《现代食品工程高新技术》课程教学资源（文献资料）食品粉碎、微胶囊包埋新技术 Encapsulation, protection, and delivery of bioactive proteins and peptides using nanoparticle and microparticle systems：A review David Julian McClements.pdf_大学文库

Historical perspective Encapsulation, protection, and delivery of bioactive proteins and peptides using nanoparticle and microparticle systems: A review David Julian McClements Department of Food Science, University of Massachusetts Amherst, Amherst, MA 01003, USA article info abstract Available online 16 February 2018 There are many examples of bioactive proteins and peptides that would benefit from oral delivery through functional foods, supplements, or medical foods, including hormones, enzymes, antimicrobials, vaccines, and ACE inhibitors. However, many of these bioactive proteins are highly susceptible to denaturation, aggregation or hydrolysis within commercial products or inside the human gastrointestinal tract (GIT). Moreover, many bioactive proteins have poor absorption characteristics within the GIT. Colloidal systems, which contain nanoparticles or microparticles, can be designed to encapsulate, retain, protect, and deliver bioactive proteins. For instance, a bioactive protein may have to remain encapsulated and stable during storage and passage through the mouth and stomach, but then be released within the small intestine where it can be absorbed. This article reviews the application of food-grade colloidal systems for oral delivery of bioactive proteins, including microemulsions, emulsions, nanoemulsions, solid lipid nanoparticles, multiple emulsions, liposomes, and microgels. It also provides a critical assessment of the characteristics of colloidal particles that impact the effectiveness of protein delivery systems, such as particle composition, size, permeability, interfacial properties, and stability. This information should be useful for the rational design of medical foods, functional foods, and supplements for effective oral delivery of bioactive proteins. © 2018 Elsevier B.V. All rights reserved. Keywords: Microencapsulation Insulin Lipase Lactase Nanoparticles Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Protein characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.1. Molecular dimensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2. Electrical characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.3. Polarity, solubility, and surface activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.4. Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3. Challenges to oral protein delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.1. Product stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.1.1. Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.1.2. Potential solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.2. Gastrointestinal stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.2.1. Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.2.2. Potential solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.3. Absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.3.1. Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.3.2. Potential solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 4. Product requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.1. Matrix compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.2. Product stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.3. Dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.4. Gastrointestinal stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Advances in Colloid and Interface Science 253 (2018) 1–22 E-mail address: mcclements@foodsci.umass.edu. https://doi.org/10.1016/j.cis.2018.02.002 0001-8686/© 2018 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Advances in Colloid and Interface Science journal homepage: www.elsevier.com/locate/cis

4.5. Ingredient selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.6. Production economics and feasibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5. Particle characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5.1. Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5.2. Size and shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 5.3. Interfacial properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 5.4. Aggregation state . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6. Particle functionality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6.1. Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6.2. Retention/release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6.2.1. Simple diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6.2.2. Swelling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 6.2.3. Specific molecular interactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 6.2.4. Particle dissociation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 6.3. Bioactive protection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 6.4. Particle stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 6.4.1. Gravitational stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 6.4.2. Aggregation stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 6.5. Particle permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 7. Particle impact on end product quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 7.1. Appearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 7.2. Texture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 7.3. Mouthfeel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 8. Delivery system selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 8.1. Microemulsions and emulsified microemulsions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 8.2. Emulsions, nanoemulsions, and multiple emulsions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 8.3. Solid lipid nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 8.4. Liposomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 8.5. Biopolymer microgels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 9. