Hematopoietic stem cells (HSCs) are commonly used in the treatment of blood cancers, like leukemia, and other cancers where radiation or chemotherapy damages the native HSC population. The development of a novel system to study and maintain HSCs ex vivo would give researchers and clinicians the ability to investigate the basic biological processes of HSCs, improve current treatment regimens, and explore their use in new therapies. The work in this thesis focuses on the development of a synthetic PEG hydrogel scaffold that accurately mimics aspects of the HSC microenvironment and can expand clinically relevant HSC populations. PEG hydrogel well surfaces were covalently functionalized with bioactive factors known to be critical in controlling HSC fate in vivo. In initial studies, 32D cells, a myeloid progenitor, were cultured in the wells for 6 days. On surfaces with the adhesive RGDS peptide sequence, 32D cell adhesion increased concurrently with RGDS surface concentrations. With the immobilization of two niche cytokines, SCF and SDF1α, onto hydrogel surfaces, 32D cells demonstrated significant increases in adhesion and spreading. These results confirmed that hematopoietic cell behavior could be controlled through the design of the bioactive PEG scaffold. In studies with a primary hematopoietic cell population (c-kit + , lin - ), the effects of bioactive molecules on cell expansion and differentiation were investigated after 2 weeks in culture. The adhesive peptides sequences, RGDS and CS1, and four niche proteins, SCF, SDF1α, JAG1, and IFNγ, were covalently tethered to hydrogel well surfaces. Primary cells proliferated significantly on gels containing SCF and IFNγ though only SCF was capable of preventing HSC differentiation. Cells cultured on surfaces functionalized with JAG1 and SDF1α did not proliferate extensively, but they were able to maintain primitive HSC populations. Primary c-kit + cells were also encapsulated within biodegradable PEG hydrogels and cultured for 2-5 weeks. Cells remained viable for 5 weeks in culture, and preliminary results indicated minimal cell differentiation. In this work, biomimetic PEG hydrogels were successfully employed to expand HSC populations in both two and three dimensions. The ability to generate large populations of primitive HSCs ex vivo has broad clinical and research implications.

The overarching goal of this research was to formulate an encapsulated powder using a modified lentil protein isolate-maltodextrin mixture to encapsulate flaxseed oil by freeze drying. The primary objectives were: a) to examine the physicochemical and emulsifying properties of lentil protein isolates with different degrees of hydrolysis; b) to design and test the physicochemical properties of encapsulated flaxseed oil using a wall material with native, heat treated and partially hydrolyzed lentil proteins in combination with maltodextrin; and c) test the oxidative stability of encapsulated flaxseed oil with the capsule design with the lowest surface oil and highest encapsulation efficiency versus free oil. During the first study, the physicochemical and emulsifying properties of lentil protein isolates (LPI) were investigated as a function of their degree of hydrolysis (DH of 4, 9 and 20%) following exposure to trypsin/heat. Interfacial tension, surface characteristics (charge and hydrophobicity) and intrinsic fluorescence were determined and related to changes in the emulsification activity (EAI) and stability indices (ESI) of unhydrolyzed (u-LPI) and hydrolyzed LPI (h-LPI) in a flaxseed oil-water emulsion. Most importantly surface hydrophobicity declined from ~30 to ~24 for the u-LPI and h-LPI (DH 4-20%), respectively. The changes in physicochemical properties induced by hydrolysis had a detrimental effect on EAI and ESI values, which declined from ~51 to ~47 m2 g-1 and ~12 to ~ 11 min for u-LPI and h-LPI (DH 4-20%), respectively. In the second study, the physicochemical properties of encapsulated flaxseed oil within lentil protein-based maltodextrin microcapsules were investigated using native (n-LPI), pre-treated (heated, un-hydrolyzed (u-LPI); and heated, hydrolyzed (h-LPI)) lentil protein isolates and as a function of oil load (10.0, 20.0 and 30.0% of total solids). The moisture, water activity, surface oil and encapsulation efficiency (EE) were assessed, along with droplet size and emulsion morphology. Light microscopy imaging of the emulsions, showed that the h-LPI had slightly larger oil droplets than the n-LPI and u-LPI, which both appeared similar. Microcapsules prepared from h-LPI showed significantly higher surface oil and lower EE than both the n-LPI and u-LPI materials. The microcapsules prepared using n-LPI with 10.0% oil loading were found to have the lowest surface oil content (~3.7%) and highest EE (~62.8%) for all formulations, and were subjected to an oxidative storage stability test over a 30 d period vs. free oil. The encapsulation process however induced autooxidation leading the production of a greater amount of primary oxidative products than free oil. Findings indicate that future studies are necessary to enhance the stability of the flaxseed oil through the encapsulation process.

