Scale-up production of food grade nanoparticles aiming their application in food products

Universidade do Minho Escola de Engenharia Beatriz Sousa Afonso Scale-up production of food grade nanoparticles aiming their application in food products Mestrado Integrado em Engenharia Biológica Ramo de Tecnologia Química e Alimentar Trabalho efetuado sob orientação do: Professor Doutor Eugénio Manuel Faria Campos Ferreira e do supervisor na empresa: Doutor Miguel Ângelo Parente Ribeiro Cerqueira Janeiro de 2020 II DIREITOS DE AUTOR E CONDIÇÕES DE UTILIZAÇÃO DO TRABALHO POR TERCEIROS Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condiçõ§es não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição-NãoComercial-SemDerivações CC BY-NC-ND https://creativecommons.org/licenses/by-nc-nd/4.0/ III AGRADECIMENTOS Gostaria de expressar o meu sincero obrigada a todos que contribuíram, direta ou indiretamente, para a realização deste trabalho. Posso dizer com muito orgulho que cresci imenso com esta experiência, tanto a nível intelectual como pessoal. Primeiramente, quero agradecer ao meu supervisor do INL, Doutor Miguel Cerqueira, pela mentoria, pela disponibilidade e pela oportunidade de fazer parte de uma grande equipa de investigadores e me ter permitido uma experiência extremamente enriquecedora. Também, quero agradecer a uma contribuição essencial neste trabalho, Gabriela Azevedo, Catarina Gonçalves e Isabel Rodriguez pela ajuda incansável, pela paciência, prontividade em ajudar e disponibilidade. Ao meu orientador, Professor Eugénio Ferreira, por me clarificar dúvidas, apoiar neste trabalho e dedicação do seu tempo. Quero agradecer aos meu colegas do INL por me terem acolhido de forma tão especial e por tornarem esta experiência uma memória que guardo com muito carinho. Por fim, e com uma consideração especial, à minha família e aos meus amigos por terem contribuído para todo o meu percurso académico com muito apoio e incentivo. O desenvolvimento deste trabalho foi ainda realizado ao abrigo do Projeto “MobFood – Mobilização de conhecimento científico e tecnológico em resposta aos desafios do mercado Agroalimentar” (POCI-01-0247-FEDER-024524), pelo Consórcio “MobFood”, financiado pelo Fundo Europeu de Desenvolvimento Regional (FEDER), através do Sistema de Incentivos à Investigação e Desenvolvimento Tecnológico, no âmbito do Programa Operacional para a Competitividade e Internacionalização do Portugal2020. IV STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. V RESUMO Nanopartículas poliméricas são estruturas promissoras para o encapsulamento de vários materiais, com o objetivo de proteger o composto bioativo e libertação no local desejado. Este trabalho tem como objetivo a avaliação de mecanismos de aumento de escala para a produção de nanopartículas poliméricas e testar a sua exequibilidade a nível laboratorial para uma aplicação futura a nível industrial. Assim sendo, diferentes mecanismos foram investigados: misturador em forma de T, nanoprecipitação flash e através da utilização de uma membrana. Deste modo, o equipamento mais promissor a escalamento será o misturador em forma de T, capaz de um escalamento até 20 vezes a escala laboratorial e mais fácil de contruir. À escala laboratorial, foram produzidas nanopartículas, através do método de nanoprecipitação, capazes de encapsular b-caroteno, um composto hidrofóbico lipofílico e sensível à luz testando dois polímeros distintos como material de parede: zeína e etilcelulose. Em seguimento, foram caracterizadas por vários métodos e foi realizada uma simulação gastrointestinal in vitro . Os resultados da produção a nível laboratorial mostram que as condições ótimas para uma produção de nanopartículas com distribuição de tamanho reduzida e controlada são concentrações baixas de polímero e concentrações altas de anti-solvente. Para o nanossistema etilcelulose-b-caroteno, foi usada a concentração de 0,0004% de b-caroteno (BC) para produzir nanopartículas com tamanho (por intensidade) de 59,9 nm, PDI de 0,274, eficiência de encapsulação de 66,56% e potencial zeta de -96 mV exibindo comportamento estável. No nanossistema zeínab-caroteno, a concentração de 0,001% de b-caroteno foi usada para produzir as nanopartículas, resultando em tamanhos de 82,7 nm, PDI de 0,286, eficiência de encapsulamento de 92,69% e potencial zeta de +70,5 mV. Por fim, os resultados obtidos da digestão in vitro demonstraram que ambos os polímeros melhoraram a libertação de BC no local desejado quando comparado com o controlo (BC livre em óleo). No entanto, nenhum dos polímeros apresentou um valor de bioacessibilidade elevado, com 11,78% e 14,89% para as nanopartículas de etilcelulose e de zeína, respetivamente na fase intestinal. Ambos os polímeros melhoraram a estabilidade química do b-caroteno, retardando a sua libertação na fase gástrica. Contudo, a zeína foi o polímero que apresentou melhores valores de bioacessibilidade. Palavras chave: Misturador em forma de T, b-caroteno, Etilcelulose, Zeína, nanoprecipitação VI ABSTRACT Polymeric nanosystems are promising structures for the encapsulation of various materials with the goal of protecting the bioactive compound and delivery at the target site. This work has as final goal the assessment of scale-up techniques for the production of polymeric nanoparticles, testing its feasibility at a lab-scale for the future application at an industrial scale. Therefore, several mechanisms were investigated: T-mixer, flash nanoprecipitation and membrane contactor. The nanoparticles were produced at a laboratory scale to evaluate their feasibility using the solvent-displacement method. Thus, being the most promising technique using a T mixer device, able to a 20-fold scale-up and easier to fabricate. At laboratory scale, nanoparticles were produced, through the solvent-displacement method, with the ability to encapsulate b-carotene, a hydrophobic lipophilic light sensitive compound. Two different polymers were used as encapsulating materials: zein and ethylcellulose. Then, the nanoparticles were characterized by several methods and an in vitro gastrointestinal simulation was performed. Results show that the best conditions for a size-controlled production of nanoparticles with a narrow size distribution are lower polymer concentrations and higher antisolvent concentrations. For the ethylcellulose-b-carotene system, a concentration of 0.0004% of b-carotene was used to produce nanoparticles with size by intensity of 59.9 nm, PDI of 0.274, encapsulation efficiency of 66.56% and high stability behavior with a zeta potential of – 93.26 mV. In the zein-b-carotene system, a concentration of 0.001% of b-carotene was used to produce the nanoparticles, resulting in sizes of 82.7 nm, PDI of 0.286, encapsulation efficiency reaching 92.69% and zeta potential of + 70.5 mV. Lastly, results obtained from the in vitro digestion demonstrated that both polymers have improved release of BC at the target site comparing to the control sample of free BC in oil. However, none of the polymers showed high bioaccessibility, having 11.78 % and 14.89 % for the ethylcellulose nanoparticles and zein nanoparticles, respectively in the intestinal phase. Both polymers improved chemical stability of entrapped beta-carotene by retarding its release in the gastric phase. Nonetheless, zein was the polymer with better results for the bioaccessibility. Key words: T mixer, b-carotene, Ethylcellulose, Zein, Solvent-displacement Scale-up production of Food grade nanoparticles VII TABLE OF CONTENTS AGRADECIMENTOS ................................................................................................................. III RESUMO .................................................................................................................................. V ABSTRACT ............................................................................................................................... VI TABLE OF CONTENTS ............................................................................................................ VII LIST OF FIGURES ..................................................................................................................... X LIST OF TABLES ..................................................................................................................... XII LIST OF NOMENCLATURE ..................................................................................................... XIV CHAPTER 1 INTRODUCTION ................................................................................................. 1 1.1 Motivation .................................................................................................................. 1 1.2 Approach ................................................................................................................... 1 1.3 Nanoencapsulation in food products ........................................................................... 2 1.4 b-carotene ................................................................................................................. 3 1.5 Polymers .................................................................................................................... 6 1.5.1 Ethylcellulose ...................................................................................................... 6 1.5.2 Zein .................................................................................................................... 8 1.6 Solvent Displacement Method ................................................................................... 10 CHAPTER 2 OVERVIEW OF SCALE-UP PRODUCTION OF FOOD GRADE NANOPARTICLES ..... 13 2.1 Introduction .............................................................................................................. 13 2.2 T mixer .................................................................................................................... 14 2.3 Flash nanoprecipitation ............................................................................................ 15 2.4 Membrane contactor ................................................................................................ 17 2.5 Comparison of scale-up devices ................................................................................ 17 CHAPTER 3 PRODUCTION AND CHARACTERIZATION OF FOOD GRADE NANOPARTICLES AT LABORATORY-SCALE .......................................................................................................... 19 3.1 MATERIALS AND METHODS ..................................................................................... 19 3.1.1 Nanoparticle production .................................................................................... 19 3.1.1.1. Materials ................................................................................................... 19 3.1.1.2. Production of ethylcellulose nanoparticles .................................................. 19 VIII 3.1.1.3. Production of zein nanoparticle ................................................................. 20 3.1.2 Nanoparticle characterization ............................................................................ 21 3.1.2.1. Dynamic Light Scattering analysis ............................................................. 21 3.1.2.2. Encapsulation Efficiency ............................................................................ 21 3.1.2.3. Transmission electron microscopy ............................................................. 22 3.1.2.4. Fourier transform infrared (FTIR) ............................................................... 23 3.1.2.5. X-Ray Diffraction ........................................................................................ 23 3.1.3 In vitro gastrointestinal digestion ........................................................................ 23 3.1.3.1. PHASE 1 - Enzyme activities, bile concentration and stock solutions ........... 24 Enzyme activity assays ....................................................................................... 24 Bile concentration assay ..................................................................................... 26 Simulated digestion fluids preparation ................................................................ 27 3.1.3.2. PHASE 2 – Digestion procedure ................................................................ 28 Oral phase ......................................................................................................... 