Ana Cristina Braga Pinheiro

Universidade do Minho Escola de Engenharia

Ana Cristina Braga Pinheiro behaviour in vitro heir Development of nanostructures for tructures for application in food application in food aluation of t v technology – evaluation of their in vitro behaviour velopment of nanos De technology – e o aga Pinheir Ana Cristina Br 3 1

UMinho|20 August, 2013

Universidade do Minho Escola de Engenharia

Ana Cristina Braga Pinheiro

Development of nanostructures for application in food technology – evaluation of their in vitro behaviour

Dissertation for PhD degree in Chemical and Biological Engineering

Supervisors of the thesis Professor António Augusto Vicente Professor Manuel A. Coimbra

August, 2013

AGRADECIMENTOS

Durante este percurso muitos foram os que me ajudaram e contribuíram para que eu chegasse ao fim desta etapa e aos quais não poderia deixar de agradecer.

Em primeiro lugar, os meus agradecimentos vão para o meu orientador, Professor António Vicente, pela confiança que depositou em mim, por todo o apoio, disponibilidade e amizade. As suas excepcionais qualidades científicas e humanas foram sem dúvida determinantes para a concretização deste trabalho. De igual modo agradeço ao meu co-orientador, Professor Manuel Coimbra, por todo o apoio científico, pela confiança que depositou em mim e pela disponibilidade demonstrada.

À instituição de acolhimento, Centro de Engenharia Biológica da Universidade do Minho pela disponibilização do espaço e equipamento necessários à realização deste trabalho. Agradeço ao Sr. Santos e à Engª Madalena Vieira o apoio técnico e toda a disponibilidade demonstrada sempre que necessário.

Aos meus colegas e amigos do LIP agradeço o espírito de entreajuda, a partilha de ideias, a constante boa disposição e o excelente ambiente de trabalho que me proporcionaram! Agradeço em especial à Isa, ao Miguel e ao Hélder pela colaboração activa neste trabalho. Sem dúvida que pertencer a uma equipa tão dinâmica tornou a realização deste trabalho uma tarefa muito mais fácil! Gostaria também de expressar uma palavra de agradecimento à Mafalda Quintas e à Professora Graça que muito contribuíram para o desenvolvimento deste trabalho.

Aos colegas do departamento de química da Universidade de Aveiro, em especial à Cláudia e à Élia, agradeço a ajuda na análise dos polissacarídeos.

I would also like to express my gratitude to Dr. Mike Boland for giving me opportunity to do part of my work at Riddet Institute in New Zealand and to Dr. Mita Lad for being so supportive and helpful! I am thankful to all the staff and research fellows in Riddet Institute for receiving so well, especially to Amit and Janiene for the technical support and to Natascha, Ajit and Noelia for making my passage through the beautiful New Zealand so much pleasant!

Um obrigado muito especial ao Bruno pelo amor, companheirismo e por me incentivar a ser sempre melhor! Obrigada por estares sempre presente!

Finalmente agradeço à minha Família, especialmente aos meus pais e irmã, a quem devo tudo o que sou, pelo incentivo e apoio que sempre me deram!

iii

This thesis was financially supported by a PhD scholarship from Fundação para a Ciência e Tecnologia (Ref.: SFRH/BD/48120/2008), inserted in the Programa Potencial Humano Quadro de Referência Estratégico Nacional (POPH - QREN) Tipologia 4.1 – Formação Avançada. The POPH-QREN is co-financed by Fundo Social Europeu (FSE) and by Ministério da Ciência, Tecnologia e Ensino Superior (MCTES).

ABSTRACT

The emerging field of nanotechnology offers new challenges to the food industry either by offering novel tools for the development of strategies to improve food quality and human health, or by the introduction of questions about the behaviour of nanostructures within the human body. Nanotechnology holds a great potential to generate very innovative solutions and to provide food technologists and manufacturers with instruments to meet the ever- growing consumer demands in very diverse aspects related with the foods they eat: safety, quality, health-promotion and novelty. However, the application of nanostructures to foods is hindered by very pertinent problems, which could be summarized in two issues: edibility (only edible materials must be used for their production) and functionality/behaviour once inside the human body, that is raising safety concerns among the consumers, and therefore demands an evaluation (ideally) in vivo, or at least in vitro.

In this context, the two main challenges addressed in this thesis were to develop stable nanostructures for food applications and to evaluate their in vitro behaviour. The strategy adopted included the development and characterization of edible nanostructures, incorporation of bioactive compounds and evaluation of their behaviour when subjected to digestion in artificial gastrointestinal (GI) systems. In particular, the research undertaken was based on three different nanostructures: nanofilms composed of κ-carrageenan and chitosan, curcumin nanoemulsions stabilized by different emulsifiers and multilayer nanocapsules composed of chitosan and fucoidan.

The interactions between κ-carrageenan and chitosan in nanofilms have been investigated and the results showed that κ-carrageenan/chitosan interaction is an exothermic process and that the alternate deposition of κ-carrageenan and chitosan results in the formation of a stable multilayer structure mainly due to the electrostatic interactions existing between the two polyelectrolytes. The κ-carrageenan/chitosan nanolayers exhibit good gas barrier properties and therefore offer great potential to be used e.g. to coat food systems.

Also, the model cationic compound methylene blue (MB) was incorporated in different positions of the κ-carrageenan/chitosan nanolayered coating and its loading and in vitro release behaviour were evaluated. Loading results suggest that MB was able to diffuse into the κ-carrageenan/chitosan nanolayered coating and not only adhered to the surface of the layer immediately below it. For most of the tested conditions, MB release from the κ- carrageenan/chitosan nanolayered coatings was successfully described by the linear superimposition model, which allowed concluding that MB transport is due to both

concentration gradient and polymer relaxation in the nanolayers. However, depending on temperature and pH of the medium and on the position of MB incorporated on the nanolayered coatings, different mechanisms prevailed. These results lead to conclude that the κ-carrageenan/chitosan nanolayered coatings represent a promising platform from which the controlled release of different bioactive compounds may be explored.

Regarding curcumin nanoemulsions, they were prepared using four different emulsifiers: Tween 20 (non-ionic), Sodium Dodecyl Sulphate (SDS, anionic), Dodecyltrimethylammonium Bromide (DTAB, cationic) and lactoferrin (protein). Their formulation was optimized and their stability was evaluated during 90 days. DTAB-stabilized emulsions revealed to be the least stable, showing a phase separation, the highest colour change and the highest decrease in the magnitude of ζ-potential. Nanoemulsions stabilized by lactoferrin revealed to be at least as stable as nanoemulsions stabilized by synthetic emulsifiers, thus showing the great potential of lactoferrin to substitute synthetic emulsifiers, which is in line with the current trend of food industry.

A human gastric simulator was used as in vitro digestion model (in which stomach, duodenum, jejunum and ileum steps were performed) to evaluate the impact of emulsifier (Tween 20, SDS and DTAB) charge on the digestion of curcumin nanoemulsions. The emulsifier charge had a significant effect on droplet size, particle electric charge and microstructure of curcumin nanoemulsions during the simulated digestion, which consequently influenced free fatty acids release and curcumin bioavailability. Results showed the positively charged DTAB-stabilized emulsions to be the least stable during the digestion process, exhibiting the largest increase in droplet size and eventual phase separation. This also contributed to the observed low bioavailability of curcumin. Conversely, emulsions stabilized with Tween 20 showed retention of emulsion structure (high surface area) and greater free fatty acid production, which could explain increased curcumin bioavailability.

A dynamic digestion model, comprising the simulation of stomach, duodenum, jejunum and ileum, was developed and used to evaluate the impact of alginate coating on the digestion of curcumin nanoemulsions stabilized by lactoferrin. Results suggested that alginate coating may be able to control the rate of lipid digestion and free fatty acids adsorption within the GI tract, but encapsulated lipids are digested to the same extent, releasing the lipophilic bioactive compound.

Biodegradable hollow nanocapsules were developed through alternate deposition of 10 chitosan/fucoidan layers on polystyrene (PS) nanoparticles, used as templates, followed by vi

removal of the PS core. Poly-L-lysine (PLL) was entrapped in the nanocapsules and its loading and release behaviour was evaluated. Chitosan/fucoidan nanocapsules showed a good capacity for the encapsulation and loading of PLL and the results of fitting the linear superimposition model to the experimental data of PLL release suggested an anomalous release behaviour, with one main polymer relaxation and PLL release was found to be pH- dependent. Due to their bioactive and non-toxic nature and to their responsiveness at different pHs, chitosan/fucoidan hollow nanocapsules showed great potential to act as a controlled delivery system for bioactive compounds.

In conclusion, the nanostructures developed in this work can be used as platforms for the production of new products with improved characteristics targeted at the most recent consumer trends. This work contributes to the understanding of the behaviour of those nanostructures inside the human body during digestion (e.g. release phenomena involved at the nano-scale and bioavailability of lipophilic compounds during in vitro digestion), which is important to determine their functional performance, aiming at the optimization of nano- based delivery systems to protect and release bioactive compounds. Therefore, the outcome of this thesis is a very positive contribution to the evolution of the state-of-art on the application of nanotechnology to the food sector.

vii

RESUMO

A nanotecnologia é uma área emergente que oferece novos desafios à indústria alimentar, quer pela oferta de novas ferramentas para o desenvolvimento de estratégias para melhorar a qualidade dos alimentos e a saúde humana, quer pela introdução de questões sobre o comportamento das nanoestruturas dentro do corpo humano. A nanotecnologia tem um elevado potencial para gerar soluções inovadoras e para fornecer instrumentos que permitem ir ao encontro das exigências dos consumidores, em diversos aspetos relacionados com os alimentos: segurança: qualidade, promoção de saúde e novidade. Contudo, a aplicação de nanoestruturas em alimentos está limitada por problemas bastante pertinentes que se enquadram em duas temáticas: comestibilidade (apenas materiais comestíveis podem ser usados para a sua produção) e funcionalidade/comportamento dentro do corpo humano (questão que está a gerar preocupação entre os consumidores, exigindo, desta forma, uma avaliação in vivo (idealmente) ou pelo menos in vitro).

Neste contexto, os dois principais desafios propostos nesta tese são o desenvolvimento de nanoestruturas estáveis para aplicações alimentares e a avaliação do seu comportamento in vitro. A estratégia adotada incluiu o desenvolvimento e caracterização de nanoestruturas, incorporação de compostos bioativos e avaliação do seu comportamento quando submetidos a uma digestão num sistema gastrointestinal artificial. Em particular, a investigação realizada baseou-se em três nanoestruturas diferentes: filmes nanolaminados compostos por κ- carragenato e quitosana, nanoemulsões de curcumina, estabilizadas por diferentes emulsificantes e nanocápsulas multicamadas compostas por quitosana e fucoidana.

As interações entre o κ-carragenato e a quitosana em filmes nanolaminados foram investigadas e os resultados demostraram que a interação κ-carragenato/quitosana é um processo exotérmico e que a deposição alternada de κ-carragenato e quitosana resulta na formação de uma estrutura multicamadas estável, principalmente devido às interações eletrostáticas entre os dois polieletrólitos. O filme nanolaminado de κ-carragenato/quitosana exibe boas propriedades de barreira aos gases e por isso apresenta elevado potencial para ser usado como revestimento em sistemas alimentares.

Para além disso, incorporou-se um composto catiónico modelo, azul de metileno (MB) em diferentes posições do filme nanolaminado de κ-carragenato/quitosana e avaliou-se o seu comportamento de adesão e libertação in vitro. Os resultados da adesão do MB sugerem que este composto se difundiu no filme nanolaminado em vez de aderir à superfície da camada imediatamente abaixo desta. Para a maior parte das condições testadas, a libertação do MB

dos filmes nanolaminados de κ-carragenato/quitosana foi descrita com sucesso pelo modelo da sobreposição linear, que permitiu concluir que o transporte do MB ocorre devido ao gradiente de concentração e à relaxação das nanocamadas. Contudo, dependendo da temperatura e do pH do meio e da posição do MB no filme nanolaminado, podem prevalecer diferentes mecanismos. Estes resultados permitem concluir que o filme nanolaminado de κ- carragenato/quitosana representa uma plataforma de trabalho promissora a partir da qual se pode explorar a libertação de diferentes compostos bioativos.

Em relação às nanoemulsões de curcumina, preparam-se usando quatro emulsificantes diferentes: Tween 20 (não iónico), Dodecil sulfato de sódio (SDS, aniónico), brometo de dodeciltrimetilamónio (DTAB, catiónico) e lactoferrina (proteína). Otimizaram-se as formulações das nanoemulsões e avaliou-se a sua estabilidade durante 90 dias. A nanoemulsão estabilizada pelo DTAB revelou ser a menos estável, pois verificou-se separação de fases e sofreu a maior alteração de cor e o maior decréscimo no valor absoluto do potencial zeta. As nanoemulsões estabilizadas com lactoferrina revelaram ser pelo menos tão estáveis como as estabilizadas pelos emulsificantes sintéticos, o que demonstra o grande potencial da lactoferrina em substituir os emulsificantes sintéticos, que é a tendência atual da indústria alimentar.

De forma a avaliar o impacto da carga do emulsificante (Tween 20, SDS e DTAB) na digestão das nanoemulsões de curcumina, utilizou-se um digestor gástrico como modelo de digestão in vitro (no qual se simularam o estômago, o duodeno, o jejuno e o íleo). A carga do emulsificante teve um impacto significativo no tamanho, carga elétrica e microestrutura das nanoemulsões de curcumina durante a simulação da digestão, influenciando consequentemente a libertação de ácidos gordos livres e a biodisponibilidade da curcumina. Os resultados demostraram que as emulsões estabilizadas pelo DTAB (carregado positivamente) são as menos estáveis durante o processo de digestão, pois exibiram o maior aumento de tamanho e separação de fases. Isto contribuiu para a baixa biodisponibilidade da curcumina. Pelo contrário, as emulsões estabilizadas com Tween 20 preservaram a sua estrutura (elevada área de superfície) e verificou-se uma maior produção de ácidos gordos livres, o que pode explicar a elevada biodisponibilidade da curcumina.

No decurso do trabalho foi desenvolvido um modelo de digestão dinâmico, que inclui a simulação do estômago, duodeno, jejuno e íleo, utilizado para avaliar o impacto do revestimento de alginato na digestão das emulsões de curcumina estabilizadas por lactoferrina. Os resultados obtidos sugerem que o revestimento de alginato pode ser capaz de controlar a taxa de digestão dos lípidos e a adsorção de ácidos gordos livres no tracto x

intestinal, contudo os lípidos encapsulados são digeridos na mesma extensão, libertando o composto bioativo lipofílico.

Desenvolveram-se nanocápsulas ocas biodegradáveis através da deposição alternada de 10 camadas de quitosana/fucoidana em nanopartículas de poliestireno (PS), usadas como suporte, seguido pela remoção do núcleo de PS. A poli-L-lisina (PLL) foi aprisionada nas nanocápsulas e o seu comportamento de adesão e libertação foi avaliado. As nanocápsulas de quitosana/fucoidana revelaram uma boa capacidade para encapsular a PLL e os resultados de ajuste do modelo de sobreposição linear aos dados experimentais da libertação da PLL sugerem um comportamento de libertação anómalo, com uma relaxação principal do polímero, sendo a libertação da PLL dependente do pH. Devido sua à natureza bioativa e não- tóxica e à sua capacidade de resposta a diferentes pHs, as nanocápsulas de quitosana/fucoidana demostram elevado potencial para ser usadas como sistema de libertação controlada para compostos bioativos.

Em conclusão, as nanoestruturas desenvolvidas neste trabalho podem ser usadas como plataformas para a produção de novos produtos com características melhoradas e focalizadas nas mais recentes tendências dos consumidores. Este trabalho contribuiu para a compreensão do comportamento de tais nanoestruturas dentro do corpo humano durante a digestão (ex. fenómeno de libertação envolvida à escala nano e biodisponibilidade de compostos lipofílicos durante a digestão), que é importante para determinar o seu desempenho funcional, visando a otimização dos sistemas de libertação à escala nano como veículos para proteger e libertar compostos bioativos. Desta forma, os resultados desta tese são um contributo bastante positivo para a evolução do estado-da-arte da aplicação da nanotecnologia ao setor alimentar.

TABLE OF CONTENTS

SECTION I. INTRODUCTION ...... 1

1 CHAPTER 1. MOTIVATION AND OUTLINE ...... 3

1.1 Thesis motivation ...... 5

1.2 Research aims ...... 7

1.3 Thesis outline ...... 7

1.4 References ...... 9

2 CHAPTER 2. LITERATURE REVIEW ...... 11

2.1 Food Nanotechnology ...... 13

2.2 Biobased materials ...... 13

2.2.1 Polysaccharides ...... 14

2.2.1.1 Carrageenans ...... 15

2.2.1.2 Alginate ...... 16

2.2.1.3 Fucoidans...... 17

2.2.1.4 Chitosan ...... 18

2.2.2 Proteins ...... 19

2.2.2.1 Lactoferrin ...... 20

2.2.3 Lipids...... 21

2.3 Development of multilayer nanostructures through the layer-by-layer technique 22

2.3.1 Planar substrates ...... 23

2.3.2 Colloidal substrates ...... 23

2.4 Nanostructures ...... 25

2.4.1 Nanofilms ...... 25

2.4.1.1 Release of bioactive compounds ...... 25

2.4.1.2 Application in food systems ...... 26

2.4.2 Nanoparticles ...... 27

2.4.2.1 Hollow polyelectrolyte multilayer nanocapsules ...... 28

2.4.3 Nanoemulsions ...... 29

xiii

2.4.3.1 Production methods ...... 30

2.4.3.2 Emulsifiers ...... 31

2.4.3.3 Lipophilic bioactive compounds ...... 34

2.4.3.4 Multilayer nanoemulsions ...... 35

2.5 Nanostructures characterization ...... 37

2.5.1 Physicochemical Characterization Techniques ...... 37

2.5.1.1 Dynamic Light Scattering ...... 37

2.5.1.2 ζ-potential...... 37

2.5.1.3 Quartz Crystal Microbalance ...... 38

2.5.1.4 Ultraviolet–Visible Spectroscopy ...... 39

2.5.1.5 Contact Angle ...... 39

2.5.1.6 Fourier Transform Infrared ...... 39

2.5.2 Imaging Techniques ...... 39

2.5.2.1 Transmission Electron Microscopy ...... 40

2.5.2.2 Scanning Electron Microscopy ...... 40

2.5.2.3 Confocal Laser Scanning Microscopy ...... 40

2.5.3 Transport Properties ...... 41

2.5.4 Controlled Release ...... 41

2.6 In vitro digestions ...... 43

2.6.1 Human gastrointestinal digestion ...... 43

2.6.2 In vitro digestion models ...... 45

2.6.2.1 pH-stat ...... 46

2.6.2.2 Dynamic Gut Model (DGM) ...... 46

2.6.2.3 Human Gastric Simulator (HGS) model ...... 47

2.6.2.4 TIM-1 model ...... 48

2.6.3 Behaviour of nanostructures in the GI tract ...... 49

2.7 References ...... 52

SECTION II. NANOFILMS ...... 67 xiv

3 CHAPTER 3. INTERACTIONS BETWEEN κ-CARRAGEENAN AND CHITOSAN IN NANOLAYERED COATINGS – STRUCTURAL AND TRANSPORT PROPERTIES EVALUATION ...... 69

3.1 Introduction...... 71

3.2 Materials and Methods ...... 72

3.2.1 Preparation of the polyelectrolyte solutions ...... 72

3.2.2 PET aminolysis ...... 73

3.2.3 Evaluation of the interactions between the polyelectrolytes ...... 73

3.2.3.1 Electrostatic properties ...... 73

3.2.3.2 Microcalorimetry analyses ...... 73

3.2.3.3 Quartz crystal microbalance analyses ...... 74

3.2.4 Preparation of nanolayered coating ...... 74

3.2.5 Characterization of the nanolayered coating ...... 75

3.2.5.1 UV/VIS spectroscopy ...... 75

3.2.5.2 Contact angle measurements ...... 75

3.2.5.3 Scanning electron microscopy (SEM) ...... 75

3.2.5.4 Water vapour permeability (WVP) measurements ...... 76

3.2.5.5 Oxygen permeability ...... 76

3.2.6 Evaluation of the behaviour of the κ-carrageenan/chitosan nanolayered coating during in vitro digestion ...... 77

3.2.6.1 Statistical Analyses ...... 78

3.3 Results and discussion ...... 79

3.3.1 Evaluation of the interactions between κ-carrageenan and chitosan ...... 79

3.3.1.1 Electrostatic properties of the polyelectrolytes ...... 79

3.3.1.2 Microcalorimetry analyses ...... 79

3.3.1.3 Quartz crystal microbalance analysis ...... 84

3.3.2 Characterization of the nanolayered coating ...... 85

3.3.2.1 UV/VIS spectroscopy ...... 85

3.3.2.2 Contact angle measurements ...... 86

3.3.2.3 Scanning electron microscopy (SEM) ...... 87

3.3.2.4 Gas barrier properties ...... 88

3.3.3 Evaluation of the behaviour of the κ-carrageenan/chitosan nanolayered coating during in vitro digestion ...... 90

3.4 Conclusions...... 91

3.5 References ...... 93

4 CHAPTER 4. Κ-CARRAGEENAN/CHITOSAN NANOLAYERED COATING FOR CONTROLLED RELEASE OF A MODEL BIOACTIVE COMPOUND ...... 97

4.1 Introduction...... 99

4.2 Materials and Methods ...... 100

4.2.1 Experimental Procedures ...... 100

4.2.1.1 Preparation of nanolayered coating ...... 100

4.2.1.2 MB loading ...... 102

4.2.1.3 MB release ...... 102

4.2.2 Statistical Procedures ...... 102

4.3 Results and Discussion ...... 103

4.3.1 MB loading ...... 103

4.3.2 MB release ...... 105

4.3.2.1 Assessment of Fick´s transport in nanolayered coatings ...... 107

4.3.2.2 Anomalous transport in nanolayered coatings ...... 107

4.3.2.3 Other mechanisms of transport in nanolayered coatings...... 111

4.4 Conclusions...... 112

4.5 References ...... 113

SECTION III. NANOEMULSIONS ...... 117

5 CHAPTER 5. DEVELOPMENT OF CURCUMIN NANOEMULSION-BASED DELIVERY SYSTEMS . 119

5.1 Introduction...... 121

5.2 Materials and Methods ...... 122

5.2.1 Materials ...... 122 xvi

5.2.2 Emulsion preparation ...... 122

5.2.3 Emulsion characterization ...... 123

5.2.3.1 Particle size measurements ...... 123

5.2.3.2 ζ-potential measurements ...... 123

5.2.3.3 Transmission Electron Microscopy (TEM) ...... 123

5.2.3.4 Colour evolution ...... 123

5.2.3.5 Statistical Analyses ...... 124

5.3 Results and Discussion ...... 124

5.3.1 Development of the curcumin nanoemulsions ...... 124

5.3.1.1 Effect of emulsifier concentration on particle size ...... 124

5.3.1.2 Effect of oil/emulsifier solution ratio on particle size ...... 125

5.3.1.3 Particle charge ...... 126

5.3.1.4 Morphology ...... 127

5.3.2 Storage stability of curcumin nanoemulsions ...... 127

5.3.2.1 Particle size ...... 128

5.3.2.2 Particle charge ...... 129

5.3.2.2 Colour evolution ...... 130

5.4 Conclusions...... 131

5.5 References ...... 133

6 CHAPTER 6. UNRAVELLING THE BEHAVIOUR OF CURCUMIN NANOEMULSIONS DURING IN VITRO DIGESTION: IMPACT OF SURFACE CHARGE ...... 135

6.1 Introduction...... 137

6.2 Materials and methods ...... 139

6.2.1 Materials ...... 139

6.2.2 Emulsion preparation ...... 139

6.2.3 In vitro digestion ...... 139

6.2.4 Particle size ...... 140

6.2.5 ζ-potential ...... 141

xvii

6.2.6 Confocal microscopy ...... 141

6.2.7 Free fatty acids released ...... 141

6.2.8 Curcumin bioavailability ...... 142

6.2.9 Preparation of multilayer emulsions ...... 142

6.2.10 Statistical analyses...... 143

6.3 Results and Discussion ...... 143

6.3.1 Influence of digestion on particle size ...... 143

6.3.2 Influence of digestion on emulsion microstructure ...... 146

6.3.3 Influence of digestion on droplet charge ...... 148

6.3.4 In vitro digestibility ...... 149

6.3.5 In vitro curcumin bioavailability ...... 150

6.3.6 Curcumin nanoemulsions stabilized by a multilayer coating ...... 151

6.4 Conclusions...... 152

6.5 References ...... 153

7 CHAPTER 7. INFLUENCE OF INTERFACIAL COMPOSITION ON IN VITRO DIGESTIBILITY OF CURCUMIN NANOEMULSIONS STABILIZED BY BIOPOLYMER EMULSIFIERS ...... 157

7.1 Introduction...... 159

7.2 Materials and Methods ...... 161

7.2.1 Materials ...... 161

7.2.2 Emulsion preparation ...... 161

7.2.3 In vitro digestion ...... 161

7.2.3.1 Gastro-intestinal model ...... 161

7.2.3.2 Experimental Conditions ...... 163

7.2.4 Particle size ...... 163

7.2.5 ζ-potential ...... 164

7.2.6 Transmission Electron Microscopy (TEM) ...... 164

7.2.7 Free fatty acids released ...... 164

7.2.8 Curcumin bioavailability ...... 165 xviii

7.2.9 Statistical analyses...... 165

7.3 Results and discussion ...... 165

7.3.1 Optimization of the secondary emulsion preparation ...... 165

7.3.2 In vitro digestion ...... 167

7.3.2.1 Influence of alginate on particle characteristics during in vitro digestion 167

7.3.2.2 Influence of alginate on lipid digestion ...... 170

7.3.2.3 Influence of alginate on curcumin bioavailability ...... 172

7.4 Conclusions...... 173

7.5 References ...... 174

SECTION IV. POLYMERIC NANOPARTICLES ...... 177

8 CHAPTER 8. CHITOSAN/FUCOIDAN MULTILAYER NANOCAPSULES FOR CONTROLLED RELEASE OF BIOACTIVE COMPOUNDS ...... 179

8.1 Introduction...... 181

8.2 Materials and Methods ...... 182

8.2.1 Materials ...... 182

8.2.2 Polysaccharides analyses ...... 183

8.2.3 Antioxidant activity of fucoidan ...... 183

8.2.4 Antimicrobial activity of chitosan ...... 183

8.2.5 Evaluation of the interactions between chitosan and fucoidan ...... 184

8.2.5.1 Quartz crystal microbalance ...... 184

8.2.6 Preparation of nanocapsules...... 184

8.2.7 Characterization of nanoparticles and nanocapsules ...... 185

8.2.7.1 ζ-potential measurements ...... 185

8.2.7.2 Morphology ...... 185

8.2.7.3 Fourier transform infrared (FTIR) spectroscopy ...... 185

8.2.8 PLL loading ...... 185

8.2.9 Determination of PLL release kinetics ...... 186

8.2.10 Statistical Procedures ...... 187

xix

8.2.10.1 Non-Linear Regression Analysis ...... 187

8.3 Results and Discussion ...... 187

8.3.1 Polyelectrolytes characterization ...... 187

8.3.1.1 Composition ...... 187

8.3.1.2 Bioactive properties ...... 188

8.3.2 Interactions between chitosan and fucoidan ...... 190

8.3.3 Characterization of core-shell nanocapsules ...... 191

8.3.3.1 ζ-potential measurements ...... 191

8.3.4 Characterization of hollow nanocapsules ...... 192

8.3.4.1 Fourier transform infrared (FTIR) spectroscopy ...... 192

8.3.4.2 Morphology ...... 193

8.3.5 Encapsulation and in vitro release of a model bioactive compound ...... 194

8.4 Conclusion ...... 196

8.5 References ...... 198

SECTION V. CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK ...... 201

9 CHAPTER 9. GENERAL CONCLUSIONS AND FUTURE PERSPECTIVES ...... 203

9.1 General Conclusions ...... 205

9.2 Guidelines for future work ...... 207

LIST OF FIGURES

CHAPTER 1 Figure 1.1 Schematic representation of the motivation for the work presented in this thesis. .. 6

CHAPTER 2 Figure 2.1 Representative repetitive structures of the main carrageenan types (adapted from Relleve et al., 2005)...... 16 Figure 2.2 Chemical structure of alginate (adapted from Burana-osot et al., 2009)...... 17 Figure 2.3 Chemical structure of the repeating dimeric units of fucoidans (adapted from Park et al., 2011)...... 18 Figure 2.4 Structure of a fully deacetylated chitosan (adapted from Ryzhkova et al., 2011). ... 19 Figure 2.5 Layer-by-layer assembly on planar surfaces (Wohl and Engbersen, 2012)...... 23 Figure 2.6 Layer-by-layer assembly on colloidal templates to obtain hollow capsules (Wohl and Engbersen, 2012)...... 24 Figure 2.7 Utilization of LbL technique for the production of multilayer nanoemulsions...... 35 Figure 2.8 Schematic diagram of physicochemical conditions in the different regions of the human GI tract (adapted from Guerra et al., 2012)...... 44 Figure 2.9 Schematic representation of pH-stat in vitro digestion model used to determine the digestion and release of lipid droplets.(Li and McClements, 2010) ...... 46 Figure 2.10 DGM showing the main body/fundal area (1) and antrum (2) (McAllister, 2010). . 47 Figure 2.11 Human gastric simulator. (1) Motor; (2) latex lining; (3) mesh bag; (4) secretion tubing; (5) roller; (6) belt; (7) light bulb for temperature control; (8) plastic foam insulation (Kong and Singh, 2010)...... 48 Figure 2.12 Front panel of the TIM-1 system and schematic diagram showing the stomach compartment (A) and three small- intestinal compartments: duodenum (C), jejunum (E), and ileum (G) connected by vertical peristaltic valves, including the pyloric sphincter (B) and the ileo-cecal valve (H). Each compartment with secretion tubes, such as gastric acid and pepsin in the stomach (I), and pancreatin and bile in the duodenum (J), pH electrodes (P), pressure (S) or level sensors (Q) and temperature sensors (R). Connected to the jejunum (left) and ileum (right) are semi-permeable hollow-fibre membrane units (M) for continuous absorption of digested products and water absorption (N)...... 49

xxi

CHAPTER 3 Figure 3.1 Schematic representation of the experimental installation used for in vitro digestion experiments: (A) Glass jacket reactor, (B) pH electrode, (C) Water bath heated at 37 °C, (D) Syringe pump, (E) Stir plate ...... 77

Figure 3.2 a) Dilution enthalpy change (ΔdilH) resulting from the titration of chitosan solution on water alone as function of chitosan solution concentration. b) Observed enthalpy change

(ΔobsH) resulting from the titration of chitosan solution on the κ-carrageenan solution as function of chitosan solution concentration...... 80

Figure 3.3 κ-carrageenan/chitosan apparent interaction enthalpy (Δap-intH) as a function of chitosan solution concentration...... 82 Figure 3.4 Adsorbed A/C PET/κ-carrageenan/chitosan apparent interaction enthalpy change

(Δap-intHA/C PET) as a function of chitosan solution concentration...... 83 Figure 3.5 Change in frequency (a) and in motional resistance (b) as a function of time for κ- carrageenan (0.2% w/v, pH 7) and chitosan (0.2% w/v, pH 3) adsorption and for rinsing with distilled water (pH 7 or pH 3); 1 - distilled water at pH 7, 2 - deposition of κ-carrageenan; 3- deposition of chitosan, 4 - distilled water at pH 3...... 84 Figure 3.6 Multilayer assembly monitored by UV/VIS spectroscopy at 260 nm. Each data point is the average of three determinations and the error bars show the standard deviation...... 86 Figure 3.7 Contact angle measurements for the original PET, the A/C PET and for the nanolayered coatings. The measurements were performed at 0, 15 and 30 s. Each data point is the average of ten measurements and the error bars show the standard deviation...... 86 Figure 3.8 SEM images of the surface morphology of the original PET (a); surface morphology of the A/C PET with the 5 nanolayers (b); first layer of κ-carrageenan (c); coating with two layers (κ-carrageenan and chitosan) (d)...... 88

CHAPTER 4 Figure 4.1 Schematic representation of the MB loading in the different positions of the κ- carrageenan/chitosan nanolayered: second layer (a), fourth layer (b) and sixth layer (c)...... 101 Figure 4.2 UV-Vis absorption spectra of the κ-carrageenan/chitosan coatings as a function of the MB layer position in the multilayer structure: MB in second (___), forth (___) and sixth layer (___)...... 103 Figure 4.3 Real time monitoring of the deposition of MB forming the second (), fourth () and sixth () layer of the κ-carrageenan/chitosan multilayered coatings using a quartz crystal microbalance (QCM)...... 104

xxii

Figure 4.4 SEM images of the κ-carrageenan/chitosan coating with MB in the sixth layer. Thickness of the layers (a). Detail of the structure of the layers (b)...... 105 Figure 4.5 Example of Fick´s (Equation 4.1) and LSM (Equation 4.3) description of Methylene Blue release at 37 °C from: (a) the sixth layer at pH=7 and (b) the fourth layer at pH=2. Insets show the detail of the model fitting to the initial experimental data. Experimental data (x), results from fitting Equation 4.1 (-----) and Equation 4.3 (____)...... 108 Figure 4.6 Fick´s (Equation 4.1) and LSM (Equation 4.3) description of Methylene Blue release at pH 7. Example is for the fourth layer at 4 °C. Experimental data (x), results from fitting Equation 4.1 (-----) and Equation 4.3 (____)...... 111

CHAPTER 5 Figure 5.1 Molecular structure of curcumin (adapted from Duvoix et al., 2005) ...... 121 Figure 5.2 Effect of emulsifier concentration on emulsions size (a) and polydispersity index (PDI) (b). Tween 20 (), SDS (), DTAB () and lactoferrin () -stabilized nanoemulsions. Results obtained for an oil/ emulsifier aqueous solution ratio of 0.5/9.5. Bars indicate standard deviation...... 124 Figure 5.3 Effect of oil/emulsifier solution ratio on size (a) and polydispersity index (PDI) (b). Tween 20 (), SDS (), DTAB () and lactoferrin () -stabilized nanoemulsions. Bars indicate standard deviation...... 125 Figure 5.4 ζ-potential of nanoemulsions stabilized by Tween 20, SDS, DTAB (concentration of 0.5%, oil/ emulsifier aqueous solution of 0.5/9.5) and lactoferrin (concentration of 2%, oil/ emulsifier aqueous solution ratio of 0.5/9.5)...... 126 Figure 5.5 TEM micrographs of Tween 20 (a), SDS (b), DTAB (c) and lactoferrin-stabilized nanoemulsions (d)...... 127 Figure 5.6 Evolution of Z-average diameter (a) and of polydispersity index (PDI) (b) of the nanoemulsions during 90 days of storage at 4 °C in the dark: Tween 20 (), SDS (), DTAB () and lactoferrin () -stabilized nanoemulsions. Bars indicate standard deviation...... 128 Figure 5.7 Visual appearance of the curcumin nanoemulsions (a) at day 0 and (b) after 90 days of storage in the dark at 4 °C. The insert figure represents a detail of the DTAB-stabilized nanoemulsion...... 129 Figure 5.8 Changes of ζ-potential of nanoemulsions stabilized by Tween 20 (), SDS (), DTAB () and lactoferrin () as a function of time. Bars indicate standard deviation...... 130 Figure 5.9 Evolution of total colour difference (TCD) of nanoemulsions stabilized by Tween 20 (), SDS (), DTAB () and lactoferrin (), during 90 days of storage. Bars indicate standard deviation...... 130

xxiii

CHAPTER 6 Figure 6.1 Human gastric simulator used in the in vitro digestion experiments: (A) latex lining, (B) secretion tubing, (C) roller, (D) belt, (E) peristaltic pump, (F) temperature control, (G) motor ...... 140 Figure 6.2 Dynamic light scattering-based mean particle diameter of curcumin nanoemulsions stabilized by different emulsifiers: Tween 20 (), SDS () and DTAB (), as they pass through an in vitro digestion model...... 143 Figure 6.3 Static laser light scattering-based particle size distribution curves of curcumin nanoemulsions stabilized by different emulsifiers: Tween 20 (a); SDS (b) and DTAB(c). Each stage of in vitro digestion is represented as follows: initial (◆), stomach (■), duodenum (▲), jejunum (×) and ileum (○)...... 144 Figure 6.4 Confocal images (a) and visual appearance of the samples (b) of curcumin nanoemulsions stabilized by different emulsifiers (Tween 20, SDS and DTAB) as they pass through an in vitro digestion model. The scale bar for all confocal images is 25 µm...... 147 Figure 6.5 ζ-potential values of curcumin nanoemulsions stabilized by different emulsifiers: Tween 20 (), SDS () and DTAB (), as they pass through an in vitro digestion model...... 148 Figure 6.6 Influence of emulsifier type on the production of total free fatty acids from curcumin nanoemulsions stabilized by Tween 20 (), SDS () and DTAB ()...... 149 Figure 6.7 Effect of initial emulsifier type on curcumin bioavailability as nanoemulsions pass through an in vitro digestion model: nanoemulsions stabilized by Tween 20 (), SDS () and DTAB ()...... 150 Figure 6.8 Confocal images of curcumin nanoemulsions stabilized by SDS and chitosan. Scale bar represents 2.5 µm...... 151

CHAPTER 7 Figure 7.1 Dynamic gastrointestinal system used in the in vitro digestion experiments: (A) gastric compartment, (B) duodenal compartment, (C) jejunal compartment, (D) ileal compartment, (E) glass jacket, (F) flexible wall, (G) pH electrodes, (H) syringe pumps, (I) hollow fibre membrane, (J) water bath, (K) peristaltic pumps (L) pump controllers, (M) recovery of jejunal filtrate, ileal filtrate and ileal delivery on ice...... 162 Figure 7.2 Influence of alginate concentration on the ζ-potential (a) and size (b) of secondary emulsions ...... 166 Figure 7.3 Mean particle diameter of curcumin nanoemulsions stabilized by lactoferrin () and lactoferrin/alginate () as they pass through the dynamic in vitro digestion model...... 167 xxiv

Figure 7.4 TEM images of curcumin nanoemulsions stabilized by lactoferrin and lactoferrin/alginate as they pass through the dynamic in vitro digestion model. The scale bar for all TEM images is 0.5 µm...... 169 Figure 7.5 ζ-potential values of curcumin nanoemulsions stabilized by lactoferrin () and lactoferrin/alginate () as they pass through the dynamic in vitro digestion model...... 170 Figure 7.6 Percentage of free fatty acids (FFA) released from lactoferrin (LF) and lactoferrin/alginate (LF+Al) stabilized emulsions as they pass through the dynamic in vitro digestion model: Initial (); Jejunal filtrate (); Ileal filtrate () and Ileal delivery ()...... 171 Figure 7.7 Total bioavailability of curcumin during gastrointestinal transit of curcumin nanoemulsions stabilized by lactoferrin (a) and lactoferrin/alginate (b) in the dynamic in vitro digestion model. Total bioavailability of curcumin is shown in the filtrate samples of jejunal () and ileal () compartments at 1 h intervals and expressed as % related to curcumin intake. The results are presented as the mean values of triplicate experiments...... 172

CHAPTER 8 Figure 8.1 Scavenging effects on DPPH radical of fucoidan () compared with the positive controls BHA () and BHT ()...... 188 Figure 8.2 Diameter of the inhibition zones (cm) of S. aureus and E. coli produced by a solution of 1mg.mL-1 of chitosan in the disk diffusion assay. Error bars represent the standard deviation obtained in the triplicate experiments...... 189 Figure 8.3 Change in frequency as a function of time for chitosan (C, 1 mg.mL-1 (w/v), pH 3) and fucoidan (F, 1 mg.mL-1 (w/v), pH 7) adsorption...... 190 Figure 8.4 ζ-potential change of polystyrene nanoparticles after each chitosan (even layers) or fucoidan (odd layers) deposition. The first point refers to the bare templates...... 191 Figure 8.5 FTIR spectra of polystyrene nanoparticles (─), chitosan/fucoidan core-shell nanoparticles (─) and hollow chitosan/fucoidan nanocapsules (─)...... 192 Figure 8.6 SEM images of polystyrene nanoparticles (a) and hollow chitosan/fucoidan nanocapsules (b)...... 193 Figure 8.7 TEM images of polystyrene nanoparticles (a) and hollow chitosan/fucoidan nanocapsules (b)...... 194 Figure 8.8 Effect of PLL concentration on encapsulation efficiency () and loading capacity () of nanocapsules...... 195 Figure 8.9 Profile of PLL release from chitosan/fucoidan nanocapsules at 37 °C and pH 2 (a) and pH 7 (b): experimental data (x); description of Fick’s model (i=0) (--) and of Linear Superimposition Model (i=1) (__)...... 195

xxv

LIST OF TABLES

CHAPTER 2

Table 2.1 Most relevant polyelectrolytes derived from natural polysaccharides ...... 15 Table 2.2 Most relevant proteins used in food systems ...... 20 Table 2.3 Commonly used lipid vehicles for lipid-based formulations...... 21 Table 2.4 Examples of nanolayered systems produced with bio-based materials ...... 25 Table 2.5 Hollow multilayer nanocapsules composed of bio-based polyelectrolytes...... 28 Table 2.6 Most commonly used emulsifiers in the production of nanoemulsions ...... 33 Table 2.7 Examples of lipophilic bioactive compounds encapsulated in food nanoemulsions and their potential benefits ...... 34 Table 2.8 Examples of multilayer emulsions produced using biopolymers ...... 36

CHAPTER 3

Table 3.1 Water vapour permeability (WVP) and oxygen permeability (O2P) of the A/C PET, of the PET with 5 nanolayers and of the 5 polysaccharide nanolayers alone. The values represent the experimental average ± standard deviation ...... 89 Table 3.2 Monosaccharide composition of the PET + nanolayers samples at different stages of in vitro digestion ...... 90

CHAPTER 4

Table 4.1 Results of fitting the Linear Superimposition Model (LSM) (Equation 4.3) to experimental data of the MB release. Evaluation of the quality of the regression on the basis of RMSE and R2. Estimates precision is evaluated using the SHW% (in parenthesis) ...... 109

CHAPTER 8

Table 8.1 Composition of the fucoidan used to prepare the multilayer nanocapsules ...... 188 Table 8.2 Results of fitting the Linear Superimposition Model (LSM) (i=1) to experimental data of the PLL release. Evaluation of the quality of the regression on the basis of RMSE and R2. Estimates precision is evaluated using the SHW% (in parenthesis)...... 196

xxvii

LIST OF GENERAL NOMENCLATURE

Symbols

A - permeation area kF - Fickian diffusion rate constant kR - relaxation rate constant

L – average thickness

M∞ - mass released at equilibrium

M∞,F – mass released by Fick diffusion at equilibrium

M∞,R – mass released due to relaxation process at equilibrium

Mt - total mass released

Mt,F - mass released by Fick diffusion at time t

Mt,R - mass released due to relaxation process at time t

O2P - oxygen permeability t - time

WVP – water vapour permeability

X - fraction of compound released by Fickian transport

Greek symbols

Δap-intH - apparent enthalpy of interaction

ΔdilH - dilution enthalpy change

ΔF - variation of resonance frequency

ΔHi - enthalpy change for injection i

ΔobsH - observed enthalpy change

ΔP – difference of partial vapour pressure

ΔR - variation of the motional resistance

xxix

Abbreviations

A/C PET - aminolized/ charged polyethylene terephthalate

CLSM - confocal laser scanning microscopy

DAG - diacylglycerides

DGM - dynamic Gut Model

DLS – dynamic light scattering

DTAB - dodecylTrimethylAmmonium Bromide

FFA – free fatty acids

FTIR - Fourier transform infrared

GI – gastrointestinal

HGS - human gastric simulator

LbL – layer-by-layer

LF – lactoferrin

LF + Al – lactoferrin + alginate

LSM - linear superimposition model

MAG - monoacylglycerides

MB – methylene blue

MW – molecular weight

PBS - phosphate buffer solution

PDI – polydispersity index

PET - polyethylene terephthalate

PLL – poly-L-lysine

PS – polystyrene

QCM - quartz crystal microbalance

RMSE - squared root mean square error

SDS - sodium Dodecyl Sulphate

SEM – scanning electron microscopy xxx

SHW - standardized Halved Width

SSE - sum of the squared residues

TAG – triglycerides

TEM – transmission electron microscopy

WPI – whey protein isolate

xxxi

LIST OF PUBLICATIONS

This thesis is based on the following publications:

Papers in peer reviewed journals

Pinheiro, A. C., Bourbon, A. I., Medeiros, B. G. S., da Silva, L. H. M., da Silva, M. C. H., Carneiro- da-Cunha, M. G., Coimbra, M. A., Vicente, A. A. (2012). Interactions between κ-carrageenan and chitosan in nanolayered films – structural and transport properties evaluation. Carbohydrate Polymers, 87, 2, 1081-1090.

Pinheiro, A. C., Bourbon, A. I., Quintas, M. A. C., Coimbra, M. A., Vicente, A. A. (2012). κ- carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound. Innovative Food Science and Emerging Technologies, 16, 227-232.

Pinheiro, A. C., Lad, M., Silva, H. D., Coimbra, M. A., Boland, M., Vicente, A. A. (2013). Unravelling the behaviour of curcumin nanoemulsions during in vitro digestion: effect of the surface charge. Soft Matter, 9, 3147-3154.

Pinheiro, A. C., Coimbra, M. A., Vicente, A. A. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers. Soft Matter (submitted).

Pinheiro, A. C., Bourbon, A. I., Cerqueira, M. A., Maricato, E.; Nunes, C., Coimbra, M. A., Vicente, A. A. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds. Journal of Food Engineering (submitted).

Cerqueira, M. A., Pinheiro, A. C., Silva, H. D., Ramos, P. E., Azevedo, A. A., López, M. L., Rivera, M. C., Bourbon, A. I., Ramos, O. L., Vicente, A. A. Design of bio-nanosystems for functional compounds delivery. Food Engineering Reviews (submitted).

Book Chapter: Cerqueira, M. A. Bourbon, A. I., Pinheiro, A. C., Silva, H. D., Quintas, M. A. C. and Vicente, A. A. (2013). Edible nanolaminate coatings for food applications, In: Ecosustainable Polymer Nanomaterials for Food Packaging. Editors: Clara Silvestre, Istitutodi Chimica e Tecnologia dei Polimeri (ICTP), Consiglio Nazionale delle Ricerche (CNR), Pozzuoli, Italy, publicado por Koninklijke Brill NV, The Netherlands (ISBN: 978-9-00420-738-7).

xxxiii

SECTION I INTRODUCTION

1 CHAPTER 1. MOTIVATION AND OUTLINE

MOTIVATION AND OUTLINE

1.1 Thesis motivation ...... 5

1.2 Research aims ...... 7

1.3 Thesis outline ...... 7

1.4 References ...... 9

Pinheiro, A. C. (2013)

Chapter 1. Motivation and outline

1.1 THESIS MOTIVATION

The increase in consumers’ awareness of the impact that food has on health and the consequent increase in demand for functional foods are stimulating the innovation and the development of new products in the food industry, in which nanotechnology could play a decisive role. Recent studies involving the use of nanotechnology in the food sector deal with very diverse nanostructures. These include nanofilms for, e.g. food packaging applications (Medeiros et al., 2012) and nanocapsules (Shu et al., 2010) and nanoemulsions (Qian et al., 2012), encapsulating functional ingredients. In general, these systems are reported to be promising means of improving the bioavailability of incorporated active compounds (Yu and Huang, 2012) as well as their physical and chemical stability while offering a greater control over the rate of release of these (McClements and Xiao, 2012).

Although nanotechnology is an emerging scientific area that has potential to generate innovative products and processes in the food sector, there are significant challenges that must be overcome so this technology can be entirely embraced by food industry. These main challenges are the need to produce edible delivery systems using economic processing operations and the need to formulate them to be effective and safe for human consumption (McClements and Xiao, 2012). The first issue can be solved by producing nanostructures entirely from food-grade ingredients and using simple and economic approaches (such as layer-by-layer technique) and the second issue is a broader one, it is raising a safety concern among the consumers, and must therefore be evaluated (ideally) in vivo, or at least in vitro. The danger of not doing this is to jeopardize the public image of a very interesting and very promising area of development and opportunity for the food industry. Nowadays, the existing information about the behaviour of the release systems based on nanotechnology in the gastrointestinal (GI) tract during digestion is very scarce and this knowledge will be crucial to evaluate their biological activity in vivo and to know the potential health risks from their use.

In this context, the great potential of nanotechnology in food sector and the emerging need to evaluate the behaviour of nanostructures in the GI tract were the major motivations for developing this thesis. The schematic representation of the motivation for this thesis can be seen in Figure 1.1.

Pinheiro, A. C. (2013)

Figure 1.1 Schematic representation of the motivation for the work presented in this thesis.

Chapter 1. Motivation and outline

1.2 RESEARCH AIMS

The main objectives of this thesis were to provide more options for nanostructures which are edible and therefore applicable in the food industry; to build a model gastrointestinal system which is able to simulate the in vivo behaviour of the human body as closely as possible, in order to provide a first approach to obtain information on the behaviour of nanostructures inside the human body during digestion; and to provide information on the behaviour of the nanostructures inside the human body during digestion.

The specific aims of this work were:

. Production of nanofilms through the layer-by-layer deposition of two oppositely charged polysaccharides, their characterization in terms of surface and gas barrier properties and evaluation of their behaviour during in vitro digestion; . Incorporation of a model bioactive compound in the nanofilm and evaluation of its loading and release behaviour; . Production of nanoemulsions incorporating a valuable lipophilic compound using different emulsifiers and evaluation of emulsifier effect on the behaviour of nanoemulsions during in vitro digestion. . Development of multilayer emulsions incorporating a bioactive compound and evaluation of their behaviour in a dynamic gastrointestinal system. . Development of multilayer nanocapsules for controlled release of bioactive compounds: incorporation of a model bioactive compound and evaluation of its loading and release behaviour.

1.3 THESIS OUTLINE

The present thesis is dived in five thematic sections, which are also divided in chapters, in a total of nine, with six of them reporting experimental results and their discussion (Chapter 3- 8).

Section I (containing Chapters 1 and 2) provides an overview on the state-of-art of nanotechnology applied to food sector and enables to contextualize the work reported in the following sections. The results obtained were organized in the subsequent sections according to the type of nanostructure studied: nanofilms (Section II – Chapters 3 and 4), nanoemulsions (Section III – Chapters 5, 6 and 7) and polymeric nanoparticles (Section IV - Chapter 8). Each one of the chapters containing experimental results is provided with a specific introduction and a specific list of references.