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 9.1. Hormones: insulin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 9.2. Digestive enzymes: lipase and lactase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 9.3. Vaccines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 9.4. Antimicrobials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 9.5. ACE inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 10. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1. Introduction There is great interest in the oral delivery of various types of bioactive proteins and peptides because of their potential health benefits, such as hormones, enzymes, vaccines, antimicrobials, and nutraceuticals [1–7]. For the sake of concision, these types of biologically active proteins and peptides will be referred to collectively as “bioactive proteins” for the remainder of this article. Bioactive proteins exhibit a broad spectrum of biological activities, which make them of interest for application in foods, supplements, and medicines. For instance, oral delivery of lactase aids in the breakdown of lactose into galactose and glucose within the small intestine, which is important for individuals with lactose intolerance [8,9]. Similarly, oral delivery of lipase can help patients with pancreatitis, i.e., the inability to breakdown lipids within the small intestine [10]. Bioactive proteins may also include various kinds of hormones, such as insulin and glucagon-like peptide-1 (GLP-1) which are used to treat diabetes [11] or angiotensin converting enzyme (ACE) inhibitors which are used to treat hypertension [12]. Certain peptides have strong antimicrobial activity, and can therefore be utilized as therapeutic agents [5]. However, there are a number of important technical challenges that have to be overcome before these bioactive proteins can be successfully delivered through the oral route. Typically, bioactive proteins must have a specific three-dimensional structure to exhibit their beneficial biological activities [13]. Proteins may undergo appreciable changes in their molecular structure within commercial products (such as functional foods, supplements, and drugs) during manufacturing, transport or storage, as well as inside the gastrointestinal tract (GIT) after ingestion. These structural changes may be brought on by alterations in environmental conditions, such as pH, ionic strength, denaturants, temperature, and enzyme activity [14]. In particular, many proteins are susceptible to degradation within the highly acidic and protease-rich environment of the human stomach [15]. Consequently, bioactive proteins often have to be encapsulated so as to protect them during storage and after ingestion, but then release them at the appropriate site of action within the human body [3,16,17]. Numerous types of colloidal delivery systems with different structural designs have been developed to encapsulate bioactive proteins (Fig. 1), with each system having its own advantages and disadvantages. The selection of the most efficacious oral delivery system for a specific application depends on a thorough understanding of the factors that impact the loading, retention, stability, and release of the proteins in that specific system. The aim of this review article is therefore to provide a critical evaluation of some of the most important factors that impact the development of oral delivery systems for bioactive proteins based on food-grade nanoparticles and microparticles. This type of colloidal particle is assembled from food-grade ingredients (such as proteins, polysaccharides, lipids, surfactants, and mineral oils) using food-grade processing operations. These colloid systems could therefore be widely utilized in medical foods, functional foods, or supplements specifically designed to deliver bioactive proteins via the oral route. 2. Protein characteristics The first factor to consider when identifying an appropriate colloidal delivery system for a specific application is the molecular and physicochemical properties of the bioactive proteins to be encapsulated. 2 D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22

Bioactive proteins vary considerably in their molecular weights, conformations, electrical characteristics, polarities, and stabilities [13], which will impact their loading, retention, stability, and release in colloidal delivery systems. In this section, a brief overview of the impact of molecular and physicochemical properties of bioactive proteins that may impact the design of colloidal delivery systems is given. 2.1. Molecular dimensions The dimensions of proteins in aqueous solutions depend on their molecular weight, conformation, and aggregation state, and may have a major impact on their retention and release within colloidal delivery systems. The molecular weight of individual bioactive proteins may vary from around 1 kDa for relatively small peptides to around 100 kDa for relatively large proteins. The conformations of bioactive proteins are mainly determined by their specific biological functions, and can be conveniently classified as globular, random coil or helical [13]. At the same molecular weight, the dimensions of proteins in solution depend strongly on the configuration they tend to adopt, with globular proteins being considerably smaller than random oil or helical proteins. Proteins may exist as individual molecules, small clusters, or large aggregates depending on solution conditions, such as pH, ionic strength, and temperature [18,19]. Consequently, proteins may vary considerably in their molecular dimensions, from a few nm (for small isolated globular proteins) to a few 100 nm (for aggregated proteins). Knowledge of the molecular dimensions of proteins under different solution conditions is therefore important for developing appropriate delivery systems. For W/O microemulsions or emulsions (Fig. 1), a bioactive protein should be smaller than the hydrophilic domains (water droplets) inside this kind of colloidal delivery system if it is going to be successfully encapsulated [20]. Conversely, for polymeric colloidal particles such as microgels (Fig. 1), a bioactive protein should be considerably larger than the pore size if it is going to be trapped inside the particles through steric hindrance effects. The impact of the molecular dimensions of proteins on their retention and release from polymeric colloidal particles is discussed in a later section. 