29 Gastric phase ..................................................................................................... 29 Intestinal phase .................................................................................................. 29 3.1.3.3. PHASE 3 – Sample Analysis ...................................................................... 29 Extraction of BC ................................................................................................. 30 b-carotene quantification through HPLC ............................................................. 30 3.1.4 Statistical analysis ............................................................................................. 31 3.2 RESULTS AND DISCUSSION ..................................................................................... 32 3.2.1 Ethylcellulose nanoparticles ............................................................................... 32 3.2.2 Zein nanoparticles ............................................................................................. 34 3.2.3 Encapsulation Efficiency .................................................................................... 35 3.2.4 Size, PDI and Zeta Potential ............................................................................... 37 3.2.5 Transmission electron microscopy ..................................................................... 38 3.2.6 Fourier Transform Infrared spectroscopy ............................................................ 40 3.2.7 X-ray diffraction ................................................................................................. 42 3.2.8 In vitro digestion ................................................................................................ 43 CHAPTER 4 CONCLUSIONS AND FUTURE WORK ................................................................ 49 REFERENCES ......................................................................................................................... 51 ANNEXES ............................................................................................................................... 58 Scale-up production of Food grade nanoparticles IX ANNEX A OPTIMIZATION OF NANOPARTICLE’S PARAMETERS .............................................. 58 ANNEX B CALIBRATION CURVE OF BC ................................................................................. 60 ANNEX C ENZYME ACTIVITIES AND BILE CONCENTRATION ASSAYS ..................................... 61 ANNEX D HPLC CALIBRATION CURVE .................................................................................. 63 ANNEX E EXTRACTION EFFICIENCY ASSAYS ......................................................................... 64 2 For this purpose, both ethylcellulose and zein nanoparticles were prepared following the solventdisplacement method (nanoprecipitation). Then, the final particles were submitted to several tests to assess their functionality and morphology. An in vitro simulation of the digestion process was carried out to evaluate the bioaccessibility of b-carotene after digestion representing the amount of b-carotene that would be available for intestinal absorption. This work also presents a review of scale-up strategies for the production of the produced foodgrade polymeric nanoparticles based on the solvent-displacement method. 1.3 Nanoencapsulation in food products Nanotechnology has a great potential to improve the quality and safety of foods. When applied to the food industry can be used to modify some of the food characteristics, such as stability during shelf life, coloring strength, texture and taste. (Silva et al., 2012) Nano can play a major role in incorporating bioactive ingredients such as carotenoids, polyphenols, vitamins and minerals into food products (Wan et al., 2015). For that purpose, food-grade polymer nanoparticles as carriers of bioactive ingredients have been used as delivery systems for controlled release and the protection of these ingredients against adverse conditions present in the human body (Fernandez et al., 2009) The encapsulation of bioactive ingredients may be achieved in nanoparticles with different structural features, namely i) nanospheres, that are homogeneous matrix particles having their entire arrangement solid, constituted only by the polymer where active compounds can be trapped inside, dissolved in it or adsorbed at the sphere surface, ii) nanocapsules, that are nonhomogeneous vesicles having a polymeric shell surrounding a liquid core in which the active compounds are dissolved but may also be absorbed in the surface of the capsule (Pinto Reis et al., 2006; Vauthier & Bouchemal, 2009) The encapsulation of bioactive compounds offers a wide range of solutions in the food industry in terms of health, safety and high-quality products (Barreras-Urbina et al., 2016). Improved solubility and bioavailability are achieved due to the controlled release of functional compounds targeting specific sites of action, in particular compounds with poor solubility in certain matrices. The protective capsule masks astringency tastes, having less impact on the sensory attributes of foods and the bioactive compound is sheltered from degradation from the conditions it can be exposed (Lepeltier et al., 2015). Scale-up production of food grade nanoparticles 3 1.4 b -carotene In recent years, there has been a growing interest in foods that provide health benefits. These socalled functional foods provide added benefits beyond the basic nutrition of the initial food product. Therefore, secondary plant metabolites namely, carotenoids, are promising ingredients to be incorporated in foods due to their proposed health benefits. There are more than 60 different carotenoids present in vegetable products that are consumed by humans. Among them, the most common are b-carotene, lutein, zeaxanthin and lycopene (Dasgupta & Klein, 2014; Gunstone, 2003). Carotenoids are naturally occurring pigments that absorb blue and purple light with maxima at 450 nm (Pénicaud et al., 2011) thus displaying the color red, orange and yellow, responsible for the bright colors of many vegetables (Gunstone, 2003). They are held in the chloroplasts as oil droplets, for instance in fruits (Alminger et al., 2014). b-carotene can be found predominantly in vegetables and fruits such as cantaloupe, carrots, mangos, sweet potatoes and several greens (spinach, lettuce) (Boon et al., 2010; Gunstone, 2003). b-carotene is fat-soluble and best-known for its function as a natural antioxidant and high provitamin A activity (Packer et al., 2005). It comprises properties such as the ability to scavenge active free radicals, neutralizing peroxy radicals as well as singlet oxygen, thus protecting cellular tissues from lipid peroxidation (Dasgupta & Klein, 2014; Paiva & Russell, 1999). It is oxidized by light due to the presence of conjugated double bounds in their molecules functioning as lightharvesting pigments and photoprotective agents in vegetables (Gunstone, 2003). b-carotene is currently being used as a food additive with the European denomination E160 (European Comission, 2008) acting as a coloring agent, antioxidant and vitamin A capacities (de Freitas Zômpero et al., 2015). Structurally, b-carotene is constituted by a polyene chain with 11 conjugated double bounds and two rings at each end of the chain (Wang et al., 2018). As a provitamin A carotenoid, it can be cleaved by the enzyme b-carotene 15,15’ monooxygenase present in the human intestine and liver to form two equivalent retinol molecules (Pénicaud et al., 2011). Vitamin A designate the group of compounds such as retinol, retinal, retinoic acid and retinyl esters. The enzyme acts when the body is in need for vitamin A, using b-carotene as a form of vitamin A. If the body has sufficient vitamin then b-carotene will be stored for futures needs or 4 absorbed intact. For vitamin A, 1 IU corresponds to 0.0006 mg of b-carotene (Dasgupta & Klein, 2014). The properties that gives carotenoids beneficial properties in healthy body function creates challenges when carotenoids are added to food products. Because of b-carotene hydrophobic character, it is insoluble in water and displays poor bioavailability in crystalline form thus making it difficult the absorption through diet (Yusuf et al., 2012). b-carotene is slightly soluble in ethanol. Três et al. (2007) predicted its solubility at 0.360 mg/mL at 20 ºC. Bioaccessibility of a certain compound is intended to be the portion of the nutrient from the food that is available for absorption, the amount that is released from the food matrix during digestion and accessible for absorption into the mucosa (Hedrén, Diaz, & Svanberg, 2002). Much attention has focused on b-carotene bioaccessibility due to several factors that might influence it, such as solubility, the viscosity of the medium, interactions with other compounds present in the food matrix (e.g. competition with other fat-soluble nutrients such as vitamin E), the food processing method applied and the presence of factors that interfere with micelle formation (Hedrén et al., 2002; Paiva & Russell, 1999). Encapsulation Encapsulation of labile compounds has been a solution for many challenges in the food industry. b-carotene displays several limiting factors that hinder its use in food products in its free form. Those limitations consist of water insolubility, light, oxygen and heat sensitivity, chemical instability and low bioavailability (Deng et al., 2014; Wang et al., 2018). Also, they limit the amount of bcarotene to reach the specific sites where its action is needed (Chuacharoen & Sabliov, 2016). The bioaccessibility of carotenoids from raw vegetables is approximately 10% (Ribeiro et al., 2008). A compelling approach to protect and deliver labile ingredients is to incorporate them within a coating or wall material (Deng et al., 2014). Said nanoparticles can be developed into polymerbased systems that allow a controlled release of b-carotene and help decrease their degradation (Boon et al., 2010). In vitro , ex vivo and in vivo experiments proved that b-carotene nanostructures are superior to pure b-carotene (Jain et al., 2018). There are diversified methods to design b-carotene nanoparticles. Those include the microencapsulation of b-carotene with soy protein isolate and OSA-modified starch by one of the most common techniques which is spray-drying (Deng et al., 2014). Through emulsion- Scale-up production of food grade nanoparticles 5 electrospray, b-carotene particles were made within zein and whey protein concentrate to improve bioaccessibility (Gómez-Mascaraque et al., 2017). Furthermore, homogenization-evaporation method was employed to produce b-carotene nanoparticles with different protein materials (Yi, Lam, et al., 2015). Solvent-displacement (or nanoprecipitation) was also performed with different wall material such as PLA, PLGA (Ribeiro et al., 2008), sodium caseinate (Yin, et al., 2009) and zein (Chuacharoen & Sabliov, 2016). Health related properties A diet with a large consumption of fruits and vegetables rich in b-carotene is less likely to develop lung cancer, cardiovascular diseases or age-related macular degeneration (Albanes, 1999; Gunstone, 2003). Once b-carotene is converted to vitamin A, the health benefits derived from the vitamin comprise the maintenance of normal eye health and immune system function, an important role in cell differentiation and organ development (Boon et al., 2010). b-carotene supplementation can be also be used to treat patients with a photosensitive disorder (Packer et al., 2005). Despite all the benefits stated above, b-carotene supplementation in a high-risk population, such as smokers and asbestos-exposed workers did not show protective effects against cancer or cardiovascular diseases (Boon et al., 2010; Paiva & Russell, 1999). The environment of the lungs of smokers have lower circulating concentrations of antioxidants, among them is b-carotene (Dasgupta & Klein, 2014). This environment enhances b-carotene breakdown to produce oxidative by-products that may induce lung carcinogenesis (Packer et al., 2005). b-carotene does not have recommended dietary allowance, however, the recommendation for vitamin A for women and men is 800 and 1000 µg of retinol (or retinol equivalents), respectively. Assuming that all b-carotene consumed is converted to retinol the dietary allowance would be approximately 4.8 and 6.0 mg for women and men, respectively (Albanes, 1999). The dietary allowance of vitamin A is expressed in retinol equivalents due to the existence of two sources: preformed vitamin from animal products such as animal tissues, milk and eggs and provitamin A from carotenoids in vegetables and fruits. Most of the humans are omnivores and consume a mixture of the two sources of the vitamin. However, consuming high doses of vitamin A from animal products can cause hypervitaminosis but the overconsumption of b-carotene foods is tightly regulated in which the body only converts the vitamin needed, storing or absorbing the remaining b-carotene (Packer et al., 2005). As previously stated, risk groups can be harmed by ingesting b- 6 carotene, which may result in increased lung cancer. Nonetheless, it does not exhibit toxicity or a health hazard in non-at-risk groups. When ingested in large amounts, can cause the appearance of carotenoderma, i.e., the accumulation of b-carotene in the skin giving it a yellow or orange tone (Packer et al., 2005). 