Probiotics and prebiotic are becoming increasingly important to consumers to alleviate issues surrounding gut health, despite the lack of definitive efficacy studies to support health claims. The addition of both probiotics and prebiotics to foods is challenging due to the harsh environmental conditions within the food itself and during transit through the gastrointestinal (GI) tract. To circumvent these challenges encapsulation technology is being explored to protect sensitive ingredients and to control their release within the lower intestines thereby maximizing the health benefiting effects. The overall goal of this research was to design a protein delivery capsule using phase separated pea protein isolate (PPI)-alginate (AL) mixtures for the entrapment of the synbiot which includes the probiotics, Bifidobacterium adolescentis, and the prebiotic, fructooligosaccharides (FOS), such that the capsule design provides highly effective protection and release within the GI tract. Research was carried out in three studies.<p> In study 1, PPIn (native isolate) and AL interactions were studied in dilute aqueous solutions as a function of pH and biopolymer mixing ratio. Turbidimetric analysis and electrophoretic mobility during an acid titration was used to determine conditions where phase separation occurred. Critical structure forming events associated with the formation of soluble and insoluble complexes in a 1:1 PPIn-AL mixture were found to occur at pH 5.00 and 2.98, respectively, with optimal interactions occurring at pH 2.10. As the PPIn-AL ratio increased, critical pH values shifted towards higher pH until a mixing ratio between 4:1 and 8:1was reached, above which structure formation became independent of the ratios through to ratios of 20:1. Electrophoretic mobility measurements showed a similar trend, where the isoelectric point (pI) shifted from pH 4.00 (homogeneous PPIn) to pH 1.55 (1:1 PPIn-AL). As the ratio increased towards 8:1 PPIn-AL, net neutrality values shifted to higher pHs (~3.80) before becoming constant at higher ratios. Maximum coacervate formation occurred at a mixing ratio of 4:1. Based on these findings, capsule design by segregative phase separation was only used in future studies, due to the acidic nature associated with associative phase separation.<p> In study 2, capsule formation using a native and commercial PPI was studied, and showed no difference between the two formulations during challenge experiments in simulated gastric juice (SGJ). As a result study 3 focused on optimization and characterization of capsules prepared using the commercial PPI. Capsule designs were investigated as a function of protein concentration, prebiotic level, and extrusion conditions (20 vs. 27 G needle) in order to determine protective ability for B. adolescentis within SGJ. Capsule designs were also measured in terms of protein and prebiotic retention during the encapsulation process, geometric mean diameter and size distribution, swelling behaviour and release characteristics within simulated intestinal fluids (SIF). All capsules provided adequate protection over the 2 h duration within SGJ. Capsule breakdown and release was similar for all designs within SIF, with a release mechanism believed to be tied to enzymatic degradation of the PPI material within the wall matrix and/or the amount of excessive Na+ present in the SIF. Capsule size was found to be dependent only on the needle gauge used in the extrusion process. Swelling behaviour of the capsules with SGJ was also found to be dependent only on the protein concentration, where capsules shrank once immersed in SGJ.<p> A 2.0% PPI-0.5% AL capsule without FOS and extruded through a 20 G needle represents the best and most cost effective design for entrapping, protecting and delivering probiotic bacteria. Future work to establish the role FOS could play post-release as the entrapping probiotics colonize the GI tract, and the protective effect of the capsules wall on FOS structure during transit is recommended.