Pinheiro, A. C. (2013)

In Chapter 3 the interactions between κ-carrageenan and chitosan were investigated. Also, the κ-carrageenan/chitosan nanofilm assembled on a polyethylene terephthalate (PET) support was characterized in terms of its surface and gas barrier properties.

In Chapter 4, methylene blue, used as a model compound, was incorporated in different positions of the κ-carrageenan/chitosan nanofilm and its loading and release behaviour was evaluated.

In Chapter 5, oil-in-water nanoemulsions containing curcumin were prepared with different synthetic emulsifiers (Tween 20, non-ionic, sodium dodecyl sulphate, anionic, dodecyltrimethylammonium bromide) and a protein (lactoferrin) in order to evaluate the impact of emulsifier type on the stability of curcumin nanoemulsions.

The impact of surface charge on the behaviour of curcumin nanoemulsions during in vitro digestion was evaluated in Chapter 6. The oil-in-water nanoemulsions were stabilized using three emulsifiers with different charge and a human gastric simulator was used as the in vitro digestion model.

In Chapter 7, a dynamic digestion model comprising stomach, duodenum, jejunum and ileum was used to evaluate the impact of alginate coating on the digestion of curcumin nanoemulsions stabilized by lactoferrin.

Biodegradable hollow nanocapsules were developed through LbL assembly of chitosan and fucoidan in Chapter 8. A model bioactive compound (poly-L-lysine) was entrapped in the nanocapsules and its loading and release behaviour was evaluated.

The last section (Section V – Chapter 9) contains the most relevant conclusions of this thesis and perspectives for future work.

Chapter 1. Motivation and outline

1.4 REFERENCES

McClements, D. J., & Xiao, H. (2012). Potential biological fate of ingested nanoemulsions: influence of particle characteristics, Food & Function 3(3), 202-220. Medeiros, B. G. S., Pinheiro, A. C., Carneiro-da-Cunha, M. G., & Vicente, A. A. (2012). Development and characterization of a nanomultilayer coating of pectin and chitosan – Evaluation of its gas barrier properties and application on ‘Tommy Atkins’ mangoes, Journal of Food Engineering 110(3), 457-464. Qian, C., Decker, E. A., Xiao, H., & McClements, D. J. (2012). Physical and chemical stability of β-carotene-enriched nanoemulsions: Influence of pH, ionic strength, temperature, and emulsifier type, Food Chemistry 132(3), 1221-1229. Shu, S., Sun, C., Zhang, X., Wu, Z., Wang, Z., & Li, C. (2010). Hollow and degradable polyelectrolyte nanocapsules for protein drug delivery, Acta Biomaterialia 6(1), 210-217. Yu, H., & Huang, Q. (2012). Improving the Oral Bioavailability of Curcumin Using Novel Organogel-Based Nanoemulsions, Journal of Agricultural and Food Chemistry 60(21), 5373- 5379.

2 CHAPTER 2. LITERATURE REVIEW

LITERATURE REVIEW

2.1 Food Nanotechnology ...... 13

2.2 Biobased materials ...... 13

2.3 Development of multilayer nanostructures through the layer-by-layer technique 22

2.4 Nanostructures ...... 25

2.5 Nanostructures characterization ...... 37

2.6 In vitro digestions ...... 43

2.7 References ...... 52

Pinheiro, A. C. (2013)

This Chapter was adapted from:

Cerqueira, M.A. Bourbon, A.I., Pinheiro, A.C., Silva, H.D., Quintas, M.A.C. and Vicente, A.A., (2013). Edible nanolaminate coatings for food applications, In: Ecosustainable Polymer Nanomaterials for Food Packaging. Editors: Clara Silvestre, Istitutodi Chimica e Tecnologia dei Polimeri (ICTP), Consiglio Nazionale delle Ricerche (CNR), Pozzuoli, Italy, publicado por Koninklijke Brill NV, The Netherlands (ISBN: 978-9-00420-738-7).

Chapter 2. Literature Review

2.1 FOOD NANOTECHNOLOGY

Nanotechnology is an emerging scientific area that involves the characterization, fabrication and/or manipulation of structures, devices or materials that have one or more dimension that is smaller than 100 nm. Structures on this scale exhibits physical and chemical properties that are substantially different from their macroscopic counterparts (Duncan, 2011). Therefore, nanotechnology holds a great potential to generate new materials with unique properties and to produce very innovative products for numerous applications. Many industries, including microelectronics, aerospace and pharmaceutical industries, have already recognized the potential benefits of nanotechnology and many nanotechnology-based products are already available in the market (Weiss et al., 2006). However, achievements and discoveries in nanotechnology are only slowly impacting the food industry and associated industries (Duncan, 2011). The nanotechnology-derived foods are new to consumers and the lack of knowledge of their potential effects can raise concerns amongst the public and could impact the future application of nanotechnology in the food sector (Chaudhry et al., 2008). Even though, applications of nanotechnology in food are predicted to grow rapidly in the coming years due to the increase in demand e.g. for functional foods. In the near future, in addition of being a good source of nutrients with good sensory appeal, food also needs to contribute to the well-being and health of individuals. Consequently there are also requirements to control bioaccessibility during gastrointestinal transit, whilst maintaining food texture, structure and sensory appeal (Sanguansri and Augustin, 2006). Nanotechnology has the potential to impact many aspects of food systems and the main areas of application include food packaging and food products that contain nanosized or nanoencapsulated ingredients and additives (Sekhon, 2010 ).

Nanotechnologies promise many stimulating changes to enhance health, wealth and quality of life, while reducing environmental impact. However, their application by the food industry is hindered by facts that: 1) only edible materials must be used for nanostructures production; and 2) existing information about the behaviour of nanostructures during digestion is very scarce.

2.2 BIOBASED MATERIALS

The growing interest in nanotechnology has stimulated the discovery and development of new materials that can self-assemble into well-ordered structures at the nanometre scale and much effort has been focused on investigating the use of biological molecules for nanotechnology applications (Zhang et al., 2002). They present several advantages over

Pinheiro, A. C. (2013)

conversional materials, such as biocompatibility, biodegradability and nontoxicity. Biopolymers, such as polysaccharides and proteins, include a large variety of natural polyelectrolytes. Polyelectrolytes are polymers with ionisable groups, that under appropriate conditions, such as in aqueous solutions, dissociate, leaving ions on polymer chains and counterions in the solution (Dobrynin and Rubinstein, 2005).

Out of all biobased compounds available, this part of literature review focus mainly on polyelectrolytes (either polysaccharides or proteins) and lipids, once they were used in the present thesis work as materials for the formation of multilayer nanostructures and for the formation of nanoemulsions, respectively.

2.2.1 Polysaccharides

Polysaccharides can derive from plant biomass, algae, microbial fermentation or can be animal-derived. Polysaccharides chemical diversity is due to type, number, sequence, and bonding of the monosaccharides within the polymer chain. These chemical differences lead to differences in molecular properties (e.g. molecular weight, degree of branching, structure, flexibility, electrical charge and interactions) and consequently to differences in functional properties (e.g. solubility, thickening, gelation, water holding capacity, surface activity, emulsification and digestibility) (Matalanis et al., 2011).

Electrostatic interactions may be used to assemble specific biopolymer structure (e.g. nanolayered coatings through layer-by-layer deposition technique); therefore, it is important to know the electrical characteristics of polysaccharides. The magnitude of the electrical charge on ionic polysaccharides depends on the nature of their functional groups and varies with the pH of the environment, depending on the pKa value of the ionisable groups. The most common charged groups present in polysaccharides are sulphate groups (e.g., carrageenan, pKa < 2), carboxyl groups (e.g., pectin, alginate, xanthan, carboxymethylcellulose, pKa about 2.5 to 4.5) and amino groups (e.g., chitosan, pKa about 6.2) (Matalanis et al., 2011). Some of the polyelectrolytes derived from natural polysaccharides most relevant for food industry, their origins and main properties are presented in Table 2.1.

Chapter 2. Literature Review

Table 2.1 Most relevant polyelectrolytes derived from natural polysaccharides

Sources Polysaccharides Main structure type Main properties

Gelling or thickening polymer depending on the structure, cations Pectin Branched coil and temperature

Anionic polysaccharide

Arabic gum Branched coil domains on Emulsifier and stabilizer agent (Arabinogalactan) protein scaffold Botanical Anionic polysaccharide

Cellulose ionic derivatives Helical Gelling or thickening polymers

Starch ionic derivatives Coil Thickening and emulsifying polymers

Galactomannans ionic Linear/Branched Thickening agents derivatives

Gelling polymer in the presence of

Alginate Linear divalent counterions

Anionic polysaccharides

Gelling or thickening polymers,

Algae depending on the structure, cations Carrageenan Linear/Helical and temperature

Anionic polysaccharides

Wide range of biological activities Fucoidan Branched random coil Anionic polysaccharides

Xanthan Linear/helical Thickening agents Microbial Succinoglycan Linear/helical Anionic polysaccharides

Antimicrobial activity Animal Chitosan Linear Cationic polysaccharide

2.2.1.1 Carrageenans

Carrageenans are sulphated polysaccharides extracted from red seaweeds (Rhodophyceae). These linear polymers are composed of repeating and alternating α(1-3)-galactose and β(1-4)- 3,6-anhydrogalactose units, that can be sulphated (Figure 2.1). The various carrageenans’ structures differ in 3,6-anhydrogalactose and ester sulphate content and these variations

Pinheiro, A. C. (2013)

influence gel strength, texture, solubility, melting and setting temperatures, syneresis and synergistic properties (Imeson, 2009).

Figure 2.1 Representative repetitive structures of the main carrageenan types (adapted from Relleve et al., 2005).

The main carrageenan types, lambda, kappa and iota, differ in their thickening and gelling properties. Lambda carrageenan is a thickening polymer that is not affected by the presence of salts, kappa forms strong and rigid gels with potassium ions, while iota interacts with calcium ions to give soft, elastic gels (Rinaudo, 2008). All carrageenans are soluble in hot water but only the sodium salts of kappa and iota are soluble in cold water. When heated at pH values below 4.3, carrageenans’ solutions loose viscosity and gel strength due to the occurrence of autohydrolysis at low pH (Hoffmann et al., 1996).

Due to the sulphation, carrageenans are anionic polyelectrolytes: a solution of 0.2% of κ- carrageenan at pH 7 exhibits a ζ-potential value of -60.53 ± 0.15 mV (Medeiros et al., 2011).

2.2.1.2 Alginate

Alginate is an anionic polysaccharide extracted from brown algae (mainly from the orders Laminariales and Fucales), where it is a structure-forming component, giving the plant both mechanical strength and flexibility. It is considered to be biocompatible, non-toxic, non- immunogenic and biodegradable (Klöck et al., 1997).

Alginate is a linear block copolymer composed of varying proportions of 1,4-linked β-D- mannuronic acid (M) and α-L-guluronic acid (G), forming regions of M-blocks, G-blocks and blocks of alternating sequence (MG-blocks) (Figure 2.2). Alginate physical properties depend on the M/G ratio (Chan et al., 2009).

Chapter 2. Literature Review

Figure 2.2 Chemical structure of alginate (adapted from Burana-osot et al., 2009).

Alginate exhibits thickening properties (increase of solution viscosity upon dissolution) and has the ability to form gels in the presence of multivalent counterions, which is a direct consequence of the fact that alginate is a polyelectrolyte. The viscosity of alginate aqueous solutions depends on their concentration, molecular weight and on the ionic strength of the medium. However, it is nearly constant between pH 6 and 8 and, at moderate alginate concentrations, increases below pH 4.5, reaches a maximum around 3-3.5 and then decreases. This behaviour may be explained by the fact that H-bonds attractions dominate over electrostatic repulsions (Rinaudo, 2008).

Alginate gels have the particular characteristic of being cold setting, which means that alginate gels’ setting is essentially independent from temperature (Draget, 2009). Also, alginate gels are heat stable and their gelling properties depend on the molecular characteristics of alginate; in particular stability and physical properties of the gels depend directly on their G content and on the length of their G blocks (Rinaudo, 2008). Alginates show also interesting ion-binding properties, which represent the basis for their gelling properties. Most monovalent counterions form soluble alginate salts, while divalent and multivalent cations (such as calcium) form gels or precipitates (Draget, 2009).

Purified alginate is produced under different salt forms, e.g., sodium, calcium and potassium. Sodium alginate presents a ζ-potential value of −62.13 ± 4.01 mV at pH 7 and at a concentration of 0.2% (Carneiro-da-Cunha et al., 2010).

2.2.1.3 Fucoidans

Fucoidans belong to a family of sulphated homo- and heteropolysaccharides isolated from brown algae (Figure 2.3). Each brown algae species synthesizes highly branched polysaccharides with specific sugar composition, originating diverse fucoidans’ structures.

Pinheiro, A. C. (2013)

Moreover, every algae species are composed by several types of fucoidans (Kusaykin et al., 2008). Fucoidans exhibit highly complex structures due to the presence of branching, random distribution of the sulphate units, different types of glycosidic linkages and also the presence of acetylation, methylation and pyruvilation (Pomin and Mourão, 2008). Percival and Ross (1950) suggested that fucoidan was mainly composed of 1,2-α-fucose and that most of sulphate groups were located at position C-4 of the fucose units (Percival and Ross, 1950). Recently, new fucoidans has been fully characterized, and it was shown that fucoidans can have an α-1,3 backbone or repeating disaccharide units of α-1,3 and α-1,4 linked fucose residues with branching attached at C2 positions and may be sulphated at C4, C2 or at both positions of fucose units (Bilan et al., 2004; Chevolot et al., 2001; Zvyagintseva et al., 2003). Due to the presence of sulphate groups, fucoidans exhibit negative charge at neutral pH values (Indest et al., 2009).

Figure 2.3 Chemical structure of the repeating dimeric units of fucoidans (adapted from Park et al., 2011).

Fucoidans exhibit a wide range of biological activities such as anticoagulant (Nishino et al., 1989), antitumor (Aisa et al., 2005), anti-inflammatory (Yang et al., 2006), antiviral (Ponce et al., 2003), antibacterial (Hirmo et al., 1995) and antioxidant (Zhao et al., 2008).

2.2.1.4 Chitosan

Chitosan is a linear polysaccharide obtained by deacetylation of chitin, which is a polysaccharide widely distributed in nature (e.g. crustaceans, insects and certain fungi)

(Peniche, 2008). Chitosan is composed of N-acetyl-2-amino-2-deoxy-D-glucopyranose and 2- amino-2-deoxy-D-glucopyranose units, linked by (1,4)-β-glycosidic bonds (Dash et al., 2011) (Figure 2.4). The amino group confer a polycationic character to the chitosan.

Chitosan is a very interesting polymer for numerous applications due to its biocompatibility, biodegradability, non-toxicity and due its intrinsic antimicrobial activity (Helander et al., 2001; Shahidi et al., 1999).

Chapter 2. Literature Review

Figure 2.4 Structure of a fully deacetylated chitosan (adapted from Ryzhkova et al., 2011).

In contrast with chitin, chitosan is soluble in aqueous acid media through the protonation of the amino groups and the formation of the corresponding chitosan salts (Peniche, 2008). As a cationic polymer (ζ-potential of +58.28 ± 4.18 mV for 0.2% solution at pH 3), chitosan may be associated with anionic polyelectrolytes, leading to a formation of a polyelectrolyte multilayer (Carneiro-da-Cunha et al., 2010).

2.2.2 Proteins

Proteins can adopt different structures (e.g. random coil, fibrous or globular shapes), which depend on their amino acid sequence, environmental conditions and environment history (Matalanis et al., 2011). The type, number and particular sequence of amino acids in a polypeptide chain determines the molecular weight, conformation, electrical charge, hydrophobicity, physical interactions and functionalities of proteins (Jones and McClements, 2010). It is important to establish the electrical characteristics of proteins, since electrostatic interactions are the key for the formation of some nanostructures (e.g. nanolayered coatings). The electrical charge of proteins goes from positive, at pH below their isoelectric point (pI), to zero at the pI and to negative at pH above the pI. The pI depends on the biological origin of the protein and on the processing procedures used to extract it (Matalanis et al., 2011). Protonated amino side groups contribute with positive charges below pH 10, whereas deprotonated carboxylate side groups contribute with negative charges above pH 2 (Jones and McClements, 2010).

Proteins exhibit unique functional properties, including emulsification, gelation, foaming and water binding capacity and offers the possibility of developing delivery systems for both hydrophilic and lipophilic bioactive compounds (Wang et al., 2011). Food proteins such as: soy proteins, milk proteins (caseins and whey proteins) and egg proteins are frequently used in food matrices (Table 2.2).

Pinheiro, A. C. (2013)

Table 2.2 Most relevant proteins used in food systems

Source Proteins Main structural type pI Functional properties

Emulsifying, foaming and gelling β-lactoglobulin Globular 4.8-5-1 properties

Bovine serum Solubility, foaming, and Globular 4.7 albumin emulsifying properties Milk proteins Emulsifying, foaming and gelling Casein Rheomorphic 4.6 properties,

Gelling, antimicrobial, anti- Lactoferrin Globular 8-9 inflammatory and antitumor properties

Milk or egg proteins Lysozyme Globular 10.7 Antimicrobial properties

Gelling and emulsifying

Egg protein Ovalbumin Globular 4.5-4.7 properties and foam

stability

7.0-9.4 Animal or vegetable (type A), Gelatin Linear Emulsifying, gelling properties proteins 4.8-5.5 (type B)

Gelling properties and thermal Vegetable protein Soy protein Globular 4.5 stability

2.2.2.1 Lactoferrin

Lactoferrin is a basic iron-binding glycoprotein of the transferrin family (Farnaud and Evans, 2003). It is abundant in milk and in most biological fluids (including tears and saliva) and is a cell-secreted molecule that bridges innate and adaptive immune function in mammals (García- Montoya et al., 2012). Lactoferrin has a molecular weight of 80 kDa and is composed by ca. 700 amino acids. It consists in a simple polypeptide chain folded into two symmetrical globular lobes (N and C lobes), which are connected by a hinge region that provides additional flexibility to the molecule (González-Chávez et al., 2009). Each lobe can bind one metal atom (such as Fe2+, Fe3, Cu2+, Zn2+ and Mn2+ ions) in synergy with the carbonate ion (van der Strate et al., 2001) .

This multiple bioactive protein has been extensively studied over the past decades (Farnaud and Evans, 2003). Indeed, it exhibits a diverse range of biological activities, including

Chapter 2. Literature Review antimicrobial, antiviral, antioxidant, anticarcinogenic, anti-inflammatory and immune modulator activities (González-Chávez et al., 2009). This wide range of biological activities is due to not only its capacity to bind but also due to interactions of lactoferrin with molecular and cellular components of both hosts and pathogens (García-Montoya et al., 2012).

2.2.3 Lipids

Lipids have the potential of providing endless opportunities in the area of delivery of lipophilic bioactive compounds due to their unique properties, such as their physicochemical diversity, biocompatibility and ability to enhance gastrointestinal bioavailability of poorly water soluble compounds (Chakraborty et al., 2009). There are a large variety of nanostructured systems that contain a lipophilic core, such as nanoemulsions, nanostructured multiple emulsions, nanostructure multilayer emulsions and solid lipid nanoparticles (Prombutara et al., 2012; Weiss et al., 2006). Commonly used lipid vehicles are listed in Table 2.3.

Table 2.3 Commonly used lipid vehicles for lipid-based formulations.

Classification Lipophilic vehicles

Fatty acids Oleic acid, Myristic acid, Caprylic acid, Capric acid,

Ethyl esters Ethyl Oleate

Soybean Oil, Peanut Oil, Corn Oil, Olive Oil, Sesame Oil, Canola oil, Triglycerides of long-chain fatty acids Grape seed oil

Triglycerides of medium-chain fatty acids Miglyol® 812, Captex® 355, Labrafac®

Flavor oils Orange oil, limonene

Fatty acids are often classified according to their chain length. Conventional fats and oils are composed of long chain triglycerides (LCT), which contain fatty acids with more than 12 carbons, whereas medium chain triglycerides (MCT) contain fatty acids with 6-12 carbons. Chain length imparts important differences in the absorption and metabolism. LCTs are transported via chylomicrons into the lymphatic system, allowing for extensive uptake into adipose tissue. MCTs, however, are directly absorbed in to portal circulation and transported to the liver for rapid oxidation (St-Onge and Jones, 2002).

Pinheiro, A. C. (2013)

2.3 DEVELOPMENT OF MULTILAYER NANOSTRUCTURES THROUGH THE LAYER-BY-LAYER TECHNIQUE

In the last 60 years, various methods have been developed and implemented for the construction of thin films. Langmuir (Langmuir, 1917) started the work on liquid monolayers that was complemented later by Blodgett (Blodgett, 1934). In this method, the films from the liquid surface were transferred to solid supports through dipping or removing the solid support from the aqueous sub-phase that intercepted the monolayer. This process, when repeated, lead to the formation of a multilayer film (Pavinatto et al., 2010).

In 1991, Decher and Hong changed the area of thin film assembly by the introduction of the Layer-by-Layer (LbL) technique (Decher and Hong, 1991). This method consists in the sequential adsorption of oppositely charged species onto a solid support and involves as forces ionic interactions. It does not involve covalent bonding; however, secondary forces such as hydrogen bonding may act. During the construction of a nano-laminate coating, the removal of the polymer excess is of extreme importance to avoid aggregation and formation of complexes. A variety of different adsorbing substances could be used to create the different layers, including natural polyelectrolytes (proteins, polysaccharides), charged lipids (phospholipids, surfactants) and colloidal particles (micelles, vesicles, droplets). The type of adsorbing substances used to produce each layer, the total number of layers incorporated into the nano-laminate coating, the sequence of the different layers and the conditions used to prepare each layer will determine the functionality of the final nano-laminate coating (e.g. their permeability to gases or water; their mechanical properties, such as rigidity or flexibility; their swelling and wetting characteristics; and their environmental sensitivity to pH, ionic strength and temperature) (Weiss et al., 2006).

The main advantage of the LbL technique is the use of water as the main solvent, allowing the control of thickness, composition, and structure of the nano-laminate coatings at nanometre scale. In fact, the LbL technique allows precise control over the thickness and properties of the interfacial films, enabling the creation of thin films (1 to 100 nm per layer) (Weiss et al., 2006), being the obtained thickness value dependent on the charge density, molecular weight and ionic strength of the adsorbing materials.

The LbL technique can be applied to modify the surface properties of a substrate of interest (Carneiro-da-Cunha et al., 2010; Fu et al., 2005) or the substrate can be used just as a support to allow the formation and the characterization of the nano-laminate coating (Medeiros et al., 2011). The LbL technique allows film formation on surfaces of almost any kind and shape,

Chapter 2. Literature Review however, the prerequisite for successful LbL coating is the presence of a minimal surface charge, which is one of the few disadvantages of the technique (de Villiers et al., 2011). Indeed, the key issue of LbL assembly is the need for surface recharging at each adsorption step. The molecules used for assembly should have a sufficient number of charged groups to provide stable adsorption on an oppositely charged surface and non-compensated charges exposed to the exterior (Mora-Huertas et al., 2010).

The versatility of the LbL technique allows the use of diverse templates (e.g. planar and colloidal substrates) (Ochs et al., 2010), resulting in formation of different multilayer nanostructured systems such as nanofilms, nanoemulsions and nanocapsules.

2.3.1 Planar substrates

The LbL assembly on nonporous planar substrates typically involves solid supports for the production of nanofilms. The most commonly used substrates are glass, quartz, synthetic polymers (e.g. polyethylene terephthalate, PET), mica and gold-coated substrates. The use of porous planar substrates for LbL assembly allows the production of nanotubes. The procedure of the LbL assembly on planar surfaces involves the alternating immersion of the template in solutions of two oppositely charged polymers and washing the template between the immersion steps in order to remove the excess of polymers (Figure 2.5).

Figure 2.5 Layer-by-layer assembly on planar surfaces (Wohl and Engbersen, 2012).

2.3.2 Colloidal substrates

The most common colloidal substrates are spherical colloids (e.g. solid spheres, emulsions) and high-aspect-ratio colloids (such as rods and fibres) (Sukhorukov et al., 1998). However, on

Pinheiro, A. C. (2013)

colloidal templates, LbL method can raise some difficulties such as the formation of counterion aggregates, the separation of the excess of polyelectrolytes from the particles prior to the next deposition cycle and polyelectrolyte-induced bridging during centrifugation (Mora-Huertas et al., 2010). Another difficulty is the size of the particles that is often higher than 500 nm, however this problem has been overcome by subjecting the particles to ultrasonic treatment to decrease the size of individual particles to nano-scale (100-200 nm) (Sun et al., 2005).

The LbL technique has been highly used in the construction of multilayers on sacrificial templates to obtain free-standing structures of various morphologies (Wohl and Engbersen, 2012). This has first been demonstrated by Caruso et al., by the construction of novel hollow capsules via LbL assembly on colloidal templates which were subsequently removed (Caruso et al., 1998). The LbL on colloidal templates assembly is achieved by suspending the templates alternatively in solutions of two oppositely charged polymers and after each assembly step, the templates are isolated through centrifugation and washed to remove the excess of polymer. Hollow LbL capsules are obtained by dissolving the colloidal template (Figure 2.6). Being so, free-standing nanostructured materials with different morphologies and functions can be obtained using a template with the required shape for LbL deposition, followed by removal of the template. This approach is known as LbL templating technique (Wang et al., 2007b).

Figure 2.6 Layer-by-layer assembly on colloidal templates to obtain hollow capsules (Wohl and Engbersen, 2012).

The simplicity of the LbL technique and the availability of established characterization methods render nano-laminate coatings extremely versatile in a wide range of applications such as cell/tissue scaffolds, optics and optoelectronics, electrochemistry and drug delivery systems. Food industry is one of the areas that can benefit from these nanostructures, once

Chapter 2. Literature Review nanolayered coatings could be created entirely from food-grade ingredients (i.e. proteins, polysaccharides and lipids).

2.4 NANOSTRUCTURES

2.4.1 Nanofilms

Nanofims are more likely to be used as coatings that are attached to surfaces (e.g. food), rather than as self-standing films, because their extremely thin nature makes them very fragile (Kotov, 2003). The formation of nanofilms using LbL assembly technique has been explored over the last years (Carneiro-da-Cunha et al., 2010; Fu et al., 2005; Lavalle et al., 2002; Rudra et al., 2006). Nanofilms can give food scientists some advantages for the preparation of edible coatings and films over conventional technologies, such as the simplicity of their preparation (simple processing operations such as dipping and washing procedures), the low concentrations of polymers used and the possibility of activating the release of functional agents in response to environmental changes, such as dilution, pH, ionic strength or temperature. Table 2.4 shows some examples of nanofilms composed by bio-based materials and their functional properties.

Table 2.4 Examples of nanolayered systems produced with bio-based materials

Polyelectrolytes Properties Reference Polycation Polyanion

Lysozyme Poly-L-glutamic acid Antimicrobial action Rudra et al., 2006

Chitosan Hyaluronan Antimicrobial action Richert et al., 2004

Low water vapor and oxygen Lysozyme κ-carrageenan Medeiros et al., 2011 permeabilities

Chitosan Alginate Good gas barrier properties Carneiro-da-Cunha et al., 2010

Anti-adhesive and anti- Chitosan Heparin Fu et al., 2005 bacterial properties

Bioactive compound Poly-L-lysine Poly-L-glutamic acid Jiang and Li, 2009 release system

2.4.1.1 Release of bioactive compounds

Nanofilms have been investigated as bioactive compounds release systems (Chung and Rubner, 2002; Jiang and Li, 2009; Zhong et al., 2007) due to the possibility of controlling the compound release through the manipulation of film properties and to incorporate a wide

Pinheiro, A. C. (2013)

range of functional biomolecules without substantial loss of their biological functions (Wang et al., 2007a). Functional LbL nanofilms can be obtained by chemical grafting of polyelectrolytes by functional moieties (Kaschak et al., 1999), alternate deposition of polyelectrolytes and functional molecules (Rousseau et al., 2000) or by post diffusion of the functional molecules into the multilayers (Jiang and Li, 2009; Zhong et al., 2007). A diversity of nanofilms has been developed to control bioactive compounds release by deploying different triggers such as pH, ionic strength, temperature and enzymes (Quinn and Caruso, 2004; Serizawa et al., 2002; Wood et al., 2005).

Jiang and Li (2009) developed polypeptide nanofilms composed of poly-L-lysine and poly-L- glutamic acid using LbL technique and studied the loading and release behaviour of small charged bioactive molecules (e.g. cefazolin, gentamicin and methylene blue). Their loading and release were found to be pH-dependent and could also be controlled by changing the number of film layers and drug incubation time, and applying heat-treatment after film formation. Antibiotic-loaded polypeptide nanofilm showed controllable antibacterial properties against Staphylococcus aureus. The authors concluded that the developed biodegradable polypeptide nanofilms are capable of loading both positively- and negatively-charged drug molecules and promise to serve as drug delivery systems.

In another work, polypeptide nanofilms have been prepared by LbL self-assembly to study the post-fabrication loading and release of a model therapeutic, methylene blue (MB) (Zhong et al., 2007). The results showed that the amount of MB loaded and the rate of MB release were influenced by the choice of polypeptide film species, pH, number of capping layers, and release medium.

2.4.1.2 Application in food systems

A nanofilm can be applied to a food system (such as fruits, vegetables or cheese) by using simple processing operations such as dipping. The food product to be coated can be dipped into a series of solutions containing substances that will adsorb to the surface of the food or, alternatively, the solutions containing the adsorbing substances could be spayed onto the surface of the food product. The composition, thickness, structure, and properties of the nanolayered coating formed around the food product could be controlled by a) changing the type of adsorbing substances in the dipping solutions; b) changing the total number of dipping steps used; c) changing the order that the food product is introduced into the various dipping solutions; or d) changing the solution and environmental conditions used, such as pH, ionic strength, dielectric constant and temperature (Weiss et al., 2006).

Chapter 2. Literature Review

It was not until recent years that some work has been developed on the effect of the application of nanofilms to food products. Medeiros et al. (2011) applied a nano-laminate coating (κ-carrageenan-lysozyme), through the LbL technique, on ‘Rocha’ (Pyrus communis L.) fresh-cut and whole pears. The results showed that the nanolayered coating assembled on the fruits’ surface had a positive effect on fruit quality and contributed to extend their shelf-life. In another work, the same authors applied a nanofilm composed of pectin and chitosan on whole “Tommy Atkins” mangoes and observed a positive effect of the nanofilm on gas flow reduction and on the consequent extension of the shelf-life of mangoes. These results were explained based on the combination of the antimicrobial and gas barrier properties of chitosan, with the low oxygen permeability of pectin layers. Also, the authors suggest that barrier properties were possibly improved by the nano-structure of the coating (e.g. by improving the tortuosity of the polymer network and/or by imposing successive interfaces to the gas molecules while migrating through the structure) (Medeiros et al., 2012).

2.4.2 Nanoparticles

Nanoparticles can be defined as solid colloidal particles that include both nanospheres and nanocapsules (Mora-Huertas et al., 2010). One of the main criteria in using particles in the nanometre-sized range as a delivery system for bioactive compounds is that they are nontoxic. Nanoparticles can be produced using biodegradable polymers such as proteins or polysaccharides (de Britto et al., 2012; Naoe et al., 2011). Biopolymeric nanoparticles can act as efficient vehicles for sustained, controlled and targeted release of the bioactive compound (Sundar et al., 2010). They offer several advantages over traditional delivery systems, including the ease of their preparation, their high stability in biological fluids and during storage (Mu and Seow, 2006), their subcellular size that allows relatively higher intracellular uptake than other particulate systems (Mosqueira et al., 2001) and their capacity to improve the stability of bioactive compounds (Ourique et al., 2008). The bioactive compounds encapsulated can be released in response to specific environmental triggers by altering the solution conditions to induce complete particle dissolution or changes in particle porosity (Weiss et al., 2006).

Generally, the classical methods for the preparation of nanocapsules are: nanoprecipitation, emulsion–diffusion, double emulsification, emulsion-coacervation, polymer-coating and LbL assembly (Mora-Huertas et al., 2010). As previously mentioned, using the LbL assembly makes it possible to obtain vesicular nanocapsules, such as hollow polyelectrolyte multilayer nanocapsules, with well-defined chemical and structural properties.

Pinheiro, A. C. (2013)

2.4.2.1 Hollow polyelectrolyte multilayer nanocapsules

Hollow nanocapsules are nano-vesicular systems that exhibit a typical core-shell structure in which the bioactive compound is within a cavity surrounded by polyelectrolyte layers. The permeability properties of hollow polyelectrolyte multilayer nanocapsules can be altered: they can shift from open to closed states (and vice-versa) through changes in environmental conditions (e.g. temperature or presence of organic solvents) (Ai and Gao, 2004; Anton et al., 2008). This unique feature allows the efficient loading of molecules inside the hollow capsules and their release. Table 2.5 shows some examples of hollow multilayer nanocapsules composed by bio-based polyelectrolytes.

The LbL technique allows to readily tailor the hollow nanocapsules properties (e.g., size, composition, porosity, stability, surface functionality, colloidal stability) and the step-wise formation of these capsules allows the introduction of multiple functionalities, thus providing opportunities to engineer a new class of materials with unprecedented structure and function (Johnston et al., 2006).

Table 2.5 Hollow multilayer nanocapsules composed of bio-based polyelectrolytes

Polyelectrolytes Bioactive Template Reference Polycation Polyanion compound

Water-soluble Dextran Bovine serum Silica particles Shu et al., 2010 chitosan sulphate albumin

Acridine Polystyrene particles Chitosan Alginate Ye et al., 2005 hydrochloride

Amino-modified silica Chitosan Carrageenan ---- Liu et al., 2012b

Didodecyldimethylammonium Chitosan Alginate ---- Cuomo et al., 2010 bromide

2.4.2.1.1. Encapsulation of bioactive compounds

There are three main techniques to incorporate bioactive compounds within multilayer nanocapsules. One of the most used approaches is the diffusion of the bioactive molecules from the surrounding medium in which the nanocapsules are dispersed into their interior. When the polyelectrolyte nanocapsule shell is assembled from polyelectrolytes that are responsive to salt, pores within the shell can open to allow molecules to diffuse into the capsules (Sukhorukov et al., 2001). The limitations of this technique are the low loadings achieved (once the maximum concentration of bioactive compounds inside the nanocapsules

Chapter 2. Literature Review is often limited to the concentration in the solution) and the fact that the molecular weight of diffusing molecules should be low enough to allow their permeation though the nanocapsule pores (Johnston et al., 2006).

In the cases where bioactive compound form crystals, polyelectrolyte multilayers can be directly assembled on the crystal template, encapsulating the compound (Caruso et al., 2000). The main advantage of this technique is that a very high concentration of compound is encapsulated. However, this strategy is most effective when the bioactive compound is insoluble or poorly soluble in water (once multilayers are typically deposited from aqueous solutions), and it is limited to compounds that form crystals (Johnston et al., 2006).

The other most commonly used technique for encapsulating bioactive compounds involves adsorbing the compound within a porous nanoparticle that can be subsequently LbL-coated and the porous template is subsequently removed. This technique allows a high loading of the bioactive compound (due to the high surface area of the porous particle) and is applicable to a range of materials of different sizes (Johnston et al., 2006).

2.4.3 Nanoemulsions

Lipophilic bioactive compounds present some characteristics that make them difficult to incorporate directly into food matrices, such as low water-solubility, poor chemical stability, high melting point and low bioavailability (McClements and Li, 2010b). The bioactivity of poorly water-soluble bioactive compounds can be greatly enhanced by incorporating them within emulsion-based delivery systems (McClements and Xiao, 2012). In these systems the bioactive compound is dissolved within a carrier oil, which is then emulsified with an aqueous emulsifier solution forming the oil-in-water emulsion. Emulsion-based delivery systems can be used to control the release of lipophilic bioactive compounds at specific locations within the GI tract, such as the mouth, stomach, small intestine, or colon.

Various types of emulsion-based delivery systems have been developed, such as conventional emulsions, nanoemulsions, multilayer emulsions, solid lipid nanoparticles and filled hydrogel particles (McClements and Li, 2010b). Both conventional emulsions and nanoemulsions consist of lipid droplets surrounded by a thin emulsifier layer, dispersed in an aqueous medium, being the main difference between these two systems, the size of lipid droplets (typically between 100 nm to 100 µm for conventional emulsions and between 10 nm to 100 nm for nanoemulsions). Therefore, nanoemulsions can be defined as nanoscale droplets of multiphase colloidal dispersions formed by dispersing one liquid in another immiscible liquid by physical shear-induced rupturing (Fathi et al., 2012).

Pinheiro, A. C. (2013)

Nanoemulsions exhibit a wide range of potential advantages over conventional emulsions for encapsulation and delivery of lipophilic bioactive compounds (McClements and Xiao, 2012). In nanoemulsions, droplet sizes are much smaller than visible wavelengths, so they only scatter light waves weakly and they appear optically transparent (McClements and Li, 2010b). In contrast, conventional emulsions tend to be white opaque in appearance because the size of lipid droplets is of the same order as the wavelength of light. This is a very favourable feature of nanoemulsions, once they are suitable to be incorporated as nutrient carriers in products that are optically clear or slightly turbid such as beverages (Fathi et al., 2012). Although nanoemulsions are only kinetically stable (whereas microemulsions are thermodynamically stable) (Tadros et al., 2004), they often exhibit long-term physical stability, which can lead to extended shelf-lives (Liu et al., 2006). Nanoemulsions have much better stability against gravitational separation over conventional emulsions due to Brownian motion of nano sized droplets, caused by entropic driving forces (Mason et al., 2006). They also have good stability against droplet aggregation because the range of attractive forces acting between the droplets decreases with decreasing particle size, while the range of steric repulsion is less dependent on particle size (Tadros et al., 2004). Once nanoemulsions have a very high surface area-to- volume ratio, chemical reactions, such as lipid digestion, are accelerated, thus leading to the increase of bioavailability of lipophilic bioactive compounds encapsulated within nanoemulsions compared with conventional emulsions (McClements and Li, 2010b).

2.4.3.1 Production methods

Nanoemulsions can be produced using two different approaches: high-energy (mechanical) and low energy approaches (non-mechanical). High-energy methods involve the use of mechanical devices capable of generating intense disruptive forces that breakup macroscopic droplets into smaller droplets and include high-pressure homogenization, microfluidization and ultrasonication (Silva et al., 2011). The choice of the particular homogenizer and of the operating conditions used depends on the characteristics of materials being homogenized (e.g., viscosity, interfacial tension, and shear sensitivity) and of the required final properties of the emulsion (e.g., droplet concentration, droplet size and viscosity) (McClements and Li, 2010b). In the high-pressure homogenization method, the coarse dispersion of oil and aqueous phase containing the emulsifier is pumped through a small inlet orifice at high pressures (in the range of 500 and 5000 psi) (Silva et al., 2012). Microfluidization method is based on combination of cavitation and mechanical shear and uses a very high pressure (up to 20,000 psi) to force the liquid through the microchannels, and then through a collision chamber which

Chapter 2. Literature Review leads to the formation of fine nano-scale droplets (Fathi et al., 2012). Using the ultrasonication method, the two immiscible liquids are submitted to high-frequency sound waves (typically 20 kHz or larger) and nanoemulsions are generated by cavitation (Patil and Pandit, 2007). Low- energy methods consist on the spontaneous formation of emulsions when the solution or the environment conditions are altered and include phase inversion and spontaneous emulsification methods (McClements and Xiao, 2012).

High energy approaches are the most widely used in food industry because they are capable of large-scale production and are the most versatile means of producing food-grade nanoemulsions due to the wide variety of oils and emulsifiers that can be used.

2.4.3.2 Emulsifiers

For a liquid–liquid biphasic system, the thermodynamically most stable state is the one of minimal interfacial area between the dispersed phase and the dispersion medium and, therefore, emulsions tend to the separation of the two liquid phases, oil and water (Grigoriev and Miller, 2009). Absorption of an emulsifier at the interface between droplets and dispersion medium is the most common way to stabilize emulsions because the interfacial tension is reduced.

Different types of emulsifiers are commonly used to stabilize nanoemulsions in food and pharmaceutical applications, including natural and synthetic surfactants (non-ionic, anionic or cationic) and amphiphilic polymers (Li et al., 2012). In the case of anionic or cationic surfactants, stability of the nanoemulsions is achieved by the electrical charge provided by the emulsifier to the emulsion droplets, whereas in the case of non-ionic surfactants, stability is achieved creating a strong steric barrier via bulky molecular groups directed toward the dispersion medium (Silva et al., 2012).

For biopolymers to be effective as emulsifying agents, they must be surface-active, i.e., they must have the capacity to lower surface tension at the oil–water interface (Dickinson, 2003). Proteins are a class of natural polyelectrolytes that exhibit simultaneously several hydrophobic groups in the structure and reveal therefore a certain interfacial activity. Many proteins are able to improve significantly the stability of nanoemulsions, once their hydrophobic groups are attached to the surface of the oil phase or even penetrate into it, whereas hydrophilic parts protrude into the aqueous phase and provide a thick and bulky interfacial layer (Grigoriev and Miller, 2009). Also, using proteins as emulsifiers allows controlling droplet charge because the sign and magnitude of their charge can be altered simply by varying solution pH: proteins are

Pinheiro, A. C. (2013)

positively charged below their isoelectric point (pI) and negatively charged above their pI (Guzey and McClements, 2006).

Some polysaccharides also have surface-active molecules that are capable of reducing the interfacial tension between the oil phase and the water phase, presenting therefore emulsification properties (Burapapadh et al., 2010). Due to their amphiphilic structure and highly surface-active characteristics, also phospholipids can be used as emulsifiers to stabilize nanoemulsions.

The type of emulsifier used to produce a nanoemulsion may impact its properties in a variety of ways: the initial droplet size produced by homogenization; emulsion stability during storage; droplet interactions with other components and formation of crystals within an oil–water system (Li et al., 2012). The emulsifiers most commonly used in the production of nanoemulsions can be seen in Table 2.6.

The impact of food-grade non-ionic surfactant type (Tween 20, 40, 60, 80, and 85) on the size of vitamin E-enriched nanoemulsions prepared by spontaneous emulsification was investigated by Saberi et al. (2013). The mean particle diameter and the polydispersity index of these nanoemulsions increased in the following order: Tween 80 < Tween 40 < Tween 20 < Tween 60 < Tween 85, and these results were attributed, not only to the HLB (hydrophilic- lipophilic balance) but also to the molecular geometry of the surfactant, that is characterized by a packing parameter (ratio between the effective cross-sectional areas of the tail group and the head group of the emulsifier).

Other authors compared the stability of β-carotene nanoemulsions stabilized by a non-ionic surfactant (Tween 20) with those stabilized by a globular protein (β-lactoglobulin) and observed that the rate of β-carotene degradation was appreciably faster in the nanoemulsion stabilized by Tween 20 than in the one stabilized by β-lactoglobulin. The authors suggested that this phenomenon may be due to numerous reasons, namely, the antioxidant properties of the β-lactoglobulin, the ability of proteins to form molecular complexes with carotenoids through hydrophobic interactions, the physical barrier provided by the layer of adsorbed β- lactoglobulin molecules at the oil–water interface that prevented any pro-oxidants in the aqueous phase from contacting the β-carotene, or the size of the droplets that was larger in the β-lactoglobulin-stabilized nanoemulsions (smaller oil–water interfacial that lead to a slower reaction rate) (Qian et al., 2012 ).

Chapter 2. Literature Review

Table 2.6 Most commonly used emulsifiers in the production of nanoemulsions

Type Emulsifier Reference

Tween 20 Troncoso et al., 2012

Tween 40 Saberi et al., 2013

Tween 60 Yuan et al., 2008

Tween 80 Rao and McClements, 2012

Non-ionic surfactants Tween 85 Ostertag et al., 2012

Span 20 de Araújo et al., 2007

Span 80 Schwarz et al., 2012

Brij 35 Li and McClements, 2011

Sucrose monopalmitate (SMP) Rao and McClements, 2011

Sodium dodecyl sulphate (SDS) Qian and McClements, 2011 Anionic surfactants Alkanol-XC Howe and Pitt, 2008

Dodecyl trimethylammonium bromide Li et al., 2012

Cationic surfactants (DTAB)

Lauric arginate Ziani et al., 2011

Whey protein isolate (WPI) Lee and McClements, 2010

Whey protein concentrate (WPC) Chu et al., 2007

Proteins Lactoferrin Mao and McClements, 2012

Sodium caseinate Qian and McClements, 2011

β-lactoglobulin Qian et al., 2012

Pectin Burapapadh et al., 2010 Polysaccharides Modified starch Jafari et al., 2007

Lipoid E-80® (egg lecithin) Fasolo et al., 2007 Phospholipid Lipoid S75-3® (soy lecithin) Hoeller et al., 2009

(Qian et al., 2012).

The influence of emulsifier type (β-lactoglobulin, lyso-lecithin, Tween, and DTAB) on the formation, properties and stability of polymethoxyflavone-loaded nanoemulsions was examined by Li et al. (2012). These authors showed that nanoemulsions (r < 100 nm) could be formed using high pressure homogenization for all emulsifier types, except DTAB and that the lipid droplet charge could be altered from highly cationic (DTAB), to near neutral (Tween 20),

Pinheiro, A. C. (2013)

to highly anionic (β-lactoglobulin, lyso-lecithin) by varying emulsifier type. Moreover, there were some visible differences in the size, morphology and aggregation of the polymethoxyflavone crystals observed in the emulsions containing different types of emulsifiers (Li et al., 2012).

Other authors investigated the influence of interfacial composition on the in vitro digestion of corn oil-in-water emulsions stabilized by various emulsifiers (sodium caseinate, WPI, lecithin and Tween 20). Their results suggested that the access of pancreatic lipase to emulsified fats decreases in the following order: proteins (caseinate and WPI) > phospholipids (lecithin) > non- ionic surfactants (Tween 20) (Mun et al., 2007).

2.4.3.3 Lipophilic bioactive compounds

Nanoemulsions are one of the most interesting emulsion-based delivery systems to act as carriers for lipophilic compounds such as nutraceuticals, drugs, flavours, antioxidants, antimicrobial agents and lipophilic vitamins. Table 2.7 shows some of the lipophilic bioactive compounds that can be encapsulated in nanoemulsions and have potential applications in food and beverages. (Donsì et al., 2011). Table 2.7 Examples of lipophilic bioactive compounds encapsulated in food nanoemulsions and their potential benefits

Lipophilic bioactive compound Potential benefit Reference

ω-3 fatty acids Anticarcinogenic, prevent heart Chalothorn and Warisnoicharoen, 2012 diseases

α-tocopherol Antioxidant Hatanaka et al., 2010

Curcumin Antioxidant, anti-inflammatory and Ahmed et al., 2012 anticarcinogenic

β-carotene Antioxidant, vitamin A precursor Silva et al., 2011

Lycopene Antioxidant, anticarcinogenic Garti et al., 2005

Phytosterols Cholesterol absorption inhibitor Garti et al., 2005

Quercetin Anti-inflammatory, antioxidant Chen-yu et al., 2012

Co-enzyme Q Antihypertensive, anticarcinogenic, Belhaj et al., 2012 antioxidant, prevents heart diseases

Carvacrol Antimicrobial activity Donsì et al., 2012

Resveratrol Antioxidant, antimicrobial, anti- Donsì et al., 2011 inflammatory, chemopreventive and anti-cancer activity

Chapter 2. Literature Review

The encapsulation of lipophilic bioactive compounds in nanoemulsions contributes to the improvement of their dispersibility in aqueous solutions (minimizing the tendency of phase separation), to protect the bioactive compounds from interaction with food ingredients, keeping their functional properties and preventing deterioration of the food itself, to minimize the impact on organoleptic properties of foods, as well as to improve absorption and bioavailability due to the subcellular size of nanoemulsions, which enhances passive transport mechanisms (i.e. related to the concentration gradient) across the cell membrane

2.4.3.4 Multilayer nanoemulsions

Multilayer nanoemulsions can be produced using the LbL electrostatic deposition technique, which allows the formation of multiple layers of emulsifiers and/or polyelectrolytes around oil droplets. Initially, this approach involves adsorption of an ionic emulsifier to the surface of lipid droplets during homogenization to produce the “primary” emulsion, then an oppositely charged polyelectrolyte is added and adsorbs to the droplet surface, forming the “secondary” emulsion that is now coated with a two-layer interface. This procedure can be repeated to form emulsions coated by three or more layers (McClements, 2012) (Figure 2.7).

Figure 2.7 Utilization of LbL technique for the production of multilayer nanoemulsions.

Polyelectrolytes adsorb to the emulsion surfaces due to electrostatic interactions and if the concentration of polyelectrolyte is high enough, emulsions become saturated and charge reversion will occur. Before adding the polyelectrolyte that will form the second layer around the lipid droplet, it is necessary to ensure that there is a minimum concentration of polyelectrolyte present in the solution; otherwise, it will interact with the oppositely charged polyelectrolyte of the next solution, interfering with the formation of multilayers around the emulsion droplets (Guzey and McClements, 2006).

Pinheiro, A. C. (2013)

In order to avoid droplet aggregation due to the presence of free polyelectrolyte, therefore producing stable multilayer emulsions, different preparation methods have been developed. A washing step (centrifugation or filtration) may be included between each electrostatic deposition step to remove any excess of non-adsorbed polyelectrolyte remaining in the continuous phase. Alternatively, the solution conditions may be optimized so that just enough polyelectrolyte to completely coat the emulsions is added and therefore the non-adsorbed polyelectrolyte remaining in the aqueous phase after the coating step is minimized (saturation method) (McClements, 2012).

Different biopolymers (both proteins and polysaccharides) have been used to form the subsequent layers on top of the primary emulsions in order to provide the desirable properties to the emulsion droplets (Table 2.8).

Multilayer emulsions have been found to have better stability to environmental stresses (e.g. pH, salt, thermal processing, chilling, freezing, dehydration and mechanical agitation) than conventional emulsions with single layer interfaces (Gu et al., 2007; Tokle et al., 2010).

The properties of multilayers can change in response to specific environmental triggers, such as dilution, pH, ionic strength, temperature or enzyme activity, offering the possibility of activating the release of functional agents. Also, their release rate can be controlled by designing thickness and permeability of multilayer coatings surrounding the oil droplets.