1.5 Polymers The polymers used to produce the food-grade nanoparticles have to fulfill certain requirements regarding food safety. they must be non-toxic, non-immunogenic and biodegradable or at least totally eliminated from the body without any risk of accumulation, and its degradation products must follow the same requirements (Vauthier & Bouchemal, 2009). In the US they have to be selected from the materials granted GRAS (generally recognized as safe) status (Wan et al., 2015), and in the EU they must be approved for food use as ingredient, additive of novel food. 1.5.1 Ethylcellulose Ethylcellulose is a non-toxic, water-insoluble, biodegradable polymer used in the development of a variety of polymer nanoparticles (Urbán-Morlán et al., 2015). It is a semisynthetic material having properties such as wall-forming and biocompatibility (Avanço & Braschi, 2008). Moreover, it is a tasteless, white to light tan-colored powder which has a wide range of applications in the food, cosmetic and pharmaceutical fields (Lai, Pitt, & Craig, 2010; Rowe et al., 2009). Cellulose derivatives possess the polymeric backbone of cellulose, which contains a basic repeating structure of anhydroglucose units, each unit comprises three replaceable hydroxyl groups (Rekhi & Jambhekar, 1995). Ethylcellulose (EC) is produced via etherification of alkali cellulose using ethyl chloride. Figure 1.2 shows the structure of ethylcellulose, where it can be seen the anhydroglucose unit which has the three reactive hydroxyl groups ethoxylated (Porter, 1989). The repeating anhydroglucose units (n) are modified into ethyl ether groups to form the so-called ethyl ether of cellulose or ethylcellulose. The number of units joined by acetal linkages can vary to provide a wide variety of molecular weights (Rowe et al., 2009). The physical properties of EC are determined by the main chain properties such as the degree of polymerization and by the side-chain properties such as degree of substitution and the distribution of these substituents (Lai et al., 2010). Scale-up production of food grade nanoparticles 7 Figure 1.1. Molecular structure of ethylcellulose. Image adapted from (Rekhi & Jambhekar, 1995). Ethylcellulose is soluble in a wide variety of solvents such as esters, aromatic hydrocarbons, alcohols, ketones and chlorinated solvents. It is also soluble in ethanol and methanol yielding solutions with lower viscosity (Rekhi & Jambhekar, 1995). Moreover, EC is insoluble in water, therefore it is widely used for the controlled release of hydrophobic bioactive compounds (Murtaza, 2012). Ethylcellulose materials vary according to the degree of etherification or ethoxyl substitution, which is measured by the ethoxyl content. These materials are also produced in different viscosity grades of differing molecular weight: the higher the polymer molecule chain the higher is the viscosity (Rekhi & Jambhekar, 1995; Rowe et al., 2009). It can range from 7 to 100 mPas (7-100 cP), however for controlled release purposes, higher EC viscosities formulations resulted in slower drug release rates, while lower viscosities formulation resulted in the fastest drug release rate (Murtaza, 2012; Rowe et al., 2009). Ultimately, for the assemble of nanoparticles with the goal of controlled release of ingredients the optimal conditions combine a lower concentration of ethylcellulose with a lower viscosity ethylcellulose formulation. Applications The growing interest in renewable polymer resources, as opposed to petroleum-based polymers and chemicals, leads to an increasing importance of cellulose ethers handling. Among cellulose derivatives, ethylcellulose applications go from a coating material to a thickening agent (Rekhi & Jambhekar, 1995). Several uses have been reported for this polymer, as well as a wide range of fields of application. It is most widely used in the pharmaceutical field for a number of functions such as a viscosity 8 increasing agent and film former (Rowe et al., 2009); a hydrophobic coating agent in the release of a drug to mask unpleasant taste, improve the stability, control odor or taste, protect drugs from moisture or oxidation and to alter solubility or prevent incompatibilities (Avanço & Braschi, 2008; Murtaza, 2012); in topical formulation EC is used as a thickening agent for creams, lotions and gels (Rowe et al., 2009); and in the food industry as a stabilizer for emulsions and as an emulsifier (Rowe et al., 2009) Safety Several synthetic polymers have been used to formulate delivery systems, and for its application in foods they need to be approved. EC is approved as a food additive in Europe (E462) (European Comission, 2008), considered GRAS and approved by the FDA as safe for human consumption and it is included in the Canadian List of Acceptable Non-medicinal Ingredients (Rowe et al., 2009) Nanoparticles Literature shows that ethylcellulose can be used to encapsulate compounds with different properties and that a great variety of methods can be used to produce those nanoparticles. Among those methods, the most used in the literature, are the production of nanoparticles by solventdisplacement (or nanoprecipitation) (Fessi, et al., 1989), by solvent evaporation (Avanço & Braschi, 2008; Leitner et al., 2019), and by water-in-oil-in-oil double emulsion solvent diffusion method (Rama, Senapati, & Das, 2005). Some of the works with ethylcellulose nanoparticles comprise encapsulation of drugs such as ciclosporin, taxol, betaxolol, essential oils, lipiodol (contrast agent) (Fessi et al., 1989), zidovudine (Rama et al., 2005), ibuprofen (Swathi & Krishna Sailaja, 2014) and biobased photoprotectants (Hayden et al., 2018). 1.5.2 Zein Corn is mainly composed by the endosperm and germ, which are where the proteins are mainly located (Anderson & Lamsal, 2011). There are several types of proteins on corn, which represents approximately 10% of the dry mass. However, zein proteins comprise about 80% of the whole protein content (Papalia & Londero, 2015). Zein belongs to the group of corn proteins named prolamins (Patent No. US 9,381,252 B2, 2016; Zhang, Cui, Che, et al., 2015) and it is not constituted by a single protein but a mixture of four different classes grouped by solubility and sequence similarity, it can be divided into a-, b-, gand d-zeins. a-zein is the most abundant fractions, approximately 80% of the total zein mass (Li et al., 2017). Scale-up production of food grade nanoparticles 9 Zein is an amphiphilic biopolymer since it has both hydrophobic and hydrophilic attributes. The hydrophobic aminoacid content is represented by leucine, proline and alanine and the hydrophilic counterpart is represented by glutamine (Corradini et al., 2014). However, zein almost completely lacks lysine and tryptophan amino acids which is why it has poor nutritional quality (Patent No. US 9,381,252 B2, 2016; Holding, 2014). The combination of zein’s different amino acids is responsible for its particular solubility: zein is insoluble in pure water and in pure ethanol, making it a material of interest for the controlled release of a loaded compound, but is soluble in aqueousethanol solutions at certain concentrations (Hurtado-López & Murdan, 2005; Jain et al., 2018). It is also soluble in mixtures of water with aliphatic alcohols and other organic solvents containing hydroxyls, carbonyls, amines and other polar groups (Corradini et al., 2014). Commercial zein is usually extracted from corn gluten, which is a renewable source, through the conventional process of aqueous-alcohol extraction (Anderson & Lamsal, 2011). The extract is then centrifuged and chilled to precipitate the zein. At last, it is dried and form a yellowish powder (Corradini et al., 2014). Zein is a renewable, biodegradable and biocompatible (Corradini et al., 2014) powder with a yellow color due to the carotenoids (b-carotene, zeaxanthin and lutein) present in the corn (Sessa et al., 2003). In addition, is thermally stable up to 280 ºC, shows great oxygen barrier properties (Corradini et al., 2014; Penalva et al., 2017) and can withstand gastric pH (Zhang et al., 2016). Applications Zein has been investigated as carriers in various forms such as nanoparticles, films, hydrogels, micelles and nanofibers (Zhang, Cui, Chen, et al., 2015; Zhang et al., 2016) with several purposes being used for controlled release, improved stability of bioactive agents, controlled delivery and enhanced oral bioavailability of bioactive agents (Zhang, Cui, Che, et al., 2015; Zhang et al., 2016). As a result of these properties, zein can be used for the production of adhesives, chewing gum, coating for food products, textiles, ceramic, cosmetic powders, as an encapsulator for pesticides and inks (Patent No. US 9,381,252 B2, 2016; Anderson & Lamsal, 2011; Holding, 2014). It can serve as a biodegradable and sustainable material for the replacement of oil-based plastic (Fernandez et al., 2009) and, in the food industry, zein-based nanoparticles are used for encapsulation, stabilization and controlled release of bioactive compounds such as essential oils, polyphenols, food coloring agents (Wan et al., 2015), carotenoids (Chuacharoen & Sabliov, 2016; Wang et al., 2018) and vitamins (Kasaai, 2018). 10 Safety Plant-based proteins have become a material of increasing interest due to their abundance, relatively low cost in comparison with animal proteins and due to their safety (Zhang et al., 2015). Zein has been regarded with “Generally Recognized as Safe” (GRAS) status by the Food & Drug Administration (U.S. Food & Drug Administration, 2019). and it is also included in non-parenteral medicines licensed in the UK and in the Canadian List of Acceptable Non-medicinal Ingredients (Rowe et al., 2009). Therefore, it is a product considered food-grade and is safe for use in human or animal food according to the Codex Alimentarius of a country (Patent No. US 9,381,252 B2, 2016). Nanoparticles Zein with spherical morphology has the optimal shape for controlled release and protection of nutrients: provides a minimum contact with the environment and longest diffusion pathways (Kasaai, 2018). Zein nanoparticles has been used in the pharmaceutical and food fields, and different methods have been employed for the development of those nanoparticles. Among them are the anti-solvent precipitation (Li et al., 2017), phase separation and electrospraying (Wan et al., 2015). The distinct methods produce particles with the ability to encapsulate several bioactive compounds such as insulin (Patent No. US 9,381,252 B2, 2016), b-carotene (Chuacharoen & Sabliov, 2016; Jain et al., 2018), folic acid, resveratrol (Peñalva et al., 2015), quercetin (Penalva et al., 2017) and even essential oils (Wan et al., 2015). 