2.2. Electrical characteristics The electrical characteristics of proteins are also important in determining their encapsulation properties, as changes in electrostatic interactions between proteins and colloidal particles are often used to tune their retention and release properties [21–23]. The electrical properties of proteins depend on the number of exposed anionic (e.g., \\COOH ↔ \\COO− + H+) and cationic (e.g.,\\NH2 + H+ ↔\\NH3 +) groups on their surfaces, and the pH of the surrounding solution. Typically, the electrical charge goes from positive to neutral to negative as the pH is increased from below to above the isoelectric point (pI) of the protein (Fig. 2). Some of the isoelectric points of common bioactive proteins are included in Table 1. Experimentally, the electrical characteristics of proteins can be conveniently characterized by measuring the change in ζ-potential with pH using micro-electrophoresis methods. Knowledge of the electrical characteristics of bioactive proteins can be extremely important in designing effective colloidal delivery systems. For instance, the retention and release of a bioactive protein from a polymeric colloidal particle depends on the charge characteristics of the biopolymers from which it is constructed. Thus, proteins will be attracted to networks consisting of anionic biopolymers (such Fig. 1. Schematic diagrams of some common types of colloidal delivery systems for encapsulation of hydrophilic bioactive proteins so that they can be incorporated into aqueous-based products. D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22 3

as alginate, carrageenan, or pectin) at pH values below their pI, but released at higher pH values [24,25]. Conversely, it would be expected that proteins would be attracted to networks consisting of cationic biopolymers (such as chitosan or polylysine) above their pI, but released at lower values. It should be highlighted that the strength of electrostatic interactions is weakened in the presence of salts due to electrostatic screening effects [26]. Consequently, for practical applications, it may be difficult to retain bioactive proteins inside colloidal particles using this approach because of the relatively high salt contents found in many commercial products. 2.3. Polarity, solubility, and surface activity The polarity of proteins is important because it determines their solubility characteristics, as well as their interactions with each other and with other substances. Proteins vary from being very hydrophilic to very hydrophobic depending on the relative proportion of polar and non-polar groups exposed at their surfaces [27]. Consequently, bioactive proteins may be either soluble or insoluble in aqueous solutions depending on their surface polarities (Table 1). Some bioactive proteins have good surface activity because they have an appropriate balance of polar and non-polar groups on their surfaces, i.e., they can adsorb to air-water, oil-water, or solid-water interfaces [28]. Knowledge of the polarity of proteins can be particularly important for designing effective colloidal delivery systems. 2.4. Stability The native structure of proteins may be altered appreciably when environmental conditions are changed, such as pH, ionic composition, or temperature [29,30]. Changes in protein conformation often lead to a loss in biological activity, and therefore it is important to clearly establish the major physical and chemical factors that impact the stability of the protein to be encapsulated. Globular proteins undergo appreciable conformational changes when they are heated above their thermal denaturation temperature (Tm), whose value depends on protein type and local environmental conditions (such as pH, ionic strength, and dielectric constant) [27]. In addition, they may undergo conformational changes when they adsorb to certain interfaces, which is known as surface denaturation, and can also lead to a loss of activity [31,32]. They may also become denatured under highly acidic or alkaline conditions [33], when exposed to certain types of salts [34], or in the presence of certain types of surfactant [35]. These changes in structure and activity may be reversible or irreversible depending on the nature of the protein and the environmental conditions. Consequently, it is important to identify the range of temperatures, pH values, and ingredient interactions where a bioactive protein retains its activity. Colloidal delivery systems can sometimes be designed to extend this range of conditions, and therefore enhance protein stability and functionality. 3. Challenges to oral protein delivery There are a number of major hurdles that must be overcome before bioactive proteins and peptides can be successfully delivered via the oral route [36,37]. In this section, some of the most important hurdles are highlighted, as well as some possible strategies to overcome them. 3.1. Product stability 3.1.1. Challenge Bioactive proteins may be incorporated into functional foods, supplements, or medical foods that have different physicochemical properties and storage requirements. For instance, proteins may be delivered in the form of fluids, gels, pastes, powders, or bulk solids, which may be exposed to different temperature, light, oxygen, and humidity levels. As a result, the proteins may become denatured, aggregated, or hydrolyzed during the manufacture, storage, transport or utilization of commercial products, thereby reducing their biological activity and efficacy [38–41]. Consequently, knowledge of the composition and structure of the matrix surrounding proteins in commercial products is important, as well as information about the environmental stresses that they might encounter during the lifetime of the product. In addition, the range of conditions where the bioactive proteins maintain their structure and activity should also be clearly defined. 