Table 2.8 Examples of multilayer emulsions produced using biopolymers

Primary layer Secondary layer Tertiary layer Properties Reference (Emulsifier)

β-lactoglobulin Alginate or ι–carrageenan Better stability to ionic Harnsilawat et al., or gum arabic ____ strength and thermal 2006 processing.

Lactoferrin Low methoxyl pectin or Improved thermal stability Tokle et al., 2010 high methoxyl pectin or ____ alginate

Soybean soluble Chitosan Improved physicochemical Hou et al., 2010 ____ polysaccharides stability

Lecithin Chitosan Pectin Good stability to Ogawa et al., 2004 aggregation over a wide range of pH values and salt concentrations

β-lactoglobulin Chitosan Pectin or alginate Lower lipid digestibility rate Li et al., 2010

Chapter 2. Literature Review

2.5 NANOSTRUCTURES CHARACTERIZATION

The analytical techniques that can be used for characterization of the nanostructured systems are described in this section. The analytical approaches have been subdivided into two groups: physicochemical techniques and imaging techniques, where some are more appropriate for planar systems and other for colloidal systems.

2.5.1 Physicochemical Characterization Techniques

In this section, techniques involving essentially physical determinations (e.g., size, size distribution, ζ-potential, and crystallinity) of the nanostructures are described.

2.5.1.1 Dynamic Light Scattering

Dynamic light scattering (DLS) is a technique that measures Brownian motion and relates this to the size of the particles through Stokes–Einstein equation. DLS is the most versatile and useful analytical tool for measuring in situ the size and size distributions of nanoparticles and nanoemulsions in liquid environment (Liu et al., 2012b; Saberi et al., 2013). However, this technique has a size limitation: it is suitable for the measurement of particle diameters up to a few microns (≤ 10 µm).

2.5.1.2 ζ-potential

ζ-potential analysis is a technique for determining the surface charge of nanoparticles in solution (colloids). The magnitude of the ζ-potential is predictive of the colloidal stability: generally, ζ-potential values above 30 mV (positive or negative) lead to more stable dispersions, because repulsion between the particles prevents their aggregation (Preetz et al., 2010). Dispersions with a low ζ-potential value will eventually aggregate due to Van der Waals inter-particle attractions.

In ζ-potential measurements, an electrical field is applied across the sample and the movement of the nanoparticles (electrophoretic mobility) is measured by laser doppler velocimetry (LDV). Then, the ζ-potential can be calculated using the Henry equation. Nanoparticles with a ζ-potential between −10 and +10 mV are considered approximately neutral, while nanoparticles with ζ-potentials higher than +30 mV or lower than −30 mV are considered strongly cationic and strongly anionic, respectively (Clogston and Patri, 2011).

Pinheiro, A. C. (2013)

In addition of being used for predicting the stability of colloidal nanostructures, the measurement of ζ-potential has been widely used to monitor the alternate deposition of anionic and cationic polyelectrolytes in multilayer nanoparticles (Shu et al., 2010).

2.5.1.3 Quartz Crystal Microbalance

Quartz crystal microbalance (QCM) is a simple, cost effective, high resolution mass sensing technique based on the piezoelectric effect: the mechanical stress applied to the surface of the quartz crystal originates a corresponding electrical potential across the crystal whose magnitude is proportional to the applied stress (Buttry and Ward, 1992). As the mass is deposited onto the crystal surface, the piezoelectric properties of the quartz crystal change its oscillation frequency (Marx, 2003). It has the ability to sensitively measure mass changes associated with liquid–solid interfacial phenomena, as well as to characterize energy dissipation or viscoelastic behaviour of the mass deposited on the electrode surface of the quartz crystal.

If the adsorbed material is rigidly attached, evenly distributed and is much thinner than the mass of the quartz crystal, the resulting decrease in oscillation frequency is proportional to the adsorbed mass and thus can be quantified from the Sauerbrey equation (Sauerbrey, 1959). However, when using polymers, in addition to the measurement of frequency shift, it is necessary to measure the damping of the crystal oscillation provoked by the viscous character of polymer solutions, and the previous relation is no longer valid. The Sauerbrey equation can only be used to calculate elastic mass upon the QCM surface after it has been determined experimentally that the bound mass dissipates no energy (Marx, 2003). Other physical parameters can be measured in addition to the resonance frequency, such as motional resistance, which can be used to measure the viscoelastic change of the deposited film. The simultaneous measurements of resonance frequency and motional resistance can differentiate an elastic mass attachment from viscosity-induced effects: for an elastic mass attachment, the motional resistance will be zero and the resonance frequency will be proportional to the mass increase on the surface of the quartz crystal; when viscosity-induced effects are present, the decrease of resonance frequency is accompanied by a proportional increase of motional resistance (Fakhrullin et al., 2007).

Indest et al. (2009) observed in situ the adsorption behaviour of fucoidan to a chitosan layer, using QCM measurements.

Chapter 2. Literature Review

2.5.1.4 Ultraviolet–Visible Spectroscopy

Ultraviolet–visible (UV/Vis) spectroscopy refers to absorption spectroscopy or reflectance spectroscopy in the ultraviolet–visible spectral region. UV/Vis spectroscopy is being used as a technique to verify if the adsorption of polysaccharides in nanolayered coatings is occurring (Carneiro-da-Cunha et al., 2010; Medeiros et al., 2011), which is evaluated by changes of the absorbance at specified wavelengths.

2.5.1.5 Contact Angle

Contact angle measurement is a technique for surface analysis (in terms of surface energy and tension). It provides information regarding the bonding energy of the solid surface and surface tension of the droplet. Because of its simplicity, contact angle measurement has been used to characterize the surface wettability, adhesion and absorption properties of nanolayered coatings (Yoo et al., 1998). The differences observed in the wettability of nanolayered coatings as the outermost layer changes has been attributed to the differences in the hydrophilicity of the functional groups of the adsorbed layer and allow to confirm the alternate deposition of the polyelectrolytes (Carneiro-da-Cunha et al., 2010; Medeiros et al., 2011; Yoo et al., 1998).

2.5.1.6 Fourier Transform Infrared

Fourier transform infrared (FTIR) technique is extremely accurate and reproducible, and the resulting spectrum represents the molecular absorption and transmission, creating a molecular fingerprint of the sample.

Lawrie et al. studied the interactions between alginate and chitosan in alginate-chitosan polyelectrolyte complexes in the form of a film trough FTIR (Lawrie et al., 2007). Also, other authors verified by ATR-FTIR analysis the coverage of the multilayer film on the treated PET substrate (Channasanon et al., 2007) .

2.5.2 Imaging Techniques

Imaging techniques can be applied in order to enable information and visualization of the shape and aggregation state and to confirm the size of the nanostructures. Some of the imaging methods that are used to characterize nanostructures are presented in the following sections.

Pinheiro, A. C. (2013)

2.5.2.1 Transmission Electron Microscopy

Transmission electron microscopy (TEM) is an imaging technique capable of a resolution of the order of 0.2 nm (Luykx et al., 2008; Wang, 2000). However, TEM has some drawbacks: it requires extensive sample preparation to produce a sample thin enough to be electron transparent (making TEM a relatively time-consuming process), the samples suffer changes during the preparation process, the field of view is small (raising the possibility that the region analysed may not be characteristic of the whole sample) and the electron beam may damage the sample (Luykx et al., 2008; Wang, 2000).

Preetz et al. used TEM to determine the shape, morphology and surface structure of polyelectrolyte nanocapsules and for estimation of the shell thickness (Preetz et al., 2010).

2.5.2.2 Scanning Electron Microscopy

Scanning electron microscopy (SEM) is able to produce high-resolution images of a sample surface (Luykx et al., 2008). SEM images have a characteristic three-dimensional appearance and are useful for the visualization of the surfaces’ structure. SEM allows a large amount of sample to be in focus at one time, the higher magnification produces high-resolution images, and this combined with the larger depth of focus, greater resolution and easy sample observation makes SEM one of the most used techniques in nanostructures characterization. Despite of this, SEM requires a high vacuum and sample conductivity (Luykx et al., 2008; Wang, 2000) thus rendering it inappropriate for certain types of samples.

SEM has been widely used to measure the thickness of the layers that compose a nanolayered coating (Carneiro-da-Cunha et al., 2010) and the diameter of nanocapsules (Shu et al., 2010).

2.5.2.3 Confocal Laser Scanning Microscopy

Confocal laser scanning microscopy (CLSM) allows obtaining high-resolution optical images with depth selectivity. The key feature of confocal microscopy is the ability to acquire in-focus images from selected depths, a process known as optical sectioning, allowing three- dimensional reconstructions of topologically complex objects, which is useful for surface profiling in opaque specimens. In non-opaque specimens, it is possible to see interior structures with enhanced quality of image over simple microscopy. Confocal microscopy allows a direct, non-invasive, serial optical sectioning of intact, thick, living specimens with a minimum of sample preparation as well as a marginal improvement in lateral resolution. Biological samples are often treated with fluorescent dyes to make selected objects visible.

Chapter 2. Literature Review

Moreover, the actual dye concentration can be low to minimize the disturbance of biological systems: some instruments can track single fluorescent molecules.

The CLSM has been used to observe the structure of multilayer emulsions, being the oil phase stained with an oil-specific fluorescent dye (nile red) (Gudipati et al., 2010; Li et al., 2010).

2.5.3 Transport Properties

In food systems, the application of edible coatings has as main objective the control of gas and water vapour exchanges between the food and its surrounding atmosphere, eventually leading to the extension of the products’ shelf-life.

This control may slow down the typical degradation reactions that occur in food systems, being also protected from physical damage along the distribution chain. Although coatings have been applied for a large number of years to food products, the use of these materials at the nanoscale may improve their functionality (Weiss et al., 2006). It is then of utmost importance to characterize the transport of gases and water through nanolayered coatings.

Due to nanoscale nature of these systems, experimental determination of a standalone nano- coating transport properties is not an easy task. To solve this issue, experiments can be done using a well-characterized support on which the coating is constructed using virtually any of the techniques mentioned in previous sections. The transport properties of the coating alone can then be determined by (Cooksey et al., 1999):

Equation 2.1 ( ⁄ ) ( ⁄ ) where

TP refers to the transport property in study

L is the thickness

This approach has been successfully used in nanolayered systems like κ-carrageenan/lysozyme (Medeiros et al., 2011) and chitosan/alginate (Carneiro-da-Cunha et al., 2010).

2.5.4 Controlled Release

In the past years, the controlled release from nanostructures has been extensively studied. Most of these studies are devoted to developing drug delivery systems. In foods, the main applications of controlled release from nanostructures are related with human GI environment when developing particles to deliver nutrients in the intestine (Li et al., 2011a). Also, research

Pinheiro, A. C. (2013)

on “intelligent” packaging of foods shows strong potential to extend shelf-life, which may be achieved e.g. through the release of antimicrobial agents when a packed food is exposed to a temperature abuse (Han and Gennadios, 2005).

Release of an active compound from a nanostructure can be due to the collapse of the nanostructure itself. However, in cases like nanolayered coatings aimed at extending shelf-life of perishable foods, the release may be due to transport phenomena within the nanostructure itself. In these cases, understanding the transport properties and how they are related with the system’s structural rearrangement will be crucial. Moreover, understanding the transport of bioactive compounds in edible nanolayered films/coatings can provide valuable information for the incorporation of compounds able to extend products’ shelf-life, such as antioxidants or antimicrobials. Release of bioactive compounds from bio-polymeric matrices may occur due to mechanisms of Fick’s diffusion, polymer matrix swelling, polymer erosion, and degradation (Faisant et al., 2002; Jain, 2000) and a different mechanism may prevail, depending on the system and environmental conditions. These mechanisms may be generally classified into three different types: ideal Fickian diffusion (Brownian transport), case II transport (polymer relaxation-driven), and anomalous behaviour (ranging from Fickian to case II transport). In recent years, several works concerning the release of functional compounds from conventional edible films have been published (Del Nobile et al., 2008; Flores et al., 2007). Comparatively, few works can be found on release mechanisms involving systems at the nanoscale. Literature suggests that in the case of nanolayered coatings, the release behaviour depends on the permeability and on the disassembly or erosion of the multilayer structure and on other experimental variables (Jiang and Li, 2009). Depending on the nanolayered coatings’ composition and on the released compound, release may follow a Fickian (Polakovic et al., 1999) or an anomalous transport behaviour (Wang et al., 2007a). Also, some authors attribute the observed behaviour to a Fickian transport (Equation 2.2) through the nanolayers, followed by release due to polymer dissolution (Antipov et al., 2001). However, there are still a lot of unknown variables that certainly play a role during the release of compounds from nanolayered coatings, especially regarding the effect of environmental conditions.

( ) Equation 2.2

where

Mt is the total mass released from the polymer at time t

M∞ is the mass of the compound release at equilibrium

Chapter 2. Literature Review

kF is Fickian diffusion rate constant

To account for both Fickian and case II transport effects on the observed anomalous behaviour in hydrophilic matrices, a linear superimposition (Equation 2.3) can be used (Berens and Hopfenberg, 1978; Crank, 1975; Flores et al., 2007). This approach assumes that the observed transport of molecules within the polymer can be described by the sum of the molecules transported due to Brownian motion (Fick’s transport), with the molecules transported due to polymer relaxation:

[ ∑ ( ) ] ( )[ ( )] Equation 2.3

where

Mt is the total mass released from the polymer at time t

M∞ is the mass of the compound released at equilibrium kF is the Fickian diffusion rate constant kR is the relaxation rate constant

X is the fraction of compound released by Fickian transport

2.6 IN VITRO DIGESTIONS

Although in vivo feeding studies usually provide most accurate results than in vitro systems, they are usually expensive, time-consuming, susceptible to subject-to-subject variations, often involve subjecting animals to sacrifice and are limited by ethical constrains when potential harmful compounds are involved. In the recent years, much effort has been dedicated to the development of in vitro models that closely mimic the physiological processes occurring during human digestion and could provide a useful alternative to animal studies, i.e. that provide accurate results in the short time, serving as a tool for rapid screening foods or delivery systems with different compositions and structures (McClements and Li, 2010a).

2.6.1 Human gastrointestinal digestion

Human digestion is a complex process in which two main processes occur simultaneously: mechanical transformations that reduce the size of the food particles and enzymatic transformations where macromolecules are hydrolysed into smaller constituents that are absorbed into the bloodstream (Guerra et al., 2012).

Pinheiro, A. C. (2013)

Figure 2.8 shows the physicochemical conditions on the main stages of the human digestion process. Digestion starts in the mouth, where the food is converted into a form suitable for swallowing. Mouth secretes saliva that contains mucus and the enzyme amylase and mastication reduces the particle size and hydrates the food by mixing it with saliva (Kong and Singh, 2008). Although being a short step, mastication has significant influence on the overall digestive process (Woda et al., 2010).

The food bolus resulting from mechanical and enzymatic degradation in the mouth is then transported through the oesophagus to the stomach by the mechanism of peristalsis, which consists in the advancing of contraction of the walls, forcing the contents forward (Siddiqui et al., 1991).

Figure 2.8 Schematic diagram of physicochemical conditions in the different regions of the human GI tract (adapted from Guerra et al., 2012).

The stomach has three main motor functions: storage, mixing and emptying and is divided into four major regions: fundus, body, antrum and pylorus, each one with different physiological functions (Kong and Singh, 2008). The proximal part (fundus and body) acts as a reservoir for undigested food and secretes gastric juice that contains digestive enzymes (pepsin and lipase) responsible for protein and lipid digestion, and acids that promote hydrolysis. The main function of the antrum is to generate mechanical forces that mix, disrupt large particles and transport the gastric contents. The small particles pass from the stomach into the duodenum via pylorus, whereas particles with larger dimension remain in the stomach, by a mechanism of retropulsion, until further degradation. Gastric emptying is a crucial parameter of digestion influenced by many factors such as food composition or structure and biological factors (Guerra et al., 2012).

Chapter 2. Literature Review

The final stages of digestion occur in the small intestine that is divided into three parts: a short section receiving digestion secretion from the pancreas and liver - the duodenum, and two longer ones - the jejunum and ileum. Its two main roles in digestion are the breakdown of macromolecules and absorption of water and nutrients. In the duodenum, chyme is neutralized with sodium bicarbonate to give an appropriate pH for enzyme activity, pancreatic enzymes (mixture of proteases, amylases and lipases) act to breakdown food constituents and bile (produced in the liver) plays a specific role in lipid digestion by emulsifying lipids and thus promoting pancreatic lipase activity. All of the digested nutrients and water are absorbed through the intestinal walls (villus enterocytes), whereas waste products (non-absorbed material) are propelled into the large intestine for excretion (Kong and Singh, 2008). The main functions of the colon are absorption of water and electrolytes, fermentation of polysaccharides and proteins by colonic microbiota, reabsorption of bile salts and formation, storage and elimination of faeces (Guerra et al., 2012)

2.6.2 In vitro digestion models

Although during the past few years a number of in vitro digestion models have been used to test the structural and chemical changes that occur in different foods or in different delivery systems under simulated GI conditions, none has yet been widely accepted (Hur et al., 2011). This is due to the fact that, despite their complexity, the in vitro digestion models remain simplified compared to the in vivo situation: they do not include feedback mechanisms, resident microbiota, immune system or specific hormonal controls (Guerra et al., 2012).

Many attempts to model the human stomach and small intestine have been made in the last years. The different in vitro digestion models differ from one another in the number of steps included in the digestion, the composition of digestive fluids used in each step and on the mechanical stresses and fluid flows used. Also, in vitro digestion models can be static or dynamic. In static models, there is no absorption process, i.e., the products of digestion are not removed during digestion and they do not mimic the physical processes that occur in vivo (e.g. shear, mixing). In dynamic models the products of digestion may or may not be removed, but they include the physical processing and temporal changes in luminal conditions that mimic conditions in vivo (Wickham et al., 2009). Most of in vitro models are static, include a limited number of simulated parameters and are dedicated to a particular application. In this approach, the gastric phase is reproduced by pepsin hydrolysis of homogenized food, under fixed pH and temperature for a set period of time and this step is usually followed, in the same reactor, by an intestinal phase involving pancreatic enzymes and bile acids (Guerra et al.,

Pinheiro, A. C. (2013)

2012). None of the static models reproduce the dynamic processes occurring during human digestion (gastric emptying, continuous change in pH, peristaltic movements and secretion flow rates).

2.6.2.1 pH-stat

The pH-stat method is an analytical tool that is finding increasing applications within pharmaceutical and food research for the in vitro characterization of lipid digestion under simulated small intestinal conditions (Li and McClements, 2010). This method is based on measurements of the amounts of free fatty acids (FFA) released from lipids, usually triglycerides (TAG), after lipase addition at pH values close to neutral (Li et al., 2011b). The sample containing the lipids is placed in a temperature-controlled reaction chamber with appropriate concentration of small intestinal fluids (e.g. lipase, bile salts and minerals) (Figure 2.9). The lipase catalyses lipid digestion leading to the formation of two FFA and one monoacylglycerides (MAG) per TAG molecule.

Figure 2.9 Schematic representation of pH-stat in vitro digestion model used to determine the digestion and release of lipid droplets (Li and McClements, 2010).

The concentration of NaOH that must be titrated into the digestion chamber to neutralize the FFA produced during lipid digestion, and thereby maintain the pH at the initial pre-set value (e.g., pH 7.0) is recorded versus time (Li et al., 2011b).

This method is simple and rapid to carry on and can be used to rapidly screen the impact of different physicochemical factors on lipid digestion (Li and McClements, 2010).

2.6.2.2 Dynamic Gut Model (DGM)

This dynamic monocompartmental model was developed by the Institute of Food Research (IFR, Norwich, U.K.) from knowledge gained from echo-planar magnetic resonance imaging

Chapter 2. Literature Review studies on the gastric processing of complex meals in human volunteers (McAllister, 2010). These studies allowed to collect essential data on the digestion of multiphase meals and the influence of structure, hydration, mixing, shear, transport and delivery within the GI tract (Wickham et al., 2009). The DGM is claimed to replicate these conditions and to provide an accurate in vitro simulation. This model was designed to take into account the region specificity of the stomach and it is composed of two successive compartments (Figure 2.10). The first stage mimics the main body of the stomach (fundal region), where the gastric secretions are mixed with food and the second stage replicates the shear forces and mixing observed in the antral region (McAllister, 2010). Despite its complexity, the DGM does not accurately reproduce the in vivo peristaltic forces (Guerra et al., 2012).

Figure 2.10 DGM showing the main body/fundal area (1) and antrum (2) (McAllister, 2010).

2.6.2.3 Human Gastric Simulator (HGS) model

The HGS is a dynamic stomach model that has been designed in a way to produce continuous peristaltic movements of stomach wall comparable to the ones observed in vivo. The HGS is composed of a latex chamber, simulating the stomach, and a mechanical driving system composed of 12 rollers secured on belts pushing the stomach walls and creating a continuous contraction of the latex walls (Figure 2.11). It also incorporates gastric secretion, emptying systems and temperature control, enabling accurate simulation of dynamic gastric digestion process (Kong and Singh, 2010).

The HGS provides a reasonably realistic set of conditions that mimic the human digestion process. The incorporation of peristaltic movements into the wall of the latex chamber enables the successful reproduction of peristaltic contractions of stomach wall in amplitude, intensity and frequency, creating patterns similar to those of in vivo mechanical forces and providing a more accurate simulation of gastric digestion process. This model is a useful tool to evaluate

Pinheiro, A. C. (2013)

the transformation of food constituents during simulated gastric digestion (Kong and Singh, 2010).

Figure 2.11 Human gastric simulator. (1) Motor; (2) latex lining; (3) mesh bag; (4) secretion tubing; (5) roller; (6) belt; (7) light bulb for temperature control; (8) plastic foam insulation (Kong and Singh, 2010).

2.6.2.4 TIM-1 model

The TIM-1 system was developed at TNO Nutrition and Food Research centre (Zeist, The Netherlands) and patented (Minekus and Havenaar, 1996). This model is a multicompartimental dynamic computer controlled system that simulates the human GI tract (McAllister, 2010). The TIM-1 model combines multi-compartmentalization and dynamism and is so far the only gastric and small-intestinal system characterized as “full” (Minekus et al., 1995) and the one that allows the closest simulation of the in vivo dynamic events occurring within the human gastric and small intestinal lumen (Guerra et al., 2012). The parameters used in the TIM-1 model design are based on data obtained from human volunteer studies and it integrates key parameters of human digestion, such as temperature, pH, gastric and ileal deliveries, transit time, peristaltic mixing and transport, sequential addition of gastric secretion (lipase, pepsin, HCl) and small intestine secretion (pancreatic juice, bile and sodium bicarbonate) and passive absorption of water and small molecules through a dialysis system (McAllister, 2010).

This model consists in four successive compartments simulating the stomach, the duodenum and the ileum (Figure 2.12). Each compartment is formed by two connected units consisting of a glass jacket with a flexible wall inside. Water is pumped into the area between the glass jacket and the flexible walls to maintain the temperature at 37 °C and to squeeze the walls. The changes in water pressure that originate the alternate compression and relaxation of the

Chapter 2. Literature Review flexible walls are achieved by computer-activated rotary pumps and ensure the mixing the chyme. The transit of the contents through the TIM-1 is regulated by opening and closing peristaltic valves that connect each compartment.

Figure 2.12 Front panel of the TIM-1 system and schematic diagram showing the stomach compartment (A) and three small- intestinal compartments: duodenum (C), jejunum (E), and ileum (G) connected by vertical peristaltic valves, including the pyloric sphincter (B) and the ileo-cecal valve (H). Each compartment with secretion tubes, such as gastric acid and pepsin in the stomach (I), and pancreatin and bile in the duodenum (J), pH electrodes (P), pressure (S) or level sensors (Q) and temperature sensors (R). Connected to the jejunum (left) and ileum (right) are semi-permeable hollow-fibre membrane units (M) for continuous absorption of digested products and water absorption (N).

The compartments are equipped with pH electrodes and the pH values are computer-

-1 -1 controlled by secreting 1 mol.L HCl into the stomach and 1 mol.L NaHCO3 into the small- intestine compartments, via syringe pumps (Krul et al., 2000). The absorption phase is simulated by the use of dialysis membranes which removes water and small molecules

(including the products of digestion and the dissolved bioactive compound) (McAllister, 2010).

The jejunum and ileum compartments are each connected to filtration units (semi-permeable hollow-fibre devices with a molecular cut-off of 5 kDa) which allow the bioaccessibility to be quantified (Verwei et al., 2006). Nonbioaccessible fractions (ileal delivery) are collected at the end of ileum compartment and represent the unabsorbed material that will enter the large intestine.

2.6.3 Behaviour of nanostructures in the GI tract

The knowledge of the behaviour of nanostructures as well as the fate of the bioactive compounds encapsulated within them in the GI tract is of utmost importance for optimizing the bioactivity of encapsulated compounds and to ensure that these structures are safe for

Pinheiro, A. C. (2013)

human consumption (McClements and Xiau, 2011). Conversely to the extensive work focusing on the behaviour of lipid-based delivery systems during in vitro digestion, few works can be found in literature about the behaviour of nanocapsules or nanofilms within the GI tract.

The fate of nanoemulsions within the GI tract can be potentially altered by their initial characteristics, such as particle size, initial emulsifier and oil type, therefore the impact of those characteristics has been extensively studied in the last years. Salvia-Trujillo and co- workers evaluated the influence of particle size on lipid digestion and on β-carotene bioaccessibility using emulsions with different initial droplet diameters. These authors observed that the rate and extent of lipid digestion increased with decreasing droplet diameter, which was attributed to the increase in lipid surface area exposed to lipases with decreasing droplet size and also that β-carotene bioaccessibility increased as the initial droplet size decreased, which could be related to the fact that there was more undigested oil present for large droplets, retaining the β-carotene (Salvia-Trujillo et al., 2013). Recently, Liu et al. evaluated the influence of the interfacial structure in the bioaccessibility and microstructural changes of β-carotene emulsions during in vitro digestion. Results of this work showed that the initial emulsifier (whey protein isolates (WPI), soybean soluble polysaccharides or decaglycerol monolaurate (ML750)) used to stabilize oil-in-water emulsions had a significant effect on the droplet size, particle electric charge and microstructure change during the digestion of β- carotene droplets. The emulsion stabilized by ML750 exhibited the highest release rate in the gastric fluids, whereas WPI-emulsion showed drastic droplet changes and the most significant release rate in intestine (Liu et al., 2012a). The influence of droplet composition on lipid digestibility and curcumin bioaccessibility was examined by other authors (Ahmed et al., 2012). They found that long chain triglycerides (LCT) were digested to a lesser extent than short or medium chain triglycerides (SCT or MCT), which was attributed to the fact that longer chain fatty acids digestion products, contrary to shorter chain fatty acids, present low water dispersibility and tend therefore to accumulate at the oil/water interface, preventing lipase from accessing the non-digested emulsified lipids. On the other hand, the bioaccessibility of curcumin was very low when SCT was used as the oil phase and the highest bioaccessibility was obtained with MCT. The authors attributed these results to the fact that MCT and LCT can form mixed micelles capable of solubilising curcumin, whereas SCT cannot.

One of the few works regarding the evaluation of the behaviour of polymeric nanoparticles in the GI tract was recently developed. In this work, lutein, a non-provitamin-A carotenoid, was encapsulated in nanoparticles composed of water-soluble low molecular weight chitosan and the lutein bioavailability was studied in vitro and in vivo, using lutein in mixed micelles as

Chapter 2. Literature Review control. The bioavailability of lutein in nanoparticles was significantly higher (27.7%) than control, and also the lutein level in plasma (54.5%), liver (53.9%) and eyes (62.8%) of mice fed on nanoencapsulated lutein were higher compared with the control (Arunkumar et al., 2013).

There is still a lack of knowledge regarding the biological fate of ingested bio-nanostructures and further research is needed to both access its safety and to produce tailored delivery systems (i.e. with optimized bioactivity). Also, being infeasible the use of in vivo models (high costs and ethical constrains often involved), there is a need of using more realistic in vitro gastrointestinal models, i.e. models that can accurately simulate the complex physicochemical and physiological processes that occur within the human gastrointestinal tract.

Pinheiro, A. C. (2013)

2.7 REFERENCES

Ahmed, K., Li, Y., McClements, D. J., & Xiao, H. (2012). Nanoemulsion- and emulsion- based delivery systems for curcumin: Encapsulation and release properties, Food Chemistry 132(2), 799-807. Ai, H., & Gao, J. (2004). Size-controlled polyelectrolyte nanocapsules via layer-by-layer self-assembly, Journal of Materials Science 39(4). Aisa, Y., Miyakawa, Y., Nakazato, T., Shibata, H., Saito, K., Ikeda, Y., & Kizaki, M. (2005). Fucoidan induces apoptosis of human HS-Sultan cells accompanied by activation of caspase-3 and down-regulation of ERK Pathways, American Journal of Hematology 78(1), 7-14. Antipov, A. A., Sukhorukov, G. B., Donath, E., & Möhwald, H. (2001). Sustained Release Properties of Polyelectrolyte Multilayer Capsules, The Journal of Physical Chemistry B 105(12), 2281-2284. Anton, N., Benoit, J.-P., & Saulnier, P. (2008). Design and production of nanoparticles formulated from nano-emulsion templates—A review, Journal of Controlled Release 128(3), 185-199. Arunkumar, R., Harish Prashanth, K. V., & Baskaran, V. (2013). Promising interaction between nanoencapsulated lutein with low molecular weight chitosan: Characterization and bioavailability of lutein in vitro and in vivo, Food Chemistry 141(1), 327-337. Belhaj, N., Dupuis, F., Arab-Tehrany, E., Denis, F. M., Paris, C., Lartaud, I., & Linder, M. (2012). Formulation, characterization and pharmacokinetic studies of coenzyme Q10 PUFA’s nanoemulsions, European Journal of Pharmaceutical Sciences 47(2), 305-312. Berens, A. R., & Hopfenberg, H. B. (1978). Diffusion and relaxation in glassy polymer powders. 2. Separation of diffusion and relaxation parameters, Polymer 19(5), 489-496. Bilan, M. I., Grachev, A. A., Ustuzhanina, N. E., Shashkov, A. S., Nifantiev, N. E., & Usov, A. I. (2004). A highly regular fraction of a fucoidan from the brown seaweed Fucus distichus L, Carbohydrate Research 339(3), 511-517. Blodgett, K. B. (1934). Monomolecular films of fatty acids on glass, Journal of the American Chemical Society 56(2), 495-495. Burana-osot, J., Hosoyama, S., Nagamoto, Y., Suzuki, S., Linhardt, R. J., & Toida, T. (2009). Photolytic depolymerization of alginate, Carbohydrate Research 344(15), 2023-2027. Burapapadh, K., Kumpugdee-Vollrath, M., Chantasart, D., & Sriamornsak, P. (2010). Fabrication of pectin-based nanoemulsions loaded with itraconazole for pharmaceutical application, Carbohydrate Polymers 82(2), 384-393.

Chapter 2. Literature Review

Buttry, D. A., & Ward, M. D. (1992). Measurement of interfacial processes at electrode surfaces with the electrochemical quartz crystal microbalance, Chemical Reviews 92(6), 1355- 1379. Carneiro-da-Cunha, M. G., Cerqueira, M. A., Souza, B. W. S., Carvalho, S., Quintas, M. A. C., Teixeira, J. A., & Vicente, A. A. (2010). Physical and thermal properties of a chitosan/alginate nanolayered PET film, Carbohydrate Polymers 82(1), 153-159. Caruso, F., Caruso, R. A., & Möhwald, H. (1998). Nanoengineering of Inorganic and Hybrid Hollow Spheres by Colloidal Templating, Science 282(5391), 1111-1114. Caruso, F., Yang, W., Trau, D., & Renneberg, R. (2000). Microencapsulation of Uncharged Low Molecular Weight Organic Materials by Polyelectrolyte Multilayer Self- Assembly†, Langmuir 16(23), 8932-8936. Chakraborty, S., Shukla, D., Mishra, B., & Singh, S. (2009). Lipid – An emerging platform for oral delivery of drugs with poor bioavailability, European Journal of Pharmaceutics and Biopharmaceutics 73(1), 1-15. Chalothorn, K., & Warisnoicharoen, W. (2012). Ultrasonic emulsification of whey protein isolate-stabilized nanoemulsions containing omega-3 oil from plant seed, American Journal of Food Technology 7(9), 532-541. Chan, E.-S., Lee, B.-B., Ravindra, P., & Poncelet, D. (2009). Prediction models for shape and size of ca-alginate macrobeads produced through extrusion-dripping method, Journal of Colloid and Interface Science 338(1), 63-72. Channasanon, S., Graisuwan, W., Kiatkamjornwong, S., & Hoven, V. P. (2007). Alternating bioactivity of multilayer thin films assembled from charged derivatives of chitosan, Journal of Colloid and Interface Science 316, 331-343. Chaudhry, Q., Scotter, M., Blackburn, J., Ross, B., Boxall, A., Castle, L., Aitken, R., & Watkins, R. (2008). Applications and implications of nanotechnologies for the food sector, Food Additives & Contaminants: Part A 25(3), 241-258. Chen-yu, G., Chun-fen, Y., Qi-lu, L., Qi, T., Yan-wei, X., Wei-na, L., & Guang-xi, Z. (2012). Development of a Quercetin-loaded nanostructured lipid carrier formulation for topical delivery, International Journal of Pharmaceutics 430(1–2), 292-298. Chevolot, L., Mulloy, B., Ratiskol, J., Foucault, A., & Colliec-Jouault, S. (2001). A disaccharide repeat unit is the major structure in fucoidans from two species of brown algae, Carbohydrate Research 330(4), 529-535. Chu, B.-S., Ichikawa, S., Kanafusa, S., & Nakajima, M. (2007). Preparation of Protein- Stabilized β-Carotene Nanodispersions by Emulsification–Evaporation Method, Journal of the American Oil Chemists' Society 84(11), 1053-1062.

Pinheiro, A. C. (2013)

Chung, A. J., & Rubner, M. F. (2002). Methods of Loading and Releasing Low Molecular Weight Cationic Molecules in Weak Polyelectrolyte Multilayer Films, Langmuir 18(4), 1176- 1183. Clogston, J., & Patri, A. (2011) Zeta Potential Measurement, In Characterization of Nanoparticles Intended for Drug Delivery (McNeil, S. E., Ed.), pp 63-70, Humana Press. Cooksey, K., Marsh, K. S., & Doar, L. H. (1999). Predicting permeability & transmission rate for multilayer materials, Food Technology 53(9), 60-63. Crank, J. (1975) The mathematics of diffusion, 2 ed., Clarendon Press, Oxford. Cuomo, F., Lopez, F., Miguel, M. G., & Lindman, B. r. (2010). Vesicle-Templated Layer- by-Layer Assembly for the Production of Nanocapsules, Langmuir 26(13), 10555-10560. Dash, M., Chiellini, F., Ottenbrite, R. M., & Chiellini, E. (2011). Chitosan--A versatile semi-synthetic polymer in biomedical applications, Progress in Polymer Science 36(8), 981- 1014. de Araújo, S. C., de Mattos, A. C. A., Teixeira, H. F., Coelho, P. M. Z., Nelson, D. L., & de Oliveira, M. C. (2007). Improvement of in vitro efficacy of a novel schistosomicidal drug by incorporation into nanoemulsions, International Journal of Pharmaceutics 337(1–2), 307-315. de Britto, D., de Moura, M. R., Aouada, F. A., Mattoso, L. H. C., & Assis, O. B. G. (2012). N,N,N-trimethyl chitosan nanoparticles as a vitamin carrier system, Food Hydrocolloids 27(2), 487-493. de Villiers, M. M., Otto, D. P., Strydom, S. J., & Lvov, Y. M. (2011). Introduction to nanocoatings produced by layer-by-layer (LbL) self-assembly, Advanced Drug Delivery Reviews 63(9), 701-715. Decher, G., & Hong, J.-D. (1991). Buildup of ultrathin multilayer films by a self- assembly process, 1 consecutive adsorption of anionic and cationic bipolar amphiphiles on charged surfaces, Makromolekulare Chemie. Macromolecular Symposia 46(1), 321-327. Del Nobile, M. A., Conte, A., Incoronato, A. L., & Panza, O. (2008). Antimicrobial efficacy and release kinetics of thymol from zein films, Journal of Food Engineering 89(1), 57- 63. Dickinson, E. (2003). Hydrocolloids at interfaces and the influence on the properties of dispersed systems, Food Hydrocolloids 17(1), 25-39. Dobrynin, A. V., & Rubinstein, M. (2005). Theory of polyelectrolytes in solutions and at surfaces, Progress in Polymer Science 30(11), 1049-1118. Donsì, F., Annunziata, M., Vincensi, M., & Ferrari, G. (2012). Design of nanoemulsion- based delivery systems of natural antimicrobials: Effect of the emulsifier, Journal of Biotechnology 159(4), 342-350.

Chapter 2. Literature Review

Donsì, F., Sessa, M., Mediouni, H., Mgaidi, A., & Ferrari, G. (2011). Encapsulation of bioactive compounds in nanoemulsion- based delivery systems, Procedia Food Science 1(0), 1666-1671. Draget, K. I. (2009) Alginates, In Handbook of hydrocolloids (Phillips, G. O., and Williams, P. A., Eds.) 2nd ed., Woodhead Publishing Limited, Cambridge, UK. Duncan, T. V. (2011). Applications of nanotechnology in food packaging and food safety: Barrier materials, antimicrobials and sensors, Journal of Colloid and Interface Science 363(1), 1-24. Faisant, N., Siepmann, J., & Benoit, J. P. (2002). PLGA-based microparticles: elucidation of mechanisms and a new, simple mathematical model quantifying drug release, European Journal of Pharmaceutical Sciences 15(4), 355-366. Fakhrullin, R. F., Vinter, V. G., Zamaleeva, A. I., Matveeva, M. V., Kourbanov, R. A., Temesgen, B. K., Ishmuchametova, D. G., Abramova, Z. I., Konovalova, O. A., & Salakhov, M. K. (2007). Quartz crystal microbalance immunosensor for the detection of antibodies to double- stranded DNA, Analytical and Bioanalytical Chemistry 388(2), 367-375. Farnaud, S., & Evans, R. W. (2003). Lactoferrin—a multifunctional protein with antimicrobial properties, Molecular Immunology 40(7), 395-405. Fasolo, D., Schwingel, L., Holzschuh, M., Bassani, V., & Teixeira, H. (2007). Validation of an isocratic LC method for determination of quercetin and methylquercetin in topical nanoemulsions, Journal of Pharmaceutical and Biomedical Analysis 44(5), 1174-1177. Fathi, M., Mozafari, M. R., & Mohebbi, M. (2012). Nanoencapsulation of food ingredients using lipid based delivery systems, Trends in Food Science & Technology 23(1), 13-27. Flores, S., Conte, A., Campos, C., Gerschenson, L., & Del Nobile, M. (2007). Mass transport properties of tapioca-based active edible films, Journal of Food Engineering 81(3), 580-586. Fu, J., Ji, J., Yuan, W., & Shen, J. (2005). Construction of anti-adhesive and antibacterial multilayer films via layer-by-layer assembly of heparin and chitosan, Biomaterials 26(33), 6684- 6692. García-Montoya, I. A., Cendón, T. S., Arévalo-Gallegos, S., & Rascón-Cruz, Q. (2012). Lactoferrin a multiple bioactive protein: An overview, Biochimica et Biophysica Acta (BBA) - General Subjects 1820(3), 226-236. Garti, N., Spernath, A., Aserin, A., & Lutz, R. (2005). Nano-sized self-assemblies of nonionic surfactants as solubilization reservoirs and microreactors for food systems, Soft Matter 1(3), 206-218.

Pinheiro, A. C. (2013)

González-Chávez, S. A., Arévalo-Gallegos, S., & Rascón-Cruz, Q. (2009). Lactoferrin: structure, function and applications, International Journal of Antimicrobial Agents 33(4), 301.e301-301.e308. Grigoriev, D. O., & Miller, R. (2009). Mono- and multilayer covered drops as carriers, Current Opinion in Colloid & Interface Science 14(1), 48-59. Gu, Y. S., Decker, E. A., & Julian McClements, D. (2007). Application of multi- component biopolymer layers to improve the freeze–thaw stability of oil-in-water emulsions: β-Lactoglobulin–ι-carrageenan–gelatin, Journal of Food Engineering 80(4), 1246-1254. Gudipati, V., Sandra, S., McClements, D. J., & Decker, E. A. (2010). Oxidative Stability and in Vitro Digestibility of Fish Oil-in-Water Emulsions Containing Multilayered Membranes, Journal of Agricultural and Food Chemistry 58(13), 8093-8099. Guerra, A., Etienne-Mesmin, L., Livrelli, V., Denis, S., Blanquet-Diot, S., & Alric, M. (2012). Relevance and challenges in modeling human gastric and small intestinal digestion, Trends in Biotechnology 30(11), 591-600. Guzey, D., & McClements, D. J. (2006). Formation, stability and properties of multilayer emulsions for application in the food industry, Advances in Colloid and Interface Science 128– 130(0), 227-248. Han, J. H., & Gennadios, A. (2005) Edible films and coatings:a Review, In Innovations in food packagins (Han, J. H., Ed.), pp 239-262, Elsevier, Amesterdam. Harnsilawat, T., Pongsawatmanit, R., & McClements, D. J. (2006). Stabilization of Model Beverage Cloud Emulsions Using Protein−Polysaccharide Electrostatic Complexes Formed at the Oil−Water Interface, Journal of Agricultural and Food Chemistry 54(15), 5540- 5547. Hatanaka, J., Chikamori, H., Sato, H., Uchida, S., Debari, K., Onoue, S., & Yamada, S. (2010). Physicochemical and pharmacological characterization of α-tocopherol-loaded nanoemulsion system, International Journal of Pharmaceutics 396(1–2), 188-193. Helander, I. M., Nurmiaho-Lassila, E. L., Ahvenainen, R., Rhoades, J., & Roller, S. (2001). Chitosan disrupts the barrier properties of the outer membrane of Gram-negative bacteria, International Journal of Food Microbiology 71(2-3), 235-244. Hirmo, S., Utt, M., Ringner, M., & Wadström, T. (1995). Inhibition of heparan sulphate and other glycosaminoglycans binding to Helicobacter pylori by various polysulphated carbohydrates, FEMS Immunology And Medical Microbiology 10(3-4), 301-306. Hoeller, S., Sperger, A., & Valenta, C. (2009). Lecithin based nanoemulsions: A comparative study of the influence of non-ionic surfactants and the cationic phytosphingosine

Chapter 2. Literature Review on physicochemical behaviour and skin permeation, International Journal of Pharmaceutics 370(1–2), 181-186. Hou, Z., Gao, Y., Yuan, F., Liu, Y., Li, C., & Xu, D. (2010). Investigation into the Physicochemical Stability and Rheological Properties of β-Carotene Emulsion Stabilized by Soybean Soluble Polysaccharides and Chitosan, Journal of Agricultural and Food Chemistry 58(15), 8604-8611. Howe, A. M., & Pitt, A. R. (2008). Rheology and stability of oil-in-water nanoemulsions stabilised by anionic surfactant and gelatin 2) addition of homologous series of sugar-based co- surfactants, Advances in Colloid and Interface Science 144(1–2), 30-37. Hur, S. J., Lim, B. O., Decker, E. A., & McClements, D. J. (2011). In vitro human digestion models for food applications, Food Chemistry 125(1), 1-12. Imeson, A. P. (2009) Carrageenan and furcellaran, In Handbook of hydrocolloids (Phillips, G. O., and Williams, P. A., Eds.) 2nd ed., Published by Woodhead Publishing Limited, Cambridge, UK. Indest, T., Laine, J., Johansson, L. S., Stana-Kleinschek, K., Strnad, S., Dworczak, R., & Ribitsch, V. (2009). Adsorption of Fucoidan and Chitosan Sulfate on Chitosan Modified PET Films Monitored by QCM-D, Biomacromolecules 10(3), 630-637. Jafari, S. M., He, Y., & Bhandari, B. (2007). Production of sub-micron emulsions by ultrasound and microfluidization techniques, Journal of Food Engineering 82(4), 478-488. Jain, R. A. (2000). The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices, Biomaterials 21(23), 2475-2490. Jiang, B. B., & Li, B. Y. (2009). Tunable drug loading and release from polypeptide multilayer nanofilms, International Journal of Nanomedicine 4(1), 37-53. Johnston, A. P. R., Cortez, C., Angelatos, A. S., & Caruso, F. (2006). Layer-by-layer engineered capsules and their applications, Current Opinion in Colloid & Interface Science 11(4), 203-209. Jones, O. G., & McClements, D. J. (2010). Functional Biopolymer Particles: Design, Fabrication, and Applications, Comprehensive Reviews in Food Science and Food Safety 9(4), 374-397. Kaschak, D. M., Lean, J. T., Waraksa, C. C., Saupe, G. B., Usami, H., & Mallouk, T. E. (1999). Photoinduced energy and electron transfer reactions in lamellar polyanion/polycation thin films: Toward an inorganic "leaf", Journal of the American Chemical Society 121(14), 3435- 3445. Klöck, G., Pfeffermann, A., Ryser, C., Gröhn, P., Kuttler, B., Hahn, H.-J., & Zimmermann, U. (1997). Biocompatibility of mannuronic acid-rich alginates, Biomaterials 18(10), 707-713.

Pinheiro, A. C. (2013)

Kong, F., & Singh, R. P. (2008). Disintegration of Solid Foods in Human Stomach, Journal of Food Science 73(5), R67-R80. Kong, F., & Singh, R. P. (2010). A Human Gastric Simulator (HGS) to Study Food Digestion in Human Stomach, Journal of Food Science 75(9), E627-E635. Kotov, N. A. (2003) Layer-by-Layer Assembly of Nanoparticles and Nanocolloids: Intermolecular Interactions, Structure and Materials Perspectives, Wiley-VCH Verlag GmbH & Co. KGaA. Krul, C., Luiten-Schuite, A., Baan, R., Verhagen, H., Mohn, G., Feron, V., & Havenaar, R. (2000). Application of a dynamic in vitro gastrointestinal tract model to study the availability of food mutagens, using heterocyclic aromatic amines as model compounds, Food and Chemical Toxicology 38(9), 783-792. Kusaykin, M., Bakunina, I., Sova, V., Ermakova, S., Kuznetsova, T., Besednova, N., Zaporozhets, T., & Zvyagintseva, T. (2008). Structure, biological activity, and enzymatic transformation of fucoidans from the brown seaweeds, Biotechnology Journal 3(7), 904-915. Langmuir, I. (1917). The constitution and fundamental properties of solids and liquids. II. Liquids.1, Journal of the American Chemical Society 39(9), 1848-1906. Lavalle, P., Gergely, C., Cuisinier, F. J. G., Decher, G., Schaaf, P., Voegel, J. C., & Picart, C. (2002). Comparison of the structure of polyelectrolyte multilayer films exhibiting a linear and an exponential growth regime: An in situ atomic force microscopy study, Macromolecules 35(11), 4458-4465. Lawrie, G., Keen, I., Drew, B., Chandler-Temple, A., Rintoul, L., Fredericks, P., & Grøndahl, L. (2007). Interactions between Alginate and Chitosan Biopolymers Characterized Using FTIR and XPS, Biomacromolecules 8(8), 2533-2541. Lee, S. J., & McClements, D. J. (2010). Fabrication of protein-stabilized nanoemulsions using a combined homogenization and amphiphilic solvent dissolution/evaporation approach, Food Hydrocolloids 24(6–7), 560-569. Li, Q., Liu, C.-G., Huang, Z.-H., & Xue, F.-F. (2011a). Preparation and Characterization of Nanoparticles Based on Hydrophobic Alginate Derivative as Carriers for Sustained Release of Vitamin D3, Journal of Agricultural and Food Chemistry 59(5), 1962-1967. Li, Y., Hu, M., & McClements, D. J. (2011b). Factors affecting lipase digestibility of emulsified lipids using an in vitro digestion model: Proposal for a standardised pH-stat method, Food Chemistry 126(2), 498-505. Li, Y., Hu, M., Xiao, H., Du, Y., Decker, E. A., & McClements, D. J. (2010). Controlling the functional performance of emulsion-based delivery systems using multi-component biopolymer coatings, European Journal of Pharmaceutics and Biopharmaceutics 76(1), 38-47.

Chapter 2. Literature Review

Li, Y., & McClements, D. J. (2010). New Mathematical Model for Interpreting pH-Stat Digestion Profiles: Impact of Lipid Droplet Characteristics on in Vitro Digestibility, Journal of Agricultural and Food Chemistry 58(13), 8085-8092. Li, Y., & McClements, D. J. (2011). Inhibition of lipase-catalyzed hydrolysis of emulsified triglyceride oils by low-molecular weight surfactants under simulated gastrointestinal conditions, European Journal of Pharmaceutics and Biopharmaceutics 79(2), 423-431. Li, Y., Zheng, J., Xiao, H., & McClements, D. J. (2012). Nanoemulsion-based delivery systems for poorly water-soluble bioactive compounds: Influence of formulation parameters on polymethoxyflavone crystallization, Food Hydrocolloids 27(2), 517-528. Liu, W., Sun, D., Li, C., Liu, Q., & Xu, J. (2006). Formation and stability of paraffin oil-in- water nano-emulsions prepared by the emulsion inversion point method, Journal of Colloid and Interface Science 303(2), 557-563. Liu, Y., Hou, Z., Lei, F., Chang, Y., & Gao, Y. (2012a). Investigation into the bioaccessibility and microstructure changes of β-carotene emulsions during in vitro digestion, Innovative Food Science & Emerging Technologies 15(0), 86-95. Liu, Y., Yang, J., Zhao, Z., Li, J., Zhang, R., & Yao, F. (2012b). Formation and characterization of natural polysaccharide hollow nanocapsules via template layer-by-layer self-assembly, Journal of Colloid and Interface Science 379(1), 130-140. Luykx, D. M. A. M., Peters, R. J. B., van Ruth, S. M., & Bouwmeester, H. (2008). A Review of Analytical Methods for the Identification and Characterization of Nano Delivery Systems in Food, Journal of Agricultural and Food Chemistry 56(18), 8231-8247. Mao, Y., & McClements, D. J. (2012). Modulation of emulsion rheology through electrostatic heteroaggregation of oppositely charged lipid droplets: Influence of particle size and emulsifier content, Journal of Colloid and Interface Science 380(1), 60-66. Marx, K. A. (2003). Quartz crystal microbalance: A useful tool for studying thin polymer films and complex biomolecular systems at the solution-surface interface, Biomacromolecules 4(5), 1099-1120. Mason, T. G., Wilking, J. N., Meleson, K., Chang, C. B., & Graves, S. M. (2006). Nanoemulsions: formation, structure, and physical properties, Journal of Physics: Condensed Matter 18(41), R635. Matalanis, A., Jones, O. G., & McClements, D. J. (2011). Structured biopolymer-based delivery systems for encapsulation, protection, and release of lipophilic compounds, Food Hydrocolloids 25(8), 1865-1880. McAllister, M. (2010). Dynamic Dissolution: A Step Closer to Predictive Dissolution Testing?, Molecular Pharmaceutics 7(5), 1374-1387.