1.6 Solvent Displacement Method The solvent displacement method also known as nanoprecipitation was described by Fessi et al. (1989) and consists of the development of nanoparticles in a reproducible, easy and scalable way. Because nowadays there is an increasing interest to employ processes that are environmentally sustainable, safe and energy-saving, the solvent displacement method has been widely used. This technique enables the production of nanospheres as well as nanocapsules in one step, costefficient and high yield encapsulation of hydrophobic compounds (Calderó et al., 2019). Nanoprecipitation requires three basic ingredients to be performed: the polymer, the polymer solvent and the non-solvent of the polymer (Vauthier & Bouchemal, 2009). In the selection of a suitable solvent, it has to comprise the ability of complete miscibility with the non-solvent, in addition to good solubilization of the bioactive compound and lower boiling point from the solvent, Scale-up production of food grade nanoparticles 11 so that removal by evaporation can occur (Horn & Rieger, 2001). Acetone is the most used solvent for this method. Other solvents such as ethanol, hexane and methylene chloride or a binary solvent blend can also be used. The polymers commonly used in this technique are PLA and PLGA (Rao & Geckeler, 2011). The polymer, the bioactive compound to be encapsulated and the solvent constitute the organic phase. The non-solvent of the polymer constitutes the aqueous phase which is usually water. Other materials could be added to this phase such as hydrophilic surfactants to avoid particles’ aggregation and coating materials (Miladi, Sfar, Fessi, & Elaissari, 2016). To produce the nanoparticles, the organic phase is poured into large amounts of aqueous phase under moderate stirring, usually with a magnetic stirrer. Reversing the order by adding large amounts of aqueous phase to the organic phase can also produce nanoparticles. When the phases come in contact, the solvent rapidly diffuses into the water, the polymer and the bioactive compound are insoluble in the mixed binary solution formed and precipitate producing nanoparticles (Rao & Geckeler, 2011). Then, the suspension of nanoparticles is submitted to evaporation to remove the organic solvent. To remove the aqueous phase so that a powder can be obtained, freeze drying can be used (Miladi et al., 2016). Several operating conditions influence the characteristics of the nanoparticles such as, the ratio of organic phase to aqueous phase, stirring rate, injection rate of the organic phase, which phase is poured and polymer concentration (Rao & Geckeler, 2011). The particle formation in the nanoprecipitation technique can be explained by nucleation theory which involves three phases: nucleation, growth and aggregation showed by Lince et al. (2008). When a solution of a polar water-miscible solvent containing a hydrophobic solute is mixed with large amounts of water, the concentration of the solute reaches its solubility limit causing nucleation of small solute molecules. Therefore, nucleation is driven by the saturation limit, i.e., when the aqueous phase breaks the interface between polymer and solvent. Supersaturation determines nucleation rate, thus mixing has an important effect: poor mixing of phases leads to low nucleation rate and produces big nanoparticles, high nucleation rates produces large amounts of smaller particles. Growth occurs through the capture of non-aggregated solute molecules added to the core from the surrounding solution. The separation of nucleation and growth allows obtaining a uniform particle suspension. Operating conditions well controlled aim for high nucleation rates and consequently low growth rates. Aggregation occurs when the particles formed previously collide with already existing particles, forming larger aggregates. In this phase maintaining control of the 18 Table 2.1. Comparison of the three scale-up methods T-mixer Flash nanoprecipitation devices Membrane contactor Provides appropriate mixing of the phases (Briançon et al., 1999). Impinging jets creates supersaturation producing a homogeneous solution (Chow et al., 2015). Uses a filtration device to produce nanoparticles. Complex set-up. Device with a simple design. Zein NPs were successfully produced in a CIJM by Wan et al. (2015). Scaled up to 8 -fold the lab-scale (Khayata et al., 2012). Production in a facile and reproducible way (GalindoRodriguez et al., 2005). Limitation: In CIJM the NPs solution showed low stability and aggregation (Chow et al., 2015). Particles’ size decreased from lab to pilot-scale (Khayata et al., 2012). Continuous feeding process, thus continuous production of NPs (Vauthier & Bouchemal, 2009). MIVM is composed by 4 inlets: each can be loaded with solvents at different ratios, materials and flow rates (Tao et al., 2019). Encapsulation efficiency remained high (Khayata et al., 2012). NPs batches up to 20-fold larger than lab-scale (Vauthier & Bouchemal, 2009). MIVM allows better control of particle size than CIJM (D’Addio & Prud’homme, 2011). Size of NPs is controlled by modifying the membrane (Charcosset & Fessi, 2005). Adjusting flow rate of the phases allows production of batches with different volumes (Galindo-Rodriguez et al., 2005). Both devices can be operated in parallel with several units scaled-up (Tao et al., 2019). Vitamin E loaded NPs were successfully produced in a pilot-scale set-up (Khayata et al., 2012). Limitation: low polymer concentration, otherwise polymer precipitates (Briançon et al., 1999). Good reproducibility (Tao et al., 2019). Limitation: Of the three methods, membrane contactor showed lower batch volume production. Scale-up production of food grade nanoparticles 19 CHAPTER 3 PRODUCTION AND CHARACTERIZATION OF FOOD GRADE NANOPARTICLES AT LABORATORY-SCALE 3.1 MATERIALS AND METHODS 3.1.1 Nanoparticle production 3.1.1.1. Materials Ethylcellulose polymer with a viscosity ranged between 40 to 52 cP, molecular weight 160.0 g/mol and ethoxyl content of 48 to 49.5% was obtained from Ashland (Wilmington, DE, USA) in the powder form with product number 44331. Zein (product number Z3625) and b-carotene (>97% UV, lot number BCBR8106V) with molecular weight 536.87 g/mol were purchased from Sigma Aldrich (Saint Louis, MO, USA). b-carotene was kept at -22 ºC and in darkness until use. All samples containing b-carotene during the experiments were protected from light. Ethanol absolute (99.8% purity, molecular weight 46 g/mol) was obtained from Honeywell Riedel-de-Haën (Muskegon, MI, USA). Pure water was obtained by a Milli-Q system (Milipore). 3.1.1.2. Production of ethylcellulose nanoparticles The development of ethylcellulose nanoparticles was based on the methodology described by Fessi based on the interfacial polymer deposition following solvent displacement (Briancon, et al., 1999). Briefly, EC were dissolved in ethanol during at least 4 h at room temperature. Afterwards, water (5 mL at a time) was injected into the organic phase with continuous stirring (240 rpm using a magnetic stirrer) at room temperature. The solution was kept stirring for 15 min. An experimental design (Table A.1, ANNEX A) was used to evaluate different conditions (EC concentration and antisolvent concentration) in the size and polydispersity index of the nanoparticles. The optimal formulations were chosen based on the size by intensity and polydispersity index of the nanoparticles. To produce b-carotene loaded EC nanoparticles, several concentrations of b-carotene were tested based on its maximum solubility in ethanol. All formulations tested are presented in Table A.2 (ANNEX A). The organic phase was prepared by dissolving EC and BC in ethanol and then the same methodology used for the production of blank EC nanoparticles production. 20 After the production of nanoparticles, the solvent was evaporated using a rotary evaporator (IKA®- Werke GmbH & CO. KG, Germany) at 60 ºC, under reduced pressure of at least 30 mbar until there was no ethanol left. The final volume was expected to be 40 mL, if water started to evaporate, the final volume was adjusted with more water. In some characterization tests (i.e. FTIR and X-Ray Diffraction) the samples needed to be dried before the analysis. For that, the solution of nanoparticles without ethanol was frozen over night at -22 ºC. Then, they were transferred to a freezer at -80 ºC for at least 4 h and then freeze-dried using a Freeze Dryer (Lyoquest -55 ºC Plus Eco, Telstar, Spain) for 3 days to get a solid powder of nanoparticles. 3.1.1.3. Production of zein nanoparticle Zein nanoparticle’s preparation was also based on the methodology described by Fessi. Blank zein nanoparticles were firstly prepared to determine the optimal formulations (Briançon et al., 1999). The experimental design used is presented in ANNEX A Table A.3. Three different conditions were tested, namely zein concentration, antisolvent percentage and the flow rate in which the organic phase was dropped into the aqueous phase. The formulations chosen were 0.4% zein, 90% antisolvent (water) concentration and flow rate 0.7 mL/min based on the size by intensity and polydispersity index. Briefly, 0.4% (0.4 g in 100 mL) of zein was prepared in ethanol solution (75%, v/v) making the organic phase. This solution was then added to the aqueous phase (water) in an amount sufficient for the final proportion of ethanol to be 10%. Therefore, for a total volume of 50 mL, the volume of zein solution added was 6.67 mL and the corresponding amount of water was 43.33 mL. The addition of the organic phase to the aqueous phase was controlled by a syringe pump (NE-1000 Programmable Single Syringe Pump, New Era Pump Systems Inc., USA) with a 12 mL Luer lock syringe (Soft-Ject®, Henk Sass Wolf, Germany) and a 15.80 mm diameter steel needle at a constant flow rate of 0.7 mL/min. The zein solution was added dropwise to the water at a constant stirring of 235 rpm using a magnetic stirrer and at room temperature. The final nanoparticle solution was maintained stirring for 15 min. The same methodology was used for the different concentrations of zein. For the production of b-carotene loaded zein nanoparticles, a variety of BC concentrations were tested starting on its maximum solubility in ethanol. The formulations assayed are presented in Scale-up production of food grade nanoparticles 21 ANNEX A Table A.4. After testing several concentrations, the final concentration was 0.001% BC chosen based on the size by intensity and polydispersity index. Because BC does not dissolve in aqueous ethanol, two different solutions were made for the preparation of the organic phase: i) 0.5 g of zein was dissolved in 100 mL of 75% aqueous ethanol solution and ii) 0.0025 g of BC was dissolved in 50 mL of 100% ethanol (0.005% BC concentrated solution). The two solutions were mixed by adding 10 mL of the BC solution to 40 mL of zein solution under constant stirring of 200 rpm thus making the final organic phase. Then, 6.25 mL of the latest solution was added dropwise to 43.75 mL of water, following the procedure described above. An evaporation process was followed to remove ethanol by a rotary evaporator under the temperature 60 ºC and reduce pressure of at least 40 mbar until all ethanol evaporated. In case of water evaporated, the volume would be adjusted to the expected volume of 45 mL. A freezedrying step was also used when needed following the same procedure presented for ethylcellulose nanoparticles. 3.1.2 Nanoparticle characterization 3.1.2.1. Dynamic Light Scattering analysis Parameters such as the mean size of the particles by intensity and by number, polydispersity index (PDI) and zeta potential were determined using dynamic light scattering (Nano Partica SZ-100, HORIBA, Japan). A disposable cuvette (ThermoFisher Scientific) and a carbon zeta potential cuvette were used for the measurements. All measurements were performed at 25 ºC. For each sample 5 measurements were performed, each with a duration of 180 seconds and using a detection angle of 90º. For the zeta potential determination were performed 3 measurements for 90 seconds each. 3.1.2.2. Encapsulation Efficiency The encapsulation efficiency was determined to assess the percentage of b-carotene that was successfully entrapped into the nanoparticles. Therefore, freshly produced nanoparticles were submitted to ultracentrifugation for the separation of nanoparticles from free non-encapsulated b- 22 carotene. The separation was achieved using an Ultracentrifuge (OPTIMA XE-100, Beckman Coulter Life Sciences, USA) with a type 70 Ti Fixed-Angle Titanium Rotor. Briefly, approximately 26 mL of nanoparticle solution was poured into a polycarbonate bottle and were centrifuged at 50 000 rpm, for 1 h. After the centrifugation, there was a visible solid pellet containing the nanoparticles with the encapsulated BC for both zein NPs and EC NPs and the free BC would be dispersed in the supernatant. The BC dispersed in the supernatant was analyzed spectrophotometrically in a Microtiter plate reader (Synergy H1, BioTek Instruments, USA). 200 µL of supernatant was put in triplicate in a 96 well plate (ThermoFisher Scientific Inc., USA) as well as a blank solution of pure ethanol and measured at 453 nm, which corresponds to the maximum absorbance of BC. The concentration of free BC was estimated by a previously made calibration curve. All dilutions, concentrations and respective absorbances measured for the curve are present in Table B.1 and Figure B.1 (ANNEX B). The encapsulation efficiency for both EC nanoparticles and zein nanoparticles was calculated by the following equation: EE (%) ! = ! A — B A !