1976. [59] I. Cotrell, K. Kang, and P. Kovacs. Handbook of water-soluble gums and resins, chapter Xanthan gum, pages 24 (1){24 (31). McGraw-Hill, New York, 1980. [60] Y. M. Lo, S. T. Yang, and D. B. Min. Efiects of yeast extract and glu- cose on xanthan production and cell growth in batch culture of Xanthomonas campestris. Applied Microbiology and Biotechnology, 47:689{694, 1997. [61] B. T. Stokke, B. E. Christensen, and O. Smidsrod. Macromolecular properties of xanthan. In S Dumitriu, editor, Polysaccharides: Structural diversity and functional versatility, pages 433{472. Marcel Dekker, Inc, USA, 1998. [62] B. Urlacher and O. Noble. Xanthan gum. In A. Imeson, editor, Thickening and gelling agents for food, pages 284{311. second edition, 1997. [63] G. Sworn. Handbook of hydrocolloids, chapter Xanthan gum, pages 103{115. CRC Press, Boca Raton, FL, 2000. [64] G. Holzwarth and E. B. Prestridge. Multistranded helix DNA in xanthan polysaccharide. Science, 197:757{759, 1977. [65] T. Sato, T. Norisuye, and H. Fujita. Double-stranded helix of xanthan: Di- mensional and hydrodynamic properties in 0.1 m aqueous sodium chloride. Macromolecules, 17(12):2696{2700, 1984. [66] H. Lee and D.A. Brant. Rheology of concentrated isotropic and anisotropic xanthan solutions.1. a rodlike low molecular weight sample. Macromolecules, 35:2212{2222, 2002a. [67] H. Lee and D.A. Brant. Rheology of concentrated isotropic and anisotropic xanthan solutions.2. a semi exible wormlike intermediate molecular weight sample. Macromolecules, 35:2223{2234, 2002b. [68] E.L. Meyer, G.G. Fuller, R.C. Clark, and W.M. Kulicke. Investigation of xanthan gum solution behavior under shear ow using rheooptical techniques. Macromolecules, 26(3):504{511, 1993. [69] G. R. Sanderson. Gums and their use in food systems. Food Technology, 50(3):81{84, 1996. [70] A. Nussinovitch. Xanthan gum. In A. Nussinovitch, editor, Hydrocolloid applications: Gum technology in the food and other industries, pages 154{168. 1997. 63 [71] F. Garcia-Ochoa, V.E. Santos, J.A. Casas, and E.Gomez. Xanthan gum: production, recovery, and properties. Biotechnology Advances, 18:549{579, 2000. [72] K. S. Kang and D. J. Pettitt. Industrial Gums: Polysaccharides and their derivatives, chapter Xanthan, gellan,wellan and rhamsan, pages 341{371. Aca- demic Press: San Diego, CA, 1993. [73] M. Milas and M. Rinaudo. Properties of xathan gum in aqueous solutions: Role of the conformational transition. Carbohydrate Research, 158:191{204, 1986. [74] W.E. Rochefort and S. Middleman. Rheology of xanthan gum: Salt, temper- ature, and strain efiects in oscillatory and steady shear experiments. Journal of Rheology, 31(4):337{369, 1987. [75] J.A. LopesdaSilva and M.A. Rao. Rheological behavior of food gel systems. In M.A. Rao, editor, Rheology of Fluid and Semisolid Foods: Principles and Applications. Aspen, Gaithersburg, 1999. [76] R. Lapasin and S. Pricl. Rheology of Industrial Polysaccharides: Theory and Applications. Blackie Academic & Professional Glasgow, 1995. [77] R.K. Richardson and S.B. Ross-Murphy. Non-linear viscoelasticity of polysac- charide solutions. 