This thesis describes the development of microfluidic (MF) platforms for the study of biological and chemical processes. In particular the thesis is divided into two distinct parts: (i) development of a MF methodology to generate tunable cell-laden microenvironments for detailed studies of cell behavior, and (ii) the design and fabrication of MF reactors for studies of chemical reactions. First, this thesis presented the generation of biopolymer microenvironments for cell studies. In the first project we demonstrated a high-throughput MF system for generating cell-laden agarose microgels with a controllable ratio of two different types of cells. The MF co-encapsulation system was shown to be a robust method for identifying autocrine and/or paracrine dependence of specific cell subpopulations. In the second project we studied the effect of the mechanical properties on the behavior of acute myeloid leukemia (AML2) cancer cells. Cell-laden macroscopic agarose gels were prepared at varying agarose concentrations. A modest range of the elastic modulus of the agarose gels were achieved, ranging from 0.62 kPa to 20.21 kPa at room temperature. We observed a pronounced decrease in cell proliferation in stiffer gels when compared to the gels with lower elastic moduli. The second part of the thesis focuses on the development of MF platforms for studying chemical reactions. In the third project presented in this thesis, we exploited the temperature dependent solubility of CO2 in order to: (i) study the temperature mediated CO2 transfer between the gas and the various liquid phases on short time scales, and (ii) to generate bubbles with a dense layer of colloid particles (armoured bubbles). The fourth project involved the fabrication of a multi-modal MF device with integrated analytical probes. The MF device comprised a pH, temperature, and ATR-FTIR probes for in-situ analysis of chemical reactions in real-time. Furthermore, the MF reactor featured a temperature controlled feedback system capable of maintaining on-chip temperatures at flow rates up to 50 mL/hr.

Research was undertaken to examine different <i>in vitro</i> characteristics of probiotic bacteria, including <i>Lactobacillus acidophilus</i> ATCC® 11975, <i>Bifidobacterium infantis</i> ATCC 15697D, <i>Bifidobacterium catenulatum</i> ATCC® 27675 and <i>Bifidobacterium adolescentis</i> ATCC® 15703 in order to identify suitable strain(s) for encapsulation. Under simulated gastric conditions (pH 2.0), <i>L. acidophilus</i> was the most acid-tolerant strain (D-value 10.2 ± 0.8 min), and was able to survive for 30 min; whereas, the other tested probiotics underwent a rapid (within the first 5 min at pH 2.0) 4-5 log colony forming units (cfu)/mL loss in viability. All probiotics tested were able to survive 5 h exposure to 0.3% Oxgall bile at pH 5.8. The relative ranking of probiotic adherence to Caco-2 cells was determined to be: <i>L. acidophilus</i> > <i>B. catenulatum</i> > <i>B. adolescentis</i> > <i>B. infantis</i>, which correlated with 4.5 104, 3.1 103, 2.6 101, and 1.5 101 cfu/mL associated with Caco-2 cell monolayers, respectively. The most hydrophobic probiotics included <i>L. acidophilus</i> (46.5 ± 6.1%) and B. catenulatum (65.5 ± 5.2%); their hydrophobicity were positively correlated with auto-aggregation ability. Addition of divalent cations, EDTA, and bile salts were found to affect hydrophobicity as well; for example, 0.5 mM MgCl2 resulted in a 20% increase in cell surface hydrophobicity of <i>L. acidophilus</i> from baseline levels; whereas, the addition of 0.1 and 0.5% bile salts decreased <i>L. acidophilus</i> hydrophobicity from control levels by 60 and 90%, respectively. Cell free culture supernatant of <i>L. acidophilus</i> effectively inhibited the growth of <i>Escherichia coli</i> O157:H7, and <i>Clostridium sordelli</i>. Bactericidal activity of <i>L. acidophilus</i> cell-free supernatant (the lethal factor was determined to be both heat and trypsin-resistant) against Escherichia coli O157:H7 and <i>Clostridium sordelli</i> ATCC 9714 over 24 h resulted in reductions of 5.5 and 3.5 log cfu/mL, respectively. Further examination of probiotics revealed varying degrees of resistance to the iv antimicrobial agents ciprofloxacin (4 ìg/mL), naladixic acid (32 ìg/mL), kanamycin (64 ìg/mL) and sulfisoxazone (256 ìg/mL). Determination of carbon source utilization patterns indicated that <i>B. catenulatum</i> utilized a number of carbohydrates including -methyl-D-glucoside, D-xylose, D-cellobiose, and -D-lactose; whereas,<i>L. acidophilus, B. infantis</i>, and <i>B. adolescentis</i> utilized D-xylose. <i>Lactobacillus acidophilus</i> was ultimately selected for encapsulation in a 3 mm diameter pea protein-alginate matrix followed by <i>in vitro</i> challenge to simulated gastric conditions (pH 2.0). Encapsulation of <i>L. acidophilus</i> demonstrated a significant (P < 0.05) protective effect during the 2 h exposure to simulated acidic stomach conditions; within capsules, there was approximately 1 log cfu/mL loss in cell viability, whereas unprotected cells experienced > 6 log/mL loss in cell viability over the same period.