× 100 being, EE (%) the percentage of encapsulation efficiency, A the initial concentration of b-carotene in the suspension of particles, B the concentration of b-carotene in suspension after centrifugation. 3.1.2.3. Transmission electron microscopy Parameters such morphology, particle size distributions and aggregation were analyzed through a transmission electron microscope (JEM-2100, JEOL Ltd., Japan) operating at 200 kV accelerating voltage.!TEM micrographs were analyzed using the public domain software ImageJ. A drop of the sample solution (» 10 µl) was placed on a grid (ultra-thin carbon film on Lacey carbon support film, 400 mesh, Copper, Ted Pella Inc., USA). Then, contrast solution UranyLess EM Stain (Electron Microscopy Sciences (EMS), USA) was dropped on parafilm and the grid was placed on top of the drop so it could be stained. It was left to dry for approximately 5 hours at room temperature. Eq 3.1. Scale-up production of food grade nanoparticles 23 3.1.2.4. Fourier transform infrared (FTIR) Measurements were made using a FTIR VERTEX 80/80v spectrometer (Bruker Corporation, MA, USA) in Attenuated Total Reflectance mode (ATR) with a platinum crystal accessory in the wavelength range: 4000—400 cm-1, using 32 scans at a resolution of 4 cm-1. All data is presented in transmittance percentage after a normalization to the maximum transmittance value. The samples analyzed were empty and loaded zein and EC nanoparticles, and BC, zein and EC powder. All nanoparticle’s samples were freeze-dried before the measurements. 3.1.2.5. X-Ray Diffraction An X-ray diffractometer was used to evaluate the crystallographic structure of the samples through their diffraction pattern. Measurements were made using an X-ray diffraction system (X’Pert Pro MRD, Malvern Panalytical Ltd, UK). PANanalytical X’Pert HighScore Plus was the software used to gather data and analyse peak diffractions. Background noise was also measured. The powder sample was added to the glass slide through an adhesive glue and put on the sample holder for detection. The XRD diffractograms were acquired at room temperature, angular scans from 5º to 50º (2 𝜃 ) were performed with a Cu source, X-ray tube ( 𝜆 = 1.54056 Å) at 45 kV and 40 mÅ. The fine calibration offset for 2 𝜃 = −0.0372º. The samples analyzed were empty and loaded zein and EC nanoparticles, and BC, zein and EC powder. The nanoparticles were freeze-dried before the measurements. 3.1.3 In vitro gastrointestinal digestion An in vitro simulated digestion was performed to study the bioaccesibility of b-carotene entrapped in the polymeric nanoparticles made of ethylcellulose or zein, following the standardized static in vitro digestion protocol – an international consensus established by the COST Infogest network (Brodkorb et al., 2019a). In the protocol the sample is subjected to sequential oral, gastric and intestinal digestion using parameters of digestion (electrolytes, enzymes, bile, dilution, pH and time of digestion) based on available physiological data. The protocol can be divided into three phases: a preparation phase where enzyme activities and bile concentrations are assessed and stock solutions of simulated fluids are prepared, a digestion phase which exposes the food to oral, gastric and intestinal conditions; and the last phase is the sample preparation and treatment for 24 subsequent analysis (Brodkorb et al., 2019a). 3.1.3.1. PHASE 1 - Enzyme activities, bile concentration and stock solutions The digestion involves the action of enzymes such as amylase, pepsin, lipase, trypsin and chymotrypsin. These enzymes need to be added in a specific amount (U/mL) defined in the protocol, therefore their activities must be accurately determined. This way, the pepsin activity and trypsin activity in pancreatin were determined to calculate the amount (mg/mL) of each, needed in the in vitro digestion. Amylase was not used since there was no starch present in the digested sample. The addition of gastric lipase was omitted due to the limited access of the commercially available enzyme. Enzyme activity assays Materials Pepsin from porcine gastric mucosa was obtained from Sigma-Aldrich (St Louis, MO, USA) in powder form (product number P7012) was stored at -20 ºC. Hemoglobin from bovine blood was purchased from Sigma-Aldrich (St Louis, MO, USA) with product number H2625, stored at 4 ºC. Trichloroacetic acid with molecular weight 163.69 g/mol (product number T6399) was obtained from Sigma-Aldrich (Germany) and stored at 4 ºC. Sodium chloride with molecular weight 58.44 g/mol and product number S1639 was obtained from Sigma-Aldrich (St Louis, MO, USA). Pancreatin from porcine pancreas (product number P7545), TAME (Na-p-Tosyl-L-arginine methyl ester hydrochloride, product number T4626) and calcium chloride (product number C3306, molecular weight 110.89 g/mol) were purchased from Sigma-Aldrich (St Louis, MO, USA). Trizma base from Sigma-Aldrich (St Louis, MO, USA) with product number T1503. Hydrochloric acid purchased from ThermoFisher Scientific with molecular weight 36.46 g/mol. Pure water was obtained from a Milli-Q system (Milipore). Pepsin activity assay Pepsin activity assay is based on the spectrophotometric stop reaction method. The principle of the assay is that hemoglobin diluted in water in the presence of pepsin produces TCA-soluble Scale-up production of food grade nanoparticles 25 tyrosine peptides. One unit will produce a DAbs280 of 0.001 per minute measured at TCA-soluble products (pH 2 and 37ºC) (Brodkorb et al., 2019, supplementary information). Two different solutions were prepared previously to the assay: substrate and enzyme solutions. Hemoglobin was used as substrate dissolving 0.5 g of 25 mL of water and adjusting the pH to 2 with 0.3M HCl. To prepare the enzyme solution firstly 88 mg NaCl plus 12 mg Trizma base were dispersed in 10 mL water and the pH adjusted to 6.5 with 0.3M HCl. Then, 10 mg of pepsin were dissolved in the previous solution to obtain a final concentration of 1 mg/mL of pepsin. For the assay, pepsin was diluted to prepare 6 different concentrations: 5, 10, 15, 20, 25, 30 µg/mL. 500 µL of substrate solution was pipetted into 2 mL Eppendorf tubes and incubated for 4 minutes at 37 ºC and 300 rpm in a ThermoMixer® C (Eppendorf). 100 µL of each pepsin concentration was added and incubated for 10 min exactly. After that time, 1 mL of 5% (w/v) TCA solution was added to stop the reaction. For each enzyme concentration, a blank sample was prepared the same way, but pepsin was added after the addition of TCA, which stops the reaction. All tubes were then centrifuged to precipitate the remaining hemoglobin for 30 minutes at 6000x g . The supernatant was put on disposable cuvettes (BRAND®, Germany) and the absorbance was read at 280 nm in a nanodrop (NanoDrop 3300, Fisher Scientific, Lda.). The activity of pepsin (U/mg) was calculated by equation 3.2: Units/mg ! = ! (Abs Test ! — Abs Blank) × 1000 Dt × m × 0.001 ! Being Dt the duration of reaction in minutes (10 min), m the amount of pepsin (µg) in 1 mL of assay solution, 0.001 is the DAbs280 per unit of pepsin and 1000 the factor to convert µg to mg. Trypsin from pancreatin activity assay To measure the trypsin activity in pancreatin the kinetic spectrophotometric rate determination method was used. The principle is that TAME diluted in water in the presence of trypsin produces ptoluene-sulfonyl-L-arginine and methanol. One unit corresponds to the hydrolysis of 1 µmol of TAME per minute (pH 8.1 and 25 ºC) (Brodkorb et al., 2019, supplementary information). Substrate solution was prepared by dissolving 19 mg of TAME in 5 mL of water. The enzyme solution was prepared by dissolving 10 mg of pancreatin in 10 mL of 1 mM HCl resulting in 1 mg/mL concentration. Two dilutions were made: 0.5 mg/mL and 0.1 mg/mL. An assay solution Eq 3.2. 26 was also prepared by dissolving 51 mg of CaCl2 and 223 mg Trizma base in 40 mL of water adjusting the pH to 8.1. For the assay, 2.6 mL of assay solution and 0.3 mL of substrate solution were pipetted into disposable cuvettes (BRAND®, Germany) and put in the nanodrop for 3 minutes at 25 ºC while being mixed with a magnetic stirrer. Then 100 µL of pancreatin solution was pipetted into the cuvette and the absorbance at 247 nm was measured each 30 seconds for 10 min. For the blank, the same procedure was followed adding 100 µL of assay solution instead of enzyme solution. The slopes ΔA247 [unit absorbance/minute] were calculated for both the blank and the test reactions by using the maximum linear rate over at least 5 minutes. The trypsin activity was calculated by the equation 3.3: Units/mg ! = ! (DAbs Test ! — DAbs Blank) × 1000 !×" 3 540 × m Being DAbs Test and DAbs Blank the slope of the initial portion of the curve [unit absorbance/minute] for the enzyme and blank (without enzyme), respectively, 540 is the molar extinction coefficient (L/(mol × cm) of TAME at 247 nm, m is the amount (mg) of pancreatin in the final reaction volume and 3 is the total volume of the reaction mix, in mL. Bile concentration assay Materials Bile bovine was purchased from Sigma-Aldrich (St Louis, MO, USA) in powder form with product number B3883. The concentration of bile acids in bile were measured with a commercial assay kit also obtained from Sigma-Aldrich (St Louis, MO, USA), catalog number MAK309. The bile acid kit assay contains several pre-made solutions: an assay buffer (cat. no. MAK309A), NAD solution (cat.no. MAK309B), probe (cat.no. MAK309C), Enzyme A (cat.no. MAK309D), Enzyme B (cat.no. MAK309E) and standard solution (cat.no. MAK309F). Method Bile acids concentration was measured following the supplier’s protocol which provides a fluorometric method to measure the total bile acids. Firstly, a bile solution (60 mg/mL) was prepared with water. Dilutions were made to achieve three different final concentrations to be analyzed: 0.06 mg/mL, 0.03 mg/mL and 0.01 mg/mL. Three more stock solutions were Eq 3.3. Scale-up production of food grade nanoparticles 27 prepared: - Internal standard reagent: 20 µL of standard solution + 230 µL water; - Working reagent: 75 µL assay buffer + 8 µL NAD solution + 4 µL probe + 1 µL enzyme A + 1 µL enzyme B; - Blank reagent: 75 µL assay buffer + 8 µL NAD solution + 4 µL probe + 1 µL enzyme B. All samples were prepared in a 96 well opaque black plate and run in duplicate. Three different wells were prepared for each sample: sample, sample blank and internal standard. 20 µL of sample (different concentrations) were pipetted into the respective wells. 5 µL of water was pipetted into the sample and sample blank wells and 5 µL of internal standard reagent was pipetted into the internal standard wells. Then, 80 µL of working reagent was added to the sample and internal standard wells and 80 µL of blank reagent was added to the sample blank wells. The plate was then incubated for 20 minutes in the dark and the fluorescence intensity was read at lex=530 nm and lem=585 nm. The bile concentration was calculated using the equation 3.4: [Bile acids] ! = ! A ! — B C ! — A ! × 20 × n Being A the fluorescence intensity values of sample, B the fluorescence intensity values of sample blanks and C the fluorescence intensity values of internal standard; 20 µM is the concentration of the internal standard and n is the dilution factor. Final bile acids concentration is given in µM. Simulated digestion fluids preparation The stock electrolytic solutions were prepared according to the INFOGEST static in vitro simulation of gastrointestinal digestion protocol (Brodkorb et al., 2019a), as shown in the Table 3.1. SSF, SGF and SIF were prepared 1.25 times concentrated, considering the later dilution (4:1) with enzymes and CaCl2(H2O)2 added in the day of use to avoid precipitation. Eq 3.4. 