Pinheiro, A. C. (2013)

McClements, D. J. (2012). Advances in fabrication of emulsions with enhanced functionality using structural design principles, Current Opinion in Colloid & Interface Science 17(5), 235-245. McClements, D. J., & Li, Y. (2010a). Review of in vitro digestion models for rapid screening of emulsion-based systems, Food & Function 1(1), 32-59. McClements, D. J., & Li, Y. (2010b). Structured emulsion-based delivery systems: Controlling the digestion and release of lipophilic food components, Advances in Colloid and Interface Science 159(2), 213-228. McClements, D. J., & Xiao, H. (2012). Potential biological fate of ingested nanoemulsions: influence of particle characteristics, Food & Function 3(3), 202-220. Medeiros, B., Pinheiro, A., Teixeira, J., Vicente, A., & Carneiro-da-Cunha, M. (2011). Polysaccharide/Protein Nanomultilayer Coatings: Construction, Characterization and Evaluation of Their Effect on ‘Rocha’ Pear Shelf-Life, Food and Bioprocess Technology, 1-11. Medeiros, B. G. S., Pinheiro, A. C., Carneiro-da-Cunha, M. G., & Vicente, A. A. (2012). Development and characterization of a nanomultilayer coating of pectin and chitosan – Evaluation of its gas barrier properties and application on ‘Tommy Atkins’ mangoes, Journal of Food Engineering 110(3), 457-464. Minekus, M., & Havenaar, R. (1996). United States Patent nr. 5,525,305. Minekus, M., Marteau, P., Havenaar, R., & ., J. H. J. H. I. t. V. (1995) A multicompartmental dynamic computer-controlled model simulating the stomach and small intestine, Vol. 23, Fund for the Replacement of Animals in Medical Experiments, Nottingham, ROYAUME-UNI. Mora-Huertas, C. E., Fessi, H., & Elaissari, A. (2010). Polymer-based nanocapsules for drug delivery, International Journal of Pharmaceutics 385(1-2), 113-142. Mosqueira, V. C. F., Legrand, P., Gulik, A., Bourdon, O., Gref, R., Labarre, D., & Barratt, G. (2001). Relationship between complement activation, cellular uptake and surface physicochemical aspects of novel PEG-modified nanocapsules, Biomaterials 22(22), 2967-2979. Mu, L., & Seow, P. H. (2006). Application of TPGS in polymeric nanoparticulate drug delivery system, Colloids and Surfaces B: Biointerfaces 47(1), 90-97. Mun, S., Decker, E. A., & McClements, D. J. (2007). Influence of emulsifier type on in vitro digestibility of lipid droplets by pancreatic lipase, Food Research International 40(6), 770- 781. Naoe, K., Yoshimoto, S., Naito, N., Kawagoe, M., & Imai, M. (2011). Preparation of protein nanoparticles using AOT reverse micelles, Biochemical Engineering Journal 55(2), 140- 143.

Chapter 2. Literature Review

Nishino, T., Yokoyama, G., Dobashi, K., Fujihara, M., & Nagumo, T. (1989). Isolation, purification, and characterization of fucose-containing sulfated polysaccharides from the brown seaweed ecklonia kurome and their blood-anticoagulant activities, Carbohydrate Research 186(1), 119-129. Ochs, C. J., Such, G. K., Yan, Y., van Koeverden, M. P., & Caruso, F. (2010). Biodegradable Click Capsules with Engineered Drug-Loaded Multilayers, ACS Nano 4(3), 1653- 1663. Ogawa, S., Decker, E. A., & McClements, D. J. (2004). Production and Characterization of O/W Emulsions Containing Droplets Stabilized by Lecithin−Chitosan−Pectin Mutilayered Membranes, Journal of Agricultural and Food Chemistry 52(11), 3595-3600. Ostertag, F., Weiss, J., & McClements, D. J. (2012). Low-energy formation of edible nanoemulsions: Factors influencing droplet size produced by emulsion phase inversion, Journal of Colloid and Interface Science 388(1), 95-102. Ourique, A. F., Pohlmann, A. R., Guterres, S. S., & Beck, R. C. R. (2008). Tretinoin-loaded nanocapsules: Preparation, physicochemical characterization, and photostability study, International Journal of Pharmaceutics 352(1–2), 1-4. Park, H. Y., Han, M. H., Park, C., Jin, C.-Y., Kim, G.-Y., Choi, I.-W., Kim, N. D., Nam, T.-J., Kwon, T. K., & Choi, Y. H. (2011). Anti-inflammatory effects of fucoidan through inhibition of NF-κB, MAPK and Akt activation in lipopolysaccharide-induced BV2 microglia cells, Food and Chemical Toxicology 49(8), 1745-1752. Patil, M. N., & Pandit, A. B. (2007). Cavitation – A novel technique for making stable nano-suspensions, Ultrasonics Sonochemistry 14(5), 519-530. Pavinatto, F. J., Caseli, L., & Oliveira, O. N. (2010). Chitosan in Nanostructured Thin Films, Biomacromolecules 11(8), 1897-1908. Peniche, C. (2008) Chitin and Chitosan: Major Sources, Properties and Applications, In Monomers, Polymers and Composites from Renewable Resources (Belgacem, M. N., and Gandini, A., Eds.) 1st ed., Elsevier, Oxford, UK. Percival, E. G. V., & Ross, A. G. (1950). 145. Fucoidin. Part I. The isolation and purification of fucoidin from brown seaweeds, Journal of the Chemical Society (Resumed), 717- 720. Polakovic, M., Görner, T., Gref, R., & Dellacherie, E. (1999). Lidocaine loaded biodegradable nanospheres: II. Modelling of drug release, Journal of Controlled Release 60(2- 3), 169-177. Pomin, V. H., & Mourão, P. A. S. (2008). Structure, biology, evolution, and medical importance of sulfated fucans and galactans, Glycobiology 18(12), 1016-1027.

Pinheiro, A. C. (2013)

Ponce, N. M. A., Pujol, C. A., Damonte, E. B., Flores, M. L., & Stortz, C. A. (2003). Fucoidans from the brown seaweed Adenocystis utricularis: extraction methods, antiviral activity and structural studies, Carbohydrate Research 338(2), 153-165. Preetz, C., Hauser, A., Hause, G., Kramer, A., & Mäder, K. (2010). Application of atomic force microscopy and ultrasonic resonator technology on nanoscale: Distinction of nanoemulsions from nanocapsules, European Journal of Pharmaceutical Sciences 39(1–3), 141- 151. Prombutara, P., Kulwatthanasal, Y., Supaka, N., Sramala, I., & Chareonpornwattana, S. (2012). Production of nisin-loaded solid lipid nanoparticles for sustained antimicrobial activity, Food Control 24(1–2), 184-190. Qian, C., Decker, E. A., Xiao, H., & McClements, D. J. (2012). Physical and chemical stability of β-carotene-enriched nanoemulsions: Influence of pH, ionic strength, temperature, and emulsifier type, Food Chemistry 132(3), 1221-1229. Qian, C., & McClements, D. J. (2011). Formation of nanoemulsions stabilized by model food-grade emulsifiers using high-pressure homogenization: Factors affecting particle size, Food Hydrocolloids 25(5), 1000-1008. Quinn, J. F., & Caruso, F. (2004). Facile tailoring of film morphology and release properties using layer-by-layer assembly of thermoresponsive materials, Langmuir 20(1), 20- 22. Rao, J., & McClements, D. J. (2011). Food-grade microemulsions, nanoemulsions and emulsions: Fabrication from sucrose monopalmitate & lemon oil, Food Hydrocolloids 25(6), 1413-1423. Rao, J., & McClements, D. J. (2012). Food-grade microemulsions and nanoemulsions: Role of oil phase composition on formation and stability, Food Hydrocolloids 29(2), 326-334. Relleve, L., Nagasawa, N., Luan, L. Q., Yagi, T., Aranilla, C., Abad, L., Kume, T., Yoshii, F., & dela Rosa, A. (2005). Degradation of carrageenan by radiation, Polymer Degradation and Stability 87(3), 403-410. Richert, L., Lavalle, P., Payan, E., Shu, X. Z., Prestwich, G. D., Stoltz, J.-F., Schaaf, P., Voegel, J.-C., & Picart, C. (2004). Layer by Layer Buildup of Polysaccharide Films: Physical Chemistry and Cellular Adhesion Aspects, Langmuir 20(2), 448-458. Rinaudo, M. (2008) Polyelectrolytes Derived from Natural Polysaccharides, In Monomers, Polymers and Composites from Renewable Resources (Belgacem, M. N., and Gandini, A., Eds.) 1st ed., Elsevier, Oxford, UK.

Chapter 2. Literature Review

Rousseau, E., Van der Auweraer, M., & De Schryver, F. C. (2000). Steady-state and time-resolved spectroscopy of a self-assembled cyanine dye multilayer, Langmuir 16(23), 8865-8870. Rudra, J. S., Dave, K., & Haynie, D. T. (2006). Antimicrobial polypeptide multilayer nanocoatings, Journal of Biomaterials Science-Polymer Edition 17(11), 1301-1315. Ryzhkova, A., Jarzak, U., Schäfer, A., Bäumer, M., & Swiderek, P. (2011). Modification of surface properties of thin polysaccharide films by low-energy electron exposure, Carbohydrate Polymers 83(2), 608-615. Saberi, A. H., Fang, Y., & McClements, D. J. (2013). Fabrication of vitamin E-enriched nanoemulsions: Factors affecting particle size using spontaneous emulsification, Journal of Colloid and Interface Science 391(0), 95-102. Salvia-Trujillo, L., Qian, C., Martín-Belloso, O., & McClements, D. J. (2013). Influence of particle size on lipid digestion and β-carotene bioaccessibility in emulsions and nanoemulsions, Food Chemistry 141(2), 1472-1480. Sanguansri, P., & Augustin, M. A. (2006). Nanoscale materials development – a food industry perspective, Trends in Food Science & Technology 17(10), 547-556. Sauerbrey, G. (1959). Verwendung von Schwingquarzen zur Wägung dünner Schichten und zur Mikrowägung, Zeitschrift für Physik A Hadrons and Nuclei 155(2), 206-222. Schwarz, J. C., Klang, V., Karall, S., Mahrhauser, D., Resch, G. P., & Valenta, C. (2012). Optimisation of multiple W/O/W nanoemulsions for dermal delivery of aciclovir, International Journal of Pharmaceutics 435(1), 69-75. Sekhon, B. S. (2010 ). Food nanotechnology - an overview Nanotechnology, Science and Applications 3(1), 1-15. Serizawa, T., Yamaguchi, M., & Akashi, M. (2002). Enzymatic hydrolysis of a layer-by- layer assembly prepared from chitosan and dextran sulfate, Macromolecules 35(23), 8656- 8658. Shahidi, F., Arachchi, J. K. V., & Jeon, Y.-J. (1999). Food applications of chitin and chitosans, Trends in Food Science & Technology 10(2), 37-51. Shu, S., Sun, C., Zhang, X., Wu, Z., Wang, Z., & Li, C. (2010). Hollow and degradable polyelectrolyte nanocapsules for protein drug delivery, Acta Biomaterialia 6(1), 210-217. Siddiqui, A. M., Provost, A., & Schwarz, W. H. (1991). Peristaltic pumping of a second- order fluid in a planar channel, Rheologica Acta 30(3), 249-262. Silva, H., Cerqueira, M., & Vicente, A. (2012). Nanoemulsions for Food Applications: Development and Characterization, Food and Bioprocess Technology 5(3), 854-867.

Pinheiro, A. C. (2013)

Silva, H. D., Cerqueira, M. A., Souza, B. W. S., Ribeiro, C., Avides, M. C., Quintas, M. A. C., Coimbra, J. S. R., Carneiro-da-Cunha, M. G., & Vicente, A. A. (2011). Nanoemulsions of β- carotene using a high-energy emulsification–evaporation technique, Journal of Food Engineering 102(2), 130-135. St-Onge, M.-P., & Jones, P. J. H. (2002). Physiological Effects of Medium-Chain Triglycerides: Potential Agents in the Prevention of Obesity, The Journal of Nutrition 132(3), 329-332. Sukhorukov, G. B., Antipov, A. A., Voigt, A., Donath, E., & Möhwald, H. (2001). pH- Controlled Macromolecule Encapsulation in and Release from Polyelectrolyte Multilayer Nanocapsules, Macromolecular Rapid Communications 22(1), 44-46. Sukhorukov, G. B., Donath, E., Lichtenfeld, H., Knippel, E., Knippel, M., Budde, A., & Mˆhwald, H. (1998). Layer-by-layer self assembly of polyelectrolytes on colloidal particles, Colloids and Surfaces A: Physicochemical and Engineering Aspects 137(1-3), 253-266. Sun, B., Mutch, S. A., Lorenz, R. M., & Chiu, D. T. (2005). Layered Polyelectrolyte−Silica Coating for Nanocapsules, Langmuir 21(23), 10763-10769. Sundar, S., Kundu, J., & Kundu, S. C. (2010). Biopolymeric nanoparticles, Science and Technology of Advanced Materials 11(1), 014104. Tadros, T., Izquierdo, P., Esquena, J., & Solans, C. (2004). Formation and stability of nano-emulsions, Advances in Colloid and Interface Science 108–109(0), 303-318. Tokle, T., Lesmes, U., & McClements, D. J. (2010). Impact of Electrostatic Deposition of Anionic Polysaccharides on the Stability of Oil Droplets Coated by Lactoferrin, Journal of Agricultural and Food Chemistry 58(17), 9825-9832. Troncoso, E., Aguilera, J. M., & McClements, D. J. (2012). Fabrication, characterization and lipase digestibility of food-grade nanoemulsions, Food Hydrocolloids 27(2), 355-363. van der Strate, B. W. A., Beljaars, L., Molema, G., Harmsen, M. C., & Meijer, D. K. F. (2001). Antiviral activities of lactoferrin, Antiviral Research 52(3), 225-239. Verwei, M., Freidig, A. P., Havenaar, R., & Groten, J. P. (2006). Predicted Serum Folate Concentrations Based on In Vitro Studies and Kinetic Modeling are Consistent with Measured Folate Concentrations in Humans, The Journal of Nutrition 136(12), 3074-3078. Wang, R., Tian, Z., & Chen, L. (2011). Nano-encapsulations liberated from barley protein microparticles for oral delivery of bioactive compounds, International Journal of Pharmaceutics 406(1–2), 153-162. Wang, X. Y., Hu, X., Daley, A., Rabotyagova, O., Cebe, P., & Kaplan, D. L. (2007a). Nanolayer biomaterial coatings of silk fibroin for controlled release, Journal of Controlled Release 121, 190-199.

Chapter 2. Literature Review

Wang, Y., Angelatos, A. S., & Caruso, F. (2007b). Template Synthesis of Nanostructured Materials via Layer-by-Layer Assembly†, Chemistry of Materials 20(3), 848-858. Wang, Z. L. (2000). Transmission Electron Microscopy of Shape-Controlled Nanocrystals and Their Assemblies, The Journal of Physical Chemistry B 104(6), 1153-1175. Weiss, J., Takhistov, P., & McClements, D. J. (2006). Functional materials in food nanotechnology, Journal of Food Science 71(9), R107-R116. Wickham, M., Faulks, R., & Mills, C. (2009). In vitro digestion methods for assessing the effect of food structure on allergen breakdown, Molecular Nutrition & Food Research 53(8), 952-958. Woda, A., Mishellany-Dutour, A., Batier, L., François, O., Meunier, J. P., Reynaud, B., Alric, M., & Peyron, M. A. (2010). Development and validation of a mastication simulator, Journal of Biomechanics 43(9), 1667-1673. Wohl, B. M., & Engbersen, J. F. J. (2012). Responsive layer-by-layer materials for drug delivery, Journal of Controlled Release 158(1), 2-14. Wood, K. C., Boedicker, J. Q., Lynn, D. M., & Hammond, P. T. (2005). Tunable drug release from hydrolytically degradable layer-by-layer thin films, Langmuir 21(4), 1603-1609. Yang, J. W., Yoon, S. Y., Oh, S. J., Kim, S. K., & Kang, K. W. (2006). Bifunctional effects of fucoidan on the expression of inducible nitric oxide synthase, Biochemical and Biophysical Research Communications 346(1), 345-350. Ye, S., Wang, C., Liu, X., & Tong, Z. (2005). Multilayer nanocapsules of polysaccharide chitosan and alginate through layer-by-layer assembly directly on PS nanoparticles for release, Journal of Biomaterials Science, Polymer Edition 16, 909-923. Yoo, D., Shiratori, S. S., & Rubner, M. F. (1998). Controlling bilayer composition and surface wettability of sequentially adsorbed multilayers of weak polyelectrolytes, Macromolecules 31(13), 4309-4318. Yuan, Y., Gao, Y., Zhao, J., & Mao, L. (2008). Characterization and stability evaluation of β-carotene nanoemulsions prepared by high pressure homogenization under various emulsifying conditions, Food Research International 41(1), 61-68. Zhang, S., Marini, D. M., Hwang, W., & Santoso, S. (2002). Design of nanostructured biological materials through self-assembly of peptides and proteins, Current Opinion in Chemical Biology 6(6), 865-871. Zhao, X., Xue, C.-H., & Li, B.-F. (2008). Study of antioxidant activities of sulfated polysaccharides from Laminaria japonica, Journal of Applied Phycology 20(4), 431-436.

Pinheiro, A. C. (2013)

Zhong, Y., Whittington, C. F., Zhang, L., & Haynie, D. T. (2007). Controlled loading and release of a model drug from polypeptide multilayer nanofilms, Nanomedicine: Nanotechnology, Biology and Medicine 3(2), 154-160. Ziani, K., Chang, Y., McLandsborough, L., & McClements, D. J. (2011). Influence of Surfactant Charge on Antimicrobial Efficacy of Surfactant-Stabilized Thyme Oil Nanoemulsions, Journal of Agricultural and Food Chemistry 59(11), 6247-6255. Zvyagintseva, T. N., Shevchenko, N. M., Chizhov, A. O., Krupnova, T. N., Sundukova, E. V., & Isakov, V. V. (2003). Water-soluble polysaccharides of some far-eastern brown seaweeds. Distribution, structure, and their dependence on the developmental conditions, Journal of Experimental Marine Biology and Ecology 294(1), 1-13.

SECTION II NANOFILMS

3 CHAPTER 3. INTERACTIONS BETWEEN

κ - CARRAGEENAN AND CHITOSAN IN NANOLAYERED COATINGS – STRUCTURAL AND TRAN SPORT PROPERTIES EVALUATION INTERACTIONS BETWEEN κ-CARRAGEENAN AND CHITOSAN IN NANOLAYERED COATINGS – STRUCTURAL AND TRANSPORT PROPERTIES EVALUATION

3.1 Introduction...... 71

3.2 Materials and Methods ...... 72

3.3 Results and discussion ...... 79

3.4 Conclusions...... 91

3.5 References ...... 93

Pinheiro, A. C. (2013)

The results presented in this Chapter were adapted from:

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

3.1 INTRODUCTION

Nanotechnology holds a great potential to generate innovative solutions and to provide food technologists with instruments to meet consumers’ demands in very diverse aspects related with foods, such as safety, quality, health-promotion and novelty. Layer-by-Layer (LbL) deposition technique is one of the most powerful methods to create nanolayered coatings with tailored properties (Decher and Schlenoff, 2003). Typically, this technique is based on the alternating deposition of oppositely charged polyelectrolytes and can be applied to produce multilayers of nanometre thickness (1 to 100 nm per layer).(Weiss et al., 2006). The versatility of the LbL has allowed a deposition of a broad range of materials (e.g. polymers, lipids, proteins, nanoparticles, dye molecules) on various templates (e.g. planar and colloidal) on basis of not only electrostatic interactions but also hydrogen bonding, hydrophobic interactions, covalent bonding and complementary base pairing (Wang et al., 2007). The functionality of the final films/coatings, such as their permeability to gases, mechanical properties, swelling and wetting characteristics and their environmental sensitivity to pH and temperature, will be determined by the type of adsorbing substances used to create each layer, the total number of layers, the sequence of the different layers and by the conditions (e.g. pH and ionic strength) used to prepare each layer (Weiss et al., 2006). Moreover, the nanolayered coatings can be specially engineered to incorporate bioactive compounds and act as controlled release systems and can be used to coat food products such as fruits and vegetables.

There has been an increasing interest in nanolayered systems due to their potential applications in a wide range of areas including pharmaceutical, biomedical and food packaging. Several works concerning the production of nanolayered coatings with synthetic polymers (Yoo et al., 1998) or for biomedical applications (Fu et al., 2005; Martins et al., 2010a) have been done, while few works have used biodegradable polymers from renewable sources and have been directed to the food industry. Examples include the construction of nanolayered coatings composed of poly(L-glutamic acid) and lysozyme (Rudra et al., 2006) and of κ- carrageenan and lysozyme (Medeiros et al., 2011), as strategies for food preservation. Therefore, there is a need to create and characterize (e.g. in terms of gas permeabilities) other multilayer systems that combine both biofunctionality and biocompatibility. Biopolymers such as chitosan and κ-carrageenan are competitive candidates for the formation of nanolayered coatings, due to their opposite electrostatic properties, together with their bioactive and non- toxic properties. Chitosan is a natural polysaccharide obtained by deacetylation of chitin, which is the major constituent of the exoskeleton of crustaceous animals, and has been

Pinheiro, A. C. (2013)

extensively used to produce biodegradable films (Begin and Van Calsteren, 1999; Pranoto et al., 2005; Shi et al., 2009). Chitosan is an excellent film component due to its good oxygen barrier properties and due to its intrinsic antimicrobial activity (Begin and Van Calsteren, 1999). Carrageenan is a fraction of the sulphated polysaccharides family extracted from certain red seaweeds. It is extensively used in the food industry as gelling, emulsifying, and stabilizing agent and has been reported as having excellent film forming properties (Park et al., 2001). Polyethylene terephthalate (PET) is a good candidate to be used as a support for the deposition of the polyelectrolytes using the LbL technique due to its excellent physicochemical properties such as good mechanical properties, thermal stability and optical transparency (Fu et al., 2005; Indest et al., 2009).

Understanding the electrostatic interactions of the polyelectrolytes within a nanolayered system can be of utmost importance when considering the development of increasingly specific systems. The driving force for the polyelectrolytes interaction is the enthalpic contribution related to the interaction of oppositely charged groups and the increase in entropy due to the release of their counterions (Vinayahan et al., 2010).

The present work aims at evaluating the interactions between κ-carrageenan and chitosan in a nanolayered coating produced through LbL assembly on a PET support and to characterize the nanolayered coating in terms of their permeabilities and surface properties. A first approach to understanding the behaviour of κ-carrageenan/chitosan nanolayered coating in the gastrointestinal system (using a static digestion model) is also presented.

3.2 MATERIALS AND METHODS

3.2.1 Preparation of the polyelectrolyte solutions

The κ-carrageenan solution was prepared dissolving 0.2% (w/v) of κ-carrageenan (MW = 3.8 x 105 Da, 76% purity, GENUGEL carrageenan type WR-78, CPKelco, Denmark) in distilled water under magnetic stirring at approximately 200 rpm during 2 h at room temperature (20 °C). The pH of the solution was adjusted to 7.0 with a solution of 1 mol.L-1 sodium hydroxide (Riedel-de Haёn, Germany).

The chitosan solution was prepared dissolving 0.2% (w/v) of chitosan (90% deacetylation, MW = 3.4 x 105 Da, 99% purity; Aqua Premier Co. Ltd, Thailand) in a 1.0% (v/v) lactic acid (Acros Organics, Belgium) solution under agitation (using a magnetic stirrer) at approximately 200 rpm during 2 h at room temperature. The pH of the solution was adjusted to 3.0 with a solution of 1 mol.L-1 sodium hydroxide (Riedel-de Haёn, Germany).

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

3.2.2 PET aminolysis

PET sheets were obtained from Canson (Annonay Cedex, France) and were cut into rectangular pieces of 0.8 x 5.0 cm for the UV-Vis, contact angle and SEM analysis and into circular pieces of 5.0 cm of diameter for the gas permeability measurements and in vitro digestion, and were aminolyzed according to previous studies (Fu et al., 2005). Briefly, PET pieces were cleaned through immersion in ethanol (Panreac, Spain)/water (1:1, v/v) solution for 3 h, followed by rinsing with distilled water and drying at 30 °C and at a RH of 42.0 ± 1.5% in an oven for 24 h. Then, the pieces were immersed in 0.06 g.mL-1 1,6-hexanediamine/propanol (Aldrich-Germany and Sigma–Aldrich-USA, respectively) solution at 37 °C for 4 h, thoroughly washed with distilled water to remove free 1,6-hexanediamine, and finally dried in an oven at 37 °C and at a RH of 42.0 ± 1.5% for 24 h. These aminolyzed PET pieces were treated with 0.1 mol.L-1 hydrochloric acid (Merck, Germany) solution for 3 h at room temperature (20 °C) and then washed with a large amount of distilled water, and dried in an oven at 30 °C and at a RH of 42.0 ± 1.5% for 24 h. The obtained polymer was termed aminolyzed/charged PET (A/C PET).

3.2.3 Evaluation of the interactions between the polyelectrolytes

3.2.3.1 Electrostatic properties

The ζ-potential of the polyelectrolyte solutions was calculated by determining the electrophoretic mobility and then applying Henry’s equation. Electrophoretic mobility is obtained by performing an electrophoresis experiment on the sample and measuring the velocity of particles using Laser Doppler Velocimetry (Zetasizer Nano series, Malvern Instruments, UK). The measurements (three readings for each assay) were performed in a folded capillary cell. The assays were performed in triplicate, so the results are given as the average ± standard deviation of the nine values obtained.

3.2.3.2 Microcalorimetry analyses

For the microcalorimetry analyses, A/C PET pieces were reduced to powder with a mill (Retsch, Germany), at room temperature (20 °C). Calorimetric measurements were carried out using a CSC (Calorimeter Science Corp., USA) microcalorimeter, model 4200 controlled by ItcRun software with a 1.75 mL reaction cell (sample and reference) and sensitivity of 0.02 µW. Determinations of the enthalpy of interactions between chitosan and κ-carrageenan were carried out at 20 °C, in triplicate and the whole calorimetric procedure was chemically and electrically calibrated as recommended (da Silva et al., 2008). During each experiment, aliquots of 10 µL of an aqueous solution of chitosan (0.2% w/v, pH 3.0) were injected in a sample cell

Pinheiro, A. C. (2013)

containing i) 1.75 mL of ultra-pure water (dilution enthalpy determination) or ii) 1.75 mL of a κ- carrageenan aqueous solution (0.2% w/v, pH 7.0) or iii) 1.75 mL of a suspension containing 40 mg of A/C PET microspheres with κ-carrageenan adsorbed (adsorbed A/C PET-κ-carrageenan- chitosan apparent interaction enthalpy determination).

For each injection, the experimental heat (enthalpy) change Hi (in kJ) resulting from injection i was obtained by the raw data peak integration. The values of integrated molar enthalpy

-1 change for injection i, ΔHi (in kJmol ), were obtained by dividing Hi by the number of moles of chitosan added, ni; hence ΔHi = Hi/ni. Then, plots of ΔHi against the total chitosan concentration added were obtained and are referred as enthalpy curves.

3.2.3.3 Quartz crystal microbalance analyses

The adsorption behaviour of κ-carrageenan and chitosan was evaluated using a quartz crystal microbalance (QCM 200, purchased from Stanford Research Systems, SRS, USA), equipped with AT-cut quartz crystals (5 MHz) with optically flat polished chrome/gold electrodes on contact and liquid sides. Before carrying out the experiments, the crystal was cleaned by successive sonications (40 KHz, 30 min) in ultra-pure water, ethanol (Panreac, Spain) and ultra- pure water, followed by drying with a gentle flow of nitrogen. QCM measurements were started with distilled water at pH 7.0 as a baseline. Adsorption measurements were performed by alternate immersion of the crystal in κ-carrageenan (pH 7.0) and chitosan (pH 3.0) solutions, for 15 min. A washing step of 15 min with distilled water (at a pH 7.0 or pH 3.0, depending if the previous adsorption was with κ-carrageenan or chitosan solutions, respectively) was carried out after the deposition of each polyelectrolyte, in order to remove the unbound polyelectrolytes and to prevent the cross-contamination of solutions. The variations of the resonance frequency (ΔF) and of the motional resistance (ΔR) were simultaneously measured as a function of time and the analyses were performed at 20 °C, in triplicate.

3.2.4 Preparation of nanolayered coating

The nanolayered coating was composed of an A/C PET support layer adsorbed with a polysaccharide multilayer constituted of 5 polysaccharide layers (three κ-carrageenan and two chitosan layers).

In order to prepare the nanolayered coating, A/C PET pieces were firstly dipped into the κ- carrageenan solution for 15 min and subsequently rinsed with distilled water at pH 7.0. The samples were dried under a flow of nitrogen. The procedure was repeated, this time using chitosan as the polyelectrolyte and rinsed with distilled water at pH 3.0. The

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation dipping/washing/drying process was repeated until the deposition of a total of 5 layers (κ- carrageenan/chitosan/κ-carrageenan/chitosan/κ-carrageenan).

As a result of the immersion into polyelectrolyte solutions, both sides of the A/C PET pieces were coated with the same set of layers. Therefore, for the determination of water vapour and oxygen permeabilities, it was necessary to multiply the thickness value of each of the five layers by 2.

The nanolayered coatings were then maintained at 20 °C and 50% relative humidity (RH) before analyses.

3.2.5 Characterization of the nanolayered coating

3.2.5.1 UV/VIS spectroscopy

In order to follow the multilayer construction, a UV-VIS spectrophotometer (Jasco 560, Germany) was used to measure the absorbance of the prepared nanolayered coatings at a wavelength of 260 nm. This wavelength was chosen according to other authors (Carneiro-da- Cunha et al., 2010; dos Santos et al., 2003; Wasikiewicz et al., 2005). Three replicates of each measurement were obtained.

3.2.5.2 Contact angle measurements

The contact angle of the nanolayered coating surface was measured by the sessile drop method, using a face contact angle meter (OCA 20, Dataphysiscs, Germany). A 2 µL droplet of ultra-pure water was placed on the horizontal surface of each film with a 500 µL syringe (Hamilton, Switzerland), with a needle of 0.75 mm of diameter, and observed under the contact angle meter. The contact angle measurements were performed at three different surfaces and at each surface, ten replicates of contact angles were obtained at contact times of 0, 15, and 30 s.

3.2.5.3 Scanning electron microscopy (SEM)

The surface morphology of the nanolayered coating was examined using scanning electron microscopy (Nova NanoSEM 200, Netherlands) with an accelerating voltage from 10 to 15 kV. Before analyses, all samples were mounted on aluminium stubs using carbon adhesive tape and sputter-coated with gold (thickness of about 10 nm). The SEM images allowed the direct measurement of the layers thicknesses (average of three measurements) using the equipment’s image analysis software (XT microscope control software).

Pinheiro, A. C. (2013)

3.2.5.4 Water vapour permeability (WVP) measurements

The methodology used was based on the ASTM E96-92 (ASTM E96-92, 1990) method. The films were sealed on the top of a permeation cell containing 55 mL of distilled water (100% RH; 2337 Pa vapour pressure at 20 °C). The cells were then placed in a desiccator at 20 °C and 0% RH (0 Pa water vapour pressure) containing previously dried silica (105 °C overnight) and were weighed during 10 h at intervals of 2 h, assuming steady-state and uniform water pressure conditions (Ziani et al., 2008). In order to ensure uniform relative humidity, fan speeds were installed inside the desiccators. Three replicates were obtained for each sample.

The WVP of the A/C PET and of the A/C PET + 5 nanolayers was estimated using regression analysis from Equation 3.1 adapted from literature (Sobral et al., 2001):

Equation 3.1

where L is the average thickness of the films, A is the permeation area (0.005524 m2). ΔP is the difference of partial vapour pressure of the atmosphere (2337 Pa at 20 °C) and w is the weight loss.

The WVP of the polysaccharide nanolayers was determined by the following equation (Cooksey et al., 1999):

Equation 3.2 ( ) ( )

where A, B and T correspond to the A/C PET, to the five polysaccharide nanolayers and to the A/C PET with polysaccharide nanolayered coating, respectively. L corresponds to the thickness of the films. The thickness of A/C PET was measured by a digital micrometer (Mitutoyo, Japan) (100 µm) and the thickness of the polysaccharide nanolayered coating was determined by the SEM analyses.

3.2.5.5 Oxygen permeability

Oxygen permeability (O2P) was determined based on the ASTM D3985-02 (ASTM D3985-02, 2002) method. The films were sealed between two chambers, having each one two channels.

In the lower chamber O2 was supplied at a constant flow rate controlled by a gas flow meter (J & W Scientific, ADM 2000, USA) to maintain its pressure constant in that compartment. The other chamber was purged by a stream of nitrogen, also at controlled flow. Nitrogen acted as a carrier for the O2. The flow leaving this chamber was connected to an O2 sensor (Mettler

Toledo, Switzerland) that measured O2 concentration in that flow on-line.

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

The flows of the two chambers were connected to a manometer to ensure the equality of pressures (both at 1 atm) between both compartments. As the O2 was carried continuously by the nitrogen flow, it was considered that partial pressure of O2 in the upper compartment is null, therefore ΔP is equal to 1 atm. Three replicates were performed for each sample.

The polysaccharide nanolayers O2P were determined by the following equation (Cooksey et al., 1999):

Equation 3.3 ( ) ( )

where A, B and T correspond to the A/C PET, to the five polysaccharide nanolayers and to the A/C PET with polysaccharide nanolayered coating, respectively. L corresponds to the thickness of the films. The thickness of A/C PET was measured by a digital micrometre (Mitutoyo, Japan) (100 µm) and the thickness of the nanolayered coating was determined by SEM.

3.2.6 Evaluation of the behaviour of the κ-carrageenan/chitosan nanolayered coating during in vitro digestion

In vitro digestion of the κ-carrageenan/chitosan nanolayered coating was performed in a static system (glass jacket reactor with magnetic stirrer), simulating the gastric, duodenal, jejunal and ileal portions of the GI tract as described by other authors (Reis et al., 2008) with the following modifications. Approximately 6 g of the κ-carrageenan/chitosan nanolayered coating (assembled on the PET support) were introduced into the reactor (Figure 3.1) and the experiment was run for a total of 5 h, simulating average physiological conditions of GI tract by the continuous additions of gastric, duodenal, jejunal and ileal secretions.

Figure 3.1 Schematic representation of the experimental installation used for in vitro digestion experiments: (A) Glass jacket reactor, (B) pH electrode, (C) Water bath heated at 37 °C, (D) Syringe pump, (E) Stir plate

The gastric secretion consisted of pepsin (600 U.mL-1) and lipase (40 U.mL-1) (Sigma-Aldrich, St.

-1 -1 -1 Louis, MO) in a gastric electrolyte solution (NaCl 4.8 g.L , KCl 2.2 g.L , CaCl2 0.22 g.L and -1 -1 NaHCO3 1.5 g.L ), secreted at a flow rate of 0.5 mL.min . The pH was controlled to follow a

Pinheiro, A. C. (2013)

predetermined curve (from 4.8 at t=0 to 1.7 at t=120 min) by secreting hydrochloric acid (1 mol.L-1). The duodenal secretion consisted of 4% (w/v) porcine bile extract (Sigma-Aldrich, St. Louis, MO) at a flow rate of 0.5 mL.min-1, a 7% (w/v) pancreatin (from porcine pancreas 8xUSP, Sigma-Aldrich, St. Louis, MO) solution at a flow rate of 0.25 mL.min-1 and a small intestinal

-1 -1 -1 electrolyte solution (SIES) (NaCl 5 g.L , KCl 0.6 g.L , CaCl2 0.25 g.L ) at a flow rate of 0.15 mL.min-1. The jejunal secretion fluid consisted of SIES containing 10% (v/v) porcine bile extract solution at a flow rate of 3.2 mL.min-1. The ileal secretion fluid consisted of SIES at a flow rate of 3.0 mL.min-1. The pH in the different parts of small intestine was controlled by the addition of 1 mol.L-1 sodium bicarbonate solutions to set-points of 6.5, 6.8 and 7.2 for simulated duodenum, jejunum and ileum, respectively. At pre-set times, samples of PET were taken from the reactor and analysed for its monosaccharide content. This analysis was performed indirectly, once when the samples were collected from the liquid medium, the monosaccharide analysis revealed a considerable interference from the enzymes (that contained significant amounts of carbohydrates). The monosaccharide composition of κ-carrageenan was determined by the reductive hydrolysis described by other authors (Stevenson and Furneaux, 1991), in which the hydrolysis of the κ-carrageenan glycosidic bonds is performed simultaneously with the reduction reaction using 4-methylmorpholine borane (MMB). The hydrolysis was conducted in two steps, the first with trifluoroacetic acid (TFA) 2.4 mol.L-1 at 80 °C during 5 min and the second with TFA 2 mol.L-1 at 120 °C during 1 h. After this hydrolysis, the PET was discarded. The alditol acetates were obtained by acetylation with 1-methylmidazol and acetic anhydride and analysed by GC- FID as described by Coimbra et al. (1996). For chitosan determination, the sample was hydrolysed with HCl 6 mol.L-1 at 100 °C during 20 h, followed by the glucosamine conversion to alditol acetate derivatives by reduction with sodium borohydrate and acetylation, as previously described (Coimbra et al., 1996). The alditol acetate derivatives were analysed by GC-FID.

3.2.6.1 Statistical Analyses

The statistical analyses were carried out using analysis of variance, Tukey mean comparison test (p<0.05) and linear regression analysis (SigmaStat, trial version, 2003, USA).

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

3.3 RESULTS AND DISCUSSION

3.3.1 Evaluation of the interactions between κ-carrageenan and chitosan

3.3.1.1 Electrostatic properties of the polyelectrolytes

Since polyelectrolyte deposition is primarily governed by electrostatic interactions and in order to guarantee a correct interaction between the surface of the A/C PET and the κ-carrageenan nanolayer (the first one on the A/C PET surface) and between the subsequent κ- carrageenan/chitosan layers, it is necessary to assure that these polyelectrolytes exhibit opposite electrostatic properties. Therefore, to confirm the opposite charges of κ-carrageenan and chitosan solutions, their ζ-potential values were determined. The obtained values were - 56.90 ± 5.11 mV for κ-carrageenan solution at pH 7.0 and + 45.83 ± 3.35 mV for chitosan solution at pH 3.0, which means that these polyelectrolytes can interact by electrostatic forces. The ζ-potential values obtained for κ-carrageenan and chitosan solutions are in agreement with the ones obtained by Medeiros et al. (2011) (60.53 ± 0.15 mV) and by Carneiro-da-Cunha et al. (2010) (58.28 ± 4.18 mV), respectively.

Other pH values were tested for the κ-carrageenan and chitosan solutions (data not shown) and the pH values corresponding to the highest ζ-potential values were the ones chosen to perform the experiments in order to enhance the electrostatic interactions between these polyelectrolytes.

3.3.1.2 Microcalorimetry analyses

Microcalorimetry is a thermodynamic technique that allows the study of the interactions between two species, by directly measuring the heat released or absorbed during a biomolecular binding event. Although the prevalent kind of interactions in our system is possibly electrostatic interactions, one of the objectives of this work is precisely to evaluate/confirm (through microcalorimetry) the specific energy of those interactions. Therefore, the measurement of the energy evolved or absorbed upon interaction between chitosan and κ-carrageenan compared to that obtained upon dilution of a concentrated chitosan solution can give thermodynamic information about the nature of the interactions between both polyelectrolytes (electrostatic, hydrogen bonds, van der Waals, among others). Whenever we refer to “interactions” in the subsequent text, we mean by default all types of interactions that may be occurring; the conclusion on the prevalent type of interaction in our system is presented at the end of this section. The curves of ΔH (resulting from the integration of the raw microcalorimetric data) (Denadai et al., 2006) versus total chitosan concentration

Pinheiro, A. C. (2013)

added to water in the absence (dilution process, ΔdilH) and in the presence of a 0.2% (w/v) κ- carrageenan solution (dilution and interaction process, ΔobsH) are shown in Figure 3.2 a) and b).

Figure 3.2 a) Dilution enthalpy change (ΔdilH) resulting from the titration of chitosan solution on water alone as function of chitosan solution concentration. b) Observed enthalpy change (ΔobsH) resulting from the titration of chitosan solution on the κ-carrageenan solution as function of chitosan solution concentration.

The titration curve for chitosan added to water (Figure 3.2a)) shows that in the concentration interval between 0 and 4.0 µmol.L-1 of chitosan, the dilution process of this polyelectrolyte is

-1 exothermic, with very low values of enthalpy of dilution in the range of -0.59 kJ.mol < ΔdilH < - 0.013.kJ.mol-1. However, above 4.0 µmol.L-1, the dilution process becomes endothermic with the enthalpy of dilution having a value of 0.50 kJ.mol-1 for 6 µmol.L-1. The enthalpy changes obtained in these concentration intervals include mainly contributions from the decrease of chitosan-chitosan interaction (due to increase of distance between chitosan molecules during the dilution process - this increase of distance occurs when the concentrated chitosan solution is added to the water) and from the enthalpy changes due to modification of chitosan conformation/solvation. As the resulting enthalpy change is negative (exothermic) this means that the endothermic processes absorb less energy than the energy released by the exothermic processes. However, once ΔdilH magnitudes are so small it is possible to conclude that both processes (endo and exothermic) have very similar absolute values of enthalpy change. The slope of the dilution curve is a measurement of the energy involved in the chitosan-chitosan interaction (i.e., the interaction between chitosan molecules). The system’s enthalpy increases during the dilution process possibly due to changes in the repulsion of charges present on the double electric layers of the polyelectrolyte molecules.

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

The enthalpic curves for addition of chitosan solution into the κ-carrageenan aqueous solution are shown in Figure 3.2 b). The initial additions of chitosan solution result in exothermic enthalpy change values, caused by the intermolecular interaction between chitosan and κ- carrageenan macromolecules. The enthalpy change at 0.134 µmol.L-1 of chitosan is around - 19.8 kJ.mol-1, showing an enthalpically favourable interaction. From Figure 3.2 b) it can also be observed that all subsequent additions of chitosan to the κ-carrageenan solution lead to an exothermic effect, showing that in this concentration range chitosan macromolecules bind to κ-carrageenan and this interaction continues to be enthalpically favourable. However, when more chitosan is added, the interaction between the chitosan and κ-carrageenan becomes less exothermic. The slope of the ΔobsH versus chitosan concentration plot is possibly caused by changes in the chitosan/κ-carrageenan interaction (reflected in changes in enthalpy), since for increasing chitosan concentrations, the free chitosan concentration decreases somewhat, and also by changes in aggregate-aggregate interactions with the increase of chitosan/κ- carrageenan aggregates’ concentration. At a total chitosan concentration of 5.6 µmol.L-1, the κ- carrageenan chain became saturated with chitosan macromolecules, the influence of the κ- carrageenan on the dilution of chitosan solution in the calorimeter cell ceased, and ΔobsH and

ΔdilH curves became equal. This critical concentration, defined as the concentration where the chitosan titration curve in κ-carrageenan solution joins the chitosan dilution curve in water corresponds to a concentration ratio (r) between chitosan and κ-carrageenan of 1.2 in the aggregates. The apparent enthalpy of interaction, Δap-intH, is equal to the enthalpy difference between ΔobsH and ΔdilH curves.

Equation 3.4

For the application of (Equation 3.4), it must be kept in mind that the concentration of chitosan that is free in solution during the microcalorimetric titration experiment is different in the absence and in the presence of κ-carrageenan for the same total amount of chitosan added. As pointed out above, this difference occurs because a fraction of the added chitosan binds to κ-carrageenan. Therefore, the energy change caused by chitosan dilution will be just little different in the microcalorimetric experiments performed in the presence and absence of κ-carrageenan. As the amount of chitosan molecules bound per κ-carrageenan molecule is not known, it is not possible to calculate the exact molar enthalpy change of interaction, only an apparent molar enthalpy change, Δap-intH (Barbosa et al., 2010). However, the features of the

Δap-intH curve are able to provide qualitative information about the progress of κ- carrageenan/chitosan interaction for increasing chitosan concentration. Due to the small contribution of chitosan dilution process enthalpy, the Δap-intH curve between chitosan and κ-

Pinheiro, A. C. (2013)

carrageenan is very similar to that obtained by addition of chitosan in the presence of κ- carrageenan. In Figure 3.3 the Δap-intH versus chitosan concentration values are shown for chitosan in the presence of 0.2% (w/v) κ-carrageenan aqueous solution.

-5 -1

-10

H / / H mol kJ

ap-intl  -15

-20 0 1 2 3 4 5 6 chitosan concentration / mol L-1

Figure 3.3 κ-carrageenan/chitosan apparent interaction enthalpy (Δap-intH) as a function of chitosan solution concentration.

The measured Δap-intH represents a superposition of the heat exchanges associated with different effects. In the initial additions of chitosan to the κ-carrageenan solution, chitosan dilution results in separation of chitosan macromolecules from each other, followed by chitosan interaction with κ-carrageenan. In the concentration interval studied, the Δap-intH -1 -1 values are in the range of -19.2 kJ.mol < Δap-intH < 0.0 kJ.mol . The magnitude of this enthalpy change can be caused by electrostatic interactions as well as dipole-dipole, hydrogen bonds and conformation changes of both polyelectrolytes.

In order to evaluate the kind of κ-carrageenan/chitosan interactions on the A/C PET interface, titrations of chitosan solutions on a dispersion of κ-carrageenan adsorbed on the A/C PET powder particles were performed. The apparent enthalpy of interaction on A/C PET surface,

Δap-intHA/C PET, was obtained after the subtraction of the enthalpy change observed during titration of chitosan on κ-carrageenan/PET from that of the dilution of the chitosan solution. Figure 3.4 shows the change in the apparent enthalpy of interaction between adsorbed κ- carrageenan/A/C PET and chitosan (Δap-intHA/C PET) as a function of chitosan concentration.

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

0.0

-0.2 -1

-0.4

H / / H mol kJ ap-intl

 -0.6

-0.8

0.0 0.7 1.4 2.1 2.8 3.5 chitosan concentration / mol L-1

Figure 3.4 Adsorbed A/C PET/κ-carrageenan/chitosan apparent interaction enthalpy change (Δap-intHA/C PET) as a function of chitosan solution concentration.

-1 -1 For chitosan concentrations between 0.13 µmol.L and 3.4 µmol.L , the values of Δap-intHA/C PET were in the range -0.71 kJ.mol-1 and -0.01 kJ.mol-1. Figure 3.4 shows that, for lower chitosan

-1 concentrations (<1.5 µmol.L ), Δap-intHA/C PET becomes slightly less negative, reaches a maximum value and then, in the vicinity of 1.5 µmol.L-1, abruptly decreases, remaining the adsorption process exothermic. This initial increase may arise from electrostatic interactions between chitosan and κ-carrageenan adsorbed as first layer on the A/C PET surface, while the abrupt decrease could be caused by changes in both polyelectrolytes conformation associated with film formation process (such as pH changes). After the minimum value of Δap-intHA/C PET (-0.73 kJ.mol-1) at 2.0 µmol.L-1 is attained, the addition of more chitosan promotes an abrupt increase in the adsorption energy of chitosan. This increase in the enthalpy of interaction could be attributed to the chitosan-chitosan interaction that occurs at surrounding sites on the κ- carrageenan/A/C PET surface, caused by the decrease in the distance between adsorbed chitosan molecules. These features suggest that the occurrence of the chitosan-chitosan interaction at that surface is the responsible by the formation of a dense film.

Moreover, when comparing the magnitude of Δap-intH and Δap-intHA/C PET as a function of chitosan solution concentration (Figure 3.3 and Figure 3.4, respectively) it can be observed that the presence of the A/C PET causes a large decrease of the magnitude of the apparent enthalpy of interaction between κ-carrageenan and chitosan. This is possibly due to the electrostatic repulsion existing between A/C PET and chitosan molecules, once both exhibit positive charges. The influence of the substrate is eliminated only after more than 6 to 8 layers are deposited (Yoo et al., 1998).

Pinheiro, A. C. (2013)

3.3.1.3 Quartz crystal microbalance analysis

The real-time build-up of the κ-carrageenan and chitosan nanolayered assemblies is reported in Figure 3.5 a) and b).

Figure 3.5 Change in frequency (a) and in motional resistance (b) as a function of time for κ-carrageenan (0.2% w/v, pH 7) and chitosan (0.2% w/v, pH 3) adsorption and for rinsing with distilled water (pH 7 or pH 3); 1 - distilled water at pH 7, 2 - deposition of κ-carrageenan; 3- deposition of chitosan, 4 - distilled water at pH 3.

The decrease of the frequency after each polyelectrolyte deposition, observed in Figure 3.5 a), indicates that mass is being deposited at the crystal surface. Δf changed from 0 Hz to -20 Hz, from -67 to -156 Hz and from -186 to -293 Hz, after the adsorption of the first, second and third κ-carrageenan layer, respectively and changed from -19 to -70 Hz and from -156 to -192 Hz, after the deposition of the first and second chitosan layer, respectively. The adsorbed κ- carrageenan acts as a sublayer and offers binding sites for the subsequent chitosan adsorption and contrariwise, therefore the Δf caused by the adsorption of the first κ-carrageenan layer is significantly lower than the Δf originated by the deposition of the following layers. Moreover, the roughness and the relative hydrophilicity of the substrate (crystal surface) affect the measurement of frequency and this effect is more pronounced in the first layer.

During the washing steps, the frequency did not decrease but remained stable, indicating that no desorption of the polyelectrolytes took place. From Figure 3.5 a) it can also be concluded that the adsorption equilibrium is attained and stable films are obtained.

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

The decrease of frequency due to increasing film mass is accompanied by a resistance increase (Figure 3.5 b). The motional resistance represents the energy loss and the damping process within the system and is closely related to the physical properties of the adsorbed layers and adjacent liquid and its variation can be used to measure the viscoelastic change of the deposited film (Fakhrullin et al., 2007). From Figure 3.5 b) it can be seen that the nanolayered κ-carrageenan/chitosan film dissipates energy, exhibiting viscoelastic behaviour (Marx, 2003).