3.1.2. Potential solutions Knowledge of the environmental factors and ingredient interactions that adversely alter the structure and activity of a specific bioactive protein can be utilized to design a product matrix and processing operations that will minimize any damage to it. For instance, if the temperature, pH, and water-activity ranges that promote protein denaturation are known, then the product can be designed to avoid these Table 1 Molecular and physicochemical properties of some important proteins and peptides that may need to be delivered orally. Note, thermal denaturation data may depend on protein type, pH and ionic strength, and should just be used as a guide. Protein MW (Da) Conformation pI Tm (°C) Polarity Biological activity Lipase (pancreatic) 51,000 Globular 4.9 70 Amphiphilic Digestive enzyme: hydrolyzes lipids Lactase 465,400 Globular (tetramer) 4.61 86 Amphiphilic Digestive enzyme: hydrolyzes lactose Amylase 55,000 Globular 6.5–7.0 61 Amphiphilic Digestive enzyme: hydrolyzes starch Insulin 5808 Dimer 5.3 76 Amphiphilic Hormone: modulates glucose levels in the blood GLP-1 3298 Flexible coils 4.6 – Amphiphilic Hormone: enhances insulin secretion Ghrelin 3371 Flexible coils 11.5 – Amphiphilic Hormone: regulates appetite Nisin 3354 Flexible coils 8.5 – Hydrophobic Antimicrobial: inhibits microorganisms Lysozyme 14,000 Globular 11 72 Amphiphilic Antimicrobial: inhibits microorganisms Lactoferrin 77,000 Globular 8.7 61 and 93 Amphiphilic Antioxidant: inhibits oxidation of lipids Fig. 2. The electrical charge on proteins typically goes from positive at low pH to negative at high pH, with a point of zero charge at intermediate pH, which is known as the isoelectric point. Key: β-Lg – β-lactoglobulin; LF = lactoferrin. (Data from, Mao and McClements, 2011, Food Hydrocolloids 25 (2011) 1201–1209). 4 D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22

environmental stress factors. In some cases, encapsulation of bioactive proteins in delivery systems can be used to improve the stability of bioactive proteins by altering their local environment [42,43]. 3.2. Gastrointestinal stability 3.2.1. Challenge Many bioactive proteins are highly susceptible to denaturation, aggregation, and hydrolysis when exposed to gastrointestinal fluids [39,41]. In particular, the highly acidic environment of the gastric fluids within the stomach may promote protein denaturation and aggregation [44,45]. The extent of these effects depends on the structure and properties of the bioactive proteins, as well as the nature of any foods consumed with them. In addition, digestive enzymes (such as proteases) in the mouth, stomach, and small intestine can promote hydrolysis of proteins [46,47]. The extent, rate, and nature of hydrolysis depend on the molecular structure of the bioactive proteins involved, as well as their environment. Consequently, many bioactive proteins may lose their biological activity when they are exposed to the fluids within the gastrointestinal tract. 3.2.2. Potential solutions Knowledge of the impact of specific GIT conditions on the structure and activity of a bioactive protein can be utilized to design an effective delivery system that inhibits these changes. For instance, if a bioactive protein is normally denatured and hydrolyzed in the stomach due to the high acidity and enzyme activity of the gastric fluids, then a delivery system can be developed to isolate it from these stressors. For instance, it has been shown that digestive enzymes (such as lactase and lipase) can be trapped inside biopolymer microgels that maintain a neutral internal pH under gastric conditions, which greatly enhances their stability and activity [42,43]. 3.3. Absorption 3.3.1. Challenge Another major factor that limits the efficacy of bioactive proteins is their relatively low absorption within the gastrointestinal tract [39,41]. The proteins must diffuse through the gastrointestinal fluids and across the mucus layer before they reach the surfaces of the epithelium cells (Fig. 3). The rheological properties of the gastrointestinal fluids impact the mixing and transport of the bioactive proteins, thereby impacting their residence time within certain regions of the GIT, as well as their absorption rate. The gastrointestinal fluids may vary from relatively low viscosity fluids to highly viscous gels depending on the type and amount of foods consumed [48–50]. The mucus layer consists of a highly viscous network of cross-linked mucin molecules and other substances (e.g., enzymes, lipids, and mineral ions) that coats the GIT and protects it from damage [51–53]. Bioactive proteins (or the colloidal particles containing them) may not be able to enter the mucus layer, or they may be trapped within the mucus layer and hydrolyzed by digestive enzymes before reaching the epithelium cells [37]. Moreover, once the bioactive proteins do encounter the epithelium cells they must be absorbed, which is often challenging because of their relatively large size and polarity. In principle, absorption may occur through a variety of physiological mechanisms, including transcellular (passive or active), paracellular (T-junctions), and endocytosis mechanisms [37]. Proteins (or colloidal particles containing them) may be absorbed by enterocytes or M-cells depending on their dimensions and surface characteristics. Typically, the overall extent of intact protein absorption is relatively low because of these barriers, and so special approaches must be developed to increase it. 3.3.2. Potential solutions A number of different methods can be used to increase the absorption of bioactive proteins by the epithelium cells. First, permeation enhancers can be included in a protein delivery system that increase Fig. 3. Proteins or protein-loaded particles must move through the lumen and mucus layer and be absorbed by the epithelium cells before they can reach the systemic circulation. D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22 5