34 3.2.2 Zein nanoparticles The optimization of the production of zein nanoparticles has as independent variables the polymer concentration, antisolvent concentration and flow rate. The assayed formulations and obtained values are presented in Table A.3 (ANNEX A). Figures 3.3 and 3.4 shows the effects of varying those variables on size by intensity and PDI, respectively. Polymer concentration varied between 0.4, 0.6 and 0.8%, antisolvent concentration varied between 80, 85 and 90% (v/v) and flow rate varied between 0.3 and 0.7 mL/min. Figure 3.3. A) Pareto chart of standardized effects. B) Plot of marginal means and conf. limits (95%). Variable: Size by intensity. Figure 3.4. A) Pareto chart of standardized effects. B) Plot of marginal means and conf. limits (95%). Variable: PDI. Scale-up production of food grade nanoparticles 35 Results shown in Figure 3.3 A) reveal that the parameters that most influence the size by intensity are mostly the antisolvent concentration and zein concentration with statistical significance (p<0.05). In Figure 3.3 B) it can be seen how these parameters influence the particles: higher concentrations of antisolvent shows a significant drop on the size by intensity, e.g., for a 0.7 mL/min flow rate and 0.4% zein concentration values of size goes from approximately 140 to 90 nm, when the antisolvent concentration goes from 80% to 90%. Lower concentrations of zein also revealed the smaller nanoparticles’ sizes. These results are similar to the ones for ethylcellulose nanoparticles and can be explained by the same phenomenon: lower polymer concentration and higher amounts of antisolvent lowers the viscosity of the suspension inducing fast miscibility of phases and smaller particles formed. Also, when less zein and more water is used, it results in less accessible zein polymer in suspension that forms nanoparticles, thus decreasing the agglomeration of particles as well as its size (Xu, Jiang, Reddy, & Yang, 2011). Li et al. (2017) reported similar results, increasing the zein concentration also resulted in increased particle size and PDI. For the polydispersity index, significant statistical differences were found for the antisolvent concentration (p<0.05). However, in Figure 3.4 B), it can be seen that the antisolvent concentration has an opposite effect of what would be expected, i.e., in this case, the higher amount of antisolvent resulted in higher PDI, varying from 0.275 to 0.200, for flow rate 0.7 mL/min and 0.4% zein concentration. Nevertheless, the PDI remained under the value of 0.5, which means good homogeneity. Because the flow rate did not have a significant influence on both variables tested, 0.7 mL/min was chosen to produce the nanoparticles due to it being higher rate. The other conditions were chosen based on the results of its combination: higher antisolvent concentration (90%) with lower zein concentration (0.4%). After the selection of the conditions to obtain the nanoparticles with small sizes and PDI values, was evaluated the encapsulation of b-carotene. 3.2.3 Encapsulation Efficiency The encapsulation efficiency of nanoparticles was tested using different concentrations of b-carotene. One of the first aspects considered was the effect of BC in nanoparticles size and PDI, 36 and it was observed that the highest concentration that allowed to obtain nanoparticles with small sizes and PDI values was 0.0004% (Table A.2., ANNEX A) for EC and 0.001% (Table A.4., ANNEX A) for zein nanoparticles. Based on these BC concentrations limits, the EE was determined for these and lower concentrations of BC. Once the encapsulation efficiency is determined by the amount of BC that remained free in the suspension, and therefore, was not encapsulated, higher encapsulation efficiencies may mean two different issues: i) low amount of BC used, not enough to attach to all polymer available that resulted in some polymer being empty, therefore no free BC remained in suspension; ii) the right amount of BC used that resulted in all the polymer being saturated with BC and therefore no free BC remained in suspension. Lower values of encapsulation efficiency are a cause of larger amounts of BC used that exceeded the saturation limit for the polymer available, leaving BC in solution. Table 3.2 presents the EE of BC obtained for EC and zein nanoparticles. Results show that the use of higher concentrations of BC resulted in a decrease of the encapsulation efficiency for EC, but the opposite for zein nanoparticles. Encapsulation efficiency curves typically reach a peak at a certain optimal BC concentration, and then drop into much lower values of encapsulation efficiencies. The preferred value for the encapsulation efficiency is the BC concentration value below the one where the encapsulation efficiency drops. Table 3.2. Percentage of encapsulation efficiency on different formulations of ethylcellulose nanoparticles and zein nanoparticles varying the BC concentration BC Concentration (%) Encapsulation efficiency (%) Ethylcellulose nanoparticles 0.00004 85.75 0.00008 78.63 0.0004 66.56 Zein nanoparticles 0.0005 82.65 0.001 92.69 Scale-up production of food grade nanoparticles 37 During the production of BC loaded EC nanoparticles, some BC would precipitate on the surface of the dispersion. This phenomenon can be explained by the large amount of BC used. Because there is a saturation limit for the incorporation of free BC into the dissolved ethylcellulose polymer, the excess BC precipitates and is not be encapsulated. This hypothesis is supported by the observation that a precipitate is formed when higher BC concentrations are used (Hayden et al., 2018). In the case of EC nanoparticles, 0.00004% is the concentration for which the maximum EE is obtained. It is important to mention that higher EE could be obtained if the concentrations of BC were lower. However, as one of the goals is to improve the bioaccessibility of BC, higher concentrations are needed, and therefore the concentration of BC 0.0004% was chosen to move forward. For the zein nanoparticles, two concentrations were tested, being chosen a concentration of 0.001% of BC since it shows a greater value of encapsulation efficiency (92.69%). The EE for higher concentrations of BC was not evaluated due to the high PDI values (>0.5) obtained for the nanoparticles (Table A.4, ANNEX A). In this case, to higher BC concentrations is observed an increase of the EE. This may mean that a higher amount of BC can be encapsulated (and the amount of zein used hasn’t yet reach its saturation limit), however based on the effect of higher BC concentrations in the zein nanoparticles, the value obtained for the EE (92.69%) could be the maximum possible value without changing the size and PDI of the nanoparticles. Similar results were obtained by Wang et al. (2018). The encapsulation efficiency of BC loaded zein nanoparticles would reach its peak approximately at 50% of encapsulation efficiency and then slightly decrease if the concentration of BC was lowered, since there was no enough BC attaching to the zein. 3.2.4 Size, PDI and Zeta Potential The size by intensity and number, PDI and zeta potential of the loaded nanoparticles was determined using the BC concentrations. The values obtained for blank EC nanoparticles and loaded EC nanoparticles were —60 ± 1.7 and —93.29 ± 2.6 mV, respectively, meaning that the samples showed good stability behavior. Blank zein nanoparticles and loaded zein nanoparticles presented values of +63.5 mV and +70.5 mV, respectively. The zeta potential can be an indicator 38 of how stable a suspension of nanoparticles is, i.e. higher absolute values indicate higher repulsion between particles, meaning that the suspension is stable. In general, zeta potential values outside the range +30 mV to —30 mV have high stability behavior (Trivedi et al., 2017). The zeta potential values obtained for unloaded and loaded NPs either for EC or zein are different ( p <0.05). Table 3.3. Size by intensity and number, PDI and zeta potential for ethylcellulose and zein nanoparticles with and without BC measured by DLS. (Data are expressed as the mean ± SD) Size by intensity (nm) Size by number (nm) PDI Zeta potential (mV) Blank EC NPs 69.4 ± 2.1 ab 49.1 ± 2.3 a 0.176 ± 0.038 a —60.7 ± 1.7 a 0.0004% BC loaded EC NPs 59.9 ± 9.0 a 39.1 ± 8.5 b 0.274 ± 0.015 b — 93.26 ± 2.6 b Blank zein NPs 81.5 ± 7.3 bc 53.8 ± 4.4 0.341 ± 0.049 b 63.5 ± 1.6 c 0.001% BC loaded zein NPs 82.7 ± 8.1 c 62.4 ± 7.9 0.286 ± 0.058 b 70.5 ± 0.7 d Values reported are the mean ± standard deviation (sd). Different letters (a–d) in the same column indicate a statistically significant difference ( p <0.05). Ethylcellulose nanoparticles shows a drop in the size by intensity, as well as number, when loaded with BC. Decreasing from 69.4 (± 2.1) to 59.9 (± 9.0) nm. However, the PDI shows higher values for the loaded particles, which may be due to the surface area variation that can occur as a result of BC addition. Zein nanoparticles showed similar results for blank and loaded nanoparticles, approximately 80 nm for the size by intensity. However, for the loaded nanoparticles, higher values were obtained for the number distribution and PDI values were lower. In general, all nanoparticles were below the size 100 nm and with PDI lower than 0.5. 3.2.5 Transmission electron microscopy Morphological analysis was performed on the produced nanoparticles. Results are shown in Figure 3.5 and Figure 3.6. TEM images provides information from the nanoparticles’ direct observation. Images are obtained through the interaction of a beam of electrons and the sample (with contrast solution previously added) (Reimer, 2013). Scale-up production of food grade nanoparticles 39 From the observation of Figure 3.5., all EC nanoparticles showed spherical shape, which was consistent with the results obtained by DLS. Nanoparticles’ diameters were roughly similar. However, it can also be seen a broader range of particles, large amounts of smaller sizes (<100 nm) and a few of larger sizes (>100 nm), which confirms the number distributions obtained by DLS. For loaded BC ethylcellulose nanoparticles, aggregation seemed to occur, and it may lead to an increase of the PDI obtained by DLS. Figure 3.5. TEM images of blank EC nanoparticles (a) and BC loaded EC nanoparticles (b) at 100 000x. Figure 3.6. TEM images of blank zein nanoparticles (a) and beta-carotene loaded zein nanoparticles (b) at 50 000x. (a) (b) (a) (b) 40 Size measurements using dynamic light scattering yielded a size by intensity of about 80 nm which was compatible with the TEM image for zein nanoparticles. In addition, various-sized nanoparticles were present, which contributed for smaller number distributions. All particles had spherical shape. When comparing the blank nanoparticles with loaded nanoparticles it is important to mention that they remained without significant changes to the morphology of the particles. 3.2.6 Fourier Transform Infrared spectroscopy Fourier transform infrared (FTIR) spectroscopy was used to determine the nanoparticles’ chemical structure at molecular scale. Each molecule has a characteristic infrared spectrograph equivalent to a fingerprint, making it possible to identify the constituting elements and bonding arrangement of a sample. In addition, FTIR can be used to find the compatibility between drug and polymers (Baudot, Tan, & Kong, 2010). Pure samples spectrum was used as a reference for evaluating their presence in the nanoparticle form. b-carotene spectra show a clear peak at 964 cm-1, which is the trans conjugated alkene CH out of plane deformation mode (Yi et al., 2015). Figure 3.7. FTIR spectra of b-carotene powder, ethylcellulose powder, ethylcellulose nanoparticles and b-carotene loaded ethylcellulose nanoparticles. Scale-up production of food grade nanoparticles 41 The absorption characteristic peaks of ethylcellulose (Figure 3.7) at 1051 cm-1corresponds to C— O—C stretching and the wavelengths 2870 cm-1 and 2974 cm-1corresponded to C—H. The band at 3471 cm-1is due to the hydroxyl (O—H) stretching vibrations (Guo & Gray, 1994; Madni et al., 2014). From pure EC to the nanoparticle form there was only a peak shift from 3471 cm-1 to 3477 cm-1 (difference greater than the resolution of 4 cm-1). These results reveal no significant change on the structure of EC when nanoparticles are produced. Even as loaded BC nanoparticles, no significant BC peaks were found. This may be due to the small concentration of BC used compared to EC concentration, since the intensity of the peaks of FTIR spectra are proportional to the concentration of the sample. Figure 3.8. FTIR spectra of b-carotene powder, zein powder, zein nanoparticles and b-carotene loaded zein nanoparticles. For pure zein, the characteristic bands of proteins can be seen in Figure 3.8. Protein bands that most stands out are amide I or C=O, usually in the range 1600-1700 cm-1, stretching at 1641 cm-1 and amide II or N—H, usually present at 1500-1530 cm-1, bending and stretching at 1516 cm-1. Amide A or N—H is stretching at 3288 cm-1; amide B or asymmetric stretching vibration of =C—H at 2928 cm-1; and amide III or C—N stretching at 1236 cm-1. Similar values can be seen in the encapsulation structures containing zein (Gómez-Mascaraque et al., 2017; Yi et al., 2015). 