During deposition of mass on the QCM surface, an alteration in the hydrophilicity of the surface can result in higher frequency changes. From Figure 3.5 a) and Figure 3.5 b) it can be observed that the adsorption of κ-carrageenan results in higher changes in frequency and resistance compared with the adsorption of chitosan, which is possibly due to the more pronounced hydrophilic character of the κ-carrageenan. Hydrophilic surfaces (such as κ- carrageenan layer) are rough and can entrap solvents, resulting in a more viscoelastic multilayered system and contributing to the mass increase sensed by the QCM and therefore to higher frequency changes. On the other hand, hydrophobic surfaces (such as chitosan layer) often do not wet and air can be entrapped, which results in smaller measured mass and energy losses. This is just one of the possibilities presented in the literature to explain the observed behaviour; we advanced it as an hypothesis once a similar behaviour has been previously observed by other authors (Martins et al., 2010a) for alginate/chitosan multilayered systems assembled by LbL technique.

It has to be taken into consideration that the mass corresponding to the frequency shift is the total adsorbed mass, which also includes the water entrapped in the adsorbed layer (Indest et al., 2008) and therefore a mass calculation can only be obtained for rigid thin films and was not performed in this work.

The QCM measurements confirm that the alternating deposition of κ-carrageenan and chitosan, at pH values in which polyelectrolytes carried opposite charge, results in the formation of a stable multilayer structure.

3.3.2 Characterization of the nanolayered coating

3.3.2.1 UV/VIS spectroscopy

The multilayer coating growth on A/C PET surface was followed by UV/VIS spectroscopy. Figure 3.6 shows the UV absorbance of κ-carrageenan/chitosan nanolayered coating plotted against the assembly step. The successful layer-by-layer deposition was confirmed by the increase in absorbance at 260 nm with an increasing assembly step. The multilayer growth has been

Pinheiro, A. C. (2013)

characterized by a variety of growth regimes (e.g. linear and exponential) (Fu et al., 2005; Lavalle et al., 2002). Figure 3.6 shows that, for the deposition of five layers, the amount of polyelectrolytes adsorbed on A/C PET surface is a linear function of the number of nanolayers.

Figure 3.6 Multilayer assembly monitored by UV/VIS spectroscopy at 260 nm. Each data point is the average of three determinations and the error bars show the standard deviation.

3.3.2.2 Contact angle measurements

The surface wettability of the nanolayered coating was investigated. The effect of the contact time between the drop and the surface of the coating was evaluated (Figure 3.7) performing measurements at different contact times (after drop application).

Figure 3.7 Contact angle measurements for the original PET, the A/C PET and for the nanolayered coatings. The measurements were performed at 0, 15 and 30 s. Each data point is the average of ten measurements and the error bars show the standard deviation.

From Figure 3.7 it can be concluded that the contact time does not influence the global trends. Also, the contact angles at 0 s for nanolayered coatings (from 1 to 5 layers) are notoriously different when compared to the values obtained at 15 and 30 s. On the other hand, the results

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation obtained at 15 and 30 s are very similar, showing that there are no significant changes in the system for contact times above 15 s.

The hydrophobic character of the original PET substrate was confirmed by the high value obtained for its contact angle (78.37 ± 5.11 ° at t = 30 s). This value is in good agreement with previous studies (Carneiro-da-Cunha et al., 2010; Xu et al., 2008). After aminolysis, the PET surface exhibited a significantly (p<0.05) lower contact angle (63.21 ± 4.81 ° at t = 30 s), which means that the A/C PET is more hydrophilic than the original PET. Other authors (Xu et al., 2008) reported a decrease in contact angle from 73.6 ± 1.6 ° to 61.8 ± 1.0 ° after the aminolysis of PET. Also, the difference obtained in contact angle between original and A/C PET confirms that the aminolysis process was correctly performed.

The contact angle results also show a distinct oscillation when the outermost layer changed between κ-carrageenan and chitosan, suggesting the progressive construction of the coating by alternate deposition of the nanolayers. The deposition of the κ-carrageenan layers induced a decrease in contact angle, whereas the deposition of chitosan led to the increase of contact angle. These results can be explained by the higher hydrophobicity of chitosan when compared to κ-carrageenan. This behaviour is in agreement with the QCM measurements (Figure 3.5). Since the outermost layer is κ-carrageenan, the produced coating exhibits a contact angle of approximately 40 °, which makes it a relatively hydrophilic coating.

Moreover, the differences in wettability that were observed as the outermost layer changed between κ-carrageenan and chitosan may also be due to other factors besides the hydrophilicity of the functional groups of the adsorbed layer; such factors may include its chemical composition, the level of interpenetration of the outermost layer by segments of the previously adsorbed polymer layer (Yoo et al., 1998) and the swelling of the layers when in contact to water droplet. These are hypothesis than need to be confirmed by further work.

3.3.2.3 Scanning electron microscopy (SEM)

Scanning electron microscopy (SEM) images of the different samples can be seen in Figure 3.8. From these images it is possible to confirm the formation of the nanolayered coating on the A/C PET surface, corroborating the previous results. Comparing the surface of the original PET (Figure 3.8 a) and the surface of the nanolayered coating (Figure 3.8 b) allows significant differences to be seen. The surface of the nanolayered coating exhibits a more pronounced roughness and a higher number of particles, compared with the original PET surface.

Pinheiro, A. C. (2013)

Figure 3.8 SEM images of the surface morphology of the original PET (a); surface morphology of the A/C PET with the 5 nanolayers (b); first layer of κ-carrageenan (c); coating with two layers (κ-carrageenan and chitosan) (d).

Figure 3.8 c) shows the first layer of κ-carrageenan and in Figure 3.8 d) the two first layers (one of κ-carrageenan and one of chitosan) can be distinguished. SEM images allowed the estimation of the thicknesses of κ-carrageenan (37.9 nm) and chitosan (28.7 nm) layers and therefore of the 5 nanolayers (0.171 µm). However, since both sides of the A/C PET pieces were coated with the same set of layers, to obtain the total thickness it was necessary to multiply the obtained value by 2. Thus, the total thickness of the nanolayers is 0.342 µm.

3.3.2.4 Gas barrier properties

Gas barrier properties of the nanolayered coating were evaluated, using the A/C PET film as reference. The water vapour and oxygen permeabilities (WVP and O2P, respectively) were measured and the results are shown in Table 3.1. The WVP values of the nanolayers were determined based on Equation 3.2, using a thickness of 100 µm, which was measured by a

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation micrometer, for the A/C PET and of 0.342 µm, measured by SEM, for the polysaccharide nanolayers.

Table 3.1 Water vapour permeability (WVP) and oxygen permeability (O2P) of the A/C PET, of the PET with 5 nanolayers and of the 5 polysaccharide nanolayers alone. The values represent the experimental average ± standard deviation -11 -1 -1 -1 -14 -1 -1 -1 Sample WVP x 10 (g.m .s .Pa ) O2P x 10 (g.m .s .Pa .)

A/C PET 2.07 ± 0.34 2.19 ± 0.81

PET + 5 nanolayers 1.53 ± 0.39 1.74 ± 0.35

5 nanolayers 0.020 ± 0.002 0.043 ± 0.027

Since the main function of food packaging is to avoid or decrease water loss, the WVP should be as low as possible (Gontard et al., 1992). Several factors have been shown to influence WVP, including film/coating composition, film/coating thickness and the technique used for film/coating preparation (Bifani et al., 2007). The produced nanolayers exhibited a lower WVP when compared with the values obtained for conventional edible films composed of chitosan (WVP = (8.60 ± 0.14) x 10-11 g.m-1.s-1.Pa-1, ± 50 μm of thickness) (Fajardo et al., 2010) and of sulphated polysaccharides such as κ-carrageenan (WVP between (11.80 ± 0.30) x 10-11 and (235 ± 19.8) x 10-11 g.m-1s-1Pa-1, depending on temperature and humidity gradient; ± 50 μm of thickness) (Hambleton et al., 2008). Also, the κ-carrageenan/chitosan nanolayered coating exhibits a WVP very similar to the value obtained for an alginate/chitosan (WVP = (0.014 ± 0.001) x 10-11 g.m-1s-1Pa-1, 0.12 μm of thickness) (Carneiro-da-Cunha et al., 2010) and κ- carrageenan/lysozyme (WVP = (0.013 ± 0.003) x 10-11 g.m-1s-1Pa-1, 0.47 μm of thickness) (Medeiros et al., 2011) nanolayered coating, both also composed of five nanolayers.

The WVP is strongly governed by the interaction between polymer and water molecules. The hydrophobic character of chitosan nanolayers, confirmed by contact angle measurements, partially explains the lower WVP values of the nanolayers in comparison with the WVP value of A/C PET. These good results may also be explained based on interactions established between the κ-carrageenan and chitosan layers that lead to an increase of the tortuosity of the material and consequently to a decrease of the permeability to the water molecules (Jang et al., 2008; Tieke et al., 2003). Finally, the fact that water molecules must cross several interfaces between the five nanolayers of material may also be contributing to the lower values of WVP.

The food quality can be improved and consequently, the food shelf life can be extended by the application of films/coatings with proper oxygen barrier properties, since the presence of

Pinheiro, A. C. (2013)

oxygen may cause oxidation, which initiates several food alterations such as odor, color, flavor and nutrients deterioration (Sothornvit and Pitak, 2007).

The κ-carrageenan/chitosan nanolayered coating exhibits a O2P lower than a κ- -14 -1 -1 carrageenan/lysozyme nanolayered coating (O2P = (0.1 ± 0.01) x 10 gm s Pa, 0.47 μm of thickness) (Medeiros et al., 2011) and than conventional edible films composed of chitosan

-14 -1 -1 -1 (O2P = (0.71 ± 0.02) x 10 gm s Pa , ± 50 μm of thickness) (Fajardo et al., 2010) and of -14 -1 -1 -1 sulphated polysaccharides such as κ-carrageenan (O2P = (720 ± 280) x 10 gm s Pa , ± 50 μm of thickness) (Hambleton et al., 2008). Generally, polysaccharide films/coatings are good oxygen barriers, since their hydrogen-bonded network structure is firmly packed and arranged (Martins et al., 2010b).

3.3.3 Evaluation of the behaviour of the κ-carrageenan/chitosan nanolayered coating during in vitro digestion

The κ-carrageenan/chitosan nanolayered film was subjected to an in vitro digestion and PET samples were collected, in order to detect the differences on the polysaccharide content of the PET films at different stages of digestion. The results can be seen in Table 3.2.

Table 3.2 Monosaccharide composition of the PET + nanolayers samples at different stages of in vitro digestion

-1 Stage of Monosaccharide composition (µgmonosaccharide.gPET+nanolayers ) Total monosaccharides -1 digestion 3,6-AnGal Man Gal Glc (µg.gPET+nanolayers ) Initial 469.6 ± 62.0 10.1 ± 7.6 353.8 ± 15.6 41.0 ± 1.1 874.5 ± 71.1 Stomach 28.2 ± 0.9 3.0 ± 1.9 34.9 ± 18.2 10.1 ± 2.2 76.2 ± 23.2 Duodenum 17.0 ± 6.0 10.6 ± 3.7 26.6 ± 7.3 37.8 ± 9.1 92.0 ± 26.0 Jejunum 12.1 ± 3.6 10.3 ± 1.2 14.1 ± 1.8 34.4 ± 1.2 70.9 ± 4.2 Ileum 39.8 ± 5.5 5.9 ± 0.6 42.4 ± 6.4 18.4 ± 0.4 106.4 ± 12.2

The analysis of the monosaccharides attached to the PET revealed the presence of 3,6 anhydrogalactose and galactose as the major sugars and also the residual presence of mannose and glucose. These monosaccharides are derived from κ-carrageenan, once this polysaccharides is mainly composed of galactose and 3,6-anhydrogalactose linked by alternating α-1,3 and β-1,4 glycosidic linkages (Spichtig and Austin, 2008). Therefore, the κ- carrageenan content on PET films decreased with digestion, which is probably related to the disassembling of the multilayer structure. Early at stomach stage, there is a significant loss of κ-carrageenan from the PET film, which could be explained by the acidic conditions of the gastric digestion: at pH 2.0, the electrostatic interaction between κ-carrageenan and PET surface and between κ-carrageenan and chitosan is very low, since the pKa value of the anionic

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation sulphate groups on κ-carrageenan is around pH 2 and at this pH the sulphate groups on the κ- carrageenan molecules are half protonated and consequently partially uncharged (Gu et al., 2005). Therefore, the decrease of electrochemical interaction leaded to a partial detachment of the multilayer structure from PET surface. The increase in 3,6-anhydrogalactose and galactose content observed at the ileum stage can be due to the reassembling and adsorption of the carrageenan charged species previously released on top of the charged PET.

The chitosan analysis on PET films was also performed, however, it was not possible to detect any peak (not even the peak corresponding to 2-deoxyglucose, used as internal standard). This was probably due to the PET degradation during the acid hydrolysis with HCl 6 mol.L-1. The products of PET degradation may have interfered in the reduction and/or acetylation reaction. Other reduction and acetylation conditions were tested (e.g. increase of the sodium borohydrate and acetic anhydride concentration, increase of reaction time and addition of a pre-hydrolysis step of the PET with the nanolayers in TFA 120 °C during 1 h to detach the nanolayers from PET) in order to overcome this interference, however all of the tested conditions revealed ineffective in the conversion to alditol acetate derivatives.

3.4 CONCLUSIONS

The obtained results allowed concluding on the nature of interactions between κ-carrageenan and chitosan, two polyelectrolytes of opposite charges. Microcalorimetric measurements showed that the interaction between κ-carrageenan and chitosan is an exothermic process and that the formation of a nanolayered coating occurs mainly due to electrostatic interactions existing between the two polyelectrolytes, though other types of interactions such as dipole- dipole, hydrogen bonds and conformation changes of both polyelectrolytes may also be involved.

From the quartz crystal microbalance measurements it could be concluded that the alternating deposition of κ-carrageenan and chitosan results in the formation of a stable multilayer structure. The κ-carrageenan/chitosan nanolayers exhibit good gas barrier properties and therefore offer great potential to be used e.g. to coat food systems. When subjected to an in vitro digestion, the κ-carrageenan content on PET films decreased with digestion time, which is probably related to the disassembling of the multilayer structure due to the weakening of the electrostatic interactions.

These results contribute to a deeper understanding of the interactions between the polyelectrolytes in a multilayer system and can be of utmost importance when considering the

Pinheiro, A. C. (2013)

development of systems tailored for increasingly specific applications (e.g multilayer systems for the controlled release of bioactive compounds).

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

3.5 REFERENCES

ASTM D3985-02. (2002) Standard test method for oxygen gas transmission rate through plastic film and sheeting using a coulometric sensor, In In: Annual book of ASTM, Amer. Soc. for Testing & Materials, Philadelphia, PA. ASTM E96-92. (1990) Standard test methods for water vapor transmission of materials, In In: Annual book of ASTM standards, Amer. Soc. for Testing & Materials, Philadelphia, PA. Barbosa, A. M., Santos, I. J. B., Ferreira, G. M. D., da Silva, M. D. H., Teixeira, A., & da Silva, L. H. M. (2010). Microcalorimetric and SAXS Determination of PEO-SDS Interactions: The Effect of Cosolutes Formed by Ions, Journal of Physical Chemistry B 114(37), 11967-11974. Begin, A., & Van Calsteren, M. R. (1999). Antimicrobial films produced from chitosan, International Journal of Biological Macromolecules 26(1), 63-67. Bifani, V., Ramirez, C., Ihl, M., Rubilar, M., Garcia, A., & Zaritzky, N. (2007). Effects of murta (Ugni molinae Turcz) extract on gas and water vapor permeability of carboxymethylcellulose-based edible films, LWT-Food Science and Technology 40(8), 1473- 1481. Carneiro-da-Cunha, M. G., Cerqueira, M. A., Souza, B. W. S., Carvalho, S., Quintas, M. A. C., Teixeira, J. A., & Vicente, A. A. (2010). Physical and thermal properties of a chitosan/alginate nanolayered PET film, Carbohydrate Polymers 82(1), 153-159. Coimbra, M. A., Delgadillo, I., Waldron, K. W., & Selvendran, R. R. (1996) Isolation and Analysis of Cell Wall Polymers from Olive Pulp, In Plant Cell Wall Analysis (Linskens, H., and Jackson, J., Eds.), pp 19-44, Springer Berlin Heidelberg. Cooksey, K., Marsh, K. S., & Doar, L. H. (1999). Predicting permeability & transmission rate for multilayer materials, Food Technology 53(9), 60-63. da Silva, L. H. M., da Silva, M. C. H., Francisco, K. R., Cardoso, M. V. C., Minim, L. A., & Coimbra, J. S. R. (2008). PEO- M(CN)(5)NO (x-) (M = Fe, Mn, or Cr) interaction as a driving force in the partitioning of the pentacyanonitrosylmetallate anion in ATPS: Strong effect of the central atom, Journal of Physical Chemistry B 112(37), 11669-11678. Decher, G., & Schlenoff, J. B. (2003) Polyelectrolyte Multilayers, an Overview, Wiley- VCH Verlag GmbH & Co. KGaA, Weinheim Denadai, A. M. L., Santoro, M. M., Da Silva, L. H., Viana, A. T., Santos, R. A. S., & Sinisterra, R. D. (2006). Self-assembly characterization of the beta-cyclodextrin and hydrochlorothiazide system: NMR, phase solubility, ITC and QELS, Journal of Inclusion Phenomena and Macrocyclic Chemistry 55(1-2), 41-49.

Pinheiro, A. C. (2013)

dos Santos, D. S., Riul, A., Malmegrim, R. R., Fonseca, F. J., Oliveira, O. N., & Mattoso, L. H. C. (2003). A Layer-by-Layer Film of Chitosan in a Taste Sensor Application, Macromolecular Bioscience 3(10), 591-595. Fajardo, P., Martins, J. T., Fucinos, C., Pastrana, L., Teixeira, J. A., & Vicente, A. A. (2010). Evaluation of a chitosan-based edible film as carrier of natamycin to improve the storability of Saloio cheese, Journal of Food Engineering 101(4), 349-356. Fakhrullin, R. F., Vinter, V. G., Zamaleeva, A. I., Matveeva, M. V., Kourbanov, R. A., Temesgen, B. K., Ishmuchametova, D. G., Abramova, Z. I., Konovalova, O. A., & Salakhov, M. K. (2007). Quartz crystal microbalance immunosensor for the detection of antibodies to double- stranded DNA, Analytical and Bioanalytical Chemistry 388(2), 367-375. Fu, J., Ji, J., Yuan, W., & Shen, J. (2005). Construction of anti-adhesive and antibacterial multilayer films via layer-by-layer assembly of heparin and chitosan, Biomaterials 26(33), 6684- 6692. Gontard, N., Guilbert, S., & Cuq, J.-L. (1992). Edible Wheat Gluten Films: Influence of the Main Process Variables on Film Properties using Response Surface Methodology, Journal of Food Science 57(1), 190-195. Gu, Y. S., Decker, E. A., & McClements, D. J. (2005). Influence of pH and carrageenan type on properties of beta-lactoglobulin stabilized oil-in-water emulsions, Food Hydrocolloids 19(1), 83-91. Hambleton, A., Debeaufort, F., Beney, L., Karbowiak, T., & Voilley, A. (2008). Protection of active aroma, compound against moisture and oxygen by encapsulation in biopolymeric emulsion-based edible films, Biomacromolecules 9(3), 1058-1063. Indest, T., Laine, J., Johansson, L. S., Stana-Kleinschek, K., Strnad, S., Dworczak, R., & Ribitsch, V. (2009). Adsorption of Fucoidan and Chitosan Sulfate on Chitosan Modified PET Films Monitored by QCM-D, Biomacromolecules 10(3), 630-637. Indest, T., Laine, J., Ribitsch, V., Johansson, L.-S., Stana-Kleinschek, K., & Strnad, S. (2008). Adsorption of Chitosan on PET Films Monitored by Quartz Crystal Microbalance, Biomacromolecules 9(8), 2207-2214. Jang, W. S., Rawson, I., & Grunlan, J. C. (2008). Layer-by-layer assembly of thin film oxygen barrier, Thin Solid Films 516(15), 4819-4825. Lavalle, P., Gergely, C., Cuisinier, F. J. G., Decher, G., Schaaf, P., Voegel, J. C., & Picart, C. (2002). Comparison of the structure of polyelectrolyte multilayer films exhibiting a linear and an exponential growth regime: An in situ atomic force microscopy study, Macromolecules 35(11), 4458-4465.

Chapter 3. Interactions between κ-carrageenan and chitosan in nanolayered coatings- Structural and transport properties evaluation

Martins, G. V., Mano, J. F., & Alves, N. M. (2010a). Nanostructured self-assembled films containing chitosan fabricated at neutral pH, Carbohydrate Polymers 80(2), 570-573. Martins, J. T., Cerqueira, M. A., Souza, B. W. S., Avides, M. C., & Vicente, A. A. (2010b). Shelf Life Extension of Ricotta Cheese Using Coatings of Galactomannans from Nonconventional Sources Incorporating Nisin against Listeria monocytogenes, Journal of Agricultural and Food Chemistry 58(3), 1884-1891. Marx, K. A. (2003). Quartz crystal microbalance: A useful tool for studying thin polymer films and complex biomolecular systems at the solution-surface interface, Biomacromolecules 4(5), 1099-1120. Medeiros, B., Pinheiro, A., Teixeira, J., Vicente, A., & Carneiro-da-Cunha, M. (2011). Polysaccharide/Protein Nanomultilayer Coatings: Construction, Characterization and Evaluation of Their Effect on ‘Rocha’ Pear Shelf-Life, Food and Bioprocess Technology, 1-11. Park, S. Y., Lee, B. I., Jung, S. T., & Park, H. J. (2001). Biopolymer composite films based on [kappa]-carrageenan and chitosan, Materials Research Bulletin 36(3-4), 511-519. Pranoto, Y., Rakshit, S. K., & Salokhe, V. M. (2005). Enhancing antimicrobial activity of chitosan films by incorporating garlic oil, potassium sorbate and nisin, LWT-Food Science and Technology 38(8), 859-865. Reis, P., Raab, T., Chuat, J., Leser, M., Miller, R., Watzke, H., & Holmberg, K. (2008). Influence of Surfactants on Lipase Fat Digestion in a Model Gastro-intestinal System, Food Biophysics 3(4), 370-381. Rudra, J. S., Dave, K., & Haynie, D. T. (2006). Antimicrobial polypeptide multilayer nanocoatings, Journal of Biomaterials Science-Polymer Edition 17(11), 1301-1315. Shi, P., Zuo, Y., Zou, Q., Shen, J., Zhang, L., Li, Y., & Morsi, Y. S. (2009). Improved properties of incorporated chitosan film with ethyl cellulose microspheres for controlled release, International Journal of Pharmaceutics 375(1-2), 67-74. Sobral, P. J. A., Menegalli, F. C., Hubinger, M. D., & Roques, M. A. (2001). Mechanical, water vapor barrier and thermal properties of gelatin based edible films, Food Hydrocolloids 15(4-6), 423-432. Sothornvit, R., & Pitak, N. (2007). Oxygen permeability and mechanical properties of banana films, Food Research International 40(3), 365-370. Spichtig, V., & Austin, S. (2008). Determination of the low molecular weight fraction of food-grade carrageenans, Journal of Chromatography B 861(1), 81-87. Stevenson, T. T., & Furneaux, R. H. (1991). Chemical methods for the analysis of sulphated galactans from red algae, Carbohydrate Research 210(0), 277-298.

Pinheiro, A. C. (2013)

Tieke, B., Pyrasch, M., & Toutianoush, A. (2003) Functional Layer-by-Layer Assemblies with Photo- and Electrochemical Response and Selective Transport of Small Molecules and Ions, In Multilayer thin films: Sequential assembly of nanocomposite materials (Decher, G., and Schlenoff, J. B., Eds.), pp 427–458, Wiley - VCH Verlag GmbH & Co. KGaA, Weinheim Vinayahan, T., Williams, P. A., & Phillips, G. O. (2010). Electrostatic Interaction and Complex Formation between Gum Arabic and Bovine Serum Albumin, Biomacromolecules 11(12), 3367-3374. Wang, Y., Angelatos, A. S., & Caruso, F. (2007). Template Synthesis of Nanostructured Materials via Layer-by-Layer Assembly†, Chemistry of Materials 20(3), 848-858. Wasikiewicz, J. M., Yoshii, F., Nagasawa, N., Wach, R. A., & Mitomo, H. (2005). Degradation of chitosan and sodium alginate by gamma radiation, sonochemical and ultraviolet methods, Radiation Physics and Chemistry 73(5), 287-295. Weiss, J., Takhistov, P., & McClements, D. J. (2006). Functional materials in food nanotechnology, Journal of Food Science 71(9), R107-R116. Xu, J. P., Wang, X. L., Fan, D. Z., Ji, J., & Shen, J. C. (2008). Construction of phospholipid anti-biofouling multilayer on biomedical PET surfaces, Applied Surface Science 255(2), 538-540. Yoo, D., Shiratori, S. S., & Rubner, M. F. (1998). Controlling bilayer composition and surface wettability of sequentially adsorbed multilayers of weak polyelectrolytes, Macromolecules 31(13), 4309-4318. Ziani, K., Oses, J., Coma, V., & Maté, J. I. (2008). Effect of the presence of glycerol and Tween 20 on the chemical and physical properties of films based on chitosan with different degree of deacetylation, LWT - Food Science and Technology 41(10), 2159-2165.

4 CHAPTER 4. Κ - CARRAGEENAN/CHITOSAN

NANOLAYERED COATING FOR CONTROLLED RELEASE OF A MODEL BIOACTIVE COMPOUND Κ-CARRAGEENAN/CHITOSAN NANOLAYERED COATING FOR CONTROLLED RELEASE OF A MODEL BIOACTIVE COMPOUND

4.1 Introduction...... 99

4.2 Materials and Methods ...... 100

4.3 Results and Discussion ...... 103

4.4 Conclusions...... 112

4.5 References ...... 113

Pinheiro, A. C. (2013)

The results presented in this Chapter were adapted from:

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

4.1 INTRODUCTION

Over the past years, there has been an increasing interest in the field of bioactive compounds delivery, resulting in new controlled release systems that provide continuous liberation of the functional compounds. However, particularly in the food industry, the development of versatile and highly responsive release systems still constitutes a challenge. Nanolayered coatings have promising applications in a wide range of areas including pharmaceutical, biomedical and food packaging. Particularly, when applied on food products, they can be used as a strategy for shelf-life extension. However, there is a need to create systems of multilayers that combine both biofunctionality and edibility. Conventional layer-by-layer (LbL) assembly is based on the alternating deposition of oppositely charged polyelectrolytes and can be used to produce multilayers with thickness control on the nanometre scale (Decher and Schlenoff, 2003). In general, the resulting polyelectrolyte multilayer films/coatings are formed by interpenetrated polyelectrolyte chains and not by a strictly stratified structure (Decher, 1997).

More recently, LbL nanolayered film assemblies have been investigated as drug release systems (Chung and Rubner, 2002; Jiang and Li, 2009; Zhong et al., 2007) due to the possibility to control the drug release through manipulating the film/coating properties and to incorporate a wide range of functional biomolecules without substantial loss of their biological functions (Wang et al., 2007). The functional LbL multilayers can be obtained by chemical grafting of polyelectrolytes by functional moieties (Kaschak et al., 1999), alternate deposition of polyelectrolytes and functional molecules (Rousseau et al., 2000) or by post diffusion of the functional molecules into the multilayers (Jiang and Li, 2009; Zhong et al., 2007). A diversity of multilayered films has been developed to control bioactive compounds release by deploying different triggers such as pH, ionic strength, temperature and enzymes (Quinn and Caruso, 2004; Serizawa et al., 2002; Wood et al., 2005).

Chitosan, a cationic polysaccharide obtained by deacetylation of chitin, is an excellent edible film component due to its good oxygen barrier properties and due to its intrinsic antimicrobial activity (Begin and Van Calsteren, 1999) and antioxidant activity (Xia et al., 2011). Moreover, properties such as biodegradability, non-toxicity and biocompatibility make chitosan suitable for application in pharmaceutical and food products (Dutta et al., 2009; Illum et al., 2001). - carrageenan, a sulphated anionic polysaccharide extracted from certain red seaweeds, is extensively used in food industry as gelling and stabilizing agent and has been reported as having excellent film forming properties (Park et al., 2001). A nanolayered coating composed of κ-carrageenan and chitosan was previously developed and it has proved to exhibit good gas barrier properties (Chapter 3). Therefore, this system offers great potential to be used to coat

Pinheiro, A. C. (2013)

food systems such as fruit, vegetables or cheese and to act as a support for the incorporation of bioactive compounds. Methylene blue (MB), a monovalent cation of 374 Da, is highly soluble in water and is a suitable model for loading and release experiments of small molecules due to its large absorption peak in the visible range. MB has been used to investigate multilayer surface properties due to its ability to bind to available negatively charged functional groups (Chung and Rubner, 2002). Moreover, the loading of dye molecules, such as MB, to preformed layers offers the possibility to evaluate the extent of electrostatic interactions between oppositely charged polyelectrolytes in the multilayer films/coatings (Tedeschi et al., 2000; Yoo et al., 1998).

The release of bioactive compounds from polymeric matrices may occur due to mechanisms of Fick’s diffusion, polymer matrix swelling, polymer erosion, and degradation (Faisant et al., 2002; Jain, 2000) and a different mechanism may prevail, depending on the system and environmental conditions. In recent years, several works concerning the release of functional compounds from conventional edible films have been done (Del Nobile et al., 2008; Flores et al., 2007). Comparatively, few works can be found on release mechanisms involved at the nano-scale. Literature suggests that in the case of nanolayered coatings, the release behaviour depends on the permeability and on the disassembly or erosion of the multilayer structure and on other experimental variables (Jiang and Li, 2009). Understanding the release mechanisms involved, by recurring to mathematical modelling, is very important for food products development and the potential applications thereof.

In this research, multilayer nanocoatings were prepared by LbL deposition using κ-carrageenan and chitosan onto a substrate (polyethylene terephthalate - PET). The model compound MB was incorporated at different positions of the nanolayered coating and its loading and release behaviour was evaluated. Diffusion of MB into liquid medium was studied at different isothermal conditions and mathematical models were used to understand the transport mechanisms.

4.2 MATERIALS AND METHODS

4.2.1 Experimental Procedures

4.2.1.1 Preparation of nanolayered coating

The polyelectrolyte solution of κ-carrageenan was prepared dissolving 0.2% (w/v) of κ- carrageenan (GENUGEL carrageenan type WR-78, CPKelco, Denmark) in distilled water and the pH of the solution was adjusted to 7.0 with a solution of 1 mol.L-1 sodium hydroxide (Riedel-de

100

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

Haёn, Germany). The polyelectrolyte solution of chitosan was prepared dissolving 0.2% (w/v) of chitosan (90% deacetylation, MW = 3.4 x 105, Aqua Premier Co. Ltd, Thailand) in a 1.0% (v/v) lactic acid solution and the pH of the solution was adjusted to 3.0 with a solution of 1 mol.L-1 sodium hydroxide (Riedel-de Haёn, Germany). MB solution was prepared dissolving 0.3% (w/v) of MB (Riedel-de Haёn, Germany) in 50% (v/v) 0.01 mol.L-1 Phosphate Buffer Solution (PBS) and ultra-pure water, and the pH was adjusted to 7.0 with a solution of 1 mol.L-1 hydrochloric acid (Merck, Darmstadt, Germany).

The pH values chosen for κ-carrageenan, chitosan and MB solutions are those at which these polyelectrolytes exhibit the highest ζ-potential values (positive or negative), thus maximizing the electrostatic interactions between these polyelectrolytes.

The nanolayered coating was composed by an aminolyzed/charged PET (A/C PET) (Canson, France) sustaining layer adsorbed with a polysaccharide multilayer constituted of 5 layers (three κ-carrageenan and two chitosan layers). MB was incorporated in the κ- carrageenan/chitosan nanolayered coating, forming the second (κ-carrageenan-MB-κ- carrageenan-Chitosan-κ-carrageenan-Chitosan), fourth (κ-carrageenan-Chitosan-κ- carrageenan-MB-κ-carrageenan-Chitosan) or the sixth layer (κ-carrageenan-Chitosan-κ- carrageenan-Chitosan-κ-carrageenan-MB), thus replacing the corresponding chitosan layer in those positions (Figure 4. 1).

a) b) c)

Figure 4.1 Schematic representation of the MB loading in the different positions of the κ-carrageenan/chitosan nanolayered: second layer (a), fourth layer (b) and sixth layer (c).

In order to prepare the nanolayered coating, the PET pieces were aminolyzed (Carneiro-da- Cunha et al., 2010) and were firstly dipped into the κ-carrageenan solution for 15 min and subsequently rinsed with distilled water at pH 7. The samples were dried with a flow of nitrogen. The procedure was repeated, this time using chitosan (or MB) as the polyelectrolyte and rinsed with distilled water at pH 3.0 (or 7.0). The dipping/ washing/ drying process was repeated until the deposition of the desirable layers conformation was achieved. The

101

Pinheiro, A. C. (2013)

nanolayered coatings were then maintained at 20 °C and 50% relative humidity (RH) before analysis.

4.2.1.2 MB loading

The loading of MB on each position of the multilayer structure was monitored by UV-VIS spectroscopy (Jasco 560, Germany), in the range 190-800 nm. The maximum MB absorption peak in the nanolayered coating was 570 nm. Three replicates were performed.

Real time deposition of MB on the coated electrode was recorded by Quartz Crystal Microbalance (QCM 200 purchased from Stanford Research Systems, SRS, USA), equipped with AT-cut quartz crystals (5 MHz) with optically flat polished titanium/gold electrodes on contact and liquid sides. Three replicates were performed.

The deposition of the layers and their nano dimensions were also confirmed by scanning electron microscopy (SEM) (Nova NanoSEM 200, Netherlands) with an accelerating voltage from 10 to 15 kV. Before analyses, all samples were mounted on aluminium stubs using carbon adhesive tape and sputter-coated with gold (thickness of about 10 nm). The SEM images allowed the direct measurement of the layers thicknesses using the equipment’s image analysis software (XT microscope control software).

4.2.1.3 MB release

The release of MB from the multilayer coating on PET was evaluated by incubating the loaded coatings in 40 mL of 0.01 mol.L-1 PBS of a certain pH (2.0 or 7.0) and temperature (4 or 37 °C). The release medium was continuously stirred at 60 rpm. At pre-set intervals, 0.25 mL supernatant was taken and 0.25 mL of fresh PBS was added to keep the volume of the release medium constant. The sample solutions of MB were analysed using UV-VIS spectroscopy (Jasco 560, Germany) by measuring their absorbance at 600 nm – the maximum absorption peak of MB in solution. Three replicates were conducted for each condition.

4.2.2 Statistical Procedures

The equations mentioned along the text were fitted to data by non-linear regression analysis, using a package of STATISTICA 7.0 (Statsoft. Inc, USA). The Levenberg-Marquadt algorithm for the least squares function minimization was used.

The quality of the regressions was evaluated on the basis of the determination coefficient, R2, the squared root mean square error, RMSE (i.e., the square root of the sum of the squared

102

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound residues (SSE) divided by the regression degrees of freedom) and residuals visual inspection for randomness and normality. R2 and SSE were obtained directly from the software.

The precision of the estimated parameters was evaluated by the Standardized Halved Width (SHW %), which was defined as the ratio between the 95% Standard Error (obtained from the software) and the value of the estimate.

4.3 RESULTS AND DISCUSSION

4.3.1 MB loading

Nanolayered film/coating architecture is important for bioactive compound loading. The MB loading at different positions of the κ-carrageenan/chitosan multilayered coating was evaluated by UV–Vis spectroscopy (Figure 4.2). For all cases, the maximum absorbance was observed at 570 nm, which corresponds to a characteristic absorption peak of MB, thus substantiating the incorporation of this compound in the κ-carrageenan/chitosan nanolayered coatings.

Figure 4.2 UV-Vis absorption spectra of the κ-carrageenan/chitosan coatings as a function of the MB layer position in the multilayer structure: MB in second (___), forth (___) and sixth layer (___).

It can be seen that the MB absorbance increased with the distance from the first layer. This can be due to the fact that full establishment of the layer organization can occur only after deposition of a minimum number of layers, depending on the system (Yoo et al., 1998). Literature proposes that before this minimum number of layers is achieved, interpenetration between layers can occur (Decher, 1997). This phenomenon may lead to uneven deposition in the first layers of the nano-structure and thus less charged chain segments for adhesion of MB are available (Yoo et al., 1998) (Decher, 1997). Also, it is possible that the MB adsorption is not

103

Pinheiro, A. C. (2013)

confined to the outermost surface but MB was able to diffuse into the κ-carrageenan/chitosan nanolayered coating. If that was the case, as the multilayer coating was built, coating´s total thickness increased and consequently more MB was loaded (Chung and Rubner, 2002).

The interpenetration and diffusion phenomena can also be inferred by the results of real-time adsorption of MB on κ-carrageenan/chitosan pre-coated gold electrode surface, which were monitored using a quartz crystal microbalance (QCM). The mass changes as a function of time for the deposition of MB constituting the second, fourth or sixth layer of the multilayer structure are shown in Figure 4.3.

Figure 4.3 Real time monitoring of the deposition of MB forming the second (), fourth () and sixth () layer of the κ-carrageenan/chitosan multilayered coatings using a quartz crystal microbalance (QCM).

It was found that MB loaded mass increased with the incubation time, exhibiting a very fast initial adsorption phase followed by a slower phase upon approaching a steady state.

The initial fast adsorption phase shows an increase of total MB loaded with the number of layer, with the MB loading capacity in the sixth layer of the multilayer structure being about ten times that in the second layer. Since the incorporation of MB in the κ- carrageenan/chitosan nanolayered coating is mainly driven by the electrostatic interaction between MB and κ-carrageenan molecules and the amount of MB adsorbed depends mainly on the number of anionic groups of κ-carrageenan not forming contact ion pairs with cationic groups of chitosan (i.e., on the number of anionic groups available for further interactions) (Soedjak, 1994), the increase of MB loading with layer number supports the existence of high interpenetration between layers in the innermost layers and of a more homogenous surface as the layers are consecutively deposed.

As for the slower phase of MB loading, it can be a reflection of MB diffusion into the nanolayered coating. The increase of the slope of the curve with layer number observed in

104

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

Figure 4.3 supports the existence of such phenomenon and that it is proportional to coating’s thickness.

From these results, one can conclude that it is possible to control the amount of bioactive compound adsorbed by simply controlling its position in the multilayer structure.

The scanning electron microscopy (SEM) images of the κ-carrageenan/chitosan coating allowed confirming the nano-scale of the obtained layers. Figure 4.4 shows the SEM image of the coating with MB in the sixth layer as an example. The total thickness of the κ- carrageenan/chitosan layers was 234.9 nm (Figure 4.4 a). Also, a detailed image (Figure 4.4 b) shows the structure of the layers, revealing that κ-carrageenan/chitosan coating is formed by interpenetrated layers and not by strictly stratified layers.

a) b)

Figure 4.4 SEM images of the κ-carrageenan/chitosan coating with MB in the sixth layer. Thickness of the layers (a). Detail of the structure of the layers (b).

4.3.2 MB release

Experimentally, all the release profiles were characterized by three different parts: an initial burst phase, a continuous release phase and a stagnant phase. These results can be interpreted on the basis of mass transport theory: pure diffusion, polymer matrix swelling, polymer erosion and degradation are mechanisms that lead to compound release from polymeric devices (Faisant et al., 2002; Jain, 2000; Polakovic et al., 1999). Depending on both the system (polymer/active compound) and environmental conditions, a different mechanism may prevail. These mechanisms may be generally classified as of three different types: ideal Fickian diffusion (Brownian transport); Case II transport (polymer relaxation driven) and anomalous behaviour (ranging from Fickian to Case II transport). At the nano-scale, few works in literature address compound release mechanisms from layered coatings. Literature suggests

105

Pinheiro, A. C. (2013)

that, depending on nanolayered coatings composition and on the released compound, release may follow a Fickian or an anomalous transport behaviour (Wang et al., 2007).

The linear superimposition model (LSM) accounts for both Fickian and Case II transport effects on the observed anomalous behaviour in hydrophilic matrices (Berens and Hopfenberg, 1978) and is then a suitable approach to investigate transport mechanism in these systems.

This approach assumes that the observed transport of molecules within the polymer can be described by the sum of the molecules transported due to Brownian motion (Fick´s transport) with the molecules transported due to polymer relaxation.

For a thin slab of polymer immersed in a sufficiently large amount of water, molecular transport due to Brownian motion can be described by the solution of Fick’s second law for a plane sheet with constant boundary conditions (Crank, 1975), which can be simplified using the first term of the Taylor series (Rathore and Kapuno, 2011):

( ) Equation 4.1

where Mt is the total mass released from the polymer at time t, M∞,F is the compound release at equilibrium and kF is Fickian diffusion rate constant.

As for polymer relaxation, it is driven by the swelling ability of the polymer and is then related to the dissipation of stress induced by the entry of the penetrant and can be described as a distribution of relaxation times, each following a first order-type kinetic equation (Berens and Hopfenberg, 1978).

∑ [ ( )] Equation 4.2

here, M∞,Ri are the contributions of the relaxation processes for compound release and kRi are the relaxation ith rate constants. For most cases, there is only one main polymer relaxation that influences transport ant thus the above equation can be simplified using i = 1.

Assuming the above conditions, the linear superimposition model for compound release from a hydrophilic polymer slab yields (Flores et al., 2007):

[ ( )] ( )[ ( )] Equation 4.3

where X is the fraction of compound released by Fickian transport.

This “general” model can then be used to describe Fickian (M∞,F ≠ 0 and i = 0); anomalous

(M∞,F and i ≠ 0) or Case II transport (M∞,F = 0 and i ≠ 0).

106

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

To assess the transport mechanism at the nano-scale and investigate the phenomena observed at this scale, release experimental results were analysed by fitting Equation 4.1 and Equation 4.3 to the obtained data. The effects of temperature, pH and layer position on MB release were evaluated.

4.3.2.1 Assessment of Fick´s transport in nanolayered coatings

In all release experiments carried under different conditions (temperature 4 and 37 °C; pH 2 and 7) and from the different layers (second, fourth and sixth), results from fitting Equation 4.1 lead to a poor description of the data (see examples on Figure 4.5 and 4.6). This indicates that transport mechanism for the tested nanolayered coatings cannot be described by only Brownian motion of MB molecules in the polymeric matrix, i.e. it does not follow Fick’s behaviour.

4.3.2.2 Anomalous transport in nanolayered coatings

As for fitting Equation 4.3 - i.e. assuming anomalous behaviour with one main relaxation - to the experimental data, a successful description was achieved for most of the tested conditions. However, different results can be observed depending on the position of MB incorporated on the nanolayered coating or the pH and temperature of the medium.

4.3.2.2.1 MB release from the outermost layer

Figure 4.5 a) shows an example of the adequate description of the LSM (Equation 4.3) of release of MB from the sixth layer and Table 4.1 presents the regression analysis results. It can be observed that this model adequately describes the experimental data with good regression quality and parameters are estimated with good precision (maximum SHW% was 57%).

These results show that in the outermost layer, the mechanism of diffusion is governed by both Fickian and Case II transport, with only one main relaxation of the polymeric coating. The fact that polymer relaxation has an influence on MB release supports the existence of interpenetration between layers even between MB in the outermost and κ-carrageenan on the fifth layers and/or diffusion of MB into the inner layers of the nano-device.

107

Pinheiro, A. C. (2013)

Figure 4.5 Example of Fick´s (Equation 4.1) and LSM (Equation 4.3) description of Methylene Blue release at 37 °C from: (a) the sixth layer at pH=7 and (b) the fourth layer at pH=2. Insets show the detail of the model fitting to the initial experimental data. Experimental data (x), results from fitting Equation 4.1 (-----) and Equation 4.3 (____).

The knowledge of the effects of the environmental conditions (such as temperature and pH) on the quantity of mass released and on the rate of release can be of greatest importance for the application of these systems on food products. Furthermore, once applied on a food system, the release of the bioactive compound can occur after or before ingestion. Therefore, the release behaviour of MB from κ-carrageenan/chitosan nanolayered coatings was evaluated at 37 and 4 °C, which are the temperatures within the human body and the standard refrigeration temperature of a food product, respectively, and at a pH of 2.0 and 7.0, which are the pHs corresponding to the conditions within a human gastrointestinal tract (particularly in the stomach and in the small intestine, respectively). As can be evaluated from Table 4.1, MB release from the sixth layer was influenced by both temperature and pH. Although these effects were not significant (p>0.05) they provide a good insight on the behaviour of the polymeric nanolayered coating at the different environmental conditions tested.

108

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

Layer 2 -1 -1 pH Temp (°C) RMSE R X kF (min ) kR (min ) Position

0.775 3.58x10-2 2.81x10-1 37 3.82x10-2 0.9854 (15%) (18%) (49%) 7 0.661 1.27x10-2 3.51x10-2 4 7.96x10-2 0.9461 (44%) (41%) (57%) 6 0.464 2.93x10-2 3.19x10-1 37 7.52x10-2 0.9378 (40%) (55%) (40%) 2 0.421 2.25x10-2 8.16x10-2 4 4.05x10-2 0.9800 (45%) (41%) (20%)

0.124 7.20x10-3 1.30x10-2 37 0.9635 6.60x10-2 (258%) (228%) (26%) 2 2 -2 0.734 5.70x10-3 3.56x10 4 0.9246 9.12x10-2 (35%) (33%) (88%)

-1 0.656 1.67x10-2 2.21x10 37 0.9068 9.52x10-2 (29%) (44.24%) (76%) 4 2 -2 0.650 5.40x10-3 5.77x10 4 0.8994 1.01x10-1 (31%) (37%) (83%)

Fick’s diffusion contribution to the total release from the nanolayered coating can be evaluated by the estimate of X (which is defined as ). It can be observed that for neutral pH (7), X is higher than 0.5, indicating that Fick’s is the main release mechanism in this condition. As for the MB release in pH = 2 relaxation is the governing phenomenon (X < 0.5). At pH 2.0, the electrostatic interaction between κ-carrageenan and MB is very low, since the pKa value of the anionic sulphate groups on κ-carrageenan is around pH 2 and at this pH the sulphate groups on the κ-carrageenan molecules are half protonated and consequently partially uncharged (Gu et al., 2005). The decrease of electrochemical interaction with chitosan or MB facilitates the loosening of the polymer chain network and promotes release due to polymer relaxation.

109

Pinheiro, A. C. (2013)

In both studied pH conditions, the amount of MB released through Fick’s transport (evaluated by X estimate) increased with process temperature – this is of course not unexpected – a higher internal energy of a system will promote increased molecular vibration and movement, i.e. Brownian motion; and increased solubility of diffusing molecules (Myint et al., 1996;

Vojdani and Torres, 1990). Similarly, the Fickian and relaxation rate constants (kF and kR) increased with temperature.

The relaxation rate constant (kR), decreased with increasing pH for both tested temperatures. This supports the previously discussed effect of acidic pH on promoting transport through relaxation of the polymeric network.

As for kF, the Fickian rate constant, the effect of pH was different at the two tested temperatures. While at 37 °C kF increased with pH, thus reflecting Fick´s behaviour for the tightly packed polymer network at neutral pH, at 4 °C this parameter showed an unexpected inverse behaviour – kF was higher at pH = 2. This observation does not mean that Fick’s diffusion was higher than polymer relaxation under this condition, but it shows that for a system with low internal energy (i.e. low temperature) and for a tightly packed polymer network (pH = 7) the molecular mobility can be decreased to a point where even Fick´s diffusion is limited.

4.3.2.2.2 MB release from the second and fourth Layers

MB release in the inner layers (second and fourth) presented a different behaviour than the one previously discussed.

Under acidic conditions (pH = 2), the transport of MB can be described by the LSM (Equation 4.3), as shown on Figure 4.5 b). Results of the regression analysis (Table 4.1) show that although the model describes the experimental data, parameters are estimated with low precision (maximum SHW% was 258%), hindering any conclusion on the effect of environmental conditions on the diffusion kinetics.

For the inner layers, at neutral pH, both Equation 4.1 and Equation 4.3 were not able to describe the experimental data (see Figure 4.6). This is related with the fact that, as discussed above, κ-carrageenan is negatively charged at pH 7, while chitosan and MB are positively charged, leading to electrochemical interaction. This stronger interaction affects the transport mechanism, which cannot be described by either Fickian or anomalous behaviour.

110

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

Figure 4.6 Fick´s (Equation 4.1) and LSM (Equation 4.3) description of Methylene Blue release at pH 7. Example is for the fourth layer at 4 °C. Experimental data (x), results from fitting Equation 4.1 (-----) and Equation 4.3 (____).

These results show that the MB position on the nanolayered coating influences its release. As expected, at the same release conditions, the MB in the sixth layer enabled a more extensive and faster release of this compound, when compared with the MB in the fourth or second layers. Therefore, depending on the application, the quantity of the mass loaded and consequently released and the rate of release can be controlled by simple altering the position of the compound of interest within the nanolayered structure.

4.3.2.3 Other mechanisms of transport in nanolayered coatings

One should consider that other factors that are not accounted in the Linear Superimposition Model (such as the disassembly or erosion of the multilayer structure) may be involved on MB release (Jiang and Li, 2009). As an initial approach to this question, the diffusion and relaxation rate constants estimated at the different nano-layers can be compared. For pH = 2, the condition where kR and kF could be estimated in all the studied layers, it was observed that the resistance to transport was higher in the inner layers. This may be an indication that the disassembly or the erosion of the multilayer structure was not a predominant factor in this case.

From these results, one can conclude that the main physical mechanism of the MB transport in nanolayered κ-carrageenan/chitosan coatings is anomalous transport, with one main polymer relaxation. Similar results were obtained by other authors that observed that the release of azoalbumin is a complex combination of pure diffusion and Case II transport (Wang et al., 2007).

111

Pinheiro, A. C. (2013)

4.4 CONCLUSIONS

The results obtained in this work provide useful information on the morphology and architecture of the nanolayered coatings. UV-VIS spectroscopy and quartz crystal microbalance analysis demonstrated that the amount of MB loaded increased with the distance from the first layer, suggesting layer interpenetration of the inner layers and that MB was able to diffuse into the κ-carrageenan/chitosan nanolayered coating.