the permeability of the cell membranes (transcellular) or that open up the tight junctions separating the cells (paracellular), thereby promoting greater protein absorption [54]. Second, efflux inhibitors can be included in a protein delivery system that blocks the active transport mechanisms within the cell membrane that normally expel proteins or particles out of the epithelium cells [36]. Third, the proteins can be encapsulated within colloidal particles that are absorbed by the cells, and then released into the systemic circulation [37,55]. The design of colloidal delivery systems to achieve this goal depends on a good understanding of the various cellular absorption mechanisms, and the factors that impact them (such as particle size, shape, charge, and polarity). 4. Product requirements Once the molecular and physicochemical properties of the bioactive proteins have been clearly defined, and the challenges to their delivery have been identified, it is then necessary to specify the requirements of the end product, which will depend on the particular application. In the case of medical foods, functional foods, or supplements, one should define the required appearance, texture, mouthfeel, and stability characteristics of the end product. For example, the bioactive protein may be delivered in the form of a cloudy low viscosity beverage, a transparent gummy type product, or an opaque solid tablet. In addition, it is important to define the functional attributes of the end product. For instance, it may have to protect the bioactive protein from degradation within the end product, mouth and stomach, but then release it within the small intestine. Some of the most important factors that should be considered when developing a delivery system for bioactive proteins intended for oral ingestion are highlighted in this section [56]. 4.1. Matrix compatibility If the delivery system is going to be incorporated into a medical food, functional food, or supplement intended for oral ingestion, then it should not adversely impact the desirable quality attributes of the end product, such as its appearance, texture, mouthfeel, taste, or shelf-life (see later). Particle characteristics, such as their concentration, size, shape, and charge, will determine their impact on end product properties. 4.2. Product stability The delivery system should prevent any undesired changes in the activity of the bioactive protein during the manufacture, storage, and utilization of the end product. The approach used to ensure protein stability will depend on the nature of the product, e.g., whether it is a fluid, gel, or solid. Moreover, the delivery system itself should be resistant to any undesired changes in its properties throughout the lifetime of the end product. Consequently, the delivery system may have to be designed to be resistant to changes in pH, ionic strength, temperature, light, oxygen, and mechanical stresses. This can be achieved by careful selection of the composition and structure of the colloidal delivery system, as well as by controlling the composition and structure of the food matrix and packaging material. 4.3. Dose The delivery system should be capable of encapsulating the level of bioactive proteins required to have the intended biological effect, and then consistently delivering the proteins to the intended site of action at this level. The level of bioactive proteins delivered will depend on the loading capacity, retention, and release properties of the colloidal particles. Factors that may affect the reliability of the dose received should also be carefully considered, and the delivery system should be designed to overcome any problems, e.g., variable processing or storage conditions, or food matrix effects. 4.4. Gastrointestinal stability After ingestion, the delivery system should be designed to protect the bioactive proteins from degradation within certain regions of the GIT (such as the mouth and stomach), but then release them in other regions (such as the small intestine). In addition, the delivery system may have to be designed to have a prolonged residence time in the region of the GIT where the bioactive proteins are supposed to be absorbed, which may require that the colloidal particles have mucoadhesive properties. 4.5. Ingredient selection Colloidal delivery systems intended for oral ingestion may be fabricated from a variety of synthetic and/or natural constituents, including surfactants, phospholipids, proteins, polysaccharides, and lipids. For certain applications, it may be important to select particular types of ingredients, e.g., for individuals who have vegan, vegetarian, Kosher, or non-allergenic dietary requirements. In addition, the cost, shelf-life, ease of use, and reliability of the ingredients used to assemble the colloidal delivery system should be considered. 4.6. Production economics and feasibility The colloidal delivery system should be capable of being consistently produced at an appropriate scale and cost. Many of the methods of producing colloidal delivery systems described in the literature involve ingredients or processing operations that are too costly or inappropriate for commercial applications. 5. Particle characteristics Once the properties of the bioactive proteins to be encapsulated have been clearly defined, and the requirements of the end product have been specified, then it is necessary to establish the particle characteristics required to create an appropriate oral delivery system [56]. In this section, some of the most important particle properties that may impact the efficacy of colloidal delivery systems for bioactive proteins are highlighted. 5.1. Composition Colloidal particles can be assembled from a variety of food-grade ingredients, including proteins, polysaccharides, lipids, surfactants, and minerals [57–59]. The ingredients used to fabricate the colloidal particles impact their functional attributes, and so ingredient selection is an important consideration when developing colloidal delivery systems for bioactive proteins. For instance, the composition of colloidal particles impacts the region they are digested in the GIT, as well as their ability to inhibit protein degradation. Some of the most important ways that particle composition impacts the encapsulation, protection, and release of bioactive proteins are discussed in later sections. 