2: Xanthan polysaccharide solutions. International Journal of Biological Macromolecules, 9:257{264, 1987. [78] C.S.H. Chen and E.W. Sheppard. Conformation and shear stability of xanthan gum in solution. Polymer Engineering and Science, 20(7):512{516, 1980. [79] J.G. Southwick, A.M. Jamieson, and J. Blackwell. Conformation of xanthan dissolved in aqueous urea and sodium chloride solutions. Carbohydrate Re- search, 99(2):117{127, 1982. [80] K.C. Tam and C. Tiu. Rheology of water-soluble polymers: A comparative study on the efiect of monovalent salt. Polymer-Plastics Technology and En- gineering, 32(1&2):123{138, 1993. [81] H. Lee and D.A. Brant. Rheology of concentrated isotropic and anisotropic xanthan solutions: 3. temperature dependence. Biomacromolecules, 3:742{ 753, 2002c. [82] R. A. A. Muzzarelli. Chitin. Pergamon, Oxford, 1977. [83] P. Sandford. Chitosan: Commercial uses and potential applications. In G. Skjak-Braek, T. Anthonsen, and P. Sandford, editors, Chitin and Chitosan, pages 51{69. Elsevier Science Publishers Ltd, Northern Ireland, 1989. 64 [84] L. K. Han, Y. Kimura, and H. Okuda. Reduction in fat storage during chitin- chitosan treatment in mice fed a high-fat diet. International Journal of Obe- sity, 23:174{179, 1999. [85] N. Kubota and Y. Kikuchi. Macromolecular complexes of chitosan. In S. Du- mitriu, editor, Polysaccharides: Structural diversity and functional versatility. Marcel Dekker Inc., USA, 1998. [86] P. K. Dutta, J. Dutta, and V. S. Tripathi. Chitin and chitosan: Chemistry, properties and applications. Journal of Scientiflc and Industrial Research, 63:20{31, 2004. [87] M. N. V. R. Kumar, R. A. A. Muzzarelli, C. Muzzarelli, H. Sashiwa, and A. J. Domb. Chitosan chemistry and pharmaceutical perspectives. Chemical Reviews, 104:6017{6084, 2004. [88] H. Sashiwa and S. I. Aiba. Chemically modifled chitin and chitosan as bio- materials. Progress in Polymer Science, 29:887{908, 2004. [89] E. Kohr. Chitin: Fulfllling a Biomaterials Promise. Elsevier, Netherlands, 2001. [90] P. Sandford. High purity chitosan and alginate: Preparation, analysis and ap- plications. In R. Chandrasekaran, editor, Frontiers in Carbohydrate Research- 2, pages 250{269. Elsevier Science Publishers Ltd, 1992. [91] S. Lu, X. Song, D. Cao, Y. Chen, and K. Yao. Preparation of water-soluble chitosan. Journal of Applied Polymer Science, 91:3497{3503, 2004. [92] P. A. Felse and T. Panda. Studies on applications of chitin and its derivatives. Bioprocess Engineering, 20:505{512, 1999. [93] H. Bokura and S. Kobayashi. Chitosan decreases total cholesterol in women: a randomized, double-blind, placebo-controlled trial. European Journal of Clinical Nutrition, 57:721{725, 2003. [94] L.W. Myra. An overview of the regulatory status and of the safety of chitin and chitosan as food and pharmaceutical ingredients. In C. J. Brine, P. A. Sandford, and J. P. Zikakis, editors, Advances in Chitin and Chitosan, pages 663{670. Elsevier Science Publishers Ltd, Oxford, 1992. [95] R. A. A. Muzzarelli. Chitosan-based dietary foods. Carbohydrate Polymers, 29:309{316, 1996. [96] J. D. McCurdy. FDA and the use of chitin and chitosan derivatives. In C. J. Brine, P. A. Sandford, and J. P. Zikakis, editors, Advances in Chitin and Chitosan, pages 659{662. Elsevier Science Publishers Ltd, Oxford, 1992. 