The results of fitting the Linear Superimposition Model to the experimental data of MB release suggest an anomalous behaviour, with one main polymer relaxation. The effects of layer position, temperature and pH on MB release were evaluated and different results were observed depending on the position of MB incorporated on the nanolayered coating or the pH and temperature of the medium.

This work allows clarifying the release mechanisms involved at nano-scale, which is of the utmost importance for the application of nanolayered systems in food products, as a strategy for shelf-life extension.

112

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

4.5 REFERENCES

Begin, A., & Van Calsteren, M. R. (1999). Antimicrobial films produced from chitosan, International Journal of Biological Macromolecules 26(1), 63-67. Berens, A. R., & Hopfenberg, H. B. (1978). Diffusion and relaxation in glassy polymer powders. 2. Separation of diffusion and relaxation parameters, Polymer 19(5), 489-496. Carneiro-da-Cunha, M. G., Cerqueira, M. A., Souza, B. W. S., Carvalho, S., Quintas, M. A. C., Teixeira, J. A., & Vicente, A. A. (2010). Physical and thermal properties of a chitosan/alginate nanolayered PET film, Carbohydrate Polymers 82(1), 153-159. Chung, A. J., & Rubner, M. F. (2002). Methods of Loading and Releasing Low Molecular Weight Cationic Molecules in Weak Polyelectrolyte Multilayer Films, Langmuir 18(4), 1176- 1183. Crank, J. (1975) The mathematics of diffusion, 2 ed., Clarendon Press, Oxford. Decher, G. (1997). Fuzzy nanoassemblies: Toward layered polymeric multicomposites, Science 277(5330), 1232-1237. Decher, G., & Schlenoff, J. B. (2003) Polyelectrolyte Multilayers, an Overview, Wiley- VCH Verlag GmbH & Co. KGaA, Weinheim Del Nobile, M. A., Conte, A., Incoronato, A. L., & Panza, O. (2008). Antimicrobial efficacy and release kinetics of thymol from zein films, Journal of Food Engineering 89(1), 57- 63. Dutta, P. K., Tripathi, S., Mehrotra, G. K., & Dutta, J. (2009). Perspectives for chitosan based antimicrobial films in food applications, Food Chemistry 114(4), 1173-1182. Faisant, N., Siepmann, J., & Benoit, J. P. (2002). PLGA-based microparticles: elucidation of mechanisms and a new, simple mathematical model quantifying drug release, European Journal of Pharmaceutical Sciences 15(4), 355-366. Flores, S., Conte, A., Campos, C., Gerschenson, L., & Del Nobile, M. (2007). Mass transport properties of tapioca-based active edible films, Journal of Food Engineering 81(3), 580-586. Gu, Y. S., Decker, E. A., & McClements, D. J. (2005). Influence of pH and carrageenan type on properties of beta-lactoglobulin stabilized oil-in-water emulsions, Food Hydrocolloids 19(1), 83-91. Illum, L., Jabbal-Gill, I., Hinchcliffe, M., Fisher, A. N., & Davis, S. S. (2001). Chitosan as a novel nasal delivery system for vaccines, Advanced Drug Delivery Reviews 51(1-3), 81-96. Jain, R. A. (2000). The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices, Biomaterials 21(23), 2475-2490.

113

Pinheiro, A. C. (2013)

Jiang, B. B., & Li, B. Y. (2009). Tunable drug loading and release from polypeptide multilayer nanofilms, International Journal of Nanomedicine 4(1), 37-53. Kaschak, D. M., Lean, J. T., Waraksa, C. C., Saupe, G. B., Usami, H., & Mallouk, T. E. (1999). Photoinduced energy and electron transfer reactions in lamellar polyanion/polycation thin films: Toward an inorganic "leaf", Journal of the American Chemical Society 121(14), 3435- 3445. Myint, S., Daud, W. R. W., Mohamad, A. B., & Kadhum, A. A. H. (1996). Temperature- dependent diffusion coefficient of soluble substances during ethanol extraction of clove, Journal of the American Oil Chemists Society 73(5), 603-610. Park, S. Y., Lee, B. I., Jung, S. T., & Park, H. J. (2001). Biopolymer composite films based on [kappa]-carrageenan and chitosan, Materials Research Bulletin 36(3-4), 511-519. Polakovic, M., Görner, T., Gref, R., & Dellacherie, E. (1999). Lidocaine loaded biodegradable nanospheres: II. Modelling of drug release, Journal of Controlled Release 60(2- 3), 169-177. Quinn, J. F., & Caruso, F. (2004). Facile tailoring of film morphology and release properties using layer-by-layer assembly of thermoresponsive materials, Langmuir 20(1), 20- 22. Rathore, M. M., & Kapuno, R. (2011) Engineering Heat Transfer, 2nd Ed ed., Jones & Bartlett Learning, London, UK. Rousseau, E., Van der Auweraer, M., & De Schryver, F. C. (2000). Steady-state and time-resolved spectroscopy of a self-assembled cyanine dye multilayer, Langmuir 16(23), 8865-8870. Serizawa, T., Yamaguchi, M., & Akashi, M. (2002). Enzymatic hydrolysis of a layer-by- layer assembly prepared from chitosan and dextran sulfate, Macromolecules 35(23), 8656- 8658. Soedjak, H. S. (1994). Colorimetric determination of carrageenans and other anionic hydrocolloids with methylene-blue, Analytical Chemistry 66(24), 4514-4518. Tedeschi, C., Caruso, F., Mohwald, H., & Kirstein, S. (2000). Adsorption and desorption behavior of an anionic pyrene chromophore in sequentially deposited polyelectrolyte-dye thin films, Journal of the American Chemical Society 122(24), 5841-5848. Vojdani, F., & Torres, J. A. (1990). Potassium sorbate permeability of methylcellulose and hydroxypropyl methylcellulose coatings - Effect of fatty acids, Journal of Food Science 55(3), 841-846.

114

Chapter 4. κ-carrageenan/chitosan nanolayered coating for controlled release of a model bioactive compound

Wang, X. Y., Hu, X., Daley, A., Rabotyagova, O., Cebe, P., & Kaplan, D. L. (2007). Nanolayer biomaterial coatings of silk fibroin for controlled release, Journal of Controlled Release 121, 190-199. Wood, K. C., Boedicker, J. Q., Lynn, D. M., & Hammond, P. T. (2005). Tunable drug release from hydrolytically degradable layer-by-layer thin films, Langmuir 21(4), 1603-1609. Xia, W., Liu, P., Zhang, J., & Chen, J. (2011). Biological activities of chitosan and chitooligosaccharides, Food Hydrocolloids 25(2), 170-179. Yoo, D., Shiratori, S. S., & Rubner, M. F. (1998). Controlling bilayer composition and surface wettability of sequentially adsorbed multilayers of weak polyelectrolytes, Macromolecules 31(13), 4309-4318. Zhong, Y., Whittington, C. F., Zhang, L., & Haynie, D. T. (2007). Controlled loading and release of a model drug from polypeptide multilayer nanofilms, Nanomedicine: Nanotechnology, Biology and Medicine 3(2), 154-160.

115

SECTION III NANOEMULSIONS

5 CHAPTER 5. DEVELOPMENT OF

CURCUMIN NANOEMULSION - BASED DELIVERY SYSTEMS DEVELOPMENT OF CURCUMIN NANOEMULSION-BASED DELIVERY SYSTEMS

5.1 Introduction...... 121

5.2 Materials and Methods ...... 122

5.3 Results and Discussion ...... 124

5.4 Conclusions...... 131

5.5 References ...... 133

Pinheiro, A. C. (2013)

120

Chapter 5. Development of curcumin nanoemulsion-based delivery systems

5.1 INTRODUCTION

Curcumin is a low molecular weight polyphenol extracted as a yellow pigment from the powdered rhizomes of turmeric (Curcuma longa) spice (Ahmed et al., 2012). It is widely used as food colorant and it has also been found to have a wide range of biological and pharmacological activities (e.g. antitumoral, antioxidant, antimicrobial and anti-inflammatory properties) and therefore it has attracted considerable attention in recent years (Duvoix et al., 2005). Its chemical structure presents two o-methoxy phenols attached symmetrically through α,β-unsaturated β-diketone linker, which also induces keto-enol tautomerism (Souguir et al., 2013) (Figure 5.1).

Figure 5.1 Molecular structure of curcumin (adapted from Duvoix et al., 2005)

However, the application of curcumin as a functional ingredient is limited due to its extremely low water solubility (11 ng.mg-1) and its low bioavailability, which means that its beneficial attributes may not be realized when ingested. Thus, there is a significant need to improve delivery and bioavailability of curcumin.

Encapsulation of bioactive lipophilic compounds using emulsions is a common approach to protect active ingredients against extreme conditions, to enhance their stability, and to increase their bioavailability. Emulsifiers such as proteins and small molecular weight surfactants are commonly used for stabilization of emulsion droplets against coalescence, creaming, sedimentation and flocculation (Tikekar et al., 2013). Nowadays, food industry has an increasing interest in the replacement of synthetic emulsifiers by natural ones, such as proteins (Huck-Iriart et al., 2011). For example, recent studies have shown that lactoferrin can be successfully used as an emulsifier to stabilize oil-in-water emulsions (Tokle and McClements, 2011). Lactoferrin is a globular glycoprotein derived from milk and other mammalian fluids that has been reported to have numerous biological activities such as antioxidant capacity, antimicrobial activity, cancer prevention, and modulation of immune responses, which make very desirable its incorporation into foods as a functional ingredient (Brisson et al., 2007). Lactoferrin has also an unusually high isoelectric point (pI > 8), which

121

Pinheiro, A. C. (2013)

means it tends to be cationic at pH values where most of the major dairy globular proteins are anionic (Steijns and van Hooijdonk, 2000).

Nanoemulsions (50-200 nm) are emerging as an attractive alternative to conventional emulsions (1-100 µm) to protect and deliver lipophilic functional compounds since they present better stability to particle aggregation and gravitational separation, high kinetic stability, improved optical clarity, and increased bioaccessibility (Salvia-Trujillo et al., 2013).

The specific objectives of this work were to investigate the impact of different emulsion-based formulations on size, morphology and ζ-potential of curcumin nanoemulsions and to evaluate their stability during 90 days. Oil-in-water nanoemulsions were prepared with different emulsifiers: Tween 20 (non-ionic), Sodium Dodecyl Sulphate (SDS, anionic), DodecylTrimethylAmmonium Bromide (DTAB, cationic) and lactoferrin (protein), in order to evaluate the impact of emulsifier type on the stability of curcumin nanoemulsions and to access the potential of lactoferrin to substitute synthetic emulsifiers.

5.2 MATERIALS AND METHODS

5.2.1 Materials

Corn oil (Fula, Portugal) was purchased from a local supermarket in Braga (Portugal) and was used without further purification. Curcumin, Tween 20 and sodium dodecyl sulphate (SDS) were purchased from Sigma-Aldrich (St. Louis, MO, USA), dodecyltrimethylammonium bromide (DTAB) was acquired from Acros Organics (Belgium) and lactoferrin was purchased from DMV International (Veghel, The Netherlands).

5.2.2 Emulsion preparation

Curcumin nanoemulsions were prepared by homogenizing 5 wt.% corn oil containing 0.1 wt.% of curcumin with 95 wt.% aqueous emulsifier solution (Tween 20, SDS, DTAB or lactoferrin at different concentrations), using an Ultra-Turrax homogenizer (T 25, Ika-Werke, Germany) for 2 min followed by passage through a high pressure homogenizer (NanoDeBee, Bee International, South Easton, Massachusetts, USA) at 20 000 Psi (137.9 MPa), for 20 cycles. The nanoemulsions were kept at 4 °C in the dark before characterization.

122

Chapter 5. Development of curcumin nanoemulsion-based delivery systems

5.2.3 Emulsion characterization

5.2.3.1 Particle size measurements

The particle size of the curcumin nanoemulsions was determined by dynamic light scattering (DLS) (Zetasizer Nano ZS, Malvern Instruments, UK). Each sample was analysed in a folded capillary cell. The measurements were made in triplicate, with three readings for each of them. The results are given as the average ± standard deviation of the nine values obtained.

5.2.3.2 ζ-potential measurements

The ζ-potential of the different curcumin nanoemulsions was determined by dynamic light scattering (DLS; Zetasizer Nano ZS, Malvern Instruments, UK). Each sample was analysed in a folded capillary cell. The results are given as the average ± standard deviation of the nine values obtained.

5.2.3.3 Transmission Electron Microscopy (TEM)

TEM micrographs were conducted on a Zeiss EM 902A microscope (Germany) at an accelerating voltage of 50 kV and 80 kV. The samples were prepared by dropping nanoemulsion solutions onto copper grids coated with carbon film, followed by staining with uranyl acetate and natural drying.

5.2.3.4 Colour evolution

Colour parameters were determined using a tristimulus colorimeter (Minolta Chroma Meter CR400, Japan), programmed to use illuminant C as light source and the 2° observer for colour interpretation. A white tile (Minolta calibration plate) with following standard value: Y = 93.9, x = 0.3133, y = 0.3193, was used to calibrate the equipment. Results were calculated by the equipment into the Hunter Lab colour scale. In this scale, L ranges from 0 (black) to 100 (white), a indicates degree of greenness (for negative a values) and degree of redness (for positive a results). Axis b also ranges from negative to positive values indicating, respectively, degree of blueness to yellowness. Colour changes were assessed using TCD (Silva et al., 2011):

√( ) ( ) ( ) Equation 5.1

where L*, a* and b* are the colour parameters at the end of the period under analysis and L0*, a0* and b0* are the colour parameters at the beginning of that period.

123

Pinheiro, A. C. (2013)

5.2.3.5 Statistical Analyses

The statistical analyses were carried out using analysis of variance, Tukey mean comparison test (p<0.05) and linear regression analysis (SigmaStat, trial version, 2003, USA).

5.3 RESULTS AND DISCUSSION

5.3.1 Development of the curcumin nanoemulsions

A wide range of emulsifiers are available for use in the food technology, including small- molecule surfactants, phospholipids, proteins and polysaccharides (Hur et al., 2009). It is therefore useful to establish the impact of emulsifier type on nanoemulsion formation. In this work, the influence of four different emulsifiers on the droplet size and charge was evaluated. The emulsifiers examined were: a non-ionic surfactant (Tween 20), an anionic surfactant (SDS), a cationic surfactant (DTAB) and a protein (lactoferrin).

5.3.1.1 Effect of emulsifier concentration on particle size

Emulsions with an oil/emulsifier aqueous solution ratio of 0.5/9.5 were prepared using corn oil and different emulsifiers with different concentrations (0.5–3 wt%) and their particle sizes were measured (Figure 5.2). a) b)

Figure 5.2 Effect of emulsifier concentration on emulsions size (a) and polydispersity index (PDI) (b). Tween 20 (), SDS (), DTAB () and lactoferrin () -stabilized nanoemulsions. Results obtained for an oil/ emulsifier aqueous solution ratio of 0.5/9.5. Bars indicate standard deviation.

These results reflect the ability of all emulsifiers to produce small droplets (diameter lower than 200 nm, i.e. nanoemulsions) at concentrations between 0.5 and 3%. However, the type and concentration of emulsifier used to stabilize the curcumin emulsions influenced the size of the droplets obtained. In general, there was a decrease in mean droplet diameter with increasing emulsifier concentration, which might be expected because there is more emulsifier

124

Chapter 5. Development of curcumin nanoemulsion-based delivery systems present to cover any new droplet surfaces formed during homogenization, and the droplet surfaces will be more rapidly covered by a layer of emulsifier.

Although the differences are not statistically significant, these results are in agreement with previous works that showed that small-molecule surfactants (such as Tween 20, SDS and DTAB) are more effective at producing small droplets (at least at low concentrations) than biopolymers (such as lactoferrin) under similar homogenization conditions because they adsorb to the droplet surfaces more rapidly (Qian and McClements, 2011).

The monodipersity of the nanoemulsions can be evaluated by its polydispersity index (PDI), which can range from 0 to 1 (with 0 being monodisperse and 1 being polydisperse). All the emulsions produced exhibit PDI values lower than 0.5 (Figure 5.1 b), which means that they are relatively monodisperse.

In general, due to economic, sensory, or regulatory reasons, it is often desirable to minimize the concentration of the emulsifier used to prepare an emulsion (Qian and McClements, 2011). For example, some small-molecule surfactants generate off-flavours when used at high concentrations and there is often a concentration limitation for their use in food products. Therefore, for the small-molecule surfactants the concentration chosen for the following tests was the minimum concentration tested: 0.5%, once although the size usually decreases with increasing emulsifier concentration, in most cases this difference is not statistically significant and also the PDI increases. For the lactoferrin-stabilized emulsions the concentration selected was 2% that was the concentration for which the best compromise between size and PDI was obtained.

5.3.1.2 Effect of oil/emulsifier solution ratio on particle size

The impact of the oil/emulsifier solution ratio was also evaluated (Figure 5.3), using the emulsifier concentrations chosen above. a) b)

Figure 5.3 Effect of oil/emulsifier solution ratio on size (a) and polydispersity index (PDI) (b). Tween 20 (), SDS (), DTAB () and lactoferrin () -stabilized nanoemulsions. Bars indicate standard deviation.

125

Pinheiro, A. C. (2013)

The oil/emulsifier solution ratio had a major impact on the size and PDI of the droplets produced. Results show that for all systems stabilized by the different emulsifiers, the ratio at which the smallest sizes and PDI values are obtained is 5/95. At higher oil/emulsifier solution ratios, there is not enough emulsifier present to cover the new droplets formed during homogenization and their agglomeration occurs.

5.3.1.3 Particle charge

Figure 5.4 shows the particle charge of curcumin nanoemulsions as a function of the emulsifier used.

Figure 5.4 ζ-potential of nanoemulsions stabilized by Tween 20, SDS, DTAB (concentration of 0.5%, oil/ emulsifier aqueous solution of 0.5/9.5) and lactoferrin (concentration of 2%, oil/ emulsifier aqueous solution ratio of 0.5/9.5).

As expected, SDS-stabilized emulsions had a highly negative surface charge whereas DTAB- stabilized emulsions were positively charged. Tween 20 and lactoferrin-stabilized emulsions presented neutral charge. It should be pointed out that these ζ-potential values were obtained at neutral pH and, contrary to what happened in the case of Tween 20 nanoemulsions, the charge of the nanoemulsions stabilized by lactoferrin varied with pH (results not shown), once this emulsifier is a protein. Lactoferrin has an isoelectric point (pI) between 8.0 and 9.0 (Brisson et al., 2007), which means that it carries a positive charge at pH values below its pI and a negative charge at pH values above its pI. Lactoferrin-stabilized nanoemulsions have a maximum charge at pH 4 (ζ-potential=27.47 ± 1.11 mV). The use of proteins such as lactoferrin as emulsifiers to stabilize emulsions allows the possibility of manipulating the charge of the emulsion by simply changing pH, which could be useful for example to form multilayered emulsions (Tokle et al., 2012).

126

Chapter 5. Development of curcumin nanoemulsion-based delivery systems

5.3.1.4 Morphology

The morphology of the nanoemulsions stabilized by the different emulsifiers was observed by TEM (Figure 5.5).

All curcumin nanoemulsions present a spherical morphology, however some differences can be found depending on the emulsifier used. In the TEM micrograph of the Tween 20-stabilized emulsions clustering between the lipid droplets is clearly visible (Figure 5.5), which contributes to the relatively high PDI values observed for this nanoemulsion (Figure 5.2 b). This is a direct consequence of the neutral charge of Tween 20, once in this case attractive forces between droplets dominate repulsive forces. This phenomenon did not occur in the SDS or DTAB- stabilized nanoemulsions due to the strong repulsive interactions existent between charged droplets (Figure 5.5 b) and c)). As mentioned before, the nanoemulsion stabilized by lactoferrin exhibits neutral charge at pH 7, which explains the aggregation observed in Figure 5.5 c).

a) b)

c) d)

Figure 5.5 TEM micrographs of Tween 20 (a), SDS (b), DTAB (c) and lactoferrin-stabilized nanoemulsions (d).

5.3.2 Storage stability of curcumin nanoemulsions

The stability of an emulsion consists in its ability to resist changes in its properties over time. Physical instability results in an alteration in the spatial distribution or structural organization of molecules and may occur due to creaming, flocculation, coalescence, partial coalescence,

127

Pinheiro, A. C. (2013)

phase inversion or Ostwald ripening (Huck-Iriart et al., 2011). The physical stability of all nanoemulsions was evaluated over 90 days of storage.

5.3.2.1 Particle size

The changes in droplet size and polydispersity index as a function of the storage time for nanoemulsions stabilized with different emulsifiers are shown in Figure 5.6.

a) b)

Figure 5.6 Evolution of Z-average diameter (a) and of polydispersity index (PDI) (b) of the nanoemulsions during 90 days of storage at 4 °C in the dark: Tween 20 (), SDS (), DTAB () and lactoferrin () -stabilized nanoemulsions. Bars indicate standard deviation.

From Figure 5.6 a) it is possible to observe that SDS and lactoferrin-stabilized nanoemulsions maintained their initial size (p>0.05) over 90 days of storage. In the case of Tween 20 nanoemulsion, the z-diameter increased from 155.4 ± 1.6 nm at the beginning of storage to 189.9 ± 2.5 nm at the end of 90 days. The increase of droplet size with time is due to the movement of dispersed droplet through the continuous phase that increases the opportunities for droplets’ collisions (Li and Chiang, 2012). Particularly, Tween 20 is a neutral surfactant and therefore there is no charge repulsion between the droplets and this emulsifier is inefficient to keep the emulsions droplets separated from each other (as shown in Figure 5.5 a). Ostwald ripening was found to be the main mechanism producing instability of nanoemulsions (Liu et al., 2006). This mechanism occurs due to the difference in radius between droplets, and the driving force is the difference in chemical potential of the oil phase between different droplets (Li and Chiang, 2012).

However, in the case of DTAB-stabilized emulsions, the z-average diameter actually decrease (p<0.05) with time (from 157.9 ± 1.6 to 133.6 ± 0.4 nm). However, this result must be carefully analysed, because after observation of this sample (Figure 5.7 b), a phase separation is clearly visible (insert of Figure 5.7 b), which means that, contrary to what DLS results indicate, this sample is not stable. The misleading DLS results may be explained by the fact that the larger coalesced droplets rapidly creamed to the top of the Zetasizer cell and thus escaped to

128

Chapter 5. Development of curcumin nanoemulsion-based delivery systems detection by the laser; the detection beam thus passed through an unrepresentative fraction of emulsions containing relatively small droplets (Mun et al., 2007). Other authors evaluated the stability of emulsions stabilized by different surfactants (cationic, non-ionic and small polymers) and observed that the most unstable emulsion was the one stabilized with DTAB that has a lifetime of about 6 days (Georgieva et al., 2009). These results were correlated with the surface elastic moduli, both at high and low frequencies, once the DTAB presented lowest modulus (Georgieva et al., 2009).

a) b)

Figure 5.7 Visual appearance of the curcumin nanoemulsions (a) at day 0 and (b) after 90 days of storage in the dark at 4 °C. The insert figure represents a detail of the DTAB-stabilized nanoemulsion.

Regarding the evolution of the polydispersity index, there were no significant changes (p>0.05) of this parameter for all nanoemulsions over time (Figure 5.6 b). At the end of storage time, nanoemulsions stabilized by SDS, DTAB and lactoferrin continue to exhibit a narrow size distribution (PDI = 0.21 ± 0.01, 0.25 ± 0.01 and 0.27 ± 0.01 respectively), and Tween 20- stabilised emulsions, although being not so monodisperse systems, maintained PDI values (PDI = 0.41 ± 0.01).

5.3.2.2 Particle charge

The possibility that electrical charge effects may be of significance in stabilizing emulsions has been early recognized (Sperandio, 1965). Changes in particle charge over time were therefore evaluated and are represented in Figure 5.8.

129

Pinheiro, A. C. (2013)

Figure 5.8 Changes of ζ-potential of nanoemulsions stabilized by Tween 20 (), SDS (), DTAB () and lactoferrin () as a function of time. Bars indicate standard deviation.

Over the 90 days of storage, Tween 20 and lactoferrin-stabilized nanoemulsions maintained their neutral charge, whereas nanoemulsions with SDS maintained their negative charge. On the contrary, the ζ-potential of DTAB-stabilized emulsions significantly decreased (p<0.05) from 50.67 ± 5.08 mV (day 0) to 34.20 ± 2.39 mV (day 90). The decrease of the magnitude of ζ- potential (i.e. its absolute value) is associated with the decrease in emulsion stability (Rambhau et al., 1977).

5.3.2.2 Colour evolution

The evolution of the total colour difference of all nanoemulsions over time can be observed in Figure 5.9.

Figure 5.9 Evolution of total colour difference (TCD) of nanoemulsions stabilized by Tween 20 (), SDS (), DTAB () and lactoferrin (), during 90 days of storage. Bars indicate standard deviation.

130

Chapter 5. Development of curcumin nanoemulsion-based delivery systems

During 90 days of storage, the total colour difference remained constant (p>0.05) for Tween 20, SDS and lactoferrin-stabilised nanoemulsions. On the other hand, for DTAB-stabilized nanoemulsions, the TCD significantly increased (p<0.05) in a linear way. Other authors evaluated the physical stability of β-carotene nanoemulsions during 21 days of storage and observed that the samples with the highest increase of TCD with time were the samples that were more affected during storage in terms of size distribution and β-carotene retention (Silva et al., 2011). Although in the size results (Figure 5.6) this is not clear due the reasons explained above, it was observed that DTAB-stabilized nanoemulsions were highly unstable with time once a phase separation occurred during the period of storage, which is in agreement with the results obtained by Silva et al. (2011). The greatest difference in the colour of DTAB-stabilized nanoemulsions over time is also clearly visible comparing the images of DTAB nanoemulsions in the beginning of storage time (t=0 days) and at the end (t=90 days) (Figure 5.6).

Other authors (Mei et al., 1998) observed, on the contrary, that the surface charge of emulsion droplets plays an important role in their oxidative stability and that the utilization of positively charged emulsifiers, such as DTAB could be an effective way to control iron-catalysed lipid oxidation. These authors showed that cationic droplets are more stable to lipid oxidation than anionic droplets because they do not attract positively charged transition metal ions that catalyse oxidation to the droplet surfaces (Mei et al., 1998).

5.4 CONCLUSIONS

The impact of different emulsion-based formulations on size, morphology and ζ-potential of curcumin nanoemulsions was evaluated. Tween 20 (non-ionic), Sodium Dodecyl Sulphate (SDS, anionic), DodecylTrimethylAmmonium Bromide (DTAB, cationic) and lactoferrin (protein) were the emulsifiers used to prepare curcumin nanoemulsions. The type, concentration of emulsifier and oil/emulsifier solution ratio used to prepare curcumin nanoemulsions influenced the size of the droplets obtained. For Tween 20, SDS and DTAB-stabilized emulsions, the concentration chosen was 0.5% and for lactoferrin-stabilized emulsions the concentration selected was 2%. For all systems, the oil/emulsifier solution ratio at which the smallest sizes and PDI values were obtained was 5/95.

After optimizing the emulsifier concentration and the oil/emulsifier solution for each system, stability of curcumin nanoemulsions stabilized by different emulsifiers was evaluated during 90 days. DTAB-stabilized emulsions revealed to be the least stable, showing a phase separation, the highest colour change and the highest decrease in the magnitude of ζ-potential. Nanoemulsions stabilized by lactoferrin revealed to be at least as stable as those stabilized by

131

Pinheiro, A. C. (2013)

synthetic emulsifiers, during 90 days of storage, which shows the great potential of proteins, in particular of lactoferrin, to substitute synthetic emulsifiers. This is the current trend of food industry.

There is still a lack of knowledge regarding the biological fate of ingested nanoemulsions and further research is needed to access their safety and to produce tailored delivery systems (i.e. with optimized bioactivity).

132

Chapter 5. Development of curcumin nanoemulsion-based delivery systems

5.5 REFERENCES

Ahmed, K., Li, Y., McClements, D. J., & Xiao, H. (2012). Nanoemulsion- and emulsion- based delivery systems for curcumin: Encapsulation and release properties, Food Chemistry 132(2), 799-807. Brisson, G., Britten, M., & Pouliot, Y. (2007). Heat-induced aggregation of bovine lactoferrin at neutral pH: Effect of iron saturation, International Dairy Journal 17(6), 617-624. Duvoix, A., Blasius, R., Delhalle, S., Schnekenburger, M., Morceau, F., Henry, E., Dicato, M., & Diederich, M. (2005). Chemopreventive and therapeutic effects of curcumin, Cancer Letters 223(2), 181-190. Georgieva, D., Schmitt, V. r., Leal-Calderon, F., & Langevin, D. (2009). On the Possible Role of Surface Elasticity in Emulsion Stability, Langmuir 25(10), 5565-5573. Huck-Iriart, C., Álvarez-Cerimedo, M. S., Candal, R. J., & Herrera, M. L. (2011). Structures and stability of lipid emulsions formulated with sodium caseinate, Current Opinion in Colloid & Interface Science 16(5), 412-420. Hur, S. J., Decker, E. A., & McClements, D. J. (2009). Influence of initial emulsifier type on microstructural changes occurring in emulsified lipids during in vitro digestion, Food Chemistry 114(1), 253-262. Li, P.-H., & Chiang, B.-H. (2012). Process optimization and stability of d-limonene-in- water nanoemulsions prepared by ultrasonic emulsification using response surface methodology, Ultrasonics Sonochemistry 19(1), 192-197. Liu, W., Sun, D., Li, C., Liu, Q., & Xu, J. (2006). Formation and stability of paraffin oil-in- water nano-emulsions prepared by the emulsion inversion point method, Journal of Colloid and Interface Science 303(2), 557-563. Mei, L., McClements, D. J., Wu, J., & Decker, E. A. (1998). Iron-catalyzed lipid oxidation in emulsion as affected by surfactant, pH and NaCl, Food Chemistry 61(3), 307-312. Mun, S., Decker, E. A., & McClements, D. J. (2007). Influence of emulsifier type on in vitro digestibility of lipid droplets by pancreatic lipase, Food Research International 40(6), 770- 781. Qian, C., & McClements, D. J. (2011). Formation of nanoemulsions stabilized by model food-grade emulsifiers using high-pressure homogenization: Factors affecting particle size, Food Hydrocolloids 25(5), 1000-1008. Rambhau, D., Phadke, D. S., & Dorle, A. K. (1977). Evaluation of O/W emulsion stability through zeta potential-I Prediction of O/W emulsions stability through zeta potential by accelerated ageing tests, Journal of the Society of Cosmetic Chemists 28, 183-196.

133

Pinheiro, A. C. (2013)

Salvia-Trujillo, L., Qian, C., Martín-Belloso, O., & McClements, D. J. (2013). Influence of particle size on lipid digestion and β-carotene bioaccessibility in emulsions and nanoemulsions, Food Chemistry 141(2), 1472-1480. Silva, H. D., Cerqueira, M. A., Souza, B. W. S., Ribeiro, C., Avides, M. C., Quintas, M. A. C., Coimbra, J. S. R., Carneiro-da-Cunha, M. G., & Vicente, A. A. (2011). Nanoemulsions of β- carotene using a high-energy emulsification–evaporation technique, Journal of Food Engineering 102(2), 130-135. Souguir, H., Salaün, F., Douillet, P., Vroman, I., & Chatterjee, S. (2013). Nanoencapsulation of curcumin in polyurethane and polyurea shells by an emulsion diffusion method, Chemical Engineering Journal 221(0), 133-145. Sperandio, G. J. (1965). Emulsions: Theory and practice. 2nd Ed., Acs Monograph No. 162. By Paul Becher. Reinhold Publishing Corp., 430 Park Ave., New York, N. Y., 1965. xi + 440 pp. Price $22, Journal of Pharmaceutical Sciences 54(8), 1227-1227. Steijns, J. M., & van Hooijdonk, A. C. M. (2000). Occurrence, structure, biochemical properties and technological characteristics of lactoferrin, British Journal of Nutrition 84(SupplementS1), 11-17. Tikekar, R. V., Pan, Y., & Nitin, N. (2013). Fate of curcumin encapsulated in silica nanoparticle stabilized Pickering emulsion during storage and simulated digestion, Food Research International 51(1), 370-377. Tokle, T., Lesmes, U., Decker, E. A., & McClements, D. J. (2012). Impact of dietary fiber coatings on behavior of protein-stabilized lipid droplets under simulated gastrointestinal conditions, Food & Function 3(1), 58-66. Tokle, T., & McClements, D. J. (2011). Physicochemical properties of lactoferrin stabilized oil-in-water emulsions: Effects of pH, salt and heating, Food Hydrocolloids 25(5), 976- 982.

134

6 CHAPTER 6. U NRAVELLING THE

BEHAVIOUR OF CURCUMIN NANOEMULSIONS DURING IN VITRO DIGESTION: IMPACT OF SURFACE CHARGE UNRAVELLING THE BEHAVIOUR OF CURCUMIN NANOEMULSIONS DURING IN VITRO DIGESTION: IMPACT OF SURFACE CHARGE

6.1 Introduction...... 137

6.2 Materials and methods ...... 139

6.3 Results and Discussion ...... 143

6.4 Conclusions...... 152

6.5 References ...... 153

Pinheiro, A. C. (2013)

The results presented in this Chapter were adapted from:

136

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

6.1 INTRODUCTION

Curcumin is a yellow pigment derived from the rhizomes of Curcuma longa (commonly known as turmeric). This natural polyphenol is an important natural colorant used in food (Wang et al., 2012) and has gained considerable attention in recent years due to its wide range of biological activities, such as anti-oxidant (Sharma, 1976), anti-microbial (Wang et al., 2009), anti-inflammatory (Srimal and Dhawan, 1973), wound healing (Sidhu et al., 1998) and antitumoral effects (Kuttan et al., 1985). However, the applicability of curcumin as health- promoting agent is limited by its poor solubility in aqueous solutions and low bioavailability. The reduced bioavailability of curcumin is related to its poor absorption, rapid metabolism and rapid systemic elimination (Anand et al., 2007). The bioavailability of poorly water-soluble compounds such as curcumin can be greatly enhanced by incorporating them into emulsion- based delivery systems. An oil-in-water emulsion can be produced by homogenizing the oil phase containing the lipophilic bioactive compound with an aqueous phase containing a water- soluble emulsifier (Ahmed et al., 2012). Emulsions can have a wide range of droplet sizes, depending on the composition of the system and on the method of homogenization used. Emulsions formed at the nanometre scale have better stability against droplet aggregation once attractive forces acting between droplets decreases with decreasing particle size. The bioavailability of encapsulated lipophilic components within the GI tract may be, in principle, increased by using nanoemulsions due to their small particle size and high surface-to-volume ratio (Silva et al., 2012).

In humans, hydrolysis of lipids starts in the stomach (pH 1 to 3) by the action of gastric lipase, which hydrolyses about 10 to 30% of the ingested triglycerides, generating mainly free fatty acids and diacylglycerides. This facilitates subsequent lipid hydrolysis in the duodenum by pancreatic lipase by allowing fat emulsification and promoting enzyme activity (Pafumi et al., 2002). Encapsulated lipophilic bioactive compounds are released upon digestion of the emulsion and are incorporated into bile salt/phospholipid micelles (mixed micelles), being subsequently transported through the mucous layer to the surfaces of the enterocytes where it is absorbed (McClements and Xiao, 2012). The formation of mixed micelles facilitates molecular absorption of lipophilic compounds from the small intestine into the bloodstream (Rozner et al., 2010).

In recent years, different in vitro digestion models have been developed as a means for understanding the physicochemical processes associated with the digestion process. The pH- stat method is a simple in vitro lipolysis model commonly used to characterize lipid digestion under simulated small intestinal conditions. This method allows measuring the free fatty acids

137

Pinheiro, A. C. (2013)

(FFA) released from triglycerides over time and can be used to rapidly evaluate the impact of different formulations or conditions on lipid digestion (Li and McClements, 2010). Some of the complex dynamic gastrointestinal models include the dynamic gastric model (DGM), that was designed to replicate real-time changes in pH, enzyme addition, shearing, mixing and retention time of an adult human stomach (Vardakou et al., 2011), the TIM model, which simulates the physiological conditions of human stomach and small intestine (Reis et al., 2008), and the Human Gastric Simulator (HGS), a model that allows simulating the continuous peristaltic movements of stomach walls with similar amplitude and frequency of contraction forces as reported in vivo (Kong and Singh, 2010). This model also incorporates gastric secretion, emptying systems and temperature control that enable a more realistic simulation of the dynamics of the digestion process.

Most of the works that evaluate the behaviour of emulsions during digestion use simple in vitro models that do not include a gastric stage (Lesmes et al., 2010; Li et al., 2011; Troncoso et al., 2012). The passage of nanoemulsions through the highly acidic conditions on the stomach involves physical destabilization of emulsions, once the composition, size and interfacial characteristics of lipid droplets is altered (Sarkar et al., 2009). Therefore, more realistic in vitro digestion models should be used.

The emulsifier type will impact the susceptibility of lipid droplets to coalescence and breakup within the GI tract, thereby altering the total surface area of lipid exposed to lipase absorption and activity (Hur et al., 2009). The characteristics of the interfacial layer surrounding lipid droplets will influence the ability of lipase to interact with the lipid phase, and consequently, the rate of lipid digestion.

In the present work, the Human Gastric Simulator model was used to mimic mechanical and chemical environments of the human stomach, duodenum, jejunum and ileum. The impact of emulsifier type on structural changes of curcumin nanoemulsions and on curcumin bioavailability during in vitro digestion was evaluated. In particular, the impact of emulsifier charge was evaluated using Tween 20, SDS and DTAB as emulsifiers. Also, the possibility of a multilayer coating formation was evaluated by adding a chitosan layer to the SDS-stabilized emulsions.

138

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

6.2 MATERIALS AND METHODS

6.2.1 Materials

Corn oil was purchased from a local supermarket in Palmerston North (New Zealand) and was used without further purification. Curcumin, Tween 20 and sodium dodecyl sulphate (SDS) were purchased from Sigma-Aldrich (St. Louis, MO) and dodecyltrimethylammonium bromide (DTAB) was acquired from Acros Organics (Geel, Belgium). Pepsin from porcine gastric mucosa, amano lipase A from Aspergillus niger, pancreatin from porcine pancreas (8xUSP), bile extract porcine and the salts used for preparing the gastric electrolyte solution and the small intestinal electrolyte solution, hydrochloric acid and sodium bicarbonate were purchased from Sigma- Aldrich (St. Louis, MO). Chloroform was obtained from BDH chemicals (Poole, England).

Milli-Q water (water purified by treatment with Milli-Q apparatus, Millipore Corp., Bedford, MA, USA) was used to prepare all solutions.

6.2.2 Emulsion preparation

Curcumin nanoemulsions were prepared by homogenizing 5 wt.% corn oil containing 0.1 wt.% of curcumin with 95 wt.% aqueous emulsifier solution (0.5 wt.% Tween 20, SDS or DTAB), using a high-speed blender for 2 min followed by passage through a microfluidizer (Microfluidics M- 110P) at 20 000 Psi (137.9 MPa), for 5 cycles.

6.2.3 In vitro digestion

In vitro digestion was performed simulating the gastric, duodenal, jejunal and ileal portions of the GI tract as described by other authors (Reis et al., 2008) with the following modifications. A volume of 300 mL of curcumin nanoemulsions with the selected emulsifier was introduced into the human gastric simulator (Figure 6.1) and the experiment was run for a total of 5 h, simulating average physiological conditions of GI tract by the continuous additions of gastric, duodenal, jejunal and ileal secretions.

The gastric secretion consisted of pepsin and lipase in a gastric electrolyte solution (NaCl 4.8

-1 -1 -1 -1 -1 g.L , KCl 2.2 g.L , CaCl2 0.22 g.L and NaHCO3 1.5 g.L ), secreted at a flow rate of 0.5 mL.min . The pH was controlled to follow a predetermined curve (from 4.8 at t=0 to 1.7 at t=120 min) by secreting hydrochloric acid (1 mol.L-1). The duodenal secretion consisted of a mixture of 30 mL of 4% (w/v) porcine bile extract, 15 mL of 7% (w/v) pancreatin solution and 9 mL of a small

-1 -1 -1 intestinal electrolyte solution (SIES) (NaCl 5 g.L , KCl 0.6 g.L , CaCl2 0.25 g.L ) at a flow rate of 0.9 mL.min-1. The jejunal secretion fluid consisted of SIES containing 10% (v/v) porcine bile

139

Pinheiro, A. C. (2013)

extract solution at a flow rate of 3.2 mL.min-1. The ileal secretion fluid consisted of SIES at a flow rate of 3.0 mL.min-1. The pH in the different parts of small intestine was controlled by the addition of 1 mol.L-1 sodium bicarbonate solutions to set-points of 6.5, 6.8 and 7.2 for simulated duodenum, jejunum and ileum, respectively.

Figure 6.1 Human gastric simulator used in the in vitro digestion experiments: (A) latex lining, (B) secretion tubing, (C) roller, (D) belt, (E) peristaltic pump, (F) temperature control, (G) motor

6.2.4 Particle size

The particle size of emulsions was measured at different stages of digestion by both dynamic light scattering (Zetasizer Nano ZS, Malvern Instruments, Worcestershire, UK) and static light scattering (Mastersizer 2000 Hydro MU, Malvern Instruments Ltd, Malvern, Worcestershire, UK).

The Z-average mean diameter was calculated by the Zetasizer instrument using Stokes-Einstein equation, assuming that nanoemulsion droplets were spherical. Before measurement, samples were diluted at a ratio of 1:100 (v/v) in an appropriate buffer solution with the same pH as the aqueous phase of the sample (citrate buffer (10 mmol.L-1, pH 2) for stomach step and phosphate buffer (10 mmol.L-1, pH 7) for before digestion and small intestine steps (Hur et al., 2009)) at room temperature to avoid multiple scattering effects. The measurements were performed at least in triplicate.

Particle size distribution was also measured by the Mastersizer instrument, which measures the angular dependence of the intensity of laser light scattered by a dilute emulsion and then finds the particle size distribution that gives the best fit between experimental measurements and predictions made using light-scattering theory (Mie theory). To avoid multiple scattering effects the emulsions were diluted in distilled water at a ratio of 1:100 (v/v).

140

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

Data fitting – Particle size distribution (d4,3 = (∑ nidi4 / ∑ nidi3) where ni is the number of droplets of diameter di) of emulsions was recorded before and during simulated digestion using refractive indices of 1.47 and 1.33 for droplet and dispersed (water) phase respectively.

6.2.5 ζ-potential

The ζ-potential values of nanoemulsions at different stages of digestion were measured using a particle electrophoresis instrument (Malvern Instruments, Worcestershire, UK). The ζ-potential was determined by measuring the direction and velocity of droplet movement in the applied electric field and was calculated by the instrument using the Smoluchowski model.

Samples were diluted at a ratio of 1:100 (v/v) in appropriate buffer solution (citrate buffer (10 mmol.L-1, pH 2) for stomach step and phosphate buffer (10 mmol.L-1, pH 7) for before digestion and small intestine steps (Hur et al., 2009)) at room temperature before measurement and each individual ζ-potential measurement was determined from the average of three readings.

6.2.6 Confocal microscopy

The microstructure of oil droplets in the emulsion was studied using a confocal scanning laser microscope (Leica TCS SP5 DM6000B, Heidelberg, Germany) with an x100 oil immersion objective lens. Samples were stained with Nile Red (9-diethylamino-5H-benzo[α]phenoxazine- 5-one, 0.25 mg.mL-1 in dimethyl sulfoxide, 1:10 (dye:sample), v/v), which enabled the oil droplets to be visible. Slides were prepared by taking a portion of the stained emulsion solution, placing it in a concave glass microscope slide and covering with a glass cover slip.

6.2.7 Free fatty acids released

A measure of lipolysis is the production of free fatty acids (FFA) with increased digestion time, which leads to a decrease in the pH of the system. A pH-stat automatic titration unit (TIM 856 titration manager, Titralab, Radiometer analytical, France) was used to automatically monitor pH and maintain it at a pre-set value by titrating with 0.05 mol.L-1 NaOH solution. The volume of NaOH added to the emulsion was recorded, and used to calculate the concentration of FFA generated by lipolysis.

A sample of emulsion was collected from the HGS after 120 min (i.e. after stomach simulation) and diluted in an appropriate quantity of small intestinal electrolyte solution (SIES), to obtain the same emulsion concentration as in the HGS at the end of digestion. Pancreatin and bile extract were added and a back titration was used to measure total free fatty acids. After

141

Pinheiro, A. C. (2013)

incubation of emulsions at pH 7.0 for a certain time, pH was rapidly increased with NaOH to pH 9.0 to stop the reaction and to favour the release of FFA. Therefore, total FFA were assessed by titration, measuring the volume necessary to reach pH 9.0 (Helbig et al., 2012). Back titration was performed at pH 9.0 in order to ensure full ionization and titration of free fatty acids (Amara et al., 2009).

Blank experiments (without pancreatin) were performed in order to determine the volume of NaOH required to raise pH to 9.0. This was required for the calculation of total free fatty acids liberated according to Equation 4.1. The quantity of fatty acids liberated was calculated based on the following equation (Pinsirodom and Parkin, 2001):

( ) Equation 6.1

where C is the molar concentration of NaOH titrant used (0.05 mol.L-1 in this case).

6.2.8 Curcumin bioavailability

It was assumed that the fraction of original curcumin that ended up in the micelle phase was a measure of curcumin bioavailability (Ahmed et al., 2012). Therefore, curcumin bioavailability at different stages of digestion was determined based on the methodology described by other authors (Ahmed et al., 2012). Briefly, 10 mL of emulsions were centrifuged (Sorvall evolution RC. Thermo Scientific, USA) (18675 g) at room temperature for 30 min. The micelle phase (5 mL) was collected, vortexed with 5 mL of chloroform, and then centrifuged at 651 g, at room temperature, for 10 min. The bottom chloroform layer was collected and the extraction procedure was repeated with the top layer. The second bottom chloroform layer was added to the previously set aside chloroform layer, mixed, and analysed with a UV–VIS spectrophotometer (Genesys 10uV, Thermo Scientific, USA) at 422 nm (absorbance peak). The concentration of curcumin was determined from a previously prepared calibration curve of absorbance versus curcumin concentration in chloroform.

6.2.9 Preparation of multilayer emulsions

Multilayer emulsions were prepared by adding the SDS-stabilized nanoemulsion (pH 4, at which lactoferrin has the maximum charge) dropwise to a chitosan solution (0.02%, pH 3, at which chitosan has the maximum charge) under stirring for 30 min. The multilayer emulsion was then treated with sonication (during 2 min at a frequency of 50-60 kHz) (Brason 5510, Bransonic® ultrasonic cleaner, USA) to disrupt any flocculated droplets formed during its preparation (Li et al., 2010).

142

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

6.2.10 Statistical analyses

All experiments were carried out in triplicate using freshly prepared samples. Results were then reported as averages and standard deviations of these measurements. Statistical analyses were carried out using analysis of variance, Tukey mean comparison test (p<0.05) and linear regression analysis (SigmaStat, trial version, 2003, USA).

6.3 RESULTS AND DISCUSSION

6.3.1 Influence of digestion on particle size

Initially, the change in particle size distribution of nanoemulsions as a function of digestion time was characterized. Figure 6.2 shows the mean hydrodynamic diameter of curcumin nanoemulsions at different stages of the simulated digestion process namely initial (before digestion), stomach, duodenum, jejunum and ileum, determined using a dynamic light scattering methodology.

All emulsifiers tested formed stable nanoemulsions with z-average diameter of 124.2 ± 8.4, 152.9 ± 2.2 and 119.5 ± 4.5 nm, for Tween 20, SDS and DTAB stabilized nanoemulsions, respectively. The monodipersity of nanoemulsions can be evaluated by their polydispersity index (PDI), which can range from 0 to 1 (with 0 being monodisperse and 1 being polydisperse). Nanoemulsions stabilized by SDS and DTAB exhibited a narrow size distribution (PDI = 0.19 ± 0.02 and 0.22 ± 0.01, respectively), whereas Tween 20 stabilised emulsions present a lower monodipersity (PDI=0.42 ± 0.03).

Figure 6.2 Dynamic light scattering-based mean particle diameter of curcumin nanoemulsions stabilized by different emulsifiers: Tween 20 (), SDS () and DTAB (), as they pass through an in vitro digestion model.

During the simulated digestion, all nanoemulsions showed an increase in size, although the precise mechanism was not identified. The increase in droplet size observed could be attributed to aggregation, coalescence or flocculation due to the action of digestive enzymes

143

Pinheiro, A. C. (2013)

as well as changes in pH and ionic strength. From the data presented it would appear that emulsions stabilized with DTAB were the least stable, showing the greatest increase in particle diameter (p<0.05).

It should be noted that dynamic light scattering has a size limitation: it is suitable for the measurement of particle diameters up to few microns (≤ 10 µm). Therefore, in order to detect the extent to which emulsion drop diameter had increased, emulsions were also analysed using static light scattering (Figure 6.3).

Figure 6.3 Static laser light scattering-based particle size distribution curves of curcumin nanoemulsions stabilized by different emulsifiers: Tween 20 (a); SDS (b) and DTAB(c). Each stage of in vitro digestion is represented as follows: initial (◆), stomach (■), duodenum (▲), jejunum (×) and ileum (○).

The droplet size distribution of emulsions obtained using static light scattering showed that the initial nanoemulsions had a monomodal size distribution for all emulsifier types. For Tween 20 nanoemulsions (Figure 6.3 a), after the addition of simulated gastric fluids, emulsions started

144

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge to show a bimodal distribution, with a second peak in the region of 10 µm and a corresponding decrease in the area of the first peak.

The addition of simulated intestinal fluids resulted in a multimodal distribution, with a third peak appearing in the range of 100 µm, and a gradual decrease of the initial peak (centred ~ 0.1 µm). This result suggests that the change in size distribution was an effect of digestive enzymes and not an effect of the change in pH, ionic strength or shear, as there was no change in the size distribution for the control experiment performed with everything but the enzyme (data not shown). In the case of SDS nanoemulsions (Figure 6.3 b), a more rapid decrease in the nanometre size emulsion droplets was observed and a new population of relatively large particles (around 1 µm) was formed as the emulsions passed through the in vitro system.