5.2. Size and shape The size and shape of the colloidal particles used to encapsulate bioactive proteins should also be carefully selected for the particular application. The size of colloidal particles may vary from around 10 nm for small nanoparticles (such as microemulsions) to around 1 mm for large microparticles (such as hydrogel beads), and depends on the ingredients and processing operations used to fabricate them. The colloidal particles in delivery systems are often spherical, but they can have other shapes, such as ellipsoid, cylindrical, or irregular, which can impact the optical, rheological, stability, and release characteristics of colloidal dispersions. The impact of the size and shape of colloidal 6 D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22

particles on their ability to encapsulate and deliver bioactive proteins is discussed in later sections. 5.3. Interfacial properties The interfacial properties of colloidal particles can be manipulated by fabricating them from different ingredients, or by coating them with other ingredients after they have been formed. Consequently, the thickness, composition, charge, and permeability of the interfacial layer can be controlled, which allows one to manipulate the retention, protection, and release of encapsulated bioactive proteins. The interfacial properties of colloidal particles with some lipophilic character (such as lipid droplets or hydrophobic protein nanoparticles) can be modified after fabrication by adsorbing surface-active emulsifiers to their surfaces, such as surfactants, phospholipids, proteins, or polysaccharides [60,61]. The interfacial properties of colloidal particles with an electrical charge can be altered by depositing oppositely charged substances, such as biopolymers or solid particles, onto their surfaces [62]. This electrostatic deposition method can be used to coat colloidal particles with multiple layers of charged substances, which allows the thickness, permeability, and electrical properties of the interfacial layers to be manipulated. 5.4. Aggregation state Colloidal particles may be present within a colloidal dispersion as individual entities or as clusters (“flocs”). The aggregation state of the particles in a system may have a major impact on their functional attributes. For example, it influences the stability of the particles to gravitational separation (flocs usually move faster than individual particles because of their larger size) and rheology (flocs usually give a higher viscosity than individual particles because they trap more solvent). Moreover, aggregated particles may be digested more slowly that individual particles in the GIT, which can impact the stability and release of bioactive proteins [63,64]. Consequently, it is often important to control the aggregation state of the colloidal particles in the product, as well as in the GIT. 6. Particle functionality In this section, the most important functional attributes of the particles in colloidal delivery systems are highlighted, with special reference to their relevance to the encapsulation, protection, and delivery of bioactive proteins. 6.1. Loading An important property of any protein delivery system is the maximum amount of bioactive protein that can be successfully loaded into the colloidal particles, i.e., the loading capacity (LC) [60]. The LC will determine the level of colloidal particles required to achieve the intended dose of the bioactive protein. Moreover, the fraction of bioactive protein that is actually incorporated into the colloidal particles (rather than remaining outside of them) during the encapsulation process is also important, i.e., the encapsulation efficiency (EE). The EE will determine the fraction of bioactive protein that is lost during the manufacturing process, which obviously has important economic consequences. The following expressions can be used to calculate these values for a particular bioactive protein-colloidal particle combination: LC ¼ mB;E=mP ð1aÞ EE ¼ mB;E=mB;T ð1bÞ Here, mB,E and mB,T are the masses of the encapsulated and total bioactive protein used to produce the colloidal delivery system, and mP is the total mass of the colloidal particles (carrier material + bioactive proteins). The loading capacity and encapsulation efficiency depend on the molecular and physicochemical properties of the bioactive proteins, as well as of the carrier materials used to assemble the colloidal delivery system. Most bioactive proteins are predominantly hydrophilic and so colloidal particles should have some hydrophilic domains within them, which means they must be assembled from ingredients that have appreciable numbers of polar groups (such as hydroxyl, carboxyl, sulfate, phosphate, or amino groups). Some examples of colloidal delivery systems with hydrophilic domains (usually water) are reverse micelles, W/O microemulsions, W/O emulsions, W/O/W emulsions, liposomes, and microgels (Fig. 1). Bioactive proteins may also be held inside colloidal particles by electrostatic interactions between oppositely charged groups or by hydrophobic interactions between non-polar groups (see next section). In this case, the ingredients used to form the interior of the colloidal particles should have electrically charged or non-polar groups that are strongly attracted to the bioactive proteins. In some cases, the sign or magnitude of the interactions between bioactive proteins and carrier materials can be altered by changing environmental conditions (such as pH, ionic strength or temperature), which can be used to develop triggered release mechanisms. As a specific example, the encapsulation of bioactive proteins within W1/O/W2 emulsions is considered, where W1 and W2 are the internal and external aqueous phases, and O is the oil phase (Fig. 1). Assuming