65 [97] M. L. Weiner. An overview of the regulatory status and of the safety of chitin and chitosan as food and pharmaceutical ingredients. In C. J. Brine, P. A. Sandford, and J. P. Zikakis, editors, Advances in Chitin and Chitosan, pages 663{672. Elsevier Science Publishers Ltd, Oxford, 1992. [98] A. Mart??nez, E. Chornet, and D. Rodrigue. Steady-shear rheology of concen- trated chitosan solutions. Journal of Texture Studies, 35:53{74, 2004. [99] M. Mucha. Rheological characteristics of semi-dilute chitosan solutions. Macromolecular Chemistry and Physics, 198:471{484, 1997. [100] R. H. Chen, M. L. Tsaih, and W-C. Lin. Efiects of chain exibility of chi- tosan molecules on the preparation, physical, and release characteristics of the prepared capsule. Carbohydrate Polymers, 31:141{148, 1996. [101] M. W. Anthonsen, K. M. V?arum, and O. Smidsr?d. Solution properties of chitosan: conformation and chain stifiness of chitosans with difierent degrees of n-acetylation. Carbohydrate Polymers, 22:193{201, 1993. [102] R. H. Chen, W-C. Lin, and J. H. Lin. Efiects of pH, ionic strength, and type of anion on the rheological properties of chitosan solutions. Acta Polymerica, 45:41{46, 1994. [103] D.Braun, H.Cherdron, M.Rehahn, H.Ritter, andB.Voit. Polymer Synthesis: Theory and Practice. Springer, Germany, 4th edition, 2005. [104] F. Oosawa. Polyelectrolytes. Marcel Dekker, Inc, New York, 1971. [105] W. E. Krause, J. S. Tan, and R. H. Colby. Semidilute solution rheology of polyelectrolytes with no added salt. Journal of Polymer Science: Part B: Polymer Physics, 37:3429{3437, 1999. [106] A. S. Michaels and R. G. Miekka. Polycation-polyanion complexes preparation and properties of poly-(vinylbenzyltrimethylammonium) poly- (styrenesulfonate). Journal of Physical Chemistry, 65:1765{1773, 1961. [107] J. W. Lee, S. Y. Kim, S. S. Kim, Y. M. Lee, K. H. Lee, and S. J. Kim. Synthesis and characteristics of interpenetrating polymer network hydrogel composed of chitosan and poly(acrylic acid). Journal of Applied Polymer Science, 73:113{ 120, 1999. [108] D. Thacharodi and K. P. Rao. Development and in vitro evaluation of chitosan-based transdermal drug delivery systems for the controlled delivery of propanolol hydrochloride. Biomaterials, 16:145{148, 1995. [109] V. R. Patel and M. M. Amiji. Preparation and characterization of freeze-dried chitosan-poly(ethylene oxide) hydrogels for site-speciflc antibiotic delivery in the stomach. Pharmaceutical Research, 13:588{593, 1996. 66 [110] M. Risbud, A. A. Hardikar, S. V. Bhat, and R. R. Bhonde. pH-sensitive freeze- dried chitosan-polyvinyl pyrrolidone hydrogels as controlled release system for antibiotic delivery. Journal of Controlled Release, 68:23{30, 2000. [111] M. V. Risbud and R. R. Bhonde. Polyacrylamide-chitosan hydrogels: In vitro biocompatibility and sustained antibiotic release studies. Drug Delivery, 7:69{ 75, 2000. [112] K. C. Gupta and M. N. V. R. Kumar. Studies on semi-interpenetrating poly- mer network beads of chitosan-poly(ethylene glycol) for the controlled release of drugs. Journal of Applied Polymer Science, 80:639{649, 2001. [113] D. Magnin, J. Lefebvre, E. Chornet, and S. Dumitriu. Physicochemical and structural characerization of a polyionic matrix of interest in biotechnology, in the pharmaceutical and biomedical flelds. Carbohydrate Polymers, 55:437{453, 2004. [114] J. Berger, M. Reist, J. M. Mayer, O. Felt, N. A. Peppas, and R. Gurny. Structure and interactions in covalently and ionically crosslinked chitosan hy- drogels for biomedical applications. European Journal