For DTAB-stabilized emulsions (Figure 6.3 c), the nanometre-range peak disappeared and the distribution became multimodal when nanoemulsions were subjected to simulated gastric conditions. Under simulated intestinal conditions, droplet sizes increased further, reaching approximately 1000 µm. For nanoemulsions stabilized with both SDS and DTAB, there was a significant decrease in droplets in the nanometre size range, especially under simulated gastric conditions. The increase in droplet size and the presence of a multimodal size distribution cannot be simply explained by the activation or presence of digestion enzymes. Control results (without enzymes - data not shown) also showed an increase in droplet diameter with digestion time suggesting that the change in pH, ionic strength and shear contribute to weakening of the emulsion interface promoting coalescence of oil drops. In the case of DTAB this interfacial forces weakening could result from the strong binding of bile salts (an interaction promoted by the strong electrostatic binding), that are known to displace emulsifiers at the emulsions interface, facilitating the binding of lipase and thus enabling lipolysis (Hur et al., 2009). According to other authors, emulsions stabilized by non-ionic surfactants (such as Tween 20) are more stable in the acidic gastric environment due to steric repulsion provided by the polyoxyethylene head group (Golding et al., 2011).

Results obtained with static light scattering (Mastersizer) are in agreement with the trend observed with dynamic light scattering (Zetasizer). However, in general, the hydrodynamic diameters measured by Zetasizer are lower than those measured by Mastersizer, which may be explained by the fact that the larger coalesced droplets rapidly creamed to the top of the Zetasizer cell and thus escaped detection by the laser and the detection beam passed through an unrepresentative fraction of emulsions containing relatively small droplets (Mun et al., 2007).

145

Pinheiro, A. C. (2013)

These results have shown that the nature and extent of changes in lipid droplet size as the emulsions pass through an in vitro digestion model depends on the emulsifier type, confirming the results of previous work (Hur et al., 2009; Vardakou et al., 2011). This work, however, also brings new insight in suggesting that this is the result of a contribution made by changes in pH, ionic strength and shear, and highlights the importance of subjecting emulsions to the simulated gastric environment in order to evaluate the full range of variables at stake here.

6.3.2 Influence of digestion on emulsion microstructure

The microstructure of curcumin nanoemulsions stabilized by different emulsifiers at each stage of the simulated digestion process was determined by confocal microscopy (Figure 6.4 a). Due to the size of emulsion droplets (nm) before digestion, Brownian motion prevents them from being clearly visible. However, with digestion time droplet size increases thus enabling emulsion droplets to be clearly detected under the confocal microscope.

Microscopy images show that before digestion nanoemulsions stabilised by Tween 20 contain a greater population of larger droplets compared with the other initial nanoemulsions, which is in agreement with their higher PDI value. All curcumin nanoemulsions present a spherical morphology and during simulated digestion a significant increase in nanoemulsions’ particle size was observed, showing the same tendency obtained with the size results. As demonstrated with the particle size data in Figure 6.3, confocal images show that as in vitro digestion progresses, emulsions stabilized with Tween 20 appear to have a larger droplet size when compared to emulsions formed with SDS. This size however, is much smaller than the oil droplet size visible after 120 min of simulated gastric digestion for the emulsion formed with DTAB. Confocal images for the later stages of the simulated digestion of DTAB emulsion also show a reduction in the amount of oil drops present. Our explanation for this observation is that this particular emulsion was highly unstable, leading to eventual phase separation (Figure 6.4 b). Therefore, the confocal images of DTAB-stabilized emulsions from the simulated intestinal stage show small lipid droplets remaining in the water phase. Our confocal images, in general, suggest a coalescence-based mechanism for the increase in emulsion drop diameter. There also appears to be possible clustering of emulsion drops of the emulsion stabilized with Tween 20 in the ileum.

146

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

Figure 6.4 Confocal images (a) and visual appearance of the samples (b) of curcumin nanoemulsions stabilized by different emulsifiers (Tween 20, SDS and DTAB) as they pass through an in vitro digestion model. The scale bar for all confocal images is 25 µm.

These results suggest that the stability of nanoemulsions to coalescence during digestion is strongly affected by the initial emulsifier type. The emulsifier type influences the ability of lipase to come into contact with the emulsified lipid, and this, together with the production of free fatty acids (FFA), monoacylglycerides (MAG) and diacylglycerides (DAG) at the droplet surfaces during lipid digestion, will modify the emulsion interface. These surface-active products of digestion are ineffective at stabilizing oil-in-water emulsions against coalescence (Mun et al., 2007). This could explain the observed changes in emulsion structure as digestion progressed through the simulated GI tract.

147

Pinheiro, A. C. (2013)

6.3.3 Influence of digestion on droplet charge

The surface electrical charge (ζ-potential) of curcumin nanoemulsions as they passed through the different stages of the in vitro digestion model provides information about changes in interfacial composition (Figure 6.5).

Figure 6.5 ζ-potential values of curcumin nanoemulsions stabilized by different emulsifiers: Tween 20 (), SDS () and DTAB (), as they pass through an in vitro digestion model.

The ζ-potential of the initial nanoemulsions stabilized by Tween 20, SDS and DTAB was -1.03 ± 0.64, -92.86 ± 5.89 and 89.42 ± 7.18 mV, respectively. These electrical charges are to be expected as Tween 20, SDS and DTAB are non-ionic, anionic and cationic emulsifiers, respectively.

Under simulated gastric conditions, there was no significant (p>0.05) change in ζ-potential for Tween 20 and DTAB nanoemulsions. However, in the case of SDS-stabilized nanoemulsions, the electrical charge became less negative. The pKa of SDS is about 1.9 (Chakraborty et al., 2009), therefore, as the environment of emulsions became more acidic, more SDS molecules are in the non-ionic state.

The addition of simulated intestinal fluids resulted in all emulsions having a net negative charge. This led to a reversal of charge for DTAB-stabilized nanoemulsions from positive to negative and Tween 20-stabilized nanoemulsions gained a further increase in negative charge. The observed change in electrical charge can be attributed to the adsorption of anionic components present in the intestinal juices (e.g. bile salts) or presence of free fatty acids (anionic), which are surface-active products of lipolysis. Bile salts are anionic bio-surfactants with a high affinity for the oil–water interface and are capable of displacing surface-active compounds from an oil–water interface (Hur et al., 2009; Maldonado-Valderrama et al., 2008; Mun et al., 2007). Bile salts then remove digestion products from the interface and solubilize them into the bulk aqueous phase (Golding et al., 2011; Malaki Nik et al., 2011; Maldonado-

148

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

Valderrama et al., 2008). Although a precise mechanism is difficult to define from the current measurements, the high negative charge reported for DTAB-stabilized emulsions after digestion could be a result of the presence of bile salts at the interface or digestion products of TAG (i.e. mono and di-glycerides). These data are in agreement with previous work that has shown that the charge of emulsions stabilized by Tween 20, lysolecithin, caseinate and whey protein became more negative when they moved from a simulated stomach to a small intestine environment (Hur et al., 2009).

At the end of simulated digestion (ileum), ζ-potential values of all the emulsions were fairly similar irrespective of the initial droplet charge.

6.3.4 In vitro digestibility

A measure of digestibility is the production of free fatty acids (FFA). Figure 6.6 shows the results for the production of FFA during simulated intestinal digestion of curcumin nanoemulsions stabilized by different emulsifiers.

Figure 6.6 Influence of emulsifier type on the production of total free fatty acids from curcumin nanoemulsions stabilized by Tween 20 (), SDS () and DTAB ().

The total FFA produced as measured from the back titration was the greatest for Tween 20- stabilized emulsion. This increased production is attributed to the reduced size of emulsion droplets (observed throughout simulated digestion), which maintains a large surface area for the binding of lipase and bile salts.

Emulsions stabilized with SDS and DTAB showed a reduced amount of FFA production in comparison to Tween 20. With SDS-stabilized emulsions, charge repulsion could be preventing effective binding of bile salts and thus the binding of lipase is not being promoted. In this case the further increase in negative charge observed in the simulated small-intestine (Figure 6.5) could have been mainly due to the presence of free fatty acids and not due to the effective

149

Pinheiro, A. C. (2013)

binding of bile salts. In the case of DTAB-stabilized emulsions, a clear phase separation (as seen in Fig 3b) led to a significant reduction in surface area for enzyme binding, decreasing the rate of FFA production. Another explanation presented in literature is the binding of FFA to DTAB altering the extent of digestion (Dahan and Hoffman, 2006). For all three emulsifier types, FFA production reached an equilibrium value. This constant value could be attributed to the inhibition of lipase activity by the FFA released (Silva et al., 2012). Our observations suggest that the size of the emulsion droplets as well as emulsifier surface charge affected the total production of FFA.

6.3.5 In vitro curcumin bioavailability

The influence of the emulsifier type on bioavailability of curcumin at different stages of the simulated digestion process is presented in Figure 6.7.

Figure 6.7 Effect of initial emulsifier type on curcumin bioavailability as nanoemulsions pass through an in vitro digestion model: nanoemulsions stabilized by Tween 20 (), SDS () and DTAB ().

The bioavailability of curcumin, measured as the curcumin concentration in the mixed micelle phase, depends on the emulsifier type and on the digestion stage. Curcumin bioavailability for nanoemulsions stabilized by Tween 20 increased with digestion time and appears to have significance only for SDS-stabilized emulsion in the jejunum and ileum. This behaviour can be attributed to the formation of digestion products that have the ability to form mixed micelles capable of solubilizing highly lipophilic components. Also, the greater free fatty acid production observed in the case of Tween 20-stabilized nanoemulsions (Figure 6.6) could explain the high curcumin bioavailability (in relation to the initial bioavailability) obtained for Tween 20-stabilized nanoemulsions.

150

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

In contrast, curcumin had very low bioavailability under simulated intestinal conditions when it was encapsulated in an emulsion containing DTAB. This effect can be attributed to the fact that under simulated intestinal conditions, DTAB-stabilized emulsions are highly unstable, leading to phase separation. Curcumin, which is soluble only in the oil phase, will have remained in the separated bulk oil, and with this reduced surface area, digestive enzymes will not have acted as efficiently to enable mixed micelle formation, thus reducing bioavailability. Similar results were obtained in another work, in which it was observed that the total amount of β-carotene present in the micellar phase during each stage of digestion is much higher in small-sized emulsions compared to emulsions that are larger in size, which was attributed to the increased surface area for enzyme interaction (Liu et al., 2012).

6.3.6 Curcumin nanoemulsions stabilized by a multilayer coating

Previous works showed that digestion and release of bioactive lipophilic components encapsulated within emulsion-based delivery systems can be controlled by coating the lipid droplets with multilayer coatings (Li et al., 2010). As a preliminary work, chitosan was added to SDS-stabilized emulsions, in order to evaluate the possibility of formation of a multilayer emulsion. In order to establish the optimum chitosan concentration required to prepare stable multilayer emulsions, different chitosan concentrations (0-0.1 wt %) were used and their effect on ζ-potential and particle size were measured (results not shown). The optimum chitosan concentration, i.e. the concentration at which SDS-stabilized emulsions were completely coated by a chitosan layer and no significant chitosan excess in solution was detected, was found to be 0.02%. Figure 6.8 shows the morphology of the SDS/chitosan-stabilized curcumin emulsions.

SDS-stabilized emulsions

Chitosan layer

2.5 µm

Figure 6.8 Confocal images of curcumin nanoemulsions stabilized by SDS and chitosan. Scale bar represents 2.5 µm.

151

Pinheiro, A. C. (2013)

From Figure 6.8 it is possible to observe the multilayer structure around the lipid droplets, proving the deposition of the chitosan layer. The chitosan layer is stained in green and can be clearly distinguished from the SDS-stabilized emulsions. The successful addition of a biopolymer layer around SDS-coated lipid droplets offers the possibility of a tailored design of delivery systems for lipophilic bioactive compounds.

6.4 CONCLUSIONS

A digestion model involving in vitro simulation of stomach, duodenum, jejunum and ileum was used to evaluate the impact of surface charge on the digestion of curcumin nanoemulsions. Although all the nanoemulsions tested were markedly affected by the digestion process, DTAB- stabilized emulsions showed the most drastic change in droplet size and electrical charge during digestion. The positively charged surface of DTAB-stabilized nanoemulsions could have promoted the adsorption of anionic lipase and anionic bile salts to the oil–water interface, however, the reduction in surface area would not have promoted enzyme binding to the emulsion interface. The initial emulsifier type was also found to impact curcumin bioavailability due to the differences in the formation of digestion products that have the ability to form mixed micelles capable of solubilizing highly lipophilic components. Curcumin exhibited a very low bioavailability under intestinal conditions in the presence of DTAB as emulsifier. The highest curcumin bioavailability, observed for Tween 20-stabilized nanoemulsions (in relation to the initial bioavailability), could be related with the greater free fatty acid production obtained for this nanoemulsion as a direct result of emulsion droplet size.

The results from this work showed that the behaviour of nanoemulsions (e.g. free fatty acids release and bioavailability of the bioactive compound incorporated) during in vitro digestion depends on the charge/type of the initial emulsifier. This work also highlighted the importance of using a more realistic digestion model, particularly of subjecting the emulsions to a simulated gastric environment, once changes in pH, ionic strength, gastric enzymes activity and shear will impact the emulsions properties’ when passing through the small-intestine. The obtained results will contribute to the development of optimized nanoemulsion-based delivery systems to protect and release bioactive lipophilic compounds within the human body, for food and pharmaceutical applications.

152

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

6.5 REFERENCES

Ahmed, K., Li, Y., McClements, D. J., & Xiao, H. (2012). Nanoemulsion- and emulsion- based delivery systems for curcumin: Encapsulation and release properties, Food Chemistry 132(2), 799-807. Amara, S., Lafont, D., Fiorentino, B., Boullanger, P., Carrière, F., & De Caro, A. (2009). Continuous measurement of galactolipid hydrolysis by pancreatic lipolytic enzymes using the pH-stat technique and a medium chain monogalactosyl diglyceride as substrate, Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 1791(10), 983-990. Anand, P., Kunnumakkara, A. B., Newman, R. A., & Aggarwal, B. B. (2007). Bioavailability of Curcumin: Problems and Promises, Molecular Pharmaceutics 4(6), 807-818. Chakraborty, S., Shukla, D., Jain, A., Mishra, B., & Singh, S. (2009). Assessment of solubilization characteristics of different surfactants for carvedilol phosphate as a function of pH, Journal of Colloid and Interface Science 335(2), 242-249. Dahan, A., & Hoffman, A. (2006). Use of a Dynamic in Vitro Lipolysis Model to Rationalize Oral Formulation Development for Poor Water Soluble Drugs: Correlation with in Vivo Data and the Relationship to Intra-Enterocyte Processes in Rats, Pharmaceutical Research 23(9), 2165-2174. Golding, M., Wooster, T. J., Day, L., Xu, M., Lundin, L., Keogh, J., & Clifton, P. (2011). Impact of gastric structuring on the lipolysis of emulsified lipids, Soft Matter 7(7), 3513-3523. Helbig, A., Silletti, E., Timmerman, E., Hamer, R. J., & Gruppen, H. (2012). In vitro study of intestinal lipolysis using pH-stat and gas chromatography, Food Hydrocolloids 28(1), 10-19. Hur, S. J., Decker, E. A., & McClements, D. J. (2009). Influence of initial emulsifier type on microstructural changes occurring in emulsified lipids during in vitro digestion, Food Chemistry 114(1), 253-262. Kong, F., & Singh, R. P. (2010). A Human Gastric Simulator (HGS) to Study Food Digestion in Human Stomach, Journal of Food Science 75(9), E627-E635. Kuttan, R., Bhanumathy, P., Nirmala, K., & George, M. C. (1985). Potential anticancer activity of turmeric (Curcuma longa), Cancer Letters 29(2), 197-202. Lesmes, U., Baudot, P., & McClements, D. J. (2010). Impact of Interfacial Composition on Physical Stability and In Vitro Lipase Digestibility of Triacylglycerol Oil Droplets Coated with Lactoferrin and/or Caseinate, Journal of Agricultural and Food Chemistry 58(13), 7962-7969. Li, Y., Hu, M., & McClements, D. J. (2011). Factors affecting lipase digestibility of emulsified lipids using an in vitro digestion model: Proposal for a standardised pH-stat method, Food Chemistry 126(2), 498-505.

153

Pinheiro, A. C. (2013)

Li, Y., Hu, M., Xiao, H., Du, Y., Decker, E. A., & McClements, D. J. (2010). Controlling the functional performance of emulsion-based delivery systems using multi-component biopolymer coatings, European Journal of Pharmaceutics and Biopharmaceutics 76(1), 38-47. Li, Y., & McClements, D. J. (2010). New Mathematical Model for Interpreting pH-Stat Digestion Profiles: Impact of Lipid Droplet Characteristics on in Vitro Digestibility, Journal of Agricultural and Food Chemistry 58(13), 8085-8092. Liu, Y., Hou, Z., Lei, F., Chang, Y., & Gao, Y. (2012). Investigation into the bioaccessibility and microstructure changes of β-carotene emulsions during in vitro digestion, Innovative Food Science and Emerging Technologies 15(0), 86-95. Malaki Nik, A., Wright, A. J., & Corredig, M. (2011). Impact of interfacial composition on emulsion digestion and rate of lipid hydrolysis using different in vitro digestion models, Colloids and Surfaces B: Biointerfaces 83(2), 321-330. Maldonado-Valderrama, J., Woodward, N. C., Gunning, A. P., Ridout, M. J., Husband, F. A., Mackie, A. R., Morris, V. J., & Wilde, P. J. (2008). Interfacial Characterization of β- Lactoglobulin Networks: Displacement by Bile Salts, Langmuir 24(13), 6759-6767. McClements, D. J., & Xiao, H. (2012). Potential biological fate of ingested nanoemulsions: influence of particle characteristics, Food & Function 3(3), 202-220. Mun, S., Decker, E. A., & McClements, D. J. (2007). Influence of emulsifier type on in vitro digestibility of lipid droplets by pancreatic lipase, Food Research International 40(6), 770- 781. Pafumi, Y., Lairon, D., de la Porte, P. L., Juhel, C., Storch, J., Hamosh, M., & Armand, M. (2002). Mechanisms of Inhibition of Triacylglycerol Hydrolysis by Human Gastric Lipase, Journal of Biological Chemistry 277(31), 28070-28079. Pinsirodom, P., & Parkin, K. L. (2001) Lipase Assays, In Current Protocols in Food Analytical Chemistry, John Wiley & Sons, Inc. Reis, P., Raab, T., Chuat, J., Leser, M., Miller, R., Watzke, H., & Holmberg, K. (2008). Influence of Surfactants on Lipase Fat Digestion in a Model Gastro-intestinal System, Food Biophysics 3(4), 370-381. Rozner, S., Shalev, D. E., Shames, A. I., Ottaviani, M. F., Aserin, A., & Garti, N. (2010). Do food microemulsions and dietary mixed micelles interact?, Colloids and Surfaces B: Biointerfaces 77(1), 22-30. Sarkar, A., Goh, K. K. T., Singh, R. P., & Singh, H. (2009). Behaviour of an oil-in-water emulsion stabilized by β-lactoglobulin in an in vitro gastric model, Food Hydrocolloids 23(6), 1563-1569.

154

Chapter 6. Unravelling the behavior of curcumin nanoemulsions during in vitro digestion: Effect of surface charge

Sharma, O. P. (1976). Antioxidant activity of curcumin and related compounds, Biochemical Pharmacology 25(15), 1811-1812. Sidhu, G. S., Singh, A. K., Thaloor, D., Banaudha, K. K., Patnaik, G. K., Srimal, R. C., & Maheshwari, R. K. (1998). Enhancement of wound healing by curcumin in animals, Wound Repair and Regeneration 6(2), 167-177. Silva, H., Cerqueira, M., & Vicente, A. (2012). Nanoemulsions for Food Applications: Development and Characterization, Food and Bioprocess Technology 5(3), 854-867. Srimal, R. C., & Dhawan, B. N. (1973). Pharmacology of diferuloyl methane (curcumin), a non-steroidal anti-inflammatory agent, Journal of Pharmacy and Pharmacology 25(6), 447- 452. Troncoso, E., Aguilera, J. M., & McClements, D. J. (2012). Fabrication, characterization and lipase digestibility of food-grade nanoemulsions, Food Hydrocolloids 27(2), 355-363. Vardakou, M., Mercuri, A., Barker, S., Craig, D., Faulks, R., & Wickham, M. (2011). Achieving Antral Grinding Forces in Biorelevant In Vitro Models: Comparing the USP Dissolution Apparatus II and the Dynamic Gastric Model with Human In Vivo Data, AAPS PharmSciTech 12(2), 620-626. Wang, Y.-F., Shao, J.-J., Zhou, C.-H., Zhang, D.-L., Bie, X.-M., Lv, F.-X., Zhang, C., & Lu, Z.- X. (2012). Food preservation effects of curcumin microcapsules, Food Control 27(1), 113-117. Wang, Y., Lu, Z., Wu, H., & Lv, F. (2009). Study on the antibiotic activity of microcapsule curcumin against foodborne pathogens, International Journal of Food Microbiology 136(1), 71- 74.

155

7 CHAPTER 7. I N F L U E N C E

OF INTERFACIAL COMPOSITION ON IN VITRO DIGESTIBILITY OF CURCUMIN NANOEMULSIONS STABILIZED BY BIOPOLYMER EMULSIFIERS Influence of interfacial composition on in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

7.1 Introduction...... 159

7.2 Materials and Methods ...... 161

7.3 Results and discussion ...... 165

7.4 Conclusions...... 173

7.5 References ...... 174

Pinheiro, A. C. (2013)

The results presented in this Chapter were adapted from:

158

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

7.1 INTRODUCTION

The growing consumers’ awareness of the link between food and health and the consequent increase in demand for functional foods are stimulating the innovation and the development of new products in the food industry. Many health-promoting compounds are lipophilic (e.g. curcumin, ω-3 fatty acids, lycopene, phytosterols, quercetin, resveratrol), however, their utilization by food industry is limited due to their low bioavailability and difficulties associated with their incorporation into food matrices (McClements and Xiao, 2012). In particular, curcumin (a natural polyphenolic phytochemical extracted from the powdered rhizomes of turmeric spice) has been reported to have many biological activities, such as antitumor, antioxidant, anti-microbial and anti-inflammatory properties. However, its application as a bioactive ingredient is currently limited by its poor water solubility (hindering its incorporation into many food products) and low bioavailability (which means that its health-promotion effects may not be realised) (Ahmed et al., 2012). Emulsion-based delivery systems are one of the vehicles that are being increasingly used to both protect and deliver lipophilic bioactive compounds (increasing their bioavailability) and to achieve the incorporation of into food products (Salminen et al., 2013). These systems can be produced entirely from bio-based materials, using simple processing operations such as blending or homogenization (Tokle and McClements, 2011). For example, lactoferrin, a globular protein derived from milk with a wide range of reported biological activities (e.g. antioxidant and antimicrobial activity, cancer prevention, and modulation of immune responses) can be used as natural emulsifier to enhance the formation and stabilization of oil-in-water emulsions (Balcão et al., 2013; Shimoni et al., 2013).

The design of emulsion-based delivery systems to control digestion, release and absorption of encapsulated lipophilic bioactive compounds within the GI tract is one of the current challenges of food industry (McClements, 2013). Digestibility of lipids and release of bioactive lipophilic components encapsulated within emulsion-based delivery systems can be controlled by coating lipid droplets with biopolymer coatings. These multilayer emulsions can be formed by electrostatic deposition of ionic biopolymers onto oppositely charged lipid droplets (layer- by-layer (LbL) technique) (Li et al., 2010). Alginate, a naturally occurring anionic polymer typically obtained from brown seaweed, is one of the polyelectrolytes most used for LbL deposition (Maurstad et al., 2008; Zhou et al., 2010), which is probably due to its biocompatibility, low toxicity and relatively low cost (Lee and Mooney, 2012). The design of the nanolayered coatings interface will depend on their specific application: nanolayered coatings can be used to increase the bioavailability of incorporated bioactive compound, being

159

Pinheiro, A. C. (2013)

in this case essential that it is broken down within the GI tract so that the bioactive component can be released and absorbed, or it can be designed to be completely indigestible within the GI tract then it could be used to reduce the calorie content of foods by preventing digestive enzymes (that is, lipases) from accessing encapsulated lipids (McClements, 2010).

There is wide range of emulsion-based delivery systems that can be assembled entirely from food grade materials. Nanoemulsions (radius < 100 nm) present some possible advantages over conventional emulsions, such as higher stability to gravitational separation and droplet aggregation, higher optical clarity and enhanced bioavailability of encapsulated lipophilic bioactive compounds (due to their small size) (McClements and Xiao, 2012). Despite of these potential benefits of nanoemulsions, their application to foods is hindered by some concerns about potential risks associated with the ingestion of nanostructures, such as the potential toxicity of some components used to produce them (e.g. surfactants and organic solvents) and their ability to alter the biological fate of ingested materials and encapsulated bioactive compounds, which could potentially have some adverse effects on human health (McClements, 2013). The first issue can be solved by using only bio-based food grade materials when producing the nanoemulsions; the second issue is a broader one and must be evaluated (ideally) in vivo, or at least in vitro. The knowledge of the behaviour of nanoemulsions as well as the fate of bioactive compounds encapsulated within them in the GI tract is of utmost importance to either assess their safety for human consumption and to produce tailored delivery systems (i.e. with optimized bioactivity). In recent years, considerable efforts have been made in understanding the physicochemical behaviour of nanoemulsions within the GI tract (Ahmed et al., 2012; Salvia-Trujillo et al., 2013), however further work is clearly needed because there is still a lack of knowledge regarding the major factors that influence the biological fate of ingested nanoemulsions. Also, most of the works that evaluate the behaviour of emulsions under gastrointestinal conditions use static in vitro digestion models and simplified gastrointestinal conditions that do not accurately simulate the complex physicochemical and physiological processes that occur within the human gastrointestinal tract (Li et al., 2010; Mun et al., 2007). Therefore being very difficult, often infeasible, the use of in vivo models (high costs and ethical constrains often involved), there is a need of using more realistic in vitro gastrointestinal models.

In this work, a dynamic digestion model, comprising the simulation of stomach, duodenum, jejunum and ileum, was developed and used to evaluate the behaviour of curcumin nanoemulsions stabilized by biopolymer emulsifiers (lactoferrin and lactoferrin/alginate multilayer structure) under gastrointestinal conditions.

160

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

7.2 MATERIALS AND METHODS

7.2.1 Materials

Corn oil was purchased from a local supermarket in Braga (Portugal) and was used without further purification. Curcumin was purchased from Sigma-Aldrich (St. Louis, MO), lactoferrin was acquired from DMV International (Veghel, The Netherlands) and sodium alginate was purchased from Kelco International, Ltd. (UK). Pepsin from porcine gastric mucosa (600 U.mL- 1), lipase (40 U.mL-1), pancreatin from porcine pancreas (8xUSP), bile extract porcine and the salts used for preparing the gastric electrolyte solution and the small intestinal electrolyte solution, hydrochloric acid and sodium bicarbonate were purchased from Sigma-Aldrich (St. Louis, MO). Chloroform was obtained from Fisher Scientific (NJ, USA), acetone from Fisher Chemical (UK) and sodium hydroxide and phenolphthalein from Panreac (Spain).

7.2.2 Emulsion preparation

The primary curcumin nanoemulsions (stabilized by lactoferrin) were prepared by homogenizing 5 wt.% corn oil containing 0.1 wt.% of curcumin with 95 wt.% aqueous emulsifier solution (lactoferrin at a concentration of 2%), using a Ultra-Turrax homogenizer (T 25, Ika-Werke, Germany) for 2 min followed by passage through a high pressure homogenizer (NanoDeBee, Bee International, South Easton, Massachusetts, USA) at 20 000 Psi (137.9 MPa), for 20 cycles. The secondary emulsions (stabilized by lactoferrin/alginate) were prepared by adding the primary nanoemulsion (pH 4, at which lactoferrin has the maximum charge) dropwise with a syringe pump (NE-1000, New Era Pump Systems, Inc., USA) to an alginate solution (pH 7, at which alginate has the maximum charge) under stirring for 30 min. The secondary emulsion was then treated with sonication (during 2 min at a frequency of 50-60 kHz) (Brason 5510, Bransonic® ultrasonic cleaner, USA) to disrupt any flocculated droplets formed during its preparation (Li et al., 2010). Both emulsions were kept at 4 °C in the dark before characterization and in vitro digestion.

7.2.3 In vitro digestion

7.2.3.1 Gastro-intestinal model

A dynamic gastrointestinal system (Figure 7.1) was develop during this PhD project and was used in the in vitro digestion experiments. This system was constructed based on the TIM model developed by TNO (Zeist, The Netherlands) (Minekus and Havenaar, 1996).

161

Pinheiro, A. C. (2013)

Figure 7.1 Dynamic gastrointestinal system used in the in vitro digestion experiments: (A) gastric compartment, (B) duodenal compartment, (C) jejunal compartment, (D) ileal compartment, (E) glass jacket, (F) flexible wall, (G) pH electrodes, (H) syringe pumps, (I) hollow fibre membrane, (J) water bath, (K) peristaltic pumps (L) pump controllers, (M) recovery of jejunal filtrate, ileal filtrate and ileal delivery on ice.

This model simulates the major events occurring during digestion and consists of four compartments simulating the stomach (A), duodenum (B), jejunum (C) and ileum (D). Each compartment is formed by two connected basic units consisting of a glass jacket (E) with a flexible wall (F) inside (produced with plastic from stomacher bags). Water is pumped from a water bath (J) around the flexible walls to maintain the temperature at 37 °C and to make pressure on the flexible walls, which will enable the mixing of the chyme by the alternate compression and relaxation of the flexible walls. The changes in water pressure are achieved by peristaltic pumps (K), which alter the flow direction according to the timed controller (L) connected to them. The compartments are connected by silicone tubes as every 20 min a constant volume of chyme is transferred.

All compartments are equipped with pH electrodes (G) and pH values are controlled by the

-1 -1 secretion of HCl (1 mol.L ) into the stomach and NaHCO3 (1 mol.L ) into the intestine compartments. The gastric and intestinal secretions are added via syringe pumps (H) at pre-set flow rates. The jejunum and ileum compartments are connected with hollow-fibre devices (I) (SpectrumLabs Minikros®, M20S-100-01P, USA) to absorb digestion products and water from the chyme and to modify electrolyte and bile salts concentration in the chyme.

162

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

7.2.3.2 Experimental Conditions

In vitro digestion was performed as described by other authors (Reis et al., 2008) with the following modifications. A volume of 60 mL of curcumin nanoemulsions (both primary and secondary emulsions) was introduced into the dynamic gastrointestinal system (gastric compartment) and the experiment was run for a total of 5 h, simulating average physiological conditions of GI tract by the continuous addition of gastric, duodenal, jejunal and ileal secretions. The gastric secretion consisted of pepsin and lipase in a gastric electrolyte solution

-1 -1 -1 -1 (NaCl 4.8 g.L , KCl 2.2 g.L , CaCl2 0.22 g.L and NaHCO3 1.5 g.L ), secreted at a flow rate of 0.33 mL.min-1. The pH was controlled to follow a predetermined curve (from 4.8 at t=0 to 1.7 at t=120 min) by secreting HCl (1 mol.L-1). The duodenal secretion consisted of a mixture of 66 mL of 4% (w/v) porcine bile extract, 34 mL of 7% (w/v) pancreatin solution and 20 mL of a

-1 -1 -1 small intestinal electrolyte solution (SIES) (NaCl 5 g.L , KCl 0.6 g.L , CaCl2 0.25 g.L ) at a flow rate of 0.66 mL.min-1. The jejunal secretion fluid consisted of SIES containing 10% (v/v) porcine bile extract solution at a flow rate of 2.13 mL.min-1. The ileal secretion fluid consisted of SIES at a flow rate of 2.0 mL.min-1. The pH in the different parts of small intestine was controlled by

-1 the addition of 1 mol.L NaHCO3 solution to set-points of 6.5, 6.8 and 7.2 for simulated duodenum, jejunum and ileum, respectively. During in vitro digestion, samples were collected directly from the lumen of the different compartments, from the jejunal and ileal filtrates and from the ileal delivery. The jejunal and ileal filtrates were used to determine the bioavailable fraction of curcumin. The samples were analysed for size, ζ-potential and free fatty acids.

Curcumin nanoemulsions were tested in the dynamic gastrointestinal model at least in triplicate.

7.2.4 Particle size

The particle size of the emulsions was measured at different stages of digestion by dynamic light scattering (Zetasizer Nano ZS, Malvern Instruments, Worcestershire, UK). The Z-average mean diameter was calculated by the instrument using Stokes-Einstein equation, assuming that nanoemulsion droplets were spherical. Before measurement, the samples were diluted at a ratio of 1:100 (v/v) in an appropriate buffer solution with the same pH as the aqueous phase of the sample (KCl-HCl (10 mmol.L-1, pH 2) for stomach step and phosphate buffer (10 m mmol.L-1, pH 7) for before digestion and small intestine steps (Hur et al., 2009)) at room temperature to avoid multiple scattering effects. The measurements were performed in triplicate.

163

Pinheiro, A. C. (2013)

7.2.5 ζ-potential

The samples were diluted at a ratio of 1:100 (v/v) in appropriate buffer solution (KCl-HCl buffer (10 mmol.L-1, pH 2) for stomach step and phosphate buffer (10 mmol.L-1, pH 7) for before digestion and small intestine steps (Hur et al., 2009) at room temperature before measurement and each individual ζ-potential measurement was determined from the average of three readings.

7.2.6 Transmission Electron Microscopy (TEM)

7.2.7 Free fatty acids released

The digestion activity was measured by determining the amount of free fatty acids (FFA) released from curcumin nanoemulsions using a titration method (Pinsirodom, 2005). Briefly, 5 mL of jejunal filtrate, ileal filtrate and ileal delivery samples were collected and 10 mL of acetone were added to quench the enzymes’ activity and 3 drops of 1% (w/v) phenolphthalein were added as an indicator. A direct titration with 0.1 mol.L-1 NaOH using a burette was performed and the volume of NaOH added until the titration end point was determined and used to calculate the concentration of FFA produced by lipolysis. Therefore, the percentage of free fatty acids released was calculated from the number of moles of NaOH required to neutralize the FFA divided by the number of moles of FFA that could be produced from triglycerides if they were all digested (assuming 2 FFA produced per 1 triacylglycerol molecule) (Li et al., 2011):

( ) Equation 7.1

where vNaOH is the volume of sodium hydroxide required to neutralize the FFA generated (in -1 mL), mNaOH is the molarity of the sodium hydroxide used (in mol.L ), wlipids is the total weight of

164

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

corn oil initially present and Mlipid is the molecular weight of the corn oil (based on their average fatty acid composition the molecular weight of corn oil was considered to be 800 g.mol-1).

7.2.8 Curcumin bioavailability

It was assumed that the fraction of the original curcumin that ended up in the micelle phase was a measure of curcumin bioavailability (Ahmed et al., 2012) and that the mixed micelles that contained the bioavailable curcumin fraction were able to pass the hollow-fibre membranes (i.e. jejunal filtrate and ileal filtrate), while undigested emulsions were retained (Minekus et al., 2005). Curcumin bioavailability was determined based on the methodology described by other authors (Ahmed et al., 2012). Briefly, 5mL of the jejunal and ileal filtrate were vortexed with 5 mL of chloroform, and then centrifuged (Sigma 4K15, Germany) at 1750 rpm, at room temperature, for 10 min. The bottom chloroform layer was collected and the extraction procedure was repeated with the top layer. The second bottom chloroform layer was added to the previously set aside chloroform layer, mixed, and analysed in a UV–VIS spectrophotometer (Jasco V560, USA) at 419 nm (absorbance peak). The concentration of curcumin was determined from a previously prepared calibration curve of absorbance versus curcumin concentration in chloroform.

7.2.9 Statistical analyses

All experiments were carried out in triplicate using freshly prepared samples. The results were then reported as averages and standard deviations of these measurements. The statistical analyses were carried out using analysis of variance, Tukey mean comparison test (p<0.05) and linear regression analysis (SigmaStat, trial version, 2003, USA).

7.3 RESULTS AND DISCUSSION

7.3.1 Optimization of the secondary emulsion preparation

In Chapter 5, the optimum lactoferrin concentration required to form the primary curcumin nanoemulsion composed of 5 wt% corn oil was determined. A concentration of 2% was found to be the minimum amount of lactoferrin required to form stable nanoemulsions and, therefore, this concentration was used to prepare the primary nanoemulsions.

Regarding secondary emulsions, the biopolymer concentration used for their preparation must be carefully controlled to avoid particle aggregation (due to charge neutralization, bridging or

165

Pinheiro, A. C. (2013)

depletion effects) (McClements, 2005). In order to establish the optimum alginate concentration required to prepare stable secondary emulsions, different alginate concentrations (0-0.3 wt %) were used and their effects on ζ-potential and particle size were measured (Figure 7.2).

Figure 7.2 Influence of alginate concentration on the ζ-potential (a) and size (b) of secondary emulsions

In the absence of alginate, lactoferrin-stabilized nanoemulsions exhibited a size of 149.3 ± 4.87 nm and ζ-potential of 27.47 ± 1.11 (at pH 4). As the concentration of alginate increased, the ζ- potential of the secondary emulsions became increasingly negative until it reached a constant value (≈ -30 mV) (at an alginate concentration of 0.20%) (Figure 7.2 a). This suggests that the anionic alginate molecules adsorbed to the surfaces of lactoferrin-stabilized nanoemulsions until they became saturated with alginate.

The increase in alginate concentration also influenced the emulsions’ size (Figure 7.2 b). At low alginate concentrations (< 0.06%), a significant increase in emulsion size occurred, suggesting extensive droplet aggregation which can be attributed to the low net charge on droplets as well as to bridging flocculation effects that occurred when droplet surfaces were not saturated with alginate (Li et al., 2010). At alginate concentrations higher than 0.08%, an appreciable decrease in emulsions size was observed, suggesting that the emulsions were not susceptible to droplet aggregation at high alginate concentrations. This behaviour can be due to the strong electrostatic repulsion between the droplets and to the saturation of the droplets surfaces with alginate so that there was less bridging flocculation (Thanasukarn et al., 2006).

Therefore, 0.2% was the alginate concentration selected for the preparation of secondary emulsions, since at this concentration lactoferrin-stabilized nanoemulsions are completely coated by an alginate layer and there is no significant alginate excess in solution.

166

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

7.3.2 In vitro digestion

In vitro digestions were conducted in order to assess if the presence of a dietary ﬁbre (in this cases alginate) coating around lactoferrin-stabilized curcumin nanoemulsions inﬂuence their behaviour within a simulated GI tract.

7.3.2.1 Influence of alginate on particle characteristics during in vitro digestion

The effect of the alginate coating addition on particle size during in vitro digestion can be seen in Figure 7.3.

Figure 7.3 Mean particle diameter of curcumin nanoemulsions stabilized by lactoferrin () and lactoferrin/alginate () as they pass through the dynamic in vitro digestion model.

At the stomach stage, the mean particle diameter of the primary nanoemulsion (stabilized by lactoferrin) remained relatively small, although some coalesce phenomenon occurred (Figure 7.4). On the contrary, the secondary emulsion (stabilized by lactoferrin and alginate) were highly unstable under gastric conditions, exhibiting a large increase in the particle size (indicating that extensive droplet aggregation occurred) and a high evidence of creaming (result not shown). This different behaviour can be related to differences in their interfacial characteristics. At acidic conditions, the lactoferrin-stabilized emulsions are highly positively charged (Figure 7.5) and consequently there is a strong electrostatic repulsion between them, avoiding the emulsions aggregation, coalescence and flocculation. On the other hand, alginate molecules have a linear anionic chain with no side chains, and therefore they are more susceptible to droplet aggregation through charge neutralization and bridging effects since there is little steric repulsion contribution (Tokle et al., 2012). Also, at acidic pH, the electrostatic attraction between alginate and the cationic lactoferrin-stabilized emulsions are weakened because of the partial loss of negative charge on alginate below their pKa values, which could have led to bridging flocculation (i.e. a single alginate molecule may have partially

167

Pinheiro, A. C. (2013)

detached from one lipid droplet surface and became attached to another lipid droplet) (Tokle et al., 2010). The droplet aggregation of lactoferrin/alginate-stabilized emulsions at the stomach stage is clearly visible in Figure 7.4.

Under intestinal conditions, both primary and secondary emulsions exhibited high particle diameters, due to particle aggregation (flocculation and coalescence of emulsions) (Figure 7.3 and Figure 7.4). At these stages of digestion, bile salts may have been able to displace some lactoferrin from the lipid surface and lipase molecules were probably able to access and hydrolyse the lipids, producing free fatty acids (FFA), monoacylglycerides (MAG) and diacylgycerides (DAG) at the droplet surfaces. These surface-active products of digestion are ineffective at stabilizing oil-in-water emulsions against coalescence (Mun et al., 2007), which can explain the observed emulsions instability (i.e. high particle sizes).

As expected, jejunal and ileal filtrates (i.e. the fraction absorbed at jejunum and ileum compartments, respectively) exhibit particle sizes significantly (p<0.05) lower than the ileal delivery (i.e. fraction that is not absorbed in the small intestine).

Regarding the effect of digestion on droplet charge (Figure 7.5), it was observed that in the stomach stage, the electrical charge of both emulsions became more positive which can be attributed to changes in solutions conditions (pH and ionic strength), competitive displacement of the original emulsifier by surface active materials present in the gastric electrolyte solution, or adsorption of charged species from gastric electrolyte solution on top of the original emulsifier coating (Hur et al., 2009). Also, the gastric electrolyte solution contains pepsin, a protease that may have hydrolysed the adsorbed lactoferrin molecules thus facilitating their competitive displacement. Comparatively with primary emulsions, secondary emulsions exhibit a lower positive charge under gastric conditions, which suggests that, despite the fact that at acid pH electrostatic attraction between alginate and lactoferrin is weakened, alginate remained partly bound to lactoferrin-coated lipid droplets, protecting lactoferrin from pepsin hydrolysis.

168

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers Lactoferrin + Alginate Lactoferrin

Inicial

Stomach

Duodenum

Jejunal Filtrate

Ileal Filtrate

Ileal Delivery

169

Pinheiro, A. C. (2013)

Figure 7.5 ζ-potential values of curcumin nanoemulsions stabilized by lactoferrin () and lactoferrin/alginate () as they pass through the dynamic in vitro digestion model.

At duodenal conditions (pH=6.5) both lactoferrin and lactoferrin/alginate-stabilized emulsions became highly negatively charged. If there were no changes in the interfacial composition, the lactoferrin-stabilized emulsions would be slightly positively charged at neutral pH, once the isoelectric point (pI) of lactoferrin is between 8.4 and 9.0 (Farnaud and Evans, 2003). Therefore, the fact that the primary emulsion has high negative charge suggest that there was a displacement of the lactoferrin layer from lipid surfaces by surface-active anionic molecules (e.g free fatty acids, bile salts or lipase), the adoption of anionic species (e.g. free fatty acids, bile salts or lipase) on top of the lactoferrin layer, or adoption of small negatively charged ions from the small intestinal electrolyte solution (Tokle et al., 2012). Also, the negative charge of lactoferrin-stabilized emulsions at this stage of digestion means that in the case of lactoferrin/alginate-stabilized emulsions, alginate is no longer electrostatically attracted to their surface, which leaves the emulsions more susceptible to lipase hydrolysis. There are no significant differences (p>0.05) between ζ-potential values of primary and secondary emulsions at the subsequent stages of digestion process.

7.3.2.2 Influence of alginate on lipid digestion

The effect of alginate coating on free fatty acids release was also examined at the different stages of small intestine digestion (Figure 7.6).

The overall extent of lipid digestion was fairly similar for both primary and secondary emulsions, which suggests that alginate coating did not prevent lipid digestion. However, there are some differences in the percentage of FFA released in the different stages and fractions of small intestine digestion.

170

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

Figure 7.6 Percentage of free fatty acids (FFA) released from lactoferrin (LF) and lactoferrin/alginate (LF+Al) stabilized emulsions as they pass through the dynamic in vitro digestion model: Initial (); Jejunal filtrate (); Ileal filtrate () and Ileal delivery ().

The major difference was observed in the jejunal filtrate fraction, with lactoferrin/alginate- stabilized emulsions having a significantly (p<0.05) lower amount of FFA as compared to lactoferrin-stabilized emulsions. These results suggest that lactoferrin/alginate stabilized- emulsions were digested at a slower rate than lactoferrin-stabilized emulsions, which is in agreement with previous works (Tokle et al., 2012). Alginate is resistant to digestion in the stomach and small-intestine and has the ability to strongly bind calcium ions, preventing them from precipitating free fatty acids produced at the oil droplet surface (Hu et al., 2010). The decrease in the rate of lipid digestion in the presence of relatively small amounts of alginate has been thus mainly attributed to its ability to bind calcium ions. However, other physicochemical mechanisms may have occurred, e.g., formation of a coating around the lipid droplets that inhibited lipase from reaching their surface and binding multivalent cations required for lipase activity (Hu et al., 2010).

Another interesting result is that the percentage of FFA absorbed in small intestine (jejunum and ileum) is higher for lactoferrin-stabilized emulsions comparing to lactoferrin/alginate- stabilized emulsions (64.84 ± 10.27% and 48.82 ± 12.69%, respectively), which suggests that calories absorption in small-intestine is reduced by the addition of the alginate coating. This means that, although FFA were released almost in the same extent for both primary and secondary emulsions (92.34 ± 10.69 and 81.98 ± 18.67, respectively), the presence of alginate hindered FFA absorption.

These results suggest that alginate coating may be useful for controlling the rate of lipid digestion and the FFA adsorption within the gastrointestinal tract, but it does not prevent the adsorption of lipase to the lipid droplet and their consequent digestion. If the objective of the multilayer system is to increase the bioavailability of a water-insoluble bioactive compounds

171

Pinheiro, A. C. (2013)

(such as curcumin), it is necessary for the carrier oil surrounding them to be digested first, releasing the bioactive compounds so they can be taken up by mixed micelles (Tso, 2000).

7.3.2.3 Influence of alginate on curcumin bioavailability

Figure 7.7 shows curcumin bioavailability, measured as curcumin concentration in the mixed micelle phase, in the filtrate samples of jejunal and ileal compartments as a function of time for primary and secondary emulsions.

a) b)

Figure 7.7 Total bioavailability of curcumin during gastrointestinal transit of curcumin nanoemulsions stabilized by lactoferrin (a) and lactoferrin/alginate (b) in the dynamic in vitro digestion model. Total bioavailability of curcumin is shown in the filtrate samples of jejunal () and ileal () compartments at 1 h intervals and expressed as % related to curcumin intake. The results are presented as the mean values of triplicate experiments.

Curcumin bioavailability tends to increase with digestion time, and this behaviour can be attributed to the formation of digestion products that have the ability to form mixed micelles capable of solubilizing highly lipophilic components such as curcumin.

For both lactoferrin and lactoferrin/alginate-stabilized emulsions the relative amount of curcumin was higher in the filtrate sample of jejunal compartment compared to the ileal compartment, however lactoferrin/alginate-stabilized emulsions present a significantly (p<0.05) higher curcumin bioavailability in ileal filtrate at all times. This result is in agreement with FFA release results (Figure 7.6), i.e. for lactoferrin/alginate-stabilized emulsions the higher curcumin bioavailability observed in ileal filtrate was related with the higher amount of FFA found in this fraction. The access of lipases to encapsulated lipids is a critical step in determining the bioactivity of bioactive compounds, once the digestion products from TAG digestion (DG, MG, and FFA) may themselves become part of the mixed micelles, thereby increasing the overall solubilizing capacity of the small intestine, increasing the bioavailability of encapsulated bioactive compounds (Porter et al., 2007).

172

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

For both emulsions, the total curcumin bioavailability is relatively low (10.77 ± 1.19% and 11.11 ± 1.52% for lactoferrin and lactoferrin/alginate-stabilized emulsions, respectively) with most of the curcumin being recovered in the ileal deliveries and residues in the dynamic digestion model after the digestion was completed.

7.4 CONCLUSIONS

A dynamic digestion model comprising the stomach, duodenum, jejunum and ileum was developed and used to evaluate the impact of alginate coating on the digestion of curcumin nanoemulsions stabilized by lactoferrin. The interfacial characteristics of curcumin nanoemulsions had a significant impact on their physicochemical stability within the simulated GI tract: lactoferrin-stabilized emulsions were stable in the stomach but aggregated in the small intestine, while lactoferrin/alginate-stabilized emulsions were unstable under both gastric and small intestinal conditions. The overall extent of lipid digestion was fairly similar for both primary and secondary emulsions, which suggests that alginate coating did not prevent lipid digestion. However, for lactoferrin-stabilized emulsions the highest fraction of FFA was obtained in the jejunal filtrate whereas for lactoferrin/alginate-stabilized emulsions the higher fraction of FFA was found in the ileal filtrate. Also, the higher curcumin bioavailability observed in ileal filtrate of lactoferrin/alginate-stabilized emulsions is related with the higher amount of FFA found in this fraction. These results suggest that alginate coating may be able to control the rate of lipid digestion and FFA adsorption within the GI tract, but the encapsulated lipid is digested at the same extent, releasing the lipophilic bioactive compound, which is then taken up by mixed micelles.

This work contributed to an improved understanding of how multilayer emulsions behave within the human GI tract and this knowledge can be useful for the optimization of delivery systems that improve the physicochemical stability of emulsions in food products, while still releasing encapsulated lipophilic components in the GI tract.