that the bioactive proteins cannot diffuse through the oil phase and that the internal water droplets are not disrupted during the second homogenization stage, the encapsulation efficiency would be relatively high (EE ≈ 100%) because all of the proteins would remain trapped within the internal water phase. The loading capacity depends on the highest level of protein that can be dissolved in the internal water phase and the highest level of water droplets that can be incorporated in the W/O emulsion. If it is assumed that the maximum amount of protein that can be dissolved into an aqueous solution is around 20% and the maximum level of water droplets that can be packed into an oil phase is around 50%, then the loading capacity (LC) of the capsules in a W/O/W emulsion would be around 10%. Similar kinds of calculations could be carried out for other kinds of colloidal delivery systems. However, in practice it is always important to measure the masses of encapsulated and non-encapsulated bioactive protein to determine the EE and LC values more accurately. 6.2. Retention/release Colloidal delivery systems are often designed to retain bioactive proteins under one set of environmental conditions, but then release them when exposed to another set of conditions, such as a change in pH, ionic strength, temperature, or enzyme activity [60]. A number of physicochemical mechanisms can be utilized to control the retention and release of bioactive proteins (Fig. 4). 6.2.1. Simple diffusion In the simplest case, bioactive proteins may be released from polymeric colloidal particles by simple diffusion. To a first approximation, the release of a bioactive protein from a spherical particle due to simple diffusion can be described by the following expression [65]: M tð Þ Mð Þ ∞ ¼ 1− exp −1:2π2DP r2 t ð2Þ Here, M(t) and M(∞) are the concentrations of the bioactive protein trapped within the colloidal particles at time t and at equilibrium (infinite time), r is the radius of the colloidal particles, and DP is the diffusion coefficient of the bioactive proteins through the colloidal particles. Bioactive proteins are often encapsulated within biopolymer microgels whose interior consists of a network of crosslinked polymer molecules with a certain pore size. In this case, the diffusion coefficient can be obtained using the following expression, D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22 7

which describes the diffusion of small molecules through a network of polymer chains [66,67]: DP ¼ DW exp −π rH þ rf ξ þ 2rf ! ! ð3Þ Here, DP and DW are the diffusion coefficients of the bioactive protein through the polymer network and through pure water, rH is the hydrodynamic radius of the bioactive protein,rf is the cross-sectional radius of the polymer chains, and ζ is the pore diameter. The translational diffusion coefficient of proteins through water can be estimated using the following equation: Dw ¼ kBT=6πηrH ð4Þ Here, kB is Boltzmann's constant, T is absolute temperature, and η is the viscosity of water. These equations can provide valuable insights into the impact of the pore size and external dimensions of polymeric colloidal particles on protein retention and release. A prediction of the influence of the protein dimensions relative to the pore size (rH/ζ) on the normalized diffusion coefficient (DP/DW) is shown in Fig. 5. This prediction indicates that the bioactive proteins must have dimensions appreciably larger than those of the pores in the polymer network before their diffusion is appreciably retarded. Consequently, polymeric colloidal particles would only retain proteins when they have pore sizes appreciably smaller than that of the bioactive proteins. This may be difficult to achieve for small peptides (d b 1 nm), but may be possible for larger proteins or protein aggregates. This equation also predicts that the release of bioactive proteins from colloidal particles decreases as the size of the particles increases (Fig. 6), since the proteins have a greater distance to travel before they reach the external environment. Conversely, the retention of proteins will increase as the particle size Fig. 4. Encapsulated bioactive proteins can be released from a colloidal particle due to a variety of mechanisms, including changes in molecular interactions, an increase in pore size, particle dissociation (network disintegration), or simple diffusion. Fig. 5. The restricted diffusion of bioactive proteins through a hydrogel depends on the hydrodynamic radius of the proteins (rH) relative to the pore size (ζ) of the polymer network. Here the diffusion coefficient of the proteins through the polymer network (DP) is normalized relative to their diffusion coefficient in water (DW). 8 D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22

increases for the same reason. Predictions made using the above equation show that the release of proteins from polymeric colloidal particles would be quite rapid (b5 min), even if they had relatively small pore sizes (ζ = 1 nm) and large external radii (r = 1000 μm) (Fig. 6). This simple calculation therefore suggests that other physicochemical mechanisms are required to ensure that proteins are retained within smaller particles, e.g., specific molecular interactions or physical barriers. It should be noted that different kinds of mathematical models may be required to describe the release of proteins from other kinds of colloidal delivery systems, such as W/O/W emulsions, emulsified microemulsions, or liposomes (Fig. 1). Indeed, it is likely that simple diffusion is not the most important release mechanism for these systems. 