of Pharmaceutics and Biopharmaceutics, 57:19{34, 2004. [115] C. H. Chu, T. Sakiyama, and T. Yano. pH-sensitive swelling of a polyelec- trolyte complex gel prepared from xanthan and chitosan. Bioscience Biotech- nology and Biochemistry, 59(4):717{719, 1995. [116] C. H. Chu, H. Kumagai, and K. Nakamura. Application of polyelectrolyte complex gel composed of xanthan and chitosan to the immobilization of Corynebacterium glutamicum. Journal of Applied Polymer Science, 60:1041{ 1047, 1996. [117] F. Chellat, M. Tabrizian, S. Dumitriu, S. Chornet, C. H. Rivard, and L. Yahia. Study of biodegradation behavior of chitosan-xanthan microspheres in simu- lated physiological media. Journal of Biomedical Materials Research, 53:592{ 599, 2000. [118] S. Dumitriu, P. Magny, D. Montane, P. F. Vidal, and E. Chornet. Polyionic hydrogels obtained by complexation betweeen xanthan and chitosan: Their properties as supports for enzyme immobilization. Journal of Bioactive and Compatible Polymers, 9:184{209, 1994. [119] S. Dumitriu and S. Chornet. Immobilization of xylanase in chitosan-xanthan hydrogels. Biotechnology Progress, 13:539{545, 1997. [120] A. Axelsson, C. Sisak, B. A. Westrin, and B. Szajani. Difiusional character- istics of a swelling gel and its consequences for bioreactor performance. The Chemical Engineering Journal, 55:B35{B39, 1994. 67 [121] S. Dumitriu. Polymeric biomaterials, chapter Polysaccharides as biomaterials, pages 1{61. Marcel Dekker Inc,New York, 2002. [122] N. A. Peppas, P. Bures, W. Leobandung, and H. Ichikawa. Hydrogels in pharmaceutical formulations. European Journal of Pharmaceutics and Bio- pharmaceutics, 50:27{46, 2000. [123] N. A. Peppas, M. W. Wood, and J. O. Blanchette. Hydrogels for oral delivery of therapeutic proteins. Expert Opinion on Biological Therapy, 4:881{887, 2004. [124] J. Mao, S. Kondu, H-F. Ji, and M. J. McShane. Study of the near-neutral pH- sensitivity of chitosan/gelatin hydrogels by turbidimetry and microcantilever de ection. Biotechnology and Bioengineering, 85:333{341, 2006. [125] H. Yi, L-Q Wu, R. Ghodssi, G. W. Rublofi, G. F. Payne, and W. E. Bent- ley. Signal-directed sequential assembly of biomolecules on patterned surfaces. Langmuir, 21:2104{2107, 2005. [126] S. Ikeda, H. Kumagai, T. Sakiyama, C. H. Chu, and K. Nakamura. Method for analyzing pH-sensitive swelling of amphoteric hydrogels-application to a polyelectrolyte complex gel prepared from xanthan and chitosan. Bioscience Biotechnology and Biochemistry, 59(8):1422{1427, 1995. [127] P. S. Gill, S. R. Sauerbrunn, and M. Reading. Modulated difierential scanning calorimetry. Journal of Thermal Analysis, 40:931{939, 1993. [128] M. Reading, D. Elliott, and V. L. Hill. A new approach to the calorimetric in- vestigation of physical and chemical transitions. Journal of Thermal Analysis, 40:949{955, 1993. [129] K K?ohler, H M?ohwald, and GB Sukhorukov. Thermal behavior of polyelec- trolyte multilayer microcapsules: 2.insight into molecular mechanisms for the pdadmac/pss system. Journal of Physical Chemistry B, 110:24002{24010, 2006. [130] L. Brannon-Peppas and N. A. Peppas. Solute and penetrant difiusion in swellable polymers. IX. the mechanism of drug release from ph-sensitive swelling-controlled systems. Journal of Controlled Release, 8:267{274, 1989. [131] T. Higuchi. Theoretical analysis of rate of release of solid drugs dispersed in solid matrices. Journal of Pharmaceutical Sciences, 52:1145{1149, 1963. [132] R.W.Korsmeyer, R.Gurny, E.Doelker, PBuri, andN.A.Peppas. Mechanism of solute release from porous hydrophilic polymers. Internationa Journal of Pharmaceutics, 15:25{35, 1983. 