42 Amide A and amide III revealed different values for zein NPs and BC loaded zein NPs, however they were not significant due to the resolution being 4 cm-1. Amide B showed a shift from 2928 to 2958 cm-1 when the material was in the form of nanoparticle. This implies that the formation of nanoparticles impacts the structure and conformation of the protein (Yi et al., 2015). In the loaded BC nanoparticles, there was no obvious difference in the amide bands, implying that the BC incorporated in the polymer may not show an obvious effect on the structural conformation of the proteins from the blank nanoparticles or that the concentration of BC used in the nanoparticles hampered the unequivocal identification of BC in the FTIR spectra of encapsulated particles (Gómez-Mascaraque et al., 2017). 3.2.7 X-ray diffraction The X-ray diffraction (XRD) spectrum of a crystalline structure shows sharp peaks. Thus, the spectrum of the pure BC (Figure 3.9 and 3.10) showed that the compound is a crystalline material. However, the BC peaks were not visible in the loaded BC nanoparticles, either being EC or zein nanoparticles suggesting the BC is amorphous in both nano systems. Changes in BC crystallinity may be due to the precipitation without crystallization when in contact with the antisolvent in the nanoprecipitation procedure or during solvent evaporation (Yi et al., 2015). Figure 3.9. X-ray diffraction patterns of pure BC, pure EC, EC NPs, EC-BC NPs and background. Scale-up production of food grade nanoparticles 43 Pure EC spectra revealed sharp peaks, which confirmed the presence of crystalline structure (Khan et al., 2010). All nanoparticles’ samples showed to be amorphous. However, EC characteristic peak showed an intensity able to be observed and can be also seen in EC nanoparticles and loaded EC nanoparticles. Pure zein spectra did not show sharp peaks, instead it shows two humps, indicating the amorphous structure of the protein. A peak can be observed, however as it matches the background peak, it was not considered. Zein nanoparticles and loaded nanoparticles show similar amorphous spectra. Figure 3.10. X-ray diffraction patterns of pure BC, pure zein, zein NPs, zein-BC NPs and background. Results reported by Tao et al. (2019) revealed that nanoparticles produced by nanoprecipitation were usually amorphous and less stable during storage compared to their crystalline counterpart. Since amorphous compounds easily recrystallize in water therefore increasing aggregation, a solution desirable is a freeze-drying step to improve their stability. 3.2.8 In vitro digestion The determination of the bioaccessibility of an encapsulated compound is one of the best ways to assess if the encapsulation system will allow the bioactive compound available to be absorbed. The nanoparticle must be designed to reach the targeted tissues. Even though particles sized below 50 - The investigation and optimization of new methods of extraction the allow the accurate way to quantify b-carotene; - Increasing exposure of humans to nanosystems with its diverse properties demands to predict the toxicity and hazard capability of the produced nanoparticles as a method of characterization; - The storage of the nanoparticles as a suspension can present several problems, among them is microbiological contaminations, polymer degradation, particle aggregation and sedimentation and the loss of the functionality of the bioactive compound. Therefore, the shift for another form is necessary. A form of powder by freeze-drying is a common technique and the insurance that the suspension is resuspendable without aggregation. Furthermore, freeze-drying compliments well this work because it can be used at an industrial scale. Scale-up production of food grade nanoparticles 51 REFERENCES Agüeros Bazo, M., Esparza Catalán, I., González Ferrero, C., González Navarro, C. J., Irache Garreta, J. M., & Romo Hualde, A. (2016). Patent No. US 9,381,252 B2 . Washington, DC. Ahmed, B., Yusuf, M., Khan, M., & Khan, R. (2012). Plausible antioxidant biomechanics and anticonvulsant pharmacological activity of brain-targeted beta-carotene nanoparticles. International Journal of Nanomedicine , 7 , 4311. https://doi.org/10.2147/IJN.S34588 Albanes, D. (1999). β-Carotene and lung cancer: a case study. The American Journal of Clinical Nutrition , 69 (6), 1345S-1350S. https://doi.org/10.1093/ajcn/69.6.1345S Alminger, M., Aura, A.-M., Bohn, T., Dufour, C., El, S. N., Gomes, A., … Santos, C. N. (2014). In Vitro Models for Studying Secondary Plant Metabolite Digestion and Bioaccessibility. Comprehensive Reviews in Food Science and Food Safety , 13 (4), 413–436. Anderson, T. J., & Lamsal, B. P. (2011). REVIEW: Zein Extraction from Corn, Corn Products, and Coproducts and Modifications for Various Applications: A Review. Cereal Chemistry Journal , 88 (2), 159–173. Avanço, G. B., & Braschi, M. L. (2008). Preparation and characterisation of ethylcellulose microparticles containing propolis. Revista de Ciências Farmacêuticas Básica e Aplicada , 29 (2), 129–135. Barreras-Urbina, C. G., Ramírez-Wong, B., López-Ahumada, G. A., Burruel-Ibarra, S. E., MartínezCruz, O., Tapia-Hernández, J. A., & Félix, F. R. (2016). Nanoand Micro-Particles by Nanoprecipitation: Possible Application in the Food and Agricultural Industries. International Journal of Food Properties , 19 (9), 1912–1923. Baudot, C., Tan, C. M., & Kong, J. C. (2010). FTIR spectroscopy as a tool for nano-material characterization. Infrared Physics and Technology , 53 (6), 434–438. Boon, C. S., McClements, D. J., Weiss, J., & Decker, E. A. (2010). Factors Influencing the Chemical Stability of Carotenoids in Foods. Critical Reviews in Food Science and Nutrition , 50 (6), 515– 532. Briançon, S., Fessi, H., Lecomte, F., & Lieto, J. (1999). STUDY AND SCALE-UP OF A NANOPRECIPITATION PROCESS. In 14th International Symposium on Industrial Crystallization . IChemE. Brodkorb, A., Egger, L., Alminger, M., Alvito, P., Assunção, R., Ballance, S., … Recio, I. (2019a). INFOGEST static in vitro simulation of gastrointestinal food digestion. Nature Protocols , 14 (4), 991–1014. https://doi.org/10.1038/s41596-018-0119-1 Brodkorb, A., Egger, L., Alminger, M., Alvito, P., Assunção, R., Ballance, S., … Recio, I. (2019b). INFOGEST static in vitro simulation of gastrointestinal food digestion supplementary information. Nature Protocols , 14 (4), 991–1014. Calderó, G., Leitner, S., García-Celma, M. J., & Solans, C. (2019). Modulating size and surface charge of ethylcellulose nanoparticles through the use of cationic nano-emulsion templates. Carbohydrate Polymers , 225 , 115201. https://doi.org/10.1016/j.carbpol.2019.115201 Charcosset, C., & Fessi, H. (2005). Preparation of nanoparticles with a membrane contactor. 52 Journal of Membrane Science , 266 (1–2), 115–120. Chow, S. F., Wan, K. Y., Cheng, K. K., Wong, K. W., Sun, C. C., Baum, L., & Chow, A. H. L. (2015). Development of highly stabilized curcumin nanoparticles by flash nanoprecipitation and lyophilization. European Journal of Pharmaceutics and Biopharmaceutics , 94 , 436–449. https://doi.org/10.1016/j.ejpb.2015.06.022 Chuacharoen, T., & Sabliov, C. M. (2016). The potential of zein nanoparticles to protect entrapped β-carotene in the presence of milk under simulated gastrointestinal (GI) conditions. LWT - Food Science and Technology , 72 , 302–309. Corradini, E., Curti, P., Meniqueti, A., Martins, A., Rubira, A., & Muniz, E. (2014). Recent Advances in Food-Packing, Pharmaceutical and Biomedical Applications of Zein and Zein-Based Materials. International Journal of Molecular Sciences , 15 (12), 22438–22470. https://doi.org/10.3390/ijms151222438 D’Addio, S. M., & Prud’homme, R. K. (2011). Controlling drug nanoparticle formation by rapid precipitation. Advanced Drug Delivery Reviews , 63 (6), 417–426. Dasgupta, A., & Klein, K. (2014). Antioxidants in foods, vitamins and suplements: Prevention and Treatment of Disease . Academic Press. de Freitas Zômpero, R. H., López-Rubio, A., de Pinho, S. C., Lagaron, J. M., & de la Torre, L. G. (2015). Hybrid encapsulation structures based on β-carotene-loaded nanoliposomes within electrospun fibers. Colloids and Surfaces B: Biointerfaces , 134 , 475–482. https://doi.org/10.1016/j.colsurfb.2015.03.015 Deng, X.-X., Chen, Z., Huang, Q., Fu, X., & Tang, C.-H. (2014). Spray-drying microencapsulation of β-carotene by soy protein isolate and/or OSA-modified starch. Journal of Applied Polymer Science , 131 (12). European Comission. (2008). Regulation (EC) No 1333/2008 of the European Parliament and of the Council of 16 December 2008 on food additives. Off J Eur Union , (354), 16–33. Fernandez, A., Torres-Giner, S., & Lagaron, J. M. (2009). Novel route to stabilization of bioactive antioxidants by encapsulation in electrospun fibers of zein prolamine. Food Hydrocolloids , 23 (5), 1427–1432. Fessi, H., Puisieux, F., Devissaguet, J. P., Ammoury, N., & Benita, S. (1989). Nanocapsule formation by interfacial polymer deposition following solvent displacement. International Journal of Pharmaceutics , 55 (1), R1–R4. https://doi.org/10.1016/0378-5173(89)90281-0 Galindo-Rodríguez, S. A., Puel, F., Briançon, S., Allémann, E., Doelker, E., & Fessi, H. (2005). Comparative scale-up of three methods for producing ibuprofen-loaded nanoparticles. European Journal of Pharmaceutical Sciences , 25 (4–5), 357–367. Gómez-Mascaraque, L. G., Perez-Masiá, R., González-Barrio, R., Periago, M. J., & López-Rubio, A. (2017). Potential of microencapsulation through emulsion-electrospraying to improve the bioaccesibility of β-carotene. Food Hydrocolloids , 73 , 1–12. https://doi.org/10.1016/j.foodhyd.2017.06.019 Gunstone, F. D. (2003). Lipids for functional foods and nutraceuticals (1st ed., Vol. 13). The Oily Scale-up production of food grade nanoparticles 53 Press. Guo, J.-X., & Gray, D. G. (1994). Preparation, characterization, and mesophase formation of esters of ethylcellulose and methylcellulose. Journal of Polymer Science Part A: Polymer Chemistry , 32 (5), 889–896. https://doi.org/10.1002/pola.1994.080320510 Hayden, D. R., Kibbelaar, H. V. M., Imhof, A., & Velikov, K. P. (2018). Fully-biobased UV-absorbing nanoparticles from ethyl cellulose and zein for environmentally friendly photoprotection. RSC Advances , 8 (44), 25104–25111. https://doi.org/10.1039/C8RA02674B Hedrén, E., Diaz, V., & Svanberg, U. (2002). Estimation of carotenoid accessibility from carrots determined by an in vitro digestion method. European Journal of Clinical Nutrition , 56 (5), 425–430. Holding, D. R. (2014). Recent advances in the study of prolamin storage protein organization and function. Frontiers in Plant Science , 5 , 276. https://doi.org/10.3389/fpls.2014.00276 Horn, D., & Rieger, J. (2001). Organic Nanoparticles in the Aqueous Phase—Theory, Experiment, and Use. Angewandte Chemie International Edition , 40 (23), 4330–4361. https://doi.org/10.1002/1521-3773(20011203)40:23<4330::AID-ANIE4330>3.0.CO;2-W Hurtado-López, P., & Murdan, S. (2005). Formulation and characterisation of zein microspheres as delivery vehicles. Journal of Drug Delivery Science and Technology , 15 (4), 267–272. https://doi.org/10.1016/S1773-2247(05)50048-0 Jain, A., Sharma, G., Kushwah, V., Ghoshal, G., Jain, A., Singh, B., … Katare, O. P. (2018). Beta carotene-loaded zein nanoparticles to improve the biopharmaceutical attributes and to abolish the toxicity of methotrexate: a preclinical study for breast cancer. Artificial Cells, Nanomedicine, and Biotechnology , 46 (sup1), 402–412. https://doi.org/10.1080/21691401.2018.1428811 Johnson, B. K., & Prud’homme, R. K. (2003). Chemical processing and micromixing in confined impinging jets. AIChE Journal , 49 (9), 2264–2282. Kasaai, M. R. (2018). Zein and zein -based nano-materials for food and nutrition applications: A review. Trends in Food Science and Technology , 79 , 184–197. Khan, S. A., Ahmad, M., Aamir, M. N., Murtaza, G., Rasool, F., & Akhtar, M. (2010). Study of Nimesulide Release From Ethylcellulose Microparticles and DrugStudy of Nimesulide Release From Ethylcellulose Microparticles and Drug-Polymer Compatibility Analysis. Lat. Am. J. Pharm, 29 (4), 554–561. Khayata, N., Abdelwahed, W., Chehna, M. F., Charcosset, C., & Fessi, H. (2012). Preparation of vitamin e loaded nanocapsules by the nanoprecipitation method: From laboratory scale to large scale using a membrane contactor. International Journal of Pharmaceutics , 423 (2), 419–427. https://doi.org/10.1016/j.ijpharm.2011.12.016 Kim, M. S., Yeom, D. W., Kim, S. R., Yoon, H. Y., Kim, C. H., Son, H. Y., … Choi, Y. W. (2016). Development of a chitosan based double layer-coated tablet as a platform for colon-specific drug delivery. Drug Design, Development and Therapy , Volume11 , 45–57. https://doi.org/10.2147/DDDT.S123412 Lai, H. L., Pitt, K., & Craig, D. Q. M. (2010). Characterisation of the thermal properties of ethylcellulose using differential scanning and quasi-isothermal calorimetric approaches. 