6.2.2. Swelling The pore size in a polymeric colloidal particle can sometimes be altered by changing the environmental conditions (such as pH, ionic strength, or temperature) to induce swelling or shrinkage of the particles [68,69]. For instance, polymer networks comprised of ionized polyelectrolytes tend to swell under conditions where there is a strong electrostatic repulsion between the polymer chains, i.e., at low ionic strengths or at pH values where the polymer has a high charge density. Conversely, these types of polymer networks tend to shrink when the solution conditions reduce the electrostatic repulsion between the polymer chains, i.e., at high ionic strengths or at pH values where the polymer has a low charge density. This phenomenon has been reported in alginate microgels that tend to swell under neutral conditions because of the high negative charge density on the alginate chains, but shrink under acid conditions because the carboxyl groups on the alginate chains lose some of their charge [70,71]. Changing the ionic strength of the aqueous solution surrounding polymeric particles containing charged polymers may also lead to swelling or shrinking. For instance, it has been shown that adding salts (sodium or calcium chloride) to transglutaminase cross-linked caseinate gels caused them to shrink, due to a reduction in the electrostatic repulsion between the biopolymer chains [72]. This phenomenon can be utilized to develop delivery systems that can trigger the release of bioactive proteins in response to a change in ionic strength. Polymeric particles can also be prepared that will swell or shrink when the temperature is changed, because this alters the conformation of the polymer chains from expanded to collapsed [69]. A classic example of natural food-grade colloidal particles that undergo swelling upon heating is starch granules. Native starch granules have dense structures with small pore sizes at low temperatures, but they swell when they are heated in the presence of water [73,74]. Starch granules have been used to encapsulate various types of bioactive components by incubating them with a solution of the bioactives [74–76]. The swelling and shrinking of polymeric particles can be utilized to load, retain, and release bioactive proteins. For instance, a proteinloaded biopolymer microgel could be prepared under conditions where the polymer network is shrunken and has small pores, thereby inhibiting the release of the proteins. The solution conditions could then be changed to cause the polymer network to swell, thereby facilitating the release of the proteins due to simple diffusion (see previous section). 6.2.3. Specific molecular interactions An alternative mechanism for controlling the retention and release of proteins from polymeric colloidal particles is to utilize specific attractive or repulsive interactions between the proteins and the polymer molecules [23,26]. If there is a strong attraction between the proteins and polymers, then the protein will tend to be retained inside the colloidal particles, but if there is no attraction or a repulsion then the proteins will tend to be released. One of the most commonly used means of controlling molecular interactions is based on changing the pH or ionic strength to weaken or strengthen the electrostatic interactions within the colloidal particles [22]. In particular, the net charge on proteins usually goes from positive to negative as the pH is increased from below to above their isoelectric points (Fig. 7a). Consequently, the proteins tend to be attracted (retained) to anionic polymers at low pH values, but repelled (released) at high pH values. To a first approximation, the strength of the electrostatic interaction can be estimated using the following expression [22]: EI ¼ −ζ Protein ζ Polymer ð5Þ µ µ Fig. 6. Prediction of the release of bioactive proteins from polymeric colloidal particles. The parameters used in the predictions were: hydrodynamic radius of proteins (rH) = 5 nm; pore size (ζ) of polymer network = 1 nm; radius of polymer chains = 1 nm. The diffusion coefficient of the protein through the particles was calculated using the equation given in text. The radius of the colloidal particles is given next to lines. D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22 9

Here, ζProtein and ζPolymer are the effective surface potentials (ζ- potentials) of the protein and polymer, respectively. The value of the electrostatic interaction will be attractive when this term is positive, and repulsive when it is negative. The change in the electrostatic interaction parameter with pH for the whey protein–alginate system is shown in Fig. 7b. The strongest electrostatic attraction occurs around pH 3.5, and therefore this would be expected to be the pH where the protein was held inside the particles most strongly. In practice, one must also take into account the change in charge in the polymer molecules due to any ion-binding effects associated with their cross-linking inside the colloidal particles. In addition, bioactive proteins typically have a mixture of both cationic and anionic groups on their surfaces, and so they may be able to bind under conditions where both the protein and polymer have similar net charges [77]. An example of the retention/release of proteins from polymeric colloidal particles due to a change in electrostatic interactions is highlighted in Fig. 8, which shows the impact of pH on the retention of a model globular protein (whey protein) from alginate microgels [25]. The proteins are held inside the alginate microgels at low pH values where the two biopolymers have opposite charges (positive and negative), but are released at high pH values where they have similar charges (both negative). It should be noted that the strength of the electrostatic attraction may be Fig. 7. The electrical surface potential (ζ-potential) of bioactive proteins and of the polymers used to construct colloidal particles often depends on pH: (a) ζ-potential versus pH profile of whey protein and alginate solutions; (b) Effective electrostatic interaction versus pH for a mixed whey protein and alginate system, calculated from ζ-potential measurements. Fig. 8. Confocal fluorescence microscopy images of the impact of pH on the retention and release of a model protein (whey protein isolate) encapsulated in calcium alginate microgels. The protein was stained green. Data from Zhang et al. (2016), Food Hydrocolloids, 58, 308–315. 10 D.J. McClements / Advances in Colloid and Interface Science 253 (2018) 1–22