68 [133] N. A. Peppas and J. J. Sahlin. A simple equation for the description of solute release. III: Coupling of difiusion and relaxation. International Journal of Pharmaceutics, 57:169{172, 1989. [134] J. M. Llabot, R. H. Manzo, and D. A. Allemandi. Drug release from car- bomer:carbomer sodium salt matrices with potential use as mucoadhesive drug delivery system. Internationa Journal of Pharmaceutics, 276:59{66, 2004. [135] B. Kim, K. La Flamme, and N. A. Peppas. Dynamic swelling behavior of pH-sensitive anionic hydrogels used for protein delivery. Journal of Applied Polymer Science, 89:1606{1613, 2003. [136] V. Pillay and R. Fassihi. In vitro release modulation from crosslinked pellets for site-speciflc drug delivery to the gastrointestinal tract. I. comparison of pH- responsive drug release and associated kinetics. Journal of Controlled Release, 59:229{242, 1999. [137] C. Ferrero, A. Mu~noz Ruiz, and M. R. Jim?enez-Catellanos. Fronts move- ment as a useful tool for hydrophilic matrix release mechanism elucidation. Internationa Journal of Pharmaceutics, 202:21{28, 2000. [138] Jr. T. Alfrey, E. F. Gurnee, and W. G. Lloyd. Difiusion in glassy polymers. Journal of Polymer Science: Part C, 12:249{261, 1966. [139] H. B. Hopfenberg and K. C. Hsu. Swelling-controlled, constant rate delivery systems. Polymer Engineering and Science, 18:1186{1191, 1978. [140] P. L. Ritger and N. A. Peppas. A simple equation for description of solute release. II. flckian and anomalous release from swellable devices. Journal of Controlled Release, 5:37{42, 1987. [141] W. J. Simpson and H. Taguchi. The genera of lactic acid bacteria, chapter The genus Pediococcus, with notes on the genera Tetragenococcus and Aerococcus, pages 125{172. Blackie Academic and Professional, London, 1995. [142] S. Erkkil?a and Pet?aj?a E. Screening of commercial meat starter cultures at low ph and in the presence of bile salts for potential probiotic use. Meat Science, 55:297{300, 2000. [143] N. P. Guerra, P. F. Bernardez, J. Mendez, P. Cachaldora, and L. P. Castro. Production of four potentially probiotic lactic acid bacteria and their evalua- tion as feed additives for weaned piglets. Animal Feed Science and Technology, 134:89{107, 2007. [144] S. Mandal, A. K. Puniya, and K. Singh. Efiect of alginate concentrations on survival of microencapsulated Lactobacillus casei ncdc-298. International Dairy Journal, 16:1190{1195. 69 [145] S. M. Koo, Y. H. Cho, C. S. Huh, Y. J. Baek, and J. Park. Improvement of the stability of Lactobacillus casei YIT 9018 by microencapsulation using alginate and chitosan. Journal of Microbiology and Biotechnology, 11:376{383, 2001. [146] G.R. Sanderson, V.L. Bell, and D. Ortega. A comparison of gellan gum, agar, ?-carrageenan and algin. Cereal Foods World, 34:991, 994{995,997{998, 1989. [147] L. Truelstrup Hansen, P. M. Allan-Wojtas, Y. L. Jin, and A. T. Paulson. Survival of ca-alginate microencapsulated Bifldobacterium spp. in milk and simulated gastrointestinal conditions. Food Microbiology, 19:35{45, 2002. 70