ABSTRACT ENCAPSULATION OF WHEAT GERM OIL Yazicioglu, Basak M.Sc., Department of Food Engineering Supervisor: Prof. Dr. Serpil Sahin Co- Supervisor: Prof. Dr. G&uuml / l&uuml / m Sumnu February 2013, 82 pages Wheat germ oil is a rich source of omega 3 and omega 6, octacosanol and tocopherol which has vitamin E activity. Due to these properties it is beneficial for health but it is prone to oxidation in free form. The aim of this study was to encapsulate wheat germ oil in micron size and determine the best encapsulation conditions by analysing encapsulation efficiency, particle size distribution and surface morphology of the capsules. The effects of core to coating ratio, coating materials ratio and ultrasonication time on encapsulation of wheat germ oil were investigated. Maltodextrin (MD) and whey protein concentrate (WPC) at different ratios (3:1, 2:2, 1:3) were used as coating materials. Total solid content of all samples was 40% (w/w). Five different core to coating ratios (1:8, 1:4, 2:4, 3:4, 4:4) were experimented. Ultrasound was used at 320 W and 20 kHz frequency for three different times (2, 5, 10 min). Prepared emulsions were frozen and then freeze dried for 48 hours to obtain microcapsules. Encapsulation efficiency analysis, particle size analysis and scanning electron microscopy (SEM) analysis were performed. Increasing WPC content in coating led to an increase in encapsulation efficiency. Microcapsules prepared with MD:WPC ratio of 1:3 were found to have higher encapsulation efficiencies (65.62%-89.62%) than the other ratios. Increase in oil load led to decrease in encapsulation efficiency thus 1:8 core to coating ratio gave better results. The best conditions for microcapsules were determined as ultrasonication time 10 min, core to coating ratio of 1:8 and MD:WPC ratio 1:3.

A novel injectable hydrogel system for cell delivery in craniofacial bone tissue engineering was developed in this work. The hydrogel employs a dual solidification mechanism by containing units that gel upon temperature increase to physiological temperature and groups that allow for covalent crosslinking. The successful synthesis of macromers for hydrogel fabrication was demonstrated and structure-property relations were established. The hydrophilic-hydrophobic balance of the macromers was found to be an important design criterion towards their resulting thermal gelation properties. When tested with cells in vitro , macromers with different molecular compositions, molecular weights and transition temperatures were all found to be cytocompatible. The introduction of a chemically crosslinkable group in the macromers resulted in hydrogels with improved stability. The effect of the addition of these highly reactive groups on cell viability was evaluated and parameters that enable viable cell encapsulation in the hydrogels were determined. It was shown that there was a dose- and time-dependent effect of the macromers on cell viability. Increased degrees of modification were found to decrease the thermal transition temperature as well as the cytocompatibility of the macromers. Hydrogels were fabricated at physiological temperature upon physical gelation and chemical crosslinking with the addition of a thermal free radical initiator system. The swelling behavior of the hydrogels was characterized and it was found to be controlled by the chemistry of the macromer end group, the concentration of the initiator system used, the fabrication interval as well as the incubation temperature and medium. In order to evaluate the hydrogels as cell carriers, mesenchymal stems cells were encapsulated in the hydrogels over a 21-day period. Cells retained their viability over the duration of the study and exhibited markers of osteogenic differentiation when cultured with appropriate supplements. These findings hold promise for the use of these hydrogel systems for cell encapsulation in tissue engineering applications.