173

Pinheiro, A. C. (2013)

7.5 REFERENCES

Ahmed, K., Li, Y., McClements, D. J., & Xiao, H. (2012). Nanoemulsion- and emulsion- based delivery systems for curcumin: Encapsulation and release properties, Food Chemistry 132(2), 799-807. Balcão, V. M., Costa, C. I., Matos, C. M., Moutinho, C. G., Amorim, M., Pintado, M. E., Gomes, A. P., Vila, M. M., & Teixeira, J. A. (2013). Nanoencapsulation of bovine lactoferrin for food and biopharmaceutical applications, Food Hydrocolloids 32(2), 425-431. Farnaud, S., & Evans, R. W. (2003). Lactoferrin—a multifunctional protein with antimicrobial properties, Molecular Immunology 40(7), 395-405. Hu, M., Li, Y., Decker, E. A., & McClements, D. J. (2010). Role of calcium and calcium- binding agents on the lipase digestibility of emulsified lipids using an in vitro digestion model, Food Hydrocolloids 24(8), 719-725. Hur, S. J., Decker, E. A., & McClements, D. J. (2009). Influence of initial emulsifier type on microstructural changes occurring in emulsified lipids during in vitro digestion, Food Chemistry 114(1), 253-262. Lee, K. Y., & Mooney, D. J. (2012). Alginate: Properties and biomedical applications, Progress in Polymer Science 37(1), 106-126. Li, Y., Hu, M., Du, Y., Xiao, H., & McClements, D. J. (2011). Control of lipase digestibility of emulsified lipids by encapsulation within calcium alginate beads, Food Hydrocolloids 25(1), 122-130. Li, Y., Hu, M., Xiao, H., Du, Y., Decker, E. A., & McClements, D. J. (2010). Controlling the functional performance of emulsion-based delivery systems using multi-component biopolymer coatings, European Journal of Pharmaceutics and Biopharmaceutics 76(1), 38-47. Maurstad, G., Mørch, Y. A., Bausch, A. R., & Stokke, B. T. (2008). Polyelectrolyte layer interpenetration and swelling of alginate–chitosan multilayers studied by dual wavelength reflection interference contrast microscopy, Carbohydrate Polymers 71(4), 672-681. McClements, D. J. (2005). Theoretical Analysis of Factors Affecting the Formation and Stability of Multilayered Colloidal Dispersions, Langmuir 21(21), 9777-9785. McClements, D. J. (2010). Design of Nano-Laminated Coatings to Control Bioavailability of Lipophilic Food Components, Journal of Food Science 75(1), R30-R42. McClements, D. J. (2013). Edible lipid nanoparticles: Digestion, absorption, and potential toxicity, Progress in Lipid Research 52(4), 409-423. McClements, D. J., & Xiao, H. (2012). Potential biological fate of ingested nanoemulsions: influence of particle characteristics, Food & Function 3(3), 202-220.

174

Chapter 7. Influence of interfacial composition on the in vitro digestibility of curcumin nanoemulsions stabilized by biopolymer emulsifiers

Minekus, M., & Havenaar, R. (1996) In vitro model of an in vivo digestive tract, United States. Minekus, M., Jelier, M., Xiao, J.-z., Kondo, S., Iwatsuki, K., Kokubo, S., Bos, M., Dunnewind, B., & Havenaar, R. (2005). Effect of Partially Hydrolyzed Guar Gum (PHGG) on the Bioaccessibility of Fat and Cholesterol, Bioscience, Biotechnology, and Biochemistry 69(5), 932- 938. Mun, S., Decker, E. A., & McClements, D. J. (2007). Influence of emulsifier type on in vitro digestibility of lipid droplets by pancreatic lipase, Food Research International 40(6), 770- 781. Pinsirodom, P. P., K. L. (2005) Lipase assays, In Handbook of Food Analytical Chemistry (Wrolstad, R. E., Ed.), pp 371–383, Wiley: Hoboken, NJ. Porter, C. J. H., Trevaskis, N. L., & Charman, W. N. (2007). Lipids and lipid-based formulations: optimizing the oral delivery of lipophilic drugs, Nature Reviews Drug Discovery 6, 231-248. Reis, P., Raab, T., Chuat, J., Leser, M., Miller, R., Watzke, H., & Holmberg, K. (2008). Influence of Surfactants on Lipase Fat Digestion in a Model Gastro-intestinal System, Food Biophysics 3(4), 370-381. Salminen, H., Herrmann, K., & Weiss, J. (2013). Oil-in-water emulsions as a delivery system for n-3 fatty acids in meat products, Meat Science 93(3), 659-667. Salvia-Trujillo, L., Qian, C., Martín-Belloso, O., & McClements, D. J. (2013). Influence of particle size on lipid digestion and β-carotene bioaccessibility in emulsions and nanoemulsions, Food Chemistry 141(2), 1472-1480. Shimoni, G., Shani Levi, C., Levi Tal, S., & Lesmes, U. (2013). Emulsions stabilization by lactoferrin nano-particles under in vitro digestion conditions, Food Hydrocolloids 33(2), 264- 272. Thanasukarn, P., Pongsawatmanit, R., & McClements, D. J. (2006). Utilization of layer- by-layer interfacial deposition technique to improve freeze–thaw stability of oil-in-water emulsions, Food Research International 39(6), 721-729. Tokle, T., Lesmes, U., Decker, E. A., & McClements, D. J. (2012). Impact of dietary fiber coatings on behavior of protein-stabilized lipid droplets under simulated gastrointestinal conditions, Food & Function 3(1), 58-66. Tokle, T., Lesmes, U., & McClements, D. J. (2010). Impact of Electrostatic Deposition of Anionic Polysaccharides on the Stability of Oil Droplets Coated by Lactoferrin, Journal of Agricultural and Food Chemistry 58(17), 9825-9832.

175

Pinheiro, A. C. (2013)

Tokle, T., & McClements, D. J. (2011). Physicochemical properties of lactoferrin stabilized oil-in-water emulsions: Effects of pH, salt and heating, Food Hydrocolloids 25(5), 976- 982. Tso, P. (2000) Overview of digestion and absorption, In Biochemical and physiological aspects of human nutrition (K.D. Crissinger , M. H. S., Ed.), pp 75–90, W.B. Saunders Co, Philadelphia. Zhou, J., Romero, G., Rojas, E., Ma, L., Moya, S., & Gao, C. (2010). Layer by layer chitosan/alginate coatings on poly(lactide-co-glycolide) nanoparticles for antifouling protection and Folic acid binding to achieve selective cell targeting, Journal of Colloid and Interface Science 345(2), 241-247.

176

SECTION IV POLYMERIC NANOPARCTICLES

8 CHAPTER 8. CHITOSAN/FUCOIDAN

MULTILAYER NANOCAPSULES FOR CONTROLLED RELEASE OF BIOACTIVE COMPOUNDS CHITOSAN/FUCOIDAN MULTILAYER NANOCAPSULES FOR CONTROLLED RELEASE OF BIOACTIVE COMPOUNDS

8.1 Introduction...... 181

8.2 Materials and Methods ...... 182

8.3 Results and Discussion ...... 187

8.4 Conclusion ...... 196

8.5 References ...... 198

Pinheiro, A. C. (2013)

The results presented in this Chapter were adapted from:

180

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

8.1 INTRODUCTION

Nanoencapsulation systems exhibit high potential as carriers of bioactive substances. Their subcellular size allows relatively higher intracellular uptake and their permanence in circulation for longer, therefore extending their biological activity compared to microcapsules (Mora- Huertas et al., 2010). Also, nanoencapsulation may be beneficial regarding improved stability and protection capability of labile substances against degradation factors (Preetz et al., 2008). Layer-by-Layer (LbL) deposition technique is one of the most powerful methods to create multilayer nanocapsules. LbL assembly is based on the electrostatic interaction between oppositely charged polyelectrolytes alternatively adsorbed onto an appropriate template (Decher and Schlenoff, 2003). Nanocapsules prepared through LbL technique can be specially engineered with controlled sizes, composition, porosity, stability, surface functionality and colloidal stability and can be used as carriers for bioactive compounds. Also, the step-wise formation of multilayer nanocapsules allows introducing multiple functionalities (Johnston et al., 2006).

Core-shell nanoparticles can be produced through the deposition of polyelectrolyte layers onto colloidal nanoparticles; by subsequently removing the core, by dissolution or decomposition, from the core-shell structure, it is possible to obtain hollow nanocapsules with different properties, depending on the polyelectrolytes used (Cuomo et al., 2010). Solid particles such as polystyrene, silica and CaCO3 are the most often used sacrificed templates for formation of capsules (Szczepanowicz et al., 2010). In addition, diameter, membrane thickness and permeability of hollow capsules prepared by LbL assembly can be easily controlled (Caruso et al., 1998).

Multilayer nanocapsules have promising applications in the release of bioactive compounds in the pharmaceutical and food industries; however for these applications to be possible, biofunctionality and the use of non-toxic materials are the most fundamental conditions to be met. Therefore, natural polyelectrolytes such as chitosan and fucoidan are competitive candidates as materials for the formation of multilayer nanocapsules. Chitosan is a cationic polysaccharide obtained by deacetylation of chitin which is the major constituent of exoskeleton of crustaceous animals. Chitosan is nontoxic, biodegradable, and biocompatible and has intrinsic antimicrobial activity, inhibiting the growth of a wide variety of bacteria (Helander et al., 2001; Shahidi et al., 1999). Fucoidan, an anionic sulphated polysaccharide found in brown algae display diverse biological activities, including antioxidant, anticoagulant, antibacterial, antiviral, anti-inflammatory and antitumor effects (Kusaykin et al., 2008).

181

Pinheiro, A. C. (2013)

Several techniques have been proposed to incorporate bioactive compounds within nanocapsules (Johnston et al., 2006). In general, the bioactive compound can be loaded during the preparation of capsules (adsorption of the bioactive compound to the surface of a particle that can be subsequently coated using the LbL technique) or after the formation of the capsules (diffusion of bioactive compound from the surrounding medium in which the capsules are dispersed into the capsule) (Shu et al., 2010).

Poly-L-lysine (PLL) is a short cationic polypeptide composed of 27–33 identical L-lysine residues and is industrially produced by fermentation using Streptomyces albulus ssp. Lysinololymerus (Najjar et al., 2007). This polypeptide exhibits a strong antimicrobial activity, is stable at high temperatures and under both acidic and alkaline conditions and is safe for human consumption (Najjar et al., 2007; Yoshida and Nagasawa, 2003); therefore it is widely used as natural food preservative (Yamanaka and Hamano, 2010). PLL inhibits the grow of a wide spectrum of microorganisms including Gram-positive and Gram-negative bacteria; it also has anti-phage activity (Hamano et al., 2007). Its mechanism of action against microbial growth is the electrostatic adsorption to the cell surface and subsequent interference with cell membranes (Najjar et al., 2007).

In this work, biodegradable hollow nanocapsules were developed through LbL assembly of chitosan and fucoidan. Nanocapsules were built through the alternate deposition of 10 chitosan/fucoidan layers on polystyrene (PS) nanoparticles (diameter ≈ 100 nm), used as templates, followed by removal of the PS core. PLL was entrapped in the nanocapsules and its loading and release behaviour was evaluated.

8.2 MATERIALS AND METHODS

8.2.1 Materials

Dispersion of polystyrene nanoparticles with a diameter of 93.8 ± 1.5 nm and a ζ-potential of - 44.8 ± 0.3 mV was obtained from Polysciences, Inc. (Polybead® Polystyrene 0.10 µm Microspheres, Warrington, PA, USA). Fucoidan with a MW = 57,260 was obtained from Sigma- Aldrich (Fucoidan from Fucus vesiculosus, St. Louis, MO, USA) and chitosan (deacetylation degree ≥ 95%) was purchased from Golden-Shell Biochemical CO., LTD (Zhejiang, China). Lactic acid (90%) was obtained from Acros Organics (Geel, Belgium). Tetrahydrofuran (THF) and poly- L-lysine hydrobromide (mol. Wt 1,000-5,000) were purchased from Sigma-Aldrich (St. Louis, MO, USA).

182

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

8.2.2 Polysaccharides analyses

-1 Fucoidan sample was hydrolyzed with H2SO4 1 mol.L at 100 °C during 2.5 h according to other authors (Selvendran et al., 1979). Neutral sugars were converted to alditol acetates, as previously described (Coimbra et al., 1996) and uronic acids were quantified through the 3- phenylphenol colorimetric method (Blumenkrantz and Asboe-Hansen, 1973), modified by Coimbra et al. (1996).

Chitosan sample was hydrolyzed with HCl 6 mol.L-1 at 100 °C during 20 h, followed by glucosamine conversion to alditol acetates by reduction with sodium borohydride and acetylation, as previously described (Coimbra et al., 1996). Alditol acetate derivatives were analysed by GC-FID.

8.2.3 Antioxidant activity of fucoidan

The free-radical scavenging capacity of fucoidan was analysed using the 1,1-diphenyl-2- picrylhydrazyl (DPPH) test according to the method of Blois (1958), with some modifications. BHT and BHA were used as positive controls. Briefly, 0.2 mL of MeOH and 0.3 mL of the sample dissolved in MeOH (concentrations ranging from 0.05 to 1.0 mg.mL-1) were mixed in a 10 mL test tube. DPPH (2.5 mL of 75 μmol.L-1 in MeOH) were then added to achieve a final volume of 3.0 mL. The solution was kept at room temperature for 30 min and the absorbance was measured at 517 nm (Blois, 1958).

The DPPH scavenging effect was calculated as follows:

( ) ( ) Equation 8.1

where A0 is the absorbance at 517 nm of DPPH without sample, A is the absorbance at 517 nm of sample and DPPH and Ab is the absorbance at 517 nm of sample without DPPH.

8.2.4 Antimicrobial activity of chitosan

The antibacterial activity of chitosan was tested against two bacterial strains: Staphylococcus aureus (Gram-positive) and Escherichia coli (Gram-negative) by the disc agar diffusion test according to other authors (Bauer et al., 1966). Briefly, the chitosan solution (1 mg.mL-1) was absorbed in sterilized filter paper discs (of 0.6 cm in diameter) and placed on the lawn cultures of S. aureus and E. coli. The agar plates were incubated for 24 h at 37 °C and diameters of the inhibitory zone of clearance (cm) surrounding the discs were measured to estimate the antimicrobial activity. Sterile distilled water was used as control.

183

Pinheiro, A. C. (2013)

8.2.5 Evaluation of the interactions between chitosan and fucoidan

8.2.5.1 Quartz crystal microbalance

The adsorption behaviour of fucoidan and chitosan was evaluated using a quartz crystal microbalance (QCM 200, purchased from Stanford Research Systems, SRS, USA), equipped with AT-cut quartz crystals (5 MHz) with optically flat polished chrome/gold electrodes on contact and liquid sides.

Before carrying out the experiments, the crystal was cleaned by successive sonications (40 KHz, 30 min) in ultra-pure water, ethanol (Panreac, Spain) and ultra-pure water, followed by drying with a gentle flow of nitrogen. Adsorption measurements were performed by alternate immersion of the crystal in chitosan (pH 3.0) and fucoidan (pH 7.0) solutions, for 15 min. The variations of the resonance frequency (ΔF) were simultaneously measured as a function of time and the analyses were performed at 20 °C, in triplicate.

8.2.6 Preparation of nanocapsules

The chitosan/fucoidan multilayers were assembled on the polystyrene nanoparticles by LbL deposition technique. The multilayer build-up was carried out according to other work (Ye et al., 2005) with some modifications. Briefly, 1 mL of chitosan solution (1 mg.mL-1, pH 3) in 1% of lactic acid was added drop-by-drop to 1 mL of polystyrene particles’ suspension (0.5% v/v), under gentle agitation and the mixture was incubated for 15 min. The excess of polysaccharide was removed by two repeated cycles of centrifugation (14000 rpm, 20 min)/washing/redispersion in water/sonication (40 KHz, 60 min). The washing steps after every deposition were performed to minimize occurrence of artefacts, i.e. polyelectrolyte aggregates formed in solution rather than on the core template, whereas the sonication step was implemented in order to disperse the nanocapsules in solution and minimize their aggregation. This procedure was repeated, this time using fucoidan (1 mg.mL-1, pH 7) as polyelectrolyte. The alternate deposition of chitosan and fucoidan was repeated until the deposition of 10 layers.

In order to remove the core of polystyrene particles, chitosan/fucoidan nanoparticles were treated with THF for 12 h. The suspension was then centrifuged at 14000 rpm for 15 min to remove THF. This procedure was repeated two times to assure complete removal of the polystyrene core and the resulting nanocapsules were washed three times with water, centrifuged and redispersed in distilled water.

184

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

8.2.7 Characterization of nanoparticles and nanocapsules

8.2.7.1 ζ-potential measurements

The ζ-potential of coated polystyrene nanoparticles was determined by dynamic light scattering (DLS; Zetasizer Nano ZS, Malvern Instruments, UK). Each sample was analysed in a folded capillary cell. The ζ-potential values are the average of nine successive measurements.

8.2.7.2 Morphology

The morphology of nanoparticles and nanocapsules was evaluated by scanning electron microscopy (SEM) (Nova NanoSEM 200, the Netherlands) with an accelerating voltage from 10 to 15 kV) and by transmission electron microscopy (TEM) (EM 902A, ZEISS, Germany).

SEM samples were prepared by depositing the suspension of nanoparticles or nanocapsules on Si slides. After drying the samples, they were coated with a thin gold layer (thickness of about 10 nm). TEM samples were prepared by depositing the same suspensions on a carbon-coated copper grid. Before being analysed, samples were air-dried.

8.2.7.3 Fourier transform infrared (FTIR) spectroscopy

In order to confirm the removal of the polystyrene core, FTIR analyses were carried out with a Perkin Elmer 16 PC spectrometer (Perkin Elmer, Boston, MA, USA) equipped with an ATR probe in the wavenumber region of 600–4000 cm−1 using 16 scans for each sample. The nanocapsules were dried and then embedded in KBr pellets.

8.2.8 PLL loading

A PLL solution (with concentration ranging from 0.25 to 2 mg.mL-1) was added to a suspension of polystyrene nanoparticles (0.5% v/v) and allowed to interact for 15 min.

The PLL encapsulation efficiency (EE) and loading capacity (LC) of the nanocapsules were determinate using the following equations (Shu et al., 2010):

Equation 8.2

Equation 8.3

Free and bound PLL were separated by centrifugation (14000 rpm, 30 min) using Amicon Ultra (10 K) (Millipore, Ireland). The amount of free PLL in the supernatant was quantified by

185

Pinheiro, A. C. (2013)

measuring the absorbance at 215 nm (absorbance peak) (Elisa Biotech Synergy HT, Biotek, USA).

8.2.9 Determination of PLL release kinetics

The in vitro release kinetics of PLL was performed by a dialysis method. The PLL-loaded nanocapsule suspension (1.5 mL) was added into a dialysis membrane (molecular weight cut- off 25 kDa; Cellu-Sep H1, Membrane filtration products, USA) that was subsequently placed into 40 mL of buffer solution (phosphate buffer for pH 7 and KCl-HCl buffer for pH 2) under magnetic stirring. At appropriate time intervals, 0.5 mL of supernatant were taken and 0.5 mL of fresh buffer were added to keep the volume of the release medium constant. The amount of PLL released from nanocapsules was evaluated by measuring the absorbance at 215 nm (absorbance peak) (Elisa Biotech Synergy HT, Biotek, USA). All release tests were run at least in triplicate. In order to evaluate the release mechanism of PLL from chitosan/fucoidan nanocapsules a kinetic model that accounts for both Fickian and Case II transport effects in hydrophilic matrices, the linear superimposition model (LSM), was applied (Berens and Hopfenberg, 1978):

Equation 8.4

where Mt,F and Mt,R are the contributions of the Fickian and relaxation processes, respectively, at time t. The purely Fickian process is described by:

[ ∑ ( )] Equation 8.5

where M∞,F is the compound release at equilibrium and kF is the Fickian diffusion rate constant. Equation 8.5 can be simplified using the first term of the Taylor series (Chapter 4). As for polymer relaxation, it is driven by the swelling ability of the polymer and is then related to the dissipation of stress induced by the entry of the penetrant and can be described as a distribution of relaxation times, each assuming a first order-type kinetic equation (Berens and Hopfenberg, 1978).

∑ [ ( )] Equation 8.6

here, are the contributions of the relaxation processes for compound release and are the relaxation ith rate constants. For most cases, there is only one main polymer relaxation that influences transport and thus the above equation can be simplified using .

186

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

Therefore, the linear superimposition model for compound release from nanocapsules can be described by:

[ ( ) ] ( )[ ( )] Equation 8.7

where X is the fraction of compound released by Fickian transport. The experimental results were analysed by fitting Equation 8.5 (Fick’s second law) and Equation 8.7 (linear superimposition model) in order to assess the transport mechanism involved in the PLL release from nanocapsules at two different pH values.

8.2.10 Statistical Procedures

8.2.10.1 Non-Linear Regression Analysis

The equations were fitted to data by non-linear regression, using a package of STATISTICA v 7.0 (Statsoft. Inc, USA). The Levenberg-Marquadt algorithm for the least squares function minimization was used. The quality of the regressions was evaluated on the basis of the determination coefficient, R2, the squared root mean square error, RMSE (i.e., the square root of the sum of the squared residues (SSE) divided by the regression degrees of freedom) and residuals visual inspection for randomness and normality. R2 and SSE were obtained directly from the software. The precision of the estimated parameters was evaluated by the Standardised Halved Width (SHW %), which was defined as the ratio between the 95% Standard Error (obtained from the software) and the value of the estimate.

8.3 RESULTS AND DISCUSSION

8.3.1 Polyelectrolytes characterization

8.3.1.1 Composition

The bioactive properties of polysaccharides usually depend on their chemical structure. The chemical composition of fucoidan is listed in Table 8.1.

Regarding the monosaccharide composition, the fucoidan sample is mainly composed of fucose and uronic acids. Minor amounts of other monosaccharide constituents such as galactose, xylose, mannose, glucose and rhamnose were also detected.

187

Pinheiro, A. C. (2013)

Table 8.1 Composition of the fucoidan used to prepare the multilayer nanocapsules

-1 Monosaccharide composition of fucoidan (µgmonosaccharide.mgsample ) Rha Fuc Xyl Man Gal Glc Uronic acids

1.1 ± 0.0 402.1 ± 9.8 29.8 ± 2.3 5.5 ± 3.3 36.0 ± 4.1 2.1 ± 0.3 91.7 ± 4.1

Other authors reported that fucoidan from F. vesiculosus is mainly composed be fucose, sulphate and ash (Black, 1952), therefore the low percentage of sugars found in this sample (about 58%) is probably due to the presence of sulphate and ash.

-1 The chitosan is mainly composed by D-glucosamine (408.26 ± 6.11 µg.mgsample ), with only 7 ± 2% of N-acetyl-D-glucosamine.

8.3.1.2 Bioactive properties

Chitosan and fucoidan are known for their bioactive properties (mainly antimicrobial and antioxidant activities, respectively), however, a preliminary study was conducted in order to verify the functional properties of these polyelectrolytes at concentrations as low as the ones typically used for the formation of layer-by-layer nanocapsules (Sun et al., 2005; Ye et al., 2005).

The antioxidant activity of fucoidan was investigated by the DPPH scavenging assay (Figure 8.1).

Figure 8.1 Scavenging effects on DPPH radical of fucoidan () compared with the positive controls BHA () and BHT ().

The reduction of DPPH (1,1-diphenyl-2-picrylhydrazyl), a stable free radical, to DPPH-H after encountering a substance acting as a hydrogen atom donor can be used as an indicator of antioxidant activity, once when the stable nonradical form of DPPH is formed, a simultaneous change of solution’s colour occurs from violet to pale yellow (Szabo et al., 2007).

188

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

The DPPH scavenging activity of fucoidan is concentration-dependent, but this polysaccharide exhibits a strong effect on inhibiting the formation of DPPH-H at concentration between 0.1

-1 -1 -1 and 1 mg.mL (IC50 of 0.23 mg.mL ). Moreover, at a concentration of 1 mg.mL , fucoidan reaches a scavenging activity of 90.74%. This value of scavenging activity is close to the value obtained by other authors for crude fucoidan extracted from Undaria pinnatifida (86.80% at 1 mg.mL-1) (Mak et al., 2013)

These results show that at a concentration as low as 1 mg.mL-1, fucoidan presents a very strong antioxidant activity, therefore this was the concentration chosen for nanocapsules formation.

Chitosan was also tested in order to assess its antimicrobial activity at the same concentration of 1 mg.mL-1. Therefore, the antimicrobial activity of a chitosan solution against S. aureus and E. coli was evaluated using the disk diffusion technique (Figure 8.2).

Figure 8.2 Diameter of the inhibition zones (cm) of S. aureus and E. coli produced by a solution of 1mg.mL-1 of chitosan in the disk diffusion assay. Error bars represent the standard deviation obtained in the triplicate experiments.

At a concentration of 1 mg.mL-1, chitosan exhibited antimicrobial activity against both S. aureus (Gram-positive) and E. coli (Gram-negative), although this activity was significantly (p<0.05) higher against S. aureus. Chitosan has high antimicrobial activity against a wide range of microorganisms, including fungi, Gram-positive and Gram-negative bacteria, however, literature shows that chitosan is more active against Gram-positive than against Gram- negative bacteria (Dutta et al., 2012). The microbial target of protonated chitosan’s actions would be the cytoplasmic membrane and cellular damage can lead to the disruption of the cellular integrity of the membrane. In the case of Gram-negative bacteria, the existence of an outer membrane protects the cells, once it acts as a barrier preventing chitosan from reaching the cytoplasmic membrane in the same extent as in the Gram-positive bacteria (Coma et al., 2003).

189

Pinheiro, A. C. (2013)

These results show that 1 mg.mL-1 is also a suitable concentration for the chitosan solution, once this cationic polysaccharide exhibits a relevant antimicrobial activity against both Gram- positive and Gram-negative bacteria.

8.3.2 Interactions between chitosan and fucoidan

Since LbL deposition is primarily governed by electrostatic interactions and in order to guarantee a correct deposition of chitosan and fucoidan layers, it is necessary to assure that these polyelectrolytes exhibit opposite charge. Once these polysaccharides are weak polyelectrolytes, their dissociation degree depends strongly on the solution’s pH. The chitosan solution was found to have a maximum positive charge (62.5 ± 4.9 mV) at pH 3 and the fucoidan solution exhibited a maximum negative charge (-48.4 ± 1.1 mV ) at pH 7.

Also before LbL deposition of chitosan and fucoidan on the polystyrene nanoparticles, the ability of these polyelectrolytes to form layers was evaluated through quartz crystal microbalance analysis. Real-time build-up of chitosan and fucoidan nanolayered assemblies is reported in Figure 8.3.

Figure 8.3 Change in frequency as a function of time for chitosan (C, 1 mg.mL-1 (w/v), pH 3) and fucoidan (F, 1 mg.mL-1 (w/v), pH 7) adsorption.

Figure 8.3 shows the consecutive decrease of frequency after the alternate immersion of the crystal in chitosan and fucoidan solutions, which means that mass is being subsequently deposited and confirms the electrostatic interaction between these polyelectrolytes and the consequent formation of a stable multilayer structure. Also, from Figure 8.3 it can be observed that the adsorption of fucoidan results in higher changes in frequency compared with the adsorption of chitosan, which is possibly due to the more pronounced hydrophilic character of fucoidan. Hydrophilic surfaces, such as fucoidan layer, are rough and can entrap solvents, which results in a more viscoelastic multilayered system and contributes to the mass increase

190

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

sensed by the QCM and therefore to higher frequency changes. On the other hand, hydrophobic surfaces, such as chitosan layer, often do not wet and air can be entrapped, which results in smaller measured mass. A similar behaviour was observed for κ- carrageenan/chitosan (Chapter 3) and for alginate/chitosan multilayers (Martins et al., 2010).

The mass corresponding to the observed decrease of frequency also accounts for the water entrapped in the adsorbed layer (Indest et al., 2008), therefore a correct mass calculation can only be obtained for rigid thin films and was not performed in this work.

8.3.3 Characterization of core-shell nanocapsules

Once core-shell nanoparticles are the precursors of hollow nanocapsules, it is important to verify the successive deposition of polyelectrolytes layers on the polystyrene template and to follow shell formation.

8.3.3.1 ζ-potential measurements

The alternate deposition of chitosan and fucoidan on the polystyrene nanoparticles was monitored measuring ζ-potential values at each assembly step (Figure 8.4).

Figure 8.4 ζ-potential change of polystyrene nanoparticles after each chitosan (even layers) or fucoidan (odd layers) deposition. The first point refers to the bare templates.

Polystyrene nanoparticles templates have a negative charge due to the presence of sulphate ester groups (Sun et al., 2005), so the initial step of polysaccharide multilayer construction consisted in the adsorption of the positively charged chitosan solution. The ζ-potential alternated between positive (more than 30 mV) and negative (less than -30 mV) values after the deposition of chitosan and fucoidan, respectively, indicating the successfully deposition of these polysaccharides on the polystyrene nanoparticles up to 10 layers. Similar results were

191

Pinheiro, A. C. (2013)

obtained by other authors for chitosan and alginate assembled on polystyrene (Ye et al., 2005) and SiO2–NH2 templates (Liu et al., 2012) .

The ζ-potential also gives considerable information on the stability of nanoparticles’ suspensions, once particles with ζ-potential values higher than 30 mV or lower than -30 mV are generally considered to be stable because of the strong repulsion forces that prevent the occurrence of aggregation phenomena among them (Liu et al., 2012). Therefore, the absolute values of ζ-potential are large enough to indicate that the nanoparticles of chitosan/fucoidan are stable, due to the strong electrostatic repulsion existing among them. Also, the magnitude of ζ-potential values did not decrease with the subsequent layer deposition, suggesting that the mobility of nanoparticles did not decrease and therefore no significant aggregation occurred (Ye et al., 2005).

8.3.4 Characterization of hollow nanocapsules

After the removal of the polystyrene core by repeated dipping in THF, hollow capsules can be obtained. The hollow capsules usually have the shape of the template and their stability and permeability features depend on the polyelectrolyte complex used (Cuomo et al., 2010).

8.3.4.1 Fourier transform infrared (FTIR) spectroscopy

In order to confirm the removal of the polystyrene core from chitosan/fucoidan nanoparticles, FTIR spectroscopy was carried out. IR spectra of the polystyrene nanoparticles, chitosan/fucoidan core-shell nanoparticles and hollow chitosan/fucoidan nanocapsules can be seen in Figure 8.5.

Figure 8.5 FTIR spectra of polystyrene nanoparticles (─), chitosan/fucoidan core-shell nanoparticles (─) and hollow chitosan/fucoidan nanocapsules (─).

192

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

The characteristic bands of benzene residue in polystyrene appears at 3000-3108 cm-1, 762 cm- 1 and 703 cm-1 for both polystyrene nanoparticles and chitosan/fucoidan core-shell nanoparticles, indicating the presence of the polystyrene core. However, after core removal by dissolution on THF, the characteristic polystyrene peaks disappear, confirming that the polystyrene core has been completely removed and that hollow nanocapsules have been formed.

8.3.4.2 Morphology

Morphological characteristics of polystyrene nanoparticles and hollow chitosan/fucoidan nanocapsules were examined using TEM and SEM (Figure 8.6 and 8.7, respectively).

Microscopy images of the polystyrene template (Figure 8.6 a and Figure 8.7 a) show that the particles are spherical and smooth and present an average diameter of 90 nm. After the deposition of chitosan and fucoidan layers, the dissolution of the supporting core by repeated dipping in THF resulted in the shrinking of the nanostructure and on a consequent decrease of its size (to about 50 nm), but they were found to maintain their spherical shape without rupturing of the wall as can be seen in Figure 8.6 b) and Figure 8.7 b). Nanocapsules’ diameter decreased due to the disorganization of the supporting core that led to a skink of the cyst wall (Liu et al., 2012). Contrary to the behaviour observed by other authors (Ye et al., 2005), the swelling of nanocapsules (i.e. increase of size) during the polystyrene core removal did not occur.

From Figure 8.7 it can also be observed that nanocapsules exhibit a reduced electron density compared with the core-shell nanoparticles, which is another indication of their hollow structure.

a) b)

Figure 8.6 SEM images of polystyrene nanoparticles (a) and hollow chitosan/fucoidan nanocapsules (b).

193

Pinheiro, A. C. (2013)

a) b)

Figure 8.7 TEM images of polystyrene nanoparticles (a) and hollow chitosan/fucoidan nanocapsules (b).

8.3.5 Encapsulation and in vitro release of a model bioactive compound

Chitosan/fucoidan nanocapsules present per se bioactive potential due to the properties of their constituents; however, in order to evaluate the ability of these nanocapsules to encapsulate and release a bioactive compound, PLL was incorporated. Two methods for PLL encapsulation were evaluated: adsorption of PLL in the polystyrene nanoparticles that were subsequently coated with 10 layers of chitosan/fucoidan and the core was removed (method 1) and diffusion of PLL into the capsules’ interior after their formation (method 2). For the same PLL initial concentration (1 mg.mL-1), the higher encapsulation efficiency (EE) was obtained using method 1 (EE = 4 5.1 ± 1.5% against EE = 26.01 ± 5.5% for method 2). The loadings achieved are generally higher when the bioactive compound is loaded during the preparation of capsules due to the high surface area of the capsule and due to the fact that, when the bioactive compound is loaded after capsules’ formation, the maximum concentration of bioactive inside the capsules is often limited to the concentration in the solution (Johnston et al., 2006). Being so, method 1 was chosen as the method for PLL encapsulation. The mechanism of PLL encapsulation is based on the strong electrostatic interactions that are established between negatively charged polystyrene nanoparticles and positively charged PLL (ζ-potential = 26.0 ± 1.5 mV).

The encapsulation efficiency and loading capacity of nanocapsules as a function of PLL concentrations are represented in Figure 8.8.

194

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

Figure 8.8 Effect of PLL concentration on encapsulation efficiency () and loading capacity () of nanocapsules.

The encapsulation efficiency (EE) and the loading capacity (LC) of the chitosan/fucoidan nanocapsules strongly depend on the initial PLL concentration used. The highest EE and LC values were observed at a PLL concentration of 1 mg.mL-1 and this was the concentration chosen for subsequent in vitro release experiments.

The mechanisms of PLL release from chitosan/fucoidan nanocapsules were evaluated at 37 °C (temperature within the human body) and at two different pH values: 2 (pH of the stomach) and 7 (pH of the small intestine).

From Figure 8.9 a) (profile release of PLL at pH 2) and b) (profile release of PLL at pH 7) it can be seen that nanocapsules present a pH-dependent release pattern. In order to investigate the mechanism of PLL release from nanocapsules, the exprimental results were analysed by fitting Fick’s second law (Equation 8.5) and LSM (Equation 8.7). For both pH conditions, LSM fitting curves show a better description of the experimental data (Figure 8.9). This indicates that this transport mechanism cannot only be described by Brownian motion of PLL in the nanoparticles, i.e. it does not strictly follow Fick’s behavior, but is governed by both Fickian and Case II transport, with only one main relaxation of the nanocapsules. a) b)

Figure 8.9 Profile of PLL release from chitosan/fucoidan nanocapsules at 37 °C and pH 2 (a) and pH 7 (b): experimental data (x); description of Fick’s model (i=0) (--) and of Linear Superimposition Model (i=1) (__).

195

Pinheiro, A. C. (2013)

Table 8.2 presents the regression analysis results of the LSM fitting, showing that this model adequately describes the experimental data with relatively good regression quality (R2>0.90) and that almost all parameters were estimated with good precision.

Table 8.2 Results of fitting the Linear Superimposition Model (LSM) (i=1) to experimental data of the PLL release. Evaluation of the quality of the regression on the basis of RMSE and R2. Estimates precision is evaluated using the SHW% (in parenthesis).

2 -1 -1 pH RMSE R X kF (min ) KR (min )

0.299 0.008 0.059 2 0.160 0.903 (78.40%) (148.01%) (38.40%)

0.565 0.145 0.033 7 0.113 0.915 (21.42%) (106.97%) (46.30%)

The parameter X is defined as the Fick’s diffusion contribution to the total release from the nanocapsules ( ) and it can be observed from Table 8.2 that for pH 2 X < 0.5 and for pH 7 X

> 0.5, which indicates that at pH 2 relaxation is the governing phenomenon and at pH 7 Fick’s diffusion is the main release mechanism. These results may be explained based on the magnitude of the electrostatic interactions between the polyelectrolytes at different pH values. The pKa value of the anionic sulphate groups of fucoidan is around pH 2, therefore, at this pH these groups are half protonated and consequently partially uncharged, which means that the electrostatic interactions between fucoidan and PLL or chitosan are weaker. This leads to the loosening of nanocapsules’ structure and promotes the release due to polymer relaxation. Also the relaxation rate constant (kR) decreased with the increase of pH, supporting this hypothesis. As for the Fickian rate constant (kF), as expected this parameter increased with pH, reflecting the predominance of Fick’s behaviour at pH 7. A similar behaviour was observed when studying the release of methylene blue from a nanolayered coating composed of κ- carrageenan (also a sulphated anionic polysaccharide) and chitosan (Chapter 4).

8.4 CONCLUSION

This work demonstrated that the layer-by-layer deposition of chitosan/fucoidan layers on a sacrificial template allows the production of biodegradable hollow nanocapsules with great potential to act as a controlled delivery system for bioactive compounds. Chitosan and fucoidan were alternatively adsorbed on polystyrene templates until the deposition of 10 layers and the polystyrene core was subsequently removed. The obtained nanocapsules were characterized by means of ζ-potential, scanning and transmission electron microscopy in order

196

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

to establish their surface charge, size and morphology. Poly-L-lysine, a cationic bioactive compound, was encapsulated in the nanocapsules and its loading and release behaviour was evaluated. Chitosan/fucoidan nanocapsules showed a good capacity for the encapsulation and loading of PLL. The results of fitting the linear superimposition model to the experimental data of PLL release suggested an anomalous behaviour, with one main polymer relaxation and the PLL release was found to be pH-dependent. This work allows clarifying the release mechanisms involved at two different pH values found within a human GI tract, which is of the utmost importance to control the rate of release of bioactive compounds under those circumstances.

Due to the bioactive and non-toxic nature of chitosan/fucoidan nanocapsules and due to their responsiveness at different pHs, these capsules are envisaged as a nanocarrier system to control the release of bioactive compounds for food and pharmaceutical applications.

197

Pinheiro, A. C. (2013)

8.5 REFERENCES

Bauer, A., Kirby, W., Sherris, J., & Turck, M. (1966). Antibiotic susceptibility testing by a standardized single disk method, American Journal of Clinical Pathology 45, 493–496. Berens, A. R., & Hopfenberg, H. B. (1978). Diffusion and relaxation in glassy polymer powders. 2. Separation of diffusion and relaxation parameters, Polymer 19(5), 489-496. Black, W. A. P., Dewar, E.T., and Woodward, F.N. (1952). Manufacture of algal chemicals. IV. -Laboratory-scale isolation of fucoidan from brown marine algae, Journal of the Science of Food and Agriculture 3, 122-129. Blois, M. S. (1958). Antioxidant Determinations by the Use of a Stable Free Radical, Nature 181(4617), 1199-1200. Blumenkrantz, N., & Asboe-Hansen, G. (1973). New method for quantitative determination of uronic acids, Analytical Biochemistry 54(2), 484-489. Caruso, F., Caruso, R. A., & Möhwald, H. (1998). Nanoengineering of Inorganic and Hybrid Hollow Spheres by Colloidal Templating, Science 282(5391), 1111-1114. Coimbra, M. A., Delgadillo, I., Waldron, K. W., & Selvendran, R. R. (1996) Isolation and Analysis of Cell Wall Polymers from Olive Pulp, In Plant Cell Wall Analysis (Linskens, H., and Jackson, J., Eds.), pp 19-44, Springer Berlin Heidelberg. Coma, V., Deschamps, A., & Martial-Gros, A. (2003). Bioactive Packaging Materials from Edible Chitosan Polymer—Antimicrobial Activity Assessment on Dairy-Related Contaminants, Journal of Food Science 68(9), 2788-2792. Cuomo, F., Lopez, F., Miguel, M. G., & Lindman, B. r. (2010). Vesicle-Templated Layer- by-Layer Assembly for the Production of Nanocapsules, Langmuir 26(13), 10555-10560. Decher, G., & Schlenoff, J. B. (2003) Polyelectrolyte Multilayers, an Overview, Wiley- VCH Verlag GmbH & Co. KGaA, Weinheim Dutta, J., Tripathi, S., & Dutta, P. K. (2012). Progress in antimicrobial activities of chitin, chitosan and its oligosaccharides: a systematic study needs for food applications, Food Science and Technology International 18(1), 3-34. Hamano, Y., Nicchu, I., Shimizu, T., Onji, Y., Hiraki, J., & Takagi, H. (2007). ɛ-Poly-L- lysine producer, Streptomyces albulus, has feedback-inhibition resistant aspartokinase, Applied Microbiology and Biotechnology 76(4), 873-882. Helander, I. M., Nurmiaho-Lassila, E. L., Ahvenainen, R., Rhoades, J., & Roller, S. (2001). Chitosan disrupts the barrier properties of the outer membrane of Gram-negative bacteria, International Journal of Food Microbiology 71(2-3), 235-244.

198

Chapter 8. Chitosan/Fucoidan multilayer nanocapsules for controlled release of bioactive compounds

Indest, T., Laine, J., Ribitsch, V., Johansson, L.-S., Stana-Kleinschek, K., & Strnad, S. (2008). Adsorption of Chitosan on PET Films Monitored by Quartz Crystal Microbalance, Biomacromolecules 9(8), 2207-2214. Johnston, A. P. R., Cortez, C., Angelatos, A. S., & Caruso, F. (2006). Layer-by-layer engineered capsules and their applications, Current Opinion in Colloid & Interface Science 11(4), 203-209. Kusaykin, M., Bakunina, I., Sova, V., Ermakova, S., Kuznetsova, T., Besednova, N., Zaporozhets, T., & Zvyagintseva, T. (2008). Structure, biological activity, and enzymatic transformation of fucoidans from the brown seaweeds, Biotechnology Journal 3(7), 904-915. Liu, Y., Yang, J., Zhao, Z., Li, J., Zhang, R., & Yao, F. (2012). Formation and characterization of natural polysaccharide hollow nanocapsules via template layer-by-layer self-assembly, Journal of Colloid and Interface Science 379(1), 130-140. Mak, W., Hamid, N., Liu, T., Lu, J., & White, W. L. (2013). Fucoidan from New Zealand Undaria pinnatifida: Monthly variations and determination of antioxidant activities, Carbohydrate Polymers 95(1), 606-614. Martins, G. V., Mano, J. F., & Alves, N. M. (2010). Nanostructured self-assembled films containing chitosan fabricated at neutral pH, Carbohydrate Polymers 80(2), 570-573. Mora-Huertas, C. E., Fessi, H., & Elaissari, A. (2010). Polymer-based nanocapsules for drug delivery, International Journal of Pharmaceutics 385(1-2), 113-142. Najjar, M. B., Kashtanov, D., & Chikindas, M. L. (2007). ε-Poly-l-lysine and nisin A act synergistically against Gram-positive food-borne pathogens Bacillus cereus and Listeria monocytogenes, Letters in Applied Microbiology 45(1), 13-18. Preetz, C., Rübe, A., Reiche, I., Hause, G., & Mäder, K. (2008). Preparation and characterization of biocompatible oil-loaded polyelectrolyte nanocapsules, Nanomedicine : nanotechnology, biology, and medicine 4(2), 106-114. Selvendran, R. R., March, J. F., & Ring, S. G. (1979). Determination of aldoses and uronic acid content of vegetable fiber, Analytical Biochemistry 96(2), 282-292. Shahidi, F., Arachchi, J. K. V., & Jeon, Y.-J. (1999). Food applications of chitin and chitosans, Trends in Food Science & Technology 10(2), 37-51. Shu, S., Sun, C., Zhang, X., Wu, Z., Wang, Z., & Li, C. (2010). Hollow and degradable polyelectrolyte nanocapsules for protein drug delivery, Acta Biomaterialia 6(1), 210-217. Sun, B., Mutch, S. A., Lorenz, R. M., & Chiu, D. T. (2005). Layered Polyelectrolyte−Silica Coating for Nanocapsules, Langmuir 21(23), 10763-10769. Szabo, M., Idiţoiu, C., Chambre, D., & Lupea, A. (2007). Improved DPPH determination for antioxidant activity spectrophotometric assay, Chem Pap 61(3), 214-216.

199

Pinheiro, A. C. (2013)

Szczepanowicz, K., Hoel, H. J., Szyk-Warszynska, L., Biela ska, E., Bouzga, A. M., Gaudernack, G., Simon, C., & Warszynski, P. (2010). Formation of Biocompatible Nanocapsules with Emulsion Core and Pegylated Shell by Polyelectrolyte Multilayer Adsorption, Langmuir 26(15), 12592-12597. Yamanaka, K., & Hamano, Y. (2010) Biotechnological Production of Poly-Epsilon-l- Lysine for Food and Medical Applications, In Amino-Acid Homopolymers Occurring in Nature (Hamano, Y., Ed.), pp 61-75, Springer Berlin Heidelberg. Ye, S., Wang, C., Liu, X., & Tong, Z. (2005). Multilayer nanocapsules of polysaccharide chitosan and alginate through layer-by-layer assembly directly on PS nanoparticles for release, Journal of Biomaterials Science, Polymer Edition 16, 909-923. Yoshida, T., & Nagasawa, T. (2003). e-Poly-l-lysine: microbial production, biodegradation and application potential, Applied Microbiology and Biotechnology 62(1), 21- 26.

200

SECTION V CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK

9 CHAPTER 9. G E N E R A L

CONCLUSION S AND FUTURE PERSPECTIVES GENERAL CONCLUSIONS AND FUTURE PERSPECTIVES

9.1 General Conclusions ...... 205

9.2 Guidelines for future work ...... 207

Pinheiro, A. C. (2013)

204

Chapter 9. General conclusions and future perspective

9.1 GENERAL CONCLUSIONS

The recent advances in nanotechnology have brought to the attention of food technologists a huge potential for the development of novel solutions to be implemented by the food industry. However, application of nanotechnology by the food industry is hindered by very significant problems, which could be summarized in two issues: edibility and functionality/behaviour once inside the human body.

In this context, the work presented in this thesis is a result of a plan that aimed at developing stable nanostructures for food applications and evaluating their in vitro behaviour. The strategy adopted included the development and characterization of edible nanostructures, incorporation of bioactive compounds and evaluation of their behaviour when subjected to digestion in an artificial gastrointestinal system. In particular, the research undertaken was based on three different nanostructures: nanolayered coatings composed of κ-carrageenan and chitosan, curcumin nanoemulsions stabilized by different emulsifiers and chitosan/fucoidan hollow nanocapsules.

The main contributions of this thesis can be summarized as follows:

. A deeper understanding of the interactions between polyelectrolytes in a multilayer system was obtained (Chapter 3). The interaction between κ-carrageenan and chitosan is an exothermic process and the formation of a nanolayered coating occurs mainly due to electrostatic interactions existing between the two polyelectrolytes. Also, the alternating deposition of κ-carrageenan and chitosan results in the formation of a stable multilayer structure. The κ-carrageenan/chitosan nanolayers exhibit good gas barrier properties and therefore offer great potential to be used e.g. to coat food systems and can act as platforms for development of systems tailored for increasingly specific applications. . Useful information on the morphology and architecture of nanolayered coatings was obtained (Chapter 4). The amount of methylene blue (MB) loaded into the κ- carrageenan/chitosan nanolayered coating increased with the distance from the first layer, suggesting layer interpenetration of the inner layers and that MB was able to diffuse into the κ-carrageenan/chitosan nanolayered coating. The release of MB follows an anomalous behaviour, with one main polymer relaxation. However, the prevailing release mechanism depends on the position of MB incorporated on the nanolayered coating and on the values of pH and temperature of the medium. The release mechanisms involved at nano-scale were clarified, which is of the utmost

205

Pinheiro, A. C. (2013)

importance for the application of nanolayered systems in food products, as a strategy for shelf-life extension. . The stability of curcumin nanoemulsions stabilized by different emulsifiers (Tween 20, SDS, DTAB and lactoferrin) was evaluated (Chapter 5). The DTAB-stabilized emulsions revealed to be the least stable, showing phase separation, also displaying the highest colour change and the highest decrease in the magnitude of ζ-potential. The nanoemulsions stabilized by lactoferrin revealed to be at least as stable as those made with synthetic emulsifiers, showing the potential of lactoferrin to substitute synthetic emulsifiers. . The impact of surface charge on the digestion of curcumin nanoemulsions was evaluated using as digestion model the Human Gastric Simulator (Chapter 6). Although all the nanoemulsions tested were markedly affected by the digestion process, DTAB- stabilized emulsions showed the most drastic change in droplet size and electrical charge. The initial emulsifier used was also found to impact curcumin bioavailability: curcumin exhibited a very low bioavailability under intestinal conditions in the presence of DTAB as emulsifier. The higher curcumin bioavailability observed for Tween 20-stabilized nanoemulsions could be related with the greater free fatty acid production obtained for this nanoemulsion as a direct result of emulsion droplet size. Therefore, the behaviour of nanoemulsions during in vitro digestion was clarified, contributing to the development of optimized nanoemulsion-based delivery systems to protect and release bioactive lipophilic compounds within the human body. . A dynamic digestion model comprising stomach, duodenum, jejunum and ileum was developed and used to evaluate the impact of alginate coating on the digestion of curcumin nanoemulsions stabilized by lactoferrin (Chapter 7). The results suggested that alginate coating may be able to control the rate of lipid digestion and FFA adsorption within the GI tract, but encapsulated lipids are digested at the same extent, releasing the lipophilic bioactive compound, which is then taken up by mixed micelles. Therefore, lactoferrin/alginate multilayer coating can improve the physicochemical stability of emulsions in food products, while still releasing the encapsulated lipophilic components in the GI tract. . The layer-by-layer deposition of chitosan/fucoidan layers on a sacrificial template allowed producing biodegradable hollow nanocapsules with potential to act as a controlled delivery system for bioactive compounds (Chapter 8). Chitosan/fucoidan

nanocapsules showed good capacity for the encapsulation and loading of poly-L-lysine and their release behaviour depends on pH.

206

Chapter 9. General conclusions and future perspective

In conclusion, this work has provided more options of edible nanostructures for incorporation of both hydrophilic and lipophilic compounds. These nanostructures can be used as platforms for the production of virtually thousands of new products with improved characteristics targeted at the most recent consumer trends. Further, this work contributed to the understanding of the behaviour of those nanostructures inside the human body during digestion (e.g. release phenomena involved in structures at the nano-scale and bioavailability of lipophilic compounds during in vitro digestion), which is important to determine their functional performance, aiming at the optimization of nano-based delivery systems to protect and release bioactive compounds.

9.2 GUIDELINES FOR FUTURE WORK

Although this thesis has provided very important insights about the development of stable nanostructures for food applications and their in vitro behaviour, a work like this is never complete. Therefore some recommendations and guidelines for future work can be given:

. Further evaluation of behaviour of chitosan/fucoidan nanocapsules in vitro, using this time an artificial gastrointestinal system; . Application of the nanostructures produced in model food systems and evaluation of their effects on foods’ shelf-life; . Assessment of the cytotoxicity of the nanostructures and their cellular uptake, and quantification of the absorption of bioactive compounds incorporated in the nanostructures in human intestinal cells, using e.g. Caco-2 cells as model system; . Combining the dynamic gastrointestinal simulation system developed in this work with a Caco-2 cell culture model to assess bioactive compounds bioavailability from nanostructures.

207