54 International Journal of Pharmaceutics , 386 (1–2), 178–184. https://doi.org/10.1016/j.ijpharm.2009.11.013 Leitner, S., Solans, C., García-Celma, M. J., & Calderó, G. (2019). Low-energy nano-emulsification approach as a simple strategy to prepare positively charged ethylcellulose nanoparticles. Carbohydrate Polymers , 205 , 117–124. Lepeltier, E., Bourgaux, C., & Couvreur, P. (2014). Nanoprecipitation and the “Ouzo effect”: Application to drug delivery devices. Advanced Drug Delivery Reviews , 71 , 86–97. https://doi.org/10.1016/j.addr.2013.12.009 Li, F., Chen, Y., Liu, S., Qi, J., Wang, W., Wang, C., … Zhang, Y. (2017). Size-controlled fabrication of zein nano/microparticles by modified anti-solvent precipitation with/without sodium caseinate. International Journal of Nanomedicine , 12 , 8197–8209. https://doi.org/10.2147/IJN.S143733 Lince, F., Marchisio, D. L., & Barresi, A. A. (2008). Strategies to control the particle size distribution of polyε -caprolactone nanoparticles for pharmaceutical applications. Journal of Colloid and Interface Science , 322 (2), 505–515. Liu, Y., Cheng, C., Prud’homme, K. R., & Fox, R. O. (2008). Mixing in a multi-inlet vortex mixer (MIVM) for flash nano-precipitation. Chemical Engineering Science , 63 (11), 2829–2842. Madni, A., Ekwal, M., Ahmad, S., Din, I., Hussain, Z., Ranjha, N., & Khan, M. I. (2014). FTIR DrugPolymer Interactions Studies of Perindopril Erbumine FTIR Drug-Polymer Interactions Studies of Perindopril Erbumine. Journal of the Chemical Society of Pakistan , 36 (6). Miladi, K., Sfar, S., Fessi, H., & Elaissari, A. (2016). Nanoprecipitation Process: From Particle Preparation to In Vivo Applications. In Polymer Nanoparticles for Nanomedicines (pp. 17– 53). https://doi.org/10.1007/978-3-319-41421-8_2 Murtaza, G. (2012). Ethylcellulose microparticles: A review. Acta Poloniae Pharmaceutica - Drug Research , 69 (1), 11–22. Neha, T., Shishir, T., & Ashutosh, D. (2017). Fourier transform infrared spectroscopy (FTIR) profiling of red pigment produced by Bacillus subtilis PD5. African Journal of Biotechnology , 16 (27), 1507–1512. https://doi.org/10.5897/AJB2017.15959 Packer, L., Kraemer, K., Obermüller-Jevic, U., & Sies, H. (2005). Carotenoids and Retinoids: Molecular Aspects and Health Issues . Champaing, Illinois: AOCS Publishing. Paiva, S. A. R., & Russell, R. M. (1999). β-Carotene and Other Carotenoids as Antioxidants. Journal of the American College of Nutrition , 18 (5), 426–433. Papalia, Í. da S., & Londero, P. M. G. (2015). Extração de zeína e sua aplicação na conservação dos alimentos. Ciência Rural , 45 (3), 552–559. Penalva, R., Esparza, I., Agüeros, M., Gonzalez-Navarro, C. J., Gonzalez-Ferrero, C., & Irache, J. M. (2015). Casein nanoparticles as carriers for the oral delivery of folic acid. Food Hydrocolloids , 44 , 399–406. Peñalva, R., Esparza, I., González-Navarro, C. J., Quincoces, G., Peñuelas, I., & Irache, J. M. (2015). Zein nanoparticles for oral folic acid delivery. Journal of Drug Delivery Science and Scale-up production of food grade nanoparticles 55 Technology , 30 , 450–457. Penalva, R., González-Navarro, C. J., Gamazo, C., Esparza, I., & Irache, J. M. (2017). Zein nanoparticles for oral delivery of quercetin: Pharmacokinetic studies and preventive antiinflammatory effects in a mouse model of endotoxemia. Nanomedicine: Nanotechnology, Biology and Medicine , 13 (1), 103–110. https://doi.org/10.1016/j.nano.2016.08.033 Pénicaud, C., Achir, N., Dhuique-Mayer, C., Dornier, M., & Bohuon, P. (2011). Degradation of βcarotene during fruit and vegetable processing or storage: reaction mechanisms and kinetic aspects: a review. Fruits , 66 (6), 417–440. Pinto Reis, C., Neufeld, R. J., Ribeiro, A. J., & Veiga, F. (2006). Nanoencapsulation I. Methods for preparation of drug-loaded polymeric nanoparticles. Nanomedicine: Nanotechnology, Biology and Medicine , 2 (1), 8–21. Porter, S. C. (1989). Controlled-Release Film Coatings Based on Ethylcellulose. Drug Development and Industrial Pharmacy , 15 (10), 1495–1521. Qian, C., Decker, E. A., Xiao, H., & McClements, D. J. (2012). Nanoemulsion delivery systems: Influence of carrier oil on β-carotene bioaccessibility. Food Chemistry , 135 (3), 1440–1447. Quintanar-Guerrero, D., Allémann, E., Fessi, H., & Doelker, E. (1998). Preparation Techniques and Mechanisms of Formation of Biodegradable Nanoparticles from Preformed Polymers. Drug Development and Industrial Pharmacy , 24 (12), 1113–1128. Rama, K., Senapati, P., & Das, M. K. (2005). Formulation and in vitro evaluation of ethyl cellulose microspheres containing zidovudine. Journal of Microencapsulation , 22 (8), 863–876. Rao, J. P., & Geckeler, K. E. (2011). Polymer nanoparticles: Preparation techniques and sizecontrol parameters. Progress in Polymer Science , 36 (7), 887–913. https://doi.org/10.1016/j.progpolymsci.2011.01.001 Reimer, L. (2013). Transmission electron microscopy: physics of image formation and microanalysis (Vol. 36). Springer. Rekhi, G. S., & Jambhekar, S. S. (1995). Ethylcellulose - A Polymer Review. Drug Development and Industrial Pharmacy , 21 (1), 61–77. Ribeiro, H. S., Chu, B.-S., Ichikawa, S., & Nakajima, M. (2008). Preparation of nanodispersions containing β-carotene by solvent displacement method. Food Hydrocolloids , 22 (1), 12–17. Rowe, R. C., Sheskey, P. J., & Quinn, M. E. (2009). Handbook Pharmaceutical Excipients (6th ed.). Pharmaceutical Press. Sadeghi, F., Garekani, H. A., & Sadeghi, R. (2003). COMPARISON OF ETHYLCELLULOSE MATRIX CHARACTERISTICS PREPARED BY SOLID DISPERSION TECHNIQUE OR PHYSICAL MIXING. DARU Journal of Pharmaceutical Sciences , 11 (1), 7–3. Sessa, D. J., Eller, F. J., Palmquist, D. E., & Lawton, J. W. (2003). Improved methods for decolorizing corn zein. Industrial Crops and Products , 18 (1), 55–65. https://doi.org/10.1016/S0926-6690(03)00033-5 Silva, H. D., Cerqueira, M. Â., & Vicente, A. A. (2012). Nanoemulsions for Food Applications: 56 Development and Characterization. Food and Bioprocess Technology , 5 (3), 854–867. Swathi, P., & Krishna Sailaja, A. (2014). Formulation of ibuprofen loaded ethyl cellulose nanoparticles by nanoprecipitation technique. Asian Journal of Pharmaceutical and Clinical Research , 7 (3), 44–48. Tao, J., Chow, S. F., & Zheng, Y. (2019). Application of flash nanoprecipitation to fabricate poorly water-soluble drug nanoparticles. Acta Pharmaceutica Sinica B , 9 (1), 4–18. https://doi.org/10.1016/j.apsb.2018.11.001 Tewa-Tagne, P., Briançon, S., & Fessi, H. (2007). Preparation of redispersible dry nanocapsules by means of spray-drying: Development and characterisation. European Journal of Pharmaceutical Sciences , 30 (2), 124–135. U.S. Food & Drug Administration. (2019). CFR - Code of Federal Regulations Title 21: zein. Retrieved September 16, 2019, from https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=184.1984 Urbán-Morlán, Z., Mendoza-Elvira, S. E., Hernández-Cerón, R. S., Alcalá-Alcalá, S., RamírezMendoza, H., Ciprián-Carrasco, A., … Quintanar-Guerrero, D. (2015). Preparation of ethyl cellulose nanoparticles by Solvent-Displacement using the conventional method and a recirculation system. Journal of the Mexican Chemical Society , 59 (3), 173–180. Valente, I., Celasco, E., Marchisio, D. L., & Barresi, A. A. (2012). Nanoprecipitation in confined impinging jets mixers: Production, characterization and scale-up of pegylated nanospheres and nanocapsules for pharmaceutical use. Chemical Engineering Science , 77 , 217–227. Vauthier, C., & Bouchemal, K. (2009). Methods for the Preparation and Manufacture of Polymeric Nanoparticles. Pharmaceutical Research , 26 (5), 1025–1058. Wan, Z.-L., Guo, J., & Yang, X.-Q. (2015). Plant protein-based delivery systems for bioactive ingredients in foods. Food & Function , 6 (9), 2876–2889. Wang, M., Fu, Y., Chen, G., Shi, Y., Li, X., Zhang, H., & Shen, Y. (2018). Fabrication and characterization of carboxymethyl chitosan and tea polyphenols coating on zein nanoparticles to encapsulate β-carotene by anti-solvent precipitation method. Food Hydrocolloids , 77 , 577– 587. https://doi.org/10.1016/j.foodhyd.2017.10.036 Wasilewska, K., & Winnicka, K. (2019). Ethylcellulose–A Pharmaceutical Excipient with Multidirectional Application in Drug Dosage Forms Development. Materials , 12 (20), 3386. https://doi.org/10.3390/ma12203386 WRIGHT, A., PIETRANGELO, C., & MACNAUGHTON, A. (2007). Influence of simulated upper intestinal parameters on the efficiency of beta carotene micellarisation using an in vitro model of digestion. Food Chemistry , 107 (3), 1253–1260. https://doi.org/10.1016/j.foodchem.2007.09.063 Xu, H., Jiang, Q., Reddy, N., & Yang, Y. (2011). Hollow nanoparticles from zein for potential medical applications. Journal of Materials Chemistry , 21 (45), 18227. https://doi.org/10.1039/c1jm11163a Yi, J., Lam, T. I., Yokoyama, W., Cheng, L. W., & Zhong, F. (2015). Beta-carotene encapsulated in food protein nanoparticles reduces peroxyl radical oxidation in Caco-2 cells. Food Scale-up production of food grade nanoparticles 57 Hydrocolloids , 43 , 31–40. https://doi.org/10.1016/j.foodhyd.2014.04.028 Yin, L.-J., Chu, B.-S., Kobayashi, I., & Nakajima, M. (2009). Performance of selected emulsifiers and their combinations in the preparation of β-carotene nanodispersions. Food Hydrocolloids , 23 (6), 1617–1622. Zhang, Y., Cui, L., Che, X., Zhang, H., Shi, N., Li, C., … Kong, W. (2015). Zein-based films and their usage for controlled delivery: Origin, classes and current landscape. Journal of Controlled Release , 206 , 206–219. https://doi.org/10.1016/j.jconrel.2015.03.030 Zhang, Y., Cui, L., Chen, Y., Zhang, H., Zhong, J., Sun, Y., … Kong, W. (2015). Zein-Based Nanofibres for Drug Delivery: Classes and Current Applications. Current Pharmaceutical Design , 21 (22), 3199–3207. Zhang, Y., Cui, L., Li, F., Shi, N., Li, C., Yu, X., … Kong, W. (2016). Design, fabrication and biomedical applications of zein-based nano/micro-carrier systems. International Journal of Pharmaceutics , 513 (1–2), 191–210. https://doi.org/10.1016/j.ijpharm.2016.09.023 58 ANNEXES Annex A Optimization of nanoparticle’s parameters I. Ethylcellulose nanoparticles experimental design Table A.1. Tested formulations assayed for the production of blank ethylcellulose nanoparticles EC concentration (organic phase) (%) Antisolvent concentration (%) Size by intensity (nm) Size by number (nm) PDI 0.1 60 163.57 109.81 0.201 0.1 80 103.18 68.47 0.287 0.2 70 182.75 126.98 0.153 0.4 60 258.29 176.31 0.242 0.4 80 187.47 127.58 0.253 Table A.2. Concentrations tested for BC loaded nanoparticles of 0.1% ethylcellulose and 80% of antisolvent concentration BC concentration (%) Size by intensity (nm) Size by number (nm) PDI 0.04 ND* ND > 0.5 0.02 ND ND > 0.5 0.01 ND ND > 0.5 0.005 ND ND > 0.5 0.001 ND ND > 0.5 0.0004 109.06 75.62 0.369 0.0001 112.23 69.86 0.213 0.00004 105.82 73.56 0.259 0.00005 101.80 63.12 0.315 0.00008 111.85 76.82 0.244 * ND – non detectable Scale-up production of food grade nanoparticles 59 II. Zein nanoparticles experimental design Table A.3. Tested formulations assayed for the production of blank zein nanoparticles Zein concentration (%) Antisolvent concentration (%) Flow rate (ml/min) Size by intensity (nm) Size by number (nm) PDI 0.4 80 0.3 131.04 88.90 0.214 0.4 80 0.7 136.74 133.42 0.195 0.4 90 0.3 90.98 63.90 0.226 0.4 90 0.7 95.47 66.84 0.285 0.6 85 0.5 128.37 82.87 0.222 0.8 80 0.3 156.06 104.44 0.181 0.8 80 0.7 162.28 112.0 0.213 0.8 90 0.3 104.5 74.42 0.243 0.8 90 0.7 104.1 70.14 0.290 Table A.4. Different concentrations tested for BC loaded nanoparticles of 0.4% zein, 90% of antisolvent concentration and 0.7 ml/min flow rate BC concentration (%) Size by intensity (nm) Size by number (nm) PDI 0.04 81.72 62.75 > 0.5 0.02 114.29 85.82 > 0.5 0.005 72.17 58.10 > 0.5 0.001 94.38 66.43 0.233 0.0005 83.91 61.26 0.211