Research was undertaken to examine different <i>in vitro</i> characteristics of probiotic bacteria, including <i>Lactobacillus acidophilus</i> ATCC® 11975, <i>Bifidobacterium infantis</i> ATCC 15697D, <i>Bifidobacterium catenulatum</i> ATCC® 27675 and <i>Bifidobacterium adolescentis</i> ATCC® 15703 in order to identify suitable strain(s) for encapsulation. Under simulated gastric conditions (pH 2.0), <i>L. acidophilus</i> was the most acid-tolerant strain (D-value 10.2 ± 0.8 min), and was able to survive for 30 min; whereas, the other tested probiotics underwent a rapid (within the first 5 min at pH 2.0) 4-5 log colony forming units (cfu)/mL loss in viability. All probiotics tested were able to survive 5 h exposure to 0.3% Oxgall bile at pH 5.8. The relative ranking of probiotic adherence to Caco-2 cells was determined to be: <i>L. acidophilus</i> > <i>B. catenulatum</i> > <i>B. adolescentis</i> > <i>B. infantis</i>, which correlated with 4.5 104, 3.1 103, 2.6 101, and 1.5 101 cfu/mL associated with Caco-2 cell monolayers, respectively. The most hydrophobic probiotics included <i>L. acidophilus</i> (46.5 ± 6.1%) and B. catenulatum (65.5 ± 5.2%); their hydrophobicity were positively correlated with auto-aggregation ability. Addition of divalent cations, EDTA, and bile salts were found to affect hydrophobicity as well; for example, 0.5 mM MgCl2 resulted in a 20% increase in cell surface hydrophobicity of <i>L. acidophilus</i> from baseline levels; whereas, the addition of 0.1 and 0.5% bile salts decreased <i>L. acidophilus</i> hydrophobicity from control levels by 60 and 90%, respectively. Cell free culture supernatant of <i>L. acidophilus</i> effectively inhibited the growth of <i>Escherichia coli</i> O157:H7, and <i>Clostridium sordelli</i>. Bactericidal activity of <i>L. acidophilus</i> cell-free supernatant (the lethal factor was determined to be both heat and trypsin-resistant) against Escherichia coli O157:H7 and <i>Clostridium sordelli</i> ATCC 9714 over 24 h resulted in reductions of 5.5 and 3.5 log cfu/mL, respectively. Further examination of probiotics revealed varying degrees of resistance to the iv antimicrobial agents ciprofloxacin (4 ìg/mL), naladixic acid (32 ìg/mL), kanamycin (64 ìg/mL) and sulfisoxazone (256 ìg/mL). Determination of carbon source utilization patterns indicated that <i>B. catenulatum</i> utilized a number of carbohydrates including -methyl-D-glucoside, D-xylose, D-cellobiose, and -D-lactose; whereas,<i>L. acidophilus, B. infantis</i>, and <i>B. adolescentis</i> utilized D-xylose. <i>Lactobacillus acidophilus</i> was ultimately selected for encapsulation in a 3 mm diameter pea protein-alginate matrix followed by <i>in vitro</i> challenge to simulated gastric conditions (pH 2.0). Encapsulation of <i>L. acidophilus</i> demonstrated a significant (P < 0.05) protective effect during the 2 h exposure to simulated acidic stomach conditions; within capsules, there was approximately 1 log cfu/mL loss in cell viability, whereas unprotected cells experienced > 6 log/mL loss in cell viability over the same period.

Probiotics and prebiotic are becoming increasingly important to consumers to alleviate issues surrounding gut health, despite the lack of definitive efficacy studies to support health claims. The addition of both probiotics and prebiotics to foods is challenging due to the harsh environmental conditions within the food itself and during transit through the gastrointestinal (GI) tract. To circumvent these challenges encapsulation technology is being explored to protect sensitive ingredients and to control their release within the lower intestines thereby maximizing the health benefiting effects. The overall goal of this research was to design a protein delivery capsule using phase separated pea protein isolate (PPI)-alginate (AL) mixtures for the entrapment of the synbiot which includes the probiotics, Bifidobacterium adolescentis, and the prebiotic, fructooligosaccharides (FOS), such that the capsule design provides highly effective protection and release within the GI tract. Research was carried out in three studies.<p> In study 1, PPIn (native isolate) and AL interactions were studied in dilute aqueous solutions as a function of pH and biopolymer mixing ratio. Turbidimetric analysis and electrophoretic mobility during an acid titration was used to determine conditions where phase separation occurred. Critical structure forming events associated with the formation of soluble and insoluble complexes in a 1:1 PPIn-AL mixture were found to occur at pH 5.00 and 2.98, respectively, with optimal interactions occurring at pH 2.10. As the PPIn-AL ratio increased, critical pH values shifted towards higher pH until a mixing ratio between 4:1 and 8:1was reached, above which structure formation became independent of the ratios through to ratios of 20:1. Electrophoretic mobility measurements showed a similar trend, where the isoelectric point (pI) shifted from pH 4.00 (homogeneous PPIn) to pH 1.55 (1:1 PPIn-AL). As the ratio increased towards 8:1 PPIn-AL, net neutrality values shifted to higher pHs (~3.80) before becoming constant at higher ratios. Maximum coacervate formation occurred at a mixing ratio of 4:1. Based on these findings, capsule design by segregative phase separation was only used in future studies, due to the acidic nature associated with associative phase separation.<p> In study 2, capsule formation using a native and commercial PPI was studied, and showed no difference between the two formulations during challenge experiments in simulated gastric juice (SGJ). As a result study 3 focused on optimization and characterization of capsules prepared using the commercial PPI. Capsule designs were investigated as a function of protein concentration, prebiotic level, and extrusion conditions (20 vs. 27 G needle) in order to determine protective ability for B. adolescentis within SGJ. Capsule designs were also measured in terms of protein and prebiotic retention during the encapsulation process, geometric mean diameter and size distribution, swelling behaviour and release characteristics within simulated intestinal fluids (SIF). All capsules provided adequate protection over the 2 h duration within SGJ. Capsule breakdown and release was similar for all designs within SIF, with a release mechanism believed to be tied to enzymatic degradation of the PPI material within the wall matrix and/or the amount of excessive Na+ present in the SIF. Capsule size was found to be dependent only on the needle gauge used in the extrusion process. Swelling behaviour of the capsules with SGJ was also found to be dependent only on the protein concentration, where capsules shrank once immersed in SGJ.<p> A 2.0% PPI-0.5% AL capsule without FOS and extruded through a 20 G needle represents the best and most cost effective design for entrapping, protecting and delivering probiotic bacteria. Future work to establish the role FOS could play post-release as the entrapping probiotics colonize the GI tract, and the protective effect of the capsules wall on FOS structure during transit is recommended.

Sol-gel processing routes for encapsulation of double stranded DNA in solid porous silica hosts have been established. The encapsulation was carried out in two steps: hydrolysis of a silica-forming alkoxide-based sol was followed by condensation/gelation to a solid form upon addition of a buffer solution containing DNA molecules. The effects of gelation chemistry and DNA amount on chemical and microstructural properties of resultant silica matrices and on DNA encapsulation efficiency were investigated. The analytical characterization was performed by UV-vis spectroscopy, 29Si nuclear magnetic resonance spectroscopy and by nitrogen adsorption studies. It was demonstrated that DNA incorporation had a pH-dependent catalytic effect on gelation kinetics and promoted silica network completion. In addition, the scale of porosity and the average pore size of the resultant silica increased with gelation pH and also with DNA-buffer solution in the starting sol-gel formulation. The chemistry-derived pore size variation controls the DNA encapsulation efficiency in the silica matrices and the DNA holding capacity strongly depends on the scale of the porosity attained. The selective adsorption of ethidium bromide- a DNA-intercalative reagent molecule- on DNA-silica gels confirmed that the DNA molecules remained entrapped within the silica host in their native state without any deterioration. Besides pure silica, amine-functionalized hybrid silica hosts were also formed by sol-gel. The hybrid gels were found not to be suitable for DNA encapsulation, as these matrices dissolve in aqueous environment due to incomplete silica network formation. The DNA-doped silica hosts may provide promising matrices for development of biosensors, bioreactors and bioassay platforms.